Dutch Contributions to the Fourteenth International Congress of Slavists: Dutch Contributions to the Fourteenth International Congress of Slavists. (Studies in Slavic and General Linguistics)

Citation preview

Dutch Contributions 2008 Linguistics

Studies in Slavic and General Linguistics Series Editors:

Peter Houtzagers · Janneke Kalsbeek · Jos Schaeken

Editorial Advisory Board: R. Alexander (Berkeley) · A.A. Barentsen (Amsterdam) · B. Comrie (Leipzig) B.M. Groen (Amsterdam) · F.H.H. Kortlandt (Leiden) · W. Lehfeldt (Göttingen) G. Spieß (Cottbus) · R. Sprenger (Amsterdam) · W.R. Vermeer (Leiden)

Amsterdam – New York 2008

Studies in Slavic and General Linguistics, vol. 34

Dutch Contributions to the Fourteenth International Congress of Slavists Ohrid September 10-16, 2008

Linguistics

edited by Peter Houtzagers Janneke Kalsbeek Jos Schaeken

The paper on which this book is printed meets the requirements of “ISO 9706:1994, Information and documentation - Paper for documents Requirements for permanence”. ISBN: 978-90-420-2442-7 ©Editions Rodopi B.V., Amsterdam - New York, NY 2008 Printed in the Netherlands

CONTENTS AԌӢӠԂӰ ԮԂӢӾӰԽԙӾӰ ғѳҩүѭѣѢѨѣ ҭҫҋһѣѩҫғүѷѣһѻѢҫҋѷѨ ѩѣҢҋѷғѨҢ ҭҩѨ ҭҫғѷҫҩѤѣҴҫҋѷѨ ғ ҭҩҫҹһҫҴ ғ ҋҫғҩѣҴѣѢѢѳҤ ҋһүғѤѢҋҘѨҤ Ѥ҈ѳҘүҤ

1

Rick Derksen NOTES ON THE SLAVIC METATHESIS OF LIQUIDS

37

Egbert Fortuin SEMANTICS AND SYNTAX OF THE CONSTRUCTION OF DEGREE IN RUSSIAN

45

René Genis PERDURATIVITY IN POLISH – 100 YEARS AFTER AGRELL

109

W. Andries van Helden VICISSITUDES OF THE GENITIVE RULE

145

Pepijn Hendriks and Jos Schaeken ON THE COMPOSITION AND LANGUAGE OF THREE EARLY MODERN RUSSIAN-GERMAN PHRASEBOOKS

217

Pepijn Hendriks THE EXPRESSION OF FUTURE TENSE IN THE PHRASEBOOK OF TÖNNIES FENNE (PSKOV, 1607): A PHILOLOGICAL APPROACH

259

Wim Honselaar RECIPROCAL MARKERS IN RUSSIAN

279

Peter Houtzagers ON BURGENLAND CROATIAN ISOGLOSSES

293

Janneke Kalsbeek PAST TENSE HABITUAL EXPRESSIONS IN OLD CROATIAN TEXTS

333

Cornelia E. Keijsper FROM OVS ORDER TO CONVERSE STRUCTURE IN RUSSIAN AND OTHER SLAVIC LANGUAGES

349

Frederik Kortlandt SLAVIC HISTORICAL MORPHOLOGY: NOMINAL PARADIGMS

397

Cecilia Odé TRANSCRIPTION OF RUSSIAN INTONATION, TORI, AN INTERACTIVE RESEARCH TOOL AND LEARNING MODULE ON THE INTERNET

431

CONTENTS

Tijmen Pronk FONOLOŠKI OPIS GOVORA POTOƀ V ZILJSKI DOLINI

451

EԊӾӰԂ TӢӠԼӫԶӠӰӠӰԂ ѢѣҋҘҫһѻҘҫ үҘҋѨҫҴ ҫ ҭҩѨһүҐүѷѣһѻѢѳҤ-үѢѷҫѢѨҴүҤ: ҘҫѢѷҩүҩҐѰҴѣѢѷѳ Ѩ҈ ҩѰҋҋҘҫҐҫ Ѥ҈ѳҘү

461

William R. Veder THE GLAGOLITIC BARRIER

489

Willem Vermeer PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION

503

Dutch Contributions to the Fourteenth International Congress of Slavists, Ohrid: Linguistics (SSGL 34). Amsterdam – New York: Rodopi, 2008, 1-36.

ғѳҩүѭѣѢѨѣ ҭҫҋһѣѩҫғүѷѣһѻѢҫҋѷѨ ѩѣҢҋѷғѨҢ ҭҩѨ ҭҫғѷҫҩѤѣҴҫҋѷѨ ғ ҭҩҫҹһҫҴ ғ ҋҫғҩѣҴѣѢѢѳҤ ҋһүғѤѢҋҘѨҤ Ѥ҈ѳҘүҤ 1 үѩҩѨүѢ ҷүҩѣѢѷҋѣѢ

1. ғӊѦҦѦӁѹѦ ҭҝѹ ѹӁҲѦҝҰҝѦҲҽ҃ѹѹ ҲѦҔҡҲҽ ӁѦҡҸѸӁѦӁӁҸ ӊҽңӁҸ ҸҰҝѦҦѦґѹҲҧ ҸҲӁҸҗѦӁѹҾ ѸѦңҦҏ ҏҔҽӉҽӁӁӃѸѹ ӊ ӁѦѸ ҡѹҲҏҽ҃ѹҾѸѹ. ҭҸҙҲҸѸҏ Ұҝѹ ҽӁҽґѹӉѦ ҲҸҊҸ ѹґѹ ѹӁҸҊҸ ҾӉӃҔҽ ӁѦҸқѽҸҦѹѸҸ ѹӉҏҍѹҲҧ ҝҽӉґѹҍӁӃѦ ҾӉӃҔҸӊӃѦ ҡҝѦҦҡҲӊҽ, ӊӃҝҽңҽӀӂѹѦ ҙҲѹ ҸҲӁҸҗѦӁѹҾ. ҷҸґҧҗҸѦ ӊӁѹѸҽӁѹѦ ҙҲѹѸ ӊҸҰҝҸҡҽѸ ҏҦѦґҾӀҲ ҲҽҔѹѦ ґѹӁҊӊѹҡҲӃ, ҔҽҔ Ҍ.ҋ. ҴҽҡґҸӊ, ү.ғ. ҷҸӁҦҽҝҔҸ ѹ ғ.ҋ. ҤҝҽҔҸӊҡҔѹҮ, ҝҽҡҡѸҽҲҝѹӊҽӀӂѹѦ ҦҽӁӁӃѦ ҾӊґѦӁѹҾ ӊ ҝҽѸҔҽѽ ӍҏӁҔ҃ѹҸӁҽґҧӁҸ-ҡѦѸҽӁҲѹҍѦҡҔҸҮ ҔҽҲѦҊҸҝѹѹ “ҲҽҔҡѹҡ”. 2 ҐґҽӊӁҸѦ ӊӁѹѸҽӁѹѦ Ұҝѹ ҙҲҸѸ ӁҽҰҝҽӊґѦӁҸ Ӂҽ ӊӃҝҽңѦӁѹѦ ҸҡӁҸӊӁӃѽ ӊҝѦѸѦӁӁӃԀѽ ҸҲӁҸҗѦӁѹҮ: ҸҦӁҸӊҝѦѸѦӁӁҸҡҲҧ – ҰҝѦҦҗѦҡҲӊҸӊҽӁѹѦ – ҡґѦҦҸӊҽӁѹѦ. ғ ҡґҽӊҾӁҡҔѹѽ ҾӉӃҔҽѽ ҦҸҡҲҽҲҸҍӁҸ ҦѹӍӍѦҝѦӁ҃ѹҝҏѦѸҽҾ ѹӁӍҸҝѸҽ҃ѹҾ Ҹ ӊҝѦѸѦӁӁӃѽ ҸҲӁҸҗѦӁѹҾѽ ҡѹҲҏҽ҃ѹҮ ҦҽѦҲҡҾ ӊ ҝҽѸҔҽѽ ҡґҸңӁҸҰҸҦҍѹӁѦӁӁӃѽ ҰҝѦҦґҸңѦӁѹҮ ҡ ӊҝѦѸѦӁӁӃѸѹ ҡҸӀӉҽѸѹ. ѢҽҰҝѹѸѦҝ, ӊ ҝҏҡҡҔҸѸ ҾӉӃҔѦ ҸҰҝѦҦѦґѦӁӁӃѦ ҝҽӉӁҸӊѹҦӁҸҡҲѹ ҏҔҽӉҽӁӁӃѽ ӊӃҗѦ ҲҝѦѽ ҸҡӁҸӊӁӃѽ ҸҲӁҸҗѦӁѹҮ ӊӃҝҽңҽӀҲҡҾ ҡҸҸҲӊѦҲҡҲӊѦӁӁҸ ҡҸӀӉҽѸѹ ԗӹӪԂ, ӪԂӪ ԽӹԊӸӪӹ ѹ ԗӹӪԂ ӰӾ. ҭҝѹӊѦҦѦѸ ӁѦҡҔҸґҧҔҸ ҰҝѹѸѦҝҸӊ ѹӉ ҝҸѸҽӁҽ ҷҏґҊҽҔҸӊҽ ԠԂԙԽӾӢ Ӡ ԠԂӢԄԂӢӠԽԂ: (1) ҭҸҔҽ ҰҝѹҗѦҦҗѹҮ Ұѹґi ѹ Ѧґi, ҭѹґҽҲ, ҰҝѹѽґѦқӃӊҽҾi ӊѹӁҸ, ԗӹԄԊԋԌՂӴԂԊi ҰҝѹӂҏҝѦӁӁӃѸѹ ҊґҽӉҽѸѹ Ӂҽ ҡӊҸѦҊҸ ҊҸҡҲҾ. 3 (2) Ѥ ҲҏҲ ҰҸҔҽ ґѦңҽґi [...] ҸҍѦӁҧ ѸӁҸҊҸѦ ԗӹӰԋԊp. 1

ѩҽӁӁҸѦ ѹҡҡґѦҦҸӊҽӁѹѦ ӊҸӉӁѹҔґҸ ӊ ҝҽѸҔҽѽ ҦѦҾҲѦґҧӁҸҡҲѹ “ҽҡҰѦҔҲҸґҸҊѹҍѦҡҔҸҊҸ Ҕґҏқҽ”, ӁѦӍҸҝѸҽґҧӁҸҮ ҝҽқҸҍѦҮ ҊҝҏҰҰӃ Ұҝѹ ҔҽӍѦҦҝѦ ҡґҽӊҾӁҡҔѹѽ ҾӉӃҔҸӊ үѸҡҲѦҝҦҽѸҡҔҸҊҸ ҏӁѹӊѦҝҡѹҲѦҲҽ. ҘҝҸѸѦ ҽӊҲҸҝҽ ӊ ҊҝҏҰҰҏ ӊѽҸҦҾҲ ҩѦӁѦ ҐѦӁѹҡ (ҰҸґҧҡҔѹҮ ҾӉӃҔ), ҴҽҊҦҽ ӊҽӁ ѩҺҮҔѦҝѦӁҐҝҽқҸӊҽ (ҍѦҗҡҔѹҮ ҾӉӃҔ), ѤӁӁѦҔѦ ҘҽґҡқѦѦҔ ѹ ҩҽҦҸӊҽӁ һҏҍѹҍ (ѽҸҝӊҽҲҡҔѹҮ ѹ ҡѦҝқҡҔѹҮ ҾӉӃҔѹ). үӊҲҸҝ ҰҝѹӁҸҡѹҲ ѹѸ ӊҡѦѸ қґҽҊҸҦҽҝӁҸҡҲҧ Ӊҽ ҰҸѸҸӂҧ ӊ ҝѦҗѦӁѹѹ ҝҾҦҽ ӊҸҰҝҸҡҸӊ, ӊҸӉӁѹҔҗѹѽ ӊ ѽҸҦѦ ѹҡҡґѦҦҸӊҽӁѹҾ. ҈ҽ ӊҽңӁҏӀ ҰҸѸҸӂҧ Ұҝѹ ҰҝѦҦҸҡҲҽӊґѦӁѹѹ ѹ ҸқҡҏңҦѦӁѹѹ ѸҽҔѦҦҸӁҡҔҸҊҸ ѸҽҲѦҝѹҽґҽ ҡѦҝҦѦҍӁҸ қґҽҊҸҦҽҝѹѸ ҲҽҔңѦ ҢҽҽҰҽ ѹ ҋӁѦңҽӁҏ ҘҽѸҰѽҺҮҡ. 2 ҋѸ. ҝҽқҸҲҏ (ҤҝҽҔҸӊҡҔѹҮ 2003), ҊҦѦ ҏҔҽӉҽӁҽ ѹ қҸґѦѦ ҝҽӁӁҾҾ ґѹҲѦҝҽҲҏҝҽ. 3 ғ ҰҝѹѸѦҝҽѽ ӊѹҦ ҸқҡҏңҦҽѦѸӃѽ ӍҸҝѸ ҏҔҽӉӃӊҽѦҲҡҾ ҡҏҰѦҝҡҔҝѹҰҲҽѸѹ: p (ҡҸӊѦҝҗѦӁӁӃҮ ӊѹҦ) ѹ i (ӁѦҡҸӊѦҝҗѦӁӁӃҮ). (ҋҝ. ҝҽқҸҲӃ (Forsyth 1970), (Mønnesland 1984) ѹ (Dickey 2000).)

2

үѩҩѨүѢ ҷүҩѣѢѷҋѣѢ

(3) ҘҽҔ ҲҸґҧҔҸ ҰѦҝӊӃҮ ҔҏҰґѦҲ ҰҝѹҗѦґp Ҕ ҔҸӁ҃ҏ, ҰѦӁѹѦ ԙԽӠԖԊӹp ӊӁѦӉҽҰӁҸ [...] (4) ѷҽҔ ԶԊӠi ѸҸґҍҽ ӁѦҔҸҲҸҝҸѦ ӊҝѦѸҾ, ҰҸҔҽ ҸӁҽ ӁѦ ӊӃӁҏґҽp ҏ ѸѦӁҾ ѹӉ ҝҏҔ ҃ӊѦҲӃ [...] ҘҽҔ ҰҸҔҽӉӃӊҽӀҲ ҙҲѹ ҰҝѹѸѦҝӃ, ҡҏӂѦҡҲӊҏѦҲ ҦҸҡҲҽҲҸҍӁҸ ҡѹґҧӁҽҾ ѹӉқѹҝҽҲѦґҧӁҸҡҲҧ ҰҸ ҸҲӁҸҗѦӁѹӀ Ҕ ҡҸҍѦҲҽѦѸҸҡҲѹ ҦҽӁӁӃѽ ҡҸӀӉҸӊ ҡ ӊѹҦҸӊӃѸѹ ӍҸҝѸҽѸѹ. ҫҦӁҸӊҝѦѸѦӁӁҸҡҲҧ Ұҝѹ ҡҸӀӉѦ ԗӹӪԂ ҰҸҦҦѦҝңѹӊҽѦҲҡҾ ӁѦҡҸӊѦҝҗѦӁӁӃѸ ӊѹҦҸѸ (Ѣҋғ) ӊ ҰҝѹҦҽҲҸҍӁҸҮ ҍҽҡҲѹ ҰҝѦҦґҸңѦӁѹҾ. ҫҦӁҽҔҸ ӉӁҽҍѦӁѹѦ ҡґѦҦҸӊҽӁѹҾ ҡҸӀӉҽ ԗӹӪԂ ӰӾ, ҔҽҔ ҰҝҽӊѹґҸ, ҲҝѦқҏѦҲ ҔҽҔ ҝҽӉ ӊӃқҸҝҽ ҡҸӊѦҝҗѦӁӁҸҊҸ ӊѹҦҽ (ҋғ). ҭҝѹ ҡҸӀӉѦ ӪԂӪ ԽӹԊӸӪӹ ҸқӃҍӁҸ ӊ ҸқѦѹѽ ҍҽҡҲҾѽ ҰҝѦҦґҸңѦӁѹҾ ҏҰҸҲҝѦқґҾѦҲҡҾ ҋғ. ҭҝѹ ҡҸӀӉѦ ӪӹԄԌԂ, ѹѸѦӀӂѦѸ ӁѦҦѹӍӍѦҝѦӁ҃ѹҝҸӊҽӁӁҸѦ ӉӁҽҍѦӁѹѦ (ҩҐ 1980: 542), ӊӃқҸҝ ӊѹҦҸӊӃѽ ӍҸҝѸ ѸҸңѦҲ ѹҊҝҽҲҧ ҝѦҗҽӀӂҏӀ ҝҸґҧ Ұҝѹ ҝҽӉґѹҍѦӁѹѹ ҸҦӁҸӊҝѦѸѦӁӁҸҡҲѹ ѹґѹ ҰҝѦҦҗѦҡҲӊҸӊҽӁѹҾ. ғ ҰѦҝӊҸѸ ҡґҏҍҽѦ ӊӃқѹҝҽӀҲҡҾ ҲҽҔѹѦ ңѦ ӍҸҝѸӃ, ҔҽҔ Ұҝѹ ҡҸӀӉѦ ԗӹӪԂ, ҽ ӊҸ ӊҲҸҝҸѸ, ҔҽҔ Ұҝѹ ҡҸӀӉѦ ӪԂӪ ԽӹԊӸӪӹ: (5) (6)

[ғҸҝҸқѦҮ] ӴԊӾԽӾԊp, ҔҸҊҦҽ Ҿ ҸҲѽҸҦѹґi ҸҲ ҸҔӁҽ [ҡҝ. (2)]. ҘҸҊҦҽ ңѦ ӁҽҡҲҽґҽp ҲѹҗѹӁҽ, ҽҝҲѹҡҲ ԙӪԂԒԂԊp [...] [ҡҝ. (3)].

ғѦҡҧѸҽ ӊҽңӁҸ ҸҲѸѦҲѹҲҧ, ҍҲҸ ӊ ҝҏҡҡҔҸѸ ҾӉӃҔѦ ҏҔҽӉҽӁӁҽҾ ҰҝѹҔҝѦҰґѦӁӁҸҡҲҧ ҲҸҊҸ ѹґѹ ѹӁҸҊҸ ӊѹҦҽ Ҕ ҸҲҦѦґҧӁӃѸ ҲҽҔҡѹҡӁӃѸ ҸҲӁҸҗѦӁѹҾѸ ѸҸңѦҲ ҃ѦґѹҔҸѸ ҸҲѸѦӁҾҲҧҡҾ Ұҝѹ ҰҸӊҲҸҝҾѦѸҸҡҲѹ, ҸҡҸқѦӁӁҸ ӊ ҲѦѽ ҡґҏҍҽҾѽ, ҔҸҊҦҽ ҝѦҍҧ ѹҦѦҲ Ҹ ҦѦҮҡҲӊѹҾѽ ӊ ҰҝҸҗґҸѸ. ѩґҾ ҝҏҡҡҔҸҊҸ ҾӉӃҔҽ ѽҽҝҽҔҲѦҝӁҸ, ҍҲҸ ӊ ҲҽҔѹѽ ҡґҏҍҽҾѽ ҦҸҰҏҡҔҽѦҲҡҾ ӍҽҔҲѹҍѦҡҔѹ ҲҸґҧҔҸ Ѣҋғ. ѷҽҔѹѸ ҸқҝҽӉҸѸ, ӊ ҝҾҦѦ ҡґҏҍҽѦӊ ӉҽѸѦӁҽ ҋғ Ӂҽ Ѣҋғ ӊ ҡґҸңӁӃѽ ҰҝѦҦґҸңѦӁѹҾѽ ҏҔҽӉҽӁӁҸҊҸ ҲѹҰҽ ӁѦҰҸҡҝѦҦҡҲӊѦӁӁҸ ӊӃҝҽңҽѦҲ ҰҝҸҲѹӊҸҰҸҡҲҽӊґѦӁѹѦ ѦҦѹӁѹҍӁҸҡҲҧ – ҰҸӊҲҸҝҾѦѸҸҡҲҧ. ғ ҡӊҸѦѸ ѹӉӊѦҡҲӁҸѸ ҰҸҡҸқѹѹ ҰҸ ҏҰҸҲҝѦқґѦӁѹӀ ӊѹҦҸӊ ҊґҽҊҸґҽ ҫ.ҭ. ҩҽҡҡҏҦҸӊҽ ѹґґӀҡҲҝѹҝҏѦҲ ҦҽӁӁҸѦ ҰҸґҸңѦӁѹѦ ҦѦґ ҡґѦҦҏӀӂѹѸѹ ҰҝѹѸѦҝҽѸѹ (1968: 48-49): (7ҽ) ҘҽҔ ҲҸґҧҔҸ ҦҸҔґҽҦҍѹҔ ҰҸҦӁҾґҡҾp Ӂҽ ҲҝѹқҏӁҏ, ӢԂԒԌԂԊӠԙӸp ҽҰґҸҦѹҡѸѦӁҲӃ [ѦҦѹӁѹҍӁҸҡҲҧ]; (7қ) ҘҽҔ ҲҸґҧҔҸ ҦҸҔґҽҦҍѹҔ ҰҸҦӁѹѸҽґҡҾi Ӂҽ ҲҝѹқҏӁҏ, ӢԂԒԌԂӴԂԊӠԙӸi ҽҰґҸҦѹҡѸѦӁҲӃ [ҰҸӊҲҸҝҾѦѸҸҡҲҧ]. (8ҽ) ҘҸҊҦҽ ӊѦҲѦҝ ҏҡѹґѹґҡҾp, ҊҝѦҡҲѹi ԙԽԂԊӹp ҲҝҏҦӁѦѦ [ѦҦѹӁѹҍӁҸҡҲҧ]; (8қ) ҘҸҊҦҽ ӊѦҲѦҝ ҏҡѹґѹӊҽґҡҾi, ҊҝѦҡҲѹi ԙԽԂӰӹӴӠԊӹԙӸi ҲҝҏҦӁѦѦ [ҰҸӊҲҸҝҾѦѸҸҡҲҧ]. ҫқҾӉҽҲѦґҧӁҸҡҲҧ ӊӃқҸҝҽ Ѣҋғ ӊ ҰҸҦҸқӁӃѽ ҰҝѦҦґҸңѦӁѹҾѽ Ұҝѹ Ӂҽґѹҍѹѹ ҰҸӊҲҸҝҾѦѸҸҡҲѹ ҰҝѹӊҸҦѹҲ Ҕ ҲҸѸҏ, ҍҲҸ Ұҝѹ ҡҸӀӉҽѽ ҲѹҰҽ ӪӹԄԌԂ ӉҽҲҏҗѦӊӃӊҽѦҲҡҾ ҝҽӉґѹҍѹѦ ѸѦңҦҏ (ҍҽҡҲѹҍӁҸҮ) ҸҦӁҸӊҝѦѸѦӁӁҸҡҲҧӀ (ҔҽҔ ӊ ҰҝѹѸѦҝѦ (5)) ѹ ҰҝѦҦҗѦҡҲӊҸӊҽӁѹѦѸ (ҔҽҔ ӊ ҰҝѹѸѦҝѦ (6)). ѣҡґѹ, ӁҽҰҝѹѸѦҝ, ӊ ҰҝѹѸѦҝѦ (7қ) ӉҽѸѦӁѹҲҧ ҡҸӀӉ ӪԂӪ ԽӹԊӸӪӹ Ӂҽ ӪӹԄԌԂ, ҡҲҽӁҸӊѹҲҡҾ ӁѦҾҡӁӃѸ, ҔҸҊҦҽ ҝҽӉҦҽӊҽґѹҡҧ

ҭҫҋһѣѩҫғүѷѣһѻѢҫҋѷѻ ѩѣҢҋѷғѨҢ ҭҩѨ ҭҫғѷҫҩѤѣҴҫҋѷѨ

3

ҽҰґҸҦѹҡѸѦӁҲӃ: ѹѸѦӁӁҸ ӊ ҲѦ ѸҸѸѦӁҲӃ, ҔҸҊҦҽ ҦҸҔґҽҦҍѹҔ ѦӂѦ ҰҸҦӁѹѸҽґҡҾ Ӂҽ ҲҝѹқҏӁҏ, ѹґѹ ӊ ҲҸ ӊҝѦѸҾ, ҔҸҊҦҽ ҏңѦ ҡҲҸҾґ Ӂҽ ӁѦҮ. ѷ.Ѧ. ҸҲӁҸҗѦӁѹҾ ѸѦңҦҏ ҦѦҮҡҲӊѹҾѸѹ, ҔҸҲҸҝӃѦ Ұҝѹ ѦҦѹӁѹҍӁӃѽ ҦѦҮҡҲӊѹҾѽ ҦҸҡҲҽҲҸҍӁҸ ҾҡӁҸ ӊӃҝҽңҽӀҲҡҾ ӊӃқҸҝҸѸ ӊѹҦҸӊӃѽ ӍҸҝѸ, Ұҝѹ ҰҸӊҲҸҝҾѦѸҸҡҲѹ ѸҸҊҏҲ ҡҲҽҲҧ ѸѦӁѦѦ ҍѦҲҔѹѸѹ. ғӃҗѦҏҔҽӉҽӁӁҸѦ ҰҸґҸңѦӁѹѦ ҦѦґ ҡґѦҦҏѦҲ ҡҍѹҲҽҲҧ ҦґҾ ҝҏҡҡҔҸҊҸ ҾӉӃҔҽ ӊѦҡҧѸҽ ѦҡҲѦҡҲӊѦӁӁӃѸ ѹ ҡҏӂѦҡҲӊѦӁӁӃѸ. ҫҦӁҽҔҸ ҡґѦҦҏѦҲ ҸҲѸѦҲѹҲҧ, ҍҲҸ ҲҽҔҏӀ ҲѦҡӁҏӀ ҡӊҾӉҧ ҰҸӊҲҸҝҾѦѸҸҡҲѹ ҡ Ѣҋғ ӁѦґҧӉҾ ҡҍѹҲҽҲҧ ѽҽҝҽҔҲѦҝӁҸҮ ҦґҾ ӊҡѦѽ ҡґҽӊҾӁҡҔѹѽ ҾӉӃҔҸӊ. ғ ӁҽҏҍӁҸҮ ґѹҲѦҝҽҲҏҝѦ ҦҽӊӁҸ ҸҲѸѦҍѦӁҸ, ҍҲҸ ӊ ҦҽӁӁҸѸ ҸҲӁҸҗѦӁѹѹ ѸѦңҦҏ ҸҲҦѦґҧӁӃѸѹ ҡґҽӊҾӁҡҔѹѸѹ ҾӉӃҔҽѸѹ ҡҏӂѦҡҲӊҏӀҲ ӉҽѸѦҲӁӃѦ ҝҽӉґѹҍѹҾ. 4 ҶҲҸ ҸҲӁҸҡѹҲҡҾ ѹѸѦӁӁҸ Ҕ ҡґҏҍҽҾѸ Ҳ.Ӂ. ӁѦҸҊҝҽӁѹҍѦӁӁҸҔҝҽҲӁҸҊҸ ӉӁҽҍѦӁѹҾ, ҸҡҸқѦӁӁҸ Ұҝѹ ӍҸҝѸҽѽ ҰҝҸҗѦҦҗѦҊҸ ӊҝѦѸѦӁѹ. 5 ѢѦ ҝҽӉ ҸҲѸѦҍҽґҸҡҧ, ӁҽҰҝѹѸѦҝ, ҍҲҸ ӊ ҸҲґѹҍѹѦ ҸҲ ҝҏҡҡҔҸҊҸ ҾӉӃҔҽ, ӊ ҍѦҗҡҔҸѸ ҾӉӃҔѦ ӊ ҲҽҔѹѽ ҡґҏҍҽҾѽ җѹҝҸҔҸ ҏҰҸҲҝѦқґҾѦҲҡҾ ҋғ. 6 ҋҔҽӉҽӁӁҸѦ ҸҲӁҸҡѹҲҡҾ ҲҽҔңѦ Ҕ ҰҝѦҦґҸңѦӁѹҾѸ ҸқҡҏңҦҽѦѸҸҊҸ ҲѹҰҽ, ѹѸѦӀӂѹѸ ӊҝѦѸѦӁӁҏӀ ҰҝѹҦҽҲҸҍӁҏӀ ҍҽҡҲҧ. ҋҝҽӊӁѹҲѦ ҡґѦҦҏӀӂѹҮ ҰҝѹѸѦҝ ѹӉ ҷҏґҊҽҔҸӊҽ, ҡ ѦҊҸ ҰѦҝѦӊҸҦҸѸ Ӂҽ ҍѦҗҡҔѹҮ ҾӉӃҔ: (9ҽ) һѹҗҧ ҲҸґҧҔҸ ӁҽҍѹӁҽґi ӉӊѦӁѦҲҧi ҲѦґѦӍҸӁ, ғҽҝѦӁҏѽҽ ԼӢԂԊi ҲҝҏқҔҏ [...] (9қ) Sotva zazvonilp telefon, Varenucha zvedlp sluchátko [...] ғ ҲҸ ӊҝѦѸҾ ҔҽҔ ӊ ҸқѦѹѽ ҍҽҡҲҾѽ ҝҏҡҡҔҸҊҸ ҰҝѦҦґҸңѦӁѹҾ (9ҽ) ҦґҾ ӊӃҝҽңѦӁѹҾ ҰҸӊҲҸҝҾѦѸҸҡҲѹ ҦҽӁӁҸҮ ҃ѦҰҸҍҔѹ ҡҸқӃҲѹҮ ӁѦҸқѽҸҦѹѸ Ѣҋғ, ӊ ҍѦҗҡҔҸѸ ҡҸҸҲӊѦҲҡҲӊѹѹ (9қ) ӊ ҡѹґҏ ӉӁҽҍѦӁѹҾ ҦҽӁӁҸҊҸ ҡҸӀӉҽ ӍҽҔҲѹҍѦҡҔѹ ҸқҾӉҽҲѦґҧӁӃѸ ҾӊґҾѦҲҡҾ ҔҽҔ ҝҽӉ ҋғ. ѷҽҔѹѸ ҸқҝҽӉҸѸ ӊ ҍѦҗҡҔҸѸ ҾӉӃҔѦ ӊ ҰҸҦҸқӁӃѽ ҡґҏҍҽҾѽ ҍѦҲҔҸ ҏҔҽӉӃӊҽѦҲҡҾ ӊѹҦҸӊҽҾ ѽҽҝҽҔҲѦҝѹҡҲѹҔҽ ҔҽңҦҸҊҸ ҸҲҦѦґҧӁҸҊҸ ҰҝҸҾӊґѦӁѹҾ ҦѦҮҡҲӊѹҾ, ӊҡґѦҦҡҲӊѹѦ ҍѦҊҸ ӊӉҽѹѸӁӃѦ ҸҲӁҸҗѦӁѹҾ ѸѦңҦҏ ҏҔҽӉҽӁӁӃѸѹ ҦѦҮҡҲӊѹҾѸѹ ҰҝѦҦҡҲҽӊґѦӁӃ ҲҽҔ ңѦ ҲҸҍӁҸ, ҔҽҔ ҙҲҸ қӃґҸ қӃ ӊ ҡґҏҍҽѦ ӁѦҰҸӊҲҸҝҾѦѸҸҡҲѹ. ҫҦӁҽҔҸ ӊҡґѦҦҡҲӊѹѦ ҙҲҸҊҸ ҡҽѸ ӊӃқҸҝ ӊѹҦҸӊӃѽ ӍҸҝѸ ҏңѦ ӁѦ ҏҔҽӉӃӊҽѦҲ, ѹҦѦҲ ґѹ ӉҦѦҡҧ ҝѦҍҧ Ҹ ѦҦѹӁѹҍӁӃѽ ѹґѹ Ҹ ҰҸӊҲҸҝҾӀӂѹѽҡҾ ҡґҏҍҽҾѽ, Ҳ.Ѧ. 4

ҋѸ., ӁҽҰҝѹѸѦҝ, ҝҽқҸҲӃ (ѨӊҽӁҍѦӊ 1973) ѹ (Galton 1976). ҋҸҦѦҝңҽҲѦґҧӁӃѦ ҸқӉҸҝӃ ҡ ҰҸҦҝҸқӁӃѸ ҏҔҽӉҽӁѹѦѸ ґѹҲѦҝҽҲҏҝӃ Ҹқ ҙҲҸѸ ҰҝѦҦѸѦҲѦ ҦҽӀҲҡҾ ӊ ҝҽқҸҲҽѽ (ҭѦҲҝҏѽѹӁҽ 2000) ѹ (Dickey 2000). 5 ғ ҦҽӁӁҸҮ ҝҽқҸҲѦ ҝҽҡҡѸҽҲҝѹӊҽѦѸ ҲҸґҧҔҸ ӁѦҸҊҝҽӁѹҍѦӁӁҸ-ҔҝҽҲӁӃҮ ӊҽҝѹҽӁҲ ҰҸӊҲҸҝҾѦѸҸҡҲѹ. ҤҸҝҸҗҸ ѹӉӊѦҡҲӁҸ, ҍҲҸ Ұҝѹ ҸҊҝҽӁѹҍѦӁӁҸҮ ҔҝҽҲӁҸҡҲѹ, Ҳ.Ѧ. Ұҝѹ ҸқҡҲҸҾҲѦґҧҡҲӊҽѽ ҲѹҰҽ ԌӴԂԞԌՂ, ԌӴԂ ӢԂԒԂ, ӢԂԒԂ ԌӴԂ, ӰӾԙӪӹԊӸӪӹ ӢԂԒ, ѹ ӊ ҝҏҡҡҔҸѸ ҾӉӃҔѦ ҍҽҡҲҸ ӊҡҲҝѦҍҽѦҲҡҾ ҋғ. ҷҸґѦѦ ҦѦҲҽґҧӁҸ Ҹқ ҙҲҸѸ ҡѸ. ҝҽқҸҲӃ (ҷҽҝѦӁҲҡѦӁ 1992, 1994 ѹ 1997) ѹ ҏҔҽӉҽӁӁҏӀ ӊ Ӂѹѽ ґѹҲѦҝҽҲҏҝҏ. ҫ ҡҸҸҲӁҸҗѦӁѹѹ ҝҏҡҡҔҸҊҸ ѹ ҰҸґҧҡҔҸҊҸ ҾӉӃҔҸӊ ӊ ҙҲҸҮ ҸқґҽҡҲѹ ҡѸ. ҝҽқҸҲҏ (ҋҲҽӊӁѹ҃Ҕҽ 2003). ѢҽҡҔҸґҧҔҸ ӁҽѸ ѹӉӊѦҡҲӁҸ, ҰҸҔҽ ӁѦҲ ҦѦҲҽґҧӁӃѽ ҦҽӁӁӃѽ Ҹқ ҙҲҸѸ ҲѹҰѦ ҰҸӊҲҸҝҾѦѸҸҡҲѹ ӊ ҦҝҏҊѹѽ ҡґҽӊҾӁҡҔѹѽ ҾӉӃҔҽѽ. ѢѦҔҸҲҸҝӃѦ ҡӊѦҦѦӁѹҾ ҦҽӀҲҡҾ ӊ ҝҽқҸҲѦ Ҍ.ҭ. ҘӁҾӉѦӊҽ (1989: 139–140) 6 ҫқ ҙҲҸѸ, ӁҽҰҝѹѸѦҝ, ҏңѦ ҰҸґӊѦҔҽ ӁҽӉҽҦ Ұѹҡҽґҽ ѣґѦӁҽ ҘҝңѹңҔҸӊҽ (Kǥížková 1958: 198).

4

үѩҩѨүѢ ҷүҩѣѢѷҋѣѢ

ӊ ҍѦҗҡҔҸѸ ҾӉӃҔѦ ҦҽӁӁӃҮ ҽҡҰѦҔҲ ѹӁҲѦҝҰҝѦҲҽ҃ѹѹ ӊӃҡҔҽӉӃӊҽӁѹҾ ӊ ӉӁҽҍѹҲѦґҧӁҸ қҸґҧҗѦҮ ҡҲѦҰѦӁѹ ӉҽӊѹҡѹҲ ҸҲ ҔҸӁҲѦҔҡҲҽ, ҍѦѸ ҏ ҡҸҸҲӊѦҲҡҲӊҏӀӂѹѽ ҝҏҡҡҔѹѽ ӊӃҡҔҽӉӃӊҽӁѹҮ. ҭҝѹѸѦҝӃ (9ҽ) ѹ (9қ) ҰҸҦҲӊѦҝңҦҽӀҲ ѸӁѦӁѹѦ, ҍҲҸ ҝҏҡҡҔѹҮ ѹ ҍѦҗҡҔѹҮ ҾӉӃҔѹ ӉҽӁѹѸҽӀҲ ҰҝҾѸҸ ҰҝҸҲѹӊҸҰҸґҸңӁӃѦ ҰҸӉѹ҃ѹѹ ҰҸ ҸҲӁҸҗѦӁѹӀ Ҕ ӍҏӁҔ҃ѹҸӁѹҝҸӊҽӁѹӀ ҔҽҲѦҊҸҝѹѹ ӊѹҦҽ Ұҝѹ ӊӃҝҽңѦӁѹѹ ҦҽӁӁӃѽ ҡґҏҍҽѦӊ ҰҸӊҲҸҝҾѦѸҸҡҲѹ. ҫҦӁҽҔҸ ҽӁҽґѹӉ ҰҸҦҸқӁӃѽ ҡҲҝҏҔҲҏҝ ӊ ҦҝҏҊѹѽ ҡґҽӊҾӁҡҔѹѽ ҾӉӃҔҽѽ ҰҸҔҽӉӃӊҽѦҲ, ҍҲҸ ҡҏӂѦҡҲӊҏӀҲ ѦӂѦ ѹ ҦҝҏҊѹѦ ҡѹҡҲѦѸӁӃѦ ӊҸӉѸҸңӁҸҡҲѹ. ѢҽҰҝѹѸѦҝ, ӊ ӁѹңѦҰҝѹӊѦҦѦӁӁҸѸ ҡѦҝқҡҔҸѸ ҰѦҝѦӊҸҦѦ ҲҸҊҸ ңѦ ӍҝҽҊѸѦӁҲҽ ӁҽѽҸҦѹѸ ҡӊҸѦҸқҝҽӉӁҸѦ ҡҸҍѦҲҽӁѹѦ ҸқҸѹѽ ӊѹҦҸӊ (ӊ ҰҝѹҦҽҲҸҍӁҸҮ ҍҽҡҲѹ – ҋғ, ҽ ӊ ҊґҽӊӁҸҮ – Ѣҋғ). ғ ҦҽӁӁҸѸ ҡґҏҍҽѦ ҙҲҸ ҡӊҾӉҽӁҸ ѦӂѦ ѹ ҡ ӊӃқҸҝҸѸ ҊґҽҊҸґҧӁҸҮ ӍҸҝѸӃ ҦҝҏҊҸҊҸ ӁҽҔґҸӁѦӁѹҾ (ҔҸӁҦѹ҃ѹҸӁҽґҽ): (9ӊ) ѯѹѸ қѹ ӉҽӉӊҸӁѹҸp ҲѦґѦӍҸӁ, ғҽҝѦӁҏѽҽ ԼӠ ԌӠԒԂӹi ҡґҏҗҽґѹ҃ҏ [...] ҭҸ ӁҽҗѦѸҏ ѸӁѦӁѹӀ, ӊѦҡҧѸҽ ңѦґҽҲѦґҧӁҸ ҰҸґҏҍѹҲҧ қҸґѦѦ ҰҸґӁҏӀ ҔҽҝҲѹӁҏ Ҹ ҲҸѸ, ҔҽҔ ҡѦқҾ ӊѦҦҏҲ ҝҽӉґѹҍӁӃѦ ҡґҽӊҾӁҡҔѹѦ ҾӉӃҔѹ ӊ ҡґҏҍҽҾѽ ҏҔҽӉҽӁӁҸҊҸ ҲѹҰҽ. ғѦҦҧ ҲҸґҧҔҸ Ӂҽ ҸҡӁҸӊѦ ҰҸҦҸқӁӃѽ ѸҽҲѦҝѹҽґҸӊ ѸҸңӁҸ қҏҦѦҲ ҰҝѹҮҲѹ Ҕ қҸґѦѦ ҲҸҍӁҸѸҏ ҸҰҝѦҦѦґѦӁѹӀ ҡѦѸҽӁҲѹҔѹ ҸҡӁҸӊӁӃѽ ҊґҽҊҸґҧӁӃѽ ҔҽҲѦҊҸҝѹҮ ӊ ҡґҽӊҾӁҡҔѹѽ ҾӉӃҔҽѽ ѹ Ұҝҽӊѹґ ѹѽ ӍҏӁҔ҃ѹҸӁѹҝҸӊҽӁѹҾ. 7 ғ ӁҽҡҲҸҾӂѦҮ ҡҲҽҲҧѦ ҦѦґҽѦҲҡҾ ҰҸҰӃҲҔҽ ҰҸҦҸқӁҸҊҸ ҸқӉҸҝҽ. ҭҝѹӊҸҦҾҲҡҾ ӁѦҔҸҲҸҝӃѦ ҝѦӉҏґҧҲҽҲӃ ҡҸҰҸҡҲҽӊѹҲѦґҧӁҸҊҸ ҽӁҽґѹӉҽ ѸҽҲѦҝѹҽґҽ ӊӃҗѦҏҔҽӉҽӁӁҸҊҸ ҲѹҰҽ ѹӉ ҰҸҍҲѹ ӊҡѦѽ ҡҸӊҝѦѸѦӁӁӃѽ ҡґҽӊҾӁҡҔѹѽ ҾӉӃҔҸӊ. ѷҽҔѹѸ ҸқҝҽӉҸѸ, Ӂҽҗҽ ҝҽқҸҲҽ ҸҲӁҸҡѹҲҡҾ Ҕ ҸқґҽҡҲѹ ҡҸҰҸҡҲҽӊѹҲѦґҧӁҸҊҸ ѹҡҡґѦҦҸӊҽӁѹҾ ҲҽҔҡѹҡҽ, ҡ ҸҡҸқӃѸ ӊӁѹѸҽӁѹѦѸ Ӂҽ ҝҸґҧ ҰҸӊҲҸҝҾѦѸҸҡҲѹ Ұҝѹ ӊӃҝҽңѦӁѹѹ ҸҰҝѦҦѦґѦӁӁӃѽ ҲҽҔҡѹҡӁӃѽ ҸҲӁҸҗѦӁѹҮ. ғ ҃ѦӁҲҝѦ ӁҽҗѦҊҸ ӊӁѹѸҽӁѹҾ ӁҽѽҸҦѹҲҡҾ ҔҽҲѦҊҸҝѹҾ ӊѹҦҽ. ҭҝѹ ҙҲҸѸ ӊ ҸҡӁҸӊӁҸѸ ҝҽҡҡѸҽҲҝѹӊҽӀҲҡҾ ҡґҏҍҽѹ ҡ ҰҝѦҦѦґҧӁӃѸѹ ҦѦҮҡҲӊѹҾѸѹ, ҸқҝҽӉҏӀӂѹѸѹ ӁѦҔҸҲҸҝҏӀ ҃ѦҰҸҍҔҏ. ѨѸѦӁӁҸ ӊ ҲҽҔѹѽ ҏҡґҸӊѹҾѽ ӊ ҡґҏҍҽҾѽ ӁѦҰҸӊҲҸҝҾѦѸҸҡҲѹ, ҔҽҔ ҰҝҽӊѹґҸ, ӊӃқѹҝҽѦҲҡҾ ҋғ ѹ ҲҸґҧҔҸ ҏ ҰҝѦҦѦґҧӁӃѽ ҊґҽҊҸґҸӊ ҡҏӂѦҡҲӊҏѦҲ ҍѹҡҲҸӊѹҦҸӊҸѦ ҰҝҸҲѹӊҸҰҸҡҲҽӊґѦӁѹѦ. 8 ѩҽґѦѦ ѸӃ ҸҊҝҽӁѹҍѹӊҽѦѸҡҾ ҲҸґҧҔҸ ҝҽҡҡѸҸҲҝѦӁѹѦѸ ҡґҏҍҽѦӊ ҰҸӊҲҸҝҾѦѸҸҡҲѹ ӊ ҰҝҸҗґҸѸ, ҰҸҡҔҸґҧҔҏ ѹѸѦӁӁҸ ӉҦѦҡҧ ҝҽӉґѹҍѹҾ ѸѦңҦҏ ҡґҽӊҾӁҡҔѹѸѹ ҾӉӃҔҽѸѹ ҾӊґҾӀҲҡҾ ӁҽѹқҸґѦѦ ҡҏӂѦҡҲӊѦӁӁӃѸѹ. ҫқӂѦѹӉӊѦҡҲӁҸ, ҍҲҸ ѹ ӊ ҝҏҡҡҔҸѸ ҾӉӃҔѦ ӍҸҝѸҽ ҋғ ҰҝѦӉѦӁҡҽ җѹҝҸҔҸ ҏҰҸҲҝѦқґҾѦҲҡҾ ӊӁѦ ҡӍѦҝӃ ҏҔҽӉҽӁѹҾ ѦҦѹӁѹҍӁҸҊҸ ҦѦҮҡҲӊѹҾ, Ҳ.Ѧ. Ұҝѹ ҸҰѹҡҽӁѹѹ ҰҝѹӊӃҍѦҔ, Ҹқӂѹѽ ҲѦӁҦѦӁ҃ѹҮ ѹґѹ ӉҽҔҸӁҸѸѦҝӁҸҡҲѦҮ. ѣҡґѹ ӊ ҰҝѹѸѦҝѦ (8ҽ) ӉҽѸѦӁѹҲҧ ӍҸҝѸӃ ҰҝҸҗѦҦҗѦҊҸ ӊҝѦѸѦӁѹ Ӂҽ 7 ғ ҡҲҽҲҧѦ ҴѹґҔѹ Ѩӊѹҍ (Iviǀ 1983) ҦҽѦҲҡҾ ѹӁҲѦҝѦҡӁӃҮ ҸқӉҸҝ ҝҽӉґѹҍӁӃѽ ҡҰҸҡҸқҸӊ ҰѦҝѦҦҽҍѹ ҰҸӊҲҸҝҾѦѸҸҡҲѹ ӊ ҡґҽӊҾӁҡҔѹѽ ҾӉӃҔҽѽ, ӁҸ ҸҡҸқѦӁӁҸҡҲѹ ҡґҸңӁҸҰҸҦҍѹӁѦӁӁӃѽ ҰҝѦҦґҸңѦӁѹҮ ҲҽѸ ӁѦ ҝҽҡҡѸҽҲҝѹӊҽӀҲҡҾ. 8 ғҡґѦҦ Ӊҽ ҷҸӁҦҽҝҔҸ (1991: 198) ѸӃ ҰҸӁѹѸҽѦѸ ҰҝѦҦѦґҧӁҸҡҲҧ ҔҽҔ ҡѦѸҽӁҲѹҍѦҡҔѹҮ ҰҝѹӉӁҽҔ, ҰҝѹҡҏҲҡҲӊҏӀӂѹҮ ҏ ӊҡѦѽ ҊґҽҊҸґҸӊ ҋғ (ҷҽҝѦӁҲҡѦӁ 1995: 5–13; 1998: 46–49).

ҭҫҋһѣѩҫғүѷѣһѻѢҫҋѷѻ ѩѣҢҋѷғѨҢ ҭҩѨ ҭҫғѷҫҩѤѣҴҫҋѷѨ

5

ӍҸҝѸӃ ҰҝѦӉѦӁҡҽ, ҰҸґҏҍѹѸ ҰҝѦҦґҸңѦӁѹѦ, ҔҸҲҸҝҸѦ ѸҸңӁҸ ѹӁҲѦҝҰҝѦҲѹҝҸӊҽҲҧ ҏңѦ ҦӊҸҾҔҸ: ґѹқҸ ҔҽҔ ҸҰѹҡҽӁѹѦ ѦҦѹӁѹҍӁҸҮ ҃ѦҰҸҍҔѹ ҦѦҮҡҲӊѹҮ ӊ қҏҦҏӂѦѸ, ґѹқҸ ҔҽҔ ҏҔҽӉҽӁѹѦ ҲҸҊҸ, ҍҲҸ ҰҸҦҸқӁҽҾ ҃ѦҰҸҍҔҽ ѸҸңѦҲ ӊҸӉӁѹҔӁҏҲҧ ӊ ґӀқҸѦ ӊҝѦѸҾ ѹ ӊҡґѦҦҡҲӊѹѦ ҙҲҸҊҸ ӊ ҦѦҮҡҲӊѹҲѦґҧӁҸҡҲѹ ґѦҊҔҸ ѸҸңѦҲ қӃҲҧ ҰҸӊҲҸҝҾӀӂѦҮҡҾ: 9 (8ҽ' ) ҘҸҊҦҽ ӊѦҲѦҝ ҏҡѹґѹҲҡҾp, ҊҝѦҡҲѹ ԙԽԂӰӾԽp ҲҝҏҦӁѦѦ. ҫҡӁҸӊҸҮ ӁҽҗѦҊҸ ҽӁҽґѹӉҽ ҾӊґҾӀҲҡҾ ҰҝѹѸѦҝӃ ѹӉ ҰҽҝҽґґѦґҧӁӃѽ ҲѦҔҡҲҸӊ, ҡҸқҝҽӁӁӃѽ ҽӊҲҸҝҸѸ Ӊҽ ҰҸҡґѦҦӁѹѦ ҊҸҦӃ ӊ ҝҽѸҔҽѽ ҝҽқҸҲӃ ӁҽҦ ҡҸӉҦҽӁѹѦѸ ҙґѦҔҲҝҸӁӁҸҊҸ ҔҸҝҰҏҡҽ ҦґҾ ҡҸҰҸҡҲҽӊѹҲѦґҧӁҸҊҸ ѹӉҏҍѦӁѹҾ ҡґҽӊҾӁҡҔѹѽ ҾӉӃҔҸӊ – ASPAC (Amsterdam Slavic Parallel Aligned Corpus). ғ ӁҽҡҲҸҾӂѦѦ ӊҝѦѸҾ ҙҲҸҲ ҔҸҝҰҏҡ ѦӂѦ ҦҽґѦҔ ҸҲ ңѦґҽѦѸҸҊҸ ҡҸҡҲҸҾӁѹҾ, ӁҸ ҲѦѸ ӁѦ ѸѦӁѦѦ ӁҽѸ ҰҝѦҦҡҲҽӊґҾѦҲҡҾ, ҍҲҸ Ұҝѹ ѹӉҏҍѦӁѹѹ ѹӁҲѦҝѦҡҏӀӂѹѽ Ӂҽҡ ӊҸҰҝҸҡҸӊ ѹӉ ӁѦҊҸ ҏңѦ ѸҸңӁҸ ѹӉӊґѦҍҧ ӁѦҔҸҲҸҝӃѦ ҦҸҡҲҽҲҸҍӁҸ ѹӁҲѦҝѦҡӁӃѦ ҡӊѦҦѦӁѹҾ. 10 ѩґҾ ӊӃҾӊґѦӁѹҾ ҡҰѦ҃ѹӍѹҔѹ ҰҸӊѦҦѦӁѹҾ ҊґҽҊҸґҧӁӃѽ ӍҸҝѸ ӊ ҡґҸңӁҸҰҸҦҍѹӁѦӁӁӃѽ ҰҝѦҦґҸңѦӁѹҾѽ ҝҽҡҡѸҸҲҝѹѸ ҡӁҽҍҽґҽ ӁѦҡҔҸґҧҔҸ ҡґҏҍҽѦӊ ӊӃҝҽңѦӁѹҾ ҰҸӊҲҸҝҾѦѸҸҡҲѹ ӊ ӁѦӉҽӊѹҡѹѸӃѽ ҰҝѦҦґҸңѦӁѹҾѽ. 2. ҫ ӁѦҔҸҲҸҝӃѽ ҡѽҸҦҡҲӊҽѽ ѹ ҝҽӉґѹҍѹҾѽ ӊ ӊӃҝҽңѦӁѹѹ ҰҸӊҲҸҝҾѦѸҸҡҲѹ ӊ ӁѦӉҽӊѹҡѹѸӃѽ ҰҝѦҦґҸңѦӁѹҾѽ ҝҽӉґѹҍӁӃѽ ҡґҽӊҾӁҡҔѹѽ ҾӉӃҔҸӊ ҭҝѹӁҾҲҸ ҡҍѹҲҽҲҧ, ҍҲҸ ӊ ҡґҽӊҾӁҡҔѹѽ ҾӉӃҔҽѽ ӊӃҝҽңѦӁѹѦ ҰҸӊҲҸҝҾӀӂѹѽҡҾ ҦѦҮҡҲӊѹҮ ҾӊґҾѦҲҡҾ ҸҦӁҸҮ ѹӉ ҃ѦӁҲҝҽґҧӁӃѽ ӍҏӁҔ҃ѹҮ ӁѦҡҸӊѦҝҗѦӁӁҸҊҸ ӊѹҦҽ. ҭҝѹ ҙҲҸѸ ҸҦӁҽ ѹӉ ҸҡӁҸӊӁӃѽ ҰҝҸқґѦѸ ҽӁҽґѹӉҽ ӍҏӁҔ҃ѹҸӁѹҝҸӊҽӁѹҾ ӊѹҦҽ ӉҽҔґӀҍҽѦҲҡҾ ӊ ҲҸѸ, ҍҲҸ Ұҝѹ ҰҸӊҲҸҝҾѦѸҸҡҲѹ ѸҸңѦҲ ӁѦҮҲҝҽґѹӉҸӊҽҲҧҡҾ ҰҝҸҲѹӊҸҰҸҡҲҽӊґѦӁѹѦ ҃ѦґҸҡҲӁҸҡҲѹ/ӉҽҔҸӁҍѦӁӁҸҡҲѹ – ӁѦӉҽҔҸӁҍѦӁӁҸҡҲѹ ҰҸ ҸҲӁҸҗѦӁѹӀ Ҕ ҔҽңҦҸѸҏ ҸҲҦѦґҧӁҸ ӊӉҾҲҸѸҏ ҰҝҸҾӊґѦӁѹӀ ҦѦҮҡҲӊѹҾ. ѷҽҔѹѸ ҸқҝҽӉҸѸ ҃ѦґҸҡҲӁҸѦ ҰҸӊҲҸҝҾӀӂѦѦҡҾ ҦѦҮҡҲӊѹѦ ѸҸңѦҲ ӊӃҝҽңҽҲҧҡҾ ӍҸҝѸҸҮ Ѣҋғ, ӊ ҲҸ 9 ҭҝѹ ҦҸҡҲҽҲҸҍӁҸҮ ҰҸҦҦѦҝңҔѦ ҔҸӁҲѦҔҡҲҽ, ӁҽҰҝѹѸѦҝ Ұҝѹ Ӂҽґѹҍѹѹ ҍҽҡҲѹ҃Ӄ ԼՂӴԂԊӹ, ҲҽҔҸѦ ҸқӂѦѦ ҰҸґҸңѦӁѹѦ ҦѦґ ѸҸңѦҲ ҸҲӁҸҡѹҲҧҡҾ ҦҽңѦ Ҕ ҰҝҸҗґҸѸҏ. 10 ғ ҦҽӁӁӃҮ ѸҸѸѦӁҲ ҰҽҝҽґґѦґҧӁӃҮ ҔҸҝҰҏҡ ASPAC ҡҸҦѦҝңѹҲ қҸґҧҗѦ 30 ҲѦҔҡҲҸӊ, ҰҝѦҦҡҲҽӊґѦӁӁӃѽ, ҔҽҔ ѸѹӁѹѸҏѸ, Ӂҽ ҍѦҲӃҝѦѽ ҡґҽӊҾӁҡҔѹѽ ҾӉӃҔҽѽ. ғ ҔҸҝҰҏҡѦ ӁҽѹқҸґҧҗѦѦ ҔҸґѹҍѦҡҲӊҸ ҰѦҝѦӊҸҦҸӊ ѹѸѦӀҲ: Lewis Carroll Alice’s Adventures in Wonderland (Ӂҽ 11 ҡґҽӊҾӁҡҔѹѽ ҾӉӃҔҽѽ), Alan A. Milne Winnie-the-Pooh (11), J.R.R. Tolkien The Hobbit (11), Antoine de SaintExupéry Le petit prince (11), Ҵѹѽҽѹґ ҷҏґҊҽҔҸӊ ԠԂԙԽӾӢ Ӡ ԠԂӢԄԂӢӠԽԂ (11), Rudyard Kipling The Jungle Book (ҝҽҡҡҔҽӉӃ Ҹ ҴҽҏҊґѹ) (10), J.K. Rowling Harry Potter and the Philosopher’s Stone (10). ҋѦҝқҸґҏңѹ҃ҔѹҮ ҾӉӃҔ, Ҕ ҡҸңҽґѦӁѹӀ, ҰҝѦҦҡҲҽӊґѦӁ ҲҸґҧҔҸ ҸҦӁѹѸ ҲѦҔҡҲҸѸ (ҰѦҝѦӊҸҦҸѸ ҔӁѹҊѹ Winnie-the-Pooh). ҭҸ ҙҲҸҮ ҰҝѹҍѹӁѦ ѸӃ ӊӃӁҏңҦѦӁӃ ҸҡҲҽӊѹҲҧ ҡѦҝқҸґҏңѹ҃ҔѹѦ ҾӉӃҔѹ ӊ ҦҽӁӁҸҮ ҡҲҽҲҧѦ ӊӁѦ ӊӁѹѸҽӁѹҾ (ҸҦӁҽҔҸ ҡѸ. ҰҝѹѸѦҝ (18)). ғҡѦ ҸҡҲҽґҧӁӃѦ ҸҡӁҸӊӁӃѦ ҡґҽӊҾӁҡҔѹѦ ґѹҲѦҝҽҲҏҝӁӃѦ ҾӉӃҔѹ, ҽ ѹѸѦӁӁҸ: ҝҏҡҡҔѹҮ, қѦґҸҝҏҡҡҔѹҮ, ҏҔҝҽѹӁҡҔѹҮ, ҰҸґҧҡҔѹҮ, ҍѦҗҡҔѹҮ, ҡґҸӊҽ҃ҔѹҮ, ҡґҸӊѦӁҡҔѹҮ, ѽҸҝӊҽҲҡҔѹҮ, ҡѦҝқҡҔѹҮ, ѸҽҔѦҦҸӁҡҔѹҮ ѹ қҸґҊҽҝҡҔѹҮ ӊ ҡҲҽҲҧѦ ҝҽҡҡѸҽҲҝѹӊҽӀҲҡҾ.

6

үѩҩѨүѢ ҷүҩѣѢѷҋѣѢ

ӊҝѦѸҾ ҔҽҔ Ұҝѹ ӁѦҰҸӊҲҸҝҾѦѸҸҡҲѹ ҙҲҸ ҦѦҮҡҲӊѹѦ қӃґҸ қӃ ӊӃҝҽңѦӁҸ ҰҝҸҲѹӊҸҰҸґҸңӁӃѸ, ҡҸӊѦҝҗѦӁӁӃѸ ӊѹҦҸѸ (ҡѸ. Ұҽҝ. 1). ҭҝѹѸѦҝӃ ҏҰҸҲҝѦқґѦӁѹҾ ӍҸҝѸ Ѣҋғ ҦґҾ ӊӃҝҽңѦӁѹҾ ҰҸӊҲҸҝҾӀӂѦҊҸҡҾ ҃ѦґҸҡҲӁҸҊҸ ҰҝѦҦѦґҧӁҸҊҸ ҦѦҮҡҲӊѹҾ ҦѦҮҡҲӊѹҲѦґҧӁҸ ѸҸңӁҸ ӁҽҮҲѹ ӊҸ ӊҡѦѽ ҡґҽӊҾӁҡҔѹѽ ҾӉӃҔҽѽ. ғ ҰҽҝҽґґѦґҧӁӃѽ ҲѦҔҡҲҽѽ ӁѦҝѦҦҔҸ ҰҝѦҦҡҲҽӊґѦӁӃ ҲҽҔңѦ ҡґҏҍҽѹ, ҊҦѦ ӊҡѦ ҡґҽӊҾӁҡҔѹѦ ҾӉӃҔѹ ҦѦґҽӀҲ ӍҽҔҲѹҍѦҡҔѹ ҸҦѹӁҽҔҸӊӃҮ ӊӃқҸҝ. ғ ҔҽҍѦҡҲӊѦ ҰҝѹѸѦҝҽ ҰҝѹӊѦҦѦѸ ӍҝҽҊѸѦӁҲ ѹӉ ѹӉӊѦҡҲӁҸҮ ҦѦҲҡҔҸҮ ҔӁѹҊѹ ԛӠӰӰӠ ӷӫԖ ҡ ѦҊҸ ҰѦҝѦӊҸҦҽѸѹ Ӂҽ ҃ѦґӃҮ ҝҾҦ ҡҸӊҝѦѸѦӁӁӃѽ ҡґҽӊҾӁҡҔѹѽ ҾӉӃҔҸӊ. 11 ғ ҦҽӁӁҸѸ ӍҝҽҊѸѦӁҲѦ ҸҰѹҡӃӊҽѦҲҡҾ, ҔҽҔ ҸҡґѹҔ Ѩҽ-Ѩҽ ѹҊҝҽѦҲ ҡ ҲҸґҧҔҸ ҍҲҸ ҰҸҦҽҝѦӁӁӃѸѹ ѦѸҏ ҊҸҝҗҔҸѸ ѹ (ґҸҰӁҏӊҗѹѸ) җҽҝѹҔҸѸ: (10) Alan Alexander Milne: Winnie-The-Pooh үӁҊґ.

But Eeyore wasn’t listening. He was taking the balloon out, and putting it back again, as happy as could be.... ҩҏҡҡҔѹҮ ѢҸ Ѩҽ-Ѩҽ ѹ ӁѦ ҡґҏҗҽґ. ҫӁ ҸҰҏҡҔҽґi җҽҝѹҔ ӊ ҊҸҝҗҸҔ ѹ ӊӃӁѹѸҽґi ѦҊҸ ҡӁҸӊҽ ѹ ҡӁҸӊҽ. ҫӁ қӃґ ҡҍҽҡҲґѹӊ (ҴҽҔҡѹѸѹҗѹӁҽ). 12 ѰҔҝ. ѷҽ ӎҽ-ӎҽ ӁҟҍҸҊҸ ӁѦ ҍҏӊ. ҢҸѸҏ қҏґҸ ӁѦ ҦҸ ҲҸҊҸ: ӊҟӁ ҲҸ Ҕґҽӊi ҡӊҸӀ ҔҏґҧҔҏ ӊ ҊҸҝӂѹҔ, ҲҸ ӊѹҲҾҊҽӊi ѶѶ ӁҽӉҽҦ, ҟ қҏґҸ ӊѹҦӁҸ, ӂҸ ӊҟӁ ҲҽҔѹ ҰҸ-ҡҰҝҽӊңӁҧҸѸҏ ӂҽҡґѹӊѹҮ... ҭҸґҧҡҔ. Lecz K’apouchy nie s’ucha’. Wk’ada’i balonik i wyjmowa’i go z powrotem, szczNjǧliwy jak jeszcze nigdy w ǻyciu (Tuwim). ѯѦҗҡҔ. Ale Ijáǁek je neposlouchal. Vyndávali balónek a zase jej ukládali, celý šǬastný. ҋґҸӊҽ҃Ҕ. Ale Híkaj ich už nepoǁúval. Vyberali balón z pohára a zase ho tam vkladali a bol najšǬastnejší zo všetkých... ҋґҸӊѦӁҡҔ. Toda Sivǁek ju ni poslušal. Ves sreǁen je jemali balon iz lonca in ga spet devali vanj...

11

ғ ӁҽҗѦѸ ҔҸҝҰҏҡѦ ӁѦ ѹѸѦѦҲҡҾ қѦґҸҝҏҡҡҔҸҊҸ ҰѦҝѦӊҸҦҽ ҙҲҸҊҸ ҰҝҸѹӉӊѦҦѦӁѹҾ. ѢҽҮҦѦӁӁӃѦ ӊ ҔҸҝҰҏҡѦ ҰѦҝѦӊҸҦӃ ҦҽӁӁҸҊҸ ӍҝҽҊѸѦӁҲҽ Ӂҽ ҡѦҝқҸґҏңѹ҃ҔҸѸ ѹ ѽҸҝӊҽҲҡҔҸѸ ѸѦӁѦѦ ҰҝѹҊҸҦӁӃ ҦґҾ ҡҝҽӊӁѦӁѹҾ, ҰҸҡҔҸґҧҔҏ ҸӁѹ ҰҸ ҡҲҝҸѦӁѹӀ ҡґѹҗҔҸѸ ҡѹґҧӁҸ ҸҲґѹҍҽӀҲҡҾ ҸҲ ҸҡҲҽґҧӁӃѽ. ғ ѽҸҝӊҽҲҡҔҸѸ ҰѦҝѦӊҸҦѦ ѹҡҰҸґҧӉҸӊҽӁӃ ҸҲҊґҽҊҸґҧӁӃѦ ҡҏӂѦҡҲӊѹҲѦґҧӁӃѦ. ҭҸ-ҝҏҡҡҔѹ қҏҔӊҽґҧӁҸ: ӹӰ ԗӹԙӴԋԽӠԊ ԙӾԼԋ ӴՂӰӠԐԂӰӠԇ Ӡ ӴӪԊԂԌՂӴԂӰӠԇ... ғ ҡѦҝқҸґҏңѹ҃ҔҸѸ ҰѦҝѦӊҸҦѦ ѹҡҰҸґҧӉҸӊҽӁӃ ӍҸҝѸӃ ҋғ, ӁҸ ҙҲҸ ѸҸҲѹӊѹҝҸӊҽӁҸ ҲѦѸ, ҍҲҸ ӊѸѦҡҲҸ ҰҸӊҲҸҝҾӀӂѦҮҡҾ ҃ѦҰҸҍҔѹ ҸҰѹҡӃӊҽѦҲҡҾ ѦҦѹӁѹҍӁҽҾ ҃ѦҰҸҍҔѹ, ҡ қóґҧҗѹѸ ҔҸґѹҍѦҡҲӊҸѸ ӉӊѦӁҧѦӊ – ӴԒԋԊ ԶԂӢӠӪ ӠԒ ԄӹӢԶӪԂ, ԙӫӰӫԊ ӾԄӹ ӹԼӢԂԽӰӹ, ӴՂӰӫԊ ӾԄӹ, ԙӫӰӫԊ ӾԄӹ ӹԼӢԂԽӰӹ... 12 ҭѦҝѦӊҸҦҍѹҔ ҏҔҽӉҽӁ ҲҸґҧҔҸ ӊ ҡґҏҍҽѦ ӁҽґѹҍѹҾ ӊ ҔҸҝҰҏҡѦ ҝҽӉӁӃѽ ҰѦҝѦӊҸҦҸӊ ҰҝҸѹӉӊѦҦѦӁѹҾ Ӂҽ ҸҦӁҸѸ ѹ ҲҸѸ ңѦ ҾӉӃҔѦ.

ҭҫҋһѣѩҫғүѷѣһѻѢҫҋѷѻ ѩѣҢҋѷғѨҢ ҭҩѨ ҭҫғѷҫҩѤѣҴҫҋѷѨ

ҋѦҝқҡҔѹҮ ҴҽҔѦҦҸӁ.

ҷҸґҊҽҝҡҔ.

7

Ali Iar ih nije slušao. Vadioi je balon i vraǀaoi ga natrag, sav sreǀan ... ѢҸ ѰҗґѦ ӁѦ ҡґҏҗҽҗѦ. ѷҸҎ ҡѹ ҊҸ ӊҽҦѦҗѦi [impf.] 13 ѹ ҡѹ ҊҸ ҡҲҽӊҽҗѦi [impf.] қҽґҸӁҍѦҲҸ ӊҸ ҊҝӁѦҲҸ, ѹ ҰҸӊѦҡѦґ ӁѦ ѸҸңѦҗѦ Ӂѹ Ҧҽ қѹҦѦ... ѢҸ ѨҮҸҝѹ ӁѦ ҍҏӊҽҗѦ ӁѹӂҸ. ѷҸҮ ӊҽҦѦҗѦi [impf.] қҽґҸӁҽ ѹ ҊҸ ҡҰҏҡҔҽҗѦi [impf.] ҰҽҔ, ӂҽҡҲґѹӊ ҔҽҲҸ ӁѹҔҸҊҽ.

ғҸ ӊҡѦѽ ҙҲѹѽ ҲѦҔҡҲҽѽ ҦѦҮҡҲӊѹҾ ҊґҽҊҸґҸӊ, ӊӃҦѦґѦӁӁӃѽ ңѹҝӁӃѸ җҝѹӍҲҸѸ, ӁѦҡҸѸӁѦӁӁҸ ҾӊґҾӀҲҡҾ ҰҝѦҦѦґҧӁӃѸѹ. Ѣҽ ҏҝҸӊӁѦ ҸҲҦѦґҧӁҸ ӊӉҾҲӃѽ ҰҝҸҾӊґѦӁѹҮ ҙҲѹѽ ҦѦҮҡҲӊѹҮ ӊ ҦҽӁӁҸѸ ҡґҏҍҽѦ ҸқҝҽӉҏѦҲҡҾ ҃ѦҰҸҍҔҽ, ӊ ҔҽңҦҸѸ ӉӊѦӁѦ ҔҸҲҸҝҸҮ ҾӊӁҸ ҰҝѹҡҏҲҡҲӊҏѦҲ ҰҝѹӉӁҽҔ ҃ѦґҸҡҲӁҸҡҲѹ, Ҳ.Ѧ. ҦҸҡҲѹҊӁҏҲҸҡҲѹ ҦҽӁӁҸҊҸ ҰҝѦҦѦґҽ. ғ ҡґҏҍҽѦ ҸҦӁҸҮ ѦҦѹӁҡҲӊѦӁӁҸҮ ҃ѦҰҸҍҔѹ ӊҡѦ ҡґҽӊҾӁҡҔѹѦ ҾӉӃҔѹ ӁѦҡҸѸӁѦӁӁҸ ӊӃқҝҽґѹ қӃ ҡҸӊѦҝҗѦӁӁӃҮ ӊѹҦ, ҔҽҔ, ӁҽҰҝѹѸѦҝ, ӊ ҝҏҡҡҔҸѸ ӊҽҝѹҽӁҲѦ: ӹԗӫԙԽӠԊp ԶԂӢӠӪ Ӡ ԙӰӹӴԂ ӴՂӰӫԊp ӾԄӹ. 14 ғӃқҸҝ ҰҝҸҲѹӊҸҰҸґҸңӁҸҊҸ, ӁѦҡҸӊѦҝҗѦӁӁҸҊҸ ӊѹҦҽ ӊҸ ӊҡѦѽ ҦҽӁӁӃѽ ҡґҽӊҾӁҡҔѹѽ ҰҝѹѸѦҝҽѽ ѸҸҲѹӊѹҝҸӊҽӁ ҊґҽӊӁӃѸ ҸқҝҽӉҸѸ ҲѦѸ, ҍҲҸ ҏҔҽӉҽӁӁҽҾ ҃ѦҰҸҍҔҽ ҦѦҮҡҲӊѹҮ ҰҸҡҲҸҾӁӁҸ ҰҸӊҲҸҝҾѦҲҡҾ. ҘҝҸѸѦ ҲҸҊҸ, ӁҽӊѦҝӁҸѦ, ӁѦѸҽґҸӊҽңӁҸ, ҍҲҸ ҦѦҮҡҲӊѹҾ ҾӊґҾӀҲҡҾ ҽӁҲҸӁѹѸѹҍӁӃѸѹ ѹ ӁѦҰҝѦҝӃӊӁҸ ҡґѦҦҏӀҲ ҦҝҏҊ Ӊҽ ҦҝҏҊҸѸ ӊ ҲѦҍѦӁѹѦ ҦҸӊҸґҧӁҸ ҔҸҝҸҲҔҸҊҸ ҰѦҝѹҸҦҽ. ҋѽѦѸҽҲѹҍѦҡҔѹ ҙҲҸ ѸҸңӁҸ қӃґҸ қӃ ѹӉҸқҝҽӉѹҲҧ ҰҝѹѸѦҝӁҸ ҲҽҔ: ... ٕٕٕٕٔٔٔٔ ... ғҡѦ ҙҲҸ ҡҰҸҡҸқҡҲӊҏѦҲ ӊҸҡҰҝѹҾҲѹӀ ҦѦҾҲѦґҧӁҸҡҲѹ ҸҡґѹҔҽ ҔҽҔ ҡӊҸѦҊҸ ҝҸҦҽ ҰҝҸ҃Ѧҡҡҽ қҸґѦѦ ӊӃҡҸҔҸҊҸ ҰҸҝҾҦҔҽ, Ӂҽ ҍҲҸ, ҰҸңҽґҏҮ, ҲҽҔңѦ ҏҔҽӉӃӊҽѦҲ ҏҰҸҲҝѦқґѦӁѹѦ ҰҝҸҊҝѦҡҡѹӊҽ ӊ ҽӁҊґѹҮҡҔҸѸ ҸҝѹҊѹӁҽґѦ. ҘҽҔ қӃґҸ ҏңѦ ҸҲѸѦҍѦӁҸ ӊ ҰҝѦҦӃҦҏӂѦѸ ҰҽҝҽҊҝҽӍѦ, ҦґҾ ҝҏҡҡҔҸҊҸ ҾӉӃҔҽ ӊӃқҸҝ Ѣҋғ ӊ ҲҽҔѹѽ ҡґҏҍҽҾѽ ҾӊґҾѦҲҡҾ ӊѦҡҧѸҽ “ѦҡҲѦҡҲӊѦӁӁӃѸ”, ҸҡҸқѦӁӁҸ Ұҝѹ ӍҸҝѸҽѽ ҰҝҸҗѦҦҗѦҊҸ ӊҝѦѸѦӁѹ. ѢҸ ӊ ҦҽӁӁҸѸ ҡґҏҍҽѦ, ӊ ҡѹґҏ ҏҔҽӉҽӁӁӃѽ ҰҝѹҍѹӁ, ҲҽҔҸҮ ңѦ ӊӃқҸҝ қӃґ ҡҦѦґҽӁ ҦҽңѦ ӊ ҲѦѽ ҾӉӃҔҽѽ, ӊ ҔҸҲҸҝӃѽ ӊ ҏҡґҸӊѹҾѽ ӁѦҸҊҝҽӁѹҍѦӁӁҸҮ ҔҝҽҲӁҸҡҲѹ ӉӁҽҍѹҲѦґҧӁҸ ҡӊҸқҸҦӁѦѦ ҏҰҸҲҝѦқґҾѦҲҡҾ ҋғ, ҔҽҔ ӁҽҰҝѹѸѦҝ ӊ ҍѦҗҡҔҸѸ. ѯҲҸқӃ ҰҸґҏҍѹҲҧ ҰҝѦҦҡҲҽӊґѦӁѹѦ Ҹ ӍҽҔҲҸҝҽѽ, ҡҰҸҡҸқҡҲӊҏӀӂѹѽ ҲҽҔҸѸҏ ҏҰҸҲҝѦқґѦӁѹӀ ҋғ, ҝҽҡҡѸҸҲҝѹѸ ҡґѦҦҏӀӂѹҮ ҰҝѹѸѦҝ, ѽҸҝҸҗҸ ҸҲҝҽңҽӀӂѹҮ ҙҲҸ ҝҽӉґѹҍѹѦ ѸѦңҦҏ ҡґҽӊҾӁҡҔѹѸѹ ҾӉӃҔҽѸѹ. ғ ҦҽӁӁҸѸ ӍҝҽҊѸѦӁҲѦ ҸҰѹҡӃӊҽѦҲҡҾ, ҔҽҔ, қӃӊҽґҸ, ӁѦҔҽҾ ҊҸҡҰҸңҽ ѷҸӍҽӁҽ ҰҸѸҸҊҽґҽ ѸҸґҸҦӃѸ ӁѦҽҰҸґѹҲҽӁҔҽѸ, “ҔҸҲҸҝӃѸ ӁҽҦҸѦґѹ ѹѽ ѸҏңҧҾ”, ѹ ҍҲҸӿ ӊ ҲҽҔѹѽ ҡґҏҍҽҾѽ, ҔҽҔ ҰҝҽӊѹґҸ, ҰҝҸѹҡѽҸҦѹґҸ.

13

ҫ ҡҸҔҝҽӂѦӁѹҾѽ ӊ ҔӊҽҦҝҽҲӁӃѽ ҡҔҸқҔҽѽ ҡѸ. ӁѹңѦ. ѰҔҽӉҽӁӁӃҮ ҰҸҝҾҦҸҔ ҦѦҮҡҲӊѹҮ ӊ ҝҏҡҡҔҸѸ, ҏҔҝҽѹӁҡҔҸѸ ѹ ҰҸґҧҡҔҸѸ ҰѦҝѦӊҸҦҽѽ ҸҲґѹҍҽѦҲҡҾ ҸҲ ҸҝѹҊѹӁҽґҽ. ҫҦӁҽҔҸ ҙҲҸ ӁѦ ѸѦҗҽѦҲ ҰҝҽӊѹґҧӁҸѸҏ ӊҸҡҰҝѹҾҲѹӀ ҸҰѹҡҽӁӁҸҊҸ ҰҸґҸңѦӁѹҾ ҦѦґ. ҈ҦѦҡҧ ӊҽңӁҸ ѹѸѦӁӁҸ ҲҸ, ҍҲҸ Ӊҽ ҔҽңҦӃѸ ҸҲҦѦґҧӁӃѸ ҦѦҮҡҲӊѹѦѸ ҰҸҡҲҸҾӁӁҸ ҡґѦҦҏѦҲ ҦѦҮҡҲӊѹѦ ҡ ҸқҝҽҲӁӃѸ ҙӍӍѦҔҲҸѸ.

14

8

үѩҩѨүѢ ҷүҩѣѢѷҋѣѢ

ғ ҰѦҝѦӊҸҦҽѽ, ҰҝѹӊѦҦѦӁӁӃѽ ӊ ҰҝѹѸѦҝѦ (11), ҡҝҽӊӁѹӊҽѦѸӃѦ ҰҸӉѹ҃ѹѹ ӍҸҝѸ ҏҔҽӉҽӁӃ ӁҸѸѦҝҽѸѹ ӊ ҏҊґҸӊӃѽ ҡҔҸқҔҽѽ. ҐґҽҊҸґҧӁӃѦ ӍҸҝѸӃ ӊ Ӂѹѽ ҸҲѸѦҍѦӁӃ ңѹҝӁӃѸ җҝѹӍҲҸѸ. ҋҸҔҝҽӂѦӁѹҾѸѹ ӊ ҔӊҽҦҝҽҲӁӃѽ ҡҔҸқҔҽѽ ҦҽѦҲҡҾ ҊҝҽѸѸҽҲѹҍѦҡҔҽҾ ѹӁӍҸҝѸҽ҃ѹҾ Ҹ ӁѦҔҸҲҸҝӃѽ ӍҸҝѸҽѽ, ѹҡҰҸґҧӉҸӊҽӁӁӃѽ ӊ ҰѦҝѦӊҸҦҽѽ Ӂҽ ӀңӁҸҡґҽӊҾӁҡҔѹѦ ҾӉӃҔѹ: [pres.] – ҰҝѦӉѦӁҡ; [cond.] – ҔҸӁҦѹ҃ѹҸӁҽґ; [impf.] – ѹѸҰѦҝӍѦҔҲ; [exp.] – “expectative” ӊ ѸҽҔѦҦҸӁҡҔҸѸ ҾӉӃҔѦ (Ҹқ ҙҲҸѸ ҡѸ. ӁѹңѦ). ѰҰҸҲҝѦқґѦӁѹѦ Ҕҏҝҡѹӊҽ ѹ ҰҸҍѦҝҔѹӊҽӁѹҾ ҸқӆҾҡӁѦӁҸ ӁѹңѦ. (11)

Ҵѹѽҽѹґ ҷҏґҊҽҔҸӊ: ԠԂԙԽӾӢ Ӡ ԠԂӢԄԂӢӠԽԂ

ҩҏҡҡҔѹҮ

ҷѦґҸҝҏҡ.

ѰҔҝ.

15

ѩҽ, ҲҽҔ ӊҸҲ-ҡ, ҊҸҡҰҸңҽ ѷҸӍҽӁҽ ӊѽҸҦѹґҽi ӊ ҰҸґҸңѦӁѹѦ 15 ҙҲѹѽ қѦҦӁӃѽ ңѦӁӂѹӁ ѹ ҰҝҸҦҽӊҽґҽi ѹѸ ҔҽҔҏӀ-ҲҸ ӊҸҦҏ ӊ ҰҏӉӃҝҧҔҽѽ. ѭѦӁҽ ӊґѹӊҽґҽi ҙҲҏ ӊҸҦҏ ӊ ҡҏҰ ҡҏҰҝҏҊҏ, ҲҸҲ ѦҊҸ ҡӆѦҦҽґi, қґҽҊҸҦҽҝѹґi Ӊҽ ґҽҡҔҏ ѹ ҍҏӊҡҲӊҸӊҽґi ҡѦқҾ ҰҝѦӊҸҡѽҸҦӁҸ. ҭҝҽӊҦҽ, ҍѦҝѦӉ ӁѦҡҔҸґҧҔҸ ҍҽҡҸӊ ѦѸҏ ӁҽҍѹӁҽґҸi ҸҍѦӁҧ ҡѹґҧӁҸ ѽҸҲѦҲҧҡҾ ҰѹҲҧ, ӉҽҲѦѸ ҸӁ ґҸңѹґҡҾi ӊ ҰҸҡҲѦґҧ, ѹ ҍѦҝѦӉ ҦѦӁҧ ҰҝѦҔҝҽҡӁҽҾ ӁѦҽҰҸґѹҲҽӁҔҽ, ӁҽҔҸҝѸѹӊҗҽҾp ҡӊҸѦҊҸ Ѹҏңҽ ҡҏҰҸѸ, қӃґҽi ҡӊҸқҸҦӁҽ, ҔҽҔ ӊѦҡѦӁӁѹҮ ӊѦҲѦҝ. ѩӃҔ ӊҸҡҧ, ҰҽӁҟ ѷҽӍҽӁҽ ҝҽӉҏѸѦґҽi ҊҙҲӃѽ қѦҦӁӃѽ ңҽӁҍӃӁ ҟ ҰҝҽҦҽӊҽґҽi ҟѸ ӁѦҮҔҏӀ ӊҽҦҏ ҕ қҏҲҙґѦҍҔҽѽ. ѭҸӁҔҽ ӊӃґҟӊҽґҽi ҊҙҲҏӀ ӊҽҦҏ ҕ Ѹҏңҽӊҏ ҡҲҝҽӊҏ, ҲҸҮ Ӊ'ҾҦҽҕi, (2ӊ) ҦӉҾҔҽӊҽҕi Ӊҽ ґҽҡҔҏ ҟ ҽҦҍҏӊҽҕi ҡҾқѦ ҃ҏҦҸҕӁҽ. ҭҝҽҕҦҽ, ҰҝҽӉ ӁѦҔҽґҧҔҟ ҊҽҦӉҟӁ ҾѸҏ ҰҽҍӃӁҽґҽi ҡҲҝҽҗҙӁӁҽ ѽҽ҃Ѧ҃҃ҽ Ұҟ҃ҧ, ҰҸҲӃѸ ҺӁ ҔґҽҕҡҾi ҕ ґҸңҽҔ, ҽ ҰҝҽӉ ҦӉѦӁҧ ҰҝӃҊҸңҽҾ ӁѦҽҰҽґҟҲҽӁҔҽ, ҾҔҽҾ ӁҽҔҽҝѸҟґҽp ҡӊҽҮҊҸ Ѹҏңҽ, қӃґҽi ӊҸґҧӁҽҾ, ҾҔ ӊҾҡӁҸӊӃ ӊѦ҃Ѧҝ. ѷҽҔ Ҹҡҧ, ҃Ҿ ҰҽӁҟ ѷҸӍҽӁҽ, ҡҰҟӊҍҏӊҽӀҍѹi ҃ѹѸ ңҟӁҔҽѸ ҏ Ѷѽ ҊҟҝҔҟҮ ҦҸґҟ, ҰҝҸҦҽӊҽґҽi ѶѸ ҾҔҏҡҧ ӊҸҦѹҍҔҏ ӊ ҡґҸѶҔҽѽ. ѩҝҏңѹӁҽ ӊѹґѹӊҽґҽi Ҳҏ ӊҸҦѹҍҔҏ ѸҏңѦӊҟ ӊ ҡҏҰ, ҲҸҮ ӊѹѶҦҽӊi ҮҸҊҸ, ҦҾҔҏӊҽӊi Ӊҽ ґҽҡҔҏ ҟ ѸҽӊҡҾi ҰҝѦҍҏҦҸӊҸ. ҭҝҽӊҦҽ, ҰҸ ҔҟґҧҔҸѽ ҊҸҦѹӁҽѽ ҮҸѸҏ ҡҲҝҽѽ ҾҔ ҰҝҽҊґҸҡҾi ҰѹҲѹ, ҰҸҲҟѸ ӊҟӁ ҔґҽӊҡҾi ӊ ґҟңҔҸ, ҽ ҍѦҝѦӉ ҦѦӁҧ ҰҝѦҔҝҽҡӁҽ ӁѦҽҰҸґҟҲҽӁҔҽ, ӂҸ ӁҽҊҸҦҏӊҽґҽp ҡӊҸҊҸ ҍҸґҸӊҟҍѦӁҧҔҽ ҡҏҰҸѸ, ҡҲҽӊҽґҽi ӊҟґҧӁҽ, ҾҔ ӊѦҡӁҾӁѹҮ ґѦҊҟҲ.

ғ ҙҲҸѸ ҸҲҝӃӊҔѦ ҡҸҍѦҲҽӁѹѦ ӴԖӹԌӠԊԂ Ӵ ԗӹԊӹԞӾӰӠӾ ҏҔҽӉӃӊҽѦҲ Ӂҽ ҰҸӊҲҸҝҾӀӂҏӀҡҾ ҡѸѦӁҏ ҡѹҲҏҽ҃ѹҮ, ҽ ѹѸѦӁӁҸ Ӂҽ ҰҸӊҲҸҝҾӀӂѦѦҡҾ ӊҸӉӁѹҔӁҸӊѦӁѹѦ ҰҸӁѹѸҽӁѹҾ ѹ ҡҸҍҏӊҡҲӊѹҾ. ҫҦӁҽҔҸ ӊ ҝҾҦѦ ҰѦҝѦӊҸҦҸӊ ӉҦѦҡҧ ѹҡҰҸґҧӉҏѦҲҡҾ ҦҝҏҊҸҮ ґѦҔҡѹҍѦҡҔѹҮ ѸҽҲѦҝѹҽґ ѹ ҏҔҽӉӃӊҽѦҲҡҾ ґѹҗҧ Ӂҽ ҡҏӂѦҡҲӊҸӊҽӁѹѦ ҦҽӁӁҸҊҸ ҡҸҡҲҸҾӁѹҾ. ғ ҲҽҔѹѽ ҡґҏҍҽҾѽ ҡѸӃҡґ ҰҸӊҲҸҝҾѦѸҸҡҲѹ ҸҡґҽқґѦӁ ѹґѹ ҸҲҡҏҲҡҲӊҏѦҲ. ѩґҾ ҃ѦґѦҮ ӁҽҗѦҊҸ ѹҡҡґѦҦҸӊҽӁѹҾ ҰҸҦҸқӁӃѦ ҰҝѹѸѦҝӃ ӁѦ ҰҝѦҦҡҲҽӊґҾӀҲ ѹӁҲѦҝѦҡҽ.

ҭҫҋһѣѩҫғүѷѣһѻѢҫҋѷѻ ѩѣҢҋѷғѨҢ ҭҩѨ ҭҫғѷҫҩѤѣҴҫҋѷѨ

ҭҸґҧҡҔ.

9

Tak wiNjc signora Tofana wczuwa’ai siNj w sytuacjNj tych biednych kobiet i sprzedawa’ai im jakowƾǧ wodNj we flaszeczkach. ƺona wlewa’ai tNj wodNj mNjǻowi do zupy, mƾǻ to spoǻywa’i, piNjknie dziNjkowa’i i czu’i siNj znakomicie. Co prawda po paru godzinach zaczyna’i mieǀ ogromne pragnienie, potem k’ad’i siNj do ’óǻka i nie mija’ dzieǝ, a piNjkna neapolitanka która poda’ap swemu mNjǻowi tak znakomitƾ zupNj, by’ai juǻ wolna jak wiosenny wiatr. ѯѦҗҡҔ. Tak tedy signora Toffanová chápalai tyhle ubožaǁky a prodávalai jim jakousi záhadnou vodiǁku. Žena ji nalilap manželovi do polévky, ten zbaštilp polévku, podLJkovalp za péǁi a cítili se v sedmém nebi. Pravda, za pár hodin ho popadlap hrozná žízeǞ, ulehlp do postele a za dva dny pDzvabná Neapolitánka, která podalap muži takovou polévku, bylai volná jako ptáǁe. ҋґҸӊҽ҃Ҕ. Nuž tak, prosím, signora Toffana sa vedelai vžiǬp do situácie nešǬastných žien a predávalai im akúsi vodiǁku vo fǚaštiǁkách. Žena nalialap vodiǁku mužovi do polievky, ten ju zjedolp, poDŽakovalp sa jej za dobrotu a cítili sa znamenite. Pravda, o pár hodín ho pochytilp hrozný smäd, potom si ǚaholp do postele, a na druhý deǞ krásna Neapolǁanka, ktorá nachovalap muža takou polievoǁkou, bolai voǚná ako jarný vánok. ҋґҸӊѦӁҡҔ. No, torej, gospa Tofana je imelai razumevanje za položaj teh ubogih žen in jim je prodajalai neko vodo v stekleniǁkah. Žena je nalilap to vodo možu v juho, ta jo (2қ) je pojedelp, se zahvalilp za ljubeznivost in imeli imeniten obǁutek. Res pa je, ǁez nekaj ur se ga je polotilap strašanska žeja, potem je legelp v posteljo, ǁez en dan pa je bilai prelepa Neapeljǁanka, ki je nahranilap svojega moža z juho, svobodna kakor pomladni veter. ҤҸҝӊҽҲҡҔ. Da, dakle, gospoda Tofana razumjelai/p bi [cond.] položaj tih jadnih žena i prodavalai im neku vodu u boǁicama. Žena bi [cond.] ulilap tu vodu suprugu u juhu, on bi [cond.] je pojeop, zahvaliop na brižljivosti i divno se osjeǀaoi. Istina, za nekoliko sati on bi [cond.] bioi jako žedan, zatim bi [cond.] legaop u postelju, i za jedan dan prekrasna Napuljka koja je svojeg muža nahranilap juhom, bilai bi [cond.] slobodna kao proljetni vjetar.

10 ҋѦҝқҡҔѹҮ

ҴҽҔѦҦҸӁ.

ҷҸґҊҽҝҡҔ.

үѩҩѨүѢ ҷүҩѣѢѷҋѣѢ

ѩҽ, ѹ ҲҽҔҸ ҡѦ, ѦҲҸ, ҊҸҡҰҸ҉ҽ ѷҸӍҽӁҽ ҏӁҸҡѹґҽi ҏ ҰҸґҸңҽҎѦ Ҳѹѽ ҎҽҦӁѹѽ ңѦӁҽ ѹ ҰҝҸҦҽӊҽґҽi ѹѸ ҎѦ ӁѦҔҽҔӊҏ ӊҸҦѹ҃ҏ ҏ Ӎґҽҗѹ҃ҽѸҽ. ѭѦӁҽ қѹ [cond.] ҡѹҰҽґҽi Ҳҏ ӊҸҦѹ҃ҏ ҡҏҰҝҏҊҏ ҏ ҡҏҰҏ, ҸӊҽҎ ԼӠ [cond.] ҡҏҰҏ ԗӹԦӾӹp, ԒԂԖӴԂԊӠӹp ԼӠ [cond.] Ӂҽ ҝҏҍҔҏ ѹ ҸҦґѹҍӁҸ ҡѦ ҸҡѦѴҽҸi. ѨҡҲѹӁҽ, ӁѦҔҸґѹҔҸ ҍҽҡҸӊҽ ҔҽҡӁѹҎѦ ԼӠ [cond.] ԗӹԓӾӹp Ҧҽ ҸҡѦѴҽ ҏңҽҡӁҏ ңѦ҉, ҰҸҡґѦ ҍѦҊҽ қѹ [cond.] ѸҸҝҽҸi Ҧҽ ґѦҊӁѦp ҏ ҰҸҡҲѦҒҏ, ҽ ҦҽӁ ҔҽҡӁѹҎѦ қѹ [cond.] ҰҝѦґѦҰҽ ѢҽҰҸґѹҲҽӁҔҽ, ҔҸҎҽ ҎѦ ӁҽѽҝҽӁѹґҽp ҡӊҸҊ ҡҏҰҝҏҊҽ ҡҏҰҸѸ, қѹґҽi ҡґҸқҸҦӁҽ ҔҽҸ ҰҝҸґѦѴӁѹ ӊѦҲҽҝ. ѩҽ, ѹ ҲҽҔҽ, Ҳҽҽ ҊҸҡҰҸҖҽ ѷҸӍҽӁҽ ѹѸ ҡҸҍҏӊҡҲӊҏӊҽҗѦi [impf.] Ӂҽ ҲѹѦ қѦҦӁѹ ңѦӁѹ ѹ ѹѸ ҰҝҸҦҽӊҽҗѦi [impf.] ӁѦҔҽҔӊҽ ӊҸҦѹҍҔҽ ӊҸ җѹҗѦӁ҃ҽ. ѭѦӁҽҲҽ ԀӾ [exp.] Ҏҽ ӰԂԊӾӾԶӾp [impf.] Ҳҽҽ ӊҸҦѹҍҔҽ ӊҸ ҡҏҰҽҲҽ Ӂҽ ѸҽңҸҲ, ҲҸҎ ԀӾ [exp.] Ҏҽ ӠԒӾԌӾԶӾp [impf.], ѱѦ [exp.] ѹӿ қѦҗѦi [impf.] қґҽҊҸҦҽҝѦӁ Ӊҽ ӊӁѹѸҽӁѹѦҲҸ ѹ ѱѦ [exp.] ҡѦ ҍҏӊҡҲӊҏӊҽҗѦi [impf.] ҰҝѦҔҝҽҡӁҸ. ѢҸ, Ӊҽ ңҽґ, ҰҸ ӁѦҔҸґҔҏ ҍҽҡҽ ҲҸҎ ԀӾ [exp.] ԗӹԓӰӾԶӾp [impf.] Ҧҽ ҍҏӊҡҲӊҏӊҽ ҊҸґѦѸҽ ңѦҦ, ҰҸҲҸҽ ԀӾ [exp.] ԊӾԄӰӾԶӾp [impf.] ӊ ҰҸҡҲѦґҽ, ѹ ҰҸ ѦҦѦӁ ҦѦӁ ҏқҽӊҽҲҽ ӁѦҽҰҸґѹҲҽӁҔҽ, җҲҸ Ѹҏ Ҏҽ Ҧҽґҽp ҡҏҰҽҲҽ Ӂҽ ҡӊҸҎҸҲ Ѹҽң, ѱѦ [exp.] қѦҗѦi [impf.] ҡґҸқҸҦӁҽ ҔҽҔҸ ҰҝҸґѦҲѦӁ ӊѦҲѦҝ. ѢҦҽ, Ҳҽ ҊҸҡҰҸңҽ ѷҸӍҽӁҽ ӊґѹӉҽҗѦi [impf.] ӊ ҰҸґҸңѦӁѹѦҲҸ Ӂҽ ҊҸҝҔѹҲѦ ңѦӁѹ ѹ ѹѸ ҰҝҸҦҽӊҽҗѦi [impf.] ҡҲӆҔґѦӁѹҍҔѹ ҡ ӁҾҔҽҔӊҽ ӊҸҦҽ. ѭѦӁҽҲҽ ѹӉҡѹҰӊҽi [pres.] ӊҸҦҽҲҽ ӊ ҡҏҰҽҲҽ Ӂҽ ҡӆҰҝҏҊҽ, ҲҸҮ Ҿ ѹӉҡӆҝқӊҽi [pres.], қґҽҊҸҦҽҝѹi [pres.] Ӊҽ ҊҝѹңѹҲѦ ѹ ҡѦ ҍҏӊҡҲӊҽi [pres.] ҰҝѦӊӆӉѽҸҦӁҸ. ѩҽ, ӁҸ ҡґѦҦ ӁҾҔҸґҔҸ ҍҽҡҽ ҏҡѦӂҽi [pres.] ҡѹґӁҽ ңҽңҦҽ, ҰҸҡґѦ ґҾҊҽi [pres.] қҸґѦӁ ѹ ҡґѦҦ ѦҦѹӁ ҦѦӁ ҰҝѦҔҝҽҡӁҽҲҽ ӁѦҽҰҸґѹҲҽӁҔҽ, ӁҽҊҸҡҲѹґҽp ҡӆҰҝҏҊҽ ҡѹ ҡӆҡ ҡҏҰҽ, ҡҲҽӊҽi [pres.] ҡӊҸқҸҦӁҽ ҔҽҲҸ ҰҝҸґѦҲӁѹҾ ӊҾҲӆҝ.

ҋҸӊѦҝҗѦӁӁҸ ҾҡӁҸ, ҍҲҸ ѹ ӊ ҙҲҸѸ ҰҝѹѸѦҝѦ ӊҡѦ ҊґҽҊҸґҧӁӃѦ ӍҸҝѸӃ, ӊӃҦѦґѦӁӁӃѦ ңѹҝӁӃѸ җҝѹӍҲҸѸ, ӊӃҝҽңҽӀҲ ӁѦҸҊҝҽӁѹҍѦӁӁҸ-ҔҝҽҲӁҸѦ ӉӁҽҍѦӁѹѦ. ғѦҦҧ ӊ ӁѦѸ ҸҰѹҡӃӊҽѦҲҡҾ ӁѦ ҸҦѹӁ ҔҸӁҔҝѦҲӁӃҮ ҡґҏҍҽҮ, ҽ ӁѦҔҸҲҸҝҸѦ ҡҸҍѦҲҽӁѹѦ ҡҸқӃҲѹҮ, ҰҸӊҲҸҝҾӀӂѦѦҡҾ ӁѦѹӉӊѦҡҲӁҸѦ ҔҸґѹҍѦҡҲӊҸ ҝҽӉ. ѷѦѸ ӁѦ ѸѦӁѦѦ, ӊ җѦҡҲѹ ѹӉ ҸҦѹӁӁҽҦ҃ҽҲѹ ҰҝѹӊѦҦѦӁӁӃѽ ҡґҽӊҾӁҡҔѹѽ ҾӉӃҔҸӊ ҸҰҝѦҦѦґѦӁӁӃѦ ҊґҽҊҸґӃ ѹѸѦӀҲ ӍҸҝѸҏ ҋғ. (ѩґҾ ӁҽҊґҾҦӁҸҡҲѹ ҙҲѹ ӍҸҝѸӃ ӊ ҰҝѹѸѦҝҽѽ ҦҸҰҸґӁѹҲѦґҧӁҸ ӊӃҦѦґѦӁӃ ҔҏҝҡѹӊҸѸ. ) ѢҽѹқҸґѦѦ ҡѹґҧӁҸ ѹ ҰҸҡґѦҦҸӊҽҲѦґҧӁҸ ҋғ ҰҝѦҦҡҲҽӊґѦӁ ӊ ҍѦҗҡҔҸѸ, ҡґҸӊҽ҃ҔҸѸ ѹ ҡґҸӊѦӁҡҔҸѸ ҲѦҔҡҲҽѽ. ғ Ӂѹѽ ҙҲҸҲ ӊѹҦ ӊҡҲҝѦҍҽѦҲҡҾ ҰҸ 6 ҝҽӉ, ҰҝѹҲҸѸ ӊ ҸҦӁѹѽ ѹ ҲѦѽ ңѦ ҰҸӉѹ҃ѹҾѽ – – ѹ –. ғҽңӁҸ ҲҽҔңѦ ҸҲѸѦҲѹҲҧ, ҍҲҸ ӊҡѦ ҙҲѹ ӍҸҝѸӃ ҸҲӁҸҡҾҲҡҾ Ҕ ҸқӃҍӁӃѸ ӍҸҝѸҽѸ ҰҝҸҗѦҦҗѦҊҸ

ҭҫҋһѣѩҫғүѷѣһѻѢҫҋѷѻ ѩѣҢҋѷғѨҢ ҭҩѨ ҭҫғѷҫҩѤѣҴҫҋѷѨ

11

ӊҝѦѸѦӁѹ (Ҳ.Ѧ. ӉҦѦҡҧ ӁѹҔҽҔѹѽ ҸҡҸқӃѽ ѸҸҦҽґҧӁӃѽ ҏҔҽӉҽҲѦґѦҮ ӁѦ ӁҽқґӀҦҽѦҲҡҾ). ғ ѽҸҝӊҽҲҡҔҸѸ, ҡѦҝқҡҔҸѸ ѹ ѸҽҔѦҦҸӁҡҔҸѸ ҲѦҔҡҲҽѽ ӍҸҝѸӃ ҋғ ҰҝѦҦҡҲҽӊґѦӁӃ ӁѦѸӁҸҊҸ ѸѦӁҧҗѦ, ӁҸ ҲҽҔңѦ ӊ ҰҝѦҦѦґҽѽ ҏҔҽӉҽӁӁӃѽ ҰҸӉѹ҃ѹҮ. 16 ѯҝѦӉӊӃҍҽҮӁҸ ӊҽңӁҸ ҸҲѸѦҲѹҲҧ, ҍҲҸ ӊ ҙҲѹѽ ҾӉӃҔҽѽ ҰҝѦҦҡҲҽӊґѦӁӁӃѦ ӍҸҝѸӃ ҸҲӁҸҡҾҲҡҾ Ҕ ѸҸҦҽґҧӁҸ ѸҽҝҔѹҝҸӊҽӁӁӃѸ. ғ ѽҸҝӊҽҲҡҔҸѸ ѹ ҡѦҝқҡҔҸѸ ҲѦҔҡҲҽѽ ӁҽѽҸҦѹѸ ҽӁҽґѹҲѹҍѦҡҔҏӀ ӍҸҝѸҏ ҔҸӁҦѹ҃ѹҸӁҽґҽ, ҡҸҡҲҸҾӂҏӀ ѹӉ ҰҝѹҍҽҡҲѹҾ Ӂҽ -Ԋ ѹ ҸҡҸқҸҮ ӍҸҝѸӃ ӊҡҰҸѸҸҊҽҲѦґҧӁҸҊҸ ҊґҽҊҸґҽ ԼӠԽӠ. ҨҸҝѸҽ ҔҸӁҦѹ҃ѹҸӁҽґҽ, ҸқҸӉӁҽҍҽѦѸҽҾ ӊ ҰҝѹѸѦҝҽѽ ҡҸҔҝҽӂѦӁѹѦѸ [cond.] Ұҝѹ ӊҡҰҸѸҸҊҽҲѦґҧӁҸѸ ҊґҽҊҸґѦ, ӊ ҦҽӁӁӃѽ ҾӉӃҔҽѽ ҝѦҊҏґҾҝӁҸ ҏҰҸҲҝѦқґҾѦҲҡҾ ҦґҾ ҸқҸӉӁҽҍѦӁѹҾ ҰҸӊҲҸҝҾӀӂѹѽҡҾ ҡѹҲҏҽ҃ѹҮ ӊ ҰҝҸҗґҸѸ. 17 ҘҽҔ ҰҸҔҽӉӃӊҽѦҲ ӊӃҗѦҰҝѹӊѦҦѦӁӁӃҮ ҰҝѹѸѦҝ, Ұҝѹ ҙҲҸѸ ѸҸҊҏҲ ӊӃҡҲҏҰҽҲҧ Ҹқҽ ӊѹҦҽ. ғ ѸҽҔѦҦҸӁҡҔҸѸ ҲѦҔҡҲѦ ҏҰҸҲҝѦқґҾѦҲҡҾ ҸҡҸқҸѦ ҡҸҍѦҲҽӁѹѦ ӍҸҝѸӃ ѹѸҰѦҝӍѦҔҲҽ ҡ ҍҽҡҲѹ҃ѦҮ ԀӾ, ӊҸҡѽҸҦҾӂѦҮ Ҕ ӊҡҰҸѸҸҊҽҲѦґҧӁҸѸҏ ҊґҽҊҸґҏ ҲѹҰҽ ԖӹԽӾԽӸ. ҋҸҍѦҲҽӁѹѦ ҙҲҸҮ ҍҽҡҲѹ҃Ӄ ҡ ӍҸҝѸҸҮ ҰҝѦӉѦӁҡҽ (ҋғ ѹ Ѣҋғ) ҰҝѹӁҾҲҸ ӁҽӉӃӊҽҲҧ ӍҸҝѸҸҮ қҏҦҏӂѦҊҸ ӊҝѦѸѦӁѹ, ӁҸ ҸқӂѦҰҝѹӁҾҲҸҊҸ ҸқҸӉӁҽҍѦӁѹҾ ҦґҾ ҡҸҍѦҲҽӁѹҾ ԀӾ + ѹѸҰѦҝӍѦҔҲ, ҰҸ-ӊѹҦѹѸҸѸҏ, ҰҸҔҽ ӁѦҲ. ѩҽӁӁҸѦ ҡҸҍѦҲҽӁѹѦ ѹѸѦѦҲ ҦҸӊҸґҧӁҸ җѹҝҸҔѹҮ ҔҝҏҊ ӍҏӁҔ҃ѹҮ, ѹӉ ҔҸҲҸҝӃѽ ӁѦҔҸҲҸҝӃѦ ӊҡҲҝѦҍҽӀҲҡҾ ҲҽҔңѦ ҏ ҡѦҝқҡҔҸҊҸ/ѽҸҝӊҽҲҡҔҸҊҸ ҔҸӁҦѹ҃ѹҸӁҽґҽ ѹ ҝҏҡҡҔҸҊҸ ҡҸҡґҽҊҽҲѦґҧӁҸҊҸ ӁҽҔґҸӁѦӁѹҾ (ҡҝ. ҰҝѹѸ. 17). ғ ҰҝѹѸѦҝҽѽ, ҰҝѹӊѦҦѦӁӁӃѽ ӊ ӁҽҡҲҸҾӂѦҮ ҡҲҽҲҧѦ, ҸҡӁҸӊӁҸҮ ҊґҽҊҸґҧӁӃҮ ҔҸѸҰҸӁѦӁҲ, ҡҸҍѦҲҽӀӂѹҮҡҾ ҡ ҍҽҡҲѹ҃ѦҮ ԀӾ, ҸқҸӉӁҽҍѦӁ ҡҸҔҝҽӂѦӁѹѦѸ [impf.] (ѹѸҰѦҝӍѦҔҲ), ҽ ҡҽѸҽ ҍҽҡҲѹ҃ҽ – [exp.] (ҡҝ. ҲѦҝѸѹӁ “expectative” ӊ ҸҍѦҝҔѦ Ҹ ѸҽҔѦҦҸӁҡҔҸѸ ҾӉӃҔѦ ғ. ҨҝѹҦѸҽӁҽ (Friedman 1993: 269). 18 ҭҝѹѸѦҝ (11) ҰҸҔҽӉӃӊҽѦҲ, ҍҲҸ ҦҽӁӁҽҾ ҽӁҽґѹҲѹҍѦҡҔҽҾ ӍҸҝѸҽ, ҰҸҦҸқӁҸ 16

ғӃқҸҝ Ѣҋғ ӊ ӁѦҔҸҲҸҝӃѽ ѹӉ ҙҲѹѽ ҰҸӉѹ҃ѹҮ ҊґҽӊӁӃѸ ҸқҝҽӉҸѸ ҡӊҾӉҽӁ ҡ ґѦҔҡѹҍѦҡҔѹѸѹ ӍҽҔҲҸҝҽѸѹ. ғ ҰҸҦҸқӁӃѦ ҦѦҲҽґѹ ѸӃ ӉҦѦҡҧ ӊѽҸҦѹҲҧ ӁѦ қҏҦѦѸ. 17 ҋґѦҦҏѦҲ ҰҸҦҍѦҝҔӁҏҲҧ, ҍҲҸ ҸқҸӉӁҽҍѦӁѹѦ ҸҰҝѦҦѦґѦӁӁҸҊҸ ҲѹҰҽ ҰҸӊҲҸҝҾѦѸҸҡҲѹ ҾӊґҾѦҲҡҾ ґѹҗҧ ҸҦӁҸҮ ѹӉ ӍҏӁҔ҃ѹҮ ҔҸӁҦѹ҃ѹҸӁҽґҽ ӊ ҡѦҝқҡҔҸѸ ѹ ѽҸҝӊҽҲҡҔҸѸ ҾӉӃҔҽѽ. ғ қҸґҧҗѹӁҡҲӊѦ ҦҝҏҊѹѽ ӍҏӁҔ҃ѹҮ ҦҽӁӁҽҾ ӍҸҝѸҽ ҾӊґҾѦҲҡҾ ҦҸӊҸґҧӁҸ қґѹӉҔѹѸ ҙҔӊѹӊҽґѦӁҲҸѸ ҝҏҡҡҔҸҊҸ ҡҸҡґҽҊҽҲѦґҧӁҸҊҸ ӁҽҔґҸӁѦӁѹҾ. 18 ҤҸҲҾ ѸҽҔѦҦҸӁҡҔҽҾ ҍҽҡҲѹ҃ҽ ԀӾ ӁѦҡҸѸӁѦӁӁҸ ҝҸҦҡҲӊѦӁӁҽ қҸґҊҽҝҡҔҸҮ ҍҽҡҲѹ҃Ѧ ԷӾ, ҔҸҲҸҝҽҾ ҡѽҸҦӁӃѸ ҸқҝҽӉҸѸ ҏҰҸҲҝѦқґҾѦҲҡҾ ҦґҾ ҸқҝҽӉҸӊҽӁѹҾ ӍҸҝѸ қҏҦҏӂѦҊҸ ӊҝѦѸѦӁѹ, ҙҲѹ ҍҽҡҲѹ҃Ӄ ӉӁҽҍѹҲѦґҧӁҸ ҸҲґѹҍҽӀҲҡҾ ҰҸ ҦҽґҧӁѦҮҗѦѸҏ ҝҽҡҰҝѦҦѦґѦӁѹӀ ѹ ҏҰҸҲҝѦқґѦӁѹӀ. ҋ ҏҔҽӉҽӁӁҸҮ ѸҽҔѦҦҸӁҡҔҸҮ ҔҸӁҡҲҝҏҔ҃ѹѦҮ ԀӾ + ѹѸҰѦҝӍѦҔҲ ӊ ҰҝѹӁ҃ѹҰѦ ѸҸңӁҸ ҡҝҽӊӁѹҲҧ қҸґҊҽҝҡҔҏӀ ҔҸӁҡҲҝҏҔ҃ѹӀ, ҡҸҡҲҸҾӂҏӀ ѹӉ ѹѸҰѦҝӍѦҔҲҽ ӊҡҰҸѸҸҊҽҲѦґҧӁҸҊҸ ҊґҽҊҸґҽ ԷԂ + ԌԂ + ҰҝѦӉѦӁҡ (Belyavski-Frank 2003: 4). ѢҸ, ӁҽҡҔҸґҧҔҸ ӁҽѸ ѹӉӊѦҡҲӁҸ, ҰҸҡґѦҦӁҾҾ ҔҸӁҡҲҝҏҔ҃ѹҾ ӁѦ ӊӃҝҽңҽѦҲ ҰҸӊҲҸҝҾѦѸҸҡҲѹ. ҭҸ-ӊѹҦѹѸҸѸҏ, ҦҽӁӁӃѦ ҔҸӁҡҲҝҏҔ҃ѹѹ ӊ ҡҸӊҝѦѸѦӁӁӃѽ ҾӉӃҔҽѽ ҡѽҸҦҾҲҡҾ ҲҸґҧҔҸ ӊ ӍҏӁҔ҃ѹѹ ӊӃҝҽңѦӁѹҾ ӁѦҔҸҲҸҝҸҮ ҝҽӉӁҸӊѹҦӁҸҡҲѹ ѹҝҝѦҽґҧӁҸҡҲѹ. ҋҝ. ҡґѦҦҏӀӂѹҮ ҰҝѹѸѦҝ ѹӉ ӁҽҗѦҊҸ ѸҽҲѦҝѹҽґҽ (ѹӉ ҰѦҝѦӊҸҦҸӊ ҔӁѹҊѹ Ҹ Ҥҽҝҝѹ ҭҸҲҲѦҝѦ): үӁҊґ. If brains were gold you’d be poorer than Weasley; ѸҽҔ. ԟԂ ԼӾԶӾ ԐӹԒӹӪӹԽ ԒԊԂԽӹ, ԀӾ ԼӾԶӾ ԗӹԙӠӢӹԐԂԶӾӰ ӹԌ ԛӠԒԊӠ; қҸґҊ. ԣӪӹ ӫԐԔԽ ԼӾԶӾ ԒԊԂԽӹ, ԽӠ ԷӾԶӾ ԌԂ ԙӠ ԗӹ-ԼӾԌӾӰ Ӡ ӹԽ ԙӾԐӾӡԙԽӴӹ ӯӠԒԊӠ; ҝҏҡ. ӺԙԊӠ ԼՂ ԐӹԒԄӠ ԼՂԊӠ ӠԒ ԒӹԊӹԽԂ, ԽՂ ԼՂ ӴԙӾ ӢԂӴӰӹ ԼՂԊ ԼӾԌӰӾӾ ӯӠԒԊӠ.

12

үѩҩѨүѢ ҷүҩѣѢѷҋѣѢ

ҡѦҝқҡҔҸѸҏ/ѽҸҝӊҽҲҡҔҸѸҏ ҔҸӁҦѹ҃ѹҸӁҽґҏ, ӊ ҡґҏҍҽҾѽ ӁѦҸҊҝҽӁѹҍѦӁӁҸ-ҔҝҽҲӁҸҊҸ ӉӁҽҍѦӁѹҾ ӊӃҡҲҏҰҽѦҲ ҔҽҔ ӊ Ѣҋғ, ҲҽҔ ѹ ӊ ҋғ. 19 ҷҸґҊҽҝҡҔѹҮ ҰѦҝѦӊҸҦ ҦҽӁӁҸҊҸ ӍҝҽҊѸѦӁҲҽ ҦҸӊҸґҧӁҸ ҡѹґҧӁҸ ҸҲґѹҍҽѦҲҡҾ ҸҲ ҰѦҝѦӊҸҦҸӊ Ӂҽ ҦҝҏҊѹѦ ҾӉӃҔѹ. ҭҸҡґѦ ҸҲҝӃӊҔҽ ҰҸҦ ӁҸѸѦҝҸѸ ҰҝҸѹҡѽҸҦѹҲ ҰѦҝѦҔґӀҍѦӁѹѦ Ӂҽ ӍҸҝѸӃ ҰҝѦӉѦӁҡҽ Ѣҋғ. ѢҽѸ ҰҸҔҽ ҲҝҏҦӁҸ ҡҏҦѹҲҧ Ҹ ҲҸѸ, ӁҽҡҔҸґҧҔҸ ҙҲҸ ҲѹҰѹҍӁҸ ҦґҾ қҸґҊҽҝҡҔҸҊҸ ҾӉӃҔҽ. ҫҦӁҽҔҸ Ӂҽ ҸҡӁҸӊѦ ҲҸҊҸ, ҍҲҸ ӊ ӁѦѸ ӁҽѹқҸґѦѦ ҝҽҡҰҝҸҡҲҝҽӁѦӁӁӃѸѹ ҦґҾ ӊӃҝҽңѦӁѹҾ ӁѦҸҊҝҽӁѹҍѦӁӁҸҮ ҔҝҽҲӁҸҡҲѹ ӊ ҰҝҸҗґҸѸ ҾӊґҾӀҲҡҾ ӍҸҝѸӃ ѹѸҰѦҝӍѦҔҲҽ Ѣҋғ, ѸӃ ҰҝѦҦҰҸґҽҊҽѦѸ, ҍҲҸ ӍҸҝѸӃ ҰҝѦӉѦӁҡҽ ӊ ҦҽӁӁҸѸ ҸҲҝӃӊҔѦ ҡҲѹґѹҡҲѹҍѦҡҔѹ ѸѦӁѦѦ ӁѦҮҲҝҽґҧӁӃ. ғ ҰҝѹӊѦҦѦӁӁӃѽ ӍҝҽҊѸѦӁҲҽѽ ӍҸҝѸӃ ҋғ ӊҡҲҝѦҍҽӀҲҡҾ ӊ ҸҰҝѦҦѦґѦӁӁӃѽ ҰҸӉѹ҃ѹҾѽ: ҸқҸӉӁҽҍҽѦѸӃѦ ѹѸѹ ҃ѦґҸҡҲӁӃѦ ҰҝѦҦѦґҧӁӃѦ ҦѦҮҡҲӊѹҾ ӊҸҡҰҝѹӁѹѸҽӀҲҡҾ ҔҽҔ ӉӊѦӁҧҾ ӁѦҔҸҲҸҝҸҮ ҃ѦҰѹ ҡҸқӃҲѹҮ. ҭҸӊҲҸҝҾѦѸҸҡҲҧ ҸҲӁҸҡѹҲҡҾ ҔҸ ӊҡѦҮ ҙҲҸҮ ҃ѦҰѹ ӊ ҃ѦґҸѸ. 20 ғ ҾӉӃҔҽѽ, ҦҸҰҏҡҔҽӀӂѹѽ ӊ ҲҽҔѹѽ ҡґҏҍҽҾѽ ҋғ, ӊӁѹѸҽӁѹѦ ӁҽҰҝҽӊґѦӁҸ ҰҝѦңҦѦ ӊҡѦҊҸ Ӂҽ ҸҲӁҸҗѦӁѹҾ ӊӁҏҲҝѹ ҦҽӁӁҸҮ ҃ѦҰѹ, Ӂҽ ҲҸ, ҍҲҸ ҔҽңҦҽҾ ҸҲѸѦҍѦӁӁҽҾ ҡѸѦӁҽ ҡѹҲҏҽ҃ѹҮ ҍѦҝѦҦҏѦҲҡҾ ҡ ҦҝҏҊҸҮ ҡѸѦӁҸҮ (ѹґѹ ҰҝѹҔҝѦҰґҾѦҲҡҾ Ҕ ҔҽҔҸҮ-ҲҸ ҦҝҏҊҸҮ ҡѹҲҏҽ҃ѹѹ). ғ ҙҲҸѸ ҰґҽӁѦ ҸҲҝӃӊҔѹ ҰҸҦ ӁҸѸѦҝҸѸ 2 (– ѹ –) ӉҽѸѦҲӁҸ ҸҲґѹҍҽӀҲҡҾ ҸҲ ҸҲҝӃӊҔҽ , ӁѦҡѸҸҲҝҾ Ӂҽ ҲҸ, ҍҲҸ ѹ ҲҽѸ ҰҝѹҡҏҲҡҲӊҏѦҲ ҲѹҰѹҍӁӃҮ ҰҝѦҦѦґҧӁӃҮ ҊґҽҊҸґ, ӊѽҸҦҾӂѹҮ ӊ ҸқҝҽӉ҃ҸӊҏӀ ӊѹҦҸӊҏӀ Ұҽҝҏ (ԗӢӹԌԂԽӸp – ԗӢӹԌԂӴԂԽӸi). ҤҸҲҾ ѹӉ ҔҸӁҲѦҔҡҲҽ ҦҸҡҲҽҲҸҍӁҸ ҾҡӁҸ ӊӃҲѦҔҽѦҲ, ҍҲҸ ѹ ӉҦѦҡҧ ҸҲҦѦґҧӁҸ ӊӉҾҲӃѦ ҰҝҸҾӊґѦӁѹҾ ҦѦҮҡҲӊѹҾ ҾӊґҾӀҲҡҾ ҃ѦґҸҡҲӁӃѸѹ, ӉҽҔҸӁҍѦӁӁӃѸѹ, ӊҸ ӊҡѦѽ ҰѦҝѦӊҸҦҽѽ, ҸҦӁҽҔҸ, ҏҰҸҲҝѦқґѦӁҽ ӊ ҙҲҸѸ ѸѦҡҲѦ ӍҸҝѸҽ Ѣҋғ. ҫҍѦӊѹҦӁҸ ӊ ҙҲҸѸ ҡґҏҍҽѦ ҡѸѦӁҽ ҡѹҲҏҽ҃ѹҮ, ҡӊҾӉҽӁӁҽҾ ҡ ҦҽӁӁӃѸ ҦѦҮҡҲӊѹѦѸ, ҔҽҔ қӃ ҸҡҲҽѦҲҡҾ ӊ ҲѦӁѹ, ӁѦ “ҽҔҲҏҽґѹӉѹҝҏѦҲҡҾ”. ҭҝѹ ҦҽӁӁҸҮ ҰҸӉѹ҃ѹѹ ӊ ҲѦҔҡҲѦ ҦѦҮҡҲӊѹѦ ӊҸҡҰҝѹӁѹѸҽѦҲҡҾ ӁѦҡҔҸґҧҔҸ ѹӁҽҍѦ, ҍѦѸ ӊ ҡґҏҍҽҾѽ , ѹ Ҳ.Ҧ., ӊ қҸґѦѦ ҸқӂѦѸ ӊѹҦѦ, қѦӉ ҸҡҸқҸҊҸ ӊӁѹѸҽӁѹҾ Ҕ ӁѦҰҸҡҝѦҦҡҲӊѦӁӁӃѸ ҡӊҾӉҾѸ ҡ ҰҝѦҦӃҦҏӂѹѸѹ ѹґѹ ҰҸҡґѦҦҏӀӂѹѸѹ ҦѦҮҡҲӊѹҾѸѹ, ҽ ҡҔҸҝѦѦ ҔҽҔ ԅӹӰ ҦґҾ ҦҝҏҊѹѽ ҦѦҮҡҲӊѹҾѽ. ҶҲҸ, ҸҍѦӊѹҦӁҸ, ҡӊҾӉҽӁҸ ҡ ӊӊҸҦӁӃѸ ѽҽҝҽҔҲѦҝҸѸ ҸҲҝӃӊҔҽ . ҭѦҝѦѽҸҦ Ӂҽ ӍҸҝѸӃ ҋғ ӊ ҸҲҝӃӊҔҽѽ ҰҸҦ ӁҸѸѦҝҸѸ ҰҝѹҦҽѦҲ ҙҲҸҮ ҍҽҡҲѹ ҲѦҔҡҲҽ ӁѦҡҔҸґҧҔҸ ѹӁҸҮ ҸҲҲѦӁҸҔ. ҨҸҝѸӃ ҋғ ӉҽҡҲҽӊґҾӀҲ Ӂҽҡ ӉҦѦҡҧ ҔҸӁ҃ѦӁҲҝѹҝҸӊҽҲҧҡҾ Ӂҽ ҲҸѸ, ҔҽҔ

19

ѰҰҸҲҝѦқґѦӁѹѦ ӊ ѽҸҝӊҽҲҡҔҸѸ, ҡѦҝқҡҔҸѸ ѹ ѸҽҔѦҦҸӁҡҔҸѸ ҾӉӃҔҽѽ ҏҔҽӉҽӁӁӃѽ ѸҸҦҽґҧӁӃѽ ӍҸҝѸ ҦґҾ ҸқҸӉӁҽҍѦӁѹҾ ҸҰҝѦҦѦґѦӁӁӃѽ ҝҽӉӁҸӊѹҦӁҸҡҲѦҮ ҰҸӊҲҸҝҾѦѸҸҡҲѹ ӁҽҰҸѸѹӁҽѦҲ ҽӁҊґѹҮҡҔҏӀ ҔҸӁҡҲҝҏҔ҃ѹӀ ҡҸ ӊҡҰҸѸҸҊҽҲѦґҧӁӃѸ ҊґҽҊҸґҸѸ will, ҸҡҸқѦӁӁҸ ӊ ӍҸҝѸѦ ҰҝҸҗѦҦҗѦҊҸ ӊҝѦѸѦӁѹ ҙҲҸҊҸ ҊґҽҊҸґҽ (would). ѨӁҲѦҝѦҡӁҸ ҸҲѸѦҲѹҲҧ, ҍҲҸ ӊ ҽӁҊґѹҮҡҔѹѽ ҰѦҝѦӊҸҦҽѽ ҦҽӁӁҸҊҸ ӍҝҽҊѸѦӁҲҽ ҙҲҽ ҔҸӁҡҲҝҏҔ҃ѹҾ ҦѦҮҡҲӊѹҲѦґҧӁҸ ҰҝѦҦҡҲҽӊґѦӁҽ, ҽ ѹѸѦӁӁҸ ӊ ҰҸӉѹ҃ѹѹ : However, after a few hours he would begin to feel a terrible thirst (Glenny). 20 ү.ғ. ҷҸӁҦҽҝҔҸ ҰҸ ҸҲӁҸҗѦӁѹӀ Ҕ ҲҽҔѹѸ ҡґҏҍҽҾѸ ҏҰҸҲҝѦқґҾѦҲ ҲѦҝѸѹӁӃ “ҔҝҽҲӁҸ-ҰҽҝӁҽҾ ҔҸӁҡҲҝҏҔ҃ѹҾ” ѹґѹ “ҔҝҽҲӁҸ-҃ѦҰӁҽҾ ҔҸӁҡҲҝҏҔ҃ѹҾ”, ӁҽҰҝѹѸѦҝ ӊ ѦҊҸ ѹӉӊѦҡҲӁҸҮ ҝҽқҸҲѦ (ҷҸӁҦҽҝҔҸ 1971).

ҭҫҋһѣѩҫғүѷѣһѻѢҫҋѷѻ ѩѣҢҋѷғѨҢ ҭҩѨ ҭҫғѷҫҩѤѣҴҫҋѷѨ

13

ҸҲҦѦґҧӁҸ ӊӉҾҲӃѦ ҡѸѦӁӃ ҡѹҲҏҽ҃ѹҮ ҍѦҝѦҦҏӀҲҡҾ ѸѦңҦҏ ҡҸқҸҮ ѹ ҰҸҙҲҸѸҏ ӊҸӉӁѹҔҽѦҲ ӁѦҔҸҲҸҝӃҮ ҸҲҲѦӁҸҔ “ҰҽҝҲѹҔҏґҾҝѹӉҽ҃ѹѹ” ѹ қҸґҧҗѦҮ ӁҽҊґҾҦӁҸҡҲѹ. ѰҰҸҲҝѦқґѦӁѹѦ ҋғ ҡ ҰҸҦҸқӁӃѸѹ ҸҲҲѦӁҔҽѸѹ, ӊ ҰҝѹӁ҃ѹҰѦ, ӊҸӉѸҸңӁҸ ҲҽҔңѦ ӊ ҝҏҡҡҔҸѸ ҾӉӃҔѦ, ӁҸ ӊ ҲҽҔѹѽ ҡґҏҍҽҾѽ ґѹҍӁӃѦ ӍҸҝѸӃ ҊґҽҊҸґҽ ҸқӃҍӁҸ ҸҲӁҸҡҾҲҡҾ Ҕ ҰҝѦӉѦӁҡҏ, Ҳ.Ѧ. ҲҽҔҸѦ ҏҰҸҲҝѦқґѦӁѹѦ ӁҽқґӀҦҽѦҲҡҾ ӊ ҸҡӁҸӊӁҸѸ ҏ ӍҸҝѸ ӁҽҡҲҸҾӂѦҊҸ-қҏҦҏӂѦҊҸ ҋғ. ѷҽҔҸѦ “ӁҽҊґҾҦӁҸ-ҰҝѹѸѦҝӁҸѦ” ҏҰҸҲҝѦқґѦӁѹѦ ҙҲѹѽ ӍҸҝѸ (ӴӹԊӸӾԽp.., ԙԔӾԙԽp..., ԗӹԼԊԂԄӹԌԂӢӠԽp...) ӊ ҰҝѹӁ҃ѹҰѦ қӃґҸ қӃ ӊҸӉѸҸңӁҸ ѹ ӊ ӁҽҗѦѸ ӍҝҽҊѸѦӁҲѦ, ӁҸ ҲҸҊҦҽ ѹӉѸѦӁѹґҡҾ қӃ ҡҲѹґѹҡҲѹҍѦҡҔѹҮ ѽҽҝҽҔҲѦҝ ҲѦҔҡҲҽ. ҭҸ ҸҲӁҸҗѦӁѹӀ Ҕ ҡҸқӃҲѹҾѸ ӊ ҰҝҸҗґҸѸ ӁҽҊґҾҦӁҸ-ҰҝѹѸѦҝӁҸѦ ҏҰҸҲҝѦқґѦӁѹѦ ҋғ ӊ ҝҏҡҡҔҸѸ ҾӉӃҔѦ ӁѦҡҸѸӁѦӁӁҸ ҡґѦҦҏѦҲ ҡҍѹҲҽҲҧ ҡҲѹґѹҡҲѹҍѦҡҔѹ ѸҽҝҔѹҝҸӊҽӁӁӃѸ. ғ ҙҲҸѸ, ҰҸңҽґҏҮ, ѹ ҡҸҡҲҸѹҲ ҸҦӁҸ ѹӉ ҰҝѹӁ҃ѹҰѹҽґҧӁӃѽ ҝҽӉґѹҍѹҮ ѸѦңҦҏ ҝҏҡҡҔѹѸ ҾӉӃҔҸѸ 21 ѹ ҍѦҗҡҔѹѸ, ҡґҸӊҽ҃ҔѹѸ, ҡґҸӊѦӁҡҔѹѸ, ѽҸҝӊҽҲҡҔѹѸ ѹ ҡѦҝқҡҔѹѸ ҾӉӃҔҽѸѹ, ӊ ҔҸҲҸҝӃѽ ҏҔҽӉҽӁӁӃѦ ҡґҏҍҽѹ ҏҰҸҲҝѦқґѦӁѹҾ ҋғ ӊ ҸҲҝӃӊҔҽѽ ҰҸҦ ӁҸѸѦҝҸѸ ѸҸңӁҸ ҡҍѹҲҽҲҧ ҡҲѹґѹҡҲѹҍѦҡҔѹ ӊҰҸґӁѦ ӁѦҮҲҝҽґҧӁӃѸѹ. 22 ҋҔҽӉҽӁӁҸѦ ӊӃҗѦ ҰҸҦҲӊѦҝңҦҽѦҲ ѸӁѦӁѹѦ, ҍҲҸ, ҰҸ ҸҲӁҸҗѦӁѹӀ Ҕ ӊҸӉѸҸңӁҸҡҲѹ ӊӃқҸҝҽ ҋғ Ұҝѹ ҰҸӊҲҸҝҾѦѸҸҡҲѹ, ѸѦңҦҏ ҡґҽӊҾӁҡҔѹѸѹ ҾӉӃҔҽѸѹ ҡҏӂѦҡҲӊҏӀҲ ҦҸӊҸґҧӁҸ ҰҝѹӁ҃ѹҰѹҽґҧӁӃѦ ҝҽӉґѹҍѹҾ. ҭҝѹ ҙҲҸѸ ҸҦӁҸ ѹӉ ҸҡӁҸӊӁӃѽ ҝҽӉґѹҍѹҮ, ҔҽӉҽґҸҡҧ қӃ, ҡҸҡҲҸѹҲ ӊ ҲҸѸ, Ӂҽ ҍҲҸ ҰҝѦңҦѦ ӊҡѦҊҸ ҸҝѹѦӁҲѹҝҏӀҲҡҾ ҸҰҝѦҦѦґѦӁӁӃѦ ҊґҽҊҸґҧӁӃѦ ӍҸҝѸӃ ҦҽӁӁҸҊҸ ҾӉӃҔҽ: Ӂҽ ҏҝҸӊѦӁҧ “ѸҽҔҝҸҡѹҲҏҽ҃ѹѹ”, Ҳ.Ѧ. ӁѦҸҰҝѦҦѦґѦӁӁҸҊҸ ҝҾҦҽ ҰҝҸҾӊґѦӁѹҮ ҰҸӊҲҸҝҾӀӂѦҊҸҡҾ ҦѦҮҡҲӊѹҾ, ѹґѹ Ӂҽ ҏҝҸӊѦӁҧ “ѸѹҔҝҸҡѹҲҏҽ҃ѹѹ” – ҔҽңҦҸҊҸ ҸҲҦѦґҧӁҸ ӊӉҾҲҸҊҸ ҰҝҸҾӊґѦӁѹҾ ҦѦҮҡҲӊѹҾ. 23 ғ ҰѦҝӊҸѸ ҡґҏҍҽѦ ѦҡҲѦҡҲӊѦӁӁӃѸ ҾӊґҾѦҲҡҾ ӊӃқҸҝ Ѣҋғ, ҰҸҡҔҸґҧҔҏ ӁѦҸҊҝҽӁѹҍѦӁӁҽҾ ҰҸӊҲҸҝҾѦѸҸҡҲҧ ҰҝѹҦҽѦҲ ѸҽҔҝҸҡѹҲҏҽ҃ѹѹ ѽҽҝҽҔҲѦҝ, ҡѽҸңѹҮ ҡ ӁѦҰҝѦҦѦґҧӁҸҡҲҧӀ. ғҸ ӊҲҸҝҸѸ ҡґҏҍҽѦ, Ҳ.Ѧ. Ұҝѹ ҔҸӁ҃ѦӁҲҝҽ҃ѹѹ ӊӁѹѸҽӁѹҾ Ӂҽ ѸѹҔҝҸҡѹҲҏҽ҃ѹѹ, ӊ ҰҝѹӁ҃ѹҰѦ, ѸҸңѦҲ ҰҸҾӊѹҲҧҡҾ ҋғ, ҔҸҊҦҽ ҦҸҡҲѹҊӁҏҲҸҡҲҧ ҰҝѦҦѦґҽ ҸҲҦѦґҧӁҸ ӊӉҾҲҸҊҸ ҰҝҸҾӊґѦӁѹҾ ҦѦҮҡҲӊѹҾ ҔҽҔѹѸ-ҲҸ ҸқҝҽӉҸѸ ҽҔҲҏҽґѹӉѹҝҏѦҲҡҾ. ҶҲҸ ҰҝҸѹҡѽҸҦѹҲ ѹѸѦӁӁҸ ӊ ҲѦѽ ҡґҏҍҽҾѽ, ҊҦѦ ҾӊӁҸ ӊҽңӁӃ ӊӉҽѹѸӁӃѦ ҸҲӁҸҗѦӁѹҾ ҦѦҮҡҲӊѹҾ ҡ ҰҝѦҦҗѦҡҲӊҏӀӂѹѸѹ ѹґѹ ҡґѦҦҏӀӂѹѸѹ ҡѹҲҏҽ҃ѹҾѸѹ. ҫҦӁҽҔҸ, ҔҽҔ ҰҸҔҽӉӃӊҽӀҲ ҰҝѹѸѦҝ (10) ѹ ҸҲҝӃӊҸҔ ҰҝѹѸѦҝҽ (11), ҲҽҔҸѸҏ ӊҸҡҰҝѹҾҲѹӀ ѸҸҊҏҲ ҔҽҔ ҝҽӉ ѸѦҗҽҲҧ ҝҽӉґѹҍӁӃѦ ҔҸӁҲѦҔҡҲҏҽґҧӁӃѦ ӍҽҔҲҸҝӃ, ѹ Ұҝѹ ӊӃқҸҝѦ ӊѹҦҽ ѸҸҊҏҲ ҦѦҮҡҲӊҸӊҽҲҧ ҦҸӊҸґҧӁҸ ҲҸӁҔѹѦ ҡҸҸқҝҽңѦӁѹҾ ҸҝҊҽӁѹӉҽ҃ѹѹ ҲѦҔҡҲҽ, ҲҝѦқҏӀӂѹѦ ҦѦҲҽґҧӁҸҊҸ ѹӉҏҍѦӁѹҾ.

21

ғ ҦҽӁӁҸѸ ҸҲӁҸҗѦӁѹѹ ҲҽҔѹѸ ңѦ ҸқҝҽӉҸѸ ӊѦҦҏҲ ҡѦқҾ ҦҝҏҊѹѦ ӊҸҡҲҸҍӁҸҡґҽӊҾӁҡҔѹѦ ҾӉӃҔѹ – қѦґҸҝҏҡҡҔѹҮ ѹ ҏҔҝҽѹӁҡҔѹҮ, ӁҸ ҲҽҔңѦ қҸґҊҽҝҡҔѹҮ ҾӉӃҔ, қҏҦҏҍѹ ӁҽѹқҸґѦѦ ӊҸҡҲҸҍӁӃѸ ҾӉӃҔҸѸ ӀңӁҸҡґҽӊҾӁҡҔҸҮ ҊҝҏҰҰӃ. 22 ѢҽѸ ҰҸҔҽ ѸѦӁѦѦ ҾҡѦӁ ҡҲѹґѹҡҲѹҍѦҡҔѹҮ ҡҲҽҲҏҡ ҏҰҸҲҝѦқґѦӁѹҾ ѸҽҔѦҦҸӁҡҔҸҮ ҔҸӁҡҲҝҏҔ҃ѹѹ ԀӾ + ѹѸҰѦҝӍѦҔҲ ӊ ӁѦӉҽӊѹҡѹѸӃѽ ҰҝѦҦґҸңѦӁѹҾѽ, ҔҽҔ ӊ ҰҝѹѸѦҝѦ (11). 23 ҫ ҰҸҦҸқӁҸѸ ҝҽӉґѹҍѦӁѹѹ ҙҲѹѽ Ҧӊҏѽ ҏҝҸӊӁѦҮ ҡѸ. (Timberlake 1982: 315, Stunová 1986: 470– 471, 1993: 35).

14

үѩҩѨүѢ ҷүҩѣѢѷҋѣѢ

Ҙ ҰҸҦҸқӁӃѸ ӍҽҔҲҸҝҽѸ, ҸҍѦӊѹҦӁҸ, ҡґѦҦҏѦҲ ҸҲӁѦҡҲѹ ѹ ҝҽӉґѹҍѹѦ ѸѦңҦҏ ӁѦӉҽӊѹҡѹѸӃѸ ѹ ӉҽӊѹҡѹѸӃѸ ҡҲҽҲҏҡҸѸ ҦҽӁӁҸҮ ҊґҽҊҸґҧӁҸҮ ӍҸҝѸӃ ӊ ҲѦҔҡҲѦ. ҶҲҸ ѸҸңӁҸ ҰҝҸѹґґӀҡҲҝѹҝҸӊҽҲҧ ҲҽҔңѦ Ӂҽ ҰҝѹѸѦҝѦ (11). ѯҽҡҲѹ ѹ ҝҽӉҦѦґѦӁӃ ӉҽӊѹҡѹѸҸҮ ҸҰҝѦҦѦґѹҲѦґҧӁҸҮ ҔҸӁҡҲҝҏҔ҃ѹѦҮ (ҰҝѹҍҽҡҲѹѦѸ ѹґѹ ҰҝѹҦҽҲҸҍӁӃѸ ҰҝѦҦґҸңѦӁѹѦѸ ҸҰҝѦҦѦґѹҲѦґҧӁҸҊҸ ҲѹҰҽ). ѨӁҲѦҝѦҡӁҸ ҸҲѸѦҲѹҲҧ, ҍҲҸ ӊҸ ӊҡѦѽ ҰѦҝѦӊҸҦҽѽ ӉҦѦҡҧ ѹҡҰҸґҧӉҏѦҲҡҾ ӍҸҝѸҽ ҋғ (ҡѸ. ҰҸҦҍѦҝҔӁҏҲҏӀ ҍҽҡҲҧ ӊ ҰҝѹѸѦҝҽѽ), ѽҸҲҾ ҦѦҮҡҲӊѹѦ, ҏҔҽӉҽӁӁҸѦ ҦҽӁӁӃѸѹ ӍҸҝѸҽѸѹ, ӁѦҡҸѸӁѦӁӁҸ ҲҽҔңѦ ҰҸӊҲҸҝҾѦҲҡҾ. ҫҍѦӊѹҦӁҸ, ҦҽңѦ ӊ ҝҏҡҡҔҸѸ ҾӉӃҔѦ ҰҸӉѹ҃ѹҾ ӊ ҲҽҔҸҮ ҔҸӁҡҲҝҏҔ҃ѹѹ ҡѹґҧӁҸ ҸқґѦҊҍҽѦҲ ӊҸӉѸҸңӁҸҡҲҧ ӊӃқҸҝҽ ҋғ. ғ ҡґѦҦҏӀӂѹѽ ҰҽҝҽҊҝҽӍҽѽ ӊѦҝӁѦѸҡҾ Ҕ ҝҽҡҡѸҸҲҝѦӁѹӀ ҡґҸңӁҸҰҸҦҍѹӁѦӁӁӃѽ ҰҝѦҦґҸңѦӁѹҮ, ӁҽҍҽҲҸѸҏ ӊ Ұҽҝ. 1. ҫӁѹ ѹӁҲѦҝѦҡӁӃ ѹѸѦӁӁҸ ҲѦѸ, ҍҲҸ ҏңѦ ҡҽѸҽ ҡҲҝҏҔҲҏҝҽ ҲҽҔѹѽ ҰҝѦҦґҸңѦӁѹҮ ҰҝѦҦҰҸґҽҊҽѦҲ ӁҽґѹҍѹѦ ҸҡҸқҸҊҸ ӊӁѹѸҽӁѹҾ Ҕ ӊӉҽѹѸҸҡӊҾӉҾѸ ҦѦҮҡҲӊѹҮ, ҏҔҽӉҽӁӁӃѽ ӊ ҊґҽӊӁҸҮ ѹ ҰҝѹҦҽҲҸҍӁҸҮ ҍҽҡҲҾѽ. ҴҸңӁҸ ҸңѹҦҽҲҧ, ҍҲҸ ҙҲҸ ҡѹґҧӁҸ ҡҰҸҡҸқҡҲӊҏѦҲ ӊӃқҸҝҏ ҋғ ӊ ҲѦѽ ҡґҏҍҽҾѽ, ҔҸҊҦҽ ҝѦҍҧ ѹҦѦҲ Ҹ ҰҸҡґѦҦҸӊҽҲѦґҧӁҸҡҲѹ ҦѦҮҡҲӊѹҮ (ҡѸ. ҰҝѹѸѦҝӃ (9қ) ѹ (9ӊ)). ѩҽӁӁӃѦ ҡґҸңӁӃѦ ҰҝѦҦґҸңѦӁѹҾ ҰҸӉӊҸґҾӀҲ ӁҽѸ ҲҽҔңѦ ҰҝҸҡґѦҦѹҲҧ, ҔҽҔҏӀ ҝҸґҧ ѹҊҝҽѦҲ ҲҸҲ ӍҽҔҲ, ҍҲҸ ҊґҽҊҸґҧӁҽҾ ӍҸҝѸҽ ҰҝѹҦҽҲҸҍӁҸҮ ҍҽҡҲѹ ҰҝѦҦґҸңѦӁѹҾ ӁҽѽҸҦѹҲҡҾ ӊ ӉҽӊѹҡѹѸҸҮ ҰҸӉѹ҃ѹѹ. 3. ҭҝѹҦҽҲҸҍӁӃѦ ҰҝѦҦґҸңѦӁѹҾ ҡ ҡҸӀӉҽѸѹ ҲѹҰҽ ӪԂӪ ԽӹԊӸӪӹ ѢҽѹқҸґѦѦ қґҽҊҸҰҝѹҾҲӁӃѦ ҏҡґҸӊѹҾ ҦґҾ ҏҰҸҲҝѦқґѦӁѹҾ ҋғ ӊ ҸқѦѹѽ ҍҽҡҲҾѽ ҰҝѦҦґҸңѦӁѹҾ ѹѸѦӀҲҡҾ Ұҝѹ ҡҸӀӉҽѽ ҲѹҰҽ ӪԂӪ ԽӹԊӸӪӹ, ҏҔҽӉӃӊҽӀӂѹѽ Ӂҽ ҰҸҡґѦҦҸӊҽҲѦґҧӁҸҡҲҧ ҦѦҮҡҲӊѹҮ ҡ ѸѹӁѹѸҽґҧӁӃѸ ѹӁҲѦҝӊҽґҸѸ. 24 ғӃҗѦ ҏңѦ қӃґѹ ҰҝѹӊѦҦѦӁӃ ҰҝѹѸѦҝӃ (9ҽ), (9қ) ѹ (9ӊ), ҰҸҔҽӉӃӊҽӀӂѹѦ, ҍҲҸ ӊ ӁѦҔҸҲҸҝӃѽ ҾӉӃҔҽѽ ӊ ҰҸҦҸқӁҸѸ ҡґҏҍҽѦ ҦѦҮҡҲӊѹҲѦґҧӁҸ ҏҰҸҲҝѦқґҾѦҲҡҾ ҋғ. ҋґѦҦҏӀӂҽҾ Ҳҽқґѹ҃ҽ ҰҸӉӊҸґҾѦҲ ҡҝҽӊӁѹҲҧ ѹѽ ҡ ҦҝҏҊѹѸѹ ҰѦҝѦӊҸҦҽѸѹ ҦҽӁӁҸҊҸ ӍҝҽҊѸѦӁҲҽ, ѹѸѦӀӂѹѸѹҡҾ ӊ ӁҽҗѦѸ ҝҽҡҰҸҝҾңѦӁѹѹ. (12) Ҵѹѽҽѹґ ҷҏґҊҽҔҸӊ: ԠԂԙԽӾӢ Ӡ ԠԂӢԄԂӢӠԽԂ ҩҏҡҡҔѹҮ ҷѦґҸҝҏҡ. ѰҔҝ. ҭҸґҧҡҔ.

24

һѹҗҧ ҲҸґҧҔҸ ӁҽҍѹӁҽґi ӉӊѦӁѦҲҧi ҲѦґѦӍҸӁ, ғҽҝѦӁҏѽҽ ԼӢԂԊi ҲҝҏқҔҏ [...] ѤҔ ҲҸґҧҔҟ ҰҽҍӃӁҽҕi ӉӊҟӁѦ҃ҧi ҲҙґѦӍҸӁ, ғҽҝҙӁҏѽҽ ԼӢԂ԰i ҡґҏѽҽҕҔҏ [...] ѷҟґҧҔѹ-ӁҸ ҰҸҍѹӁҽӊi ҦӉѦґѦӁҍҽҲѹi ҲѦґѦӍҸӁ, ғҽҝѦӁҏѽҽ ԼӢԂӴi ҝҏҝҔҏ [...] Skoro tylko telefon zaczyna’i dzwoniǀi, Warionucha podnosi’i s’uchawkNj [...]

ҭҸҦҸқӁӃѦ ҡґҏҍҽѹ ҔґҽҡҡѹӍѹ҃ѹҝҏӀҲҡҾ ғ.ҋ. ҤҝҽҔҸӊҡҔѹѸ ҔҽҔ “ҔҸӁҲҽҔҲӁҸѦ ҰҝѦҦҗѦҡҲӊҸӊҽӁѹѦ” (2003: 42).

ҭҫҋһѣѩҫғүѷѣһѻѢҫҋѷѻ ѩѣҢҋѷғѨҢ ҭҩѨ ҭҫғѷҫҩѤѣҴҫҋѷѨ

ѯѦҗҡҔ. ҋґҸӊҽ҃Ҕ. ҋґҸӊѦӁҡҔ. ҤҸҝӊҽҲҡҔ. ҋѦҝқҡҔѹҮ ҴҽҔѦҦҸӁ. ҷҸґҊҽҝ.

15

Sotva zazvonilp telefon, Varenucha zvedlp sluchátko [...] Len ǁo zazvonilp telefón, Varenucha zdviholp slúchadlo [...] Samo da je zazvonilp telefon, pa je Varenuha vzelp slušalko [...] Tek što bi [cond.] poǁeop zvonitii telefon, Varenuha bi [cond.] uzimaoi slušalicu [...] ѯѹѸ қѹ [cond.] ӉҽӉӊҸӁѹҸp ҲѦґѦӍҸӁ, ғҽҝѦӁҏѽҽ ԼӠ [cond.] ԌӠԒԂӹi ҡґҏҗҽґѹ҃ҏ [...] ҋҽѸҸ ѱѦ [exp.] ӉҽҠӊҸӁѦҗѦp [impf.] ҲѦґѦӍҸӁ, ҽ ғҽҝѦӁҏѽҽ Ҏҽ ӪӢӾӴԂԶӾi [impf.] ҡґҏҗҽґҔҽҲҽ [...] ӅҸѸ ӉӊӆӁӁѦҗѦp [impf.] ҲѦґѦӍҸӁӆҲ, ғҽҝѦӁҏѽҽ ӴԌӠԄԂԶӾi [impf.] ҡґҏҗҽґҔҽҲҽ [...]

ҭҝѹ ҡҝҽӊӁѦӁѹѹ ҙҲѹѽ ҰҝѹѸѦҝҸӊ қҝҸҡҽѦҲҡҾ ӊ ҊґҽӉҽ, ҍҲҸ ҋғ ҰҝѦҸқґҽҦҽѦҲ ҰҝѦңҦѦ ӊҡѦҊҸ ӊ ҰҝѹҦҽҲҸҍӁҸҮ ҍҽҡҲѹ. ѷҸґҧҔҸ ӊ ҍѦҲӃҝѦѽ ѹӉ ҸҦѹӁӁҽҦ҃ҽҲѹ ҰҝѹӊѦҦѦӁӁӃѽ ҾӉӃҔҸӊ ҸӁҽ ҡҸҦѦҝңѹҲ ӍҸҝѸҏ Ѣҋғ. ҘҽҔ қӃґҸ ҏңѦ ҏҔҽӉҽӁҸ ӊ Ұҽҝ. 1, Ҕ ӁѹѸ ҸҲӁҸҡѹҲҡҾ ҝҏҡҡҔѹҮ ҾӉӃҔ. Ѣҽ ҸҡӁҸӊѦ ҡҸҸҲӁҸҗѦӁѹҾ ӊѹҦҸӊ ҰҝѹҦҽҲҸҍӁҸҮ ѹ ҊґҽӊӁҸҮ ҍҽҡҲѦҮ ӊӃҦѦґҾӀҲҡҾ Ҳҝѹ ҲѹҰҽ: Ѣҋғ–Ѣҋғ (Ѣҋғ ӊ ҸқѦѹѽ ҍҽҡҲҾѽ), ҋғ–ҋғ (ҋғ ӊ ҸқѦѹѽ ҍҽҡҲҾѽ), ҋғ–Ѣҋғ (ҋғ ӊ ҰҝѹҦҽҲҸҍӁҸҮ ҍҽҡҲѹ, Ѣҋғ ӊ ҊґҽӊӁҸҮ). Ҙ ҲѹҰҏ ҋғ–ҋғ ҸҲӁҸҡѹҲҡҾ ҍѦҗҡҔѹҮ ҰҝѹѸѦҝ, ҽ ҲҽҔңѦ ҰҝѹѸѦҝ ѹӉ қґѹӉҔҸҝҸҦҡҲӊѦӁӁҸҊҸ ҡґҸӊҽ҃ҔҸҊҸ. ѨӁҲѦҝѦҡӁҸ ҸҲѸѦҲѹҲҧ, ҍҲҸ ҙҲҸҲ ңѦ ҲѹҰ ҰҝѦҦҡҲҽӊґѦӁ ѹ ӊ ҡґҸӊѦӁҡҔҸѸ ҰѦҝѦӊҸҦѦ, ѽҸҲҾ ӊ ҲҝҽҦѹ҃ѹҸӁӁҸҮ ҔґҽҡҡѹӍѹҔҽ҃ѹѹ ҡґҽӊҾӁҡҔѹѽ ҾӉӃҔҸӊ ҡґҸӊѦӁҡҔѹҮ ҸҲӁҸҡѹҲҡҾ Ҕ ҦҝҏҊҸҮ ҊҝҏҰҰѦ. ғӉҽѹѸӁҽҾ қґѹӉҸҡҲҧ ҏҔҽӉҽӁӁӃѽ ҲҝѦѽ ҾӉӃҔҸӊ ӊ ҸҲӁҸҗѦӁѹѹ ӍҏӁҔ҃ѹҸӁѹҝҸӊҽӁѹҾ ӊѹҦҸӊ Ұҝѹ ҰҸӊҲҸҝҾѦѸҸҡҲѹ ҰҸҦҲӊѦҝңҦҽѦҲҡҾ ѸӁҸңѦҡҲӊҸѸ ҦҝҏҊѹѽ ҰҝѹѸѦҝҸӊ, ӁҽҮҦѦӁӁӃѽ ӁҽѸѹ ӊ ҰҽҝҽґґѦґҧӁҸѸ ҔҸҝҰҏҡѦ. 25 ҘҽҔ ҏңѦ қӃґҸ ҸҲѸѦҍѦӁҸ ӊӃҗѦ, ҔҝҽҮӁѦ ҰҝҸҲѹӊҸҰҸґҸңӁҏӀ ҰҸӉѹ҃ѹӀ ӁҽѽҸҦѹѸ ӊ ҝҏҡҡҔҸѸ ҾӉӃҔѦ, ҦґҾ ҔҸҲҸҝҸҊҸ ѦҡҲѦҡҲӊѦӁӁӃѸ ҾӊґҾѦҲҡҾ ҔҽҔ ҝҽӉ ҲѹҰ Ѣҋғ–Ѣҋғ. ѷҽҔѹѸ ңѦ ҸқҝҽӉҸѸ ӊѦҦҏҲ ҡѦқҾ қґѹӉҔҸҝҸҦҡҲӊѦӁӁӃѦ қѦґҸҝҏҡҡҔѹҮ ѹ ҏҔҝҽѹӁҡҔѹҮ ҾӉӃҔѹ. ғ ҰҝѹѸѦҝѦ (12) ҲѹҰ Ѣҋғ–Ѣҋғ ҰҝѦҦҡҲҽӊґѦӁ ҲҽҔңѦ ӊ ҰҸґҧҡҔҸѸ ҰѦҝѦӊҸҦѦ. ҫҦӁҽҔҸ Ұҝѹ ҝҽҡҡѸҸҲҝѦӁѹѹ қҸґѦѦ ҸқҗѹҝӁҸҊҸ ѸҽҲѦҝѹҽґҽ ҸҔҽӉӃӊҽѦҲҡҾ, ҍҲҸ ҦґҾ ҰҸґҧҡҔҸҊҸ ҾӉӃҔҽ ҙҲҸҲ ҲѹҰ ӊҡѦ ңѦ ӁѦґҧӉҾ ҡҍѹҲҽҲҧ ҡҽѸӃѸ ѽҽҝҽҔҲѦҝӁӃѸ. ғ қҸґҧҗѹӁҡҲӊѦ ҦҝҏҊѹѽ Ӂҽҗѹѽ ҰҝѹѸѦҝҸӊ ѹӉ ҰҸґҧҡҔѹѽ ҲѦҔҡҲҸӊ ӁҽқґӀҦҽѦҲҡҾ ҦҝҏҊҸѦ ҡҸҸҲӁҸҗѦӁѹѦ ӊѹҦҸӊ, ҽ ѹѸѦӁӁҸ ҰҸ ҲѹҰҏ ҋғ–Ѣҋғ. ҭҸ ѸӁѦӁѹӀ ӁҸҡѹҲѦґѦҮ ҾӉӃҔҽ, ѹ ӊ ҦҽӁӁҸѸ ҡґҏҍҽѦ ӍҸҝѸҏ Ѣҋғ (zaczyna’) ѸҸңӁҸ ґѦҊҔҸ ӉҽѸѦӁѹҲҧ ӍҸҝѸҸҮ ҋғ (zaczƾ’).

25

Ѣҽ қґѹӉҸҡҲҧ ҙҲѹѽ ҾӉӃҔҸӊ ӊ ҽҡҰѦҔҲҏҽґҧӁҸѸ ҸҲӁҸҗѦӁѹѹ ҏҔҽӉӃӊҽѦҲҡҾ ҲҽҔңѦ ӊ ҦҝҏҊѹѽ ҝҽқҸҲҽѽ ҰҸ ҡҝҽӊӁѹҲѦґҧӁҸҮ ҡґҽӊҾӁҡҔҸҮ ҽҡҰѦҔҲҸґҸҊѹѹ, ӁҽҰҝ. (Mønnesland 1984) ѹ (Dickey 2000).

16

үѩҩѨүѢ ҷүҩѣѢѷҋѣѢ

ғҸ ӊҡѦѽ ҸҡҲҽґҧӁӃѽ ҾӉӃҔҽѽ ӊ ҦҽӁӁҸѸ ҰҝѹѸѦҝѦ ҰҝѦҦҡҲҽӊґѦӁ ҲѹҰ ҋғ–Ѣҋғ. ҫҦӁҽҔҸ ӊ қҸґҧҗѹӁҡҲӊѦ ѹӉ Ӂѹѽ ҲҽҔңѦ ӊҸӉѸҸңѦӁ ҲѹҰ ҋғ–ҋғ, Ҳ.Ѧ. ҸӁѹ ҦҸҰҏҡҔҽӀҲ ҋғ ѹ ӊ ҊґҽӊӁҸҮ ҍҽҡҲѹ (ҡҝ., ӁҽҰҝѹѸѦҝ (13)). ҶҲҸ ѹѸѦӁӁҸ ҲѦ ҾӉӃҔѹ, ҔҸҲҸҝӃѦ ҸҲґѹҍҽӀҲҡҾ ѹҡҰҸґҧӉҸӊҽӁѹѦѸ ѸҸҦҽґҧӁӃѽ ӍҸҝѸ ҦґҾ ӊӃҝҽңѦӁѹҾ ҰҸӊҲҸҝҾѦѸҸҡҲѹ – ѽҸҝӊҽҲҡҔѹҮ, ҡѦҝқҡҔѹҮ ѹ ѸҽҔѦҦҸӁҡҔѹҮ. ғ ҝҽҡҡѸҽҲҝѹӊҽѦѸӃѽ ҡґҏҍҽҾѽ ҡ ҰҸӊҲҸҝҾӀӂѦҮҡҾ ҃ѦҰҸҍҔҸҮ ҦѦҮҡҲӊѹҮ ҦґҾ ҡѦҝқҡҔҸҊҸ ѹ ѽҸҝӊҽҲҡҔҸҊҸ ҾӉӃҔҸӊ ӁҽґѹҍѹѦ ҔҸӁҦѹ҃ѹҸӁҽґҽ ҋғ ӊ ҰҝѹҦҽҲҸҍӁҸҮ ҍҽҡҲѹ ѸҸңӁҸ ҡҍѹҲҽҲҧ ҡҲҽӁҦҽҝҲӁӃѸ. ҘҝҸѸѦ ҲҸҊҸ ҙҲҽ ӍҸҝѸҽ җѹҝҸҔҸ ҰҝѦҦҡҲҽӊґѦӁҽ ѹ ӊ ҊґҽӊӁҸҮ ҍҽҡҲѹ (ҡѸ. ҰҝѹѸѦҝ (13). ҴҽҔѦҦҸӁҡҔҽҾ ӍҸҝѸҽ ԀӾ + ѹѸҰѦҝӍѦҔҲ ҋғ ҲҽҔңѦ ҾӊґҾѦҲҡҾ ҡҲҽӁҦҽҝҲӁҸҮ ҦґҾ ҰҝѹҦҽҲҸҍӁҸҮ, ӁҸ ӊҡҲҝѦҍҽѦҲҡҾ ѹ ӊ ҊґҽӊӁҸҮ ҍҽҡҲѹ. ҫҦӁҽҔҸ ӊ ҙҲҸѸ ҾӉӃҔѦ ҲѹҰ ҋғ–ҋғ ҰҝѦҦҡҲҽӊґѦӁ ӊҡѦ ңѦ ӉӁҽҍѹҲѦґҧӁҸ ҝѦңѦ, ҍѦѸ ҲѹҰ ҋғ–Ѣҋғ. ҭҸҦҍѦҝҔӁѦѸ, ҍҲҸ ӊ ҲҝѦѽ ӁҽӉӊҽӁӁӃѽ ҾӉӃҔҽѽ Ұҝѹ ҰҸҡґѦҦҸӊҽҲѦґҧӁҸҡҲѹ ҦѦҮҡҲӊѹҮ ӊӃқҸҝ ѸѦңҦҏ ҋғ ѹ Ѣҋғ ҡҏӂѦҡҲӊҏѦҲ ҲҸґҧҔҸ ӊ ҊґҽӊӁҸҮ ҍҽҡҲѹ, ӊ ҰҝѹҦҽҲҸҍӁҸҮ ңѦ ҏҔҽӉҽӁӁӃѦ ѸҸҦҽґҧӁӃѦ ӍҸҝѸӃ ҋғ ҾӊґҾӀҲҡҾ ӍҽҔҲѹҍѦҡҔѹ ҸқҾӉҽҲѦґҧӁӃѸѹ (ҡҝ. ҰҝѹѸѦҝӃ, ҰҝѹӊѦҦѦӁӁӃѦ ӁѹңѦ). ғ ҝҽҡҡѸҸҲҝѦӁӁӃѽ ӁҽѸѹ ҰѦҝѦӊҸҦҽѽ Ӂҽ қҸґҊҽҝҡҔѹҮ ҾӉӃҔ ӊҡѦ ҰҸҦҸқӁӃѦ ҰҝѹѸѦҝӃ ҸҲӁҸҡҾҲҡҾ Ҕ ҲѹҰҏ ҋғ–Ѣҋғ. ѨӁҲѦҝѦҡӁҸ ҸҲѸѦҲѹҲҧ, ҍҲҸ ҡҲҽӁҦҽҝҲӁҸҮ ӍҸҝѸҸҮ ӊ ҰҝѹҦҽҲҸҍӁҸҮ ҍҽҡҲѹ ҰҸҦҸқӁӃѽ ҰҝѦҦґҸңѦӁѹҮ ҾӊґҾѦҲҡҾ ѹѸҰѦҝӍѦҔҲ ҋғ, ҔҸҲҸҝӃҮ ӊ ҦҝҏҊѹѽ ҰҸӉѹ҃ѹҾѽ ҏҰҸҲҝѦқґҾѦҲҡҾ ҝѦҦҔҸ. ғӃқҸҝ ҙҲҸҮ ӍҸҝѸӃ ґѦҊҔҸ ҸқӆҾҡӁҾѦҲҡҾ ѦѦ ҸҡҸқҸҮ ҡѦѸҽӁҲѹҔҸҮ, ҡҸҡҲҸҾӂѦҮ ѹӉ ҡӊҸѦҸқҝҽӉӁҸҊҸ ҡҸҍѦҲҽӁѹҾ Ҧӊҏѽ ҰҝҸҲѹӊҸҰҸґҸңӁӃѽ ҽҡҰѦҔҲҏҽґҧӁӃѽ ҙґѦѸѦӁҲҸӊ. ҷґҽҊҸҦҽҝҾ ҙҲҸѸҏ ҦҽӁӁҽҾ ӍҸҝѸҽ ҡҰҸҡҸқӁҽ ҸҦӁҸӊҝѦѸѦӁӁҸ ӊӃҝҽӉѹҲҧ ҽҡҰѦҔҲҏҽґҧӁӃҮ ѽҽҝҽҔҲѦҝ ҸқҸѹѽ ҏҝҸӊӁѦҮ ҰҸӊҲҸҝҾӀӂѦҊҸҡҾ ҃ѦґҸҡҲӁҸҊҸ ҰҝѦҦѦґҧӁҸҊҸ ҦѦҮҡҲӊѹҾ. ғ ҙҲѹѽ ҡґҏҍҽҾѽ ҋғ, ӊӃҝҽңѦӁӁӃҮ ӊ ӊѹҦҸӊҸҮ ҸҡӁҸӊѦ ӍҸҝѸӃ, ҸҝѹѦӁҲѹҝҏѦҲҡҾ Ӂҽ ѸѹҔҝҸҡѹҲҏҽ҃ѹѹ, ӊ ҲҸ ӊҝѦѸҾ ҔҽҔ ҸқӂҽҾ ҡѦѸҽӁҲѹҔҽ ѹѸҰѦҝӍѦҔҲҽ ѽҸҝҸҗҸ ҸҲҝҽңҽѦҲ ѽҽҝҽҔҲѦҝ ѸҽҔҝҸҡѹҲҏҽ҃ѹѹ. 26 ғѦҡҧѸҽ ҰҝѹѸѦҍҽҲѦґҧӁҸ, ҍҲҸ ҦҽӁӁҽҾ ҡҰҸҡҸқӁҸҡҲҧ ӍҸҝѸӃ Ӂҽҗґҽ ҰҝѹѸѦӁѦӁѹѦ ҰҝҽҔҲѹҍѦҡҔѹ ѹҡҔґӀҍѹҲѦґҧӁҸ ӊ ӉҽӊѹҡѹѸҸҮ ҰҸӉѹ҃ѹѹ. 27 ғ ҙҲҸѸ ҰґҽӁѦ ӁҽқґӀҦҽѦҲҡҾ ӉҽѸѦҲӁҸѦ ҝҽӉґѹҍѹѦ ҡ ѸҽҔѦҦҸӁҡҔѹѸ ҾӉӃҔҸѸ, ҊҦѦ ѹѸҰѦҝӍѦҔҲ ҋғ ӊ ҔҸӁҡҲҝҏҔ҃ѹѹ ҡ ҍҽҡҲѹ҃ѦҮ ԀӾ, ҏҰҸҲҝѦқґҾѦҲҡҾ қҸґѦѦ ҡӊҸқҸҦӁҸ (ҡҝ. ҰҝѹѸѦҝ (11)).

26

҈ӁҽҍѦӁѹѦ ҰҝҸҲҸҲѹҰѹҍѦҡҔҸҊҸ ѹѸҰѦҝӍѦҔҲҽ ѸӃ ҰҸӁѹѸҽѦѸ ҡґѦҦҏӀӂѹѸ ҸқҝҽӉҸѸ: ѹѸҰѦҝӍѦҔҲ ҰҝѦҦҸҡҲҽӊґҾѦҲ ҸҝѹѦӁҲҽ҃ѹӀ ӊӁҏҲҝҧ ҸқҸӉӁҽҍҽѦѸҸҮ ҡѹҲҏҽ҃ѹѹ, ҰҝѹҍѦѸ ҙҲҽ ҡѹҲҏҽ҃ѹҾ ӊҸҡҰҝѹӁѹѸҽѦҲҡҾ ҔҽҔ ҡҏӂѦҡҲӊҏӀӂҽҾ ҸҦӁҸӊҝѦѸѦӁӁҸ ҡ ҔҽҔҸҮ-ӁѹқҏҦҧ ҦҝҏҊҸҮ ҡѹҲҏҽ҃ѹѦҮ (ѹґѹ ҔҽҔ ӊҔґӀҍҽӀӂҽҾ ѦѦ). 27 ҋѸ. ҏҲӊѦҝңҦѦӁѹѦ Ҍ.ҋ. ҴҽҡґҸӊҽ Ҹ ҲҸѸ, ҍҲҸ “ӊ ҊґҽӊӁҸѸ ҰҝѦҦґҸңѦӁѹѹ ҰҸҡҲҽӁҸӊҔҽ ӁѦҡҸӊѦҝҗѦӁӁҸҊҸ ѹѸҰѦҝӍѦҔҲҽ ҾӊґҾѦҲҡҾ ӊ ҡҸӊҝѦѸѦӁӁҸѸ ҾӉӃҔѦ ҰҝҽӊѹґҸѸ [...] ҽ ҰҸҡҲҽӁҸӊҔҽ ҡҸӊѦҝҗѦӁӁҸҊҸ ѹѸҰѦҝӍѦҔҲҽ – ѹҡҔґӀҍѦӁѹѦѸ” (ҴҽҡґҸӊ 1959: 272). ҩѦҦҔҸҡҲҧ ѹѸҰѦҝӍѦҔҲҽ ҋғ ӊ ӁѦӉҽӊѹҡѹѸҸҮ ҰҸӉѹ҃ѹѹ ҸҲѸѦҍҽѦҲҡҾ ҲҽҔңѦ ӊ ҝҽқҸҲѦ (Mønnesland 1984: 68). ғ ҝҽӉґѹҍӁӃѽ ҝҽқҸҲҽѽ Ҍ.ҋ. ҴҽҡґҸӊҽ ҏҦѦґҾѦҲҡҾ қҸґҧҗҸѦ ӊӁѹѸҽӁѹѦ ҏҰҸҲҝѦқґѦӁѹӀ қҸґҊҽҝҡҔҸҊҸ ѹѸҰѦҝӍѦҔҲҽ ҋғ. ҤҸҝҸҗѹҮ ҸқӉҸҝ ҦҽѦҲҡҾ ӊ ҝҽқҸҲѦ (ҴҽҡґҸӊ 1984: 170-177).

ҭҫҋһѣѩҫғүѷѣһѻѢҫҋѷѻ ѩѣҢҋѷғѨҢ ҭҩѨ ҭҫғѷҫҩѤѣҴҫҋѷѨ

17

ҋҔҽӉҽӁӁҸѦ ӊӃҗѦ ҰҝҸѹґґӀҡҲҝѹҝҏѦѸ Ӂҽ ҦҝҏҊѹѽ ҰҝѹѸѦҝҽѽ, ӊ ҔҸҲҸҝӃѽ ҡґҽӊҾӁҡҔѹҮ ѸҽҲѦҝѹҽґ ҡҸҡҲҸѹҲ ѹӉ ҰҽҝҽґґѦґҧӁӃѽ ҰѦҝѦӊҸҦҸӊ ҡ ҽӁҊґѹҮҡҔҸҊҸ ѹґѹ ӍҝҽӁ҃ҏӉҡҔҸҊҸ. ҋґѦҦҏӀӂѹҮ ҰҝѹѸѦҝ ѹӉ ҔӁѹҊѹ ԣԊӠԙԂ Ӵ ԙԽӢԂӰӾ ԓӫԌӾԙ ҾӊґҾѦҲҡҾ ҸҲҝӃӊҔҸѸ ѹӉ ҸҰѹҡҽӁѹҾ ӁѦҸқӃҍӁҸҮ ѹҊҝӃ ӊ ҔҝҸҔѦҲ ҏ ҘҸҝҸґѦӊӃ. (13) Lewis Carroll: Alice’s Adventures in Wonderland үӁҊґ.

[The chief difficulty Alice found at first was in managing her flamingo: she succeeded in getting its body tucked away, comfortably enough, under her arm, with its legs hanging down, but] generally, just as she had got its neck nicely straightened out, [and was going to give the hedgehog a blow with its head,] it would twist itself round and look up in her face, [with such a puzzled expression that she could not help bursting out laughing]. ҩҏҡҡҔѹҮ [...] ѦҦӊҽ ҸӁҽ ҏҡҰѦӊҽґҽi ҽҔҔҏҝҽҲӁҸ ӊӃҰҝҾѸѹҲҧp ӍґҽѸѹӁҊҸ җѦӀ [...] ҔҽҔ ҸӁ ҸқҾӉҽҲѦґҧӁҸ ӹԼӹӢԂԓӠӴԂԊԙԋi ѹ ԒԂԄԊԋԌՂӴԂԊi ѦҮ ӊ ҊґҽӉҽ [...] (҈ҽѽҸҦѦҝ). ѰҔҝ. ѷҽ ҲҟґҧҔѹ-ӁҸ ӊҸӁҽ ӊѹҰҝҸҡҲҏӊҽґҽi ҰҲҽѽҸӊҟ җѹӀ, ҾҔ ӍґҽѸҟӁҊҸ ҲҏҲ-ҲҽҔѹ ӴӠԄӠӰԂӴԙԋi ӁҽӉҽҦ ҟ ԒԂԒӠӢԂӴi ѶҮ ҏ ӊҟҍҟ [...] ҭҸґҧҡҔ. [...] ilekroǀ 28 wyprostowa’ap mu naleǻycie szyjNj [...] on jƾ zaraz wykrNjca’i i spoglƾda’i jej w twarz [...] (Stiller). ѯѦҗҡҔ. [...] sotva mu pLJknLJ narovnalap krk [...] plameǞák se pokaždé stoǁilp dozadu a [...] se na ni zadívalp [...] ҋґҸӊҽ҃Ҕ. [...] zakaždým, keDŽ mu narovnalap krk [...] plameniak sa obrátilp a pozeralp jej do tváre [...] ҋґҸӊѦӁҡҔ. [...] brž ko mu je poravnalap vrat [...] se je še vsakiǁ zasukalp in ji pogledalp v oǁi [...] ҤҸҝӊҽҲҡҔ. [...] kad bi [cond.] mu opružilap vrat [...] on bi [cond.] uzviop vrat i pogledaop je u lice [...] ҋѦҝқҡҔѹҮ үґѹ ҔҽҦ ҊҸҦ қѹ [cond.] ӍґҽѸѹӁҊҏ ґѦҰҸ ѹҡҰҝҽӊѹґҽp ӊҝҽҲ [...] ҸӁ ԼӠ [cond.] ԙӾ ӠԒӴӠӹp ѹ [...] ԗӹԄԊӾԌԂӹp ҏ ѬѦӁҸ ґѹ҃Ѧ [...] ҴҽҔѦҦҸӁ. [...] җҲҸѸ ѱѦ [exp.] Ѹҏ ҊҸ ѹҡҰҝҽӊѦҗѦp [impf.] ҏқҽӊҸ ӊҝҽҲҸҲ [...] ҲҸҎ ԀӾ [exp.] Ԅӹ ԙӴӠԽӪԂԶӾp [impf.] ӊҝҽҲҸҲ ҸҦӁҸӊҸ ѹ ԀӾ [exp.] ԗӹԄԊӾԌӰӾԶӾp [impf.] ӊ ґѹ҃Ѧ [...] ҷҸґҊҽҝҡҔ. ғҡѦҔѹ ҰӆҲ ҔҸҊҽҲҸ ѹӉҰӆӁѦҗѦp [impf.] җѹҾҲҽ Ѹҏ ҡ ҊґҽӊҽҲҽ ӁҽҦҸґҏ [...] ӍґҽѸѹӁҊҸҲҸ ҲҏҲҽҔҡѹ ԙӾ ӠԒӴӠӴԂԶӾi [impf.], ӫԙӫӪӴԂԶӾi [impf.] ԙӾ ӁҽҊҸҝѦ ѹ ԒԋԗӴԂԶӾi [impf.] үґѹҡҽ ӊ ґѹ҃ѦҲҸ [...]

28

ҭҸґҧҡҔѹҮ ҡҸӀӉ ilekroǀ ҏңѦ ҡҽѸ ҰҸ ҡѦқѦ ҏҔҽӉӃӊҽѦҲ Ӂҽ ҰҸӊҲҸҝҾѦѸҸҡҲҧ ‘ҔҽңҦӃҮ ҝҽӉ, ҔҸҊҦҽ’.

18

үѩҩѨүѢ ҷүҩѣѢѷҋѣѢ

ғ ҽӁҊґѹҮҡҔҸѸ ҸҝѹҊѹӁҽґѦ ҏҰҸҲҝѦқґѦӁѹѦѸ ҸқҡҲҸҾҲѦґҧҡҲӊҽ generally ѹ ҔҸӁҡҲҝҏҔ҃ѹѦҮ ҡ ӍҸҝѸҸҮ would ҾҡӁҸ ҏҔҽӉҽӁҸ, ҍҲҸ ҝѦҍҧ ѹҦѦҲ Ҹ ҰҸӊҲҸҝҾӀӂѹѽҡҾ ҡҸқӃҲѹҾѽ. ғ ҝҾҦѦ ҰѦҝѦӊҸҦҸӊ ҲҽҔңѦ ҏҰҸҲҝѦқґҾӀҲҡҾ ҙҔҡҰґѹ҃ѹҲӁӃѦ ҰҸҔҽӉҽҲѦґѹ ҰҸӊҲҸҝҾѦѸҸҡҲѹ. ғ ҰҝѹѸѦҝҽѽ ҸӁѹ ҰҸҦҍѦҝҔӁҏҲӃ. ғ Ҹқӂѹѽ ҍѦҝҲҽѽ ӊӃқҸҝ ӍҸҝѸ ҰҸҦҲӊѦҝңҦҽѦҲ ҡҔҽӉҽӁӁҸѦ ӊӃҗѦ. ғ ҝҏҡҡҔҸѸ ѹ ҏҔҝҽѹӁҡҔҸѸ ҰѦҝѦӊҸҦҽѽ ҏҰҸҲҝѦқґѦӁҸ ҰҝҸҗѦҦҗѦѦ ӊҝѦѸҾ Ѣҋғ ӊ ҸқѦѹѽ ҍҽҡҲҾѽ, ҍҲҸ ѽҽҝҽҔҲѦҝӁҸ ҦґҾ ӊҸҡҲҸҍӁҸҡґҽӊҾӁҡҔѹѽ ҾӉӃҔҸӊ. 29 ғ ҍѦҗҡҔҸѸ, ҡґҸӊҽ҃ҔҸѸ ѹ ҡґҸӊѦӁҡҔҸѸ, ӁҽҸқҸҝҸҲ, ҏҰҸҲҝѦқґҾӀҲҡҾ ҲҸґҧҔҸ ӍҸҝѸӃ ҋғ. ѨѸѦӁӁҸ ӊ ҙҲѹѽ ҾӉӃҔҽѽ ӊҡѦ ҰѦҝѦӊҸҦӃ ҡҸҦѦҝңҽҲ ҲҽҔңѦ ҙҔҡҰґѹ҃ѹҲӁӃҮ ҰҸҔҽӉҽҲѦґҧ ҰҸӊҲҸҝҾѦѸҸҡҲѹ, ӊ ҲҸ ӊҝѦѸҾ ҔҽҔ ӊ ҝҏҡҡҔҸѸ ѹ ҏҔҝҽѹӁҡҔҸѸ, ҸӁѹ ҸҲҡҏҲҡҲӊҏӀҲ. ҶҲҸ ӊҰҸґӁѦ ҰҸӁҾҲӁҸ, ҰҸҡҔҸґҧҔҏ ҏңѦ ҡҽѸ ӊӃқҸҝ Ѣҋғ Ұҝѹ ҦҽӁӁӃѽ ҰҝѦҦѦґҧӁӃѽ ҊґҽҊҸґҽѽ ҦҸҡҲҽҲҸҍӁҸ ҾҡӁҸ ҡѹҊӁҽґѹӉѹҝҏѦҲ ҰҸӊҲҸҝҾѦѸҸҡҲҧ. ѨӁҲѦҝѦҡӁҸ ҸҲѸѦҲѹҲҧ, ҍҲҸ ӊ ҝҏҡҡҔҸѸ ҰѦҝѦӊҸҦѦ ҲҽҔңѦ ҰҝѦҦҡҲҽӊґѦӁ ҋғ, ӁҸ ҲҸґҧҔҸ ӁѦ ӊ ґѹҍӁҸҮ ӍҸҝѸѦ ҊґҽҊҸґҽ. ѨҡҰҸґҧӉҸӊҽӁӁҽҾ ӉҦѦҡҧ ҔҸӁҡҲҝҏҔ҃ѹҾ, ҡҸҡҲҸҾӂҽҾ ѹӉ ґѹҍӁҸҮ ӍҸҝѸӃ Ѣҋғ ѸҸҦҽґҧӁҸҊҸ ҊґҽҊҸґҽ ѹ ѹӁӍѹӁѹҲѹӊҽ ҋғ, ҰҸӉӊҸґҾѦҲ ҡӊҸѦҸқҝҽӉӁҸ ӊӃҝҽӉѹҲҧ ӊѹҦҸӊӃѦ ҔҽҍѦҡҲӊҽ ҸқҸѹѽ ҏҝҸӊӁѦҮ ҰҸӊҲҸҝҾӀӂѦҊҸҡҾ ҃ѦґҸҡҲӁҸҊҸ ҰҝѦҦѦґҧӁҸҊҸ ҦѦҮҡҲӊѹҾ: Ѣҋғ ґѹҍӁҸҮ ӍҸҝѸӃ ӫԙԗӾӴԂԊԂ ӊӃҝҽңҽѦҲ ӁѦҰҝѦҦѦґҧӁӃҮ ѽҽҝҽҔҲѦҝ ѸҽҔҝҸҡѹҲҏҽ҃ѹѹ, ҽ ҋғ ѹӁӍѹӁѹҲѹӊҽ ӴՂԗӢԋԐӠԽӸ ӊӃҝҽңҽѦҲ ҰҝѦҦѦґҧӁҸҡҲҧ ѹ ҃ѦґҸҡҲӁҸҡҲҧ Ӂҽ ҏҝҸӊӁѦ ѸѹҔҝҸҡѹҲҏҽ҃ѹѹ. ҘҝҸѸѦ ҊґҽҊҸґҽ ӫԙԗӾӴԂԽӸi ӊ ҰҸҦҸқӁӃѽ ҡґҏҍҽҾѽ җѹҝҸҔҸ ҰҝѦҦҡҲҽӊґѦӁ ҊґҽҊҸґ ԙԽӹӠԽӸi. ѨѸѦӁӁҸ ҙҲҸҲ ӊҽҝѹҽӁҲ ӊӃқҝҽґ ӊ ҡӊҸѦѸ ҰѦҝѦӊҸҦѦ Ҍ. ѢѦҡҲѦҝѦӁҔҸ: (13ҽ) [...] ҡҲҸѹґҸi ѦҮ ӊӃҰҝҾѸѹҲҧp ѦѸҏ җѦӀ [...] ӪԂӪ ӍґҽѸѹӁҊҸ ӠԒԄӠԼԂԊi җѦӀ ѹ ԒԂԄԊԋԌՂӴԂԊi ѦҮ ӊ ґѹ҃Ҹ. ѭѦґҽӁѹѦ ӁҽҮҲѹ ӊҸӉѸҸңӁҸҡҲҧ ӊҔґӀҍѹҲҧ ӍҸҝѸҏ ҋғ, ҔҸҲҸҝӃҮ ґҏҍҗѦ ҡҸҸҲӊѦҲҡҲӊҏѦҲ ѽҽҝҽҔҲѦҝҏ ѸѹҔҝҸҡѹҲҏҽ҃ѹѹ, ҸҲҝҽңҽѦҲҡҾ ҲҽҔңѦ ӊ ҡґѦҦҏӀӂѹѽ ҝҏҡҡҔѹѽ ҰѦҝѦӊҸҦҽѽ ҦҽӁӁҸҊҸ ҰҝҸѹӉӊѦҦѦӁѹҾ: (13қ) [...] ҔҽҔ ҲҸґҧҔҸ ҸӁҽ ӊӃҰҝҾѸґҾґҽi ѦѸҏ җѦӀ [...] – ӍґҽѸѹӁҊҸ ӰӾԽ-ӰӾԽ, ԌԂ Ӡ ӴՂԄӰӾԽԙԋp [pres.] ӁҽӉҽҦ, ԄԊԋԌԋi30 ѦҮ ӊ ґѹ҃Ҹ (ѢҽқҸҔҸӊ); (13ӊ) [...] ҲҸґҧҔҸ ҡҏӁѦҲp [pres.] ѦҊҸ ӊӁѹӉ ҊҸґҸӊҸҮ ҰҸҦ ѸӃҗҔҏ [...] ӪԂӪ ҸӁ ӠԒӹԄӰӾԽp [pres.] җѦӀ ѹ ԗӹԄԊԋԌӠԽp ѦҮ ҰҝҾѸҸ ӊ ҊґҽӉҽ (ѩѦѸҏҝҸӊҽ). 29 ғ ѹѸѦӀӂѦѸҡҾ ҏ Ӂҽҡ қѦґҸҝҏҡҡҔҸѸ ҰѦҝѦӊҸҦѦ ҸқѦ ҍҽҡҲѹ ҡҸҦѦҝңҽҲ ӍҸҝѸӃ ҋғ: ԋӪ ԽӹԊӸӪӬ ԣԊӾԙԋ ӴՂԗӢԂԙԽԂԊԂp ԌԂ ӰԂԊӾԞӰԂԄԂ ԙԽԂӰӫ ԗԽӫԶՂӰӫԇ ԶՂԇ [...] ԗԽӫԶӪԂ ӊӃҊӁҏґҽҡҾp [...] ҶҲҸ ӁҽҡҲҸґҧҔҸ ӁѦ ҡҸҸҲӊѦҲҡҲӊҏѦҲ ҦҝҏҊѹѸ ѹѸѦӀӂѹѸҡҾ ҏ Ӂҽҡ ҰҝѹѸѦҝҽѸ ѹӉ ҙҲҸҊҸ ҾӉӃҔҽ, ҍҲҸ ѸӃ ѸҸңѦѸ ҙҲҸ ҸқӆҾҡӁѹҲҧ ҲҸґҧҔҸ ҲѦѸ, ҍҲҸ ҰѦҝѦӊҸҦҍѹҔ ӁѦ ҏҍѦґ, ҍҲҸ ӉҦѦҡҧ ҝѦҍҧ ѹҦѦҲ Ҹ ҰҸӊҲҸҝҾӀӂѹѽҡҾ ҦѦҮҡҲӊѹҾѽ. ҭҸҙҲҸѸҏ ҦҽӁӁӃҮ қѦґҸҝҏҡҡҔѹҮ ӊҽҝѹҽӁҲ ӊ Ҳҽқґѹ҃Ѧ ӁѦ ҰҝѦҦҡҲҽӊґѦӁ. 30 ҭҸҡҝѦҦҡҲӊҸѸ ҦѦѦҰҝѹҍҽҡҲѹҾ Ѣҋғ ҦҽӁӁҸѦ ҦѦҮҡҲӊѹѦ ӊ ҙҲҸѸ ҰѦҝѦӊҸҦѦ ҰҝѦҦҡҲҽӊґѦӁҸ ӁѦ ҔҽҔ ҡґѦҦҏӀӂѹҮ җҽҊ ӊ ҃ѦҰѹ ҡҸқӃҲѹҮ, ҽ ҔҽҔ ҡҸҰҝҸӊҸңҦҽӀӂѦѦ ҸқҡҲҸҾҲѦґҧҡҲӊҸ.

ҭҫҋһѣѩҫғүѷѣһѻѢҫҋѷѻ ѩѣҢҋѷғѨҢ ҭҩѨ ҭҫғѷҫҩѤѣҴҫҋѷѨ

19

҈ҦѦҡҧ ҰѦҝѦӊҸҦҍѹҔѹ ӊ ҝҽӉґѹҍӁҸҮ ҡҲѦҰѦӁѹ ҰҝѹқѦҊҽӀҲ Ҕ ҡҲѹґѹҡҲѹҍѦҡҔѹ ѸҽҝҔѹҝҸӊҽӁӁҸѸҏ ҏҰҸҲҝѦқґѦӁѹӀ ҰҝѦӉѦӁҡҽ ҋғ. ҘҽҔ қӃґҸ ҏңѦ ҸҲѸѦҍѦӁҸ ӊ Ұҽҝ. 2, ҙҲҽ ӍҸҝѸҽ җѹҝҸҔҸ ҏҰҸҲҝѦқґҾѦҲҡҾ ӊ ҦҽӁӁҸѸ, ӁҽҊґҾҦӁҸ-ҰҝѹѸѦҝӁҸѸ ӉӁҽҍѦӁѹѹ. ҫҦӁҽҔҸ ҡґѦҦҏѦҲ ҸҲѸѦҲѹҲҧ, ҍҲҸ ӊ ҝҽҡҡѸҽҲҝѹӊҽѦѸҸѸ ӁҽѸѹ ѸҽҲѦҝѹҽґѦ, ҡҸҡҲҸҾӂѦѸ ѹӉ ҡґҸңӁҸҰҸҦҍѹӁѦӁӁӁӃѽ ҰҝѦҦґҸңѦӁѹҮ, ҸҰѹҡӃӊҽӀӂѹѽ ҰҸӊҲҸҝҾѦѸҸҡҲҧ ӊ ҰҝҸҗґҸѸ, ҰҸҦҸқӁӃѦ ҡґҏҍҽѹ ӊҡҲҝѦҍҽӀҲҡҾ ҝѦҦҔҸ. ғ ҰҝѹѸѦҝѦ (13) ҰҸґҧҡҔѹҮ ҰѦҝѦӊҸҦ ҸҲґѹҍҽѦҲҡҾ ӊ ӊѹҦҸӊҸѸ ҸҲӁҸҗѦӁѹѹ ҸҲ ҝҏҡҡҔҸҊҸ: ӊ ҰҝѹҦҽҲҸҍӁҸҮ ҍҽҡҲѹ ѹҡҰҸґҧӉҸӊҽӁҽ ӍҸҝѸҽ ҰҝҸҗѦҦҗѦҊҸ ӊҝѦѸѦӁѹ ҋғ. ғ ҝҏҡҡҔҸѸ ҾӉӃҔѦ ҲҽҔҸҮ ӊӃқҸҝ ӁѦҦҸҰҏҡҲѹѸ, ӁҸ ҸӁ ҦҸҡҲҽҲҸҍӁҸ ѽҽҝҽҔҲѦҝѦӁ ҦґҾ ҰҸґҧҡҔҸҊҸ ҾӉӃҔҽ. ғ ӁҽҗѦѸ ѸҽҲѦҝѹҽґѦ ҰҸґҧҡҔѹѦ ҰҝѹѸѦҝӃ ҦҽӁӁҸҊҸ ҲѹҰҽ (ҋғ–Ѣҋғ) ӊҡҲҝѦҍҽӀҲҡҾ ӉӁҽҍѹҲѦґҧӁҸ ҍҽӂѦ, ҍѦѸ ҲѹҰ Ѣҋғ–Ѣҋғ, ҰҝѦҦҡҲҽӊґѦӁӁӃҮ ӊ ҰҝѦҦӃҦҏӂѦѸ ҰҝѹѸѦҝѦ (12). 31 ғҽңӁҸ, ҸҦӁҽҔҸ, ҸҲѸѦҲѹҲҧ, ҍҲҸ ӊ ҊґҽӊӁҸҮ ҍҽҡҲѹ ҰҸҦҸқӁӃѽ ҰҝѦҦґҸңѦӁѹҮ ӊӃқҸҝ ӊѹҦҽ ӊ ҰҸґҧҡҔҸѸ ҾӉӃҔѦ ӁѦ ҸҲґѹҍҽѦҲҡҾ ӉҽѸѦҲӁҸ ҸҲ ҝҏҡҡҔҸҊҸ. ғ ҸҲґѹҍѹѦ ҸҲ ҰҝѹѸѦҝҽ (12), ӊ ѽҸҝӊҽҲҡҔҸѸ, ҡѦҝқҡҔҸѸ ѹ ѸҽҔѦҦҸӁҡҔҸѸ ҾӉӃҔҽѽ ѸҸҦҽґҧӁҽҾ ҔҸӁҡҲҝҏҔ҃ѹҾ ҋғ ѹҡҰҸґҧӉҏѦҲҡҾ ӁѦ ҲҸґҧҔҸ ӊ ҰҝѹҦҽҲҸҍӁҸҮ ҍҽҡҲѹ, ӁҸ ѹ ӊ ҊґҽӊӁҸҮ. ҶҲҸҲ ҲѹҰ җѹҝҸҔҸ ҰҝѦҦҡҲҽӊґѦӁ ѹѸѦӁӁҸ ӊ ѽҸҝӊҽҲҡҔѹѽ ѹ ҡѦҝқҡҔѹѽ ҰҝѹѸѦҝҽѽ ӁҽҗѦҊҸ ҔҸҝҰҏҡҽ. ҷҸґҊҽҝҡҔѹҮ ҰѦҝѦӊҸҦ ӊ ҰҝѹѸѦҝѦ (13) ҰҸ ҲѹҰҏ ҡҸҸҲӁҸҗѦӁѹҾ ӍҸҝѸ ҰҸґӁҸҡҲҧӀ ҡҸҸҲӊѦҲҡҲӊҏѦҲ ҰҝѹѸѦҝҏ (12). ҘҽҔ қӃґҸ ҏңѦ ҸҲѸѦҍѦӁҸ, ӊ қҸґҊҽҝҡҔҸѸ ҾӉӃҔѦ ҲѹҰ ҋғ–Ѣҋғ ҾӊґҾѦҲҡҾ ҸқӃҍӁӃѸ, ҡҲҽӁҦҽҝҲӁӃѸ. ҶҲҸ ҡӊҾӉҽӁҸ ҡ ҡҏӂѦҡҲӊҸӊҽӁѹѦѸ ӊ қҸґҊҽҝҡҔҸҮ ҊґҽҊҸґҧӁҸҮ ҡѹҡҲѦѸѦ ӍҸҝѸӃ ѹѸҰѦҝӍѦҔҲҽ ҋғ, ҡҰѦ҃ѹӍѹҍѦҡҔҽҾ ҡѦѸҽӁҲѹҔҽ ҔҸҲҸҝҸҮ ҸҡҸқҸ ҰҝѹҊҸҦӁҽ ҦґҾ ҏҰҸҲҝѦқґѦӁѹҾ ӊ ӉҽӊѹҡѹѸҸҮ ҍҽҡҲѹ ҰҸҦҸқӁӃѽ ҰҝѦҦґҸңѦӁѹҮ. ғ ӉҽҔґӀҍѦӁѹѦ ҙҲҸҊҸ ҰҽҝҽҊҝҽӍҽ ҰҝѦҦґҽҊҽѦѸ ҡґѦҦҏӀӂѹҮ ҰҝѹѸѦҝ. ҭѦҝѦӊҸҦӃ, ҰҝѦҦҡҲҽӊґѦӁӁӃѦ ӊ ӁѦѸ, ӊ ҸҡӁҸӊӁҸѸ, ҸҲґѹҍӁҸ ҸҲҝҽңҽӀҲ ҸқӂѹѦ ҲѦӁҦѦӁ҃ѹѹ, ҸҲѸѦҍѦӁӁӃѦ ӊӃҗѦ: (14) Rudyard Kipling: The Jungle Book үӁҊґ.

So as soon as Messua pronounced a word Mowgli would imitate it almost perfectly, [and before dark he had learned the names of many things in the hut].

31 ѨӁҲѦҝѦҡӁҸ ҸҲѸѦҲѹҲҧ, ҍҲҸ ӊ ҦҝҏҊҸѸ ҰҸґҧҡҔҸѸ ҰѦҝѦӊҸҦѦ ҏҰҸҲҝѦқґҾѦҲҡҾ ӊҡѦ ңѦ ӍҸҝѸҽ Ѣҋғ. ѢҸ ӉҦѦҡҧ ҰҝѦҦҡҲҽӊґѦӁҽ ѹѸѦӁӁҸ ҏҰҸѸѹӁҽѦѸҽҾ ӊӃҗѦ ҔҸӁҡҲҝҏҔ҃ѹҾ ҡ ѸҸҦҽґҧӁӃѸ ҊґҽҊҸґҸѸ, ҰҸӉӊҸґҾӀӂҽҾ ҏҰҸҲҝѦқґҾҲҧ Ҹқҽ ӊѹҦҽ ҸҦӁҸӊҝѦѸѦӁӁҸ: [...] gdy tylko udawa’oi jej siNj wyprostowaǀp mu szyjNj wraz z g’owƾ [...] wykrNjca’i jƾ do ty’u i spoglƾda’i w twarz dziewczynki (S’omczyǝski). ѢѦҸқѽҸҦѹѸҸ ҸҲѸѦҲѹҲҧ, ҍҲҸ ӊ ҰҸҦҸқӁӃѽ ҡґҏҍҽҾѽ ӊҡҲҝѦҍҽѦҲҡҾ ҲҽҔңѦ ҡҸҸҲӊѦҲҡҲӊҏӀӂѹҮ ҊґҽҊҸґ ҋғ – udaǀ siNj, ӊ ҲҸ ӊҝѦѸҾ ҔҽҔ ӊ ҝҏҡҡҔҸѸ ҾӉӃҔѦ ӊҸӉѸҸңѦӁ ҲҸґҧҔҸ ҊґҽҊҸґ Ѣҋғ ӫԌԂӴԂԽӸԙԋ. ҭҸ ѸӁѦӁѹӀ ӁҸҡѹҲѦґѦҮ ҾӉӃҔҽ ѹ ӊ ҰҝѹӊѦҦѦӁӁҸѸ ӍҝҽҊѸѦӁҲѦ ѸҸңӁҸ ґѦҊҔҸ ӉҽѸѦӁѹҲҧ udawa’oi siNj Ӂҽ uda’op siNj.

20

үѩҩѨүѢ ҷүҩѣѢѷҋѣѢ

ҩҏҡҡҔѹҮ

ҘҽҔ ҲҸґҧҔҸ ҴѦҡҡҏҽ ҰҝҸѹӉӁҸҡѹґҽi ҔҽҔҸѦ-ӁѹқҏҦҧ ҡґҸӊҸ, ҴҽҏҊґѹ ҸҍѦӁҧ ҰҸѽҸңѦ ԗӹӴԽӹӢԋԊi ѦҊҸ Ӊҽ ӁѦҮ [...] ҷѦґҸҝҏҡ. ѤҔ ҲҸґҧҔҟ ҴѦҡҏҽ ҊҽӊҽҝӃґҽi ҾҔҸѦ-ӁѦқҏҦӉҧ ҡґҸӊҽ, ҴҽҕҊґҟ ӊѦґҧѸҟ ҰҽҦҸқӁҽ ԗԂ԰ԽԂӢԂ԰i ҾҊҸ Ӊҽ ҺӀ [...] ѰҔҝ. [...] ҾҔ ҲҟґҧҔѹ ҴѦҡҡҏҽ ӊѹѸҸӊґҾґҽi ҾҔѦ-ӁѦқҏҦҧ ҡґҸӊҸ, ҴҽҏҊґҟ ԗӹӴԽӹӢԇӴԂӴi ҮҸҊҸ ҦҸҡѹҲҧ ҦҸқҝѦ [...] ҭҸґҧҡҔ. Ilekroǀ Messua wymówi’ap jakieǧ s’owo, on natychmiast powtarza’i je g’oǧno [...] ѯѦҗҡҔ. Sotva tedy proneslap Mésúa nLJjaké slovo, hned je skoro bez chyby napodobilp [...] (Skoumalovi). ҋґҸӊҽ҃Ҕ. Len ǁo Mesúa vyslovilap nejaké slovo, Mauglí ho napodobnilp takmer dokonale [...] ҋґҸӊѦӁҡҔ. Brž ko je Mesua izgovorilap kako besedo, jo je Mavgli skoraj brez napake ponovilp [...] ҤҸҝӊҽҲҡҔ. ƀim bi [cond.] Mesua izgovorilap koju rijeǁ, Mogli bi [cond.] je gotovo bez pogreške ponoviop [...] ҋѦҝқҡҔѹҮ Ѩ ӉҽҲҸ, ҍѹѸ қѹ [cond.] ҴѦҡҏҽ ѹӉҊҸӊҸҝѹґҽp ӁѦҔҏ ҝѦҍ, ҴҸҊґѹ ԼӠ [cond.] ҎѦ ҊҸҲҸӊҸ ҡҽӊҝҗѦӁҸ ԗӹӰӹӴӠӹp [...] ҴҽҔѦҦҸӁ. ҋҽѸҸ җҲҸ ҴѦҡҏҽ ѱѦ [exp.] ѹӉҊҸӊҸҝѦҗѦp [impf.] ѦҦѦӁ ӉқҸҝ, ҴҸҊґѹ ҊҸ ӠԐӠԽӠӢԂԶӾi [impf.] ҝѦҍѹҡѹ ҡҸӊҝҗѦӁҸ [...] ҷҸґҊҽҝҡҔ. ҋѦҊҽ ҦҸҡҲҽҲӆҍӁҸ қѦ [impf.] ҴѦҡҏҽ Ҧҽ ҰҝҸѹӉӁѦҡѦp [pres.] ѦҦӁҽ ҦҏѸҽ ѹ ҴҽҏҊґѹ Ҿ ԗӹӴԽԂӢԋԶӾi [impf.] ҰҸҍҲѹ қѦӉҊҝѦҗӁҸ. ҷҸґҊҽҝҡҔѹҮ ӊҽҝѹҽӁҲ ӁѦҡҔҸґҧҔҸ ҸҲґѹҍҽѦҲҡҾ ҸҲ ӁҽѹқҸґѦѦ ѽҽҝҽҔҲѦҝӁҸҊҸ ҲѹҰҽ. ғ ҦҽӁӁҸѸ ҡґҏҍҽѦ ҰҝѦҦҡҲҽӊґѦӁҽ ҔҸӁҡҲҝҏҔ҃ѹҾ ҲѹҰҽ ҝҏҡҡҔҸҊҸ ԙԽӹӠԊӹi + ѹӁӍѹӁѹҲѹӊ ҋғ. 32 ҘҽҔ қӃґҸ ҸҲѸѦҍѦӁҸ ӊӃҗѦ, ҙҲҽ ҔҸӁҡҲҝҏҔ҃ѹҾ ҲҽҔңѦ ӊӃҝҽңҽѦҲ ӊѹҦҸӊҸҮ ѽҽҝҽҔҲѦҝ Ҧӊҏѽ ҏҝҸӊӁѦҮ – ѸѹҔҝҸҡѹҲҏҽ҃ѹѹ ѹ ѸҽҔҝҸҡѹҲҏҽ҃ѹѹ. ѷѦѸ ҡҽѸӃѸ ҸӁҽ ҰҸ ҡѦѸҽӁҲѹҔѦ ҸҍѦӁҧ қґѹӉҔҽ Ҕ қҸґѦѦ ҏҰҸҲҝѦқѹҲѦґҧӁҸҮ ӊ ҦҽӁӁҸҮ ҰҸӉѹ҃ѹѹ ӍҸҝѸѦ ѹѸҰѦҝӍѦҔҲҽ ҋғ. ғ ӊӃҗѦҰҝѹӊѦҦѦӁӁҸѸ ҍѦҗҡҔҸѸ ҰѦҝѦӊҸҦѦ ҰҝѦҦҡҲҽӊґѦӁ ҲѹҰ ҋғ–ҋғ, ѽҽҝҽҔҲѦҝӁӃҮ ҦґҾ ҦҽӁӁҸҊҸ ҾӉӃҔҽ. ҫҦӁҽҔҸ ѹӁҲѦҝѦҡӁҸ ҸҲѸѦҲѹҲҧ, ҍҲҸ ӊ ҦҝҏҊҸѸ ҍѦҗҡҔҸѸ ҰѦҝѦӊҸҦѦ ӊ ҊґҽӊӁҸҮ ҍҽҡҲѹ ҏҰҸҲҝѦқґҾѦҲҡҾ Ѣҋғ: (14ҽ) Jakmile Messua vyslovilap slovo, opakovali je Maugli témLJǥ bezvadnLJ (Maixner). ҭҸ ѸӁѦӁѹӀ ӁҸҡѹҲѦґҾ ҾӉӃҔҽ, ҲҏҲ, ӊҰҝҸҍѦѸ, қӃґҽ қӃ ӊҸӉѸҸңӁҽ ҲҽҔңѦ ӍҸҝѸҽ ҋғ: zopakoval. ѢҽҡҔҸґҧҔҸ ѸӃ ѸҸңѦѸ ҡҏҦѹҲҧ, ӊ ҍѦҗҡҔҸѸ ҾӉӃҔѦ ӊӃқҸҝ Ѣҋғ 32

ѨӉ-Ӊҽ ҸҲҡҏҲҡҲӊѹҾ ѹӁӍѹӁѹҲѹӊҽ ӊ қҸґҊҽҝҡҔҸѸ ҾӉӃҔѦ ӉҦѦҡҧ, ҔҸӁѦҍӁҸ, ҏҰҸҲҝѦқґҾѦҲҡҾ ѽҽҝҽҔҲѦҝӁҽҾ ӀҊҸ-ӊҸҡҲҸҍӁҸ-ҡґҽӊҾӁҡҔҽҾ ҔҸӁҡҲҝҏҔ҃ѹҾ ԌԂ + ҰҝѦӉѦӁҡ.

ҭҫҋһѣѩҫғүѷѣһѻѢҫҋѷѻ ѩѣҢҋѷғѨҢ ҭҩѨ ҭҫғѷҫҩѤѣҴҫҋѷѨ

21

ӊ ҊґҽӊӁҸҮ ҍҽҡҲѹ ҰҸҦҸқӁӃѽ ҰҝѦҦґҸңѦӁѹҮ ҰҸӊҲҸҝҾѦѸҸҡҲҧӀ ҸҡҸқҸ ӁѦ ѸҸҲѹӊѹҝҏѦҲҡҾ (ҔҽҔ, ӁҽҰҝѹѸѦҝ, ӊ ҝҏҡҡҔҸѸ ѹ ҰҸґҧҡҔҸѸ ҾӉӃҔҽѽ). ғ ҦҽӁӁҸѸ ҡґҏҍҽѦ Ѣҋғ ҡҔҸҝѦѦ ӊҡѦҊҸ ҰҸҦҍѦҝҔѹӊҽѦҲ ҰҝҸ҃ѦҡҡӁҸҡҲҧ ѹґѹ ҸҲҲѦӁҸҔ “ҰҝѹҡҲҏҰ Ҕ ҦѦҮҡҲӊѹӀ”. ғ ӁҽҏҍӁҸҮ ґѹҲѦҝҽҲҏҝѦ ҸҲѸѦҍѦӁҸ, ҍҲҸ ӊ ҍѦҗҡҔҸѸ ҾӉӃҔѦ ҲҽҔѹѦ ӍҽҔҲҸҝӃ ѹҊҝҽӀҲ ҲҽҔңѦ ӁѦѸҽґҸӊҽңӁҏӀ ҝҸґҧ Ұҝѹ ҸқҸӉӁҽҍѦӁѹѹ ӁѦҰҸӊҲҸҝҾӀӂѦҮҡҾ ҃ѦҰҸҍҔѹ ҦѦҮҡҲӊѹҮ. 33 4. ҭҝѹҦҽҲҸҍӁӃѦ ҰҝѦҦґҸңѦӁѹҾ ҡ ҦҝҏҊѹѸѹ ҡҸӀӉҽѸѹ 34 ѢѦ ӊҸ ӊҡѦѽ ҰѦҝѦӊҸҦҽѽ ҰҝѦҦӃҦҏӂѦҊҸ ҰҽҝҽҊҝҽӍҽ қӃґѹ ѹҡҰҸґҧӉҸӊҽӁӃ ҡҸӀӉӃ ҲѹҰҽ ӪԂӪ ԽӹԊӸӪӹ. ѢҽҰҝѹѸѦҝ, ҽӁҊґѹҮҡҔѹҮ ҡҸӀӉ just ӊ ҰҝѹѸѦҝѦ (13), Ӂҽ ҰҸґҧҡҔѹҮ, ѽҸҝӊҽҲҡҔѹҮ, ҡѦҝқҡҔѹҮ ѹ қҸґҊҽҝҡҔѹҮ ҰѦҝѦӊҸҦѹҲҡҾ ҡҸӀӉҽѸѹ, ҸҡҸқҸ ӁѦ ӊӃҝҽңҽӀӂѹѸѹ ҡҰѦ҃ѹӍѹҍѦҡҔҏӀ ҡѦѸҽӁҲѹҔҏ (ѸѹӁѹѸҽґҧӁӃҮ ѹӁҲѦҝӊҽґ, қӃҡҲҝҽҾ ҰҸҡґѦҦҸӊҽҲѦґҧӁҸҡҲҧ) ҽӁҊґѹҮҡҔҸҊҸ ҡҸӀӉҽ. ғ ѸҽҲѦҝѹҽґѦ ҰҽҝҽґґѦґҧӁҸҊҸ ҔҸҝҰҏҡҽ ҰҸҦҸқӁӃѦ ҡґҏҍҽѹ ӁѦҡҸҸҲӊѦҲҡҲӊѹҾ ӊ ӊӃқҸҝѦ ҡҸӀӉҸӊ ӁҽқґӀҦҽӀҲҡҾ ӁѦҝѦҦҔҸ. ҫҍѦӊѹҦӁҸ, ҙҲҸ ҡӊҾӉҽӁҸ ҡ ҲѦѸ, ҍҲҸ ҡӍѦҝӃ ҏҰҸҲҝѦқґѦӁѹҾ ҡҸӀӉҸӊ ҝҽӉӁӃѽ ҾӉӃҔҸӊ ӉҽѸѦҲӁҸ ҝҽӉґѹҍҽӀҲҡҾ ҦҽңѦ ӊ ҲѦѽ ҡґҏҍҽҾѽ, ҔҸҊҦҽ ҙҲѹ ҡҸӀӉӃ ӊ ҡґҸӊҽҝҾѽ ҰҝѹӊҸҦҾҲҡҾ ҔҽҔ қґѹӉҔѹѦ ҡҸҸҲӊѦҲҡҲӊѹҾ. ҭҸ ӁҽҗѦѸҏ ѸӁѦӁѹӀ, ҸҦӁҸҮ ѹӉ ҰҝѹҍѹӁ ҲҽҔѹѽ ҝҽӉґѹҍѹҮ ѸҸңѦҲ қӃҲҧ ҲҸ, ҍҲҸ ҲѦ ѹґѹ ѹӁӃѦ ҡѦѸҽӁҲѹҍѦҡҔѹѦ ҰҝѹӉӁҽҔѹ, ҸҰҝѦҦѦґҾӀӂѹѦ ӉӁҽҍѦӁѹѦ ҰҸҦҸқӁӃѽ ҡҸӀӉҸӊ, ҰҝѦҦҡҲҽӊґѦӁӃ ӊ ҡҸӀӉҽѽ ҝҽӉӁӃѽ ҾӉӃҔҸӊ ӁѦҸҦѹӁҽҔҸӊҸ ҡѹґҧӁҸ. 35 ғ ҦҽӁӁҸѸ ҰҽҝҽҊҝҽӍѦ ҝҽҡҡѸҽҲҝѹӊҽӀҲҡҾ ӊ ҸҡӁҸӊӁҸѸ ҰҝѹѸѦҝӃ ҡ ҡҸӀӉҽѸѹ ҲѹҰҽ ӪӹԄԌԂ. ѢҸ ӊ ӁѦҔҸҲҸҝӃѽ ҰѦҝѦӊҸҦҽѽ ӊѸѦҡҲҸ Ӂѹѽ ҏҰҸҲҝѦқґҾӀҲҡҾ ҡҸӀӉӃ ҲѹҰҽ ӾԙԊӠ, Ҳ.Ѧ. ҡҸӀӉӃ, ҔҸҲҸҝӃѦ, ӊ ҰҝѹӁ҃ѹҰѦ, ҸҲӁҸҡҾҲҡҾ Ҕ Ҕґҽҡҡҏ ҏҡґҸӊӁӃѽ ҡҸӀӉҸӊ. ҶҲҸ ҏҔҽӉӃӊҽѦҲ Ӂҽ ҲҸ, ҍҲҸ ҡѦѸҽӁҲѹҍѦҡҔѹѦ ҝҽӉґѹҍѹҾ ѸѦңҦҏ ҙҲѹѸѹ ҡҸӀӉҽѸѹ ӊ ҝҾҦѦ ҡґҏҍҽѦӊ ҡҲҽӁҸӊҾҲҡҾ ѸѦӁѦѦ ҸӂҏҲѹѸӃѸѹ. ҫҡҸқѦӁӁҸ ҍҽҡҲҸ ҙҲҸ ӁҽқґӀҦҽѦҲҡҾ ѹѸѦӁӁҸ Ұҝѹ ҰҸӊҲҸҝҾѦѸҸҡҲѹ. Ѣҽ ҡѦѸҽӁҲѹҍѦҡҔҏӀ қґѹӉҸҡҲҧ ҦҽӁӁӃѽ ҡҸӀӉҸӊ ҏҔҽӉӃӊҽѦҲ ѦӂѦ ѹ ҲҸҲ ӍҽҔҲ, ҍҲҸ ӊ ҝҾҦѦ ҾӉӃҔҸӊ ҸқҸѹѸ ҝҏҡҡҔѹѸ ҡҸӀӉҽѸ ҡҸҸҲӊѦҲҡҲӊҏѦҲ ҸҦѹӁ ѹ ҲҸҲ ңѦ ҡҸӀӉ. 36

33

ҋѸ. ҝҽқҸҲӃ ҋҲҏӁҸӊҸҮ (Stunová 1988, 1993: 105–167) ѹ ѩѹҔѹ (Dickey 2000: 203–218) ѹ ҏҔҽӉҽӁӁҏӀ ҲҽѸ қҸґѦѦ ҝҽӁӁӀӀ ґѹҲѦҝҽҲҏҝҏ. 34 ҴӃ ӁҽѸѦҝѦӊҽґѹҡҧ ҝҽҡҡѸҸҲҝѦҲҧ ӉҦѦҡҧ ҲҽҔңѦ ҰҝѹѸѦҝӃ ҡ ҡҸӀӉҸѸ ҲѹҰҽ ԗӹԙԊӾ ԽӹԄӹ ӪԂӪ, ӁҸ ҸҔҽӉҽґҸҡҧ, ҍҲҸ Ұҝѹ ҸқҸӉӁҽҍѦӁѹѹ ҰҸӊҲҸҝҾѦѸҸҡҲѹ ӊ ҰҝҸҗґҸѸ ҸӁѹ ӊҡҲҝѦҍҽӀҲҡҾ ҔҝҽҮӁѦ ҝѦҦҔҸ. ҭҸҙҲҸѸҏ ѸӃ ҰҸҔҽ ҸҡҲҽӊґҾѦѸ ѹѽ қѦӉ ӊӁѹѸҽӁѹҾ. 35 ҋѸ. ҡҸҰҸҡҲҽӊѹҲѦґҧӁӃҮ ҽӁҽґѹӉ ҸҊҝҽӁѹҍѹҲѦґҧӁӃѽ ҡҸӀӉҸӊ ҡѦҝқҡҔҸҊҸ, ҝҏҡҡҔҸҊҸ ѹ ҰҸґҧҡҔҸҊҸ ҾӉӃҔҸӊ (ҡҸҸҲӊѦҲҡҲӊѦӁӁҸ ԌӹӪ (ӰӾ), ԗӹӪԂ (ӰӾ) ѹ (do)póki (nie)) ӊ ҝҽқҸҲѦ (ҷҽҝѦӁҲҡѦӁ 2007). 36 ѢҽҰҝѹѸѦҝ, ӊ ѹҡҰҸґҧӉҏѦѸӃѽ ӁҽѸѹ ҦӊҏҾӉӃҍӁӃѽ ҡґҸӊҽҝҾѽ қѦґҸҝҏҡҡҔѹҮ ҡҸӀӉ ӪԂԊӬ, ҏҔҝҽѹӁҡҔѹҮ ӪӹԊӠ ѹ ҍѦҗҡҔѹҮ kdyzև ҰѦҝѦӊҸҦҾҲҡҾ ҔҽҔ ҡҸӀӉҸѸ ӪӹԄԌԂ, ҲҽҔ ѹ ҡҸӀӉҸѸ ӾԙԊӠ. ѢѹҦѦҝґҽӁҦҡҔѹҮ ҡҸӀӉ als ҲҽҔңѦ ӊӃҝҽңҽѦҲ ӁѦ ҲҸґҧҔҸ ҏҡґҸӊӁҸѦ, ӁҸ ѹ ӊҝѦѸѦӁӁҸѦ ӉӁҽҍѦӁѹѦ. ҭҝѹҲҸѸ ӊ ҰҸҡґѦҦӁѦѸ ҡґҏҍҽѦ ҸӁ Ұҝѹ ҸқҸӉӁҽҍѦӁѹѹ ҦѦҮҡҲӊѹҮ ӊ ҰҝҸҗґҸѸ ҦҸҰҸґӁѹҲѦґҧӁҸ ӊӃҝҽңҽѦҲ ҰҸӊҲҸҝҾѦѸҸҡҲҧ. (ҭҝѹ ѦҦѹӁѹҍӁӃѽ ҦѦҮҡҲӊѹҾѽ ӊ ҰҝҸҗґҸѸ ҏҰҸҲҝѦқґҾѦҲҡҾ toen.)

22

үѩҩѨүѢ ҷүҩѣѢѷҋѣѢ

ғ Ұҽҝ. 1 қӃґҸ ҸҲѸѦҍѦӁҸ, ҍҲҸ ҲҽҔҡѹҡӁӃѦ ҸҲӁҸҗѦӁѹҾ ҰҝѦҦҗѦҡҲӊҸӊҽӁѹҾ ѹ ҸҦӁҸӊҝѦѸѦӁӁҸҡҲѹ ҡҽѸѹѸ ҡҸӀӉҸѸ ӪӹԄԌԂ ӁѦ ҦѹӍӍѦҝѦӁ҃ѹҝҏѦҲҡҾ. ҶҲҸ ҸҲӁҸҡѹҲҡҾ ҲҽҔңѦ Ҕ ҡҸӀӉҏ ӾԙԊӠ ѹ Ҕ ҡҸҸҲӊѦҲҡҲӊҏӀӂѹѸ ҡҸӀӉҽѸ ӊ ҦҝҏҊѹѽ ҡґҽӊҾӁҡҔѹѽ ҾӉӃҔҽѽ. ҭҸҙҲҸѸҏ Ұҝѹ ҦҽӁӁӃѽ ҡҸӀӉҽѽ ӊӃқҸҝ ӊѹҦҸӊӃѽ ӍҸҝѸ, ӊ ҰҝѹӁ҃ѹҰѦ, ѸҸңѦҲ ѹҊҝҽҲҧ ӊҽңӁҏӀ ҝҸґҧ Ұҝѹ ҝҽӉґѹҍѦӁѹѹ ҙҲѹѽ Ҧӊҏѽ ҡѸӃҡґҸӊ. ғ ҡґѦҦҏӀӂѦѸ ҰҝѹѸѦҝѦ ѹӉ ҰҸӊѦҡҲѹ ԠԂԊӾӰӸӪӠӡ ԗӢӠӰՁ ҽӊҲҸҝ ҝҽҡҡҔҽӉӃӊҽѦҲ Ҹ ҲҸѸ, ҍҲҸӿ ҸқӃҍӁҸ ҰҝҸѹҡѽҸҦѹґҸ Ұҝѹ ѦҊҸ ӊҡҲҝѦҍҽѽ ҡҸ ӊӉҝҸҡґӃѸѹ ґӀҦҧѸѹ. (15) Antoine de Saint-Exupéry: Le petit prince ҨҝҽӁ҃.

Quand j’en rencontrais une [qui me paraissait un peu lucide], je faisais l'expérience sur elle de mon dessin numéro 1 [que j’ai toujours conservé]. ҩҏҡҡҔѹҮ ҘҸҊҦҽ Ҿ ӊҡҲҝѦҍҽґi ӊӉҝҸҡґҸҊҸ [...] Ҿ ԗӹӪԂԒՂӴԂԊi ѦѸҏ ҡӊҸҮ ҝѹҡҏӁҸҔ ҥ 1 [...] (Ґҽґҧ). ҷѦґҸҝҏҡ. Ҙҽґҟ Ҿ ҡҏҡҲҝҽҔҽҕi ҔҽҊҸ-ӁѦқҏҦӉҧ Ӊ ҟѽ ҽҡҾҝҸҦҦӉҾ [...] Ҿ ԗԂӪԂԒӴԂ԰i ҾѸҏ ѸҽґӀӁҽҔ ҥ 1 [...] ѰҔҝ. ҘҸґѹ ѸѦӁҟ ҲҝҽҰґҾӊҡҾi ҦҸҝҸҡґѹҮ [...] Ҿ ՁӬӪԂӴӠӴԙԋi ҮҸҊҸ ҦҏѸҔҸӀ ҰҝҸ ҡӊҟҮ ѸҽґӀӁҸҔ ҥ 1 [...] ҭҸґҧҡҔ. Gdy spotyka’emi doros’ƾ osobNj [...] robi’emi na niej doǧwiadczenie z moim rysunkiem numer 1 [...] (Szwykowski). ѯѦҗҡҔ. Když jsem mezi nimi potkalp nLJkoho [...] ovLJǥilp jsem si na nLJm svou zkušenost s kresbou ǁíslo 1 [...] ҋґҸӊҽ҃Ҕ. KeDŽ som stretolp niekoho [...] skúsilp som to uǞho s kresbou ǁíslo jeden [...] ҋґҸӊѦӁҡҔ. Kadar sem sreǁalp koga [...] sem ga preizkusilp s sliko številka 1 [...] ҤҸҝӊҽҲҡҔ. Kad bih [cond.] sreop odrasloga [...] okušaop bih [cond.] na njemu svoj crtež br. 1 [...] (Kušan). ҋѦҝқҡҔѹҮ Kada bih sreop [cond.] nekog [...] proveravaoi sam ga svojim crtežom broj 1 [...] (Vukšiǀ). ҴҽҔѦҦҸӁ. ҘҸҊҽ ѱѦ [exp.] ҡҝѦҲӁѦӊp [impf.] ӁѦҔҸҎ [...] ԙӾ ӹԼӠԌӫӴԂӴi [impf.] ҔҽҎ ӁѦҊҸ ҡҸ ѸҸҎҸҲ ҃ҝҲѦң қҝҸҎ 1 [...] ҷҸґҊҽҝҡҔ. ҋҝѦӂӁѦѽp [impf.] ґѹ ӊӆӉҝҽҡҲѦӁ [...] ҊҸ ԗӢӹӴӾӢԋӴԂԖi [impf.] ҡ ѸҸҾҲҽ ҝѹҡҏӁҔҽ ӁҸѸѦҝ 1 [...] ҭҸҍҲѹ ӊҡѦ ҰҝѹӊѦҦѦӁӁӃѦ ҰѦҝѦӊҸҦӃ ѹҡҰҸґҧӉҏӀҲ ӉҦѦҡҧ ҡҽѸӃѦ ӁѦҮҲҝҽґҧӁӃѦ ҡҸӀӉӃ. ҫҲґѹҍҽѦҲҡҾ ҲҸґҧҔҸ қҸґҊҽҝҡҔѹҮ. ѨҡҰҸґҧӉҸӊҽӁӁҽҾ ӊ ӁѦѸ ҔҸӁҡҲҝҏҔ҃ѹҾ ҰҸ ҡѸӃҡґҏ қґѹңѦ Ҕ ҝҏҡҡҔҸѸҏ ӾԙԊӠ. ѢҸ Ӂҽ ҸҡӁҸӊѦ ҦҝҏҊѹѽ ҡѽҸҦӁӃѽ ҰҝѹѸѦҝҸӊ ѸӃ ҰҝѦҦҰҸґҽҊҽѦѸ, ҍҲҸ ӊ ҦҽӁӁҸѸ ҡґҏҍҽѦ ӊҰҸґӁѦ ҏѸѦҡҲѦӁ қӃґ қӃ ҲҽҔңѦ ӁҽѹқҸґѦѦ ӁѦҮҲҝҽґҧӁӃҮ ӊҝѦѸѦӁӁҸҮ ҡҸӀӉ: ԹӹԄԂԽӹ ԙӢӾԷӰӾԖ ...

ҭҫҋһѣѩҫғүѷѣһѻѢҫҋѷѻ ѩѣҢҋѷғѨҢ ҭҩѨ ҭҫғѷҫҩѤѣҴҫҋѷѨ

23

ғӃқҸҝ ҊґҽҊҸґҧӁӃѽ ӍҸҝѸ ӉҦѦҡҧ ҰҝѹӁ҃ѹҰѹҽґҧӁҸ ӁѦ ҸҲґѹҍҽѦҲҡҾ ҸҲ ҲҸҊҸ, ҍҲҸ ѸӃ ӁҽѽҸҦѹѸ ӊ ѸҽҲѦҝѹҽґѦ ҰҝѦҦӃҦҏӂѦҊҸ ҰҽҝҽҊҝҽӍҽ. ғ ҦҽӁӁҸѸ ҡґҏҍҽѦ ӊ ҔҸѸѸѦӁҲҽҝѹѹ ӁҏңҦҽӀҲҡҾ ҲҸґҧҔҸ ҰѦҝѦӊҸҦӃ Ӂҽ ҰҸґҧҡҔѹҮ ѹ ҡѦҝқҡҔѹҮ ҾӉӃҔѹ. ѢѦҡѸҸҲҝҾ Ӂҽ ӁҽҗѦ ҏҲӊѦҝңҦѦӁѹѦ, ҍҲҸ ӊ ҰҸґҧҡҔҸѸ ҾӉӃҔѦ Ұҝѹ ӊӃҝҽңѦӁѹѹ ҰҸҡґѦҦҸӊҽҲѦґҧӁҸҡҲѹ ҦѦҮҡҲӊѹҮ ӊ ҰҝѹҦҽҲҸҍӁҸҮ ҍҽҡҲѹ ӁҽѹқҸґѦѦ ҏҰҸҲҝѦқѹҲѦґҧӁҸҮ ҾӊґҾѦҲҡҾ ӍҸҝѸҽ ҋғ, ӊ ҦҽӁӁҸѸ ҰѦҝѦӊҸҦѦ ѹҡҰҸґҧӉҸӊҽӁҽ ӊҡѦ ңѦ ӍҸҝѸҽ Ѣҋғ. ҋҏҦҾ ҰҸ ҝѦҽҔ҃ѹҾѸ ӁҸҡѹҲѦґѦҮ ҾӉӃҔҽ, ѸҸңӁҸ ҡҦѦґҽҲҧ ӊӃӊҸҦ, ҍҲҸ ӉҦѦҡҧ ѸҸңѦҲ ѹҊҝҽҲҧ ӁѦҔҸҲҸҝҏӀ ҝҸґҧ ҲҸҲ ӍҽҔҲ, ҍҲҸ ҊґҽӊӁҽҾ ҍҽҡҲҧ ҸҲҦѦґѦӁҽ ҸҲ ҰҝѹҦҽҲҸҍӁҸҮ ҰҝҸѸѦңҏҲҸҍӁӃѸ ҸҰҝѦҦѦґѹҲѦґҧӁӃѸ ҰҝѦҦґҸңѦӁѹѦѸ. ҭҝѹ ҲҽҔѹѽ ҸқҡҲҸҾҲѦґҧҡҲӊҽѽ ӊӃқҸҝ ӊ ҰҝѹҦҽҲҸҍӁҸҮ ҍҽҡҲѹ ӍҸҝѸӃ ҋғ (spotka’em) ѸҸҊ қӃ ҰҝѹӊѦҡҲѹ Ҕ ҲҸѸҏ, ҍҲҸ ҰѦҝӊҸӁҽҍҽґҧӁҸ ӊҸӉӁѹҔґҽ қӃ ҸҗѹқҸҍӁҽҾ ѹӁҲѦҝҰҝѦҲҽ҃ѹҾ, ҍҲҸ ҝѦҍҧ ѹҦѦҲ Ҹ ѦҦѹӁѹҍӁӃѽ ҦѦҮҡҲӊѹҾѽ. ғѦҦҧ Ҹ ҰҸӊҲҸҝҾѦѸҸҡҲѹ ѹӁӍҸҝѸѹҝҏѦҲ Ӂҽҡ ҲҸґҧҔҸ ӊӃқҸҝ Ѣҋғ ӊ ҊґҽӊӁҸҮ ҍҽҡҲѹ, ҽ ҙҲҽ ѹӁӍҸҝѸҽ҃ѹҾ ҰҸҡҲҏҰҽѦҲ ҰҸӉҦӁҸ. ѷҽҔѹѸ ҸқҝҽӉҸѸ ӁѦ ѹҡҔґӀҍѦӁҸ, ҍҲҸ ӊ ҦҽӁӁҸѸ ҡґҏҍҽѦ ӊӃқҸҝ ӍҸҝѸӃ Ѣҋғ (spotyka’em) ӊ ҸҡӁҸӊӁҸѸ ѸҸҲѹӊѹҝҸӊҽӁ ңѦґҽӁѹѦѸ ѹӉқѦңҽҲҧ ӁѦҰҝҽӊѹґҧӁҸҮ ѹӁҲѦҝҰҝѦҲҽ҃ѹѹ. Ѣҽ ҙҲҸ ҏҔҽӉӃӊҽѦҲ ѹ ҲҸ ҸқҡҲҸҾҲѦґҧҡҲӊҸ, ҍҲҸ ӍҸҝѸҽ ҋғ ҡҲҽӁҸӊѹҲҡҾ ӊҰҸґӁѦ ҰҝѹѦѸґѦѸҸҮ, Ѧҡґѹ ӊҔґӀҍҽѦѸ ӊ ҰҝѹҦҽҲҸҍӁҏӀ ҍҽҡҲҧ ҔҽҔҸѦ-ӁѹқҏҦҧ ҏҔҽӉҽӁѹѦ Ӂҽ ҰҸӊҲҸҝҾѦѸҸҡҲҧ. ҘҽҔ ӊҸӉѸҸңӁӃҮ ӊҽҝѹҽӁҲ қӃґ, ӁҽҰҝѹѸѦҝ, ҰҝѹӁҾҲ ҡґѦҦҏӀӂѹҮ: Gdy spotka’emp byle jakƾ osobӾ֗ ... ‘ӊҡҲҝѦҍҽҾ ґӀқҸҊҸ ҍѦґҸӊѦҔҽ’. ҭҸ ҸҲӁҸҗѦӁѹӀ Ҕ ҡѦҝқҡҔҸѸҏ ҰҝѹѸѦҝҏ ӁҏңӁҸ ҸҲѸѦҲѹҲҧ, ҍҲҸ ӊ ҊґҽӊӁҸҮ ҍҽҡҲѹ ҰҝѦҦҡҲҽӊґѦӁҽ ӁѦ ӍҸҝѸҽ ҔҸӁҦѹ҃ѹҸӁҽґҽ, ҽ ӍҸҝѸҽ ѹӁҦѹҔҽҲѹӊҽ. ѷҽҔҸҮ ӊҽҝѹҽӁҲ ӊҡҲҝѦҍҽѦҲҡҾ ҦҸӊҸґҧӁҸ ҍҽҡҲҸ ӁѦ ҲҸґҧҔҸ ӊ ҡѦҝқҡҔҸѸ, ӁҸ ѹ ӊ ѽҸҝӊҽҲҡҔҸѸ ҾӉӃҔѦ. ҫҦӁҽҔҸ ҡґѦҦҏѦҲ ҰҸҦҍѦҝҔӁҏҲҧ, ҍҲҸ Ұҝѹ ѹӁҦѹҔҽҲѹӊѦ ҏҰҸҲҝѦқґҾѦҲҡҾ ҲҸґҧҔҸ Ѣҋғ, Ҳ.Ѧ. Ұҝѹ ҰҸӊҲҸҝҾѦѸҸҡҲѹ ӊҸӉѸҸңӁҸҡҲҧ ӊӃқҸҝҽ ҋғ ҡҏӂѦҡҲӊҏѦҲ ҲҸґҧҔҸ Ұҝѹ ҔҸӁҦѹ҃ѹҸӁҽґѦ. ѷҽҔѹѸ ҸқҝҽӉҸѸ ӊ ҝҽҡҡѸҽҲҝѹӊҽѦѸӃѽ ҡѦҝқҡҔѹѽ ѹ ѽҸҝӊҽҲҡҔѹѽ ҰҝѦҦґҸңѦӁѹҾѽ 37 ӊҸӉѸҸңӁӃ ҡґѦҦҏӀӂѹѦ Ҳҝѹ ҲѹҰҽ: ҋғҔҸӁҦ–ҋғҔҸӁҦ, ҋғҔҸӁҦ–ѢҋғҔҸӁҦ ѹ ҋғҔҸӁҦ–ѢҋғѹӁҦ. ғ ҰҝѹѸѦҝҽѽ, ӁҽҮҦѦӁӁӃѽ ӁҽѸѹ ӊ ҔҸҝҰҏҡѦ, ӁҽѹқҸґѦѦ ҏҰҸҲҝѦқѹҲѦґҧӁӃѸѹ ҸҔҽӉҽґѹҡҧ ҡҽѸӃѦ ҔҝҽҮӁѹѦ ҲѹҰӃ (Ҳ.Ѧ. ҋғҔҸӁҦ– ҋғҔҸӁҦ ѹ ҋғҔҸӁҦ–ѢҋғѹӁҦ). ҭҝѦҦҰҸґҽҊҽѦѸ, ҍҲҸ ҙҲҸ ѸҸңӁҸ ҸқӆҾҡӁѹҲҧ ҲѦѸ, ҍҲҸ ҰҸ ҡѹҡҲѦѸӁӃѸ ҡҸҸқҝҽңѦӁѹҾѸ ҡҝѦҦӁѹҮ ҲѹҰ ҋғҔҸӁҦ–ѢҋғҔҸӁҦ, ҡҸқҡҲӊѦӁӁҸ ҊҸӊҸҝҾ, ҾӊґҾѦҲҡҾ ѹӉґѹҗӁѹѸ. ғѦҦҧ ӊ ҦҽӁӁӃѽ ҾӉӃҔҽѽ Ѣҋғ (ӊ ӁѦӉҽӊѹҡѹѸҸҮ ҰҸӉѹ҃ѹѹ) ҡҽѸ ҰҸ ҡѦқѦ ѸҸңѦҲ ҡҝҽӊӁѹҲѦґҧӁҸ ґѦҊҔҸ ҸқҸӉӁҽҍҽҲҧ ҰҸӊҲҸҝҾӀӂѹѦҡҾ ҃ѦґҸҡҲӁӃѦ ҦѦҮҡҲӊѹҾ, Ҳ.Ѧ. Ұҝѹ Ѣҋғ ҰҸҦҦѦҝңҔҽ ҸҡҸқҸҮ, ѸҸҦҽґҧӁҸ ѸҽҝҔѹҝҸӊҽӁӁҸҮ ӍҸҝѸӃ ҔҸӁҦѹ҃ѹҸӁҽґҽ ӁѦ ӁҏңӁҽ. ҭҝѹ ӊҡѦѽ ҲҝѦѽ ҲѹҰҽѽ ҰҝѹѸѦҝӁҸ ҸҦѹӁҽҔҸӊҸ ҾҡӁҸ ӊӃҝҽңҽѦҲҡҾ ҰҸӊҲҸҝҾѦѸҸҡҲҧ ӊҡѦҮ ҃ѦҰҸҍҔѹ ҦѦҮҡҲӊѹҮ ѹ ҰҝѦҦҗѦҡҲӊҸӊҽӁѹѦ ҦѦҮҡҲӊѹҾ ҰҝѹҦҽҲҸҍӁҸҮ ҍҽҡҲѹ. ѢѦҡҸѸӁѦӁӁҸ, қӃґҸ қӃ ӊҽңӁҸ ҰҝҸҡґѦҦѹҲҧ ҦҸҰҸґӁѹҲѦґҧӁӃѦ ӍҽҔҲҸҝӃ, ҡҰҸҡҸқҡҲӊҏӀӂѹѦ ӊӃқҸҝҏ ѸѦ37

ѨѸѦӀҲҡҾ ӉҦѦҡҧ ӊ ӊѹҦҏ ҲҽҔңѦ ҰҝѦҦґҸңѦӁѹҾ ҡ ҡҸӀӉҸѸ ԓӠԐ/ǁim, Ҳ.Ѧ. ҲѹҰҽ, ҝҽҡҡѸҽҲҝѹӊҽѦѸҸҊҸ ӊ ҰҝѦҦӃҦҏӂѦѸ ҰҽҝҽҊҝҽӍѦ.

24

үѩҩѨүѢ ҷүҩѣѢѷҋѣѢ

ңҦҏ ҏҔҽӉҽӁӁӃѸѹ ҲҝѦѸҾ ҲѹҰҽѸѹ, ӁҸ ҦґҾ ҙҲҸҊҸ ҰҸҲҝѦқҏѦҲҡҾ ҸҲҦѦґҧӁҸѦ ѹҡҡґѦҦҸӊҽӁѹѦ Ӂҽ ҸҡӁҸӊѦ ҸқҗѹҝӁҸҊҸ ѸҽҲѦҝѹҽґҽ. 38 ғ ҡґѦҦҏӀӂѦѸ ҰҝѹѸѦҝѦ ҡҸӀӉӃ ҲѹҰҽ ӾԙԊӠ ҰҝѦҦҡҲҽӊґѦӁӃ ӊ ҸҝҊѹӁҽґѦ ѹ ӊ Ҧӊҏѽ ҰѦҝѦӊҸҦҽѽ. 39 ғ Ӂѹѽ ӁѦ ӁҽқґӀҦҽѦҲҡҾ ӁѹҔҽҔѹѽ ҸҲҔґҸӁѦӁѹҮ Ұҝѹ ӊӃқҸҝѦ ҊґҽҊҸґҧӁӃѽ ӍҸҝѸ ҰҸ ҡҝҽӊӁѦӁѹӀ ҡ ҦҝҏҊѹѸѹ ҰҝѹѸѦҝҽѸѹ, ҝҽҡҡѸҽҲҝѹӊҽѦѸӃѸѹ ӉҦѦҡҧ ѹ ӊӃҗѦ. ѩґҾ ґҏҍҗѦҊҸ ҰҸӁѹѸҽӁѹҾ ҰҝѹѸѦҝҽ (16) ӁѦҸқѽҸҦѹѸҸ ӉӁҽҲҧ ҸқӂҏӀ ҡѹҲҏҽ҃ѹӀ, Ҕ ҔҸҲҸҝҸҮ ҸӁ ҸҲӁҸҡѹҲҡҾ. ғҸґҗѦқӁѹҔ ҐҙӁҦҽґҧӍ ҡѹҦѹҲ Ӂҽ ӊѦҝѽҏҗҔѦ ҦѦҝѦӊҽ, ӊ ҸҔҝҏңѦӁѹѹ ӉґӃѽ ӊҸґҔҸӊ. ҫӁ ҰҸҦңѹҊҽѦҲ ҡҸҡӁҸӊӃѦ җѹҗҔѹ ѹ қҝҸҡҽѦҲ ѹѽ ҸҦӁҏ Ӊҽ ҦҝҏҊҸҮ ӊ ӊҸґҔҸӊ. (16) J.R.R. Tolkien: The Hobbit or There and Back Again үӁҊґ.

If a spark got in their coats it stuck and burned into them, [and unless they rolled over quick they were soon all in flames]. ҩҏҡҡҔѹҮ ѣҡґѹ ѽҸҲҧ ҸҦӁҽ ѸҽґѦӁҧҔҽҾ ѹҡҔҸҝҔҽ ҰҸҰҽҦҽґҽi Ӂҽ ӊҸґҍҧӀ җҔҏҝҏ, ҸӁҽ ҲҏҲ ңѦ ԗӢӾӴӢԂԷԂԊԂԙӸi ӊ ҰґҽѸҾ [...] (ӅҏҝҸӊ). ҷѦґҸҝҏҡ. Ҙҽґҟ ҟҡҔҝҽ ҲҝҽҰґҾґҽi Ӂҽ ҡҔҏҝҏ, ҲҸ ԗӢՂԊӬԗԂԊԂi ҦӃ ԰ԗԂԊӸӴԂԊԂԙԋi ҕ ҾѦ [...] ѰҔҝ. ѩҸҡѹҲҧ қҏґҸi ҟҡҔҝҟ ҰҸҲҝҽҰѹҲѹp ҏ ӊҸӊҍҏ ӊҸӊӁҏ, ҾҔ ӊҸӁҽ ӊңѦ ӉӊҟҦҲѹp ӰӾ ӴӠԽӢӫԶӫӴԂԊԂԙӸi, ҽ ԙԗԂԊԂԖӫӴԂԊԂi ҰҸґҏѸ’ҾѸ. ҭҸґҧҡҔ. Gdy choǀ jedna iskra pad’ap na ich futro, natychmiast roznieca’ai p’omieǝ [...] (Braiter). ѯѦҗҡҔ. Kdykoli jim nLJjaká jiskra vletLJlap do kožichu, uvízlap tam a propalovalai se do nich [...] ҋґҸӊҽ҃Ҕ. KeDŽ dopadlap niektorému vlkovi do kožucha iskra, zavǤtalap sa mu doǞ a tak ho pálilai, že [...] ҋґҸӊѦӁҡҔ. Kadarkoli je iskra padlap kateremu na kožuh, se je zažrlap in zažgalap vanj [...] ҤҸҝӊҽҲҡҔ. Kad bi [cond.] ih pogodilap iskra, zalijepilap bi [cond.] im se za kožu i upalilap je [...] 38

ѢѦҸқѽҸҦѹѸҸ ҲҽҔңѦ ҰҸҦҝҸқӁҸ ѹҡҡґѦҦҸӊҽҲҧ ӊӉҽѹѸӁҸѦ ҝҽҡҰҝѦҦѦґѦӁѹѦ ҲҝѦѽ ҙҲѹѽ ӍҸҝѸ (ѢҋғѹӁҦ – ѢҋғҔҸӁҦ – ҋғҔҸӁҦ) Ұҝѹ ҰҸӊҲҸҝҾѦѸҸҡҲѹ ӊӁѦ ҝҽѸҸҔ ҡґҸңӁҸҰҸҦҍѹӁѦӁӁҸҊҸ ҰҝѦҦґҸңѦӁѹҾ. 39 ү ѹѸѦӁӁҸ ӊ ҝҏҡҡҔҸѸ ѹ ӊ ѸҽҔѦҦҸӁҡҔҸѸ ҰѦҝѦӊҸҦҽѽ. ѢҽѸ ӁѦ ҏҦҽґҸҡҧ ӁҽҮҲѹ ҲҽҔѹѦ ҰҝѹѸѦҝӃ, ҊҦѦ қӃ ӊҡѦ ҰѦҝѦӊҸҦӃ ҡҸҦѦҝңҽґѹ ҡҸӀӉ ҙҲҸҊҸ ҲѹҰҽ. ҭҸҡҔҸґҧҔҏ ӊ қҸґҊҽҝҡҔҸѸ ҰѦҝѦӊҸҦѦ ҰҝѹѸѦҝҽ (16) ҰҝѦҦҡҲҽӊґѦӁ ҔҽҔ ҝҽӉ ҲѹҰѹҍӁӃҮ ӊҝѦѸѦӁӁҸҮ ҡҸӀӉ, ҡҝҽӊӁѹѸӃҮ ҡ ҝҏҡҡҔѹѸ ӪԂӪ ԽӹԊӸӪӹ, ҰҝѹӊҸҦѹѸ ҦҸҰҸґӁѹҲѦґҧӁӃҮ ҰҝѹѸѦҝ ҡ қҸґҊҽҝҡҔѹѸ ҏҡґҸӊӁӃѸ ҡҸӀӉҸѸ ԂӪӹ: and if an intruder removed a piece of artwork, compartmentalized exits would seal around that gallery – ԂӪӹ ӴԔӰԶӰӹ ԊӠՁӾ ԗӢӾԐӾԙԽӾԶӾp [impf.] ӰԋӪӹӡ ӾӪԙԗӹӰԂԽ, ӠԒԖӹԌӠԽӾ ӰԂ ԙԔӹԽӴӾԽӰԂԽԂ ԄԂԊӾӢӠԋ ҡѦ ӉҽҲӊҽҝҾѽҽi [impf.] (Dan Brown: The Da Vinci Code).

ҭҫҋһѣѩҫғүѷѣһѻѢҫҋѷѻ ѩѣҢҋѷғѨҢ ҭҩѨ ҭҫғѷҫҩѤѣҴҫҋѷѨ

ҋѦҝқҡҔѹҮ ҴҽҔѦҦҸӁ. ҷҸґҊҽҝҡҔ.

25

Kad bi [cond.] im se varnica dotaklap krzna, prosto se lepilai i zadiralai u njih [...] үҔҸ ҡҽѸҸ ѹҡҔҝҽ ҦҸҰҝѦҗѦp [impf.] ҦҸ ӁѹӊӁѹҲѦ ҔҝӉӁҽ ԙӾ ԊӾԗӾԶӾi [impf.] ѹ Ҋѹ ԄӹӢӾԶӾi [impf.] [...] ӅҸѸ ӁҾҔҸҾ ѹҡҔҝҽ ҰҸҰҽҦӁѦҗѦp [impf.] ӊӆҝѽҏ ҔҸӉѹӁҽҲҽ, ӹԗԂӢӴԂԶӾi [impf.] Ҋѹ, ӢԂԒԄԂӢԋԶӾi [impf.] ԙӾ [...]

ғӃқҸҝ ҊґҽҊҸґҧӁӃѽ ӍҸҝѸ ӊ ҦҽӁӁҸѸ ҰҝѹѸѦҝѦ ҰҸҍҲѹ ҰҸґӁҸҡҲҧӀ ҡҸҸҲӊѦҲҡҲӊҏѦҲ ҸҲѸѦҍѦӁӁӃѸ ӊӃҗѦ ҲѦӁҦѦӁ҃ѹҾѸ. 40 ѢҽѹқҸґѦѦ ӉҽѸѦҲӁҸѦ ҝҽӉґѹҍѹѦ ҡҸҡҲҸѹҲ ӊ ҲҸѸ, ҍҲҸ ӊ ѸҽҔѦҦҸӁҡҔҸѸ ҰѦҝѦӊҸҦѦ ҸҲҡҏҲҡҲӊҏѦҲ ҍҽҡҲѹ҃ҽ ԀӾ. ҶҲҸ ҡӊҾӉҽӁҸ ѹѸѦӁӁҸ ҡ ӊӃқҸҝҸѸ ҡҸӀӉҽ – Ұҝѹ ҏҡґҸӊӁҸѸ ҡҸӀӉѦ ԂӪӹ ҙҲҽ ҍҽҡҲѹ҃ҽ ҸҰҏҡҔҽѦҲҡҾ. 41 ѩҽґѦѦ ҸҲѸѦҲѹѸ ҲҸґҧҔҸ, ҍҲҸ ӊ ҏҔҝҽѹӁҡҔҸѸ ҰѦҝѦӊҸҦѦ ѹҡҰҸґҧӉҏѦҲҡҾ ҔҸӁҡҲҝҏҔ҃ѹҾ, ҡѽҸҦӁҽҾ ҡ ҝҏҡҡҔѹѸ ԙԽӹӠԊӹ + ѹӁӍѹӁѹҲѹӊ ҋғ (ҡҝ. (13ҽ)) ѹ ҍҲҸ ѽҸҝӊҽҲҡҔѹҮ ѹ ҡѦҝқҡҔѹҮ ҰҝѹѸѦҝӃ ҸҲӁҸҡҾҲҡҾ Ҕ ҦӊҏѸ ҝҽӉґѹҍӁӃѸ, “ҔҝҽҮӁѹѸ” ҲѹҰҽѸ. 42 ѨӁҲѦҝѦҡӁҸ ҸҲѸѦҲѹҲҧ, ҍҲҸ ӊ ҸҦӁҸѸ ѹӉ ҦҝҏҊѹѽ ѹѸѦӀӂѹѽҡҾ ҝҏҡҡҔѹѽ ҰѦҝѦӊҸҦҸӊ ѹҡҰҸґҧӉҏѦҲҡҾ ҸҡҸқӃҮ ҡҰҸҡҸқ ҰҝѦҸҦҸґѦӁѹҾ ӊѹҦҸӊҸҮ ӁѦҮҲҝҽґѹӉҽ҃ѹѹ, ѽҽҝҽҔҲѦҝӁҸҮ ҦґҾ ҝҏҡҡҔҸҊҸ ҾӉӃҔҽ ӊ ҸқӃҍӁӃѽ ҡґҸңӁҸҰҸҦҍѹӁѦӁӁӃѽ ҰҝѦҦґҸңѦӁѹҾѽ ҝҽҡҡѸҽҲҝѹӊҽѦѸҸҊҸ ҲѹҰҽ: ӊѸѦҡҲҸ ҰҝѹҦҽҲҸҍӁҸҮ ҍҽҡҲѹ ӊ ҙҲҸѸ ҰѦҝѦӊҸҦѦ ѹҡҰҸґҧӉҏѦҲҡҾ ҦѦѦҰҝѹҍҽҡҲѹѦ ҋғ: (16ҽ) ҴҽґѦҮҗҽҾ ѹҡҔҸҝҔҽ, ҰҸҰҽӊp Ӂҽ ӊҸґҔҽ, ӴԗӠӴԂԊԂԙӸi ӊ җҔҏҝҏ (үӁҦҝѦѦӊ). ҩҏҡҡҔҸѦ ҦѦѦҰҝѹҍҽҡҲѹѦ ҋғ “ҡӊҸқҸҦӁҸ ҏҰҸҲҝѦқґҾѦҲҡҾ ҦґҾ ҸқҸӉӁҽҍѦӁѹҾ ӁѦҸҊҝҽӁѹҍѦӁӁҸ-ҔҝҽҲӁӃѽ ҡѹҲҏҽ҃ѹҮ” (ҘӁҾӉѦӊ 1998: 199). ҴҸңӁҸ ҡҦѦґҽҲҧ ӊӃӊҸҦ, ҍҲҸ ҦѦѦҰҝѹҍҽҡҲѹѦ, ӊ ҸҲґѹҍѹѦ ҸҲ ґѹҍӁӃѽ ӍҸҝѸ ҰҝҸҗѦҦҗѦҊҸ ӊҝѦѸѦӁѹ, ҸҝѹѦӁҲѹҝҏѦҲҡҾ Ӂҽ ѸѹҔҝҸҡѹҲҏҽ҃ѹѹ. ҋҸҍѦҲҽӁѹѦ ҙҲҸҮ ӍҸҝѸӃ ҡ ҊґҽӊӁӃѸ ҰҝѦҦѹҔҽҲҸѸ ҰҝѦҦѦґҧӁҸҊҸ ҊґҽҊҸґҽ ӊ ӍҸҝѸѦ ҰҝҸҗѦҦҗѦҊҸ ӊҝѦѸѦӁѹ Ѣҋғ ҾӊґҾѦҲҡҾ ӁҽѹқҸґѦѦ ҽҦѦҔӊҽҲӁӃѸ ѹ ҙҔҸӁҸѸӁӃѸ ҡҰҸҡҸқҸѸ ҙҔҡҰґѹ҃ѹҲӁҸҊҸ ӊӃҝҽңѦӁѹҾ ҃ѦҰҸҍҔѹ ҦѦҮҡҲӊѹҮ, ҰҸӊҲҸҝҾӀӂѦҮҡҾ ӊ ҰҝҸҗґҸѸ. ҭҸҦҸқӁӃѦ ҰҝѦҦґҸңѦӁѹҾ ӊ ӊѹҦҸӊҸѸ ҸҲӁҸҗѦӁѹѹ ӁҽҰҸѸѹӁҽӀҲ ҝҽҡҡѸҸҲҝѦӁӁӃѦ ӊӃҗѦ ҰҸґҧҡҔѹѦ ҰҝѹѸѦҝӃ, ҸҲӁҸҡҾӂѹѦҡҾ Ҕ ҲѹҰҏ ҋғ–Ѣҋғ. ѰҔҽӉҽӁӁӃѦ ҰҝѦҦґҸңѦӁѹҾ ӁѦҝѦҦҔҸ ӊҡҲҝѦ40

ҫқҝҽӂҽѦѸ ӊӁѹѸҽӁѹѦ Ӂҽ ҲҸ, ҍҲҸ ӊ ҍѦҗҡҔҸѸ ѹ ҡґҸӊҽ҃ҔҸѸ ҰҝѹѸѦҝҽѽ ҙҲҸ ҸҲӁҸҡѹҲҡҾ ҲҸґҧҔҸ Ҕ ҰѦҝӊҸѸҏ ҊґҽҊҸґҏ ҊґҽӊӁҸҮ ҍҽҡҲѹ. ғ ҦѦҲҽґѹ ӊѹҦҸӊҸҮ ҸҝҊҽӁѹӉҽ҃ѹѹ ӊӁҏҲҝѹ ҊґҽӊӁҸҮ ҍҽҡҲѹ ҰҝѦҦґҸңѦӁѹҾ ѸӃ ӉҦѦҡҧ ӊѽҸҦѹҲҧ ӁѦ қҏҦѦѸ. 41 ҘҽҔ ҰҝҽӊѹґҸ, ҸӁҽ ҸҰҏҡҔҽѦҲҡҾ ҲҽҔңѦ Ұҝѹ ҡҸӀӉѦ ҡґѦҦҸӊҽӁѹҾ – ԌӹӪԂԽӹ ӰӾ. 42 ҭҝѹѸѦҝӃ ҡ ѹӁҦѹҔҽҲѹӊҸѸ Ѣҋғ ӊ ҊґҽӊӁҸҮ ҍҽҡҲѹ ӊҡҲҝѦҍҽӀҲҡҾ ҦҸҡҲҽҲҸҍӁҸ ҍҽҡҲҸ ҔҽҔ ӊ ҡѦҝқҡҔҸѸ, ҲҽҔ ѹ ӊ ѽҸҝӊҽҲҡҔѹѽ ҲѦҔҡҲҽѽ. ѷҸҲ ӍҽҔҲ, ҍҲҸ ӊ ҰҝѹѸѦҝҽѽ (15) ѹ (16) ҲҸґҧҔҸ ҡѦҝқҡҔѹҮ ҰѦҝѦӊҸҦ ҡҸҦѦҝңѹҲ ҙҲҸҲ ӊҽҝѹҽӁҲ, ҡґѦҦҏѦҲ ҡҍѹҲҽҲҧ ҡҔҸҝѦѦ ӊҡѦҊҸ ҡґҏҍҽҮӁҸҡҲҧӀ.

26

үѩҩѨүѢ ҷүҩѣѢѷҋѣѢ

ҍҽӀҲҡҾ ӊ ҲѦҔҡҲҽѽ, ѹ Ҳ.Ҕ. ҦҽӁӁҽҾ ҔҸӁҡҲҝҏҔ҃ѹҾ ѸҸңѦҲ ґҏҍҗѦ ӊӃҝҽӉѹҲҧ ӊѹҦҸӊҸҮ ѽҽҝҽҔҲѦҝ ѸѹҔҝҸҡѹҲҏҽ҃ѹѹ, ѦѦ ҡґѦҦҏѦҲ ҡҍѹҲҽҲҧ ѽҸҝҸҗѦҮ ҽґҧҲѦҝӁҽҲѹӊҸҮ ҦґҾ ҔҸӁҡҲҝҏҔ҃ѹѹ ҡ ҰҝѹҦҽҲҸҍӁӃѸ ҰҝѦҦґҸңѦӁѹѦѸ. ҐґҽӊӁҸѦ ҰҝѦҰҾҲҡҲӊѹѦ ҦґҾ ҏҰҸҲҝѦқґѦӁѹҾ ҙҲҸҮ ҔҸӁҡҲҝҏҔ҃ѹѹ ӉҽҔґӀҍҽѦҲҡҾ ӊ ҲҸѸ, ҍҲҸ ӊ ґѹҲѦҝҽҲҏҝӁҸѸ ҾӉӃҔѦ ҸӁҽ ӊҸӉѸҸңӁҽ ҲҸґҧҔҸ Ұҝѹ ҸҦӁҸҝѦӍѦҝѦӁҲӁҸҡҲѹ ҡҏқӆѦҔҲҸӊ ҊґҽӊӁҸҊҸ ѹ ӉҽӊѹҡѹѸҸҊҸ ҰҝѦҦѹҔҽҲҸӊ. ѩҽґѦѦ ҡґѦҦҏѦҲ ҸҲѸѦҲѹҲҧ, ҍҲҸ ҦѦѦҰҝѹҍҽҡҲѹѦ ѸѦӁѦѦ ҍѦҲҔҸ ӊӃҝҽңҽѦҲ ѽҽҝҽҔҲѦҝ ҸҲӁҸҗѦӁѹҮ ѸѦңҦҏ ҦҽӁӁӃѸѹ ҦѦҮҡҲӊѹҾѸѹ. ҫҰҝѦҦѦґѦӁӁӃѦ ҝҽӉґѹҍѹҾ, ҡӊҾӉҽӁӁӃѦ ҡ ӊӃқҸҝҸѸ ҲҸҊҸ ѹґѹ ѹӁҸҊҸ ҡҸӀӉҽ (ӪԂӪ ԽӹԊӸӪӹ, ӪӹԄԌԂ, ӾԙԊӠ ѹ Ҳ.Ҧ.), ҏ ҦѦѦҰҝѹҍҽҡҲѹҮ ӁѦ ҦѹӍӍѦҝѦӁ҃ѹҝҏӀҲҡҾ. ғҰҝҸҍѦѸ, ӊ ҝҾҦѦ ҡґҏҍҽѦӊ ҲҽҔҏӀ ӁѦҦѹӍӍѦҝѦӁ҃ѹҝҸӊҽӁӁҸҡҲҧ ҡґѦҦҏѦҲ ҡҍѹҲҽҲҧ ҡҔҸҝѦѦ ҰҝѦѹѸҏӂѦҡҲӊҸѸ, ҍѦѸ ӁѦҦҸҡҲҽҲҔҸѸ. ғѦҡҧѸҽ ѹӁҲѦҝѦҡӁҸ ҸҲѸѦҲѹҲҧ, ҍҲҸ ӊ ѹѸѦӀӂѹѽҡҾ ҏ Ӂҽҡ җѦҡҲѹ ҝҽӉґѹҍӁӃѽ ҝҏҡҡҔѹѽ ҰѦҝѦӊҸҦҽѽ ҔҸӁҡҲҝҏҔ҃ѹҾ ҡ ҦѦѦҰҝѹҍҽҡҲѹѦѸ ҋғ ӊҡҲҝѦҍҽѦҲҡҾ ӊҡѦҊҸ ҸҦѹӁ ҝҽӉ, ӊ ҲҸ ӊҝѦѸҾ ҔҽҔ ӁѦҔҸҲҸҝӃѦ ҰѦҝѦӊҸҦҍѹҔѹ (ү. ҘҸӁҽҰӃѽѹӁ, Ѣ. ҭҝҸѽҸҝҸӊҽ, Ѣ. ҩҽѽѸҽӁҸӊҽ) ӊӃқҝҽґѹ ӉҦѦҡҧ ҦѦѦҰҝѹҍҽҡҲѹѦ ҰҝҸҲѹӊҸҰҸґҸңӁҸҊҸ ӊѹҦҽ: (16қ) ѨҡҔҝҽ, ҰҸҰҽҦҽҾi Ӂҽ җҔҏҝҏ, ӴԗӠӴԂԊԂԙӸi ӊ ӁѦҺ, ԗӢӹԞӠԄԂԊԂi ӊҸґҔҽ ӁҽҡҔӊҸӉҧ (ҩҽѽѸҽӁҸӊҽ). ҶҲҸҲ ӊҽҝѹҽӁҲ ӊ ӊѹҦҸӊҸѸ ҸҲӁҸҗѦӁѹѹ қґѹңѦ Ҕ ҰҝѹӊѦҦѦӁӁҸѸҏ ӊ (16) ӊҽҝѹҽӁҲҏ ҡ ҰҝѹҦҽҲҸҍӁӃѸ ҰҝѦҦґҸңѦӁѹѦѸ. ҫҦӁҽҔҸ, ҰҸҡҔҸґҧҔҏ Ұҝѹ ҦѦѦҰҝѹҍҽҡҲѹѹ ӊҸӉѸҸңѦӁ ӊѹҦҸӊҸҮ ҔҸӁҲҝҽҡҲ, Ѣҋғ ӊ (16қ) ӊӁҸҡѹҲ ҸҡҸқӃҮ ҸҲҲѦӁҸҔ. ҫӁ ҔҽҔ қӃ ҔҸӁ҃ѦӁҲҝѹҝҏѦҲ ӁҽҗѦ ӊӁѹѸҽӁѹѦ Ӂҽ ӁѦҔҸҲҸҝҸѸ ҰҽҝҽґґѦґѹӉѸѦ ѸѦңҦҏ ҦѦҮҡҲӊѹҾѸѹ, ӊ ҲҸ ӊҝѦѸҾ ҔҽҔ ҡҸҸҲӊѦҲҡҲӊҏӀӂҽҾ ӍҸҝѸҽ ҋғ (ӊ (16ҽ)) ҔҽҔ ҝҽӉ ҰҸҦҍѦҝҔѹӊҽѦҲ ҙґѦѸѦӁҲ ҃ѦґҸҡҲӁҸҡҲѹ ҦѦѦҰҝѹҍҽҡҲӁҸҊҸ ҦѦҮҡҲӊѹҾ ѹ, ҔҽҔ ҡґѦҦҡҲӊѹѦ ҙҲҸҊҸ, ҰҝѦҦҗѦҡҲӊҸӊҽӁѹѦ ҦѦҮҡҲӊѹӀ ҊґҽӊӁҸҮ ҍҽҡҲѹ. ғ ҦҽӁӁҸѸ ҡґҏҍҽѦ ӊӃқҸҝ Ѣҋғ, ӊҸӉѸҸңӁҸ, ѸҸҲѹӊѹҝҸӊҽӁ ѦӂѦ ѹ ҲѦѸ, ҍҲҸ ѸѹҔҝҸҡѹҲҏҽ҃ѹѹ ҰҸӊҲҸҝҾӀҲҡҾ ҡ ӁѦӉӁҽҍѹҲѦґҧӁӃѸѹ ѹӁҲѦҝӊҽґҽѸѹ, ӊҡґѦҦҡҲӊѹѦ ҍѦҊҸ ѸҽҔҝҸҡѹҲҏҽ҃ѹҾ ҰҝѹҸқҝѦҲҽѦҲ ѽҽҝҽҔҲѦҝ ҰҸҍҲѹ ӁѦҰҝѦҝӃӊӁҸҊҸ ҰҸҲҸҔҽ ҦѦҮҡҲӊѹҮ. ҘҽҔ ӁҽѸ ҔҽңѦҲҡҾ, Ұҝѹ ҦҽӁӁҸѸ ҰҸґҸңѦӁѹѹ ҦѦґ Ҹқҽ ҰҝѦҦҡҲҽӊґѦӁѹҾ (ҰҸҡґѦҦҸӊҽҲѦґҧӁҸҡҲҧ ѹ ҰҽҝҽґґѦґѹӉѸ) ӍҽҔҲѹҍѦҡҔѹ ҸҦѹӁҽҔҸӊҸ ҰҝѹѦѸґѦѸӃ. Ѣҽ ҸҡӁҸӊѦ ҰҸҦҸқӁӃѽ ҰҝѹѸѦҝҸӊ ѸҸңӁҸ ҡҦѦґҽҲҧ ӊӃӊҸҦ, ҍҲҸ ӊ ӁѦҔҸҲҸҝӃѽ ҡґҏҍҽҾѽ ҝҽӉґѹҍѹѦ ѸѦңҦҏ ӉӁҽҍѦӁѹѦѸ ҰҝѦҦҗѦҡҲӊҸӊҽӁѹҾ ѹ ҸҦӁҸӊҝѦѸѦӁӁҸҡҲѹ ѸҸңѦҲ ӊ ҔҽҔҸҮ-ҲҸ ѸѦҝѦ ҡҲѹҝҽҲҧҡҾ. ғ ӉҽҔґӀҍѦӁѹѦ Ҕ ҦҽӁӁҸѸҏ ҰҽҝҽҊҝҽӍҏ ҰҝѹӊѦҦѦѸ ҰҝѹѸѦҝ, ҰҸҔҽӉӃӊҽӀӂѹҮ, ҍҲҸ ӊӃқҸҝ Ѣҋғ ӊ ҊґҽӊӁҸҮ ҍҽҡҲѹ ѸҸңѦҲ қӃҲҧ ѸҸҲѹӊѹҝҸӊҽӁ ҦҝҏҊѹѸѹ ӍҽҔҲҸҝҽѸѹ, ҍѦѸ ҰҸӊҲҸҝҾѦѸҸҡҲҧ:

ҭҫҋһѣѩҫғүѷѣһѻѢҫҋѷѻ ѩѣҢҋѷғѨҢ ҭҩѨ ҭҫғѷҫҩѤѣҴҫҋѷѨ

27

(17) Alan Alexander Milne: Winnie-The-Pooh үӁҊґ.

[But his arms were so stiff from holding on to the string of the balloon all that time that they stayed up straight in the air for more than a week, and] whenever a fly came and settled on his nose he had to blow it off. ҩҏҡҡҔѹҮ ѣҡґѹ ѦѸҏ Ӂҽ ӁҸҡ ҡҽҦѹґҽҡҧi Ѹҏѽҽ, ѦѸҏ ԗӢӠԖӹԌӠԊӹԙӸi ԙԌӫӴԂԽӸi ѦѦ (҈ҽѽҸҦѦҝ). ѰҔҝ. ӎ ҔҸґѹ Ӂҽ Ӂҟҡ ҮҸѸҏ ҡҟҦҽґҽi Ѹҏѽҽ, ӊҟӁ ԐӫԙӠӴi ѶѶ ԒԌԐӫԖӫӴԂԽӠi (ҋҸґҸӁҧҔҸ). ҭҸґҧҡҔ. Ile razy mucha siad’ap mu na nosie [...] tylko zdmuchiwa’i jƾ (Tuwim). ѯѦҗҡҔ. [...] když si mu sedlap nLJjaká moucha na nos, musili ji vždycky odfouknoutp. ҋґҸӊҽ҃Ҕ. [...] keDŽ mu na nos sadlap mucha, museli ju odfúknuǬp. һҏңѹ҃Ҕ. 43 A hdyž w tym ǁasu nLJkajka mucha pǥileǀap a so jemu na nos sydnyp, dyrbješei wón ju preǁ duǀi. ҋґҸӊѦӁҡҔ. [...] kadar mu je priletelap muha na smrǁek, si jo je morali odpihnitip. ҤҸҝӊҽҲҡҔ. [...] kad bi [cond.] god naišlap kakva muha i sletjelap mu na nos, moraoi je puhatii. ҋѦҝқҡҔѹҮ [...] kad god mu neka muva sletip na nos, moraoi je da je oduvai. ҴҽҔѦҦҸӁ. [...] ҡѦҔҸҊҽҗ ҔҸҊҽ ѱѦ [exp.] Ѹҏ ӉҽҡҲҽӁѦҗѦp [impf.] Ѹҏӊҽ Ӂҽ ӁҸҡҸҲ, ҭҏ ԐӹӢԂԶӾi [impf.] ԌԂ Ҏҽ ԌӫӴԂi [pres.]. ҷҸґҊҽҝҡҔ. ү ҔҸҊҽҲҸ ӁҾҔҸҾ Ѹҏѽҽ Ѹҏ Ҕҽ҃ӁѦҗѦp [impf.] Ӂҽ ӁҸҡҽ, ԽӢԋԼӴԂԶӾi [impf.] ԌԂ Ҿ ԌӫԖԂi [pres.]. ҷҝҸҡҽѦҲҡҾ ӊ ҊґҽӉҽ, ҍҲҸ ӊҸ ӊҡѦѽ ҰҝѹӊѦҦѦӁӁӃѽ ҰѦҝѦӊҸҦҽѽ ґѹҍӁҽҾ ӍҸҝѸҽ ҊґҽҊҸґҽ ӊ ҊґҽӊӁҸҮ ҍҽҡҲѹ ҸҲӁҸҡѹҲҡҾ Ҕ Ѣҋғ ҦҽңѦ ӊ ҲѦѽ ҾӉӃҔҽѽ, ҊҦѦ ҸқӃҍӁҸ ґѦҊҔҸ ӊӃқѹҝҽѦҲҡҾ ҋғ. ғ ҦҽӁӁҸѸ ҡґҏҍҽѦ ҙҲҸ ӁѦҡҸѸӁѦӁӁҸ ҡӊҾӉҽӁҸ ҡ ҲѦѸ, ҍҲҸ ӊӃҝҽңҽѦҲҡҾ ӁѦ ҸқӃҍӁҸѦ ҦѦҮҡҲӊѹѦ, ҽ ҸҰҝѦҦѦґѦӁӁҸѦ ѸҸҦҽґҧӁҸѦ ҸҲӁҸҗѦӁѹѦ Ҕ ҦѦҮҡҲӊѹӀ (ѹґѹ, ѸҸңѦҲ қӃҲҧ, ҦҽңѦ ҔҸ ӊҡѦҮ ҸқҡҲҽӁҸӊҔѦ ӊ ҃ѦґҸѸ). 44 ғ ҡґҽӊҾӁҡҔѹѽ ҾӉӃҔҽѽ ӁҽѹқҸґѦѦ ҸқӃҍӁӃѦ ѸҸҦҽґҧӁӃѦ ҰҝѦҦѹҔҽҲӃ ҦҸґңѦӁҡҲӊҸӊҽӁѹҾ/ӁѦҸқѽҸҦѹѸҸҡҲѹ (ӁҽҰҝ. ҝҏҡҡҔ. ԌӹԊԞӾӰ, ҰҸґҧҡҔ. musieǀ, ҡѦҝқҡҔ. ԐӹӢԂԽӠ, қҸґҊ. ԽӢԋԼӴԂԐ) ҾӊґҾӀҲҡҾ ӁѦҰҝѦҦѦґҧӁӃѸѹ ѹ ҰҸҙҲҸѸҏ ҸҦӁҸӊѹҦҸӊӃѸѹ ѹ ҸҲӁҸҡҾӂѹѸѹҡҾ Ҕ Ѣҋғ. ҫҍѦӊѹҦӁҸ, ҸқӃҍӁҸ ӁѦ ҸӂҏӂҽѦҲҡҾ ҸҡҸқҸҮ ҰҸҲҝѦқӁҸҡҲѹ ӊ 43

ҭҝѹӊҸҦѹѸ ӉҦѦҡҧ ҲҽҔңѦ ѦҦѹӁҡҲӊѦӁӁӃҮ ѹѸѦӀӂѹҮҡҾ ҏ Ӂҽҡ ҰҝѹѸѦҝ ѹӉ ӊѦҝѽӁѦ-ҡѦҝқҸґҏңѹ҃ҔҸҊҸ ҾӉӃҔҽ. ҐґҽҊҸґҧӁӃѦ ӍҸҝѸӃ ӊ ӁѦѸ ҰҝѦҦҡҲҽӊґҾӀҲ ҰҝҸҡҲҸҮ ҰҝѦҲѦҝѹҲ. ѷҝҽҦѹ҃ѹҸӁӁҸ ҸӁѹ ӁҽӉӃӊҽӀҲҡҾ ҽҸҝѹҡҲҸѸ Ұҝѹ ҋғ ѹ ѹѸҰѦҝӍѦҔҲҸѸ Ұҝѹ Ѣҋғ. 44 ғ ҰҸґҧҡҔҸѸ ҰѦҝѦӊҸҦѦ ҙҲҸ ѸҸҦҽґҧӁҸѦ ӉӁҽҍѦӁѹѦ ӁѦ ӊӃҝҽңҽѦҲҡҾ.

28

үѩҩѨүѢ ҷүҩѣѢѷҋѣѢ

ҙҔҡҰґѹ҃ѹҲӁҸѸ ҏҔҽӉҽӁѹѹ ӊ Ҹ Ӊ Ӂ ѹ Ҕ Ӂ Ҹ ӊ Ѧ Ӂ ѹ Ҿ ҡҸҡҲҸҾӁѹҾ ӁѦҸқѽҸҦѹѸҸҡҲѹ. ҭҸҍҲѹ ӊҸ ӊҡѦѽ ҰҝѹӊѦҦѦӁӁӃѽ ҰѦҝѦӊҸҦҽѽ ҙҲҽ ӍҽӉҽ ҔҽҔ қӃ ҰҝҸҰҏҡҔҽѦҲҡҾ. ѰҔҽӉӃӊҽѦҲҡҾ ҲҸґҧҔҸ ҡҽѸҸѦ ӉӁҽҍѹѸҸѦ, ҽ ѹѸѦӁӁҸ ҡ ҏ ӂ Ѧ ҡ Ҳ ӊ Ҹ ӊ ҽ Ӂ ѹ Ѧ ҙҲҸҊҸ ҡҸҡҲҸҾӁѹҾ. 45 ѷҽҔҸѦ ҰҸґҸңѦӁѹѦ ҦѦґ ӁѦҡҔҸґҧҔҸ ӁҽҰҸѸѹӁҽѦҲ ҲҸ, ҍҲҸ ѸӃ ӊѹҦѦґѹ ӊ ҍѦҗҡҔҸѸ ӊҽҝѹҽӁҲѦ ҰҝѹѸѦҝҽ (14ҽ), ӊ ҡӊҾӉѹ ҡ ҔҸҲҸҝӃѸ ѸӃ ҸқҡҏңҦҽґѹ ӊҸӉѸҸңӁҸҡҲҧ ҸқӆҾҡӁѦӁѹҾ ӊӃқҸҝҽ Ѣҋғ ӊ ҍѦҗҡҔҸѸ ҾӉӃҔѦ ҸҲҲѦӁҔҽѸѹ ҲѹҰҽ “ҰҝѹҡҲҏҰ Ҕ ҦѦҮҡҲӊѹӀ”. ѩҝҏҊҽҾ ҸҡҸқѦӁӁҸҡҲҧ ҰҝѹѸѦҝҽ (17) ҡҸҡҲҸѹҲ ӊ ҲҸѸ, ҍҲҸ ӊ ӉҽӊѹҡѹѸҸҮ ҍҽҡҲѹ ҡѦҝқҡҔҸҊҸ ҰѦҝѦӊҸҦҽ ӊӃқҝҽӁ ӁѦ ҋғ ҔҸӁҦѹ҃ѹҸӁҽґҽ, ҽ ҋғ ҰҝѦӉѦӁҡҽ. ҭҸҡҔҸґҧҔҏ ҏ Ӂҽҡ ӍҽҔҲѹҍѦҡҔѹ ӁѦҲ ҦҝҏҊѹѽ ҰҸҦҸқӁӃѽ ҰҝѹѸѦҝҸӊ, ӁҽѸ ҰҸҔҽ ӁѦҦҸҡҲҽҲҸҍӁҸ ҾҡӁҸ, ҔҽҔ ӁҏңӁҸ Ҹ҃ѦӁѹӊҽҲҧ ҡҲҽҲҏҡ ҙҲҸҊҸ ӊҽҝѹҽӁҲҽ, 46 ӁҽҡҔҸґҧҔҸ ҸӁ ӊҸӉѸҸңѦӁ ҲҽҔңѦ ӊ ҦҝҏҊѹѽ ҾӉӃҔҽѽ ѹ ҔҽҔѹѦ ҸӁ ҲҸҊҦҽ ѹѸѦѦҲ ҡҲѹґѹҡҲѹҍѦҡҔѹѦ ѹґѹ ҡѦѸҽӁҲѹҍѦҡҔѹѦ ҙӍӍѦҔҲӃ. ҘҽҔ қӃґҸ ҸҲѸѦҍѦӁҸ ӊ Ұҽҝ. 2, ӊ ҝҏҡҡҔҸѸ ҾӉӃҔѦ ӍҸҝѸҽ ҰҝѦӉѦӁҡҽ ҋғ җѹҝҸҔҸ ҏҰҸҲҝѦқґҾѦҲҡҾ ӊ ӁҽҊґҾҦӁҸ-ҰҝѹѸѦҝӁҸѸ ӉӁҽҍѦӁѹѹ, ӁҸ ӊ ҰҝѦҦґҸңѦӁѹҾѽ ҡ ҸқӃҍӁӃѸѹ ҡҸӀӉҽѸѹ ҰҝѦҦҗѦҡҲӊҸӊҽӁѹҾ 47 ҙҲҸ ҡҲѹґѹҡҲѹҍѦҡҔѹ ҸҔҝҽҗѦӁҸ ѹ ҰҸҍҲѹ ӁѦ ӊҡҲҝѦҍҽѦҲҡҾ. 48 5. ҫқӉҸҝ ӁҽҮҦѦӁӁӃѽ ҡҸҸҲӊѦҲҡҲӊѹҮ ҴҽҲѦҝѹҽґ, ҰҝѹӊѦҦѦӁӁӃҮ ӊ ҰҽҝҽҊҝҽӍҽѽ 3 ѹ 4, ҰҸӉӊҸґҾѦҲ ҡҦѦґҽҲҧ ӊӃӊҸҦ, ҍҲҸ ҰҸ ҸҲӁҸҗѦӁѹӀ Ҕ ӊӃҝҽңѦӁѹӀ ҰҸӊҲҸҝҾӀӂѦҮҡҾ ҰҸҡґѦҦҸӊҽҲѦґҧӁҸҡҲѹ ҦѦҮҡҲӊѹҮ ӊ ҰҝҸҗґҸѸ, ӊ ҝҽѸҔҽѽ ҡґҸңӁҸҰҸҦҍѹӁѦӁӁҸҊҸ ҰҝѦҦґҸңѦӁѹҾ ҡҸ ӉӁҽҍѦӁѹѦѸ ҰҝѦҦҗѦҡҲӊҸӊҽӁѹҾ ҦѦҮҡҲӊѹҾ ӉҽӊѹҡѹѸҸҮ ҍҽҡҲѹ, ѸѦңҦҏ ҡҸӊҝѦѸѦӁӁӃѸѹ ҡґҽӊҾӁҡҔѹѸѹ ҾӉӃҔҽѸѹ ӁҽқґӀҦҽӀҲҡҾ ӉҽѸѦҲӁӃѦ ҝҽӉґѹҍѹҾ. Ѣҽ ҸҡӁҸӊѦ ӁҽҮ-

45

҈ҽѸѦҲӁӃѸ ѹҡҔґӀҍѦӁѹѦѸ ҾӊґҾѦҲҡҾ ҝҏҡҡҔѹҮ ҾӉӃҔ, ӊ ҔҸҲҸҝҸѸ ҡҏӂѦҡҲӊҏѦҲ ҦҸҡҲҽҲҸҍӁҸ ҏҰҸҲҝѦқѹҲѦґҧӁҽҾ ӊѹҦҸӊҽҾ Ұҽҝҽ ԗӢӠӡԽӠԙӸp–ԗӢӠԖӹԌӠԽӸԙԋi. ҫҲѸѦҲѹѸ ҲҽҔңѦ ҊґҽҊҸґ ԗӹӰԂԌӹԼӠԽӸԙԋp, ҏҔҽӉӃӊҽӀӂѹҮ Ӂҽ ӊҸӉӁѹҔӁҸӊѦӁѹѦ ҡѹҲҏҽ҃ѹѹ ҲѹҰҽ ӰԂԌӹ. ѨӁҲѦҝѦҡӁҸ ҸҲѸѦҲѹҲҧ, ҍҲҸ ӊ ӊѹҦҸӊҸѸ ҸҲӁҸҗѦӁѹѹ ҝҏҡҡҔѹҮ ҾӉӃҔ ҸҲґѹҍҽѦҲҡҾ ҸҲ ҦҝҏҊѹѽ ҡґҽӊҾӁҡҔѹѽ ҾӉӃҔҸӊ ҲҽҔңѦ Ұҝѹ ӊӃҝҽңѦӁѹѹ ӊҸӉѸҸңӁҸҡҲѹ/ҡҰҸҡҸқӁҸҡҲѹ. ҋѸ. Ҹқ ҙҲҸѸ ҝҽқҸҲҏ (ҷҽҝѦӁҲҡѦӁ 2002). 46 ѢѦ ѹҡҔґӀҍѦӁҸ, ҍҲҸ ҸҰҝѦҦѦґѦӁӁҏӀ ҝҸґҧ ҲҏҲ ѹҊҝҽѦҲ ӁҽґѹҍѹѦ ӊ ҊґҽӊӁҸҮ ҍҽҡҲѹ ҔҸӁҡҲҝҏҔ҃ѹѹ ҡ ѸҸҦҽґҧӁӃѸ ҊґҽҊҸґҸѸ. 47 Ѱ ҡҸӀӉҽ ӾԙԊӠ, ҸҍѦӊѹҦӁҸ, ӊҸӉѸҸңӁҸҡҲѹ җѹҝѦ. ҋѸ. ҡґѦҦҏӀӂѹҮ ҰҝѹѸѦҝ: Then there were doors that wouldn’t open unless you asked politely – ӷӹԽӹԐ ӾԷӾ ԼՂԊӠ ԌӴӾӢӠ, ӪӹԽӹӢՂӾ ӰӾ ҸҲҔҝӃӊҽґѹҡҧ, ӾԙԊӠ ӠԖ ӴӾԞԊӠӴӹ ӰӾ ԗӹԗӢӹԙӠԶӸ (“ѢҽҝҸҦӁӃҮ ҰѦҝѦӊҸҦ”, ӊ ѨӁҲѦҝӁѦҲѦ) (J.K. Rowling: Harry Potter and the Philosopher’s Stone). ҋґѦҦҏѦҲ ҸҲѸѦҲѹҲҧ, ҍҲҸ Ұҝѹ ҡҸӀӉҽѽ ҡґѦҦҸӊҽӁѹҾ, ӁҽҰҝѹѸѦҝ Ұҝѹ ҝҏҡҡҔҸѸ ҡҸӀӉѦ ԗӹӪԂ ӰӾ, ӍҸҝѸӃ ҰҝѦӉѦӁҡҽ ҲҽҔңѦ ҏҰҸҲҝѦқґҾӀҲҡҾ ҡҝҽӊӁѹҲѦґҧӁҸ ґѦҊҔҸ. ѢҸ ҝҽҡҡѸҸҲҝѦӁѹѦ ҰҝѦҦґҸңѦӁѹҮ ҡ ҲҽҔѹѸѹ ҡҸӀӉҽѸѹ ӊӃѽҸҦѹҲ Ӊҽ ҝҽѸҔѹ ӁҽҡҲҸҾӂѦҮ ҝҽқҸҲӃ. 48 ҶҲҸҲ ӊҽҝѹҽӁҲ ҰҝѦҦҡҲҽӊґѦӁ ӊ ҰҝѹѸѦҝѦ (13ӊ), ӁҸ ҲҽѸ ҦҽӁӁҽҾ ӍҸҝѸҽ ҏҰҸҲҝѦқґҾѦҲҡҾ ӊ ҸқѦѹѽ ҍҽҡҲҾѽ ҰҝѦҦґҸңѦӁѹҾ.

ҭҫҋһѣѩҫғүѷѣһѻѢҫҋѷѻ ѩѣҢҋѷғѨҢ ҭҩѨ ҭҫғѷҫҩѤѣҴҫҋѷѨ

29

ҦѦӁӁӃѽ ҡҸҸҲӊѦҲҡҲӊѹҮ ҝҽҡҡѸҸҲҝѦӁӁӃѦ ҾӉӃҔѹ ҝҽӉҦѦґҾӀҲҡҾ Ӂҽ 6 ҊҝҏҰҰ. ҋѸ. ҡґѦҦҏӀӂҏӀ ҡѽѦѸҏ: 49 ҝҏҡҡҔѹҮ қѦґҸҝҏҡҡҔѹҮ ҏҔҝҽѹӁҡҔѹҮ ҰҸґҧҡҔѹҮ ҍѦҗҡҔѹҮ ҡґҸӊҽ҃ҔѹҮ ҡґҸӊѦӁҡҔѹҮ ѽҸҝӊҽҲҡҔѹҮ ҡѦҝқҡҔѹҮ ѸҽҔѦҦҸӁҡҔѹҮ қҸґҊҽҝҡҔѹҮ

ҭҩѨѩ. ѯүҋѷѻ

ҐһүғѢүѤ ѯүҋѷѻ

ҰҝҸҗ. Ѣҋғ

ҰҝҸҗ. Ѣҋғ

ҰҝҸҗ. ҋғ / (Ѣҋғ)

ҰҝҸҗ. Ѣҋғ

ҰҝҸҗ. ҋғ

ҰҝҸҗ. ҋғ 50

ҔҸӁҦ. ҋғ

ҔҸӁҦ. ҋғ / (Ѣҋғ) ѹґѹ ҰҝҸҗ. Ѣҋғ (ԀӾ + ѹѸҰӍ. ҋғ) ѹѸҰӍ. Ѣҋғ ѹѸҰӍ. Ѣҋғ

(ԀӾ) 51 + ѹѸҰӍ. 52 ҋғ ѹѸҰӍ. ҋғ

ѤӉӃҔѹ ӊ ҡѽѦѸѦ ҝҽҡҰҸґҸңѦӁӃ ҔҽҔ ӊ ҰҝѦҦӃҦҏӂѹѽ ҰҝѹѸѦҝҽѽ. ҭҸҝҾҦҸҔ ӊӃҦѦґѦӁӁӃѽ ҊҝҏҰҰ ӊ Ҹқӂѹѽ ҍѦҝҲҽѽ ҡҸҸҲӊѦҲҡҲӊҏӀҲ ѹѽ ҊѦҸҊҝҽӍѹҍѦҡҔҸѸҏ ҝҽҡҰҸґҸңѦӁѹӀ. ҭҝѹ ҙҲҸѸ ѸӃ ҦӊѹңѦѸҡҾ ҰҸ ҡѦӊѦҝӁӃѸ ҾӉӃҔҽѸ ӊ ҸҡӁҸӊӁҸѸ ҡ ӊҸҡҲҸҔҽ Ӂҽ ӉҽҰҽҦ, ҽ ӁҽҍѹӁҽҾ ҡҸ ҡґҸӊѦӁҡҔҸҊҸ ҾӉӃҔҽ, ҰҸ ҲѦҝҝѹҲҸҝѹѹ ӀңӁҸҡґҽӊҾӁҡҔѹѽ ҾӉӃҔҸӊ – ҡ ӉҽҰҽҦҽ Ӂҽ ӊҸҡҲҸҔ. ҋѹґҽ ҰҝѹҔҝѦҰґѦӁӁҸҡҲѹ Ҕ ҲҸѸҏ ѹґѹ ҦҝҏҊҸѸҏ ӊѹҦҏ ӊ ҡѽѦѸѦ ҦҸҰҸґӁѹҲѦґҧӁҸ ҸҲѸѦҍѦӁҽ ҃ӊѦҲҸѸ: ҸҲ ѹҡҔґӀҍѹҲѦґҧӁҸҊҸ ҏҰҸҲҝѦқґѦӁѹҾ Ѣҋғ (ҲѦѸӁҸҡѦҝӃҮ ҃ӊѦҲ) ҦҸ ѹҡҔґӀҍѹҲѦґҧӁҸҊҸ ҏҰҸҲҝѦқґѦӁѹҾ ҋғ (қѦґӃҮ ҃ӊѦҲ). ҘҸѸҰҸӁѦӁҲӃ ӍҸҝѸӃ, ҡӊҾӉҽӁӁӃѦ ҡ ӁѦѹӁҦѹҔҽҲѹӊӁҸҮ ѸҸҦҽґҧӁҸҡҲҧӀ ҏҔҽӉҽӁӃ ңѹҝӁӃѸ җҝѹӍҲҸѸ.

49

ҭҸҦҍѦҝҔӁѦѸ, ҍҲҸ ӊ ҦҽӁӁҸҮ ҡѽѦѸѦ ѹӁӍҸҝѸҽ҃ѹҾ Ҹ ӊӃқҸҝѦ ӍҸҝѸ ӊ ҊґҽӊӁҸҮ ҍҽҡҲѹ ҸҲӁҸҡѹҲҡҾ ҲҸґҧҔҸ Ҕ ҲѦѸ ҡґҏҍҽҾѸ, ҔҸҊҦҽ ӊ ӁѦҮ ҸқҸӉӁҽҍѦӁҸ ҰҝѦҦѦґҧӁҸѦ ҦѦҮҡҲӊѹѦ. ҶҲҸ ӉӁҽҍѹҲ, ҍҲҸ ҡґҏҍҽѹ ҡ ҸҦӁҸӊѹҦҸӊӃѸѹ ҰҝѦҦѹҔҽҲҽѸѹ Ѣҋғ, ҔҽҔ ӊ ҰҝѹѸѦҝѦ (17), ӉҦѦҡҧ ӁѦ ҏҍҲѦӁӃ. ѷҽҔѹѸ ҸқҝҽӉҸѸ ҡѽѦѸҽ ҸҲҸқҝҽңҽѦҲ ҲѦ ҡґҏҍҽѹ, ҔҸҲҸҝӃѦ Ұҝѹ ӁѦҰҸӊҲҸҝҾѦѸҸҡҲѹ ҡҸҦѦҝңҽґѹ қӃ ӊ ҸқѦѹѽ ҍҽҡҲҾѽ ӍҸҝѸҏ ҰҝҸҗѦҦҗѦҊҸ ӊҝѦѸѦӁѹ ҋғ. ҶҲҸ ҰҸӉӊҸґҾѦҲ ѸҽҔҡѹѸҽґҧӁҸ ҰҝҸҡґѦҦѹҲҧ ӊґѹҾӁѹѦ ҰҸӊҲҸҝҾѦѸҸҡҲѹ Ӂҽ ӊӃқҸҝ ҊґҽҊҸґҧӁӃѽ ӍҸҝѸ ӊ ҝҽҡҡѸҽҲҝѹӊҽѦѸҸҮ ҡҲҝҏҔҲҏҝѦ. 50 ҈ҦѦҡҧ ӁѦ ҏҍѹҲӃӊҽӀҲҡҾ ҸҲҦѦґҧӁӃѦ ҡґҏҍҽѹ ҏҰҸҲҝѦқґѦӁѹҾ Ѣҋғ ӊ ҍѦҗҡҔҸѸ ҾӉӃҔѦ. ҋѸ. ҡҔҽӉҽӁӁҸѦ ӊӃҗѦ ҰҸ ҰҸӊҸҦҏ ҰҝѹѸѦҝҽ (14ҽ). 51 ҭҝѹ ӁѦҔҸҲҸҝӃѽ ҡҸӀӉҽѽ ҙҲҸҲ ҙґѦѸѦӁҲ ҸҰҏҡҔҽѦҲҡҾ, ҡѸ. (16). 52 ғ ҸҰҝѦҦѦґѦӁӁӃѽ ҏҡґҸӊѹҾѽ ӍҸҝѸҽ ѹѸҰѦҝӍѦҔҲҽ ӊ ѸҽҔѦҦҸӁҡҔҸѸ ѹ қҸґҊҽҝҡҔҸѸ ҾӉӃҔҽѽ ӉҽѸѦӁҾѦҲҡҾ ҡҸҸҲӊѦҲҡҲӊҏӀӂѹѸ “ҰѦҝѦҡҔҽӉӃӊҽҲѦґҧӁӃѸ” ҙҔӊѹӊҽґѦӁҲҸѸ. ғ ҙҲѹ ҰҸҦҝҸқӁҸҡҲѹ ѸӃ ӉҦѦҡҧ ӊѽҸҦѹҲҧ ӁѦ қҏҦѦѸ.

30

үѩҩѨүѢ ҷүҩѣѢѷҋѣѢ

ҋѽѦѸҽ ѽҸҝҸҗҸ ҰѦҝѦҦҽѦҲ, ҍҲҸ ҰҝѹѸѦҝӁҸ ҰҸґҸӊѹӁҽ ҾӉӃҔҸӊ ҸҲӁҸҡѹҲҡҾ Ҕ ҸҦӁҸҮ ѹӉ Ҧӊҏѽ ҔҝҽҮӁѹѽ ҲѹҰҸӊ – ӊ ҸқѦѹѽ ҍҽҡҲҾѽ ҰҝѦҦґҸңѦӁѹҾ ґѹқҸ ҲҸґҧҔҸ Ѣҋғ (ҝҏҡҡҔѹҮ, қѦґҸҝҏҡҡҔѹҮ, ҏҔҝҽѹӁҡҔѹҮ), ґѹқҸ ҲҸґҧҔҸ ҋғ (ҍѦҗҡҔѹҮ, ҡґҸӊҽ҃ҔѹҮ, ҡґҸӊѦӁҡҔѹҮ). ҶҲѹ ҔҝҽҮӁѹѦ ҰҸӉѹ҃ѹѹ ӉҽӁѹѸҽӀҲ ѹѸѦӁӁҸ ҲѦ ҾӉӃҔѹ, ҔҸҲҸҝӃѦ ѹ ӊ ҊѦҸҊҝҽӍѹҍѦҡҔҸѸ ҸҲӁҸҗѦӁѹѹ ҝҽҡҰҸґҸңѦӁӃ Ӂҽ ҰҝҸҲѹӊҸҰҸґҸңӁӃѽ ӉҸӁҽѽ ҡґҽӊҾӁҡҔҸҊҸ ҽҝѦҽґҽ. ѷҽҔѹѸ ҸқҝҽӉҸѸ ҰҝѦҸқґҽҦҽӁѹѦ Ѣҋғ Ұҝѹ ҰҸӊҲҸҝҾѦѸҸҡҲѹ ѸҸңӁҸ ҸѽҽҝҽҔҲѦҝѹӉҸӊҽҲҧ ҔҽҔ ҲѹҰѹҍӁҸ “ӊҸҡҲҸҍӁҸѦ” ҾӊґѦӁѹѦ, ҽ ҰҝѦҸқґҽҦҽӁѹѦ ҋғ, ҔҽҔ “ӉҽҰҽҦӁҸѦ”. 53 ѩҽґѦѦ ҡѽѦѸҽ ҰҸҔҽӉӃӊҽѦҲ, ҍҲҸ ӊ ҾӉӃҔҽѽ, ҝҽҡҰҸґҸңѦӁӁӃѽ ѸѦңҦҏ ҙҲѹѸѹ ҔҝҽҮӁѹѸѹ ҸқґҽҡҲҾѸѹ ҏҦѦґҧӁӃҮ ӊѦҡ Ѣҋғ ҰҸ ѸѦҝѦ ҦӊѹңѦӁѹҾ ҡ ӊҸҡҲҸҔҽ Ӂҽ ӉҽҰҽҦ ҡӁѹңҽѦҲҡҾ. ѨӁҲѦҝѦҡӁҸ ҸҲѸѦҲѹҲҧ, ҍҲҸ ҏ ҲѹҰѹҍӁӃѽ ҰҝѦҦҡҲҽӊѹҲѦґѦҮ “ӉҽҰҽҦӁҸҊҸ” ҲѹҰҽ ӁҽґѹҍѹѦ ҋғ ӊ ҊґҽӊӁҸҮ ҍҽҡҲѹ ҝҽҡҡѸҽҲҝѹӊҽѦѸӃѽ ҰҝѦҦґҸңѦӁѹҮ ҾӊґҾѦҲҡҾ ӁѦ ҲҸґҧҔҸ ҦҸҰҏҡҲѹѸӃѸ, ӁҸ, ҡҔҸҝѦѦ, ҸқҾӉҽҲѦґҧӁӃѸ. ҶҲҸ ҏҔҽӉӃӊҽѦҲ Ӂҽ ҲҸ, ҍҲҸ ҦҽӁӁӃѦ ҰҝѦҦґҸңѦӁѹҾ ӉҽӁѹѸҽӀҲ ҸҡҸқҏӀ ҰҸӉѹ҃ѹӀ ӊ ҰґҽӁѦ ӊӃқҸҝҽ ӊѹҦҽ. ғѦҦҧ ӊ ҝҾҦѦ ҦҝҏҊѹѽ ҡґҏҍҽѦӊ ӊ “ӉҽҰҽҦӁӃѽ” ҾӉӃҔҽѽ ҦґҾ ҸқҸӉӁҽҍѦӁѹҾ ҰҸӊҲҸҝҾӀӂѹѽҡҾ ҃ѦґҸҡҲӁӃѽ ҦѦҮҡҲӊѹҮ ҏҰҸҲҝѦқґҾѦҲҡҾ Ѣҋғ (ҡѸ. Ұҽҝ. 2). ѢҸ ҙҲҸ, ӊѹҦѹѸҸ, ӊҸӉѸҸңӁҸ ҲҸґҧҔҸ ҲҸҊҦҽ, ҔҸҊҦҽ ҔҸӁҲѦҔҡҲ ҲѦѸ ѹґѹ ѹӁӃѸ ҸқҝҽӉҸѸ ҸҲӊґѦҔҽѦҲ ӊӁѹѸҽӁѹѦ ҸҲ ҃ѦґҸҡҲӁҸҡҲѹ, “ҦѦӉҽҔҲҏҽґѹӉѹҝҏѦҲ” ҰҝѹӉӁҽҔ ҃ѦґҸҡҲӁҸҡҲѹ. ҫҍѦӊѹҦӁҸ, ҰҸҦҸқӁҽҾ ҦѦӉҽҔҲҏҽґѹӉҽ҃ѹҾ ҰҝҸҲѹӊҸҝѦҍѹҲ ѽҽҝҽҔҲѦҝҏ ҰҝѦҦґҸңѦӁѹҮ, ҝҽҡҡѸҸҲҝѦӁӁӃѽ ӊ Ұҽҝ. 3 ѹ 4. ғҡҾ ҡҏҲҧ ҰҸҦҸқӁӃѽ ҰҝѦҦґҸңѦӁѹҮ ҔҽҔ ҝҽӉ ҰҝѦҦҰҸґҽҊҽѦҲ ҸҡҸқҸѦ ӊӁѹѸҽӁѹѦ Ҕ ӊҝѦѸѦӁӁӃԀѸ ҸҲӁҸҗѦӁѹҾѸ ѸѦңҦҏ ҦѦҮҡҲӊѹҾѸѹ ҊґҽӊӁҸҮ ѹ ҰҝѹҦҽҲҸҍӁҸҮ ҍҽҡҲѦҮ. ҭҸҙҲҸѸҏ Ұҝѹ ҰҸҡґѦҦҸӊҽҲѦґҧӁҸҡҲѹ ҙҲѹѽ ҦѦҮҡҲӊѹҮ ӊ ҃ѦӁҲҝѦ ӊӁѹѸҽӁѹҾ ӁҽѽҸҦҾҲҡҾ ѹѸѦӁӁҸ ѹѽ ҊҝҽӁѹ҃Ӄ ѹ ӊ ҦҽӁӁӃѽ ҾӉӃҔҽѽ ҙҲҸ ѦҡҲѦҡҲӊѦӁӁӃѸ ҸқҝҽӉҸѸ ӊѦҦѦҲ Ҕ ӊӃқҸҝҏ ҋғ ҔҽҔ ӊ ҰҝѹҦҽҲҸҍӁҸҮ, ҲҽҔ ѹ ӊ ҊґҽӊӁҸҮ ҍҽҡҲҾѽ. ҫҦӁҽҔҸ ҰҝѹӊѦҦѦӁӁӃѦ ҡҸҸқҝҽңѦӁѹҾ ѹѸѦӀҲ ҡѹґҏ ҲҸґҧҔҸ ҰҸ ҸҲӁҸҗѦӁѹӀ Ҕ ҾӉӃҔҽѸ ҔҝҽҮӁѦ ӉҽҰҽҦӁҸҊҸ ҲѹҰҽ. ғ қҸґҧҗѹӁҡҲӊѦ ҸҡҲҽґҧӁӃѽ ҾӉӃҔҸӊ ӊ ҊґҽӊӁҸҮ ҍҽҡҲѹ ӊҸӉѸҸңѦӁ ҲҸґҧҔҸ Ѣҋғ. ѨҡҔґӀҍѦӁѹҾѸѹ ҾӊґҾӀҲҡҾ ѸҽҔѦҦҸӁҡҔѹҮ ѹ ӊ ҸҡҸқѦӁӁҸҡҲѹ ҡѦҝқҡҔѹҮ ѹ ѽҸҝӊҽҲҡҔѹҮ, ӁҸ ҰҝѹѸѦҍҽҲѦґҧӁҸ, ҍҲҸ ӊ ҙҲѹѽ ҲҝѦѽ ҾӉӃҔҽѽ ӊӃқҸҝ ҋғ ӊ ҊґҽӊӁҸҮ ҍҽҡҲѹ ҦҸҰҏҡҔҽѦҲҡҾ ҲҸґҧҔҸ Ұҝѹ ґѹҍӁҸҮ ӍҸҝѸѦ ҡ ѸҸҦҽґҧӁӃѸ ҔҸѸҰҸӁѦӁҲҸѸ.

53 ғ ѹӁҲѦҝѦҡӁҸҮ ҝҽқҸҲѦ ѩѹҔѹ (Dickey 2000) ҏқѦҦѹҲѦґҧӁҸ ҰҸҔҽӉҽӁҸ, ҍҲҸ ӊҸҡҲҸҍӁӃҮ ѹ ӉҽҰҽҦӁӃҮ ҲѹҰӃ ҝҽӉґѹҍҽӀҲҡҾ ҲҽҔңѦ ҰҸ ҃ѦґҸѸҏ ҝҾҦҏ ҦҝҏҊѹѽ ӊѹҦҸӊӃѽ “ҰҽҝҽѸѦҲҝҸӊ”. Ҙ ҲѹҰѹҍӁӃѸ ӉҽҰҽҦӁӃѸ ҾӉӃҔҽѸ ѩѹҔѹ ҸҲӁҸҡѹҲ ѹ ҡѦҝқҸґҏңѹ҃ҔѹѦ ҾӉӃҔѹ. ҭҝѹӊѦҦѦӁӁӃҮ ӊӃҗѦ ѦҦѹӁҡҲӊѦӁӁӃҮ ҰҝѹѸѦҝ ѹӉ ӊѦҝѽӁѦґҏңѹ҃ҔҸҊҸ (17) ҙҲҸѸҏ ӁѦ ҰҝҸҲѹӊҸҝѦҍѹҲ. ҷҸґҊҽҝҡҔѹҮ ҾӉӃҔ ѩѹҔѹ ҸҲӁҸҡѹҲ Ҕ ӊҸҡҲҸҍӁҸҮ ҊҝҏҰҰѦ. ѢҽҗѦ ѹҡҡґѦҦҸӊҽӁѹѦ ҰҸҔҽӉӃӊҽѦҲ, ҍҲҸ ҡ ҦҝҏҊѹѸѹ ҾӉӃҔҽѸѹ ҙҲҸҮ ҊҝҏҰҰӃ ѦҊҸ ҦѦҮҡҲӊѹҲѦґҧӁҸ ҡӊҾӉӃӊҽѦҲ ҸқҾӉҽҲѦґҧӁҸѦ ҏҰҸҲҝѦқґѦӁѹѦ Ѣҋғ ӊ ҊґҽӊӁҸҮ ҍҽҡҲѹ. ѢҸ ӊ ҸҲґѹҍѹѦ ҸҲ Ӂѹѽ ӊ ӉҽӊѹҡѹѸҸҮ ҍҽҡҲѹ ӊ қҸґҊҽҝҡҔҸѸ ҏҰҸҲҝѦқґҾѦҲҡҾ ҋғ, ҍҲҸ ҾӊӁҸ ҡӊҾӉҽӁҸ ҡ ҡҏӂѦҡҲӊҸӊҽӁѹѦѸ ҡҰѦ҃ѹӍѹҍѦҡҔҸҮ ӍҸҝѸӃ ѹѸҰѦҝӍѦҔҲҽ ҋғ.

ҭҫҋһѣѩҫғүѷѣһѻѢҫҋѷѻ ѩѣҢҋѷғѨҢ ҭҩѨ ҭҫғѷҫҩѤѣҴҫҋѷѨ

31

ғ Ҹқӂѹѽ ҍѦҝҲҽѽ ѸҸңӁҸ ҡҔҽӉҽҲҧ, ҍҲҸ ӊӃқҸҝ ӍҸҝѸӃ ӊ ҊґҽӊӁҸҮ ҍҽҡҲѹ ӍҽҔҲѹҍѦҡҔѹ ӁѦ ҸҲґѹҍҽѦҲҡҾ ҸҲ ҲҸҊҸ, ҍҲҸ ӁҽѽҸҦѹѸ ӊ ҸқӃҍӁӃѽ ӁѦӉҽӊѹҡѹѸӃѽ ҰҝѦҦґҸңѦӁѹҾѽ. (ѷҸґҧҔҸ ҏ ҔҝҽҮӁѹѽ ӉҽҰҽҦӁӃѽ ҾӉӃҔҸӊ ӁҽқґӀҦҽѦҲҡҾ ҰҸӊӃҗѦӁӁҽҾ ҰҸҲҝѦқӁҸҡҲҧ ӊӃқҸҝҽ ҋғ.) ҘҽҔ ҰҸҔҽӉӃӊҽѦҲ ҡѽѦѸҽ, ҡѹҲҏҽ҃ѹҾ ҰҸ ҸҲӁҸҗѦӁѹӀ Ҕ ӉҽӊѹҡѹѸҸҮ ҍҽҡҲѹ ҾӊґҾѦҲҡҾ ҡҸӊѦҝҗѦӁӁҸ ҦҝҏҊҸҮ. ғ қҸґҧҗѹӁҡҲӊѦ ҾӉӃҔҸӊ ӊӃқҸҝ ӍҸҝѸӃ ӊ ҙҲҸҮ ҰҸӉѹ҃ѹѹ ҡѹґҧӁҸ ҸҲґѹҍҽѦҲҡҾ ҸҲ ҲҸҊҸ, ҍҲҸ ӁҽѽҸҦѹѸ ӊ ӁѦӉҽӊѹҡѹѸӃѽ ҰҝѦҦґҸңѦӁѹҾѽ. ҭҝҸҾӊґҾѦҲҡҾ ӊѦҡҧѸҽ ҡѹґҧӁҽҾ ҲѦӁҦѦӁ҃ѹҾ Ҕ ӊӃқҸҝҏ ҋғ. ѷҸґҧҔҸ ӁҽѹқҸґѦѦ ҔҝҽҮӁѹѦ ҰҝѦҦҡҲҽӊѹҲѦґѹ ӊҸҡҲҸҍӁҸҊҸ ӊѹҦҸӊҸҊҸ ҲѹҰҽ ҦҸҰҏҡҔҽӀҲ ѹҡҔґӀҍѹҲѦґҧӁҸ Ѣҋғ. 54 ѢҸ ҰҸҍҲѹ ӊҸ ӊҡѦѽ ҸҡҲҽґҧӁӃѽ ҾӉӃҔҽѽ ѦҦѹӁҡҲӊѦӁӁҸ ӊҸӉѸҸңӁӃѸ ҾӊґҾѦҲҡҾ ҔҽҔ ҝҽӉ ҋғ. ҫҦӁҽҔҸ ӊӁѦ ҝҽѸҸҔ ҔҝҽҮӁѦ ӉҽҰҽҦӁҸҊҸ ҲѹҰҽ ҙҲҸ ӊѦҦѦҲ Ҕ ҦҸҰҸґӁѹҲѦґҧӁӃѸ ҸҊҝҽӁѹҍѦӁѹҾѸ: ӊ ѽҸҝӊҽҲҡҔҸѸ, ҡѦҝқҡҔҸѸ, ѸҽҔѦҦҸӁҡҔҸѸ ѹ қҸґҊҽҝҡҔҸѸ ӁѦ ҦҸҰҏҡҔҽѦҲҡҾ ҸқӃҍӁҽҾ ӍҸҝѸҽ ҰҝҸҗѦҦҗѦҊҸ ӊҝѦѸѦӁѹ ҋғ ѹӁҦѹҔҽҲѹӊҽ. ғ ѽҸҝӊҽҲҡҔҸѸ ѹ ҡѦҝқҡҔҸѸ ӊӃқѹҝҽѦҲҡҾ ӍҸҝѸҽ ҔҸӁҦѹ҃ѹҸӁҽґҽ, ҔҸҲҸҝӃҮ ѹѸѦӁӁҸ ӊ ҙҲѹѽ ҾӉӃҔҽѽ ӉҽѸѦҲӁҸ ҝҽҡҗѹҝѹґ ҔҝҏҊ ҡӊҸѹѽ ӉӁҽҍѦӁѹҮ. ү ӊ ѸҽҔѦҦҸӁҡҔҸѸ ѹ қҸґҊҽҝҡҔҸѸ ѹҡҰҸґҧӉҏӀҲҡҾ ӍҸҝѸӃ ѹѸҰѦҝӍѦҔҲҽ ҋғ, ҔҸҲҸҝӃѦ, ҸҡҸқѦӁӁҸ ӊ қҸґҊҽҝҡҔҸѸ ҾӉӃҔѦ, ӊӁѦ ҏҔҽӉҽӁӁҸҮ ӉҽӊѹҡѹѸҸҮ ҰҸӉѹ҃ѹѹ ҏҰҸҲҝѦқґҾӀҲҡҾ ҝѦҦҔҸ. ѨӁҲѦҝѦҡӁҸ, ҍҲҸ ӊ ѸҽҔѦҦҸӁҡҔҸѸ ҾӉӃҔѦ Ұҝѹ ҡҸӀӉҽѽ ҲѹҰҽ ԶԽӹԐ, ԙԂԐӹ (ԶԽӹ) ѹ ӪӹԄԂ ӍҸҝѸҽ ѹѸҰѦҝӍѦҔҲҽ ҋғ ҡҸҍѦҲҽѦҲҡҾ ҡ ѸҸҦҽґҧӁҸҮ ҍҽҡҲѹ҃ѦҮ ԀӾ, ҍѦѸ ҸӁ ҸҲґѹҍҽѦҲҡҾ ҸҲ қҸґҊҽҝҡҔҸҊҸ ҾӉӃҔҽ. 55 ѣҦѹӁҡҲӊѦӁӁӃѸ ҾӉӃҔҸѸ, ҏ ҔҸҲҸҝҸҊҸ ӊ ӉҽӊѹҡѹѸҸҮ ҍҽҡҲѹ ӊҡҲҝѦҍҽӀҲҡҾ ҊґҽҊҸґӃ ҸқҸѹѽ ӊѹҦҸӊ ҾӊґҾѦҲҡҾ ҰҸґҧҡҔѹҮ. 56 ҩҽҡҡѸҽҲҝѹӊҽѦѸӃѦ ӁҽѸѹ ҡґҏҍҽѹ ҰҸҦҲӊѦҝңҦҽӀҲ, ҍҲҸ ҙҲҸҲ ҾӉӃҔ ӉҽӁѹѸҽѦҲ ҰѦҝѦѽҸҦӁҏӀ ҰҸӉѹ҃ѹӀ ѸѦңҦҏ ӊҸҡҲҸҍӁҸҮ ѹ ӉҽҰҽҦӁҸҮ ҡґҽӊҾӁҡҔѹѸѹ ӊѹҦҸӊӃѸѹ ҡѹҡҲѦѸҽѸѹ (ҡѸ. Dickey 2000: 5). ҋѽҸҦҡҲӊҸ ҡ ӊҸҡҲҸҍӁӃѸѹ ҾӉӃҔҽѸѹ ҰҝҸҾӊґҾѦҲҡҾ ӊ ӍҽҔҲѹҍѦҡҔҸҮ ҸқҾӉҽҲѦґҧ54

ѣҡґѹ ҍҏӊҡҲӊҏѦҲҡҾ ӁѦҸқѽҸҦѹѸҸҡҲҧ ӁѦҦӊҏҡѸӃҡґѦӁӁҸ ӊӃҝҽӉѹҲҧ ӊѹҦҸӊҸҮ ѽҽҝҽҔҲѦҝ ѸѹҔҝҸҡѹҲҏҽ҃ѹѹ, ҙҲѹ ҾӉӃҔѹ ѸҸҊҏҲ ѹҡҰҸґҧӉҸӊҽҲҧ ҔҸӁҡҲҝҏҔ҃ѹѹ ҡ ҊґҽҊҸґҽѸѹ ҲѹҰҽ ӫԙԗӾӴԂԽӸ ѹґѹ ԙԽӹӠԽӸ (ҡѸ. ҰҝѹѸѦҝӃ (13), (13ҽ) ѹ (16)). ҭҝѹ ҸҦӁҸҝѦӍѦҝѦӁҲӁҸҡҲѹ ҡҏқӆѦҔҲҸӊ ѽҸҝҸҗѦҮ ҽґҧҲѦҝӁҽҲѹӊҸҮ ҦґҾ ҰҝѹҦҽҲҸҍӁҸҊҸ ҰҝѦҦґҸңѦӁѹҾ ѸҸңѦҲ қӃҲҧ ҦѦѦҰҝѹҍҽҡҲѹѦ ҋғ (ҡѸ. (16ҽ)). 55 ѨӁҲѦҝѦҡӁҸ ҸҲѸѦҲѹҲҧ, ҍҲҸ ҡҸҍѦҲҽӁѹѦ ҍҽҡҲѹ҃Ӄ ԀӾ ҡ ӍҸҝѸҸҮ ҰҝѦӉѦӁҡҽ ҋғ ҲҽҔңѦ җѹҝҸҔҸ ҏҰҸҲҝѦқґҾѦҲҡҾ ҦґҾ ҸқҸӉӁҽҍѦӁѹҾ ҰҸӊҲҸҝҾӀӂѹѽҡҾ ҡѹҲҏҽ҃ѹҮ. ғ қҸґҊҽҝҡҔѹѽ ҡҸҸҲӊѦҲҡҲӊѹҾѽ ѹҡҰҸґҧӉҏѦҲҡҾ ҰҝҸҡҲҽҾ ӍҸҝѸҽ ҰҝѦӉѦӁҡҽ ҋғ. ѩҽӁӁҸѦ ҡҸҸҲӊѦҲҡҲӊѹѦ ӁҽѽҸҦѹѸ ҲҽҔңѦ ӊ ҝҽѸҔҽѽ ӊҝѦѸѦӁӁҸҮ ҰҝѹҦҽҲҸҍӁҸҮ ҍҽҡҲѹ ҰҝѦҦґҸңѦӁѹҾ, ҸҰѹҡӃӊҽӀӂѦҊҸ ҰҸӊҲҸҝҾӀӂҏӀҡҾ ҃ѦҰҸҍҔҏ ҦѦҮҡҲӊѹҮ, ӁѦ ҸҲӁҸҡҾӂҏӀҡҾ Ҕ ҰҝҸҗґҸѸҏ. ҋҝ. ѸҽҔѦҦҸӁҡҔѹҮ ѹ қҸґҊҽҝҡҔѹҮ ҰѦҝѦӊҸҦӃ ҸҦӁҸҊҸ ҰҝѦҦґҸңѦӁѹҾ ѹӉ ҦӁѦӊӁѹҔҽ үӁӁӃ ҨҝҽӁҔ: ԹӹԄԂ ԀӾ [exp.] ӴԊӾԒӾp [pres.] Ԅ. Թ[...], ԙӹӰՁӾԽӹ ӉҽҡӊѦҲҏӊҽi [pres.] – ԹӹԄԂԽӹ ԄӹԙԗӹԌӠӰ Թ[...] ӴԊӾԒӾp [pres.], ӴԙӾ ӾԌӰӹ ԙԊԔӰՁӾԽӹ ѹӉҊҝҾӊҽi [pres.]. (ҋѸ. ҲҽҔңѦ ҊҸґґҽӁҦҡҔѹҮ ҸҝѹҊѹӁҽґ ѹ ҝҏҡҡҔѹҮ ҰѦҝѦӊҸҦ: Als mijnheer K[...] binnenkomt [pres.], gaat [pres.] de zon op – ԹӹԄԌԂ ԗӢӠԖӹԌӠԽi ԄӹԙԗӹԌӠӰ Թ[...], Խӹ ӊҸҡѽҸҦѹҲi ԙӹԊӰՁӾ.) 56 Ѣҽ ҦҽӁӁҸѦ ҝҽӉґѹҍѹѦ ѸѦңҦҏ ҰҸґҧҡҔѹѸ ѹ ҝҏҡҡҔѹѸ ҾӉӃҔҽѸѹ ҏҔҽӉӃӊҽѦҲ ү. ҤҸґӊҏҲ (Holvoet 1991: 123). ҫҦӁҽҔҸ ӊ ѦҊҸ ҡҲҽҲҧѦ ӁѦ ҸҲѸѦҍѦӁҸ, ҍҲҸ ӊ ҰҸґҧҡҔҸѸ ҾӉӃҔѦ ӊ ҙҲҸҮ ҰҸӉѹ҃ѹѹ ӁҽѹқҸґѦѦ ҸқӃҍӁҸҮ ӍҸҝѸҸҮ ҾӊґҾѦҲҡҾ ѹѸѦӁӁҸ ҋғ.

32

үѩҩѨүѢ ҷүҩѣѢѷҋѣѢ

ӁҸҡҲѹ Ѣҋғ ӊ ҊґҽӊӁҸҮ ҍҽҡҲѹ. ѢҸ ҰҸ ҸҲӁҸҗѦӁѹӀ Ҕ ҰҝѦѹѸҏӂѦҡҲӊѦӁӁҸѸҏ ӊӃқҸҝҏ ҋғ ӊ ҰҝѹҦҽҲҸҍӁҸҮ ҍҽҡҲѹ ҰҸґҧҡҔѹҮ ҾӉӃҔ ҡѹґҧӁҸ ҡқґѹңҽѦҲҡҾ ҡ ӉҽҰҽҦӁӃѸѹ ҾӉӃҔҽѸѹ. ҫҦӁҽҔҸ, ҔҽҔ қӃґҸ ҸҲѸѦҍѦӁҸ Ұҝѹ ҸқҡҏңҦѦӁѹѹ ҰҝѹѸѦҝҸӊ (12) ѹ (15), ӊ ҰҸґҧҡҔҸѸ ҾӉӃҔѦ ӊҡҲҝѦҍҽӀҲҡҾ ҲҽҔңѦ ҰҝѹѸѦҝӃ, ӊ ҔҸҲҸҝӃѽ ҰҝѹҦҽҲҸҍӁҽҾ ҍҽҡҲҧ ҡҸҦѦҝңѹҲ ӍҸҝѸҏ Ѣҋғ, ӊ ҲҸ ӊҝѦѸҾ ҔҽҔ ӊ ҡҸҸҲӊѦҲҡҲӊѹҾѽ ѹӉ ҰҝҸҲҸҲѹҰѹҍѦҡҔѹѽ ӉҽҰҽҦӁӃѽ ҾӉӃҔҸӊ ӁҽѽҸҦѹѸ ҲҸґҧҔҸ ҋғ. ҭҸҔҽ ӁҽѸ ӁѦ ҾҡӁҸ, ѸҸңӁҸ ґѹ ҸқӆҾҡӁѹҲҧ ҙҲҸ ӁѦҔҸҲҸҝҸҮ ӍҽҔҏґҧҲҽҲѹӊӁҸҡҲҧӀ ҦҽӁӁҸҊҸ ӊӃқҸҝҽ ѹґѹ ҡҏӂѦҡҲӊҏѦҲ ҦҝҏҊҸѦ ҸқӆҾҡӁѦӁѹѦ. ҴӃ ӁѦ ѹҡҔґӀҍҽѦѸ ҲҽҔҏӀ ӊҸӉѸҸңӁҸҡҲҧ, ҍҲҸ ӊӃқҸҝ Ѣҋғ ӊ ҦҽӁӁҸҮ ҰҸӉѹ҃ѹѹ ӊҡѦ ңѦ ӊ ҸҡӁҸӊӁҸѸ ѸҸҲѹӊѹҝҏѦҲҡҾ ҡѦѸҽӁҲѹҍѦҡҔѹѸѹ ӍҽҔҲҸҝҽѸѹ ӊӁѦ ҰҸӊҲҸҝҾѦѸҸҡҲѹ. ҘҽҔ қӃґҸ ҸҲѸѦҍѦӁҸ ӊӃҗѦ, ӊ ҝҾҦѦ ҡґҏҍҽѦӊ ҊҝҽӁѹ҃ҽ ѸѦңҦҏ ҰҝѦҦҗѦҡҲӊҸӊҽӁѹѦѸ ѹ ҸҦӁҸӊҝѦѸѦӁӁҸҡҲҧӀ ҡҲѹҝҽѦҲҡҾ ӁҽҡҲҸґҧҔҸ, ҍҲҸ ҡҸҸҲӁҸҗѦӁѹҾ ҸҰѹҡӃӊҽѦѸӃѽ ҡѹҲҏҽ҃ѹҮ ґѦҊҔҸ ӊҸҡҰҝѹӁѹѸҽӀҲҡҾ ҦӊҸҾҔѹѸ ҸқҝҽӉҸѸ. ҶҲҸ ѸҸҊґҸ қӃ ѹҊҝҽҲҧ ӁѦҔҸҲҸҝҏӀ ҝҸґҧ ѹ ӊ ҰҝѹѸѦҝѦ (15). ҭҸӁҾҲѹѦ “ӊҡҲҝѦҍҽ” ѸҸңӁҸ ӊѦҦҧ ѹҡҲҸґҔҸӊҽҲҧ ҔҽҔ ѸҸѸѦӁҲҽґҧӁҸѦ “ӊҡҲҏҰґѦӁѹѦ ӊ ҔҸӁҲҽҔҲ”, ӁҸ ѹ ҔҽҔ ҸқҸӉӁҽҍѦӁѹѦ ӊҸӉӁѹҔҗѦҮ Ұҝѹ ҙҲҸѸ ҡѹҲҏҽ҃ѹѹ. ҘҽҔ ӁҽѸ ҰҝѦҦҡҲҽӊґҾѦҲҡҾ, ҰҸҡґѦҦӁѹҮ ҡѸӃҡґ ҰҝѹҡҏҲҡҲӊҏѦҲ ӊ ӊӃҝҽңѦӁѹѹ ԗӢӠ ӴԙԽӢӾԓӾ. ѷҽҔѹѸ ҸқҝҽӉҸѸ ӊѹҦҸӊҸѦ “ҸҲҔґҸӁѦӁѹѦ” ӊ ҰҸґҧҡҔҸѸ ҰѦҝѦӊҸҦѦ ҰҝѹѸѦҝҽ (15) ѸҸҊґҸ қӃ ҸқӆҾҡӁҾҲҧҡҾ ңѦґҽӁѹѦѸ ҰҸҦҍѦҝҔӁҏҲҧ ӁѦҔҸҲҸҝӃҮ ҸҲҲѦӁҸҔ ҸҦӁҸӊҝѦѸѦӁӁҸҡҲѹ. ғ ҡӊҾӉѹ ҡ ҲҽҔѹѸѹ ҰҝҸқґѦѸҽѸѹ ӊҸҰҝҸҡ Ҹ ѽҽҝҽҔҲѦҝѦ ӊӃқҸҝҽ ӊѹҦҽ ӊ ҰҸґҧҡҔѹѽ ҰҝѦҦґҸңѦӁѹҾѽ ҝҽҡҡѸҽҲҝѹӊҽѦѸҸҊҸ ҲѹҰҽ ӉҽҡґҏңѹӊҽѦҲ ҸҲҦѦґҧӁҸҊҸ ѹҡҡґѦҦҸӊҽӁѹҾ Ӂҽ ҸқҗѹҝӁҸѸ ѸҽҲѦҝѹҽґѦ ѹ, ҰҸ ӊҸӉѸҸңӁҸҡҲѹ, ӊ ҡҸҲҝҏҦӁѹҍѦҡҲӊѦ ҡ ӁҸҡѹҲѦґҾѸѹ ҰҸґҧҡҔҸҊҸ ҾӉӃҔҽ. 6. ҈ҽҔґӀҍѦӁѹѦ ҴҽҲѦҝѹҽґ, ҰҝѦҦҡҲҽӊґѦӁӁӃҮ ӊ ҦҽӁӁҸҮ ҡҲҽҲҧѦ, ҰҸҔҽӉӃӊҽѦҲ, ҍҲҸ ӊ қҸґҧҗѹӁҡҲӊѦ ҡґҽӊҾӁҡҔѹѽ ҾӉӃҔҸӊ Ұҝѹ ҸқҸӉӁҽҍѦӁѹѹ ҰҸӊҲҸҝҾӀӂѦҮҡҾ ӊ ҰҝҸҗґҸѸ ҃ѦҰҸҍҔѹ ҦѦҮҡҲӊѹҮ ӊӃқҸҝ ҊґҽҊҸґҧӁҸҮ ӍҸҝѸӃ ӊ ҰҝѹҦҽҲҸҍӁҸҮ ҍҽҡҲѹ ӊҝѦѸѦӁӁҸҊҸ ҡґҸңӁҸҰҸҦҍѹӁѦӁӁҸҊҸ ҰҝѦҦґҸңѦӁѹҾ ҸҰҝѦҦѦґҾѦҲҡҾ ҦҸӊҸґҧӁҸ ҡҰѦ҃ѹӍѹҍѦҡҔѹѸѹ ҰҝҽӊѹґҽѸѹ. ғ ҸҡӁҸӊӁҸѸ ѹѽ ѸҸңӁҸ ѹҡҲҸґҔҸӊҽҲҧ ҔҽҔ ҝҽӉӁӃѦ ҡҰҸҡҸқӃ ӊӃҝҽңѦӁѹҾ ӊѹҦҸӊҸҊҸ ѽҽҝҽҔҲѦҝҽ ѸѹҔҝҸҡѹҲҏҽ҃ѹѹ. ҭҸӊӃҗѦӁӁҽҾ ӊҸӉѸҸңӁҸҡҲҧ ҸҝѹѦӁҲѹҝҸӊҽҲҧҡҾ Ӂҽ ѸѹҔҝҸҡѹҲҏҽ҃ѹӀ ҰҝҸҾӊґҾѦҲҡҾ ӊ қҸґҧҗѹӁҡҲӊѦ ҡґҽӊҾӁҡҔѹѽ ҾӉӃҔҸӊ ѹѸѦӁӁҸ ӊ ҝҽѸҔҽѽ ҡґҸңӁҸҰҸҦҍѹӁѦӁӁҸҊҸ ҰҝѦҦґҸңѦӁѹҾ, ҸҡҸқѦӁӁҸ ӊ ѦҊҸ ҰҝѹҦҽҲҸҍӁҸҮ ҍҽҡҲѹ. ҫҍѦӊѹҦӁҸ, ҙҲҸ ҡӊҾӉҽӁҸ ҡ ӉҽӊѹҡѹѸӃѸ ѽҽҝҽҔҲѦҝҸѸ ҰҝѹҦҽҲҸҍӁҸҮ ҍҽҡҲѹ. ҭҝѹѸѦҍҽҲѦґҧӁҸ, ҍҲҸ ӊ ҝҏҡҡҔҸѸ ҾӉӃҔѦ ҲҽҔҽҾ ӊҸӉѸҸңӁҸҡҲҧ ӁҽқґӀҦҽѦҲҡҾ ѹѸѦӁӁҸ Ұҝѹ ҦѦѦҰҝѹҍҽҡҲѹѹ, ҾӊґҾӀӂѦѸҡҾ ҡҰѦ҃ѹӍѹҍѦҡҔѹ ӉҽӊѹҡѹѸҸҮ ӍҸҝѸҸҮ (ҡҝ. (16ҽ)). ҫҦӁҽҔҸ ӊ ҝҏҡҡҔҸѸ ҾӉӃҔѦ (ҔҽҔ ѹ ӊ ҦҝҏҊѹѽ ҾӉӃҔҽѽ ҔҝҽҮӁѦ ӊҸҡҲҸҍӁҸҊҸ ҲѹҰҽ) ӉҽӊѹҡѹѸӃҮ ѽҽҝҽҔҲѦҝ ҰҝѹҦҽҲҸҍӁҸҮ ҍҽҡҲѹ ӊҡѦ ңѦ ӁѦ ӊ ҡѹґҽѽ ҰҝѦҸҦҸґѦҲҧ ҡѹґҧӁҏӀ ҡӊҾӉҧ ѸѦңҦҏ ҰҸӊҲҸҝҾѦѸҸҡҲҧӀ ѹ Ѣҋғ, ҔҸҲҸҝҽҾ ҰҝҸҾӊґҾѦҲҡҾ ӊ ҸҡҸқѦӁӁҸҡҲѹ Ұҝѹ ӍҸҝѸҽѽ ҰҝҸҗѦҦҗѦҊҸ ӊҝѦѸѦӁѹ. ҭҝѦҦҡҲҸѹҲ ѦӂѦ ҝҽҡҡѸҸҲҝѦҲҧ, ӁҽҡҔҸґҧҔҸ ҙҲҸҲ ӍҽҔҲҸҝ ҦѦҮҡҲӊҏѦҲ ӊ ҰҝѦҦґҸңѦӁѹҾѽ, ҸқҸӉӁҽҍҽӀӂѹѽ ӁѦҰҝҸҗѦҦҗѹѦ ҦѦҮҡҲӊѹҾ.

ҭҫҋһѣѩҫғүѷѣһѻѢҫҋѷѻ ѩѣҢҋѷғѨҢ ҭҩѨ ҭҫғѷҫҩѤѣҴҫҋѷѨ

33

ҩҽҡҡѸҸҲҝѦӁӁӃѦ ҰҝѹѸѦҝӃ ҰҸҔҽӉӃӊҽӀҲ, ҍҲҸ ӊ ҦҽӁӁҸҮ ҸқґҽҡҲѹ ѸѦңҦҏ ҡґҽӊҾӁҡҔѹѸѹ ҾӉӃҔҽѸѹ ӁҽқґӀҦҽӀҲҡҾ ӊѦҡҧѸҽ ӉҽѸѦҲӁӃѦ ҝҽӉґѹҍѹҾ. ҭҸ ӁҽҗѦѸҏ ѸӁѦӁѹӀ, ҦҽӁӁӃҮ ѸҽҲѦҝѹҽґ ӁҽҊґҾҦӁҸ ҰҸҔҽӉӃӊҽѦҲ, ҍҲҸ ҸқҡҏңҦҽѦѸӃѦ ҝҽӉґѹҍѹҾ ӊ ӊӃқҸҝѦ ӊѹҦҽ ҡґҸңӁӃѸ ҸқҝҽӉҸѸ ҡӊҾӉҽӁӃ ҡ ҝҽӉґѹҍѹҾѸѹ ӊ ҡҏӂѦҡҲӊҸӊҽӁѹѹ ѹ ҏҰҸҲҝѦқґѦӁѹѹ ҦҝҏҊѹѽ ҊґҽҊҸґҧӁӃѽ ҔҽҲѦҊҸҝѹҮ (ӁҽҰҝ. ҔҸӁҦѹ҃ѹҸӁҽґҽ ѹ ѹѸҰѦҝӍѦҔҲҽ ҋғ). ҫҡҸқѦӁӁҸ ѹӁҲѦҝѦҡӁӃѸ ӁҽѸ ҰҝѦҦҡҲҽӊґҾѦҲҡҾ ҲҸҲ ӍҽҔҲ, ҍҲҸ ӊ ҝҾҦѦ ӀңӁҸҡґҽӊҾӁҡҔѹѽ ҾӉӃҔҸӊ ҲҽҔҏӀ ӊҽңӁҏӀ ҝҸґҧ ѹҊҝҽӀҲ ӉҦѦҡҧ ѸҸҦҽґҧӁҸ ѸҽҝҔѹҝҸӊҽӁӁӃѦ ӍҸҝѸӃ (ҔҸӁҦѹ҃ѹҸӁҽґ ӊ ѽҸҝӊҽҲҡҔҸѸ ѹ ҡѦҝқҡҔҸѸ ѹ ҍҽҡҲѹ҃ҽ ԀӾ ӊ ѸҽҔѦҦҸӁҡҔҸѸ). ғ ҏҡҲҽӁҸӊґѦӁѹѹ ҰҸҦҸқӁӃѽ ӍҽҔҲҸҝҸӊ ѸӃ ѹ ӊѹҦѹѸ ӉӁҽҍѦӁѹѦ ҡҸҰҸҡҲҽӊѹҲѦґҧӁӃѽ ѹҡҡґѦҦҸӊҽӁѹҮ. ҴӃ ӁҽҦѦѦѸҡҾ, ҍҲҸ ӊ ҙҲҸҮ ҡҲҽҲҧѦ ӁҽѸ ҏҦҽґҸҡҧ ҰҸҔҽӉҽҲҧ, ҍҲҸ Ұҝѹ ҲҽҔѹѽ ѹҡҡґѦҦҸӊҽӁѹҾѽ қҸґҧҗҏӀ ҰҸґҧӉҏ ѸҸңѦҲ ҰҝѹӁѦҡҲѹ ҝҽқҸҲҽ ҡ ҰҽҝҽґґѦґҧӁӃѸѹ ҲѦҔҡҲҽѸѹ. ҘҽҔ ӁҽѸ ҔҽңѦҲҡҾ, ѸҽҲѦҝѹҽґ ҰҽҝҽґґѦґҧӁҸҊҸ ҔҸҝҰҏҡҽ, ҦҽңѦ ӊ ӁӃӁѦҗӁѦҮ, ѦӂѦ ҰҝѦҦӊҽҝѹҲѦґҧӁҸҮ ҡҲҽҦѹѹ ҝҽӉҝҽқҸҲҔѹ, ҏңѦ ҰҸӉӊҸґѹґ ӁҽѸ ҰҸґҏҍѹҲҧ ӁѦҔҸҲҸҝӃѦ ѹӁҲѦҝѦҡӁӃѦ ҦҽӁӁӃѦ. ѯҲҸқӃ ґҏҍҗѦ ҰҸӁҾҲҧ ҡҏҲҧ ҝҽҡҡѸҸҲҝѦӁӁӃѽ ӊӃҗѦ ҾӊґѦӁѹҮ, ҡґѦҦҸӊҽґҸ қӃ ҝҽҡҗѹҝѹҲҧ ѸҽҲѦҝѹҽґ ѹ ҰҝҸҽӁҽґѹӉѹҝҸӊҽҲҧ ҲҽҔңѦ ҰҝѦҦґҸңѦӁѹҾ ҡ ҡҸӀӉҽѸѹ ҡґѦҦҸӊҽӁѹҾ ѹ ҰҝѦҦґҸңѦӁѹҾ, ҸқҸӉӁҽҍҽӀӂѹѦ ҦѦҮҡҲӊѹҾ, ӁѦ ҸҲӁҸҡҾӂѹѦҡҾ Ҕ ҰҝҸҗґҸѸҏ. ѢҽґѹҍѹѦ ҲҽҔҸҊҸ қҸґѦѦ ҰҸґӁҸҊҸ ѹ ҦѦҲҽґҧӁҸҊҸ ѸҽҲѦҝѹҽґҽ ѸҸҊґҸ қӃ ҲҽҔңѦ ҰҸѸҸҍҧ ӁҽҮҲѹ ҏҦҸӊґѦҲӊҸҝѹҲѦґҧӁҸѦ ҲѦҸҝѦҲѹҍѦҡҔҸѦ ҸқӆҾҡӁѦӁѹѦ ӊӃҾӊґѦӁӁӃѽ ҝҽӉґѹҍѹҮ ѸѦңҦҏ ҡґҽӊҾӁҡҔѹѸѹ ҾӉӃҔҽѸѹ. ѩҽӁӁӃѦ ҝҽӉґѹҍѹҾ ӍҽҔҲѹҍѦҡҔѹ ӁѦҸқӆҾҡӁѹѸӃ, Ѧҡґѹ ѹҡѽҸҦѹҲҧ ѹӉ ҲҸҊҸ, ҍҲҸ ҡѦѸҽӁҲѹҔҽ ҊґҽҊҸґҧӁҸҊҸ ӊѹҦҽ ӊҸ ӊҡѦѽ ҡґҽӊҾӁҡҔѹѽ ҾӉӃҔҽѽ ҸҰҝѦҦѦґҾѦҲҡҾ ѹҡҔґӀҍѹҲѦґҧӁҸ ҰҝѹӉӁҽҔҸѸ ҃ѦґҸҡҲӁҸҡҲѹ. ѨѸѦӁӁҸ ҰҸҙҲҸѸҏ ҡҝҽӊӁѹҲѦґҧӁҸ ӁѦҦҽӊӁҸ қӃґѹ ҰҝѦҦґҸңѦӁӃ ҰҝѹӉӁҽҔѹ “ҡѦҔӊѦӁҲӁҽҾ ҡӊҾӉҧ” (ҷҽҝѦӁҲҡѦӁ 1995; 1998) ѹ “ҲѦѸҰҸҝҽґҧӁҽҾ ҸҰҝѦҦѦґѦӁӁҸҡҲҧ” (Dickey 2000), ҔҽҔ ҦҸҰҸґӁѦӁѹѦ Ҕ ҰҝѹӉӁҽҔҏ ҃ѦґҸҡҲӁҸҡҲѹ (ѹґѹ ҔҽҔ ѦҊҸ ӉҽѸѦӁҽ). ҤҸҲҾ ҸҰҝѦҦѦґѦӁѹҾ ҙҲѹѽ ӁҸӊӃѽ ҰҝѹӉӁҽҔҸӊ ѹѸѦӀҲ ҦҸҡҲҽҲҸҍӁҸ ѸӁҸҊҸ ҸқӂѦҊҸ, ѹѽ ӁѦґҧӉҾ ҰҸґӁҸҡҲҧӀ ҸҲҸңҦѦҡҲӊѹҲҧ. ѢҸ, ҔҽҔ ӁҽѸ ҔҽңѦҲҡҾ, ҸӁѹ ӊ ҰҝѹӁ҃ѹҰѦ ҦҸҰҸґӁҾӀҲ ҦҝҏҊ ҦҝҏҊҽ ѹ ѹѽ ѸҸңӁҸ қӃґҸ қӃ ҔҽҔѹѸ-ҲҸ ҸқҝҽӉҸѸ ҡҸѦҦѹӁѹҲҧ. ѢҽҰҝѹѸѦҝ, ҏҔҽӉҽӁѹѦѸ Ӂҽ ҲҸ, ҍҲҸ “ҡѦҔӊѦӁҲӁҽҾ ҡӊҾӉҧ” ҰҝѦҦҰҸґҽҊҽѦҲ Ҳҏ ѹґѹ ѹӁҏӀ ҡҲѦҰѦӁҧ “ҲѦѸҰҸҝҽґҧӁҸҮ ҸҰҝѦҦѦґѦӁӁҸҡҲѹ” ҡѹҲҏҽ҃ѹѹ, ҡ ҔҸҲҸҝҸҮ ҏҡҲҽӁҽӊґѹӊҽѦҲҡҾ ҙҲҽ ҡӊҾӉҧ. ҩҽӉґѹҍѹҾ ӊ ҰҸӊѦҦѦӁѹѹ ӊѹҦҸӊ ӊ ҝҽӉӁӃѽ ҡґҽӊҾӁҡҔѹѽ ҾӉӃҔҽѽ ѸҸңӁҸ қӃґҸ қӃ ҲҸҊҦҽ ҸқӆҾҡӁѹҲҧ ҸҲӁҸҡѹҲѦґҧӁҸҮ ҡѹґҸҮ ҦѦҮҡҲӊѹҾ ҰҝѹӉӁҽҔҽ “ҡѦҔӊѦӁҲӁҽҾ ҡӊҾӉҧ”, ҽ ҲҽҔңѦ ҡҲѦҰѦӁҧӀ “ҲѦѸҰҸҝҽґҧӁҸҮ ҸҰҝѦҦѦґѦӁӁҸҡҲѹ”, ѽҽҝҽҔҲѦҝӁҸҮ ҦґҾ ҲѦѽ ѹґѹ ѹӁӃѽ ҊґҽҊҸґҧӁӃѽ ӍҸҝѸ ҝҽӉґѹҍӁӃѽ ҾӉӃҔҸӊ. ԣԐԙԽӾӢԌԂԐԙӪӠӡ ӫӰӠӴӾӢԙӠԽӾԽ

34

үѩҩѨүѢ ҷүҩѣѢѷҋѣѢ һѨѷѣҩүѷѰҩү

ҷҽҝѦӁҲҡѦӁ, үҦҝѹҽӁ ү. 1992 “ҫқ ҸқҡҲҸҾҲѦґҧҡҲӊҽѽ ҸҊҝҽӁѹҍѦӁӁҸҮ ҔҝҽҲӁҸҡҲѹ ҦѦҮҡҲӊѹҾ ӊ ҝҏҡҡҔҸѸ ҾӉӃҔѦ. ѯҽҡҲҧ 1”. ғ: Studies in Russian Linguistics (= SSGL 17), 1–66. Amsterdam: Rodopi. 1994 “ҫқ ҸқҡҲҸҾҲѦґҧҡҲӊҽѽ ҸҊҝҽӁѹҍѦӁӁҸҮ ҔҝҽҲӁҸҡҲѹ ҦѦҮҡҲӊѹҾ ӊ ҝҏҡҡҔҸѸ ҾӉӃҔѦ. ѯҽҡҲҧ 2”. ғ: Dutch Contributions to the Eleventh International Congress of Slavists, Linguistics (= SSGL 22), 1–50. Amsterdam: Rodopi. 1995 “ѷҝѦѽҡҲҏҰѦӁҍҽҲҽҾ ѸҸҦѦґҧ ѹӁӊҽҝѹҽӁҲҽ ҡҸӊѦҝҗѦӁӁҸҊҸ ӊѹҦҽ ӊ ҝҏҡҡҔҸѸ ҾӉӃҔѦ”. ғ: ҋҲҽӁѹҡґҽӊ ҘҽҝҸґҾҔ (ҝѦҦ.), ՀӾԐԂӰԽӠӪԂ Ӡ ԙԽӢӫӪԽӫӢԂ ԙԊԂӴԋӰԙӪӹԄӹ ԄԊԂԄӹԊӸӰӹԄӹ ӴӠԌԂ 1, 1–26. Kraków: Wydawnictwo Naukowe WSP. 1997 “ҩҸґҧ ґѦҔҡѹҍѦҡҔҸҊҸ ӉӁҽҍѦӁѹҾ ҊґҽҊҸґҽ Ұҝѹ ӊӃқҸҝѦ ӊѹҦҽ ӊ ҔҸӁҲѦҔҡҲѦ ҸҊҝҽӁѹҍѦӁӁҸҮ ҔҝҽҲӁҸҡҲѹ”. ғ: ҋҲҽӁѹҡґҽӊ ҘҽҝҸґҾҔ (ҝѦҦ.), ՀӾԐԂӰԽӠӪԂ Ӡ ԙԽӢӫӪԽӫӢԂ ԙԊԂӴԋӰԙӪӹԄӹ ԄԊԂԄӹԊӸӰӹԄӹ ӴӠԌԂ 2, 1–30. Kraków: Wydawnictwo Naukowe WSP. 1998 “ҭҝѹӉӁҽҔ «ҡѦҔӊѦӁҲӁҽҾ ҡӊҾӉҧ» ѹ ӊѹҦҸӊҸѦ ҰҝҸҲѹӊҸҰҸҡҲҽӊґѦӁѹѦ ӊ ҝҏҡҡҔҸѸ ҾӉӃҔѦ”. ғ: Ҍ.Ҵ. ѯѦҝҲҔҸӊҽ (ҝѦҦ.), ԁӠԗӹԊӹԄӠԋ ӴӠԌԂ. ӷӢӹԼԊӾԐՂ, ԗӹӠԙӪӠ, ӢӾԶӾӰӠԋ, 43–58. ҴҸҡҔӊҽ: «ѤӉӃҔѹ ҝҏҡҡҔҸҮ ҔҏґҧҲҏҝӃ». 2002 “ҫ ҝҏҡҡҔѹѽ ҊґҽҊҸґҽѽ ԙԐӹԓӸ ѹ ԙӫԐӾԽӸ”. ғ: ғ.ғ. ѨӊҽӁѹ҃ҔѹҮ ѹ Ҧҝ. (ҝѦҦ.), ԜԙӰӹӴӰՂӾ ԗӢӹԼԊӾԐՂ ӢӫԙԙӪӹӡ ԂԙԗӾӪԽӹԊӹԄӠӠ, 7–20. ҋҽӁҔҲ-ҭѦҲѦҝқҏҝҊ: ѢҽҏҔҽ. 2007 “ҫқ ҸҊҝҽӁѹҍѹҲѦґҧӁӃѽ ӊҝѦѸѦӁӁӃѽ ҡҸӀӉҽѽ ӊ ҡѦҝқҡҔҸѸ, ҝҏҡҡҔҸѸ ѹ ҰҸґҧҡҔҸѸ ҾӉӃҔҽѽ”. ԩԼӹӢӰӠӪ ԠԂԽӠՁӾ ԙӢԗԙӪӾ ԒԂ ԙԊԂӴӠԙԽӠӪӫ 71–72, 237–253. ҷҸӁҦҽҝҔҸ, ү.ғ. 1971 ԛӠԌ Ӡ ӴӢӾԐԋ ӢӫԙԙӪӹԄӹ ԄԊԂԄӹԊԂ. ҴҸҡҔӊҽ: ҭҝҸҡӊѦӂѦӁѹѦ. 1991 “ҭҝѦҦѦґҧӁҸҡҲҧ ѹ ҊґҽҊҸґҧӁӃҮ ӊѹҦ (Ӂҽ ѸҽҲѦҝѹҽґѦ ҝҏҡҡҔҸҊҸ ҾӉӃҔҽ”. ԚԒӴӾԙԽӠԋ ԣԾ ՀՀՀԪ, ՀӾӢӠԋ ԊӠԽӾӢԂԽӫӢՂ Ӡ ԋԒՂӪԂ 50, 3, 195–206. ѨӊҽӁҍѦӊ, ҋӊ. 1973 “҈ҽ ӁҾҔҸѹ ҸҡӁҸӊӁѹ ҡѽҸҦҡҲӊҽ ѹ ҝҽӉґѹҔѹ ѸѦңҦҏ ҡґҽӊҾӁҡҔѹҲѦ ѦӉѹ҃ѹ ӊ ҸқґҽҡҲҲҽ Ӂҽ ҊґҽҊҸґӁѹҾ ӊѹҦ”. ՀԊԂӴԋӰԙӪԂ ԅӠԊӹԊӹԄӠԋ. ԁӹԐ XII. ӺԒӠӪӹԒӰԂӰӠӾ, 131–152. ҘӁҾӉѦӊ, Ҍ.ҭ. 1989 “ғӃҝҽңѦӁѹѦ ҰҸӊҲҸҝҾѦѸҸҡҲѹ ҦѦҮҡҲӊѹҾ ӊ ҝҏҡҡҔҸѸ ѹ ҦҝҏҊѹѽ ҡґҽӊҾӁҡҔѹѽ ҾӉӃҔҽѽ”. ғ: ғ.ҋ. ҤҝҽҔҸӊҡҔѹҮ (ҝѦҦ.), ԁӠԗӹԊӹԄӠԋ ӠԽӾӢԂԽӠӴӰՂԖ ӪӹӰԙԽӢӫӪՁӠӡ, 132–145. һѦӁѹӁҊҝҽҦ: ѢҽҏҔҽ. 1998 “ҭҽҝҽѸѦҲҝӃ ҦґҾ ҲѹҰҸґҸҊѹѹ ӊѹҦҽ ѹ ҝҏҡҡҔѹҮ ҾӉӃҔ”. ғ: Ҵ.Ҍ. ѯѦҝҲҔҸӊҽ (ҝѦҦ.), ԁӠԗӹԊӹԄӠԋ ӴӠԌԂ. ӷӢӹԼԊӾԐՂ, ԗӹӠԙӪӠ, ӢӾԶӾӰӠԋ, 193–206. ҴҸҡҔӊҽ: ҹҔҸґҽ «ѤӉӃҔѹ ҝҏҡҡҔҸҮ ҔҏґҧҲҏҝӃ». ҴҽҡґҸӊ, Ҍ.ҋ. 1959 “ҐґҽҊҸґҧӁӃҮ ӊѹҦ ӊ ҡҸӊҝѦѸѦӁӁҸѸ қҸґҊҽҝҡҔҸѸ ґѹҲѦҝҽҲҏҝӁҸѸ ҾӉӃҔѦ. ҈ӁҽҍѦӁѹѦ ѹ ҏҰҸҲҝѦқґѦӁѹѦ”. ғ: ҋ.ҷ. ҷѦҝӁҗҲѦҮӁ (ҝѦҦ.) ԛӹԗӢӹԙՂ ԄӢԂԐԐԂԽӠӪӠ ԼӹԊԄԂӢԙӪӹԄӹ ԊӠԽӾӢԂԽӫӢӰӹԄӹ ԋԒՂӪԂ, 157–312. ҴҸҡҔӊҽ: үѢ ҋҋҋҩ. 1984 ԜԓӾӢӪӠ ԗӹ ԂԙԗӾӪԽӹԊӹԄӠӠ. һѦӁѹӁҊҝҽҦ: ѨӉҦ. һѦӁѹӁҊҝҽҦҡҔҸҊҸ ҏӁ-Ҳҽ. ҭѦҲҝҏѽѹӁҽ, ѣ.ғ. 2000 ԣԙԗӾӪԽӫԂԊӸӰՂӾ ӪԂԽӾԄӹӢӠӠ ԄԊԂԄӹԊԂ Ӵ ӢӫԙԙӪӹԐ ԋԒՂӪӾ (Ӵ ԙӹԗӹԙԽԂӴԊӾӰӠӠ ԙ ԓӾԶԙӪӠԐ, ԙԊӹӴԂՁӪӠԐ, ԗӹԊӸԙӪӠԐ Ӡ ԼӹԊԄԂӢԙӪӠԐ ԋԒՂӪԂԐӠ). ҴҸҡҔӊҽ: ѨӉҦ. ҴҸҡҔҸӊҡҔҸҊҸ ҏӁ-Ҳҽ.

ҭҫҋһѣѩҫғүѷѣһѻѢҫҋѷѻ ѩѣҢҋѷғѨҢ ҭҩѨ ҭҫғѷҫҩѤѣҴҫҋѷѨ

35

ҩҽҡҡҏҦҸӊҽ, ҫ.ҭ. 1968 ӯԗӹԽӢӾԼԊӾӰӠӾ ӴӠԌӹӴ Ӵ ӢӫԙԙӪӹԐ ԋԒՂӪӾ. ҴҸҡҔӊҽ: ѨӉҦ. ҴҸҡҔҸӊҡҔҸҊҸ ҏӁ-Ҳҽ. ҩҐ 1980 1980 ԪӫԙԙӪԂԋ ԄӢԂԐԐԂԽӠӪԂ, ԁӹԐ II, ՀӠӰԽԂӪԙӠԙ. ҴҸҡҔӊҽ: ѢҽҏҔҽ. ҋҲҽӊӁѹ҃Ҕҽ, ѤҦӊѹҊҽ 2003 “ғӃқҸҝ ӊѹҦҽ ӊ ҔҸӁҲѦҔҡҲѦ ҸҊҝҽӁѹҍѦӁӁҸҮ ҰҸӊҲҸҝҾѦѸҸҡҲѹ ӊ ҰҸґҧҡҔҸѸ ҾӉӃҔѦ ӊ ҡҸҸҲӁҸҗѦӁѹѹ ҡ ҝҏҡҡҔѹѸ”. Opera Slavica XIII, 20–36. ҤҝҽҔҸӊҡҔѹҮ, ғ.ҋ. 2003 “ҘҽҲѦҊҸҝѹҾ ҲҽҔҡѹҡҽ (ҸқӂҽҾ ѽҽҝҽҔҲѦҝѹҡҲѹҔҽ)”. ԛӹԗӢӹԙՂ ԋԒՂӪӹԒӰԂӰӠԋ 2, 32–54. Belyavski-Frank, Masha 2003 The Balkan Conditional in South Slavic. A Semantic and Syntactic Study. München: Otto Sagner. Dickey, Stephen M. 2000 Parameters of Slavic Aspect. A Cognitive Approach. Stanford: CSLI Publications. Forsyth, J. 1970 A Grammar of Aspect. Usage and Meaning in the Russian Verb. Cambridge: University Press. Friedman, Victor A. 1993 “Macedonian”. In: Bernard Comrie and Greville Corbett (eds.), The Slavonic Languages, 249–305. London and New York: Routledge. Galton, Herbert 1976 The Main Functions of the Slavic Verbal Aspect. Skopje: Macedonian Academy of Sciences and Arts. Holvoet, Axel 1991 “Uǻycie aspektu czasownikowego przy oznaczaniu czynnoǧci wielokrotnych w jNjzyku polskim i rosyjskim (Na tle s’owiaǝskim)”. Acta Universitatis Lodziensis. Folia Linguistica 25, 121–128. Iviǀ, Milka 1983 “Naǁini na koje slovenski glagol ovremenjuje ponavljanu radnju”. In: Milka Iviǀ, Lingivistiǁki ogledi, 37–56. Beograd: Prosveta. Kǥížková, Helena “K problematice aktuálního a neaktuálního užití ǁasových a vidových forem v 1958 ǁeštinLJ a v ruštinLJ”. ƀeskoslovenská rusistika 4, 185–200. Mønnesland, Svein 1984 “The Slavonic frequentative habitual”. In: Casper de Groot and Hannu Tommola (eds.), Aspect Bound. A voyage into the realm of Germanic, Slavonic and Finno-Ugrian aspectology, 53–76. Dordrecht – Holland / Cinnamon – U.S.A.: Foris publications. Stunová, Anna 1986 “Aspect and Iteration in Russian and Czech. A Contrastive Study”. In: Dutch Studies in Russian Linguistics (= SSGL 8), 467–510. Amsterdam: Rodopi. 1993 A Contrastive Study of Russian and Czech Aspect: Invariance vs. Discourse. Dissertation. Amsterdam.

36

үѩҩѨүѢ ҷүҩѣѢѷҋѣѢ

Timberlake, Alan 1982 “Invariance and the Syntax of Russian Aspect”. In: Paul J. Hopper (ed.), Tense–Aspect: Between Semantics & Pragmatics, 305–331. Amsterdam/Philadelphia: John Benjamins.

SUMMARY The Expression of Repeated Past Consecutive Events in the Modern Slavic Languages Slavic verbal aspect plays an important role in the expression of taxis relations between events. For instance, consecutive events are usually both expressed by a perfective form, at least in the case of a non-repeated string. In cases of repetition, matters become more complicated as a result of the strong association between unlimited repetition and imperfectivity, especially with past events. The article concentrates on the different ways the conflict between the aspectual properties of the individual events and the macrolevel of the whole unbounded series of repeated events is resolved. On the basis of the study of parallel examples of sentences with a temporal clause of the type as soon as or when (with a consecutive interpretation) it is shown that the various Slavic languages differ considerably here in their choice of verbal forms, especially in the temporal clause. An interesting phenomenon is the very consistent use of modally marked forms in some of the southern languages. On the basis of the typical distribution of aspecto-temporal-modal forms, the 11 studied languages have to be divided in no less than 6 different groups.

Dutch Contributions to the Fourteenth International Congress of Slavists, Ohrid: Linguistics (SSGL 34). Amsterdam – New York: Rodopi, 2008, 37-43.

NOTES ON THE SLAVIC METATHESIS OF LIQUIDS RICK DERKSEN

1. Metathesis and lengthening In my contribution to the forthcoming Kortlandt Festschrift, which deals with the development of vowel quantity in Upper Sorbian, I made an excursus into the relative chronology of the metathesis of liquids. Here I would like to return to the subject, while paying special attention to South Slavic forms that apparently escaped the metathesis. The relative chronology that served as a point of reference in the article mentioned above was established by Kortlandt (2003: 232). If we limit ourselves to the developments relevant to our subject, the chain of events is: (1) (2) (3) (4) (5) (6)

Lengthening before tautosyllabic resonants in South Slavic. Word-initial metathesis. Lengthening before tautosyllabic resonants in Czecho-Slovak. Non-initial metathesis in South Slavic and Czecho-Slovak. Rise of the new timbre distinctions. Lengthening under the stress before tautosyllabic resonants in Polish and Sorbian. (7) Non-initial metathesis in Polish and Sorbian. In comparison with earlier versions of Kortlandt’s relative chronology, the one represented above is more explicit about the origin of length reflected in forms such as SCr. brázda, Cz. brázda, Pl. bruzda ‘furrow’ (note that the metathesis preceded Dybo’s law). Here lengthening is not a phenomenon that under certain circumstances accompanies metathesis, but an independent development. 1 As I pointed out in my article, the advantage of this scenario is that it provides a solution for the fact that forms that seem to have escaped the metathesis as a rule show vowel lengthening. Since forms of this type are scattered over the scholarly literature, it may be useful to arrange them together. 1

The chronology in which the metathesis is preceded by lengthening had already been suggested by a number of scholars (e.g. Arumaa 1964: 152), but never within a wider chronological perspective.

38

RICK DERKSEN

2. Liquid diphthongs that were not affected by metathesis 2.1 Word-initially The verb *òlkati ‘be hungry’, cf. OCz. lákati ‘crave for’, Sln. lákati ‘be hungry, be greedy’, is widely attested as àëúêàòè in Old Church Slavic. 2 The non-metathesized variant is actually more frequently attested than ëàêàòè, which is found in the Codex Zographensis (Zogr.), the Psalterium Sinaiticum (Sin.), the Codex Marianus (Mar.), and the Codex Assemanianus (Ass.). The root appears as àëúê- in Sava’s Book (Sav.) and the Euchologium Sinaiticum (Euch.). 3 The spelling àë'ê- occurs in Zogr. and the Codex Suprasliensis (Supr.), while àëê- is found in Supr., Mar. and Ass. In Church Slavic manuscripts not belonging to the canon we also find àëüê-. The verb àëúêàòè has found its way into Modern Russian as alkát’ ‘crave for, hunger for’. The prefixed verb âúçàëúêàòè (ñ) ‘get hungry, hunger for’, Ru. vzalkát’, shows a similar distribution as the simplex. The variant âúçëàêàòè (ñ) is limited to Zogr. (8x) and Sin. (2x). It is outnumbered by the non-metathesized variant (28x). In Old Church Slavic we find various derivatives of the root àëúê-, e.g. àëúêàíè¬ (Euch., Supr.) ‘hunger, fast’, àëú÷üáà (Supr.) ‘fast’, and àëú÷üíú (Supr.) ‘hungry’. The latter adjective, which occurs only once, is actually written àë'÷åíú. It survives as Ru. álǁnyj, Bulg. álǁen, Mcd. alǁen (E. dial.) ‘greedy’, where it occurs alongside Ru. lakómyj, Bulg. lákom ‘id.’, cf. OCS ëàêîìûè (Supr. 1x). An early attestation of the metathesized form is laǁna Asg. m. in the Freising Fragments. The word for ‘ship, boat’ is àëúäèè in Supr. (2x) and Zogr. (1x). To be more exact, we find âú àë'äè¶ (Zogr.), âú àëäèè, and âú àë'äû (Supr.). Zogr. more frequently has ëàäèè (4x), which is also the form in Mar. (6x), cf. also ëàäèöà (Zogr., Mar.) ‘boat’. I shall leave the Asg. ãàëè« (Zogr.) for what it is. Forms with the root àë(ú)ä- also occur in later Church Slavic texts, for instance in John the Exarch’s Hexaemeron (see 2.2.). In Russian Church Slavic, too, there are a number of attestations of àëäèÿ or àëäèè (see Sreznevskij 1893-1912: s.v.). In Supr., there is a Gpl. ìüíèè that is usually emendated to àëúíèè ‘does’. The form ëàíèè or ëàíèÿ, cf. OCz. laní, is not attested in Old Church Slavic, but occurs in later texts. In Russian Church Slavic, we also find attestations of àëúíèè or àëúíèÿ, for instance in Gregory of Nazianzus’s Thirteen Sermons (11th c.). Furthermore, there is a Bulgarian dialect form alné (also alnénce) ‘young chamois’ (Mirǁev 1958: 140, BER I: 9, Shevelov 1964: 394). The same etymon may be 2 For the Old Church Slavic material I rely on the Slovník jazyka staroslovLJnského (SJS) and the Staroslavjanskij slovar’ (SS). 3 According to the SJS, àëúêàòè occurs four times in Sin., which I assume to be a mistake for Sav.

NOTES ON THE SLAVIC METATHESIS OF LIQUIDS

39

present in the place-name Alnéto (NE Bulgaria). Less clear is the significance of Ru. dial. alyn’já (Vlad., Kostrom.) ‘cow’ (Arumaa 1964: 151, Shevelov 1964: 395). It is not obvious that this noun is a transformation of *oln-, but on the other hand it is strongly reminiscent of Russian dialect forms such as skoryn’já (Astrax., ORu. skorynija) ‘jaw’, sporyn’já ‘abundance’, polyn’já (Arxangel’sk, etc., ORu. sporynija) ‘ice-hole’, pólymja (Vladimir, etc., also Belorussian) ‘flame’, golymjá (Arxangel’sk, Vjatka) ‘open sea’ (Shevelov 1964: 411). According to Shevelov, CoRyC instead of CoRoC is typical of North Russian and attested from the 12th-13th century onwards. The development is said to occur before palatalized m and n under falling pitch. I wonder if the latter condition is valid. In any case, alyn’já can only be an instance of the phenomenon mentioned by Shevelov if the form escaped the initial metathesis of liquids, which would make it unique within East Slavic. Furthermore, the initial a would have to be due to akan’e. Besides Bulg. latica ‘patch’, Mcd. latica ‘gusset, patch’, SCr. làtica (ƀak. lբtica) ‘petal, gusset, patch’, we find Mcd. dial. altica, oltica (Korubin 1953: 216, Malenko 1954: 106). Without the suffix *-ica we find ORu. lata ‘patch’, Ru. láty ‘armour’, Cz. lata ‘patch’, Pl. ’ata ‘id.’, etc. The ƇSSJa (1987: 47-49) advocates a connection with *lat- seize’, cf. Sln. látati se ‘undertake, make an effort’, which verb is supposed to be cognate with *letLJti ‘fly’. In my opinion, this etymology is unconvincing. The same holds for the connection with Gk. ʌ̮Ӑʂ ‘hem, fringe’. Thus, there is no objection to a Proto-Slavic form *oltica. 2.2 Word-internally Word-internal instances of non-metathesized liquid diphthongs are, generally speaking, limited to North West Slavic (Polabian, Pomeranian), where *CoRC developed in the same way as *CԔRC, and Church Slavic. South Slavic dialect material containing non-metathesized sequences is very scarce. Shevelov (1964: 406) mentions Bulg. dial. dálta (Sliven) ‘chisel’ and kargúj ‘falcon’ (*korgujӸ, a Proto-Slavic borrowing from Turkish). He also draws attention to the placenames Varna, Kavarna (Bulg.), and Smarda (Rom.). The Church Slavic evidence may be gathered from various manuals. Van Wijk (1931: 56) mentions the Gsg. çîëúòà (Sin.) ‘gold’, the noun çàëòàðèíú ‘goldsmith (ʖʑʔʒʏʖ˶ʏʤ)’, which is attested in a translation of a text by Athanasius of Alexandria, as well as a number of forms from John the Exarch’s Hexaemeron which will be discussed below. Shevelov (1964: 406) also mentions xaldodavica ‘bringing dew’ and svarbԔ ‘itch’ from a 13th-14th-century Oktoix. According to Koneski (1983: 4), xaldodavica is recorded in the Belgrade Macedonian Oktoix (13th c.). The Hexaemeron (or ŠestodӸnevԔ) by John the Exarch, one of the leading figures of the Preslav school, has been handed down through a number of manuscripts, of which a Serbian recension dating from 1263 is the oldest. This is

40

RICK DERKSEN

a copy of a Middle Bulgarian exemplar. I shall now present the relevant material according to Aitzetmüller’s edition and add the actually attested forms 4 between square brackets. The forms are áàëòèíû [áàëoò¸íû] Npl. ‘swamps’ (alongside áëàòèíû); ìàëäè÷è¬ [ìàëoä¸]¸¬] Nsg. ‘twigs’; ïàëòèöà [ïàëòüöà] Nsg. ‘crescent’, ïàëòèöå« [ïàëò¸öåþ“] Isg.; ïàëòüöà 2x [ïàëoòúöà 2x] Nsg. ‘half’, ïàëòüöè [ïàëòúö¸] Adu.; ïàëòüìü [ïàëúòüìü] Isg. ‘half’, ïàëúòè [ïàëoò¸] Ndu.; ïàëòú [ïàëoòú] Asg. ‘half’, ïàëúòú [ïàëüòú] Asg., ïàëúòà [ïàëoòà] Gsg.; ïàëúòüíýíà [ïàëüòåíoíà] Nsg. f. ‘weaving-’; ñàëíîñòü [ñàëíîñòü] Asg. ‘saltiness’. The NDu. ïàëòà ‘halves’ in Bodjans’kyj’s edition was corrected to ïëàòà by Popov. Later manuscripts usually show metathesis in these forms or sequences that are characteristic of East Slavic, such as ïîëòèöà, ïîëúòèöà. As one might expect, the Hexaemeron also has instances of initial àë(ú): àëäèª 2x [à‡ë丬“ 2x] APl. ‘ships’; àëêàíè¬ìü [à‡ëoêàí¸¬“ìú 2x] Isg. ‘fast’; àë÷èìú [àëü]åìü 2x] 1pl. ‘we shall fast’; àë÷üáà [à‡ëo]üáà] Nsg. ‘fast’, àëü÷üá© [à‡ëú]üá¹], àë÷üá© [à‡ëü]oá¹], àëü÷üá© [à‡ëü]oá¹] Asg. The word for ‘ship’ is twice attested with metathesis: ëàäèª [ëà丬“] Npl., ëàäèÿìè [ëàä¸ÿ“ì¸] Ipl. The only other forms showing metathesis of *ol- are case-forms of the word for ‘elbow’, e.g. ëàêîòú [ëàêúòú] Gpl. It appears that in the Hexaemeron word-initially as well as word-internally the non-metathesized liquid diphthong al is well attested. The spellings àëú, àëü, and àëo seem to indicate that in most cases the sequence was followed by an anaptyctic vowel. Aitzetmüller’s normalized edition of the text inconsistently renders àëo as àëú or àë. As a result, the anaptyctic vowel seems less frequently represented in the orthography than it actually is. 3. Geographical distribution The evidence from the Hexaemeron by John the Exarch seems to place the phenomenon under investigation in East Bulgaria. As far as ColC groups are concerned, Shevelov (1964: 406, cf. also Koneski 1983: 9) holds the same view. With respect to the non-metathesized liquid diphthongs in anlaut, Shevelov states that “these forms are found in OCS as well as Mo M and, to a lesser extent, Bg dialects” (o.c.: 394). The scarce dialect material does not exactly bear this out. It is true that altica (oltica) occurs in SW Macedonian dialects, but alné is attested in the dialect of Tetevensko (NE of Sofia) and the place-name Alnéto is located in NE Bulgaria. Forms continuing the adjective *alǁӸnԔ must be of Church Slavic origin.

4

That is to say the forms as they appear in Bodjans’kyj’s edition of the oldest manuscript (1879), while taking into account the corrections by A. Popov. These corrections can be found in Popov’s preface to Bodjans’kyj 1879, which was published posthumously.

NOTES ON THE SLAVIC METATHESIS OF LIQUIDS

41

An analysis of the Old Church Slavic material presented in 2.1 does not yield a clear-cut distribution. We may observe, however, that the metathesized variants occur almost exclusively in West Bulgarian (Macedonian) manuscripts. The East Bulgarian Codex Suprasliensis always has àë(ú)- except in ëàêîìûè (1x) and ëàêúòú ‘elbow’, where the non-metathesized variant is nowhere attested. 4. Relative chronology It is generally agreed upon that in part of the Slavic linguistic territory the outcome of the metathesis in word-initial position is connected with a prosodic contrast, which is usually designated as acute vs. circumflex or rising vs. falling. This terminology is confusing, as the cardinal issue seems to be whether or not we are dealing with an acute initial syllable that is stressed throughout the paradigm. According to Kortlandt (l.c.), acute syllables were at this stage still characterized by the presence of a glottal stop continuing either a laryngeal or originating from Winter’s law. In this view, there is no need to assume lengthening to explain, for instance, the sequence ra- of Ru. rálo, Cz. rádlo ‘plough’ < *ràսdla < *àrսdla. 5 As a matter of fact, the term “lengthening” may be considered misleading here, cf. SCr. rբlo, with the regular short reflex of an old acute. It is theoretically possible that in South Slavic (and Central Slovak) the lengthening before tautosyllabic resonants was limited to non-glottalized syllables. The forms that apparently escaped the metathesis seem to contradict this, however, the most widely attested root being alԔk- ‘be hungry’ < BSl. *olսk-, cf. Lith. álkti, Latv. a‫ך‬kt ‘id.’. Thus, it is likely that in the South Slavic area metathesis generated sequences *lƽսk- and *lƽk-. There is no evidence, however, that *lƽսk- was phonologically distinct from *laսk-. Apparently, glottalized syllables were neutral with respect to quantity. As can be gathered from Kortlandt’s chronology, the lengthening before tautosyllabic resonants spread from South Slavic to Czecho-Slovak and subsequently to Polish and Sorbian. The non-initial metathesis seems to have followed a similar trajectory. It is likely that the initial metathesis, which is early and occurred in virtually the entire Slavic area, also spread from the South. The area where metathesis of *(C)olC occurred in historical times or even never at all must be located on the periphery of the South Slavic linguistic territory. East Bulgaria is the most likely candidate for this area. The hypothesis that forms without metathesis are reborrowings from languages that had borrowed the relevant Slavic etyma at an early stage (Vaillant 1950: 162), e.g. Rom. baltƼ, altiǭƼ, seems insufficient to me (cf. Shevelov 1964: 407). The metathesis may have 5

The vocalism *a (rather than *o) of the reconstructed forms is in agreement with early Slavic borrowings in neighbouring languages, e.g. Ȳʑʅʂʄʂʒʓ˶ʤ (Radogost).

42

RICK DERKSEN

been slowed down, however, due to language contact with speakers of Romance and Greek. One may wonder if there was a South Slavic area where the initial metathesis (at least in the case of *or-) preceded the lengthening before resonants. 6 OCS ðîáú (more frequently ðàáú), attested almost exclusively in Supr., is usually assumed to have originated from West Slavic through slave trade, but Supr. also has a number of occurrences of the prefix ðîç-. Vaillant regards ðîç- as a relic from the language of the Moravian mission, but the one manuscript that is linked directly to the Moravian-Pannonian period, the Kiev Folios, has ðàç- (1x). Another form in Supr. without lengthening may be ðåòü ‘strife, contest’ (also 1x Zogr.²), if from *hōer-ti-. The same manuscript has numerous occurrences of ðàòú, however. I shall refrain from speculating on the origin of the isolated Gsg. çîëúòà (Sin.). Leiden University REFERENCES Aitzetmüller, R. 1958-1975 Das Hexaemeron des Exarchen Johannes. I-VII. Graz: Institut für Slavistik der Universität Graz. Arumaa, P. 1964 Urslavische Grammatik: Einführung in das vergleichende Studium der Slavischen Sprachen. I: Einleitung, Lautlehre. Heidelberg: Winter. BER 1962 BƼlgarski etimologiǁen reǁnik. I: A-zjapvam. SƼstavili Vl. Georgiev, Iv. GƼlƼbov, J. Zajmov and St. Ilǁev. Sofija: BƼlgarska Akademija na naukite. Bodjans’kyj, O.M. [Bodjanskij] 1879 Šestodnev sostavlennyj Ioannom Eksarxom Bolgarskim. Po xaratejnomu spisku Moskovskoj Sinodal’noj biblioteki 1263 goda. Slovo v slovo i bukva v bukvu (ƀtenija v Imperatorskom obšǁestve istorii i drevnostej rossijskix pri Moskovskom universitetLJ, kn. 4). Moskva: Universitetskaja tipografija. Derksen, R. forthc. “Quantity patterns in the Upper Sorbian noun”. To appear in: A. Lubotsky, J. Schaeken and J. Wiedenhof (eds.), Evidence and Counter-Evidence. Linguistic essays in honour of Frederik Kortlandt. Volume I: Balto-Slavic and IndoEuropean linguistics (Studies in Slavic and General Linguistics 32). Amsterdam-New York: Rodopi. ƇSSJa 1987 Ƈtimologiǁeskij slovar’ slavjanskix jazykov. Praslavjanskij leksiǁeskij fond. Vypusk 14 (*labati-*lLJteplԔjӸ). Pod redakciej O.N. Trubaǁeva. Moskva: Nauka.

6

The Bulgarian river-name Lom, cf. Lat. almus, Gk. ȳʌӐʏʤ, may also be viewed in this context.

NOTES ON THE SLAVIC METATHESIS OF LIQUIDS Koneski, B. 1983 Kortlandt, F. 2003

Korubin, V. 1953 Malenko, D. 1954 Mirǁev, K. 1958 Shevelov, G.Y. 1964

43

A historical phonology of the Macedonian language. Translated by Victor A. Friedman. Heidelberg: Winter. “Early dialectal diversity in South Slavic II”. In: J. Schaeken, P. Houtzagers and J. Kalsbeek (eds.), Dutch Contributions to the Thirteenth International Congress of Slavists, Ljubljana: Linguistics, August 15-21, 2003 (Studies in Slavic and General Linguistics 30), 215-235. Amsterdam-New York: Rodopi. “Altica (Prilog kon reǁnikot na mak. jazik)”. Makedonski jazik 4/9, 216. “Nekoliki primeri za “altica””. Makedonski jazik 5/1, 106. Istoriǁeska gramatika na bƼlgarskija ezik. Sofija: Nauka i izkustvo. A prehistory of Slavic: the historical phonology of Common Slavic. Heidelberg: Winter.

SJS 1958-1997 Sreznevskij, I.I. 1893-1912

Slovník jazyka staroslovLJnského (Lexicon linguae palaeoslovenicae). I-IV. J. Kurz (ed.). Praha: ƀeskoslovenská Akademie vLJd / Euroslavica. Materialy dlja slovarja drevnerusskago jazyka po pismennym pamjatnikam. IIII. Sankt-Peterburg: Tipografija Imperatorskoj Akademii nauk.

SS 1994 Vaillant, A. 1950 van Wijk, N. 1931

Staroslavjanskij slovar’ (po rukopisjam X-XI vekov). Pod redakciej R.M. Cejtlina, R. Veǁerki i Ƈ. Blagovoj. Moskva: Russkij jazyk. Grammaire comparée des langues slaves. I: Phonétique. Lyon-Paris: IAC. Geschichte der altkirchenslavischen Sprache. I: Laut- und Formenlehre. BerlinLeipzig: de Gruyter.

SUMMARY An attempt is made to establish the origin of South Slavic forms that appear to have escaped the metathesis of liquids and view them within the context of a relative chronology. The article includes a discussion of the relevant material in John the Exarch’s Hexaemeron.

Dutch Contributions to the Fourteenth International Congress of Slavists, Ohrid: Linguistics (SSGL 34). Amsterdam – New York: Rodopi, 2008, 45-108.

SEMANTICS AND SYNTAX OF THE CONSTRUCTION OF DEGREE IN RUSSIAN EGBERT FORTUIN 1. Introduction In this paper, I will provide a semantic-syntactic analysis of the construction of degree in Russian. In the introduction, I will provide a general overview of this construction and the main features of the construction that I will analyze in this paper. I will use the term ‘construction of degree’ for sentences consisting of the degree word ԌӹԙԽԂԽӹԓӰӹ (dostatoǁno ‘enough’) or the degree word ԙԊӠԶӪӹԐ (sliškom ‘too’), and usually a subordinate clause introduced by ԓԽӹԼՂ (ǁtoby, that-IRR ‘to’). Two examples are given below: 1 (1)

ү ҲѦ, ҔҲҸ ӁҏңҦҽѦҲҡҾ, ӁѦ ѸҸҊҏҲ ҦҽңѦ ҝҲҽ ҦҸҡҲҽҲҸҍӁҸ җѹҝҸҔҸ ҝҽҡҔҝӃҲҧ, ҍҲҸқӃ ѦҊҸ ҰҝҸҊґҸҲѹҲҧ. (ԠӹԙӪӹӴԙӪӠӾ ӰӹӴӹԙԽӠ, 2002.07.18) and those, who need, not can even mouth enough wide-ADV open-INF, that-IRR it swallow-INF ‘And those who need it, cannot even open their mouths wide enough to swallow it.’

(2)

ѢҸ ҰҝoқґѦѸҽ-ҲҸ ҡґѹҗҔҸѸ ҡѦҝҧѦӉӁҽҾ, ҍҲҸқӃ ҸҡҲҽӊґҾҲҧ ѦѦ қѦӉ ӊӁѹѸҽӁѹҾ. (ԠӾԼӾԊӸӰՂӡ ԼӠԒӰӾԙ, 2002.05.15) but problem-FEM-NOM PART too serious-ADJ-FEM-NOM, that-IRR leave-INF her without attention ‘But this problem is too serious not to pay attention to it.’

1

The examples are either from the ԾԂՁӠӹӰԂԊӸӰՂӡ ӪӹӢԗӫԙ ӢӫԙԙӪӹԄӹ ԋԒՂӪԂ (NKRJa) or the internet. In the case of examples taken from the NKRJa which are from the 19th century, I have indicated the date. – Throughout the paper, I will provide glosses (and translations) for most of the Russian sentences for those readers who do not read Russian. I do, however, not strive for a complete representation of all grammatical and morphological features, and only provide as much information as is relevant for the discussion. – With regard to the grammatical status of the degree words, it must be noted that the degree word ԙԊӠԶӪӹԐ is not morphologically marked as an adverb. Etymologically, it can be seen as the contraction of [ԙ + ԊӠԶӪӹԐ] ‘with surplus’. As such it differs from ԌӹԙԽԂԽӹԓӰӹ or its (near) synonym ԌӹӴӹԊӸӰӹ. Both these forms are adverbs derived form adjectives (ԌӹԙԽԂԽӹԓӰՂӡ, ԌӹӴӹԊӸӰՂӡ).

46

EGBERT FORTUIN

In the literature (e.g. Meier 2003), the meanings of these forms are usually stated in terms of a scale. In the case of expressions like ԌӹԙԽԂԽӹԓӰӹ, the degree on the scale is equal to or larger than some contextually given minimum, whereas in the case of expressions like ԙԊӠԶӪӹԐ, the degree is larger than some contextually given maximum. It is also remarked (e.g. Heim 2001; Meier 2003), that these degree words are associated with a specific kind of modality. More specifically, expressions like ԌӹԙԽԂԽӹԓӰӹ are used to indicate that the degree is such that the situation referred to in the subordinate clause can be realized, whereas sentences with expressions like ԙԊӠԶӪӹԐ express that the degree is such that the situation in the subordinate clause cannot be realized. 2 Note, however, that the Russian data are not fully in accordance with the semantic characterizations given here. Firstly, there are also sentences where the modality associated with the subordinate clause is closer to necessity than possibility, for example: (3)

ғҡѦ ҙҲҸ ҡґѹҗҔҸѸ ҸҍѦӊѹҦӁҸ, ҍҲҸқӃ ҰҏҡҔҽҲҧҡҾ ӊ қҸґѦѦ ҰҸҦҝҸқӁӃѦ ҝҽӉӆҾҡӁѦӁѹҾ. (ԩӴӾԒԌԂ, 2001) all that too obvious, that-IRR engage-INF in more detailed explanations ‘This is too obvious, for us to (have to) start giving even more detailed explanations.’

Furthermore, in some contexts ԌӹԙԽԂԽӹԓӰӹ can be interpreted in such a way that the realization of the situation in the subordinate clause is in fact presented as something that cannot or should not be realized (see Švedova 1980: 601): (4)

ҫӁ ҦҸҡҲҽҲҸҍӁҸ ҲҽґҽӁҲґѹӊ, ҍҲҸқӃ ҝѦҗҽҲҧ ҲҽҔѹѦ ӉҽҦҽҍѹ. (Švedova 1980: 601) he enough talented, that-IRR solve-INF such problems Possible interpretation: ‘He is talented enough not to have to solve such problems.’

In the examples of the construction of degree above, the degree word is applied to an adverb or adjective, but the degree words can also be applied to verbs: (5)

҈Ӂҽґҽ, ҍҲҸ ҸӁ ҡґѹҗҔҸѸ ґӀқѹҲ ѸѦӁҾ, ҍҲҸқӃ ҦҸҰҏҡҲѹҲҧ Ӂҽҗ ҝҽӉҝӃӊ. (ғ. ҹҽѽѹҦңҽӁҾӁ, 1001 ӴӹԗӢӹԙӹӴ ԗӢӹ ӳԽӹ) knew-SG-FEM, that he too loves me, that-IRR allow-INF our break up ‘I knew that he loved me too much to allow us to break up.’

2 In most sentences, ԙԊӠԶӪӹԐ or ԌӹԙԽԂԽӹԓӰӹ occur before the element which they modify. In neutral speech, especially in written language, it often occurs immediately before this element, but this is not necessarily the case. If accented, the degree word may also follow the element it modifies. In this paper, I will not go into the word order of the construction.

THE CONSTRUCTION OF DEGREE IN RUSSIAN

47

With some gradable verbs, the degree word is not always applied directly to the verb as in (5), but can also occur with an adverb that modifies the verb. An example is the verb ԒӰԂԽӸ (znat’ ‘know’), which can occur both with and without the adverb of quality ԖӹӢӹԶӹ (xorošo ‘well’): (6)

ҫӁ ҡґѹҗҔҸѸ ѽҸҝҸҗҸ ӉӁҽґ ҡӊҸҮ ҊҸҝҸҦ, ҍҲҸқӃ ѦҊҸ ѸҸҊґѹ ѹӁҲѦҝѦҡҸӊҽҲҧ ҦѦҲҽґѹ. (ѩ. ҐҝҽӁѹӁ, ԠӾԙԋՁ ӴӴӾӢԖ ӰӹԄԂԐӠ) he too good-ADV knew his city, that-IRR him could interest-INF details ‘He knew his city too well to be interested in details.’

(7)

[Ѥ] ҡґѹҗҔҸѸ ӉӁҽґ ҐґѹӁҔҏ, ҍҲҸқӃ ҰӃҲҽҲҧҡҾ ӊҲѹҝҽҲҧ ѦѸҏ ҸҍҔѹ. (ү. ѷҽҲѹӂѦӊ, ԩӾԐԊӠ Ӡ ԊԇԌӠ: ԛ ԄӫԷӾ ԗӾӢӾԙӾԊӾӰԓӾԙӪӹԄӹ) I too knew-SG-MASC Glinka-ACC, that-IRR try-INF rub-INF he-DAT glasses ‘I knew Glinka too well to try to pull the wool over his eyes.’

Minimal pairs can also be found with or without the quantifier ԐӰӹԄӹ (mnogo ‘much’) with verbs such as ԙԽӢԂԌԂԽӸ (stradat’ ‘to suffer’). Consider for example the following sentences, both from the nineteenth century: (8)

[Ѥ] ҡґѹҗҔҸѸ ѸӁҸҊҸ ҡҲҝҽҦҽґҽ ѹ ҡґѹҗҔҸѸ ҦҸґҊҸ ңҦҽґҽ ҙҲҸҊҸ, ҍҲҸқӃ ҲѦҰѦҝҧ ҸңѹҦҽҲҧ ѦӂѦ ҦҸґҧҗѦ. (Ѣ. һѦҡҔҸӊ, ԱӹӢԂ (1888)) I too much suffered-SG-FEM and too long waited-SG-FEM that-DEMGEN, that-IRR now wait-INF still longer ‘I have suffered too much, and I have waited too long to wait any longer now.’

(9)

ѷӃ ҦҏҝҽҔ ҰҸҡґѦ ҙҲҸҊҸ ҡҸӊѦҝҗѦӁӁѦҮҗѹҮ, Ѧҡґѹ ӁѦ ҰҸӁѹѸҽѦҗҧ ҲҸҊҸ, ҍҲҸ Ҿ ҡґѹҗҔҸѸ ҡѦҊҸҦӁҾ ӁѦҡҍҽҡҲґѹӊ ѹ ҡґѹҗҔҸѸ ҡҲҝҽҦҽӀ, ҍҲҸқ ѹӉҦѦӊҽҲҧҡҾ ӁҽҦҸ ѸӁҸҮ ѹ ҦѦґҽҲҧ ӁҽҡҲҽӊґѦӁѹҾ ѸӁѦ. (ү.Ҩ ҭѹҡѦѸҡҔѹҮ, ԛԂԂԊ (1873)) you fool after that-GEN uttermost, if not understand-2SG that-DEMGEN, that I too today unhappy-ADJ-SHORT-SG-MASC and too suffer1SG, that-IRR mock-INF before me-INSTR and do-INF admonition meDAT ‘You are the biggest fool, if you do not understand that today I am too unhappy, and suffer too much for you to mock me and admonish me.’

The degree words can also be applied to nouns. In the case of ԙԊӠԶӪӹԐ this use occurs with the quantifier ԐӰӹԄӹ, and in the case of ԌӹԙԽԂԽӹԓӰӹ this quantifier is not used in most sentences: (10) Ѱ ҰҝҸҔҏҝҽҲҏҝӃ қӃґҸ ҦҸҡҲҽҲҸҍӁҸ ӊҝѦѸѦӁѹ ѹ җҽӁҡҸӊ, ҍҲҸқӃ ҦҸҔҽӉҽҲҧ ӊѹӁҏ Ҋ-Ӂҽ ғҽӊѹґҸӊҽ ҰѦҝѦҦ ӉҽҔҸӁҸѸ. (ԚԒӴӾԙԽӠԋ, 2002.06.16) at public.prosecutor was-SG-NEUT enough time-GEN-SG and chanceGEN-PL, that-IRR prove-INF guilt mister-GEN Vavilov-GEN before law

48

EGBERT FORTUIN

‘The public prosecutor had enough time and opportunity to prove that mister Vavilova was guilty before the law.’ (11)

[ҋ]ґѹҗҔҸѸ ѸӁҸҊҸ ӁҽѸѦҔҸӊ, ҍҲҸқӃ ҸҰӃҲӁӃҮ ӉҝѹҲѦґҧ ҰҸґҏҍѹґ ҸӂҏҲѹѸҏӀ ҸҰґѦҏѽҏ ӊ ӍѹӁҽґѦ. (Ծ. ԾӹӴԄӹӢӹԌ, 2002.03.11) too much hints-GEN, that-IRR experienced audience received perceptible slap.in.the.face in final ‘Too much hints for the experienced audience to get a slap in the face at the end.’ [i.e. the audience of the opera knows what is coming]

In many sentences, in fact in all the examples given above, the main clause with the degree word occurs with a subordinate clause introduced by ԓԽӹԼՂ (ǁtoby, that-IRR ‘to’). In most cases, the predicate of the subordinate clause is an infinitive, as is the case in the examples above such as (1). The subject of the infinitive is either coreferent with the subject of the main clause (as in (5)), or generic (as in (2), ‘people in general’). There are also rare instances where the subject of the infinitive is expressed by a dative noun or pronoun. In such cases the subject of the infinitive and the subject of the main clause are usually non-coreferent: (12) ѷӃ ӊѦҦҧ ҰҝѹӊѦӉ ҡґѹҗҔҸѸ ѸӁҸҊҸ ӁҸӊҸҡҲѦҮ, ҍҲҸқӃ ӁҽѸ ӊӃҡґҏҗѹӊҽҲҧ ѹѽ ӊ ҡҰѦҗҔѦ. (ғ. ҋҔӊҸҝ҃Ҹӊ, ՀӠӰԄԂԗӫӢԙӪӠӡ ӪӴԂӢԽӾԽ) you after.all brought too much news-GEN-PL, that-IRR we-DAT listenINF them in hurry ‘After all, you brought too much news, for us to listen to it quickly.’ In sentences where the object of the infinitive is coreferential with the subject of the main clause, the object is usually expressed (as in (2) above), but there are also sentences where this is not the case: (13) ѢҽѸѦҔѹ қӃґѹ ҡґѹҗҔҸѸ ҰҝҸӉҝҽҍӁӃѸѹ, ҍҲҸқӃ ӁѦ ҰҸӁҾҲҧ. (ғ. ҋґѹҰѦӁҍҏҔ, ԩӠӰԒӠӴӾӢ) hints were too obvious, that-IRR not understand-INF ‘The hints were too obvious, not to understand them.’ In some sentences a finite verb is used in the subordinate clause, especially in contexts where the subject of the main clause and the subject of the subordinate clause are non-coreferent. In many contexts, sentences with a finite verb have a modal expression in the subordinate clause, typically verbs that express possibility: (14) ғҡѦ-ҲҽҔѹ ѸҸѦ ҰҝѦқӃӊҽӁѹѦ Ӂҽ ӊѦҝҗѹӁѦ қӃґҸ ҡґѹҗҔҸѸ ҔҸҝҸҲҔѹѸ, ҍҲҸқӃ Ҿ ѸҸҊ ҡҰҸҔҸҮӁҸ ҏҮҲѹ. (ԚԒӴӾԙԽӠԋ, 2003.01.10) nevertheless my stay at top was too short, that-IRR I could quietly leaveINF

THE CONSTRUCTION OF DEGREE IN RUSSIAN

49

‘Nevertheless, my stay at the top was too short to be able to leave without any worries .’ However, finite sentences without modal expression occur as well, for example (11) given earlier. Although I will use the term construction of degree for sentences with ԙԊӠԶӪӹԐ and ԌӹԙԽԂԽӹԓӰӹ, it can also be used for sentences with ԌӹӴӹԊӸӰӹ and ԓӾӢӾԙԓӫӢ. ѷhe adverb ԌӹӴӹԊӸӰӹ (dovol’no), can be seen as a (near) synonym of ԌӹԙԽԂԽӹԓӰӹ (‘enough’), and the degree word ԓӾӢӾԙԓӫӢ (ǁeresǁur) is a (near) synonym of ԙԊӠԶӪӹԐ (‘too’): (15) TӃ ѦӂѦ ҍѦҝѦҡҍҏҝ ѸҸґҸҦҽҾ, ҍҲҸқӃ ҲҽҔ ӊѦҡҲѹ ҡѦқҾ. (һ. үҝҔҽҦѹҮ, ԟӴӹӢ) you still too young-ADJ, that-IRR so behave-INF yourself ‘You are still too young, to behave yourself like that.’ (16) ҴӃ ҲѦҰѦҝҧ ҦҸӊҸґҧӁҸ ӉӁҽѦѸ һҸҰҏѽҸӊҽ, ҍҲҸқӃ ӊѹҦѦҲҧ, ҍҲҸ ҸӁ қӃґ ҍѦґҸӊѦҔ ӁѦ ҡҽӁҲѹѸѦӁҲҽґҧӁӃҮ. (ѯѦҝӁӃҗѦӊҡҔѹҮ, өԽӹ ԌӾԊԂԽӸ?) we now enough know Lopuxov, that-IRR see-INF, that he was man not sentimental ‘We now know Lopuxov well enough, to see that he was not a sentimental man.’ The semantic element ‘too’, which is usually expressed by a degree word, may also be expressed by a (predicatively used) short form of an adjective, which in Russian may express that the degree is too high (see Groen 1998): (17) ѷӃ ѦӂѦ Ѹҽґ ѹ ҊґҏҰ, ҍҲҸқӃ ѸѦӁҾ ҸқѸҽӁӃӊҽҲҧ. you still small-ADJ.SHORT and stupid-ADJ.SHORT, that-IRR me foolINF ‘You are still too small and too stupid to fool me.’ The construction of degree shares particular semantic-syntactic features with the construction consisting of ӰԂԙԽӹԊӸӪӹ (nastol’ko), ԙԽӹԊӸ(ko) (stol’(ko)), or ԽԂӪ (tak) (all mean ‘so (much)’) and a subordinate clause introduced by ԓԽӹԼՂ (ǁtoby, that-IRR; ‘to’) or ԓԽӹ (ǁto, that; ‘that’): (18) һѹҦѦҝӃ ҘҭҩҨ ӁѦ ӁҽҡҲҸґҧҔҸ ҊґҏҰӃ, ҍҲҸқӃ ҙҲҸҊҸ ӁѦ ҰҸӁѹѸҽҲҧ. (ԛӾԙԽӸ ԹԂԊӫԄԂ, 2002.10.17) leaders KPRF not so stupid-ADJ.SHORT-PL, that-IRR that-DEM-GEN not understand-INF ‘The leaders of the Communist Party of the Russian Federation are not so stupid, that they wouldn’t understand that.’

50

EGBERT FORTUIN

(19) ҭѦҝѦѽҸҦҾӂѹѦ ӉҽҰҽҡӃ ҸҔҽӉҽґѹҡҧ ӁҽҡҲҸґҧҔҸ ӊѦґѹҔѹ, ҍҲҸ ҦҽңѦ ҡҝѦҦӁѹҮ ҏҝҸңҽҮ ҙҲҸҊҸ ҊҸҦҽ ҸқӊҽґѹҲ Ӂҽ ҝӃӁҔѦ ҃ѦӁӃ. (ԚԒӴӾԙԽӠԋ, 2002.04.24). carry-over stock turned.out so huge-ADJ.SHORT-PL, that even average harvest of.this year will.decrease on market prices ‘The carry-over-stock turned out to be so huge, that even the average harvest of this year will decrease the prices on the market.’ As I will show, there are also important semantic differences between the construction of degree and this construction. In the literature, the construction of degree has received relatively little attention. The existing analyses are mainly from people working within formal semantic frameworks (e.g. Bierwisch 1989, Meier 2003, Heim 2001). These analyses deal with the construction in English or German, and focus on the question of how one can represent the semantics of the construction in formal logic frameworks. To my knowledge, no systematic analysis of the construction of degree in Russian exists, with the exception of Apresjan et al. 2004, who provide a small semantic analysis of the degree word ԙԊӠԶӪӹԐ. In the existing analyses the modal character of the construction is either explained by postulating a possibility operator in the semantic representation of the degree word (Heim 2001), or in the semantic representation of the subordinate clause (Meier 2003). However, the specific relation between the meaning of the main clause, containing the degree word, and the meaning of the modal subordinate clause remains largely unexplained. Furthermore, because the existing analyses do not deal with Russian, specific Russian phenomena such as the occurrence of sentences like (4), where ԌӹԙԽԂԽӹԓӰӹ is used to express that the situation in the subordinate clause should not be realized, are not taken into account. Such sentences pose a problem for the existing formal analyses where enough is defined in terms of minimum. The construction is also very interesting from a syntactic-semantic point of view. The degree words can be combined with different parts of speech: adjectives, adverbs, verbs, and nouns. As I remarked, in the case of verbs the adverbs ԐӰӹԄӹ and ԖӹӢӹԶӹ are sometimes used. The specific combinatory rules remain, however, largely unexplained. To give an example, it is not fully clear what constitutes the class of verbs that can occur both with and without ԖӹӢӹԶӹ (see (6) and (7)), and whether there is a difference in meaning between sentences with or without ԖӹӢӹԶӹ. The same question can be asked for verbs that can occur both with and without ԐӰӹԄӹ (see (8) and (9)). In this paper, I will provide an extensive analysis of the semantics and syntax of the construction of degree, which will address the above questions. This analysis may shed more light on the construction of degree in Russian. Furthermore, it may contribute to the study of the semantic-syntactic properties of

THE CONSTRUCTION OF DEGREE IN RUSSIAN

51

degree expressions in general (see for example Kennedy and McNally 2005, who deal with this issue). The paper is divided into two main parts. In the first part (section 2), I will discuss the meanings of ԙԊӠԶӪӹԐ and ԌӹԙԽԂԽӹԓӰӹ respectively. In the second part (section 3), I will focus on the semantic-syntactic side of the construction. This part deals with the way in which the degree words are combined with the different parts of speech: adjectives, adverbs, verbs and nouns. Finally, in section 4, I will give my conclusion. 2. Meaning of the degree words 2.1 ՀԊӠԶӪӹԐ (‘too’) The meaning of ԙԊӠԶӪӹԐ (‘too’) can be defined as follows: ԙԊӠԶӪӹԐ X:

‘Such a degree of X, that it exceeds the appropriate limit (maximum) with respect to Y (such that is leads to negative consequences)’

The meaning given here can also be formulated differently by making explicit that exceeding the maximum means that a lower degree is needed: ԙԊӠԶӪӹԐ X:

‘A lower degree of X is necessary with respect to Y’

The semantic element of ‘appropriate limit’ or ‘maximum’ is always connected to a standard Y (see also Meier 2003, for discussion of the term maximum). This can be illustrated with regard to a sentence such as the following: (20) ҩѦқѦӁҸҔ ҡґѹҗҔҸѸ ҲҸґҡҲӃҮ. ‘The child is too heavy.’ In this sentence, ԙԊӠԶӪӹԐ (‘too’) presupposes that children can be more or less heavy, and that when one is ‘adding up’ this property, at some point one can say that the weight has such a degree that it exceeds a maximum, and is evaluated negatively. Here, Y can be thought of as an implicit standard of a healthy weight. In other cases the evaluation of having too much of a particular property, is relative to a contextually given standard, for example: (21) ҷҏѸҽҊҽ ҡґѹҗҔҸѸ ҲҸґҡҲҽҾ, ҍҲҸқӃ ѦѦ ҝҽӉҸҝӊҽҲҧ. ‘The paper is too thick to tear apart.’ Here, it is only expressed that the thickness of the paper exceeds the appropriate limit of thickness, with respect to the realization of the action given in the subordinate clause (Y). Note that the subordinate clause is introduced by the conjunction ԓԽӹԼՂ (that-IRR, ‘to’), which has an instrumental meaning in sentences like these:

52

EGBERT FORTUIN

ԓԽӹԼՂ Y:

‘instrumental in the realization of Y’

The instrumental meaning is sometimes stressed by using ԌԊԋ ԽӹԄӹ (for that, ‘in order’): (22) [Ѣ]Ҹ ҸӁ қӃґ ҡґѹҗҔҸѸ ҡҽѸҸҡҲҸҾҲѦґҧӁӃѸ ҦґҾ ҲҸҊҸ, ҍҲҸқӃ ҡҦѦґҽҲҧ ҔҽҝҧѦҝҏ ӊ ҡѹҡҲѦѸѦ ӊґҽҡҲѹ. (ԚԒӴӾԙԽӠԋ, 2002.04.28) but he was too independent for that-DEM-GEN, that-IRR make-INF career in system power-GEN ‘But he was too independent to make a career in the system of power.’ The general structure of the whole construction (including the degree word and the subordinate clause) can be defined as follows: X ԙԊӠԶӪӹԐ Y, ԓԽӹԼՂ Z:

‘degree Y of X exceeds the maximum with respect to being instrumental in the realization of Z’

A similar instrumental meaning occurs in sentences where the standard is expressed by the prepositional phrase [ԌԊԋ (‘for’) + noun], which expresses a noun associated with the realization of some situation: (23) ѢҽӊѦҝӁҸ, Ҿ ҡґѹҗҔҸѸ ҰҝҸҡҲ ҦґҾ ӊҡѦѽ ҙҲѹѽ ѹӁҲҝѹҊ. (ғ. ҭѦґѦӊѹӁ, ԜԽӪӢӹӴӾӰӠӾ ԹӢӾԄӾӢԂ) probably, I too simple for all these intrigues ‘Probably, I am too simple for all these intrigues.’ In most cases, the use of an instrumental subordinate clause implies that because of the (relatively) high degree the situation expressed in the subordinate clause cannot be realized. The interpretation of a sentence like (21) can be paraphrased as follows: We consider the situation where someone wants to tear apart the paper, and presuppose that one needs a particular degree of thickness to be able to do that. Put differently, at a particular degree of thickness, tearing apart the paper becomes impossible. We determine that the degree of thickness is such that tearing apart the paper is not possible. That is, the degree exceeds the maximum with respect to being instrumental (here making it possible) to tear apart the paper. In the literature, the meaning of possibility is even treated as part of the semantics of the degree word (Heim 2001) or the subordinate clause (Meier 2003) (see also section 1.). The first option does not account for all sentences, because there are many sentences without a subordinate clause which do not express or imply impossibility of any kind (for example (20)). Although the postulation of a possibility operator in the subordinate clause would account for the majority of

THE CONSTRUCTION OF DEGREE IN RUSSIAN

53

cases, there are, however, also sentences with a different modality type, such as (3) given earlier. This sentence could be paraphrased as follows: We consider the situation where there is no intention or desire to give even more detailed explanations, and presuppose that one needs a particular degree of clarity, for such a situation to be the case. Put differently, at a particular degree of clarity, giving detailed explanations becomes senseless, useless and unnecessary. We determine that the degree of clarity is such that giving detailed explanations is not necessary. That is, the degree exceeds the maximum with respect to being instrumental (here making it necessary) to give detailed explanations. Furthermore, in many sentences labels such as possibility or necessity are not fully adequate to describe the modality associated with the construction: (24) Ѥ ҡґѹҗҔҸѸ ҍҲҏ ґӀҦѦҮ, ҍҲҸқӃ ҸҡҔҸҝқґҾҲҧ ѹѽ ҰґҽҲӁҸҮ ґӀқҸӊҧӀ. (Ҵ. ѵӊѦҲҽѦӊҽ, Ԝ ԼԊԂԄӹԌԂӢӰӹԙԽӠ) I too honor people, that-IRR insult-INF them paid-INSTR love-INSTR ‘I honor people too much, for me to insult them by means of paid love.’ [in other words, I honor them to such an extent, that I would not do such a thing] (25) ѩѦґҸ ҡґѹҗҔҸѸ ѸѦґҔҸѦ, ҍҲҸқӃ ӊӃ ӊ ӁѦҊҸ ӊѸѦҗѹӊҽґѹҡҧ, ѸҸҮ ӊӃҡҸҔѹҮ ҔҸґґѦҊҽ. (Ҍ. ѩҸѸқҝҸӊҡҔѹҮ, ԜԼӾԒӸԋӰԂ ԗӢӠԖӹԌӠԽ ԒԂ ԙӴӹӠԐ ԓӾӢӾԗӹԐ, ԓԂԙԽӸ 3) case too trivial, that-IRR you in him interfere-PL-PAST, my esteemed colleague ‘That case is too trivial for you to interfere in it, esteemed colleague.’ [in other words, that case is so trivial that it makes no sense for you to interfere in it] This suggests that the modality must be seen as an interpretation, rather than as a discrete meaning, and that it cannot be described on the par with the modality expressed by modal forms. It is therefore better to treat the specific modality associated with the subordinate clause as a specific interpretation of the subordinate clause, which arises because of the interaction of the meaning of the degree word, the meaning of the subordinate clause, especially the meaning of ԓԽӹԼՂ, and the pragmatic context in which the sentence occurs. This means that it cannot be ascribed to one of the constituents of the construction alone (degree word, ԓԽӹԼՂ, predicate of the subordinate clause), but occurs because of the interaction of all these elements together. In my view, the reason why the most frequent interpretation is that of possibility is essentially pragmatic in nature. In contexts where we talk about the appropriate degree for something, the most natural interpretation is that the degree is appropriate for something we

54

EGBERT FORTUIN

would like to do. Similarly, in contexts where we talk about a degree which exceeds the appropriate degree, we usually talk about circumstances blocking something that we want to do. Therefore, in order to arrive at an interpretation of necessity specific clues are needed. The specific way the modality comes about is a topic which deserves a separate analysis, but which fall outside the scope of this paper. It should be noted that in most sentences ԙԊӠԶӪӹԐ does not necessarily express a high degree, but only a high degree with respect to some (implicitly) given standard. A possible exception is constituted by sentences with negation, like the following where ԙԊӠԶӪӹԐ is used to express that the degree of the property expressed by the adjective to which it is applied is not very high, but rather low: (26) ҘҸѸӁҽҲҽ қӃґҽ ҲѦҡӁҸҮ ѹ ӁѦ ҡґѹҗҔҸѸ ҏӀҲӁҸҮ. room was narrow and not too cosy ‘The room was narrow and not too cosy.’ (27) ѢѦ ҡґѹҗҔҸѸ ѽҸҝҸҗҸ ҰҸӁѹѸҽӀ. not too well understand-1SG-PRES ‘I do not understand that too well.’ This use of ԙԊӠԶӪӹԐ is rather close to that of an intensifier (ӰӾ ԙԊӠԶӪӹԐ X | ‘not very X’). In most uses, however, the use of ԙԊӠԶӪӹԐ only implies a relative degree. This means that some instance of X may be portrayed as having a high degree with respect to some contextually given standard, but a low degree from a more general perspective, for example because it has a low degree with respect to the class of X. This is the case in (28) where a gun is judged to be small, for example because it is smaller than the average gun, but big with regard to someone’s hands: (28) Ѥ ҰҸҡѸҸҲҝѦґ Ӂҽ ҰѹҡҲҸґѦҲ, ҡҸӊѦҝҗѦӁӁҸ қѦҡҰҸґѦӉӁӃҮ ҦґҾ ѸѦӁҾ, ѽҸҲҧ ѹ ҡ ҰҸґӁҸҮ ҸқҸҮѸҸҮ, ҲҽҔҸҮ ѸҽґѦӁҧҔѹҮ, ӁҸ ӊҡѦ ҝҽӊӁҸ, ҡґѹҗҔҸѸ қҸґҧҗҸҮ ҦґҾ ѸҸѦҮ ҝҏҔѹ. (www.kras.fatal.ru/greger.htm) I looked at gun, totally useless for me, although and with full charger, such small, but anyway too big for my hand ‘I looked at the gun, which was totally useless for me, although it was fully loaded, and a small one, but it was nevertheless too big for my hands.’ As such, it differs from degree words like ԓӢӾԒԐӾӢӰӹ (ǁrezmerno ‘overly’), which also express that the degree exceeds a maximum, but which always expresses an absolute high degree of something (e.g. a high degree with respect to the class of X):

THE CONSTRUCTION OF DEGREE IN RUSSIAN

55

(29) [ҫ]қҲҾӁҏҲӃҮ ҊґҽҦҔҸҮ ҔҸңѦҮ ҝҏґҧ ҍҝѦӉѸѦҝӁҸ ҡҔҸґҧӉҔѹҮ, ҽ ҝӃҍҽҊ ҰѦҝѦҔґӀҍѦӁѹҾ ҰѦҝѦҦҽҍ ҡґѹҗҔҸѸ ҸҲҸҦӊѹӁҏҲ ӊҰҝҽӊҸ ѹ ӁҽӉҽҦ. (ԩԂ ӢӫԊӾԐ, 2004) covered smooth-INSTR leather-INSTR steering.wheel overly slippery, but lever shift gears-GEN too moved.aside to.the.right and backwards ‘The steering wheel, which is covered with smooth leather, is overly slippery, and the gear shift lever is positioned too much to the right and to the back.’ This explains why the following sentence where ԓӢӾԒԐӾӢӰӹ occurs with a subordinate clause which provides a standard of comparison is not acceptable (see also Apresjan et al, 2004): (30) ?ҫӁ ҍҝѦӉѸѦҝӁҸ ҊґҏҰ, ҍҲҸқӃ ҦҸҦҏѸҽҲҧҡҾ ҦҸ ҙҲҸҊҸ ҡҽѸҸѸҏ. (Apresjan et al 2004: 1044) he overly stupid, that-IRR come.up.with-INF to that self corresponding sentence in English: ?‘He is overly stupid, to be able to come up with that idea by himself.’ Someone who is overly stupid is so, in the mind of the speaker, without any restriction, hence it is not possible to use a subordinate clause that provides a relative standard (‘with respect to’). ԓӢӾԒԐӾӢӰӹ X: ‘The degree of X is higher than the upper boundary of appropriateness (maximum), such that the degree is at the extreme end of the scale (which is evaluated negatively)’ Apresjan et al (2004: 1045) argue that verbs with the prefix ԗӾӢӾ- can also express the meaning of ԙԊӠԶӪӹԐ, for example in the case of the verb ԗӾӢӾԙԽԂӢԂԽӸԙԋ (‘try too hard’), ԗӾӢӾԙӹԊӠԽӸ (‘use too much salt’), or ԗӾӢӾԗӠԽӸ (‘drink too much’). The use of the prefix ԗӾӢӾ- (lit. ‘over’) suggests that some maximum has been exceeded. It should be noted, though, that like ԓӢӾԒԐӾӢӰӹ, these verbs do not indicate a relative degree. Because of this they cannot occur with a subordinate clause that provides a standard. In those sentences where these verbs are used with a subordinate clause with ԓԽӹԼՂ, the conjunction expresses the goal of the action mentioned in the main clause, for example: (31) ҶҲҸ ҡҽӁѹӁҡҲҝҏҔҲҸҝ ҰѦҝѦҡҲҽҝҽґҡҾ, ҍҲҸқӃ ӁѦ ӉҽҊҝҾӉӁѹґҽҡҧ ҝҽӁҽ. (militera.lib.ru/memo/russian/altunin_at/04.html) that sanitary-officier tried.too.hard, that-IRR not became.dirty wound ‘It’s because the sanitary officer tried too hard not to let the wound become dirty.’

56

EGBERT FORTUIN

As such it differs from sentences with ԙԊӠԶӪӹԐ, where the subordinate clause can provide a standard: (32) ҋґѹҗҔҸѸ ҦҸґҊҸ қӃґ ҡҸґҦҽҲҸѸ ѹ ҡґѹҗҔҸѸ ҡҲҽҝҽґҡҾ қӃҲҧ ҡҸґҦҽҲҸѸ, ҍҲҸқӃ ҰҸѸӁѹҲҧ ӊҡҾҔҸѦ-ҲҽҔҸѦ. (zhurnal.lib.ru/s/sterhow_a/rubezh.shtml) too long was soldier and too tried be-INF soldier, that-IRR rememberINF such things ‘For too long he had been a soldier, and tried too hard to be a soldier to remember such things.’ Some verbs with the form [ԒԂ-......-ԙԋ] like ԒԂԙӠԞӠӴԂԽӸԙԋ (‘sit/‘stay (too) long’), ԒԂԄӹӴӹӢӠԽӸԙԋ (‘talk (too) long/much’) which can express that the subject overly indulges in some action, show a similar restriction. These verbs never occur with a restrictive subordinate clause, for example: (33) ҷҸѹҲѦҡҧ, ҍҲҸ ӉҽҡѹҦѹѸҡҾ, – ҡѸѦѦҲҡҾ һѹҦҽ. (ԾӹӴՂӡ ԠӠӢ, 2002) be.afraid.2PL-PRES, that sit.too.long.1PL-FUT, laughs Lida ‘‘Are you afraid, that we will keep on sitting here for too long?’, Lida said jokingly.’ Besides ԙԊӠԶӪӹԐ, the form ԓӾӢӾԙԓӫӢ (ǁeresǁur, ‘too’) is used. Like ԙԊӠԶӪӹԐ it can occur with all parts of speech (see section 3 for an overview of the syntactic possibilities of ԙԊӠԶӪӹԐ). Apresjan et al. (2004: 1044) argue that ԓӾӢӾԙԓӫӢ expresses a subjective evaluation of the speaker, and has a more emphatic meaning. According to them, this explains why ՀӾӡԓԂԙ ԙԊӠԶӪӹԐ ӢԂӰӹ, ԐԂԄԂԒӠӰ ӾԷӾ ԒԂӪӢՂԽ (‘It’s too early. The shop is still closed.’) sounds better than the same sentence with ԓӾӢӾԙԓӫӢ. This opinion is in accordance with the observation of the native speakers that I consulted that ԓӾӢӾԙԓӫӢ, and not ԙԊӠԶӪӹԐ, may also be used to indicate that the degree exceeds the standard to a large extent (‘much too much’). In the case of ԙԊӠԶӪӹԐ, special intonational emphasis or reduplication (ԙԊӠԶӪӹԐ-ԙԊӠԶӪӹԐ) is needed to achieve a similar effect. However, I am not sure whether the specific semantics observed by Apresjan plays a part in all contexts. The degree words ԙԊӠԶӪӹԐ and ԓӾӢӾԙԓӫӢ are quite often used in the same text for, what at least seems, stylistic reasons, that is to avoid repetition of the same word: (34) ѷѦҰѦҝҧ ҸӁѹ ҽҲҽҔҏӀҲ ѦѦ Ӊҽ ҲҸ, ҍҲҸ ҸӁҽ ҡґѹҗҔҸѸ ӊѦґѹҔҽҾ ѹ ҍѦҝѦҡҍҏҝ ҡҰҝҽӊѦҦґѹӊҽҾ. (ԚԒӴӾԙԽӠԋ, 2001.08.12) now they attack her for that, that she too great and too righteous ‘Now they attack her for being to great and righteous.’ In such sentences ԓӾӢӾԙԓӫӢ is a stylistically marked synonym of ԙԊӠԶӪӹԐ. Further research is needed here.

THE CONSTRUCTION OF DEGREE IN RUSSIAN

57

2.2 ԟӹԙԽԂԽӹԓӰӹ (‘enough’) The basic meaning of the adverb ԌӹԙԽԂԽӹԓӰӹ (‘enough’) can be formulated as follows: ԌӹԙԽԂԽӹԓӰӹ1 X:

‘A higher degree of X is not necessary or desirable (appropriate degree) with respect to Y’

The meaning of ԌӹԙԽԂԽӹԓӰӹ (and its (near) synonym ԌӹӴӹԊӸӰӹ) can be illustrated with the following sentence: (35) ҫӁ Ѧґ ҦҸҡҲҽҲҸҍӁҸ. ‘He ate enough.’ Here, the adverb ԌӹԙԽԂԽӹԓӰӹ presupposes that one can eat more or less, and that when someone is ‘adding up’ the amount of food that is eaten, it is possible to say at some point that this person can stop eating, and that there is no necessity to continue eating. The sentence implies that some contextually given minimum has been reached with respect to eating (see Meier 2003, for the notion of minimum). Note, however, that the sentence itself does not provide any information about the question whether eating more would be appropriate or not. In (35) the standard Y remains implicit, and is interpreted differently depending on the context (with respect to someone’s health; with respect to not being hungry, etc.). Like in the case of ԙԊӠԶӪӹԐ, the standard can also be expressed by a subordinate clause introduced by (ԌԊԋ ԽӹԄӹ) ԓԽӹԼՂ, or a prepositional phrase with ԌԊԋ (‘for’). Again, the subordinate clause is prototypically interpreted as a case of possibility, but there are also sentences which are closer to necessity than possibility: (36) ҋ ҙҲҸҮ ҡҲҸҝҸӁӃ ѸӃ қӃґѹ ҦҸҡҲҽҲҸҍӁҸ ӊҸҸҝҏңѦӁӃ, ҍҲҸқӃ ӁѦ ѹѸѦҲҧ ҸҡӁҸӊҽӁѹҾ қѦҡҰҸҔҸѹҲҧҡҾ. (ғ. ҘҸӁѦӁҔҸ, ӯ ԘӸӴԂ ԾӠӪӹԊԂӾӴӠԓԂ ԁӹԊԙԽӹԄӹ). from that side we were enough equipped, that-IRR not have-INF reason worry-INF ‘With regard to that, we were equipped well enough, not to have any reasons to worry.’ (37) ғ ҊҸҝҽѽ, ҦҸҡҲҽҲҸҍӁҸ ӊӃҡҸҔҸ, ҍҲҸқӃ ҰҸ ӊѦҍѦҝҽѸ ѽҸҦѹҲҧ ӊ ҔҏҝҲҔҽѽ. (www.moto.ru) in mountains enough high, that-IRR during evenings go-INF in jackets ‘In the mountains, high enough (for it to be a good idea/to have) to wear a jacket during the evenings.’

58

EGBERT FORTUIN

In other contexts the modality cannot be classified with labels such as possibility or necessity, and is better described as ‘making sense to engage in an action’: (38) ҭҸҲҸѸ, ӊѦҦҧ ѹ ѸҽҲѦҝѹҽґҧӁӃҮ ҡҲҽҲҏҡ ӍѹӉѹҔҸӊ қӃґ ӊ ҋҋҋҩ ҦҸҡҲҽҲҸҍӁҸ ӊӃҡҸҔ, ҍҲҸқӃ ѹҦҲѹ ӊ ӍѹӉѹҔѹ, ҽ ӁѦ ӊ ѹҡҲҸҝѹҔѹ. (www.altruism.ru/sengine.cgi/5/7/8/8/2) besides, after.all and material status physicists-GEN was in SSSR enough high, that-IRR go-INF in physics, and not in history ‘Besides that, the material status of physicists in the Soviet Union was high enough to become a physicist, in stead of a historian.’ [i.e. for it to be a good idea, etc.] With regard to the modality, see my analysis for ԙԊӠԶӪӹԐ given in section 2.1. At this point, I would like to point at the conceptual difference between ԌӹԙԽԂԽӹԓӰӹ, which indicates that the degree is appropriate, and ԙԊӠԶӪӹԐ, which indicates that the degree exceeds the appropriate degree. This fundamental difference in meaning can be illustrated with the following combinations: (39) ?ӁѦѸӁҸҊҸ ҦҸҡҲҽҲҸҍӁҸ little enough (40) ӁѦѸӁҸҊҸ ҡґѹҗҔҸѸ little too (41) ?ѦҦӊҽ ҡґѹҗҔҸѸ hardly/barely too (42) ѦҦӊҽ ҦҸҡҲҽҲҸҍӁҸ hardly/barely enough Note that with ԌӹԙԽԂԽӹԓӰӹ the use of ӰӾԐӰӹԄӹ (nemnogo ‘a little (bit)’) is marked because the degree is either conceptualized as appropriate or not (ok or not), and not as partly appropriate. The form ԙԊӠԶӪӹԐ displays a different picture. To indicate that the degree is a little bit above the maximum, the form ӰӾԐӰӹԄӹ (‘a little’) can be used. This is possible, because in contrast to ԌӹԙԽԂԽӹԓӰӹ, this degree word indicates the deviation from the appropriate degree. This deviation may be smaller or larger. In the case of ԌӹԙԽԂԽӹԓӰӹ, it is also possible to use ӾԌӴԂ (edva ‘hardly’, ‘barely’). This form emphasizes that if the degree were a little less, this would amount to a degree under the minimum. With ԙԊӠԶӪӹԐ the use of ӾԌӴԂ is marked, because ӾԌӴԂ suggests a negative evaluation, which is not in accordance with the idea that the speaker wants to express that a little bit less, and the degree would have been perfectly fine (see Verhagen 2005: 45–50, for an analysis of the English barely, which shows similar features as ӾԌӴԂ).

THE CONSTRUCTION OF DEGREE IN RUSSIAN

59

Besides ԌӹԙԽԂԽӹԓӰӹ, the form ԌӹӴӹԊӸӰӹ is used as in (16) given earlier. In modern Russian, this form seems to have the same meaning as ԌӹԙԽԂԽӹԓӰӹ. A different opinion is expressed by Dagurov (1994: 59), who argues that according to the correct norm, ԌӹԙԽԂԽӹԓӰӹ is used to indicate a sufficient degree with respect to some condition, whereas ԌӹӴӹԊӸӰӹ is used as a moderator, indicating a rather high degree of something (in his terminology ԗӢӠԼԊӠԒӠԽӾԊӸӰӹ ‘approximately’, Ԍӹ ӰӾӪӹԽӹӢӹӡ ԙԽӾԗӾӰӠ ‘to a certain extent’): (43) ҴҽґҧҍѹҔ ҝѹҡҏѦҲ ҦҸӊҸґҧӁҸ ѽҸҝҸҗҸ. (Dagurov, 1994: 59) boy draws rather well ‘The boy draws rather well.’ He also remarks that in modern Russian, ԌӹԙԽԂԽӹԓӰӹ is sometimes incorrectly used as a moderator: (44) ҋҏҦѦқӁӃҮ ҰҝҸ҃Ѧҡҡ җѦґ ҦҸҡҲҽҲҸҍӁҸ ѸѦҦґѦӁӁҸ. (Dagurov, 1994: 59; judged incorrect by him) judicial process went rather slowly ‘The judicial proceedings proceeded rather slowly.’ The data at my disposal however, show that in modern Russian (both in newspapers and literature), both forms allow for both readings. 3 Despite the remarks by Dagurov (1994), there are no clear signs of a semantic division of labor between these degree words in modern Russian. See further below, for further discussion of use of ԌӹԙԽԂԽӹԓӰӹ as a moderator. Note that in the existing analyses for English (e.g. Meier 2003) the meaning of enough is defined in terms of a value greater than or equal to the minimum of all values such that some action can be realized. Such an analysis is, however, not in accordance with all the Russian data. In Russian, ԌӹԙԽԂԽӹԓӰӹ can also be used in sentences where a higher degree is in fact undesirable with respect to Y. In the latter case, one can speak of an interpretation of non-occurrence, because one can infer that because of the degree, the situation in the subordinate clause can or will not be realized. The following example is provided by the Russian Grammar (= Švedova 1980). According to the Russian Grammar, it can be interpreted in two ways: 3

The data from the NKRJa show that in the nineteenth century, ԌӹӴӹԊӸӰӹ was the standard form in both meanings. In the twentieth century, ԌӹԙԽԂԽӹԓӰӹ is the most frequently used form. To give an indication, in the NKRJa there are 177 instances of ԌӹԙԽԂԽӹԓӰӹ from the end of the eighteenth century till 1850 against about 2000 instances of ԌӹӴӹԊӸӰӹ. In the second half of the twentieth century, sentences with ԌӹԙԽԂԽӹԓӰӹ are more frequent than sentences with ԌӹӴӹԊӸӰӹ (8000 instances of ԌӹӴӹԊӸӰӹ against 15000 with ԌӹԙԽԂԽӹԓӰӹ).

60

EGBERT FORTUIN

(45) ҫӁ ҦҸҡҲҽҲҸҍӁҸ ҲҽґҽӁҲґѹӊ, ҍҲҸқӃ ҝѦҗҽҲҧ ҲҽҔѹѦ ӉҽҦҽҍѹ. (Russian Grammar (Švedova 1980: 601); repeated example (4)) 1. ‘He is talented enough to be able to solve such problems.’ 2. ‘He is talented enough not to have to solve such problems, i.e. it makes no sense to engage in the action’. There are also sentences where the interpretation of non-occurrence is the only likely interpretation: (46) Ѱ ӉҦѦҗӁѦҮ ҰҸґѹ҃ѹѹ, ҰҝҽӊҦҽ, ҦҸҡҲҽҲҸҍӁҸ ҡӊҸѹѽ ҰҝҸқґѦѸ, ҍҲҸқӃ ӊҸґӁҸӊҽҲҧҡҾ ѦӂѦ ѹӉ-Ӊҽ ҔҽҔѹѽ-ҲҸ ҏқѹҮҡҲӊ ӊ ҴѦҔҡѹҔѦ. (ѩ. ҴҸҝҝѦґґ, өӫԞӹӾ ԊӠՁӹ) 4 at local police, really, enough own problems, that-IRR worry-INF still because.of some murders in Mexico ‘The local police really have too many problems of their own, to have to worry about some murders in Mexico.’ Although such sentences constitute the minority of sentences with ԌӹԙԽԂԽӹԓӰӹ/ԌӹӴӹԊӸӰӹ, they are accepted by many speakers of Russian, and can be found in different genres, ranging form spoken language to newspapers, and novels. This also explains why they are mentioned in the Russian Grammar (Švedova 1980). Note, however, that some speakers reject this use of ԌӹԙԽԂԽӹԓӰӹ or ԌӹӴӹԊӸӰӹ. This point of view is expressed in the following fragment from a radio program on language (ԪԂԌӠӹԗӾӢӾԌԂԓԂ «ԱӢԂԐӹԽӾӡ». ԛՂԗӫԙӪ 9), where the point of view of a Russian linguist/mathematician (V.A. Uspenskij) regarding sentences like (46) is discussed: ‘‘ғ.ү. ѰҡҰѦӁҡҔѹҮ ҸҲѸѦҍҽѦҲ ҲҽҔңѦ қґѹӉҔҏӀ ҸҗѹқҔҏ – ҰҝҸҰҏҡҔ ҸҲҝѹ҃ҽӁѹҾ ҲҽѸ, ҊҦѦ ҸӁҸ ӁҏңӁҸ. ҶҲҽ ҸҗѹқҔҽ, ҔҽҔ ҰѹҗѦҲ ѰҡҰѦӁҡҔѹҮ, ѹ ѸӃ ҡ ӁѹѸ ҡҸӊѦҝҗѦӁӁҸ ҡҸҊґҽҡӁӃ, ҰҝѹҸқҝѦҲҽѦҲ ѽҽҝҽҔҲѦҝ ҙҰѹҦѦѸѹҍѦҡҔҸҊҸ ӉҽқҸґѦӊҽӁѹҾ. ѨѸѦӀҲҡҾ ӊ ӊѹҦҏ ҸқҸҝҸҲӃ ҲѹҰҽ ҫӁ ҦҸҡҲҽҲҸҍӁҸ ҏѸѦӁ, ҍҲҸқӃ ҦѦґҽҲҧ ҰҸҦҸқӁӃѦ ҸҗѹқҔѹ ӊѸѦҡҲҸ ҰҝҽӊѹґҧӁҸҊҸ ҫӁ ҦҸҡҲҽҲҸҍӁҸ ҏѸѦӁ, ҍҲҸқӃ ӁѦ ҦѦґҽҲҧ ҰҸҦҸқӁӃѽ ҸҗѹқҸҔ ѹґѹ ңѦ ҫӁ ҦҸҡҲҽҲҸҍӁҸ ҊґҏҰ, ҍҲҸқӃ ҦѦґҽҲҧ ҰҸҦҸқӁӃѦ ҸҗѹқҔѹ. ѩҸқҽӊѹѸ, ҍҲҸ ӊ ҙҲҸѸ ҡґҏҍҽѦ ҰҝѹҍѹӁҽ ҾӉӃҔҸӊҸҮ ҸҗѹқҔѹ, ӊѹҦѹѸҸ, ҲҽҔңѦ ѹ ӊ ҡѸѦҗѦӁѹѹ ҡґҸӊ ҦҸҡҲҽҲҸҍӁҸ ѹ ҡґѹҗҔҸѸ. ҴӃ ҊҸӊҸҝѹѸ ҫӁ ҋһѨҹҘҫҴ ҏѸѦӁ, ҍҲҸқӃ ҦѦґҽҲҧ ҸҗѹқҔѹ, ӁҸ ҫӁ ѩҫҋѷүѷҫѯѢҫ ҏѸѦӁ, ҍҲҸқӃ Ѣѣ ҦѦґҽҲҧ ҸҗѹқҸҔ. һӀҦѹ, ӁѦқҝѦңӁҸ ҸҲӁҸҡҾӂѹѦҡҾ Ҕ ҾӉӃҔҏ, ҰҏҲҽӀҲ ҙҲѹ ҦӊѦ ҔҸӁҡҲҝҏҔ҃ѹѹ. ѯҲҸ ң, ҸҡҲҽѦҲҡҾ ҲҸґҧҔҸ ҰҝѹӉӊҽҲҧ ѹѽ қӃҲҧ ӊӁѹѸҽҲѦґҧӁѦѦ.’’ 5

4 5

www.aldebaran.ru/tri/morrel/morrel6/?11 rus.1september.ru/articlef.php?ID=200304705.

THE CONSTRUCTION OF DEGREE IN RUSSIAN

61

‘V. A Uspenskij also remarks a similar mistake, namely the omitting of the negation where it is necessary. That mistake, as Uspenskij writes, and we totally agree with him, is getting the character of a contagious disease. We mean sentences like [lit. He is smart enough, to make such mistakes], in stead of the correct [lit. He is smart enough, not to make such mistakes] or [lit. He is stupid enough, to make such mistakes]. We want to add that in this case, it is obvious that the reason for these grammatical mistakes also has to do with the mixing up of the words [lit. enough], and [lit. too]. We say [lit. He is TOO smart, to make such mistakes], but [lit. He is smart ENOUGH, NOT to make such mistakes]. People, who do not have a careful attitude towards language, mix up these two constructions. The only thing left to do is to appeal to them to be more careful.’ According to these authors, sentences like (45) or (46) are incorrect, and the result of sloppy use of language, which has become a trend in recent years. The authors also provide an explanation for these ‘mistakes’, namely the mixing up of ԙԊӠԶӪӹԐ with ԌӹԙԽԂԽӹԓӰӹ, or the omission of negation in the subordinate clause when ԌӹԙԽԂԽӹԓӰӹ is used. At first hand, there is something to be said for arguing that the omitting of the negation is a mistake, because it enhances the potential ambiguity in the linguistic system. The idea underlying this point of view is probably that the meaning of ԙԊӠԶӪӹԐ itself contains a negation, whereas ԌӹԙԽԂԽӹԓӰӹ does not, hence the need to express a negation with ԌӹԙԽԂԽӹԓӰӹ if the infinitive situation does not occur. However, if we look closer at the use of ԌӹԙԽԂԽӹԓӰӹ and an inference of non-occurrence, we can perceive an interpretational logic which is not random, but governed by general principles of interpretation. Two types of uses can be distinguished. Firstly, the inference of nonoccurrence can occur in sentences with indications like ӾԷӾ (‘still’) or ӴԙӾ ӴӢӾԐԋ (‘all the time’) in the subordinate clause: (47) ѢҽҰҝѹѸѦҝ, ҰҝҸѹӉӊѦҦѦӁѹҮ ӊ ӊѹҦѦ ҔҸѸѸѦӁҲҽҝѹѦӊ Ҕ ӁѦҔҸѦѸҏ ҲѦҔҡҲҏ (…) ҏңѦ ӁҽҔҸҰѹґҸҡҧ ҦҸӊҸґҧӁҸ ѸӁҸҊҸ, ҍҲҸқӃ ѸҸҊґҽ ѦӂѦ ѹҦҲѹ ҝѦҍҧ Ҹқ ҏӁѹҔҽґҧӁҸҡҲѹ ѹ ӁѦҸқӃҍӁҸҡҲѹ (…). (ԾӹӴՂӡ ԠӠӢ, 2000, ҥ2) for.example, works-GEN in form commentaries-GEN to some text (…) already accumulated enough much, that-IRR could still go-INF talk on uniqueness and uncommonness ‘For example, works in the form of commentaries to some text are copied in such a big amount, that they cannot be considered unique or uncommon.’ (48) ѢҽҰҝҽҡӁҸ ӊӃ ӊҡѦ ҡӊҸҦѹҲѦ Ӂҽ ѦӊҝѦѦӊ. Ѱ Ӂҽҡ ҦҸҡҲҽҲҸҍӁҸ ҡӊҸѹѽ ҰҝҸқґѦѸ, ҍҲҸқӃ ӊҡѦ ӊҝѦѸҾ ҦҏѸҽҲҧ: ‘ѦӊҝѦѹ..ѦӊҝѦѹ..’. (www.kp.ru/daily/forum/article/5382/)

62

EGBERT FORTUIN

in.vain you everything blame on Jews. at us enough own-GEN problemsGEN, that-IRR all time think-INF: "Jews. Jews.." ‘You are wrong to blame the Jews for everything. We have too many problems of our own, to think all the time: ‘the Jews, the Jews’.’ In all these examples the situation in the main clause is characterized as something where ‘adding up’ properties would be evaluated negatively because the degree is close to a point which would or already has exceeded the maximum. Because of this, realizing an extra situation (‘in addition to what is already realized’), or realizing another situation in a big quantity (‘all the time’), would increase the degree of the situation expressed in the main clause to such an extent that it would be evaluated negatively. Secondly, in some contexts, the interpretation of non-occurrence also occurs without the above-mentioned indications: (49) ҭҸҡҽҦҔҽ – ҦҸҡҲҽҲҸҍӁҸ ӊҽңӁӃҮ ҰҝҸ҃Ѧҡҡ, ҍҲҸқӃ ҦҸӊѦҝҾҲҧ ѦҊҸ ҸҦӁҸѸҏ ҍѦґҸӊѦҔҏ. (ғ. ғҽҡѹґҧѦӊ, Ufo: ԛӢԂԄ ԾӾӠԒӴӾԙԽӾӰ) 6 landing – enough important process, that-IRR entrust-INF him one-DAT person-DAT ‘The landing is too important, to entrust it to one person.’ (50) ҴҸӁҲѹ ҡѹґҧӁӃҮ, ҏѸӁӃҮ ѹ ҦҸҡҲҽҲҸҍӁҸ ҸҰҽҡӁӃҮ қҸҔҡѦҝ, ҍҲҸқӃ ҸҲӁҸҡѹҲҧҡҾ Ҕ ӁѦѸҏ ґѦҊҔҸѸӃҡґѦӁӁҸ. (www.klitschko.com/news/archive.php3?scmon=200007&read=200007/1 6&part=ru) Monty strong, wise and enough dangerous boxer, that-IRR act-INF towards him lightly ‘Monty is a strong and wise boxer, and too dangerous to be lightly regarded.’ In these sentences it is expressed that the degree is rather high such that it is not a good idea to engage in the realization of the action expressed by the subordinate clause. The reason why this interpretation occurs is pragmatic. A possible clue for this interpretation is that the situation expressed by the subordinate clause itself is seen as something negative, or as something that could easily lead to negative consequences. There may, however, also be formal factors associated with this type of use, such as intonation. 7 6

books.rusf.ru/add-on/xussr_av/vasilv04/vasilv04.34.html According to the Russian Grammar (= Švedova 1980) the aspect of the verb also plays an important role in distinguishing the interpretation of non-occurrence from the interpretation of occurrence, because the interpretation of non-occurrence is correlated with an imperfective aspect. In all the examples I have attested, the infinitive does indeed occur in the imperfective 7

THE CONSTRUCTION OF DEGREE IN RUSSIAN

63

It is possible that the possibility of the inference of occurrence is connected to the fact that ԌӹԙԽԂԽӹԓӰӹ (ԌӹӴӹԊӸӰӹ) can also be used as a moderator, often to indicate that the degree is higher than could be expected (cf. English quite): ԌӹԙԽԂԽӹԓӰӹ2 X:

‘Not a low degree of X’ (cf. English ‘rather’, ‘reasonably’, ‘quite’, fairly’) 8

This meaning is exemplified by the following sentences: (51) ҤҸҍҏ ҡҔҽӉҽҲҧ ҦҸҡҲҽҲҸҍӁҸ ҸҍѦӊѹҦӁҏӀ ӊѦӂҧ. (ԚԒӴӾԙԽӠԋ, 2002.12.19) want-1SG-PRES say-INF rather obvious thing ‘I would like to say something rather obvious.’ (52) ҶҲѹ Ҧӊҽ ҲѹҰҽ ҝҽҔҽ ӊҡҲҝѦҍҽӀҲҡҾ ҦҸҡҲҽҲҸҍӁҸ ҍҽҡҲҸ. (ԚԒӴӾԙԽӠԋ, 2002.02.12) these two types cancer-GEN occur quite frequently ‘These two types of cancer occur quite frequently.’ According to Vasil’ev (2001), there is a trend in modern Russian to use ԌӹԙԽԂԽӹԓӰӹ especially with phenomena or objects that are evaluated negatively, for example ԌӹԙԽԂԽӹԓӰӹ ԄӢӫԙԽӰՂӡ (‘sad’), ԼӾԌӾӰ (‘poor’), ӰӾԗӢӠԋԽӰӹ (‘unpleasant’), etc. 9 Although this meaning must be listed separately, it can be linked to the basic meaning of ԌӹԙԽԂԽӹԓӰӹ, because the meaning of ԌӹԙԽԂԽӹԓӰӹ1 presupposes that a higher degree of X is not necessary, for example because it is already rather/quite high. 10 A similar use can be found in other aspect. This seems to be related to the meaning of the construction that expresses that the degree of X is such, that it does not make sense to engage in the realization of the infinitive situation. The focus on engaging in a situation, rather than on the possible completion is typical for the imperfective aspect, rather than the perfective aspect. Note, however, that the imperfective aspect is also typical of sentences with the inference of occurrence, which means that aspect cannot be a decisive factor. Furthermore, in sentences like (49) the use of the perfective is not precluded. 8 Although this meaning of ԌӹԙԽԂԽӹԓӰӹ sometimes seems close to an intensifier, the occurrence of sentences like the following (in this case a review of the control panel of the remote control on zoom.cnews.ru) shows that it is closer to that of a moderator: (…) ӪӰӹԗӹӪ ӰӾԐӰӹԄӹ, ӢԂԙԙԽӹԋӰӠӾ ԐӾԞԌӫ ӰӠԐӠ ԌӹԙԽԂԽӹԓӰӹ ӴӰӫԶӠԽӾԊӸӰӹӾ (ӠӰӹԄԌԂ ԌԂԞӾ ӹԓӾӰӸ)…; ‘…there are few buttons, and the distance between them is quite large (sometimes even very)…’. 9 This also seems to play a role with moderators in other languages. Compare for example Dutch nogal (‘rather’, ‘quite’), which seems to be used more easily with forms that express negative phenomena. 10 Note that like English rather, this use does not occur with negation (put differently, one cannot negate the moderately high degree). If the speaker wants to express that the degree is lower than expected, or lower than the average, forms like ӰӾ ӰԂԙԽӹԊӸӪӹ/ӰӾ ԽԂӪ (‘not that’) or ӰӾ ӹԓӾӰӸ/ӰӾ ԙԊӠԶӪӹԐ (‘not very’) are used.

64

EGBERT FORTUIN

languages such as French, where the degree word assez (‘enough’), can also be used as a moderator. Note that because of the close relation between these two meanings, it is often difficult to tell them apart. This probably also explains the occurrence of sentences with an interpretation of non-occurrence like (48), (49) and (50). It should be noted, however, that when used to indicate a minimum, ԌӹԙԽԂԽӹԓӰӹ does not necessarily express a high degree, but only a high degree with respect to some (implicitly) given standard. It is therefore possible to say, that something is small from a general perspective (small for a particular class of objects, or small from a general perspective), but big enough with regard to some contextually given standard: (53) ҴҽґѦӁҧҔѹҮ ҦӊҸҝѦ҃ ҡѹҾѦҲ ҡӊҸѹѸѹ 7,600 ҔҝѹҡҲҽґґҽѸѹ ѹ ҦҸҡҲҽҲҸҍӁҸ ӊѦґѹҔ, ҍҲҸқӃ ҝҽӉѸѦҡҲѹҲҧ ҲҽѸ (…) ҡҸқҽҍҔҏ, ҲѹҰҽ ҍѹѽҏҽ-ѽҏҽ ѹґѹ ҲѦҝҧѦҝҍѹҔҽ. (www.infodog.ru/index.php?go=News&in=view&id=2715) small palace shines own-INSTR 7,600 crystals-INSTR and enough huge, that-IRR accommodate-INF there (…) dog-ACC, type-GEN Chihuahua, or Yorkshire.Terrier ‘The small palace shines with its 7,600 crystals and is large enough to accommodate a small dog, like the Chihuahua, or the Yorkshire Terrier.’ (54) ҋѦҝӃҮ, қґѦҡҲҾӂѹҮ ѹ ҡҸӊҡѦѸ ѸҽґѦӁҧҔѹҮ. ҫҦӁҽҔҸ ҦҸҡҲҽҲҸҍӁҸ қҸґҧҗҸҮ, ҍҲҸқӃ ҏқѹҲҧ. (www.lib.ru/NATUR/HERRIOT/sozdaniya.txt) grey, shining and really small. however, enough big that-IRR kill-INF ‘Grey, shining and really small. However, big enough to kill.’ As such, ԌӹԙԽԂԽӹԓӰӹ differs from the degree word ӰԂԙԽӹԊӸӪӹ as in (18) and (19), which functions as an intensifier in most contexts, with the exception of sentences where the exact degree of something is given, usually when the degree is situated in-between two points of a scale: (55) ҋѦҔҝѦҲ ҡҲҝѹңҔѹ ӉҽҔґӀҍҽѦҲҡҾ ӊ ҲҸѸ, ҍҲҸ Ҳҝҽӊҽ ҦҸґңӁҽ қӃҲҧ ӁҽҡҲҸґҧҔҸ ӊӃҡҸҔҽ, ҍҲҸқӃ ҔҸҝӁѹ ҰҸґҏҍҽґѹ ҦҸҡҲҽҲҸҍӁҸѦ ҔҸґѹҍѦҡҲӊҸ ҰѹҲҽҲѦґҧӁӃѽ ӊѦӂѦҡҲӊ, ѹ ӁҽҡҲҸґҧҔҸ ҔҸҝҸҲҔҽ, ҍҲҸқӃ ҊҽӉҸӁ ӊӃҊґҾҦѦґ ҔҝҽҡѹӊҸ. (Ѩ. ҈қҝҸҗҔҸ, ԜԽ ԗӫԙԽՂӢԋ Ԍӹ ԗӢӾӪӢԂԙӰӹԄӹ ԙԂԌԂ // Homes & Gardens, 2002) ‘The secret of mowing is that the grass should be long enough [lit. so long], for the roots to receive enough nutritive materials, and short enough [lit. so short], for the lawn to look nice.’ (56) ҈ҽҲѦѸ – ӁҽҦҽӊґѹӊҽѦѸ ӁҽҡҲҸґҧҔҸ ҡѹґҧӁҸ, ҍҲҸқӃ ҰҝҸҦҽӊѹҲҧ ҊҝҏҦѹӁҏ Ӂҽ ӁѦҡҔҸґҧҔҸ ҡҽӁҲѹѸѦҲҝҸӊ. (ӮӫԊӠԄԂӰ, 2004) ‘After that, we push hard enough [lit. so hard], that the chest is pushed downwards a few centimeters.’

THE CONSTRUCTION OF DEGREE IN RUSSIAN

65

According to Bierwisch (1989: 97) enough, or its German counterpart genug, behaves differently from too (German zu) with respect to the relativity of the degree. He argues that in the case of the positive pole of a set of antonyms (+Pol) such as tall, the degree is measured relative to the subordinate clause, whereas a combination of enough and the negative pole of a set of antonyms (Pol) (short enough), [enough + adjective] has a norm related value (with respect to the norm or dimension indicated by the noun). It may indeed be true, that the use of ԌӹԙԽԂԽӹԓӰӹ, both with negative and polar adjectives, easily leads to an interpretation where the degree is evaluated as high from a general (norm related) perspective. This inference can, however, be overruled by the context, as in the following sentence, where something is considered quite big from a general perspective, but small enough with respect to the situation in the subordinate clause: (57) ҐҸҝҸҦ ҦҸҡҲҽҲҸҍӁҸ қҸґҧҗҸҮ, ӁҸ ҦҸҡҲҽҲҸҍӁҸ ѸҽґѦӁҧҔѹҮ, ҍҲҸқӃ ӉҽӊҲҝҽ ӊҡѦ ҏӉӁҽґѹ ӁҸӊҸҡҲҧ’. (www.executive.ru/discussions/forum_20756/msg_194866_1414743/) ‘The city is rather/quite big, but small enough for everyone to know about the news tomorrow.’ Nevertheless, in some contexts it is difficult to overrule the inference that the degree indicated by ԌӹԙԽԂԽӹԓӰӹ is high. In Russian, this inference is probably strengthened by the fact that ԌӹԙԽԂԽӹԓӰӹ can also indicate a (rather/quite) high degree. The close relation between the relative use (where ԌӹԙԽԂԽӹԓӰӹ does not necessarily express a (rather) high degree) and the moderator use (where it does) probably also accounts for the observation that at least for some native speakers of Russian (59), where the degree word has the form of short adjective, sounds better than (58), where we find the degree word in its normal form: 11 (58) ҤҸҲҾ Ҿ ӁѹӉҔҸҊҸ ҝҸҡҲҽ, Ҿ ҦҸҡҲҽҲҸҍӁҸ ӊӃҡҸҔѹҮ, ҍҲҸқӃ ѸѦӁҾ ҰҝѹӁҾґѹ. although I short-GEN height-GEN, I enough high, that-IRR me accepted ‘Although I am small, I am tall enough to get the job.’ (59) ҤҸҲҾ Ҿ ӁѹӉҔҸҊҸ ҝҸҡҲҽ, ѸҸҮ ҝҸҡҲ ҦҸҡҲҽҲҸҍѦӁ ҦґҾ ҲҸҊҸ, ҍҲҸқӃ ѸѦӁҾ ҰҝѹӁҾґѹ. although I short-GEN height-GEN, my height-MASC-NOM enoughADJECTIVE-SHORT-MASC for that, that-IRR me accepted ‘Although I am small, my height is sufficient for me to get the job.’

11

Thanks to Elena Tribušinina for this example, and this specific linguistic judgement.

66

EGBERT FORTUIN

Put differently, the difference in acceptability seems to be related to the fact that [ԌӹԙԽԂԽӹԓӰӹ + Adj], but not the short form ԌӹԙԽԂԽӹԓӾӰ, can be used as a moderator, which is sometimes close in meaning to an intensifier. Besides the forms ԌӹԙԽԂԽӹԓӰӹ and ԌӹӴӹԊӸӰӹ, it is possible to find instances of the quantifier ԐӰӹԄӹ (mnogo ‘much’) without ԌӹԙԽԂԽӹԓӰӹ. In the case of the instances I found, the quantifier functioned as the object of the verb. Two examples are given below: (60) Ѥ ӉӁҽӀ ѸӁҸҊҸ ҍҲҸқӃ ҔҽҔ-ҲҸ ӊӃңѹҲҧ ѹ ҡґѹҗҔҸѸ ѸӁҸҊҸ ҍҲҸқӃ ҰҝҸҡҲҸ ңѹҲҧ. (www.karendjangirov.com/belaya.html) I know much that-IRR somehow survive-INF and too much that-IRR just live-INF ‘I know enough to somehow survive, and too much to just live.’ (61) ѣҡґѹ ҲӃ ҰӃҲҽѦҗҧҡҾ ѸѦӁҾ ѹӉѸѦӁѹҲҧ/Ѥ ӊѹҦѦґҽ ѸӁҸҊҸ, ҍҲҸқӃ ӊҡѦ ҰҝѦҔҝҽҲѹҲҧ. (forum.fanzona.tv/archive/index.php/t-37.html) 12 if you try me change/I saw much, that-IRR all stop-INF ‘If you are trying to change me/I have seen enough, to stop everything.’ In these sentences, ԐӰӹԄӹ indicates that the degree is evaluated as much, even too much (that is, there is a negative evaluation of the degree). At the same time this degree does not lead to the non-realization of the situation given in the subordinate clause, and as such its meaning is close to ‘enough’. Consultation with native speakers of Russian indicated that sentences like (60) and (61) are considered acceptable but stylistically marked. Note that there are also sentences where ԐӰӹԄӹ is interpreted as ‘too much’: (62) [ғ]ҡҰҸѸѹӁҽҾ Ҹ ґҽҊѦҝӁҸҮ ҰҽҮҔѦ, Ӂҽ ҔҸҲҸҝҸҮ ҸӁѹ ҡѹҦѦґѹ ҰҸ ҦѦҡҾҲҧ ѹ қҸґѦѦ ґѦҲ, "ҡѹқѹҝҾҔѹ" ҊҸӊҸҝҾҲ: ‘ҴҽґҸ, ҍҲҸқӃ ңѹҲҧ, ѸӁҸҊҸ, ҍҲҸқӃ ҏѸѦҝѦҲҧ...’ (www.memorial.krsk.ru/public/00/20010223.htm) remembering about camp ration, on which they sat for ten and more years, ‘Siberians’ say: ‘little, that-IRR live-INF, much, that-IRR die-INF… ‘Remembering the ration of food that they were put on in the camp for ten years or more, the ‘Siberians’ say: ‘Too little to live, too much to die.’ The degree word ԌӹԙԽԂԽӹԓӰӹ also shows similarities to the degree word ԗӢӾԌӹԙԽԂԽӹԓӰӹ (‘more than enough’), which consists of the prefix ԗӢӾ-, indicating a high degree, and ԌӹԙԽԂԽӹԓӰӹ. This word occurs with nouns (in the

12

Russian translation of a pop song. The original text: If you’re trying to turn me into something else/I’ve seen enough and I’m over that. (Avril Lavigne, Nobody's Fool)

THE CONSTRUCTION OF DEGREE IN RUSSIAN

67

genitive case) and is used to express that the quantity exceeds the minimum degree of appropriateness to a large extent: 13 (63) ғҝѦѸѦӁѹ қӃґҸ ҰҝѦҦҸҡҲҽҲҸҍӁҸ, ҍҲҸқӃ ҸҲѸѦҲѹҲҧ ҔҸѸҽӁҦѹҝҸӊҸҍӁӃѦ ҏҦҸҡҲҸӊѦҝѦӁѹҾ ѹ ҰҸҡѦҲѹҲҧ ҸҊҝҸѸӁӃҮ ҸҲґѹҍӁҸ ҡѦҝӊѹҝҸӊҽӁӁӃҮ қҏӍѦҲ ҡ ҸҍѦӁҧ ҰҝѹґѹҍӁӃѸ ҔҸӍѦ ѹ қҏҲѦҝқҝҸҦҽѸѹ. (ԛӹԙԽӹԓӰӹ-ՀӠԼӠӢԙӪԂԋ ԗӢԂӴԌԂ (ѨҝҔҏҲҡҔ), 2003.06.25) ‘There was more than enough time, to meet the authorities and visit the large, full range buffet with excellent service and very good coffee and sandwiches.’ In many cases it is used as a marker of a high degree without explicit relative function: (64) ѢѦҸңѹҦҽӁӁҸҡҲѦҮ қӃґҸ ҰҝѦҦҸҡҲҽҲҸҍӁҸ. (ӯӢԂԊӸԙӪӠӡ ԂӴԽӹԐӹԼӠԊӸ, 2004.01.13) ‘There were more than enough unexpected things.’ Besides these degree words, the main clause may also contain a verb that has the meaning of ‘enough’ as part of its lexical meaning, for example ԖӴԂԽӠԽӸ (xvatit’) (‘be enough’, ‘suffice’) (see Kor-Chahine (forthc.) for an analysis): (65) ҋҸқҝҽӁӁӃѽ ҡҝѦҦҡҲӊ ѽӊҽҲѹґҸ, ҍҲҸқӃ ҰҸҝҽҦҸӊҽҲҧ ҡґҽҦҸҡҲҾѸѹ ҡӊӃҗѦ 500 ҝѦқҾҲѹҗѦҔ. (ԠԂӢӠӡԙӪԂԋ ԗӢԂӴԌԂ, 2003.06.01) collected-GEN-PL means-GEN-PL sufficed-3SG-NEUT, that-IRR make.happy-INF sweets-INSTR more 500 boys ‘The collected means sufficed to make more than 500 boys happy with sweets.’ ѷhe adverbs ԙԊӠԶӪӹԐ and ԌӹԙԽԂԽӹԓӰӹ can occur with different parts of speech, such as adjectives, adverbs, verbs and nouns. In section 3, I will discuss the semantics of these syntactic combinations in more detail. 3 Syntax and semantics of ԙԊӠԶӪӹԐ (‘too’) and ԌӹԙԽԂԽӹԓӰӹ (‘enough’) 3.1 Adjectives and adverbs In the following sentences ԙԊӠԶӪӹԐ modifies an adjective. The adjective can occur with a noun [ԙԊӠԶӪӹԐ Adj N] as in (66), or without noun [ԙԊӠԶӪӹԐ Adj] as in (67):

13

Note that this degree word does not occur with adjectives or verbs: *ԗӢӾԌӹԙԽԂԽӹԓӰӹ ɤɪɚɫɢɜɵɣ (more.than.enough beautiful); *ӵ ԗӢӾԌӹԙԽԂԽӹԓӰӹ ԙԗԂԊ (I more.than.enough slept).

68

EGBERT FORTUIN

(66) ғҸҮӁҽ ҡґѹҗҔҸѸ ҡѦҝҧѦӉӁҸѦ ҦѦґҸ, ҍҲҸқӃ ҦҸӊѦҝҾҲҧ ѦѦ ҊѦӁѦҝҽґҽѸ. (ԪӹԙԙӠӡԙӪԂԋ ԄԂԒӾԽԂ, 2003.07.07) war too serious-ADJ-NEUT-SG-NOM matter-N-NEUT-SG-NOM, thatIRR entrust-INF her generals-DAT ‘War is too serious a matter to entrust to generals.’ (67) ҴҸҝҽґҧ ҡґѹҗҔҸѸ ҰҸҦӊѹңӁҽ, ҍҲҸқӃ ӉҽҦҽӊҽҲҧ ҙҲѹ ҊҝҽӁѹ҃Ӄ. (ԚԒӴӾԙԽӠԋ, 2001.09.04) morality-N-FEM-SG-NOM too flexible-ADJ-SHORT-FEM, that-IRR draw-INF these lines ‘Morality is too flexible to draw such lines.’ In these sentences the degree of the property expressed by the adjective is presented as too high. In sentences where the adjective does not modify a noun, the adjective usually occurs in the short form. A similar use occurs with adverbs which express a property of a situation expressed by a verb: (68) ҭҝѹѦѽҽґ ӊҸ һҧӊҸӊ ҡґѹҗҔҸѸ ҰҸӉҦӁҸ, ҍҲҸқӃ ӊ ҲҸҲ ңѦ ҦѦӁҧ ҸҲҰҝҽӊѹҲҧҡҾ ӊ ҴҸҡҔӊҏ. (ԩӴӾԒԌԂ, ҥ 5, 2002) arrived in L’vov too late-ADV, that-IRR in that PRT day leave-INF for Moscow ‘I arrived in L’vov too late to leave for Moscow the same day.’ The adverb ԌӹԙԽԂԽӹԓӰӹ behaves in the same way. When it is applied to an adjective (as in (69) or (70)) or adverb (as in (1)), it indicates that the degree of the property expressed by the adjective or adverb is sufficient: (69) ҫӁ ҦҸҡҲҽҲҸҍӁҸ ҊҝѦҗѦӁ, ҍҲҸқӃ ҡҲҽҲҧ ҰҽҡҲҸҝҸѸ, ѹ ҦҸҡҲҽҲҸҍӁҸ ҡӊҾҲ, ҍҲҸқӃ қӃҲҧ ѹҡҰҸӊѦҦӁѹҔҸѸ. (ԁӾԂԽӢԂԊӸӰԂԋ ԞӠԒӰӸ, 2004.06.28) he enough sinful-ADJ-SHORT-MASC, that-IRR become pastor, and enough holy-ADJ-SHORT-MASC, that-IRR be-INF confessor ‘He is sinful enough to become a pastor, and holy enough to be a confessor.’ (70) Ѥ ӁѦ ҦҸҡҲҽҲҸҍӁҸ ѽҸҝҸҗѹҮ ҝѦңѹҡҡѦҝ, ҍҲҸқӃ ҡҲҽӊѹҲҧ ҰґҸѽѹѦ ҰҧѦҡӃ. (ѨӉӊѦҡҲѹҾ, 2002.01.14) I not enough good-ADJ-MASC-SG-NOM director-N-MASC-SG-NOM that-IRR stage-INF bad plays ‘I am not a good enough director to stage bad plays.’ In comparison to the other parts of speech there are few restrictions on the class of adjectives or adverbs that can occur with ԙԊӠԶӪӹԐ or ԌӹԙԽԂԽӹԓӰӹ. Note, however, that some adjectives, for example ԐӾӢԽӴՂӡ (‘dead’) are usually not

THE CONSTRUCTION OF DEGREE IN RUSSIAN

69

viewed as gradable, because of their lexical semantics. In other cases, for example ԌӾӢӾӴԋӰӰՂӡ (‘wooden’), the property is not associated with a particular degree or intensity, and other conceptualizations are used (for example ԙԊӠԶӪӹԐ ԐӰӹԄӹ ԌӾӢӾӴԂ (‘too much wood’), ԙӹӴԙӾԐ ԌӾӢӾӴԋӰӰՂӡ (‘completely wooden’). See Bolinger 1972, for a discussion about English adjectives and degree words. 14 Sentences with an adjective or adverb differ from uses with verbs or nouns, because the quantifier ԐӰӹԄӹ (‘much’) is not used: (71) ҡґѹҗҔҸѸ ѸӁҸҊҸ Ҕҝѹҍҽґ too much shouted-V ‘shouted to much’ (72) ҡґѹҗҔҸѸ ѸӁҸҊҸ ҦѦӁѦҊ too much money-N-GEN ‘too much money’ (73) *ҡґѹҗҔҸѸ/ҦҸҡҲҽҲҸҍӁҸ ѸӁҸҊҸ ҡѦҝҧѦӉӁҽҾ too/enough much serious-ADJ (74) *ҡґѹҗҔҸѸ/ҦҸҡҲҽҲҸҍӁҸ ѸӁҸҊҸ ҰҸӉҦӁҸ too /enough much late-ADV In Russian, the use of ԐӰӹԄӹ (‘much’) is typical of countable entities, or other entities that are associated with a particular quantity such as mass nouns or nouns expressing more abstract phenomena like ԙӠԊԂ (‘strength’), ӴӰӠԐԂӰӠӾ (‘attention’), etc. These are prototypically things expressed by nouns, or, in some cases, situations expressed by verbs (see 3.2.3). The restriction on ԐӰӹԄӹ with degree words runs parallel to the restriction in sentences without degree word: (75) *ѸӁҸҊҸ ҡѦҝҧѦӉӁҽҾ/ҰҸӉҦӁҸ much serious-ADJ/late-ADV In modern Russian, the adverb ԐӰӹԄӹ cannot be used to express that the intensity of the property expressed by the adjective or adverb is high (but see 3.2.3 for observations about the situation in old Russian). Instead, the most neutral way

14 Note that many prepositional phrases behave like adjectives in this respect, for example ӾԖԂԽӸ ԙԊӠԶӪӹԐ ӰԂ ԙӾӴӾӢ (lit. ‘drive too much to the north’). In contrast to languages like Dutch or English, in Russian the use of a quantifier is not possible here (*ԙԊӠԶӪӹԐ ԐӰӹԄӹ ӰԂ ԙӾӴӾӢ). This means that these PP’s are conceptualized as expressing gradable phenomena rather than quantifiable phenomena. In this specific case, there is also a corresponding adjective, which has a slightly different meaning, namely ԙӾӴӾӢӰՂӡ (lit. ‘northern’), as in ԄӹӢӹԌ ԙԊӠԶӪӹԐ ԙӾӴӾӢӰՂӡ (lit. ‘the city is too northern’).

70

EGBERT FORTUIN

to express an intensifying meaning with an adjective is by using ӹԓӾӰӸ (oǁen’; ‘very’) 15 : (76) ҸҍѦӁҧ ҡѦҝҧѦӉӁҽҾ/ҰҸӉҦӁҸ very serious-ADJ/late-ADV Note, however, that the quantifier ӰԂԐӰӹԄӹ (namnogo ‘much too much’, ‘much to a large extent’) or ԐӰӹԄӹ can be used in the case of comparatives: (77) ӁҽѸӁҸҊҸ/ѸӁҸҊҸ ҡѦҝҧѦӉӁѦѦ/ҰҸӉҦӁѦѦ much/much serious-COMP/late-COMP In such sentences, the quantifier does not indicate that the degree of the quality expressed by the comparative is high, but the (big) extent to which the degree differs from some other contextually given degree. The degree word ԙԊӠԶӪӹԐ can be combined with the adverb ԐԂԊӹ (‘few’; ‘little’) to indicate that the degree is too low. Such cases are very rare. ѷhe only two examples I have attested in the NKRJa are from the 19th century. An example is given below: (78) ғҸҸқӂѦ ңѦ ҸӁ қӃґ ҡґѹҗҔҸѸ ѸҽґҸ ҡѦҝҧѦӉѦӁ[.] (Ѣ.ү. ѩҸқҝҸґӀқҸӊ, ԩԂԼӠԽՂӾ ԊԇԌӠ (1861)) in.general PRT he was too little-ADV serious-ADJ ‘In general, he was not serious enough.’ This indicates that in the case of ԙԊӠԶӪӹԐ and ԌӹԙԽԂԽӹԓӰӹ the idea of a high degree (or quantity) is the unmarked dimension. This can be compared to the use of contrary antonyms in English like big-small, old-young, etc., where the adjective that refers to the upper part of the scale can be used in how-questions like How old are you? The concept of a low degree (or small quantity) therefore has to be expressed additionally in sentences like (78). However, in many of such contexts the adverb ԌӹԙԽԂԽӹԓӰӹ plus negation is preferred (ӰӾ ԌӹԙԽԂԽӹԓӰӹ, or as one word ӰӾԌӹԙԽԂԽӹԓӰӹ, with the meaning ‘not really enough’). This shows that these expressions are not synonymous. Although [ԙԊӠԶӪӹԐ + Adj] expresses that the degree of the quality expressed by the adjective is too high, it is not used with the intensifier ӹԓӾӰӸ (oǁen’ ‘very’), which indicates that the degree of the quality expressed by the adjective or adverb is very high: (79) ҸҍѦӁҧ қҸґҧҗҸҮ very big-ADJ 15

See Kennedy and McNally 2005 for the difference between much+ADJ, well+ADJ and very +ADJ in English.

THE CONSTRUCTION OF DEGREE IN RUSSIAN

71

(80) ?ҡґѹҗҔҸѸ ҸҍѦӁҧ қҸґҧҗҸҮ too very big-ADJ The explanation is that ‘x is too y’ does not imply that ‘x is very y’, because the high degree of y is only contextually determined (a high degree with respect to Z). There are, however, exceptions, where we find the combination [ԙԊӠԶӪӹԐ ӹԓӾӰӸ ADJ]: (81) ҭҸҲҸѸҏ ҍҲҸ ҏ ѸѦӁҾ қӃґѹ ѹӁҊґѹҗѹ. ҫҍѦӁҧ ґӀқѦӉӁӃѦ. ѢѦѸӁҸңҔҸ ҦҽңѦ ҡґѹҗҔҸѸ ҸҍѦӁҧ ґӀқѦӉӁӃѦ. (Ҍ. ҐѦҝѸҽӁ, ԟӹӢӹԄӹӡ Ԑӹӡ ԓӾԊӹӴӾӪ) because at me were Englishmen. very nice-ADJ-PL-NOM. little even too very nice-ADJ PL-NOM ‘Because there were Englishmen. Very nice people. Even a little bit too nice.’ In such sentences, the degree word is applied to an explicitly or implicitly given idea of ‘very X’. Such a reference may, however, also be absent, especially in the spoken language or written language with colloquial features: (82) ѢҸ ҝѦҽґҧӁҸ, 18 ҝҏқ. Ӊҽ ґѹҲҝ 95 - ҙҲҸ ҔҝҏҲҸ, ѸӃ ңѦ ӁѦ ӊ ѣӊҝҸҡҸӀӉѦ! ғѹҦѹѸҸ, ҡґѹҗҔҸѸ ҸҍѦӁҧ қҸґҧҗҽҾ ҃ѦҰҸҍҔҽ (ҔҽҔ ҸқӆҾҡӁҾґ ѸӁѦ ҸҦѹӁ ҽӊҲҸҝѹҲѦҲӁӃҮ ҍѦґҸӊѦҔ) Ӂҽ ҙҲҸѸ ҔҸҝѸѹҲҡҾ! (Forum) (www.ttw.ru/index.php?name=PNphpBB2&file=viewtopic&p=58348) ‘But really, 18 rubles for 95 litres of petrol, that is madness, after all, we do not live in the European Union! Clearly, it really is far too big a chain of people [lit. too very big a chain] (as an authority on this matter explained to me) that make their living on that.’ 3.2 Verbs 3.2.1 Verb classes If ԙԊӠԶӪӹԐ or ԌӹԙԽԂԽӹԓӰӹ are applied directly to a verb, they indicate that the degree to which the situation expressed by the verb is performed, is too high or sufficient. An example was already given with ԙԊӠԶӪӹԐ in (5), and in (83) an example is given with ԌӹԙԽԂԽӹԓӰӹ: (83) Ѩ ӊҸҲ Ҿ ҏңѦ ҦҸҡҲҽҲҸҍӁҸ ҰҸҡҲҽҝѦґ, ҍҲҸқӃ ҰҸґӀқѹҲҧ ңѹӉӁҧ. (Ҷ. ҩҽҦӉѹӁҡҔѹҮ, ӷӢӹԌӹԊԞӾӰӠӾ ԟӹӰ ԑӫԂӰԂ) and PRT I already enough aged, that-IRR love-INF life ‘And I already grew sufficiently old to start loving life.’ Different classes of verbs can occur in the construction of degree. A search in the corpus (NKRJa) for constructions with ԙԊӠԶӪӹԐ in the main clause, and a subordinate clause (ԓԽӹԼՂ) resulted in the following verb classes:

72 (i)

(ii)

(iii)

(iv)

(v)

EGBERT FORTUIN

verbs that indicate mental activities, for example ԗӹӰӠԐԂԽӸ (‘understand’), ԒӰԂԽӸ (‘know’), ԗӹԐӰӠԽӸ (‘remember’), that one can do fully or partly or with different levels of quality [degree of quality/completeness of the action] verbs that express a positive or negative attitude towards someone/something such as ԊԇԼӠԽӸ (‘love’), ӫӴԂԞԂԽӸ, ԌӹӢӹԞӠԽӸ, ՁӾӰӠԽӸ (‘honor’, etc.); ԼԂԊӹӴԂԽӸ (‘spoil’), ӰӢԂӴӠԽӸԙԋ (‘like’), ӴӾӢӠԽӸ (‘believe’), ԗӢӾԒӠӢԂԽӸ (‘despise’), ӰӾӰԂӴӠԌӾԽӸ (‘hate’), etc. [degree of intensity; level of positiveness or negativeness of feeling] verbs that express a negative experience of a subject such as ԙԽӢԂԌԂԽӸ, ӠԒԐӹԽԂԽӸԙԋ, ԒԂԼӹԽӠԽӸԙԋ (‘suffer’) [degree of intensity; level of severeness of negative experience] verbs that express physical sensations, for example ӴԒӴӹԊӰӹӴԂԽӸ (‘be exited’), ӫԙԽԂԽӸ (‘get tired’) [degree of fulfilment or intensity of process] verbs that express (physical) incremental changes, for example ӹӪԂԐӰӾӰӾԽӸ (‘turn into stone’) [degree of fulfilment or intensity of process].

The same search with ԌӹԙԽԂԙԽӹԓӰӹ resulted in more or less the same verb classes: (i)

(ii)

(iii)

(iv)

verbs that express activities related to knowing something or someone: ԗӹӰӠԐԂԽӸ (‘understand’), ԒӰԂԽӸ (‘know’), ԗӹԐӰӠԽӸ (‘remember’), ӠԒӫԓӠԽӸ (‘study’), ӢԂԒԼӠӢԂԽӸԙԋ (‘investigate’) [degree of quality or completeness of the action] verbs that indicate a positive attitude towards a person or thing: ԊԇԼӠԽӸ (‘love’), ӫӴԂԞԂԽӸ (‘honor’) [degree of intensity; level of positiveness of feeling] verbs that indicate changes in age: ԗӹԙԽԂӢӾԽӸ(ԙԋ) (‘getting older’), ԗӢӹԞӠԽӸ (‘live’), ӴՂӢԂԙԽӠ (‘grow’), ԙӹԒӢӾԽӸ (‘’mature’) [degree of fulfillment of process] verbs that indicate dynamic physical processes in a person, or states resulting from such processes, e.g. ӫԙԗӹӪӹӠԽӸԙԋ (‘be/become calm’) [degree of fulfillment of process].

This is not an exhaustive list of verbs that can occur with the degree words in the construction of degree, especially because sentences without subordinate clause are not taken into account. In sentences without subordinate clause and ԙԊӠԶӪӹԐ the list of verbs is much bigger, and also includes verbs like ӪӢӠԓԂԽӸ (‘cry’), ԗԂԖӰӫԽӸ (‘smell of/like’), ԖӹԽӾԽӸ/ԞӾԊԂԽӸ (‘want/desire’), ԗӹӴӾԒԽӠ (ԗӹӴӾԒԊӹ) (‘be lucky’), ӢԂԒԐӰӹԞԂԽӸԙԋ (‘multiply’), ӰԂԌӹӾԙԽӸ (‘bore’), ӴՂԙԐӾӠ-

THE CONSTRUCTION OF DEGREE IN RUSSIAN

73

ӴԂԽӸ (‘make fun of’), ӢԂԙӪӢՂӴԂԽӸ (‘uncover’), ӰԂԄӢԂԌӠԽӸ (‘reward’), ӹԼӾԙԗӾԓӠӴԂԽӸ (‘guarantee’), ԽӹӢӹԗӠԽӸԙԋ, ԗӹԙԗӾԶӠԽӸ (‘hurry up’), ԙԐӹԽӢӾԽӸ ӰԂ (‘pay attention to’), ԞԌԂԽӸ (‘wait’), ԒԂԓԂԙԽӠԽӸ (‘visit often’), and many more. In general one can say that verbs that can occur with the intensifier ӹԓӾӰӸ (‘very’) outside the construction of degree, or with forms such as ԙӹӴԙӾԐ (‘completely’) can also be combined with ԙԊӠԶӪӹԐ. Note that the class of verbs that can occur with ԌӹԙԽԂԽӹԓӰӹ is even bigger, because it also includes verbs that are not typically seen as gradable, but which can be associated with a particular quantity. To give an example, in the sentence ӵ ԌӹԙԽԂԽӹԓӰӹ ԄӫԊԋԊ ԗӹ ԠӹԙӪӴӾ, ԓԽӹԼՂ ӴԙӾ ԲԽӹ ԒԂԐӾԽӠԽӸ (‘I walked around in Moscow enough, to notice all that’), ԌӹԙԽԂԽӹԓӰӹ is combined with the non-gradable verb ԄӫԊԋԽӸ (‘walk’), and indicates that enough time has been spent on walking. A similar case can be found below: (84) ҘҏӉҧѸѹҍ ҏҡҲҝҸѹґ ҝҽӉӁҸҡ ӁҽҍҽґҧӁѹҔҏ ҏҍҽҡҲҔҽ, ѸҸґҸҦҸҮ ӉҽҲӀҔҽӁӁҸҮ ңѦӁӂѹӁѦ, ҔҸҲҸҝҽҾ ҦҸҡҲҽҲҸҍӁҸ ҰҝҸҝҽқҸҲҽґҽ Ӂҽ ҰҝҸѹӉӊҸҦҡҲӊѦ, ҍҲҸқӃ ҏҾҡӁѹҲҧ ҰҝҸҡҲҏӀ ѹҡҲѹӁҏ (…). Kuz’miǁ excoriated boss-DAT department-GEN, young-DAT timid-DAT woman, who enough worked in firm, that-IRR understand-INF simple truth ‘Kuz’miǁ excoriated the head of the department, a young timid woman, who worked long enough in the firm, to understand the simple truth.’ In the case of some gradable verbs, it is possible to use an adverb which expresses the dimension of the degree with respect to the situation expressed by the verb. In such cases it is possible to construct minimal pairs: [degree word + verb] versus [degree word + adverb + verb]. This is possible with verbs that indicate a mental activity related to an object such as ԒӰԂԽӸ ‘know’. These verbs can occur both with and without ԖӹӢӹԶӹ (‘well’). Examples are (6) and (7) given earlier. Another example is the verb ԙԽӢԂԌԂԽӸ (‘suffer’), which can occur both with and without the quantifier ԐӰӹԄӹ (‘much’) (see examples in (8) and (9) given earlier). The difference in meaning between such minimal pairs is either absent or very subtle. In the next section, I will give a detailed analysis of these cases. Besides ԖӹӢӹԶӹ or ԐӰӹԄӹ, there are also other adverbs which can occur in such minimal pairs, for example ԙӠԊӸӰӹ (sil’no ‘strong’) in phrases like ԙԊӠԶӪӹԐ (ԙӠԊӸӰӹ) ԊԇԼӠԽӸ (‘love too (strongly)’) or ԌԂԊӾӪӹ (daleko ‘far’), which may, or may not be expressed with verbs such as ӹԽӹӢӴԂԽӸԙԋ (‘break away from’): (85) ҴӁѦ ҔҽӉҽґҸҡҧ, ҍҲҸ Ҿ ҦҸҡҲҽҲҸҍӁҸ ҸҲҸҝӊҽґҡҾ, ҍҲҸқӃ ҏҦѦҝңҽҲҧҡҾ ӊҰѦҝѦҦѹ[.] (ԢӹӢԐӫԊԂ, 2001.05.15) me-DAT seemed, that I enough broke.away, that-IRR keep-INF ahead

74

EGBERT FORTUIN

‘I seemed to me, that I broke away sufficiently, to keep ahead.’ (86) ҴӁѦ ҔҽӉҽґҸҡҧ, ҍҲҸ Ҿ ҦҸҡҲҽҲҸҍӁҸ ҦҽґѦҔҸ ҸҲҸҝӊҽґҡҾ, ҍҲҸқӃ ҏҦѦҝңҽҲҧ ҡӊҸӀ ҰҸӉѹ҃ѹӀ[.] (ԣӴԽӹԼӠԒӰӾԙ – Weekly, 2001.4.5) me-DAT seemed, that I enough far broke.away, that-IRR keep-INF my position ‘I seemed to me that I broke away far enough to keep my position.’ Although the meaning of the verb ӹԽӹӢӴԂԽӸԙԋ implies that ԙԊӠԶӪӹԐ ӹԽӹӢӴԂԽӸԙԋ means that the degree concerns the distance, it is possible to make this explicit by using the adverb ԌԂԊӾӪӹ (‘far’). Other examples are the verbs ԗԂԖӰӫԽӸ (paxnut’ ‘smell like’), or ԗӹԽӾԽӸ (potet’ ‘sweat’) which can occur both with and without ԙӠԊӸӰӹ (sil’no ‘strong’). For example: (87) ҨѦҦҸҝҸӊ ҡ ҲҝҏҦҸѸ җѦґ Ҕ ӊѹҡѦґѹ҃Ѧ, ѹ ѦѸҏ ҔҽӉҽґҸҡҧ, ҍҲҸ ѹѸѦӁӁҸ ҰҸҲҸѸҏ, ҍҲҸ ҡґѹҗҔҸѸ ҡѹґҧӁҸ ҰҽѽӁѦҲ ґѹҰҽѸѹ. (ҩ. Ґҏґҧ, ԣԒӾԅ) Fedorov with difficulty went to gallows, and he-DAT seemed, that exactly because that too strong-ADV smells of.linden.trees ‘Fedorov went to the gallows with difficulty, and it seemed to him, that this was because the linden trees smelled too strong.’ (88) ѤқґҸҔѹ ѹ ҲѹѸҧҾӁ ѽҸҝҸҗѹ ӊ ҰҸѸѦӂѦӁѹҾѽ, ҊҦѦ ҔҏҝҾҲ ѹґѹ ҲҽѸ, ҊҦѦ ҡґѹҗҔҸѸ ҰҽѽӁѦҲ ѦҦҸҮ. (ԟӹԐӹӴӹӡ, 2002.12.04) apples and thyme good in rooms, where smoke, or there where too smells of.food.’ ‘You can use apples and thyme for rooms where people smoke, or where it smells too much of food.’ A similar phenomenon occurs in the case of the intensifier ӹԓӾӰӸ (‘very’). Bitextina (1975: 67) gives pairs like the following: ӹԓӾӰӸ (ԙӠԊӸӰӹ) ӪӢӠԓԂԽӸ (lit. very (hard) shout); ӹԓӾӰӸ (ӴӰӠԐԂԽӾԊӸӰӹ) ԙԊӾԌӠԽӸ ԒԂ ԙӹԼӹӡ (lit. very (carefully) look after oneself); ӹԓӾӰӸ (ӪӢӾԗӪӹ) ԌӢӫԞӠԽӸ (lit. very (strongly) be friends with). She argues that sentences without adverb are more typical of the spoken language. Which class of verbs can occur both with and without adverb, and whether there is a slight difference in meaning or style, is a matter that deserves more research. I would like to point out, though, that there are differences between verbs with respect to the use of adverbs that are difficult to predict. Take for example a verb like ԊԄԂԽӸ (lgat’ ‘lie’), which can occur without adverb: (89) ҶҲѹ ґӀҦѹ ҡґѹҗҔҸѸ ґҊҏҲ, ҍҲҸқӃ ӁѦ ӉҽґҊҽҲҧҡҾ[.] (ғ.ҭ. ҴѦӂѦҝҡҔѹҮ, ԠӹӠ ӴӹԙԗӹԐӠӰԂӰӠԋ (1897)) these people too lie-V, that-IRR not be.caught.in.a.web.of.lies-INF ‘These people lie too much not to be caught in a web of lies.’

THE CONSTRUCTION OF DEGREE IN RUSSIAN

75

With this verb, however, it is much more common to use an adverb, or to convey the meaning of quantity with ԐӰӹԄӹ (mnogo ‘much’), or the meaning of frequency with ԓԂԙԽӹ (ǁasto ‘often’), even in those cases where the speaker wants to emphasize the intensity of the situation (‘lie in a serious way’). In the next sections, I will say more about the use of the adverbs ԖӹӢӹԶӹ and ԐӰӹԄӹ in the construction of degree, and the question whether there are differences in meaning between sentences with, or without these adverbs. 3.2.2 The use of ԖӹӢӹԶӹ with gradable verbs In this section, I will discuss the use of gradable verbs that can occur both with and without the adverb ԖӹӢӹԶӹ (‘good’) if they are combined with ԙԊӠԶӪӹԐ or ԌӹԙԽԂԽӹԓӰӹ (see for example (6) or (7) given earlier). 16 These verbs belong to the class of verbs that express mental activities that are related to an object. In combinations with ԙԊӠԶӪӹԐ, the following verbs belonging to this class are attested in the NKRJa: ԒӰԂԽӸ (‘know’), ԗӹӰӠԐԂԽӸ (‘understand’), ԗӹԐӰӠԽӸ (‘remember’), ӴӠԌӾԽӸ (lit. ‘see’), and ԓӫӴԙԽӴӹӴԂԽӸ (lit. ‘feel’). The mental activity directed at the object can be seen as gradable, because it can be realized with different levels of quality, and at the same time different levels of intensity. Below, some general descriptions are given with regard to the three most frequent verbs in this class: X ԒӰԂԽӸ Y (‘know’):

possessing information with regard to Y (e.g. such that one can predict things about Y); Ռ the amount and quality of information with regard to Y

X ԗӹӰӠԐԂԽӸ Y (‘understand’): the ability to make sense of something or someone, e.g. such that one can use or interpret it (a language); such that one can provide reasons for it (someone’s behaviour), etc. Ռ the extent to which one can make sense of Y X ԗӹԐӰӠԽӸ Y (‘remember’):

16

the ability to have or bring back experiences about Y, which were experienced in the past by X;

This means that sentences like the following are not considered here: ԜӰ ԙԊӠԶӪӹԐ ԖӹӢӹԶӹ ԄӹӴӹӢӠԽ ԗӹ-ӢӫԙԙӪӠ (‘He speaks Russian too well.’) Here, the degree word modifies the adverb ɯɨɪɨɲɨ (‘well’), and the omission of the adverb would be unacceptable: ?ԜӰ ԙԊӠԶӪӹԐ ԄӹӴӹӢӠԽ ԗӹ-ӢӫԙԙӪӠ. It is, however, possible to find instances without adverb on the internet, but in these sentences ɧɟ ɫɥɢɲɤɨɦ is similar in meaning to ɧɟ ɨɱɟɧɶ (‘not very’): Ɉɧ ɧɟ ɫɥɢɲɤɨɦ/ɧɟ ɨɱɟɧɶ ɝɨɜɨɪɢɬ ɩɨ-ɪɭɫɫɤɢ.

76

EGBERT FORTUIN

Ռ the extent to which the information that X has, or brings back about Y, is thought of as being identical to the actual experience of X about Y Note that the verbs ӴӠԌӾԽӸ (‘see’) and ԓӫӴԙԽӴӹӴԂԽӸ (‘feel’) also belong to this class, when they are used to convey the meaning of ‘understanding something’: 17 (90) Ѥ ҡґѹҗҔҸѸ ӊѹңҏ, ҍҲҸ ӊӃ қҸҮҔҽҾ! (Ҩ. ѩҸҡҲҸѦӊҡҔѹҮ, ӷӢӾԙԽӫԗԊӾӰӠӾ Ӡ ӰԂӪԂԒԂӰӠӾ (1866)) I too see, that you smart ‘I see (understand) too well that you are smart!’ (91) Ѥ, ҰҸӊѦҝҧҲѦ, ҡґѹҗҔҸѸ ѽҸҝҸҗҸ ӉӁҽӀ, ҡґѹҗҔҸѸ ҍҏӊҡҲӊҏӀ; ҙҲҸ ѸҸҮ ҦҸґҊ. (Ѩ. ѷҏҝҊѦӁѦӊ, ԩԂӴԽӢԂӪ ӫ ԗӢӾԌӴӹԌӠԽӾԊԋ (1846)) I, believe-IMP, too good know, too feel-1SG-PRES, that my duty ‘I, believe me, know too well, and realize (feel) too well, that is my duty.’ If the verbs ӴӠԌӾԽӸ or ԓӫӴԙԽӴӹӴԂԽӸ are used in their literal meaning (‘see’ and ‘feel’ respectively), for example ӵ ԙӾԼԋ ӰӾ ԙԊӠԶӪӹԐ ԖӹӢӹԶӹ ԓӫӴԙԽӴӫԇ (‘I do not feel too well’), the adverb ԖӹӢӹԶӹ cannot be omitted. Note that the use of the adverb ԖӹӢӹԶӹ with these verbs is not restricted to the construction of degree, but also occurs without degree word, in sentences where it indicates a high degree or quality: (92) Ѥ ӉӁҽӀ/ҰҸѸӁӀ/ҰҸӁѹѸҽӀ ѽҸҝҸҗҸ. I know/remember/understand well ‘I know/remember/understand well.’ With these verbs, ԖӹӢӹԶӹ is not the only adverb that is used to indicate a high degree. Other adverbs that may also be used to indicate a high degree include for example boosters (ԗӢӾӪӢԂԙӰӹ ‘perfectly’, ӹԽԊӠԓӰӹ ‘excellent’, ӴӾԊӠӪӹԊӾԗӰӹ ‘splendidly’), adverbs that denote that the situation is performed with great detail (ԓӾԽӪӹ ‘neatly’, ӹԽԓӾԽԊӠӴӹ ‘distinctly’, ԋԙӰӹ ‘clearly’, ԽӹԓӰӹ ‘exactly’), adverbs that indicate that the situation is performed with regard to the whole object (ԙӹӴԙӾԐ ‘completely’, ԙӹӴӾӢԶӾӰӰӹ ‘totally’). The adverb ԖӹӢӹԶӹ, however, differs from these adverbs because it indicates the dimension of quality in its most neutral sense, without additional meaning. Note that the use of the adverb 17

It should be noted that there are probably more verbs that belong to this class, but which are not attested in the NKRJa. An example is the verb ӠԒӫԓԂԽӸ (izuǁat’ ‘study’), which is attested both with and without ԖӹӢӹԶӹ on the internet.

THE CONSTRUCTION OF DEGREE IN RUSSIAN

77

ԖӹӢӹԶӹ with regard to verbs like ‘know’, ‘remember’, ‘understand’, etc. is not a typical trait of Russian, but occurs in other European languages as well, for example in English (well), Dutch (goed), and French (bien). There are, however, also differences between the function of forms like ԖӹӢӹԶӹ in these languages. An important difference between Russian and French on the one hand, and Dutch and English on the other is that in Dutch and English, the use of the adverb goed (‘well’) is obligatory in contexts with te or too respectively and verbs like ‘know’, ‘understand’, etc., whereas this is not the case in Russian and French. 18 Table 1 shows the division between [ԙԊӠԶӪӹԐ + V] and [ԙԊӠԶӪӹԐ ԖӹӢӹԶӹ + V] in the NKRJa, both in the 19th and 20th century. Because the number of attested instances is so small, I have given the number of occurrences. 19 Note that in the second part of the nineteenth century, most instances without ԖӹӢӹԶӹ are attested with just one author, namely Dostoevskij. With all verbs, Dostoevskij is inclined not to use ԖӹӢӹԶӹ, whereas most of his contemporaries prefer ԖӹӢӹԶӹ. It may be that with respect to this construction, the speech of Dostoevskij is to some extent idiosyncratic, or part of a trend, confined to a smaller number of people, which started in the second half of the nineteenth century with some writers, and ended somewhere in the first part of the twentieth century. Note, furthermore, that in the nineteenth century, Dostoevskij is also attested most frequently with the construction of degree as such (with and without ԖӹӢӹԶӹ), whereas the NKRJa also contains many texts by his contemporaries, for example Tolstoj. Table 2 provides insight into the

18 Another important difference is that in both Russian and French, ԖӹӢӹԶӹ or bien always indicates a high degree, whereas in English or Dutch, the forms well or goed can indicate the dimension of quality as such, without necessarily expressing a high degree. Because of this, well or goed is used in questions such as the following in both Dutch and English: Hoe goed spreek je Russisch? (lit. ‘How well do you speak Russian?’). In Russian and French, the adverb ԖӹӢӹԶӹ or bien is not used in such contexts: ԹԂӪ ӴՂ ԄӹӴӹӢӠԽӾ ԗӹ-ӢӫԙԙӪӠ?; Comment vous parlez le français? 19 Unfortunately, the NKRJa does not indicate the number of words, but only information with respect to the number of documents: 924 before 1849; 719 from 1850 till 1899; 1080 from 1900 till 1949, and 38000 from 1950. The oldest text in the corpus is ѩ.Ѩ. ҨҸӁӊѹӉѹӁ, ѩӊҽ ѸҽҝҔѹҲҽӁҲҽ (1788). Note that there are also texts which start in one period, and end in another. NKRJa classifies such texts according to the starting date. – I have looked for [ԙԊӠԶӪӹԐ * ԖӹӢӹԶӹ * ԒӰԂԽӸ], [ԒӰԂԽӸ * ԙԊӠԶӪӹԐ * ԖӹӢӹԶӹ], [ԙԊӠԶӪӹԐ * ԒӰԂԽӸ] and [ԒӰԂԽӸ * ԙԊӠԶӪӹԐ] (* indicates possible words between the forms). This search also includes adverbial participles and participles. I have excluded sentences like the following, where the syntactic structure could not be established: ԜӰ, ӴԗӢӹԓӾԐ, ԒԂԐӾԽӠԊ, ԓԽӹ ԾԂԙԽԂԙӸԋ ԢӠԊӠԗԗӹӴӰԂ ԙԊӠԶӪӹԐ ԖӹӢӹԶӹ ԒӰԂԊԂ Ӡ ԗӹӰӠԐԂԊԂ, ԓԽӹ ԒӰԂԓӠԊԂ ԌԊԋ ӰӾԄӹ ԣԄԊԂԋ. (Ҩ. ѩҸҡҲҸѦӊҡҔѹҮ, ԚԌӠӹԽ (1869)). (lit. too good knew and understood).

78

EGBERT FORTUIN

percentages for the most frequent verbs of table 1, both with and without Dostoevskij. If we look only at the second half of the twentieth century, we can conclude that with both ԒӰԂԽӸ and ԗӹԐӰӠԽӸ, the use of ԖӹӢӹԶӹ is much more common than without this adverb. The verb ԗӹӰӠԐԂԽӸ, however, occurs relatively frequently without ԖӹӢӹԶӹ. What about diachronic trends? The data show that ԖӹӢӹԶӹ was preferred with ԒӰԂԽӸ already in the beginning of the nineteenth century. In general, the use of ԖӹӢӹԶӹ with this verb seems relatively stable if we look at the development in the nineteenth and twentieth century. In the second half of the nineteenth century (with Dostoevskij), and the first half of the twentieth century, there was a slight decrease of ԖӹӢӹԶӹ, but this trend did not continue. It is difficult to draw any conclusions about the verbs ԗӹӰӠԐԂԽӸ and ԗӹԐӰӠԽӸ, due to the small number of instances. However, the data imply that with both verbs there may have been a small trend in particular parts of the population not to use ԖӹӢӹԶӹ at the end of the nineteenth century and the beginning of the twentieth century (with ԗӹԐӰӠԽӸ), or only at the end of the nineteenth century (with ԗӹӰӠԐԂԽӸ).

79

THE CONSTRUCTION OF DEGREE IN RUSSIAN

17881849

1850-1899

19001949

1950-2007

ԙԊӠԶӪӹԐ ԒӰԂԽӸ

1

5

4

ԙԊӠԶӪӹԐ ԖӹӢӹԶӹ ԒӰԂԽӸ

17

23

102

ԙԊӠԶӪӹԐ ԗӹӰӠԐԂԽӸ

0

2

11

ԙԊӠԶӪӹԐ ԖӹӢӹԶӹ ԗӹӰӠԐԂԽӸ

2

8

19

ԙԊӠԶӪӹԐ ԗӹԐӰӠԽӸ

0

2

2

ԙԊӠԶӪӹԐ ԖӹӢӹԶӹ ԗӹԐӰӠԽӸ

0

21 (18 from Dostoevskij) 41 (7 from Dostoevskij) 23 (23 from Dostoevskij) 10 (8 from Dostoevskij) 5 (5 from Dostoevskij) 6 (2 from Dostoevskij) 3 (3 from Dostoevskij) 2

2

24

1

0

1

4

6 (5 from evskij) 0

0

0

ԙԊӠԶӪӹԐ ӴԂԽӸ

ԓӫӴԙԽӴӹ- 2

ԙԊӠԶӪӹԐ ԖӹӢӹԶӹ ԓӫӴԙԽӴӹӴԂԽӸ ԙԊӠԶӪӹԐ ӴӠԌӾԽӸ ԙԊӠԶӪӹԐ ԖӹӢӹԶӹ ӴӠԌӾԽӸ

0 0

0

Dosto1 (from 1 (literal ‘to 1900) see’)

Table 1 Number of instances with [ԙԊӠԶӪӹԐ (+ ԖӹӢӹԶӹ) + verb]

80

EGBERT FORTUIN

ԒӰԂԽӸ %ѽҸҝ. ԒӰԂԽӸ total ԗӹӰӠԐԂԽӸ %ѽҸҝ. ԗӹӰӠԐԂԽӸ total ԗӹԐӰӠԽӸ %ѽҸҝ. ԗӹԐӰӠԽӸ total

17881849

1850-1899 with Dost.

1850-1899 1900without Dost. 1949

19502007

94% 18 100%

66% 62 30%

92% 37 100%

82% 28 80%

96% 106 63%

2 0

33 55% 11

2 100% 4

10 50% 4

30 92% 26

Table 2: Percentage of instances with [ԙԊӠԶӪӹԐ (+ ԖӹӢӹԶӹ) + verb]

With respect to the data given above, the following questions can be asked: (i) (ii)

(iii)

Is there a subtle difference in meaning between sentences with and sentences without ԖӹӢӹԶӹ? How can one account for the fact that in the second half of the century, there is a trend with some writers not to use ԖӹӢӹԶӹ with ԒӰԂԽӸ, ԗӹӰӠԐԂԽӸ and ԗӹԐӰӠԽӸ? How can one account for the differences in the use of ԖӹӢӹԶӹ with respect to different verbs?

Firstly, with regard to the first question: there is no clear indication that there is a difference in meaning between sentences with and sentences without ԖӹӢӹԶӹ. This means that ԖӹӢӹԶӹ makes the dimension of quality explicit, which is already part of the meaning (or rather interpretation) of the construction itself. The use or non use of ԖӹӢӹԶӹ is therefore merely a stylistic phenomenon. A possible exception has to be made with regard to (pseudo) past passive participles. With these forms the use or non use of ԖӹӢӹԶӹ may be connected to the question whether the form is primarily associated with verbal semantics or not. This phenomenon does not occur with the verbs ԒӰԂԽӸ, ԗӹԐӰӠԽӸ and ԓӫӴԙԽӴӹӴԂԽӸ, but it does with the verb ӴӠԌӾԽӸ (‘see’), when it refers to the literal action of seeing. Consider for example the use of the adjective (pseudo imperfective past participle) ӴӠԌӾӰ, in (93) without ԖӹӢӹԶӹ, and in (94) with ԖӹӢӹԶӹ: (93) ѣҡґѹ ҸӁ ҍҏӊҡҲӊҸӊҽґ, ҍҲҸ ҡґѹҗҔҸѸ ӊѹҦѦӁ ҡҸ ӊҡѦѽ ҡҲҸҝҸӁ, ҸӁ ҦӁѦѸ ҸҲґѦңѹӊҽґҡҾ, ҽ ӁҸҍҧӀ җѦґ. (ү. ҨҽҦѦѦӊ, ԠӹԊӹԌԂԋ ԄӴԂӢԌӠԋ) if he felt, that too visible-ADJ/see-PARTICPLE-PAST-PASSIVE-IMPERF from all sides, have.a.thorough.rest, and at.night went ‘If he felt that he was too visible from every side, he had a good rest during the day, and traveled at night.’

THE CONSTRUCTION OF DEGREE IN RUSSIAN

81

(94) ҫӁ қӃґ ҡґѹҗҔҸѸ ѽҸҝҸҗҸ ӊѹҦѦӁ Ӂҽ ӍҸӁѦ ӁѦқҽ (…). (www.svoboda.org/programs/ex/2004/ex.012504.asp) He was too good visible-ADJ/see-PARTICIPLE-PAST-PASSIVEIMPERF on background sky-GEN ‘He was too visible against the sky.’ Because this form is associated both with verbal and adjectival semantics, and because the adverb ԖӹӢӹԶӹ is not used with ԙԊӠԶӪӹԐ in combination with regular adjectives, one may expect some variation between speakers with regard to the use of ԖӹӢӹԶӹ. The same phenomenon occurs in the case of the past passive participle ӴӹԙԗӠԽԂӰ (vospitan ‘educated/raised’, ‘well mannered’): (95) ѢҸӊӃҮ ѩңѦҮѸҡ ҷҸӁҦ ӁѦ ҡґѹҗҔҸѸ ӊҸҡҰѹҲҽӁ (…). (search.vedomosti.ru/newspaper/article.shtml?2006/11/21/116119) new James Bond not too well-mannered-ADJ/raise-PARTICPLE-PASTPASSIVE-PERF ‘The new James Bond is not too/very well mannered.’ (96) ѢҸ ѢҽқҸҔҸӊ қӃґ ҡґѹҗҔҸѸ ѽҸҝҸҗҸ ӊҸҡҰѹҲҽӁ, ҍҲҸқӃ ҡҝҽӊӁѹӊҽҲҧ ңѦӁӂѹӁҏ ҡ ґҸҗҽҦҧӀ. (magazines.russ.ru/novyi_mi/1997/4/obz101-p.html) but Nabokov was too well raise-PARTICPLE-PAST-PASSIVE-PERFSHORT-MASC-SG, that-IRR compare-INF woman with horse ‘But Nabokov was too well-bred to compare a woman to a horse’ Secondly, the above mentioned facts suggest that in both centuries, but especially in the nineteenth century, there is some variation within the population with respect to the use of adverbs: some authors always use ԖӹӢӹԶӹ, whereas others never do. Some authors may prefer ԖӹӢӹԶӹ only with some verbs. This may for example be an explanation for the use of ԖӹӢӹԶӹ in (91) by Turgenev: ԙԊӠԶӪӹԐ ԖӹӢӹԶӹ ԒӰԂԇ, ԙԊӠԶӪӹԐ ԓӫӴԙԽӴӫԇ (too good know, too feel; ‘know too well, and feel too well’). 20 Furthermore, there may be authors that sometimes will, and sometimes will not use ԖӹӢӹԶӹ with the same verb. Dostoevskij is for example attested with 7 instances of ԙԊӠԶӪӹԐ ԖӹӢӹԶӹ ԒӰԂԽӸ, and with 18 instances of ԙԊӠԶӪӹԐ ԒӰԂԽӸ. The data do not suggest that the use of ԖӹӢӹԶӹ or leaving it out with this author is connected to the different norms that Dostoevskij wants to express. The following sentences are both taken from the same source (his diaries from 1877):

20

Alternatively, one may argue that ԖӹӢӹԶӹ is not used with ԓӫӴԙԽӴӹӴԂԽӸ because it is already expressed with ԒӰԂԽӸ (suggesting an ellipsis).

82

EGBERT FORTUIN

(97) … ѹ һѦӊѹӁ ҡґѹҗҔҸѸ ӉӁҽѦҲ, ҍҲҸ ҸӁ ҊҸӊҸҝѹҲ Ҹ ҡґҽӊҾӁҽѽ. (Ҩ. ѩҸҡҲҸѦӊҡҔѹҮ, ԟӰӾӴӰӠӪ ԗӠԙԂԽӾԊԋ. 1877. ҐҸҦ II-Ү (1877)) …and Levin too knows, that he talks about Slavs ‘…and Levin knows too well, that he talks about the Slavs.’ (98) ҩҽҮҾ ҡґѹҗҔҸѸ ѽҸҝҸҗҸ ӉӁҽѦҲ, ҍҲҸ ҲҸґҧҔҸ ҙҲѹѽ ҍҏӊҡҲӊ ҸӁҽ ѹ ѸҸңѦҲ ҸңѹҦҽҲҧ ҲѦҰѦҝҧ ҸҲ ѣӊҝҸҰӃ. (Ҩ. ѩҸҡҲҸѦӊҡҔѹҮ, ԟӰӾӴӰӠӪ ԗӠԙԂԽӾԊԋ. 1877. ҐҸҦ II-Ү (1877)) Raja too well knows, that only these feelings she and can expect now from Europe ‘Raja knows too well, that she may only expect these feelings from Europe right now.’ ү possible explanation for the small trend not to use ԖӹӢӹԶӹ in the second half of the twentieth century could be the influence of French. In French, the construction of degree occurs both without and with bien with verbs like connaître (‘know’), savoir, se souvenir, and comprendre: (99) Je te connais trop (bien) pour ça. French differs from Russian, because with trop the use of bien is less frequent than cases without adverb. In the corpus Frantext, in the second half of the 19th century, about 40% of the instances with trop and connaître (72 randomly selected sentences) occurred with bien. In the case of the verb savoir the percentage of bien was even lower. About 90% of the 81 randomly selected instances did not contain bien. Note that of these, 90% occurred with negation (Je ne sais trop). A writer like Dostoevskij, or other writers in the second half of the nineteenth century, may well have been influenced by French. Evidence for this is the fact that in his work, Dostoevskij actually uses the French construction of degree with comprendre without bien: (100) Enfin, je n’ai pas trop compris. ѢҸ Ҿ ѹѽ ҰѦҝѦѽѹҲҝѹґ, ѹ ҔҽҔҸѦ ѸӁѦ ҦѦґҸ ҦҸ ѩѸѹҲҝѹҾ ҴѹҲҝѹҍҽ. (Ҩ. ѩҸҡҲҸѦӊҡҔѹҮ, ԮӾԙՂ (1871–1872)) ‘Enfin, je n’ai pas trop compris. But I outwitted them, and what do I have to do with Dmitrij Mitriǁ.’ One should of course be careful here to jump to conclusions, because there are also indications which could be interpreted as counterarguments. An important counterargument is that writers like Tolstoj or Turgenev, who spoke French as well, prefer the use with ԖӹӢӹԶӹ. Note for example that Turgenev is attested 6 times with ԙԊӠԶӪӹԐ ԖӹӢӹԶӹ ԒӰԂԽӸ, and zero times with ԙԊӠԶӪӹԐ ԒӰԂԽӸ. Thirdly, there is a difference between ԒӰԂԽӸ on the one hand and ԗӹӰӠԐԂԽӸ on the other. It must be noted that the reason why in the second part of the twentieth century a relatively high number of instances of ԗӹӰӠԐԂԽӸ occurs

83

THE CONSTRUCTION OF DEGREE IN RUSSIAN

without ԖӹӢӹԶӹ is correlated with the use of negation. More specifically, the absence of ԖӹӢӹԶӹ can relatively frequently be found with the expression ӰӾ ԙԊӠԶӪӹԐ ԗӹӰӠԐԂԽӸ. In this use, ԙԊӠԶӪӹԐ behaves like an intensifier, and is semantically similar to ӹԓӾӰӸ (ӰӾ ӹԓӾӰӸ ԗӹӰӠԐԂԇ). Table 3 shows that if we compare sentences with and without ԖӹӢӹԶӹ with ԙԊӠԶӪӹԐ + ԗӹӰӠԐԂԽӸ in the twentieth century, there is indeed a negative correlation between the use of negation and the use of ԖӹӢӹԶӹ (0,41), and consequently a positive correlation between negation and sentences without ԖӹӢӹԶӹ (-0,41). According to the Chisquare test, this correlation is significant (ʖ² = 6,7; with df=1, this result has a 99% reliability). However, because of the small sample size, the Chi-square test is less reliable. The Fisher exact test, which takes the small sample size into account, shows that the correlation is significant indeed (observed value = 4, expected value = 7.43; Fisher exact two-tailed: p=0.02, which means that the reliability is 98%). negation Number of cases ԙԊӠԶӪӹԐ 7 ԗӹӰӠԐԂԽӸ Number of cases ԙԊӠԶӪӹԐ 4 ԖӹӢӹԶӹ ԗӹӰӠԐԂԽӸ Number 11

no negation 6

number 13

23

27

29

40

cor cases ԙԊӠԶӪӹԐ ԗӹӰӠ- 0,41 ԐԂԽӸ cor cases ԙԊӠԶӪӹԐ ԖӹӢӹԶӹ -0,41 ԗӹӰӠԐԂԽӸ Table 3: correlation ԗӹӰӠԐԂԽӸ and negation (twentieth century)

These data suggest that when ԙԊӠԶӪӹԐ is used as an intensifier, the use of ԖӹӢӹԶӹ is less typical. In order to see if there is indeed a relation, let us have a look at the use of ӹԓӾӰӸ. With this intensifier we find both uses with and without ԖӹӢӹԶӹ: (101) ѯѦҡҲӁҸ ҊҸӊҸҝҾ, ѸӃ ӁѦ ҸҍѦӁҧ ӉӁҽѦѸ, ҰҝҸ ҍҲҸ ҔҸӁҔҝѦҲӁҸ ҲҏҲ ҊҸӊҸҝѹҲҧ. (ԜӪԽԋԼӢӸ, 2003) honestly, speaking, we not very know, about what specifically here speakINF ‘To be honest, we do not know very well about what exactly we have to speak.’ (102) ҫҲӊѦҲѹҲҧ Ӂҽ ҙҲҸҲ ӊҸҰҝҸҡ ҸӁ ӁѦ ѸҸҊ, ҰҸҲҸѸҏ ҍҲҸ ѦӂѦ ӁѦ ҸҍѦӁҧ ѽҸҝҸҗҸ ӉӁҽґ ѦѦ. (ԁӢӫԌ, 1989.01.01)

84

EGBERT FORTUIN

answer-INF that question he not could, because he not.yet not very well knew her ‘He couldn’t answer that question, because he didn’t know her very well yet.’ Table 4 lists the use of ӹԓӾӰӸ with or without ԖӹӢӹԶӹ in the first and second half of the nineteenth and twentieth century for the verbs ԒӰԂԽӸ, ԗӹӰӠԐԂԽӸ, and ԗӹԐӰӠԽӸ.

ӹԓӾӰӸ ԒӰԂԽӸ ӹԓӾӰӸ ԖӹӢӹԶӹ ԒӰԂԽӸ ӹԓӾӰӸ ԗӹӰӠԐԂԽӸ ӹԓӾӰӸ ԖӹӢӹԶӹ ԗӹӰӠԐԂԽӸ ӹԓӾӰӸ ԗӹԐӰӠԽӸ ӹԓӾӰӸ ԖӹӢӹԶӹ ԗӹԐӰӠԽӸ

< 1850

1850-1900

1900-1950

t 1950

29 37 (56%) 16 7 (30%)

21 190 (90%) 28 21 109 (93%) 22

5 18 (78%) 19 32 (60%)

2 74 (97%) 191 119 (37%)

18 8 (31%)

26 42 (62%)

6 18 (75%)

13 117 (90%)

Table 4: the use of ԖӹӢӹԶӹ with ӹԓӾӰӸ

Table 4 shows that in modern Russian (second half of the twentieth century), the use of ԖӹӢӹԶӹ with ӹԓӾӰӸ is to a great extent similar to the use with ԙԊӠԶӪӹԐ. With the verbs ԒӰԂԽӸ and ԗӹԐӰӠԽӸ the use of ԖӹӢӹԶӹ is by far the most frequent, whereas with ԗӹӰӠԐԂԽӸ the majority of the uses occur without ԖӹӢӹԶӹ. 23 This means that there is no general negative correlation between the function of intensifier and the use of ԖӹӢӹԶӹ. However, like in the case of ԙԊӠԶӪӹԐ, the use of ԖӹӢӹԶӹ with ԗӹӰӠԐԂԽӸ is influenced by the presence of negation. Table 5 shows that with ԗӹӰӠԐԂԽӸ, there is a negative correlation of 0,69 between the presence of negation and the use of ԖӹӢӹԶӹ. This correlation is significant (ʖ² =176; whereas with df=1, and a reliability of 95%, the ʖ² should be above 3,84). 21

10 are from Dostoevskij. 18 are from Dostoevskij. 23 With respect to diachronic tendencies, table 4 suggests that there are differences between verbs. Both with ԒӰԂԽӸ and ԗӹԐӰӠԽӸ there is an increase in the use of ԖӹӢӹԶӹ. In the case of ԗӹӰӠԐԂԽӸ, however, the use of ԖӹӢӹԶӹ first increased till the second half of the nineteenth century, and then decreased again. In modern Russian most instances of ԗӹӰӠԐԂԽӸ occur without ԖӹӢӹԶӹ. The trend to not use ԖӹӢӹԶӹ which occurred with ԙԊӠԶӪӹԐ at the end of the nineteenth century cannot be established for ӹԓӾӰӸ. It should be stressed, though, that a statistical analysis is necessary to see whether these trends are statistically significant. 22

85

THE CONSTRUCTION OF DEGREE IN RUSSIAN

Number of cases ӹԓӾӰӸ ԗӹӰӠԐԂԽӸ Number of cases ӹԓӾӰӸ ԖӹӢӹԶӹ ԗӹӰӠԐԂԽӸ Number cor cases ӹԓӾӰӸ ԗӹӰӠԐԂԽӸ cor cases ӹԓӾӰӸ ԖӹӢӹԶӹ ԗӹӰӠԐԂԽӸ

negation 164

no negation 46

number 210

12

139

151

176

193

369

0,69 -0,69

Table 5: correlation between negation and the use of ԖӹӢӹԶӹ with ӹԓӾӰӸ and ԗӹӰӠԐԂԽӸ

It is possible that the negative correlation between ӰӾ ӹԓӾӰӸ/ӰӾ ԙԊӠԶӪӹԐ X (‘not very X’) and the use of ԖӹӢӹԶӹ can be explained in terms of the negative meanings of these expressions. More specifically, they both express that the degree, or rather the quality of X, is rather low. It is possible that the use of the adverb ԖӹӢӹԶӹ easily triggers an absolute reading, where the quality of X is perceived as high. Whether or not this analysis is correct, is a topic that deserves a separate study. I will now turn to sentences with ԌӹԙԽԂԽӹԓӰӹ (or ԌӹӴӹԊӸӰҸ). Like ԙԊӠԶӪӹԐ, the degree word ԌӹԙԽԂԽӹԓӰӹ can occur both with and without ԖӹӢӹԶӹ with the verbs ԒӰԂԽӸ (‘know’), ԗӹӰӠԐԂԽӸ (‘understand’), and ԗӹԐӰӠԽӸ (‘remember’). In addition, in the NKRJa minimal pairs are also attested with the verbs ӴԊԂԌӾԽӸ (‘have a command of’), ӠԒӫԓԂԽӸ (‘study’), and ӢԂԒԼӠӢԂԽӸԙԋ (‘investigate’): (103) ҭҝҽӊҦҽ, үґѦҔҡҽӁҦҝ ғҸґҧӍ ѸҸҊ қӃҲҧ ѸҸѹѸ ҡҸҸҲѦҍѦҡҲӊѦӁӁѹҔҸѸ ѹ ҦҸҡҲҽҲҸҍӁҸ ѽҸҝҸҗҸ ӊґҽҦѦҲҧ ҽӁҊґѹҮҡҔѹѸ ҾӉӃҔҸѸ, ҍҲҸқӃ ӁѦ ҰҝѹқѦҊҽҲҧ Ҕ ҰҸѸҸӂѹ ҰѦҝѦӊҸҦҍѹҔҽ (…). (Ґ. ҐҽӉҦҽӁҸӊ, ӷӢӠԒӢԂӪ ԣԊӾӪԙԂӰԌӢԂ ԛӹԊӸԅԂ). true, Aleksandr Vol’f could be-INF my compatriot and enough well command-INF English language, that-IRR not run for help interpreterGEN ‘True, Aleksandr Vol’f could be a compatriot of mine and have a sufficient command of English not to run to the interpreter for help.’

86

EGBERT FORTUIN

(104) Ҙ ҲҸѸҏ ӊҝѦѸѦӁѹ Ҿ ҏңѦ ҦҸҡҲҽҲҸҍӁҸ ӊґҽҦѦґ ѸҸӁҊҸґҧҡҔѹѸ ҾӉӃҔҸѸ, ҍҲҸқӃ ѦѦ ҰҸӁҾҲҧ (һ.ҌӉѦӍҸӊѹҍ, ԹӰԋԒӸ ӴӾԽӢԂ) by that time I already enough commanded Mongolian language, that-IRR her understand ‘By then, I had a sufficient command of Mongolian, to understand her.’ (105) ѤҔҏҗѦӊ ҏңѦ ҦҸҡҲҽҲҸҍӁҸ ѽҸҝҸҗҸ ҝҽӉқѹҝҽґҡҾ ӊ ҽқѽҽӉҡҔѹѽ ѹ ҊҝҏӉѹӁҡҔѹѽ ҦѦґҽѽ, ҍҲҸқӃ ҰҸӁҾҲҧ, ҍҲҸ ӉҽҲѦҾӁӁҸѦ ҘҸӁҦҝҽҲҧѦӊӃѸ «ӊҸӉӊҝҽӂѦӁѹѦ қѦңѦӁ҃Ѧӊ» ҸӉӁҽҍҽѦҲ ӁҸӊҏӀ қҸґҧҗҏӀ ӊҸҮӁҏ. (forums.ng.ru/archive/part/www.ng.ru,cis,2007-06-29,100_gruzoset/116) Jakušev already enough well investigated in Abchaz and Georgian matters, that-IRR understand-INF, that undertaken Kondrat’ev-INSTR ‘‘re turn refugees-GEN’’ means new big war. ‘Jakušev investigated the Abchaz and Georgian situation well enough to understand that the decision taken by Kondrat’ev means the beginning of a new war.’ (106) ѷҽҲҽҝҡҔѹҮ ҏңѦ ҦҸҡҲҽҲҸҍӁҸ ҝҽӉқѹҝҽґҡҾ ӊ қѹӉӁѦҡѦ, ҍҲҸқӃ ҰҸӁҾҲҧ, ҍҲҸ ѸҽҗѹӁҽ ӁҽӁҾҲҽ ҍҽҡҽ Ӂҽ Ҧӊҽ (ғ. ҭѦґѦӊѹӁ, Generation "ӷ"). Tatarskij already enough investigated in business, that-IRR understandINF, that car rented hour for two ‘Tatarskij already investigated the business enough to understand that the car was rented for two hours.’ It should be noted that ԌӹԙԽԂԽӹԓӰӹ (and ԌӹӴӹԊӸӰӹ) occurs without subordinate clause as well, and in many cases the degree word is best interpreted as a moderator. Furthermore, the line between the relative and the moderator use of ԌӹԙԽԂԽӹԓӰӹ is not always easy to draw. This is the case in the following sentence: (107) Ѥ ҡҝҽӉҏ ңѦ ҰҝѹҡҲҏҰѹґ Ҕ ҝҽқҸҲѦ, ҽ ҰҸҡҔҸґҧҔҏ ҲѦҔҡҲ қӃґ ҦҸӊҸґҧӁҸ қҸґҧҗҸҮ, ҽ ӊҝѦѸѦӁѹ ѸҽґҸ, ҰҝѹҊґҽҡѹґ Ӂҽ ҰҸѸҸӂҧ ҡҸҲҝҏҦӁѹҔҽ ҴѹӁѹҡҲѦҝҡҲӊҽ ѹӁҸҡҲҝҽӁӁӃѽ ҦѦґ, ҦҸҡҲҽҲҸҍӁҸ ѽҸҝҸҗҸ ӉӁҽӊҗѦҊҸ ҽӁҊґѹҮҡҔѹҮ ҾӉӃҔ. (ҫ. ѷҝҸҾӁҸӊҡҔѹҮ, өӾӢӾԒ ԄӹԌՂ Ӡ ӢԂԙԙԽӹԋӰӠԋ) …, enough well know-PARTICIPLE-ACTIVE-PAST-IMPERF-SG-GEN English language ‘I immediately started my job, but as the text was rather big, and the time limited, I invited an official of the Ministry of foreign affairs, who knew English rather well/well enough.’ Table 6 shows the number of instances with and without ԖӹӢӹԶӹ with ԌӹԙԽԂԽӹԓӰӹ and ԌӹӴӹԊӸӰӹ in the 19th and 20th century. Between brackets the number of instances with a subordinate clause with ԓԽӹԼՂ is indicated.

THE CONSTRUCTION OF DEGREE IN RUSSIAN

ԌӹԙԽԂԽӹԓӰӹ/ԌӹӴӹԊӸӰӹ ԒӰԂԽӸ ԌӹԙԽԂԽӹԓӰӹ/ԌӹӴӹԊӸӰӹ ԖӹӢӹԶӹ ԒӰԂԽӸ ԌӹԙԽԂԽӹԓӰӹ/ԌӹӴӹԊӸӰӹ ԗӹӰӠԐԂԽӸ ԌӹԙԽԂԽӹԓӰӹ/ԌӹӴӹԊӸӰӹ ԖӹӢӹԶӹ ԗӹӰӠԐԂԽӸ ԌӹԙԽԂԽӹԓӰӹ/ԌӹӴӹԊӸӰӹ ԗӹԐӰӠԽӸ ԌӹԙԽԂԽӹԓӰӹ/ԌӹӴӹԊӸӰӹ ԖӹӢӹԶӹ ԗӹԐӰӠԽӸ ԌӹԙԽԂԽӹԓӰӹ/ԌӹӴӹԊӸӰӹ ӴԊԂԌӾԽӸ ԌӹԙԽԂԽӹԓӰӹ/ԌӹӴӹԊӸӰӹ ԖӹӢӹԶӹ ӴԊԂԌӾԽӸ ԌӹԙԽԂԽӹԓӰӹ/ԌӹӴӹԊӸӰӹ ӢԂԒԼӠӢԂԽӸԙԋ ԌӹԙԽԂԽӹԓӰӹ/ԌӹӴӹԊӸӰӹ ԖӹӢӹԶӹ ӢԂԒԼӠӢԂԽӸԙԋ ԌӹԙԽԂԽӹԓӰӹ/ԌӹӴӹԊӸӰӹ ӠԒӫԓԂԽӸ ԌӹԙԽԂԽӹԓӰӹ/ԌӹӴӹԊӸӰӹ ԖӹӢӹԶӹ ӠԒӫԓԂԽӸ

19th century 29 ԌӹӴӹԊӸӰӹ (1 ԓԽ.); 9 ԌӹԙԽԂԽӹԓӰӹ (2 ԓԽ.) 8 ԌӹӴӹԊӸӰӹ (0 ԓԽ.); 0 ԌӹԙԽԂԽӹԓӰӹ 4 ԌӹӴӹԊӸӰӹ (0 ԓԽ.); 7 ԌӹԙԽԂԽӹԓӰӹ (0 ԓԽ.) 2 ԌӹӴӹԊӸӰӹ (0 ԓԽ.); 0 ԌӹԙԽԂԽӹԓӰӹ

20th century 7 ԌӹӴӹԊӸӰӹ (0 ԓԽ.); 37 ԌӹԙԽԂԽӹԓӰӹ (12 ԓԽ.) 13 ԌӹӴӹԊӸӰӹ (0 ԓԽ.); 36 ԌӹԙԽԂԽӹԓӰӹ (4 ԓԽ.) 0 ԌӹӴӹԊӸӰӹ; 3 ԌӹԙԽԂԽӹԓӰӹ (0 ԓԽ.) 3 ԌӹӴӹԊӸӰӹ (0 ԓԽ.); 9 ԌӹԙԽԂԽӹԓӰӹ (1 ԓԽ.)

1 ԌӹӴӹԊӸӰӹ (0 ԓԽӹԼՂ); 0 ԌӹԙԽԂԽӹԓӰӹ 1 ԌӹԙԽԂԽӹԓӰӹ (0 ԓԽ.); 0 ԌӹӴӹԊӸӰӹ 1 ԌӹԙԽӹԽӹԓӰӹ (0 ԓԽ.); 0 ԌӹӴӹԊӸӰӹ -

-

1 ԌӹӴӹԊӸӰӹ (0 ԓԽ.); 3 ԌӹԙԽԂԽӹԓӰӹ (0 ԓԽ.) -

4 ԌӹӴӹԊӸӰӹ(0 ԓԽ.); 1 ԌӹԙԽԂԽӹԓӰӹ (0 ԓԽ.) 0 ԌӹӴӹԊӸӰӹ; 4 ԌӹԙԽԂԽӹԓӰӹ (1 ԓԽ.) 1 ԌӹӴӹԊӸӰӹ (0 ԓԽӹԼՂ); 2 ԌӹԙԽԂԽӹԓӰӹ (2 ԓԽ.) 0 ԌӹӴӹԊӸӰӹ; 1 ԌӹԙԽԂԽӹԓӰӹ (1 ԓԽ.) 2 ԌӹӴӹԊӸӰӹ (0 ԓԽ.); 2 ԌӹԙԽԂԽӹԓӰӹ (0 ԓԽ.) 0 ԌӹӴӹԊӸӰӹ; 11 ԌӹԙԽԂԽӹԓӰӹ (2 ԓԽ.) 4 ԌӹӴӹԊӸӰӹ; 21 ԌӹԙԽԂԽӹԓӰӹ (0 ԓԽ.)

Table 6 Number of instances with [ԌӹԙԽԂԽӹԓӰӹ/ԌӹӴӹԊӸӰӹ (+ ԖӹӢӹԶӹ) + verb]

87

88

EGBERT FORTUIN

Table 7 lists the percentages from table 6 for the most frequent verbs. ԒӰԂԽӸ %ѽҸҝ. ԒӰԂԽӸ total ԗӹӰӠԐԂԽӸ %ѽҸҝ. ԗӹӰӠԐԂԽӸ total ԗӹԐӰӠԽӸ %ѽҸҝ ԗӹԐӰӠԽӸ total ӠԒӫԓԂԽӸ%ѽҸҝ. ӠԒӫԓԂԽӸ total

19th century 17% 46 13% 16 50% 2 0% 4

20th century 53% 93 80% 15 100% 5 70% 36

Table 7 Percentage of instances with [ԙԊӠԶӪӹԐ (+ ԖӹӢӹԶӹ) + verb]

Note that in the twentieth century with ԒӰԂԽӸ the use of ԖӹӢӹԶӹ is more typical for ԙԊӠԶӪӹԐ (see table 2), than for ԌӹԙԽԂԽӹԓӰӹ. With respect to the other verbs one should be careful to draw any conclusions, in view of the small numbers. A possible explanation is that the meaning of ԌӹԙԽԂԽӹԓӰӹ itself is more easily associated with the idea of an adequate quality than ԙԊӠԶӪӹԐ, because of which is does not have to be expressed by a separate form. 24 Furthermore, another factor may be that ԌӹԙԽԂԽӹԓӰӹ has a more independent syntactic status than ԙԊӠԶӪӹԐ. To give an example, ԌӹԙԽԂԽӹԓӰӹ can occur on its own in sentences where it functions as predicate or object, whereas ԙԊӠԶӪӹԐ always needs another constituent on which it is dependent. It should be noted, though, that in the twentieth century there are differences between verbs. More specifically, the verbs ԗӹԐӰӠԽӸ and ԗӹӰӠԐԂԽӸ seem to occur more frequently with ԖӹӢӹԶӹ than ԒӰԂԽӸ. To give an example, on the basis of the data given in table 6 one can calculate that there is a positive correlation of around 20% between the use of ԌӹԙԽԂԽӹԓӰӹ/ԌӹӴӹԊӸӰӹ ԗӹӰӠԐԂԽӸ, and the use of ԖӹӢӹԶӹ in comparison to sentences with ԒӰԂԽӸ (observed value of ԌӹԙԽԂԽӹԓӰӹ/ԌӹӴӹԊӸӰӹ ԖӹӢӹԶӹ ԗӹӰӠԐԂԽӸ = 12, expected value = 8.47; Fisher exact two-tailed p = 0.05, which means that the reliability is 95%). Another relevant, and possibly related factor is the status of the degree word. The data show that if the degree word is used to indicate a relative degree (‘enough with respect to’) and not as a moderator, the likelihood of ԖӹӢӹԶӹ decreases. This can be shown with regard to ԌӹԙԽԂԽӹԓӰӹ ԒӰԂԽӸ and ԌӹԙԽԂԽӹԓӰӹ ԖӹӢӹԶӹ ԒӰԂԽӸ in the twentieth century. There is a negative correlation of around 25% between the use of ԖӹӢӹԶӹ and the occurrence of a subordinate clause with ԓԽӹԼՂ (observed value of ԌӹԙԽԂԽӹԓӰӹ ԖӹӢӹԶӹ ԒӰԂԽӸ with ԓԽӹԼՂ 24

This explanation would also explain why a search in the corpus Frantext suggests that in French, the use of bien is probably more typical of trop than of assez.

THE CONSTRUCTION OF DEGREE IN RUSSIAN

89

= 4, expected value =7.89; Fisher exact two-tailed p = 0.04, which means that the reliability is 96%). In view of the small number of sentences with ԗӹԐӰӠԽӸ and ԗӹӰӠԐԂԽӸ, it is difficult to determine whether the same correlation also plays a part in the case of these verbs. It should be noted, though, that consultation of native speakers indicated that even in sentences with a subordinate clause and ԖӹӢӹԶӹ like (103) or (105), speakers are inclined to interpret ԌӹԙԽԂԽӹԓӰӹ as a modifier and the degree in absolute terms. To give an example, sentence (103) expresses that the subject speaks the language quite well, and therefore does not need an interpreter, whereas (104) without ԖӹӢӹԶӹ only expresses that the degree is sufficient to understand someone. 3.2.3 The use of ѸӁҸҊҸ with ԙԊӠԶӪӹԐ and gradable verbs In this section, I will discuss sentences with ԙԊӠԶӪӹԐ that occur both with and without the quantifier ԐӰӹԄӹ as in (8) and (9) given earlier. Before I propose an explanation of this use of ԐӰӹԄӹ, I will first provide a brief discussion of the meaning and function of ԐӰӹԄӹ with verbs in general. If used with a verb, and not functioning as the object of the verb, the quantifier ԐӰӹԄӹ indicates the large amount of occurrences of a situation, rather than the intensity of the (realization of the) situation (see also Bitextina 1975 for some observations about this use of ԐӰӹԄӹ in Russian, and Doetjes 2007, for an analysis of similar use of beaucoup in French): (108) ѩҽ, Ҿ ӊ ҦѦҲҡҲӊѦ ѸӁҸҊҸ ҡѹҦѦґҽ Ӂҽ ҊҸґҸѸ ҃ѦѸѦӁҲӁҸѸ ҰҸґҏ ӉѹѸҸҮ (…). (ҋ. ҫҡѹҰҸӊ, ՀԽӢԂԙԽӠ ԗӹ ԢӹԐӾ. ԹӰӠԄԂ ԽӢӾԽӸԋ. ԹӰӠԄԂ ӷӾӢӾԐӾӰ) yes, I in childhood much sat on bare concrete floor in.winter ‘Yes, in my childhood I spent a lot of time during the winter on the bare concrete floor.’ (109) [ү]ѸѦҝѹҔҽӁ҃Ӄ ҡґѹҗҔҸѸ ѸӁҸҊҸ ѦӉҦҾҲ Ӂҽ ѸҽҗѹӁҽѽ ѹ ӊҸҸқӂѦ ҸҲҸҗґѹ ҸҲ ҰҝѹҝҸҦӃ. (Ѣ. ҘҽҝҰҸӊҽ, ӷӠԙӸԐӹ ӠԒ ԣԐӾӢӠӪӠ ӹԼ ԂԐӾӢӠӪԂӰԙӪӹӡ ӪӫԖӰӾ) Americans too much drive in cars and in.general became.detached from nature ‘Americans drive too much, and in general have become detached from nature.’ In these sentences, the adverb ԐӰӹԄӹ in combination with a verb presupposes different (individual) events which can be taken together in order to evaluate the totality of events as ‘much’. The additional effect of using ԙԊӠԶӪӹԐ is that the quantity is presented as ‘too much’ (exceeding the appropriate quantity). Sentences like (109) must be distinguished from sentences without ԐӰӹԄӹ. In the latter case the speaker conceptualizes one (individuated) event as having a too

90

EGBERT FORTUIN

high degree. Since the verb ӾԒԌӠԽӸ (‘drive’) as in (109) is not a gradable verb, such a conceptualization is not possible (?ԙԊӠԶӪӹԐ ӾԒԌӠԽӸ). An important feature of the use of ԐӰӹԄӹ in (108) and (109) is that it refers to a collection of situations, which can be seen as similar and therefore be taken together in order to be evaluated as ‘much’. Note, however, that with some verbs, one cannot really speak of individuation. Take for example the atelic verb ӢԂԒԄӹӴԂӢӠӴԂԽӸ (razgovarivat’ ‘talk’), which has no accusative object valence, but which may be used with an oblique object with the preposition o ‘about’. If applied to this verb, ԐӰӹԄӹ can express both the long duration of one individual talking situation, and the content of the speaking of this situation, which is evaluated as ‘much’ (‘there was a lot of talking; a lot of things said’): (110) ҋҽӁѹӁ ҝҽӉҊҸӊҽҝѹӊҽґ ѸӁҸҊҸ, ҰҸ-ӊҍѦҝҽҗӁѦѸҏ, ӁҸ ӁѦ Ҹ ҩҸҡҡѹѹ ѹ ӁѦ Ҹ ҝҏҡҡҔҸҮ ңѹӉӁѹ. Sanin talked much, as.in.the.old.days, but not about Russia and not about Russian life Sanin talked a lot, as in the old days, but not about Russia and Russian life.’ As such it differs from the verb ԙӠԌӾԽӸ (sidet’ ‘sit’). The phrase ԙӠԌӾԽӸ ԐӰӹԄӹ (sit much) refers to a number of different sit-occurrences, and cannot refer to the large amount of sitting of one individual sitting occurrence. For this, other adverbs have to be used (e.g. ԌӹԊԄӹ, dolgo ‘long’). Sentences like (110) show similarity with uses where ԐӰӹԄӹ functions as the object of the verb, and does not modify the situation itself: (111) ҊҸӊҸҝѹҲҧ/ҍѹҲҽҲҧ ѸӁҸҊҸ say-INF/read-INF much‘say/read much/a lot (that is a lot of contextually given things)’ Note that the interpretation of ԐӰӹԄӹ as an object of the verb is often the most natural interpretation in the case of transitive verbs, especially when the aspect of the verb is perfective (see Bitextina 1975: 67). Take for example the following sentences with ҰҸӪӫԗԂԽӸ (‘buy’) and ӴӠԌӾԽӸ (‘see’): (112) Ѥ ѸӁҸҊҸ ҰҸҔҏҰҽґ. I much bought ‘I bought a lot (of things).’ (113) ҫӁ ѸӁҸҊҸ ӊѹҦѦґ ѹ ҰѦҝѦңѹґ. (ԠӫӢԒӠԊӪԂ, ҥ1, 2003) he much saw and survived ‘He saw and survived a lot (of things).’

91

THE CONSTRUCTION OF DEGREE IN RUSSIAN

In these sentences, the interpretation of the adverb ԐӰӹԄӹ is that of object of the verb. The interpretation where ԐӰӹԄӹ modifies the situation expressed by the verb (‘many buy/see occurrences’) is extensionally close to cases where ԐӰӹԄӹ functions as an object, and sometimes cannot even be distinguished from such cases. Nevertheless, the most natural interpretation is that the speaker wants to convey that the number of objects bought or seen is large, rather than that the total amount of realizations of the buy or see situation is ‘much’ (for example ‘ I have spent much time in the shop’). There are a few verbs where one can find minimal pairs consisting of uses with and without ԐӰӹԄӹ. On the basis of a search in the NKRJa with the search phrase [ԐӰӹԄҸ + V], table 8 lists the verbs from the verbs classes that occur with ԙԊӠԶӪӹԐ given earlier which occur as such minimal pairs.

қѦҊҽҲҧ (‘run’) қѦҡҰҸҔҸѹҲҧҡҾ (‘worry’) ҊҸӊҸҝѹҲҧ (‘speak’) ҦҸҝҸңѹҲҧ (‘value’) ҦҏѸҽҲҧ (‘think’) ѹҊҝҽҲҧ (‘play’) ҔҝѹҍҽҲҧ (‘scream’) ґҊҽҲҧ (‘lie’) ґӀқѹҲҧ (‘love’)

19th century - ԐӰӹԄӹ + ԐӰӹԄӹ 1 0

20th century - ԐӰӹԄӹ 0

1

0

1 (1912)

1 (past passive participle) 1

6

3 (ipf. adv. part.) 0

1

0

2

4

0

0

12 (2 from 1900) 3 (1 ipf. adv. part.) 2

0

0

1

0

39

1

4 (1 present active participle) 0

+ ԐӰӹԄӹ 2

35

20 (1 ipf. adv. part.) 3 1

2

98 (including: 5 0 present passive participles; 3 present active participles; 1 past active participle)

92

ґӀқҸӊҽҲҧҡҾ (‘admire’) ѸҏҍѹҲҧҡҾ (‘suffer’) ӁҸҡѹҲҧҡҾ (‘make a fuss’; ‘be worried’) ҰѹҲҧ (‘drink’) ҰҝѹӊӃҔҽҲҧ (‘get used’) ҝҽқҸҲҽҲҧ (‘work’)

EGBERT FORTUIN

19th century - ԐӰӹԄӹ + ԐӰӹԄӹ 0 0

20th century - ԐӰӹԄӹ 1

+ ԐӰӹԄӹ 1 (1912)

1

1

1 (adv. part.)

0

1

0

1

1

0

1

15 25

8

0

1 (present active participle) 46

1

1

1

5

8 (1 active present participle) 1 (present active participle) 1

1

ҡҲҝҽҦҽҲҧ (‘suffer’) ҝҽӉѸӃҗґҾҲҧ (‘reflect’) ҲҝҏҦѹҲҧҡҾ (‘work hard’) ҏӊҽңҽҲҧ (‘respect’)

4

1 (present active participle) 1

0

0

0

0

12 (1 with passive ԙԋ)

1 (present passive participle)

ѽґҸҰҸҲҽҲҧ (‘take trouble to’) ҃ѦӁѹҲҧ (‘value’)

2

0

4

1

1 1 (1919)

11 (1 ipf. adv. part.; 1 present passive participle) 1 (1917-1918)

0

9 (3 past passive participle; 1 past present participle)

0

1

Table 8: Minimal pairs with ԐӰӹԄӹ

The verbs that can occur both with and without ԐӰӹԄӹ denote different classes of situations. The largest class consists of verbs that indicate positive attitudes towards a person (or thing) such as ӫӴԂԞԂԽӸ (‘respect’), ՁӾӰӠԽӸ (‘value’), ԌӹӢӹԞӠԽӸ (‘value’), ԊԇԼӠԽӸ (‘love’), ԊԇԼӹӴԂԽӸԙԋ (‘admire’). Other verbs indi25

In many of these sentences, the quantifier can also be interpreted as the object of the verb.

THE CONSTRUCTION OF DEGREE IN RUSSIAN

93

cate (i) negative physical or psychological sensations such as ԐӫԓӠԽӸԙԋ (‘suffer’), ԙԽӢԂԌԂԽӸ (‘suffer’), or ԼӾԙԗӹӪӹӠԽӸԙԋ (‘worry’); (ii) verbs referring to work (ԽӢӫԌӠԽӸԙԋ ‘work hard’, ӢԂԼӹԽԂԽӸ ‘work’); (iii) verbs indicating the act of thinking (ԌӫԐԂԽӸ ‘think’, ӢԂԒԐՂԶԊԋԽӸ ‘reflect’), and (iv) verbs that are difficult to group together such as ԼӾԄԂԽӸ (‘run’), ӠԄӢԂԽӸ (‘play’), ԄӹӴӹӢӠԽӸ (‘speak’), ӪӢӠԓԂԽӸ (‘scream’), ԊԄԂԽӸ (‘lie’), ԗӠԽӸ (‘drink’), ԗӢӠӴՂӪԂԽӸ (‘get used’), ӰӹԙӠԽӸԙԋ (‘make a fuss’; ‘be worried’), ԖԊӹԗӹԽԂԽӸ (‘take trouble to’). How can the difference in use between sentences with and without ԐӰӹԄӹ be accounted for? In explaining the data the following factors need to be taken into account: (i) the semantic contribution of ԐӰӹԄӹ; (ii) the grammatical status of the verb; (iii) the difference between nineteenth en twentieth century Russian. Firstly, the semantic contribution of ԐӰӹԄӹ. Consider for example the following sentences: (114) ғ ҲѦҽҲҝ Ҿ ӁѦ ѽҸңҏ, ҰҸҲҸѸҏ ҍҲҸ ҲҽѸ ҡґѹҗҔҸѸ ѸӁҸҊҸ ҔҝѹҍҽҲ. (ԚԒӴӾԙԽӠԋ, 2002.02.21) in theatre not go, because there too much shout-3PL-PRES ‘I don’t go to the theatre, because there they shout too much.’ (115) ҫҲҸҮҦѦѸ-Ҕҽ. ҈ҦѦҡҧ ҡґѹҗҔҸѸ ҔҝѹҍҽҲ. (ҷ. үҔҏӁѹӁ, ԾӾԅӢԽӠԽӹӴՂӾ ԓӾԽӪӠ) let’s.go. here too shout-3PL-PRES ‘Let’s go. They shout too much here.’ In (114) the speaker does not want to convey that the intensity of the shouting in theatre is high, but that in the theater the number of occurrences of shouting taken together is high (there is too much shouting there). In (115) the speaker expresses that the shouting itself is seen as having a too high degree, more specifically because it is too loud. Alternative adverbs are therefore ԄӢӹԐӪӹ (‘loudly) or ԙӠԊӸӰӹ (‘strong’), which would convey that aspect of the action more explicitly. Note that a similar phenomenon occurs with the intensifier ӹԓӾӰӸ: (116) ѩѦҲѹ ҸҍѦӁҧ ҔҝѹҍҽҲ ҰҸ ӁҸҍҽѸ. (Ҍ. ҋѦѸѦӁҸӊ, ՀӾԐӰԂԌՁԂԽӸ ԐԄӰӹӴӾӰӠӡ ӴӾԙӰՂ) children very cry-3PL-PRES during nights ‘Childeren cry very much during the night.’

94

EGBERT FORTUIN

(117) ҋѦҮҍҽҡ ҸҍѦӁҧ ѸӁҸҊҸ ҔҝѹҍҽҲ: "ѢѦҲ ҔҏґҧҲҏҝӃ!" (ғ. ҩҸӉҸӊ, ӯԌӠӴԊӾӰӠӾ ԗӾӢӾԌ ԞӠԒӰӸԇ) now very much cry-3PL-PRES: ‘there.is.no culture!’ ‘Now they cry all the time: ‘There is no culture!’ Sentence (116) expresses that there is a high degree of crying. Sentence (117) with the quantifier ԐӰӹԄӹ does not say anything about the intensity of the crying, but expresses that there are many cry instances, or that the totality of different cry instances is seen as much. The description given here explains why sentences like ԙԊӠԶӪӹԐ ԐӰӹԄӹ ӫӴԂԞԂԇ (lit. honour too much) are extremely infrequent, and, as I will explain below, confined to the nineteenth century when ԐӰӹԄӹ had a different function than today. With this verb, the event cannot be individuated and conceptualized as taking place several times. In this respect it differs from ԙԊӠԶӪӹԐ ԐӰӹԄӹ ԙԽӢԂԌԂӾԽ (lit. suffer too much). One can imagine a person suffering on different occasions. Secondly, the difference in meaning discussed here is also related to the specific grammatical category of the predicate. With some verbs, the adverbial participle is more easily interpreted as expressing intensity than a finite verb. Consider for example the following sentences: (118) ѢѦ ҡґѹҗҔҸѸ ҦҏѸҽҾ, Ҿ ҸҔҽӉҽґҡҾ ҡҲҸҾӂѹѸ ҰҝҾѸҸ ҏ ѸѦҡҲҽ ҡҸҔҝӃҲѹҾ: ҰҝҸҡҲҸ ҸӁҸ ҡҽѸҸ ҸҔҽӉҽґҸҡҧ Ӂҽ ѸҸѦѸ ҰҏҲѹ. (ғ. ҴҽҔҽӁѹӁ, ՀԇӢ Ӵ ӷӢӹԊӾԽԂӢԙӪӹԐ ӢԂӡӹӰӾ) not too think-GER-PRES-IMPERF, I turned.out standing directly at hiding.place: just it self turned.out on my way ‘Without really thinking about it, I found myself standing at the hiding place: it just turned up on my way.’ (119) Ѥ ҡҍѹҲҽӀ, ҍҲҸ ҊґҏҰҸ ҡґѹҗҔҸѸ ѸӁҸҊҸ ҦҏѸҽҲҧ Ҹ ҡӊҸѦѸ ӉҦҸҝҸӊҧѦ (ԁӴӹӾ ԒԌӹӢӹӴӸӾ) I consider, that stupid too much think-INF about own health ‘I consider it stupid to think too much about one’s health.’ In example (118) with the adverbial participle, the form ԐӰӹԄӹ (‘much’) is not expressed. Here it is expressed that the intensity of the situation is not high (‘think to deep’, ‘paying attention’), whereas the use of ԐӰӹԄӹ in (119) makes explicit that the amount of time or frequency that the situation is realized is too high (‘think too often’). Thirdly, there are differences between the nineteenth and twentieth century. Note, for example that the verbs ԊԇԼӠԽӸ (‘love’), ӫӴԂԞԂԽӸ (‘respect’), and ՁӾӰӠԽӸ (‘value’, ‘appreciate’) were attested with ԐӰӹԄӹ in the nineteenth century, but not in the twentieth century. For example:

THE CONSTRUCTION OF DEGREE IN RUSSIAN

95

(120) ҭѦҝӊҸӀ ѹӉ Ӂѹѽ қӃґҽ ҊҝҽӍѹӁҾ ҩӃӂѦӊѦҔҽҾ (…), ҔҸҲҸҝҽҾ ҡґѹҗҔҸѸ ѸӁҸҊҸ ґӀқѹґҽ ӉҽӁѹѸҽҲҧҡҾ ҰҸґѹҲѹҔҸҮ. (Ҩ. ғѹҊѦґҧ, ԩԂԗӠԙӪӠ (1850– 1860)) first-INSTR from them was countess Ryšǁevskaja (…), who too much loved be.involved.in.politics ‘The first one [to come] was countess Ryšǁevskaja, who too much liked to be involved in politics.’ This is probably connected to the different status of ԐӰӹԄӹ in the nineteenth century. In the nineteenth century, ԐӰӹԄӹ could be used as an intensifier with verbs indicating situations associated with positive feelings such as ӫӴԂԞԂԽӸ (‘respect’), ՁӾӰӠԽӸ (‘value’, ‘appreciate’), ԄӹӢԌӠԽӸԙԋ (‘be proud of), ԊԇԼӠԽӸ (‘love’), ԖӴԂԊӠԽӸ (‘praise’), ӪԊԂӰԋԽӸԙԋ (‘greet’, ‘bow’); and negative feelings such as ԙԽӢԂԌԂԽӸ (‘suffer’), ӹԄӹӢԓԂԽӸ, ԗӾԓԂԊӠԽӸԙԋ (‘grieve’), ԗӹԙӹԼԊԋԽӸ, ԖԊӹԗӹԽԂԽӸ (‘trouble’), ӹԙӪӹӢԼԊԋԽӸ (‘insult’). Below two examples are given with verbs that also occur in the construction of degree: (121) Ѥ ѸӁҸҊҸ ґӀқґӀ ѹ ҏӊҽңҽӀ ѦҊҸ. (Ѩ. һҽңѦҍӁѹҔҸӊ, ԹӹԊԌӫӰ ӰԂ ՀӫԖԂӢӾӴӹӡ ԼԂԶӰӾ (1840)) I much love and honour him ‘I love and honour him very much.’ (122) ҭҝҸӂҽҮҲѦ, ѸѹґҸҡҲѹӊҽҾ ҊҸҡҏҦҽҝӃӁҾ, ѸӁҸҊҸ ҃ѦӁӀ, ҍҲҸ ѸѦӁҾ ҰҸҡѦҲѹґѹ. (Ѣ. һѦҡҔҸӊ, ՀӹԼӹӢԋӰӾ (1872)) farewell, madam, much appreciate, that me visited ‘Farewell, madam, I appreciate it very much that you visited me.’ In modern Russian, a more neutral way would be to use the intensifier ӹԓӾӰӸ, for example ӹԓӾӰӸ ԊԇԼԊԇ (lit. very love). The use of ԐӰӹԄӹ as an intensifier can be considered archaic. In modern Russian, the phrase ӵ ԐӰӹԄӹ ԊԇԼӠԊ (lit. I much loved) therefore usually refers to the total amount of love occurrences (‘I was many times in love’), rather than the intensity of the love (‘I loved very intensively’). This use presupposes that there are different (individual) events that occur in different or the same time interval, but which can be taken together in order to evaluate the totality of events as much. It can therefore occur with situations such as ԙԽӢԂԌԂԽӸ (‘suffer’), ԌӫԐԂԽӸ (‘think’), or ԗӹԐӹԄԂԽӸ (‘help’), but not with ӫӴԂԞԂԽӸ. As such, Russian differs from languages like French or English, where one can use the quantifiers beaucoup or a lot with verbs like aimer/to love. The use of ԐӰӹԄӹ as an intensifier and its relation to its use with ԙԊӠԶӪӹԐ and ӹԓӾӰӸ is a topic that deserves a separate study. There are a great number of details that are unclear, and which need further analysis.

96

EGBERT FORTUIN

Firstly, the specific boundaries of the different classes of verbs that occur with ԐӰӹԄӹ as an intensifier in the nineteenth century are not fully clear to me. The examples from the nineteenth century seem to be restricted to gradable verbs indicating situations associated with positive and negative feelings and attitudes. It is possible that in the eighteenth century, this class of verbs was larger, because the NKRJa contains examples from the end of the eighteenth century with ӹԙӠԗӰӫԽӸ (‘get hoarse’), and ӴӾӢӠԽӸ (‘believe in‘), which both seem to fall outside of the typical nineteenth century gradable verb classes that occur with ԐӰӹԄӹ. Secondly, in the nineteenth century, ԐӰӹԄӹ could probably be used both as an intensifier and as a quantifier with some verbs. This is for example the case with ԙԽӢԂԌԂԽӸ. As such, the use of ԐӰӹԄӹ in (8) may be interpreted both as an indication of quantity or frequency (‘too many suffer instances’), and as an indication of intensity (‘suffer deeply’), whereas in modern Russian it can only indicate repetition (Bitextina 1975: 67). Thirdly, not all gradable verbs that could occur in the nineteenth century with [ԐӰӹԄӹ + V], are also attested with [ԙԊӠԶӪӹԐ ԐӰӹԄӹ + V]. Consider for example the verb ԄӹӢԌӠԽӸԙԋ (‘be proud of’), which is attested three times in the nineteenth century with the intensifier ԐӰӹԄӹ (‘be very proud of’). However, the only two instances with ԙԊӠԶӪӹԐ both occur without ԐӰӹԄӹ. Fourthly, although in the nineteenth century ԐӰӹԄӹ was used as an intensifier with some gradable verbs, the use of the intensifier ӹԓӾӰӸ was more frequent in the same contexts. This may suggest that this use of ԐӰӹԄӹ was already decreasing, or that the semantics of these two intensifiers was not identical. It is also interesting to notice that the four examples from the end of the eighteenth century that I attested with ԐӰӹԄӹ ԊԇԼӠԽӸ all occur with an intensifier: ԙԽӹԊӸ/ӴӾԙӸԐԂ/ӹԓӾӰӸ ԐӰӹԄӹ ԊԇԼӠԽӸ (so/very/very much love). The specific semantic difference and relation between ӹԓӾӰӸ (or its semantic equivalents) and ԐӰӹԄӹ needs more study. Finally, it is interesting to notice that ԐӰӹԄӹ is also attested in the medieval birchbark texts from Novgorod (see the site gramoty.ru, October 11, 2007; ҊҝҽѸҸҲҽ 31, 283, 302, and 414). In these fragments it is used with phrases that indicate greetings of the type ‘bowing deeply for someone’, e.g.: ԐӰӹԄӹ ՁӾԊӹԐ ԼӸԇ ‘much forehead bow-1SG-PRES’, ‘bow deeply’, or with the verb ӪԊԂӰԋԽӸԙԋ (‘bow’). In such phrases the quantifier is (probably) used to indicate that the action is performed to a high degree (gramoty.ru translates it with ӰӠԒӪӹ ‘deeply’). In old Russian, ԐӰӹԄӹ could probably be used more broadly to indicate a high degree of something. Filin (1982), indicates that in the 12-th century ԐӰӹԄӹ could be used as an intensifier with gradable verbs, but also as the first part of an adjective to indicate a high degree with compounds like ԐӰӹԄӹԽӢӫԌӰՂӡ (in modern Russian ӹԓӾӰӸ ԽӢӫԌӰՂӡ, ‘very difficult’). This suggest that ԐӰӹԄӹ originally functioned both as a quantifier and as an intensifier

THE CONSTRUCTION OF DEGREE IN RUSSIAN

97

(with verbs and probably a specific class of adjectives), but gradually lost its intensifying meaning, first in the adjectival domain, and in a later stage in the verbal domain. In the nineteenth century ԐӰӹԄӹ had retained its function of intensifier with some specific gradable verbs only (indicating positive and negative attitudes and feelings), and in modern Russian this function is lost almost entirely. What remains is the function of quantifier. 26 3.3 Nouns In this section, I will discuss the use of the degree words with nouns. Examples of this use were already given in (10) and (11). In sentences with ԙԊӠԶӪӹԐ and a noun, the adverb ԐӰӹԄӹ (‘much’) is obligatorily used, and the noun occurs in the genitive case. 27 Such sentences express that the quantity of the things expressed by N is too big. For example ԙԊӠԶӪӹԐ ԐӰӹԄӹ ԌӾӰӾԄ (too much moneyGEN, ‘too much money’). If ԌӹԙԽԂԽӹԓӰӹ is used with a noun, the noun also has genitive marking (‘enough of something’), but ԐӰӹԄӹ is not used in most sentences: ԌӹԙԽԂԽӹԓӰӹ ԌӾӰӾԄ (enough money-GEN, ‘enough money’). Instances where ԌӹԙԽԂԽӹԓӰӹ is combined with ԐӰӹԄӹ can, however, be found: (123) Ѱ ҰҝҸҔҏҝҽҲҏҝӃ ҦҸҡҲҽҲҸҍӁҸ ѸӁҸҊҸ ӊѦҡҔѹѽ ҸҡӁҸӊҽӁѹҮ, ҍҲҸқӃ ҰҝҸҦҸґңѹҲҧ ҙҲҏ ҰҝҸ҃ѦҦҏҝҏ. (www.polit.ru/archivego.html?day=15&month=09§ion=news&year= 2003) at public prosecutor enough many serious proof, that-IRR continue-INF that procedure ‘The public prosecutor has enough serious proof to continue that procedure.’ One could argue that in such sentences ԌӹԙԽԂԽӹԓӰӹ does not indicate a minimum but functions as a moderator (‘rather much’, ‘quite much’). This is however not in accordance with the use of the subordinate clause with ԓԽӹԼՂ, which provides a standard of comparison (‘enough with respect to’). It is possible, however, that this use of ԐӰӹԄӹ is a relatively new phenomenon which is 26 The development of ԐӰӹԄӹ in Russian is possibly part of a more general typological trend that quantifiers evolve into degree expressions and vice versa (see Doetjes (forthc.) for some suggestions). With respect to ԐӰӹԄӹ it should also be kept in mind that the grammatical status of the word itself has changed. In older stages of Russian, ԐӰӹԄӹ could be used as an adjective, where this is not possible in modern Russian (see for example Filin 1982). 27 This includes sentences where the adjective refers to a group or type of entities, for example ԙԊӠԶӪӹԐ ԐӰӹԄӹ ԖӹӢӹԶӾԄӹ (too much good-ADJ-SG-GEN) ‘too many good things; ԙԊӠԶӪӹԐ ԐӰӹԄӹ ԄӹԊӫԼՂԖ (too many blue-ADJ-PL-GEN (=gays); ‘too many gays’).

98

EGBERT FORTUIN

influenced by the increase in the use of ԌӹԙԽԂԽӹԓӰӹ as a moderator. One might argue that the semantics of sentences like these differs from normal instances of the construction of degree, and conveys something like ‘rather much X such that Y is the case’. With respect to the use of the quantifier (ԐӰӹԄӹ) Russian is similar to English or Dutch, where much/many and Dutch veel are not used with enough and genoeg respectively, for example too much money, versus enough money. This suggests that because of its meaning, ԌӹԙԽԂԽӹԓӰӹ (or its Dutch and English counterparts) is more intrinsically associated with quantity than ԙԊӠԶӪӹԐ (and its Dutch and English counterparts). One should be careful, however, to see this as a logical consequence of the different meanings of the degree words. Note, for example, that there are also languages where both the form expressing ‘too’ and the form expressing ‘enough’ occur without quantifier. Such is for example the case in French: assez de livres (enough of books, ‘enough books’ and trop de livres (too of books, ‘too many books’). Whether or not there are languages where only the form expressing ‘enough’ occurs with a quantifier is unfortunately not known to me. Another usage where ԙԊӠԶӪӹԐ or ԌӹԙԽԂԽӹԓӰӹ occur with a noun is a construction consisting of a subject, copula and a qualitative noun in the nominative case. In such sentences the nouns express notions such as a nationality, or another qualification of people such as ‘child’ or ‘human’. Such uses are quite infrequent, and rather marked for many speakers of Russian: (124) ҫӁҽ ҦҸҡҲҽҲҸҍӁҸ ѦӊҝѦҮҔҽ, ҍҲҸқӃ ңѹҲҧ ӊ ѨӉҝҽѹґѦ, ѹ ӁѦҦҸҡҲҽҲҸҍӁҸ – ӊ ҐѦҝѸҽӁѹѹ! (ԾӹӴՂӡ ԠӠӢ, ҥ1–2, 1999) she enough jew-N-F-NOM, that-IRR live in Israel, and not.enough in Germany ‘She is Jewish enough to live in Israel, and not enough to live in Germany.’ (125) ѩґҾ ҙҲҸҊҸ ҘҸӉҽҔ ӊҡѦ-ҲҽҔѹ ҡґѹҗҔҸѸ ҍѹӁҸӊӁѹҔ. (www.gazeta.ru/comments/2004/12/17_e_213289.shtml) for that Kozak anyway too bureaucrat-N-MASC-NOM ‘For that, Kozak is too much a bureaucrat anyway.’ Here, the noun has a similar function as an adjective, and qualifies the subject. The specific use depends on the noun. There may be differences between ԙԊӠԶӪӹԐ and ԌӹԙԽԂԽӹԓӰӹ here. To give an example, with ӾӴӢӹԗӾӾՁ and ԓӠӰӹӴӰӠӪ only instances with ԙԊӠԶӪӹԐ were attested. Note that in these sentences, ԐӰӹԄӹ is not inserted, and the noun occurs in the nominative case: (126) *ҡґѹҗҔҸѸ ѸӁҸҊҸ ѦӊҝҸҰѦѦ҃

THE CONSTRUCTION OF DEGREE IN RUSSIAN

99

too much European-N (cf. English: ‘too much of a European’; Dutch: ‘teveel een Europeaan’) As such, these nouns behave like adjectives. It must be noted that not all qualitative nouns allow for the use of ԙԊӠԶӪӹԐ. This is the case, for example, with nouns that express qualifications that are semantically (and sometimes morphologically) related to situations, for example: (127) *ҡґѹҗҔҸѸ ґҊҏӁ (ґңѦ҃)/қҸґҲҏӁ too liar/talker Instead, adjectives like ԼӹԊӸԶӹӡ (bol’šoj; ‘big’) or ӰԂԙԽӹԋԷӠӡ (nastojašǁij; ‘real’), ԙԽӢԂԶӰՂӡ (strašnyj; ‘terrible’), etc. are used: (128) ҡґѹҗҔҸѸ қҸґҧҗҸҮ ґҊҏӁ/ґңѦ҃/қҸґҲҏӁ too big liar/liar/talker ‘too much of a liar’ Although the corresponding verbs allows for the use of the adverb ԐӰӹԄӹ, this is not the case with these nouns: (129) ґңѦҲ ѸӁҸҊҸ/қҸґҲҽѦҲ ѸӁҸҊҸ lies much/talks much ‘he lies too much/talks too much’ (130) *(ҡґѹҗҔҸѸ) ѸӁҸҊҸ ґҊҏӁҽ (ґңѦ҃ҽ)/қҸґҲҏӁҽ too much liar/talker 3.4 Non-dependent uses In this section, I will discuss three uses where the degree word does not depend directly on another constituent. Firstly, as I indicated earlier, both ԙԊӠԶӪӹԐ (with the quantifier ԐӰӹԄӹ) and ԌӹԙԽԂԽӹԓӰӹ may occur as objects of a transitive verb, for example: (131) ҴӃ ҡґѹҗҔҸѸ ѸӁҸҊҸ ӉӁҽѦѸ, ҍҲҸқӃ Ӂҽҡ ҰҝҸҦҽӊҽҲҧ Ӊҽ ҊҝҽӁѹ҃ҏ. (ѣ. ҐѹӁӉқҏҝҊ, ԹӢӫԽӹӡ ԐԂӢԶӢӫԽ) we too much know, that-IRR us sell-INF abroad ‘We know too much, for them to sell us to another country.’

100

EGBERT FORTUIN

(132) Ҙ ҲҸѸҏ ӊҝѦѸѦӁѹ үӁҦҝѦҮ ҏңѦ ҦҸҡҲҽҲҸҍӁҸ ӉҽҝҽқҽҲӃӊҽґ, ҍҲҸқӃ ҰҸҔҏҰҽҲҧ ѸӁѦ ҰҝѦҔҝҽҡӁӃѦ қҏҔѦҲӃ ҃ӊѦҲҸӊ. (ԟӾԊӹ (ҋҽѸҽҝҽ), 2002.04.11) by then, Andrej already enough earned, that-IRR buy-INF me beautiful flowers ‘By then, Andrej already earned enough to buy me beautiful flowers.’ In such sentences, the degree words modify contextually given things. Such, uses are similar to cases where the degree word occurs with a noun (see 3.3). A similar example, with ԌӹԙԽԂԽӹԓӰӹ ԐӰӹԄӹ, can be found below, where the degree word functions as predicate of the sentence (modifying a situation referred to by ԲԽӹ (èto ‘that’): (133) ҶҲҸ ҦҸҡҲҽҲҸҍӁҸ ѸӁҸҊҸ, ҍҲҸқӃ ѹҡҰӃҲҽҲҧ ҏҦҸӊґѦҲӊҸҝѦӁѹѦ ҸҲ ҰҝҸҦѦґҽӁӁҸҮ ҝҽқҸҲӃ. (ԾӹӴՂӡ ԠӠӢ, 2002) that enough much, that-IRR experience satisfaction from carried.out work ‘That is enough, to experience satisfaction from the work that is carried out.’ On the internet, similar cases can be found without ԐӰӹԄӹ. Secondly, in some sentences, ԙԊӠԶӪӹԐ may be used metaphorically. This use can be found in sentences like the following: (134) ѢѦҲ, 600 ҦҸґґҽҝҸӊ – ҙҲҸ ҡґѹҗҔҸѸ. (ԩԂ ӢӫԊӾԐ, 2004) no, 600 dollar – that too ‘No, 600 dollar, that is too much.’ (135) ѩӊҸѦ – ҙҲҸ ҡґѹҗҔҸѸ. (ՀԽӹԊӠՁԂ, 1997.09.02) two.people that too ‘Two (people) – that is too much’. In thѦse sentences, ԙԊӠԶӪӹԐ modifies a non expressed idea of quantity, based on the first clause. Note, however, that the idea of quantity may be rather abstract, as is evident from the following sentence: (136) ҘҏӊѦҮҲ: ңѦӁҡҔѹҮ ӍҏҲқҸґ – ‘ҙҲҸ ҡґѹҗҔҸѸ’ (ԮӠ-ԼӠ-ԙӠ, 2001.04.09) Kuwait: women soccer – ‘that too’ ‘Kuwait: women soccer- ‘that is too much’.’ In such sentences, ԲԽӹ ԙԊӠԶӪӹԐ is better analyzed as an idiomatic expression expressing the feeling that a particular event ‘is too much too handle’ or ‘too much of a good thing’. Finally, the adverb ԌӹԙԽԂԽӹԓӰӹ is used in sentences where it occurs in a subject-predicate construction. About something (a thing or action) is said that it

THE CONSTRUCTION OF DEGREE IN RUSSIAN

101

suffices. These sentences have a copular element, but in the present tense the copular element is not expressed in the case of the verb ԼՂԽӸ. Two types can be distinguished, namely sentences with an infinitive and sentences with a noun in the genitive case: (137) ѩҸҡҲҽҲҸҍӁҸ ҰѦҝѦӊѦҝӁҏҲҧ ҡҲҝҽӁѹ҃ҏ, ҍҲҸқӃ ҏҡґӃҗҽҲҧ ҡҸӊѦҝҗѦӁӁҸ ѹӁҸҮ ҊҸґҸҡ. (ԜӪԽԋԼӢӸ, ҥ5, 2002) enough turn-INF page, that-IRR listen-INF totally other voice ‘It’s enough to turn the page, in order to hear an altogether different voice.’ (138) ѩҸҡҲҽҲҸҍӁҸ ҰҽҡҡѹӊӁҸҊҸ ҡҸӉѦҝ҃ҽӁѹҾ, ҍҲҸқӃ ҝҽҦҸӊҽҲҧҡҾ ҙҲѹѸ ҃ѦӁӁҸҡҲҾѸ. (ԛӹԗӢӹԙՂ ԗԙӠԖӹԊӹԄӠӠ, ҥ5, 2003) enough passive-GEN contemplation-GEN, that-IRR enjoy these valuable.things ‘Passive contemplation is enough to enjoy these valuable things.’ In the case of the infinitive, a dative noun may be used to express the experiencer of the infinitive predicate: (139) ѢҽѸ қӃґҸ ҦҸҡҲҽҲҸҍӁҸ ҡӃҊҝҽҲҧ ӊ ҡӊҸӀ ҸқӃҍӁҏӀ ҡѹґҏ, ҍҲҸқӃ ҰҸқѦҦѹҲҧ ѹҡҰҽӁ҃Ѧӊ. (ԚԒӴӾԙԽӠԋ, 2001.09.07) we-DAT it.was enough play-INF in own normal strength, that-IRR beatINF Spaniards ‘It was sufficient for us to play with our normal strength to be able to beat the Spaniards.’ See Kor-Chahine 2002 for an analysis of this construction. 4. Conclusion and some final remarks I have given a semantic-syntactic analysis of the construction of degree in Russian. Let me first sum up the main conclusions. The degree words ԙԊӠԶӪӹԐ (‘too’) and ԌӹԙԽԂԽӹԓӰӹ (‘enough’) both indicate a relative degree. The meaning of ԙԊӠԶӪӹԐ can be defined in terms of a degree which exceeds the appropriate degree with respect to some standard (and which has negative consequences). Put differently, ԙԊӠԶӪӹԐ expresses that a lower degree is necessary with respect to some standard. The standard relative to which the degree is measured can be expressed by a subordinate clause introduced by ԓԽӹԼՂ (that-IRR, ‘to’).The meaning of ԌӹԙԽԂԽӹԓӰӹ is that the degree is appropriate with respect to some standard, and that a higher degree is not necessary. In some contexts ԌӹԙԽԂԽӹԓӰӹ can also express that a higher degree is not desirable with respect to some standard, because it already is high, or even too

102

EGBERT FORTUIN

high. This use is possibly related to the non-relative use of ԌӹԙԽԂԽӹԓӰӹ as an indication of a rather or quite high degree (the so-called moderator use). In contrast to most treatments of these degree words in the literature, I have argued that it is not necessary to contribute the modality of the construction (more specifically the subordinate clause) to one of the constituents in the construction. The modal character of the construction is an interpretation of the meaning of the construction as a whole (degree word, and the subordinate clause introduced by ԓԽӹԼՂ) and can have a character of possibility or necessity, depending on the context in which it occurs. The most prototypical interpretation, however, is that of possibility (permission). I have also discussed the syntactic (combinatory) possibilities of ԙԊӠԶӪӹԐ and ԌӹԙԽԂԽӹԓӰӹ in quite some detail. As I showed, both ԙԊӠԶӪӹԐ and ԌӹԙԽԂԽӹԓӰӹ can be combined with adjectives and adverbs in the same way. With respect to other parts of speech, there are, however, differences. In my analysis I have focused mainly on the combinatory possibilities with verbs. The degree word ԙԊӠԶӪӹԐ is applied directly to gradable verbs, for example ԙԊӠԶӪӹԐ ԊԇԼӠԽӸ (too love, ‘love too much’). In the case of nongradable verbs, for example ӾԖԂԽӸ (‘drive’), it is necessary to use an adverb to indicate the dimension of the degree (e.g. ԙԊӠԶӪӹԐ ԌӹԊԄӹ (‘long’)/ԓԂԙԽӹ (‘often’)/ԐӰӹԄӹ (‘much’)/ԖӹӢӹԶӹ (‘well’) ӾԖԂԽӸ). Verbs like ԒӰԂԽӸ (‘know’) constitute a special class, because they occur both with and without the adverb ԖӹӢӹԶӹ (‘well’) (ԙԊӠԶӪӹԐ (ԖӹӢӹԶҸ) ԒӰԂԽӸ). With respect to this adverb, I have pointed at the difference between the twentieth century and the nineteenth century, and between the different verbs in this class. I have also pointed at other of such ‘minimal pairs’, for example ԙԊӠԶӪӹԐ (ԙӠԊӸӰӹ) ԗԂԖӰӫԽӸ/ԗӹԽӾԽӸ (too (strongly) smell.like/sweat). Minimal pairs could also be established with and without the quantifier ԐӰӹԄӹ, for example ԙԊӠԶӪӹԐ (ԐӰӹԄӹ) ӪӢӠԓԂԽӸ (too (much) shout). I have shown that the use of ԐӰӹԄӹ with ԙԊӠԶӪӹԐ can point at two things. In modern Russian it contributes the meaning of quantity. In the nineteenth century, ԐӰӹԄӹ was also used as an intensifier with gradable verbs like ԊԇԼӠԽӸ (‘love’), and made the feature of intensity, inherent to the construction of degree, explicit. Now let’s turn to the degree word ԌӹԙԽԂԽӹԓӰӹ. Like ԙԊӠԶӪӹԐ, the degree word ԌӹԙԽԂԽӹԓӰӹ is applied directly to gradable verbs. However, a similar tendency can be perceived with verbs like ԒӰԂԽӸ, where the adverb of quality ԖӹӢӹԶӹ (‘well’) is used to a lesser extent than in the case of ԙԊӠԶӪӹԐ. The data suggest that a possible factor here is that ԌӹԙԽԂԽӹԓӰӹ can also be used as a modifier (‘quite’). Another difference with ԙԊӠԶӪӹԐ is that if the verb is associated with a particular quantity, this is not additionally expressed by the quantifier ԐӰӹԄӹ (cf. ԌӹԙԽԂԽӹԓӰӹ ӾԖԂԽӸ (enough drive, ‘drive enough’) versus ԙԊӠԶӪӹԐ ԐӰӹԄӹ ӾԖԂԽӸ (too much drive, ‘drive too much’)). A similar phenomenon

THE CONSTRUCTION OF DEGREE IN RUSSIAN

103

can be perceived in the case of nouns. Here, ԙԊӠԶӪӹԐ is not applied directly to the noun, and the quantifier ԐӰӹԄӹ is used. As such its use differs from ԌӹԙԽԂԽӹԓӰӹ, which is applied directly to the verb. The data suggest that ԙԊӠԶӪӹԐ behaves like a degree modifier comparable to an intensifier such as ӹԓӾӰӸ (‘very’). It can therefore only be applied to forms that have a gradable meaning. ԟӹԙԽԂԽӹԓӰӹ, however, both functions as a degree modifier (when it is applied to gradable adjectives or verbs), and as a quantifier. In the latter case it functions as an indicator of the degree of the quantity expressed by verbs or nouns. The main question which remains is: where does this difference come from? In order to explain this difference, it is interesting to look at ԓӢӾԒԐӾӢӰӹ (‘overly’), which shares properties both with ԙԊӠԶӪӹԐ and ԌӹԙԽԂԽӹԓӰӹ. The data from the NKRJa show that it is most frequently used with adjectives as in (29) given earlier. It may, however, also be used with gradable verbs as in (140). With these verbs the use of the quantifier ԐӰӹԄӹ is not acceptable. In the case of verbs that are usually not seen as gradable such as ӾԖԂԽӸ (‘drive’) or ԙԗԂԽӸ (‘sleep’) as in (141), the degree word is not applied directly to the verb. In such sentences ԐӰӹԄӹ can be used to indicate quantity, but the use of other adverbs, for example ԓԂԙԽӹ (‘often’) or ԌӹԊԄӹ (‘long’) is possible as well, depending on the meaning the speaker wishes to convey: (140) ҍҝѦӉѸѦҝӁҸ ҝҽҦҸӊҽҲҧҡҾ/ґӀқѹҲҧ/ҏӊґѦҔҽҲҧҡҾ overly enjoy-INF/love-INF/be.carried.away.by-INF ‘overly enjoy/love/be carried away by’ (141) ҍҝѦӉѸѦҝӁҸ ѸӁҸҊҸ/ҍҽҡҲҸ/ҦҸґҊҸ ѦѽҽҲҧ/ҡҰҽҲҧ overly much/often/long drive-INF ‘drive/sleep too much’ Note, however, that there are also instances where ԓӢӾԒԐӾӢӰӹ is used with verbs that are not typically seen as gradable, but where the degree word is applied directly to the verb. In similar contexts with ԙԊӠԶӪӹԐ, an adverb would be obligatory with these verbs. This is for example the case with transitive verbs that have objects that can easily be conceptualized as having a particular quantity: (142) ѷҸґҧҔҸ ӁѦ ӁҏңӁҸ ҍҝѦӉѸѦҝӁҸ ӁҽґѹӊҽҲҧ ѦѸҏ ѹӉ ӍґҾҊѹ, ҽ ҲҸ ҰѦҲҧ ӊҸӉҧѸѦҲҡҾ. (ѣ. ҤҽѦ҃ҔҽҾ, ҴҝҽҔҸқѦҡ/ҋӊҸҝҽ ҰҝҸҰҽӂѹѽ) just not need overly pour-INF him from flask, otherwise sing-INF starts ‘It’s just that one shouldn’t give him too much to drink [lit. overly pour him from the flask], otherwise he will start to sing.’ (143) ѢѦґҧӉҾ ѹ ҍҝѦӉѸѦҝӁҸ ѦҡҲҧ ѹ ҰѹҲҧ. (suscribe.ru/archive/science.health.beauty.aibolit2001/200609/24070508. html)

104

EGBERT FORTUIN

may.not also overly eat-INF and drink-INF ‘One should not overly eat and drink either.’ In the case of sentences like (142) and (143) the degree word does not function as the object of the verb, because the object can be expressed, for example ԓӢӾԒԐӾӢӰӹ ӫԗӹԽӢӾԼԊԋԽӸ ԂԊӪӹԄӹԊӸ (lit. overly use alcohol). Furthermore, data from the internet show that the boundary between gradable and non gradable verbs is not clear-cut. Take for example the verb ԗӹԽӾԽӸ (‘sweat’). In combinaton with ԓӢӾԒԐӾӢӰӹ this verb occurs without quantifier. It could be argued that this is because with ԓӢӾԒԐӾӢӰӹ this verb itself is associated with a very large quantity of sweat. Evidence for this point of view is that this verb can be used with the intensifier ӹԓӾӰӸ in combination with the quantifier ԐӰӹԄӹ (ӹԓӾӰӸ ԐӰӹԄӹ ԗӹԽӾԽӸ, lit. ‘to sweat overly much’). However, the same verb is also used with ӹԓӾӰӸ without quantifier (ӹԓӾӰӸ ԗӹԽӾԽӸ, lit. overly sweat). The difference between ԓӢӾԒԐӾӢӰӹ and ԙԊӠԶӪӹԐ can also be perceived when it is used with nouns. In such cases it is most common to use ԐӰӹԄӹ: (144) ҍҏӊҡҲӊҽ қӃґҸ ҍҝѦӉѸѦҝӁҸ ѸӁҸҊҸ. (Ґ. үґѦҔҡѦѦӊ, ҈ѦґѦӁӃѦ қѦҝѦҊҽ) feeling-GEN be-3SG-NEUT-PAST overly much ‘there were overly many feelings’ (145) ҭҸҙҲҸѸҏ ҦґҾ Ӂҽҡ ӁѦӊҸӉѸҸңӁҸ ӁҽқѹҝҽҲҧ ҍҝѦӉѸѦҝӁҸ ѸӁҸҊҸ ҡҲҏҦѦӁҲҸӊ (…). (www.spbumag.nw.ru/2006/14/6.shtml) therefore for us impossible select-INF overly many students-GEN ‘Therefore it is impossible for use to select too many [lit. overly many] students.’ On the internet examples can also be found with nouns in the genitive case without ԐӰӹԄӹ. These are either examples of nouns that express abstract concepts such as ԙӠԊԂ (‘strength’), ӴӰӠԐԂӰӠӾ (‘attention’), ӴӢӾԐԋ (‘time’), ӹԼԋԒԂӰӰӹԙԽӸ (‘obligation’), or concrete objects such as ԌӾӰӸԄӠ (‘money’), ӴӹԌԂ (‘water’), or ӴӹԊӹԙ (‘hair’). Most of these uses are probably considered substandard Russian by the majority of speakers of Russian, but some sentences are acceptable: (146) ѯҝѦӉѸѦҝӁҸ ҦѦӁѦҊ ӁѦ ѸҸңѦҲ қӃҲҧ, ҰҸҲҸѸҏ ҍҲҸ ҰҸҲҝѦқӁҸҡҲѹ ҍѦґҸӊѦҔҽ ҝҽҡҲҏҲ қӃҡҲҝѦѦ, ҍѦѸ ҔҸґѹҍѦҡҲӊҸ ҦѦӁѦҊ. (www.old.samara.ru/paper/565/5604/98706/) overly money-GEN not can be-INF, because needs man-GEN grow faster, than amount money-GEN ‘You can never have too much money, because the needs of a man grow faster than the amount of money.’

THE CONSTRUCTION OF DEGREE IN RUSSIAN

105

In this context, ԓӢӾԒԐӾӢӰӹ can be replaced by the quantifier ԐӰӹԄӹ ‘much’ (ԐӰӹԄӹ ԌӾӰӾԄ ӰӾ ԐӹԞӾԽ ԼՂԽӸ), or by ԙԊӠԶӪӹԐ ԐӰӹԄӹ (ԌӾӰӾԄ ӰӾ ԐӹԞӾԽ ԼՂԽӸ ԙԊӠԶӪӹԐ ԐӰӹԄӹ). Finally, in the NKRJa examples can also be found where the degree word functions as a predicate, and where it has the form of a short adjective: (147) ѢҽӊѦҝӁҸѦ, ӊґѹҾӁѹѦ Ҩ. Ҵ. ѩҸҡҲҸѦӊҡҔҸҊҸ Ӂҽ ҋѦґѦӉӁҺӊҽ қӃґҸ ҍҝѦӉѸѦҝӁҸ (Ѣҽҗ ҡҸӊҝѦѸѦӁӁѹҔ, 2004) probably, influence-NEUT-NOM F.M. Dostoevskij-GEN on SeleznevGEN was-3SG-NEUT-PAST overly-ADJ-SHORT-NEUT ‘Probably, the influence of F.M. Dostoevskij on Seleznev was huge’. At this point, it is interesting to turn to Doetjes (forthc.). Doetjes shows that degree words can be classified according to the parts of speech they can be combined with. She lists 7 types: (A) adjectives only; (B) adjectives and gradable verbs; (C) adjectives, gradable verbs, eventive (non gradable) verbs, mass nouns and plural nouns; (D) gradable verbs, eventive verbs, mass nouns, plural nouns; (E) eventive verbs, mass nouns, plural nouns; (F) mass nouns and plural nouns; and (G) plural nouns only. See table 9, which shows how the different degree words that I have discussed in this paper combine with different parts of speech; note that types (A) and (G) were not attested among the degree words studied by me. adjective/ adverbs gradable verbs nongradable (eventive) verbs

nouns

Type B ԙԊӠԶӪӹԐ ӹԓӾӰӸ

mixed type ԓӢӾԒԐӾӢӰӹ

Type C ԌӹԙԽԂԽӹԓӰӹ Type D ԐӰӹԄӹ (19th century; only with a specific class of gradable verbs)

Type E ԐӰӹԄӹ (20-th century)

Type F ԗӢӾԌӹԙԽԂԽӹԓӰӹ

Table 9: Overview of syntactic properties of degree words

if the verb is associated with quantity, without ԐӰӹԄӹ (ӾԙԽӸ, ԗӠԽӸ, etc.) the use of ԐӰӹԄӹ obligatory in most contexts

106

EGBERT FORTUIN

In this classification, ԙԊӠԶӪӹԐ is a type B degree word, and ԌӹԙԽԂԽӹԓӰӹ a type C degree word. Note that the behavior of all degree words, except ԓӢӾԒԐӾӢӰӹ, is in accordance with the predication by Doetjes that degree expressions do not cover non-adjacent fields of the continuum from adjective to plural noun. How can the difference between ԌӹԙԽԂԽӹԓӰӹ and ԙԊӠԶӪӹԐ be explained, and why does ԓӢӾԒԐӾӢӰӹ share features with both degree words? The reason why ԌӹԙԽԂԽӹԓӰӹ has a more independent status is possibly that it indicates a conceptually less complex concept than ԙԊӠԶӪӹԐ. Whereas ԌӹԙԽԂԽӹԓӰӹ indicates an appropriate degree or quantity, ԙԊӠԶӪӹԐ indicates that the degree exceeds the appropriate degree. The degree word ԓӢӾԒԐӾӢӰӹ shares semantic features with both. In contrast to ԙԊӠԶӪӹԐ, it does not indicate a relative degree, but like ԙԊӠԶӪӹԐ it expresses that some norm has been exceeded. Another possible relevant factor is whether the degree word can function as an object or predicate. This is not the case for ԙԊӠԶӪӹԐ, which never occurs on its own, and always modifies another element. In contrast, however, ԌӹԙԽԂԽӹԓӰӹ can function as the object of a verb, and as a predicate. The degree word ԓӢӾԒԐӾӢӰӹ seems to take an intermediate position. It can have a function close to that of an object with verbs like ӾԙԽӸ or ԗӠԽӸ, and is also used as a predicate. I think that these syntactic differences are, at least partly, related to the meanings of these forms, and probably also the grammatical status of these forms (both ԌӹԙԽԂԽӹԓӰӹ and ԓӢӾԒԐӾӢӰӹ can function as adjectives, and as predicates). The behavior of ԓӢӾԒԐӾӢӰӹ suggests that the model given by Doetjes is correct, but that a more fine-grained version is necessary. In my view, such a model should not make universal claims about which grammatical categories are gradable or not, and where the boundaries are between such categories. In the end, whether or not a particular category or instance of that category is seen as gradable or not is determined by the question of whether it can be combined with intensifiers. The particular boundaries of concepts like gradable verbs differ, however, from language to language. Furthermore, they may even differ within one language, since different intensifiers may behave differently with respect to particular verbs. This means that one cannot postulate linguistic universals here, but only formulate cross-linguistic statistical tendencies. Further research should therefore focus both on the details of specific languages, and on differences between languages. Leiden University

THE CONSTRUCTION OF DEGREE IN RUSSIAN

107

REFERENCES Apresjan, Jurij (ed.) 2004 Novyj ob’’jasnitel’nyj slovar’ sinonimov russkogo jazyka [New Explanatory Dictionary of the Synonyms of the Russian Language]. Izd. 2 ispravl. i dopoln. pod obšǁim rukovodstvom akademika Ju.D. Apresjana. Moskva: Škola ‘Jazyki russkoj kul’tury’. Filin, F.P. (ed.) 1982 Slovar’ russkogo jazyka XI-XVII vv, V. 9. Mosvka: Nauka. Bierwisch, Manfred 1989 “The semantics of gradation”. In: Manfred Bierwisch, Ewald Lang (eds.), Dimensional adjectives. Grammatical structure and conceptual interpretation, 71261. Berlin-Heidelberg: Springer. Bitextina, G.A. 1975 “Ob upotreblenii koliǁestvennyx nareǁij ӹԓӾӰӸ i mnogo”. Russkij jazyk za rubežom 1, 66-68. Bolinger, Dwight 1972 Degree words. The Hague: Mouton de Gruyter. Dagurov, G. V. 1994 “Dostatoǁno i dovol’no”. Russkaja Reǁ’: nauǁno-populjarnyj žurnal Rossijskoj akademii nauk 2, 59-60. Doetjes, Jenny forthc. “Adjectives and degree modification”. In: McNally, Louise and Chris Kennedy (eds.), Adjectives and Adverbs: Syntax, Semantics and Discourse. Oxford: Oxford University Press. 2007 “Adverbs and quantification: degrees versus frequency”. Lingua, 117(4), 685720. Groen, Ben 1998 “The Use of the Long and Short Adjectival Forms in Contemporary Standard Russian.” In: Adriaan Barentsen, Ben Groen, et al (eds.), Dutch Contributions to the Twelfth International Congress of Slavists, Cracow: Linguistics (Studies in Slavic and General Linguistics 24), 151-173. Amsterdam-Atlanta: Rodopi. Heim, Irene 2001 “Degree Operators and Scope”. In: Caroline Féry, Wolfgang Sternefeld (eds.), Audiatur Vox Sapientiae. A Festschrift for Arnim von Stechow. Internet version: www2.sfs.unituebingen.de/~arnim10/Festschrift/Heim-7-6komplett %20fertig.pdf (7-1-2007). Kennedy, Christopher and Louise McNally 2005 “Scale structure and the semantic typology of gradable predicates”. Language 81.2, 345-381. Kor-Chahine, Irina 2002 “La suffisance quantitative : sémantique des prédicats russes”. In: Essais sur le discours de l’Europe éclatée. 18, 111-131. Grenoble: Université Stendhal, Centre d’études slaves contemporaines, n° 18. forthc. “Le verbe impersonnel xvatat’ / xvatit’ : de la syntaxe à la sémantique”. In: Questions de linguistique slave. Études offertes à M. Guiraud-Weber. Aix-enProvence: Université de Provence.

108

EGBERT FORTUIN

Meier, Cécile 2003

“The meaning of too, enough and so...that”. Natural Language Semantics 11, 69-107. Švedova, Natalija (ed.) 1980 Russkaja Grammatika. V.2. Moskva: Nauka. Vasil’ev, A.D. 2001 ‘‘Zametki. Slovo-kompensator’’. In: Rossijskij lingvistiǁeskij ežegodnik Sibiri. Vypusk 3, 160–166. Internet version: library.krasu.ru/ft/ft/_articles/0088625.pdf (9-20-2007) Verhagen, Arie 2005 Constructions of Intersubjectivity. Oxford: Oxford University Press.

SUMMARY This article provides a detailed semantic and syntactic analysis of the construction of degree in Russian. This construction consists of the degree words ԙԊӠԶӪӹԐ (‘too’) or ԌӹԙԽԂԽӹԓӰӹ (‘enough’), and a subordinate clause introduced by ԓԽӹԼՂ (‘to’). An explanation is given of why the two degree words share particular semantic features such as the association with modality, and why they differ in other respects. The paper also addresses the way in which the degree words are combined with different parts of speech (adjectives, verbs and nouns), and whether such combinations require the use of ԖӹӢӹԶӹ (‘ well’) or the quantifier ԐӰӹԄӹ (‘much’). It is shown that in analyzing such combinations both the synchronic and the diachronic dimension have to be taken into account.

Dutch Contributions to the Fourteenth International Congress of Slavists, Ohrid: Linguistics (SSGL 34). Amsterdam – New York: Rodopi, 2008, 109-143.

PERDURATIVITY IN POLISH – 100 YEARS AFTER AGRELL RENÉ GENIS

0. Introduction The Swedish poet, slavist and runologist (Per) Sigurd Agrell introduced the concept of PERDURATIVITY to Slavic and indeed general linguistics exactly 100 years ago. 1 He uses the (German) term PERDURATIV for one his AKTIONSARTEN (1908: 52) and renders it POSTAſ PERDURATYWNA (1918: 65) in the later Polish version of his publication. 2 It serves to typify situations such as appear in the following example. Przegada’ ca’y wieczór. ‘He has through-chatted the whole evening.’3 The prefixed verb przegadaǀ ‘through-chat’ is accompanied by ca’y wieczór ‘whole evening’ to convey the meaning that that particular portion of time was spent chatting. A comprehensive description of Polish perdurativity, including its component parts and other considerations such as aspectual matters, has not appeared to date. In section 1 of this paper I will present as much information as can be gathered from the two versions of Agrell’s work as a basis for the following discussion. In section 2 I will take a look at more recent publications on this topic as well as add my own views, in an attempt to fine-tune the concept of perdurativity. Following that, in section 3, I will discuss imperfective perdurativity. 1. Agrell’s perdurativity Agrell made a formidable contribution to the study of Slavic verbal aspect by introducing the highly important differentiation between aspect and Aktions1

In most of his publications Agrell does not mention his first name, Per. The two versions of Agrell’s work are slightly different in presentation of the material and in content. Moreover, the later Polish version (1918) includes a dictionary part that is lacking in the earlier German work (1908). 3 Please note that the prefix prze- is translated as ‘through-’. Here and there in this paper I have approached the matter of translating the examples in a similar way in order to elucidate the structure. 2

110

RENÉ GENIS

art(en). In his introduction (1908: 1-6, 1918: 4-11) he states that the main aim of his work is to clarify the existence in Polish of so many, differently prefixed, perfective verbs, which in dictionaries are usually given the same translation into German, and which all seem to function as counterparts for single imperfective simplexes. In both his publications, Agrell discusses at length the principles underlying the functioning of prefixes in respect to lexical meaning and aspect. In his view, prefixes can on the one hand provide a base-verb with a completely new meaning, but on the other, merely cause aspectual change from imperfective to perfective.4 For the latter, so Agrell points out, we need to differentiate between two cases. The first is what nowadays is often termed an “empty” prefix.5 The second Agrell describes as follows: “Forma z’oǻona wyraǻa takǻe, jak rozwija’a siNj czynnoǧǀ (lub stan); przedrostek nie tylko sprowadza zmianNj w aspekcie (rodzaju) treǧci czasownikowej, lecz zarazem zaznacza takǻe pewien charakter, pewnƾ postaǀ (= niem. ‘Aktionsart’) czynnoǧci lub stanu.” (Agrell 1918: 5).6 ‘The compound form also expresses how an action (or state) developed; the prefix does not only introduce aspectual change of the contents of the verb, but at the same time indicates a certain character, a certain shape (= German ‘Aktionsart’) of the action or state.’ Importantly then, Agrell would have it that his Aktionsarten, perdurativity included, are firmly connected with perfectivity. Agrell points out that the perdurative Aktionsart concerns time (as opposed to space) and that it brings to the fore the beginning and end of a temporal course (1908: 52, 78, 122, 1918: 64, 65, 105) without qualitative shades of meaning (1918: 69). In his summary he visualises it thus (1908: 82, 1918: 111):

Agrell describes his Aktionsarten and sets them off against each other whilst discussing particular characteristics of groups of base-verbs. In his discussion of 4

The term BASE-VERB is used here to refer to the simplex verb that lies at the basis of the compound, prefixed formation under discussion. 5 The appropriateness of the term “empty” prefix (as well as the concept of “perfectivisation”) is discussed in Genis (2008: 157-198). Here it is argued that the term is in fact unhandy and misleading; prefixes are never “empty” and, with the exception of very few cases, mark lexical semantic terminativity. 6 I have opted to quote this later, probably reconsidered, Polish version of this passage rather than the earlier, slightly less elaborate German version (1908: 2).

PERDURATIVITY IN POLISH – 100 YEARS AFTER AGRELL

111

the prefix prze-, marker of perdurativity in Polish, he contrasts perdurativity with his PRÄTERITIV / POSTAſ PRETERYTYWNA ‘PRETERITIVE AKTIONSART’ and his DURATIV / POSTAſ DURATYWNA ‘DURATIVE AKTIONSART’. 7 The preteritive Aktionsart (formed with prefix po- and corresponding to what nowadays is referred to as the DELIMITATIVE AKTIONSART) defines undefined time, which is also shorter than with the perdurative prze-compound, and lacks a clear beginning and end. His durative Aktionsart (often, but not exclusively formed with prefix u-) presents an action without regard to whether it takes a longer or shorter amount of time, whilst conveying a much clearer resultative sense. Perdurativity is here characterised as pointing to defined time (1908: 52 1918: 65). Agrell writes furthermore, that there is a quite specific resultative quality to perdurativity: “das Resultat = ein Zeitverlauf” ‘the result = a passage of time’ (1908: 78, 79, cf. 1918: 108). As far as matters semantic are concerned, this is probably as much as we can gather. Let us turn to some formal considerations. Agrell explicitly mentions that prze- is the only marker of the perdurative Aktionsart in Polish (1908: 122, 1918: 105). Agrell does not actually name or categorise the base-verbs that form perdurative compounds according to semantic specifics other than their behaviour in conjunction with certain Aktionsart prefixes. His text is somewhat confusing, when at one point he describes such verbs as spaǀi ‘sleep’ and trwaǀi ‘last’ as forming “pure” perfectives with prefix prze- (1918: 64). Only a little further on he clearly names the very formations przespaǀp ‘through-sleep’ and przetrwaǀp ‘through-last’ as perduratives (1918: 65, 66). 8 His description of these particular base-verbs – “czasowoǧǀ tkwi tu juǻ w samem pojNjciu prostego czasownika” ‘temporality is embedded here already in the very concept of the base-verb’ – is poignant and interesting. I think the latter verb could be seen as a “pure” perfective form of trwaǀi (although an other candidate might be potrwaǀp which occurs slightly more frequently on the internet). I think przespaǀp and spaǀi do not have this kind of relationship. In a section on “pure” perfectivity, which only 7

For the sake of completeness I should mention that in his earlier publication, Agrell recognises two further Aktionsart functions for the prefix prze-: PRÄTERITIV and EFFEKTIV-DEFINITIV (1908: 53, 54, 111, 112). Interestingly, the use of prze- as preteritive marker is toned down quite considerably in (1918: 66), where he speaks of a ‘meaning that is related (pokrewne) to preteritive po-’. It concerns a meaning of prze-, which I prefer to describe as ‘execute not (quite) totally’ or ‘with lesser intention’; cf. Genis (2008: 51, 52) It is beyond the scope and need for this paper to elaborate on these here. 8 Agrell also mentions at this point (1908: 52, 1918: 64) the verbs ǻyǀi ‘live’, nocowaǀi ‘spend the night’, zimowaǀi ‘spend the winter’ and czekaǀi ‘wait’. Other than is the case with spaǀi and trwaǀi, these do not return in his further discussion of the prefix prze- and the perdurative Aktionsart, although some are mentioned as forming perdurative compounds in the dictionary part of his work: przenocowaǀp (1918: 552), przezimowaǀp (1918: 618), przeczekaǀp (1918: 176).

112

RENÉ GENIS

occurs in Agrell (1908: 126) he writes that the prefix prze- is used to form “pure” perfective counterparts of verbs, which express an act without actual activity, and which only have bearing on the passing of a certain (amount of) time. Examples of this are przenocowaǀp ‘through-stay the / a night’, przezimowaǀp ‘through-spend the / a winter’ and przespaǀp ‘through-sleep’. The latter verb is not a good example and the former two might not be perduratives, as I will discuss at the end of 2.2.9. Other matters are best discussed on the basis of Agrell’s examples in the following list. These were gathered from both versions of his work. The order here is that of 1918 to which the examples unique to 1908 were added.9 (1)

Przesiedzia’p w wiNjzieniu trzyA kwarta’y. (AgP) ‘He has through-sat/spent three quarters in prison.’10

(2)

Przesta’p ca’ƾ nocA pode drzwiami na mrozie. (Ag) ‘He has through-stood the whole night at the door in the cold.’

(3)

Przecierpia’emp te okropne chwileA. (AgP) ‘I have through-suffered those awful moments.’

(4)

Przeczeka’p nocA ca’ƾ. (Ag) ‘He has through-waited the whole night.’

(5)

Przegraliǧmyp z sobƾ partyNjA bilardu. (Ag) ‘We have through-played with each other a game of billiards.’

(6)

Ca’e latoA przemieszka’p w Warszawie. (Ag) ‘The whole summer I have through-lived in Warsaw.’

(7)

Przespa’p ca’e dwaA dni. (Ag) ‘He through-slept the whole two days.’

9 In the examples in this paper I have added indexes for aspect (subscript “p” for perfective, “i” for imperfective) and case with the main component of relevant noun phrases (subscript small capital “A” for the accusative, “G” for the genitive and “I” for the instrumental). Also, I have marked the source with each example. This list of markers refers to the bibliography and other sources. Ag Agrell (1908 and 1918) Du Dunaj Po Polaǝski AgG Agrell (1908 only) Ho Holvoet Ʀm Ʀmiech AgP Agrell (1918 only) In Internet Wr Wróbel Ba Baǝko Ni Native informant Do Doroszewski Pi Piernikarski When an example is drawn from literature, the source is mentioned in a footnote. 10 The verb przesiedzieǀ literally means ‘through-sit’, but often is used in its meaning ‘through-spend’. In this paper I have translated it in various ways to reflect its particular meaning in each example.

PERDURATIVITY IN POLISH – 100 YEARS AFTER AGRELL

(8)

Ten dƾb przetrwa’p aǻ do dziǧ. (Ag) ‘That oak tree has through-lasted until today.’

(9)

Przegada’p ze mnƾ ca’y wieczórA. (Ag) ‘He has through-chatted with me the whole evening.’

113

(10) Przeczyta’p gazetNjA. (Ag) ‘He has through-read the/a newspaper.’ (11)

Tej chwiliG ja nie przeǻyjNjp. (AgG) ‘This moment I will not through-live.’

(12) Dzieciak przebecza’ pó’A nocy. (AgG) ‘The child through-cried half the night.’ Although Agrell does not point this out himself, all these sentences, with the exception of (8), include what I will provisionally term an accusative or genitive complement. 11 These are either clear indications of units of time or entities that can readily be interpreted to “take time”, i.e. to have an implied temporal semantic component of duration. So, (1), (3), (7) and (11) have indicators of units of time whilst (2), (4), (6), (9) and (12) have indicators with a strong link to time (night, evening and such like are ‘times of day’, whereas summer is ‘a time of year’). Example (8) has no accusative / genitive complement but the adjunct is clearly one of time, with a pronounced end and a beginning that is readily interpretable as ‘the beginning of the existence of the oak tree’. The complements in (5) and (10) are more complicated and I suppose that Agrell interprets them as indicators of time indirectly, with implied duration. I do not regard these examples of perdurativity, which I will discuss in 2.2.1. Of course, Agrell’s publications are nowadays considered to be somewhat dated. The particular term perdurativity has survived with its meaning more or less intact, although additional bits of information can be gathered from publications since those of Agrell. Certain notions and restrictions have been added and I think a few more can be introduced. In the light of this we need to take a fresh look at some matters concerning perdurativity. 2. The component parts of perdurativity In this section I will take a look at several publications that touch the subject of Polish perdurativity in order to lay down a somewhat clearer picture of the concept. I will include some more recently expressed views as well as my own and delineate the restrictions employed in the following discussion. I think the mat11

In this paper the general term TEMPORAL COMPLEMENT is used to indicate any non-verbal provider of the duration in a given sentence.

114

RENÉ GENIS

ter is best approached by dealing with the various component parts of perdurativity separately. I will tackle this in the following order: the prefix prze-, the temporal complements and the nature of the base-verbs. 2.1 The prefix przeThere seems to be complete consensus among authors on the subject that the prefix prze- is the primary, indeed in Polish, the only marker of perdurativity. However, certain authors have defined other kinds of perdurativity in addition to the “classic” version we found with Agrell. Wróbel (1984: 475) mentions CONATIVE-PERDURATIVE formations such as wysiedzieǀp godzinNj ‘sit out an hour’ with prefix wy- and NORMATIVE-PERDURATIVE formations such as odsiedzieǀp godzinNj ‘sit out an hour’ with prefix od-. Both prefixes transitivise – a quality they have in common with prze- in perdurative compounds – and the direct objects with such compounds are clearly temporal in nature. The first of these two formations is more neutral as to the act of “sitting out”, whilst the second provides the additional information that external force was at the basis of why the hour had to be sat out. Czarnecki (1989: 53, 54) adds to these a PERDITIVE-PERDURATIVE Aktionsart, which comprises such formations as przebaraszkowaǀp czasA ‘fritter away / waste time’, przespaǀp ca’y dzieǝA ‘waste the whole day sleeping’, which always include prefix prze-. 12 This seems rather redundant, as the aspect of ‘wastefulness’ is not actually inherent in the prefix or in the compound verb as such. This part of the meaning is either embedded in the meaning of the base-verbs (baraszkowaǀi ‘frolic, gambol, caper, romp’), or in the context. In this respect it is worth noting that przespaǀp caly dzieǝA does not necessarily express wastefulness and can simply mean ‘sleep throughout the whole day’. Moreover, examples like przepi’p swój ca’y majƾtekA ‘he wasted his entire fortune (drinking)’ show that this perditive meaning, again one of the variants of prze-, is not restricted to temporal predicates. Czarnecki also combines perdurativity with fortativity into his FORTATIVE-PERDURATIVE Aktionsart, which has two formations: na- + siNj and wy- + siNj. Examples are: Piotr naczeka’p siNj trzy godziny ‘Piotr has waited no less than three hours’ and wystalip siNj ca’y dzieǝ w kolejce ‘they spent the whole day standing in a queue’. In this paper I will restrict myself to the more “classic” cases of perdurativity that are purely about spending time and in which the prefix prze- plays a part. Therefore “conative”, “normative” and “fortative” perdurativity will not be taken into account any further. In my recent publication of studies on the Polish verbal prefix prze- (Genis 2008), I pointed out that the meaning variants of this prefix can be reduced to 12

Czarnecki (1989: 53, 54) points out that he combined the perdurative and the perditive Aktionsart to form this term. The latter was based on Faßke ((1979) 1981: 119).

PERDURATIVITY IN POLISH – 100 YEARS AFTER AGRELL

115

two basic variants, which in other Slavic languages correspond by and large with the meanings of cognates of the two Proto-Slavic prefixes *pro- / *pra- (e.g. Russian ԗӢӹ-) and *per(i)- / (*prLJ-) (e.g. Russian ԗӾӢӾ-) respectively (Genis 2008: 66, 67). The group of variant *pro- meanings includes the perdurative function,13 which is rendered visually thus (Genis 2008: 41):

This is but one variant of the invariant:

The following three points describe the invariant: •

• •

Points P and Q; the existence of both points is clear, however, Q can be clearly located as ‘immediately adjacent to the outer edge of R, coinciding with the exact end of the TRAJECTOR’, whilst P is merely situated ‘at the beginning of the trajector’. A landmark R (for the variant meaning we are dealing with: R = ‘time’) at whose edge is marked the border with Q. A trajector linking P to Q.14

It is not difficult to spot the resemblance between the basic underlying idea of perdurativity as expressed by Agrell with the above invariant, or specific variant. The landmark is the period of time, which is bordered by a beginning and ending, all of which are crossed by way of the trajector. Importantly, the begin13 Bogus’awski (1963: 117) mentions that in “isolated cases” in Russian, the prefix ԗӾӢӾ- forms perdurative verbs. His example is ԗӾӢӾԼӠԽӸԙԋ Ԍӹ ӴӾԙӰՂ ‘manage until spring’, which implies ‘making it through winter’. Russian native speakers confirm, however, that it carries a strong notion of ‘reaching spring’ and so, the focus is more on ‘getting there’, than on ‘getting through the winter’. This is in fact the primary meaning of nepe- as opposed to npo- in Russian. 14 It is beyond the scope and aim of this contribution to go over the process through which this basic invariant was arrived at. For the full discussion, see Genis (2008: 15-67).

116

RENÉ GENIS

ning of the trajector in this representation is necessarily vague and it is by no means as clear as the end. In the perdurative situations of sentences (1) through (4), (6), (7), (8), (9), (11) and (12) the two borders suggested are, however, very clear, due to the nature of the temporal complements. Prze- on its own does not provide the perdurative Aktionsart; complements play a vital role. The prefix easily allows landmarks of an entirely non-temporal nature such as seen in przejechaliǧmyp ca’ƾ PolskNjA ‘we drove through the whole of Poland’. This particular example is very unlikely to be intended temporally due to the directionally definite verb of motion that lies at the base of the compound, and due to the non-temporal character of the complement. 15 I will now discuss temporal complements that crop up in perdurative situations. 2.2 Temporal complements The title of this section is necessarily vague as to the nature of the provider of the temporal information that specifies the temporal landmark in perdurative structures. As was already mentioned in passing here and there in this contribution, there are more candidate formations than merely the accusative complement, which figures with most authors. With example (8) above, Agrell may have hinted at other formations also applying, and Piernikarski (1969: 119, 120) actually mentions that an “adwerbium czasowe” ‘temporal adverb’ (by which he seems to also mean przez + NPA constructions) can also fulfil this function. He does, however, not develop the idea. Grzegorczykowa (1975: 59) speaks of perduratives and delimitatives as connoting temporal determinators but does not elaborate, and in her perdurative examples we only find temporal accusative objects. Wróbel (1984: 474) mentions that “wyraǻenie nazywajƾce ten odcinek czasowy pojawia siNj przy derywacie [perduratywnym] w sposób obligatoryjny” ‘the expression of the temporal period appears with [perdurative] derivatives in an obligatory way’. He then provides examples with the temporal accusative object as well as with a prepositional complement od + NPG do + NPG. Only the latter two structures are mentioned by Przybylska (2006: 159).16 15

See 2.3. Temporal objects are not likely to combine with this kind of verb as, in the words of Piernikarski (1969: 120), “they specialised to indicate spatial measure”. Actually, Piernikarski states that for this reason these combinations are impossible, which is in fact not true as I will discuss in 2.3. Also, with verbs of motion a complement with the preposition przez ‘through’ (cognate of prefix prze-) is possible as well, and so we can say przejechaliǧmyp przez ca’ƾ PolskNj ‘we have through-driven through the whole of Poland’, although there are some semantic differences to consider as I will explain in 2.2.2. Cf. footnote 17 and Genis (2008: 197-217). 16 Przybylska mentions at this point (2006: 159) that Bogus’awski (1963: 70) calls both of these formations “czasowo-tranzytywne” ‘temporally transitive’. My own impression is that Bogus’awski is only writing about the temporal accusative object.

PERDURATIVITY IN POLISH – 100 YEARS AFTER AGRELL

117

In the following subsections I will discuss and expand on the various complements that have been mentioned by other authors (2.2.1-2.2.3) as well as add some more that are plausible in perdurative structures (2.2.4-2.2.7). I will also briefly turn to some complements that native speakers reject (2.2.8). In section 2.2.9 I will sum up these matters. 2.2.1 The accusative / genitive complement: NPA|G Most authors on perdurativity mention and illustrate the accusative object as relevant to the expression of perdurativity. Grochowska (1979: 70) mentions only the temporal accusative object in connection with perdurativity and she specifically states that it is connoted by the prze-compound. It expresses the relevant time frame (“rama czasowa”). Most of Agrell’s examples contain such an accusative complement, but (5) and (10) are of a different nature. Przybylska (2006: 157) accepts structures with a prze-compound and objects such ca’y wyk’adA ‘the whole lecture’ as perdurative. I rather disagree with Przybylska, as the temporality of such objects is in my view at best secondary and there is no need to consider these in the light of perdurativity. For this reason, (5) and (10) will no longer be included in this discussion and I will restrict myself to such complements that refer to an amount of time and duration explicitly. This stance finds considerable support in Polaǝski (1980-1992), whose identified perduratives are always explicated with examples showing noun phrases of the type ca’e piNjtnaǧcieA lat ‘the whole 15 years’ rather than the type partiNjA bilardu ‘a game of billiards’ or gazetNj ‘newspaper’. Holvoet (1991: 81 ff.) provides a very detailed description of the temporal object with perdurativity. In his discussion on objecthood, this author compares temporal objects with locative ones, such as in przejechaliǧmyp miastoA ‘we drove through the town’. 17 In this construction the object is not affected by the action and it is the reference point of relative localisation: “… its spatial dimensions determine the range of the spatial parameter of the predicate”. This works also for temporal objects and the temporal dimensions determine the range of the temporal parameter of the predicate. With prze-compounds, then, the meaning of the prefix implies the limits to this temporally localised object. So far I have refrained from calling this temporal complement simply accusative / genitive object or even direct object, although the similarity to this is obvious. We need to establish the appropriateness of the term direct object for this kind of complement in perdurative structures. We turn again to Holvoet, 17 Holvoet (1991: 81) gives the alternative construction przejechaliǧmyp przez miasto ‘we throughdrove through the/a town’ and seems to suggest that these two constructions are synonymous. These constructions, when referring to locative situations are not altogether synonymous, as I will explain in 2.2.2 below. Cf. footnote 15 and Genis (2008: 197-217).

118

RENÉ GENIS

who writes that one of the conditions for objecthood is the fact that, in Polish, clear direct objects are put in the genitive when the predicate is negated. This is also the case with the temporal accusative complement in perdurative constructions. His example is as follows. (13) Piotr nie przesiedzia’p tam dwóch godzinG. (Ho) ‘Piotr didn’t spend two hours (sitting) there.’ Agrell’s example (11) also shows this. Compare the following negated version of earlier example (9), which confirms that we are dealing with real direct objects. (14) Nie przegada’p ze mnƾ ca’ego wieczóraG, (tylko do dziewiƾtej.) (Ni) ‘He has not through-chatted with me the whole evening, (only until nine.)’ This goes for all the other Agrell examples with temporal direct objects: (1), (2), (3), (4), (6), (7), (9) and (12). 18 For this reason we will refer to this temporal complement as NPA|G. Matters are more complicated in the case of the following example, which has two NPA. (15) Przeskroba’emp ca’y dzieǝA ziemniakiA, a teraz fajrant. (Pi) ‘I have through-scrubbed the whole day potatoes, and now I am free.’ For this, there are two negated versions some native informants could produce.19 (16) Nie przeskroba’emp ca’ego dniaG ziemniakówG … (Ni) ‘I have not through-scrubbed the whole day potatoes …’ (17) Nie przeskroba’emp ca’y dzieǝA ziemniakówG … (Ni) ‘I have not through-scrubbed potatoes the whole day …’ According to native informants (16) has a clear emphasis on the temporal notion and so could represent perdurativity, whilst in (17) perdurativity is questionable; here the emphasis is very much on the potatoes and the action expressed by the verb itself. Native informants felt another, non-temporal variant meaning of prefix prze- coming through: ‘execute with less(er) intensity’. Here: ‘I did not really scrub those potatoes very well throughout the whole day’, which is not perdurative. 18

The word pó’ ‘half’ in (12) is indeclinable; its form does not change when put in the genitive. The native informants who could not produce these negated versions probably could not on account of the unfamiliarity of the verb. 19

PERDURATIVITY IN POLISH – 100 YEARS AFTER AGRELL

119

Holvoet also employs certain transformational devices (1991: 88, 89) to establish objecthood. These also seem to point to a clear object status of this kind of temporal complement in perdurative constructions. Compare the following examples, which show that with the perdurative verb the object can be a relative pronoun, whilst the true accusativus temporalis with verbs like siedzieǀ ‘sit’ can not. (18) Godziny, któreA przesiedzia’emp za biurkiem. (Ho) ‘The hours I spent (sitting) at my desk.’ (19) *Godziny, któreA siedzia’emi za biurkiem. (Ho) ‘The hours I spent sitting at my desk.’ A further criterion involves the possibility to transform certain structures into the passive voice. The occurrence of przesiedzieǀp ‘through-sit/spend’ type verbs in such situations is quite infrequent, but, according to Holvoet, passive participle constructions do occur: (20) Nieprzespana noc. (Ho) ‘A sleepless night.’ (21) Iloǧǀ przesiedzianych godzin.20 (Ho) ‘The number of hours spent (on sitting) there.’ All in all, the accusative / genitive temporal complement can be treated as a real direct object in perdurative structures and from here onwards I shall refer to it as NPA|G. 2.2.2 The prepositional complement: przez + NPA Piernikarski (1969: 119-121) and Mindak (1987: 197) explicitly mention the formation przez + NPA as a possible complement in perdurative constructions such as przesiedzieǀp przez dzieǝ ‘through-sit/spend through the day’. Mindak mentions that this structure occurs less frequently than the simple temporal direct object and adds, as does Piernikarski, that these perfective verb structures are linked to imperfective siedzieǀi dzieǝA ‘sit (for) a day’ or siedzieǀi przez dzieǝ ‘sit through the day’. This latter observation relinquishes acceptability as Holvoet (1991: 89) points out that the imperfective verb siedzieǀi ‘sit’ plus accusativus temporis construction does not allow for the kind of transformational exercise 20 This example as well as (18) shows in some way, that plural objects should be possible. This is so indeed as shown in przegada’yǧmyp ca’e wieczoryA ‘we have chatted through many evenings’. Natives indicate, as expected with perfective aspect, that the amount of evenings this happened is closed of and therefore finite.

120

RENÉ GENIS

we can perform on przesiedzieǀp to establish objecthood as we saw in (18) through (21).21 Piernikarski (1969: 121) adds that certain verbs can occur with the NPA|G construction as well as the przez + NPA construction, whilst certain verbs can only occur in przez + NPA formations. He gives this example: (22) Przeczyta’p coǧA przez dzieǝ. (Pi) ‘He has through-read something through(out) the day.’ I do not count this kind of example as perdurative, since the direct object here is coǧ ‘something’ and, although this is perhaps not immediately apparent from Piernikarski’s example, this must clearly be non-temporal (ksiƾǻkNj, “Pana Tadeusza” etc.). The przez-complement here has the status of temporal adjunct, whilst the prze-meaning first and foremost relates to the non-temporal direct object. We need to look further for examples of the przez + NPA construction as complement in perdurative predicates. Compare the following, based on Piernikarski. (23) Przesiedzieliǧmyp dzieǝA. (Pi, In) ‘We have spent the day (sitting).’ (24) ?Przesiedzieliǧmyp przez dzieǝ. (Pi, In) ‘We have spent the day (sitting).’ Piernikarski (1969: 121 n3) remarks that constructions like these are probably not entirely synonymous, but he does not elaborate. I have not been able to persuade my native informants at all to (24) and they were all quite adamant, hence the question mark. The general opinion amongst my group of native informants was that a perdurative prze-compound does not take a przez-complement to set the duration of the perdurative action. This conviction gains validity when one realises that searches in a Polish text corpus did not produce any hits. However, in one of the dictionaries I used, this structure was also found in an example with the verb przetrwaǀ ‘through-last’. (25) Dinozaury by’y doskonale przystosowane do ǧrodowiska i przetrwa’y przez sto milionów lat. (Ba) ‘The dinosaurs were brilliantly adjusted to [their] environment and through-lasted through 100 million years.’

21

Holvoet (1991: 88-92) discusses this point extensively, provides further argumentation and actually gives such accusative temporal complements with verbs like siedzieǀi the status of adverb rather than direct object.

PERDURATIVITY IN POLISH – 100 YEARS AFTER AGRELL

121

Furthermore, in spite of my informants’ adamancy, Internet searches for [przesiedzieǀ przez] and other verbs yielded a very restricted number of examples. The following is a selection. (26) Tam przesiedzieliǧmyp przez prawie dwa lata. (In) ‘There we have through-spent through almost two years.’ (27) … no i na tym strychu przesiedzieliǧmyp przez dziewiNjǀ dni. (In) ‘… well, and in that attic we through-spent through nine days.’ (28) Niewinnie jednak przesiedzielip przez kilka miesiNjcy w wiNjzieniu ǧledczym. (In) ‘Even though there were not guilty they through-spent through several months in custody.’ (29) WiNjc do dzie’a. Wcisnƾ’em “play” parNj minut po 20:00. W fotelu bez najmniejszego ruchu przesiedzia’emp przez 46 minut. Dok’adnie w momencie gdy skoǝczy’ siNj ostatni kawa’ek, z dok’adnoǧciƾ do sekundy, zegarynka og’osi’a, ǻe “wybi’a dwudziesta pierwsza”. (In) ‘And so on to the piece (of music). I pressed “play” a few minutes after 8PM. In my armchair without the slightest movement I through-spent through 46 minutes. Exactly at the moment the last fragment finished, accurate to the second, the clock announced that “it had shown 9 PM”.’ In comparable non-temporal predicates we can recognise a landmark that is, as it were, doubly marked for the trajector. 22 The course of the trajector certainly crosses the border that marks the end of the landmark and takes away the ambiguity of the construction przeszliǧmyp ulicNjA Kochanowskiego, which can mean: 1) ‘we have crossed Kochanowski Street’ or 2) ‘we walked past / we missed Kochanowski Street’. The version przeszliǧmyp przez ulicNj Kochanowskiego usually only means ‘we have crossed Kochanowski street’. The lack of this ambiguity in perdurative contexts probably accounts for the fact that examples with a przez-complement are so rare. When this double marking for perlativity (‘passage across’) takes place in perdurative sentences it is often in contexts where the word prawie ‘almost’ occurs and where there is talk of sitting in prison or in hiding. The use of prawie, such as in (26) interferes with the perspective of the message conveyed, which is compensated by the use of this more explicit prepositional complement rather than NPA|G. This provides, then, a heightened sense of the features concerning the passage through the landmark and its boundaries. In (25), (27) and (28) the length of the duration and so the 22

Compare in this respect footnote 17 about the spatial przez-complements occurring with prze-compounds.

122

RENÉ GENIS

specification of the period involved is highlighted. That is also particularly apparent in (29) where a very strong sense of the importance of the time frame is conveyed, especially as one considers the slightly broader context provided. That time frame is actually linked to the duration of the piece of music, and it is commented on in quite precise terms. The przez-complement also figures in a different way in perdurative situations. Compare the following example. (30) [Synek …] przespa’p przez pierwszy tydzieǝ wszystkie noceA. (In) ‘My son through-slept all nights through the first week.’ We are dealing here with a combination of przez NPA and NPA|G. In this kind of structure the NPA|G is the true object of the verb (the son slept through the nights) and przez NPA is an additional adjunct (the son did not sleep through the whole week). A good English translation might be ‘During the first week my son slept through all nights’, which also has the perdurative notion clearly situated with the nights and not the week. Perhaps further research should be done to establish the exact acceptability of the przez-complement in perdurative structures. On the basis of my native speakers’ adamancy against it, the failure to detect such usage in text corpora and other “official” sources except the internet and a single dictionary example, the structure can be qualified as quite rare. 2.2.3 Prepositional complements: od + NPG do + NPG and z + NPG The structure in example (31), based on Wróbel (1984: 474), is of a somewhat more complex nature. In its full form it posits the initial as well as the final border quite explicitly. (31) Przeb’ƾka’p siNj od rana do wieczora. (Wr) ‘He has through-roamed from morning to evening.’ Example (8) from the original set of examples from Agrell quoted at the beginning of this paper also fits this structure, and its initial border is obvious although not mentioned, as I pointed out in 1. above. The next example seems at first to suggest final border vagueness. (32) Ten dƾb przetrwa’p od czasu wyleczenia z tej choroby. (In) ‘That oak tree has through-lasted from the time it recovered from that disease.’ In fact, this is an elliptic construction, and the suggestion aǻ do dzisiaj ‘until this day’ (or such like, always referring, though, to the moment of utterance) is definitely and clearly felt by native informants. Some informants felt it so compel-

PERDURATIVITY IN POLISH – 100 YEARS AFTER AGRELL

123

ling that it should have been mentioned. Others understood it anyway. This points to the fact that the final border is implied very strongly by the perdurative prze-verb and this is not surprising in view of the invariant meaning of prze- presented above in 2.1. A similar case is presented by the following example with preposition z: (33) ƺe takie miasto istnia’o, ǧwiadczƾ teǻ historyczne dokumenty, które z tych czasów przetrwa’y. (Do) ‘That such a city existed is also attested by historic documents, which through-lasted from those times.’ So, with a perdurative verb in the past tense it suffices to have the beginning of the landmark set by a temporal complement; the end can be deduced from the context and can be identified as I indicated. This seems to work as such for at least some native speakers. The formula z + NPG has a slightly different meaning and conveys clearly that the documents in (33) were made in the period under discussion and lasted from then onward. The group od + NPG is less precise and could in this context have meant ‘from that period onward’ (but not necessarily ‘from within that period onward’). Although it is quite feasible that z + NPG could also be extended to z + NPG do + NPG, no such examples have been found. Variant forms of the first formula are (aǻ) do + NPG as in example (8) and od + NPG (do + “now”) as in (32). To the former I have added the optional emphatic particle aǻ in between brackets as it crops up quite frequently in examples of this structure. It can also easily be used in the other variants of the formula, although it occurs less frequently. 2.2.4 The prepositional complement: (o)ko’o + NPG A somewhat more vague postulation of the temporal landmark, and especially also its limits, is presented in examples such as the following: (34) Dzwoni’ do mnie wczoraj wieczorem, przegadalismyp oko’o 20 minut, ale to i tak nie wystarcza … (In) ‘He phoned me yesterday evening, we through-chatted about 20 minutes, but that anyway is not enough …’ (35) Z zarzutów zosta’ oczyszczony, niemniej przesiedzia’p oko’o dwóch miesNjcy w wiNjzieniach. (In) ‘He was cleared of all accusations, he nevertheless through-spent about two months in [various] prisons.’ There is no good reason to exclude complements expressing such vagueness from perdurativity.

124

RENÉ GENIS

2.2.5 The prepositional complement: po + NPA Like (o)ko’o + NPG the next encountered construction can be considered an alternative to the NPA|G compliment. Where (o)ko’o + NPA presents the borders of the temporal landmark as vague, this complement provides an additional sense of distributivity. As examples are really quite rare and usually somewhat convoluted, I constructed the following examples and had them vetted by several native informants, who all acknowledged their comprehensibility and acceptability. (36) Przepracowa’emp po 2 dwie godziny dziennie. (Ni) ‘I have worked two hours every day.’ (37) Przez ca’e ǻycie przepracowa’emp po 12 godzin dziennie. (Ni) ‘Throughout my life I have worked 12 hours each day.’ In these examples the perdurativity is clear to each native speaker. The kind of context is quite specific, though, and as these samples are not gathered but made up, perhaps some further remarks of native informants should be noted. They indicated that (36) could be uttered in a situation where someone is explaining how he distributed the total amount of hours that were given as a task over the days that were spent working. Example (37) points to a situation in which the speaker is thinking of the total amount of hours worked throughout his life. 23 Whichever way one looks at these matters, po + NPA should be reckoned with the complements that function in perdurative predicates. 2.2.6 Particle complements such as: z / blisko / ponad /etc. + NPA Continuing our quest for other possible perdurative temporal complements we turn to Polaǝski (1980-1992 III: 445), who gives the following example in his entry for przedrzemaǀ: (38) PrzedrzemiNjp siNj z pó’A godziny, a potem pójdziemy oboje na spacer. (Po) ‘I’ll through-nap some half an hour, and afterwards we will go for a walk together.’ Compared to (o)ko’o + NPG, this construction, z + NPA, expresses an even more vague specification of the duration, crossed, passed over by the action expressed by the base-verb. Other than this element of vagueness the principle of an amount of time is still present. This construction too should be considered to be perdurative as it clearly expresses duration and fills the place connoted by the

23

Some Poles came up with a context of having to justify certain pension claims.

PERDURATIVITY IN POLISH – 100 YEARS AFTER AGRELL

125

perdurative verb. It is sometimes found in more complex structures such as in this example from the Internet: (39) W ogóle to mój ma’y wczoraj w sumie moǻe z 2 godzinki przespa’p przez ca’y dzieǝ, i to jeszcze budzi’ siNj co 5 min. (In) ‘On the whole, my little one yesterday all in all maybe through-slept some two hours throughout the whole day, and on top of that he woke every five minutes.’ In this example, we see another combination of przez + NPA with a further temporal complement, in this case z + NPA. Again, as in (30), the przez + NPA complement is not the temporal complement of perdurativity, but it gives the time frame against which the real focus of perdurativity is presented. The perdurative complement in this kind of structure is the totality of the separate bits of action expressed by the base-verb during the time frame presented by the przez + NPA adjunct. The following examples are of similar complements, in which the particles add different shades of meaning to the temporal landmark. (40) Zjad’yǧmy razem przygotowany przez niƾ obiad, wypi’yǧmy kawNj i przegada’yǧmyp blisko trzy godziny. (In) ‘We ate dinner made by her, drank coffee and we through-chatted nearly three hours.’ (41) Zakopany w ziemi, przetrwa’p ponad szeǧǀdziesiƾt lat. (In) ‘Buried in the ground, he through-lasted more than sixty years.’ In Polish there are more of these particles, such as nieomal, niemalǻe, bez ma’a ‘almost’ etc. and it is not impossible that these would be able to function in perdurative structures. However, none were found. 2.2.7 Adverbs of time In the following examples we see that an adverb of time can also fill the connoted place for a temporal complement in perdurative structures. (42) We wnNjtrzu jego, w izbie duǻej, na bardzo prostym drewnianym sprzNjcie Hersz przesiedzia’p d’ugo z chmurnym wzrokiem i z czo’em w d’oniach ukrytym.24 ‘Inside, in a large room, on a very simple, wooden piece of furniture, Hersz through-spent a long time with a clouded gaze and with his forehead hidden in his hands.’ 24

From: Eliza Orzeszkowa, Meir Ezofowicz.

126

RENÉ GENIS

(43) Piszƾc w/w tekst przesiedzia’amp d’ugo nad zapisami rozporzƾdzeǝ. (In) ‘Writing the above text I through-spent a long time [working] on the instruction notes.’ (44) … przegada’yp d’ugo co z tym zrobiǀ. (In) ‘… they through-chatted a long time [about] what to do with that.’ (45) … ǻo’nierzy Armii Krajowej, którzy zostali straceni w tych najgorszych latach piNjǀdziesiƾtych lub przesiedzielip d’ugo w wiNjzieniach. (In) ‘… The soldiers of the AK, who were lost in those horrible years of the fifties or through-spent a long time in prison.’ Here are further examples showing that not only d’ugo ‘long’ can be used. (46) … przesiedzialp krotko na Lubiance. (In) ‘… he through-spent a short time in Lubianka prison.’ (47) Przeczekaliǧmyp trochNj w trzcinach, po czym pop’ynNjliǧmy dalej. (In) ‘We waited a little in the reeds, after which we swam on.’ Examples such as these are not particularly frequent, but do occur, especially with the almost idiomatically used verb przesiedzieǀ ‘through-spend’. Those internet and text corpora examples that I have been able to study often (but not always) include a further specifying element after the temporal adverb. As the examples (42) through (47) demonstrate, these contexts can have various structures but always relate in some way or another to the time spent. They further characterise the temporal landmark. In (42) the narrative points to the temporally bounded period being characterised by Hersz’s particular posture: “a clouded gaze and with his forehead hidden in his hands”. The temporal duration in (43) coincides with the work done on the instruction notes. In (45) and (46) it is quite possible that we are dealing with the more idiomatic phraseology concerning ‘doing time in prison’, in which the time spent is, of course, quite characteristic and always perceived as having limits. 2.2.8 Rejected structures The complement structures presented above (with the possible exception of przez + NPA) all appear to satisfy the requirements of perdurative prze-compounds that the boundaries of a temporal landmark must be specified. Examples of these structures can be found in dictionaries (e.g. Doroszewski 19581969, Dunaj 1996, Baǝko 2000, Polaǝski 1980-1992), literature as well as on the Internet. This is not the case with other seemingly likely complements that meet the apparently minimal requirement of positing at least the final border of the temporal landmark. Here are some of these rejected examples.

PERDURATIVITY IN POLISH – 100 YEARS AFTER AGRELL

127

(48) *Przegada’yǧmyp z Nadiƾ, dopóki jej Mama (nie) wroci’a do domu. (Ni) ‘We have through-chatted with Nadia until her Mama returned home.’ (49) *Przegada’yǧmyp z Nadiƾ, póki nie by’a zmNjczona. (Ni) ‘We have through-chatted with Nadia as long as she was not tired.’ (50) *Przegada’yǧmyp, aǻ Ania zasnNj’a. (Ni) ‘We chatted until Ania fell asleep.’ It is actually rather surprising that Poles reject these, especially as one realises that similar structures in Russian (employing the conjunction poka ne) are quite possible.25 A further complement that does not seem to function in perdurative structures is the noun phrase in the instrumental case: NPI. (51) *Przegada’yǧmyp ca’ym wieczoremI. (Ni) ‘We have through-chatted the/a whole evening.’ (52) *Przegada’yǧmyp ca’ymi wieczoramiI. (Ni) ‘We have through-chatted (the) whole evenings.’ These needed to be considered on account of the possibility to use such complements in spatial predicates: DolinƾI jej przechodzii droga prowadzƾca przez Ba’kany do Kazanliku ‘Along the valley runs a road that leads through the Balkans to Kazanlik’.26 Admittedly, the verb in this example is imperfective and this probably has some bearing. Cf. section 3.1 examples (74) and (75). 2.2.9 Concluding remarks Summing up, it would seem right to confirm that the perdurative prze-compound connotes a temporal complement obligatorily. It is not sensible to speak of perdurativity in cases such as presented in (5) and (10), as these are not clear and unequivocal as to a temporal meaning. Here is an overview of the complements I have found possible and discussed above.

25 26

I am grateful to Adriaan Barentsen for pointing this out to me. From: Teodor Tomasz JeĪ, W zaraniu.

128 complement NPA|G (przez + NPA (od + NPG)(aǻ) do + NPG (o)ko’o + NPG po + NPA z / blisko / ponad /etc. + NPA adverb of time (+ …)

RENÉ GENIS

highlights (none; neutral perdurativity.) specific attention, perspective) (initial and) final boundary. vagueness distribution. boundary/landmark further specified by particle characterisation of duration per se (often related to a further description.)

The complement przez + NPA is included in this overview in spite of its being quite uncommon. There are at least two verbs that were considered perdurative by Agrell, but which I would like to exclude on account of the fact that they do not connote a temporal complement obligatorily. They are the verbs przenocowaǀ ‘through-stay the / a night’ and przezimowaǀ ‘through-stay the / a winter’. The content of these verbs, visible in their morphological derivation (noc ‘night’ and zima ‘winter’), sets the temporal landmark by itself.27 2.3 Base-verbs I already pointed out in 1. that Sigurd Agrell does not actually give a clear-cut characterisation or classification of the simplexes that form perduratives with the prefix prze-. Let us have a look at what other authors have to say about this. Piernikarski (1969: 119) and Grochowska (1979: 70) both discuss base-verbs that form perdurative prze-compounds. 28 They each provide a good number of examples, which are gathered on the following list: 29 balowaǀi ‘dance, party’,

27

In Russian, the cognates of these verbs do not have prefix ԗӢӹ-, which in that language is the usual prefix for perdurativity. Compare: ԗӾӢӾӰӹԓӾӴԂԽӸ ‘spend the night’ and ԗӾӢӾԒӠԐӹӴԂԽӸ ‘spend the winter’. I would like to express my gratitude to Adriaan Barentsen for pointing this out to me. Compare also footnote 13. 28 Interestingly, Bogus’awski (1963: 117), writing on Russian, states that there is an enormous amount of verbs with perdurative formational prefix ԗӢӹ-, whilst Grochowska (1979: 70) states that Doroszewski contains more than 120. This might be a reason to doubt whether Grochowska and Bogus’awski are thinking along the same lines about perdurativity, although neither is specific enough to determine this. It may also simply be that Russian and Polish differ in this respect. In any case, since the formation of perdurative compounds is a productive process, there is no point in counting. 29 I have added the following markers to the verbs on this list; none of these matters have much bearing on the present discussion, but they will come in handy later on:

PERDURATIVITY IN POLISH – 100 YEARS AFTER AGRELL

129

ba’aganiǀi ‘mess, waste’, ba’akaǀi ‘chat’, baraszkowaǀi ‘frolic, gambol, caper, romp, waste’, bar’oǻyǀi ‘litter’, ?¿bƾblowaǀi, biedowaǀi ‘live in poverty’, ?!bumblowaǀi ‘waste’, !chorowaǀi ‘be ill’, ¿chrapaǀi ‘snore’, cierpieǀi ‘suffer’, !czekaǀi ‘wait’, *czuwaǀi ‘feel’, drzemaǀi ‘nap’, dumaǀi ‘ponder’, dyskutowaǀi ‘discuss’, ¿fantazjowaǀi ‘fantasise’, *frymarczyǀi ‘barter’, fiukaǀi ‘whistle’, ?!*fujarzyǀi ‘(waste through inadequacy)’, !gadaǀi ‘chat’, gawNjdziǀi ‘chat’, ¿gderaǀi ‘grumble’, gwarzyǀi ‘chat’, gospodarowaǀi ‘run a farm, household’, graǀi 30 ‘play’, gwizdaǀi ‘whistle’, harowaǀi ‘toil’, hulaǀi ‘make merry’, hultaiǀi ‘squander’, jNjczeǀi ‘groan’, klNjczeǀi ‘kneel’, koczowaǀi ‘tramp, live rough’, ¿kontemplowaǀi ‘contemplate’, leǻeǀi ‘lie’, medytowaǀi ‘meditate’, !mieszkaǀi ‘live, inhabit’, ¿mitrNjǻyǀi ‘dawdle’, nocowaǀi ‘spend the night’, *patrzyǀi ‘look (at)’, !p’akaǀi ‘cry’, !pracowaǀi ‘work’, próǻniaczyǀi 31 ‘waste time’, próǻnowaǀi ‘waste time’, robiǀi ‘execute work’, 32 !siedzieǀi ‘sit’, !* skakaǀi ‘jump’, !spaǀi ‘sleep’, staǀi ‘stand’, stNjkaǀi ‘sigh’, strzelaǀi ‘shoot’, szaleǀi ‘make merry’, ǧmiaǀi ‘laugh’, ǧniǀi ‘dream’, !ǧpiewaǀi ‘sing’, ?taǝcowaǀi ‘dance’, taǝczyǀi ‘dance’, tNjskniǀi ‘long’, trwaǀi ‘last, endure’, wa’Njsaǀi 33 ‘loiter’, wa’koniǀi 34 ‘maroon’, wNjdrowaǀi ‘roam’, !ǻyǀi ‘live’. Although Grochowska in particular lists some semantic features apparent in these verbs, she also states that to almost all of these there are quite some exceptions. For our purposes then, her “classes”, lacking in relevance, are not so very useful. However, I will list the verbs she has found not to form perduratives, as they will come in handy later: byǀ ‘be’, znajdowaǀi siNj ‘be located’, tkwiǀi ‘be located’, drNjczyǀi siNj ‘worry’, rozpaczaǀi ‘despair’, denerwowaǀi siNj ‘be angry’, wǧciekaǀi siNj ‘be angry’, narzekaǀi ‘complain’, skarǻyǀi siNj ‘complain’, pasjonowaǀi −

?

(a small superscript question mark): the simplex verbs Piernikarski mentions here does not appear in major dictionaries such as Doroszewski. When the prze-compound is noted by Doroszewski, the simplex meaning provided here is derived from that. − ¿ (an inverted superscript question mark): the suggested prze-compound does not appear in the dictionaries. − When neither of the above forms appears I have left out the translation. − ! (a small superscript exclamation mark): the perdurative prze-compounds is accompanied by a secondary imperfective with perdurative meaning by Doroszewski. (The imperfective aspect of perdurative compounds will be discussed in sections 3 through 3.2 below.) − (asterisk): although the prze-compound is included in Doroszewski, the perdurative meaning is not provided there. 30 Doroszewski’s provides the secondary imperfective przegrywaǀ in the entry for przegraǀ, but specifically states that this imperfective form does not share the temporal (perdurative) meaning ‘spend time playing’ with the perfective. 31 Doroszewski refers to przepróǻnowaǀ, which means the same but is more common. 32 Please note the rather specific translation; it is this specific meaning that is intended by Piernikarski (1969: 120) rather than the more general one ‘do, make’ for this verb. 33 The simplex only occurs with reflexive pronoun siNj. 34 The simplex only occurs with reflexive pronoun siNj.

130

RENÉ GENIS

siNj ‘be passionate’, broiǀi ‘frolic, romp, gambol’, krzyczeǀi ‘shout’, jƾkaǀi siNj ‘stammer’, protestowaǀi ‘protest’, rozumowaǀi ‘reason’, pamiNjtaǀi ‘remember’, sƾdziǀi ‘judge’, mniemaǀi ‘suppose, imagine’. There is a class of verbs with peculiarities that warrant a little more attention. Grochowska (1979: 70) and Piernikarski (1969: 117, 120) both point out that only simplex verbs of motion that are undefined as to direction can be base-verbs for perdurative compounds. 35 The “classic” set of verbs of motion have two imperfective forms, one for expressing defined direction, the other for expressing undefined direction, e.g. respectively: iǧǀi ~ chodziǀi ‘go, walk’, jechaǀi ~ jeǺdziǀi ‘go, drive’, lecieǀi ~ lataǀi ‘go, fly’, etc. The directionally undefined imperfective forms of these verbs form perfective compounds with prefixes such as prze- (przechodziǀp, przejeǺdziǀp, przelataǀp 36 , etc.) and po- (pochodziǀp, pojeǺdziǀp, polataǀp, etc.), which are then used, among others, to express temporal meanings such as perdurativity and delimitativity respectively.37 The directionally defined simplex verbs of motion form perfectives with all kinds of prefixes and also with prze- (przejǧǀp, przejechaǀp, przelecieǀp) and po- (pójǧǀp, pojechaǀp, polecieǀp). These are not usually mentioned as verbs to express temporal meanings such as perdurativity and delimitativity, and according to Piernikarski (1969: 120) they actually specialised in denoting a measure of space. Compare: przejeǺdziǀp dwie godzinyA : *przejechaǀp dwie godzinyA ‘drive for two hours’ *przejeǺdziǀp ca’ƾ PolskNjA : przejechaǀp ca’ƾ PolskNjA ‘drive (straight) through the whole of Poland’38 35 As both Piernikarski (1969: 120) and Grochowska (1979: 70) point out implicitly, all verbs of motion, whether they belong to the “classic” set or not, potentially form perdurative compounds. It is only the verbs of the “classic” set that pose the morphologic problems dealt with in this section. 36 This particular verb is only mentioned in Doroszewski (in passing), but is quite prominent in Ʀmiech (1986: 60). Interestingly, the delimitative polataǀ ‘fly’ appears in all major dictionaries. 37 Examples such as przejeǺdziliǧmyp ca’ƾ benzynNjA ‘we drove through all our petrol’ (i.e. ‘we went as far as we could on the amount of petrol that was at our disposal’) and even more so PrzejeǺdzi’p calƾ KotlinNjA K’odzkƾ ‘He drove through the whole valley of K’odzko’, show that their correct use includes non-directional spatial perfective events. For this reason, examples such as Mój taunus 2.3 przejeǺdzi’p przesz’o 60 tyǧ km ‘my Taunus 2.3 got through more than 60,000 km’, occur with some frequency on the Internet. This last example was labelled incorrect by some native informants, who would have preferred przejecha’p. It would seem, then, that there is a grey area between real temporally intended and interpretable structures and nondirectional perfective structures such as these, making matters at times rather cumbersome to deal with for the linguist. Cf. footnotes 15 and 17 and Genis (2008: 42, 43). 38 Cf. footnotes 15 and 37 and Genis (2008: 42, 43).

PERDURATIVITY IN POLISH – 100 YEARS AFTER AGRELL

131

The morphological side of this matter is complicated by the fact that the directionally defined derivatives come in pairs with secondary imperfectives that are very similar and often homonymous with the perdurative and delimitative compounds. The following table presents some sample verb sets and their formations with prefix prze-. Homonymous forms are highlighted and the categories have been named indef(inite) dir(ection) and def(inite) dir(ection) for both simplex and compound verbs. simplex imperfective indef. dir. def. dir. biegaǀi chodziǀi jeǺdziǀi lataǀi ’aziǀi nosiǀi p’ywaǀi

bieci biegnƾǀi iǧǀi jechaǀi lecieǀi leǺǀi nieǧǀi p’ynƾǀi

compound perfective indef. dir. def. dir. przebiegaǀp przechodziǀp przejeǺdziǀp przelataǀp prze’aziǀp przenosiǀp przep’ywaǀp

przebiecp przebiegnƾǀp przejǧǀp przejechaǀp przelecieǀp przeleǺǀp przenieǧǀp przep’ynƾǀp

sec. imperf. przebiegaǀi przechodziǀi przejeǻdǻaǀi przelatywaǀi prze’aziǀi przenosiǀi przep’ywaǀi

Ʀmiech (1986: 60), however, produces the following example: (53) Ca’ƾ nocA przejecha’p w t’oku. (Ʀm) ‘Throughout the whole night he travelled/drove in a crowd.’ He explicates that sometimes also prze-compounds with base-verbs with defined directionality form constructions denoting time. Native informants approved and even preferred the following version over that of Ʀmiech when undefined direction is intended. (54) Ca’ƾ nocA przejeǺdzi’p w t’oku. (Ni) ‘Throughout the whole night he travelled/drove in a crowd.’ Ʀmiech’s version clearly combines a verb for defined direction with an accusative temporal complement. In my view this example ought also to be classed perdurative; the added notion of defined direction (referring to space) is in this respect of no consequence since the conveyed notion of ‘passage through’ is clearly there and has its bearing on the temporal object. It is true to say, though, that the directional definiteness is an extra notion that is lacking in the nondirectional variant (54). So far, we have seen lists of verbs but no real definition of the characteristics of base-verbs that form perdurative compounds. It is important to point out

132

RENÉ GENIS

though, that the formation of perdurative compounds is a productive process, and so, we must use these lists, as well as the dictionaries they are based on, only as a starting point in our search for the underlying principles. Barentsen (1995: 11, 2003: 381) provides a very clear, first restriction to candidate base-verbs for perdurativity (as well as in fact delimitativity) for Russian. He states that only aterminative verbs qualify. However, not all aterminative verbs form perdurative compounds, as can be surmised from Grochowska’s reject list. Wróbel (1984: 474) writes that for all verbs utilised in temporal derivatives an important characteristic is that the base-verbs express “akcje rozwijajƾce siNj w czasie” ‘actions that develop in time’. “Wyk’adniki statycznych relacji omnitemporalnych” ‘expressions of static, omnitemporal accounts’ (e.g. naleǻeǀi ‘belong to’ and równaǀi siNj ‘to be equal’) are thus excluded from forming such derivatives. These verbs are classed as STATES in the Vendler (1967) classification of verbs and all of these are excluded from forming perdurative derivatives. Vendler’s verb class of ACTIVITIES, on the other hand, covers most of the accepted perduratives. 39 Actually, looking at the verbs on the lists as well as those defined with a perdurative meaning in the major dictionaries, it is quite striking that most of them refer to controllable activities. Notable exceptions are przechorowaǀp ‘be ill’, przetNjskniǀp ‘long’, przeczuwaǀp ‘feel’, przecierpieǀp ‘suffer’, and possibly also przebiedowaǀp ‘live in poverty’, przechrapaǀp ‘snore’. This observation, for what it is worth, does not provide a clear-cut solution to our problem of characterising base-verbs for perdurativity. The exceptions, however, expose meanings for which it is sensible to speak of in terms of an amount of time and duration. Chorowaǀ ‘be ill’, as well as the others mentioned just above, have this temporal relevance, wiedzieǀ ‘know’ and such state verbs, do not. Having established the minimum criteria – aterminative activities – it is possible to address Grochowska’s rejects. Indeed, at least one of the activities on this list is not actually impossible as a base for perdurative compounds. Compare in this respect: (55) Chore dziecko (mi) przekrzycza’op ca’ƾ nocA. (Ni) ‘The sick child cried throughout the whole night.’ It is true, the perdurative use of this compound is not listed in any of the major dictionaries, which just goes to show that the formation of perdurative structures is a productive process. I can add here a further reason why certain activities do not form perduratives in Polish. Other than in Russian, perdurative compounds are not formed 39

Pasich-Piasecka (1993: 24) speaks of “kinds of activities” but is a little unclear since she does not mention what kinds of activities nor does she refer to Vendler or any such classification.

PERDURATIVITY IN POLISH – 100 YEARS AFTER AGRELL

133

on the basis of imperfective prefixed verbs, which is quite as we would expect, for prefixed verbs are usually terminative.40 Moreover, perduratives are not even formed of simplex verbs with a prefix-like beginning (that historically may actually have been a prefix). 41 It would seem that some (residual) prefix meaning, and so terminativity, is still felt. Grochowska’s deselected verbs rozpaczaǀi ‘despair’, wǧciekaǀi siNj ‘be angry’, narzekaǀi ‘complain’ and skarǻyǀi siNj ‘complain’ are examples of this. Summing up then, none of the above-mentioned publications presents a clear-cut and exhaustive list of characteristics for base-verbs to form perdurative compounds. Most authors discarded directional verbs of motion in this respect although they too should be considered as base-verbs. It remains an open question whether clear-cut specifications could be drawn up, but such a list would surely feature aterminative verbs. It would include Vendler’s activities, especially controllable ones, but exclude seemingly aterminative verbs with prefix-like beginnings as well as Vendler’s states. Further research is needed to point out whether a clear set of criteria can in fact be drawn up. 3. Perdurativity and imperfectivity Most authors consider only perfective perdurative situations. This is not surprising as Agrell’s original concept was connected with Aktionsart and therefore perfectivity. As far as I am aware, only Wróbel (1984: 475) mentions imperfective perdurativity as a possibility but he refrains from expanding on this.42 Consider the following lot of examples. (56) Latem zwykle bywam zdrowszƾ, zimyA ca’e prawie przechorowujNji. (Do) ‘In summer I am usually healthier, throughout entire winters I am usually ill.’ 40 The Russian prefixed verbs that form the derivational basis for compounds such as ԗӢӹԒԂӰӠԐԂԽӸԙԋ ‘through-busy oneself’ are polysemic in this respect: they are both terminative as well as aterminative. Similar formations are absent in Polish. I am grateful to Adriaan Barentsen for this information on Russian. 41 Two examples from Doroszewski that might seem exceptions to this are przezapustowaǀ ‘through-celebrate the last days of Carnival’ and przepodróǻowaǀ ‘through-travel’. The first, a word not in general use today, can be likened to the verbs przenocowaǀ ‘spend the night’ and przezimowaǀ ‘spend the winter’, as it too does not connote a temporal complement obligatorily (cf. section 2.2.9) and is formed denominally from zapust ‘the last days of Carnival’. Przepodróǻowaǀ has the same denominal feel to it, and historically speaking, the element po was not a prefix, but a preposition that merged with droga ‘road’ in the locative case, droǻe to form the noun podróǻ ‘trip’. 42 Czarnecki (1989: 50) gives examples and even pairs of perdurative verbs, but neither mentions nor explains this.

134

RENÉ GENIS

(57) Przechorowywa’i kaǻdƾ zimNjA. (Du) ‘He was ill throughout every winter.’ (58) On czNjsto przesypiai pó’A dnia. (Du) ‘He often through-sleeps half the day.’ (59) Bywa’y tygodnie, gdy noc w noc parNjA godzin przepNjdzaliǧmyi w lesie. (Po) ‘There were weeks, when night upon night we spent a few hours in the wood.’ These examples all have in common, that they express so-called “open” (non-quantified) repetition, consisting of an unspecified number of separate units of event. This is one of the main applications of any secondary imperfective verb.43 In these examples this repetitive nature is made clear by other means than merely the secondary imperfective. Where temporal direct objects occur, they are accompanied by modifiers such as zwykle ‘usually’ (56), czNjsto ‘often’ (58) or noc w noc ‘night upon night’ (59). In (57) the accusative temporal complement itself contains the word kaǻdƾ ‘each’. In addition to adjuncts expressing repetition, direct objects can be plural as in (56), which feature, in combination with the imperfective verb, also brings out the “open” repeated nature. Let us now consider the next example. (60) Nikt nie przepracowywa’i tutaj d’uǻej niǻ dwa lataA. (Po) ‘Nobody worked here for longer than two years.’ The “open” repetition is embedded here in the speaker’s consideration of an unspecified number of different cases concerning individual employees, who each worked there for a different length of time, but none made it to two years. The ‘portion of two years of work’ repeatedly did not take place. Many secondary imperfectives can also be used to describe single event situations. The following example includes a non-perdurative secondary imperfective in such a situation, which is concurrent with the moment of speaking and processual in nature. (61) W’aǧnie w tej chwili przepisujemyi ca’y artyku’. (Ni) ‘Just at this moment we are rewriting the whole article.’ Such a reading is not possible with perdurative secondary imperfectives: 43

All imperfective perdurative verbs are secondary imperfectives. There are no perdurative imperfectiva tantum.

PERDURATIVITY IN POLISH – 100 YEARS AFTER AGRELL

135

(62) *W’asnie w tej chwili przegadujemyi calƾ nocA. (Ni) ‘Just at this moment we are through-talking the whole night.’ We have stumbled here on a universal impossibility: the duration of an event cannot be specified as it is still going on. 44 Here are some further examples to illustrate this point. (63) *Ania dwie godziny przepracowywa’ai nocA. (Ni) ‘Ania two hours through-worked the night.’ (64) *Jak zadzwoni’em Ania akurat przepracowywa’ai nocA. (Ni) ‘When I phoned, Ania was just through-working the night.’ (65) *Ania przepracowywa’ai, i przepracowywa’ai, aǻ wreszcie przepracowa’ap nocA. (Ni) ‘Ania through-worked and through-worked, until finally, she through-worked the night.’ A further use of imperfective verbs, secondary imperfectives included, is in the historic present, which native informants could form for (1), (4), (6), (7), (9), (13), (14), (23), (26) – (30), (34), (36), (37), (39), (40), (42) – (44), (46), (47), (53) and (54), albeit, that often there were more common ways of expressing the meaning intended. The major stumbling blocks in forming imperfective historic present equivalents to past tense examples is the unfamiliarity of the secondary imperfectives needed. When asked, native informants could not form historic present versions of (3), (8), (11), (12), (15) – (17), (25), (31), (32), (33), (38), (41) and (55) and they were very doubtful about (2), (24), (35), and (45). 45 Imperfective perdurativity is a rare phenomenon in Polish and so we may not expect too much, but its use does include the historic present. Perdurative secondary imperfectives are excluded from concurrent, processual, single event readings, leaving only the multiple, iterative application and the historic present.

44 In English, though, the following is quite possible: – Where is father? – He is having an hour’s sleep. This is some kind of elliptic answer, which could actually imply ‘Father went up to sleep for an hour’. Compare also: ?father is sleeping for an hour, about which native informants have doubts. There are no doubts about *Look at father. He’s just at this moment sleeping (for) an hour. 45 The remaining examples from this paper were not considered relevant.

136

RENÉ GENIS

3.1 Imperfective perdurativity and temporal complements In examples (56) through (59) we have seen accusative temporal complements in imperfective perdurative structures, whilst (60) sports an adverb of time. There are some matters we must address here. We will also look at other possible temporal complements. Again, as with perfective situations, the temporal accusative objects we found with imperfective perdurative verbs can be considered to be real direct objects, as they must be put into the genitive when sentences are negated. Such sentences are not readily available and the nature of the repeated event makes it impossible to produce direct negations of the earlier series. Here are two not directly related examples that nevertheless contain a negation. (66) Synek nawet nie przesypiai dwóchG godzin dziennie. (In) ‘My son does not even sleep through two hours a day.’ (67) Ja nie przechorowujNji zimG, a siostra przechorowuje. (Ni) ‘I am not ill throughout the winters, my sister is.’ Although (66) and (67) are not direct negations of the earlier examples, they show the grammatical point and so, for imperfective verbs too, we can from here onwards speak of NPA|G. In addition, we can say that certain transformations that can be used to establish objecthood, such as suggested by Holvoet and discussed in 2.2.1 above, are also possible with imperfective prze-compounds. Compare: (68) Iloǧǀ przesiadywanych godzin. (Ni)46 ‘The amount of through-chatted hours.’ This confirms that these NPA|G can be classed as direct objects. Example (69) shows that do + NPG also works in imperfective predicates and the adjunct dzieǝ w dzieǝ ‘day upon day’ explicates the indicated repetition. (69) Prawie dzieǝ w dzieǝ przepijaliǧmyi w gospodzie aǻ do bia’ego ǧwitu. (Po) ‘Almost day upon day we drank in the pub until the early morning.’ In the next example, (70), the adverb codziennie works together with the distributive noun phrase po 5 / 6 godzin ‘5 / 6 hours each time’ to effectuate repetition:

46

The difference between (68) and (21) is explained by native informants as: (68) is presented as ongoing, there will be more such hours, whilst (21) involves closure and there will be no more such hours.

PERDURATIVITY IN POLISH – 100 YEARS AFTER AGRELL

137

(70) Niemalǻe codziennie przegadywaliǧmyi po 5-6 godzin … (In) ‘Almost daily we chatted for about 5 to 6 hours (each day).’ Particle complements such as z + NPA can also function in imperfective perdurative situations: (71) Moja ma’a przesypiai z trzy godziny dziennie. (Ni) ‘My little daughter through-sleeps some three hours every day.’ This example was constructed and approved by native informants, who acknowledged that other particles such as ponad ‘more than’, blisko ‘nearly’ etc. must also be possible. However, none were found in examples. In example (60) we already saw an adverbial structure as temporal complement in a perdurative situation. Here are two further examples with adverbs: (72) By’ ogromnie pracowity i przesiadywa’i d’ugo w biurze. (In) ‘He was enormously diligent and through-sat/spent long in his office.’ (73) Bierze siNj to stƾd, ǻe ptak jest szarobrƾzowy, nie rzuca siNj w oczy gdyǻ przesiadujei d’ugo w jednym miejscu, a i zachowuje siNj cicho. (In) ‘That is caused by the fact, that the bird is grey-brown, he does not catch the eye as he through-sits/spends long in one place and conducts itself quietly.’ So far, all the temporal complement structures I mentioned for imperfective perdurative verbs also apply for perfectives. Now, consider the following examples: (74) … wieczorami albo siedzNj sama w pokoju i czytam albo przegadujNj godzinami z kumplem. (In) ‘… in the evenings I either sit alone in my room and read or I through-chat for hours with a friend.’ (75) Ca’ymi dniamiI u nas przesiadywa’i … na obiadach, na kolacjach. (Do) ‘For whole days he would through-sit/spend with us … during lunch, dinner.’ (76) Janka przesiadujei w pracy nawet nocamiI. (Ba) ‘Janka through-spends at work even at night.’ These stand out for having a NPI, which was rejected for use with perfective perdurative verbs such as in sentences (51) and (52) above. Here, as in (56), the plural temporal complement points to repetition. Native informants point out that the instrumental, when opposed to the accusative in these sentences, is not precise as to the duration, especially not the otherwise important final border, such as perdurative structures usually imply. The NPI implies something that

138

RENÉ GENIS

was described as ‘really almost the whole time’ and it often seems to point to habituality. All in all, the NPI must be accepted as a possible way to fill the obligatory connotation of imperfective perdurative verbs. No examples were found with przez + NPA. When such a formation occurs with a perdurative verb other than przesiadywaǀi (for which, see the next section), it was in examples such as the following: (77) … iloǧǀ godzin, jakƾA cz’owiek zgodnie z przepisami przepracowujei przez piNjǀ dni roboczych. (In) ‘… an amount of hours, which one, according to the regulations, through-works through five working days.’ Here, the perdurative object is jakƾA, which refers back to iloǧǀ godzin. The przez + NPA does not itself set the landmark but sets the background, against which the amount of hours, worked in separate instances, are usually worked. This structure can be likened to (30) and (39) above. As examples are so rare it is difficult to be absolutely unequivocal about which temporal complements can, and which cannot partake in imperfective perdurative structures. Most of the ones listed for perfective perdurativity seem to apply and NPI seems possible in habitual, imperfective situations but not in perfective contexts. Good examples of przez + NPA are lacking and native informants were also unable to produce these. We will, however, briefly return to this formation in the next section. 3.1.1 The verb przesiadywaǀi Imperfective perdurative verbs are very rare indeed. By far the most frequent is przesiadywaǀi, which, like its perfective counterpart przesiedzieǀ, means ‘through-sit/spend’. It already featured in some of the examples above and it is the only verb that I have found to occur without a temporal complement. Compare: (78) Przesiadujei tam z córkƾ bƾdǺ koleǻankami, zajada siNj lodami, pije kawNj. (In) ‘He through-sits/spends there with his daughter or with girl friends, he enjoys ice creams, drinks coffee.’ (79) … a ja tego pana widzia’am na ǻywo – przesiadujei na rynku w Krakowie … (In) ‘… I saw that man live, he through-sits/spends on the square in Cracow …’

PERDURATIVITY IN POLISH – 100 YEARS AFTER AGRELL

139

Native informants interpret this to mean ‘sit throughout the whole time / day, every time I come there’. As this does not seem to happen with other verbs, it should be classed as an isolated case. Interestingly, formations involving przesiadywaǀi and przez + NPA also occur on the Internet: (80) Nigdy do Jerozolimy nie dotar’, bo przesiadywa’i przez blisko dwadzieǧcia lat (z krótkimi przerwami) w paryskiej Bibliotece Narodowej … (In) ‘He never made it to Jerusalem, for he through-sat through nearly twenty years (with short intervals) in the Parisian National Library …’ This example can be likened to (77). There, the temporal framework put into place by the adjunct przez + NPA was described as the background, against which the multiple event takes place, but which itself is not the temporal complement of perdurativity. In (80) a similar background is posited by przez + NPA, this time for an unspecified number of occurrences of the event expressed by the verb przesiadywaǀi, without further temporal complement. 3.2 Imperfective perdurativity and verbs of motion None of the examples so far in this section have included verbs of motion. None of the dictionaries provide imperfective perdurative versions of the verbs of motion, except, possibly, Doroszewski, who notes the aspectual pair przejeǺdziǀp : przejeǻdǻaǀi ‘through-drive’. Unfortunately, examples are not provided there, nor are they available in text corpora or on the Internet. Apart from this one exception, all perdurative verbs of motion in the dictionaries are indicated as being perfective. It is difficult to ascertain which exact verb forms would apply for imperfective use, but Doroszewski’s mention of przejeǻdǻaǀi suggests that it would be the series dubbed “secondary imperfective” in the table in section 2.3. As most of these are homonyms of the directionally undefined perfective verb forms, matters become quite blurry. The only other example that might be formally distinct is przelatywaǀi ‘through-fly’. Again, no examples could be found. I confronted my Polish native informants with the following invented sentences: (81) Przejeǻdǻamyi tym autkiem dwie godzinyA dziennie, tylko dla przyjemnoǧci. (Ni) ‘We through-drive in this little car two hours every day, just for fun.’ (82) Przelatujemyi tym samolotkiem dwie godzinyA dziennie, tylko dla przyjemnoǧci. (Ni) ‘We through-fly in this little plane two hours every day, just for fun.’

140

RENÉ GENIS

These sentences did not meet with any disapproval. In fact, they are quite acceptable. To say the least, though, verbs of motion are highly uncommon in imperfective perdurative contexts, but not all of them seem to be excluded. 4. Concluding remarks In perdurative constructions the different components each play their own part. The temporal landmark is set by the temporal complement and the prefix prze- expresses the crossing of that landmark and its limits. The base-verb presents the action that by way of the trajector crosses the domain. The activity expressed by the base-verb has to include a quality that makes it sensible, meaningful to imagine that one could spend time on it. As many potentially perdurative verbs can also connote non-temporal complements (landmarks) we cannot say that these verbs connote temporal complements obligatorily. We can, however, say that with these verbs, the temporal complement is obligatory in order to bring out the (potentially) perdurative meaning of the prze-compound. The need for a specifically mentioned temporal complement is peculiar to perdurativity amongst the two durative Aktionsarten; the other, delimitativity, does not need a temporal complement in this way. E.g. Pawe’ pospacerowa’p (sobie) ‘Pawe’ walked for a while’. Here is an attempt to contrast these two Aktionsarten that both deal with temporal quantification. In delimitative structures the implied limitations, and so quantification, of the action are already expressed by the compound verb itself, and perhaps the temporal interpretation many authors on the subject give to it, is in fact secondary and a result of the action itself being presented as limited. In the view of Barentsen (1995: 11, 2003: 381) for Russian, delimitativity expresses the execution of ‘a certain amount / a portion’ of the action inherent in the base-verb, with the temporal notion only present through implication. This probably holds for Polish too. The mere fact that the delimitative verb does not connote the temporal complement in an obligatory fashion, points to the difference in perspective expressed by these forms. For perdurativity that may already be clear from the above: the temporality of the action denoted has a feeling of being externally imposed. The limits of the temporal domain / landmark, expressed by a(n external) temporal complement, are the limits of the action. Delimitativity on the other hand seems to have the limitation enforced from within: it is the amount of action that determines the quantification of the action and so implies the temporal boundaries, that do not even need to be (made) explicit. Perhaps this lies at the basis of the observation by Maslov (1959: 184) writing on Bulgarian, cited by Forsyth (1970: 24) for Russian, who both point out that delimitatives imply that the duration of time for an action or state is (perceived as) short, whereas perdurativity indicates a longer period, or one that is perceived to be

PERDURATIVITY IN POLISH – 100 YEARS AFTER AGRELL

141

long. Compare Forsyth’s two Russian examples with both kinds of verb in conjunction with temporal complements and their interpreted meanings: ԋ ԗӹԙԽӹԋԊp ԽԂԐ ӴԙӾԄӹ ӹԌӠӰ ԓԂԙA means ‘I stood there for only an hour’, whilst ԋ ԗӢӹԙԽӹԋԊp ԽԂԐ ՁӾԊՂӡ ԓԂԙA means ‘I stood there for a whole hour’. A final definition of perdurativity needs to cope with the fact that it occurs in both aspects, but that the secondary imperfective verbs can only be used in cases of open repetition (multi-event readings) or the historic present, not the single, ongoing, processual or concurrent event. Considering that of Agrell’s original Aktionsarten quite a few have been rejected or have fallen into disuse over the last 100 years, it is amazing that the notion of perdurativity as it was formulated by him has survived almost intact. Perduratywnoǧǀ przetrwa’ap aǻ do dziǧ ‘perdurativity through-lasted until today’. University of Amsterdam REFERENCES Agrell, (Per) Sigurd 1908 Aspektänderung und Aktionsartbildung beim polnischen Zeitworte: ein Beitrag zum Studium der Indogermanischen präverbia und ihrer Bedeutungsfunktionen (Lunds Universitets Årsskrift. N.F. Afd. 1. Bd 4. Nr 2.). Lund: Håkan Ohlssons Buchdruckerei. 1918 Przedrostki postaciowe czasowników polskich (Materia’y i prace Komisji jNjzykowej Akademii umiejNjtnoǧci w Krakowie 8). Kraków: Nak’adem Akademii UmiejNjtnoǧci. Baǝko, Miros’aw (ed.) 2000 Inny s’ownik jNjzyka polskiego PWN. Warszawa: Wydawnictwo Naukowe PWN. Barentsen, Adria(a)n 1995 “Trexstupenǁataja model’ invarianta soveršennogo vida v russkom jazyke”. In: Stanis’aw Karolak (ed.), Semantika i struktura slavjanskogo glagol’nogo vida 1, 1-26. Kraków: Wydawnictwo Naukowe WSP. 2003 “De betekenis van het aspect”. In: Peter Houtzagers (ed.), Russische grammatica, 369-397. Bussum: Coutinho. Bogus’awski, Andrzej 1963 Prefiksacja czasownikowa we wspó’czesnym jNjzyku rosyjskim. Wroc’aw et al.: Komitet S’owianoznawstwa Polskiej Akademii Nauk. Czarnecki, Tomasz 1989 “Die Ausdruckmittel der Perdurativität im Polnischen und Deutschen”. Zeszyty naukowe uniwersytetu jagielloǝskiego CMVII, Prace jNjzykoznawcze 94, 47-58. Doroszewski, Witold (ed.) 1958-1969 S’ownik jNjzyka polskiego I-XI. Warszawa: Polska Akademia Nauk / Paǝstwowe Wydawnictwo Naukowe.

142

RENÉ GENIS

Dunaj, Bogus’aw (ed.) 1996 S’ownik wspó’czesnego jNjzyka polskiego. Warszawa: Wilga. Faßke, H. (1979) 1981 Grammatik der obersorbischen Schriftsprache der Gegenwart. Morphologie. Bautzen: Domowina. Forsyth, J. 1970 A Grammar of Aspect – Usage and Meaning in the Russian Verb. Cambridge: University Press. Genis, René 2008 Studies on the Polish Verbal Prefix prze- (Pegasus Oost-Europese Studies 7). Amsterdam: Uitgeverij Pegasus. Grochowska, Alina 1979 “Próba opisu regu’ ’ƾczliwoǧci przedrostka prze- z tematami czasownikowymi”. Polonica V, 59-74. Grzegorczykowa, Renata 1975 Funkcje semantyczne i sk’adniowe polskich przys’ówków (Prace JNjzykoznawcze 77). Wroc’aw et al.: Polska Akademia Nauk / Komitet JNjzykoznawstwa. Holvoet, Axel 1991 Transitivity and Clause Structure in Polish. A Study in Case Marking. Warszawa: Polska Akademia Nauk / Instytut S’owianoznawstwa. Maslov, Ju.S. 1959 “Glagol’nyj vid v sovremennom bolgarskom literaturnom jazyke (znaǁenie i upotreblenie)”. In: S.B. Bernštejn (ed.), Voprosy grammatiki bolgarskogo literaturnogo jazyka, 157-307. Moskva: Izdatel’stvo Akademii nauk SSSR / Institut slavjanovedenija. Mindak, Jolanta 1987 “O polskich czasownikach typu przegadaǀ (ca’ƾ noc) widzianych oczami ba’kanisty”. In: Mieczys’aw Basaj (ed.), Slawistyczne studia jNjzykoznawcze lecie, 197-201. Wroc’aw et al.: Wydawnictwo Polskiej Akademii Nauk. Pasich-Piasecka, Agnieszka 1993 “Polysemy of the Polish Verbal Prefix prze-”. In: E. Górska (ed.), Images from the Cognitive Scene, 11-26. Kraków: Universitas. Piernikarski, Cezary 1969 Typy opozycji aspektowych czasownika polskiego na tle ogólnos’owiaǝskim (Monografie Slawistyczne 17). Wroc’aw et al.: Komitet S’owianoznawstwa Polskiej Akademii Nauk. Polaǝski, Kazimierz (ed.) 1980-1992 S’ownik syntaktyczno-generatywny czasowników polskich I-V. Wroc’aw et al.: Polska Akademia Nauk / Instytut JNjzyka Polskiego. Przybylska, Renata 2006 Schematy wyobraǻeniowe a semantyka polskich prefiksów czasownikowych do-, od-, prze-, roz-, u- (JNjzykoznawstwo kognitywne, studia i analizy). Kraków: Universitas. Ʀmiech, Witold 1986 Derywacja prefiksalna czasowników polskich. Wroc’aw et al.: Œódzkie Towarzystwo Naukowe.

PERDURATIVITY IN POLISH – 100 YEARS AFTER AGRELL Vendler, Zeno 1967

143

“Verbs and Times”. In: Z. Vendler (ed.), Linguistics in philosophy, 97-121. Ithaca, New York. (Revised version of “Verbs and Times”, Philosophical Review 46, 1957, 143-160.)

Wróbel, Henryk 1984 “S’owotwórstwo czasowników”. In: R. Gregorczykowa, R. Laskowski, H. Wróbel (eds.), Gramatyka wspó’czesnego jNjzyka polskiego. Morfologia. Warszawa: Polska Akademia Nauk / Instytut JNjzyka Polskiego.

SUMMARY The Swedish poet, slavist and runologist Per Sigurd Agrell introduced the concept of perdurativity exactly 100 years ago as one of his Aktionsarten. This paper contains an overview of the scant information on this subject gathered from extant literature. I discuss the three component parts of perdurative expressions separately: the prefix prze-, the temporal complement and the base-verb. The perdurative meaning variant of prefix przefunctions in constructions with temporal complements, of which there are more than the two usually mentioned: the accusative object and the od + NPG do + NPG construction. A list is provided of complements drawn from text corpora and internet. Although it is difficult to pinpoint exactly which simplexes can form perdurative compounds certain classes of verbs can be isolated. In my view, perdurativity is expressed only when these three component parts are present in a given structure. They are subject to various restrictions and have particular characteristics that are made explicit. Usually, only perfective perdurativity is discussed in publications on this subject. However, imperfective perdurativity also exists. On the basis of text corpus, internet and native speaker research I have been able to establish that imperfective perdurativity is used only to express terminal iterativity.

Dutch Contributions to the Fourteenth International Congress of Slavists, Ohrid: Linguistics (SSGL 34). Amsterdam – New York: Rodopi, 2008, 145-216.

VICISSITUDES OF THE GENITIVE RULE W. ANDRIES VAN HELDEN

Case selection for the direct object of negated verbs in Russian (henceforth referred to as GENACCNEG) is a popular topic of linguistic reflection. Corbett’s 1986 bibliography of works devoted to it contains as many as 106 items. Some dozens of recent papers and monographs can meanwhile be added. If we include the general grammars of Russian, which usually contain a few pages on the subject, the number of publications containing original contributions on GENACCNEG must by now be well over two hundred. The present article does not add fresh data or novel linguistic theories to this vast body of results. Instead, the phenomenon is looked at from the outside. It is attempted to provide a consistent and plausible account of the interaction between scholarly reflection on GENACCNEG and available data on spoken and written usage between 1600 and the first quarter of the 20th century. In the account, which is preceded by a brief sketch of current approaches to GENACCNEG, it will be claimed that it is not necessary to resort to the assumption of linguistic change having taken place over the inspected period to account for the available scholarly statements and data. The article concludes with a discussion of the question why teachers and writers tolerated the difference between spoken and written practice over the investigated period. 1. An overview of approaches to GENACCNEG 1.1 Theoretical linguistics Linguistics as a core discipline has produced both synchronic and diachronic explanations for GENACCNEG usage in Russian. Recent synchronic approaches include semantic explanations (e.g. Meintema 1986, Ueda 1992: 4-128) 1 as well as syntactic ones (e.g. Bailyn 1997, Brown 1999) and aim to account for case choice in terms of unified principles determining case choice, viz. semantic invariants or syntactic rules, which are regarded as

1

As space does not permit a serious enumeration of the relevant available literature, only one or two examples of each approach are provided.

146

W. ANDRIES VAN HELDEN

intrinsic to the language as an autonomous organism. In some approaches (e.g. Babby 1986), semantic and syntactic explanations are complementary. 1.2 Diachronic linguistics Many linguists regard straightforward syntactic and semantic accounts of GENACCNEG usage as unsatisfactory. They doubt the possibility of providing a consistent linguistic explanation for the present pattern and believe that a diachronic explanation is the best linguistics can offer. According to the standard diachronic scenario, the genitive was at some point in the past the only possible case to be used to mark the direct object of a negated verb and is presently making way for the accusative (e.g. Comrie et al. 1996: 309; Benigni 2002: 49) in a process of linguistic change that is likely to result in its eventual fossilization, as in Czech (Timberlake ((1975)1986: 354). In some approaches, this diachronic scenario is set in a synchronic framework. Timberlake ((1975)1986: 357), for example, regards the present oscillation between the genitive and the accusative as a temporary, transitional stage between two stable states of the language, viz. the state in which the syntax of the language contains a rule changing an underlying accusative into a genitive in sentences containing a negation, and a state in which this rule no longer exists. The time-scale of the process of linguistic change entails a complication for any diachronic explanation of GENACCNEG: how must case use be accounted for while the change is going on? This would have been a minor detail if the transitional period lasted for a generation or so but analyses of historic sources prompt the assumption that the GENACCNEG change has been taking much longer. Save for a minority of linguists who question the standard scenario (see 8.3), historical linguists (e.g. Borkovskij 1978: 347) believe that the genitive was obligatory, or at least used exclusively, in Common Slavic. But accusatives turn up in even early Old Church Slavonic texts. While they used to be accounted for by underlying Greek or Latin originals (e.g. Buslaev (1858)18754: 249), they are presently assumed to be spontaneous: Veǁerka (1963: 204-209), after examining individual instances of accusative use in Old Church Slavonic, presents a set of semantic and syntactic conditions for the use of the accusative that look like a subset of the conditions that account for present-day Russian usage (e.g. in Mustajoki and Heino 1991). Veǁerka observes, furthermore, that the accusative occurs more often in Old Church Slavonic than in the oldest sources in West and East Slavic (1963: 209) and refers to the general tendency of South Slavic to use the genitive less often than in the northern branches of Slavic. He attributes this tendency to the mixed linguistic environment of the Slavic speaking population in the Balkans besides a more intensive exposure to Greek texts.

VICISSITUDES OF THE GENITIVE RULE

147

In early East Slavic sources, in fact, direct objects of negated verbs are always in the genitive (Morozova 1968: 152). Except for one or two very specific circumstances, this situation is also found in the Novgorod birch bark texts (Zaliznjak 20042: 159). Accusatives started occurring occasionally in 15th-century Russian texts (Xarpaleva 1985: 102) and seem to have gained a bridgehead in 17th century (Borkovskij 1978: 347-348). If such observations are taken seriously, it must be assumed that the transitional period during which the accusative has been ousting the genitive commenced several centuries ago. Explanations that fail to account for GENACCNEG usage during this long intermediate stage in synchronic terms may then be regarded as incomplete or unsatisfactory. Timberlake’s syntactic explanation cited above requires a plausible (biological, psychological or sociological) scenario that accounts for the transmission of a damaged syntactic rule over many generations. 1.3 Eclectic linguistics The fact that the change was not yet completed when linguistics developed as a discipline added to the topicality of the issue. In the second half of the 20th century, as non-native professionals started learning Russian, GENACCNEG became a popular topic among slavists. They produced a host of publications on the possibility of providing reliable guidelines for case choice (starting with Magner 1955 and culminating in Bogus’awski 1998). Soviet structuralists joined the slavists in these efforts when machines started being taught Russian (e.g. Raviǁ 1971). The proposed guidelines are mostly eclectic: the syntactic and semantic factors as furnished by core linguistics are supplemented with lexical, phraseological and stylistic as well as contextual, pragmatic and sociolinguistic factors, so as to produce multifactor algorithms emulating “correct” case usage in as many instances as possible. The usual bottom line of such efforts is that no factor or combination of factors provides a perfect fit for actual usage, as a great deal of variation can only be ascribed to incidental factors and individual speakers’ habits. Efforts to pinpoint, or neutralize, such effects have given rise to a number of quantitative surveys of actual usage (e.g. Safarewiczowa 1959, 1960, Restan 1960). Quantitative research on GENACCNEG has meanwhile reached a considerable level of sophistication, especially since Mustajoki and Heino’s 1991 project. 1.4 Prescriptive linguistics In Russia, a considerable part of the literature deals with GENACCNEG in a prescriptive setting. The Academy grammars, various spravoǁniki and other sources on codification and normalization of standard usage (e.g. Ickoviǁ 1982)

148

W. ANDRIES VAN HELDEN

that have appeared in the second half of the 20th century, usually contain a few pages on GENACCNEG. The recommendations for case choice found here are even more eclectic in that they include considerations of all categories identified in descriptive analyses but also ad hoc considerations, such as reading convenience. 1.5 Written vs. spoken Russian Whatever approach is adopted, all sources agree on one general tendency in GENACCNEG usage: the genitive occurs much more frequently in text (and spoken discourse that is associated with it) than in speech (and written discourse that reflects it). The divergence between speech and writing is usually explained by the inherent conservatism characterizing a written code. In the standard view on linguistic change, the spoken mode is the natural domain of genuine language change. The written standard is regarded as a frozen early stage of the spoken language: it accommodates changes occurring in the spoken language reluctantly and with a considerable delay. Timberlake illustrates this for GENACCNEG pointing to the increasing tolerance of the accusative in normative grammar: “The loss of the genitive of negation is also evident from the increasing tolerance of accusatives in grammars. In his [18th-century] Rossijskaja grammatika, Lomonosov [((1757)1952: 561)] admits no exceptions to the genitive of negation. [In the 19th century,] Vostokov (1831: 256) allows the accusative for infinitives and questions. Finally, the [20th-century] Academy Grammar [(Vinogradov and Istrina 1954: 562-563)] allows the accusative under various conditions […].” (Timberlake (1975)1986: 359) Meanwhile, the spoken language is reported to be moving on, always ahead of the written norm, eliminating the genitive in more and more positions (e.g. Bocale 2004: 29-30). For the standard view on the relationship between the written norm and spoken usage to hold good for GENACCNEG, two conditions must be met: 1. the standard language reflected spoken usage when it was fixed; 2. the observed relationship between written and spoken usage cannot be explained otherwise. The evidence presented in the following sections suggests that these conditions are not fulfilled.

VICISSITUDES OF THE GENITIVE RULE

149

2. The genitive rule and the normalization of Russian Before going into the codification of GENACCNEG in the Russian standard language, it is important to note an important distinction: the observation that the genitive is used to mark the direct object of a negated verb is not identical to the rule (henceforth referred to as the genitive rule) that the genitive be used to mark the direct object of a negated verb. The genitive rule, regardless of its ontological status (e.g. syntactic or normative), may be submitted as an explanation for the corresponding observation. Without additional evidence, however, it is not necessarily the only possible explanation: there may be broader factors governing the use of the genitive, the exclusive or predominant use of the genitive for the direct object of a negated verb being just a side effect. Veǁerka (1963: 204) seems to assume that such a situation existed in older stages of Slavic when he states that: “[ӊ ҡҲҽҝҸҡґҽӊҾӁҡҔҸѸ] ҾӉӃҔѦ ҝҸҦѹҲѦґҧӁӃҮ Ұҝѹ ҊґҽҊҸґҽѽ ҡ ҸҲҝѹ҃ҽӁѹѦѸ ҰҝѦҦҡҲҽӊґҾѦҲ ҏңѦ ҡӊҸѦҊҸ ҝҸҦҽ “ҡѽѦѸҏ ҝѦҍѹ, ҾӉӃҔҸӊҏӀ ӍҸҝѸҏґҏ”, ҸѽӊҽҲӃӊҽӀӂҏӀ ҡӍѦҝҏ ѹҡҔҸӁӁҸҮ ҽқґҽҲѹӊӁҸҡҲѹ ѹ ҡӍѦҝҏ ѹҡҔҸӁӁҸҮ ҰҽҝҲѹҲѹӊӁҸҡҲѹ.” (Veǁerka 1963: 204) On the other hand, independent evidence may be available pointing to the existence of a rule that is not derived from direct observation of living usage. Such evidence will be explored in the sections below. 2.1 Smotrickij’s Slavonic grammar Grammatical reflection on GENACCNEG predates the normalization of Great Russian. In the late 16th century, Church Slavonic was promoted in the western Rus’ as a vehicle for religious and intellectual communication. It should represent a fully-fledged alternative to Neo-Latin, which was at the time being imposed by Polish rule to serve these functions, and should therefore be supplied with an elaborate set of teachable grammatical rules (Mešǁerskij 1981: 121) so as to ensure consistent usage, to reduce ambiguity and to produce the overall impression that serious, objective, universal, authoritative and well-considered messages are being transferred. The textbooks that appeared as a part of this Neo Church Slavonic enterprise include Smotrickij’s highly influential Normative Compendium of Slavonic Grammar, which was published in various editions and in various places, including Moscow, between 1619 and 1721. It contains a statement on GENACCNEG.

150

W. ANDRIES VAN HELDEN

“ҐґҽҊҸґӃ ӊѹӁѹҲѦґҧӁҸѸҏ ҡҸҍѹӁҾѦѸӃѦ ҸҲҝѹ҃ҽӁѹҾ ҍҽҡҲѹ҃ҏ ӰӾ ҰҝѹѦѸґӀӂѦ ӊѹӁѹҲѦґҧӁӃҮ ҰҝѦҲӊҸҝҾӀҲ ӊ ҝҸҦѹҲѦґѦӁ.” (Smotrickij (1619)1974: 209) 2 Since the Compendium dealt with a linguistic variety that was not acquired naturally, as a mother tongue, but explicitly, as a dedicated code to be used in specific social and pragmatic functions, it must be treated as a set of instructions for linguistic behaviour rather than an empirical description of linguistic habits. Smotrickij’s GENACCNEG statement must therefore be regarded as an early occurrence of the genitive rule. The origins of the rule remain to be investigated. If Smotrickij did not simply follow an existing tradition, he may have been inspired by living usage in the Slavic vernaculars that he knew, e.g. Polish, in which he often published. More likely, the genitive rule was a generalization based on observed usage in Church Slavonic texts. As is evident from Veǁerka’s observations cited in 1.2, the genitive rule does not provide a satisfactory fit for Old Church Slavonic. But it was good enough as a rule of thumb for East Slavs desiring to emulate Church Slavonic usage in the functions for which the code was intended. As will be shown presently, it is plausible that it was the Neo Church Slavonic genitive rule that was imported into Standard Russian in the 18th century. 2.2 The standard language laboratory The sociolinguistic situation that existed in Russia when Peter the Great laid the foundations for the nation state at the turn of the 18th century is assumed to be familiar (see, e.g., Živov 1996 for a detailed exposition). The reader is also likely to be acquainted with Peter’s linguistic reform. But some relevant developments that occurred after Peter’s death will be recalled here. The script reform had accentuated the existing linguistic diversity in Russia by causing the parallel existence, in a largely illiterate society, of two separate categories of literates. While the civil education system trained a small elite in the new script, the ecclesiastical school system continued to produce literates in the old one. This situation was frozen when the expansion of the civil institutions stagnated after Peter’s death. Some schools closed down (de Madariaga 1979: 370); student enrolment in the Academy of Sciences declined from 114 in 1727 to 19 in 1737 (Marker 1985: 247); the production of books dropped from a yearly average of 22 published titles in the period between 1698 and 1725 to only 13 titles each year between 1726 and 1740 (Sokolov 2002: 171). As a result, the influence of the civil education system was restricted, both geographically and socially. At the same time, the literacy reached in the eccle2

All quotes in this paper are rendered in the modern orthography.

VICISSITUDES OF THE GENITIVE RULE

151

siastical system was so limited that many of its literates could not read texts written in the civil script (Marker 1994: 11-12). 3 The supply of civil content, too, was limited and centralized. As a result of a 1727 ukaz, nearly all printing presses in Russia were suppressed (Marker 1985: 41-42): only the Academy of Sciences could continue to publish in the new script (i.e. provide secular content), besides publishing in Latin and German. In the decades to follow, the Academy press gradually expanded its activities, printing verse and journals as well as scientific books and textbooks for secondary and higher education. 4 Employing almost a hundred persons by the 1760s, it also took orders from government institutions, printing anything from decrees, charters and public addresses down to calendars and stationery (Ostrovitjanov et al. 1958: 297). So for some crucial decades the Academy printers virtually controlled the outward appearance of the texts printed in the new script. Their control included linguistic aspects. In 1733, the Academy adopted a set of orthographic guidelines to be used in its composing shop (tipografija). In 1735, it set up the Rossijskoe sobranie, comprising the philologists who were entrusted with the translation into Russian of scientific and technical literature, to sort out linguistic issues (Ostrovitjanov et al. 1958: 120-124). Compared to the standard languages of Western Europe, Standard Russian was developed in vitro. 2.3 Trediakovskij’s Novyj i kratkij sposob složenija russkix stixov The 1733 orthographic guidelines are lost (Ostrovitjanov et al. 1958: 120) but we know that they included morphological issues: the 1746 dispute between Trediakovskij and Lomonosov over the nominative plural endings of the adjectives was triggered by Trediakovskij’s proposal to modify the relevant guideline (cf. Lomonosov (1746)1952, ann.: 801). There is no direct evidence of the genitive rule also being included in the guidelines but we know that the codifiers of the Academy Press were aware of it. In 1735 Trediakovskij, who was the most active member of the Rossijskoe sobranie, published a treatise on Russian versification that includes a section on poetic license. One of the issues dealt with is GENACCNEG: “ҫҡӁҸӊҽҲѦґҧӁҸѦ ҰҝҽӊѹґҸ, ҔҸҲҸҝҸѦ ҏҲӊѦҝҦѹҲѦґҧӁҸҮ ҝѦҍѹ ҰҝҽӊѹѸӃҮ ҊґҽҊҸґҸѸ ӊѹӁѹҲѦґҧӁӃҮ ҰҽҦѦң ӊҡѦҊҦҽ ҰѦҝѦѸѦӁҾѦҲ ӊ ҝҸҦѹҲѦґҧӁӃҮ ӊ ӍҝҽӉѹҡѦ ҸҲҝѹ҃ҽҲѦґҧӁҸѸ, ѸҸңѦҲ ҦґҾ ҔҝҽҮӁѦҮ ӁҏңҦӃ ӊ ҡҲѹѽѦ ҸӁӃҮ ӊѹӁѹҲѦґҧӁӃҮ ѹ ӁѦ 3

The consequences of this rift were felt as late as 1817, when count Arakǁeev had his regulations for the establishment of military-agricultural colonies printed in the old script so as to reach the broadest possible readership in the countryside (Živov 1996: 492). 4 Textbooks for primary education were provided by the church printers, who continued to use the old script (cf. Marker 1994: 11ff.).

152

W. ANDRIES VAN HELDEN

ҰѦҝѦѸѦӁҾҲҧ. ѢҸ ҡѹҾ ӊҸґҧӁҸҡҲҧ ҸҍѦӁҧ ӊѦґѹҔҽ; ҲҸҊҸ ҝҽҦѹ ҝѦҦҔҸ ѹґѹ, ҔҽҔ ѸҸңӁҸ, ӁѹҔҸҊҦҽ ѦѦ ӁѦ ҏҰҸҲҝѦқґҾҲҧ.” (Trediakovskij (1735)19632: 380) Trediakovskij explicitly refers to the genitive rule as a well-established rule, not as an original observation of actual usage. 2.4 The propagation of the civil standard In the 1750s the standard language broke out of the cubicle. Peter’s nation state programme regained some momentum during the reign of Elizabeth (17411761), when the limited outreach of the Academy of Sciences into Russian society started to be felt as a problem. Count Ivan Šuvalov had the opportunity to carry through a notable programme of education and technology. He cooperated with Lomonosov, who emphasized the importance of the creation of a national elite, arguing that: “[…] ҏ Ӂҽҡ ӁѦҲ ҰҝѹҝҸҦӁӃѽ ҝҸҡҡѹҾӁ Ӂѹ ҽҰҲѦҔҽҝѦҮ, Ҧҽ ѹ ґѦҔҽҝѦҮ ѸҽґҸ, ҲҽҔңѦ ѸѦѽҽӁѹҔҸӊ ѹҡҔҏҡӁӃѽ, ҊҸҝӁӃѽ ґӀҦѦҮ, ҽҦӊҸҔҽҲҸӊ ѹ ҦҝҏҊѹѽ ҏҍѦӁӃѽ ѹ ӁѹңѦ ҡӊҸѹѽ ҰҝҸӍѦҡҡҸҝҸӊ ӊ ҡҽѸҸҮ үҔҽҦѦѸѹѹ ѹ ӊ ҦҝҏҊѹѽ ѸѦҡҲҽѽ.” (Solov’ev (1876)1965: 516) Trediakovskij and Lomonosov were inaugurated as the first Russian members of the Academy in 1745. Šuvalov adopted Lomonosov’s proposal to set up, in 1756, Moscow University, including its preparatory gymnasium. These developments triggered the propagation of the Academy guidelines. Printing shops outside St Petersburg started out in the 1750s and 1760s as branches of the Academy press and were set up by printers and compositors who had been trained in the Academy. The compositor Poljaninov, for example, was sent to Moscow with two apprentices in 1755 to set up the printing shop of Moscow University (Ostrovitjanov et al. 1958: 297). 2.5 Lomonosov’s Rossijskaja grammatika The development of the Russian education system secured the propagation of the Academy guidelines in the long run. In the mid 18th century it was no longer regarded as a virtue that the majority of the (Academy) students could express themselves better in Latin than in their native language (Šmelëva 2003: 116). The need for a state-of-the-art grammar of civil Russian was supplied by Lomonosov’s Rossijskaja grammatika (Lomonosov (1757)1952). Lomonosov’s grammar was completed in 1755 and came out in 1757. Even though the first run comprised only 1,200 copies and its contents were subject

VICISSITUDES OF THE GENITIVE RULE

153

to serious criticism, 5 it must had a considerable impact: it was specifically intended for use in the lower classes of the three gymnasia that were attached to Moscow University, the Army Cadet Corps and the Academy of Sciences (Marker 1994: 15), where adolescents were being prepared for key roles in the civil elite. The Army Cadet Corps, in particular, acted as a forcing house for the first generation of Russian intellectuals (de Madariaga 1979: 370), which was consequently exposed to Lomonosov’s grammar. Catherine II’s access to power in 1762 is likely to have secured further leverage for the Rossijskaja grammatika. Catherine strongly believed in education and started developing and expanding a fully-fledged civil education system (de Madariaga 1979: 373). Schools, including private schools, were subjected to central control, at least as far as contents, methods and teacher training were concerned (de Madariaga 1979: 384). This time, the Russian education system would sustain an uninterrupted (albeit excruciatingly slow) expansion of literacy until the Revolution. At the outset of this process, the Rossijskaja grammatika was the only grammar around. It was reprinted six times before the end of the century. Other grammars were written, notably Kurganov’s 1769 Rossijskaja universal’naja grammatika (cf. Kurganov (17935)1978), the two homonymous anonymous grammars entitled Kratkie pravila (1773 and 1784 respectively: cf. Schütrumpf 1982: i-iv) and the 1802 Academy grammar (cf. Schütrumpf 1984). The Kratkie pravila were published eight times before the end of the century (Marker 1994: 15); Kurganov’s book had eleven editions (Setchkarev 1978: ii). In many respects, these grammars built on Lomonosov’s grammar. Kurganov often literally repeated Lomonosov’s rules and examples. So whatever virtues and flaws Lomonosov’s grammar possessed, civil literates of many generations were exposed to its rules. The expanding school system ensured a multiplier effect: in each subsequent generation, the share of civil literatures increased. According to many writers, including Puškin, Lomonosov’s grammar was a cornerstone of both the Russian language and Russian linguistics. 2.6 Lomonosov and the genitive rule The links between Lomonosov’s and Smotrickij’s grammars are welldocumented. According to Lomonosov’s numerous biographies, the Compen5

Sumarokov disqualified Lomonosov’s linguistic competence, styling him as a dialect speaker (Sumarokov 17872b: 38; cf. 4.2). The Academy philologist Schlözer disqualified Lomonosov’s competence as a linguist, styling him as an undiluted naturalist (Makeeva 1961: 38), probably meaning somebody who collects disparate facts and items without being guided by a unitary philosophical principle.

154

W. ANDRIES VAN HELDEN

dium is one of the two books that he obtained at the age of thirteen and reread time and again during his adolescence in faraway Xolmogory (cf. Van het Reve 1985: 43). In explaining the need for his own grammar, Lomonosov referred to Smotrickij’s. 6 The extant notes on which the Rossijskaja grammatika is based include adaptations of sections from Smotrickij’s work (Lomonosov 1952, ann.: 908). 7 It may therefore be surmised that Smotrickij’s genitive rule was familiar to Lomonosov. Similarly, we may assume that Lomonosov was aware of Trediakovskij’s genitive rule. He was not involved in the Academy’s codification activities in the 1730s but took part in the discussion that ensued Trediakovskij’s rules on versification (cf. Silbajoris 1968), which included the latter’s statement on the genitive rule and poetic license. So it comes to no surprise that Lomonosov’s grammar contains the genitive rule: “§ 501: ғҡҾҔѹҮ ҊґҽҊҸґ ҦѦҮҡҲӊѹҲѦґҧӁӃҮ ҲҝѦқҏѦҲ ӊѹӁѹҲѦґҧӁҸҊҸ ҰҽҦѦңҽ ҡӊѦҝѽ ѹѸѦӁѹҲѦґҧӁҸҊҸ: (1)

KҲҸ ѽҝҽӁѹҲ ԒԂӪӹӰՂ, ӉҽҔҸӁӃ ҡҸѽҝҽӁҾҲ ѦҊҸ ӊӉҽѹѸӁҸ.

§ 502: ҘҸҊҦҽ ң ҰҝѦҦ ӁѹѸ ҰҝѹҡҸӊҸҔҏҰґҾѦҲҡҾ ҸҲҝѹ҃ҽҲѦґҧӁҽҾ ҍҽҡҲѹ҃ҽ ӰӾ, ӊѹӁѹҲѦґҧӁӃҮ ҰҽҦѦң ҸқҝҽӂҽѦҲҡҾ ӊ ҝҸҦѹҲѦґҧӁӃҮ: (2)

ѢѦ ҦҽӊҽҮ ӴӹԊӠ ҾӉӃҔҏ ӊ ҰѹҝҸӊҽӁѹѹ.” (Lomonosov (1757)1952: 560-561)

The rule is also included in the textbooks of Russian grammar that followed Lomonosov’s (cf. Kurganov 17935(1978): 63; Kratkie pravila (1773)1982: 79; A.A. Barsov (1788)1980: 304; Schütrumpf 1984: 170). So, while the Academy composing shop may have been the channel that facilitated the genitive rule’s migration from Neo Church Slavonic into 18thcentury civil printing, it is likely that Lomonosov’s grammar paved its way into the writing habits of expanding generations of secular literates in Russian. 3. The genitive rule and 18th-century usage The scenario provided above is independent of the question whether the genitive rule was applied in the 18th-century spoken vernacular. It is generally taken

6 Lomonosov 1952: 690: "ҋѹӀ ҊҝҽѸѸҽҲѹҔҏ ӁѦ ӊӃҦҽӀ Ҿ ӉҽҰҸґӁҏӀ, ӁҸ ҲҸґҧҔҸ ҸҰӃҲ, ѹқҸ ѦӂѦ ӁѹҔҽҔҸҮ ӁѦҲ, ҔҝҸѸѦ ҡґҽӊѦӁҡҔҸҮ [viz. Smotrickij’s Compendium] ѹ ѸҽґѦӁҧҔҸҮ ӊ ґѦҔҡѹҔҸӁѦ, ӊѦҡҧѸҽ ӁѦҡҸӊѦҝҗѦӁӁҸҮ ѹ ӊҸ ѸӁҸҊѹѽ ѸѦҡҲҽѽ ӁѦѹҡҰҝҽӊӁҸҮ." 7 The notes were written by an unidentified philologist who assisted Lomonosov in the preparation of the grammar.

VICISSITUDES OF THE GENITIVE RULE

155

for granted that Lomonosov’s genitive rule reflected “the language” at large. 8 As we shall see below, this is not necessarily justified. 3.1 Compliance in the 18th century According to Safarewiczowa’s (1959) counts conducted on various 18th-century texts, the genitive is used virtually without exception at the outset of the century, in the writings of Peter the Great (1672-1725), as well as at the end, in Karamzin’s works (1766-1826), while only sporadic accusatives are found in the prose of other 18th-century writers, including Dolgorukova (1714-1771), Fonvizin (1745-1792) and Krylov (1769-1844). So recorded 18th-century usage bears out the genitive rule. In view of the history of the development of the standard language, however, this evidence must be put into perspective. We know from the dispute between Trediakovskij and Lomonosov over the plural endings of the adjectives that the Academy composing shop applied its orthographic conventions to the manuscripts that were handed in (Lomonosov (1746)1952: 802). There is no reason a priori why compositors and correctors would leave case endings alone: it requires some sophistication to distinguish between correcting the spelling of an ending and correcting the ending itself; for some nouns, the difference between the genitive and the accusative was not even audible. Sumarokov in fact included the genitive rule in his treatise on orthography (Sumarokov 17872a: 21-22). Any principles guiding the composition of the endings, however, should be simple: compositors were not expected to study the contents of manuscripts in order to be able to compose them. The genitive rule fulfils this requirement. It cannot be excluded, therefore, that any accusatives in the texts that were used for frequency counts in the 20th century had been weeded out in the 18th century before the texts went to print for the first time. A quick scan of Safarewiczowa’s material suggests that much of it may originally have been printed at the Academy press (e.g. Lomonosov, Trediakovskij) or the Moscow university press (e.g. Fonvizin), while late 18th-century writers (Karamzin, Krylov) are likely to have learned the genitive rule at the gymnasium. The letters of Peter the Great, which have zero accusatives, may have been edited as well. This would leave Dolgorukova’s memoirs, for which Safarewiczowa used a 1913 scientific edition, to be accounted for. Further research is needed here but any evidence supporting the overall use of the genitive must be balanced against various types of counter-evidence that are discussed below. 8

Note that usage alone did not warrant the inclusion of a rule in Lomonosov’s grammar. His syntax does not, for example, include the rule accounting for his use of the genitive nikakoj with net in the quote given in note 6.

156

W. ANDRIES VAN HELDEN

3.2 The transformationalist phrasing of the genitive rule In 17th- and 18th-century grammatical discourse, the genitive rule is phrased as if it were a transformation. The terms pretvorjaet (Smotrickij), peremenjaet (Trediakovskij), obrašǁaetsja (Lomonosov, Kurganov), peremenjaetsja (A.A. Barsov (1788)1980: 304) evoke a process in which one case is converted into another. Since such a process cannot be observed, it must be assumed that grammarians postulated it as a mental operation and on the basis of the theoretical assumption that affirmative sentences are in some sense more basic, as in late 20thcentury syntax. Alternatively, the transformational phrasing can be interpreted as an instruction of the grammarian to a writer or speaker to change what he intended to write or say. In that case it must be assumed that grammarians presupposed that the use of the genitive did not always come naturally. 3.3 The transformational capacities of non-object accusatives In fact the expansion of the genitive at the cost of the accusative in certain types of negative sentences in written sources can be interpreted as an indication of the advance of the accusative. According to Krys’ko (20062: 70), the accusative had always been used in negative sentences to mark adjuncts of duration and distance, i.e. for non-object functions. From the 16th century onward, however, genitive forms gained a foothold in this function. Cf. Krys’ko’s example from Peter the Great’s correspondence: (3)

ү ԌӴԂԌՁԂԽӠ ӴӾӢԙԽ ӁѦ Ѧѽҽґ ҰҽҝҏҡҸѸ … ӊҡѦ җґѹ Ӂҽ ҊҝѦқґѹ. (Krys’ko 20062: 70)

So accusatives of duration and distance acquired, in Krys’ko’s terminology, transformational capacities (transformacionnye sposobnosti) in a period when the first direct object accusatives slipped into negative sentences. This suggests that, at that point, writers started overcomplying: they no longer spontaneously used the genitive in all instances of negation but acquired the habit of replacing accusatives by genitives. 9 3.4 The genitive rule and poetic licence The way GENACCNEG is dealt with in the discussion on versification suggests that the use of the genitive was not under all circumstances natural. Tre9

Vaillant’s (1977: 77) observation that the genitive indicates duration in negated sentences in older manuscripts detracts from the value of this suggestion. More research is needed anyway, as two kinds of duration can occur in negated sentences and may engender different case use: the duration of the negated event and the duration of the negation of the event.

VICISSITUDES OF THE GENITIVE RULE

157

diakovskij’s tolerance of the accusative in verse implies that its use was conceivable. Poetic licence did not imply that anything goes, at least in Trediakovskij’s days. While the rules for civil prose were intended to reduce dysfunctional variety in current usage, poetic licence was not supposed to expand it. In Trediakovskij’s conception, poetic licence should not lead to admitting forms that lack any basis whatsoever in existing practice: “ғҸґҧӁҸҡҲѹ ӊҸҸқӂѦ ҲҽҔҸӊҸҮ ӁҽҦґѦңѹҲ қӃҲҧ, ҍҲҸқ ҝѦҍѦӁѹѦ, ҰҸ ӊҸґҧӁҸҡҲѹ ҰҸґҸңѦӁӁҸѦ, ӊѦҡҧѸҽ ҝҽҡҰҸӉӁҽҲҧ қӃґҸ ѸҸңӁҸ, ҍҲҸ ҸӁҸ ҰҝҾѸҸѦ ҝҸҡҡѹҮҡҔҸѦ, ѹ ѦӂѦ ҲҽҔ, ҍҲҸқ ҸӁҸ ӁѦҡҔҸґҧҔҸ ѹ ҏҰҸҲҝѦқѹҲѦґҧӁҸѦ қӃґҸ.” (Trediakovskij (1735)19632: 380) Lomonosov himself saw no bones in dropping an occasional accusative into his verse. Here are a few examples from his Pis’mo o pol’ze stekla: (4)

Ѩ ҔҽҔ Ҿ ҸӁҸѦ ѽӊҽґҾ ӊҸҡҰҸѸѹӁҽӀ ѢѦ ԊӹԐӪӹԙԽӸ ґңѹӊҸҊҸ Ҿ ҡҍҽҡҲҧҾ ҰҝѦҦҡҲҽӊґҾӀ.

(5)

ѢѦ ԌԂӢ ґѹ ѸӃ ӊ ҋҲѦҔґѦ ԼӹԞӾԙԽӴӾӰӰՂӡ ѹѸѦѦѸ ѯҲҸ ҍѦҡҲҧ ҦҸҡҲҸҮӁҏӀ ӊҸӉҦҽҲҧ ѦѸҏ ҔҸҡӁѦѦѸ?

(6)

ѢѦ ӹԄӰӸ ґѹ ҸӁ ҋҲѦҔґҸѸ ҏѸѦґ ҡӊҸҦѹҲҧ ҡ ӁѦқѦҡ Ѩ ҰҽҊҏқҏ ҡѦқѦ ҸҲ ӊҽҝӊҽҝҸӊ ӁҽӁѦҡ …

(7)

ҫ ҔҸґҧ ӊѦґѹҔѹѸ ҸӁ ӊҸҡҲҸҝҊҸѸ қӃ ҰґѦӁѹґҡҾ ҘҸҊҦҽ қ ӢԂԒӫԐӰӫ ԽӴԂӢӸ ҲҸґҧ ҲѦҡӁҸ ӁѦ ӊҔґӀҍҽґ … (Lomonosov (1752)1959)

All this suggests that the codifiers rejected the use of the accusative in the standard language that they were developing, not that it was alien to them. 3.5 Sumarokov’s genitive rule apologetics The vehemence of the discourse in which non-compliance with the genitive rule is dissuaded or even condemned may represent relevant evidence disproving the overall use of the genitive. Trediakovskij’s insistence on extreme restraint in using the accusative in verse, as quoted in 2.3, may well reflect the force of the temptation to do so. In this respect, Sumarokov’s position is most illuminating. Although disagreeing with many of Lomonosov’s choices, he fully endorses the genitive rule: “ԾӾ: ӊ ҦҝҏҊѹѽ ҾӉӃҔҽѽ Ұҝѹ ҊґҽҊҸґҽѽ қѦҝѦҲ ҔҸ ҡґѦҦҏӀӂѦѸҏ ѹѸѦӁѹ ҡҏӂѦҡҲӊѹҲѦґҧӁҸѸҏ, ҰҽҦѦң ӊѹӁѹҲѦґҧӁӃҮ, ҽ ӊ ӁҽҗѦѸ ҾӉӃҔѦ қѦҝѦҲ ҸӁҸ ҰҽҦѦң ҝҸҦѹҲѦґҧӁӃҮ: ӊ ҍѦѸ Ӂҽҗѹ ѸӁҸҊѹѦ Ұѹҡ҃Ӄ ҰҸҍҲѹ ӊҡѦҊҦҽ ҊҝѦҗҽҲ; ѹқҸ ҊҝҽѸѸҽҲѹҔѹ ӁѦ ӉӁҽӀҲ. ѢҽҰҝ. ӁѦґҧӉҾ ҡҔҽӉҽҲҧ: (8)

Ѥ ӁѦ ґӀқґӀ ѸҸҮ ԗӹӪӹӡ.

158 (9)

W. ANDRIES VAN HELDEN

ѢѦ ӁҽҦeӊҽӀ ԶӫԼӫ.

(10) ѢѦ ӊѹңҏ ԊӾԙ. ӁҸ ԗӹӪӹԋ, ԶӫԼՂ, ԊӾԙԂ ѹ ҲҽҔ ҦҽґѦѦ.” (Sumarokov 17872a: 21-22) By condemning the “many writers” who “nearly always” violate the rule, Sumarokov acknowledges that the use of the accusative is widespread. His observation is hardly compatible with Safarewiczowa’s evidence cited in 3.1 and feeds the suspicion that the accusatives had been weeded out in the texts she examined. Note, furthermore, that Sumarokov does not fault the writers because they fail to conform to existing usage but because they do not know grammar. 3.6 Kantemir’s non-compliance Antiox Kantemir’s (1708-1744) well-documented non-compliance with the genitive rule is another piece of evidence supporting the assumption that the use of the accusative was widespread in the 18th century. Fleckenstein (1961: 215) even finds 58% accusatives (sample size n = 48) in Kantemir’s satires. Safarewiczowa (1959: 99) believes that this observation should not shatter the general picture. She points to Kantemir’s origins: he was born in Constantinople from a Greek mother and a Moldavian father and spent his early youth in Moldavia. Safarewiczowa surmises that Russian was not Kantemir’s native language and concludes that his accusatives do not represent authentic 18th-century Russian usage. She attributes Kantemir’s frequent use of the accusative to the exclusive use of this case in negative sentences in his native language (whatever was his native language; Safarewiczowa 1959: 77). Safarewiczowa is not the only one to call Kantemir’s linguistic credentials into question. Fleckenstein (1961: 215) mentions that his accusative use has been ascribed to French influence. Karskij accuses Kantemir of vulgarisms and polonisms: “ѢҸ ҸӁ, ҔҽҔ ѹӁҸҡҲҝҽӁѦ҃ ҰҸ ҰҝҸѹҡѽҸңҦѦӁѹӀ, ӁѦ ѸҸҊ ӁҽҡҲҸҾӂѹѸ ҸқҝҽӉҸѸ ҰҸҡҲѹҊӁҏҲҧ Ҧҏѽҽ ҝҏҡҡҔҸҊҸ ҾӉӃҔҽ: ҝҾҦҸѸ ҡ ҸҍѦӁҧ ѸѦҲҔѹѸѹ ѹ ҡѹґҧӁӃѸѹ ӊӃҝҽңѦӁѹҾѸѹ ҏ ӁѦҊҸ ӊҡҲҝѦҍҽӀҲҡҾ ҡґѹҗҔҸѸ ҊҝҏқӃѦ ѹ ӊҏґҧҊҽҝӁӃѦ ҡґҸӊҽ, ӁҽҰҝ. ӊ I-ҸҮ ҡҽҲѹҝѦ: “ԪӫԐԋӰՂӡ, ԽӢӹԞԌՂ ӢՂԄӰӫӴ, ԘӫӪԂ ԗӹԌԗӾӴԂӾԽ”. ѢѦѸҽґҸ ҏ ӁѦҊҸ ѹ ҰҸґҸӁѹӉѸҸӊ.” (Karskij 1912: 70) As far as GENACCNEG is concerned, Karskij’s disqualifications of Kantemir’s command of Russian seem to cancel out Safarewiczowa’s. His example suggests that he accuses Kantemir of providing too realistic a picture of the spoken language rather than of incorrect observations. Also, since Polish is the one Slavic standard language that most consistently uses the genitive for the direct object of negated verbs, Kantemir’s accusatives cannot be polonisms.

VICISSITUDES OF THE GENITIVE RULE

159

In fact there are few reasons to doubt Kantemir’s reliability as a source of information on contemporary GENACCNEG usage. He had arrived in St Petersburg as early as 1711, at the age of three, and had received his complete education in the immediate environment of the court. His home tutor was the poet and Academy translator Ivan Il’inskij (Ostrovitjanov et al. 1958: 119). Kantemir was an erudite, wrote on Slavonic and in Russian, and left a manuscript of a Russian-French dictionary (Uspenskij (1992)1997: 529). It is, therefore, unlikely that Kantemir was less proficient in civil Russian than the majority of literate persons of his generation and social standing. But even if he had failed to master Russian to the extent of writing stylistically flawless works of art in it, it is implausible that he would have failed to notice, and adopt in writing, the simple mechanical rule stipulating that a negated verb always takes the genitive. Kantemir must, moreover, have been aware of the existence of the genitive rule: he was the third participant in the aforementioned discussion on Russian versification started by Trediakovskij (cf. Silbajoris 1968) and must have read the latter’s opinion. Yet his abundant accusative use cannot be reduced to excessive use of poetic licence: as shown by Meintema (1988: 399ff.), Kantemir’s high accusative frequency is not only found in his satires but also in the dispatches that he wrote as a diplomat in Paris and London, which may be expected to be free of mannerisms. This may explain why Kantemir’s accusatives were never weeded out: his dispatches had not been published in the 18th century. His satires, too, circulated only in manuscript, at least during his lifetime. The history of the posthumous publication of the satires is complex; 10 but it is likely that the texts that Safarewiczowa, Fleckenstein and Meintema analysed had not passed through the hands of regular 18th-century Academy correctors and typographers. More important, Meintema’s (1988) detailed analysis of Kantemir’s usage shows that his GENACCNEG choices are not arbitrary: they follow a semantic pattern, which is something a non-native speaker is not likely to produce unless he could import a pattern from his native language. Meintema demonstrates, however, that Kantemir’s usage can largely be accounted for by the same set of principles that she established for 20th-century Russian (Meintema 1986; see 8.10 for some exceptions). So it is not unreasonable to assume that a variety of spoken Russian existed in Lomonosov’s days in which the accusative was frequently used to mark the 10 The editor of the 19th-century edition (1867) of Kantemir’s works P.A. Efimov accused the editor of the posthumous 1762 edition I.S. Barkov of distorting Kantemir’s text and comment (Kantemir 1867: v). In a comment accompanying the 20th-century edition of Kantemir’s verse, Gerškoviǁ (1956) points out that Efimov is not to be trusted either.

160

W. ANDRIES VAN HELDEN

direct object of a negated verb, that variety being sufficiently widespread and neutral to be tolerable among the civil elite and even in official correspondence. 4. Why was the genitive rule adopted? If it can be assumed that, in Lomonosov’s days, at least some prominent Russians could be heard using the accusative to mark the direct object of negated verbs, the question arises why this was not taken into account in the Rossijskaja grammatika. If confronted with a discrepancy between the Church Slavonic genitive rule and existing usage, why did Lomonosov, like Trediakovskij and Sumarokov, decide to maintain the genitive rule? Several plausible reasons can be submitted. 4.1 Lomonosov as a reasonable codifier Lomonosov’s attitude toward grammatical tradition is often characterized as realistic. The norms to be applied should be reasonable (rassuditel’nyj: cf. Martel 1933: 38). Archaisms are artificial: rules must be derived from actual usage. But how does Lomonosov define usage? His approach transpires from his position in the dispute with Trediakovskij over the nominative plural endings of the adjectives. According to Trediakovskij (cf. Lomonosov (1746)1952, ann.: 801), each gender was entitled to its own ending: -i should be used with masculine nouns (istinnyi), -e with feminine nouns (istinnye) and -ja with neuter nouns (istinnyja). Lomonosov observed that there is no gender distinction at all in the plural in spoken Russian: -i was heard nowhere, at least in Great Russian, while -e and -ja were both heard but did not differentiate gender: “[…] ӊҡҾҔ ҏҡґӃҗѹҲ ҏ ӊҡѦѽ ӊѦґѹҔҸҝҸҡҡѹҾӁ, ҔҸҊҦҽ ҸӁѹ ҊҝҸѸҔҸ ҊҸӊҸҝҾҲ, ѹ ҏ ӊҡѦѽ ҏӊѹҦѹҲ, ҔҸҲҸҝӃѦ, ӁѦ ҦҏѸҽҾ Ҹ Ұҝҽӊѹґҽѽ, ҰѹҗҏҲ, ҍҲҸ ҸӁӃѦ ҔҸӁҍҽҲҡҾ Ӂҽ Ӿ ѹґѹ Ӂҽ ԋ ӊҸ ӊҡѦѽ ҝҸҦҽѽ қѦӉ ҝҽӉқҸҝҏ, ӁҸ қҸґҧҗѦ Ӂҽ Ӿ, ӁѦңѦґѹ Ӂҽ ԋ, ѹ Ӂҽ Ӡ ӁѹҔҸҊҦҽ.” (Lomonosov (1746)1952: 86) A non-existent distinction, Lomonosov points out, should not be reinvented: “ҫҦӁҽҔҸ Ҿ ҝҽҡҡҏңҦҽӀ, ҍҲҸ ҲҽҔҸҊҸ ҝҽӉґѹҍѹҾ ҝҸҦҸӊ, ҔҸҲҸҝҸҊҸ ӊ ҝҸҡҡѹҮҡҔҸѸ ҾӉӃҔѦ ӁѦҲ, ӊӁҸӊҧ ӉҽѸӃҗґҾҲҧ ӁѦ ӁҽҦґѦңѹҲ.” (Lomonosov (1746)1952: 86) By this token, the 1933 Academy guidelines, distinguishing between masculine nouns on the one hand (istinnye) and feminine and neuter nouns on the other (istinnyja), lack a sufficient basis: “[…] ӊӊѦҦѦӁӁҸѦ Ӊҽ 10 ѹ қҸґҧҗѦ ґѦҲ ӊ ҽҔҽҦѦѸѹҍѦҡҔҸҮ ҲѹҰҸҊҝҽӍѹѹ ҏҰҸҲҝѦқґѦӁѹѦ […] ҦҸӊҸґҧӁҸҊҸ ҸҡӁҸӊҽӁѹҾ ӁѦ ѹѸѦѦҲ […]” (Lomonosov (1746)1952: 87)

VICISSITUDES OF THE GENITIVE RULE

161

Yet, Lomonosov proposed to leave the guidelines as they are. For one thing, “[…] ҔҽҔ ѹ ӊҸ ӊҡѦҮ ҊҝҽѸѸҽҲѹҔѦ, ҲҽҔ ѹ ӊ ҡѦѸ ҡґҏҍҽѦ ҸҦӁҸѸҏ ҏҰҸҲҝѦқґѦӁѹӀ ҰҸӊѹӁҸӊҽҲҧҡҾ ҦҸґңӁҸ.” (Lomonosov (1746)1952: 84) Besides, the current practice of the composing shop was not “protivno” to Great Russian (Lomonosov (1746)1952: 87). Lomonosov often uses this term, apparently to underpin judgements that cannot be buttressed by reasoning. These arguments made Lomonosov stick to the existing plural rule (which was only to be discarded in the 20th century). To him, the type of existing usage to be respected was current Academy practice, even if it was unfounded and had been in existence for only a decade, rather than the spoken vernacular. The genitive rule was current practice in the Academy: this alone may have made Lomonosov decide to maintain it. He may never have given it a second thought. 4.2 Lomonosov as a scientific syntactician As an additional consideration, the state of the art in 18th-century reflection on language may have kept Lomonosov from disposing of the genitive rule. The genitive rule presupposes a connection between negation and case choice, i.e. concepts that are usually allotted to different departments of linguistic analysis. Negation had been dealt with in semantic terms since antiquity. Lomonosov was aware of that. His work on rhetoric discusses Artistotle’s distinction between generic and specific or partial negation (Lomonosov (1748)1952: 118). This distinction might to a certain extent have served to account for the use of the accusative and genitive for the direct object of negated verbs (cf. 6.4). In Lomonosov’s days, however, the core business of grammar schools consisted in teaching applied Latin. Grammars of national languages were modelled on the format of Latin grammars. In production grammars of Latin, case was dealt with in syntactic terms. The dominant paradigm was government: cases are basically assigned on the basis of syntactic relationships, especially those between nouns and verbs, which are the essential building bricks of human communication, representing the two components of the visible world that can be communicated about: things (vešǁi) and acts (dejanija) (Lomonosov (1757)1952: 405-406). Verbs are said to require (trebovat’) cases. Transitive verbs usually require the accusative besides the nominative, except for a few that require a different case or a preposition-case combination (Lomonosov (1757)1952: 561). Instances of case selection that do not fit this scheme tend to be ignored. But Lomonosov allows for two such “anomalies”.

162

W. ANDRIES VAN HELDEN

The first instance is what Peškovskij calls dvojakoe upravlenie: the same role may be filled by more than one case, as in: (11)

ѩҽҮ ӊҸҦҏ.

(12) ѩҽҮ ӊҸҦӃ. (Lomonosov (1757)1952: 561) So as to cover this instance, Lomonosov refines the syntactic concept of government. Instead of just positing a discrete government relationship between verbs and nouns, he postulates a Newtonian (i.e. electric or magnetic) force, 11 which verbs are supposed to exert on nouns. The effect of this governing force may vary, depending on its strength in specific conditions, resulting in different case usage: “ҭҝѹӁѹѸҽӀҲ ӁѦҝѦҦҔҸ ҦѦҮҡҲӊѹҲѦґҧӁӃѦ ҊґҽҊҸґӃ ѹ ҝҸҦѹҲѦґҧӁӃҮ ҰҽҦѦң, ҔҸҊҦҽ ѹѽ ҡѹґҽ ӁѦ ҔҸ ӊҡѦҮ ӊѦӂѹ, ӁҸ Ҕ ҍҽҡҲѹ […] ҰҝҸҡҲѹҝҽѦҲҡҾ.” (Lomonosov (1757)1952: 561) This approach would have offered an explanation for GENACCNEG as an instance of dvojakoe upravlenie: if the genitive is used whenever a verb fails to govern its object completely, as in (12), it may be expected to be used a fortiori if the verb fails to affect the object at all, as in the case of negation. Unexpectedly, Lomonosov does not invoke the Newtonian force image to account for the genitive rule as an instance of zero governing force: this option is only found in the 19th-century treatments on GENACCNEG (see 7.2). Lomonosov adopts a different strategy. Voice presents another anomaly in the government approach. Identical relationships between things and acts may be represented by either active or passive verbs, which have different case assignment. So as to account for case choice in passive sentences, Lomonosov ((1757)1952: 562-563) invokes transformations. The idea of transformation was familiar to natural scientists and grammarians (cf. 3.2 and Klenin 1988), and ultimately goes back to antiquity: alteration (alloiôsis) is on Aristotle’s list of forms of movement. Lomonosov and his Marburg tutor Christian Wolff are said to have innovated Aristotle’s transformation concept: instead of regarding it as a special

11 The term sila was already used in grammars (including Smotrickij’s) but its Newtonian qualities may be due to Lomonosov. Being a natural scientist, he is likely to have imported concepts from physics and chemistry into his grammatical framework. Another possible example is his use of the concept of indivisible part (nerazdelimaja ǁast’) in the phonetic part of his grammar (Lomonosov (1757)1952: 398), which might be regarded as an early attempt to get to grips with the intuitive notion of phoneme as the atom of language: atoms, defined as hypothetical bodies, so small as to be incapable of further division, were a known concept in physics.

VICISSITUDES OF THE GENITIVE RULE

163

type of change, qualitatively distinct from mechanical movement, they treated it as a complex of mechanical movements (cf. Uëmov 2003: 127). 12 The phrasing of Lomonosov’s version of the genitive rule quoted in 2.6 shows that it is this conception of transformation as a complex mechanical movement that Lomonosov resorts to when accounting for GENACCNEG. No attempt is made to reconcile the syntax of government with the semantics of negation: in presenting the rule, he does not refer to the semantics of negation at all, only to the negative particle. The addition 13 of the negative particle to the verb converts the accusative into the genitive. One movement triggers another. Does this mechanistic approach offer the possibility to distinguish between situations in which ne triggers a case transformation on the one hand and situations in which it fails to do so on the other, as would be required to account for Kantemir’s usage? When encountering kindred problems as he tried to provide rules for the passive transformation by means of adding the particle -sja, Lomonosov simply gives up: his bottom line is that it is always safer (bezopasnee) to employ a passive participle (Lomonosov (1757)1952: 563). Similarly, he may have considered it to be impossible, given his linguistic framework, to formulate criteria for the use of two cases for the direct object of a negated verb. It must be realized that the Rossijskaja grammatika was intended as a textbook. In the ideology of Enlightenment, knowledge was acquired by observation and transferred by demonstration. If two sentences referred to different states of affairs in observable reality, they had different meanings; if two sentences containing the same lexical material referred to the same state of affairs, i.e. were simultaneously true or false, their difference was syntactic, as in the case of the passive transformation. Similarly, an affirmative and its corresponding negative sentence intrinsically referred to the same state of affairs (except of course that the existence of the state of affairs was denied in the negative sentence). So it made sense to account for deviating case usage in the negative sentence in terms of a transformation rule. This approach could not take Lomonosov very far in dealing with GENACCNEG in Kantemir’s usage. Here direct object genitives and accusatives can replace each other in most negative sentences without affecting the state of affairs referred to. The states of affairs that are referred to by genitive negative sentences and accusative negative sentences are usually identical.

12

It is tempting to speculate that this innovation may have underlain both Lomonosov’s conviction that mass is preserved in chemical transformations (the Lomonosov-Lavoisier law) and his apparent conviction that meaning is preserved in syntactic transformations. 13 Lomonosov uses the verb prisovokupljat’sja. He employs the same term to refer to addition in physical processes.

164

W. ANDRIES VAN HELDEN

In the Age of Enlightenment, it is appropriate to discard what cannot be shown to exist. Now if there is neither a way to explain why the genitive and accusative should both be available for marking the direct object of negated verbs, nor a way to exploit them for making a useful distinction, the coexistence of the two cases in GENACCNEG can only be a source of confusion, an instance of dysfunctional variety. A grammar that does not offer a choice is more efficient. Given the goals of the Rossijskaja grammatika, discarding the distinction was by far the most elegant solution. 14 4.3 Lomonosov as a dialect speaker It has been suggested that Lomonosov’s reluctance to discard the genitive rule was inspired by his own usage. Although Lomonosov’s native village Xolmogory was not in the middle of nowhere at the time (before the foundation of St Petersburg the Dvina delta was a major channel for imports by sea from Western Europe), Lomonosov was sometimes faulted for being a dialect speaker and introducing provincialisms: “ҐҝҽѸѸҽҲѹҔҽ Ҋ. һҸѸҸӁҸҡҸӊҽ ӁѹҔҽҔѹѸ ҏҍѦӁӃѸ ҡҸқҝҽӁѹѦѸ ӁѦ ҏҲӊѦҝңҦѦӁҽ, ѹ ҰҸ ҰҝѹҍѹӁѦ ҍҲҸ ҸӁ ѸҸҡҔҸӊҡҔҸѦ ӁҽҝѦҍѹѦ ӊ ҔҸґѸҸҊҸҝҡҔҸѦ ҰҝѦӊҝҽҲѹґ, ӊҸҗґҸ ӊ ӁѦѦ ѸӁҸңѦҡҲӊҸ ҰҸҝҍѹ ҾӉӃҔҽ.” (Sumarokov 17872b: 38) As historical information on GENACCNEG usage in dialects is scarce, 15 it is hard to verify whether Lomonosov’s native dialect inspired his GENACCNEG choice. But Grandilevskij 1907 includes an account of the major differences between Xolmogory and Standard Russian case usage that may provide a clue. If Xolmogory had the genitive rule, Grandilevskij would not explicitly tell us so, since the rule applied in Standard Russian as well. But he reports that the genitive is used more often to mark direct objects in Xolmogory than in Standard Russian (Grandilevskij 1907: 69). Assuming that the use of the genitive had not expanded since Lomonosov’s days, chances are that all negated verbs had 14

It may be argued that Lomonosov could also have opted for generalizing the accusative rather than the genitive. But this would not have provided a better fit: Kantemir used both cases. Moreover, as the genitive rule existed already, adoption of the accusative risked being stigmatized as a barbarism: elsewhere, Lomonosov complains about a construction (the use of ǁrez to mark the agent in a passive sentence) having been imposed on Russian from French (Lomonosov (1757)1952: 562). 15 Safarewiczowa’s analysis of Durnovo and Ušakov’s 1910 collection of folk texts of unspecified origin produces an average 25% accusative score (Safarewiczowa 1960: 106; n = 57), which is within the range found for modern written Russian. Zelenin’s 1914 collection of texts from the Perm’ region, however, contains an exceptional 71% accusative frequency (Safarewiczowa 1960: 104; n = 163). Peter the Great had populated Perm’ from various places in central Russia in the early 18th century (cf. Zelenin 1913: 520-522).

VICISSITUDES OF THE GENITIVE RULE

165

always had genitive direct objects in Xolmogory. One of Grandilevskij’s examples hints at this situation. (13) ѣҦӊҽ ңѦ ҸӁ ԌӴӫԖ ԌӫԶ ӁѦ ӉҽҊҏқѹґ. (Grandilevskij 1907: 69) Here dvux duš fits the genitive rule in its strictest interpretation. The standard language has given up applying the rule to sentences containing edva ne or ǁut’ ne, which are nowadays regarded as being functionally affirmative and always have an accusative object (Mustajoki and Heino 1991: 24). As Grandilevskij adduces the example to illustrate the difference between Xolmogory and the standard language, this is likely to have been the case in the early 20th century as well. In fact, as will be shown in 6.3, Vostokov exempted confirmative sentences containing ne from the genitive rule in the first half of the 19th century. Example (13) prompts the suspicion that Lomonosov may have felt quite comfortable with Smotrickij’s genitive rule: the rule generated sentences that were familiar in his own dialect. Lomonosov’s attitude towards his dialect was positive and respectful: he considered it to be closest to “old Slavonic” (Makeeva 1961: 27-28). The genitive rule may have confirmed this conviction. His dialect may also have nurtured the aesthetic considerations he also allowed to tilt the scale in his grammatical decisions: certainly the genitive rule prevented generating sentences that sounded protivno to his Xolmogory ears. 16 This does not imply that Xolmogory “had” the genitive rule. For one thing, the border between affirmative and negative sentences does not demarcate the use of the direct object genitive in Xolmogory. As Grandilevskij points out, the genitive is used more often in affirmative sentences as well, replacing either the object accusative or the subject nominative of the standard language. Cf. his examples: (14) ѢҽҗѦґ қҸҝҸӊѹҔҽ. (15) Ѱ ӁѦҊҸ ҦѦӁѦҊ ѦҡҲҧ ѹ ӊҡҾҔѹѽ ӁҽҝҾҦҸӊ. (Grandilevskij 1907: 69)

16

Note the difference between Lomonosov’s and Kantemir’s linguistic biographies. Kantemir arrived in St Petersburg in 1711 at the age of three, only eight years after its foundation. The population increased from virtually zero in 1703 to 70,000 in 1725 (ƀernyšev (1913)1970: 339) and to 150,000 in 1750 as a result of immigration from all over Russia, from and into all social classes, each immigrant bringing along his speech habits. While dialects of capitals often carry a certain prestige and tend to linguistically absorb newcomers from the provinces, St Petersburg lacked a native population base with a vernacular that could serve as a target for linguistic assimilation: as ƀernyšev ((1913)1970: 340) put it, “[…] ӊ ҾӉӃҔѦ ҭѦҲѦҝқҏҝҊҽ ӁѦҲ ҃ѦґҧӁҸҡҲѹ ѹ ҸҰҝѦҦѦґѦӁӁҸҡҲѹ.” Kantemir was one of the first children to grow up in this linguistic environment.

166

W. ANDRIES VAN HELDEN

So it cannot be excluded that the use of the genitive for direct objects of negated verbs in Xolmogory must be accounted for by a broader principle. This would be in line with Veǁerka’s statements on the use of the genitive in West and East Slavic as cited in Section 2. 5. The genitive rule and 19th-century usage The evidence that was presented so far challenges the assumption that the genitive rule reflected 18th-century usage. Under that assumption, the accusative progressively ousted the genitive in the course of the ensuing centuries, the written standard following spoken usage at a distance. The gradual retreat of the genitive is in fact said to be observable in written usage (cf. Uglitsky 1956: 377) as well as in 19th-century grammars (cf. Timberlake’s statement quoted in 1.5). If it is assumed, however, that the accusative was there all the time, alternative explanations must be given for the 19th-century data and statements that apparently underlie the observations. Such explanations will be explored below. 5.1 Puškin’s poetic and narrative licence A.S. Puškin was possibly the first user of the standard language to challenge the genitive rule, or at least its scope. In the journal Atenej, the critic M.A. Dmitriev had faulted him for ignoring it in stanza 33 of Chapter 4 of Evgenij Onegin: (16) ԟӴԂ ӴӾӪԂ ҡҡҸҝѹҲҧ ӁѦ ѽҸҍҏ. According to Dmitriev, the use of the accusative dva prompts a reading in which dva veka is interpreted as an accusative of duration instead of an object: this could have been avoided if Puškin had observed the genitive rule (Vinokur (1937)1959: 201). While ignoring the ambiguity point (which is interesting as it suggests that the applicability of the genitive rule to adjuncts of time and distance was not self-evident), Puškin pleaded poetic licence: “҈ҽѸѦҍҏ, ӊҸ-ҰѦҝӊӃѽ, ҍҲҸ ҲҽҔ ӁҽӉӃӊҽѦѸҽҾ ҡҲѹѽҸҲӊҸҝҍѦҡҔҽҾ ӊҸґҧӁҸҡҲҧ ҦҸҰҏҡҔҽѦҲ Ӂҽҡ ҡҸ ӊҝѦѸѦӁ һҸѸҸӁҸҡҸӊҽ ҏҰҸҲҝѦқґҾҲҧ indifféremment ҰҸҡґѦ ҸҲҝѹ҃ҽҲѦґҧӁҸҮ ҍҽҡҲѹ҃Ӄ ӰӾ ҝҸҦѹҲѦґҧӁӃҮ ѹ ӊѹӁѹҲѦґҧӁӃҮ ҰҽҦѦң.” (Puškin (1828)1949: 78) Puškin only raised this point in the rough copy of his reply: in the published version, which will be dealt with in 6.2, he declared himself willing to comply rather than referring to his privileged position. Yet he seemed to experience the genitive rule as an arbitrary imposed constraint that curbed his creative potential. He returned to the point in the first chapter of his Egyptian nights:

VICISSITUDES OF THE GENITIVE RULE

167

“[…] ҔҝҸѸѦ Ұҝҽӊҽ ҡҲҽӊѹҲҧ ӊѹӁѹҲѦґҧӁӃҮ ҰҽҦѦң ӊѸѦҡҲҸ ҝҸҦѹҲѦґҧӁҸҊҸ ѹ ѦӂѦ ҔҸҮ-ҔҽҔѹѽ, ҲҽҔ ӁҽӉӃӊҽѦѸӃѽ ҰҸҙҲѹҍѦҡҔѹѽ ӊҸґҧӁҸҡҲѦҮ, ѸӃ ӁѹҔҽҔѹѽ ҸҡҸқѦӁӁӃѽ ҰҝѦѹѸҏӂѦҡҲӊ Ӊҽ ҝҏҡҡҔѹѸѹ ҡҲѹѽҸҲӊҸҝ҃ҽѸѹ ӁѦ ӊѦҦҽѦѸ.” (Puškin (1835)1971: 424) Here there seems to be little left of the extreme restraint in applying poetic license that Trediakovskij insisted on (cf. 2.3): using the accusative is a prerogative, not a responsibility. Puškin exercised his prerogative not just in verse. Safarewiczowa, counting GENACCNEG usage in Puškin’s prose, reports a 20% accusative score (Safarewiczowa 1960: 75; n = 404). This contrasts sharply with the zero accusative score of Karamzin (Safarewiczowa 1959: 99), who was born three decades before Puškin. As far as the genitive rule is concerned, Puškin was the first to stretch poetic licence so far as to turn it into narrative licence. 5.2 Compliance in the 19th century Puškin was not alone in occasionally ignoring the genitive rule. The various frequency counts that have been conducted on 19th-century narrative writing and drama (e.g. Uglitsky 1956, Donǁeva 1962, Papina 1964, Safarewiczowa 1960) reveal that no writer succeeded in avoiding the accusative altogether, although ƀexov came close. The statistics show that non-compliance varies strongly between writers and styles. But does it reveal the clear trend in time that would be expected to surface if writers’ usage were to reflect a progressive linguistic change from genitive to accusative? The accusative scores of writers covered in Safarewiczowa’s admirable 1960 survey can be arranged by year of birth: Puškin (prose) Lermontov Turgenev Dostoevskij Ostrovskij (plays) Tolstoj Mamin-Sibirjak

(1799-1837) (1814-1841) (1818-1883) (1821-1881) (1823-1886) (1828-1910) (1852-1912)

20% 11% 24% 11% 11% 27% 17%

(n = 404) (n = 89) (n = 240) (n = 84) (n = 93) (n = 629) (n = 306)

ƀexov (1860-1904) is not included in Safarewiczowa’s survey but Uglitsky’s (1956: 378) and Donǁeva’s (1962: 27) counts yield extremely low accusative scores for him, including even an absolute zero for ƀajka. In Restan’s 20thcentury count, Ƈrenburg’s (1891-1957) mixed narrative prose and dialogue contains 20% accusatives (Restan 1960: 105; n = 125), i.e. the same score that was found for Puškin’s prose, which was written a century before Ƈrenburg’s. All in

168

W. ANDRIES VAN HELDEN

all, the figures fail to corroborate any progress of the accusative over the 19th century in narrative prose and artistic dialogue. To the knowledge of the present writer, counts for 19th-century case usage in other types of text are rare. As a general impression, writers of such texts largely observed the genitive rule. The accusative score for V.G. Belinskij’s (1811-1848) essayistic work is 4% (Safarewiczowa 1960: 78; n = 72). It may be assumed that the accusative score was even lower in the neutral, non-expressive prose of official reports, scholarly publications or newspapers. So it seems that, in 19th-century artistic writing, GENACCNEG usage is to a certain extent a matter of personal choice. In the expanding education system, responsibilities for maintaining the conventions had been handed over to individual writers. While educated people internalized the rules and applied them without much reflection, artistic writers were more self-confident in linguistic matters and consciously defined their own styles. Note that the available figures are not unambiguous: it is not always clear how various fuzzy cases (see Section 6 below) have been counted. Also, results vary significantly between surveys. Papina (1964: 200) only finds 10% accusatives for Puškin. Uglitsky’s (1956: 376) count, yielding almost zero accusatives for Puškin and 25% for Dostojevskij, even “shows with unmistakable clarity” that the accusative “is becoming increasingly prevalent”; but Uglitsky mentions no sample sizes. In Safarewiczowa’s survey, Gor’kij (1868-1936) has a very high accusative score of 40% (Safarewiczowa 1960: 99; n = 202), albeit in material that was mostly written in the 1930s. But even if the accusative turns out to have gained ground in 19th-century narrative discourse, this does not necessarily corroborate the linguistic change assumption: any advance of the accusative may just as well reflect a change of writers’ attitudes with respect to the standard. 6. Fine-tuning the genitive rule If narrative writers tampered with the genitive rule, grammarians stuck to it until well into the 19th century. But around 1830s they started adding clauses to the categorical version, either allowing for exceptions in certain specific conditions or insisting that the rule be applied even in certain specific conditions. As shown by the quote from Timberlake in 1.5, this can be regarded as evidence supporting the assumption of progressive relaxation of the genitive rule, the receding genitive progressively giving up syntactic positions to the advancing accusative. For this scenario to be plausible, the body of exceptions to the rule must have been increasing slowly but steadily until the present day. If there was no linguistic change to guide the changes, however, any increasing tolerance of accusatives in grammars should either be refuted or be accounted for otherwise, e.g. in terms of shifting attitudes with respect to linguis-

VICISSITUDES OF THE GENITIVE RULE

169

tic norms, shifting views on language, or a shifting division of roles between written and spoken codes. These issues will be explored below. 6.1 The inherent instability of the genitive rule It may be argued that the relaxations of the genitive rule found in 19th-century grammars do not result from linguistic change but from a change in the pattern of interaction between grammarians and language users. At first sight, the genitive rule looks robust and seems simple to apply. This may be true for compositors. But at the turn of the century, the education system had produced a body of active users of the new conventions. In that stage of codification, the new code represented progress. Its use stood for literacy, secularism, creativity and prestige. The group who mastered the standard was eager to apply it and to develop it further, in mutual interaction, outside the schools. 17 As illustrated by Puškin’s attitude described above, the jeunesse dorée who felt at home in the new standard regarded themselves as its new owners. To them, Trediakovskij’s and Lomonosov’s mechanistic transformational approach GENACCNEG would no longer do: negation was more complex than just the addition of the particle ne. (This does not imply that the original codifiers had been short-sighted: they may not have intended their instructions to be followed as literally as they formulated them.) The new users confronted the grammarians with some unintended side effects of the rule, forcing them to refine it. Three loci of inherent instability for the genitive rule are easily pinpointed: long-distance government, semantically affirmative sentences containing a negative particle, and partial negation. In sentences containing these phenomena, a syntactic rule that makes case choice dependent on the presence of the negative particle is bound to produce arbitrary or counter-intuitive decisions. It is on these three issues that 19th-century grammarians started refining the genitive rule. 6.2 long-distance government In Puškin’s sentence (16) adduced in 5.1, which Dmitriev had reprehended, the accusative dva veka is not in fact the direct object of a negated verb: it is the direct object of the infinitive ssorit’, which in turn depends on the negated verb xoǁu.

17

Typically, Karamzin codified Russian word order – a field that Lomonosov had neglected –, not by expounding and prescribing his ideas in grammars but by experimenting with them in his own writing (cf. Kovtunova 1969: 151-185).

170

W. ANDRIES VAN HELDEN

This circumstance buttresses Puškin’s second argument in defence of his use of the accusative. “ҐҝҽѸѸҽҲѹҔҽ Ӂҽҗҽ ѦӂѦ ӁѦ ҰҸҾҡӁѦӁҽ.” (Puškin (1828)1949: 78) Puškin expresses his commitment to the rules of the standard language, using the phrase our grammar. He hints that he is quite willing to comply but sees himself faced with a defective rule: the genitive rule is not transparent because its syntactic scope is not specified. Puškin’s opinion on this point is worked out in the published version of his reply. “ҋҲѹѽ [(16)] ҔҝѹҲѹҔҏ ҰҸҔҽӉҽґҡҾ ӁѦҰҝҽӊѹґҧӁӃѸ. ѯҲҸ ҊґҽҡѹҲ ҊҝҽѸѸҽҲѹҔҽ? ѯҲҸ ҦѦҮҡҲӊѹҲѦґҧӁӃҮ ҊґҽҊҸґ, ҏҰҝҽӊґҾѦѸӃҮ ҸҲҝѹ҃ҽҲѦґҧӁҸӀ ҍҽҡҲѹ҃ѦӀ, ҲҝѦқҏѦҲ ҏңѦ ӁѦ ӊѹӁѹҲѦґҧӁҸҊҸ, ҽ ҝҸҦѹҲѦґҧӁҸҊҸ ҰҽҦѦңҽ. ѢҽҰҝѹѸѦҝ: (17) Ѥ ӁѦ Ұѹҗҏ ԙԽӠԖӹӴ. ѢҸ ӊ ѸҸѦѸ ҡҲѹѽѦ ҊґҽҊҸґ ԙԙӹӢӠԽӸ ҏҰҝҽӊґҾѦѸ ӁѦ ҍҽҡҲѹ҃ѦӀ ӰӾ, ҽ ҊґҽҊҸґҸѸ Ԗӹԓӫ. Ergo, ҰҝҽӊѹґҸ ҡӀҦҽ ӁѦҮҦѦҲ. ғҸӉҧѸѦѸ, ӁҽҰҝѹѸѦҝ, ҡґѦҦҏӀӂѦѦ ҰҝѦҦґҸңѦӁѹѦ: (18) Ѥ ӁѦ ѸҸҊҏ ӊҽѸ ҰҸӉӊҸґѹҲҧ ӁҽҍҽҲҧ ҰѹҡҽҲҧ … ԙԽӠԖӠ, ҽ ҏң ҔҸӁѦҍӁҸ, ӁѦ ԙԽӠԖӹӴ. ѢѦҏңҲҸ ҙґѦҔҲҝѹҍѦҡҔҽҾ ҡѹґҽ ҸҲҝѹ҃ҽҲѦґҧӁҸҮ ҍҽҡҲѹ҃Ӄ ҦҸґңӁҽ ҰҝҸҮҲѹ ҡҔӊҸӉҧ ӊҡӀ ҙҲҏ ҃ѦҰҧ ҊґҽҊҸґҸӊ ѹ ҸҲҸӉӊҽҲҧҡҾ ӊ ҡҏӂѦҡҲӊѹҲѦґҧӁҸѸ? ѢѦ ҦҏѸҽӀ.” (Puškin (1830)1949: 173) Puškin gives Lomonosov his due by using the force metaphor and identifying its origin in physics. But he reinterprets its precise linguistic content: whereas Lomonosov attributes the governing force to the verb, Puškin assigns it to the negative particle. The strength of Puškin’s force is expressed in terms of syntactic distance instead of actual involvement of the object referred to in the action expressed by the verb, as in Lomonosov’s (12). Although Puškin’s position has become the most frequently cited text in the literature on GENACCNEG, the linguistic purport of his argument is rarely taken up. Possibly Raviǁ (1971: 261) was the first to do so: she adduces (18) to illustrate the relevance of the depth of embedding for case choice: each additional infinitive separating the object from the head in the dependency structure of a sentence reduces the probability of the genitive being used (although even the lowest infinitive may trigger the genitive with ni-words such as ni … ni). 18 18 In the literature on GENACCNEG examples are often recycled. Raviǁ (1971: 261) presented (18) without explicitly mentioning its origin. Timberlake ((1975)1986: 347) picked it up there and confirmed Puškin’s timelessness by including it in his collection of contemporary Russian examples.

VICISSITUDES OF THE GENITIVE RULE

171

The use of the genitive at a long syntactic distance from the negative particle is as old as the use of the genitive with direct objects of negated verbs. Even if an infinitive only remotely depended on a negated verb, its direct object could be in the genitive, as shown by this Old Church Slavonic example: (19) Snӆ ǁskӆi ne imatӆ kde glavy podӆkloniti. (Veǁerka 1963: 205) Here the genitive form of the object glavy is considered to be motivated by the negated verb form imatԔ, even though glavy is the object of podԔkloniti, which itself depends only indirectly on imatԔ. The genitive rule had not explicitly provided for such instances; Lomonosov may have regarded it as a minor detail. (Note his own use of a long-distance accusative in his verse (6) adduced in 3.4: ogn’ depends on svodit’, while umel is negated.) The spoken language may have admitted both the accusative and the genitive here: A.A. Barsov had in fact made Puškin’s point in the 1780s, in his Rossijskaja grammatika: “1. ҘҸҊҦҽ ҊґҽҊҸґ ҏҰҝҽӊґҾӀӂѹҮ ӊѹӁѹҲѦґҧӁӃѸ ѹѸѦѦҲ Ұҝѹ ҡѦқѦ ҸҲҝѹ҃ҽҲѦґҧӁҏӀ ҍҽҡҲѹ҃ҏ ӰӾ: ҲҸ ҰҽҦѦң ҡѦҮ ҰѦҝѦѸѦӁҾѦҲҡҾ ӊ ҝҸҦѹҲѦґҧӁӃҮ […]. 2. Ѩ ҡѹҾ ҰѦҝѦѸѦӁҽ ѹѸѦѦҲ ѸѦҡҲҸ ҦҽңѦ ѹ ҰҸҡґѦ ӁѦҸҔҸӁҍҽҲѦґҧӁҸҊҸ ӁҽҔґҸӁѦӁѹҾ ҏҰҝҽӊґҾѦѸҸҊҸ ҦҝҏҊѹѸ ҊґҽҊҸґҸѸ ҸҲҝѹ҃ҽҲѦґҧӁӃѸ […].” (A.A. Barsov (1788)1980: 304). Barsov’s examples include: (20) Ѥ қѦҝҏ ԲԽӫ ӪӰӠԄӫ. (21) Ѥ ңѦґҽӀ ӊӉҾҲҧ ԲԽӫ ӪӰӠԄӫ. (22) Ѥ ӁѦ қѦҝҏ ԲԽӹӡ ӪӰӠԄӠ. (23) Ѥ ӁѦ ңѦґҽӀ ӊӉҾҲҧ ԲԽӹӡ ӪӰӠԄӠ. Barsov continues: “ҫҦӁҽҔҸң ӊ ҡѦѸ ҰҸҡґѦҦӁѦѸ ҡґҏҍҽѦ ѸҸңӁҸ ‹ѹ› ҸқҝҽҲӁҸ ҏҰҸҲҝѦқѹҲҧ ӊѹӁѹҲѦґҧӁӃҮ ӊѸѦҡҲҸ ҝҸҦѹҲѦґҧӁҸҊҸ, ҔҽҔ ҲҸ: (24) Ѥ ‹ӁѦ› ңѦґҽӀ ӊӉҾҲҧ ԲԽӫ ӪӰӠԄӫ.” (A.A. Barsov (1788)1980: 304) 19

19

Newman inserts ne in his edition of Barsov’s text so as to make the preceding statement meaningful (A.A. Barsov (1788)1980: 304). In Tobolova and Uspenskij’s edition, the sentence is reproduced without ne (A.A. Barsov (1788)1981: 212).

172

W. ANDRIES VAN HELDEN

But Barsov’s grammar was never published: the Imperial Education Committee, which had commissioned it, decided that it was too voluminous for a textbook (Uspenskij (1981)1997: 637-643). 20 So Puškin could only appeal to reason, not to authority. Grammarians of Puškin’s own epoch, however, had abandoned A.A. Barsov’s liberal position. Greǁ, in his Praktiǁeskaja russkaja grammatika, stipulated that: “ҭҽҦѦң ҝҸҦѹҲѦґҧӁӃҮ ҰҸґҽҊҽѦҲҡҾ ѹ ҲҸҊҦҽ, ҔҸҊҦҽ ӁҽҝѦҍѹѦ ҸҲҝѹ҃ҽҲѦґҧӁҸѦ ӁҽѽҸҦѹҲҡҾ ҰѦҝѦҦ ҊґҽҊҸґҸѸ, ҰҝѦҦҗѦҡҲӊҏӀӂѹѸ ҏҰҝҽӊґҾӀӂѦѸҏ ҊґҽҊҸґҏ […].” (Greǁ 1834: 666, quoted from Vinokur (1937)1959: 202) Greǁ’s example: (25) ѢѦ ґӀқґӀ ҲѦҝҾҲҧ ӴӢӾԐӾӰӠ. Greǁ’s categorical nyet to Puškin’s suggestion did not last for all of the century: Buslaev ((1858)18754: 249) tolerated it by remarking that “the latest writers” use the accusative freely with infinitives depending on negated xoǁu, mogu or stanu. But this can hardly be regarded as evidence supporting the standard position on the progressive loss of the genitive rule. A.A. Barsov’s grammar was just as tolerant of the long-distance accusative as modern grammars are. Moreover, although long-distance negation represented an early foothold for the accusative, the genitive has survived here until the present day: in recent frequency counts, it still occurs in about three out of ten instances (Mustajoki and Heino 1991: 33), while it has meanwhile disappeared in other positions. 6.3 Affirmative sentences containing ne Any conflict between negative form and affirmative substance is likely to present problems for the application of the genitive rule. What prevails in such instances: the semantics of affirmation or the syntactics of the negative particle? In Russian, some types of sentence containing ne are used to convey messages that are not, or not necessarily, negative. The improbability of a proposition, or doubt about it, is expressed by the insertion of modal adverbs. When inserted in negative sentences, such adverbs could turn them into positive messages: expressing doubt about non(X) boils down to expressing the assumption that X is actually the case. Here marking the direct object with the accusative seemed to make sense. Vostokov therefore introduced an exception to the genitive rule. 20

It was nonetheless influential: all early 19th-century grammarians consulted Barsov’s manuscript (Uspenskij (1981)1997: 629). Buslaev published an extract of it in the mid 19th century.

VICISSITUDES OF THE GENITIVE RULE

173

“ѢҸ ҔҸҊҦҽ ҸҲҝѹ҃ҽҲѦґҧӁҸѦ ӁҽҝѦҍѹѦ ӰӾ ҡҸҍѹӁѦӁҸ ҡ ҰҝѦҦҰҸґҸңѹҲѦґҧӁӃѸѹ ӾԌӴԂ ԊӠ, ԓӫԽӸ ԊӠ, ѹ ҡ ӊҸҰҝҸҡѹҲѦґҧӁӃѸ ҡҸӀӉҸѸ ԊӠ, ҲҸҊҦҽ, ҲѦҝҾҾ ҡѹґҏ ҸҲҝѹ҃ҽӁѹҾ ѹ ӊѽҸҦҾ ӊ ҡҸҡҲҽӊ ӁҽҝѦҍѹҾ ҰҝѦҦҰҸґҸңѹҲѦґҧӁҸҊҸ, ӁѦ ҡҸҸқӂҽѦҲ ҊґҽҊҸґҽѸ ґѹҍӁӃѸ ҡҰҸҡҸқӁҸҡҲѹ ҏҰҝҽӊґҾҲҧ ҝҸҦѹҲѦґҧӁӃѸ ҰҽҦѦңҸѸ […].” (Vostokov 1831: 255-256, quoted from Mustajoki 1985: 17-18) Vostokov’s examples include: (26) ҫӁ ѦҦӊҽ ґѹ ҲѦҝҾѦҲ ӴӢӾԐԋ. (27) ѢѦ ҲѦҝҾѦҗҧ ґѹ ҲӃ ӴӢӾԐԋ? Vostokov’s explicit tolerance of these sentences represents the first official breach in the genitive rule, the oldest exception to be endorsed in a 19th-century grammar. Buslaev ((1858)18754: 250) upheld it. In Mustajoki and Heino’s (1991: 24) survey of modern Russian GENACCNEG, edva li ne and ǁut’ li ne always require the accusative. In questions containing ne that implicitly expect a positive answer (i.e. a denial of the negation), the accusative is strongly dominant (Mustajoki and Heino 1991: 35). Vostokov’s concession cannot, however, be adduced as a symptom of the slow but inexorable advance of the accusative. As in all grammatical discussion on negation (which is characterized by frequent shifts between logical, psychological and communicative vantage points), the new tack may be rather ascribed to changing attitudes on language, or simply be regarded as a refinement of the overly categorical formulation of the genitive rule in the 18th century. 21 6.4 Partial negation Partial negation presents another obvious problem for the genitive rule. What happens if ne is not used to deny a complete proposition but only a particular aspect of it? In written standard Russian, ne is usually placed before the element that is specifically denied, thus avoiding the cleft constructions that are so widespread in the languages spoken on the Atlantic coasts. Such sentences can be regarded as limited modifications of their affirmative counterparts. Cf. Tomson’s example: (28) ѢѦ Ҿ ҔҏҰѹґ ԌӹԐ. (Tomson (1902)1903: 199)

21 The present writer is aware of a potential Achilles’ heel in the present argument: if syntactic considerations, such as the presence of ne, tend to determine case choice in normalized language while semantic considerations are likely to prevail in non-normalized speech, how come sentence (13) is found in Xolmogory? Any considerations on this point will be welcomed.

174

W. ANDRIES VAN HELDEN

Although the sentence denies my buying a house, it is not denied that a house was bought. Using the genitive must have been quite cumbersome here. This applies as well if the direct object itself is the very element to be denied. In Tomson’s following example, the genitive must have been impossible. (29) ғӃқѹҝҽҮ ӁѦ ӰӾӴӾԙԽӫ ҽ ҡӊҽѽҏ. (Tomson (1902)1903: 196) The codifiers probably never intended the genitive rule to apply to instances of partial negation. As the reader may have noticed in 2.3, Trediakovskij uses an accusative for the direct object of a negated verb in the very sentence in which he brings up the genitive rule: (30) ҫҡӁҸӊҽҲѦґҧӁҸѦ ҰҝҽӊѹґҸ, ҔҸҲҸҝҸѦ ҏҲӊѦҝҦѹҲѦґҧӁҸҮ ҝѦҍѹ ҰҝҽӊѹѸӃҮ ҊґҽҊҸґҸѸ ӊѹӁѹҲѦґҧӁӃҮ ҰҽҦѦң ӊҡѦҊҦҽ ҰѦҝѦѸѦӁҾѦҲ ӊ ҝҸҦѹҲѦґҧӁӃҮ ӊ ӍҝҽӉѹҡѦ ҸҲҝѹ҃ҽҲѦґҧӁҸѸ, ѸҸңѦҲ ҦґҾ ҔҝҽҮӁѦҮ ӁҏңҦӃ ӊ ҡҲѹѽѦ ӹӰՂӡ ӴӠӰӠԽӾԊӸӰՂӡ ѹ ӁѦ ҰѦҝѦѸѦӁҾҲҧ. (Trediakovskij (1735)19632: 380) The accusative onyj vinitel’nyj is the direct object of the negated infinitive ne peremenjat’: so the genitive rule, in its categorical interpretation, applies here. But few writers would have considered replacing the accusative by a genitive in this complex instance of partial negation. We may expect the successors of the codifiers to accommodate the users and teachers of the standard language and specify the circumstances under which they would formally exempt writers from applying the genitive rule. Greǁ made a modest attempt to do so by allowing the use of the accusative if ne is separated from the verb by an adverb, as in his example (31) below, in which it is not denied that I take the medicine, just that I am glad to do so. (31) Ѥ ӁѦ ҸѽҸҲӁҸ ҰҝѹӁѹѸҽӀ ԊӾӪԂӢԙԽӴӹ. (Cited from Mustajoki 1985: 17) By implicit extension, Greǁ’s exception could be applied to other parts of speech preceded by ne, as in (28). But Greǁ’s exception remains inadequate. It fails to cover many other instances of partial negation. The verb itself may be the focus of negation and there is no way of telling whether ne is put in front of it to negate the complete proposition or to negate the verb. Tomson’s following example makes this clear: (32) ҫӁ ӁѦ ҏҔҝҽґ ԲԽӫ ԊӹԶԂԌӸ (ҽ ҔҏҰѹґ ѦѦ). (Tomson (1902)1903: 202) Greǁ also fails to cover instances of partial negation where ne occurs before the verb in spite of the focus of negation being elsewhere. Cf. Tomson’s example

VICISSITUDES OF THE GENITIVE RULE

175

(33) from Dal’’s collection of proverbs, where ne precedes the verb although the focus of negation is on the object: 22 (33) ѢѦ ӊӃқѹҝҽҮ ӰӾӴӾԙԽӫ, ӊӃқѹҝҽҮ ҡӊҽѽҏ. (Tomson (1902)1903: 195) All in all, Greǁ’s exception looks like a reluctant concession to the most inescapable linguistic facts. Vostokov’s and Greǁ’s extensions and exceptions had already increased the complexity of what started out as a simple rule in Lomonosov’s days, and rendered its explanation to the population of the expanding school system more difficult. Devising a formulation that covered all instances of partial negation would entail the introduction of complex and unverifiable auxiliary concepts, rendering the rule unmanageably complex while it would not even assist the intelligent language user insofar as it still failed to match the even broader range of the accusative in spoken usage. 23 For a grammarian, treating the other instances of partial negation as ungrammatical was clearly more preferable. Consequently, Greǁ’s exception can hardly be adduced as evidence supporting the progressive advance of the accusative: as it fails to account for Trediakovskij’s accusative in (30), it would even seem to point to a retreat rather than an advance of the accusative in normative grammar.

22 It is tempting to regard the tendency of the negative particle to seek out the specific item to be negated as another product of codification. In the spoken language, and more generally in texts reflecting older stages of Russian, ne tends to remain in place before the finite verb, any specific focus of the negation being identified by the context and the intonation pattern. Cf. on ne byl doma a na rabote instead of on byl ne doma a na rabote, or Savel’eva’s (1989: 145) example ne sut’ to bozi, no drevo, from the Povest’ vremennyx let, where ne precedes the verb although bozi is obviously the focus of the negation. The written language, pursuing precision and lacking intonation, developed ne-movement, or at least expanded it (cf. Savel’eva 1989: 145). As Crockett (1977: 233, 239) points out, ne-movement continues to be subject to various limitations in the written standard as well. On edet ne na sobstvennoj mašine is an instance of partial negation, since ne does not appear in front of the finite verb, thus excluding the reading that no travelling takes place. But not all ambiguity has been removed. The focus of negation may be anywhere within na sobstvennoj mašine: it may be the word group as a whole (the negation opposing his car to public transport), or sobstvennoj (the negation opposing it to someone else’s car), or mašine (the negation opposing it to his own private plane). Ne cannot be inserted inside the group. 23 Conversely, partial negation does not warrant the use of the accusative, at least in 20th-century Russian. According to Paduǁeva (1974: 143), on ne rešil vsex vašix zadaǁ is synonymous with on rešil ne vse vaši zadaǁi, while *on ne rešil vse vaši zadaǁi is categorized as ungrammatical (Paduǁeva 1974: 155). To account for sentences containing such semantic quantifiers, the concept of partial negation itself should be refined or replaced.

176

W. ANDRIES VAN HELDEN

7. Professional linguistics and the genitive rule In the second half of the 19th century, grammarians appear to have made no further concessions to spoken usage. But, as a result of the further expansion of literacy and the rise of the prestige of the natural sciences, reflection on language started to change. The new professional discipline of linguistics endowed it with scientific claims. Two currents arose: comparative and psychological linguistics. One might expect that reliable information on the status of the genitive rule would henceforth be available. Unlike grammarians, linguists were dispensed from the task of giving guidance to the language; they could concentrate on describing what they saw and heard (cf. Mustajoki 1985: 18). Comparative linguists could be expected to sort out the diachronic development of GENACCNEG, while the psychologically oriented linguists could sort out the contemporary situation. It turns out that, at least until the end of the century, the new linguists did not question the genitive rule: they started explaining it rather than debunking it. Does the fact that they did not observe the use of the accusative imply that it was not yet there? In order to reject this hypothesis, the linguists’ blindness to the accusative must be accounted for otherwise. 7.1 The reification of the genitive rule The new linguists were no tabula rasa. They were products of the education system and active educators themselves. They had become linguists because they had been successful in assimilating the contents of the gymnasium curriculum. They were not necessarily eager to decompose what they had learned. Rather, they considered raising the subject matter to a higher level as their mission: their research was expected to provide better explanations for existing educational content rather than to question it. The linguists started out acting as apologists rather than sceptics of received wisdom. By the second half of the 19th century, the rules of the standard language were there as if they had always existed. The genitive rule no longer represented an analytical construct, created to account for a series of observations, let alone an essentially arbitrary convention. Rather, the rule was regarded as part of the language as a natural organism, a discrete entity stemming from the mists of time and travelling along with the language, like a lexical item. Historical linguists virtually imposed the rule on the scribes of the Old Church Slavonic manuscripts as their teachers had imposed it on themselves, and described ancient GENACCNEG usage in terms of compliance with the rule: “ғ ҦҝѦӊӁѦҮҗѹѽ ҡґҽӊҾӁҡҔѹѽ ҲѦҔҡҲҽѽ ҋӊ. ҭѹҡҽӁѹҾ ӊѦҝӁѦѦ ѹ ҰҸҡҲҸҾӁӁѦѦ ҰҸӉҦӁѦҮҗѹѽ ҡҸқґӀҦҽѦҲҡҾ ҙҲҸ ҰҝҽӊѹґҸ Ҹқ ҏҰҸҲҝѦқґѦӁѹѹ ҝҸҦѹҲѦґҧӁҸҊҸ ҰҽҦѦңҽ Ұҝѹ ҦѦҮҡҲӊѹҲѦґҧӁҸѸ ҊґҽҊҸґѦ ҡ ҸҲҝѹ҃ҽӁѹѦѸ.” (Buslaev (1858)18754: 249).

VICISSITUDES OF THE GENITIVE RULE

177

The new linguists considered it their task to explain why it was natural or rational for the genitive rule to exist. A few examples will be adduced below. 7.2 Miklosich’s explanation of the genitive rule Miklosich essentially follows Lomonosov’s transformational view when stating that the direct object genitive that occurs with negated transitive verbs corresponds to the accusative in affirmative sentences. The genitive rule is no longer a feature of standard Russian, East Slavic or Church Slavonic: it is a feature of Slavic. Miklosich explains the origin and essence of the genitive of negation as follows: “Er ist wohl Partitiv und die Partition hat ihren Grund in der Kraft der Negation, indem die durch das Verbum ausgedrückte Tätigkeit vom ganzen und von jedem, selbst dem kleinsten Teile desselben negiert wird.” (Miklosich 18832: 498) Here another variety of Lomonosov’s government force can be distinguished. Miklosich’s force is assigned to the negation as a semantic item, whereas Puškin assigned his force to the negative particle as a syntactic item. It is not really obvious to the present writer what Miklosich’s force of the negation is supposed to extend over. “Desselben” seems to refer to the verb. Potebnja renders the last part of Miklosich’s quote as: “ѩѦҮҡҲӊѹѦ, ӊӃҝҽңѦӁӁҸѦ ҊґҽҊҸґҸѸ (ҡ ҸҲҝѹ҃ҽҲѦґҧӁҸҮ ҍҽҡҲѹ҃ѦҮ), ҸҲҝѹ҃ҽѦҲҡҾ ҸҲӁҸҡѹҲѦґҧӁҸ ҔҽҔ ҃ѦґҸҊҸ ҰҝѦҦѸѦҲҽ, ҲҽҔ ѹ ҔҽңҦҸҮ ҦҽңѦ ӁҽѹѸѦӁҧҗѦҮ ѦҊҸ ҍҽҡҲѹ.” (Potebnja (18882)1958: 320) In Potebnja’s rendering, Miklosich’s force recalls Lomonosov’s variable syntactic force as applied to the partitive genitive (cf. (12) in 4.2): negation entails zero government force. 7.3 Potebnja’s explanation of the genitive rule Potebnja himself, as ҽ psychologically oriented linguist, adopts a novel view on He does not agree with Miklosich’s suggestion that the genitive is essentially partitive. He does away with Lomonosov’s governing force and even with transformations, replacing them by a fresh metaphor: the negative particle as a germ.

GENACCNEG.

“ғӃҝҽңѦӁѹѦ “ҦѦҮҡҲӊѹҲѦґҧӁӃҮ ҊґҽҊҸґ ҡ ҸҲҝѹ҃ҽӁѹѦѸ ҲҝѦқҏѦҲ ҝҸҦѹҲѦґҧӁҸҊҸ” ҔҽңѦҲҡҾ ҸҗѹқҸҍӁӃѸ ӊ ҲҸѸ ҸҲӁҸҗѦӁѹѹ, ҍҲҸ ӉҽҔґӀҍҽѦҲ ӊ ҡѦқѦ ҰҝҸҲѹӊҸҝѦҍѹѦ, ѹқҸ ҦѦҮҡҲӊѹҲѦґҧӁӃҮ ҊґҽҊҸґ ѦҡҲҧ ҲҽҔҸҮ, ҔҸҲҸҝӃҮ ҦҸҰҏҡҔҽѦҲ Ұҝѹ ҡѦқѦ ҰҝҾѸҸҮ ҸқӆѦҔҲ ӊ ӊѹӁѹҲѦґҧӁҸѸ, ѹѸѦӁӁҸ ѹ ӁѦ ҦҸҰҏҡҔҽѦѸӃҮ (Ѧҡґѹ ӁѦ қҝҽҲҧ ӊ ӊӁѹѸҽӁѹѦ ҸҲҔґҸӁѦӁѹҮ ҸҲ Ұҝҽӊѹґҽ) ҊґҽҊҸґҸѸ ҡ ҸҲҝѹ҃ҽӁѹѦѸ, ҔҸҲҸҝҸѦ, ҰҝҸӁѹҔҽҾ ӊҡѦ ӉҽґҸңӁҸѦ ӉӁҽҍѦӁѹѦ ҊґҽҊҸґҽ ҦѦҮҡҲӊѹҲѦґҧӁҸҊҸ, ҝҽӉҝҏҗҽѦҲ ѦҊҸ

178

W. ANDRIES VAN HELDEN

ҰҝѦңӁӀӀ ҡӊҾӉҧ ҡ ҸқӆѦҔҲҸѸ. ҭҸҙҲҸѸҏ ӁѦґҧӉҾ ҸҰҝҽӊҦҽҲҧ ҸқӃҍҽҾ ҰѹҡҽҲҧ ҸҲҝѹ҃ҽҲѦґҧӁҏӀ ҍҽҡҲѹ҃ҏ ӰӾ Ұҝѹ ҊґҽҊҸґѦ ҝҽӉҦѦґҧӁҸ. ѩѦҮҡҲӊѹҲѦґҧӁӃҮ ҊґҽҊҸґ ҡ ҸҲҝѹ҃ҽӁѹѦѸ ӁѦ ѦҡҲҧ ҏңѦ ҦѦҮҡҲӊѹҲѦґҧӁӃҮ, ѹ ҰҸҲҸѸҏ ҡҲҽӊѹҲҡҾ ҾӉӃҔҸѸ Ӂҽ ҸҦӁҏ ҦҸҡҔҏ ҡҸ ҡҝѦҦӁѹѸѹ, 24 ҡӁҽқңѦӁӁӃѸѹ ҸҲҝѹ҃ҽӁѹѦѸ.” (Potebnja (18882) 1958: 320) While it does not exactly account for the genitive obyǁaja in the quote itself, Potebnja’s analysis reveals a substantial shift away from earlier views on the genitive rule. In Lomonosov’s conception, the negative particle triggers a formal rearrangement of the syntactic relations between the verb and the roles that are defined by it while the underlying, referential, semantics remained unaffected (except for the negation). Potebnja’s analysis is deeper in the sense that the negation does not transform the syntactic relations of the verb but the verb itself. The new verb, the voice value of which has been “permeated” by the negative particle, defines a different class of situations, containing a new role, which is consequently expressed by a different case. This is an innovation inasmuch as the genitive of negation is no longer a syntactic but a semantic phenomenon. The genitive is not mechanically assigned to a role: it is used because it expresses a role. Although the genitive is no longer accounted for in terms of a syntactic rule, the data on which the analysis is based remain unaltered. Potebnja’s analysis applies to the variety of Russian in which the direct object of negated verbs is always in the genitive, i.e. the set of sentences that are generated by the genitive rule. As the quote makes clear, Potebnja is interested in explaining the nature of the genitive of negation, not in accounting for the exceptions. 7.4 A.V. Barsov’s explanation of the genitive rule A.V. Barsov’s explanation of the direct object genitive of negation is essentially a schoolteacher’s version of Potebnja’s. He uses simple words and an example that evokes the etymologies of perexodnyj and stradatel’nyj. “ҫҲҝѹ҃ҽӁѹѦ Ұҝѹ ҊґҽҊҸґҽѽ ҰҸҔҽӉӃӊҽѦҲ, ҍҲҸ ҰҝѦҦѸѦҲӃ […] ӁѦ ҰҸҦӊѦҝҊҽӀҲҡҾ ҦѦҮҡҲӊѹӀ, ӰӾԒԂӴӠԙӠԐՂ: (34) Ѥ ӁѦ қѹґ ҡҸқҽҔѹ. 24 By srednimi Potebnja refers to verbs with the particle -sja, for which Lomonosov ((1757)1952: 563) had ruled out the use of the direct object accusative: bojat’sja, for instance, takes the genitive. This rule has survived in the present-day standard but has, like the genitive rule, a long tradition of non-compliance. Butorin (1966: 128-132) provides examples from (even early) 19thcentury texts where the accusative is used with -sja-verbs: Djadju bojalis’ vse.

VICISSITUDES OF THE GENITIVE RULE

179

ѣҡґѹ Ҿ ӁѦ қѹґ, ҲҸ ҡҸқҽҔҽ ѹ ӁѦ ҍҏӊҡҲӊҸӊҽґҽ қҸґѹ, ҰҸҲҸѸҏ ҍҲҸ ҦѦҮҡҲӊѹѦ ѦѦ ӁѦ ҔҸҡӁҏґҸҡҧ, ӁѦ ԗӾӢӾԖӹԌӠԊӹ Ӂҽ ӁѦѦ.” (A.V. Barsov 1880: 14) This explanation is so obvious and convincing that Barsov needs another explanation to account for those languages in which the direct object of negated verbs is not in the genitive: “ғ ґҽҲѹӁҡҔҸѸ ӊѹӁѹҲѦґҧӁӃҮ ҰҸ ҽӁҽґҸҊѹѹ ҡ ӊѹӁѹҲѦґҧӁӃѸ Ұҝѹ ҊґҽҊҸґҽѽ ҦѦҮҡҲӊѹҲѦґҧӁӃѽ қѦӉ ҸҲҝѹ҃ҽӁѹҾ.” (A.V. Barsov 1880: 14) So the new semantic linguists raised the level of sophistication of the explanations for the use of the genitive. But, with one notable exception, to be discussed in the next section, they only concealed the genitive rule: they did not discard it. As the syntactic genitive rule continued to determine the facts to be accounted for, the new semantic explanations can still be regarded as apologies for the rule. 7.5 Nekrasov’s refutation of the genitive rule As far as the present writer is aware, N.P. Nekrasov (1828-1913) was the only 19th-century renegade of the genitive rule. It is possible to view his exposure of the accusative as an early symptom of a language change going on. But is also possible to regard it as an example of discovery driven by a new ideological programme. Nekrasov was not a grammarian or a prominent member of the new guild of linguists. Being a member of the Council of the Minister of Education, he was interested in the cultural development of Russian society. As a thinker, Nekrasov was a slavophile. He rejected the application of western mental models with universal claims to the minds of the unique Russian people. This view had implications for the Russian language, which was considered to be the expression of the collective spirit (dux) of the people. This implied for Russian grammar that the western models that had been adopted when creating and developing the standard language in the days of Lomonosov and Trediakovskij were unfit for it. The rules acted as a straitjacket, stifling the expression of the living Russian soul: “[…] Ӂҽҗҽ ҊҝҽѸѸҽҲѹҔҽ ѹӉҸқѹґҏѦҲ ҡѽҸґҽҡҲѹҔҸӀ ѹ ӊҸӊҡѦѸ ӁѦҰҝѹѸѦӁѹѸҽ Ҕ ҰҸӁѹѸҽӁѹӀ ҝҸҦӁҸҊҸ ҾӉӃҔҽ. ғҝѦѸҾ ӁҽґҸңѹґҸ Ӂҽ ӁѦѦ ҲҾңѦґӃѸ ҊӁѦҲҸѸ ҸқӂѹѦ ҡҏѽѹѦ ҰҝѹѦѸӃ ѹ ӊӉҊґҾҦӃ, ҸҲ ҔҸҲҸҝӃѽ ӁѦӊҸґҧӁҸ ҡҲҽӁѦҗҧ ӊ ҲҏҰѹҔ Ұҝѹ ѹӉҏҍѦӁѹѹ ңѹӊҸҮ ҝѦҍѹ.” (Nekrasov 1865: 1) As far as textbooks of grammar are concerned,

180

W. ANDRIES VAN HELDEN

“ғѦӉҦѦ ӊѦѦҲ ҸҦӁѹѸ ѹ ҲѦѸ ңѦ ҲґѦҲӊҸҝӁӃѸ ҦҏѽҸѸ Ұҝҽӊѹґ ѹ ѹҡҔґӀҍѦӁѹҮ. ғѦӉҦѦ ҾӉӃҔ ҝҽҡҡѸҽҲҝѹӊҽѦҲҡҾ ҔҽҔ ҲҝҏҰ, ӁѦҡѸҸҲҝҾ Ӂҽ ҰҝѦҦѹҡґҸӊѹҾ, ӊ ҔҸҲҸҝӃѽ ҽӊҲҸҝӃ ҸқѦӂҽӀҲ ҸқҝҽӂҽҲҧҡҾ ҡ ҾӉӃҔҸѸ, ҔҽҔ ҡ ңѹӊӃѸ ҸҝҊҽӁѹӉѸҸѸ.” (Nekrasov 1865: 1) Nekrasov’s way out of the deadlock is a radical semantic approach to the formal phenomena that are traditionally dealt with in terms of syntactic rules. He assumes that, for every member of a morphological category, it is possible to establish a general, essential meaning, which is manifested in a set of special meanings that in turn surface in actual usage, their interpretation depending on contextual, situational and intonational factors. While applying his semantic theory to the categories of the verb, Nekrasov (1865: 39ff.) makes an excursion into some Russian cases. He observes that the rules that dictate case government in the grammars obscure the fact that the use of cases is much freer in living usage. He adduces, as an example, the accusatives of duration and distance and the (“so-called”) direct object accusative, all of which are, in his view, just instantiations of the general meaning of the accusative. Thus, so-called intransitive verbs can be combined with the accusative, just like so-called transitive verbs: (35) ҋҰҽҲҧ ӊҡӀ ӁҸҍҧ. Conversely, the genitive turns up with verbs that traditional grammar does not allow governing it, as in: (36) ѢҽѦѽҽґҸ ҊҸҡҲѦҮ. This turns government into a redundant concept, a mail of the straitjacket that the codifiers of neo-languages have imposed on living language to curb its expressive potential. So Nekrasov’s programme implied for case: 1. replacing explanation of case usage in terms of syntactic government by explanation in terms of semantic invariants; 2. replacing analysis of grammatical rules by analysis of observed living speech; 3. systematic checking of the language on material that justifies these choices by failing to conform to government rules. Implementing the programme inevitably resulted in the unveiling of the use of the accusative to mark direct objects of negated verbs. As Nekrasov notes:

VICISSITUDES OF THE GENITIVE RULE

181

“[…] ҝҸҦѹҲѦґҧӁӃҮ ҰҽҦѦң ҡҲҽӊѹҲҡҾ ҰҸҡґѦ ҊґҽҊҸґҸӊ ҡ ҸҲҝѹ҃ҽӁѹѦѸ ӊҡѦҊҦҽ, ҔҸҊҦҽ ҡѹґҽ ҝѦҍѹ ҡҸҡҝѦҦҸҲҸҍѹӊҽѦҲҡҾ Ӂҽ ҸҲҝѹ҃ҽӁѹѹ ҦѦҮҡҲӊѹҾ ҸҲ ҰҝѦҦѸѦҲҽ, ѦҊҸ ҸҰҝѦҦѦґҾӀӂѦҊҸ, ӁҽҰҝ. (37) ѢѦ ҍѹҲҽҲҧ ҰѹҡҧѸҽԀ. (38) ѢѦ ҡҰҽҲҧ ӁҸҍѦҮ. ѣҡґѹ ңѦ ҡѹґҽ ҝѦҍѹ ҡҸҡҝѦҦҸҲҸҍѹӊҽѦҲҡҾ қҸґѦѦ Ӂҽ ҰҝѦҦѸѦҲѦ, ҍѦѸ Ӂҽ ҸҲҝѹ҃ҽӁѹѹ ҸҲ ӁѦҊҸ ҦѦҮҡҲӊѹҾ, ҲҸ ҏҰҸҲҝѦқґҾѦҲҡҾ ѹ ӊѹӁѹҲѦґҧӁӃҮ ҰҽҦѦң: (39) ѢѦ ҍѹҲҽҲҧ ҰѹԀҡҧѸҽ. (40) ѢѦ ҡҰҽҲҧ ӁҸҍѹ.” (Nekrasov 1865: 41) Now whether or not Nekrasov was the first to discover the accusative depends on the definition of discovery. If discovery requires: 1. observing a fact, 2. deciding to regard the observation as relevant, and: 3. proposing a plausible explanation for the observation and the fact, Nekrasov may be regarded as the discoverer of the accusative. The grammarians and linguists who went before him are likely to have observed the use of the accusative but they failed to regard the observation as relevant. Puškin had both made the observation and regarded it as relevant but he did not provide an explicit explanation for it. Nekrasov was the first to propose an underlying principle that determined case choice for GENACCNEG, thus presenting an alternative explanation for the facts that the genitive rule is supposed to handle. According to some historians of science, however, for an observation to be a discovery a fourth requirement must be fulfilled: 4. acceptance among the prospective discoverer’s peers. If this requirement is added, Nekrasov did not discover the accusative. Although an advisor to the minister of education, he did not impose his slavophile cultural agenda. Only linguists practising generally accepted frameworks counted as serious professionals; while they were not ready for radical changes, as we have seen above, they identified internationally accepted with universal. By rejecting the universality of their methods, Nekrasov marginalized himself. Three decades later, Baudouin de Courtenay adduced the fate of Nekrasov’s linguistic efforts when exposing the complacency of the powers that be in linguistics, rejecting innovative, unprejudiced and courageous approaches that force them to think for themselves:

182

W. ANDRIES VAN HELDEN

“ѷҝѹҦ҃ҽҲҧ ҰҾҲҧ ґѦҲ ҲҸѸҏ ӁҽӉҽҦ Ѣ. ҭ. ѢѦҔҝҽҡҸӊ ӊ ҡӊҸѦѸ ҡҸҍѹӁѦӁѹѹ ҡҦѦґҽґ ҰҸҰӃҲҔҏ ҸҲӁѦҡҲѹҡҧ ҡҽѸҸҡҲҸҾҲѦґҧӁҸ Ҕ ҝҏҡҡҔҸѸҏ ҊґҽҊҸґҏ; ӁҸ ѦҊҸ ҰѦҝѦҔҝѹҍҽґѹ ѹ ӁҽҔѹӁҏґѹҡҧ Ӂҽ ӁѦҊҸ ҡ ҸңѦҡҲҸҍѦӁѹѦѸ. ҘҽҔ-ҦѦ ҰҸҡѸѦґ ҸӁ, қҏҦҏҍѹ ҲҸґҧҔҸ ҝҏҡҡҔѹѸ, ӊӉҊґҾӁҏҲҧ ҡҸқҡҲӊѦӁӁӃѸѹ ҊґҽӉҽѸѹ Ӂҽ ӍҽҔҲӃ ҝҏҡҡҔҸҊҸ ҾӉӃҔҽ ѹ ӊѹҦѦҲҧ ӊ ӁѦѸ ҲҸ, ҍҲҸ ӊ ӁѦѸ ҦѦҮҡҲӊѹҲѦґҧӁҸ ѦҡҲҧ, ҽ ӁѦ ҲҸ, ҍҲҸ ѦѸҏ ӁҽӊҾӉӃӊҽѦҲҡҾ ҰҸ җҽқґҸӁҏ ҡҝѦҦӁѦӊѦҔҸӊӃѽ ґҽҲѹӁҡҔѹѽ ҊҝҽѸѸҽҲѹҔ. ҋӊҸѦҸқҝҽӉӁҸѦ “ӉҽҰҽҦӁѹҍѦҡҲӊҸ”, ӊӃӉӊҽӁӁҸѦ, ҔҸӁѦҍӁҸ, ҸҰҽҡѦӁѹѦѸ, ҍҲҸ ӊ ҡґҏҍҽѦ ҰҝѹӁҾҲѹҾ ҏҍѦӁѹҾ ѢѦҔҝҽҡҸӊҽ ҰҝѹҦѦҲҡҾ ҰҸҗѦӊѦґѹҲҧ ѸҸӉҊҽѸѹ, ҽ ӊѦҦҧ “Denken ist schwer und gefährlich!”. һҏҍҗѦ ҏқҽӀҔѹӊҽҲҧ ҡѦқҾ ҰҸӊҲҸҝѦӁѹѦѸ ҍҏңѹѽ ѸӃҡґѦҮ, – ґѹҗҧ қӃ ҲҸґҧҔҸ ӁѦ ҲҝѦӊҸңѹҲҧ!, ґѹҗҧ қӃ ҲҸґҧҔҸ ӁѦ ҲҝѦӊҸңѹҲҧ!” (Baudouin de Courtenay (1901)1963: 363) So nobody picked up Nekrasov’s observations on GENACCNEG, hidden in a misjudged book about verbs. 25 Nekrasov turned to classic comparative linguistics and never returned to the subject. 8. The demise of the genitive rule So throughout the 19th century, the genitive rule continued to control both reflection on language and actual written discourse, with the partial exception of artistic writing. The first decades of the 20th century, on the other hand, are generally considered to have witnessed the final breakthrough of the accusative. Under the standard assumption, the appearance of the accusative in the early 20th century must be ascribed to the progressive linguistic change having reached a critical mass that could no longer be ignored. In the alternative scenario, it must be demonstrated that, if a breakthrough took place at all, it is due to changing linguistic attitudes, triggering the disappearance of the rule or at least an extension of the domain in which it was ignored. It must, furthermore, be shown that the changes affected usage to the extent of accounting for the shifts that actually took place in the 20th century. Three early 20th-century shifts will be checked here on their relevance for GENACCNEG: a shift in linguistics, a shift in normative grammar and a shift in education. Their relative impacts on the genitive rule will be assessed. 8.1 The progress of linguistics If the accusative had been there all the time, what finally made it discoverable? A brief historical account of comparative linguistics may elucidate this question. 25

Nekrasov’s approach to case, with its general and special meanings, presages Jakobson’s ((1936)1971) classic case theory. Jakobson does not mention Nekrasov there but acknowledges his influence in his structural analysis of the Russian verb (Jakobson (1932)1971: 4-5).

VICISSITUDES OF THE GENITIVE RULE

183

At the beginning of the 19th century, reflection on language was perceived from a utilitarian viewpoint: by promoting the development the standard language and enhancing serious thought, it supported the development of the nation. In the course of the century, additional reasons for linguistic reflection came up. The quest for credentials for the nation state produced comparative linguistics, which was popularly identified with the construction of pedigrees for nations. Comparative linguistics went through three stages of methodological expansion. In the classicist stage of comparative linguistics, which, as far as Slavic is concerned, is often associated with A.X. Vostokov (1781-1864), documents represented the dominant source of information on the origins of the language. In 1828, for example, the Academy of Sciences organized the Great Archaeographical Expedition, in which 3,000 manuscripts were uncovered. In the romantic stage of comparative linguistics, which is associated with Vostokov’s pupil I.I. Sreznevskij (1812-1880), contemporary rural dialects became a second source of information on earlier stages of a language. This broadened the view on the role of language in society: it was Sreznevskij who acknowledged that the coexistence of spoken and written codes could characterize older as well as contemporary linguistic communities. In the last part of the 19th century, comparative linguistics entered its scientist stage, which can be associated with Sreznevskij’s pupil J. Baudouin de Courtenay (1845-1929). New concepts, terminology and discourse were imported from neighbouring disciplines, including psychology, bi0logy and geology. More important, linguistics started pursuing unprejudiced fact-finding and developed an international professional community. Linguists from different countries published in the same journals, participated in joint networks and scrutinized each other’s results. Even non-Slavs could become slavists. Erich Berneker (1874-1937) and Aleksandr I. Tomson (1860-1935) were exponents of the third wave. Besides being well-trained comparative linguists, they were used to looking at language from the outside. To Berneker, Russian case usage was a phenomenon to be observed, recorded, reported and explained as any other. Tomson had started his career in the 1880s describing and analysing the Armenian dialect of Tbilisi. 26 He, too, was trained to detach spoken usage from its written counterpart and to take it seriously. His work on Russian is full of subtle observations of spoken usage. He had assimilated the concepts of psychological linguistics and was particularly aware of the distorting effect of lan-

26

N.Ja. Marr coarsely attacked Tomson’s work in the early 1890s, forcing him eventually to abandon the field (Larcev 1988: 197-200).

184

W. ANDRIES VAN HELDEN

guage attitudes, as can be illustrated by his remarks on the genitive of temporal limitation, which had been declared deceased by other grammarians: “ѣҡґѹ ҸқҝҽҲѹҲҧҡҾ ҡ ӊҸҰҝҸҡҸѸ Ҹ ҰҝҽӊѹґҧӁҸҡҲѹ ҙҲѹѽ ӍҝҽӉ Ҕ ҸқҝҽӉҸӊҽӁӁӃѸ ҍѹҡҲҸ ҝҏҡҡҔѹѸ ґӀҦҾѸ, ҲҸ ҸӁѹ ҸқӃҔӁҸӊѦӁӁҸ ҡӁҽҍҽґҽ ҸҲҔҝѦӂѹӊҽӀҲҡҾ ҸҲ Ӂѹѽ. ҭҸҲҸѸ ҸҔҽӉӃӊҽѦҲҡҾ, ҍҲҸ ҦґҾ ҡґҏѽҽ ҊҝҸѸҽҦӁҸҊҸ қҸґҧҗѹӁҡҲӊҽ ҲҽҔѹѦ ӍҝҽӉӃ ҡҸӊѦҝҗѦӁӁҸ ҰҝѹӊӃҍӁӃ ѹ ӁѦ ӊҸӉқҏңҦҽӀҲ ѹѽ ӊӁѹѸҽӁѹҾ ҔҽҔ ӁѦҰҝҽӊѹґҧӁҸҡҲҧ, ѹ ҍҲҸ ѸӁҸҊѹѦ ѹӉ Ӂѹѽ ҏҰҸҲҝѦқґҾӀҲ ҡҽѸѹ ҲҽҔѹѦ ӍҝҽӉӃ ӊ ҦҸѸҽҗӁѦѸ ҾӉӃҔѦ, ҸҡҸқѦӁӁҸ ҡ ҦѦҲҧѸѹ.” (Tomson 1908: 296) At the end of his first article on GENACCNEG, Tomson ((1902)1903: 232ff.) even sketches the elicitation procedures that are needed to obtain a reliable description of stigmatized linguistic items, proposing methods that would eventually be applied by Labov or, in Russia, Andreev (1963). Finally, Berneker and Tomson, unlike Nekrasov, met the fourth requirement for scientific discovery: they belonged to the professional network. The scientific community could be expected to take their findings seriously. 8.2 Berneker and Tomson’s discovery of the accusative Berneker is often regarded as the first linguist to point to the existence of the negated direct object accusative in Russian. On the last page of a paper on animacy he remarks that exceptions to the genitive rule are not uncommon (Berneker 1904: 386). He adduces accusative examples from narrative prose, thus introducing the first official counterexamples to the genitive rule into linguistics. He does not propose an explanation. Tomson, who also worked on animacy, devoted a series of publications partly or wholly to GENACCNEG. 27 He makes it absolutely clear that the use of the accusative is widespread in all types of Russian. He observes that the accusative occurs much more frequently in both popular speech and the ordinary speech of educated Russian society than in text (Tomson (1902)1903: 207) and provides a host of examples, not only from the classics but also from folk sources and the colloquial speech of various social groups. Tomson demonstrates that the use of the accusative is not at all limited to the classic borderline cases that were discussed in Section 6: the accusative occurs in core negation, i.e. in negative sentences that are in no way syntactically 27

Tomson (1902)1903 is his first article on the subject (the 1902 version appeared in a festschrift for Fortunatov in Warsaw). The printed version of Berneker 1904, the manuscript of which is dated January 1901, is cited in Tomson 1902(1903), which suggests that they corresponded. Tomson 1903 also deals with GENACCNEG but has been unavailable to the present writer. Some new aspects of GENACCNEG are dealt with in Tomson 1908. Tomson’s most important findings are resumed in Tomson (Thomson) 1912.

VICISSITUDES OF THE GENITIVE RULE

185

complex, referentially affirmative or partially negative (Tomson (1902)1903: 206ff.). But this does not, in his view, imply that the accusative and the genitive are interchangeable: “Aber soviel ist Tatsache, […] daß in den negierten Sätzen, in welchen ihn die einen gebrauchen, die anderen ihn bei derselben seelischen Stimmung auch gebrauchen, daß also eine Scheidung dieser Kasus durch das unbewußte Sprachgefühl in den meisten Fällen fast strikt durchgeführt wird.” (Tomson 1912: 253) The principles that rule the strict separation of the two cases are semantic and recall the explanation proposed by Nekrasov. Tomson in fact presented a theory of GENACCNEG. It will be dealt with in more detail in 9.1. 8.3 Implications for linguistics Professional linguists accepted Tomson’s findings and started thinking about their implications. How was the new viewpoint accommodated? Since, in the early 20th century, linguists were almost by definition comparative linguists, their discussion concentrated on the implications of the nonexistence of the genitive rule for Common Slavic. Tomson’s representation of the linguistic facts now seemed so self-evident that it was hard to imagine that the data of modern Russian could ever have been different in earlier periods. Tomson himself, adducing the coexistence of accusative and genitive direct objects in negated sentences in West Latvian dialects, surmises that it must have been possible to use the accusative in earlier stages of the language as well, all the way back to Balto-Slavic (Tomson 1912: 259). Notwithstanding the absence of any reliable instances of the accusative in the written sources until the end of the Middle Ages (e.g. Borkovskij 1978: 347), this assumption is sometimes found in modern handbooks as well (e.g. Vlasto 1986: 209). 8.4 Implications for grammar While, to comparative linguists, the discovery of the accusative was just a result to be used in further research, the matter was slightly more complex to the writers of grammars of Russian. Literacy had become a mass phenomenon. Even if only a third of the population of the Russian empire was literate by the end of the 19th century, this still implied that negative sentences were written by millions and read by dozens of millions of persons every day. The production of books had risen from 1,500 titles each year in the late 1850s to 12,000 in the 1890s. In the first decade of the 20th century Russia became the world leader in book production, in terms of both titles and copies (Sokolov 2002: 172). The conventions of Standard Russian

186

W. ANDRIES VAN HELDEN

were firmly established. A century and a half after Lomonosov, grammar had developed its own momentum. Linguistics and grammar had grown apart. What, in this situation, can be the impact of the linguists’ decision that the genitive rule is no longer a linguistic fact? If Lomonosov was not prepared to change the standard practice of a dozen or two Academy compositors, who would be interested in trying to alter the linguistic habits of millions of language users? This presented a dilemma for scholars who were both linguists and grammarians, which is visible in the writings of Bogorodickij, a linguist of renown from the Kazan’ School. In an article on the cases that is intended for schoolteachers, Bogorodickij takes up an apologetic position, presenting the umpteenth self-evident historical explanation for the genitive rule, which is qualified this time as a variety of the genitive of deprivation: “[…] ӁѦҲҝҏҦӁҸ ҰҸӁҾҲҧ ҏҰҸҲҝѦқґѦӁѹѦ ҝҸҦѹҲѦґҧӁҸҊҸ ҰҽҦѦңҽ ҰҸҡґѦ ҦѦҮҡҲӊѹҲѦґҧӁӃѽ ҊґҽҊҸґҸӊ ҡ ҸҲҝѹ҃ҽӁѹѦѸ, ӁҽҰҝ. (41) Ѥ ҦҽӀ ԌӾӰӸԄӠ. (42) Ѥ ӁѦ ҦҽӀ ԌӾӰӾԄ. ғҸ ӊҲҸҝҸѸ ҡґҏҍҽѦ ҝҸҦѹҲѦґҧӁӃҮ ҰҽҦѦң ҡҲҽґ ҏҰҸҲҝѦқґҾҲҧҡҾ ԗӹ ԂӰԂԊӹԄӠӠ ҡ ҲҽҔѹѸѹ ҡґҏҍҽҾѸѹ, ҔҽҔ (43) һѹҗҽӀҡҧ ӰԂԙԊӾԌԙԽӴԂ. ҲҽҔ ҔҽҔ ҍҽҡҲѹ҃ҽ ӰӾ ҲҸңѦ ѹѸѦѦҲ ҡѸӃҡґ ґѹҗѦӁѹҾ.” (Bogorodickij 19092: 457) In an article for linguists, Bogorodickij presents a balanced discussion on the negated direct object genitive in which it is proposed to regard it as a variety of the partitive genitive, plus a footnote in which the use of the accusative is acknowledged and a semantic explanation is suggested: “ғҰҝҸҍѦѸ ӁѦҝѦҦҔҸ ӊ ҲҽҔѹѽ ҸҲҝѹ҃ҽҲѦґҧӁӃѽ ӊӃҝҽңѦӁѹҾѽ ҦҸҰҏҡҔҽѦҲҡҾ ѹ ӊѹӁѹҲѦґҧӁӃҮ ҰҽҦѦң, ҰҸҍҲѹ қѦӉ ӊҡҾҔҸҊҸ, Ӂҽ ҰѦҝӊӃҮ ӊӉҊґҾҦ, ҡѦѸҽӉѹҸґҸҊѹҍѦҡҔҸҊҸ ҝҽӉґѹҍѹҾ. ҫҦӁҽҔҸ ӉҦѦҡҧ ѸҸңӁҸ ҏҡҲҽӁҸӊѹҲҧ ҝҽӉӁѹ҃ҏ ӊ ҸҲҲѦӁҔѦ ӉӁҽҍѦӁѹҾ […]: (44) Ѥ ӁѦ ҍѹҲҽґ ԲԽӠ ԗӠԙӸԐԂ. (ҰҝҸҡҲҸ ҏҔҽӉӃӊҽѦҲҡҾ ҸқӆѦҔҲ ҦѦҮҡҲӊѹҾ ӊ ѦҊҸ ҃ѦґҸҡҲѹ ѹґѹ ҡҸӊҸҔҏҰӁҸҡҲҧ); (45) Ѥ ӁѦ ҍѹҲҽґ ԲԽӠԖ ԗӠԙӾԐ. (ҏңѦ ҡ қҸґѦѦ ҍҽҡҲӁӃѸ ӉӁҽҍѦӁѹѦѸ, ӊ ҡѸӃҡґѦ: ӁѦ ҰҝѹҡҲҏҰҽґ Ҕ ҍҲѦӁѹӀ).” (Bogorodickij 1912: 219-220)

VICISSITUDES OF THE GENITIVE RULE

187

It may be concluded that the rejection of the genitive rule as a linguistic fact did not automatically entail its abolition in the reality of the standard language. Was its disappearance in the course of the 20th century just a matter of time, or did it require specific actions and circumstances to make it happen? 8.5 ƀernyšev’s Pravil’nost’ i ǁistota russkoj reǁi Linguistics was not the only type of reflection on language to change. The rise of literacy and the social prestige of the standard language gradually modified the mission and position of prescriptive grammar. In the early days of nation building, grammar had been the business of codifiers: they developed the standard language as a means to enhance the adequacy of collective linguistic behaviour. After Karamzin, the code was largely done: the new generation of grammarians turned to fine-tuning. In the course of the 19th century, they moved on to explication of the code, both for teaching and apologism, while some of them eventually moved away to engage in linguistics. Meanwhile the social realities of a stratified society produced a new need for prescriptive reflection on language. This time, it was not the nation state that promoted collective linguistic behaviour: by now, there was a homogeneous class of users of the standard. Instead, as the social advantages of access to the standard had become obvious, individual language users wishing to assimilate the standard (or feeling uncertain about having assimilated it in a sufficient degree) sought assistance in doing so. Stylistic handbooks were to meet their needs. They were designed to enhance individual language behaviour, addressing adults wishing to write more effectively, i.e. conform to their readers’ expectations. ƀernyšev’s stylistic handbook is the only such effort to appear before the Revolution. Its author (1867-1949) has the background one would imagine for a linguistic advisor. Born into a family of former serfs, ƀernyšev taught in various places in various types of school before integrating into the academic world as a lexicographer. He made sketches of the Moscow and St Petersburg dialects and brought subtle differences to light. 28 His stylistic handbook appeared in 1911 and was very popular; two more editions appeared before the Revolution. ƀernyšev operates a consistent, principled view on linguistic norms: at the outset, he ranks his criteria for determining best usage. “ҭҝҽӊѹґҽ ҡҲѹґѹҡҲѹҍѦҡҔҸҮ ҊҝҽѸѸҽҲѹҔѹ ѹӉӊґѦҔҽӀҲҡҾ ӁҽѸѹ ѹӉ ҲҝѦѽ ҸҡӁҸӊӁӃѽ ѹҡҲҸҍӁѹҔҸӊ ґҏҍҗѦҮ ҝҏҡҡҔҸҮ ҝѦҍѹ: 28

Cf. ƀernyšev (1913)1970 and ƀernyšev 1970. As may be expected in view of the population history of the cities, he concludes that pronunciation and morphology are closer to the written language in St Petersburg than in Moscow.

188

W. ANDRIES VAN HELDEN

1. ҸқӂѦҰҝѹӁҾҲҸҊҸ ҡҸӊҝѦѸѦӁӁҸҊҸ ҏҰҸҲҝѦқґѦӁѹҾ; 2. ҰҝҸѹӉӊѦҦѦӁѹҮ ҸқҝҽӉ҃ҸӊӃѽ ҝҏҡҡҔѹѽ ҰѹҡҽҲѦґѦҮ; 3. ґҏҍҗѹѽ ҊҝҽѸѸҽҲѹҔ ѹ ҊҝҽѸѸҽҲѹҍѦҡҔѹѽ ѹҡҡґѦҦҸӊҽӁѹҮ ґѹҲѦҝҽҲҏҝӁҸҊҸ ҝҏҡҡҔҸҊҸ ҾӉӃҔҽ.” (ƀernyšev (19143)1970: 443) So the classics override the grammars. But generally accepted usage prevails over the classics: “Ѩ ҡҽѸѹ ҰѹҡҽҲѦґѹ ҾӊґҾӀҲҡҾ ҦґҾ Ӂҽҡ ҽӊҲҸҝѹҲѦҲӁӃѸѹ ӊ ҲѦѽ ҲҸґҧҔҸ ҡґҏҍҽҾѽ, ҊҦѦ ҸӁѹ ӁѦ ҰҸҊҝѦҗҽӀҲ ҰҝҸҲѹӊ ҸқӂѦҰҝѹӁҾҲӃѽ ӁҸҝѸ.” (ƀernyšev (19143)1970: 443) The order of the criteria testifies to the declining prestige of normative grammar and its ambassadors in education. The circle was round. Lomonosov had referred to what he considered to be acceptable usage. His successors adapted and repaired his constructions but rarely referred to usage themselves. The first generations of users of the standard language regarded grammars as a reference and only proposed amendments when they encountered problems. But the early 20th-century users of the standard language, as a self-conscious segment of society, regarded their own collective usage as the most important reference and relied on their internal social dynamics to sort out any differences. They could refer directly to their intuitions for determining the norms. Individuals with failing intuitions were outsiders, who should make an effort to assimilate. 8.6 ƀernyšev’s discovery of the accusative ƀernyšev’s stylebook contains a lengthy section on GENACCNEG, which starts with a discussion of the genitive rule in one of Buslaev’s versions, including long-distance government (cf. 6.2) and the exceptions made for some types of non-core negation (cf. 6.3, 6.4). Applying his system of criteria, according to which the masterworks of the classics override even the best grammars, ƀernyšev observes: “ѰҔҽӉҽӁѹѦ ҷҏҡґҽѦӊҽ, ҸҦӁҽҔҸ, ӁѦґҧӉҾ ҰҝѹӉӁҽҲҧ ҸҍѦӁҧ ҲӊѦҝҦӃѸ, ҲҽҔ ҔҽҔ ҏ ҽӊҲҸҝҸӊ ӁҽҮҦѦҲҡҾ ҸҍѦӁҧ қҸґҧҗҸѦ ҔҸґѹҍѦҡҲӊҸ ѹҡҔґӀҍѦӁѹҮ.” (ƀernyšev (19143) 1970: 594) Similarly, ƀernyšev adduces Puškin’s electric force to refute Buslaev’s position on the long-distance accusative (ƀernyšev (19143)1970: 594). 29 He lists a long series of instances of core negation containing direct object accusatives taken

29

Buslaev was actually rather liberal on this point, at least on the spot cited in 6.2.

VICISSITUDES OF THE GENITIVE RULE

189

from the classics. But, unlike Nekrasov and Tomson, he cannot derive a workable principle governing the use of the genitive and the accusative: “ғ ҸқӂѦѸ, ҰҸ-ӊѹҦѹѸҸѸҏ, ӊҸ ѸӁҸҊѹѽ ҡґҏҍҽҾѽ ӁѦґҧӉҾ ҲҸҍӁҸ ҸҰҝѦҦѦґѹҲҧ ҏҡґҸӊѹҮ, ҔҸҊҦҽ Ұҝѹ ҸҲҝѹ҃ҽӁѹѹ ҡҲҽӊѹҲҡҾ ҝҸҦѹҲѦґҧӁӃҮ ѹ ҔҸҊҦҽ ӊѹӁѹҲѦґҧӁӃҮ ҰҽҦѦң. ҋҏӂѦҡҲӊҏӀӂѦѦ ӊ җҔҸґҧӁӃѽ ҊҝҽѸѸҽҲѹҔҽѽ ҰҝҽӊѹґҸ ҡҲҽӊѹҲҧ ӊҡѦҊҦҽ Ұҝѹ ҸҲҝѹ҃ҽӁѹѹ ҝҸҦѹҲѦґҧӁӃҮ ҰҽҦѦң ӊѸѦҡҲҸ ӊѹӁѹҲѦґҧӁҸҊҸ ӁѦ ѹѸѦѦҲ ҰҸҦ ҡҸқҸҮ ҲӊѦҝҦҸҊҸ ҸҡӁҸӊҽӁѹҾ. ѷҽҔѹѸ ҸҡӁҸӊҽӁѹѦѸ ѸҸҊґҸ қӃ қӃҲҧ ҲҸґҧҔҸ ҡҸӉӁҽӁѹѦ ҍѹҡҲҸ ҊҸӊҸҝҾӂѹѽ ҰҸ-ҝҏҡҡҔѹ ѹґѹ, ҰҸ ҔҝҽҮӁѦҮ ѸѦҝѦ, ҍҽҡҲӁҸѦ ӊӃҝҽңѦӁѹѦ ҙҲҸҊҸ ҡҸӉӁҽӁѹҾ – ҰҸҡґѦҦҸӊҽҲѦґҧӁҸѦ ҏҰҸҲҝѦқґѦӁѹѦ ҦҽӁӁӃѽ ӍҸҝѸ ѽҸҲҾ ӁѦҔҸҲҸҝӃѸѹ ҽӊҲҸҝҽѸѹ. Ѣѹ ҲҸҊҸ Ӂѹ ҦҝҏҊҸҊҸ ҏ Ӂҽҡ ӁѦҲ, ѹ, ҡґѦҦҸӊҽҲѦґҧӁҸ, ӁҏңӁҸ ҰҝѹӉӁҽҲҧ, ҍҲҸ ҔҽңҦӃҮ ѹӉ Ӂҽҡ ӊ ӉҽҲҝҏҦӁѹҲѦґҧӁӃѽ ҡґҏҍҽҾѽ, қҸґѦѦ ѹґѹ ѸѦӁѦѦ ҡӊҸқҸҦӁҸ, ѸҸңѦҲ ҏҰҸҲҝѦқґҾҲҧ ӍҸҝѸӃ, ҔҸҲҸҝӃѦ ѦѸҏ қҸґѦѦ ҡӊҸҮҡҲӊѦӁӁӃ, ӁѦ ҝҏҔҸӊҸҦҡҲӊҏҾҡҧ ӁҽӊҾӉӃӊҽѦѸӃѸ ҾӉӃҔҏ ӁѦҸҡӁҸӊҽҲѦґҧӁӃѸ ҰҝҽӊѹґҸѸ.” (ƀernyšev (19143)1970: 594) So ƀernyšev is confronted with a loop in his system of criteria. The classic masterworks (criterion 2) are at odds with the rules of the grammarians (criterion 3) and should, therefore, override them. But the same classics, as well as generally accepted contemporary usage (criterion 1), fail to prompt alternative rules, leaving the rule-abiding language user to his own device. While sharing Tomson’s observation that the classics contain accusatives, ƀernyšev’s analysis is different in most respects: semantic and pragmatic distinctions hardly enter his deliberation. In a sense, ƀernyšev pursues Puškin’s approach to GENACCNEG. Whereas the latter extended Trediakovskij’s poetic licence to narrative prose, ƀernyšev makes it available to writers of all types of text. But, like Trediakovskij, he holds on to responsible use of the prerogative: as we saw in the quote, the user must choose the case ending he deems appropriate. It is doubtful whether ƀernyšev really triggered an advance of the accusative in the 20th century. Nearly all of his accusative examples from the classics are taken from dialogue. His own GENACCNEG usage suggests that he foresees no radical changes in other types of discourse: he rarely deems the use of the accusative appropriate himself. Other things being equal, everybody is expected to continue to write as one was taught at school. 8.7 Peškovskij as a Soviet Lomonosov For a well-established norm to change or disappear it requires initiatives to be taken, and favourable conditions. Just as in Lomonosov’s case, it requires the right person in the right place at the right time. A.M. Peškovskij (1878-1933) had the credentials the period required. In 1899, after his expulsion from Moscow University for having taken part in student

190

W. ANDRIES VAN HELDEN

unrest, he continued his physics studies in Berlin. Upon his return to Moscow he started to study philology, was again expelled and spent six months in jail. He was, like Tomson twenty years earlier, a student of Fortunatov. After obtaining his degree in 1906 with the highest marks, he did not stay at the university but started teaching Russian and Latin in various gymnasia (Belov 1958: 5). In 1914 he published Russian Syntax in the Light of Science, the title of which speaks for itself. The preface shows that Peškovskij intended it to restore the ties linking linguistics with education and society at large: “ҭҝѦҦґҽҊҽѦѸҽҾ ӊӁѹѸҽӁѹӀ ҍѹҲҽҲѦґҾ ҔӁѹҊҽ ӊҸӉӁѹҔґҽ ѹӉ ҰѦҦҽҊҸҊѹҍѦҡҔҸҮ ҦѦҾҲѦґҧӁҸҡҲѹ ҽӊҲҸҝҽ. ҭҝѦҰҸҦҽӊҽҾ ҝҏҡҡҔѹҮ ҾӉӃҔ ӊ ҸҦӁҸҮ ѹӉ ҍҽҡҲӁӃѽ ѸҸҡҔҸӊҡҔѹѽ ҊѹѸӁҽӉѹҮ, ҽӊҲҸҝ ӁҽҲҸґҔӁҏґҡҾ Ӂҽ Ҕґҽҡҡ, ӁѦ ҏҡҰѦӊҗѹҮ, ҰҸ ҡґҏҍҽҮӁӃѸ ҰҝѹҍѹӁҽѸ, ҰҝѹҸқҝѦҡҲѹ ӊ ҲѦҍѦӁѹѦ ҰѦҝӊӃѽ 3-ѽ ґѦҲ ҰҝѦқӃӊҽӁѹҾ ӊ җҔҸґѦ ҡҔҸґҧҔҸ-ӁѹқҏҦҧ ҰҝҸҍӁӃѽ ҡӊѦҦѦӁѹҮ ѹӉ ҡѹӁҲҽҔҡѹҡҽ. ҫқґҽҦҽҾ ӁѦҔҸҲҸҝӃѸ ӁҽҏҍӁӃѸ қҽҊҽңҸѸ ӊ ҙҲҸҮ ҸқґҽҡҲѹ, ҸӁ ӁѦ ҏҡҲҸҾґ ҰѦҝѦҦ ҡҸқґҽӉӁҸѸ ѹӉґҸңѹҲҧ ҰѦҝѦҦ ҡӊҸѦҮ ҡҰґҸҗҧ 14-Ҳѹ – 15-Ҳѹ-ґѦҲӁѦҮ ҽҏҦѹҲҸҝѹѦҮ ҸҡӁҸӊӁӃѦ ӊҸҰҝҸҡӃ ҡѹӁҲҽҔҡѹҡҽ ӊ ҲҸѸ ӊѹҦѦ, ӊ ҔҽҔҸѸ ҸӁѹ ҲҝҽҔҲҏӀҲҡҾ ҡҸӊҝѦѸѦӁӁӃѸ ҾӉӃҔҸӉӁҽӁѹѦѸ. ҫқҝҽӉҸӊҽӊҗѹѦҡҾ ӊ ҝѦӉҏґҧҲҽҲѦ җҔҸґҧӁӃѦ ӉҽҰѹҡҔѹ, ҝҽҡҗѹҝѦӁӁӃѦ ѹ ҦҸҰҸґӁѦӁӁӃѦ, ѹ ҰҸҾӊґҾӀҲҡҾ ҲѦҰѦҝҧ ҰѦҝѦҦ ҰҏқґѹҔҸҮ.” (Peškovskij 19202: i) Peškovskij’s syntax does not just comprise state-of-the-art viewpoints. Encountering numerous gaps in the existing literature on syntax, he fills them on his own accord, anticipating future developments in linguistic theory. His innovations include dependency analysis, intonation and word order analysis, as well as an effort to define the technical terms of linguistics in terms of their mutual relationships so as to obtain a logical deductive system. Yet the book was written in a non-technical language that was accessible for all. The linguistic community was sharply divided on the virtues of the book. In spite of accusations of inconsistency, creation of unnecessary terms, confusing the public instead of enlightening it, etc. (Belov 1958: 15ff.), he received the Academy Prize for his work on the basis of a review by Kudrjavskij (Belov 1958: 18). 30 Peškovskij’s book was hailed as the basis for the innovation of the Russian school curriculum at the All-Russian Teachers’ Congress in 1916-1917, which planned radical educational reform after the war (Šapiro (1956)2001: 5). The second edition appeared in 1920 and was one of the first textbooks printed in the new orthography. Peškovskij’s syntax started being used just when the new government started its offensive to eradicate illiteracy, the likbez campaign, 30

D.N. Kudrjavskij is known as a friend of Lenin, as the first linguist to use statistics in his analyses, and as the author of the textbook of linguistics that Stalin used when writing his article on Marxism and linguistics.

VICISSITUDES OF THE GENITIVE RULE

191

which required a considerable expansion of the teaching population in the 1920s: literacy rose from 56.6% in 1926 to 87.4% in 1939. Peškovskij’s grammar may be assumed to have had some leverage, as it was the only modern handbook that was available for teacher training in this crucial period. He also worked out his ideas in textbooks to be used in the lower classes (e.g. Peškovskij 1923). The French slavist Tesnière, who is often considered to be the first theoretician of dependency analysis, reports having brought home a collection of textbooks containing dependency trees from a mission to the Soviet Union in 1936 (Tesnière 19662: 15). Throughout the 20th century, young Russian linguists dealing with syntactic problems (e.g. Papina 1964) started out with Peškovskij’s dicta on the subject. 8.8 Peškovskij’s discovery of the accusative Like Lomonosov, Peškovskij is a syntacticist: his default way of dealing with case selection is government. While applying the method in a systematic manner, he does not hesitate to resort to semantic distinctions wherever case usage cannot be accounted for in terms of government. As Klobukov (1986: 15-16) points out, the seminality of Peškovskij’s modern version of government is evident in the mathematical approaches to case that were developed in the late 1950s and 1960s by the so-called Set-theoretical School. In the second edition of his book, Peškovskij states that some verbs may govern more than one object case, either simultaneously (dvojnoe upravlenie) or alternatively (dvojakoe upravlenie). In the last instance there is a semantic difference (Peškovskij 19202: 177). He presents the usual semantic distinctions between the accusative and the partitive genitive (cf. (11) and (12)) but regards GENACCNEG as a different matter altogether: “ҭѦҝӊҸӁҽҍҽґҧӁҸ ѹ ҲҏҲ ҝҽӉґѹҍҽґѹҡҧ Ҧӊҽ ӉӁҽҍѦӁѹҾ: (46) ѢѦ Ұѹҗҏ ҰѹҡҧѸҸ. ҏҔҽӉӃӊҽґҸ Ӂҽ ѹӉӊѦҡҲӁҸѦ ҰѹҡҧѸҸ, Ҹ ҔҸҲҸҝҸѸ ҝҽӁҧҗѦ ҊҸӊҸҝѹґҸҡҧ, ҽ (47) ѢѦ Ұѹҗҏ ҰѹҡҧѸҽԀ. Ӂҽ ҰѹҡҧѸҸ ӊҸҸқӂѦ, қѦӉ ҙҲҸҊҸ ҸҲҲѦӁҔҽ ѹӉӊѦҡҲӁҸҡҲѹ. ѢҸ ӊҰҸҡґѦҦҡҲӊѹѹ Ҹқҽ ҡҸҍѦҲҽӁѹҾ ҡѸѦҗѹӊҽґѹҡҧ, ѹ ҲѦҰѦҝҧ ӊѹӁѹҲѦґҧӁӃҮ ѹ ҝҸҦѹҲѦґҧӁӃҮ ҏҰҸҲҝѦқґҾӀҲҡҾ ӉҦѦҡҧ қѦӉҝҽӉґѹҍӁҸ. ҹҔҸґҧӁӃѸѹ ҊҝҽѸѸҽҲѹҔҽѸѹ ҏӉҽҔҸӁѦӁ, ҰҝҽӊҦҽ, ҲҸґҧҔҸ ҝҸҦѹҲѦґҧӁӃҮ, ӁҸ ӊ ӉӁҽҍѹҲѦґҧӁҸҮ ѸѦҝѦ ѹҡҔҏҡҡҲӊѦӁӁҸ.” (Peškovskij 19202: 177)

192

W. ANDRIES VAN HELDEN

The third edition of Peškovskij’s syntax 31 pays more attention to GENACCNEG and contains some original observations on the negated genitive of duration and distance. 32 But the bottom line remains: “[…] ӊ ӁҽҡҲҸҾӂѦѦ ңѦ ӊҝѦѸҾ ӊѹӁѹҲѦґҧӁӃҮ ӉҦѦҡҧ ԙӹӴӾӢԶӾӰӰӹ ӹԼՂԓӾӰ, ѹ ӊ ҏҰҸҲҝѦқґѦӁѹѹ ѦҊҸ ѸӃ ӁѦ ѸҸңѦѸ ҏґҸӊѹҲҧ ӁѹҔҽҔҸҮ ҝҽӉӁѹ҃Ӄ ѸѦңҦҏ ӁѹѸ ѹ ҝҸҦѹҲѦґҧӁӃѸ.” (Peškovskij (19283)20018: 297) Like the earlier exponents and opponents of the genitive rule, Peškovskij goes on to impose his analysis of the present state of Russian GENACCNEG on the earlier stages of the language: “ҤҸҲҾ ӊ ҰҽѸҾҲӁѹҔҽѽ ҏҰҸҲҝѦқґҾѦҲҡҾ Ұҝѹ ҸҲҝѹ҃ҽӁѹѹ ѹҡҔґӀҍѹҲѦґҧӁҸ ҝҸҦѹҲѦґҧӁӃҮ, ҸҦӁҽҔҸ ҡҸѸӁѦӊҽҲѦґҧӁҸ, ҍҲҸқӃ ңѹӊҸҮ ҾӉӃҔ ѸҸҊ ҰҝҸӊѦҡҲѹ ҔҸҊҦҽґѹқҸ ҲҽҔҸѦ ѹҡҔґӀҍѹҲѦґҧӁҸѦ ҏҰҸҲҝѦқґѦӁѹѦ: ӊѦҦҧ Ұҝѹ ҰҸҡҲҽӁҸӊҔѦ ҡҏӂѦҡҲӊѹҲѦґҧӁҸҊҸ ԗӾӢӾԌ ҊґҽҊҸґҸѸ ҊҸӊҸҝҾӂѹҮ ѸҸңѦҲ ӁѦ ҦѦҝңҽҲҧ ӊ ҏѸѦ Ұҝѹ ҰҝҸѹӉӁҸҗѦӁѹѹ ҡҏӂѦҡҲӊѹҲѦґҧӁҸҊҸ ҊґҽҊҸґҽ ҡ ҸҲҝѹ҃ҽӁѹѦѸ, ѹ ҲҸҊҦҽ ҰҽҦѦң ӁѦѹӉқѦңӁҸ қҏҦѦҲ ӊѹӁѹҲѦґҧӁӃҮ.” (Peškovskij (19283)20018: 297) 8.9 The impact of Tomson’s, ƀernyšev’s and Peškovskij’s discoveries As pointed out in 8.7, it is likely that Peškovskij exerted a considerable influence on the teaching of Russian in as far as his grammar dominated the textbooks and teacher training after the revolution. So his approach to GENACCNEG was in a better position than Tomson’s and ƀernyšev’s to make an impact on actual usage. But for an analysis to be accepted and applied, it should also be workable in educational practice. It is on this inherent quality that the three approaches presented above will be compared. Tomson, ƀernyšev and Peškovskij agree that the use of the accusative with direct objects of negated verbs is a legitimate option. But their positions diverge in their further analyses. Tomson believes in a guiding principle for GENACCNEG. As there are no indications that this principle is of a syntactic nature, it must be semantic. We shall see in Section 9 that Tomson goes a long way in defining it. ƀernyšev is a semantic agnostic. He admits that he cannot distinguish a guiding principle that rules GENACCNEG but he does not exclude that there is 31

Peškovskij’s syntax was almost entirely rewritten for the third edition (Šapiro 1956(2001): 5-6), which appeared in 1928 and was the basis for all subsequent, posthumous, editions. 32 He points out, for example, that, whereas pel pesni may transform into ne pel pesen’ and sidel den’ may transform into ne sidel dnja, the sentence pel pesni vsju noǁ’ can only transform into ne pel pesen’ vsju noǁ’, i.e. not into *ne pel pesni vsej noǁi or *ne pel pesen’ vsej noǁi (Peškovskij (19283)20018: 297).

VICISSITUDES OF THE GENITIVE RULE

193

one: he leaves it up to the individual language user to make the appropriate choices. Peškovskij, finally, is an unbeliever. While failing to discern any syntactic rule governing GENACCNEG, he rejects the existence of any semantic principle. But he can live with the fact that there is no guiding principle that determines case selection. Now what must be assessed here is not the relative validity of these viewpoints but their likely impact. Consider the options for a new teacher or a likbez educationalist, facing the immense task of eliminating hardcore illiteracy. As compared with Tomson’s subtle semantic analysis and ƀernyšev’s injunction to select a case as deemed appropriate by the educated, the genitive rule had the advantage of being simple and clear. Only one option is simpler than the genitive rule, viz. no rule at all. The only educational proposition that was stronger than the genitive rule is saying that there is no rule. Consequently, one effect is particularly likely to occur: the army of newly trained teachers and likbez educationalists for whom Peškovskij was a reference is not likely not have put a great deal of effort into teaching the genitive rule. So in the twenties and thirties, when Russia made its decisive leap toward universal literacy, the genitive rule was most probably not actively taught. If the use of the accusative increased in 20th-century writing, this circumstance may be part of the explanation. 8.10 Compliance in the 20th century? Although an enormous amount of research has been conducted on GENACCNEG in the second half of the 20th century, it is not simple to assess unambiguously what changed after the abolition of the genitive rule. The achievement of universal literacy, the increased participation of the population in linguistic communication in formalized settings, the appearance of new media, and the shifts in the stratification of society seem to have resulted in a pragmatic and sociolinguistic situation that is much more complex than in the preceding centuries. A thorough exploration of the 20th-century developments goes beyond the scope of the present paper but the following observations may illustrate the complexity of the problem. The default assumption about what happens when a normative item ceases to be taught in the schools is that written usage gravitates towards spoken usage. But how long does such a process take? And does it always occur, regardless of the sociolinguistic situation? It is only since the 1970s that the grammar of the informal spoken vernacular has been subject to sophisticated descriptive research. Zemskaja’s team, examining the informal spoken vernacular (razgovornaja reǁ’; RR) as an independent

194

W. ANDRIES VAN HELDEN

system, to be distinguished from the standard language (kodificirovannyj literaturnyj jazyk; KLJa: cf. Zemskaja 1973: 5-39), observes that the use of the genitive is much more limited in RR than in KLJa: sometimes the genitive is not even used in negated existential sentences, while direct objects of negated verbs may even be in the nominative (at least in specific communicative situations): (48) ԛԂԊӠԌӹԊ ӁѹҊҦѦ ӊ ҊҸҝҸҦѦ ӁѦҲ. (49) ԪӾԌӠԙӪԂ ӁѦ ҰҝѹӊҸӉѹґѹ. (Zemskaja 1973: 258) In her analysis of the speech used in the Russian version of Big Brother, Bocale (2004: 29-30) points to the overwhelming preponderance of the accusative for direct objects of negated verbs in informal speech, which she views as symptomatic of a linguistic long-term trend toward analyticity (cf. Comrie et al. 1996: 149). Whether this is a recent trend or an artefact of the introduction of more sophisticated methods of gathering data on the spoken language may be a matter of debate. But if, in accordance with the default assumption, such observed RR is to be considered the natural target for the development of written GENACCNEG usage, the future must have looked dim for the genitive after the disappearance of the rule that protected it. The prose of Soviet narrative writers does not show such an immediate and unequivocally progressive shift. Professionals of literature are likely to continue to be aware of the rule, if only by standing on the shoulders of their predecessors: they decide for themselves whether to apply or ignore the rule, or to play with it to achieve stylistic effects. According to Restan’s (1960: 105) counts, individual accusative scores of 20thcentury writers vary between 11% (Šoloxov; n = 38) and 48% (Paustovskij; n = 135) for narrative prose, and between 21% (Ƈrenburg; n = 68) and 67% (Šoloxov; n = 12) for dialogue. Paustovskij scores 48% for both narrative prose and dialogue (n = 135 and n = 29, respectively). Under the default assumption, the phasing out of the genitive should be more visible in non-artistic prose as its writers may be expected to continue to observe the rules they learned at school. The advance of the accusative should, then, keep pace with the replacement of the older generation, although even here the trend may be mitigated, since journalists, scholars and civil servants tend to learn the tricks of the trade from their elder colleagues on the work floor. Restan (1960: 108) observes a clear shift here: in sizeable samples taken from Izvestija, the use of the accusative turns out to have almost doubled between about 1920 and 1959: from 21.7% (n = 507) to 38.3% (n = 504). It is tempting to attribute this speedy change to the excessive turnover among writers of political content that occurred during the Stalin era, preventing neophytes from learning

VICISSITUDES OF THE GENITIVE RULE

195

the tricks of the trade. The scale of the shift is not, however, borne out by the 30% accusative score for a 1955 Pravda sample (Safarewiczowa 1960: 123; n = 132). Whatever trend occurred in written Russian seems to have slowed down in the second half of the 20th century. In relatively recent counts the overall accusative scores, while varying considerably depending on sample choice and the applied definitions, do not exceed 40% (cf. Mustajoki and Hein0 1991: 213; Ueda 1992: 9). The advance of the accusative seems to have come to a halt at about a quarter or a third of all occurrences of core negation. This is a long way from the situation reported for informal conversational Russian. Factors that can be adduced to explain the resistance of the genitive in text include the return of GENACCNEG in the prescriptivist literature in the second half of the 20th century. GENACCNEG is treated in the Academy grammars that appeared at relatively short intervals starting with Vinogradov and Istrina 1954, as well as in the stylebooks that started appearing in the 1950s. Rozental’’s reference books on stylistic matters, starting with Mamonov and Rozental’ 1957, were extremely popular and had numerous reprints. The 1971 edition of Rozental’’s (1967)19712 spravoǁnik for press workers alone was printed in 50,000 copies, the 1978 edition in 100,000 copies. The new guidelines do not resuscitate the genitive rule. They usually contain a miscellany of syntactic, semantic, lexical, morphological, pragmatic, stylistic and phraseological factors that are said to determine or codetermine case selection. The factors are distributed over three, sometimes five, classes: 1. 2. 3. 4. 5.

conditions (contexts or circumstances) in which the genitive is mandatory; conditions in which the genitive is favoured; conditions in which no particular case is favoured; conditions in which the accusative is favoured; conditions in which the accusative is mandatory.

The stylebooks thus provide a blend of Lomonosov’s rules (i.e. conditions 1 and 5), ƀernyšev’s recommendations (2 and 4), and Peškovskij’s laissez-faire (3). As to the substance of the new rules, a novel consideration that has turned up is disambiguation for readability: Rozental’ ((1967)19712: 250), for example, recommends avoiding such forms as gazety when marking the direct object of a negated verb because of the syncretism of the genitive singular and the accusative plural. Using gazety in a context that admits both the genitive and the accusative would blur the number distinction, i.e. semantic information with obvi-

196

W. ANDRIES VAN HELDEN

ous referential implications: instead, the accusative (gazetu) is recommended for the singular and the genitive (gazet) for the plural. 33 The impact of this new, complex, prescriptive approach on actual usage in the second half of the 20th century has not been assessed. It cannot be ruled out that the present fuzzy picture of GENACCNEG is due to the simultaneous occurrence of two developments: increasing visibility of the spoken conversational language, in which the accusative is dominant, and expansion of the spheres in which the written standard, where the genitive is dominant, is functionally used. On the surface, the 20th-century developments produce the impression that the accusative has progressed. But in some areas of the language the genitive rule has left its mark. The genitive is nowadays strongly preferred for direct objects of negated gerunds, where the accusative was not uncommon the 18th century (Meintema 1988: 412-413; cf. Mustajoki and Heino 1991: 59-60). The verb imet’ passes for the iron bastion of the negated object genitive in analyses of modern Russian (e.g. Guiraud-Weber 2003: 367; cf. Mustajoki and Heino 1991: 28-29 for numerous references), whereas it occurred not infrequently with the accusative in Kantemir’s dispatches: (50) Ґ. ҔҽҝҦѹӁҽґ ҏӊѦҝѹґ ѸѦӁҾ, ҍҲҸ ҔҸҝҸґҧ ԌӢӫԄӫԇ ԗӢӠԓӠӰӫ ӁѦ ѹѸѦѦҲ. (Meintema 1988: 408) Note also the accusative of the direct object dar of negated imet’ in Lomonosov’s verse (5) quoted in 3.4. It is possible to qualify such minor movements running counter to the postulated trend as (lexical, phraseological, stylistic) noise obscuring the strong underlying main current toward the accusative. But the question remains how much noise a trend can stand. How many centuries are needed for a language change to just give up? If the accusative was already there in the 18th century, there may not even have been much of a trend toward the accusative since then. If the imposed rule secured the genitive a few centuries of prestige in a largely illiterate population, can it be ruled out that an educated nation is giving the case another lease of life? The 20th-century position of the genitive of negation might, then, even be accounted for as resulting from a rather successful intervention to impose its use, rather than as the direct continuation of an ailing relic from prehistoric times it is often assumed to be.

33 Berneker signalled the dependence of GENACCNEG choice on morphological characteristics of nouns in 1904: 386, long before disambiguation was a legitimate normative argument. Rendering the discussion on this mysterious issue is a complex exercise, which will not be undertaken here.

VICISSITUDES OF THE GENITIVE RULE

197

9. Why was the genitive rule tolerated? In the preceding sections it was shown that it is possible to provide a historical scenario for GENACCNEG usage and reflection on it over the last centuries without having to resort to the assumption of gradual and progressive linguistic change. The key piece of evidence supporting its plausibility is Meintema’s analysis of GENACCNEG case choice in Kantemir’s dispatches (cf. 3.6), which leads her to the conclusion that: “Kantemir’s choice was indicated by real usage, and not by artificial rules.” (Meintema 1988: 419) As discussed in Section 4, the genitive rule was not in every sense artificial. It is likely to have been a logical choice given the state of the language and the reflection on it when the language was codified, being dictated by practical considerations as well as by the syntacticist approach to linguistic analysis that dominated reflection on language, while perhaps being supported by the crucial codifier’s own dialectal usage. This leaves one objection unmet. Even if the codifiers had valid, or at least understandable, reasons to impose the genitive rule, why did the generations of language users who succeeded them sustain it when the reasons were no longer valid or understandable? Why did everyone put up with the genitive rule? The standard answer, in various guises, is path dependence: sub-efficient conventions are perpetuated as long as the short-term costs involved in the required change are perceived to outweigh the expected long-term cost reduction resulting from it. 34 It is not feasible to quantify the costs and benefits of the genitive rule; but some qualitative assessments can be made. As was shown in 8.9, when discussing the impact of Peškovskij’s analysis of the genitive rule, there was a point when the costs of change were minimal while the benefits seemed enormous and immediate. But the costs of change had always been minimal. The genitive rule had no orthographic implications: it is simple to discard a syntactic rule. Several rules in Lomonosov’s grammar disappeared in the course of time without anyone taking notice. The costs of maintaining the genitive rule, on the other hand, surfaced in the early 19th century. Even if Puškin c0uld not explain why he needed the accusa34 The classic example of path dependence is the QWERTY keyboard layout, which is supposed to have been scientifically designed to slow down the typing so as to keep the hammers of a mechanical typewriter from getting entangled (Beeching 1974: 39-40). Later research has questioned the historical correctness of the example.

198

W. ANDRIES VAN HELDEN

tive, it is clear that he regretted the ban on it, many artistic writers implicitly sharing his regrets by ignoring the rule. The costs were also manifest in the increase of the rule’s complexity when the 19th-century grammarians tried to accommodate instances for which it produced blatantly counter-intuitive results. Tomson states that the accusative and the genitive are not always interchangeable: “ѩҽґѦѦ ҸҔҽӉӃӊҽѦҲҡҾ, ҍҲҸ Ұҝѹ ҲѦѽ ҏҡґҸӊѹҾѽ, Ұҝѹ ҔҸҲҸҝӃѽ ҏҰҸҲҝѦқґҾӀҲҡҾ ҸҲҝѹ҃ҽҲѦґҧӁӃѦ ҰҝѦҦґҸңѦӁѹҾ ҡ ӊѹӁѹҲѦґҧӁӃѸ ҰҽҦѦңҸѸ, ҦҽґѦҔҸ ӁѦ ӊҡѦҊҦҽ ӊҸӉѸҸңӁҽ, қѦӉ ѹӉѸѦӁѦӁѹҾ ҡѸӃҡґҽ, ӉҽѸѦӁҽ ѹѽ ҰҸҡҝѦҦҡҲӊҸѸ ҰҝѦҦґҸңѦӁѹҮ ҡ ҝҸҦѹҲѦґҧӁӃѸ ҰҽҦѦңҸѸ ҸқӆѦҔҲҽ.” (Tomson (1902)1903: 207) If that observation holds good, the ban on the accusative implies that certain types of message that the language enables its speakers to express could not be made available to its writers. Tomson believes that such messages exist: “Ein Genitiv statt dieses Akkusativs würde oft gekünstelt erscheinen oder sogar den Sinn entstellen.” (Tomson 1912: 253) So Puškin and other subtle language users were right in feeling uneasy with the rule. The rule also put a load on ordinary writers. Tomson suggests that avoiding the accusative in writing may require a certain effort. “Unter dem Einfluß der Grammatik […] suchen allerdings Gebildete, besonders wenn sie sich mit Bewußtheit zu der Sache verhalten, wie Schreibende oder Lehrer beim Korrigieren, den Akkusativ zu vermeiden.” (Tomson 1912: 253) But although the costs of the genitive rule were real, Tomson also indicates that they were not always experienced as being considerable. Speakers using the accusative in speech switched to the genitive when writing without noticing: “Ѱ ӊҡѦѽ ҰѹҡҽҲѦґѦҮ, ҏ ҸҦӁѹѽ ӊ қҸґҧҗѦҮ, ҏ ҦҝҏҊѹѽ ӊ ѸѦӁҧҗѦҮ ҡҲѦҰѦӁѹ, ӊҡҲҝѦҍҽӀҲҡҾ ҰҝѹѸѦҝӃ ҡ ӊѹӁѹҲѦґҧӁӃѸ ҰҽҦѦңҸѸ. ѢҸ ҊҸҝҽӉҦҸ ҍҽӂѦ ҏҰҸҲҝѦқґҾѦҲҡҾ ҙҲҽ ҔҸӁҡҲҝҏҔ҃ѹҾ ӊ ӁҽҝҸҦӁҸѸ ҾӉӃҔѦ ѹ ӊ ҸқӃҦѦӁӁҸҮ ҝѦҍѹ ҸқҝҽӉҸӊҽӁӁҸҊҸ ҝҏҡҡҔҸҊҸ ҸқӂѦҡҲӊҽ, ҔҝҸѸѦ ҝҽӉӊѦ ҸҲҦѦґҧӁӃѽ ґӀҦѦҮ, ҏ ҔҸҲҸҝӃѽ ӊґѹҾӁѹѦ җҔҸґӃ ҏӁѹҍҲҸңѹґҸ ҍҏҲҧѦ Ҕ ҰҸӁѹѸҽӁѹӀ ҝҽӉґѹҍѦӁѹҾ ӊ ӉӁҽҍѦӁѹҾѽ ҙҲѹѽ Ҧӊҏѽ ҔҸӁҡҲҝҏҔ҃ѹҮ ѹ ҰҝѹҏҍѹґҸ ѹѽ Ҕ ѹҡҔґӀҍѹҲѦґҧӁҸѸҏ ҏҰҸҲҝѦқґѦӁѹӀ ҝҸҦѹҲѦґҧӁҸҊҸ ҰҽҦѦңҽ. ѷѦ ңѦ ҰҝѹҝҸҦӁӃѦ ҝҏҡҡҔѹѦ, ҔҸҲҸҝӃѦ ҊҸӊҸҝҾҲ: […] (51) ѨӊҽӁ ӁѦ ҰҝѹӁѦҡ ԽӴӹӡ ԙӫӰԌӫӪ. ҔҸҊҦҽ ѨӊҽӁ ҦҸґңѦӁ қӃґ ҰҝѹӁѦҡҲѹ ѹӉӊѦҡҲӁӃҮ ҊҸӊҸҝҾӂѹѸ ҡҏӁҦҏҔ […], ҲѦ ңѦ ґѹ҃ҽ ӉҽѸѦӁҾӀҲ ӊ ҰѹҡҧѸѦ ҙҲҸҲ ӊѹӁѹҲѦґҧӁӃҮ ҰҽҦѦң ҸқӃҔӁҸӊѦӁӁҸ ҰҸҡҝѦҦҡҲ-

VICISSITUDES OF THE GENITIVE RULE

199

ӊҸѸ ҝҸҦѹҲѦґҧӁҸҊҸ ҰҽҦѦңҽ, ѹ ҦҽңѦ ӊҸҡҡҲҽӀҲ ӊ ҲѦҸҝѹѹ ҰҝҸҲѹӊ ҡҏӂѦҡҲӊҸӊҽӁѹҾ ҲҽҔҸҮ ҔҸӁҡҲҝҏҔ҃ѹѹ.” (Tomson (1902)1903: 207) This would imply that the costs of the rule are minimal, at least for habitual writers. By adducing personalities whose sensitivity to the distinction has been completely distorted by education, Tomson seems to ascribe the public’s compliant behaviour to acquired habit, to internalization of the rules of normative grammar. But Tomson’s own analysis of the difference between accusative messages and genitive messages allows for a different, or at least additional, reason why writers (and public speakers) of all sorts tolerated a ban on expressing whatever message the accusative was supposed to express. Speech and writing are different media. The possible messages that are conveyed in writing do not necessarily coincide with the possible messages that are conveyed in speech. It is conceivable that the accusative is needed to convey messages that occur less often in writing, and vice versa. In that case, rank-andfile writers tolerated the genitive rule because they felt that the use of the genitive was more appropriate anyway, or even assisted them to write more efficiently, i.e. make it easier to read them. This would provide an intrinsic explanation for the tolerance of the genitive rule until its abolition and the limited decline of the genitive afterward. The plausibility of this idea, which requires a plausible semantic analysis of GENACCNEG and a plausible analysis of the pragmatics of speech and writing, as well as a match between the two analyses, will be explored below. Tomson’s semantic analysis of GENACCNEG will be adopted as a point of departure. Note that, for the purpose of the present argument, it is not necessary to demonstrate the validity of a specific semantic analysis of GENACCNEG. It is sufficient to identify a class of plausible semantic analyses that entail different pragmatic implications for spoken and written discourse. Tomson’s analysis will be confronted with some pragmatic characteristics of speech and writing, which are loosely based on Ehlich’s (1982) work on deixis. 9.1 Tomson’s GENACCNEG semantics Tomson’s semantic analysis dates from a period when the literacy rate was still below 50%: at the time, the spoken usage is likely to have diverged more from the written standard than at present. Other available semantic analyses reflect more recent situations, when the standard language and the vernacular may have become more entangled as a result of universal literacy and other 20thcentury developments indicated in 8.10.

200

W. ANDRIES VAN HELDEN

As pointed out in 8.1, Tomson was a subtle observer of spoken usage; his analysis is borne out by a large number of examples. Moreover, as will be shown below, his approach to negation anticipates recent views on the semantics of negation. Finally, Tomson’s views on the semantics of GENACCNEG are not idiosyncratic: later analysts of the semantics of GENACCNEG arrive at similar results. Tomson sets the exceptions discussed in 6.2 and 6.3 aside, regarding them as trivial, and focuses on core negation. He characterizes the semantic distinction corresponding to the use the genitive or the accusative as follows. “Ein negiertes Verb beim Genetiv bezeichnet eine abstrakte Handlung und weist nur auf die Umstände hin, die aus der Abwesenheit der Handlung gefolgert werden – auf einen negativ bestimmten Zustand. Das Objekt weist nur auf die Richtung (die Sphäre) der fiktiven Handlung und kam daher in den Genetiv zu stehen. Z.B. mit Russischen (52) Ѥ ӁѦ ӊӃӁҏґ Ҧҏҗѹ. ‘ich habe die Seele nicht herausgenommen’ bezeichnet man, daß die Seele im Körper geblieben ist, oder daß dem Täter nicht solche Eigenschaften oder Umstände zuzuschreiben sind, die als Motiv oder Folge dieser Handlung ihm zuzuschreiben wären. 35 Wenn aber der Redende die Handlung selbst als nichtexistierend zurückweist, die also dann in der Seele des Hörers in dem entsprechenden positiven Urteil schon erregt ist oder als erwartet, erwünscht usw. vom Redenden vorausgesetzt werden kann, so wird im heutigen Russisch, und wurde wohl stets die Konstruktion mit dem Akkusativ, nicht dem Genetiv, gebraucht, weil nur so eine vollkommene Widerlegung des positiven Urteils in demselben Verhältnis der Handlung zum Objekt möglich war. Daher heißt [es]: (53) ѢѦ ӊӃӁҏґ Ҿ Ҧҏҗҏ. ‘Ich habe nicht die Seele herausgenommen (die ich herausnehmen sollte)’ […].” (Tomson 1912: 252-253) Tomson’s analysis is essentially included in Jakobson’s case theory ((1936)1971: 41) and converges with several analyses that were made in the 1960s and 1970s following Morison’s observation that:

35 Examples (52) and (53) are inspired by a passage in L.N. Tolstoj’s story ƀem ljudi živy, in which an angel is sent out to take the soul of a woman but returns empty-handed. In the exchange with the woman that makes him change his mind, negated synonyms of brat’ alternately take the accusative and the genitive of duša, depending on the perspective of the message. As shown by Safarewiczowa’s count cited in 5.2, Tolstoj was a heavy accusative user. Tomson ((1902)1903: 233) characterized him as an author who was deeply imbued with the spirit of the Russian language.

VICISSITUDES OF THE GENITIVE RULE

201

“The [accusative] is used if the sentence expresses what was (is, etc.) not done, whereas the [genitive] appears if the idea emphasized is TO what (whom) was (etc.) it not done.” (Morison 1959: 23) Space does not permit discussing, or even mentioning, all the available variants here but many of them tally with Meintema’s observation that: “negations with the accusative often refer to an event which does not take place; the genitive, on the other hand, often refers to a situation which does not exist.” (Meintema 1986: 391) Tomson’s situational semantics imply a breach with the classic approach to negation. In the latter, “logical”, approach, the analyses of a negative sentence and its corresponding affirmative sentence tend to coincide except for the addition of the negation: the negation of a meaning X is identical to X plus the assertion that X is rejected or is not a fact. X itself is left intact, while the mode of its nonexistence is not specified: the non-existence of a situation is treated as a universal, absolute fact. In practice, however, negative messages are rarely universal. They are intended to mark the absence of X in a specific situation (context, frame of reference). From a pragmatic viewpoint, a negated state of affairs cannot be communicated if it is not bound by another state of affairs. In some communication-oriented analyses, therefore, negation is treated as the simultaneous evocation of two specific states of affairs: (A) the state of affairs that is denied (rejected, negated, etc.); (B) the state of affairs that binds the negation of (A), i.e. is characterized by the absence of (A). Such bisituational approaches to negation have been proposed, for example, by Ebeling and Fauconnier. In Ebeling’s semantics, a sentence purports to evoke two projections in the hearer’s frame of reference, a narrated event and a narrated period, which may, but do not necessarily, coincide (Ebeling 1978: 239). Negation is the instruction to the hearer to project the narrated event without assigning it to the narrated period that is simultaneously evoked (cf. Ebeling 1978: 242-243). Similarly, Fauconnier views negation as an instruction to construct a second, counterfactual, mental space, in which the denied state of affairs exists and which serves as a reference in the current, parent, mental space (Fauconnier 1984: 124-125). Tomson did not explicitly adopt a communication-oriented approach to negation, phrasing his analysis in psychological terms. But the bisituational nature of linguistic negation is discernable in his approach to GENACCNEG.

202

W. ANDRIES VAN HELDEN

The two states of affairs (projections, mental spaces) involved in negation are linked by the participation of the negation of the denied state of affairs (A) in the binding state of affairs (B). Other items may participate in either or both states of affairs: usually, most participants of (A) and (B) coincide. It is up to the language either to provide the means to determine explicitly which item participates in which state of affairs, or to remain ambiguous on this point, leaving it up to the hearer to infer which item belongs where. Tomson appears to imply that Russian offers the choice to a speaker to indicate by means of case selection in which of the two evoked states of affairs the object is presented as participating: 1. the accusative marks the direct object of the verb in the denied (rejected, etc.) state of affairs (A); 2. the genitive presents the object as a separate participant in the state of affairs (B) that binds (A). This does not mean that the use of the accusative excludes that the object participates in (B) as well: it is left up to the hearer to make that inference. Similarly, the hearer may infer that the object that is in the genitive is also the direct object in (A). As a consequence of this division of roles between the cases as imposed by the language, the denied state of affairs (A) is not fully specified if the genitive is used: it lacks a direct object. This entails that, if the genitive is used, (A) stands for a class of events being absent rather than for a single specific event. The genitive specifies the object in the binding state of affairs (B) that is affected by the absence of a class of events {A1, …, An} of which it could be the object. Ueda exposed this implication: “[… Genitive] clauses invoke a type or set of entities X and report the existence of non-existence of any of its members (x1…n), without distinguishing any particular member or members from others.” (Ueda 1992: 213) As a consequence, the use of the genitive is often associated with abstraction and abstract discourse (e.g. Fleckenstein 1961: 216). Tomson envisages this: “Bei der Akkusativkonstruktion hat das Objekt stets bestimmte konkrete Bedeutung und die Handlung wird mehr oder weniger konkret aufgefaßt. Beim Genetiv wird die fiktive Handlung abstrakt vorgestellt.” (Tomson 1912: 253) Tomson’s distinction in fact combines two earlier approaches to GENACCNEG. Lomonosov, who had adopted Latin and logic as his beacons, treated GENACCNEG from the viewpoint of the denied state of affairs (A). The negative particle leaves the internal structure of the sentence intact: it just adds the negation to it, turning X into non-(X). Consequently, Lomonosov’s had to produce

VICISSITUDES OF THE GENITIVE RULE

203

his genitives by means of a mechanical transformation rule stipulating that the deep accusative surfaces as a genitive. Potebnja, on the other hand, interpreted the relationships between the components of the sentence in terms of the binding state of affairs (B). He regarded the predication underlying the negated sentence as fundamentally different from that of any affirmative sentence, its basic structure consisting of a proposition the subject of which is the absence of a class of events, the predicate being the object marked with the genitive. Tomson more or less agrees with Potebnja’s analysis for instances in which the genitive is used (Tomson 1908: 299), while reserving the accusative for the instances to which Lomonosov’s analysis applies. This of course rendered Lomonosov’s transformation, i.e. the genitive rule, redundant. 9.2 GENACCNEG as a tool for deixis In Tomson’s analysis, the semantic distinction that corresponds to GENACCNEG does not necessarily have referential implications. In many instances both options are available for making reference to what is one and the same situation in reality: the use of the genitive to invoke the absence of a class of events in a particular situation does not preclude the inference on the part of the hearer that the specific event that is prominent in his frame of reference is absent in that situation: a concrete or specific inference may complement an abstract or general meaning. Rather, the semantic difference is available to the speaker for carrying out different communicative strategies. A single state of affairs can be referred to by using either the genitive or the accusative depending on the aspect of the situation that the speaker wishes the hearer to focus on. This is Tomson’s (1912: 253) seelische Stimmung: the psychological term is not too felicitous as it suggests that case choice is determined by the speaker’s mental disposition, while it is rather determined by the change in the hearer’s mental disposition that the speaker wishes to bring about. The semantics of Tomson’s GENACCNEG distinction are, therefore, comparable with those of deictic categories, such as the articles in West European languages. By means of deixis the speaker instructs the hearer to focus on a specific aspect of the situation referred to in the setting of the utterance. Consider the sentences: (54) I see a man. (55) I see the man. Sentences (54) and (55) may refer to an identical state of affairs; but by choosing the definite or indefinite article the speaker instructs the hearer to relate it in different ways to the situation in which the sentence is uttered, or rather to the

204

W. ANDRIES VAN HELDEN

way in which he assumes the hearer perceives the situation. It is possibly no coincidence that several studies of GENACCNEG view the case distinction as a mapping of the article distinction in other languages (e.g. Birkenmaier 1979: 120143). In fact Tomson himself hints at the similarity as well (cf. Tomson (1902)1903: 223-228). The interpretation of deictic categories depends on the situation in which they are used. The interpretation of the non-deictic word house does not vary much between different settings: the properties characterizing houses are largely independent of the situation in which the word is uttered. The interpretation of this in this house, on the other hand, is determined by the situation in which the words are uttered. 9.3 Deictic elements in speech and text The interpretational potential of many deictic elements varies particularly strongly between speech and text. In dialogue, the speaker and the hearer share the same setting and are largely aware of each other’s situation and mental disposition. The speaker communicates successfully because he can make plausible assumptions on the state of the frame of reference of the hearer, enabling him efficiently to modify it by applying deictic categories. The writer of a text, however, does not generally share a specific situation with his readers and can only make rough assumptions on the state of a reader’s frame of reference. In dialogue, a speaker chooses such deictic words as this and that on the basis of his assessment of the situation that he shares with the hearer and of his assumptions about the latter’s occasional and permanent characteristics. Deictic words may refer to items that have just been brought up or are about to be brought up in the exchange, or to items physically present in the shared situation, or just to items that the speaker assumes to be prominent in the hearer’s frame of reference on the basis of his knowledge about the hearer. But the situational landmarks for deictic categories that are present in live conversation are largely absent in text. A writer usually does not know the situation in which his text is read. He sometimes does not know the reader, and he may intend the text to be read by many readers in different situations. The semantics of deictic elements is often of little use to him. Their use is then avoided or conventionalized. The use of the demonstrative pronoun, for example, is usually limited to anaphora, i.e. to refer to items that have been invoked in the part of the text that a reader is assumed to have read just before reading the pronoun. Uses of this requiring a non-anaphoric interpretation tend to be avoided: special lexical items may even be resorted to for reference to the text itself, although that is the only thing the writer knows to be present in the read-

VICISSITUDES OF THE GENITIVE RULE

205

ing situation and, therefore, to be identifiable for the reader: nastojašǁaja stat’ja; the present paper. 9.4 GENACCNEG semantics and the pragmatics of speech and writing The pragmatic distinction that was introduced above may match Tomson’s bisituational GENACCNEG distinction. In his semantic analysis, the semantic difference between the genitive and the accusative is typically relevant in interactive speech and much less so in text. We can illustrate this with his example (51) given above. In dialogue it is convenient for a speaker to have the option to choose between: 1. the message that an expected event, viz. Ivan’s delivery of the trunk, has been cancelled, and 2. the message that a situation has occurred that is characterized by Ivan’s failure to deliver the trunk. It is this difference that can conveniently be expressed by Tomson’s GENACCthe accusative projects the trunk in the negated state of affairs (A); the genitive assigns a role to it in the description of the binding state of affairs (B): “as to the trunk, Ivan did not bring it”. But when the speaker sits down to write a note about the incident to be read afterward by his wife, the distinction is hardly relevant. Most rank-and file writers will, in that case, opt for the second version of the reported event, i.e. the one in which the absence of the trunk is focused on, as this is the more relevant message for the person reading the note later: there will usually be little interest, by then, in reporting from the viewpoint the aborted event of delivery itself. So when Tomson notes, implicitly observing himself at his desk, that:

NEG:

“[…] durch den Prozeß des langsamen Niederschreibens die seelische Disposition bei solchen feinen Bedeutungsdifferenzen leicht verschoben wird” (Tomson 1912: 253), he implies that writers do not always have to make an effort to avoid the accusative. The accusatives were not necessarily weeded out, at least not all of them: the word “leicht” suggests that the shift of perspective occurred naturally when switching from speech to writing. The switch to the genitive is in many cases the practical consequence of the pragmatic switch to a mode of communication that lacks the deictic possibilities in which the use of the accusative is functional.

206

W. ANDRIES VAN HELDEN

9.5 The genitive is serious In the preceding sections it was shown that it is possible to devise a plausible semantic theory for GENACCNEG that may account for divergent GENACCNEG usage between speech and writing because of the pragmatic differences between these media: the meaning of the accusative is expressed less often in writing because the messages for the expression of which the accusative is required are particularly relevant in interactive communication in shared situations, which is impossible in written communication anyway. This may in turn provide a plausible explanation for the strong position that the genitive rule has had for so long in the standard language. Consider Peškovskij’s sociolinguistic law on the use of a standard language: “ѯѦѸ ґѹҲѦҝҽҲҏҝӁѦѦ ҝѦҍҧ, ҲѦѸ ѸѦӁҧҗҏӀ ҝҸґҧ ѹҊҝҽѦҲ ӊ ӁѦҮ ҸқӂҽҾ ҸқҡҲҽӁҸӊҔҽ ѹ ҸқӂѹҮ ҰҝѦҦӃҦҏӂѹҮ ҸҰӃҲ ҊҸӊҸҝҾӂѹѽ.” (Peškovskij (1924)1959: 58) The national standard language was, in the first place, intended to abstract as far as possible from shared situations and experience in a physical sense (replacing them by the virtual shared experience of the school curriculum): it started out as a public language, to be used in writing, science and technology, government, etc. If the distinction between the genitive and the accusative is typically useful in shared situations, the ban on the distinction in writing is not entirely artificial or semantically arbitrary. Reference to elements of a shared situation if the situation is not shared may be dysfunctional. If Tomson’s analysis is not entirely incorrect, it is conceivable that the ban on the accusative was in fact supported by implicit semantic considerations: reporting about reality in terms of situations evokes in the reader a sense of serenity, stability, thoughtfulness and objectivity, producing the impression that the writer is reliable and serious and his message is credible and relevant; whereas by writing about reality in terms of cancelled expectations, intentions, etc., the writer impresses the reader as trying to involve him in the reported event, evoking a sense of immediacy, unrest and emotion and possibly producing the impression that the writer is self-centred or immature. Teachers aspiring to educate rational, knowledgeable, responsible adults will be inclined to dissuade their pupils from presenting themselves as self-centred and immature and encourage them to radiate a serious and reliable, adult-like attitude by avoiding semantic elements that convey emotion and involvement to the reader and detract from the informative impact of the message: such teachers will naturally recommend communicating “objective” messages in terms of situations instead of cancelled events. It is, therefore, possible that the genitive rule, or rather the stigmatization of the accusative, was not as artificial and arbitrary as it has always been depicted: it may have been an indirect consequence of the enlightened attitude that char-

VICISSITUDES OF THE GENITIVE RULE

207

acterized education since the start of the nation building process. The future elite and, later on, future citizens were not so much taught to conform to an arbitrary rule but to conform to the norms of reason and reliability, to present thoughtful conclusions rather than subjective experiences. They may have internalized the semantic implications of the rule. All this applies to a lesser degree to artistic writers, who must command a much broader range of ways to present events, situations and experiences than other writing individuals. Narration is about events: narrators are interested in presenting events inclusive of the way in which they are experienced. Long before the frequency counts were conducted, Tomson ((1902)1903: 207) noted that no writer of narrative prose denied himself the convenience of using the accusative altogether. As he put it, “Beim Akkusativ ist der Sinn sozusagen narrativ, beim Genitiv deskriptiv.” (Tomson 1912: 253) This aspect has produced some confusion in GENACCNEG research. While it was easily acknowledged that dialogue could be expected to show deviant GENACCNEG usage, as it reflected speech, it was not realized that the deictics of narrative text also stimulated the use of the accusative. As demonstrated by ƀernyšev’s stylebook, this eventually produced a certain misunderstanding when the classics’ usage was presented as a model to be emulated in all text. The plausibility of the explanation that has been suggested in this paper can, in principle, be assessed. If the use of the genitive fulfils a pragmatic and communicative need, it will be continue to be used even if there is no genitive rule. If the rule has been semantically internalized, i.e. if the seriousness that it evokes continues to be felt as an important value by the population, it can only disappear as a manifestation of a cultural shift, leading to a society in which it is less important to present oneself as a representative of the values conveyed by the use of the negated direct object genitive. We can, to a certain extent, assess the validity of the explanation by exploring what happened to GENACCNEG in the 20th and early 21st centuries. Leiden University

208

W. ANDRIES VAN HELDEN REFERENCES 36

Andreev, Nikolaj Dmitrieviǁ 1963 “Ob odnom Ljksperimente v oblasti russkoj orfoLjpii”. Voprosy kul’tury reǁi 4, 49-52. Babby, Leonard H. 1986 “The locus of case assignment and the direction of percolation: Case theory and Russian”. In: Richard D. Brecht, James S. Levine (eds.), Case in Slavic, 170-219. Ohio: Slavica Publishers. Bailyn, John F. 1997 “Genitive of negation is obligatory”. Formal Approaches to Slavic Linguistics 4: The Cornell Meeting 1995, 84-114. Barsov, Anton Alekseeviǁ (1788)1980 “Barsov’s comprehensive grammar of Russian”. In: Lawrence W. Newman (ed.), The Comprehensive Grammar of A.A. Barsov / Obstojatel’naja grammatika A.A. Barsova (Moscow and Leningrad Manuscripts 1783-1788), 1-374. Ohio: Slavica Publishers. (1788)1981 “Moskovskij spisok”. In: M.P. Tobolova, B.A. Uspenskij (eds.), Rossijskaja grammatika Antona Alekseeviǁa Barsova, 35-240. Moskva: Izdatel’stvo Moskovskogo universiteta. Barsov, A.V. 1880 “Znaǁenie i zavisimost’ padežej i ix sootnošenija meždu soboj”. Filologiǁeskie zapiski 19/3, 1-16. Baudouin de Courtenay (BoduLjn de KurtenLj), Jan Niecis’aw (Ivan Aleksandroviǁ) (1901)1963 “O smešannom xaraktere vsex jazykov”. In: Izbrannye trudy po obšǁemu jazykoznaniju 1, 362-372. Moskva: Izdatel’stvo Akademii nauk SSSR. Beeching, Wilfred A. 1974 Century of the Typewriter. London: Heinemann. Belov, Aleksandr Iosifoviǁ 1958 A.M. Peškovskij kak lingvist i metodist. Moskva: Uǁpedgiz. Benigni, Valentina 2002 “L’evoluzione della grammatica”. In: Claudia Lasorsa Siedina, Valentina Benigni, Il russo in movimento: Un’indagine sociolinguistica (Linguistica applicata e glottodidattica (nuova serie) 41), 43-105. Roma: Bulzoni. Berneker, Erich 1904 “Der genetiv-accusativ bei belebten wesen im Slavischen”. Zeitschrift für vergleichende Sprachforschung 37, 364-386. Birkenmaier, Willy 1979 Artikelfunktionen einer artikellosen Sprache: Studien zur nominale Determination im Russischen (Forum slavicum 34). München: Wilhelm Fink. Bocale, Paola 2004 “Il russo in vetrina: indagini sulla lingua dei partecipanti a ‘Za steklom’”. eSamizat 2/2, 27-33.

36

Bibliographical and page references are to the non-bracketed edition. Works the present author has not seen are marked with *.

VICISSITUDES OF THE GENITIVE RULE

209

Bogorodickij, Vasilij Alekseeviǁ “O naǁal’nom prepodavanii russkoj grammatiki”. In: Oǁerki po jazykovedeni19092 ju i russkomu jazyku: Posobie pri izuǁenii nauki o jazyke, 439-460. Kazan’: Tipo-litografija Imperatorskogo Universiteta. 1912 “Sintaksis roditel’nogo padeža v russkom jazyke”. Russkij filologiǁeskij vestnik 68/3, 211-225. Bogus’awski, Andrzej 1998 “Nochmals zu Genitiv und Akkusativ in russischen verneinten Verbalphrasen”. Die Welt der Slaven 43, 1-32. Borkovskij, Viktor Ivanoviǁ 1978 Istoriǁeskaja grammatika russkogo jazyka 1: Sintaksis: prostoe predloženie. Moskva: Akademija nauk SSSR. Brown, Sue 1999 The Syntax of Negation: A Minimalist Approach. Stanford: Center for the Study of Language and Information. Buslaev, Fedor Ivanoviǁ (1858)18754 Istoriǁeskaja grammatika russkogo jazyka: sintaksis. Moskva: Brat’ja Salaevy. Butorin, D.I. 1966 “Ob osobyx sluǁajax upotreblenija vinitel’nogo prjamogo ob”ekta v sovremennom russkom literaturnom jazyke”. In: G.A. Kaǁevskaja, K.S. Gorbaǁeviǁ (eds.), Normy sovremennogo russkogo literaturnogo upotreblenija, 125-136. Moskva-Leningrad: Nauka. ƀernyšev, Vasilij Il’iǁ (1913)1970 “Kak govorjat v Peterburge”. In: Izbrannye trudy 2, 338-346. Moskva: Prosvešǁenie. (19143)1970 “Pravil’nost’ i ǁistota russkoj reǁi: opyt stilistiǁeskoj grammatiki”. In: Izbrannye trudy 1, 338-346. Moskva: Prosvešǁenie 1970 “Dialekt Moskvy i russkij literaturnyj jazyk”. In: Izbrannye trudy 2, 396-416. Moskva: Prosvešǁenie. Comrie, Bernard, Gerald Stone, and Maria Polinsky 1996 The Russian Language in the Twentieth Century. Oxford: Clarendon Press. Corbett, Greville G. 1986 “The use of the genitive or accusative for the direct object of negated verbs in Russian: A bibliography”. In: Richard D. Brecht, James S. Levine (eds.), Case in Slavic, 361-372. Columbus, Ohio: Slavica. Crockett, Dina B. 1977 “The scope of denial in Russian negative sentences”. Lingua 43, 229-245. de Madariaga, Isabel 1979 “The foundation of the Russian educational system by Catherine II”. Slavonic and East European Review 57/3, 369-395. Donǁeva, L. 1962 *“Njakoi sposobi na izrazjavane kategorijata opredelenost/neopredelenost v ruskite otricatelni izreǁenija v sravnitelen plan s bƼlgarski ezik”. Ezik i literatura 1962/2, 25-40. Durnovo, Nikolaj Nikolaeviǁ, and Dimitrij Nikolaeviǁ Ušakov 1910 *Xrestomatija po velikorusskoj dialektologii. Moskva. Ebeling, Carl L. 1978 Syntax and Semantics: A Taxonomic Approach. Leiden: Brill.

210 Ehlich, Konrad 1982

W. ANDRIES VAN HELDEN

“Anaphora and deixis: same, similar or different?” In: R.J. Jarvella, J. Klein (eds.), Speech, Place and Action, 315-338. Chichester-New York: John Wiley & Sons. Fauconnier, Gilles 1984 Espaces mentaux: Aspects de la construction du sens dans les langues naturelles. Paris: Les éditions de minuit. Fleckenstein, Christa 1961 “Zur Frage des Kasusgebrauchs nach verneinten Transitiva im Russischen”. Wissenschaftliche Zeitschrift der Martin-Luther-Universität Halle-Wittenberg. Gesellschafts- und sprachwissenschaftliche Reihe 10/1, 213-220. Gerškoviǁ, Z.I. 1956 “Satiry Kantemira v pervonaǁal’noj redakcii”. In: A.D. Kantemir, Sobranie stixotvorenij, 501-503. Leningrad: Sovetskij pisatel’. Grandilevskij, Arkadij Nikandroviǁ 1907 Rodina Mixaila Vasil’eviǁa Lomonosova: Oblastnyj krest’janskij govor (Sbornik Otdelenija russkogo jazyka i slovesnosti Imperatorskoj akademii nauk 83/5). Sankt-Peterburg. Greǁ, Nikolaj Ivanoviǁ 1834 *Praktiǁeskaja russkaja grammatika. Sankt-Peterburg. Guiraud-Weber, Marguerite 2003 “Ešǁe raz o russkom genitive otricanija: vzgljad so storony”. Russian Linguistics 25, 363-384. Ickoviǁ, Viktor Aleksandroviǁ 1982 Oǁerki sintaksiǁeskoj normy. Moskva: Nauka. Jakobson, Roman Osipoviǁ (1932)1971 “Zur Struktur des russischen Verbums”. In: Selected Writings II: Word and Language, 3-15. The Hague: Mouton. (1936)1971 “Beitrag zur allgemeinen Kasuslehre”. In: Selected Writings II: Word and Language, 23-71. The Hague: Mouton. Kantemir, Antiox Dmitrieviǁ 1867 Soǁinenija 1. Sankt-Peterburg: Glazunov. Karskij, Evfimij Fedoroviǁ 1912 “Znaǁenie M.V. Lomonosova v razvitii russkogo literaturnogo jazyka”. Russkij filologiǁeskij vestnik 67/1, 65-80. Klenin, Emily 1988 “Timkovskij’s syntax: paraphrase and ‘transformation’ in Lomonosov, Barsov, and Timkovskij”. In: J. Biedermann, G. Freidhof (eds.), Texts and Studies on Russian Universal Grammar 1806-1812, 3: Linguistische, philosophische und wissenschaftsgeschichtliche Grundlagen (Specimina philologiae slavicae, Supplementband 24), 113-123. München: Otto Sagner. Klobukov, Evgenij Vasil’eviǁ 1986 Semantika padežnyx form v sovremennom russkom literaturnom jazyke. Moskva: Izdatel’stvo Moskovskogo universiteta. Kovtunova, Irina Il’iniǁna 1969 Porjadok slov v russkom literaturnom jazyke XVIII - pervoj treti XIX v. Moskva: Nauka.

VICISSITUDES OF THE GENITIVE RULE Kratkie pravila (1773)1982

211

“Kratkija právila rossijskoj grammatiki”. In: Michael Schütrumpf (ed.), Kratkie pravila rossijskoj grammatiki, sobrannye iz raznyx rossijskix grammatik v pol’zu obuǁajušǁegosja junošestva v gimnazijax imperatorskogo universiteta (Specimina philologiae slavicae 41), 1-105. München: Sagner. Krys’ko, Vadim Borisoviǁ Istoriǁeskij sintaksis russkogo jazyka: Ob”ekt i perexodnost’. Moskva: Azbu20062 kovnik. Kurganov, Nikolaj (17935)1978 “Pismovnik 1”. In: Vsevolod Setchkarev (ed.), Pismovnik, 1-394. Würzburg: JAL-reprint. (Title of the 1769 first edition: Rossijskaja universal’naja grammatika.) Larcev, Vasilij Grigor’eviǁ 1988 Evgenij Dmitrieviǁ Polivanov: stranicy žizni i dejatel’nosti. Moskva: Nauka. Lomonosov, Mixajlo (Mixail) Vasil’eviǁ (1746)1952 “Primeǁanija na predloženie o množestvennom okonǁenii prilagatel’nyx imen”. In: Polnoe sobranie soǁinenij 7: Trudy po filologii 1739-1758 gg., 81-87, annotations 801-805. Moskva-Leningrad: Izdatel’stvo AN SSSR. (1748)1952 “Kratkoe rukovodstvo k krasnoreǁiju”. In: Polnoe sobranie soǁinenij 7: Trudy po filologii 1739-1758 gg., 89-378, annotations 805-839. Moskva-Leningrad: Izdatel’stvo AN SSSR. (1752)1959 “Pis’mo o pol’ze stekla”. In: Polnoe sobranie soǁinenij 8: PoLjzija, oratorskaja proza, nadpisi 1732-1764 gg., 508-522, annotations 1003-1008. Moskva-Leningrad: Izdatel’stvo AN SSSR. (1757)1952 “Rossijskaja grammatika”. In: Polnoe sobranie soǁinenij 7: Trudy po filologii 1739-1758 gg., 389-578, annotations 844-888. Moskva-Leningrad: Izdatel’stvo AN SSSR. 1952 “Materialy k rossijskoj grammatike”. In: Polnoe sobranie soǁinenij 7: Trudy po filologii 1739-1758 gg., 595-760, annotations 900-943. Moskva-Leningrad: Izdatel’stvo AN SSSR. Magner, Thomas F. 1955 “Negation and case selection in Russian”. Word 11/4, 531-541. Makeeva, Valentina Nikolaevna 1961 Istorija sozdanija ‘Rossijskoj grammatiki’ M.V. Lomonosova. Moskva-Leningrad: Izdatel’stvo Akademii nauk. Mamonov, V.A., and Ditmar Ƈl’jaševiǁ Rozental’ 1957 Praktiǁeskaja stilistika sovremennogo russkogo jazyka. Moskva: Iskusstvo. Marker, Gary 1985 Publishing, Printing, and the Origins of Intellectual Life in Russia, 1700-1800. Princeton, NY: Princeton University Press. 1994 “Faith and secularity in eighteenth-century Russian literacy, 1700-1775”. In: Robert P. Hughes, Irina Paperno (eds.), Christianity and Eastern Slavs: Volume II: Russian Culture in Modern Times (California Slavic Studies 17), 3-24. Berkeley-Los Angeles-London: University of California Press. Martel, Antoine 1933 Michel Lomonosov et la langue littéraire russe (Bibliothèque de l’Institut français de Leningrad 13). Paris: Imprimerie nationale.

212

W. ANDRIES VAN HELDEN

Meintema, Annie 1986 “The case of the two cases: genitive and accusative in Russian negative constructions”. In: A.A. Barentsen, B.M. Groen, R. Sprenger, (eds.), Dutch Studies in Russian Linguistics (Studies in Slavic and General Linguistics 8), 373-394. Amsterdam: Rodopi. 1988 “Antioch’s accusatives: the object case in A.D. Kantemir’s dispatches and political letters (1732-1744)”. In: A.A. Barentsen, B.M. Groen, R. Sprenger (eds.), Dutch Contributions to the Tenth International Congress of Slavists (Studies in Slavic and General Linguistics 11), 397-419. Amsterdam: Rodopi. Mešǁerskij, Nikita Aleksandroviǁ 1981 Istorija russkogo literaturnogo jazyka. Leningrad: Izdatel’stvo Leningradskogo universiteta. Miklosich, Franz Xaver von 18832 Vergleichende Grammatik der slavischen Sprachen 4: Syntax. Wien: Braumüller. Morison, Walter Angus 1959 Studies in Russian Forms and Uses: The Present Gerund and Active Participle. London: Faber & Faber. Morozova, S.E. 1968 “Roditel’nyj padež”. In: V.I. Borkovskij (ed.), Sravnitel’no-istoriǁeskij sintaksis vostoǁnoslavjanskix jazykov: ǁleny predloženija, 151-178. Moskva: Nauka. Mustajoki, Arto 1985 Padež dopolnenija v russkix otricatel’nyx dopolnenijax 1: izyskanija novyx metodov v izuǁenii staroj problemy (Slavica Helsingiensia 2). Helsinki: Helsinki University Press. Mustajoki, Arto, and Hannes Heino 1991 Case Selection for the Direct Object in Russian Negative Clauses, 2: Report on a Statistical Analysis (Slavica Helsingiensia 9). Helsinki: Helsinki University Press. Nekrasov, Nikolaj Petroviǁ 1865 O znaǁenii form russkogo glagola. Sankt-Peterburg: Paul’son & co. Ostrovitjanov, Konstantin Vasil’eviǁ, et al. 1958 Istorija Akademii nauk SSSR: tom 1 (1724-1803). Moskva-Leningrad: Izdatel’stvo akademii nauk SSSR. Paduǁeva, Elena Viktorovna 1974 O semantike sintaksisa. Moskva: Nauka. Papina, Aza Feodosievna 1964 “Roditel’nyj i vinitel’nyj padeži pri glagolax s otricaniem v russkom literaturnom jazyke konca XVIII - naǁala XIX v.” In: Y.K. Bilodid et al. (eds.), Doslidžennja z ukrajins’koji ta rosijs’koji mov, 197-211. Kyjiv: Radjan’ska Škola. Peškovskij, Aleksandr Matveeviǁ Russkij sintaksis v nauǁnom osvešǁenii. Moskva: Gosudarstvennoe izdatel’stvo. 19202 1923 *Naš jazyk: uǁebnaja kniga po grammatike dlja škol pervoj stupeni. Moskva. (1924)1959 “Ob”ektivnaja i normativnaja toǁka zrenija na jazyk”. Izbrannye trudy, 50-62. Moskva: Gosudarstvennoe Uǁebno-pedagogiǁeskoe izdatel’stvo Ministerstva Prosvešǁenija RSFSR. (19283)20018 Russkij sintaksis v nauǁnom osvešǁenii. Moskva: Jazyki slavjanskoj kul’tury. Potebnja, Aleksandr Afanas’eviǁ (Oleksandr Opanasovyǁ) (18882)1958 Iz zapisok po russkoj grammatike: tom 1-2. Moskva: Prosvešǁenie.

VICISSITUDES OF THE GENITIVE RULE

213

Puškin, Aleksandr Sergeeviǁ (1828)1949 “Vozraženie na stat’ju ‘Ateneja’”. In: Polnoe sobranie soǁinenij 7: kritika i publicistika, 76-80, annotations 669-670. Moskva-Leningrad: Akademija Nauk SSSR; Institut literatury. (1830)1949 “Oproverženie na kritiki”. In: Polnoe sobranie soǁinenij 7: kritika i publicistika, 166-198, annotations 684-689. Moskva-Leningrad: Akademija Nauk SSSR; Institut literatury. (1835)1971 “Egipetskie noǁi”. In: Soǁinenija v trex tomax 3, 424-436. Moskva: Xudožestvennaja literatura. Raviǁ, R.D. 1971 “O vybore padeža prjamogo dopolnenija pri perexodnyx glagolax s otricaniem v russkom jazyke”. In: R.I. Avanesov et al. (eds.), Fonetika. Fonologija. Grammatika: k semidesjatiletiju A.A. Reformatskogo, 254-265. Moskva: Nauka. Restan, Per A. 1960 “The objective case in negative clauses in Russian: the genitive or the accusative?” Scando-Slavica 6, 92-112. Rozental’, Ditmar Ƈl’jaševiǁ (1967)19712 Spravoǁnik po pravopisaniju i literaturnoj pravke dlja rabotnikov peǁati. Moskva: Kniga. Safarewiczowa, Halina 1959 “Forma dope’nienia bliǻszego w rosyjskim zdaniu zaprzeczonym, czNjǧǀ 1”. Slavia orientalis 8/4, 77-109. 1960 “Forma dope’nienia bliǻszego w rosyjskim zdaniu zaprzeczonym, czNjǧǀ 2”. Slavia orientalis 9/1, 69-137. Šapiro, Abram Borisoviǁ (1956)2001 “A.M. Peškovskij i ego ‘russkij sintaksis v nauǁnom osvešǁenii’”. In: Peškovskij (19283)20018, 3-6. Savel’eva, Lidija Vladimirovna 1989 Formy otricanija v russkom jazyke donacional’nogo perioda. Leningrad: Izdatel’stvo Leningradskogo universiteta. Schütrumpf, Michael 1982 “Die kratkija pravila von 1773 und 1784 und einige angrenzende Probleme”. In: Michael Schütrumpf (ed.), Kratkie pravila rossijskoj grammatiki, sobrannye iz raznyx rossijskix grammatik v pol’zu obuǁajušǁegosja junošestva v gimnazijax imperatorskogo universiteta (Specimina philologiae slavicae 41), IXIII. München: Sagner. 1984 Die russische Akademiegrammatik von 1802: Eine sprachwissenschaftliche Analyse (Specimina philologiae slavicae, Supplementband 1). München: Sagner. Setchkarev, Vsevolod 1978 “Preface”. In: Kurganov (17935)1978, 4 pages (no page numbers). Silbajoris, Rimvydas 1968 Russian Versification: The Theories of Trediakovskij, Lomonosov, and Kantemir. New York-London: Columbia University Press. Šmelëva, Tat’jana Viktorovna 2003 “Tradicii universitetskoj slovesnosti”. Vestnik Novgorodskogo gosudarstvennogo universiteta 25, 115-121.

214

W. ANDRIES VAN HELDEN

Smotrickij (Smotryǀkyj), Meletij (1619)1974 “Hrammatiki slavenskija pravilnoe syntagma”. In: Olexa Horbatsch (ed.), Meletij Smotryǀkyj: Hrammatiki slavenskija pravilnoe syntagma (Specimina philologiae slavicae 4), 1-270. Frankfurt am Main-München: Kubon & Sagner. Sokolov, Aleksandr Vasil’eviǁ 2002 Obšǁaja teorija social’noj kommunikacii. Sankt-Peterburg: Izdatel’stvo Mixajlova V.A. Solov’ev, Sergej Mixajloviǁ (1876)1965 Istorii Rossii s drevnejšix vremen 13 (toma 25-26), 471-583. Moskva: Mysl’. Sumarokov, Aleksandr Petroviǁ “O pravopisanii”. In: Polnoe sobranie vsex soǁinenij 10: Raznye prozaiǁeskie so17872a ǁinenija i perevody, 5-38. Moskva. (Date of first publication not found.) 17872b “Primeǁanie o pravopisanii”. In: Polnoe sobranie vsex soǁinenij 10: Raznye prozaiǁeskie soǁinenija i perevody, 38-47. Moskva. (Date of first publication not found.) Tesnière, Lucien 19662 Éléments de syntaxe structurale. Paris: Klincksieck. Timberlake, Alan (1975)1986 “Hierarchies in the genitive of negation”. In: Richard D. Brecht, James S. Levine (eds.), Case in Slavic, 338-360. Columbus, Ohio: Slavica. Tomson (Thomson), Aleksandr Ivanoviǁ (Alexander) (1902)1903 “Vinitel’nyj padež prjamogo dopolnenija v otricatel’nyx predloženijax v russkom jazyke”. Russkij filologiǁeskij vestnik 49, 192-234. 1903 *K sintaksisu i semasiologii russkogo jazyka. Odessa. 1908 “K voprosu o vozniknovenii rod.-vin. p. v slav. jazykax. Priglagol’nyj rod. p. v praslav. jazyke”. Izvestija Otdelenija russkogo jazyka i slovesnosti Akademii nauk 13/3, 281-302. 1912 “Beiträge zur Kasuslehre III: Zur Genetivrektion des Verbums im Baltischslavischen”. Indogermanische Forschungen 29 (1911-1912), 249-259. Trediakovskij, Vasilij Kiriloviǁ (1735)19632 “Novyj i kratkij sposob složenija russkix stixov”. In: Izbrannye proizvedenija, 365-420, commentary 530-536. Moskva-Leningrad: Sovetskij pisatel’. Ueda, Masako 1992 The Interaction between Clause-level Parameters and Context in Russian Morphosyntax: Genitive of Negation and Predicative Adjectives (Slavistische Beiträge 295). München: Otto Sagner. Uëmov, A.I. 2003 “Lomonosov i filosofija”. Vestnik Moskovskogo universiteta. Serija 7: Filosofija 5, 123-128. Uglitsky, Zinaida 1956 “Accusative and genitive with transitive verbs preceded by a negative in contemporary Russian”. The Slavonic and East European Review 34/83, 377-387. Uspenskij, Boris Andreeviǁ (1981)1997 “O ‘rossijskoj grammatike’ A.A. Barsova (1783-188 gg.)”. Izbrannye trudy 3: Obšǁee i slavjanskoe jazykoznanie, 628-656. Moskva: Škola “Jazyki russkoj kul’tury”. (1992)1997 “Dolomonosovskie grammatiki russkogo jazyka (itogi i perspektivy)”. In: Izbrannye trudy 3: Obšǁee i slavjanskoe jazykoznanie, 437-572. Moskva: Škola “Jazyki russkoj kul’tury”.

VICISSITUDES OF THE GENITIVE RULE

215

Vaillant, André 1977 Grammaire comparée des langues slaves 5: la syntaxe. Paris: Klincksieck. van het Reve, Karel 1985 Geschiedenis van de Russische literatuur. Amsterdam: G.A. van Oorschot. Veǁerka, Radoslav 1963 “Sintaksis bespredložnogo roditel’nogo padeža v staroslavjanskom jazyke”. In: Josef Kurz (ed.), Issledovanija po sintaksisu staroslavjanskogo jazyka, 183-223. Praga: ƀexoslovackaja Akademija Nauk. Vinogradov, Viktor Vladimiroviǁ, and Evgenija Samsonovna Istrina 1954 Grammatika russkogo jazyka: Tom II: Sintaksis: ƀast’ vtoraja. Moskva: Izdatel’stvo Akademii nauk SSSR. Vinokur, Grigorij Osipoviǁ (1937)1959 “Puškin i russkij jazyk”. Izbrannye raboty po russkomu jazyku, 189-206. Moskva: Uǁpedgiz. Vlasto, Alexis P. 1986 A Linguistic History of Russia to the End of the Eighteenth Century. Oxford: Clarendon Press. Vostokov, Aleksandr Xristoforoviǁ 1831 *Russkaja grammatika (po naǁertaniju ego že sokrašǁennoj Grammatiki polnee izložennaja). Sankt-Peterburg. Xarpaleva, V.F. 1985 “O variativnosti form ob”ekta v pamjatnikax russkoj pis’mennosti XV-XVII vv.”. In: Ju.N. Karaulov (ed.), Vostoǁnye slavjane: jazyki, istorija, kul’tura, 100107. Moskva: Nauka. Zaliznjak, Andrej Anatol’eviǁ Drevnenovgorodskij dialekt. Moskva: “Jazyki slavjanskoj kul’tury”. 20042 Zelenin, Dimitrij Konstantinoviǁ 1913 Velikorusskie govory s neorganiǁeskim i neperexodnym smjagǁeniem zadnenebnyx soglasnyx. Sankt-Peterburg: Otdelenie russkogo jazyka i slovesnosti Imperatorskoj Akademii Nauk. 1914 *Velikorusskie skazki permskoj gubernii (Zapiski Imperatorskogo Russkogo Geografiǁeskogo Obšǁestva po otdeleniju Ljtnografii 12). Petrograd. Zemskaja, Elena Andreevna 1973 Russkaja razgovornaja reǁ’. Moskva: Nauka. Živov, Viktor Markoviǁ 1996 Jazyk i kul’tura v Rossii XVIII veka. Moskva: Škola “Jazyki russkoj kul’tury”.

SUMMARY The article treats the history of the rule that prescribes the use of the genitive to mark the object of a negated verb in Russian. On the basis of assembled evidence on contemporary reflection on language and available descriptions of written usage, it is argued that the appearance of the rule in 17th-century Church Slavonic and 18th-century Russian grammars, as well as its disappearance in the first half of the 20th century, must be understood in terms of specific historical events and generic social and political mechanisms rather than in terms of autonomous linguistic change occurring over the investigated period. The paper concludes with an attempt to find an explanation why the literate part of the population tolerated a rule that lacked support in spoken usage for so long. Here a linguistic

216

W. ANDRIES VAN HELDEN

explanation is submitted. It is suggested that the semantics of the accusative of negation was to a certain extent incompatible with the pragmatics of written communication, as in the case of some deictic categories. As a consequence, non-narrative writing was not seriously affected by the ban on the use of the accusative, the genitive being more amenable to expressing negation in serious descriptive prose anyway.

Dutch Contributions to the Fourteenth International Congress of Slavists, Ohrid: Linguistics (SSGL 34). Amsterdam – New York: Rodopi, 2008, 217-258.

ON THE COMPOSITION AND LANGUAGE OF THREE EARLY MODERN RUSSIAN-GERMAN PHRASEBOOKS PEPIJN HENDRIKS AND JOS SCHAEKEN

1. Introduction In the 1930s and 1940s, bilingual phrasebooks compiled by foreigners were discovered as a source of information on older stages of Russian. Approximately fifteen such phrasebooks from the 16th and 17th centuries have been handed down to our times. The present contribution will focus on three genetically related phrasebooks linked to the Northwest-Russian town of Pskov: the phrasebook by Tönnies Fenne (1607; henceforth F), Einn Russisch Buch by Thomas Schroue (S) and the Anonymous Ein Rusch Boeck … (A), both of which are slightly older.1 Before anything else, the interest in the three phrasebooks has been of a lexicographic and lexicological nature. The lexicological research is often explicitly comparative and treats the phrasebooks as representatives of the long-standing tradition they are part of. The comparative attitude of this research has led to the realisation that F, S, and A share a protograph, whose origin has been projected back as far as the last third of the 13th century (Xoroškeviǁ 2000). The present contribution offers a concordance to the text of F, S, and A (see the Appendix). On the basis of this concordance, we will show how the compiler of F selected, reorganised, and critically examined the material available to him. We will argue that close textual comparison of the composition and language of F and S can shed light on the tradition these texts are part of, on the protograph of F and S, as well as on the input of the compiler of F. In a separate paper elsewhere in this volume (Hendriks 2008), one of the authors will show how the linguistic research into the phrasebooks – which has so far treated the manuscript as a much more independent source – can benefit from the philological insights presented in this contribution.

1

All references to F are based on the electronic text edition by Hendriks and Schaeken (version 1.1, July 2008). References to S and A point to the text in Fa’owski 1997 and Fa’owski 1994 respectively.

218

PEPIJN HENDRIKS AND JOS SCHAEKEN

2. State of the field 2.1 The three manuscripts The interest in foreigners’ accounts of the Russian language as a source of information about its older stages dates from the late 1930s and the years immediately following the Second World War.2 The Soviet scholars who turned to these foreign sources motivated their choice with the argument that they more reliably reflect the East-Slavic vernacular than other sources, which were influenced by Church Slavonic (see Larin 2002: 5-20).3 Approximately fifteen such accounts have survived, varying in size, quality and contents. 4 Typically, they contain a section with lists of lexical items, and a section with phrases or little dialogues. In the present contribution we will focus on three phrasebooks: • • •

the phrasebook of Tönnies Fenne (F), 502 pages, date of 1607 in the text, 5 Low German, four-volume edition (TF I-IV) and electronic text edition (Hendriks and Schaeken 2008); Einn Russisch Buch by Thomas Schroue (S), 224 pages, dated by the watermarks between 1582 and 1591, 6 date of 1546 in the text, High German, facsimile and text edition (Fa’owski and Witkowski 1992, Fa’owski 1997); the Anonymous Ein Rusch Boeck … (A), 183 pages, dated mid-16th century, 7 High German, facsimile, text edition, and analysis (Fa’owski 1994, 1996).

Historically, these three phrasebooks are connected to circles of the Hanseatic League and to the League’s trading activities with the Russian cities of Novgorod and Pskov, through Reval (Tallinn), Narva, and Dorpat (Tartu) (see, e.g., Angermann and Endell 1988: 96f.). Merchants could use these manuscripts when dealing with their Russian counterparts. As such, these phrasebooks are

2

The most important name in this respect is B.A. Larin, whose publications on the topic culminated in a doctoral dissertation in 1946 (republished as Larin 2002). 3 Krys’ko points out that the political situation in the Soviet Union played a role in the use of these secular documents as well. At the time, everything Old Russian – let alone Church Slavonic – was suspect (Krys’ko 2007: 107). 4 See for an overview Günther 1994 and the introductory chapter in Mžel’skaja 2003 (5-40). 5 A renewed attempt to pinpoint the watermarks of the paper of F in 2006 (cf. TF I: 7f.) has not been successful. The authors express their gratitude to Andrea Lothe, section Papierhistorische Sammlungen of the Deutsches Buch- und Schriftmuseum der Deutschen Nationalbibliothek Leipzig. 6 Fa’owski and Witkowski 1992: 14, based on Stone 1991: 23. 7 Fa’owski 1994: 10.

EARLY MODERN RUSSIAN-GERMAN PHRASEBOOKS

219

firmly rooted in a Western tradition of compiling manuals to cope directly with foreign languages.8 Being aware of this tradition, scholars have realised from the onset that F, S, and A cannot be considered independent works of one single author. The authorship of S was discussed by M.P. Alekseev as early as 1951, and in the preface to the facsimile edition of F (1961), Roman Jakobson and Elizabeth van Schooneveld concluded that F and S were based on a “common model” (TF I: 19). For a long time, however, the exact relationship between the three phrasebooks remained unclear, as both S and A – previously in the Prussian State Library in Berlin – were considered to be lost after the Second World War. 2.2 Research to date Lexicographers welcomed the edition of F, and quickly started to use the Russian data from the manuscript as a valuable source for dictionaries such as the ӷԙӪӹӴԙӪӠӡ ӹԼԊԂԙԽӰӹӡ ԙԊӹӴԂӢӸ ԙ ӠԙԽӹӢӠԓӾԙӪӠԐӠ ԌԂӰӰՂԐӠ (POS) and the ՀԊӹӴԂӢӸ ӢӫԙԙӪӹԄӹ ԋԒՂӪԂ XI-XVII ӴӴ. (SRJa XI-XVII). After the manuscripts of S and A had resurfaced in Cracow in the late 1970s (see Whitehead 1976, 1980, also Stone 1990: 341-344, Bolek 2003: 213), S and A became available to a wider audience in the 1990s. In lexical studies, scholars such as Ol’ga Mžel’skaja, Larisa Kostjuǁuk, Adam Fa’owski, and Anna Bolek examined the lexicon of F, S, and A – and often of a number of other phrasebooks – from an impressive range of perspectives (see, e.g., the bibliography of Mžel’skaja 2003, and Bolek 2003). Their research into these three phrasebooks is an important contribution to our knowledge about the historical lexicon of the Russian language. As a result of the mainly comparative nature of the lexicological research – which has ensured that time and again the scholars were confronted with the tradition the phrasebooks are rooted in – it became clear very soon that the similarity between F, S, and A goes far beyond a mere thematic similarity and a “common model”. In fact, the manuscripts of F, S, and A ultimately share the same protograph. Within the tradition, F and S are more closely related to each other than to A. To give a rough impression: approximately 80% of the material in S also occurs in F (see Fa’owski 1997: 10, and sections 3.3 and 3.4 below). Anna Xoroškeviǁ and Anna Bolek date the protograph for F and S at the end of the 15th century (Xoroškeviǁ 2000: 85-91, Bolek 2003: 215). On the basis of a number of lexical archaisms – mainly referring to coins and scales – the protograph of all three 8

The oldest merchant phrasebook known of – with High German and Italian as its languages – dates back as far as 1424, and was compiled by a Master George of Nuremberg (see Gernentz et al. 1988: 21-23). The tradition itself is even older.

220

PEPIJN HENDRIKS AND JOS SCHAEKEN

manuscripts is pushed back as far as the last third of the 13th century (Xoroškeviǁ 2000: 85-91).9 When we take a look at the more linguistic research into the three phrasebooks, we face quite a different picture. Of the three phrasebooks, F has received most attention. Not only is F by far the largest phrasebook, but it has also been accessible to scholars since the 1960s, and as such has received more attention than S and A.10 Yet even after more than half a century, there is no full description and analysis of the language of F,11 and its material is usually used rather than explicitly studied. Publications that exceptionally do look at the text of F as an object of study are Jakobson 1971 (1966), Zaliznjak 1986: 133, and Schaeken 1992 (all on the nominative singular in -e), Mürkhein 1979 (on polnoglasie), Gluskina and Bol’šakova 1988 (on sokan’e/šokan’e and on initial v-/f- in positions where Old Russian dialects typically had initial u-), and Le Feuvre 1993 (on e > o). The most notable exception, however, is Zaliznjak 1998. In this article Zaliznjak made a case for the manuscript’s reliability as a linguistic source. In doing so he drew F into the ambit of Russian birchbark research. 12 He takes the data at their face value, and shows that individual words and their morphology, but also syntactic constructions that seemed strange or downright wrong to the editors of the manuscript, fit the Northwest-Russian data wonderfully well.13 His conclusion is that “ҰҸҍҲѹ ӊҸ ӊҡѦѽ ҡґҏҍҽҾѽ ҰҝѦҦґҸңѦӁӁӃѦ ӊ ӁҽҡҲҸҾӂѦҮ ҡҲҽҲҧѦ ѹӁҲѦҝҰҝѦҲҽ҃ѹѹ ӊѦҦҏҲ Ҕ ҲҸѸҏ, ҍҲҸ ӉҽҰѹҡҧ ҨѦӁӁѦ ѸҸңѦҲ қӃҲҧ ҰҝѹӉӁҽӁҽ қҸґѦѦ ҦҸҡҲҸӊѦҝӁҸҮ, ҍѦѸ ҰҝѦҦҰҸґҽҊҽґҸҡҧ ҝҽӁѦѦ” (Zaliznjak 1998: 274). We seem to be faced here with two strands of research. One qualifies Tönnies Fenne as a mere “kolejn[y] kompilator[]” (Bolek 1997: 63), the other treats F as a “қѦҡ҃ѦӁӁӃҮ ѹҡҲҸҍӁѹҔ” (Zaliznjak 2004: 14). These qualifications are not mutually exclusive per se. Yet if one thing is clear, it is that once and for all we must forsake the idea that Tönnies Fenne went to Pskov to compile his phrasebook merely based on what he heard in the streets of the town, aided in his undertaking by one or more native informants (cf. also Mžel’skaja 2003: 200). This raises 9

This would make it older than the oldest phrasebook known to exist; see footnote 8. Nevertheless, A is the only phrasebook of which a linguistic analysis is available (Fa’owski 1996). Unfortunately, it is hardly exhaustive: syntactic features, for instance, are barely touched upon (see Krys’ko and Šalamova 1998). S has received almost no linguistic attention (see the introductory chapters to Fa’owski and Witkowski 1992 and Fa’owski 1997). 11 This description is now under preparation as part of the PhD project (2005-2010) of Hendriks. 12 The manuscript of F is approximately 150 years younger than the most recent birchbark letter (mid-15th century), but stems from the same general dialect area. The Old Pskov dialect belonged to the Old Novgorod dialect “ӊ җѹҝҸҔҸѸ ҡѸӃҡґѦ” (Zaliznjak 2004: 13). 13 Something which earlier had been suspected by Helge Poulsen (1972: 214). 10

EARLY MODERN RUSSIAN-GERMAN PHRASEBOOKS

221

the question, however, as to the exact position of F within the tradition, and of the compiler’s attitude towards his sources and his own input. 2.3 The present contribution In the light of the above, we will argue that a philological and a linguistic approach should be combined in order to come to a full appreciation of the language of F. For our observations, we will use a concordance of the text of F, S, and A, which we present as an Appendix to this article. This concordance is based on the references given in the edition of S, but has been considerably expanded and corrected, and for the first time also includes corresponding phrases in A. First we will look at the textual organisation of F as compared to that of the other two phrasebooks (section 3.1). The various occurrences of introductory phrases in the three manuscripts will be discussed (section 3.2) and we will very briefly touch upon the contents of their lexical sections in order to get a full picture of the relationship between the manuscripts (section 3.3). The main focus, however, will be on the composition of the phraseological sections of F and S (section 3.5), as they turn out to be the most closely and intimately related. The phraseological section of A is related as well, albeit more loosely (section 3.6). Our observations do not only illustrate the nature of the tradition, but also the position of F within it, and the way the compiler of F approached his sources. Then, we will turn to the language of the phrasebooks (section 4). We will see that the attention of the compiler of F for the composition of his phrasebook also extends to its language. Close textual comparison will lead to new insights on the protograph shared by F and S (section 5). The insights gained from the observations about composition and language shed light on the input of the compiler. This clears the view on the data as well as on the underlying linguistic system that can be gleaned from them. 3. Textual organisation of F, S, and A 3.1 General The topics covered in the three phrasebooks F, S, and A largely overlap. However, the phrasebook of F reveals the most logical and systematic way in which a learner’s guide of a foreigner to the Russian language can be compiled (cf. Bolek 1997: 65). The composition of F in comparison with that of S and A is illustrated by Tables 1 and 2: with regard to the different sections and the order in which they

222

PEPIJN HENDRIKS AND JOS SCHAEKEN

appear in the manuscripts (Table 1), as well as with regard to the textual length of each section (Table 2).14 Fenne

Schroue

Anonymous

Introduction 1-24

Introduction 1r-2v

Introduction 1r-4v

Phraseology I 3r-64v

Religious 5rv

Religious 65r-67v

Phraseology I 7r-8v

Lexicon 31-130

Lexicon I 68r-88r

Lexicon I 9r-36v

Grammar 131-184

Grammar 88v-94v

Numbers 37rv

Phraseology 187-464

Lexicon II 95r-98v

Lexicon II 38r-39v

Proverbs 469-494

Phraseology II 99r-113r

Grammar 40r

Religious 497-510

Phraseology II 40v-46v

Polish 527-538

Lexicon III 47r-53r

Numbers 545-566

Phraseology III 54r-94v

Table 1: Schematic comparison of the sections in F, S and A

14

For detailed tables of contents see TF I: 11-16, Fa’owski and Witkowski 1992: 15-18, Fa’owski 1994: 12-14.

223

EARLY MODERN RUSSIAN-GERMAN PHRASEBOOKS

300 250 200 150 100 50 0 Introduction

Lexicon

Grammar

Phraseology

Proverbs

Religious

Polish

Numbers

Fenne (502)

14

100

50

272

24

10

12

20

Schroue (224)

3

49

13

153

6

0

0

Anonymous (183)

8

71

1

99

2

0

2

Table 2: Proportional comparison of the sections in F, S and A (in number of pages)

The lexical section of F starts out with single words in thematic clusters, ranging from very general to more specific (commercial) topics, and grouped under headings of the type Van vngeschickten luden boser artt ‘Of unpleasant people of evil character’ and Van allerley lakenn ‘Of all kinds of cloth’. Then, in between the lexical and the phraseological sections, F provides a grammatical sketch,15 including lists of frequently used ‘little words’.16 The phrases in the phraseological section constitute the main part of the phrasebook. As was the case in the lexical section, these phrases are arranged logically, starting with sentences of a more general type – mainly concerning social affairs – and leading up to phrases that can be used in commercial discussions and negotiations (see Table 3 in section 3.5). In close relationship, the phraseological part is followed by more or less fixed expressions, with a separate section of “indecent proverbs, riddles, swear words, bywords and turns of speech” (TF I: 15) at the end, which, of course, also belong to the language of everyday life. The manuscript concludes with several sections that can be considered useful appendices to the main body of the text: religious phrases, Polish texts, and lists of numbers and letters. A closer look at Table 1 shows that – with the exception of the Polish texts – all sections in F also occur in S and A. However, in comparison with F the textual structure of S, and especially A, is less systematic. In S, the lexical and phraseological sections are each located in two different places (with the lexical sec-

15

We will stick to the term ‘grammatical’ used in the table of contents in TF I: 13-14. However, for the section 131-184 (as well as the corresponding parts in S and A), it would be more appropriate to speak of ‘word derivation and word classes’. 16 Cf. the heading Vonn Kleinenn Worten (S 88v 12).

224

PEPIJN HENDRIKS AND JOS SCHAEKEN

tions placed between the phraseological ones), and in A even in three different places. In A, the list of numbers (37r 1–38r 2) is actually more a part of the lexicon (preceded by the unit Van Gelde Vnderschede) than a separate section. 17 In general, A does not always clearly distinguish between the different sections, such as the lexicon and the phraseology (cf. 11v 15-20, 13v 20-24, 18v 13-26, 20r, 20v 19-20, 21r, 27v 1-4, etc.),18 or the brief grammatical part (40r Van Kleinen vockabulen) and the following pages of the second phraseological section. In S, the religious and grammatical sections are located in the middle of manuscript, between the phraseological and lexical parts, although they are still clearly marked off as separate units. The religious section has a few (rough) correspondences with the one in F: F 497 2-11 ~ S 66r 7-8, 498 7-19 ~ 66r 21–66v 6, 499 1-4 ~ 66v 7-8, 499 12-16 ~ 66v 15-17, 501 12-18 ~ 66v 18-23.19 The grammatical part of S is not as elaborate as the one of F (especially when it comes to the list of simple and prefixed verbs on pages 161-184), although occasional matches can be found between clusters of words in the lists of adjectives (cf., e.g., F 133 11-18 ~ S 90v 13-20), pronouns (e.g., F 149 20–150 3 ~ S 88v 19-22), prepositions, adverbs (e.g., F 157 1-6 ~ S 89rb 12-16), conjunctions, and verbs. 3.2 Introductory phrases The introduction of F (14 pages) at the beginning of the manuscript is considerably larger than that of A (8 pages), and especially that of S (3 pages). At the end of the section, F marks the beginning of his enterprise with the following statement: (1) Voima Svætaia Troitza ia potzinall piszatt tu ruschuiu knigu Gospodi Isus Christus Sin Boseÿ pridi komne da posobbi mne ti ruschigi retzi præmo vtzitza da pochvatitt ÿ oposle præmo piszatt ÿ daÿ mne svoiu milost sdorovie ffsvoiem ffstrachu szԂiti. à tie ruschigi retzi vollodiett. Amin. Ihm nahmen der hilligen drefoldicheitt begunde ich dutt rusche boek tho schriuen. herr Iesus 17 The list roughly corresponds with F 545-554. In contrast with F, there is no list of letters in A (see F 565-566). S contains neither numbers nor letters. 18 The division of the lexical and phraseological parts of F and S is usually straightforward. We found one instance of a phrase in F which is located in the lexicon (80 18-19 pust kon segodni opotzinutt – latt datt perdtt van dage rowenn) and which has a counterpart in the second phraseological section of S (103v 17 Pust koenn sogodne opottsÿne – Las das pferdt ruhen). 19 There is one phrase in the religious section of F that has a counterpart in the one of A: F 508 34 ~ A 5r 10-11+12-13.

EARLY MODERN RUSSIAN-GERMAN PHRASEBOOKS

Christus söhne gades kum mÿ tho hulpe de rusche sprake recht tho leren vnd tho vahten vnd darna recht tho schriuen vnd giff mÿ dine gnade here mÿtt gesundttheÿdtt in dinen fruchten tho leuen vnd de rusche sprake mitt leue tho gebrukenn. Amen.

225

(F 22 7-20)

These introductory lines are not the only statement of this kind. The phraseological section is introduced with a similar statement on pages 187-188, and the part of this section that deals with trading issues – starting on page 273, see Table 3 below – is also clearly marked off: (2) Sdies ias bosǖeiu pomotziu potziu piszatt ruskuiu retzi kak nemtzinu sruszinom poruskÿ govorit […] Hÿr wÿll ich mÿtt der hulpe gotts anfangen tho schriuen de rusche sprake alse de dutzschen mÿt den rußen behouen rusch tho spreken […]

(F 188 1-3, 10-13)

(3) Woimæ svætaia troÿtza ias potzinu sdies piszat kak nadob nemptzinu sruszinom torgovat […] Im namen der hilligen drefoldicheitt. wÿll ich hir anfangen tho schriuen wo de dutschen behouen mÿtt den rußen tho koepslagen […]

(F 273 12-14, 6-9)

Like F, the phrasebook of S also has several sets of introductory phrases. There are three introductory statements, located at the beginning of the manuscript (3r), preceding the religious section (64v-65r) and the second phraseological section (99r): (4) Ißde Jagotzu ißboßÿu pomotzÿu potßÿnath kack nadop ißrusszÿmum turguwath […] Itzunndt will ich mit der Hilffe gottes beginnen, Als man mit denn Russen behufet zu kaufschlagen […] (5) Iste Ja gottzu mÿloßzÿrdÿe boßzÿe rußkÿm Jassÿkum pottsÿnath tacko menne aspodÿ boch ffmoÿe ßÿrtze prÿßlath pottsÿnath BockBossobÿ Imne tho pottwattÿth premo rußkaÿa kuÿka pÿssath da Issdorouo bes strachu boßÿe wolodÿeÿth Imnogo dobbro ÿmeth.

(S 3r 3-4, 7-8)

226

PEPIJN HENDRIKS AND JOS SCHAEKEN

Hier will ich mit der Hulff gottes die russische sprache Anfahenn, vnndt so viell mir gott in das Hertze senndt gott help mir sie zu vassen vnd ein recht russisch buch zuschreibenn mit gesundtheit in der furchte gottes zu brauchenn Manniche Liebe zeidt. (6) Issde Ja gottßne mÿlos sÿodÿe boßÿe Pottsÿnath Da Pÿsath kack nadup Ißrußÿmum gauwerÿth. Hier will ich mit der hulffe gottes beginnen vnd Schreibenn, als man behuffet mit den Russenn zu sprechenn, als hiernach volgdt. p.

(S 64v 21–65r 8)

(S 99r 8-12)

We can see that the main introductory statement of F, (1), roughly corresponds to (5), which introduces the religious section of S. The link between (1) in F and (5) in S is strengthened by the fact that the main introduction of F on page 24 gives us the Lord’s Prayer in Russian (without translation), which corresponds with the end of the religious section of S, where the bottom of page 67v reads: Nu volgett das Reussiche Vatter Vnser Auff dieselb sprache.20 Furthermore, the second introductory statement in F, (2), corresponds with (6), the introduction to the second phraseological section of S. The introductions (3) and (4), in F and S respectively, can also be matched (cf. also TF I: 1921). As A is more loosely related to F and S, the introductory statements have fewer correspondences. Still, the one real introductory statement to A, located at the beginning of the first lexical section (9r), can be considered a shortened version of F’s (1) and S’s (5): (7) Hospody Blahoslawi Otze. Gott der herr helpe zw dem Ahnfange. Hospody boch posobbi mene potsinati da konsati Ruskomu Jasicku Vtziti sa Gott helpe mi Ahnfangen Vnd Enden Rusüsche schprak to leren.

(A 9r 1-6)21

20 The next folio of S does not in fact give us the Lord’s Prayer. Instead, the lexical section starts right off with the heading Vonn gott vnd Himlischenn Dinngenn. The catchword Otze nas at the bottom of page 67v suggests that one of the folios disappeared before they were numbered and bound (see Fa’owski and Witkowski 1992: 17). 21 After these introductions S and A continue with the same phrase: S 65r 9-11 Aspoddi Iseus Christus sÿm bosie bomÿluÿ nas grechnich (Her Jesu Christe du sohne gottes Erbarme dich vber vnns Armen sunders) ~ A 9r 7-9 Isus Christus sine bosie pomilui nas gresnych (Jesus Chrüstus sone gades vorbarme die unser Armen sünders). The phrase does not occur in F.

EARLY MODERN RUSSIAN-GERMAN PHRASEBOOKS

227

On top of this, an echo of introduction (3) of F and (4) of S is heard in a loose phrase (German only) in the third phraseological section of A: (8) Hiernach vollgett wo man met den Rossenn schall kopschlagenn.

(A 59r 1-2)

Despite the more distant relationship between F and S on one side, and A on the other, the similarities are still striking. For example, after the introductory statements (5) and (7), S and A continue with the same phrase, absent in F: (9) Aspoddi Iseus Christus sÿm bosie bomÿluÿ nas grechnich. p Her Jesu Christe du sohne gottes Erbarme dich vber vnns Armen sunders. p. Isus Christus sine bosie pomilui nas gresnych. Jesus Chrüstus sone gades vorbarme die unser Armen sünders

(S 65r 9-11)

(A 9r 7-9)

Also, the very first (rhymed) sentences of A can be matched with the ones of S, again to the exclusion of F: (10) Einn Russisch Buch binn ich genanndt, Ihm deutschenn Lanndt gantz vnbekanndt,

(S 1r 1-2)

Ein Rusch Boeck Bin Ick Genanth / mit velen Ehrlücken lüden sey Ick Allto wol nicht Bekhant

(A 1r 1-3)22

Several of the next rhymed sentences on pages S 1r, 2r and A 1r are also rather similar: (11) geschriebenn […] Lebenn23 erdenn […] werde Lerenn […] bekerenn geschrewen […] leben Erden […] werd(en) leren […] keren

(S 1r, 2r)

(A 1r)

22 Cf. the very similar passage S 25 2-3, and M.P. Alekseev’s comments concerning their authorship (Alekseev 1951: 107-109). 23 The words only rhyme in Low German; see section 5 on the translation of S from Low German into High German.

228

PEPIJN HENDRIKS AND JOS SCHAEKEN

Close textual correspondences between the introduction of A and F – this time excluding S – can be found on the verso side of the same folio in A and page 14 in F, for example: (11) Wiltu in Rußlandtt de sprake lehren so laht dÿ van den rußen nicht vorföhren (12) Wultu Ihn Rußlandt de schpracke leren/ so lathe die vor den hungerigen Russen nicht vorferen/

(F 14 1-2) (A 1v 4-5)24

The repeated introductions in S have led some scholars to assume that S unites several previously existing phrasebooks (see Bolek 1997: 64, Xoroškeviǁ 2000: 83f.). The somewhat haphazard impression made by the introductions, contrasts with the more rational arrangement of the same introductory statements in F (Xoroškeviǁ 2000: 84). The fact that the compiler of F has retained the various introductory sections – and arranged them in a more rational way – suggests that he considered the various sections as more or less independent parts or chapters that could be arranged in different ways.25 3.3 Lexical sections As has been pointed out above (section 2.2), the lexicon of F, S, and A has been studied in considerable detail, including the vocabulary in the lexical sections; a recent work in the field is Mžel’skaja 2003. In this section, we therefore restrict ourselves to a few general comments. The lexical section of F contains approximately 1,950 entries. It is subdivided into a general vocabulary with 40 different chapters (running from Van den veer elementenn to Van schepenn), and a vocabulary for trading with another 14 chapters (starting with Van allerleÿ pelterienn and ending with Van allerleÿ lakenn). More than half of the entries in the lexical section of F are also attested in S and almost every single chapter of F can be encountered in S (see Gernentz et al. 1988: 63-76 for some detailed comparisons). 24

The other correspondences are F 14 7-8 ~ A 1r 16-17; 14 9-10 ~ 1v 1-2; 14 11-12 ~ 1v 9-10; 14 13-14 ~ 1v 11-12; 14 21-22 ~ 1v 13-14; 14 25-26 ~ 1v 6. 25 The retention of double occurrences of the same phrase in the various sections could be seen as a confirmation of this perceived independence. Examples of phrases that are repeated literally or with some variation: 189 9-11 ~ 191 3-8 ~ 274 1-5; 214 17-18 ~ 472 8-11; 217 15-17 ~ 277 10-13; 219 4-7 ~ 332 1-4, 219 11-12 ~ 390 21-23, 225 13-17 ~ 348 16-20; 229 19-20 ~ 472 3-4, 257 1-4 ~ 331 57. Note that within the large section on trading repetitions also occur: 285 13-14 ~ 322 13-14, 288 16-20 ~ 417 4-7, 293 1-4 ~ 316 15-22, 296 20-21 ~ 318 17-18, 301 1-6 ~ 336 17-20, 305 6-8 ~ 339 2022, 321 1-2 ~ 384 6+10, 336 10-12 ~ 439 13-15, 339 12-15 ~ 349 18-23, 356 1-5 ~ 463 9-13.

EARLY MODERN RUSSIAN-GERMAN PHRASEBOOKS

229

Although S normally contains fewer entries per chapter, the way in which the lexemes are ordered is remarkably the same. It is striking that the first lexical section of S (68r-88r; cf. Table 1 above) nicely corresponds with the general vocabulary of F (31-107), whereas the second section (95r-98v, and also the first five lines of 99r) precisely covers the vocabulary for trading in F (108-130). The connection between F and A – whose combined lexical sections make up a considerable part of the entire document (cf. Table 2) – is less obvious than between F and S. Nevertheless more than one third of the lexemes of F can also be found in A, frequently arranged in exactly the same order. Examples of such corresponding passages are F 189-190 ~ A 20v and F 259 ~ A 20v, 21v. In conclusion: comparison with S and A shows that at least two thirds of the vocabulary in F is not original. 3.4 Phraseology: general The Appendix to the present contribution provides a list of textual correspondences in the phraseological sections of F, S, and A. It should come as no surprise that in general, the list shows far more correspondences between F and S than between F and A. A rough statistical comparison (cf. also Table 3 below) yields the following figures: • •

• •

Of a total of 685 entries in the section which could be labelled ‘Phraseology: social’ (187-261), 198 phrases can also be found in S (29%). If we also include A, there are 232 corresponding phrases (34%). In the section ‘Phraseology: trading’ (273-464), the overlap between F and S is considerably higher: of a total of 991 entries in F, 707 are also attested in S (71%). If we also take into account the small amount of exclusive correspondences between F and A (29 instances), the percentage rises to 74. The section ‘Proverbs: various’ (469-482) has 86 entries, of which 16 correspond with S (19%) and 20 with S and A (23%). In the section ‘Indecent sayings’ (485-494) 26 we only found one phrase that is attested elsewhere.27

26

This section is called “indecent proverbs, riddles, swear words, bywords and turns of speech” in TF I (15). 27 Tÿ ne sibaies særini gusi seres ogoroda da krasnoi sonki. Du werpest keÿne braden honer göse vnd suuerliche fruwen auer den tuhen. (F 487 9-12) Ty ne schibass sarenych da warenych gusy tzeres tyn Du werpest keine bradene nach gesadene guse Auer den tom It looks as if F added da krasnoi sonki to the standard proverb.

(A 90v 12-13)

230

PEPIJN HENDRIKS AND JOS SCHAEKEN

Thus, 0f the total sum of 1,762 entries in F, 989 phrases are also attested in S and/or A. It is mainly due to the overwhelming number of corresponding phrases in the trading sections of F and S (71%), that the overall percentage of non-original phrases in the main phraseological sections in F is at least 56%. 3.5 Phraseology: F versus S The composition of F and S reveals a clear correspondence between the different phraseological sections: Fenne

Schroue

Social 187-261 Phraseology

Phraseology I 3r-64v

187-464

Trading 273-464 Social 99r-108r Phraseology II 99r-113r

Various 469-482 Trading

Proverbs 469-494 Indecent

108v-113r

485-494

Table 3: Close-up comparison of the phraseological sections of F and S

The concordance further shows an overwhelming amount of exact parallel sequences of phrases, testifying to the extremely close textual relationship that F and S enjoy. Speaking in broad terms, the section ‘Phraseology: social’ (F 187-261) first runs parallel with S 99r-108r, then with parts of S 3r-64v. The section ‘Phraseology: trading’ (F 273-464) first closely follows S 3r-64v, then matches S 108v-113r, and at the end again relates to parts of S 3r-64v. The majority of correspondences in the section ‘Proverbs: various’ (F 496-482) can be found in the section S 3r-64v. A closer look at parallel sequences of phrases reveals the keen eye of the compiler of F for the layout of his text. In quite a few cases, a sequence which runs parallel with S is disrupted by the last phrase at the bottom of the page. A typical example is F 341-342, where F follows the sequence of S 32rv, with the exception of the last phrase on page 341:

EARLY MODERN RUSSIAN-GERMAN PHRASEBOOKS Fenne … 341 16-19 341 20-21 341 22-23 (end of page) 342 1-6 342 7-10 342 11-12 …

231

Schroue … 32r 15-18 32r 19-20 32v 5-6 32r 21-24 (end of page) 32v 1-4 32v 7-8 …

Table 4: Typical order of phrases at the end of a page in F in relation to S

The phrase after 341 20-21 in F is not the counterpart of S 32r 21-24, but a phrase (341 22-23) which corresponds with a phrase located on the next page of S (32v 5-6). On the next page, F picks up the order of S again – leaving out, of course, the already used 32v 5-6. The principle that seems to govern this deviation from the standard pattern is that the compiler of F does not split up a single phrase (including the Russian and German sentences) over two different pages. In this particular case, F had room for one more phrase after 341 20-21. We assume that the next phrase in his source corresponded with S 32r 21-24, which, however, was too long for the space left – it turns up as F 342 1-6, taking up six lines. His solution was to pick two lines somewhat further on in his copy.28 This procedure may also explain another category of cases where parallel sequences of phrases are disrupted. These are instances in which the last phrase of a page in F lacks a counterpart in S. For example, the phrases on pages 456 and 457 in F closely follow the phrases on S 15v and 16r, with the exception of the last phrase on page 456 (20-21):

28 Other typical examples of this procedure can easily be found throughout the phraseological section of F, e.g., 201 22, 285 18-22, 286 21-23, 296 20-21, 300 21-22, 301 19-22, 308 18-21, 312 19-21, 315 16-21, 325 20-21, 331 20-21, 336 21-22, 341 22-23, 352 22-23, 364 21-23, 366 20-23, 367 21-22, 384 21-23, 385 21-22, 386 21-22, 389 20-21, 394 21-22, 397 21-22, 405 20-21.

232

PEPIJN HENDRIKS AND JOS SCHAEKEN Fenne … 456 14-16 456 17-19 456 20-21 (end of page) 457 1-3 457 4-6 457 7-9 …

Schroue … 15v 15-1629 15v 21-23 15v 24-26 (end of page) 16r 1-2 16r 3-4 …

Table 5: A parallel sequence in F and S is disrupted by a missing phrase in S

In a number of cases these phrases in F – which lack a counterpart in S – correspond with a phrase in A. Examples of this are F 278 23-24 ~ A 75r 18-19; F 288 21-23 ~ A 83v 8; F 305 16-20 ~ A 72r 16-18. This means that these phrases in F stem from some written source.30 There is a final set of phrases which can be related to layout considerations on the part of the compiler of F. There are pages in F that have no correspondences in S with the exception of the last phrase of the page. The following table illustrates a typical case: Fenne … 213 15-16 213 17-18 213 19-20 (end of page) 214 1-2 214 3-4 214 5-6 …

Schroue … 67v 12-14 …

Table 6: A phrase from S shows up as the last item on a page without correspondences

Other cases showing this pattern are F 237 22-23 ~ S 106v 5-6 and F 426 21-22 ~ S 101v 11. Of course, this final pattern of missing correspondences can be explained in different ways. All three sets of phrases combined, however, do make clear how close the textual composition of the phraseological part of F to that of S actually is.

29

The missing lines S 15v 17-20 roughly correspond with F 452 3-4 and 5-6. If neither S nor A show the phrase, they could stem from a written or from an oral source. In the case of F 456 20-21, a written source will have to be assumed. The strange form menetzu in this phrase, transliterated in TF II as ԐӥӰԋԽӸԙԇ, points to the syntactic conversion of earlier Ԗӹԓӫ … ԐӾӰԋԽӸԙԋ (see Hendriks 2008). 30

EARLY MODERN RUSSIAN-GERMAN PHRASEBOOKS

233

3.6 Phraseology: F versus A, S versus A Like in the case of the lexicon, the connection between the composition of F and A is less strong than between F and S. We find correspondences throughout the text of A. In the section ‘Phraseology: social’ (F 187-261) we see some concentration of corresponding phrases on pages 54-59, 86 and 91 of A. The part ‘Phraseology: trading’ here and there shows rough parallel sequences of sentences (e.g., F 273-277 ~ A 59-60, F 297-302 ~ A 68-70), but far less frequently and systematically than in the case of F versus S. Although we have not systematically searched for exclusive correspondences between S and A only, we feel safe in concluding that there are only few cases in which similar phrases in S and A are not shared by F.31 4. Language 4.1 F versus S The close correspondence between F and S is not limited to the composition of the texts. The following examples, which can easily be found throughout the phraseological sections, further illustrate the direct relationship between both manuscripts on the orthographical, grammatical, and syntactic levels: • • • • • •

31

F 190 16 Dobranitz ~ S 99r 16 Dobranithz (cf. TF II: 128, n. 16: “Ukrainian influence in secular greeting”). F 190 17 Doboszdorouie (for ԟӹԼӢӹ ԒԌӹӢӹӴӸӾ) ~ S 99r 14 Dabes drowe (both written without r). F 205 6-7 svoi polat ~ S 103r 8-9 swoÿ pollat. In both cases ԗӹԊԂԽԂ is treated as masculine. F 208 21 fftorg (vp den markede) ~ S 105r 20-21 offturk (both lacking a locative ending). F 279 9-12 sprimka (vp winst) ~ S 5v 3-6 Isprÿmka. In both instances an instrumental case is to be expected. The word ԗӹӴӹԌӰӹ (‘ӊӃҊҸҦӁҸ’, cf. Zaliznjak 1998: 261-262) occurs for the first time in F as povodno (279 19-23). In the corresponding phrase S writes pouodua (5v 11-18), where -du- has to be read as -dn- (as in numerous other instances in S). There are three other cases of ԗӹӴӹԌӰӹ in F, all without d and with a single n (povono 280 6-10, 286 8-14, povonu 376 1-7), most probably reflecting assimilation of dn > nn (idem: 262). In the corresponding cases in S, the word is written in the same way (pouono 5v 19-23, 7v 20-25, powono 46v 22–47r 2).

Examples are S 4r 17-20 ~ A 60r 3-5; 15r 21-24 ~ 80r 4-7; 16v 15-16 / 31r 8-9 ~ 72r 7-8; 33r 6-7 ~ 77r 7-8; 41r 1-2 ~ 85r 7-8; 63r 20-24 ~ 85v 16+17+18; 106v 1-2 ~ 54v 5; 106v 9 ~ 54v 11.

234 • •

• • • • • • • • • • • • •

•

PEPIJN HENDRIKS AND JOS SCHAEKEN

F 287 18-22 prigourival (TF II: 247 ԗӢӠԄӹӴӢӠӴԂԊ) ~ S 9r 5-8 prÿgowrÿuall. F 309 5-12 Moi aspodar velil mnie ottebe dengi ffzæt has the German translation Mÿn herr heffft my tho dÿ gesandtt vnd gehehten van dÿ dat geldt tho entfangen. The words that are crossed out can be found in the corresponding phrase in S: Meinn Her hat mich zu dir gesanndt, das geldt zu holenn (18r 23-28). The explicit addition including iestze at the end of F 311 17-20 (al. iestze bolse sudit) corresponds with the text in S 19r 22-23 (Towar Jeße bolße ßudÿth). F 320 13-14 Sol dorogo büil (for ԼՂԊԂ) ~ S 22v 3-4 Soall dorogo buÿll. F 320 21-23 snakon (for ԒӰԂӪӹԐ) ~ S 23r 10-11 Ißnakonn. F 334 6-9 polübitz ÿ tÿ (ԗӹԊԇԼӠԽԙ, Ӡ ԽՂ, cf. Günther 1974: 790, Zaliznjak 1998: 272) ~ S 29r 8-11 pololubÿtz Ithÿ. F 351 17-18 na moim tovarum (vp myner wahre) ~ S 36r 12-13 na moÿum thowarum. F 354 11-17 ias buit tebe (for ԋԒ ԼՂ ԽӾԼӾ) ~ S 37r 22–37v 3 Jas buth debbe. F 360 1-7 tzob tovar for ԓӹԼ ԽӹӴԂӢ (TF II: 328) exactly corresponds with S 40r 17-23 tzoep thowar. F 371 10-11 Tuoi krasenina (for ԽӴӹԋ) ~ S 45r 8 Tuoÿ kraßmÿna. F 377 1-14 tÿ mecha (3 sack) ~ S 47v 16–48r 5 thÿ mechga (Dreÿ Secke). Here, of course, the letter r is omitted in the numeral in both F and S. F 377 15-23 nasch drusba (for ӰԂԶԂ) ~ S 47r 15-21 nas trusba. F 407 3-8 da kotorum tÿ slubuies (TF II: 381 ӪӹԽӹӢӫ) ~ S 60v 11-16 da kotoroÿm thÿ Ißlubuÿes. F 412 1-5 prismotritz: prigleditz (TF II: 387 ԗӢӠԙԐӹԽӢӠԽԙ / ԗӢӠԄԊԋԌӠԽԙ) ~ S 62v 12-16 prÿßmotritz: prÿgleditzs. F 469 14-17 mosit dielat (kan doen) notes the alternative infinitive (without German gloss) spiraitze as a secondary addition at the end of the phrase (cf. F 210 21-22 spiraitze – twisten). The addition corresponds with the reading of S 112r 3-5: mossÿ spÿratse (kann Zwÿtrechtich sein). The German text of F 471 13-17 reads so erkendt he sine hefft he noch de qwahle tho nehmen (~ S 39v 4-7 so hat er noch dem willenn zuleben. vnnd zuhandlen). The words that are crossed out in F cannot possibly be the translation of the Russian text (ÿno iestze ÿsumilsa ffzæt ~ S Imo Jeße om Ißmueles). However, they occur in S a few lines further on the corresponding page in a phrase that is not attested in F: Wenn ein Mann Ihn Nottenn ist, so erkent er seine vorachtigenn freunde (39v 20-21). This suggests that the the direct source of F also included the phrase.

4.2 F versus A The following examples illustrate that the text of A belongs to the same manuscript tradition as F and S, yet is further removed from F and S in the sense that

EARLY MODERN RUSSIAN-GERMAN PHRASEBOOKS

235

often phrases match only partially and that the correspondences are more indirect, vague and associative: • •

•

• •

•

•

•

•

F 195 21-22 Na velikum saluange bog dall pollno is a combination of two separate phrases on page 27r of A: Dal boch polno in line 16 and Na welicko salowange in line 20. F 215 5 Ia chotzu stoboi sreditza roughly corresponds with A 76v 8-9 Jas stoboi otom pomirymsa. The next phrase in F (215 6-7 Otzum tÿ na menæ varavis – Worumb luchstu vp mÿ) is thematically close to the previous phrase in A (76v 6-7 Mnoho ty Chwastajesch da lsesch – Du lawest die sehr vnd liehst). F 256 16-17 Dolgo tÿ szօiues poru tebe vmerett (Du leuest lange idtt is tÿt dattu steruest) corresponds with A 18v 16-17 Dolgo ty syl na sem swete (Du hefst lange gelewet Auf diesser Erd(en)) and the next line pora tebbe vmerety (Ist tyth dat du sterwest). The preceding sentences in F and A touch upon the theme of sin (“sunde”), although different expressions are used. F 297 7-10 Pravo vbütka mnie licho potzinat ia torguiu spributka da tÿ takovos (~ S 12r 24-27) is a free combination of two short phrases in A: 61r 15-16 Prawo na vbytek chudo torgowat and 17-18 Torgui na pributeck. F 297 11-15 corresponds with A 68v 9-15 and deals with the need of finding a broker to mediate in the trade. Interestingly, the concluding part of the phrase in A (togo dla sto Ja ne snaiu gorasdo ruskogo Jasycka – darumb das Ick nicht khan die Ruschische schpracke woll) is missing in F. In view of the sequence of phrases, F 311 7-9 Ias stoboi tebe rosplatilsa da tebe ne vinovate (~ S 19r 13-15) can be matched with A 74v 14-15 Jas tebe ne winowat da nesnaJu toba. Only tebe ne vinovate corresponds with the text in A. F 319 15-20 (~ S 22r 13-17) is a full phrase consisting of seventeen words. It can be matched with A 80v 9-10, which, however, only records the beginning of the sentence (F Otzum tÿ moi tovar roskladivais: perekladivais … ~ A Czemu perekladass towar). There are more examples in A where only the first words of corresponding phrases in F and S are attested: F 321 1-2 ~ A 81v 1, F 322 20-23 ~ A 82r 3-4, F 323 1-4 ~ A 82r 5-6, F 328 1-4 ~ A 83r 4, F 328 5-7 ~ A 83r 5, F 332 1-4 ~ A 56r 1, F 332 9-14 ~ A 44r 13-14, F 365 1-5 ~ A 73r 3-4, F 393 7-13 ~ A 93r 6-7, etc. In view of the sequence of phrases, F 320 21-23 Moi tovar snakon ne nadob mnie iovo klemit (~ S 23r 10-11) corresponds with A 81r 10-11 and 12-13. However, the correspondence between F and the two different sentences in A is restricted to a few words only: Czemu ty na swoi towar Kleym ne Polosyl and Nenadob wet snackom tzoloweck. F 328 10-12 matches A 83r 8-9, and F 328 16-17 matches A 83r 12-13. The phrases in between do not correspond textually: cf. F 328 13-15 Tzto tebe ff-

236

•

•

•

PEPIJN HENDRIKS AND JOS SCHAEKEN

tom tovaru stalo protoroff na dorogi and A 83r 10-11 Nonetza protoroff welick day desat Rublow. However, both show the word ԗӢӹԽӹӢԔ ‘costs’, which is otherwise rare. The only other phrase in which ԗӢӹԽӹӢԔ is used in F is 319 21-22: Protori mnie stalo velikÿ. This phrase, which is not attested in S, vaguely alludes to A 83r 10-11. Whereas the last three phrases on page 329 in F accurately match the last ones on pages 26r in S, the corresponding part in A (83v 6-7+8-9) only uses some similar words (ԌӹԼՂӴԂԽӸ, ӰӾ ӴӥӢӠԽӸ and ԗӹӢӫԓӰӠӪԔ) in the same theme. F 400 12-15 Ia tebæ otovo spaszaiu: vgimaiu kak druga, vieri tÿ mnie (~ S 57r 3-4) corresponds fairly well with the first part of A 84v 11-12: Ja toba beregu kack druga … The second part of the sentence in A (… podi opet nasady) is completely different and is attested elsewhere in F (214 21 Podi opæd nasad). The two phrases F 408 10-18 and 410 1-7 are combined in a single entry in S 61v 1-16. A 69r 2 my chotzem tebbe dati wo twoJe rutze alludes to F 408 11-12 mÿ tebe dadim ffsuoiu volu da ruku (~ S 61v 3), whereas A 69r 2-3 da skasi pramo ne wosmi po sulu da ne prodai duschu dyawolu corresponds with F 410 1-2 and S 61v 5-7 (both reading ԼӥԙԔ instead of ԌӸԋӴӹԊԔ).

5. The protograph of F and S In our opinion, the observations presented in 4.1 – which can easily be multiplied – give rise to the hypothesis that F and S share a direct common source. In more cautious terms, we cannot agree with the complicated stemma of Xoroškeviǁ (2000: 91), who presupposes a common source for F and S dating back to the end of the 15th century, with two intermediate copies for S and as much as three for F. This stemmatological distance is unlikely in view of the strong philological and linguistic similarity which we have shown above. If we assume that the scribes of F and S used the same source, we could imagine the following scenario for S. 32 At the end of the 15th century, Thomas Schroue, mayor of Dorpat, ordered a Russian phrasebook, which was copied or compiled for him from earlier sources. The language used was his native Low German. This manuscript was copied in 1546, and retained the name of Thomas Schroue. A next copy – the copy we have been calling S – was made between 1582 and 1591 (cf. section 2.1 above). On page 1r, the scribe paid homage to Thomas Schroue, whom he thought was the author, and included the year 1546 of the source he copied (cf. Alekseev 1951: 108). With regard to language, this scribe was obviously not competent in Russian at all (cf., e.g., Fa’owski 1997: 9). His main linguistic merit was the translation of the Low German text of the 32

Cf. also Alekseev 1951, Bolek 1997, Xoroškeviǁ 2000.

EARLY MODERN RUSSIAN-GERMAN PHRASEBOOKS

237

source into High German. A piece of direct evidence of this translation is Low German menn moeth immediately preceding High German Man mus (S 51v 1921), which exactly corresponds with Men modt in F (391 6-8).33 This scenario raises a number of questions. These questions pertain to the nature of the source that F and S copied, and especially to the scribe of F, his knowledge of Russian and his input. Given the lack of knowledge of Russian on the part of the scribe of S, we assume that the scribe of S copied the source of his manuscript in a rather mechanical fashion. He may have rearranged or left out phrases here and there – based on German –, but S is to be regarded as a fairly accurate representation of the text and composition of the common source of F and S. The scribe of F, on the other hand, composed his work in a much more logical and rational way (see above). Moreover, he obviously also knew how to read and write the Cyrillic alphabet. It is used not only in the lexical section, but occasionally in the phraseological section as well.34 Any assumption that the Cyrillic words and passages wѦre written by a different scribe (see, e.g., Prowatke 1985: 69) should be discarded in the light of cases such as Ԍvedolone (59.16), pekiԽtt (108.8), Ԍadut (146.19), Bog Ԍato (190.3), ᖇtzum (209.20), and the remarks ԹӹԊӠ ԌӹԼӢӹ and ԹӹԊӠ ԖӫԌӹ (236.3).35 One could wonder whether the common source of F and S was as corrupt as S. We assume that this was not the case, and that the extreme distortion of the text of S on the orthographical and grammatical levels is due to the scribe of S. 36 Had the common source been equally corrupt, we would have to assume that F corrected his copy from cover to cover in a fundamental way only a native speaker could be capable of. Although we assume that F used a native speaker to make corrections and additions, in the light of the imperfections and violations against the Russian language witnessed in F, we have to conclude that the assistance was only limited. Still, the scribe of F did know Russian to some extent. Structural differences between F and S should be regarded as the input by the scribe of F, with or 33 Gernentz et al. provide a list of cases where the Low German original shines through in the High German text (1988: 39f). 34 Single words (206 22, 212 17, 213 6, 224 18, 228 10, 242 6, 243 6, 245 16, 301 3, 327 20, 418 2, 429 18, 435 2, 435 5 (crossed out), and 442 16), as well as whole phrases (as on pages 444, 445, 488, 489, and 492). 35 It is an interesting question whether or not the direct source of S and F contained Cyrillic. If it did, the scribe of S – not knowing Russian – removed it. If it did not, the scribe of F must have added it. For the moment, we shall leave this question aside. 36 A typical example is S 12r 12-13 Imno thÿ piÿproßÿnaÿs vboff, which is almost unintelligible without the accompanying German equivalent (Du Hieschest zuuiel zu, schlagk etwas ab) and, in fact, the corresponding phrase in F (296 12-13): Mnogo tÿ priproszivaies vbaff.

238

PEPIJN HENDRIKS AND JOS SCHAEKEN

without the aid of a native informant. The attitude of the scribe of F towards his source was free and conscious. This attitude is to be expected considering the very practical nature of the genre. Copies were compiled for individual use in everyday life and the available source (or sources) were adapted, corrected, updated and new – first- or second-hand – information was added. All this means that the list of structural morphological and syntactic differences between F and S noted by Bolek (1997: 65-67), must be considered innovations by F with regard to its source. The most important example is the structural change in the way how future tense is expressed (item 7 on Bolek’s list). Hendriks 2008 illustrates how the scribe of F largely removed the construction ‘Ԗӹԓӫ + INF’ from his source, and discusses the linguistic implications of this change. The following differences between F and S further illustrate what in our opinion should primarily be regarded as the result of the compiler’s own input in F (again, with or without the help of an informant or any other sources he had at his disposal): •

• •

•

•

F 204 20 Tzto tack pachnett (Wadtt stinckett so) corresponds with S 103r 5 Stho kack pachane. The next two phrases on page 204 continue the subject of ‘smelling’: F 204 21 Tzto tack nuchatt, F 204 22 Nuchai na tutt traffka. Both phrases are missing in S. After the digression in F, both texts continue with the same phrases. The German phrases in F 209 17 and S 106r 3 are very similar: Im warhafftiger warheidtt and Inn rechter warheit, respectively. However, they are translated in a totally different way: F Voistinnu praffdu and S Proprauomo slaua. F 228 10-11 reads Perevesi tÿ menæ seres reku (Vor mÿ auer de beke). Thematically it does not belong in the set of phrases on pages 227 and 228, which deal with invitations and guests. It seems that the sentence, which begins with Perevesi, was included as a kind of association to the last word(s) of the previous phrase: F 228 5-9 besz perewodoff: du (~ S 57r 5-9 bes verewodaff perewodu). Etymologically, of course, the association is incorrect. Interestingly, at the end of the line ԗӾӢӾӴӾԒӠ is added in cyrillic by the same hand. F 230 15-17 Nichto isbohu ne dumall to vedait bog odin corresponds with S 112r 6-8. The next phrases in F, which are not included in S, can be considered a string of lexical associations: 230 18-21 Ne nadob mne oboronætza, bog mnie oboronitt da praffda alludes to 15-17 (isbohu; bog) and also to 11-14, beginning with the word Nadob; 230 22-25 Ne vpaddÿvai: torropis: vrobe ot ioga da oboroni sebe otiogo alludes to 18-21 (oboronætza). The long phrase in S 5v 11-18 is split up into two separate phrases in F; the first one at the bottom of page 279 (19-23), the second one at the top of page 280 (1-5). This is in accordance with the layout principle that guides the scribe of F (see above). In a number of other cases a single phrase in S is

EARLY MODERN RUSSIAN-GERMAN PHRASEBOOKS

•

•

•

239

split up in F. ; cf., e.g., S 6r 3-10 ~ F 280 11-15 and 16-20, S 6r 14-23 ~ F 280 1-5 and 6-10, S 7v 7-10 ~ F 285 13-14 and 15-17, S 41r 3-15 ~ F 361 5-11 and 12-19. An interesting case is S 105v 4-7 Ja gottsu stoboÿ oßaklath sa Lossÿth thrÿ grÿuen tho tack kack Ja sthassalle (Ich will dreÿ Marck darauff vorwetten, es ist also wie ich gesagt habe), which was rearranged in F 209 3-4 Ia stoboi osaklu to takkak ia schasale (Ich wÿll mÿtt dÿ wedden dat idt so is alse ich sede) and 209 5-6 Ia na tom saklad salöszu tri griuena to tack (Ich wÿll vm dre ൂ wedden datt is so). Sometimes we also find the opposite procedure, where F combines two separate phrases of S: cf., e.g., F 387 20-23 ~ S 51r 1819 and 20-21, F 414 11-12 ~ S 63v 9-10 and 11-12. In the case of F 404 6-11 ~ S 59r 8-9 and 10-14 the concluding remark in F ffnasche semla vaschi dengi chodæt (in vnsem lande iß iuwe geldt gankbahr) says the opposite of S: off semlÿ waßÿ kumÿ ne godde (In vnserm Lande gildt eur geldt nicht). The phrases in F and S just mentioned above also show that F uses the word ԌӾӰӸԄӠ in stead of ӪӫӰՂ, as in quite a few other cases: F 208 5-6 ~ S 105r 5-6, F 280 21-23 ~ S 6r 11-13, F 296 4-7 ~ S 12r 4-7, F 309 5-12 ~ S 18r 23-28, F 315 9-15 ~ S 20r 26–20v 3, F 321 17-21 ~ S 23v 12-15, F 337 10-16 ~ S 30r 11-16, F 390 17-20 ~ S 52r 19-22, etc. Also note F 281 11-15 dengami inserted right after kunami. Similar cases where different monetary units are involved are F 201 22 dengi ~ S 101v 1 pennÿsÿ, F 395 12-18 trÿ rubloff ~ S 54v 3-8 teÿ grÿuennÿ, and F 396 1-6 due rubli ~ S 54v 9-13 due grÿuenÿ. The changes made in F obviously reflect the actual monetary system of the early seventeenth century (cf. Fa’owski 1997: 9, Xoroškeviǁ 2000: 85f.). Besides ԌӾӰӸԄӠ versus ӪӫӰՂ we find other cases of lexical variation between F and S; cf., e.g., F 275 15-21 na dorogo ~ S 3v 18-22 na puthÿ, F 291 5-10 vootzi ~ S 10r 15-21 off glaßzach, and F 344 19-22 kasaki ~ S 33r 16-19 dregeloff. The word ӪԂԒԂӪԔ does not seem to occur in S, whereas F translates German dreger either as ӪԂԒԂӪԔ or ԌӢԋԄӠԊӸ (e.g., F 345 4-8 dregila ~ S 33r 23–33v 3 dregÿla). On the different words for German wage, wachttschaell in F (ӴӥԙԓԂ ԗӹԊԂԽԂ; Ӵӥԙ, ӴӥԷԔ, ԙӪԂԊӴԂ), S (ԙӪԂԊӴӾӰӠՁԂ; ԙӪԂԊӴԂ, ӴӥԙԔ), and also A (ӴԂԞӰԋ, ӴԂԞӰӠՁԂ), see Mžel’skaja 2003:182f., Xoroškeviǁ 2000: 88f. F 291 5-10 seems to have left out the last part of the corresponding phrase in S 10r 15-21: szitele thÿ Ja slepe (meinstu das ich blinndt seÿ). In other cases we find additions in F, like for instance in 298 1-4 (~ S 12v 13-16), where F adds besz koluverte (sunder argerlist) at the end of the phrase, or in 209 18-19 (~ S 106r 6-7) where F notes ili tusis at the end of a sentence as an alternative reading for the verb soblis in the phrase. F shows many other examples of additional variant readings (mostly of verbs and verb forms) in contrast with S; cf., e.g., F 223 1-4 Prikin: pribaff ~ S 4v 26–5r 2 prÿbaff, F 226 1-4 Buitt veszoll: rodosten: da poteszon ~ S 35v 15-16 Buth weßoll, F 227 8-10 porukudam: porutzilsa ~ S 44r 1-3 porutsull, F 279 5-8 peremetit: peretzest ~ S 5v 1-2 parametÿth, F 308 1-6 naloszit: nakladat ~ S 17v 27–18r 2 na loßÿth, F 311 21-

240

PEPIJN HENDRIKS AND JOS SCHAEKEN

22 rosdumal: rosmuislil ~ S 19r 24-25 rossdumall, F 346 19-22 oblitzu: vlitzu: dovedu ~ S 34r 8-10 oblÿttßu: dauedÿe, F 364 5-7 sa naleiski: nachotki ~ S 42r 12-13 ßa naßotkÿ, F 371 1-5 vnelo: ielo ~ S 45v 13-17 vnello, F 371 19-22 dovöl: vlitzil: oblitzil: perevörnul ~ S 45v 23-25 dowaell: vlÿttsÿll: blittsÿs, F 373 1-11 Ialsa: isimalsa ~ S 109r 15-23 Jalsse, F 402 8-10 Vÿmeti: vÿpachi: vÿtzisti ~ S 58r 14-15 Weÿpas. On the other hand, we also find cases where F leaves out variant readings that are recorded in S; cf., e.g., F 228 12-15 fftzest ~ S 52r 1416 off radost: offzest, F 230 7-10 smuschaies ~ S 64v 1-4 skaßÿs: smußaÿes, F 287 10-17 plosze ~ S 8r 18-23 guße: ploße, F 311 7-9 rosplatilsa ~ S 19r 13-15 platÿles: rosplattÿles, F 341 22-23 dotzsolsa: doszla ~ S 32v 5-6 dotzolssze: detzkles: doßlo. Although S produced a text which is highly corrupt, especially when it comes to orthography and grammar, it still turns out to be a valuable source to trace and to clarify possible mistakes made by F, which partly have already been emended in the edition (TF II). Here are some examples: • •

•

• •

•

F 201 11-12 Sapovedall (Idtt is vorbaden) should probably be read as ԩԂԗӹӴӥԌԂ (TF II: 142 “҈ҽҰҸӊҬҦҽґ (!)”); cf. S 100v 3-5 Tho Jest sapowedan. In F 223 1-4 the Russian translation probably by mistake leaves out the German word wedder; cf. S 4v 26–5r 2 opeth. Similar cases are F 299 19-22, where German noch has no translation (cf. S 13r 26–13v 2 Jeßze) and F 301 19-22, where the negation in Al iestzo tÿ tzitall (oder heffstudt noch nicht gerekendt) is missing (cf. S 14r 21-24 all Jeßze thÿ ne ßÿtall). F 299 11-15 na in tzto promesznik na sgovorit (wat vns de mekeler affsprÿktt) should be read as ӰԂԐ, not ӰԂԶ (as in TF II: 260); cf. S 13r 18-22 stho promeßnÿck mam Ißgouwarÿth and also F 408 10-18 tzto tÿ nam sgovoris (wadt du vnß sprikst) ~ S 61v 61v 1-16 stho thÿ nam Ißgouwerÿs. The emendation in TF II: 278 of F 316 9-12 kupli = Ӫӫԗ-ԊӠ is confirmed by the infinitive kupÿth in S 20v 15-18. In F 320 17-18 Otzum tÿ moi tovar smotzill: pomotzil (Worumb heffstu dÿne wahre genettedt) the possessive pronoun should be ԙӴӹӡ, not Ԑӹӡ; cf. S 23r 4-5 Vtßum tuÿ suoÿ thowar pomotzÿle. There are more cases where F mixes up ԙӴӹӡ with Ԑӹӡ and ԽӴӹӡ, e.g. in 304 1-4 Smetli mnie suoiogo tovaru smotrit (Mach ich dÿne wahre woll besehen). This phrase is not attested in S, but the mistake is obvious if we look at the corresponding text in A (77v 8-9): Smejutli twoi towar posmotryti. In F 335 10-14 Ia ne sameril suoi tovar (Ich hebbe nicht auer de mahte mine wahr vorlauet) the preposition ԒԂ is most probably missing; cf. S 28v 22–29r 3 Ja ne samerÿll ßa ßuoÿ thowar and also F 279 19-23 Tÿ sa suoi tovar samerivaies.

EARLY MODERN RUSSIAN-GERMAN PHRASEBOOKS

•

•

• •

•

• •

241

In view of the variants Ißobißaÿ: Ißobÿt (ӠԒӹԼӠԞԂӡ: ӠԒӹԼӠԌӸ) in S 36r 7-9, the addition isobÿ in F 351 13-16 (after the sentence which contains the verb sobÿsai) should probably be understood as the perfective imperative ӠԒӹԼӠ, and not as a simple “correction for sobysai without the appropriate initial i” (TF II: 319). F 358 21-23 Primeti te dom (Merke dat hueß) is emended in TF II: 327 as ԗӢӠԐӥԽӠ Խ ԌӹԐ. However, in view of S 40v 3-5 Prÿmetÿe the that podworÿe: pallath and also S 51v 6-8 Prÿmeth thÿ, the corrupt passage in F may well include the personal pronoun ԽՂ (ԗӢӠԐӥԽӠ Խ ԌӹԐ) or even both ԽՂ and the demonstrative ԽӹԽ (ԗӢӠԐӥԽӸ Ղ Խ ԌӹԐ). F 372 7-10 rukÿ (TF II 342 “ҝҏҔѹ (!)”) in Schupai rukÿ ffmech (Taste mÿt der handt in den sack) should be corrected into the instrumental singular form ӢӫӪӡ; cf. S 46r 7-10 Szupaÿ ruckoÿ off mech. 378 5-8: F sa to|menæ ne poveszui (darumb vorkerdt mÿ nicht) should be corrected in sa to na menæ ne poveszui; cf. S 48r 10-12 ßa tho namenna ne poueßuÿ and also F 229 1-5 (satim na menæ ne poveszui), 291 1-4 (ne poveszui tÿ namenæ), and 326 8-12 (ne poveszuiu tÿ na menæ, where na was inserted afterwards). Note that in the last phrase the correction in TF II: 290 ԗӹӴӥԷӫ for poveszuiu is confirmed by S 25r 7-10 poneßuÿ. F 386 1-4 Zakuni (De vorkoper) in Sakuni iesdet, da tovaru sakupaiu, da iovo dorosaiut (De vorkoper rÿden, vnd vörkopen, de wahre vnd maken se duer). If we take into account the corrupt passage Sakup mÿ (Die vorkauffer) in the corresponding phrase in S (50r 8-10), F should be emended ԒԂӪӫӰӠ (< NOM ԒԂӪӫԗӾӰӸ), and not ԒԂӪӫӠ (from NOM ԒԂӪӫԗԔ) as in TF II: 357. See also SRJa XI-XVII, s.v. ԒԂӪӫԗӾӰӸ = ԒԂӪӫԗԷӠӪԔ ‘ѷҸҲ, ҔҲҸ ӉҽҔҏҰҽѦҲ ҔҝҏҰӁӃѦ ҰҽҝҲѹѹ ҲҸӊҽҝҽ ҦґҾ ҰѦҝѦҰҝҸҦҽңѹ’, which nicely fits the meaning of Zakuni in F. The emendation in TF II: 398 of F 420 14-17 Mnogl = ԠӰӹԄ-ԊӸ is corroborated by Imnogoll in S 111v 8-10. The same holds for dielal = ԌӾԊԂ (TF II: 399) one page further on in F (421 15-19); cf. delath in S 112r 17-22. In F 463 14-20 dobro na na obemæ ne builo obidno (datt vnß beÿden keÿn vnrecht geschee) na should be corrected in nam; cf. S 46v 16-21 bobbre nam obemo nebude obedduo).

6. Conclusions At the end of section 2 we have raised the questions of the exact position of F within the tradition of compiling and copying phrasebooks, of the scribe’s attitude towards his sources and of his own inputs. What were his merits if he was indeed only one of many successive compilers – as supposed by Bolek? And how should we evaluate the manuscript’s status as a priceless source – a qualification by Zaliznjak – in the light of the tradition it is rooted in?

242

PEPIJN HENDRIKS AND JOS SCHAEKEN

In this contribution, we have argued that F and S go back to a direct common source. This conclusion was drawn on the basis of the closeness of the textual organisation (section 3), as well as of a multitude of exact correspondences brought to light by close textual comparison of both texts (section 4). These exact correspondences show that tradition plays a crucial role in the phrasebook of Tönnies Fenne. At the same time, we have seen that the scribe took an independent and conscious attitude towards his sources, and have suggested a number of corrections to the reading of the text of F (section 5). If we want to come to a full appreciation of the Russian material in F, of the underlying linguistic system, and of the input of the scribe of F, close textual comparison is indispensible. In other words: the study of the language of F would be unthinkable without including the data that S has to offer. The study of the historical Old Novgorod dialect – as reflected in the birchbark documents – could profit from the study of S as well. Here too, close textual comparison of S and F is indispensible. It is the only way to see through the distortions that block the view on the underlying linguistic system of S, which more closely represents that of its source than the system which underlies the data of F. Leiden University REFERENCES Alekseev, M.P. 1951

“‘Kniga russkogo jazyka’ T. Šrove 1546 g. i ee avtor”. In: B.A. Larin et al. (eds.), Pamjati akademika L’va Vladimiroviǁa Šǁerby (1880-1944). Sbornik statej, 103122. Leningrad. Angermann, Norbert, and Ulrike Endell 1988 “Die Partnerschaft mit der Hanse”. In: D. Herrmann et al. (eds.), Deutsche und Deutschland aus russischer Sicht. 11.-17. Jahrhundert, 83-115. München. Bolek, Anna 1997 “Czy Tönnies Fenne zas’uguje na miano pioniera slawistyki?”. In: A. Bolek et al. (eds.), S’owianie Wschodni. MiNjdzy jNjzykiem a kulturƾ. KsiNjga jubileuszowa dedykowana Profesorowi Wies’awowi Witkowskiemu w siedemdziesiƾtƾ rocznicNj urodzin, 63-67. Kraków. 2003 “Pskovskie razgovorniki-slovari v rabotax krakovskix rusistov”. In: Pskov v rossijskoj i evropejskoj istorii (k 1100-letiju letopisnogo upominanija) 2, 213-218. Moskva. Fa’owski, Adam 1994 Ein Rusch Boeck. Ein russisch-deutsches anonymes Wörter- und Gesprächsbuch aus dem XVI. Jahrhundert. Köln-Weimar-Wien. 1996 “Ein Rusch Boeck …”. Rosyjsko-niemiecki anonimowy s’ownik i rozmówki z XVI wieku. Analiza jNjzykowa. Kraków.

EARLY MODERN RUSSIAN-GERMAN PHRASEBOOKS 1997

243

“Einn Russisch Buch” Thomasa Schrouego. S’ownik i rozmówki rosyjsko-niemieckie z XVI wieku. CzNjǧǀ II. Transliteracja tekstu. Indeks wyrazów i form rosyjskich. Kraków. Fa’owski, Adam, and Wies’aw Witkowski 1992 “Einn Russisch Buch” by Thomas Schroue. The 16th-century Russian-German dictionary and phrase-book. Part one. Introduction. Photocopies. Cracow. Gernentz, H.J., et al. (eds.) 1988 Untersuchungen zum russisch-niederdeutschen Gesprächsbuch des Tönnies Fenne, Pskov 1607. Ein Beitrag zur deutschen Sprachgeschichte. Berlin. Gluskina, S.M., and N.V. Bol’šakova 1988 “Razgovornik T. Fenne kak istoǁnik dlja izuǁenija pskovskogo dialekta XVII v.”. In: Pskovskie govory v ix prošlom i nastojašǁem. Mežvuzovskij sbornik nauǁnyx trudov, 16-24. Leningrad. Günther, Erika 1974 Review of TF II. Zeitschrift für Slawistik 19, 788-790. 1994 “Die ersten Russischlehrbücher für Deutsche im 16. und 17. Jahrhundert”. In: E. Eichler, K. Krüger, A. Thiele (eds.), Wort und Text. Slavistische Beiträge zum 65. Geburtstag von Wolfgang Sperber, 49-59. Frankfurt am Main. Hendriks, Pepijn 2008 “The expression of future tense in the phrasebook of Tönnies Fenne: A philological approach”. This volume. Hendriks, Pepijn, and Jos Schaeken 2008 Tönnies Fenne’s Low German Manual of Spoken Russian. Pskov 1607: An electronic text edition. http://website.leidenuniv.nl/~hendriksp1/fenne (version 1.1, July 2008). Jakobson, Roman 1971 (1966) “Dublety tipa soǁkle/soǁël v pokazanijax T. Fenne o jazyke Pskova na poroge XVII veka”. In: R. Jakobson, Selected writings. II: Word and Language, 203208. The Hague-Paris. Krys’ko, V.A. 2007 “Russkaja istoriǁeskaja leksikografija (XI-XVII vv.): Problemy i perspektivy”. Voprosy jazykoznanija 2007/1, 103-118. Krys’ko, V.A., and A.N. Šalamova 1998 Review of Fa’owski 1994 and 1996. Voprosy jazykoznanija 1998/2, 129-134. Larin, B.A. 2002 Tri inostrannyx istoǁnika po razgovornoj reǁi Moskovskoj Rusi XVI-XVII vekov. Sankt-Peterburg. Le Feuvre, Claire 1993 “The sound change e > o in the birchbark letters of Novgorod and T. Fenne’s manual, and the N.sg.m. ending -e”. Harvard Ukrainian Studies 17/3-4, 219250. Mürkhein, V. [Mjurkxejn] 1979 “O leksike s nepolnoglasiem v ‘Razgovornike’ Tonnisa Fenne”. In: S.M. Gluskina et al. (eds.), Pskovskie govory. Sbornik nauǁnyx trudov, 19-23. Leningrad. Mžel’skaja, O.S. Leksika obixodno-razgovornogo jazyka Moskovskoj Rusi XVI-XVII vv. (po 2003 dannym inostrannyx rukovodstv dlja izuǁenija russkogo jazyka). Sankt-Peterburg.

244 POS 1967Poulsen, Helge 1972

PEPIJN HENDRIKS AND JOS SCHAEKEN

Pskovskij oblastnoj slovar’ s istoriǁeskimi dannymi. Leningrad/Sankt-Peterburg.

Review of TF II. International Journal of Slavic Linguistics and Poetics 15, 212215. Prowatke, Christa 1985 “Gesprächsbücher des 17. Jahrhunderts und ihre sprachwissenschaftliche Auswertung. Ein Beitrag zur Schreibung des Niederdeutschen”. Beiträge zur Erforschung der deutschen Sprache 5, 66-79. Schaeken, Jos 1992 “Zum nordrussischen Nominativ Singular auf -e im Gesprächsbuch des Tönnies Fenne (Pskov 1607)”. In: A.A. Barentsen, B.M. Groen, R. Sprenger (eds.), Studies in Russian Linguistics (Studies in Slavic and General Linguistics 17), 285-293. Amsterdam-Atlanta. SRJa XI-XVII 1975Slovar’ russkogo jazyka XI-XVII vv. Moskva. Stone, Gerald 1990 “Thomas Schrove’s ‘Russisch Buch’: a Source for the Language of Muscovite Russia”. In: R. Lachmann et al. (eds.), Tgolí chole Mêstró. Gedenkschrift für Reinhold Olesch, 341-348. Köln-Wien. 1991 “PorjedǺenki k dotalnymaj wudaǀomaj A. Molleroweje zbLJrki lLJkarskich zelow z lLJta 1582”. LLJtopis Instituta za serbski ludospyt A/38, 19-29. (n.v.) TF I-IV 1961-1986 Tönnies Fenne’s Low German Manual of Spoken Russian. Pskov 1607. I: L.L. Hammerich, R. Jakobson et al. (eds.), Facsimile copy (1961). II: L.L. Hammerich, R. Jakobson (eds.), Transliteration and translation (1970). III: A.H. van den Baar (ed.), Russian-Low German Glossary (1985). IV: H.J. Gernentz (ed.), Mittelniederdeutsch-neuhochdeutsches Wörterbuch zum Russischniederdeutschen Gesprächsbuch (1986). Copenhagen. Whitehead, P.J.P. 1976 “The lost Berlin manuscripts”. Notes (series 2) 33/1, 7-15. 1980 “The Berlin manuscripts recovered”. Notes (series 2) 36/3, 773-776. Xoroškeviǁ, A.L. 2000 “K istorii sozdanija nemecko-russkix slovarej-razgovornikov Tomasa Šrove i Tennisa Fenna ”. Slavia orientalis 49/1, 77-91. Zaliznjak, A.A. 1986 “Novgorodskie berestjanye gramoty s lingvistiǁeskoj toǁki zrenija“. In: V.L. Janin, A.A. Zaliznjak (eds.), Novgorodskie gramoty na bereste (iz raskopok 1977-1983 gg.), 89-219. Mosvka. 1998 “Iz nabljudenij nad ‘Razgovornikom’ Fenne”. In: T.M. Nikolaeva (ed.), Ȑʏʌʔʓʑʏʐʏʍ. K 70-letiju Vladimira Nikolaeviǁa Toporova, 235-275. Moskva. 2004 Drevnenovgorodskij dialekt. Vtoroe izdanie, pererabotannoe s uǁetom materiala naxodok 1995-2003 gg. Moskva.

245

EARLY MODERN RUSSIAN-GERMAN PHRASEBOOKS APPENDIX: CONCORDANCE OF F, S, AND A

The concordance lists all the corresponding phrases in the phraseological sections of the manuscripts of F, S, and A. Compared to the references to corresponding phrases in the text edition of S (Fa’owski 1997), the present concordance contains a considerable number of corrections and additional correspondences. A concordance of textual correspondences between F and A was so far unavailable.37 References to F are based upon the electronic text edition by Hendriks and Schaeken (version 1.1, July 2008). References to S and A point to the editions Fa’owski 1997 and Fa’owski 1994 respectively. Page numbers in italics indicate a rough or partial textual correspondence between the manuscripts. To some extent the distinction made between exact and rough correspondences can only be arbitrary. Fenne

Schroue

Anonymous

Phraseology: Social (187-261) 187-188 189 9-11 (191 3-8, 274 1-5) 189 12-14 189 15-16 189 19-20 190 1 190 2 190 4 190 7 190 9 190 10 190 12 190 13 190 14 190 15 190 16 190 17 190 18 190 19

37

66r 12-15 / 99r 8-12 3r 19-20 / 99v 13 / 99v 20-21 / 100r 1 99v 15-16 99v 22 65r 13 / 99v 18 99r 19 99v 19 100r 15 100r 14 99v 7 99r 13 99r 15 99r 16 99r 14 99r 17 99r 18

59r 6-7+8

20v 17 20v 11 20v 13

20v 7 / 59r 3 27r 13-14 20v 8

20v 2 / 59r 4 20v 4 20v 5 20v 6

191 1-2 191 3-8 (189 9-11, 274 1-5) 192 5-7 193 4-5 194 1 195 21-22 196 1-2 196 18-19 197 1-9 198 10 201 3-6 201 7-8 201 9-10 201 11-12 201 13-14 201 15-16 201 17-18 201 19-20 201 21 201 22 202 1-3 202 6-8 202 11-12 202 13

65r 22-23 3r 19-20 / 99v 13 / 99v 20-21 101v 12-13 52v 19-20

59r 6-7

58v 11 27r 2021+16-17 27r 22-23 56r 8 104v 9-10 102v 10 100r 20-22 100r 23-24 100v 1-2 100v 3-5 100v 10-11 100v 12-13 101r 6-7 101r 8-9 101r 10 101v 1 101r 11-12 101r 13-14 101r 15-16 101r 17

56v 8 91v 9-10 90r 9

We have not systematically searched for exclusive correspondences between S and A only.

246 202 14-15 202 16-17 202 18-20 202 21-22 203 1-2 203 3 203 4-5 203 6-7 203 8 203 9 203 10-11 203 12 203 13-15 203 16-17 203 18 203 19-20 204 1-2 204 3-4 204 5-6 204 7 204 8-11 204 12 204 13 204 14-15 204 16-17 204 18 204 19 204 20 205 1 205 2 205 5 205 6-7 205 8 205 9 205 11 205 12 205 13 205 14-15 205 16-17 206 1-2 206 7

PEPIJN HENDRIKS AND JOS SCHAEKEN 101r 18 101r 19-20 101r 21 101r 22-23 101v 2-3 101v 4 101v 5-6 101v 9-10 101v 8 103v 3-4 101v 14-15 101v 17 101v 18–102r 2 102r 5-6 102r 7 64r 1-2 / 102r 8-9 102r 19 102r 20+2122 / 102v 6 102v 1-2 57r 23 / 102v 5 102v 7-9 102v 13 102v 14-15 102v 20 103r 1 103r 2 103r 4 103r 5 103r 6 103r 7 103r 10 103r 8-9 103r 13 103r 14 57r 22 / 103r 18 103r 15 103r 19-20 103v 18-19 103v 11-12 104r 6-7 108r 1

77r 7-8

33r 16-19

56v 6-7

58r 7 57v 10-11 57v 12

58r 4

86v 2-3

206 13-14 206 15-16 206 19 206 22 207 6 207 7 207 8 207 9 207 10 207 11-12 207 13-14 207 15 207 16 207 19-20 207 21-22 208 1-2 208 3-4 208 5-6 208 7-8 208 9-10 208 11-12 208 13-15 208 16-20 208 21 209 1 209 3-4 209 5-6 209 8 209 9 209 10-11 209 12-13 209 15-16 209 17 209 18-19 209 20-22 210 1 210 2-3 210 4-5 – 210 6-8 210 9-10 – 210 11-12

104r 8 104r 3 104r 1-2 103v 9-10 104r 4-5 103v 5 103v 6-7 104r 9 104r 10-11 104r 12-13 104r 14-15 104r 17 104v 2 104v 11-12 104v 14-15 104v 18-19 103v 1-2 105r 5-6 107r 9 105r 1-3 105r 7-9 105r 12-14 105r 15-19 105r 20-21 105r 4 105v 4-7 105v 4-7 48v 12-13 / 105v 8 105v 9 105v 10-11 106r 4-5 106r 1-2 106r 3 106r 6-7 105v 17-19 106r 14-15 106r 16-17 106v 1-2 106v 3-4 106v 7-8 106v 9 107r 1-2 / 107r 3-4

76v 3

54r 6 54v 1 54v 2 54v 5 54v 6-7 54v 11 83v 9 / 87r 9

247

EARLY MODERN RUSSIAN-GERMAN PHRASEBOOKS 210 13-14 210 17-18 210 19-20 210 21-22 211 1-2 211 3-4 211 5-6 211 7 211 8 211 9 211 10-11 211 12-13 211 14 211 15 211 16-17 211 18 212 1-2 213 19-20 214 17-18 (472 8-11) 214 21 215 5 215 17-18 216 10-11 216 16-18 217 15-17 (277 10-13) 219 4-7 (332 1-4) 219 11-12 (390 21-23) 220 8 223 1-4 223 5-9 223 10-13 223 14-16 223 17-21 223 22-24 224 1-5 224 6-8 224 9-12 224 13-16 224 17-20

107r 5-6 107r 7-8 107v 5-6 107r 10-11 107r 14-15 107r 16-17 / 107v 3-4 107v 7-8 107v 9-10 107v 11 107v 15 107v 16-17 108r 6-7 108r 16 65v 9-10 / 108r 8 108r 11-12 108r 13

55r 1

86v 11-12 67v 12-14 113r 1-4 84v 11-12 76v 8-9 57v 21-22 14v 1-3 27v 7-8 / 100r 11 5r 7-8

84v 5-6

27r 19-21

56r 1

52v 3-4 34r 1 4v 26–5r 2 5v 24–6r 2 9r 9-12 10r 22-25 23r 20–23v 3 27r 3-6 26v 18-22 27v 1-2 27v 13-16 28r 14-17 28r 18-21

40v 20-21

224 21-22 225 1-4 225 5-8 225 11-12 225 13-17 (348 16-20) 225 18-19 225 20-21 226 1-4 226 12-15 226 19-21 226 22-23 227 1-3 227 4-7 227 8-10 227 11-13 227 14-17 227 18-22 228 1-4 228 5-9 228 12-15 228 16-17 228 21-23 229 9-12 229 13-14 229 15 229 17-18 229 19-20 (472 3-4) 229 21-23 230 1-2 230 3-4 230 5-6 230 7-10 230 11-14 230 15-17 232 4-6 232 7-8 233 15-18 234 10-12 235 22-23 236 5-7 236 10-11 236 12-13 236 19-20

28v 16-17 28v 3-5 31v 5-8 32r 13-14 34v 20-22 35r 1-2 35r 3-4 35v 15-16 41r 16-19 42v 14-16 52r 17 43v 17-19 43v 20-23 44r 1-3 48v 14-16 48r 13-16 43r 18-21 56v 16-18 57r 5-9 52r 14-16 52v 1-2 55r 21-23 57r 10-12 57r 13-14 57r 19 57v 11-12 59v 1-2 / 106v 16-17 60v 6-8 64r 8-9 60v 20-21 64r 6-7 64v 1-4 111v 18-21 112r 6-8

86v 8

58v 16-17 103v 8 35r 7-9 75r 1-2 54r 8 86r 8 54r 1 54r 2 106r 8-9

248 237 22-23 238 5-7 239 1-4 239 5-6 239 19-20 240 6-10 245 21 251 9-13 252 11-13 252 23-24 256 16-17 257 1-4 (331 5-7) 257 5-7 257 8-9 257 10 258 1-9 258 10-11 258 15-17 259 4-7 259 11 259 12 259 13 259 14

PEPIJN HENDRIKS AND JOS SCHAEKEN 106v 5-6

275 1-3

3v 8-9

275 4-6 275 11-12 275 13-14 275 15-21 276 1-6 276 7-11 276 16-22

3v 10-11 3v 12-13 3v 14-15 3v 18-22 3v 23–4r 4 4r 5-8 4r 13-16

18v 16-17+18

– 277 1-4

4r 17-20 4r 21-24

277 10-13 (217 15-17) 277 14 277 15-16 277 17-21 278 3-6 278 7-10 278 11-14 278 15-18

5r 7-8

94r 4-5 94r 8 43v 1819+20-22 27r 22-23 91r 13-14 33r 22-25

92v 17-19 86v 14-15 55r 5 100r 12 91v 15-16 7v 9 3r 11-16 101r 4-5 65v 16-17

91r 15-16

27r 7-8 104v 16-17

99v 2 20v 19-20 21v 7 20v 19-20 / 21v 3 8r 9 / 21v 8 / 77v 16-17 60v 14

259 15 261 18

Phraseology: Trading (273-464) 273 2-16 274 1-5 (189 9-11, 191 3-8) 274 6-8 274 9-10 274 11-12 274 17-18 274 19-22

3r 3-10 3r 19-20 / 99v 13 / 99v 20-21 3r 23-24 / 100r 6-7 3v 1-2

3v 3-4 3v 5-7 / 100r 3

59r 1-2 59r 6-7

59r 9 59r 11 / 59r 13-14 59r 15 59r 16-17

278 19-22 278 23-24 279 1-4 279 5-8 279 9-12 279 15-18 279 19-23 280 1-5 280 6-10 280 11-15 280 16-20 280 21-23 281 1-5 281 6-10 281 11-15 281 16-17 281 18-21 282 1-3 282 6-9 282 10-14 283 1-11 283 14-20

4v 10-11 4v 12-13 4v 14-17 4v 18-21 4v 22-25 5r 3-6 5r 9-12

8v 6-7+8 / 59r 18 59v 1 59v 2-3 / 8r 5 59v 4-8 59v 9-14 8v 3-5 / 59v 15–60r 2 60r 3-5 6or 6-9 / 62r 11-14 84v 5-6

63r 10-11 / 93r 1-3

5r 17-20 5r 21-24 5v 1-2 5v 3-6 5v 7-10 5v 11-18 5v 11-18 5v 19-23 6r 3-10 6r 3-10 6r 11-13 6r 14-23 6r 14-23

75r 18-19 71r 1-2

75r 16-17 + 75v 1-2 18v 25-26 74v 5-8 6r 24-26 6v 1-4 6v 5-8 6v 9-18 7r 3-8

63r 16-17

249

EARLY MODERN RUSSIAN-GERMAN PHRASEBOOKS 284 1-6 284 7-11 284 12-14 284 15-18 284 19-25 285 1-5 285 6-10 285 11-12 285 13-14 (322 13-14) 285 15-17 285 18-22 286 1-7 (312 1-8) 286 8-14 286 15-20 286 21-23 287 1-5 287 6-9 287 10-17 287 18-22 288 1-7 288 8-11 288 12-15 288 16-20 (417 4-7) 288 21-23 289 1-5 289 6-14 289 15-18 289 19-22 290 1-2 290 3-8 290 9-11 290 12-15 290 16-20 291 1-4 291 5-10 291 11-16 291 17-20 291 21-22 292 5-8 292 9-13 292 14-18 292 19-23

6v 23–7r 2 6v 19-22 7r 9-11 7r 12-15 7r 16-21 7r 22-25 7v 1-4 7v 5-6 7v 7-10 7v 7-10 8r 1-4 7v 11-16 / 19r 26–19v 4 7v 20-25 8r 5-9 10v 5-8 8r 10-13 8r 14-17 8r 18-23 9r 5-8 8v 1-7 8v 8-11 8v 17-18 9r 13-16 / 109r 11-14

82r 2

11r 14-20 / 20v 25–21r 6

294 11-18 294 19-22 295 1-5 295 6-11

11r 24–11v 3 11v 4-7 11v 12-16 11v 17-21

295 12-16

11v 8-11

84r 4-5 64r 3-4+5

295 17-22 296 1-3

11v 22-26 12r 1-3

296 4-7 296 8-11

12r 4-7 12r 8-11

296 12-13

12r 12-13

296 14-19 296 20-21 (318 17-18) 297 1-6

12r 14-17 12v 1-2

297 7-10

12r 24-27

297 11-15 297 16-22 298 1-4 298 5-8 298 9-13 298 14-19 298 20-23 299 1-4 299 5-10

12v 7-12 12v 13-16 12v 17-20 12v 21-24 12v 25–13r 5 13r 6-9 13r 10-13 13r 14-17

299 11-15 299 16-18 299 19-22 300 1-15 300 16-20 300 21-22 301 1-6 (336 17-20)

13r 18-22 13r 23-25 13r 26–13v 2 13v 3-15 13v 16-20 14r 3-4 13v 21-25 / 29v 6-9

65v 7-8

86v 1

79r 8-9 63v 15-16

11r 21-23 75r 14-15 / 93r 4-5

82r 11-13

83v 8 9r 17-21 9v 3-10 9r 24–9v 2 10r 4-6 + 26v 14-15 9r 22-23 9v 20-25 9v 26–10r 3 10r 7-10 10r 11-14 8r 24-27 10r 15-21 10r 26–10v 4 10v 9-12 10v 13-14 10v 15-18 10v 19-22 10v 23–11r 3 11r 4-7

293 1-4 (316 15-22) 293 5-6 293 20-22 294 9-10

12r 18-23

89v 16-17 67r 8-9 62r 16–62v 2 / 67r 10-12 64v 1-2 / 66v 12-13 66v 10-11 8r 11 / 78r 7-8 61r 6-7 / 65v 12-13 61r 8-9 / 65v 16-17 61r 10-12 61v 1 / 90r 14 61r 13-14 / 67v 5-8 61r 15-16+1718 68v 9-15

68r 6-9 68r 10-13 62v 5-11 68r 14-16 68v 6-8 68r 19-20 + 68v 1-2

69r 8-11 67v 13–68r 5 69v 6-11 69v 13-15

250

PEPIJN HENDRIKS AND JOS SCHAEKEN

301 7-10

13v 26–14r 2

301 11-14

13v 26–14r 2 / 14r 5-8 14r 9-12 14r 21-24 14r 13-20 14r 25-28 / 54r 6-7 14v 4-7 14v 11-12 14v 13-16 14v 17-20 15r 3-6 15r 9-14 15r 15-16

301 15-18 301 19-22 302 1-11 302 12-16 302 17-21 303 1-3 303 4-6 303 7-10 303 11-14 303 15-20 303 21-22 304 1-4 304 5-8 304 9-13 304 14-17 304 18-21 305 1-5 – 305 6-8 (339 20-22) 305 9-11 305 12-15 305 16-20 306 1-5 306 6-10 306 11-15 306 16-21 307 1-5 307 6-9 307 10-14 307 15-18 307 19-22 308 1-6 308 7-12 308 13-15 308 16-17 308 18-21 309 1-4 309 5-12

69v 16-18 / 70r 3-4 70r 5-6 70r 7-8 70r 11-12 70r 9-10

70r 13-14

77v 8-9 16r 26-28 5r 13-16 16v 1-4 16v 5-8 16v 9-14 16v 15-16 / 31r 8-9 16v 17-20 / 31r 13-15 16v 21-22 17r 1-4 16v 27-30 17r 5-9 17r 10-13 17v 1-4 17v 9-12 17v 5-8 17v 13-16 17v 17-19 17v 20-23 17v 27–18r 2 18r 3-7 18r 8-9 18r 15-16 18r 19-22 18r 17-18 18r 23-28

72r 7-8 77v 13-14

72r 16-18 72r 11-15 73r 11-12

90v 18-19 / 92r 3-4

309 13-17

18v 1-4

309 18-21 310 1-8 310 9-11 310 12-15 310 16-19 310 20-23 311 1-6 311 7-9 311 10-11 311 12-16

315 16-21 316 1-8 316 9-12 316 13-14 316 15-22 (293 1-4) 317 1-6 317 7-12 317 13-16 317 17-21 318 1-5 318 6-11 318 12-13

18v 5-8 18v 9-14 18v 15-16 18v 17-20 18v 21-24 19r 9-12 19r 1-6 19r 13-15 19r 16-17 19r 18-21 / 52v 15-18 19r 22-23 19r 24-25 19r 26–19v 4 / 7v 11-16 19v 5-9 19v 10-13 20r 1-2 19v 14-17 19v 18-23 19v 24-28 20r 4-5 20r 6-8 20r 9-18 22r 1-6 20r 19-25 20r 26–20v 3 20v 10-14 20v 4-9 20v 15-18 20v 19-20 11r 14-20 / 20v 25–21r 6 21r 7-11 21r 16-20 21r 21-24 21v 1-4 21v 5-8 21v 9-14 21v 15-16

318 14-16 318 17-18

12v 1-2

311 17-20 311 21-22 312 1-8 (286 1-7) 312 9-13 312 14-18 312 19-21 313 1-5 313 6-12 313 13-16 313 17-19 313 20-22 314 1-12 314 13-19 315 1-8 315 9-15

73v 11-12 / 92r 5-6 78r 13 74r 9-12 74r 13-14 74r 15-16 74v 1-4 74v 16-18 74v 14-15 75r 1-2 75r 3-5 / 86r 1 75v 15-17 76r 1-2 82r 11-13

78v 5-6 78v 1-4

77r 17 77r 15+16 80r 11-16

78v 13-14 78v 15-18 64r 16-17 79r 8-9

79v 6-9

79v 18–80r 3 61r 15-16 / 67v 9-12 79v 16-17 61v 1 /

251

EARLY MODERN RUSSIAN-GERMAN PHRASEBOOKS (296 20-21) 318 19-22 319 9-12 319 13-14 319 15-20 320 1-4 320 5-8 320 9-12 320 13-14 320 15-16 320 17-18 320 21-23 321 1-2 (384 6+10) 321 3-5 321 6-7 321 8-11 321 12-16 321 17-21 322 1-7 322 8-12 322 13-14 (285 13-14) 322 15-19 322 20-23 323 1-4 323 5-7 323 8-12 323 13-15 323 16-21 324 1-3 324 4-7 324 8-16 324 17-21 325 1-8 325 9-12 325 13-16 325 17-19 325 20-21 326 1-7 326 8-12 326 13-16 326 17-22 327 1-4

90r 14 21v 17-20 21v 21-24 22r 11-12 22r 13-17 22r 18-21 22r 22-25 22r 7-10 22v 3-4 22v 14 23r 4-5 23r 10-11 23r 12-13 23r 16-17 23r 18-19 23v 4-7 23v 8-11 23v 12-15 23v 16-21 23v 22-25 7v 7-10 24r 5-8 24r 1-4 24r 11-14 24r 9-10 24r 15-18 24r 19-20 24r 21-24 22v 15-16 22v 17-19 22v 20–23r 3 24v 8-12 24r 25–24v 7 24v 15-18 24v 19-22 24v 23-24 25v 11-12 25r 1-6 25r 7-10 25r 11-13 25r 14-19 25r 20-21

80v 2-4 80v 9-10 80v 15-17 80r 17–80v 1 81r 5 81r 8-9 81r 10-11+1213 81v 1 81r 6-7 65r 13–65v 2 / 81v 12-15 81v 16-17 81v 18-19

82r 2

82r 3-4 82r 5-6 82r 7-8 67r 6-7 82r 9-10 82r 14-15

327 5-8 327 9-15 327 16-19 327 20-23

25r 22-25 25v 1-6 25v 7-10 25v 13-16

328 1-4 328 5-7 328 8-9 328 10-12 328 16-17 328 18-20 328 21-22 329 1-5 329 6-9 329 10-13 329 14-17 329 18-20 330 1-4 330 4-8 330 9-15 330 20-21 331 1-4 331 5-7 (257 1-4) 331 8-10 331 11-13 331 14-16 331 17-19 331 20-21 332 1-4 (219 4-7) 332 5-8 332 9-14 332 15-22 333 1-7

25v 17-20 25v 21-22 8v 12-13 8v 14-16 25v 23-24 26r 1-2 26r 3-4 26r 9-12 26r 13-16 26r 17-20 26r 21-23 26r 24-25 26v 1-3 26r 5-8 26v 8-13 26v 16-17 26v 23–27r 2 27r 7-8

82v 1-2

83r 1-2 82v 9-10

82v 11-12

333 8-11 333 14-18 333 19-20 334 1-5 334 6-9 334 10-14 334 15-21 335 1-4 335 5-9 335 10-14

27r 11-12 27r 13-14 27r 15-16 27r 17-18 27v 17-18 27r 19-21 27v 3-6 28r 22–28v 2 28v 6-13 28r 1-6 / 107v 1-2 28r 11-13 28r 7-10 29r 12-13 29r 14-17 29r 8-11 29r 4-7 29r 18-22 27v 19-22 27v 9-12 28v 22–29r 3

82v 13

83r 1-2 / 83r 3 83r 4 83r 5 83r 6-7 83r 8-9 83r 12-13 83r 14-15 83r 16 83v 4 83v 4 83v 6-7+8-9 83v 6-7+8-9 83v 6-7+8-9

56r 1

44r 13-14

86v 4+5

252 335 15-19 335 20-21 336 1-4 336 5-6 336 7-9 336 10-12 (439 13-15) 336 13-16 336 17-20 (301 1-6) 336 21-22 337 1-3 337 4-5 337 6-9 337 10-16 337 17-20 337 21-24 338 1-6 338 7-10 338 11-14 338 15-22 339 1-4 339 5-11 339 12-15 (349 18-23) 339 16-19 339 20-22 (305 6-8) 339 23-24 340 1-3 340 4-6 340 7-8 340 9-10 340 11-14 340 15-22 341 1-5 341 6-11 341 12-15 341 16-19 341 20-21 341 22-23 342 1-6 342 7-10 342 11-12

PEPIJN HENDRIKS AND JOS SCHAEKEN 28v 18-21 29v 19-20 29r 23–29v 2 29v 3-5 29v 14-15 29v 16-18 29v 21–30r 3 13v 21-25 / 29v 6-9 30r 24-25 29v 10-11 29v 12-13 30r 7-10 30r 11-16 15r 7-8 / 30r 17-19 30r 20-23 30v 1-4 30v 5-8 30v 9-12 30v 13-18 30v 19-22 30v 23–31r 3 31r 4-7 31r 10-12 16v 17-20 / 31r 13-15 31r 16 31r 17-18 31r 19-20 31r 21–31v 2 31v 3-4 31v 9-11 31v 12-19 31v 20-24 32r 1-5 32r 6-8 32r 15-18 32r 19-20 32v 5-6 32r 21-24 32v 1-4 32v 7-8

67r 1-2 63r 14-15 70r 9-10 69v 13-15

94v 1-5 94v 6-7 63v 3-4

62v 1415+16-17+20 63r 4-7 61v 7-8 72r 3-6

77v 13-14

77r 3-6

342 13-14 342 15-16 342 17-18 342 19-20 343 1-3 343 4-7 343 8-9 343 10-13 – 343 14-17 343 18-21 344 1-4 344 5-8 344 9-10 344 11-13 344 14-15 344 16-18 344 19-22 345 1-3 345 4-8 345 9-10 345 11-14 345 15-16 345 17-18 345 21-23 346 1-4 346 5-10 346 11-14 346 15-18 346 19-22 347 1-4 347 5-8 347 9-11 347 12-13 347 14-15 347 16-21 348 1-5 348 6-11 348 12-15 348 16-20 (225 13-17) 348 21-23 349 1-4 349 5-7

32v 9-10 32v 11-12 32v 13-14 32v 15-16 32v 17-18 32v 19-22 32v 23–33r 1 33r 2-5 33r 6-7 33r 8-10 33r 13-15 38r 4-7 38r 8-10 38r 11-12 38r 13-14 36r 3 33r 11-12 33r 16-19 / 36r 10-11 33r 20-22 33r 23–33v 3 33v 4-6 33v 7-9 33v 12-13 33v 10-11 33v 14-16 33v 17-19 33v 20-24 34r 2-5 34r 6-7 34r 8-10 34r 11-14 34r 15-17 34r 18-20 34r 23-24 34r 21-22 34v 1-5 34v 6-9 34v 10-15 34v 16-19 34v 20-22 34v 23-24 4v 1-2 / 35r 10-11 35r 12-13

77r 7-8

79v 1-2

253

EARLY MODERN RUSSIAN-GERMAN PHRASEBOOKS 349 8-11 349 14-17 349 18-23 (339 12-15) 350 1-5 350 6-12 350 13-18 350 19-22 351 1-4 351 5-8 351 9-12 351 13-16 351 17-18 351 19-22 352 1-4 352 5-9 352 10-17 352 18-21 352 22-23 353 1-6 353 7-11 353 12-13 353 14-19 353 20-22 354 1-4 354 5-8 354 9-10 354 11-17 354 18-22 355 1-8 355 9-12 355 13-17 355 18-22 356 1-5 (463 9-13) 356 6-11 356 12-15 356 16-22 356 23-24 357 1-5 357 6-9 357 10-14 357 15-18 357 19-21

35r 14-17 35r 18-21 31r 4-7 35r 22–35v 3 35v 9-14 35v 4-8 35v 17-18 35v 19-20 35v 21–36r 2 36r 4-6 36r 7-9 36r 12-13 36r 14-16 36r 17-18 36r 19-22 36v 4-9 36v 10-12 37r 19 36v 13-17 36v 18-21 36v 22-23 37r 1-6 37r 7-8 37r 9-12 37r 13-15 19r 7-8 / 37r 16-18 37r 22–37v 3 37v 4-7 37v 8-15 37v 16-18 37v 19-22 37v 23–38r 3 38r 18-21 38r 22–38v 3 39r 10-12 38v 4-9 39r 1-2 38v 12-15 38v 16-19 38v 20-23 39r 4-6 39r 7-9

72r 3-6 72v 10-11 / 80v 5-8

358 1-7 358 8-12 358 13-16 358 17-20 358 21-23

362 13-16 362 17-22 363 1-9 363 10-16 363 17-22

39r 16-21 39r 22–39v 3 39v 8-10 39r 13-15 40v 3-5 / 51v 6-8 39v 11-17 40r 1-5 40r 11-16 40r 17-23 40r 24–40v 2 40v 6-9 40v 10-17 40v 21-24 41r 1-2 41r 3-15 41r 3-15 41r 20-21 41v 1-4 41r 22-24 4v 6-9 / 41v 5-8 41v 9-11 41v 12-16 41v 18-25 42r 1-7 42r 14-18

364 1-4 364 5-7 364 8-10 364 11-14 364 15-20 364 21-23 365 1-5 365 6-9 365 10-14 365 15-23 366 1-5 366 6-10 366 11-14 366 15-19 366 20-23 367 1-6 367 7-9

42r 8-11 42r 12-13 42r 19-21 42r 22-25 42v 1-6 43r 16-17 42v 7-10 42v 17-20 42v 21-24 43r 1-8 43r 9-11 43r 12-15 43r 22-24 43v 1-6 44r 4-6 43v 7-11 43v 12-13

359 1-9 359 10-14 359 15-22 360 1-7 360 8-11

92v 9-10 92v 3-4

360 12-15 360 16-23 361 1-4 – 361 5-11 361 12-19 361 20-22 362 1-4 362 5-8 362 9-12

84r 11-12

85r 7-8

64r 14-15 / 72v 4-7

73r 3-4

89r 3

254 367 10-13 367 14-16 367 17-20 367 21-22 368 1-6 368 7-10 368 11-15 368 16-22 369 1-24 370 1-7 370 8-14 370 15-21 371 1-5 371 6-9 371 10-11 371 12-18 371 19-22 372 1-6 372 7-10 372 11-14 372 15-18 372 19-22 373 1-11 373 12-18 373 19-22 374 1-17 374 18-21 375 1-20 376 1-7 376 8-12 376 13-17 376 18-19 376 20-23 377 1-14 377 15-23 378 1-4 378 5-8 378 9-10 379 3-4 379 5-6 379 7-8 379 17-20 382 1-2

PEPIJN HENDRIKS AND JOS SCHAEKEN 43v 14-16 44r 7-8 44r 9-11 44r 22-23 44r 12-15 44r 18-21 45r 9-13 44v 1-6 44v 7-28 45r 1-7 45r 20-26 45v 1-6 45v 13-17 45r 14-19 45r 8 45v 18-22 45v 23-25 / 106v 13 46r 1-6 46r 7-10 46r 11-14 46r 15-16 46r 17-19 109r 15-23 109v 1-6 46r 22–46v 2 46v 3-15 109v 7-10 47r 22–47v 15 46v 22–47r 2 47r 3-6 47r 7-10 46r 20-21 47r 11-14 47v 16–48r 5 47r 15-21 48r 6-9 48r 10-12 48r 17-18

382 8-11 382 12-15 382 16-20 382 21 383 1-5 383 6-10

85r 10-13 87v 15-16

383 11-15 383 16-19 383 20-23 384 1-5 384 6+10 (321 1-2) 384 7-9 384 11-15 384 16-20 384 21-23 385 1-4 385 5-8 385 9-15 385 16-20 385 21-22 386 1-4 386 5-9 386 10-12 386 13-16 386 17-20 386 21-22 387 1-9 387 10-14 (436 4-7) 387 15-19 387 20-23 388 1-5

76r 7-8 76r 9 76r 11-12 48r 19-22 87v 12

388 6-9 388 10-13 388 14-16 388 17-21 389 10-14 389 15-19 389 20-21 390 1-4 390 5-8

48v 1-4 48v 5-7 48v 8-11 48v 8-11 48v 17-22 / 110r 1-4 48v 23–49r 2 / 59r 19-23 49r 3-6 49r 7-10 49r 11-14 49r 15-18 23r 12-13

81v 1

49r 20-21 49v 1-4 49v 9-12 49v 20-21 49v 13-17 49v 18-19 49v 22–50r 3 50r 4-7 52r 18 50r 8-10 50r 11-14 50r 15-16 50r 17-19 50r 20-22 53v 7-8 50v 1-7 50v 8-13

72r 9-10

51r 1-5 51r 1819+20-21 41v 17-18 / 52v 11-14 50v 14-17 50v 18-21 50v 22-23 51r 6-9 51r 14-17 51r 22–51v 2 51r 12-13 51v 3-5 51v 9-11

255

EARLY MODERN RUSSIAN-GERMAN PHRASEBOOKS 390 9-13 390 14-16 390 17-20 390 21-23 (219 11-12) 391 1-5 391 6-8 391 9-13 391 14-18 391 19-21 391 22-23 392 1-4 392 5-8 392 9-14 392 19-22 393 1-6 393 7-13 393 14-18 393 19-20 394 1-4 394 5-12 394 13-16 394 17-20 394 21-22 395 1-6 395 7-11 395 12-18 395 19-22 396 1-6 396 7-11 396 12-16 396 17-23 397 1-2 397 3-7 397 8-10 397 11-17 397 18-20 397 21-22 398 1-4 398 5-9 398 10-11 398 12-15 398 16-19

51v 12-14 52r 12-13 52r 19-22 52v 3-4 51v 15-18 51v 19-21 52r 1-4 52r 8-11 52r 5-7 52v 7-8 52v 5-6 52v 21–53r 2 53r 5-9 53r 10-11 53r 12-15 + 53r 18-19 53v 1-6 53v 9-12 53v 16-17 53v 18-20 53v 21–54r 5 54r 8-10 54r 11-13 55v 6 54r 14-18 54r 19-23 54v 3-8 54r 24–54v 2 54v 9-13 54v 14-17 54v 18-19 + 55r 1-2 54v 20-25 55r 3-4 55r 9-12 55r 7-8 / 106v 14-15 55r 13-18 55v 1-2 55r 19-20 55v 3-5 55v 7-10 55v 15-16 55v 17-19 55v 23–56r 2

398 20-23 399 1-3 399 4-8 399 9-12 399 13-17 399 18-21 400 1-8 400 9-11 400 12-15 400 16-19 400 20-22 401 1-5 401 6-8 401 9-15 401 16-18 401 19-22 93r 6-7 93v 6-7

67v 3-4 93v 10-11

84r 1-3

84r 13-14 84r 15 46v 3-4

402 1-5 402 8-10 402 11-14 402 15-21 403 1-6 403 7-13 403 14-16 403 17-20 404 1-5 404 6-11 404 12-14 404 15-19 404 20-22 405 1-3 405 4 405 5 405 6-9 405 10 405 11 405 12-15 405 16-19 405 20-21 406 1-4 406 5-8 406 9-12 406 13-15

56r 7-10 55v 20-22 / 107v 12-13 56r 3-6 56v 11-13 56r 17-21 56r 22-25 56v 1-6 57r 1-2 57r 3-4 57r 15-18 57v 6-7 57v 1-4 57v 13-14 57v 15-22 58r 1-3 58r 8-9 / 58r 10-13 58r 4-7 58r 14-15 58r 16-19 58r 20–58v 4 11r 8-13 / 58v 5-10 58v 14-20 58v 21-22 58v 23–59r 2 59r 3-7 59r 8-9+1014 59r 17-18 59v 3-8 59v 21-23 59v 9-11 59v 12-13 59v 14 59v 15-18 59v 19 59v 20 60r 1-4 60r 5-7 60v 3 60r 8-10 60r 11-14 60r 15-18 60r 19-22

84v 7-8

84v 9-10 84v 11-12

94r 13-14 94r 16-17 94r 15 44r 17

76r 5-6

256 406 16-20 406 21-22 407 1-2 407 3-8 407 9-12 407 13-22 408 1-2 408 3-9 408 10-18 408 19-22 409 1-11 409 12-22 410 1-7 410 8-11 410 12-16 410 17-20 411 1-9 411 10-15 411 16-22 412 1-5 412 6-9 – 412 10-14 412 15-18 412 19-20 413 1-4 413 5-8 414 11-12 414 13-14 414 15-18 414 19-22 415 1-4 415 5-7 415 8-9 415 10-14 415 17-20 415 21-24 416 1-5 416 6-9 416 10-14 416 15-19 416 20-23

PEPIJN HENDRIKS AND JOS SCHAEKEN 60r 23–60v 2 60v 4-5 60v 9-10 60v 11-16 60v 17-19 61r 5-13 61r 14-15 61r 16-23 61v 1-16 61r 1-4 61v 21–62r 5 62r 6-16 61v 1-16 61v 17-20 62r 17-22 62v 8-11 63r 8-16 62r 23–62v 3 62v 17-23 62v 12-16 63r 17-19 63r 20-24 63v 3-8 63r 3-5 63r 6-7 63v 15-18 63v 19-22 63v 9-10+1112 63v 23-24 64r 3-5 64r 18-21 64r 14-17 64r 22-23 64r 12-13 64v 11-15 64v 16-18 108r 17–108v 2 108v 3-7 108v 8-11 108v 12-15 108v 16-19 109r 3-6

85r 1-6 69r 1-7

69r 1-7

85v 19-22 85r 17-18 / 85v 1-2

85v 16+17+18 85v 11-13 87r 5-6 87r 11-13

417 1-3 417 4-7 (288 16-20) 417 8-12 417 13-16 417 17-20 417 21-24 418 1-4 418 5-8 (460 5-8) 418 9-13 418 14-17 418 18-22 419 1-4 419 5-9 419 10-13 419 14-17 419 18-19 420 1-5 420 6-7 420 8-13 420 14-17 420 18-21 420 22-23 421 1-4 421 5-9 421 10-12 421 13-14 421 15-19 421 20-22 422 1-4 422 5-10 422 11-15 423 1-3 424 15-17 425 14-18 425 19-21 426 21-22 427 11-14 428 10-16 434 9-12 436 4-7 (387 10-14) 438 5-7

109r 7-10 9r 13-16 / 109r 11-14 109v 11-15 109v 20-22 110r 5-8 110r 19-22 110r 9-11 53v 13-15 / 110v 3-6 110v 7-11 110v 12-15 110v 16-19 110r 14-18 110v 20–111r 2 111r 3-6 111r 18-21 79v 16-17 111r 22-23 111r 13-14 111v 3-7 111v 8-10 111v 15-17 112v 7-8 111v 22–112r 2 112r 9-12 111r 7-9 111r 10-12 112r 17-22 112v 5-6 112v 1-4 112v 13-18 112v 19-22 104v 4-5 64v 5-8 64v 9-10 101v 11

50v 8-13

88r 1 88r 4 76r 14-15 75v 9-10

81v 2-4 75r 8-9 70v 15-17 72r 9-10 70v 13-14

257

EARLY MODERN RUSSIAN-GERMAN PHRASEBOOKS 439 5 439 13-15 (336 10-12) 440 13-14 441 12-13 452 3-4 452 5-6 452 11-12 452 13-14 452 15-16 453 1-4 453 5-6 453 20-21 454 15-16 454 17-18 – 455 12-14 455 15-16 456 1-2 456 3-4 456 5-6 456 7-9 456 10-11 456 12-13 456 14-16 456 17-19 457 1-3 457 4-6 457 7-9 457 10-17 457 18-19 457 20-23 458 1-4 458 5-7 458 8-10 458 11-15 460 1-4

106r 12 29v 16-18

63r 14-15 66r 5-6

108r 3 15v 17-20 15v 17-20

53v 13-15 / 110v 3-6 38r 18-21 46v 16-21

Proverbs: Various (469-482) 72v 1-2 77v 10-11 77v 10-11 77v 12 77v 12

15v 7

15r 21-24 15r 28-29 15v 1-2 15v 3-4 15v 5-6 15v 8-9 15v 10-11 15v 12-13 15v 14 15v 15-16 15v 21-23 15v 24-26 16r 1-2 16r 3-4 16r 5-10 16r 11-12 16r 13-15 16r 16-18 16r 19-20 16r 21-23 16r 24-25 / 18r 10-14 38r 15-17

460 5-8 (418 5-8) 463 9-13 (356 1-5) 463 14-20

81r 14-15 81r 14-15 80r 4-7

469 9-11 469 14-17 471 5-7 471 8-12 471 13-17 471 18-21 472 3-4 (229 19-20) 472 5-7 472 8-11 (214 17-18) 472 12-13 474 12-13 476 6-8 476 9-12 476 13-14 476 15-16 478 1-2 478 3-5 478 11-14 478 15-16 478 17-18

78r 11-12 / 80v 20–81r 2

104v 7-8 112r 3-5 38v 10-11 39v 22-24 39v 4-7 52v 9-10 59v 1-2 / 106v 16-17 111r 15-17 113r 1-4 88r 13-14 88r 12 66r 1 / 66r 2-3 66r 4-6 7v 1-2 / 43v 13 7v 1-2 37r 20-21 40v 18-20 55r 5-6 55v 11-12 55v 13-14

83r 16 46v 1-2

258

PEPIJN HENDRIKS AND JOS SCHAEKEN SUMMARY

This contribution deals with the tradition of bilingual Early Modern Russian merchant phrasebooks, and the position of Tönnies Fenne (Pskov, 1607) within this tradition. Can the qualification of Tönnies Fenne as one of many successive compilers be united with that of the phrasebook as a priceless source of information for research into older stages of Russian? Based on a concordance of three genetically related phrasebooks linked to the NorthwestRussian town of Pskov, the textual organisation of Fenne’s phrasebook is examined. Close textual comparison of corresponding phrases enhances the impression that the phrasebook is so intimately related to the phrasebook of Thomas Schroue that a direct common source can be posited. Comparison of the data of these phrasebooks makes it possible to determine the attitude of the scribe to his sources and to find out the extent of his own input, clearing the view on the data. A philological approach is of vital importance for coming to a full appreciation of the language of the phrasebook of Tönnies Fenne as well as of the underlying linguistic system.

Dutch Contributions to the Fourteenth International Congress of Slavists, Ohrid: Linguistics (SSGL 34). Amsterdam – New York: Rodopi, 2008, 259-278.

THE EXPRESSION OF FUTURE TENSE IN THE PHRASEBOOK OF TÖNNIES FENNE (PSKOV, 1607): A PHILOLOGICAL APPROACH PEPIJN HENDRIKS

1. Introduction Elsewhere in this volume, Jos Schaeken and I discuss the composition and the language of three genetically closely related 16th- and 17th-century bilingual merchant phrasebooks linked to the Northwest-Russian town of Pskov: the phrasebook of Tönnies Fenne (1607; henceforth F) and the slightly older Einn Russisch Buch by Thomas Schroue (S) and Anonymous Ein Rusch Boeck … (A).1 That contribution presents various observations on the composition and the language of the phrasebooks and their common protograph. On the basis of close textual comparison we argue that F and S are so closely related that we can assume a direct common source for these two phrasebooks. This means that in order to ascertain and appreciate the linguistic system underlying the Russian material of F, the mainly linguistic approach taken to its data thus far has to be supplemented by a philological approach. The present contribution will illustrate how this philological approach can shed light on one of the structural linguistic differences between F and S: the expression of future tense. Textual comparison of F and S allows us to discard or at least reinterpret some of F’s data, thereby improving the reliability of a description and analysis of the language of F. 2. Comparing Tönnies Fenne and Thomas Schroue 2.1 General Tönnies Fenne did not go to Pskov to compile his phrasebook based on what he heard in the streets of Pskov, aided in his undertaking by one or more native informants. Although he may very well have spent time in Russia, may have

1

F is available as a four-volume edition (TF I-IV) and as an electronic text edition (Hendriks and Schaeken 2008b). Editions of S (Fa’owski and Witkowski 1992, Fa’owski 1997) and A (Fa’owski 1994) are likewise available.

260

PEPIJN HENDRIKS

learnt Russian whilst there and has compiled an extensive phrasebook, his phrasebook can hardly be called an original, independent work.2 Rather, Fenne copied most of the phrases from earlier sources. At least two thirds of the vocabulary of F can be proven to be non-original, and the same can be said for at least 56% of the phrases in the main phraseological sections of F (see Hendriks and Schaeken 2008b). 3 Any linguistic research into the language of F must come to terms with the historical dimension of the text. The present contribution will focus on the light that a philological approach to the data can shed on the most structural difference between F and S: the expression of future tense. When comparing corresponding phrases in F and S, one cannot help noticing that S abundantly uses the periphrastic construction ‘Ԗӹԓӫ + INF’ to express future-tense meaning, whilst F prefers synthetic, present-tense forms. The first scholar to draw attention to this fact was Anna Bolek: “Doǧǀ licznym w RS [= S – PH] formom analitycznym czasu przysz’ego typu chotLJti + bezokolicznik w RF [= F – PH] z regu’y odpowiadajƾ formy proste” (1997: 66).4 Two examples illustrate the difference between F and S in this respect: (1)

Ia suoi tovar prodam [...] Ich wÿll mÿne wahre vorkopen [...] ‘Ѥ ҡӊҸҮ ҲҸӊҽҝ ҰҝҸҦҽѸ’ ‘I will sell my goods’ Ja gotßu suoÿe thowar prodath [...] Ich will meine wahre vorkauffenn [...]

(2)

Ia sa sebe otvetzaiu [...] Ich wÿll vor mÿ suluest antworden [...] ‘Ѥ Ӊҽ ҡѦқѦ ҸҲӊҬҍҽӀ’ ‘I answer for myself’ Ja gotzu sa sabbÿ otweßath [...] Ich will mich vorantworten [...]

2

(F 280 21-22)5

(S 6r, 11-12) (F 290 9-10)

(S 9v, 26; 10r, 1)

Biographical data of Tönnies Fenne (or rather: Fonne) have been available since Jeannin 1973; see also Erpenbeck 1993, especially pages 556-559. 3 In the large subsection ‘Phraseology: trading’ (273-464), this percentage of non-original phrases rises to 74. 4 The question whether the verb ԖӹԽӥԽӸ was in fact a future-tense auxiliary or whether it retains a shade of its modal meaning will be discussed in section 3.2. I shall refer to the construction ‘Ԗӹԓӫ + INF’ as periphrastic future tense, if only for practical reasons. 5 All references to F point to the electronic text edition (Hendriks and Schaeken 2008a).

THE EXPRESSION OF FUTURE TENSE IN TÖNNIES FENNE (PSKOV, 1607)

261

In her article, Bolek did not further quantify this observation. To gain a better understanding of the phenomenon, this will have to be done first. 2.2 The data In order to be able to draw a good comparison between the two manuscripts, I have selected the data that can actually be compared. Using the concordance between F and S, I have sifted through F’s data to select those phrases which featured a future tense (either in Russian or in German) and which showed a direct correspondence with S. From the Russian data, I have most obviously selected present-tense forms of perfective verbs, which in 17th-century Russian – as in today’s – could have a future-tense meaning. The situation for the periphrastic future, however, is not as clear-cut, as the present-day construction ‘ԼӫԌӫ + INF’, is a relatively recent innovation in Russian (see section 3.4 below). Of the periphrastic forms we find in older stages of the language, we only find ‘Ԗӹԓӫ + INF’.6 By the second half of the 17th century, two new auxiliaries were in general use: ԙԽԂӰӫ ‘begin’ and ԼӫԌӫ ‘to be (future tense)’ (see Cocron 1961: 240), which we find several times. Phrases containing either of these three constructions have also been selected. For the German phrases, I have selected constructions with the verbs willen ‘want’, schöllen ‘shall, will’ and werden ‘become’, which could be used as auxiliaries to form an analytic future tense (see Sarauw 1924: 225; similarly Lasch 1914: 222f.). I have ignored cases where a present-tense form has future meaning, as there is no way to distinguish these cases morphologically, and the future-tense reading can only be inferred from the context. From all these cases, I have winnowed out correspondences between F and S which were less than certain. Finally, we are left with 538 cases where a future tense in F – either in Russian or German – has a direct correspondence in S. Table 1 shows how the 538 selected cases from F and S are distributed over the various constructions.

6

Other auxiliaries used to build the periphrastic future tense were ӠԐԂԐӸ (inf ӠԐӥԽӠ) ‘have to, must’, ӠԐӫ (INF ԋԽӠ), and various verbs on -ԓӰӫ (ӰԂԓӰӫ, ԗӹԓӰӫ, ӫԓӰӫ) ‘begin’ (see AndersѦn 2006b: 71).

262

PEPIJN HENDRIKS

synthetic periphrastic

other

present tense Ԗӹԓӫ ԙԽԂӰӫ ԼӫԌӫ DAT + INF various

total

F 4407 28 3 4 42 218 538

% 82% 5% 0.5% 0.5% 8% 4% 100%

S 344 147 2 4 19 22 538

% 64% 27% 0.5% 0.5% 4% 4% 100%

Table 1: corresponding future-tense constructions in F and S

These numbers indeed bear out Bolek’s observation: S uses ‘Ԗӹԓӫ + INF’ a lot more often than F. The table also shows which constructions F favours instead: a morphological present tense (both perfective and imperfective; see examples (1) and (2) above) and – to a lesser extent – the ‘DAT + INF’ construction, as the following example shows: (3)

Mnie stoboi ne beszeduvat [...] Ich will nicht mit dÿ tho doende hebben [...] ‘ҴӁҬ ҡ ҲҸқҸҮ ӁѦ қѦҡҬҦҸӊҽҲҧ’ ‘I will have nothing to do with you’ Ne gottsu Ja stobuÿ boßeduwath [...] Ich wil nicht mit dir zuthun haben [...]

(F 397 18-19)

(S 55v, 1-2)

In Hendriks and Schaeken 2008b, we have argued that structural differences between F and S have to be considered innovations on the part of the scribe of F. The data in our subcorpus can be used to test this hypothesis. A closer look reveals that the differences are indeed not random. There are no cases where F

7

This number includes present-tense forms of both perfective and imperfective verbs. In the latter case, the future tense can be found in the German phrase. 8 This number includes 3 cases in which ԋԊԙԋ (or ԋԊӾԙӸ, with the typical Novgorod NSg in -e), the past-tense form of ԋԽӠԙԋ, is combined with an infinitive and used with future-time reference (366 20, 314 15, 407 14). In all of F, there are 5 more cases of this construction (206 7, 237 22, 249 5, 249 8, 278 1), e.g.: [...] on ialsa komne buit. [...] he werdtt tho mÿ kamen. ‘ҫӁ ҾґҡҾ ҔҸ ѸӁҬ қӃҲҧ’ ‘He is to come to me’

(F 249 5-7)

From the same verb ԋԽӠԙԋ is the 2SG ӠԐӾԶӸԙԋ (206 3; not included in the 538 cases), which is combined with an infinitive to achieve future-tense meaning. Cf. also Issatschenko 1983: 388 on Slovak jal sa hovoriǬ ‘he began to speak’, which is reflexive as well.

THE EXPRESSION OF FUTURE TENSE IN TÖNNIES FENNE (PSKOV, 1607)

263

has Ԗӹԓӫ, but S does not. The number of cases for which the reverse holds true – S has Ԗӹԓӫ, but F does not – are shown in Table 2. Ԗӹԓӫ (S) present tense DAT + INF ԙԽԂӰӫ other total

no. of cases 89 22 1 7 119

Table 2: cases where S has Ԗӹԓӫ, but F does not

Compelling evidence, both direct and indirect, shows that the construction ‘Ԗӹԓӫ + INF’ in S was indeed replaced by various other constructions in F.9 2.3 Scribal corrections The direct evidence for the change of ‘Ԗӹԓӫ + INF’ into other constructions takes the form of scribal corrections in F. The following examples show that the protograph of F still contained the construction ‘Ԗӹԓӫ + INF’, which was adapted by the scribe: (4)

[...] da cho ÿch rosdelim [...] [...] vnd wÿlle(n)se van ander dehlen [...] ‘Ҧҽ ѽҸ ѹѽ ҝҸӉҦҬґѹѸ’ ‘we are going to divide them’

[...] da gottÿm muÿ gÿch roßdelÿth [...] [...] vnndt wollenn sie von einander theilenn [...]

9

(F 463 14, 18)

(S 46v, 16-17, 19-20)

Admittedly, a number of correspondences between F and S do not seem to fit the general pattern. In all of these, S combines Ԗӹԓӫ not with an infinitive but with another present-tense form. For example, S Ja […] gottzu dobbe dam ‘Ҿ ѽҸҍҏ ҲѦқҬ ҦҽѸ’ (S 101r, 15) corresponds with F Ia […] tebe dam (F 202 11). The other examples can be found in F 286 21 ~ S 10v, 5; F 290 12 ~ S 10r, 7 (2×); F 313 1 ~ S 19v, 14; F 402 15 ~ S 58r, 23.

264 (5)

PEPIJN HENDRIKS

[...] da ia ch i rad deszovo kuplu [...] [...] vnd ich wil gerne guden kop kopen [...] ‘Ҧҽ Ҿ ѽ ѹ ҝҽҦ ҦѦҗҺӊҸ ҔҏҰґӀ’ ‘and I like to buy cheap’

(F 411 11, 14-15)

[...] da Ja gottßu rade deßeuwo kupÿth [...] [...] vnnd ich wolte gerne gutten kauff kauffenn [...] (S 62r, 24-25; 62v, 2-3) In (4) we see that the scribe has struck out cho (of chotim) and continued with ÿch at the same place where S has the verb form gottÿm. Similarly, in (5) the scribe started to write chotzu, changed his mind, crossed it out, then started to write ia again, noticed his mistake, crossed it out, and proceeded to copy the rest of the phrase. Scribal corrections also provide us with indirect evidence of the same process: (6)

[...] ia stim torguvaiu. [...] mÿtt deme wolde ich gerne koepslagen, ‘Ҿ ҡ ҲҬѸ ҲҸҝҊҏӊҽӀ’ ‘I trade with him’

[...] Ja rade gotßu Isgÿm turguwath. p [...] Ich wolte gerne mit ihm kaufschlagenn. p. (7)

(F 276 8, 10-11)

(S 4r, 6, 8)

Mne ne 2 tebe 1 poklanitza, ia mnie stoboi tegatza. Ich wÿll mÿ nichtt vor dÿ nÿgen, ich wÿll mitt dÿ tho rechte gahen. (F 208 13-15) ‘ҴӁҬ ҲѦқҬ ӁѦ ҰҸҔґҸӁѹҲҧҡҾ, ѸӁҬ ҡ ҲҸқҸҮ ҲҾҊҽҲҧҡҾ.’ ‘I will not bow down before you, I will go to court with you’

THE EXPRESSION OF FUTURE TENSE IN TÖNNIES FENNE (PSKOV, 1607)

Ja gottsu stoboÿ tegasthe Ja ne gottsu tebbe poklanÿtze, Ich will mit dir rechnen, Ich will dir nicht vorgnugenn. p. (8)

Koli tÿ suoi saklat opæt vÿnet vÿnes: koli tÿ chotzis suoi saklat opæt vÿnet [...] Wanner wiltu dÿn pandt wedder inlosen [...] ‘ҘҸґѹ ҲӃ ҡӊҸҮ ӉҽҔґҽҦ ҸҰҾҲҧ ӊӃӁѦҗҧ: ҔҸґѹ ҲӃ ѽҸҍѦҗҧ ҡӊҸҮ ӉҽҔґҽҦ ҸҰҾҲҧ ӊӃӁҾҲҧ?’ ‘When [do /] will you redeem your pledge again?’

265

(S 105r, 12-14)

(F 294 11-12, 15)

Kallÿ thÿ gottzÿs ßuaÿe ßackladt opedt wuÿed polmo [...] Wan ehr wiltu dein pfandt wied[er] losen [...] (S 11r, 24; 11v, 1) In all three examples from F above, the corresponding phrase in S shows the construction ‘Ԗӹԓӫ + INF’. The scribal corrections are then most easily explained by assuming that the protograph of F still contained an infinitive, as part of that same construction. In (6), the auxiliary Ԗӹԓӫ was removed, but torguvaiu still shows traces of the infinitive: va was struck out, and u was emended from earlier o (from the infinitive, regularly spelt as torgovat). In (7) the order of the two synonymous expressions in F has been reversed compared to S, but we also see clear traces of the conversion from ‘Ԗӹԓӫ + INF’ to ‘DAT + INF’. In (8), F shows two synonymous constructions for S’s ‘Ԗӹԓӫ + INF’: one with a perfective present, the other with Ԗӹԓӫ. The crossed-out infinitive vÿnet ‘ӊӃӁҾҲҧ’ in the first construction could of course be interpreted as anticipating the infinitive in the next construction, but it could also reflect an infinitive in the original. 2.4 Mistakes In the examples from the previous section, the resulting Russian phrase is grammatically flawless. However, the elimination of ‘Ԗӹԓӫ + INF’ was not always that smooth. Of the 119 cases where S has Ԗӹԓӫ and F does not, F shows morphological or syntactic mistakes in 28 (or 24%), shown in Table 3.

266

PEPIJN HENDRIKS

Ԗӹԓӫ > present tense DAT + INF stanu other total

no. of cases 89 22 1 7 119

no. of mistakes 24 (27%) 2 (9%) 0 (0%) 2 (29%) 28 (24%)

Table 3: mistakes made during the conversion of ‘Ԗӹԓӫ + INF’

These mistakes can also be seen as indirect evidence of the conversion of ‘Ԗӹԓӫ + INF’. The following examples illustrate the kind of mistakes that were made: (9)

Ia ffperot ssuoim tovariszum, podumu [...] Ich wÿll mÿ ersten mÿt mÿnem maschoppe bedenken [...] (F 363 17, 20-21) ‘Ѥ ӊҰѦҝҺҦ ҡ ҡӊҸѹѸ ҲҸӊҽҝѹӂҺѸ ҰҸҦҏѸҏ’ ‘First I will consider with my companion’ Ja gotßu off piruo Ißuoÿm thowarÿßÿm podumath [...] Ich will mich erst mit meinem Mastkop bedennckenn [...] (S 42r, 14, 16-17)

(10) Ia isvostzikoff ponaimovaiu [...] Ich wÿll forlude huren [...] ‘Ѥ ѹӉӊҸӉҍѹҔҸӊ ҰҸӁҽѹѸҸӊҽӀ’ ‘I will hire carters’ Ja gotßu Ißuosmÿkoff ponaÿmowath [...] Ich will furleute Heurenn [...] (11)

Ti mnie to otnett : sapovedatt Wultu mÿ dat vorbeden. ‘ѷӃ ѸӁҬ ҲҸ ҸҲӁҾҲҧ: ӉҽҰҸӊҬҦҽҲҧ?’ ‘Will you deny: prohibit me that?’ Gottsÿs thÿ Imne tho Sapuwedath, p. Wiltu mir das vorbietenn. p.

(12) Ia suoi tovar prodam, da frutz dengi ffzæt. Ich wÿll mÿne wahre vorkopen vnd redtt geldtt nemen. Ja gotßu suoÿe thowar prodath da ffnuths kunÿ ffÿsedth Ich will meine wahre vorkauffenn vnd bahr geldt daruon nhemenn, p.

(F 288 8, 10)

(S 8v, 8-9) (F 201 9-10)

(S 100v, 1-2) (F 280 21-23)

(S 6r, 11-13)

In (9) and (10), ‘Ԗӹԓӫ + INF’ was meant to be replaced by a present-tense form. In (9) the infinitive ending -ԽӸ and the preceding vowel Ԃ were cut off from ԗӹԌӫԐԂԽӸ, and replaced by -ӫ, the prototypical ending for the 1Sg present tense.

THE EXPRESSION OF FUTURE TENSE IN TÖNNIES FENNE (PSKOV, 1607)

267

Example (10) is largely analogous to (6), except that this time, the verb in -ӹӴԂԽӸ was treated as a regular verb in -Ԃԇ, -ԂӾԶӸ. In (11) – analogous to the second part of (7) – we see that the conversion from ‘Ԗӹԓӫ + INF’ to ‘DAT + INF’ was not completed: the originally nominative subject remained unchanged. In (12), we see another incomplete conversion: the first verb construction (Ԗӹԓӫ ԗӢӹԌԂԽӸ > ԗӢӹԌԂԐ) went well, but in the second (Ԗӹԓӫ ӴԒԋԽӸ > ӴӹԒӸԐӫ) the infinitive was left untouched. When ‘Ԗӹԓӫ + INF’ was changed into a present-tense form, the mistakes are largely morphological in nature. 10 The morphological mistakes can be roughly divided into the following categories: •

•

•

•

•

reflexive verbs – mollitzu (191 1) (< ԐӹԊӠԽӸԙԋ) – vtzitzv (191 1) (< ӫԓӠԽӸԙԋ) – ismoluitzu (207 10) (< ԙԐӹԊӴӠԽӸԙԋ) – menetzis (296 4) (< ԐӾӰԋԽӸԙԋ) – salitzu (330 2) (< ԞԂԊӠԽӸԙԋ)11 verbs in -ӹӴԂԽӸ: – torgovais (3×: 276 16, 306 6, 384 11) (< ԽӹӢԄӹӴԂԽӸ) – ponaimovaiu (288 8) (< ԗӹӰԂӠԐӹӴԂԽӸ) – isorduvu (333 14) (< ӠԒӹӢӫԌӹӴԂԽӸ) – vÿdolguvu (385 2) (< ӴՂԌӹԊԄӹӴԂԽӸ) consonant alternation within present-tense conjugation – vbafflis (280 1) (< ӫԼԂӴӠԽӸ) – daszadu (401 1) (< ԂԙԂԌӠԽӸ) – naredu (406 5) (< ӰԂӢԋԌӠԽӸ) less straightforward relationship between present-tense and infinitive stem: – ԗӹԓӠӰώ/potziu (187 4/188 1) and ԗӹԓӠӰώ potzinu (273 2/13) (< ԗӹԓӠӰԂԽӸ) – poslas (325 20) (< ԗӹԙԊԂԽӸ) – prinu (337 11) (< ԗӢӠӰԋԽӸ) – sdais (409 14) (< ԞԌԂԽӸ) overgeneralisation of ԌԂԽӸ – rosgadam (205 14) (< ӢӹԒԄԂԌԂԽӸ)

10 The editors of TF II have actually noticed the conversion in a few cases. They considered the verb forms sdaisli (409 14), rosvestu (442 5), priroveneis (458 18), ԗӹԓӠӰώ (469 2), rosplodu (502 3), and blagoslovu (502 4), as well as the construction mnie tebæ rad sluszu ‘ѸӁҬ ҲѦқҾ ҝҽҦ ҡґҏҗҏ’ (399 5) to be blends of some kind. Since they had no access to S (or A), they could not know the origin of the contamination. 11 In these cases, the reflexive suffix -ԙԋ/-ԙӸ was not recognised. As a result, the conjugational ending did not substitute the infinitive ending, but instead deformed the reflexive suffix.

268 •

PEPIJN HENDRIKS

various – prirovenes (345 9) (< ԗӢӠӢӹӴӰԋԽӸ) – perenetzis (357 6) (< ԗӾӢӾӰԋԽӸ) – podumu (363 17) (< ԗӹԌӫԐԂԽӸ)

In case of conversion to the construction ‘DAT + INF’, the mistakes are more syntactic in nature. These mistakes are so few that categorisation is superfluous: • • • • •

Ti mnie to [...] sapovedatt ‘ѷӃ ѸӁҬ ҲҸ ӉҽҰҸӊҬҦҽҲҧ’ (201 9); Tott tzölloveck pÿr naresatt [...] ‘ѷҸҲ ҍҬґҸӊѦҔ Ұѹҝ ӁҽҝҾңҽҲҧ’ (202 1); Mne rad tvoi orudio isoruduvat ‘ҴӁҬ ҝҽҦ ҲӊҸ ҸҝҏҦҧҺ ѹӉҸҝҏҦҸӊҽҲҧ’ (211 3); Ia [...] frutz dengi ffzæt ‘Ѥ ӊҝҏҍҧ ҦѦӁҧҊѹ ӊӉҾҲҧ’ (280 21); Rad ia tebe dat [...] ‘ҩҽҦ Ҿ ҲѦқҬ ҦҽҲҧ’ (406 1).

Especially when replacing ‘Ԗӹԓӫ + INF’ with a present-tense form, the scribe was prone to overgeneralising the procedure of replacing the infinitive ending (with or without the preceding vowel) with the proper conjugational ending.12 3. A linguistic look at the conversion 3.1 Ӯӹԓӫ fallen into disfavour The strange verb forms and constructions we have seen in section 2.4 show that the removal of ‘Ԗӹԓӫ + INF’ was largely an automatism on the side of the scribe who copied the material. Its mechanical nature is illustrated by the fact that even the religious and the highly formulaic parts flanking the main body of the text apparently did not escape conversion, as the following example shows: (13) ғҸѹѸҾ ҡӊҾҲҽҾ ҲҝҸѹ҃ҽ. λǬ ҰҸҍѹӁώ ӉҦҬ ҰѹҡҽҲ [...] Im namen der hilligen drefoldicheitt. wÿll ich hir anfangen tho schriuen [...] Woimæ svætaia troÿtza ias potzinu sdies piszat [...] ‘In the name of the Holy Trinity, I shall begin to write here’ Ißde Jagotzu Ißboßÿu pomotzÿ potßÿnath [...] Itzunndt will ich mit der Hilffe gottes beginnen [...] 12

(F 273 1-2, 6-7, 12-13)

(S 3r, 3, 7)13

Still, the scribe did not always proceed without careful consideration, given cases like omanivais (331 20) for S gottsÿs omanÿth (27v, 17), prodam (307 19) for S gottzu [...] prodath (17v, 20). 13 This correspondence is not completely literal.

THE EXPRESSION OF FUTURE TENSE IN TÖNNIES FENNE (PSKOV, 1607)

269

Here, cutting off -ԂԽӸ from ԗӹԓӠӰԂԽӸ and replacing it by -ӫ yielded the hybrid form ԗӹԓӠӰӫ, instead of either perfective ԗӹԓӰӫ or imperfective ԗӹԓӠӰԂԇ. Yet despite the automatic execution on the part of the scribe, in and of itself the conversion was a very conscious effort. For some reason or other, the ‘Ԗӹԓӫ + INF’ construction had fallen into disfavour compared to the earlier phrasebooks.14 3.2 Ӯӹԓӫ as an auxiliary Of course, one could ask oneself whether the verb Ԗӹԓӫ in combination with an infinitive is in fact an auxiliary verb used to form a future tense, or whether it has retained its lexical meaning ‘want, will’. Although ‘Ԗӹԓӫ + INF’ was used as a periphrastic future tense in Church Slavonic, the construction is considered alien to East Slavic (Andersen 2006b: 71). Whereas Bolek speaks about “form[y] analityczn[e] czasu przysz’ego typu chotLJti + bezokolicznik” (1997: 66) without reservation, Adam Fa’owski, in his analysis on the language of the Anonymous phrasebook, is more hesitant: “W tekǧcie omawianego zabytku tylko dwa z wymienionych wyǻej czasowników [i.e. ӠԐӥԽӠ, ԖӹԽӥԽӠ, ӰԂԓԂԽӠ (ӫԓԂԽӠ, ԗӹԓԂԽӠ), ԙԽԂԽӠ, ԼՂԽӠ – PH] pe’ni’y funkcjNj pomocniczƾ przy tworzeniu z’oǻonego czasu przysz’ego: STAT’ i – byǀ moǻe – CHOTET’.” (Fa’owski 1996: 61) Not only may Ԗӹԓӫ have retained its modal meaning, but also – and more importantly –, in all but a few cases the Russian construction ‘Ԗӹԓӫ + INF’ in S shows a suspicious correspondence with the German construction ‘willen + INF’. This could indicate that ‘Ԗӹԓӫ + INF’ is no more than a literal translation of the German construction. 15 Both Bolek and Fa’owski have in fact suggested that the abundant use of Ԗӹԓӫ may be influenced by the corresponding German construction: The ‘Ԗӹԓӫ + INF’ construction of S was not the only future-tense expression which was changed in F. In a few cases, our corpus shows traces of other changes in future-tense expressions: – mnie isvorotzus ‘ѸӁҬ ѹӉӊҸҝҸҍҏҡҧ’ (325 3) for S Ja Ißworatsuß (24v, 2); – Samli tÿ sa ihim iechas [...] ‘ҡҽѸ-ґѹ ҲӃ Ӊҽ ѹѸ Ҭѽҽҗҧ’ (328 1) for S Sammomull tobbÿ ßagÿm ÿochgath (25v, 17); – [...] tochdi mnie tebæ rad sluszu ‘ҲҸҊҦӃ ѸӁҬ ҲѦқҾ ҝҽҦ ҡґҏҗҏ’ (399 5) for S [...] thochdÿ Ja tebbe vade sslussÿth (56r, 4). 15 To make things even more complex: we cannot even be sure that the corresponding German construction, ‘willen + INF’, has no modal meaning (cf. Günther 1964: 61, Lunden 1972: 94). In this respect, it is worthy of note that despite the introduction of new constructions in F, the German phrases retain the verb willen. 14

270

PEPIJN HENDRIKS

“Byǀ moǻe na wysokƾ frekwencjNj konstrukcji analitycznych w RS wp’ynƾ’ wzorzec niemiecki i dlatego trudno mieǀ pewnoǧǀ, ǻe posiadajƾ one znaczenie czysto gramatyczne.” (Bolek 1997: 66-67) “Nieda siNj wykluczyǀ, ǻe o duǻej czNjstotliwoǧci po’ƾczeǝ rosyjskich typu chotet’ + infinitivus w KR [= A – PH] zadecydowa’ wzorzec niemiecki wollen + infinitivus.” (Fa’owski 1996: 62) Nevertheless, the construction ‘Ԗӹԓӫ + INF’ can hardly be called ungrammatical; after all, it is used in Russian to this day. If we accept that ‘Ԗӹԓӫ + INF’ was indeed a literal translation from German, the motivating factor behind its abandonment may have been that it was considered unidiomatic for Russian. This fact may have been pointed out to the compiler of F by a native informant.16 3.3 Verbal aspect The editors of TF II remarked “Fenne’s hesitant search for expression of future” in Russian, which should “undergo a critical examination” (TF II: XVII). It is not entirely clear what they mean by “hesitant search”, but it is interesting in this light to take a look at those cases where the Ԗӹԓӫ-type conversion resulted in a morphological present tense. Table 4 shows how the verbal aspect of 76 present-tense forms (out of a total of 89; see Table 3) in F correlates with that of the infinitive following Ԗӹԓӫ in S. These 76 forms constitute those cases where the infinitive in S could confidently be assigned as being perfective or imperfective.17 F PERF IMPERF unclear18 total

PERF (S) 41 7 3 51

imPERF (S) 1 20 4 25

Table 4: aspect of present-tense forms in F after conversion from ‘Ԗӹԓӫ + INF’

16

This does not mean that the scribe had no competence in Russian, and blindly recorded whatever he was told by his informants. On the contrary, the scribal corrections and the mistakes discussed in sections 2 and 3 show that the scribe knew Russian to a considerable extent. 17 I have excluded 13 cases. In 12 cases, the verb involved is ӪӫԗӠԽӸ; the other case concerns the verb ԙӪԂԒԂԽӸ. According to SRJa XI-XVII, these are biaspectual verbs (see vol. 8: 127f. and vol. 24: 165ff.). 18 This category includes 7 cases where the verb form in F is either wrong or so obviously converted that it is difficult to confidently decide whether it is perfective or imperfective. These are ԗӹԓӠӰώ/potziu (2×: 187 4/188 1, 273 2/13), rosgadam (205 14), poslas (325 20), podumu (363 17), perebiru (414 13) and rosbiru (414 13).

THE EXPRESSION OF FUTURE TENSE IN TÖNNIES FENNE (PSKOV, 1607)

271

A clear pattern emerges from this table. Typically, the aspect of the presenttense form in F follows that of the infinitive in S. If the infinitive in S is perfective, then so is the present tense in F, and vice versa, as illustrated by the following two examples. (14) Ia nimnuscha opotzinu. Ich wÿl ein weinig rowen ‘Ѥ ӁѦѸӁҸңҔҸ ҸҰҸҍѹӁҏ’ ‘I will rest a little’ Ja gottsu neÿmsofka opottsÿnet, Ich will ein wenig ruhenn. p.

(F 206 19)

(S 104r, 1-2)

(15) Ia stoboi torguiu [...] Ich wÿl [...] mÿtt dÿ kopslagen. ‘Ѥ ҡ ҲҸқҸҮ ҲҸҝҊҏӀ’ ‘I (will) trade with you’ Ja gotzu turguwath stoboÿou [...] Ich will mit dir kaufschlagenn [...]

(F 303 4-5)

(S 14v, 13, 15)

The pattern is most notably disturbed by 7 counterexamples where a perfective infinitive in S corresponds with an imperfective present tense in F. However, 5 of these concern one and the same verb: perfective ԗӢӹԌԂԽӸ in S, corresponding with the imperfective 2Sg ԗӢӹԌԂӾԶӸ in F. In cases such as (15), where an imperfective present tense corresponds with a clear future tense in German, one could indeed get the impression of a “hesitant search” for the expression of future tense, at least when verbal aspect is used for this purpose. However, looking at the presented 76 forms, we have to conclude that inconsistencies in the expression of future tense can to some extent be attributed to the abandonment of ‘Ԗӹԓӫ + INF’.19 3.4 Periphrastic future-tense constructions in the early 17th century Another unanswered question pertains to the extent to which the attested numbers of cases for the various constructions hold up to what we would expect, based on the relevant literature. Older stages of Russian knew a number of periphrastic future-tense constructions, combining an auxiliary with an infinitive. Andersen (2006b: 71) lists the following auxiliaries for Old Russian: Ԗӹԓӫ ‘will’, ӠԐԂԐӸ (inf ӠԐӥԽӠ) ‘have 19

Several facts testify to the fact that the compiler was not completely insensitive to matters of verbal aspect. If you take all the data into account, you find that attestations of the perfective verb ԙԽӹӢԄӹӴԂԽӸ ‘trade’ in F typically correspond with imperfective ԽӹӢԄӹӴԂԽӸ in S.

272

PEPIJN HENDRIKS

to, must’, ӠԐӫ (INF ԋԽӠ), and various verbs on -ԓӰӫ (ӰԂԓӰӫ, ԗӹԓӰӫ, ӫԓӰӫ) ‘begin’. 20 For the Novgorod birchbark corpus, he concludes that the periphrastic future in the Old Novgorod dialect was formed using the phasal verb ԗӹԓӰӫ (2006b: 76). 21 The most recent example of the ‘ԗӹԓӰӫ + INF’ construction that Andersen found in the birchbark corpus, dates from the end of the 14th or the beginning of the 15th century (letter 129).22 We have also seen that there were two new constructions, employing the auxiliaries ԼӫԌӫ and ԙԽԂӰӫ. We know that the ‘ԼӫԌӫ + inf’ construction spread across the Russian language area from west to east from the mid-16th century onwards: it is first attested in the mid-16th century (Cocron 1961: 249), and had replaced the older constructions by the time Lomonosov published his ԪӹԙԙӠӡԙӪԂԋ ԄӢԂԐԐԂԽӠӪԂ in 1755 (Kiparsky 1967: 235). 23 Modern dialects in the Pskov region use the ‘ԼӫԌӫ + INF’ construction, e.g. the dialect of Ostrovcy, near Gdov (Honselaar 2001: 159). For the second half of the 17th century, Friedrich Cocron (1961: 248) concludes that the spoken language of that time preferred the ‘ԙԽԂӰӫ + INF’ construction, whilst ‘ԼӫԌӫ + INF’ was essentially restricted to the written language, where it was gaining popularity.24 All in all, at the beginning of the 17th century the replacement of the older constructions with the newer ‘ԼӫԌӫ + INF’ and ‘ԙԽԂӰӫ + INF’ must have been a process in full swing. When we now look at our data, we must first of all conclude that by the mid-16th century, when S was compiled, ԗӹԓӰӫ was no longer used as an auxiliary for the periphrastic future tense. There is no attestation of it in our data, nor, in fact, a single attestation in all of F and S.

20

The present contribution deals with 17th-century Russian; these verbs are – slightly anachronistically – cited in their form after the loss of yers and nasal vowels. 21 Andersen argues that ӠԐԂԐӸ and Ԗӹԓӫ belonged to the (Church Slavonic) high code in Russian, and – even when encountered in the birchbark documents – can be assumed to retain their original, modal meaning (Andersen 2006b: 73). The verb ӠԐӫ is only attested once in the birchbark corpus (Andersen 2006b: 75), and had disappeared from the language well before the beginning of the 17th century (see Andersen 2006a: 27, Birnbaum 1958: 267). 22 See http://gramoty.ru, letter 129. 23 Its rise in Russian is usually attributed to Polish influence, which ties in nicely with the earliest attestations in the West of the Russian language area (see Issatschenko 1983: 389f.). It could also tie in well with the Polish influence on the Muscovite language in the 17th century, although this assumption does not answer all questions connected with the spread of the construction over present-day dialects all over the Russian language area (see Kuznecov 1959: 247). 24 A popular auxiliary in the chancery language in the 17th century was ӫԓӰӫ, but it seems to be confined largely to that variety (Cocron 1961: 239). It disappeared in the early 18th century (Kiparsky 1967: 234).

THE EXPRESSION OF FUTURE TENSE IN TÖNNIES FENNE (PSKOV, 1607)

273

Yet the new auxiliary ԼӫԌӫ is only attested marginally in both manuscripts. For all of F, there are only 4 reliable instances of this auxiliary. 25 Also, ‘Ԗӹԓӫ + INF’ was never changed into ‘ԼӫԌӫ + INF’. The same holds true for ԙԽԂӰӫ, the other new auxiliary: in all of F’s data, only 5 cases can be found, 26 of which 1 comes in the place of earlier Ԗӹԓӫ. This is at odds with the general rise of the auxiliaries ԼӫԌӫ and ԙԽԂӰӫ in the early 17th century, which raises the suspicion that the resulting constructions in F need not be representative of the spoken language. 4. Implications for F’s data In section 2, we have seen how the scribe of F replaced the construction ‘Ԗӹԓӫ + INF’ by a number of other constructions. We have also seen that this change in some cases led to mistakes on the part of the scribe. In the previous section we have taken a linguistic look at the conversion. In this section we will discuss the philological implications. Sometimes, individual odd verb forms in the text of F can be explained linguistically. An example of a linguistic explanation are the verb forms primetu (211 16) and naredu (406 5). Although they were transliterated in TF II as ԗӢӠԐӥӫ and ӰԂӢԋӫ, Zaliznjak (1998: 269, 273) suggests that analogical levelling may have resulted in the verb forms ԗӢӠԐӥԽԇ and ӰԂӢԋԌԇ, in which case F’s primetu and naredu could in fact be correct. However, the corresponding phrases in S (108r 11-12 and 60r 11-14) contain ‘Ԗӹԓӫ + INF’, which makes this explanation unlikely. Instead, we must conclude that primetu and naredu do not reliably reflect either ԗӢӠԐӥԓӫ/ӰԂӢԋԞӫ or ԗӢӠԐӥԽԇ/ӰԂӢԋԌԇ. Although an explanation of a linguistic nature should of course never be discarded offhand, a philological explanation is best able to explain the oddities as a group, and as such should take precedence. The pattern is so clear that it can even be extrapolated to cases where F shows a suspicious verb form without a direct correspondence in S. The criteria for this extrapolation are that the corresponding German phrase shows willen, and that the odd verb form can be explained by assuming the presence of the construction ‘Ԗӹԓӫ + INF’. In light of the above, some of the data of F II have to be reconsidered or even discarded. Below, I will mention several cases (including the ones adduced al-

25

3 are impersonal constructions (323 8, 334 10, 406 16), the other can be found in 300 3-4. 2 other cases have to be dismissed: in 302 3, bude can be considered a conditional conjunction (see SRJa XI-XVII, vol. 1: 344); in 416 6, ‘Ia ktebe budu posmotrit’ should be read as ‘Ѥ Ҕ ҲѦқҬ қҏҦҏ, ҰҸҡѸҸҲҝѹҲҧ’ (cf. the corresponding German, as well as comparable cases in 206 1, 206 5, 221 17, 341 17, 380 13 and 414 15). 26 Viz. in 202 1, 210 1, 352 2, 422 12 and 459 8.

274

PEPIJN HENDRIKS

ready in section 2.4), without pretending to be exhaustive. The proposed transliterations of F II have to be corrected for the following forms: • • • • • •

poslas (325 20): ԗӹԙԊԂԶӸ, not ԗӹԶԊԶӸ (< ԗӹԙԊԂԽӸ); isorduvu (333 14): ӠԒӹӢԌӫӴӫ, not ӠԒӹӢԌӫ (< ӠԒӹӢӫԌӹӴԂԽӸ); perenetzis (357 6): ԗӾӢӾӰԋԓӠԶӸ, not ԗӾӢӾӰԋԽӸ ԓӾԶӸ (< ԗӾӢӾӰԋԽӸ); vÿdolguvu (385 1): ӴՂԌӹԊԄӫӴӫ, not ӴՂԌӹԊԄӫ (< ӴՂԌӹԊԄӹӴԂԽӸ); rosbiru, perebiru (414 13): ӢӹԒԼӠӢӫ, ԗӾӢӾԼӠӢӫ, not ӢӹԒԼӾӢӫ, ԗӾӢӾԼӾӢӫ (< ӢӹԒԼӠӢԂԽӸ, ԗӾӢӾԼӠӢԂԽӸ); ԗӹԓӠӰώ (187 4 and 469 2), potzinu (273 13): ԗӹԓӠӰӫ, not ԗӹԓӰӫ (< ԗӹԓӠӰԂԽӸ).

For the following forms, the normalised transliteration is adequate, but the forms themselves must be considered uninformative: • • • • • • • • • • •

voiovu ‘ӊҸҺӊҏ’ (199 6); rosgadam ‘ҝҸӉҊҽҦҽѸ’ (205 14, 368 8);27 torgovais ‘ҲҸҝҊҸӊҽѦҗҧ’ (3×: 276 16, 306 6, 384 11); potzinas ‘ҰҸҍѹӁҽҗҧ’ (279 15); vbafflis ‘ҏқҽӊґѹҗҧ’ (280 1); ponaimovaiu ‘ҰҸӁҽѹѸҸӊҽӀ’ (288 8); iechas ‘Ҭѽҽҗҧ’ (328 1); prinu ‘ҰҝѹӁҏ’ (3×: 337 11, 337 17, 411 12);28 podumu ‘ҰҸҦҏѸҏ’ (363 17); prines ‘ҰҝѹӁѦҗҧ’ (408 5);29 sdais ‘ңҦҽѦҗҧ’ (409 12).30

For other forms, it is impossible to come to an adequate normalised transliteration for the conjugational ending and the phonemes immediately preceding it; they too must be left out of any analysis: • • • • • 27

mollitzu (191 1) (< ԐӹԊӠԽӸԙԋ); vtzitzv (191 1) (< ӫԓӠԽӸԙԋ); pairitzu (196 8) (< ԗԂӢӠԽӸԙԋ); ismoluitzu (207 10) (< ԙԐӹԊӴӠԽӸԙԋ); primetu (211 16) (< ԗӢӠԐӥԽӠԽӸ);

Cf. correct rosgadaiu, also in 368 8. Not entirely certain, since S has guttße [...] prÿedt (30r, 12) (without n) and correct primu (30r, 17) for prinu in 337 11 and 337 17, respectively. 29 Not entirely certain, since S has correct preÿmes (61r, 19). 30 Unconnected to all this, the verb form chotizs (349 9), now transliterated as ‘ѽҸҲѹҗҧ’, could also be a scribal error for more common chotzis ‘ѽҸҍѦҗҧ’. There is no other instance of zs for /š/ in F. 28

THE EXPRESSION OF FUTURE TENSE IN TÖNNIES FENNE (PSKOV, 1607)

• • • • • • • • • • • • • • • • • •

275

bositzu (218 1) (< ԼӹԞӠԽӸԙԋ); pomiritzu (258 19) (< ԗӹԐӠӢӠԽӸԙԋ); omenu (278 16) (< ӹԐӥӰӠԽӸ); menetzis (296 4) (< ԐӾӰԋԽӸԙԋ); salvatzu (330 1) (< ԞԂԊӴԂԽӸԙԋ); salitzu (330 2) (< ԞԂԊӠԽӸԙԋ); ponarovu (339 6) (< ԗӹӰԂӢӹӴӠԽӸ); prirovenes (345 9) (< ԗӢӠӢӹӴӰԋԽӸ);31 posnatzu (396 13) (< ԗӹԒӰԂԽӸԙԋ); posnavatzu (396 13) (< ԗӹԒӰԂӴԂԽӸԙԋ); daszadu (401 1) (< ԂԙԂԌӠԽӸ); naredu (406 5) (< ӰԂӢԋԌӠԽӸ); klemu (426 17) (< ӪԊӾԐӠԽӸ); savetzætzu (2×: 435 1, 435 5) (< ԒԂӴӥԓԂԽӸԙԋ); schavitzu/ԶԂӴӠԽՁԇ (435 1) (< ԶԂӴӠԽӸԙԋ); menetzu (456 20) (< ԐӾӰԋԽӸԙԋ); rosplodu (502 3) (< ӢӹԙԗԊӹԌӠԽӸ); blagoslovu (502 4) (< ԼԊԂԄӹԙԊӹӴӠԽӸ).

Looking at syntactic idiosyncrasies, the construction ia mnie podumaiu (199 19) could very well have originated from ia chotzu podumath, as a contamination of ia podumaiu and mnie podumath. Still, interference with the corresponding German ‘Ich will mÿ bedenken’ (199 20) cannot be excluded. 5. Conclusions The present contribution set out to show how a philological approach to F can lead to a better insight into the language system underlying the Russian data in the manuscript. The striking difference in the use of the periphrastic future of the type ‘Ԗӹԓӫ + INF’ was used to illustrate this point. This abundantly used construction in S corresponds with a number of other constructions in F. Direct and indirect evidence point towards the removal of this construction by the scribe of F (section 2). The subsequent linguistic discussion of this material in section 3 allows for a number of conclusions. First of all, we have seen that the abandonment of the periphrastic future tense ‘Ԗӹԓӫ + INF’ was both conscious and automatic in nature (section 3.1). In all likelihood, the change was motivated by the fact that the original construction was considered unidiomatic, rather than ungrammatical (section 3.2). In the change, the scribe favoured a morphological present tense. 31

Cf. also 458 18 and 455 6.

276

PEPIJN HENDRIKS

This verb form could be either imperfective or perfective, a choice which a closer look also shows to have been mostly automatic (section 3.3). The automatic and uniform nature of the conversion as a whole raises the question of how representative of the spoken language the new construction is. The fact that the new periphrastic future-tense constructions which were on the rise in the first half of the 17th century are both hardly attested in the data and not introduced by the conversion, confirm the validity of this question (section 3.4). These insights shed light on the history of the text and the input of the scribe of F, and allow us to identify a category of mistakes and to discard or at least reinterpret some of the data (section 4). This makes the forthcoming description and analysis of F more reliable on a number of levels: it is beneficial to the phonetic/phonological, morphological as well as the syntactic analysis. Only an approach of the phrasebook’s data which not only takes the data at face value, but also takes the genetic relationship and its consequences into account, can be truly fruitful. Leiden University REFERENCES Andersen, Henning 2006a “Periphrastic futures in Slavic. Divergence and convergence”. In: K. Eksell, Th. Vinter (eds.), Change in verbal systems. Issues on explanation, 9-45. Frankfurt am Main. 2006b “Future and future perfect in the Old Novgorod dialect”. Russian Linguistics 30, 71-88. Birnbaum, Henrik 1958 Untersuchungen zu den Zukunftsumschreibungen mit dem Infinitiv im Altkirchenslavischen. Ein Beitrag zur historischen Verbalsyntax des Slavischen (Acta Universitatis Stockholmiensis. Études de philologie Slave 6). Stockholm. Bolek, Anna 1997 “Czy Tönnies Fenne zas’uguje na miano pioniera slawistyki?”. In: A. Bolek et al. (eds.), S’owianie Wschodni. MiNjdzy jNjzykiem a kulturƾ. KsiNjga jubileuszowa dedykowana Profesorowi Wies’awowi Witkowskiemu w siedemdziesiƾtƾ rocznicNj urodzin, 63-67. Kraków. Cocron, Friedrich 1961 La langue russe dans la seconde moitié du XVIIe siècle (morphologie). [Wien]. Erpenbeck, Dirk 1993 “Die Kaufmannsfamilien Fonne aus Westfalen im Lübecker Rußlandhandel. Biographische Anmerkungen zum Schreiber des Pleskauer Gesprächsbuches von 1607, Tönnies Fonne”. Zeitschrift für Ostforschung 42, 548-562. Fa’owski, Adam 1994 Ein Rusch Boeck. Ein russisch-deutsches anonymes Wörter- und Gesprächsbuch aus dem XVI. Jahrhundert. Köln-Weimar-Wien.

THE EXPRESSION OF FUTURE TENSE IN TÖNNIES FENNE (PSKOV, 1607) 1996

277

“Ein Rusch Boeck …”. Rosyjsko-niemiecki anonimowy s’ownik i rozmówki z XVI wieku. Analiza jNjzykowa. Kraków. 1997 “Einn Russisch Buch” Thomasa Schrouego. S’ownik i rozmówki rosyjskoniemieckie z XVI wieku. CzNjǧǀ II. Transliteracja tekstu. Indeks wyrazów i form rosyjskich. Kraków. Fa’owski, Adam, and Wies’aw Witkowski 1992 “Einn Russisch Buch” by Thomas Schroue. The 16th-century Russian-German dictionary and phrase-book. Part one. Introduction. Photocopies. Cracow. Günther, Erika 1964 Zwei russische Gesprächsbücher aus dem 17. Jahrhundert (Dissertation Humboldt-Universität). Berlin. Hendriks, Pepijn, and Jos Schaeken 2008a Tönnies Fenne’s Low German Manual of Spoken Russian. Pskov 1607: An electronic text edition. http://website.leidenuniv.nl/~hendriksp1/fenne (version 1.1, July 2008). 2008b “On the composition and language of three Early Modern Russian-German phrasebooks”. This volume. Honselaar, Zep [Zep Xonselaar] 2001 Govor derevni Ostrovcy Pskovskoj oblasti (Studies in Slavic and General Linguistics 29). Amsterdam-Atlanta. Issatschenko, Alexander 1983 Geschichte der russischen Sprache. 2. Band: Das 17. und 18. Jahrhundert. Heidelberg. Jeannin, Pierre 1973 “Der Lübecker Tönnies Fonne – ein Pionier der Slawistik”. Hansische Geschichtsblätter 91, 50-53. Kiparsky, Valentin 1967 Russische historische Grammatik. Band II: Die Entwicklung des Formensystems. Heidelberg. Kuznecov, P.S. 1959 Oǁerki istoriǁeskoj morfologii russkogo jazyka. Moskva. Lasch, Agathe 1914 Mittelniederdeutsche Grammatik (Sammlung kurzer Grammatiken germanischer Dialekte IX). Halle a.S. Lunden, Siri Sverdrup 1972 The Trondheim Russian-German MS Vocabulary. A contribution to 17th-century Russian lexicography. Oslo-Bergen-Tromsö. Sarauw, Chr. 1924 Die Flexionen der mittelniederdeutschen Sprache (Niederdeutsche Forschungen II). København. SRJa XI-XVII 1975Slovar’ russkogo jazyka XI-XVII vv. Moskva.

278 TF I-IV 1961-1986

Zaliznjak, A.A. 1998

PEPIJN HENDRIKS

Tönnies Fenne’s Low German Manual of Spoken Russian. Pskov 1607. I: L.L. Hammerich, R. Jakobson et al. (eds.), Facsimile copy (1961). II: L.L. Hammerich, R. Jakobson (eds.), Transliteration and translation (1970). III: A.H. van den Baar (ed.), Russian-Low German Glossary (1985). IV: H.J. Gernentz (ed.), Mittelniederdeutsch-neuhochdeutsches Wörterbuch zum Russisch-niederdeutschen Gesprächsbuch (1986). Copenhagen. “Iz nabljudenij nad ‘Razgovornikom’ Fenne”. In: T.M. Nikolaeva (ed.), Ȑʏʌʔʓʑʏʐʏʍ. K 70-letiju Vladimira Nikolaeviǁa Toporova, 235-275. Moskva.

SUMMARY This contribution applies a philological approach to the text of the Russian-German merchant phrasebook of Tönnies Fenne (Pskov, 1607; F), as advocated elsewhere in this volume by Jos Schaeken and myself. It shows how this approach to one grammatical feature can fruitfully supplement the linguistic approach to the data that has been taken thus far. The feature under investigation is the periphrastic future tense: F does not show the same abundant use of ‘Ԗӹԓӫ + INF’ as the older, genetically related phrasebook of Thomas Schroue. Direct and indirect evidence point towards the removal of this construction by the scribe of F, probably for reasons of idiomaticity. In order to accomplish this removal, the scribe had to be competent in Russian to a certain degree. Still, new mistakes were introduced as a result of the automatic nature of this process and the scribe’s non-native fluency. The history of the text allows us to identify this category of mistakes, and to discard or at least reinterpret some of the data. This makes the forthcoming description and analysis of the language of F more reliable on a number of levels: it is beneficial to the phonetic/phonological, morphological as well as the syntactic analysis.

Dutch Contributions to the Fourteenth International Congress of Slavists, Ohrid: Linguistics (SSGL 34). Amsterdam – New York: Rodopi, 2008, 279-292.

RECIPROCAL MARKERS IN RUSSIAN WIM HONSELAAR

0. Introduction In Honselaar (forthcoming) I make a distinction between reciprocity as a feature of extra-linguistic states of affairs and reciprocity as (an aspect of) the meaning or the interpretation of linguistic forms. At the referential level reciprocity is defined by Knjazev as follows: “[…] ҡѹҲҏҽ҃ѹѹ ҡ ҦӊҏѸҾ ѹґѹ қҸґѦѦ ҏҍҽҡҲӁѹҔҽѸѹ, ҔҽңҦӃҮ ѹӉ ҔҸҲҸҝӃѽ ҾӊґҾѦҲҡҾ ҸҦӁҸӊҝѦѸѦӁӁҸ ѹ ҦѦҮҡҲӊҏӀӂѹѸ ҰҝѦҦѸѦҲҸѸ (ҡҏқӆѦҔҲҸѸ), ѹ ҰҝѦҦѸѦҲҸѸ, ҰҸҦӊѦҝҊҽӀӂѹѸҡҾ ҦѦҮҡҲӊѹӀ (ҸқӆѦҔҲҸѸ).” (ҘӁҾӉѦӊ: 1998, quoting ҭѦҗҔҸӊҡҔѹҮ) ‘[…] situations with two or more participants, each of which is the agent (subject) and at the same time the patient (object) in the action referred to.’

or as: “[…] ҲҽҔѹѦ ҡѹҲҏҽ҃ѹѹ ҡ ҦӊҏѸҾ ҏҍҽҡҲӁѹҔҽѸѹ, ү ѹ ҷ, ӊ ҔҸҲҸҝӃѽ ү ӁҽѽҸҦѹҲҡҾ ӊ ҲҸѸ ңѦ ҸҲӁҸҗѦӁѹѹ Ҕ ҷ, ҍҲҸ ѹ ҷ Ҕ ү.” (ҘӁҾӉѦӊ: 1998, quoting Lichtenberk) ‘[…] situations with two participants, A and B, in which A has the same relation with B as B with A.’

The latter definition is broader than the former in that the “symmetrical” relation between A and B is not restricted to participants which are mentioned in subject and object positions but to participants as such. Besides, the Knjazev/Lichtenberk definition does not make a difference between situations in which participants are equally actively involved in their contribution to the state of affairs and situations in which they are not. Referential definitions specify a necessary condition for a state of affairs. If a speaker wants to refer to a state of affairs using reciprocal forms, or, as Frajzyngier and Curl put it, using “reciprocal markers” (Frajzyngier and Curl 1999: vii), this state of affairs has to meet the conditions given in a referential definition. Given the fact that reciprocal markers do not presuppose that the relationship of reciprocity holds between the subject-participant and the objectparticipant but can hold between other participants as well, the Knjazev/Lichtenberk definition seems to be more adequate than the more restricted Knjazev/Peškovskij definition.

280

WIM HONSELAAR

As a linguistic (semantic) concept, reciprocity is (a part of) the meaning or the interpretation of words, word combinations or morphemes, specifying symmetry in the relation between participants. The reciprocal meaning may be implicit in the meaning of words, such as: ԙԗӹӢӠԽӸ ‘to discuss’, ԌӢӫԞӠԽӸ ‘to become friends’, ԼӾԙӾԌӹӴԂԽӸ ‘to have a conversation’, ԙԙӹӢӠԽӸ ‘to quarrel’, ԐӠӢӠԽӸ ‘to reconcile’, ԒӰԂӪӹԐӠԽӸ ‘to introduce to each other’ and ԙӹӾԌӠӰԋԽӸ ‘to join together’ (ҘӁҾӉѦӊ, 1998). These verbs refer to states of affairs in which two participants play identical roles. For example: (1)

ҶҲҽ ҲѦѽӁҸґҸҊѹҾ ҰҸӉӊҸґҾѦҲ ҡѦҊҸҦӁҾ ҡҸѦҦѹӁҾҲҧ ґӀқӃѦ ҏҡҲҽӁҸӊҔѹ ҦґҾ ҡӊҾӉѹ ӊ ѸѹҝѦ – ҔҸѸҰҧӀҲѦҝ, ҡҲҽ҃ѹҸӁҽҝӁӃҮ ѹґѹ ѸҸқѹґҧӁӃҮ ҲѦґѦӍҸӁ!. (Yandex) ‘This technology allows us to connect all kinds of communication equipment all over the world: computers, stationary and mobile telephones!.’

In Honselaar (forthcoming) I list additional implicit reciprocal words, such as ӴӢԂԞԌӹӴԂԽӸ ‘to be enemies’, ӢӹԌӰӠԽӸ ‘to make related’, ԙԼӠԽӸ ‘to push together’, ԙӹԙӾԌӠԽӸ ‘to neighbour’, etc. The implicit reciprocal verbs may, however, also be used to refer to states of affairs in which participants play less symmetrical roles, that is to say, situations in which one of the participants is presented as more prominent, more active, whereas the other participant is presented as having a less prominent, more passive role; in cognitive semantic terminology we could speak here of “figures” and “grounds”. For instance, in example (2) it is said that a balcony has been made part of the living room in order to create more room for a nursery: (2)

ҴѦҡҲҽ ҏ Ӂҽҡ ӁѦѸӁҸҊҸ ѹ ӁҽѸ ҰҝѹҗґҸҡҧ ҏҲѦҰґҾҲҧ қҽґҔҸӁ, ҡҸѦҦѹӁҾҲҧ ѦҊҸ ҡ ҔҸѸӁҽҲҸҮ ѹ ҦѦґҽҲҧ ҲҽѸ “ҦѦҲҡҔҏӀ”. (Yandex) ‘We did not have enough space; so we had to insulate the balcony, join it to the room and create a nursery there.’

In this example the verb ԙӹӾԌӠӰԋԽӸ ‘to join (to)’ is used without a clear suggestion of equivalence or symmetry of the balcony and the room; on the contrary, the speaker focuses on the balcony which has been insulated in order to become suitable as part of the room. The asymmetry is reflected in the syntactic makeup of the sentence: the ԼԂԊӪӹӰ ‘balcony’ has the status of direct object whereas ӪӹԐӰԂԽԂ ‘room’ is pushed away into a prepositional phrase with the preposition ԙ ‘with’. In many cases, however, conceptually reciprocal states of affairs may be referred to by words having a meaning that does not reflect any reciprocity; good examples are the reflexive verb ԞӾӰӠԽӸԙԋ and the non-reflexive verb ӴՂӡԽӠ ԒԂԐӫԞ. Both verbs mean ‘to get married’ but ԞӾӰӠԽӸԙԋ (lit. ‘to provide oneself with a wife’) is said of male persons and ӴՂӡԽӠ ԒԂԐӫԞ (lit. ‘to go out after one’s

RECIPROCAL MARKERS IN RUSSIAN

281

husband’) of females. If one wishes to refer to the situation of marrying with a verb that is neutral as to male or female perspective, ӴԙԽӫԗӠԽӸ Ӵ ԼӢԂӪ ‘to get married’ 1 can be used. We may conclude that referential reciprocity and semantic reciprocity are closely connected in the sense that referential reciprocity specifies a necessary condition for the linguistic expression of semantic reciprocity; on the other hand, speakers have the freedom to highlight a hierarchy between participants by using specific biased words (such as ԞӾӰӠԽӸԙԋ ‘to get married’ and ӴՂӡԽӠ ԒԂԐӫԞ ‘to get married’) or to use words which may or may not explicitly refer to reciprocal situations, thus leaving it to the hearer to figure out whether the situation referred to is symmetrical, non-symmetrical or somewhere between these two extremes. More explicit is the expression of reciprocity in one of the many standard interpretations of reflexive verbs in Russian, i.e. verbs with the reflexive suffix -ԙԋ. According to Knjazev the number of reflexive verbs with reciprocal interpretations is, however, very low; Israeli, quoting Knjazev and Nedjalkov, specifies their number as “about forty” (Israeli 1997: 74). In Honselaar (forthcoming) I show that the inventory of reciprocal reflexive verbs is considerably larger than Knjazev and Israeli suggest, numbering at least more than a hundred. Israeli (1997: 74) categorises the reciprocal reflexives into two groups according to the state of affairs they refer to: x

“naturally reciprocal events” (“ԼӢԂԽԂԽӸԙԋ ‘to fraternize’, ԞӾӰӠԖԂԽӸԙԋ ‘to court each other’, ӹԼӰӠԐԂԽӸԙԋ ‘to embrace’, ӴԙԽӢӾԓԂԽӸԙԋ ‘to meet’ (on purpose, not by accident), ӴӠԌӾԽӸԙԋ ‘to see each other, to meet’, ԙԙӹӢӠԽӸԙԋ ‘to quarrel’, ԐӠӢӠԽӸԙԋ ‘to reconcile’, ӪӹӰԽԂӪԽӠӢӹӴԂԽӸԙԋ ‘to get in touch with each other’, ӹԼԐӾӰӠӴԂԽӸԙԋ ‘to exchange’ and so on.”);

x

“reciprocal aggressive actions” (“ԼӹԌԂԽӸԙԋ ‘to butt each other’, ԊԋԄԂԽӸԙԋ ‘to kick each other’, ӪԊӾӴԂԽӸԙԋ ‘to peck each other’, ӪӫԙԂԽӸԙԋ ‘to bite each other’, ԼӢԂӰӠԽӸԙԋ ‘to argue’, ԄӢՂԒԽӸԙԋ ‘to bitterly argue’, ԼӢӹԙԂԽӸԙԋ to throw to each other’, ԼӢՂԒԄԂԽӸԙԋ ‘to splash on each other’, and so on.”).

Israeli claims that all of them are accommodated by her definition “Vsja [= reflexive verb] in A and B Vsja is reciprocal if the actions involved are performed by both A and B to one another, at the same time or in continuous alternation.” (Israeli 1997: 75). 1

According to the dictionary ՀԪӵ, the verb ԞӾӰӠԽӸԙԋ too can be used to refer to the process of marrying from a neutral perspective, but only under specific grammatical circumstances: the subject is plural and the verbal aspect is perfective. Besides, this meaning is marked as colloquial.

282

WIM HONSELAAR

If we take into account my inventory of reciprocal verbs, Israeli’s list has to be expanded by some additional categories, such as “actions of exchange” ((ӹԼ)ԐӾӰԋԽӸԙԋ ‘to exchange’) and “actions of separation” (ӢԂԒԔӾԖԂԽӸԙԋ ‘to separate’); all these categories are not, however, strictly separated. 2 Reciprocity is most explicitly expressed in the (complex) reciprocal pronouns ԌӢӫԄ ԌӢӫԄԂ ‘each other’, ӹԌӠӰ ԌӢӫԄӹԄӹ ‘one another’, ԐӾԞԌӫ ԙӹԼӹӡ ‘among each other’. In his article, Knjazev analyses subtle differences between these three complex pronouns and mentions additional shades of meaning which go beyond pure reciprocity. In this paper I will be dealing with an evaluation of Knjazev’s analysis and will try to elaborate on it. I will be dealing with ԌӢӫԄ ԌӢӫԄԂ ‘each other’ (section 1), ӹԌӠӰ ԌӢӫԄӹԄӹ ‘one another’ (section 2) and ԐӾԞԌӫ ԙӹԼӹӡ ‘among each other’ (section 3). In the last section I will present a tentative algorithm for the linguistic expression of referential reciprocity. 1. ҦҝҏҊ ҦҝҏҊҽ Knjazev mentions two restrictions on the normally practically unrestricted use of ԌӢӫԄ ԌӢӫԄԂ ‘each other’, without, however, trying to give an explanation for these restrictions. The first restriction refers to the fact that ԌӢӫԄ ԌӢӫԄԂ cannot co-occur with the majority of reciprocal reflexive verbs (‘ӊӉҽѹѸӁҸ-ӊҸӉӊҝҽҲӁӃѦ ҊґҽҊҸґӃ’, Knjazev 1998). The second restriction concerns the fact that ԌӢӫԄ ԌӢӫԄԂ cannot be used if the symmetrical participants differ in their surface realizations. 3 Restrictions on combinability with reciprocal reflexive verbs are illustrated in unacceptable examples such as (3a-f): (3a) *ҫӁѹ ҲҸґҔҽӀҲҡҾ ҦҝҏҊ ҡ ҦҝҏҊҸѸ. ‘They are pushing each other.’

2

The implicit reciprocal verbs may be classified in the same way as the reciprocal reflexive verbs, e.g. “naturally reciprocal events” (ӢӹԌӰӠԽӸ ‘to make related’), “reciprocal aggressive actions” (ӴӢԂԞԌӹӴԂԽӸ ‘to be enemies’). 3 Knjazev makes a distinction between ӹԌӰӹԽӠԗӰӹӾ ӹԅӹӢԐԊӾӰӠӾ ԂӪԽԂӰԽӹӴ ‘identical expression of actants’ and ӢԂԒӰӹԽӠԗӰӹӾ ӹԅӹӢԐԊӾӰӠӾ ԂӪԽԂӰԽӹӴ ‘non-identical expression of actants’. In ԜӰ ӰӾ ӢԂԒԊӠԓԂӾԽ ԒӹԊӹԽӹ Ӡ ԐӾԌӸ ‘He cannot discern gold and cupper’ the two symmetrical actants gold and cupper have the same surface form (both are direct objects in the accusative); in ԜӰ ӰӾ ӹԽԊӠԓԂӾԽ ԒӹԊӹԽӹ ӹԽ ԐӾԌӠ ‘He cannot discern gold from cupper’ the actant gold is the direct object in the accusative whereas the actant cupper shows up as a prepositional phrase with the preposition ӹԽ ‘from’ with the genitive. Knjazev’s claim is that ԌӢӫԄ ԌӢӫԄa cannot be used with verbs having ‘non-identical expression of actants’.

RECIPROCAL MARKERS IN RUSSIAN

283

(3b) *ҫӁѹ ҔҏҡҽӀҲҡҾ ҦҝҏҊ ҡ ҦҝҏҊҸѸ. ‘They are biting each other.’

(3c) *ҫӁѹ ҃ҽҝҽҰҽӀҲҡҾ ҦҝҏҊ ҡ ҦҝҏҊҸѸ. ‘They are scratching each other.’

(3d) *ҫӁѹ ҸқґѹӊҽӀҲҡҾ ҦҝҏҊ ҡ ҦҝҏҊҸѸ (ӊҸҦҸҮ). ‘They are pouring (water) over each other.’

(3e) *ҫӁѹ ҰѹӁҽӀҲҡҾ ҦҝҏҊ ҡ ҦҝҏҊҸѸ. ‘They are kicking each other.’

(3f) *ҫӁѹ қҸҦҽӀҲҡҾ ҦҝҏҊ ҡ ҦҝҏҊҸѸ. ‘They are butting each other.’

Combinations with a limited number of other reciprocal reflexive verbs are considered to be well-formed such as (4a-c): (4a) ҫӁѹ ҝҏҊҽӀҲҡҾ ҦҝҏҊ ҡ ҦҝҏҊҸѸ. ‘They are arguing with each other.’

(4b) ҫӁѹ ҃ѦґҏӀҲҡҾ ҦҝҏҊ ҡ ҦҝҏҊҸѸ. ‘They are kissing (each other).’

(4c) ҫӁѹ ҸқӁѹѸҽӀҲҡҾ ҦҝҏҊ ҡ ҦҝҏҊҸѸ. ‘They are embracing (each other).’

It must be said that Knjazev’s qualification “unacceptable” with respect to (3a-f) is not always completely justified. An internet search with Yandex revealed that examples of the combination of these reciprocal reflexive verbs with ԌӢӫԄ ԌӢӫԄԂ occur, e.g. with the verb ԼӹԌԂԽӸԙԋ ‘to butt each other’: (5a) ҫҦӁҽңҦӃ ѸӃ ҡѹҦѦґѹ ӊѸѦҡҲѦ ҡ ҭҝҸҝҸҔҸѸ [...] ѹ ӊҦҝҏҊ ҏӊѹҦҽґѹ, ҔҽҔ Ҧӊҽ ҔҸӉґҽ қҸҦҽӀҲҡҾ ҦҝҏҊ ҡ ҦҝҏҊҸѸ. (Yandex) ‘Once we were sitting somewhere with Prorok [...] and suddenly we saw how two goats were butting each other.’

(5b) ү ҰҸҍѦѸҏ ҔҸҝҸӊӃ ӁѦ қҸҦҽӀҲҡҾ ҦҝҏҊ ҡ ҦҝҏҊҸѸ, [...]? (Yandex) ‘And why do cows not butt each other, [...]?’

(5c) ғ ҝѦҦҔѹѦ ѸѹӁҏҲӃ ѸѦңҦҏ қѹҲӊҽѸѹ ҍҺҝӁӃѦ ҸҝҔѹ [...] ѹґѹ ңѦ қҸҦҽӀҲҡҾ ҦҝҏҊ ҡ ҦҝҏҊҸѸ, [...] (Yandex) ‘In rare minutes between battles black whales [...] or butt each other, [...]?’

(5d) ѩѦҮҡҲӊѹҲѦґҧӁҸ, ҡҽѸ҃Ӄ ҊҽӉѦґѦҮ, ӁҽҰҝѹѸѦҝ, [...] ҝҽӉ ҦѦҡҾҲҧ Ӂҽ ҦѦӁҧ қҸҦҽӀҲҡҾ ҦҝҏҊ ҡ ҦҝҏҊҸѸ. (Yandex) ‘Really, male gazelles for example [...] butt each other some ten times a day.’

(5e) ғ ҲҸ ӊҝѦѸҾ ҔҽҔ ҸґѦӁѹ қҸҦҽӀҲҡҾ ҦҝҏҊ ҡ ҦҝҏҊҸѸ Ӊҽ ҰҝҽӊҸ ҸқґҽҦҽӁѹҾ ҡҽѸҔҸҮ, ҏ ґӀҦѦҮ ҡҸӊҡѦѸ ҦҝҏҊѹѦ ҸҲӁҸҗѦӁѹҾ. (Yandex) ‘If deer butt each other for the right to possess a female, humans have other relations.’

284

WIM HONSELAAR

Besides, in its metaphorical meaning ‘to fight’ the verb ԼӹԌԂԽӸԙԋ occurs very often with ԌӢӫԄ ԌӢӫԄԂ: (6a) ҭҸҔҽ ҰҝѦӉѹҦѦӁҲ ѹ ҸҰҰҸӉѹ҃ѹҾ “қҸҦҽӀҲҡҾ” ҦҝҏҊ ҡ ҦҝҏҊҸѸ, ҸӁҽ ҝҽҡҲѦҲ. (Yandex) ‘While the president and the opposition are fighting, it grows.’

(6b) ҭҸ ѸҸѹѸ ӁҽқґӀҦѦӁѹҾѸ Daikin ѹ Fujitsu General ҰҝҽҔҲѹҍѦҡҔѹ ӁѦ қҸҦҽӀҲҡҾ ҦҝҏҊ ҡ ҦҝҏҊҸѸ. (Yandex) ‘As far as I have noticed, Daikin and Fujitsu General hardly fight.’

(6c) Ѩ ҰҏҡҲҧ ҲҝҽӁҡҰҸҝҲӁѹҔѹ қҸҦҽӀҲҡҾ ҦҝҏҊ ҡ ҦҝҏҊҸѸ ҔҲҸ ӊѹӁҸӊҽҲ ѹ ҰҸҍѦѸҏ. (Yandex) ‘And let the transporters fight on the question who is guilty and why.’

Many more examples can be found in the Amsterdam Corpus of Russian Texts in which reciprocal reflexive verbs co-occur with ԌӢӫԄ ԌӢӫԄԂ, for instance with verbs like the following: ԼӹӢӹԽӸԙԋ ‘to wrestle’ ӴԙԽӢӾԓԂԽӸԙԋ ‘to meet’ ԌӢԂԽӸԙԋ ‘to fight’ ԒԌӹӢӹӴԂԽӸԙԋ ‘to greet’ ӹԼԷԂԽӸԙԋ ‘to mix’ ԗӹӴӠԌԂԽӸԙԋ ‘to see one another’ ԙӹӢӾӴӰӹӴԂԽӸԙԋ ‘to compete’ ԙԽӹԊӪӰӫԽӸԙԋ ‘to collide’ ԙՁӾԗӠԽӸԙԋ ‘to be coupled’ ԓӹӪԂԽӸԙԋ ‘to clink glasses’ If we consider this rather confusing collection of data, it is striking that sentences with verbs that refer to situations in which two participants are active at the same moment or in which the participants perform the action mentioned together or simultaneously (instead of successively, at different moments) are generally more acceptable than sentences in which the actions referred to must necessarily be successive (as in the case of (3b) *ӪӫԙԂԽӸԙԋ ԌӢӫԄ ԙ ԌӢӫԄӹԐ ‘to bite each other’) or accidentally successive (as in the case of (3a) *ԽӹԊӪԂԽӸԙԋ ԌӢӫԄ ԙ ԌӢӫԄӹԐ ‘to push each other’). If we take into account Knjazev/Peškovskij’s referential definition of reciprocity, which specifies that we are dealing with ‘situations with two or more participants, each of which is the agent (subject) and at the same time the patient (object) in the action referred to’, it is likely that Knjazev’s severe judgment on the unacceptability of (3a-f) is strongly influenced by the phrase “at the same time”. Moreover he considers situations in which actions take place successively rather than simultaneously to be less prototypical

RECIPROCAL MARKERS IN RUSSIAN

285

reciprocal situations. If he had taken the second definition of referential reciprocity as a criterion, his judgment might have been considerably milder. Knjazev’s second restriction will not be discussed here since it seems to be completely justified and self-evident. 2. ҸҦѹӁ ҦҝҏҊҸҊҸ The combination ӹԌӠӰ ԌӢӫԄӹԄӹ ‘one another’ is said to be a synonym of ԌӢӫԄ ԌӢӫԄa but differs from it by the fact that the former is considerably less frequent than the latter. Moreover, ӹԌӠӰ ԌӢӫԄӹԄӹ requires a more specific referential situation than ԌӢӫԄ ԌӢӫԄa in the sense that it is used if reference is made to situations ‘that deviate to a greater or lesser extent from prototypical reciprocal situations.’ 4 Knjazev’s two examples are given here as (7) and (8); (7) refers to the successive striking of clocks in different locations while (8) refers to unrelated situations in which the participants separately swear that they will perform the same action with respect to the other (i.e. to destroy him). The idea of reciprocity comes up as a result of the conceptualisation of these events as a series: (7)

ҭҸ ӊҡѦѸ ҔҸѸӁҽҲҽѸ ҍҽҡӃ ҸҦӁѹ Ӊҽ ҦҝҏҊѹѸѹ ҰҝҸӉӊҸӁѹґѹ ҦӊѦӁҽҦ҃ҽҲҧ – ѹ ӊҡѦ ҏѸҸґҔґҸ ҸҰҾҲҧ. ‘In all the rooms the clocks one after the other struck twelve and then it became silent again.’

(8)

ҩҽӁҧҗѦ ҸӁѹ қӃґѹ ӊҡѦҊҦҽ ҡҸӀӉӁѹҔҽѸѹ, ҽ ҲѦҰѦҝҧ ҔҽҔҽҾ-ҲҸ ҍѦҝӁҽҾ ҔҸҗҔҽ ҰҝҸқѦңҽґҽ ѸѦңҦҏ ӁѹѸѹ, ѹ ҸӁѹ ҰҸҔґҾґѹҡҧ ҰҸҊҏқѹҲҧ ҸҦѹӁ ҦҝҏҊҸҊҸ. ‘In the past they always were mates but now a black cat seems to have run between them, and they swore to destroy each other.’

A search in the Amsterdam Corpus of Russian Texts has made it clear that Knjazev is right when he says that ӹԌӠӰ ԌӢӫԄӹԄӹ is less frequent than ԌӢӫԄ ԌӢӫԄa; their rate of occurrence is about 1:5. It is a matter of taste whether this rate justifies Knjazev’s qualification of a considerably lower frequency for ӹԌӠӰ ԌӢӫԄӹԄӹ. In any case, ӹԌӠӰ ԌӢӫԄӹԄӹ is not a rare reciprocal marker. As opposed to ԌӢӫԄ ԌӢӫԄa, ӹԌӠӰ ԌӢӫԄӹԄӹ is often used with a non-reciprocal meaning, e.g.:

4

“[...] ӊ ҲҸҮ ѹґѹ ѹӁҸҮ ѸѦҝѦ ҸҲҔґҸӁҾӀӂѹѽҡҾ ҸҲ ҰҝҸҲҸҲѹҰѹҍѦҡҔѹѽ ҝѦ҃ѹҰҝҸҔҽґҧӁӃѽ ҡѹҲҏҽ҃ѹҮ.” (ҘӁҾӉѦӊ, 1998)

286 (9)

WIM HONSELAAR

ѨӉӊѦҡҲѦӁ ҽӁѦҔҦҸҲ Ӂҽ ҙҲҏ ҲѦѸҏ: ӊҡҲҝѦҍҽӀҲҡҾ ҦӊҸѦ ҰҝѹҾҲѦґѦҮ Ӂҽ ҏґѹ҃Ѧ, ҸҦѹӁ ҦҝҏҊҸҊҸ ҡҰҝҽҗѹӊҽѦҲ: «ҋґҏҗҽҮ, ғҽӁҧ, ҲӃ ҲҽҔҸҮ ҡѹӁѹҮ – ӉҽѸѦҝӉ, ҍҲҸ ґѹ?» (ғѹҗӁѦӊҡҔҽҾ) ‘There is a well-known joke on the same topic: two friends meet in the street and one asks the other: ‘Say Vanya, you look so blue – are you frozen, or what?’ (Vishnevskaya)

My qualification “non-reciprocal” for the meaning of ӹԌӠӰ ԌӢӫԄӹԄӹ in this case may be a bit too strong; probably it is more adequate to speak of some kind of indeterminateness on the part of the speaker. He is non-committal about the actual division of roles: he does not specify whether he considers participant A or participant B to be the one who asks the question or the one who is addressed. This is appropriate in case of a joke about two fictional characters: who did what is totally irrelevant. I will call this usage-type the non-committal use of reciprocals. 5 In the case of (9) it is obvious that the passage refers to only one instance of asking the question. In other cases the action (or situation) referred to might occur more than once, or, in other words, the predicate might be applied to more than one situation. In (10), for example, there is at least a possibility of a multiple application of the predicate ‘not better than’; in (11-13) the number of applications depends on the number of referents of the subject: clearly two in (11) and possibly more than two in (12) and (13): (10) – ғӃѽҸҦѹҲ, ҸҦѹӁ ҦҝҏҊҸҊҸ ӁѦ ґҏҍҗѦ, – ҏҡѸѦѽӁҏґҡҾ ѷ. ‘– So, the one is not better than the other, – T. said laughing.’

(11)

[...] ҦӊҸѦ ҍҽҡҸӊӃѽ ѹҊҝҽґѹ ӊ ҔҽҔҏӀ-ҲҸ ѹҊҝҏ, ҡҸҡҲҸҾӊҗҏӀ ӊ ҲҸѸ, ҍҲҸ ҸҦѹӁ ҦҝҏҊҸҊҸ қѹґ ҦӊҏѸҾ Ұҽґҧ҃ҽѸѹ ҰҸ ґҽҦҸӁѹ. (ҐҸҊҸґҧ) ‘[...] two guards were playing a game that consisted of beating the other on the hand with two fingers.’ (Gogol)

(12) ѨӉӊѦҡҲӁҸ қӃґҸ, ҍҲҸ ҸӁѹ ӉҽѸѦҍҽҲѦґҧӁҸ ґӀқѹґѹ ҦҝҏҊ ҦҝҏҊҽ ѹ ҸҦӁҽ ҦҝҏҊҏӀ ҰҸҦҦѦҝңѹӊҽґѹ. (ѩҸҡҲҸѦӊҡҔѹҮ) ‘It was well known that they loved each other very much and supported each other.’ (Dostoevsky)

(13) ҋ ҰѦҝӊҸҊҸ ӊӉҊґҾҦҽ, ҡґҸӊҽ ѹ ҦӊѹңѦӁѹҾ ҸӁѹ ҰҸӁҾґѹ ѹ Ҹ҃ѦӁѹґѹ ҸҦӁҽ ҦҝҏҊҏӀ. (ѷҏҝҊѦӁѦӊ) ‘From the first look, word and gesture they understood and respected each other .’ (Turgenev)

5

In Dutch, the direct equivalent of ԌӢӫԄ ԌӢӫԄԂ, elkaar ‘each other’ has a non-committal use in cases as De twee schilderijtjes hingen onder elkaar aan de muur litt. ‘The two paintings hang under each other on the wall.’; in contrast, Russian ԌӢӫԄ ԌӢӫԄԂ does not seem to be used in noncommittal situations. (Honselaar, forthcoming)

RECIPROCAL MARKERS IN RUSSIAN

287

The meaning of ӹԌӠӰ ԌӢӫԄӹԄӹ can, then, be formulated as an instruction to the hearer to apply the predicate to the set of referents as often as is relevant and each application of the predicate is intended to refer to a separate situation. 3. ѸѦңҦҏ ҡҸқҸҮ The combination ԐӾԞԌӫ ԙӹԼӹӡ ‘among each other’ is characterized by Knjazev as a typically Russian (in contrast to the majority of Slavic languages) “ҡҽѸҸҡҲҸҾҲѦґҧӁҸѦ ӊҸӉӊҝҽҲӁҸѦ ѸѦҡҲҸѹѸѦӁѹѦ [...] ҦґҾ ӊӃҝҽңѦӁѹҾ ӊӉҽѹѸӁҸҡҲѹ” ‘independent reciprocal pronoun [...] expressing mutuality’. He regards ԐӾԞԌӫ ԙӹԼӹӡ as an almost perfect equivalent of ԌӢӫԄ ԌӢӫԄԂ; they may occur in similar contexts, as illustrated in (14) and (15): (14) ҋҏӂѦҡҲӊѹҲѦґҧӁӃѦ ҸқҸѹѽ ҲѹҰҸӊ ҡѽҸҦӁӃ ѸѦңҦҏ ҡҸқҸҮ ӁѦ ҲҸґҧҔҸ ҰҸ ѸҸҝӍҸґҸҊѹҍѦҡҔҸѸҏ ҸӍҸҝѸґѦӁѹӀ, ӁҸ ҲҽҔңѦ ѹ ҰҸ ҡѹӁҲҽҔҡѹҍѦҡҔҸѸҏ ӍҏӁҔ҃ѹҸӁѹҝҸӊҽӁѹӀ. (Ѩ. ҭѦҝѦґҧѸҏҲѦҝ) ‘Nouns of both types are similar not only because of their morphological form, but also because of their syntactic functioning.’ (I. Perelmuter)

(15) ѢҽҦ ҰҸқґѦҔҗѦҮ ҲҝҽӊҸҮ, [...], ӁҸҡҾҲҡҾ Ҋҝҽҍѹ; ӊҡѦ ҸӁѹ ҰҸѽҸңѹ ҦҝҏҊ Ӂҽ ҦҝҏҊҽ ѹ ҦѦґҽӀҲ ҡҲѦҰҧ ѦӂѦ қҸґѦѦ ҸҦӁҸҸқҝҽӉӁҸҮ. (ү. ѯѦѽҸӊ) ‘High above the faded grass, [...], there are rooks flying; they are all similar and make the steppe look even more monotonous.’ (A. Chekhov)

According to Knjazev, ԐӾԞԌӫ ԙӹԼӹӡ frequently occurs with all kinds of symmetrical predicates, such as “ӊӉҽѹѸӁҸ-ӊҸӉӊҝҽҲӁӃѦ ҊґҽҊҸґӃ ҡ ӊӉҽѹѸӁӃѸ ӉӁҽҍѦӁѹѦѸ, ґѦҔҡѹҍѦҡҔѹѦ ҝѦ҃ѹҰҝҸҔѹ ѹ ҸқҝҽӉҸӊҽӁӁӃѦ ҸҲ Ӂѹѽ ӊҸӉӊҝҽҲӁӃѦ ҊґҽҊҸґӃ, ѸҏґҧҲѹҰґѹҔҽҲѹӊӁҸ-ӊӉҽѹѸӁӃѦ ҊґҽҊҸґӃ” ‘reciprocal reflexive verbs with reciprocal meanings, lexical reciprocals and derived reflexive verbs, multiplicativereciprocal verbs’, as illustrated by (16-19) (16) ѷѦҰѦҝҧ ҸӁѹ җґѹ, ҝҽӉҊҸӊҽҝѹӊҽҾ ѸѦңҦҏ ҡҸқҸҮ, ѹ ӁѦ ҝҽӉҏ ӁѦ ҸҊґҾӁҏґѹҡҧ. (ҷ. ҷҸҊҸѸҸґҸӊ) ‘Now they went on, talking among each other, and they did not glance back even a single time.’ (B. Bogomolov)

(17) ҤҝӃѸѹӁӃ ҋҲҽҝҗѹѦ ҰҸҡҲҸҾӁӁҸ ҡҏҦѹґѹҡҧ ҡ ѸґҽҦҗѹѸѹ, ѹӁҸҊҦҽ ѹ ѸґҽҦҗѹѦ ҡҡҸҝѹґѹҡҧ ѸѦңҦҏ ҡҸқҸҮ ѹ ӁҽҍѹӁҽґѹ ҡҏҦѹҲҧҡҾ. (ү. ѯѦѽҸӊ) ‘The Older Khrymins were constantly at law with the younger ones and sometimes even the younger ones quarrelled among each other and started to institute legal proceedings.’ (A. Chekhov)

(18) ҭҝѹҍѹӁӃ ҡѦѸҽӁҲѹҍѦҡҔѹѽ ѹӉѸѦӁѦӁѹҮ, ҔҽҔ ҰҝҽӊѹґҸ, ҲѦҡӁҸ ҰѦҝѦҰґѦҲҽӀҲҡҾ ѸѦңҦҏ ҡҸқҸҮ. (ѩ. ҹѸѦґѦӊ) ‘The causes of semantic changes interweave, as a rule, closely among each other.’ (D. Shmelev)

288

WIM HONSELAAR

(19) Ѥ ҊҸҝҦҸ ҰҸҡѸҸҲҝѦґ Ӂҽ ҍѹӁҸӊӁѹҔҸӊ, ҔҸҲҸҝӃѦ ѸѦңҦҏ ҡҸқҸҮ ҰѦҝѦҗѦҰҲӃӊҽґѹҡҧ ҡ ӊѹҦҸѸ ӁѦҏҦҸӊҸґҧҡҲӊѹҾ ѹ қѦҡҰҸҔҸҮҡҲӊҽ. (ү. ҭҏҗҔѹӁ) ‘I proudly looked at the officials, who were whispering to one another with a look full of displeasure and anxiety.’ (A. Pushkin)

Consequently, the combination ԐӾԞԌӫ ԙӹԼӹӡ is used to refer to the second participant of a symmetrical relationship. As such, it stresses that the state of affairs is restricted to the members of the group identified in the subject, often in contrast with other people. In that case, ԐӾԞԌӫ ԙӹԼӹӡ may refer to the internal organisation or structure of that group, as illustrated in (20) – found in the Amsterdam Corpus of Russian Texts –: (20) ҋҸӊѦҝҗѦӁӁҸ ҸҍѦӊѹҦӁҸ, [...], ҍҲҸ ҙҲѹ Ҧӊҽ ҰҝѦҡҲҏҰґѦӁѹҾ ҡӊҾӉҽӁӃ ѸѦңҦҏ ҡҸқҸҮ ҔҽҔ ҰҸ ҡҏқӆѦҔҲҏ, ҲҽҔ ѹ ҰҸ ҡҏқӆѦҔҲѹӊӁҸҮ ҡҲҸҝҸӁѦ, [...] (ѢѦӉӁҽӁҡҔѹҮ) ‘It is as clear as can be, [...], that those two crimes are connected, with respect to the subject, as well as to the subjective side, [...]’ (Neznansky)

Other verbs referring to the internal organisation of a group are, apart from the verb ԙӴԋԒԂԽӸ ‘to connect’ in (20), words which have the feature of ‘taking up positions in relation to one another’ or ‘making each other take up such positions’ in common; for example ԄԂӢԐӹӰӠӢӹӴԂԽӸ ‘to harmonise’, ԙӹӾԌӠӰӠԽӸ ‘to unite’, ӫӢԂӴӰӠӴԂԽӸ ‘to make equal’, (ԼՂԽӸ) ӰӾԙԖӹԌӰՂӡ ‘(to be) dissimilar’, ԙӢԂӴӰӠԽӸ ‘to compare’, verbs of speaking and communication, such as ԗӾӢӾԄӹӴԂӢӠӴԂԽӸԙԋ ‘to discuss’, ԙԗӹӢӠԽӸ ‘to argue’ and verbs referring to other forms of social interaction, such as ԗӾӢӾԄԊԋӰӫԽӸԙԋ ‘to exchange glances’. The natural effect of focusing on the subject group is – as Knjazev says – that this group is contrasted with other people who do not belong to that group: (21) ҘҽӉҽѽҡҔѹҮ ҾӉӃҔ ҡҲҽґ ҊҸҡҏҦҽҝҡҲӊѦӁӁӃѸ, ӁҸ ҔҽӉҽѽѹ, ңѹӊҏӂѹѦ ӊ ҊҸҝҸҦѦ, ҰҸҍҲѹ ӊҡѦ ҦҽңѦ ѸѦңҦҏ ҡҸқҸҮ ҊҸӊҸҝҾҲ ҰҸ-ҝҏҡҡҔѹ. (үҝҊҏѸѦӁҲӃ ѹ ӍҽҔҲӃ, 1966, 3) ‘The Kazakh language became the official standard language but almost all the Kazakhs who live in the city speak Russian even amongst themselves.’ (Argumenty i fakty, 1966, 3)

or from the Amsterdam Corpus of Russian Texts: (22) [...] ҽ ҴҽҲҝҽ҃ (ҲҽҔ ѸѦңҦҏ ҡҸқҸҮ Ӊӊҽґѹ Ҷҝѹѽҽ ҰҸҝҲӁѹѽѹ) [...] (ҋҸґҸҏѽѹӁ) ‘[...] and Mattress (as the dressmakers used to call Eric amongst themselves) [...]’ (Soloukhin)

A search in the Amsterdam Corpus of Russian Texts reveals that ԐӾԞԌӫ ԙӹԼӹӡ is also used with a non-reciprocal meaning, e.g. in (23) where the literal meaning of ԙӹԼӹӡ ‘himself’ is used:

RECIPROCAL MARKERS IN RUSSIAN

289

(23) ѢѦ ҡҲҽӊѹҲ ґѹ ҲҸӊҽҝѹӂ ҘѹҝҸӊ ӉӁҽҔ ҝҽӊѦӁҡҲӊҽ ѸѦңҦҏ ҡҸқҸҮ ѹ ҰҽҝҲѹѦҮ? (ҩӃқҽҔҸӊ) ‘Does comrade Kirov place a sign of equality between himself and the party?’ (Rybakov)

It also does not always refer to a participant of the predicate, as is, incidentally, also the case in Knjazev’s own example (see (21)), and in (24-25): (24) ҴѦңҦҏ ҲѦѸ җҰҝѹ҃, ҔҸҲҸҝӃҮ ҸӁѹ ѸѦңҦҏ ҡҸқҸҮ ӁҽӉӃӊҽґѹ «қҽҾӁҸѸ», ҰѦҝѦҗҺґ Ҕ ҦҝҏҊҸѸҏ ѸҽґҧҍѹҗҔѦ ѹ [...] (ѢѦӉӁҽӁҡҔѹҮ-ѷҸҰҸґҧ) ‘In the meantime, the syringe, which they called “bayan” among themselves, passed to another boy and [...]’ (Neznansky-Topol)

(25) [...] ҔӊҽҦҝҽҲѹҔѹ ҰҸӊѹҡҽӀҲ ӊ ӊҸӉҦҏѽѦ, ҽ ҰҝҾѸҸҏҊҸґҧӁѹҍҔѹ ҸқҝҽӉҏӀҲ ѸѦңҦҏ ҡҸқҸҮ ӉҽѸҔӁҏҲӃҮ ҔҝҏҊ. (ѢѦӉӁҽӁҡҔѹҮ) ‘[...] the little squares were hanging in the air and the tiny rectangles formed a closed circle.’ (Neznansky)

Finally, Knjazev remarks that ԐӾԞԌӫ ԙӹԼӹӡ can be used if the subject group consists of two or more participants, such as in his examples (26-27) but that it is hardly used to refer to situations in which two individuals act together 6 , such as in his example (28): (26) ҶҲѹ Ҧӊҽ ҏҊґҽ ҝҽӊӁӃ ѸѦңҦҏ ҡҸқҸҮ. ‘These two angles are equal.’

(27) ғҡѦ ҰҝҾѸӃѦ ҏҊґӃ ҝҽӊӁӃ ѸѦңҦҏ ҡҸқҸҮ. ‘All right angles are equal.’

(28) ?ҫӁѹ ҰҸ҃ѦґҸӊҽґѹҡҧ (ҰҸӉӁҽҔҸѸѹґѹҡҧ, ҰҸңѦӁѹґѹҡҧ) ѸѦңҦҏ ҡҸқҸҮ. ‘They kissed (made the acquaintance, married) among each other.’

Knjazev does not explain the reason for the discrepancy between the acceptable (26) and the less acceptable (28), in spite of the fact that both in (26) and (28) we are dealing with two participants – two right angles and two persons, respectively. It seems to me that the acceptability of (26) is accounted for by the fact that (26) specifies a timeless continuous situation of equality – and (27) even a generalised timeless situation of equality – whereas (28) deals with once-only mutual actions in the past, normally denoted by reflexive verbs (ԗӹՁӾԊӹӴԂԽӸԙԋ ‘to kiss (each other)’, ԗӹԒӰԂӪӹԐӠԽӸԙԋ ‘to meet’ and ԗӹԞӾӰӠԽӸԙԋ ‘to marry’), which by themselves indicate that the action is restricted to the people specified 6 “ҫҦӁҽҔҸ, Ѧҡґѹ ҰҝѦҦѹҔҽҲ ҡґҏңѹҲ ҦґҾ ҸқҸӉӁҽҍѦӁѹҾ ҡѹҲҏҽ҃ѹҮ, ӊ ҔҸҲҸҝӃѽ ҸқӃҍӁҸ ҦѦҮҡҲӊҏӀҲ ҲҸґҧҔҸ Ҧӊҽ ҏҍҽҡҲӁѹҔҽ, ҏҰҸҲҝѦқґѦӁѹѦ ӊҸӉӊҝҽҲӁҸҊҸ ѸѦҡҲҸѹѸѦӁѹҾ ҸҔҽӉӃӊҽѦҲҡҾ ӉҽҲҝҏҦӁѹҲѦґҧӁӃѸ [...]” ‘If the predicate, however, is used to refer to situations in which usually act only two participants, then the use of the reciprocal pronoun turns out to be problematic [...]’ (ҘӁҾӉѦӊ, 1998).

290

WIM HONSELAAR

by the subject. Any further specification of the group is therefore felt to be communicatively superfluous. Finally, Knjazev remarks that ԐӾԞԌӫ ԙӹԼӹӡ, other than ԌӢӫԄ ԌӢӫԄԂ, can freely be used with verbs having ‘identical surface realizations of actants’, as illustrated by his example (29) from Pushkin’s Yevgeni Onegin: (29) ғҸґӁҽ ѹ ҔҽѸѦӁҧ, ҡҲѹѽѹ ѹ ҰҝҸӉҽ, ґѦҦ ѹ ҰґҽѸѦӁҧ ӁѦ ҡҲҸґҧ ҝҽӉґѹҍӁӃ ѸѦңҦҏ ҡҸқҸҮ (*ҦҝҏҊ ѸѦңҦҏ ҦҝҏҊҸѸ). ‘Gulf and stone, poetry and prose, ice and flame do not differ so much [among each other].’

4. Conclusion If we do not take into account the highly exceptional non-idiomatic meaning of ԌӢӫԄ ԌӢӫԄԂ as ‘a friend of a friend’, then ԌӢӫԄ ԌӢӫԄԂ ‘each other’ is the only fully fledged explicit marker of reciprocity, with the reservation, however, that ԌӢӫԄ ԌӢӫԄԂ cannot be used in cases where reciprocity is metaphorically understood as the non-committal position of the speaker with respect to the actual division of roles in the situation referred to while the predicate can be “applied” only once. For reflexive and non-reflexive verbs referring to situations with symmetrical participants, reciprocity is just one of the possible interpretations: reflexive verbs can be interpreted as passives, causatives, true reflexives, etc. and nonreflexive verbs have an even broader range of interpretations. The complex reciprocal pronoun ӹԌӠӰ ԌӢӫԄӹԄӹ ‘one another’ is a reciprocal marker as well, but its use is restricted either to cases in which the situation referred to occurs more than once and every instance of the application of the predicate is seen as a separate, discrete event, with participants in different roles or to situations in which the situation referred to occurs only once and the speaker is non-committal as to the actual division of roles. The third complex reciprocal pronoun, ԐӾԞԌӫ ԙӹԼӹӡ ‘among each other’, is suitable for reciprocal and non-reciprocal situations but functions primarily as a noun phrase specifying the place where the situation takes place rather then as an explicit marker of reciprocity. Finally, I will present a tentative algorithm for the process of selecting an appropriate reciprocal marker and the criteria that have to be met during that process: x

The basic condition for the use of any reciprocal marker with a reciprocal meaning is that the situation referred to is characterized by the presence of “symmetrical” participants.

x

If the situation referred to is a situation in which the individual contribution of the participants to the event is not the primary focus (in case of so-

RECIPROCAL MARKERS IN RUSSIAN

291

cial etiquette) or if the speaker specifically focuses on the action as such, as part of a series of actions by the same agents, as potential rather than actual or otherwise distances himself from the action and its participants, a reflexive verb is used (Honselaar, forthcoming), but only if the reflexive verb has a reciprocal interpretation, that is to say, if the reflexive verb belongs to one of the categories “naturally reciprocal events”, “reciprocal aggressive actions”, “actions of exchange” or “actions of separation”. The same applies to implicit reciprocal non-reflexive verbs. The contribution of the participants to the event may be either more or less simultaneous or successive. NB If the speaker wants to take the perspective of one of the participants, he has to use specific “biased” words.

x

If the situation referred to is a situation in which the individual contribution of the participants to the event is the focus of attention and the contribution of the participants to the event is more or less simultaneous or successive, then the reciprocal marker ԌӢӫԄ ԌӢӫԄԂ ‘each other’ can be used. In Honselaar (forthcoming) I exemplify this perspective as follows: focus on what individual participants feel, behave or show, focus on unconventional behaviour, and focus on the individual body parts involved.

x

If reference is made to a series of separate situations in which the roles of the participants alternate, then the reciprocal marker ӹԌӠӰ ԌӢӫԄӹԄӹ ‘one another’ can be used. NB Note that ӹԌӠӰ ԌӢӫԄӹԄӹ does not necessarily mean that the situation is reciprocal for it can be used with respect to once-only situations in which only two participants occur, if the speaker does not want to specify which participant plays which role.

x

If the participants of the situation referred to are to be explicitly demarcated from other groups, then the reciprocal marker ԐӾԞԌӫ ԙӹԼӹӡ ‘among each other’ may be used. NB Note that ԐӾԞԌӫ ԙӹԼӹӡ does not necessarily mean that the situation is reciprocal.

University of Amsterdam REFERENCES ҘӁҾӉѦӊ, Ҍ.ҭ. 1998

“ҴѣҋѷҫѨҴѣѢѢѳѣ ҔҸӁҡҲҝҏҔ҃ѹѹ ҡ ӊӉҽѹѸӁӃѸ ӉӁҽҍѦӁѹѦѸ: ҦҝҏҊ ҦҝҏҊҽ, ҸҦѹӁ ҦҝҏҊҸҊҸ, ѸѦңҦҏ ҡҸқҸҮ”. ԛӾԙԽӰӠӪ ӰӹӴԄӹӢӹԌԙӪӹԄӹ ԄӹԙӫԌԂӢԙԽӴӾӰӰӹԄӹ ӫӰӠӴӾӢԙӠԽӾԽԂ. [ҋѦҝѹҾ “ҐҏѸҽӁѹҲҽҝӁӃѦ ӁҽҏҔѹ”] 9, 95-99 [quoted from http://www.admin.novsu.ac.ru/uni/vestnik.nsf/All/E18F8B7F4D0A93F8C32567 27002E7BA3]

292 ҋҩѤ 1981-1984

WIM HONSELAAR

ՀԊӹӴԂӢӸ ӢӫԙԙӪӹԄӹ ԋԒՂӪԂ Ӵ ԓӾԽՂӢӾԖ ԽӹԐԂԖ. ү.ҭ. ѣӊҊѦӁҧѦӊҽ (ҝѦҦ.). ҴҸҡҔӊҽ: ҩҏҡҡҔѹҮ ҾӉӃҔ.

Frajzyngier, Z., and T.S. Curl (eds.) 1999 Reciprocals. Forms and Functions. Amsterdam-Philadelphia: John Benjamins Publishing Company. Honselaar, Wim forthcoming “Reflections on reciprocity in Russian and Dutch”. In: Alexander Lubotsky, Jos Schaeken, Jeroen Wiedenhof (eds.), Evidence and Counter-evidence. Linguistic essays in honour of Frederik Kortlandt (Studies in Slavic and General Linguistics 32). Amsterdam-New York: Rodopi. Israeli, Alina 1997 Semantics and Pragmatics of the “Reflexive” Verbs in Russian. München: Verlag Otto Sagner.

SUMMARY In this paper a distinction is made between referentially based “conceptually” reciprocal situations (as defined by Knjazev/Peškovskij) and “semantical” reciprocity as expressed by reciprocal forms/markers in Russian, such as non-reflexive symmetrical predicates, reflexive verbs and – most explicitly – the (complex) reciprocal pronouns ԌӢӫԄ ԌӢӫԄԂ, ӹԌӠӰ ԌӢӫԄӹԄӹ and ԐӾԞԌӫ ԙӹԼӹӡ. For all of these reciprocal markers the paper discusses the specific relationship between the meaning of the reciprocal marker and character of the situation referred to. Finally, a tentative algorithm for the linguistic expression of referential reciprocity in Russian is presented.

Dutch Contributions to the Fourteenth International Congress of Slavists, Ohrid: Linguistics (SSGL 34), Amsterdam – New York, Rodopi, 293-331.

ON BURGENLAND CROATIAN ISOGLOSSES PETER HOUTZAGERS 1. Introduction Among the Croatian dialects spoken in the Austrian province of Burgenland and the adjoining areas1 all three main dialect groups of central South Slavic2 are represented. However, the dialects have a considerable number of characteristics in common.3 The usual explanation for this is (1) the fact that they have been neighbours from the 16th century, when the Ottoman invasions caused mass migrations from Croatia, Slavonia and Bosnia; (2) the assumption that at least most of them were already neighbours before that. Ad (1) Map 1 4 shows the present-day and past situation in the Burgenland. The different varieties of Burgenland Croatian (henceforth “BC groups”) that are spoken nowadays and from which linguistic material is available each have their own icon. 5

1

For the sake of brevity the term “Burgenland” in this paper will include the adjoining areas inside and outside Austria where speakers of Croatian dialects can or could be found: the province of Niederösterreich, the region around Bratislava in Slovakia, a small area in the south of Moravia (Czech Republic), the Hungarian side of the Austrian-Hungarian border and an area somewhat deeper into Hungary east of Sopron and between Bratislava and Gyǡr. As can be seen from Map 1, many locations are very far from the Burgenland in the administrative sense. 2 With this term I refer to the dialect continuum formerly known as “Serbo-Croatian”. The main dialect groups are ƀakavian, Štokavian and Kajkavian. 3 Neweklowsky mentions 13 “wichtigste Burgenland-Kroatismen”, one of which, the lexicon, forms a large set of common features by itself (1969: 99-101). For a dialectometric illustration of the linguistic closeness of the Burgenland dialects to each other see Sujoldžiǀ et al. (1990). 4 Based on Breu (1970: 220-229) and Neweklowsky (1978: 346). From the latter I also adopted the division into dialect varieties and the assignment of the individual dialects to those varieties. The dotted line represents the west boundary of the province of Burgenland. 5 The Moravian Croats are also represented by an icon, although their dialect is extinct or at least untraceable since the Second World War. There is some linguistic material from these dialects (see Neweklowsky 1978: 89-93).

294

PETER HOUTZAGERS

Map 1: present-day and former Croatian dialects in the Burgenland

ON BURGENLAND CROATIAN ISOGLOSSES

X

Moravian Croats Haci and Poljanci Kajkavian Dolinci Vlahi

295

Legend to Map 1 Štoji Southern ƀakavians intermediate Štoji – Southern ƀ. + not belonging to any of the groups formerly Croatian villages

The three dialects indicated by “+” do not belong to any BC group. These are Devínska Nová Ves and Chorvátski Grob (resp. NW and NE of Bratislava) and Weingraben (west of the Dolinci group). There is also a considerable number of small dots. Each dot stands for a location where a Croatian dialect was once spoken but disappeared between the 16th and the 20th century. In the case of the locations indicated by dots there is almost never any evidence as to what variety of Croatian was spoken there. However, the dots do give information on the former geographical spreading of Burgenland Croatian, which can have played a role in the diffusion of linguistic characteristics. The map shows that, whereas Burgenland Croatian today consists of a number of speech islands, it once more or less formed a dialect continuum. Only the Moravian dialects and those in the middle of Niederösterreich were already relatively isolated. Ad (2) There is no full agreement among linguists as to where the predecessors of the Burgenland dialects come from. Yet most authors locate at least the majority of them in or around a compact area surrounded by the rivers Sava, Kupa and Una. Map 2 below illustrates this. For a number of reasons it would be interesting to know which common characteristics of Burgenland Croatian are premigratory and which are postmigratory: (a) Such knowledge could help us clarify the chronology of some of the linguistic changes in the dialects under study. (b) In certain cases it could corroborate or contradict assumptions about the origins of the various BC groups. (c) The Burgenland Croats have been living for centuries in a complicated bior multilingual environment, in which the other languages (German, Hungarian, Slovak, Czech) have always had a higher prestige than Croatian. Many dialectologists would probably like to know, as I do, if under such circumstances the Croatians kept influencing each other linguistically and if so where, in which direction, to what extent, in which aspects of the language, etc. Finding out which instances of spreading of linguistic features are pre- and which are postmigratory would contribute to such knowledge.

296

PETER HOUTZAGERS

The aim of the present paper is to see if something can be said about the chronology of the linguistic phenomena that interest us by looking at them from the point of view of dialect geography, more specifically by comparing isoglosses in present-day Burgenland Croatian to hypothetical isoglosses in the (also hypothetical) area where the dialects came from. I shall do so from section 3 onward. But first I shall briefly discuss the various opinions on the origins of the Burgenland Croats.

Map 2: provenance of Burgenland Croatian according to N78 (281)6

2. The provenance of the Burgenland Croats The most extensive and in-depth discussion on the origin of the various Burgenland Croatian dialects is found in the standard work about these dialects, viz. Neweklowsky 1978, henceforth abbreviated “N78” (264-281). I shall summarize Neweklowsky’s views and compare them with those of a number of other authors.

6

Of the dialects that do not belong to any BC group, Neweklowsky locates only “WG” (Weingraben) and “CHG” (Chorvátsky Grob) on his premigratory map (with question-marks), not Devínska Nová Ves. He considers the Moravian Croats part of the Haci and Poljanci, with the only difference that less is known about them (see N78: 89-93). The dialects that are intermediate between Vlahi and Štoji are not discussed separately. It stands to reason that their transitional character arose after the migration.

ON BURGENLAND CROATIAN ISOGLOSSES

297

2.1 ƀakavian i/ekavian Neweklowsky’s assumption about the provenance of the dialects of the Haci and Poljanci, the Dolinci, Chorvátski Grob and Weingraben is chiefly based on comparison with present-day ƀakavian i/ekavian dialects in the Croatian inland. 7 In this comparison accentuation, diphthongization and vowel lengthening in final syllables play the most important role. In the Burgenland dialects discussed here the stress was in principle retracted from all final syllables and *e and *o were diphthongized if they were long. With the exception of part of the Dolinci dialects, stressed *e and *o were also diphthongized if they were short and already stressed before the retraction. In most of the ƀakavian dialects that the Burgenland dialects are compared with we also find stress retractions and either closing or diphthongization of *e and *o. In a number of cases there is also vowel lengthening in closed final syllables. However, the retractions often apply under more restricted conditions and there never is diphthongization of originally short *e and *o. The place of the Haci and Poljanci, the Dolinci, Chorvátsky Grob and Weingraben on Map 2 is chosen on the basis of the degree of similarity with the other inland ƀakavian dialects in these respects, to which are added some other linguistic as well as historical arguments. 8 Decisions of this kind – and this also holds for Neweklowsky’s propositions on the provenance of the other BC groups – are in fact compromises and therefore in themselves not very satisfying, but Neweklowsky does not suggest any great precision, makes all proper reservations and under the circumstances this is probably all that can be achieved.9 2.2 Ikavian All other Burgenland Croatian dialects except the Kajkavian ones are ikavian. 10 Neweklowsky argues that their original location should have been

7

The term “i/ekavian” means that the dialects in question have either an [e]- or an [i]-like vowel as the reflex of *LJ, in principle depending on the phonetic environment (see Jakubinskij 1925). 8 Although he does not say so, it seems probable that in Neweklowsky’s location of the different BC groups relatively to each other, differences and shared features between these groups themselves have played a role as well. 9 An example of such a compromise is the location of the Dolinci on Map 2 (east of the Haci and Poljanci), part of which, as far as the diphthongization is concerned, are closest to the dialects of Raǁice, Brinje, Oštarije and Generalski Stol (all to the west), cf. Neweklowsky (1978: 271). Another example is the place of the Haci and Poljanci on Map 2 in spite of the fact that, according to Neweklowsky, in a number of respects their dialect is closer to the one in the area around Otoǁac (1978: 271). 10 These dialects have an [i]-like vowel as the reflex of *LJ.

298

PETER HOUTZAGERS

(1) near the isogloss between the ikavian and the i/ekavian area. He assumes that this isogloss must have run from a point south-east of Dugi Otok (the most southeastern i/ekavian-speaking island today), almost straight north to the middle-course of the Una and then more or less parallel to that river to the left of it; (2) near the point where ƀakavian, Štokavian and Kajkavian bordered on each other. He bases this on the great number of similarities with other varieties of Burgenland Croatian and he locates the various dialects with respect to each other according to these similarities. This, of course, is only valid if it is improbable that the similarities concerned are postmigratory, which question we shall address later in this article. Neweklowsky ends his discussion on the origin of the ikavian dialects as follows: “Meiner Meinung nach könnte man die ikavischen ƀakaver des südlichen Burgenlandes entlang der Jat-Isoglosse lokalisieren, und zwar die rein ǁakavischen Mundarten etwas südwestlicher, die ƀakaver mit štokavischen und kajkavischen Elementen aber etwas nordwestlicher [obviously “nordöstlicher” is meant, P.H.] in die Nachbarschaft der Vlahi und Štoji. Darauf, daß diese ƀakaver nicht aus Dalmatien stammen, weisen die zahlreichen Übereinstimmungen besonders im Wortschatz mit dem übrigen Burgenländischen Material” (N78: 278-279). The first sentence suggests that Neweklowsky thinks that the dialects that are now transitional – both geographically and linguistically – between ƀakavian and Štokavian already were so in the premigratory situation. I think that this would be too much of a coincidence. If there are no strong indications to the contrary, it is more logical to assume that the dialects in question acquired their transitional character after the migration. What is said in the second sentence cited from Neweklowsky is only valid if there is reason to assume that the distribution of common lexicon is for the most part premigratory. 2.3 Kajkavian In the Burgenland, Kajkavian is only spoken in two villages: Hidegség and Fertǡhomok near Sopron. There is some (inconclusive) historical evidence that the inhabitants came from Kraljeva Velika and MeDžuriǁ (near Kutina on Map 2, see also Houtzagers 1999: 20-25). N78 gives examples of Kajkavian characteristics lacking in the dialect of Hidegség and Fertǡhomok and concludes that “Die sprachlichen Merkmale der beiden Mundarten lassen jedenfalls darauf schließen, daß sie sich einst in unmittelbarer Nachbarschaft der übrigen burgenländischen Kroaten befunden haben müssen, vor allem in der der Štoji und Vlahi”

ON BURGENLAND CROATIAN ISOGLOSSES

299

(279). I do not know if by “die sprachlichen Merkmale” Neweklowsky means the absent Kajkavian characteristics or other common Burgenland Croatian traits. With respect to the latter, I think that, in order to use them as an argument, there should be some indication that the distribution of the common traits is pre- and not postmigratory. Specific similarities between the Štoji and Vlahi and Burgenland Kajkavian are of course good candidates for being premigratory, since Kajkavian does not border on the other two in the new homeland. On the whole, as I have argued elsewhere (1999: 24-25), I find the location of Kajkavian as it appears on Map 2 plausible. Because the dialect lacks certain characteristics that Kajkavian dialects have in common today and that are assumed to be old (such as neo-circumflex in the present tense of verbs with fixed stem-stress), the original location of the dialect must be looked for outside the present-day Kajkavian speaking area. If one takes as a starting-point the hypothetical premigratory boundaries between ƀakavian, Štokavian and Kajkavian according to Ivšiǀ (1971: [799]) and Brozoviǀ (1970: 31), the area around Kraljeva Velika and MeDžuriǁ and to the northeast of them is the only one that has to be considered. If Lonǁariǀ (1995: 97) is right and Kajkavian stretched out considerably farther to the east, the dialect could originate from a more eastward location in Slavonia. However, this would not be an option if it could be shown that the similarities to the other varieties of Burgenland Croatian are premigratory.11 2.4 Other authors Neweklowsky’s ideas are in most cases very similar to and almost always compatible with those of Ivšiǀ (1971), Iviǀ (1961-62) and Brozoviǀ (1963, 1970). In Ivšiǀ (1971) and Brozoviǀ (1970) we find maps of the premigratory borders between Štokavian, Kajkavian and ƀakavian. As can be seen from Map 3, for the area we are discussing now, Ivšiǀ’s and Brozoviǀ’s borders are the same and almost fully compatible with Map 2. The only exception to this compatibility is the location of the Štoji on Map 2, for which we would have to stretch out the Štokavian area somewhat to form a Štokavian peninsula between Kajkavian and ƀakavian. 12 Iviǀ’s view (1961-62: 119-122) on the original locations of the dialects

11

In Lonǁariǀ’s view (1984, 1990) the dialect is a result of mixture of western Slavonian and MeDžimurje Kajkavian. Although I am not against the assumption of mixed origin for this and other Burgenland dialects, I do not think that by assuming mixture Lonǁariǀ explains the things he wants to explain (see Houtzagers 1999: 29). 12 The map in Ivšiǀ 1971 was made by the editor, Božidar Finka, on the basis of Ivšiǀ’s texts and notes from the 1930s. Ivšiǀ (1971) seems to contradict himself when on map nr. 5 he assumes that the Štoji came from western Slavonia, as well as the population of two other villages that he considers Štokavian but that Neweklowsky assigns to the south ƀakavian group (see also N78: 277 n60).

300

PETER HOUTZAGERS

of Tömörd (Dolinci, with vowel system identical to that of the Haci and Poljanci) and Narda (Štoji) is also compatible with Neweklowsky’s. 13 Lonǁariǀ (1988: 95) draws the border between Kajkavian and the other two main groups farther to the south than the other authors do, 14 but his view is compatible with Neweklowsky’s as well. The only difference is that the “Štokavian peninsula” mentioned in the preceding paragraph would diminish the Kajkavian and not the ƀakavian area somewhat.

Map 3: premigratory boundaries between Štok., ƀak. and Kajk. according to Ivšiǀ (1971) southern boundary of Kajk. west of the Sava according to Lonǁariǀ (1988) 13 Iviǀ gives two alternative possibilities for the origin of the dialect of Narda. The first one allows for the idea that the Štoji were originally located immediately north of the Una, which agrees with Neweklowsky. The second possible location would be farther to the north (near the Kupa). If we would want to combine this with Ivšiǀ’s and Brozoviǀ’s borders between the main dialect groups, we would need an even longer “Štokavian peninsula” than the one just described. Iviǀ’s views are incompatible with Lonǁariǀ’s (1988, 1995). Iviǀ supposes a strip of ƀakavian and, in one scenario, Štokavian dialects very near to the Kupa, whereas Lonǁariǀ assumes that south of the Kupa and southwest of the Sava there was a considerable area where Kajkavian was spoken. 14 Northeast of the point where it meets the Sava, Lonǁariǀ’s southern Kajkavian border runs just as it does on Ivšiǀ’s and Brozoviǀ’s maps. In his 1995 article, however, Lonǁariǀ describes another view, according to which (a) Kajkavian and transitional Kajkavian-Štokavian dialects could be found much farther east in Slavonia than Ivšiǀ and Brozoviǀ assumed and (b) Kajkavian also stretched out rather far to the south, so that Štokavian in this area was spoken only in a relatively narrow strip north of the Sava (96-97).

ON BURGENLAND CROATIAN ISOGLOSSES

301

2.5 Historical vs. linguistic evidence In his chapter on the provenance of the Burgenland Croatian dialects, Neweklowsky distinguishes between historical and linguistic data and treats them in two different sections (N78: 264-266 and 266-281). In the beginning of the chapter he states “Uns soll die kroatische Besiedlung des Burgenlandes und der angrenzenden Gebiete nur in Zusammenhang mit der Dialektologie interessieren” (264). In keeping with this, the map that he presents at the end of the chapter and that is here reproduced as Map 2 seems to be based on the linguistic data only. Map 2 suggests that all the Burgenland Croats came from the area between the Kupa, Sava and Una or the direct vicinity of that area. However, many sources, several of which are cited by Neweklowsky himself, mention migrations into the Burgenland from other areas. The most prominent of these other areas is western Slavonia, followed by the Lika (a region in Croatia well to the south of the area shown on Map 2). Ujeviǀ assumes that between 1522 and 1527 migrations have taken place from Lika to the area around Sopron (1934: 8). According to Paviǁiǀ there have been mass migrations between 1536 and 1540 from the western part of Slavonia (until approximately 100 km. east of Sisak),15 first to the south of Hungary but later to the north of the Burgenland and the area around Bratislava (1920: 216-218). Paviǁiǀ also reports that in 1540 peasants from the west of Slavonia settled in the southern part of the Burgenland (1953: 208209). Dobrovich mentions settlers from southern Bosnia in the county of Forchtenstein, between Sopron and Wiener Neustadt (1963: 88, cited by Koschat 1978: 35). According to Valentiǀ: “Die größte Zahl der Auswander kam jedoch aus den Gegenden um den unteren Lauf der Kupa und der Una, ganz besonders viele aber aus Westslawonien” (1984: 15). Ernst writes: “Während sich die Herkunftsgebiete der meisten kroatischen Immigranten nur ziemlich global festlegen lassen, wie die Moslavina in Slawonien, Nordbosnien und die Region zwischen Una und Kupa sowie die Lika und Krbava (bis zum Velebitgebirge) […]” (1987: 250). It is not clear whether Neweklowsky sees the contrast between the historical data and his linguistic reconstruction as a problem. In my opinion, however, it is a matter that should be accounted for, especially where western Slavonia is concerned. 16 Lonǁariǀ (1984: 121-124) makes an attempt to solve the apparent conflict between historical sources and linguistic reconstruction. He suggests the possibility that the immigrants from Slavonia settled later, in relatively small 15

Paviǁiǀ describes an area of approximately 70 km. from north to south. The Lika perhaps presents less of a problem. The migrations from these regions were not directed exclusively to the north (cf. Valentiǀ 1984: 15), so that the number of settlers in the Burgenland was not necessarily very great. Moreover, it is possible that the dialects spoken in (part of) the Lika were not too different from Burgenland ƀakavian. 16

302

PETER HOUTZAGERS

numbers at a time, into villages previously occupied by settlers from the area indicated on Map 2. These “new” immigrants for the most part adapted linguistically to the population that was already there, but did have a certain influence, especially where the lexicon is concerned. This would explain the remarkable quantity of characteristics common to Burgenland ƀakavian that are usually considered Kajkavian. “Navedene se crte mogu shvatiti na dva naǁina. Prvo, možemo ih najveǀim dijelom dovesti u vezu s podrijetlom gradišǀanskih Hrvata nekajkavaca, tj. s podruǁja izmeDžu Kupe, Save i Une te iz srednje Slavonije, dakle iz kajkavskoga susjedstva, koje su zahvaǀale neke kajkavske izoglose. Drugo, utjecajem kajkavskih emigranata iz zapadne Slavonije koji su se u manjem broju naseljavali po Gradišǀu s naseljenicima iz drugih krajeva, kojih je govor prevladao, dok su od kajkavaca ostale pojedine crte. Naravno, u nekim mjestima moguǀe je djelovanje obaju ǁinitelja” (124).17 Lonǁariǀ’s explanation sounds convincing. It provides a solution for the apparent linguistic disappearance into thin air of a considerable number of people and at the same time explains the Burgenland kajkavisms.18 2.6 Conclusion I conclude that Neweklowsky’s reconstruction of the former dwelling-places of the Burgenland Croats provides a good starting point for the comparison of present and former isoglosses. As I have tried to show in the preceding sections, there are a few instances in which Neweklowsky assumes that certain common Burgenland Croatian traits are premigratory, while I think that this does not necessarily has to be the case. Moreover, apart from the dialect of Hidegség and Fertǡhomok, Neweklowsky does not seem to take into account the potential linguistic input of migrants from Slavonia. I shall return to these issues in due course.

17

Neweklowsky gives only the first explanation by Lonǁariǀ for the kajkavisms in Burgenland Croatian (1982: 263). There is a possibility that he does not assume, as Lonǁariǀ does, that Kajkavian was originally spoken until far into Slavonia. 18 Of course, if we would choose only Lonǁariǀ’s first alternative (“Prvo …”), the problem would remain unsolved. It would be worth while to check Lonǁariǀ’s scenario against other cases of mixed settlement.

ON BURGENLAND CROATIAN ISOGLOSSES

303

3. Isoglosses 3.1 Preliminary remarks In this and the following sections we shall have a closer look at a number of isoglosses in Burgenland Croatian from a dialect geographic point of view and see if this contributes anything to the discussion about their being pre- or postmigratory. In order to make clear what I mean I shall first give a few examples. On Map 4 (N78M26) 19 we see the isoglosses for the form of the l-participle of *xԔtLJti, which can be either štil or til. 20 The map consists of two parts. Map 4a represents the present-day state of affairs, Map 4b a hypothetical former situation. 21 This hypothetical situation is nothing more than a projection of the present-day situation on a hypothetical premigratory map. Thus, not only the locations of the BC groups are hypothetical, but also the assignment of linguistic characteristics to specific BC groups. On Map 4b, the BC groups have the same locations as on Map 2, i.e. the premigratory locations as proposed by N78. The BC groups have labels “ƀ1”, “Š2”, etc., where “ƀ” stands for “ƀakavian”, “Š” for “Štokavian” and “K” for “Kajkavian”. ƀ1 are the Haci and Poljanci, ƀ2 the Dolinci, ƀ3 the Southern ƀakavians, Š1 the Vlahi and Š2 the Štoji. On Map 4b we see, for obvious reasons, only these labels and no icons that indicate individual villages as on Map 4a. “CHG” means “Chorvátsky Grob”, WG means “Weingraben”.22 On Map 4a the areas that share the same linguistic characteristic are discontinuous, but on Map 4b they are continuous. One tends to infer from that that the distribution of this characteristic is, at least for the most part, premigratory.

19 The maps with numbers 4 and higher can be found at the end of the present article. N78Mx means “map x in N78” (after page 376). The sets of villages shown on the maps differ, as they do on the maps in N78, because not all relevant forms have been attested in all villages. The Moravian Croats are not shown on these maps. They only have separate relevance in the case of the phenomenon discussed in 3.2.3 below and shown on Map 8a. 20 By underlining št and t I indicate that these are the phonemes that the discussion is about. Variation in other parts of the forms, e.g. the reflex of *LJ, is ignored. Other cases of underlining in this paper must be interpreted in the same sense. 21 This map and almost all the maps that follow are based on the maps and/or the data in N78. In some cases I have completed or corrected the data from the three villages from which I have my own field-work material. These are the two Kajkavian villages on the south bank of the Fertǡ Lake and the neighbouring ƀakavian village of Kópháza. 22 On maps 4b, 5b, etc. some icons that stand for linguistic characteristics are smaller than others. The smaller icons are used when the feature is found in Chorvátsky Grob, Weingraben, or only in a small part of a certain BC group.

304

PETER HOUTZAGERS

An exception to all this seems to be Chorvátsky Grob, which can have taken over til from the Haci and Poljanci after the migration.23 Of course, this is all highly hypothetical, because: 1. We assume that the locations on Map 4b are correct, which is not certain. 2. Even if we assume that they are correct in principle, we have no idea about the shape and size of the area occupied by each BC group and about the borders between them. 3. We have very little knowledge about the varieties of Croatian spoken around the dialects shown on Map 4b. It is not excluded that we should draw different conclusions if we had more. 4. We assume that the BC groups on Map 4b have a one-to-one relationship with those on Map 4a. It is highly improbable that it is for 100% the case. Dialects that belonged to the same premigratory BC groups in the area shown on Map 4b did not necessarily move to contiguous areas in the new homeland and it is very unlikely that they always did. Moreover, it is certain that the representatives of these groups were not the only Croatian-speaking immigrants in the area (see 2.5 above). 5. In addition to all this, we have the usual caveats that apply when one wants to draw conclusions about the history of a linguistic phenomenon by looking at the dialect map. To mention only two: (a) areal diffusion of a linguistic characteristic does not always take place just in the way most linguists would predict; (b) very often a linguistic characteristic and its development cannot be seen in isolation but must be viewed in a broader synchronic and diachronic context. Yet I think that, if we keep these reservations in mind, try to avoid the pitfalls and use all the other knowledge we have about the characteristics under discussion, it is worth while to look at the Burgenland isoglosses from this point of view. Especially if the number of maps we take into account is not too small, we should be able to draw some tentative conclusions. 3.2 More examples 3.2.1 On Map 5 (N78M51) we see the different lexemes for the word ‘corn’ that are used in the Burgenland. If we compare Map 5a with Map 5b, we see that it is improbable that the present-day distribution of the lexeme among the BC groups was “brought along” by the Croatian immigrants, especially where the Štoji (Š2) are concerned. Probably the Štoji originally said žito/žitak and part of 23

There is also a possibility that the Chorvátsky Grob dialect bordered on that of the Southern ƀakavians before the migration (see also 3.4.2 below).

ON BURGENLAND CROATIAN ISOGLOSSES

305

them took over teg from the neighbouring Southern ƀakavians. One Southern ƀakavian village took over žito/žitak from the neighbouring Vlahi and Kópháza took over zrnje from the Kajkavian Croats. 24 Other scenarios are possible as well. It cannot be excluded that only the Vlahi (Š1) originally had žito/žitak and that the spreading to both the Štoji (Š2) and Dolinci (ƀ2) is postmigratory. This is slightly less probable in view of the greater geographical distance to the Dolinci, with few intermediate villages (see also Map 1). However, if convincing examples can be found of spreading of linguistic characteristics between the Vlahi and the Dolinci, that would substantiate this alternative hypothesis.25 3.2.2 Of course the dialect geographic picture taken by itself often gives us no clue as to whether the present-day distribution is pre- or postmigratory. The distribution of žetva vs. žatva ‘harvest’, for instance, makes perfect sense, whether we look at the present situation or at the hypothetical old one (Map 6, N78M12). 26 On Map 6b the areas which share the same characteristic are adjacent to one another. On Map 6a there is one area with a in the middle which separates the two larger areas with e, but if we regard a in this particular word as an innovation, the present distribution can be as recent as we want it to be. However, historical linguistics tells us that we have to do with two different reflexes of *Nj after a palatal consonant and it stands to reason that the difference between e and a in this word arose as early as the denasalization of the front nasal, viz. centuries before the earliest migrations. Since the reflexes of *Nj after palatal consonants are to a high extent lexicalized, it is possible that the isogloss moved somewhat in the course of the centuries, but it runs so neatly around one BC group (the Dolinci) that there is no reason to assume that it did. Map 7 (N78M11), which shows us the reflexes of *Nj in the same phonetic environment in another lexeme (žedan/žajan ‘thirsty’), demonstrates how lexically dependent the distribution of this feature is (note that the Haci and Poljanci have žetva but žajan) and at the same time corroborates the assumption that the isogloss originated in the old homeland. As was the case with Map 6, there is little reason to suppose that the isogloss moved after the migration. 3.2.3 Map 8 shows the locative singular endings of o-stem nouns (both masculine and neuter) as they are presented on N78M30, completed with some infor24 The reader might object that part of the Štoji can already have taken over teg from the Southern ƀakavians before the migration. However, this is highly improbable, since it would mean that the same Štoji dialects that now border on Southern ƀakavian already did so before the migration (cf. the last paragraph of section 2.2). 25 In the course of this article we shall see that such examples are rare. 26 Map 13 of Neweklowsky 1978 suggests that in the westernmost Dolinci village the form is žetva, which is contradicted op page 105. I assume that the icon on the map is a mistake.

306

PETER HOUTZAGERS

mation from the main text of N78 (195). According to N78, the dialects that have -i also have -u as a free variant ending, except the Moravian Croats and Chorvátsky Grob, which exclusively have/had -i. 27 (I left area 2 on Map 8a open to symbolize that it includes the Moravian Croats; the small club under ƀ1 on Map 8b stands for Chorvátsky Grob and the Moravian Croats.) Literary Burgenland Croatian also had free variation until the 19th century (Hadrovics 1974: 190, 192-193), whereas, according to N78, now only -u is written (195, no sources given). All this suggests that in the premigratory situation the Vlahi and Štoji had -u and the other BC groups -i. After the migration -u spread to the rest of Burgenland Croatian, resulting in the disappearance of -i in part of the Southern ƀakavian dialects and variation in the other BC groups, except in the extreme north, which was not reached. If one assumes that prior to the migration the Štoji and Vlahi formed part of a larger area with -u, the premigratory location of the dialect of Chorvátsky Grob is improbable (see also 3.4.2 below). Summarizing, one can conclude that Map 10 reflects a large scale postmigratory development departing from a premigratory isogloss. 3.2.4 Map 9 (N78M27) tells us which dialects reflect original final -m as -n and which ones preserved -m. We see that the only BC group that has final -m consistently reflected as -n is that of the Štoji. Part of the Vlahi has retained -m and the picture of the Dolinci and the Southern ƀakavians suggests that for these BC groups -m is the relict form. The only Haci and Poljanci dialect that has -n is that of Kópháza, which neighbours on a Dolinci area with -n. It is clear that -n spread at least partially after the migration. I propose a premigratory situation in which the Štoji had -n and were the center from which -n diffused afterwards. It is probable that the diffusion of -n from the Štoji to the Dolinci had started before the migration. 3.2.5 If we would want the maps 4b-9b that we have discussed so far to represent the situation as it probably was before the migration, we would have to “correct” four of them. On Map 4b we would have to change the feature attested for Chorvátsky Grob and on Map 5b one of the features for the Štoji and the minority features for the Haci and Poljanci and the Southern ƀakavians. On Map 8b we would have to remove the ending -u for the Southern ƀakavians and the possibility of free variation between -i and -u for the Dolinci, Kajkavians and Haci and Poljanci. Chorvátsky Grob does not fit in very well (see 3.4.2 be27 This free variation between -i and -u is confirmed by Koschat 1978: 82. For Kajkavian and the neighbouring Haci and Poljanci dialect of Kópháza I must refine the picture somewhat. There the normal ending is -i and -u is only attested for a very restricted set of lexemes (Houtzagers 1999: 76 and forthc.). The preference for -u in svet ‘world’ agrees with that found in older written texts (see Hadrovics 1974: 192-193).

ON BURGENLAND CROATIAN ISOGLOSSES

307

low). On Map 9b we would have to restrict final -n < -m to the Štoji and part of the Dolinci. All this would result in the “corrected” premigratory maps numbered 10 through 13 below. Maps 6b and 7b are not in need of “correction”. It goes without saying that “corrected” premigratory maps like Maps 10-13 are even one step more hypothetical than those of the type 4b and 5b, because they include my reconstruction, which does not have to be correct. 3.3 Comparison: pre- or postmigratory I made 52 comparisons of the type described above, based on the maps presented in N78 (after page 376). The reason why I chose these maps as a basis for the comparison between the new situation and the hypothetical old one is the fact that almost all of them are about features that are explicitly used to distinguish the BC groups from each other. 28 Because the BC groups are assumed in some way or other to be continuations of premigratory dialect varieties, it is a reasonable hypothesis that the distribution of the features themselves is in principle premigratory.29 For reasons of space I cannot present the details of each comparison in this article, but I shall give my conclusions, illustrated by examples and maps when I consider that helpful. 3.3.1 In most cases (37 out of 52) my comparison of the maps in N78 with the hypothetical premigratory maps along the lines shown in 3.1 and 3.2 corroborates Neweklowsky’s explicit or implicit hypothesis that we must consider the relevant characteristics premigratory, by which is meant that both the isoglosses themselves and their general geographic picture arose before the migration. 30 This does not mean that after the migration nothing happened (see 3.5 below). 3.3.2 In more than half of these cases (20 out of 37) the main clue against the assumption that the spreading of the relevant features was postmigratory are features shared by Kajkavian and presently non-adjacent BC groups to the south. The following possibilities can be distinguished:

28 On pages 59, 60, 94, 108, 115, 138, 153, 170 and 186 I count 45, but Neweklowsky’s enumerations are often followed by “u.a.” (‘and others’). 29 This hypothesis is by far not always explicit in N78, but sometimes it is, especially in the chapter on the origins of the Burgenland Croats (264-281). 30 With respect to the question whether isoglosses were pre- or postmigratory I could find no statistically significant differences between isoglosses pertaining to different levels of the language such as phonology, morphology or lexicon.

308

PETER HOUTZAGERS

(a) The relevant feature is shared by Kajkavian and the Vlahi (3 cases: N78M4, 5, 34). (b) The relevant feature is shared by Kajkavian, the Vlahi and the Štoji (8 cases: N78M21, 26, 28, 32, 35, 36, 37, 48). These cases can also be seen as cases in which the ƀakavian dialects are opposed to the other BC groups (see 3.3.3 sub (i)). In almost all of these cases part of the Southern ƀakavian BC group shares the same feature, but, in contradistinction to the cases under (c) below, there is reason to believe that this arose after the migration. I shall give two examples. The first is Map 14 (N78M36). In view of the marginal location of the Southern ƀakavian villages with uho it is probable that this is the original form for this lexeme in this BC group and that the other form spread from the Vlahi and Štoji after the migration. If we believe that the original location of Chorvátsky Grob is correct, we have to assume that it took over uho from the Haci and Poljanci (but see 3.4.2). Map 15 (N78M37) shows us the same diffusion from the Vlahi and Štoji to the Southern ƀakavians, but on a smaller scale. Chorvátsky Grob must have taken along its characteristic from the old homeland and probably spread it to Devínska Nová Ves. There is one Štoji village that does not fit in very well, viz. Miedlingsdorf (see 3.4.3 below). (c) The relevant feature is shared by Kajkavian, the Vlahi, Štoji and Southern ƀakavians (4 cases: N78M6, 11, 18, 38). (d) The relevant feature is shared by Kajkavian and the Štoji (5 cases: N78M9, 25, 41, 42, 50). In most cases the feature spread to part of the Southern ƀakavians after the migration (see (b) above). In accordance with Neweklowsky’s premigratory map, no features are shared by Kajkavian and the Southern ƀakavians exclusively. 3.3.3 The remaining 17 cases in which the dialect map seems to corroborate the hypothesis “premigratory” fall apart into a number of small categories. (i)

(ii)

The relevant feature is shared by non-adjacent ƀakavian BC groups (8 cases; the 8 cases in 3.3.2 sub (b) can also be argued to belong here). In one case the Dolinci share the relevant feature with the Southern ƀakavians only (N78M19, see last paragraph of 3.5.1 below and Map 19), in three cases the feature is shared by the Southern ƀakavians and the Haci and Poljanci (N78M47, 49, 51), in the other 4 (or 12) cases the feature is common to all ƀakavian dialects (N78M1, 10, 24, 45). In part of these cases postmigratory developments have blurred the picture, especially in the Southern ƀakavian area. The relevant feature is shared by the Dolinci and the Štoji (1 case: N78M13).

ON BURGENLAND CROATIAN ISOGLOSSES

309

(iii) The relevant feature is shared by the Haci and Poljanci, Dolinci and the Štoji (1 case: N78M46). (iv) The relevant feature opposes one BC group to all the other ones (2 cases), and there is reason to believe that the feature did not occur separately in that one BC group. There is a single case in which the “one group” is the Dolinci and one in which it is the Vlahi (N78M12 and 15, respectively). (v) The relevant feature is obviously old, because it opposes large areas within Central South Slavic to each other and it also agrees with Neweklowsky’s premigratory map. The present dialect geographic picture does not corroborate much apart from the fact that the isoglosses run neatly between the BC groups (1 case: N78M14) (vi) The data from only a small number of villages corroborate the hypothesis “premigratory”. This occurs in the case of Miedlingsdorf (2 instances, one of which together with Zuberbach; both belong to the Vlahi BC group: N78M2, 3) and Chorvátsky Grob together with Devínska Nová Ves (1 instance: N78M39). (vii) There is one complicated case, for which it cannot be said to which of the categories (i-v) it belongs (N78M22). We see this case on Map 16. The original form is undoubtedly bil and the marginal location of the villages with bil on Map 16a suggests that at least part of the diffusion of the other forms has been taking place after the migration. At the same time the discontinuous areas with bi indicate that part of the innovating forms are also premigratory. Several scenarios are possible. One of them is that before the migration the Vlahi already had bi(j)a, part of the Haci and Poljanci had biuԘ and part of the continuum of the Štoji, Dolinci and Southern ƀakavians already had bi. After the migration these innovating forms kept spreading at the expense of bil. 3.3.4 In 2 cases it is more probable that the general geographical picture of the relevant feature is postmigratory. One of these was discussed in 3.2.3 above (Map 8 in the present article, N78M30). The other one is that of the distribution of the preservation or loss of the phoneme Ǜ (Map 17, N78M23). Apart from the Štoji, in every BC group larger than one village there is a mixture of some kind. Kajkavian has two possibilities (Ǜudi/judi), the other BC groups are divided: some villages have Ǜudi, others have judi. The Haci and Poljanci group is almost homogeneous, with consistent Ǜudi, except in the village nearest to the Kajkavians (Kópháza), which has Ǜudi/judi. The situation in Kópháza and Kajkavian suggests that the process of loss of the phoneme Ǜ is going on right now, and the location of most of the villages in the middle and south of the Burgenland which have preserved Ǜ seem to indicate that the spread of the loss of Ǜ has at least in part taken place after the migration. The only thing that speaks against assuming this spread as entirely postmigrational is the slight discontinu-

310

PETER HOUTZAGERS

ity of the areas with loss of Ǜ. Possibly in the premigratory continuum between Vlahi, Southern ƀakavians and Dolinci there already were a few villages where Ǜ was being lost and after the migration loss of Ǜ continued spreading from several – now discontinuous – locations. 3.3.5 There are 13 cases in which I think that it cannot be established whether the relevant feature arose before or after the migration (N78M7, 8, 16, 17, 20, 27, 29, 31, 33, 40, 43, 44, 52). Map 18 (N78M7) shows what appear to be two different degrees of diphthongization resulting in rising diphthongs, the isogloss between which splits Burgenland Croatian in a northern and a southern part. 31 To the south only originally long e and o have diphthongized (with some phonetic variation); to the north, in addition, short e and o have diphthongized if they were already stressed before the stress retraction. What the areas marked “2” on the map have in common is the absence of the change of long e and o to rising diphthongs. Chorvátsky Grob and some Vlahi dialects have no diphthongization at all. Other Vlahi dialects have different other kinds of diphthongization, often resulting in falling diphthongs as reflexes of (part of the) old long vowels.32 The Kajkavian dialects diphthongized old long e and o to ei and ou. It is not excluded that the tendency to diphthongize long vowels into falling diphthongs must be accounted for by the assumption that before the migration the Vlahi and Kajkavians were neighbours. On the other hand the Vlahi dialects do not form a unity at all in this respect (very often other vowels than long e and o are involved) and in the case of the Vlahi the variety of German spoken in the area could have served as a model for this type of diphthongization (cf. N78: 29-31). In the case of Kajkavian the diphthongization had probably taken place before the migration: 1. it is very unlikely that the Kajkavian dialect was deeply influenced by German and 2. the languages by which the speakers of Burgenland Kajkavian – which has probably never consisted of more than a few villages – have always been surrounded, viz. western Hungarian and Burgenland ƀakavian, diphthongize their long vowels in exactly the opposite direction (cf. Houtzagers 1996: 127). Regarding the diphthongization resulting in rising diphthongs, what makes the chronology difficult is that it can be doubted whether it represented (and represents)33 more than a phonetic characteristic of long e and o which can have 31

On N78M7, the south easternmost Dolinci village (Tömörd) belongs to the northern type within the Dolinci BC group. I corrected that on the basis of the main text (N78: 94). 32 The phonological and phonetic accounts of N78 (170-180) and Tornow (1989: 3-4) on the vowels in the Vlah dialects do not always agree. 33 In N78, diphthongization is consistently treated as a phonetic and not a phonemic characteristic, with the exception of Weingraben, which has lost distinctive length.

ON BURGENLAND CROATIAN ISOGLOSSES

311

come about at any time, affecting different sets of vowels according to the lengthenings that had previously taken place in the different dialects. In the north, where originally long and originally stressed short vowels are reflected as diphthongs (tièlo, gliédat) but originally pretonic short vowels are not (sտelo), we can assume that before the retraction stressed short e and o were lengthened, in the south they were not. I do not agree with Neweklowsky that the situation in the north proves that the diphthongization took place before the retraction (N87: 73). In my opinion the situation in the southern half of the Dolinci group forms a stronger case for Neweklowsky’s chronology. Here we have three sets of stressed o- and e- like vowels. Original long e and o are reflected as diphthongs, original stressed short e and o as long rising monophthongs and short e and o that received the stress as a result of the retraction as short monophthongs: mìeso, lé:to, sտelo (N78: 94-95). An elegant way to account for this would be to assume (1) diphthongization of long e and o; (2) lengthening of short stressed e and o; (3) retraction. However, this solution is not the only one possible. It is also imaginable that there were other types of timbre distinctions between the different sets of vowels before the present situation arose.34 Summarizing, I conclude that little can be said about the chronology of the various types of diphthongization. In view of the frequency with which rising diphthongs are found in ƀakavian dialects, it is reasonable to assume that at least a tendency toward diphthongization was brought along by (an unknown part of) the ƀakavian dialects and spread to most of the Burgenland afterwards. There is an (albeit inconclusive) indication for a terminus ante quem in the case of part of the Dolinci. The course of isogloss 1 on Map 18 makes it improbable that the whole picture as it is now is premigratory: it would be too much of a coincidence if the neat internal division between north and south within the Dolinci group would reflect a premigratory situation. 3.4 Isoglosses on the premigratory dialect map 3.4.1 Those 37 cases in which we can assume that both the relevant characteristics themselves and their general distribution are premigratory agree very well with the premigratory map proposed in N78. In each case, if we would make “corrected” premigratory maps by removing the part of the distribution that we can reasonably consider postmigratory, the result would be a neat dialect map on which the linguistic features are shared by continuous areas. If we take all these maps together, we see that the isoglosses run through the area in almost every imaginable way and unite almost every possible set of premigratory neighbours. This, in my opinion, strongly corroborates Neweklowsky’s assump34

As a matter of fact, Neweklowsky assumes that on the phonemic level this type of Dolinci dialects has four degrees of openness, one of which is phonetically diphthongized (N78: 94).

312

PETER HOUTZAGERS

tions. There are 19 different distribution patterns, one of which is found in 9 cases: here the area is divided into two parts with Kajkavian and Štokavian (Štoji and Vlahi) on one side of the isogloss and the ƀakavian BC groups (Dolinci, Southern ƀakavians and Dolinci) on the other (N78M21, 24, 26, 28, 32, 35, 36, 37, 48). One distribution pattern occurs 4 times, two occur 3 times, the other 15 occur once only or twice. This suggests that before the migration the dialects have been neighbours for a considerable time and that the situation was one of intensive contact in every possible direction. 3.4.2 There are also two individual villages that figure on Neweklowsky’s premigratory map: Weingraben and Chorvátsky Grob (both with a question mark). The location of the former (between the Dolinci, Štoji and Kajkavians) is plausible and presents no problems. There are a few cases in which Weingraben shares characteristics with the Kajkavians and not with the other premigratory neighbours (final -m, the lexeme ide as opposed to gre, Isg of a-stems in -um). This could lead one to move the premigratory location of Weingraben somewhat to the east, but this is not necessary since the Kajkavian influence might have come from the north as well (from the north bank of the Kupa, see Map 3). The location of Chorvátsky Grob south of the Vlahi is more problematic. This village shares several characteristics with the Dolinci and Haci and Poljanci that cannot reasonably be assumed to have been taken over after the migration, e.g. the i/ekavian reflex of jat. In addition it has several traits in common with the Štoji, either together with Kajkavian (e.g. prothetic v- in vuǁiti, menom as opposed to manom) or the Southern ƀakavians (distinctive length in accented and pretonic syllables). It seems more probable that the premigratory location was to the northwest of the Vlahi, between the Vlahi, Štoji, Dolinci and Southern ƀakavians. The only feature shared exclusively with Kajkavian (viz. the interrogative pronoun kaj) could be due to mixed settlement (see the last paragraph of 2.5). 35 The small number of villages in which this type of Burgenland Croatian was spoken (cf. N78: 108) and its peripheral position within Burgenland Croatian can have it made more susceptible to influences from outside. Devínska Nová Ves has no separate location on Neweklowsky’s premigratory map. This is probably due to the fact that the dialect is very similar to that of the Haci and Poljanci and there are no characteristics shared with other BC groups that would suggest historical connections. There are a few instances in which the dialect was probably influenced by those of the Chorvátsky Grob type

35

Menom and vuǁiti could also be due to mixed settlement, but are already sufficiently accounted for if we assume that the Štoji were among the premigratory neighbours.

ON BURGENLAND CROATIAN ISOGLOSSES

313

(see for instance N78M37, 39).36 The original location of the Devínska Nová Ves dialect can in principle have been any not too central place in the Haci and Poljanci area or a place near that area. 3.4.3 The two north westernmost Štoji villages Zuberbach and Miedlingsdorf also deserve separate attention. These villages have a more or less i/ekavian reflex of jat and several other characteristics that connect them historically with i/ekavian ƀakavian, which is why Neweklowsky calls them “štojisierte, ehemals aber ǁakavische Mundarten” (N78: 153). The problem is, however, that among the many characteristics that Zuberbach and Miedlingsdorf share with their present-day neighbours, there are several that they cannot have taken over after the migration, such as the absence of stress shift or the presence of distinctive length in stressed and pretonic syllables (instead of stressed and posttonic). We must assume that the premigratory location of these two dialects was near the western border of the Štoji area, not far from the Dolinci. 3.5 Postmigratory areal diffusion 3.5.1 After the migration many individual Burgenland Croatian dialects have been taking over linguistic characteristics from neighbouring BC groups (including the “single-village groups” Chorvátsky Grob and Weingraben). Examples: Kajkavian from the neighbouring Haci and Poljanci dialect of Kópháza (e.g. several instances of i-reflexes of jat such as crikfa, see Houtzagers 1996: 123) and vice versa (choice between lj and j in judi ‘people’, see Houtzagers, forthc.), the Dolinci dialect of Kaisersdorf from neighbouring Weingraben (e.g. N78M10), Devínska Nová Ves from Chorvátsky Grob (see 3.3.2 (b)). Instances of postmigratory areal diffusion on a larger scale are primarily found in the southern part of the Burgenland, in the area occupied by the Vlahi, Štoji and Southern ƀakavians. By far the most instances of areal spreading took place in the eastern half of that area, mostly involving the Štoji and the eastern half of the Southern ƀakavian dialects. The characteristics concerned can be either phonological, morphological or lexical. The direction of the spreading was almost always north to south. For example, the interrogative pronoun što (Vlahi and Štoji) spread to the south to such an extent that nowadays the majority of the Southern ƀakavian villages has što (N78M1). Other examples can be seen on Maps 14 and 15 discussed in 3.3.2 above. Similar areal diffusion is found on N78M10, 21, 24, 25, 26, 27, 32, 35, 42, 46, 47, 48, 50. Only in the case of menom vs. manom (N78M9) there is reason to believe that the influence went from south to north (from Southern ƀakavians to Štoji). There is a possibility that the 36

Devínska Nová Ves and Chorvátsky Grob are not very near to each other, but from Map 1 it can be seen that in the past there were Croatian villages in between.

314

PETER HOUTZAGERS

same happened in the case of the word for ‘corn’ (N78M51, see 3.2.1 above and Map 5 in the present article). There are only a few instances of other patterns of postmigratory areal spreading on a larger scale. In one instance the Vlahi influenced part of the Štoji (lexeme for ‘forest’ N78M41), in another part of the Dolinci took over a lexeme from the Haci and Poljanci (‘kitchen’, N78M44). In the complicated case of the form for ‘pull’ (vliǀ vs. vuǀ vs. vluǀ; see Map 19, N78M19) there is a possibility that for the Southern ƀakavians and the Dolinci vuǀ was the original form and vliǀ spread from all other BC groups, resulting in Southern ƀakavian and Dolinci relict areas with vuǀ and a Dolinci area with intermediate vluǀ. What is not indicated on the map is the fact that Kópháza (the Haci and Poljanci village west of Kajkavian) has both vuǀ (together with one of its neighbours, the only Dolinci village with vuǀ) and vliǀ. Another instance of large scale postmigratory diffusion was discussed in 3.2.3 and shown on Map 8. 3.5.2 What strikes the eye when one looks at the maps in N78 is the internal unity of the Haci and Poljanci dialects. There are three maps (N78M21, 48, 49) on which one or two villages in the extreme north differ from the rest of their BC group, always together with Devínska Nová Ves and/or Chorvátsky Grob. There is one map (N78M31) on which Kajkavian and a few adjoining Haci and Poljanci dialects share the same characteristic (Isg -um in a-stem nouns). What I know from my own data is that Kópháza, a Haci and Poljanci village in Hungary geographically close to Kajkavian, is in a number of respects intermediate between the Haci and Poljanci and Dolinci and shares a few characteristics with Kajkavian (see Houtzagers, forthc.). The explanation for the homogeneity of the Haci and Poljanci group could be its size, in combination with its distance to the other BC groups. 4. Conclusion By far the most maps in Neweklowsky 1978 (37 out of 52) show isoglosses that are probably premigratory, by which I mean that both the existence itself of the isoglosses and the general picture of the sets of dialects divided by them must be assumed to date from before the migration. Only in two cases I think that the general dialect geographical picture is postmigratory: the spread of the desinence -u in the Lsg of o-stems and the loss of the phoneme Ǜ. In 13 cases I am of the opinion that it cannot be established whether it is pre- or postmigratory. In the first part of this article I have argued that Neweklowsky’s proposed premigratory map is a good starting-point for drawing hypothetical premigratory isoglosses. Comparison of the new and the hypothetical old map corroborates in a very convincing way the relative premigratory locations of the BC

ON BURGENLAND CROATIAN ISOGLOSSES

315

groups proposed by Neweklowsky, with the exception of that of Chorvátsky Grob. The patterns of the premigratory isoglosses suggest intensive contacts over a long period. After the migration areal diffusion on a small scale (one or two villages) has taken place over the whole area. Larger-scale spreading of linguistic features has been quite common in the south, notably from the Štoji to the eastern part of the Southern ƀakavian area, and has been much more restricted in other parts of the Burgenland. An example of a large scale diffusion from south to north is the just mentioned spread of the desinence -u in the Lsg of o-stems. It must be noted that the validity of almost everything that has been said above rests heavily on the assumption that the present BC groups more or less correspond to dialect groups showing a comparable kind of similarity before the migration, which is not more than a working hypothesis.37 University of Groningen REFERENCES N78 = Neweklowsky 1978; N78Mx = map nr. x in Neweklowsky 1978 (after page 376) Breu, J. 1970 Brozoviǀ, D. 1963 1970

Die Kroatensiedlung im Burgenland und den anschließenden Gebieten. Wien. “O rekonstrukciji predmigracionog mozaika hrvatskosrpskih dijalekata”. Filologija 4, 45-55. “Dijalekatska slika hrvatskosrpskoga jeziǁnog prostora”. Radovi, Filozofski fakultet Zadar, razdio lingvistiþko-filološki 8, 5-32 (+ maps).

Dobrovich, J. 1963 Volk an der Grenze – Schicksal und Auftrag. Eisenstadt. (n.v.) Ernst, A. 1987 Geschichte des Burgenlandes. München. Hadrovics, L. 1974 Schrifttum und Sprache der burgenländer Kroaten. Wien. Houtzagers, Peter 1996 “The development of the Hidegség and Fertohǡmok vowel system”. In: A.A. Barentsen et al. (eds.), Studies in South Slavic and Balkan Linguistics (Studies in Slavic and General Linguistics 23), 111-142. Amsterdam-Atlanta.

37

There are instances in which there is some proof that the migrants moved from one possession of the same landowner to another (e.g. in the case of the Kajkavians, see Houtzagers 1999: 20-25), but (a) this does not mean that this was always the case; (b) one landowner could have various, non-contingent properties in the Burgenland region; (c) it would be highly coincidental if the boundaries between the properties of the various landowners in the premigratory situation would coincide with bundles of isoglosses between dialects.

316 1999

PETER HOUTZAGERS

The Kajkavian dialect of Hidegség and Fertǡhomok (Studies in Slavic and General Linguistics 23). Amsterdam-Atlanta. forthc. “On the ƀakavian dialect of Koljnof hear Sopron”. In: Jos Schaeken, Peter Houtzagers, Janneke Kalsbeek (eds.), Evidence and Counter-evidence. Linguistic essays in honour of Frederik Kortlandt (Studies in Slavic and General Linguistics 32). Amsterdam-New York. Iviǀ, Pavle et al. (eds.) 1961-62 “Prilog rekonstrukciji predmigracione dijalekatske slike srpskohrvatske jeziǁke oblasti”. Zbornik za filologiju i lingvistiku 4-5, 117-130. Ivšiǀ, S. 1971 “Hrvatska dijaspora u 16. stoljeǀu i jezik Hrvata Gradišǀanaca”. In: S. Ivšiǀ, Izabrana djela iz slavenske akcentuacije (Slavische Propyläen 96), 723-798. München. Jakubinskij, L. 1925 “Die Vertretung des urslavischen LJ im ƀakavischen”. Zeitschrift für slavische Philologie 1, 381-396. Koschat, Helene 1978 Die ǁakavische Mundart von Baumgarten im Burgenland (Schriften der Balkankommission, Linguistische Abteilung 24/2). Wien. Lonǁariǀ, M. 1984 “Podrijetlo gradišǀanskih kajkavaca”. In: B. Finka et al. (eds.), Gradišǀanski hrvati 1533-1983, 121-131. Zagreb. 1988 “Rani razvitak kajkavštine”. Rasprave ZJ 14, 79-104 (+ maps). 1990 Kaj – juǁer i danas. ƀakovec. 1995 “Prostiranje kajkavštine u prošlosti”. Rasprave ZHJ 21, 79-102. Neweklowsky, Gerhard 1969 “Die kroatischen Mundarten im Burgenland. Überblick”. Wiener Slavistisches Jahrbuch 15, 94-115. 1978 Die kroatischen Dialekte des Burgenlandes und der angrenzenden Gebiete (Schriften der Balkankommission, Linguistische Abteilung 25). Wien. 1982 “O kajkavskim osobinama u nekajkavskim govorima Gradišǀa”. Hrvatski dijalektološki zbornik 6, 257-263. Paviǁiǀ, S. 1920 “O govoru u Slavoniji do turskih ratova i velikih seoba u 16. i 17. stoǛeǀu”. RAD JAZU 222, 194-269. 1953 Podrijetlo naselja i govora u Slavoniji (Djela JAZU 47). Zagreb. Sujoldžiǀ, A. et al. 1990 “Gradišǀanskohrvatski govori u meDžusobnom odnosu i prema nekim govorima u Hrvatskoj (Hemmingove mjere sliǁnosti)”. In: G. Holzer (ed.), Croatica - Slavica - Indoeuropaea (Wiener Slavistisches Jahrbuch, Ergänzungsband VIII), 223-238. Wien. Tornow, S. 1989 Burgenländisches Dialektwörterbuch. Die vlahischen Ortschaften (Balkanologische Veröffentlichungen 15). Berlin. Ujeviǀ, T. 1934 Gradišǀanski hrvati (Jeronimska knjižnica 426). Zagreb. Valentiǀ, M. 1984 “Turski ratovi i velike seobe hrvata u XV. i XVI. stoljeǀu”. In: B. Finka et al. (eds.), Gradišǀanski hrvati 1533-1983, 5-15. Zagreb.

ON BURGENLAND CROATIAN ISOGLOSSES

317

APPENDIX: LIST OF MAPS IN N78 1. 2. 3. 4. 5. 6. 7.

8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26.

Interrogative pronoun: ǁa, što or kaj? The reflex of jat: i-, e- or i/ekavian? Reflex of jat in *gnLJzdo. Presence/absence of length oppositions in various positions. Place of the stress: old, retracted, etc. Presence or absence of lengthening of (once) stressed final closed syllables. Presence or absence of diphthongization of (i) old long *lj, *Ǣ and (ii) old short *e, *o with original stress, except in final syllables. Reflex of syllabic r. Reflex of weak jer in Isg *sԔ mӸnom. Reflex of weak jers in *dӸnӸsӸ. Reflex of front nasal in *žNjdӸnԔ. Reflex of front nasal in *žNjtva. Presence or absence of change *ra > re in vrabac, rasti, krasti. Preposition vԔ: u, va [vu in Kajkavian, PH]. Reflex of *vԔ in *vԔzNjti. Accentuation in the present of the verb morati. Presence or absence of -i in mati. Vowel in tepal/topal. Form of the verb ‘pull’: vliǀ [including vleǀ, PH], vuǀ or vluǀ. Presence or absence of initial i- in (i)šli. Presence or absence of the opposition ǁ vs. ǀ. Masculine singular of the l-participle: bէu֛, bil, bi, bi(j)a. Retention or loss of the phoneme Ǜ as opposed to j. Presence or absence of prothetic j-, DŽor dž- before initial i-: iskat, jiskat, etc. Presence or absence of prothetic v- in uǁiti. Presence or absence of initial š in štil.

27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52.

Final -m > n? Comparative of velik: veǀi or vekši? Ordinal numeral treti or treǀi? Lsg ending of o-stems: -i or -u? Isg ending of a-stems: -u, -um or -om? Lpl ending of neuter pluralia tantum: na vrati(h) or na vrato(h)? Po nǖmški or po nǖmšku ‘in German’? Enclitic accusative of neuter personal pronoun: je or ga? Habitual past: oni su tili povidat or oni su mogli povidat? Form of the word ‘ear’: uho or (v)ušeto/(v)ušeso/šeto? Pri nas or kod nas? ‘Always’: (v)senek or (u)vik? ‘He goes’: gre or ide? Does the word poredan exist? The word for ‘forest’: loza, gora, lug, kiǀe or grmlje? The word for ‘cock’: pLJtLJh or pLJvac? The word for ‘Monday’: pandiǛak/ pondiǛak or prvi dan? The word for ‘kitchen’: kuhnja, veža, krepljet? The word for ‘lazy’: lLJn, taman/tamal or maǝi? The word for ‘spring’: protuliǀe, prmaliǀe/premaliǀe, proliǀe? The word for ‘wedding’: pir, veseǛe or svadba? The lexeme for ‘dog’: kucak, cucak, pas? The word for ‘harrow’: brana or zubatka/zubatica? The word for ‘acre (parcel of arable land)’: lapat or poǛe? The lexeme for ‘corn’: teg or žito/žitak? The word for ‘farmer’: paur, or pugar/pogar?

318

PETER HOUTZAGERS SUMMARY

In the present article the isoglosses in the Burgenland Croatian dialect area and hypothetical isoglosses in the old homeland of the Burgenland Croats are discussed from a dialect geographical point of view. The author has three purposes: (1) to verify existing theories about the provenance of the Burgenland Croats, especially that of Neweklowsky (1978); (2) to establish the chronology of a number of changes in Burgenland Croatian; (3) to shed some light on the dynamics of postmigratory areal diffusion within Burgenland Croatian.

ON BURGENLAND CROATIAN ISOGLOSSES

Map 4a Presence or absence of initial š in štil 4a: within the isoglosses the l-participle of *xԔtLJti is štil, elsewhere til til 4b: štil;

319

Map 4b

320

PETER HOUTZAGERS

Map 5a Lexeme for ‘corn’ 5a: area 1 žito/žitak, area 2 zrnje, elsewhere teg 5b: žito/žitak; zrnje; teg

Map 5b

321

ON BURGENLAND CROATIAN ISOGLOSSES

Map 6a Reflex of *Nj in *žNjtva ‘harvest’ 6a: within the isogloss žatva, elsewhere žetva; 6b:

Map 6b žatva;

žetva

322

PETER HOUTZAGERS

Map 7a Reflex of *Nj in *žNjdӸnԔ ‘thirsty’’ 7b: 7a: within the isoglosses žedan, elsewhere žajan;

Map 7b žedan;

žajan

ON BURGENLAND CROATIAN ISOGLOSSES

Map 8a Lsg ending of o-stems: -i or -u? 8a: area 1 -u, area 2 -i, elsewhere -i/-u; 8b: -u; -i;

323

Map 8b -i/-u

324

PETER HOUTZAGERS

Map 9a Original final *-m: > -n or > -m 9a: within the isoglosses -n, elsewhere -m; 9b:

Map 9b -n;

-m

ON BURGENLAND CROATIAN ISOGLOSSES

Map 10: corrected version of Map 4b

Map 11: corrected version of Map 5b

Map 12: corrected version of Map 8b

Map 13: corrected version of Map 9b

325

326

PETER HOUTZAGERS

Map 14a (v)ušeto/(v)ušeso/šeto or uho ‘ear’ 14a: within the isoglosses vušeto, etc., elsewhere uho; 14b: vušeto, etc.;

Map 14b uho

ON BURGENLAND CROATIAN ISOGLOSSES

Map 15a pri nas or kod nas 15a: within the isoglosses pri nas, elsewhere kod nas; 15b:

pri nas;

327

Map 15b kod nas

328

PETER HOUTZAGERS

Map 16a Final -l in the l-participle 16a: areas 1 biu֛, area 2 bi(j)a, areas 3 bi, elsewhere bil 16b: biu֛; bi ( j)a; bi; bil

Map 16b

ON BURGENLAND CROATIAN ISOGLOSSES

Map 17a Preservation or loss of *Ǜ 17a: unnumbered isoglosses: judi, isogloss 1: judi/Ǜudi, elsewhere Ǜudi 17b: judi; judi/ Ǜudi; Ǜudi

329

Map 17b

330

PETER HOUTZAGERS

Map 18a Diphthongization of *e and *o only when originally long? Map 18b 18a: S of line 1: only when long; N of line 1: not only when long; areas 2: no diphthongs 18b: only when long; not only when long; no (rising) diphthongs

331

ON BURGENLAND CROATIAN ISOGLOSSES

Map 19a Form of the word for ‘pull’ 19a: areas 1 vuǀ, area 2 vluǀ, elsewhere vliǀ/vleǀ; 19b: vuǀ;

vluǀ;

Map 19b vliǀ/vleǀ

Dutch Contributions to the Fourteenth International Congress of Slavists, Ohrid: Linguistics (SSGL 34). Amsterdam – New York: Rodopi, 2008, 333-348.

PAST TENSE HABITUAL EXPRESSIONS IN OLD CROATIAN TEXTS JANNEKE KALSBEEK

1. Past tense habitual expressions in Slavic It is a well-known fact that besides the many similarities, there are considerable differences between the aspectual systems of the individual Slavic languages. Dickey concludes that, on the basis of systematic differences observed in the behaviour of the imperfective and perfective aspects in the individual Slavic languages, “the Slavic languages pattern into two distinct groups, a western group consisting of Cz, Sk, Sor and Sn, and an eastern group consisting of Ru, Uk, Br and Bg” (Dickey 2000: 287). Moreover, “two languages, SC and Pol, show an intermediate patterning, and are thus best viewed as transitional zones between the two groups” (ibidem). 1 This patterning manifests itself clearly in the differences between the Slavic languages in the expression of repeated actions in the past, where so-called unbounded repetition is concerned. 2 Several studies have pointed to the fact that especially in these cases there exists a broad east-west division of Slavic, wherein preference for the imperfective aspect (imperfective preterite) figures as a typical feature of the eastern half of the Slavic language area. Russian and Polish tend to use the imperfective preterite, while Bulgarian normally uses the imperfective imperfect (the perfective imperfect does occur, but “only in habitual-correlative constructions, and never in single main clauses”, Dickey 2000: 76). On the other hand, the most western Slavic languages, Czech, Slovak and Slovene, frequently use the perfective preterite, especially when repetition is expressed by an adverbial or is otherwise obvious from the context. 3 In this respect the C/B/S continuum not only “shows an intermediate patterning”, as Dickey concludes (2000: 287), but also constitutes a transitional area by its internal geographical divergence, which is also the tendency Dickey 1

Here, of course, Cz stands for Czech, Sk for Slovak, Sor for Sorbian, Sn for Slovene, Ru for Russian, Uk for Ukrainian, Br for Belarusian, Bg for Bulgarian, SC for Serbo-Croatian, Pol for Polish (Dickey 2000: 2). 2 “[I].e. cases of repetition in which the number of repetitions is not limited” (Dickey 2000: 50). 3 Cf. besides Dickey (2000), especially Galton (1976), Iviǀ (1983), Mønnesland (1984) and Barentsen (2008); see for further references Kalsbeek and Luǁiǀ (forthc.).

334

JANNEKE KALSBEEK

observes (2000: 87, fn. 31). 4 Whereas the western part of this area (kajkavian dialects, and the north-western part of the ǁakavian dialect area) tolerates perfective preterite referring to repeated events in the past, like Slovene or Czech, the eastern part favours imperfective preterite.5 Apart from the use of perfective preterite, a second striking correspondence in western aspectual features between the north-westernmost dialects of Croatian and the western group of Slavic aspect is the occurrence of specific habitual/iterative verbs, which is also a well-known feature of Czech (e.g. Žminj hiǀievբt ‘throw regularly, be in the habit of throwing’, beside hզtati and hզtitp, cf. also Kalsbeek 1998: 179).6 In addition, most of the C/B/S continuum distinguishes itself from almost all other Slavic languages by the use of modal forms and constructions for the denotation of repeated actions in the past, among which the use of the conditional is on the one hand the most common, and on the other forms a more or less unique feature within the Slavic languages (the possibility of using the conditional, beside the imperfective preterite, Croatian, Bosnian and Serbian share with Upper Sorbian only). 7 Interestingly, Macedonian, too, can use a modal construction to express a repeated event in the past: the particle ԀӾ plus imperfect. “When ѱѦ/Ԥe is added to the imperfect […] it forms the future-in-the-past (anterior future), the expectative unfulfillable (irreal) conditional or the past iterative: ѱѦ ҦҸҎҦѦҗѦ/Ԥe dojdeše can be translated ‘he will have come’, ‘he would have come’, or ‘he would come’” (Friedman 2002: 270).

4

In the following pages, I will use the term “C/B/S continuum” referring to “the region in which Croatian, Bosnian and Serbian are spoken”. 5 See also Iviǀ (1983: 50-51, fn. 30). 6 See for Czech e.g. Danaher (2003). 7 Apart from the conditional, several other habitual expressions occur in the C/B/S continuum, e.g. znati or umjeti/umeti plus infinitive or plus da + present (cf. Hellman 2005). In some dialects, užբt + infinitive is used (e.g. Žminj (Istria), cf. Kalsbeek 1998: 287). Both constructions are used to denote habituals in the past as well as extratemporal habituals. Furthermore, the construction šǀaše plus infinitive, the so-called Balkan conditional, can have an iterative/habitual function in some Montenegrin dialects, cf. Belyavski-Frank (2003: 33 ff.), in some Bosnian dialects a habitual imperative occurs (simple imperative or bi + imperative, cf. Valjevac 2002: 240), and in some Montenegrin dialects šǀaše plus imperative occurs (Sladojeviǀ 1953: 222). The conditional and the construction šǀaše plus the infinitive (or plus da + present) share several functions, especially that of the counterfactual past conditional, habitual/iterative in the past, and attenuated modal expression (“softened volition”), cf. Belyavski-Frank (2003: 31-36). Apart from these functions, the conditional covers the functions of “potential, including gnomic and non-past potentials (but not the future-in-the-past), subjunctive expressions, optative expressions, […] and the non-past conditional. There is one invariant component in the above list: potentiality” (Belyavski-Frank 2003: 22).

PAST TENSE HABITUAL EXPRESSIONS IN OLD CROATIAN TEXTS

335

All Slavic languages share the possibility of usage of the imperfective indicative referring to repeated events in the past in which completion of the individual events in question is irrelevant. Nevertheless, in the denotation of repeated complete events in the past, the individual Slavic languages differ considerably. These differences concern the problem of maintaining or ignoring the explicit expression of the completion of the individual events (i.e. at the microlevel). The following fragments from translations of Antoine de Saint-Exupéry’s Le petit prince 8 (1a) into Russian (1b), Czech (1c), Croatian (1d) and Macedonian (1e), respectively, illustrate the general east-west division of Slavic as far as past tense habituals are concerned: (1a) Quand j'en rencontrais une qui me paraissait un peu lucide, je faisais l'expérience sur elle de mon dessin numéro 1 [que j'ai toujours conservé. Je voulais savoir si elle était vraiment compréhensive.] Mais toujours elle me répondait: “C'est un chapeau.” (1b) ҘҸҊҦҽ Ҿ ӊҡҲҝѦҍҽґi ӊӉҝҸҡґҸҊҸ, ҔҸҲҸҝӃҮ ҔҽӉҽґҡҾ ѸӁѦ ҝҽӉҏѸӁѦҮ ѹ ҰҸӁҾҲґѹӊѦҮ ҦҝҏҊѹѽ, Ҿ ҰҸҔҽӉӃӊҽґi ѦѸҏ ҡӊҸҮ ҝѹҡҏӁҸҔ ҥ 1 […] ѢҸ ӊҡѦ ҸӁѹ ҸҲӊѦҍҽґѹi ѸӁѦ: “ҙҲҸ җґҾҰҽ.” (1c) Když jsem mezi nimi potkalp nLJkoho, kdo se mi zdál trochu bystrý, ovLJǥilp jsem si na nLJm svou zkušenost s kresbou ǁíslo 1 […] Ale vždycky mi odpovLJdLJlp: “To je klobouk.” (1d) Kada bih, meDžu odraslima, naišaop na nekoga tko mi se ǁinio bistrijim, provjeravaoi bih ga pokazujuǀi mu svoj crtež broj 1 […] No, uvijek bi odgovaralii: “To je šešir.” (1e) ҘҸҊҽ ѱѦ ҡҝѦҲӁѦӊp ӁѦҔҸҎ җҲҸ ѹӉҊґѦҦҽҗѦ ѸҽґҔҏ ҍҏҦѦӁ, ҡѦ ҸқѹҦҏӊҽӊi ҔҽҎ ӁѦҊҸ ҡҸ ѸҸҎҸҲ ҃ҝҲѦң қҝҸҎ 1 […] ѢҸ ҡѦҔҸҊҽҗ Ѹѹ ҸҦҊҸӊҽҝҽҽi: “ѷҸҽ Ѧ җҽҰҔҽ.” In this example, Russian (1b), as a representative of the eastern group, uses the imperfective preterite (so that the completion of the individual events remains implicit). On the other end of the scale, Czech (1c), as a typical representative of the western group, uses the perfective preterite (so that completion of the event at the microlevel is explicit, while repetition is obvious from the context). Macedonian (1e) uses ԀӾ plus perfective imperfect with iterative meaning in the 8

These examples are taken from the Amsterdam Slavic Parallel Aligned corpus (ASPAC) of A.A. Barentsen. I would like to thank Adriaan Barentsen for making this material available to me. This contribution has benefited a lot from comments especially of Adriaan Barentsen, but also of the other members of the “Amsterdam Aspect Circle”, Magda van Duijkeren-Hrabová, René Genis and Radovan Luǁiǀ.

336

JANNEKE KALSBEEK

temporal clause, and imperfective imperfect in the two other cases. The Croatian translation (1d) uses the conditional, but in the main clauses of (1d), imperfective preterite would also be possible. Instead of “provjeravao bih”, the imperfective indicative “provjeravao sam” would be an alternative, and instead of “uvijek bi odgovarali”, “uvijek su odgovarali”. In temporal clauses in which the completion of each individual event has to be stressed because there is a sense of succession between the events of temporal clause and main clause, like the temporal clause in (1d), imperfective preterite forms are not found (cf. Barentsen 2008). Barentsen argues that dependent clauses like “Kada bih (…) naišao na nekoga (…)” should be seen as a specific position in which the rules defining the usage of verbal forms are different from those defining the usage of verb forms in main clauses and independent sentences; a complete event preceding the action of the main clause requires a perfective verb, which in the case of repeated past events means selecting a perfective conditional. In dependent temporal clauses like “Kada bih (…) naišao na nekoga (…)” (1d), the (perfective) indicative does not form an alternative, as it would have the interpretation of a single event only. On the other hand, in main clauses and in independent sentences, perfective conditional is generally possible: in (1d), the perfective conditionals “provjerio bih ga (…)” and “uvijek bi odgovorili” would be acceptable. The function of the perfective conditional to denote repeated complete events in the past is evident, particularly in dependent clauses, where it typically cannot be replaced by an imperfective indicative. In independent sentences and main clauses, the conditional can usually be replaced by an imperfective indicative, but in general, the main clause more typically contains a perfective conditional in this type of sentences, too). 9 All in all, the perfective conditional is the most obvious form to refer to repeated complete events in the past. On the whole, the exact function of the perfective or the imperfective conditional versus the imperfective indicative referring to repeated past events in Croatian (and Bosnian and Serbian) has not yet been fully investigated (but cf. Kalsbeek and Luǁiǀ forthc. for some preliminary observations). Within Slavic, it is evidently an innovation. In many respects it resembles one type of usage of English “would”; one English translation of example (1a) is (1f): (1f)

9

Whenever I met one of them who seemed to me at all clear-sighted, I tried the experiment of showing him my Drawing Number One, which I have always kept. I would try to find out, so, if this was a person of true understanding. But, whoever it was, he, or she, would always say: “That is a hat.”

Information extracted from ASPAC by A. Barentsen (personal communication).

PAST TENSE HABITUAL EXPRESSIONS IN OLD CROATIAN TEXTS

337

However, the English construction is not as frequent as the C/B/S one, and besides it does not occur in dependent clauses (which is confirmed by the example quoted above), whereas in C/B/S the habitual conditional is more frequent in dependent clauses.10 In Kalsbeek and Luǁiǀ (forthc.) a tentative attempt was made at an inventarization of factors which possibly play a role in selecting a conditional instead of an imperfective indicative. Among these factors are: • • • •

the aspect of the verb (the conditional of perfective verbs in this function is more frequent than the conditional of imperfective verbs); the occurrence in independent clauses or sentences or in dependent ones (the habitual conditional is more frequent in dependent clauses); explicit or implicit external conditions which “trigger” the event expressed by the conditional; the type of repetition in question, expressed by adverbs of frequency (the habitual conditional does not occur in combination with adverbs like “constantly”, but is on the other hand very common in combination with adverbial like “every day” or “sometimes”).

Thus, on the one hand western South Slavic (i.e. the north-western dialects of the C/B/S continuum, versus the rest of the continuum including the standard languages) forms part of the general east-west division of Slavic as a whole, on the other hand, the south-western “corner” of Slavic as a whole (i.e. the C/B/S continuum, plus Macedonian) uses de-modal forms to express iterative events in the past. This usage of de-modal forms is clearly an innovation. Early stages of Croatian continue the use of the perfective imperfect referring to repeated complete actions in the past which is found in Old Church Slavonic, as well as in Old Russian and Old Czech (see further section 2 on the function of the perfective imperfect in Old Croatian). 11 It has been retained (almost?) to this day in the ǁakavian dialects of the Kvarner area in the north-west (see Kušar 1894 12 on the dialect of Rab; Hamm, Hraste and Guberina 1956 on the dialect of Susak etc.; Houtzagers 1991 on the dialects of Pag). To the east of the C/B/S continuum, Bulgarian (in dependent clauses) and Macedonian still preserve the perfective

10

Information extracted from ASPAC by A. Barentsen (personal communication). An analysis of the meaning of the perfective imperfect (and of the remaining imperfect and aorist forms) in Codex Zographensis is given by Hercigonja (1961). 12 “Imperfektom glagola perfektivnih (tako kako i imperfektivnih) kazuje se, da ǝešto ǁesto biva svaki put u ǝeko vrijeme, kad to vrijeme doDže, na pr. kakը ostգvljaše, tako najdգše (kako bi ostavio, tako bi našao)” (Kušar 1894: 49). 11

338

JANNEKE KALSBEEK

imperfect with this meaning, although in Macedonian it is usually accompanied by the modal particle ԀӾ. Besides its function of past tense habitual expression, the perfective imperfect also had the function of referring to past counterfactual (irreal) events. When the perfective imperfect was lost, in the vast majority of dialects of the C/B/S/ continuum, its habitual/iterative function was apparently taken over by another modal expression which already shared its function of referring to past counterfactual events: the conditional. Apart from this shared function, there is another important clue that leads one to the conclusion that the conditional took the place of the perfective imperfect: both the conditional and the (perfective) imperfect frequently occur/occurred in habitual-correlative sentences (repeated sequences of (mainly complete) events), of the type “when …, then …”, or “whoever …, he …”. In sequences of this type, the conditional and the perfective imperfect frequently occur both in the dependent clause and in the main clause (see section 2 for examples of the perfective imperfect in this type of construction). The conditional with the function of denoting a past tense habitual is found already with certainty e.g. in the works of fra Nikola Lašvanin (first half of the eighteenth century), cf. example (2) below. In example (2), the cases (a), (b), (c), (d), (e), (g) and (h), all with perfective conditionals, can be seen as habitualcorrelative; in each case, one event is conditioned by another one (in (a), (b) and (d), past counterfactual conditions shade into concessivity; (h) can be seen as purely habitual-correlative). However, the independent sentence (f), which functions without an explicitly mentioned external condition, represents an iterative expressed by a conditional: (2)

“1690. […] Bijaše na Bosni Husejin-paša. […] Iste godine pade snig i mraz na žita. I bi glad koga nije nitko zapamtio. […] Pomr mlogi narod od glada. A bižanija bižaše od Save prid vojskom cesarovom. Kud god bi se makǢp (a), ležahu mrt/va/ci: nit se kopahu, nit imadijaše tko. […] u Banjoj Luci, koga bi obisilip (b), obnoǀ bi ga gladni ljudi svega izilip (c). A u to vrime paša sicijaše i višaše i uskoke i raju, koga god bi dovelip (d), i te bi ljudi mrt/va/ce sve izilip (e). I davali bismo (f) jist ubogim, ali kako bi se najilop (g), ta/ko/ bi i umrlop (h).” Lašvanin (2003: 196-197). ‘1690. […] Husejin-paša ruled in Bosnia. […] In the same year snow and ice fell onto the grain. And there was a famine like no one remembered. […] Many people died from famine. And refugees fled from the Sava before the emperor’s army. Where ever one would turn, there were dead men lying: neither were they buried, nor was there any one to bury them. […] in Banja Luka, whoever they hanged, at night the hungry people would eat them. And at that time the paša slayed and hanged both Uskoks and Christian subjects, whomever they brought to him, and those

PAST TENSE HABITUAL EXPRESSIONS IN OLD CROATIAN TEXTS

339

dead men, too, the people would eat. And we used to feed the poor, but as they ate, they died too’. In all cases within example (2), some consequence is brought forward by some external condition (which may or may not be explicit): one event “triggers” another, so that habitual-correlative events appear. However, at what point in history the conditional extended its meaning to the semantic function of habitual/iterative, and took over the role of the perfective imperfect in this function (still quite common in the sixteenth century, see section 2) is not completely clear. It is difficult to get hold of information on the usage of verb forms in Old Croatian texts. For attestations of a conditional, or of a perfective imperfect, with past habitual function, a specific type of narrative texts with generalizations is needed; besides, the majority of studies on the language of Old Croatian texts focuses on phonological and morphological characteristics, and the semantic function of e.g. verb forms is rarely touched upon. Definitive conclusions will therefore have to wait until more research on the usage of verb forms in these texts is done. However, by way of a first small pilot study, in section 2 I will take a preliminary look at some characteristics of the usage of the perfective imperfect in Old Croatian texts (sixteenth century), especially texts by Marko Maruliǀ (Split, 1450-1524), and at the use of the conditional in these texts, particularly at the occurrence of the conditional referring to unbounded repetition. Maruliǀ’s texts belong to the well-documented writing tradition Vermeer has labelled “Classical ƀakavian”. 13 In these texts the imperfect of both imperfective and perfective verbs is still in use. 2. Verb forms expressing repeated past events in Old Croatian 2.1 The imperfect As was pointed out already by Rešetar in his analysis of the language of the Primorski lekcionari, “[u] pogledu p e r f e k t i v n i j e h glagola udara u oǁi što se od ǝih može graditi i imperfekt” (1898b: 192-193). 14 Rešetar quotes the observa13

“[…] the practice of using the Latin alphabet for producing texts on a coastal or insular ƀakavian basis […]. It will be referred to here as “Classical ƀakavian” or “Clƀak”” (Vermeer 1996: 255). “Like the contemporary written language of Dubrovnik, Clƀak was a forwardlooking, vernacular-based tradition. Clƀak texts constitute a major source of information about the language actually spoken in Dalmatia from the fourteenth to the seventeenth century.” (ibidem). 14 Two of the texts Rešetar (1898a and 1898b) analysed, Zadarski lekcionar (fifteenth century) and Lekcionar Bernardina Spliǀanina (1495) belong to the Classical ƀakavian tradition. The third, Ranjinin lekcionar or Ranjinin zbornik (1508, but compiled from texts from the second

340

JANNEKE KALSBEEK

tions by Jagiǀ on the iterative meaning of the perfective imperfect in Marijnskoe ǁetveroevangelie (1883) and on the occurrence of comparable usage of this form in Old Church Slavonic, Old Russian, and Old Croatian and Czech, and concludes: “[s]pecijalno za naš jezik može se slobodno kazati da imperfekt perfektivnijeh glagola (u starije doba) naznaǁuje djelo koje se u prošlosti ponavǛa, pa na taj naǁin ovaki imperfekt odgovara sasvim dobro sadašǝemu vremenu perfektivnijeh glagola koje, kako je poznato, naznaǁuje takoDžer djelo koje se u sadašǝosti ponavǛa”.15 Judging by Rešetar’s examples, this type of usage appears to be frequently referring to habitual-correlative events in sentences of the type: (3)

i ne vraǀahu se kada pojdihu ‘and they did not come back when they went away’ (Zadarski lekcionar 59, Rešetar 1898b: 192)

(4)

i koliko godi jih se dotakniše ozdrafǛeni bihu ‘and as he touched them they were cured’ (Lekcionar Bernardina Spliǀanina 27, ibidem)

(5)

i ki najprija sliziše u lokvu […] zdraf budiše ‘whoever then first stepped in the pool […] was made well’ (Lekcionar Bernardina Spliǀanina 32, ibidem)

As one would expect in view of the general east-west division of Slavic in the expression of repeated complete events in the past (see also section 1), the modern Croatian translation of example (5) (from the Gospel of John, the Pool of Bethesda) uses the habitual conditional of perfective verbs: tko bi prvi ušao […], ozdravio bi, whereas the Russian translation uses an imperfective indicative: Ӡ ӪԽӹ ԗӾӢӴՂӡ ӴԖӹԌӠԊ Ӵ ӰӾӾ (i.e. Ӵ ӪӫԗԂԊӸӰԇ, J.K.) […] ԽӹԽ ӴՂԒԌӹӢԂӴԊӠӴԂԊ. Rešetar adds that the perfective imperfect with a similar function also occurs in texts by e.g. Maruliǀ, Menǁetiǀ and Vetraniǀ (1898b: 193). Vaillant mentions the same construction for Zlatariǀ’s work: “L’imparfait se tire exceptionnellement de verbes perfectifs, et nous en trouvons 4 exemples chez Zlatariǀ: proDžaše […], ponikǝah […], umraše […], prosjaše […]. Ces imparfaits ont une valeur spé-

half of the fifteenth century; it also contains poems by Š. Menǁetiǀ and Dž. Držiǀ, Rešetar 1898a: 81-2) is from Dubrovnik, so it represents the written language of Dubrovnik (which, despite the differences between these two writing traditions, also had much in common with Classical ƀakavian). 15 In modern Croatian, the factors which apparently play a role in the selection of a perfective present instead of an imperfective present in the expression of extratemporal iterativity really seem to bear a considerable resemblance to the factors which determine the choice of a perfective conditional instead of an imperfective indicative for repeated complete events in the past.

PAST TENSE HABITUAL EXPRESSIONS IN OLD CROATIAN TEXTS

341

ciale: ils indiquent ordinairement la répétition dans le passé, ou bien ils servent à exprimer l’irréel.” (1931: 239).16 In Marko Maruliǀ’s epic poems Judita (printed in Venice, 1521) and Suzana, both imperfective and perfective imperfect forms occur, cf. Mladenoviǀ (1957: 135). Imperfective imperfect (Mladenoviǀ mentions about 70 different forms, ibidem) is used referring to the following types of events: •

•

•

16

past processes or actions which form the background to other past actions, e.g. On podstupi Betuliju, odvrati vodu ka u grad teciše, zdence pridgradske ǁini ǁuvati (Maruliǀ 1970: 39) 17 ‘He approached Bethulia, turned away the water that flowed into the city, and placed a guard upon the wells beneath the city walls.’ (Cooper 1991: 9); repeated complete and incomplete actions in the past, e.g. Ovo je bio Polidektes, kralj od Serifa otoka, ki se okameni kad mu ukaza glavu od Meduze Perzej, sin Jovetov i Danaje; ku glavu tko godi vijaše, zakamenjaše se. (Maruliǀ 1970: 76 fn.) ‘This was Polydectes, the King of the Isle of Seriphus, who did turn to stone when Perseus, the son of Zeus and Danae, showed him the head of Medusa; whose head whoever looketh upon it turneth to stone.’ (Cooper 1991: 91); a druzi služahu vino iz bokare,/ ǁrljeno livahu u zlate pehare. (Maruliǀ 1970: III 107 ff.) ‘Others served them wine from a pitcher,/ Pouring crimson into golden goblets.’ (Cooper 1991: 61); irreal conditions, e.g. Bolje ti mu biše u boj smart prijati,/ još kada imiše sa Rimnjani rati,/ nere živu stati do stotin lit roka/ pak konac imati tolika priroka. (Maruliǀ 1970: V 107 ff.) ‘It would have been better for him to die in battle/ Whilst he made war on the Romans,/ Than to live to his hundredth year/ And find such a shameful end.’ (Cooper 1991: 107).

Mavro Vetraniǀ (Vetranoviǀ, 1482-1576), Šiško Menǁetiǀ (1457-1527) and Dominko Zlatariǀ (1558-1613) are all poets from Dubrovnik., so, like Ranjinin lekcionar (which also contains poems by Menǁetiǀ), they do not belong to the Classical ƀakavian writing tradition. However, there are many similarities between their language and the language of sixteenth century poetry (Maruliǀ, Hektoroviǀ, Zoraniǀ) and prose (Maruliǀ, Zoraniǀ) within the Classical ƀakavian tradition (Maruliǀ’s epic poem Judita also contains prose sections; Zoraniǀ’s novel Planine “[i]n addition to prose […] contains more than 1600 lines of verse”, Vermeer 1996: 260). 17 For the purpose of this contribution, I used modern (and accessible) editions of the sixteenth century texts, as it focuses on syntactic and semantic features, not e.g. phonological details. In the references to Maruliǀ’s work, Roman numerals refer to books of Judita, Arabic numerals refer to lines; Arabic numerals without Roman ones refer to pages in the 1970 edition (prose sections and footnotes); S refers to Suzana.

342

JANNEKE KALSBEEK

Perfective imperfect forms (Mladenoviǀ 1957: 135 mentions 12 different cases) 18 have habitual/iterative meaning, they indicate repeated complete events, e.g. (6)

na zemlji padašei ptica sa visine,/ kad zavapijašep vojska iz dubine. (Maruliǀ 1970: I 291-2) ‘Birds fall to the ground from on high/ When e’er the army letteth loose with a shout.’ (Cooper 1991: 31)

(7)

Jer u zakonu ockvrnjen bišei ki se taknišep mrtva ǁlovika; pop jih oǁišǀašep kropeǀi vodom u koj biše lug krave ǁrljene […] (Maruliǀ 1970: 95 fn.) ‘For in the law he is defiled who toucheth the garments of a dead man; the priest did purify them, sprinkling them with water in which was the ash of a red heifer […]’ (Cooper 1991: 129, fn. 12)

(8)

Mojses na goru moleǀi stašei; kada dvignišep ruke, dobivašei puk njegov, kad jih spušǀašep, dobivahui neprijatelji. (Maruliǀ 1970: 55 fn.) ‘Moses stood in prayer on the mountain; when he raised his arms, his people prevailed, and when he lowered them, the foe prevailed.’ (Cooper 1991: 47 fn. 13)

(9)

(a) Nestanišep t’ vode gdino postojahup,/ zato vred na brode prid se popeljahup./ (b) A kad se brojahui, sklopiv moste nike,/ deset dan zbrojahup brodeǀ se prik rike. (Maruliǀ 1970: I 297 ff.) ‘Nestalo bi vode gdje bi oni proboravili, pa su zato što prije kretali dalje preko gazova: kad su sastavivši nekakve skele prelazili rijeku, trajalo je to ǁitavih deset dana.’ (Maruliǀ 1970: 189)19

(10) (a) Niki mrt, niki živ u hlapstvo idiše,/ niki sve izgubiv plaǁan ostanišep;/ (b) opet jih vrnišep Bog u prvo stan’je,/ kad jih vidiše ǁineǀi kajan’je. (Maruliǀ 1970: II 281 ff.) ‘Some died, others went into captivity;/ Others, losing everything, remained to weep;/ Once again God returned them to their former state/ When He saw them truly repenting.’ (Cooper 1991: 51)

18

It is occasionally difficult to determine the aspect of a verb form in these texts (superscript i denotes “imperfective”, superscript p “perfective”); most verbs in -iti have 3sg imperfect in -aše, so oǁišǀaše in example (7) could be 3sg imperfect from perfective oǁistiti (cf. poǀaše se from potiti se), or 3sg imperfect from imperfective oǁišǀati (? JK). Similarly, spušǀaše in example (8) could be perfective or imperfective imperfect. In example (10): vidjeti is biaspectual in modern Croatian, Maruliǀ has a 3sg imperfect form vijaše too, which is clearly imperfective (I 232, S 163, cf. Mladenoviǀ 1957: 135), therefore 3sg imperfect vidiše in example (10) probably has to be interpreted as perfective imperfect. 19 For this example I quote the translation of the first book of Judita into modern Croatian by Ivan Slamnig (Maruliǀ 1970: 183-189) instead of the English translation, because I think it is closer to the original. Besides, in (9a) it illustrates the equivalency of habitual perfective conditional in modern Croatian to the perfective imperfect in Maruliǀ’s language.

PAST TENSE HABITUAL EXPRESSIONS IN OLD CROATIAN TEXTS

(11)

343

A on ti sijašei oholo, visoko,/ a sam pogledašep po vojsku široko; (Maruliǀ 1970: I 233 ff.) ‘And he sat in arrogance, on high,/ And glanced at his troops far and wide’ (Cooper 1991: 29)

(12) Sva mista njih koje godir gdi imihui/ uǁinivši svoje, u miru živihui./ Dobrosriǀni bihui vazda der do vika,/ dokol ne padihup u prezpravdja nika. (Maruliǀ 1970: II 289 ff.). ‘Having conquered their lands where e’er they went,/ They made them their own, and lived in peace;/ And they prospered always, for as long/ As they did not fall into sin.’ (Cooper 1991: 51) (13) motahui t’ ureda skakǀuǀi nogami,/ plešuǀi poreda, zavrg se glavami./ Plaho ti bedrami pojdihup svrtaje,/ razmašuǀ prsami, stegna podžimaje; (Maruliǀ 1970: I 263 ff.) ‘They would twist about stamping their hooves,/ Prancing against one another and tossing their heads./ They moved coyly, shaking their shanks,/ Expanding their chests, pressing their flanks;’ (Cooper 1991: 31) Perfective imperfect forms occur both in dependent and in main clauses, and in independent position. In examples (6), (7), (8), (9a), and probably (10) (vidiše, see footnote 18). The dependent clauses all have perfective imperfect. In the cases (9a) (nestaniše) and (10) (vrniše), the imperfect form in the main clause is perfective, too, as is the imperfect form in the independent clauses in (7) (oǁišǀaše, but see footnote 18), (9) (popeljahu) and (10) (ostaniše). In the majority of cases, cf. examples (6), (7), (8) (in both sequences), (9a) and (10b) above, habitual-correlative actions are involved: there is a kind of cause and effect relation between two events in each case. There is a sense of concessivity present in example (9) (wherever they stopped, the water disappeared). In example (13), a parallel with the usage of the conditional in example (15) can be seen, see section 2.2. 2.2 The conditional Maruliǀ uses the conditional in several modal functions, e.g. • •

20

in final clauses: Ozija, knez od grada, moli jih da bi još ǁekali pet dan pomoǀi božje. (Maruliǀ 1970: 39), ‘Ozias, ruler of the city, entreated them to wait five days more for God’s help.’ (Cooper 1991: 9); with potential meaning, e.g. Tko je toliko smiv ki bi jih doǁekal?/ al nadaleǁ vidiv da se ne bi pripal? (Maruliǀ 1970: I 305-6) ‘Who might be so bold as to await them?/ Or, seeing them from afar, might not faint?’ 20 (Cooper 1991: 33); ‘Might not be frightened’ seems closer to the original.

344 •

JANNEKE KALSBEEK

expressing a counterfactual, irreal event, e.g. mnju ti, bi Apolo lin tirati Dafnu bil,/ tad kon tesalskih stin ovu da bi vidil. (Maruliǀ 1970: IV 139-140) ‘I believe Apollo would not have chased Daphne/ Had he first seen Judith at the crags of Thessaly.’ (Cooper 1991: 85).

I think a first onset to the usage of the conditional with habitual/iterative function is found in cases like: (14) Obrvi ter viju, nad kimi staše val,/ ako reǀi smiju, Bog biše upisal./ Tko bi podaleǁ stal, rekal bi: rumen cvit/ od ust je nje procval ali ružice list. (Maruliǀ 1970: S 33-36) ‘Her eyebrows and eyelids,/ if I may say so, God had drawn./ Whoever would stand a bit further off, would say: a red flower/ bloomed from her mouth or a rose leaf.’ Here, as in the case of the fragment from fra Nikola Lašvanin’s Bosanski ljetopis (cf. section 1, example (2)), we find a rather clear sense of concessivity which seems to weaken the sense of irreality/counterfactuality that is typically associated with the conditional. In this kind of sentences the concessive nuance leads to the thought of various different instantiations of the actions (for instance by different subjects). As soon as it becomes possible to assume that at least some of these instantiations may have occurred in reality, the conditional starts to be associated with real repeated events in the past. In the following example we find a comparison: (15) Suknje bihu svargli, župe pripasali;/ rukavce uzvargli, biǁve podpasali;/ barže t’ bi ticali skaǁuǀi dubravom,/ ner kad bi bigali jelini prid lavom. (Maruliǀ 1970: I 213-216) ‘They had stripped off their garments, girded their shirts,/ Rolled back their sleeves, laced up their boots./ Swifter did they dash through the oak grove on their mounts,/ Than deer fleeing before a lion.’ (Cooper 1991: 27) The conditional in kad bi bigali retains its sense of counterfactuality (the fleeing deer are imaginary). But in bi ticali, a real event is referred to. Maybe the use of the conditional was triggered by parallelism with the dependent clause, but the fact that a conditional obviously can be used to refer to a non-counterfactual event probably has to to with the tendency mentioned in the discussion of example (14) above. The use of the conditional bi ticali to refer to repeated past events in example (15) strongly reminds one of the use of the perfective imperfect pojdihu in example (13).

PAST TENSE HABITUAL EXPRESSIONS IN OLD CROATIAN TEXTS

345

Cases similar to example (14) (i.e. with a concessive sense, in example (16) explicitly expressed by god) are found in Hektoroviǀ’s Riban’je i ribarsko prigovaran’je (1568)21, cf.: (16) A od ruk tvojih teg kî god bi imio/ Knez, ban ali herceg, njim bi se diǁio,/ Kî bi ga vidili mladi ali stari,/ Svi bi ga hvalili, kralji ter cesari. (Hektoroviǀ 1968: 227) ‘And whoever would have the work of your hands/ Be he prince, governor or duke, he would be proud of it,/ Whoever would see it, whether young or old,/ All would praise it, kings and emperors.’ Comparable cases are also found in Zoraniǀ’s22 Planine, cf.: (17) Tko b’ liǁca rumena i mile oǁice i usta medvena/ vidil me Ružice, pustil bi tužice, prenda/ b’ v žalosti bil, radostju odiv lice peǁal/ bi ostavil. (Zoraniǀ 1988: 150) ‘Whoever would see the rosy face and lovable eyes and lips full of honey/ of my Ružica, would let go of sorrow, even if/ he would be troubled, he would let go of grief, joy adorning his face.’ 3. Tentative conclusion In view of the cases in Maruliǀ’s, Hektoroviǀ’s and Zoraniǀ’s works discussed in section 2.2, I think an early onset of usage of the conditional referring to repeated events in the past in the C/B/S language area may possibly be located in sixteenth century Classical ƀakavian. This usage must have become much more common between the sixteenth and eighteenth centuries (cf. the example from Lašvanin’s chronicle in section 1). Besides the fact that the conditional and the perfective imperfect shared the function of referring to counterfactual events, correlative cases, especially with some concessive sense, may have played a decisive role in the extension of the meaning of the conditional to habitual/iterative in the past. It would be interesting to examine this development in relation to several other large structural changes in the verbal systems of the C/B/S continuum that must have taken place somewhere between the fifteenth and eighteenth centuries: in the first place the loss of the imperfect in the entire central and western area (but its retention in the Kvarner area, with the specific habitual/iterative meaning of the perfective imperfect, and retention of the imperfective imperfect in eastern dialects), but also the loss of the perfective present with future meaning in inde21

Petar Hektoroviǀ (1487-1572) is a Renaissance poet from Hvar; his works belong to the Classical ƀakavian tradition. 22 Petar Zoraniǀ (born probably 1508, died before 1569) lived in Zadar. His Classical ƀakavian Planine (printed in Venice in 1569) is the earliest prose novel of the C/B/S language area, cf. Vermeer (1996: 255).

346

JANNEKE KALSBEEK

pendent sentences, possibly connected with the generalization of periphrastic future forms (consisting of the de-modal auxiliary ǀu etc. plus the infinitive in most of the C/B/S area, but (also) with budem etc. plus the infinitive in part of the ǁakavian dialects, and with budem etc. plus l-participle in kajkavian dialects (as in Slovene).23 University of Amsterdam REFERENCES Barentsen, Adrian 2008 “Vyraženie posledovatel’nosti dejstvij pri povtorjaemosti v prošlom v sovremennyx slavjanskix jazykax”. This volume. Belyavski-Frank, Masha 2003 The Balkan Conditional in South Slavic. A Semantic and Syntactic Study (Slavistische Beiträge 421). München: Verlag Otto Sagner. Cooper, Henry R. (ed.) 1991 Judith. Translated from the Croatian by Henry R. Cooper, Jr. (East European monographs 203). New York: Columbia University Press. Danaher, David 2003 The semantics and discourse function of habitual-iterative verbs in contemporary Czech. Munich: Lincom Dickey, Stephen M. 2000 Parameters of Slavic Aspect. A Cognitive Approach. Stanford: CSLI Publications. Friedman, Victor A. “Macedonian”. In: Bernard Comrie and Greville G. Corbett (eds.), The Sla20023 vonic Languages, 249-305. London and New York: Routledge. Galton, Herbert 1976 The Main Functions of the Slavic Verbal Aspect. Skopje: Macedonian Academy of Sciences and Arts. Hamm, J., M. Hraste, and P. Guberina 1956 “Govor otoka Suska”. Hrvatski dijalektološki zbornik 1, 7-213. Hektoroviǀ, Petar 1968 “Bogoljubnoj i svake hvale dostojnoj mladici Graciosi Lovrinǁevi”. In: M. Franiǁeviǀ (ed.), Hanibal Luciǀ, Skladanja izvarsnih pisan razlicih, Petar Hek-

23 In Maruliǀ’s language, the future is usually formed from the auxiliary hoǀu etc. + the infinitive, e.g. Dike ter hvaljen’ja presvetoj Juditi/ smina nje stvore[n]ja hoǀu govoriti. (Maruliǀ 1988: 119). Perfective present with futural function also occurs, but infrequently, e.g. stav stražu, kano pad kon zdenac, ne dade/ vazimat vodu kad ki od njih napade./ Tim ti se pridade vas puk on od volje, / ali se raspade od toke nevolje; […]. The construction budu or budem plus the infinitive with futural function (and very rarely also budem etc. plus the l-participle) is found only in dependent clauses (“Predbuduǀe vreme”, Mladenoviǀ 1957: 139).

PAST TENSE HABITUAL EXPRESSIONS IN OLD CROATIAN TEXTS

347

toroviǀ, Ribanje i ribarsko prigovaranje i razlike stvari ine (Pet stoljeǀa hrvatske književnosti 7), 226-231. Zagreb: Matica hrvatska, Zora. Hellman, M. 2005

Hercigonja, E. 1961

The verbs znati and um(j)eti in Serbian, Croatian and Bosnian: a case study in the grammaticalisation of habitual auxiliaries. Helsinki: Helsinki University Press. “O nekim znaǁenjima osnovnih preteritalnih vremena u jeziku Zografskog kodeksa”. Radovi za slavensku filologiju 4, 87-124

Houtzagers, H.P. 1991 “Imperfekt v ǁakavskix dialektax ostrova Pag”. Sovetskoe slavjanovedenie 5, 77-82. Iviǀ, Milka 1983 “Naǁini na koje slovenski glagol ovremenjuje ponavljanu radnju”. In M. Iviǀ, Lingvistiǁki ogledi, 37-56. Beograd: Prosveta. Kalsbeek, Janneke 1998 The ƀakavian Dialect of Orbaniǀi near Žminj in Istria (Studies in Slavic and General Linguistics 25). Amsterdam-Atlanta: Rodopi. Kalsbeek, J., and R. Luǁiǀ forthc. “Oblik kondicionala u funkciji oznaǁavanja ponavljane radnje u prošlosti”. To appear in: Croatica et Slavica Iadertina 4. Zadar: Sveuǁilište u Zadru. Kušar, M. 1894 “Rapski dijalekat”. Rad JAZU 118, 1-54. Lašvanin, fra Nikola 2003 “/Bosanski ljetopis/. /Od 1682. do 1750./”. In: fra Ignacije Gavran (ed.), Nikola Lašvanin, Ljetopis, 193-233. Sarajevo-Zagreb: Synopsis. Maruliǀ, Marko 1970 (I. Slamnig, ed.) Judita, Suzana, pjesme (Pet stoljeǀa hrvatske književnosti 4). Zagreb: Matica hrvatska, Zora. Mladenoviǀ, Aleksandar 1957 “Fonetske i morfološke osobine Maruliǀevog jezika”. Godišnjak Filozofskog fakulteta u Novom Sadu 6, 89-144. Mønnesland, S. 1984 “The Slavonic Frequentative Habitual”. In: C. de Groot and H. Tommola (eds.), Aspect Bound. A voyage into the realm of Germanic, Slavonic and Finno-Ugrian aspectology, 53-76. Dordrecht etc.: Foris Publications. Rešetar, Milan 1898a “Primorski lekcionari XV. vijeka”. Rad JAZU 134, 80-160. “Primorski lekcionari XV. vijeka (Konac)”. Rad JAZU 136, 97-199. 1898b Sladojeviǀ, P. 1953 “O imperfektu u srpskohrvatskom jeziku”. Južnoslovenski filolog 20, 219-245. Vaillant, André 1931 La langue de Dominko Zlatariǀ, poète ragusain de la fin du XVIe siècle II, Morphologie (Travaux publiés par l’Institut d’études slaves 6). Paris: Imprimerie nationale. Valjevac, Naila 2002 Govor u slivu Lašve (Bosanskohercegovaǁki dijalektološki zbornik 9). Sarajevo: Institut za jezik u Sarajevu.

348

JANNEKE KALSBEEK

Vermeer, Willem 1996 “The twofold origin of Classical ƀakavian”. In: A.A. Barentsen et al. (eds.), Studies in South Slavic and Balkan Linguistics (Studies in Slavic and General Linguistics 23), 255-318. Amsterdam-Atlanta: Rodopi. Zoraniǀ, Petar 1988 Planine. Izvornik i prijevod. (Prijevod i komentari: Marko Grǁiǀ). Zagreb: Grafiǁki zavod Hrvatske.

SUMMARY Considerable differences exist between the Slavic languages with respect to the expression of repeated complete events in the past, where so-called unbounded repetition is concerned. In the eastern zone of Slavic aspect, Russian typically uses the imperfective preterite, while Bulgarian uses the imperfective imperfect. The westernmost Slavic languages, e.g. Czech, can use the perfective preterite when repetition is expressed by an adverbial. Croatian and Serbian frequently use the perfective conditional (besides the imperfective preterite); Macedonian can use a construction consisting of the particle ԀӾ plus perfective imperfect (see Dickey 2000). In Old Croatian texts the perfective imperfect (a form which was lost, except in some Kvarner area dialects) was used to express repeated complete events in the past (see Rešetar 1898). The imperfect and the conditional shared the function of referring to counterfactual events. Apparently, in modern Croatian the conditional took over the role of referring to habitual events which the perfective imperfect had in Old Croatian. This contribution is a preliminary investigation of the usage of the perfective imperfect and the conditional in 16th century Classical ƀakavian texts, and of the first indications of the taking over of the role of the habitual perfective imperfect by the conditional.

Dutch Contributions to the Fourteenth International Congress of Slavists, Ohrid: Linguistics (SSGL 34). Amsterdam – New York: Rodopi, 2008, 349-396.

FROM OVS ORDER TO CONVERSE STRUCTURE IN RUSSIAN AND OTHER SLAVIC LANGUAGES CORNELIA E. KEIJSPER

1. Introduction Nobody who knows Russian is surprised to see sentences like ҶҲҏ ҔӁѹҊҏ ҰҝҸӁѹӉӃӊҽѦҲ ѹҝҸӁѹҾ or ҋҰѦҔҲҽҔґҧ ҸҲґѹҍҽѦҲ ҸҝѹҊѹӁҽґҧӁҸҡҲҧ (ҘҸӊҲҏӁҸӊҽ 1976: 173). Yet, in general linguistics sentences in which the object (ҙҲҏ ҔӁѹҊҏ, ҡҰѦҔҲҽҔґҧ) precedes the subject (ѹҝҸӁѹҾ, ҸҝѹҊѹӁҽґҧӁҸҡҲҧ) are considered to be very exceptional. In 1963 Joseph Greenberg even devoted his First Universal to the issue: “In declarative sentences with nominal subject and object, the dominant order is almost always one in which the subject precedes the object” (1963: 110). Closer to Russian is the approach represented by ҘҸӊҲҏӁҸӊҽ (1976), which allows for the possibility that what Kovtunova calls the “ҡѹӁҲҽҊѸҽҲѹҍѦҡҔѹ ӁѦӉҽӊѹҡѹѸӃҮ ҍґѦӁ ҰҽҝҽҦѹҊѸӃ” is different in different cases: in ҐҝҏӁҾ ӁҸҡѹґҽ ҰҸҍҲҏ or үӁҦҝѦҮ ҡҦѦґҽґ ҔҽҝҧѦҝҏ the “syntagmatically independent member of the paradigm” has SVO order (and the last accent on O, here indicated by underlining), but in ҶҲҏ ҔӁѹҊҏ ҰҝҸӁѹӉӃӊҽѦҲ ѹҝҸӁѹҾ or ҋҰѦҔҲҽҔґҧ ҸҲґѹҍҽѦҲ ҸҝѹҊѹӁҽґҧӁҸҡҲҧ the “syntagmatically independent member of the paradigm” has OVS order (and the last accent on S, here indicated by underlining) (ҘҸӊҲҏӁҸӊҽ 1976: 170-183). This article reviews the linguistic issues supposedly involved in OVS sentences like ҶҲҏ ҔӁѹҊҏ ҰҝҸӁѹӉӃӊҽѦҲ ѹҝҸӁѹҾ or ҋҰѦҔҲҽҔґҧ ҸҲґѹҍҽѦҲ ҸҝѹҊѹӁҽґҧӁҸҡҲҧ, and replaces them by the starting point for investigations of socalled converse structure in Russian and other Slavic languages. Examples are from the Amsterdam Slavic Parallel Aligned Corpus (ASPAC, version March 2007) developed by A.A. Barentsen. For the present article I searched ASPAC mainly on the basis of the Russian originals in it (ca. 555 pages), with some Russian verbs known to occur in OVS order as query term. 1 Other examples are as 1 I used the following verbs that are known to occur in OVS sentences at least sometimes: First, some so-called verbs of emotion (cf. ѨҸҝҦҽӁҡҔҽҾ 1970, Wierzbicka 1992: 401-403, Keijsper 2003: 195-205), viz. ӊҸґӁҸӊҽҲҧ/ӊӉӊҸґӁҸӊҽҲҧ, ѹӁҲѦҝѦҡҸӊҽҲҧ, ҰҸҝҽңҽҲҧ/ҰҸҝҽӉѹҲҧ, ҝҽҦҸӊҽҲҧ/(ҰҸ/)ҸқҝҽҦҸӊҽҲҧ, ҰҏҊҽҲҧ/ѹҡҰҏҊҽҲҧ. Secondly, some so-called ҊґҽҊҸґӃ ҰҸґӁҸҊҸ ҸѽӊҽҲҽ and ҊґҽҊҸґӃ ҡҸѦҦѹӁѦӁѹҾ (ҭҽҦҏҍѦӊҽ ѹ ҩҸӉѹӁҽ 1993, ҐҽӊҝѹґҸӊҽ 1975, Keijsper 2003: 181-184,

350

CORNELIA E. KEIJSPER

far as possible from originals in other Slavic languages translated into Russian (ca. 816 pages Polish, ca. 326 pages Czech, ca. 310 pages in Serbian or Croatian). I used the remainder of the corpus, including small amounts of texts translated into other Slavic languages, to find OVS examples of the verbs mentioned in note 1 if none occurred in the Slavic originals; also, I used the remainder for additional verification of some issues discussed here. I concentrate on Russian, although it will appear from the discussion that progress in the study of other Slavic languages is what we really need next. 2. Word order and constituency as the expression of S and O in English In the general linguistic literature, often based on English, the influence of the English language is so preponderant that the properties of this language are often mistakenly regarded as universal. One example is the expression of subject and object: in the simple case that concerns us here, English word order and related constituency make it clear which noun phrase is the subject and which the object. Consider, for example, Chomky’s (1965: 64 ff.) explanation of his idea that the notions Subject, Predicate, Main-Verb, and Object are relational and already represented in a phrase-marker like the following: S NP

Aux

N

M

sincerity

may

VP

V frighten

NP Det N the boy

195-205): ӊѦӁҍҽҲҧ/ҏӊѦӁҍҽҲҧ, ӉҽӊѦҝҗҽҲҧ/ӉҽӊѦҝҗѹҲҧ, ӁҽҰҸґӁҾҲҧ/ӁҽҰҸґӁѹҲҧ, ҸҔҝҏңҽҲҧ/ҸҔҝҏңѹҲҧ, ҰҸҔҝӃӊҽҲҧ/ҰҸҔҝӃҲҧ, ҡҸѦҦѹӁҾҲҧ/ҡҸѦҦѹӁѹҲҧ, ҏҔҝҽҗҽҲҧ/ҏҔҝҽҡѹҲҧ, plus the verb from the title of the group: ҸѽӊҽҲӃӊҽҲҧ/ҸѽӊҽҲѹҲҧ. Thirdly, some verbs from ҘҸӊҲҏӁҸӊҽ’s (1976: 173-177) examples of verbs with OVS as the or a “syntagmatically independent member of the paradigm” (see further section 5). These examples contain some of the verbs already mentioned, but also, among others, қҏҦѹҲҧ/ҝҽӉқҏҦѹҲҧ, ӊҡҲҝѦҍҽҲҧ/ӊҡҲҝѦҲѹҲҧ, ҸӊґҽҦѦӊҽҲҧ/ҸӊґҽҦѦҲҧ, ҸҲґѹҍҽҲҧ/ҸҲґѹҍѹҲҧ, ҰҸҡѦӂҽҲҧ/ҰҸҡѦҲѹҲҧ. ҫҦҸґѦӊҽҲҧ/ҸҦҸґѦҲҧ represents verbs with an instrumental rather than an accusative object. This list is not intended to be exhaustive; it is also not homogeneous. But it suffices to bring to light the linguistic issues we are dealing with.

FROM OVS ORDER TO CONVERSE STRUCTURE IN RUSSIAN

351

Namely, the Noun Phrase directly under the Sentence node is the subject (“Subject-of” = [NP, S]), the Noun Phrase under the Verb Phrase node is the object (“Direct-Object-of” = [NP, VP]), “Predicate-of” is [VP, S], and “Main-Verb-of” is [V, VP] (1965: 71). On page 69 Chomsky mentions, casually, that his proposal is about English. But already on page 71-72 we learn that “these definitions must be thought of as belonging to general linguistic theory”. This treatment by Chomsky was the prelude to a demonstration of total failure of the Chomskyan tradition to deal with matters of word order and constituency: treatments in this tradition systematically confuse the situation in English with that in other languages, because the fact that other languages, such as Russian, use other means for the expression of subject and object than English is not recognized. In off-springs of the Chomskyan traditions such as Dik (1978, 1989) the failure showed up in enormities like the proposal that (some) Slavic languages have no Subject and/or Object-functions (Dik and Gvozdanoviǀ 1980, Bubenik 1987: 58, Siewierska 1988: 125, Stanchev 1987: 34, 36, cf. Keijsper 1990: 28-35). This is the result of the fact that Dik’s so-called Functional Grammar started from the Greenbergian idea that “in the functional patterns of the vast majority of languages the Subj normally precedes the Obj both in main and subordinate clauses” (Dik 1978: 176), and soon became worse by replacing this idea by “the constituent ordering rules of any given language will be sensitive to Subj and Obj function” (Dik 1980: 18). In reality, of course, Russian and other Slavic languages with case forms have both syntactic hierarchy, with notions such as subject and object, on the one hand, and, on the other hand, word order, constituency, accentuation and intonation (together nowadays called linear-intonational hierarchy – see ҭҽҦҏҍѦӊҽ 1985: 108, ҋѦґѹӊѦҝҡҲҸӊҽ ѹ ҭҝҸӉҸҝҸӊҽ 1992: 204), 2 and they do not express the former hierarchy by means of the latter hierarchy but by means of case forms (nominative, accusative etc.) and verb forms (active, reflexive, -n/-t participle short form, as far as finite verb forms are concerned). There is a huge literature in which the history of the Chomsky disaster can be traced (Keijsper 1994: 297-307 is an introduction). Here, it suffices to realize that Russian sentences like ҶҲҏ ҔӁѹҊҏ ҰҝҸӁѹӉӃӊҽѦҲ ѹҝҸӁѹҾ or ҋҰѦҔҲҽҔґҧ ҸҲґѹҍҽѦҲ ҸҝѹҊѹӁҽґҧӁҸҡҲҧ illustrate precisely the structure [O[VS]] which is im2

The separation of accentuation and intonation, and the inclusion of constituency (with hierarchy) is my own contribution to the study of what I then called Information Structure (Keijsper 1985: 118). The arrows used in Keijsper (1985) for the indication of hierarchy within and between constituents are not the same as those used in ҋѦґѹӊѦҝҡҲҸӊҽ ѹ ҭҝҸӉҸҝҸӊҽ (1992: 204 ff.) for the indication of the “ѽҽҝҽҔҲѦҝѹӉҏѦѸҸѦ” and the “ѽҽҝҽҔҲѦҝѹӉҏӀӂѦѦ”: in my view, these notions do not always coincide with the relevant hierarchy (see, e.g., Keijsper 1985: 58-60).

352

CORNELIA E. KEIJSPER

possible according to the English-based definition of Chomsky cited above, and have the order OVS which is almost non-existent according to typological studies of the Greenberg tradition. 3. Case syncretism The approach to word order represented by ҘҸӊҲҏӁҸӊҽ 1976 is an example of the tradition founded by Vilém Mathesius (1882-1945) which is nowadays common in Slavistic treatments of word order in Slavic languages. It starts from the knowledge that word order (and implied sentence accent placement) in Russian, for example, does not serve to make it clear which parts of the sentence are subject and which object: the syntactic hierarchy is expressed by case forms (nominative, accusative etc.) and by forms of verbs (active, reflexive, -n/-t participle). Linear-intonational hierarchy (word order, constituency, sentence accent and intonation) serves to express the informational content of a sentence, which makes it clear how the sentence must be integrated into its context. The existence of both syntactic and linear-intonational hierarchy creates a wealth of subtle distinctions. It is sometimes suggested that this wealth is restricted by the existence of case syncretism: if the form of a noun does not enable us to detect its syntactic function, is it true that word order information is used instead? Although some authors say that in sentences like ҴҽҲҧ ґӀқѹҲ ҦҸҍҧ the first noun phrase always is the subject and the second the object (e.g. Benoist 1979: 32-34), in reality sentences like ҋҰѦҔҲҽҔґҧ ҸҲґѹҍҽѦҲ ҸҝѹҊѹӁҽґҧӁҸҡҲҧ are recognized as OVS occurrences despite the fact that object and subject have the same form. The discussion in ҘҸҝҸґѦӊ (1966) makes it clear why the issue is hypothetical rather than real: in reality, factors other than case forms resolve the potential homonyms most often even without further context. Consider the following sentence, for example: ғҽңӁҸѦ ӉӁҽҍѦӁѹѦ ѹѸѦѦҲ ҰҝҸ҃Ѧҡҡ ҰҝѦҦӊҽҝѹҲѦґҧӁҸҮ ҡҲҽҲѹҡҲѹҍѦҡҔҸҮ ҸқҝҽқҸҲҔѹ ҲѦҔҡҲҽ. When you read it, you can have no doubt that you are reading an OVS sentence, because the OV combination is phraseological. In other examples the reality referred to is decisive: ҭѹҔҽҡҡҸ ѹӁҲѦҝѦҡҏѦҲ ҝҽҡҔҝӃҲѹѦ ҦҏѽҸӊӁҸҊҸ Ѹѹҝҽ ҍѦґҸӊѦҔҽ or ҈ҽ ӁҸҍҧ ҡҔӊѦҝ ҰҸҔҝӃґ ҡӁѦҊ, for example, also leave no doubt about subject and object functions. Cases like ҴҽҲҧ ґӀқѹҲ ҦҸҍҧ or ҋѦҊҸҦӁҾ ӍѹӉѸҽҲ ҰҝҸӊѦҝҾѦҲ ґѹҲӍҽҔ, which are indeed homonyms without further context, are rare (ҘҸҝҸґѦӊ 1966). One of such rare examples showed up during my searches in ASPAC. It should have drawn the attention of translators to the disambiguating context, but evidently not all translators took the trouble to look that far:3 3

The language of the examples is abbreviated after the number preceding the examples. If more than a single translation into the same language is cited, an abbreviation of the translator is

FROM OVS ORDER TO CONVERSE STRUCTURE IN RUSSIAN

353

(1R)

ҫҰґӃӊҽҾ ҰҸҲҸѸ, ҸӍѹ҃ѹҽӁҲӃ ӁѦҡґѹ ӁҽҦ ҊҸґҸӊҽѸѹ ӉҽҰҸҲѦӊҗѹѦ ҔҝҏңҔѹ ҡ ҰѹӊҸѸ, ѽҝѹҰґҸ ѹ ҡ ӁѦӁҽӊѹҡҲҧӀ Ҕҝѹҍҽґѹ: “ғѹӁҸӊҽҲ, ҊҝҽңҦҽӁѹӁ!” ҐҦѦ-ҲҸ ӊ ҝҏҰҸҝѦ ҊҸґҸҡ ҔҸѸҽӁҦҸӊҽґ: “ҘҽҝҡҔѹҮ ҝҽӉ! ҈ҏқҝѹҔ Ҧӊҽ! ҨґҾҔѹ ҊҸҡҰҸҦҽҝҡҔѹѦ!!” ҲҸӁҔѹҮ ҊҸґҸҡ ҏңѦ ӁѦ ҰѦґ, ҽ ӉҽӊӃӊҽґ: “үґґѹґҏҮҾ!”. ҐҝҸѽҸҲ ӉҸґҸҲӃѽ ҲҽҝѦґҸҔ ӊ ҦңҽӉѦ ѹӁҸҊҦҽ ҰҸҔҝӃӊҽґ ҊҝҸѽҸҲ ҰҸҡҏҦӃ, ҔҸҲҸҝҏӀ ҡҏҦҸѸҸҮҔѹ ҰҸ ӁҽҔґҸӁӁҸҮ ҰґҸҡҔҸҡҲѹ ҡҰҏҡҔҽґѹ ӊ ҔҏѽӁӀ. ҋґҸӊҸѸ ҽҦ. (ҷҏґҊҽҔҸӊ, ҴҽҡҲѦҝ ѹ ҴҽҝҊҽҝѹҲҽ)

(1P)

Zlani potem kelnerzy nieǧli ponad g’owami spotnia’e kufle piwa, krzyczeli ochryple, z nienawiǧciƾ: “Najmocniej przepraszam, obywatelu!” G’os skƾdǧ z g’oǧnika komenderowa’: “Szasz’yk karski raz! Dwie gicze! Flaki po polsku!!” Cienki g’os nie ǧpiewa’ juǻ, ale wy’: “Alleluja!” Œoskot z’otych talerzy jazzbandu zag’usza’o od czasu do czasu szczNjkanie naczyǝ, które pomywaczki zsuwa’y po równi pochy’ej do kuchni. Jednym s’owem – piek’o.

(1Cz)

Z ǁíšníkDz se lil proudem pot, když nosili nad hlavou tácy s orosenými pDzllitry piva a sípavLJ, nenávistnLJ pokǥikovali: “PromiǞte, soudruhu.” Dopálený hlas v ampliónu zavelel: “Šašlyk jednou! Skopové na pepǥi dvakrát! Domácí dršǬky!” Tenor už nezpíval, ale kvílel: “Aleluja!” Ƥinkot zlatých ǁinelDz v orchestru obǁas pǥehlušil ǥinǁení nádobí, které myǁky pouštLJly po lince do kuchynLJ. Jedním slovem, peklo.

(1Sb)

ҫқґѹӊѦӁѹ ӉӁҸҎѦѸ, ҔѦґӁѦҝѹ ҡҏ ѹӉӁҽҦ Ҋґҽӊҽ ӁҸҡѹґѹ ӉҽѸҽҊҒѦӁѦ ҔҝѹҊґѦ Ұѹӊҽ ѹ ҡҽ ѸҝңѬҸѸ ҏ Ҋґҽҡҏ ҏӉӊѹҔѹӊҽґѹ: “ѨӉӊѹӁѹҲѦ, Ҋҝҽ҉ҽӁѹӁѦ!” ѢѦҊҦѦ ҏ ҊґҽҡӁҸҊҸӊҸҝӁѹҔҏ Ҋґҽҡ ҎѦ ҔҸѸҽӁҦҸӊҽҸ: “ҘҽҝҡҔѹ ҎѦҦҽӁ! ҈ҏқҝѹҔ Ҧӊҽ!” ү ӊѹҡҸҔѹ Ҋґҽҡ ӊѹҗѦ ӁѹҎѦ ҰѦӊҽҸ, ӊѦѴ ҎѦ ӉҽӊѹҎҽҸ: “үґѦґҏҎҽ!” һҏҰҔҽѬѦ ӉґҽҲӁѹѽ ҲҽҡҸӊҽ ҏ ѫѦӉҏ ҰҸӁѦҔҽҦ ҎѦ ӁҽҦҎҽҍҽӊҽґҽ ґҏҰҽ ҰҸҡҏ҉ҽ ҔҸҎѦ ҡҏ ҡҏҦҸҰѦҝѦ ҡҰҏҗҲҽґѦ ӁѹӉ ӁҽҊѹқ ҏ ҔҏѽѹѬҏ. ҀѦҦӁҸѸ ҝѦҍҎҏ: ҰҽҔҽҸ.

(1Hr)

Obliveni znojem, konobari su nad glavama nosili orošene vrǁeve piva, promuklo su i s mržnjom vikali: “Oprostite, graDžanine!” Negdje u zvuǁniku glas je zapovijedao: “Šašlik jedan! Bizonǁiǀ dva! Fileki domaǀi!!” Visoki glas više nije pjevao veǀ zavijao: “Aleluja!” Buku zlatnih tanjura u džezu ponekad je zaglušivala buka posuda koje su sudopere po kosini spuštale u kuhinju. Jednom rijeǁju, pakao.

added (for abbreviations see the end of the article). The crucial text is in bold print. The first place in a series of examples is in the language illustrated by the example, if the discussion pertains only to that language. The source is mentioned after the original example, which is given as the second in the series (if the source is different from the first in the series). The third and subsequent places are, if different, Russian, Polish, Czech, Serbian, Croatian, Bulgarian, other Slavic; non-Slavic translations (but not originals) are omitted.

354

CORNELIA E. KEIJSPER

(1Bg)

ҭґҏӊӁҽґѹ ӊ ҰҸҲ, ҔѦґӁѦҝѹҲѦ ӁҸҡѦѽҽ ӁҽҦ ҊґҽӊѹҲѦ ҡѹ ӉҽҰҸҲѦӁѹ ѽҽґқѹ қѹҝҽ, ҦҝѦӉҊҽӊҸ ѹ ӉґҸқӁҸ ӊѹҔҽѽҽ: “ҭҽҝҦҸӁ, ҊҝҽңҦҽӁѹӁҸ!” ѢҾҔӆҦѦ ѦҦѹӁ Ҋґҽҡ ҔҸѸҽӁҦӊҽҗѦ ҰҝѦӉ ҝҏҰҸҝ: “ѣҦѹӁ ҔҽҝҡҔѹ ҔѦқҽҰ! ѩӊѦ ӉҏқҝҸӊҔѹ! ҹҔѦѸқѦ ҰҸ ҡѦґҡҔѹ!” ѷӆӁҔѹҾҲ Ҋґҽҡ ӊѦҍѦ ӁѦ ҰѦѦҗѦ, ҽ ӊѹѦҗѦ: “үґѹґҏҾ!” ҐҝҸѽҸҲӆҲ Ӂҽ ӉґҽҲӁѹҲѦ ҍѹӁѦґѹ ӊ ҦңҽӉҽ ӉҽҊґҏҗҽӊҽҗѦ ҸҲ ӊҝѦѸѦ Ӂҽ ӊҝѦѸѦ ҊҝҸѽҸҲҽ Ӂҽ ҡӆҦҸӊѦҲѦ, ҔҸѹҲҸ ѸѹҾҍҔѹҲѦ ҡҰҏҡҔҽѽҽ ҰҸ ӁҽҔґҸӁѦӁҽ ҰґҸҡҔҸҡҲ ӊ ҔҏѽӁҾҲҽ. ҋ ѦҦӁҽ ҦҏѸҽ – ҽҦ.

(1BR)

ҋҰґӃӊҽґҟ ҰҸҲҽѸ ҽӍҟ҃ӃҾӁҲӃ ҟ Ӂҽҡҟґҟ ӁҽҦ ҊҽґҸӊҽѸҟ ӉҽҰҽ҃ѦґӃҾ ҔҏӍґҟ Ӊ ҰҟӊҽѸ, ѽҝӃҰґҽ ҟ Ӊ ӁҾӁҽӊҟҡ҃Ӏ ҔҝӃҍҽґҟ: “ғӃқҽҍҽҮ҃Ѧ, ҊҝҽѸҽҦӉҾӁҟӁ!” ѢѦҦӉѦ Ӊ ҝҏҰҽҝҽ ҊҸґҽҡ ҔҽѸҽӁҦҽӊҽҕ: “ҘҽҝҡҔҟ ҝҽӉ! ҈ҏқҝӃҔ Ҧӊҽ! ҨґҾҔҟ ҊҽҡҰҽҦҽҝҡҔҟҾ!” ҐҝҏҔҽҲ ӉҽґҽҲӃѽ ҦңҽӉҽӊӃѽ ҲҽґѦҝҽҔ ҏ ҽҝҔѦҡҲҝӃ ҍҽҡҽѸ ҰѦҝҽҔҝӃӊҽҕ ҊҝҏҔҽҲ ҰҸҡҏҦҏ, ҾҔҟ ҰҽҡҏҦҽѸӃҮҔҟ Ұҽ ңҸґҽқѦ ҡҰҏҡҔҽґҟ Ӂҽ ҔҏѽӁӀ. үҦӁӃѸ ҡґҸӊҽѸ, ҰѦҔґҽ.

As will be clear, the Czech and Bulgarian translations miss the point of the Russian OVS sentence. 4 Again, the example is unique, it does not represent the normal situation, which is that case syncretism as in ҋҰѦҔҲҽҔґҧ ҸҲґѹҍҽѦҲ ҸҝѹҊѹӁҽґҧӁҸҡҲҧ does not hinder the understanding of OVS order. 4. Nouns vs. pronouns It is not the case that case syncretism in general leads to restrictions of word order freedom (section 3). Is it true that if case distinctions are (almost) restricted to pronouns, OS orders (with various positions of V) are (almost) restricted to pronouns? Bulgarian is a Slavic language to consider in this connection. According to the examples given by ҐѦҸҝҊѹѦӊҽ (1974: 69), OS orders with nominal objects are not impossible: she gives SVO (ѷҸҮ ҔҽӉӊҽ ѹҡҲѹӁҽҲҽ), and SOV (ѷҸҮ ѹҡҲѹӁҽҲҽ ҔҽӉӊҽ), as well as OSV (ѨҡҲѹӁҽҲҽ ҲҸҮ ҔҽӉӊҽ). The restriction according to ҐѦҸҝҊѹѦӊҽ (ibidem) pertains to accent, not to order: Russian sentences like Ѥ ҙҲҸҲ ҝҽӉҊҸӊҸҝ ӉҽқӃґ (SOV), with the object in the so-called Theme (i.e. not in the scope of the last sentence accent – see section 6), are impossible without pronominal reduplication, so that the Bulgarian translation is (ibidem): (2Bg)

4

үӉ ҲҸӉѹ ҝҽӉҊҸӊҸҝ Ԅӹ Ӊҽқҝҽӊѹѽ

For considerations of space subsequent examples only give the sentence with the crucial (bold) text in all languages but not the larger context. Available languages are omitted if they are irrelevant to the point discussed.

FROM OVS ORDER TO CONVERSE STRUCTURE IN RUSSIAN

355

Further, ѷҸҡҔҽ ѸѦӁҾ ӊӉҾґҽ (SOV) is called impossible (ibidem), but at least nominal SVO is not, according to this example from ASPAC (valid if Bulgarian retains the implied accentuation of the Russian original): (3Bg)

ҐҝҽҦҡҔѹҾҲ ӉҝѦґѹӂѦӁ Ӎѹґѹҽґ ҡѦ ҰҸѸѦӂҽӊҽҗѦ ӊ ҰҸҸґӀӂѦӁҽ ҸҲ ӊҝѦѸѦҲҸ қҸҊҽҲҽҗҔҽ Ҕӆӂҽ ӊ ҦӆӁҸҲҸ Ӂҽ ѦҦѹӁ ҦӊҸҝ ѹ қѦҗѦ ҰҝҸҍҏҲ ҡ ҰҸҝӍѹҝӁѹҲѦ ҔҸґҸӁѹ ӊӆӊ ӊѦҡҲѹқӀґҽ. ѢҸ ӁѦ ҔҸґҸӁѹҲѦ ҡѸҽҮӊҽѽҽ ҲҸӉѹ ҦѦӁ ҰҸҡѦҲѹҲѦґѹҲѦ Ӂҽ Ӎѹґѹҽґҽ, ҽ ҸӁҸӊҽ, ҔҸѦҲҸ ҡҲҽӊҽҗѦ ѸѦңҦҏ ҲҾѽ. ѢҾҔҸґҔҸ ҰҸҡѦҲѹҲѦґѹ ҡҲҸҾѽҽ ӊ҃ѦҰѦӁѦӁѹ ѹ ҊґѦҦҽѽҽ ҰґҽҍѦӂҽҲҽ ҊҸҡҰҸңѹ҃ҽ, ҡѦҦӁҽґҽ ҦҸ ѦҦӁҽ ѸҽҡѹҍҔҽ ҡ ҰҸҦҝѦҦѦӁҽ Ӂҽ ӁѦҾ ҡҰѦ҃ѹҽґѹӉѹҝҽӁҽ ӉҝѦґѹӂӁҽ ґѹҲѦҝҽҲҏҝҽ, ҔҸҾҲҸ ҊҸҡҰҸңѹ҃ҽҲҽ ҰҝҸҦҽӊҽҗѦ.

(3R)

ҐҸҝҸҦҡҔҸҮ ӉҝѦґѹӂӁӃҮ Ӎѹґѹҽґ ҰҸѸѦӂҽґҡҾ ӊ ҸқґҏҰґѦӁӁҸѸ ҸҲ ӊҝѦѸѦӁѹ ҸҡҸқӁҾҔѦ ӊ ҊґҏқѹӁѦ ҦӊҸҝҽ ѹ ӉӁҽѸѦӁѹҲ қӃґ ҡӊҸѹѸѹ ҰҸҝӍѹҝҸӊӃѸѹ ҔҸґҸӁӁҽѸѹ ӊ ӊѦҡҲѹқӀґѦ. ѢҸ ӁѦ ҔҸґҸӁӁӃ ҰҸҝҽңҽґѹ ӊ ҙҲҸҲ ҦѦӁҧ ҰҸҡѦҲѹҲѦґѦҮ Ӎѹґѹҽґҽ, ҽ ҲҸ, ҍҲҸ ҰҝҸѹҡѽҸҦѹґҸ ҰҸҦ ӁѹѸѹ. ѢѦҡҔҸґҧҔҸ ҰҸҡѦҲѹҲѦґѦҮ ҡҲҸҾґѹ ӊ Ҹ҃ѦҰѦӁѦӁѹѹ ѹ ҊґҾҦѦґѹ Ӂҽ ҰґҽҍҏӂҏӀ қҽҝӃҗӁӀ, ҡѹҦѦӊҗҏӀ Ӊҽ ҡҲҸґѹҔҸѸ, Ӂҽ ҔҸҲҸҝҸѸ ґѦңҽґҽ ҡҰѦ҃ѹҽґҧӁҽҾ ӉҝѦґѹӂӁҽҾ ґѹҲѦҝҽҲҏҝҽ, ҰҝҸҦҽӊҽѦѸҽҾ қҽҝӃҗӁѦҮ. (ҷҏґҊҽҔҸӊ, ҴҽҡҲѦҝ ѹ ҴҽҝҊҽҝѹҲҽ)

(3P)

Nie owe kolumny wszakǻe, ale to, co siNj wǧród nich dzia’o, robi’o owego dnia niesamowite wraǻenie na interesantach.

(3Cz)

Dnes ale nepǥekvapovaly návštLJvníka sloupy, nýbrž to, co se dLJlo pod nimi.

(3Sb)

үґѹ ҲҸҊҽ ҡѦ ҦҽӁҽ ҰҸҡѦҲѹҸ҃ѹ Ӂѹҡҏ Ҧѹӊѹґѹ ҡҲҏқҸӊѹѸҽ ӊѦѴ ҸӁҸѸѦ җҲҸ ҡѦ ҰҸҦ ѬѹѸҽ ҦѦҗҽӊҽґҸ.

(3Hr)

Ali nisu stupovi ǁudili tog dana posjetioce filijale nego ono što se zbivalo pored njih.

(3BR)

үґѦ ӁѦ ҊҙҲӃҾ ҔҽґҸӁӃ ҕҝҽңӊҽґҟ ӁҽӊѦҦӊҽґҧӁҟҔҽҕ ҏ ҲҸҮ ҦӉѦӁҧ, ҽ ҲҸѦ, җҲҸ ҽҦқӃӊҽґҽҡҾ ҰҽҦ ҟѸҟ.

As the reader will notice, Czech and Serbian change the order of the words (see section 8 below). Dyer (1992) also gives Bulgarian examples of nominal OS orders without pronominal reduplication: the order is at least possible in phraseological verbobject combinations and in other disambiguating circumstances (1992: 86-87, 62-63, Dyer’s English translations): (4Bg)

ғ ӁѦҊҸӊҽҲҽ ҝҽқҸҲҽ ӊӉѦ ҏҍҽҡҲѹѦ ҰӆҝӊѹҾҲ ӉҽѸѦҡҲӁѹҔ-ҰҝѦҦҡѦҦҽҲѦґ Ӂҽ Ѣҋ Ӂҽ ҫҲѦҍѦҡҲӊѦӁѹҾ ӍҝҸӁҲ ҋҽӊҽ ѩӆґқҸҔҸӊ.

356

CORNELIA E. KEIJSPER

(4E)

The first deputy chairman of the NS of the Fatherland’s Front, Sava Dalbokov, took part in its work (VN, 9/6/85:1).

(5Bg) (5E)

ҘӁѹҊҽҲҽ ҍѦҲҽҲ ҡҲҏҦѦӁҲѹҲѦ The students are reading the book

(6Bg) (6E)

ҘӁѹҊҽҲҽ ҍѦҲҽ ҡҲҏҦѦӁҲӆҲ The student is reading the book

(7Bg) (7E)

ҘҝҏҗҽҲҽ ѽҽҰѦ ҔҏҍѦҲҸ The dog is munching on the pear

(8Bg) (8E)

ҘҝҽӊҽҲҽ ҝѹҡҏӊҽ ңѦӁҽҲҽ The woman is painting the cow

But in ambiguous cases (1992: 63): (9Bg) (9E)

ҭҝѦҦҰҝѹҾҲѹҾҲҽ ԄӠ ҡӁҽқҦҾӊҽҲ ѸҽҊҽӉѹӁѹҲѦ The stores are stocking the firms

This information can as yet not entirely be confirmed by means of ASPAC. Although example (10) below illustrates the possibility of nominal OS orders, it was extremely difficult to find the example, so at least the frequency of the nominal OS orders allowed by Georgieva and Diver is not very high in literary texts.5 The normal case in ASPAC is that OS orders have (at least) a pronominal object or a pronominal reduplication of a nominal object. (10Bg) ѷҽҔҽ ҰҸҡҝѦӂӁҽ ҝҽӉҡӆѸӊҽӁѦҲҸ Ӂҽ ҰѦҲӁҽҮҡѦҲѹ, ӁѹҡҽӁ ҰѦҲѹҾҲ ҰҝҸҔҏҝҽҲҸҝ Ӂҽ ѨҏҦѦҾ ҭѹґҽҲ ҭҸӁҲѹҮҡҔѹ. (10R)

ҋӊѦҲѹґҧӁѹҔѹ ҦҽӊӁӃѸ-ҦҽӊӁҸ ҰҸҊҽҡґѹ. Ѣҽ ґҸңѦ ґѦңҽґ ҰҝҸҔҏҝҽҲҸҝ. ҭҸҦґҸңѹӊ ҝҏҔҏ ҰҸҦ ӂѦҔҏ, ҸӁ ҡҰҽґ ѹ ҦӃҗҽґ қѦӉӉӊҏҍӁҸ. ҩҾҦҸѸ ҡ ӁѹѸ ҡҰҽґ ҷҽӁҊҽ. ѷҽҔ ӊҡҲҝѦҲѹґ ҝҽҡҡӊѦҲ ҰҾҲӁҽҦ҃ҽҲҸҊҸ ӁѹҡҽӁҽ ҰҾҲӃҮ ҰҝҸҔҏҝҽҲҸҝ ѨҏҦѦѹ ҭҸӁҲѹҮ ҭѹґҽҲ. (ҷҏґҊҽҔҸӊ, ҴҽҡҲѦҝ ѹ ҴҽҝҊҽҝѹҲҽ)

(10P)

Tak spotka’ ǧwit piNjtnastego nisana piƾty procurator Judei, Poncjusz Pi’at.

5

ASPAC contains ca. 278 pages Bulgarian translations of Russian texts (ҷҏґҊҽҔҸӊ, ҋҲҝҏҊҽ҃ҔѹѦ), and ca. 1706 pages translated into Bulgarian from non-Slavic sources; both have been searched (but the non-Slavic sources not yet completely) on the basis of the Russian verbs in note 1 and on the basis of a few Bulgarian stems. Ca. 1045 pages translated into Macedonian from non-Slavic sources, searched on the basis of a few Bulgarian stems, do not change the picture.

FROM OVS ORDER TO CONVERSE STRUCTURE IN RUSSIAN

357

(10Cz) Tak pǥivítal úsvit patnáctého dne mLJsíce nísánu pátý prokurátor Judeje Pilát Pontský. (10Sb)

ѷҽҔҸ ҦҸҍѦҔҽ ӉҸҝҏ ҰѦҲӁҽѦҡҲҸҊ ӁѹҡҽӁҽ ҰѦҲѹ ҰҝҸҔҏҝҽҲҸҝ ҀҏҦѦҎѦ ҭҸӁҲѹҎѦ ҭѹґҽҲ.

(10Hr) Tako je doǁekao svitanje petnaestog nisana peti prokurator Judeje, Poncije Pilat. The example suggests that it is indeed accentuation which is the relevant parameter: normally, the example would be spoken as a series of separate pieces of information, every piece introduced by ҽ full accent (see the end of section 6). Further, apart from pronominal reduplication as in (9Bg), we find cases where a nominal object in a Russian original is replaced by a pronominal object in Bulgarian, so that the problem is avoided: (11R)

ҘҝҸѸѦ ҲҸҊҸ, ҷѦҝґѹҸӉҽ ҸѽӊҽҲѹґ ӁѦҸқҸҡӁҸӊҽӁӁӃҮ, ӁҸ ҡҲҸґҧ ҡѹґҧӁӃҮ ҡҲҝҽѽ, ҍҲҸ ѦѸҏ ӉҽѽҸҲѦґҸҡҧ ҲҸҲҍҽҡ ңѦ қѦңҽҲҧ ҡ ҭҽҲҝѹҽҝҗѹѽ қѦӉ ҸҊґҾҦҔѹ. (ҷҏґҊҽҔҸӊ, ҴҽҡҲѦҝ ѹ ҴҽҝҊҽҝѹҲҽ)

(11Bg)

ҫҡӊѦӁ ҲҸӊҽ Ԅӹ ҸқӉѦ ӁѦҸҡӁҸӊҽҲѦґѦӁ, ӁҸ ҲҸґҔҸӊҽ ҡѹґѦӁ ҡҲҝҽѽ, ҍѦ Ѹҏ ҡѦ ҦҸӂҾ ӊѦҦӁҽҊҽ Ҧҽ ѹӉқҾҊҽ ҰҝѦӉҊґҽӊҽ ҸҲ ҭҽҲҝѹҽҝҗѹҲѦ ѦӉѦҝҽ.

Or in this way: (12R)

ѢҽѦӊҗҏӀҡҾ ҴҽҝҊҽҝѹҲҏ ҸѽӊҽҲѹґҸ ҍҏӊҡҲӊҸ қґҽңѦӁҡҲӊҽ. (ҷҏґҊҽҔҸӊ, ҴҽҡҲѦҝ ѹ ҴҽҝҊҽҝѹҲҽ)

(12Bg)

ҘҽҲҸ ҡѦ ӁҽѽҝҽӁѹ, ҴҽҝҊҽҝѹҲҽ ҏҡѦҲѹ, ҍѦ ԋ ҸқӉѦѸҽ ҍҏӊҡҲӊҸ Ӂҽ қґҽңѦӁҡҲӊҸ.

Another possibility of avoiding nominal OS orders is the use of converse structures (see example (38Bg) in section 11). As to other Slavic languages in ASPAC, no OS examples were found in Sorbian in cases where Russian had OVS with the verbs of note 1, but since the parallel text available is only ca. 142 pages (Milne), this result is negligible. The ca. 204 pages Slovenian (Kipling, Milne, Schmidt) prove the possibility of nominal OVS order in principle by the following example (pronominal examples are the normal case): (13Sn)

“Tu so bili,” je rekel Willem. “Vse sem pokazal, tako kot so mi naroǁili.” “Ampak, kako so ti zadnji prišli zraven?!” “Tega pa ne vem,” je odgovoril Willem. Dvorano je tedaj napolnil velikanski hrušǁ. Ljudje so bili strašno besni in gospod Smit se je trudil, da bi jih vsaj malo pomiril. “Dame in gospodje, pomirite se! Samo za velik nesporazum gre,” je rekel.

358

CORNELIA E. KEIJSPER

(13D)

‘Ik draai gewoon het hele rijtje af,’ zei Willem. ‘Zoals het moet.’ ‘Maar hoe komen die laatste er dan bij?’ ‘Dat kan ik toch niet weten,’ zei Willem. Er was nu een hele opschudding in de zaal en meneer Smit probeerde de herrie te sussen. ‘Dames en heren, dit alles berust op een beklagenswaardig misverstand,’ zei hij. (Schmidt, Minoes)

(13R)

– Ѥ ҰҝҸҡҲҸ ҰҸҔҽӉӃӊҽґ ӊҡѦ ҰҸ ҰҸҝҾҦҔҏ, – ҰҸңҽґ ҰґѦҍҽѸѹ ғѹґґѦѸ. – ѯҲҸ ѹ ҲҝѦқҸӊҽґҸҡҧ. – ѢҸ ҸҲҔҏҦҽ ңѦ ҸӁѹ ӊӉҾґѹҡҧ? – ҭҸӁҾҲѹҾ ӁѦ ѹѸѦӀ. ғҸґӁѦӁѹѦ ӊ ӉҽґѦ ӁҽҝҽҡҲҽґҸ, ѹ ҊҸҡҰҸҦѹӁ ҋѸѹҲ ҰҸҰӃҲҽґҡҾ ҝҽӉҝҾҦѹҲҧ ҸқҡҲҽӁҸӊҔҏ. – ѩҽѸӃ ѹ ҊҸҡҰҸҦҽ, ӊҡѦ ҙҲҸ ҦҸҡҽҦӁҸѦ ӁѦҦҸҝҽӉҏѸѦӁѹѦ, – ҏӊѦӂѦӊҽґ ҸӁ ҰҏқґѹҔҏ.

(13Cz)

V sále bylo nyní hotové pozdvižení a pan Smit se pokoušel halas utišit.

The example is remarkable because Russian OVS orders with ӁҽҰҸґӁҾҲҧ/ӁҽҰҸґӁѹҲҧ are relatively rare (see section 11, point 9). The fact that the Slovenian object refers to a place/container is probably notable. As will be clear, 204 pages of Slovenian text is too little to draw further conclusions. Slovak (272 pages: Carroll, Exupéry, Milne, Tolkien) is as yet also relatively poorly represented. The same word dvorana provides a nominal OVS example: (14Sk)

Fili a Kili mali takmer veselú náladu. KeDŽ našli na stene hǤbu zlatých hárfso striebornými strunami, pozvesovali ich a brnkali na nich. Harfy boli ǁarovné (a šarkan sa ich ani nedotkol, lebo o hudbu sa veǚmi nezaujímal), nuž zostali naladené. Temnú dvoranu zaliala melódia, aká tu už dlho nezaznela.

(14E)

Fili and Kili were almost in merry mood, and finding still hanging there many golden harps strung with silver they took them and struck them; and being magical (and also untouched by the dragon, who had small interests in music) they were still in tune. The dark hall was filled with a melody that had long been silent. (Tolkien, The Hobbit)

(14RQu) үҝӍӃ қӃґѹ ӁҽҡҲҝҸѦӁӃ, ѹ ҰѦӂѦҝҏ ӉҽҰҸґӁѹґѹ ҦѹӊӁӃѦ ӉӊҏҔѹ, ӁҽҝҏҗѹӊҗѹѦ ѸӁҸҊҸґѦҲӁѦѦ қѦӉѸҸґӊѹѦ. ѢҸ ѸӁҸҊѹѦ ҔҽҝґѹҔѹ қӃґѹ ҰҸҊґҸӂѦӁӃ ѹӁӃѸѹ ҦҏѸҽѸѹ: (14RKor) ҭҸҦ ѸҝҽҍӁӃѸѹ ҡӊҸҦҽѸѹ ҰҸҰґӃґҽ ҦҝѦӊӁҾҾ ѸѦґҸҦѹҾ. (14PBra) Mrocznƾ salNj wype’ni’a dawno nie s’yszana tu melodia. (14PSki) Mroczna pieczara rozbrzmia’a melodiƾ od dawna nie s’yszanƾ. (14Cz) Temná síǞ se rozezvuǁela melodiemi, které byly dlouho nLJmé.

FROM OVS ORDER TO CONVERSE STRUCTURE IN RUSSIAN

(14Sb)

359

Tamna sala ispuni se melodijom koja je dugo pripadala muku. Ali veǀina patuljaka bila je praktiǁnija:

(14Hr) Mraǁna dvorana ispunila se melodijama koje odavno nisu u njoj odjekivale. Ali veǀina patuljaka bila je praktiǁnija: (14Bg)

… ѹ ҲӆѸӁҸҲҸ ҰҸҦӉѦѸѹѦ ҡѦ ѹӉҰӆґӁѹ ҡ ҸҲҦҽӊӁҽ ӉҽқҝҽӊѦӁҽ ѸѦґҸҦѹҾ. ҫҡҲҽӁҽґѹҲѦ ҦңҏҦңѦҲҽ ҸқҽҍѦ, қҾѽҽ ҰҸ-ҰҝҽҔҲѹҍӁѹ

(14Uk) ѷѦѸӁҽ Ӊҽґҽ ӁҽҰҸӊӁѹґҽҡҾ ѸѦґҸҦҟ҄Ӏ, ӂҸ ӁѦ Ӊӊҏҍҽґҽ ҲҽҔ ҦҸӊҊҸ. үґѦ ҝѦҗҲҽ ҊӁҸѸҟӊ қҏґҽ ҰҝҽҔҲѹҍӁҟҗҽ: (14BR) ҭҽ ҃ҺѸӁҽҮ Ӊҽґҟ ҝҽӉӁѦҡґҽҡҾ ѸѦґҸҦӃҾ, ҾҔҸҮ ҊҙҲӃҾ ҡ҃ѦӁӃ ӁѦ ҍҏґҟ ҦҸҕҊҟҾ, ҦҸҕҊҟҾ ҍҽҡӃ. үҡҲҽҲӁҟҾ қӃґҟ қҸґҧҗ ҰҝҽҔҲӃҍӁӃҾ, Note that the first of the Russian translations also has OVS order, but contains the stem ӉҽҰҸґӁ- rather than ӁҽҰҸґӁ-. The fact that the first Polish translation has OVS with wype’n- is less remarkable (see further section 11, point 9). Also, nominal objects in the genitive are normal in Slovak OS orders, for example with the stem zmocn- (as in Czech, see example (19) in section 7). But pronominal OS prevails. The occurrence of nominal as well as pronominal OS orders in all other Slavic languages represented in ASPAC has been verified by me and cannot be doubted. 5. “Dominant order” vs. “syntagmatically independent member” The Greenberg tradition of word order typology (see the Introduction) discusses issues such as whether a language has VSO or SVO or SOV as the dominant order. This tradition is originally based on the idea that a single order is dominant in an entire language, or at least in some type of simple sentence, for example in declarative sentences, with possible differences between nouns and pronouns. In this tradition, Russian is usually regarded as a SVO language (but see e.g. King 1995, who proposes that Russian has underlying VSO order). This classification is not without problems if we also take into account spoken Russian, which does not conform to the written norm of placing the last accented word in final position (Keijsper 1985: 44-49, see also section 8). The Mathesian tradition represented by ҘҸӊҲҏӁҸӊҽ (1976) and other publications has made it clear, using other terms, that it is not useful to try and determine “the” “dominant” order in Slavic languages, because there is no such thing independently of meaning. As I said in the introduction, the Mathesian tradition recognizes that the “ҡѹӁҲҽҊѸҽҲѹҍѦҡҔѹ ӁѦӉҽӊѹҡѹѸӃҮ ҍґѦӁ ҰҽҝҽҦѹҊѸӃ” is different for different verbs; the same holds true for different uses of

360

CORNELIA E. KEIJSPER

verbs. For the simple Subject-Verb-Object combinations that are of interest here ҘҸӊҲҏӁҸӊҽ (1976: 170-183) recognizes four groups: 1. A group that has [S[VO]] as the “syntagmatically independent member”. The object need not be in the accusative case, and there can be a prepositional phrase instead of an object: ҐҝҏӁҾ ҫӍѹ҃ѦҝҸӊҽ ӁҸҡѹґҽ ҰҸҍҲҏ, үӁҦҝѦҮ ҡҦѦґҽґ ҔҽҝҧѦҝҏ, ҷѹҝӀҔ ҰҸңѹѸҽґ ҰґѦҍҽѸѹ, ҭҽҝҸӊҸӉ ҏҰѦҝҡҾ ӊ ҲҏҰѹҔ, ҴӃ ҰҸҦҝҏңѹґѹҡҧ ҡ ҏҍѹҲѦґѦѸ, and so on. 2. A group that has [O[VS]] as the “syntagmatically independent member”. The object need not be in the accusative case, and there can be a prepositional phrase instead of an object: ҶҲҏ ҔӁѹҊҏ ҰҝҸӁѹӉӃӊҽѦҲ ѹҝҸӁѹҾ, ҋҰѦҔҲҽҔґҧ ҸҲґѹҍҽѦҲ ҸҝѹҊѹӁҽґҧӁҸҡҲҧ, ғҽґҧҔҸӊӃѸ ҸӊґҽҦѦґҽ ҾҝҸҡҲҧ, ҴӁѦ ҰҝѹҡӁѹґѹҡҧ ҏґѹ҃Ӄ, Ѣҽ ѸѦӁҾ ӁҽҰҽҦҽґҽ ҡѦҝҽҾ ҲҸҡҔҽ, ғ ѽҸґҸҦӁҸѸ ӁѦқѦ ҡѹӁѦґѹ ӉӊѦӉҦӃ, and so on. 3. A group that has both [S[VO]] and [O[VS]] as “syntagmatically independent members”. Kovtunova, probably accidentally, has no examples of nonaccusative objects, but there can be a prepositional phrase instead of an object: ғӁҏҔѹ ҸҔҝҏңѹґѹ ҦѦҦҽ / ѩѦҦҽ ҸҔҝҏңѹґѹ ӊӁҏҔѹ, ҭҝҸӍѦҡҡҸҝ ҰҸҡѦҲѹґ ґҽқҸҝҽҲҸҝѹӀ / һҽқҸҝҽҲҸҝѹӀ ҰҸҡѦҲѹґ ҰҝҸӍѦҡҡҸҝ, ѩѦҲѹ ѹҊҝҽӀҲ ӊ ҡҽҦҏ / ғ ҡҽҦҏ ѹҊҝҽӀҲ ҦѦҲѹ, ҴҸґҸҔҸ ӊӃҲѦҔґҸ ѹӉ қҏҲӃґҔѹ / ѨӉ қҏҲӃґҔѹ ӊӃҲѦҔґҸ ѸҸґҸҔҸ, and so on. 4. A group that has the last accent on the verb, the verb specifying the relationship between S and O, either with SVO order (if the subject is the so-called semantic subject), or with SVO or OVS order (if the object is both the socalled ҡҏқӆѦҔҲ ҡҸҡҲҸҾӁѹҾ and the ҸқӆѦҔҲ ҦѦҮҡҲӊѹҾ). The object need not be in the accusative case, and there can be a prepositional phrase instead of an object: ҤҸѽґҽҲҔҽ ґӀқѹґҽ ҡӊҸӀ ҝҽқҸҲҏ, ҴӃ ӁѦ ҝҽӉқѹӊҽґѹ ҊҸґҏқҸҮ ҍҽҗҔѹ, ҭҝѹҊґҽҗѦӁѹѦ ѸҏҍѹґҸ ѦҊҸ / ѣҊҸ ѸҏҍѹґҸ ҰҝѹҊґҽҗѦӁѹѦ, ҴҸѹ ӊҸҰҝҸҡӃ ӊҡҲҝѦӊҸңѹґѹ ҦҾҦӀ / ѩҾҦӀ ӊҡҲҝѦӊҸңѹґѹ ѸҸѹ ӊҸҰҝҸҡӃ, ҶҲҽ ѸӃҡґҧ ҰҸҝҽӉѹґҽ ѸѦӁҾ / ҴѦӁҾ ҰҸҝҽӉѹґҽ ҙҲҽ ѸӃҡґҧ, and so on. The examples with a prepositional phrase instead of a direct object, or with a dative object instead of an accusative object lack the theoretical piquancy of [O[VS]] explained in section 2, but they play a role in translations (see section 9 ff.). The “syntagmatically independent member of the paradigm” used for this classification is the linear-intonational hierarchy of the given elements that ensures that the resulting sentence has, with the given elements, a minimal Theme and a maximal Rheme: the syntagmatically independent member of the paradigm contains a minimal amount of so-called “old”, so-called contextually dependent information (i.e. selected or repeated from the context), and it contains

FROM OVS ORDER TO CONVERSE STRUCTURE IN RUSSIAN

361

a maximal amount of so-called “new”, so-called contextually independent information (i.e. introduced into the context). The classification does not predict the frequency of SVO/OVS or of the “syntagmatically independent member of the paradigm”: it may but need not be the case that most verbs are in group 1 (which is what the idea that Russian is an SVO language amounts to), and it may but need not be the case that the verbs classified in one of the groups are the most frequent members of their sets of socalled converse structures (see section 9 ff.). The classification also does not prescribe a linear-intonational hierarchy: it only says that another choice than the “syntagmatically independent member of the paradigm” will result in a sentence with a larger Theme and a smaller Rheme than the “syntagmatically independent member of the paradigm” has. For example, if in the first group we choose OVS, the Theme will be OV, and the Rheme will be S alone. Thus, ҭҸҍҲҏ ӁҸҡѹґҽ ҐҝҏӁҾ means “the one who carried the post is Grunja”: the sentence has OVS order but not the constituency [O[VS]] of the second group. Instead, in the first group S on OVS is a separate constituent specifying the agent already implied in the OV group. Likewise, if in the second group we choose an SVO order, the Theme will be SV, and the Rheme will be O alone. Thus, ҫҝѹҊѹӁҽґҧӁҸҡҲҧ ҸҲґѹҍҽѦҲ ҡҰѦҔҲҽҔґҧ means, when read as SVO: “the thing characterized by originality is the show”, where the accent on O specifies the object already implied in the combination SV. Various uses of the same verb can be in different groups. ҴҽҲҧ ҸҲґѹҍҽѦҲ ҦҸҍҧ (ҸҲ/ҡҝѦҦѹ ҦҝҏҊѹѽ ҦѦҲѦҮ), for example, belongs in the first group: the sentence tells us what mother does ([S[VO]], rather than what property the object has in comparison with other things (as a sentence with the same verb does in group 2). When a verb systematically has two possibilities, it is in the third group: ғӁҏҔѹ ҸҔҝҏңѹґѹ ҦѦҦҽ ([S[VO]]) tells us what the grandchildren did (they surrounded grandfather), while ѩѦҦҽ ҸҔҝҏңѹґѹ ӊӁҏҔѹ ([O[VS]]) conveys what happened to grandfather (he became surrounded by his grandchildren). The fourth group contains verbs that specify the relationship between two referents but are unable to introduce the fact that the relationship exists (according to Kovtunova). As appears from these examples, what can be the maximal Rheme in these simple combinations depends on the lexical meaning of the verb and the boundaries of its uses (the tolerance boundaries of the meaning of the verb). This is a consequence of the functioning of sentence accent (see section 6). Despite the fact that the Mathesian tradition contains many linguistic mistakes (Keijsper 1985: 3-145 is a detailed criticism), it has contributed the insight that linear-intonational hierarchy in Slavic languages expresses informational content, and not which constituents are subject and object. This was a step forward not only with respect to typological studies like Greenberg’s (1963), but also

362

CORNELIA E. KEIJSPER

with respect to earlier Slavistic studies of word order in Slavic. Thus, Berneker (1900), ҷҽҍӊҽҝҸӊ (1974) and Schaller (1966) discuss issues such as the position of the verb with respect to the subject, clitics, the position of attributive adjuncts, and other traditional issues of word order, but, apart from Schaller (in pre-Mathesian terms), they do not raise the point of informational content. To be sure, the orders mentioned here are not the only possibilities, of course. The order VOS, for example, is perfectly normal in many cases, for example following a piece of direct speech: (15R)

ѩңѦӁҲґҧѸѦӁӃ, – ӉҽӊѦҝҗѹґ ҡӊҸѦ ӊӃҡҲҏҰґѦӁѹѦ ѢҽҰҸґѦҸӁ (Orwell, Animal farm)

This is because the direct speech introduces the fact that speech is produced, which fact is repeated by VO (in other words), while S specifies the speaker. 6. Lexical meaning and constituency Kovtunova’s classification explained in section 5 is formulated in terms of minimal Themes and maximal Rhemes. If one reduces these interpretational notions to their causes (as I did in Keijsper 1985), one sees that the classification is based on the possible scope of the last sentence accent: the more words are included into this scope the larger the so-called Rheme is. As Kovtunova’s list shows, the number of words that can be included in the scope of the last sentence accent depends (inter alia) on lexical meaning. This is a consequence of the fact that, in order to form a constituent with a subsequent noun phrase or prepositional phrase, a finite verb must fit semantically into the idea that it is the sentence accent on the subsequent noun phrase or prepositional phrase (in writing implied) which introduces the constituent consisting of verb plus noun phrase or prepositional phrase: [ҋҰѦҔҲҽҔґҧ [ҸҲґѹҍҽѦҲ ҸҝѹҊѹӁҽґҧӁҸҡҲҧ]] conveys, in effect, “the performance is original” or “the performance has originality”: the connection between the performance and originality is introduced by the accent on S, while the verb “ҸҲґѹҍҽѦҲ” only specifies the broader idea of being or having introduced by the accent. In terms I used elsewhere, the last accent expresses an intonational predication, plus an intonational nexus if, as in this case, a noun phrase or a prepositional phrase precedes the verb (Keijsper 1985: 240-246, 2003: 141-151, 2004: 221). The verb fills in this general idea of intonational nexus with a slightly more precise idea. In still other terms (after Keijsper 1994: 259 ff.), Russian [O[VS]] conveys that within the span of attention defined by O the idea of the existence or presence of S is introduced, while V only specifies the way in which this existence or presence manifests itself. Not all verbs have this “way of being” among their possible uses, and agentive uses of verbs counteract it. In ҭҸҍҲҏ ӁҸҡѹґҽ ҐҝҏӁҾ, for example, there is no way of seeing Grunja’s being within boundaries defined by post and consist-

FROM OVS ORDER TO CONVERSE STRUCTURE IN RUSSIAN

363

ing in delivering the post; instead, delivering the post is something Grunja does in addition to being (existing, having a life): Grunja is what we call an agent. The Rheme now is the subject alone: the intonational predication and nexus introduced by the accent on S conveys “is Grunja”. [ҐҝҏӁҾ [ӁҸҡѹґҽ ҰҸҍҲҏ]] has the larger VO-constituent as Rheme and therefore is the “syntagmatically independent member of the paradigm” (S[VO] conveys what Grunja – selected (+accent) or repeated (-accent) from the context – did. This [S [VO]] sentence is, then, the normal “syntagmatically independent member of the paradigm” for agentive verbs or agentive uses of verbs. Sentences like ҭҸҍҲҏ ӁҸҡѹґҽ ҐҝҏӁҾ do not have the theoretical piquancy explained in section 2, because they lack the relevant VS constituent (in generative “explanations” they would probably be derived by means of a movement transformation). We will see in section 11 (point 9) that the precise possibilities of [O[VS]] are more subtle than nonagentive vs. agentive. Cases like Kovtunova’s ҭҝҸӍѦҡҡҸҝ ҰҸҡѦҲѹґ ґҽқҸҝҽҲҸҝѹӀ / һҽқҸҝҽҲҸҝѹӀ ҰҸҡѦҲѹґ ҰҝҸӍѦҡҡҸҝ have both [S[VO]] and [O[VS]] possibilities. On the one hand, the professor is an agent who willingly and on his own initiative pays a visit to the laboratory ([S[VO]], what did the professor do?), but on the other hand it is possible to view what happens from within the laboratory: the place is visited by the professor, which is a way of being present on the scene defined by the laboratory ([O[VS]], what happened to/in the laboratory?). In so far as it is possible to hide the professor’s agentive role (willingly, on his own initiative) in the case of OVS (a verb supporting the idea of appearing or being on the scene helps), the [O[VS]] type is possible. But the relevant VS constituent is more obvious with non-agents, for example: (16R)

ѷҸҊҦҽ ѢѹҔҽӁҸҝҽ ѨӊҽӁҸӊѹҍҽ ҰҸҡѦҲѹґҸ ҡӁҸӊѹҦѦӁѹѦ, ӊ ҸҡӁҸӊѦ ҔҸҲҸҝҸҊҸ, ӁѦҡҸѸӁѦӁӁҸ, қӃґѹ ѦҊҸ ҡѦҊҸҦӁҾҗӁѹѦ ҰѦҝѦңѹӊҽӁѹҾ. (ҷҏґҊҽҔҸӊ, ҴҽҡҲѦҝ ѹ ҴҽҝҊҽҝѹҲҽ)

It goes without saying that OVS examples in the corpus include both the type with and the type without [VS] constituent. More importantly, it must be added that in reality the difference between the two is not so simple as that between “the verb belongs to the Rheme or new information” and “the verb belongs to the Theme or old information”: often there is an additional accent on O, which can have V in its scope, or there are additional accents on O as well as on V (if the text is long enough). In that case first [OV] is introduced and then is already introduced when S is introduced after that, or all pieces of information are introduced separately. The difference between can amount, then, to a difference in the intonational grouping of V (with S or with O), or even no grouping at all (see example (10)). This explains why Russian OVS sentences which Kovtunova would analyze as [O[VS]] can be translated into other Slavic languages which

364

CORNELIA E. KEIJSPER

group O with V as a result of clitic placement (see section 7), with hardly any effect on the information conveyed. 7. “Automatic” differences in word order In good translations the linear-intonational hierarchy of the original text is usually retained as far as possible. Yet, the original text must often be adapted to the possibilities of the target language in a translation. Automatic adaptations pertain to the position of unaccented verb forms and clitic pronouns. They hardly affect the informational content, as long as the order of sentence accents remains the same (for other cases see section 8 below). Most word order differences between languages in ASPAC are of this type, and no difference whatsoever can be found between treatments of SO or OS orders. The rules of pronoun placement will not be discussed here; they can be found in grammars of the individual languages. An example may suffice to illustrate that they have effect on word order, of course (R SVO vs. P SOV vs. Cz VOS vs. Sb OSV vs. Hr OSV vs. Bg SOV): (17R)

ҭҸҡґѦ ӊҡѦѽ ӊҸґҗѦқҡҲӊ ѹ ҍҏҦѦҡ ҡѦҊҸҦӁҾҗӁѦҊҸ ӊѦҍѦҝҽ ҸӁҽ ҏңѦ ҦҸҊҽҦӃӊҽґҽҡҧ, Ҕ ҔҸѸҏ ѹѸѦӁӁҸ ӊ ҊҸҡҲѹ ѦѦ ӊѦӉҏҲ, ӁҸ ҙҲҸ ӁѦ ҰҏҊҽґҸ ѦѦ. ѢҽҦѦңҦҽ Ӂҽ ҲҸ, ҍҲҸ ҲҽѸ ѦҮ ҏҦҽҡҲҡҾ ҦҸқѹҲҧҡҾ ӊҸӉӊҝҽӂѦӁѹҾ ҡӊҸѦҊҸ ҡҍҽҡҲҧҾ, ҡҦѦґҽґҽ ѦѦ қѦҡҡҲҝҽҗӁҸҮ. (ҷҏґҊҽҔҸӊ, ҴҽҡҲѦҝ ҴҽҝҊҽҝѹҲҽ)

(17P)

Po wszystkich cudach i cudeǝkach dzisiejszego wieczoru domyǧla’a siNj juǻ, do kogo jƾ wiozƾ, ale to jej nie przeraǻa’o.

(17Cz)

Po všech tLJch kouzlech a zázracích dnešního veǁera se dovtípila, ke komu ji vlastnLJ vezou na návštLJvu, ale nedLJsilo ji to.

(17Sb)

ҭҸҡґѦ ҡӊѹѽ ҍҏҦҽ ѹ ҍҽҝҸґѹҎҽ ҸӊѦ ӊѦҍѦҝѹ, ҸӁҽ ҎѦ ӊѦѴ ӁҽҊҽ҉ҽґҽ ҔҸҦ ҔҸҊҽ ҎѦ ҲҸ ӊҸҦѦ ҏ ҊҸҡҲѦ, ҽґѹ ҎѦ ҲҸ ӁѹҎѦ Ӂѹ ӁҽҎѸҽѬѦ ҰґҽҗѹґҸ.

(17Hr) Poslije svih ǁarobnjaštava i ǁudesa današnje veǁeri, ona se veǀ domišljala kome je zapravo voze u goste, ali je to nije plašilo. (17Bg)

ҋґѦҦ ӊҡѹҍҔѹ ӊӆґҗѦқҡҲӊҽ ѹ ҍҏҦѦҡҽ Ӂҽ ҲҽӉѹ ӊѦҍѦҝ ҲҾ ӊѦҍѦ ҡѦ ҦҸҡѦӂҽҗѦ Ұҝѹ ҔҸҊҸ Ҿ ӊҸҦҾҲ Ӂҽ ҊҸҡҲѹ, ӁҸ ҲҸӊҽ ӁѦ Ҿ ҰґҽҗѦҗѦ.

As is well known, pronominal subjects are normally absent in a number of Slavic languages, so that OS in one language can correspond to O alone in another language, with various positions of V: (18R)

– ѩҽ. ҋҔҽңҏ ҏң ҲѦқѦ ӊҡѦ. ҩѦҍҧ ҡӊҸӀ ӹӰԂ ӉҽҔҸӁҍѹґҽ ҡґҸӊҽѸѹ, ҔҸҲҸҝӃѦ ҰҸҝҽӉѹґѹ ѸѦӁҾ ѹ ӊӉӊҸґӁҸӊҽґѹ: “ѷӃ, ҘҽӉҾ, ҰҝѹҰҸѸӁѹҗҧ ѸҸѹ ҡґҸӊҽ ҰҸӉңѦ, ҔҸҊҦҽ ҸӁѹ ҡқҏҦҏҲҡҾ, ӊѦҦҧ ҏѸѹҝҽӀӂѹѦ ҰҝҸӉҸҝґѹӊӃ …”

FROM OVS ORDER TO CONVERSE STRUCTURE IN RUSSIAN

(18P)

365

– Tak! Zresztƾ powiem ci wszystko. Swojƾ rozmowNj zakoǝczy’a s’owami, które dziwnie mnie poruszy’y … “Wspomnisz sobie moje s’owa, Kaziu, ǻe tak bNjdzie, bo umierajƾcy widzƾ jaǧniej …” (Prus, Lalka)

Subjects are of course always absent in impersonal sentences, so that translations of impersonal sentences by means of personal sentences, or the other way round, may have additional OS occurrences (here a genitive object): (19R)

ҭҸҦҰҸҝҏҍѹҔҽ ѩҏқҽ қҝҸҡѹґҸ ӊ ңҽҝ, ҸҡҸқѦӁӁҸ ҔҸҊҦҽ ҰҸҡґѦ ҸқӁҽҦѦңѹӊҽӀӂѦҊҸ җҏҝҗҽӁѹҾ қҏѸҽҊѹ ҰҝҸҗґҸ ѦӂѦ ҡѦѸҧ ѸѹӁҏҲ, ҽ ҦӊѦҝҧ ӊҡѦ ӁѦ ҸҲҔҝӃӊҽґҽҡҧ.

(19Cz) Poruǁíka Duba zmocnila se horeǁka, zejména když po nadLJjném šustLJní papíru uplynulo dalších sedm minut, aniž by se dveǥe otevǥely. (Hašek, Švejk) Object pronouns can be absent as well, so that OS can correspond to S alone, with various positions of V: (20R)

ҘҽҔ-ҲҸ ҡѸҾҊҍѹґҡҾ ӊ ҰҽѸҾҲѹ ҰҝҸҔґҾҲӃҮ қѦҡҸӊҡҔѹҮ ҔҸҲ, ӁѦ ҰҏҊҽґҽ қҸґѦѦ ҸҲҝѦӉҽӁӁҽҾ ҊҸґҸӊҽ, ѹ, ҰҸҔѹӁҏӊ ѸӃҡґҧ Ҹ ӁѦҮ, ҡҲҽґ ҝҽӉѸӃҗґҾҲҧ ѨӊҽӁ Ҹ ҲҸѸ, ҍҲҸ, ҰҸ ҡҏҲѹ ҦѦґҽ, ӊ ҔґѹӁѹҔѦ ҸҍѦӁҧ ӁѦҰґҸѽҸ, ҍҲҸ ҋҲҝҽӊѹӁҡҔѹҮ ҏѸӁѹ҃ҽ ѹ ӉӁҽѸѦӁѹҲҸҡҲҧ ѹ ҍҲҸ ѹѸѦҲҧ ҡ ӁѹѸ ҦѦґҸ ҍҝѦӉӊӃҍҽҮӁҸ ҰҝѹҾҲӁҸ. (ҷҏґҊҽҔҸӊ, ҴҽҡҲѦҝ ѹ ҴҽҝҊҽҝѹҲҽ)

(20P)

Zatar’ siNj nieco w pamiNjci przeklNjty diabelski kot, nie straszy’a juǻ odciNjta g’owa, a Iwan, zarzuciwszy rozwaǻania o niej, rozmyǧla’ o tym, ǻe w klinice jest w gruncie rzeczy zupe’nie nieǺle, ǻe Strawiǝski to bardzo mƾdry cz’owiek i w ogóle s’awa, ǻe mieǀ z nim do czynienia to wielka przyjemnoǧǀ.

(20Cz) Vzpomínka na DŽábelského kocoura se zmírnila a už ho nedLJsila uǥíznutá hlava. Pǥestal na ni myslet a napadlo ho, že v podstatLJ to není na zdejší klinice tak zlé, Stravinský že je kapacita a hlava otevǥená a jednat s ním je pǥíjemné. (20Sb) ѢѦҔҽҔҸ ҎѦ ѹӉқґѦҦѦҸ ҏ ѸѹҡґѹѸҽ ҰҝҸҔґѦҲѹ ӊҝҽҗҔѹ ѸҽҍҽҔ, ӁѹҎѦ ԄԂ ӊѹҗѦ Ұґҽҗѹґҽ ҸҦҝҏқҒѦӁҽ Ҋґҽӊҽ ѹ, ӁҽҰҏҡҲѹӊҗѹ ѹ ҰҝѦҡҲҽӊҗѹ ҡҽҡӊѹѸ Ҧҽ Ѹѹҡґѹ Ӂҽ Ѭҏ, ѨӊҽӁ ҰҸҍѦ Ҧҽ ҝҽӉѸѹҗҒҽ Ҹ ҲҸѸѦ, Ҧҽ ҏ ҡҽѸҸҎ ҡҲӊҽҝѹ ҏ ҔґѹӁѹ҃ѹ ҏҸҰҗҲѦ ӁѹҎѦ ґҸҗѦ, Ҧҽ ҎѦ ҋҲҝҽӊѹӁҡҔѹ – ҰҽѸѦҲҽӁ ѹ ӉӁҽѸѦӁѹҲ ҍҸӊѦҔ ѹ Ҧҽ ҎѦ ӊѹҗѦ ӁѦҊҸ ҰҝѹҎҽҲӁҸ ҝҽӉҊҸӊҽҝҽҲѹ ҡҽ ѬѹѸ. (20Hr) U njegovu sjeǀanju nekako se smekšao prokleti, demonski maǁak, više ga nije plašila odrezana glava i, zaboravivši na nju, Ivan je poǁeo razmišljati kako zapravo na klinici nije loše, kako je Stravinski mudrac i znamenit ǁovjek i kako je vrlo ugodno s njim imati posla.

366

CORNELIA E. KEIJSPER

(20Bg) ҭҝҸҔґѦҲѹҾҲ ҦҾӊҸґҡҔѹ ҔҸҲҽҝҽҔ ӁҾҔҽҔ ҡѦ ҡѸѦҔҍѹ ӊ ҰҽѸѦҲҲҽ Ѹҏ, ӊѦҍѦ ӁѦ Ԅӹ ҰґҽҗѦҗѦ ҸҲҝҾӉҽӁҽҲҽ Ҋґҽӊҽ ѹ ҔҽҲҸ ҡѦ ҸҡӊҸқҸҦѹ ҸҲ ѸѹҡӆґҲҽ Ӊҽ ӁѦҾ, ѨӊҽӁ ҡѦ ҝҽӉѸѹҡґѹ, ҍѦ ӊҡӆӂӁҸҡҲ ӊ ҔґѹӁѹҔҽҲҽ ӁѦ Ѧ ӁѹҔҽҔ ґҸҗҸ, ҍѦ ҋҲҝҽӊѹӁҡҔѹ Ѧ ҏѸӁѹҔ ѹ ҡӊѦҲѹґҸ ѹ Ҧҽ Ҹқӂҏӊҽҗ ҡ ӁѦҊҸ Ѧ ѸӁҸҊҸ ҰҝѹҾҲӁҸ. (20BR) ѢѦҾҔ ҰҽҦҽқҝҙҕ ҏ ҰҽѸҾ҃ҟ ҰҝҽҔґҾҲӃ Ҧ’ҾқҽґҧҡҔҟ ҔҸҲ, ӁѦ ҰҽґҸѽҽґҽ қҸґѦҮ ҽҦҝҙӉҽӁҽҾ Ҋҽґҽӊҽ, ҟ, ҽҦҝӃӁҏҕҗӃ ҾѦ, Ұҽҍҽҕ ҝҽӉҦҏѸӊҽ҃ҧ ӎӊҽӁ Ұҝҽ ҲҸѦ, җҲҸ Ӂҽ ҡҽѸҽҮ ҡҰҝҽӊѦ ӁҾҦҝҙӁӁҽ ҟ ҕ қҽґҧӁҟ҃Ӄ, җҲҸ ҋҲҝҽӊҟӁҡҔҟ ҝҽӉҏѸӁӃ ҟ ӉӁҽҔҽѸҟҲӃ ҟ җҲҸ қӃ҃ҧ ҏ ҾҊҸ ӁҽҦӉӊӃҍҽҮ ҰҝҸҡҲҽ ҟ ҰҝӃѦѸӁҽ. Expressions that in one language use OS order can be absent in other languages, for example the following so-called ironic negation (ҹѸѦґѦӊ 1958) in ѨґҧӍ ѹ ҭѦҲҝҸӊ: (21R)

ҫҍѦӁҧ ѸѦӁҾ ҙҲҸ ҰҏҊҽѦҲ! (ѨґҧӍ ѹ ҭѦҲҝҸӊ, 12 ҡҲҏґҧѦӊ)

(21P)

– Nic a nic siNj nie bijNj. Na Kaukazie nie sƾ one jadowite.

(21Cz)

“TLJch se zrovna leknu! A na Kavkaze nejsou jedovatí!”

(21Sb)

– Šta imam da se plašim! Oni na Kavkazu nisu otrovni!

All this amounts to saying that OS orders in general, or their frequency, are not a relevant issue in the languages illustrated here. 8. Other tolerable changes of linear-intonational hierarchy Two regular possibilities of changing the linear-intonational hierarchy must be mentioned that almost do not affect informational content, although they are not as automatic as the rules exemplified in section 7: the position of the word with the last accent can be changed, and in some cases the order of two accented items can be reversed. The result of these operations can be that one language has OS and another SO (with various V positions). First, the last accented item is usually placed in final position in written sentences, but in Czech, for example, this rule is obeyed more consistently than in Russian (Adamec 1956: 182-185, Kǥížková 1953: 295-298, Keijsper 1985: 363). As a result we find regular word order differences as in example (3) and (22): (22R)

ғҸҰҝҸҡӃ ґӀқӊѹ ѹ ҡѸѦҝҲѹ ӁѦ ӊҸґӁҸӊҽґѹ ѨҰҰҸґѹҲҽ ҴҽҲӊѦѦӊѹҍҽ ғҸҝҸқҧҾӁѹӁҸӊҽ, ѽҸҲҾ ҙҲѹѸѹ ӊҸҰҝҸҡҽѸѹ ҰҸ ҝҸҦҏ ҡӊҸѦҮ ҡґҏңқӃ ҸӁ ӊѦҦҽґ ҡ ҦѦӊҾҲѹ ҏҲҝҽ ҦҸ ҰҾҲѹ ӊѦҍѦҝҽ ѦңѦҦӁѦӊӁҸ ҡ ҰҸґҍҽҡҸӊӃѸ ҰѦҝѦҝӃӊҸѸ ҦґҾ ӉҽӊҲҝҽҔҽ. (ѨґҧӍ ѹ ҭѦҲҝҸӊ, 12 ҡҲҏґҧѦӊ)

FROM OVS ORDER TO CONVERSE STRUCTURE IN RUSSIAN

367

(22Cz) Hypolita MatLJjoviǁe Vrabǁinského otázky lásky a smrti nevzrušovaly, tǥebaže se jimi zabýval z titulu svého úǥadu dennLJ od devíti do pLJti s pDzlhodinovou polední pǥestávkou. (22P)

Kwestie mi’oǧci i ǧmierci bynajmniej nie wzrusza’y Hipolita Matwiejewicza Worobjaninowa, mimo iǻ zajmowa’ siNj nimi z urzNjdu codziennie od 9-tej rano do 5-tej po po’udniu z pó’godzinnƾ przerwƾ na drugie ǧniadanie.

(22Sb)

Pitanja ljubavi i smrti nisu baš mnogo muǁila Ipolita Matvejeviǁ Vorobjanjinova, mada se on, veǀ po prirodi svoga posla, upravo s tim pitanjima susretao svakodnevno od devet ujutru do pet posle podne, s prekidom od pola sata za doruǁak.

The informational contents of (22R) and (22Cz) are so close that in the Mathesian tradition the difference is called “stylistic” (so-called subjective and objective word order, respectively). This is incorrect (see Keijsper 1985: 313 ff.), but the difference need not be essential in translations. The case mainly pertains to Kovtunova’s group 4 of section 5 above, because Russian relatively often has the last accent on a verb in non-final position there. Other languages can make the same choice as Czech, and Russian can choose an “objective” word order in translations of “subjective” word order. The second possibility of changing the linear-intonational hierarchy almost without affecting informational content consists in interchanging the order of two accented constituents. It occurs relatively often in Kovtunova’s group 2 of section 5. ѨґҧӍ ѹ ҭѦҲҝҸӊ, for example, systematically use SVO instead of OVS with the verb ҸѽӊҽҲѹҲҧ, and translations change it back again, as it were: (23R)

ѢѦҲѦҝҰѦӁѹѦ ҸѽӊҽҲӃӊҽґҸ ҸҲ҃ҽ ҨѦҦҸҝҽ. ҭҸҦ ҰҸґҸӀ ҏ ӁѦҊҸ Ӊҽ ӊѹҲҸҮ җӁҏҝҸҔ қӃґ ӉҽҲҔӁҏҲ ҲҸҰҸҝѹҔ. ҫҲѦ҃ ҨѦҦҸҝ ҡѦґ ҝҾҦҸѸ ҡ ҔҏҍѦҝҸѸ ѹ, ҰҸѸѹӁҏҲӁҸ ҸҊґҾҦӃӊҽҾҡҧ Ӂҽ ҡҲҏґҧҾ, ӊӃѦѽҽґ Ҕ ҷҽҲҏѸҏ. (ѨґҧӍ ѹ ҭѦҲҝҸӊ, 12 ҡҲҏґҧѦӊ)

(23P)

Ojca Fiodora rozsadza’a po prostu niecierpliwoǧǀ.

(23Cz) Otce Fjodora se zmocnila netrpLJlivost. (23Sb)

Oca Fjodora obuzme nestrpljenje.

Kovtunova mentions type SVO as a so-called expressive (“subjective”) variant of OVS (although she adds that the degree of expressiveness is weak (ҘҸӊҲҏӁҸӊҽ 1976: 175), which may mean that she has in mind SVO with the second accent lower than the first accent. Alternatively, the SVO type is Kovtunova’s ҭҽӁѹҔҽ ҸѽӊҽҲѹґҽ / ӊѦҡҧ ҊҸҝҸҦ (ҘҸӊҲҏӁҸӊҽ 1976: 175-176), where the verb is included into the Theme; it can, however, be newly introduced informa-

368

CORNELIA E. KEIJSPER

tion, just as the verb in SV (ѩҸңҦҧ ѹҦѦҲ) can be included in the scope of the accent on the subject. At any rate, the possibility is based on the fact that (23R), for example, has two intonational predications (two full accents), conveying, approximately: “there was impatience” and “it gained control over Father Fyodor”. If the verb is of the type that allows [O[VS]] (see section 5), viz. is relatively empty in that it only fills in the meaning of the linear-intonational hierarchy somewhat more precisely, then SVO can amount to the same as OVS, although it separates intonationally the introduction of the existence/presence of S and of the verbal meaning (via an accent on O), whereas OVS does not. If the verb is of another type the difference is unmistakable, to the effect that SVO can be possible when OVS hardly is (see further section 11, point 9). There exists an attempt to explain the difference in contextual connections created by SVO and OVS when they are “almost the same” (as in (23)). ҋѦґѹӊѦҝҡҲҸӊҽ ѹ ҭҝҸӉҸҝҸӊҽ (1992: 222-224) analyze the type SVO (or SVO with pronominal O). It occurs “ӊ ҲѦѽ ҔҸӁҲѦҔҡҲҽѽ, ӊ ҔҸҲҸҝӃѽ ҸҰѹҡӃӊҽѦѸҸѦ ҍҏӊҡҲӊҸ ҡҽѸҸ ҡґҏңѹҲ ҰҝѹҍѹӁҸҮ ҦґҾ ҔҽҔѹѽ-ҲҸ ҰҸҡґѦҦҏӀӂѹѽ ҦѦҮҡҲӊѹҮ «ҍҏӊҡҲӊҏӀӂѦҊҸ ҡҏқӆѦҔҲҽ»”, which “ҰҸӉӊҸґҾѦҲ ѦѸҏ ӉҽӁҾҲҧ ҃ѦӁҲҝҽґҧӁҸѦ ѸѦҡҲҸ ӊ ҔҸѸѸҏӁѹҔҽҲѹӊӁҸҮ ҰѦҝҡҰѦҔҲѹӊѦ”; and “ҔҸҊҦҽ ҰҝѦҦҗѦҡҲӊҏӀӂѹҮ ҔҸӁҲѦҔҡҲ ҡҸҸқӂҽѦҲ Ҹ ҰҸҡҲѦҰѦӁӁҸѸ ӁҽҔҽҰґѹӊҽӁѹѹ ҰҝѦҦҰҸҡӃґҸҔ ҦґҾ ӊҸӉӁѹҔӁҸӊѦӁѹҾ ҡҸҸҲӊѦҲҡҲӊҏӀӂѦҊҸ ҍҏӊҡҲӊҽ, ҍҲҸ ѹ ҦѦґҽѦҲ ҍҏӊҡҲӊҸ «҃ѦӁҲҝҽґҧӁӃѸ ҰѦҝҡҸӁҽңѦѸ»”; and “ҲҝѦҲҧѦҮ ѹ ҸҡӁҸӊӁҸҮ ҰҝѦҦҰҸҡӃґҔҸҮ […] ѸҸңѦҲ ҾӊґҾҲҧҡҾ ҡҲҝѦѸґѦӁѹѦ ҽӊҲҸҝҽ ҰҸҦҍѦҝҔӁҏҲҧ ҰҸґӁҏӀ ҸқҏҡґҸӊґѦӁӁҸҡҲҧ ӊҸӉӁѹҔӁҸӊѦӁѹҾ ҍҏӊҡҲӊҽ ӊӁѦҗӁѹѸѹ ҏҡґҸӊѹҾѸѹ, ҸқӆѦҔҲѹӊӁӃѸѹ ҰҝѦҦҰҸҡӃґҔҽѸѹ, ӉҽҦҽӁӁӃѸѹ ҡҽѸҸҮ ҡѹҲҏҽ҃ѹѦҮ”. “ѢҽҰҝҸҲѹӊ, Ѧҡґѹ ӊҸӉӁѹҔӁҸӊѦӁѹѦ ҍҏӊҡҲӊҽ ӁѦ ҦѹҔҲҏѦҲҡҾ ҰҸҍҲѹ ҸҦӁҸӉӁҽҍӁҸ ҸқӆѦҔҲѹӊӁӃѸѹ ҰҝѦҦҰҸҡӃґҔҽѸѹ, ҸқҏҡґҸӊґѦӁҸ ӊ қҸґҧҗѦҮ ҡҲѦҰѦӁѹ ҸҡҸқѦӁӁҸҡҲҾѸѹ «ҍҏӊҡҲӊҏӀӂѦҊҸ ҡҏқӆѦҔҲҽ», ҲҸ қҸґѦѦ ѦҡҲѦҡҲӊѦӁӁҸ ӊӃқҝҽҲҧ ӊ ҔҽҍѦҡҲӊѦ ҸҡӁҸӊӁҸҊҸ ѽҽҝҽҔҲѦҝѹӉҏѦѸҸҊҸ «ҍҏӊҡҲӊҏӀӂѹҮ ҡҏқӆѦҔҲ»” (ҋѦґѹӊѦҝҡҲҸӊҽ ѹ ҭҝҸӉҸҝҸӊҽ 1992: 222-224). But, as appears from example (23) and others of the same type, the difference between SVO and OVS can sometimes be neglected in translations without serious consequences for the informational content that is conveyed. 9. Converse structure Although the Mathesian approach is a step forward with respect to earlier treatments of word order (section 5), it is not sufficient to cover all phenomena relevant to the study of OVS order, not even after the mistakes in the approach are eliminated (as I tried to do in Keijsper 1985). This is because the approach only compares various linear-intonational hierarchies of the same syntactic

FROM OVS ORDER TO CONVERSE STRUCTURE IN RUSSIAN

369

structure: ѨӊҽӁ ҍѹҲҽѦҲ ҔӁѹҊҏ, ѨӊҽӁ ҍѹҲҽѦҲ ҔӁѹҊҏ, ѨӊҽӁ ҔӁѹҊҏ ҍѹҲҽѦҲ, ѨӊҽӁ ҔӁѹҊҏ ҍѹҲҽѦҲ, ҘӁѹҊҏ ҍѹҲҽѦҲ ѨӊҽӁ, ҘӁѹҊҏ ҍѹҲҽѦҲ ѨӊҽӁ, and so on (all sentences together constitute a paradigm in Kovtunova’s sense – section 5). But the approach is unable to compare, say, ҋҰѦҔҲҽҔґҧ ҸҲґѹҍҽѦҲ ҸҝѹҊѹӁҽґҧӁҸҡҲҧ with ҋҰѦҔҲҽҔґҧ ҸҲґѹҍҽѦҲҡҾ ҸҝѹҊѹӁҽґҧӁҸҡҲҧӀ: these have (more or less) the same linear-intonational hierarchy, but different syntactic structures (in the example: nominative-accusative sentence in OVS order vs. reflexive verb and corresponding case forms). Such pairs are called converse structures (“ҔҸӁӊѦҝҡѹӊӃ”) (e.g. үҰҝѦҡҾӁ 1967: 127 ff., 1995: 256 ff., ғҡѦӊҸґҸҦҸӊҽ ѹ ѩѦѸѦӁҲҧѦӊҽ 1997: 214 ff., ғҡѦӊҸґҸҦҸӊҽ 2000). Note especially that the term “converse structure” is not about word order (although the term “basic member of a set of converse structures” can include word order – see below, note 7). Thus, ҫҝѹҊѹӁҽґҧӁҸҡҲҧ ҸҲґѹҍҽѦҲ ҡҰѦҔҲҽҔґҧ (SVO) and ҫҝѹҊѹӁҽґҧӁҸҡҲҧӀ ҸҲґѹҍҽѦҲҡҾ ҡҰѦҔҲҽҔґҧ (instrumental case – reflexive verb – subject) illustrate the same pair of converse structures. It is converse structure that is the relevant issue when one compares translations: except for the cases mentioned in sections 7 and 8, if a sentence must be adapted to the possibilities of the target language in a translation, it is better to retain the linear-intonational hierarchy of the original text as far as possible (this is the invariant of translation), and to adapt the syntactic hierarchy to the possibilities of the target language. This phenomenon is well known from translations of English into German, for example (Zydatiʃ 1974: 41): (24E) (24G)

Examples like these were used by Boas to … (“passive”) Beispiele wie diese führte Boas an, um … (OVS)

The same holds true for translations from and into Slavic languages. The real issue of OVS sentences, then, is not the expression of subject and object functions (section 2), and not the informational content of sentences which makes that this order can be the best choice for creating contextual relations and can even be the “syntagmatically independent member of the paradigm” (section 5); the real issue is the choice between OVS and other members of the set of converse structures with more or less the same informational content: when and why is one preferred over others? Or, to restrict the issue to the theme of the present article: is there a reason to assume that OVS order is avoided when such a choice is made? The term “converse structures” is broader than “active-passive” and is also applicable to languages without active-passive oppositions. It also includes replacement of noun phrases by prepositional phrases, replacement of finite verb forms containing a certain morpheme by other word sorts with the same morpheme plus another verb, and even (with sometimes larger semantic differences) so-called lexical converses. Examples:

370

CORNELIA E. KEIJSPER

OVS (R en P) vs. Prepositional Phrase plus VS (reflexive in Cz) (Cz en Sb): (25R)

ѷҸ, ҍҲҸ ҸӁ ҏӊѹҦѦґ, қӃґҸ ҰҸґӁҸҮ ҦґҾ ӁѦҊҸ ӁѦҸңѹҦҽӁӁҸҡҲҧӀ. ҭѹҡҧѸѦӁӁӃҮ ҡҲҸґ, Ӊҽ ҔҸҲҸҝӃѸ ҡѹҦѦґ ҰҝҸҔҏҝҸҝ, ҸҔҝҏңҽґѹ ҍґѦӁӃ ѸҸҊҏҍѦҮ ҸҝҊҽӁѹӉҽ҃ѹѹ “ҴѦҍҽ ѹ Ҹҝҽґҽ”. ҋҏҦҾ ҰҸ ѹѽ ңѦҡҲҽѸ ѹ ҰґҽҔҡѹӊӃѸ ҊҸґҸҡҽѸ, ҸӁѹ ҡҸӉӁҽӊҽґѹҡҧ ӊҸ ӊҡѦѸ. (ѨґҧӍ ѹ ҭѦҲҝҸӊ, 12 ҡҲҏґҧѦӊ)

(25P)

Biurko, przy którym siedzia’ prokurator, otaczali cz’onkowie potNjǻnej organizacji “Miecza i lemiesza”. Wnioskujƾc z ich gestów, przyznawali siNj do wszystkiego.

(25Cz) Kolem psacího stolu, za nímž sedLJl prokurátor, kupili se ǁlenové mocné organizace Meǁe a rádla. (25Sb)

Oko pisaǀeg stola za kojim je sedeo tužilac stajali su ǁlanovi moǀne organizacije “Maǁ i ralica”.

OSV (R) vs. the same morpheme in a substantive plus the verb ‘to have’ (Cz): (26R)

ҫҦѹӁ ҔҝѦҡҲҧҾӁѹӁ, ҰҸ ӍҽѸѹґѹѹ ҭѹѽҽ, ҔҸҲҸҝҸҊҸ ҲҽҔҸҮ ӊѹӉѹҲ ӁѹҍҏҲҧ ӁѦ ҸқҝҽҦҸӊҽґ, ӁѦ ӊӉҾґ ӊҸҡѦѸӁҽҦ҃ҽҲҧ ҔҝҸӁ, ҔҸҲҸҝӃѦ ҔҽӉӁҽ ѦѸҏ Ҧҽӊҽґҽ Ӊҽ ҰҸҲҸҰҲҽӁӁӃѦ ҰҾҲҧ ѸѦҝ ҰҸґҾ, ӉҽѽҸҲѦґ, ҊҸҡҰҸҦѹӁ ҸқѦҝґѦҮҲѦӁҽӁҲ, ҡҏҦѹҲҧҡҾ ѹ ҰҸґҏҍѹґ Ӊҽ ҙҲҸ ӊҸҡѦѸӁҽҦ҃ҽҲҧ ѸѦҡҾ҃Ѧӊ.

(26Cz) NLJjakej sedlák Pícha nemLJl žádnou radost z tý návštLJvy a nepǥijmul vod eráru osmnáct korun náhrady za zdupanejch pLJt mLJr pole, chtLJl se vám, pane obrlajtnant, soudit a dostal za to osmnáct mLJsícDz. (Hašek, Švejk) Such differences are, of course, not restricted to OVS orders, for example: Reflexive verb (R, P, Sb) vs. the same morpheme in a substantive plus the verb ‘to have’ (Cz): (27R)

҈ҽѸѦҲҧ, ҍҲҸ ӊҡѦ ҡҲҏґҧҾ ҰҸҰҽґѹ Ҕ ҸҦӁҸѸҏ ҍѦґҸӊѦҔҏ, ҍѦҊҸ Ҿ ӁѹҔҽҔ ӁѦ ҸңѹҦҽґ (қҸҾґҡҾ, ҍҲҸ ҡҲҏґҧҾ ҰҸҰҽҦҏҲ ӊ ҝҽӉӁӃѦ ѸѦҡҲҽ). Ѥ ҸҍѦӁҧ ҙҲҸѸҏ ҸқҝҽҦҸӊҽґҡҾ. (ѨґҧӍ ѹ ҭѦҲҝҸӊ, 12 ҡҲҏґҧѦӊ)

(27P)

Bardzo siNj z tego powodu ucieszy’em.

(27Cz) MLJl jsem z toho velkou radost. (27Sb)

Jako sam se obradovao zbog toga.

Lexical converses: ‘to meet’ (R, Bg, BR – all three in a passive converse: ‘was met by’; P SVO – with another informational content; Cz OVS) vs. ‘to wait’ (OVS) (Sb, Hr):

FROM OVS ORDER TO CONVERSE STRUCTURE IN RUSSIAN

371

(28R)

ғ ҏҊґҏ ҦҸҰѹӊҽґҽ ҔҽҔҽҾ-ҲҸ ҔҸѸҰҽӁѹҾ, ѹ ӊ ҃ѦӁҲҝѦ ѦѦ ҡҏѦҲѹґҡҾ ӉӁҽҔҸѸӃҮ ҔҸӁӍѦҝҽӁҡҧѦ ӊ ҲӀқѦҲѦҮҔѦ ѹ ҡ қҸҔҽґҸѸ “үқҝҽҏ” ӊ ҝҏҔѦ. ҩӀѽѹӁ, ҸқҝѦѸѦӁѦӁӁӃҮ ҰҸґҸҲѦӁ҃ҽѸѹ, қӃґ ӊҡҲҝѦҍѦӁ үҝҍѹқҽґҧҦҸѸ үҝҍѹқҽґҧҦҸӊѹҍѦѸ ҸҍѦӁҧ ҰҝѹӊѦҲґѹӊҸ ѹ ҲҸҲҍҽҡ ѹӉқҽӊґѦӁ ҸҲ ҰҝҸҔґҾҲӃѽ ҲҝҾҰҸҔ. (ҷҏґҊҽҔҸӊ, ҴҽҡҲѦҝ ѹ ҴҽҝҊҽҝѹҲҽ)

(28P)

Archibald Archibaldowicz serdecznie powita’ ob’adowanego serwetami Riuchina i niezw’ocznie uwolni’ poetNj od przeklNjtych szmat.

(28Cz) SviǞkova obtíženého utLJrkami srdeǁnLJ uvítal Archibald Archibaldoviǁ a okamžitLJ ho zbavil nepǥíjemného nákladu. (28Sb) ҩҎҏѽѹӁҽ, ӁҽҲҸӊҽҝѦӁҸҊ ҏқҝҏҡѹѸҽ, ҦҸҍѦҔҽ үҝҍѹқҽґҦ үҝҍѹқҽґҦҸӊѹҍ ҒҏқҽӉӁҸ ѹ ҸҦѸҽѽ Ҋҽ ҸҡґҸқҸҦѹ ҰҝҸҔґѦҲѹѽ ҔҝҰҽ. (28Hr) Rjuhina, natovarena ubrusima, ljubazno je doǁekao Arǁibald Arǁibaldoviǁ i odmah ga izbavio od prokletih krpa. (28Bg) ѢҽҲҸӊҽҝѦӁѹҾҲ ҡ ҰѦҗҔѹҝѹҲѦ ҩӀѽѹӁ қѦ ҰҸҡҝѦӂӁҽҲ ґӀқѦӉӁҸ ҸҲ үҝҍѹқҽґҦ үҝҍѹқҽґҦҸӊѹҍ ѹ ӁѦӉҽқҽӊӁҸ ҸҡӊҸқҸҦѦӁ ҸҲ ҰҝҸҔґѦҲѹҲѦ Ұҽҝ҃ҽґѹ. (28BR) ҩҸѽҟӁ, ӁҽҊҝҏңҽӁӃ ҝҏҍӁҟҔҽѸҟ, қӃҕ ҡҏҡҲҝҙҲӃ үҝҍӃқҽґҧҦҽѸ үҝҍӃқҽґҧҦҽӊҟҍҽѸ ӊѦґҧѸҟ ҰҝӃҾӉӁҽ ҟ ҽҦҝҽӉҏ ң ҰҽӉқҽҕґѦӁӃ ҽҦ ҰҝҽҔґҾҲӃѽ ҽӁҏҍ. Comparisons between languages add additional problems. For example, Russian dictionaries give both ҦҸңҦҽҲҧҡҾ and ӊҡҲҝѦҲѹҲҧ (and ҡѽӊҽҲѹҲҧ) as translations of ҦҸҍѦҔҽҲѹ (28Sb/Hr), so perhaps we may only call these Russian translations lexical converses of each other (with large semantic differences). A special type of translation has ‘to be’ or ‘to have’ or another relatively empty verb in a translation of a verb with at first sight more lexical content, or vice versa. The verb ҸҲґѹҍҽҲҧҡҾ specializes in this use: (29R)

ғҸҔҏґҧҡҔѹҮ ѸѹѸҸѽҸҦҸѸ ҰҸҦѸѦҲѹґ, ҍҲҸ ҡҝѦҦѹ ҙҲҸҊҸ ҡқҸҝѹӂҽ ӊҸӉӁѹ҃ ҍѦґҾҦҧ ӉӁҽҲӁӃѽ ҊҸҡҰҸҦ ӊӃҦѦґҾґҽҡҧ ӊҽңӁҸҮ ҡҲѦҰѦӁӁҸҡҲҧӀ, ҔҏҍѦҝҽ қҽӁҔѹҝҸӊ ҰӃҲҽґѹҡҧ ӊѦҝѽҸӊҸҦѹҲҧ, ҍҲҸ ӊӃӉӃӊҽґҸ ѹӉҦѦӊҔѹ ѹ қҝҽӁҧ, ѹӉӊҸӉҍѹҔѹ ңѦ ҸҲґѹҍҽґѹҡҧ ҡҽѸҸҏӊѦҝѦӁӁҸҮ қҸҮҔҸҡҲҧӀ. ҘҏҍѦҝҽ ӁҽѦѸӁӃѽ ҔҽҝѦҲ ҦѦҝңҽґѹҡҧ ҸҡҸқӁҾҔҸѸ, қҝѦӉҊґѹӊҸ ҡҲҸҝҸӁѹґѹҡҧ ҸҡҲҽґҧӁӃѽ, ҽ ҲѦ, ӊ ҡӊҸӀ ҸҍѦҝѦҦҧ, қҝѦӉҊҸӊҽґѹ ѹѸѹ.

(29P)

Dostrzeg’ teǻ mimochodem, ǻe w gronie zebranych woǺniców s’uǻba wielkich panów zachowywa’a siNj w sposób pe’en godnoǧci, bankierscy chcieli rej wodziǀ, za co im wymyǧlano, a doroǻkarze byli najrezolutniejsi. Furmani zaǧ powozów najNjtych trzymali siNj blisko siebie, gardzƾcy resztƾ i przez niƾ pogardzani. (Prus, Lalka)

372

CORNELIA E. KEIJSPER

(29Sb) … қҽӁҔҽҝҡҔѹ ҔҸҍѹҎҽҗѹ ѽҲѦґѹ ҡҏ Ҧҽ ӊҸҦѦ ҊґҽӊӁҏ ҝѦҍ, Ӊҽ җҲҽ ҡҏ ѹѽ ҦҝҏҊѹ ӁҽҰҽҦҽґѹ, ҽ ҏ ҲҸѸѦ ҡҏ ӍѹҎҽҔѦҝѹҡҲѹ қѹґѹ ӁҽҎҸҦґҏҍӁѹҎѹ. 10. Basic member? The notion “set of converse structures” introduces another type of paradigm than Kovtunova’s (section 5) set of sentences with the same syntactic structure and different linear-intonational hierarchies. In this new type of paradigm, viz. a set of different syntactic structures with more or less the same linearintonational hierarchy, too, a basic member is recognized: authors dealing with converse structures (үҰҝѦҡҾӁ 1967, 1995, ғҡѦӊҸґҸҦҸӊҽ ѹ ѩѦѸѦӁҲҧѦӊҽ 1997, ғҡѦӊҸґҸҦҸӊҽ 2000, among others) find it necessary to determine which member of a pair of converse structures is basic and which derived, unless the relationship is one of parity, as in ҋҽҗҽ – қҝҽҲ ҘҸґѹ ҘҸґҾ – қҝҽҲ ҋҽҗѹ (ғҡѦӊҸґҸҦҸӊҽ 2000: 453). The procedure for doing so gives some information about semantic differences between converse structures. First, ғҡѦӊҸґҸҦҸӊҽ (2000: 36-39), following ҈ҸґҸҲҸӊҽ (1982: 122-132) starts from the notion of isosemy. An isosemic (prototypical) word has a referent of the type that conforms to the class of words to which the word belongs: nouns denote things, verbs activities or processes, and so on. For example, ҘҝҽҡѹӊҽҾ ҦѦӊҏҗҔҽ Ҋҝҽ҃ѹҸӉӁҸ ҲҽӁ҃ҏѦҲ is isosemic, in contrast to ѷҽӁѦ҃ ҔҝҽҡѹӊҸҮ ҦѦӊҏҗҔѹ Ҋҝҽ҃ѹҸӉѦӁ, where the action is referred to by a noun, or ѷҽӁѦ҃ ҔҝҽҡѹӊҸҮ ҦѦӊҏҗҔѹ – ҡҽѸҽ Ҋҝҽ҃ѹҸӉӁҸҡҲҧ where a property of the action is referred to by a substantive (ғҡѦӊҸґҸҦҸӊҽ 2000: 36-39). A basic member of a set of converse structures is an isosemic structure. 6 This reasoning explains to a certain extent why ҴҽґҧҍѹҔ қѦҊґҸ ҍѹҲҽѦҲ, for example, is easier to understand than ѯҲѦӁѹѦ ѸҽґҧҍѹҔҽ ҸҲґѹҍҽѦҲҡҾ қѦҊґҸҡҲҧӀ and comparable non-isosemic constructions characteristic of Russian scientific and journalistic texts (ғҡѦӊҸґҸҦҸӊҽ 2000: 39). Next, if there is an agent or a possible agent, the basic member has the constituent concerned in the nominative case: ҘҏҔґӃ ӊҝҽӂҽӀҲ ҊґҽӉҽѸѹ is probably not agentive, but the same structure in ѢҽґѦҲҍѹҔ ҡӊѹҝѦҰҸ ӊҝҽӂҽґ ҊґҽӉҽѸѹ shows that it can have an agentive interpretation. Therefore, this structure is 6

More precisely, if all isosemic words are used in syntactic positions which agree with their semantic roles an isosemic structure is isomorphic, otherwise it is not isomorphic: Ѣҽ ғү҈Ѧ ҝҽқҸҍѹѦ ӊӃҰҏҡҔҽӀҲ «ѭѹҊҏґѹ» is isomorphic because the place (Ѣҽ ғү҈Ѧ) is referred to by a prepositional phrase with the preposition Ӂҽ, the agent is referred to by the nominative case, the action by the verb, and the object by the accusative case. If not all isosemic words are in syntactic positions which conform to their semantic role the isosemic structure is not isomorphic: ҩҽқҸҍѹѦ ғү҈ҽ ӊӃҰҏҡҔҽӀҲ ҽӊҲҸѸҸқѹґѹ «ѭѹҊҏґѹ» is not isomorphic, because it has the place (ғү҈ҽ) in the syntactic position of a non-agreeing attribute (ғҡѦӊҸґҸҦҸӊҽ ѹ ѩѦѸѦӁҲҧѦӊҽ 1997: 61).

FROM OVS ORDER TO CONVERSE STRUCTURE IN RUSSIAN

373

basic, while Ѱ ҔҏҔҸґ ӊҝҽӂҽӀҲҡҾ ҊґҽӉҽѸѹ / Ѱ ӁҽґѦҲҍѹҔҽ ҡӊѹҝѦҰҸ ӊҝҽӂҽґѹҡҧ ҊґҽӉҽ is derived, because it cannot have an agentive interpretation (үҰҝѦҡҾӁ 1995: 269, who calls these quasi-converses). When no agent or possible agent is involved, it is the most active role which is considered to be basically in the nominative: In ҈ѦѸґӀ ҰҸҔҝӃґ ҡӁѦҊ Ռ ҈ѦѸґҾ ҰҸҔҝӃґҽҡҧ ҡӁѦҊҸѸ (ғҡѦӊҸґҸҦҸӊҽ 2000: 456, 458) or ҫӉѦҝҸ ҸҔҝҏңҽӀҲ ҊҸҝӃ Ռ ҫӉѦҝҸ ҸҔҝҏңѦӁҸ ҊҸҝҽѸѹ (ғҡѦӊҸґҸҦҸӊҽ 2000: 457) the first member is basic because there the subject has the more active role. Note especially that the order OVS plays no role in determining which member of the pair is basic. In sentences referring to emotional states (ҍҏӊҡҲӊҽ-ҡҸҡҲҸҾӁѹҾ) the reflexive verb (e.g. ӊҸґӁҸӊҽҲҧҡҾ) is considered basic and the nominative-accusative version (e.g. ӊҸґӁҸӊҽҲҧ) derived, because the emotions involved “ӊҡѦҊҦҽ ҲҝҽҔҲҏӀҲҡҾ ҾӉӃҔҸѸ ҔҽҔ ӊӃӉӃӊҽѦѸӃѦ ӊ ҡҏқӆѦҔҲѦ ҸҰҝѦҦѦґѦӁӁҸҮ Ҹ҃ѦӁҔҸҮ ӁѦҔҸҲҸҝҸҊҸ ҡҸқӃҲѹҾ, ҽ ӁѦ ӊҸӉӁѹҔҽӀӂѹѦ ҡҽѸѹ ҰҸ ҡѦқѦ” (үҰҝѦҡҾӁ 1995: 271, following ѨҸҝҦҽӁҡҔҽҾ 1970). The reflexive verb in such pairs is sometimes called reactive, i.e. it refers to a self-initiated reaction to a stimulus (e.g. Keijsper 2003: 174, 186). So even in Ѥ ӊҸҡѽѹӂҽӀҡҧ ҴҽҝѹѦҮ Ҙҽґґҽҡ Ռ ҴѦӁҾ ӊҸҡѽѹӂҽѦҲ ҴҽҝѹҾ Ҙҽґґҽҡ, the first member is basic, although Maria Callas has the active role of singing, because the activity of singing is not aimed at the initiation of my emotion: the latter is self-initiated, with the activity of singing as a stimulus (in my words, which abbreviate ғҡѦӊҸґҸҦҸӊҽ 2000: 453). In an earlier publication (ғҡѦӊҸґҸҦҸӊҽ 1976: 70) Vsevolodova adhered to the opposite point of view (the nominative-accusative version is basic), which shows that disagreement about such issues is possible. Finally, there can be pairs of nominal and verbal stems such as ҴӃ ҝҽҦӃ ӊҡҲҝѦҍѦ ҡ ӊҽѸѹ Ռ Ѣҽҡ ҝҽҦҏѦҲ ӊҡҲҝѦҍҽ ҡ ӊҽѸѹ (ғҡѦӊҸґҸҦҸӊҽ 2000: 461), as well as series like Ѥ ѹӁҲѦҝѦҡҏӀҡҧ ѹҡҲҸҝѹѦҮ Ռ ҴѦӁҾ ѹӁҲѦҝѦҡҏѦҲ ѹҡҲҸҝѹҾ Ռ ҴӁѦ ѹӁҲѦҝѦҡӁҸ ѹҡҲҸҝѹҾ (ғҡѦӊҸґҸҦҸӊҽ 2000: 456) or Ѥ ӊҸҡѽѹӂҽӀҡҧ ҴҽҝѹѦҮ Ҙҽґґҽҡ Ռ ҴѦӁҾ ӊҸҡѽѹӂҽѦҲ ҴҽҝѹҾ Ҙҽґґҽҡ Ռ Ѥ ӊҸҡѽѹӂѦӁ ҰѦӁѹѦѸ Ҵҽҝѹѹ Ҙҽґґҽҡ (ғҡѦӊҸґҸҦҸӊҽ 2000: 453, 456), with primary and secondary converses. Note, again, that word order does not belong to these considerations.7 Now, this new notion of basicness, like Kovtunova’s one (section 5), does not predict frequency and does not prescribe which member of a set of converse structures must be chosen. True, in ASPAC Russian reflexive verbs of emotion, 7

However, even word order is sometimes brought into the notion of a basic member of a set of converse structures: according to ғҡѦӊҸґҸҦҸӊҽ ѹ ѩѦѸѦӁҲҧѦӊҽ (1997: 61-62), an isosemic structure must also conform to the rule that the semantic subject (protagonist) is the Theme (what the authors call the ҰҝѦҦѹ҃ѹҝҏѦѸӃҮ ҔҸѸҰҸӁѦӁҲ), and the action the Rheme (ҰҝѦҦѹ҃ѹҝҏӀҗѹҮ ҔҸѸҰҸӁѦӁҲ). Thus, in this view ҩҽқҸҍѹѦ Ӂҽ ғү҈Ѧ ӊӃҰҏҡҔҽӀҲ ҽӊҲҸѸҸқѹґѹ «ѭѹҊҏґѹ» is isosemic, while ғӃҰҏҡҔҽӀҲ (ҝҽқҸҍѹѦ) Ӂҽ ғү҈Ѧ ҽӊҲҸѸҸқѹґѹ «ѭѹҊҏґѹ» or үӊҲҸѸҸқѹґѹ «ѭѹҊҏґѹ» (ҝҽқҸҍѹѦ) ӊӃҰҏҡҔҽӀҲ Ӂҽ ғү҈Ѧ are not. In this version, too, OVS can be basic.

374

CORNELIA E. KEIJSPER

for example, are more frequent than nominative-accusative sentences with the corresponding non-reflexive verbs. But it is not the case that ѹӁҲѦҝѦҡҸӊҽҲҧҡҾ, for example, is more frequent than ѹӁҲѦҝѦҡ or ѹӁҲѦҝѦҡӁҸ, as the idea of primary and secondary converses might suggest. It is also not the case that the basic member ҈ѦѸґӀ ҰҸҔҝӃґ ҡӁѦҊ is more frequent than the derived sentence ҈ѦѸґҾ ҰҸҔҝӃґҽҡҧ ҡӁѦҊҸѸ, but it is the case that the type ҫӉѦҝҸ ҸҔҝҏңҽӀҲ ҊҸҝӃ is more frequent than ҫӉѦҝҸ ҸҔҝҏңѦӁҸ ҊҸҝҽѸѹ, for example (in Russian originals in ASPAC). And, more importantly, there are differences between languages and verbs that are not explained by the notion “basic member of a set of converse structures”. So the notion of basicness in application to converse structures gives an impression of semantic differences (which participant has the most active role?), but it helps us no further than that. The whole idea that there must be a basic member is inspired by the idea that the relationship between members of a set of converse structures is one of transformation (үҰҝѦҡҾӁ 1967) or that they are syntactic synonyms (e.g. ѯҽҊѹӁҽ 1990). This inspiration tends to blur distinctions rather than bring them to light.8 We can study semantic differences between converse structures directly by regarding pairs like the following as semantic minimal pairs, without deriving one from the other (Keijsper 2003: 188 ff.): (30R) (31R) (32R) (33R) (34R) (35R)

Ҵ.ғ. һҸѸҸӁҸҡҸӊ ӹԙӰӹӴԂԊ ҴҸҡҔҸӊҡҔѹҮ ҏӁѹӊѦҝҡѹҲѦҲ Ҵ.ғ. һҸѸҸӁҸҡҸӊӃѸ ӹԙӰӹӴԂӰ ҴҸҡҔҸӊҡҔѹҮ ҏӁѹӊѦҝҡѹҲѦҲ or ҴҸҡҔҸӊҡҔѹҮ ҏӁѹӊѦҝҡѹҲѦҲ ӹԙӰӹӴԂԊ Ҵ.ғ. һҸѸҸӁҸҡҸӊ ҴҸҡҔҸӊҡҔѹҮ ҏӁѹӊѦҝҡѹҲѦҲ ӹԙӰӹӴԂӰ Ҵ.ғ. һҸѸҸӁҸҡҸӊӃѸ ҩѦҰѹӁ ӢԂԌӹӴԂԊԙԋ ӊѦҡѦґҸҮ, ҸҡҲҝҸҏѸӁҸҮ қѦҡѦҦѦ ѯҏҔҸӊҡҔҸҊҸ ҩѦҰѹӁҽ ӢԂԌӹӴԂԊԂ ӊѦҡѦґҽҾ, ҸҡҲҝҸҏѸӁҽҾ қѦҡѦҦҽ ѯҏҔҸӊҡҔҸҊҸ (҈ӊѦӉҦҽ 1994) (or another linear-intonational hierarchy)

Unfortunately, it is as yet uncommon to do so. Most linguistic models treat linear-intonational hierarchy as a phenomenon “more to the surface” than syntactic hierarchy: they first choose a syntactic hierarchy, and only after that com8

The term “synonyms” does not deny semantic differences in the terminology of the so-called Moscow school of semantics (see, for example, үҰҝѦҡҾӁ ѹ Ҧҝ. 1997). It rather means “set of semantic (near) minimal pairs”. Vsevolodova’s idea that word order must be brought under the same heading (see note 7) is still more confusing: it blurs the difference between linearintonational hierarchy and syntactic hierarchy instead of stating that the two are in Russian to a large extent independent (in contrast to English, where the former expresses the latter – see section 2).

FROM OVS ORDER TO CONVERSE STRUCTURE IN RUSSIAN

375

pare different linear-intonational possibilities within that choice (as Kovtunova in section 5). Also, a consequence of overestimating the importance of English (section 2) is the fact that many linguists change linear-intonational hierarchy if they compare converse structures: (36E) (37E)

Boas used examples like these to … Examples like these were used by Boas to… (= 24E)

(30R) (33R)

Ҵ.ғ. һҸѸҸӁҸҡҸӊ ӹԙӰӹӴԂԊ ҴҸҡҔҸӊҡҔѹҮ ҏӁѹӊѦҝҡѹҲѦҲ ҴҸҡҔҸӊҡҔѹҮ ҏӁѹӊѦҝҡѹҲѦҲ ӹԙӰӹӴԂӰ Ҵ.ғ. һҸѸҸӁҸҡҸӊӃѸ

If one does so, pairs such as (30)-(31), (32)-(33), (34)-(35) are not recognized as minimal pairs, and the separate contributions to meaning of linear-intonational hierarchy and of syntactic hierarchy in a language like Russian are not isolated (cf. also Keijsper 1985: 52-58). This isolation is in my view a prerequisite for progress in the investigation of these contributions. 11. Issues for investigation The investigation of converse structures in Slavic languages is too large a subject for a single article. I confine myself here to listing ten observations about differences between Slavic languages that require further research. 1. First, of course, apart from the possibilities mentioned in section 4 above, languages without nominal cases can have recourse to converse structures in order to retain a given linear-intonational hierarchy within the boundaries of their possibilities (R and all others vs. Bg): (38R)

ѷҽ, ґѹҗҧ ҲҸґҧҔҸ ҏӊѹҦѦґҽ ҔҸҲҽ, ґѦӉҏӂѦҊҸ ӊ ҲҝҽѸӊҽҮ, ҡҸ ӉґҸқҸҮ, ҸҲ ҔҸҲҸҝҸҮ ҦҽңѦ ҲҝҾҡґҽҡҧ, ӉҽҔҝѹҍҽґҽ: – ҘҸҲҽѸ ӁѦґҧӉҾ! ҋ ҔҸҲҽѸѹ ӁѦґҧӉҾ! ҷҝӃҡҧ! ҋґѦӉҽҮ, ҽ ҲҸ Ѹѹґѹ҃ѹӀ ҰҸӉҸӊҏ! Ѣѹ ҔҸӁҦҏҔҲҸҝҗҏ, Ӂѹ ҰҽҡҡҽңѹҝҸӊ ӁѦ ҰҸҝҽӉѹґҽ ҡҽѸҽҾ ҡҏҲҧ ҦѦґҽ: ӁѦ ҲҸ, ҍҲҸ ҔҸҲ ґѦӉѦҲ ӊ ҲҝҽѸӊҽҮ, ӊ ҍѦѸ қӃґҸ қӃ ѦӂѦ ҰҸґқѦҦӃ, ҽ ҲҸ, ҍҲҸ ҸӁ ҡҸқѹҝҽѦҲҡҾ ҰґҽҲѹҲҧ! (ҷҏґҊҽҔҸӊ, ҴҽҡҲѦҝ ѹ ҴҽҝҊҽҝѹҲҽ)

(38P)

Ani konduktorki, ani pasaǻerów nie zdziwi’o to, co by’o najdziwniejsze – nie to wiNjc, ǻe kot pakuje siNj do tramwaju, to by’oby jeszcze pó’ biedy, ale to, ǻe zamierza zap’aciǀ za bilet!

(38Cz) Ani prDzvodǁí, ani cestující nepǥekvapovala sama podstata vLJci: že kocour leze do tramvaje, na tom by ještLJ nebylo nic divného, ale že chce platit! (38Sb)

Ѣѹ ҔҸӁҦҏҔҲѦҝҔҏ Ӂѹ ҰҏҲӁѹҔѦ ӁѹҎѦ ӉҽҍҏҦѹґҽ ҡҏҗҲѹӁҽ ҡҲӊҽҝѹ: ҲҸ җҲҸ ѸҽҍҽҔ ѽҸѴѦ Ҧҽ ҏ҉Ѧ ҏ ҲҝҽѸӊҽҎ, Ӂѹ ҰҸ’ ҎҽҦҽ: Ѭѹѽ ҎѦ ҍҏҦѹґҸ җҲҸ ѸҽҍҽҔ ѽҸѴѦ Ҧҽ ҰґҽҲѹ ҔҽҝҲҏ.

376

CORNELIA E. KEIJSPER

(38Hr) Ni kondukterku ni putnike nije zaǁudila sama bit stvari: ne to što maǁak ulazi u tramvaj, što je tek pola zla, nego to što pokušava platiti! (38Bg) ѢѹҲҸ ҔҸӁҦҏҔҲҸҝҔҽҲҽ, ӁѹҲҸ ҰӆҲӁѹ҃ѹҲѦ қҾѽҽ ѹӉӁѦӁҽҦҽӁѹ ҸҲ ҡҽѸҽҲҽ ҡӆӂӁҸҡҲ Ӂҽ ӍҽҔҲҽ: ҡҾҔҽҗ ҸқҡҲҸҾҲѦґҡҲӊҸҲҸ, ҍѦ ӁҾҔҽҔӆӊ ҔҸҲҽҝҽҔ ѹҡҔҽ Ҧҽ ҡѦ ӊѸӆҔӁѦ ӊ ҲҝҽѸӊҽҾ, ӁѦ қѦҗѦ ӁҽҮ-ҡѸҽҮӊҽӂҸҲҸ, ҽѸѹ ҲҸӊҽ, ҍѦ ӁҽҡҲҸҾӊҽ Ҧҽ ҡѹ ҰґҽҲѹ! (38BR) ӎ ҔҽӁҦҏҔҲҽҝҔҏ, ҟ ҰҽҡҽңӃҝҽҕ ҏҝҽӉҟґҽ ҲҸѦ, җҲҸ ҔҸҲ ґѦӉѦ ҕ ҲҝҽѸӊҽҮ, ҽ ӁѦ ҲҸѦ, җҲҸ ҺӁ ӉқҟҝҽѦ҃҃ҽ ӉҽҰґҽ҃ҟ҃ҧ! 2. A phenomenon that can for the time being be called “syntactic parallism” is illustrated by repetitions like those in the following Russian example: (39R)

Ѣҽ ѸѦҡҲѦ ҲҸҊҸ, ҔҲҸ ӊ ҦҝҽӁҸҮ ҃ѹҝҔҸӊҸҮ ҸҦѦңҦѦ ҰҸҔѹӁҏґ ғҸҝҸқҧѦӊӃ ҊҸҝӃ ҰҸҦ ѹѸѦӁѦѸ ҘҸҝҸӊҧѦӊҽ-ҨҽҊҸҲҽ, ҲѦҰѦҝҧ ҡҔҽҔҽґ, ҲѹѽҸ ӉӊѦӁҾ ӉҸґҸҲҸӀ ҃ѦҰҧӀ ҰҸӊҸҦҽ, ҲѦѸӁҸ-ӍѹҸґѦҲҸӊӃҮ ҝӃ҃ҽҝҧ ҡ ѸҝҽҍӁѦҮҗѹѸ ѹ ӁѹҔҸҊҦҽ ӁѦ ҏґӃқҽӀӂѹѸҡҾ ґѹ҃ҸѸ. ҫӁ ҏҰѦҝҡҾ ҰҸҦқҸҝҸҦҔҸѸ ӊ ҊҝҏҦҧ, ҸӁ ӁѦ ҊґҾҦѦґ Ӂҽ ґҏӁҏ, ҸӁ ӁѦ ѹӁҲѦҝѦҡҸӊҽґҡҾ ӉѦѸґѦӀ ҰҸҦ ҡҸқҸӀ, ҸӁ ҦҏѸҽґ Ҹ ҍѦѸ-ҲҸ ҡӊҸѦѸ, ґѦҲҾ ҝҾҦҸѸ ҡ ғҸґҽӁҦҸѸ. (ҷҏґҊҽҔҸӊ, ҴҽҡҲѦҝ ѹ ҴҽҝҊҽҝѹҲҽ)

(39P)

Wspiera’ brodNj na piersi, nie patrzy’ na ksiNjǻyc, nie interesowa’a go ziemia, pNjdzƾc obok Wolanda rozmyǧla’ o jakowychǧ swoich sprawach.

(39Cz) Nedíval se ani na mLJsíc, ani ho nezajímala zemLJ. LetLJl po WolandovLJ boku, bradu zaboǥenou na prsa, zamLJstnán svými myšlenkami. (39Sb)

ҋҰҏҡҲѹҸ ҎѦ қҝҽҦҏ Ӂҽ ҊҝҏҦѹ, ѸѦҡѦ҃ ӁѹҎѦ ҊґѦҦҽҸ ӁѹҲѹ ҡѦ ѹӁҲѦҝѦҡҸӊҽҸ Ӊҽ ӉѦѸҒҏ ҰҸҦ ҡҸқҸѸ, ѸѹҡґѹҸ ҎѦ Ӂҽ ӁѦҗҲҸ ҡӊҸҎѦ, ґѦҲѦѴѹ ҰҸҝѦҦ ғҸґҽӁҦҽ.

(39Hr) Podbradak je upro u prsa, nije gledao mjesec, nije ga zanimala zemlja, zadubio se u svoje misli, leteǀi uz Wolanda. (39Bg) ҫҰҝҾґ қҝҽҦѹҍҔҽ ӊ ҊӆҝҦѹҲѦ ҡѹ, ҲҸҮ ӁѦ ҰҸҊґѦңҦҽҗѦ ґҏӁҽҲҽ, ӁѦ ҊҸ ѹӁҲѦҝѦҡҏӊҽҗѦ ѹ ӉѦѸҾҲҽ ҰҸҦ ӁѦҊҸ, ҲҸҮ ѸѹҡґѦҗѦ Ӊҽ ӁѦӂҸ ҡӊҸѦ ѹ ґѦҲѦҗѦ ҝѦҦҸѸ ҡ ғҸґҽӁ. As the translations show, the Russian repetition sustained by the reflexive verb ѹӁҲѦҝѦҡҸӊҽҲҧҡҾ does not prevent the choice of OVS orders in other languages. The tendency towards syntactic parallelism may be less influential in languages without subject pronouns (as suggested by (39) and comparable examples), but other issues prevent drawing this conclusion already: the Russian inside vs. outside perspective discussed in Keijsper (2003: 200-203) seems less evident in

FROM OVS ORDER TO CONVERSE STRUCTURE IN RUSSIAN

377

other Slavic languages, and lexical issues (see point 3 below) sometimes make the comparison between languages difficult. Other interrelated choices within a language must also first be unravelled: if, for example, a gerund is chosen, which is more often the case in Russian than in other Slavic languages, the subject in the main clause cannot be chosen independently, of course, which may lead to the choice of another converse than in the original: (40R)

ҘҸҊҦҽ ҰҸҝҏҍѹҔ ӊҡҲҽґ, ҹӊѦҮҔ, ҰҸҦҽӊҽҾ ѦѸҏ ӉҽӊҲҝҽҔ, ҰҸҝҽӉѹґ ѦҊҸ ӁҸӊӃѸ ӊҸҰҝҸҡҸѸ: – ҫҡѸѦґӀҡҧ ҡҰҝҸҡѹҲҧ, ҊҸҡҰҸҦѹӁ ҸқѦҝ-ґѦҮҲѦӁҽӁҲ, ӁѦ ҰҝѹҔҽңѦҲѦ ґѹ ҰҸҦӃҡҔҽҲҧ ӊҽѸ ҦҝҏҊҏӀ ҡҸқҽҍҔҏ?

(40Cz) Když pak byl již vzhDzru a Švejk mu pǥinesl snídani, byl nadporuǁík pǥekvapen novou otázkou Švejka: (Hašek, Švejk) But when the wish to use a gerund overrules the free choice from a set of converse structures is not immediately clear. 3. The availability of converses is of course an important factor influencing the choice of one of them. Of the verbs mentioned in note 1, ҰҸ/ҸқҝҽҦҸӊҽҲҧ, ѹҡҰҏҊҽҲҧ and ҸҲґѹҍѹҲҧ have no occurrences of the -n/-t participle (short form) in Slavic originals. In the same sources, no reflexive finite verb forms are found for ҸҔҝҏңҽҲҧ/ҸҔҝҏңѹҲҧ, ҸѽӊҽҲӃӊҽҲҧ/ҸѽӊҽҲѹҲҧ, ҰҸҡѦӂҽҲҧ/ҰҸҡѦҲѹҲҧ, ҏҔҝҽҗҽҲҧ/ҏҔҝҽҡѹҲҧ. Neither short forms of the -n/-t participle nor reflexive finite verb forms appear for қҏҦѹҲҧ/ҝҽӉқҏҦѹҲҧ, ҸӊґҽҦѦӊҽҲҧ/ҸӊґҽҦѦҲҧ, ҸҦҸґѦӊҽҲҧ/ҸҦҸґѦҲҧ. Of course, in some cases the accidental prefix in the verbs selected is responsible for this outcome. Thus, if instead of ҝҽӉқҏҦѹҲҧ we take ҰҝҸқҏҦѹҲҧ we do find reflexive verb forms, of course. For prefixes see also point 9 below. The availability of lexical alternatives has syntactic consequences in cases like (41) and (42), where the Russian difference between reactivity (reflexive verb) and passive involvement of an object (OVS) is a translation of lexically different verbs (see also (39)): (41R)

– ҶҲҸ ҰҝҽӊҦҽ, Ҿ ӁѹҔҸҊҦҽ ӁѦ ѹӁҲѦҝѦҡҸӊҽґҡҾ ӉӊѦӉҦҽѸѹ, – ҲѹѽҸ ҡҔҽӉҽґ ѨҊӁҽ҃ѹҮ. – ҋ ҲҸҊҸ ҦӁҾ Ӂҽҍҽґҽҡҧ ҏ ѸѦӁҾ ҡҲҝҽӁӁҽҾ қҸґѦӉӁҧ, – ҰҝҸҦҸґңҽґ ғҸҔҏґҧҡҔѹҮ.

(41P)

– Prawda, ǻe nigdy nie interesowa’y mnie gwiazdy – szepnƾ’ pan Ignacy. (Prus, Lalka)

(41Sb)

– ѨҡҲѹӁҽ ҎѦ, ӁѹҔҽҦҽ ѸѦ Ӂѹҡҏ ѹӁҲѦҝѦҡҸӊҽґѦ ӉӊѦӉҦѦ, – җҽҰӁҏ ҊҸҡҰҸҦѹӁ ѨҊӁҽ҃ѹ.

378

CORNELIA E. KEIJSPER

(42R)

ѯѦҝѦӉ Ҳҝѹ ҦӁҾ ҰҸҡґѦ ҡҲҝҽӁӁҸҊҸ ҰҸѦҦѹӁҔҽ ғҸҔҏґҧҡҔѹҮ ҡѹҦѦґ, ӉҽҰѦҝҗѹҡҧ ӊ ҔҽқѹӁѦҲѦ, ҡ ӁѦҔѹѸ ҰҽӁҸѸ ғѹґҧҾѸҸѸ ҘҸґґѹӁӉҸѸ. ҋґҏҊҽ, ҔҸҲҸҝҸҊҸ ҦҽӊӁҸ ҏңѦ ѹӁҲѦҝѦҡҸӊҽґѹ ҙҲѹ ҔҸӁӍѦҝѦӁ҃ѹѹ, ҰҝҸѹҡѽҸҦѹӊҗѹѦ ӁѦҡҔҸґҧҔҸ ҝҽӉ ӊ ӁѦҦѦґӀ, ӊӃҲѹҝҽҾ ҰӃґҧ ӊ ҔҸѸӁҽҲѦ ҝҾҦҸѸ, ӊҝѦѸҾ ҸҲ ӊҝѦѸѦӁѹ ҰҝѹҔґҽҦӃӊҽґ Ҕ ӉҽѸҸҍӁҸҮ ҡҔӊҽңѹӁѦ ҲҸ ҏѽҸ, ҲҸ ҊґҽӉ.

(42P)

W trzy dni po dziwnym pojedynku siedzia’ Wokulski zamkniNjty w gabinecie z niejakim panem Wiliamem Colins. S’uǻƾcy, którego od dawna intrygowa’y te konferencje odbywajƾce siNj po kilka razy na tydzieǝ, ǧciera’ kurze w pokoju obocznym i od czasu do czasu przysuwa’ bƾdǺ oko, bƾdǺ ucho do dziurki od klucza. (Prus, Lalka)

(42Sb) ѷҝѹ ҦҽӁҽ ҰҸҡґѦ ҍҏҦӁҸҊ ҦӊҸқҸҎҽ ҡѦҦѦҸ ҎѦ ғҸҔҏґҡҔѹ ӉҽҲӊҸҝѦӁ ҏ ҡӊҸѸ ҔҽқѹӁѦҲҏ ҡ ӁѦҔҽҔӊѹѸ ҊҸҡҰҸҦѹӁҸѸ ғѹҒѦѸҸѸ ҘҸґѹӁҡҸѸ. ҋґҏҊҽ, ҔҸҎѦҊ ҡҏ ӊѦѴ ҸҦҽӊӁҸ ӉҽӁѹѸҽґѦ ҲѦ ҔҸӁӍѦҝѦӁ҃ѹҎѦ, җҲҸ ҡҏ ҡѦ ҸҦҝңҽӊҽґѦ ӁѦҔҸґѹҔҸ ҰҏҲҽ ӁѦҦѦҒӁҸ, қҝѹҡҽҸ ҎѦ ҰҝҽҗѹӁҏ ҏ ҡҏҡѦҦӁҸҎ ҡҸқѹ ѹ, ҡ ӊҝѦѸѦӁҽ Ӂҽ ӊҝѦѸѦ, ӁҽҡґҽѬҽҸ ѹґѹ ҸҔҸ ѹґѹ ҏѽҸ Ӂҽ ҝҏҰҏ ҔҒҏҍҽҸӁѹ҃Ѧ. The availability of verbal and nominal stems is relevant in many cases (see (26) and (27) for examples). The question whether languages prefer one or the other in a systematic way, and whether the preference conforms to ҴҝҽӉѦҔ’s 1990 division of verbal and nominal tendencies in Slavic languages is a study in itself. 4. If converses are available, there can still be lexical factors preventing their use. For example, Polish pokrywaǀ siNj / pokryǀ siNj has among its meanings ‘to coincide with’ (reciprocal reflexive: ‘to cover each other’), which is probably one of the causes (see also point 9 below) responsible for the fact that the Polish verb pokrywaǀ/pokryǀ shows up in OVS orders in cases where Russian has the reflexive verb ҰҸҔҝӃӊҽҲҧҡҾ/ҰҸҔҝӃҲҧҡҾ. Moreover, the use of OVS can be different in different applications of a verb. For example, Russian ҰҸҔҝӃӊҽҲҧ/ҰҸҔҝӃҲҧ is normal in OVS in application to sounds that become inaudible because of other sounds (see example (1R)), in application to debts that are paid, in application to distances that are covered by moving, and in other cases, but in the meaning ‘covering a surface so that the thing covered becomes invisible’ we often see a Russian reflexive verb or a short form of the -n/-t participle where Polish has OVS: (43R)

ү ҨҽҊҸҲ, ҡҰҝҸӊҽҦѹӊ ҰҸҡҲҝҽҦҽӊҗѦҊҸ ҔҸӁӍѦҝҽӁҡҧѦ, ҸқӆҾӊѹґ ҰҏқґѹҔѦ ҲҽҔ: – ѷҽҰѦҝѹҍҽ, ҔҸҊҦҽ ҙҲҸҊҸ ӁҽҦҸѦҦҽґҏ ҡҰґҽӊѹґѹ, ҦҽӊҽҮҲѦ ҸҲҔҝҸѦѸ ҦҽѸҡҔѹҮ ѸҽҊҽӉѹӁ! Ѩ ҲҸҲҍҽҡ ҰҸґ ҡ҃ѦӁӃ ҰҸҔҝӃґҡҾ ҰѦҝҡѹҦҡҔѹѸѹ ҔҸӊҝҽѸѹ, ӊҸӉӁѹҔґѹ ҊҝҸѸҽҦӁӃѦ ӉѦҝҔҽґҽ, ҡ қҸҔҸӊ ҸҡӊѦӂѦӁӁӃѦ ӉѦґѦӁҸӊҽҲӃѸѹ ҲҝҏқҔҽѸѹ, ҽ ѸѦң ӉѦҝҔҽґ ӊѹҲҝѹӁӃ, ѹ ӊ Ӂѹѽ

FROM OVS ORDER TO CONVERSE STRUCTURE IN RUSSIAN

379

ӉҝѹҲѦґѹ ӊ ӊѦҡѦґҸѸ ҸҗѦґҸѸґѦӁѹѹ ҏӊѹҦѦґѹ ҝҽӉӁӃѽ ҃ӊѦҲҸӊ ѹ ӍҽҡҸӁҸӊ ҰҽҝѹңҡҔѹѦ ңѦӁҡҔѹѦ ҰґҽҲҧҾ. (ҷҏґҊҽҔҸӊ, ҴҽҡҲѦҝ ѹ ҴҽҝҊҽҝѹҲҽ) (43P)

I natychmiast deski sceny pokry’y perskie dywany, pojawi’y siNj olbrzymie lustra oǧwietlone po bokach zielonkawymi rurkami, miNjdzy lustrami zaǧ witryny, w których mile oszo’omieni widzowie zobaczyli paryskie sukienki najróǻniejszych kolorów i fasonów.

(43Cz) Polovina scény se okamžitLJ pokryla perskými koberci, kde se vzala tu se vzala obrovská zrcadla, osvLJtlená ze stran nazelenalými trubiǁkami, mezi nimi se objevily vitríny a v nich radostnLJ ohromení diváci zahlédli paǥížské toalety nejrDzznLJjších barev a stǥihDz. (43Sb)

Ѩ ѹҡҲҸҊ ҍҽҡҽ ҰҸҦ ҡ҃ѦӁѦ қѹ ҰҝѦҔҝѹӊѦӁ ҰѦҝҡѹҎҡҔѹѸ ѴѹґѹѸѹѸҽ, ҰҸҎҽӊѹҗѦ ҡѦ ҸҊҝҸѸӁҽ ҸҊґѦҦҽґҽ, ҸҡӊѦҲҒѦӁҽ ҡҽ ҡҲҝҽӁѦ ӉѦґѦӁҔҽҡҲѹѸ ҃ѦӊѹѸҽ ҽ ѹӉѸѦ҉ҏ ҸҊґѦҦҽґҽ ѹӉґҸӉѹ, ҽ ҏ ѬѹѸҽ ӉҽҰҽѬѦӁѹ ҊґѦҦҽҸ҃ѹ ҏҊґѦҦҽҗѦ ҰҽҝѹҡҔѦ ҦҽѸҡҔѦ ҲҸҽґѦҲѦ ӁҽҎҝҽӉґѹҍѹҲѹҎѹѽ ҔҝҸҎѦӊҽ ѹ ӍҽӉҸӁҽ.

(43Hr) I odmah se pod pozornice prekrio perzijskim sagovima, pojavila su se ogromna zrcala osvijetljena sa strane zelenkastim cijevima, a izmeDžu zrcala – vitrine u kojima su gledaoci, ošamuǀeni od veselja, ugledali pariške ženske haljine raznih boja i krojeva. (43Bg) ѷҏҲҽҔҡѹ ҰҸҦӆҲ Ӂҽ ҡ҃ѦӁҽҲҽ ҡѦ ҸҔҽӉҽ ҰҸҡҲґҽӁ ҡ ҰѦҝҡѹҮҡҔѹ ҔѹґѹѸѹ, ѹӉӁѹҔӁҽѽҽ ҸҊҝҸѸӁѹ ҸҊґѦҦҽґҽ, ҸҡӊѦҲѦӁѹ ҸҲҡҲҝҽӁѹ ҡӆҡ ӉѦґѦӁѹҔҽӊѹ ӁѦҸӁҸӊѹ Ҳҝӆқѹ, ҽ ѸѦңҦҏ ҸҊґѦҦҽґҽҲҽ ӊѹҲҝѹӁѹ ѹ ӊ ҲҾѽ ӊѦҡѦґҸ ӉҽҗѦѸѦҲѦӁѹҲѦ ӉҝѹҲѦґѹ ӊѹҦҾѽҽ ҰҽҝѹңҔѹ ҦҽѸҡҔѹ ҝҸҔґѹ, ӁҽҮ-ҝҽӉґѹҍӁѹ ҃ӊѦҲҸӊѦ ѹ ѸҸҦѦґѹ. (43BR) ӎ ҽҦҝҽӉҏ ҰҽҦґҸҊҽ ҡ҃ҙӁӃ ҰҽҔҝӃґҽҡҾ ҰѦҝҡҟҦҡҔҟѸҟ ҦӃӊҽӁҽѸҟ, ҏӉӁҟҔґҟ ӊҾґҟӉӁӃҾ ґӀҡҲҝӃ, Ӊ қҽҔҸҕ ҽҡӊѦҲґѦӁӃҾ ӉѦґѦӁҽӊҽҲӃѸҟ ҲҝҏқҔҽѸҟ, ҽ ҰҽѸҟң ґӀҡҲҝҽѸҟ ӊҟҲҝӃӁӃ, ҏ ҟѽ ҊґѦҦҽҍӃ ҕқҽҍӃґҟ Ӊ ӊҾҡҺґӃѸ ҽҗҽґҽѸґѦӁӁѦѸ ҝҸӉӁҽҊҽ ҔҸґѦҝҏ ҟ ӍҽҡҸӁҏ ңҽӁҸҍӃҾ ҡҏҔѦӁҔҟ. (44R)

һҸӂѹӁҽ қӃґҽ ҰѦҝѦҦ ӁѹѸѹ, ѹ үҝҲҏҝ ҏңѦ ҡҲҸҾґ, ҝҽҡҲѦҝҾӁӁҸ ҊґҾҦҾ Ӂҽ ҩҙҦҝѹҔҽ. ѩӁҸ ґҸӂѹӁӃ қӃґҸ ҰҸҔҝӃҲҸ ҊӁҸҮӁҸ-ӉѦґѦӁҸҮ, ңѹҝӁҸ ҸҲҡӊѦҍѹӊҽӀӂѦҮ Ӂҽ ҡҸґӁ҃Ѧ ңѹңѦҮ. ѢҽҦ ҰҸӊѦҝѽӁҸҡҲҧӀ ѦѦ ҔҏҝѹґҡҾ ґѦҊҔѹҮ ҰҽҝҸҔ, ѸѦңҦҏ ѽҸґѸҽѸѹ ҸӁ ҡҲҽӁҸӊѹґҡҾ ҊҏӂѦ, ѹ ӊ ҲҝѹҦ҃ҽҲѹ җҽҊҽѽ ҏңѦ ӁѹҍѦҊҸ ӁѦ қӃґҸ ӊѹҦӁҸ. Ѩ ҡѸҝҽҦ. (ҋҲҝҏҊҽ҃ҔѹѦ, ҭѹҔӁѹҔ Ӂҽ ҸқҸҍѹӁѦ)

(44P)

Kotlinka by’a tuǻ przed nimi i Artur juǻ sta’, niepewnie patrzƾc na Reda. Dno kotliny pokrywa’a t’usto po’yskujƾca na s’oǝcu ǻó’to - zielona maǺ.

380

CORNELIA E. KEIJSPER

(44Bg) ѩѦҝѦҲҸ қѦҗѦ ҰҝѦҦ ҲҾѽ ѹ үҝҲӆҝ ӊѦҍѦ ҡѦ қѦҗѦ ҡҰҝҾґ ѹ қѦӉҰҸѸҸӂӁҸ ҊґѦҦҽҗѦ ҩѦҦҝѹҔ. ѩӆӁҸҲҸ Ӂҽ ҦѦҝѦҲҸ қѦҗѦ ҰҸҔҝѹҲҸ ҡ ҊӁҸҮӁҸӉѦґѦӁҽ, ѸҽӉӁҸ ҰҝҸқґҾҡӊҽӂҽ Ӂҽ ҡґӆӁ҃ѦҲҸ ҊӆҡҲҽ ҰҸѸѹҾ, ӁҽҦ ӁѦҮӁҽҲҽ ҰҸӊӆҝѽӁҸҡҲ ҡѦ ӊѹѦѽҽ ґѦҔѹ ѹӉҰҽҝѦӁѹҾ, ѸѦңҦҏ ѽӆґѸҸӊѦҲѦ ҲѦ ҡҲҽӊҽѽҽ ҰҸ-ҊӆҡҲѹ ѹ Ӂҽ ҲҝѹҮҡѦҲѹӁҽ ҔҝҽҍҔѹ ӊѦҍѦ ӁѹӂҸ ӁѦ ҡѦ ӊѹңҦҽҗѦ. Ѩ ӊҸӁѦҗѦ. The cause of this Russian distribution may be that mentioned in point 9 below. In translations from Polish the difference between the two languages seems to be illustrated by a case like the following, where Russian translates Polish OVS by means of OVS, but uses another verb or rather explains the meaning of the Polish text: (45R)

ғҡҔҸҝѦ ґѹҗҧ ҔҸѦ-ҊҦѦ ѦӂѦ қӃґѹ ӊѹҦӁӃ ҸҦѹӁҸҔѹѦ ҡҲҸҾӂѹѦ Ӂҽ ҔҸґѦӁҾѽ ӍѹҊҏҝӃ, ӁҸ ѹѽ қӃҡҲҝҸ ӉҽҡґҸӁҾґѹ ҸҲ ҊґҽӉ ҦӊѹңҏӂѹѦҡҾ, ӊҸӀӂѹѦ ҡӊҸҝӃ.

(45P)

Zapach krwi i poszarpanych wnNjtrznoǧci zg’uszy’ arabskie wonie i nape’ni’ ca’y cyrk. W koǝcu juǻ tylko gdzieniegdzie widaǀ by’o pojedyncze klNjczƾce postacie, które wnet pokrywa’y ruchome wyjƾce kupy. (Sienkiewicz, Quo vadis)

However, there are also cases where Russian translations take over the Polish use of OVS, so we must reckon with the possibility of Polonisms and other cases of interference (see further point 9 below). 5. As will be remembered, the so-called “ҊґҽҊҸґӃ ҰҸґӁҸҊҸ ҸѽӊҽҲҽ” have originally been grouped together on the basis of the fact that they exhibit regular polysemy of the following type (ҭҽҦҏҍѦӊҽ ѹ ҩҸӉѹӁҽ 1993: 8): (46R) (47R) (48R)

ҋҲҸҝҸң ӁҽҰҸґӁҾѦҲ қҽҡҡѦҮӁ ӊҸҦҸҮ (ҦѦҮҡҲӊѹѦ) ғҸҦҽ ҰҸҡҲѦҰѦӁӁҸ ӁҽҰҸґӁҾѦҲ қҽҡҡѦҮӁ (ҰҝҸ҃Ѧҡҡ) ғҸҦҽ ӁҽҰҸґӁҾѦҲ қҽҡҡѦҮӁ ҦҸ ҔҝҽѦӊ (ҡҸҡҲҸҾӁѹѦ)

The verbs exhibiting this polysemy are not a homogeneous group. Thus, of the verb selected in note 1, ӁҽҰҸґӁҾҲҧ/ӁҽҰҸґӁѹҲҧ and ҰҸҔҝӃӊҽҲҧ/ҰҸҔҝӃҲҧ have the fullest pattern: type (47R) (process) is possible in both aspects. According to ҭҽҦҏҍѦӊҽ ѹ ҩҸӉѹӁҽ (1993), ҸҔҝҏңҽҲҧ/ҸҔҝҏңѹҲҧ, ҡҸѦҦѹӁҾҲҧ/ҡҸѦҦѹӁѹҲҧ, ӊѦӁҍҽҲҧ/ҏӊѦӁҍҽҲҧ, and ҏҔҝҽҗҽҲҧ/ҏҔҝҽҡѹҲҧ in one application (ҏҔҝҽҗҽҲҧ ѦґҔҏ ѹҊҝҏҗҔҽѸѹ), do not have the process interpretation in the imperfective aspect, and ӉҽӊѦҝҗҽҲҧ/ӉҽӊѦҝҗѹҲҧ and ҏҔҝҽҗҽҲҧ/ҏҔҝҽҡѹҲҧ in another application (ҏҔҝҽҗҽҲҧ қҽӁҲҸѸ) have neither the process interpretation nor the action interpretation (cf. 46R) in the imperfective aspect. If we take into account OVS orders, reflexive verbs and short forms of -n/-t participles, the diversity is still

FROM OVS ORDER TO CONVERSE STRUCTURE IN RUSSIAN

381

larger, and the possibilities differ in different languages. For the study of converse structure it does not seem useful to treat the “ҊґҽҊҸґӃ ҰҸґӁҸҊҸ ҸѽӊҽҲҽ” as a group rather than as individual verbs whose uses are of great interest but require more attention than is possible here. 6. The -n/-t participle has a number of aspects that require further attention. First, as will be remembered (ҘӁҾӉѦӊ 1989: 191), Polish has a formal difference between -n/-t participles with byǀ (stative/resultative) and zostawaǀ/zostaǀ (actional, mostly perfective). In Russian, ҘӁҾӉѦӊ 1989 recognizes stative/resultative and actional interpretations of the -n/-t participle (without direct formal differences), and his investigations show that there are many cases where stative/resultative is not comparable to imperfective (as appears when the -n/-t participle is replaced by the l-participle), and many cases where actional is not the same as perfective; Keijsper (2003: 173-176) contains a list of ҘӁҾӉѦӊ’s crucial cases. One case is verbs of emotion: the -n/-t participles are of the type which Knjazev calls resultative, but they have the property that they occur in contexts where l-participles would be perfective. Other Slavic languages, or at least some of them, choose other forms there: (49R)

ҭҸӊѦҦѦӁѹѦ ҔҸҲҽ ӁҽҡҲҸґҧҔҸ ҰҸҝҽӉѹґҸ ѨӊҽӁҽ, ҍҲҸ ҸӁ ӊ ӁѦҰҸҦӊѹңӁҸҡҲѹ ӉҽҡҲӃґ ҏ қҽҔҽґѦҮӁҸҊҸ ѸҽҊҽӉѹӁҽ Ӂҽ ҏҊґҏ ѹ ҲҏҲ ӊҲҸҝѹҍӁҸ, ӁҸ ҊҸҝҽӉҦҸ ҡѹґҧӁѦѦ, қӃґ ҰҸҝҽңѦӁ ҰҸӊѦҦѦӁѹѦѸ ҔҸӁҦҏҔҲҸҝҗѹ. ѷҽ, ґѹҗҧ ҲҸґҧҔҸ ҏӊѹҦѦґҽ ҔҸҲҽ, ґѦӉҏӂѦҊҸ ӊ ҲҝҽѸӊҽҮ, ҡҸ ӉґҸқҸҮ, ҸҲ ҔҸҲҸҝҸҮ ҦҽңѦ ҲҝҾҡґҽҡҧ, ӉҽҔҝѹҍҽґҽ: - ҘҸҲҽѸ ӁѦґҧӉҾ! ҋ ҔҸҲҽѸѹ ӁѦґҧӉҾ! ҷҝӃҡҧ! ҋґѦӉҽҮ, ҽ ҲҸ Ѹѹґѹ҃ѹӀ ҰҸӉҸӊҏ! (ҷҏґҊҽҔҸӊ, ҴҽҡҲѦҝ ѹ ҴҽҝҊҽҝѹҲҽ)

(49P)

Zachowanie siNj kota wstrzƾsnNj’o Iwanem do tego stopnia, ǻe zastyg’ nieruchomo obok sklepu kolonialnego na rogu i wtedy zdumia’ siNj po raz drugi, i to znacznie silniej, tym razem za przyczynƾ konduktorki. Ta, skoro tylko zobaczy’a w’aǻƾcego do tramwaju kota, wrzasnNj’a dygocƾc z wǧciek’oǧci:

(49Cz) Kocourovo chování básníka natolik vyvedlo z míry, že zDzstal strnule stát na rohu pǥed obchodem s lahDzdkami. Daleko víc ho však udivilo, jak reagovala prDzvodǁí. Sotva uvidLJla kocoura, jak leze do tramvaje, vykǥikla zlostnLJ: (49Sb) ҭҸӁҽҗҽѬѦ ѸҽҍҔҽ ҦҸ ҲѦ ѸѦҝѦ ӉҽҰҝѦҰҽҡҲѹ ѨӊҽӁҽ, Ҧҽ ҎѦ ҔҽҸ ҡҔҽѸѦѬѦӁ ҡҲҽҎҽҸ ҔҝҽҎ қҽҔҽґӁѹ҃Ѧ Ӂҽ ҏҊґҏ, ҽ ҎҸҗ ӊѹҗѦ Ҋҽ ӉҽҰҝѦҰҽҡҲѹ ҔҸӁҦҏҔҲѦҝҔѹӁҸ ҰҸӁҽҗҽѬѦ. ҫӁҽ, ҍѹѸ ӊѹҦѦ ѸҽҍҔҽ ҔҽҔҸ ҰҸҔҏҗҽӊҽ Ҧҽ ҡѦ ҏӊҏҍѦ ҏ ҲҝҽѸӊҽҎ, ҒҏҲѹҲҸ ҰҸӊѹҔҽ: (49Bg) ҭҸӊѦҦѦӁѹѦҲҸ Ӂҽ ҔҸҲҽҝҽҔҽ ҲҽҔҽ ҡѸҽҾ ѨӊҽӁ, ҍѦ ҲҸҮ ӉҽѸҝӆӉӁҽ ӁѦҰҸҦӊѹңѦӁ ҰҝѦҦ қҽҔҽґӁѹ҃ҽҲҽ Ӂҽ ӆҊӆґҽ, ѹ ҡѦҊҽ ӊҲҸҝѹ ҰӆҲ, ӁҸ ѸӁҸҊҸ ҰҸҡѹґӁҸ ҡѦ ҡѸҽҾ, ӊѦҍѦ ҸҲ ҰҸӊѦҦѦӁѹѦҲҸ Ӂҽ ҔҸӁҦҏҔҲҸҝҔҽҲҽ. ӅҸѸ ӊѹҦҾ

382

CORNELIA E. KEIJSPER

ҔҸҲҽҝҽҔҽ, ҔҸҮҲҸ ҡѦ ҸҰѹҲӊҽҗѦ Ҧҽ ҡѦ ӊѸӆҔӁѦ ӊ ҲҝҽѸӊҽҾ, ҲҾ ҃ҾґҽҲҽ ҡѦ ҝҽӉҲҝѦҰѦҝѹ ҸҲ ӉґҸқҽ ѹ ҡѦ ҝҽӉӊѹҔҽ: In cases where the Russian -n/-t participle (also resultative) could be replaced by an imperfective l-participle, we see the -n/-t participle elsewhere as well (in Polish with byǀ): (50R)

ғҸҲ ҦҸ ҍѦҊҸ ҙҲѹ ҲҝҽѸӊҽѹ ҦҸӊҸҦҾҲ! – ѹ, қҏҦҏҍѹ, ӊѹҦѹѸҸ, ӁѦ ӊ ҡѹґҽѽ ҡҦѦҝңҽҲҧ ҡѦқҾ, ҘҸҝҸӊҧѦӊ ҔґӀӁҏґ ӁҸҡҸѸ ӊ ҡҲѦӁҏ ҝҾҦҸѸ ҡ ӉѦҝҔҽґҸѸ ѹ ҡҲҽґ ҡҸҦҝҸҊҽҲҧҡҾ ӊ ҝӃҦҽӁѹҾѽ. ѩҾҦҾ ҷѦҝґѹҸӉҽ қӃґ ѹҡҔҝѦӁӁѦ ҰҸҝҽңѦӁ ҰҸӊѦҦѦӁѹѦѸ ӁѦѹӉӊѦҡҲӁҸҊҸ. (ҷҏґҊҽҔҸӊ, ҴҽҡҲѦҝ ѹ ҴҽҝҊҽҝѹҲҽ)

(50P)

Wujaszek Berlioza by’ szczerze wzruszony zachowaniem nieznajomego.

(50Cz) BerliozDzv strýc byl dojat jeho upǥímnou lítostí. (50Sb) ҷѦҝґѹҸӉҸӊ ҲѦҍҽ ҎѦ қѹҸ ѹҡҔҝѦӁҸ Ӊҽҍҏ҉ѦӁ ҰҸӁҽҗҽѬѦѸ ӁѦӉӁҽӁ҃ҽ. (50Bg) ѯѹҍҸҲҸ Ӂҽ ҷѦҝґѹҸӉ қѦҗѦ ѹҡҔҝѦӁҸ ҡѸҽҾӁ ҸҲ ҰҸӊѦҦѦӁѹѦҲҸ Ӂҽ ӁѦҰҸӉӁҽҲѹҾ. Unfortunately, the emotion verbs mentioned in note 1 do not provide enough examples to resolve this matter here. The taxis/time relations of Russian -n/-t participles (Keijsper 2003: 177-180) are also worth attention. For example, the verb ҸѽӊҽҲѹҲҧ in most cases has OVS order when reference is made to an emotion (S) arising within the boundaries defined by O (the person having the emotion); see example (11). However, one of the cases where we find a (rare) short form9 of the -n/-t participle of this verb is when the l-participle would probably have the interpretation that the emotion arose after the preceding event, while the -n/-t participle allows for the possibility that the emotion arose before the preceding event (before others looked at the person involved in (51)): (51R1)

9

ѢҽҡҲҏҰѹґҽ ҲѹҗѹӁҽ, ѹ ҊґҽӉҽ ӊҡѦѽ ҸқҝҽҲѹґѹҡҧ Ҕ ҨҝҸҦҸ. ҫӁ қӃґ ҸѽӊҽҍѦӁ ӊӁѦӉҽҰӁӃѸ ҡѸҏӂѦӁѹѦѸ ѹ ҡҲҝҽѽҸѸ, Ҕ ҲҸѸҏ ңѦ ѦѸҏ ҸҍѦӁҧ ӁѦ ѽҸҲѦґҸҡҧ ҦҸҡҲҽӊҽҲҧ ҘҸґҧ҃Ҹ.

Note that in this article I take into account only the short form of the -n/-t participle. I mentioned in Keijsper (2003: 189, 202-203) that the long form is another matter; this is confirmed by ASPAC. The use of ҸѽӊҽҍѦӁӁӃҮ, for example, combines uses of the verb observed in OVS order and in the short form of the -n/-t participle.

FROM OVS ORDER TO CONVERSE STRUCTURE IN RUSSIAN

383

(51E)

There was a hush, and all turned their eyes on Frodo. He was shaken by a sudden shame and fear; and he felt a great reluctance to reveal the Ring, and a loathing of its touch. (Tolkien, Ring)

(51R2)

ғ ӉҽґѦ ӊҸ҃ҽҝѹґҽҡҧ ѸѦҝҲӊҽҾ ҲѹҗѹӁҽ, ӊҡѦ ӁҽҰҝҾңѦӁӁҸ ҡѸҸҲҝѦґѹ Ӂҽ ҨҝҸҦҸ, ҽ ҸӁ, ҝҽҡҲѦҝҾӁӁӃҮ ѹ ҡҲҝҽӁӁҸ ѹҡҰҏҊҽӁӁӃҮ, ҍҏӊҡҲӊҸӊҽґ, ҍҲҸ ӁѦ ѽҸҍѦҲ ҰҸҔҽӉӃӊҽҲҧ ҘҸґҧ҃Ҹ, ѦѸҏ ҦҽңѦ ҦҸҲҝҸӁҏҲҧҡҾ ҦҸ ӁѦҊҸ қӃґҸ ҲҝҏҦӁҸ.

(51Sb)

Nastupi tajac, i svi upraviše pogled ka Frodu. On je bio potresen nekim iznenadnim stidom i strahom; osetio je veliku neodluǁnost da pokaže Prsten, i neko gaDženje od njegovog dodira.

(51Hr)

Nastane tajac i svi upru oǁi u Froda. On bijaše potresen iznenadnim stidom i strahom.

(51Bg)

ѢҽҡҲӆҰѹ ҲѹҗѹӁҽ ѹ ӊҡѹҍҔѹ ҰҸҊґѦҦѹ ҡѦ ҸқӆҝӁҽѽҽ ҔӆѸ ҨҝҸҦҸ. ғӁѦӉҽҰӁҸ ӉҽҡҝҽѸѦӁ ѹ қҸҾӉґѹӊ, ҲҸҮ қѦ ѹӉҰӆґӁѦӁ ҸҲ ѸҸҊӆӂҸ ӁѦңѦґҽӁѹѦ Ҧҽ ҝҽӉҔҝѹѦ ҭҝӆҡҲѦӁҽ ѹ ҦҸҝѹ Ҧҽ ҊҸ ҦҸҔҸҡӁѦ.

(51Mc) ѢҽҡҲҽҰѹ ѸҸґҔ ѹ ҡѹҲѦ Ҋѹ ӁҽҡҸҍѹҎҽ ҰҸҊґѦҦѹҲѦ ҔҸӁ ҨҝҸҦҸ. ѷҸҎ қѦҗѦ ҸқӉѦѸѦӁ ҸҦ ӁѦҔҽҔҸӊ ӁѦӁҽҦѦѦӁ ҡҝҽѸ ѹ ҡҲҝҽӊ; ҰҸҍҏӊҡҲӊҏӊҽ ҊҸґѦѸҽ ӁѦҝѦҗѹҲѦґӁҸҡҲ Ҧҽ ҊҸ ҰҸҔҽңѦ ҭҝҡҲѦӁҸҲ, ѹ ӁѦҔҽҔӊҸ ҊҽҦѦѬѦ ҸҦ ӁѦҊҸӊѹҸҲ ҦҸҰѹҝ. 7. A further issue regarding the -n/-t participle is the fact that the existence of an external agent can be implied even though he/she is not part of the referent (ҘӁҾӉѦӊ 1989: 127-131, Keijsper 2003: 176). This is a motivation for using the -n/-t participle in Russian, but Polish can still use OVS in such cases: (52R)

ҫӁ ҡѸҸҲҝѦґ Ӂҽ ѦѦ ґѹ҃Ҹ, ҰҸқґѦҦӁѦӊҗѦѦ ѹ ҸҡҏӁҏӊҗѦѦҡҾ, Ӂҽ ӉҽӊѹҲҔѹ ҲѦѸӁӃѽ ӊҸґҸҡ, Ӂҽ қѦҦӁҸѦ ҰґҽҲҧѦ, ҡѸҸҲҝѦґ ҲҽҔ ҏҰҸҝӁҸ, ҍҲҸ ҰҸҦ ѦҊҸ ӊӉҊґҾҦҸѸ қѦґҸҡӁѦңӁӃҮ ѦѦ ґҸқ Ӂҽҍҽґ ҡґѦҊҔҽ ҝҸӉҸӊѦҲҧ, – ѹ ҲҏҲ ҸӁ ҰҸҦҏѸҽґ, ҍҲҸ ґӀқѹҲ ѦѦ ҰҸ-ҰҝѦңӁѦѸҏ ѹ ҍҲҸ ӊ ҙҲҸҮ ӊҸҲ қґѦҦӁҸҡҲѹ ѦѦ ѹ ӁѹӂѦҲѦ ҰҸӊѹӁѦӁ ҸӁ, ҍҲҸ ҙҲҸ ҸӁ Ӊҽқҝҽґ ѦѦ ѹӉ ҦҸѸҽ, ҊҦѦ ѦѦ ґӀқѹґѹ, ҊҦѦ ҸӁҽ қӃґҽ ҸҔҝҏңѦӁҽ ҦҸҡҲҽҲҔҸѸ ѹ ҏҦҸқҡҲӊҽѸѹ, ҸӁ ӉҽҊӁҽґ ѦѦ ӊ ҙҲҏ ңҽґҔҏӀ ґҽҍҏҊҏ ѹ ҸҦѦґ ӊ қѦҦӁҾ҃ҔѹҮ Ұґҽӂ ѹӉ ҲѦѸӁҸҮ җѦҝҡҲѹ.

(52P)

Patrzy’ na jej twarz, bledszƾ i szczuplejszƾ niǻ dawniej, na skrNjty ciemnych w’osów; na ubogi ubiór robotnicy; patrzy’ tak uporczywie; ǻe aǻ pod wp’ywem jego wzroku jej ǧnieǻne czo’o poczNj’o róǻowieǀ – i naprzód pomyǧla’, ǻe jƾ kocha zawsze, a po wtóre, ǻe ta jej bladoǧǀ i to jej ubóstwo sƾ jego dzie’em, ǻe on to wypNjdzi’ jƾ z domu, gdzie jƾ kochano i gdzie jƾ otacza’ dostatek i wygody, a wtrƾci’ do tej mizernej izby i odzia’ w ten nNjdzny p’aszcz z ciemnej we’ny. (Sienkiewicz, Quo vadis)

384

CORNELIA E. KEIJSPER

(53R)

ҴҸґҸҦ҃Ӄ ѹӉ “ҷӃҡҲҝҸҏҰҽҔҽ”, ӉҽҔҏҰѹӊ ӊҡӀ ѸҏҔҏ ӊ ґҽқҽӉѦ, ҰѦҝѦҗґѹ Ӂҽ қҽҔҽґѦӀ ѹ ҸқҝҽӉҸӊҽґѹ ҍҽҮӁҸ-ҡҽѽҽҝӁҏӀ ҸҍѦҝѦҦҧ. ғ Ҳҝѹ ҦӁҾ ҋҲҽҝҊҸҝҸҦ қӃґ ҸѽӊҽҍѦӁ ҰҝҸҦҸӊҸґҧҡҲӊѦӁӁӃѸ ѹ ҲҸӊҽҝӁӃѸ ҔҝѹӉѹҡҸѸ. (ѨґҧӍ ѹ ҭѦҲҝҸӊ, 12 ҡҲҏґҧѦӊ)

(53P)

W ciƾgu trzech dni nastƾpi’ w Starogrodzie kryzys ǻywnoǧciowy.

(53Cz) Ve tǥech dnech se projevil ve mLJstLJ nedostatek potravin a všeho zboží. (53Sb)

U toku tri dana Stargorod je zahvatila kriza namirnica i robe.

8. I have mentioned in the foregoing that the choice from a set of converse structures is a choice between alternatives with more or less the same informational content. Now, if we go a bit further we see that there are cases where the linear-intonational hierarchy influences the choice (this is the point where the “to a large extent independent” of note 8 ends). First, in Keijsper (2003: 189-190) I called attention to an observation of KҸҝҸґѦӊ (1969: 181-182), who said that a nominative subject is undesirable in a position between the first and the last accent of a sentence, as a non-final Rheme. In that position the agent of a passive sentence, in the instrumental case, is preferred in Russian. An example was (28R), partly repeated here as (54): (54R)

ҩӀѽѹӁ, ҸқҝѦѸѦӁѦӁӁӃҮ ҰҸґҸҲѦӁ҃ҽѸѹ, қӃґ ӊҡҲҝѦҍѦӁ үҝҍѹқҽґҧҦҸѸ үҝҍѹқҽґҧҦҸӊѹҍѦѸ ҸҍѦӁҧ ҰҝѹӊѦҲґѹӊҸ ѹ ҲҸҲҍҽҡ ѹӉқҽӊґѦӁ ҸҲ ҰҝҸҔґҾҲӃѽ ҲҝҾҰҸҔ (ҷҏґҊҽҔҸӊ, ҴҽҡҲѦҝ ѹ ҴҽҝҊҽҝѹҲҽ)

The first accent is on ҩӀѽѹӁ; the last accent in the part before the conjunction ѹ is on ҰҝѹӊѦҲґѹӊҸ; the agent is in between these accents (and is probably also lightly accented). In Russian the peripheral instrumental case is normal in that position while S in OVS is not. As the translations of (28) show, Serbian replaces the Russian passive by OVS: (54Sb)

ҩҎҏѽѹӁҽ, ӁҽҲҸӊҽҝѦӁҸҊ ҏқҝҏҡѹѸҽ, ҦҸҍѦҔҽ үҝҍѹқҽґҦ үҝҍѹқҽґҦҸӊѹҍ ҒҏқҽӉӁҸ ѹ ҸҦѸҽѽ Ҋҽ ҸҡґҸқҸҦѹ ҰҝҸҔґѦҲѹѽ ҔҝҰҽ.

Other translations avoid the problem by moving the adverb to a position before the agent, which then is in final position in the part of the sentence before the conjunction, so that OVS is no problem (also in Russian), although the informational content is different from that of the Russian original then. In short, it may be the case that in Slavic languages other than Russian the use of passive constructions still more than is the case in Russian (Keijsper 2003: 189) serves to avoid mentioning an agent. A factor favouring OVS, then, can be the wish to mention an agent.

FROM OVS ORDER TO CONVERSE STRUCTURE IN RUSSIAN

385

To be sure (cf. Dik and Gvozdanoviǀ 1980, for example), it is not the case that passive constructions with an agent specification are impossible in, in our case, Serbian: (55R)

ҫӁ ҸқҝҽҲѹґҡҾ Ҕ ӁѹѸ ҡ ҰҝҸҡҧқҸҮ Ҹқ ҏқѦңѹӂѦ, ѹ ҸӁѹ ҲҸҲҍҽҡ ңѦ ҰҸҡґҽґѹ Ӊҽ ӁѹѸ ҡҰѦ҃ѹҽґҧӁҸѦ ҡҏҦӁҸ, ҔҸҲҸҝҸѦ ҰѦҝѦӊѦӉґҸ Ӂҽ ҩҸҦҸҡ ѹ ӊҡӀ ѦҊҸ ҡӊѹҲҏ – ҸҔҸґҸ ҲҝѹҦ҃ҽҲѹ ҍѦґҸӊѦҔ. ҴҾҲѦңӁѹҔ ѹ ҰҝѦҲѦӁҦѦӁҲ Ӂҽ ҰҝѦҡҲҸґ қӃґ ӊҡҲҝѦҍѦӁ ҡ ҃ҽҝҡҔѹѸѹ ҰҸҍѦҡҲҾѸѹ ѸҽҊѹҡҲҝҸѸ ҸҝҦѦӁҽ Ҧ’ҫқѹҡҡҸӁҸѸ (Pierre d’Aubisson), ѸҸӁҽѽҽѸѹ-ҝӃ҃ҽҝҾѸѹ ѹ ӁҽҡѦґѦӁѹѦѸ ҸҡҲҝҸӊҽ.

(55Sb)

Obratio im se sa molbom za utoǁište i oni su, jedva doǁekavši, poslali odmah naroǁitu galiju koja je sa obale prebacila njega i celu njegovu pratnju, oko tridesetak lica, na Rod. Odmetnik i pretendent na presto doǁekan je sa carskim poǁastima od velikog majstora viteškog reda D’Obisona (Pierre d’Aubusson), svih vitezova-redovnika i celokupnog stanovništva. (Andriǀ, Prokleta avlija)

9. A second, still more important point where linear-intonational hierarchy influences the choice from a set of converse structures concerns verbs on the borderline of the meaning of Russian [O[VS]] explained in section 6: the verb fills in the intonational nexus introduced by the accent on S: “there is/was S manifesting itself within the boundaries of O by Ving”. We saw in section 8 that socalled “stylistic” reversal of two accented constituents (SVO) separates intonationally the predications “there is/was S” and “this S did VO”. Now, verbs whose meaning excludes the possibility that S is absent, for example by referring to the degree in which S is present (including a zero degree), are in conflict with the intonational predication “there is/was S” (which replaces the absence of S by its presence) if this predication is merged in a single constituent with the verb ([O[VS]]). The same verbs are perfectly normal when the predication “there is/was S” is separated intonationally from the meaning of V, as in SVO. Consider, for example, (56) and (57): (56RNP) ҫӁ ҔҽҔ ҝҽӉ ҰҝѹҔҽӉҽґ ӊҡѦѸ ҰҸҡѸҸҲҝѦҲҧ, ҔҽҔ ӊѦґѹҔҸґѦҰӁҸ ҴҽґӍҸҮ ҰҸҦңҽҝѹґ ҝҸҊҽҲӃѽ ҡґѹӉӁҾҔҸӊ, ҔҸҊҦҽ ҸқґҽҔҸ ҔѹҡґҸҲӁҸ-ӉѦґѦӁҸҊҸ ҦӃѸҽ ѹ ҊҝҸѸҔҸѦ җѹҰѦӁѹѦ ӁҽҰҸґӁѹґѹ ҰҸҦӉѦѸѦґҧѦ. (56E)

Things didn’t improve for the Gryffindors as the Potions lesson continued. Snape put them all into pairs and set them to mixing up a simple potion to cure boils. He swept around in his long black cloak, watching them weigh dried nettles and crush snake fangs, criticizing almost everyone except Malfoy, whom he seemed to like. He was just telling everyone to look at the perfect way Malfoy had stewed his horned slugs when clouds of acid green smoke and a loud hissing

386

CORNELIA E. KEIJSPER

filled the dungeon. (Rowling, Harry Potter and the Philosopher’s Stone) (56ROr) ҋӁѦҮҰ ҝҽҡҔҝѹҲѹҔҸӊҽґ ӊҡѦѽ, ҔҝҸѸѦ ҴҽґӍҸҾ, ҔҸҲҸҝҸѸҏ, ҸҍѦӊѹҦӁҸ, ҡѹѸҰҽҲѹӉѹҝҸӊҽґ, ѹ ҔҽҔ ҝҽӉ ҰҝѹӉӃӊҽґ ӊҡѦѽ ҰҸґӀқҸӊҽҲҧҡҾ ҲѦѸ, ҔҽҔ ҴҽґӍҸҮ ӊҽҝѹҲ ҝҸҊҽҲӃѽ ҡґѹӉӁҾҔҸӊ, ҔҸҊҦҽ ҲѦѸӁѹ҃ҽ ӊҦҝҏҊ ӁҽҰҸґӁѹґҽҡҧ ҾҦҸӊѹҲҸ-ӉѦґѦӁӃѸ ҦӃѸҸѸ ѹ ҊҝҸѸҔѹѸ җѹҰѦӁѹѦѸ. (56RSp) ҈ґѦҮ ҔҽҔ ҝҽӉ ҰҝѹӉӃӊҽґ ӊҡѦѽ ҰҸҡѸҸҲҝѦҲҧ, ҔҽҔ ѹҦѦҽґҧӁҸ ҴҽґӍҸҮ ӊӃӊҽҝѹґ ҝҸҊҽҲҏӀ ҏґѹҲҔҏ, ҔҸҊҦҽ ӊҦҝҏҊ ҸҲҔҏҦҽ-ҲҸ ҰҸӊҽґѹґѹ ҔґҏқӃ ӉѦґѦӁҸҊҸ ѦҦҔҸҊҸ ҦӃѸҽ, ѹ ҰҸ ӊҡѦѸҏ ҰҸҦӉѦѸѦґҧӀ ҝҽӉӁѦҡґҸҡҧ ҊҝҸѸҔҸѦ җѹҰѦӁѹѦ. (56P)

Mówi’ w’aǧnie, by wszyscy siNj przyjrzeli, jak Malfoy znakomicie uwarzy’ swoje rogate ǧlimaki, kiedy nagle loch wype’ni’a chmura gryzƾcego, zielonego dymu i rozleg’ siNj g’oǧny syk.

(56Cz) PrávLJ všechny vyzýval, aǬ se podívají, jak dokonale Malfoy podusil hlemýždLJ, když celé sklepení naplnil oblak nakyslého zeleného dýmu a hlasitý sykot. (56Sb) ѰҰҝҽӊҸ ҎѦ ҊҸӊҸҝѹҸ ҔҽҔҸ ҡӊѹ ҲҝѦқҽ Ҧҽ ҡѦ ҏҊґѦҦҽҎҏ Ӂҽ ҡҽӊҝҗѦӁҡҲӊҸ ҡ ҔҸҎѹѸ ҎѦ ҴѦґӍҸҎ ҔҏӊҽҸ ҡӊҸҎѦ ҰҏңѦӊѦ ҒҏҡҔѦ, ҔҽҦҽ ҰҸҝҏѸҡҔҏ ҸҦҽҎҏ ѹҡҰҏӁѹҗѦ Ҹқґҽ҃ѹ ҔѹҡѦґҔҽҡҲҸҊ ӉѦґѦӁҸҊ ҦѹѸҽ ѹ ҊґҽҡӁҸ җѹҗҲҽѬѦ. (56Hr) Govorio je ostalima neka samo pogledaju kako je Malfoy savršeno zgnjeǁio rogate puževe, pri ǁemu su tamnicu ispunili oblaci otrovnozelenog dima i glasno pištanje. (56Bg) ѷӆҔѸҸ ҔҽӉӊҽҗѦ Ӂҽ ӊҡѹҍҔѹ Ҧҽ ҰҸҊґѦҦӁҽҲ ҔҸґҔҸ ҡӆӊӆҝҗѦӁҸ ҴҽґӍҸҮ Ѧ ӉҽҦҏҗѹґ ҝҸҊҽҲѹҲѦ ҡѹ ҰґҏңѦ҃ѹ, ҔҸҊҽҲҸ Ҹқґҽ҃ѹ ґӀҲѹӊ ӉѦґѦӁ ҦѹѸ ѹ ҡѹґӁҸ ҡӆҡҔҽӁѦ ѹӉҰӆґӁѹѽҽ ҰҸҦӉѦѸѹѦҲҸ. (56Mc) ѷҽѸҽӁ ѹѸ ҔҽңҏӊҽҗѦ ҡѹҲѦ Ҧҽ ҰҸҊґѦҦӁҽҲ ҔҸґҔҏ ҏқҽӊҸ Ҋѹ ҡӊҽҝѹґ ѩҝҽҔҸ ҡӊҸѹҲѦ ҊҸґѹ ҰҸґңҽӊѹ, ҔҸҊҽ ҏҍѹґӁѹ҃ҽҲҽ ҡѦ ӁҽҰҸґӁѹ ҡҸ Ҹқґҽ҃ѹ ҔѹҡѦґ ӉѦґѦӁ ҍҽҦ ѹ ҡҸ ҊґҽҡӁҸ җѹҗҲѦѬѦ. (56Uk) ҋӁѦҮҰ ҡҽѸѦ ҰҝҸҰҸӁҏӊҽӊ ҏҡҟѸ ҰҸҦѹӊѹҲѹҡҾ, ҾҔ ҦҸқҝѦ ҴѦґӍҸӀ ӊҦҽґҸҡҾ ӁҽҡҲҸҾҲѹ ҡґѹѸҽҔҟӊ, ҽң ҝҽҰҲҸѸ ҰҟҦӊҽґ ӁҽҰҸӊӁѹӊҡҾ ҾҦҏҍѹѸ ӉѦґѦӁѹѸ ҦѹѸҸѸ ҟ ҰҝҸӁѹӉґѹӊѹѸ җѹҰҟӁӁҾѸ. (57R)

ғҡѦѸ ѹӉӊѦҡҲӁҸ, ҍҲҸ ҔҽңҦҸѦ ӊӁҸӊҧ ҸҲҔҝӃҲҸѦ ҰҸґѦӉӁҸѦ ӊѦӂѦҡҲӊҸ, ҔҽңҦҽҾ ӊӁҸӊҧ ҸҲҔҝӃҲҽҾ ҡѹґҽ – ҙҲҸ ӁҸӊҽҾ ҡҲҏҰѦӁҧ ӊ ҝҽӉӊѹҲѹѹ ҃ѹӊѹґѹӉҽ҃ѹѹ. ҷҝҸӁӉҽ ҡҸӉҦҽґҽ ҽӁҲѹҍӁҏӀ ҃ѹӊѹґѹӉҽ҃ѹӀ, ңѦґѦӉҸ – ҡҝѦҦӁѦӊѦҔҸӊҏӀ, ҰҸҝҸѽ ӉҽӊѦҝҗѹґ ҡҝѦҦӁѦӊѦҔҸӊҧѦ, ҽ ҔҽѸѦӁӁӃҮ ҏҊҸґҧ ҸҲҔҝӃґ ҙҰҸѽҏ ҦѦӊҾҲӁҽҦ҃ҽҲҸҊҸ ҡҲҸґѦҲѹҾ.

FROM OVS ORDER TO CONVERSE STRUCTURE IN RUSSIAN

387

(57P)

Bo juǻci jest faktem, ǻe kaǻdy nowy a waǻny materia’, kaǻda nowa si’a to nowe piNjtro cywilizacji. Brƾz stworzy’ cywilizacjNj klasycznƾ, ǻelazo wieki ǧrednie; proch zakoǝczy’ wieki ǧrednie, a wNjgiel kamienny rozpoczƾ’ wiek dziewiNjtnasty. (Prus, Lalka)

(57Sb)

ҷҝҸӁӉҽ ҎѦ ҡҲӊҸҝѹґҽ ҔґҽҡѹҍӁҏ ҃ѹӊѹґѹӉҽ҃ѹҎҏ, ҊӊҸң҉Ѧ ҡҝѦҦѬѹ ӊѦҔ; қҽҝҏҲ ҎѦ ӉҽӊҝҗѹҸ ҡҝѦҦѬѹ ӊѦҔ, ҽ ҔҽѸѦӁѹ ҏҊҽҒ ҰҸҍѦҸ ҎѦ ҦѦӊѦҲӁҽѦҡҲѹ.

The subjects are not agentive (cf. group 1 in section 5), so at first sight the examples with SVO may seem to be “stylistic” reversals of OVS (section 8), which order actually appears in (56P), (56Cz), (56Sb), and (56Hr). However, there is only one OVS occurrence of ӁҽҰҸґӁҾҲҧ/ӁҽҰҸґӁѹҲҧ in Russian originals, and no OVS occurrence of ӉҽӊѦҝҗҽҲҧ/ӉҽӊѦҝҗѹҲҧ, so there is little to reverse. The cause of this is, as far as I understand, that the verbs tend to refer to the final phase of presence (a full degree of presence) but presuppose the fact of presence: the latter must then be introduced separately by intonational means: “there was S” / “the existence of S” and “it filled the dungeon” / “put an end to the Middle Ages” (SVO). The single OVS occurrence of ӁҽҰҸґӁҾҲҧ/ӁҽҰҸґӁѹҲҧ in Russian originals is (58): (58R)

ҫӁ ҔҸҍѦӊҽґ ѹӉ ҸҍѦҝѦҦѹ ӊ ҸҍѦҝѦҦҧ, ҰҝѹҡґҏҗѹӊҽґҡҾ Ҕ ҝҽӉҊҸӊҸҝҽѸ, ҦѦґҽґ ѦҦҔѹѦ ӉҽѸѦҍҽӁѹҾ, ѸӁҸҊҸӉӁҽҍѹҲѦґҧӁҸ ӉҽҦѹҝҽґ қҝҸӊѹ ѹ ҰҝҸҝҸҍѦҡҲӊҸӊҽґ. ҋґѦҦҡҲӊѹѦѸ ѦҊҸ ӁѦҦҸѸҸґӊҸҔ қӃґҸ ҲҸ, ҍҲҸ ҊҸҝҸҦ ӁҽҰҸґӁѹґѹ ҡґҏѽѹ Ҹ ҰҝѹѦӉҦѦ ҔҽҔҸҮ-ҲҸ ҡ ҴѦҍѹ ѹ Ѱҝҽґҽ ҰҸҦҰҸґҧӁҸҮ ҸҝҊҽӁѹӉҽ҃ѹѹ. (ѨґҧӍ ѹ ҭѦҲҝҸӊ, 12 ҡҲҏґҧѦӊ)

(58P)

Skutek jego pó’s’ówek by’ taki, ǻe w mieǧcie zaczNj’y krƾǻyǀ g’uche wieǧci o przyjeǺdzie spiskowej organizacji.

(58Cz) Výsledkem jeho náznakových ǥeǁí bylo, že po mLJstLJ se zaǁaly šíǥit zvLJsti o pǥíjezdu nLJjaké ilegální organizace z Meǁe a Bradla. (58Sb)

Posledica njegovih aluzija bila je da se u gradu proneo glas o dolasku neke ilegalne organizacije s Meǁi i Urala.

A (rare) OVS occurrence of ӉҽӊѦҝҗҽҲҧ/ӉҽӊѦҝҗѹҲҧ is (59R) (see (15) for a normal case of VOS with this verb): (59R)

Ѱ ӊҽҡ ѹѸѦѦҲҡҾ ҔґӀҍ ѹ ӁҸѸѦҝ, ӁѦ ѹӉӊѦҡҲӁӃҮ қҸґҧҗѦ ӁѹҔҸѸҏ. ҘґӀҍ – ҙҲҸ ҲҸґҧҔҸ ҍҽҡҲҧ ҰҝҸ҃ѦҦҏҝӃ ѹҦѦӁҲѹӍѹҔҽ҃ѹѹ. ѣѦ ӉҽӊѦҝҗҽѦҲ ґѹҍӁӃҮ ӁҸѸѦҝ. ғҡѦ ҙҲҸ ҦѦґҽѦҲҡҾ ӊ ҃ѦґҾѽ ҰҸҦҡҲҝҽѽҸӊҔѹ. ғѦҦҧ Ѧҡґѹ ӊӃ, ҦҸҰҏҡҲѹѸ, ҰҸҲѦҝҾѦҲѦ ҔґӀҍ, ѹѸ ѸҸңѦҲ ӊҸҡҰҸґҧӉҸӊҽҲҧҡҾ ҔҲҸ-ҲҸ ҦҝҏҊҸҮ.

388

CORNELIA E. KEIJSPER

(59E)

The man paused at the door, looking pale. “But of course. Like most Swiss banks, our safe-deposit boxes are attached to a NUMBER, not a name. You have a key and a personal account number known only to you. Your key is only half of your identification. Your personal account number is the other half. Otherwise, if you lost your key, anyone could use it.” (Brown, The Da Vince Code)

(59P)

Macie paǝstwo klucz, a wasz osobisty numer rachunku znacie tylko wy. Klucz jest zaledwie po’owƾ identyfikacji skrytki. Osobisty numer rachunku jest drugƾ po’owƾ.

(59Cz1) Máte klíǁ a osobní ǁíslo, které znáte jenom vy. Klíǁ je pouze jednou polovinou identifikace. Vaše osobní ǁíslo je druhou polovinou. (59Cz2) Máte klíǁ a ǁíslo úǁtu, které nikdo jiný nezná. Klíǁ je jen polovina identifikaǁního kódu, druhá polovina je vaše ǁíslo. (59Bg) ҘґӀҍӆҲ Ѧ ҡҽѸҸ ҰҸґҸӊѹӁҽҲҽ ҸҲ ӊҽҗҽҲҽ ѹҦѦӁҲѹӍѹҔҽ҃ѹҾ. һѹҍӁѹҾҲ ӊѹ ӁҸѸѦҝ Ѧ ҦҝҏҊҽҲҽ ҰҸґҸӊѹӁҽ. (59Mc) ҘґҏҍҸҲ ѹ ґѹҍӁѹҸҲ қҝҸҎ Ӂҽ ҡѸѦҲҔҽҲҽ Ҋѹ ӉӁҽѦҲѦ ҡҽѸҸ ӊѹѦ. ҘґҏҍҸҲ Ѧ ҡҽѸҸ ҰҸґҸӊѹӁҽ ѹҦѦӁҲѹӍѹҔҽ҃ѹҎҽ. һѹҍӁѹҸҲ қҝҸҎ Ӂҽ ҡѸѦҲҔҽҲҽ Ѧ ҦҝҏҊҽҲҽ ҰҸґҸӊѹӁҽ. Now, to begin with, the issue is subtle enough for another prefix to turn the scales. ғҡѦӊҸґҸҦҸӊҽ (2000: 45) calls ҰѦҝѦҰҸґӁҾҲҧ/ҰѦҝѦҰҸґӁѹҲҧ the basic member of ҽ pair of converses in (60): (60R)

Ҙ ҙҲҸѸҏ ӊҝѦѸѦӁѹ ғӃҗҊҸҝҸҦ ҰѦҝѦҰҸґӁѹґѹ ҡґҏѽѹ Ҹ ҍҏҦѦҡӁӃѽ ӉӁҽѸѦӁѹҾѽ

The -n/-t participle (… қӃґ ҰѦҝѦҰҸґӁѦӁ …) is called derived (ibidem). See also example (14) in section 4 (ӉҽҰҸґӁҾҲҧ/ӉҽҰҸґӁѹҲҧ). Presumably, then, in these cases the verb can successfully be seen as a manifestation of being present. More important is the fact that the borderlines of the issue are possibly language-specific. For example, there is a whole series of Polish examples with, inter alia, wype’n- that have OVS in cases where Russian chooses the reflexive verb instead of the problematic OVS of ӁҽҰҸґӁҾҲҧ/ӁҽҰҸґӁѹҲҧ, for example: (61R)

ҩҙҦҝѹҔ ҰҝҸӊҸҦѹґ ѦҊҸ ҊґҽӉҽѸѹ, ҰҸҦҸӉӊҽґ ҸӍѹ҃ѹҽӁҲҔҏ, ҡҰҝҸҡѹґ ҰҽҍҔҏ “һҽҔѹ ҡҲҝҽҮҔ”, ҝҽҡҰґҽҲѹґҡҾ ѹ, ӊӉҾӊҗѹ ҰҸҝҲӍѦґҧ, ӁѦҲҸҝҸҰґѹӊҸ ҰҸҗѦґ ҍѦҝѦӉ ҏґѹ҃ҏ Ҕ ҸҲѦґӀ. ҋҸґӁ҃Ѧ ҏңѦ ѹӉҝҾҦӁҸ ҰҝѹҰѦҔҽґҸ, ҏґѹ҃ҽ қӃҡҲҝҸ ӁҽҰҸґӁҾґҽҡҧ ӊґҽңӁҸҮ ҦҏѽҸҲҸҮ, ѹ ҩҙҦҝѹҔ ҸӂҏҲѹґ ңңѦӁѹѦ ҰҸҦ ӊѦҔҽѸѹ. (ҋҲҝҏҊҽ҃ҔѹѦ, ҭѹҔӁѹҔ Ӂҽ ҸқҸҍѹӁѦ)

FROM OVS ORDER TO CONVERSE STRUCTURE IN RUSSIAN

(61P)

389

S’oǝce przypieka’o juǻ mocno, ulicNj szybko wype’nia’ wilgotny zaduch i Red poczu’, jak go piekƾ powieki.

(61Bg) ҋґӆӁ҃ѦҲҸ ӊѦҍѦ ҰҝѹґѹҍҽҗѦ ҦҸҡҲҽ, ӊґҽңѦӁ ӉӁҸҮ қӆҝӉҸ ѹӉҰӆґӊҽҗѦ ҏґѹ҃ҽҲҽ ѹ ҩѦҦҝѹҔ ҏҡѦҲѹ, ҍѦ Ѹҏ ґӀҲѹ ҰҸҦ ҔґѦҰҽҍѹҲѦ. This can be a matter of prefix meaning, but since the issue recurs in a number of verbs (see, for example, pokrywaǀ/pokryǀ and ҰҸҔҝӃӊҽҲҧ/ҰҸҔҝӃҲҧ in point 4) I suspect that more is at stake. In some Russian translations of Polish texts we more often then elsewhere find OVS, more specifically in cases where the scanty Belorussian information systematically avoids the “Polish” option: (62R)

ѢѦ ӉӁҽӀ, ҦҸґҊҸ ґѹ Ҿ ҰҝҸҡҲҸҾґ ҲҽҔ, ҸҰѦҝҗѹҡҧ Ҹ ѽҸґҸҦҾӂѹҮ ѸѦҲҽґґ ҡҲѦӁӃ. ҋҲҽӁ҃ѹӀ ӁҽҰҸґӁҾґҽ ҲѹҗѹӁҽ, ѹ ґѹҗҧ ѸҸӁҸҲҸӁӁҸ җҏѸѦґѹ ҔҸѸҰҝѦҡҡҸҝӃ ҔґѹѸҽҲѹҍѦҡҔѹѽ ҏҡҲҽӁҸӊҸҔ.

(62P)

Nie wiem, jak d’ugo sta’em oparty o ch’odny metal ǧciany. StacjNj wype’nia’a cisza, a jedynym s’yszalnym dǺwiNjkiem by’ odleg’y, monotonny odg’os klimatyzacyjnych sprNjǻarek. (Lem, Solaris)

(62BR) ѢѦ ӊѦҦҽӀ, ҔҸґҧҔҟ Ҿ ҰҝҽҡҲҽҾҕ, ҽқҽҰҺҝҗӃҡҾ Ӂҽ ѽҽґҸҦӁӃ ѸѦҲҽґ ҡ҃ҾӁӃ. Ѣҽ ҋҲҽӁ҃Ӄҟ ҰҽӁҽӊҽґҽ ҃ҟҗӃӁҾ, ҽҦӉҟӁӃѸ ңӃӊӃѸ ҰҟҡҔҽѸ қӃҕ ҦҽґҺҔҟ ѸҽӁҽҲҸӁӁӃ Ҋҏґ ҔҽӁҦӃ҃ӃҾӁѦҝҽҕ. So I suggest that the over-use of OVS in Russian translations can be a Polonism. 10 But, evidently, this over-use is not so strikingly incorrect that it does not occur. There is even a case where one of four Russian translations of the same English text has OVS while the others have an adjective: (63RVeb) ҘҽҔ-ҲҸ ҝҽӉ ҘҝҸґѹҔ ѹ ҤҝӀҔҽ ҡѹҦѦґѹ ҸҔҸґҸ ҦҸѸѹҔҽ ғѹӁӁѹ-ҭҏѽҽ ѹ ҡґҏҗҽґѹ, ҍҲҸ ӊѦӂҽѦҲ ҘҝҸґѹҔ. ҘҸѸҰҽӁѹӀ ѹѸ ҡҸҡҲҽӊґҾґ ѹ ҭҏѽ. һѦҲӁѹҮ ҦѦӁҧ ӊӃҦҽґҡҾ ҲѦҰґӃѸ, ҦҝѦѸҸҲӁӃѸ, һѦҡ ӁҽҰҸґӁҾґѹ ҰҝѹҾҲӁӃѦ ҦґҾ ҡґҏѽҽ ҊҸґҸҡҽ ѹ, ҔҽӉҽґҸҡҧ, ӊҡѦ ҸӁѹ ҏқѦңҦҽґѹ ғѹӁӁѹ: “ѢѦ ҡґҏҗҽҮ ҘҝҸґѹҔҽ, ҡґҏҗҽҮ Ӂҽҡ”. (63E)

One day Rabbit and Piglet were sitting outside Pooh’s front door listening to Rabbit, and Pooh was sitting with them. It was a drowsy summer afternoon, and the Forest was full of gentle sounds, which all seemed to be saying to Pooh, “Don’t listen to Rabbit, listen to me.” (Milne, The House at Pooh Corner)

(63RRud) ҫҦӁҽңҦӃ ҘҝҸґѹҔ ѹ ҭҸҝҸҡѦӁҸҔ ҡѹҦѦґѹ ӊҸӉґѦ ҰҽҝҽҦӁҸҮ ҭҏѽҸӊҽ ҦҸѸҽ, ѹ ҭҏѽ ҲҸңѦ ҡ ӁѹѸѹ ҡѹҦѦґ. ҷӃґ ҲҽҔҸҮ ҦҝѦѸҸҲӁӃҮ ґѦҲӁѹҮ 10

To be sure, they do not contradict my observations in Keijsper (2003: 188-205).

390

CORNELIA E. KEIJSPER

ҦѦӁҧ, ѹ һѦҡ қӃґ ҰҸґҸӁ ҰҝѹҾҲӁӃѽ ӉӊҏҔҸӊ, ҔҸҲҸҝӃѦ ӊҡѦ, ҔҽҔ ҸҦѹӁ, ҔҽӉҽґҸҡҧ, ҊҸӊҸҝѹґѹ ҭҏѽҏ: “ѢѦ ҡґҏҗҽҮ ҘҝҸґѹҔҽ, ҡґҏҗҽҮ ѸѦӁҾ”. (63RMa) ҫҦӁҽңҦӃ ҘҝҸґѹҔ ѹ ҭҾҲҽҍҸҔ ҡѹҦѦґѹ Ӂҽ ҰҸҝҸҊѦ ҦҸѸѹҔҽ ғѹӁӁѹҭҏѽҽ ѹ ҡґҏҗҽґѹ, ҍҲҸ ҊҸӊҸҝѹҲ ҘҝҸґѹҔ. ҭҏѽ ҰҝѹѸҸҡҲѹґҡҾ ҝҾҦӃҗҔҸѸ. ҷӃґ ҡҸӁӁӃҮ ґѦҲӁѹҮ ҰҸґҦѦӁҧ. һѦҡ қӃґ ҰҸґҸӁ ӁѦңӁӃѽ ҍҽҝҏӀӂѹѽ ӉӊҏҔҸӊ ѹ, ҔҽӉҽґҸҡҧ, ҊҸӊҸҝѹґ ҭҏѽҏ: “ѢѦ ҡґҏҗҽҮ ҘҝҸґѹҔҽ, ҡґҏҗҽҮ ѸѦӁҾ”. (63RZa) ҘҝҸґѹҔ ѹ ҭҾҲҽҍҸҔ ҡѹҦѦґѹ ӊҸӉґѦ ҰҽҝҽҦӁҸҮ ҦӊѦҝѹ ҦҸѸҽ ғѹӁӁѹ-ҭҏѽҽ ѹ ҡґҏҗҽґѹ, ҍҲҸ ҊҸӊҸҝѹҲ ҘҝҸґѹҔ. ғѹӁӁѹ-ҭҏѽ ҲҸңѦ ҡѹҦѦґ ҡ ӁѹѸѹ. ҷӃґ ҦҝѦѸҸҲӁӃҮ ґѦҲӁѹҮ ҰҸґҦѦӁҧ, ѹ һѦҡ қӃґ ҰҸґҸӁ Ҳѹѽѹѽ, ӁѦҾҡӁӃѽ ӉӊҏҔҸӊ, ѹ ӊҡѦ ҸӁѹ, ҔҽӉҽґҸҡҧ, ҊҸӊҸҝѹґѹ ҭҏѽҏ: “ѢѦ ҡґҏҗҽҮ ҘҝҸґѹҔҽ. ҋґҏҗҽҮ ѸѦӁҾ”. (63Cz) Jednoho dne sedLJl Králíǁek s Prasátkem pǥed Pilovým domkem a poslouchali Králická a Pú sedLJl s nimi. Bylo ospalé letní odpDzldne a Les byl plný nLJžných šelestDz a Púovi se zdálo, že na nLJho všecky volají: “Neposlouchej Králiǁka, poslouchej nás”. (63Sk)

Jedného dǞa sedeli Prasiatko a Králik pred Púovými dverami a poǁúvali Králika. Pú tam sedel s nimi. Bolo ospalé letné popoludnie a z Lesa sa ozývali príjemné zvuky, ktoré Púovi akoby vraveli: “Nepoǁúvaj Králika, poǁúvaj nás.”

(63Uk) ҷҏӊ ҦҝҟѸҸҲӁѹҮ ґҟҲӁҟҮ ҰҸґҏҦѦӁҧ, ҟ һҟҡ қҏӊ ҡҰҸӊӁѦӁѹҮ ҍҽҝҟӊӁҸҔҸґѹҡҔҸӊѹѽ ӉӊҏҔҟӊ, ҔҸңѦӁ ҟӉ ҾҔѹѽ, ӉҦҽӊҽґҸҡҾ, ҔҽӉҽӊ ҭҏѽҸӊҟ: “ѢѦ ҡґҏѽҽҮ ҘҝҸґѹҔҽ. ҋґҏѽҽҮ ѸѦӁѦ.” (63Sn) Nekega dne sta Zajec in Pujsek sedela pred Pujevo hišo in poslušala Zajca, Pu pa je sedel z njima. Bil je zaspan poletni popoldan in Gozd je bil poln tihih glasov; Puju se je zdelo, da mu vsi pravijo: “Ne poslušaj Zajca, poslušaj nas!.” The OVS occurrence in (63RVeb) cannot be ascribed to influence of the English original, of course. The issue is moreover not only about Russian and Polish, although as yet most clear examples are from these languages. A Russian-Czech example is (64) (see also (56)): (64R)

ҫӁ ҰѦҝѦҡҲҽґ қӃҲҧ ҡҔѦҰҲѹҔҸѸ, ѹ Ҧҏҗҽ ѦҊҸ ӁҽҰҸґӁѹґҽҡҧ ҡҸҡҲҝҽҦҽӁѹѦѸ Ҕ ҹӊѦҮҔҏ.

(64Cz) … Nemám pravdu, Toníku?” obrátil se k ǁahounovi, který potichu ǥekl: “Bodla bychom mohli sundat. Je to pǥece náš ǁlovLJk.” Pǥestal být skeptikem a duši jeho naplnila soustrast k Švejkovi. (Hašek, Švejk)

FROM OVS ORDER TO CONVERSE STRUCTURE IN RUSSIAN

391

This issue is the most important one of my OVS study, because I am unable to say at this point whether the difference between Russian and other Slavic languages is entirely a matter of lexical meaning and of the meaning of prefixes, or also implies that my analysis of Russian [O[VS]] (section 6) is not applicable to other Slavic languages without modifications (preferably related to the different placement of the verb in these languages). If the latter will appear to be the case in the future, it means that the close-knit constituent [VS] in [O[VS]] which is so problematic for some general linguists (see section 2) is less close-knit in languages which, unlike Russian, even need a verb rather than linear-intonational hierarchy alone, to express the meanings ‘to be’ and ‘to have’. This links up the study of OVS order with ҴҝҽӉѦҔ’s 1990 study of the expression of esse / habere as one of the most important syntactic issues in Slavic languages. 10. Finally, there is an issue of style. In section 10 we saw that ҴҽґҧҍѹҔ қѦҊґҸ ҍѹҲҽѦҲ is easier than ѯҲѦӁѹѦ ѸҽґҧҍѹҔҽ ҸҲґѹҍҽѦҲҡҾ қѦҊґҸҡҲҧӀ because of the non-isosemic character of the latter. In ASPAC the heavy style occurs in the Russian translation of Orwell’s Animal farm, for example, but is rare elsewhere. Still worse is the case for OVS with the non-reflexive verb ҸҲґѹҍҽҲҧ as far as ASPAC is concerned: we must even go outside the Slavic originals to locate a few examples. One of them is (65): (65R)

ҴҸҝҲҽҲѹ, ҔҽҔ ҰҸґҽҊҽґѹ ѸӁҸҊѹѦ, ѸҸҊ ҡҲҽҲҧ ҰҽҰҸҮ ӊ қҸґѦѦ ҝҽӁӁѦѸ ӊҸӉҝҽҡҲѦ, Ѧҡґѹ қӃ ӁѦ ҸқґҽҦҽґ ҸҦӁѹѸ ӊѦҡҧѸҽ ҡѦҝҧѦӉӁӃѸ ӁѦҦҸҡҲҽҲҔҸѸ. ҘҽҝҦѹӁҽґҽ ҴҸҝҲҽҲѹ ҸҲґѹҍҽґҽ җѹҝҸҲҽ ӊӉҊґҾҦҸӊ, ҍҲҸ ҰҝҸҲѹӊҸҝѦҍѹґҸ ҏҡґҸӊѹҾѸ ҋӊҾҲҸҮ ҲҝѹҽҦӃ, ҡҸқґӀҦѦӁѹѦ ҔҸҲҸҝӃѽ ҲҝѦқҸӊҽґҸҡҧ ҦґҾ ѹӉқҝҽӁѹҾ Ӂҽ ҰҸҡҲ ҰҽҰӃ. ҶҲѹ ҲҝѹҽҦҽ ӉҽҔґӀҍҽґҽҡҧ ӊ ҲҝѦѽ ҡґҸӊҽѽ – ҔҸӁҡѦҝӊҽҲѹӉѸ, ҔҸӁҡѦҝӊҽҲѹӉѸ ѹ ҔҸӁҡѦҝӊҽҲѹӉѸ.

(65E)

Mortati, many believed, could have been Pope in his younger days had he not been so broad-minded. When it came to pursuing the papacy, there was a Holy Trinity – Conservative. Conservative. Conservative. (Brown, Angels and Demons)

(65P)

Wiele osób uwaǻa’o, ǻe Mortati móg’by zostaǀ papieǻem za swoich m’odszych dni, gdyby nie mia’ tak szerokich horyzontów. Kiedy chodzi’o o stanowisko papieǻa, liczy’a siNj tylko ƦwiNjta Trójca – Konserwatysta, Konserwatysta, Konserwatysta.

(65Sb) Mnogi su verovali da je Mortati mogao postati papa u mlaDžim danima da nije bio toliko slobodouman. Kada su u pitanju kriterijumi za papsko zvanje, postojalo je Sveto trojstvo – Konzervativan. Konzervativan. Konzervativan. (65Bg) ҴӁҸӉѹӁҽ ҡѸҾҲҽѽҽ, ҍѦ ҔҽҲҸ ҰҸ-ѸґҽҦ ҴҸҝҲҽҲѹ Ѧ ѸҸңѦґ Ҧҽ ҡҲҽӁѦ ҰҽҰҽ, ҡҲѹҊҽ Ҧҽ ӁѦ Ѧ қѹґ ҲҸґҔҸӊҽ ҡӊҸқҸҦҸѸѹҡґѦӂ. ҋӊѦҲҽҲҽ ҲҝҸѹ҃ҽ Ӂҽ ӊҽ-

392

CORNELIA E. KEIJSPER

ҲѹҔҽӁҡҔҽҲҽ ҔҽҝѹѦҝҽ ҡѦ ҡӆҡҲҸѦҗѦ ҸҲ ҔҸӁҡѦҝӊҽҲѹӊӁҸҡҲ, ҔҸӁҡѦҝӊҽҲѹӊӁҸҡҲ ѹ ҰҽҔ ҔҸӁҡѦҝӊҽҲѹӊӁҸҡҲ. At the same time, a quick look on the internet gives us ca. 25 OVS occurrences of ҸҲґѹҍҽҲҧ on the first 10 pages returned by Google, so we must conclude that the verb is perfectly normal in OVS order in other than the literary styles represented in ASPAC. In terms of most active member (section 10), in OVS the active referent is the distinguishing feature, while in the reflexive verb it is the carrier of the feature. The fact that S is a property of O in OVS may give the impression that O is the largest referent and should be the subject. However, it is the negation of the role of S which would remove the entire situation (if there is no distinguishing there is no situation in which O is distinct from others by having the property S), therefore even this OVS occurrence is a regular nominative-accusative sentence in Russian (cf. Keijsper 2003: 180). Of course, as the original English text and the other Slavic translations show the verb is an almost empty filling of the meaning of the linear-intonational hierarchy. 12. Conclusion It will be clear that the problem of converse structures in Slavic is not exhausted by the present article. I hope that the elimination of non-existent issues introduced by linguistic theories is the first step towards the investigation of real issues. In no case encountered so far is there any reason to suppose that O(V)S order per se is avoided in Slavic languages, so general linguistic theories which would lead us to expect that outcome (see sections 1-2) are not corroborated. University of Amsterdam ABBREVIATIONS R = Russian, BR = Belorussian, Uk = Ukrainian, P = Polish, Cz = Czech, Sk = Slovak, So = (High-)Sorbian, Bg = Bulgarian, Mc = Macedonian, Sb = Serbian, Hr = Croatian, Sn = Slovenian, D = Dutch, E = English, G = German. Translations − Brown, Dan, The Da Vince Code (2003): Cz1 = Zdík Dušek, Cz2 = Barbora Michálková − Milne, Alan Alexander, The House at Pooh Corner (1928): RMa = ҋ.Ѥ. ҴҽҔҡѹѸѹҗѹӁ, RRud = ҩҏҦӁѦӊ, ѷ. ҴѹѽҽҮґҸӊҽ, RVeb = ғѹҔҲҸҝ ғѦқѦҝ, RZa = ҷ. ҈ҽѽҸҦѦҝ (one chapter not translated) − Rowling, J.K., Harry Potter and the Philosopher’s Stone (1997): RNP = “ѢҽҝҸҦӁӃҮ ҰѦҝѦӊҸҦ”, ROr = Ѩ.ғ. ҫҝҽӁҡҔѹҮ, RSp = ҴҽҝѹҾ ҋҰѹӊҽҔ − Tolkien, J.R.R., The Hobbit or There and Back Again (1937): PBra = Paulina Braiter, RKor = Ҙ. ҘҸҝҸґѦӊ, PSki = Maria Skibiniewska, RQu = ү.ғ. ӅҏҝҸӊ

FROM OVS ORDER TO CONVERSE STRUCTURE IN RUSSIAN −

393

Tolkien, J.R.R., The fellowship of the Ring, being the first part of The Lord of the Rings (1954, 1966): R1 = not indicated, R2 = ғ. ҴҏҝҽӊҧѦӊҽ (R2 may be based on the first edition and differs strongly from R1, which is more close to the English text included in the ASPAC corpus)

REFERENCES Adamec, Pǥemysl 1956 “K rozdílDz mezi ǁeským a ruským slovosledem”. Kapitoly ze srovnávací mluvnice ruské a ǁeské I. Studie syntaktické, 180-189. Praha: Nakladatelství ǁeskoslovenské akademie vLJd. Benoist, Jean-Pierre 1979 Les fonctions de l’ordre des mots en russe moderne. Romans et nouvelles de Gorki. Paris: Institut d’études slaves. Berneker, Erich 1900 Die Wortfolge in den slavischen Sprachen. Berlin: B. Behr’s Verlag (E. Bock). Bubenik, V. 1987 “The study of clitics in Functional Grammar”. In: J. Nuyts, G. de Schutter (eds.), Getting One’s Words into Line. On Word Order and Functional Grammar (Functional Grammar Series 5), 45-60. Dordrecht-Providence RI: Foris Publications. Chomsky, N. 1965 Aspects of the theory of syntax. Cambridge: MIT Press. Dik, S.C. 1978 Functional Grammar (North-Holland Linguistics Series 37). Amsterdam-New York-Oxford: North Holland Publishing Company. 1980 Studies in Functional Grammar. London etc.: Academic Press. 1989 The Theory of Functional Grammar. Part I: The Structure of the Clause (Functional Grammar Series 9). Dordrecht-Providence RI: Foris Publications. Dik, S.C., and J. Gvozdanoviǀ 1980 “Subject and Object in Serbo-Croatian”. In: T. Hoekstra, H. van der Hulst, M. Moortgat (eds.), Perspectives on Functional Grammar, 21-39. Dordrecht: Foris. Dyer, Donald L. 1992 Word order in the simple Bulgarian sentence: A study in grammar, semantics and pragmatics. Amsterdam-Atlanta: Rodopi. Greenberg, Joseph H. 1963 “Some universals of grammar with particular reference to the order of meaningful elements”. In: Joseph H. Greenberg (ed.), Universals of Language, 73113. Cambridge: MIT Press. Keijsper, C.E. 1985 Information Structure. With examples from Russian, English and Dutch (Studies in Slavic and General Linguistics 4). Amsterdam: Rodopi. 1990 Waar gaat de FG heen? Haarlem, Amsterdam: Rodopi. 1994 “From accent to transitivity”. In: A.A. Barentsen, B.M. Groen, R. Sprenger (eds.), Dutch Contributions to the Eleventh International Congress of Slavists, Bratislava: Linguistics (Studies in Slavic and General Linguistics 22), 193-329. Amsterdam-Atlanta: Rodopi.

394

CORNELIA E. KEIJSPER

2003

“Notes on intonation and voice in modern Russian”. In: J. Schaeken, P. Houtzagers, J. Kalsbeek (eds.), Dutch Contributions to the Thirteenth International Congress of Slavists, Ljubljana: Linguistics (Studies in Slavic and General Linguistics 30), 141-214. Amsterdam-New York: Rodopi. 2004 “Typically Russian”. Russian Linguistics 28, 189-226. King, Tracy Holloway 1995 Configuring Topic and Focus in Russian (Dissertations in linguistics). Stanford, California: CSLI Publications. Kǥížková, Helena 1953 “Problémy ǁeského a ruského slovosledu”. In: Kniha o pǥekládání, PǥíspLJvky k otázkám pǥekladu z ruštiny, 280-298. Praha: Nakladatelství ǁeskoslovenskosovLJtského institutu. Schaller, Helmut Wilhelm 1966 Die Wortstellung im Russischen (Slavistische Beiträge 21). München: Sagner. Siewierska, Anna 1988 Word Order Rules (Croom Helm linguistic series). London-New York-Sydney: Croom Helm. Stanchev, S. 1987 “Some observations on the order of constituents in Bulgarian”. In: J. Nuyts, G. de Schutter (eds.), Getting One’s Words into Line. On Word Order and Functional Grammar (Functional Grammar Series 5), 33-43. Dordrecht-Providence RI: Foris Publications. Wierzbicka, Anna 1992 Semantics, Culture, and Cognition. Universal Human Concepts in CultureSpecific Configurations. New York-Oxford: Oxford U.P. Zydatiʃ, Wolfgang 1974 “Some means for rendering the English passive in German”. Linguistische Berichte 29, 34-48. үҰҝѦҡҾӁ, Ҍ.ѩ. 1967

ӳӪԙԗӾӢӠԐӾӰԽԂԊӸӰӹӾ ӠԙԙԊӾԌӹӴԂӰӠӾ ԙӾԐԂӰԽӠӪӠ ӢӫԙԙӪӹԄӹ ԄԊԂԄӹԊԂ. ҴҸҡҔӊҽ: ѢҽҏҔҽ. 1995 ԘӾӪԙӠԓӾԙӪԂԋ ԙӾԐԂӰԽӠӪԂ (ԙӠӰӹӰӠԐӠԓӾԙӪӠӾ ԙӢӾԌԙԽӴԂ ԋԒՂӪԂ) (ѨӉқҝҽӁӁӃѦ ҲҝҏҦӃ, ҲҸѸ I), 2-e ѹӉҦҽӁѹѦ, ѹҡҰҝҽӊґѦӁӁҸѦ ѹ ҦҸҰҸґӁѦӁӁҸѦ. ҴҸҡҔӊҽ: ҹҔҸґҽ «ѤӉӃҔѹ ҝҏҡҡҔҸҮ ҔҏґҧҲҏҝӃ», ѨӉҦҽҲѦґҧҡҔҽҾ ӍѹҝѸҽ «ғҸҡҲҸҍӁҽҾ ґѹҲѦҝҽҲҏҝҽ», ҩүѢ. үҰҝѦҡҾӁ, Ҍ.ѩ. ѹ Ҧҝ. 1997 ԾӹӴՂӡ ӹԼԔԋԙӰӠԽӾԊӸӰՂӡ ԙԊӹӴԂӢӸ ԙӠӰӹӰӠԐӹӴ ӢӫԙԙӪӹԄӹ ԋԒՂӪԂ. ӷӾӢӴՂӡ ӴՂԗӫԙӪ (ҰҸҦ ҸқӂѦҮ ҝҏҔҸӊҸҦҡҲӊҸѸ ҽҔҽҦѦѸѹҔҽ Ҍ.ѩ. үҰҝѦҡҾӁҽ). ҴҸҡҔӊҽ: «ѤӉӃҔѹ ҝҏҡҡҔҸҮ ҔҏґҧҲҏҝӃ». ҷҽҍӊҽҝҸӊ, Ѥ. 1974 “ҭҸҝҾҦҸҔ ҡґҸӊ ҔҽҔ ҊҝҽѸѸҽҲѹҍѦҡҔҸѦ ҡҝѦҦҡҲӊҸ ӊ ҡґҽӊҾӁҡҔѹѽ ҾӉӃҔҽѽ (ӊ ҡҸҰҸҡҲҽӊґѦӁѹѹ ҡ ӁѦҔҸҲҸҝӃѸѹ ӁѦҡґҽӊҾӁҡҔѹѸѹ)”. ғ: ԱӢԂԐԐԂԽӠԓӾԙӪӹӾ ӹԗӠԙԂӰӠӾ ԙԊԂӴԋӰԙӪӠԖ ԋԒՂӪӹӴ, ԹӹӰՁӾԗՁӠӠ Ӡ ԐӾԽӹԌՂ, 211-218. ҴҸҡҔӊҽ: ѢҽҏҔҽ. ғҡѦӊҸґҸҦҸӊҽ, Ҵ.ғ. 1976 “ѢҸѸѹӁҽҲѹӊӁҸ-ҽҔҔҏӉҽҲѹӊӁӃѦ ҡҲҝҏҔҲҏҝӃ ѹ ѹѽ ҔҸӁӊѦҝҡѹӊӃ ӊ ҝҏҡҡҔҸѸ ҾӉӃҔѦ (Ҙ ӊҸҰҝҸҡҏ Ҹқ ѹҡҍѹҡґѦӁѹѹ ҲѹҰҸӊ ҔҸӁӊѦҝҡӁӃѽ ҡҲҝҏҔҲҏҝ)”. ԾԂӫԓӰՂӾ ԌӹӪԊԂԌՂ ӴՂԙԶӾӡ ԶӪӹԊӾ (ԢӠԊӹԊӹԄӠԓӾԙӪӠӾ ӰԂӫӪӠ) 6, 67-73.

FROM OVS ORDER TO CONVERSE STRUCTURE IN RUSSIAN

395

ԁӾӹӢӠԋ ԅӫӰӪՁӠӹӰԂԊӸӰӹ-ӪӹԐԐӫӰӠӪԂԽӠӴӰӹԄӹ ԙӠӰԽԂӪԙӠԙԂ: ԢӢԂԄԐӾӰԽ ԗӢӠӪԊԂԌӰӹӡ (ԗӾԌԂԄӹԄӠԓӾԙӪӹӡ) ԐӹԌӾԊӠ ԋԒՂӪԂ: ӯԓӾԼӰӠӪ. ҴҸҡҔӊҽ: ѨӉҦҽҲѦґҧҡҲӊҸ ҴҐѰ. ғҡѦӊҸґҸҦҸӊҽ, Ҵ.ҷ., ѹ ҫ.Ҍ. ѩѦѸѦӁҲҧѦӊҽ 1997 ӷӢӹԼԊӾԐՂ ԙӠӰԽԂӪԙӠԓӾԙӪӹӡ ԗԂӢԂԌӠԄԐԂԽӠӪӠ: ӪӹԐԐӫӰӠӪԂԽӠӴӰԂԋ ԗԂӢԂԌӠԄԐԂ ԗӢӾԌԊӹԞӾӰӠӡ (ӰԂ ԐԂԽӾӢӠԂԊӾ ԌӴӫԙӹԙԽԂӴӰՂԖ ԄԊԂԄӹԊӸӰՂԖ ԗӢӾԌԊӹԞӾӰӠӡ, ӴӪԊԇԓԂԇԷӠԖ ӠԐԋ ԊӹӪӫԐԂ). ҴҸҡҔӊҽ: ҘҝҸӁҰҝѦҡҡ. ҐҽӊҝѹґҸӊҽ, ғ.Ѩ. 1975 “ҫҡҸқѦӁӁҸҡҲѹ ҡѦѸҽӁҲѹҔѹ, ҡѹӁҲҽҔҡѹҡҽ ѹ ѸҸҝӍҸґҸҊѹѹ ҊґҽҊҸґҸӊ ҰҝѹҡҸѦҦѹӁѦӁѹҾ”. ғ: ՀӾԐӠӹԽӠӪԂ Ӡ ӠӰԅӹӢԐԂԽӠӪԂ (җѦҡҲҸҮ ӊӃҰҏҡҔ), 144-164. ҴҸҡҔӊҽ: ғѨѢѨѷѨ. ҐѦҸҝҊѹѦӊҽ, ѣґѦӁҽ 1974 ՀԊӹӴӹӢӾԌ ӰԂ ԗӢӹԙԽӹԽӹ ӠԒӢӾԓӾӰӠӾ Ӵ ԼԔԊԄԂӢԙӪӠԋ ӪӰӠԞӹӴӾӰ ӾԒӠӪ. ҋҸӍѹҾ: ѨӉҦҽҲѦґҡҲӊҸ Ӂҽ қӆґҊҽҝҡҔҽҲҽ ҽҔҽҦѦѸѹҾ Ӂҽ ӁҽҏҔѹҲѦ. ҈ҸґҸҲҸӊҽ, Ґ.ү. 1982 ԹӹԐԐӫӰӠӪԂԽӠӴӰՂӾ ԂԙԗӾӪԽՂ ӢӫԙԙӪӹԄӹ ԙӠӰԽԂӪԙӠԙԂ. ҴҸҡҔӊҽ: ѢҽҏҔҽ. ѨҸҝҦҽӁҡҔҽҾ, һ.Ѣ. 1970 “ҭҸҰӃҲҔҽ ґѦҔҡѹҔҸҊҝҽӍѹҍѦҡҔҸҊҸ ҲҸґҔҸӊҽӁѹҾ ҊҝҏҰҰӃ ҝҏҡҡҔѹѽ ҡґҸӊ ҡҸ ӉӁҽҍѦӁѹѦѸ ҍҏӊҡҲӊҽ”. ԠԂԶӠӰӰՂӡ ԗӾӢӾӴӹԌ Ӡ ԗӢӠӪԊԂԌӰԂԋ ԊӠӰԄӴӠԙԽӠӪԂ 13, 326. ҘӁҾӉѦӊ, ҌҝѹҮ ҭҽӊґҸӊѹҍ 1989 ԣӪՁӠӹӰԂԊӸӰӹԙԽӸ Ӡ ԙԽԂԽԂԊӸӰӹԙԽӸ: ԚԖ ԙӹӹԽӰӹԶӾӰӠӾ Ӵ ӢӫԙԙӪӠԖ ӪӹӰԙԽӢӫӪՁӠԋԖ ԙ ԗӢӠԓԂԙԽӠԋԐӠ ӰԂ -Ӂ, -Ҳ. München: Sagner. ҘҸӊҲҏӁҸӊҽ, Ѩ.Ѩ. 1976 ՀӹӴӢӾԐӾӰӰՂӡ ӢӫԙԙӪӠӡ ԋԒՂӪ. ӷӹӢԋԌӹӪ ԙԊӹӴ Ӡ ԂӪԽӫԂԊӸӰӹӾ ԓԊӾӰӾӰӠӾ ԗӢӾԌԊӹԞӾӰӠԋ. ӯԓӾԼӰӹӾ ԗӹԙӹԼӠӾ ԌԊԋ ԙԽӫԌӾӰԽӹӴ ԗӾԌԂԄӹԄӠԓӾԙӪӠԖ ӠӰԙԽӠԽӫԽӹӴ ԗӹ ԙԗӾՁӠԂԊӸӰӹԙԽӠ «ԪӫԙԙӪӠӡ ԋԒՂӪ Ӡ ԊӠԽӾӢԂԽӫӢԂ». ҴҸҡҔӊҽ: ҭҝҸҡӊѦӂѦӁѹѦ. ҘҸҝҸґѦӊ, Ҷ.Ѩ. 1966 “ҫқҝҽқҸҲҔҽ ҸҦӁҸҊҸ ҡґҏҍҽҾ ҸѸҸӁѹѸѹѹ Ұҝѹ ҽӊҲҸѸҽҲѹҍѦҡҔҸѸ ҽӁҽґѹӉѦ ҝҏҡҡҔҸҊҸ ҾӉӃҔҽ”. ԾԂӫԓӰӹ-ԽӾԖӰӠԓӾԙӪԂԋ ӠӰԅӹӢԐԂՁӠԋ 10, 39-41. 1969 “ѨҡҰҸґҧӉҸӊҽӁѹѦ Ұҽҡҡѹӊҽ Ұҝѹ ҽӊҲҸѸҽҲѹҍѦҡҔҸѸ ҡѹӁҲѦӉѦ ҝҏҡҡҔҸҊҸ ҰҝѦҦґҸңѦӁѹҾ”. ԠԂԶӠӰӰՂӡ ԗӾӢӾӴӹԌ Ӡ ԗӢӠӪԊԂԌӰԂԋ ԊӠӰԄӴӠԙԽӠӪԂ 11, 177-192. ҴҝҽӉѦҔ, ҩҸѸҽӁ 1990 ՀӢԂӴӰӠԽӾԊӸӰՂӡ ԙӠӰԽԂӪԙӠԙ ԙԊԂӴԋӰԙӪӠԖ ԊӠԽӾӢԂԽӫӢӰՂԖ ԋԒՂӪӹӴ. ԚԙԖӹԌӰՂӾ ԙԽӢӫӪԽӫӢՂ ԗӢӹԙԽӹԄӹ ԗӢӾԌԊӹԞӾӰӠԋ. Brno: Univerzita J.E. PurkynLJ v BrnLJ. ҭҽҦҏҍѦӊҽ, ѣ.ғ. 1985 ԛՂԙӪԂԒՂӴԂӰӠӾ Ӡ ӾԄӹ ԙӹӹԽӰӾԙӾӰӰӹԙԽӸ ԙ ԌӾӡԙԽӴӠԽӾԊӸӰӹԙԽӸԇ (ӢӾԅӾӢӾӰՁӠԂԊӸӰՂӾ ԂԙԗӾӪԽՂ ԙӾԐԂӰԽӠӪӠ ԐӾԙԽӹӠԐӾӰӠӡ). ҴҸҡҔӊҽ: ѢҽҏҔҽ. ҭҽҦҏҍѦӊҽ, ѣ.ғ., ѹ ҩ.Ѩ. ҩҸӉѹӁҽ “ҋѦѸҽӁҲѹҍѦҡҔѹҮ Ҕґҽҡҡ ҊґҽҊҸґҸӊ ҰҸґӁҸҊҸ ҸѽӊҽҲҽ: ҲҸґҔҸӊҽӁѹѦ ѹ ґѦҔҡѹҔҸ1993 ҡѹӁҲҽҔҡѹҍѦҡҔѹѦ ҡӊҸҮҡҲӊҽ”. ԛӹԗӢӹԙՂ ԋԒՂӪӹԒӰԂӰӠԋ 6, 5-16. ҋѦґѹӊѦҝҡҲҸӊҽ, ҫ.Ѣ., ѹ һ.ү. ҭҝҸӉҸҝҸӊҽ 1992 “ҘҸѸѸҏӁѹҔҽҲѹӊӁҽҾ ҰѦҝҡҰѦҔҲѹӊҽ ӊӃҡҔҽӉӃӊҽӁѹҾ”. ғ: ՀӫԼԔӾӪԽӰӹԙԽӸ. ԜԼԔӾӪԽӰӹԙԽӸ. ԹӹԐԐӫӰӠӪԂԽӠӴӰԂԋ ԗӾӢԙԗӾӪԽӠӴԂ ӴՂԙӪԂԒՂӴԂӰӠԋ. ԜԗӢӾԌӾԊӾӰӰӹԙԽӸ/ӰӾӹԗӢӾԌӾԊӾӰӰӹԙԽӸ (ѷѦҸҝѹҾ ӍҏӁҔ҃ѹҸӁҽґҧӁҸҮ ҊҝҽѸѸҽҲѹҔѹ 4), 189231. ҋҽӁҔҲ-ҭѦҲѦҝқҏҝҊ: ѢҽҏҔҽ. 2000

396 ѯҽҊѹӁҽ, O.ғ. 1990 ҹѸѦґѦӊ, ѩ.Ѩ. 1958

CORNELIA E. KEIJSPER

ԹԂӪ ԙӪԂԒԂԽӸ ӠӰԂԓӾ? (ԪԂԼӹԽԂ ӰԂԌ ԙӠӰԽԂӪԙӠԓӾԙӪӹӡ ԙӠӰӹӰӠԐӠӾӡ ӢӫԙԙӪӹԄӹ ԋԒՂӪԂ Ӵ ӠӰӹԙԽӢԂӰӰӹӡ ԂӫԌӠԽӹӢӠӠ). ҴҸҡҔӊҽ: ҩҏҡҡҔѹҮ ҾӉӃҔ. “ҶҔҡҰҝѦҡҡѹӊӁҸ-ѹҝҸӁѹҍѦҡҔҸѦ ӊӃҝҽңѦӁѹѦ ҸҲҝѹ҃ҽӁѹҾ ѹ ҸҲҝѹ҃ҽҲѦґҧӁҸҮ Ҹ҃ѦӁҔѹ ӊ ҡҸӊҝѦѸѦӁӁҸѸ ҝҏҡҡҔҸѸ ҾӉӃҔѦ”. ԛӹԗӢӹԙՂ ԋԒՂӪӹԒӰԂӰӠԋ 6, 63-75.

SUMMARY This article discusses the linguistic issues involved in so-called OVS sentences (sentences with the word order Object-Verb-Subject). These issues have been raised ever since general linguistic theories (e.g. Greenberg, Chomsky) found OVS sentences problematic. On the basis of Slavic examples from the Amsterdam Slavic Parallel Aligned Corpus it is shown that the presumed problems do not exist in Slavic languages. Instead, a whole series of questions involving socalled converse structure awaits further investigation.

Dutch Contributions to the Fourteenth International Congress of Slavists, Ohrid: Linguistics (SSGL 34). Amsterdam – New York: Rodopi, 2008, 397-429.

SLAVIC HISTORICAL MORPHOLOGY: NOMINAL PARADIGMS FREDERIK KORTLANDT

Elsewhere I have presented my relative chronology of Slavic phonological developments (1989, 2003a, 2006) and compared it with the chronologies proposed by Greenberg (2000) and Holzer (2005). Here I intend to examine the consequences for the morphology of the nominal paradigms. The period covered in the following is from the disintegration of the Balto-Slavic protolanguage up to the loss of the jers. The reconstructions reflect my stages 5.0 (end of 4 Late Balto-Slavic), 6.0 (end of 5 Early Slavic), 7.0 (end of 6 Early Middle Slavic), 8.0 (end of 7 Late Middle Slavic), 9.0 (end of 8 Young Proto-Slavic), and 10.0 (end of 9 Late Proto-Slavic). The main dialect chosen is that of Ohrid. For the later stages I also give dialectal variants which I reconstruct for Graz (i.e. the dialect of the Freising Fragments, cf. Kortlandt 2003a: 234), Kraków and Novgorod. The paradigms chosen reflect masculines, feminines and neuters, hard and soft a- and o-stems, u-, i- and consonantal stems, and accent classes (a), (b) and (c). The examples are: krava (a) ‘cow’, žena (b) ‘woman’, glava (c) ‘head’, osnova (a) ‘base’, volja (a) ‘will’, ovӸca (c) ‘sheep’, dymԔ (a) ‘smoke’, konjӸ (b) ‘horse’, z՛bԔ (c) ‘tooth’, lLJto (a) ‘year’, igo (c) ‘yoke’, polje (c) ‘field’, synԔ (c) ‘son’, p՛tӸ (b) ‘way’, kostӸ (c) ‘bone’, dӸnӸ (c) ‘day’, imNj (c) ‘name’ and dLJtNj (c) ‘child’. As the word konjӸ evidently did not exist in Balto-Slavic times, I have substituted the equivalent of equus at the oldest stage; for the other words I have simply reconstructed the expected proto-form. The initial (final Balto-Slavic) phonological system is the following: p t ǀ k ‫ٳ‬

b d ՝փ g

s

m n

i e

ǔ lj

l

r

j

w u o

a

ƽ

ǰ Ǣ

398

FREDERIK KORTLANDT

5.0 Balto-Slavic nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

kòr‫ٳ‬we‫ٳ‬ kòr‫ٳ‬wljs kòr‫ٳ‬wlji kòr‫ٳ‬wljn kòr‫ٳ‬we‫ٳ‬ kòr‫ٳ‬we‫ٳ‬i kòr‫ٳ‬wljs kuru‫ٳ‬ùn kurù‫ٳ‬mus kòr‫ٳ‬we‫ٳ‬ns kurù‫ٳ‬mi‫ٳ‬s kurù‫ٳ‬su

gèna‫ٳ‬ gènƽs gènƽi gènƽn gèna‫ٳ‬ gèna‫ٳ‬i gènƽs gènun gèna‫ٳ‬mus gèna‫ٳ‬ns gèna‫ٳ‬mi‫ٳ‬s gèna‫ٳ‬su

gol‫ٳ‬wà‫ٳ‬ gol‫ٳ‬wás gòl‫ٳ‬wƽi gòl‫ٳ‬wƽn gol‫ٳ‬wà‫ٳ‬ gol‫ٳ‬wà‫ٳ‬i gòl‫ٳ‬wƽs gol‫ٳ‬wùn gol‫ٳ‬wà‫ٳ‬mus gòl‫ٳ‬wa‫ٳ‬ns gol‫ٳ‬wà‫ٳ‬mi‫ٳ‬s gol‫ٳ‬wà‫ٳ‬su

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

dù‫ٳ‬mos dù‫ٳ‬mǢ dù‫ٳ‬mǢi dù‫ٳ‬mun dù‫ٳ‬mo‫ٳ‬ dù‫ٳ‬moi dù‫ٳ‬moi dù‫ٳ‬mun dù‫ٳ‬momus dù‫ٳ‬mons dù‫ٳ‬mǢis dù‫ٳ‬moisu

èǀwos èǀwǢ èǀwǢi èǀwun èǀwo‫ٳ‬ èǀwoi èǀwoi èǀwun èǀwomus èǀwons èǀwǢis èǀwoisu

՝փòmbos ՝փòmbǢ ՝փòmbǢi ՝փòmbun ՝փòmbo‫ٳ‬ ՝փòmboi ՝փòmboi ՝փòmbun ՝փòmbomus ՝փòmbons ՝փòmbǢis ՝փòmboisu

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

sù‫ٳ‬nus sù‫ٳ‬nous sù‫ٳ‬nowei sù‫ٳ‬nun su‫ٳ‬numì sù‫ٳ‬nlju sù‫ٳ‬nowes sù‫ٳ‬nun su‫ٳ‬numùs sù‫ٳ‬nuns su‫ٳ‬numì‫ٳ‬s su‫ٳ‬nusù

pònte‫ٳ‬s pintès pìntei pòntljn pintè‫ٳ‬ pìnte‫ٳ‬i pòntljs pintùn pintmùs pònte‫ٳ‬ns pintmì‫ٳ‬s pintsù

kòsti kostèis kòsteiei kòsti kostimì kòstlji kostèi‫ٳ‬ kostiùn kostimùs kostèi‫ٳ‬ kostimì‫ٳ‬s kostisù

SLAVIC HISTORICAL MORPHOLOGY: NOMINAL PARADIGMS

5.0 Balto-Slavic nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

òpsnowa‫ٳ‬ òpsnowƽs òpsnowƽi òpsnowƽn òpsnowa‫ٳ‬ òpsnowa‫ٳ‬i òpsnowƽs òpsnowun òpsnowa‫ٳ‬mus òpsnowa‫ٳ‬ns òpsnowa‫ٳ‬mi‫ٳ‬s òpsnowa‫ٳ‬su

wòlei‫ٳ‬ wòljljs wòljlji wòlein wòlje‫ٳ‬ wòlje‫ٳ‬i wòljljs wòljun wòlje‫ٳ‬mus wòlje‫ٳ‬ns wòlje‫ٳ‬mi‫ٳ‬s wòlje‫ٳ‬su

owikà‫ٳ‬ owikás òwikƽi òwikƽn owikà‫ٳ‬ owikà‫ٳ‬i òwikƽs owikùn owikà‫ٳ‬mus òwika‫ٳ‬ns owikà‫ٳ‬mi‫ٳ‬s owikà‫ٳ‬su

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

lè‫ٳ‬to lè‫ٳ‬tǢ lè‫ٳ‬tǢi lè‫ٳ‬to lè‫ٳ‬to‫ٳ‬ lè‫ٳ‬toi lè‫ٳ‬ta‫ٳ‬ lè‫ٳ‬tun lè‫ٳ‬tomus lè‫ٳ‬ta‫ٳ‬ lè‫ٳ‬tǢis lè‫ٳ‬toisu

jù‫ٳ‬go jù‫ٳ‬gǢ jù‫ٳ‬gǢi jù‫ٳ‬go ju‫ٳ‬gò‫ٳ‬ jù‫ٳ‬goi ju‫ٳ‬gà‫ٳ‬ ju‫ٳ‬gùn ju‫ٳ‬gomùs ju‫ٳ‬gà‫ٳ‬ ju‫ٳ‬góis ju‫ٳ‬goisù

pòljo pòljǢ pòljǢi pòljo poljò‫ٳ‬ pòljoi poljà‫ٳ‬ poljùn poljomùs poljà‫ٳ‬ poljóis poljoisù

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

dèin dinès dìnei dèinin dinmì dèini dìnes dinùn dinmùs dìnins dinmì‫ٳ‬s dinsù

ìn‫ٳ‬men in‫ٳ‬menès ìn‫ٳ‬menei ìn‫ٳ‬men in‫ٳ‬menmì ìn‫ٳ‬meni ìn‫ٳ‬men‫ٳ‬ in‫ٳ‬menùn in‫ٳ‬menmùs ìn‫ٳ‬men‫ٳ‬ in‫ٳ‬menmì‫ٳ‬s in‫ٳ‬mensù

dè‫ٳ‬ten de‫ٳ‬tentès dè‫ٳ‬tentei dè‫ٳ‬ten de‫ٳ‬tentmì dè‫ٳ‬tenti dè‫ٳ‬tei‫ٳ‬ de‫ٳ‬tiùn de‫ٳ‬timùs dè‫ٳ‬tei‫ٳ‬ de‫ٳ‬timì‫ٳ‬s de‫ٳ‬tisù

399

400

FREDERIK KORTLANDT

6.0 Early Slavic nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

kàr‫ٳ‬we‫ٳ‬ kàr‫ٳ‬wljh kàr‫ٳ‬wei kàr‫ٳ‬weN kàr‫ٳ‬wejoN kàr‫ٳ‬we‫ٳ‬i kàr‫ٳ‬wljh kàr‫ٳ‬wu kàr‫ٳ‬we‫ٳ‬muh kàr‫ٳ‬wuN kàr‫ٳ‬we‫ٳ‬mǔh kàr‫ٳ‬we‫ٳ‬su

gèna‫ٳ‬ gènƽh gènai gènoN gènajoN gèna‫ٳ‬i gènƽh gènu gèna‫ٳ‬muh gènuN gèna‫ٳ‬mǔh gèna‫ٳ‬su

galwà‫ٳ‬ galwáh gàlwai gàlwoN galwàjoN galwà‫ٳ‬i gàlwƽh galwù galwà‫ٳ‬muh gàlwuN galwà‫ٳ‬mǔh galwà‫ٳ‬su

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

dù‫ٳ‬mah dù‫ٳ‬mƽ dù‫ٳ‬mau dù‫ٳ‬mu dù‫ٳ‬ma‫ٳ‬ dù‫ٳ‬mai dù‫ٳ‬muih dù‫ٳ‬mu dù‫ٳ‬mamuh dù‫ٳ‬muN dù‫ٳ‬mǰh dù‫ٳ‬maixu

kàbnjah kàbnjƽ kàbnjau kàbnju kàbnja‫ٳ‬ kàbnjai kàbnjuih kàbnju kàbnjamuh kàbnjuN kàbnjǰh kàbnjaixu

zòNbah zòNbƽ zòNbau zòNbu zòNba‫ٳ‬ zòNbai zòNbuih zòNbu zòNbamuh zòNbuN zòNbǰh zòNbaixu

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

súnuh súnauh súnawei súnu sǰnumì súnau súnaweh súnu sǰnumùh súnǰh sǰnumíh sǰnuxù

pòNtih pòNteih pòNtei pòNti pòNtimi pòNte‫ٳ‬i pòNteieh pòNtiu pòNtimuh pòNtǔh pòNtimǔh pòNtixu

kàsti kastèih kàstei kàsti kastimì kàstei kastèi‫ٳ‬ kastiù kastimùh kastèi‫ٳ‬ kastimíh kastixù

SLAVIC HISTORICAL MORPHOLOGY: NOMINAL PARADIGMS

6.0 Early Slavic nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

àpsnawƽ àpsnawƽh àpsnawai àpsnawoN àpsnawajoN àpsnawai àpsnawƽh àpsnawu àpsnawƽmuh àpsnawuN àpsnawƽmǔh àpsnawƽsu

wàli‫ٳ‬ wàljljh wàljei wàlǔ wàljejoN wàlje‫ٳ‬i wàljljh wàlju wàlje‫ٳ‬muh wàljuN wàlje‫ٳ‬mǔh wàlje‫ٳ‬su

awikà‫ٳ‬ awikáh àwikai àwikoN awikàjoN awikà‫ٳ‬i àwikƽh awikù awikà‫ٳ‬muh àwikuN awikà‫ٳ‬mǔh awikà‫ٳ‬su

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

lè‫ٳ‬ta lè‫ٳ‬tƽ lè‫ٳ‬tau lè‫ٳ‬ta lè‫ٳ‬ta‫ٳ‬ lè‫ٳ‬tai lè‫ٳ‬ta‫ٳ‬ lè‫ٳ‬tu lè‫ٳ‬tamuh lè‫ٳ‬ta‫ٳ‬ lè‫ٳ‬tǰh lè‫ٳ‬taixu

júga júgƽ júgau júga jǰgà‫ٳ‬ júgai jǰgà‫ٳ‬ jǰgù jǰgamùh jǰgà‫ٳ‬ jǰgúh jǰgaixù

pàlja pàljƽ pàljau pàlja paljà‫ٳ‬ pàljai paljà‫ٳ‬ paljù paljamùh paljà‫ٳ‬ paljúh paljaixù

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

dìN dinèh dìnei dìni diNmì dìni dìneh dinù diNmùh dìnǔh diNmíh diNsù

ìmeN imenèh ìmenei ìmeN imeNmì ìmeni imenà‫ٳ‬ imenù imeNmùh imenà‫ٳ‬ imeNmíh imeNsù

déteN dljteNtèh déteNtei déteN dljtentmì déteNti détei‫ٳ‬ dljtiù dljtimùh détei‫ٳ‬ dljtimíh dljtixù

401

402

FREDERIK KORTLANDT

7.0 Early Middle Slavic nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

ka֐rwa֐ ka֐rwƽ ka֐rwlj ka֐rwoN ka֐rwaj‫׀‬N ka֐rwe֐ ka֐rwƽ ka֐rwu ka֐rwa֐mu ka֐rwuN ka֐rwa֐mǔ ka֐rwa֐xu

žֻւna֐ žֻւnƽ žֻւnlj žֻւnoN žֻւnaj‫׀‬N žֻւne֐ žֻւnƽ žֻւnu žֻւna֐mu žֻւnuN žֻւna֐mǔ žֻւna֐xu

galwa֐ galwá gգlwlj gգlwoN galwàj‫׀‬N galwe֐ gգlwƽ galwù galwa֐mu gգlwuN galwa֐mǔ galwa֐xu

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

du֐ma du֐mƽ du֐mǢ du֐mu du֐ma֐ du֐mlj du֐m٤ du֐mu du֐mamu du֐muN du֐mǰ du֐mljxu

kàbnjֻ kàbnjƽ֓ kàbnjǢ֓ kàbnju֓ kàbnjֻ֐ kàbnj‫־‬ֆ kàbnj٤ kàbnju֓ kàbnjֻmu kàbnju֓N kàbnj٤ kàbnj‫־‬ֆxu

zoNbà zթNbƽ zթNbǢ zթNbu zoNba֐ zթNblj zթNb٤ zoNbù zoNbamù zթNbuN zoNbú zoNbljxù

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

sխnu sǰnó sխnaw‫־‬ֆ sխnu sǰnumì sǰnó sխnawe sǰnowù sǰnumù sխnǰ sǰnumí sǰnuxù

póNti póNt‫־‬ֆ póNt‫־‬ֆ póNti póNtimi póNt‫־‬ֆ póNtejֻ póNtiju֓ póNtimu póNtǔ póNtimǔ póNtixu

kբsti kast‫־‬փ kբst‫־‬ֆ kբsti kastimì kast‫־‬փ kast‫֐־‬ kastijù֓ kastimù kast‫֐־‬ kastimí kastixù

SLAVIC HISTORICAL MORPHOLOGY: NOMINAL PARADIGMS

7.0 Early Middle Slavic nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

àpsnawƽ àpsnawƽ àpsnawlj àpsnawoN àpsnawaj‫׀‬N àpsnawlj àpsnawƽ àpsnawu àpsnawƽmu àpsnawuN àpsnawƽmǔ àpsnawƽxu

wàlı֐ wàljƽ֓ wàlj‫־‬ֆ wàlj‫׀‬N wàljֻj‫׀‬N wàlj‫֐־‬ wàljƽ֓ wàlju֓ wàljֻ֐mu wàlju֓N wàljֻ֐mǔ wàljֻ֐xu

awiǀֻ֐ awiǀá֓ բwiǀlj բwiǀ‫׀‬N awiǀà֓j‫׀‬N awiǀe֐ բwiǀƽ֓ awikù awiǀֻ֐mu բwikuN awiǀֻ֐mǔ awiǀֻ֐xu

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

le֐ta le֐tƽ le֐tǢ le֐ta le֐ta֐ le֐tlj le֐ta֐ le֐tu le֐tamu le֐ta֐ le֐tǰ le֐tljxu

jխ֓ga jխ֓gƽ jխ֓gǢ jխ֓ga jǰ֓ga֐ jխ֓՝փlj jǰ֓ga֐ jǰ֓gù jǰ֓gamù jǰ֓ga֐ jǰ֓gú jǰ֓՝փljxù

pբljֻ pբljƽ֓ pբljǢ֓ pբljֻ paljֻ֐ pբlj‫־‬ֆ paljֻ֐ paljù֓ paljֻmù paljֻ֐ palj΋ palj‫־‬ֆxù

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

dզni dinè dզn‫־‬ֆ dզni dinimì dinè dզne dinù dinimù dզnǔ dinimí dinixù

զmeN imenè զmen‫־‬ֆ զmeN imeNmì imenè imena֐ imenù imeNmù imena֐ imeNmí imeNsù

dեteN dljteNtè dեteNt‫־‬ֆ dեteN dljteNtimì dljteNtè dեt‫֐־‬ dljtijù֓ dljtimù dեt‫֐־‬ dljtimí dljtixù

403

404

FREDERIK KORTLANDT

8.0 Late Middle Slavic (Ohrid) nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

kra֐wa kra֐wy kra֐wՙ kra֐woN kra֐woøN kra֐wæ kra֐wy kra֐wԔ kra֐wamԔ kra֐wy kra֐wamǔ kra֐waxԔ

žèna žèny žènՙ žènoN žènoøN žènæ žèny žènԔ žènamԔ žèny žènamǔ žènaxԔ

glawa֐ glawǹ glգwՙ glգwoN glawòøN glawæ֐ glգwy glawԔւ glawa֐mԔ glգwy glawa֐mǔ glawa֐xԔ

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

dy֐mo dy֐mƽ dy֐mǰ dy֐mԔ dy֐momӸ dy֐mՙ dy֐mǔ dy֐mԔ dy֐momԔ dy֐my dy֐mձ dy֐mՙxԔ

kòǟlj kòǟՙ kòǟ٤ kòǟӸ kòǟljmӸ kòǟǔ kòǟǔ kòǟӸ kòǟljmԔ kòǟeN kòǟǔ kòǟǔxԔ

zoNbò zթNbƽ zթNbǰ zթNbԔ zթNbomӸ zթNbՙ zթNbǔ zoNbԔւ zoNbomԔւ zթNby zoNbý zoNbæxԔւ

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

sy֏nԔ synú sy֏nowǔ sy֏nԔ synԔmӸւ synú sy֏nowe synowԔւ synԔmԔւ sy֏ny synԔmí synԔxԔւ

póNtӸ póNtǔ póNtǔ póNtӸ póNtӸmӸ póNtǔ póNtӸe póNtiӸ póNtӸmԔ póNti póNtӸmǔ póNtӸxԔ

kըstӸ kostí kըstǔ kըstӸ kostӸmӸւ kostí kըsti kostiӸւ kostӸmԔւ kըsti kostӸmí kostӸxԔւ

SLAVIC HISTORICAL MORPHOLOGY: NOMINAL PARADIGMS

8.0 Late Middle Slavic (Ohrid) nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

òsnowƽ òsnowձ òsnowՙ òsnowoN òsnowoøN òsnowՙ òsnowձ òsnowԔ òsnowƽmԔ òsnowձ òsnowƽmǔ òsnowƽxԔ

wòǛՙ wòǛeN wòǛǔ wòǛøN wòǛeøN wòǛǔ wòǛeN wòǛӸ wòǛՙmԔ wòǛeN wòǛՙmǔ wòǛՙxԔ

owӸcæ֐ owӸcéN ըwӸcǔ ըwӸcøN owӸcèøN owӸcı֐ ըwӸceN owӸcӸւ owӸcæ֐mԔ ըwӸceN owӸcæ֐mǔ owӸcæ֐xԔ

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

læ֐to læ֐tƽ læ֐tǰ læ֐to læ֐tomӸ læ֐tՙ læ֐ta læ֐tԔ læ֐tomԔ læ֐ta læ֐tձ læ֐tՙxԔ

էgo էgƽ էgǰ էgo էgomӸ է՝ՙ iga֐ igԔւ igomԔւ iga֐ igý i՝æxԔւ

pըǛlj pըǛՙ pըǛ٤ pըǛlj pըǛljmӸ pըǛǔ poǛæ֐ poǛӸւ poǛemԔւ poǛæ֐ poǛí poǛixԔւ

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

dӸ֎nӸ dӸnè dӸ֎nǔ dӸ֎nӸ dӸnӸmӸւ dӸnè dӸ֎ne dӸnԔւ dӸnӸmԔւ dӸ֎ni dӸnӸmí dӸnӸxԔւ

Ӹ֎meN Ӹmenè Ӹ֎menǔ Ӹ֎meN ӸmenӸmӸւ Ӹmenè Ӹmena֐ ӸmenԔւ ӸmenӸmԔւ Ӹmena֐ ӸmenӸmí ӸmenӸxԔւ

dæ֏teN dæteNtè dæ֏teNtǔ dæ֏teN dæteNtӸmӸւ dæteNtè dæ֏ti dætiӸւ dætӸmԔւ dæ֏ti dætӸmí dætӸxԔւ

405

406

FREDERIK KORTLANDT

9.0 Young Proto-Slavic (Ohrid) nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

kra֐wa kra֐wy kra֐wæ kra֐woN kra֐wǢN kra֐wæ kra֐wy kra֐wԔ kra֐wamԔ kra֐wy kra֐wamǔ kra֐waxԔ

ženà žւenǹ ženæւ ženòN ženթN ženæւ ženǹ žènԔ ženàmԔ ženǹ ženàmǔ ženàxԔ

glawa֐ glawǹ glգwæ glգwoN glawoøւN glawæ֐ glգwy gláwԔ glawa֐mԔ glգwy glawa֐mǔ glawa֐xԔ

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

dy֐mԔ dy֐ma dy֐mu dy֐mԔ dy֐momӸ dy֐mæ dy֐mi dy֐mԔ dy֐momԔ dy֐my dy֐mձ dy֐mՙxԔ

kòǟӸ koǟà koǟΏ kòǟӸ koǟեmӸ koǟէ koǟì kòǟӸ koǟեmԔ koǟèN koǟէ koǟէxԔ

zթNbԔ zթNba zթNbu zթNbԔ zթNbomӸ zթNbæ zթNbi zóNbԔ zoNbómԔ zթNby zoNbý zoNbƿxԔ

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

sy֏nԔ synú sy֏nowi sy֏nԔ sy֏nԔmӸ synú sy֏nowe synówԔ sy֏nԔmԔ sy֏ny synԔmí sy֏nԔxԔ

póNtӸ pǢNtէ pǢNtì póNtӸ pǢNtӸւmӸ pǢNtէ pǢNtӸւe pǢNtíӸ pǢNtӸւmԔ pǢNtì pǢNtӸւmǔ pǢNtӸւxԔ

kթstӸ kostí kըsti kթstӸ kostӸøւN kostí kըsti kostíӸ kըstӸmԔ kըsti kostӸmí kըstӸxԔ

SLAVIC HISTORICAL MORPHOLOGY: NOMINAL PARADIGMS

9.0 Young Proto-Slavic (Ohrid) nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

osnòwƽ osnòwy osnòwæ osnòwoN osnòwǢN osnòwՙ osnòwy osnòwԔ osnòwƽmԔ osnòwy osnòwƽmǔ osnòwƽxԔ

woǛգ woǛեN woǛì woǛøւN woǛø֏N woǛէ woǛèN wòǛӸ woǛգmԔ woǛèN woǛգmǔ woǛգxԔ

owӸca֐ owӸcéN ըwӸci ըwӸcøN owӸceøւN owӸcı֐ ըwӸceN ówӸcӸ owӸca֐mԔ ըwӸceN owӸca֐mǔ owӸca֐xԔ

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

læ֐to læ֐ta læ֐tu læ֐to læ֐tomӸ læ֐tæ læ֐ta læ֐tԔ læ֐tomԔ læ֐ta læ֐tձ læ֐tՙxԔ

էgo էga էgu էgo էgomӸ է՝æ iga֐ ígԔ igómԔ iga֐ igý i՝ƿxԔ

pըǛe pըǛa pըǛü pըǛe pըǛljmӸ pըǛi poǛa֐ póǛӸ poǛémԔ poǛa֐ poǛí poǛíxԔ

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

dӸ֏nӸ dӸnè dӸ֎ni dӸ֏nӸ dӸ֎nӸmӸ dӸnè dӸ֎ne dӸփnԔ dӸ֎nӸmԔ dӸ֎ni dӸnӸmí dӸ֎nӸxԔ

Ӹ֎meN Ӹmenè Ӹ֎meni Ӹ֎meN Ӹ֎menӸmӸ Ӹmenè Ӹmena֐ ӸménԔ ӸménӸmԔ Ӹmena֐ ӸmenӸmí ӸménӸxԔ

dæ֏teN dæteNtè dæ֏teNti dæ֏teN dæ֏teNtӸmӸ dæteNtè dæ֏ti dætíӸ dæ֏tӸmԔ dæ֏ti dætӸmí dæ֏tӸxԔ

407

408

FREDERIK KORTLANDT

10.0 Late Proto-Slavic (Ohrid) nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

kràwa kràwy kràwæ kràwoN kràwojoN kràwæ kràwy kràwԔ kràwamԔ kràwy kràwami kràwaxԔ

ženà ženǹ ženæւ ženòN ženòjoN ženæւ ženǹ žènԔ ženàmԔ ženǹ ženàmi ženàxԔ

glawà glawǹ glգwæ glգwoN glawojòN glawæւ glգwy gláwԔ glawàmԔ glգwy glawàmi glawàxԔ

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

dǹmԔ dǹma dǹmu dǹmԔ dǹmomӸ dǹmæ dǹmi dǹmԔ dǹmomԔ dǹmy dǹmy dǹmæxԔ

kòǟӸ koǟà koǟù kòǟӸ koǟèmӸ kòǟi koǟì kóǟӸ koǟèmԔ koǟèN kòǟi kòǟixԔ

zթNbԔ zթNba zթNbu zթNbԔ zթNbomӸ zթNbæ zթNbi zóNbԔ zoNbómԔ zթNby zoNbý zoNbƿxԔ

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

sy֏nԔ sy֏nu sy֏nowi sy֏nԔ sy֏nԔmӸ synú sy֏nowe synówԔ sy֏nԔmԔ sy֏ny synԔmí sy֏nԔxԔ

póNtӸ poNtì poNtì póNtӸ poNtӸւmӸ poNtì poNtӸւje poNtìjӸ poNtӸւmԔ poNtì poNtӸւmi poNtӸւxԔ

kթstӸ kըsti kըsti kթstӸ kostӸjòN kostí kըsti kostìjӸ kըstӸmԔ kըsti kostӸmí kըstӸxԔ

SLAVIC HISTORICAL MORPHOLOGY: NOMINAL PARADIGMS

10.0 Late Proto-Slavic (Ohrid) nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

osnòwa osnòwy osnòwæ osnòwoN osnòwojoN osnòwæ osnòwy osnòwԔ osnòwamԔ osnòwy osnòwami osnòwaxԔ

wòǛa wòǛeN wòǛi wòǛoN wòǛejoN wòǛi wòǛeN wòǛӸ wòǛamԔ wòǛeN wòǛami wòǛaxԔ

owӸcà owӸcéN ըwӸci ըwӸcoN owӸcejòN owӸcì ըwӸceN ówӸcӸ owӸcàmԔ ըwӸceN owӸcàmi owӸcàxԔ

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

læւto læւta læւtu læւto læւtomӸ læւtæ læւta læւtԔ læւtomԔ læւta læւty læւtæxԔ

էgo էga էgu էgo էgomӸ է՝æ igà ígԔ igómԔ igà igý i՝ƿxԔ

pըǛe pըǛa pըǛu pըǛe pըǛemӸ pըǛi poǛà póǛӸ poǛémԔ poǛà poǛí poǛíxԔ

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

dӸ֏nӸ dӸnè dӸ֎ni dӸ֏nӸ dӸ֎nӸmӸ dӸnè dӸ֎ne dӸփnԔ dӸ֎nӸmԔ dӸ֎ni dӸnӸmí dӸ֎nӸxԔ

զmeN imenè զmeni զmeN զmenӸmӸ imenè imenà iménԔ iménӸmԔ imenà imenӸmí iménӸxԔ

dæ֏teN dæteNtè dæ֏teNti dæ֏teN dæ֏teNtӸmӸ dæteNtè dæ֏ti dætìjӸ dæ֏tӸmԔ dæ֏ti dætӸmí dæ֏tӸxԔ

409

410

FREDERIK KORTLANDT

8.0 Late Middle Slavic (Graz, Kraków, Novgorod) nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

kra֐wa kra֐wy kra֐wՙ kra֐woN kra֐woøN kra֐wæ kra֐wy kra֐wԔ kra֐wamԔ kra֐wy kra֐wamǔ kra֐waxԔ

kro֐wa kro֐wy kro֐wՙ kro֐woN kro֐woøN kro֐wæ kro֐wy kro֐wԔ kro֐wamԔ kro֐wy kro֐wamǔ kro֐waxԔ

ko֐rwa ko֐rwy ko֐rwՙ ko֐rwoN ko֐rwoøN ko֐rwæ ko֐rwy ko֐rwԔ ko֐rwamԔ ko֐rwy ko֐rwamǔ ko֐rwaxԔ

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

kòǟlj kòǟՙ kòǟ٤ kòǟӸ kòǟljmӸ kòǟǔ kòǟǔ kòǟӸ kòǟljmԔ kòǟeN kòǟǔ kòǟǔxԔ

kòǟlj kòǟՙ kòǟ٤ kòǟӸ kòǟՙ kòǟǔ kòǟǔ kòǟӸ kòǟljmԔ kòǟ‫־‬N kòǟǔ kòǟǔxԔ

kòǟlj kòǟՙ kòǟ٤ kòǟӸ kòǟՙ kòǟǔ kòǟǔ kòǟӸ kòǟljmԔ kòǟ‫־‬N kòǟǔ kòǟǔxԔ

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

póNtӸ póNtǔ póNtǔ póNtӸ póNtӸmӸ póNtǔ póNtӸe póNtiӸ póNtӸmԔ póNti póNtӸmǔ póNtӸxԔ

póNtӸ póNtǔ póNtǔ póNtӸ póNtӸmӸ póNtǔ póNtӸe póNtiӸ póNtӸmԔ póNti póNtӸmǔ póNtӸxԔ

póNtӸ póNtǔ póNtǔ póNtӸ póNtӸmӸ póNtǔ póNtӸe póNtiӸ póNtӸmԔ póNti póNtӸmǔ póNtӸxԔ

SLAVIC HISTORICAL MORPHOLOGY: NOMINAL PARADIGMS

8.0 Late Middle Slavic (Graz, Kraków, Novgorod) nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

wòǛՙ wòǛeN wòǛǔ wòǛøN wòǛeøN wòǛǔ wòǛeN wòǛӸ wòǛՙmԔ wòǛeN wòǛՙmǔ wòǛՙxԔ

wòǛՙ wòǛ‫־‬N wòǛǔ wòǛøN wòǛeøN wòǛǔ wòǛ‫־‬N wòǛӸ wòǛՙmԔ wòǛ‫־‬N wòǛՙmǔ wòǛՙxԔ

wòǛՙ wòǛ‫־‬N wòǛǔ wòǛøN wòǛeøN wòǛǔ wòǛ‫־‬N wòǛӸ wòǛՙmԔ wòǛ‫־‬N wòǛՙmǔ wòǛՙxԔ

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

էճo էճƽ էճǰ էճo էճomӸ էzՙ iճa֐ iճԔւ iճomԔւ iճa֐ iճý izæxԔւ

էgo էgƽ էgǰ էgo iga֐ է՝ՙ iga֐ igԔւ igomԔւ iga֐ igý i՝æxԔւ

էgo էgƽ էgǰ էgo iga֐ էgՙ iga֐ igԔւ igomԔւ iga֐ igý igæxԔւ

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

dæ֏teN dæteNtè dæ֏teNtǔ dæ֏teN dæteNtӸmӸւ dæteNtè dæ֏ti dætiӸւ dætӸmԔւ dæ֏ti dætӸmí dætӸxԔւ

dæ֏tæN dætæNtè dæ֏tæNtǔ dæ֏tæN dætæNtӸmӸւ dætæNtè dæ֏ti dætiӸւ dætӸmԔւ dæ֏ti dætӸmí dætӸxԔւ

dæ֏tæN dætæNtè dæ֏tæNtǔ dæ֏tæN dætæNtӸmӸւ dætæNtè dæ֏ti dætiӸւ dætӸmԔւ dæ֏ti dætӸmí dætӸxԔւ

411

412

FREDERIK KORTLANDT

9.0 Young Proto-Slavic (Graz, Kraków, Novgorod) nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

kra֐wa kra֐wy kra֐w‫־‬ kra֐woN kra֐wǢN kra֐w‫־‬ kra֐wy kra֐wԔ kra֐wamԔ kra֐wy kra֐wamǔ kra֐waxԔ

kro֐wa kro֐wy kro֐wæ kro֐woN kro֐wǢN kro֐wæ kro֐wy kro֐wԔ kro֐wamԔ kro֐wy kro֐wamǔ kro֐waxԔ

ko֐rwa ko֐rwy ko֐rw‫־‬ ko֐rwoN ko֐rwǢN ko֐rw‫־‬ ko֐rwy ko֐rwԔ ko֐rwamԔ ko֐rwy ko֐rwamǔ ko֐rwaxԔ

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

kòǟӸ koǟà koǟΏ kòǟӸ koǟեmӸ koǟէ koǟì kòǟӸ koǟեmԔ koǟèN koǟէ koǟէxԔ

kòǟӸ koǟà koǟΏ kòǟӸ koǟӸւmӸ koǟէ koǟì kòǟӸ koǟեmԔ koǟ‫־‬ւN koǟէ koǟէxԔ

koǟè koǟà koǟΏ kòǟӸ koǟӸւmӸ koǟì koǟì kòǟӸ koǟեmԔ koǟ‫־‬ւN koǟէ koǟէxԔ

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

póNtӸ pǢNtէ pǢNtì póNtӸ pǢNtӸւmӸ pǢNtէ pǢNtӸւe pǢNtíӸ pǢNtӸւmԔ pǢNtì pǢNtӸւmǔ pǢNtӸւxԔ

póNtӸ pǢNtì pǢNtì póNtӸ pǢNtӸւmӸ pǢNtէ pǢNtӸւe pǢNtíӸ pǢNtӸւmԔ pǢNtì pǢNtӸւmǔ pǢNtӸւxԔ

póNtӸ pǢNtì pǢNtì póNtӸ pǢNtӸւmӸ pǢNtէ pǢNtӸւje pǢNtӸւjӸ pǢNtӸւmԔ pǢNtì pǢNtӸւmǔ pǢNtӸւxԔ

SLAVIC HISTORICAL MORPHOLOGY: NOMINAL PARADIGMS

9.0 Young Proto-Slavic (Graz, Kraków, Novgorod) nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

woǛգ woǛեN woǛì woǛøւN woǛø֏N woǛէ woǛèN wòǛӸ woǛգmԔ woǛèN woǛգmǔ woǛգxԔ

woǛգ woǛ‫֏־‬N woǛì woǛøւN woǛø֏N woǛէ woǛ‫־‬ւN wòǛӸ woǛգmԔ woǛ‫־‬ւN woǛգmǔ woǛգxԔ

woǛգ woǛ‫֏־‬N woǛì woǛøւN woǛø֏N woǛէ woǛ‫־‬ւN wòǛӸ woǛգmԔ woǛ‫־‬ւN woǛգmǔ woǛգxԔ

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

էճo էճa էճu էճo էճomӸ էz‫־‬ iճa֐ íճԔ iճómԔ iճa֐ iճý iz‫־‬փxԔ

էgo էga էgu էgo էgԔmӸ է՝æ iga֐ ígԔ igómԔ iga֐ igý i՝ƿxԔ

էgo էga էgu էgo էgԔmӸ էg‫־‬ iga֐ ígԔ igómԔ iga֐ igý ig‫־‬փxԔ

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

d‫֏־‬teN d‫־‬teNtè d‫֏־‬teNti d‫֏־‬teN d‫֏־‬teNtӸmӸ d‫־‬teNtè d‫֏־‬ti d‫־‬tíӸ d‫֏־‬tӸmԔ d‫֏־‬ti d‫־‬tӸmí d‫֏־‬tӸxԔ

dæ֏tæN dætæNtè dæ֏tæNti dæ֏tæN dæ֏tæNtӸmӸ dætæNtè dæ֏ti dætíӸ dæ֏tӸmԔ dæ֏ti dætӸmí dæ֏tӸxԔ

d‫֏־‬tæN d‫־‬tæNtè d‫֏־‬tæNti d‫֏־‬tæN d‫֏־‬tæNtӸmӸ d‫־‬tæNtè d‫֏־‬ti d‫־‬tӸւjӸ d‫֏־‬tӸmԔ d‫֏־‬ti d‫־‬tӸmí d‫֏־‬tӸxԔ

413

414

FREDERIK KORTLANDT

10.0 Late Proto-Slavic (Graz, Kraków, Novgorod) nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

kràwa kràwy kràw‫־‬ kràwoN kràwǢN kràw‫־‬ kràwy kràwӸ kràwamӸ kràwy kràwamǔ kràwaxӸ

kròwa kròwy kròw֖æ kròwaN kròwƽN kròw֖æ kròwy kròwӸ kròwamӸ kròwy kròwam֖ǔ kròwaxӸ

korówa korówy korów‫־‬ korówu korówoju korów‫־‬ korówy korówԔ korówamԔ korówy korówami korówaxԔ

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

kòǟӸ koǟà koǟù kòǟӸ koǟèmӸ kòǟi koǟì kóǟӸ koǟèmӸ koǟèN kòǟi kòǟixӸ

kòǟӸ koǟà koǟù kòǟӸ koǟӸւm֖Ӹ kòǟǔ koǟì kóǟӸ koǟèmӸ koǟè kòǟǔ kòǟǔxӸ

koǟé koǟá koǟú kóǟӸ koǟӸւmӸ koǟí koǟí kóǟӸ koǟémԔ koǟé֓ kóǟi kóǟixԔ

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

póNtӸ pǢNtì pǢNtì póNtӸ pǢNtӸւmӸ pǢNtì pǢNtӸւje pǢNtí pǢNtӸւmӸ pǢNtì pǢNtӸւmǔ pǢNtӸւxӸ

páNǭӸ pƽNǭì pƽNǭì páNǭӸ pƽNǭӸւm֖Ӹ pƽNǭì pƽNǭӸւe pƽNǭí pƽNǭӸւmӸ pƽNǭì pƽNǭӸւm֖ǔ pƽNǭӸւxӸ

pútӸ putí putí pútӸ putӸւmӸ putí putӸւje putӸւjӸ putӸւmԔ putí putӸւmi putӸւxԔ

SLAVIC HISTORICAL MORPHOLOGY: NOMINAL PARADIGMS

10.0 Late Proto-Slavic (Graz, Kraków, Novgorod) nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

wòǛa wòǛeN wòǛi wòǛoN wòǛǢN wòǛi wòǛeN wòǛӸ wòǛamӸ wòǛeN wòǛamǔ wòǛaxӸ

wòǛƽ wòǛlj wòǛi wòǛaN wòǛƽN wòǛǔ wòǛe wòǛӸ wòǛƽmӸ wòǛe wòǛƽm֖ǔ wòǛƽxӸ

wóǛa wóǛ‫־‬ wóǛi wóǛu wóǛeju wóǛi wóǛ‫־‬ wóǛӸ wóǛamԔ wóǛ‫־‬ wóǛami wóǛaxԔ

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

զճo զճa զճu զճo զճomӸ զz‫־‬ iճà íճӸ iճómӸ iճà iճý iz‫־‬փxӸ

ìgo ìga ìgu ìgo ìgӸm֖Ӹ ì՝æ Ӹgà ígӸ ӸgómӸ Ӹgà Ӹgý Ӹ՝ƿxӸ

ìgo ìga ìgu ìgo ìgԔmӸ ìg‫־‬ igá ígԔ igómԔ igá igý ig‫־‬փxԔ

nom.sg. gen.sg. dat.sg. acc.sg. inst.sg. loc.sg. nom.pl. gen.pl. dat.pl. acc.pl. inst.pl. loc.pl.

d‫֎־‬teN d‫־‬teNtè d‫֎־‬teNti d‫֎־‬teN d‫֎־‬teNtӸmӸ d‫־‬teNtè d‫֎־‬ti d‫־‬tí d‫֎־‬tӸmӸ d‫֎־‬ti d‫־‬tӸmí d‫֎־‬tӸxӸ

d֖æւǭaN d֖æǭaNǭè d֖æւǭaNǭi d֖æւǭaN d֖æւǭaNǭӸm֖Ӹ d֖æǭaNǭè d֖æւǭi d֖æǭí d֖æւǭӸmӸ d֖æւǭi d֖æǭӸm֖í d֖æւǭӸxӸ

d‫־‬ւtæ d‫־‬tæté d‫־‬ւtæti d‫־‬ւtæ d‫־‬ւtætӸmӸ d‫־‬tæté d‫־‬ւti d‫־‬tӸւjӸ d‫־‬ւtӸmԔ d‫־‬ւti d‫־‬tӸmí d‫־‬ւtӸxԔ

415

416

FREDERIK KORTLANDT

Ru. koróva, SCr. krբva, Lith. kárvLj and OPr. curwis ‘ox’ point to a Balto-Slavic proterodynamic uH1-stem *kòr‫ٳ‬we‫ٳ‬, gen. *kòr‫ٳ‬wljs (cf. Kortlandt 1997: 162), with radical stress except in the oblique plural cases (cf. Kortlandt forthc. a: 02), which had a secondary zero grade *ur on the analogy of the full grade *or (cf. Kortlandt 2007). Ru. vólja, SCr. vըlja and Lith. valià may reflect a proterodynamic iH1-stem *wòlei‫ٳ‬, gen. *wòljljs, like OPr. giwei ‘life’, acc. warein ‘power’, Latv. dzǔve, vare (cf. Kortlandt 1997: 162). Cz. pouǬ and OPr. pintis point to a hysterodynamic H1-stem *pònte‫ٳ‬s, gen. *pintès (cf. Beekes 1995: 181). Cz. den, Lith. dienà and OPr. acc. deinan reflect a hysterodynamic n-stem *d֐èin, gen. *d֐inès, cf. Skt. -dínam. The apophonic alternations were apparently eliminated at an early stage in the a‫ٳ‬- and neuter consonant stems, cf. OPr. genno < *-ƽ, gen. gennas, dat.pl. gennƽmans, which reflects a proterodynamic H2-stem *g֐en‫ٳ‬ < PIE. *g֐wen‫‘ ڞ‬woman’, gen. *g֐na‫ٳ‬s < PIE. *g֐wne‫ڞ‬s (cf. Beekes 1995: 185), and OPr. emmens, with radical zero grade and suffixal full grade (cf. Kortlandt 2000), which reflects an original proterodynamic n-stem *no‫ٳ‬mn, Latin nǢmen < PIE. *‫ڞ‬wne‫ڞ‬wmn ‘name’, gen. *n‫ٳ‬mens, Old Irish anmae < PIE. *‫ڞ‬wn‫ڞ‬wmens (cf. Beekes 1995: 186). Deverbal a‫ٳ‬-stems had initial stress in Balto-Slavic, e.g. *òpsnowa‫ٳ‬ (cf. Kortlandt 1979a: 327f.). Denominal a‫ٳ‬-stems may represent original hysterodynamic paradigms, e.g. *gol‫ٳ‬wà‫ٳ‬, *owikà‫ٳ‬, cf. Arm. glux ‘head’ < *gǢlu- and Lith. avìs ‘sheep’. The dialectal Indo-European and common Balto-Slavic developments which affected the nominal paradigms include the following (cf. Kortlandt 1989: 42-46 and 2005a: 116f., the numbering of which is maintained here). 2.1. The opposition between PIE. fortes and lenes stops was rephonemicized as an opposition of voicedness in all branches of Indo-European except Anatolian and Tocharian. 2.2. PIE. *s was retracted to *ַ after *i, *u, *r and *k in Balto-Slavic, Albanian, Armenian and Indo-Iranian. This development remained subphonemic until the spirantization of *ǀ < PIE. *֦ (see 5.8 below). 2.3. The PIE. palatovelars were depalatalized before resonants unless the latter were followed by a front vowel, e.g. OCS. slovo ‘word’, Gr. ʋʌ˰ʏʤ, but Lith. klausýti ‘to listen’. This development was common to Balto-Slavic and Albanian. 3.1. Loss of PIE. accentual mobility. The final stress of Lith. dukt‫‘ מ‬daughter’ originated at this stage, cf. Gr. ʉʔʄˮʓʈʑ with non-final stress, gen. ʉʔʄʂʓʑ˶ʤ. Final stress was generalized in the non-neuter nom.sg. form of the u-, i- and consonantal stems, but not in the a‫ٳ‬- and e‫ٳ‬- stems. At this stage, *kòsti was still a proterodynamic neuter (cf. Meillet 1921) while for the nom.acc.sg. form of the neuter consonantal stems I assume generalization of the oblique stem form and final stress, e.g. *n‫ٳ‬mèn, *d֐e‫ٳ‬tènt, in view of the accentuation of Sln. vréme and SCr. vrijème ‘weather’, Bulg. also vremé (cf. Stang 1957: 92, Pronk forthc.). The

SLAVIC HISTORICAL MORPHOLOGY: NOMINAL PARADIGMS

417

neuter nom.acc.pl. ending was *-‫ < ٳ‬PIE. *-‫ ڞ‬at this stage (cf. Kortlandt 2005b: 155). 3.2. Pedersen’s law: the stress was retracted from inner syllables in accentually mobile paradigms, e.g. acc.sg. Lith. dùkterǕ ‘daughter’, píemenǕ ‘shepherd’, Gr. ʉʔʄʂʓ˰ʑʂ, ʐʏʊՋ˰ʍʂ. This development affected the dat.sg. and nom.pl. forms of the proterodynamic paradigms and the acc.loc.sg. and nom.acc.pl. forms of the hysterodynamic paradigms, e.g. *sù‫ٳ‬newei, *sù‫ٳ‬newes, *kòsteiei, *nւ‫ٳ‬menei, *nւ‫ٳ‬meni, *d֐è‫ٳ‬tentei, *d֐è‫ٳ‬tenti, *pnւte‫ٳ‬i, *pnւte‫ٳ‬ns, and may have been a phonetic development if there were no other accented inner syllables at this stage (cf. Kortlandt forthc. a: 03-05). 3.3. Barytonesis: the retraction of the stress spread analogically to the o-stems in the case forms where Pedersen’s law applied, e.g. acc.sg. Lith. diֿvƾ ‘god’, cf. Skt. devám, similarly in *dù‫ٳ‬mom, *dù‫ٳ‬moi, *dù‫ٳ‬mons. This development is later reflected in the paradigm of z՛bԔ, which became accentually mobile at a more recent stage (see 6.9 below). 3.4. Oxytonesis: the stress shifted from an inner syllable to the end of the word in paradigms with end-stressed forms. I now think that this development may not have affected nominal paradigms at all because the relevant forms were end-stressed in Proto-Indo-European already (cf. Kortlandt forthc. a: 03). 3.5. The nom.acc.sg. ending of oxytone neuter o-stems *-om was replaced with the corresponding pronominal ending *-od֐. The barytone neuters later merged with the masculines (see 4.5 below). 3.6. Final *-om was raised to *-um, which subsequently yielded *-un, e.g. in acc.sg. and gen.pl. *dù‫ٳ‬mun, *èǀwun, *՝փòmbun (cf. Kortlandt 1978b: 281-290). 3.7. Final *-d֐ was lost, e.g. in the neuter ending *-od֐ and in the gen.sg. ending *-Ǣd֐, as in *lè‫ٳ‬to, *lè‫ٳ‬tǢ, *dù‫ٳ‬mǢ. 4.1. Hirt’s law: the stress was retracted if the vowel of the pretonic syllable was immediately followed by a laryngeal, e.g. Lith. dúona ‘bread’, výras ‘man’, d‫ע‬mai ‘smoke’, Skt. dhƽn‫֣נ‬, vǔrá֣, dhǰmá֣, also Sln. dat.pl. goràm ‘mountains’, loc.pl. goràh, where the stress was retracted from the ending to the vowel before the stem-final laryngeal. Thus, we find retraction in *dù‫ٳ‬mos, *lè‫ٳ‬to, *sù‫ٳ‬nuַ, *nւ‫ٳ‬men, *d֐è‫ٳ‬ten, *-à‫ٳ‬muַ, *-à‫ٳ‬mi‫ַٳ‬, *-à‫ٳ‬su, but not in *su‫ٳ‬numì, *su‫ٳ‬numùַ, *su‫ٳ‬numì‫ַٳ‬, *su‫ٳ‬nuַù. The retraction under discussion was more recent than the substitution of the pronominal ending in the oxytone neuter o-stems (3.5) because neuters with retracted stress did not join the masculine gender, e.g. SCr. jբto ‘flock’, Skt. yƽtám, similarly *lè‫ٳ‬to. 4.2. The syllabic resonants dissolved into a vocalic and a consonantal part, the former of which merged with u after labiovelar stops and with i elsewhere (cf. Kortlandt 2007). The labiovelars subsequently lost their labialization. The loss of the syllabic resonants was more recent than Hirt’s law (4.1) because the stress was retracted in Latv. i‫ך‬gs ‘long’, pi‫ך‬ns ‘full’, SCr. dլg, pլn, Skt. dǔrghá֣, pǰr֭á֣. This development yielded *ìn‫ٳ‬men < *nւ‫ٳ‬men and *dèinin < *d֐èinm.

418

FREDERIK KORTLANDT

4.3. Winter’s law: the PIE. glottalic stops dissolved into a laryngeal and a buccal part (cf. Kortlandt forthc. b). The former merged with the reflex of the PIE. laryngeals and the latter with the reflex of the lenes stops. Winter’s law was more recent than the loss of final *d֐ (3.7) in view of the Slavic neuter pronoun to < *tod֐. It was more recent than Hirt’s law (4.1) because the stress was not retracted in Latv. pê֖ds ‘footstep’ < *ped֐óm, nuôgs ‘naked’ < *nog֐wós. The development yielded *jù‫ٳ‬go < *jùg֐o. 4.4. The stress was retracted from final open syllables of disyllabic word forms unless the preceding syllable was closed by an obstruent. This retraction was more recent than the loss of final *d֐ (3.7), as is clear from Lith. gen.sg. vi‫ך‬ko ‘wolf’. The stress was regularly retracted from final vowels and diphthongs, as in Lith. dat.sg. vi‫ך‬kui ‘wolf’, gálvai ‘head’, but not from syllables which ended in a fricative, a nasal, or a laryngeal, e.g. Lith. gen.sg. aviֿs ‘sheep’, gen.pl. vilkdz֗ ‘wolf’, nom.sg. galvà ‘head’. It follows that word-final nasals and laryngeals were still ordinary consonants at this stage. Thus, we find retraction in *gòl‫ٳ‬wƽi, *jù‫ٳ‬go, *jù‫ٳ‬gǢ, *jù‫ٳ‬gǢi, *jù‫ٳ‬goi, *dìnei, but not in *gol‫ٳ‬wà‫ٳ‬, *gol‫ٳ‬wás, *gol‫ٳ‬wùn, *ju‫ٳ‬gò‫ٳ‬, *ju‫ٳ‬gà‫ٳ‬, *ju‫ٳ‬gùn, *ju‫ٳ‬góiַ, *dinùn. 4.5. The merger of the original barytone neuter o-stems with the masculines in the singular must be dated to the Balto-Slavic period in view of the agreement between Slavic and Old Prussian (cf. Kortlandt 1983: 183). New barytone neuters arose as a result of the retractions at stages 4.1 and 4.4. 5.1-6.0 Early Slavic During this period Slavic developed along similar lines as its West and East Baltic sister languages. 5.1. Raising of *lj and *Ǣ before a final resonant, e.g. OCS. mati ‘mother’, kamy ‘stone’, Lith. mótLj, akmuõ, Gr. Ջ˲ʓʈʑ, ʽʋՋʘʍ. The final resonant was lost after the raising. 5.2. Labialization of *a, *ƽ and merger with *o, *Ǣ. e.g. *gol‫ٳ‬wò‫* < ٳ‬gol‫ٳ‬wà‫ٳ‬, *òpsnowo‫* < ٳ‬òpsnowa‫ٳ‬, inst.pl. *gèno‫ٳ‬mi‫* < ַٳ‬gèna‫ٳ‬mi‫ַٳ‬. 5.3. Loss of the laryngeals in pretonic and post-posttonic syllables with compensatory lengthening of an adjacent vowel, e.g. *golwò‫* < ٳ‬gol‫ٳ‬wò‫ٳ‬, inst.sg. *sǰnumì < *su‫ٳ‬numì, *òpsnowǢ < *òpsnowo‫ٳ‬, inst.pl. *gèno‫ٳ‬mǔַ < *gèno‫ٳ‬mi‫ַٳ‬. The long vowel in the final syllable of the latter words is reflected by the neocircumflex tone of Sln. osnթ֓va < 9 *osnòvƽ, ženգmi < 9 *ženàmǔ, where the middle syllable received the stress as a result of Dybo’s law (see 8.7 and 10.9 below). 5.4. Meillet’s law: on the analogy of the end-stressed forms, the laryngeals were eliminated from the barytone forms of paradigms with mobile stress, e.g. SCr. acc.sg. glգvu < 7 *gգlwoN š before *e, *lj, *i, *ǔ, *j. The velar obstruents had fronted variants before front vowels. When *e, *lj merged with the fronted variants of *a, *ƽ after *j (6.1), the sequences *ke, *klj,

420

FREDERIK KORTLANDT

*ge, *glj, *xe, *xlj were rephonemicized as *ǁֻ, *ǁƽ֓, *՞ֻ, *՞ƽ֓, *šֻ, *šƽ֓, where *ֻ, *ƽ֓ are the archiphonemes of *e, *lj and *a, *ƽ after palatals, e.g. in *՞ֻւna‫* < ٳ‬gèna‫ٳ‬. 6.3. Spirantization of the voiced affricate *՞ > ž, e.g. in *žֻւna‫ٳ‬. This development was blocked by a preceding *z. 6.4. Palatalization of the dental fricatives: *s > š and *z > ž before *j, *ǁ, *՞. 6.5. Monophthongization of diphthongs: *ai > *lj, *ei > *lj֓, *ui > *٤, *au > *Ǣ, e.g. dat.sg. *žֻւnlj, *zòNbǢ, *súnawlj֓, loc.sg. *žֻւne֐, *zòNblj, gen.sg. *súnǢh, *kasté֓h, nom.pl. *zòNb٤h, *kast‫֐־‬, loc.pl. *zòNbljxu. PIE. *eu had changed into *iou in Balto-Slavic times and into *jau at stage 5.12. The occurrence of the diphthong *ui was limited to the position before final *-h, where it had arisen at stage 5.9. After palatal consonants, the diphtongs *ֻi, *u֓i, *ֻu changed into *lj֓, *٤, *Ǣ֓, the latter of which is the phonetically conditioned variant of *Ǣ, e.g. *pàljǢ֓, *pàljlj֓, *paljlj֓xù. The rise of nasal vowels before a tautosyllabic stop can be dated to the same stage. It yielded a new nasal vowel *aN in the participial ending PIE. *-onts, which had been subject to the delabialization at stage 5.12, e.g. ORu. nesa ‘carrying’, cf. nesu < *-oN ‘I carry’. The surviving laryngeals had developed into glottal stops by this time: I shall write *ı֐, *‫֐־‬, *e֐, *a֐, *o֐, *u֐, e.g. *ka֐rwa֐, *žֻւna֐, *žֻւne֐, *kast‫֐־‬, *du֐mah. The glottalized vowels had the timbre of the corresponding long vowels, as in the case of the Latvian broken tone. 6.6. Second palatalization of velars: *k > *ǀ, *g > *՝փ, *x > *ǧ before the new front vowels *lj and *٤ which had arisen from the monophthongization of *ai, *ui (6.5), e.g. *jú֓՝փlj, *jǰ֓՝փljxù, and after the high front vowels *i, *ǔ, *iN unless followed by a consonant or by one of the high back vowels *u, *ǰ, *uN, e.g. *awiǀֻ֐, *àwiǀ‫׀‬N, *awiǀe֐, *awikù. The clusters *sk and *zg became *ǧǀ and *Ǻ՝փ before the new front vowels. The sequences *ika, *iga, *ixa were rephonemicized as *iǀֻ, *i՝փֻ, *iǧֻ, etc., e.g. in the paradigm of *awiǀֻ֐. The development restored the opposition between *lj and *ƽ after palatals, as is clear from OCS. vӸsӸ ‘all’, f.sg. and n.pl. vӸsa, gen.loc.pl. vӸsLJxԔ. Thus, the long vowel *ƽ֓ lost the status of an archiphoneme and came to be the fronted variant of *ƽ after a palatal consonant. 6.7. Rise of geminated affricates: *tj > *tǀ, *dj > *d՝փ, also *stj > *ǧtǀ, *zdj > *Ǻd՝փ. This development has a modern parallel in Ukrainian, e.g. žyttjá ‘life’. The cluster *kt yielded *tǀ before high front vowels, e.g. OCS. noštӸ ‘night’, Ru. noǁ’, SCr. nթǀ. 6.8. Loss of final *-h from *-s, e.g. in *zòNba, *zòNb٤, *zòNbamu. I date its ultimate loss toward the end of the Early Middle Slavic period because it was most probably only slightly earlier than the rise of prothetic glides (7.1). 6.9. Illiǁ-Svityǁ’s law. Accentual mobility was generalized in the masc. o-stems which did not have an acute root vowel, e.g. in the paradigm of SCr. zխb ‘tooth’, cf. Gr. ʄ˶Ջʕʏʤ ‘bolt’. The original accentuation seems to have been retained in the ƀakavian dialects of Susak and Istria.

SLAVIC HISTORICAL MORPHOLOGY: NOMINAL PARADIGMS

421

6.10. Pedersen’s law and rise of distinctive tone. The stress was retracted from inner syllables in accentually mobile paradigms (cf. 3.2 above), e.g. Ru. ná vodu ‘onto the water’, né byl ‘was not’, pródal ‘sold’, póvod ‘rein’. The stress was also retracted within the initial syllable of barytone forms in paradigms with mobile stress, yielding a falling tone. All other stressed vowels became rising by opposition. This development was more recent than Illiǁ-Svityǁ’s law (6.9) because it eliminated the identity of the two accentual paradigms in the barytone case forms on which the generalization of accentual mobility was based. It gave rise to the falling tone in e.g. acc.sg. *gգlwoN, *բwiǀ‫׀‬N, *zթNbu, *jխ֓ga, *pբljֻ, *sխnu, *kբsti, *dզni, *զmeN, *dեteN. 7.1-8.0 Late Middle Slavic This was the time when the trend toward simplification of the syllable structure reached its culmination and the major dialect divisions established themselves. 7.1. Prothesis. The hiatus between a word-final and a word-initial vowel was filled with a glide, which was j if at least one of the vowels was front and w if the preceding vowel was back and the following vowel was rounded, e.g. *jֻ֐stlj֓ ‘to eat’, Lith. ‫ם‬sti, now with the same initial as *jֻ֐xa֐tlj֓ ‘to ride’, Lith. jóti. As a consequence of this development, which was apparently posterior to 6.8, initial *j lost the status of a phoneme before front vowels, so that we can write *æ, *ՙ, *æN, *øֆ, *øN, *ü, *٤, *üN for *ֻ, *ƽ֓, *ֻN, *Ǣ֓, *‫׀‬N, *u֓, *ǰ֓, *u֓N, e.g. *pբljæ, *pբljՙ, *pբljøֆ, *wàljøN, *wàljü, *wàljüN, where *j and the following front vowel are jointly relevant (cf. Kortlandt 1972: 148f.). The twofold glide before a back vowel gave rise to doublets, e.g. OCS. utro and jutro ‘morning’, ajce and jajce ‘egg’. 7.2. Dolobko’s law. Barytone forms of accentually mobile paradigms lost the stress to an enclitic particle, e.g. Sln. lahkէ ‘light’, gen.sg. lahkegà, dat.sg. lahkemù. This development was probably posterior to the rise of distinctive tone (6.10). 7.3. First simplification of palatals: *ǀ > c, *՝փ > ՝, in South and East Slavic also *ǧ > s, *ǧǀ > sc, *Ǻ՝փ > z՝, e.g. *awicæ֐, *բwicøN, *j٤՝ՙxù. The resulting dentals continued to be palatalized for some time. 7.4. The clusters *֦w, *٥w, *xփw which had arisen before front vowels as a result of the second palatalization (6.6) shared the development of 7.3 in South and East Slavic, but were depalatalized in West Slavic. The clusters *֦n and *٥n preserved the palatalization in the nasal, e.g. in *àgǟi ‘fire’. 7.5. Loss of *t and *d before *l in South and East Slavic. As in the case of *ǧǀ (7.3) and *֦w (7.4), West Slavic preserved the original cluster. 7.6. Simplification of geminated affricates: *tǀ > *ǧǀ, *d՝փ > *Ǻ՝փ, also *ǧtǀ > *ǧǀ, *Ǻd՝փ > *Ǻ՝փ. This development was limited to Bulgarian. For the other languages I assume that length shifted from the first, occlusive element of the geminate to its second, fricative element: *tǀ > *ǀǧ, *d՝փ > *՝փǺ. This development can be iden-

422

FREDERIK KORTLANDT

tified with the general assimilation of *j to a preceding consonant: *ǁj > *ǁš, *šj > *šš, *žj > *žž, *nj > *ǟǟ, *lj > *ǛǛ, also *pj > *pǛ, *bj > *bǛ, *mj > *mǛ. The assimilation did not change the phonemic make-up of the clusters because their second components can be regarded as the realizations of the phoneme /j/ in the respective environments. 7.7. Spirantization of the ungeminated voiced affricate *՝ > z, e.g. Graz *j٤zՙxù. This development did not reach Lekhitic (Kraków) and a part of the Bulgarian dialects (Ohrid). The root-final consonant was evidently restored in Novgorod *j٤gՙxù. 7.8. Delabialization of *u, *ǰ, *uN, *ü, *٤, *üN. This development yielded *y, *ձ, *yN, *i, *ǔ, *iN, e.g. *wy֐dra֐ ‘otter’, *ly֐Nka ‘bast’, *jէga ‘yoke’, nom.pl. *zթNbǔ, *kàǟǟǔ, gen.pl. *zoNbǹ, *kàǟǟi, *paǛǛì, acc.pl. *zթNbyN, *kàǟǟiN, inst.pl. *zoNbý, *kàǟǟǔ, *paǛǛí. As a result of the delabialization, the prothetic *w before *y, *ձ received the status of a phoneme. The new *iN from *üN did not merge with earlier *iN, which had apparently merged with *eN at this stage, e.g. *xwáleN ‘praising’. 7.9. Raising of *Ǣ, *øֆ, *lj֓. The empty hole which the delabialization had left was filled by raising the remaining rounded vowels *Ǣ, *øֆ to *ǰ, *٤, e.g. dat.sg. *zթNbǰ, *pբǛǛ٤. The corresponding unrounded front vowel *lj֓ < PIE. *ei was raised to merge with *ǔ, e.g. loc.sg. *kastí, *pբǛǛǔ, nom.pl. *kastı֐, loc.pl. *paǛǛǔxù. The raising did not affect the rounded nasal vowels *oN, *øN. The phonetically complex unrounded nasal back vowel *yN lost its nasal feature, e.g. *ly֐ka ‘bast’, *sy֎ta ‘hundred’, acc.pl. *zթNby. The corresponding nasal front vowel *iN was lowered to *‫־‬N while *eN was lowered to *æN, e.g. in *kàǟǟ‫־‬N, *dեtæN, *xwálæN (cf. Kortlandt 1979b: 269). 7.10. Retraction of initial *jæ-, *j٤- to *a-, *ǰ- in East Slavic, e.g. Ru. ózero ‘lake’, útro ‘morning’, cf. SCr. jդzero, jլtro. 7.11. Dissimilation of /j/ in the word for ‘foreign’ in South Slavic, e.g. SCr. tխDž, Ru. ǁužój. 7.12. Metathesis of liquids in South Slavic and Czecho-Slovak, e.g. in *kra֐wa֐. The metathesis was often accompanied by lengthening. The timbre of the vowel shows that the metathesis was earlier than the rise of the new timbre distinctions (7.13) in Czecho-Slovak and South Slavic, but later than that development in Lekhitic and Sorbian, e.g. Cz. kráva, Slk. krava, Sln. kráva, SCr. krբva, but Po. krowa, USo. kruwa < krówa. The metathesis did not reach East Slavic except in word-initial position, where it was early in the entire Slavic area, e.g. Ru. rálo ‘plough’, Cz. rádlo < *a֐rdla (cf. Kortlandt 2003a: 232). 7.13. Rise of new timbre distinctions: *e, *Ӹ, *Ԕ, *o versus *æ, *i, *y, *a. In posttonic syllables the glottal stop was lost without compensatory lengthening whereas in stressed syllables it survived as a prosodic feature of the vowel, e.g. *kra֐wa, Kraków *kro֐wa, Novgorod *ko֐rwa. As a result, the timbre distinctions

SLAVIC HISTORICAL MORPHOLOGY: NOMINAL PARADIGMS

423

between the short vowels and the acute “long” vowels became phonemically relevant, e.g. *wy֐dra ‘otter’, *sԔ֎to ‘hundred’. As a result of the rise of the new timbre distinctions, the quantitative oppositions in pretonic syllables were rephonemicized as timbre differences, e.g. *glawa֐, *jiga֐. All pretonic vowels of this stage are reflected as short vowels in the historical languages, e.g. Cz. ruka ‘hand’ < *roNka֐, SCr. màlina ‘raspberry’ < *malı֐na. The length in SCr. rúka was introduced from the barytone forms such as acc.sg. rխku, while the original short vowel was preserved in the oblique plural form rùkama. Long vowels in posttonic syllables were not shortened, e.g. *òsnowƽ, inst.pl. *žènamǔ, where the long final vowel is reflected by the neocircumflex tone of Sln. osnթ֗va, ženգmi (see 10.9 below). The alternation between short pretonic and long posttonic vowels in paradigms with mobile stress was removed by the generalization of the long vowel in Serbo-Croatian and the short vowel in Czech and Polish, e.g. SCr. gըlǰb ‘pigeon’, žդlǰd ‘acorn’, lբbǰd ‘swan’, ըblƽst ‘region’, Cz. holub, žalud, labuǬ, oblast. The long vowel was retained everywhere if it did not alternate with a short vowel, e.g. SCr. mjդsljc ‘month’, pդnljz ‘coin’, jբstrljb ‘hawk’, pբǰk ‘spider’, Cz. mLJsíc, peníz, jestǥáb, pavouk. These words had fixed stress on the laryngealized vowel of the first syllable. Both Czech and Serbo-Croatian have a short vowel in a suffix which contained a laryngeal, e.g. SCr. bògat ‘rich’, srւdit ‘angry’. 7.14. Raising of the low nasal vowels *aN, *æN to *yN, *eN in South Slavic, e.g. OCS. nesy(˛) ‘carrying’, xvalNj ‘praising’, ORu. nesa, xvalja. 7.15. Van Wijk’s law and loss of /j/. Long consonants (see 7.6 above) were shortened with compensatory lengthening of the following vowel, e.g. SCr. pէšlj ‘writes’ < *píšše < *pé֓šjֻ < *péisje. This development was evidently more recent than 7.13, cf. *wòǛՙ < *wòǛǛæ < *wàǛǛæ֐ < *wàljֻ֐. New *lj did not merge with earlier *lj, which had become *ՙ at stage 7.13, e.g. in *dæ֏tæN. After the loss of the glottal stop in posttonic syllables and the rise of new long vowels as a result of Van Wijk’s law, case endings could have three different quantities. For example, the nom.sg. ending of the a-stems was short in *žèna, long in *wòǛՙ and *òsnowƽ, and indifferent with respect to length in *gora֐ ‘mountain’. The same distribution holds for the neuter nom.acc.pl. ending. At the next stage several levelings took place. Endings which did not occur under the stress were shortened in the whole Slavic territory. Length was generalized in the unstressed nom.acc.pl. ending of Sln. lե֓ta ‘years’, but not under the stress, e.g. drvà ‘firewood’. Conversely, the distinction between a short unstressed nasal vowel and a long nasal vowel under the stress was preserved in Sln. gen.sg. lípe ‘lime-tree’, goré֗ ‘mountain’, and in SCr. nom.acc.pl. glգve ‘heads’, gen.sg. glávlj. This difference became phonemic as a result of Dybo’s law (see 8.7 below), which reintroduced long unstressed nasal vowels and short nasal vowels under the stress.

424

FREDERIK KORTLANDT

8.1-9.0 Young Proto-Slavic The redundancies which the trend toward rising sonority had created evoked a reaction, which eventually led to the disintegration of the prosodic system and to the rise of new closed syllables. 8.1. Contractions in posttonic syllables, e.g. ƀak. (Novi) pítƽ ‘asks’, Bulg. píta, cf. ƀak. kopգ < *kopa֐(j)e ‘digs’, Bulg. kopáe, OPol. kopaje. This development was posterior to the rise of the new timbre distinctions (7.13) because new *lj did not merge with earlier *lj, which became *æ, cf. Cz. gen.sg. nového ‘new’. Contraction took place in inst.sg. *kra֐wǢN, *žènǢN, *òsnowǢN, *wòǛøֆN, but not in *glawòøN and *owӸcèøN. The uncontracted forms were partly restored after the rise of new /j/, which was early in East Slavic and late in West Slavic (cf. Kortlandt 2003b). 8.2. Retraction of the stress from final jers, e.g. Sln. gen.pl. gó֓r < *gorԔւ ‘mountains’, dán < *dӸփnԔ < *dӸnԔւ ‘days’, Pol. rƾk < *r՛kԔւ ‘hands’, also *ӸménԔ. Pretonic jers in inner syllables could not receive the stress, e.g. Sln. gen.pl. ó֓vՖc < *owӸcӸւ, Ru. dat.pl. détjam < *dætӸmԔւ (with -jam for ORu. -emԔ). This development gave rise to new long vowels, which subsequently spread to the gen.pl. forms of other accent types (cf. Kortlandt 1978b: 282-286). 8.3. Raising of *æ to *‫ ־‬in Slovene, Sorbian, Czecho-Slovak, and East Slavic, e.g. Graz *d‫֏־‬teN, Novgorod *d‫֏־‬tæN, cf. Kraków *dæ֏tæN, Ohrid *dæ֏teN. This development can be dated to approximately the same stage as the retraction of the stress from final jers (8.2) because *lj֓ became the counterpart of *Ǣ֓ in these languages. It also affected Serbo-Croatian, though perhaps slightly later and not to the same extent, cf. ƀak. (Rab) gnjƽzdը ‘nest’. 8.4. Merger of palatal fricatives: *ǧ > *š, also *ǧǀ > *šǀ, *Ǻ՝փ > *ž՝փ. 8.5. Merger of palatal clusters: *šǁ > *šǀ, *ž՞ > *ž՝փ. 8.6. Second simplification of palatals: *ǀ > c, *՝փ > ՝ in West Slavic, and subsequently *՝ > z in Czech and Sorbian; *ǀ > ǁ, *՝փ > *՞ > ž in East Slavic. The clusters *šǀ and *ž՝փ were reduced to *št and *žd in Bulgarian and the eastern dialects of Serbo-Croatian, and later in Czecho-Slovak. Similarly, the clusters *sc and *z՝ became *st and *zd in a part of the Bulgarian dialects. 8.7. Dybo’s law: rising vowels lost the stress to the following syllable, if there was one, e.g. *ženà, *osnòwƽ, *pǢNtӸւe, inst.sg. *glawoøւN. Newly stressed long vowels received a falling tone, e.g. *woǛգ. Final jers had lost their stressability (8.2) and therefore could not receive the stress, e.g. Sln. kònj < *kòǟӸ. Acute (broken, glottalized) vowels did not lose the stress, e.g. *wy֐dra ‘otter’, inst.pl. *-a֐mǔ. Dybo’s law restored distinctive vowel length in pretonic syllables, e.g. *nƽròdԔ ‘people’, *ǢNtròbƽ ‘liver’. 8.8. Lengthening of short falling vowels in monosyllables, e.g. SCr. bթg < *bըgԔ ‘god’, kթst < *kըstӸ, dգn < *dӸ֏nӸ < *dӸ֎nӸ. This development, which was apparently Common Slavic, eliminated the pitch opposition on short vowels,

SLAVIC HISTORICAL MORPHOLOGY: NOMINAL PARADIGMS

425

which had become confined to monosyllables (not counting final jers) as a result of Dybo’s law (8.7). 8.9. The inst.sg. ending -ԔmӸ of the u-stems was generalized in the paradigm of the o-stems in North Slavic. It replaced -a, which has been preserved in OCS. vӸǁera ‘yesterday’ and can be identified with Lith. -ù < *-o‫ٳ‬. The development was motivated by the merger with the gen.sg. ending *-ƽ in soft stems as a result of Van Wijk’s law (7.15) and can therefore be dated to the Young Proto-Slavic period. The rise of the South Slavic ending -omӸ requires the continued existence of the nom.sg. ending *-os and must therefore be dated to an earlier stage. 9.1-10.0 Late Proto-Slavic This is the last period of common innovations. 9.1. Pleophony in East Slavic, e.g. *koro֐wa, Ru. ogoród < *ogòrdԔ < *ògordԔ ‘kitchen-garden’, pozolóta < *pozòltƽ < *pòzoltƽ ‘gilding’. The development was evidently more recent than Dybo’s law (8.7), according to which the prefix lost the stress to the root in these words. 9.2. Loss of the acute (broken, glottalic) tone, which yielded a short rising contour, e.g. *dǹmԔ ‘smoke’, *gorà ‘mountain’. This development was more recent than the East Slavic pleophony (9.1) because the distinction between the acute and the earlier rising tone was preserved in Ukrainian, e.g. moróz < *moro֐zԔ < *mo֐rzԔ ‘frost’, gen.pl. holív < *ճolówԔ < *golwԔւ ‘heads’. 9.3. Stang’s law: the stress was retracted from long falling vowels in final syllables, e.g. *wuòǛa, Ru. dial. vôlja, Cz. vDzle, Slk. vôǚa, Sln. vó֗lja, SCr. vըlja. The long vowel was shortened, except in Lekhitic, where traces of length remain, e.g. OPol. wolå. The newly stressed vowel received a rising tone. Pretonic jers in inner syllables would not receive the stress, and final jers did not count as syllables with respect to Stang’s law. The development was more recent than the loss of the acute tone (9.2), as is clear from SCr. gen.pl. jդzǔkƽ ‘tongues’. The short vowel in the first syllable of Cz. jazyk and SCr. jèzik shows that this word had fixed stress on the second syllable before Dybo’s law operated: *(j)eNzy֐kԔ. The retraction of the stress in the gen.pl. form points to *jeNzy֏kԔ from *jeNzǹkԔ with analogical lengthening after the loss of the acute tone. If Stang’s law had been earlier than the loss of the acute tone, the lengthening would have been impossible and the retraction of the stress would not have taken place in this form. Note that the lengthening was actually more recent than Stang’s law in ƀak. (Novi) gen.pl. susեd ‘neighbors’, kolեn ‘knees’. 9.4. Shortening of long falling vowels, e.g. Cz. mladost ‘youth’, acc.sg. ruku ‘hand’, SCr. mlբdǢst ‘youth’, gen.sg. prբseta ‘sucking-pig’. The shortening did not affect monosyllables in Slovene and Serbo-Croatian and the first syllable of disyllabic word forms in the latter language, e.g. SCr. bթg ‘god’, prգse ‘sucking-

426

FREDERIK KORTLANDT

pig’, acc.sg. rխku ‘hand’. The dialect of the Kiev Leaflets sides with SerboCroatian in this respect (cf. Kortlandt 1980). 9.5. Proto-Slavic *u was fronted to *ü in Kajkavian and neighboring dialects (cf. Greenberg 2000: 116). 9.6. The rounded nasal vowels *oN, *øN were raised to *uN, *üN in SerboCroatian, Sorbian, Czecho-Slovak, and East Slavic, e.g. in *púNtӸ. This development was apparently more recent than the fronting of *u (9.5). 9.7. Denasalization of the nasal vowels in East Slavic, e.g. in *pútӸ, and subsequently in Czecho-Slovak. 9.8. Rise of the palatalization correlation in Lekhitic, e.g. in *póNǭӸ, *d֖æւǭæN, and subsequently in the other North Slavic languages. 9.9. Merger of the jers in Serbo-Croatian, Slovene, Czech, and Lekhitic, e.g. loc.pl. Graz *pǢNtӸւxӸ, Kraków *pǢNǭӸւxӸ, Novgorod *pǰtӸւxԔ, Ohrid *pǢNtӸւxԔ. 10.1– Disintegrating Slavic This is the period of parallel but not identical developments in the separate languages. 10.1. The denasalization spread to affect all Slavic languages. The nasal vowels are best preserved in modern Polish. 10.2. The rise of the palatalization correlation affected the languages differently. The correlation is especially characteristic of modern Russian. 10.3. The jers were lost or merged with other vowels under various conditions in the separate languages. They have been preserved as a separate phoneme in Slovene. 10.4. Short rising vowels were lengthened in Russian, e.g. dial. kôn’ < *kǢǟ < *kòǟӸ, cf. bog < *bթgԔ ‘god’, where the vowel had been shortened (9.4). The length has been preserved in Baltic and Fennic loan words from Russian, e.g. Latv. grãmata ‘book’, Est. raamat < *gra֐motƽ. 10.5. Short vowels were lengthened in monosyllables in Ukrainian, e.g. kin’ < *kǢǟ < *kòǟӸ. Other new long vowels originated from compensatory lengthening before a lost jer in inner syllables. 10.6. Short rising vowels in open first syllables of disyllabic words were lengthened in Czech and Upper Sorbian unless the following syllable contained a long vowel, e.g. Cz. kráva < *kràva < *kra֐va, vDzle < *vǢǛa < *wuòǛa, psáti < *pӸsàti < *pӸsa֐ti ‘to write’, USo. kruwa < *krǢwa < *kròwa < *kro֐wa, Cz. gen.pl. krav, inst.pl. kravami. This development was evidently more recent than the loss of pretonic jers. 10.7. Falling vowels lost the stress to the following syllable in Slovene, e.g. okթ֓ ‘eye’, mladթ֓st ‘youth’, acc.sg. rokթ֗ ‘hand’. The newly stressed vowel received a long falling tone. This development was evidently more recent than Stang’s law (9.3) but earlier than the loss of the nasal vowels. The Freising Fragments can be

SLAVIC HISTORICAL MORPHOLOGY: NOMINAL PARADIGMS

427

dated between Stang’s law and the progressive accent shift (cf. Kortlandt 1975: 411). The accent shift probably originated from the spread of the falling tone over two syllables as a result of the shortening (9.4). 10.8. Stressed short vowels were lengthened and received a falling tone before a non-final lost jer in Slovene, e.g. bէtka ‘battle’. This development was evidently more recent than the progressive accent shift (10.7). 10.9. Stressed short vowels were lengthened and received a falling tone in Slovene if the following syllable contained a long vowel, which was shortened, e.g. lե֓ta, osnթ֗va, inst.pl. ženգmi. The development was evidently posterior to the progressive accent shift (10.7). 10.10. The stress was retracted from a final syllable to a preceding long vowel in Lekhitic (cf. Kortlandt 1978a: 77-79), Slovene, and dialects of Serbo-Croatian, where the retraction yielded a rising tone. 10.11. Stressed short vowels in non-final syllables were lengthened and received a rising tone in Slovene, e.g. lé֓to ‘year’, vó֗lja ‘will’. This development, which was more recent than both the rise of the neocircumflex (10.8, 10.9) and the retraction of the stress to a preceding long vowel (10.10), did not reach the easternmost dialects of the language. 10.12. The stress was retracted from a final short vowel in Lekhitic, the Pannonian dialect of the Kiev Leaflets (cf. Kortlandt 1980), dialects of Slovene and Serbo-Croatian, and Bulgarian. This retraction, which generally yielded a rising tone, was followed by others in various dialectal areas. Elsewhere I have discussed the oldest isoglosses in the Slavic territories (1982, 2003a). Most of these reflect consonantal developments: simplification of palatals (7.3), loss of *t and *d before *l (7.5), simplification of geminated affricates (7.6), spirantization of the ungeminated voiced affricate *՝ > z (7.7), lenition of *g to ճ, and dissimilation of /j/ in the word for ‘foreign’ (7.11). Vocalic developments include the East Slavic retraction of initial *jæ-, *j٤- to o-, u- (7.10), the metathesis of liquids (7.12), the elimination of the alternation between short pretonic and long posttonic vowels in paradigms with mobile stress (7.13), and the South Slavic raising of *aN, *æN to *yN, *eN (7.14). As a result, the paradigms given above start diverging after the end of the Early Middle Slavic period (7.0). At the end of the Late Middle Slavic period (8.0) we have South Slavic *kra֐wa versus Polish *kro֐wa and East Slavic *ko֐rwa, acc.pl. South Slavic *-eN versus North Slavic *-‫־‬N, further *էճo beside *էgo, inst.sg. *էճomӸ, *էgomӸ, *iga֐, loc.sg. *էzՙ, *է՝ՙ, *էgՙ, loc.pl. *izæxԔւ, *i՝æxԔւ, *igæxԔւ, and *dæ֏teN, *dæteNtè beside *dæ֏tæN, *dætæNtè. At the end of the Young Proto-Slavic period (9.0) we have dat.loc.sg. *-æ and *-‫ ־‬in the ƽ-stems, inst.sg. *-omӸ beside *-ԔmӸ and locative endings *-æ, *-æxԔ and *-‫־‬, *-‫־‬xԔ in the o-stems, dialectal variation between *d‫֏־‬teN, *d‫֏־‬tæN, *dæ֏tæN, *dæ֏teN, and restoration of /j/ in East Slavic *pǢNtӸւje, *pǢNtӸւjӸ, *d‫־‬tӸւjӸ. At the end of the Late Proto-Slavic period (10.0) we have pleophony, restoration of the pronominal ending, and raising and denasaliza-

428

FREDERIK KORTLANDT

tion of the nasal vowel in East Slavic inst.sg. *korówoju, rise of palatalization and merger of jers and nasal vowels in Polish inst.sg. *pƽNǭӸւm֖Ӹ, *d֖æւǭaNǭӸm֖Ӹ and dat.pl. *pƽNǭӸւmӸ, *d֖æւǭӸmӸ, restoration of /j/ in South Slavic *pǢNtӸւje, also some of these developments elsewhere, and loss of distinctive tone outside the western South Slavic dialectal area. In West Slavic, the restoration of /j/ was evidently more recent (cf. Kortlandt 2003b, Derksen 2003). Leiden University REFERENCES Beekes, Robert S.P. 1995 Comparative Indo-European linguistics. Amsterdam: Benjamins. Derksen, Rick 2003 “Slavic *jӸ-”. In: Jos Schaeken, Peter Houtzagers, Janneke Kalsbeek (eds.), Dutch Contributions to the Thirteenth International Congress of Slavists, Ljubljana: Linguistics (Studies in Slavic and General Linguistics 30), 97-105. Amsterdam-New York: Rodopi. Greenberg, Marc L. 2000 A historical phonology of the Slovene language. Heidelberg: Winter. Holzer, Georg 2005 “Zur relativen Datierung prosodischer Prozesse im Gemeinslavischen und frühen Kroatischen”. Wiener Slavistisches Jahrbuch 51, 31-71. Kortlandt, Frederik 1972 Modelling the phoneme. The Hague: Mouton. 1975 “Jers and nasal vowels in the Freising Fragments”. Slavistiǁna Revija 23, 405412. 1978a “A history of Slavic accentuation”. Lingua 44, 67-91. 1978b “On the history of the genitive plural in Slavic, Baltic, Germanic, and IndoEuropean”. Lingua 45, 281-300. 1979a “Sur l’accentuation des noms postverbaux en slave”. In: Jan M. MeǓer (ed.), Dutch Contributions to the Eighth International Congress of Slavists. Zagreb, Ljubljana, Sept. 3-9, 1978, 325-328. Lisse: Peter de Ridder. 1979b “On the history of the Slavic nasal vowels”. Indogermanische Forschungen 84, 259-272. 1980 “Zur Akzentuierung der Kiever Blätter”. Zeitschrift für slavische Philologie 41, 1-4. 1982 “Early dialectal diversity in South Slavic I”. In: A.A. Barentsen, R. Sprenger, M.G.M. Tielemans (eds.), South Slavic and Balkan Linguistics (Studies in Slavic and General Linguistics 2), 177-192. Amsterdam: Rodopi. 1983 “On final syllables in Slavic”. Journal of Indo-European Studies 11, 167-185. 1989 “Od praindoevropskog jezika do slovenskog (fonološki razvoj)”. Zbornik za Filologiju i Lingvistiku 32/2, 41-58. (English edition: “From Proto-Indo-European to Slavic”. www.kortlandt.nl.) 1997 “Baltic lj- and ǔ/jƽ-stems”. Baltistica 32/2, 157-163. 2000 “Initial a- and e- in Old Prussian”. Linguistica Baltica 8, 125-127.

SLAVIC HISTORICAL MORPHOLOGY: NOMINAL PARADIGMS 2003a

2003b 2005a 2005b 2006 2007 forthc. a

429

“Early dialectal diversity in South Slavic II”. In: Jos Schaeken, Peter Houtzagers, Janneke Kalsbeek (eds.), Dutch Contributions to the Thirteenth International Congress of Slavists, Ljubljana: Linguistics (Studies in Slavic and General Linguistics 30), 215-235. Amsterdam-New York: Rodopi. “Indo-European e-, a-, o- in Slavic”. International Journal of Slavic Linguistics and Poetics 44-45, 279-282. “From Serbo-Croatian to Indo-European”. Wiener Slavistisches Jahrbuch 51, 113-130. “Holger Pedersen’s Études lituaniennes revisited”. Baltistica 6 priedas, 151-157. “On the relative chronology of Slavic accentual developments”. Wiener Slavistisches Jahrbuch 52, 25-41. “The development of the Indo-European syllabic resonants in Balto-Slavic”. Baltistica 42/1, 7-12. “Accent retraction and tonogenesis”. In: Thomas Olander, Jenny Helena Larsson (eds.), Stressing the Past. Papers from the Second International Workshop on Balto-Slavic Accentology (IWoBA II). University of Copenhagen, 1-3 September 2006 (Studies in Slavic and General Linguistics). Amsterdam-New York: Rodopi. (See www.kortlandt.nl.) “Winter’s law again”. (See www.kortlandt.nl.)

forthc. b Leskien, August 1907 “Über slavisches o in Endsilben”. Indogermanische Forschungen 21, 335-338. Meillet, Antoine 1921 “Le genre féminin de kostǒ et de solǒ”. In: Prace Lingwistyczne ofiarowane Janowi Baudouinowi de Courtenay, 1-3. Kraków: Uniwersytet Jagielloǝski. Pronk, Tijmen forthc. “The accentuation of the Slavic n-stems”. In: Thomas Olander, Jenny Helena Larsson (eds.), Stressing the Past. Papers from the Second International Workshop on Balto-Slavic Accentology (IWoBA II). University of Copenhagen, 1-3 September 2006 (Studies in Slavic and General Linguistics). Amsterdam-New York: Rodopi. Stang, Christian S. 1957 Slavonic accentuation. Oslo: Universitetsforlaget.

SUMMARY Elsewhere I have presented my relative chronology of Slavic phonological developments and compared it with the chronologies proposed by Greenberg (2000) and Holzer (2005). Here I intend to examine the consequences for the morphology of the nominal paradigms. The period covered is from the disintegration of the Balto-Slavic proto-language up to the loss of the jers. The main dialect chosen is that of Ohrid. For the later stages I also give dialectal variants which I reconstruct for Graz (i.e. the dialect of the Freising Fragments), Kraków and Novgorod. The paradigms chosen reflect masculines, feminines and neuters, hard and soft a- and o-stems, u-, iand consonantal stems, and accent classes (a), (b) and (c).

Dutch Contributions to the Fourteenth International Congress of Slavists, Ohrid: Linguistics (SSGL 34). Amsterdam – New York: Rodopi, 2008, 431-449.

TRANSCRIPTION OF RUSSIAN INTONATION, TORI, AN INTERACTIVE RESEARCH TOOL AND LEARNING MODULE ON THE INTERNET CECILIA ODÉ1

1. Introduction ToRI is an abbreviation for Transcription of Russian Intonation. ToRI is a new transcription system and available free of charge on www.fon.hum.uva.nl/ tori/. It uses unambiguous symbols for the transcription of pitch accents, pitch movements connecting the pitch accents and utterance boundaries marked by pitch. The descriptions of all forms of these pitch phenomena expressed in ToRI symbols are based on the results of perception experiments with native speakers of Russian. In the system, each symbol representing a pitch accent is described with rules for its realization, that is, with phonetic correlates and limits of perceptual tolerance. Each pitch accent is presented with pictures of the contours, with sound examples and with interactive audiovisual exercises training the recognition and production of Russian pitch phenomena. In the examples and exercises, main communicative functions for each accent are also given. The terminology used in the descriptions is defined in an online glossary. The system is set up in such a way that it can be used as a learning module for advanced students and linguists. It can also be used as a research tool for the description of intonation in variants of Russian pronunciation and in Russian dialects, for example during linguistic fieldwork. ToRI can be used individually or, if used as a learning module, with a teacher in a classroom situation. On top of the opening page, on the horizontal bar, the user can choose to open for reading or consultation the following documents: Acknowledgments, Introduction (what is ToRI?), Symbols (descriptions, conventions), Realization rules (phonetic correlates of pitch accents), Glossary (38 terms on intonation 1 This research is financially supported by the Netherlands Organisation for Scientific Research, NWO (dossiernr. 355-75-004). The website ToRI is financially supported by the University of Amsterdam Onderwijs Innovatie Fonds (OIF, Teaching Innovation Fund) and by the Stichting Spraaktechnologie (Foundation for Speech Technology).

432

CECILIA ODÉ

defined) and Literature (references used for the development of ToRI, and for further reading). Technical information on the use of ToRI is given on the vertical bar under Getting started. Furthermore on the vertical bar can be selected: Pitch accents, Boundaries, Non-accent lending pitch movements, Communicative functions of pitch phenomena, Interpretation of pitch phenomena, and Pitch accents in longer fragments. In my article for the previous Thirteenth International Congress of Slavists in Ljubljana 2003, the set-up and background of ToRI has been described (Odé 2003b). In the meantime, the ToRI website has been developed. In this article for the Fourteenth International Congress of Slavists in Ohrid 2007, I will discuss the most crucial terms for ToRI in the glossary, and the full set of symbols as they appear under the buttons Glossary and Symbols on the horizontal bar of the website, respectively. Reasons to describe only these two parts of ToRI are that the glossary is considered essential for my approach to intonation, and the symbols are based on my own experimental work that has been carried out for earlier descriptions of Russian intonation (e.g. Odé 1989, 1992) and for the present website (e.g. Odé 2003a, 2005 and in press). A full description of all ToRI webpages would exceed the limits of an article for the present volume. 2. Glossary In the glossary, 38 terms on pitch phenomena and related issues are explained for the ToRI user. A definition of terms was considered relevant for ToRI users in order to inform them about the approach to intonation of the author. Another reason for including a glossary of such terms on the website is the existence of the many different, sometimes conflicting, definitions for a same term as they appear in manuals, in linguistic and phonetic dictionaries, and in the literature on intonation. Those definitions highly depend on the linguistic approach applied by the various authors that will not be discussed in the present article. Most definitions in the glossary are inspired by and mainly based on the work of, in alphabetical order, Bolinger (e.g. 1958, 1986, 1989), ’t Hart et al. (1990) and Keijsper (e.g. 1983, 1987, 1992), and based on experimental evidence published in my own works on Russian intonation (Odé 1989, 1992, 2003a, 2003b, 2005, in press). As said in the Introduction, in this section I will present only a few terms that are fundamental for the approach on which my new transcription of Russian intonation is based. These terms are: pitch accent and accent-lending pitch movements, perceptual equivalence, connecting pitch movements, and register. For the other terms the reader is referred to the webpages.

TORI, AN INTERACTIVE RESEARCH TOOL

433

In the definitions discussed in the sections below, terms occur that are also explained in the glossary; these are indicated in italics. 2.1 Pitch accent and accent-lending pitch movements, perceptual equivalence and connecting pitch movements defined A pitch accent is defined as follows: “In the context of intonation languages with word stress, a pitch accent is a pitch movement or configuration of pitch movements lending perceptual prominence to a syllable in a word that thus receives an accent.” An accent-lending pitch movement is defined as “A pitch movement realized in such a way that a hearer perceives prominence on the syllable in which it occurs and thus perceives a pitch accent on that syllable. Accent-lending pitch movements are realized with various gradations, with varying salience, and usually occur in syllables carrying word stress. The position of word stress can be found in a dictionary, but the location and type of pitch accent depends on the intention of the speaker and thus cannot be found in dictionaries. Not all realizations of a given type of pitch accent are equally salient, but within one type all realizations are always perceptually equivalent.” The term perceptual equivalence is crucial in defining types of pitch accent, as follows from its definition: “In the context of intonation, all realizations of one type of pitch accent, of one type of pitch boundary or of types of configuration of pitch movements, are within that type perceptually equivalent. That is, according to native speakers, realizations of one type of pitch accent, pitch boundary or configuration of pitch movements are successful imitations of one another. The perceptual equivalence of realizations of a given type can be established by verifying realizations in listening experiments with native speakers.” First experiments on perceptual equivalence were carried out for Odé (1989) in the late eighties. A recent example of an experiment on the perceptual equivalence of pitch accents is Odé (2005). The latter paper presents the results of a perception experiment that was carried out to verify the hypothesis that in Russian the contrast between pitch accents H*L and H*H 2 on utterance-final syllables is neutralized. Recordings for the experiment were ten sets of three short utterances with word stress in the ultimate, penultimate and antepenultimate syllable of the utterance-final word. These utterances were read aloud by four female and four male native speakers. They were asked to realize accents H*L and H*H in the utterance-final word. After instructions and rehearsing, recordings were made separately for each of the two types. In the perception experiment, 30 native subjects listened to short utterances selected from the re2

Note that in the article pitch accents H*L and H*H are named LH*L and LH*. These latter names have later been changed into H*L and H*H, that are considered to more adequately represent the forms of the types of pitch accent.

434

CECILIA ODÉ

cordings and presented in 180 pairs: 120 pairs with ultimate stress and, in order to test whether listeners can hear the difference at all, 60 pairs with penultimate and antepenultimate word stress in utterance-final position. The 180 stimuli pairs consisted of short utterances with realizations of H*L and H*H on the final word, each pair containing two same or two different types of pitch accent. The task was to compare two stimuli in a pair and to indicate on a score form whether two realizations in a stimulus pair count as passable imitations of each other and thus belong to the same type of pitch accent. The same/different judgments indicate that listeners successfully distinguished between the two pitch accents in the antepenultimate and penultimate conditions, but much less so in the ultimate condition. This suggests that the two accents are truncated in final position, but not neutralized. The results of the above described experiment are reflected in the new set of symbols in section 3. Between pitch accents so-called connecting pitch movements occur. These are defined as “Pitch movements between pitch accents or between pitch accents and boundaries. Pitch does not move in a straight line from accent to accent or from accent to boundary. In sequences of connecting pitch movements, turning points occur, thus separating movements belonging to the last accent from those belonging to the next accent or boundary.” For a recent discussion on the definition of pitch accent the reader is referred to Odé (2007). 2.2 Register defined A register is defined as “The range of a given speaker between his highest and lowest pitch level. The register is subdivided into a high register and a low register, but there is no exact borderline between the two. Roughly speaking, on a logarithmic scale the low register covers a range from 0-10 semitones (ST) and the high register from 10 ST to the highest pitch level reached by a given speaker. For Russian this high level can be 25 ST above the lowest pitch level (0 ST) of a speaker. Full pitch accents H*L, H*H and H*M reach their highest point in the high register, reduced pitch accents H*L, H*H and H*M reach their highest point in the low register.” An interval of 25 ST is more than two octaves if expressed in musical terms. Russian full pitch accents H*L, H*H and H*M will be discussed in section 3. The highest and lowest pitch level of a speaker are important for establishing the phonetic specification of intervals: the excursion size. On the basis of perception experiments it was found that a given pitch level reached by a speaker is perceived by the hearer as relative to the speaker’s lowest level (Odé 1989: 89ff.; in press). That is, an interval that starts and ends within the high register is, even if the size of the interval itself is small, perceived as if it were realized starting from the low level. So it is the pitch target relative to the lowest level of a

TORI, AN INTERACTIVE RESEARCH TOOL

435

speaker that is responsible for pitch perception, and not the actual size of the interval. 3. Symbols On the horizontal bar of the ToRI website, under the button Symbols, a description of the symbols as used in and defined for ToRI is presented. These symbols are based on the experimentally verified classification into types of Russian pitch accents in Odé (1989). The names of pitch accents in this classification have not been maintained in the new transcription symbols. For the convenience of the reader familiar with the original names, an overview of the translation into new symbols is given below. Reasons for not using the original names in ToRI have been extensively discussed (e.g. in Odé 2003) and will only be summarized here. Considerations were that the new symbols as they are nowadays used worldwide by intonologists would enable a comparison with intonation systems in other languages using the same set of symbols to define pitch accents. However, symbols are to a high extent language specific. For example, a pitch accent defined with symbol H*L in Dutch will considerably differ from Russian H*L in its realizations. The symbols for pitch accent H*L express a high pitch target (H) in the accented syllable (*), indicated by H*, followed by a fall to the low pitch level (L), indicated by L. But the high pitch target in Russian is in general much higher than in Dutch, and the fall after the highest target reached is much steeper than in Dutch, to mention just two perceptually relevant differences for this accent between the two languages. Unfortunately, it is an illusion to expect that a set of symbols as nowadays used to transcribe intonation may equal a set of symbols as used for the articulation of speech sounds developed by the International Phonetic Association, the so-called IPA-alphabet. This is not to say that no transcription symbols exist to adequately represent intonation. The symbols used differ for various reasons. For example, symbols may present a phonetic or a phonological transcription. Furthermore, the linguistic school to which an intonologist belongs may be reflected in the definition of symbols. Dramatic in this respect is the publication Intonation Systems (Hirst and Di Cristo (1999)) which, despite the request from the editors to authors to follow a general outline, presents a hybrid collection of twenty systems of intonation description that does not allow a comparison of those systems. However, the phonological transcription of intonation in thirteen languages described in Sun-Ah Jun (ed. 2005) are all based on the same approach to the study of intonation, namely the Autosegmental-Metrical model of intonation phonology. As such the book is a coherent study of intonation by intonologists all using the same set of symbols developed for the transcription in ToBI, Tone and Break Indices (Beckman and Ayers (1997), Beckman et al. (2005)). Yet not

436

CECILIA ODÉ

all intonologists use the symbols in the same way. Therefore, in her final chapter Sun-Ah Jun could not present all thirteen transcriptions in the summary because of the differences between the transcriptions (ed. 2005: 430ff.). For example, in his Transcription of Dutch Intonation, ToDI, Gussenhoven describes tones but does not include break indices and therefore calls his transcription a “ToBI-like system” (Gussenhoven 2005: 122). It is beyond the scope of this article to go into more details. My set of transcription symbols resembles those used, for instance, in the autosegmental approach. My transcription is called ToRI which implies a reference to ToDI rather than to ToBI, but differs again from ToDI. Differences are not only language specific, such as distinguishing a perceptually relevant mid pitch level in Russian. In ToRI, phonetic correlates of pitch accents, realizations rules for pitch accents and main communicative functions of pitch accents are also presented in descriptions and exercises. Moreover, my approach to intonation is not in the tradition of the Autosegmental-Metrical model of intonation phonology. The initial goal of ToRI was a description of Russian intonation in terms of pitch targets with conventions for the scaling and timing of the pitch targets that are relevant for Russian intonation. It is assumed that as a result of my new transcription symbols, Russian intonation is now easier to compare with the intonation of other languages described with a same set of symbols (e.g. the languages described in Sun-Ah Jun (2005)), provided that the language-specific features described in the phonetic specification of Russian pitch phenomena are taken into account. 3.1 Overview of the new ToRI transcription symbols H*, L* H, M, L %H, %M, %L L% % ^ >>> \/\ \/\n

pitch accent with high or low pitch target reached in the accented syllable non-accent-lending pitch movements to the high, mid or low level initial high, mid or low pitch final low pitch boundary (final high pitch boundaries do not occur in Russian) boundary not marked by pitch raised peak: a small high rise, optional sawtooth pattern with sequences of reduced rising or falling pitch accents single harmonica pattern repeated harmonica pattern

TORI, AN INTERACTIVE RESEARCH TOOL

437

3.2 General conventions Depending on the type, symbols for pitch phenomena occur as singletons or as combinations of symbols without interspacing. For example, singleton H indicates a non-accent lending pitch movement to the high level. The combination of symbols L*H indicates a type of pitch accent with a low pitch target reached in the accented syllable, immediately followed by a high pitch target; both targets L and H belong to this given pitch accent. Subsequent pitch phenomena are separated from one another by spaces. For example, H*L L% for pitch accent H*L followed by low pitch boundary L%. If no new pitch symbol appears, pitch is sustained on the last indicated level. Symbols are always indicated above the text of an utterance, exactly at the syllable on which a given pitch phenomenon is realized: %L H*L L% ҸӁҽ ҰҝѹѦҦѦҲ Ӂҽ ҔҽӁѹҔҏґӃ 3.3 Symbols for pitch accents Pitch accents are marked with an asterisk (*) immediately after the pitch target that is reached in the pitch-accented syllable: H* (high) or L* (low). A pitch accent may also consist of a configuration of accent-lending pitch movements to and/or from the pitch target reached in the accented syllable, for example H*L. A symbol or combination of symbols indicating a type of pitch accent represents all realizations of that given type (see Perceptual Equivalence in section 2.1). In accent HL*, H indicates that pitch is high in the pretonic syllable or syllables. 3.4 Symbols for non-accent-lending pitch movements The melodic context of pitch accents is further specified by indicating perceptually relevant non-accent-lending pitch movements. These movements connect pitch accents, and pitch accents and boundaries. The non-accent-lending movements preceding and following the pitch accent are indicated with the following symbols: H for rising pitch reaching a high target, L for falling pitch reaching a low target, and M for rising or falling pitch reaching a mid pitch target. Symbols are always indicated above the text of the utterance exactly at the syllable in the connecting movements where a turning point is located, that is, the point where pitch switches into another direction: %L H*M Ѣ HL* L% ҸӁҽ ҰҝѹѦҦѦҲ Ӂҽ ҔҽӁѹҔҏґӃ ӊѸѦҡҲѦ ҡ қҝҽҲҸѸ

438

CECILIA ODÉ

3.5 Symbols for boundaries Boundaries are indicated as follows: initial high, mid and low pitch with %H, %M and %L, and low pitch at boundaries with L%. Pitch boundary H%, a pitch movement at boundaries that rises to the high pitch target after the last pitch target reached, does not occur in Russian. Note that initial %L does not necessarily indicate the lowest level of a speaker: utterances frequently start above the lowest level of a speaker. Complex initial pitch is expressed with a combination of symbols. For example, an initial non-accent-lending high fall is marked with symbols %HL. Single symbol % without pitch target indicates that there is a boundary, but that it is not marked by pitch. Initial pitch and final pitch at boundaries are indicated above the text on the utterance-initial and utterancefinal syllable, respectively: %L H*H % ҸӁҽ ҰҝѹѦҦѦҲ Ӂҽ ҔҽӁѹҔҏґӃ

%L H*M HL* L% ҍҲҸқӃ ҰҸӊѹҦҽҲҧҡҾ ҡ қҝҽҲҸѸ

3.6 Overview of symbols for pitch accents with stylized contours In the stylized contours below, vertical dashed lines indicate approximate boundaries of the pitch-accented syllable. The contours are stylizations and do not reflect real pitch heights. The linear distance between two boundary lines also does not refer to the actual timing of the accented syllable. It depends on the type of pitch accent where a given pitch target is reached: early or late in the accented syllable (timing), high or low in a speaker’s register (excursion size), steep or gradual (rate of change). These and other relevant features for the respective accents are not represented in the stylizations below, but can be found on the horizontal bar under the button Realization rules. symbol

contour

symbol

H*L

L*

H*H

HL*

H*M

L*H

On utterance-final syllables, H*H and H*L are truncated:

contour

TORI, AN INTERACTIVE RESEARCH TOOL

439

In utterance-final position, no syllables follow the highest point reached in the final accented syllable H*, so there is no high or low posttonic pitch target. 3.7 New and old symbols In the overview below the translation of the Odé (1989) symbols for full pitch accents into new ToRI labels is presented. Note that there are no changes as compared to the original 1989 classification: the same full pitch accents appear under a different name. The classification of pitch accents according to Odé (1989) has been evaluated in new perception experiments described in detail in Odé (2003a). In short, the question was whether the pitch accents would be directly usable for ToRI. Two listening experiments were conducted in order to verify the perceptual equivalence between types of pitch accent selected from the 1989 corpus and types of pitch accent selected from other corpora. The results of these experiments, in which pitch accents from the 1989 corpus were tested in paired comparison tests with pitch accents from the new recordings, showed that the original classification did not need to be adjusted. In the 1989 symbols the feature timing, the position in the accented syllable where the beginning or terminal frequency of a pitch movement is reached, is indicated with - (early timing) and + (late timing). In the new symbols timing is not indicated. The user must learn the timing of the pitch accents by choosing the button Realization rules on the horizontal bar of the website, where timing is indicated. As can be derived from the old symbols presented below, accents H*L, H*H, L* and L*H have early timing, accent H*M has early and late timing, and accent HL* has late timing. ToRI symbol: H*L H*H H*M L* HL* L*H

1989 symbol: RlRhRm-/+ FlFl+ Fh-

3.8 Pauses and hesitations Pauses and hesitations can occur anywhere in an utterance, that is, within an utterance or at boundaries. Speakers make pauses and hesitations if they need, for example, time to think about how to continue, to remember something, to correct an error. Pauses and hesitations can be silent or filled. If silent, there is an interruption in the speech signal: complete silence. Filled pauses and hesitations are not silent and can be of many different types, like gm ‘eh’.

440

CECILIA ODÉ

3.9 Symbol for the raised peak The optional raised peak called zanos in Odé (1989: 12; 101) is indicated with symbol ^. For the convenience of the reader, in the stylized contours below the raised peak is drawn with a bold line. Pitch movements occurring before the raised peak are not drawn here.

^HL*

^HL*

^L*

a raised peak before accents HL* (without and with a high plateau) and L* 3.10 Symbol for the sawtooth pattern The sawtooth pattern consists of a sequence of reduced pitch accents H*M or L* or HL*, and is indicated with symbol >>> followed by the given type of pitch accent. The reduced pitch accents have a much smaller excursion size than their full counterparts, that is, pitch in the accented syllable of reduced accents by no means reaches a target as high or as low as pitch in the accented syllable of full pitch accents (see also section 2.2).

>>>reduced H*M

>>>reduced L* or HL*

sawtooth with reduced H*M

sawtooth with reduced L* or HL*

Note that reduced pitch accents occur not only in the sawtooth pattern, but also elsewhere in utterances before main pitch accents and after main accents, for example in an afterthought. 3.11 Symbol for the harmonica pattern Like the sawtooth pattern, the harmonica pattern is named after its form. Coming from the high level, usually after a raised peak, pitch accent HL* (the low target is reached late in the accented syllable) is realized, after which pitch immediately rises again. In a realization of a single harmonica, pitch accent HL* is realized after which pitch reaches the mid target; in a realization of a repeated harmonica, pitch rises after accent HL* till a level high enough to make a next raised peak (indicated in the stylization with a bold line) followed by another

TORI, AN INTERACTIVE RESEARCH TOOL

441

realization of HL*, etc. For a single harmonica, the pattern is indicated with symbol /\/, and for a repeated harmonica with symbols /\/n.

/\/ ^HL* %

/\/n ^HL* %

single harmonica

repeated harmonica

3.12 Names of symbols, labels, boundaries and communicative functions In Table 1 below the names of symbols, labels, utterance boundaries, main communicative functions and other communicative functions are given. Note that the order in which functions in the column ‘other functions’ are presented does not reflect the frequency of their occurrence. One pitch accent can be used to express more communicative functions and one communicative function can be expressed by more than one type of pitch accent. However, there will always be a difference in interpretation; such differences in interpretations can be very subtle. For example, the main function of pitch accents H*H and H*M is to express incompleteness, that is, a right context is required. In a continuation, accent H*H expresses a closed enumeration, whereas accent H*M expresses an open enumeration. Compare the utterance oӰԂ ԗӢӠӾԌӾԽ ӰԂ ӷԂԙԖӫ Ӡ ӹԙԽԂӰӾԽԙԋ ӰԂԌӹԊԄӹ pronounced with H*H in the word ӷԂԙԖӫ to the utterance oӰԂ ԗӢӠӾԌӾԽ ӰԂ ӪԂӰӠӪӫԊՂ, ԗӹԽӹԐ ӰԂ ԾӹӴՂӡ ԄӹԌ Ӡ ӰԂ ԌӾӰӸ ӢӹԞԌӾӰӠԋ pronounced with H*M in the word ӪԂӰӠӪӫԊՂ. In the first utterance, accent H*H announces a final accent, but with accent H*M the utterance can continue endlessly before a final accent occurs. This difference is extensively described in Odé (in press).

442

CECILIA ODÉ

name

label

utterance main commuboundary nicative function L% yes-no question

H*L

rise-fall

H*H

risehigh

%

H*M

risemid

%

L*

L%

HL*

lowlevel low-fall

L%

L*H

low-rise

%

other communicative functions

continuation; contrast; repeated wh-question, repeated question, alternative question; prominence (emphasis) incompleteness continuation: closed enumeration; exclamation; positive qualification incompleteness continuation: open enumeration; meditation; puzzled reaction; vocative: calling from a distance completeness neutral finality, answer, confirmation completeness emphatic answer, conwith emphasis firmation; whquestion with narrow focus; imperative, command; addressing non-first ellip- polemic answer; tic question summons; enumeration; incompleteness; imperative question

Table 1: Overview of names of symbols, labels, boundaries and communicative functions.

4. Pitch accent H*M on the ToRI website As a demonstration of how forms and functions of the six pitch accents are described on the ToRI website, pitch accent H*M is presented as it appears by pressing the button Pitch accentsՌaccent H*M. By pressing the play button Źon the pages (not visible on the pages below), examples can be listened to as often as a user desires.

TORI, AN INTERACTIVE RESEARCH TOOL

443

4.1 Form of pitch accent H*M Pitch accent H*M is a high rise coming from low or mid. It tends to be less steep and less high than the rises of H*L and H*H. The high ending is reached at the end of the accented syllable. After the accented syllable, the pitch drops to mid, after which it may drop further to the end of the utterance. For a stylized representation of accent H*M see section 3.6. Typically the mid-pitched post-tonic part is often stretched out over many syllables. H*M can also be followed by another H*M within the same utterance, so as to create the sawtooth pattern, or by some other pitch accent. In the examples, H*M occurs in words in a number of different positions of the utterance. 4.2 Functions of pitch accent H*M The main communicative function of this accent is incompleteness. Other functions are: continuation in narratives, continuation in an open enumeration, vocatives (calling from a distance). It may also express a puzzled reaction, a meditation. 4.3 Examples of pitch accent H*M The pictures below are made using Praat version 4.6.38 (Boersma and Weenink 2007). The star on the contour is aligned with the star of the symbol on the accented syllable above the text of the examples (see also section 3.2). The first two examples presented in Figures 1 and 2 are neutral realizations of accent H*M with the communicative function of incompleteness: the speaker has not yet finished.

Figures 1 and 2. Realizations of accent H*M expressing incompleteness.

The next three utterances illustrated in Figures 3-5 are examples of accent H*M in a continuation in a narrative. Note that in Figure 3 accent H*M occurs in an utterance-final syllable so there is no mid-pitched post-tonic part. Yet the accent is perceived as accent H*M for its late timing: the high target H* is reached late in the accented syllable. For a discussion on post-tonic parts “cut off” after

444

CECILIA ODÉ

pitch accents occurring in utterance-final syllables with high pitch target H* the reader is referred to Odé (2005).

Figures 3-5. Realizations of accent H*M expressing a continuation in a narrative.

The next two utterances presented in Figures 6 and 7 are also examples of continuations in a narrative. The mid-pitched post-tonic part is stretched out over many syllables (see section 4.1).

TORI, AN INTERACTIVE RESEARCH TOOL

445

Figures 6 and 7. Realizations of accent H*M in a continuation where the mid-pitched post-tonic part is stretched out over many syllables.

The next utterances illustrated in Figures 8-10 are examples of continuations in an open enumeration. In contrast to closed enumerations, expressed by accent H*H (see section 3.12), an enumeration with realizations of accent H*M can continue endlessly till a final accent occurs.

Figures 8-10. Realizations of accent H*M in an open enumeration.

The following two utterances are examples of accent H*M expressing a vocative: calling someone from a distance. The examples are given in Figures 11-13.

446

CECILIA ODÉ

Figures 11-13. Realizations of accent H*M expressing vocatives: calling someone from a distance.

The following utterance is a fragment from a longer recording with many subsequent realizations of accent H*M. The selected fragment with one example of such realization of H*M is presented in Figure 14. It is pronounced by a woman who tenderly speaks to her cat.

Figure 14. Realization of accent H*M expressing a continuation.

TORI, AN INTERACTIVE RESEARCH TOOL

447

The following two utterances in Figures 15 and 16 are examples of a meditation with the typical mid-pitched post-tonic part stretched out over many syllables.

Figures 15 and 16. Realizations of accent H*M expressing a meditation with the typical midpitched post-tonic part stretched out over many syllables.

The last example of accent H*M presented in Figure 17 is an utterance in which the unaccented words in the post-tonic part after the high target H* are realized on a level starting somewhat higher than the pitch target H* itself reached in the accented syllable Լa- in ԼaԼӫԶӪԂ, followed by the mid-pitched post-tonic part stretched out over many syllables. The utterance starts with a sawtooth pattern (see section 3.10) in the words ԽԂӪӠӾ ԌӾԽԂԊӠ.

Figure 17. Initial sawtooth pattern followed by a realization of accent H*M with a post-tonic part starting high and then stretched out over many syllables in the mid-pitched post-tonic part.

448

CECILIA ODÉ

4.4 Concluding remark In the same way as accent H*M presented in sections 4.1-4.3 above, the other five accents H*L, H*H, L*, HL* and L*H are presented on the ToRI webpages with audiovisual examples. The demonstrations of the pitch accents are followed by interactive audiovisual exercises training the recognition and production of the pitch accents (see section 1). University of Amsterdam REFERENCES Beckman, M.E., and G.M. Ayers 1997 Guidelines for ToBI transcription. Version 3.0. http://www.ling.ohio-state.edu/phonetics/E_ToBI (4-12-2007). Beckman, Mary E., Julia Hirschberg, and Stefanie Shattuck-Hufnagel 2005 “The original ToBI system and the evolution of the ToBI framework”. In: Sun-Ah Jun (ed.), Prosodic Typology – The Phonology of Intonation and Phrasing, 9-54. Oxford: Oxford University Press. Boersma, P., and D. Weenink 2007 Praat version 4.3.68: http://www.fon.hum.uva.nl/praat (2007). Bolinger, D.L. 1958 “A theory of pitch accent in English”. In: Word 14, 109-149. 1986 Intonation and Its Parts. Melody in Spoken English. London: Edward Arnold. 1989 Intonation and Its Uses. Melody in Grammar and Discourse. London: Edward Arnold. Gussenhoven, C. 2005 “Transcription of Dutch Intonation”. In: Sun-Ah Jun (ed.), Prosodic Typology, The Phonology of Intonation and Phrasing, 118-145. Oxford: Oxford University Press. ’t Hart, J., R. Collier, and A. Cohen 1990 A perceptual study of intonation. Cambridge: Cambridge University Press. Hirst, Daniel, and Albert Di Cristo (eds.) 1999 Intonation systems, A Survey of Twenty Languages. Cambridge: Cambridge University Press. Keijsper, C.E. 1983 “Comparing Dutch and Russian Pitch Contours”. In: Russian Linguistics 7, 101-154. 1987 “Studying Neoštokavian Serbocroatian Prosody”. In: A.A. Barentsen, B.M. Groen, R. Sprenger (eds.), Dutch Studies in South Slavic and Balkan Linguistics (Studies in Slavic and General Linguistics 10), 101-193. Amsterdam: Rodopi. 1992 “Recent intonation research and its implications for teaching Russian”. In: A.A. Barentsen, B.M. Groen, R. Sprenger (eds.), Studies in Russian Linguistics (Studies in Slavic and General Linguistics 17), 151-214. Amsterdam-Atlanta: Rodopi.

TORI, AN INTERACTIVE RESEARCH TOOL Odé, C. 1989 1992

2003a

2003b

2005

2007 in press

449

Russian intonation: a perceptual description (Studies in Slavic and General Linguistics 13). Amsterdam-Atlanta: Rodopi. “Perceptivnaia ekvivalentnost’ realizatsii intonatsionnykh konstruktsii E.A. Bryzgunovoi”. In: A.A. Barentsen, B.M. Groen, R. Sprenger (eds.), Studies in Russian Linguistics (Studies in Slavic and General Linguistics 17), 227-284. Amsterdam-Atlanta: Rodopi. “Developing a Transcription of Russian Intonation (ToRI)”. In: Proceedings of the 15th International Congress of Phonetic Sciences, Barcelona, August 3-9, 3097-3100. “Description and Transcription of Russian Intonation (ToRI)”. In: Jos Schaeken, Peter Houtzagers, Janneke Kalsbeek (eds.), Dutch Contributions to the Thirteenth International Congress of Slavists, Ljubljana (Studies in Slavic and General Linguistics 30), 279-288. Amsterdam-New York: Rodopi. “Neutralization or truncation? The perception of two Russian pitch accents on utterance-final syllables”. In: Cecilia Odé and Rob van Son (guest eds.), Speech Communication 47/1-2, 71-79. “Zametki o poniatii tonal’nyj aktsent na primere russkogo iazyka”. In: Problemy fonetiki V, 237-249. Moskva: Nauka. “Communicative functions and prosodic labelling of three Russian rising pitch accents”. In: Alexander Lubotsky, Jos Schaeken, Jeroen Wiedenhof (eds.), Evidence and Counter-evidence. Linguistic essays in honour of Frederik Kortlandt (Studies in Slavic and General Linguistics 32), Amsterdam-New York: Rodopi.

Sun-Ah Jun (ed.) 2005 Prosodic Typology, The Phonology of Intonation and Phrasing. Oxford: Oxford University Press. Sun-Ah Jun 2005 “Prosodic Typology”. In: Sun-Ah Jun (ed.), Prosodic Typology, The Phonology of Intonation and Phrasing, 430-458. Oxford: Oxford University Press.

SUMMARY This article discusses a new system for the Transcription of Russian Intonation, ToRI, on the Internet. Section 1 presents a general outline of the system. The terminology used in ToRI is defined in an online glossary, from which Section 2 gives the following examples: pitch accent and accent-lending pitch movements, perceptual equivalence, connecting pitch movements, and register. ToRI uses unambiguous symbols for the transcription of pitch accents, pitch movements connecting the pitch accents and utterance boundaries marked by pitch. Section 3 discusses these ToRI symbols in detail. In the system, each symbol representing a pitch accent is described with rules for its realization, that is, with phonetic correlates and limits of perceptual tolerance. Each pitch accent is presented with pictures of the contours, with sound examples and with interactive audiovisual exercises training the recognition and production of Russian pitch phenomena. In the examples, main communicative functions for each accent are also given. In Section 4, pitch accent H*M is presented as a demonstration of how the pitch accents are described and how they appear on the webpages of ToRI.

Dutch Contributions to the Fourteenth International Congress of Slavists, Ohrid: Linguistics (SSGL 34). Amsterdam – New York: Rodopi, 2008, 451-459.

FONOLOŠKI OPIS GOVORA POTOƀ V ZILJSKI DOLINI TIJMEN PRONK Potoǁe so najbolj zahodna slovenska vas v Ziljski dolini. V lingvistiǁni literaturi so znane po tem, da je Tine Logar fonološki opis govora Potoǁ podal v dveh ǁlankih, in sicer v letih 1968 in 1981. Prvi ǁlanek, “Vokalizem in akcent govora Potoǁ v Ziljski dolini”, daje predvsem kratek pregled razvoja splošnoslovenskih samoglasnikov ter razvoja splošnoslovenskega naglasa. Drugi ǁlanek je kratek fonološki opis za OԼԷӾԙԊԂӴԋӰԙӪӠӡ ԊӠӰԄӴӠԙԽӠԓӾԙӪӠӡ ԂԽԊԂԙ (OLA). Logar je svoje gradivo zbral konec šestdesetih letih v Potoǁah pri družini Hebein “pr Jàmrǁu”. 1 Od leta 2001 do leta 2006 sem sam zbiral nareǁno gradivo pr Jàmrǁu, kjer sem vsega skupaj preživel veǁ kot dva meseca. Moja glavna informatorka je bila gospa Johanna Hebein, rojena leta 1923 na Brdu. Brdo ima enak govor kot Potoǁe in je osrednja slovensko govoreǁa vas v bližini Potoǁ, kjer je tudi cerkev, v katero hodijo tudi prebivalci Potoǁ.2 ƀe govorimo o ziljskem nareǁju, ne moremo ne omeniti Ivana Grafenauerja, ki je leta 1905 objavil svoj izredno natanǁen ǁlanek z naslovom “Zum Accente im Gailthalerdialekte”. Kljub naslovu najdemo v njegovem ǁlanku podatke ne samo o naglasu. ƀlanek je zaklad nareǁnega gradiva, iz katerega lahko destiliramo fonološki sistem ziljskega nareǁja. Grafenauer se je rodil v Veliki Vasi, ki leži med Potoǁami in Brdom. Njegov materni jezik je torej enak maternemu jeziku Logarjevih in mojih informatorjev. Zato moramo Grafenaurjevi sodbi o fonološkem sistemu ziljskega nareǁja posvetiti vso možno pozornost. Trideset let po izidu Grafenauerjevega ǁlanka je Viktor Paulsen napisal svojo doktorsko disertacijo o ziljšǁini z naslovom “Lautlehre des slowenischen Gailtalerdialektes in Kärnten”. Paulsenova disertacija je ostala v tipkopisu, v katero pa je Paulsen na roko dopisal naglasna in druga diakritiǁna znamenja. Zaradi tega je v rokopisu precej napak. Kljub temu nam Paulsen prinaša veliko gradiva iz

1

Takrat so tam živeli gospod in gospa Hebein, njun sin Peter Hebein 1 z svojo ženo Johanno ter njuni trije otroci. Peter in Johanna Hebein še vedno živita pr Jàmrǁu in se dobro spomnita, kako je Tine Logar domaǁinom pomagal pri delu na polju. 2 Gospa Hebein je z vztrajno navdušenostjo in s potrpežljivostjo odgovarjala na vsa moja vprašanja. Iskreno se ji zahvaljujem, da me je nauǁila naš ž‫׀‬փbar, še posebej v teh ǁasih, ko je v ostalo le še nekaj njegovih govorcev.

452

TIJMEN PRONK

celotne Ziljske doline z obsežno diahrono razpravo. Paulsenova disertacija je zato najbolj pomemben vir podatkov o drugih govorih v Ziljski dolini. Pri tem moram še dodati, da je pred kratkim Karmen Kenda-Jež objavila podroben fonološki opis govora Ovǁje vasi v Kanalski dolini v Italiji, ki je tudi ziljski govor. Fonološki sistem tega govora ima veliko skupnih pojavov s sistemom govora Potoǁ. V tem prispevku bo podan sinhroni opis fonološkega sistema potoškega govora na podlagi Logarjevih ǁlankov. Logarjevi razpravi sta tako kot vse njegovo dialektološko delo izvrsten zgled za mlajše generacije slovenskih dialektologov. Njegovi razpravi sta pravi zaklad podatkov o potoškem govoru in sta mi zelo magali, ko sem prviǁ prišel v Potoǁe. Kljub temu pa Logarjev fonološki opis ni popoln. Dopolnil ga bom s svojimi podatki, in sicer bom v primerih, kadar gradivo interpretiram drugaǁe, predlagal alternativno interpretacijo. Osnova za zasnovo tega prispevka bo Logarjeva razprava za OLA. Zapisovanje naglasov bo drugaǁa kot pri Logarju, saj bom v primerih uporabljal tradicionalno notacijo z akcenti , , in namesto za rastoǁi naglas, za padajoǁi naglas in < : > za dolžino v skladu s transkripcijo OLA. 1. Inventar 1.1 Vokalizem Dolgi naglašeni vokalizem i:

u: ‫־‬:

‫׀‬: e:

o: a:

Padajoǁi diftongi, vedno naglašeni i:՗

u:՗

Rastoǁi diftongi, vedno naglašeni i՗

u՗

Kratki naglašeni vokalizem in nenaglašeni vokalizem v nekonǁnem in v konǁnem zaprtem zlogu i

u ‫־‬

՗ e

‫׀‬ o

a

FONOLOŠKI OPIS GOVORA POTOƀ V ZILJSKI DOLINI

453

Nenaglašeni vokalizem v konǁnem odprtem zlogu ‫־‬

‫׀‬ e

o a

Funkcijo naglašenih in neglašenih silabemov imata tudi /‫׮‬/ in /n֕/, mogoǁe tudi /r֕/, ǁe ni sestavljen iz zaporedja /՗/ + /r/, kar bom zapisoval kot . ƀeprav menim, da fonološke razlike med /u/ in /w/ ter /i/ in /j/ ni (veǁ o tem kasneje v §2.1), bom zaradi berljivosti pisal in , kjer ima Logar in , torej tudi pri rastoǁih diftongih. Fonološko /gu՗ւra/, torej pišem . K Logarjevemu opisu moram še dodati, da je fonem /e/ v izgovoru precej visok in bližji slovenskemu knjižnemu /‫־‬/. Zvišanje izgovorjave fonema /e/ je verjetno povzroǁil rastoǁi diftong /i՗/, ki je nastal iz nekdanjih kratkih *e, *LJ in *Nj v zadnjem besednem zlogu in iz kratkega *e ali *LJ po umiku naglasa s konǁnega kratkega zloga. Samoglasniški del tega diftonga je fonetiǁno nižji od fonetiǁnega refleksa fonema /e/ in je pritisnil fonetiǁni refleks fonema /e/ proti fonetiǁnemu refleksu fonema /‫־‬/. Pri zadnjih samoglasnikih najdemo podobno situacijo, ǁeprav je razlika med /o/ in /‫׀‬/ malo veǁja. Zgoraj navedeni sistem se od Logarjevega sistema razlikuje v tem, da ima osem kratkih samoglasnikov proti petim kratkim naglašenim in šestim kratkim nenaglašenim samoglasnikom pri Logarju (1981). Logar tudi nima kratkih diftongov, ki sta pri njem sestavljena iz /w/ ali /j/ s sledeǁim samoglasnikom. Ne morem se strinjati z Logarjevo interpretacijo kvantitetnega nasprotja v govoru Potoǁ. Logar pravi: “Kratki naglašeni samoglasniki so v potoškem govoru v glavnem omejeni samo na zadnje ali edine besedne zloge.” (1968: 209). Po njegovem mnenju so se dolgi samoglasniki praviloma skrajšali “pred dvoǁlensko soglasniško skupino, pri ǁemer mora biti prvi soglasnik zvoǁnik, drugi pa nezvoǁnik.” (ibidem). Grafenauer pa meni, da je skrajšanje splošno pred dvoǁlenskimi soglasniškimi skupinami. V glavnem menim, da ima prav Grafenauer.3 Po mojem mnenju obstaja fonološko kvantitetno nasprotje v vseh naglašenih zlogih. V nekaterih besedah je namreǁ nastal novi kratki samoglasnik v odprtem zlogu: t‫֎׀‬t‫‘ ־‬ta’, zìgnt‫־‬, ǁզr‫‘ ־‬kjer’. V zaprtih zlogih tudi lahko najdemo dolge samoglasnike, ǁeprav je število primerov majhno: páswa, mեtra, tóšnՖw (rod. mn.) ‘torba’, kwádwo ‘kladivo’ i.t.n. Sem spada mogoǁe tudi náglna (rod.), toda rբglca. Ker ni primerov z -l՗- v odprtem ponaglasnem zlogu, bi drugo besedo lahko analizirali kot /ragl՗ca/, prvo pa kot /na:g‫׮‬na/. ƀe to dopustimo v tem in v dru3

Logar sam pravi da je “zapisovanje kvantitete ena najtežjih in tveganih stvari pri vsem dialektološkem delu na terenu.” (ibidem). Za svoje delo na terenu v Ziljski dolini bi lahko rekel, da je bilo zapisovanje kvantitete najtežje in da je trajalo kar nekaj let, da sem se jo nauǁil razlikovati.

454

TIJMEN PRONK

gih primerih, pred soglasniškimi skupinami v nezadnjem zlogu ni dolgih samoglasnikov (to se pravi, da kvantitetna opozicija ne obstaja). Grafenauer navaja dva izjemna primera z dolgim samoglasnikom v nezadnjem zlogu pred soglasniško skupino: rod. mástLj, im. mգst in rod. m‫׀‬փsta ob mխ՗sta, im. mխ՗st. Pri moji informatorki na terenu gospe Johanni Hebein je v teh dveh oblikah jasno slišati kratek samoglasnik: m‫׀‬ւsta, màst‫־‬. Logar za to, kar imenujem “rastoǁi diftongi” (t.j. /u՗/ in /i՗/), piše /wo:/ ali /wa/ in /je:/. Po njegovem mnenju imata dvofonemsko vrednost. Fonološke razlike med /wo:/ in /wa/ nisem opazil. Logar piše gwàra za gw՗ւra, toda to ne drži, saj se samoglasnik fonetiǁno razlikuje od naglašenega samoglasnika v besedi gwբnjati ‘odganjati’. Prim. tudi samoglasnik v besedi zgw՗ւnc ‘zvonec’, ki se razlikuje od samoglasnika v kwànc ‘klanec’, ǁeprav sta si fonetiǁno blizu. Ne strinjam se z Logarjem, da je samoglasniški del rastoǁih diftongov dolg. Grafenauer na primer piše in , Paulsen pa pravi, da sta diftonga /oa/ in /ea/ kratka na Brdu (1935: 38), ampak v nezadnjem zlogu dolga v bolj vzhodnih govorih (idem: 77). Samoglasniški del je fonetiǁno bolj centraliziran od /o/ in /e/, samoglasniški del sprednjega diftonga pa je pri tem fonetiǁno nižji od /e/. Po mojem mnenju je kratek samoglasnik v teh diftongih najbolj verjetno refleks polglasnika, ki ga tudi najdemo samostojno, naprimer v pՖ֎s, mՖ֎š ‘miš’ itn. Po tej interpretaciji je razlika med “kratkimi” diftongi (u՗ւ/u՗֎, i՗ւ/i՗) in “dolgimi” diftongi (ú՗/խ՗, í՗/է՗) pravzaprav razlika med padajoǁima diftongoma ter rastoǁima diftongoma. Kvantitetne opozicije pri diftongih ni, ǁeprav je samoglasniški del padajoǁih diftongov vedno fonetiǁno skrajšan pred soglasniškimi skupinami v nezadnjem zlogu. Našel sem en primer skrajšanega diftonga v odprtem zlogu, namreǁ pլ՗še ‘potlej’, toda to je pogosto uporabljana allegro varianta besede pխ՗šle, fonetiǁno [pu՗šlٰ] (prim. 'potle zraven pó:tle v Ovǁji vasi, Kenda-Jež 2005: 91). Rastoǁa diftonga sta vedno kratka, ker je naglašeni del diftonga refleks polglasnika, ki je vedno kratek. Skupini -w՗- ter -j՗- sta sovpali s padajoǁima diftongoma /u՗/ in /i՗/, ǁe sta prišli pod naglas po umiku naglasa s konǁnega kratkega zloga: ú՗k՗t ‘laket’ (’՗ւk՗t), hú՗dna ‘hladna’, sú՗d՗k ‘sladek’ (Grafenauer s’՗ւd՗k), í՗spa ‘izba’, namenilnik í՗skat in verjetno tudi dú՗r, rod. dú՗ra < *dw՗rբ (manj verjetno = štokavsko dvթr, dvóra, tako Grafenauer 1905: 200). 1.2 Konzonantizem Zvoǁniki r

l

m n

Logar loǁi tudi /w/ in /j/, ki sta po mojem mnenju fonemske variante samoglasnikov /u/ in /i/ v položaju pred ali za samoglasniki. Ni položaja, kjer [u] in [w] ter [i] in [j] niso v komplementarni distribuciji. Poleg tega je na zaǁetku stavka

FONOLOŠKI OPIS GOVORA POTOƀ V ZILJSKI DOLINI

455

slišen glotalni zapornik, tudi ǁe se pojavlja fonetiǁni /w/: wàǁna [‫ٳ‬waǁna] ‘laǁna’, j՗ւz՗k [‫ٳ‬jٰz՗k]. Pri tem mislim, da se diftongi in ter in razlikujejo samo v mestu intenzitetnega viška, o ǁem gl. zgoraj. Logar loǁi še platalni [ǝ], ki pa je po mojem mnenju bolj verjetno sestavljen iz fonemov /n/ pa /i/, na kar bi kazalo to, da so samoglasniki pred njima vedno skrajšani: ž՗blènje, p՗šlզnj‫־‬, sànje. Isto velja za skupino /r/ + /j/: šlpír ‘slepiǁ’, tož. šlpìrja, kúra, prid. kլrja. Nezvoǁniki p t c ǁ k

b d

g

f s š h

z ž

Inventar nezvoǁnikov je enak Logarjevemu inventarju, za njegov x pišem h. /b/ je spirantiziran [Փ] pred sprednjimi samoglasniki, zapornik [b] pred labialnimi soglasniki (tudi [w]) in lahko oboje v drugih položajih. /d/ je lahko spirantiziran [Dž] ali zapornik [d]. /g/ je redkokdaj spirantiziran. 1.3 Prozodija Naglas je svoboden in je lahko kratek ali dolg, akutiran ali cirkumflektiran v vseh (torej tudi v konǁnih) zlogih. Vsi nenaglašeni samoglasniki so kratki, ǁe ne štejemo zaporedja dveh enakih samoglasnikov na koncu besede (-aa ali -oo) kot en dolgi samoglasnik: v pridevnikih síraa ž. im. ed., síroo s. im. ed. od sír՗w, ter pogosto pri glagolih v obliki 3. osebe mn. sedanjika: ràstoo, pétoo ‘molijo’, prim. péto (tož. od péta ter števnik), in ž. ed., m. dv. in s. ed. preteklika: émaa, pétaa, rst՗ւrgoo. Cirkumfleks, dolg in kratek, je toniǁno visok, komaj padajoǁ. Akut, dolg in kratek, je toniǁno nizek, ne rastoǁ, in se realizira tudi na ponaglasnem zlogu kot toniǁni višek (Neweklowsky 1973: 244f.). Podobno analizo je podal Peter Jurgec za naglas govora Ovǁje vasi v Kanalski dolini (2005: 69). Višek akuta nikoli ni veǁ kot dva zloga proti koncu besede od naglasnega mesta: ž‫־‬ւgnanèga, j՗gnéǁab՗ւna ‘jagnjeǁevina’ (tako že Grafenauer 1905: 222). Klitike lahko dobijo toniǁni višek prejšnje besede: néne sè ‘vneme se’. Pravilo je torej: toniǁni višek akuta je dva zloga proti koncu besedne skupine od intenzitetnega viška; ǁe se pojavlja samo en ponaglasni zlog, je toniǁni višek na le-tem

456

TIJMEN PRONK

2. Distribucija 2.1 Vokalizem V položaju pred [j] diftong /i:՗/ in samoglasnik /i/ sovpadeta v [i]: íjo ‘jejo’, sէjan ‘sejem’. V položaju pred [w] diftong /u:՗/ in samoglasnik /u/ sovpadeta v [u]: mխwa ‘mogla’, mխ՗gu ‘mogel’, n՗ kխwo ‘na kolo’, prpխnan ‘dopoldne’, šխwa ‘šola’, mn. šխ՗le. V položaju pred neistozložnim [w], [Փ] ali [o] se rastoǁi diftong /u՗/ ne govori. Sovpal je z dolgim /‫׀‬/: g‫׀‬փwa ž. im. ed., g‫׀‬փo ž. tož. ed., gw՗ւle ž. tož. mn., n‫׀‬փwa ž. im. ed., n‫׀‬փbe ž. tož. mn. /o/ je sovpal z /‫׀‬/ v [‫ ]׀‬pred neistozložnim [w] ali [j]: dl‫׀‬փwe ž. im. mn., dl‫׀‬փj‫ ־‬m. im. mn. Pri sprednjih samoglasnikih najdemo podobno situacijo. /e/ je sovpal z /‫־‬/ v [‫ ]־‬pred neistozložnim [j]: pw‫־‬փjo ‘upelo’, pwél‫־‬, z‫־‬փja ‘vzela’ (Logar ima napaǁno zé:jà), zél‫־‬. Za položaj pred neistozložnim [w] ni primerov. Rastoǁi diftong /i՗/ se ne govori pred neistozložnim [j]. Sovpal je z dolgim /e/, tako tudi Logar 1968: ǁrnéja, kndéja, prեja, prim. Grafenauer mé‫כ‬a, d՗žé‫כ‬a. V razpravi za OLA pa Logar zapiše dvakrat ǁ‫־‬փ:jò za ǁéjo.4 Logar pripomni, da sta /i՗/ in /i/ sovpala pred istozložnim /r/: sí:՗rk, p՗pí:՗r. Izgleda, da tudi pred neistozložnim /r/ ni razlike. Logar piše šǁí:ra ‘sekira’, Paulsen ima pa šǁí՗ra, ampak síra, p՗stէr (v mojem gradivu p՗stír) (idem: 58f., za obliko p՗stէr gl. 183). Urban Jarnik v enem od svojih pisem Janezu Primcu navaja obliko sbierat nam. ‘zbirat’ (Pronk 2007). Grafenauer navaja dì՗rjatLj ‘dirjati’, toda sìrk ‘koruza’ namesto *sí՗rk. Obstaja možnost, da je dì՗rjatLj sekundarna oblika po tipu lí՗tat‫־‬, prim. p՗ší՗lat‫‘ ־‬pošiljati’. V mojem gradivu sta /i՗/ in /i/ pred vsakim /r/ sovpala v /i/. V nobenem primeru nisem slišal sklopa *-i՗r-: sìrk, p՗stír, zbírat‫־‬, b՗ǁզrja ‘veǁerja’, mէrn ‘merim’, pret. m. ed. m‫־‬փr‫׀‬, mírn ‘zidam’, pret. m. ed. mír‫׀‬. Kratki naglašeni samoglasniki so v konǁnih zlogih veǁzložnih besed redki, a se vseeno da najti nekaj primerov: s՗lբn ‘vselej’, p՗ž՗brդnǁ ‘klepetav’, p՗tr՗֎h ‘sod’, rod. p՗tríha, p՗b‫֎׀‬w ‘pobil’. Kratki rastoǁi diftong je manj redek v konǁnem zlogu: s՗n՗ži՗֎t, s՗m՗ti՗֎ž, p՗s՗di՗ւn ‘posajen’ (tudi p՗s՗d‫־‬փn), š՗rw՗֎k (tudi šw՗ւrak). Logar navaja nekaj primerov za kratki naglašeni samoglasnik v zadnjem odprtem zlogu, ki, po mojem mnenju, nimajo naglasa na zadnjem zlogu: j՗ւt‫‘ ־‬iti’, Logar j՗t‫־‬ւ; b՗ւs‫־‬, Logar Փ՗s‫־‬ւ; f՗ւž՗w, Logar f՗žòu֛; cb՗ւt՗w, Logar cu֛՗tòu֛; ú՗g՗w ‘lagal’, Logar u֛՗՚òu֛; m՗ւne, Logar m՗nè; ǁ՗ւrb՗ǁ ‘ǁrv’, Logar ǁ՗rՓ՗ւǁ itn. Vzrok razliǁnih intepretacij je verjetno fonetika akuta. Kot je bilo že reǁeno zgoraj, toniǁni višek akuta v potoškem govoru ni na naglašenem samoglasniku. To lahko da poslušalcu vtis, da je zadnji zlog naglašen, toliko bolj, ǁe je samoglasnik v naglašenem zlogu polglasnik. Da je v teh besedah naglas na polglasniku, vemo zato, ker je v tem 4

Preklicujem svoje mnenje, da je /‫־‬/ navadni refleks diftonga /i՗/ pred neistozložnim /j/ (tako Pronk 2007: 115).

FONOLOŠKI OPIS GOVORA POTOƀ V ZILJSKI DOLINI

457

zlogu intenzitetni višek. Tudi Grafenauer ima v teh primerih, ki jih omenja v svojem ǁlanku, vedno naglas na polglasniku v prvem zlogu: m՗ւndž, j՗ւtLj itn. Kratki naglašeni /a/, /՗/, /e/ in /‫־‬/ sovpadajo v [e] pred istozložnim [j]: sjáti ‘sejati’, sjդj 2. os. ed. vel., krդj, im. mn. kráj‫־‬, m‫־‬փt‫־‬, mդj 2. os. ed. vel., pէ՗ǁ im. ed., pèjǁ‫ ־‬mest. ed. V ponaglasnih besednih zlogih sovpadajo vsi samoglasniki pred istozložnim [j] v samoglasnik, ki je malo bolj centraliziran od [e]. Zapisujem ga z : fխ՗trati ‘krmiti’, 2. os. ed. vel. fխ՗tr՗j, fխ՗tr՗jnje ‘krmite’. Kratki naglašeni /o/, /‫׀‬/, /՗/ in /u՗/ sovpadajo v [‫ ]׀‬pred istozložnim [w]: w‫׀‬ւwšla ‘jelša’, p՗b‫֎׀‬w (za *pob՗֎’), w‫֎׀‬w ‘vol; vil’, mn. w՗ւl‫‘ ־‬volovi’, bؔwa, bըsta. Logar ga zapisuje z : nòu֛, Փwòu ‘ubil’, wóu֛can, Grafenauer pa ima ׀‬: š‫‘ ’֎׀‬šel’, u֛‫׀‬ւws ‘oves’, grm‫׀‬ւwla ‘mravlja’. Ni pomembno, kako ga zapisujemo, ker opozicije ni; sam ga zapisujem z ׀‬, ker je refleks najbližje fonetiǁnemu [‫]׀‬. Navedeni samoglasniki ne sovpadajo z /a/, ki ima nižji refleks: kբwc ‘tkalec’, zrբw, z՗kwբw, tràwn՗k. V ponaglasnih besednih zlogih sovpadajo vsi samoglasniki pred istozložnim [w]. Refleks je malo bolj centraliziran kot pod naglasom, zato za zapisujem z : pw՗ւtk՗w ‘podkev’, sէ՗d՗w, sín՗w, ǁís՗w ‘kisel’, gw՗ւr՗w, lí՗sk՗wc. Logar ima v tem položaju : Փrá:tròu֛, ží:Փòu֛ ‘živel’, lí:՗skòu֛c, Grafenauer piše : bèazdհwc ‘bezeg’, pòakrհw. V konǁnem nenaglašenem odprtem zlogu se /i/ in /‫־‬/ ne razlikujeta. Sovpadata v samoglasnik, ki je fonetiǁno med [i] in [‫]־‬. Grafenauer ga zapisuje z (zmrւzn֕t‫־‬, pèst‫ ־‬itn.), tako tudi Logar v prvem ǁlanku (mí:šLj). V razpravi za OLA pa ga zapisuje z ־‬sí:n‫)־‬, tako tudi Paulsen (k՗dít‫)־‬. Ker posebnega fonema /Lj/ ni, se mi zdi bolje, ǁe sledimo Logarja v OLA in ga zapisujemo z ־‬, ǁeprav ni manj pravilno, ǁe ga zapisujemo z . /u/ in /‫׀‬/ v konǁnem nenaglašenem odprtem zlogu sovpadata v samoglasnik, ki je fonetiǁno med [u] in [‫]׀‬, zapisujem ga ׀‬: n‫־‬ւhtam‫‘ ׀‬nekomu’, krգd‫׀‬, zrj՗ւb‫‘ ׀‬žebelj’. Logar še omenja, da sta konǁna /e/ in /o/ ozka ter da se bližata -‫ ־‬in -‫׀‬. Po mojem vtisu se z njima še nista zlila. V ponaglasnih nezadnjih zlogih in v zadnjih zaprtih zlogih se praviloma pojavljata samo /՗/ in /a/, v prednaglasnih zlogih samo /՗/. Izjeme so izposojenke (ib՗rfòlwal‫‘ ־‬obropali’ < nem. überfallen, b‫־‬tժnca ‘vreme’ < nem. Wetter + *-nica), zloženke (nխ՗gobբrt ‘pogosto’) ter nekaj posameznih besed (Zilán‫־‬, imեne mn., štirեd‫‘ ־‬štirideset’). Pred neistozložnim [j] se praviloma pojavlja samo [i] v nenaglašenih zlogih: bէd՗mo ‘vidimo’, 3. os. mn. bէdijo, zlխ՗d՗j, daj. ed. zlխ՗dij‫׀‬. Pred neistozložnim [w] se praviloma pojavlja samo [u] v prednaglasnih zlogih: kuwáǁ. V ponaglasnih zlogih se je ta [u] izgubil: práb՗ti, ž. ed. pret. prábwa, 1. os. mn. sed. práb՗mo, 1. os. dv. prábwa. 2.2 Konzonantizem Nekdanje omejitve v distribuciji soglasnikov so izginile zaradi analogije, kasneje prevzetih izposojenk ter fonoloških sprememb. Logar trdi, da /w/ pred /i/, /i:՗/,

458

TIJMEN PRONK

/‫־‬/ ali /e/ ni mogoǁ, da velarov ni pred sprednjimi samoglasniki, da se /s/ ne pojavlja v položaju pred /l/, pa da se /d/ ne pojavlja med /r/ in /z/ (zrբw). Tu gre bolj za opis fonoloških sprememb kot za opis sinhronega stanja. Vsak /w/ je pred omenjenimi samoglasniki postal /b/: n‫֎׀‬w, m. im. mn. n‫׀‬փb‫־‬, krգw rod. mn., krábe im. mn., prբw, p՗rbíca. Iz drugih izvirov pa je nastal novi [w] pred sprednjimi samoglasniki: wígo ‘igo’ < *vԔ igo, wét‫‘ ־‬ujeti’, wéje ‘oje’ < *w՗jդ < *ojդ (?), prim. obirsko wղփ:jٰ, wèjža ‘veža’ < *w՗jžբ < *vežբ, pwèjǁ‫‘ ־‬v peǁi’ < *vԔ peǁզ, prim. n՗ pèjǁ‫־‬, pwéne ‘vpne’ itn. Podoben pojav najdemo pri prehodu fonema /s/ v /š/ pred /l/ (šlíšat‫־‬, pխ՗šle, šlէ՗p, šlùžba). Nova skupina /sl/ je nastala v besedi sl‫׀‬ւwza ‘solza’ ter v oblikah glagola ‘sleǁi’: slէ՗ǁan 1. os. ed. sed. V izposojenkah najdemo veliko primerov velara pred sprednjimi samoglasniki: g‫־‬փpl, rod. g‫־‬փplna iz nem. Göpel, hէn ‘mrtev (o živalih); pokvarjen’, kèksi ‘piškoti’ itn. Nov /d/ se pojavlja med /r/ in /z/ v glagolu rzdrէ՗š՗t‫‘ ־‬razrešiti’. Druga pravila distribucije, ki jih Logar opisal, še veljajo, na primer da se vzglasni /ž/ ne pojavlja v položaju pred /r/: zr‫־‬փbe, zrbíca, zrj՗ւb‫׀‬, da se zveneǁi zvoǁniki govore tudi pred pavzo ter da govor ne pozna vzglasnega [w] v položaju pred soglasnikom. Univerza v Leidenu LITERATURA Grafenauer, Ivan 1905 “Zum Accente im Gailthalerdialekte”. V: Archiv für slavische Philologie 27, 195228. Jurgec, Peter 2005 “Fonetiǁni opis govora Ovǁje vasi”. V: Nataša Komac, Vera Smole (ured.), Ovǁja vas in njena slovenska govorica - Valbruna e la sua parlata slovena, 6070. Ukve-Ljubljana. Kenda-Jež, Karmen 2005 “Fonološki opis govors Ovǁje vasi”. V: Nataša Komac, Vera Smole (ured.), Ovǁja vas in njena slovenska govorica - Valbruna e la sua parlata slovena, 85104. Ukve-Ljubljana. Logar, Tine 1968 “Vokalizem in akcent govora Potoǁ v Ziljski dolini”. V: Zbornik za filologiju i lingvistiku 11, 137-143. 1981 “Potoǁe (Potschach; OLA 146)”. V: Filipoviǀ, Nedim (ured.), Fonološki opisi srpskohrvatskih/hrvatskosrpskih, slovenaǁkih i makedonskih govora obuhvaǀenih Opšteslovenskim lingvistiǁkim atlasom, 201-211. Sarajevo. Neweklowsky, Gerhard 1973 Slowenische Akzentstudien : akustische und linguistische Untersuchungen am Material slowenischer Mundarten aus Kärnten, Wien. Paulsen, Viktor 1935 Lautlehre des slowenischen Gailtalerdialektes in Kärnten, neobjavljena disertacija, Wien.

FONOLOŠKI OPIS GOVORA POTOƀ V ZILJSKI DOLINI Pronk, Tijmen 2007

459

“Gailtal Slovene in Urban Jarnik’s letters to Primic and Kopitar, 1811-1814”. V: Marko Snoj, Marc L. Greenberg (ured.), Slovenski jezik - Slovene Linguistic Studies 6, 111-132.

SUMMARY The present article is a discussion of the phonology of the Slovene dialect of Potoǁe (German Potschach) in the Gailtal valley in Austria. The description is based on fieldwork material gathered between 2001 and 2006. The findings of the author will be compared to observations and material in earlier descriptions by Ivan Grafenauer and Tine Logar, with special focus on the vowel system and contrastive vowel length.

Dutch Contributions to the Fourteenth International Congress of Slavists, Ohrid: Linguistics (SSGL 34). Amsterdam – New York: Rodopi, 2008, 461-488.

ѢѣҋҘҫһѻҘҫ үҘҋѨҫҴ ҫ ҭҩѨһүҐүѷѣһѻѢѳҤ-үѢѷҫѢѨҴүҤ: ҘҫѢѷҩүҩҐѰҴѣѢѷѳ Ѩ҈ ҩѰҋҋҘҫҐҫ Ѥ҈ѳҘү ѣһѣѢү ѷҩѨҷѰҹѨѢѨѢү

1. ғҡҲҏҰѹҲѦґҧӁӃѦ ӉҽѸѦҍҽӁѹҾ ѢҽҡҲҸҾӂҽҾ ҡҲҽҲҧҾ ҰҸҡӊҾӂѦӁҽ ҡѦѸҽӁҲѹҔѦ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ҡҔҽґҾҝӁҸ-ҽӁҲҸӁѹѸѹҍѦҡҔҸҊҸ ҔҸѸҰґѦҔҡҽ, ҲҽҔѹѽ ҔҽҔ ԼӹԊӸԶӹӡ, ӴՂԙӹӪӠӡ, ԖӹӢӹԶӠӡ ѹ ԌӾԶӾӴՂӡ. үҦӆѦҔҲѹӊӁӃѦ ҡґҸӊҽ ҙҲҸҊҸ ҲѹҰҽ ҰҸҰҽҦҽӀҲ ҰҸҦ ҦҽӁӁҸѦ һҽҮҸӁӉҸѸ ҸҰҝѦҦѦґѦӁѹѦ ҰҝҸҲҸҲѹҰѹҍѦҡҔѹѽ ҽӁҲҸӁѹѸҸӊ (Lyons 1969, 1977). ҫӁѹ ҸӉӁҽҍҽӀҲ ҡҲѦҰѦӁѹ ҰҝѹӉӁҽҔҽ, ӊ қҸґҧҗѦҮ ѹґѹ ѸѦӁҧҗѦҮ ҡҲѦҰѦӁѹ ҸҲҔґҸӁҾӀӂѹѦҡҾ ҸҲ ҲҸҍҔѹ ҝѦӍѦҝѦӁ҃ѹѹ, ҝҽҡҰҸґҽҊҽӀӂѦҮҡҾ ӊ ҃ѦӁҲҝѦ ҊҝҽҦҽ҃ѹҸӁӁҸҮ җҔҽґӃ. ғ ҡѹґҏ ҡӊҸѦҮ ҊҝҽҦҏҽґҧӁҸҡҲѹ, ҸӁѹ ҸқҝҽӉҏӀҲ ӍҸҝѸӃ ҡҲѦҰѦӁѦҮ ҡҝҽӊӁѦӁѹҾ ѹ ҸҰҝѦҦѦґҾӀҲҡҾ ӁҽҝѦҍѹҾѸѹ ѸѦҝӃ ѹ ҡҲѦҰѦӁѹ. ѩҽӁӁӃѦ ҰҝѹґҽҊҽҲѦґҧӁӃѦ ѹӉӊѦҡҲӁӃ ҲҽҔңѦ ҔҸӁҲҝҽҝӁӃѸ ҸҲӁҸҗѦӁѹѦѸ ѸѦңҦҏ ҍґѦӁҽѸѹ ҽӁҲҸӁѹѸѹҍѦҡҔҸҮ ҰҽҝӃ, Ҳ.Ѧ. ҸҲҝѹ҃ҽӁѹѦ ҸҦӁҸҊҸ ҰҝѹґҽҊҽҲѦґҧӁҸҊҸ ӁѦ ҡҽӁҔ҃ѹҸӁѹҝҏѦҲ ҏҰҸҲҝѦқґѦӁѹѦ ѦҊҸ ҽӁҲҸӁѹѸҽ, ҰҸҡҔҸґҧҔҏ ҡѦѸҽӁҲѹҍѦҡҔѹѦ ҡӍѦҝӃ ҽӁҲҸӁѹѸѹҍѦҡҔѹѽ ҡґҸӊ ҝҽӉҦѦґҾѦҲ ѸѦңҰҸґӀҡӁҽҾ ӉҸӁҽ (үҝҏҲӀӁҸӊҽ 1988, 1999, ѢѹҔҸґҽѦӊҽ 1983, ҭҸ҃ѦґҏѦӊҡҔѹҮ 1974, Croft & Cruse 2004, Cruse 1976, 1986, E.A. Klein 1980, Murphy 2003, Sapir 1944). ҩҏӉѹӁ (1994) ӁҽӉӃӊҽѦҲ ҲҽҔѹѦ ҰҝѹґҽҊҽҲѦґҧӁӃѦ ԄӢԂԌӫԂԊӸӰՂԐӠ, ҰҝҸҲѹӊҸҰҸҡҲҽӊґҾҾ ѹѽ ҰҝѹґҽҊҽҲѦґҧӁӃѸ ҙҲҽґҸӁӁӃѸ (ӁҽҰҝѹѸѦҝ, ӪӢԂԙӰՂӡ, ԊӠԐӹӰӰՂӡ, ӪӠԙԊՂӡ, ӪӢӫԄԊՂӡ). ҭҸ҃ѦґҏѦӊҡҔѹҮ (1974, 1977) ѹҡҰҸґҧӉҏѦҲ ҲѦҝѸѹӁ ӪԂԓӾԙԽӴӾӰӰӹ-ӹՁӾӰӹԓӰՂӾ ԗӢӠԊԂԄԂԽӾԊӸӰՂӾ, ҰҸҦҍѦҝҔѹӊҽҾ ҲѦѸ ҡҽѸӃѸ, ҍҲҸ ҡґҸӊҽ ҦҽӁӁҸҮ ҊҝҏҰҰӃ ӁѦ ӊӃҝҽңҽӀҲ ҡҔҸґҧҔҸ-ӁѹқҏҦҧ ҸҰҝѦҦѦґѦӁӁҸҊҸ ҰҽҝҽѸѦҲҝҽ, Ҳ.Ҕ. ѹѽ ҝѦҽґҧӁҸѦ ҡҸҦѦҝңҽӁѹѦ ҸҰҝѦҦѦґҾѦҲҡҾ ӊӉҽѹѸҸҦѦҮҡҲӊѹѦѸ ҡ ҸҰҝѦҦѦґҾѦѸӃѸ ҡҏӂѦҡҲӊѹҲѦґҧӁӃѸ. ғ ҽӁҊґҸҾӉӃҍӁҸҮ ҲҝҽҦѹ҃ѹѹ ҰҝѹґҽҊҽҲѦґҧӁӃѦ ѹӁҲѦҝѦҡҏӀӂѦҊҸ Ӂҽҡ ҲѹҰҽ ӁҽӉӃӊҽӀҲҡҾ ӹԽӰӹԙӠԽӾԊӸӰՂԐӠ (relative adjectives) ѹ ҰҝҸҲѹӊҸҰҸҡҲҽӊґҾӀҲҡҾ ҰҝѹґҽҊҽҲѦґҧӁӃѸ ԂԼԙӹԊԇԽӰՂԐ (absolute adjectives). ҶҲѹ ҦӊѦ ҊҝҏҰҰӃ ҸҲґѹҍҽӀҲҡҾ ҦҝҏҊ ҸҲ ҦҝҏҊҽ ҰҸ ҡҲѦҰѦӁѹ ӉҽӊѹҡѹѸҸҡҲѹ ѹѽ ҡѦѸҽӁҲѹҍѦҡҔѹѽ ҝѦҽґѹӉҽ҃ѹҮ ҸҲ ҔҸӁҲѦҔҡҲҽ (Chafe 1970, Katz 1972, Kennedy 2005, Kennedy & McNally 2005, Rips & Turnbull 1980, Vendler 1968). ҋҍѹҲҽѦҲҡҾ, ҍҲҸ ҔҸӁҔҝѦҲӁҸѦ ҡҸҦѦҝңҽӁѹѦ ҸҲӁҸҡѹҲѦґҧӁӃѽ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ӉҽӊѹҡѹҲ ҸҲ ҦѦӁҸҲҽҲҽ ҸҰҝѦҦѦґҾѦѸҸҊҸ ҡҏӂѦҡҲӊѹҲѦґҧӁҸҊҸ (ҡҝ. ԼӹԊӸԶӹӡ ԙԊӹӰ vs. ԼӹԊӸԶԂԋ ԐՂԶӸ). ҋҸҦѦҝңҽӁѹѦ ҽқҡҸґӀҲӁӃѽ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ңѦ, ҔҽҔ ҏҲӊѦҝңҦҽѦҲҡҾ, ӁѦ ӉҽӊѹҡѹҲ ҸҲ ҔҸӁҲѦҔ-

462

ѣһѣѢү ѷҩѨҷѰҹѨѢѨѢү

ҡҲҽ. ѷҽҔ, ӉѦґѦӁҸҊҸ ҔҏӉӁѦҍѹҔҽ ѸҸңӁҸ ӁҽӉӊҽҲҧ ӉѦґѦӁӃѸ ңѹӊҸҲӁӃѸ, ӁҸ қҸґҧҗҸҊҸ ҔҏӉӁѦҍѹҔҽ ӁѦґҧӉҾ ӁҽӉӊҽҲҧ қҸґҧҗѹѸ ңѹӊҸҲӁӃѸ, ѹӉ ҍѦҊҸ ҡґѦҦҏѦҲ, ҍҲҸ ԒӾԊӾӰՂӡ – ҰҝѹґҽҊҽҲѦґҧӁҸѦ ҽқҡҸґӀҲӁҸѦ, ҽ ԼӹԊӸԶӹӡ – ҸҲӁҸҡѹҲѦґҧӁҸѦ (Rips & Turnbull 1980: 147). ғӃҗѦҰѦҝѦҍѹҡґѦӁӁӃѦ ҲѦҝѸѹӁӃ ѹѸѦӀҲ ҝҾҦ ӁѦҦҸҡҲҽҲҔҸӊ. ѷѦҝѸѹӁ “ҊҝҽҦҏҽґҧӁҸѦ ҰҝѹґҽҊҽҲѦґҧӁҸѦ” (ҩҏӉѹӁ 1994) ҰҸҦҝҽӉҏѸѦӊҽѦҲ, ҍҲҸ ҲҸґҧҔҸ ҰҝѹґҽҊҽҲѦґҧӁӃѦ ҦҽӁӁҸҮ ҊҝҏҰҰӃ ҸҝѹѦӁҲѹҝҸӊҽӁӃ Ӂҽ җҔҽґҏ ѹ ӊӃҝҽңҽӀҲ ҊҝҽҦҏѹҝҏѦѸӃѦ ҰҸӁҾҲѹҾ. ҫҦӁҽҔҸ ҊҝҏҰҰҽ ҙҲҽґҸӁӁӃѽ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ҲҽҔңѦ ӊҔґӀҍҽѦҲ ӊ ҡѦқҾ ҊҝҽҦҏҽґҧӁӃѦ ѦҦѹӁѹ҃Ӄ. ѢҽҰҝѹѸѦҝ, ҰҝѹґҽҊҽҲѦґҧӁӃѦ ҃ӊѦҲҽ ѹ ӍҸҝѸӃ ѸҸҊҏҲ ҏҰҸҲҝѦқґҾҲҧҡҾ ӊ ҡҝҽӊӁѹҲѦґҧӁӃѽ ѹ ҰҝѦӊҸҡѽҸҦӁӃѽ ҔҸӁҡҲҝҏҔ҃ѹҾѽ ѹ ҸҰҝѦҦѦґҾҲҧҡҾ ӁҽҝѦҍѹҾѸѹ ѸѦҝӃ ѹ ҡҲѦҰѦӁѹ (Tribushinina 2006). ѷѦҝѸѹӁ “ҔҽҍѦҡҲӊѦӁӁҸ-Ҹ҃ѦӁҸҍӁҸѦ ҰҝѹґҽҊҽҲѦґҧӁҸѦ” ҸқӃҍӁҸ ѹҡҰҸґҧӉҏѦҲҡҾ ӊ қҸґѦѦ ҏӉҔҸѸ ҡѸӃҡґѦ ҦґҾ ҸқҸӉӁҽҍѦӁѹҾ ҰҝѹґҽҊҽҲѦґҧӁӃѽ, ӊӃҝҽңҽӀӂѹѽ Ҹ҃ѦӁҔҏ (ӁҽҰҝѹѸѦҝ, ԖӹӢӹԶӠӡ, ԗԊӹԖӹӡ, ӠӰԽӾӢӾԙӰՂӡ). ҭҝѹґҽҊҽҲѦґҧӁӃѦ ѹӁҲѦҝѦҡҏӀӂѦҊҸ Ӂҽҡ ҲѹҰҽ ӊҔґӀҍҽӀҲ ҔҽҔ Ҹ҃ѦӁҸҍӁӃѦ, ҲҽҔ ѹ ҰҽҝҽѸѦҲҝѹҍѦҡҔѹѦ ѦҦѹӁѹ҃Ӄ, ҡѦѸҽӁҲѹҔҽ ҔҸҲҸҝӃѽ ҸҝѹѦӁҲѹҝҸӊҽӁҽ Ӂҽ ҊҝҽҦҽ҃ѹҸӁӁҏӀ җҔҽґҏ, ӊҔґӀҍҽӀӂҏӀ ҰҸҦҗҔҽґӃ ҽӁҲҸӁѹѸѹҍѦҡҔѹѽ ҰҝѹӉӁҽҔҸӊ ѹ ѸѦңҰҸґӀҡӁҏӀ ӉҸӁҏ. ѷѦҝѸѹӁ “ҸҲӁҸҡѹҲѦґҧӁҸѦ ҰҝѹґҽҊҽҲѦґҧӁҸѦ” ҲҽҔңѦ ҰҝѦҦҡҲҽӊґҾѦҲҡҾ ӊ ҦҽӁӁҸѸ ҡґҏҍҽѦ ӁѦҏѸѦҡҲӁӃѸ, ҰҸҡҔҸґҧҔҏ ӊ ҝҏҡҡҔҸҾӉӃҍӁҸҮ ҲҝҽҦѹ҃ѹѹ ҸӁ ҸқӃҍӁҸ ѹҡҰҸґҧӉҏѦҲҡҾ ӊ ӉӁҽҍѦӁѹѹ, ҸҲґѹҍӁҸѸ ҸҲ ӊӃҗѦҸҰѹҡҽӁӁҸҊҸ ҏҰҸҲҝѦқґѦӁѹҾ ӊ ҽӁҊґҸҾӉӃҍӁҸҮ ӁҽҏҍӁҸҮ ґѹҲѦҝҽҲҏҝѦ. ҭҸҦ “ҸҲӁҸҡѹҲѦґҧӁӃѸѹ ҰҝѹґҽҊҽҲѦґҧӁӃѸѹ” ҸқӃҍӁҸ ҰҸӁѹѸҽӀҲҡҾ ҽҦӆѦҔҲѹӊӁӃѦ ҡґҸӊҽ, ӁҽӉӃӊҽӀӂѹѦ ҰҝѹӉӁҽҔ ҍѦҝѦӉ ҸҲӁҸҗѦӁѹѦ Ҕ ҰҝѦҦѸѦҲҏ, ҾӊґѦӁѹӀ ѹґѹ ҦѦҮҡҲӊѹӀ (ҡҝ. ҽӁҊґ. relational adjectives). ҭҝѹѸѦҝҽѸѹ ҾӊґҾӀҲҡҾ ҰҝѹґҽҊҽҲѦґҧӁӃѦ ӊ ҡґҸӊҸҡҸҍѦҲҽӁѹҾѽ ԗӢӹԐՂԶԊӾӰӰՂӡ ӪӹԐԗԊӾӪԙ, ӰԂӫԓӰՂӡ ԗӢӹԄӢӾԙԙ ѹ ԐӾԙԽӰՂӡ ԞӠԽӾԊӸ. ҶҲѹ ҰҝѹґҽҊҽҲѦґҧӁӃѦ ӁѦ ѹӉѸѦӁҾӀҲҡҾ ҰҸ ҡҲѦҰѦӁҾѸ ҡҝҽӊӁѦӁѹҾ ѹ ӁѦ ѸҸҊҏҲ ҏҰҸҲҝѦқґҾҲҧҡҾ ӊ ҰҝѦҦѹҔҽҲѹӊӁҸҮ ӍҏӁҔ҃ѹѹ. ҭҸ ҙҲҸҮ ҰҝѹҍѹӁѦ ѹѽ ҍҽҡҲҸ ҸҲӁҸҡҾҲ Ҕ ҰѦҝѹӍѦҝѹѹ ҽҦӆѦҔҲѹӊӁҸҮ ҔҽҲѦҊҸҝѹѹ (үҦѸҸӁѹ 1964). ҫҍѦӊѹҦӁҸ, ҍҲҸ ѹҡҡґѦҦҏѦѸӃѦ ӊ ҦҽӁӁҸҮ ҝҽқҸҲѦ ҡґҸӊҽ ӁѦ ҾӊґҾӀҲҡҾ ҸҲӁҸҡѹҲѦґҧӁӃѸѹ ӊ ҰҝѹӊӃҍӁҸѸ ҦґҾ ҝҸҡҡѹҮҡҔҸҮ ґѹӁҊӊѹҡҲѹҔѹ ҡѸӃҡґѦ ҙҲҸҊҸ ҡґҸӊҽ. ҫӁѹ ҸҲӁҸҡҾҲҡҾ Ҕ Ҕґҽҡҡҏ ҔҽҍѦҡҲӊѦӁӁӃѽ ҰҝѹґҽҊҽҲѦґҧӁӃѽ, ҔҸҲҸҝӃѦ ҡҸҡҲҽӊґҾӀҲ ҾҦҝҸ ҽҦӆѦҔҲѹӊӁҸҮ ҔҽҲѦҊҸҝѹѹ ӊ ҝҏҡҡҔҸѸ ҾӉӃҔѦ (үҔҸҦѦҡ 1987, ѢѹҔҸґҽѦӊҽ 1983, ҤҽҝѹҲҸӁҍѹҔ 1986). ғӊѹҦҏ ӊҡѦҊҸ ӊӃҗѦҡҔҽӉҽӁӁҸҊҸ, ӊ ҦҽӁӁҸҮ ҡҲҽҲҧѦ ҰҝѹґҽҊҽҲѦґҧӁӃѦ ѹӁҲѦҝѦҡҏӀӂѦҮ Ӂҽҡ ҔҽҲѦҊҸҝѹѹ қҏҦҏҲ ѹѸѦӁҸӊҽҲҧҡҾ ԙӪԂԊԋӢӰՂԐӠ ԂӰԽӹӰӠԐԂԐӠ. ҶҲҸҲ ҲѦҝѸѹӁ ҰҸҦҍѦҝҔѹӊҽѦҲ ҸҝѹѦӁҲҽ҃ѹӀ ѹѽ ҡѦѸҽӁҲѹҔѹ Ӂҽ ҊҝҽҦҽ҃ѹҸӁӁҏӀ җҔҽґҏ ѹ ӊ ҲҸ ңѦ ӊҝѦѸҾ ҰҝҸӊҸҦѹҲ ҝҽӉґѹҍѹѦ ѸѦңҦҏ ҡҔҽґҾҝӁӃѸѹ ҰҝѹґҽҊҽҲѦґҧӁӃѸѹ ҔҸӁҲҝҽҝӁҸҊҸ ҲѹҰҽ (ҡҸқҡҲӊѦӁӁҸ, ҽӁҲҸӁѹѸҽѸѹ) ѹ ҡҔҽґҾҝӁӃѸѹ ҰҝѹґҽҊҽҲѦґҧӁӃѸѹ ҔҸӁҲҝҽҦѹҔҲҸҝӁҸҊҸ ҲѹҰҽ (ӁҽҰҝѹѸѦҝ, ӹԙԽӢՂӡ : Խӫԗӹӡ; ԙӫԖӹӡ : ԐӹӪӢՂӡ). ѷҸґҧҔҸ ҰҽҝӃ ҡ ҔҸӁҲҝҽҝӁӃѸ ҡѦѸҽӁҲѹҍѦҡҔѹѸ ҸҲӁҸҗѦӁѹѦѸ қҏҦҏҲ ҸқӆѦҔҲҸѸ ҸҰѹҡҽӁѹҾ ӊ ӁҽҡҲҸҾӂѦҮ ҝҽқҸҲѦ.

ѢѣҋҘҫһѻҘҫ үҘҋѨҫҴ ҫ ҭҩѨһүҐүѷѣһѻѢѳҤ-үѢѷҫѢѨҴүҤ

463

2. ѢѦҔҸҲҸҝӃѦ ҰҸҡҲҏґҽҲӃ ҲѦҸҝѹѹ ҽҦӆѦҔҲѹӊӁҸҮ ҡѦѸҽӁҲѹҔѹ ѢѦҡѸҸҲҝҾ Ӂҽ ҝҽӉӁҸҸқҝҽӉѹѦ ҰҸҦѽҸҦҸӊ Ҕ ѹӉҏҍѦӁѹӀ ҡѦѸҽӁҲѹҔѹ ѹѸѦӁ ҰҝѹґҽҊҽҲѦґҧӁӃѽ, қҸґҧҗѹӁҡҲӊҸ ҡҏӂѦҡҲӊҏӀӂѹѽ ҲѦҸҝѹҮ ҝҽӉҦѦґҾӀҲ ӁѦҡҔҸґҧҔҸ қҽӉҸӊӃѽ ҰҸґҸңѦӁѹҮ Ҹ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ҡҔҽґҾҝӁҸ-ҽӁҲҸӁѹѸѹҍѦҡҔҸҊҸ ҲѹҰҽ. ғҸҰѦҝӊӃѽ, ҡҍѹҲҽѦҲҡҾ, ҍҲҸ ҰҸґҸңѹҲѦґҧӁӃѦ ӍҸҝѸӃ ҰҝѹґҽҊҽҲѦґҧӁӃѽ-ҽӁҲҸӁѹѸҸӊ ӊҡѦҊҦҽ ҸқҸӉӁҽҍҽӀҲ ҸҲҔґҸӁѦӁѹѦ ҸҲ ӁҸҝѸӃ (ҡҝѦҦӁѦҡҲҽҲѹҡҲѹҍѦҡҔҸҊҸ ҡҲҽӁҦҽҝҲҽ) ӊ қҸґҧҗҏӀ ѹґѹ ѸѦӁҧҗҏӀ ҡҲҸҝҸӁҏ (ҭҸ҃ѦґҏѦӊҡҔѹҮ 1974, ҩҏӉѹӁ 1994, Bierwisch 1967, 1989, Katz 1972, E.A. Klein 1980, Lyons 1969, 1977, Siegel 1980). ѷҽҔ, Ѧҡґѹ ҰҽҝҽѸѦҲҝӃ ҸқӆѦҔҲҽ ҰҝѦӊӃҗҽӀҲ ҡҝѦҦӁѹѦ ҝҽӉѸѦҝӃ ҦҽӁӁҸҊҸ Ҕґҽҡҡҽ ҸқӆѦҔҲҸӊ, ҲҸ ҏҰҸҲҝѦқґҾӀҲҡҾ ҰҝѹґҽҊҽҲѦґҧӁӃѦ қҸґҧҗҸҊҸ ҰҸґӀҡҽ, ҲҽҔѹѦ ҔҽҔ ԶӠӢӹӪӠӡ, ӴՂԙӹӪӠӡ ѹ ԌԊӠӰӰՂӡ. ѣҡґѹ ңѦ ҝҽӉѸѦҝӃ ҸҰѹҡӃӊҽѦѸҸҊҸ ҸқӆѦҔҲҽ ѸѦӁҧҗѦ ҡҝѦҦӁѦҊҸ, ҲҸ ҰҝѹѸѦӁѹѸӃ ҰҝѹґҽҊҽҲѦґҧӁӃѦ ѸҽґҸҊҸ ҰҸґӀҡҽ, ҲҽҔѹѦ ҔҽҔ ӫԒӪӠӡ, ӰӠԒӪӠӡ ѹ ӪӹӢӹԽӪӠӡ. ҭҸӁҾҲѹѦ ӁҸҝѸӃ ӁҽҡҲҸґҧҔҸ ҰҝҸҍӁҸ ӊҸҗґҸ ӊ ҲѦҸҝѹѹ ҽҦӆѦҔҲѹӊӁҸҮ ҡѦѸҽӁҲѹҔѹ, ҍҲҸ ҽқҡҸґӀҲӁҸѦ қҸґҧҗѹӁҡҲӊҸ ѹҡҡґѦҦҸӊҽҲѦґѦҮ ҰҝѹӁѹѸҽӀҲ ҙҲҸ ҰҸґҸңѦӁѹѦ ҔҽҔ қҽӉҸӊҸѦ ѹ ҡҲҝҸҾҲ Ӂҽ ӁѦѸ ҡӊҸӀ ҲѦҸҝѹӀ. ҋҏӂѦҡҲӊҏѦҲ ґѹҗҧ ҸҍѦӁҧ ӁѦқҸґҧҗҸѦ ҔҸґѹҍѦҡҲӊҸ ҝҽқҸҲ ҰҸ ҡѦѸҽӁҲѹҔѦ ѹѸѦӁ ҰҝѹґҽҊҽҲѦґҧӁӃѽ, ҊҦѦ ҽқҡҸґӀҲӁҽҾ ҝѦґѦӊҽӁҲӁҸҡҲҧ ӁҸҝѸӃ ҔҽҔ ҲҸҍҔѹ ҝѦӍѦҝѦӁ҃ѹѹ ҰҸҦӊѦҝҊҽѦҲҡҾ ҔҝѹҲѹҍѦҡҔҸѸҏ ҽӁҽґѹӉҏ (ӁҽҰҝѹѸѦҝ, Graff 2000, Pander Maat 2006, Tribushinina 2007). ғҸ-ӊҲҸҝӃѽ, ҡҍѹҲҽѦҲҡҾ, ҍҲҸ ӍҸҝѸӃ ҡҝҽӊӁѹҲѦґҧӁҸҮ ѹ ҰҝѦӊҸҡѽҸҦӁҸҮ ҡҲѦҰѦӁѹ ҾӊґҾӀҲҡҾ ҰҝҸҲҸҲѹҰѹҍѦҡҔѹ ӁѦѸҽҝҔѹҝҸӊҽӁӁӃѸѹ ѦҦѹӁѹ҃ҽѸѹ (ҭҸ҃ѦґҏѦӊҔѹҮ 1977, Bierwisch 1989, Kennedy 1999, Kennedy & McNally 2005, Lyons 1969, 1977). ҶҲҸ ҸӉӁҽҍҽѦҲ, ҍҲҸ ҰҝѹӉӁҽҔ, ҸқҸӉӁҽҍҽѦѸӃҮ ҡҔҽґҾҝӁӃѸѹ ҽӁҲҸӁѹѸҽѸѹ ӊ ӍҸҝѸѦ ҡҝҽӊӁѹҲѦґҧӁҸҮ ѹ ҰҝѦӊҸҡѽҸҦӁҸҮ ҡҲѦҰѦӁѹ, ѸҸңѦҲ ӁҽѽҸҦѹҲҧҡҾ Ӂҽ ӊҡѦѸ ҰҝҸҡҲҝҽӁҡҲӊѦ ҊҝҽҦҽ҃ѹҸӁӁҸҮ җҔҽґӃ ѹ ӁѦ ӉҽӊѹҡѹҲ ҸҲ ӁҸҝѸӃ. ѢҽҰҝѹѸѦҝ, ҰҸґҸңѹҲѦґҧӁҽҾ ӍҸҝѸҽ ҰҝѹґҽҊҽҲѦґҧӁҸҊҸ ԼӹԊӸԶӹӡ ҸӉӁҽҍҽѦҲ, ҍҲҸ ҝҽӉѸѦҝӃ ҸқӆѦҔҲҽ ҰҝѦӊӃҗҽӀҲ ҡҝѦҦӁѹѦ ҝҽӉѸѦҝӃ ҦҽӁӁҸҊҸ Ҕґҽҡҡҽ ҰҝѦҦѸѦҲҸӊ. ғ ҸҲґѹҍѹѦ ҸҲ ӍҸҝѸӃ ҰҸґҸңѹҲѦґҧӁҸҮ ҡҲѦҰѦӁѹ, ӍҸҝѸҽ ҔҸѸҰҽҝҽҲѹӊҽ (ԼӹԊӸԶӾ) ѸҸңѦҲ ҰҝѹѸѦӁҾҲҧҡҾ, ҦҽңѦ Ѧҡґѹ ҝҽӉѸѦҝӃ ҸқӆѦҔҲҽ ѸѦӁҧҗѦ ӁҸҝѸӃ. ѷҸҍҔҸҮ ҝѦӍѦҝѦӁ҃ѹѹ ӊ ҰҸҡґѦҦӁѦѸ ҡґҏҍҽѦ ҡґҏңҽҲ ҝҽӉѸѦҝӃ ҦҝҏҊҸҊҸ ҸқӆѦҔҲҽ ѹґѹ ҊҝҏҰҰӃ ҸқӆѦҔҲҸӊ. ғ-ҲҝѦҲҧѹѽ, ҍҽҡҲҸ ҏҲӊѦҝңҦҽѦҲҡҾ, ҍҲҸ җҔҽґӃ, ѹҡҰҸґҧӉҏӀӂѹѦҡҾ ҦґҾ ҔҸҊӁѹҲѹӊӁҸҮ ҸқҝҽқҸҲҔѹ ҡҔҽґҾҝӁӃѽ ҽӁҲҸӁѹѸҸӊ, ѹѸѦӀҲ ҸҲҔҝӃҲҏӀ ҡҲҝҏҔҲҏҝҏ, Ҳ.Ѧ. ҏ Ӂѹѽ ҸҲҡҏҲҡҲӊҏѦҲ ҔҽҔ ӊѦҝѽӁѹҮ, ҲҽҔ ѹ ӁѹңӁѹҮ ҰҝѦҦѦґ (Cruse 1986, Kennedy 2005, Kennedy & McNally 2005, Paradis 1997, 2000, 2001, Rotstein & Winter 2004). ѩҽӁӁҸѦ ҏҲӊѦҝңҦѦӁѹѦ ѹѸѦѦҲ ҰҸҦ ҡҸқҸҮ Ҧӊҽ ҸҡӁҸӊҽӁѹҾ – ҸӁҲҸґҸҊѹҍѦҡҔҸѦ ѹ ҡҸқҡҲӊѦӁӁҸ ҾӉӃҔҸӊҸѦ. ҫӁҲҸґҸҊѹҍѦҡҔҸѦ ҸҡӁҸӊҽӁѹѦ ӉҽҔґӀҍҽѦҲҡҾ ӊ ҲҸѸ, ҍҲҸ ҲҽҔѹѦ ҰҝѹӉӁҽҔѹ ҔҽҔ ӊӃҡҸҲҽ, ҡҲҸѹѸҸҡҲҧ ѹґѹ ҊҝҸѸҔҸҡҲҧ ѸҸҊҏҲ қӃҲҧ, ӊ ҰҝѹӁ҃ѹҰѦ, ԼӾԙӪӹӰӾԓӰӹ қҸґҧҗѹѸѹ. ѨӁӃѸѹ ҡґҸӊҽѸѹ, ҡҲҽҔҽӁ ӁѦ ѸҸңѦҲ қӃҲҧ қѦҡҔҸӁѦҍӁҸ ҰҸґҸӁ, Ҳ.Ҕ. ҏ ӁѦҊҸ ѦҡҲҧ ҔҝҽҾ (ѹӉ ҍѦҊҸ ҡґѦҦҏѦҲ, ҍҲҸ ҰҝѹґҽҊҽҲѦґҧӁҸѦ ԗӹԊӰՂӡ – ҽқҡҸґӀҲӁҸѦ), ӁҸ ҡҲҽҔҽӁ, ҔҽҔ ҏҲӊѦҝңҦҽѦҲҡҾ, ѸҸңѦҲ қӃҲҧ қѦҡ-

464

ѣһѣѢү ѷҩѨҷѰҹѨѢѨѢү

ҔҸӁѦҍӁҸ ӊӃҡҸҔ, ӊ ҲҸѸ ҡѸӃҡґѦ, ҍҲҸ ѦҊҸ ӊҡѦҊҦҽ ѸҸңӁҸ ҡҦѦґҽҲҧ ѦӂѦ ӊӃҗѦ (ҡґѦҦҸӊҽҲѦґҧӁҸ, ҰҝѹґҽҊҽҲѦґҧӁҸѦ ӴՂԙӹӪӠӡ – ҝѦґҾҲѹӊӁҸѦ). һѹӁҊӊѹҡҲѹҍѦҡҔҸѦ ҸқҸҡӁҸӊҽӁѹѦ ҦҽӁӁҸҮ ҊѹҰҸҲѦӉӃ ҡҸҡҲҸѹҲ ӊ ҲҸѸ, ҍҲҸ ӊ ҽӁҊґѹҮҡҔҸѸ ҾӉӃҔѦ ѹ ҝҾҦѦ ҦҝҏҊѹѽ ҾӉӃҔҸӊ (ӁҽҰҝѹѸѦҝ, ӊ ӁѹҦѦҝґҽӁҦҡҔҸѸ, җӊѦҦҡҔҸѸ, ѹҡҰҽӁҡҔҸѸ ѹ ҔҽҲҽґҽӁҡҔҸѸ) ҰҝѹґҽҊҽҲѦґҧӁӃѦ-ҽӁҲҸӁѹѸӃ ӁѦ ҡҸҍѦҲҽӀҲҡҾ ҡ ӁҽҝѦҍѹҾѸѹ, ӊӃҝҽңҽӀӂѹѸѹ ҰҸґӁҏӀ ҡҲѦҰѦӁҧ ҔҽҍѦҡҲӊҽ (ӁҽҰҝѹѸѦҝ, #completely slow, #completely long). ҶҲѹѸ ҡҔҽґҾҝӁӃѦ ҽӁҲҸӁѹѸӃ ҸҲґѹҍҽӀҲҡҾ ҸҲ ҡҔҽґҾҝӁӃѽ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ҡ ҔҸӁҲҝҽҦѹҔҲҸҝӁӃѸ ҡѦѸҽӁҲѹҍѦҡҔѹѸ ҸҲӁҸҗѦӁѹѦѸ, ҰҸҡҔҸґҧҔҏ ҰҸҡґѦҦӁѹѦ ӊҰҸґӁѦ ѸҸҊҏҲ ҸҰҝѦҦѦґҾҲҧҡҾ ӁҽҝѦҍѹҾѸѹ ҰҸґӁҸҮ ҡҲѦҰѦӁѹ (ӁҽҰҝѹѸѦҝ, absolutely empty, completely full). Ѣҽ ҸҡӁҸӊҽӁѹѹ ҙҲѹѽ ҦҽӁӁӃѽ ҸқӃҍӁҸ ҦѦґҽӀҲ ӊӃӊҸҦ, ҍҲҸ ҰҝѹґҽҊҽҲѦґҧӁӃѦ-ҽӁҲҸӁѹѸӃ ҸқҸӉӁҽҍҽӀҲ ҰҝѹӉӁҽҔѹ Ӂҽ ӁѦӉҽѸҔӁҏҲӃѽ җҔҽґҽѽ, ӊ ҲҸ ӊҝѦѸҾ ҔҽҔ ҰҝѹґҽҊҽҲѦґҧӁӃѦ ҔҸӁҲҝҽҦѹҔҲҸҝӁҸҊҸ ҲѹҰҽ ѹӁҲѦҝҰҝѦҲѹҝҏӀҲҡҾ ҔҽҔ ӍҏӁҔ҃ѹѹ Ӂҽ ҍҽҡҲѹҍӁҸ ѹґѹ ҰҸґӁҸҡҲҧӀ ӉҽѸҔӁҏҲӃѽ җҔҽґҽѽ. ҋґѦҦҏѦҲ ҸҲѸѦҲѹҲҧ, ҍҲҸ ҰѦҝѦҍѹҡґѦӁӁӃѦ ӊӃҗѦ ҽҔҡѹҸѸӃ ҸҡӁҸӊҽӁӃ Ӂҽ ѹҡҡґѦҦҸӊҽӁѹҾѽ ҽӁҊґѹҮҡҔѹѽ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ѹґѹ ӊ қҸґѦѦ ҝѦҦҔѹѽ ҡґҏҍҽҾѽ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ѹӉ ҦҝҏҊѹѽ ҊѦҝѸҽӁҡҔѹѽ ҾӉӃҔҸӊ, ӁҽҰҝѹѸѦҝ, ӁѦѸѦ҃ҔҸҊҸ (Bierwisch 1967, 1989, Lang 1989), ӁѹҦѦҝґҽӁҦҡҔҸҊҸ (Broekhuis 1999, H. Klein 1997, Vanden Wyngaerd 2001) ѹ җӊѦҦҡҔҸҊҸ (Paradis & Willners 2006, Vogel 2004). ѷѦѸ ӁѦ ѸѦӁѦѦ, ҽӊҲҸҝӃ ӊӃҗѦҰѦҝѦҍѹҡґѦӁӁӃѽ ҝҽқҸҲ ҰҝѦҲѦӁҦҏӀҲ Ӂҽ ӁѦҔҸҲҸҝҏӀ ҏӁѹӊѦҝҡҽґҧӁҸҡҲҧ ѹѽ ҝѦӉҏґҧҲҽҲҸӊ, ӊ ҲҸѸ ҡѸӃҡґѦ, ҍҲҸ ҸӁѹ ҦѦґҽӀҲ ӊӃӊҸҦӃ ӁѦ Ҹ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ӊ ҽӁҊґѹҮҡҔҸѸ (ӁѦѸѦ҃ҔҸѸ, җӊѦҦҡҔҸѸ ѹ Ҳ.Ҧ.) ҾӉӃҔѦ, ҽ Ҹ ҡѦѸҽӁҲѹҔѦ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ӊ ҃ѦґҸѸ. ҫҍѦӊѹҦӁҸ, ҍҲҸ ҏӁѹӊѦҝҡҽґҧӁҸҡҲҧ ҦҽӁӁӃѽ “ҽҔҡѹҸѸ” ҰҸҦґѦңѹҲ ҲӂҽҲѦґҧӁҸҮ ҰҝҸӊѦҝҔѦ Ӂҽ ѸҽҲѦҝѹҽґѦ ҦҝҏҊѹѽ ҾӉӃҔҸӊ. ғҸӉҧѸѦѸ ҦґҾ ҰҝѹѸѦҝҽ ҡҸҍѦҲҽѦѸҸҡҲҧ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ҡ ӁҽҝѦҍѹҾѸѹ ҡҲѦҰѦӁѹ. ѷҸҲ ӍҽҔҲ, ҍҲҸ ҰҝѹґҽҊҽҲѦґҧӁӃѦ-ҽӁҲҸӁѹѸӃ ӊ ҽӁҊґѹҮҡҔҸѸ ѹ ӁѦҡҔҸґҧҔѹѽ ҦҝҏҊѹѽ ҾӉӃҔҽѽ ӁѦ ѸҸҊҏҲ ҸҰҝѦҦѦґҾҲҧҡҾ ӁҽҝѦҍѹҾѸѹ, ӊӃҝҽңҽӀӂѹѸѹ ҰҸґӁҏӀ ҡҲѦҰѦӁҧ ҰҝѹӉӁҽҔҽ, ѦӂѦ ӁѦ ҸӉӁҽҍҽѦҲ, ҍҲҸ ӁѦ ҡҏӂѦҡҲӊҏѦҲ ҾӉӃҔҸӊ, ҊҦѦ ҽӁҲҸӁѹѸӃ ѸҸҊҏҲ ҡҸҍѦҲҽҲҧҡҾ ҡ ӁҽҝѦҍѹҾѸѹ ҦҽӁӁҸҊҸ ҲѹҰҽ. ѢҽҰҝѹѸѦҝ, ӊ ҝҏҡҡҔҸѸ ҾӉӃҔѦ ҡҸҍѦҲҽӁѹҾ ҲѹҰҽ ԙӹӴԙӾԐ ԌӾԶӾӴՂӡ ѹґѹ ԙӹӴԙӾԐ ԐԂԊӾӰӸӪӠӡ ӊҰҸґӁѦ ҰҝѹѦѸґѦѸӃ. ғӊѹҦҏ ӊҡѦҊҸ ӊӃҗѦҡҔҽӉҽӁӁҸҊҸ, ҃Ѧґҧ ҦҽӁӁҸҮ ҡҲҽҲҧѹ ҡҸҡҲҸѹҲ ӊ ҔҝѹҲѹҍѦҡҔҸѸ ҝҽҡҡѸҸҲҝѦӁѹѹ ҸҰѹҡҽӁӁӃѽ ӊӃҗѦ ҽҔҡѹҸѸ ӊ ҡӊѦҲѦ ҦҽӁӁӃѽ ѹӉ ҝҏҡҡҔҸҊҸ ҾӉӃҔҽ. ҘҽҔ қҏҦѦҲ ҰҸҔҽӉҽӁҸ ӁѹңѦ, ҡѦѸҽӁҲѹҔҽ ѹ ӍҏӁҔ҃ѹҸӁѹҝҸӊҽӁѹѦ ҝҏҡҡҔѹѽ ҰҝѹґҽҊҽҲѦґҧӁӃѽ-ҽӁҲҸӁѹѸҸӊ ҰҝѦҦҡҲҽӊґҾѦҲ ӊѦҡҸѸӃѦ ҔҸӁҲҝҽҝҊҏѸѦӁҲӃ ҰҝҸҲѹӊ ҡѦѸҽӁҲѹҍѦҡҔҸҮ ҲѦҸҝѹѹ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ӊ ҲҸѸ ӊѹҦѦ, ӊ ҔҽҔҸѸ ҸӁҽ ҡҏӂѦҡҲӊҏѦҲ Ӂҽ ӁҽҡҲҸҾӂѹҮ ѸҸѸѦӁҲ. 3. ѢҸҝѸҽ ѹ ҝѦӍѦҝѦӁ҃ѹҽґҧӁӃҮ Ҕґҽҡҡ ҘҽҔ ҏңѦ қӃґҸ ҡҔҽӉҽӁҸ ӊӃҗѦ, ҰҸӁҾҲѹѦ ӁҸҝѸӃ ҾӊґҾѦҲҡҾ ҸҦӁѹѸ ѹӉ ӁҽѹқҸґѦѦ ҸҡӁҸӊҸҰҸґҽҊҽӀӂѹѽ ӊ ҡѦѸҽӁҲѹҍѦҡҔҸҮ ҲѦҸҝѹѹ ҰҝѹґҽҊҽҲѦґҧӁӃѽ-ҽӁҲҸӁѹѸҸӊ.

ѢѣҋҘҫһѻҘҫ үҘҋѨҫҴ ҫ ҭҩѨһүҐүѷѣһѻѢѳҤ-үѢѷҫѢѨҴүҤ

465

ҭҸҦ “ӁҸҝѸҸҮ” ҸқӃҍӁҸ ҰҸӁѹѸҽѦҲҡҾ ҡҝѦҦӁҾҾ ҡҲѦҰѦӁҧ ҰҝѹӉӁҽҔҽ. ғ ҲҽҔҸѸ ҡґҏҍҽѦ ҰҝѹґҽҊҽҲѦґҧӁӃѦ-ҽӁҲҸӁѹѸӃ ҸқҸӉӁҽҍҽӀҲ ҸҲҔґҸӁѦӁѹѦ ҸҲ ҡҝѦҦӁѦҡҲҽҲѹҡҲѹҍѦҡҔҸҊҸ ҡҲҽӁҦҽҝҲҽ ӊ қҸґҧҗҏӀ ѹґѹ ѸѦӁҧҗҏӀ ҡҲҸҝҸӁҏ (үҰҝѦҡҾӁ 1974, 2005, ҭҸҍѦҰ҃Ҹӊ 1990, ҩҏӉѹӁ 1994, Bierwisch 1967, 1989, Chafe 1970, Katz 1972, E.A. Klein 1980, H. Klein 1997, Lang 1989, Lyons 1969, 1977, Sapir 1944). һѹҗҧ ӊ ҦҸҡҲҽҲҸҍӁҸ ҝѦҦҔѹѽ ҡґҏҍҽҾѽ ҲѦҝѸѹӁ “ӁҸҝѸҽ” ҏҰҸҲҝѦқґҾѦҲҡҾ ӊ қҸґѦѦ җѹҝҸҔҸѸ ҡѸӃҡґѦ ѹ ӊҔґӀҍҽѦҲ ӊ ҡѦқҾ ӁѦ ҲҸґҧҔҸ ҏҡҝѦҦӁѦӁӁҏӀ ҡҲѦҰѦӁҧ ҰҝѹӉӁҽҔҽ, ӁҸ ѹ ҰҸґҸңѹҲѦґҧӁӃҮ ҙҔҡҲҝѦѸҏѸ ҊҝҽҦҽ҃ѹҸӁӁҸҮ җҔҽґӃ (ӁҽҰҝѹѸѦҝ, ҫҋѷҩѳҢ ӊ ҰҽҝѦ Խӫԗӹӡ : ӹԙԽӢՂӡ), ҸңѹҦҽѦѸҏӀ ҡҲѦҰѦӁҧ ҰҝѹӉӁҽҔҽ, ҰҝҸҰҸҝ҃ѹҸӁҽґҧӁҸҡҲҧ ҰҽҝҽѸѦҲҝҸӊ ҸқӆѦҔҲҽ ѹ ҰҽҝҽѸѦҲҝӃ, ӉҽҦҽӁӁӃѦ ҸҰҝѦҦѦґѦӁӁҸҮ ҡѹҲҏҽ҃ѹѦҮ (үҝҏҲӀӁҸӊҽ 1999, ғҸґҧӍ 1985, ҭҸ҃ѦґҏѦӊҡҔѹҮ 1974, Leisi 1975). ғ ҦҽӁӁҸҮ ҡҲҽҲҧѦ, ҔҽҔ ѹ ӊ қҸґҧҗѹӁҡҲӊѦ ѹҡҡґѦҦҸӊҽӁѹҮ, ҰҸҡӊҾӂѦӁӁӃѸ ҽҦӆѦҔҲѹӊӁҸҮ ҡѦѸҽӁҲѹҔѦ, ҲѦҝѸѹӁ “ӁҸҝѸҽ” қҏҦѦҲ ҏҰҸҲҝѦқґҾҲҧҡҾ ҦґҾ ҸқҸӉӁҽҍѦӁѹҾ ҡҝѦҦӁѦҡҲҽҲѹҡҲѹҍѦҡҔҸҊҸ ҡҲҽӁҦҽҝҲҽ. ҫҡҸқѦӁӁҸҡҲҧ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ҡҔҽґҾҝӁҸ-ҽӁҲҸӁѹѸѹҍѦҡҔҸҊҸ ҲѹҰҽ ҡҸҡҲҸѹҲ ӊ ҲҸѸ, ҍҲҸ ӁҸҝѸҽ, ӁѦҸқѽҸҦѹѸҽҾ ҦґҾ ѹѽ ѹӁҲѦҝҰҝѦҲҽ҃ѹѹ, ҸҰҝѦҦѦґҾѦҲҡҾ Ӂҽ ҸҡӁҸӊҽӁѹѹ ҝѦӍѦҝѦӁ҃ѹҽґҧӁҸҊҸ Ҕґҽҡҡҽ ҸҰҝѦҦѦґҾѦѸҸҊҸ ҡҏӂѦҡҲӊѹҲѦґҧӁҸҊҸ. ҭҸ ҙҲҸҮ ҰҝѹҍѹӁѦ, ҸҦӁҸ ѹ ҲҸ ңѦ ҰҝѹґҽҊҽҲѦґҧӁҸѦ, ҏҰҸҲҝѦқґѦӁӁҸѦ ӊ ҔҸӁҲѦҔҡҲѦ ҝҽӉґѹҍӁӃѽ ҡҏӂѦҡҲӊѹҲѦґҧӁӃѽ, ҔҽҔ ҰҝҽӊѹґҸ, ҰҝѹҸқҝѦҲҽѦҲ ҝҽӉґѹҍӁҸѦ ҡѦѸҽӁҲѹҍѦҡҔҸѦ ӁҽҰҸґӁѦӁѹѦ. ѷҽҔ, ҃ѦӁҽ ҦҸҝҸҊҸ ҽӊҲҸѸҸқѹґҾ ӁҽѸӁҸҊҸ ҰҝѦӊӃҗҽѦҲ ҃ѦӁҏ ҦҸҝҸҊҸҊҸ ѽҽґҽҲҽ, ҽ ӊӃҡҸҔѹҮ ҰҾҲѹҔґҽҡҡӁѹҔ ӉӁҽҍѹҲѦґҧӁҸ ӁѹңѦ ӊӃҡҸҔҸҊҸ қҽҡҔѦҲқҸґѹҡҲҽ. ҷҸґҧҗѹѸ ӁѦҦҸҡҲҽҲҔҸѸ ҸҍѦҝҍѦӁӁҸҊҸ ӊӃҗѦ ҰҸҦѽҸҦҽ Ҕ ҽҦӆѦҔҲѹӊӁҸҮ ҡѦѸҽӁҲѹҔѦ ҾӊґҾѦҲҡҾ ҲҸ, ҍҲҸ ѦҊҸ ҡҲҸҝҸӁӁѹҔѹ ҽқҡҸґӀҲѹӉѹҝҏӀҲ ҝѦґѦӊҽӁҲӁҸҡҲҧ ӁҸҝѸӃ. ҷѹҝӊѹҗ, ӁҽҰҝѹѸѦҝ, ӁҽҡҲҽѹӊҽѦҲ, ҍҲҸ “the positive of dimensional adjectives without a complement is always (Ҕҏҝҡѹӊ ѸҸҮ, ѣ.ѷ.) norm-related” (Bierwisch 1989: 95). ҈ҽѸѦҲѹѸ, ҸҦӁҽҔҸ, ҍҲҸ ҡҝѦҦӁҾҾ ҡҲѦҰѦӁҧ ҰҝѹӉӁҽҔҽ ҸҲӁӀҦҧ ӁѦ ӊҡѦҊҦҽ ҝѦґѦӊҽӁҲӁҽ ҦґҾ ѹӁҲѦҝҰҝѦҲҽ҃ѹѹ ҰҽҝҽѸѦҲҝѹҍѦҡҔѹѽ ҰҝѹґҽҊҽҲѦґҧӁӃѽ, ҏҰҸҲҝѦқґѦӁӁӃѽ ӊ ӍҸҝѸѦ ҰҸґҸңѹҲѦґҧӁҸҮ ҡҲѦҰѦӁѹ. ҩҽҡҡѸҸҲҝѹѸ ҡґѦҦҏӀӂѹҮ ҰҝѹѸѦҝ ѹӉ ҷҝѹҲҽӁҡҔҸҊҸ Ӂҽ҃ѹҸӁҽґҧӁҸҊҸ ҔҸҝҰҏҡҽ (ҦҽґѦѦ BNC): (1)

For example, a person witnesses the following events in a swimming pool: A tall adolescent boy walks purposefully up behind a small coloured child and pushes him strongly into the pool. (BNC)

ҫҍѦӊѹҦӁҸ, ҍҲҸ ҰҸҦҝҸҡҲҸҔ ӊ ҰҝѹѸѦҝѦ (1) ӁҽӉӊҽӁ ӊӃҡҸҔѹѸ (tall), ӁѦ ҰҸҲҸѸҏ ҍҲҸ ҸӁ ӊӃҗѦ ҡҝѦҦӁѦҊҸ ҝҸҡҲҽ ҦґҾ ѦҊҸ ӊҸӉҝҽҡҲӁҸҮ ҊҝҏҰҰӃ. ҭҸҦҸқӁӃѸ ңѦ ҸқҝҽӉҸѸ, ҝѦқѦӁҸҔ, ӁҽӉӊҽӁӁӃҮ ѸҽґѦӁҧҔѹѸ (small), ӁѦ ҸқҾӉҽҲѦґҧӁҸ ӁѹңѦ ҡҝѦҦӁѦҡҲҽҲѹҡҲѹҍѦҡҔҸҊҸ ҝҸҡҲҽ ҡӊҸѹѽ ҡӊѦҝҡҲӁѹҔҸӊ. ҋҝѦҦӁѹѦ ҰҽҝҽѸѦҲҝӃ ҝѦӍѦҝѦӁ҃ѹҽґҧӁӃѽ ҔґҽҡҡҸӊ ӁѦ ҝѦґѦӊҽӁҲӁӃ ӊ ҦҽӁӁҸѸ ҔҸӁҲѦҔҡҲѦ. ѷҸҍҔҸҮ ҝѦӍѦҝѦӁ҃ѹѹ ҦґҾ ҔҸҊӁѹҲѹӊӁҸҮ ҸқҝҽқҸҲҔѹ ҡґҸӊҸҡҸҍѦҲҽӁѹҾ tall adolescent ҾӊґҾѦҲҡҾ ҝҸҡҲ ҲѦѸӁҸҔҸңѦҊҸ ҝѦқѦӁҔҽ. Ѩ, ҰҸ ҽӁҽґҸҊѹѹ, ҝҸҡҲ ҝѦқѦӁҔҽ Ҹ҃ѦӁѹӊҽѦҲҡҾ ӊ ҸҲӁҸ-

466

ѣһѣѢү ѷҩѨҷѰҹѨѢѨѢү

җѦӁѹѹ Ҕ ҝҸҡҲҏ ҰҸҦҝҸҡҲҔҽ. ѷҽҔҸѦ ҏҰҸҲҝѦқґѦӁѹѦ ҡҔҽґҾҝӁӃѽ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ҡѽҸңѦ ҡ ҡѦѸҽӁҲѹҔҸҮ ҔҸѸҰҽҝҽҲѹӊҽ. ҭҝѹӊѦҦѦѸ ҲҽҔңѦ ҰҝѹѸѦҝ ѹӉ Ѣҽ҃ѹҸӁҽґҧӁҸҊҸ ҔҸҝҰҏҡҽ ҝҏҡҡҔҸҊҸ ҾӉӃҔҽ (ҦҽґѦѦ ѢҘҩѤ): (2)

Ѥ ҝҽқҸҲҽґҽ ҡѦҔҝѦҲҽҝѦѸ ѹ ҏҍѹґҽҡҧ ӊ җҔҸґѦ ҦґҾ ҦѦқѹґҸӊ. ѣҦҏ ҔҽҔ-ҲҸ ҰҸҡґѦ ҏҝҸҔҸӊ ҦҸѸҸҮ Ӂҽ ҲҝҸґґѦҮқҏҡѦ ѸҽҝҗҝҏҲҽ “ҷ”, ҍҲҸ ѹҦѦҲ ҰҸ ҋҽҦҸӊҸѸҏ ҔҸґҧ҃ҏ. ғӃҝҽңѦӁѹѦ ѸҸѦҊҸ ґѹ҃ҽ ҡҸҸҲӊѦҲҡҲӊҏѦҲ ҰҝҸӍѹґӀ ҏҍѦқӁҸҊҸ ӉҽӊѦҦѦӁѹҾ, ҔҸҲҸҝҸѦ Ҿ ҰҸҡѦӂҽӀ. Ѣҽ ӁѦѸ, ҔҽҔ, ӊҰҝҸҍѦѸ, ѹ ӊҡѦҊҦҽ, ҸҲҝҽңҽґҸҡҧ ҰҸґӁҸѦ ҸҲҡҏҲҡҲӊѹѦ ѸӃҡґѹ. ҴҸңѦҲ қӃҲҧ, Ӊҽ ҙҲҸ, ҽ ѦӂѦ Ӊҽ ҦґѹӁӁӃѦ ңѦґҲӃѦ ӊҸґҸҡӃ, ӊӃҡҸҔѹҮ ҝҸҡҲ ѹ ҲҸӁҔҏӀ ҲҽґѹӀ ѹ ҰҸґӀқѹґ ѸѦӁҾ Ӂҽ ӊҡӀ ҡӊҸӀ ңѹӉӁҧ ҰҝѦҔҝҽҡӁӃҮ ҲҽҮҡҔѹҮ ӀӁҸҗҽ, ҔҸҲҸҝӃҮ ѦҦӊҽ ҦҸѽҸҦѹґ ѸӁѦ ҦҸ ҰґѦҍҽ. Ѥ ҸқҝҽҲѹґҽ ӊӁѹѸҽӁѹѦ Ӂҽ ҰӃҲґѹӊӃҮ ӊӉҊґҾҦ ѸҸӁҊҸґҸѹҦӁӃѽ ҍѦҝӁӃѽ ҊґҽӉ ѸҸѦҊҸ ҡґҏҍҽҮӁҸҊҸ ҰҸҰҏҲҍѹҔҽ, ӁҽҰҝҽӊґѦӁӁӃҮ ӊ ѸҸӀ ҡҲҸҝҸӁҏ. ҴӁѦ ҡҲҽґҸ ӁѦ ҰҸ ҡѦқѦ. (ѢҘҩѤ)1

ѷҸҲ ӍҽҔҲ, ҍҲҸ ҝҸҡҲ ҦѦӊҏҗҔѹ ӁҽӉӊҽӁ ӴՂԙӹӪӠԐ ӊ ҰҝѹӊѦҦѦӁӁҸѸ ӊӃҗѦ ҰҝѹѸѦҝѦ, ѦӂѦ ӁѦ ҰҸҦҝҽӉҏѸѦӊҽѦҲ ҽӊҲҸѸҽҲѹҍѦҡҔѹ, ҍҲҸ ѦѦ ҝҸҡҲ қӃґ ӊӃҗѦ ҡҝѦҦӁѦҊҸ ҦґҾ ҡҸҸҲӊѦҲҡҲӊҏӀӂѦҮ ҰҸґҸӊҸҮ ѹ ӊҸӉҝҽҡҲӁҸҮ ҊҝҏҰҰӃ. ҷҸґѦѦ ӊѦҝҸҾҲӁҸ, ҍҲҸ ҦҽӁӁҽҾ ҽҲҝѹқҏ҃ѹҾ ҦҽӁҽ ӊ ҡҝҽӊӁѦӁѹѹ ҡ ҝҸҡҲҸѸ ҲҽҮҡҔҸҊҸ ӀӁҸҗѹ. ѨӁӃѸѹ ҡґҸӊҽѸѹ, ҝҸҡҲ ѸҸґҸҦҸҊҸ ҍѦґҸӊѦҔҽ ҾӊґҾѦҲҡҾ ӊ ҦҽӁӁҸѸ ҡґҏҍҽѦ қҸґѦѦ ҡҽґѹѦӁҲӁҸҮ ҲҸҍҔҸҮ ҝѦӍѦҝѦӁ҃ѹѹ, ҍѦѸ ӁҸҝѸҽ, ҏҡҲҽӁҸӊґѦӁӁҽҾ ҦґҾ ҝѦӍѦҝѦӁ҃ѹҽґҧӁҸҊҸ Ҕґҽҡҡҽ, Ҕ ҔҸҲҸҝҸѸҏ ҸҲӁҸҡѹҲҡҾ ҦѦӊҏҗҔҽ. ҋґѦҦҏѦҲ ҸҲѸѦҲѹҲҧ, ҍҲҸ ҰҸҦҸқӁҸѦ ҏҰҸҲҝѦқґѦӁѹѦ ӍҸҝѸ ҰҸґҸңѹҲѦґҧӁҸҮ ҡҲѦҰѦӁѹ ҸҲӁӀҦҧ ӁѦ ҾӊґҾѦҲҡҾ ҝѦҦҔѹѸ ѹґѹ ѸҽҝҔѹҝҸӊҽӁӁӃѸ. ѢҽҰҝҸҲѹӊ, ҸӁҸ җѹҝҸҔҸ ҝҽҡҰҝҸҡҲҝҽӁѦӁҸ ѹ ҽқҡҸґӀҲӁҸ ҏӉҏҽґҧӁҸ. ҷҸґѦѦ ҲҸҊҸ, ҔҽҔ ҸҲѸѦҍѦӁҸ ӁѦҡҔҸґҧҔѹѸѹ ѹҡҡґѦҦҸӊҽҲѦґҾѸѹ-ҸӁҲҸґѹӁҊӊѹҡҲҽѸѹ, ѹѸѦӁӁҸ ҲҽҔѹѸ ҸқҝҽӉҸѸ ҦѦҲѹ ҍҽҡҲҸ ѸҽҝҔѹҝҏӀҲ ӉӁҽҍѦӁѹѦ ҔҸѸҰҽҝҽҲѹӊҽ (Braine 1976, Ebeling & Gelman 1994). ѢҽҰҝѹѸѦҝ, ҸӁѹ ѸҸҊҏҲ ҡҔҽӉҽҲҧ ԮӹԊӸԶӹӡ ԌӹԐ, ԐԂԊӾӰӸӪӠӡ ԌӹԐ, ҰҸҦҝҽӉҏѸѦӊҽҾ, ҍҲҸ ҸҦѹӁ ҦҸѸ қҸґҧҗѦ ҦҝҏҊҸҊҸ, ҽ ӁѦ ҲҸ, ҍҲҸ ҸҦѹӁ ҦҸѸ қҸґҧҗѦ, ҽ ҦҝҏҊҸҮ ѸѦӁҧҗѦ ҡҝѦҦӁѦҊҸ. ҈ҽҡґҏңѹӊҽѦҲ ӊӁѹѸҽӁѹҾ ҲҽҔңѦ ҲҸҲ ӍҽҔҲ, ҍҲҸ ҡҽѸ ҔҸӁ҃ѦҰҲ ҋҩѣѩѢѨҢ ҸҍѦӁҧ ҡґҸңѦӁ ҦґҾ ҦѦҲѦҮ. ҘҽҔ ҰҸҔҽӉҽґѹ ҋѦҝҽ ѹ ҋѸѹҲ (Sera & Smith 1987), ҦѦҲѹ ӊ ӊҸӉҝҽҡҲѦ ҍѦҲӃҝѦѽ ґѦҲ ѹҡҰӃҲӃӊҽӀҲ ӉӁҽҍѹҲѦґҧӁӃѦ ҲҝҏҦӁҸҡҲѹ ҡҸ ҡґҸӊҸѸ medium, ҍҲҸ, ҰҸ ѹѽ ѸӁѦӁѹӀ, ҡӊҾӉҽӁҸ ҡ ӁѹӉҔҸҮ ҍҽҡҲҸҲӁҸҡҲҧӀ ҦҽӁӁҸҊҸ ҲѦҝѸѹӁҽ ӊ ҝѦҍѹ ҝҸҦѹҲѦґѦҮ ѹ ҡ ҔҸҊӁѹҲѹӊӁҸҮ ҡґҸңӁҸҡҲҧӀ ҸҰҝѦҦѦґѦӁѹҾ ҡҝѦҦӁѦҊҸ ҙґѦѸѦӁҲҽ ѹӉ ҝҾҦҽ ҸҦӁҸҲѹҰӁӃѽ ҸқӆѦҔҲҸӊ. ѨӉ ҙҲҸҊҸ ѸҸңӁҸ ҡҦѦґҽҲҧ ӊӃӊҸҦ, ҍҲҸ ҡҝѦҦӁѦҡҲҽҲѹҡҲѹҍѦҡҔѹҮ ҡҲҽӁҦҽҝҲ ѹҊҝҽѦҲ ӊѦҡҧѸҽ ҸҊҝҽӁѹҍѦӁӁҏӀ ҝҸґҧ ӊ ҏҡӊҸѦӁѹѹ ҰҝѹґҽҊҽҲѦґҧӁӃѽ-ҽӁҲҸӁѹѸҸӊ ҦѦҲҧѸѹ. ҘҽҔ ҡҰҝҽӊѦҦґѹӊҸ ҸҲѸѦҍѦӁҸ ғѦңқѹ҃ҔҸҮ, “the idea that large (or big) means ‘larger than of average size’ seems completely incompatible with the frequent and 1

ғҸ ӊҡѦѽ ҰҝѹѸѦҝҽѽ ҡҸѽҝҽӁѦӁҽ ҽӊҲҸҝҡҔҽҾ ҸҝӍҸҊҝҽӍѹҾ ѹ ҰҏӁҔҲҏҽ҃ѹҾ.

ѢѣҋҘҫһѻҘҫ үҘҋѨҫҴ ҫ ҭҩѨһүҐүѷѣһѻѢѳҤ-үѢѷҫѢѨҴүҤ

467

competent use of the words big and little by infants in the second year of life” (Wierzbicka 1996: 55). ѰҦҽҍӁҸҮ ѹґґӀҡҲҝҽ҃ѹѦҮ ѹҡҰҸґҧӉҸӊҽӁѹҾ ҡҔҽґҾҝӁӃѽ ҽӁҲҸӁѹѸҸӊ қѦӉ ҸҝѹѦӁҲҽ҃ѹѹ Ӂҽ ӁҸҝѸҏ ҾӊґҾѦҲҡҾ ҏҰҸҲҝѦқґѦӁѹѦ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ҝҽӉѸѦҝҽ ӊ ҡӍѦҝѦ ӁҸѸѹӁҽ҃ѹѹ (ҭҸ҃ѦґҏѦӊҡҔѹҮ 1977: 62-63). ѩҽӁӁҸѦ ҾӊґѦӁѹѦ ҸҍѦӁҧ ҍҽҡҲҸҲӁҸ ӊ ҡҸӊҝѦѸѦӁӁҸѸ ҝҏҡҡҔҸѸ ҾӉӃҔѦ (ѹ ҝҾҦѦ ҦҝҏҊѹѽ ҾӉӃҔҸӊ), ҊҦѦ ҦґҾ ҦѹӍӍѦҝѦӁ҃ѹҽ҃ѹѹ ҲѦӉҸҔ җѹҝҸҔҸ ѹҡҰҸґҧӉҏӀҲҡҾ ҲҽҔѹѦ ӍҝҽӉӃ ҔҽҔ ԁӹԊӠӪ ԼӹԊӸԶӹӡ ѹ ԁӹԊӠӪ ԐԂԊӾӰӸӪӠӡ. ѣӂѦ ҸҦѹӁ ӊѦҡҸѸӃҮ ҔҸӁҲҝҽҝҊҏѸѦӁҲ ѸӃ ӁҽѽҸҦѹѸ ӊ ҡӍѦҝѦ ҝҏҡҡҔѹѽ ҔҽҍѦҡҲӊѦӁӁӃѽ ҰҝѹґҽҊҽҲѦґҧӁӃѽ, қҸґҧҗѹӁҡҲӊҸ ѹӉ ҔҸҲҸҝӃѽ ѹѸѦӀҲ ҔҽҔ ҰҸґӁҏӀ, ҲҽҔ ѹ ҔҝҽҲҔҏӀ ӍҸҝѸҏ. ҤҸҝҸҗҸ ѹӉӊѦҡҲӁҸ, ҍҲҸ ҰҸґӁӃѦ ӍҸҝѸӃ, ҔҽҔ ҰҝҽӊѹґҸ, ҸқҸӉӁҽҍҽӀҲ ҰҝѹӉӁҽҔѹ, ҰҝѹҡҏӂѹѦ ҸқӆѦҔҲҏ ҔҽҔ ҰҝѦҦҡҲҽӊѹҲѦґӀ ҸҰҝѦҦѦґѦӁӁҸҊҸ ҝѦӍѦҝѦӁ҃ѹҽґҧӁҸҊҸ Ҕґҽҡҡҽ, ӊ ҲҸ ӊҝѦѸҾ ҔҽҔ ҔҝҽҲҔѹѦ ӍҸҝѸӃ ӉҽҍҽҡҲҏӀ ҏҔҽӉӃӊҽӀҲ Ӂҽ ҰҝѹӉӁҽҔ, ҡӊҸҮҡҲӊѦӁӁӃҮ ҔҸӁҔҝѦҲӁҸѸҏ ҸқӆѦҔҲҏ ӊ ҔҸӁҔҝѦҲӁҸҮ ҡѹҲҏҽ҃ѹѹ (ғѹӁҸҊҝҽҦҸӊ 1947: 262-265, ҐӊҸӉҦѦӊ 1961: 229, ҹӊѦҦҸӊҽ 1980: 557, Aesaert & Hautekiet 2006, Groen 1998: 152, Ward 1965: 192-203). ҋҝҽӊӁѹѸ ҰҝѹѸѦҝӃ (3) ѹ (4): (3)

ҷҏҦѦҲ ӊӁѦҦҝҾҲҧҡҾ ҲҸґҧҔҸ ӁҸӊӃҮ ѸѦӁѦҦңѸѦӁҲ. ҩҾҦҸӊӃѸ ҝҽқҸҲӁѹҔҽѸ, ҰҝҽӊҦҽ, ҰҝѹҦѦҲҡҾ ҡҸҸҲӊѦҲҡҲӊҸӊҽҲҧ ӁҸӊӃѸ ҲҝѦқҸӊҽӁѹҾѸ. ғҰҝҸҍѦѸ, ҰҝҸҝҽқҸҲҔҽ ҦѦҲҽґѦҮ ҡ ҰҝѹӊґѦҍѦӁѹѦѸ ҊҸҝҸҦҡҔѹѽ ҡҰѦ҃ѹҽґѹҡҲҸӊ ѦӂѦ ӊҰѦҝѦҦѹ. ҭҝҸѦҔҲ, ҔҸӁѦҍӁҸ, ѹӁҲѦҝѦҡӁӃҮ. ҤҸҍѦҲҡҾ ӊѦҝѹҲҧ, ҍҲҸ ґѦҦ ӊ ҔҸӁ҃Ѧ ҔҸӁ҃Ҹӊ ҲҝҸӁѦҲҡҾ. (ѢҘҩѤ)

(4)

ҘҸӁҔҏҝҡ ҰҝҸӊҸҦѹґҡҾ ҽӁҸӁѹѸӁҸ ѹ қӃґ ѸѦңҦҏӁҽҝҸҦӁӃѸ. ҋѦҮҍҽҡ ӊ ѣӊҝҸҰѦ ҙҲҸ ҸқӃҍӁҽҾ ҰҝҽҔҲѹҔҽ – ҰҝҽҔҲѹҍѦҡҔѹ ӊҡѦ ҔҸӁҔҏҝҡӃ ѸѦңҦҏӁҽҝҸҦӁӃѦ. ҶҝѹҔ ӊҽӁ ҶҐѣҩүүѷ, ҊҸґґҽӁҦҡҔѹҮ ҽҝѽѹҲѦҔҲҸҝ, ҏҍҽҡҲӁѹҔ ҔҸӁҔҏҝҡҽ: – ҴӁѦ ҸҍѦӁҧ ѹӁҲѦҝѦҡѦӁ ҙҲҸҲ ҰҝҸѦҔҲ. (ѢҘҩѤ)

ғ ҰҝѹѸѦҝѦ (3) ѹӁҲѦҝѦҡӁҸҡҲҧ ҰҝҸѦҔҲҽ Ҹ҃ѦӁѹӊҽѦҲҡҾ Ӂҽ җҔҽґѦ ҰҝҸѦҔҲҸӊ ҸҰҝѦҦѦґѦӁӁҸҊҸ ҲѹҰҽ (ӁҽҰҝѹѸѦҝ, ѹӁѹ҃ѹҽҲѹӊ ӊ ҡӍѦҝѦ ҊҸҝҸҦҡҔҸҊҸ ҝҽӉӊѹҲѹҾ), ҍҲҸ ѸҸҲѹӊѹҝҏѦҲ ҏҰҸҲҝѦқґѦӁѹѦ ҰҸґӁҸҮ ӍҸҝѸӃ. ғ ҰҝѹѸѦҝѦ (4) ҰҝҸѦҔҲ ӁҽӉӊҽӁ ѹӁҲѦҝѦҡӁӃѸ ӁѦ ӊ ҡҝҽӊӁѦӁѹѹ ҡ ҦҝҏҊѹѸѹ ҰҝҸѦҔҲҽѸѹ, ҽ ӊ ҸҲӁҸҗѦӁѹѹ Ҕ ҸҰҝѦҦѦґѦӁӁҸҮ ҡѹҲҏҽ҃ѹѹ, ҲҸҍӁѦѦ Ҕ ґѹҍӁҸҮ ѸҸҲѹӊҽ҃ѹѹ ҔҸӁҔҝѦҲӁҸҊҸ ҽҝѽѹҲѦҔҲҸҝҽ. ѯҝѦӉӊӃҍҽҮӁҸ ӊҽңѦӁ ҲҸҲ ӍҽҔҲ, ҍҲҸ ӁҸҝѸҽ ѸҸңѦҲ қӃҲҧ ҝѦґѦӊҽӁҲӁҽ ҲҸґҧҔҸ ӊ ҰѦҝӊҸѸ, ӁҸ ӁѹҔҽҔ ӁѦ ӊҲҸҝҸѸ ҡґҏҍҽѦ. ѩҽӁӁҸѦ ҝҽӉґѹҍѹѦ ѸѦңҦҏ ҰҸґӁӃѸѹ ѹ ҔҝҽҲҔѹѸѹ ӍҸҝѸҽѸѹ ҸҡҸқѦӁӁҸ ҸҍѦӊѹҦӁҸ ӊ ҡӍѦҝѦ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ҝҽӉѸѦҝҽ, ҊҦѦ ҏҰҸҲҝѦқґѦӁѹѦ ҔҝҽҲҔҸҮ ӍҸҝѸӃ ӁѦҡѦҲ ҡ ҡҸқҸҮ ҡѦѸҽӁҲѹҍѦҡҔѹҮ ҡҦӊѹҊ ‘қҸґҧҗѦ ӁҸҝѸӃ’ Ռ ‘ҡґѹҗҔҸѸ қҸґҧҗҸҮ ҦґҾ ҍѦҊҸ-ґѹқҸ’ (үҰҝѦҡҾӁ 2005: 4). ҋҝҽӊӁѹѸ ҰҝѹѸѦҝ (5) ҡ ҰҝѹѸѦҝҽѸѹ (6) ѹ (7): (5)

ҭҝҽӊҦҽ, ҙҔҡҰѦҝҲӃ ѹӉ ҴѹӁѹҡҲѦҝҡҲӊҽ ҸқҝҽӉҸӊҽӁѹҾ, ҔҸҲҸҝӃѸ қҏҔӊҽҝҧ ҰҸҝҏҍѹґѹ ҰҝҸӊѦҝѹҲҧ, ӁѹҍѦҊҸ ҸҡҸқѦӁӁҸ ҽӁҲѹҡҸӊѦҲҡҔҸҊҸ ӊ ӁѦѸ ӁѦ Ҹқ-

468

ѣһѣѢү ѷҩѨҷѰҹѨѢѨѢү

Ӂҽҝҏңѹґѹ. ѰҔҽӉҽґѹ ҲҸґҧҔҸ Ӂҽ ҸҲҦѦґҧӁӃѦ ҰҸґѹҲѹҍѦҡҔѹѦ ҰҝҸҡҍѦҲӃ. ҋҔҽңѦѸ, ӊѦґѦґѹ ҰѦҝѦҸҦѦҲҧ ҏҍѹҲѦґҧӁѹ҃ҏ ҡ 54-Ү ҡҲҝҽӁѹ҃Ӄ. ҌқҔҽ, ҡҔҽӉҽґѹ, ҏ ӁѦѦ ҔҸҝҸҲҔҸӊҽҲҽ, ҽ ҔҽқґҏҔ, ӁҽҰҝҸҲѹӊ, ӊӃҡҸҔ. ѢѦѽҸҝҸҗҸ. ѢѦ ѸҸңѦҲ ҡҸӊѦҲҡҔҽҾ ҏҍѹҲѦґҧӁѹ҃ҽ ҲҽҔ ӊӃӉӃӊҽӀӂѦ ҸҦѦӊҽҲҧҡҾ. (ѢҘҩѤ) (6)

ҌқҔҽ Ӂҽ ӁѦҮ қӃґҽ ҔҸҝҸҲҔҽҾ, ӁҸҊѹ қӃґѹ ҦґѹӁӁӃѦ, ѹ ӊҸґҸҡӃ ҍҏҲҧ ӁѦ ҦҸ ҝҽӉҝѦӉҽ ӀқҔѹ, ѹ ҊҦѦ-ҲҸ ҸӁҽ ӉҽҊҸҝҽґҽ ӁѦҦҽӊӁҸ, ѹ, ѸҸңѦҲ қӃҲҧ, ҲҸҰґѦҡ ӉҽҊҸҝҽґҽ. (ѢҘҩѤ)

(7)

Ѥ ӁѦ ѸҸҊҏ қӃҡҲҝҸ қѦҊҽҲҧ, – җѹҰѦґҽ ҫҔҡҽӁѹҾ, – ҏ ѸѦӁҾ ҔҽқґҏҔѹ ӊӃҡҸҔѹѦ! (ү. ҹӊѦҦҸӊ)

ҘҝҽҲҔѹѦ ӍҸҝѸӃ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ҝҽӉѸѦҝҽ ӪӹӢӹԽӪӠӡ ѹ ӴՂԙӹӪӠӡ ӊ (5) ҸӉӁҽҍҽӀҲ, ҍҲҸ ҰҽҝҽѸѦҲҝӃ ҝѦӍѦҝѦӁҲҸӊ ҰҝѦӊӃҗҽӀҲ ѸҽҔҡѹѸҏѸ, ҦҸҰҏҡҲѹѸӃҮ ӊ ҸҰҝѦҦѦґѦӁӁҸҮ ҡѹҲҏҽ҃ѹѹ (ӊ ҰѦҝӊҸѸ ҡґҏҍҽѦ ѹҦѦҾ ӁѦңѦґҽҲѦґҧӁҸҊҸ ѹӉқӃҲҔҽ ҔҽҍѦҡҲӊҽ ҰҸҦҍѦҝҔѹӊҽѦҲҡҾ ҲҽҔңѦ Ұҝѹ ҰҸѸҸӂѹ ҡҏӍӍѹҔҡҽ -ӹӴԂԽ-). 2 ҴѦҦѹҏѸ ҦґѹӁӃ ѹ ӊӃҡҸҲӃ ҦґҾ Ҕґҽҡҡҽ ӀқҸҔ ѹ ҔҽқґҏҔҸӊ ӊ ҦҽӁӁҸѸ ҡґҏҍҽѦ ӁѦ ҝѦґѦӊҽӁҲѦӁ: ӀқҔҽ ӊ (5) ӁѦ ҸқҾӉҽҲѦґҧӁҸ ҔҸҝҸҲҔҽҾ ҦґҾ Ҕґҽҡҡҽ ӀқҸҔ, ҽ ҔҽқґҏҔѹ ӊҰҸґӁѦ ѸҸҊҏҲ қӃҲҧ ҦҽңѦ ӁѹӉҔѹѸѹ ҡ ҲҸҍҔѹ ӉҝѦӁѹҾ ӁҸҝѸӃ. ҫҦӁҽҔҸ ҦҽңѦ ҦґѹӁӁӃѦ ӀқҔѹ ѹ ӁѹӉҔѹѦ ҔҽқґҏҔѹ ѸҸҊҏҲ қӃҲҧ ԙԊӠԶӪӹԐ ҔҸҝҸҲҔѹѸѹ ѹ ӊӃҡҸҔѹѸѹ, ҡҸҸҲӊѦҲҡҲӊѦӁӁҸ, ҦґҾ ҔҽҝҲѹӁҔѹ ӊ қҏҔӊҽҝѦ, ҡѸ. (8) ѹ (9). (8)

ҌқҔҽ ҏ ӁѦѦ, ѽҸҲҧ ӊ ҸқӂѦѸ-ҲҸ ѹ ҦґѹӁӁҽҾ, ӊҡѦ ҝҽӊӁҸ ҔҸҝҸҲҔҸӊҽҲҽ ҦґҾ қҏҔӊҽҝҾ.

(9)

ҘҽқґҏҔѹ ҏ ӁѦѦ, ѽҸҲҧ ӊ ҸқӂѦѸ-ҲҸ ѹ ӁѹӉҔѹѦ, ӊҡѦ ҝҽӊӁҸ ӊӃҡҸҔѹ ҦґҾ қҏҔӊҽҝҾ.

ғ ҸҲґѹҍѹѦ ҸҲ ҔҝҽҲҔѹѽ ӍҸҝѸ ӊ (5), ҰҸґӁӃѦ ӍҸҝѸӃ ҲѦѽ ңѦ ҡҽѸӃѽ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ӊ (6) ѹ (7) ѸҸҊҏҲ ѹӁҲѦҝҰҝѦҲѹҝҸӊҽҲҧҡҾ ҸҲӁҸҡѹҲѦґҧӁҸ ӁҸҝѸӃ ҦґҾ Ҕґҽҡҡҽ ӀқҸҔ ѹ ҔҽқґҏҔҸӊ, ҡҸҸҲӊѦҲҡҲӊѦӁӁҸ. ѷҽҔ, ӀқҔҽ ѸҸңѦҲ ӁҽӉӃӊҽҲҧҡҾ ҔҸҝҸҲҔҸҮ, ҰҸҲҸѸҏ ҍҲҸ ҸӁҽ ҔҸҝҸҍѦ ҡҝѦҦӁѦҮ ҦґѹӁӃ ӀқҸҔ, ҽ ҔҽқґҏҔѹ, ҡҍѹҲҽӀӂѹѦҡҾ ӊӃҡҸҔѹѸѹ, ӊѦҝҸҾҲӁҸ, ҰҝѦӊӃҗҽӀҲ ҡҝѦҦӁӀӀ ҦґѹӁҏ ҔҽқґҏҔҸӊ. ҩҽҡҡѸҸҲҝѹѸ ҲҽҔңѦ ҰҝѹѸѦҝӃ ӊ (10) ѹ (11): (10) ҋґҽӊҽ қҸҊҏ, ҏ ѸѦӁҾ ҡҲҏґ ӁѹӉҔѹҮ қӃґ... ҭҽҦҽҲҧ ѽҸҲҧ ӁѦ қҸґҧӁҸ қӃґҸ. (Void-ӍҸҝҏѸ, http://void.neverclans.ru:8080/forum) (11)

2

ҫӁ ҡѦґ Ӂҽ ҡҲҏґ, ҔҸҲҸҝӃҮ ӁӃӁҍѦ ѹѸѦѦҲҡҾ ӊ ҔҽңҦҸѸ ҦҸѸѦ ѹ ҦҽңѦ ӊ ҔҽңҦҸҮ ѹӉқѦ. ѢӃӁҍѦ ӊѦӉҦѦ ѦҡҲҧ ҡҲҏґҧҾ ѹ ҲҽқҏҝѦҲӃ. ҫӁ ҡѦґ, ѹ ӁҽҡҲҝҸѦӁѹѦ ҍѦґҸӊѦҔҽ ӊҡѦ ҏґҏҍҗҽґҸҡҧ. Ѩ ӁѹӉҸҔ ѦѸҏ ҡҲҽґ ҡҲҏґ! ҴҽґҸ ѦѸҏ ҡҲҽґҸ ҡҲҏґҽ! ҫӁ ҰҸҡҲҽӊѹґ Ӂҽ ҡҲҏґ ҲҽқҏҝѦҲҸҍҔҏ ѹ, ӊӉҊҝҸѸҸӉҦѹӊҗѹҡҧ, ҰҝҸҦҸґ-

ҘҝҽҲҔѹѦ ҰҝѹґҽҊҽҲѦґҧӁӃѦ ҸҦӁҽҔҸ ҏҲҝҽҍѹӊҽӀҲ ҙҲҸ ӉӁҽҍѦӁѹѦ ӊ ҰҸӉѹ҃ѹѹ ҰѦҝѦҦ ҰҸҦґѦңҽӂѹѸ ѹ ҰҸҡґѦ ѹӁҲѦӁҡѹӍѹҔҽҲҸҝҸӊ (үҰҝѦҡҾӁ 2005: 4).

ѢѣҋҘҫһѻҘҫ үҘҋѨҫҴ ҫ ҭҩѨһүҐүѷѣһѻѢѳҤ-үѢѷҫѢѨҴүҤ

469

ңѹґ ҰѹҲҧѦ ѹӉ қҏҲӃґҔѹ ѹ ҡҲҽҔҽӁҽ. ѢҸ ӊѦҦҧ ӊҡѦѸ ңѦ ѹӉӊѦҡҲӁҸ, ҍҲҸ ҡѸӃҡґ ңѹӉӁѹ ӊ ҲҸѸ, ҍҲҸқӃ ӁѹҔҸҊҦҽ ӁѦ ҸҡҲҽӁҽӊґѹӊҽҲҧҡҾ Ӂҽ ҦҸҡҲѹҊӁҏҲҸѸ. ғѦҦҧ ӊҡѦѸ ңѦ ѹӉӊѦҡҲӁҸ, ҍҲҸ Ѧҡґѹ қӃ ҍѦґҸӊѦҔ ҸҡҲҽӁҽӊґѹӊҽґҡҾ Ӂҽ ҦҸҡҲѹҊӁҏҲҸѸ, ҲҸ ҸӁ қӃ ӊѦҝӁҏґҡҾ ӊ ҰѦҝӊҸқӃҲӁҸѦ ҡҸҡҲҸҾӁѹѦ ѹ ҰґҾҡҽґ ҊҸґӃҮ ӊҸҔҝҏҊ ҔҸҡҲҝҽ. (ѢҘҩѤ) ҫҰҾҲҧ ңѦ, ҰҸґӁҽҾ ӍҸҝѸҽ ӰӠԒӪӠӡ ӊ (10) ӊҰҸґӁѦ ѸҸңѦҲ ѹӁҲѦҝҰҝѦҲѹҝҸӊҽҲҧҡҾ ҔҽҔ ‘ѹѸѦӀӂѹҮ ӊӃҡҸҲҏ ӁѹңѦ ҡҝѦҦӁѦҊҸ’. ѨӁӃѸѹ ҡґҸӊҽѸѹ, ҔҸҊӁѹҲѹӊӁҽҾ ҸқҝҽқҸҲҔҽ ҰҝѹґҽҊҽҲѦґҧӁҸҊҸ ӊҔґӀҍҽѦҲ ӊ ҙҲҸѸ ҡґҏҍҽѦ ҸҰҝѦҦѦґѦӁѹѦ ѸѦҡҲҸҰҸґҸңѦӁѹҾ ҔҸӁҔҝѦҲӁҸҊҸ ҡҲҏґҽ Ӂҽ җҔҽґѦ ҡҲҏґҧѦӊ ҸҲӁҸҡѹҲѦґҧӁҸ ӁҸҝѸӃ ҦґҾ ҦҽӁӁҸҊҸ Ҕґҽҡҡҽ ҰҝѦҦѸѦҲҸӊ. ѩҽӁӁҽҾ ҔҸҊӁѹҲѹӊӁҽҾ ҸҰѦҝҽ҃ѹҾ, ҸҦӁҽҔҸ, ӁѦ ҰҝѹѸѦӁѹѸҽ ӊ (11), ҊҦѦ ҔҝҽҲҔҽҾ ӍҸҝѸҽ ҰҝѹґҽҊҽҲѦґҧӁҸҊҸ ҸӉӁҽҍҽѦҲ, ҍҲҸ ӊ ҦҽӁӁҸҮ ҡѹҲҏҽ҃ѹѹ ҡҲҏґ ҡґѹҗҔҸѸ ӁѹӉҸҔ ҦґҾ ҊѦҝҸҾ, ӁҽѽҸҦҾӂѦҊҸҡҾ ӊ ҡҸҡҲҸҾӁѹѹ ҸҰҧҾӁѦӁѹҾ. ғҽңӁҸ ҸҲѸѦҲѹҲҧ, ҍҲҸ ӊ ҰҸҡґѦҦӁѦѸ ҡґҏҍҽѦ ӁѦӊҽңӁҸ, ҾӊґҾѦҲҡҾ ґѹ ҡҲҏґ ӁѹӉҔѹѸ ҦґҾ Ҕґҽҡҡҽ ҡҲҏґҧѦӊ ӊҸҸқӂѦ. ҫӁ ӊҰҸґӁѦ ѸҸңѦҲ қӃҲҧ ѹ ӊӃҡҸҔѹѸ ҸҲӁҸҡѹҲѦґҧӁҸ ӁҸҝѸӃ ҦґҾ ҡҲҏґҧѦӊ (ҡѸ. (12)), ҦҽӁӁӃҮ ҽҡҰѦҔҲ ҰҝҸҡҲҸ ӁѦ ҾӊґҾѦҲҡҾ ҝѦґѦӊҽӁҲӁӃѸ ӊ ҦҽӁӁҸѸ ҡґҏҍҽѦ. (12) Ѩ ӁѹӉҸҔ ҡҲҽґ ѦѸҏ ѦҊҸ ӊӃҡҸҔѹҮ ҡҲҏґ! ғҡѦ ӊӃҗѦҡҔҽӉҽӁӁҸѦ ҸҲӁҸҡѹҲҡҾ ѹ Ҕ ҡґѦҦҏӀӂѹѸ ҦӊҏѸ ҡґҏҍҽҾѸ, ҊҦѦ ҰҸґӁҽҾ ӍҸҝѸҽ ӊ (13) ҰҸҦҝҽӉҏѸѦӊҽѦҲ, ҍҲҸ ҊҸҝҗҸҔ ѸѦӁҧҗѦ ҡҝѦҦӁѦҮ ӊѦґѹҍѹӁӃ ҃ӊѦҲҸҍӁӃѽ ҊҸҝҗҔҸӊ, ҽ ҔҝҽҲҔҽҾ ӍҸҝѸҽ ӊ (14) ҸӉӁҽҍҽѦҲ, ҍҲҸ ҊҸҝҗҸҔ ӁѦ ҦҸҡҲҽҲҸҍӁҸ қҸґҧҗҸҮ ҦґҾ ҍѦҲӃҝѦѽґѦҲӁѹѽ ҦѦҝѦӊ҃Ѧӊ. ѨӁҲѦҝѦҡӁҸ ҸҲѸѦҲѹҲҧ ҲҽҔңѦ, ҍҲҸ ҊҸҝҗҸҔ, ӊ ҔҸҲҸҝҸѸ ҝҸҡґѹ ԌӾӢӾӴՁԂ, ӊѦҝҸҾҲӁҸ, ҾӊґҾѦҲҡҾ қҸґҧҗѹѸ ҰҸ ҡҝҽӊӁѦӁѹӀ ҡ ҸқӃҔӁҸӊѦӁӁӃѸѹ ҃ӊѦҲҸҍӁӃѸѹ ҊҸҝҗҔҽѸѹ, ҍҲҸ ӊ ҸҍѦҝѦҦӁҸҮ ҝҽӉ ҰҸҦҲӊѦҝңҦҽѦҲ ӁѦҰҝѹѸѦӁѹѸҸҡҲҧ ӁҸҝѸӃ ҦґҾ ѹӁҲѦҝҰҝѦҲҽ҃ѹѹ ҔҝҽҲҔѹѽ ӍҸҝѸ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ҝҽӉѸѦҝҽ ӊ ҝҏҡҡҔҸѸ ҾӉӃҔѦ. (13) ҤҸҲҾ ҏ ѸѦӁҾ ҊҸҝҗҸҔ ѸҽґѦӁҧҔѹҮ - ҰѦҝѦґѹӊҽ Ұҝѹ ҲҽҔҸҮ ңҽҝѦ ӁѦ қӃӊҽѦҲ. (ӍҸҝҏѸ Ӂҽ HomeFlowers.ru) (14) ҭҸҡґѦ ҍѦҲӃҝѦѽ ґѦҲ ҦѦҝѦӊ҃ҽ ӊӃҝҽҡҲҽӀҲ қҸґҧҗѹѸѹ, ѹ ҊґѹӁҾӁӃҮ ҊҸҝҗҸҔ ҡҲҽӁҸӊѹҲҡҾ Ѹҽґ. (ѢҘҩѤ) ҭҝѦҦҡҲҽӊґѦӁӁӃҮ ӊӃҗѦ ҽӁҽґѹӉ ҰҸҔҽӉӃӊҽѦҲ, ҍҲҸ ҏҲӊѦҝңҦѦӁѹѦ Ҹқ ҸқґѹҊҽҲҸҝӁҸҡҲѹ ӁҸҝѸӃ ӊ ѹӁҲѦҝҰҝѦҲҽ҃ѹѹ ҰҸґҸңѹҲѦґҧӁӃѽ ӍҸҝѸ ҰҝѹґҽҊҽҲѦґҧӁӃѽҽӁҲҸӁѹѸҸӊ ӁѦ ӊӃҦѦҝңѹӊҽѦҲ ҔҝѹҲѹҔѹ ӊ ҡӊѦҲѦ ҦҽӁӁӃѽ ѹӉ ҝҏҡҡҔҸҊҸ ҾӉӃҔҽ, ҊҦѦ ӁѦ ҲҸґҧҔҸ ӍҸҝѸӃ ҡҝҽӊӁѹҲѦґҧӁҸҮ ѹ ҰҝѦӊҸҡѽҸҦӁҸҮ ҡҲѦҰѦӁѹ, ӁҸ ѹ ӍҸҝѸӃ ҰҸґҸңѹҲѦґҧӁҸҮ ҡҲѦҰѦӁѹ ѸҸҊҏҲ ѹӁҲѦҝҰҝѦҲѹҝҸӊҽҲҧҡҾ ӁѦӉҽӊѹҡѹѸҸ ҸҲ ӁҸҝѸӃ.

470

ѣһѣѢү ѷҩѨҷѰҹѨѢѨѢү

4. ѢѦѸҽҝҔѹҝҸӊҽӁӁҸҡҲҧ ӍҸҝѸ ҡҝҽӊӁѹҲѦґҧӁҸҮ ҡҲѦҰѦӁѹ ѣӂѦ ҸҦӁҽ ѽҸҝҸҗҸ ѹӉӊѦҡҲӁҽҾ ҽҔҡѹҸѸҽ Ҹ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ҡ ҔҸӁҲҝҽҝӁӃѸ ҡѦѸҽӁҲѹҍѦҡҔѹѸ ҸҲӁҸҗѦӁѹѦѸ ѸѦңҦҏ ҍґѦӁҽѸѹ ҽӁҲҸӁѹѸѹҍѦҡҔҸҮ ҰҽҝӃ – ҙҲҸ ӁѦѸҽҝҔѹҝҸӊҽӁӁҸҡҲҧ ӍҸҝѸ ҔҸѸҰҽҝҽҲѹӊҽ ѹ ҡҏҰѦҝґҽҲѹӊҽ. ҷѹҝӊѹҗ, ӁҽҰҝѹѸѦҝ, ӊѦҡҧѸҽ ӁѦҦӊҏҡѸӃҡґѦӁӁҸ ҏҲӊѦҝңҦҽѦҲ, ҍҲҸ “comparative, superlative and ‘too’ constructions are never norm-related” (Bierwisch 1989: 95). ҋҝҽӉҏ ҡґѦҦҏѦҲ ҸҲѸѦҲѹҲҧ, ҍҲҸ ҰҸҦ “comparative” ҷѹҝӊѹҗ ѹѸѦѦҲ ӊ ӊѹҦҏ ҲҸґҧҔҸ ӍҸҝѸӃ ҡҝҽӊӁѹҲѦґҧӁҸҮ ҡҲѦҰѦӁѹ ӊ ҏӉҔҸѸ ҡѸӃҡґѦ ҙҲҸҊҸ ҡґҸӊҽ, Ҳ.Ѧ. ӍҸҝѸӃ, ӊӃҝҽңҽӀӂѹѦ қҸґѦѦ ӊӃҡҸҔҏӀ ҡҲѦҰѦӁҧ ҔҽҍѦҡҲӊҽ (comparison of superiority/majority), ӁҸ ӁѦ ҡҝҽӊӁѹҲѦґҧӁӃѦ ҸқҸҝҸҲӃ, ӊӃҝҽңҽӀӂѹѦ ѸѦӁѦѦ ӊӃҡҸҔҏӀ ҡҲѦҰѦӁҧ ҔҽҍѦҡҲӊҽ (comparison of inferiority/minority) ѹ ҲҸңҦѦҡҲӊҸ ҡҲѦҰѦӁѹ ҔҽҍѦҡҲӊҽ (comparison of equality/equatives). ҷѹҝӊѹҗ ҡҰҝҽӊѦҦґѹӊҸ ҸҲѸѦҍҽѦҲ, ҍҲҸ ӊ ҡҝҽӊӁѹҲѦґҧӁӃѽ ҸқҸҝҸҲҽѽ, ӊӃҝҽңҽӀӂѹѽ ѸѦӁѦѦ ӊӃҡҸҔҏӀ ҡҲѦҰѦӁҧ ҔҽҍѦҡҲӊҽ, ҰҝѹґҽҊҽҲѦґҧӁӃѦ-ҽӁҲҸӁѹѸӃ ѸҽҝҔѹҝҸӊҽӁӃ (ҡҝ. Croft & Cruse 2004: 176-179), ҡѸ. (15) ѹ (16): (15) The idea was to get away from the ‘man in a suit’ which had become a cliché of space fiction and alien fantasies. In fact, the point of having him played by a 6-foot 10 inch (2.1-m) Nigerian student, Bolaji Badejo, plus a rather less tall stuntman, was largely lost. (BNC) (16) ҘҸҊҦҽ ѸӃ ҡҲҏҰѹґѹ Ӂҽ үӁӉѦҝҡҔѹҮ қѦҝѦҊ – ҸқӂѹҮ ҡѹґҏҙҲ ҐҸґҊҸӍӃ ӉҽҡґҸӁѹґҡҾ ҦҝҏҊѹѸѹ, ѸѦӁѦѦ ӊӃҡҸҔѹѸѹ ҊҸҝҽѸѹ. (ѢҘҩѤ) ҘҽҔ ӊ ҽӁҊґѹҮҡҔҸѸ, ҲҽҔ ѹ ӊ ҝҏҡҡҔҸѸ ҾӉӃҔѦ ҡҝҽӊӁѹҲѦґҧӁӃѦ ҔҸӁҡҲҝҏҔ҃ѹѹ ҡҸ ӉӁҽҍѦӁѹѦѸ ѸѦӁѦѦ ӊӃҡҸҔҸҮ ҡҲѦҰѦӁѹ ҔҽҍѦҡҲӊҽ ҰҸҦҝҽӉҏѸѦӊҽӀҲ, ҍҲҸ ҔҽҍѦҡҲӊҸ ѸҸңѦҲ қӃҲҧ ҰҝѹҰѹҡҽӁҸ ҸқӆѦҔҲҏ ѹ ӊ ҽқҡҸґӀҲӁӃѽ ҲѦҝѸѹӁҽѽ. ѢҽҰҝѹѸѦҝ, ӁѦҡѸҸҲҝҾ Ӂҽ ҲҸ, ҍҲҸ ҔҽҡҔҽҦѦҝ ӊ (15) ӁѹңѦ ҝҸҡҲҸѸ, ҍѦѸ ҡҽѸ ҽҔҲѦҝ, ҸӁ, ҲѦѸ ӁѦ ѸѦӁѦѦ, ҾӊґҾѦҲҡҾ ӊӃҡҸҔѹѸ ҦґҾ ҡӊҸѦҊҸ ҝѦӍѦҝѦӁ҃ѹҽґҧӁҸҊҸ Ҕґҽҡҡҽ. ҭҸҦҸқӁӃѸ ңѦ ҸқҝҽӉҸѸ, ҊҸҝӃ, ӁҽӉӊҽӁӁӃѦ “ѸѦӁѦѦ ӊӃҡҸҔѹѸѹ” ӊ (16), Ӂҽ ҡҽѸҸѸ ҦѦґѦ ӊӃҡҸҔѹѦ, ѽҸҲҾ ѹ ӁѦ ҲҽҔѹѦ ӊӃҡҸҔѹѦ, ҔҽҔ ҡѹґҏҙҲ ҐҸґҊҸӍӃ. ҷѹҝӊѹҗ ҲҽҔңѦ ҸҲѸѦҍҽѦҲ, ҍҲҸ ӊ ҡҝҽӊӁѹҲѦґҧӁӃѽ ҔҸӁҡҲҝҏҔ҃ѹҾѽ ҡҸ ӉӁҽҍѦӁѹѦѸ ҲҸңҦѦҡҲӊҽ ҡҲѦҰѦӁѹ ҔҽҍѦҡҲӊҽ ҰҝѹґҽҊҽҲѦґҧӁӃѦ қҸґҧҗҸҊҸ ҰҸґӀҡҽ ӁѦѸҽҝҔѹҝҸӊҽӁӃ, ѹ ҙҲѹѸ ҸҲґѹҍҽӀҲҡҾ ҸҲ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ѸҽґҸҊҸ ҰҸґӀҡҽ, ҔҸҲҸҝӃѦ ӊ ҦҽӁӁҸҮ ҔҸӁҡҲҝҏҔ҃ѹѹ ӊҡѦҊҦҽ ҸҝѹѦӁҲѹҝҸӊҽӁӃ Ӂҽ ӁҸҝѸҏ (Bierwisch 1989: 95). үӁҽґѹӉ ҽҏҲѦӁҲѹҍӁӃѽ ҰҝѹѸѦҝҸӊ ҡӊѹҦѦҲѦґҧҡҲӊҏѦҲ Ҹ ҲҸѸ, ҍҲҸ ҦҽӁӁҸѦ ҏҲӊѦҝңҦѦӁѹѦ ӊѦҝӁҸ ҦґҾ ҽӁҊґѹҮҡҔҸҊҸ ҾӉӃҔҽ. ѩґҾ ѹґґӀҡҲҝҽ҃ѹѹ ҰҝѹӊѦҦѦѸ ҡґѦҦҏӀӂѹѦ ҰҝѹѸѦҝӃ ѹӉ ҷҝѹҲҽӁҡҔҸҊҸ Ӂҽ҃ѹҸӁҽґҧӁҸҊҸ ҔҸҝҰҏҡҽ: (17) The shelves near the ground were usually well kept, but higher up they became more disorganised and larger in size, almost as tall as the pupils themselves. (BNC)

ѢѣҋҘҫһѻҘҫ үҘҋѨҫҴ ҫ ҭҩѨһүҐүѷѣһѻѢѳҤ-үѢѷҫѢѨҴүҤ

471

(18) Stuart’s parents are human, but the whole family loves him, and they make sure nobody discriminates against him because of his size and appearance. The bit where the rest of the family goes out and the cat is after Stuart bores Babur: he thinks it is probably directed at the younger reader. But then Stuart meets a little girl who is fully human but every bit as short as Stuart, and his heart is lost to her. (BNC) ҭҝѹґҽҊҽҲѦґҧӁҸѦ қҸґҧҗҸҊҸ ҰҸґӀҡҽ tall ӊ (17) ӁѦѸҽҝҔѹҝҸӊҽӁҸ, Ҳ.Ѧ. ҡҝҽӊӁѹҲѦґҧӁӃҮ ҸқҸҝҸҲ as tall as ҲҸґҧҔҸ ҡҝҽӊӁѹӊҽѦҲ ӊѦҝҲѹҔҽґҧӁҏӀ ҰҝҸҲҾңѦӁӁҸҡҲҧ Ҧӊҏѽ ҸқӆѦҔҲҸӊ, ӁҸ ӁѹҍѦҊҸ ӁѦ ҊҸӊҸҝѹҲ Ҹқ ѹѽ ӊӃҡҸҲѦ ҸҲӁҸҡѹҲѦґҧӁҸ ӁҸҝѸӃ ҦґҾ ҡҸҸҲӊѦҲҡҲӊҏӀӂѹѽ ҝѦӍѦҝѦӁ҃ѹҽґҧӁӃѽ ҔґҽҡҡҸӊ. ѨӁӃѸѹ ҡґҸӊҽѸѹ, ӁҽѸ ӁѦ ѹӉӊѦҡҲӁҸ, ҾӊґҾӀҲҡҾ ґѹ ҏҍѦӁѹҔѹ ѹ ҡҲѦґґҽңѹ ӊӃҡҸҔѹѸѹ ѹґѹ ӁѹӉҔѹѸѹ ҸҲӁҸҡѹҲѦґҧӁҸ ҡҝѦҦӁѦҡҲҽҲѹҡҲѹҍѦҡҔѹѽ ҏҍѦӁѹҔҸӊ ѹ ҡҲѦґґҽңѦҮ, ҡҸҸҲӊѦҲҡҲӊѦӁӁҸ. ѢҽҰҝҸҲѹӊ, ҰҝѹґҽҊҽҲѦґҧӁҸѦ ѸҽґҸҊҸ ҰҸґӀҡҽ short ӊ ҰҝѹѸѦҝѦ (18) ѸҽҝҔѹҝҸӊҽӁҸ ѹ ҸӉӁҽҍҽѦҲ, ҍҲҸ ѹ ҦѦӊҸҍҔҽ, ѹ ҋҲӀҽҝҲ ӁѦӊӃҡҸҔҸҊҸ ҝҸҡҲҽ. ѯҲҸ ҔҽҡҽѦҲҡҾ ҸқҸҝҸҲҸӊ ҡҸ ӉӁҽҍѦӁѹѦѸ ҲҸңҦѦҡҲӊҽ ҡҲѦҰѦӁѹ ҔҽҍѦҡҲӊҽ ӊ ҝҏҡҡҔҸѸ ҾӉӃҔѦ, ҲҸ ҔҽҝҲѹӁҽ ӁѦҡҔҸґҧҔҸ ҸҲґѹҍҽѦҲҡҾ ҸҲ ҽӁҊґѹҮҡҔҸҊҸ ҾӉӃҔҽ. ҭҝѹґҽҊҽҲѦґҧӁӃѦ ҔҽҔ қҸґҧҗҸҊҸ, ҲҽҔ ѹ ѸҽґҸҊҸ ҰҸґӀҡҽ ѸҽҝҔѹҝҸӊҽӁӃ ӊ ҦҽӁӁҸҮ ҔҸӁҡҲҝҏҔ҃ѹѹ ӊ ҝҏҡҡҔҸѸ ҾӉӃҔѦ. ҩҽҡҡѸҸҲҝѹѸ ҡґѦҦҏӀӂѹѦ ҰҝѹѸѦҝӃ: (19) ѷҸґҧҔҸ ӊҦҽґѹ, Ӂҽ ӉҽҰҽҦѦ, ҲҏѸҽӁѹґѹҡҧ ѦӂѦ ҔҽҔѹѦ-ҲҸ ҊҸҝӁӃѦ ӊѦҝҗѹӁӃ, ҲҽҔѹѦ ңѦ ӁѦӊӃҡҸҔѹѦ, қѦҡ҃ӊѦҲӁӃѦ ѹ ҸҦѹӁҸҔѹѦ, ҔҽҔ ѹ ҲѦ, ҡ ҔҸҲҸҝӃѽ Ҿ ҡѸҸҲҝѦґ. (ѢҘҩѤ) (20) ҭҝҽҦѦҦҽ ӁѹҔҲҸ ӁѦ ҰҸѸӁѹҲ ҔҝѹҍҽӂѹѸ ѹґѹ ҝҏҊҽӀӂѹѸҡҾ. ҫҦѹӁ ҲҸґҧҔҸ ҝҽӉ ҸӁ ҏҦҽҝѹґ ҔҏґҽҔҸѸ ҰҸ ҡҲҸґҏ, ҔҸҊҦҽ ѦҊҸ ґӀқѹѸҽҾ ѸґҽҦҗҽҾ ҦҸҍҧ ҏқѦңҽґҽ Ҕ үґѹ ѷҽҝҰѹӂѦӊҏ. ҭҸҲҸѸ ҸӁ ѦѦ ҰҝҸҡҲѹґ. ҫӁ, ӊҡҰҸѸѹӁҽѦҲ ѸҽѸҽ, қӃґ ҲҽҔҸҮ ңѦ ӊӃҡҸҔѹҮ, ҔҽҔ Ҿ, ҡѦҝҸҊґҽӉӃҮ ѹ ӊҡѦҊҦҽ ӁҸҡѹґ ӊ ҔҽҝѸҽӁѦ ҦґҾ ҦѦҲѦҮ ҔҸӁӍѦҲӃ. (ѢҘҩѤ) ҈ҽѸѦҲѹѸ, ҍҲҸ ӊ ҸҲґѹҍѹѦ ҸҲ ҽӁҊґѹҮҡҔҸҊҸ ҾӉӃҔҽ ӁѦ ҲҸґҧҔҸ ҰҝѹґҽҊҽҲѦґҧӁҸѦ ѸҽґҸҊҸ ҰҸґӀҡҽ ӊ (19), ӁҸ ѹ ҰҝѹґҽҊҽҲѦґҧӁҸѦ қҸґҧҗҸҊҸ ҰҸґӀҡҽ ӊ (20) ҸҝѹѦӁҲѹҝҸӊҽӁҸ Ӂҽ ӁҸҝѸҏ. ѷҽҔ, ҸқѦ ҊҸҝӁӃѦ ҃ѦҰѹ ӊ (19) ҾӊґҾӀҲҡҾ ӁѦӊӃҡҸҔѹѸѹ. ғ ҙҲҸѸ ҡѸӃҡґѦ ҝҏҡҡҔѹҮ ҾӉӃҔ ӁҽҰҸѸѹӁҽѦҲ ҽӁҊґѹҮҡҔѹҮ (ҡҝ. ҰҝѹѸѦҝ (18)). ҫҦӁҽҔҸ, ӊ ҸҲґѹҍѹѦ ҸҲ ҽӁҊґѹҮҡҔҸҊҸ ҾӉӃҔҽ, ҰҝѹґҽҊҽҲѦґҧӁҸѦ қҸґҧҗҸҊҸ ҰҸґӀҡҽ ӴՂԙӹӪӠӡ ӊ (20) ҲҸңѦ ѸҽҝҔѹҝҸӊҽӁҸ. ѨӁҽҍѦ ҊҸӊҸҝҾ, (20) ҡҸҸқӂҽѦҲ ӁҽѸ ӁѦ ҲҸґҧҔҸ Ҹ ҲҸѸ, ҍҲҸ ҰҝҽҦѦҦ қӃґ ҲҽҔҸҊҸ ңѦ ҝҸҡҲҽ, ҔҽҔ ѹ ҊҸӊҸҝҾӂѹҮ, ӁҸ ѹ Ҹ ҲҸѸ, ҍҲҸ ҸӁ қӃґ ӊӃҡҸҔѹѸ ҍѦґҸӊѦҔҸѸ (ҝҽӊӁҸ ҔҽҔ ѹ ѦҊҸ ҰҝҽӊӁҏҔ). ғ ҸҲґѹҍѹѦ ҸҲ ҡҝҽӊӁѹҲѦґҧӁӃѽ ҸқҸҝҸҲҸӊ ҡҸ ӉӁҽҍѦӁѹѦѸ ҲҸңҦѦҡҲӊҽ ѹ ѸѦӁѦѦ ӊӃҡҸҔҸҮ ҡҲѦҰѦӁѹ ҔҽҍѦҡҲӊҽ, ӍҸҝѸӃ ҡҝҽӊӁѹҲѦґҧӁҸҮ ҡҲѦҰѦӁѹ ҡҸ ӉӁҽҍѦӁѹѦѸ қҸґѦѦ ӊӃҡҸҔҸҮ ҡҲѦҰѦӁѹ ҔҽҍѦҡҲӊҽ ҡҍѹҲҽӀҲҡҾ ҰҝҸҲҸҲѹҰѹҍѦҡҔѹ ӁѦѸҽҝҔѹҝҸӊҽӁӁӃѸѹ (Bierwisch 1989, Cuzzolin & Lehmann 2004, Kennedy 1999, Kennedy & McNally 2005, Lyons 1969, 1977). ѨӁӃѸѹ ҡґҸӊҽѸѹ, ҍҽҡҲҸ ҏҲӊѦҝңҦҽѦҲҡҾ, ҍҲҸ

472

ѣһѣѢү ѷҩѨҷѰҹѨѢѨѢү

ҰҝѹґҽҊҽҲѦґҧӁӃѦ-ҽӁҲҸӁѹѸӃ ӊ ӍҸҝѸѦ ҡҝҽӊӁѹҲѦґҧӁҸҮ ҡҲѦҰѦӁѹ ӁѦ ҸҝѹѦӁҲѹҝҸӊҽӁӃ Ӂҽ ӁҸҝѸҏ. ҭҸҙҲҸѸҏ ѸӃ ӊҰҸґӁѦ ѸҸңѦѸ ҡҔҽӉҽҲҧ, ҍҲҸ ҸҦӁҽ ѸҽҗѹӁҽ ҦѦҗѦӊґѦ ҦҝҏҊҸҮ, ӁѦ ѹѸѦҾ Ұҝѹ ҙҲҸѸ ӊ ӊѹҦҏ, ҍҲҸ ҸӁѹ – ҦѦҗѦӊӃѦ. ѩҽӁӁҸѦ ҰҸґҸңѦӁѹѦ ҰҝѹӊҸҦѹҲҡҾ ӊ қҸґҧҗѹӁҡҲӊѦ ҝҽқҸҲ ҰҸ ҔҽҲѦҊҸҝѹѹ ҔҸѸҰҽҝҽҲѹӊҽ. ҫҦӁҽҔҸ ҡґѦҦҏѦҲ ҸҲѸѦҲѹҲҧ, ҍҲҸ ҦҽңѦ ҦґҾ ҽӁҊґѹҮҡҔҸҊҸ ҾӉӃҔҽ ҙҲҸ ҏҲӊѦҝңҦѦӁѹѦ ӁѦ ҾӊґҾѦҲҡҾ ҽқҡҸґӀҲӁҸ ӊѦҝӁӃѸ. ҘҽҔ ҡҰҝҽӊѦҦґѹӊҸ ҸҲѸѦҍѦӁҸ ӊ ҝҽқҸҲѦ Croft & Cruse (2004: 178), ҦҽңѦ ӊ ҽӁҊґѹҮҡҔҸѸ ҾӉӃҔѦ ӍҸҝѸӃ ҡҝҽӊӁѹҲѦґҧӁҸҮ ҡҲѦҰѦӁѹ ѸҸҊҏҲ қӃҲҧ ҸҝѹѦӁҲѹҝҸӊҽӁӃ Ӂҽ ӁҸҝѸҏ. ҶҲҸ, ӊ ҰѦҝӊҏӀ ҸҍѦҝѦҦҧ, ҸҲӁҸҡѹҲҡҾ Ҕ ҰҝѹґҽҊҽҲѦґҧӁӃѸ, ѹѸѦӀӂѹѸ ҦӊѦ ӍҸҝѸӃ ҡҝҽӊӁѹҲѦґҧӁҸҮ ҡҲѦҰѦӁѹ – ҽӁҽґѹҲѹҍѦҡҔҏӀ ѹ ҡѹӁҲѦҲѹҍѦҡҔҏӀ. ҘҝҸӍҲ ѹ ҘҝҏӉ ҰҝѹӊҸҦҾҲ ҡґѦҦҏӀӂѹѦ ҰҝѹѸѦҝӃ ҦґҾ ѹґґӀҡҲҝҽ҃ѹѹ ҦҽӁӁҸҊҸ ҝҽӉґѹҍѹҾ: (21) Process X is fast, but it’s nonetheless slower than process Y. (22) #Process X is fast, but it’s nonetheless more slow than process Y. ҘҝҸӍҲ ѹ ҘҝҏӉ ҏҲӊѦҝңҦҽӀҲ, ҍҲҸ ӊ ҡґҏҍҽѦ Ѧҡґѹ ҰҝѹґҽҊҽҲѦґҧӁҸѦ ѹѸѦѦҲ ҦӊѦ ӍҸҝѸӃ ҔҸѸҰҽҝҽҲѹӊҽ, ҡѹӁҲѦҲѹҍѦҡҔҽҾ ӍҸҝѸҽ ҾӊґҾѦҲҡҾ ӁѦѸҽҝҔѹҝҸӊҽӁӁҸҮ, ҽ ҽӁҽґѹҲѹҍѦҡҔҽҾ – ѸҽҝҔѹҝҸӊҽӁӁҸҮ. ҋґѦҦҸӊҽҲѦґҧӁҸ, ҰҝѦҦґҸңѦӁѹѦ ӊ (21) – ҰҝѹѦѸґѦѸҸ, ҰҸҡҔҸґҧҔҏ ҡѹӁҲѦҲѹҍѦҡҔѹҮ ҔҸѸҰҽҝҽҲѹӊ slower ӁѹҍѦҊҸ ӁѦ ҊҸӊҸҝѹҲ Ҹ ҡҔҸҝҸҡҲѹ ҰҝҸ҃Ѧҡҡҽ ҸҲӁҸҡѹҲѦґҧӁҸ ӁҸҝѸӃ, ҽ ҲҸґҧҔҸ ҡҸҸқӂҽѦҲ, ҍҲҸ ҰҝҸ҃Ѧҡҡ X ѸѦҦґѦӁӁѦѦ, ҍѦѸ ҰҝҸ҃Ѧҡҡ Y. Ѩ, ӁҽҰҝҸҲѹӊ, ҰҝѦҦґҸңѦӁѹѦ ӊ (22) ӁѦҰҝѹѦѸґѦѸҸ, ѹқҸ ҽӁҽґѹҲѹҍѦҡҔҽҾ ӍҸҝѸҽ more slow ҸӉӁҽҍҽѦҲ, ҍҲҸ ҰҝҸ҃Ѧҡҡ X Ӂҽ ҡҽѸҸѸ ҦѦґѦ ѸѦҦґѦӁӁӃҮ, ҍҲҸ ҰҝҸҲѹӊҸҝѦҍѹҲ ҰѦҝӊҸҮ ҍҽҡҲѹ ҰҝѦҦґҸңѦӁѹҾ. үӁҽґѹӉ ҽҏҲѦӁҲѹҍӁӃѽ ҰҝѹѸѦҝҸӊ, ӁҽҮҦѦӁӁӃѽ ӊ ҔҸҝҰҏҡѦ, ҍҽҡҲѹҍӁҸ ҰҸҦҲӊѦҝңҦҽѦҲ ҦҽӁӁҏӀ ҊѹҰҸҲѦӉҏ. ҩҽҡҡѸҸҲҝѹѸ ҰҝѦҦґҸңѦӁѹҾ (23) ѹ (24): (23) Memories obviously lingered of the War Communism strategy of trying to divide richer from poorer peasants. (BNC) (24) By comparison, socially created inequality ‘consists of the different privileges which some men enjoy to the prejudice of others, such as that of being more rich, more honoured, more powerful, or even in a position to exact obedience’. (BNC) ѨҡҰҸґҧӉҸӊҽӁѹѦ ҡѹӁҲѦҲѹҍѦҡҔҸҊҸ ҔҸѸҰҽҝҽҲѹӊҽ ӊ (23) ӁѦ ҰҸҦҝҽӉҏѸѦӊҽѦҲ, ҍҲҸ ҔҝѦҡҲҧҾӁѦ, ҸҲӁѦҡѦӁӁӃѦ Ҕ ҔҽҲѦҊҸҝѹѹ ‘richer’, Ӂҽ ҡҽѸҸѸ ҦѦґѦ қҸҊҽҲӃ. ҨҸҝѸҽ ҡҝҽӊӁѹҲѦґҧӁҸҮ ҡҲѦҰѦӁѹ richer ҲҸґҧҔҸ ҏҔҽӉӃӊҽѦҲ Ӂҽ ҲҸ, ҍҲҸ ҸҦӁѹ ҔҝѦҡҲҧҾӁѦ ӁѦҡҔҸґҧҔҸ қҸҊҽҍѦ ҦҝҏҊѹѽ ҔҝѦҡҲҧҾӁ. ѢҽҰҝҸҲѹӊ, ҽӁҽґѹҲѹҍѦҡҔҽҾ ӍҸҝѸҽ ҡҝҽӊӁѹҲѦґҧӁҸҮ ҡҲѦҰѦӁѹ ӊ (24) ҰҸҦҝҽӉҏѸѦӊҽѦҲ, ҍҲҸ ѹ ҡҝҽӊӁѹӊҽѦѸӃҮ ҸқӆѦҔҲ, ѹ ҲҸҍҔҽ ҝѦӍѦҝѦӁ҃ѹѹ (Ҳ.Ѧ. ҸқӆѦҔҲ, ҡ ҔҸҲҸҝӃѸ ҸҡҏӂѦҡҲӊґҾѦҲҡҾ ҡҝҽӊӁѦӁѹѦ) – қҸҊҽҲӃ, ҰҝҸҡҲҸ ҸҦӁѹ қҸҊҽҍѦ ҦҝҏҊѹѽ. ҘҝҸӍҲ ѹ ҘҝҏӉ ҲҽҔңѦ ҙҔҡҲҝҽҰҸґѹҝҏӀҲ ҦҽӁӁӃҮ ҝѦӉҏґҧҲҽҲ Ӂҽ ҦҝҏҊѹѦ ҾӉӃҔѹ ѹ ӊӃҦӊѹҊҽӀҲ ҊѹҰҸҲѦӉҏ Ҹ ҲҸѸ, ҍҲҸ ӊ ҾӉӃҔҽѽ, ҊҦѦ ҰҝѹґҽҊҽҲѦґҧӁӃѦ ѹѸѦӀҲ ҦӊѦ

ѢѣҋҘҫһѻҘҫ үҘҋѨҫҴ ҫ ҭҩѨһүҐүѷѣһѻѢѳҤ-үѢѷҫѢѨҴүҤ

473

ӍҸҝѸӃ ҔҸѸҰҽҝҽҲѹӊҽ (ҽӁҽґѹҲѹҍѦҡҔҏӀ ѹ ҡѹӁҲѦҲѹҍѦҡҔҏӀ), ҽӁҽґѹҲѹҍѦҡҔҽҾ ӍҸҝѸҽ қҏҦѦҲ ҸҝѹѦӁҲѹҝҸӊҽӁҽ Ӂҽ ӁҸҝѸҏ, ҽ ҡѹӁҲѦҲѹҍѦҡҔҽҾ – ӁѦҲ. ҘҝҸѸѦ ҲҸҊҸ, ҸӁѹ ҰҝѦҦґҽҊҽӀҲ, ҍҲҸ ҡҝҽӊӁѹҲѦґҧӁҽҾ ӍҸҝѸҽ ҰҝѹґҽҊҽҲѦґҧӁӃѽ, ѹѸѦӀӂѹѽ ԽӹԊӸӪӹ ҽӁҽґѹҲѹҍѦҡҔѹҮ ҔҸѸҰҽҝҽҲѹӊ, ӊҡѦҊҦҽ қҏҦѦҲ ӁѦѸҽҝҔѹҝҸӊҽӁҽ (Croft & Cruse 2004: 178). ѨӁӃѸѹ ҡґҸӊҽѸѹ, ѹѽ ҊѹҰҸҲѦӉҽ ҡҸҡҲҸѹҲ ӊ ҲҸѸ, ҍҲҸ ӍҸҝѸӃ ҽӁҽґѹҲѹҍѦҡҔҸҊҸ ҔҸѸҰҽҝҽҲѹӊҽ ӁѦѸҽҝҔѹҝҸӊҽӁӃ ҏ ҰҝѹґҽҊҽҲѦґҧӁӃѽ, ӁѦ ѹѸѦӀӂѹѽ ҡѹӁҲѦҲѹҍѦҡҔҸҊҸ ҔҸѸҰҽҝҽҲѹӊҽ, ѹ ѸҽҝҔѹҝҸӊҽӁӃ ҏ ҰҝѹґҽҊҽҲѦґҧӁӃѽ, ѹѸѦӀӂѹѽ ҔҽҔ ҡѹӁҲѦҲѹҍѦҡҔѹҮ, ҲҽҔ ѹ ҽӁҽґѹҲѹҍѦҡҔѹҮ ҔҸѸҰҽҝҽҲѹӊ. Ҙ ҡҸңҽґѦӁѹӀ, ҦҽӁӁҽҾ ҊѹҰҸҲѦӉҽ ҸҡҲҽѦҲҡҾ ӁѦҡҔҸґҧҔҸ ҏѸҸӉҝѹҲѦґҧӁҸҮ, ҰҸҡҔҸґҧҔҏ ҘҝҸӍҲ ѹ ҘҝҏӉ ҰҝѹӊҸҦҾҲ ґѹҗҧ ӁѦҡҔҸґҧҔҸ ҡҔҸӁҡҲҝҏѹҝҸӊҽӁӁӃѽ ҰҝѹѸѦҝҸӊ ѹӉ ҽӁҊґѹҮҡҔҸҊҸ ҾӉӃҔҽ, ҽ ҲҽҔңѦ ҸҲѸѦҍҽӀҲ, ҍҲҸ ҰҝѹґҽҊҽҲѦґҧӁӃѦ ӊ ҊҝѦҍѦҡҔҸѸ ҾӉӃҔѦ ҰҸґӁҸҡҲҧӀ ҡҸҸҲӊѦҲҡҲӊҏӀҲ ҰҝѦҦґҸңѦӁӁҸҮ ѸҸҦѦґѹ. ҫҦӁҽҔҸ ӊ ѹѽ ҝҽқҸҲѦ ӁѦ ҰҝѹӊҸҦѹҲҡҾ Ӂѹ ҸҦӁҸҊҸ ҰҝѹѸѦҝҽ ѹӉ ҊҝѦҍѦҡҔҸҊҸ ҾӉӃҔҽ ѹ ӁѦ ҽӁҽґѹӉѹҝҏѦҲҡҾ ҽҏҲѦӁҲѹҍӁҸѦ ҏҰҸҲҝѦқґѦӁѹѦ ҽӁҊґѹҮҡҔѹѽ ҰҝѹґҽҊҽҲѦґҧӁӃѽ. ғ ҦҽӁӁҸҮ ҡҲҽҲҧѦ, ѸӃ ҰҸҰӃҲҽѦѸҡҾ ҰҝҸӊѦҝѹҲҧ ҦҽӁӁҏӀ ҊѹҰҸҲѦӉҏ Ӂҽ ѸҽҲѦҝѹҽґѦ ҝҏҡҡҔҸҊҸ ҾӉӃҔҽ. ҩҏҡҡҔѹѦ ҰҝѹґҽҊҽҲѦґҧӁӃѦ ҡҔҽґҾҝӁҸ-ҽӁҲҸӁѹѸѹҍѦҡҔҸҊҸ ҲѹҰҽ ӊ ҙҲҸѸ ҡѸӃҡґѦ ҰҝѦҦҡҲҽӊґҾӀҲ ӊѦҡҧѸҽ ҏҦҽҍӁӃҮ ҸқӆѦҔҲ ҦґҾ ѹҡҡґѦҦҸӊҽӁѹҾ, ҰҸҡҔҸґҧҔҏ ѸӁҸҊѹѦ ѹӉ Ӂѹѽ ѹѸѦӀҲ ҔҽҔ ҡѹӁҲѦҲѹҍѦҡҔѹѦ, ҲҽҔ ѹ ҽӁҽґѹҲѹҍѦҡҔѹѦ ӍҸҝѸӃ ҔҸѸҰҽҝҽҲѹӊҽ. ҋҝҽӊӁѹѸ ҰҝѹѸѦҝӃ (25) ѹ (26): (25) ҷҽҡҔҸӊ ӊӃҗѦ ҭҏҊҽҍѦӊҸҮ Ӂҽ ҃ѦґҏӀ ҊҸґҸӊҏ. (ѢҘҩѤ) (26) ғҦҽґѹ ҍҏҲҧ ҲѦѸӁѦґѹ ҡѹґҏҙҲӃ ѣӊҝҸҰӃ – ҦӊҏѽҙҲҽңӁӃѦ ҦҸѸѹҔѹ, ѹҊґҽ ҝҽҲҏҗѹ ѹ қҸґѦѦ ӊӃҡҸҔҽҾ – ҡҸқҸҝҽ. (ѢҘҩѤ) ҋѹӁҲѦҲѹҍѦҡҔҽҾ ӍҸҝѸҽ ҔҸѸҰҽҝҽҲѹӊҽ ӊ (25) ҸӉӁҽҍҽѦҲ, ҍҲҸ ҸҦѹӁ ҰѦӊѦ҃ ӊӃҗѦ ҦҝҏҊҸҊҸ, ӁҸ ӁѦ ҲҸ, ҍҲҸ Ҹқҽ ҸӁѹ ӊӃҡҸҔѹѦ. ҷҸґѦѦ ҲҸҊҸ, ҔҽҔ ѹӉӊѦҡҲӁҸ, ѹ ҭҏҊҽҍѦӊҽ ѹ ҷҽҡҔҸӊ – ӊѦҡҧѸҽ ӁѹӉҔҸҊҸ ҝҸҡҲҽ. ѷҽҔѹѸ ҸқҝҽӉҸѸ, ҡѹӁҲѦҲѹҍѦҡҔҽҾ ӍҸҝѸҽ ӴՂԶӾ ӊ (25) ӁѦѸҽҝҔѹҝҸӊҽӁҽ. ғ ҸҲґѹҍѹѦ ҸҲ (25), ӍҸҝѸҽ ҡҝҽӊӁѹҲѦґҧӁҸҮ ҡҲѦҰѦӁѹ ӊ (26) ҏҔҽӉӃӊҽѦҲ Ӂҽ ҲҸ, ҍҲҸ ҸқѦ ѹҊґӃ Ӂҽ ҡҽѸҸѸ ҦѦґѦ ӊӃҡҸҔѹѦ, ҰҝҸҡҲҸ ѹҊґҽ ҡҸқҸҝҽ ѦӂѦ ӊӃҗѦ, ҍѦѸ ѹҊґҽ ҝҽҲҏҗѹ. ѩҽӁӁҸѦ ӁҽқґӀҦѦӁѹѦ ҰҸҦҲӊѦҝңҦҽѦҲ ҊѹҰҸҲѦӉҏ ҘҝҸӍҲҽ ѹ ҘҝҏӉҽ (ҡѸ. ҲҽҔңѦ ҭҸ҃ѦґҏѦӊҡҔѹҮ 1977). ѯҲҸқӃ ҏқѦҦѹҲҧҡҾ ӊ ҰҝҽӊѹґҧӁҸҡҲѹ ӊӃӊҸҦҽ, ҝҽҡҡѸҸҲҝѹѸ ѦӂѦ ӁѦҡҔҸґҧҔҸ ҰҝѹѸѦҝҸӊ ҡ ҦҝҏҊѹѸѹ ҰҝѹґҽҊҽҲѦґҧӁӃѸѹ-ҽӁҲҸӁѹѸҽѸѹ. (27) ѣҡґѹ ӊ ҲӊҸѦѸ ѽҸґҸҦѹґҧӁѹҔѦ ѦҡҲҧ ѦҦҽ, ҲӃ ҸҦѦҲ, ҏ ҲѦқҾ ѦҡҲҧ ҔҝӃҗҽ ӁҽҦ ҊҸґҸӊҸҮ ѹ ҰҸҡҲѦґҧ, ҲӃ қҸҊҽҍѦ, ҍѦѸ 75 ҰҝҸ҃ѦӁҲҸӊ ґӀҦѦҮ ӊ ҙҲҸѸ ѸѹҝѦ. (ѢҘҩѤ) (28) ҫҍѦӁҧ ѸӁҸҊѹѦ ңѹҲѦґѹ ӁҽҗѦҮ ҡҲҝҽӁӃԀ ҡҍѹҲҽӀҲ / ҍҲҸ ѹѸ ӊҡѦ ҦҸґңӁӃ / ҦҸґңӁҸ ҊҸҡҏҦҽҝҡҲӊҸ / ѹѸ ҦҸґңӁӃ ҸґѹҊҽҝѽѹ / ѹѸ ҦҸґңӁӃ қҸґѦѦ қҸҊҽҲӃѦ ҡҸҡѦҦѹ / ґӀҦѹ ҏӊѦҝѦӁӃ / ҍҲҸ ӊҡѦ ҙҲѹ ґӀҦѹ ңѹӊҏҲ Ӊҽ ѹѽ ҡҍҺҲ. (ѢҘҩѤ)

474

ѣһѣѢү ѷҩѨҷѰҹѨѢѨѢү

(29) ѣҡҲҧ ѸӁѦӁѹѦ, ҍҲҸ ҝҸңҦѦӁѹѦ ҝѦқѦӁҔҽ ӉҽҡҲҽӊґҾѦҲ ҊҸӁӂѹҔҽ қӃҲҧ қҸґѦѦ ҸҡҲҸҝҸңӁӃѸ ѹ ҦѦґҽѦҲ ѦҊҸ ѸѦҦґѦӁӁӃѸ. ѣҡґѹ ҍѦҡҲӁҸ, ҲҸ Ҿ ӁѦ ӉӁҽӀ. ҭҸҡґѦ ҝҸңҦѦӁѹҾ ҡӃӁҽ Ҿ ӊҡѦҊҸ ґѹҗҧ Ұҽҝҏ ҝҽӉ ҰҸҾӊґҾґҡҾ Ӂҽ ҲҝҽҡҡѦ. ѢѦҔҸҲҸҝӃѦ ҦѦҮҡҲӊѹҲѦґҧӁҸ ҏҲӊѦҝңҦҽӀҲ, ҍҲҸ ҔҽңҦӃҮ ҝѦқѦӁҸҔ ҦѦґҽѦҲ ҊҸӁӂѹҔҽ ѸѦҦґѦӁӁѦѦ Ӂҽ ҡѦҔҏӁҦҏ. (ѢҘҩѤ) (30) ѷҽҔ, ҔҸҊҦҽ ӁҏңӁҸ ҰҝѦҦҡҲҽӊѹҲҧ ӁѦѽҸҝҸҗѦҊҸ ҍѦґҸӊѦҔҽ, ѦҊҸ ҦѦґҽӀҲ ҡҏѦҲґѹӊӃѸ, қӃҡҲҝҸӁҸҊѹѸ, ңѦҡҲѹҔҏґѹҝҏӀӂѹѸ; ӊ ҝѹѸҡҔҸҮ ҔҸѸѦҦѹѹ ҰҸ ҲҝҽҦѹ҃ѹѹ, ҰѦҝѦӁҾҲҸҮ ҸҲ ҡҝѦҦӁѦҮ ҔҸѸѦҦѹѹ ѹ ӁҸӊҸҮ, ҦґҾ ҡҲҽҝѹҔҸӊ, ӀӁҸҗѦҮ, ӉҽѸҏңӁѹѽ ңѦӁӂѹӁ ѹ ӊҸѦӁӁӃѽ ҰҸґҽҊҽґҽҡҧ қҸґѦѦ ѸѦҦґѦӁӁҽҾ ҰҸѽҸҦҔҽ, ҽ қӃҡҲҝҽҾ, ҡ қҸґҧҗѹѸ ҦӊѹңѦӁѹѦѸ, ҦґҾ ҝҽқҸӊ, ҰҽҝҽӉѹҲҸӊ, ҝӃқҽҔҸӊ, ҡґҏңҽӁҸҔ. (ѢҘҩѤ) Ѩ ҦѦҮҡҲӊѹҲѦґҧӁҸ, ӊ ҡҸҸҲӊѦҲҡҲӊѹѹ ҡ ҊѹҰҸҲѦӉҸҮ ҘҝҸӍҲҽ ѹ ҘҝҏӉҽ, ҽӁҽґѹҲѹҍѦҡҔѹѦ ӍҸҝѸӃ ҔҸѸҰҽҝҽҲѹӊҽ ӊ ҰҝѹѸѦҝҽѽ (28) ѹ (30) ҰҸҦҝҽӉҏѸѦӊҽӀҲ, ҍҲҸ ҝѦӍѦҝѦӁҲӃ қҸҊҽҲӃ ѹ ѸѦҦґѦӁӁӃ, ҡҸҸҲӊѦҲҡҲӊѦӁӁҸ. ғѦҝҸҾҲӁҸ, ҙҲҸ ҡӊҾӉҽӁҸ ҡ ӁҽґѹҍѹѦѸ ҰҸґҸңѹҲѦґҧӁӃѽ (ҽ ҡҲҽґҸ қӃҲҧ ѸҽҝҔѹҝҸӊҽӁӁӃѽ) ӍҸҝѸ ӊ ҡҸҡҲҽӊѦ ҽӁҽґѹҲѹҍѦҡҔҸҊҸ ҔҸѸҰҽҝҽҲѹӊҽ. Ѩ, ӁҽҸқҸҝҸҲ, ӍҸҝѸӃ ҡѹӁҲѦҲѹҍѦҡҔҸҊҸ ҔҸѸҰҽҝҽҲѹӊҽ ӊ (27) ѹ (29) ӁѹҍѦҊҸ ӁѦ ҡҸҸқӂҽӀҲ ӁҽѸ Ҹ ҦҸҡҲҽҲҔѦ ѹ ҡҔҸҝҸҡҲѹ ѹѽ ҝѦӍѦҝѦӁҲҸӊ. ѢҽқґӀҦѦӁѹѦ, ҍҲҸ ӍҸҝѸӃ ҡҝҽӊӁѹҲѦґҧӁҸҮ ҡҲѦҰѦӁѹ ӊ (27) ѹ (29) ӁѦѸҽҝҔѹҝҸӊҽӁӃ ҰҸҦҲӊѦҝңҦҽѦҲҡҾ ҔҸӁҲѦҔҡҲҸѸ, ѹѸҰґѹ҃ѹҝҏӀӂѹѸ, ҍҲҸ ҡҏқӆѦҔҲ ӊ (27) Ӂҽ ҡҽѸҸѸ ҦѦґѦ ӁѦқҸҊҽҲ, ҽ ҊҸӁӂѹҔѹ ӊ (29) ҸҲӁӀҦҧ ӁѦ ѸѦҦґѦӁӁӃѦ. ҫҦӁҽҔҸ ӁѦ ӊҡѦ ҲҽҔ ҰҝҸҡҲҸ. үӁҽґѹӉ ҔҸҝҰҏҡӁҸҊҸ ѸҽҲѦҝѹҽґҽ ҰҸҔҽӉӃӊҽѦҲ, ҍҲҸ ҽӁҽґѹҲѹҍѦҡҔѹѦ ӍҸҝѸӃ ҡҝҽӊӁѹҲѦґҧӁҸҮ ҡҲѦҰѦӁѹ ӊҰҸґӁѦ ѸҸҊҏҲ ҸқҸӉӁҽҍҽҲҧ ҰҝѹӉӁҽҔ, ӁѦ ҸҝѹѦӁҲѹҝҸӊҽӁӁӃҮ Ӂҽ ӁҸҝѸҏ, ҡѸ. (31): (31) ҭҝѦҔҝҽҡӁҸ ҰҸҦҸҮҦѦҲ ҰҸҦҡҲѦҊѹӊҽӀӂѹҮ ҝҙҰ (112 ҏҦ/ѸѹӁ) ѹґѹ қҸґѦѦ ѸѦҦґѦӁӁӃҮ ѽѹҰ-ѽҸҰ (90 ҏҦ/ѸѹӁ). (ѢҘҩѤ) ҈ҽѸѦҲѹѸ, ҍҲҸ ҲҽӁѦ҃, ӁҽӉӊҽӁӁӃҮ “қҸґѦѦ ѸѦҦґѦӁӁӃѸ” ӊ ҰҝѹѸѦҝѦ (31), ҸҲӁӀҦҧ ӁѦ ҾӊґҾѦҲҡҾ ѸѦҦґѦӁӁӃѸ ҸҲӁҸҡѹҲѦґҧӁҸ ӁҸҝѸӃ ҦґҾ ҲҽӁ҃Ѧӊ. ҭҸҦҸқӁҸ ҡѹӁҲѦҲѹҍѦҡҔѹѸ ӍҸҝѸҽѸ ҡҝҽӊӁѹҲѦґҧӁҸҮ ҡҲѦҰѦӁѹ, ҽӁҽґѹҲѹҍѦҡҔѹҮ ҔҸѸҰҽҝҽҲѹӊ ӊ (31) ҰҝҸҡҲҸ ҏҔҽӉӃӊҽѦҲ Ӂҽ ҲҸ, ҍҲҸ ҸҦѹӁ ҲҽӁѦ҃ ѸѦҦґѦӁӁѦѦ ҦҝҏҊҸҊҸ, ӁҸ ӁѹҍѦҊҸ ӁѦ ҊҸӊҸҝѹҲ Ҹқ ѹѽ ҝѹҲѸѦ ҸҲӁҸҡѹҲѦґҧӁҸ ӁҸҝѸӃ ҦґҾ ҲҽӁ҃Ѧӊ. ҴҸңӁҸ ҰҝѦҦҰҸґҸңѹҲҧ, ҍҲҸ ӍҸҝѸӃ ҽӁҽґѹҲѹҍѦҡҔҸҊҸ ҔҸѸҰҽҝҽҲѹӊҽ қӃӊҽӀҲ ӁѦѸҽҝҔѹҝҸӊҽӁӁӃѸѹ ҲҸґҧҔҸ ӊ ӍҏӁҔ҃ѹѹ ҸҰҝѦҦѦґѦӁѹҾ, ҰҸҡҔҸґҧҔҏ ҡѹӁҲѦҲѹҍѦҡҔѹѦ ӍҸҝѸӃ, ҔҽҔ ҰҝҽӊѹґҸ, ӁѦ ѸҸҊҏҲ ҏҰҸҲҝѦқґҾҲҧҡҾ ҽҲҝѹқҏҲѹӊӁҸ, ѹ ӊ ҦҽӁӁҸѸ ҡґҏҍҽѦ ҽӁҽґѹҲѹҍѦҡҔѹѦ ӍҸҝѸӃ ѸҸҊҏҲ “қҝҽҲҧ Ӂҽ ҡѦқҾ” ӊӃҝҽңѦӁѹѦ ӁѦѸҽҝҔѹҝҸӊҽӁӁӃѽ ҡҝҽӊӁѹҲѦґҧӁӃѽ ҸҲӁҸҗѦӁѹҮ. ѣҡґѹ ҙҲҸ ҰҝѦҦҰҸґҸңѦӁѹѦ ӊѦҝӁҸ, ҲҸ ҡґѦҦҏѦҲ ҸңѹҦҽҲҧ, ҍҲҸ ҽӁҽґѹҲѹҍѦҡҔѹѦ ӍҸҝѸӃ, ҏҰҸҲҝѦқґѦӁӁӃѦ ӊ ӍҏӁҔ҃ѹѹ ҰҝѦҦѹҔҽҲѹӊӁҸҊҸ ҍґѦӁҽ, ӊҡѦҊҦҽ ѸҽҝҔѹҝҸӊҽӁӃ. үӁҽґѹӉ ӍҽҔҲѹҍѦҡҔҸҊҸ ѸҽҲѦҝѹҽґҽ, ҸҦӁҽҔҸ, ҸҰҝҸӊѦҝҊҽѦҲ ҦҽӁӁҸѦ ҰҝѦҦҰҸґҸңѦӁѹѦ, ѹқҸ ҰҝѦҦѹҔҽҲѹӊӁҸ ҏҰҸҲҝѦқґѦӁ-

ѢѣҋҘҫһѻҘҫ үҘҋѨҫҴ ҫ ҭҩѨһүҐүѷѣһѻѢѳҤ-үѢѷҫѢѨҴүҤ

475

ӁӃѦ ҽӁҽґѹҲѹҍѦҡҔѹѦ ӍҸҝѸӃ ҡҝҽӊӁѹҲѦґҧӁҸҮ ҡҲѦҰѦӁѹ ҲҽҔңѦ ѸҸҊҏҲ ҸқҸӉӁҽҍҽҲҧ ҰҝѹӉӁҽҔ, ӁѦ ҸҝѹѦӁҲѹҝҸӊҽӁӁӃҮ Ӂҽ ӁҸҝѸҏ, ҔҽҔ, ӁҽҰҝѹѸѦҝ, ӊ (32): (32) ҋѽҸңѹҮ ҰҝҸѦҔҲ “қѦӉҸҰҽҡӁҸҮ ӍҸҝѸҏґӃ” ҰҝѦҦґҽҊҽѦҲ ѹ ҔҸӁҡҲҝҏҔҲҸҝ үґҽӁ ѩңѦӁҔѹӁҡ, ҰҸҾҡӁҾҾ: “ҘҽҔ қӃ Ӂѹ ҰӃҲҽґѹҡҧ ҡҦѦґҽҲҧ ѸҽҗѹӁӃ ҨҸҝѸҏґӃ-1 қҸґѦѦ ѸѦҦґѦӁӁӃѸѹ, ӊҡѦ ҝҽӊӁҸ ҸӁѹ қҏҦҏҲ ӊӉґѦҲҽҲҧ Ұҝѹ ҔҸӁҲҽҔҲѦ ҔҸґѦҡҽѸѹ. (ѢҘҩѤ) ғ ҙҲҸҮ ҡӊҾӉѹ ѸҸңӁҸ ҡҦѦґҽҲҧ ҡґѦҦҏӀӂѹҮ ӊӃӊҸҦ. ҋѹӁҲѦҲѹҍѦҡҔѹѦ ӍҸҝѸӃ ҔҸѸҰҽҝҽҲѹӊҽ ӁѦѸҽҝҔѹҝҸӊҽӁӃ, Ҳ.Ѧ. ҸӁѹ ӊӃҝҽңҽӀҲ ҰҝѹӉӁҽҔѹ, ӁѦ ҸҝѹѦӁҲѹҝҸӊҽӁӁӃѦ Ӂҽ ӁҸҝѸҏ. 3 ғ ҸҲґѹҍѹѦ ҸҲ ҡѹӁҲѦҲѹҍѦҡҔѹѽ ӍҸҝѸ, ҽӁҽґѹҲѹҍѦҡҔѹѦ ӍҸҝѸӃ ҡҝҽӊӁѹҲѦґҧӁҸҮ ҡҲѦҰѦӁѹ ѹѸѦӀҲ ҲѦӁҦѦӁ҃ѹӀ ӊӃҝҽңҽҲҧ ҰҝѹӉӁҽҔѹ, ҸҝѹѦӁҲѹҝҸӊҽӁӁӃѦ Ӂҽ ӁҸҝѸҏ. ҋґѦҦҏѦҲ ҰҸҦҍѦҝҔӁҏҲҧ, ҍҲҸ ҙҲҸ ԽӾӰԌӾӰՁӠԋ, ҽ ӁѦ ҽқҡҸґӀҲӁҸѦ ҝҽӉґѹҍѹѦ. ҘҽҔ ҰҸҔҽӉҽґ ҽӁҽґѹӉ ҽҏҲѦӁҲѹҍӁҸҊҸ ҾӉӃҔҸӊҸҊҸ ѸҽҲѦҝѹҽґҽ, ӊ ӁѦҔҸҲҸҝӃѽ ҡґҏҍҽҾѽ ҽӁҽґѹҲѹҍѦҡҔѹѦ ӍҸҝѸӃ ҔҸѸҰҽҝҽҲѹӊҽ ѸҸҊҏҲ қӃҲҧ ӁѦѸҽҝҔѹҝҸӊҽӁӁӃѸѹ. ѷѦҰѦҝҧ ҸқҝҽҲѹѸҡҾ ҔҸ ӊҲҸҝҸҮ ҍҽҡҲѹ ҊѹҰҸҲѦӉӃ ҘҝҸӍҲҽ ѹ ҘҝҏӉҽ, ҊҦѦ ҏҲӊѦҝңҦҽѦҲҡҾ, ҍҲҸ Ѧҡґѹ ҰҝѹґҽҊҽҲѦґҧӁҸѦ-ҽӁҲҸӁѹѸ ӁѦ ѹѸѦѦҲ ҡѹӁҲѦҲѹҍѦҡҔҸҮ ӍҸҝѸӃ ҡҝҽӊӁѹҲѦґҧӁҸҮ ҡҲѦҰѦӁѹ, ҲҸ ҽӁҽґѹҲѹҍѦҡҔҽҾ ӍҸҝѸҽ ҔҸѸҰҽҝҽҲѹӊҽ ҙҲҸҊҸ ҰҝѹґҽҊҽҲѦґҧӁҸҊҸ ӊҡѦҊҦҽ қҏҦѦҲ ӁѦѸҽҝҔѹҝҸӊҽӁӁҸҮ. ҤҸҝҸҗҏӀ ҰҸҍӊҏ ҦґҾ ҰҝҸӊѦҝҔѹ ҦҽӁӁҸҊҸ ҏҲӊѦҝңҦѦӁѹҾ ҰҝѦҦҡҲҽӊґҾӀҲ ҝҏҡҡҔѹѦ ҰҝѹґҽҊҽҲѦґҧӁӃѦ ҡ ҸҲҝѹ҃ҽҲѦґҧӁҸҮ ҰҝѹҡҲҽӊҔҸҮ (ӁҽҰҝѹѸѦҝ, ӰӾԼӹԊӸԶӹӡ, ӰӾӴՂԙӹӪӠӡ, ӰӾԼӹԄԂԽՂӡ, ӰӾԌӹӢӹԄӹӡ), ҰҸҡҔҸґҧҔҏ ҙҲѹ ҰҝѹґҽҊҽҲѦґҧӁӃѦ ҡҔҽґҾҝӁҸ-ҽӁҲҸӁѹѸѹҍѦҡҔҸҊҸ ҲѹҰҽ ӁѦ ѹѸѦӀҲ ҡѹӁҲѦҲѹҍѦҡҔѹѽ ӍҸҝѸ ҡҝҽӊӁѹҲѦґҧӁҸҮ ҡҲѦҰѦӁѹ ѹ ҏҰҸҲҝѦқґҾӀҲҡҾ ҲҸґҧҔҸ ӊ ҽӁҽґѹҲѹҍѦҡҔѹѽ ҡҝҽӊӁѹҲѦґҧӁӃѽ ҔҸӁҡҲҝҏҔ҃ѹҾѽ.4 үӁҽґѹӉ ӁҽҮҦѦӁӁӃѽ ҰҝѹѸѦҝҸӊ ҰҸҔҽӉӃӊҽѦҲ, ҍҲҸ ҡҝҽӊӁѹҲѦґҧӁӃѦ ӍҸҝѸӃ ҦҽӁӁӃѽ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ӊҡѦҊҦҽ ѸҽҝҔѹҝҸӊҽӁӃ. (ҭҸ ҔҝҽҮӁѦҮ ѸѦҝѦ, ӊ ѽҸҦѦ ҦҽӁӁҸҊҸ ѹҡҡґѦҦҸӊҽӁѹҾ ӁѦ қӃґҸ ҸқӁҽҝҏңѦӁҸ Ӂѹ ҸҦӁҸҊҸ ӁѦѸҽҝҔѹҝҸӊҽӁӁҸҊҸ ҏҰҸҲҝѦқґѦӁѹҾ.) ҭҝѹӊѦҦѦѸ ӁѦҡҔҸґҧҔҸ ҰҝѹѸѦҝҸӊ: 3

ѨҡҔґӀҍѦӁѹѦ ҡҸҡҲҽӊґҾӀҲ ҡґҏҍҽѹ, ҔҸҊҦҽ ҰҝѹґҽҊҽҲѦґҧӁҸѸҏ ӊ ӍҸҝѸѦ ҡҝҽӊӁѹҲѦґҧӁҸҮ ҡҲѦҰѦӁѹ ҰҝѦҦҗѦҡҲӊҏѦҲ ҡґҸӊҽ ӾԷӾ ѹ ԌԂԞӾ (even, still ѹ yet ӊ ҽӁҊґѹҮҡҔҸѸ ҾӉӃҔѦ), ҽ ҲҽҔңѦ ҰҸґҸңѹҲѦґҧӁҽҾ ӍҸҝѸҽ ҲҸҊҸ ңѦ ҰҝѹґҽҊҽҲѦґҧӁҸҊҸ ѹґѹ ҰҝѹґҽҊҽҲѦґҧӁҸҊҸ ҡ ҲҸҮ ңѦ ҰҸґҾҝӁҸҡҲҧӀ (ҡѸ. Ҹқ ҙҲҸѸ ҰҸҦҝҸқӁѦѦ Tribushinina 2007). 4 ҷҸґҧҗѹӁҡҲӊҸ ҰҝѹґҽҊҽҲѦґҧӁӃѽ, ѹѸѦӀӂѹѽ ҲҸґҧҔҸ ҽӁҽґѹҲѹҍѦҡҔҏӀ ӍҸҝѸҏ ҡҝҽӊӁѹҲѦґҧӁҸҮ ҡҲѦҰѦӁѹ, ӁѦ ҸҲӁҸҡҾҲҡҾ Ҕ ҝҽӉҝҾҦҏ ҡҔҽґҾҝӁӃѽ ҽӁҲҸӁѹѸҸӊ. ѢҽҰҝѹѸѦҝ, ҲҽҔѹѦ ҰҝѹґҽҊҽҲѦґҧӁӃѦ ҔҽҔ ӴӢԂԞԌӾԼӰՂӡ, ԌӾӢԒӪӠӡ ѹ ӫӴԊӾԓӾӰӰՂӡ, ҸқҝҽӉҏӀӂѹѦ ӍҸҝѸҏ ҡҝҽӊӁѹҲѦґҧӁҸҮ ҡҲѦҰѦӁѹ ҲҸґҧҔҸ ҽӁҽґѹҲѹҍѦҡҔѹ, ҸҲӁҸҡҾҲҡҾ Ҕ ҝҽӉҝҾҦҏ Ҹ҃ѦӁҸҍӁӃѽ ҰҝѹґҽҊҽҲѦґҧӁӃѽ. ҫҡҸқѦӁӁҸҡҲҧӀ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ҙҲҸҊҸ ҲѹҰҽ ҾӊґҾѦҲҡҾ ҲҸ, ҍҲҸ ҔҽңҦӃҮ ѹӉ ҍґѦӁҸӊ ҽӁҲҸӁѹѸѹҍѦҡҔҸҮ ҰҽҝӃ ѹѸѦѦҲ ҡӊҸӀ ҡҸқҡҲӊѦӁӁҏӀ җҔҽґҏ (Bierwisch 1989, Croft & Cruse 2004). ҘҝҸѸѦ ҲҸҊҸ, ҰҝѹґҽҊҽҲѦґҧӁӃѦ ҙҲҸҊҸ ҲѹҰҽ ӁѦ ҸҝѹѦӁҲѹҝҸӊҽӁӃ Ӂҽ ӁҸҝѸҏ. ѣҡҲѦҡҲӊѦӁӁӃѸ ҡґѦҦҡҲӊѹѦѸ ҙҲѹѽ ҸҡҸқѦӁӁҸҡҲѦҮ ҾӊґҾѦҲҡҾ ҲҸ, ҍҲҸ ӊҡѦ ҏҰҸҲҝѦқґѦӁѹҾ ҙҲѹѽ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ӊ ӍҸҝѸѦ ҡҝҽӊӁѹҲѦґҧӁҸҮ ҡҲѦҰѦӁѹ ѸҽҝҔѹҝҸӊҽӁӃ.

476

ѣһѣѢү ѷҩѨҷѰҹѨѢѨѢү

(33) Ѱ ҔҸѸҰҽӁѹѹ «ѰӉқѦҔҲѦґѦҔҸѸ» ѹѸѦӀҲҡҾ ҰґҽӁӃ ҦґҾ җѹҝҸҔҸҊҸ ӊӁѦҦҝѦӁѹҾ ҏҡґҏҊ IP-ҲѦґѦӍҸӁѹѹ ҔҽҔ қҸґѦѦ ӁѦҦҸҝҸҊҸҮ ҽґҧҲѦҝӁҽҲѹӊӃ ҏҡґҏҊҽѸ ҲҝҽҦѹ҃ѹҸӁӁҸҮ ҲѦґѦӍҸӁӁҸҮ ҡӊҾӉѹ. (http://ru.infocom.uz/more.php) (34) Ѩ ҍѦѸ ҭҏҲѹӁҡҔѹҮ үқҝҽѸҸӊѹҍ қҸґѦѦ ӁѦѽҸҝҸҗѹҮ ӊҸҝ, ҍѦѸ ӊҸҝ ҤҸҦҸҝҔҸӊҡҔѹҮ? (http://grani.ru/Politics/Russia/yukos/d.109549.html) (35) ҴҸҊҏ ҡҔҽӉҽҲҧ ҲҸґҧҔҸ ҸҦӁҸ - ҙҲҸҲ ҡҸӁ ҸқӃҍӁҸ Ӂҽ 90% ҏҰҝҽӊґҾѦѸӃҮ, ҲҸ ѦҡҲҧ Ҿ ҸҡҸӉӁҽӀ, ҍҲҸ ҙҲҸ ҡҸӁ. ҫӂҏӂѦӁѹѦ ү`ґҾ ҔҸѸҰҧӀҲѦҝӁҽҾ ѹҊҝҽ. ғӉґѦҲҽӀ ҲҽҔ ңѦ ҔҽҔ ҲӃ ҝҽҡҡҔҽӉҽґҽ - ґѦҊҔѹҮ ҲҸґҍѦҔ ӁҸҊҽѸѹ ѹ ӁҽҰҝҾңѦӁѹѦ. ү ӊҸҲ ӊӃҡҸҲҽ ӊҡѦҊҦҽ ҰҸҍѦѸҏ-ҲҸ ӉҽӊѹҡѹҲ ҸҲ ҊґҏқѹӁӃ ҡӁҽ, Ҳ.Ѧ. ҍѦѸ қҸґѦѦ ӁѦҊґҏқҸҔѹҮ ҡҸӁ, ҲѦѸ ҡҸҸҲӊѦҲҡҲӊѦӁӁҸ ӁѹңѦ Ҿ ґѦҍҏ. (http://flasher.ru/forum/archive/index.php/t-21476.html) ҭҝѹґҽҊҽҲѦґҧӁӃѦ, ҏҰҸҲҝѦқґѦӁӁӃѦ ӊ ҰҝѹѸѦҝҽѽ (33)-(35) ӊ ӍҸҝѸѦ ҡҝҽӊӁѹҲѦґҧӁҸҮ ҡҲѦҰѦӁѹ, ҰҝѹѸѦӁѹѸӃ Ҕ ҡҸҸҲӊѦҲҡҲӊҏӀӂѹѸ ҝѦӍѦҝѦӁҲҽѸ ѹ ӊ ҰҸґҸңѹҲѦґҧӁҸҮ ӍҸҝѸѦ (Ҳ.Ѧ. ӰӾԌӹӢӹԄԂԋ ԽӾԊӾԅӹӰӠԋ, ӰӾԖӹӢӹԶӠӡ ӴӹӢ, ӰӾԄԊӫԼӹӪӠӡ ԙӹӰ), ҍҲҸ ҊҸӊҸҝѹҲ Ҹ ѸҽҝҔѹҝҸӊҽӁӁҸҡҲѹ ҦҽӁӁӃѽ ѦҦѹӁѹ҃. ҭҸ ҙҲҸҮ ҰҝѹҍѹӁѦ, ӍҸҝѸҽѸ ҔҸѸҰҽҝҽҲѹӊҽ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ҦҽӁӁҸҊҸ ҲѹҰҽ ҍҽҡҲҸ ҰҝѦҦҗѦҡҲӊҏӀҲ ҰҸҔҽӉҽҲѦґѹ ѸҽҝҔѹҝҸӊҽӁӁҸҡҲѹ, ҲҽҔѹѦ ҔҽҔ ԌԂԞӾ ѹ ӾԷӾ: (36) ҭҝѹӊҡҲҽӊҽҾ Ӂҽ ҃ӃҰҸҍҔѹ, ӁҽҦ ӁѦҮ ӁѦҡ ӉҸӁҲ ӀӁҸҗҽ - ѦӂѦ қҸґѦѦ ӁѦӊӃҡҸҔѹҮ, ҰҸґӁӃҮ ѹ ҝӃңѹҮ, ҍѦѸ Ҿ ҡҽѸ. (ҋ. һҏҔҧҾӁѦӁҔҸ) (37) ҘҽҔ ҲҸґҧҔҸ ӊҸҗґҽ ӊ ҲѦҽҲҝ ѸӁѦ ҡҝҽӉҏ ҡҔҽӉҽґѹ, ҍҲҸ Ҿ ӁѦ ѽҸҝҸҗѹҮ ҍѦґҸӊѦҔ, ҰҸҲҸѸҏ ҍҲҸ ӁѦ ѸҸҊҏ ҡҦҽҲҧ ҙҔӉҽѸѦӁ Ӂҽ Ұҝҽӊҽ ѹ ѦӂѦ қҸґѦѦ ӁѦѽҸҝҸҗѹҮ, ҰҸҲҸѸҏ ҍҲҸ ѽҸҍҏ Ұҝҽӊҽ ҔҏҰѹҲҧ, ӊҸҲ ҡѹңҏ ѹ ѸѹҝӀҡҧ ҡ ҲѦѸ, ҍҲҸ Ҿ ӊҦӊҸҮӁѦ ӁѦѽҸҝҸҗѹҮ ҍѦґҸӊѦҔ. (http://www.liveinternet.ru/users/violini/) (38) ҶҔҸӁҸѸҔґҽҡҡ ҝҽҡҡҍѹҲҽӁ Ӂҽ ґӀҦѦҮ ҡҸ ҡҝѦҦӁѹѸ ҦҸҡҲҽҲҔҸѸ ѹґѹ ҦҽңѦ қҸґѦѦ ӁѦқҸҊҽҲӃѽ. (www.izdom.ru/Rech/archive.html) ѷҽҔѹѸ ҸқҝҽӉҸѸ, ҦҽӁӁӃѦ ѹӉ ҝҏҡҡҔҸҊҸ ҾӉӃҔҽ ҰҝҸҲѹӊҸҝѦҍҽҲ ҊѹҰҸҲѦӉѦ, ҡҸҊґҽҡӁҸ ҔҸҲҸҝҸҮ ҔҸѸҰҽҝҽҲѹӊ ҽҦӆѦҔҲѹӊӁӃѽ ҽӁҲҸӁѹѸҸӊ, ѹѸѦӀӂѹѽ ҲҸґҧҔҸ ҽӁҽґѹҲѹҍѦҡҔҏӀ ӍҸҝѸҏ ҡҝҽӊӁѹҲѦґҧӁҸҮ ҡҲѦҰѦӁѹ, ӊҡѦҊҦҽ ӁѦѸҽҝҔѹҝҸӊҽӁ. ҘҝҸѸѦ ҲҸҊҸ, ҽӁҽґѹӉ ҝҏҡҡҔѹѽ ҰҝѹґҽҊҽҲѦґҧӁӃѽ-ҽӁҲҸӁѹѸҸӊ ҰҸҔҽӉҽґ, ҍҲҸ ӍҸҝѸӃ ҔҸѸҰҽҝҽҲѹӊҽ ҡҸ ӉӁҽҍѦӁѹѦѸ қҸґѦѦ ӊӃҡҸҔҸҮ ҡҲѦҰѦӁѹ ҔҽҍѦҡҲӊҽ ӁѦ ӊҡѦҊҦҽ ӁѦѸҽҝҔѹҝҸӊҽӁӃ. ѩҽӁӁӃҮ ҝѦӉҏґҧҲҽҲ ҰҝѦҦҡҲҽӊґҾѦҲ ӊѦҡҸѸӃҮ ҔҸӁҲҝҽҝҊҏѸѦӁҲ ҡҏӂѦҡҲӊҏӀӂѹѸ ҲѦҸҝѹҾѸ ҽҦӆѦҔҲѹӊӁҸҮ ҡѦѸҽӁҲѹҔѹ, ҸқӆҾҡӁҾӀӂѹѸ ҡѦѸҽӁҲѹҍѦҡҔҸѦ ҝҽӉґѹҍѹѦ ѸѦңҦҏ ӍҸҝѸҽѸѹ ҰҸґҸңѹҲѦґҧӁҸҮ ѹ ҡҝҽӊӁѹҲѦґҧӁҸҮ ҡҲѦҰѦ-

ѢѣҋҘҫһѻҘҫ үҘҋѨҫҴ ҫ ҭҩѨһүҐүѷѣһѻѢѳҤ-үѢѷҫѢѨҴүҤ

477

Ӂѹ Ӂҽ ҸҡӁҸӊҽӁѹѹ ѸҽҝҔѹҝҸӊҽӁӁҸҡҲѹ ҰѦҝӊҸҮ ѹ ӁѦѸҽҝҔѹҝҸӊҽӁӁҸҡҲѹ ҰҸҡґѦҦӁѦҮ.5 5. ҋҲҝҏҔҲҏҝҽ ҊҝҽҦҽ҃ѹҸӁӁҸҮ җҔҽґӃ: ӊҸҰҝҸҡӃ ҡҸҍѦҲҽѦѸҸҡҲѹ ѷҝѦҲѹҮ ҰҸҡҲҏґҽҲ, Ҹ ҔҸҲҸҝҸѸ ҰҸҮҦѦҲ ҝѦҍҧ ӊ ҦҽӁӁҸҮ ҝҽқҸҲѦ, ҾӊґҾѦҲҡҾ ҸҲӁҸҡѹҲѦґҧӁҸ ӁѦҦҽӊӁѹѸ ҝҽӉӊѹҲѹѦѸ ӊ ҲѦҸҝѹѹ ҽҦӆѦҔҲѹӊӁҸҮ ҡѦѸҽӁҲѹҔѹ ѹ ҡӊҾӉӃӊҽѦҲҡҾ, ӊ ҰѦҝӊҏӀ ҸҍѦҝѦҦҧ, ҡ ҝҽқҸҲҽѸѹ ҘҽҝѹҲӃ ҭҽҝҽҦѹӉ (Paradis 1997, 2000, 2001, 2005), ҘҝѹҡҲҸӍѦҝҽ ҘѦӁӁѦҦѹ ѹ һҏѹҡҽ ҴҔѢѦґґѹ (Kennedy 2005, Kennedy & McNally 2005), ҽ ҲҽҔңѦ ҰҸҡґѦҦҸӊҽҲѦґѦҮ ҘѦӁӁѦҦѹ ѹ ҴҔѢѦґґѹ (Amaral 2006, Frazier et al. 2006, Rotstein & Winter 2004, Syrett et al. 2004). ғӃҗѦҰѦҝѦҍѹҡґѦӁӁӃѦ ѹҡҡґѦҦҸӊҽҲѦґѹ ҰҝҸӊҸҦҾҲ ҝҽӉґѹҍѹѦ ѸѦңҦҏ ҰҝѦҦѦґҧӁӃѸѹ ҰҝѹґҽҊҽҲѦґҧӁӃѸѹ, ҸҝѹѦӁҲѹҝҸӊҽӁӁӃѸѹ Ӂҽ ӉҽҔҝӃҲӃѦ җҔҽґӃ (closedscale/bounded adjectives), ѹ ӁѦҰҝѦҦѦґҧӁӃѸѹ ҰҝѹґҽҊҽҲѦґҧӁӃѸѹ, ҸҝѹѦӁҲѹҝҸӊҽӁӁӃѸѹ Ӂҽ ҸҲҔҝӃҲӃѦ җҔҽґӃ (open-scale/unbounded adjectives). Ҙ ҡґҸӊҽѸ ҰѦҝӊҸҊҸ ҲѹҰҽ ҸҲӁҸҡҾҲҡҾ ҰҝѹґҽҊҽҲѦґҧӁӃѦ ҡ ҔҸӁҲҝҽҦѹҔҲҸҝӁӃѸ ҲѹҰҸѸ ҡѦѸҽӁҲѹҍѦҡҔѹѽ ҸҲӁҸҗѦӁѹҮ, ҸҝѹѦӁҲѹҝҸӊҽӁӁӃѦ Ӂҽ ѸҽҔҡѹѸҏѸ ӊӃҝҽңҽѦѸҸҊҸ ҰҝѹӉӁҽҔҽ (ҲҽҔѹѦ ҔҽҔ ԙӫԖӹӡ, ԓӠԙԽՂӡ, ԐӾӢԽӴՂӡ, ԗӫԙԽӹӡ). Ҙ ҡґҸӊҽѸ ӊҲҸҝҸҊҸ ҲѹҰҽ ҸҲӁҸҡҾҲҡҾ ѹҡҡґѦҦҏѦѸӃѦ ӊ ҦҽӁӁҸҮ ҝҽқҸҲѦ ҰҝѹґҽҊҽҲѦґҧӁӃѦ-ҽӁҲҸӁѹѸӃ, ҸҝѹѦӁҲѹҝҸӊҽӁӁӃѦ Ӂҽ ӁҸҝѸҏ ӊ ҃ѦӁҲҝѦ ҊҝҽҦҽ҃ѹҸӁӁҸҮ җҔҽґӃ. ѩѦґѦӁѹѦ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ҰҸ ҰҝѹӁ҃ѹҰҏ (ӁѦ)ӉҽѸҔӁҏҲҸҡҲѹ җҔҽґ ҸҡӁҸӊҽӁҸ Ӂҽ Ҧӊҏѽ ҔҝѹҲѦҝѹҾѽ. ғҸ-ҰѦҝӊӃѽ, ӊ ҸӁҲҸґҸҊѹҍѦҡҔҸѸ ҡѸӃҡґѦ, ҰҝѹґҽҊҽҲѦґҧӁӃѦ ҰѦҝӊҸҊҸ ҲѹҰҽ ѸҸҊҏҲ ҸқҸӉӁҽҍҽҲҧ ҽқҡҸґӀҲӁӃҮ ѸҽҔҡѹѸҏѸ ҰҝѹӉӁҽҔҽ, ӊ ҲҸ ӊҝѦѸҾ ҔҽҔ ҡҲѦҰѦӁҧ ҰҝѹӉӁҽҔҽ, ӊӃҝҽңҽѦѸҸҊҸ ҰҝѹґҽҊҽҲѦґҧӁӃѸѹ ӊҲҸҝҸҊҸ ҲѹҰҽ, ѸҸңѦҲ ҏӊѦґѹҍѹӊҽҲҧҡҾ ҦҸ қѦҡҔҸӁѦҍӁҸҡҲѹ. ѢҽҰҝѹѸѦҝ, ҡҲҸґ ѸҸңѦҲ қӃҲҧ ҰҝѦҦѦґҧӁҸ ҍѹҡҲӃѸ, ӁҸ ӁѦ ѸҸңѦҲ қӃҲҧ ҰҝѦҦѦґҧӁҸ ҦґѹӁӁӃѸ (Ҳ.Ҕ. ѦҊҸ ҦґѹӁҏ ӊҡѦҊҦҽ ѸҸңӁҸ ҏӊѦґѹҍѹҲҧ). ҭҸҦҸқӁӃѸ ңѦ ҸқҝҽӉҸѸ, қҽӁҔҽ ѸҸңѦҲ қӃҲҧ ѸҽҔҡѹѸҽґҧӁҸ ҰҏҡҲҸҮ, ӁҸ ҸӁҽ ӁѦ ѸҸңѦҲ қӃҲҧ ѸҽҔҡѹѸҽґҧӁҸ қҸґҧҗҸҮ, Ҳ.Ҕ. ӊҡѦҊҦҽ ѸҸңӁҸ ӁҽҮҲѹ ѦӂѦ қҸґҧҗҏӀ қҽӁҔҏ. ғҸ-ӊҲҸҝӃѽ, ҰҝѹґҽҊҽҲѦґҧӁӃѦ, ҸҝѹѦӁҲѹҝҸӊҽӁӁӃѦ Ӂҽ ӉҽҔҝӃҲӃѦ җҔҽґӃ, ѸҸҊҏҲ ҡҸҍѦҲҽҲҧҡҾ ҡ ҲҽҔ ӁҽӉӃӊҽѦѸӃѸѹ ԐԂӪԙӠԐԂӡԒӾӢԂԐӠ (Quirk et al. 1985), Ҳ.Ѧ. ӁҽҝѦҍѹҾѸѹ, ӊӃҝҽңҽӀӂѹѸѹ ҰҸґӁҏӀ ҡҲѦҰѦӁҧ ҔҽҍѦҡҲӊҽ (completely, absolutely), ҽ ҲҽҔңѦ ӁҽҝѦҍѹҾѸѹ, ӊӃҝҽңҽӀӂѹѸѹ ҰҝѹқґѹңѦӁѹѦ Ҕ ѸҽҔҡѹѸҽґҧӁҸѸҏ ҰҝѦҦѦґҏ (almost). ғ ҸҲґѹҍѹѦ ҸҲ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ҔҸӁҲҝҽҦѹҔҲҸҝӁҸҊҸ ҲѹҰҽ 6 , 5 ѢѦҡѸҸҲҝҾ Ӂҽ ҲҸ, ҍҲҸ ӊ ҝҽѸҔҽѽ ҦҽӁӁҸҮ ҡҲҽҲҧѹ қӃґѹ ҝҽҡҡѸҸҲҝѦӁӃ ҲҸґҧҔҸ ӍҸҝѸӃ ҡҝҽӊӁѹҲѦґҧӁҸҮ ҡҲѦҰѦӁѹ, ҰҝѹӊѦҦѦӁӁӃҮ ҽӁҽґѹӉ ѸҸңѦҲ қӃҲҧ ҲҽҔңѦ ҙҔҡҲҝҽҰҸґѹҝҸӊҽӁ Ӂҽ ӍҸҝѸӃ ҡҏҰѦҝґҽҲѹӊҽ. 6 ҋґѦҦҏѦҲ ҸҲѸѦҲѹҲҧ, ҍҲҸ ӁѦ ӊҡѦ ҔҸӁҲҝҽҦѹҔҲҸҝӁӃѦ ҔҸҝҝѦґҾҲӃ ҸҝѹѦӁҲѹҝҸӊҽӁӃ Ӂҽ ѸҽҔҡѹѸҏѸ ҰҝѹӉӁҽҔҽ; ӁѦҔҸҲҸҝӃѦ ѹӉ Ӂѹѽ ҸҝѹѦӁҲѹҝҸӊҽӁӃ Ӂҽ ѸѹӁѹѸҏѸ. ҭҝѹґҽҊҽҲѦґҧӁӃѦ ҲҽҔҸҊҸ ҲѹҰҽ ѸҸҊҏҲ ҸҰҝѦҦѦґҾҲҧҡҾ ѸѹӁѹѸҽҮӉѦҝҽѸѹ (ӰӾԐӰӹԄӹ ӴԊԂԞӰՂӡ, ԓӫԽӸ ԄӢԋԒӰՂӡ), ӁҸ ӁѦ ѸҸҊҏҲ ҡҸҍѦҲҽҲҧҡҾ ҡ ѸҽҔҡѹѸҽҮӉѦҝҽѸѹ (#ԙӹӴԙӾԐ ӴԊԂԞӰՂӡ, #ԂԼԙӹԊԇԽӰӹ ԄӢԋԒӰՂӡ) ѹ ӁҽҝѦҍѹҾѸѹ ҡҸ ӉӁҽҍѦ-

478

ѣһѣѢү ѷҩѨҷѰҹѨѢѨѢү

ҰҝѹґҽҊҽҲѦґҧӁӃѦ-ҽӁҲҸӁѹѸӃ ӁѦ ҏҰҸҲҝѦқґҾӀҲҡҾ ӊ ҡҸҍѦҲҽӁѹѹ ҡ ӊӃҗѦӁҽӉӊҽӁӁӃѸѹ ӁҽҝѦҍѹҾѸѹ; ҙҲѹ ҰҝѹґҽҊҽҲѦґҧӁӃѦ ѸҸҊҏҲ ҸҰҝѦҦѦґҾҲҧҡҾ ҲҸґҧҔҸ ӁѦҰҝѦҦѦґҧӁӃѸѹ ӁҽҝѦҍѹҾѸѹ ҡҲѦҰѦӁѹ, ҸқҸӉӁҽҍҽӀӂѹѸѹ ӊӃҡҸҔҏӀ (very, extremely), ҏѸѦҝѦӁӁҏӀ (rather, fairly) ѹ ѸҽґҏӀ (a bit, a little) ҡҲѦҰѦӁҧ ҔҽҍѦҡҲӊҽ. 7 ҋҸҍѦҲҽѦѸҸҡҲӁӃѦ ҸҡҸқѦӁӁҸҡҲѹ Ҧӊҏѽ ҲѹҰҸӊ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ҡ ӁҽҝѦҍѹҾѸѹ ѸѦҝӃ ѹ ҡҲѦҰѦӁѹ ҾӊґҾӀҲҡҾ ҝѦҗҽӀӂѹѸ ҔҝѹҲѦҝѹѦѸ ҝҽӉҦѦґѦӁѹҾ ҰҝѹґҽҊҽҲѦґҧӁӃѽ Ӂҽ ҦӊѦ ҊҝҏҰҰӃ, ҸҲґѹҍҽӀӂѹѦҡҾ ҦҝҏҊ ҸҲ ҦҝҏҊҽ ҡҲҝҏҔҲҏҝҸҮ җҔҽґ. ҋҍѹҲҽѦҲҡҾ, ҍҲҸ ҰҝѹґҽҊҽҲѦґҧӁӃѦ, ҡҸҍѦҲҽӀӂѹѦҡҾ ҡ ӁҽҝѦҍѹҾѸѹ ҰҸґӁҸҮ ҡҲѦҰѦӁѹ ѹ ӁҽҝѦҍѹҾѸѹ ҲѹҰҽ ԗӹԓԽӠ, ѹӁҲѦҝҰҝѦҲѹҝҏӀҲҡҾ ҸҲӁҸҡѹҲѦґҧӁҸ ӉҽѸҔӁҏҲӃѽ җҔҽґ, ӊ ҲҸ ӊҝѦѸҾ ҔҽҔ ҰҝѹґҽҊҽҲѦґҧӁӃѦ, ӁѦ ҡҰҸҡҸқӁӃѦ ҡҸҍѦҲҽҲҧҡҾ ҡ ҙҲѹѸѹ ӁҽҝѦҍѹҾѸѹ, ѹӁҲѦҝҰҝѦҲѹҝҏӀҲҡҾ ҸҲӁҸҡѹҲѦґҧӁҸ җҔҽґ, ӁѦ ѹѸѦӀӂѹѽ Ӂѹ ӊѦҝѽӁѦҊҸ, Ӂѹ ӁѹңӁѦҊҸ ҰҝѦҦѦґҽ. ҫҦӁҽҔҸ, ҔҽҔ ҏңѦ қӃґҸ ҡҔҽӉҽӁҸ ӊӃҗѦ, ҦҽӁӁҽҾ ҊѹҰҸҲѦӉҽ қӃґҽ ҡӍҸҝѸҏґѹҝҸӊҽӁҽ Ӂҽ ѸҽҲѦҝѹҽґѦ ҽӁҊґѹҮҡҔҸҊҸ ҾӉӃҔҽ ѹ қґѹӉҔҸҝҸҦҡҲӊѦӁӁӃѽ ѦѸҏ ҊѦҝѸҽӁҡҔѹѽ ҾӉӃҔҸӊ. ѢѦҡѸҸҲҝҾ Ӂҽ ҙҲҸ, ҽӊҲҸҝӃ, ҝҽӉҝҽқҸҲҽӊҗѹѦ ҦҽӁӁҏӀ ҡѦѸҽӁҲѹҍѦҡҔҏӀ ҔґҽҡҡѹӍѹҔҽ҃ѹӀ ҰҝѹґҽҊҽҲѦґҧӁӃѽ, ӁѦ ҸҊҝҽӁѹҍѹӊҽӀҲ ҡӊҸѹ ӊӃӊҸҦӃ ҡѦѸҽӁҲѹҔҸҮ ԂӰԄԊӠӡԙӪӠԖ (ӁѦѸѦ҃Ҕѹѽ/җӊѦҦҡҔѹѽ/ӁѹҦѦҝґҽӁҦҡҔѹѽ) ҊҝҽҦҏҽґҧӁӃѽ ҰҝѹґҽҊҽҲѦґҧӁӃѽ, ѹ ӍҸҝѸҏґѹҝҏӀҲ ҊѹҰҸҲѦӉҏ Ҹ ҡѦѸҽӁҲѹҔѦ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ӴӹӹԼԷӾ. 8 ѩҝҏҊѹѸѹ ҡґҸӊҽѸѹ, ҸӁѹ ҰҝѦҲѦӁҦҏӀҲ Ӂҽ ӁѦҔҸҲҸҝҏӀ ҏӁѹӊѦҝҡҽґҧӁҸҡҲҧ ҰҸґҏҍѦӁӁӃѽ ҝѦӉҏґҧҲҽҲҸӊ, ҏҲӊѦҝңҦҽҾ, ҍҲҸ ӊҡѦ ҝѦґҾҲѹӊӁӃѦ ҰҝѹґҽҊҽҲѦґҧӁӃѦ ӁѦҰҝѦҦѦґҧӁӃ ѹ ҰҸҙҲҸѸҏ ӁѦҰҝѹѦѸґѦѸӃ ҡ ӁҽҝѦҍѹҾѸѹ, ӊӃҝҽңҽӀӂѹѸѹ ҰҸґӁҏӀ ҡҲѦҰѦӁҧ ҔҽҍѦҡҲӊҽ (Syrett 2007). ғ ҙҲҸҮ ҍҽҡҲѹ ӁҽҡҲҸҾӂѦҮ ҝҽқҸҲӃ, ҔҽҔ ѹ ӊ ҰҝѦҦӃҦҏӂѹѽ Ҧӊҏѽ ҰҽҝҽҊҝҽӍҽѽ, қҏҦҏҲ ҰҝѦҦҡҲҽӊґѦӁӃ ӁѦҔҸҲҸҝӃѦ ҦҽӁӁӃѦ ѹӉ ҝҏҡҡҔҸҊҸ ҾӉӃҔҽ, ҔҸҲҸҝӃѦ ҾӊӁҸ ҰҝҸҲѹӊҸҝѦҍҽҲ ѹӉґҸңѦӁӁҸҮ ӊӃҗѦ ҊѹҰҸҲѦӉѦ. ҭҸҡҔҸґҧҔҏ ҊґҽӊӁӃѸ ҽҝҊҏѸѦӁҲҸѸ “ҊѹҰҸҲѦӉӃ ҰҝѦҦѦґҧӁҸҡҲѹ” ҾӊґҾѦҲҡҾ ҡҸҍѦҲҽѦѸҸҡҲҧ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ҡ ӁҽҝѦҍѹҾѸѹ ѸѦҝӃ ѹ ҡҲѦҰѦӁѹ, ӊ ҙҲҸѸ ҰҽҝҽҊҝҽӍѦ ѸӃ ҰҸҦҝҸқӁѦѦ ҸҡҲҽӁҸӊѹѸҡҾ Ӂҽ ҸҡҸқѦӁӁҸҡҲҾѽ ҡҸҍѦҲҽѦѸҸҡҲѹ ҝҏҡҡҔѹѽ ҰҝѹґҽҊҽҲѦґҧӁӃѽ-ҽӁҲҸӁѹѸҸӊ ҡ ӁҽҝѦҍѹҾѸѹ ҙҲҸҊҸ ҲѹҰҽ. ҩҏҡҡҔѹѦ ҔҽҍѦҡҲӊѦӁӁӃѦ ҰҝѹґҽҊҽҲѦґҧӁӃѦ ҡҔҽґҾҝӁҸ-ҽӁҲҸӁѹѸѹҍѦҡҔҸҊҸ ҔҸѸҰґѦҔҡҽ, ҔҽҔ ѹ ѹѽ “ҙҔӊѹӊҽґѦӁҲӃ” ӊ ҽӁҊґѹҮҡҔҸѸ ѹ ҦҝҏҊѹѽ ҊѦҝѸҽӁҡҔѹѽ ҾӉӃҔҽѽ, ҰґҸѽҸ ҡҸҍѦҲҽӀҲҡҾ ҡ ӁҽҝѦҍѹҾѸѹ, ӊӃҝҽңҽӀӂѹѸѹ ҰҝѹқґѹңѦӁѹѦ Ҕ ѸҽҔҡѹѸҽґҧӁҸѸҏ ҰҝѦҦѦґҏ (ԗӹԓԽӠ). ѷҽҔ, ӊ ѢҘҩѤ (ҸҔҸґҸ 130 ѸѹґґѹҸӁҸӊ ҡґҸӊ) ӁѦҲ ӁѹѦѸ ‘ҰҸҍҲѹ’ (#ԗӹԓԽӠ ԄӢԋԒӰՂӡ). ҋѸ. Ҹқ ҙҲҸѸ Rotstein & Winter (2004), Kennedy & McNally (2005) ѹ Kennedy (2005). 7 ҋҸҍѦҲҽѦѸҸҡҲҧ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ҡҔҽґҾҝӁҸ-ҽӁҲҸӁѹѸѹҍѦҡҔҸҊҸ ҔҸѸҰґѦҔҡҽ ҡ ѸѹӁѹѸҽҮӉѦҝҽѸѹ (ӰӾԐӰӹԄӹ, ԓӫԽӸ) ӊѦҡҧѸҽ ҸҊҝҽӁѹҍѦӁҽ. үҦӆѦҔҲѹӊӁӃѦ ҡґҸӊҸҡҸҍѦҲҽӁѹҾ ҲѹҰҽ ӰӾԐӰӹԄӹ ӴՂԙӹӪӠӡ ѸҽҝҔѹҝҸӊҽӁӃ ѹ ҸқҸӉӁҽҍҽӀҲ ӁѦ ѸҽґҏӀ ҡҲѦҰѦӁҧ ҸҲҔґҸӁѦӁѹҾ ҸҲ ӁҸҝѸӃ, ҽ ӁѦңѦґҽҲѦґҧӁӃҮ ѹӉқӃҲҸҔ ҔҽҍѦҡҲӊҽ (ҘҏҡҲҸӊҽ 1994: 22, Bolinger 1972: 50, Paradis 1997). 8 ҭҽҝҽҦѹӉ ӁҽӉӃӊҽѦҲ ҙҲҏ ҊѹҰҸҲѦӉҏ boundedness hypothesis (Paradis 2005, Paradis & Willners 2006).

ѢѣҋҘҫһѻҘҫ үҘҋѨҫҴ ҫ ҭҩѨһүҐүѷѣһѻѢѳҤ-үѢѷҫѢѨҴүҤ

479

Ӂѹ ҸҦӁҸҊҸ ҡҸҍѦҲҽӁѹҾ ԗӹԓԽӠ ҡ ҰҝѹґҽҊҽҲѦґҧӁӃѸѹ ӰӠԒӪӠӡ, ӪӹӢӹԽӪӠӡ, ԌӹӢӹԄӹӡ, ԌӾԶӾӴՂӡ, ԗԊӹԖӹӡ, ԄԊӫԼӹӪӠӡ, ԐӾԊӪӠӡ ѹ ԽӹӰӪӠӡ. ҫҦӁҽҔҸ ҙҲҸ ӁѦ ҸӉӁҽҍҽѦҲ, ҍҲҸ ҡґҸӊҸҡҸҍѦҲҽӁѹҾ ҲҽҔҸҊҸ ҲѹҰҽ ӁѦӊҸӉѸҸңӁӃ ӊ ҰҝѹӁ҃ѹҰѦ. ғ ҍҽҲҽѽ ѹ ҰҝҸҍѹѽ ѨӁҲѦҝӁѦҲ-ҝѦҡҏҝҡҽѽ ӊҡѦҊҦҽ ѸҸңӁҸ ӁҽҮҲѹ ҦҸҡҲҽҲҸҍӁҸѦ ҔҸґѹҍѦҡҲӊҸ ҲҽҔѹѽ ҏҰҸҲҝѦқґѦӁѹҮ, ҍҲҸ ҊҸӊҸҝѹҲ Ҹ ӁѦҔҸҲҸҝҸҮ ҡҲѦҰѦӁѹ ѹѽ ҰҝѹѦѸґѦѸҸҡҲѹ ӊ ҝҽӉҊҸӊҸҝӁҸѸ ҾӉӃҔѦ. ѩҽңѦ ӊ ҔҸҝҰҏҡѦ ѦҡҲҧ ҰҝѹѸѦҝӃ ҏҰҸҲҝѦқґѦӁѹҾ ԗӹԓԽӠ ҡ ҝҾҦҸѸ ҡҔҽґҾҝӁӃѽ ҽӁҲҸӁѹѸҸӊ. ѢҽҰҝѹѸѦҝ, ӊ ѢҘҩѤ ҰҝѹґҽҊҽҲѦґҧӁҸѦ ԖӹӢӹԶӠӡ ҏҰҸҲҝѦқґѦӁҸ ӊ ҡҸҍѦҲҽӁѹѹ ҡ ԗӹԓԽӠ 15 ҝҽӉ, ԽӠԖӠӡ – 9, ԼӾԌӰՂӡ ѹ ԄӢӹԐӪӠӡ – 7, ԼӹԊӸԶӹӡ – 5, ӴՂԙӹӪӠӡ ѹ ԽӹԊԙԽՂӡ – 3, ԼӹԄԂԽՂӡ – 2, ԐԂԊӾӰӸӪӠӡ ѹ ԌԊӠӰӰՂӡ – 1 ҝҽӉ. ҋґѦҦҏѦҲ ҲҽҔңѦ ҸҲѸѦҲѹҲҧ, ҍҲҸ ӊ ҡҸҍѦҲҽӁѹҾѽ ҡ ԗӹԓԽӠ ӉӁҽҍѦӁѹѦ ҰҝѹґҽҊҽҲѦґҧӁӃѽ-ҽӁҲҸӁѹѸҸӊ ҸҝѹѦӁҲѹҝҸӊҽӁҸ ӁѦ Ӂҽ ѸҽҔҡѹѸҏѸ, ҽ Ӂҽ ѸѹӁѹѸҏѸ ӊӃҝҽңҽѦѸҸҊҸ ҰҝѹӉӁҽҔҽ. ҋҝҽӊӁѹѸ ҰҝѹѸѦҝӃ (39) ѹ (40): (39) ғ ҝѦӉҏґҧҲҽҲѦ ҡґҸңӁӃѽ ӍѹӉѹҍѦҡҔѹѽ ҰҝҸ҃ѦҡҡҸӊ ҸқҝҽӉҏѦҲҡҾ ӁѦҸҦӁҸҝҸҦӁҸѦ ҰҸ ҡҸҡҲҽӊҏ ҸқґҽҍҔҸ ҡѸѦҡѹ – ӊқґѹӉѹ ҡӊѦҍѹ ҡѸѦҡҧ қҸҊҽҲҽҾ, Ӂҽ “ҰѦҝѹӍѦҝѹѹ” – қѦҦӁѦѦ, ҽ ҏ ҡҲѦӁҸҔ ҃ѹґѹӁҦҝҽ ҸҡҲҽҺҲҡҾ ҰҸҍҲѹ ҍѹҡҲӃҮ ӊҸӉҦҏѽ. (ѢҘҩѤ) (40) һӀқҸҍҔҽ ҲҸңѦ ҏңѦ ӁҽҍѹӁҽґҽ ӁҸҡѹҲҧ ҰҸҍҲѹ ҦґѹӁӁҸѦ ҰґҽҲҧѦ, ҲҽҔ ҍҲҸ ѦѦ ҊҏҡѹӁӃѦ ӁҸҊѹ қӃґѹ ҰҸҍҲѹ ӁѦ ӊѹҦӁӃ, ӁҸ ҸӁҽ қӃґҽ ҲҽҔҽҾ ңѦ ҰґҽҔҡҽ, ҔҽҔ ѹ ҰҝѦңҦѦ. (ѢҘҩѤ) өӠԙԽՂӡ – ҰҝҸҲҸҲѹҰѹҍѦҡҔѹҮ ҰҝѹѸѦҝ ҰҝѹґҽҊҽҲѦґҧӁҸҊҸ, ҸҝѹѦӁҲѹҝҸӊҽӁӁҸҊҸ Ӂҽ ѸҽҔҡѹѸҏѸ ҊҝҽҦҽ҃ѹҸӁӁҸҮ җҔҽґӃ. ҋґҸӊҸҡҸҍѦҲҽӁѹѦ ԗӹԓԽӠ ԓӠԙԽՂӡ ӊ (39) ҸӉӁҽҍҽѦҲ ҰҝѹқґѹңѦӁѹѦ Ҕ ѸҽҔҡѹѸҽґҧӁҸҮ ҡҲѦҰѦӁѹ ҍѹҡҲҸҲӃ ӊ ҲҸѸ ҡѸӃҡґѦ, ҍҲҸ ҰҸҍҲѹ ҍѹҡҲӃҮ ӊҸӉҦҏѽ ҡҸҦѦҝңѹҲ ѸѹӁѹѸҽґҧӁҸѦ ҔҸґѹҍѦҡҲӊҸ ҰҝѹѸѦҡѦҮ. ғ ҰҝѹѸѦҝѦ (40) ӁҽҝѦҍѹѦ ԗӹԓԽӠ ҸҰҝѦҦѦґҾѦҲ ӁѦҰҝѦҦѦґҧӁҸѦ ҰҝѹґҽҊҽҲѦґҧӁҸѦ ԌԊӠӰӰՂӡ. ҈ҽѸѦҲѹѸ, ҍҲҸ ӊ ҦҽӁӁҸѸ ҡґҏҍҽѦ ҽҦӆѦҔҲѹӊӁҸѦ ҡґҸӊҸҡҸҍѦҲҽӁѹѦ ӁѦ ҸӉӁҽҍҽѦҲ ҰҝѹқґѹңѦӁѹѦ Ҕ ѸҽҔҡѹѸҏѸҏ ҦґѹӁӃ. ҋҲҝҸҊҸ ҊҸӊҸҝҾ, ҰҸҍҲѹ ҦґѹӁӁҸѦ ҰґҽҲҧѦ ӊҸҸқӂѦ ӁѦ ҾӊґҾѦҲҡҾ ҦґѹӁӁӃѸ. ғѦҝҸҾҲӁҸ, ҝѦҍҧ ѹҦѦҲ Ҹ ҰґҽҲҧѦ ҡҝѦҦӁѦҮ ҦґѹӁӃ, ҔҸҲҸҝҸѦ ӁѦ ѸҸңѦҲ қӃҲҧ ӁҽӉӊҽӁҸ Ӂѹ ҔҸҝҸҲҔѹѸ, Ӂѹ ҦґѹӁӁӃѸ. ѷҽҔѹѸ ҸқҝҽӉҸѸ, ҰҝѹӉӁҽҔ, ӊӃҝҽңҽѦѸӃҮ ҦҽӁӁӃѸ ҡґҸӊҸҡҸҍѦҲҽӁѹѦѸ, ӁҽѽҸҦѹҲҡҾ ӊ ѸѦңҰҸґӀҡӁҸҮ ӉҸӁѦ ѹ ҰҝѹқґѹңҽѦҲҡҾ Ҕ ӉҸӁѦ ҰҝѹґҽҊҽҲѦґҧӁҸҊҸ қҸґҧҗҸҊҸ ҰҸґӀҡҽ. ҨҏӁҔ҃ѹҾ ӁҽҝѦҍѹҾ ӊ ҙҲҸѸ ҡґҏҍҽѦ ѸҸңѦҲ қӃҲҧ ӁҽӉӊҽӁҽ ѸѦҲҽґѹӁҊӊѹҡҲѹҍѦҡҔҸҮ: ԗӹԓԽӠ ҡѹҊӁҽґѹӉѹҝҏѦҲ ҰҝѹқґѹңѦӁѹѦ Ҕ ѸѹӁѹѸҽґҧӁҸҮ ҡҲѦҰѦӁѹ ҰҝѹӉӁҽҔҽ, ҔҸҲҸҝҽҾ ҰҸӉӊҸґѹґҽ қӃ ҊҸӊҸҝҾӂѦѸҏ ӁҽӉӊҽҲҧ ҰґҽҲҧѦ ҦґѹӁӁӃѸ. ѢҽѹқҸґѦѦ ѹӁҲѦҝѦҡӁҸҮ ҸҡҸқѦӁӁҸҡҲҧӀ ҝҏҡҡҔҸҊҸ ҾӉӃҔҽ ҾӊґҾѦҲҡҾ ҏҰҸҲҝѦқґѦӁѹѦ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ҡҔҽґҾҝӁҸ-ҽӁҲҸӁѹѸѹҍѦҡҔҸҊҸ ҲѹҰҽ ҡ ӁҽҝѦҍѹҾѸѹ, ӊӃҝҽңҽӀӂѹѸѹ ҰҸґӁҏӀ ҡҲѦҰѦӁҧ ҔҽҍѦҡҲӊҽ. ѢҽҰҸѸӁѹѸ, ҍҲҸ Ӂѹ ҸҦѹӁ ѹӉ ҍґѦӁҸӊ ҽӁҲҸӁѹѸѹҍѦҡҔҸҮ ҰҽҝӃ ӊ ҽӁҊґѹҮҡҔҸѸ ҾӉӃҔѦ ӁѦ ѸҸңѦҲ ҸҰҝѦҦѦґҾҲҧҡҾ ѸҽҔҡѹѸҽҮӉѦҝҽѸѹ (ӁҽҰҝѹѸѦҝ, #completely tall, #completely short, #absolutely wide,

480

ѣһѣѢү ѷҩѨҷѰҹѨѢѨѢү

#absolutely narrow 9 ). ғ ҝҏҡҡҔҸѸ ҾӉӃҔѦ, ӊ ҸҲґѹҍѹѦ ҸҲ ҽӁҊґѹҮҡҔҸҊҸ (ѹ ҦҝҏҊѹѽ ҊѦҝѸҽӁҡҔѹѽ ҾӉӃҔҸӊ), ҰҝѹґҽҊҽҲѦґҧӁӃѦ ѸҽґҸҊҸ ҰҸґӀҡҽ ҰҝѹѦѸґѦѸӃ ҡ ѸҽҔҡѹѸҽҮӉѦҝҸѸ ԙӹӴԙӾԐ.10 ҭҝѹӊѦҦѦѸ ӁѦҡҔҸґҧҔҸ ҰҝѹѸѦҝҸӊ ѹӉ ҔҸҝҰҏҡҽ: (41) ҋҽѸӃѦ ѸѹґӃѦ ӊҽӉҸҍҔѹ ҰҸґҏҍҽӀҲҡҾ ѹӉ ҰґҽҡҲѸҽҡҡҸӊӃѽ ҔҝӃҗѦҔ ҸҲ ҽҙҝҸӉҸґѦҮ. ҴѹґӃѦ, ҰҸҲҸѸҏ ҍҲҸ ҡҸӊҡѦѸ ѸҽґѦӁҧҔѹѦ. ғ Ӂѹѽ ҰҸѸѦҡҲҾҲҡҾ Ҧӊҽ-Ҳҝѹ ҃ӊѦҲҸҍҔҽ, ӁҸ Ӂҽ ҰҸґҔѦ ҲҽҔҽҾ ӊҽӉҸҍҔҽ ѸҸңѦҲ ҡҲҽҲҧ ҃ѦӁҲҝҸѸ ӊӁѹѸҽӁѹҾ. (ѢҘҩѤ) (42) ҘҸҊҦҽ ѸӃ ҏґҏҍҗҽѦѸ ҰҸґҸңѦӁѹѦ ҊҝҽңҦҽӁ / ѸӃ ӁѦ ѸҸңѦѸ ҊҸӊҸҝѹҲҧ Ҹ ӁҽҝҏҗѦӁѹҾѽ ҐҝҽңҦҽӁҡҔҸҊҸ ҘҸҦѦҔҡҽ. ѢҸ ӊ ҐҘ ӁҽҰѹҡҽӁҸ / ҍҲҸ ӊӃ ҔҽҔ ҡҸқҡҲӊѦӁӁѹҔ ҰҸґӁҸҡҲҧӀ ӁѦҡѦҲѦ қҝѦѸҾ ҡҸҦѦҝңҽӁѹҾ ҡӊҸѦҊҸ ѹѸҏӂѦҡҲӊҽ. ѷҽҔ ӊҸҲ / ѸӃ ҡѦҮҍҽҡ қѦҦӁҸѸҏ ҡҸқҡҲӊѦӁӁѹҔҏ ҰҸѸҸҊҽѦѸ / ѸӃ ѦѸҏ ӊӃҰґҽҍѹӊҽѦѸ ҔҸѸҰѦӁҡҽ҃ѹӀ / ҡҏқҡѹҦѹӀ ҰҸ ҸҰґҽҲѦ Ӊҽ ңѹґҧѦ ѹ ҔҸѸѸҏӁҽґҧӁӃѦ ҏҡґҏҊѹ / Ѧҡґѹ ѦҊҸ ҝҽҡѽҸҦӃ ҰҝѦӊӃҗҽӀҲ ҸҰҝѦҦѦґѦӁӁӃҮ ҏҡҲҽӁҸӊґѦӁӁӃҮ ҰҝҸ҃ѦӁҲ. ү ӊ қґѹңҽҮҗѦѦ ӊҝѦѸҾ / Ҿ ҦҏѸҽӀ / ҙҲҸҲ ҰҝҸ҃ѦӁҲ ѸӃ ѦӂѦ қҸґҧҗѦ ҡӁѹӉѹѸ ҦґҾ ҡҸӊҡѦѸ қѦҦӁӃѽ ҡѦѸѦҮ. (ѢҘҩѤ) (43) ҴӃ ҸқҝҽӂҽѦѸҡҾ Ҕ ҸӍѹ҃ѦҝҽѸ, ҰҝҽҰҸҝӂѹҔҽѸ, ҡѦҝңҽӁҲҽѸ, ҡҸґҦҽҲҽѸ – ҲѦѸ, ҔҲҸ ҰҝҸҗѦґ ҸҊӁѦӁӁӃѸѹ ҦҸҝҸҊҽѸѹ үӍҊҽӁѹҡҲҽӁҽ ѹ ѯѦҍӁѹ. ѢҽҰѹҗѹҲѦ ӁҽѸ ҰѹҡҧѸҸ. ҷҏҦѦҲ ґѹ ҸӁҸ ҦґѹӁӁӃѸ ѹґѹ ҡҸӊҡѦѸ ҔҸҝҸҲҔѹѸ, ӁѦ ӊҽңӁҸ. ғҡҰҸѸӁѹҲѦ ҙҰѹӉҸҦ ѹӉ ӊҽҗѦҮ қѹҸҊҝҽӍѹѹ, ҔҸҲҸҝӃҮ ҰҸѸҸҊ ӊҽѸ ӊ қҸӀ, ӊҸѦӁӁҏӀ ѽѹҲҝҸҡҲҧ, ҲҽҔҲѹҍѦҡҔѹҮ ҰҝѹѦѸ, ҏґҸӊҔҏ, ҰҝѹҡҰҸҡҸқґѦӁѹѦ, ҰҝѹҦҏѸҽӁӁҸѦ ӊҽѸѹ. ҭҝѦҦҡҲҽӊҧҲѦ ҡѦқѦ, ҍҲҸ ӊҽҗ ҡӃӁ ҏѽҸҦѹҲ Ӂҽ ӊҸҮӁҏ. ѰѽҸҦѹҲ ӁѦҸҰӃҲӁӃѸ, ӁѦҸқҡҲҝѦґҾӁӁӃѸ. ѩҽҮҲѦ ѦѸҏ ҡҸӊѦҲ. (ѢҘҩѤ) (44) үқҡҸґӀҲӁҸ “ҦҸҡҲҸѦӊҡҔҽҾ” ѹҡҲҸҝѹҾ ҡ ҔҸґҧ҃ҸѸ (ңѹӉӁҧ ҰҸҦҝҽңҽѦҲ ґѹҲѦҝҽҲҏҝѦ), ҔҸҲҸҝҸѦ һѦӊҽ ѹӉ ҝѦӊӁҸҡҲѹ ҡӁҽҍҽґҽ қѦҝѦҲ ҏ ҨҽѹӁӃ, ҰҸҲҸѸ ӊҸӉӊҝҽӂҽѦҲ, ӊӃҦҽӊҽҾ Ӊҽ ҦҝҏҊҸѦ, ҡҸӊҡѦѸ ҦѦҗѦӊҸѦ, ѹ ҸқӁҽңҽҾ ѦѦ ӁѦӊѹӁӁҏӀ ґҸңҧ, ҡҸҰҝҸӊҸңҦҽѦҲҡҾ ҰѦҍҸҝѹӁҡҔѹѸ қѦӉӁҽҦѦңӁӃѸ ҽӍҸҝѹӉѸҸѸ ҙҰѹҊҝҽӍҽ. (ѢҘҩѤ)

9

ҭҸѹҡҔ ӊ ҝѦҡҏҝҡҽѽ ҡѦҲѹ ѨӁҲѦҝӁѦҲ ҦҽѦҲ ӁѦҔҸҲҸҝҸѦ ҔҸґѹҍѦҡҲӊҸ ҔҸӁҲҝҰҝѹѸѦҝҸӊ. ѷҽҔѹѦ ҡґҏҍҽѹ, ҸҦӁҽҔҸ, ҦҸҡҲҽҲҸҍӁҸ ҝѦҦҔѹ. 10 ѩӊҽ ҦҝҏҊѹѦ ӁҽҝѦҍѹҾ ҡҸ ӉӁҽҍѦӁѹѦѸ ҰҸґӁҸҮ ҡҲѦҰѦӁѹ ҔҽҍѦҡҲӊҽ – ԂԼԙӹԊԇԽӰӹ ѹ ԙӹӴӾӢԶӾӰӰӹ – ӁѦ ѹҡҰҸґҧӉҏӀҲҡҾ ҡ ҽӁҲҸӁѹѸҽѸѹ ҔҸӁҲҝҽҝӁҸҊҸ ҲѹҰҽ (үҰҝѦҡҾӁ [ҝѦҦ.] 2004: 1072-1074). ҶҲѹ ӁҽҝѦҍѹҾ ѸҸҊҏҲ ҡҸҍѦҲҽҲҧҡҾ ҡ ҰҝѹґҽҊҽҲѦґҧӁӃѸѹ, ӊӃҝҽңҽӀӂѹѸѹ ӁѦҊҝҽҦҏѹҝҏѦѸӃѦ ҡӊҸҮҡҲӊҽ (ԙӹӴӾӢԶӾӰӰӹ ԼӾԙԙԐՂԙԊӾӰӰՂӡ, ԂԼԙӹԊԇԽӰӹ ԄӹԊՂӡ). ҫӁѹ ҲҽҔңѦ ҡҸҍѦҲҽӀҲҡҾ ҡ ҲҽҔ ӁҽӉӃӊҽѦѸӃѸѹ “ҙҲҽґҸӁӁӃѸѹ” ҰҝѹґҽҊҽҲѦґҧӁӃѸѹ (ҩҏӉѹӁ 1994, ҹҝҽѸѸ 1979), ҲҽҔѹѸѹ ҔҽҔ ҰҝѹґҽҊҽҲѦґҧӁӃѦ ҃ӊѦҲӃ ѹ ӍҸҝѸӃ (ԙӹӴӾӢԶӾӰӰӹ ӪӢӫԄԊՂӡ, ԂԼԙӹԊԇԽӰӹ ӪӢԂԙӰՂӡ). ғ ҰҸҡґѦҦӁѦѸ ҡґҏҍҽѦ ӁҽҝѦҍѹҾ ҸқҸӉӁҽҍҽӀҲ ҰҸґӁҸѦ ҰҸҦҸқѹѦ ҙҲҽґҸӁӁҸѸҏ ҸқӆѦҔҲҏ (ӁҽҰҝѹѸѦҝ, ҊѦҸѸѦҲҝѹҍѦҡҔҸҮ ӍҸҝѸѦ ҔҝҏҊҽ ѹґѹ ҰҝҸҲҸҲѹҰѹҍѦҡҔҸѸҏ ҔҝҽҡӁҸѸҏ ҃ӊѦҲҏ).

ѢѣҋҘҫһѻҘҫ үҘҋѨҫҴ ҫ ҭҩѨһүҐүѷѣһѻѢѳҤ-үѢѷҫѢѨҴүҤ

481

ҋґҸӊҸҡҸҍѦҲҽӁѹҾ ԙӹӴԙӾԐ ԐԂԊӾӰӸӪӠӡ/ԼӾԌӰՂӡ/ӪӹӢӹԽӪӠӡ/ԌӾԶӾӴՂӡ ҽқҡҸґӀҲӁҸ ҔҸҝҝѦҔҲӁӃ ѹ ӊѦҡҧѸҽ ҍҽҡҲҸҲӁӃ. ҫӁѹ ҸӉӁҽҍҽӀҲ, ҍҲҸ ѸҽҔҡѹѸҽґҧӁҽҾ ҡҲѦҰѦӁҧ ҰҝѹӉӁҽҔҽ ҦґҾ ҦҽӁӁҸҊҸ ҝҸҦҽ ҸқӆѦҔҲҸӊ ҦҸҡҲѹҊӁҏҲҽ. ҤҸҲѦґҸҡҧ қӃ ӉҽҸҡҲҝѹҲҧ ӊӁѹѸҽӁѹѦ Ӂҽ ӍҸҝѸҏґѹҝҸӊҔѦ ԌԊԋ ԌԂӰӰӹԄӹ ӢӹԌԂ ӹԼԔӾӪԽӹӴ, ҰҸҡҔҸґҧҔҏ ҡҸҍѦҲҽӁѹҾ ҰҝѹґҽҊҽҲѦґҧӁӃѽ-ҽӁҲҸӁѹѸҸӊ ҡ ӁҽҝѦҍѹѦѸ ԙӹӴԙӾԐ ӁѦ ҸӉӁҽҍҽӀҲ, ҍҲҸ ҰҝѹӉӁҽҔ ҰѦҝѦѽҸҦѹҲ ӊ Ӂҏґҧ ҔҽҍѦҡҲӊҽ, ҽ ҡѹҊӁҽґѹӉѹҝҏӀҲ ҦҸҡҲѹңѦӁѹѦ ӪԂԽӾԄӹӢӠԂԊӸӰӹԄӹ ѸҽҔҡѹѸҏѸҽ. ѢҽҰҝѹѸѦҝ, ѸҽҔҡѹѸҽґҧӁҸ ӁѹӉҔѹҮ ҝҸҡҲ – ҙҲҸ ӁѦ ӁҏґѦӊҸҮ ҝҸҡҲ, ҽ ӁѹңӁѹҮ ҰҝѦҦѦґ ӊѦҝҲѹҔҽґҧӁҸҮ ҰҝҸҲҾңѦӁӁҸҡҲѹ ґӀҦѦҮ. ѣҡґѹ ҝҸҡҲ ҦҽӁӁҸҊҸ ѹӁҦѹӊѹҦҽ ӁѹңѦ ҙҲҸҊҸ ҰҝѦҦѦґҽ, ѸӃ, ӊѦҝҸҾҲӁҸ, ӁҽӉҸӊѦѸ ҲҽҔҸҊҸ ҍѦґҸӊѦҔҽ ҔҽҝґѹҔҸѸ. ҭҸҦҸқӁӃѸ ңѦ ҸқҝҽӉҸѸ, ҊҸҝҽ ӁѦ ѸҸңѦҲ қӃҲҧ қѦҡҔҸӁѦҍӁҸ ӁѹӉҔҸҮ, ӁҽҝҽҡҲҽӁѹѦ ҰҝѹӉӁҽҔҽ ‘ӁѹӉҔѹҮ’ ҰҝѹӊѦҦѦҲ Ҕ ҰѦҝѦҸҡѸӃҡґѦӁѹӀ ҔҽҲѦҊҸҝѹҽґҧӁҸҮ ҰҝѹӁҽҦґѦңӁҸҡҲѹ ҦҽӁӁҸҊҸ ҸқӆѦҔҲҽ, ҔҸҲҸҝӃҮ Ұҝѹ ҙҲҸѸ ҰѦҝѦҮҦѦҲ ӊ Ҕґҽҡҡ ѽҸґѸҸӊ. ѢҏңӁҸ ҡҦѦґҽҲҧ ѦӂѦ ҦӊѦ ҸҊҸӊҸҝҔѹ. ғҸ-ҰѦҝӊӃѽ, ӁҽҝѦҍѹѦ ԙӹӴԙӾԐ ѹѸѦѦҲ ҡҝҽӊӁѹҲѦґҧӁӃҮ ҸҲҲѦӁҸҔ ӉӁҽҍѦӁѹҾ. ғ ҡҸҍѦҲҽӁѹѹ ҡ ҔҽҍѦҡҲӊѦӁӁӃѸѹ ҰҝѹґҽҊҽҲѦґҧӁӃѸѹ ҙҲҸ ӁҽҝѦҍѹѦ ҰҸҦҝҽӉҏѸѦӊҽѦҲ ҡҝҽӊӁѦӁѹѦ ҡҲѦҰѦӁѹ ҰҝѹӉӁҽҔҽ ӊ ҦҽӁӁҸѸ ҸқӆѦҔҲѦ ҡ ѦҊҸ ҰҝҸҾӊґѦӁѹѦѸ ӊ ҦҝҏҊѹѽ ҸқӆѦҔҲҽѽ ѹґѹ ӊ ҲҸѸ ңѦ ҡҽѸҸѸ ҸқӆѦҔҲѦ ӊ ҦҝҏҊҸҮ ѸҸѸѦӁҲ ӊҝѦѸѦӁѹ (үҰҝѦҡҾӁ [ҝѦҦ.] 2004: 1073, Filipenko 1998: 283). ҋґѦҦҏӀӂѹѦ ҰҝѹѸѦҝӃ ҏҦҽҍӁҸ ѹґґӀҡҲҝѹҝҏӀҲ ҦҽӁӁҏӀ ҸҡҸқѦӁӁҸҡҲҧ ӁҽҝѦҍѹҾ ԙӹӴԙӾԐ: (45) “ҭҝѹӁ҃ѹҰ ҝѦӍҸҝѸӃ ѭҘҤ ӊ ғҸґҸҊҸҦҡҔҸҮ ҸқґҽҡҲѹ ҲҽҔҸҮ: қҸҊҽҲӃҮ ҦҸґңѦӁ ҰґҽҲѹҲҧ қҸґҧҗѦ, қѦҦӁӃҮ – ҡҲҸґҧҔҸ ңѦ, ҡҔҸґҧҔҸ ҰґҽҲѹґ ҝҽӁҧҗѦ, ҽ ҡҸӊҡѦѸ қѦҦӁӃҮ – ѸѦӁҧҗѦ”, – ӉҽҾӊѹґ ҊҏқѦҝӁҽҲҸҝ ҭҸӉҊҽґѦӊ. (ѢҘҩѤ) (46) ғ ҴҽґӃҮ ҲѦҽҲҝ ѹґѹ ғҽѽҲҽӁҊҸӊҡҔѹҮ, ӊ ѷѦҽҲҝ Ӂҽ ҴҽґҸҮ ҷҝҸӁӁҸҮ ѹґѹ ӊ ѷҌ҈ ѸҸңӁҸ ѽҸҝҸҗҸ ҡѦҡҲҧ ӊ ҰҽҝҲѦҝѦ Ӊҽ 20 ҲӃҡҾҍ. Ѩ ӊ ӉҽґѦ ӊҡѦҊҦҽ ѦҡҲҧ ҡӊҸқҸҦӁӃѦ ѸѦҡҲҽ. ғ ҲҝҏҗҔѹӁҡҔҸѸ ѷѦҽҲҝѦ үӁҲҸӁҽ ѯѦѽҸӊҽ ҡӊҸқҸҦӁӃѽ ѸѦҡҲ ӁѦҲ. ҈ӁҽҍѹҲ, қѹґѦҲӃ, ҦҏѸҽґ Ҿ, ҡҸӊҡѦѸ ҦѦҗѦӊӃѦ! ҫҔҽӉҽґҸҡҧ, ҍҲҸ ҸӁѹ ҡҽѸӃѦ ҦҸҝҸҊѹѦ ӊ ҴҸҡҔӊѦ. (ѢҘҩѤ) (47) ѢѦ ѸѦӁѦѦ ӊҽңѦӁ ѹ ҰҡѹѽҸґҸҊѹҍѦҡҔѹҮ ӍҽҔҲҸҝ: Ӂҽ ҰҸӁѦҦѦґҧӁѹҔ ґӀҦѹ ҡҲҽҝҽӀҲҡҾ ҸҡҸқҸ ѸӁҸҊҸ ҦѦґ ӁѦ ҰґҽӁѹҝҸӊҽҲҧ – ҙҲҸҲ ҦѦӁҧ ҏѽҸҦѹҲ “Ӂҽ ҝҽҡҔҽҍҔҏ”. ғҸ ӊҲҸҝӁѹҔ ґӀҦѹ “ҡҸӉҝѦӊҽӀҲ” ҦҸ ӊӃѦӉҦҽ ӊ ҡӊѦҲ. ҋҝѦҦҽ ҰҝҸѽҸҦѹҲ ҔҽҔ ҰҸӁѦҦѦґҧӁѹҔ: ґӀҦѹ ӉҽӁҾҲӃ ҰґҽӁѹҝҸӊҽӁѹѦѸ ҲҸҊҸ, ҍҲҸ ѦӂѦ ӁҽҦҸ ҡҦѦґҽҲҧ Ӂҽ ӁѦҦѦґѦ, ѹ “ҸқҝҽқҸҲҔҸҮ” ҡҦѦґҽӁӁҸҊҸ ӊҸ ӊҲҸҝӁѹҔ. ѯѦҲӊѦҝҊ ѸҸңӁҸ ҡҍѹҲҽҲҧ ҰҸҡґѦҦӁѹѸ ҦӁѦѸ ӁѦҦѦґѹ, ӊѦҦҧ ӊ ҰҾҲӁѹ҃ҏ ҏ ѸӁҸҊѹѽ “ҔҸҝҸҲҔѹҮ” ҝҽқҸҍѹҮ ҦѦӁҧ (ҽ ӊ ҊҸҡҏҦҽҝҡҲӊѦӁӁӃѽ ҏҍҝѦңҦѦӁѹҾѽ ҸӁ “ҡҸӊҡѦѸ ҔҸҝҸҲҔѹҮ”). ҭҸҙҲҸѸҏ ӊҲҸҝӁѹҔ ѹ ҍѦҲӊѦҝҊ ҡҲҽӁҸӊҾҲҡҾ ҰѹҔҸӊӃѸѹ. (ѢҘҩѤ)

482

ѣһѣѢү ѷҩѨҷѰҹѨѢѨѢү

(48) ѩҸѸӁҽ ҭґҽҲҸӁҸӊӁҽ ҝҸҡҲҏ ӁѦӊӃҡҸҔҸҊҸ, ѹ ҦҽңѦ ҸҍѦӁҧ ӁѦӊӃҡҸҔҸҊҸ, ҽ ҡҔҸҝѦѦ ҡҸӊҡѦѸ ӁѹӉѦӁҧҔҽҾ, ӁҸ ӊҡѦѸ ҸӁҽ ҰҸҔҽӉӃӊҽѦҲҡҾ ҍѦґҸӊѦҔҸѸ ҔҝҏҰӁӃѸ. (ѢҘҩѤ) ғ ҰҝѹѸѦҝѦ (45) ҝѦӍѦҝѦӁҲӃ ҝҽҡҰҸґҽҊҽӀҲҡҾ Ӂҽ җҔҽґѦ ѸҽҲѦҝѹҽґҧӁҸҊҸ ҰҸґҸңѦӁѹҾ. ҭҝѹҍѦѸ җҔҽґҽ ҔҸӁҡҲҝҏѹҝҏѦҲҡҾ ӊ ҙҲҸѸ ҰҝѹѸѦҝѦ ҰҸ ҰҝѹӁ҃ѹҰҏ ӁҽҝҽҡҲҽӁѹҾ ҰҝѹӁ҃ѹҰҽ қѦҦӁҸҡҲѹ – ҸҲ ӁҏґѦӊҸҮ қѦҦӁҸҡҲѹ (қҸҊҽҲӃѦ), ҍѦҝѦӉ ҰҝҸҡҲҸ қѦҦӁҸҡҲҧ, Ҕ ҔҝҽҮӁѦҮ ҡҲѦҰѦӁѹ қѦҦӁҸҡҲѹ. ѩҽӁӁӃҮ ҝҾҦ ҡҝҽӊӁѦӁѹҾ ѸҸҲѹӊѹҝҏѦҲ ҏҰҸҲҝѦқґѦӁѹѦ ӁҽҝѦҍѹҾ ҰҸґӁҸҮ ҡҲѦҰѦӁѹ ҔҽҍѦҡҲӊҽ ҦґҾ ҸқҸӉӁҽҍѦӁѹҾ ҰҸҡґѦҦӁѦҊҸ ҙґѦѸѦӁҲҽ Ӂҽ җҔҽґѦ. ҭҸҦҸқӁӃѸ ңѦ ҸқҝҽӉҸѸ, ӊ ҰҝѹѸѦҝѦ (46) ҃ѦӁҽ қѹґѦҲҸӊ ӊ ѷѦҽҲҝѦ үӁҲҸӁҽ ѯѦѽҸӊҽ ҰҝѦҦҡҲҽӊґѦӁҽ ҊҸӊҸҝҾӂѹѸ ҔҽҔ ҔҝҽҮӁҾҾ ҲҸҍҔҽ Ӂҽ җҔҽґѦ ѸҸҡҔҸӊҡҔѹѽ ҲѦҽҲҝҸӊ. ғ ҰҝѹѸѦҝѦ (47) ҰҾҲӁѹ҃ҽ ӁҽӉӊҽӁҽ “ҡҸӊҡѦѸ ҔҸҝҸҲҔѹѸ” ҝҽқҸҍѹѸ ҦӁѦѸ ӊ ҊҸҡҏҦҽҝҡҲӊѦӁӁӃѽ ҏҍҝѦңҦѦӁѹҾѽ, ҲҽҔ ҔҽҔ ҸӁѹ, ҰҸ ѸӁѦӁѹӀ ҊҸӊҸҝҾӂѦҊҸ, ҝҽҡҰҸґҽҊҽӀҲҡҾ Ӂҽ ҙҔҡҲҝѦѸҏѸѦ ҡҸҸҲӊѦҲҡҲӊҏӀӂѦҮ җҔҽґӃ. ѩҝҏҊѹѸѹ ҡґҸӊҽѸѹ, ҊҸҡҏҦҽҝҡҲӊѦӁӁӃѦ ҏҍҝѦңҦѦӁѹҾ ҝҽқҸҲҽӀҲ ѸѦӁҧҗѦѦ ҔҸґѹҍѦҡҲӊҸ ҍҽҡҸӊ ҰҸ ҡҝҽӊӁѦӁѹӀ ҡҸ ӊҡѦѸѹ ҦҝҏҊѹѸѹ ҸҝҊҽӁѹӉҽ҃ѹҾѸѹ. ҷҸґѦѦ ҲҸҊҸ, ҰҝҸҦҸґңѹҲѦґҧӁҸҡҲҧ ѹѽ ҝҽқҸҍѦҊҸ ҦӁҾ ҰҝѦҦҡҲҽӊґѦӁҽ ӊ (47) ҔҽҔ ҰҝѦҦѦґҧӁҸ ҔҸҝҸҲҔҽҾ, ӊ ҲҸѸ ҡѸӃҡґѦ, ҍҲҸ ҔҸҝҸҍѦ ҝҽқҸҍѹҮ ҦѦӁҧ қӃҲҧ ҏңѦ ӁѦ ѸҸңѦҲ. ҭҝѹѸѦҝ (48) ӁѦҡҔҸґҧҔҸ ҸҲґѹҍҽѦҲҡҾ ҸҲ ҰҝѦҦӃҦҏӂѹѽ ҰҝѹѸѦҝҸӊ ҲѦѸ, ҍҲҸ ӉҦѦҡҧ ҡҏқӆѦҔҲ (ѩҸѸӁҽ ҭґҽҲҸӁҸӊӁҽ) ҡҝҽӊӁѹӊҽѦҲҡҾ ӁѦ ҡ ҦҝҏҊѹѸѹ ґӀҦҧѸѹ ѹ ҦҽңѦ ӁѦ ҡ ҡҽѸҸҮ ҡҸқҸҮ ӊ ҦҝҏҊҸҮ ѸҸѸѦӁҲ ӊҝѦѸѦӁѹ. ғ ҦҽӁӁҸѸ ҡґҏҍҽѦ ҊҸӊҸҝҾӂѹҮ ҦӊѹҊҽѦҲҡҾ ҰҸ җҔҽґѦ ҝҸҡҲҽ ӊ ҰҸҰӃҲҔѦ ҦҽҲҧ ҔҽҔ ѸҸңӁҸ қҸґѦѦ ҲҸҍӁҏӀ ѽҽҝҽҔҲѦҝѹҡҲѹҔҏ ҝѦӍѦҝѦӁҲҏ. ѷҽҔѹѸ ҸқҝҽӉҸѸ, ҡҝҽӊӁѦӁѹӀ ҰҸҦӊѦҝҊҽѦҲҡҾ ӁѦ ҡҽѸҽ ѩҸѸӁҽ ҭґҽҲҸӁҸӊӁҽ, ҽ ҡҏңҦѦӁѹҾ ҊҸӊҸҝҾӂѦҊҸ Ҹ ӁѦҮ. ғҸ-ӊҲҸҝӃѽ, ҲҸґҧҔҸ ҰҝѹґҽҊҽҲѦґҧӁӃѦ ѸҽґҸҊҸ ҰҸґӀҡҽ ѸҸҊҏҲ ҡҸҍѦҲҽҲҧҡҾ ҡ ӁҽҝѦҍѹѦѸ ԙӹӴԙӾԐ; ҰҝѹґҽҊҽҲѦґҧӁӃѦ қҸґҧҗҸҊҸ ҰҸґӀҡҽ ӁѦҰҝѹѦѸґѦѸӃ ӊ ҡҸҍѦҲҽӁѹѹ ҡ ѸҽҔҡѹѸҽҮӉѦҝҽѸѹ. ғӊѹҦҏ ҦҽӁӁҸҮ ҽҡѹѸѸѦҲҝѹѹ, үҰҝѦҡҾӁ (1974: 66) ӁҽӉӃӊҽѦҲ ҰҝѹґҽҊҽҲѦґҧӁӃѦ ѸҽґҸҊҸ ҰҸґӀҡҽ ԗӢӾԌӾԊӸӰՂԐӠ, ҽ ѹѽ ҽӁҲҸӁѹѸӃ ӰӾԗӢӾԌӾԊӸӰՂԐӠ. ҴҸңӁҸ ҰҝѦҦҰҸґҸңѹҲҧ, ҍҲҸ ҰҸҦҸқӁҽҾ ҽҡѹѸѸѦҲҝѹҾ ҡӊҾӉҽӁҽ ҡ ӁҽґѹҍѹѦѸ ҰѦҝ҃ѦҰҲҏҽґҧӁҸ ҡҽґѹѦӁҲӁҸҊҸ ӁѹңӁѦҊҸ ҰҝѦҦѦґҽ (ӉѦѸґҾ, ӁҸґҧ) ѹ ҸҲҡҏҲҡҲӊѹѦѸ ҰҝҸѸѹӁѦӁҲӁҸҮ ҊҝҽӁѹ҃Ӄ ӊ ӉҸӁѦ қҸґҧҗҸҊҸ ҰҸґӀҡҽ. ғҽңӁӃҮ ӊӃӊҸҦ, ҔҸҲҸҝӃҮ ѸҸңӁҸ ҡҦѦґҽҲҧ ӊ ҡӊѦҲѦ ҦҽӁӁӃѽ ѹӉ ҝҏҡҡҔҸҊҸ ҾӉӃҔҽ, ӉҽҔґӀҍҽѦҲҡҾ ӊ ҲҸѸ, ҍҲҸ ҝѦґҾҲѹӊӁӃѦ ҰҝѹґҽҊҽҲѦґҧӁӃѦ ҲҸңѦ ѸҸҊҏҲ ҰҝҸѦ҃ѹҝҸӊҽҲҧҡҾ Ӂҽ ӉҽѸҔӁҏҲӃѦ җҔҽґӃ ѹ ҸқҸӉӁҽҍҽҲҧ ѸҽҔҡѹѸҏѸ ҰҝѹӉӁҽҔҽ ӊ ҡҸҍѦҲҽӁѹѹ ҡ ѸҽҔҡѹѸҽҮӉѦҝҽѸѹ. ѷҸҲ ӍҽҔҲ, ҍҲҸ ҰҸҦҸқӁӃѦ ҡҸҍѦҲҽӁѹҾ ӁѦӊҸӉѸҸңӁӃ ӊ ҽӁҊґѹҮҡҔҸѸ ѹ ҝҾҦѦ ҝҸҦҡҲӊѦӁӁӃѽ ҾӉӃҔҸӊ, ѦӂѦ ӁѦ ҦҽѦҲ ҸҡӁҸӊҽӁѹҮ ҡҲҝҸѹҲҧ Ӂҽ ҙҲҸѸ ҡѦѸҽӁҲѹҍѦҡҔҏӀ ҲѦҸҝѹӀ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ӊҸҸқӂѦ. ҋґѦҦҏѦҲ ҲҽҔңѦ ҡҔҽӉҽҲҧ, ҍҲҸ ҦҽӁӁҽҾ ҸҡҸқѦӁӁҸҡҲҧ ҝҏҡҡҔҸҊҸ ҾӉӃҔҽ ҸҲӁӀҦҧ ӁѦ ҾӊґҾѦҲҡҾ ҍѦѸ-ҲҸ ҏӁѹҔҽґҧӁӃѸ. ҋҸҊґҽҡӁҸ ѸҸѹѸ ѹӁӍҸҝѸҽӁҲҽѸ (ґѹӁҊӊѹҡҲҽѸ), ҰҝѹґҽҊҽҲѦґҧӁӃѦ-ҽӁҲҸӁѹѸӃ ѸҸҊҏҲ ҸҰҝѦҦѦґҾҲҧҡҾ ӁҽҝѦҍѹҾѸѹ ҰҸґӁҸҮ ҡҲѦҰѦӁѹ ӊ ҦҝҏҊѹѽ ҡґҽӊҾӁҡҔѹѽ ҾӉӃҔҽѽ (ҰҸґҧҡҔҸѸ, ҍѦҗҡҔҸѸ, қҸґҊҽҝҡҔҸѸ ѹ Ҳ.Ҧ.), ҽ ҲҽҔңѦ ӊ ӊѦӁҊѦҝҡҔҸѸ, ҊҝѦҍѦҡҔҸѸ ѹ ӍҝҽӁ҃ҏӉҡҔҸѸ ҾӉӃҔҽѽ. ғ ҙҲѹѽ ҾӉӃҔҽѽ ҲҽҔңѦ Ӂҽ-

ѢѣҋҘҫһѻҘҫ үҘҋѨҫҴ ҫ ҭҩѨһүҐүѷѣһѻѢѳҤ-үѢѷҫѢѨҴүҤ

483

қґӀҦҽѦҲҡҾ ҸҲѸѦҍѦӁӁҽҾ ӊӃҗѦ ҽҡѹѸѸѦҲҝѹҾ ѸѦңҦҏ ҰҝѹґҽҊҽҲѦґҧӁӃѸѹ қҸґҧҗҸҊҸ ѹ ѸҽґҸҊҸ ҰҸґӀҡҽ. 6. ҈ҽҔґӀҍѦӁѹѦ ғ ӁҽҡҲҸҾӂѦҮ ҝҽқҸҲѦ қӃґѹ ҰҝѦҦҡҲҽӊґѦӁӃ ҦҽӁӁӃѦ ѹӉ ҝҏҡҡҔҸҊҸ ҾӉӃҔҽ, ҔҸҲҸҝӃѦ ҾӊӁҸ ҰҝҸҲѹӊҸҝѦҍҽҲ ҏҡҲҸҾӊҗѹѸҡҾ ӊӉҊґҾҦҽѸ, ӊҸ ѸӁҸҊҸѸ ҸҰҝѦҦѦґҾӀӂѹѸ ҡҸӊҝѦѸѦӁӁҸѦ ҰҸґҸңѦӁѹѦ ҡѦѸҽӁҲѹҍѦҡҔҸҮ ҲѦҸҝѹѹ ѹѸѦӁ ҰҝѹґҽҊҽҲѦґҧӁӃѽ. ҋґѦҦҏӀӂѹѦ ҸҡӁҸӊӁӃѦ ҰҸҡҲҏґҽҲӃ Ҹ ҡҔҽґҾҝӁӃѽ ҰҝѹґҽҊҽҲѦґҧӁӃѽ-ҽӁҲҸӁѹѸҽѽ қӃґѹ ҔҝѹҲѹҍѦҡҔѹ ҝҽҡҡѸҸҲҝѦӁӃ Ӂҽ ѸҽҲѦҝѹҽґѦ ҝҏҡҡҔҸҊҸ ҾӉӃҔҽ: ҸқґѹҊҽҲҸҝӁҽҾ ҸҝѹѦӁҲҽ҃ѹҾ ӍҸҝѸ ҰҸґҸңѹҲѦґҧӁҸҮ ҡҲѦҰѦӁѹ Ӂҽ ҡҝѦҦӁѦҡҲҽҲѹҡҲѹҍѦҡҔѹҮ ҡҲҽӁҦҽҝҲ, ҸқґѹҊҽҲҸҝӁҽҾ ӁѦѸҽҝҔѹҝҸӊҽӁӁҸҡҲҧ ӍҸҝѸ ҡҝҽӊӁѹҲѦґҧӁҸҮ ҡҲѦҰѦӁѹ ѹ ӁѦҰҝѦҦѦґҧӁҸҡҲҧ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ҡҔҽґҾҝӁҸ-ҽӁҲҸӁѹѸѹҍѦҡҔҸҊҸ ҔҸѸҰґѦҔҡҽ. ѯҲҸ ҔҽҡҽѦҲҡҾ ҸқҾӉҽҲѦґҧӁҸҮ ҸҝѹѦӁҲҽ҃ѹѹ ӍҸҝѸ ҰҸґҸңѹҲѦґҧӁҸҮ ҡҲѦҰѦӁѹ Ӂҽ ӁҸҝѸҏ, ҲҸ ҦҽӁӁҸѦ ҰҸґҸңѦӁѹѦ ӁѦҡҸӊѸѦҡҲѹѸҸ ҡ ӁҽґѹҍѹѦѸ ӊ ҝҏҡҡҔҸѸ ҾӉӃҔѦ Ҧӊҏѽ ӍҸҝѸ ҰҸґҸңѹҲѦґҧӁҸҮ ҡҲѦҰѦӁѹ – ҰҸґӁҸҮ ѹ ҔҝҽҲҔҸҮ. ҘҽҔ қӃґҸ ҰҸҔҽӉҽӁҸ ӊӃҗѦ, ҲҸґҧҔҸ ҰҸґӁҽҾ ӍҸҝѸҽ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ҍҽҡҲҸ (ӁҸ ҸҲӁӀҦҧ ӁѦ ӊҡѦҊҦҽ!) ҸҝѹѦӁҲѹҝҸӊҽӁҽ Ӂҽ ҡҝѦҦӁѦҡҲҽҲѹҡҲѹҍѦҡҔѹҮ ҡҲҽӁҦҽҝҲ ҰҝѹӉӁҽҔҽ. ҘҝҽҲҔҽҾ ӍҸҝѸҽ ҡҔҽґҾҝӁӃѽ ҰҝѹґҽҊҽҲѦґҧӁӃѽ, ҔҽҔ ҰҝҽӊѹґҸ, ҸқҸӉӁҽҍҽѦҲ ҰҝѹӉӁҽҔ, ӁѦ ӉҽӊѹҡѹѸӃҮ ҸҲ ӁҸҝѸӃ ҝѦӍѦҝѦӁ҃ѹҽґҧӁҸҊҸ Ҕґҽҡҡҽ. ѰҲӊѦҝңҦѦӁѹѦ Ҹқ ҸқґѹҊҽҲҸҝӁҸҮ ӁѦѸҽҝҔѹҝҸӊҽӁӁҸҡҲѹ ӍҸҝѸ ҡҝҽӊӁѹҲѦґҧӁҸҮ ҡҲѦҰѦӁѹ ӁѦҡҸӊѸѦҡҲѹѸҸ ҡ ҲѦѸ ӍҽҔҲҸѸ, ҍҲҸ қҸґҧҗѹӁҡҲӊҸ ҰҝѹґҽҊҽҲѦґҧӁӃѽ-ҽӁҲҸӁѹѸҸӊ ӊ ҝҏҡҡҔҸѸ ҾӉӃҔѦ ѹѸѦӀҲ ҦӊѦ ӍҸҝѸӃ ҔҸѸҰҽҝҽҲѹӊҽ – ҡѹӁҲѦҲѹҍѦҡҔҏӀ ѹ ҽӁҽґѹҲѹҍѦҡҔҏӀ. ҘҽҔ ҰҸҔҽӉҽґҸ ӁҽҡҲҸҾӂѦѦ ѹҡҡґѦҦҸӊҽӁѹѦ, ҡѹӁҲѦҲѹҍѦҡҔѹѦ ӍҸҝѸӃ, ҔҽҔ ҰҝҽӊѹґҸ, ӁѦѸҽҝҔѹҝҸӊҽӁӃ, ӊ ҲҸ ӊҝѦѸҾ ҔҽҔ ҽӁҽґѹҲѹҍѦҡҔѹѦ ӍҸҝѸӃ ѹѸѦӀҲ ҲѦӁҦѦӁ҃ѹӀ ӊӃҝҽңҽҲҧ ѸҽҝҔѹҝҸӊҽӁӁӃѦ ҔҸѸҰҽҝҽҲѹӊӁӃѦ ҸҲӁҸҗѦӁѹҾ. ҷҸґѦѦ ҲҸҊҸ, ҦҽӁӁҽҾ ҲѦӁҦѦӁ҃ѹҾ ҦѦҮҡҲӊѹҲѦґҧӁҽ (ѽҸҲҾ ѹ ӊ ѸѦӁҧҗѦҮ ҡҲѦҰѦӁѹ) ҲҽҔңѦ ҦґҾ ҽӁҊґѹҮҡҔҸҊҸ ҾӉӃҔҽ. ҩѦӉҏґҧҲҽҲӃ ӁҽҡҲҸҾӂѦҮ ҝҽқҸҲӃ ҰҸҔҽӉҽґѹ ҲҽҔңѦ, ҍҲҸ ӁѦҰҝѦҦѦґҧӁҸҡҲҧ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ҡҔҽґҾҝӁҸ-ҽӁҲҸӁѹѸѹҍѦҡҔҸҊҸ ҔҸѸҰґѦҔҡҽ, ҸҡӁҸӊҽӁӁҽҾ Ӂҽ ѹѽ ӁѦҡҸҍѦҲҽѦѸҸҡҲѹ ҡ ѸҽҔҡѹѸҽҮӉѦҝҽѸѹ, ҲҽҔңѦ ӁѦ ҾӊґҾѦҲҡҾ ҏӁѹӊѦҝҡҽґҧӁӃѸ ҾӊґѦӁѹѦѸ. ѢҏңӁҸ ҰҝѹӉӁҽҲҧ, ҍҲҸ ҝҏҡҡҔѹѦ ҰҝѹґҽҊҽҲѦґҧӁӃѦ-ҽӁҲҸӁѹѸӃ, ҔҽҔ ѹ ѹѽ “ҙҔӊѹӊҽґѦӁҲӃ” ӊ ҽӁҊґѹҮҡҔҸѸ ҾӉӃҔѦ, ҰґҸѽҸ ҡҸҍѦҲҽӀҲҡҾ ҡ ӁҽҝѦҍѹҾѸѹ ҲѹҰҽ ԗӹԓԽӠ. ҫҦӁҽҔҸ, ӊ ҸҲґѹҍѹѦ ҸҲ ҽӁҊґѹҮҡҔҸҊҸ ҾӉӃҔҽ, ҝҏҡҡҔѹѦ ҰҝѹґҽҊҽҲѦґҧӁӃѦ ѸҽґҸҊҸ ҰҸґӀҡҽ ҽқҡҸґӀҲӁҸ ҰҝѹѦѸґѦѸӃ ҡ ѸҽҔҡѹѸҽҮӉѦҝҸѸ ԙӹӴԙӾԐ, ҸқҸӉӁҽҍҽӀӂѹѸ ҦҸҡҲѹңѦӁѹѦ ѸҽҔҡѹѸҽґҧӁҸҮ ҡҲѦҰѦӁѹ ҰҝѹӉӁҽҔҽ ҦґҾ ҦҽӁӁҸҊҸ ҝҸҦҽ ҸқӆѦҔҲҸӊ. ҭҸ ҙҲҸҮ ҰҝѹҍѹӁѦ ԙӹӴԙӾԐ, ҔҽҔ ҰҝҽӊѹґҸ, ҰҸҦҝҽӉҏѸѦӊҽѦҲ ҡҝҽӊӁѦӁѹѦ ҡ ҦҝҏҊѹѸѹ ҙґѦѸѦӁҲҽѸѹ ҝҾҦҽ, ҏ ҔҸҲҸҝӃѽ ҰҝѹӉӁҽҔ ҰҝѦҦҡҲҽӊґѦӁ ӊ ѸѦӁҧҗѦҮ ҡҲѦҰѦӁѹ. ҫқӂѹҮ ӊӃӊҸҦ, ҔҸҲҸҝӃҮ ѸҸңӁҸ ҡҦѦґҽҲҧ ӊ ҝѦӉҏґҧҲҽҲѦ ҰҝҸӊѦҦѦӁӁҸҊҸ ѹҡҡґѦҦҸӊҽӁѹҾ, ҡҸҡҲҸѹҲ ӊ ҲҸѸ, ҍҲҸ ҽҦѦҔӊҽҲӁҽҾ ҡѦѸҽӁҲѹҍѦҡҔҽҾ ҲѦҸҝѹҾ, ҰҝѦҲѦӁҦҏӀӂҽҾ Ӂҽ ӁѦҔҸҲҸҝҏӀ ҡҲѦҰѦӁҧ ҏӁѹӊѦҝҡҽґҧӁҸҡҲѹ, ӁѦ ѸҸңѦҲ қӃҲҧ ҰҸҡҲҝҸѦӁҽ

484

ѣһѣѢү ѷҩѨҷѰҹѨѢѨѢү

Ӂҽ ѸҽҲѦҝѹҽґѦ ҸҦӁҸҊҸ ѹґѹ ӁѦҡҔҸґҧҔѹѽ ѽҸҝҸҗҸ ѹӉҏҍѦӁӁӃѽ ҾӉӃҔҸӊ. ѷҸґҧҔҸ ҰҝѹӊґѦҍѦӁѹѦ ҦҽӁӁӃѽ ѹӉ ҾӉӃҔҸӊ ҝҽӉґѹҍӁӃѽ ҲѹҰҸґҸҊѹҍѦҡҔѹѽ ҲѹҰҸӊ ҰҸӉӊҸґѹҲ ҡҏҦѹҲҧ Ҹ ҲҸѸ, ҾӊґҾѦҲҡҾ ґѹ ӁѦҔҸѦ ҏҲӊѦҝңҦѦӁѹѦ ҏӁѹӊѦҝҡҽґҧӁҸҮ ҽҔҡѹҸѸҸҮ ѹґѹ ҔҸӁҡҲҽҲҽ҃ѹѦҮ ҍҽҡҲӁҸ-ҾӉӃҔҸӊӃѽ ӍҽҔҲҸӊ, ҦѦҮҡҲӊѹҲѦґҧӁӃѽ ҲҸґҧҔҸ ҦґҾ ӁѦҡҔҸґҧҔѹѽ қґѹӉҔҸҝҸҦҡҲӊѦӁӁӃѽ ҾӉӃҔҸӊ. ԘӾӡԌӾӰԙӪӠӡ ӫӰӠӴӾӢԙӠԽӾԽ һѨѷѣҩүѷѰҩү үҦѸҸӁѹ, ғґҽҦѹѸѹҝ Ґ. 1964 ԜԙӰӹӴՂ ԽӾӹӢӠӠ ԄӢԂԐԐԂԽӠӪӠ. ҴҸҡҔӊҽ-һѦӁѹӁҊҝҽҦ: ѢҽҏҔҽ. үҔҸҦѦҡ, Ҵѹѽҽѹґ Ѩ. 1987 ԣӰԄԊӠӡԙӪӠӡ ԋԒՂӪ: ӠԐԋ ԗӢӠԊԂԄԂԽӾԊӸӰӹӾ. ҘѹѦӊ: ғѹӂҽ җҔҸґҽ. үҰҝѦҡҾӁ, ҌҝѹҮ ѩ. 1974 ԘӾӪԙӠԓӾԙӪԂԋ ԙӾԐԂӰԽӠӪԂ: ԙӠӰӹӰӠԐӠԓӾԙӪӠӾ ԙӢӾԌԙԽӴԂ ԋԒՂӪԂ. ҴҸҡҔӊҽ: ѢҽҏҔҽ. 2005 “ҫ ҡѦѸҽӁҲѹҍѦҡҔѹѽ Ұҝҽӊѹґҽѽ”. ԮӹԊԄԂӢԙӪԂԋ ӢӫԙӠԙԽӠӪԂ 3, 3-11. үҰҝѦҡҾӁ, ҌҝѹҮ ѩ. (ҝѦҦ.) 2004 ԾӹӴՂӡ ӹԼԔԋԙӰӠԽӾԊӸӰՂӡ ԙԊӹӴԂӢӸ ԙӠӰӹӰӠԐӹӴ ӢӫԙԙӪӹԄӹ ԋԒՂӪԂ. ҴҸҡҔӊҽ-ғѦӁҽ: ѤӉӃҔѹ ҡґҽӊҾӁҡҔҸҮ ҔҏґҧҲҏҝӃ. үҝҏҲӀӁҸӊҽ, ѢѹӁҽ ѩ. 1988 ԁӠԗՂ ԋԒՂӪӹӴՂԖ ԒӰԂԓӾӰӠӡ. ԜՁӾӰӪԂ. ՀӹԼՂԽӠӾ. ԢԂӪԽ. ҴҸҡҔӊҽ: ѢҽҏҔҽ. 1999 ӵԒՂӪ Ӡ ԐӠӢ ԓӾԊӹӴӾӪԂ. ҴҸҡҔӊҽ: ѤӉӃҔѹ ҝҏҡҡҔҸҮ ҔҏґҧҲҏҝӃ. ғѹӁҸҊҝҽҦҸӊ, ғѹҔҲҸҝ ғ. 1947 ԪӫԙԙӪӠӡ ԋԒՂӪ: ԄӢԂԐԐԂԽӠԓӾԙӪӹӾ ӫԓӾӰӠӾ ӹ ԙԊӹӴӾ. ҴҸҡҔӊҽ-һѦӁѹӁҊҝҽҦ: ҐѰҭѨ Ҵҭ ҩҋҨҋҩ. ғҸґҧӍ, ѣґѦӁҽ Ҵ. 1985 ԢӫӰӪՁӠӹӰԂԊӸӰԂԋ ԙӾԐԂӰԽӠӪԂ ӹՁӾӰӪӠ. ҴҸҡҔӊҽ: ѢҽҏҔҽ. ҐӊҸӉҦѦӊ, үґѦҔҡҽӁҦҝ Ѣ. 1961 ՀӹӴӢӾԐӾӰӰՂӡ ӢӫԙԙӪӠӡ ԊӠԽӾӢԂԽӫӢӰՂӡ ԋԒՂӪ. ԁ. 1. ҴҸҡҔӊҽ: ҐѰҭѨ Ҵҭ ҩҋҨҋҩ. ҘҏҡҲҸӊҽ, ҐҽґѹӁҽ Ѩ. 1994 “ҐґҽҊҸґӃ ѹӉѸѦӁѦӁѹҾ: ҰҝҸ҃Ѧҡҡ ѹ ӁҽқґӀҦҽҲѦґҧ”. ԾԂӫԓӰӹ-ԽӾԖӰӠԓӾԙӪԂԋ ӠӰԅӹӢԐԂՁӠԋ 2, 16-32. ѢѹҔҸґҽѦӊҽ, ѷҽҲҧҾӁҽ Ҵ. 1983 “ҘҽҍѦҡҲӊѦӁӁӃѦ ҰҝѹґҽҊҽҲѦґҧӁӃѦ ѹ ҸҲҝҽңѦӁѹѦ ‘ҔҽҝҲѹӁӃ Ѹѹҝҽ’”. ғ: һ.Ѣ. ҋѸѹҝӁҸӊ (ҝѦҦ.), ӷӢӹԼԊӾԐՂ ԊӾӪԙӠӪӹԊӹԄӠӠ (ҋґҽӊҾӁҡҔҸѦ ѹ қҽґҔҽӁҡҔҸѦ ҾӉӃҔҸӉӁҽӁѹѦ 8), 235-244. ҴҸҡҔӊҽ: ѢҽҏҔҽ. ҭҸ҃ѦґҏѦӊҡҔѹҮ, ѣӊҊѦӁѹҮ ү. 1974 “ѢҏґѦӊҽҾ ҡҲѦҰѦӁҧ ҔҽҍѦҡҲӊҽ ѹ ҸҰѹҡҽӁѹѦ ӉӁҽҍѦӁѹҾ ҔҽҍѦҡҲӊѦӁӁӃѽ ҰҝѹґҽҊҽҲѦґҧӁӃѽ ѹ ҡҸҍѦҲҽӁѹҮ ҡ ӁѹѸѹ”. ғ: ғ.Ҵ. ҋҸґӁ҃Ѧӊ (ҝѦҦ.), ӷӢӹԼԊӾԐՂ ԙӾԐԂӰԽӠӪӠ, 229-247. ҴҸҡҔӊҽ: ѢҽҏҔҽ. 1977 “ҋҝҽӊӁѹҲѦґҧӁҽҾ ҡҲѦҰѦӁҧ ѹ ҡӊҸқҸҦӁҸѦ ҏҰҸҲҝѦқґѦӁѹѦ ҰҝѹґҽҊҽҲѦґҧӁӃѽ”. ԛӹԗӢӹԙՂ ԋԒՂӪӹԒӰԂӰӠԋ 5, 62-71. ҭҸҍѦҰ҃Ҹӊ, ҫґѦҊ Ґ. 1990 “ѤӉӃҔҸӊҽҾ ѸѦӁҲҽґҧӁҸҡҲҧ: ҡҰҸҡҸқ ҰҝѦҦҡҲҽӊґѦӁѹҾ Ѹѹҝҽ”. ԛӹԗӢӹԙՂ ԋԒՂӪӹԒӰԂӰӠԋ 6, 110-122.

ѢѣҋҘҫһѻҘҫ үҘҋѨҫҴ ҫ ҭҩѨһүҐүѷѣһѻѢѳҤ-үѢѷҫѢѨҴүҤ

485

ҩҏӉѹӁ, ѨӊҽӁ Ґ. 1994 “ҘҸҊӁѹҲѹӊӁӃѦ ҡҲҝҽҲѦҊѹѹ ѹѸѦӁҸӊҽӁѹҾ: ѸҸҦҏҡӃ ҰѦҝ҃ѦҰ҃ѹѹ ѹ ѹѽ ӊӃҝҽңѦӁѹѦ ӊ ҾӉӃҔѦ”. ԛӹԗӢӹԙՂ ԋԒՂӪӹԒӰԂӰӠԋ 4, 79-100. ҤҽҝѹҲҸӁҍѹҔ, ҈ѹӁҽѹҦҽ ү. 1986 ԚԐӾӰԂ ԗӢӠԊԂԄԂԽӾԊӸӰՂӾ Ӵ ԊӾӪԙӠӪӹ-ԄӢԂԐԐԂԽӠԓӾԙӪӹӡ ԙӠԙԽӾԐӾ ԙӹӴӢӾԐӾӰӰӹԄӹ ԂӰԄԊӠӡԙӪӹԄӹ ԋԒՂӪԂ. ҴѹӁҡҔ: ғӃҗҙҮҗҽҾ җҔҸґҽ. ҹӊѦҦҸӊҽ, ѢҽҲҽґѹҾ Ҍ. 1980 ԱӢԂԐԐԂԽӠӪԂ ԙӹӴӢӾԐӾӰӰӹԄӹ ӢӫԙԙӪӹԄӹ ԊӠԽӾӢԂԽӫӢӰӹԄӹ ԋԒՂӪԂ. ҴҸҡҔӊҽ: ѢҽҏҔҽ. ҹҝҽѸѸ, үґѦҔҡҽӁҦҝ Ѣ. 1979 ԜԓӾӢӪӠ ԗӹ ԙӾԐԂӰԽӠӪӾ ӪԂԓӾԙԽӴӾӰӰՂԖ ԗӢӠԊԂԄԂԽӾԊӸӰՂԖ. һѦӁѹӁҊҝҽҦ: ѨӉҦҽҲѦґҧҡҲӊҸ һѦӁѹӁҊҝҽҦҡҔҸҊҸ ҏӁѹӊѦҝҡѹҲѦҲҽ. Aesaert, Kristine, and Johanna Hauttekiet 2006 “Deduška plox, a devuška xoroša: De korte vorm van het adjectief in krantentaal”. In: E. Waegemans (red.), De taal van Peter de Grote: Russisch-Nederlandse contacten en contrasten, 107-130. Leuven-Voorburg: ACCO. Amaral, Patricía 2006 “On the semantics of almost”. Paper presented at the Annual Meeting of the Linguistic Society of America, Albuquerque, NM, 5-8 January 2006. Bierwisch, Manfred 1967 “Some semantic universals of German adjectivals”. Foundations of Language 3, 1-36. 1989 “The semantics of gradation”. In: M. Bierwisch, E. Lang (eds.), Dimensional Adjectives. Grammatical Structure and Conceptual Interpretation, 71-261. Berlin, etc.: Springer. Bolinger, Dwight 1972 Degree Words. The Hague-Paris: Mouton. Braine, Martin D.S. 1976 Children’s First Word Combinations. Chicago: University of Chicago Press. Broekhuis, Hans 1999 Adjectives and Adjective Phrases (MGD Occasional Papers 2). Tilburg: University of Tilburg. Chafe, Wallace L. 1970 Meaning and the Structure of Language. Chicago, etc: University of Chicago Press. Croft, William, and D. Alan Cruse 2004 Cognitive Linguistics. Cambridge: Cambridge University Press. Cruse, D. Alan 1976 “Three classes of antonym in English”. Lingua 38, 281-292. 1986 Lexical Semantics. Cambridge: Cambridge University Press. Cuzzolin, Pierluigi, and Christian Lehmann “Comparison and gradation”. In: G. Booij et al. (eds.), Morphologie. Ein inter2004 nationales Handbuch zur Flexion und Wortbildung. Vol. 2, 1212-1220. Berlin: Mouton de Gruyter. Ebeling, Karen S., and Susan A. Gelman 1994 “Children’s use of context in interpreting ‘big’ and ‘little’”. Child Development 65, 1178-1192.

486

ѣһѣѢү ѷҩѨҷѰҹѨѢѨѢү

Filipenko, Marina V. 1998 “On certain features: adverbs with a floating sphere of action and adverbs of degree”. Russian Linguistics 22, 275-286. Frazier, Lyn, Charles Clifton, and Britta Stolterfoht 2006 “Scale structure: Processing minimum standard and maximum standard scalar adjectives”. Paper presented at the Chicago Workshop on Scalar Meaning. May 20, 2006. University of Chicago. Graff, Delia 2000 “Shifting sands: an interest-relative theory of vagueness”. Philosophical Topics 28, 45-81. Groen, Ben M. 1998 “The use of the long and short adjectival forms in contemporary standard Russian”. In: A.A. Barentsen et al. (eds.), Dutch Contributions to the Twelfth International Congress of Slavists: Linguistics (Studies in Slavic and General Linguistics 24), 151-173. Amsterdam-Atlanta: Rodopi. Katz, Jerrold J. 1972 Semantic Theory. New York: Harper and Row Publishers. Kennedy, Christopher 1999 Projecting the Adjective. The Syntax and Semantics of Gradability and Comparison. New York: Garland. 2005 “Vagueness and grammar: the semantics of relative and absolute gradable adjectives”. To appear in Linguistics and Philosophy. Kennedy, Christopher, and Louise McNally 2005 “Scale structure, degree modification, and the semantics of gradable predicates”. Language 81, 345-381. Klein, Ewan A. 1980 “A semantics for positive and comparative adjectives”. Linguistics and Philosophy 4, 1-45. Klein, Henny 1997 Adverbs of Degree in Dutch (Groningen Dissertations in Linguistics 21). Groningen: University of Groningen. Lang, Ewald 1989 “The semantics of dimensional designation of spatial objects”. In: M. Bierwisch, E. Lang (eds.), Dimensional Adjectives. Grammatical Structure and Conceptual Interpretation, 263-417. Berlin, etc.: Springer. Leisi, Ernst 1975 Der Wortinhalt: Seine Struktur im Deutschen und Englischen. Heidelberg: Quelle & Meyer. Lyons, John 1969 Introduction to Theoretical Linguistics. Cambridge: Cambridge University Press. 1977 Semantics. Vols. 1 and 2. Cambridge: Cambridge University Press. Murphy, M. Lynne 2003 Semantic Relations and the Lexicon: Antonymy, Synonymy, and Other Paradigms. Cambridge: Cambridge University Press. Pander Maat, Henk 2006 “Subjectification in gradable adjectives”. In: A. Athanasiadou, C. Canakis, B. Cornillie (eds.), Subjectification: Various Paths to Subjectivity, 279-322. BerlinNew York: Mouton de Gruyter.

ѢѣҋҘҫһѻҘҫ үҘҋѨҫҴ ҫ ҭҩѨһүҐүѷѣһѻѢѳҤ-үѢѷҫѢѨҴүҤ Paradis, Carita 1997

487

Degree Modifiers of Adjectives in Spoken British English (Lund Studies in English 92). Lund: Lund University Press. 2000 “It’s well weird. Degree modifiers of adjectives revisited: the nineties”. In: J.M. Kirk (ed.), Corpora Galore: Analyses and Techniques in Describing English, 147-160. Amsterdam-Atlanta: Rodopi. 2001 “Adjectives and boundedness”. Cognitive Linguistics 12, 47-65. 2005 “Towards a theory of lexical meaning as ontologies and construals”. Axiomathe 15, 541-573. Paradis, Carita, and Caroline Willners 2006 “Antonymy and negation: the boundedness hypothesis”. Journal of Pragmatics 38, 1051-1080. Quirk, Randolph, Sidney Greenbaum, Geoffrey Leech, and Jan Svartvik 1985 A Comprehensive Grammar of the English Language. London: Longman. Rips, Lance J., and William Turnbull 1980 “How big is big? Relative and absolute properties in memory”. Cognition 8, 145-174. Rotstein, Carmen, and Yoad Winter 2004 “Total adjectives vs. partial adjectives: scale structure and higher-order modifiers”. Natural Language Semantics 12, 259-288. Sapir, Edward 1944 “Grading: a study in semantics”. Philosophy of Science 11, 93-116. Sera, Maria, and Linda B. Smith 1987 “Big and little: “nominal” and relative uses”. Cognitive Development 2, 89-111. Siegel, Muffy E.A. 1980 Capturing the Adjective. New York-London: Garland Publishing, Inc. Syrett, Kristen 2007 Learning About the Structure of Scales: Adverbial Modification and the Acquisition of the Semantics of Gradable Adjectives. Unpublished PhD thesis, Northwestern University. Syrett, Kristen, Evan Bradley, Christopher Kennedy, and Jeffrey Lidz 2005 “Shifting standards: Children’s understanding of gradable adjectives”. Paper presented at the Annual Meeting of the Linguistic Society of America, Oakland, CA, 6-9 January 2005. Tribushinina, Elena 2006 “Is rond even rond in het Nederlands, Engels en Russisch?”. In: E. Waegemans (ed.), De taal van Peter de Grote: Russisch-Nederlandse contacten en contrasten, 87-106. Leuven-Voorburg: ACCO. 2007 Cognitive reference points: Semantics beyond the prototypes in adjectives of space and colour. Forthcoming dissertation. Vanden Wyngaerd, Guido 2001 “Measuring events”. Language 77, 61–90. Vendler, Zeno 1968 Adjectives and Nominalizations. The Hague-Paris: Mouton. Vogel, Anna 2004 Swedish Dimensional Adjectives. Stockholm: Almqvist & Wiksell International.

488

ѣһѣѢү ѷҩѨҷѰҹѨѢѨѢү

Ward, Dennis 1965

The Russian Language Today: System and Anomaly. London: Hutchinson University Library. Wierzbicka, Anna 1996 Semantics: Primes and Universals. NY: Oxford University Press.

SUMMARY Some axioms about antonymous adjectives: Counterarguments from Russian This article questions three basic assumptions about the semantics of relative adjectives by looking at the semantic peculiarities of antonymous adjectives in Russian. Firstly, the results clearly demonstrate, counter to the previous research, that the positive form of relative adjectives is not always processed vis-à-vis a norm, i.e. an average value in the middle of the scale. Russian short adjectives, for instance, are commonly interpreted irrespective of the norm. Secondly, it is not the case that the comparative of antonymous adjectives is always unmarked. For one, Russian analytic comparatives are usually marked forms denoting committed comparative relations. Thirdly, the findings strongly suggest that there is no one-to-one relationship between an adjective type and a scale type. Whereas in languages such as English and Dutch relative adjectives are (by default) infelicitous with maximizing adverbs (e.g. completely), their counterparts in Russian can be used with the maximizer sovsem, as in sovsem nizkij and sovsem ploxoj. All in all, the results reveal the necessity to revisit some of the axiomatic assumptions about adjectival semantics and to reconsider them in view of the data from non-Germanic languages.

Dutch Contributions to the Fourteenth International Congress of Slavists, Ohrid: Linguistics (SSGL 34). Amsterdam – New York: Rodopi, 2008, 489-501.

THE GLAGOLITIC BARRIER WILLIAM R. VEDER

The 21st century has witnessed two significant advances in the study of the earliest Slavic writing system, the Glagolitic alphabet, viz. the collections of studies published after the 1996 conference of the Austrian Academy of Sciences in commemoration the 100th anniversary of the death of Vatroslav Oblak 1 and the 50th anniversary of the death of Bernd von Arnim 2 (Miklas et al. 2000) and the 2002 conference of the Croatian Academy of Sciences and the Diocese of Krk in commemoration of the 100th anniversary of the Staroslavenska Akademija 3 and the 50th anniversary of the Staroslavenski Institut (Dürrigl et al. 2004). They have added to our knowledge painstaking analyses of new sources, new interpretations of known sources, and critical assessments of previous assumptions and claims. But they also show that the way to a plausible reconstruction of the state of the Glagolitic alphabet prior to its evacuation to the Balkans (886) remains barred.4 To date, the most comprehensive outline of the problems involved in the study of the Glagolitic alphabet is provided by Marti 2004b, whose findings I summarise. The barrier rests on three dilemmas: 1 Did the original alphabet have [a] 36 or [b] 38 characters? 2 Did it contain [a] only monographs, or [b] digraphs as well? 3 Did its characters have assigned numeral values [a] above 900, or [b] only up to and including 900? On these dilemmas numerous questions of detail have aggregated, which can be subsumed under three questions: 4 What was the role of phonetic analysis in the constitution of the Glagolitic alphabet? 5 What was the role of Greek Schriftdenken in its constitution? 6 Which

1

See the summary of his work and bibliography by Angelina Daskalova in KME II: 837-841. See the summary of his work and bibliography by Rumjana Zlatanova in KME I: 105-106. 3 See the summary of its work by Anica Nazor in Dürrigl et al. 2004: 25-35. 4 A third publication, Ivanova 2004, should be mentioned for the record. It is a review of some ten years of research into problems of the Glagolitic alphabet by Russian scholars, but as their approach is by and large semiotic, the result of their labour stands strangely aloof of what is achieved elsewhere. An exception should be made for the work of Proxorov 1992, which raises the question, whether St Cyril should not be credited with the authorship of both the Cyrillic and the Glagolitic alphabet, as does Lunt 2000. 2

490

WILLIAM R. VEDER

additions or alterations did it undergo in the course of its history?5 Whichever the solution proposed for the dilemmas and whichever the answer to the questions, it remains open to doubt. The ambiguity surely resides in the fact that hitherto only two sets of data have been taken into consideration: primary data, i.e. Glagolitic manuscripts and inscriptions, and secondary data, i.e. alphabetaries, numeraries and alphabetic verse (even if not preserved in Glagolitic but transcribed into Cyrillic). And 1 : 1 comparison always entails a 1 : 1 risk of error in interpretation. In order to reduce the ambiguity, we have to introduce a tertium comparationis, like the third point in topographical triangulation, which puts us in a position to measure not only relative distances but angles as well and thereby to collocate the three points precisely. Such tertium comparationis was identified by Šafaǥík 1858: 26, 28 6 as “cyrillische Copien aus glagolitischen Handschriften”. His identification has not as yet been given the attention it merits.7 In this study, I shall examine the tertiary data of the Cyrillic transcriptions of a translation to be dated before 885 and evacuated from Morava to Bulgaria in a single Glagolitic manuscript in 886. It is the Scete Patericon (CPG 5562 8 ), to be ascribed to St Methodius and dated between his return from his first visit to Rome to Morava in 870 and his death in 886. It was copied once in Glagolitic in Pliska (886-887), and the original was then transported to Ohrid, where it remained available for perusal until 1395, when the city was conquered by Hairuddin Pasha. Of the Pliska copy A, five Glagolitic copies ʂʃʄʅʆ were made at Preslav, which were all taken to Kiev, there to be transcribed into Cyrillic after 1036. At Ohrid, ten independent Cyrillic transcriptions were made (ockw and W124568), among them an edition to make the text more explicit (c) and a recollation with a younger Constantinopolitan Greek manuscript (k). Schematically, the stemma of the witnesses to this text can be represented as shown in Fig. 1.9 5

The last question is specifically addressed by Lunt 2000 and Trunte 2004. See the summary of his work and bibliography by Lili Laškova and Desislava Atanasova in KME IV: 524-529. 7 Isolated readings in Cyrillic texts, it is true, have been explained by reference to underlying Glagolitic readings (e.g. VM xv d=va pop¥ skoropis;cå Պ v@” pop;i, Mathiesen 1967, and westi[ msc; Պ Ÿ@” mesãc;, Rešetar 1913, O PismenexԔ 6:3 ʃ d œ Պ do ã, Marti 1983), but no attempt has as yet been made to systematically question Cyrillic witnesses of a Glagolitic text as to that latter’s graphic and orthographic makeup. 8 The original title is not preserved; in CPG, the work is ranged under Apophthegmata patrum as Collectio systematica; I use the traditional designation for the Slavic translation (for its contents and their structure, see Veder 2005: 259-284). The particulars of the history of the text are laid out in Veder 2007. 9 In the stemma, bold Latin letters indicate Cyrillic transcriptions of which the hyparchetype is not preserved; shading marks the work of Balkan scribes. For an overview of the structural, tex6

THE GLAGOLITIC BARRIER

ʂ ʃ A

ʄ ʅ ʆ

ʘ

o c k w

a = A1 a' A23 s S12 s' S3 b B1 b' B24 b" B5 m M1 m' M2 i I1 i' I2 O12 C1423 K12 W3 W7 W1 W2 W4 W5 W6 W8

491

Beograd NBS Deǁani 96 Moskva GIM Sin. 3; SPb RNB Pog. 267 Leiden UB Scal. 74; Moskva RGADA MGAMID 603/ii Saratov NBSGU 45 Moskva GIM Uvarov 483 Moskva RGB F.178 ҥ 8240, F.310 ҥ 219 L’viv LBAN ASP 56 SPb RNB Kir.-Bel. 20/1259 SPb RNB Sofijsk. 1391/iii SPb RNB Tixanov 552 Moskva RGB F.113 ҥ 601 Moskva GIM ƀud. 318A; Beograd NBS Deǁani 93 f.177-179v Moskva GIM ƀud. 18; RGB F.304 ҥ 37, 703; NBMGU 1310 SPb BAN 13.3.17, Belokrin. 2 Moskva RGB Popov 93 Leiden UB BPL 2290 Sinai MSC Slav. 33 Beograd NBS Peǀ 86 SPb RNB Hilf. 90 Wien ÖNB Slav. 152 Beograd MSPC Krka 4 Paris BN Slav. 10 Fig. 1: Stemma codicum of the Scete Patericon

Among the witnesses to the protograph ʘ, direct witnesses (W124568) can be distinguished from indirect (A1-K2 and W37), but the distinction should not be overrated: thanks to the conservatism of the MSS, especially those of East Slavic provenance, it is fully possible to retrieve conflicting graphic and orthographic features of ockw, ʂʃʄʅʆ, and even of A. Moreover, the direct witnesses can be shown not to be free of the pervasive tutelage of A: In ch. 8:16, the son of abba Poemen’s sister has been taken hostage in order to force abba Poemen to come and meet the local governor; when his sister reproaches him for refusing to do so, he answers her: Pastor filios non generavit. ȐʏʊӐ˱ʍ ʓʚʋʍʂ ʏ˝ʋ ˁʄʚʍʍʈʒʆʍ. PÖmïn= ne rodï håd=Î

tual and graphic features of the Cyrillic hyparchetypes (their classification not yet reflecting the insights of Veder 2007), see Veder 2005: 218-224, 231-242; for dates, provenance and descriptions of the MSS, see ibidem: 435-436 (NB. There, regrettably, the sigla I1 and I2, as well as W3 and W6 are switched).

492

WILLIAM R. VEDER

But the witnesses transmit the following: A oc kW8 W5 W2

p[mjn; nebr`Ÿi hãd;” (ʄ Ռ hãdo) pimin= nebre'et; hadaÎ pimin=Î nebr`zi hådaÎ nebr`zy håd;Î pimin; ne ro'd;Î

The degradation ne rodï Ռ nebre©ï (misimproved Ռ nebre'et=) stems from A’s effort to copy-edit the text, eliminating, inter alia, original nerodïtï ‘neglect’ (missed in no more than three instances) and marking the readings to be changed by marginal notes in Glagolitic. The later readers of the original considered the annotations on a par with the Glagolitic text and took care not to skip them (with varying success); only W2 disregarded them (the task of transcription per se seems to have sorely taxed his capacities). In the discussion below, I make no distinction between the two groups of witnesses, but eschew readings marked by interference from A. The text length (Cyrillic: 96255 occurrences of 13172 forms; Glagolitic: 96034 occurrences of 12992 forms) prohibits all but exemplary treatment of the conflicts signalled below. 10 For the examples, I have given preference to those that occur both in East and South Slavic witnesses and show that the variation is not dialect-related. Jotation of Vowels A: initial: predominant qko (all branches of transmission) and `ko (W5 2, W3 1) conflict with ako (a 4, B4W5 2, b" 1) and ak¥ or aki (ʂ 24, c 13, ʃ 6, ʄ 4, ʅʆ 2, oW34 1); postvocalic: moq (chiefly in a'scW168) conflicts with moa (b 31, k 18, W3 14, b' 12, C3 7, b" 5, o 4, W2 2). E: initial: \]e (A2 27, C1 24, W2 19, W8 18, W6 17, W4 15, A3 11, aW1 3, C4W5 2) conflicts with predominant e]e (all branches of transmission); postvocalic: mo\ (chiefly in a'scW168) conflicts with predominant moe (all branches of transmission). OU: initial: [n- (a'b' 33, c 22, bb" 16, o 15, ʅʆ 14, s 6, k 3, W12348 1) conflicts with predominant oun- (all branches of transmission); postvocalic: the predominant mnD-desinence -i[ (all branches of transmission) conflicts with -iu (W3 13, K2 9, aB4W2 3, K1 2, sbb"i'C23W4 1). Œ: initial: fA õ (W5 6; [ passim) conflicts with œ (chiefly in B4b"W6); postvo10 A searchable verse-by-verse collation of all the witnesses is published in Polata knigopisnaja 36 (2006) at http://kb.osu.edu/dspace/handle/1811/6399; machine-readable copies of the individual witnesses are being made available at the website of the Bulgarian Academy’s CyrilloMethodian Research Centre (http://kmnc.bas.bg).

THE GLAGOLITIC BARRIER

493

calic: svoõ (a 2, bmoW8 1; svo[ passim) conflicts with svoœ (W36 2, k 1). Å: initial: mfApl ã (a 9, W6 4, mW5 1) conflicts with predominant å (all branches of transmission); postvocalic: moã (a 14, W6 4, A3B4K2W5 1) conflicts with predominant moå (all branches of transmission). Consider also the misspellings ou'e adv + œ'e Ռ ['e (W6 14, W58 11, W4 10, 2 W 6, aB2ʅocW1 1), outr- + œtr- Ռ [tr- (W2 9, W48 5, W56 2). The distribution of the readings makes it plausible that neither ʘ nor A wrote jotated vowels, and that the task of introducing jotation-markers was assumed (with varying success) by the individual Cyrillic transcribers. Palatalisation of Consonants B V M P: forms with epenthetic l to mark palatalisation conflict with forms without it, e.g. zoblõ zobli (all branches of transmission) conflict with zobœ zobi (W3 2); predominant korabli korabl; korabl;nici korablå (all branches of transmission) conflict with korabi korab; korab;nici korab` (W2 4, W56 2, kW38 1); blgsvlen- (chiefly in b'W2) conflicts with predominant blgsven- (all branches of transmission); tivlqnin; (W2 4, W4 1) conflicts with predominant ƒiv`nin= (all branches of transmission); ostavl; ostavlõ (all branches of transmission) conflict with ostav; ostavå (W3 2, W456 1); predominant zemli zemlq zemlå (all branches of transmission) conflict with zemi zem` zemå (W5 10, W6 3, abC3 1); zeml;n- (W4 5, W8 3, bW12 2, ʆcW6 1) conflicts with predominant zem;n- (all branches of transmission); comp kr`plii kr`pl;wa kr`pl;wi (all branches of transmission) conflict with kr`pi kr`pwa kr`p;ci (W3 2, a'c 1); predominant s=pl[ s¥pl[ (all branches of transmission) conflicts with spå s¥på (k 4, W36 3, W25 2, AockW4 1). Consider also the variable relation of the two roots -im»;m: -eml- (ʂoW8 2:1, ʄʆcW124 1.5:1, sk 1:1, W356 1:1.5, ʅ 1:2). L N R: forms with jotated vowels (cf. above) conflict with forms without them: e.g. xvalqx= (b 2; xvalåx=/xval`x; passim) conflicts with xvalax; (W2); bol\ (W48 5, W2 4) conflicts with predominant bole (all branches of transmission); adj.mD pav;l[ (B2oW568) conflicts with pav;lou (ʂbB4b"kW2); klanq[ (W48 3, a'bmC2W126 1; klan`œ/klanå[ passim) conflicts with klana[ (W1); the desinence -n\ni\ (W24 5, W8 4, a 1) conflicts with predominant -neni\; Ldu kon[ conflicts with konou (sW123); csrq (W4; csrå/csr` passim) conflicts with csra (W28); mor\ (W8 2, W26 1) conflicts with predominant more (all branches of transmission); r\k/w; r\]i r\he (a 2, W23 1) conflict with predominant rek=w; re]i rehe (all branches of transmission); predominant tvor[ conflicts with tvorœ/tvorou (W2 35, W8 23, W4 21, W1 12, b" 6, k 5, a 4, W36 1). Consider also the variable relation of the two roots -louh- and -kl[h-, e.g. in -louhi så : -kl[hi så (ʂ 2:1, ʄ 1:2, ʆ W4 1:4, sʅW568 1:8, ockW123 1:20). The distribution of the readings makes it plausible that neither ʘ nor A marked consonants for palatalisation before full vowels and that the task of in-

494

WILLIAM R. VEDER

troducing such marking was assumed (with varying success) by the individual Cyrillic transcribers (perhaps to some extent preceded by ʂʃʄʅʆ). But the survey of palatalisation is incomplete without an examination of the two half-vowel signs = and ;, for which, by way of exception, I turn to the South Slavic witnesses alone (the East Slavic witnesses with their longer transmission history have leveled the distribution of the half-vowel signs, albeit they show a high incidence of etymologically incorrect ;-spellings). These witnesses usually prefer ; as a half-vowel sign, but none is without attestations of =: W2 has 2 (both etymologically incorrect), W8 10 correct + 2 incorrect, W1 13 correct, W4 21 correct, W3 25 correct + 6 incorrect, W5 45 correct + 17 incorrect, K1 651 correct + 166 incorrect (the witness is only fragmentarily preserved), K2 4124 correct + 3252 incorrect, and W6 6447 correct + 2921 incorrect; the locations of = in W1234 do not coincide, in W58 only in the prefixes v=(6) and s=- (1). The discrepancies in the use of = and ; in the individual witnesses make it plausible that neither ʘ nor A distinguished the two half-vowels, but that the distinction was introduced during the later phases of transmission (by a secondarily differentiated =/=), each transmitter having to decide for himself how to handle it. A vivid example of the difficulty of deciding which half-vowel to use is provided by the name of St Arsenius’ disciple Zoel (15:10b), for which we find both zoil= zoilou/zoilovi and zoil; zoil[/zoilevi (ocW3 4:0, a' 2:2, aʆk 1:3, ʄW24 0:4). "D and } In contrast to etymological ', the reflex of *Dž/٥ ('d) is a source of conflicts among the witnesses: compare qko'e (with stable ') and tako'de (chiefly W468) : tako'e (chiefly a'b'oc) : tako (chiefly ʆ). Beside ' or ø, we find: x x

x x x x 11

'g (7) e.g. d='dou bB4b"kW3568 : d='d[ moc : d='gu ʂ11 : d='g[ B2 d (27) e.g. -xo'daaxœ b"kW3 : -xo'axu ʂB4ʆ : -xodåxou bB2m : -sp`axœ ocW568; stra'd;ba b"kW5 : stra';ba ʂbb'oW268 : stradba W3; ougo'deni\ abkW3 : ougo'enie a'B2b"oc : ougodi\ B4W1; œro'daawe ʂo : ouro'awe ʄʅc : årod`we W3 : ø k zg (4) e.g. i'den- AokW56 : izgon- B24 : izg=n- cW3; ra'd;'et= aA3B4kW24 : ra';'et; A2sbB2ʅc : ra'det; ʆ : razgorit; o zd (1) ra'daã : razdaqx ʆc : razdavaå ʂʄokW1368 z (1) ra'da\t= A2 : ra'a\t; A3B4c : razdaet; b : razaet; B2 t (1) poxo'd; b"kW2568 : poxo'; aA3sbb'ʆ : poxot; A2 : poxodiv= oc

As the oldest witness of ʂ, A1 (late 12th c.), is of South Rusian provenance, 'g there cannot be explained as a North Rusian dialectism.

THE GLAGOLITIC BARRIER

x

495

] (1) sœ'd; aA2bB4b"W24568 : sou'; A3B2 : sou]e W1 : soudiv= o : soudix= c : ø ʆ.

Considering the predominance of monographic renderings, I presume the source of this variation to be a monograph in ʘ (copied by A, save in one instance), most probably the letter ˜ (pace Marti 2004a: 409-410). The reflex of *ǀ/֦ (]) shows no conflicts of interpretation, save 4 wt (W4 2, 2 C W5 1).12 OU and Ù The digraph ou varies not only with [ (cf. above), but also with the monographs u (ʂ 25-77%, ʆ 12-65%) and ö (ʄ 12-63%, o 71%, c 4-79%, k 51-73%, W5 2%, W8 30%), as well as with e, i and o: e.g. i moliwi Ռ oumoliwi i, neim`ni\ Ռ neoum`nj¡e i', s`då]e Ռ s`då]ou a, sœdi]i Ռ soudi]ou okW48, qdœ]ou Ռ qdou]e aokW35 : qdou]i B2m. This makes it plausible that ʘ and A (possibly followed by ʂʃʄʅʆ) wrote a monograph, most probably ¥ (cf. Miklas 2000: 124, Marti 2000: 65). The variation indicated above contrasts with another type of variation, restricted to loanwords from Greek, viz. i Ú ou : Ö (marginally including also o, y, [»õ, ø): e.g. egip;t- : egoup;t- : egÖp;t- (all b"iow : no i a'b'k : no ou aW2568 : no i or ou scW14 : no ou or Ö m' : no Ö bmi'), møisi : møusi/møÖsi : møsi (all b"ʆok : no i m : no ou ʂcwW56 : no ou or ø sb'W1248 : no ø b), pimin= : poumin= : pÖmin=13 (all o : no i b' : no i or ou c : no ou ʂsb"imW12456 : no ou or Ö m'wW8 : no Ö bi'k), sin=klit- : soun=klit- : sÖn=klit- (all bb"ok : no i i' : no i or ou asb'icW14 : no ou mwW2568). These conflicting readings reflect a monograph different from ¥, most probably [ (which the translator did not use to render jotated *ju, cf. above). F and Ï In the names ƒeodør= fer=m;sk¥i and ƒeofil= (16 occurrences) differentiated spellings of *ph and *th 14 conflict with non-differentiated spellings as follows: only undifferentiated b"ʅc : 1-2 differentiated ʂso : half of each b'bk : 1-3 undifferentiated ʆwW2456 : only differentiated W18. Differentiated spellings become more numerous in the second half of the text and their correctness (by Greek standards) rises from 0-30% (a-i') to almost 100% (W1-8). These conflicts, as well as what may be interpreted as a learning curve of the transcribers, make it plau12

A3 confuses with ] Glagolitic b in the heading of chapter 11: bodr¥ Ѝ ]odru. In the heading of chapter L, A wrote pojmen (reflected in a'bb' as poimen), which is surely not original. 14 Note that I consider both ƒeofil= and feoƒil= as differentiated spellings. 13

496

WILLIAM R. VEDER

sible that that neither ʘ nor A (nor ʂʃʄʅʆ) differentiated the spellings, and that the task of differentiating *ph and *th was assumed (with varying success) by the individual Cyrillic transcribers. The misspellings ƒol= Ռ pod= (W2 3 : pomost= AocW3 3) and pÖmin= Ռ fimin; (W1 17, W4 9), i.e. the confusion of Glagolitic f with p, as well as ƒivaidՌ divaid- (a') : tivaid- (W2 4, oW68 3, a'W5 2, aʆcW134 1), i.e. confusion with d and t, make it plausible that the undifferentiated character used in ʘ (and retained in A and its Glagolitic copies) was f. X and ˚ The confusion g@; Վ x@= (g@; Վ x@;), attested in all branches of transmission proves that xristos= was not written with ˚ (pace Miklas 2003: 181); xl=m=, in which ˚ is attested (Schaeken and Birnbaum 1999: 80-81), is lacking in the text, but the substitution x Ռ m in groxot= (ʂsbB4b"ʆc) Ռ xromot= (B2okW2368), x¥tr= (ʂsbB4b"ʆc) Ռ mœdr= (B2okW2368), and the confusion of desinences GLpl -x= Վ Dpl -m=/Ipl -mi (e.g. pri dv;r;x= AokW2 Ռ pr`d= dvermi cW3, neh;stiv¥ix= AockW134 Ռ neh[v;stv;nim/ W2, and inversely m=nog¥m= ʂsbB4W14 Ռ mnogyx= B2ckW3568 : m=nogy ʆ, gr`w;n¥im= AocW3 Ռ gr`w/nyx; W4568, gl@em¥im= AockW2568 Ռ gl@emyx= W3, st@im; tainam; W2 Ռ st@yx= tain= ʂʄʆokW345678, v=s`m; k Ռ vs`x= a'moc : ø aʄ, pœtem; x@v`m; b'b" ʆockW23468 Ռ pout;x= x@vx= ʂsb) might be interpreted to indicate that the protograph used ˚ in these instances.15 Á and Œ Conflicting readings with a : i : =»; : ` : ¥»y : œ»ou : å»e abound in the place of jery. The variation, most prominent in the first half of the text, affects chiefly desinences, which can significantly alter the text: e.g. 'en¥ Ռ 'enœ (akW2) : 'en` (W2) and 'en= Ռ 'en¥ (akW8); 'iv¥ Ռ 'ivi (sʄʅʆockW36) : 'ivå (W68); id¥ Ռ idi (W268) : id; (W5) : iti (W3); pol;©i Ռ pol;z¥ (ʂbb'ʆckW38) and pol;©å Ռ pol;z¥ (A 5, ok 4, sW1 3, W3 2, W2 1) : pol;zœ (ʄocW6); s¥ Ռ si (aW4 2, oW26 1); s= pr`d¥ Ռ s= pr`da (aA2) : s= pr`d` (A3) : s= pr`dou (bb"); t;r¥ Ռ tra (bB4) : tre (W1) : tri (ʂ); t¥ Ռ to (a') : t= (a); qd¥ Ռ qda (s 3) : qd= (ʂ) and qd= Ռ qd¥ (A). The variation indicates that ʘ in these instances used a monograph, and the reflexes i and ` may be taken to indicate that it bore resemblance to both i and `, i.e. most likely ÿ (pace Marti 2000: 65, Trunte 2004: 424). The massive confusion œ Վ ou in all branches of the tradition, e.g. bœdœ 15

If this can be proven, it would constitute an additional trigger for the vacillation between adnominal dative and genitive in the witnesses.

THE GLAGOLITIC BARRIER

497

(kW356 15) : predominant boudou (all branches of transmission), and the confusion œ Վ å (which appears already in A), e.g. blœd- Ռ blåd- (AW4) and blådՌ blœd- (W28 2, ʂW3456 1), s=mœt- Ռ s=måt- (C12W6) and s=måt- Ռ s=mœt- (W3 8, k 6, C3W8 3, ʅ 2, a'cW16 1), can be satisfactorily explained only if ʘ wrote a letter similar to both ¥ and ã; the reading ougotovlõ Ռ ögotovl; (W468 : ougotova[ A) shows that it was similar to ; as well. The original letter may have looked like ñ (ÿ, so Trunte 2004: 424, lacks the prerequisite similarity) and must have been abandoned, because it was opaque.16 Almost all examples in this survey could, of course, be considered individually and many of them could be explained as innovations (e.g. the forms of the imperfect and participles adduced for 'd). But such explanation, which fails to answer the crucial question: What triggered the change? 17 is, in fact, nothing more than the rationalisation of an observation. More importantly, it slights the possibility that the variation be patterned. It is, indeed, only when we regard the marking of jotation and palatalisation, the distinction of *Ǯ and *ǒ, *ph and *th, and the digraphic renderings of *Dž/٥, *u, and *y as add-ons,18 that we begin to understand that, in this text, there was a single principal trigger for the variation and the resulting discord of the witnesses: the inexplicit spelling of the Glagolitic protograph ʘ. This led the transcribers (and in some cases the Glagolitic copyists as well) to more frequent and more varied interventions than faithfulness to the text would allow. An example of these additional interventions to explicate the spelling is furnished by the name of abba Longinus (ȋʏʄʄ̢ʍʏʩ; 10 occurrences), which ʘ surely spelled loggjn= with double g: this spelling is retained only in b" (log/gin- with a hypercorrect pajerok) and k (K1 loggÖn-, K2 adding n and a hypercorrect ; lon;ggÖn-); ʅʆc undouble the g (login-), as do W1234 in half of the cases, in the other half joining W568 in substituting Slavic ng, as well as adding a hypercorrect ; (lon;gin-); ʂbb'o have an underlying spelling *lœggin-19 (a lœgin-, a' lugin-, o 16

The survey lacks data as to the distribution of o (o) and ø (ø), as well as i (j) and j (i; for the reversal of the traditional correspondences of the latter two letters, see Miklas 2003); I must leave it to a colleague with more insight and patience than I can muster to sort them out. 17 The traditional answer would be: “Language changes, and the scribes parsed the text through their language competence”. In the perspective of the data of this particular text and its tradition, it begs the counter-question: “Why then are the most recent transcriptions of the text (w and W1-8) the least innovative?”. 18 W6 exhibits a curious individual add-on to å, e.g. in hådo, regularly written håndo with a mater lectionis n, carefully erased in final editing. The concept of add-on in itself could do away with such concepts as e.g. Vertauschung der beiden Halbvokale and Aufgabe der Palatalität (Diels 1932: 96-98 and 141-146), cf. the data above on their representation in the South Slavic witnesses. 19 If A retained ʘ’s monographic œ (ñ?), the b" spelling log/gin- could be derived, just as the spellings of bb', from *lœggin-, which would then be the reading both of ʂʄ and o.

498

WILLIAM R. VEDER

loun;gin-, bb' vacillating between loun;gin-, lon;gin- and login-), which is surely to be traced to A’s copy-editing. Need it be pointed out that the development of all of the translator’s Grecisms (including aggl- Ռ an=gel- and the like) can be explained in similar terms? The terms of the development are intimately related to Nicolaas van Wijk’s concept de l’unité à la puralité. But unity, to be sought among Slavs well before Sts Cyril and Methodius’ mission to Morava, was present in the Glagolitic protograph ʘ of the Scete Patericon only as an intention (‘for all Slavs’); the means applied to realise it was parsimony, reducing the spelling to the bare essentials in order to avoid marking any feature that might be divisive. This intention and the means to realise it were evidently the first casualties of the evacuation of the book to the Balkans: there, a notation of the type zema (zema) was not recognised as being capable of representing both zemq and zemlq, but only as not representing either. The resulting tampering with the spelling (starting, to an extent as yet to be precisely determined, as early as the Pliska copy A of 886887) inevitably led to the plurality or, to put it more bluntly, prodigality of the witnesses. So far, this attempt to follow up on Šafaǥík 1858 and to question the Cyrillic witnesses to the Scete Patericon as to the evidence they can provide to the spelling of their Glagolitic source remains isolated and therefore vulnerable to criticism, especially since its findings are at variance with established scholarship. Yet there is a straightforward way to check their validity in the one text attributed expressis verbis to St Methodius, viz. the Nomocanon (VM xv). After the thorough work of Maksimoviǁ 2004, 20 who for the first time provides a coherent analysis of the lexicon and morphosyntax of the translator’s text, it should not be too difficult to collect the witnesses (including excerpts transmitted separately from the full text) and question them about the orthography of their preBalkan protograph. In conclusion, it appears that the Glagolitic barrier, which impedes our view of pre-Balkan texts and their spelling, is more than just the aggregate of problems referred to in the second paragraph of this paper. It consists of four conditions, three historical and one man-made: A the lack of any extant Glagolitic writing that can be attributed to either the first generation of Slavic men of letters (mission to Morava) or their second and third generation (reception in the Balkans); B the probable shift in spelling from parsimony to prodigality, which may have started already in the second generation, and of which all extant Glagolitic (and Cyrillic) writing bears the traces; C the retroactive influence of Glagolitic copy20

The criticism levelled against Maksimoviǁ by Stankov 2006ab in no way affects the essence of his study, but only broadens our insight into the controversial aspects of the term ‘moravism’ as a label for lexical items.

THE GLAGOLITIC BARRIER

499

editing upon pre-Balkan Glagolitic texts, readers making little or no distinction between the original and subsequent accretions; and, finally, D the neglect by established scholarship of the genealogical relations between witnesses and the resulting failure to probe beyond the extant witnesses into the realm of textual palæontology. 21 The first three conditions are irreparable, but the last can very well be remedied by the close examination of Cyrillic witnesses to texts that have a Glagolitic protograph, as Pavel Josef Šafaǥík advocated a century and a half ago. Deerfield, IL REFERENCES CPG 1974-87/1998 Clavis patrum græcorum, I-V. Maurits Geerard (ed.). Turnhout 1974-1987; Supplementum. Jacques Noret (ed.). Turnhout 1998. Diels, Paul 1932 Altkirchenslavische Grammatik, 1. Teil. Heidelberg. Dürrigl, Marija-Ana, Milan Mihaljeviǀ, and Franjo Velǁiǀ (eds.) 2004 Glagoljica i hrvatski glagolizam. Zagreb-Krk. Ivanova, Tat’jana A. 2004 “Glagolica: Novye gipotezi (neskol’ko kritiǁeskix zameǁanij po povodu novyx issledovanij o pervoj slavjanskoj azbuke)”. Trudy Otdela drevnerusskoj literatury 56, 78-93. KME 1985-2004 Kirilo-Metodievska enciklopedija, I-IV. PetƼr Dinekov (ed.). Sofia. Lunt, Horace G. 2000 “Thoughts, Suggestions, and Questions about the Earliest Slavic Writing Systems”. Wiener slavistisches Jahrbuch 46, 271-286. Maksimoviǁ, Kirill A. 2004 ZakonԔ soudӸnyi ljudӸmԔ. Istoǁnikovedǁeskie i lingvistiǁeskie aspekty issledovanija slavjanskogo juridiǁeskogo pamjatnika. Moskva. Marti, Roland W. 1983 “Textologische Probleme der Apologie Chrabrs”. Anzeiger für slavische Philologie 14, 117-147. 2000 “Die Bezeichnung der Vokale in der Glagolica”. In: Miklas et al., 54-76. 2004a “Die Bezeichnung der Konsonanten in der Glagolica”. In: Dürrigl et al., 401417. 2004b “Aktuelle Probleme des altkirchenslavischen Schrift- und Lautsystems”. Incontri linguistici 27, 11-37. 21 Since the first publication of the cod. Clozianus by Jernej Kopitar (1836), almost seven generations of slavists have gone by neglecting to establish its proper genealogical relation to the cod. Suprasliensis (remedied by Spasova and Veder, forthc.). For the term textual palæontology, see Veder 2005: 42-45.

500

WILLIAM R. VEDER

Mathiesen, Robert 1967 “An Emendation to the Vita Methodii XV.1”. Zbornik za filologiju i lingvistiku 10, 51-53. Miklas, Heinz 2000 “Zur editorischen Vorbereitung des sog. Missale Sinaiticum (Sin. slav. 5/N)”. In: Miklas et al., 117-129. 2003 “Jesus-Abbreviatur und Verwandtes: Zu einigen Rätseln der glagolitischen Schriftentwicklung am Material der Azbuǁnaja molitva”. In: W. Honselaar et al. (eds.), Time flies. Festschrift for William Veder (Pegasus Oost-Europese Studies 2), 171-204. Amsterdam. Miklas, Heinz, Sylvia Richter, and Velizar Sadovski (eds.) 2000 Glagolitica. Zum Ursprung der slavischen Schriftkultur (ÖAW, Philosophischhistorische Klasse, Schriften der Balkan-Kommission, Philologische Abteilung 41). Wien. Proxorov, Gelian M. 1992 “Glagolica sredi missionerskix azbuk”. Trudy Otdela drevnerusskoj literatury 45, 178-199. Rešetar, Milan 1913 “Zur Übersetzertätigkeit Methods”. Archiv für slavische Philologie 34, 234-239. Šafaǥík, Pavel J. 1858 Über den Ursprung und die Heimat des Glagolitismus. Praha. Schaeken, Jos, and Henrik Birnbaum 1999 Die altkirchenslavische Schriftkultur. Geschichte – Laute und Schriftzeichen – Sprachdenkmäler (mit Textproben, Glossar und Flexionsmustern) (Altkirchenslavische Studien II, Slavistische Beiträge 382). München. Spasova, Marija, and William R. Veder forthc. “Copying, Copy-Editing, Editing and Recollating Three Chrysostomian Lenten Homilies in Slavonic”. Polata knigopisnaja 38 (Preliminary Bulgarian version: “Prepisvane, popravjane, redaktirane i sverka na slavjanskija prevod na tri Zlatoustovi Velikopostni slova”. Preslavska knižovna škola 9, 2006, 53-107.) Stankov, Rostislav 2006a “O leksiǁeskix moravizmax v drevnix slavjanskix rukopisjax”. Preslavska knižovna škola 9, 29-52. 2006b “O leksiǁeskix moravizmax v drevnix slavjanskix rukopisjax (2)”. In: A. Davidov et al. (eds.), BƼlgarska filologiǁna medievistika. Nauǁni izsledvanija v ǁest na prof. dfn Ivan Haralampiev, 261-287. Veliko TƼrnovo. Trunte, Nikolaos 2004 “Zu Reformen in der glagolitischen Schrift”. In: Dürrigl et al., 419-434. Veder, William R. 2005 Hiljada godini kato edin den. ŽivotƼt na tekstovete v Pravoslavnoto slavjanstvo. Sofia. 2007 “Metodievijat prevod na Skitskija Paterik (Metodievata zla hiena)”. KiriloMetodievski studii 17, 783-793.

THE GLAGOLITIC BARRIER

501

SUMMARY The study of Glagolitic, the oldest Slavic alphabet, draws upon two sources: primary (manuscripts and inscriptions) and secondary (alphabetaries, alphabetic verse and numeraries). Šafaǥík 1858 proposed to include tertiary sources: Cyrillic transcriptions from Glagolitic originals. The paper follows his suggestion and examines the spelling conflicts in the tradition of the Scete Patericon, evacuated from Morava to Bulgaria in 886. It shows that the underlying spelling of the protograph featured a parsimony not preserved in any extant Glagolitic manuscript. It proposes to verify the results on the single text unambiguously ascribed to St Methodius, the Nomocanon (Maksimoviǁ 2004).

Dutch Contributions to the Fourteenth International Congress of Slavists, Ohrid: Linguistics (SSGL 34). Amsterdam – New York: Rodopi, 2008, 503-571.

PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION WILLEM VERMEER

1. Preliminaries1 1.1 The purpose of this contribution As is too well known to bear recounting, the velar obstruents of early ProtoSlavic (*k, *g, *x) have two types of modified reflexes in palatal contexts: 1. ǁ, ž, š. 2. Attested c, dz/z, s/š reflecting earlier *ǀ, *dǺ, *ǧ. Both types are triggered by following front vowels of various origins, the second also by preceding front vowels under conditions that have proved difficult to pin down exactly. In what follows, the two kinds of regressive palatalization will be referred to as “Reg1” and “Reg2”, the single progressive one as “Prog”. Of the problems raised by Prog, it is relative chronology that investigators have differed most persistently about, some holding that Prog was simultaneous with Reg2, others that it was later, or earlier, or even that it preceded Reg1. I can’t help finding the lack of consensus puzzling, and the inordinate length of time it has been with us even more so. Surely it is unlikely that all four possibilities are equally defensible. What has been going on here? It is the purpose of this contribution to investigate the chronology that was first formulated by Holger Pedersen in 1905 and which puts Prog earlier than Reg2. In order to grasp the implications of the various kinds of evidence that have been thought at one time or other to support Pedersen’s chronology, just about every detail with some relevance to Prog has to be taken into account. That explains the awkward length of this contribution. I am sorry about that, but there 1 I am indebted to Carl Ebeling and Peter Schrijver for their criticism of an early version of this contribution, which showed me that it was not written clearly enough to fulfil its purpose. The revision that followed took much time and effort and owes a lot to the moral support of Trille Bedarrides, Coos ’t Hoen, and Dorinde van Oort. In the final stage, critical comments by Peter Houtzagers and Frits Kortlandt led to significant improvements.

504

WILLEM VERMEER

is no alternative given the way the investigation of Prog has been plagued by misunderstandings caused largely by lack of explicitness on the part of scholars, coupled with a near-total reluctance to discuss each others’ findings. 1.2 Reg2 and Prog as a single innovation 1: historical status Before turning to Pedersen’s chronology, we need to take a brief look at the classic view that identifies Reg2 and Prog as jointly making up a single bidirectional palatalization rule. Until the early years of the twentieth century, a consensus existed that Reg2 and Prog were a single event.2 However, after Vatroslav Jagiǀ first floated the idea that Prog and Reg2 might be chronologically distinct (1901: 128), the older conception started losing visibility and by the early twenties it was hanging on by its fingernails in one or two handbooks, e.g. Lehr-Sp’awiǝski (1923: 31). The last to defend it explicitly was Kazimierz Nitsch (1926: 52). Then it went under for more than two decades. Even Meillet, who had been a staunch advocate of the classic view for decades, eventually lost faith in it. Whereas the first edition of Le slave commun presents Prog as “un cas nouveau de la seconde palatalisation” (Meillet 1924: 82), the second limits itself to observing that Reg2 and Prog have the same reflexes (Meillet 1934: 90). It is regrettable that he nowhere explains why this change of`heart occurred. If the possibility of Prog and Reg2 together making up a single event is discounted, the chronological problem is reduced to a dichotomy: Prog must be either earlier or later than Reg2. Accordingly evidence contradicting the one will tend to be perceived wrongly as positive proof for the other. This must be kept in mind when reading post-1920 publications on Prog, most of which appear to have been written in ignorance of the classic view. Quite a bit of the argumentation that has been adduced in support of Jagiǀ’s chronology, for instance, actually consists of evidence against the Pedersen view, which then is perceived – wrongly – as proof that Prog must have been later than Reg2 (Vermeer 2006: 233-234). Since the Second World War the classic view has been staging a modest comeback, beginning with Vaillant (1950: 55) and continuing in the work of such authors as Furdal (1961: 21), Besedina-Nevzorova (1962: 98), Birnbaum (1968: 15; 1970: 109), Andersen (1969: 554n.; 1998: 588), Kortlandt (1979: 266; 1982: 184), and Townsend & Janda (1996: 84).

2

See, e.g., Leskien (1871: 21-24), Miklosich (1878: 257), Sobolevskij (1889: 32), Baudouin de Courtenay (1894: 49-50), Lorentz (1904: 266), Vondrák (1906: 266), Meillet (1910: 371-372).

PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION

505

1.3 Reg2 and Prog as a single innovation 2: methodological status If one scours the literature for reasons why scholars abandoned the view of Reg2 and Prog as a single change, one finds that it has nowhere been critically evaluated. True, one or two investigators appear to have believed that progressive and regressive palatalization rules cannot coexist within a single system, most explicitly Lunt (1985: 155 = 1987: 258n). However, bidirectional palatalization rules are quite common in the languages of the world, beginning with dialectal Russian right under the nose of anybody who consults Baudouin de Courtenay’s classic article on Prog (1894: 47), so in the present state of the evidence, the belief in the impossibility of bidirectional palatalization rules cannot be used to impugn the idea of Reg2 and Prog as a single change.3 The point has important consequences for the way the problem of the chronology of Prog is to be approached. Since velars modified by Reg2 and Prog have identical reflexes, Ockham’s razor as traditionally wielded in historical phonology demands that the two phenomena be identified as long as separating them is not shown to offer explanatory advantages. By ordinary methodological standards, identifying Prog and Reg2 therefore constitutes the default solution, with respect to which more complex scenarios will have to be explicitly justified. Since that is the way historical phonology has always worked, there is no reason for it to work differently in this case.4 1.4 Proto-Slavic phonological developments relevant to Prog The debate about Pedersen’s chronology is not just about the palatalizations. Other Proto-Slavic innovations have tended to get drawn into it too, most prominently: 1. Umlaut, i.e. fronting of back vowels preceded by palatal consonants, notably (a) *j (including – depending on one’s precise view of contemporary ProtoSlavic phonology – consonants reflecting earlier clusters ending in *j), (b) the reflexes of velars modified by Prog. As is well known, the most spectacular outcome of Umlaut is the existence of inflectional subtypes traditionally known as “soft” – i.e. displaying the effects of Umlaut in the endings – alongside subtypes called “hard” with endings showing the outcomes of phonological developments found in the absence of a preceding palatal consonant. Umlaut as well as the morphological complexities it produced have 3

For some other languages that have been described as having bidirectional palatalization rules see Vermeer (2002-03: 387, note 21). 4 To the best of my knowledge the only participants in the debate to have expressed this basic point in so many words are Kazimierz Nitsch (1926: 52) and, less forcefully, Charles Bidwell (1961: 109-110). See also Vermeer (2000: 8; 2002-03: 286-287).

506

WILLEM VERMEER

cropped up time and again in the investigation of Prog. For example, the most sophisticated pieces of evidence that have been thought to support Pedersen’s chronology (Pedersen’s and Trubetzkoy’s) both involve Umlaut, as does the most important piece of counterevidence (Meillet’s). 2. Mono, i.e. the monophthongization of inherited diphthongs, which causes – speaking in traditional terms – *ei, *oi, *ou and *eu to end up as, respectively, i, LJ, u and *ju, at least in most positions. Quite a bit of what follows will be concerned with the monophthongization of *oi and *ei, as well as with the chronological position of Prog with respect to Mono. 3. The uncoupling of long and short vowels. Early Slavic as traditionally reconstructed inherited a series of four pairs of long and short vowels. In attested Slavic these pairs have been uncoupled, with vowel timbre replacing length, e.g. *e/lj > *e/LJ.5 4. The delabialization that converted *u and *ǰ into attested Ԕ and y. 1.5 Chronological assumptions shared by all participants in the debate Two important chronological insights are uncontested, providing a skeleton chronology all interested scholars are agreed upon: 1. The rise of Umlaut is universally put before Mono because the reflex of umlauted *ՌoiՌ is ՌiՌ. That fact can be explained by the assumption that Umlaut kicked in when *ՌoiՌ still was at the stage *ՌoiՌ, changing the first element of the diphthong into e, so that it became *ՌeiՌ, which subsequently developed into ՌiՌ like any other *ՌeiՌ. If, on the other hand, it is assumed that Umlaut took place after *ՌoiՌ had been monophthongized to a front vowel (attested LJ), the fact that it was subject to Umlaut remains unaccounted for. 2. Reg1 and Reg2 are near-universally agreed to be separated by the monophthongization of *oi for reasons that do not need spelling out. It follows that all attempts to deal with the relative chronology of Prog have in common the sequence of events shown in Table 1. As will become clear later, Umlaut and Mono are complex phenomena subject to more than a single interpretation, with subtle chronological consequences (see section 3.2 on Umlaut and sections 7.3-10 on Mono). However, the

5

Where others reconstruct long vowels, Frederik Kortlandt sees a distinction between long vowels proper and laryngealized vowels reflecting sequences with laryngeals or the outcome of Winter’s law. The clearest exposition of the reasons for operating with the distinction is, I think, Kortlandt (1985, notably pp. 122-123). As far as I can see, the difference between Kortlandt’s conception and the tradition is not important from the point of view of the issues addressed in the present contribution.

PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION

507

complications do not seriously diminish the value of the skeleton chronology as given here and the considerations it is based on. Table 1. Uncontested skeleton chronology 1 Reg1/Rise of Umlaut (depending on which was earlier). 2 Mono. 3 Reg2. 1.6 Transcription For the sake of convenience, examples will usually be given in the traditional transcription, i.e. one reflecting the state of the language actually attested in Old Church Slavonic, or rather normalized Old Church Slavonic, with the important restriction that the reflexes of velars modified by Prog will be transcribed as *ǀ, *dǺ, *ǧ. For that reason, examples containing these consonants will be marked with asterisks, e.g. *otӸǀӸ ‘father’, *stӸdǺa ‘path’, *vӸǧӸ ‘all, whole’. Note however that all scenarios that will be discussed put Prog at a significantly earlier stage than the one implied by the traditional transcription, notably before the uncoupling of long and short vowels and the delabialization of *u and *ǰ (for which see section 1.4). At that stage, *o was [a]-like, *Ӹ was [i]-like etcetera. All this is too familiar to bother listing in detail, but what should be retained is that the reader will be expected – in accordance with tradition – to mentally transpose examples back into whatever is the required shape at the stage that is being discussed. Transcriptions expressing the phonology of earlier stages (e.g. Asg *atiku) will not be used unless absolutely necessary. 2. The factual material It was only after the study of Prog had come of age, that scholars became interested in delimiting the relevant material. The earliest attempt at an exhaustive listing of the positives is by Knut Knutsson (1926: 48-66 and passim). Unfortunately, by not distinguishing between canonical OCS and more recent material Knutsson deprives the inventory of much of its potential value. This is not a personal quirk. Ever since Baudouin de Courtenay, scholars discussing Prog have tended to confuse the issues by mixing items of doubtful relevance among the genuinely important material. Horace Lunt’s criticism of this practice, harshly worded though it is, is amply justified (1981: 9, 25, 57). In what follows I shall try to give a general impression of the relevant material. In my view, however, the time is not yet ripe for anything approaching a definitive inventory, notably in the case of the counterevidence, which so far has received scant attention.

508

WILLEM VERMEER

2.1 Positives: velars modified as in Reg2 (*ǀ, *dǺ, *ǧ) in examples that cannot be due to Reg2 The first to present more or less accurate lists of the positives attested in OCS were Van Wijk (1931: 67-68) and Diels (1932: 133-134). The most credible noncanonical examples are included by Arumaa (1976: 32-33).6 Canonical OCS provides the following examples of c, dz/z, s (< earlier *ǀ, *dǺ, *ǧ) that cannot be attributed to Reg2: 1. 2. 3. 4. 5. 6. 7. 8.

9. 10. 11. 12. 13.

6

Njdza ‘illness’; polӸdza ‘usefulness, benefit’; stӸdza ‘path, road’; suff. Ռica e.g. tӸmӸnica ‘prison’ (< tӸmӸnԔ ‘dark’); suff. ՌӸca, e.g. dvӸrӸcNj ‘door, window’ (< dvӸrӸ, usually plural, ‘door, gate’); mLJsNjcӸ ‘moon, month’; zaNjcӸ ‘hare’; although this is based on a single attestation (Dpl zajNjcemԔ, Ps Sin, 103, 18), the word is attested all over Slavic and obviously old7; pӸcӸlԔ ‘pitch’; this item is relatively common in a single OCS manuscript (Supr), but all the rest of Slavic has unmodified k, even the only other OCS manuscript in which the word occurs (the newly discovered part of Ps Sin 27a/17, Mareš, Fetková et al. 1997); suff. ՌӸcӸ, e.g. kupӸcӸ ‘tradesman’ (< kupiti ‘buy’); lice ‘face’; suff. ՌӸce, e.g. ǁNjdӸce ‘child’ (< ǁNjdo ‘child’); adj. nicӸ ‘facing downward’; pron. vӸsӸ ‘all’ and derivations from the same stem, e.g. vӸsjakԔ, vӸsLJkԔ ‘every’, both rendering */vҧǧakӆ/, with the /ǧ/ betraying its origin as the reflex of *x modified by Prog, as in such case forms of vӸsӸ as Nsgf & NApln vӸsLJ, vӸsja rendering */vҧǧa/, or Asgf vӸsj՛ rendering */vҧǧӚ/. West Slavic has /š/ as in Cz všՌ, P wszՌ. These reflexes imply the former existence of a Proto-Slavic stem-final *ՌxՌ which was subject to Prog (Lorentz 1904: 265). Some scholars, notably Machek (1954: 224-225, supported by Lunt 1981: 3537), have objected to the reconstruction *vӸxՌ because it separates the BaltoSlavic word from Indo-Iranian forms pointing to PIE *vi֦Ռ. However, PIE Ռ֦Ռ would never have yielded the palatal ՌǧՌ attested in OCS. Nor can it account for West Slavic ՌšՌ, unless one would go so far as to assume that suffixal *ՌjՌ happened to be added to the stem precisely in those Slavic dialects where the reflex of the sequence *sj cannot be distinguished from that of *x

Lunt unaccountably leaves out some crucial items (notably nicӸ), but makes important points about a number of individual examples (1981: 15-25 and passim). 7 Lunt regards zaNjcӸ as “not canonical OCS” (1981: 22), but does not say why.

PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION

14. 15.

16. 17.

509

modified by Prog since they both yield š. As long as the attested reflexes have not been explained plausibly on the basis of *vi֦Ռ, we are forced to reconstruct *vӸxՌ whether we like it or not (Meillet 1900: 9). pron. sicӸ ‘such a one as this/here’, cf. also the infrequent derivation sicevԔ of the same meaning8; verbs with an aorist stem in Ռa, i.e. an inf in Ռati; as far as I can tell on the basis of Sadnik & Aitzetmüller (1955: 176-177, 181-186) and SSSl, 17 stems are attested with a modified velar preceding Ռa: klicaՌ ‘call’, ՌricaՌ ‘say’, ՌšticaՌ ‘allude’, dvidzaՌ ‘move, shift’, ՌmidzaՌ ‘wink’, ՌstridzaՌ ‘cut (hair)’, ՌžidzaՌ ‘kindle, ignite’; brNjcaՌ ‘make the sound of strings’, ՌlNjcaՌ ‘bend’, ՌsNjdzaՌ ‘touch’, ՌtNjdzaՌ ‘ask, discuss’; bliscaՌ (sNj) ‘shine, light up’; ՌmrӸcaՌ ‘fade, go out’, zrӸcaՌ ‘watch’ (attested only in zrӸcalo ‘mirror’, Supr once), trӸdzaՌ ‘break off, pull out’; klԔcaՌ (presumably) ‘knock’9; ՌvycaՌ ‘learn’10; msc oՌstem nouns containing the borrowed Germanic suffix *ՌingՌ, e.g. kԔnNjdzӸ ‘king, leader’; a stem *gobӸdzՌ ‘plenty, prosperity’, borrowed from Germanic and reflected in such derivations as ugobӸdziti ‘produce a plentiful harvest’.

That is all as far as OCS is concerned. Elsewhere a few additional examples have been spotted, of which to my mind the most important are: 18. PSl *elӸǧa ‘alder tree’. The modified velar is attested in toponyms in Croatia (Jelsa, e.g. near Karlovac, see ARj s.v. jelsa).11 It has been reported as occurring in in appellatives in scattered Slovene dialects, notably East Štajerski (Bezlaj 1977-2005 s.v. jélša). The relevance of this word for Prog has not often been recognized, but see Knutsson (1926: 29), Grickat-Virk (1951-52: 97), Shevelov (1964: 340). 8

Lunt (1981: 80-81) maintains that sicӸ “is surely dialectal (I suspect Moravian)”, without advancing evidence for that view, which seems arbitrary. The pronoun, though not a frequent word, is attested in all major canonical manuscripts except KBl and Ps Sin. 9 The only time the word occurs (Ps Sin 97b) it translates the rare Greek verb skállǢ ‘hoe, scrape, etc.’, which happens to be used in the metaphorical sense of ‘searching (in one’s mind)’. The meaning ‘knock’ (as given by Diels 1932: 133) is speculative, but may well be correct. It is presumably based on that of possible cognates within OCS (e.g. klӸǁӸtati ‘rattle’) and elsewhere, e.g. SCr (kucati ‘knock’), see Bezlaj (1977-2005 s.v. kolcati). 10 With the exception of ՌšticaՌ this is identical to the list compiled by Jeǻowa (1968: 56-61). All other lists I have seen are either too selective or give non-canonical material or verbs not attested in forms actually containing a reflex of *ǀ, *dǺ, or *ǧ. 11 There is also the town of Jelsa on the Central Dalmatian island of Hvar. Although I agree with Grickat-Virk (1951-52: 97) that this form is probably authentic, one cannot completely rule out the possibility that it may have been influenced by the Italian variant of the name (see Skok 1971-1974 s.v. jelha).

510

WILLEM VERMEER

19. PSl *orbotNjdǺӸ as reflected in OP robociƾdz ‘boy’ (S’StP), OCz robotLJz ‘slave’ (Novák 1934: 140, BLJliǁ et al. 1979: 416), cf. Arumaa (1976: 33). 20. In recent years, F.R. Minlos and V.A. Terent’ev (2002) have drawn attention to the possible existence of a PSl noun *vӸǧӸ denoting various toxic weeds growing in marshy surroundings, and attested in such forms as Ukr ves’, Cz (dial.) veš. As in the case of other words with stem-final *ՌǧՌ, attestations with unmodified velars occur too, e.g. R vëx (more on this in section 2.4). Evaluation is made complicated by the presence of attestations reflecting *LJ in the stem (*vLJxԔ, *vLJxa) and/or analogical interference from the verb viti, e.g. R v’ëx. See also Minlos (2001). Scholars agree about the relevance of the OCS examples apart from pӸcӸlԔ, no doubt because its status is unclear: is it inherited? is it a loan from Latin? is it inherited but contaminated by the Latin word? how to square the modified velar of pӸcӸlԔ with the retained velar found everywhere outside Supr? The noncanonical material has never drawn much attention or influenced the debate. In nearly all examples, the modified velar is preceded by a front vowel, specifically *i, *Ӹ and *Nj, but never *e or *LJ. All exceptions are verbs with an aorist stem in Ռa, notably stems containing the reflex of a syllabic *r, e.g. trӸdzati, and also pres 1pl navycajemԔ ‘learn’ (Supr 373: 11-12) and imperf 2/3sg klԔcaše (presumably) ‘knock’. Since all such examples participate in patterns linking imperfectives to correlating perfective verbs (or, in the case of rare words like klԔcati, are under reasonable suspicion of doing so), they may have arisen by analogical extension of preexisting cases of alternation, as has always been realized, beginning with Sobolevskij (1889: 31-32).12 By assuming that it is the front vowel that triggered the palatalization, the distributional pattern is accounted for in a way that makes phonetic sense. Other explanations fall short by failing to express the fact that palatalized velars are found only after front vowels.13 Part of the items listed are also attested with unmodified velars, for which I refer to section 2.4. 12

Cf. also, for instance, Baudouin de Courtenay (1894: 49), Meillet (1902-05: 48), Zubatý (1910: 153-155), Beliǀ (1921: 37-38), Trubetzkoy (1922: 226-227), Shevelov (1964: 340b, 342n), Lunt (1981: 24). 13 As far as I know there are two: (1) the early (pre-Baudouinian) view attributing palatalization to a hypothetical *j following the velar; (2) the view attributing it to morphological factors (e.g. OtrNjbski 1948, Beljaev 1986). A third view I regard as fundamentally ill-conceived is the one attributing Prog to a syllabic nasal, held to be reflected as Ӹ, preceding the velar (Jacobsson 1974, and other publications). Jacobsson’s theory clashes with the general consensus that the predominant reflex of the syllabic nasals in Slavic is *Nj and gives rise to the problem of justifying the presence of the reconstructed nasals in every individual case, e.g. *otӸǀӸ < *atnkas.

PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION

511

2.2 Two false positives Two items that have often been adduced as positives are not on the list because there are strong reasons to doubt that they should be on it: •

•

PSl *tridǺӸ ‘three-year-old’. The stem of this adjective – which happens not to be attested in OCS – has often been identified with that of Lith treigׁs ‘three-year-old horse’ and regarded as an example of Prog (e.g. Knutsson 1926: 64, Vaillant 1958b: 673, Shevelov 1964: 339, Steensland 1975: 92, Arumaa 1976: 32-33). However, there is no evidence for the stem-final *ՌdǺՌ this reconstruction requires. All attestations point to *trizԔ with stem-final *ՌzՌ, starting with the Izbornik of 1073, which has Isg fem trizoj՛ instead of **trizej՛ (123v14, 135g24, ed. Dinekov 1991-1993). At present there are no good grounds for regarding the word as an example of Prog (Milewski 1937: 9n., more explicitly Lunt 1981: 19 with note 36).14 OCS drNjselԔ ‘sad’. This word is limited to South Slavic. The view that it is an example of Prog is based on the fact that the ՌsՌ (which may or may not reflect *ǧ) alternates with ՌxՌ in the pan-Slavic synonym drNjxlԔ, suggesting that it too reflects *x, hence is the outcome of palatalization of *x. If that is correct, the ՌeՌ of the element ՌelԔ cannot reflect original *e, which would have caused a preceding *x to undergo Reg1, counterfactually yielding **drNjšelԔ. Hence we have to reconstruct an older shape *drNjxolՌ, or rather – but that goes without saying – something along the lines of *drinxalՌ or *drǕxalՌ, in which *x underwent Prog and regularly umlauted the following *o – or rather *a – to *e (Meillet 1902-05: 412). However, as Meillet points out himself, no suffix *ՌolԔ is known ever to have existed, cf. also Vaillant (1974: 545-568). For that reason I prefer the alternative explanation which interprets drNjselԔ as a local – hence possibly recent – South Slavic reshaping of inherited *drNjxlԔ on the basis of its antonym veselԔ ‘cheerful’ (cf. Skok 1971-1974 s.v. dresel).

2.3 Counterexamples Following *i, *Ӹ and *Nj, unmodified *k, *g, *x actually occur in more items than the reflexes of *ǀ, *dǺ, *ǧ. The question how to account for this evidence is among the challenges posed by Prog. Since the counterevidence has never been systematically evaluated, what follows can only be a rough outline.15 14 In the case of *dvidǺӸ ‘two-year-old’, which is documented much less abundantly, there is no evidence for *dǺ either (see for the material Knutsson 1926: 64, Arumaa 1976: 32-33). 15 Palatalization is never found if the velar involved is followed by a consonant, e.g. stӸgna ‘street’ (cf. stӸdza ‘road, path’); vԔsklikn՛ti ‘exclaim’ (cf. klicati ‘shout, cry, scream’). Other examples: igrӸ ‘play, amusement’, lixva ‘interest, usury’, stӸklՌ, as in stӸklLJnica ‘glass’. The point is

512

WILLEM VERMEER

The first thing to realize is that legitimate counterexamples are much harder to identify than positives. Whereas the latter can be recognized fairly reliably by the presence of *ǀ, *dǺ or *ǧ not attributable to a following LJ or i reflecting *oi, many cases of unmodified k, g or x that may well be valid counterexamples, may equally well be too recent to count. Under the circumstances one is forced to exclude a large amount of material, some of which may well be legitimate: •

•

•

•

Productive formations that can have arisen analogically at any stage after Prog. Suffixal ՌӸka in OCS klLJtӸka ‘hut’ may have resulted from substitution of ՌԔkՌ for ՌӸcՌ in a suffix (cf. ՌӸca). There are many transparent suffixal derivations in which the unmodified velar can easily be due to analogy, e.g. lӸgyǝi ‘relief, alleviation’ (with the ՌgՌ of lӸgԔkԔ), lixotԔkԔ ‘excess’ (with the ՌxՌ of lixԔ); such examples as ՌsNjga (as in prisNjga ‘oath’) and ՌdvigԔ (as in podvigԔ ‘struggle’) are representatives of a highly productive derivational process (for which see Vaillant 1974: 34-283, notably 265-274). Aorists of verbs with stems in front vowels that trigger Prog, e.g. 1pl pixomԔ ‘we drank’. Although Lunt (1981: 35-36) has adduced these formations as counterevidence, they can easily be analogous and/or too recent to matter, hence unfit for the role (cf. Vermeer 2002-03: 378n). 16 Suffixal ՌӸkԔ in tNjžӸkԔ ‘heavy’, gorӸkԔ (or perhaps goǤӸkԔ) ‘bitter’, and *malӸǁӸkԔ as attested in maleǁ’ko ‘a little bit’ (Euch 44a, 24). There is no reason to regard the type as particularly old: tNjžӸkԔ ‘heavy’ and gorӸkԔ ~ goǤӸkԔ may have been influenced by their antonyms lӸgԔkԔ ‘light’ and sladԔkԔ ‘sweet’; in addition tNjžӸkԔ looks like a recent substitution for *tNjgԔkԔ implied by the verb otNjgԔǁiti ‘burden’; *malӸǁӸkԔ is a double diminutive and as such a productive formation. Borrowings or words of obscure background or formation or chronology, of which there are several, e.g. grӸkԔ ‘Greek’, likԔ ‘organized communal activity, such as round dance, choral singing, or assembly’; tikԔ ‘mirror’, NApl kԔǝigy ‘book, piece of writing’, veriga ‘chain, fetters’.

Three putative counterexamples that have often been wrongly adduced in discussions of Prog are: •

*likԔ ‘face, picture’. The oldest attestations of this word are relatively recent and it is unattested in the earliest stages of the West Slavic languages, hence there are no good reasons to regard it as Proto-Slavic, all the more so be-

so obvious that the relevant material has never been collected and critically discussed. See Beliǀ (1921: 25) for a preliminary inventarization that is informative, but contains many examples that are non-canonical (e.g. igla) or otherwise objectionable (e.g. vixrԔ where *vixԔrՌ is more likely). 16 Lunt (1981: 35-36) misleadingly conveys the suggestion that those examples are given on the authority of Machek (1958).

PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION

• •

513

cause it “est un postverbal du verbe liǁiti, comme oblikӸ de obliǁiti, qu’on a pu créer à toutes époques” (Vaillant 1965: 126, cf. also Lunt 1981: 19, 63-64).17 *migԔ ‘wink’; although formally identifiable with Lith miֿgas ‘sleep’, the word can easily be a recent derivation, hence not indicative of anything (Vaillant 1974: 205). The adj. tixԔ ‘quiet, calm’ has always been identified with the family of Lith *teisùs (4) ‘right, just’ (e.g. Meillet 1934: 92). However, since tixԔ is (a)stressed, hence acute (as shown by the short ՌiՌ of SCr tզh) whereas the Lith stem is circumflex, the identification cannot be correct as it stands.

Also to be excluded are examples in which the front vowel and the velar are separated by a consonant (in practice always s). The suffix *ՌӸskԔ – e.g. mirӸskԔ ‘worldly, secular’ – suggests that in that context Prog does not operate. The fact that we find Prog in bliscati sNj ‘flash, shine’, is not conclusive, despite Van Wijk (1931: 67) and Meillet (1934: 133): since modified velars are in some cases arguably analogical in verbs with an aorist stem in Ռa, this one may well be analogical too (see section 2.1).18 As far as I can see, this leaves seven more or less credible counterexamples attested in canonical OCS: 1. suff. Ռika attested only in bližika ‘relation, relative; “neighbour” (in the christian sense)’ and ՛žika ‘relation, relative’; 2. ženixԔ ‘bridegroom’; 3. suff. ՌnՌikԔ as in dvӸrӸnikԔ ‘door-keeper’19; 4. igo ‘yoke’; 5. lixԔ ‘superfluous, excessive’; 6. one or two adjectives containing the suffix ՌԔkԔ, e.g. lӸgԔkԔ ‘light’ (cf. polӸdza), mNjkԔkԔ ‘soft’, also *tNjgԔkԔ implied by otNjgԔǁiti ‘burden’; 7. suffixal ՌikԔ as attested in tolikԔ ‘that big’ and similar formations, including velikԔ ‘big, great’. Even this reduced list may well be felt to be overly generous. True, igo and lixԔ have solid etymologies showing that they are older than Prog, and ՌnՌikԔ is at least parallelled in Baltic. But who is to say that Ռika, ženixԔ, and the types 17

Quite a few etymological dictionaries refer users looking up lice to likԔ (e.g. Berneker 1908-14, Bezlaj 1976-2005); some do not even list lice as a distinct entry (e.g. Skok 1971-1974). To the extent that this practice suggests that likԔ is primary and lice secondary it is misleading because it is lice that has a credible Proto-Slavic pedigree while likԔ has not. 18 All other examples have unmodified ՌkՌ, but none of them is completely beyond criticism. In iskati ‘seek’ the i probably reflects *ei (Meillet 1934: 92). The same is not excluded in the case of Ռtiskati ‘press’. This leaves piskati ‘whistle’ (piskaxomԔ Mar, Mt 11: 17), which is onomatopoeic. 19 Without a preceding ՌnՌ we find ՌikԔ in OCS only in a single attestation of zlatikԔ (Supr).

514

WILLEM VERMEER

lӸgԔkԔ and tolikԔ, all of which are Slavic innovations, are old enough to count? I have left Ռika in because I cannot think of a motivation for secondary reintroduction of ՌkՌ in **bližica and **՛žica, and the others because they are all panSlavic, hence at least approximately as old as Prog. But I’ll be the first to admit that those considerations are not conclusive. Of the cases of unmodified velars after *i, *Ӹ and *Nj attested outside canonical OCS, very few can plausibly be regarded as old enough to make a difference. The most persuasive ones are the following: 8. *kԔnNjgyǝi ‘female kԔnNjdzӸ; wife or daughter of kԔnNjdzӸ’. The word is widely attested, beginning with the Izbornik of 1073 and early twelfth-century birchbark Russian. Its unmodified ՌgՌ alternating with modified *ՌdǺՌ in *kԔnNjdǺӸ suggests that it arose at a stage when the latter word had a stemfinal velar, in other words before Prog took place. 9. *vixԔrӸ or *vixԔrԔ ‘whirlwind’. Judging by Carelian viehkuǤi ‘gust of wind’, Ludic vieʖkuri ‘whirlwind’, both reflecting Common Finnic *vljhkuri, this word was borrowed into Finnic before the Common Slavic merger of the monophthongized reflex of *ei with *ǔ (Kallio 2006: 155, on the basis of Kalima 1956: 140). 10. *pӸxati ‘pound (in a mortar)’ and related meanings. Though unattested in canonical OCS, the word occurs all over Slavic and has a convincing Baltic etymon, hence must have been present in the language whenever it was that Prog took place; the inf *pӸxati presupposes an aor2/3sg *pӸxa with ՌxՌ (cf. Vaillant 1966: 206). 2.4 Examples attested with both modified and unmodified velars Some examples of *ǀ, *dǺ, *ǧ are attested with unmodified *k, *g, *x as well. Even within canonical OCS, for example, the noun pӸcӸlԔ also occurs with ՌkՌ, as we have seen (section 2.1). In all other examples of this type, the velar in question is stem-final. Many of those involve the substitution of suffixes containing *ՌԔkՌ for *ՌӸǀՌ (like OCS klLJtԔka) and may have arisen at any stage. Two types of cases look more substantial and require some discussion. Several of the OCS verbs with an aorist stem in ՌaՌ (see section 2.1) also occur with unmodified velars. Alongside bliscaՌ there is bliskaՌ (once Supr), alongside ՌstridzaՌ there is ՌstrigaՌ (once Euch), alongside trӸdzaՌ there is trӸgaՌ (common). This gives only a pale impression of the variety found in Slavic as a whole, which reflects the ease with which alternating aspectual pairs (*klikn՛ti ~ kliǀati) can become non-alternating (> *klikn՛ti ~ klikati), and vice versa (*navykn՛ti ~ navykati > *navykn՛ti ~ navyǀati).

PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION

515

In East Slavic, verbs in Ռati participating in derivational patterns linking imperfectives to correlating perfectives always have unmodified velars, e.g. R dvigat’ ‘move’, except for verbs that can be understood as borrowings from OCS, such as R vosklicat’ ‘exclaim’. The absence of modified velars in productive formations obviously does not constitute evidence that Prog failed to reach East Slavic. After all, a modified velar is present all over Slavic in the reflexes of *sӸǀati ‘piss’ (e.g. R scat’, ssat’, not **skat’), which happens not to be attested in canonical OCS, but is otherwise ubiquitous and clearly old.20 As a rule, West Slavic too has unmodified velars in verbs participating in derivational patterns linking imperfectives to correlating perfectives, e.g. OCz po(v)zdvíhati ‘raise’.21 Nevertheless scattered items with modified velars are attested just about all over West Slavic, particularly in Czech. Skipping the verb *kliǀati (poorly attested and onomatopoeic) and the noun meaning ‘mirror’, six of them occur in Old and/or dialectal Czech: *klNjǀati ‘limp’, *lNjǀati (appr.) ‘set (snares)’, *tNjdǺati ‘ask’, *midǺati ‘disappear’, *myǀati (sNj) ‘stroll about’, *tyǀati ‘run’. Of the six verbs attested in Czech, three also occur in (Upper and/or Lower) Sorbian (*klNjǀati, *lNjǀati, *tNjdǺati), and of those one each in Old Polish (*tNjdǺati) and Slovincian (*klNjǀati). For the details I refer to Jeǻowa (1968: 159164). Note the presence in Czech of *myǀati, suggesting that at some stage Czech, like South Slavic, extended the range of modified velars.22 A second important case of this type of variation is found in native nouns and pronouns with stems originally ending in *ՌgՌ and *ՌxՌ. All of these are attested with unmodified velars as well: 1. Reflexes of PSl *Njga ‘illness, wound; horror; anger; witch’ with unmodified *g are the norm in Russian and Belorussian; this is the type R jagá, as in bába-jagá.23

20

Cf., e.g., Vaillant (1950: 291), Grickat-Virk (1951: 105), Jeǻowa (1968: 164-176). Unaccountably, Baudouin de Courtenay treated Church Slavonic borrowings on a par with the inherited material (1894: 48-49). With dramatic consequences, because his – influential – rule by which Prog affects velars only if they begin a stressed syllable is based largely on the alternation found in pairs like klíkat’ ~ vosklicát’, dvígat ~ vozdvizát’ (Baudouin de Courtenay 1894: 48). 21 For specific examples see StƀSl s.v. povzdvihnúti. 22 The case of *tyǀati is different. Assuming that it is a derivation from téci it looks like a lengthened grade based on present tense forms with ՌӸՌ in the stem – like OCS Ռricati – with subsequent analogical substitution of the hard initial tՌ of the basic verb for the soft tՌ that is regular in *tiǀati. 23 The semantics of this word are different in different Slavic dialects. In West and East Slavic it usually refers to witches or witch-like creatures of the imagination. In South Slavic three areas can be distinguished: Bulgarian and Macedonian, including OCS, have meanings of the type

516

WILLEM VERMEER

2. In *lӸga ‘lightness, ease’ and compounds like *polӸga ‘relief’, unmodified *g is normal in Russian and Belorussian, and attested also in Ukrainian (e.g. pil’ha) and Lechitic (Polish odelga, zelga, ulga). Modern Russian pol’za ‘use’ cannot possibly be authentically Russian because in that case the final stem consonant would have to be soft, so that we would have **pol’zja (Šaxmatov 1913: 41, Durnovo 1925: 487). 3. Attestations of *stӸga ‘path’ with unmodified *g have been reported for Russian and Belorussian.24 The stylistic value of R stezja, on the other hand, clearly betrays the Church Slavonic origin of the word. 4. Old Russian has rabotjagԔ ‘slave’, where Polish and OCz have robociƾdz and robotLJz respectively (for which see section 2.1).25 5. *vӸǧӸ/*vӸxԔ ‘whole, all’. As has been known for a long time (e.g. Ljapunov 1894: 169n; 1900: 73), unmodified ՌxՌ is attested in the medieval Novgorod dialect, e.g. NAsgn vԔxo (birchbark Novg. 893), Gsgf vԔxoLJ (Novg. 850), both twelfth century, and many other examples, for which see Zaliznjak (2004: 46). 6. *elӸǧa/*elӸxa (also *oՌ). Whereas this word is attested with a modified velar in Croatian toponyms and possibly dialectally in Slovene (section 2.1), it has an unmodified ՌxՌ almost everywhere in Slavic. In addition it is frequent with stem-final ՌšՌ, which has traditionally been explained as the reflex of suffixal *ՌjՌ (e.g. Skok 1971-1974 s.v. jelha, ÈSSJ s.v. elӸša, Andersen 1996: 128), but might reflect *ǧ in West Slavic.26 7. The noun *vӸǧӸ reconstructed by Minlos and Terent’ev (2002, see also section 2.1) is attested with unmodified stem-final *ՌxՌ in the Russian-Belorussian area, e.g. R vëx. In the Germanic borrowings the facts are the following: a. Nouns in *ՌNjdǺӸ continuing Germanic *ՌingՌ (*kԔnNjdǺӸ, *pLJnNjdǺӸ, etc.) all have specialized meanings of types that are unsuitable for documenting local developments (cf. Shevelov 1964: 340). As is well known, such nouns often have unmodified *ՌgՌ in the north, notably in Russian/Belorussian, e.g. varjag ‘Varangian’, but all examples of that type may be too recent to mat‘illness, wound’; Štokavian and ƀakavian Serbo-Croatian meanings of the type ‘horror’, whereas meanings of the type ‘anger’ are characteristic of Slovene and Kajkavian Serbo-Croatian. 24 And very marginally for Polish (Pracki 1907: 264, cf. Zierhoffer 1959: 69). 25 For other possible examples of this type see Grickat-Virk (1951-52: 102). 26 There are Polish attestations with *ՌsՌ which cannot reflect *elӸxa (or *olӸxa) and may plausibly be attributed to the Baltic substratum underlying Polish; since the same may – or may not – hold for the numerous East Slavic forms with *ՌsՌ these have not been taken into account here (cf. Andersen 1996: 127-128).

PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION

517

ter. On the other hand knjaz’ refers to an institution that probably arose in the south (Kiev), so it is not indicative either. b. The Slavic evidence for *gobӸdǺa (with *dǺ) is limited to a handful of OCS attestations of *gobӸdǺevati and *ugobӸdǺiti written with (Mar, Ass, Ps Sin). As far as I have been able to determine, all attestations outside OCS point to *gobӸza (with *z), e.g. OR Asg gobzu (not **gobzju), Cz Nsg Hobza (not **Hobze). None of the Old Russian examples reported by Sreznevskij (1890-1912 1: 630; 3: 1134) and Sl11-14 (2: 342) provides evidence for *dǺ.27 As far as I can see, there are no credible cases of modified stem-final *g in unquestionably authentic Russian or Belorussian material. The problem of how to explain this state of affairs is less straightforward than that of the verbs and is going to occupy us fleetingly below (sections 3.7-8). 2.5 Examples combining Prog and Reg2 In some forms, the modified velar involved is both preceded by *i, *Ӹ or *Nj and followed by the reflex of *oi. What is most significant about this material is the seemingly inconsistent reflex of *oi, which is ՌiՌ in some cases and ՌLJՌ in others: 1. Case forms of nouns that have modified velars throughout: Dsg dLJvici, Lsg tӸmӸnici, Lsg kԔnNjdzi, Lpl srԔdӸcixԔ, cf. also NAVdu pӸtici, lici. 2. Case forms of pronouns that have modified velars throughout, e.g. Isg vӸsLJmӸ, GLpl vӸsLJxԔ, Dpl vӸsLJmԔ, Ipl vӸsLJmi. 3. Imperatives, e.g. strizLJte ‘shear (sheep); cut (hair)’, rӸcLJte ‘say’, raždӸzLJte ‘kindle, ignite’. 4. The isolated form *lӸdǺLJ, used as in nLJstӸ mi lӸzLJ inamo iti (Supr) ‘it is not possible for me to go elsewhere’. The question whether it is ՌiՌ or ՌLJՌ that is the phonetically regular reflex here has turned out to be of crucial importance to the chronology of Prog and will be taken up at length below (sections 3.2-3). 2.6 The limitations of the factual material As we have seen, there are a mere seventeen OCS positives, ten of which involve *k (Ռica, ՌӸca, mLJsNjcӸ, zaNjcӸ, pӸcӸlԔ, ՌӸcӸ, lice, ՌӸce, nicӸ, sicӸ), five *g (Njdza, polӸdza, stӸdza, *gobӸdzՌ, the type kԔnNjdzӸ), and one *x (vӸsӸ), with the type dvidzati comprising stems ending in *ՌkՌ and *ՌgՌ. If one includes the most 27

One is reminded here of the Ԕ found instead of **Ӹ in the stem of the Novgorod reflex of the pronoun *vӸǧӸ ~ *vӸxԔ, and the traces of progressive assimilation of absence of palatality in early Russian, e.g. in the name *SԔdila (see Zaliznjak 2004: 54-55, 83).

WILLEM VERMEER

518

plausible extra-canonical material (*orbotNjdǺӸ, *elӸǧa), we get 10/6/2, or, if one is willing to give Minlos and Terent’ev’s *vӸǧӸ the benefit of the doubt, 10/6/3. The number of credible counterexamples is perceptibly smaller. There are seven OCS items, three involving *k (Ռika, ՌnikԔ, the type tolikԔ), one *g (igo), and two *x (ženixԔ, lixԔ), with the type lӸgԔkԔ/mNjkԔkԔ comprising roots ending in *ՌkՌ and *ՌgՌ. By including lӸgԔkԔ/mNjkԔkԔ and the most plausible extracanonical material (*kԔnNjgyǝi, *vixԔrӸ/Ԕ, *pӸxati), one gets 4/3/4. If one looks at the vowel preceding the velar, the numbers are the following: • • •

Preceding ՌiՌ positive vs. negative 5/6: Ռica, lice, nicӸ, sicӸ, verbs with an aorist stem in ՌaՌ; Ռika (bližika), ženixԔ, ՌnikԔ, igo, lixԔ, the type tolikԔ. Preceding ՌӸՌ positive vs. negative 9/1: polӸdza, stӸdza, ՌӸca, pӸcӸlԔ, ՌӸcӸ, ՌӸce, the pronoun vӸsӸ, verbs with an aorist stem in ՌaՌ (*sӸǀati), the borrowed stem *gobӸdǺՌ; lӸgԔkԔ. Preceding ՌNjՌ positive vs. negative 5/1: Njdza, mLJsNjcӸ, zaNjcӸ, verbs with an aorist stem in ՌaՌ, the type kԔnNjdzӸ; mNjkԔkԔ.

My reason for giving these numbers is not to let anything depend on them, but to bring out the quantitative limitations of the material, which raise the possibility that it is compatible with more than a single plausible scenario. Adding non-canonical examples brings to light quite a few items showing unmodified velars where OCS or other South Slavic dialects display palatalization. These comprise two major groups: a. the members of the type dvidzati to the extent that they are – or can be assumed to be – the imperfective member of an aspectual pair, which nearly all of them are; b. native nouns and pronouns with stem-final *g and *x. 3. Pedersen (1905) and the intermediate inflections Holger Pedersen’s reason for putting Prog before Mono (hence before Reg2) has to do with the inflection of nouns with stems in original velars modified by Prog. It is an unexpectedly complex subject. 3.1 Pedersen’s evidence The reason Pedersen puts Prog before Reg2 is because he wants to explain why oՌ and ƽ-stem nouns with stems ending in velars modified by Prog take soft endings, in other words, show the workings of Umlaut: “Denn die durch B. de C.’s gesetz [i.e. Prog, W.V.] entstandenen i-haltigen laute [meaning *ǀ/dǺ/ǧ, W.V.] haben auf ein folgendes o Ԕ und oi die gleiche wirkung wie j ausgeübt: abulg. lice, r. licó ‘gesicht’, abulg. otӸcӸ ‘vater’, lok. abulg. lici otӸci. Sie sind also älter als die monophthongierung der diphthonge” (Pedersen 1905: 385).

PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION

519

If one adds these considerations to the skeleton chronology discussed in section 1.5 what results is the chronology shown in Table 2. Table 2. Pedersen’s chronology (original version) 1 Prog. 2 Umlaut. 3 Mono. 4 Reg2. Pedersen pays no attention to the examples of ՌLJՌ reflecting *ՌoiՌ after velars modified by Prog, notably in pronouns, e.g. GLpl sicLJxԔ, and imperatives like Imp2pl strizLJte (see section 2.5). Yet the problem of these reflexes should have been faced in order to make sure the chronology accounts for a piece of reality that needs accounting for. If by any chance it would turn out that all cases of ՌiՌ are secondary and that ՌLJՌ is the regular reflex here, that would mean (a) that the cases of ՌiՌ do not also need a phonetic explanation and (b) that the reflex of *oi is not umlauted after velars modified by Prog. The former point would remove the evidence Pedersen wants to account for, thereby rendering the hypothesis redundant. The latter would actually be difficult to reconcile with the idea that Prog preceded Mono, because if it did, why did not *oi undergo Umlaut after velars modified by Prog? Ironically, the point had been decided in favour of ՌLJՌ, hence against Pedersen’s chronology, even before the latter was published. 3.2 Meillet on sicLJxԔ and Umlaut (1900) In 1900, Antoine Meillet discussed the reflex of *oi in the context of an analysis of the inflection of the pronouns *siǀӸ and *vӸǧӸ, both of which have stems ending in velars modified by Prog. The pronoun sicӸ has ՌLJՌ in endings originally starting in *ՌoiՌ: OCS Isgmn sicLJmӸ, GLpl sicLJxԔ, Ipl sicLJmi. The same holds for vӸsӸ. Meillet argued that the ՌLJՌ of sicLJxԔ is much less likely to be analogical than the ՌiՌ of srԔdӸcixԔ, hence more likely to be phonetically regular, because *ՌoiՌ is much more prominent in the paradigm of the pronouns, where it is found in the Isgmn and the oblique plural cases, than in that of the nominal ƽՌ and oՌstems, where it is limited to the DLsg and the Lsg and Lpl respectively (Meillet 1900: 8). If Meillet is right (and the least one can say is that he has a prima facie case), the reflex of *oi did not undergo Umlaut after velars modified by Prog, although in all other positions it was subject to Umlaut exactly like *o, and although we find umlauted *o after velars modified by Prog in such examples as *sӸrdӸǀe. The chronological indications with respect to Mono now look contradictory: the ordinary reflex of umlauted *oi, which is i, shows that Umlaut must have

520

WILLEM VERMEER

preceded Mono, but the fact that Umlaut leaves *oi untouched after velars modified by Prog – while affecting *o – suggests that in these cases Umlaut was later than Mono, which converted *oi into a front vowel which – being a front vowel – was no longer susceptible to Umlaut. Things look as if Umlaut hit Slavic twice, once before Mono and once afterwards. That obviously is an unattractive result. Meillet remedies the anomaly by treating Umlaut as a synchronic rule that entered the language before Mono, so that *oi was umlauted, but that remained present for some time afterwards and still was operative at the post-Mono stage when Prog took place. Since at that stage “o, Ǯ [i.e. Ԕ], y n’existaient après aucune consonne palatale”, sequences like *ǀo or *ǀԔ automatically changed into *ǀe and *ǀӸ the moment they arose. By then, however, the reflex of *oi had yielded a monophthong that was not longer susceptible to Umlaut because it was a front vowel already, so *ǀLJ was not changed. This reasoning yields the chronology shown in Table 3. Table 3. Version of the default chronology incorporating Meillet’s conception 1 Rise of Umlaut (as a synchronic rule). 2 Mono. 3 Prog (simultaneous with Reg2); at this stage, Umlaut is synchronically present, causing back vowels to be fronted if preceded by the new palatal consonants produced by Prog; *ՌLJՌ < *ՌoiՌ remains unaffected because it was a front vowel already as a consequence of Mono. 4 Cessation of Umlaut. Meillet’s conception generates the correct output and keeps Reg2 and Prog together. Pedersen’s chronology does neither. There is no need to argue which is superior. All of this gets done by the expedient of interpreting Umlaut as a synchronic rule that persisted in the language for some time, which enables Meillet to treat it as a unitary phenomenon although it appears at first sight to have operated both before and after Mono. Note that Meillet’s conception is more or less inevitable because even at as late a stage as the language underlying OCS, back vowels were still excluded after palatal consonants (e.g. Trubetzkoy 1954: 63). 3.3 The morphological implications of Meillet’s ՌLJՌ Meillet’s analysis of OCS sicӸ and vӸsӸ implies that nouns and pronouns with stems ending in velars modified by Prog were originally inflected according to paradigms that had soft endings in most case forms, but ՌLJՌ in those where soft ՌiՌ is opposed to hard ՌLJՌ.

PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION

521

Prog must have given rise to such paradigms intermediate between hard and soft in the case of at least the following declensional types: 1. 2. 3. 4. 5.

masc oՌstems, e.g. *mLJsNjǀӸ, *otӸǀӸ. neuter oՌstems, e.g. *liǀe, *sӸrdӸǀe. ƽՌstems, e.g. *stӸdǺa, *tӸmӸniǀa. adjectival oՌ/ƽՌstems, represented by *niǀӸ. pronominal oՌ/ƽՌstems, represented by *siǀӸ and *vӸǧԔ.

As we have seen (sections 2.5, 3.2), the intermediate paradigm, with its ՌLJՌ, is attested as such in the pronouns, whereas all nominal declensions have the ՌiՌ found in the soft paradigms too. These facts imply that at some stage preceding the earliest texts, soft ՌiՌ replaced ՌLJՌ in the nouns and adjectives, causing those paradigms to converge with the soft inflections. Since the substitution diminished redundant variation among inflectional patterns, it was so well motivated as to verge on the trivial. In the nominal inflections as attested in the oldest material, the only remaining difference between the intermediate paradigm and the soft subtype is the Vsg of the masc oՌstems, where the former continues inherited *Ռe whereas the latter has an ending that appears to have been borrowed from the uՌstems, e.g. otӸǁe, kԔnNjže vs. cLJsarju, zmiju. The point is going to occupy us briefly again in section 4.1. As we saw just now, Meillet explains the conservatism of the pronouns by arguing that the position of *ՌoiՌ is stronger in the pronominal than in the nominal paradigms. The force of the argument can be appreciated by looking at modern Russian, which has almost completely eliminated the difference between hard and soft paradigms in the nominal declensions, among other things by analogically transferring the hard reflex of *Ռoi into paradigms that originally had soft endings, as in Lsg kljuǁé, DLsg dušé. At the same time the two endings have been maintained side-by-side to this day in the pronouns (GLsg moíx alongside tex and vsex). A point Meillet did not make explicit – he may have found it too obvious to bother – is the following. If it is assumed (with Pedersen) that *ՌiՌ is the regular reflex of *oi after velars modified by Prog, the ՌLJՌ we find in the pronouns would have to be explained as the outcome of analogical transfer of ՌLJՌ out of the hard inflection into the paradigm of *siǀӸ and *vӸǧӸ. This gives rise to the problem of finding a motivation for that change. The advocates of Pedersen’s chronology have never faced that challenge, let alone met it. Against this background, the isolated form *lӸdǺLJ (listed in section 2.5) can be interpreted as a petrified DLsg which retained the original ending because it became unstuck from its paradigm before the substitution was carried out. In origin the word looks like a prepositionless locative of the type also seen in godLJ

522

WILLEM VERMEER

‘suitable, appropriate, pleasing’. The construction is of the type seen in such examples as “it is in order ...” in legalistic English, e.g. “Thus, if […] the vote has been taken and lost on the motion to commit, it is in order to lay on the table the resolution [...].” (Robert 1943: 105-106, my italics). The attested material cannot be reconciled with the assumption that i was the phonetically regular reflex of *oi after velars modified by Prog, hence the evidence Pedersen wants his chronology to account for turns out not to exist. 3.4 Pedersen and Meillet’s ՌLJՌ Subsequently, Pedersen appears to have realized that Meillet’s ՌLJՌ undermines his chronology. At least that is the way I interpret a later discussion he contributes in the context of an investigation into the rise of mobile stress in Slavic (Pedersen 1907: 213-217). As a background one has to know that Pedersen – like most scholars at the time – adhered to Baudouin de Courtenay’s view that Prog depended partly on the place of the stress (Baudouin de Courtenay 1893: 15-16; 1894: 48). Pedersen realizes that this implies the rise of alternations of modified and unmodified velars in the case of words with mobile stress, say, Nsg *NjdǺa (with stress on the ending) vs. NApl *Njgy (with stress on the stem).28 Most of those alternations can be assumed to have been levelled out, but: “Ein rest des lautgesetzlichen wechsels hat sich bei dem pronomen abulg. sicӸ, sikԔ ‚solcher’ erhalten. k erscheint u.a. in der form siko, die [...] den akzent auf der ersten silbe tragen musste (vgl. serb. tբko ‚so’); ferner in der ableitung sikovԔ, deren akzent nach r. takovój, takóv, taková, takovó u.s.w. zu beurtheilen ist; und schliesslich enthalten die dreisilbigen ursprünglich endbetonten formen sicLJchԔ sicLJmԔ sicLJmi (und die entsprechenden formen von vӸsӸ ‚all’: vӸsLJchԔ u.s.w.) nicht die Baudouin de Courtenay’schen palatale (nach denen i statt LJ zu erwarten wäre), sondern eine von dem folgenden LJ (aus oi) bewirkte spätere palatalisation.” (Pedersen 1907: 216-217). For the sake of the argument I am going to accept Pedersen’s version of the accentual rule (which has been abandoned generations ago), and also his assumption that *siǀӸ and *vӸǧӸ have mobile stress in the sense he intended. The real problem is that the reconstruction does not produce the desired outcome. 28 The example is mine – not Baudouin’s or Pedersen’s – and intended to be illustrative of the principle only. Note that the word is likely to have been (c)-stressed, i.e. mobile in the modern (Stangian) sense of that term. The final stress reflected in Russian jagá and SCr jéza would seem to exclude accent paradigm (a), whereas the short stem vowel reflected in Polish jNjdza excludes (b).

PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION

523

Pedersen neglects to reconstruct the exact system of alternations his vision generates. Had he done so, however, he would have discovered that the resulting paradigm differs only minimally from the hard pronominal inflexion. Let us look at the details. According to Pedersen’s version of the stress rule, Prog takes place only if the velar stands at the beginning of a stressed or post-posttonic syllable. Accordingly, velars remain unchanged in the case of stem stress and in that of bisyllabic endings, all of which have stress on the second syllable. Together these case forms make up the bulk of the paradigm. For the outcome see Table 4. Case forms displaying velars modified by Prog are underlined. Table 4. The paradigm of *siǀӸ as implied by Pedersen’s reconstruction Nsg m sikԔ (siǀe?), n siko siǀa Gsg sikogo sikojNj Dsg sikomu sikoi Asg m sikԔ, n siko sik՛ Isg siǀLJmӸ sikoj՛ Lsg sikomӸ sikoi Npl m siǀi, n siǀa siky Gpl siǀLJxԔ Dpl siǀLJmԔ Apl m siky, n siǀa siky Ipl siǀLJmi Lpl siǀLJxԔ Since the outcome differs only minimally from a normal hard paradigm in stem-final ՌkՌ as also found in pronouns such as takԔ – with which *siǀӸ is closely correlated – it is vanishingly unlikely that the few case forms with stems ending in a velar modified by Prog would have exerted enough pressure to drag the entire paradigm along with them. And assuming for a moment that that would have happened, against all odds, there is no reason why the development would have halted at forms with Meillet’s ՌLJՌ, which would have given rise to a paradigm unlike any other. Pedersen’s suggestion that sicLJchԔ, sicLJmԔ and sicLJmi have “hard” endings because the endings are bisyllabic is misleading because most other pronominal endings are bisyllabic too, and in their case there is no trace of unmodified velars. Contrary to what Pedersen appears to believe, forms like **sikԔ, **siko, **sikovԔ are not canonical OCS. They are limited to recent texts with a SCr background – beginning with the Nikolje Gospel – and some attested varieties of SCr. The simplest way to explain them is as the outcome of a trivial local levelling inspired by the stem-final ՌkՌ in the continuations of *takԔ and related words. There is no reason to project them back into Proto-Slavic.

524

WILLEM VERMEER

Pedersen’s attempt to account for Meillet’s ՌLJՌ in terms of his modification of Baudouin de Courtenay’s stress rule is deeply flawed and fails to rescue his chronology. No wonder that it has never been discussed in the literature. 3.5 The reception of Meillet’s ՌLJՌ by advocates of Pedersen’s chronology By the time Šǁepkin (1910) and Lehr-Sp’awiǝski (1911) adopted Pedersen’s chronology, Meillet’s ՌLJՌ had been in the record for a decade and had been developed further by Lorentz (1904). Since Pedersen’s attempt to account for it was clearly too inadequate to even discuss (see section 3.4), Meillet’s ՌLJՌ could not be ignored. This forced those wishing to stick to Pedersen’s chronology to acknowledge the existence of Meillet’s analysis, only to dismiss it out of hand. Accordingly Šǁepkin limits his discussion to the statement that Meillet’s ՌLJՌ is “offenbar nicht phonetischen Ursprungs” (1910: 215). By its very nature as an apodictic statement, Šǁepkin’s diagnosis is powerless against Meillet’s and Lorentz’s – and slightly later Zubatý’s – analysis of the relevant morphological structures, yet Tadeusz Lehr-Sp’awiǝski, writing shortly afterwards, referred to it approvingly as if it had conclusively disposed of the matter (1911: 145). Most later adherents of Pedersen’s chronology have treated Meillet’s ՌLJՌ as non-existent, ultimately on the authority of Šǁepkin’s insubstantial “offenbar”. As a consequence of all this, the place of Meillet’s ՌLJՌ in the history of the subject resembles the view of Reg2 and Prog as a single event, which most scholars came to ignore at some point too (see section 1.2). What makes Meillet’s ՌLJՌ different, however, is that it never disappeared so completely as to need resurrecting. It is explicitly discussed by Vondrák (1923-24: 22-24 = 1924: 354355), Beliǀ (1928: 60), Kul’bakin (1929: 231), Van Wijk (1931: 68-69; 1949-50: 306307), Diels (1932: 135), Milewski (1937: 15), Lokrantz (1950: 387-388), Vaillant (1950: 55), Bräuer (1961: 195), Bernštejn (1961: 209), Shevelov (1964: 347-349), Steensland (1975: 99-100), Aitzetmüller (1978: 16-17), Kortlandt (1991: 6), Schenker (1995: 91), and others. What is especially striking about the list is the number of handbooks on it, some of them widely used. Nevertheless, advocates of Pedersen’s chronology have tended to display no awareness of Meillet’s ՌLJՌ, cf., e.g., Porzeziǝski (1914: 50), Il’inskij (1916: 240245), Hujer (1924: 42), Jakobson (1929: 20-24), Koǥínek (1948: 55), Martinet (1952: 161-163 = 1955: 366-367), Channon (1972: 39-50), Velǁeva (1988: 32), and Holzer (1995: 248). Those few who have been aware of it appear not to have realized that Pedersen’s chronology is in serious difficulties as long as Meillet’s ՌLJՌ has not been conclusively defused in some way. František Mareš, for instance, writes that “the chief role in deciding the chronological ordering has been played by forms

PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION

525

of the type vӸsLJmӸ instead of vӸsimӸ” (1965b: 33 = 1956: 465 = 1965a: 38 = 1969: 57). He does not realize that even the mere possibility that vӸsLJmӸ is phonetically regular undermines Pedersen’s chronology by casting doubt on the evidence it has been designed to account for. I know only of two adherents of the Pedersen view who have explicitly discussed Meillet’s ՌLJՌ in the century that has elapsed since Pedersen (1907). Those are Trubetzkoy and Lunt. Trubetzkoy’s argument can only be evaluated within the larger context of his theory of Prog and will be taken up below (sections 5.1112). As for Lunt, he rejects the evidence of Meillet’s ՌLJՌ because, as he sees it, “pronouns are far more likely to develop peculiar innovations along with haphazard rearrangements of old materials” (1981: 36, cf. 1987: 280, 1989: 45n). If one puts this alongside Meillet’s, Lorentz’s and Zubatý’s analysis of the relevant declension patterns one cannot but conclude that much more work will have to be put in to make the point stick. What is needed for Meillet’s ՌLJՌ to be effectively defused along Lunt’s lines is the presence of a valid reason for expecting unpredictable – “peculiar” and “haphazard”, in Lunt’s terminology – developments modifying the inflection of words meaning ‘such’ and ‘whole, all’. Lunt does not provide such a reason. The fact that *siǀӸ and *vӸǧӸ are inflected according to what is technically known as the “pronominal” declension does not make them pronouns in the sense intended by Lunt. But even if they were, rejecting the applicability to pronouns of the laws of linguistic change constitutes a major theoretical innovation that cannot be seriously contemplated without a thoroughgoing investigation of the point.29 3.6 Meillet’s ՌLJՌ and the Brugmann conception of Prog Karl Brugmann assumes that Prog is triggered by a combination of a preceding front vowel and a following *j, in this way combining the early view that the examples of Prog are due to the presence of a following *j with Baudouin de Courtenay’s and others’ – then – recent insight that a preceding front vowel is indispensable (Brugmann and Delbrück 1897: 291-292). Velars in this context are assumed to have yielded ǀ/dǺ/ǧ, contrary to the reflexes of velars plus *j not preceded by front vowels, which are ǁ/ž/š. In this respect the Brugmann view is significantly more complex than other approaches. The Brugmann conception of Prog would provide an attractive explanation for the difference between the positives and the counterexamples, if there would be some possibility to independently substantiate the distribution of *j it requires. Traditionally that has been attempted primarily on the basis of examples from Baltic. A noun like *junӸǀӸ ‘steer, young bull’ is identified with Lith jaunì29

For some further discussion see Vermeer (2000: 11-12).

526

WILLEM VERMEER

kis ‘bridegroom’, Gsg jaunìkio, and derived from *jaunikՌjՌ, whereas, say, *tixԔ ‘quiet’ is identified with Lith teisùs ‘just, fair, correct’. The trouble is that the Brugmann conception of Prog cannot be reconciled with Meillet’s ՌLJՌ. If a *j was present after the velar one expects ՌiՌ as the reflex of a following *oi, there is no escaping from that. And there are various additional drawbacks, one of which is the impossibility to establish independently the required distribution of forms containing or lacking the required *j. Another unpleasant complication is the need to devise a scenario producing the double reflexes, with *ՌkjՌ normally yielding ՌǁՌ, but ՌǀՌ after certain front vowels. For more discussion see Jeǻowa (1968: 182-194). 3.7 The blocking principle and its morphological implications In 1910 Josef Zubatý, following a suggestion by Šaxmatov (1896: 703-705), put forward the assumption that Prog failed to operate if the velar involved was followed by *Ԕ, *y, *u, and possibly *՛ (Zubatý 1910: 150-151). Since this idea is phonetically plausible and accounts neatly for one or two otherwise intractable counterexamples (notably *kԔnNjgyǝi), most later scholarship has accepted it at least as far as *Ԕ and *y are concerned, remaining divided to this day on *u and *՛, for which the primary evidence appears to be moot (see also Vermeer 200203: 382-383; 2003: 407-410). The blocking principle has important consequences for our understanding of the intermediate paradigms generated by Prog. If it is correct, they must have differed from the soft subtype not only by endings with ՌLJՌ reflecting *ՌoiՌ, but also by the presence of case forms with retained velars and – hence – hard endings, as in the following examples: • • • •

velar followed by *Ԕ: Asg/Gpl *otӸkԔ, Gpl *stӸgԔ, Asg msc *sikԔ; velar followed by *y: Apl *otӸky, Gsg/NApl *stӸgy, Apl msc and Gsg or NApl fem *siky; perhaps also velar followed by *u: Dsg/GLdu *otӸku, GLdu *stӸgu; perhaps also velar followed by *՛: Asg *stӸg՛, Asg fem *sik՛.

It is important to realize at this point that potentially there are quite a few different reconstructions of the alternations that were originally present in the intermediate paradigms. Not only is there no consensus about the precise extent of the blocking rule (is or isn’t Prog blocked by *u and/or *՛?), but there is a fair amount of disagreement about the shape of some of the endings at the stage involved. So for the time being, we are forced to think in terms not of a single reconstruction, but of a range of reconstructions, which however has never been systematically explored and will not be explored in this contribution, because the exact formulation of the blocking rule and the precise make-up of the in-

PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION

527

termediate paradigms are not immediately relevant to the problem of Pedersen’s chronology. Table 5 is intended as an illustration of the principle.30 Table 5. A reconstruction of the intermediate paradigms Nsg otӸǀe (otҧkӆ?) stӸdǺa m vӸǧe? (vҧxӆ?), n vӸǧe Gsg otӸǀa stҧgy vӸǧego Dsg otӸǀu (otҧku?) stҧdǺLJ vӸǧemu stӸdǺ՛ (stҧgӚ?) m vҧxӆ, n vӸǧe Asg otҧkӆ Vsg otҧǁe stҧdǺe – Isg ? ? vҧǧLJmҧ Lsg otҧǀLJ stҧdǺLJ vҧǧemҧ Npl otҧǀi stҧgy m vҧǧi, n vӸǧa Gpl otҧkӆ stҧgӆ vҧǧLJxӆ Dpl otӸǀemԔ stӸdǺamԔ vҧǧLJmӆ Apl otҧky stҧgy m vҧxy, n vӸǧa Ipl otҧky stӸdǺami vҧǧLJmi Lpl otҧǀLJxӆ stӸdǺaxԔ vҧǧLJxӆ

vӸǧa vӸǧejNj vӸǧei vӸǧ՛ (vҧxӚ?) vӸǧej՛ vӸǧei vҧxy

vҧxy

Table 5 expresses the absence of consensus about the blocking effect of *u and *՛, and also about some important endings, notably the msc Nsg and the nominal Isgmn. Case forms with velars modified by Prog are printed in italic, those with unmodified velars are underlined, those that are ambiguous are left unmarked. The latter are (a) the Vsg of the msc oՌstems, (b) the msc Npl in Ռi, and (c) all case forms displaying Meillet’s ՌLJՌ. By just about all accounts, forms with retained velars must have constituted a minority in the paradigms involved, hence it is only natural that usually the modified velar is found to have been generalized, with concomitant changes in the endings, hence Asg and Gpl *otӸǀӸ, Gpl *stӸdǺӸ, Asg msc *siǀӸ, Apl *otӸǀNj, Gsg and NApl *stӸdǺNj, Aplm, Gsgf and NAplf *siǀNj, etc. On the other hand, since some of the case forms with unmodified velars occupy pivotal positions in the paradigm (e.g. Asg, Apl, NApl), and since there are quite a few ambiguous forms, generalization of the unmodified velar obviously is not completely excluded either. We have seen that it is credibly attested – though always alongside generalization of the modified velar – in nouns and pronouns with stems originally ending in *ՌgՌ or *ՌxՌ (see section 2.4). The dialects where restoration of *ՌgՌ or *ՌxՌ is found suggest a unified phenomenon having its focus in the Russian/Belorussian area.31 30

Cf. the paradigms reconstructed by Zubatý (1910: 151) and Mareš (1956: 467 = 1965a: 40, but not, unfortunately, in the English translation 1965b: 35 or thereabouts). 31 For an attempt at explanation see Vermeer (MS).

528

WILLEM VERMEER

3.8 General points having to do with the elimination of the alternations caused by the blocking rule In theory, the alternations caused by the blocking rule could be used to explain all – not just a few – counterexamples with unmodified velars in stem-final position by assuming that they too were subject to Prog, but that in their case it happens to be the unmodified velar that was subsequently generalized (“restored”). Although this approach would seem to be possible in principle, investigators have generally shrunk back from adopting it wholesale, reserving it for special cases involving *g and *x, such as the restoration of stem-final *g in nominal paradigms found primarily in the Russian/Belorussian area (see section 2.4). The reasons for the reluctance have rarely been made explicit, perhaps never. I suspect that scholars have tended to avoid the arbitrariness that would result if the principle was to be adopted wholesale. The important thing to keep in mind here is merely the existence of a tradition by which counterexamples are not ascribed to restoration of the unmodified velar except in a limited set of examples attested outside OCS. The most striking exception is Trubetzkoy, whose case we shall have occasion to examine below (section 5.4). Doublets like *Njga/*NjdǺa have bothered investigators who have not realized that the blocking rule near-ineluctably gives rise to alternating paradigms, followed quite naturally by analogical generalization of either the modified or the unmodified velar. Some have gone so far as to intimate that Prog is too irregular to be amenable to explanation by traditional means (e.g. Bernštejn 1961: 209). Since the previous existence of intermediate paradigms – which is inescapable given Meillet’s *ՌLJՌ and the blocking rule – provides a mechanism that explains the doublets, this view is not warranted, as has been stressed notably by Tadeusz Milewski (1937: 13). 4. Lehr-Sp’awiǝski (1911) Tadeusz Lehr-Sp’awiǝski’s “Nowsze poglƾdy na niektóre zjawiska t. zw. drugiej palatalizacji” of 1911 is the earliest publication aiming at a comprehensive treatment of Prog in all its aspects. In it, a new piece of presumed evidence for Pedersen’s chronology is presented. In addition, a difficulty inherent in earlier treatments of Pedersen’s chronology is solved by combining the latter with the classic view in a way that has been influential to the present day. Both points demand our attention here. 4.1 Lehr-Sp’awiǝski’s evidence for Pedersen’s chronology (Vsg otӸǁe) As we have seen (section 3.3), the only difference between the intermediate and the soft paradigms remaining in the language of the earliest texts – pronouns

PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION

529

apart – is the msc Vsg: otӸǁe vs. cLJsarju. Lehr-Sp’awiǝski holds that these forms refute Zubatý’s assumption that the intermediate inflections were transformed under the influence of the soft paradigms (Zubatý 1910: 152-153), because if that were true the soft Vsg ending would have been adopted too, so that we would find such forms as **otӸcju, **kԔnNjdzju (Lehr-Sp’awiǝski 1911: 145). The argument appears to have had no influence on the later course of the debate and it is not difficult to see why: since analogical processes cannot be counted on to always reach their natural conclusion, the point is not particularly compelling. Moreover, as Vondrák points out, the notoriously isolated position of the vocative in nominal paradigms may have sufficed to shield it from the influence of the soft inflection (Vondrák 1923-24: 22n.).32 4.2 Lehr-Sp’awiǝski’s interpretation of Pedersen’s chronology Developing hints by Šǁepkin (1910: 215), Lehr-Sp’awiǝski shored up Pedersen’s chronology by solving a major problem it left open, which is that of specifying how, why and when Reg2 and Prog came to have the same reflexes in the attested material (see section 3.1). Lehr solved the problem by splitting Prog up into two stages. He assumed that an early stage of palatalization – *k > *֦ etc. – arose before Umlaut took place and that palatalization remained in the language as a synchronic rule until after Mono; it goes without saying that the new palatal consonants umlauted following vowels; when subsequently Mono took place, velars automatically became palatalized if followed by the new front vowels Mono gave rise to; at a later stage all palatalized velars were assibilated, i.e. *֦ > c etc. (1911: 146). This yields the chronology shown in Table 6. From a general point of view, Lehr-Sp’awiǝski’s interpretation of palatalization as a synchronic rule that persisted in the language for a long time is comparable with Meillet’s approach to Umlaut (see section 3.2). Whoever rejects that kind of rule as a matter of principle – and in informal settings I have talked to otherwise well-informed people who do that – should at least be consistent and reject both Meillet’s understanding of Umlaut and Lehr’s take on Pedersen’s chronology. However, it is fairly fruitless to deny the possibility of such rules given the fact that some of them have been known to persist in many languages for lengthy periods of time, e.g. the voice assimilation in obstruent clusters that has been more or less universal in Slavic ever since the fall of the weak jers. 32 Since the original oՌstem Vsg ending is Ռe, the Ռu found in the soft paradigm is the outcome of an analogical substitution, which may well have been later than the substitution of ՌLJՌ by ՌiՌ in the Intermediate Inflections. If that was the case, there must have been a stage characterized by the absence of any difference between the intermediate inflection and the soft paradigm even in the case of the msc oՌstems.

530

WILLEM VERMEER

Table 6. Lehr-Sp’awiǝski’s interpretation of Pedersen’s chronology 1 Palatalization phase of Prog (k > ֦ etc.). 2 Rise of Umlaut. 3 Mono. 4 Palatalization phase of Reg2. 5 Assibilation of the palatalized velars (֦ > ǀ etc., both Reg2 and Prog). The mechanism devised by Lehr-Sp’awiǝski improves on Pedersen’s scheme by accounting in a natural way for the fact that Reg2 and Prog have the same reflexes. However, it fails to generate the correct output (Meillet’s ՌLJՌ), nor does it answer the fundamental question of what is gained by separating Reg2 and Prog into two distinct innovations (see section 1.3). So ultimately it fails to deliver in the way the default chronology incorporating Meillet’s conception does (see section 3.2). 4.3 Lehr-Sp’awiǝski’s interpretation as a version of the default chronology Lehr-Sp’awiǝski’s conception combines Pedersen’s chronology (first stage) with the classic view that identifies Reg2 and Prog as a single innovation that took place at some point after Mono (second stage). From a structural point of view it can be regarded as a version of the default chronology that has partly been extended backward in time in order to account for the Ռi of such forms as Pedersen’s Lsg lici and otӸci in terms of a phonological development. By choosing the label of “palatalization” for the first stage, the suggestion is conveyed that that is the stage that counts. This terminological tradition was to cause some confusion when Lehr-Sp’awiǝski’s conception of Pedersen’s chronology came to be adapted to a Prague School phonological framework (see further section 8.2). 5. Trubetzkoy (1922) and the treatment of PIE *u (attested *Ԕ) Trubetzkoy’s “Essai sur la chronologie de certains faits phonétiques du slave commun” of 1922 is the most substantial exposition of Pedersen’s chronology to have appeared after Lehr-Sp’awiǝski (1911). The novel piece of presumed evidence for Pedersen’s chronology Trubetzkoy introduces in it is closely bound up with his theory about the history of the vowel system, which therefore will have to be discussed too.33

33

For some discussion of the position of the “Essai” in the general context of the debate on Prog see also Vermeer (2003: 425-428).

PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION

531

5.1 Trubetzkoy’s evidence As in the case of Pedersen’s argument for an early placement of Prog, Trubetzkoy’s is intended to show that Prog preceded the rise of Umlaut, which, given the uncontested skeleton chronology presented in section 1.5 (Table 1), implies that it preceded Mono, hence also Reg2. In Trubetzkoy’s view, Prog must be earlier than the rise of Umlaut because *Ӹ reflecting umlauted *Ԕ does not trigger palatalization, as in *Ӹgo ‘yoke’, OCS igo < earlier *juga (cf. Beliǀ 1921: 34).34 In Trubetzkoy’s own words, “les gutturales restent toujours inaltérées lorsque le ǒ [i.e. Ӹ] inaccentué qui les précède remonte à un Ǯ [i.e. Ԕ] après consonne mouillée” (230).35 Trubetzkoy’s argument has left few traces in the literature apart from one or two bare references (e.g. Holzer 1995: 248). To the best of my knowledge it has never been explicitly discussed or evaluated. What follows is an attempt to make up for that defect. 5.2 Umlaut and *u: the problem with Trubetzkoy’s evidence For Trubetzkoy’s argument to hold it is crucial that the reflex of PIE *u (attested Ԕ) merged with the reflex of *i (attested Ӹ) immediately on being umlauted, for if the two reflexes remained distinct only to merge later, the reflex of *u may not have triggered Prog while that of *i did at some stage after the rise of Umlaut, thereby fatally undermining the argument. So the first question to ask is whether or not that condition is met. The answer would seem to depend on the phonetics of the reflex of *u at the stage involved. After all, if an [u]-like vowel is fronted, it ordinarily becomes [ü]-like, remaining distinct from any [i]-like vowel present in the system until such moment as it sheds the roundedness that still differentiates it from [i]. In the reconstruction of the history of the Slavic vowel system presented in the “Essai”, *u is assumed to have lost its rounding at a stage that was early enough for it to merge with the reflex of *i in Umlaut position immediately upon the rise of Umlaut. It would appear that Trubetzkoy’s own reconstruction supports his argument for Pedersen’s chronology. Which raises the next question: is the reconstruction plausible enough to build on? And it is here that trouble brews. On closer examination, the reconstruction turns out to depend crucially on an assumption about the conditions triggering Prog that is unwarranted.

34

I have substituted *Ӹgo, which is an uncontroversial example, for Trubetzkoy’s own material, which is drawn from modern Russian and of unproven relevance. 35 In sections 5.1-13, bracketed numbers refer to the page numbers of the “Essai”.

532

WILLEM VERMEER

5.3 Trubetzkoy’s solution to the problem of the conditions that triggered Prog As we saw earlier (section 2.3), an important challenge posed by Prog is how to explain the counterexamples. The most obvious way to meet it is by fine-tuning the formulation of the conditions under which Prog is held to have been triggered. By the time Trubetzkoy was writing, investigators had experimented with four major fine-tuning mechanisms, of which he puts three to work: a. Together with Šaxmatov (1896: 703-705), Zubatý (1910: 150-151), and most later investigators, he assumes that palatalization was blocked if the velar was followed by certain back or rounded vowels (227-228). This is the blocking principle discussed in section 3.7. Apparently unaware of his predecessors’ work, Trubetzkoy advances a highly idiosyncratic inventory of blocking vowels, comprising y, u, ՛, and LJ reflecting *oi. What is unusual here is the absence of *Ԕ and the inclusion of LJ reflecting *oi. Both features are cornerstones of his conception and will be examined in detail below (sections 5.2-8, 11-12). b. Together with Meillet (1902-05: 336), Lehr-Sp’awiǝski (1911: 146-147), and most later investigators, Trubetzkoy assumes that Prog was triggered not by all instances of preceding *i, *Ӹ and *Nj, but by a subset of those vowels. Prog is held to be regular after *i reflecting earlier *ǔ, after *Ӹ reflecting *i, and after *Nj reflecting *in, but not after *i reflecting *ei or umlauted *ǰ, after *Ӹ reflecting umlauted *u, and after *Nj reflecting *en (226-227). Trubetzkoy was the first to add *i reflecting earlier *ei to the roster. Unfortunately he does not argue the point. Subsequently several scholars have examined the – of necessity skimpy – evidence. Most of them have concluded that it supports Trubetzkoy (Meillet 1934: 92, Milewski 1937: 9, Vaillant 1950: 53, Lunt 1981: 16-19), some that it does not (Shevelov 1964: 344, Steensland 1975: 92). The point will be taken up in section 7.2. c. Together with Baudouin de Courtenay (1893: 15-16; 1894: 48), Pedersen (1907: 216-217), and Meillet (1910: 371-372), but no later authors worth speaking of (not even Meillet 1924: 83), Trubetzkoy assumes that the place of the stress played a role, more specifically that for palatalization to take place it was necessary for the syllable preceding the velar to be unstressed: “L’essentiel est que, pour devenir palatale, la gutturale devait être précédée immédiatement par une syllabe inaccentuée” (226, Trubetzkoy’s italics). In addition Trubetzkoy takes recourse to a fourth mechanism intended to explain counterexamples, this time not by fine-tuning, but by exploiting the morphological consequences of the blocking rule:

PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION

533

d. Following Zubatý (1910: 152), Trubetzkoy realizes that the alternations of modified and unmodified velars caused by the blocking rule may have been eliminated in favour not only of the former, but also of the latter (see section 3.7). He uses this insight to account for a much broader range of cases than is traditional, most strikingly suffixal ՌnikԔ and ženixԔ: “les thèmes masculins en Ռǔko (starikԔ, sԔtӸnikԔ) et en ǔxo (ženichԔ, russe ӊӃԀӊѹѽ, etc.) … ont généralisé les gutturales inaltérées” (229). 5.4 Discussion Granted that complete consensus about conditions may well be forever out of reach due to the limitations of the evidence, it does not follow that just any solution can be justified. Several points in Trubetzkoy’s account are arguably objectionable, beginning with the list of blocking vowels. The presence on it of LJ reflecting *oi has to do with Trubetzkoy’s attempt to account for Meillet’s LJ and will be taken up later (sections 5.11-12). The sole reason given for assuming that *Ԕ did not block palatalization is that, as Trubetzkoy puts it, “Vieux-slave nicӸ ( a, *ǰ > y), but left their short counterparts untouched; it goes without saying that umlauted *ǰ (i.e. *٤) was delabialized as well, merging with its front unrounded counterpart (*ǔ), giving rise to the attested facts (223, 225). Finally the inherited diphthongs were monophthongized and sequences consisting of vowels and tautosyllabic nasals developed into nasal vowels (225). Skipping issues that are irrelevant from the point of view of Prog, Trubetzkoy’s theory comprises the chronological steps shown in Table 7.

536

WILLEM VERMEER

Table 7. Trubetzkoy’s chronology 1 Merger of PIE *Ǡ/Ǣ with *Ƽ/ƽ (> *Ǡ/Ǣ). 2 Rise of a difference between two types of rounded back vowels (here types (A) and (B) respectively). 3 Prog. Palatalization is blocked by type (B) vowels and *oi. 4 Rise of Umlaut. In Umlaut position, type (A) vowels merge with their unrounded counterparts, whereas type (B) vowels retain their contrastiveness until further notice. 5 Delabialization of long rounded back vowels: *Ǣ > a, *ǰ > y; as a consequence the reflex of *ǰ now merges with that of *ǔ in Umlaut position. 6 Mono; rise of nasal vowels. 7 Reg2. 5.6 Excursus: Meillet’s chronology and Trubetzkoy’s view of umlauted *ǰ As we saw just now, Trubetzkoy assumes that umlauted *ǰ did not merge with *ǔ immediately upon the rise of Umlaut (stage 4), but only at the later stage when it lost its rounding (stage 5). This eliminates a problem implicit in all scenarios that put Prog after the rise of Umlaut, such as the classic or default chronology and the Jagiǀ view (by which Prog is later than Reg2). In the second volume of his Études sur l’étymologie et le vocabulaire du vieux slave, Meillet points out that the retained stem-final velars in the counterexamples ՛žika and bližika (listed in section 2.3) can be explained by assuming that these words reflect suffixal formations in *ՌjՌǰka and that *ǰ in Umlaut position had not yet merged with *ǔ when Prog took place, and left following velars untouched (Meillet 1902-05: 336). If Meillet is right, *ǰ and *ǔ were still distinct in Umlaut position at the stage when Prog took place. Given Meillet’s chronological identification of Reg2 and Prog this means after Mono, hence after the rise of Umlaut. Trubetzkoy’s interpretation of *ǰ shows that from a phonetic and phonological point of view this is unproblematic, on condition that *ǰ can be assumed not to have been delabialized until after Prog took place. On the basis of this, the chronology given in Table 3 in section 3.2 (which incorporates Meillet’s view of Prog and Umlaut) can be expanded as in Table 8. At this stage of the discussion the point is more neat than compelling because the evidence consists of two closely related words of purely local significance, which may conceivably have arisen after Prog took place (section 2.3). However, as we shall see presently (section 5.7) what holds for *ǰ, holds for its short counterpart *u as well, which is found in material of more solid standing.

PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION

537

Table 8. Expansion of Meillet’s conception (see also Table 3) 1 Rise of Umlaut (as a synchronic rule); in Umlaut position, *ǰ becomes [‫]خ‬, remaining distinct from *ǔ. 2 Mono. 3 Prog (simultaneous with Reg2). 4 Delabialization of *ǰ to y, as a consequence of which umlauted *ǰ, up to now realized as [‫]خ‬, loses its rounding too and merges with *ǔ. 5 Cessation of Umlaut. 5.7 The position of short *u (attested Ԕ) in Trubetzkoy’s system We have seen that Trubetzkoy distinguishes between two types of rounded back vowels in order to capture the differential treatment he claims to observe between those that blocked palatalization and remained distinct in Umlaut position on the one hand (B), and those that did neither on the other (A). We have also seen that in Trubetzkoy’s scheme of things *u – attested Ԕ – was a member of (A) whereas its long counterpart *ǰ – attested y – belonged to (B). What is important here is that it is only by classifying *u as a member of (A) that the treatment of *u in Umlaut position can be regarded as evidence for Pedersen’s chronology: since type (A) vowels merged with their unrounded front counterparts – in this case *i – the moment Umlaut arose, Prog must have preceded the rise of Umlaut. Since type (B) vowels remained distinct from their unrounded front counterparts, the argument gets into trouble the moment there is any doubt which group *u belongs in. Trubetzkoy’s view of *u as a type (A) vowel rests on the assumption that *u did not block Prog, an assumption we have seen is unwarranted (section 5.4). And if it is true that *u blocked Prog, it behaves in both diagnostic respects – Prog and Umlaut – exactly like its long counterpart *ǰ, revealing itself as a type (B) vowel, in which case it must have remained distinct from *i (> Ӹ) in Umlaut position just like its long counterpart *ǰ, so that the unmodified ՌgՌ of igo < *juga can be attributed to the same mechanism as the one in bližika and ՛žika (see section 5.6). Transferring the reflex of *u from type (A) to type (B) has more advantages than just accounting better for the low-level facts of Prog. It also frees the reconstruction of a crippling inconsistency in the way it pictures the stage at which Prog and Umlaut took place. The point is the following. While the reflex of *u is assumed to have been delabialized by the time Umlaut arose, thereby shifting from (B) to (A), it is also assumed to have exerted a rounding – and raising – effect on preceding *o strong enough to cause the latter to shift from (A) to (B).

538

WILLEM VERMEER

In other words: the reconstruction implies that *u went on exerting a rounding – and raising – effect on preceding *o even after it had been delabialized – and somewhat lowered – itself. That obviously is not plausible without strong corroborating evidence, which is not presented. So if *u is transferred from (A) to (B), Trubetzkoy’s reconstruction is improved spectacularly: it accounts better for the facts of Prog; it becomes more symmetrical because now *u and its long counterpart *ǰ are seen to behave alike; it gets rid of the anomaly by which *u is assumed to have exerted a rounding effect on preceding *o even after it had been delabialized. Since Trubetzkoy’s argument for Pedersen’s chronology depends entirely on the assumption that *u was a type (A) vowel by the stage involved, it dissolves the moment *u is seen to have belonged to type (B). 5.8 Trubetzkoy and the evidence for an [a]-like value of early Slavic *o Trubetzkoy’s assumption that his long *Ǣ – attested a – was rounded is based on the belief that short *o was. The reasoning behind that is that since short *o was rounded, its long counterpart *Ǣ had necessarily to be rounded as well because, as he puts it, “on serait bien embarrassé pour citer une langue qui transforme Ƽ en Ǡ sans altérer ƽ en Ǣ” (221). This piece of reasoning is not at all compelling. First from the material side, because it does happen from time to time that short /Ƽ/ becomes differentiated from its long counterpart by being retracted and ending up [o]-like. The short /a/ of Hungarian comes to mind. But also qua reasoning: even if it were true that a short /Ƽ/ cannot become [o]-like without its long counterpart behaving alike, there still is the possibility of a secondary development severing the link in such a way that the long vowel is lowered again subsequently with its short counterpart staying behind, as Trubetzkoy himself assumes happened in Slavic as a consequence of his delabialization rule (see section 5.5). Trubetzkoy was aware that non-Slavs perceived Slavic *o as [a]-like until relatively shortly before the rise of the OCS tradition. As is well known, an [a]like stage of Slavic *o is testified to by all languages Slavic interacted with while its geographical range was expanding, e.g. Finnic, Baltic, Germanic, Latin, Greek, Albanian, and Proto-Bulgarian. Nevertheless he decides not to take this evidence at face value, adducing two considerations (219-220): 1. Following Vasmer (1907) he assumes that the Slavic *o was rounded, but too low (open) to correspond to the various kinds of o of the various languages Slavic interacted with. 2. Adducing such examples as Finnish papu ‘bobӆ’ and turku ‘tӆrgӆ’, he argues that final Ռu here shows that the Slavic archetypes of these loans ended in an [u]-like vowel at the time and comments: “Or, ce *ՌǮ final slave remontant à *Ռon (et *Ռos?) est certainement plus récent que le changement de

PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION

539

*a en *o: le prototype slave de finn. papu n’avait pu être que *bobǮ (et non *babas)” (220). The first point shows at best that it is not completely excluded for the reflex of *o to have been rounded despite the loanword evidence; it does not even begin to show why that view is to be preferred over the assumption that the reflex of *o was in fact [a]-like at the time, hence not rounded. The second point substitutes the word “certainement” for the use of reasoned argument. As far as I can see, there is nothing particularly certain about the point, which is the raising of *o found in the attested reflex of word-final *Ռon, a subject that has exercized the minds and the emotions of scholars interested in Auslautgesetze up to the present day without very much of a consensus being in sight.37 Trubetzkoy’s insistence on this point is all the more curious because nothing in his reconstruction crucially depends on the assumption that the products of the merger of PIE *o/Ǣ and *a/ƽ were rounded – i.e. [o]-like – throughout the early history of Slavic. Everything of value remains the same if unrounded – i.e. [a]like – values are preferred. Better still, this creates the opportunity to reinterpret the difference between (A) and (B) as just the absence – in low vowels – or presence – in non-low ones – of rounding, rather than the presence of two distinct types of rounding added to a difference in tongue height: a. The type (A) vowels are [a]-like, i.e. low unrounded back vowels reflecting the outcomes of the merger of PIE *o/Ǣ and *a/ƽ. The absence of rounding accounts both for the behaviour of these vowels and for the loanword evidence. b. Short *o, though [a]-like in most positions, receives rounding, i.e. becomes [o]-like, whenever it is followed by *u or a tautosyllabic nasal, in that way behaving like a type (B) vowel. c. The remaining type (B) vowels are high rounded *u and *ǰ. 5.9 The uncoupling of long and short vowels The moment it turns out that *Ǣ may not in fact have been rounded to begin with, any justification for a delabialization rule that jointly affected *Ǣ and *ǰ disappears. That is good news in its own right because that rule is obviously problematic. To begin with is it mysterious from a phonetic point of view: why would length, 37

See, for instance, Kortlandt (1978) and the relevant passages of Orr (2000) and Halla-aho (2006).

540

WILLEM VERMEER

of all things, predispose rounded vowels to lose rounding? (Trubetzkoy does not argue the point.) In addition it gives rise to a needlessly atomistic view of the uncoupling of long and short vowels. In section 1.4 the fact was mentioned that the inherited pairs of long and short vowels of Early Slavic have been uncoupled in attested Slavic. Trubetzkoy distributes the uncoupling among (at least) three distinct processes: 1. An early uncoupling of *u and ǰ which caused the former to end up as a type (A) vowel and the latter as type (B). Trubetzkoy justifies this by referring to the attested reflexes of *u, which are weakly rounded if they are rounded at all and somewhat lowered as compared with long *ǰ (222-223). The lack of consistency is striking here: if the attested reflexes of *u are held to determine the reconstruction, why not do the same with *ǰ, which never (or hardly ever) is a rounded vowel in the attested material? Note that wordinitial *u, exactly like word-initial *ǰ, ultimately develops a prothetic v-, which is usually regarded as proof of the presence of rounding at the stage involved, a consideration that was brought into the debate by Lehr-Sp’awiǝski (1911: 144), whose insightful discussion of Pedersen’s chronology Trubetzkoy ignores completely along with nearly all of the rest of the literature. 2. The uncoupling of *o/Ǣ and *u/ǰ produced by the delabialization rule. 3. The uncoupling that ultimately causes the unrounded front pairs *e/lj and *i/ǔ to be reflected as e/LJ and Ӹ/i, which Trubetzkoy passes by in utter silence. Given the parallellism between the reflexes of the rounded back vowels (*o/Ǣ and *u/ǰ) and those of their unrounded front counterparts (*e/lj and *i/ǔ), it needs no arguing that this fragmentation of the uncoupling process is an unacceptable infringement of Ockham’s razor as long as no evidence has been presented showing that it has explanatory advantages. Trubetzkoy provides no evidence, leaving it to the reader to sort out what is going on and to discover, for instance, that the uncoupling of *e/lj and *i/ǔ is not accounted for at all. If, then, there no longer is any need to connect the fronting of *ǰ with the history of Trubetzkoy’s *Ǣ (other scholars’ *ƽ), the way is free to assume that short *u was treated exactly like long *ǰ until the uncoupling of long and short vowels took place, implying that short *u was type (B) in the terms of Trubetzkoy’s system, so that it remained distinct from *i after the rise of Umlaut until both *u and *ǰ were delabialized, ultimately to yield Ԕ and y respectively.38

38

Independently from Trubetzkoy a tentative rule treating the delabialization of *ǰ and *Ǣ as a single innovation was also proposed by Bubrix (1923: 284). From Trubetzkoy the rule migrated to Jakobson (1929: 27).

PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION

541

5.10 Repairing Trubetzkoy’s reconstruction In the light of the above considerations, the problematic sections of Trubetzkoy’s chronology can be amended as shown in Table 9. Table 9. Trubetzkoy’s chronology (repaired version) 1 Merger of PIE *o/Ǣ with *a/ƽ in a pair of unrounded low back vowels, which are however rounded by a following tautosyllabic *u or nasal. The high vowels *u/ǰ are rounded in all positions. 2 Prog. 3 Rise of Umlaut. Whereas *o, which is an [a]-like vowel lacking rounding, merges with *e in Umlaut position unless a following syllable-final nasal consonant or *u imparts rounding to it, the rounded back vowels *u and *ǰ retain their distinctiveness until further notice, being realized approximately as [ü] and [‫ ]خ‬respectively. 4 Delabialization of high rounded back vowels. As a consequence the reflexes of *ǰ and *u now merge with those of *ǔ and *i in Umlaut position. 5 Uncoupling of long and short vowels; Mono; other late developments, such as Reg2. The principal difference with the original conception is that short *u is treated on a par with its long counterpart *ǰ. The distinction between unrounded ([a]like) and rounded ([u]-like or [o]-like) back vowels corresponds with Trubetzkoy’s distinction between two types of rounded back vowels. All this shows that it is easy to repair Trubetzkoy’s reconstruction in such a way as to get rid of all features that are objectionable from the point of view of Prog and the history of *u and that have to be eliminated if the reconstruction is to retain some measure of viability. If one does so, however, the presumed new evidence for Pedersen’s chronology disappears. 5.11 Trubetzkoy on Meillet’s LJ As we have seen (section 5.3), Trubetzkoy assumes that *oi blocked Prog. To the best of my knowledge this is the only attempt to reconcile Meillet’s ՌLJՌ with Pedersen’s chronology apart from Pedersen’s own (for which see section 3.4). Trubetzkoy writes that in the pronoun vӸsӸ, “l’ancien *oi (=LJ) ne s’est pas transformé en *ei (=i), ce qui prouve que leur *x ne s’est palatalisé en ǧ que sous l’influence du LJ suivant [i.e. Reg2], et non à cause du ǒ précédent [i.e. Prog]. Toutes les autres formes du même pronom […] présentent au contraire une palatalisation provoquée par ce ǒ précédent. A une certaine époque ce pronom devait donc présenter au locatif singulier *vǒxփomǒ (avec xփփ prépalatal) et à l’instrumental *vǒxoimǒ (avec x guttural)” (228).

542

WILLEM VERMEER

Although this solution yields the correct output, this success comes at a forbidding price: since Trubetzkoy agrees with most post-Zubatý investigators in regarding *o as a non-blocking vowel (section 5.3), the blocking effect of *oi has to be attributed somehow to the second component of the diphthong, which, being a high front vowel, is in an unfavourable position to exert such an effect. Trubetzkoy is understandably reluctant to handle the anomaly in terms of his interpretation of rounded back vowels: “Il n’y a aucune raison de croire que le o de la diphtongue oi eût été plus tendu (“fermé”) que dans tout autre position” (228). The reason for the reluctance is – presumably – that umlauted *oi became *ei (ultimately to be reflected as i), showing that the first element of the diphthong belonged to his type (A) at the rise of Umlaut. The only way out is to devise an ad hoc development limited to *oi: “on pourrait bien se figurer que, tout en restant relâché, cet o comme premier élément de la diphthongue oi présentait un plus fort degré de labialisation que dans d’autres positions: c’était peut-être quelque chose comme uoi. Après le changement de eu en jou, la diphthongue oi était la seule en slave commun qui comportât deux éléments hétérogènes, le premier labiovélaire (o), le second palatal (i). Un renforcement «par contraste» du degré de labialisation de o n’aurait donc été que fort naturel pour cette diphthongue” (228). 5.12 Discussion In addition to being ad hoc, Trubetzkoy’s solution fits in poorly with the reconstruction as a whole. If *o as modified by following *u in *ou causes *ou not to merge with its front unrounded counterpart on the rise of Umlaut, why did not *uoi behave similarly? And if the initial element of *uoi resembled *u in being type (A), so that it underwent Umlaut, how was it possible for it to block palatalization? When all is said and done, Trubetzkoy simply fails to explain in an acceptable way how it could happen that *oi blocked Prog while undergoing Umlaut. Given the repaired version of Trubetzkoy’s chronology presented above (section 5.10), a slightly more down-to-earth modification of the mechanism he proposes to account for Meillet’s ՌLJՌ might look approximately as follows: at a certain stage before Prog, the first element of the diphthong *oi [ai] was retracted and rounded sufficiently to block palatalization of following velars. As a partial typological parallel one is reminded of the well-known development in Finnic that caused the stem-final vowel Ռa to be reflected as Ռo before tautosyllabic i in, say, the oblique cases of the plural (Finnish velka ‘debt’, but Inessive pl veloissa) and the preterite (kantaa ‘carry’, but pret 3sg kantoi), see, e.g., Posti (1966: 71) or any other historical phonology of Finnic. After Prog had taken place, but before the rise of Umlaut, the earlier [ai]-like realization of the diphthong was restored in order to make way for the devel-

PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION

543

opment of *oi into *ei in Umlaut position and to provide a credible startingpoint for its ultimate development into a low monophthong (> LJ, phonetically [ä]). Although this version of Trubetzkoy’s solution produces the correct output, it is ad hoc – hence methodologically vulnerable – because the putative shift is introduced solely to enable the first element of the diphthong to block Prog, and is reversed soon afterwards. And this may be the right moment to recall that the need to account for Meillet’s ՌLJՌ in this roundabout way is an artifact of Trubetzkoy’s ultimately unfounded decision to stick to Pedersen’s view putting Prog before Mono. 6. Zaliznjak (1982) and the absence of Reg2 in Novgorod A further piece of presumed evidence for Pedersen’s chronology derives from Andrej A. Zaliznjak’s work on the medieval Novgorod dialect. 6.1 The Novgorod evidence As is well known, a quarter of a century ago Zaliznjak contributed sensationally to the body of evidence relevant to the Slavic palatalizations by his discovery that Reg2 never took place in the variety of Common Slavic underlying the medieval dialects of Novgorod and Pskov (Zaliznjak 1982: 61-75 = 1986: 111-122, cf. 2004: 41-47). What is most interesting about Proto-Novgorod/Pskov is that Prog did take place in it, at least judging by the treatment of *k in such examples as Nsg z[a]dӸnicja ‘inheritance’ (607/562, probably last quarter of 11th century), Gsg pescja < *pӸsӸǀa ‘polar fox’ (724, sixties of 12th century). In view of what we saw earlier (section 3.7), it goes without saying that *g is different because analogical restoration of unmodified *g in what I have called intermediate paradigms appears to have been the norm in the Russian/Belorussian area, hence one expects to find it in Novgorod too. Unfortunately, evidence is so limited as to verge on the non-existent. Unmodified *g occurs in ne lego ‘it is not permitted’ (birchbark document 855, middle of 12th century), where lego stands for lӸgo in accordance with spelling conventions that are well understood (Zaliznjak 2004: 21-25). The word is frequent in other texts with a Novgorod background and occurs in modern dialects as well (Zaliznjak 2004: 46-47, 326). It indirectly continues the isolated Lsg *lӸdǺLJ found in OCS and elsewhere (see sections 2.6 and 3.3). Note that *lӸdǺLJ must have had a modified velar even in Novgorod, as a consequence of Prog. Attested lego is a nominative that replaced the inherited Lsg after the latter had ceased being recognizable as a locative, on the analogy of možno and similar words. Its stem-final ՌgՌ would seem to be in the spirit of what we find in the remainder of the Russian/Belorussian area in words like jaga and pol’ga.

544

WILLEM VERMEER

Unsurprisingly, modified *g is attested in *kԔnNjdǺӸ, which is not indicative owing to its meaning (see section 2.4). For *x the Novgorod evidence is limited to the pronoun *vӸǧӸ ~ vӸxԔ, which has stem-final *x (see also section 2.4). There is nothing surprising about restoration of ՌxՌ in alternating paradigms generated by the blocking rule. We have seen that restored *x is nearly pan-Slavic in the case of *elӸxa (*olӸxa) and widespread in Russian in the phytonym *vӸǧӸ/vӸxԔ (section 2.4). If we look at the evidence in the light of the elimination of the intermediate paradigms, the differences between Novgorod and its southern neighbours turn out to be limited to the morphological history of the pronoun *vӸxԔ.39 Alternatively these facts have been interpreted as showing that Prog did not affect *x in the Novgorod area at all, operating only on *k and *g (e.g. Andersen 1998: 590-591). Although I do not agree with that, I am not going to argue the point because nothing in what follows depends on it. What is significant about Novgorod is that the evidence would seem to indicate that Reg2 did not operate at all whereas Prog can be shown to have operated at least in the case of *k. On the basis of the Novgorod facts, Zaliznjak concludes: “Jasno, ǁto v tradicionnoj diskussii o tom, kakaja palatalizacija – vtoraja ili tret’ja – proisxodila ran’še, vyjavlennye zdes’ fakty javljajutsja sil’nejšim argumentom v pol’zu togo, ǁto tret’ja palatalizacija starše vtoroj” (1982: 73 = 1986: 119). Or, in different words and under a slightly different angle: “Oǁevidno, predki kriviǁej (po krajnej mere severnyx) otdelilis’ ot osnovnogo massiva slavjan ran’še, ǁem osušǁestvilsja process vtoroj palatalizacii” (Zaliznjak 1988: 166). 6.2 Discussion Zaliznjak’s argument is limited to the feeling that things are too clear to need spelling out: “Jasno, ǁto ...”, “Oǁevidno, ...”. The underlying reasoning is not made explicit, nor are the relevant mechanisms specified. I suspect the tacit premise is that innovations found in all of Slavic are necessarily older than innovations limited to certain areas. It needs no arguing that that view is not compelling. Every beginning historical linguist knows that innovations affecting only part of a linguistic territory can be followed by innovations embracing all of it. And that holds emphatically for the Novgorod dialect. The uncontested fact that it became differentiated from its neighbours at an exceptionally early stage did not prevent it from sharing numerous later innovations with them (cf. Vermeer 1997: 29-30 with references). 39

For the sociolinguistics of the elimination of *vӸxԔ see Vermeer (1997: 42-43). In recent years, examples of the type vӸsӸ have turned up in twelfth-century letters of a highly formal character (Zaliznjak 2004: 46), for the significance of which see also Vermeer (2005: 128 and passim).

PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION

545

Those who have adopted Zaliznjak’s view have not so far added anything of value, cf. the following statement by Michael Moser: “Zweifellos kann die Situation im Novgorodischen als ein wesentliches Argument für die Chronologie der “Zweiten” und “Dritten” Palatalisation aufgefaßt werden, nämlich als einer der Indikatoren dafür, daß die sogenannte “Dritte” älter als die “Zweite” Palatalisation ist. Während die Zweite Palatalisation nicht stattgefunden zu haben scheint, hat die “Dritte” Palatalisation Novgorod sehr wohl erreicht” (Moser 1998: 133).40 In the meantime, scenarios have been developed by which the Novgorod reflexes are accounted for without separating Reg2 and Prog at all (e.g. Vermeer 1986, 2000: 17-22, Andersen 1998: 590-591). As long as the Zaliznjak view has not been made explicit in terms of both reasoning and the mechanisms involved, and as long as it has not been shown to be preferable to alternative explanations that keep Prog and Reg2 together, it cannot be adduced in good faith as support for Pedersen’s chronology. 7. Pedersen’s chronology and the monophthongization of *ei As we have seen, investigators have tended to agree with Trubetzkoy that Prog is not found after *i reflecting *ei (section 5.3). Given certain assumptions about the evolution of the Proto-Slavic vowel system, this result can be construed as evidence for Pedersen’s chronology. Although I am not aware that the point has ever actually been advanced as such, it is so suitable a candidate for the role that I shall discuss it here as if it had. 7.1 No Prog after *ei: a potential piece of evidence There is a widespread tradition of assuming that Mono and the merger of *ei and *ǔ constitute a single event and that there was no intermediate phase, however brief, with the reflex of *ei being a monophthong distinct from *ǔ. If it is true that Prog took place at a stage where *ei and *ǔ were distinct, this interpretation of the merger of *ei and *ǔ implies that Prog preceded Mono, hence also Reg2. This result gives rise to a contradiction because, as we have seen, Meillet’s LJ implies that Prog, like Reg2, is later than Mono (section 3.2). What is going on here? 7.2 No Prog after *ei: a note on the evidence The principal problem with the primary evidence, as with all evidence relevant to Prog, is that it consists of a small number of items, a large part of which are ambiguous. As Vaillant puts it: “Il est sans doute conjectural dans beaucoup de 40

The same type of reasoning appears to be implicit in Hock (1998: 27).

546

WILLEM VERMEER

cas de distinguer les deux origines de sl. i, mais dans les cas les plus sûrs on voit que la palatalisation apparaît après un ancien *ǔ, et non après *ei” (1950: 53). Vaillant adduces two examples: nicӸ with *ǔ and lixԔ with *ei. I think that the situation, though not exactly overflowing with milk and honey, is just a shade less bleak than that. Prog after ՌiՌ is found in five of the items listed in section 2.1: 1. The suffix Ռica has traditionally been interpreted as suffixal *Ռka appended to a formation in *Ռǔ comparable with the one found in Skr vִkǔՌ and related formations (Rozwadowski 1897). 2. The adj. nicӸ has always been identified with Latvian and Sanskrit words reflecting *nǔkՌ < earlier *niHkwՌ.41 3. The pronoun sicӸ is a Slavic innovation containing the demonstrative pronoun sӸ plus the suffixal ՌkՌ also present in takԔ and related words; Meillet assumed – plausibly – that the i reflects long monophthongal *ǔ because the model takԔ has long *ƽ (1902-05: 329; on the demonstrative pronoun see Meillet 1934: 439-440). 4. Verbs with aorist stems in ՌaՌ are difficult to evaluate because some of them have received their modified velar analogically (see section 2.1), giving rise to the problem of determining the original set of verbs displaying Prog. Seven OCS verbs have ՌiՌ preceding the reflex of the velar. Unfortunately Ռšticati, which is attested – once – in zašticati ‘hint, allude’, cannot be used because both its status and its etymology are unclear. The six that remain are all either derived imperfectives or primary verbs correlating with perfectives in ՌnՌ. These are: Ռricati ‘say’, Ռžidzati ‘kindle, ignite’, Ռstridzati (alongside Ռstrigati) ‘cut (hair), shear’, klicati ‘call’, Ռmidzati ‘wink’, dvidzati ‘move, shift’. The derived imperfectives Ռricati, Ռžidzati and Ռmidzati have a secondary lengthened grade built on instances of ՌӸՌ occurring in the stem of at least some forms of the correlating perfective verb, so in their case it is reasonable to reconstruct long monophthongal *ǔ. Unfortunately Ռricati and Ռžidzati occur only in South Slavic and since elsewhere we find the regular lengthened grade *lj, which is also attested in OCS (e.g. prLJrLJkati, prižagati), they look like local South Slavic innovations possibly too recent to be relevant to Prog.42 Since the etymology of strigՌ and dvigՌ is too uncertain to tell what the ՌiՌ reflects here, whereas klicati, though likely to have *ǔ (judging 41

Jussi Halla-aho (2006: 99) reconstructs *neikՌ, without however explaining what is gained by separating Slavic from Latvian and Sanskrit here, and – more generally – what makes his solution superior to the traditional interpretation, which he does not even mention. (I am indebted to Petri Kallio for bringing Halla-aho’s book to my attention.) 42 Note that modern Czech ǥíkati is ambiguous because long ՌǔՌ reflects both *i and *LJ, cf. sníh ‘snow’. Old Czech has ՌieՌ, unambiguously pointing to *rLJkati.

PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION

547

by Lith klׁkti, klyknóti etc., all ‘call, scream’), happens to be onomatopoeic, we are left with Ռmidzati as the only more or less credible example of the type. Note that *midǺati is among the items attested with a modified velar in West Slavic (see section 2.4). 5. The etymology of lice is controversial. There is an old tradition of identifying the stem with the element *laikՌ attested in OPr (Elbing) laygnan ‘cheek’ (Nesselman 1873: 87 and many later investigators, including most modern etymological dictionaries).43 If lice is related to the Prussian word it is much more likely to reflect *leikՌ than *lǔkՌ, providing what looks a counterexample to the hypothesis that Prog does not take place after the reflex of *ei. There is however an alternative tradition, also plausible, of interpreting lice as a loan from Germanic (the family of English like). If that is correct it must reflect *lǔkՌ, thereby ceasing to be a counterexample. A third possibility is sketched in Derksen (2005: 59): “in view of the effects of the progressive palatalization the latter [i.e. lice] must originally have had zero grade. The eՌgrade is probably analogical after *likԔ”. Although this is an intriguing idea that deserves following up, it is problematical as it stands because the details have not been filled in and because *likԔ, which is called upon to serve as the model, is too recent for that role (see section 2.3). The most informative examples of i not followed by Prog are the following: a. The adj. lixԔ is generally agreed to reflect *leikՌsՌ (e.g. Meillet 1934: 92). b. The closest etymological fit of ՌnikԔ is Latvian Ռnieks, which cannot reflect monophthongal *ǔ, but can reflect *ei (Meillet 1934: 92). c. The type tolikԔ is derived from stems that used to be inflected according to the Ռi-declension, cf. tolӸ (tolӸma, tolӸmi), ‘in that measure’, otԔtoli ‘since then’; for that reason the ՌiՌ is much more likely to reflect *ei than *ǔ (cf. Meillet 1902-05: 329; on velikԔ see also Meillet 1934: 92). 44 The assumption that *ǔ did and *ei did not trigger Prog explains the presence of Prog in nicӸ, Ռica, sicӸ and Ռmidzati, and its absence in lixԔ, the suffix ՌnikԔ and the type tolikԔ. On the other hand, if it is true that *ei did not trigger Prog, the traditional identification of lice with OPr *laikՌ cannot be correct as it stands. In that case, 43 In the past, Irish leaca ‘cheek’ was also considered to be related, providing at least the glimmerings of an Indo-European etymology (e.g. Stokes 1893: 172). However, the Irish word offers formal difficulties – as hinted at but left implicit by Pedersen (1913: 119) – and appears no longer to be regarded as Indo-European at all (cf., e.g., Stüber 1998: 44, 116-117, with thanks to Peter Schrijver for providing the reference). 44 For igo and the type bližika see sections 5.1 and 5.6 respectively. These examples have no bearing on the problem of *ei vs. *ǔ.

WILLEM VERMEER

548

the alternative assumption that it is a Germanic loan looks inevitable (cf. Vaillant 1958a: 238, Lunt 1981: 19), unless, of course, the traditional etymology will turn out to remain viable along the lines sketched by Derksen. Those who reject the assumption that *ei did not trigger Prog, usually do so for the following reasons: •

•

They assume that the stems *dvizՌ ‘two-year-old’ and *trizՌ ‘three-year-old’ attested in SCr and non-canonical CS reflect *dvidǺՌ and *tridǺՌ, to be derived from *dveigՌ and *treigՌ with Prog, hence as a piece of genuine counterevidence for which no alternative explanations are available as in the case of lice (e.g. Shevelov 1964: 344, Steensland 1975: 92). Since at present there are no reasons to reconstruct stem-final *ՌdǺՌ in these words, this view is unwarranted (see section 2.2). They assume that if Prog differentiated between *ǔ and *ei, the reflex of *ei must have been diphthongal at the stage involved, in which case its second element must necessarily have triggered Prog. This reasoning fails to take into account – first – the possibility that *i as the second component of a diphthong may have been somewhat lowered, as in the diphthongs of, say, German or Avestan, yielding something like [äe], thereby ceasing to be able to trigger Prog while remaining diphthongal (see Lunt 1981: 19-20), and – second – the possibility that the reflex of *ei remained distinct from *ǔ for a time even after it had been monophthongized, a point we’ll be returning to presently (sections 7.7-9).

7.3 Views of Mono Different scholars have viewed the elimination of the diphthongs differently. It is useful to be aware of the variety. To begin with, although it is customary to treat the elimination of the diphthongs as a single event, there may well have been an internal chronology, with some diphthongs being eliminated earlier than others. But so far no evidence has been adduced apart from the possibility that there may have been an internal chronology. Moreover, since Mono was part of the general tendency towards the elimination of closed syllables, chances are that the diphthongs were monophthongized closely together in time even if not absolutely simultaneously. In the absence of relevant evidence to the contrary the elimination of all diphthongs will be treated as a single event in what follows.45 45 Halla-aho (2006: 199-200) has argued that *ei was monophthongized earlier than the other diphthongs because of the raising of the reflex of *e followed by *j in such cases as trӸje < *trije < *treies. This would have some force only if there were reasons to assume that the raising preceded Mono, but Halla-aho has not argued the point.

PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION

549

Furthermore, it is possible for the monophthongization of individual diphthongs to have had an internal chronology, notably in the case of *oi, where the first element was back and the second front. The former may easily have been fronted with the diphthong remaining at a diphthongal stage, yielding, say, *[äi] (recall that *o had an [a]-like realization at the time, see section 5.8). Since, obviously, fronting of the first element must have sufficed for *oi both to trigger Reg2 and to escape the workings of Umlaut, Meillet’s *LJ is compatible in principle with a scenario putting Prog at some stage before *oi ceased being a diphthong, provided that the diphthong developed by fronting its first element while remaining diphthongal (cf. Meillet 1900: 8, Mikkola 1908: 7, 13, Vondrák 1912: 304). However, for this scenario to be realistic we would need to have some idea of how an *[äi]-like diphthong reflecting earlier *oi managed to escape merging with the reflex of *ei, which was also [äi]-like. Since there is no evidence for the existence of such an intermediate stage, and since, should we assume it existed, we would have to explain how it happened that *oi avoided merging with *ei, it is best to identify the fronting of the first element of *oi with the monophthongization. Finally and most confusingly, there are two fundamentally different ideas about the mechanism by which the early Slavic diphthongs were eliminated. Whereas the most common approach is to assume that they developed straight into monophthongs, there is a minority view holding that they were first metathesized, with *oi, *ei and *ou yielding *io, *ie, and *uo respectively, and that it was only at some later stage that the attested monophthongs arose. In what follows I shall be referring to “the monophthongization view” and “the metathesis view” respectively. Both views come in several varieties. 7.4 Two versions of the monophthongization view Within the monophthongization view, there are two ways of picturing what happened to *ei and *ou. 1. According to some, the two diphthongs developed straight into ǔ and ǰ by assimilation of the first element to the second (e.g. Jakobson 1929: 25-26), with the reflex of *ei merging with pre-existing ǔ, but the reflex of *ou remaining distinct from pre-existing ǰ, which had lost its rounding, developing into y, by the time the reflex of *ou yielded *ǰ. The implication is that the delabialization of *ǰ preceded Mono. In a sense this is what could be called the textbook view. 2. Others hold that this maximally simple scenario is untenable and that there was an intermediate phase with mid or high/mid monophthongs which

550

WILLEM VERMEER

were raised to ǔ and ǰ only at a later stage (e.g. McKenzie 1919: 170, Mareš 1965a: 18-21). The view that regards the differential treatment of velars after *ei and *ǔ as evidence for Pedersen’s chronology presupposes the correctness of the textbook scenario (section 7.1). 7.5 The metathesis view and the krìevs - kuomino evidence Since it is my – admittedly highly informal – impression that most slavists are unaware of the very existence of the metathesis view, something of an introduction may come in useful. The first to explain the development of an early Slavic diphthong by means of a scenario involving a metathesis appears to have been Roman Jakobson (1929: 23-25), who limits it to *oi. He does not tell the reader what motivated him to break with the tradition in this respect. Somewhat later, Mikkola (1938: 16) extended the metathesis view to *ou in order to express the fact that some Slavic words borrowed by Latvian have uo corresponding to Slavic u (< *ou). This is interpreted as evidence for a stage with Slavic *uo as the reflex of *ou. In the case of *ei Mikkola leaves the matter open (1938: 33). Later we find the metathesis view applied to all diphthongs by such authors as Vaillant (1950: 118), Jakobson (1963: 158, cf. also Stankiewicz 1973: 182), and Shevelov (1964: 285, 298). As these names show, it is by no means a marginal aberration. The factual evidence that has been adduced in favour of the metathesis view consists of a handful of early Slavic loans in Latvian and Finnic in which ie and uo correspond to Slavic *i and *u, e.g.: • • • • •

Latv krìevs ‘Russian’, borrowed from the designation continued by Old Russian as krivՌiǁi. F. (dial.) kuomino ‘threshing-floor, barn used for threshing’ (Kalima 1956: 81), borrowed from the Slavic word attested as OCS gumӸno. F. (dial.) kuoma ‘godfather; friend’ (Kalima 1956: 80), borrowed from the Slavic word attested as OR kumԔ ‘godfather’. F. luosa ‘puddle’ (Kalima 1956: 91), borrowed from the Slavic word attested as R luža ‘puddle, pool’. F. (dial.) luota ‘plate, dish’ (Kalima 1956: 91), borrowed from the Slavic word attested as OCS bljudo.

Let us call this “the krìevs - kuomino evidence”. In order to account for it, it has been deemed necessary to assume that the reflexes of PSl *ei en *ou were [ie]like and [uo]-like for a time before they ended up as attested monophthongal i

PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION

551

and u. Put differently: the krìevs - kuomino evidence has been regarded as direct support for the metathesis view. 7.6 Criticism of the metathesis view The metathesis view has been evaluated by Schenker (1995: 86-88) and found wanting in important respects. In the light of Schenker’s discussion I see three problems that will have to be defused before there will be any reason to treat the metathesis view as a valid alternative for the monophthongization view. The first one has to do with the way the krìevs - kuomino evidence is to be interpreted. The adherents of the metathesis view do not seem to realize that the ieՌlike and uo-like diphthongs of both Latvian and Finnish have arisen fairly recently from earlier long mid monophthongs. The present state of the problem leaves little doubt that mid first millennium Finnic and pre-Latvian had not yet diphthongized their long mid vowels, hence the overwhelming likelihood is that they borrowed those Slavic words as *krljvՌ and *kǢminՌ.46 The second problem has to do with the way the metathesis view complicates our understanding of the history of the vowel system. It is reasonable to assume, for instance, that the second element of the diphthong *io which must have resulted from the presumed metathesis of *oi underwent Umlaut, causing the diphthong to end up as *ie (cf. already Jakobson 1929: 24). If *ei is held to be metathesized to *ie at the same time, an unwanted merger of the reflexes of *oi and *ei cannot easily be avoided. So how did *ei and *oi end up being different in the attested material? As far as I know the point is nowhere addressed in the literature on the metathesis view, let alone defused. This is not the only problem of this type. Take, for instance, the merger of the reflex of *oi with that of *lj. In the monophthongization view the merger is a direct consequence of Mono, which in turn is part of the drive towards open syllables. Nothing could be more straightforward. Contrast this with the metathesis view, which implies the necessity to reconstruct an unmotivated further development with the sole purpose of producing the merger; Jakobson, for instance, assumes that at a certain stage *lj was diphthongized causing it to merge with the reflex of *oi (o.c.: 28); the diphthongization is limited to *lj and completely lacks motivation (cf. Vermeer 2003: 432). The third problem is of a typological nature: is such a metathesis a natural or necessary development? Can it take place at all? Given the fact that straight 46 For Finnic see, e.g., Viitso (1998a: 106), for Latvian McKenzie (1919) and Stang (1966: 59). In the case of both Finnic and East Baltic the diphthongization has not even reached all of the relevant area, see, e.g., Viitso (1998b: 122-123) on Estonian, and Grinaveckis (1973: 190-192) on Lithuanian.

552

WILLEM VERMEER

monophthongization of diphthongs is such a well-attested type of change, what is so special about this specific case that here metathesis is to be regarded as the obvious first step? To the best of my knowledge the point has not been argued at all. True, some have stated that metathesis is preferable given the law of open syllables (notably Jakobson 1963: 158), but monophthongization produces an outcome that is perfectly satisfactory from the point of view of the law of open syllables and partisans of the metathesis view have nowhere shown that that view is actually to be preferred. Stankiewicz writes in this context: “the Common Slavic vocalic diphthongs did not become monophthongs right away, but were at first subject to a reversal of the syllabic crest which parallels the metathesis that later took place in the Slavic diphthongs with liquids” (1973: 182). That is much too rash to be credible. After all, in the case of the liquid diphthongs, metathesis was the only way – alongside polnoglasie – to get those sequences to conform to the law of open syllables, but in the case of the vocalic diphthongs monophthongization was enough, at least partisans of the metathesis view have not shown that it was not, they have never gone further than asserting it apodictically. For the time being Jules Levin’s harsh verdict appears justified: “There is no evidence for the historical validity of a metathesis rule, such that ie, uo < *ei, *ou. Such abstract exercises are not grounded in any observational evidence” (1975: 148). 7.7 The monophthongization view and the krìevs - kuomino evidence Although the metathesis view is unwarranted given the present state of the debate, its advocates have been right to insist on the importance of the krìevs kuomino evidence, which has been ignored by the overwhelming majority of the adherents of the monophthongization view, who have continued assuming that *ei and *ou developed straight into *ǔ and *ǰ. Compared with Slavic, the Baltic and Finnic languages have rich vowel systems now, and are generally agreed to have had rich vowel systems in the first millennium already. There can be no serious doubt that they distinguished between mid lj/Ǣ and high ǔ/ǰ in addition to displaying a whole range of diphthongs of the type ei, ou, ai and au. So if they adopted later Slavic *i and *u as *lj and *Ǣ there is every reason to believe that such were the approximate values of the corresponding Slavic vowels (cf. Kiparsky 1948: 33). From which it follows that nothing prevents us from reconstructing a stage with [lj]-like and [Ǣ]-like reflexes of *ei and *ou. In addition to this, scholars have wondered if straight monophthongization of *ei and *ou into the high monophthongs ǔ and ǰ is a natural development at all, particularly in view of the likelihood – verging on certainty – that the initial element of both diphthongs was in fact a low vowel at the time (see section 5.8), so that what we are dealing with is more likely to be something along the lines

PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION

553

of [äi] and [au]. It is for this reason that Mareš has argued that a scenario comprising an intermediate phase with mid monophthongal reflexes of *ei and *ou is typologically more plausible than one in which the diphthongs developed straight into high monophthongs without some intermediate step (Mareš 1965a: 18-21). Given these doubts one would certainly like to know of well-authenticated examples of [äi]-like and [au]-like diphthongs developing straight into [ǔ]-like and [ǰ]-like monophthongs. 7.8 The elimination of the Proto-Slavic diphthongs: the simplest scenario It follows from the above discussion that the simplest scenario for the elimination of the diphthongs which accounts for the relevant facts comprises the following elements: 1. it assumes that Mono had no internal chronology, e.g. that *ei and *ou were not eliminated earlier than *oi; 2. it identifies the stage at which the reflex of *oi lost the possibility of being umlauted with the elimination of the diphthongs as diphthongs; 3. it assumes that *ei and *ou did not yield ǔ and ǰ straight away, but first developed into an intermediate stage of mid or high-mid monophthongs, which is attested by the Latvian and Finnic loanword evidence and may well turn out to be more plausible from a general point of view as well; 4. it operates with the monophthongization view to the exclusion of the metathesis view. 7.9 The Prog Window If it is true that the reflex of *ei did not trigger Prog, the implication is that we not only have a terminus post quem for Prog (Mono), but also a terminus ante quem, which is the raising of the long mid monophthongs implied by the krìevs - kuomino evidence, resulting in the merger of the reflex of *ei with that of *ǔ. Let us call the chronological window defined by Mono and the merger of *ei with *ǔ the Prog Window. While the Prog Window was open, the reflex of *ei was a mid or high-mid front vowel, which was subsequently raised to merge with *ǔ, closing the window again. The reflex of *ou furnished the correlating back vowel, which however, on being raised jointly with the reflex of *ei did not merge with *ǰ, which got out of harm’s way by shedding its roundedness as part of the development that caused *ǰ and *u to be reflected as y and Ԕ in attested Slavic. As a consequence of the loss of rounding, umlauted *ǰ and *u merged with the reflexes of *ǔ and *i in Umlaut position. We saw earlier that umlauted *u, and arguably also its long counterpart *ǰ, did not trigger Prog, hence were distinct from the reflexes of *i and *ǔ until after Prog took place, hence until after Mono (e.g. sections 5.1 and 5.6).

554

WILLEM VERMEER

The Prog Window can be identified with the period that witnessed the adoption by Finnic and early Latvian of such Slavic words as *krivՌ and gumӸno. The reality of the Prog Window renders chronological identification of the merger of *ei and *ǔ with Mono impossible. As a consequence of this, the view that Prog is not found after *i reflecting *ei cannot be construed as implying that Prog preceded Mono. 8. Prog before Reg1: Martinet’s version of Pedersen’s chronology The view that Prog preceded Reg1, which has been fairly popular in recent decades, particularly in the US, grew out of Jakobson’s work on Slavic historical phonology (notably Jakobson 1929), which as far as Prog is concerned continues Lehr-Sp’awiǝski (1911) and Trubetzkoy (1922) with fairly minor modifications. 8.1 Prog and Reg1: the otӸǁe evidence Most scholars have put Prog after Reg1, notably because of the existence of vocatives of the type otӸǁe and kԔnNjže. These are simplest to account for on the assumption that the stem-final *ՌkՌ and *ՌgՌ reflected here by ՌǁՌ and ՌžՌ respectively were still velars when Reg1 operated; had Prog been earlier, one would have expected **otӸǀe and **kԔnNjdǺe, with reflexes of Prog. The point has been expressed often in the literature.47 Nevertheless some investigators have rejected the argument, preferring to put Prog simultaneous with Reg1 or even earlier, most explicitly Mikkola (1908: 13) and Lunt (1977: 173, 1981: 16 with note 21, 1987: 257, 1997: 30-31). It is obvious that views putting Prog in the vicinity of Reg1 are more complex than the mainstream interpretation because they give rise to the question as to why the velars in, say, Vsg *atike and Gsg *atikƽ ended up with the different reflexes attested in *otӸǁe and *otӸǀa respectively (see, e.g., Schenker 1995: 9192). For that reason they require additional motivation to be acceptable. 8.2 Roman Jakobson: phonemes and trends The most important advocates of Pedersen’s chronology to have been active after Trubetzkoy are: Vondrák (1923-24: 22, 1924: 354), Jakobson (1929: 19-24, 3135), Milewski (1932: 243-254), Mareš (1956: 464-468 = 1965a: 36-43 = 1965b: 3238, cf. 1969: 55-63), Ebeling (1963: 34-37), and Holzer (1995: 248). 47 Beginning with Sobolevskij (1889: 33 = 1891: 116), not to speak of Leskien’s slightly earlier lectures published as Leskien (1991: 25, originally 1885-1886). Cf. also, e.g., Baudouin de Courtenay (1893: 16), Jagiǀ (1901: 129), Il’inskij (1916: 241), Beliǀ (1921: 22), Meillet (1924: 82 = 1934: 91), Vondrák (1924: 355), Van Wijk (1931: 68), Milewski (1932: 247), Bielfeldt (1961: 102), Gorškov (1963: 260), Birnbaum (1970: 107-108), Gasparov and Sigalov (1974: 127).

PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION

555

As far as I can see, they have added little of substance to Lehr-Sp’awiǝski’s and Trubetzkoy’s findings. The chronological problem is treated everywhere in basically the same way, hence discussing the differences between these theories would go beyond the scope of this contribution. However, several points brought up by Jakobson deserve a mention because of the way they have influenced later scholarship. One of the challenges Jakobson’s account attempts to meet is that of operationalizing the Prague School phoneme, with its distinction between the allophonic and the phonemic phase of sound laws. As we have seen, Lehr-Sp’awiǝski had earlier proposed a two-stage mechanism explaining how Prog and Reg2 ended up with identical reflexes (section 4.2). In a way that is strongly reminiscent of Lehr-Sp’awiǝski’s conception, Jakobson assumes that the allophonic phase of Prog arose before Umlaut did, but that it remained in the system at least until *oi developed into a front vowel which rendered preceding velars subphonemically palatal as well. Subsequently the allophones caused by Prog and Reg2 are held to have been phonemicized in one go (Jakobson 1929: 19-24, 31). This yields a scenario identical to that shown in Table 6, with the difference that “palatalization phase” is to be read as “allophonic phase” and “assibilation” as “phonemic phase”. Although Jakobson’s scheme is straightforward enough, it has given rise to a communicative complication not everybody is aware of. The point is the following. As is well known, in Prague School practice it is traditionally the phonemic phase that is identified with the sound law involved. In talking about Prog, however, that tradition is not generally adhered to. It has become customary to refer to the allophonic phase by a label containing the word “palatalization” and to the phonemic phase – where Reg2 and Prog go together – by some different label, usually “assibilation” (e.g. Mareš 1956: 467 = 1965a: 42 = 1965b: 36 = 1969: 61). Although the point is merely terminological, it is a source of awkward misunderstandings. When adherents of the Lehr-Sp’awiǝski/Jakobson version of Pedersen’s chronology say they put Prog before Reg2, it is essential to realize that what they have in mind is the first or allophonic phase. As far as the second or phonemic phase is concerned, Lehr-Sp’awiǝski’s view (hence also Jakobson’s) coincides with the default chronology that puts Reg2 and Prog simultaneous as a single innovation (section 4.3). A second point that is new with Jakobson is the way Prog is drawn into the ambit of Reg1. He assumes – without giving reasons – that the allophonic phase of Prog was present in the language before Reg1 ceased being allophonic (1929: 19). Then the rise of Umlaut made the allophones of Reg1 into distinct phonemes, as follows: since Umlaut changed *o into *e (speaking in traditional terms), the allophones of both Prog and Reg1 could now be followed by *e. The earlier con-

556

WILLEM VERMEER

trast between such sequences as */ike/ and */iko/ (each allophonically palatalized in a different way) was now continued as one between on one hand */iǁe/ (with the brand-new phoneme /ǁ/), and on the other */ike/, with /k/ being redundantly palatalized as before by Prog. The chronology is shown in Table 10. Table 10. Jakobson’s chronology of the allophonic phase of the palatalizations 1 Allophonic phase of both Prog and Reg1. 2 Umlaut leading to phonemic phase of Reg1. Prog remains allophonic. 3 Mono. 4 Allophonic phase of Reg2. 5 Phonemic phase of both Reg2 and Prog. A striking feature of Jakobson’s take on the phonological history of Slavic is his insistence on the trend toward what he calls syllabic synharmony: “Une des tendances qui traverse comme un fil rouge toute la période considérée, c’est l’uniformisation de la syllabe. Il s’établit graduellement une harmonie des sons à l’intérieur de la syllabe (synharmonisme syllabique […]), on voit cristalliser deux types de syllabes – les molles et les dures. La tendance en question se réalise par le moyen de l’assimilation régressive aussi bien que progressive” (1929: 20). Although it needs no arguing that the trend towards syllabic synharmony was a conspicuous fact of Proto-Slavic during a certain phase of its evolution, Jakobson’s insistence on it has come to be interpreted by many as a claim that while the trend was in operation, changes not subsumed by it were impossible. And if that is correct, it goes without saying that Prog has to be moved outside the period of syllabic synharmony whatever the cost. This point constitutes the background for the subsequent rise of the assumption that Prog preceded Reg1. 8.3 Prog before Reg1 (Martinet 1952) The first to put Prog before Reg1 was André Martinet (1952: 162 = 1955: 366-367). His reason for doing so was the conviction that a change like Prog, by which one syllable influenced the next, was impossible during the period of syllabic synharmony, hence Prog either must have preceded Reg1, which initiated the period of syllabic synharmony, or followed the loss of the weak jers, which put an end to it. Since obviously Prog cannot have been later than the fall of the jers (which postdates the earliest texts, whereas Prog predates them), it had to precede Reg1. The chronology implied by Martinet is shown in Table 11. Shortly after Martinet, the belief that there is something odd about Prog being carried out during the period of syllabic synharmony was also expressed by Karel Horálek (1955: 108 = 1992: 98). Subsequently, Martinet’s reasoning has

PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION

557

been repeated with inessential modifications by several later scholars, notably Robert Channon (1972: 46-47) and Horace Lunt (1981: 25), for discussion of which see Vermeer (2003: 436-439). Table 11. Martinet’s chronology 1 Prog. 2 Onset of the period of syllabic synharmony. 3 Reg1. 4 Mono. 5 Reg2. 6 Cessation of the period of syllabic synharmony. The principal difference among the adherents of Martinet’s chronology has to do with the phase of Prog they have in mind. Whereas most of them do not differentiate between the allophonic and the phonemic phase (and one may safely assume that what they have in mind is the latter), Lunt’s version concerns the allophonic phase only, as he is at pains to argue (1981: 27-28). Indeed, as far as chronology is concerned his view differs only minimally from Jakobson’s, compare Table 12 (Lunt) with Table 10 (Jakobson, section 8.2).48 Table 12. Lunt’s chronology 1 Allophonic phase of Prog. 2 Onset of the period of syllabic synharmony; allophonic phase of Reg1. 3 Umlaut leading to phonemic phase of Reg1; Prog remains allophonic. 4 Mono. 5 Allophonic phase of Reg2. 6 Phonemic phase of both Reg2 and Prog. 7 Cessation of the period of syllabic synharmony. 8.4 Schwartz’s attempt to shore up Martinet’s chronology The most recent version of Martinet’s view I have come across is by Geoffrey Schwartz: “Early in its development, Common Slavic was characterized by a “law of open syllables”. BdC [i.e. Prog, WV] must have developed before this syllable structure was in place for perceptual reasons. If syllables were open at the time of 48 Elsewhere, Lunt assumes that Prog was phonologized prior to Reg1 by the raising of *a to *u in a number of nominal endings, notably the Gpl and the masc Asg (1987: 270). Since the chronology of the raising – which is highly controversial – is not discussed, this view has the ontological status of a shot in the dark and need not occupy us here.

558

WILLEM VERMEER

BdC, then a syllable boundary must have separated the velar from the front vowel that caused the palatalization, and the velar would have been a syllableinitial stop followed by a low back vowel. In syllables such as /ka/, the stop is very unlikely to be heard as a dental or an alveolar (Liberman – Delattre – Cooper 1952: 492-514) due to the vowel context that produces formant transitions quite distinct from those of a dental consonant. In addition, due to the syllable structure, the unambiguous noise burst in a CV syllable such as /ka/ is a robust cue for the perception of the place of articulation of the stop. However, /k/ in an /i/ environment is perceptually and spectrally difficult to distinguish from a dental. This difficulty is exacerbated in syllable-final VC sequences (Ohala 1990: 271), where listeners cannot rely on a robust stop burst, but need to use the vowel-formant transitions as a cue to identify which stop they are hearing. Thus it is likely that BdC occurred before open syllables became the norm in Slavic” (Schwartz 2001: 288). This passage illustrates the principal flaw of the Martinet view, which crucially depends on the belief that in a system characterized by – or just tending towards, as in the relevant stage of Common Slavic – open syllables, the /i/ in a sequence like /ika/ does not belong to the phonetic or perceptual environment of the /k/ because the intervening syllable boundary is held to be of a fundamentally different kind than syllable boundaries in systems not characterized by – or tending towards – open syllables, which do permit the /i/ to influence the /k/ in such sequences. The problem with this belief is that of substantiating it, a feat Schwartz’s predecessors have failed to address. Schwartz is the first to be aware of the problem. He tries to stop the gap by drawing on the results of two publications. In neither case, however, do those results even begin to accomplish what is needed: 1. The problem of the perception of velars in syllable-final position as treated by Ohala (1990) has no bearing on the problem of Prog. After all, the phonetic contexts in which Prog is found (*ika, *ǔka etc.) leave no doubt that syllable-initial velars were palatalized by preceding vowels, irrespective of the specific chronology we choose to operate with. It is up to the adherents of the Martinet view to prove by more than rhetoric alone that the syllable boundary was such as to block palatalization during the period of syllabic synharmony, but permitted it at earlier and later stages. 2. On the other hand, granted that it is the case that “in syllables such as /ka/, the stop is very unlikely to be heard as a dental or an alveolar” (as argued by Liberman et al.), there is no denying that palatalization of syllable-initial velars followed by [a]-like vowels is reliably attested even in the absence of a preceding front vowel, perhaps most notoriously in French and related Romance dialects, as in French chanter, chose reflecting Latin cantƽre, causa.

PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION

559

All published versions of the Martinet view I am familiar with (Martinet, Channon, Lunt, Schwartz) crucially lack empirical content which would invite one to take them seriously. 8.5 No Prog on *x (Machek, Lunt) In addition to Martinet’s argument, several further pieces of evidence have been adduced purporting to favour putting Prog before Reg1. Some have argued that Prog must have preceded Reg1 because they claim it did not affect *x, hence must have taken place before *x arose from earlier *s, which must have happened before Reg1, which definitely affected *x (Lunt 1977: 170-171; 1981: 35-36, similarly Velǁeva 1988: 32). The point is not at all compelling. To begin with, the external fact the chronology is designed to explain has not been shown to exist and require explanation. As we saw in sections 2.1 and 2.4, the assumption that Prog affected *x is indispensable to account for the facts with respect to *elӸǧa/elӸxa and the pronoun *vӸǧӸ/vӸxԔ, not to mention the phytonym *vӸǧӸ/vӸxԔ, on which the jury is still out. On the other hand the examples of stem-final ՌxՌ preceded by *i, *Ӹ or *Nj can be explained without taking recourse to the assumption that Prog did not affect *x: a. In the adjective *lixԔ the ՌiՌ reflects *ei. If one assumes that *ei did not trigger Prog, as most investigators have done since Trubetzkoy (1922: 227) and Meillet (1934: 92), the absence of Prog in this word is unremarkable. (For discussion see section 7.2.) b. The noun *ženixԔ is usually explained as an otherwise unparallelled derivation from the verb ženiti (cf. Meillet 1902-05: 361, Trubaǁev 1959: 95, Vaillant 1974: 671). The word is of limited relevance, if only because the ՌiՌ preceding the stem-final ՌxՌ may well reflect *ei. c. Finally there is the verb *pӸxati, presupposing an aor2/3sg *pӸxa. Derivations like pӸšenica ‘wheat’ are built on a past passive participle *pӸšenԔ, which presupposes the existence of an earlier verb *pӸsti, pres 1sg *pӸx՛ or *pӸǧ՛ (depending on whether or not Prog was blocked by a following *Ռ՛, see section 3.7), *pӸšeši, exactly corresponding to its Lithuanian counterpart pìsti, pres 3sg/pl pìsa (cf. Vaillant 1966: 206). The stem-final ՌxՌ of *pӸxa can easily have arisen on the basis of forms in which stem-final *ՌxՌ was regular, helped along by the existing alternation ՌxՌ ~ ՌšՌ, which it is reasonable to assume immediately restored the stem-final *ՌxՌ of such forms as pres 1sg *pӸx՛ even if it had been subject to Prog. Extension of ՌxՌ to the aorist may have been encouraged by the phonetic closeness of *pӸǧa to the aorist of the verb meaning ‘write, paint’ (*pӸsa).

560

WILLEM VERMEER

So while the assumption that Prog affected *x helps to explain the attested reflexes of *vӸǧӸ/vӸxԔ ‘whole, all’ and *elӸǧa/*elӸxa, there are no examples requiring the assumption that *x was left unaffected by Prog. The evidence Lunt tries to account for turns out not to exist. 8.6 Prog and the ruki rule (Velǁeva) Borjana Velǁeva has written that the shape of Prog “suggests additional evidence for the early occurrence of the progressive palatalization: it clearly resembles the unquestionably early change of s to š (e.g., aws- Ռ awš- [ҸҏѽҸ : ҸҏҗѦҡѦ]). Both are assimilations with respect to the features anterior and high; both occur before vowels. If we assume that the progressive palatalization is early – prior to the first – we may formulate it as a constraint on the distribution of noncontinuant and continuant high consonants in certain environments. At the time when s’, k, and g were the only high obstruents, the form of the progressive palatalization shows that a noncontinuant high consonant could only be anterior and strident [reference to par. 2.1, where these concepts are introduced, W.V.] under circumstances similar to those of the change of s to š” (1988: 33). Note first that the specifics are not discussed at all, beginning with the fact that early placement of Prog yields the wrong reflexes. Furthermore, no foundation has been laid for the underlying hypothesis that formally related innovations must necessarily have occurred closely together in time even in the presence of evidence to the contrary. Velǁeva rather strikingly downplays the differences between Prog and the ruki rule. Her formulation “a constraint on the distribution of noncontinuant and continuant high consonants in certain environments” glosses over the fact that the environments differ in that Prog does, but the ruki rule does not partly depend on the vowel of the syllable following the velar. Finally, the view that “s’, k, and g were the only high obstruents” presupposes the correctness of a classification by which sibilants and velar stops fall into the same natural class, and loses its force the moment a different classification is chosen. Velǁeva’s evidence for Martinet’s chronology will require a lot of additional building up before there will be any reason to take it seriously. 8.7 Further evidence Terence Carlton appears to be convinced that it is only by placing Prog before Reg1 that it becomes possible to account for the fact that the reflexes of *ei and umlauted *u/ǰ do not trigger Prog (Carlton 1991: 135). As far as I can see this rests on a misunderstanding about the implications of Pedersen’s chronology, which is perfectly able to account for these cases. If Prog preceded Mono (as it does in all versions of Pedersen’s chronology), *ei must

PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION

561

still have been distinct from *ǔ and may have had a different effect on a following velar consonant. The distinctness of umlauted *u/ǰ can be attributed to the mechanism implicit in Trubetzkoy’s account (section 5.4). 9. Pedersen’s chronology: evaluation Thus far, no single piece of credible evidence justifying an early placement of Prog has been put forward: 1. The evidence Pedersen himself wanted to account for dissolves the moment it is viewed in the light of Meillet’s analysis of the relevant morphological facts (section 3.2). 2. Lehr-Sp’awiǝski’s evidence presupposes the correctness of the unwarranted assumption that analogical processes invariably reach their natural conclusion, and fails to take into account the specific position of the vocative in a nominal paradigm (section 4.1). 3. The evidence introduced by Trubetzkoy presupposes the correctness of his interpretation of *Ԕ as a non-blocking vowel, which depends crucially on a misunderstanding about the morphological status of the adjective *niǀӸ (section 5.4). 4. The reasoning based on the absence of Reg2 in Novgorod has never been made explicit. Should that ever happen, it will have to be justified in the light of published scenarios that account for the evidence without separating Reg2 and Prog (sections 6.1-2). 5. The evidence that might be construed on the basis of the differential treatment of *ei and *ǔ presupposes the correctness of the view that *ei merged with *ǔ immediately upon the monophthongization of diphthongs. Not only is that view far from being inevitable, but there is factual evidence contradicting it (section 7.7). Conversely, the moment Prog is assumed to have preceded the rise of Umlaut, Meillet’s ՌLJՌ can no longer be accounted for in a natural way because one is forced to assume that *Ռoi blocked Prog; since *o is very unlikely to have done so, the alleged blocking effect must be due to the second component of the diphthong, which is awkward given the fact that that is a high front vowel (see sections 5.11-12). Most scholars have just ignored the difficulty (see section 3.5); the only ones to have faced it are Pedersen and Trubetzkoy. Whereas Pedersen’s solution fails to account for the facts it is designed to account for (section 3.4), Trubetzkoy’s is ad hoc and suffers from various additional disabilities, which is all the more serious because the problem does not even arise if Prog is put after Mono (section 5.12). Martinet’s version of Pedersen’s chronology (Prog before Reg1) presupposes the correctness of the unsubstantiated belief that changes by which one syllable

562

WILLEM VERMEER

influenced the next could not take place during the period of syllabic synharmony (section 8.3). Other evidence that has occasionally been adduced in its favour does not bear serious scrutiny (see sections 8.5-7). Like all versions of Pedersen’s chronology, the Martinet one gives rise to the problem of how to account for Meillet’s ՌLJՌ. In addition it does not account in a natural way for the otӸǁe evidence. A full century has not sufficed for Pedersen’s chronology to show that it has a point. It is about time to pull the plug. Leiden University REFERENCES Aitzetmüller, Rudolf 1978 Altbulgarische Grammatik als Einführung in die slavische Sprachwissenschaft. Freiburg i. Br. ARj 1880-1976 Rjeǁnik hrvatskoga ili srpskoga jezika. Zagreb. Andersen, Henning 1969 “Lenition in Common Slavic”. Language 45, 553-574. 1996 Reconstructing Prehistorical Dialects. Initial Vowels in Slavic and Baltic. Berlin and New York. 1998 “Dialektnaja differenciacija obšǁeslavjanskogo jazyka. Paradoks obšǁix tendencij razvitija s razliǁnymi lokal’nymi rezul’tatami. In: R.A. Maguire and A. Timberlake (eds.), American Contributions to the Twelfth International Congress of Slavists, 565-600. Columbus, Ohio. Arumaa, Peeter 1976 Urslavische Grammatik. Einführung in des vergleichende Studium der slavischen Sprachen II: Konsonantismus. Heidelberg. Baudouin de Courtenay, Jan Niecis’aw 1893 [Boduèn-de-Kurtenè, I. A.] Dva voprosa iz uǁenija o “smjagǁenii” ili palatalizacii v slovjanskix jazykax (= Zwei Fragen aus der Lehre über die Palatalisation in den slavischen Sprachen). Ottisk iz “Uǁenyx Zapisok Imp. Jur’evskago Universiteta” (= Sonderabdruck aus “Acta et commentationes Imp. Universitatis Jurievensis (olim Dorpatensis)”), 1893/2. 1894 “Einiges über Palatalisierung (Palatalisation) und Entpalatalisierung (Dispalatalisation)”. Indogermanische Forschungen 4, 45-57. Beliǀ, Aleksandar 1921 “NajmlaDža (treǀa) promena zadnjenepǁanih suglasnika k, g i h u praslovenskom jeziku”. Južnoslovenski filolog 2, 18-39. 1922-23 Review of Trubetzkoy 1922. Južnoslovenski filolog 3, 131-139. 1928 “La troisième ou la plus récente palatalisation des gutturales”. Revue des Études Slaves 8, 50-67. BLJliǁ, Jaromír; Adolf Kamiš, Karel Kuǁera 1979 Malý staroǁeský slovník. Praha.

PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION Beljaev, D.D. 1986

563

“Tret’ja palatalizacija praslavjanskix zadnenebnyx: mexanizm i xronologija”. Voprosy jazykoznanija 1986/2, 106-116.

Berneker, Erich, 1908-14 Slavisches etymologisches Wörterbuch. Heidelberg. Bernštejn, Samuil B. 1961 Oǁerk sravnitel’noj grammatiki slavjanskix jazykov 1. Moskva. Besedina-Nevzorova, Vera P. 1962 Staroslavjanskij jazyk. Xar’kov. Bezlaj, France 1977-2005 Etimološki slovar slovenskega jezika. Ljubljana. Bidwell, Charles E. 1961 “The chronology of certain sound changes in Common Slavic as evidenced in loans from Vulgar Latin”. Word 17, 105-127. Bielfeldt, Hans H. 1961 Altslawische Grammatik. Halle. Birnbaum, Henrik 1968 “Rekonstrukcja wewnNjtrzna, kolejnoǧǀ synchronicznych regu’ gramatyki syntetycznej i zagadnienie najdawniejszych stosunków miNjdzy jNjzykami ba’tyckimi a s’owiaǝskimi”. International Journal of Slavic Linguistics and Poetics 11, 1-24. 1970 “Internal reconstruction, order of synchronic rules in generative grammar, and the problem of early Balto-Slavic relations”. In: H. Birnbaum, Problems of Typological and Genetic Linguistics Viewed in a Generative Framework, 92122. The Hague and Paris. Bräuer, Herbert 1961 Slavische Sprachwissenschaft I: Einleitung, Lautlehre. Berlin. Brugmann, Karl and Berthold Delbrück 1897 Grundriss der vergleichenden Grammatik der indogermanischen Sprachen 1: Einleitung und Lautlehre. 2nd ed. Strassburg. Bubrix, Dmitrij V. 1923 “Iz praslovjanskoj fonetiki”. Izvestija otdelenija russkogo jazyka i slovesnosti 24, 246-284. Carlton, Terence R. 1991 Introduction to the Phonological History of the Slavic Languages. Columbus, Ohio. Channon, Robert 1972 On the place of the progressive palatalization of velars in the relative chronology of Slavic. The Hague and Paris. Derksen, Rick 2005 “The orthography of the Elbing Vocabulary and the relationship between Old Prussian and East Baltic”. Baltistica VI priedas, 51-67. Diels, Paul 1932 Altkirchenslavische Grammatik. Mit einer Auswahl von Texten und einem Wörterbuch. I Teil: Grammatik. Heidelberg. Dinekov, PetƼr (ed.) 1991-93 Simeonov sbornik: (po Svetoslavovija prepis ot 1073 g.). 3 vols. Sofija. Durnovo, Nikolaj N. 1925 Review of Meyer 1923. Zeitschrift für slavische Philologie 1, 467-507.

564 Ebeling, Carl 1963

WILLEM VERMEER

“Questions of relative chronology in Common Slavic and Russian phonology”. In: Dutch Contributions to the Fifth International Congress of Slavicists, 27-42. The Hague.

Ekblom, Richard 1935 Die Palatalisierung von k, g, ch im Slavischen (Skrifter utgivna av K. Humanistiska Vetenskaps-Samfundet i Uppsala 29/5). Uppsala and Leipzig. 1951 Die frühe dorsale Palatalisierung im Slavischen (Skrifter utgivna av K. Humanistiska Vetenskapssamfundet i Uppsala 39/2). Uppsala and Wiesbaden/Leipzig. ÈSSJ 1974Ètimologiǁeskij slovar’ slavjanskix jazykov. Praslavjanskij leksiǁeskij fond. O.N. Trubaǁev (ed.), Moskva. Furdal, Antoni 1961 Rozpad jNjzyka pras’owiaǝskiego w ǧwietle rozwoju g’osowego (Prace Wroc’awskiego towarzystwa naukowego, seria A nr 70). Wroc’aw. Gasparov, Boris M. & Pavel S. Sigalov 1974 Sravnitel’naja grammatika slavjanskix jazykov 1. Tartu. Gorškov, Aleksandr I. 1963 Staroslavjanskij jazyk. Moskva. Grickat-Virk, I. 1951-52 “Još o treǀoj palatalizaciji”. Južnoslovenski Filolog 19, 87-110. Grinaveckis, Vladas 1973 ŽemaiǁiDZ tarmiDZ istorija (fonetika). Vilnius. Halla-aho, Jussi 2006 Problems of Proto-Slavic Historical Nominal Morphology On the Basis of Old Church Slavic (Slavica Helsingiensia 26). Helsinki. Hock, Wolfgang 1998 “Das Urslavische”. In: Peter Rehder (ed.), Einführung in die slavischen Sprachen, 3d ed., 17-34. Darmstadt,. Holzer, Georg 1995 “Die ersten nachurslavischen lautlichen Innovationen und ihre relative Chronologie”. Linguistica Baltica 4 (= Kury’owicz Memorial volume 2, ed. Wojciech Smoczyǝski), 247-256. Horálek, Karel 1955 Úvod do studia slovanských jazykDz. Praha. 1992 An Introduction to the Study of the Slavonic Languages 1. Nottingham. Hujer, Oldǥich 1924 Úvod do dLJjin jazyka ǁeského. 2nd ed. Prague. Il’inskij, Grigorij A. 1916 Praslavjanskaja grammatika. NLJžin. Jacobsson, Gunnar 1974 “Odwieczny problem palatalizacji postNjpowej tylnojNjzykowych w jNjzykach s’owiaǝskich”. In: Göteborg Contributions to the Seventh International Congress of Slavists … (= Slavica Gothoburgensia 6), 49-74. Göteborg. Jagiǀ, Vatroslav 1901 “Einige Streitfragen. 6. Zum doppelten slavischen Palatalismus”. Archiv für Slavische Philologie 23, 121-129.

PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION

565

Jakobson, Roman O. 1929 Remarques sur l’évolution phonologique du russe comparée à celle des autres langues slaves (Travaux du Cercle Linguistique de Prague 2). Prague. 1963 “Opyt fonologiǁeskogo podxoda k istoriǁeskim voprosam slavjanskoj akcentologii”. American Contributions to the Fifth International Congress of Slavists 1. Linguistic Contributions, 153-178. The Hague. Jeǻowa, Maria 1968 Z problemów tak zwanej trzeciej palatalizacji tylnojNjzykowych w jNjzykach s’owiaǝskich. Wroc’aw/Warszawa/Kraków. Kalima, Jalo 1956 Die slavischen Lehnwörter im Ostseefinnischen (Veröffentlichungen der Abteilung für slavische Sprachen und Literaturen des Osteuropa-Instituts an der Freien Universität Berlin, 8). Berlin. Kallio, Petri 2006 “On the Earliest Slavic Loanwords in Finnic”. In: Juhani Nuorluoto (ed.), The Slavicization of the Russian North. Mechanisms and Chronology (Slavica Helsingiensia 27), 154-166. Helsinki. Koǥínek, Josef M. 1948 Od indoeuropského prajazyka k praslovanǁine. Bratislava. Kiparsky, Valentin 1948 “Chronologie des relations slavobaltiques et slavofinnoises”. Revue des Études Slaves 24, 29-47. Knutsson, Knut 1926 Über die sogenannte zweite Palatalisierung in den slavischen Sprachen (Lunds universitets årsskrift N.F. Avd. 1, Bd 21/9). Lund and Leipzig. Kortlandt, Frederik 1978 “On the history of the genitive plural in Slavic, Baltic, Germanic, and IndoӒEuropean”. Lingua 45, 281-300. 1979 “On the history of the Slavic nasal vowels”. Indogermanische Forschungen 84, 259-272. 1982 “Early dialectal diversity in South Slavic I”. Studies in Slavic and General Linguistics 2, 177-192. 1985 “Long vowels in Balto-Slavic”. Baltistica 21/2, 112-124. 1991 “On methods of dealing with facts and opinions in a treatment of the progressive palatalization of Slavic”. Folia Linguistica Historica 9/2, 3-12. Kul’bakin, Stepan M. 1929 Le vieux slave. Paris. Lehr-Sp’awiǝski, Tadeusz 1911 “Nowsze poglƾdy na niektóre zjawiska t. zw. drugiej palatalizacji”. Rocznik Slawistyczny 4, 141-148. 1923 Zarys gramatyki jNjzyka staro-cerkiewno-s’owiaǝskiego. Poznaǝ. Leskien, August 1871 Handbuch der altbulgarischen (altkirchenslawischen) Sprache. Weimar. 1909 Grammatik der altbulgarischen (altkirchenslavischen) Sprache. Heidelberg. [Ernst Eichler, Gerhart Schröter eds.] August Leskiens Vorlesungen zur ver1991 gleichenden Grammatik der slawischen Sprachen (Sitzungsberichte der sächsischen Akademie der Wissenschaften zu Leipzig, Philologisch-historische Klasse 131/1). Leipzig.

566 Levin, Jules F. 1975

WILLEM VERMEER

“Dynamic linguistics and Baltic historical phonology”. General Linguistics 15, 144-158. Liberman, Alvin M., Pierre Delattre, Franklin S. Cooper 1952 “The role of selected stimulus-variables in the perception of unvoiced stop consonants”. The American Journal of Psychology 65/4, 497-516. Ljapunov, Boris M. 1894 “NLJskol’ko slov o govorax Lukojanovskago uLJzda Nižegorodskoj gubernii”. Živaja Starina 4/2, 143-177. 1900 IzslLJdovanie o jazykLJ Sinodal’nago spiska 1-oj novgorodskoj lLJtopisi I (IzslLJdovanija po russkomu jazyku 2/2). Sanktpeterburg. Lokrantz, Carl 1950 “Zur Deutung der Verba auf Ռcati, Ռdzati und Ռsati”. Zeitschrift für slavische Philologie 20, 378-393. Lorentz, Friedrich 1904 “Slavische miscellen”. Zeitschrift für vergleichende Sprachforschung 37, 264273. Lunt, Horace G. 1977 [Lant, G.G.] “Praslavjanskaja progressivnaja palatalizacija”. In: Fr. Jakopin (ed.), Nahtigalov zbornik, 167-181. Ljubljana. 1981 The progressive palatalization of Common Slavic. Skopje. 1985 “On the progressive palatalization of early Slavic: Synchrony versus history”. Studies in the Linguistic Sciences 15/2, 149-169. 1987 “The progressive palatalization of early Slavic: Opinions, facts, methods”. Folia Linguistica Historica 7/2, 251-290. 1989 “The Progressive Palatalization [of] early Slavic: Evidence from Novgorod”. Folia Linguistica Historica 10/1-2, 35-59. 1997 “Common Slavic, Proto-Slavic, Pan-Slavic: What Are We Talking About? I. About Phonology”. International Journal of Slavic Linguistics and Poetics 41, 7-67. Machek, Václav 1954 Review of Ekblom 1951. Zeitschrift für Slavische Philologie 22, 223-226. 1958 “Zur Erklärung der sog. Baudouinschen Palatalisierung im Slavischen und im Baltischen”. Mélanges linguistiques offerts à Emil Petrovici (CercetƼri de LingvisticƼ, 3, Supliment.), 327-335. Bucarest. Mareš, František V. 1956 “Vznik slovanského fonologického systému a jeho vývoj do konce období slovanské jazykové jednoty”. Slavia 25/4, 443-495. 1965a Die Entstehung des slavischen phonologischen Systems und seine Entwicklung bis zum Ende der Periode der slavischen Spracheinheit (Slavistische Beiträge 18). München. 1965b The Origin of the Slavic Phonological System and Its Development up to the End of Slavic Language Unity (Michigan Slavic Materials 6). Ann Arbor. 1969 Diachronische Phonologie des Ur- und Frühslavischen (Slavistische Beiträge 40). München. Mareš, František V., Petra Fetková et al. (eds.) 1997 Psalterii Sinaitici pars nova (monasterii s. Catharinae codex slav. 2/N). (Schriften der Balkancommission, Linguistische Abteilung 38). Wien.

PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION

567

Martinet, André 1952 “Langues à syllabes ouvertes: le cas du slave commun”. Zeitschrift für Phonetik und allgemeine Sprachwissenschaft 6, 145-163. 1955 Économie des changements phonétiques. Berne. McKenzie, Roderick 1919 “Notes sur l’histoire des diphthongues ie et uo dans les langues baltiques”. Bulletin de la Société de linguistique de Paris 21/2 (= 67), 156-174. Meillet, Antoine 1900 “Notes sur quelques faits de morphologie”. Mémoires de la Société de linguistique de Paris 11, 6-21. 1902-05 Études sur l’étymologie & le vocabulaire du vieux slave. 2 vols. Paris. 1910 Review of Leskien 1909. Göttingische gelehrte Anzeigen 172/1, 362-376. 1924 Le slave commun. Paris. 1934 Le slave commun. 2nd ed. Paris. Meyer, Karl H. 1923 Historische Grammatik der russischen Sprache. Bonn. Mikkola, Jooseppi J. 1908 Review of Vondrák 1906. Rocznik Slawistyczny 1, 3-19. 1938 Die älteren Berührungen zwischen Ostseefinnisch und Russisch (Suomalaisugrilaisen Seuran toimituksia 75). Helsinki. Miklosich, Franz 1878 Altslovenische Lautlehre. 3d ed. Wien. Milewski, Tadeusz 1932 “Rozwój fonetyczny wyg’osu pras’owiaǝskiego”. Slavia 11, 1-32, 225-264. 1937 Review of Ekblom 1935. Rocznik Slawistyczny 13, 8-20. Minlos, Filipp R. 2001 “Ob odnom netrivial’nom fonetiǁeskom sootvetstvii (rus. vëx, zap.-ukr. ves’, morav. veš)”. Die Welt der Slaven 46, 117-122. Minlos, Filipp R. and V.A. Terent’ev 2002 “Rus. dial. vëx. Ètimologija”. Studia Linguarum 3/1-2 (Memoriae A.A. Korolev dicata), 517-545. Moscow. Moser, Michael 1998 “Urostslavisch oder Gemeinostslavisch?” Wiener Slavistisches Jahrbuch 44, 129-144. Nesselman, Georg H.F. 1873 Thesaurus linguae Prussicae …. Berlin. Nitsch, Kazimierz 1926 “Nature et chronologie de la seconde palatalisation en slave commun”. Revue des Études Slaves 6, 42-53. Novák, Karel 1934 Slovník k ǁeským spisDzm Husovým. Prague (quoted on the basis of the reprint, Munich, 1984). Ohala, John J. 1990 “The phonetics and phonology of aspects of assimilation”. In: John Kingston and Mary E. Beckman (eds.), Papers in Laboratory Phonology 1: Between the Grammar and Physics of Speech, 258-276. Cambridge. Orr, Robert 2000 Common Slavic nominal morphology, a new synthesis. Bloomington.

568 OtrNjbski, Jan 1948

WILLEM VERMEER

“Pochodzenie t. zw. Baudouinowskiej palatalizacji w jNjzykach s’owiaǝskich”. Slavia Occidentalis 19, 23-62, 524-525.

Pedersen, Holger 1905 “Die nasalpräsentia und der slavische akzent”. Zeitschrift für vergleichende Sprachforschung 38, 297-421. 1907 “Neues und nachträgliches”. Zeitschrift für vergleichende Sprachforschung 40, 129-217. 1913 Vergleichende Grammatik der keltischen Sprachen 2. Göttingen. Poržezinskij, Viktor K. [= Jan W. Porzeziǝski] 1914 Sravnitel’naja grammatika slavjanskix jazykov 1. Moskva. Posti, Lauri 1966 Kantasuomen äännehistorian luentojen pääkohdat. 2nd ed. Helsinki. Pracki, Witold 1907 “O mowie wsi Turowa”. Prace filologiczne 6, 187-276. Robert, Henry M. 1943 Robert’s Rules of Order Revised. Chicago etc. Rozwadowski, Jan 1897 “Quaestiones grammaticae et etymologicae III: De -ica suffixo linguarum slavicarum”. Rozprawy Akademii umiejNjtnoǧci, Serya 2 (Wydzia’ filologiczny), 10, 415-419. Sadnik, Linda and Rudolf Aitzetmüller 1955 Handwörterbuch zu den altkirchenslavischen Texten, Heidelberg. Šaxmatov, Aleksej A. 1896 “K istorii zvukov russkago jazyka”. IzvLJstija OtdLJlenija Russkago Jazyka i Slovesnosti 1, 695-743. 1913 Oǁerk sovremennago russkago literaturnago jazyka, S.-Peterburg. 1915 Oǁerk drevnLJjšago perioda istorii russkago jazyka. Petrograd. (Quoted on the basis of the reprint, The Hague, 1967.) Šǁepkin, Vjaǁeslav N. 1910 Review of Leskien 1909. Rocznik Slawistyczny 3, 203-220. Schenker, Alexander M. 1995 The Dawn of Slavic. An Introduction to Slavic Philology. New Haven & London. Schwartz, Geoffrey 2001 “Speech perception, sound change, and the Slavic palatalizations”. Folia Linguistica Historica 22/1-2, 277-300. Shevelov, George Y. 1964 A Prehistory of Slavic: The Historical Phonology of Common Slavic. Heidelberg. Skok, Petar 1971-1974 Etimologijski rjeǁnik hrvatskoga ili srpskoga jezika. Zagreb. Sl11-14 1988Slovar’ drevnerusskogo jazyka (XI-XIV vv.). Moskva. S’StP 1953S’ownik staropolski. K. Nitsch, Z. Klemensiewicz, St. Urbaǝczyk (eds.), Warszawa. Sobolevskij, Aleksej I. 1889 “Obšǁe-slavjanskija izmLJnenija zvukov”. Russkij Filologiǁeskij VLJstnik 22, 1-49.

PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION

569

1891 Drevnij cerkovno-slavjanskij jazyk: fonetika. Moskva. Sreznevskij, Izmail I. 1890-1912 Materialy dlja slovarja drevne-russkago jazyka. 3 vols. Sanktpeterburg. SSSl 1994 Staroslavjanskij slovar’. R.M. Cejtlin et al. (eds.), Moskva. Stang, Christian S. 1966 Vergleichende Grammatik der baltischen Sprachen. Oslo, Bergen, Tromsö. Stankiewicz, Edward 1973 “The historical phonology of Common Slavic (review article)”. International Journal of Slavic Linguistics and Poetics 16, 179-192. StƀSl 1968Staroǁeský slovník. B. Havránek et al. (eds.), Praha. Steensland, Lars 1975 “Ein beitrag zur diskussion anlässlich R. Channons On the Place of the Progressive Palatalization of Velars in the Relative Chronology of Slavic”. International Journal of Slavic Linguistics and Poetics 21, 90-106. Stokes, Whitley 1893 “On the assimilation of pretonic n in Celtic suffixes”. Indogermanische Forschungen 2, 1893, 167-173. Stüber, Karin 1998 The Historical Morphology of n-Stems in Celtic (Maynooth studies in Celtic linguistics 3). Maynooth. Townsend, Charles E. and Laura E. Janda 1996 Common and Comparative Slavic: Phonology and Inflection. Columbus, Ohio. Trubaǁev, Oleg N. 1959 Istorija slavjanskix terminov rodstva i nekotoryx drevnejšix terminov obšǁestvennogo stroja. Moskva. Trubetzkoy, Nikolaj S. 1922 “Essai sur la chronologie de certains faits phonétiques du slave commun”. Revue des Études Slaves 2, 217-234. 1954 Altkirchenslavische Grammatik. Schrift-, Laut- und Formensystem (Österreichische Akademie der Wissenschaften. Philosophisch-historische Klasse, Sitzungsberichte 228/4). Wien. Vaillant, André 1950 Grammaire comparée des langues slaves 1: Phonétique. Lyon and Paris. 1958a Grammaire comparée des langues slaves 2/1: Flexion nominale. Lyon and Paris. 1958b Grammaire comparée des langues slaves 2/2: Flexion pronominale. Lyon and Paris. 1965 Review of Shevelov 1964. Bulletin de la Société de linguistique de Paris 60/2, 124-127. 1966 Grammaire comparée des langues slaves 3: Le verbe. Paris. 1974 Grammaire comparée des langues slaves 4: La formation des noms. Paris. van Wijk, Nikolaas 1931 Geschichte der altkirchenslavischen Sprache 1. Laut- und Formenlehre. Berlin and Leipzig. 1949-50 “K istorii fonologiǁeskoj sistemy v obšǁeslavjanskom jazyke pozdnego perioda”. Slavia 19, 293-313.

570 Vasmer, Max 1907

WILLEM VERMEER

“Zwei kleine Abhandlungen. II. Zur slavischen Vertretung von arioeurop. o”. Zeitschrift für vergleichende Sprachforschung 41, 157-164.

Velǁeva, Borjana 1988 [= Velcheva, Boryana] Proto-Slavic and Old Bulgarian Sound Changes. Columbus, Ohio. Vermeer, Willem 1986 “The rise of the North Russian dialect of Common Slavic”. Studies in Slavic and General Linguistics 8, 503-515. 1997 “Notes on medieval Novgorod sociolinguistics”. Russian Linguistics 21, 23-47. 2000 “On the status of the earliest Russian isogloss: four untenable and three questionable reasons for separating the progressive and the second regressive palatalization of Common Slavic”. Russian Linguistics 24, 5-29. 2002-03 “Leading ideas in the study of the progressive palatalization of Proto-Slavic”. International Journal of Slavic Linguistics and Poetics 44-45 (Memorial volume H. Birnbaum), 377-394. 2003 “Comedy of errors or inexorable advance? Exploring the dysfunctionality of the debate about the progressive palatalization of Slavic”. Studies in Slavic and General Linguistics 30, 397-452. 2005 Review of Zaliznjak 2004. Voprosy jazykoznanija 2005/6, 125-129. 2006 “Jagiǀ’s chronology of the Progressive Palatalization.” In: Per Ambrosiani et al. (eds.), Jako blagopesnivaja ptica. Hyllningsskrift till Lars Steensland (= Acta Universitatis Stockholmiensis, Stockholm Slavic Studies 32), 229-240. Stockholm. MS “Why baba jaga? On restoration of velars subject to the Progressive Palatalization in the Russian-Belorussian linguistic area”. Viitso, Tiit-Rein 1998a “Fennic”. In: Daniel Abondolo (ed.), The Uralic Languages, 96-114. London [etc.]. 1998b “Estonian”. In: Daniel Abondolo, (ed.), The Uralic Languages, 115-148. London [etc.]. Vondrák, Václav 1906 Vergleichende slavische Grammatik 1: Lautlehre und Stammbildungslehre. Göttingen. 1912 Altkirchenslavische Grammatik. 2nd ed. Berlin. 1923-24 “O pozdLJjších palatalisacích v praslovanštinLJ”. Slavia 2, 17-25. 1924 Vergleichende slavische Grammatik 1: Lautlehre und Stammbildungslehre. 2nd ed. Göttingen. Zaliznjak, Andrej A. 1982 “K istoriǁeskoj fonetike drevnenovgorodskogo dialekta”. Balto-slavjanskie issledovanija 1981, 61-80. Moskva. 1986 “Novgorodskie berestjanye gramoty s lingvistiǁeskoj toǁki zrenija”. In: V.L. Janin, and A.A. Zaliznjak, Novgorodskie gramoty na bereste (iz raskopok 19771983 gg.), 89-219. Moskva. 1988 “Drevnenovgorodskij dialekt i problemy dialektnogo ǁlenenija pozdnego praslavjanskogo jazyka”. In: Slavjanskoe jazykoznanie. X Meždunarodnyj s”ezd slavistov, Sofija, sentjabr’ 1988 g., Doklady sovetskoj delegacii, 164-177. 2004 Drevnenovgorodskij dialekt. 2nd ed. Moskva.

PEDERSEN’S CHRONOLOGY OF THE PROGRESSIVE PALATALIZATION

571

Zierhoffer, Karol 1959 Ʀcieǻka i jej synonimy w gwarach i historii jNjzyka polskiego na tle ogólnos’owiaǝskim. Wroc’aw. Zubatý, Josip 1910 “Výklady etymologické a lexikální”. Sborník filologický 1, 95-164.

SUMMARY Pedersen’s chronology of the Progressive Palatalization (Prog) is the one by which Prog is held to have been significantly earlier than the Second Regressive Palatalization, which with most authors means earlier than the monophthongization of diphthongs (Pedersen), and with some even earlier than the First Regressive Palatalization (Martinet). The article critically examines all evidence that has been adduced in favour of Pedersen’s chronology and concludes that none of it is compelling.