New Perspectives and Issues in Educational Language Policy: A Festschrift for Bernard Dov Spolsky

  • 13 1,039 2
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up

New Perspectives and Issues in Educational Language Policy: A Festschrift for Bernard Dov Spolsky

New Perspectives and Issues in Educational Language Policy New Perspectives and Issues in Educational Language Policy

2,438 1,407 3MB

Pages 314 Page size 336 x 485.76 pts Year 2000

Report DMCA / Copyright

DOWNLOAD FILE

Recommend Papers

File loading please wait...
Citation preview

New Perspectives and Issues in Educational Language Policy

New Perspectives and Issues in Educational Language Policy A festschrift for Bernard Dov Spolsky

Edited by Robert L. Cooper The Hebrew University

Elana Shohamy Tel Aviv University

Joel Walters Bar-Ilan University

John Benjamins Publishing Company Amsterdam/Philadelphia

8

TMThe

paper used in this publication meets the minimum requirements of American National Standard for Information Sciences – Permanence of Paper for Printed Library Materials, ansi z39.48-1984.

Library of Congress Cataloging-in-Publication Data New perspectives and issues in educational language policy : a festschrift for Bernard Dov Spolsky / edited by Robert L. Cooper, Elena Shohamy, Joel Walters. p. cm. Includes bibliographical references and indexes. 1.Language and education. 2.Language policy. I.Spolsky, Bernard. II.Cooper, Robert Leon, 1931- III.Shohamy, Elena Goldberg. IV.Walters, Joel. P40.8. 2000 306.44’9--dc21 isbn 90 272 25613 (Eur.) / 1 55619 8558 (US) (Hb; alk. paper)

00-063085

© 2001 – John Benjamins B.V. No part of this book may be reproduced in any form, by print, photoprint, microfilm, or any other means, without written permission from the publisher. John Benjamins Publishing Co. · P.O.Box 75577 · 1070 an Amsterdam · The Netherlands John Benjamins North America · P.O.Box 27519 · Philadelphia pa 19118-0519 · usa

"wid"> "bia"> "coh"> "bac"> "dav"> "sho">

TITLE "Table of Contents"

SUBJECT ", Volume "

KEYWORDS ""

SIZE HEIGHT "220"

WIDTH "150" BOOKMARK "Table of contents">

Table of contents

Spolsky’s educational linguistics Robert L. Cooper, Elana Shohamy and Joel Walters

1

Part I Language teaching, language learning, and literacy The monolingual teaching and bilingual learning of English Henry G. Widdowson Literacy: The extension of languages through other means Ellen Bialystok Bilingual processing strategies in the social context of an undergraduate immersion program Andrew D. Cohen and Kirk C. Allison

7

19

35

Part II Language testing in educational contexts Some construct validity issues in interpreting scores from performance assessments of language ability Lyle F. Bachman

63

What second language learners can tell us about the native speaker: Identifying and describing exceptions Alan Davies

91

The social responsibility of the language testers Elana Shohamy

113

"toc">

vi

Table of contents

Part III Multilingualism, minority languages and language planning Multilingualism and diglossia in Palestine in the first century of the common era: Some thoughts on historical sociolinguistics Christina Bratt Paulston

133

Introducing Khayem Zhitlovski Joshua A. Fishman

145

Arab language education in the Hebrew state Muhammad Amara

155

Adult use and language choice in foreign language policy Richard D. Lambert

171

Linking levels in foreign language teaching through a common frame of reference Kees de Bot and Theo van Els

197

Part IV Case studies in educational linguistics: Focus on policy Micro language policy as a barometer of change: A university language policy as an example Michael Clyne

211

The genesis of a district-wide Spanish FLES program: A collaborative achievement G. Richard Tucker, Richard Donato and Kimmaree Murday

235

The distinctiveness of applied linguistics in Australia: A historical perspective Tim F. McNamara and Joseph Lo Bianco

261

Educational linguistics as a field: A view from Penn’s Program as it approaches its 25th anniversary Nancy H. Hornberger

271

Index

299



Spolsky’s educational linguistics Robert L. Cooper, Elana Shohamy and Joel Walters The Hebrew University / Tel Aviv University / Bar-Ilan University

The hallmark of Bernard Dov Spolsky’s career up to this point is, and we can say “up to this point” because (take a deep breath!) as we began work on this volume, Spolsky was completing his book with Elana Shohamy on Israeli language policy (Multilingual Matters, 1999), winding up the Concise Encyclopedia of Educational Linguistics (Elsevier, 1999), and finishing his project with Muhammad Amara and Hannah Tushiya on sociolinguistic patterns in Bethlehem; he was beginning work on a project with Norbert Dittmar and Joel Walters on language and identity among Russian immigrants; and he was continuing his indefatigable efforts in language policy and curriculum development on behalf of Israeli schools and research on the academic achievement of immigrant children in that context. His career began “Down Under” as a teacher of English and Politics (see Clyne’s reference to his New Zealand roots, this volume) and took him first to Australia and English as a high school teacher, then to Israel (his first foreign language immersion experience), then to North America (as student of linguistics at Université de Montréal and on the faculties of McGill University, Indiana University and the University of New Mexico), and finally back to Israel and Bar-Ilan University in 1980. But his institutional affiliations only tell part of the story. His work in Navajo and Maori education programs, in bilingual education efforts in North America, and in minority language concerns across the globe display recognition of a liberal duty to envelop research in social equity issues. Spolsky’s focus in this mission has been on linguistics and education, with language policy studies and policymaking the driving motivation behind his work. A career so broad in scope, so prolific and so vigorous in pursuit of social concerns makes choice of papers for a single volume a difficult task. His own publications number more than 250 (see his website http://www.biu.ac.il/ HU/lprc). The contributions to this volume are intended to reflect not only the

2

Robert L. Cooper, Elana Shohamy and Joel Walters

breadth of Spolsky’s interests, but also his conception of Educational Linguistics. In his introductory textbook on the subject (Spolsky 1978), educational linguistics (a term borrowed from educational psychology and educational sociology) is offered a model derived from theories of language, learning (psychology), language learning (psycholinguistics), and language use (sociolinguistics), all directed toward second language pedagogy. In an entry for The International Encyclopedia of Education (1985), Spolsky describes educational linguistics as a branch of applied linguistics, further specifying the following subfields: language education policy and planning; first and second language acquisition and teaching, reading, literacy, and composition; mother tongue and bilingual education, minority and immigrant education, and language testing. For Spolsky, language education integrates “the situation in which it takes place and the communicative competence of the learners who are to be educated” (1985: 3095). Four areas of inquiry are chosen to delineate the field: communicative competence, rationale/curriculum, policy, and assessment. This view has not changed substantially since then, although it has continued to be refined and enriched by both theoretical and empirical studies. Spolsky (1999) defines “the task of educational linguistics” as specifying “knowledge from the many and varied branches of the scientific study of language that may be relevant to formal or informal education”. Like Spolsky’s definition of the field, this volume reflects both the psycholinguistic and the sociolinguistic underpinnings of the intersection of language and education. The emphasis here is less on theory and theory-building, more on policy and practice. The volume is divided into four sections. The papers in Part I (by Widdowson, Cohen and Allison, and Bialystok) treat issues in language teaching, language learning and literacy, all topics of concern to Spolsky, from his early work published in TESOL Quarterly (e.g., Spolsky 1968, 1969) and climaxing in publication of his prize-winning Conditions for Second Language Learning (Spolsky 1989). Part II focuses on language testing, two papers addressing ethical issues from within the field (Bachman’s on validity and Davies on the notion of native speaker knowledge) and one which addresses language testing ethics head-on (Shohamy). Part III turns to multilingualism, minority languages and language planning. The section begins with Paulston’s look at the methodology of historical sociolinguistics using the languages of first century Palestine and Spolsky’s (1985) paper as her point of reference. Fishman takes us into the late 19th century, introducing Khayem Zhitlovski, who struggled with Russian, Hebrew and Yiddish multilingualism in making his mark on Socialism, Zionism, territorialism, Secularism and Yiddishism.

Spolsky’s educational linguistics

Amara’s paper offers a look at more current issues in multilingualism, focusing on Arabic, Hebrew and English in the Arab education system in Israel. The last two papers in the section turn to language planning, in particular with regard to foreign language learning. Lambert’s contribution, drawing on a wide variety of examples from throughout the world, examines domestic language policy from two perspectives: adult language use and language choice. De Bot and van Els discuss an issue raised in Lambert’s work, viz. linkage between levels, the paper centering on high school and university level courses in English and French in The Netherlands. Part IV converges on one of Spolsky’s most recent interests, affording the reader an insider’s view of language policy in a variety of contexts. Clyne’s paper tells the story of the creation and collapse of a language policy at Monash University. Hornberger documents the history of educational linguistics at the University of Pennsylvania. McNamara and Lo Bianco provide a chronicle of applied linguistics in Australia. In a spirit of optimism, the volume ends with Tucker, Donato and Murday’s report of an innovative elementary school program for promoting Spanish in suburban Pittsburgh. The paper describes efforts to integrate social action with educational linguistics, touching the lives of children, teachers, curriculum specialists and members of the community at large. Spolsky the educational linguist, Spolsky the writer, Spolsky the policy analyst, and Spolsky the social actionist all come through explicitly in the papers which follow. Embedded in these papers is a feature of the man and his work in a role he has practiced throughout his career — the role of mentor. A volume could have easily been assembled of work in a variety of fields from among his students. Rather, we chose contributors from among his peers, those mentored by Spolsky’s ideas. Widdowson acknowledges this mentoring debt explicitly at the outset of his paper. For this reason, and for its argument, for its literary polish, for the subtlety of its British and anti-British wit, and for the fact that it embodies the substance and politics of Bernard Spolsky, we chose it as the lead paper of this volume. A final note on convention: In the spirit of Spolsky’s plurilingualism, spelling and punctuation conventions in this volume reflect a polyphonic view of language, each author expressing his or her own rhetorical identity through these unedited conventions.

3



4

Robert L. Cooper, Elana Shohamy and Joel Walters

References Spolsky, B. 1968. “Recent research in TESOL”. TESOL Quarterly 2: 305–306. Spolsky, B. 1969. “Recent research in TESOL (2)”. TESOL Quarterly 3: 355. Spolsky, B. 1978. Educational Linguistics: An Introduction. Rowley, MA: Newbury House Publishers. Spolsky, B. 1985a. “Educational linguistics”. In The International Encyclopedia of Education, T. Husen and T. N. Postlethwaite (eds), 3095–3100. Oxford: Pergamon. Spolsky, B. 1985b. “Jewish multilingualism in the first century: An essay in historical sociolinguistics”. In Readings in the Sociology of Jewish Languages, J. A. Fishman (ed.), 35–50. Leiden: E. J. Brill. Spolsky, B. 1989. Conditions for Second Language Learning: Introduction to a General Theory. Oxford: Oxford University Press. Spolsky, B. 1999. Introduction and editor. The Concise Encyclopedia of Educational Linguistics. Amsterdam: Elsevier. Spolsky, B. and Shohamy, E. 1999. The Languages of Israel: Policy, Ideology and Practice. Clevedon: Multilingual Matters.



Part I

Language teaching, language learning, and literacy



The monolingual teaching and bilingual learning of English Henry G. Widdowson University of London, University of Vienna

Bernard Spolsky’s intellectual range is impressively wide and it would be difficult for his friends and admirers contributing to this Festschrift to take up issues relevant to its title that he has not himself touched on in one way or another. But with perspectives there is still scope for some novelty. It is always possible to take a fix on ideas from a different point of view. That is what I intend to do in this paper. I want to look at two areas of enquiry which Spolsky has himself extensively dealt with, and take up a position that shows them to be crucially inter-related. The two areas are second language learning and bilingualism. In the Preface to his book Sociolinguistics, Bernard Spolsky makes characteristically generous reference to the debt he owes me in the way of intellectual stimulation over the many years of our acquaintance. It seems appropriate that I should restore the balance, and acknowledge a corresponding debt to him by taking as the text for my contribution to his Festschrift a quotation from that book. It reads as follows: However explained neurophysiologically, the phenomenon of bilingualism is the prime example of language contact, for the two languages are in contact in the bilingual. (Spolsky 1998: 49).

I find this remark stimulating in that it sets off a whole series of thoughts about the inducement of bilingualism in foreign language learners, and what I should like to do in this paper is to pursue these thoughts, in a fairly informal way, and see where they take us. I have the feeling that they might lead us to question some of the currently orthodox assumptions about language pedagogy, but we shall see. We can begin by noting that what Spolsky is suggesting here is that language contact, traditionally conceived of in phylogenetic terms, can be seen as

8

Henry G. Widdowson

ontogenetically represented as bilingualism in the individual. But of course the only evidence of contact between languages is the change that is brought about as a result. It would seem to follow that if the contact occurs in the individual at least one of the languages must be affected, so the resulting bilingualism must be of a compound and not co-ordinate kind, since what defines the latter is precisely that the two languages are kept apart (see Spolsky 1998: 48). Spolsky speaks of bilingualism as an achieved state (although acknowledging that the definition of that state is problematic) but what of bilingualization as the process of acquiring an additional language, such as is induced in second language pedagogy? There is plenty of evidence of contact here, of course. This is often described negatively as interference, and it is a common assumption that its effects need to be minimised, and eliminated completely if at all possible. Indeed the idea that first language should be avoided at all costs in the second language classroom, and translation resolutely discouraged, is based on the belief that contact between the two languages is the last thing you want. So it would appear, paradoxically enough, that conventional language teaching procedures are designed to stifle bilingualism rather than promote it, if it is to be defined as two languages in contact in the individual. One can only suppose that the purpose of second language pedagogy in its avoidance of contact is to promote co-ordinate bilingualism in learners directly. But all the evidence is that although the teacher may strive to induce co-ordinate bilingualism, what learners do is to go through a process of compound bilingualization through interlanguage stages. And it would seem generally to be the case that, at least when it comes to the consecutive (as distinct from the simultaneous) acquisition of an additional language, learning has to pass through a compound period before co-ordinate competence can be achieved — that well-attested stage when one begins to “think in the foreign language”. It would appear, then, that we have a disparity here between teaching procedures which are directed at keeping languages co-ordinately separate, and learning that can only proceed by compounding them. This would seem to be rather a curious state of affairs, and one that warrants a little further enquiry. The conventional wisdom that holds that monolingual teaching is the best way of getting bilingual results dates back a century at least and is a legacy of the Direct Method (see Howatt 1984). As is made clear by Spolsky’s contribution to my own Festschrift it was endorsed by linguistic authority in the devising of the Army Specialised Language Program (ASTP) during the Second World War. It was sustained by the behaviourist notion that learning was essentially a matter of regulating responses by controlled exposure to the target language. This was

Monolingual teaching and bilingual learning

supposed to have the effect of establishing habits which would override interference from the learner’s mother tongue, and so counteract bilingual compounding. Such interference was anticipated by contrastive analysis (Fries 1945; Lado 1957). But this was to be drawn upon in the preparation of teaching materials in advance; it was not to be overtly exploited in classroom activities as an aspect of methodology. The presence of the L1 in the process of learning the L2 was given no pedagogic recognition. As Stern points out: Contrastive analysis was not intended to offer a new method of teaching; but it was a form of language description across two languages which was particularly applicable to curriculum development, the preparation and evaluation of teaching materials, to the diagnosis of learning problems, and to testing. (Stern 1983: 159).

Contrastive analysis was designed for diagnosis and prevention: it was to provide the prophylactic means whereby the learning of the L2 might, as far as possible, be protected from L1 contagion. It therefore made common cause with behaviourist theory, and directed at making monolingual teaching more effective, the essential purpose of which was to induce a separate modular development which at the end of the day yielded another language to be added as a coordinate to the one that learners already had. What is curious, however, is that assumptions of the validity of monolingual pedagogy should have survived the discrediting of contrastive analysis as a diagnostic device, and the demise of behaviourist ideas. For with more cognitive views of learning comes the realisation that learners cannot be immunised against the influence of their own language. There is bound to be contact, and it is of the very nature of language learning that it is of its nature, in some degree, a compound bilingual experience. Just what kind of contact this is, and the extent of its effect (how far, for example, it is mediated by principles of Universal Grammar) still remain matters of controversy and conjecture. But there seems to be no question that this effect, or transfer in SLA parlance, is a crucial phenomenon to be reckoned with, and one, furthermore, which is not to be avoided but positively welcomed. As Ellis puts it: … there is now clear evidence that the L1 acts as a major factor in L2 acquisition. One clear advance in transfer research has been the reconceptualization of the influence of the L1, whereas in behaviourist accounts it was seen as an impediment (a cause of errors), in cognitive accounts it is viewed as resource which the learner actively draws on in interlanguage development. (Ellis 1994: 343).

9

10

Henry G. Widdowson

But this raises a tricky question. If the L1 is no longer viewed negatively as an impediment, which teachers would naturally wish to avoid or remove, but positively as a resource, then why, it seems reasonable to ask, do teachers not use it as such as well? If learners actively draw on this resource why, one wonders, do teachers not take advantage of this fortunate circumstance and devise activities in class which encourage learners in the exploitation of this resource? The influence of the L1 may have been reconceptualized in SLA research, but this seems to have brought about no corresponding reconceptualization in second language teaching. Here the L1 hardly figures at all, and certainly not as a major factor. The dominant pedagogy is still monolingual. One might, of course, take the view that the beneficial effects of the L1 depend on its not being overtly recognised as a resource, that to be effective the learners’ active drawing upon it has to be an undercover operation, a process of resistant autonomy: if the resource were to be officially sanctioned, and incorporated into pedagogy, it would lose its appeal. From this arises the interesting thought that some learning at least may result from a reaction against what is taught, so that teaching (some teaching at least) might be deliberately designed to provoke a resistant response. So if it is the case that the development of learning depends on the learners naturally drawing on their L1 as a resource, one way of ensuring that they keep on doing this is to insist that they should not. It has been common knowledge in human affairs since the Garden of Eden that nothing is more enticing than what is forbidden by authority. Monolingual teaching would be justified (indeed positively required) in a provocative pedagogy of this kind. However, the deliberate provocation of resistance is not a principle which would be easy to practise, and to my knowledge it has never been seriously proposed. What has been seriously proposed is what we might call a permissive pedagogy: one which allows for, even encourages, the learners’ engagement of the L1, but again makes no acknowledgement of its existence in the design of instruction itself. Monolingual teaching is justified in this case on the grounds that input in the L2, so long as it is comprehensible, will automatically activate learning (see Krashen 1985 and passim). Since this learning proceeds through interlanguage stages which are themselves, as Ellis points out, subject to substantial L1 influence (“a major factor”), then in effect this permissive approach to teaching is based on the assumption that learners will draw on the resources of their own language anyway, and that language contact will happen in the learning process without any need for it to be explicitly promoted by actual teaching. The monolingual L2 input will be naturally converted into an intake which is modified by L1 influence.

Monolingual teaching and bilingual learning

But for this to happen the input has to be comprehensible. The question arises as to how the learner makes the input comprehensible: in part, no doubt, by reference to context, but in part also, one must suppose, by invoking L1 equivalents. It would seem unlikely, to say the least, that the learner takes in items of the L2 per se and only subsequently, in the process of interlanguage development relates them to the L1. The L1 is surely in on the act from the start in processing the input, and in a sense can be said to act as a kind of filter. If this is the case then, again, one might plausibly argue that explicit reference to the L1 would assist the learner in making the input comprehensible. But such a possibility does not seem to have been considered. In discussions as to how pedagogy might be designed to facilitate the comprehensibility of the input, translation does not figure. The only filter that is mentioned (the affective filter) concerns the temporary disposition of the learner. The well established notion that new knowledge (e.g., of L2) is only acquired, and recognised as new at all, by reference to what is familiar (e.g., in L1) has no place, it seems, in this scheme of things. So, one of the most striking features of monolingual second language teaching (when one comes to think about it) is that it takes no principled account whatever of a major factor in second language learning. The L1 resource that learners draw on so extensively is almost entirely ignored. As a result, L2 teaching practises a kind of sustained pedagogic pretence that it is dealing with only one language, whereas it is obvious that as far as L2 learning is concerned there are (at least) two languages involved. While teachers are busy trying to focus attention on the L2 as distinct from the L1, thereby striving to replicate conditions of coordinate bilingualism, the learners are busy on their own agenda of bringing the two languages together in the process of compound bilingualization. That the processes of teaching and learning should be so at odds is surely, to say the least, odd. This disparity comes about, it seems to me, because of a failure to define the second or foreign language as a pedagogic subject. It seems to be generally assumed as self-evident that the language that learners are to learn, and teachers to teach, is that which is experienced by its users in normal contexts of use. So it is supposed that if you are teaching English, then what you are teaching is the English that occurs in such contexts. And now that we have reliable and detailed evidence to hand in corpus descriptions of what users of the language actually produce, we are in a position to specify just what this English is, and so define our subject accordingly (see Sinclair 1995). But our subject is not English as experienced by its users as a first and familiar language, but by its learners as a

11

12

Henry G. Widdowson

second and foreign language. It is the foreignness of the language which is most obviously apparent and problematic for learners, and the foreignness, therefore, which must be accounted for in the definition of the subject to be taught. Now clearly the notion of a second language implies the existence of a first, and you cannot recognise what is foreign in a language without relating to another which is familiar, and, of course, for different groups of learners the second language will be foreign in different ways. So as soon as we accept that the subject we are teaching is a foreign language, then at least one other familiar language (typically the L1) is necessarily implicated. In other words, the very subject we teach is, by definition, bilingual. How then can you teach a bilingual subject by means of a monolingual pedagogy? The answer, surely, is that you can’t. And one might hazard the suggestion that it may well be that many of the problems that monolingual language pedagogy has sought to grapple with over the years are self-inflicted, and therefore, of course, also inflicted on learners; that much of its energy has been devoted to resolving difficulties of its own making rather than those that learners themselves encounter in the learning process. The question naturally arises as to why, in the face of this contradiction, monolingual language teaching, particularly in the case of English, has persisted for so long, and has never been seriously challenged. One reason springs immediately to mind: namely that those who have most cause to challenge it, are in no position to assert themselves. The learners’ challenge is covertly expressed through their own bilingual compounding, and they have no alternative but to connive in the pedagogic pretence imposed upon them. Their non native-speaker teachers are for the most part persuaded by those in native speaker authority that to acknowledge the bilingual nature of the subject is to diminish it, that any concession to the L1 in the classroom amounts to a betrayal of pedagogic principle and tantamount to unprofessional conduct. And over recent years, with the development of discourse linguistics and the availability of corpus based descriptions, these teachers have been enjoined not only to restrict themselves to the L2, but to the L2 as it really is, as it is pragmatically realised in actual contexts of native speaker use. Since these teachers, generally speaking, have had only limited, if any, experience of such use, such an injunction makes impossible demands upon them. They are, in fact, doubly disadvantaged. On the one hand, whatever the learners may do under their own steam, their teachers are not supposed to actually teach them by drawing on the shared resources of the L1 they both know. On the other hand, the teachers are told that the features of the L2 that they should focus on

Monolingual teaching and bilingual learning

are those that bear the hallmark of authentic use, which for the most part they are in no position to recognise, let alone teach. So it is that they are denied the exploitation of a bilingual competence they have and at the same time exhorted to exploit a communicative competence they do not. Against this one might argue that now that the findings of corpus linguistics are becoming available, non native teachers do have direct access to the facts of actual usage, and so can now learn to recognise authenticity when they see it. They are no longer dependent on the imperfect intuition of the native speaker: they need only consult the concordance. The difficulty here is that the concordance will only reveal usage as recurring patterns in the text that people produce. It will not reveal the discourse process whereby meanings are pragmatically achieved under various contextual conditions. It will not tell you in what particular circumstances it is appropriate to use the patterns of occurrence on display. Information about usage on a screen will not provide you with a vicarious experience of use: it will not magically transform into an ability to negotiate meaning or find your way through the socio-cultural subtleties of communication. You will know more about the language but this will not make it any less foreign. Indeed it may well be that it will indicate more clearly just what is foreign about it. And since the usage that is revealed presupposes complex conditions of use that remain unrevealed, the effect could be the discouraging one of bringing home to you how little you actually know of the language you are supposed to be teaching. It is of course in the interests of native-speaking institutions to disregard the foreignness of the language, promote monolingual pedagogy and define the subject in terms of what is to be eventually achieved rather than the process of bilingualization that must be engaged in order to achieve it. The ideas about how a foreign language should be taught do not come from an experience of its foreignness. Their most common source is the experience of native speaker teachers who typically do not even know the language of their students, at least not to the extent that they can draw confidently upon it as a resource, so they have no choice but to focus exclusively on the language they do know. And where the teaching takes place in the L2 country itself, as in English language schools in Britain, for example, students come from a variety of linguistic backgrounds, and so it is not feasible to teach in any other way. Thus monolingual teaching makes a virtue of necessity. It is necessary as an expedient solution to the problems of the bilingual incompetence of teachers and the multilingual competence of students. To make it a virtue one claims that the subject is essentially the real language as experienced by native speakers users, and this is

13

14

Henry G. Widdowson

a reality, of course, to which native speakers have privileged access. It is after all their language. As owners of it, they have special (if not exclusive) rights to it as a resource, to be exploited and marketed like any other commodity. But is it their language? If English is an international language, how can native speakers in Britain (or anywhere else for that matter) claim rights to it? This too is a question worth pondering. In the section on “Satisfying the demand for English” in the 1998 British Council annual report the opening statement reads: “English language teaching is a major British export”. And inside that section we find highlighted an endorsement by the British Foreign Secretary: I have been round many countries where I have visited the British Council and for every pound spent on the British Council it often produces much more business for Britain, particularly in English language teaching. (The British Council 1998: 10).

In the same spirit of commercial enterprise, the report on the future of English in the world (Graddol 1997) talks of the importance of establishing a British brand in the global market for the language: A debate would be timely on how Britain’s ELT providers can cooperatively prepare for the need to build and maintain the British brand and how the promotion of English language goods and services relates to the wider image of Britain as the leading-edge provider of cultural and knowledge-based products. (Graddol 1997: 63).

But there are difficulties about this attitude to English. One is that it disregards the rather obvious point that the language as an international means of communication can no longer be the property of a particular group of its native speakers (see Widdowson 1993, 1994). There is a preface to Graddol (1997), written by no less an authority than His Royal Highness The Prince of Wales. Its first sentence reads: “English has become the world’s global language”. And its last: “I commend this work to all who see a strong and vigorous future for our language”. There is an assumed equation here between English “a major British export” and English “the world’s global language”. But the two are not the same at all. English has become a global language precisely because it has extended beyond its original parochial confines and is no longer exclusively British. The language that the Prince of Wales speaks of as his own is L1 English as a national and domestic language, not English as an international and global one, and you do not provide for the latter by exporting the former. Obviously enough, once the language goes global, it necessarily loses its domestic L1 status.

Monolingual teaching and bilingual learning

And whereas speakers of English as a domestic language are typically monolingual, global English, for a large and increasing number of its users, coexists bilingually with other languages. English can only have a strong and vigorous future to the extent that it is not promoted as the monolingual property of its native speakers. And you cannot export what you have not got. So what is being branded as British and globally promoted cannot be the global language, by the very fact that it is branded as British. That is one confusion. There is a second, and it brings us back to the main theme of this paper: not only is British English not the global language, but British English language teaching is not the subject English as a foreign language either. The British cannot brand the subject because they cannot know what it is, for, as I have argued, the foreignness is a function of the particular relationship that English contracts locally with other languages, and so it will be foreign in all kinds of different ways for all kinds of different learners. And it is crucially their reality which pedagogy must account for. Britain’s ELT providers cannot meet the needs of the subject by co-operating internally with themselves: they have to co-operate externally with people for whom the language is foreign. Such external co-operation has not figured very prominently in the past. Instead, as Phillipson argues, the providers of expertise have tended to consult their own internal interests (Phillipson 1992). But this, it seems, is precisely what Graddol proposes for the future too. Promoting a British brand of the English product can only sustain the very proprietorial and hegemonic attitude to the language that Phillipson deplores. It is all the more curious to note (in passing) that Phillipson finds that Graddol’s book is one he “can warmly recommend” as “a very astute, sober and scientifically informed book on global English” (Phillipson 1999). But perhaps he did not get as far as the last chapter. This view of English as a British branded product is, I believe, mistaken on two counts. It fails to recognise that the English we are concerned with is not that which is experienced by its native speakers in Britain (or any other L1 setting) but that which on the one hand serves as an international means of communication, and, on the other hand, has to be learned in classrooms. In other words, our concern is with English which is global in its use, and local in its learning. The established position is diametrically the opposite to this, for it sees English as local in its use and global in its learning: the language to be taught is that which native speakers locally produce, and its monolingual teaching is assumed to be exportable as a global commodity, universally applicable, whatever the local circumstances of the learners.

15

16

Henry G. Widdowson

I have argued that this monolingual teaching is at odds with the bilingualization process which learners necessarily engage in when they draw on the language they know as a resource for learning the language they do not. One obvious way of dealing with this disparity is to devise a bilingual pedagogy which exploits this process and seeks to direct it. Such pedagogy would involve bringing contrastive analysis into classroom methodology in the form of translation activities of one kind or another. If transfer is such a major factor in interlanguage development, as according to Ellis the SLA literature makes plain, then it is hard to resist the conclusion that translation should be a major factor in the teaching which seeks to induce it. This is not, of course, to say that there will be no resistance to the conclusion. Translation has a bad name in language teaching circles, particularly in those which have generally called the tune of pedagogic fashion where it is not an option anyway. It is commonly associated with the universally condemned grammar-translation method. But over recent years there has been a rehabilitation of grammar, brought about in large part by the recognition of its role as a necessary resource in the achievement of pragmatic meaning. There is no reason why translation should not be rehabilitated as a resource in like manner. As the concept of grammar has changed to take on a more communicative character, so the concept of translation needs to be changed accordingly. Translation has been too long in exile, for all kinds of reasons which, as I have tried to indicate, have little to do with considered theoretical justification. It is time it was given a fair and informed appraisal. In this paper I have argued for a bilingual approach to teaching essentially on the practical pedagogic grounds that it goes with the grain of the learners’ experience whereas monolingual teaching goes against it. But there is another kind of argument that supports it: not so much a pedagogic as an educational one. Over recent years much concern has been expressed that the promotion of English as “the world’s global language” has the effect of diminishing the status, and the use, of other less prestigious languages. Thus English only flourishes at their expense (see, for example, Pennycook 1994). This concern is impressively confronted in a recent article written by two teachers of English, originally from India, but now working in South Africa. The article does not simply lament the hegemony of English but makes practical proposals, based on the experience of the authors themselves, about “what ELT professionals can do to promote indigenous languages in multilingual societies like India and South Africa” (Joseph and Ramani 1998). What they propose recognises the need to make explicit reference to the other language(s) that the learners already know, and

Monolingual teaching and bilingual learning

they lend empirical weight to the proposal by giving an account of translation activities in class. In the terms of this present paper, what Joseph and Ramani do is essentially to redefine English to make it a foreign and therefore bilingual subject. So defined, the other languages are no longer suppressed, or at best reluctantly condoned, but overtly recognised and made prominent as necessarily part of what second language learning is all about. If we acknowledge that TESOL — the teaching of English to speakers of other languages — cannot effectively be done without reference to the other languages, that guiding the development of bilinguals has to be attuned to the bilingualization process, and not by the imposition of an exclusive monolingual pedagogy, then there is some hope at least that English can be learnt without denying the legitimate rights of less privileged minority languages. I began this paper by referring to issues that have featured prominently in Spolsky’s work, and it was appropriate that I should use a quotation from him as the text to start my argument. It seems appropriate to end with a quotation which expresses a perspective on where my argument has (eventually) arrived. I can do no better than to quote from the conclusion of the Joseph and Ramani article: We strongly believe that ELT specialists need to radically reconceptualize their role in language education. As a profession we should openly support multilingualism in education, and problematize our current role as English language teachers. In addition, we need to identify and create opportunities within our own institutions for the fullest development of multilingualism. (Joseph and Ramani 1998: 222).

Knowing Bernard Spolsky as I do, I feel sure that he would approve.

References Ellis, R. 1994. The Study of Second Language Acquisition. Oxford: Oxford University Press. Fries, C. C. 1945. Teaching and Learning English as a Foreign Language. Ann Arbor: University of Michigan Press. Graddol, D. 1997. The Future of English? London: The British Council. Howatt, A. P. R. 1984. A History of English Language Teaching. Oxford: Oxford University Press. Joseph, M. and Ramani, E. 1998. “The ELT specialist and linguistic hegemony: A response to Tully and Mathew”. English Language Teaching Journal 52 (3): 214–222. Krashen, S. 1985. The Input Hypothesis. London: Longman. Lado, R. 1957. Linguistics across Cultures: Applied Linguistics for Language Teachers. Ann Arbor: University of Michigan Press.

17



18

Henry G. Widdowson

Pennycook, A. 1994. The Cultural Politics of English as an International Language. London: Longman. Phillipson, R. 1992. Linguistic Imperialism. Oxford: Oxford University Press. Phillipson, R. 1999. “Voice in Global English: Unheard Chords in Crystal Loud and Clear”. Applied Linguistics 20 (2): 265–276. Sinclair, J. M. 1995. “Selected issues”. In English in the World, R. Quirk and H. G. Widdowson (eds), 12–24. Cambridge: Cambridge University Press. Spolsky, B. 1995. Conditions for Second Language Learning. Oxford: Oxford University Press. Spolsky, B. 1998. Oxford Introductions to Language Study: Sociolinguistics. Oxford: Oxford University Press. Stern, H. H. 1983. Fundamental Concepts of Language Teaching. Oxford: Oxford University Press. The British Council Annual Report. 1998. London: The British Council. Widdowson, H. G. 1993. “Proper words in proper places”. English Language Teaching Journal 47 (4): 317–329. Widdowson, H G. 1994. “The ownership of English”. TESOL Quarterly 28 (2): 377–388.



Literacy The extension of languages through other means Ellen Bialystok York University

For the nonspecialist, understanding how children learn to speak one language seems not to have too much mystery about it. In fact, a not uncommon response from some portion of the general public to the idea that there is a minor scientific industry involved in studying how children manage to learn the language of their home is mild incredulity. The same subset of the general public may acknowledge that a slightly more interesting and less obvious case can be made for studying how children learn two languages, either simultaneously or successively. Even here, however, the question is fraught with common myth and popular legend. These conceptions are not always expressed as theories but emerge as actions: parents who are reluctant to “confuse” their children by exposing them to more than one language or educators who claim special access to knowing “when” instruction in a second language will be most effective. The possibility that scientific study has any contribution to make to these issues is not seriously considered. The case for reading is somewhat different. People generally recognise that most children need to be taught the basics of reading (although there is considerable controversy about the appropriate way to approach this instructional problem) and would probably acknowledge as well that research into the process of reading may be worthwhile. The problem of how to read in a second language, especially if proficiency in that language is weak or the language is written in a different script from the first language, is more clearly an issue requiring study. In spite of this progression of areas for which the perceived need for research is increasingly obvious, the actual output of research appears to follow precisely the reverse order. Language acquisition, in particular acquisition of a

20

Ellen Bialystok

first language, is fecund ground for research initiatives, learning to read in a first language is a respectable area with a growing proliferation of empirical study, but learning to read in a second language remains a relatively ignored domain of inquiry. There is some research investigating aspects of second-language reading for adults who are already literate and presumably fluent in the second language (e.g., Bernhardt 1991; Carrell, Devine and Eskey 1988). The issues in these cases tend to focus on the processing balance that needs to be achieved between topdown conceptual processes concerned with the meaning of the text, the background knowledge of the reader, and the linguistic proficiency in that language, and bottom-up decoding processes such as visual recognition. We also know from the reading research in a first language, conducted with monolingual populations, that this balance changes as reading becomes more skilled. Presumably, it changes as well as readers become more sophisticated, more knowledgeable, and undoubtedly, more linguistically apt. Therefore, there may be little from this kind of research that allows us to address the problem of how bilingual children learn to read in a second language. And yet, if public intuition is any guide, the issue of how young children about to enter the liberating world of literacy in a language that is either not their primary system or one of two languages they are mastering is a matter worthy of investigation. The problem of how young bilingual children learn to read is ripe for empirical analyses. The acquisition of literacy includes the need to leap many cognitive barriers and master many linguistic insights. It is undoubtedly the linguistic issues, however, that will shed the most light on those aspects of literacy acquisition that are unique to bilingual children, or to children entering literacy in a weak language. Questions such as the relation between language and the printed forms, the notational encoding of speech, and the role of syntactic structure in organising print are resolved differently in different languages and different writing systems, and children being literate in more than one language may need to grapple with alternative accounts of these issues. These are fundamentally linguistic questions, and issues for educational linguistics in particular. Importantly, these are also the questions that most depend on the languages the child speaks and the linguistic relationship between those languages. In this chapter, I will discuss how the child’s bilingualism in general and the relation between the two languages in particular influence progress in learning to read. The focus of the discussion will be to examine cases in which the specific pairs of languages used by the bilingual is the relevant variable in mastering a particular level of proficiency in learning to read.

Literacy: Languages through other means

The discussion will be organised around three stages in the path to literacy: concepts of print, entry to literacy, and fluent reading.

What it means to be written down It sounds trivial, but before children can learn to read they must figure out what the marks on the page are doing there. They learn these forms early, reciting the alphabet before they can construct full sentences and printing their names before they can properly hold a pencil or name the colour of the ink. But how do they interpret these notations? An important series of studies by DeLoache and her colleagues has shown that children between the ages of 2½ and 3½ years old make startling progress in figuring out the “meaning” of different kinds of representational systems. The progress is startling for two reasons: first, it is achieved rapidly and completely for each kind of representation, and second, the developmental sequence and age of insight is remarkably uniform. She has shown that 2½ year olds suddenly understand what is represented in a picture (DeLoache and Burns 1994) or video (Troseth and DeLoache 1998), and 3½ year olds eventually figure out how a model can represent a space (DeLoache 1987). Before these insights, children frequently encounter and apparently understand pictures, videos, and models. However, the research shows that until these landmark developments, they do not understand the significance of those forms. The same is true for figuring out the meaning or purpose of the scribbles on the page. The first compelling evidence that children did not understand the significance of print was largely anecdotal. Ferreiro (1978, 1983, 1984) interviewed children in informal interactions as they were exploring print. She discovered consistent misconceptions children had about the nature and meaning of the writing system they were learning. Children believed, for instance, that the print needed to embody characteristics of the word it signified so that big words were written with big letters, and groups of many objects were represented by repeated letters or words. In short, they treated print as an alternative to drawing, believing it to be another form of graphic representation that indicates its meaning through its visual form. For print, however, nothing could be farther from the truth. We pursued some of these ideas in more formal experiments. The significance of print is that it represents words in spoken language. Because it is a symbol, the same printed form will always represent the same word. Further, if

21

22

Ellen Bialystok

the writing system is an alphabetic language, then the means of achieving this is through the individual representation of the sounds in the word. This is the significance of print, and if children have any idea what the writing system is for, they will know that these two conclusions follow. We developed two tasks, one to assess each of these notions of print. The first was the moving word task. We showed the children pictures of common objects and named them, for example, a king and a bird. Then a card was introduced on which was printed the name of one of the pictures. The experimenter told the child what was written on the card: “This card has the word king on it. I’m going to put the card right here”. The card was then placed under the picture it named, although that correspondence was never explicitly drawn to the child’s attention. The child was asked what was written on the card. Some confusion then erupted, and a stuffed animal on the table created a disturbance that resulted in the card being displaced so it was under the other picture. Then the child was asked for the second time what was written on the card. The experimenter then noticed that the materials had been changed, commented on it to the child while returning the card to its original position, and asked the child for the third time what was written on the card. The critical question is the second one, asked when the word is under the wrong picture. If children understand that the word is a symbol for a spoken word, they will realise that its position on the table is irrelevant to determining its identity. However, if print is just some form of depiction that takes its meaning from the environment and other contextual cues, then the contiguity to a specific picture may be very relevant indeed. All three studies included groups of monolingual pre-readers who were 4- and 5-years old, but two of the studies also examined bilingual children of the same age. These children were bilingual in either English and French (Bialystok 1997), English and Chinese (Bialystok 1997), or English and Hebrew (Bialystok, Shenfield and Codd 2000). Across all the studies, the bilingual children outperformed the monolingual children by a large margin, often revealing more than a year advantage in understanding this principle. On average, the monolinguals were correct about 40% of the time and the bilinguals, about 80%. The bilingual children were comparable to each other and usually solved the problem equally in both their languages. Just being exposed to two writing systems, or two kinds of storybooks, enabled bilingual children to appreciate that the written forms are the symbolic system from which the story emerges. The second task was the word size problem. In alphabetic languages, the writing system functions because letters represent sounds. It follows, then, that

Literacy: Languages through other means

words that have more sounds will need more letters to represent them. If children realise that individual letters stand for specific sounds, then they should be able to match long strings of letters with words that have long sequences of sounds, even if they cannot read. Similarly, short words with few sounds should correspond to written strings containing few letters. This insight is a prerequisite to reading — children simply could not learn to read in an alphabetic system without understanding this relation. The task required children to match printed words to spoken words under various conditions, sometimes accompanied by pictures or written materials. For half of the items, there was incongruity between the size of the word and the size of the object it named. For example, in the pair train-caterpillar, the longer word names the smaller object. Children’s ability to achieve this depended not only on whether or not they were bilingual but also on what kind of writing system was used in the two languages. The task is very difficult, and monolingual children and French-English bilingual children struggled with the problem and showed exactly the same patterns of performance and development (Bialystok 1997). Children who were bilingual with Chinese and English revealed a different pattern. At 4-years old, these children understood virtually nothing about the task and essentially behaved at chance levels for all questions. This is significantly worse than the performance of the monolinguals and French-English bilinguals. By 5-years old, however, the Chinese-English bilinguals figured out the system and were solving the problems significantly better than either of the other two groups (Bialystok 1997). Moreover, once the rules were clear, they could solve the problem in both languages, that is, based on two different kinds of writing systems. Why were the 4-year-old Chinese-English bilinguals so baffled by the task? Their two writing systems differ from each other in two ways: they are based on different visual forms and encode different linguistic units, namely, sounds or morphemes. Either or both of these may have been responsible for the different pattern of discovery experienced by the Chinese-English bilinguals. We examined these possibilities by presenting the same problems to children who were bilingual with English and Hebrew (Bialystok, Shenfield and Codd 2000). Like English and French, Hebrew is alphabetic, but like Chinese, the visual forms are different from the Roman alphabet. The results of this study were clear — not only did the initial deficit encountered by the bilingual Chinese-English children disappear, but also there was an advantage in solving the problem already apparent by 4-years old. It appears that the need to learn two different visual forms for representing sounds help draw the child’s attention to those

23

24

Ellen Bialystok

sound-symbol correspondences. The need to learn two different correspondence rules, however, as in learning both a character and an alphabetic language, increases the difficulty of the task to the point that the younger children are simply confused. The conclusion from these studies of children’s initial concepts of print is that general statements about the effect of bilingualism on the acquisition of these concepts are simplistic. Before children can learn to read, they need to understand the function and significance of the written forms. Their progress in doing so depends on their experience with two kinds of forms, and on the nature of the two formal systems to which they are exposed. In other words, there are both general influences of bilingualism and specific effects of the languages involved that jointly determine children’s progress in establishing the foundational concepts for literacy. An analysis that omits this level of linguistic detail would fail to reveal these important relations.

Breaking the code In the early stages of reading, the main effort is on decoding the text. Children struggle with the correspondences and painstakingly attempt to construct meaningful wholes out of the arcane notations. The one metalinguistic skill that has repeatedly been shown to be predictive of children’s success in this enterprise is phonological awareness (e.g., Ellis 1990; Liberman, Shankweiler, Liberman, Fowler and Fischer 1977; Morais, Bertelson, Cary and Alegria 1986; Perfetti, Beck, Bell and Hughes 1987, 1988). At least for children who are learning to read in alphabetic languages, the concept of phoneme is crucial background to becoming a fluent reader. At the same time, these skills appear to have little formative role for children who are learning to read in nonalphabetic scripts (Mann 1986; Read, Yun-Fei, Hong-Yi and Bao-Zing 1986). Metalinguistic abilities have been a fruitful arena in which to test the developmental waters for bilingual advantages. Such advantages have been discovered for aspects of word awareness (Ben-Zeev 1977; Bialystok 1988; Cummins 1978; Ricciardelli 1992) and grammatical sensitivity (Galambos and Goldin-Meadow 1990; Galambos and Hakuta 1988). If this advantage could be found for phonological awareness, then there would be logical grounds for expecting bilingual children to be in a privileged position in the early stages of reading that rely on these insights. Some evidence for a bilingual superiority in phonological awareness has

Literacy: Languages through other means

been reported, but the results tend to be weak at best. Rubin and Turner (1989) and Yelland, Pollard and Mercuri (1993) both found limited advantages for bilinguals on different tests of phonological awareness, but the advantages were short-lived. More promising results were reported by Campbell and Sais (1995). They found that pre-school Italian-English bilinguals were more advanced than monolinguals on phonological awareness tasks that primarily involved sound deletion but the same as monolinguals on a letter identification task. It is possible that what is needed is a more detailed analysis of the relation between the two languages and a structural basis for interpreting the relation between the two phonological systems and the relation between specific tests of phonological awareness and reading. Tentative results from a study in progress designed in this way indicate again that both bilingualism in general and specific language pairs in particular influence different aspects of the emerging ability (Bialystok, Majumder and Martin, in preparation). We investigated the phonological awareness and early reading ability of children in grades 1 and 2. There were three groups of children included: monolingual English, bilingual English and Chinese, and bilingual English and Spanish. The two groups of bilingual children speak two languages that have different phonological relations to each other, and different correspondences to each other in their writing system. For the Chinese-English bilinguals, the two languages are based on completely different sound systems and are written using different representational principles. For these children, becoming aware of the sound structure of each language requires paying attention to different phonemic units, and once these are isolated, only one of the languages, namely English, uses these units as a basis for the writing system. For the Spanish-English bilinguals, the two languages share many phonemic contrasts and both languages are written by linear recording of these units. Therefore, the influence of knowing two languages should be different in these two cases when children are learning to read in English. Children were given three tests of phonological awareness and a word reading task that assessed decoding. Each phonological task was based on different subskills. The first was the sound-meaning task. Children were given a target word and asked which of two other words matched it for either sound or meaning. The incorrect alternative was the correct match for the other dimension. Thus, children would be asked: “What word sounds something like sick, ill or kick?” The task is solved by being able to separate the sound of the word from its meaning and to focus on these dimensions separately. Words in all languages have these dual properties, and speakers of all languages routinely

25

26

Ellen Bialystok

make those distinctions. Nonetheless, the problem is difficult and children improved with age, but there were no effects of language or bilingualism on the ability to attend to the sound of the word. The second task was phoneme counting. Children had to determine how many sounds were contained in individual words. This task requires both the ability to focus on the sound structure, as in the previous task, and the analytic ability to segment those sounds into phonemes. Therefore, the task makes more sophisticated demands on children’s awareness of the phonological structures of words. Here the three groups were all different from each other. The Chinese bilinguals had more difficulty with this task than the other two groups, but the Spanish bilinguals were better than everyone else. Finally, the most difficult task was phoneme substitution. Children needed to figure out what new word would be created by substituting the first sound in a word with a different sound. For example, children were asked to perform the following: “Take away the first sound from cat and put in the /m/ sound from mop. What word do you have?” The correct answer, mat, is achieved through a series of steps that include focussing on the sound, segmenting the sound, then performing a computation on the sound through substitution. Hence, this is the most difficult of the phonological awareness tasks because it requires the most processing steps. On this task, there were grade differences in performance, with older children becoming increasingly proficient in this problem, but no language group differences. The three tasks are ordered for difficulty and their reliance on insights that are essentially linguistic or cognitive. The first task is relatively simple, the second adds more difficult linguistic demands, and the third increases the cognitive processing requirements. Therefore, the most direct test of linguistic skill is in the second task. The patterns of influence of bilingualism on children’s solution reflected these attributes. The first task, requiring the most general conception of phonology, was not affected by how many languages the child knew or what those languages were. The third task was very complex and, although clearly linguistic, again indicated no role for the child’s language background. We may assume here, however, that the reason for an absence of effect is quite different. The complexity of the substitution task may mean that cognitive factors overshadowed linguistic factors in determining children’s ability to solve the problem. On this view, it is only the second task, the phoneme counting, that provides a pure measure of children’s level of phonological awareness, and in this task, the groups were different from each other. Specifically, having analytic access to the sound structure of two similar languages, both of which are written alphabetically, enhances children’s attention to the

Literacy: Languages through other means

phonemic units (Spanish bilinguals). The need to analyse the sound structure of two completely different language that are written according to different principles, however, impedes children’s progress in developing an explicit conceptualisation of the phonological structure of words (Chinese bilinguals). In spite of these differences in phonological awareness, none of the groups differed from each other in a decoding task that required them to read isolated words. Clearly, reading draws on many skills, and the balance between them does not necessarily imply that an advantage in one can override all other considerations and endow an automatic advantage in reading. Metalinguistic skills, and particularly phonological awareness, are central foundations to reading, but the specific languages that form the bilingual’s repertoire in conjunction with cognitive factors in development are still indispensable to understanding children’s acquisition of literacy.

Reading and understanding For children in school, reading is the primary channel by which information is disseminated. The reason we teach children to read at the very outset of schooling is because reading is the vehicle and essence of schooling. It is important, of course, to know how children acquire this skill, and it is essential that we determine the effects that children’s special language histories have on this achievement, but the ultimate goal is to assure that children can read texts and understand them. In other words, the goal is to be able to read for the purpose of interpreting texts. Do children’s particular language experiences interfere with their attainment of fluent reading? Fluent reading entails a balance between what are sometimes called bottomup and top-down processes. The bottom-up processes consist mainly of such decoding skills as pattern recognition, letter identification, and lexical access. If a new language is written in a familiar script, then these factors should cause little problem in learning to read. The top-down processes are the conceptual skills that include background knowledge and language proficiency. These are rarely in place in the new language to the same extent as they are in the first, even for advanced adult readers of a second language, and this is the main barrier to comprehension when reading is carried out in a weaker language (Durgunoglu 1997; Koda 1994). The vast majority of the research and theorising about the process of reading in a second language has been based on the experience of literate adults. Unless the new language is written in a different

27

28

Ellen Bialystok

writing system from the first, the primary constraint for adults learning to read in a second language is to master the top-down processes. Evidence for the role of proficiency in the second language (Barnett 1989), the difficulty of the text (Alderson 1984), and the learner’s knowledge of cultural schemata and discourse structures of the second language (Barnitz 1986; Carrell 1994) has been reported. The situation is different for children. Pre-school children who are bilingual cannot read in any language, and school-aged children who are immersed in a new language may or may not know how to read in their first language. For all these children, both top-down and the apparently simpler bottom-up processes may present formidable challenges to children’s attainment of high levels of reading skill. Evidence for the influence of first-language processes on fluent reading in a second language can be found by examining a particular cognitive strategy used during reading, namely, phonological recoding. When we convert the written notations into meaningful language, there is a sense (perhaps more metaphorical than literal) in which we reconstruct the sounds of the words from the written forms. This is essentially what is intended in the advice we give to children to “sound out words” that they do not immediately recognise. Koda (1988, 1990) developed an interesting way of investigating the role of phonological recoding for second-language readers. She studied performance in reading tasks by Japanese, Arabic, and Spanish learners of English. Unlike the other languages in the study, Japanese is written as a morphography — the written symbols represent units of meaning and not units of sound. Decoding sounds, therefore, is not routinely part of the process of reading Japanese, and her results confirmed that Japanese readers approached English text differently from the other learner groups. She presented English texts that contained words that could not be read by means of the usual decoding processes because phonological information was not available. In one study, a word was interposed in another script, namely, Sanskrit (Koda 1990), and in another, unpronounceable words were inserted (Koda 1988). Participants had to read the passage and then recall characteristics of these unpronounceable words. The Japanese learners were not impaired in their ability to solve the problems but the Arabic and Spanish learners, as well as native English controls, found the task difficult. She concluded that the Japanese speakers transferred their reading strategies from their first language where phonology is largely irrelevant to their second language in which the writing system makes phonological recoding more appropriate. Since the Japanese learners were not used to relying on

Literacy: Languages through other means

phonological recoding, they were not impeded when the strategy became impossible to use. Even for two alphabetically written languages, the likelihood of using phonological recoding as a reading strategy varies with the nature of the orthographic system. Monolingual readers of various alphabetic languages differentially rely on phonological recoding as a function of how useful or efficient it is for that language. English and French are two such languages. Native speakers of English use phonological recoding prodigiously as a process in fluent reading, but native speakers of French use the strategy less. Segalowitz and Hebert (1994) studied French-English bilinguals who were completely fluent in both languages but differed in whether or not they read the two languages at the same rate. Hence, they created two groups, an equal rate and an unequal rate reading group. Their results showed that equal rate readers used the language-appropriate strategy when reading in each language, but the unequal rate readers transferred their strategy from the stronger language. Hence, native English speakers who read French more slowly than native controls also used phonological recoding more than native speakers when reading in French, although native English speakers who read French at nativelike rates essentially did not use this strategy. This means that unequal rate readers made more errors. These are subtle differences that would never be detected without specially designed measures that can uncover these differences, but the implications are not at all subtle. Reading in the second language cannot be predicted simply on the basis of oral fluency, and the bilingual who appears to be equally proficient in both languages may nonetheless be inefficient and possible nonproficient in reading texts. Fluent reading that is the basis for comprehension is more language-specific than one might have believed, and if the reading is carried out in a second language, then the relation between the two languages is again a critical point of analysis. For children achieving high levels of reading fluency in a second or weak language, the implication is that we cannot ignore the approaches to language structure and the strategies for reading, if the child was already literate, that were engendered by the first language. We know that these levels of reading are compromised if the purely linguistic skills in the second language are weak, but it also seems to be the case that gaining autonomy over some aspects of the reading processes may be more difficult by virtue of children’s experience with another particular language.

29

30

Ellen Bialystok

The legacy of bilingualism For studies of the progress that bilingual children make in various domains of linguistic and cognitive development, the research questions are usually framed around the issue of group difference. For entirely defensible reasons, a large part of this literature is concerned with the issue of whether the experience of learning two languages in childhood has any effect on the child’s development. Indeed, both advantages and disadvantages of being bilingual have been uncovered, and the results of these studies contribute importantly to our understanding of language acquisition and the formative role of language in children’s early development. The theme that has been extracted from this cursory review of highlights of children’s path into literacy is different. Three stages were identified, corresponding to three major milestones in learning to read. In some of them, notably the first, there were indeed simple group advantages for being bilingual. The primary result, however, has been that the greatest determiner of progress is not the fact of bilingualism but its specifics, most notably, what are the two languages and how are they related to each other orally, graphemically, and orthographically. It is these relations that had the most notable impact on progress with literacy. In each of the three major stages of reading acquisition, it was moderate difference between the languages that was most beneficial in propelling readers into the next stage. For concepts of print, exposure to two writing systems that looked different but were both alphabetic was the most beneficial background; for phonological awareness, exposure to two spoken languages that were different from each other but drew on a common set of sounds led to highest levels of awareness; and for fluent reading, two alphabetic writing systems encouraged the transfer of known strategies, although here the results of such transfer were not always felicitous. The effect of language is vast and we ignore the details at our peril. It is not sufficient, it turns out, to simply know that the child speaks two languages, or even how well the child speaks both languages, but we need to know what the two languages are and how they inform and support each other. We know that phonological traces of our first language are scattered through our subsequent efforts to learn new languages in the form of telltale accents. It turns out that such “accents” are pervasive as well in all our linguistic endeavours and they shape the way we use and understand language. If we are to truly understand how bilingualism influences development, we need to understand how specific

Literacy: Languages through other means

languages affect the bilingualism profile. We need to be guided by linguistic analyses to inform our educational interpretations. Educational linguistics is a primary foundation of this endeavour.

References Alderson, J. C. 1984. “Reading in a foreign language: A reading problem or a language problem?” In Reading in a foreign language, J. C. Alderson and A. H. Urquhart (eds), 1–24. London: Longman. Barnett, M. A. 1989. More than meets the eye: Foreign language reading: Theory and practice. Englewood Cliffs, NJ: Prentice-Hall Regents. Barnitz, J. 1986. “Toward understanding the effects of cross-cultural schemata and discourse structure on second language reading comprehension”. Journal of Reading Behavior 18: 95–116. Ben-Zeev, S. 1977. “The influence of bilingualism on cognitive development and cognitive strategy”. Child Development 48: 1009–1018. Bernhardt, E. 1991. Reading development in a second language: Theoretical, empirical and classroom perspectives. Norwood, NJ: Ablex. Bialystok, E. 1988. “Levels of bilingualism and levels of linguistic awareness”. Developmental Psychology 24: 560–567. Bialystok, E. 1991. “Letters, sounds, and symbols: Changes in children’s understanding of written language”. Applied Psycholinguistics 12: 75–89. Bialystok, E. 1997. “Effects of bilingualism and biliteracy on children’s emerging concepts of print”. Developmental Psychology 33: 429–440. Bialystok, E., Majumder, S. and Martin, M. (in preparation). Effect of language background on phonological awareness and early decoding. Bialystok, E., Shenfield, T. and Codd, J. 2000. Languages, scripts, and the environment: factors in developing concepts of print. Developmental Psychology 36: 66–76. Campbell, R. and Sais, E. 1995. “Accelerated metalinguistic (phonological) awareness in bilingual children”. British Journal of Developmental Psychology 13: 61–68. Carrell, P. L. 1994. “Awareness of text structure: Effects on recall”. In Bilingual performance in reading and writing, A. H. Cumming (ed.), 23–41. Ann Arbor: John Benjamins. Carrell, P. L., Devine, J. and Eskey, D. E. (eds). 1988. Interactive approaches to second language reading. New York: Cambridge University Press. Cummins, J. 1978. “Bilingualism and the development of metalinguistic awareness”. Journal of Cross-Cultural Psychology 9: 131–149. DeLoache, J. S. 1987. “Rapid change in the symbolic functioning of very young children”. Science 238: 1556–1557. DeLoache, J. S. and Burns, N. M. 1994. “Early understanding of the representational function of pictures”. Cognition 52: 83–110.

31

32

Ellen Bialystok

Durgunoglu, A. Y. 1997. “Bilingual reading: Its components, development, and other issues”. In Tutorials in bilingualism: Psycholinguistic perspectives, A. M. B. de Groot and J. F. Kroll (eds), 255–276. Mahwah, NJ: Erlbaum. Ellis, N. 1990. “Reading, phonological skills and short-term memory: Interactive tributaries of development”. Journal of Research in Reading 13: 107–122. Ferreiro, E. 1978. “What is written in a written sentence? A developmental answer”. Journal of Education 160: 25–39. Ferreiro, E. 1983. “The development of literacy: A complex psychological problem”. In Writing in Focus, F. Coulmas and K. Ehlich (eds), 277–290. Berlin: Mouton. Ferreiro, E. 1984. “The underlying logic of literacy development”. In Awakening to Literacy, H. Goelman, A. Oberg and F. Smith (eds), 154–173. Exeter, NH: Heinemann Educational Books. Galambos, S. J. and Goldin-Meadow, S. 1990. “The effects of learning two languages on levels of metalinguistic awareness”. Cognition 34: 1–56. Galambos, S. J. and Hakuta, K. 1988. “Subject-specific and task-specific characteristics of metalinguistic awareness in bilingual children”. Applied Psycholinguistics 9: 141–162. Koda, K. 1989. “Effects of L1 orthographic representation on L2 phonological coding strategies”. Journal of Psycholinguistic Research 18 (2): 201–222. Koda, K. 1990. “The use of L1 reading strategies in L2 reading”. Studies in Second Language Acquisition 12: 393–410. Koda, K. 1994. “Second language reading research: Problems and possibilities”. Applied Psycholinguistics 15: 1–28. Liberman, I. Y., Shankweiler, D. P., Liberman, A. M., Fowler, C. and Fischer, F. W. 1977. “The structure and acquisition of reading: II. The reading process and the acquisition of the alphabetic principle”. In Toward a psychology of reading, A. S. Reber and D. L. Scarborough (eds), 207–225. Hillsdale, NJ: Erlbaum. Mann, V. A. 1986. “Phonological awareness: The role of reading experience”. Cognition 24: 65–92. Morais, J., Bertelson, P., Cary, L. and Alegria, J. 1986. “Literacy training and speech segmentation”. Cognition 24: 45–64. Perfetti, C. A., Beck, I., Bell, L. C. and Hughes, C. 1987. “Phonemic knowledge and learning to read are reciprocal: A longitudinal study of first-grade children”. Merrill-Palmer Quarterly 33: 283–319. Perfetti, C. A., Beck, I., Bell, L. C. and Hughes, C. 1988. “Phonemic knowledge and learning to read are reciprocal: A longitudinal study of first grade children”. In Children’s reading and the development of phonological awareness, K. E. Stanovich (ed.), 39–75. Detroit: Wayne State University Press. Read, C., Yun-Fei, Z., Hong-Yi, N. and Bao-Qing, D. 1986. “The ability to manipulate speech sounds depends on knowing alphabetic writing”. Cognition 24: 31–44. Ricciardelli, L. A. 1992. “Bilingualism and cognitive development in relation to threshold theory”. Journal of Psycholinguistic Research 21: 301–316. Rubin, H. and Turner, A. 1989. “Linguistic awareness skills in grade one children in a French immersion setting”. Reading and Writing: An Interdisciplinary Journal 1: 73–86.



Literacy: Languages through other means

Segalowitz, N. and Hebert, M. 1994. “Phonological recoding in the first- and secondlanguage reading of skilled bilinguals”. In Bilingual performance in reading and writing, A. H. Cumming (ed.), 103–135. Ann Arbor, MI: John Benjamins. Troseth, G. L. and DeLoache, J. S. 1998. “The medium can obscure the message: Young children’s understanding of video”. Child Development 69: 950–965. Yelland, G. W., Pollard, J. and Mercuri, A. 1993. “The metalinguistic benefits of limited contact with a second language”. Applied Psycholinguistics 14: 423–444.

33



Bilingual processing strategies in the social context of an undergraduate immersion program Andrew D. Cohen and Kirk C. Allison University of Minnesota

While university-level language immersion programs have been in existence for some years, research on the bilingual processing strategies of students participating in such programs is scant. This chapter reports on two aspects of a four-course college-level immersion program: (1) the participants’ use of both their native and the immersion language to process meaning on academic tasks, and (2) the influence of the social ecology of the immersion context on their language use. The measures used for obtaining data included a questionnaire addressing program perceptions and background, pre/post multi-modality tests and self-assessments, and a retrospective self-observation instrument to provide data on bilingual mental processing by 24 Spanish, French, and German immersion students and by 17 non-immersion students directly after classroom tasks (e.g., process writing, listening to a lecture, watching a video, or discussing an article). A third of the immersion students also volunteered to provide verbal report data outside of class for listening, reading, writing, and speaking on a central topic. In sum, immersion students reported less mental translation and more cognitive processing directly through the immersion language than did their non-immersion counterparts present in those same classrooms. Second, the language modality (i.e., listening, reading, speaking, or writing) significantly effected the type and extent of mental processing in the immersion language. Third, immersion students emphasized the unique social context of immersion as supporting and extending target language use, both within and beyond the classroom. They acknowledged that the greater extent of direct target language cognitive processing was attributable to both the linguistic

36

Andrew D. Cohen and Kirk C. Allison

and social dimensions of the immersion context. While focusing on immersion programs at the university level, the study’s quantitative and qualitative findings provided pedagogical implications for all college-level foreign language education, whether involving standard, intensive, or immersion curricula.

Introduction It is likely that for some people, if not many, the use of one language or another for thinking while performing language tasks is not viewed as a matter of strategy selection or of strategizing. Rather, it is seen as a given. However, for bilinguals and multilinguals — especially for those with at least minimal control of a second or third language, there is an element of choice involved in arriving at the language(s) used to perform cognitive operations (Cook 1994; Cohen 1995). Furthermore, the choice of the language of thought may have significant implications for ultimate success in target language learning and use in a given situation. Second language teaching and learning methods are often predicated on an assumption that learners need to think as much as possible in the foreign language — intuitively, the more thinking through the target language the better. There is, however, some evidence from research on second-language reading and writing that selective translation into the native language may play a positive role for some, if not many, language learners in the comprehension, retention, and production of written texts (regarding reading: Kern 1994; Hawras 1996; Cohen and Hawras 1996; Cohen 1998, Ch. 6; regarding writing: Jones and Tetroe 1987; Lay 1988; Friedlander 1990; Kobayashi and Rinnert 1992; Brooks 1998; Cohen and Brooks-Carson, forthcoming). For this reason, language educators are being asked to take a second look at the role of mental translation in reading (i.e., mental reprocessing of second-language (L2) source words, phrases, and sentences in the first language or another familiar language; Kern 1994) and at the role of written translation from a first language (L1) text as a means for generating a foreign language text. It is being seen that for learners with certain learning style preferences, the selective use of certain types of translation in reading and writing may constitute a set of language learning and language use strategies that are desirable and, at given stages of development, even essential. In addition, a hypothesis in the literature is that learners create their own highly personal discourse domains of second language use. These domains are

Bilingual processing strategies

“internally created contexts, within which…interlanguage structures are created differentially” (Selinker and Douglas 1985: 190). It is reasonable to assume that nonnatives will be more prone to use the target language for performing cognitive operations in discourse domains over which they have greater control. Selinker and Douglas (1985) gave the example of an English-second-language discourse domain in civil engineering created by a native Spanish-speaking graduate student, demonstrating how nonnatives may be more conversant in talking about content in a certain discourse domain than in others. Additional research indicates that nonnatives even with limited language proficiency may be more conversant in talking about content within their professional discourse domain than less knowledgeable native speakers (Zuengler 1993). While choice of language for performing cognitive tasks has been investigated in elementary-school level immersion programs (Cohen 1994; Parker, Heitzman, Fjerstad, Babbs and Cohen 1994), there appears to be little if any research data available on language choice for cognitive processing among university-level immersion students. In the elementary-school study, it was found that considerable mental translation was being used, possibly to the detriment of Spanish language acquisition. The current university-level study was designed to describe how second-language immersion students use both their native and immersion languages to process meaning on academic tasks compared with peers taking one or more immersion program courses but who are not engaged in full immersion, and to assess the impact of the social ecology of the immersion context as a factor in target language use. With regard to bilingual processing, it was our prediction going into the study that immersion students would use less mental translation than nonimmersion students. In other words, we expected them to process the course material through the immersion language more during class activities than nonimmersion students taking the same course. Likewise, we expected that immersion students would engage in more mental dialog (that is, internal mental responses to the source material) in the immersion language than non-immersion students. We also expected that both mental translation and mental dialog would vary according to the language activity or modality involved. Finally, we suspected that affective and social variables would influence both language processing and the extent of language use altogether. Thus, our research questions were as follows: (1) To what extent do immersion and non-immersion students take notes in the immersion program language? (2) To what extent do the students engage in internal mental dialog in that language? (3) To what extent do immersion and non-immersion

37

38

Andrew D. Cohen and Kirk C. Allison

students use mental translation during classroom activities and how helpful do they consider it to be? (4) To what degree does language modality (listening, reading, speaking, or writing) effect the extent of mental translation or mental dialog? (5) What role does affect play in language processing in the immersion context? (6) What role does the social ecology of the immersion program play in language processing and use?

Research design Context of the study In a full immersion program, all courses are taken exclusively in the target language. The immersion program selected for this study was the University of Minnesota’s Foreign Language Immersion Program (FLIP), offered in Spanish, French, and German. The program featured three target-language-medium courses taught in the Spring Quarter at the third-year level in a variety of disciplines (cultural anthropology, history, epidemiology, media), and a “language support course” addressing linguistic challenges. In the three content courses, the focus of instruction was not on language, but rather on a discipline in the midst of which language learning was also intended to occur. Students received content credit toward their degree programs in political science, international relations, and so forth. This content-focused format was intended to compensate for certain obstacles encountered when students take only conventional language courses, such as gaps in language continuity, and was expected to promote a reduction of L1 transfer because it was simulating to some degree a foreign language environment at home. Each class constituted in some ways a unique language learning community, with each combination of students and instructor representing a unique constellation. The internal dynamics of this learning community were different from those of a typical language class which is impacted by students whose initial (and perhaps long-term) goal is “getting an A” or surviving a language requirement rather than, say, learning French. Authentic language events were contextual and relational, fostered or hindered by the ecology of the language learning community and setting in which they took place. In this immersion context, then, the ecology of a class extended beyond a particular course, and extended beyond the composite curricular context to include student interaction

Bilingual processing strategies

during breaks, transitional periods before and after immersion classes, leisure time activities, and chores associated with living arrangements.

Sample The subjects were drawn from undergraduates participating in the University of Minnesota’s Spring 1996 Foreign Language Immersion Program (FLIP), consisting of the three sub-programs in Spanish, French, and German. These undergraduates at the University of Minnesota tended to be locals from the upper Midwest, who had studied foreign language in high school and about half of whom had spent time abroad. Aside from a general media course, the content courses varied from language to language: French had courses in cinema and North African colonial history/literature; for German, topics included the West German 60s’ student movement and postwar history; Spanish offered an anthropology-oriented course on colonialism and the historical epidemiology of Latin America. A FLIP student in the study was defined as one who registered and completed the full four-course compliment. If FLIP students dropped one course or more, they became non-FLIP students. Content courses were also individually open for non-FLIP students where space permitted (e.g., for students needing a course topic to complete a major). Instructors were native speakers or had a high level of target language proficiency in addition to expertise in their subject area. Altogether, twenty-four FLIP students (14 in Spanish, 6 in French, and 4 in German FLIP respectively) participated in this study, as well as 17 non-FLIP students taking FLIP courses. It should be noted that non-FLIP students were often more “advanced” in a program of language study than FLIP students. However, the non-FLIP students were only taking one or two courses through the target language and did not have the unique immersion environment supporting their language experience.

Instrumentation The principal type of data elicited in the processing component of the study was retrospective self-observation, using a verbal report questionnaire. In other words, it called for the inspection of specific, not generalized language behavior, shortly after the mental event had taken place (see Cohen 1996, 1998, regarding verbal report). An example of retrospective self-observation would be, “What I did during that lecture in French was to listen for key words and phrases, and

39

40

Andrew D. Cohen and Kirk C. Allison

to translate the difficult ones into English to see if they made sense to me”. The instrument included items relating to the choice of language for note taking, the extent of internal mental dialog in the target language, the extent of mental translation, and the students’ view regarding the helpfulness of mental translation. With regard to note taking, students were asked the extent to which they took notes during the task, and if they did, whether they did so in the target language. They were also asked to characterize the nature of any internal mental dialog in the immersion language during the task at hand (e.g., while listening to a lecture). With regard to mental translation, the respondents were asked about their “use of internal translation in language processing”. This item consisted of three elements: (1) a yes/no question concerning use of mental translation, (2) a check-off box concerning the extent (“all the time”, “often”, “at difficult spots”, and “a little”) and direction of mental translation (“into English”, “from English”, or “back and forth”), and (3) an open-ended inquiry as to whether mental translation helped, and if so, in what ways. Another source of data involving one-third of the FLIP cohort was an outof-class verbal report task and interview session, aimed at collecting data during the processing of academic material in listening, reading, writing, and speaking tasks on a course-related topic. The listening portion involved the student stopping the tape to report strategies used for dealing with challenging sections of a pre-recorded oral recitation. During the speaking task, students were interrupted twice — usually after hesitation points — and asked to provide verbal report concerning their language production strategies, generally after resolving a difficulty. Directly after completing the task, their performance was replayed to them, and they provided retrospective data on the processing and production strategies that they had just employed. Pre/post language performance tests for all four modalities and student selfassessments provided language proficiency data. Questionnaire items and a debriefing interview were used to gather data regarding motivation to enroll in the program, target language use outside of class, perceived program strengths and weaknesses, and the recommendability of the program.

Data collection and analysis procedures During the last three weeks of a ten-week quarter, a retrospective questionnaire was administered in all four French and German FLIP classes and in the Spanish support class. The questionnaire was completed immediately after a regularly-scheduled class task (e.g., viewing a videotaped newscast or documentary

Bilingual processing strategies

film, listening to a lecture, discussing an article, or engaging in process writing). Choice of language for cognitive processes was investigated for speaking, listening, and viewing for French; for German courses the activities were viewing a videotape and discussing an article, and for the Spanish support course the task was process writing. To compare FLIP and non-FLIP students’ language processing strategies, all students were encouraged to complete the verbal report questionnaires distributed immediately following the designated classroom task, and to complete them according to actual language use, rather than what it “ought” to have been or usually was. A total of 63 task reports were thus gathered. Data collection took place for almost all of the FLIP students in at least two separate contentsubject classes, which produced 44 task reports, one filled out for each class. The non-FLIP students were observed in at least one content-subject class, resulting in 19 task reports. Statistics were run according to FLIP status, classroom task, language, and associated processing variables, using chi-square tests and t-tests to determine statistical significance. A total of eight out-of-class verbal report task and interview sessions were conducted with three German, three French, and two Spanish volunteer FLIP students, a third of the total cohort. For listening, reading, and writing modules, students were encouraged to perform in the manner most natural to them, making use of note taking, the dictionary, and so forth. Source material for all modules was selected as representative of authentic academic content similar to that encountered in FLIP courses. During performance of these tasks, students were requested to provide commentary concerning their language processing in the language of their choice. Data from the listening task were collected as described above. During performance of these tasks, students were asked to note, among other things, the extent and nature of internal mental dialog in the target language, as well as instances of mental translation. All eight verbal report protocols for the four tasks and the interview sessions were transcribed. Then a content analysis of each subject’s responses during each task was performed.

Findings The extent of note taking in the FLIP language Sixty-seven percent of the FLIP students indicated taking notes in the language of immersion, 10% indicated bilingual notes, and the rest did not specify. Fifty-

41

42

Andrew D. Cohen and Kirk C. Allison

four percent of the non-FLIP respondents indicated note taking in the immersion language, 16% indicated the use of both, while 30% did not specify. While a somewhat higher percent of FLIP students took notes in the foreign language, this difference was not statistically significant.

The extent of internal mental dialog in the FLIP language A depiction of what a mental dialog in the FLIP language actually meant was characterized by one student as follows: “Often after the lecture or conversation is over, I will replay the conversation in my mind with my own running commentary”. Mental dialog was also reported to take place during the ongoing processing of material as well. So, in the case of a Spanish FLIP student, when doing a speaking task, he reported having an ongoing mental dialog with himself mostly in Spanish to monitor his grammar — in one case to check that he had inflected the verb for the subjunctive mood correctly (para que ofendan) and realized he had inflected it wrong (para que *ofenden). Then when he could not think of the Spanish word for “gossip” (chismosa), he reported switching over in his mental dialoguing to English, but then still could not think of what the word was in English either. It was expected that the FLIP students would engage in more mental dialog in the immersion language than non-FLIP students, as they were functioning in a set of courses conducted entirely in that language. As it turned out, almost all the FLIP students indicated use of mental dialog in the immersion language (91%) during the task that was assessed, compared to 79% of non-FLIP students. The difference between the two groups of students was not, however, statistically significant.

Mental translation during classroom activities: Extent and perceived helpfulness In order to depict mental translation, the following are two examples taken from the verbal report data in this study. The first is from a German FLIP student engaged in the writing section of the verbal report battery. While she was in the process of writing, she provided the following verbal report data: … I often write like this — very … slowly to try to find that vocabulary, um, that I already have in German even, so that I don’t have to look things up, because if I have to look things up it goes even slower in German than it is right now … I am kind of going through English phrases in my head right now, but

Bilingual processing strategies

I don’t want to use them; I don’t want to translate. So, I’m going through them and discarding them until I find a phrase that I can use in German, that I do have the vocabulary for, because I don’t want to just translate something, so I keep going through until I find something that works that I don’t have to look up …

Thus, the student differentiated here writing a “good” translation, that is, mental translation, from “bad” translation, namely, dictionary translation. In another example of mental translation used more sparingly in a reading and writing task, a German FLIP student was providing her recommendation as to how to effectively oppose modern slavery: J: [exhales] Ja … also, ja, doch, dass … ich finde das … grausam, dass solche Sachen in der Welt heute noch existieren und, um [English], es waere gut wenn, also das es gut ist, wenn wir solche Organisationen haben, die gegen diese[n] Menschenmissbrauch kaempfen, um die Lebensqualitaet zu verbessern, oder … [J: [exhales] Yes … thus, yes, in fact, that … I find that … horrific, the such things still exist today in the world, and, um [English], it would be good if we have such organizations which struggle against this abuse of human beings in order to improve the quality of life, or …]

When asked to give immediate retrospective feedback as to what she was thinking when she stopped, she replied that she was searching for the German equivalent for “standard of living”, so she was back in English thinking of how to say it in German. With regard to the quantitative findings, there was a significant difference between FLIP and non-FLIP students as to the extent of mental translation. In about 60% of the FLIP task reports (26 vs. 18), the respondents indicated that they had not used mental translation on the task, while in the case of the nineteen non-FLIP task reports, the result was the reverse, with only one-third reporting that they refrained from using mental translation. The difference between the two groups was statistically significant (chi-square, p German > French. Our expectation was confirmed using a one-tailed t-test (p < .05): results for tasks in Spanish showed significantly more mental translation than German in two categories (for task reports of FLIP and non-FLIP together and for FLIP students alone), and significantly more mental dialog in an additional two (for all task reports and for non-FLIP students alone). Spanish then significantly exceeded French in both reported mental translation and mental dialog in six categories (for all task reports, FLIP, and non-FLIP alone). German in turn exceeded French in reported mental dialog in two categories (for all task reports and for FLIP alone). No significant result contradicted the hierarchy. These results supported the thesis that control over data flow appears to be a significant factor in mental processing. The following series of quotes from a German immersion student taken from his verbal report protocols will help to illustrate the relative influence of different modalities on the extent of mental translation: Listening: … there’s vocabulary throughout there that I don’t know, but, um, you always understand the context of what’s being said. (Investigator: Do you find yourself using English as you are listening?) Um … no, not really. You don’t go back and forth a lot; I’ve been just listening to German, that’s about it … You just listen and what you can pick up … it goes that fast, or else you kind of miss everything else if you try to sit back and think, “Hm, what does that mean?” or “Where does that come from?” Speaking: it’s not that you’re going into English, but you’re just thinking, um, you just think about what you said last and how you could tie something new into that, and then your just pulling it together as a fast as you can, an when you’ve kind of run out of things to say, it’s harder to try to fake it with German … [At a vocabulary block]: I was searching for vocabulary … I knew the word

Bilingual processing strategies

in English, I was thinking of the English word, the “standard”, and now I can’t think of it in English either now, the life, just like your living standards. Reading: … I read it in German and then I talk through it in English … which is what I’m doing now … I read an article, and then I talk to my roommate about it, I expl … , I try to explain it to her in English, then it’s easier for me to see which parts I really didn’t fully understand … I usually go through it with G. [another FLIP student], we’ll talk through something in German too … [Dictionary — her guess was correct]: you just want to double check because you aren’t a hundred percent sure … like you already have it linked … but you don’t really know. Writing: I usually spend a lot of time looking … try to use different words than what I always use; you could use like one word ten times in a paper if you didn’t look up just a different way to say it.

Although the presence or absence of mental translation was found to be conditioned by the specific task, individual learning styles and strategies also played a role. One student, a history major, noted her avoidance of translation for complex reading passages: “I’m not translating it while I’m reading it, no. I almost never do that, ‘cause it’s just too complicated”. Again, although immersion students used less mental translation per category, they indicated a higher level of satisfaction with the result when used compared to non-immersion students, and this most markedly for listening, the context in which the information stream is most relentless.

Affect as a component of processing and production In an out-of-class verbal report session, several FLIP students indicated that reading content matter prompted emotions in a way that was uncommon in traditional language classes. One justification for FLIP programs is offering an opportunity to experience a broad range of ideas directly through an immersion language, rather than having the ideas distilled and filtered through English or by curriculum-based content sanitizing. An African history student, for instance, engaged in a text on French North African immigration policy, became upset in the process and commented: … I guess it’s kind of shocking to even read it in French, because I guess when I think, “I’m reading in French”, I always think that it’s going to be like nonoffensive … because it’s in French … I guess I’m surprised that I read French well enough to have an emotional response to something written in French.

49

50

Andrew D. Cohen and Kirk C. Allison

One component of her surprise was positive — her achieved fluency; the second component was that FL content could be offensive at all. French language course textbooks and materials may offer chapters on French colonial legacy and racism, but the offense is characteristically mediated through commentary or counter articles. Rarely is a student confronted by texts which actually advance an offensive attitude. The textual world of the language learner is generally sanitized, as this student expressed it, “Jane met Bob across the street and they went to lunch”. Yet engaging affect may aid in production fluency: “This (writing) will be easy because now I’m talking about the part that bugged me … instead of stopping to stumble on grammatical mistakes, or not even mistakes … hearing how it sounds in my head”. A German immersion student reacted strongly to an article concerning modern slavery and lax enforcement of antislavery laws in Britain: “That I don’t like. They don’t really enforce; they aren’t too harsh on slave owners [living in Britain from other countries with house slaves]”. While the initial mental response language is not indicated, the affective experience is authentic. At times, affect from the target language source eclipsed linguistic particularity: OK, and here, I’m talking about child prostitution, and I’m actually having an emotional reaction to what I’m reading, and I’m not just trying to figure out but I’m actually going [soto voce] “Oh, that’s terrible”, you know. And I’m not sure if that, I think that reaction doesn’t even have words. It’s just a wrinkle in my brow, I guess.

Emotions may be triggered by grammatical structures (“Ooh … compound tense at the end [a future passive]. That’s nice. There hasn’t been a lot of that”) or by not remembering a word in a text: “It’s irritating because it is a word that I have heard a lot. [She looks it up.] Generally I make flash cards for such words and then I just try to … use it in a paper if I write one, … to integrate it more through that, or use it when I speak if I’m comfortable with it, so I just try to make the connection”. Affect is also related to extraneous interruption: A student writing her final target language paper, cited English language telephone interruptions: It was starting to make me like angry: … Not angry at them obviously, but angry because … it took so much to kind of get back into it … like it feels when I’m writing an essay … in English, and I’m all focused … and then it would be like me watching a movie for two hours and then getting back to the paper and expecting to be right where I left off with it. It’s almost the same thing, although in ten minutes that can happen … just speaking in English.

Bilingual processing strategies

Another instance of affect was noted during her verbal report writing task: “I almost put the verb in the wrong place. That’s because I’ve been writing in English all day today; so far I’m in that mode, so it’s a little difficult … ” (writing for an independent study with a single paper post FLIP). Another immersion student noted: “It’s really weird; it’s kind of a pain too, because it’s really hard to get into when I am out of it … ” The challenge of maintaining language continuity for students in a mixed language curriculum is much greater: When I’m taking French and psychology, which is how I am normally doing it, because I’m a psych major … it’s kind of hard to get into the mode … 200 pages of psychology and then 50 pages of French … to get from one to another, you get into a mode, whereas with the four classes in French, each homework assignment was in French so it was always in the French mode.

The effect of the social ecology of an immersion program on language processing and use A FLIP instructor from Spain characterized the American University experience as lacking both relational and academic continuity, and as generally impersonal. She considered FLIP an exception. Learning a second language in the postsecondary context broadly means voluntary exposure to a particular language and associated traditions (if only by exercising an option within a dreaded language requirement). While the instructor is the primary personal presence and model of the target language and also the instigating agent of linguisticcultural integration, an instructor’s scope of input is quite circumscribed through contact hours and competing demands. One factor repeatedly cited in the FLIP debriefing interview was the degree to which friendships which had been made and then fostered in the program contributed to target language exposure and authenticity beyond the official classroom curriculum. Seventy-three percent of FLIP students reported an increase in “language use with friends” outside of class, the most prevalent of any category, followed by listening to target language music (67%), pleasure reading (53%), watching films (47%), watching TV (43%), and letter writing (33%). Each of these categories represented an extension of the target language into discretionary time (and money) use. One student characterized interacting with friends in the target language outside of class “about a third of the time” as not being excessive. Another wrote emphatically: “If we hadn’t had all those readings we would have had a lot more opportunities outside of class to speak French.

51

52

Andrew D. Cohen and Kirk C. Allison

But as it was, we didn’t!” Relationships mediated the FLIP experience and were being mediated through the target language. Seven students were asked the degree to which friendships were a significant part of their learning context. One student who was acquainted with two of the FLIP students prior to the program was ambivalent. Five students cited significant new friendships within the program and expressed their intention to stay in contact after the program, and one asked an established friend to take the program with her. Two also chose specifically to enroll with new friends in a particular language class for the following quarter. A few comments were illustrative of the dynamic way in which friendships extended the target language beyond the curriculum into “life as a whole”: (Investigator’s question: Did you make friends in the program?) A German FLIP student: Yeah, I did. Yeah, G. and I really became good friends. We’re going to take a class together next quarter, so I’m really glad. And with G., it’s really nice because we’re both really comfortable to speak German to each other; a lot of times you’re a little reserved or, you know, oh, maybe I’m a little embarrassed because somebody’s level is different than yours, but … with G. it’s really easy to speak German, um, so that helps. And so you get a lot of time outside of class that you just speak, for fun, and it’s nice to unwind with that, or just to practice … so I’m really, I’m really glad that worked out too. Another German FLIP student: Well, especially since it was such a small group I got to know the people better, so that was nice … (Investigator: Will you keep in touch afterwards?) Well possibly, they’re both members, or most of them are members of the German Club, both J. and G. … and so being involved with that I see them; and then if I ever want somebody to just talk some German to, I know where to go … Often times, um, G. and I would leave together and just speak in German entirely as we walked across the Washington Avenue Bridge [a good quarter mile] … A French FLIP student: (I made) one or two (friends). There weren’t very many of us … period, so that I made two friends, I’m really happy about (that). Considering my classes are usually so big, I feel like I made lots of friends, but I only made two friends.

A Spanish student stated that “at the beginning, (it was) frustrating; the main part of it was to meet the people — some really cool people — and be in the same four classes with them”. The students went out together, speaking Spanish sometimes, and did so while moving across campus to their next common class (particularly when not wanting others to understand what they were talking about!). The findings showed that the forming of relationships, together with the

Bilingual processing strategies

“familiarity” and the “trust” that these provided for performing in the target language, constituted the most frequently cited positive aspect of the program (41%), followed by immersion in theory and practice (35%), effective instructors (29%) (“the way in which they provide course work that generates some sort of uncontrollable interest amongst the students”), and finally content (24%). The single highest negative was insufficient organization and coordination (65%), concentrated specifically around one professor. But even here one student stated that the relational network within the program “(made) up for the professors that we had”. With regard to whether they would recommend the program to a friend, the characteristic response was “yes” for motivated persons who understood the work load: I would … if you really want to do it. I (don’t) think you can just sit there and have it all come easy, because it’s not going to; it’s not going to happen without a lot of work … you just really have to want it. Because if you don’t, you’ll sink.

Another added: “(To) people who are really serious about learning the language, I would suggest it to them. (But) I don’t have any friends like that … except for the two I just made in the program!” Fourteen of seventeen indicated that they would recommend FLIP outright or conditionally. Only three implied deficiencies to the extent of not recommending it, one of whom was a native speaker. One student would not recommend it to someone who would feel uncomfortable speaking or making mistakes in front of others, although “no one would laugh”. A majority would also take another quarter of FLIP if it fit their graduation curriculum. One student also anticipated a certain degree of culture shock on leaving the immersion context and reentering the regular curriculum: Yeah, I’m kind of like bummed about it. Well, I have another French course that starts tomorrow, um, but it will just be that one, and then I will have the whole rest of the summer … because I won’t take anything second summer session; and then it’ll just be Fall again, and that will be it. I don’t know; that will be kind of a bummer … That’s kind of disappointing, because (FLIP) was fun. It’s really hard, though, because there was a lot of work.

Finally, the following is an anecdote that helps to exemplify how the social context of the FLIP program extended beyond traditional notions of instruction. Late in the Spring quarter in a morning FLIP Spanish course, two student carpoolers failed to pick up a classmate. One of the students had indicated that FLIP “is like a ‘family’ and one gets to know people. If you need help, you can ask”.

53

54

Andrew D. Cohen and Kirk C. Allison

In this instance, the two drivers expressed surprise in Spanish when they encountered each other there without the third student. This event was picked up on by other students with considerable target language kidding. The winded “strandee” arrived a half hour later. Expressions of embarrassment, apologies, discussion of the miscommunication, and the instructor’s forbearance all played out in Spanish in an authentic exchange, arising not from a particular assignment, but from relationships that had developed during the program.

Discussion This study of college-level immersion and non-immersion students would seem to provide evidence that immersion encouraged cognitive processing directly through the target language due to both its unique linguistic and social dimensions. The results of this study therefore appear to be consistent with those from previous research which indicated that limited mental translation may yield benefits when not used indiscriminately. On the methodological level, it was found that simple dichotomous responses regarding the use of mental translation were less valid than the graduated scale.

Summary of findings according to research question 1. The extent of note taking in the immersion language among immersion and nonimmersion students: Note taking in the immersion language was somewhat more prevalent among FLIP students, yet the difference was not statistically significant. 2. The extent of internal mental dialog in the immersion language: As expected, more FLIP students engaged in mental dialog (91% to 79%), although two high Spanish non-FLIP scores offset the possibility of statistical significance. 3. The extent of mental translation by immersion and non-immersion students during classroom activities and its perceived helpfulness: FLIP students were found to use significantly less mental translation compared to non-FLIP students, but reported higher resulting satisfaction when they did use it. For FLIP students, mental translation functioned as a targeted strategy rather than as a general “stay afloat” strategy. Consistent with expectation, German FLIP students engaged in significantly more mental dialog vs. mental

Bilingual processing strategies

translation compared to non-FLIP students, while results for Spanish and French were inconclusive. 4. The impact of language modality/task on the extent of mental translation and mental dialog: For immersion students, modality/task proved to be a significant predictor as to the extent of mental translation, corresponding to a great extent with the control the students had over the information stream: listening (the least control), viewing, speaking, and writing (the most control). Increases in mental dialog also followed this pattern. This hierarchy was also found predictive across languages in cases where language-specific tasks favored one modality or another. The modality of the task was found to impact non-immersion students in similar ways. 5. The role of affect in language processing in the immersion context: Verbal report interviews indicated that authentically engaged affect appeared to be more prevalent in the immersion context than in traditional language curricula, and it was reported to have a positive impact on production and fluency. Likewise in contrast to the traditional language classroom, the immersion context was characterized as allowing for a greater range of exposure to authentic perspectives. 6. The role of the social ecology of the immersion program in language processing and use: It was found that continuity across the curriculum also extended to immersion language use outside the classroom. The development of “target language friendships” was the single most frequently cited positive element in the program. Correlatively, the most frequently cited stimulant for increased outside-class target language interaction was “target language use with friends”. Such relationships were also cited as a significant compensatory factor for deficits within the program against the backdrop of what was seen as the general impersonality of the university. Freeman and Long’s dictum that “there is rarely a one-to-one correspondence between what is taught to what is learned” (1991) has a salutary outcome in this case: the social ecology of immersion programs extended beyond the classroom, to a degree not usually experienced in conventional curriculum.

55

56

Andrew D. Cohen and Kirk C. Allison

Limitations The data collection for this study was limited to the number of FLIP and nonFLIP students available during the one Spring Quarter, possibly limiting its representativeness. Given the fairly small sample size, it was therefore difficult to arrive at statistically significant differences in behavior. In addition, the novelty of this kind of psycholinguistic research meant that the categories of mental translation and mental dialog were perhaps not as fully developed for research purposes as might have been desired.

Interpretation of results Our data suggest that the immersion context has a significant influence on bilingual processing strategies and practices. While the measured use of mental translation strategies can have benefit in the accomplishment of language tasks, it was seen that a full-immersion program can diminish the need for such strategizing. As predicted, there was also a greater use of mental dialog relative to mental translation among immersion students; this we attribute to greater fluency in productive modalities supported by the immersion context. We also saw that the effects of the immersion context were not limited to curricularly scheduled tasks. FLIP students reported increases in target language use outside the curricular context.

Implications for further research Research on bilingual mental processing strategies across modalities in other university-level immersion programs would allow a more ample data base for determining the impact of immersion on language and content education. The findings derived from such work can have important implications for curricular and institutional programs. Where no non-immersion students are present in immersion classes, similar or correlative activities in non-immersion classes (perhaps using identical video, for example) could be used to provide data. The initial indication of modality influences on bilingual mental processing strategies invites additional research with a view toward discerning effective strategies within particular modalities for strategy “coaching” and for providing a more differentiated response to the question of the advisability of encouraging or discouraging multilingual processing.

Bilingual processing strategies

The verbal report techniques used for describing language processing in this study could also be applied to non-Indo-European languages and to so-called LCTL’s (Less Commonly Taught Languages) for which a relative dearth of instructional materials and possibly fewer formal L1 analogies (assuming English) exist. Does the distance between, say, Chinese and English reduce the commerce in interlingual processing? Such contrasts may require modification of the verbal report interview questions to register the impact of non-Latin orthography or of root- and aspect-based grammars on strategies beyond those reflected in this survey. A goal of a foreign language curriculum is to facilitate access to the target language and culture(s), and to extend locally the language environment implied in it. Additional future research pertinent to our present discussion includes investigating the effectiveness of residential language environments (see below) for narrowing the performance gradient between regular, intensive, and immersion program learning contexts. The continuing research question is how best to foster integration and continuity for the relatively disadvantaged conventional curricular context and for those students who have not spent time abroad.

Pedagogical implications It has long been recognized that a primary mission of a language program is to create an effective language learning environment through curricular and ancillary activities (e.g., curricular offerings; access to language labs, tables and clubs; optional lectures; films, internet and FL TV; interaction with target language speakers in non-course contexts, etc.). The immersion program studied confirms that “foreign languages across the curriculum” programs, such as the one offered at the University of Minnesota since 1987, can be successfully integrated to provide a complete quarter or semester-based multi-discipline immersion context in which content is engaged in the target language in a structure applicable to a variety of majors. With regard to this point, 90% of the FLIP students acknowledged enrolling with a goal of improving language skills, while only 23% cited interest in a specific academic topic as primary. Although such courses have largely been in the humanities, science topics such as geography have also been offered. Foreign language immersion experiences embrace the view that fostering an optimal language environment implies extending the academic target language environment beyond the competency of language department disciplines per se into other distinct fields (history, anthropology, geography, international

57

58

Andrew D. Cohen and Kirk C. Allison

relations, media) which have their own distinctive vocabularies, theoretical frameworks, methodologies, and practices. To accomplish such expansion, issues of “turf” in terms of field and enrollments — institutional dynamics that often militate against realizing optimal educational contexts for the student — will have to be addressed. Establishing such a program requires a coordinating agency within the university (at the University of Minnesota, it was the Institute for Global Studies). One byproduct, however, is a greater degree of integration in the student’s overall educational experience, a potential particularly relevant for those double majors, who were living a “disparate” existence: Especially I liked A. K.’s class because it was really cool studying African Studies in French … it was really interesting; I really liked that. And then it made me feel a little more like, you know, that my two majors kind of go together in a practical way …

Much also can be done to increase the density of social context in support of a more holistic target language experience. Although a “university language village” such as that created by a language camp program such as the Concordia Language Villages one in Minnesota is not practical for the university, a real option to the benefit of immersion, intensive, and regular curriculum students is to establish target language living contexts within the college or university, preferably in conjunction with an institution’s housing authority (whether via rental houses or dorm wings). Such arrangements also may be pursued independently by students. For example, several FLIP Spanish students elected to live together in their own casa español. Yet such private initiatives presuppose a high level of prior commitment and focus (and are naturally more likely to transpire in an immersion context than among students in a mixed curriculum). An initiative at the University of Minnesota has been taken to establish language dedicated dormitory wings for German and Spanish, open to majors and minors, extending the precedent of residential honors colleges and pre-med wings. Many of the target language leisure activities given increased attention by FLIP students are locally supportable in such an environment (FL magazines, papers, TV, films on video). Such dormitory arrangements also extend to traditional curriculum students some benefits of immersion by providing a living context that converts the target language from an “exotic” to an intrinsic element of daily life. Such an arrangement likewise offers greater continuity for students returning from L2 study-abroad experiences, should they so choose, and addresses a situation, which arose between two FLIP students, one of whom was encouraged by the other to take the program:

Bilingual processing strategies

I think I’m a little more serious about speaking French than she is … I always wanted to speak French outside of class, and she really didn’t, which is, you know, fine … but it kind of bummed me out, because I was kind of hoping to have this “all French experience” …

Conclusions Cognitive processing in the target language appears to be significantly enhanced by a foreign language immersion program, both through the explicit linguistic curricular context and through extending target language integration beyond the classroom, mediated by relationships formed within it. Mental translation in particular was both reduced in degree and increased in effectiveness among immersion students. In addition, the language modality of the task was seen to have a significant effect on the type and extent of bilingual processing. A university-based immersion program provides a somewhat sheltered “middle territory” deemed helpful for maintaining and enhancing fluency also for students returning from study abroad, but with one advantage: The nice thing about FLIP, though, is that you can be punished all day … and go home and you don’t have the cultural environment, which is fortunate and unfortunate, because it’s a little relief, it’s not so intensive then because you do get to escape.

References Brooks, A. W. 1996. Coherence in FL writing: What translation can tell us. Unpublished manuscript. Miami, FL: Dept. of Foreign Languages and Literatures, University of Miami, Coral Gables. Cohen, A. D. 1994. “The language used to perform cognitive operations during full-immersion math tasks”. Language Testing 16 (2): 171–195. Cohen, A. D. 1995. “The role of language of thought in foreign language learning”. Working Papers in Educational Linguistics 11 (2): 1–23. Cohen, A. D. 1996. “Verbal reports as a source of insights into second language learner strategies”. Applied Language Learning 7 (1–2): 5–24. Cohen, A. D. 1998. Strategies in learning and using a second language. Harlow, Essex: Longman. Cohen, A. D. and Brooks-Carson, A. Forthcoming. “Research on direct vs. translated writing processes: Implications for assessment”. Modern Language Journal 85 (summer, 2001).

59



60

Andrew D. Cohen and Kirk C. Allison

Cohen, A. D. and Hawras, S. 1996. “Mental translation into the first language during foreignlanguage reading”. The Language Teacher 20 (2): 6–12. Cook, V. J. 1994. Internal and external uses of a second language. Unpublished manuscript. Essex: Department of Language and Linguistics, University of Essex. Friedlander, A. 1990. “Composing in English: Effects of a first language on writing in English as a second language”. In Second language writing: Research insights for the classroom, B. Kroll (ed.), 109–125. Cambridge: Cambridge University Press. Hawras, S. 1996. Towards describing bilingual and multilingual behavior: Implications for ESL instruction. [Double Plan B Paper]. Minnesota, MN: English as a Second Language Department, University of Minnesota. Jones, S. and Tetroe, J. 1987. “Composing in a second language”. In Writing in real time, A. Matsuhashi (ed.), 34–57. Norwood, NJ: Ablex. Kern, R. G. 1994. “The role of mental translation in second language reading”. Studies in Second Language Acquisition 16 (4): 441–461. Kobayashi, H. and Rinnert, C. 1992. “Effects of first language on second language writing: Translation versus direct composition”. Language Learning 42 (2): 183–215. Larsen-Freeman, D. and Long, M. H. 1991. An introduction to second language acquisition research. London: Longman. Lay, N. D. S. 1988. “The comforts of the first language in learning to write”. Kaleidoscope 4 (1): 15–18. Parker, J., Heitzman, J., Fjerstad, A. J., Babbs, L. M. and Cohen, A. D. 1994. “Exploring the role of foreign language in immersion education: Implications for SLA theory and L2 pedagogy”. In Second language acquisition theory and pedagogy, F. R. Eckman, D. Highland, P. W. Lee, J. Milcham and R. R. Weber (eds), 235–253. Mahwah, NJ: Lawrence Erlbaum. Selinker, L. and Douglas, D. 1985. “Wrestling with ‘context’ in interlanguage theory”. Applied Linguistics 6 (2): 190–204. Spolsky, B. 1985. “The limits of authenticity in language testing”. Language Testing 2 (1): 31–40. Zuengler, J. 1993. “Encouraging learners’ conversational participation: The effect of content knowledge”. Language Learning 43 (3): 403–432.



Part II

Language testing in educational contexts



Some construct validity issues in interpreting scores from performance assessments of language ability* Lyle F. Bachman University of California, Los Angeles

Introduction Performance assessment and the complexities that accompany it, are issues with which language testing theory has been grappling for at least since Spolsky (1968, 1973) asked the fundamental and difficult questions about validation: “What is the nature of competence and performance?” and “What does it mean to know a language?”1 These issues have been debated among language testers, in the past quarter of a century, in the context of what have variously been called performance tests, direct tests, authentic tests and communicative tests, (e.g., Alderson 1981a, 1981b, 1983; Bachman 1990; Carroll 1961; Clark 1972, 1975; Morrow 1979; Oller 1979; Skehan 1984; Spolsky 1986; Stevenson 1985; Upshur 1979).2 Furthermore, at least two language testing textbooks focus on the concerns of performance testing: Carroll and Hall (1985), provide practical guidance in the construction and use of performance tests, while more recently McNamara (1996) deals with many of the conceptual and measurement issues that face language testing, situating these in the context of performance assessment. In practice, of course, performance assessments have been used in language testing for well over a century (Spolsky 1995), and written compositions and oral interviews have been and still are considered by many to be the premier techniques for testing language ability. The term “performance test” or “performance assessment” has been used in a number of different ways over the years, and I discuss its history in language testing very briefly below. For the purposes of this paper, let me adopt the definition of performance assessment

64

Lyle F. Bachman

offered by McNamara: its “defining characteristic is that actual performances of relevant tasks are required of candidates (McNamara 1996: 6). Performance assessment (sometimes called alternative assessment) is also currently of considerable interest and controversy in educational measurement in many countries, and this approach to assessment has led many researchers in this field to re-examine traditional views of test qualities, especially validity (e.g., Baker, O’Neil and Linn 1993; Delandshere and Petrosky 1998; Dunbar, Koretz and Hoover 1991; Linn 1994; Linn, Baker and Dunbar 1991; Messick 1994; Moss 1992, 1995; Shavelson, Baxter and Pine 1992). Since the field of language testing has had considerable experience in addressing many of these issues, the approaches to these concerns that have evolved for language testing might well be of interest to developers and users of psychological and educational tests, as well as to researchers in psychological and educational measurement. Performance assessments of language ability have tended to focus on achieving authenticity of assessment tasks, often with little attention to the nature of constructs these tasks are intended to measure. The constructs that are tapped in performance assessment, however, are likely to be more complex than those engaged by more restricted task types, such as the multiple-choice. This is because the tasks typically presented in performance assessments of language ability are intended to engage test takers in language use, which, by its very nature, is likely to involve the full range of areas of language ability, as well as other areas of ability, such as topical knowledge. With greater opportunity for the full range of language ability, as well as topical knowledge, to be involved, the process of demonstrating the validity of intended inferences about specific areas of language ability becomes more difficult. In performance assessments, the fundamental dilemma of language testing — that language is both the object and instrument of measurement (Bachman 1990) — is particularly evident, as the aspects of language ability we want to measure become almost inextricably meshed with other abilities and with the methods used to elicit language. In addition, because performance assessment tasks themselves are of greater richness, in terms of both language use and topical content, the domains to which our inferences about language ability generalize become potentially more restricted. Thus, in the performance assessment of language ability, the potential interactions between language ability and other abilities, on the one hand, and between language ability and the characteristics of the assessment tasks, on the other, complicate two validity issues: (1) investigating and demonstrating the validity of the construct interpretations we derive from test

Some construct validity issues

performance, and (2) identifying the domains to which these inferences generalize, and providing evidence of such generalizations. These validity issues are recurring themes in Spolsky’s work, and are concerns that have also consumed much of my own research for the past twenty years or so. In this paper I will discuss my current thinking and approach to addressing these issues. I will first suggest several questions relevant to construct validity and validity generalization that I believe need to be addressed for any given performance assessment task.3 I will then argue that in order to adequately address these questions, three elements are needed in every assessment procedure we design, develop and use. Next, I will briefly describe a current model of language use and language ability and a framework of task characteristics. I will then describe how these frameworks can be used to define two qualities — authenticity and interactiveness — that are frequently claimed to be distinguishing characteristics of performance assessment tasks, and relate these qualities to construct validity and validity generalization. Finally, I will illustrate how these elements might apply to performance assessments, with example tasks from various kinds of tests. The main message that I would like to get across is that in language testing we have claimed, or sought, to achieve the qualities of authenticity and interactiveness in performance assessment tasks without an adequate analytic framework for doing so, and hence no adequate mechanism for evaluating their usefulness. What is critical, I would argue, is that we must do more than simply consider the roles of language ability, assessment task characteristics, and test taker characteristics when we design and use performance assessments of language ability. What we need to do is utilize logical analysis and empirical investigation, grounded in a sound theoretical framework, to define the constructs we want to measure, to design and specify our assessment tasks, and to justify the kinds of inferences we make on the basis of our assessments.

Construct validity questions Construct validity can be defined as the extent to which the results of an assessment can be interpreted as an indicator of the ability we intend to measure, with respect to a specific domain of generalization (Bachman and Palmer 1996: 21). For any given performance assessment task, I would argue that the following questions related to construct validity need to be addressed:

65

66

Lyle F. Bachman

(1) What construct(s) is(are) assessed? (What specific inferences about test takers’ language ability can we make on the basis of their performance on the language assessment task?) (2) What is(are) the domain(s) of generalization of assessment-based inferences? • To what domain(s) of language learning tasks or real-world language use tasks do we expect our inferences about test takers’ language ability to generalize, based on their performance on the language assessment task? • How representative of language learning tasks and real-world language use tasks is the language assessment task? That is, to what extent do the characteristics of the language assessment task correspond to the characteristics of language learning tasks and real-world language use tasks? • To what extent do test takers’ responses to the test task correspond to individuals’ responses to similar language learning tasks and real-life language use tasks? Because of the claims that have been made about the specific qualities of performance assessment, and because of the interrelatedness of these qualities and construct validity, we also need to ask the following questions: (1) How authentic are test takers’ responses to the test task? (To what extent and in what specific ways do test takers perceive the assessment task as corresponding to language learning tasks, or to real-life language use tasks?) (2) How interactive are test takers’ responses to the test task? (To what extent does the test task engage the areas of language ability to be assessed?)

Critical design elements for addressing construct validity questions In order to adequately address these construct validity questions, I believe that three elements are needed in the design, development and use of performance assessments: (1) an interactionalist definition of language ability that characterizes the interactions among the components of language ability and other cognitive processes, and, in turn, the interactions among these and features of the language use setting, (2) a clearly identified domain of target language use tasks, and (3) a framework of characteristics for describing both the assessment task and the target language use tasks.

Some construct validity issues

First element: Abilities to be assessed Language use and language ability A number of models of language use and language ability have been proposed in the language testing literature over the years, (e.g., Bachman 1990; Bachman and Palmer 1996; Canale 1983; Canale and Swain 1981; Carroll 1961; Chapelle 1998; Cooper 1968; Lado 1961). A recent approach by Bachman and Palmer (1996) illustrates what Chapelle (1998) calls an “interactionalist” approach to defining language ability. According to Chapelle, “from an interactionalist perspective, performance is viewed as a sign of underlying traits which are influenced by the context in which it occurs, and is therefore a sample of performance in similar contexts” (Chapelle 1998:43, italics in original). An interactionalist approach to defining the language ability construct must therefore include the specification of not only the trait and contextual components, but also the metacognitive strategies that mediate the interactions among the trait components and between these and the contextual components. Bachman and Palmer (1996) describe language use in terms of two types of interactions: (1) among several attributes of individuals — language knowledge, topical knowledge, affective schemata, and strategic competence, and (2) among these attributes and the characteristics of the language use situation. They illustrate these interactions in Figure 1. Bachman and Palmer (1996) define language ability as consisting of two components: language knowledge and strategic competence, or metacognitive strategies. Language knowledge, which they define as “a domain of information that is available for use by the metacognitive strategies in creating and interpreting discourse”(1996: 66) includes two broad categories: organizational knowledge (with subcategories of grammatical and textual knowledge), which enables language users to create and interpret grammatically well-formed sentences or utterances and to organize these into cohesive texts, both oral in written, and pragmatic knowledge (with subcategories of functional and sociolinguistic knowledge), which enables language users to produce or interpret utterances or sentences with meaning and intentionality, and that are appropriate to the particular language use situation. For Bachman and Palmer, strategic competence consists of three metacognitive strategies — goal-setting, assessment and planning — which they characterize as “higher order executive processes that provide a management function in language use” (1996: 70). These metacognitive strategies manage, or regulate, the interactions among the various components of language use — language knowledge, topical knowledge,

67

68

Lyle F. Bachman

Language knowledge Topical knowledge

Personal characteristics Affect

Strategic competence

Characteristics of the language use or test task and setting

Figure 1.

Some components of language use and language test performance (Bachman and Palmer 1996: 63)

personal characteristics and affect. In addition, they provide the cognitive link between these components and the other aspects of the speech event, including, or course, other participants in that event. Topical knowledge Because language ability is the construct of primary interest in language assessment, I have focussed on this part of Bachman and Palmer’s description. This is not to suggest, however, that the other elements in their framework of language use, particularly topical knowledge, or knowledge schemata, do not affect language use, or that these can be ignored in the design, development and use of language assessments. Since language use is always about “something” and thus includes and refers to topical content, it will engage the topical knowledge of language users to some degree. A number of language testing researchers have grappled with technical definitions of this term (e.g., Alderson 2000; Carrell 1983; Clapham 1996; Johnson 1981), but for practical test development purposes, Bachman and Palmer’s definition of topical knowledge as “knowledge structures in long-term memory” (1996: 65) will suffice. The accepted wisdom, in language testing, has been to assume that we are not interested in measuring test takers’ topical knowledge, and hence to view the

Some construct validity issues

effects of topical content as potential sources of either measurement error or bias, that need to be minimized by selecting input that is appropriate for the particular group of test takers. However, given the considerable research, particularly in reading assessment, indicating that topical knowledge interacts with language ability, (e.g., Alderson and Urquhart 1985; Clapham 1996) to affect test performance, we can no longer treat this simply as a factor to be somehow eliminated from or minimized in language tests. That is, as Bachman and Palmer argue, we cannot ignore the interaction between the topical content of the test task and test takers’ topical knowledge when we design a test task, but must decide how to deal with this in the way we define the construct to be measured. Bachman and Palmer describe three alternatives for dealing with topical knowledge in the way we define constructs as we design a language test, and discuss the implications of these alternatives for the inferences of ability we can make. Their alternatives are: (1) defining the construct as language ability only, and attempting to eliminate or at least minimize the effects of topical knowledge in the way we design test tasks, (2) defining the construct as a combination of language ability and topical knowledge, and attempting to design assessment tasks that will engage this construct, and (3) defining language ability and topical knowledge as separate constructs and attempting to measure these in separate assessment tasks. McNamara (1996) addresses the issue from a slightly different perspective, that of the criteria used for evaluating test takers’ performance on assessment tasks, and distinguishes between what he calls strong and weak senses of the term “second language performance assessment”. In the strong sense of second language performance assessment, “tasks will represent real-world tasks, and performance will be primarily judged on real-world criteria, that is, the fulfillment of the task” (1996: 43). While this sense encompasses more than test takers’ topical knowledge, it corresponds roughly to Bachman and Palmer’s second alternative, namely defining the ability to be assessed as including knowledge and skills in addition to language ability. In McNamara’s weak sense of performance assessment, inferences focus on language ability: “tasks may resemble or simulate real-world tasks … The candidate is required to perform on a task which may represent tasks he or she may subsequently face in the real world; however, the capacity to perform the task is not actually the focus of assessment. Rather, the purpose of assessment is to elicit a language sample so that second language proficiency … may be assessed” (McNamara 1996: 44). This sense corresponds closely to Bachman and Palmer’s first alternative, namely, defining the construct only in terms of language ability.

69

70

Lyle F. Bachman

Second element: Identifying and defining the target use situation and tasks A characteristic that has been claimed for performance assessments in the language testing literature is that they mirror “real-life”, or “authentic language use contexts”, and in some of the earliest writings on the subject (e.g., Clark 1975; Jones 1979), there was little indication of how such contexts might be characterized, much less considered in the design of language tests.4 One can see much of this same vagueness in current discussions of authenticity in the educational measurement literature (e.g., Newman, Brandt and Wiggins 1998; Terwilliger 1997, 1998). Language testers subsequently recognized that not just any non-test language use situation was relevant to the authenticity of assessment tasks. Wesche (1987), for example, stated that in performance-based tests “examinees must demonstrate their second language proficiency through tasks whose content and contextual features represent the situations in which the second language will eventually be used” (1987: 28), and uses the term “target use situation” to characterize these situations.5 Similarly, Jones (1985) argued that what distinguishes performance tests from other types of tests is “the degree to which test procedures approximate the reality of the situation in which the actual task would be performed” (1985: 16). Bachman and Palmer also recognize that the real-life domain which is relevant to the authenticity of a particular performance assessment task is the specific one in which test takers are likely, or expect, to use language. They thus make the notion of a target language use domain a critical part of their approach to the test development process, tying this to the domain to which inferences of ability generalize. They begin by defining a language use task as “an activity that involves individuals in using language for the purpose of achieving a particular goal or objective in a particular situation” (1996: 45). Under this definition, it is possible to consider language test tasks as specific instances of language use tasks, and to characterize these within the same descriptive framework (described below). They further define a target language use (TLU) domain as “a set of specific language use tasks that the test taker is likely to encounter outside of the test itself, and to which we want our inferences about language ability to generalize” (1996: 44). They argue that for any given testing situation, one or more TLU domains need to be specified, and this specification provides the criterion domain for the design and development of language tests and for the interpretation of test performance.

Some construct validity issues

Third element: A descriptive framework of task characteristics As indicated above, one of the distinguishing characteristics that has been claimed for performance assessments is their correspondence to TLU tasks. One of the problems that plagued early discussions of performance assessment in language testing was that of characterizing TLU tasks with sufficient precision to adequately investigate or support claims for authenticity. The TLU tasks that were described were frequently fairly complex speech events (e.g., the interactions between a flight attendant and airplane passengers), while the domains referred to were often very broad (e.g., coursework in an academic discipline) and the tasks were generally described more or less as holistic entities. There were several problems with this approach to defining TLU tasks. First, with a finite, open set of tasks to choose from in virtually any TLU domain, there was a problem of sampling, or of how to establish content coverage and representativeness — the extent to which any given TLU task adequately represented the variety of tasks in the TLU domain. Another problem was that discussions typically left out formal language learning situations as potential TLU domains. A third problem was that tasks in a given TLU domain vary along many dimensions, so that there was a need to characterize these different dimensions. Largely in response to these problems, as well as to the growing body of research indicating sizeable effects associated with different test methods, Bachman (1990) and Bachman and Palmer (1996) developed a framework of task characteristics that can be used for describing types of target use tasks and test tasks. In brief, their framework consists of five sets of characteristics, as follows: a. the setting consists of the physical circumstances under which either language use or language testing takes place. It can be characterized in terms of its physical characteristics, the participants involved, and the time of the task. b. the rubric is essentially the context for the task, and includes those characteristics of the test or language use situation that (1) provide the structure for the task, (2) constrain how language users or test takers are expected to respond to these tasks, and (3) describe the criteria by which the quality of the language user’s performance will be evaluated.6 c. the input consists of the material contained in the task, which test takers or language users need to process in some way, and to which they are expected to respond. Input may be either aural or visual or both; it may include verbal or non-verbal visual material, or both. If the input includes verbal

71

72

Lyle F. Bachman

material, this can be characterized in terms of its linguistic characteristics, as well as its topical content. d. the expected response consists of the language use that is expected, given the way in which the rubric, or context, for the task is configured, and the particular input that is provided e. the relationship between input and response can be described in terms of reactivity, which is the degree of reciprocity, or interaction involved, scope, which is the amount and range of input that needs to be processed in order to respond, and directness, which is the extent to which the response can be made by using information in the input by itself, or whether the language user or test taker must also rely on information in the context or in his or her own real world knowledge. Describing tasks as types, consisting of sets of characteristics, rather than as single, holistic elements, accomplishes a number of things. First, it adds precision to the analysis TLU tasks and assessment tasks and thus facilitates the evaluation of their comparability. Similarly, it facilitates the evaluation of the comparability of different assessment tasks. Finally, since the description of the TLU domain and tasks serves as the description of the domain of generalization, the use of a task characteristics framework enables us to describe the domain to which our inferences of ability generalize with much greater precision. Two characteristics that are often associated performance assessment tasks are authenticity and interactiveness.7 I believe that we can utilize a framework of task characteristics and a model of language ability to define authenticity and interactiveness and to relate these qualities to construct validity and validity generalization, as illustrated in Figure 2 below. The correspondence that is of central concern here is that between characteristics of test takers, or language users — in particular, their language ability — and the score interpretation. This correspondence pertains to construct validity, which can be defined as the meaningfulness and appropriateness of the interpretations of test scores as indicators of the ability defined in the construct definition, with respect to a specific target language use domain. The correspondence between characteristics of test tasks and tasks in the target domain pertain to authenticity and the domain of generalization of score interpretations.8 Bachman and Palmer define authenticity in terms of the degree of correspondence between the characteristics of language assessment task and those of a TLU task, and point out that authenticity is a matter of perception.

Some construct validity issues

Score interpretation

Language ability

Figure 2.

Test score

Interactiveness

Domain of generalization

Authenticity

Construct Validity

Inferences about language ability (construct definition)

Characteristics of the test task

Construct validity of score interpretations, authenticity and interactiveness (Bachman and Palmer 1996: 22)

What may thus be critical, in terms of test takers’ performance on an assessment task, is their perception of authenticity — of the relevance of the characteristics of the assessment task to the characteristics of tasks in a particular TLU domain. That is, one could hypothesize that test takers’ perceptions of authenticity will affect the way in which they approach a particular assessment task, the strategies they use in completing it, and ultimately their performance on that task. The work of Douglas (Douglas 1998; Douglas and Selinker 1985) on discourse domains in language testing has shed some light on this hypothesis, and this remains, I believe, a fascinating and potentially very productive avenue for further research. Bachman and Palmer argue that the characteristics of the assessment task provide a de facto definition of tasks in TLU domain, and hence of the domain of generalization. That is, the distinctive characteristics of the assessment task constitute an hypothesis about the types of TLU tasks to which inferences of ability generalize, and hence provide a basis for investigating the validity generalization of inferences to TLU domains.

73

74

Lyle F. Bachman

Finally, the correspondence between language ability and the characteristics of assessment tasks pertains to the quality of interactiveness, which Bachman and Palmer (1996) define as the extent to which the test task engages the test taker’s areas of language knowledge, metacognitive strategies, and topical knowledge (1996: 25). Relating this to Messick’s (1989) formulation of construct validity, we can say that to the extent that the assessment task engages those abilities that we want to measure, it yields construct relevant variance, or variations in test scores that are related to the construct we intend to measure, and hence enhances the construct validity of score interpretations. To the extent that it engages abilities that we do not intend to measure, it introduces construct irrelevant variance, or score variations related to factors other than the construct we want to measure, and hence diminishes the construct validity of score interpretations. The point of this discussion has been to hopefully add some clarity to the discussion of qualities of performance assessments, by suggesting that it is useful to consider authenticity, interactiveness and construct validity to be functions of different relationships among the abilities of the test takers, the characteristics of the assessment tasks and two aspects of score interpretations, rather than attempting to lump them all together under the rubric of validity. I would argue that this analytic separation is useful not only for guiding research into factors that affect the way individuals perform on performance assessments, but also in the design and development of such measures.9

Illustrative examples Let me now turn to some examples to illustrate how the use of these three elements — interactionalist construct definition, target language use domain, and descriptive task characteristics — can help address the construct validity questions posed above. The first two examples are of performance assessment tasks that require test takers to write compositions. For these examples, I will focus on the question of the construct, and consider only the scoring rubrics for these tasks. The first scoring rubric is from the test of English for Professional Employment, which is a test of English for professional purposes developed in Australia (Clarkson and Jensen 1995), presented in Figure 3 below. What construct is assessed in this scoring rubric? As this is intended to be a measure of English for professional purposes, we might expect the wording in the scoring rubric to reflect this, and so we do. In the rating scale for the

Some construct validity issues

Please note that in the following rating scales, Level A is excellent, Level B is equivalent to satisfactory performance of the task and Level C is unsatisfactory. Rating scale for the standard letter of application B. Client presents letter of application which reflects the convention of layout of standard business letters after some modifications. Content has been organized into paragraphs in such a way as not to detract from the purpose of the letter. After some adjustments the letter has correct spelling, is grammatically accurate and usually uses appropriate expressions. Correct register is sustained throughout most of the letter. Content may include some slightly irrelevant information and details included may be sketchy. Style of presentation is predictable: overall impression is of a competent letter, but with no items of notable interest. Rating scale for the standard job interview A. Client presents required documents and shows detailed knowledge of the position advertised, particularly the criteria for selection of required staff. Able to understand all questions, requesting clarification where necessary. Able to answer questions clearly and appropriately, giving accurate and detailed information related wherever possible to the selection criteria. Terminology and selection of detail is consistent with the professional field. Uses appropriately polite register. Communication is smooth and accent does not cause serious difficulty for the interviewer.

Figure 3.

The EPE rating scales for six course tasks (selected from Clarkson and Jensen 1995: 189, 191)

standard letter of application, level B, for example, we find the phrases, “content has been organized into paragraphs”, “the letter has correct spelling, is grammatically accurate and usually used appropriate expressions”, and “correct register is sustained throughout”. In the rating scale for the standard job interview, we find, “able to understand all questions, requesting clarification when necessary”, “uses appropriately polite register”, and “communication is smooth and accent does not cause serious difficulty”. At the same time we find phrases that suggest that the candidate’s knowledge of relevant topical information is also being assessed. In the rating scale for the standard letter of application, for example, we find, “reflects the conventions of layout of standard business letters” and ‘content may include some slightly irrelevant information and details included may be sketchy”, while in the rating scale for the standard job interview we find, “detailed knowledge of the position advertised, particularly the criteria for selection of required staff”, and “terminology and selection of detail is consistent with the professional field”. What construct underlies these scoring rubrics? Language ability? (grammatical competence, textual

75

76

Lyle F. Bachman

competence, illocutionary competence, sociolinguistic competence)? Knowledge of the conventions of the genre elicited by the specific tasks? Knowledge of the relevant professional information generally required in such tasks as a letter of application and a job interview? Or is it something like the ability to use language to interpret and express professional information? Or perhaps the ability to understand and communicate professional information through language? The second scoring rubric is for essay-type mathematics problems from the California Assessment Program (cited in Herman, Aschbacher and Winters 1992), and is presented in Figure 4 below. What construct is assessed in this scoring rubric? Since it is ostensibly a mathematics problem, we might assume that it measures knowledge or skills in mathematics, such as might be emphasized in a mathematics syllabus. Indeed, if we look at the specific wording of the level descriptors, we find phrases that would support this claim. In the wording of level 6, for example, we find the phrases, “includes a clear and simplified diagram”, “shows understanding of the open-ended problem’s mathematical ideas and processes”, and “identifies important elements of the problem”, while at level 3, we find “may make major computational errors”. On the other hand, we also find phrases that refer clearly to the quality of the language. In level 6, for example, we find, “clear, coherent, unambiguous and elegant explanation” and “communicates effectively with the identified audience”, while in level 2 we find, “words to not reflect the problem”. What, then, is the construct that underlies this scoring rubric? Knowledge, understanding and skill in mathematics? Language ability? Or is it something like the ability to understand and communicate mathematical information through language? What these two scoring rubrics illustrate, I believe, is that when we score test takers’ performance by rubrics such as these we are no longer assessing a narrow construct of either language ability or of academic content achievement. What we are, in fact assessing is a construct that involves both. Hence the validity of narrow inferences about either language ability or mastery of academic content is questionable. Let us now consider two example assessment tasks, keeping in mind the question of what construct is being assessed. The first example is as follows: Test takers hear the following conversation, from an audio tape: (1) (male speaker) “Hey, Tina. What’s up? You still down to go tonight?” (2) (female speaker) “Yeah. It ought to be a good game. I heard their outside shooter’s been on fire for the last week, and they’re tough on the boards, too”.

Some construct validity issues

CAP Generalized Rubric (California State Department of Education 1989) Demonstrated Competence Exemplary Response . . . Rating = 6 Gives a complete response with a clear, coherent, unambiguous, and elegant explanation; includes a clear and simplified diagram; communicates effectively to the identified audience; shows understanding of the open-ended problem’s mathematical ideas and processes; identifies all the important elements of the problem; may include examples and counterexamples; presents strong supporting arguments. Competent Response . . . Rating = 5 Gives a fairly complete response with reasonably clear explanations; may include an appropriate diagram; communicates effectively to the identified audience; shows understanding of the problem’s mathematical ideas and processes; identifies the most important elements of the problems; presents solid supporting arguments. Satisfactory Response Minor Flaws But Satisfactory . . . Rating = 4 Completes the problem satisfactorily, but the explanation may be muddled; argumentation may be incomplete; diagram may be inappropriate or unclear; understands the underlying mathematical ideas; uses mathematical ideas effectively. Serious Flaws But Nearly Satisfactory . . . Rating = 3 Begins the problem appropriately but may fail to complete or may omit significant parts of the problem; may fail to show full understanding of mathematical ideas and processes; may make major computational errors; may misuse or fail to use mathematical terms; response may reflect an inappropriate strategy for solving the problem. Inadequate Response Begins, But Fails to Complete Problem . . . Rating = 2 Explanation is not understandable; diagram may be unclear; shows no understanding of the problem situation; may make major computational errors. Unable to Begin Effectively . . . Rating = 1 Words do not reflect the problem; drawings misrepresent the problem situation; copies parts of the problem but without attempting a solution; fails to indicate which information is appropriate to problem. No Attempt . . . Rating = 0

Figure 4.

CAP scoring rubric (Herman, Aschbacher and Winters 1992: 60)

77

78

Lyle F. Bachman

(3) (male speaker) “Yeah, I heard they’re dope. There’s supposed to be a cool band, too”. (4) (female speaker) “Yeah. I heard them last week and their alto sax’s got mad skills. The game starts at 7:30, so I’ll pick you up at 7”. (5) (male speaker) “That’s cool, but there might be some traffic, so come a little earlier”. (6) (female speaker) “I’m down with that. I’ll be there around 6:30, then. That’ll give us about 45 minutes to get there before the band starts”. (7) (third voice) “What kind of an event are the speakers going to, and what time will they arrive there?”

(Test takers are asked to write their response in a blank space in the test book.) Here is the second example: Test takers read the following: Mr. K. Nine has two dogs named Barnum and Bailey. He has 35 dog biscuits to split between the two dogs so that Barnum will get exactly 11 more than Bailey. How many biscuits will each dog get? First, solve the problem, showing your calculations. Then, write a story problem of your own. Make sure that your story is clearly written and that you include all the data needed to solve the problem.

What constructs are assessed? Looking at these tasks first from the perspective of language testing, the construct validity question would be, “What specific inferences about test takers’ language abilities can we make on the basis of their performance on these assessment tasks?” The first task would appear to be a fairly standard listening comprehension question, of the kind found on many standardized English proficiency examinations (with the exception that it requires a limited production, rather than a selected response.) The question appears to require test takers to make some inferences based on their comprehension of the information presented in the short dialogue. The second task engages the test takers’ language ability, both to comprehend the story problem, and to create and describe a task similar to the one they are asked to solve, and could thus be viewed as an assessment of their ability to comprehend specific details in a reading passage and to describe their own thought processes in writing. Although the first task is intended to assess listening comprehension, the answer is dependent on the test taker’s knowledge of American culture (i.e., that “outside shooters” and “boards” refer to basketball games, which take place

Some construct validity issues

in gyms and are frequently accompanied by the music of a live band), and their mathematical reasoning (identifying 6:30 and 45 minutes as the relevant information, and then adding 45 minutes to 6:30 to get 7:15). Similarly, the second task engages test takers’ knowledge of mathematics. From the perspective of language assessment, the fact that both of these tasks engage abilities other than language ability would be seen as a potential source of construct irrelevant variance, and perhaps content bias. From the perspective of subject matter assessment, the construct validity question would be, “What specific inferences about test taker’s knowledge schemata and skills in the subject can we make on the basis of their performance on these test tasks?” Looking at these same tasks from this perspective, we reach a different interpretation from that above. Although it is a bit of a stretch, we might imagine a task such as the first one occurring in an orientation program for prospective students or immigrants to the U.S., where participants are learning about popular culture and ways of expressing time in colloquial American English. In this case, the construct measured would be largely related to topical knowledge. The second task, which is from an actual 6th grade mathematics classroom achievement test, is aimed at assessing students’ ability to solve problems through gradual approximation, as a precursor to algebra. Both of these tasks, however, rely heavily on test taker’s language ability, which from the educational measurement perspective would be viewed as a source of construct irrelevant variance, and perhaps, linguistic bias. What both of these example assessment tasks illustrate, I believe, is that in performance assessment tasks such as these, both language ability and topical knowledge are involved in test takers’ performance, thus raising questions about the validity of inferences about either language ability or content knowledge. These four examples — two scoring rubrics from performance assessments and two example performance assessment tasks — illustrate several alternative interpretations to scores from performance assessment, as described by Bachman and Palmer (1996: 120–127). The interpretation from the traditional perspective of language assessment, as indicators of language ability alone, can be illustrated as in Figure 5. A second interpretation, as indicators of academic content, or topical knowledge only, can be illustrated as in Figure 6. A third interpretation, as indicators of a portmanteau construct, “the capacity for understanding and communicating content information through language” or “the capacity for using language to understand and communicate content information”, depending, perhaps, on your particular focus, can be

79

80

Lyle F. Bachman

Score interpretation Inferences about topical knowledge

Domain of generalization

Test score

Characteristics of the test tasks

Language ability

Topical knowledge

Figure 5.

Alternative interpretation 1: Language ability only (Bachman and Palmer 1996: 121)

Score interpretation Inferences about topical knowledge

Domain of generalization

Test score

Characteristics of the test tasks

Language ability Topical knowledge

Figure 6.

Alternative interpretation 2: Topical knowledge only

Some construct validity issues

Score interpretation Inferences about the capacity for using language to understand and communicate topical information

Domain of generalization

Test score

Language ability

Figure 7.

Topical knowledge

Characteristics of the test tasks

Alternative interpretation 2: Capacity for using language to understand and communicate topical information

illustrated as in Figure 7. This interpretation corresponds to what Douglas (2000) calls specific purpose language ability, which he defines as the interaction between specific purpose background knowledge and language ability. I would argue that even though this may not be the intended interpretation in many performance assessments, it is the most defensible for most performance assessments, whether these are intended to focus on language ability or academic achievement. Bringing what is intended and what is defensible in line with each other should be, I would argue, a major concern and area for research for both language testing and educational measurement. There are however, and will continue to be, I am sure, situations in which we do need to make distinct inferences about either language ability or topical knowledge, or both. Bachman and Palmer (1996) discuss another option for the way we define and operationalize ability constructs, which might be appropriate in such situations, and this is to assess them separately in separate tasks. They also point out a practical problem with this approach, in that it entails giving at least two separate assessments — one for language ability and one for topical knowledge. Jacoby and McNamara (1999) point out another problem with this approach, in that it artificially separates the language and knowledge components

81

82

Lyle F. Bachman

of what is essentially an integrated speech event, in that distinct sets of tasks, often with different interlocutors, and distinct assessment criteria, often applied by entirely different raters, are likely to be employed. A third problem is that this approach essentially vitiates most of the advantages that are claimed for performance assessments, in that the types of tasks that are most likely to result from such narrow definitions of the constructs are precisely those that performance assessments are intended to replace — tasks which may not correspond to tasks test takers are likely to encounter outside the test itself.

What is the domain of generalization? Judging from the content of the first task, the most likely domain of generalization is that of informal conversations between high school or college age students. Given the function performed in the task — arranging for a time to meet — we might consider a potentially broader language use domain, that of giving and requesting information, and might be tempted to consider generalizing our inferences to a broader population, perhaps adult non-native speakers of English. In this regard, we would need to describe the target language use domains that are relevant for this group, to determine the extent to which this task corresponds to tasks in that domain. This description would need to include not only the topical content of tasks in the TLU domain, but also the functions that these tasks perform and the register that is most likely to be used. For this, I would argue, a framework of task characteristics such as that proposed by Bachman and Palmer is essential. What about the second task? One can assume that the writer of this task wanted to provide content that generalizes beyond the mathematics classroom, and that does not limit the domain to which inferences about math ability generalize. But does the fact that the topical content is about a dog owner and his two dogs limit our inferences to this particular domain? I would argue that questions such as this about validity generalization can only be answered by empirical investigation, and not on the basis of assumptions or claims. These examples illustrate, I believe, the need for test developers and users to carefully examine the assumptions they make about the domains to which they can expect their inferences about ability to generalize. This examination can best be done, I would argue, by beginning with a careful analysis of the characteristics of tasks in the TLU domain, which should provide evidence of the extent to which assessment-based inferences can be expected to generalize. This logical examination needs to be followed by empirical construct validation

Some construct validity issues

studies, using methods such as those that have been described in the literature (e.g., Bachman 1990; Chapelle 1998; Messick 1989).

How authentic is the test takers’ response to the assessment task? In addition to investigating the validity of assessment-based inferences with respect to the domain of generalization, we also need to consider the authenticity of the assessment tasks for the intended test takers. This will entail a description of the characteristics of the intended test takers, in particular, the presuppositions we can make about what topical knowledge — cultural, or academic — they may possess. As indicated above, authenticity is a matter of perception, so without actual evidence on how test takers respond to these tasks we can only speculate about their perceived authenticity. We can surmise that the test developer who wrote the basketball game task probably felt that this was quite authentic, and if the intended test takers were American high school or college students, we might expect the input of the task to be perceived by the test takers as relatively authentic as well. But would it be equally authentic for adult nonnative speakers of English? How much could we expect them to know about the sport of basketball and the features of a typical high school or college basketball game? How much of the slang would they understand? If the answer to this question is “very little”, then how authentic will this task be for them? How valid will our inferences about language ability for this group be? How fair will any decisions we make on the basis of the assessment results be? For the second task, which was intended for grade six pupils, we can also assume that the test writer has tried to make this as authentic a task as possible. Some insight into how this task was perceived by at least one test taker, a sixth grade pupil, can be seen from the following response to the second part of this task, to make up another similar problem: Kay Commo has 2 dogs named Manuel and Taco Grande. She has 25 biscuits to split between the two. Greedy Taco Grande took 11 more for himself so Manuel has less. Now, how many biscuits will each dog get?

This student’s story problem includes names that she may feel are more authentic, that is, more relevant to her target language use domain, than “Barnum” and “Bailey”. Furthermore, she very perceptively provides some motivation for the fact that one dog gets 11 more biscuits than the other. The fact the she was able to successfully create a task that she presumably perceives as authentic suggests that the original task itself may have been perceived as

83

84

Lyle F. Bachman

reasonably authentic. It also provides some support for the generalizability of the original task to another domain, where people and animals have Hispanic names and dogs, rather than their capricious owners, determine how the food is distributed. What these two tasks illustrate, I believe, is the need to carefully consider the characteristics of the intended test takers, particularly their topical knowledge, in the design, development and use of language tests.

How interactive is the test takers’ response to the assessment task? Looking at the first task, for students who provide the ‘correct’ information, that the two speakers are going to a basketball game and will arrive at approximately 7:15, this task will have engaged both their language ability and their topical knowledge — cultural and academic. In Douglas’s (1998, 2000) terms, this task will have evoked the relevant discourse domain for these students. With respect to the second task, the example student response given above indicates that for this pupil, at least, the task engaged her language ability. Whether or not it also engaged knowledge of mathematics is difficult to say without more information, such as might be obtained through introspection, since the relevant data for solving the problem are exactly the same as in the assessment task itself. What these two tasks illustrate, I believe, again, is that we cannot simply assume that performance assessment tasks will engage the abilities we want to measure. Rather, we need to examine these assumptions and investigate their tenability through empirical research-either qualitative or quantitative, or, ideally, both.

Summary In performance assessments of language ability, topical content and test takers’ topical knowledge play much more important roles than in selected-response task types. Similarly, language ability is likely to affect performance on alternative assessment tasks in educational measurement much more than it does in multiple-choice tests. Performance assessment tasks, in both language testing and educational measurement, typically require test takers to use language to process topical content, both in interpreting the input in the test task and in responding to that task, thus engaging not only their language ability but also



Some construct validity issues

their knowledge schemata. What this implies, I believe, is that the inferences that are made on the basis of scores from performance assessment tasks must be based on construct definitions that include both language ability and content knowledge. Furthermore, this also implies that performance-based approaches to language testing and educational measurement cannot consider either language ability or topical content and subject matter knowledge to be merely potential sources of measurement error or bias. The increased involvement of language ability and content knowledge in performance assessments also has implications for the way we specify the domain(s) to which we intend our inferences of ability to generalize. Assuming that test use inevitably involves generalizing beyond the specific test task, then it is imperative that we define the target use situation and tasks with sufficient specificity to permit us to determine the extent to which our test tasks are representative of these target use tasks. That is, if we want to be able to provide evidence for validity generalization, we must clearly specify the characteristics of the tasks that are included in our intended domain of generalization. Taking all of this into consideration, it is my belief that in order to provide a sound basis for demonstrating the construct validity of interpretations and the generalizability of these interpretations to target use domains beyond the test task itself, our approach to both language assessment and educational performance assessment must be both construct-based and task-based.

Notes * I would like to thank Adrian Palmer and John Schumann for their valuable comments and suggestions. I am indebted to Tina and Melissa Bachman for the basketball and dog biscuit tasks. 1. McNamara (1996) provides an excellent historical overview and theoretical discussion of the issues involved in second language performance assessment. 2. Although I find none of the terms that have been used (performance, direct, authentic, communicative) to be particularly useful in general, as all focus on singular aspects of this approach to assessment, I will use the term “performance” assessment, primarily because this seems to have had a longer history in the field of language testing, is generally more familiar to researchers and practitioners alike, and seems to have the greatest currency at present. 3. I use the term “construct validity”, rather than “validity”, because I associate the former with score-based inferences, and distinguish this from other qualities of test usefulness such as impact, or consequences of test use (See Bachman and Palmer 1996). The consequences of test use are considered by many to be part of validity — what Messick (1989; 1994) has

85

86

Lyle F. Bachman

called the “consequential basis” of validity-and considerable discussion has taken place in language testing in this area in the past few years (e.g., Alderson and Wall 1993; Davies 1997; Hamp-Lyons 1997; Lynch 1997; 1993a; Shohamy 1993b; 1997; Shohamy, Donitsa-Schmidt and Ferman 1997; Spolsky 1981; Spolsky 1997). I recognize the importance of considering the consequences, or impact, of the ways in which we develop and use language tests. However, in this paper, my focus is on issues related to the construct validity of score interpretations. I use the term “validity generalization” to refer to the extent to which scorebased inferences generalize to domains outside the test itself. This is of particular importance in performance assessments in which inferences may be intended to generalize to either instructional or “real-life” domains, or both, so that the specification of the domain of generalization is critical for demonstrating that score-based inferences are appropriate, or generalize to these domains. 4. Early discussions of performance assessment in the language testing literature (e.g., Jones 1979) drew on definitions of performance assessment from the field of educational measurement (Fitzpatrick and Morrison 1971; Ryans and Frederiksen 1951), which typically discussed this in contrast to paper-and-pencil tests. 5. The notion of target use situation can be found in Fitzpatrick and Morrison’s (1971: 237) discussion of “criterion situations” as “those in which the learning is to be applied”. 6. Jacoby (1998) describes “indigenous assessment” in a professional setting and Jacoby and McNamara (1999) discuss the implications for specifying scoring criteria for performance assessments of indigenous assessment criteria of communicative competence that are used by interlocutors in professional settings. Early discussions of performance assessment in the educational measurement literature (e.g., Ryans and Frederiksen 1951) refer very generally to job analysis and “success on the job” as a basis for evaluating performance. McNamara (1996) correctly points out that in addition to assessment criteria or, more broadly speaking, rating procedures, another set of characteristics that typically enter into performance assessments, and that thus must be specified in the design and development of performance assessments are the characteristics of the individuals, the raters, who will apply these rating procedures to samples of language use elicited in performance assessments. It would seem logical to include the characteristics of individuals as raters in Bachman and Palmer’s category of rubric, while the characteristics of individuals as interlocutors might be included in the category of setting, as participants. This analytic separation would permit the empirical investigation of the potentially differential effects, on scores from an oral assessment task, of the characteristics of individuals as interlocutors and the characteristics of individuals a raters. 7. Although the term interactiveness is not used in the educational measurement literature, I believe this notion is similar to what has been called “cognitive complexity” (Linn et al. 1991) and the idea of measuring higher-order cognitive processes, since it involves the engagement of metacognitive strategies. 8. I would note that this conceptualization of authenticity is quite different from that of Messick (1994; 1996), who ties authenticity to construct representation — the way the construct is defined, and the extent to which this is represented in the assessment task.

Some construct validity issues

9. Bachman and Palmer (1996) include the qualities of construct validity, authenticity and interactiveness, along with reliability, impact and practicality, as part of a more general quality of usefulness, which they argue provides a guiding framework for both the design and development of language tests and for the interpretation and use of results from these tests.

References Alderson, J. C. 1981a. “Reaction to the Morrow paper”. In Issues in Language Testing. ELT Documents 111, J. C. Alderson and A. Hughes (eds), 45–54. London: The British Council. Alderson, J. C. 1981b. “Report of the discussion on communicative language testing”. In Issues in Language Testing. ELT Documents 111, J. C. Alderson and A. Hughes (eds), 21–39. London: British Council. Alderson, J. C. 1983. “Response to Harrrison: Who needs jam?” In Current Developments in Language Testing, A. Hughes and D. Porter (eds), 87–92. London: Academic Press. Alderson, J. C. 2000. Assessing Reading Ability. Cambridge: Cambridge University Press. Alderson, J. C. and Urquhart, A. H. 1985. “The effect of students’ academic discipline on their performance on ESP reading tests”. Language Testing 2 (2): 192–204. Alderson, J. C. and Wall, D. 1993. “Does washback exist?” Applied Linguistics 14: 115–129. Bachman, L. F. 1990. Fundamental Considerations in Language Testing. Oxford: Oxford University Press. Bachman, L. F. and Palmer, A. S. 1996. Language Testing in Practice: Designing and Developing Useful Language Tests. Oxford: Oxford University Press. Baker, E. L., O’Neil, H. F. and Linn, R. L. 1993. “Policy and validity prospects for performance-based assessment”. American Psychologist 48 (12): 1210–1218. Canale, M. 1983. “On some dimensions of language proficiency”. In Issues in Language Testing Research, J. W. Oller (ed.), 333–342. Rowley, MA: Newbury. Canale, M. and Swain, M. (1981). “Theoretical bases of communicative approaches to second language teaching and testing”. Applied Linguistics 1 (1): 1–47. Carrell, P. L. 1983. “Three components of background knowledge in reading comprehension”. Language Learning 33: 183–207. Carroll, B. J. and Hall, P. J. 1985. Make Your Own Language Tests: A Practical Guide to Writing Language Performance Tests. Oxford: Pergamon Press. Carroll, J. B. 1961. “Fundamental considerations in language testing”. In Testing the English Proficiency of Foreign Students, 30–40. Washington, DC: Center for Applied Linguistics. Chapelle, C. 1998. “Construct definition and validity inquiry in SLA research”. In Interfaces Between Second Language Acquisition and Language Testing Research, L. F. Bachman and A. D. Cohen (eds), 32–70. New York: Cambridge University Press. Clapham, C. 1996. The Development of IELTS: A Study of the Effect of Background Knowledge on Reading Comprehension. Cambridge: University of Cambridge Local Examinations Syndicate/Cambridge University Press. Clark, J. L. D. 1972. Foreign Language Testing: Theory and Practice. Philadelphia: Center for Curriculum Development.

87

88

Lyle F. Bachman

Clark, J. L. D. 1975. “Theoretical and technical considerations in oral proficiency testing”. In Testing language proficiency, R. L. Jones and B. Spolsky (eds), 10–24. Arlington, VA: Center for Applied Linguistics. Clark, J. L. D. 1978. “Interview testing research at Educational Testing Service”. In Direct Testing of Speaking Proficiency: Theory and Application, J. L. D. Clark (ed.), 211–228. Princeton: Educational Testing Service. Clarkson, R. and Jensen, M.-T. 1995. “Assessing achievement in English for professional employment programs”. In Language Assessment in Action, G. Brindley (ed.), 165–194. Sydney: National Center for English Language Teaching and Research, Macquarie University. Cooper, R. L. 1968. “An elaborated language testing model”. In J. A. Upshur and J. Fata (eds), Problems in foreign language testing [Language Learning Special Issue 3], 57–65. Ann Arbor: Research Club in Language Learning. Davies, A. 1997. “Demands of being professional in language testing”. Language Testing 14: 328–339. Delandshere, G. and Petrosky, A. R. 1998. “Assessment of complex performances: Limitations of key measurement assumptions”. Educational Researcher 27 (2): 14–24. Douglas, D. D. 1998. “Testing methods in context-based second language research”. In Interfaces between second language acquisition and language testing research, L. F. Bachman and A. D. Cohen (eds), 105–117. New York: Cambridge University Press. Douglas, D. D. 2000. Assessing Languages for Specific Purposes: Theory and Practice. Cambridge: Cambridge University Press. Douglas, D. D. and Selinker, L. 1985. “Principles for language tests within the ‘discourse domains’ theory of interlanguage: Research, test construction and interpretation”. Language Testing 2: 205–226. Dunbar, S. B., Koretz, D. and Hoover, H. D. 1991. “Quality control in the development and use of performance assessments”. Applied Measurement in Education 4: 289–304. Fitzpatrick, R. and Morrison, E. J. 1971. “Performance and product evaluation”. In Educational measurement, R. L. Thorndike (ed.), 237–270. Washington, DC: American Council on Education. Hamp-Lyons, L. 1997. “Washback, impact and validity: Ethical concerns”. Language Testing 14: 295–303. Herman, J. L., Aschbacher, P. R. and Winters, L. 1992. A Practical Guide to Alternative Assessment. Alexandria, VA: Association for Supervision and Curriculum Development. Jacoby, S. 1998. Science as Performance: Socializing Scientific Discourse through the Conference Talk Rehearsal. Unpublished Ph.D. University of California, Los Angeles. Jacoby, S. and McNamara, T. 1999. “Locating competence”. English for Specific Purposes 18 (3): 213–241. Johnson, P. 1981. “Effects on reading comprehension of language complexity and cultural background of a text”. TESOL Quarterly 16: 503–516. Jones, R. L. 1979. “Performance testing of second language proficiency”. In Concepts in Language Testing: Some Recent Studies, E. J. Briére and F. B. Hinofotis (eds), 50–57. Washington, DC: TESOL.

Some construct validity issues

Jones, R. L. 1985. “Second language performance testing: An overview”. In Second language performance testing, P. C. Hauptman, R. LeBlanc and M. B. Wesche (eds), 15–24. Ottawa: University of Ottawa Press. Lado, R. 1961. Language Testing. New York: McGraw-Hill. Linn, R. L. 1994. “Performance assessment: Policy promises and technical measurement standards”. Educational Researcher 23 (9): 4–14. Linn, R. L., Baker, E. L. and Dunbar, S. B. 1991. “Complex, performance-based assessment: Expectations and validation criteria”. Educational Researcher 20 (8): 15–21. Lynch, B. K. 1997. “In search of the ethical test”. Language Testing 14: 315–327. McNamara, T. F. 1996. Measuring Second Language Performance. London: Longman. Messick, S. 1989. “Validity”. In Educational Measurement, R. L. Linn (ed.), 13–103. New York: American Council on Education. Messick, S. 1994. “The interplay of evidence and consequences in the validation of performance assessments”. Educational Researcher 23 (2): 13–23. Messick, S. 1996. “Validity and washback in language testing”. Language Testing 13 (3): 241–256. Morrow, K. 1979. “Communicative language testing: Revolution or evolution?” In The Communicative Approach to Language Teaching, C. J. Brumfit and K. Johnson (eds), 9–25. Oxford: Oxford University Press. Moss, P. A. 1992. “Shifting conceptions of validity in educational measurement: Implications for performance assessment”. Review of Educational Research 62: 229–258. Moss, P. A. 1995. “Themes and variations in validity theory”. Educational Measurement: Issues and Practice 14 (2): 5–13. Newman, F., Brandt, R. and Wiggins, G. 1998. “An exchange of views on ‘semantics’, psychometrics and assessment reform: A close look at ‘authentic’ assessments”. Educational Researcher 27 (6): 19–22. Oller, J. W. J. 1979. Language Tests at School. London: Longman. Ryans, D. G. and Frederiksen, N. 1951. “Performance tests of educational achievement”. In Educational Measurement, E. F. Lindquist (ed.), 455–494. Washington, DC: American Council on Education. Shavelson, R. J., Baxter, G. P. and Pine, J. 1992. “Performance assessments: Political rhetoric and measurement reality”. Educational Researcher 21 (4): 2–22. Shohamy, E. 1993a. “The exercise of power and control in the rhetorics of testing”. In Language Testing: New Openings, A. Huhta, K. Sajavaara and S. Takala (eds), 23–38. Jyväskylä: University of Jyväskylä. Shohamy, E. 1993b. “The power of tests: The impact of language tests on teaching and learning”. NFLC Occasional Papers. Washington, DC: National Foreign Language Center. Shohamy, E. 1997. “Testing methods, testing consequences: Are they ethical?” Language Testing 14: 340–349. Shohamy, E., Donitsa-Schmidt, S. and Ferman, I. 1997. “Test impact revisited: Washback effect over time”. Language Testing 13 (3): 298–317. Skehan, P. 1984. “Issues in the testing of English for specific purposes”. Language Testing 1 (2): 202–220.

89



90

Lyle F. Bachman

Spolsky, B. 1968. “Language testing: The problem of validation”. TESOL Quarterly 2 (2): 88–94. Spolsky, B. 1973. “What does it mean to know a language? Or, how do you get someone to perform his competence?” In Focus on the Learner: Pragmatic Perspectives for the Language Teacher, J. John, W. Oller and J. Richards (eds), 164–176. Rowley, MA: Newbury House. Spolsky, B. 1981. “Some ethical questions about language testing”. In Practice and Problems in Language Testing, Vol. 1, C. Klein-Braley and D. K. Stevenson (eds), 5–21. Frankfurt: Peter D. Lang. Spolsky, B. 1986. “A multiple-choice for language testers”. Language Testing 3 (2): 31–40. Spolsky, B. 1995. Measured Words. Oxford: Oxford University Press. Spolsky, B. 1997. “The ethics of gatekeeping tests: What have we learned in a hundred years?” Language Testing 14: 242–247. Stevenson, D. K. 1985. “Authenticity, validity and a tea party”. Language Testing 2 (1): 41–47. Terwilliger, J. 1997. “Semantics, psychometrics and assessment reform: A close look at ‘authentic’ assessments”. Educational Researcher 26 (8): 24–27. Terwilliger, J. S. 1998. “Rejoinder: Response to Wiggins and Newman”. Educational Researcher 27 (6): 22–23. Upshur, J. A. 1979. “Functional proficiency theory and a research role for language tests”. Concepts in Language Testing: Some Recent Studies, E. Briére and F. B. Hinofotis (eds), 75–100. Washington, DC: TESOL. Wesche, M. B. 1987. “Second language performance testing: The Ontario Test of ESL as an example”. Language Testing 4 (1): 28–47.



What second language learners can tell us about the native speaker Identifying and describing exceptions* Alan Davies University of Edinburgh

The problem in language proficiency testing During the last 30 years there appears to have been a loss of nerve about the native speaker goal among language testers. Here is Robert Lado (1961): When the (test) items have been written and the instructions prepared the test is ready for an experimental administration to native speakers of the language. … Items eliciting the desired response from native speakers 95% of the time or better should probably be kept. (Lado 1961: 93–94).

And now Lyle Bachman (1990): there are serious problems in determining what kind of language use to consider as the ‘native speaker’ norm, while the question of what constitutes a native speaker, or whether we can even speak of individuals who are native speakers, is the subject of much debate. (Bachman 1990: 248).

These are indeed serious problems, as Bachman reminds us, and the solution, that of delineating the language proficiency continuum from zero to ultimate attainment in terms of the native speaker, is, as he indicates, now in doubt. In doubt therefore are scales such as the Foreign Service Institute (FSI), now the Inter Agency Roundtable (ILR) and the Australian Second Language Proficiency Ratings (ASLPR) and the like, which typically fix a criterion of native speaker ability, thus: FSI Level 6: Native pronunciation, with no trace of “foreign accent” FSI Level 6: Understands everything in both formal and colloquial speech to be expected of an educated native speaker ASLPR Level 5: Able to use the second language as effectively as native speakers

92

Alan Davies

(“there is nothing about the way I speak that suggests I am not a native speaker”)

The parallel with nation Benedict Anderson’s discussion of the concept nation as an “imagined political community” provides a useful analogy to our consideration of the native speaker. He notes how difficult it is to define nation: It is an imagined political community — and imagined as both inherently limited and sovereign. It is imagined because the members of even the smallest nation will never know more of their fellow-members, meet them, or even hear of them, yet in the minds of each lives the image of their communion. (Anderson 1991: 6–7).

While the imagination may be sufficient for identifying with a nation, it will not do when we need to operationalise the native speaker. It is to that I now turn.

Defining the native speaker The Native Speaker, like Lewis Carroll’s snark, is a useful and enduring linguistic myth; again, like the snark, itself a product of the debate over idealism in philosophy, it must be taken with a large pinch of salt. Linguists may have given a special place to the native speaker as “the only true and reliable source of language data” (Ferguson 1983: vii), but there is little detailed discussion of the concept, which is often appealed to but difficult to track down. Full-length treatments of the topic (Coulmas 1981; Paikeday 1985; Davies 1991) are recent. Ferguson’s argument has to do with language use rather than with language knowledge: “much of the world’s verbal communication” he writes, “takes place by means of languages which are not the users’ mother tongue but their second, third or nth language, acquired one way or another and used when appropriate. This kind of language use merits the attention of linguists as much as do the more traditional objects of their research”. (Ferguson op. cit.). We can agree with Ferguson’s desire that linguistics pay more attention to language use without agreeing with his dismissal of the native speaker: “In fact the whole mystique of native speaker and mother tongue should preferably be quietly dropped from the linguist’s set of professional myths about language”. (ibid.).

Learners and native speakers

Theoretically, the Native Speaker concept is rich in ambiguity. It raises, quite centrally, the issue of the relation between the particular and the universal. Responding to Paikeday (1985), Chomsky maintains that to question the difference between native and non-native is just pointless. Chomsky’s argument depends on a rationalist opposition to “incorrect metaphysical assumptions, in particular the assumption that among the things in the world there are languages or dialects, and that individuals come to acquire them”. (Paikeday 1985: 49). What such views indicate is the accuracy of Coulmas’ statement that a tension exists between the flesh and blood and the idealisation definitions. The practical importance of the term is emphasised by Paikeday (1985), pointing to the employment discrimination against those who lack the “ideal” Native Speaker attributes: “native speakership should not be used as a criterion for excluding certain categories of people from language teaching, dictionary editing, and similar functions”. (Paikeday 1985: 88). Paikeday’s own solution seems to be to separate the ideal and the operative meanings of native speaker, making proficiency the criterion for employment, and personal history the criterion for ideal membership. (Davies 1995). The challenge which Paikeday sets us is essentially which native speaker to choose, and lurking behind all such choices is undoubtedly his dilemma of whether a new model (which can be supported by acknowledged proficiency) matches a distant “historically authentic” model; for example Indian English models or Nigerian English models versus British or American models. But this dilemma is in fact just one example of the more general case. There is equally a dispute between the British and American models just as there is among other metropolitan models, and just also as there is between any Standard and other dialects. The important choice of a model therefore raises issues of acceptability, of currency and of intelligibility.

The second language learner as native speaker I now address the question of whether a second language learner can be/become a native speaker of that language. Let me suggest six criteria, to which I respond in turn: 1. Childhood acquisition No, the second language learner does not acquire the target language in early childhood. If s/he does then s/he is a (bilingual) native speaker of both the first language (L1) and the target language (TL).

93

94

Alan Davies

2. Intuitions about the individual’s idiolectal grammar Yes, it must be possible, with sufficient contact and practice, for the second language learner to gain access to intuitions about his/her own idiolectal grammar of the target language. 3. Intuitions about standard language grammar Yes again, with sufficient contact and practice, the second language learner can gain access to the standard grammar of the target language. Indeed in many formal learning situations it is exactly through exposure to a TL standard grammar that the TL idiolectal grammar emerges, the reverse of L1 development. 4. Discourse and pragmatic control Yes, there may indeed be a descriptive difference between native speaker and non-native speaker groups but it is not in any way explanatory: that is to say it in no way argues that a particular second language learner cannot become a native speaker, given a favourable set of circumstances. 5. Creative performance Yes again, with practice it must be possible for a second language learner to become an accepted creative writer or speaker in the TL. There are of course well known, for example Conrad, Becket and Senghor. 6. Interpreting and translating Even though international organisations generally require that interpreters should interpret into rather than out of their L1, there are attested cases of nonnative speakers being employed as official interpreters and translators. All except (1) are contingent issues. In that way the question: “can a second language learner become a native speaker of a target language?” reduces to: is it necessary to acquire a code in early childhood in order to be a native speaker of that code? Now the answer to that question is to ask a further question, what is it that the child acquires in acquiring his/her L1? But I have already answered that question in my criteria (2)–(6) above, and so the question again becomes a contingent one. Attempts at definition are therefore vitiated by the unhelpful reductionism of the analytic statement which equates native speakers to one characteristic: childhood acquisition. The consequent lack of contingency appears to invalidate all empirical investigation. There is also the cultural dimension we probably need to consider (Nos 2 and 3 above) since the first language child acquirer does have access to the

Learners and native speakers

resources of the culture attached to the language and particularly to those learnt and encoded or even imprinted early. But to what extent are such differences critical? What of subcultural differences between for example the Scots and the English; of different cultures with the same standard language (for example the Swiss, the Austrians, the former two Germanies); or of different cultures with different standard languages (for example the British and the Australian)? What too of International English and of an isolated L1 in a multilingual setting (for example Indian English)? Given such interlingual differences and the lack of agreement on norms that certainly occurs among such groups it does appear that the second language learner has a difficult but not an impossible task to become a native speaker of a target language.

The Second Language Acquisition Research Approach Such a conclusion may be more sociological than linguistic or psychological. For in addition to the tension referred to above between the ideal and the flesh and blood approaches to the native speaker, there is a further opposition between the sociological and the psychological views; they are not easily reconcilable. Coppieters (1987) nicely points up the lack of fit in her account of a grammatical judgement experiment. She took a group of 27 non-native adult speakers of French who had “so thoroughly mastered French that it was no longer clearly possible to distinguish them from native speakers by mistakes which they made, or by the restricted nature of their choice of words and constructions” (1987: 544). For baseline data she took 20 native speakers of French, matched with the experimental group as far as possible. She used 107 sentences illustrating various aspects of French and asked her subjects individually for acceptability judgements. Her results indicated that the two groups belonged to two different populations, with no overlap between, even at extremes. Yet in spite of the clear divide she finds between native and nonnative speakers, she accepts that there is an argument in favour of identity theory (Tajfel 1981): “A speaker of French is someone who is accepted as such by the community referred to as that of French speakers, not someone who is endowed with a specific formal underlying linguistic system”. Nevertheless, for Coppieters, such an argument is strongly sociological and in her view competence must include a psychological dimension. She continues: “it is also clear that the variation between native speakers and non-native speakers cannot simply be subsumed as a special case of the variation among

95

96

Alan Davies

native speakers: that is non-native speakers have been found to lie outside the boundaries on native speaker variation” (1987: 545). Native speakers, reports Coppieters, “did not need the help of an explicit context. No matter how skilful non-native speakers might be at deriving the appropriate interpretation of a sentence in context, their inability to do so in the absence of an explicit context indicates a fundamental difference between their knowledge of the language and that of native speakers” (1987: 566–567). In view of the idiolectal and dialectal differences among native speakers themselves, Coppieters’ claim is a strong one, raising, as Spolsky notes, “the interesting possibility that native-speaker-like performance may be based on different underlying competence” (from that of learners) (Spolsky 1989: 45). Coppieters’s argument for cognitive rather than formal dissonance between native and non-native speakers concerns the grammar of the standard or common language learned before the critical period (Lenneberg 1967). Even so, this claim has been challenged. Birdsong (1992) takes issue with researchers such as Long (1990) who appear to make an absolute distinction between the native speaker and the nonnative speaker, viz that “ultimate attainment” for the non-native speaker can never be equal to native speaker competence. Birdsong reexamines the Coppieters (1987) experiment with learners of French and reports also on his own parallel study. What he concludes is that while as a group his French language learners and the French native speaker subjects differed significantly, the large amount of overlap suggests that “this general lack of difference is taken as evidence that ultimate attainment by nonnatives can coincide with that of natives” (1992: 739). Of course those who overlap are, as Birdsong admits, “exceptional learners”; but the implication here is that “our attention should turn to the issue of trainability: what can be discovered from exceptional learners that could be applied to improve other learners’ chances of attaining native norms”. (1992: 742). Birdsong’s challenge to the view that ultimate attainment (native speaker ability) is not possible for exceptional second language learners has been supported by Ellen Bialystok (1997), who has shown that on the basis of her experiments it is not only the exceptional learner who is capable of such attainment. She accepts that “the general success of younger learners in acquiring a second language is true” (ibid.: 133) but points out that the evidence does not therefore mean that “this advantage is the reflection of a sensitive period in

Learners and native speakers

learning” (ibid.: 133). She rejects the casuistry of those who wish to add qualifications to the claim that a second language learner can become a native speaker; as she writes: “there either is a maturational constraint on second language acquisition or there is not” (ibid.: 134). And she concludes that it is not in fact only the exceptional learner who can overcome the problems of reaching ultimate attainment after the sensitive period. Indeed it is quite a normal feat: “it is prudent to assume that successful second language acquisition remains a possibility for all those who have learnt a natural language in childhood and can organize their lives to recreate some of the social, educational and experiential advantages that children enjoy” (ibid.: 134). The fact of the success of these exceptional learners suggests that the native speaker is as much a sociolinguistic construct as a developmental one.

The sociolinguistic approach Coppieters represents the strong psychological position. According to that the native speaker is defined by early acquired knowledge. Bartsch (1988) takes a more sociological view, allowing for the importance of attitude and identity. Although both views concern control of the standard language, they are probably not reconcilable. But why should they be? The concept “native speaker” is used entirely appropriately in these quite different ways. It is probable that what is most enduring about the concept has nothing to do with truth and reality, whether or not individuals are native speakers; what matters most is the enduring native speaker myth combining both knowledge and identity: in that myth the two views have an equal role. Bartsch (1988) states that “norms are the social reality of the correctness notions … In this way correctness concepts which are psychic entities have a social reality”. (Bartsch 1988: 4). The correctness notions are the “how to behave notions”, similar to all other forms of learnt behaviour: how to ski, drive, play an instrument, dress and so on. What distinguishes language from these other skilled behaviours is that in addition to the psychic entities (knowing whether or not you are doing it well, right and so on) there is the social reality which carries and provides sanctions. The rules which are attributed to language by linguists, those which are constructed for the grammar of what Jespersen callled the common language,

97

98

Alan Davies

are therefore in part an acknowledgement and a working out of the intricate normative system acquired in taking on a standard language. And it is important to remember that for Bartsch (and for us) correctness is not restricted to a few shibboleths such as in English: it’s I/me; who/whom; will/shall) however frequently they may occur in teaching programmes, in primers and as examples of the supposed uselessness of the whole notion of correctness put forward by libertarian descriptivists. Correctness for Bartsch includes: the basic means of expression, lexical items, syntactic form, texts, semantic expression, pragmatic correctness. There is no argument here for triviality and no want of indication of the importance for language acquisition of correctness. Norms are established in terms of central models in speaking and writing ; and those models may be individuals or more likely elite groups. Bloomfield (1927) writes: “the Menonimi will say that one person speaks well and another badly, that such-and-such a form of speech is incorrect and sounds bad, and another too much like a shaman’s preaching or archaic” (“the way the old, old people talked”). Bloomfield notes that though a foreigner he was able to share in these judgements of the Menonimi. The nearest approach to an explanation of “good” and “bad” language seems to be this, then, that by a cumulation of obvious superiorities, both of character and standing, as well as of language, some persons are felt to be better models of conduct and speech than others. (Bloomfield 1927).

Standard languages, like religions, provide a commonality but not a homogeneity. That is why it makes sense to speak of norms rather than of rules which the standard language member accepts.

Empirical studies Grammatical judgements Ross (1979) in his paper Where’s English? discusses grammaticality judgements. He dismisses the strong hypothesis: English consists of that set of sentences which all speakers agree is grammatical, in favour of the much weaker: Beyond the core there is no single continuum of acceptability. (Ross 1979: 129, 131). Ross (1979) reports an empirical study of grammaticality judgements of 13 English sentences, selected so as to offer a range of sentences types on what he

Learners and native speakers

regards as the continuum of grammaticality. Ross’s results, based on the 12 sentences in the Appendix (he discarded one of the original 13), may be summed up thus: the sentences of a language seem to be viewed by speakers as falling into three groups, a Core (Sentences 1, 10), a Bog, and a Fringe (Sentences 12, 5). (Ross 1979: 156).

He finds general (though even here not universal) agreement about the grammaticality of his Core sentences; there is some variability in consensus that the Fringe sentences are not grammatical, and then comes the greatest variability in agreement about the Bog sentences (all the rest). He continues There are fairly clear differences between linguists and normals (the latter view themselves as more conservative than the former, and … reject more sentences, and with greater confidence than the former do). There are also differences between native and foreign speakers, with the latter tending to reject more sentences than the former do, and also tending to make fuller use of all four grades than the former do (ibid.: 156). Table 1. Choices of grammaticality sentences by Non-Native Speakers NNS

1

2

3

4

5

6

7

8

9

10 11 12

Siswati Greek Hungarian German Japanese Hungarian Urdu Japanese Korean Dutch Japanese Hebrew Chinese Chinese Kirundi Spanish Japanese Sinhala

1 1 1 1 1 1 1 1 1 2 1 2 1 1 1 1 1 1

4 3 3 2 4 3 2 3 4 2 3 4 3 4 1 3 3 3

2 1 1 1 1 1 2 2 2 2 2 4 2 2 1 2 1 3

2 1 1 3 2 2 1 1 1 1 2 3 1 2 1 1 2 2

4 4 4 4 4 4 2 3 4 3 4 4 2 4 1 4 4 4

3 3 4 3 2 3 3 4 1 2 1 2 2 1 4 1 3 3

3 3 4 3 4 3 1 3 3 4 4 3 3 2 1 2 3 4

1 2 2 4 4 4 1 3 3 2 2 1 3 3 1 1 4 1

4 2 2 2 1 2 1 3 3 3 3 4 4 4 1 2 2 2

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Mean

1.1 3.0 1.8 1.6 3.5 2.5 3.0 2.2 2.5 1.0 1.5 3.3

1 2 3 1 1 1 1 2 1 1 2 2 2 1 1 1 2 1

4 3 3 4 3 4 3 3 2 4 3 4 4 3 4 2 3 4

Mean 2.5 2.2 2.4 2.3 2.4 2.4 1.6 2.4 2.2 2.2 2.3 2.8 2.3 2.3 1.5 1.7 2.4 2.4

99

100 Alan Davies

Table 2. Choices of grammaticality sentences by Native Speakers NS

1

2

3

4

5

6

7

8

9

10

11

12

Mean

ENS 1 ENS 2 ENS 3 ENS 4 ENS 5 ENS 6 ENS 7 ENS 8 ENS 9 ENS 10 ENS 11 ENS 12 ENS 13 ENS 14 ENS 15 ENS 16

1 1 1 2 1 1 1 1 1 1 1 1 1 1 1 1

4 2 3 3 3 3 3 2 2 3 2 3 1 4 2 2

1 1 1 3 2 4 1 3 1 1 1 2 1 2 1 1

2 2 1 2 1 2 2 3 1 2 2 2 1 2 2 1

3 3 1 2 3 3 2 1 4 4 4 2 3 4 2 3

1 2 1 2 1 1 1 1 4 2 2 1 1 2 2 2

2 4 2 4 4 4 3 4 2 3 3 3 4 3 3 4

1 1 1 1 1 2 1 1 1 1 1 1 1 2 1 2

2 1 2 3 3 3 1 1 4 2 2 3 2 4 2 2

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

2 2 1 2 1 2 2 3 1 1 1 1 2 2 2 1

2 4 2 2 3 4 1 4 4 3 3 4 4 2 2 3

1.8 2.0 1.4 2.2 2.0 2.5 1.6 2.1 2.2 2.0 2.3 2.0 1.8 2.4 1.7 1.9

Mean

1.1 2.7 1.6 1.7 2.7 1.7 3.3 1.2 2.3 1.0 1.7 3.0

Ross invited his subjects (n = 30) to rate the sentences on the 4-point scale as follows: 1. The sentence sounds perfect. You would use it without hesitation. 2. The sentence is less than perfect-something in it just doesn’t feel comfortable. Maybe lots of people could say it, but you never feel quite comfortable with it. 3. Worse than (2), but not completely impossible. Maybe somebody might use the sentence, but certainly not you. The sentence is almost beyond hope. 4. The sentence is absolutely out. Impossible to understand, nobody would say it. Un-English. (Ross 1979: 161).

I replicated Ross’s study (using a slightly localised version of his questionnaire) on a sample of applied linguists, all of whom had had experience as English language teachers. The sample (n = 34) was made up of staff (faculty) members, research students and Master’s course-work students, containing both native speakers (NS, n = 16) and non native speakers (NNS, n = 18) of English. The non-native speakers of English were in all cases fluent, highly proficient English

Learners and native speakers

Table 3. Means and SDs for NS and NNS on grammaticality sentences

NS NNS

Sentence mean

Sentence mean

1.99 2.23

0.78 0.85

speakers. The NS in the sample were mostly speakers of British English). My results were similar to Ross’s on his test. I report here only one feature, the Native Speaker (NS)–Non-native speaker (NNS) difference in terms of the overall sentence acceptance and rejection rate. Here are the results for NS/NNS selection of the four categories for all 12 sentences, ranging from Category 1 (“the sentence sounds perfect”) to Category 4 (“the sentence is absolutely out … Un-English”).

Results 1. The two groups are significantly different at the 1% level [for independent samples] = 18.5, df = 32 ). As the Total Sentence Mean and SD for the two groups indicate (Table 3), the NS range is narrower than the NNS range (Mean = 1.99: 2.23; SD = 0.78: 0.85). Not that this means that the NS are more conservative. Indeed, on the contrary, they are more adventurous, or perhaps we should say they are more tolerant of uncertainty with regard to grammaticality. If they were more conservative than the NNS, their ratings for sentences would focus narrowly in the middle range, with predominant choices of the 2s and 3s (between the perfect 1s and the absolutely out 4s). But in fact what they do (relative to the NNS) is to cluster their choices on the 1s and 2s, giving the benefit of the doubt to doubtful sentences. The NNS on the other hand choose 3s and 4s more frequently. It is they, as a group, who are more conservative, or perhaps here we should say, less aware of possibilities, less wide-ranging in their knowledge of potential contexts of use, more concerned with the risk of being wrong. The Mean choices for the 4 ratings for the two groups are given in Table 4: 2. At the same time, the two groups are similar. A product moment correlation between the two sets of sentence means gives an r of 0.88. In other words, in spite of the difference in mean amount (reported in 1 above), in terms of variance the two groups tend to order the sentences in the same way. A Sign Test (n = 12; t = 11 (one pair identical), l = 3) indicated a ns result since the obtained value of L is greater than the highest required value of L (1) where

101

102

Alan Davies

Table 4. Mean ratings by NS and NNS on grammaticality sentences

NS NNS

1

2

3

4

4.9 4.3

3.6 2.7

1.9 2.6

1.5 2.3

t = 112 At the same time, since in 8 of the paired values NNS is higher, there does seem to be a suggestion that higher scores are more likely to occur for NNS than for NS. But of course we knew this already from the differences reported between group means and SDs. 3. In terms of Ross’s three categories of grammatical uncertainty (Core: Sentences 1,10; Fringe: Sentences 5,12; Bog: Sentences 2,3,4,6,7,8,9,11), the two groups agreed almost exactly on the grammaticality of the two Core sentences (Tables 1 and 2). On the two Fringe sentences, the NNS were less accepting than the NS, but most disagreement was on certain Bog sentences, notably Sentence 8 (Mean difference of 0.91) and Sentence 6 (Mean difference of 0.82) (8) That is a frequently talked about proposal. (6) I urge that anything he touch be burned.

In both cases the differences were in the expected direction. These are extreme examples of the general case we have already referred to, that the NNS are less likely to accept as grammatical sentences about which they are uncertain. The NS on the other hand are either more knowledgeable or they are prepared to give such sentences the benefit of the doubt, and it is precisely in what Ross calls the Bog that this difference exhibits itself most acutely. A crude summary of the NS/NNS group differences then might be that NNs have internalised the same grammatical judgements as NS but don’t as a group know as much. 4. What about their performance as individuals? Does the NNS group contain any of Birdsong’s “exceptional learners”? In a non-linguistic sense they are all candidates for such ascription. They are all fluent adult speakers of English; all are or have been teachers of English, all have studied English for at least 10 years. None of them started acquiring English as a young child. In order to support my claim that in selected cases NNS individuals, on this evidence, are indistinguishable linguistically from NS, I present below ratings

Learners and native speakers 103

for the 12 sentences for three pairs of respondents having the same or similar total scores. The first pair contains 2 NS, the second and third pairs have one NS and one NNS. As will be seen, the choices by the pairs of NS/NNS are just as similar/dissimilar as those made by the pair of NS. Since it is the case that all of the NS produce a unique set of ratings and no two profiles are identical, it seems reasonable to argue that, in certain cases, individual NNS cannot be distinguished from what is normal variation within a similar group of NS. The 2 NS (both with a mean for sentence ratings of 2.0) agree with one another in 5 ratings (Table 5); they differ by one level in 6 cases and by two levels in one case. Similarly, NS/NNS pair 1 (both with a mean for sentence ratings of 1.7) agree with one another in 5 ratings (Table 6); they differ by one level in 6 cases and by two levels in one case. In the second NS/NNS pair (NS/NNS pair 2) while the overall means differ in the expected direction (2.3: 1.0) the two members of the pair agree with one another in 7 of the 12 cases and differ by only one level in the other five. (Table 7). Table 5. Comparison of ratings by NS/NS pair ENS ENS

2 12

1 1

2 3

1 2

2 2

3 2

2 1

4 3

1 1

1 3

1 1

2 1

4 4

2.0 2.0

Table 6. Comparison of ratings by NS/NNS pair 1 Spanish ENS 15

1 1

3 2

2 1

1 2

4 2

1 2

2 3

1 1

2 2

1 1

1 2

2 2

1.7 1.7

4 1

2 1

1 1

1 2

4 4

2.3 2.0

Table 7. Comparison of ratings by NS/NNS pair 2 German ENS 2

1 1

2 2

1 1

3 2

4 3

3 2

3 4

104

Alan Davies

Discussion The question to which I address myself is what it is that binds the NS together (and to a large extent these NNS, who are, after all, near NS) as speakers of what Jespersen called a common language, given that they show such idiolectal variability when presented with grammaticality judgements. And the solution I offer is indeed the common language of Jespersen, or more relevantly perhaps, the Standard Language to which they (like the NNS) have been exposed through the Bartschian norms of education and training. If therefore native speakers become native speakers, at least in part, through education and training, NNS can also become native speakers of the target language if they too are exposed to similar education and training.

Pragmatic competence As a further check on the NNS potentiality to become “exceptional learners”, I replicated the Eisenstein and Bodman (1986) study of pragmatic competence in expressing gratitude in English. In their study Eisenstein and Bodman report an analysis of their questionnaire data relating to seven gratitude situations. Their analysis shows that native and non-native speakers of English varied in their responses. They emphasise the importance of those differences: advanced non-native speakers of English had considerable difficulty adequately expressing gratitude in the target language. Some problems were pragmalinguistic in nature, exhibiting divergence from native use on lexical and syntactic levels. Learners were often unable to approximate native idioms and routines. In our judgement, socio-pragmatic limitations were more severe, because the socio-cultural incongruities they revealed created the potential for more serious misunderstandings. (Eisenstein and Bodman 1986: 176).

In Table 2 of their article (ibid. 1986: 173; reproduced in the Appendix) they summarise the results for all non-native students. In spite of their general conclusion, quoted above, that the differences between native and non-native speakers were considerable, what this Table reveals is that their sample contained a substantial number of “exceptional learners” who were not distinguishable in this highly sophisticated communicative competence area from native speakers. Their Table 2 provides columns for the following categories of response: No Response; Not Acceptable; Problematical; Acceptable; Perfect; Not Comprehensible; Resistant.

Learners and native speakers 105

Table 8. NS/NNS gratitude expressions in three categories NNS

NS

Loan

Bald Bald + Expansion Elaborate

05 03 08

02 04 10

Sweater

Bald Bald + Expansion Elaborate

00 07 09

00 06 10

Dinner

Bald Bald + Expansion Elaborate

00 07 09

01 06 09

Totals

Bald Bald + Expansion Elaborate

05–10% 17–35% 26–54%

03–6%0 16–33% 29–60%

Summing across responses by non-native speakers gives a total mean for Perfect + Acceptable of 51%. If we exclude Acceptable (‘clear and appopriate language, but containing small errors which do not interfere seriously with native speakers’ understanding”) we still retain a total mean for Perfect of 31%. In other words of the 67 non-native speakers of American English in the USA, “in advanced-level ESL classes”, 20 were producing native speaker responses. Eisenstein and Bodman stress the limitations of their data, pointing in particular to the lack of fit between the respondents’ written responses to what were presented as situations requiring oral expressions of gratitude. But of course the same lack of fit existed for their native speaker subjects whom they used as a control. In a partial replication (slightly localised to the UK context) of the Eisenstein and Bodman investigation (n = 32; 16 native speakers of English; 16 nonnative speakers of English) I collected responses to three of their gratitude situations (the $5 loan, the sweater gift, the dinner invitation: see Appendix). These were all analysed in three categories, from “Bald” to “Bald plus Expansion” to “Elaborate” expressions of gratitude. Thus, an example of a Bald response to the offer of the loan would be: “Thanks a lot!”; of a Bald + Expansion: “Thanks very much. I really appreciate this.”; and of an Elaborate response (in this case with a promise and/or a comment): “Are you sure this is OK? Thanks very much. I’ll give it back on Monday, definitely”.

106

Alan Davies

Across all three situations (Table 8) the differences between native and nonnative speakers were minimal: non-native speakers had slightly more Bald + Expansion and slightly fewer Elaborate responses (Bald+Expansion : native speakers=16; non-natives=17; Elaborate: native speakers=29; non-natives=26).

Discussion In terms of Eisenstein and Bodman’s “Perfect” category, across all four of my situations, 56% of non-native speaker responses could be regarded as “perfect”. No doubt what counts as “perfect” is a matter of interpretation, and the figure of 56% refers to responses rather than to individual non-native speakers. Even so, a check of individual total responses indicates that at least three of these non-native speakers had completely “perfect” responses. It is therefore hard to avoid the conclusion that several of these non-native speakers of English were pragmatically “exceptional learners”.

Further research: Operationalising language test data to describe the NS Morton’s fork,3 o ered two equally unpalatable choices. Is that where we are with the native speaker? If the native speaker still seems problematic as a goal, is that, as I have suggested, because we remain doubtful whether learners can ever reach that goal, or is it because we are no clearer about how to define the native speaker? Can we avoid both choices by instead using second language test data to help describe the native speaker? After all, the measurement question will not go away, since we still need to demonstrate how it is we describe the attainment of these exceptional learners. There are several ways of demonstrating this. The first is to take examples of performances on tests by the kinds of second language learners who find careers in international agency interpreting or of course spying. It does not have to be test data of course. It could equally well be examples of writings (writings rather than speeches because of the distraction of accent) by such well-known learners as Nabokov, Senghor, Becket or Narayan. But such demonstration is perhaps supererogatory because we all know that learners can — in exceptional cases — reach such ultimate attainment. And if it is then countered that these are always limited tasks or domains, that there are other areas of native speaker control that are not within the demonstration, then the answer must be that this is true also of sub-groups of native speakers.

Learners and native speakers 107

A second way of demonstrating is to use test descriptors to describe exceptional learners’ attainment in order to examine to what extent they fail to reach ultimate attainment. For example, the descriptors on a test such as the Italian Teachers’ Proficiency Test (see below) purport to describe what a second language learner is required to achieve, But what is intriguing about performance at the highest level (Level 3) on these scales is how difficult it is to distinguish what is expected of the learner from what is assumed of the native speaker (Elder 1994). Certificate of Proficiency for Language Teachers: Italian Description of Test Performance: Level 3: In performing the range of test tasks the speaker demonstrated communicative abilities and strategies which were highly appropriate for the classroom context. The tone of voice, intonation, phrasing and pace of delivery and level of language used reflected a high degree of confidence and audience awareness. In explaining student error, the speaker was able to articulate the rules of Italian grammar and phonology in a clear and accurate manner with judicious use of specialist terminology.

A third demonstration is to use test data to chart language separation, that is performance on a test (like writing in a code) which illustrates a late stage of acceptance within a speech community of a new standard. In other words what is needed are answers to an EFL test (eg TOEFL) which would not receive a high mark from an American or British etc judge but would do so from a compatriot. Such evidence would support the claims of New English writers that the standard they are modelling is now localised. Exceptional learners are needed for all three demonstrations: in the first and second cases (joining the target standard) they indicate individual ability. In the third case (here we would call them prestige speakers) they show how individuals can influence the direction of a social group. Lowenberg (1995) argues that ‘diversification on a societal level is clearly a significant variable that can no longer be ignored in the measurement of English proficiency.’ (Lowenberg: 1995: 64). Interesting as such an argument is, it seems to me quite unlikely that any international language proficiency test would build in local variation in this way. What is more likely is the development of localised proficiency tests. The fourth type of demonstration is to establish an acceptable local foreign language level of attainment as the norm. This issue has been considered in relation to a teacher proficiency test for Indonesian teachers of English (Brown and Lumley 1995: 122–128, and Hill, this volume). What this test sets out to do is: (1) to assess English language proficiency as relevant to teachers, (2) to base

108

Alan Davies

its content on topics and situations relevant, familiar and hence accessible to people in Indonesia, (3) to emphasise assessment of the ability of inhabitants of the region to communicate effectively in English with each other, rather than relating their proficiency to metropolitan norms, (4) to rely entirely on local Indonesian teacher trainers as raters. The top level of performance on this test could, in essence, represent, in a reportable way, an approximation to native speaker proficiency.

Conclusion A standard language needs as its ‘members’ those who uphold its norms by taking on the responsibility of being its native speakers. Native speakers represent standard languages: it is the standard language they are native speakers of. What exceptional learners tell us is (1) that native speaker attainment is achievable in the target standard language and (2) that an alternative standard is possible when such learners are accorded the status of prestige speakers of the alternative standard. As we saw with both the Ross replication and the Eisenstein and Bodman replication in certain cases, both in grammaticality judgements and in pragmatic selections, individual NNS are indistinguishable from NS; these are exceptional learners. The native speaker is a fine myth: we need it as a model, a goal, almost an inspiration. But it is useless as a measure; it will not help us define our goals. So, in spite of my firm agreement with Birdsong and my conviction that there is a continuum between native speakers and non-native speakers, nevertheless, I recognise that for language teaching purposes what is crucial is the description of adequate partial proficiencies. It is to language tests that we look for these partial proficiency descriptions. This is what the second language learner can tell us about the native speaker which the native speaker’s lack of boundaries cannot.

Appendix A J. R. Ross Grammaticality Sentences (Ross J. R. 1979. “Where’s English?” In Individual Differences in Language Ability and Language Behavior, C. J. Fillmore, D. Kempler, and W. S.-Y. Wang (eds), 127–163. New York: Academic Press. 1. 2. 3.

Under no circumstances would I accept that offer Nobody who I get along with is here who I want to talk to We don’t believe the claim that Jimson ever had any money

Learners and native speakers 109

4. 5. 6. 7. 8. 9. 10. 11. 12.

The fact he wasn’t in the store shouldn’t be forgotten What will the grandfather clock stand between the bed and? I urge that anything he touch be burned All the further we got was to Sudbury That is a frequently talked about proposal Nobody is here who I get along with who I want to talk to The doctor is sure that there will be no problems The idea he wasn’t in the store is preposterous Such formulas should be writable down

(NB Sentences (11) and (12) in this list were (12) and (13) in Ross’s. I omitted Ross’s Sentence (11) in my study for the reason Ross himself gives: I have omitted from discussion the results of the eleventh questionnaire sentence, where some additional questions of a semantic kind, were asked, because the variation among the respondents was so overwhelming as to defy analysis. (Ross 1979: 134, fn. 2).

Appendix B Eisenstein and Bodman’s Gratitude Situations 1.* It’s Friday. You look in your wallet and notice that you only have $2.00. Your good friend at work notices this and hears you say “Darn, I’ll have to go to the bank.” Your friend asks if you need money, and you say that you forgot to go to the bank. Your friend says: “I have plenty. How much do you need?” You say: “Could you lend me $5.00? I’ll pay you back on Monday.” Your friend says “Sure. Are you sure you don’t need more than that?” You say you don’t. Your friend gives you the $5.00. 3.* It’s your birthday and you are having a few people over for dinner. A friend brings you a present. You unwrap it and find a blue sweater. 4. You work for a large company. The Vice-President of Personnel calls you into his office. He tells you to sit down, You feel a little nervous, because you have only been working there for 6 months. The Vice-President says “You’re doing a good job. In fact, we are so pleased with you that I am going to give you a $20.00 a week raise. 7. You find yourself in sudden need of money — $500.00. You mention this to a friend. Your friend immediately offers to lend it to you. You are surprised and very grateful. Your friend writes out a check for $500.00 and gives it to you. At first you say: “Oh, no, I didn’t mean for you to lend it to me. I couldn’t take it.” Your friend says: “Really, it’s all right. What are friends for?” After your friend insists again, you take the check. 9. Your friend suggests going out to lunch. You say that you’d like to go, but you only have $2.00. Your friend says, ”Ah, don’t worry. I’ll take you today.” Your friend takes you to a very nice restaurant-a much more expensive one than the ones you usually go to. You have a wonderful meal. Your friend pays and you get up to leave.



110

Alan Davies

10. You have just gotten a new and better job. A friend at the office tells you she has organized a farewell party for you. 14.* You have been invited to the home of a rather new friend. You have dinner with him and his wife and a few other friends of theirs. The food was great, and you really enjoyed the evening. As you leave, your hosts accompany you to the door. (original numbering: the three used in my study are asterisked)

Appendix C Summary of results of all students on individual questions (Eisenstein, M. and Bodman, J., 1986: 173, Table 2). Q

Q.topic

No Resp

Not accept Prob

Accept

Perf

NC

Resist

1

$5.00

5 7.5%

3 4.5%

9 13.4%

9 13.4%

39 58.2%

1 1.5%

1 1.5%

3

Sweat

2 3.0%

3 4.5%

19 28.4%

14 20.9%

29 43.3%

0 0%

0 0%

4

Raise

5 7.5%

10 14.9%

16 23.9%

14 20.9%

18 26.9%

1 1.5%

3 4.5%

7

$500

9 13.4%

3 4.5%

14 20.9%

15 22.4%

21 31.3%

1 1.5%

4 6.0%

9

Lunch

5 3.0%

11 16.4%

22 32.8%

9 13.4%

14 20.9%

1 1.5%

5 7.5%

10

Farewell

14 20.9%

7 10.4%

21 31.3%

11 16.45%

13 0 19.45% 0%

0 0%

14

Dinner

7 10.4%

2 3.0%

34 35.8%

23 34.3%

11 16.4%

0 0%

0 0%

Notes * A version of this chapter was read at the 1996 meeting of AAAL in Chicago. 1. l = Frequency of the less frequent sign. 2t = Frequency of both pluses and minuses. 2. John Morton, (1420–1500) was Archbishop of Canterbury and a minister of the English King Henry the Seventh. As a way of raising forced loans he would apply his “fork” — the

Learners and native speakers

argument that if people were obviously rich, then they could afford to pay, and if people looked poor, then they were obviously holding something back and so could also afford to pay. An early form of Catch 22.

References Anderson, B. 1983/1991. Imagined Communities. London: Verso. Bachman, L. 1990. Fundamental Considerations in Language Testing. Oxford: Oxford University Press. Bartsch, R. 1988. Norms of Language. London: Longman. Bialystok E. 1997. “The structure of age: in search of barriers to second language acquisition”. Second Language Research 13: 116–137. Birdsong, D. 1992. “Ultimate attainment in second language acquisition”. Language 68: 706–755. Bloomfield L. 1927. “Literate and illiterate speech”. American Speech 10: 432–439. Brown, A. and Lumley, T. 1995. “How can English proficiency tests be made more culturally appropriate? A case study: the assessment of English teacher proficiency”. In Intelec’94: International and English Language Education Conference: National and International Challenges and Responses, S. K. Gill (ed.), 122–128. Kuala Lumpur: Language Centre, Universiti Kebangsaan Malaysia. Coppieters, R. 1987. “Competence differences between native and near-native speakers”. Language 63: 544–557. Coulmas, F. (ed.). 1981. A Festschrift for Native Speaker. The Hague: Mouton. Davies, A. 1991. The Native Speaker in Applied Linguistics. Edinburgh: Edinburgh University Press. Davies, A. 1995. “Proficiency or the native speaker: what are we trying to achieve in ELT?” In Principle and Practice in Applied Linguistics: Studies in Honour of H. G. Widdowson, G. Cook and B. Seidlhofer (eds), 145–157. Oxford: Oxford University Press. Eisenstein, M. and Bodman, J. W. 1986. “’I very appreciate’: expressions of gratitude by native and non-native speakers of American English”. Applied Linguistics 7 (2): 167–185. Ferguson, C. 1983. “Language planning and language change”. In Progress in Language Planning: International Perspectives, H. Cobarrubias and J. Fishman (eds), 29–40. Berlin: Mouton. Elder, C. 1994. “Are raters’ judgements of language teacher effectiveness wholly language based?” Melbourne Papers in Language Testing 3 (2): 41–61. Giles, H. and Powesland, P. F. 1975. Speech Style and Social Evaluation London: Academic Press. Haugen, E. 1966. “Dialect, language and nation”. American Anthropologist 68 (4): 922–934. Lado, R. 1961. Language Testing. London: Longman. le Page, R. B. and Tabouret-Keller, A. 1985. Acts of Identity. Cambridge: Cambridge University Press. Lennerberg, E. 1967. Biological Foundations of Language. New York: Wiley.

111



112

Alan Davies

Long, M. 1990. “Maturational constraints on language development”. Studies in Second Language Acquisition 12 (3): 251–285. Lowenberg, P. H. 1995. “What do ESL tests test? Issues of cross-cultural norms”. In Intelec’94: International and English Language Education Conference: National and International Challenges and Responses, S. K. Gill (ed.), 57–65. Kuala Lumpur: Language Centre, Universiti Kebangsaan Malaysia. Paikeday, T. M. 1985. The Native Speaker is Dead! Toronto and New York: Paikeday. Ross, J. R. 1979. “Where’s English?” In Individual Differences in Language Ability and Language Behavior, C. J. Fillmore, D. Kempler, and W.S-Y Wang (eds), 127–163. New York: Academic Press. Spolsky, B. 1989. Conditions for Second Language Learning. Oxford: Oxford University Press.



The social responsibility of the language testers Elana Shohamy Tel Aviv University

Introduction Tests have a very powerful role in modern societies. Their results have detrimental effects on individuals as they can create winners and losers, successes and failures, rejections and acceptances. Test scores are often the sole indicators for placing people in class levels, for granting certificates and prizes, for determining whether persons will be allowed to continue in future studies, for deciding on professions, for entering special education classes, for participating in honor classes, for getting accepted to higher education and for obtaining jobs. Bourdieu (1991: 120) gives the following example to illustrate this detrimental effect: The paradigmatic example of this, and my starting point, is the competitive academic examination (concours): between the last person to pass and the first person to fail, the competitive examination creates differences of all or nothing that can last a lifetime. The former will graduate from an elite institution like the Ecole Polytechnique and enjoy all the associated advantages and perks, while the latter will become a nobody.

The detrimental effects of tests on test takers cause test takers to change their behaviors in order to maximize their performances on tests. Shohamy (1998) argues that it is this willingness of test takers to change their behavior along the demands of the test that drives decision makers and those in authority to introduce tests in order to cause test takers to change their behavior to along their lines. This way of using tests, it was further claimed, is unethical and undemocratic as it creates new de facto systems that over-rides existing ones, it re-defines content and has a negative effects on the quality of knowledge. Tests,

114

Elana Shohamy

therefore, are not neutral but rather embedded in political, social, educational, ideological and economic contexts. As a result of the strong and unchallenged power of tests, test taker develop fear and respect towards tests. They feel powerless as they perceive themselves as being in the hands of the testers that have control over them. They further feel that testers are capable of manipulating them while they have to follow the orders and the demands of tests. In discussing the symbolic power of tests Bourdieu (1985) argues that the power of those in authority is granted to them by the subjects of their power as those have vested interest in maintaining social order. As part of this agreement, Bourdieu notes, subjects grant those in authority unlimited, almost magic power. There is therefore, an unwritten contract between those in authority and those who are subject to the authority to maintain the power. Tests, serves them in carrying out this goal.

The purpose of the paper The purpose of this paper is to discuss the role of the tester, given the powerful uses of tests in society as tools for mostly political, social, educational and ideological agendas. Davies (1997) admits that there is a growing feeling among scholars in testing that challenges the morality of testing both within and outside the field. He therefore poses some difficult questions about the ethics and morality of language testers: … should testing specialists be responsible for decision beyond test construction? Who decides what is valid? Does professionalism conflict with (1) public and (2) individual morality?

Language testers, Davies (1997: 330) argues have both professional and moral obligations. No social science is immune from such criticism, as it becomes less and less tenable for academic discipline to allow their researchers the liberty to make up their own minds as to the conduct and practice of their professional research. There is urgent reason therefore both to examine the state of ethics in academic language testing and language testing research and to encourage a move towards an explicit statement of good conduct and practice.

Specifically, the topics which are brought up focus on the relationships with stake holders, the relationship between bias and fairness, backwash, the politics of

Social responsibility of the language testers

gatekeeping and a conflict between fairness and face validity. In other words, once testers become aware of misuses of tests in society they are faced with serious dilemmas as to their professional roles and responsibilities. Is the test developer responsible for uses and misuses of tests? What is the role of the tester once s/he notices misuses? Is his/her role at that point to warn against misuses or to take actual steps such as using sanctions against misuses? Or, perhaps it is not the responsibility of the tester at all to worry about the test takers after a test had been handed to the users. In this paper various perspectives regarding the responsibility of the tester are being explored.

Identifying the tester Before discussing the responsibility of the tester, there is a need for a clear definition of who should share it? Or, in other words what do we mean by ‘a language tester’? Such definitions are needed as they relate directly to the types of responsibilities. While references is often made to “the tester” or “the language tester”, it cannot be referred to as a monolithic concept. Is the tester the person who writes the questions and tasks of the test? Or alternatively, is it the policy maker who makes the decisions to introduce tests in a given system? Is it a government agency that funds testing operations? or, is it the principal who decides to administer tests in a given school? Is it the researcher that writes the test? Is it the researcher who administers the test in order to obtain information to answer important research questions? Is it the classroom teacher who writes a test and administers it to his/her students? or, are these the parents who put pressure on the school or a school systems to introduce tests for their children for accountability purposes? The answer to these questions is — all of the above, and more. A tester may be anyone who is involved with the test in any shape and form — writing test items, computing the tests’ statistics, designing the test, examining its reliability and validity, its norms and quality of items. The tester, then, refers to all those who take part in the decisions and actions surrounding the testing event and the testing experience. These include policy makers, researchers, question writers, statisticians, groups that pay for the test, language supervisors who make a decision to introduce and administer the tests, parents who support or motivate the introduction of tests, or even presidents who believe that tests will save their nations. ‘A tester’ then is all those who have some contribution to the act of testing. It is all these agents that need to share the social responsibility.

115

116

Elana Shohamy

This view is in line with Hamp-Lyons (1997: 298) who noted that it is not just the test developer who has responsibility but other bodies who take part in the introduction of the tests. … it is not only test developers whose work has ‘impact’; it is also testing agencies who make policy and economic decisions about the kinds of testing to support and the kinds that will not be supported; it is textbook publishers, who make economic decisions about the kinds of textbooks teachers and parents will buy to ‘ensure’ their children are ready for the test’; it is school districts, boards and ministries of education, and national or federal governments who bow to pressure to account for the progress of pupils and the value added effect of education.

What are the responsibilities? Five views Now that the di erent bodies that share the responsibilities have been identified, it is important to discuss the type of responsibilities. 1. An ethical perspective Davies argues for an ethical perspective of professional morality which is defined as a contract for the profession and the individual with the public, thereby safeguarding all three. Yet, such a contract cannot occur without clear dilemmas, such as ensuring a balance between a professional ethical code and the individual moral conscience and even raising questions about the right of the profession to exist. Yet, at the same time he argues for a limit on what is achievable in terms of responsibility of consequences: … the apparent open-ended offer of consequential validity goes too far. I maintain that it is not possible for a tester as a member of a profession to take account of all possible social consequences. What can be done is the internal (technical) bias analysis and a willingness to be accountable for a test’s fairness. In other words, limited and predictable social consequences we can take account and regard ourselves as responsible for. (Davies 1997: 336).

2. Making others aware Another view, not necessarily in contradiction with the first, is that the only responsibility a tester has with regards to the use of tests is to point out to society the intentions, e ects and consequences of tests. Accordingly, the role of the tester is to collect data about dimensions and aspects of use and make this information known

Social responsibility of the language testers

and accessible to the users and the public at large. Such a view follows Foucault (1991: 12) in Remarks of Marx with regards to the role of the intellectual in the society: The role of the intellectual does not consist in telling others what they must do. What right would they have to do that? … The job of an intellectual does not consist in molding the political will of others but rather to challenge the way we think … There is always something ludicrous in philosophical discourse when it tries, from the outside, to dictate to others, to tell them where their truth is and how to find it …

Spolsky (1998) notes that the danger of tests comes in their misuses, and that misuse is made more likely by our failure to continue to publicize their inevitable uncertainty as well as by our slowness to study and incorporate into our work the user’s perspective. 3. All consequences A di erent view is expressed by Hamp-Lyons (1997) who claims that the role of the tester is significantly broader. The tester needs to accept responsibility for all those consequences which the tester is aware of and more: The responsibility of the language testers is clear: we must accept responsibility for all those consequences which we are aware of. Furthermore, there needs to be a set of conditions and parameters inside which we are sure of the consequences of our work and we need to develop a conscious agenda to push outward the boundaries of our knowledge of the consequences of language tests and their practices.

4. Impose sanctions on misuses Yet, another view is that the responsibility of the tester is not just to collect data on intentions and e ects and make the public aware of such findings, but also to impose sanctions, punish and forbid the use of tests which are misused and of those who violate the standard of correct testing practices. This argument views the tester as responsible to the product, and the tester therefore has the obligation to forbid its use when it is found to be defective. According to this argument constructing a test is identical to the manufacturing of other products in society. It is similar, for example, to the responsibility of a manufacturer in designing a defective car, a poisonous drug or engaging in malpractice. In these situations it is understood that the manufacturer has a responsibility to generate a high level product, and failing to do so results in a penalty for damaged outcomes.

117

118

Elana Shohamy

5. Shared responsibility A different view follows the notion of shared responsibility — the responsibilities for good conduct are in the hands of all those who are involved in the testing process. It is argued that there is a need o change the balance of power with regards to tests. While up to now the testers had the authority of testing while the test taker was the passive object, calls are currently being made to change the balance of power by having the testers and test takers, and others who are involved in the process, share this authority. The notion of shared authority implies, according to this view, shared responsibilities as well. This view is based on the notion that testers need to admit that their knowledge is limited, that measuring knowledge by a single test administered by testing bodies cannot cover all domains of knowledge and that the testers do not have all the answers when undertaking assessment. Thus, testers and test takers should engage in a mutual constructive effort working together in constructing meaning of knowledge. The responsibility of the tester then, is to admit the limitations of the profession and to construct knowledge in a responsible way, by working together with a number of groups of users who accumulate evidence of the knowledge that is being assessed. The responsibility also means using a variety of critical testing methods in order to examine the uses of tests and become critical when misuses occur by collecting data on tests’ effects and consequences of tests, warning against misuses and protecting all those involved in the acts of testing. It should be viewed as the responsibility of all to work for better instrument and uses of tests and to warrant against misuses. Madaus (1991) as well, claims that there is a need for people from outside the testing community to address the social, technical and value issues associated with testing. Specifically, he notes that this suggestion is not limited to the involvement of scientists or technicians from other communities but rather to extend it to the wider non-technical communities. Thus, the responsibility for collecting data about the quality of the instruments as well as their uses is not only the responsibility of the tester but also of those with whom the tests are used and on whom the test has an effect (intended or not). These include test takers as well as others who are in charge of civil and fair behavior in society, such as courts and human rights groups. Although it is expected that testers will carry out some of this research they are clearly less motivated to do so as often the results of their studies are not trusted. A case in point is the low trust that the public has regarding research on the tests which are developed and marketed by testing institutions in the same way that there is low trust the public has in research conducted by drug companies regarding

Social responsibility of the language testers

the drugs they produce. Clearly, research is trusted more when it is performed by independent groups. One does not expect testers to do the research in the same way that malpractice is not being conducted by doctors, but rather by their patients. Thus, research on consequences should be done by others — test takers, human rights activists and others who are affected or not affected by test results. After all it is the role of society to defend itself from powerful institutions. It is therefore only logical that others, not only the testers, should be motivated to carry out research in order to defend themselves as is not expected that those who develop tests will be the sole bodies to carry out that research and examine its consequences. It is therefore suggested here that research on tests’ use should be conducted in collaboration with other stakeholders. It is essential that testers cooperate with experts not involved in testing and acquaint all those concerned with tests and their use with the vocabulary of testing. This process will enable others to enter the discourse without being dismissed as naïve, thus improving communication between groups and having more people outside the testing community address the social and technical values of testing. This is also related to the need for shared discourse and for specifying the type of information that needs to be collected.

Shared discourse A related issue to the shared responsibility is the shared discourse. So far, Madaus (1990: 14) notes, there has been a tense relationship between the testing community and lay people. The testing community’s language and value systems do not coincide with the language and value systems of many groups critical to testing. Consequently, in testing — as in other important technological areas — there is almost no middle ground of rational discourse, no available common language with which persons of differing background can discuss matters of technology in thoughtful, critical terms.

Madaus, therefore claims that an additional responsibility of testers is to acquaint all groups who have concerns about tests and their use, with the techniques and vocabulary of the testing community, so that they can enter into the discourse without being dismissed as naïve. There is a need to improve communication between groups, words in the common lexicon that have been

119

120

Elana Shohamy

appropriated and narrowed by the testing fraternity need to be abandoned so that we no longer talk on parallel non-intersecting tracks.

The type of information needed Adopting any of the above views regarding the types of responsibilities — whether it is to make the public aware or the sanction, or the shared responsibility, all require the collection of data about tests’ use. Thus, a relevant issue concerns the type of data that needs to be collected. Traditional testing excluded dimension of tests’ use, yet it should be clear by now that such information is crucial for upgrading the quality of tests. Thus, the responsibility of the tester is to collect data about the quality of the instrument, the relevant types of reliability and validity relevant for the different users and adapted to different needs. This also includes the use of multiple ways of disseminating information, including scores such as diagnostic information, comparisons, profiles, verbal descriptors, narratives, progress profiles, interpretive summaries and more. It also includes information about test use such as intentions and rationales for introducing the test, long and short term effects on test takers, education and society and other relevant information for tests’ consequences.

Professional morality Davies (1997: 328) brings up the topic of professional morality, which relates to the moral and ethical concerns that testers face once the field of testing encompass test’s use and consequences. He notes the need for professional morality in addition to public and individual moralities. The profession, he claims, establish contracts with the public but at the same time must protect their members. House (1990) notes regarding ethics that “Ethics are the rules or standards of right conduct or practice, especially the standards of a profession”. Thus, he claims, it is important for all stakeholders, both within the institutions and without, that professional standards of behavior are stated explicitly. So that stakeholders know what is meant by right conduct and practice and therefore what can and cannot be expected.

Social responsibility of the language testers

Following critical theorists in other social sciences, critical applied linguists have been asking questions about the ethics of applied linguistics and whether an ideologically neutral study of applied linguistics is possible. And where critical applied linguistics goes, critical language testing follows. There is urgent reason therefore both to examine the state of ethics in academic language testing and language testing research and to encourage a move towards explicit statement of good conduct or practice. (House 1990: 328–329).

Davies justifies the need for professional morality and provides examples of the type of issues that it can focus on: … professional morality … provides a contract for the profession and the individual with the public, thereby safeguarding all three. Professional research of a social nature is likely to state the profession’s ethical position on issues having to do with informed consent, a taboo on invasions of privacy, confidentiality, precautions against injury, and distress, covert methods of conducting research, and agreeing on conventions for writing and publishing. (Davies 1997: 333).

On the other hand, Davies notes that members of the profession need protection as well, clearly without reaching overprotection (especially the kind that can be associated with incestuous concern for the protection of members of the profession, by avoiding and covering over complaints and offences to avoid legal actions in a climate of increasing litigation). After all, there have already been cases, and surely there will be more, when public assessment provides for potential complaints and legal actions because of the use of assessment in selection and certification. Those excluded or rejected often believe that the methods of assessment used were not valid and therefore they were wrongly excluded or rejected. Thus, safeguards for professional practitioners are needed.

Standards for professional behavior One strategy of focussing on these issues is by developing standards of “professional behaviors” which are expected to protect both the tester as well as the test taker. Nevo and Shohamy (1986) introduced an adaptation they made of the Standards for Evaluation of Educational Programs, Projects and Materials (1981). These standards had been developed over a number of years by the Joint Committee on Standards for Educational Evaluation, made up of members from the AERA, the APA, the NCME and nine other organizations. They were designed in order to give guidance to professionals involved in the evaluation

121

122

Elana Shohamy

of educational programs, and it was hoped that the establishment of a common set of principles would help to upgrade the practice of educational evaluation. The Joint Committee developed a set of 30 Standards, divided into four headings: Utility Standards, Feasibility Standards, Propriety Standards and Accuracy Standards. Alderson, Clapham and Wall (1993) note the unique dimension of these standards: these standards show a stronger interest in the context of the testing situation and how the public views the test itself; the reports; and the effect of the test on the candidates, education and society. The political viability of a test — Testing is planned and conducted with anticipation of the different positions of various interest groups, so that their cooperation may be obtained’) is an issue which does not appear in other sets of testing standards.

Nevo and Shohamy extended these Standards to testing methods. From the Joint Committee’s original 30 Standards, they produced a list of 23 that they suggested were suitable for testing methods. All the standards were re-worded so that they would apply to test methods rather than to evaluation programs. The Standards were organized as follows: 1. Utility Standards: The Utility Standards are intended “to ensure that a testing method will serve the practical information needs of given audiences”. Utility assures that a test serves the practical information needs of a given audience; for example it is concerned with the impact that tests may have on instruction and learning in the classroom, such as backwash effects. Also included are other utility factors such as finding the best ways for training testers and raters and for presenting test results. They are also concerned with tester credibility, information scope, justified criteria, report clarity, dissemination timeliness and impact. 2. Accuracy Standards: The Accuracy Standards are intended “to ensure that a testing method will reveal and convey technically adequate information on the educational achievement of those that are being tested”. Thus accuracy assures that the test represents reliable and valid measurements, testing conditions, data analysis and objective reporting. 3. Feasibility Standards: Feasibility Standards are intended “to ensure that a testing method will be realistic, prudent and frugal”. Feasibility relates to whether tests are feasible to administer within different contexts, whether they are conducted with the support of certain political groups, whether they produce information of sufficient value to justify their costs, and whether it is practical to train people to conduct such tests.

Social responsibility of the language testers

4. Fairness Standards: These are intended “to ensure that a testing method is conducted legally, ethically, and with due regard to the welfare of tested individuals as well as those affected by test results”. Fairness relates to whether the tests are conducted legally and ethically, as well as for the welfare of the respondents. Included are questions on whether the tests are based on materials that examinees are expected to know, if they are conducted so that the rights and welfare of human subjects are respected, if they are conducted fairly with regard to the strength and weaknesses of the individuals tested, and if test-takers have a positive attitudes towards the tests. Nevo and Shohamy also note that each of these elements can be viewed as another dimension in a broader view of test validation. They tried the standards out on relevant professionals. They asked two groups to study the Standards and use them to rank four alternative testing methods which were being considered for a new national matriculation examination. The first group consisted of policy makers who were supposed to make the final decision about which of the testing methods would be included in the new exam; the second group consisted of testing experts attending a language testing conference. They also designed a sample test which contained all four of the testing methods under consideration. This was administered to 1000 students, to find out not only how the test performed in the real world, but also to provide a basis for assessing whether several of the new testing Standards could be used to evaluate such testing methods. It is important to note that Alderson, et al. claim that not enough attention was paid in the adaptation to the political aspects of the original standard. Specifically, they note that there is a need to refer more to the use of tests as political weapons and that testers need to be aware of all the possible ways in which their tests can be misused so that they can pre-empt any wrong-doing. They specifically focus on a distinction with regards to misuses of tests between misuse as the result of ignorance or carelessness or whether the parties may deliberately misuse information in the way that it implied in the Standard above. Thus, in making decisions as to the quality of tests, consideration must be given to a whole set of validation questions, which are related not only to accuracy but also to test utility, feasibility and fairness. Empirical data needs to be collected on each aspect of the test. Thus, there is more than one way to examine a test; different criteria and standards, as those listed above, are needed to provide answers regarding quality of tests.

123

124

Elana Shohamy

ILTA Codes of Ethics In the past few years a number of efforts have been taking place to promote the use of codes of Ethics in order to professionalize the field. There is currently a growing awareness for the need of a code of Ethics for language testers that will consider both the traits of the test as well as its uses and consequences. The Code of Ethics that is included in the Appendix is based on current discussion within the field of language testing. In March 2000 the Code of Ethics was finally adapted at the annual meeting of ILTA (the International Language Testing Association) held in Vancouver. This Code represents an example of an initiative to guard testers against misuses of tests. The Principles are set out, followed by Annotations. These elaborate and generally clarify the nature of the principles; they prescribe what ILTA members ought to do or not do, or more generally how they ought to comport themselves or what they, or the profession, ought to aspire to; and they identify the difficulties and exceptions inherent in the application of the principles. The Annotations further elaborate the Code’s sanctions, making clear that failure to uphold the Code may have serious penalties. Codes of this sort give guidance on how to act responsibly and they cover areas such as the professional role in society, integrity, conflicts of interest, diligence and due care, confidentiality and communication with clients and the public (Nitko 1996). The ability of such Codes of Ethics to actually change behavior to a more responsible one is clearly an open question.

Conclusion This article attempted to analyze various aspects related to the social responsibility of language testers; the issues are clearly to general testers as well. The topic is relatively new and as could be seen in this article there are still many resolved issues and unanswered questions regarding the types of responsibilities and the specific actions that need to be taken in order to pursue them. Yet, the importance of this discussion is that it is becoming an issue that is being dealt with, argued, discussed and debated in the open. Testers are now in a situation when they feel that they do have moral obligations to society, yet, justifiably so, do not want to and should not, take and share the entire responsibility for the bad as well as for the good consequences of tests.

Social responsibility of the language testers

Appendix: Code of Ethics for ILTA (adopted at the annual meeting of ILTA held in Vancouver, March 2000) This, the first Code of Ethics prepared by the International Language Testing Association (ILTA), is a set of principles which draws upon moral philosophy and serves to guide good professional conduct. It is neither a statute nor a regulation and it does not provide guidelines for practice, but it is intended to offer a benchmark of satisfactory ethical behaviour by all language testers. It is associated with a separate Code of Practice (in progress). The Code of Ethics is based on a blend of the principles of beneficence, nonmaleficence, justice, a respect for autonomy and for civil society. This Code of Ethics identifies 9 fundamental principles, each elaborated on by a series of annotations which generally clarify the nature of the principles; they prescribe what ILTA members ought to do or not do, or more generally how they ought to comport themselves or what they, or the profession, ought to aspire to; and they identify the difficulties and exceptions inherent in the application of the principles. The Annotations further elaborate the Code’s sanctions, making clear that failure to uphold the Code may have serious penalties, such as withdrawal of ILTA membership on the advice of the ILTA Ethics Committee. Although this Code derives from other similar ethical codes (stretching back into history), it does endeavour to reflect the ever changing balance of societal and cultural values across the world, and for that reason should be interpreted by language testers in conjunction with the associated Code of Practice. All professional codes should inform professional conscience and judgement. This ILTA Code of Ethics does not release language testers from the obligations and responsibilities laid on them by other Codes to which they have subscribed or from their duties under the legal codes, both national and international, to which they may be subject. Language testers are independent moral agents and sometimes they may have a personal moral stance which conflicts with participation in certain procedures. They are morally entitled to refuse to participate in procedures which would violate personal moral belief. Language testers accepting employment positions where they foresee they may be called on to be involved in situations at variance with their beliefs have a responsibility to acquaint their employer or prospective employer with this fact. Employers and colleagues have a responsibility to ensure that such language testers are not discriminated against in their workplace. The Code of Ethics is instantiated by the Code of Practice (currently under preparation by ILTA). While the Code of Ethics focuses on the morals and ideals of the profession, the Code of Practice identifies the minimum requirements for practice in the profession and focuses on the clarification of professional misconduct and unprofessional conduct. Both the Code of Ethics and the Code of Practice need to be responsive to the needs and changes within the profession and, in time, these Codes will require revision in response to changes in language testing and in society. The Code of Ethics will be reviewed within five years, or earlier if necessary.

125

126

Elana Shohamy

Principle 1 • Language testers shall have respect for the humanity and dignity of each of their test takers. They shall provide them with the best possible professional consideration and shall respect all persons’ needs, values and cultures in the provision of their language testing service. Annotation • Language testers shall not discriminate against nor exploit their test takers on grounds of age, gender, race, ethnicity, sexual orientation, language background, creed, political affiliations or religion, nor knowingly impose their own values (for example social, spiritual, political and ideological), to the extent that they are aware of them. • Language testers shall never exploit their clients nor try to influence them in ways that are not related to the aims of the service they are providing or the investigation they are mounting. • Sexual relations between language testers and their test takers are always unethical. • Teaching and researching language testing involving the use of test takers (including students) requires their consent; it also requires respect for their dignity and privacy. Those involved should be informed that their refusal to participate will not affect the quality of the language tester’s service (in teaching, in research, in development, in administration). The use of all forms of media (paper, electronic, video, audio) involving test takers requires informed consent before being used for secondary purposes. • Language testers shall endeavour to communicate the information they produce to all relevant stakeholders in as meaningful a way as possible. • Where possible, test takers should be consulted on all matters concerning their interests. Principle 2 Language testers shall hold all information obtained in their professional capacity about their test takers in confidence and they shall use professional judgement in sharing such information. Annotation • In the face of the widespread use of photocopied materials and facsimile, computerized test records and data banks, the increased demand for accountability from various sources and the personal nature of the information obtained from test takers, language testers are obliged to respect test takers’ right to confidentiality and to safeguard all information associated with the tester-test taker relationship. • Confidentiality cannot be absolute, especially where the records concern students who may be competing for admissions and appointments. A careful balance must be maintained between preserving confidentiality as a fundamental aspect of the language tester’s professional duty and the wider responsibility the tester has to society. • Similarly, in appropriate cases, the language tester’s professional colleagues also have a right to access data of test takers other than their own ion order to improve the service the profession offers. In such cases, those given access to data should agree to maintain confidentiality. • Test taker data collected from sources other than the test taker directly (for example from teachers of students under test) are subject to the same principles of confidentiality. • There may be statutory requirements on disclosure, for example where the language tester is called as an expert witness in a law court or tribunal. In such circumstances, the language tester is released from his/her professional duty to confidentiality.

Social responsibility of the language testers 127

Principle 3 Language testers should adhere to all relevant ethical principles embodied in national and international guidelines when undertaking any trial, experiment, treatment or other research activity. Annotation • Language testing progress depends on research, which necessarily involves the participation of human subjects. This research shall conform to generally accepted principles of academic inquiry, be based on a thorough knowledge of the professional literature; and be planned and executed according to the highest standards. • All research must be justified; that is proposed studies shall be reasonably expected to provide answers to questions posed. • The human rights of the research subject shall always take precedence over the interests of science or society. • Where there are likely discomforts or risks to the research subject, the benefits of that research should be taken into account but must not be used in themselves to justify such discomforts or risks. If unforeseeable harmful effects occur, the research should always be stopped or modified. • An independent Ethics Committee should evaluate all research proposals in order to ensure that studies conform to the highest scientific and ethical standards. • Relevant information about the aims, methods, risks and discomforts of the research shall be given to the subject in advance. The information shall be conveyed in such a way that it is fully understood. Consent shall be free, without pressure, coercion or duress. • The subject shall be free to refuse to participate in or to withdraw from, the research at any time prior to publication of research results. Such refusal shall not jeopardise the subject’s treatment. • Special care shall be taken with regard to obtaining prior consent in the case of subjects who are in dependent relationships (for example, students, the elderly, proficiency challenged learners). • In the case of a minor, consent shall be obtained from a parent or guardian but also from the child if he is of sufficient maturity and understanding. • Confidential information obtained in research shall not be used for purposes other than those specified in the approved research protocol. • Publication of research results shall be truthful and accurate. • Publication of research reports shall not permit identification of the subjects who have been involved. Principle 4 Language testers shall not allow the misuse of their professional knowledge or skills, in so far as they are able. Annotation • Language testers shall not knowingly use their professional knowledge or skills to advance purposes inimical to their test takers’ interests. When the progress of the tester’s intervention is not directly to the benefit of the test takers(for example, when they are asked to act as trial

128

Elana Shohamy

subjects for a proficiency test designed for some other situation), its nature shall be made absolutely clear. • Non-conformity with a society’s prevailing moral, religious etc. values, or status as an unwelcome migrant, shall not be the determining factor in assessing language ability. • Whatever the legal circumstances, language testers shall not participate, either directly or indirectly in the practice of torture or other forms of cruel, inhuman or degrading punishment (see Declaration of Tokyo 1975). Principle 5 Language testers shall continue to develop their professional knowledge, sharing this knowledge with colleagues and other language professionals. Annotation • Continued learning and advancing one’s knowledge are fundamental to the professional role; failure to do so constitutes a disservice to test takers. • Language testers shall make use of the various methods of continuing education that are available to them. These may involve participation in continuing language testing programmes and professional conferences, and the regular reading of relevant professional publications. • Language testers shall take the opportunity to interact with colleagues and other relevant language professionals as an important means of developing their professional knowledge. • Language testers shall share new knowledge with colleagues by publication in recognized professional journals or at meetings. • Language testers shall be expected to contribute to the education andprofessional development of language testers in training and to the drawing up of guidelines for the core requirements of that training. • Language testers shall be prepared to contribute to the education of students in the WIDER language professions. Principle 6 Language testers shall share the responsibility of upholding the integrity of the language testing profession. Annotation • Language testers shall promote and enhance the integrity of their profession by fostering a sense of trust and mutual responsibility among colleagues. In the event of differences of opinion, viewpoints should be expressed with candour and respect rather than by mutual denigration. • Language testers develop and exercise norms on behalf of society. As such theirs is a privileged position which brings with it an obligation to maintain appropriate personal and moral standards in their professional practice, and in those aspects of their personal life which may reflect upon the integrity of that practice. • Language testers who become aware of unprofessional conduct by a colleague shall take appropriate action; this may include a report to the relevant authorities. • Failure to uphold this Code of Ethics will be regarded with the utmost seriousness and could lead to severe penalties including withdrawal of ILTA membership.

Social responsibility of the language testers 129

Principle 7 Language testers in their societal roles shall strive to improve the quality of language testing, assessment and teaching services, promote the just allocation of those services and contribute to the education of society regarding language learning and language proficiency. Annotation • Language testers have a particular duty to promote the improvement of language testing provision/services in that many of their test takers are disenfranchised and lack power on account of their non-native speaker status. • Language testers shall be prepared by virtue of their knowledge and experience to advise those responsible for the provision of language testing services. • Language testers shall be prepared to act as advocates and join with others in ensuring that language testing test takers have available to them the best possible language testing service. • Language testers shall be prepared to work with advisory, statutory, voluntary and commercial bodies that have a role in the provision of language testing services. • Language testers shall take appropriate action if services, by reason of fiscal restriction or otherwise, fall below minimal standards. Exceptionally, language testers may have to dissociate themselves from such services provided that this is not harmful to their test takers. • Language testers shall be prepared to interpret and disseminate relevant scientific information and established professional opinions to society. In so doing, language testers shall clarify their status as either spokespersons for a recognised professional body or not. If the views expressed are contrary to those generally held, they shall so indicate. • It is reasonable for language testers to make scientifically substantiated contributions to public debate on sensitive socio-political issues, such as race, disadvantage and child rearing. • Language testers shall differentiate between their role as educators based on professional knowledge and their role as citizens. • In fulfilling their responsibilities under this principle, language testers shall take care to avoid self-promotion and the denigration of colleagues. • Language testers shall make clear that they do not claim (and are not seen to claim) that they alone possess all the relevant knowledge. Principle 8 Language testers shall be mindful of their obligations to the society within which they work, while recognising that those obligations may on occasion conflict with their responsibilities to their test takers and to other stakeholders. Annotation • When test results are obtained on behalf of institutions (government departments, professional bodies, universities, schools, companies) language testers have an obligation to report those results accurately, however unwelcome they may be to the test takers and other stakeholders (families, prospective employers etc). • As members of the society in which they work, language testers should recognise their obligation to the testing requirements of that society, even when they may not themselves agree with them. Where their disagreement is of sufficient strength to qualify as a conscientious objection, they should have the right to withdraw their professional services.



130

Elana Shohamy

Principle 9 Language testers shall regularly consider the potential effects, both short and long term on all stakeholders of their projects, reserving the right to withhold their professional services on the grounds of conscience. Annotation • As professionals, language testers have the responsibility to evaluate the ethical consequences of the projects submitted to them. While they cannot consider all possible eventualities, they should engage in a thorough evaluation of the likely consequences and, where those consequences are in their view professionally unacceptable, withdraw their services. In such cases, they should as a matter of course consult with fellow language testers to determine how far their view is shared, always reserving the right, where their colleagues take a different view, to make an individual stand on the grounds of conscience.

References Alderson, J. C., Clapham, C. and Wall, D. 1995. Language Test Construction and Evaluation. Cambridge: Cambridge University Press. Bourdieu, P. 1991. Language and Symbolic Power. Cambridge, MA: Harvard University Press. Bourdieu, P. 1996. On Television. New York: The New Press. Davies, A. 1997. “Demands of being professional in language testing”. Language Testing 14 (3): 328–339. Foucault, M. 1991. Remarks on Marx: Conversations with Duccio Trombadori. (Translated by R. James Goldstein and James Cascaito). Brooklyn: New York. Hamp-Lyons, L. 1997. “Washback, impact and validity: Ethical concerns”. Language Testing 14 (3): 295–303. House, E. 1998. Schools for Sale. New York, NY: Teacher College Press. International Language Testing Association. 1999. Code of Practice (draft). Madaus, G. 1990. Testing as a Social Technology: The Inaugural Annual Boise Lecture on Education and Public Policy. Boston College, MA. Nevo, D. and Shohamy, E. 1986. “Evaluation standards for the assessment of alternative testing methods: An application”. Studies in Educational Evaluation 12 (2): 149–158. Shohamy, E. 1998. “Critical language testing and beyond”. Studies in Educational Evaluation 24 (4): 331–345. Shohamy, E. (in preparation). The power of tests and its uses: Critical perspective of the case of language tests. Spolsky, B. 1995. Measured Words: The Development of Objective Language Testing. Oxford: Oxford University Press. Spolsky, B. 1997. “The ethics of gatekeeping tests: What have we learned in a hundred years?” Language Testing 14 (3): 242–247. Spolsky, B. 1998. “What is the user’s perspective in language testing? Paper presented at the Colloquium (15. June 1998). The State of the Art in Language Testing. The Users’ Perspective”. Washington DC: National Foreign Language Center, The Johns Hopkins University.



Part III

Multilingualism, minority languages and language planning



Multilingualism and diglossia in Palestine in the first century of the common era Some thoughts on historical sociolinguistics* Christina Bratt Paulston University of Pittsburgh

Palestine, the crossroads of mighty empires in Mesopotamia and Egypt, has long been of interest to scholars in history and religion and in recent times, i.e., the last century, to linguists. The traffic of these crossroads, of this multitude of ethnic groups in contact for purposes of administration, trade, conquest, and worship, resulted in a veritable profusion of multilingualism. Palestine was also the birthplace of two of the world’s great religions. Exegesis of the Torah and exegesis of the New Testament have the same purpose in their linguistic analysis; a better understanding of the word of God. But the Old Testament and the New Testament are also studied for their historical significance — they are some of our very few literary sources from this time. My own interest lies in accounting for the language shift of the Jewish people from Hebrew to Aramaic, which to this date has found no satisfactory explanation why God’s chosen people shifted from their sacred Hebrew to Aramaic, the common lingua franca of the Fertile Crescent for centuries. As Paper puts it “there still remains the conundrum inherent in the spread of both Aramaic language and writing … ” (1982: 116). Whatever our questions, a better understanding of the perplexing situation of the languages in contact and their use in first century Palestine clearly will give us a better insight into the lives of the people who used those languages, what their values and priorities were, and where their loyalties lay. Palestine of the first century was a site of many languages but also of various language shifts between the ethnic groups, a condition which enormously confuses the situation. Not that language shift was a new phenomenon in Palestine; the Philistines, who gave their name to Palestine (the land of the Philistines, from Hebrew, via Greek), left no record of their language and

134

Christina Bratt Paulston

shifted — probably — to Canaanite, then to Aramaic and later to Koiné Greek. The Jews can be said to have begun their shift to Aramaic with the Babylonian exile 537 BCE, and the Egyptian Diaspora.1 The Idumeans shifted from Arabic to Hebrew/Aramaic; King Herod was by lineage all Arab although raised as a Jew. The final shift of the region came later, with Islam and the shift to Arabic. The real difficulty in attempting to account for the language situation in first century Palestine lies in the paucity of data. Working with No Data2 necessitates very robust concepts, models, and theories. By robust I mean, like Webster’s 3rd, “powerfully constructed”, i.e., concepts and models which hold great power for generalization and are not situation-sensitive but have, where necessary, context as part of the concept construction, as does e.g., diglossia; concepts that are, to borrow another definition, from Rogets’ Thesaurus, “incontestable”. One can of course simply work, as does Rendsburg (1990), “in analogy with other situations” but that is risky business because without data one cannot tell whether other situations are actually analogous. One can also use the few data there are, preferably from text in the original language, with models which allow for a quantitative approach (maybe with statistical analysis for significance). Quantification is not proof, nor do correlations imply causality, but they do give us a powerful summary of dispersed and sparse data as Porter’s charts make clear. Porter (2000) uses Hallidayan register features for his theoretical framework, but his chapter holds as much, if not more, significance as a model for possible alternative analytical procedures as do the actual findings of his register analysis (see also Watt 1997). Another promising approach is that of historical sociolinguistics. Bernard Spolsky points out that the Jewish languages provide an excellent field for the study of historical sociolinguistics, which he defines as “the application of sociolinguistic principles and models to questions of language use at earlier times” (1985: 35). He goes on to demonstrate the contribution of theory to the solution of a problem of historical language use, examining the way the notion of diglossia and Jackendoff’s preference linguistics (1983) cast “light on the available contemporary evidence and reveal the pattern of language use among Jews of Palestine in the first century of the Common Era” (Spolsky 1985: 35). Historical sociolinguistics then becomes a search for the most appropriate models and key concepts to use in the explanation and even basic description of historical situations. With present day scholarly work in sociolinguistics, we seek to reach an adequate explanation so that we can predict the coming situation. As an example, in 1982 the Swedish National Board of Education asked me to do a review and analysis of Swedish research on bilingual education

Multilingualism in Palestine in the first century

as the Board was responsible for the language instruction of immigrants: “Sweden is an immigration country. Within a few decades, Swedish society has been transformed from ethnic homogeneity and virtual monolingualism to heterogeneity and a plurality of languages”.3 In other words, the National Board of Education needed to predict the future language situation for the purpose of efficient language-in-education planning. Given certain social factors, among which the most important were the immigrants’ willing shift to Swedish (much like the United States a hundred years ago) and the children’s desire to assimilate, the prediction that mother tongue instruction was laudable but only indispensable for linguistic minority groups with a record of back-migration, was fairly obvious (if at the time very controversial). Time has proved that report accurate. But with historical sociolinguistics, we begin with the consequence or results; we already know what happened, and instead we are looking for description and explanation. We know that the Galilean Jew Jesus spoke Aramaic, and we know that 500 years earlier the Jews spoke Hebrew. So, says Spolsky, “we must assume a situation in which the status attached to being a Hebrew speaker was less than that attached to speaking the other language” (1985: 48). One difficulty in selecting assumptions and speculations based on our present day theories and models of multilingualism is that those theories are almost all based on present day situations, such as the modern nation-state. (Sociolinguistics itself is less than a half-century old). The notion of Palestine itself is easily misleading. It was neither a country nor an empire but a geographic area, consisting of petty kingdoms and citystates, variously independent, colonized, or occupied with in-between stages like client-kingdoms; with steadily shifting borders due to conquests and reconquests, much like eastern Europe. Most of our models for multilingualism refer to ethnic groups within a modern nation-state.4 We do not really know to what degree such generalizations will hold for an area like Palestine, and we need to specify whether we are talking about multilingualism in Judea, Galilee, Caesarea, Scythopolis or Elephantine as the situations all differed with regard to languages and language use. Comparing the Palestinian situation to historical empires where we do have adequate data about language use, such as the Hapsburg, the Romanov, or the Ottoman Empires and where we can test the strengths of modern day generalizations, might be a fruitful approach. We know, for instance, that language shifts in multilingual nations takes place only if there are opportunities and incentives to learn the national language. Such opportunities of access include

135

136

Christina Bratt Paulston

participation in social institutions like universal schooling, exogamy, military service and religious institutions; access to mass media such as television; demographic factors such as urbanization, and so on. The delayed urbanization and industrialization in Eastern Europe contributed to ethnic group and language maintenance there (Paulston 1998), and the flip side also holds true for the Palestine area: urbanization, like Caesarea resulted in the spread of Greek, whether in bilingualism or in shift. We know that negative sanctions for language use (such as forbidding or urging against the use of a particular language) do not seem to be very effective. It did not work in the Hapsburg Empire, and it did not work in Judea where the rabbis’ strictures against Greek were not effective. Many modern conditions do not apply, like mass media or universal schooling but some of the same principles can be extracted, inherent in those situations: Jesus’ knowledge of Biblical Hebrew is probably tied up with reading (and learning to read) the Torah in the Temple (Luke 4: 16). The Greek city-states may not have had television, but they provided public entertainment in Greek and so provided access to willing listeners. Many of the other social factors were as effective then as they are today. Exogamy was most certainly a major factor in the Jewish shift to Aramaic as were forced population movements (cf. Stalin’s deportation of the Tartars). That Hebrew is the national language of modern day Israel is directly due to its role as a sacred language and its function in religious worship. My point here is that this kind of comparison with similar historical situations where we do have data will allow us to test the robustness of our theoretical models and so allow us to speculate with more assurance about language situations where there is a paucity of data. These then are some background thoughts that bring us to the topic of language repertoire and diglossia in first century Palestine. What can we hope to achieve by using the notion of diglossia? Watt (2000) suggests using the concept of diglossia “for the purpose of describing the first century Palestinian situation” and continues some pages later with the (rhetorical) question whether some order can be imposed on the “chaos” of the biblical literature about these language issues. While in principle using the theoretical model of diglossia to describe a historical situation may be well worth the effort, I have my doubts about imposing order on chaos. One of the side effects of working with no or little data is that it becomes very difficult to refute scholars with whom one disagrees: the strength of the argument lies within the argument, not with the facts. The more important then the clarity of the theoretical concepts.

Multilingualism in Palestine in the first century

Charles Ferguson himself was very clear about his intentions with his diglossia model. He used four “defining languages” to examine carefully a language situation “where two varieties of a language exist side by side throughout the community, with each having a definite role to play” (1971: 1). Later on he elaborated on his intentions: I wanted to characterize a particular kind of language situation, taking a clear case that was relatively easy and uncontroversial to characterize. However, the idea of doing that was to make the clear case just one slot in a taxonomy of some sort … Ultimately the taxonomy would be replaced by some set of principles or frame of reference in terms of which this kind of thinking about language and this kind of research should be done. My goals, in ascending order were: clear case, taxonomy, principle, theory. (Ferguson 1991: 215).

Ferguson clearly had theory building in mind, and he was interested in language situations “which seemed to have implications for the writing of grammars and for theories of language and language change” (1991: 216–217). Ferguson expressed the hope that other people would write articles about other cases, and so they have. The writing of grammars took another turn with the Chomskyan revolution in linguistics, while historical linguistics have turned to Fergusonian thoughts and sociolinguistics for elucidation in accounting for language change.5 Ferguson continues to discuss Switzerland as example of “four different language situations in a nation” (1991: 215) and it is perfectly clear that he recognized bi-/multilingualism as co-existing with diglossia as well as with dialectic diversity at the same time. If we apply this description to the various (semi-)autonomous units in Palestine, it is obvious that Galilee’s language situations was not the same as Judea’s, nor like Caesarea and the Decapolis, and that Jerusalem (much like Brussels) was an exception to all. There is obviously no need to extend Ferguson’s clear case of Arabic diglossia to all these language situations (nor to the Jewish Diaspora); the result is conceptual confusion. Watt (2000) comments several times that one cannot “copyright” a theoretical concept like diglossia (Ferguson himself had after all borrowed the term),6 one more comment here from Ferguson: “ … diglossia when it is used by a French linguist nowadays always implies the oppression of some lower classes by upper classes and I never even thought of that when I was writing about diglossia” (1998: 83). Oppression of an ethnic minority rather than of a social class may be more accurate because most French writers have used Fishman’s definition and have applied diglossia to bi/multilingualism.

137

138

Christina Bratt Paulston

On a ainsi l’impression que le succès du concept de diglossie s’explique par le moment historique où il fut lancé. A l’époque des indépendances africaines, de nombreux pays étaient confrontés à une situation linguistique complexe: plurilinguisme d’une part et prédominance officielle de la langue coloniale d’autre part. Donnant un cadre théorique à cette situation, la diglossie tendait à la présenter comme normale, stable, à gommer le conflit linguistique dont elle témoignait, à justifier en quelque sorte qu’on n’y change rien (ce qui fut d’ailleurs le cas dans la plupart des pays décolonisés). (Calvet 1993: 45)7

Slight wonder that Ferguson had not thought of this application of diglossia because he was trying to describe the language situation of one ethnic group, be it Arabic, Greek or Swiss German. He certainly did not have in mind the “tendance à gommer les conflits qui caracterisent les situations de diglossie et à presenter comme normale une situation de domination” (Calvet 1993: 45),8 as Calvet quite inaccurately put it, but his statement illustrates two important points: (1) extending diglossia to include bilingualism merely contributes to confusion and dilutes the concept, and more importantly (2) different scholars will differently interpret what they see as the most salient feature of diglossia. For the French it was the conflict situation inherent in the contact situation between ethnic minorities and the dominant group in ex-colonial settings and later extended to any kind of group conflict involving languages. Joshua Fishman saw functional complementary distribution of the language varieties as the most salient feature of diglossia and extended it to bilingual situations. This is not a new insight in multilingual settings. Charles V is reputed to have claimed that he spoke German to the horses, Italian with the ladies, French with the men, but Spanish to God; in other words, he chose his language according to the addressee. Maybe more to the point here is the frequently cited saying from the Jerusalem Talmud: “Rabbi Jonathan of Bet Gubrin said, ‘Four languages are of value: Greek for song, Latin for war, Aramaic for dirges, and Hebrew for speaking’”.9 The choice here is according to genre or purpose. Ferguson divided his language varieties according to what he called “situation” but these are reminiscent of Rabbi Jonathan’s genre division, such as sermon in church, speech in parliament, conversation with family, etc., but he also includes addressee, instruction to servants and workmen (who do not know the H form). The major difference between Rabbi Jonathan and Ferguson is that the former writes about different languages echoing Ancient Greece’s distribution of dialects such as Ionic for history, Doric for choral and lyric words, and Attic for tragedy (Haugen 1972), while Ferguson’s break with tradition in writings on diglossia lay in that he was

Multilingualism in Palestine in the first century 139

attempting a description of the speech community, not linguistic descriptions as had earlier writers. It is quite clear that the notion of functional complementary distribution can include separate languages (and language families) as well as varieties of the same language. It is also quite clear that all bi-/multilingual situations do not necessarily involve functional distribution. Where languages are legally protected by territorial linguistic rights, as in Switzerland and Belgium, those languages co-exist with no functional distribution. When languages co-exist within a nation-state without the protection of language rights and without functional distribution in a super-subordinate relationship, the norm is shift to the dominant language although the rate of that shift may vary as with Gaelic in Scotland over hundreds of years compared to Swedish in Minnesota over a three generation shift. But shift is normal, given the absence of internally or externally imposed boundary maintenance (Paulston 1994: 20–21). The Jewish dietary and marital regulations with insistence on endogamy reinforced the group’s religious boundary maintenance — an “us against them” consciousness, but for the Jews language did not seem to have been an absolute for group membership; the translation of the Torah into Greek for the Alexandrian Jews is evidence of that. What we need at this point is a very clear set of guidelines for selecting robust theoretical models and frameworks for use in historical sociolinguistics, and obviously we have not yet arrived at that point. Let me here very briefly outline some thoughts in considering whether diglossia is a useful construct in describing the language situation in first century Palestine. As I hope to have made clear, Palestine is a misleading notion to begin with and should be reconsidered in this context. We do know that the region we call Palestine was bi-/multilingual so we need to consider what sub-concepts are implied or entailed in using a concept like diglossia in contrast to bilingualism. In diglossia the Low (L) form is not standardized and often not written while in bilingualism the L form (read subordinate language) can be standardized and written just as well as the High (H) form, as e.g., Spanish and English in the US Southwest. In a putative Hebrew-Aramaic diglossia, Aramaic had been written for centuries and presumably owed its spread and what prestige it had partially to its written nature. Diglossia is not helpful to explain this feature, but it is useful for the shift to Aramaic. Diglossia is characterized by a stable language situation with an increasingly larger split between the written and spoken word and this situation we find in the Biblical Hebrew and Aramaic.

140

Christina Bratt Paulston

When we do get shift in a diglossic situation, the shift is to the L form, here Aramaic, while in bilingualism the shift typically is to the H form, the dominant language (as Spolsky suggests with the Jewish shift to Greek). In diglossia, there are no native speakers of the H form while there are in bilingualism. Having no native speakers of the H form entails great pressure on book learning and education, especially in the form of economic resources and leisure time for study, a situation which very much favors the elite classes and ultimately serves to legitimate their status (Horsley 1995). Jewish rabbinical studies neatly fits this characterization but interestingly enough, it also extends to the study of Greek within the Jewish community. Ferguson’s notion of diglossia specified dialects of the same language or very closely related varieties,10 which in turn entailed that both the H and the L form were spoken by members of the same ethnic group, and for present day conditions with frequent strife and competition between groups along ethnic lines, this is of crucial importance in diglossic versus bilingual situations. But ethnicity as we think of it did not really exist in first century Judea and Galilee. Israel was a theocracy as much as an ethnic nation (nationalism in the modern sense did not exist either; political scientists typically date the birth of nationalism to the time of the French Revolution), and there were monolingual Greekspeaking Jews. King Herod, in fact, was ethnically Arab, but this was not of major concern because religiously he was Jewish. So this highly salient feature in contrasting diglossia with bilingualism explains very little in our first century situation. Finally, as discussed above, one does not need the concept of diglossia to deal with complementary distribution, which also occurs with bilingualism. In other words, I find very little explanatory power in the concept of diglossia applied to first century Palestine and considerable — and unnecessary — conceptual confusion or in Watt’s terms the chaos which follows from these implications. On the other hand, I find Watt’s suggestion of a diglossic continuum quite intriguing although at this time I would avoid the term diglossic and simply call it a language continuum, calqued on the notion of a Creole continuum. In a Creole continuum, the opposing poles of the continuum are simply (H) acrolect and (L) basilect and there is no reason I can see that this notion could not be extended to plural acrolects (Greek, Latin, Biblical Hebrew) and basilects (Aramaic, Arabic, unwritten languages) with mesolects sketched in (maybe Mishnaic Hebrew, “educated” Aramaic, etc.) where the conceptual weight of the continuum is placed on the individual speaker’s repertoire, similarly but not identically to a Creole continuum where the proficiency includes a continuum of the same language variety but here would include



Multilingualism in Palestine in the first century

different language varieties. The locus, however, remains the individual speaker’s proficiency. I am here obviously only playing with ideas of how best to characterize and explain the various language situations in first century Palestine, but I do think that this is the direction that it would be helpful for us to take in future studies: the development of solid theoretical models and concepts for use in a historical sociolinguistics which will allow us to speculate with more accuracy and in so doing maybe discovering data we had not thought of as significant before. The use of quantificational analysis of text within a theoretical framework may allow us a way to test such new speculations and generalizations, be they linguistic or social. I can only hope that such a direction would give impetus to a more general development of historical sociolinguistics as a recognized field of study with its own methodology.

Notes * This paper began life as a discussion paper of the Biblical Greek Language and Linguistics Section “Language Repertoire and Diglossia in First Century Palestine” of the Society of Biblical Literature Annual Meeting, Orlando, Florida in November 1998. That version of the paper is published in Stanley E. Porter. (ed.). 2000. Diglossia in the New Testament and Other Linguistic Essays. Journal for the Study of the New Testament Supplement Series. [Studies in New Testament Greek Series]. Sheffield: Sheffield Academic Press. 1. There is no agreement in the scholarly literature on when the Jews shifted to Aramaic. The existence of Aramaic as a lingua franca from Egypt to Babylon is generally acknowledged as well as an Aramaic speaking Jewry; the disagreement rather concerns the duration of Hebrew-Aramaic bilingualism and the demise of Hebrew as a common, spoken language. Porter (2000) argues for a shift, complete by the first century; for another view see Bernard Spolsky (1991: 215). I expect both are partially right; in all likelihood, different groups of Jews shifted at various rates, depending on region and social class. As long as Biblical Hebrew and Mishnaic Hebrew were prestige languages, the upper strata were slower to shift than the lower; Judea slower than Galilee. Once Greek had become the prestige language, the upper classes became bilingual in Greek-Aramaic or shifted completely to Greek at a quicker rate than the lower strata. 2. The title of a Festschrift for Professor Lambdin of Harvard University: Golumb, David M. (ed.). 1987. Working with No Data: Semitic and Egyptian Studies Presented to Thomas O. Lambdin. Winona Lake: Eisenbraus. 3. From Inger Marklund, Head of Research Division, National Swedish Board of Education, introduction to Christina Bratt Paulston. 1982. Swedish Research and Debate about Bilingualism. Stockholm: National Board of Education.

141

142

Christina Bratt Paulston

4. See Christina Bratt Paulston. 1994. Linguistic Minorities in Multilingual Settings. Amsterdam: Benjamins, and its references. 5. Unfortunately, this approach of socio-historical linguistics is often misnamed historical sociolinguistics, which it is not, as in e.g., T.W. Machan and Charles T. Scott (eds). 1982. English in its Social Contexts: Essays in Historical Sociolinguistics. Oxford: Oxford University Press, or James Milroy. 1992. Linguistic Variation and Change. On the Historical Sociolinguistics of English. Oxford: Blackwell. Suzanne Romaine discusses under the heading of “Historical Sociolinguistics: Problems and Methodology” in Ulrich Ammon et al. (eds). 1988. Sociolinguistics: An International Handbook of the Science of Language and Society. Berlin: Walter de Gruyter, the goals, methods and data of what she quite accurately refers to as socio-historical linguistics, accounting for language change. 6. D. Sotiropoulos. 1977. “Diglossia and the National Language Question in modern Greece”. Linguistics 19: 5–31, mentions that diglossia was first used by Karl Krumbacher in 1902 in his book Das Problem der Modernen Griechischen Schriftsprache. Other references are to the coinage of the term by the French linguist W. Marcais. 1930. “La diglossie arabe”. L’enseignement public 97. 7. Louis-Jean Calvet. 1993. La Sociolinguistique. Paris: Presses Universitaires de France, p. 45. “One has thus the impression that the success of the concept of diglossia can be explained by the historical moment in which it was introduced. At the time of the African independencies, numerous countries were confronted with a complex linguistic situation: on the one hand, multilingualism, and on the other, the official predominance of the colonial language. Providing a theoretical framework for this situation, diglossia tended to present this situation as normal, stable and to erase the linguistic conflict to which it bore witness, to justify in some way that nothing changed (which was after all the case in most of the decolonized countries)”. [My translation]. 8. “tendency to erase the conflicts which characterized diglossic situations and to present as normal a situation of domination”. [My translation]. 9. This version of the translation is taken from Bernard Spolsky and Robert L. Cooper. 1991. The Languages of Jerusalem. Oxford: Clarendon Press, p. 24. 10. Language and dialect are notoriously difficult to define linguistically; see David Crystal. 1987. The Cambridge Encyclopedia of Language. Cambridge: Cambridge University Press.

References Calvet, L.-J. 1993. La Sociolinguistique. Paris: Presses Universitaires de France. Ferguson, C. 1971. “Diglossia”. In Language Structure and Language Use, 5–15. Stanford: Stanford University Press. Ferguson, C. 1991. “Diglossia Revisited”. Studies in Diglossia, Southwest Journal of Linguistics 10 (1): 103–127. Ferguson,C. 1998. “History of Sociolinguistics”. In The Early Days of Sociolinguistics: Memories and Reflections, C. B. Paulston and G. R. Tucker (eds), 79–91. Dallas: SIL.



Multilingualism in Palestine in the first century 143

Haugen, E. 1972. “Dialect, Language, Nation”. In Sociolinguistics, J. Pride and J. Holmes (eds), 79–91. Harmondsworth: Penguin Books. Horsley, R. A. 1995. Galilee: History, Politics, People. Valley Forge, PA: Trinity Press International. Jackendoff, R. 1983. Semantics in Cognition. Cambridge, MA.: MIT Press. Macfarlane, R. T. 1997. “Hebrew, Aramaic, Greek and Latin: Languages of New Testament Judea”. In Masada and the World of the New Testament, J. F. Hall and J. W. Welch (eds), 213–221. Provo, Utah: BYU Studies. Paper, H. 1982. “Language Spread: The Ancient Near Eastern World”. In Language Spread: Studies in Diffusion and Social Change, R. L. Cooper (ed.), 20–31. Bloomington: Indiana University Press. Paulston, C. B. 1994. Linguistic Minorities in Multilingual Settings. Amsterdam: J. Benjamins. Paulston, C. B. 1998. “Linguistic Minorities in Central and Eastern Europe: An Introduction”. In Linguistic Minorities in Central and Eastern Europe, C. B. Paulston and Donald Peckham (eds), 144–171. Clevedon: Multilingual Matters. Porter, S. E. 2000. “The Functional Distribution of Koine Greek in First Century Palestine”. In Diglossia in the New Testament and Other Linguistic Essays, S. E. Porter (ed.), 80–89. Sheffield: Sheffield Academic Press. Rabin, C. 1976. “Hebrew and Aramaic in the First Century”. In The Jewish People in the First Century, Vol. II., S. Safrai and M. Stern (eds), 374–387. Amsterdam: Van Gorcum/ Assen. Rendsburg, G. 1990. Diglossia in Ancient Hebrew. New Haven: American Oriental Society. Spolsky, B. 1983. “Triglossia and Literacy in Jewish Palestine of the First Century”. International Journal of the Sociology of Language 42: 95–109. Spolsky, B. 1985. “Jewish Multilingualism in the First Century: An Essay in Historical Sociolinguistics”. In Readings in the Sociology Of Jewish Languages, Joshua A. Fishman (ed.), 111–131. Leiden: E. J. Brill. Spolsky, B. 1991. “Diglossia in Hebrew in the Late Second Temple Period”. Studies in Diglossia: Southwest Journal of Linguistics 10 (1): 47–54. Spolsky, B. and Cooper, R. L. 1991. The Languages of Jerusalem. Oxford: Clarendon Press. Sotiropoulos, D. 1977. “Diglossia and the National Language Question in Modern Greece”. Linguistics 19: 5–31. Watt, J. M. 1997. Code-Switching in Luke and Acts. Berlin: Peter Lang. Watt, J. M. 2000. “The Current Landscape of Diglossia Studies”. In Diglossia in the New Testament and Other Linguistic Essays, S. E. Porter, (ed.), 90–111. Sheffield: Sheffield Academic Press.



Introducing Khayem Zhitlovski Joshua A. Fishman Yeshiva University, Stanford University

The sociology of Jewish languages has far too few heros who have successfully made the transition to the general sociology of language. Although books have been written about Eliyezer ben Yehuda (Fellman 1973) and about Nosn Birnboym (Fishman 1987), that about the former is largely deconstructive (ben Yehuda wasn’t as much of a factor in the “revival” of Hebrew as is commonly assumed) and that about the latter largely emphasizes other-than-language concerns of the protagonist (in his Zionist and Ultra Orthodox stages Birnboym paid comparatively little attention to either Yiddish or Hebrew). Not a single hero of Jewish languages is as well known as are Ivar Aasen, Vuk Karadzic, Takdir Alisjahbana, A. Korais, Ludovit Stur or other “language activists and language makers” who have come to be recognized in the general sociolinguistic literature. The case of Khayem Zhitlovski1 (1865–1942) is instructive in that from the time he proclaimed his interest in Yiddish, still as an Es-Er (acronym for members of the “Russian Socialist” party) in the 1880’s, he never de-emphasized that interest, reinterpreting it and relating it to all of his other interests (in Jewish socialism, Jewish secularism, Jewish Diaspora cultural autonomism and territorialism/Zionism). Thus, if the Viennese-born and German educated Birnboym is the first Western derived Yiddishist-pluralist-activist, Zhitlovski is every bit his Eastern European counterpart (born in Ushatsi, near Vitebsk, Russian educated). However, unlike Birnboym, Khayem Zhitlovski never strayed in directions that took him away from primary loyalty to and advocacy of Yiddish, although he too explored many other supplementary paths as well.

Socialism Socialism was Zhitlovski’s beginning stage in his life-long pursuit of solutions

146

Joshua A. Fishman

to the problems that modernization brought for the Jewish people in Eastern Europe. It was his fundamental or most basic belief and all of his subsequent specifically Jewish positions went hand in hand with it. This should serve to remind us that language planners (and language promoters, activists, champions and ideologues should all be included within the language planning fold) always have societal goals above and beyond language on their minds and that “their” language is both symbolic and functional with respect to those societal goals). Initially, Zhitlovski’s “socialism with Yiddish” convictions were founded on the postulate that a socialist triumph in any of the countries of the Jewish Diaspora would automatically also bring with it equal cultural and political rights for Jews and an end to their poverty, exploitation and persecution. This was a position significantly to the right of the mainstream of openly assimilationist Eastern European Jewish intellectuals. At that time their cosmopolitan views led them to conduct their efforts in Russian and (through the underground organization Narodniya Volya) to focus upon rendering literate the Russian peasantry so that it could be more successfully influenced by the illegal Socialist press. The conviction that a united proletariat was an absolute desideratum also motivated Khayem Zhitlovski (hereafter: Kh Zh) to initially follow this linguistically more parsimonious path. It is clear, however, that this view was accompanied in his mind by the conviction — expressed in his Russian writings — that Jews should never be completely swallowed up within the huge “all Russian” proletariat. It is equally clear that he was among the first modernized Eastern European Jewish socialists to come to the realization that Russian literacy was no way in which to successfully bring the overwhelmingly nonRussian-speaking Jewish masses to socialism. In this respect, Kh Zh was also to the right of Moses Hess (1812–1875). The latter never saw fit to champion any Jewish language in his even earlier efforts to move German Jews (then still overwhelmingly Yiddish-speaking) to an amalgam of socialism and Zionism. In the German context, Yiddish could be viewed by Jewish intellectuals as simply “bad German” (and was so viewed even initially by Birnboym, two generations later); in multiethnic Eastern Europe, however, Kh Zh fostered the recognition that Yiddish was the vernacular of the bulk of the “Jewish nation” and, therefore, the only proper tool for advancing socialism among the members of that nation, regardless of the degree of its “abstand” from German. He was later to repeat this argument re the Byelorussian language (vis-a-vis Russian), as the proper tool for spreading socialism among Byelorussian workers. The above realization that Jewish poverty could only be combatted by achieving Jewish literacy in Yiddish, then the declared mother-tongue of 97

Introducing Khayem Zhitlovski 147

percent of all Jews in the Czarist Empire (see the tables of the Czarist census of 1897 reproduced in Fishman 1991: 392), was first activated when Kh Zh spearheaded the turn of the century struggle for instituting Yiddish as the medium of instruction in the “new schools” (“new” because they offered both secular and traditional Judaic studies) that were then being established, with Czarist government approval, by the Jewish “Society to Spread Enlightenment” (1863–1918). Unlike Birnboym, who was then also slowly turning ever more toward Yiddish — which he had until then only passively understood — as the language of the “Jewish nation” within the Austro-Hungarian Monarchy, Zhitlovski was, quite naturally, already fully fluent in Yiddish and steeped in its budding modern literature. Nevertheless, in his pre-1880-focused retrospective “When and How I became a Yiddishist” (1914), as well as in the original version of his “Why particularly Yiddish?” (1897/1900)2 Kh Zh refers only to socialismrelated rationales for preferring Yiddish. Although Kh Zh came to consider himself a Yiddishist, even before the Czarist pogroms of 1881, as is evident from his account, he had not yet arrived at a stage of Yiddishism in which Yiddish was considered as worthy of cultivation in order to further both a distinctly Jewish identity and specifically Jewish functional and symbolic goals, rather than general socialist ones.

Yiddishism Regarding Yiddish as an ideological cause separate from but allied to socialism was a view Kh Zh began to espouse and ultimately to champion only after allying himself with the cause of the above-mentioned “school struggle”. Moses Hess’ work (Rome and Jerusalem 1862), combining socialism and Zionism, was apparently not yet known to him when, in his own Russian context, Kh Zh became one of a very small and still young and untested group of intellectuals who realized that Yiddish was an invaluable tool for fostering Jewish identity, modernization and mobilization. In this respect the early Kh Zh was not unlike the much earlier proponents of English, French, German and other major and minor modern Western European vernaculars who also first regarded these vernaculars only as means of communication for reaching other, desired material and intellectual goals, rather than being as symbolically essential for their various national identities, cultures and histories. Zhitlovski’s first major step in the latter direction occurred with the publication of his famous Russian brochure, “Yevrey kiyevreyom” (A Jew to Jews).

148

Joshua A. Fishman

It is dated 1890/91, but was not released until 1892, and was intended for Russian and Jewish socialist intellectuals. In this brochure (published in Yiddish: 1917), Jews were proclaimed to be an independent nation (“We are not 4% of some other nation but 100% of ourselves!”). Socialism and nationalism were considered not to be opposites but complementary to each other, and the existence of a Jewish working class unmistakably implied the existence of that class’ special problems and the need for its own socialist movement. Western European Jews and Eastern European Jews were viewed by Kh Zh as being fundamentally different; Nosn Birnboym also held this view and distinguished between the two on the dimension of authenticity (Western Jews were mostly “minus” authenticity and Eastern Jews, mostly “plus”). Zhitlovski’s distinction was based on the related dimension of peoplehood. On that dimension, the Western Jews had demanded only individual rights in the course of their emancipation, whereas the Eastern Jews required group rights as well. Jews needed to participate in the (larger Russian and worldwide) Socialist revolutionary movement as members of the Jewish nation and in the Socialist movement as members of the Jewish working class. Jews needed to be productivized (by becoming workers in industry and farmers on the soil) and modernized (via secular educations), and Jewish intellectuals needed to lead them in directions that corresponded to these needs. National (rather than only political) equality for Jews required recognition, utilization and cultivation of their mother tongue, Yiddish. The latter view was a revolutionary one, particularly within its socialist context, and it is one that Kh Zh developed with increasing clarity and singularity both in Eastern Europe and in the United States. Zhitlovski’s Yiddishism had the effect of moving him away from extending a welcome to Zionism. A day after the First Zionist Congress, Basel, 1897, Zhitlovski declared Zionism to be bourgeois and, at best, attuned to the needs of Westernized Jews, a view which he then shared with the post-Zionist Birnboym (1902). Yiddishism also drew Kh Zh closer to the newly organized Jewish Workers Bund (also founded in 1897), although the latter initially found Zhitlovski to be too much of a Jewish nationalist for its then pan-Socialist tastes (later revised [1905] in the direction of vociferously advocating the national needs and rights of Jewish workers). In the very same year as the Basel Congress, Kh Zh also published his famous essay “Zionism or Socialism” in a Bundist journal, expressing for the first time in Yiddish the views which he propounded for the rest of his life, namely, the right of Jews to demand national recognition and to foster their own national renaissance in the diaspora. He

Introducing Khayem Zhitlovski 149

deplored the fact that Jewish writers and thinkers had not yet sufficiently worked in Yiddish, claiming that it was: dear and holy to us (Jews), just as is German to Germans, Russian to Russians, and as is his Holy Tongue to the old-fashioned Jew … As soon as languageartists begin to warm the language, pour it into new vessels, create three new words out of each old one, learn how to hit the bull’s-eye of folksy expression and folksy good taste, then Yiddish will become rich and soft and flowing, suitable for expressing whatever one wishes … Then we Jews will have our own secular schools, our own intelligently modernized religious schools, our high schools for humanities and for sciences, and even our own Yiddish universities (1904: 51 [my translation]).

At this point Yiddish-and-Yiddishism became an independent principle, a drive, in Zhitlovski’s thinking. Whereas Yiddish had previously been regarded only as a useful tool for spreading socialism among Jewish workers, it now became something “dear and holy”, and would, he believed, become even more so as Jewish poets, writers, journalists and essayists expressed their talents in this medium, making it the most accessible ethno-national medium of a modern and mobilized Jewish people living among other modern peoples in the world at large.

Territorialism Just as the Czarist-inspired pogroms of the early 1880’s had led Kh Zh from Russian to Jewish socialism and, ultimately, to Yiddishism, so the infamous Kishinev pogrom of 1903 prompted him to recognize “the importance of a land of our own”, without in any way withdrawing from either his Jewish national or socialist revolutionary goals in the Diaspora as a whole. Initially, his territorialism was clearly non-Zionist (i.e., not Land of Israel) in focus. However, at that very time, the Herzlian Zionist Congress too was considering Jewish settlement in Uganda, either as a “way-station” or as a substitute for Zion. Such straying from the historical and hallowed homeland was no more than a brief episode for Herzl (who was, at any rate, outvoted by the Congress in this particular very controversial connection). In Zhitlovski’s case, Zionism in its Labour-Zionist guise ultimately became a specific instance of his own broader territorialist search and support for planned Jewish agricultural concentration and autonomous cultural development. Unlike the main body of Zionism, which appeared to Zhitlovski to be opposed to Jewish activism on behalf of

150

Joshua A. Fishman

socialist revolutionary efforts and diaspora Yiddish cultural autonomy efforts, territorialism appeared to him to be a furtherance of both. Arriving in the USA in 1904, Zhitlovski added these new territorial emphases to his mesmerizing talks and essays on socialism, Jewish nationalism and Yiddishism, proclaiming that the true realization of American democracy lay not in the assimilationist melting pot ideal pursued by many capitalists and socialists alike, but in a future “United nations of the United States”. In this respect he both anticipated and was more concrete in his views than either Horace Kallen (1924) or most other noteworthy American structural pluralists (viz., Gunier 1994; Walzer 1983, 1992), whether then or now. It was in America (indeed, in a Jewish immigrant neighborhood of the Bronx), in 1907, that Birnboym (on his first tour of the New World) arranged to meet with Kh Zh and interested him in becoming a co-signatory (together with Doved Pinski, a Yiddish literary giant, and two now relatively unknown Yiddish theater personalities) to a proclamation convening the famous “First World Conference for the Yiddish Language” (at Tshernovits, in the AustroHungarian Monarchy but close to the Ukrainian border of the Czarist Empire) in 1908. Indeed, Kh Zh became the virtual Chairman of that Conference and one of its keynote speakers, his elegantly cultivated Yiddish eliciting admiration for being “more beautiful than French” (for my comments on his subsequently [1928] moderately critical evaluation of the Conference, see Fishman 1998; his initially more positive comments, published in 1909, are reproduced in v. 4 of his Collected Writings [1912]). He continued for many years to lecture and publish widely, both in the Old World and in the New, providing the philosophical foundations for “a Jewish thinking and a Yiddish writing intelligentsia”, particularly in non-religious and non-Zionist circles. He belatedly made two attempts to establish some links to Zionism, one successful and one unsuccessful. When Labour Zionism arose (Doved Pinski being among its American founders), Kh Zh’s emphases on the importance of the diaspora influenced some of its leaders in the USA to found the first Yiddish secular schools, called “National Radical Schools” (1910), to stress Zhitlovski’s view that these two characterizations were in no way contradictory. These schools evolved into the philosophically pluralistic Jewish National Workers Alliance (“Farband”) schools that soon strove to be even more pluralistic that their initiator by stressing Hebrew, Yiddish, socialism, Zionism and Jewish traditional holidays and observances as well. On the other hand, on Kh Zh’s first visit to Palestine (1914), his views and his reputation elicited so much opposition in anti-Yiddishist circles that these circles made it impossible, by means of their

Introducing Khayem Zhitlovski

disruptive behavior, for him to give any public lectures in Yiddish or make any public appearances at all. He quickly left the country, never to return (not even after joining the American Labour Zionist movement, c. 1915).

Summary and discussion Zhitlovski and Birnboym overlap and are similar in many respects. They are both pluralistic in their views with respect to the desired directions of Eastern European Jewish modernization and the centrality of Yiddish in that connection. They ultimately differ profoundly, however, with respect to left-right emphases and secular-religious preferences, Zhitlovski being consistently partisan to socialist secularism, and Birnboym finally ending up close to religious fundamentalism. They both reveal the ease with which languageideology labels can be ascribed, but also some of the fundamental simplifications in many such ascriptions, particularly among pluralistic thinkers and leaders. Birnboym was a champion of Yiddish primarily in his secularistautonomist stage, after very slowly becoming such in his prior Zionist stage and equally slowly leaving off being such in his final ultra-Orthodox stage when his premonitions of an impending Holocaust pushed all other issues to the sidelines. Zhitlovski is far less a “stagewise” personality than Birnboym. His pluralism is additive rather than substitutive in nature. Socialism is always present and so is secularism. Yiddish emphases represent the first added stage and are particularly intriguing in Kh Zh’s thought since they caused him to revise his earlier view of socialism (thereby changing him from a secular [Russian] Socialist Revolutionary to a secular Jewish Socialist-Nationalist). Territorialism (initially anti-Zionist but subsequently including a Labour Zionist facet too) was his final add-on. It was a sufficiently significant addition to cause him to revise his former views of the advisability or feasibility of cultural autonomy arrangements for Jews in the absence of territorially concentrated self-management. Yiddish too needed its own “place” he came to believe, so that it could become deeply rooted during a time of increasingly profound social and cultural change (1928). At the beginning of the 20th century, Zhitlovski became the darling of a younger generation of Yiddishists and autonomists in both Eastern Europe and the USA, although he was then already more than a generation older than they were. He had the capacity to continually reinvent and reintegrate himself. He was frequently controversial and remained so to the very end. Even more so

151

152

Joshua A. Fishman

than Birnboym, he reminds us of how impossible it may be to pigeon-hole an entire language (particularly, a “language of the masses”) as pertaining to only one political ideology or transcendental philosophy. Certainly Yiddish cannot be so pigeon-holed, as Kh Zh would doubtlessly agree today, were he to note that the leading-edge of efforts on behalf of Yiddish in daily life are now precisely among ultra-Orthodox khareydim in Israel. Overlooking Zhitlovski merely because he carried to a logical extreme the widespread secular view that Jews were “just like any other people with respect to the possibility of separating religion from nationality” (therefore allowing for the “logical possibility” of Protestant Italians and Catholic Jews), or because he initially misinterpreted Nazi anti-Semitism as an over-reaction to Jewish middle class exploitation of German workers, or because he rose to champion the Soviet cause precisely after the Hitler-Stalin pact burst upon a startled world (the USSR being the only country in which both a version of socialism and of Jewish territorial autonomy [Birobidzhan, on the Amur River near the Korean border in eastern Siberia] had been implemented), would be to judge his lifelong impact by his very least influential views. Few, if any of his early adherents followed him in these final, desperate directions, yet his earlier impact on them continued to be substantial indeed. New languages of late modernization that aspire to High functions, always arise in contexts where the ongoing discharge of these functions is already established and is defended by those who already hold power, be they rightwing, centrist or left-wing. Zhitlovski fearlessly fought the opponents of the “language of the people”, whether they were right-wing (in his view: Hebraists, Jewish clergy and Zionists) or left-wing (other socialists, Jewish or gentile). Adopting a single prism (socialism plus Yiddish) through which to judge all of the events of the world is always a dangerous thing. In the polarised times in which he lived, Zhitlovski adopted just such a prism very early in life and exhibits all the assets and the debits of so doing. His language views were a good example of the Jamesian theory as to the functional autonomy of motives (“a mechanism often tends to become a drive per se”) and of the possible stability of language loyalty even during periods of ongoing ideological change. After the Holocaust, the position of Yiddish changed radically, from being the darling of Jewish left-wing secularists to being a bastion of fundamentalist Ultra-Orthodoxy. These two worlds have hardly interacted, let alone formed a single speech community, in more than a century. Zhitlovski belonged to only one of these two worlds, and the fact that it is now no more the one in ascendancy should not keep us from being aware of his important role then, when that world was in ascendancy, nor of the existence of Jewish circles today that still selectively



Introducing Khayem Zhitlovski

admire and support various of his nuanced views pertaining to Yiddish and to the desiderata for Jewish ethno-national modernization more generally. Jewish languages and their advocates clearly reveal the “more than language” focus that underlies all movements on behalf of particular languages. They also starkly reveal the multiplicity of functional and symbolic/ideological roles that languages can discharge and be associated with, not only at different times but even at the very same period of a people’s history, not to mention the lifetime of one protagonist.5

Notes * Parts of this paper were originally presented, in Yiddish, at the 1998 Award Meeting of “The Zhitlovski Foundation for Yiddish Culture”, on which occasion I was the recipient of one of its awards. 1. The transliteration key followed in this article for all Yiddish words or names is that of the YIVO/Library of Congress. Please note that in many sources not following this transliteration key our main protagonist’s name is transliterated as “Chaim Zhitlowsky”. 2. The earlier date pertains to the original Russian version of this article (which remained unpublished because the journal for which it was intended rejected it “in principle”). The second date pertains to its publication in Yiddish. In the latter version Kh Zh already adds footnotes commenting on the rapid growth in the literary and educational value of Yiddish in the brief period of only four years. 3. Kh Zh’s cited publications are listed under the dates of their original sources, rather than in the order of their translated or collected (re-)publication. 4. Note that various volumes of Kh Zh’s Gezamlte Shriftn were published in various years, rather than all at the same time.

References Birnboym, N. 1902. “Losungen der Judenfrage”. Ost und West 2 (6): 5–13. Fellman, J. 1973. The Revival of a Classical Tongue. The Hague: Mouton. Fishman, J. A. 1981. “Attracting a following to high-culture functions for a language of everyday life”. In Never Say Die! A Thousand Years of Yiddish in Jewish Life and Letters, J. A. Fishman (ed.), 369–394. The Hague: Mouton. Fishman, J. A. 1987. Ideology, Society and Language: The Odyssey of Nathan Birnbaum. Ann Arbor: Karoma. Fishman, J. A. 1991. Yiddish: Turning to Life. Amsterdam: Benjamins. Fishman, J. A. 1998. “Zhitvovski un tshernovits”. Afn shvel 311: 1–4. Gunier, L. 1994. The Tyranny of the Majority. New York: Free Press.

153



154

Joshua A. Fishman

Hess, M. 1862/1965. Rome and Jerusalem. New York: Philosophical Library. Kallen, H. 1924. Culture and Democracy in the United States. New York: Boni Liveright. Walzer, M. 1983. Spheres of Justice: A Defense of Pluralism and Equality. Oxford: Blackwell. Walzer, M. 1992. What it Means to Be an American. New York: Marsilio. Zhitlovski, Kh.3 1892/1917. “A yid tsu yidn”[A Jew to Jews]. In Gezamlte shriftn [Collected Writings4] 6: 29–43. New York: Dr. Zhitlovski farlag gezelshaft. Zhitlovski, Kh.1897/1900. “Far vos davke yidish” [Why necessarily Yiddish]. In Gezamlte shriftn [1912] 5: 29–43. New York: Dr. zhitlovski farlag gezelshaft. Zhitlovski, Kh.1904/1917. “Dos yidishe folk un di yidishe shprakh”. [The Jewish people and the Yiddish language]. In Gezamlte shriftn [1912] 5: 79–100. New York: Dr. zhitlovski farlag gezelshaft. Zhitlovski, Kh. 1914. “Ven un vi ikh bin gevorn a yidishist” [When and how I became a Yiddishist]. Yidishe velt (Vilne) 1 (1): 50–53 and (2): 221–227.



Arab language education in the Hebrew state Muhammad Amara Bar-Ilan University

Introduction The Arabs in Israel constitute 18 percent of the total population, numbering 1,130,0001 including East Jerusalem.2 Language education is considered an important part of the curriculum at all stages of education. The languages studied are Arabic, Hebrew, English; French is studied as well in several private schools. The Bagrut examinations3 and the National Assessment for Educational Progress4 (known as MASHOV) clearly show that the Arab pupils in Israel do not have a satisfactory command of the three main languages (Arabic, Hebrew and English) that are taught.5 The unsatisfactory command of these languages is attributed to two main groups of factors, which can be labeled external and internal. The chief external factors are: the policy of control and supervision that the Ministry of Education exercises on Arab education in general (Al-Haj 1995) and on language education in particular; the dominance of Hebrew; the status of Arabic as “official” in name only; and the definition and perception of Israel as a Jewish State. The internal factors are related to social processes within Arab society, that is, the socio-economic situation, the diglossic situation of Arabic, and the increasing use of Hebrew features into Spoken Arabic. As Al-Haj (1995: 216) explains “the ruling groups [in Israel], without exception, have attempted to delegitimize Arab and Palestinian nationalism and to use the education system as an instrument for legitimizing the official ideology alongside the transmission of vague universal values”. Formulation of the language education policy in the Arab sector is under the exclusive responsibility of the Ministry of Education. The implementation of this policy is evident, for instance, in the curricula, in the number of hours allocated to language instruction, in the hiring of teachers, and in the approval of textbooks. The teaching materials in the textbooks do not meet the needs of the Arab pupil, and they are designed, among other things, to educate the

156

Muhammad Amara

learner about the values of the Jewish society and its customs. For instance, in Hebrew language classes, Arab pupils learn more about Judaism than Jewish pupils do in State-sponsored public schools. In Arabic more literary texts from the classical period are learnt than from the modern period. These texts are distant from the life and the identity of the Arab learner. In English there are many texts imbued with Jewish themes and Western culture, also distant and even alien to the Arab pupil’s world. Language education emphasizes the technical side of language instruction. That is to say, language is emphasized as a means of communication, and the importance of language in fostering the character of the Arab pupil as an independent individual; in conveying national symbols is almost ignored. Arabic in Israel is recognized as an official language along with Hebrew. However, in reality it is official in name only (Spolsky and Shohamy 1999). Jews, as many studies show,6 avoid studying Arabic, while Arabs study Hebrew formally (from the third grade on) and informally for both instrumental and communicative needs. Highly educated Arabs in most cases have a better command of Modern Standard Hebrew than of Standard Arabic (see Amara and Abu-Akel 1998). A linguistic repertoire is considered an important asset for every minority. Good command of a variety of languages is needed by minorities to increase their chances to cope successfully with the socio-economics of their lives. Appropriate language education policy may contribute to success in schooling and raise the level of their achievement. It is important to emphasize that success in education contributes significantly to the integration of the Arab graduate in the life of the country, giving him/her better opportunities for progress in his/her career. The purpose of this chapter is to highlight the major forces shaping Arab language education in Israel. The forces influencing Arabic, Hebrew and English and the effects of these forces on pupil achievement will be considered. First, a general background of Arab society in Israel is given. This is followed by a brief description of the group’s language repertoire. Finally, the major factors influencing and shaping Arab education are described and discussed.

Arabs in the Hebrew state: General background The point of departure here is the impact of the extra-linguistic forces on the Arab language education. Arab society in Israel has been considerably influenced

Arab language education in the Hebrew state

by the changed socio-political circumstances in the last five decades. Three interrelated factors need to be considered in order to understand the main characteristics of the society and the collective identity of the Arabs in Israel: (1) Israel’s policy towards the Arab minority; (2) internal developments within the Arab society; and (3) external regional developments and their influence on the Arabs in Israel. Relevant to the first of these is the Israeli-Arab conflict and the perception and definition of Israel as a Jewish State. Together these help determine the nature of relations between the Arab minority and the Jewish majority in the Jewish State. These relations are characterized by continuous tension and friction. The basic assumption after the founding of Israel was that because the Arabs in Israel are a national minority, they are an integral part of the Arab world and because they identify with that world emotionally and physically, they constitute a risk to Israel’s security. As long as no appropriate solution is found to the conflict, the Arabs will continue to be a security risk (Riter 1996). Consequently, the Arabs are perceived as citizens whose loyalty to the country is “questionable and … a potential for risk to its Zionist character (in the best case) or even to its very existence (in the worst case)” (Benziman and Mansur 1992: 211). The security perception has brought about the exemption of the Arabs in Israel from the duty of military service and has created the perception of “loyalty conflict” (Riter 1996) or “dual loyalties” (Landau 1971). According to these perceptions, the Arabs are more loyal to the Arab world than to Israel. The results of these perceptions, especially military exemption, has helped to institutionalize discrimination against Arabs in Israel. For example, many financial benefits are granted only to those who serve in the army. Israel performs many activities in the name of security. Land appropriation is one of the most salient; in this and other domains, laws have been enacted, and emergency regulations have been implemented. A well-known example concerns the villages of Iqrit and Biram.7 Discrimination against the Arabs is also evident in allocating government resources and payments, such as national child allowances (up until 1998) and university fees, as determined by the Katsav Commission. The security issue is not the only factor influencing the situation and status of Arabs. The main influence on their situation and status lies in the very definition and perception of Israel as a Jewish State. The Independence Charter states this clearly in its formulation that Israel is a Jewish national state, and many laws have been enacted to enhance the Jewish character of the country (Kretzmer 1990).

157

158

Muhammad Amara

Extensive and continuous concern for security issues has prevented genuine discussion of the essence and identity of the country, and has made the question of relations between the Arab minority and the Jewish majority a marginal issue. Policymakers in Israel have not clearly determined the type of policy to be carried out regarding the Arabs in Israel, either in the short or long run. Decisions have been made under the pressure of events (Benziman and Mansur 1992). However, because these events were few and scattered, there was no active pressure on decision makers to take decisive steps. Generally speaking, problem solving in the Arab sector is ad hoc, such as after strikes or violent demonstrations. A good example is the first Land Day8 marked in 1976 when the Israel government for the first time considered seriously the demands of the Arab minority. It is worthwhile mentioning that in the last decade considerable changes have occurred among Israeli leaders from both left and right wings about the need to narrow the gap between the minority and majority in various domains of life. This is clear in some areas, but in spite of the changes, the status of the Arabs in Israel has not been essentially altered. In other words, there is clear preference of Jews over the Arabs. So far we have dealt with Israel’s policy towards the Arab minority. We turn now to internal and external developments which shape Arab society in Israel. The main internal factors are: discrimination against Palestinians in Israel in the civil and national domains, modernization and urbanization processes in the Arab sector, and contact with Israeli Jews. One of the consequences of modernization and contact with Jews was the internalization of some aspects of Western culture, some of which had previously been alien to Arab society in Palestine. Modernization and urbanization have led to socioeconomic changes, and the Arab sector, mainly rural, has begun to acquire urban characteristics. In the 1960s and 1970s, in spite of the acquisition of urban characteristics and an increase in the standard of living, local councils were not able to introduce sufficient services and infrastructure to accommodate these emerging changes (Bar-Gal and Soffer 1981). The traditional hamula (a clan of extended families related through a common ancestor and carrying the same family name), which was initially strengthened after 1948, became gradually weaker. Nakhleh (1975) argued that urbanization, changes in the Arab economy, and the resultant decline in its political power weakened the ability of the traditional hamula to control its members. By the early 1970s, this led to increased factionalism in inter-village politics. In the civil domain, Arabs experienced increased discrimination, at the same time becoming more aware of their status and the effects of

Arab language education in the Hebrew state 159

their unfavorable position. In the national domain, the definition of Israel as a Jewish State made it extremely difficult for Arab citizens to identify with symbols which are representative of the Jewish majority, e.g., the flag and the state’s national anthem. Finally, several regional developments over the last fifty years have contributed to the shaping of Arab society in Israel: Pan-Arabism in the 1950s and 1960s, the defeat of the Arab countries in the Six Day War, the increasing power and recognition of the PLO in the international arena, the success of the Islamic revolution in Iran and the growth of religious fundamentalism in the neighboring countries, the Lebanon War in 1982, the Palestinian Intifada 1987–1993, and more recently, the Oslo accords.9 In sum, Israel’s policy towards the Arab minority together with internal and external developments all contribute to the current texture of Arab society in Israel.

Language education The remainder of the paper seeks to highlight the major factors affecting Arab language education in Israel. First, a brief description of the Israeli Palestinian language repertoire is in order.

Linguistic repertoire The Palestinian Arab linguistic repertoire during the British Mandate was simple, and to some degree, uniform. Since the majority of people were villagers and since Palestinian society was predominantly agrarian, schooling was not available to all, and contact with the outside world was infrequent. The majority of the Palestinians knew and spoke mainly the local Palestinian dialect, and only limited sectors of the population knew standard Arabic, English; even fewer knew and used Hebrew. After the establishment of the State of Israel, the Israeli Palestinian linguistic repertoire gradually became more complex and diverse. The linguistic repertoire of Palestinians in Israel is at the present time changing rapidly, with many of the changes coming from increasing contacts with standard Arabic, from contact with other varieties of Arabic, and from contact with Hebrew and English (e.g., Amara 1986, 1995, 1999a). Arabic is the mother tongue and the main national language of the Arab citizens of Israel. It serves as the sole official language of Israel’s neighboring countries and enjoys a unique status in most Muslim countries.10 Arabic in

160

Muhammad Amara

Israel is unique in a different way: It is both a minority language and a recognized second official language. Arabic was a major language until the founding of Israel, and due to changes in socio-political circumstances, it became marginalized. Arabic is learnt as the first language in all the Arab schools from the first to the twelfth grades, and is the language of instruction at several teacher training institutes. Hebrew is the dominant national language. It is learned formally and informally. Since Hebrew is one of the official languages of Israel, Israeli Palestinians learn Hebrew as the language of the country and not as a foreign language (see Winter 1981). Hebrew is taught in Israeli Palestinian schools from the third grade on, but the influence of informal learning from outside contact is even more important (see Reves 1983). All age groups, regardless of gender, have informal contact with Hebrew speakers, though in varying degrees (see Spolsky and Cooper 1991). Arabs learn Hebrew for pragmatic reasons: for work, for communication in everyday matters, to obtain services from governmental and private institutions, and for continuing studies at institutions of higher education (Amara 1986; Amara and Spolsky 1986). Consequently, Hebrew is an important tool for every Arab citizen in Israel, since it enables him to function effectively in all domains of life. English is taught in Israeli Arab schools from the fourth grade. In other words, it is studied formally. Outside contact is very slight because there is no direct contact between the Arabs in Israel and an English-speaking community. However, English is important because of its role as the international language of science, technology and commerce, the popularity of American culture, and the close relationship between the USA and Israel.

Major factors affecting Arab language education The results of the National Language Tests (Mashov) of 1996 and 1997 show consistent and wide gaps between Arab and Jewish pupils in performance on tests in their mother tongues, Arabic for Arabs and Hebrew for Jews. (See Table 1, and the same results are displayed in Figure1), and also in English (Table 2 and Figure 2). The gap in English between the Arab and Jewish pupils is striking, amounting to two standard deviations. Only 40% of the pupils in the Arab sector reached a satisfactory level of achievement, among whom fewer than 10%

Arab language education in the Hebrew state

Table 1. Averages for achievement in reading comprehension and writing in the mother tongues in both Arab and Jewish schools in grades 4 and 8. ReadComp-4

ReadComp-8

Writing-4

Writing-8

60.3 072.0

56.1 73.6

39.5 72.1

54.1 70.9

Arab (Arabic) Jewish (Hebrew)

Source: The Ministry of Education, Culture and Sport and The National Institute for Testing and Evaluation 1998a, 1998b, 1998c and 1998d. The National Mashov for the Education System, grades 4 and 8 in Arabic and Hebrew as mother tongue in both Arab and Jewish sectors.

80 60 40 20

Jewish Arab

Figure 1.

Writ iting-8 ing-8

Writ iting-4 ing-4

Rea eading ding Comp mprrehe hensio nsion-8 n-8

Rea eading ding Comp mprrehe hensio nsion-4 n-4

0

Achievements in reading comprehension and writing in both Arab and Jewish schools in grades 4 and 8 (%)

reached an advanced level. Close to 60% fell below the level desired. Simply put, 60% of the Arab pupils failed the test, in contrast to approximately 15% of the Jewish pupils. The data presented above indicate that approximately half of the Arab pupils fail to reach a satisfactory level of achievement in their mother tongue, Arabic, and the majority fail to do so in English (60%). What are the explanations for this low level of achievement? As explained above, many forces are involved in shaping Arab language education. To understand the complex nature of the issues in Arab language education and the low level of achievement in the various languages taught, including the mother tongue, linguistic, sociological, political and pedagogical explanations are all relevant. We begin first with diglossia,11 offering linguistic as well as ideological explanations. Diglossia, as many studies have shown (e.g., Amara and Abu Akel

161

162

Muhammad Amara

Table 2. Averages of achievements in English in five skills in both Arab and Jewish schools in grade 8.

Arab sector Jewish sector

Reading

Listening

Speaking

Writing

Grammar

42.6 73.0

40.4 83.9

40.7 61.0

37.2 68.8

27.0 50.7

Source: The Ministry of Education, Culture and Sport and The National Institute for Testing and Evaluation 1998e. The National Mashov for the Education System, English, grade 8, in both Jewish and Arab sectors.

Arab

100

Jewish

80 60 40 20 0 Rea eading ding

Figure 2.

Listeening List

Speaking Sp eaking

Writ iting ing

Grammar

Achievements in English in grade eight, according to skill by sector (%)

1998; Maamouri 1998) is a major burden on the Arab learner. Standard Arabic is a different language, with syntax, morphology, and lexicon all substantially different from the spoken variety. The problematic nature of a diglossic situation is not only linguistic, but also social and ideological. The standard is limited in its use to only a few domains and settings, making it limited as a spoken variety. Competence in the standard variety requires competence in the four major language skills, which are acquired only through education. In school standard Arabic is mainly used in language classes. Thus, the learner speaks the standard with difficulty only when the situation demands it. This limitation in use explains why the standard is not an everyday language, and probably explains the low proficiency of most Arab learners in either writing or speaking the standard. The distinction between the spoken and written varieties is to a large degree ideological, with the major purpose being to maintain the purity of the Arabic language.12 The standard is limited to formal domains, such as school, media, courts, and mosques or churches etc. So it is probably not surprising that the various Arabic reform attempts have met with almost total failure (Abu-Absi 1986). Arabic diglossia in Israel is even more complicated and constitutes a bigger

Arab language education in the Hebrew state 163

burden on its speakers than it does in Arab countries. As mentioned already, the official status of Arabic in Israel is only titular (Spolsky and Shohamy 1999). Its use is limited to the local public sphere. Beyond that sphere, mostly Hebrew that is used, e.g., in the Knesset (parliament), in the media, in higher education, on road signs, etc. This means that Arabic in Israel is different from Arabic in the Arab world, since the public sphere for the use of the standard is extremely limited, mainly used in education (and even in education, as we will see below, it competes with other languages) and in religious places. In other domains, Hebrew is dominant. Furthermore, there is a major influence of Modern Standard Hebrew on educated Arabs. The majority acquire their higher education in Hebrew13 and feel much more comfortable expressing themselves in writing and formal speaking in Hebrew (Amara and Abu-Akel 1998). The burden of diglossia is evident not only in the case of the mother tongue, but also in the case of Hebrew, as a second language, and English, a foreign and third language for Arabs. It is also important to mention that Arab pupils must learn three writing systems. A second factor in Arab language education is the socio-political milieu. The changed socio-political circumstances after the founding of Israel turned the Arabs in Israel into a numerical and marginal minority. Life necessities and priorities heavily influence language proficiency and use among the Arabs in Israel. From the various studies conducted by Amara (e.g., 1986, 1995, 1999a) on the language repertoire of Israeli Palestinians, it was found that Arabs use Hebrew not just for filling gaps left by the absence of equivalent elements in Arabic. That is to say, Hebrew features are not only used for communicative needs, but also to “show off”.14 Hebrew fulfills a major symbolic function among Palestinians in Israel and symbolizes the desire and aspiration to associate oneself with the outside modern world. The prestige of Hebrew is related to Israeli progress in many fields. That Israel is seen as a modern country with advanced technology has encouraged Palestinian youth in Israel to learn Israeli patterns of behavior in order to join this progress. The fact that Israeli Palestinian youth read Hebrew papers and watch Hebrew programs on television is an indication of their desire for some of this modernity. Nevertheless, they attach different values to the two languages. Because Israeli Palestinians are aware of the fact that Arabic is a rich, beautiful and prestigious language, for them the mastery of Hebrew is a method of achieving social, educational and economic levels similar to those of Israeli Jews (see Amara 1986). This implies that Palestinians in Israel learn Hebrew for practical or instrumental motivations rather than integrative purposes. This

164

Muhammad Amara

situation reflects the nature of Palestinian-Jewish relations in Israel. First, Israel is considered a Jewish state, rather than a country for all its citizens. Consequently, Palestinians seek to enhance their unique identity in the Jewish state with Arabic being an important vehicle (see Amara 1995). Second, the IsraeliArab conflict enhanced the differences between the majority and minority. Third, residential patterns which can be described as defacto segregation do not contribute to extensive contact between the two nations. All of the above factors together do not lead to a high level of social convergence towards the majority culture and its language among Palestinians in Israel. This means that each side preserves its identity and the associated language (Amara 1999b). Though Hebrew is the other significant language among Palestinians in Israel, due to contact with Israeli Jews in various domains of life, and though it serves as a modernization agent, there are nevertheless sociolinguistic constraints on language convergence. In the words of Ben-Rafael (1994: 176): … a barrier impedes this convergence, as expressed in a retention of Arabic. The limits each case imposes on convergence towards the dominant culture respond to the nature of the commitment to the dominant culture. For the Muslim and Christian Arabs, the legitimate language remains Arabic, as an expression of their fundamental identity. The penetration of Hebrew as the dominant language does not subtract anything from Arabic, though its deeper influence comes out in borrowing and substitutions.

Though the Arabs in Israel express positive attitudes towards English, there is a lower level of priority for learning English. Arabs see learning Hebrew as the first priority (Shohamy and Donitsa-Schmidt 1998). Another related issue is place of residence. A large proportion of Arabic speaking children comes from villages (more than 65%) and not towns, where they have less exposure to the English speaking Western culture that has come to dominate Israeli cities. A third factor is teachers’ qualifications and status. In the three languages taught, there is a problem of teacher qualification. Most of the university graduates of Arabic language and literature receive their education at Israeli universities, where Arabic is taught as a second (or even foreign) language, and teacher education programmes are not designed for Arabic as a mother tongue. Consequently, teaching Arabic is based on intuition rather than knowledge and expertise. This is also true to a large degree of Hebrew language pedagogy, since most of the Arab teachers receive their higher education at Israeli universities. This means that their qualification in Hebrew is as a mother tongue, rather than as a second language. In the case of English, the gap is even wider when

Arab language education in the Hebrew state 165

compared with the Jewish sector. A large proportion of the teachers in the Jewish sector have university degrees in English, which means that they spent several years in classes taught in English. Another sizable group of English teachers includes either native speakers of English or those who have spent extended periods in English-speaking countries. It is also to be noted here that decisions about the appointment of language teachers in the Arab sector are made by general Ministry of Education inspectors and not, as is increasingly true in the Jewish sector, by language inspectors. Still another relevant issue is the way Arab teachers are hired. As Rouhana (1997: 86) explains “the security principle was used to restrict teacher appointments relentlessly in the early years of the state”. Though the security principle has attenuated over the years, hiring Arab teachers is still influenced by security and political considerations. Arab teachers must fill out a “security form”, which does not exist in the Jewish sector. This affects the level of teaching since some qualified teachers are not hired. A fourth factor concerns educational goals, curricula, and textbooks. One of the major goals of the Israeli education in the Arab sector is to make Arab education devoid of any national content. Al-Haj (1995: 121) explains that “policymakers sought from the very beginning to reinforce the religiouscultural component and Israel-citizenship component instead of the Arabnational component”. This policy is in line with the definition of Israel and its perception as a Jewish State. This policy has been implemented through curricula and materials. Former curricula and textbooks that were used in the British Mandate were completely removed. The new curricula and textbooks was aimed to tighten the control of the state over the content of Arab education (Al-Haj 1995). Analyses of the educational goals, curricula and content of textbooks in the Arab sector shows clearly that the State aims at weakening the Palestinian Arab identity (Peres et al. 1968; Mari 1978; Lustick 1980). The textbooks in the Arab sector in the three languages do not meet the needs of the Arab pupil. Through Arabic education, the country aspires to achieve a policy of denationalization of the Arab minority, that is to say, to weaken and devalue Palestinian identity and to expose Arab pupils primarily to texts from the classic Arab period. In the early period of Israeli statehood, Arabic was emphasized in the curriculum as a tool for self-expression, totally excluding its role in conveying national symbols. In the recent curriculum the national pride of the Arabs in their language is mentioned as one of the goals of teaching Arabic. For example, one of the goals is “the pride of the Arab in his

166

Muhammad Amara

national language”.16 However, there is still little congruence between the declared goals of the curriculum and the teaching materials. In Hebrew, emphasis is placed on “exposure of Arab students to the culture and heritage of the Jewish people and on the development of Israeli citizenship” (Al-Haj 1995: 133). This becomes clear when we compare Hebrew in Arab schools with Arabic in Jewish schools. In the Arab schools the aim of teaching Hebrew is to expose the Arab student to Hebrew culture and values in the past and present (Mari 1978), while the aim of teaching Arabic in the Jewish schools is instrumental, i.e., learning about Arabic language and literature. Though educational goals and the curriculum for teaching English are the same for both Arabs and Jews, the content of the textbooks put the Arab pupils at a disadvantage. There is a lack of cultural adaptation of the textbooks to Arab minority pupils. The textbooks represent the majority culture, which no doubt further weakens motivation. The factors influencing and shaping Arab language education examined above are not the only ones, but they are the most influential. Other factors related to those mentioned above are, for instance, allocation of resources and pedagogical issues, e.g., the much lower level of support that the schools in the Arab sector receive (Amara and Kabaha 1996). There is a tremendous shortage of teaching hours allotted to Arabic language classes. The curriculum for teaching Arabic is outmoded. Currently, the Arab student learns about Arabic, rather than how to use the Arabic language. Because of the diglossic situation, the skill of speaking is largely neglected.

Conclusion The unique case of Arab language education in Israel has been influenced and shaped by internal and external factors. As Spolsky (e.g., 1972; Spolsky and Shohamy 1999) wisely suggests, language education is a complex matter, its understanding depends on a great number of sociological, economic, historical, political educational, religious, cultural, and other factors. Unfortunately, in language education both in the Arab world (Amara and Abu-Akel 1998; Maamouri 1998), and in Arab society in Israel, linguistic and pedagogical factors are those advanced to describe and account for language education in general and for the curriculum and textbooks in particular. External factors, such as the political and socio-cultural, are not seriously considered by policymakers. This paper has attempted to bring some of these factors to light.

Arab language education in the Hebrew state 167

Notes 1. Muslims 15% (901,000), Christians 1.7% (129,000) and Druze 1.3% (99,000) (Yediot Ahronot, 1.1.1999). 2. Shortly after the 1967 War, Israel unilaterally unified East and West Jerusalem, declaring the whole city of Jerusalem as the capital of Israel. The Arab population of East Jerusalem, which includes the Old City and the inhabited areas outside it, numbers about 180,000 people. East Jerusalem is not included in describing Arab language education, not only for political reasons, but because linguistically and in terms of language education it has more to do with the West Bank than the Arabs in Israel. 3. The Bagrut is the Hebrew term for matriculation examinations given at the end of high school in all schools in Israel. Great importance is accorded to Bagrut achievement in Israel, especially for college and university admission. 4. These tests were carried out by the Ministry of Education with the assistance of the National Institute for Testing and Evaluation. They are considered the most comprehensive and reliable language achievement tests in the Israeli schools. 5. School achievement of Arab pupils in comparison with Jewish pupils is reported below. 6. Most of the studies on teaching Arabic in Jewish schools (e.g., Ben-Rafael and Brosh 1991; Kraemer 1990) show negative attitudes of both pupils and parents towards Arabic. Arabic suffers from a low status among Jews. 7. These villages, located near the Lebanese border, were evacuated during the 1948 War by the Israeli army. The Israeli authorities promised them that they would come back once the war was over. However, so far the villagers have not been allowed to go back to their villages, and for about 50 years they have run a legal and public struggle to return to their villages. 8. Land Day was first marked in 1976, when Israeli Palestinians participated in massive protests against the confiscation of Arab lands and continued discrimination with respect to civil rights. On March 30, 1976 violent clashes took place between Palestinians in Israel and the Israeli police. Six Palestinians were killed, and several were wounded. This day came to be known as Land Day, which is remembered and celebrated annually by Palestinians in Israel. 9. For greater details on the influence of regional developments on the Arabs in Israel, see Amara and Kabaha (1996) and Rouhana (1997). 10. Arabic is the sole official language in the Middle East, excluding Turkey and Iran, and the Arab countries of North Africa. In Israel, Somalia and Chad, Arabic shares official status with other languages. Because it is the language of the Qur’aan, it has a special status in many Muslim countries. 11. Arabic is considered a diglossic language (Ferguson 1959; Brosh 1996). One of the main characteristics of a diglossic situation is that the functional division between the standard (according to Ferguson, the High variety) and a local dialect (low variety) is absolute. That is to say, the standard is used for specific functions and the spoken for others. The use of functions of one variety in the other is artificial if not ridiculous.

168

Muhammad Amara

12. The spread of Islam, the learning of Arabic by non-Arabs, and the threat of a wide diversification of Arabic contributed to this distinction. 13. There are no Arab universities in Israel. Arabs tried to establish an Arab university, but failed to do so, mainly because the Hebrew State is not interested in such a university, lest it becomes the center for Arab intellectuals and for nationalism. 14. Observation of signs in an Israeli Palestinian city, Tira, located in the area known as the Little Triangle, showed that more than 90% of shop signs were written in Hebrew — though in many of the stores, clients are only Arabs. One explanation for this may be that Hebrew enjoys a high status among Israeli Palestinians; writing in Hebrew adds prestige and credibility to the goods being offered for sale. 15. This goal was formulated in the recent curriculum. In former curricula, Arabic teaching was not associated with nationality or identity.

References Abu-Absi, S. 1986. “The modernization of Arabic: Problems and prospects”. Anthropological Linguistics 28 (3): 337–348. Al-Haj, M. 1995. Education, Empowerment, and Control: The Case of the Arabs in Israel. New York: State University of New York Press. Amara, M.1986. The Integration of Hebrew and English Lexical Items into the Arabic Spoken in an Arab Village in Israel. Unpublished MA Thesis, Bar-Ilan University, RamatGan. Amara, M. 1995. “Hebrew and English lexical reflections of socio-political change, in Palestinian Arabic”. Journal of Multilingual and Multicultural Development. 15 (6): 165–172. Amara, M. 1999a. Political and sociolinguistic reflexes: Palestinian Border Villages. Amsterdam: John Benjamins. Amara, M. 1999b. “Hebrew and English borrowings in Palestinian Arabic in Israel: A sociolinguistic study in lexical integration and diffusion”. In Language and Society in the Middle East and North Africa: Studies in Variation and Identity, Y. Suleiman (ed.), 81–103. London: Curzon Press. Amara, M. and Abu-Akel, A. 1998. “Arabic Diglossia and Language Policy in the Arab World”. Alrisala 6: 289–308. Amara, M. and Kabaha, S. 1996. Split Identity: Political Division and Social Reflections in a Divided Palestinian Village. Giv’at Haviva: The Institute for Peace Studies. Amara, M. and Spolsky, B. 1986. “The diffusion and integration of Hebrew and English lexical items in spoken Arabic in an Israeli village”. Anthropological Linguistics 28 (1): 43–58. Bar-Gal, Y. and Soffer, A. 1981. Geographical Changes in the Traditional Arab Villages in Northern Israel. Durham, England: Center for Middle Eastern and Islamic Studies, University of Durham.

Arab language education in the Hebrew state 169

Ben-Rafael, E. 1994. Language, Identity and Social Division: The Case of Israel. Oxford: Oxford Studies in Language Contact. Ben-Rafael, E. and Brosh, H. 1991. “A sociological study of language diffusion: the obstacles to Arabic teaching in the Israeli school”. Language Planning and Language Problems 15 (1): 1–23. Benziman, U. and Mansur, A. 1992. Sub-tenants: Arabs in Israel, their Status and the Policy towards them. Jerusalem: Keter. (In Hebrew) Brosh, H. 1996. “The connection between the spoken language and the written language in teaching Arabic in the Hebrew school”. Dappim 23: 64–75. (In Hebrew) Ferguson, C. A. 1959. “Diglossia”. Word 15: 325–340. Kretzmer, D. 1990. The Legal Status of the Arabs in Israel. Boulder, CO. Kraemer, R. 1990. Social psychological factors related to the study of Arabic among Israeli Jewish High School students. Unpublished Ph.D., School of Education, Tel Aviv University, Ramat-Aviv. Landau, J. 1971. Arabs in Israel: Political Readings. Tel-Aviv: Ma’rakhot, Ministry of Defense. (In Hebrew) Lustick, I. 1980. Arabs in the Jewish State: Israel’s Control over a National Minority. Austin: University of Texas. Maamouri, M. 1998. Language Education and Human Development: Arabic Diglossia and its Impact on the Quality of Education in the Arab Region. [Discussion paper prepared for the World Bank. The Mediterranean Development Forum, Marrakech, 3–6 September 1998]. Mari, S. 1978. Arab Education in Israel. Syracuse, N. Y.: Syracuse University Press. The Ministry of Education, Culture and Sport and The National Institute for Examinations and Evaluation 1998a. The National Mashov for the Education System — Mother Tongue (Arabic), Grade 8. Jerusalem. The Ministry of Education, Culture and Sport and The National Institute for Examinations and Evaluation 1998b. The National Mashov for the Education System — Mother Tongue (Hebrew), Grade 8. Jerusalem. The Ministry of Education, Culture and Sport and The National Institute for Examinations and Evaluation 1998c. The National Mashov for the Education System — Mother Tongue (Arabic), Grade 4. Jerusalem. The Ministry of Education, Culture and Sport and The National Institute for Examinations and Evaluation 1998d. The National Mashov for the Education System — Mother Tongue (Hebrew), Grade 4. Jerusalem. The Ministry of Education, Culture and Sport and The National Institute for Examinations and Evaluation 1998e. The National Mashov for the Education System — English, Grade 8. Jersalem. Nakhleh, K. 1975. “The Direction of Local-Level Conflict in two Arab Villages in Israel”. American Ethnologist 2 (3): 497–516. Peres, Y, Erlich, A. and Yuval-Davis, N. 1968. “National education for Arab youth in Israel: A comparative analysis of curricula”. Magamot 17 (1): 26–36. Reves, T. 1983. What makes a good language learner? Personal characteristics contributing to successful language acquisition. Unpublished Ph.D., The Hebrew University of Jerusalem.



170

Muhammad Amara

Riter, I. 1996. The Civil Status of the Arabs in Israel in a Time of Peace. Beit Berl: The Institute for Israeli Arab Studies. (In Hebrew) Rouhana, N.1997. Palestinian Citizens in an Ethnic Jewish State: Identities in Conflict. New Haven and London: Yale University Press. Shohamy, E. and Donitsa-Schmidt, S. 1998. Jews vs. Arabs: Language Attitudes and Stereotypes. Tel Aviv: The Tami Steinmetz Center for Peace Research, Tel Aviv University. Spolsky, B. 1972. “Introduction”. In The Language Education of Minority Children, B. Spolsky (ed.), 1–10. Rowley, MA: Newbury House. Spolsky, B. and Cooper, R. 1991. The Languages of Jerusalem. Oxford: Clarendon Press. Spolsky, B. and Shohamy, E. 1999. Languages of Israel: Policy, Ideology and Practice. Clevedon: Multilingual Matters. Winter, M. 1981. “Basic problems in the Arab education system”. In Arabs in Israel: Continuity and Change, A. Lish (ed.), 168–179. Jerusalem: Magnes Press. (In Hebrew) Yediot Ahronot, 1.1.1999.



Adult use and language choice in foreign language policy Richard D. Lambert University of Pennsylvania

In the last several decades, the field of language policy analysis has grown immensely. A rich set of case studies of language policy in a large number of countries is now available, and several attempts have been made to develop a comparative, analytic framework. (e.g., Spolsky 1999; Schiffman 1996; Lambert 1999). Most studies of language policy, however, concentrate on what I would call “domestic” languages, those in use by major portions of the population within a country. Its two principal divisions, “corpus policy” that deals with the prescription of the proper form of a country’s language(s); and “status policy” that is concerned with the relative standing of the languages of ethnic minorities are both focused within the country. It is striking, how little attention is paid in discussions of general language policy to the use and teaching of foreign languages, languages whose home domain is in another country. The distinction between domestic and foreign languages is, of course, a slippery one, as any prairie province Canadian student learning French, or American student studying Spanish can attest. There is also a debate about whether ex-colonial languages in, for instance, Africa (Tengan 1994) and India are domestic or foreign languages. Even allowing for this definitional imprecision, however, it is surprising how little discussion of foreign languages there is in analyses of domestic language policy. While discussions of foreign languages do not often find their way into domestic language policy, there is a vast literature dealing solely with what is, in effect, foreign language policy, although it is usually not referred to as policy. A number of comprehensive foreign language policy analyses within single countries have appeared recently. They include the Dutch National Action Plan (Van Els, et al. 1990), the Australian National Plan (Lo Bianco 1987), and most recently in England a national review of foreign language policy sponsored by the Nuffield Foundation entitled Where Are We Going with Languages (Moys 1998).

172

Richard D. Lambert

We are also beginning to see research into the implementation of countryspecific foreign language plans (Djité 1994; Nikki 1992; Sajavaara 1997; Tuin and Westhoff 1997). Further, in recent years a number of carefully constructed comparative studies have appeared (Dickson and Cumming 1996; Bergentoft 1994) which describe the language instructional systems in a number of countries. The growing analytic literature provides the basis for the development of a more general theoretical framework for the analysis of foreign language policy. In this paper I would like to concentrate on two topics that have the greatest potential for interrelating the theories of domestic and foreign language policy. The first domain is one that tends to get the least attention from policy makers: the relationship of foreign languages to the larger society, in particular the relationship of the foreign language system to adult use and learning. I propose to discuss in turn four topics: (1) societal characteristics that determine the extensiveness and success of the system; (2) the determination of national and occupational needs; (3) adult foreign language education; and (4) “language enablers”, that is mechanisms for helping the language-limited user deal with foreign languages. The second part of the paper will deal with another area of fruitful overlap between domestic and foreign language policy. A great deal of status language policy is concerned with language choice, that is which languages will be taught to and used by whom. This decision is one of the principal concerns of foreign language policy makers as well. Sometimes it is determined by the use of languages in the larger society, sometimes, as in the United States and the United Kingdom, it is not. A discussion of language choice, then, presents another opportunity for overlap with the theoretical structure of domestic language instruction.

Foreign language and society Determinants of the extensiveness of the system Countries vary immensely in the extensiveness of their foreign language education establishment. While many analysts have a general idea of the amount of foreign language instruction available in various parts of the world, it is curious that there is so little organized speculation about what determines

Adult use and language choice in foreign language policy

how important foreign language teaching is in individual countries. An exception is the following hypothesis proposed by John L. Trim (Trim 1994). A country is more likely to value L2 proficiency highly if: 1. It is a small country but not geographically isolated 2. The neighbors speak a different language 3. Its own language is not widely spoken and not used as a vehicle of international communication 4. Its export/import trade is a high proportion of GDP 5. Its travel trade is a high proportion of foreign trade and GDP.

The important implication of Trim’s hypothesis is that the scale of a foreign language system is determined by characteristics of the larger society. There can be quarrels with the predictive universality of Trim’s paradigm. The Dutch provide an example of an essentially monolingual society in which foreign language learning and use are highly valued. Switzerland is quintessentially multilingual but, as in the case of the Netherlands, foreign language use is extensive. And the countries where foreign languages play the least role in both the larger society and the educational system are what I have called “mosaic societies”, those with five or more major indigenous languages (Lambert 1999a). This term applies to more than half of the countries of the world. Those countries are usually too busy coping with the many indigenous languages, dealing with the linguistic remnants of their colonial past, or searching for a lingua franca to have room for foreign languages. Moreover, in dealing with a country like the United Sates or Israel, it is the perception rather than the reality of monolingualism that matters (Fishman 1969; Spolsky and Shohamy 2000). The United States is de facto a bi- or multi-lingual society, but it is ideologically monolingual, and its foreign language educational system is underdeveloped and undervalued.

Adult use Another way of viewing Trim’s hypothesis is that it argues that use of foreign languages among adults in a society determines the extensiveness and strength of the foreign language system. However, we do not have the systematic data to enable us to test this hypothesis. Some twenty-five years ago an opportunity to collect such data was missed. Early in the formulation of the Council of Europe’s language development projects (Trim 1983: 230–231) it resolved to measure as precisely as possible adult use of foreign languages in each of its member states:

173

174

Richard D. Lambert

They resolved that as a basis for effective planning of a unit/credit system information should be collected (and kept updated) on the statistics of language learning and use in Europe. The statistics would concern: (a) the general demographics of the adult population in terms of age, sex, education, social class, etc.; (b) the existing knowledge of languages distributed over these demographic categories; (c) projects as to what the schools are and what they will be producing in the way of language competence in young adults; (d) the subjectively-felt language needs of different categories of adults; (e) the expressed needs of society for different kinds of language ability; (f) the objective pattern of different foreign language use among adults in different demographic categories; (g) the short and long term projections of language use, and thus of needs.

Unfortunately, this grand scheme of multinational data collection was never implemented. The Council concluded that the language needs of its member countries were ethnographically so varied and the scale of the project was so immense that the proposed parallel data collection system was unfeasible. While this decision is certainly understandable, it is unfortunate that the data on adult use were not collected. For one thing, it would have allowed for a precise testing and elaboration of Trim’s hypothesis. More generally, it would have given us some basis for investigating systematically the relationship of adult foreign language use and foreign language education. However, we do have a little questionnaire data on self-rated language competencies in European countries. The Economist, (Economist 1997: 60) quoting a survey in Eurobarometer reported that: More than half the European Union’s people say they speak at least one European language other than their mother tongue. One in three now speaks English well enough to get along in conversation, making it the Union’s lingua franca. French is spoken by 15% of EU citizens (outside France), while 9% of them (excluding Germans) say they speak German. Only 5% of non-Spaniards in the Union claim to speak Spanish. Multi-lingualism varies hugely within Europe. Not surprisingly, those from small places — Luxembourg, the Netherlands, Denmark — do best, with the Luxembourgers winning the prize for Europe’s paragons. Their official language is Letzeburgisch, a German-French amalgam, but 86 % of people in the duchy also speak French and 77% German as their second languages, while nearly half can chat in English, and a tenth in Italian. The Dutch are fine linguists too. Only 14% of them admit to knowing no second tongue: 79% speak English, two-thirds German, and nearly a quarter French.

Adult use and language choice in foreign language policy

It is the English-speaking countries that have the lowest percentage of adults speaking a language other than their mother tongue. “The British and Irish are the duds. … Two thirds of the people in both countries speak English only, though a quarter of Britons now say they speak French (Economist 1997)”. There are no equivalent figures for the United States, but if the large number of people who speak a language other than English as their home language is left aside (some 30 million people), the proportion of the population of the United States that can carry on a conversation in another language is probably well below even the British. It is interesting to note that these differences among countries in the proportion of polyglots does not precisely follow the total number of hours dedicated to foreign language study. Luxembourg is off the scale. It not only dedicates very large amounts of time to the teaching of French and German as well as Letzeburgisch, but at an early age it uses French, then, German as media of instruction. It maintains no higher educational system of its own. Hence its students desiring to enroll in a university must go elsewhere, most often France. For other countries, the total clock hours given to the study of either a first foreign language or of all languages together (Bergentoft 1994) does not parallel the percentage of adults reporting an ability to carry on a conversation in another language. Indeed, aside from the Russians, Spain has the lowest proportion of adults reporting conversational language skills, but the highest number of clock hours of classroom hours required to be spent in studying a foreign language. If one adds the optional enrollments in multiple languages, the Russians are also near the top in number of hours spent in a language classroom.

National need There was, however, another important theoretical shift behind the demise of the proposed collection of data on adult foreign language. The notion that policy should be a straight line projection from adult use was rejected in favor of an estimation of national need that included but went beyond current adult use. It was argued that the use of foreign languages should be higher than it is at present, that the globalization of the society and the economy, and the creation of a united Europe called for a citizenship with greater language skills than can be found at present. Using this mixture of advocacy and analysis of current practice, the Council of Europe commissioned and published a series of monographs estimating the national need for foreign languages in a number of

175

176

Richard D. Lambert

countries. In the 1970’s and 1980’s, this notion of a national need for foreign language need was picked up by other scholars in Europe as well (Van Els and Oud-de-Glas 1983). It not only served the general rhetoric advocating a general increase in foreign language education, but it served as the basis for the development of new full-scale national plans. For instance, as a precursor to the development of the Dutch National Action Plan for improving foreign language instruction, (Horizon Taal) several national surveys were conducted to determine the actual adult use of foreign languages. The results of these surveys provided the basis for the establishment of priorities in the choice of which languages to teach. Furthermore, the evidence of extensive adult use in the Netherlands served as a rationale for increasing the amount of foreign language instruction provided in the school system and expanding the number and kinds of students who should receive such instruction.

Business needs In the next step in national needs analysis, the domain of national need was narrowed. It came to focus almost exclusively on the needs of business and trade, or more generally, occupational use of foreign language skills. In the l980’s a whole series of surveys was undertaken to determine actual use of foreign language in corporations. Sometimes the notion of business needs was quite general and expressed in terms of the growing international (or European) economy, sometimes it referred more specifically to the projected employment needs of businesses. As a result, corporate needs surveys and company language audits flourished (CILT Information Sheet 46; Van Hest and Oud-de Glas 1991), particularly in England (Hagen 1993, 1994; Emmans, Hawkins and Westoby 1974) and Australia (Ingram, forthcoming). While coverage of the entire corporate sector is quite unusual, and thus projection to aggregate national need is difficult, some of the surveys of corporate language need have been quite extensive. Great Britain has been a leader in this activity. The classic York Report (Emmans et al. 1974) was an early example of the broad scale business needs survey. The LINGUA-sponsored FLAIR survey (Hagen 1993) covering 1306 companies in 5 countries is another example. And some historical data are available. For over a hundred years, The London Chamber of Commerce Examinations Board has been collecting data on business demand for languages and information on the specific tasks requiring foreign language use. The Department of Trade and Industry conducted a National Languages for Export Campaign. The European Commission has sponsored both survey and

Adult use and language choice in foreign language policy 177

training projects for language use in business through LINGUA and LEONARDO DA VINCI. These, and similar surveys of business need for foreign languages tend to take four different approaches to the definition of need (Van Hest and Oud-de Glas 1991): (1) actual language use; (2) estimates of effective demand (Ingram 1998); (3) the deleterious effect of a lack of foreign language competence;1 and (4) estimates of how improved business would be if it added to the foreign language capacity of its employees. Most company-specific language audits use all four of these approaches. However, the more general advocates of greater business use of foreign languages usually emphasize argument (5), that is that foreign language policy should be keyed to externally perceived need rather than current utilization or demand. It is interesting to note that surveys of business use of foreign languages are not numerous in the United States. The few cursive studies (Fixman 1990; Adelman 1994) do not suggest a felt need in corporate America for a substantial supply of language-competent employees. Nor do they find a pattern of extensive foreign language utilization among graduates of educational programs that are specifically focused on combining business and foreign language education. A survey (Lambert and Moore 1990) of the graduates of the three major internationally focused MBA programs who were specifically trained to combine language competency with business skills found a very limited use of foreign language in the workplace. In the United States, in any event, if foreign language advocates are truly convinced of the fourth argument (d above), they need to concentrate their attention on fostering real corporate demand for and use of foreign languages. Several countries have taken a number of major policy steps to deal with this perceived need for greater foreign language competency in business. These include the setting of standards for occupational use of languages, monitoring progress, and funding the development of new programs. For instance, the British established a governmental organization called the Language Lead Body to develop standards for judging the level of language competency needed by various occupations. This guide, entitled Standards for Languages: Units of Competency and Assessment and Guidance (Language Lead Body 1992) is used as a guide both for employers and job applicants as well as serving as a standard for the certification of foreign language instructional providers. It is interesting to note that this function as of May 1998, has been passed on to a private body called the National Languages Training Organization whose leaders are primarily corporate officers. It is also interesting to note that this organization has

178

Richard D. Lambert

expanded the purpose of the organization to explicitly include an advocacy role for the expansion of foreign language use in business. In addition to surveying, advocacy and the setting of standards, many European countries have established training programs whose principal function is to provide language training for occupational use. Sometimes such units are created by universities such as the program at the University of Jyväskylä which both creates an increasingly detailed set of standards of language competency for various occupations and provides the language training for students to meet these qualifications. More often, business language training institutes are free standing, and the government’s role is to assure quality through certification. In the United States characteristically such efforts to control quality have been taken by non-governmental organizations. An organization of proprietary language schools (APPLS) was created and, under the auspices of one of the most prestigious private standards organization, the Association for Standards, Testing and Measurement (ASTM), and in collaboration with representatives of governmental and corporate consumers of language training, created a set of standards for language schools. The standards, however, were not set for student outcomes but for the professionalization of the teaching organization. Currently a more general organization of language providers — bilingual court reporters, interpreters, etc. — is being formed to create an additional set of standards. In Europe, in addition to the introduction of standards and certification, business needs have encouraged, with government help, the development of occupation-specific courses in languages for special purposes. In England, the Further Education Council was established to fund adult education programs primarily with a vocational orientation. In fact, the promotion of such language learning has received a great deal of attention among international language policy makers. Vocational language use has been a primary focus of LINGUA’s interest. It has sponsored both the development of vocationally oriented language teaching facilities and materials and student apprenticeships in which the learning of an occupationally oriented foreign language is a deliberate part. Vocational foreign language orientation is one of the principal concerns in the Council of Europe’s multinational plans for the development of language education. Nonetheless, national foreign language policy tends to deal almost exclusively with the school system and touches only lightly on the language needs of business. For instance, vocational language education is not mentioned in any of the synopses of language policy presented in a recent survey of language

Adult use and language choice in foreign language policy 179

education systems in 35 countries (Dickson and Cumming 1996). There are, of course, some “language for special purposes” courses in upper level secondary schools, universities, and trade schools, and some science departments and business schools offer their own foreign language courses. However, more commonly students who choose a vocational path for their secondary schooling have less foreign language instruction than students pursuing an academic path.

Adult language education While serving the needs of internationally oriented business is one objective of national foreign language policy, business use is only one of the ways that adults make use of foreign languages. Adults need foreign language competencies for a wide range of purposes: travel, trans-national communication, professional linkages, entertainment, participation in the world community. In many countries an underbrush of privately-held language teaching institutions has grown up to serve these needs. Except for a few attempts to set minimal standards for their products, this world of proprietary language schools tends to be unattended and unsupported by language policy makers. Pauline Swanson (Swanson 1998: 56) gives some idea of the variety of such private language learning facilities: Language learning opportunities occur in the workplace, in venues ranging from front rooms, church halls, schools, FE colleges, adult colleges, HE establishments, private language schools. Managed and funded in many ways the potential learner can go accredited, non-accredited, vocational, nonvocational, self-access, distance, one-to-one, group. There are day courses, short courses, intensive courses, courses at all levels in all kinds of languages. The richness of provision is quite startling.

This varied collection of adult language learning opportunities plays a vital role in a country’s language instructional system. Adult language schools, like the formal education system, tend to concentrate in the traditionally taught languages. However, in many countries these schools are the places where instruction is given in the less commonly taught languages. The Dutch National Action Plan, for instance, deliberately assigns instruction in the non-European languages to such schools. In most countries the semi-monopoly of proprietary schools in teaching all but the most commonly taught languages happens by default. For instance, among the westernized nations only the United States and Australia have deliberately encouraged the teaching of Asian languages in their schools and universities. Additionally, the proprietary language schools provide

180

Richard D. Lambert

advanced and specialized language instruction, for instance, advanced English for specialized purposes such as business negotiation, or science. In the United States, adult, occupationally oriented foreign language instruction is primarily found in the very substantial language schools maintained within a number of government agencies (Lambert 1991). For many agencies, the early stages of language learning for government use are carried out in a centralized Defense Language Institute. Others carry out their own instruction at both the elementary and advanced levels. Except for the FALCON program at Cornell, government schools are the only long-term, full-time foreign language teaching schools in the United States. They are not only an important part of adult foreign language instruction in the United States but they have introduced major changes in pedagogical style and assessment procedures throughout the rest of the language educational system. Indeed, a set of computer-interactive language learning materials created by a consortium of intelligence and other government agencies is currently being successfully marketed to secondary schools. If foreign language policy makers are truly concerned with the national capacity to cope with foreign languages they need to address the world of proprietary language instruction. There are a number of issues that will have to be faced. First, there is an almost complete lack of statistical information that would be required to manage adult foreign language education: how many people are studying which languages, where, to what level, to what purpose, and in what kind of learning environment. Second, how can the overall profile of language learning be managed so as to minimize overlap, fill in lacunae, and inhibit the lemming effect of all programs rushing to fill only the most popular requirements. The market-driven system tends to concentrate instruction in the beginning levels of language instruction and in the languages for which there is the greatest demand. How can instruction in the upper skill levels and in the less popular languages be encouraged and managed? Third, how is and should financial support be assured? The current market competition-based system has the advantage that it is flexible and responsible to adult demand, but it makes for high turnover, and the constant threat of discontinuation. Fourth, there is almost no attention paid to teacher preparation and professionalization. The dependence on part-time, short-contract, untrained native speakers makes the upgrading of the skills of the tutors difficult. And fifth, teaching materials focused specifically on adult learners, particularly those with some prior exposure to the language, or experience in learning another language are difficult to find. Moreover, existing teaching materials tend to focus on a fixed

Adult use and language choice in foreign language policy

progression to over-all fluency while the goals of adult learners may be more focused and passive. There is little thought given to the diffusion of the opportunities in adult language throughout the country instead of exclusive concentration in one or a few major cities. In short, there are many relevant policy issues relating to organization and management of this important part of a country’s language teaching system that should be faced. Fifth, an important part of any country’s capacity to provide its citizens with foreign language instruction resides overseas. Whether language training abroad should come early or later in the learning sequence is a matter of some debate. What is clear is that the development of high-level, usable skills demands some exposure to use in the countries where the languages are spoken. However, currently the system of overseas language training opportunities is almost entirely uncoordinated. There is some control and certification of this overseas language learning when it is affiliated with the formal education system. However, overseas programs serving adults are largely unregulated and uncoordinated. Nor is their any planning for the role they play within the larger foreign language educational system. There is another aspect of adult use of foreign languages that comprises an important and growing part of a country’s foreign language system that should be brought into deliberate national planning. Clearly, not everyone in a society is likely to have gained, let alone retained, a high competency level in all of the languages he or she is likely to need to deal with as an adult. Foreign language education for adults is obviously only one way of meeting those needs, but, and in view of the demands on time and effort and the limited use that is likely to be made of foreign languages by most people, it is an inefficient way of solving this problem. This situation has led to the creation of a set of what may be called “foreign language enablers”, that is, facilities for making cross-language communication possible where the parties do not have full competency in the language(s) of communication. The classic way of meeting these needs is through the employment of human interpreters and translators. National foreign language planning rarely includes planning for the production and use of such specialists. Nor is it generally clear whether such people are to be exclusively drawn from among native speakers of a language or those trained in the formal or adult language educational system. In recent years, the growing international mobility of large segments of the public has increased the demand for human translators. Indeed, several corporations have established lists of certified interpreters at widely dispersed locations throughout the world who would be available for employment to

181

182

Richard D. Lambert

travelers or to businesses. The alternative response is to concentrate translation services at centralized locations. In the United States, the principal long distance telephone company, AT&T, has established a center in Monterey, California to provide instant interpreting services in some 30 or so languages to clients on immediate demand. In Great Britain, 11,000 call centers (Graddol 1998) handle cross-language interpretation requests for individual corporations at centralized locations. For instance, IBM established its centralized helpline at Glasgow, where it maintains a staff knowledgeable in both electronics and languages to handle calls for assistance. For more general use, a half dozen computer-based programs are now available with voice recognition capacity to provide for or assist in interpreting between more than a score of language pairs. Similarly, the need for and supply of translation services have expanded as well. In the European Community alone with its need to translate all official documents into the expanding number of national and some of the regional languages of its member states has created an immense demand. Similarly, the practice of some newspapers and magazines of transmitting current publications throughout the world via satellite and having them locally translated for domestic consumption creates a huge demand for translation services. In part as a response to this expanded demand, translation services have become more mechanized and commoditized. Machine translation, still an imperfect art, is becoming more widely used. In some domains with restricted vocabulary, such as weather forecasts, scientific communication, or scanning international patents, machine translation is reasonably adequate, particularly if supplemented by a skilled human translator. Several computerbased translation services are available for a general audience. As the internet becomes more important as a vehicle for international communication, instant translation facilities are becoming increasingly more commonplace. While these machine translation and machine interpreting devices are still only approximate in accuracy, they expand immensely the ability of people to communicate across languages. The general point that should be made is that national foreign language policy serving the foreign language needs of adults should include planning for these “language enablers” as well. In the future, they are as likely to meet the needs of adult foreign language use as are the competencies most students gain during their school years. Indeed, it can be persuasively argued that language instruction in the schools should include education in how to use these language enablers as well as providing a limited competency in one or a few foreign languages. This discussion leads back to the more general question of the relationship

Adult use and language choice in foreign language policy

of adult use, adult education, and language enablers to the domain where almost all of the attention of language policy makers is concentrated — language learning in the formal education system. As noted at the outset, the pattern and extent of adult use determines the importance of foreign language education in a particular society. Where adult use is low, foreign language courses tend to be fewer, the number of languages learned by individual students is lower, student motivation to study a language is lower, the level of competency which students reach tends to be lower, and the choice of languages which students choose to study is more limited and can be random and fluctuate substantially over time. In the past when the accepted purpose for foreign language learning was humanistic — it made for a cultured person — there was no necessary link between language acquisition and language use. With the shift to a communicative goal for language learning, the presumption of probable use is stronger. Particularly in monoglot, insular societies such as those described by Trim (the United States and Russia are classic examples), a low level of adult use makes it difficult to convince individual students that they should invest the time in learning a foreign language. An extreme statement of the consequences for a language education system of this lack of fit between adult use and student motivation is illustrated in a lecture given by Bjorn Jernudd in New Delhi (Jernudd 1990): The enormous costs of educating all in a public school system in a foreign language is indefensible when only some ever benefit from its use, and valuable educational time is thus taken away from other activities that could directly serve a development effort or enhance individual futures, and when foreign language knowledge can easily be alternatively provided for those who eventually need it. Opportunities should be made available [by good planning] at those points in time when the student is motivated to learn and this may not be until business school or as the result of an on-the-job overseas assignment.

This advice would, of course, put the entire burden of foreign language education on the adult sector. Few, if any, countries accept this advice, except perhaps where instruction in the less commonly taught languages is concerned, where it does describe the de facto policy of many nations. Rather, the trend in most countries, including all of the European countries (Bergentoft 1994; Dickson and Cumming 1997) is to expand, not displace, school-based foreign-language instruction. Most countries are moving the starting age for foreign language study earlier to allow for more time to acquire the competencies, and they are extending the language study requirement to more and more types of students. The United States is a conspicuous exception to that trend. While some states

183

184

Richard D. Lambert

are introducing foreign language instruction into primary school, general foreign language enrollments are static at best, and language and literature departments on many university campuses are in real danger of severe contraction (Lambert 1999b). I do not propose to discuss here the many policy issues that must be addressed in language instruction in the formal education system. The point I want to make is that the link between that system and adult language learning and use should be strengthened. An effective national foreign language plan should include all levels and forms of instruction and use.2 Special attention should be paid to what Americans call articulation — that is, links between various layers of the language education system. The British more properly call this progression — that is the path that serves individual language learners throughout their lives. There is some move in this direction with the Council of Europe’s concern for the development of skills for self-directed, life-long learning skills, and its use-oriented Common European Framework. Several of the Council of Europe’s international workshops have dealt specifically with adult and vocational language learning. However, all too often, language education policy concentrates exclusively on instruction within schools and to a lesser extent colleges and universities. It is as if for most language policy makers the national concern for foreign language stops with the diploma. What is needed is an integrated system, planning for the variety of foreign language education venues, encompassing adult use as well as adult learning, and including the foreign language enablers. The school-based system should not only give students a start toward a language competency, but some preparation for adult learning and use, and for ways of coping with language differences when competencies are limited or non-existent. Facilities should be developed to assist learners in up-grading or retaining skills lost with aging (Lambert and Freed 1982; Weltens, de Bot and van Els 1986).

Language choice The other major area of overlap between domestic and foreign language policy is the decision of which languages to teach. Most discussions of status language policy concern the selection of which languages are to be taught and used for various official purposes. There is, however, surprising little theory of language choice in foreign language policy, although one would expect it to be an issue

Adult use and language choice in foreign language policy

of paramount importance. Indeed, it should be one of the principal linkages between adult language use and foreign language policy. John Trim in his seminal article on factors affecting the extensiveness of foreign language systems quoted at the beginning of this article (Trim 1994) also proposes a set of factors influencing which foreign languages will be highly valued. A country is more likely to value proficiency in a particular foreign language highly if that language: 1. Is widely used as a vehicle of international communication 2. Is spoken by a country (or countries) with which there are active interpersonal, industrial, commercial, leisure, entertainment, touristic relations 3. Has a positive image 4. Is spoken by high prestige groups.

There are few other statements of this general nature on the determinants of language choice in foreign language policy. In the language plans of some individual countries, there are explicit statements concerning why particular languages are to be taught. The Dutch limited the languages to be taught in compulsory education to German, English and French. They based their selection on the results of surveys of adult use, anticipated business requirements, and the demands of citizenship in an increasingly united Europe. Australia, based on its geographic location, deliberately shifted its emphasis in language instruction away from the European to the Asian languages, in effect, going in a direction away from current patterns of language use toward desired use. While explanations of language choice such as Trim’s do capture very generally the determinants of language choice, it is also useful to examine more specifically some of the elements which influence which languages will be offered. In general, language choice will be made according to the following priorities: (1) the demands of domestic language policy; (2) languages of wider communication; (3) other foreign languages commonly taught; and, (4) languages introduced through a desire for diversification.

Priority given to corpus and status language policy The first priority in language choice reflects the demands of corpus and status policy with respect to domestic languages. The highest priority goes to instruction in the national language(s) (Coulmas 1988), then to the languages of ethnolinguistic minorities (Lambert 1999b; Paulston 1996). Foreign languages are allocated curricular time only after the demands of domestic language policy

185

186

Richard D. Lambert

are met. Such demands are easily met in monolingual countries, but in many countries there are questions about the number of national languages to be recognized. Some countries, like France, are quite insistent that the country should have only French taught in schools, but there is continuous agitation at the margin by such groups as the Bretons and the Basques to accord official language status to their languages. Others like Switzerland, Cyprus and Belgium give two or three territorially distinct languages equal status as official languages. The tendency in such countries is to treat the different language regions as monolingual, giving the language of each region sole official status within its domain. Out of 26 cantons in Switzerland, only 4 are officially multi-lingual. They allow several languages to be taught as official languages. While municipalities differ in their language choice, within the municipality only one language is taught. In Cyprus, the forced separation of Turks and Greeks in 1974 led to a fully bifurcated language educational system. Such countries also differ in the extent to which they actively pursue bilingualism. In Switzerland, the first “foreign language” taught is the official language of a canton different from one’s own. The official language of the other cantons can be the second or the third “foreign language” to be learned. In Cyprus, Greek is the official language in the Greek part of the island, Turkish in the Turkish part. A long tradition of bilingualism, particularly among the Turks, is dying out. Canada has a long history of deliberately trying to promote competency in both French and English among all its citizens, but the trend toward Quebec separateness is threatening to produce a monolingual Quebec and an undermining of the bilingual education tradition in the rest of Canada. Belgium provides an interesting case not only of bifurcation but of the consequences of a shift in relative status between the Walloons and the Flemings. Countries also differ in how full an official status they give to regional languages. Increasingly, languages such as Catalan in Spain achieve nearsovereign status within their geographic domain. Basque and Galician are moving in the same direction. The autonomous regional governments in Spain differ in the way they treat language diversity. Some establish complete educational systems using the dominant language of the territory. Others allow for parallel educational systems, and still others make the teaching of minority languages compulsory, but as a subject matter, not as a medium of instruction. Countries also differ in their treatment of the languages of scattered linguistic units and diverse immigrant groups. The Swiss policy is somewhat extreme. The courts have decided that using a language other than an official language as a medium of instruction is illegal. Other countries allow for a brief period of

Adult use and language choice in foreign language policy 187

transition from the home language to the national languages. The debate can be quite acrimonious between spokesmen for the linguistic groups and the national government over how long that transition period should be, or, more generally, whether education in the home language should serve only as a transition or as a device for longer-term cultural maintenance. The political debate in the United States over bilingual education (Tatalovitch 1995) is precisely over this question. More generally there is a durable tension between the national desire for homogeneity and the minorities’ concern for the maintenance of their cultural integrity. The movement of Sami in Finland and Maori in New Zealand illustrate a typical passage over time along this continuum from a strategy of transition to one of cultural maintenance. Linguistic complexity is more manageable when there are only one or a few territorial languages, plus a relatively small scattering of other minority languages. Where the number of important domestic languages is large — more than half of the countries in the world contain at least five important linguistic minorities, language choice is even more complex. This is particularly true when a country is trying to struggle for, perhaps even invent, a lingua franca. The case of the Philippines is instructive. It has eight major languages plus 156 minor languages. It has declared Filipino to be the official language. Filipino is a language based on Tagalog, and like Indonesian is in the process of creation. For an unspecified period of time English is also an official language in the Philippines. The concern within the Philippines, interestingly enough, is whether the eventual abandonment of English as an official language will diminish the scholastic achievement of students in both languages. In Hawaii a similar argument by the autochthonous population stopped a drive by ethnic leaders to make Pidgin a co-official language. Similarly, the argument against proposals for a language maintenance strategy among Hispanics gave impetus to the English Only drive in California and elsewhere. India has some of the same problems as the Philippines. Originally it had fourteen official languages. Now there are eighteen. The central government is working to install Hindi, a sanskritized version of Hindustani, as a lingua franca. After much debate India developed a three language policy (Khubchandani 1997): instruction first in a mother tongue, usually defined as the official language of the state; then Hindi or English, supplemented by another Indian or foreign language. This somewhat demanding linguistic fiat is often not implemented, particularly in South India where the languages are of non-Indo-Aryan origin and where Hindi is not in common use. Moreover, for the large number of Indians whose mother tongue is a local language or a dialect that is not an official language of a state,

188

Richard D. Lambert

the formula means the study of four languages. Many countries in Africa are even more polyglot than India. Nigeria has more than a hundred languages and dialects, and must balance them at various levels of the educational system. One common way of dealing with these expanding demands is to layer language choice, that is to teach different languages at different levels of the educational system. Normally, the more localized languages are used as media of instruction at the lower levels; then a shift is made to an existing or an advocated lingua franca, as is the case with Hindi in India, Swahili in East Africa, or Putunghua in Hong Kong and parts of China. The general point is that the national, regional, and other minority languages identified by the domestic language agenda use up a considerable amount of the curricular space available for language instruction. Teaching other languages must fit into the remainder.

Instruction in one of the languages of world communication Once the needs of domestic language policy are met, the next priority in language choice is usually the teaching of the languages of international communication. In the ex-colonial countries this is most commonly the language of the formerly dominant metropolitan country, usually French or English. In the former colonies of Africa and Asia, these languages have become de facto domestic as well as foreign languages (Tengan 1994). They remain the languages of the ruling elites, and are used in elite schools from the primary grades onwards. One of the social functions of this arrangement is that it is unnecessary for the children of the elite to make the difficult transition from the local vernacular to the world language used in secondary and advanced education. Increasingly, these languages, particularly English, have become in a domesticated form a quasi-lingua franca well down in the class hierarchy, and are seen as a means of upward mobility for middle class children. These advantages support the continued use of the ex-colonial languages in the face of immense political pressures within the countries for nativization of official languages. India, the Philippines, Hong Kong, the former ex-colonial countries of Africa illustrate this situation. Former colonies aside, throughout the world the overwhelming choice for the first foreign language taught is English. While a remarkable variety of Englishes has grown up in different countries (Gorlach 1991), particularly in lexicon and presentation style, it is standard British or American that is taught in schools, usually the British variety. As Bergentoft notes (Bergentoft 1994: 24):

Adult use and language choice in foreign language policy 189

Despite the variety offered, English is by far the predominant first language in all the countries where it is not the mother tongue. Even in Japan, where the study of a foreign language is technically optional, 98% of the students choose English. In the Netherlands 100% of the students study English in primary school and 88% in secondary school; in Norway it is 99% and 69%; in Sweden it is 99% and 100%. Even in France, where the cultural bias in favor of the mother tongue is especially strong, 78% of students study English in primary school and 94% in secondary school.

The widespread choice of English as the first foreign language is a reflection of the widespread use of English by adults as a lingua franca of international communication. For instance, English is the language of the scientific research community. An analysis of 896,740 scientific publications indexed in the Institute of Scientific Information data base (Garfield and Welljams-Dorof 1990: 28–29) which covers all science research publications in any language throughout the world, indicated the extent of the dominance of English. That analysis enumerated by country both the language in which their scientific publications were presented, and which language materials were most often cited by other sources. Their overall conclusion was: The data presented here document the predominance of English as the primary language of international research. More source items are published in English by both the native and non-native speakers than any other language, and they have the highest impact. Also, most major scientific nations, regardless of their native language or languages, cite the English native-language literature almost exclusively.

A similar point could be made of scientific conferences. Moreover, until recently English was almost exclusively the language used on the internet, and in a substantial portion of the international entertainment culture. There is a considerable debate about the origins and equity of this situation (Phillipson 1992), but one consequence is to make English the first choice for languages offered in schools throughout the world. David Graddol (Graddol 1998) estimates that there are 375 million individuals in the world for whom English is their first language, an equivalent number for whom it is their second language, and perhaps 750 million who study English as a foreign language. The ubiquity of English in the Netherlands is so overwhelming that some Dutch language planners have concluded that most students will learn that language in any event. Accordingly, a suggestion has been made by Dutch language policy makers that there is no need to continue to make the study of English compulsory in Dutch primary and secondary schools (de Bot 2000). They

190

Richard D. Lambert

propose that one of the European languages should be substituted. However, the hegemony of English continues in the Netherlands, as elsewhere. And there, as well as elsewhere, its widespread use by adults gives it a supportive envelope within which English can be taught most effectively.

Choice of other foreign languages The widespread study of foreign languages, other than the languages of world communication is, in the main, a European culture pattern. Another priority claimant on language choice is what might be called the “significant other”, that is, the language of a country or language that has, or has had, a special relationship. Sometimes this has been historically a quasi-colonial relationship as with Russian in East Europe, or Swedish in Finland. Sometimes it reflects a larger culture zone, as is the case with German in East Europe. Sometimes it reflects the importance of an historical language as in the case of classical Arabic in North Africa, of Sanskrit in India. Such significant other languages are frequently the second foreign language studied. Of special interest are the occasions when a significant other language loses its political backing. Sudden political shifts that led to the dropping of Russian and the substitution of English in China, or substituting English for German in Eastern Europe are cases in point. It will be interesting to see what happens to French in western Canada if Quebec ever becomes a separate nation. Once the demands of domestic language policy and significant others are met, in many countries it leaves little time for the choice of other languages. However, within the remaining curricular space the selection of which foreign languages to offer must be made. Moreover, the teaching of a variety of other foreign languages is primarily a Western European cultural pattern. In the United States, without the European expectation of substantial use by adults, foreign language education remains largely a reflection of this historical culture pattern. The teaching of other foreign languages is not widespread throughout the rest of the world. Even in Europe, the teaching of foreign languages other than English is given the status primarily as second and third languages, with consequent decrease in the amount of time spent and the proportion of the student body required to learn them. Second and third foreign language usually appear only in secondary schools. The proportion of students studying a foreign language other than English in primary schools rarely reaches ten percent. Bergentoft (Bergentoft 1994: 29) summarizes the situation of language prioritizing in Europe as follows:

Adult use and language choice in foreign language policy

From the academic year 1992–93 onward, the study of at least one foreign language is obligatory in all countries examined with the exception of Japan. Several countries require knowledge of two foreign languages for entry to the upper secondary school: in the Netherlands the requirement is three foreign languages. In a number of countries there is currently a debate on whether the study of a second foreign language should be made obligatory. The number of students enrolled in a third language differs by country. In the Netherlands it is a high of 85 per cent; in Finland 77 per cent of upper secondary students study a third foreign language. In other countries, the percentage is very much lower. Moreover, the number of hours dedicated to the learning of third languages is lower, rarely passing 300 hours, and they are usually limited to those on an academic rather than a vocational track.

The most commonly taught languages in Western Europe are German and French, but even these are taught to only a minority of the students. Of those reporting in the International Study for Evaluation of Educational Achievement (IEA) comparative survey of language instruction in 35 countries (Dickson and Cumming 1996) the median proportion of students in compulsory education who were enrolled in German was 43 percent, while 14 percent, were enrolled in French. Bergentoft (Bergentoft 1994) reports a similar pattern. Few northern Europeans study southern European languages. France is an exception: 46% of the students in compulsory education study Spanish (Dickson and Cumming 1996). Students in Portugal and to a lesser extent Italy, study French. The language choice situation in English-speaking countries is quite different. First, the percent of students studying foreign languages is lower. In England, since 1991 every student between ages 11 to 16 must study one foreign language. Students aged 11 to 14 take on average 12 hours of foreign language instruction per week and students aged 14 to 16 take 90 minutes. However, by upper secondary school only about ten percent of the students take at least one language. Language choice tends to be limited. While the National Curriculum for England and Wales specifies 19 languages which may be offered, French is the language most commonly taken. In 1997 French had 62.1% of the Modern Language GCSE entries, and German had 25.3% (Boaks 1998). The United States, like Britain, is not troubled by the need for students to study English as a foreign language. Moreover, with its even more limited adult use of foreign languages, the United States has not adopted a universal foreign language requirement as has the United Kingdom. All 50 states require that secondary schools offer foreign languages, but none of them require students to have studied a foreign language for graduation. Language instruction is

191

192

Richard D. Lambert

required of students by only about 25% of colleges and universities and less than ten percent of students take any language instruction in college. The overwhelming proportion of students who do enroll in language classes in the United States take only one language. Which language they choose is neither determined by, as is the case in England, nor is guided by expectations of adult use. Hence, language choice in the United States is a constrained free market, reflecting the tidal swings of student preferences constrained by the competencies of the language teaching staff. In the early decades of this century, Latin and Greek were the most common languages taught in the United States, followed in subsequent decades by German and French, in that order. More recently, French enrollments declined from 34.4% of all foreign language enrollments in 1968 to 18.0% in 1995. In 1968 German had 19.2% of all foreign language enrollments. By 1995 they had dropped to 8.5%. The largest increase in language enrollments in the United States has been in Spanish. Spanish has grown from a 32.4% share in 1995 to 53.2% share in 1995. The causes of these shifts over time are somewhat obscure, since most students give non-instrumental reasons for studying foreign languages in the United States and few students ever get enough competency to use them effectively in adult life. The implications of this great shift in enrollments have been quite troublesome for American language teachers (Lambert 1999b), particularly at the collegiate level where departments, professorships, and fellowships are in jeopardy.

Diversification The emphasis, particularly in Europe, on learning English and another language of the European Community has raised concern about the narrowing of the range of languages taught to only a few languages. Important parts of the world — all of Asia and Latin America — where relationships in the future are likely to become increasingly important are left out. Moreover, even within Europe the tension between giving equal status to a large and expanding number of official languages and the limited number of languages in the curriculum is a matter of great concern. The narrowing of language choices in the midst of an increasingly varied linguistic world is a cause of concern for language policy makers. Diversification is one of the central themes in the development of the Council of Europe’s language projects, but it refers primarily to the variety of languages within Europe. Several countries outside of Europe have opted for diversification as well. Australia faced the issue head on and has instituted a shift away from

Adult use and language choice in foreign language policy 193

language choice based on its mostly-European ethnic composition. A major effort has been made in Australia to develop Asian, particularly Japanese, language instruction, so much so that Japanese has the highest number of enrollees in Australian secondary schools. The United States has had three decades of language policy promoting diversification. In one of the few examples of centralized foreign language policy making in the United States there have been continuous efforts to expand the small — still less than ten percent — but significant and growing enrollments in languages other than those of Western Europe, in both the secondary schools and universities. The federal government has invested $30 million a year in the expansion of teaching programs in what are called the less commonly taught languages, the languages of Africa, Asia, Eastern Europe, and the Middle East. A number of private foundations and overseas governments have invested heavily in the maintenance of our teaching capacity in the less commonly taught languages in secondary schools, universities, and now primary schools. Enrollments in individual languages have fluctuated over time as public interest in the countries where the languages are spoken has varied. Russian had a brief surge in enrollments in the 1980’s but there has been a substantial decline in the last few years. Enrollments in Japanese had surged in the early 1990’s but they seem to have leveled off. Chinese now seems to be a rising star. In the United States, diversification is helped along by the fact that if there is little expectation of the use of a school-learned foreign language, why not study one that is a bit unusual?

Conclusion In this paper I have been able to discuss only two aspects of foreign language policy: adult language use and language choice. These obviously are only two of a large number of issues which foreign language policy makers must face. Indeed, both national plans and the day-to-day administration of foreign language instruction tend to concentrate on quite different topics. They are concerned with such matters as: when should language instruction begin, how long should it last and at what levels of the educational system, what kind of students should be enrolled, how centralized or dispersed should be the management, what are the proper modes of pedagogy and assessment, how are teachers to be trained and certified, and how are text materials to be produced and controlled? Hopefully, a full theory of foreign language policy will include all of these topics, and, equally hopefully, the interconnections among them.

194

Richard D. Lambert

Notes 1. Linguatel reported that in Great Britain “75% of incoming foreign calls to a group of London-based companies were lost at the switchboard through incomprehension”. 2. The Nuffield survey does cover the full range, including adult learning and technological enablers. See Nuffield Inquiry: Where are we going with languages? Nuffield Foundation, 1998. The Dutch National Action Plan also covers adult language learning.

References Adelman, C. 1994. “The borderless career: Corporate hiring in the 90s”. International Educator 12 (15): 32–33. Bergentoft, R. 1994. “Foreign language instruction: A comparative perspective”. In Foreign Language Policy: An Agenda for Change, R. D. Lambert (ed.), 17–46. Philadelphia: American Academy of Political and Social Change. Boaks, P. 1998. “Languages in schools”. In Where Are we Going with Languages? A. Moys (ed.), 34–43. London: Nuffield Foundation. Boergarts, T. and de Bot, K. 1997. Perspectives on Foreign Language Policy: Studies in Honour of Theo van Els. Amsterdam: John Benjamins. CILT. Foreign Languages in Business and Industry. Information Sheet 46. Coulmas, F. (ed.). 1988. With Forked Togue, What Are National Languages Good For? Singapore: Karoma. De Bot, K. 2000. “An early start for foreign languages, but not English, in the Netherlands?” In Language Policy and Language Pedagogy, R. Lambert and E. Shohamy (eds). Amsterdam: John Benjamins. Dickson, P. and Cumming, A. (eds). 1996. Profiles of Language Education in 25 Countries. Slough, UK: National Foundation for Educational Research. Djité, P. G. 1994. Foreign Language Policy to Language Planning. Deakin: National Languages and Institutes of Australia. Emmans, K., Hawkins, E. and Westoby, A. 1974. The Use of Foreign Languages in the Private Sector of Industry and Commerce. York: University of York Language Teaching Centre. “Euro-tongues wag in English”. 1997. The Economist. October 25: 60. Fishman, J. A. 1969. “National Languages and Languages of Wider Communication in the Developing Nations”. Anthropological Linguistics 11: 111–135. Fishman, J. A. 2000. “The status agenda in corpus planning”. In Language Policy and Language Pedagogy, R. Lambert and E. Shohamy (eds), 43–52. Amsterdam: John Benjamins. Fixman, C. 1990. “The foreign language needs of U. S.-based corporations”. In Foreign Language in the Workplace, R. Lambert and S. Moore (eds), 25–46. Philadelphia: American Academy of Political and Social Science.

Adult use and language choice in foreign language policy 195

Garfield, E. and Welljams-Dorof, A. 1990. “Language Use in International Research: A Citation Analysis”. In Foreign Language in the Workplace, R. Lambert and S. Moore (eds), 10–24. Philadelphia: American Academy of Political and Social Science. Gorlach, M. 1991. Englishes: Studies in Varieties of English 1984–1988. Amsterdam: John Benjamins. Graddol, D. 1998. “Will English be enough?” In Where Are we Going with Languages? A. Moys (ed.), 24–33. London: Nuffield Foundation. Hagen, S. 1998. “What does global trade mean for UK languages?” In Where Are we Going with Languages? A. Moys (ed.), 14–23. London: Nuffield Foundation. Hagen, S. (ed.). 1993. Languages in European Business: The FLAIR Report. London: CILT. Hagen, S. 1994. “Language Policy and Planning for Business in Great Britain”. In Foreign Language Policy: An Agenda for Change, R. D. Lambert (ed.), 111–130. Philadelphia: American Academy of Political and Social Change. Ingram, D. E. Forthcoming. Languages in Australian Industry. Jernudd, B. 1990. Lectures on Language Problems. New Delhi: Bahri. Khubchandani, L. 1997. “Language policy and education in the Indian sub-continent”. In Encyclopedia of Language and Education, Vol I., R. Wodak and D. Corson (eds), 179–187. Dordrecht: Kluwer. Lambert, R. and Freed, B. (eds). 1982 The Loss of Language Skills. Rowley, MA: Newbury. Lambert, R. and Moore, S. (eds). 1990. Foreign Language in the Workplace. Philadelphia: American Academy of Political and Social Science. Lambert, R. D. and Moore, S. 1990a. “Foreign language use among international business graduates”. In Foreign Language in the Workplace, R. Lambert and S. Moore (eds), 47–59. Philadelphia: American Academy of Political and Social Science. Lambert, R. 1991. A National Plan for a Use-Oriented Foreign Language System. Washington, D. C.: National Foreign Language Center Occasional Paper. Lambert, R. (ed.) 1994. Foreign Language Policy: An Agenda for Change. Philadelphia: American Academy of Political and Social Change. Lambert, R. 1994a. “Language policy: An overview”. In Foreign Language Policy: An Agenda for Change, R. D. Lambert (ed.), 4–21. Philadelphia: American Academy of Political and Social Change. Lambert, R. 1994b. Language Planning Around the World: Contexts and Systematic Change. Washington, National Foreign Language Center. Lambert, R. 1999a. “A scaffolding for foreign language policy”. International Journal of the Sociology of Language 137: 3–25. Lambert, R. 1999b. “Winds of change in foreign language policy”. International Educator 8: 31–35. Lambert, R. and Shohamy, E. (eds). 2000. Language Policy and Language Pedagogy. Amsterdam: John Benjamins. Languages Lead Body. 1992. National Language Standards. London: LLB, CILT. Lo Bianco, J. 1987. National Policy on Languages. Canberra: Australian Government Publishing Service. Moys, A. (ed.). 1998. Where are we going with languages? London: Nuffield Foundation. Nikki, M.-L. 1992. The implementation of the Finnish national plan for foreign language teaching. Jyväskylä: University of Jyväskylä.



196

Richard D. Lambert

Paulston, C. B. 1996. ”Linguistic consequences of ethnicity and nationalism in multilingual settings”. In Language and Education in Multilingual Settings, B. Spolsky, (ed.), 119–144. London: Multilingual Matters. Phillipson, R. 1992. Linguistic Imperialism. Oxford: Oxford University Press. Pütz, M. 1994. (ed.). Language Contact and Language Planning. Amsterdam: John Benjamins. Sajavaara, K. 1997. “Implementation of foreign language policy in Finland”. In Perspectives on Foreign Language Policy: Studies in Honour of Theo van Els, T. Bongaerts and K. de Bot (eds), 113–128. Amsterdam: John Benjamins. Schiffman, H. 1996. Language Culture and Language Policy. London: Routledge. Spolsky, B. 1996. (ed.). Language and Education in Multilingual Settings. London: Multilingual Matters. Spolsky, B. and Shohamy, E. 2000. “Language practice, language ideology and language Policy”. In Language Policy and Language Pedagogy, R. Lambert and E. Shohamy (eds), 1–42. Amsterdam: John Benjamins. Swanton, P. 1988. “Adults learning languages”. In Where Are we Going with Languages? A. Moys (ed.), 54–61. London: Nuffield Foundation. Tatalovich, R. 1995. Nativism Revisited? The Official English Movement and the American States. Lexington, Kentucky: The University of Kentucky Press. Tengan, A. B. 1994. “European languages in African society and culture: A view on cultural authenticity”. In Language Contact and Language Planning, M. Pütz (ed.), 125–138. Amsterdam: John Benjamins. Trim, J. L. M. 1983. “The Place of Needs Analysis in the Council of Europe Modern Language Project”. In Research into Foreign Language Needs, T. van Els and M. Oud-de Glas (eds), 223–242. Augsburg: Universität Augsburg. Trim, J. L. M. 1994. “Some factors influencing national foreign language policymaking in Europe”. In Foreign Language Policy: An Agenda for Change, R. D. Lambert (ed.), 1–16. Philadelphia: American Academy of Political and Social Change. Tuin, D. and Westhoff, G. 1997. “The task force of the Dutch national action programme as an instrument for developing and implementing foreign language policy”. In Perspectives on Foreign Language Policy: Studies in Honour of Theo van Els, T. Bongaerts and K. de Bot (eds), 21–34. Amsterdam: John Benjamins. Van Els, T. and Oud-de Glas, M. (eds). 1983. Research into Foreign Language Needs. Augsburg: Universität Augsburg. Van Els, T. et al. 1990. Horizon Taal: Nationaal Actieprogramma Moderne Vreemde Talen: Nota van Aanbevelingen. Zoetermeer: Ministerie van Onderwijs en Wetenschappen. Van Hest, E. and Oud-de Glas, M. 1991. A Survey of the Techniques Used in the Diagnosis and Analysis of Foreign Language Needs in Trade and Industry: Lingua. Human Resources, Education, Training and Youth. Commission of the European Communities. Verdoodt, A. and Sente, A. 1983. “Interest shown by secondary school pupils in modern languages and adult language needs in Belgium”. In Research into Foreign Language Needs, T. van Els and M. Oud-de Glas (eds), 263–283. Augsburg: Universität Augsburg. Weltens, B., de Bot, K. and Van Els, T. (eds). 1986. Language Attrition in Progress. Dordrecht: Foris. Wodak, R. and Corson, D. (eds). 1997. Encyclopedia of Language and Education. Dordrecht: Kluwer.



Linking levels in foreign language teaching through a common frame of reference Kees de Bot and Theo van Els University of Nijmegen

In his discussion of the architecture of the foreign language teaching system in the US, Lambert (1994: 49) indicates that … (the US system) is unique among countries in that a large portion of our basic language instruction takes place at the college or university level. In most other countries, such instruction is provided almost exclusively in secondary schools, hence problems of articulation are less pressing.

Also Bergentoft (1994: 33) in his comparative analysis of foreign language instruction in various countries is rather optimistic about articulation between levels in the system: “Continuity in curricula across levels prevents students who have already studied the language for some time from having to start at the beginning”. While this last point may be true generally, it is definitely not the case that in most countries this fit between levels and school types is seamless. Speaking from the perspective of a country in which languages are taught at different levels, Van Els (1994: 40) comments on the role different parts have to play: “In the Netherlands, despite the magnitude of the need for foreign languages, policy proposals have traditionally ignored the existence of other sectors, usually resulting in overcharging secondary education in particular”. Bergentoft’s (1994) analysis of language policy suggests that in comparable countries like Denmark and Norway, articulation between levels is high on the agenda. In this article we want to discuss the position of tertiary education in the architecture of the foreign language teaching system in the Netherlands and the specific problems of articulation between pre-university education and university education in this respect. The “fit” between levels can be defined along two axes: one which is concerned with goals and levels of proficiency, the other with methods of instruction. For the latter we have to establish what the differences in methods

198

Kees de Bot and Theo van Els

of instructions are, and to what extent such differences are problematic. For the former, which will be the main focus of the present article, we need to know what are the goals of language instruction and the degree of proficiency aimed at for a given level, or in other words what the desired output of that part of the system is. If entrance to language related programs at the university were the only or most logical step in secondary students’ educational career, the output of one part of the system should ideally have a perfect fit with the next part. In order not to lose time, language teaching at the next level should start where the preceding part ended. In reality there is no such direct relation between secondary education and language related tertiary curricula. Secondary education has to prepare students for many other types of education, both within the tertiary system and in-company.

Articulation and fit in the system If we look at the structure of the educational system in various countries, the global picture presented in Figure 1 emerges. We see three joints in the system where a fit between levels is needed: between primary and lower secondary, between lower and higher secondary and between higher secondary and tertiary. There are clear signs that in the Netherlands there are problems of articulation at all joints: foreign language teachers in lower secondary education complain that teaching English in primary schools has little or no effect (Edelenbos 1993), while teachers in upper secondary complain that the fit between the curricula in lower secondary and upper secondary is far from perfect (Withagen et al. 1996). Universities have indicated repeatedly that at the moment students cannot be asked to read course-books and articles in other languages than Dutch and English, and that even for languages that have been taught for 4 years, like French, proficiency at the end of secondary education is low. Here, we will concentrate on the third joint mainly. The labels used for the levels Figure 1 are very general and they may actually make different situations look more similar than they actually are. It could be argued, for instance, that the levels attained in pre-university education in several European countries is comparable to and probably even higher than those attained in community colleges in the US-system. For the first axis of articulation presented earlier (goals and proficiency levels) we would need a scale that can cover the whole range of proficiencies in

Linking levels in foreign language teaching 199

primary education/FLES

1

lower secondary

2

upper secondary

3

tertiary secondary

Figure 1.

Articulation in the educational system

the system. As Spolsky (1994: 235) points out: “A quasi-absolute scale, expressed in functional terms, is an attractive and intelligent way to specify the goals of a national language policy”. This leads us to the main purpose of this article. In discussions on articulation between levels, a common system of reference is needed. Without such a system it remains unclear what the different participants in the discussion mean when they use terms like “proficiency”, “fluency” and “functional competence”. The data presented here form part of larger project aiming at applying one specific system of reference in the discussions on articulation. We want to find out to what extent the scales developed on the basis of the European Language Framework (North 1992, Council of Europe 1996) may serve this purpose.

Tertiary education as part of the foreign language teaching structure Before we discuss the findings from a pilot study, we need to clarify the role

200 Kees de Bot and Theo van Els

tertiary education can, or has to, play in the architecture of FLT, in particular with regard to the development of highly proficient speakers of those FLs. Do we need people who have studied the language at university if the goal is to acquire a near-native command of FLs? Is language proficiency in a country a pyramid in which both a wide basis of many incipient learners and a small top of experts is needed, or — to use Richard Lambert’s terms — what is the shape of the (ideal) national language capacity? What role would this FL elite have to play? It could be argued that in order to have sufficiently high levels of proficiency at the base we need to have full expertise at the top. Language proficiency would then cascade from the top to the base. The main point here is whether we want to have a system of FLT that will generate near-native speakers that have as such a role to play, or that those functions may be served more effectively by native speakers of that language. The question of the articulation of FLT between secondary and tertiary education starts with a definition of the aims of FLT in a given country or community. It is ultimately the question of what we need (very) high proficiency in non-native speakers for. We want to argue that a high proficiency in itself is not a sensible goal, it always has to be instrumental for another purpose, like studying the linguistic or literary aspects of the language, negotiating in that language or to be able to read and understand everything being said or written in that language (interpreters/translators/intelligence/secret services) or even to pass oV as native speaker (the famous spy-argument). Therefore, curricula that only have a near-native command of the language as their goal are basically pointless. The line of argument could then be whether for those professions in which near-native command is called for, government sponsored parts of the educational system are the most likely providers. If a multinational wants to have engineers that can communicate with Japanese colleagues in Japanese, it could be argued that those skills are part of the (in-company) training such staff should get from their employer. It is simply inconceivable to have programs at technical universities in which foreign languages are taught up to that level in addition to the normal curriculum. Some form of Content Based Instruction might be an option to do at least something to enhance the proficiency, but recent empirical research on the effectiveness of this type of teaching at the university level in the Netherlands (Vink 1997) has cast some doubt on the feasibility of such an approach. In the articulation of secondary and tertiary education five groups can be distinguished from the university’s perspective: (1) Students who opt for the foreign language as a Major or primary subject including students training for

Linking levels in foreign language teaching 201

foreign language teachers. (2) Students for whom the foreign language is an important and integral part of their curriculum (e.g students of Business Communication). (3) Students in programs preparing for a job in which the foreign language will be needed (e.g., students in International Law programs). (4) Students in programs in which knowledge of specific languages is needed (e.g., Italian for Archeology, Latin/Hebrew for Theology, German for Historical Sociology). (5) Students in programs for translators and interpreters. It is not at all clear whether for all five groups the same skills and levels of proficiency are required. The only group for which (near-)native command of the foreign language is the ultimate goal are the translators and interpreters, and even for them a specialization in written vs. spoken skills may be an option: a good translator does not necessarily have to be a good interpreter, since the skills needed for those two activities differ considerably. In addition, there is a tendency to develop specialized programs for each of these groups: technical translators will need skills different from those of literary translators, and interpreters who assist business men in negotiations need skills different from those of court interpreters. For the other groups the needs also vary. For students in group 2 speaking proficiency will be important, while for students of groups 3 and 4 advanced reading skills in the foreign language may be more important. As Van Els (1992: 14) indicates, professional prospects for students taking foreign languages as their main topic are very limited: they can either enter the teaching profession or aim at an academic career. In the last decade unemployment among those highly qualified language specialists has been high. This seems to support the idea that “combination studies” in which one or more foreign languages are an integral part of the curriculum can serve the purpose to educate professionals with sufficiently high levels of proficiency in foreign languages. This suggestion in the Dutch National Foreign Language Plan is also endorsed by the Dutch Ministry of Education. In addition, changes taking place in secondary education at the moment which allow students to opt for specific skills, like reading proficiency, will make the group of students entering universities less homogeneous: some may have advanced listening skills but much less advanced speaking skills. It would be wasteful to submit all students entering language programs to the same programs which are intended to make them more similar, rather than make use of differences between individuals both in terms of knowledge and in terms of strategies in learning. Such an emphasis on autonomous learning is now very much en vogue in secondary education in the Netherlands and elsewhere.1

202 Kees de Bot and Theo van Els

Articulation as a language planning goal In his book Language Planning and Social Change, Cooper (1989: 98) presents what he calls “an accounting scheme for the study of language planning”. His scheme consists of a number of components that can be summarized as follows: “What actors attempt to influence what behaviors of which people for what ends under what conditions by what means, through what decision-making process, with what effect”. The application of such an analytic scheme makes it clear why articulation and fit between levels in the educational system is a complicated issue. It is not clear who should be the actors in this case: the two subsystems of secondary and tertiary education are controlled by two different regulatory bodies which rarely interact, and neither feel the urge to take the lead for this type of developments and go beyond their respective responsibilities. The strategic decision to carve up the educational fields along the lines of the different school types inevitably leads to problems in the border areas. Neither is it clear whose behaviors should be changed and how: Should the secondary school teachers change their aims in teaching and redefine them on the basis of the wishes of the university, or should teachers in tertiary education adapt their teaching approaches to those of secondary education? What means can be used to make the necessary changes? Are the conditions in either or both of the two subsystems such that changes can be implemented fairly easily? In the Netherlands, the autonomy of universities in the definition of their curricula is clearly greater than that of secondary education, due to central regulations and the requirements for the final examinations. There is little formal decision making involved in the striving for a better articulated foreign language teaching system. The kind of changes aimed at are basically beyond central control: it is the local school and universities that should get together to define the problem and find solutions for it. What the effects should be of such actions depends on the goals, which are still to be defined. One of the main themes in Cooper (1989) is that language planning always takes place in a specific context and that language itself in most cases, is not the real issue: language planning quite often has to do with control by elites and economic motives. There are good reasons why schools and universities should aim at better articulation and fit. From the schools’ perspective it is important that a considerable proportion of the pupils go to university, and it is clear that the school should prepare them for this. That will contribute to a school’s status, which will likely lead to an increase in enrollments. Since pupils are free to choose their school, there is increasingly competition between

Linking levels in foreign language teaching 203

schools to attract more pupils. Good connections with universities and having the explicit target to send as many pupils as possible to the university is an asset in many parents’ eyes. For the universities there are various motives of a quantitative nature, besides the more qualitatively based aim of producing graduates with the highest possible level of proficiency. Universities also have to attract as many students as they can, since their income depends largely on student numbers. Good relations with the local schools, from which most students will eventually come, may have a positive effect on the numbers of first year students. In addition, it is important to have both a good system of selection of students in the first year and a high success-rate for those students continuing their studies after the first year. Entering a system in which the teaching approach is totally new, and in which the levels of proficiency of students are not taken into account is likely to have detrimental effects both on the motivation and the academic success of students.2 For all those motives a good fit with secondary education will help. The main problem seems to be that at the moment there is a general feeling of a problematic lack of fit between the levels of secondary and tertiary education, but in fact there is very little hard evidence that can be used to point out what the improved situation should look like. Of the two axes of articulation mentioned earlier, teaching approaches and levels of proficiency aimed at, the latter is clearly the easier one to get to grips with. In the second part of this article we will deal with this in more detail. In the Dutch National Foreign Language Program (van Els and van Hest 1992) there is concern about the level of proficiency at different levels, and in particular about the specification of the knowledge required in higher education: “In any comprehensive foreign language policy it is necessary that the position and the role of foreign languages in higher education — a sector that at the same time ‘consumes’ and ‘offers’ FLT should be specified in more detail”. (van Els and van Hest 1992: 46). What the specific skills required when entering the university are for the groups mentioned is not clear in all respects, and probably there is little awareness at the university level of what is achieved in secondary education with the limited number of teaching hours available in secondary education, as there is little insight into what teachers at the university level actually expect of students who have previously been taught a foreign language for 4 or 6 years.

204 Kees de Bot and Theo van Els

A pilot study on the use of the Common European Framework of Reference In the last part of this article we will present some preliminary data of a project on articulation between secondary and tertiary education in The Netherlands. We contacted language teachers in secondary education and members of staff at the University of Nijmegen. We pointed out to them that, while skills in one or more foreign languages are part of the final examination programs in secondary education, it is not clear to what extent foreign language instruction in secondary education actually leads to the levels of proficiency demanded by the universities. We asked both groups to complete a questionnaire. The questionnaire listed a number of language proficiency scales from the project set up by the Council of Europe mentioned earlier (North 1994, Council of Europe 1996: 153). According to the official report the descriptors in the scales “are relevant to the description of actual learner achievement in lower and upper secondary, vocational and adult education and could thus represent realistic objectives”. Each scale consisted of six levels, represented by a number of activities that call for a specific level of proficiency in the foreign language.3 We asked two things for each scale: what can the students do (“can”), and what should students be able to do in the teachers’/staff members’ opinion (“should be able”). It was pointed out that the scores for “can” and “should be able” might be the same for some statements, but different for others. E.g., it might be the case that in their view students should have a perfect pronunciation of the foreign language, but that they do not actually have such perfect pronunciation. There were separate questionnaires for two foreign languages: English and French, both of which play a central role in the debate on FLT in the Netherlands. In order to get a first impression of estimated language proficiency and desired level of proficiency, we will discuss data that have been gathered from 16 university teachers who teach first year students. Due to time constraints not enough questionnaires from secondary school teachers had been returned to include them in this analysis. The university teachers were asked to fill out the questionnaires with the scales from the Common European Framework for English and French. There were 6 teachers of English from the English Department and the Business Communication Studies Department, and 8 nonlanguage teachers (History, History of Art, Communication). There are also

Linking levels in foreign language teaching 205

data from 2 teachers of French, but their number is obviously too small to treat them as a separate group. The categories used in the scales are the following: (1) Breakthrough, (2) Waystage, (3) Threshold, (4) Vantage, (5) Effective operational proficiency, (6) Mastery. The categories are supposed to represent three types of foreign language users: Basic User (Breakthrough+Waystage), Independent User (Threshold+ Vantage) and Proficient User (Effective operational proficiency+Mastery). In the questionnaire these labels have not been used. None of the teachers involved had previous knowledge of the Common European Framework and the categories and scales used. On the response sheets, the teachers made numerous comments on the scales used. They indicated that they had problems interpreting the scales in many ways: was the use of a dictionary allowed in doing those tasks, was there time pressure, was the interaction mentioned with a native or a non-native speaker and so on. Another comment was that the various can-do statements that were grouped as one step in the scale did not seem to belong together, in the sense that they seemed to represent different levels of proficiency. Also the scales were not always seen as scales: some of the activities mentioned in the higher scales were actually judged to be easier than some in the lower scales. The teachers also indicated very explicitly that they had serious problems judging the level of the language(s) they did not teach themselves. Some of them refrained from using the scales and added general comments such as “students have to able to read the scientific literature in various foreign languages”. They also had much clearer ideas about what the students should be able to do than about what they can actually do. There seems to be very little awareness of problems students might have using the foreign languages. It could also be that teachers avoid using texts in foreign languages on the basis of earlier experiences. This is certainly the case for French, which for the more recent cohorts of students entering university was not an obligatory subject in the last two years of secondary education. In earlier discussions teachers from various disciplines indicated that they no longer asked students to read texts in French and even in German, because this would be too much of a problem for the students. This also means that many teachers in the first year simply have no means to assess what students can do in French. This doesn’t apply to the same extent to English. In most disciplines part of the reading material is in English, mostly because there are no adequate handbooks or introductory texts in Dutch on the market. It should be added that his is generally more so in the second

206 Kees de Bot and Theo van Els

and third year than in the first year. In the later years of their study students are expected to be able to read in English rather fluently. To give an indication of the expected level of proficiency: for the estimation of the students working load, the assumption is that they can read 10 pages of literature per hour in their first language, and five pages per hour in English. Student evaluations on their perception of their work load indicate that this assumption is valid. In order to make some comparisons between languages and groups, a quantitative analysis of the questionnaire data was carried out. We want to stress that those quantitative data are merely indicative of trends, and that it is hard to draw firm conclusions from these data. One of the problems is that calculating means implies that successive categories of the scales are equidistant, i.e. that the distance between the various categories is equal. There are no indications in the description of the framework that this is actually the case. In our description two dimensions are relevant: English vs. French, and what students can do (“Can”) and what they should be able to do (“Should be able”). To give a first impression of the whole group of informants means are given along these two dimensions (Table 1). Table 1. Means for French and English

French English

Can

Should be able

2.3 3.7

3.3 4.2

According to the means in Table 1, students entering university can be labeled as Basic Users for French and Independent Users for English. The difference between French and English hardly comes as a surprise. The dominant position of English in Dutch society is undeniable. What is interesting is that both for English and for French the desired level of proficiency is considerably higher than the estimated level. So even though English has been taught for two years in primary school and for six years in secondary education, the level attained is clearly not high enough to deal with academic language use. For French, the gap between “can” and “should be able” is much larger, but the desired level is much lower than for English. The next step in the analysis was to look at the two larger groups: teachers of English, and non-foreign-language teachers. Tables 2 and 3 present the four cells for these two groups.

Linking levels in foreign language teaching 207

Table 2. Means for French and English for English teachers

French English

Can

Should be able

2.2 3.2

3.4 4.3

Table 3. Means for French and English for non-language teachers

French English

Can

Should be able

2.3 3.7

3.3 4.2

The data in Tables 2 and 3 show that on the whole the two groups react quite similar. Not surprisingly, the English teachers notice a bigger gap between “Can” and “Should be able”. The small differences between the two groups for “Can” suggest that the non-language teachers have a good idea of the students’ proficiency. The teachers reacted differently to the 6 scales. Correlations between scales over subjects range from −.14 (ns) to .80 (p < .001) with not clear picture on what goes with what. The highest agreement is on reading skills, which are consistently between 4 and 5 for both “can” and “should be able”. Obviously correlations were higher between the various “can” scales than between the “should be able” and “can”: for a fair number of scales the gap between “can” and “should be able” was considerable (e.g., French reading: “can”: 2.4, “should be able”: 4.1).

Conclusion The data from this pilot study show that the scales developed for the Common European Framework of Reference can be a good starting point for discussions on articulation between levels. Though teachers at the university level report a number of problems using the scales, the outcomes are consistent enough to develop this line of research further. The next step is to include data from more secondary school teachers and compare their data with the university teachers. An interesting further development could be to compare the Dutch data with data from countries with language education systems that are similar on



208 Kees de Bot and Theo van Els

relevant aspects. A connection with the language policy center at Bar-Ilan would be a useful step in this.

Notes 1. Cf. the work that is being done in the Scientific Commission on Learner Autonomy of the Association Internationale de Linguistique Appliquée. 2. Informal impressions suggest that underestimating the skills of students is at least as detrimental as overestimating them. 3. The scales used were: Interviewing and being interviewed, Reading, Reading fiction for pleasure, Production (spoken), Creative writing and Orthographic control.

References Bergentoft, R. 1994. “Foreign language instruction: A comparative perspective”. In Language Planning around the world: Contexts and systemic change, R. Lambert (ed.), 17–46. Washington DC: The National Foreign Language Center. Cooper, R. 1989. Language Planning and Social Change. Cambridge: Cambridge University Press. Edelenbos, P. 1993. De Aansluiting tussen Engels in het Basisonderwijs en Engels in het Voortgezet Onderwijs. Groningen: RION. Europe, Council of. 1996. Common European Framework of Reference for Language Learning and Teaching: Draft 1 of a Framework Proposal. Strassbourg: Council of Europe. Lambert, R. 1998. “Language policy: An overview”. International Journal of the Sociology of Language 129: 65–78. North, B. 1993. The Development of Descriptors on Scales of Language Proficiency. [NFLC Occasional Papers]. Washington: NFLC. Oud-de Glas, M., Buis, T. and Withagen, V. 1993. Onderzoek naar Vreemde-talenonderwijs in Nederland. Den Haag: SVO. Spolsky, B. 1994. “Policy issues in testing and evaluation”. The Annals of the American Academy of Political and Social Science 532: 226–237. van Els, T. 1994. “Foreign language planning in the Netherlands”. In Language Planning around the World: Contexts and Systemic Change, R. Lambert (ed.), 47–68. Washington DC: The National Foreign Language Center. van Els, T. and van Hes, E. 1992. The Dutch National Action Programme on Foreign Languages. Den Haag: Ministry of Education. Vinke, D. 1996. “De haalbaarheid van Engelstalig onderwijs in Nederland”. Toegepaste Taalwetenschap in Artikelen 54: 71–82. Withagen, V., Oud-de Glas, M., Smeets, E. and Buis, T. 1996. Vernieuwingen in het Vreemdetalenonderwijs. Nijmgen: ITS.



Part IV

Case studies in educational linguistics Focus and policy



Micro language policy as a barometer of change A university language policy as an example Michael Clyne University of Melbourne

Introduction In this paper I will discuss how a university language policy, which drew on the involvement of an entire institution, faltered soon after, in a context of financial cuts, corporatization, economic rationalism, and a realignment of power structures. The policy and its development will be described and its fate will be discussed in relation to a changing political situation in Australia, a reduction in the responsibilities of government, and the decline of national language policy. Implications for micro language policy in general will be drawn. I can only state that it represents the state of affairs at the time of writing. It is some 44 years since I first met Bernard Spolsky. I was in my final year at a boys’ secondary school in Melbourne when this young New Zealander joined the teaching staff. He was on the first leg of a journey teaching his way around the world — in this case, teaching English and Politics. He did not actually teach me but we had a chance to meet and talk as he marked my midyear examination in Politics. I was subsequently able to follow his career — his balance between research and commitment to social justice, focussing attention on linguistics, education and politics has been an example to all of us. Hence my decision to dedicate to him this modest attempt to combine linguistic, educational and political issues in an institutional context geographically close to the one in which we first met.

212

Michael Clyne

The context of the project As has been discussed elsewhere (e.g., Lo Bianco 1990; Clyne 1991; Ozolins 1993), the adoption of multiculturalism as a national policy in Australia in the 1970s resulted in both provision of services in what came to be called community languages and government support for the maintenance and development of such languages. Among the new initiatives of the time were a telephone interpreter service (now operating in 90 languages), government-financed multilingual radio stations (and, as from 1980, a national multilingual TV network), final secondary school examinations in a large number of languages (currently 43), and public library holdings in the major community languages of the area. The orientation towards multiculturalism also led to a push on the part of academic linguists, teachers, ethnic, indigenous and deaf community groups for an explicit and comprehensive national language policy. The actual policy, which was drafted by Joseph Lo Bianco (1987) and informed by local and overseas research, was preceded by a Senate inquiry involving hearings and written submissions on all aspects of language in Australia, including English as a first, second, and foreign language, indigenous, community and sign languages, languages in the media, external language needs, and interpreting and translating needs. The Senate inquiry set the guiding principles, which were followed by most Australian language policies: Opportunities to achieve competence in English for all, maintenance and development of languages other than English (both indigenous and community languages), provision of services in such languages, and opportunities to learn a second language. Lo Bianco stressed the balance of social, economic and cultural objectives and advantages of a multilingual Australia. In keeping with the general policy shift in Australia towards short-term economic goals, training and globalization, the Australian Language and Literacy Policy, brainchild of the then Minister for Employment, Education and Training, John Dawkins (1991), more strongly emphasized English literacy and languages for economic objectives. However, several of the 14 languages prioritized for teaching (e.g., Italian, Modern Greek, Vietnamese) were thus earmarked predominantly because of their use in the Australian community. Monash University was founded in 1961 as Melbourne’s second university. It grew rapidly in the educational explosion that gave a large proportion of the population an opportunity to study at university level. With the upgrading of colleges of advanced education and their amalgamation with universities, commencing in 1988, Monash became Australia’s largest university, by 1994

Micro language policy as a barometer of change

reaching 36,500 students, and operating from five campuses. It was one of Australia’s most prestigious universities. It had developed a strong push towards “international education”, i.e., attracting students from the Asian region but also giving its own students the opportunity to study abroad. In 1994, international students represented 13% of the enrolment. Many of the other students were from non-English-speaking backgrounds; 80 countries constituted the birthplaces of the Monash population. 27% of the students spoke a language other than English at home.

Project In June 1993, the writer approached the Vice Chancellor (President), on behalf of the Department of Linguistics, with the offer to conduct a research project leading to a language policy for the University, which would cover a similar scope to the National Policy on Languages. It would examine the University’s present position and make recommendations on such issues as (1) training for effective communication, (2) the provision of support for the English language and academic discourse of students from both English- and non-Englishspeaking backgrounds, (3) the multilingual and intercultural resources of the University, (4) the preparation of students in all faculties to be proficient in languages other than English, (5) non-discriminatory language, (6) facilities for sign language, and (7) the status of languages and language issues within the University. Most decisions on language policy had been taken, often independently and in an ad hoc manner, on a daily basis by different people, on matters ranging from the wording of documents to the assessment of language proficiency. As in most institutions of its kind, this had led to uncoordinated and even contradictory decisions. The project was approved, and three members of the Department, Michael Clyne, Mark Newbrook, and Anne Pauwels (now at the University of Wollongong), were given fractional time release to undertake the project with the help of a Senior Research Assistant, Deborah Neil. The project would take 12 months, i.e., the whole of 1994. Our Department hoped that through the exercise we would demonstrate the centrality of linguistic issues to the University. The brief was to (1) document current practices in the areas of language and communication across the campuses of the University and identify language and communication needs, (2) identify the needs of faculties, departments, and centres in these areas, (3) compile a report summarizing the

213

214

Michael Clyne

main issues and concerns, (4) develop recommendations for the implementation of the policy. A steering committee was formed, comprising the four team members, the Senior Deputy Vice Chancellor (Vice-President), and the Pro-Vice Chancellor (International). The project was able to build on a review of language and learning services (i.e., services to help students of all backgrounds with communication skills) and a languages development plan, which initiated a large-scale “study abroad” scheme for language students. Information for the language policy project was gathered through: (1) interviews with deans, some heads of departments and other senior staff, and student groups on issues of concern to them, and especially on key issues such as developing communication skills in international students and Australian students from non-English-speaking backgrounds; also focussed consultations with focus of colleagues and different interest groups on all campuses of the university; (2) questionnaires on matters such as class size and medium of instruction in language subjects and the use of and competence in languages other than English by staff of all faculties for lecturing, conference papers, and collaboration in international research projects; (3) feasibility studies on the introduction of courses in languages of limited enrolment, languages other than English as part of primary teacher training, and of pilot immersion programs in some languages through vocational faculties; (4) archival work. Monash was the first Australian university but not the only one, embarking on a language policy, and there was vibrant discussion on such matters. The Australian Linguistic Society (ALS) annual conference in September 1993 opened with a forum on University Language Policy, at which Elizabeth Dines, then Academic Registrar of the University of Adelaide, urged fellow-linguists to “contribute their expertise to developing policies for their universities that address languages issues in a systematic and coordinated way” (1994: 15) as the National Policy on Languages had done for the nation. The 1995 ALS Conference at the Australian National University included a report from three members of the project team on the process and recommendations of the Monash Language Policy. Other Australian universities which have undertaken institutional language planning include: the University of Adelaide, which has taken on board the full range of issues and involved Linguistics students in the fact-finding, the Australian National University and the University of Western Australia, which both limited their policy scope to languages other than English, and LaTrobe University, where a Mission Statement on Languages (i.e.,

Micro language policy as a barometer of change

languages other than English and ESL) was completed in November 1995. Although this LaTrobe policy became official in 1996, little implementation has resulted (personal communication David Bradley). The University of Melbourne developed a cultural diversity policy in 1997, and the Royal Melbourne Institute of Technology completed one in 1998. These policies touched on a number of language issues concerning international students and Australian students from non-English speaking backgrounds. Kaplan and Baldauf (1997: 256–262) develop a contextual framework for the development of university language policies in Australia, focussing especially on literacy issues.

Response The response to the inquiry indicated great interest in, and enthusiasm for language issues throughout the university. Response rates to questionnaires were high and there was, on the whole, a strong willingness to talk about language and communication matters. The actual discussions and early drafts contributed to debate, awareness raising, and the resolution of some issues before the final report. There were a number of industrial issues, such as the workload and leave entitlement of people running low candidature language programs and the academic status of teachers in the Language and Learning Centre. Our audit of the University’s language resources indicated that over 300 academic staff members in 57 departments had expertise in a total of about 50 languages. Some issues, such as the need for a bank of software in a wide range of languages, were initiated by colleagues responding to a report on the project in the University news-sheet of the time. The 43-page report of the project contains fifteen recommendations and eight strategies for implementing them. The latter are more specific than the recommendations. For instance, a recommendation: That the University see it as part of its national responsibility to teach some languages with limited enrolments especially those otherwise not available in Australian universities

has, attached to it, three implementational strategies: a. requesting the Vice Chancellor to make representations to the appropriate government department for special funding for such languages, b. asking the University to give support to the development of Distance Education curricula for some of the Asian languages currently with low enrolments, and

215

216

Michael Clyne

c. appealing to the University to do its utmost to continue to teach Khmer, Korean, Thai, and Vietnamese.

Results Effective communication Training for effective communication Our report cites a number of very successful practices, such as the integration of language and communication components into core courses in Engineering, including oral communication components, a Communications Liaison Group involving staff from the Language and Learning Centre and the Engineering Faculty, and a Professional Disputation course for postgraduate Mechanical Engineering students in which they “formally attack and defend a thesis”. The report also records the understanding that has been expressed for students from cultural backgrounds for whom argumentation of this kind is considered inappropriate and who may take some time to develop biculturalism. The first recommendation, addressed to the faculties and to the University’s Education Committee, expresses the need for all faculties to follow these examples: Rec.1. That explicit communication activities be made to constitute an integral part of training in all disciplines and that this be included in the University education policy and in the objectives of degree programs.

In view of the cultural diversity of the staff as well as of the student population, it was recommended that cross-cultural communication be a component of induction courses for new staff conducted by the University’s Professional Development Centre but also suggested that this be extended to existing staff and included in training for heads of departments conducting staff evaluations. Equity in communication As the University had already adopted some language guidelines in favour of non-sexist language in all official guidelines and publications, with a short section on language use for minority groups (e.g., ethnic minorities, people with disabilities), it was sufficient to recommend the reiteration of this commitment to non-discriminatory use and extend it to inclusive language covering all categories defined in the University’s Equal Opportunity Policy. This section of the report also provided information on facilities in Australian

Micro language policy as a barometer of change 217

Sign Language (Auslan) and Braille and indicated interest in the availability of courses in Auslan among hearing as well as deaf and hearing-impaired students. Communication within the University This section of the report constituted a premonition. It can be expected from all members of a university that they display critical awareness of, and sensitivity to the ways in which one talks about the educational community. Currently, nomenclature from business and economics is in vogue to refer to the university, its members and activities. For example, metaphors from business corporations such as products and markets are frequently used to refer to courses offered and students. Similarly … students (are designated as) clients …

This proved to be one of the more controversial sections of the report, as I shall discuss later, and although no specific recommendation was formulated, members of the university community, especially the administration, were advised that these terms redefine the identity of members of the university, their work and relations to each other “in ways which do not always reflect the(ir) expectations and perceptions”. Communication and new technologies In recognizing the advantages of utilizing new technologies, the report called for academics to understand the influences of the technologies on teaching and learning, to maximize the enhancement of the students’ written communication.

English and Monash’s role as an international university Our surveys indicated uncertainty among staff as to whether they were assessing expression as well as content and who was responsible for the English of students’ dissertations when students had English as a second or foreign language, or as a first language but a different national variety. Facilities for helping international and non-English-speaking background students were explored and recommendations offered. There were indications of a demand for postgraduate programs in the Teaching of English to Speakers of Other Languages for overseas teachers, and it was suggested that this demand should be investigated and referred to the Faculty of Education so that, if appropriate, the necessary programs could be put into operation. Concern about the TOEFL/ IELTS score of 6 being too low gave rise to a proposal that research be carried out tracking the academic performance of students selected on this basis.

218

Michael Clyne

To assist staff and students across the University, it was recommended that a uniform set of correction symbols be devised covering the main grammatical and stylistic difficulties to act as a diagnostic tool and to promote awareness on the relation between content and expression in assessment. Advice to staff included in the report covers areas ranging from cultural differences in academic discourse and taboo subjects to students’ appeal to authority. This latter point led to the issue of plagiarism, which has a cultural dimension, many students from Asian and some other backgrounds taking ideas from a respected person such as an author or a lecturer, whose expertise they could neither challenge nor surpass. The report therefore recommended (Rec.6) “that cultural variation in the perception of plagiarism be taken into account in any university actions on plagiarism”.

Utilizing and strengthening the multilingual and intercultural resources of the University Mandarin, Cantonese, French, and German emerged as the languages in which members of the University were most likely to have the capacity to give lectures and conference papers. Some interest was shown, especially in the Faculty of Engineering, in conducting some courses through the medium of a second language to provide students with an immersion situation comparable to that at the University of Ottawa or the FLAC (Foreign Language Across the Curriculum) at some American universities. There were already a considerable number of students enrolled in combined degrees in Engineering and Arts, majoring in a language. It was recommended that a pilot program be trialled involving one or more Engineering lecturers, the relevant language department (initially Chinese or German) and the Language and Society Centre (attached to the Linguistics Department) for course development, advice and evaluation. On the other hand, it was suggested that short professional courses be developed for people from overseas in disciplines such as Business and Engineering in appropriate languages other than English where the respective departments considered there was a demand. It was recommended that (Rec.8) “the university purchase computer software in languages other than English (e.g., scripts, thesaurus, dictionaries)” in order to enable staff to utilize their skills in the languages mentioned in the languages audit. Proposals in an earlier Languages Development Plan to enhance the opportunities of students wishing to specialize in languages or combine them with other courses were strongly supported — notably through the establishment of a specialist Bachelor of Arts

Micro language policy as a barometer of change 219

(Languages) with at least a minor sequence in two languages and Linguistics and a major in two of these subjects, and a Diploma of Languages that can be taken concurrently with a degree course in any other faculty. The report addressed an issue that was concerning the Board of Studies which is responsible for the end-of-secondary school assessment. — In the interests of equity, should students with and without a home background in a language they are studying be assessed differently? We found serious problems in differentiating groups, especially since any system of discriminating in favour of “non-background students” could discriminate against, and demotivate those with a limited background (e.g., second and third language, only one bilingual parent, very little home exposure to the language) who had worked hard to build on it (cf. Clyne et al. 1995). A study of the academic results of those who had been helped to a university place through their very good results in a language other than English (which carry a special bonus in the state of Victoria) showed that many of them picked up additional languages and performed outstandingly in the new languages.

Decisions on the teaching of languages other than English Of the eight Asian languages taught at Monash at the time, Japanese, Mandarin, and Indonesian had strong participation rates. Korean appeared to be moving in that direction, but enrolments in Malaysian, Thai, Vietnamese and Khmer were low. Our report addressed the recommendations of an earlier Working Party on Asian Languages that a feasibility study be conducted over three years to make these languages more viable and that these languages be limited to minor sequences. We believed that such a limitation would lower the enrolment potential of the languages even further and urged the University to do its utmost to keep the languages, in the case of the most precarious, Khmer, arguing that Monash is the only Australian university to offer it, that it had relevance to the teaching and research in the University’s Centre for South-East Asian Studies and that it was spoken by a sizeable community in the immediate environs of the University. It was suggested that the Vice Chancellor raise with the Government the issue of funding of programs in low candidature languages (Baldauf 1995) represented in only one university and that distance education programs be developed so that major sequences can be continued in the languages. The arguments for teaching low candidature languages were discussed also in relation to Dutch, a language for which a feasibility study was conducted following its abandonment as a subject at another university —

220 Michael Clyne

existing expertise within the University, relevance for other departments, student potential, and community support. The report recorded some interest in the offering of a living, viable Aboriginal language in addition to existing overview courses in Aboriginal languages or the in-depth study of a now extinct one and in the availability of Auslan as a language subject through LaTrobe University. A survey of class sizes in language departments found that they were generally too large (13 : 1) for effective input and output and therefore for the facilitation of teaching through the medium of the target language. A more comfortable staffing formula was operating for language departments than for traditional humanities departments. This is a recognition of the heavy demands of labour-intensive language teaching. However, the level of staffing was still not adequate and the more favourable formula was not being utilized to the best advantage because English-medium literature, linguistics and civilization classes were conducted in groups of similar sizes as language and target-language classes. It was therefore recommended that “class sizes in practical LOTE teaching be such as to facilitate adequate language input and output”. In order to increase the provision of primary and secondary language teachers, three measures were suggested: (1) a combined degree in languages and teaching on the campus that offers primary teacher training with timetabling to facilitate this combination, (2) a follow-on Diploma of Education for language graduates wishing to teach languages at primary school, (3) a postgraduate certificate in LOTE Education for primary and secondary teachers without a LOTE qualification, involving advanced language (including classroom discourse), LOTE methodology, and second language acquisition. A survey among local school teachers indicated substantial interest across a range of languages. In order to ensure adequate levels of graduates, it was recommended that the Language Competency Committee, a sub-committee of the inter-departmental Language Studies Committee, establish proficiency levels to apply across languages, taking into account the relative distance between English and the target language, where possible, pegged to international competency certificates. Finally, it was proposed that the recommendations of the report be addressed as soon as possible and that the entire language policy be reviewed in three years.

Micro language policy as a barometer of change 221

Revisions by Education Committee The draft report was circulated for comment among many of the people who had been involved in discussions and submissions. Suggestions were received, especially from the Pro-Vice Chancellor (International) and some changes were made, including the correction of some factual errors and disagreements on interpretation. Objections to the critique of the use of business metaphors relating to the University caused disagreements, but the section was left in the report. The final report was accepted in July 1995, by the University’s main governing body, the Academic Board. The Board requested the Education Committee, comprising Associate Deans (Teaching) of the various faculties and a student representative, to form from within itself a working party to “consider the most effective way of accommodating within the Monash Education Policy” specific recommendations of the language policy document. The Working Party decided that the key recommendations should be adopted as a separate Monash Language Policy “which can be visibly identified as a universal and coherent statement about Monash’s position in relation to language issues, as an integral part of its role as Australia’s leading international university”. In addition, certain sections of the Draft Language Policy would be included in the University’s Education Policy, which so far had paid little attention to language issues. Recommendation 1 found its way three times into the Education Policy: (1) Under “Effective teaching”: “Communication skills (would be emphasized) as an integral component of all courses”. (2) “Both general and discipline-specific communication skills” were included under skills and attributes of graduates in the University’s teaching objectives. (3) Academic units (faculties, departments, centres) would have to specify communication skills as one of the learning outcomes of their courses as one of their responsibilities. Teaching and learning support offered by the Professional Development Centre would include a component in cross-cultural training, as would mentoring of new staff at the Faculty and Departmental level. In order to implement the common set of correction symbols, the Department of Linguistics, in co-operation with the Language and Learning Services Unit, was asked to present to the Education Committee a list of useful correction symbols which, if approved, could be recommended to faculties across subjects. Our recommendation on plagiarism was referred to the Discipline Committee which was asked to “separate out the teaching issues from the discipline issues involved” so that a formulation for inclusion in the Education Policy could be found.

222

Michael Clyne

A number of other recommendations were referred to other bodies: a. The extension of inclusive language to cover categories other than gender in the University’s Equal Opportunity Policy — to the Equal Opportunity Committee. b. The possibility of subjects in various faculties taught in a language other than English to provide an immersion experience — to the Faculties themselves. c. Measures to reduce the class size of language classes and classes conducted in languages other than English — to the Faculty of Arts. d. Common language competency standards for students majoring in a language — to the Language Studies Committee. The Working Party proposed to the Library that “any useful software (in scripts of other languages) which it does not already have should be purchased, noting that the cost will be relatively small”. Moreover, it was noted that a number of recommendations had already been implemented: a. Support for low-candidature Asian languages and the development of some programs through distance education and cooperation with other universities. b. The approval of the introduction of two subjects in the Aboriginal language Bandjalang in 1996. c. The opportunity for Monash students to take Auslan at LaTrobe University was now available under a general cooperative arrangement.

Guiding principles: A new basic discussion with a different reaction The report of the Working Party next had to come before the Academic Board for final approval. It was accompanied by a statement of Guiding Principles, included in our final report. At the relevant meeting, in November 1995, the Academic Board accepted the amendments to the Education Policy incorporating the relevant sections from the Language Policy. It also referred the recommendation on Plagiarism to the University’s Policy and Legal Counsel for inclusion in the draft statute on plagiarism, approved the suggestion that the Department of Linguistics and the Language and Learning Services Unit develop the set of correction symbols, and referred the other recommendations of the Language Policy to the bodies which would implement them. The whole issue of the free-standing University Language Policy and the guiding principles

Micro language policy as a barometer of change 223

was referred to the Deans of Faculties, as would be customary for matters to be implemented largely at Faculty level. The Guiding Principles were reworded by the Dean of Arts without affecting the spirit of the statement: UNIVERSITY LANGUAGE POLICY Guiding principles In adopting a University Language Policy, Monash University recognizes the centrality of language in academic, professional and social life, the rich linguistic resources available within the institution, and the language needs generated by globalisation. The Policy establishes that: 1. Monash University accepts the responsibility of encouraging effective and efficient communication in all its operations. 2. The University recognises that communication is a two-way process, equally the responsibility of students, teachers, and administrators. 3. The University accepts its responsibility to assist students of all backgrounds in improving their communication skills. 4. Monash is a multicultural community operating on equitable principles in an international setting. Communication in the University should be inclusive of race, ethnicity, and also gender, age and disability. 5. Facilities should be available to assist members of the University who have any kind of communication impairment.

This statement was followed by a commitment to a. making communication activities explicit in the professional training in all disciplines; b. including cross-cultural communication training in staff induction and encouraging faculties and department to make use of it, and c. accepting it as part of its national responsibility to teach some low-candidature languages otherwise unavailable in our state. The response of the deans to the whole Language Policy issue was less supportive than any so far. Deans were asked to give a written response. While the deans of Arts, Education, and Engineering supported the policy as revised by the Working Party, Engineering indicating that many of the recommendations in the language and learning area had already been implemented independently in that Faculty, the Dean of Computing and Information Technology agreed with the recommendations being incorporated into the Education Policy but saw no point in also having a free-standing comprehensive Language Policy. Responses

224 Michael Clyne

from Law and Medicine showed confusion between the Language Policy and the recently tabled report on the review of the Language and Learning Services Unit. There was also fallout from an ongoing power struggle between the Deans and the Central Administration. In his submission, the Dean of Law argued that deans should have decided on the need for services since they have “a keen appreciation of the need for financial stringency”. Instead of a central professional Language and Learning Services Unit, he contended that “teachers and former teachers” could be employed “at casual salary rates” within the Faculties to undertake language and learning services for international students. After the Faculty of Medicine had supported the Working Party’s recommendations, the Dean of Medicine wrote agreeing with the Dean of Law’s objection. At the subsequent Academic Board meeting in April 1996, the confusion between the Language Policy and the Review of Language and Learning Services was clarified. In the discussion on the Guiding Principles, a tone of hostility was apparent which had not been present previously during the development of the Language Policy. The Dean of Law argued that the principles, being general, implied an accountability and financial commitment of unknown proportions. “We shouldn’t have started it in the first place without considering the financial implications”. Two speakers questioned the need for a language policy on the grounds that a numeracy policy was needed just as much. One board member moved that Guiding Principle 1 be amended so that “communication” be preceded by the word “English” contending (erroneously) that English is the official language of the University. The four sentences of commitment were dropped due to financial anxieties. The guidelines were then endorsed and referred to faculties for consideration. When they came up again at Academic Board, at the June 1996 meeting, the amendment to include the word “English” lapsed through want of a seconder.

The wider political context By 1996, Australia was well along the path of economic rationalism. Everything had to be reassessed to ensure its profitability. Governments were divesting themselves through privatization of many of their traditional responsibilities, including water, gas and electricity supplies, transport and, to an increasing extent, health and welfare. The “user pays” principle now presides almost everywhere.

Micro language policy as a barometer of change 225

Few areas have been as shabbily treated as education. Federally, the new direction was initiated by the establishment of the amalgamated Department of Employment, Education and Training (the order is revealing). In Victoria, for instance, many primary and secondary schools were closed and large numbers of teachers sacked. Education budgets have been slashed. This has affected universities most of all. They are now expected to raise a substantial part of their income from outside government-derived operating budgets. This has led to a pretence that universities are business corporations, an emphasis on costcutting, investment, income generating activities, and what Clarke (1998) describes as “dummed-down” managerialism. Like governments, universities generally opted for economic rather than social agendas. The paradox is that the main source of extra revenue was supposed to be full fees paid by overseas students as the institutions of higher education tried to reap the harvest of globalization. This entailed providing services that were useful to the “clients”. The Asian financial crisis did not reward those who had placed all their confidence in Asian students’ fees making Australian universities profitable businesses. It should be added that Australia has had a long history of anti-intellectualism as well as a culture of materialism and upward socioeconomic mobility. Economic rationalism, introduced by a Labor government and stepped up by the subsequent conservative government as well as by conservative state governments, is extremely strong, although there is much discontent with this, which fed into support for the extreme right-wing nationalist One Nation Party. In 1997, after many years of protracted claims, the industrial court granted academics a salary increase. The Federal Government recognized this but would not fund it, requiring universities to pay it out of their declining budgets. The situation with languages other than English fitted the general context. As has often been the case in Australia, languages were labelled, and a particular “type” of language was the “flavour of the year” for political or short-term economic reasons. In the mid-1990s it was “Asian trade languages” rather than “European” or “community languages”. In 1994 the Rudd Report (Rudd 1994) was accepted, giving special priority status to four Asian languages, Indonesian, Japanese, Korean, and Mandarin, and setting a target that 65% of all school students and 15% of Year 12 students should study one of those languages. Large additional funding was made available to the states to implement this but this too gradually declined. Victoria, the state in which Monash is based, continued to prioritize four Asian and four European languages, French, German, Greek, Indonesian, Italian, Japanese, Mandarin, and Vietnamese,

226 Michael Clyne

several of which have significant communities in Melbourne, and the Victorian government has a strong policy on the teaching of languages in primary and secondary schools. While some federal governments in the past, for instance the Whitlam and Fraser governments, had taken special steps to fund languages of particular importance in Australia at university level, this is not the case now. The policy of productive diversity propagated by the Keating government, and still continued in Victoria as the Multicultural Advantage, has lost its impact on language policy in universities in recent years.

The university: A new beginning At the end of 1996, the Vice-Chancellor/President who had been the chief executive of the University for ten years retired. It should be stated that ViceChancellors and their deputies have unprecedented power in the running of Australian universities, which had previously been collegiate institutions. Before the arrival of Monash’s new Vice- Chancellor, arrangements were made for outside appointments to replace all the deputy and pro-vice chancellors except one. This meant that, in a sense, planning for the University would start from scratch again. Existing policies, including the Language Policy, were disregarded. In the middle of 1997, neither the Deputy Vice-Chancellor (Teaching) nor the Deputy Vice-Chancellor (International) was aware of the Language Policy. During 1997, a new “Monash Plan” was developed, and in late March 1998, the draft of a new “Learning and Teaching Operational Plan” (LTOP) was released so that implementational strategies could be developed to meet the plan. Some of the recommendations in the Monash Language Policy correspond to the following strategies in the LTOP: Develop Monash graduates’ attributes Develop graduates’ … life-long learning skills of written and oral communication Extend internationalization and study abroad opportunities for students

This would involve the preparation of further discussion papers on internationalizing the curriculum, “develop(ing) university wide guidelines and policy”, and incorporating this perspective in all course proposals and terms of reference of academic reviews; and development of suitable curricula for “off-shore courses”; develop and extend subjects taken in on-site locations overseas, and provision of study abroad programs for students in all faculties. The exact nature of the language aspects of this are not spelt out, apart from

Micro language policy as a barometer of change 227

the general objective, “Expand access to intercultural and language studies” and the mention of the Diploma of Languages. However, a further “strategy” is to “Develop strategies (sic) to improve international knowledge and perspectives and cross-cultural awareness of academic staff” by exchange programs and study leave in relevant countries and universities. In order to improve student learning support, proposals would be prepared (including resourcing implications) for systematic support for students with English as a second language and for the integration of language and learning support into faculty teaching. These are but a few of some 130 strategies. Targets were being developed at the time of writing for the early implementation of LTOP. It will be seen that some backtracking would have to take place.

Implementation in an environment of cuts Restructuring Before reviewing what has become of some of the recommendations in the Language Policy which were being implemented, we need to relate that, during 1997 and 1998, plans were made to “strengthen” (i.e., decimate) the Faculties of Arts and Science which were running at a deficit, the former especially because of the large proportion of the budget spent on academic staff. It was believed by the university administration, that these two faculties had less of a future because they were not primarily providing much vocational training. Although 70 positions had been cut from the Faculty of Arts in 1996 and 1997, it was required to give up another 55 academic posts through a combination of voluntary early retirements and forced redundancies. Two alternatives were available to the Faculty: dismissals on the basis of questionnaires which were to be completed on one’s performance and a complete restructuring of the Faculty with some courses and areas being axed. Despite widespread opposition from the Faculty members to both of these, the latter was imposed. In July 1998, a Review Group was charged with developing a new blueprint for the Faculty of Arts, in the interests of cutting out complete areas and therefore positions in order to make up for a financial deficit. The first report, distributed in mid-August 1998, proposed the abolition of departments and the establishment of nine schools of which Languages (including the Arts support staff from the Language and Learning Services and the Language Centre) was one.

228

Michael Clyne

This was to be the largest school in number of staff members, suggesting that much of the cutting would occur here. In fact, it was assumed that many of the language programs would be closed down. The inclusion of the Language and Learning Unit and the Language Centre, it was argued, would contribute research on language teaching pedagogy and “facilitate the movement towards on-line delivery of course elements which must occur within all language areas over the past few years”, respectively. The use of new technologies and webbased/interactive learning is cited as a possible way out of a major financial problem by facilitating a reduction in face-to-face teaching. Minimum subject enrolments (60 in first year, 25 for second and third year subjects and 15 for third year only subjects) were impossible for most of the languages departments. The Languages staff were given the alternatives of reducing expenses and especially the cost of staff or else a considerable reduction in the number of languages taught. At the same time, the labour-intensive nature of language teaching was recognized and the question raised if “some language staff should be dedicated class-room teachers”, i.e., with a much larger number of classes and no responsibility to undertake research. In determining that “not all languages currently taught can be maintained”, co-operation between the universities in Melbourne was invoked. The language of the original blueprint idealized the panacea of change — “the way forward” — in which greater emphasis would be placed on vocational training and subjects would be “repackaged” to that end. Incidentally, underlying the restructuring of the Faculty (named “Academic planning”) was the assumption that departments that were large at the time would constitute the core of the new Faculty, and the other areas would have to be accommodated around them. For that reason, the Review Committee found it hardest to find a location for Linguistics, which had multiple disciplinary links. Their proposal was for Linguistics to become part of a School of Literary and Visual Communication, around English literature and visual arts, which would not give linguists much scope. Within the severe time constraints imposed on initial responses to the blueprint, the Department of Linguistics negotiated a merger with the Department of Philosophy and the Centre for Human Bioethics which it was believed would allow Linguistics to continue to exist in a more independent way. The role of the Language departments seems to have been understood, in the Review Committee, as predominantly language teaching and therefore subacademic. In a document that is full of the jargon of internationalization, the importance of languages is not recognized, and they are seen mainly in terms of their over proportionate cost to the Faculty. The languages departments

Micro language policy as a barometer of change 229

appealed against their merger, instead proposing successfully the amalgamation of Asian languages and of European languages into a School of Asian Studies and a School of European Studies respectively, with the Language Centre and the Language and Learning Services Unit remaining independent. The restructuring was subject to widespread protests from staff and students and the wider community, and there was much media coverage of the plight of Arts faculties throughout Australia. The Arts Faculty Board (September 9, 1998), while acknowledging the blueprint as “a basis for a restructuring of the Faculty”, rejected the idea that it was necessary and passed various amendments to preserve many of the existing departments within the new schools. Now to discuss the implementation or otherwise of the actual policy recommendations. In brief, it could be said that very little positive implementation had taken place in the absence of follow-up.

Induction in cross-cultural training No progress has been made with this initiative (see above: Training for effective communication), except for staff seconded to the new Malaysian campus of the University. The more general induction is being considered in relation to a draft on the internationalization of the University as part of LTOP.

Correction symbols A set of symbols, devised by the working group, was trialled in three departments in first semester 1998 prior to a proposal being made to faculties.

Software At the time of writing, nothing had been done about obtaining software in a full range of languages although part of this purpose had been served by the facilities available through Microsoft Word.

Languages It will be seen from the discussion of the restructuring of the Faculty that the cuts would have a catastrophic effect on languages other than English. This was somewhat alleviated by the subsequent interim decision of the Vice-Chancellor

230 Michael Clyne

to increase the funding based on student enrolments by 15%. However, even before the restructuring proposals, the implementation of the policy recommendations was slow and half-hearted. The committee investigating a common set of competency criteria was unable to come up with a clear cut answer as there were many differences according to the type of language, its distance from language, and the entry level of the students. The Bachelor of Languages was introduced in 1996. It has attracted students wishing to specialize in languages and linguistics. The Diploma of Languages did not have its first intake until 1998. Its establishment had been marred by bureaucratic delays. The diploma enables students to study one or more languages alongside a degree course in any other faculty. The Cambodian (Khmer) program was discontinued in 1997 and the Vietnamese program phased out in 1998–99 following the departure of the lecturer. Malaysian had been dropped earlier. With the departure of the Senior Lecturer in Thai, that program was reduced and completely phased out by the end of 2000. A grant received from the Vice Chancellor in 1995 was spent on the development of distance materials based on a program conducted at the Australian National University but, as far as I can ascertain, the task was never completed. Courses in Latin, Classical Greek, and Classical Cultures were to be closed down. In 1996, a Languages Consortium had been established by the Deans of Arts at the various universities in the state of Victoria. This was to coordinate offerings in Languages other than English so that any student enrolled at any of the universities could take at any other university a language not offered at their own. This enabled each university to “rationalize” their languages, especially those with low or declining enrolments. For instance, as from 1999, Russian at the University of Melbourne would be taught by staff from Monash, while Melbourne staff would teach Latin at Monash. Among unresolved problems are the time and monetary cost of travel and the timetable clashes. Also, the cooperative arrangement did not prevent universities from dropping languages, as in the case of Hindi. Monash stopped teaching Hindi because it was offered at LaTrobe, which in turn abandoned its program. No Monash student has so far taken advantage of the opportunity to study Auslan as a complementary subject at LaTrobe.

Micro language policy as a barometer of change

Plagiarism A sub-committee of the Heads of Departments in the Faculty of Arts discussed new disciplinary arrangements for plagiarism. It decided that two things had to be distinguished: (1) plagiarism which is defined (and covered) by new statutes as “cheating” because it is intentional, i.e., involves “seeking to obtain an unfair advantage”; (2) plagiarism “where no such intent is established and which thus is not covered by the draft of the new University statute, but which needs to be covered by Faculty policy”. The Faculty recognized that plagiarism was a “scholarly and pedagogical issue as well as an ethical one” and should be approached in the first instance as a “problem of misunderstanding the nature of citation and reference” (i.e., as a cultural matter). Special reference is made to students for whom plagiarism is a “cultural obligation” as evidence of being well-read, displaying superior verbal memory, or showing deference. In such cases, discussion with the student, resubmission or deduction of marks and not disallowing the work, suspension or exclusion of the student would be the consequence. The sub-committee recommended that a statement on the interpretation of plagiarism be included in all student handbooks, that the question be discussed in first year tutorials, and that the Associate Deans (Teaching and Graduate Studies) report annually to the Faculty Board on cases that have come before them.

TESOL qualifications for international students Some international students now take a Master of Education (TESOL). There is a joint venture between Monash and Pukyong University (Korea) in which the Korean students take courses at Monash during the summer. The undergraduate IELTS entry score throughout the University (see above: English and Monash’s role as an international university) remains 6.0 (TOEFL 577), but it is 6.5 for postgraduate students in the Faculties of Arts and Education, for instance. However, for Master of Education (TEFL) students, because the course has English-based subjects, the Faculty requires TOEFL 577 or IELTS 6.0.

Language teacher education The Postgraduate Certificate in LOTE Education was introduced in 1997. There is a graduate diploma (Primary) in place now but the language specialists have problems simultaneously satisfying the generalist requirements of the course. A

231

232

Michael Clyne

concurrent Bachelor of Arts/Bachelor of Teaching course was introduced on the Peninsula Campus, where primary school teachers are trained. Unfortunately enrolments in the languages stream are very low.

Inclusive language The formation of the Languages Policy coincided with a review of the Equal Opportunity Policy in 1995. A small booklet of Language Guidelines published some years ago has not been updated because there is an Australian style manual (Australian Government Printing Service, 1988) the inclusive language chapter of which, written by our former colleague, Anne Pauwels, covers the issues requiring attention. The University Solicitor’s Office changes university regulations and legislation into inclusive language, but this is mainly genderinclusive only.

The new look university administrators and their discourse The unfunded salary increases (see above) brought to an end the long tradition of uniform salaries across Australian universities. The National Tertiary Education Union, the academics’ trades union, then negotiated on behalf of its members with the employers, the “universities”. This reinforced an increasing tendency for Vice-Chancellors and Deputy Vice-Chancellors to speak as “the University” and to describe the staff (and students) in terms that exclude them from this umbrella title. “The university will not condone staff cancelling classes for the time of the ‘stop work’ forum”. (Memo from the Vice-Chancellor 23 July 1998). “The deputy vice-chancellor of Monash University … said students and not the university were the most a ected by the strike”. (The Australian, 8 April, 1998). The Vice Chancellor also refers to himself as the CEO (Chief Executive Officer) of the University (e.g., to a group of student demonstrators, 25 August 1998).

Closing comments With the serious cutbacks throughout the Faculty of Arts, academic staff entered into a defensive mode in which academic freedom, the core functions of the University, and the current general position of the humanities and social

Micro language policy as a barometer of change 233

sciences, as opposed to vocational training, needed to be protected (James 1998). This paralyzed the position of the individual languages although there were voices from all over the Faculty lamenting the treatment of languages in the restructuring. It rendered impossible the advancement of the implementation of the language policy. Our journey through the development and fate of a micro language policy has demonstrated the importance of the general context of enabling or disabling changes for the policy opportunities. The emphasis on starting afresh frustrates ongoing policy implementation. Corporatization and managerialization detract from the collegiality that enables cooperative policy implementation as well as the utilization of expertise within the academic staff of the institution. Moreover, the overwhelming importance of short-term economic profitability as a key criterion in decision making changes the parameters of the intended policy. These are changes that have resulted from the way in which the entire country is running, something that is also reflected in the diminution in national language policy initiatives, so that the fate of our micro policy can indeed serve as a barometer of general change. It has been shown what a difference a few short years have made in promoting or impeding positive language policy. Any language policy requires relatively stable conditions for its implementation. When there are major changes in power arrangements, it may be useful to enable new leaders to claim ownership of the policy, e.g., by adopting an alternative nomenclature. Micro policy developers should consider from the outset what are the most important issues that need to be saved in times of economic austerity and political change. Was the whole undertaking of the University language policy wasted effort? I don’t think so. Too many members of the university were involved in the process for the spirit of the results to be lost completely. They are far more likely to resurface in different forms, perhaps in better times.

References Australian Government Publishing Service. 1988. Style Manual for Authors, Editors and Printers. 4th edition. Canberra: Australian Government Publishing Service. Baldauf, R. B. 1995. Viability of Low Candidature LOTE Courses in Universities. Canberra: National Languages and Literacy Institute of Australia. Clarke, H. 1998. “Dumbing-down in Australian universities”. Quadrant 4 (2), September 1998: 55–59. Clyne, M. 1991. Community Languages: The Australian Experience. Cambridge: Cambridge University Press.



234

Michael Clyne

Clyne, M., Fernandez, S., Chen, I. and Summo-O’Connell, R. 1995. Background Speakers. Canberra: Language Australia. Dines, E. 1994. “The public face of linguistics”. In The Public Face of Linguistics, P. Mühlhäusler (ed), 12–15. Adelaide: Centre for Language Teaching and Research, University of Adelaide. James, P. 1998. “How Monash gained the world and lost its soul”. The Age, p. 10. Kaplan, R. B. and Baldauf, R. B. 1997. Language Planning from Practice to Theory. Clevedon: Multilingual Matters. Lo Bianco, J. 1987. National Policy on Languages. Canberra: Australian Government Publishing Service. Lo Bianco, J. 1990. Making Language Policy: Australia’s Experience. Clevedon: Multilingual Matters. Learning and Teaching Operational Plan. 1998. Monash University. Mühlhäusler, P. (ed.). 1994. The Public Face of Linguistics. Adelaide: Centre for Language Teaching and Research, University of Adelaide. Ozolins, U. 1993. The Politics of Language in Australia. Cambridge: Cambridge University Press. The Review of the Faculty of Arts. Draft Blueprint — a progress report and various versions, August 20, 28, 1998.



The genesis of a district-wide Spanish FLES program A collaborative achievement* G. Richard Tucker, Richard Donato and Kimmaree Murday Carnegie Mellon University (Tucker, Murday) / University of Pittsburgh (Donato)

In this paper, we describe the systematic planning and subsequent implementation of a new system-wide Spanish program at the primary level in one of the school districts in suburban Pittsburgh, PA. We discuss the steps involved in selecting the target language — involving community, teacher, and board consultation; the incorporation of the language program into the overall curriculum of the district; and the active participation by all senior administrators, including the superintendent, in a year-long planning effort that culminated in teacher selection, curriculum development and in-service training for all primary school faculty members. We then describe the current program which is presently in its third full year of implementation. This program involves all children in kindergarten, grade one, and grade two with plans to add one additional cohort each year until all children in the school system are included. We present information concerning the language and attitudinal development of the participating children. Through a series of detailed interviews with the stakeholders of the program, we analyze the planning and implementation process. We identify and discuss six themes that emerge across interviewees. We conclude with a discussion of factors that have contributed to the successful implementation of this exemplary program.

Introduction Recent federal legislation enacted in the United States, the Goals 2000: Educate America Act, calls for American students to leave grades 4, 8, and 12 having

236

G. Richard Tucker, Richard Donato and Kimmaree Murday

“demonstrated competence over challenging subject matter including English, mathematics, science, [and] foreign languages, ….” Although every European country has a national policy for introducing at least one foreign language into the elementary school curriculum of every child (Dickson and Cumming 1996; Pesola 1992), it is estimated that foreign languages are only offered in approximately 31% of the elementary schools in the United States (Branaman and Rhodes 1997). However, studies indicate that, as a nation, we are generally receptive to teaching foreign languages in the elementary school. Forty percent of the American population believes that there should be a foreign language requirement in the elementary schools and seventy-five percent think that foreign language study should be an option available in the elementary school (Eddy 1980). If American students are to complete grades 4, 8 and 12 with demonstrable proficiency in a foreign language, the number of programs at all levels will need to be significantly expanded and improved. This is particularly true at the elementary level. The importance of including foreign language study in the elementary school is also supported by the research on the amount of instructional time required for developing functional proficiency in a foreign language (Carroll 1967) and by the widely held professional view that language competence can only be achieved by children who follow well articulated, sustained sequences of foreign language instruction (Donato and Terry 1995). By expanding foreign language instruction in the elementary school, students will have an extended opportunity to achieve the goals that have recently been articulated and disseminated as the National Standards for Foreign Language Education (1996) and they will have an opportunity to develop truly functional ability in a language other than their first language. A major objection to incorporating foreign language instruction into the elementary school curriculum seems to be that there is not enough time in the instructional day (Baranick and Markham 1986). Our present national concerns with systemic educational reform and with competitiveness make this a critical time to explore more fully factors related to the implementation of elementary school foreign language programs. A number of major issues are often raised when considering foreign language education in the elementary school (FLES) for majority language speakers in the United States, that is for speakers of English as a native language: (1) which model of instruction should be implemented — an immersion or standard FLES model; (2) at what grade level should foreign language instruction begin; (3) in what language(s) should instruction be offered; (4) what are realistic proficiency expectations for elementary school students studying a particular language within a given model;

The genesis of a district-wide Spanish FLES program 237

and (5) how can we best assess the language proficiency of young children. Moreover, very little empirical research has been conducted on these critical issues in FLES and responses to these questions often rely on opinion and impressions about the early language learning (see Donato, Antonek and Tucker 1994; Donato, Tucker and Antonek 1996; Tucker, Donato and Antonek 1996). The goal of this article is to describe the systematic planning and subsequent implementation and formative evaluation of a new system-wide Spanish program at the primary level in a school district in suburban Pittsburgh, PA. The school system is a relatively small one comprising approximately 2,800 students who come from mostly European-American, working-class families. To facilitate this description we will draw on two sources of data: (1) our own experiences and observations as consultants to this program over a four-year period, and (2) an analysis of a series of extensive interviews carried out with various stakeholders who comprised the planning committee. In the sections that follow we will discuss: (1) the active participation by all senior administrators including the superintendent in a year-long planning effort that culminated in choice of language, teacher selection, curriculum development and in-service training for all primary school faculty members, (2) the ways in which the language program which is intended ultimately for all children in the district has been incorporated into the overall curriculum of the district, (3) the status of the current program currently in its third year of implementation, (4) a thematic analysis of the interview data, and (5) plans for the future. We conclude by discussing factors that have contributed to the initial success of this exemplary program.

Genesis of the program In May 1995, Tucker and Donato were invited to attend an informal meeting with the Superintendent of Chartiers Valley School District and several of his administrative staff. The invitation resulted in part from our previous research examining diverse aspects of the implementation of a Japanese program at the elementary school (see, for example, Donato, Tucker and Antonek 1996; Tucker, Donato and Antonek 1996), and partly from the fact that Donato directs the major foreign language education teacher preparation program in the region. This meeting marked the beginning of a mutually beneficial and thoroughly enjoyable school district — university partnership which continues to the present day. Bernie Sulkowski, the superintendent, opened the meeting by articulating

238

G. Richard Tucker, Richard Donato and Kimmaree Murday

a vision for his students and for his district — a vision that included doing something different, something daring. He proposed that a new program be developed so that all of the district’s pupils would study a common foreign language throughout their entire scholastic career. He described how American secondary school graduates in the 21st century will be competing for positions in which numeracy, literacy, problem-solving and communication skills will be increasingly valued, and moreover, how students with bilingual language proficiency will possess a comparative advantage in comparison with their monolingual English-speaking counterparts. He predicted that tomorrow’s graduates will compete for positions in Beijing, Paris, Tokyo and Zurich — not only for jobs in Baltimore, Chicago, Detroit and New York. A number of questions were raised at the initial meeting. Was his vision plausible? If so, in which language(s) should instruction be offered? Were teachers available? Would the community support such a program? Would members of the School Board support such a program — and provide the necessary budgetary authorization? How could the school district and the local universities (Carnegie Mellon and the University of Pittsburgh) work collaboratively to their mutual benefit? The group decided to form a “Foreign Language Program Committee” to oversee the planning and implementation of a new and innovative foreign language program. Committee members consist of the director of curriculum, Kathy Gori, who chairs the group, the superintendent, principals from the primary and intermediate schools, selected primary school teachers, the chair of the secondary school foreign languages department, and the university collaborators. To date (that is, through October 1998), the group has met on 13 occasions to plan, review accomplishments, and make decisions concerning priorities for future work. As appropriate, subgroups or individuals carry out specific activities which they report back to the Committee.

Choice of language and timetable for implementation One of the first issues confronting the group was choice of language(s). A number of options were considered including French, German, Japanese and Spanish. At the time (academic year 1995/1996) the district offered French, German and Spanish to students in grades 9–12 on an elective basis. In addition, members of the university partnership proposed that Japanese be considered because of their work with a local innovative program. A number of factors were evaluated such as likely availability of prospective teachers and materials,

The genesis of a district-wide Spanish FLES program 239

potential community support, utility of proficiency in the target language for graduates, etc. For pragmatic reasons, the decision was made to select only one language, and to make its study compulsory for all children. At this point, the Committee decided that it would be useful to conduct a community survey to ascertain level of support for the program and to obtain feedback concerning the choice of language. A survey instrument was developed, piloted, revised and administered to a representative sample of parents as well as to all members of the school board. The results revealed broad general support for an innovative foreign language education program and support specifically for the teaching of Spanish. Data from the FLES Community Climate Survey developed by a Committee member and distributed to selected parents, the School Board, and a district Steering Committee indicated that approximately 65% of the respondents preferred a program that introduced children to Spanish (the other third of the respondents were split among Japanese, German, French, or indicated that they had no preference) and that this new FLES program should have the goals of developing cultural knowledge (93%), engaging students in the excitement of language learning (80%), and building basic language proficiency (19%).1 The second major question was whether to begin the program from the bottom up, that is at the kindergarten level; from the top down, that is working backwards a year at a time from grade 9 where instruction then began; or from both ends to meet in the middle. After much discussion centering around issues such as scheduling, teacher availability and the necessity of developing long term articulation, a decision was made to propose to the School Board the implementation of a Spanish FLES program in September 1996 for all kindergarten children in the district. The proposal recommended extending the program with the systematic introduction of new cohorts of kindergarten youngsters in subsequent years (i.e., to grade 1 in September 1997, to grade 2 in September 1998, etc.). With this plan, the district will have a fully articulated foreign language program from kindergarten through grade 12 by 2,004. The Board of School Directors formally approved this plan in April 1996 and authorized an initial 5-year pilot project.

Development of an action plan After deciding on the target language (Spanish) and the model for implementation (bottom up), the Committee next turned its attention to: (1) recruiting an appropriate teacher for the first cohort of students, (2) planning for curriculum

240 G. Richard Tucker, Richard Donato and Kimmaree Murday

development activities, (3) informing members of the community about the new program, and (4) orienting other teachers and administrators working in the system. These activities continued during late winter and spring of 1996. A major benchmark was the hiring of the first kindergarten Spanish teacher, Ms. Cassie (Quince) Kuzniewski who was a graduate from the University of Pittsburgh dually certified in elementary education and in Spanish. The early hiring of Ms. Kuzniewski meant that she was able to devote a substantial block of time to curriculum development activities during summer 1996 preceding the start of the program. She worked with other curriculum specialists in the district and in continuing consultation with Donato. Curriculum development activities benefited from and reflected work that had been done on the National Standards as well as innovative work that had been completed by other staff in the school district on “Standards in the Arts”.

Overview of the Spanish program The Spanish program began in September 1996 with 11 classes comprising a total enrollment of 223 kindergarten students. Each class meets for 20 minutes per day, five days a week. The Spanish specialist (Ms. Kuzniewski) goes to the students in their regular classroom, and in effect, team teaches with the regular classroom teacher. A strong collaboration between the kindergarten teachers and the language specialist developed almost immediately and continues to mark this program as unique (see Donato, Tucker and Antonek 1994, concerning the problem of marginality of FLES programs). Rather than expressing indifference toward the new program by working neither to support it nor to repudiate it, the kindergarten teachers — and subsequently the grade one and grade two teachers — have established close contact with the FLES teachers and freely shared materials during the curriculum development phase of the program. They have also actively studied Spanish during the FLES lessons in their classrooms. The program and its teacher were clearly positioned from its inception as an integrated part of the kindergarten program and equal participants in the total school curriculum. The curriculum was developed following the school-district template for planned courses of study. That is, each thematically organized unit (e.g., colors and shapes, numbers, greetings, calendar and weather, clothing and body parts, fiesta and foods, etc.) was specified according to: (1) student learning outcome, (2) content, materials, and activities, and (3) procedures for assessment. The

The genesis of a district-wide Spanish FLES program 241

main focus of each lesson is on vocabulary building and comprehension rather than production. The curriculum reflects this orientation in its assessment procedures including such activities as coloring, baking brownies, movement activities, and the playing of games. Every attempt is made to integrate Spanish with ongoing activities in art, music, library, physical education, and the computer curriculum. The integrated nature of the Spanish class is explicit and obvious in the curriculum. Children learn numbers by accompanying jumping jack calisthenics with counting, listen to age-appropriate fairy tales in Spanish, and are introduced to days of the week in Spanish when learning them in their regular classes. The Spanish teacher uses Spanish whenever possible in the classroom for classroom management and outside the classroom to greet her students in the hallways. She makes extensive use of manipulatives and visuals and brings in a wide variety of authentic materials. Her classes are enriched with visits by Spanish speakers and through a partnership begun in collaboration with students studying Spanish at the secondary level (e.g., the FLORES program — Foreign Language On Request Elementary Spanish program). Ms. Kuzniewski kept parents informed about the goals and the actual content of the program by means of a monthly newsletter, and by her frequent tape-recorded updates on the “homework hotline”. There have also been a number of special presentations for parents, and other interested community members, at regular back-to-school nights and by means of an informational videotape that was prepared for broadcast on a local cable channel. In addition, on several occasions parents have been asked to complete questionnaires designed to elicit their attitudes toward various aspects of the program. During the course of the 1996/1997 school year, the Committee continued to meet regularly to monitor the implementation of the program, to plan for an assessment of student progress during Spring 1997, and to begin planning for the extension of the program into grade 1 together with the introduction of a new kindergarten cohort in September 1997.

Expansion of the program The Committee continued to meet quarterly to discuss various aspects of the program and to plan for its expansion in the 1997/1998 school year. Ms. Lisa Bischoff, another dually certified teacher from the University of Pittsburgh program, was hired and once again the two teachers spent time with Donato during Summer 1997 revising the kindergarten curriculum in light of the first year’s experience and developing the curriculum for the new grade 1 program.

242 G. Richard Tucker, Richard Donato and Kimmaree Murday

When classes began in September 1997, all kindergarten and grade 1 children in the district participated in this innovative Spanish FLES program. They followed the model established during the first year — namely 20 minutes of instruction in Spanish 5 days a week with the specialist teacher who comes to their home classroom. Curriculum development for the second year of the program built on concepts and vocabulary learned during the first year and expanded students’ participation in the lesson to include more oral production. That is, the curriculum retained its integrated, thematic focus but moved toward greater oral participation in the lessons by the students. The Committee addressed a number of major issues during the 1997/1998 school year. In general, members wanted to ensure that parents of current students, parents of prospective students, members of the school board, and other teachers in the district were as well informed about the program as they could be. To this end the Committee drafted and is currently piloting a “report card” for parents intended to convey information about the content of the curriculum and whether their children are progressing to master the material presented. The reporting procedure has undergone several revisions attempting to create a “report card” that capitalizes on what children can do well rather than on where they are deficient. Beginning this year parents will receive a checklist of several functional language abilities (e.g., saying the date and month, identifying classroom items, telling time) with an indication of whether the child has mastered the material or is progressing. Care has been taken in the design of the report card not to discourage children by sending the unintended message to parents that their children lack an aptitude for language study or are not progressing adequately. Given that we do not know what comprises “adequate achievement” at this level of instruction, we wished to highlight achievement over failure in a formal report procedure that would inspire confidence in parents concerning their children’s ability to learn another language and bolsters the image of the program in the home. This report card is also an innovative development since the school district had not previously assigned grades or provided other formative information to parents about children’s progress in foreign language programs at the primary grades. In addition, several back-to-school nights were held for parents with a focus on the Spanish program; an evening orientation program for parents of prospective new pupils was held; an informational video was produced featuring the children and the teachers that was shown on local cable television; and an in-service program was presented in the Spring to inform other teachers in the middle and intermediate schools about the FLES program.

The genesis of a district-wide Spanish FLES program 243

This same general model was followed for the 1998/1999 school year with the recruitment of Ms. Heather Foley, another dually certified teacher, to participate in the program. With the opening of the school year, the program has now been expanded to include all kindergarten, grade one and grade two pupils in the district.

Formative evaluation of the program Members of the Committee decided that it would be important for all stakeholders (e.g., the pupils, their parents, members of the school board, etc.) to conduct a systematic assessment of the progress of the pupils near the end of each year of instruction. In addition, the university partners decided that it would be useful to document as fully as possible the process of program development and implementation in a manner analogous to that described by Markee (1997).

Assessment of student progress To carry out the first of these objectives — a systematic assessment of student progress — a curriculum-based interview protocol was developed, pretested and revised with the assistance of the university partners. Approximately 44 pupils (two boys and two girls from each of the 11 classes) were randomly selected to participate in a 10 to 12 minute “interview” conducted by the high school Spanish coordinator and one of the primary school counselors during June 1997. The subtasks of the interview provided a basis for assessing the students’ listening comprehension (e.g., responding to a command with an appropriate action point to the letter M on the rectangle), their range of vocabulary (asking the child to name in Spanish a range of visuals such as elephant, book, school), and their emerging sense of grammaticality (by asking them to make grammaticality judgments and by asking them to perform sentence repetition tasks with increasingly long sentences designed to exceed short term memory capacity). Interviews were recorded for later transcription and analysis. In terms of interview data, children’s scores ranged from 55% to 100% with a median of 89% leading the evaluator to conclude that “based on the results it is evident that our children are certainly learning Spanish”. The results of the formative assessment of the children coincided with our general expectations about their expected progress (e.g., the children’s listening comprehension exceeded their

244 G. Richard Tucker, Richard Donato and Kimmaree Murday

oral production; their production was limited to learned material; production began as single word utterances and formulaic expressions; signs of emerging syntax have begun to appear but the children focus more on content than on the form of their utterances; language mixing was not uncommon; and the children have developed good pronunciation ability in Spanish). The children and their parents were unanimously positive about the Spanish program and in wanting it to continue. Likewise the views of the regular classroom teachers were positive and supportive. None expressed the view that the Spanish program was somehow detracting from other elements of the school district’s program. The classroom teacher noted that “we are most pleased with the level of achievement our students have attained [and] inspired by the enthusiasm they demonstrate in so doing”. During 1997/1998, the Committee revised the assessment measures for Spring 1998 administration. Once again a randomly selected sample of four children from each kindergarten classroom (2 boys and 2 girls) was interviewed as well as those children now in grade 1 who had been interviewed the previous year. The interviewing was done by the high school Spanish coordinator and one of the primary school counselors. The second interview protocol was revised to include more opportunities for the students to demonstrate creative use of language. The goal was to be able to describe whether the children were able to produce phrases or sentences that involve their combining previously learned material into novel utterances. The assessment instrument was based on tasks with visuals linked to what the children do in their classes (e. g., describing a picture, singing a song, talking about the family) which were selected to elicit language that would allow for fine-grained evaluation of discrete phonological, lexical and grammatical items yet be sufficiently broad to assess creative and spontaneous production that draws upon, combines and recombines these discrete features of language. The students were tested in late Spring 1998. Although data analyses are not complete, the preliminary results are encouraging — and mirror those from the assessment conducted in Spring 1997.

Collaborating on curricular innovation: The insider’s perspective Markee (1997: 46) defines curricular innovation as “a managed process of development whose principal products are teaching (and/or testing) materials, methodological skills, and pedagogical values that are perceived as new by potential adopters”. The definition provided by Markee reveals the underlying

The genesis of a district-wide Spanish FLES program 245

complexity and various components of innovation that must be accounted for in an accurate and thorough description of the phenomenon. To analyze “how change is designed, implemented and maintained” (p. 43), Markee suggests that information concerning the roles of participants, the decision-making process, and the nature of the innovation must be gathered, examined, and understood. Through our participation in the project, we realized that we were uniquely positioned to analyze this planning and implementation process in systematic ways that could ultimately contribute to other school districts wanting to launch similar foreign language programs. As university consultants, we participated fully and regularly in the discussions and decision-making concerning the shape and the future of the program. In this capacity, we found ourselves regularly adopting, rather unintentionally at first, the role of participant observer allowing us access to the thinking of the various individuals across institutional roles and responsibilities during deliberations about the program and while the program was being implemented. As foreign language education researchers, we soon realized the value of our participation to the district and to ourselves and the uniqueness of our dual position. We were true co-participants in a full blown educational innovation and, at the same time, informed observers of this process because of our research experiences with similar early language learning initiatives in FLES and immersion programs. In what follows, we construct a narrative report by uniting the opinions and perspectives of the stakeholders themselves and our participant observations of one school district’s experience in collaborating on the design and implementation of a district-wide Spanish FLES program.

Interview procedure, analysis, and emerging themes We developed an interview protocol (see Appendix A) intended to elicit the participants’ recollections of topics such as the early stages of program development, as well as their opinions of the relative success of implementation to date and problems likely to be encountered in the future, and comparisons with the implementation of other curricular innovations in the district. After consultation with district officials, we selected eleven individuals to be interviewed. Interviewees were chosen, for the most part, from members of the Steering Committee. They included the superintendent, the director of instruction, principals, regular classroom teachers from various levels and two of the

246 G. Richard Tucker, Richard Donato and Kimmaree Murday

Spanish teachers (see Appendix B). In addition, one member of the school board was interviewed. The interviews were conducted over a several week period by Murday who had not participated in earlier Steering Committee discussions, and thus, brought a “clean slate” to the task. The interviewees were advised via a memo from the director of instruction that they would be contacted, following which the individual interviews were set up. The majority took place over the phone; two were conducted in the interviewees’ offices, and two were conducted in restaurants. All but one of the interviews were tape-recorded; in addition, copious written notes were taken. Shortly after each interview, the notes were transcribed. All participants expressed remarkable enthusiasm for the FLES program, and reported that they considered it to be a success to date. During our analysis of the interview data, six overarching and consistent themes emerged across respondents. 1. 2. 3. 4. 5. 6.

A vision that resonated Careful planning Empowerment Support of, and for, the teachers The physical contribution of consolidation Concerns for the future

We turn now to a consideration of each of these themes in the form of discussion and analysis of the collective thinking of the respondents that emerged during these interviews.

Articulating a shared vision The superintendent wanted a foreign language program for the Chartiers Valley school system “because of a sense that American education was behind [the rest of the world] with regard to exposure to foreign languages”. From the time that he first proposed the idea of a foreign language program beginning at the elementary level as part of the district’s plans for consolidating several community schools in one new facility, his vision resonated positively throughout the group. That the language be Spanish was not particularly important; instead, most respondents “felt strongly that a foreign language should be introduced, but it didn’t matter [to them] which one”. The members of the Steering Committee, with representatives from

The genesis of a district-wide Spanish FLES program 247

administration, teachers, community members, parents, and students, all agreed that for their students to be viable citizens for the 21st century, “[they] needed knowledge of another language”. Their students should be prepared to compete for jobs in a global market, where other students would have the benefit of school systems that emphasized the learning of several languages. The students’ representatives expressed dissatisfaction with the status quo; after several years of foreign language instruction at the secondary school level, they often could remember no more than a few key phrases. Many of the respondents also echoed this frustration with regards to their own language experience in school. In addition, the respondents reported that a foreign language program would help elevate the reputation of the Chartiers Valley School District. As one respondent pointed out, “This definitely seemed a way of making them better, and improving their reputation as a leadership district … . It was also a way to show the people that this was something they could do for the children, to improve cultural awareness, and improve their self esteem as well”. Not only would it show their willingness to try innovative new programs, but it would demonstrate their dedication to doing things that benefit the students. The belief that this program would truly benefit the students resonated powerfully and positively across the various interviewees. Some held this belief based on personal experience, or through knowledge of child development. Others were convinced by the research disseminated by the university partners. As one regular classroom teacher mentioned, “The research really did seem to support it. Having taught kindergarten for a while … [I know that] children are really open to learning”. The benefits of this program could go beyond simply providing an additional skill that would help the students compete for jobs; they would extend to other areas of learning: “students studying another language can pick up reading and language skills quicker also”. And respondents noted that success breeds success; some students might excel at language learning who don’t show promise in other more traditional subjects, and this success would bring increased interest and confidence to other areas.

Careful planning All respondents reported that the success of the program to date was due in large part to the careful planning that went into the development and implementation of the program. The most crucial part of this planning revolved around the involvement of all of the stakeholders: the school board and the administrators, of course, but also the teachers, the parents, student representatives,

248 G. Richard Tucker, Richard Donato and Kimmaree Murday

community members, and university partners. Each stakeholder was continually encouraged to “raise their concerns so that they could be addressed, instead of complaining in isolation later”. As one administrator pointed out, “This one was done right. Sometimes when school districts make a decision to implement a program, they’re not always careful to get all of the stakeholders to buy in”. The use of an action plan, developed by the director of instruction, to organize tasks was considered crucial. The action plan included “objectives, specific activities, areas of responsibility, and time lines”. Not only was it developed, but the committee “[does] actually follow the action plan; if something is set to be done by a certain date, they expect it to be done by that date”. Part of the planning involved visiting other school systems that had implemented a foreign language program. Although many respondents felt this was important, some also cautioned that “every district must tailor [a program] to their own school community needs”. The continued and active participation of the university partners provided the district with many valuable benefits. As the first Spanish teacher pointed out, “they didn’t try to reinvent the wheel; they went to find experts … who gave them the background information”. In addition to research, the university partners also “helped them stay focused, and find candidates … even just the positive feedback and reinforcement makes them feel better”. In particular, the superintendent cautioned that working with university people is crucial in terms of staffing; the universities need to be preparing graduates “to do what we’re asking them to do … . Planning won’t help if no one is qualified to fill the positions”. There was a strong sense that the communication in this project between the world of academia and that of the public school provided incalculable benefits for both. Repeated again and again among the respondents was the sense that anticipation of concerns prevented them from becoming obstacles. As one respondent mentions, “they anticipated concerns, and they were built into the action plan … . The proactive nature minimized problems”. In addition, the program is under constant scrutiny and evaluation, and is revised as needed. The Steering Committee, as one respondent put it, “helps them keep on track of things, and anticipate issues”. This gives the respondents confidence when contemplating potentially problematic issues that they feel will need to be addressed in the future.

The genesis of a district-wide Spanish FLES program 249

Empowerment The notion of “empowerment” was central to many of the interviews. Respondents were unanimous that they felt “ownership” of the foreign language program from the beginning. The early discussions surrounding the implementation of the program were seen as drawing upon “ … the philosophy of teamwork in the district”. Respondents reported that they were “ … never hesitant to voice a very small concern because … you know any comments are treated with respect and addressed”. This empowerment that teachers, department heads, principals and others felt was clearly the result of the strong leadership provided by Superintendent Sulkowski who saw his position as one that would “allow him to guide ‘the powers that be’ toward the inclusion of the program [centrally within the core curriculum for the district]”. It is noteworthy that both the superintendent and the director of instruction have personally participated in all meetings of the steering committee over the past three and a half years. As one respondent noted, “Dr. Gori was phenomenal; she would establish all of the meetings … develop the action plan, … made people responsible, decided the questions that needed to be answered and set dates by when they needed to be answered”. Members of the Foreign Language Committee were carefully selected to include representatives from the key constituent groups — administration, building management, supervision, classroom teaching and the university partners. This was seen by many as a new and desirable approach to curricular innovation for Char Valley. One respondent noted “Back in the old days [before Superintendent Sulkowski had arrived] things didn’t seem as well thought out as these initiatives have been in recent years … communication, teacher involvement, representatives from different aspects of the whole procedure, that type of thing”. A major outcome of the Foreign Language Committee meetings was that each stakeholder felt personally responsible for the potential success or failure of the program. The basic philosophy that guided committee discussions was “what would you like to see happen? Don’t worry about cost, but tell us how to do it [well]”. Thus, the program was, in a sense, constructed from “the bottom up” drawing on the insights and the experiences of “the people who are with the kids every day” and drawing on the results of extant research concerning foreign and second language learning and teaching. The discussions were conducted within an ambiance that favored creative, albeit task-oriented, dialogue. Respondents reported that the committee was a “forum … that

250 G. Richard Tucker, Richard Donato and Kimmaree Murday

promoted responsibility of the individual participants in the outcome”, a roundtable in which “everyone had a voice”.

Support of, and for, the teachers Another central thread that was woven throughout the interviews was that of “support of, and for, the teachers”. This thread took many forms. For example, there is a continuing search for teachers with dual certification (in Elementary Education and Spanish) so that classroom teachers would not be expected to learn, and then to teach Spanish, as has been done with other programs in the United States, often with less than optimal results. A good deal of attention was paid to ensuring that the Spanish program was incorporated into the regular curriculum of the primary school with a minimum of disruption. Care was taken to provide assistance to the Spanish teachers through continuing linkage with the university partners and for classroom teachers by the systematic provision of in-service training. This support was manifested in several concrete ways after the selection of the initial Spanish teacher. The foreign language department head noted that “the selection of the first [Spanish] teacher was very important … ” and they were fortunate to find an experienced, credentialled teacher. Several respondents remembered that “they put together a committee involving the Spanish teacher, the primary school department head, and a couple of kindergarten teachers who went through the kindergarten curriculum scope and sequence and looked for how they could best integrate the Spanish with the regular curriculum in a natural and developmentally appropriate way”. This continuing collaboration became a hallmark of the program as it expanded over time from kindergarten to grade one to grade two. Respondents also noted that the Spanish teachers “had the support of the classroom teachers; they’ve really accepted it, they’re learning the language and speaking it with their students, and they reinforce what they’re learning all throughout the day”. The principal of the primary school noted that he devotes a good deal of attention to ensuring that the regular classroom teachers are “attempting to implement the Spanish instruction in their daily lessons as well”. All involved have tried continually to inform and involve teachers from the intermediate school so that they will be prepared to receive the students when they reach their level — a source of some anxiety to which we turn in a later section. Throughout the interviews, one gets a sense of concern to ensure that the program “fits into the regular schedule, so that it’s not a major interfering

The genesis of a district-wide Spanish FLES program

factor; [that time is found so that] Spanish does not cause a major disruption” for the classroom teacher. This same sense of support carries over to the second, and now the third year, as evidenced by the ways in which the Spanish teachers have worked together taking “a lot of time for planning the curriculum … and anticipating problems that might occur”. The first Spanish teacher also provided support for the parents. She started a newsletter for them; did a survey to find out their attitudes toward the program; and participated in the district’s homework “hotline” which is updated daily. In summary, one gets a sense of teamwork, of collaboration, of staff working together to make the Spanish program the best that it can be — and to ensure that the program is central, rather than peripheral, to the life of the school.

The contribution of physical consolidation Although not appearing as extensively or elaborately in the interviews as other themes, reflections on the consolidation of several community schools into one new facility in 1996 was cited as one factor facilitating the implementation and contributing to the overall feasibility of the program. Comments concerning the physical setting in which the program was housed were offered across a majority of the respondents. In the words of the superintendent, the district was a “sleeping giant”, “a good district” with the potential “to become great one”. According to Sulkowski “the combination of new people, the consolidation of buildings, and fiscal responsibility came together to create the right situation” to realize this transition to an exemplary school district. But the benefits gained through consolidation for the district in general, and the FLES program in particular, were not realized without effort or anxiety. The former elementary school principal pointed out that although “a new building would improve education”, the physical move itself, coupled with the simultaneous restructuring effort, was a major point of stress. Adding to this anxiety were the former principal’s concerns about the myriad curricular innovations and staff changes that a newly consolidated school would present — a new technology lab, more complex scheduling, the need to establish rapport with the community, a new faculty configuration, and the FLES program. He was responsible for the “building of buildings” and for closing and making the move from community schools to this state-of-the-art teacher-designed facility. He would be “scrutinized by the school board, by the community, and

251

252

G. Richard Tucker, Richard Donato and Kimmaree Murday

by the teachers”. Despite the understandable apprehension that consolidation created for this former principal, he remained throughout a strong proponent of the FLES program. In spite of the complex changes that were brought about because of consolidation, the former principal explains that his expectations for the program did not go unrealized. During that first year, he “saw the kids speaking phrases, and their enthusiasm and excitement; it was enlightening. ( … ) These little people have an enormous capacity for learning”, demonstrated in the children’s ability to acquire another language, according to their former principal. From his perspective, it was clear that consolidation into one “school of schools”, although not an effortless task, created possibilities for creative and beneficial educational changes, such as the FLES program, that “benefited the kids” and “created a sense of pride in the district”. From a FLES teacher’s perspective, the consolidation and the derivative benefits of providing all instruction in one building was also a critical factor in alleviating her apprehensions about the physical demands of teaching 11 sections a day of Spanish. Reflecting on her thinking during an interview for the position of kindergarten teacher in 1997, the now present second grade teacher stated that she “was very confused as to how the program was going to work, because she didn’t realize [at the time] that the children were situated in one building”. It is commonplace for FLES teachers to be itinerants, traveling from classroom to classroom and building to building (Donato, Antonek and Tucker 1994), and prior knowledge of this unfortunate reality clearly shaped this teacher’s perceptions of her instructional day during the interview. Her confusion, however, was resolved when she learned that “[they] were all in one building, so it seemed easy”. As the intermediate school principal points out “consolidation was a strength; no one teacher would have been able to do what the first grade teacher was asked to do if there had been moving around to consider”.

Concerns for the future Respondents were asked to discuss their long term prognosis for the program and, thus, several concerns related to the future of foreign language learning in the district were brought sharply into focus. The most resounding theme reflected a realization that issues of articulation will be critical if the district is to have a coherent and viable foreign language program across 13 years of instruction. Several respondents mentioned the need to re-think the curriculum in the middle school (grades 6–8) and high school (grades 9–12). As the chairperson of the high school foreign language department stated, “There will need to be

The genesis of a district-wide Spanish FLES program 253

drastic changes in the curriculum in the later years of schooling” but she hastened to add that she sees this “as a wonderful problem”. Similar comments across several respondents made clear that the district anticipates the effect of such a program on all level of language instruction and recognizes the centrality of language instruction for the total school curriculum. What is striking in the interview data are the possible solutions envisioned to the “wonderful problem” of restructuring a foreign language curriculum upwards toward high school. Many respondents provided concrete suggestions for a well articulated program model, course content, and teaching personnel. The second grade teacher spoke of later courses containing challenging subject matter in Spanish that could be taught by highly qualified instructors trained to deal with academic content. In the words of the intermediate school principal, the “domino effect” of an elementary foreign language program on later years of instruction requires careful planning and active preparation for the language curriculum in the upper grades. Her suggestions included recruiting highly qualified teachers with teaching certificates in both elementary education and foreign language instruction, and constant attention to the curriculum to ensure that it is developmentally appropriate for the learners and reflective of advances in the field of language learning. The intermediate school principal, a non-foreign language specialist, anticipated that the students will be better able to acquire “additional foreign languages because of their early learning experiences with Spanish”, thus possibly foreshadowing an expansion of foreign language offerings to students in the high school. Finally, the elementary school principal, aware of the current exploratory model in the middle school, stated that this program “will have to be redesigned [and] when [the students] reach high school, Spanish I will be archaic for most students”. Clearly the committee already realizes the dangers of senseless repetition in the foreign language curriculum and the need to build upon what students know and can do (see Conclusion for a fuller discussion of this point). Based upon our participation in meetings of the Foreign Language Committee and our analysis of these interview data, we detected anxiety on the part of non-foreign language intermediate school teachers toward their perceived need to know the language their students will be learning. To address this fear, plans have been made to provide in-service workshops for the intermediate school teachers about the Spanish program and to encourage them to visit the elementary school to observe classes in action. The familiarization programs will also include rudimentary Spanish lessons for small groups of 3 to 5 intermediate teachers given by the high school Spanish instructor.

254

G. Richard Tucker, Richard Donato and Kimmaree Murday

Serious discussions of the impact of the FLES program in the future of foreign language learning in the district and some minor concern for budget cuts under a new administration with a possible set of different priorities were tempered with optimism for the program and the achievements of the children. From the classroom teacher to the superintendent, all expressed a hope that, through this program, the children will build strong functional language proficiency, develop global awareness, and be highly desirable and competitive candidates for jobs after completing their education. As the former elementary school principal pointed out “the ultimate goal is bilingual kids by the time of graduation”. His optimism for the future was also echoed, somewhat metaphorically, by the superintendent — the program “will turn foreign language instruction upside-down … foreign language instruction should start in kindergarten; they will grasp it quickly and I hope I will see this come true … the impact will be dramatic”.

From local themes to professional practice The profile which emerged from the interviews provides, we believe, important lessons for others who may embark upon similar curricular innovations. This study has examined in actual practice key elements in developing an apparently successful educational innovation and in so doing has told the story of one district’s lived experience with FLES. The direction and decisions of this district rested upon the concerns of several important constituents and reflects Markee’s (1997) observation that innovative projects are affected, positively or negatively, by complex sociocultural variables, such as cultural beliefs, political climate, historical and economic conditions, administrative attitudes, institutional support, and technological, sociolinguistic, and language planning factors (see also Holliday 1994). When viewed globally these 6 themes — vision, planning, empowerment, support, physical setting, and future concerns — refract and reflect all of the sociocultural variables listed by Markee and attest to their importance, as well as the need to acknowledge and address openly these factors when designing and implementing new programs. Others in the profession contemplating the development of a program, such as the one we have just described, or monitoring and evaluating current FLES programs, might well be advised to benchmark successes and failures against these themes.

The genesis of a district-wide Spanish FLES program 255

Conclusion At present, the committee has several short-term tasks, and one major longer term goal. In the short run, another teacher must be hired for September 1999 when the program will again expand to encompass all kindergarten, grade 1, grade 2 and grade 3 pupils in the district. Then the group will need to turn its attention to curriculum revision and development, materials selection, etc. for the expansion of the program to the Intermediate School (grades 3, 4 and 5). In addition, the Middle School staff has also expressed a strong desire to “be prepared” when foreign language enters their program in three years. As the Middle School representative on the Committee voiced during a recent meeting “our teachers want to be prepared and some have even started to review their college Spanish”. In addition, as discussed in the interview data, the current high school Spanish program will need to be thoroughly revised for subsequent cohorts of students who will bring to the high school language class a “beyond-the-basic” level of proficiency. The district and the current Spanish teachers will face a major challenge in developing a rich content-based Spanish program that will allow these students to continue to develop cognitive and academic language proficiency in both English and in Spanish by the time of their graduation. This concern for articulation is well founded in light of the failures of FLES programs in the 1960s (Rosenbusch 1995). One commonly observed phenomenon during that period was that former FLES students often repeated basic language lessons when they entered high school. This repetition of previously learned material resulted in a severe lack of motivation in students and diminished their interest and enthusiasm for language study. The source of this instructional discontinuity in language study has been traced to the lack of clearly articulated shared goals and outcomes for language learning in a seamless sequence of instruction. It is not surprising that in foreign language education today articulation of programs and studies still dominate the professional literature, grant-funded projects, and conference presentations. The Committee is fully aware of these issues and is currently taking steps to assure that transitions between instructional units and course outcomes will expand student proficiency rather than recycle rudimentary skills each year.

256

G. Richard Tucker, Richard Donato and Kimmaree Murday

Reflections on a successful and satisfying collaboration There are a number of factors that have contributed to the success of this project to date. The first that comes to mind is the key role played by the superintendent. By his active participation in all of the Committee meetings, he has provided immediate and visible credibility and value to the activity. He continually reminds the Committee that they are embarking on an innovation by “navigating uncharted waters” that will have far-reaching consequences for the school district in terms of its visibility and reputation. We have also been struck by the extent to which the Committee members — themselves mostly monolingual and monocultural — have embraced the goal of multiple language proficiency and cross-cultural competence for their students and themselves “act as if” they were multilingual and multicultural. Throughout our association with the Committee, we have found the representatives from the district to be continually and genuinely concerned with providing the best possible education for their students that they can. We have never heard a disparaging remark about a pupil, a parent or a community member — rather Committee members express genuine knowledge about and concern for the students’ educational, social, and personal well being. We have found it enormously satisfying to be a part of this Committee. What are some of the issues that have intrigued us over the past three and a half years? Clearly, we have appreciated the opportunity to attempt to extend the generalizeability of some of our ideas about language program evaluation to another setting. But perhaps more importantly, we have enjoyed the challenges (in the words of the superintendent) of “navigating uncharted waters” in helping to write a curriculum for the early grades; in examining the relationship between what goes on in the language arts curriculum to what we are trying to accomplish in the FLES curriculum; in thinking through and sharing with other Committee members various issues related to the introduction of second language literacy in relationship to native language literacy; in examining the complex set of issues related to reporting student progress to parents (we are intrigued, for example, by how parents evaluate the progress of their children in areas in which they — the parents — themselves have no experience); and more generally with the responsibility of injecting substantive issues from time to time into what might otherwise become a procedural dialogue. We have enjoyed witnessing the genesis of this program and being a part of this collaborative achievement thus far, and look forward to its continuation in the future.

The genesis of a district-wide Spanish FLES program 257

Appendix A Interview Protocol 1. All but superintendent: What are your recollections about the initial discussions that led to the implementation of a foreign language program for Chartiers Valley youngsters? …were you in favor of establishing such a program? …dubious about establishing such a program? …were you a proponent? …a healthy skeptic? Tell me a bit about your recollections of the early discussions…and the role that you played in them. 1a. Superintendent: What prompted you to start thinking about this program for Char Valley? …what role, if any, did Chapter 5 have in your thinking about this program for your students? Tell me a bit about your recollections of the early discussions…and the role that you played in them. 2. All: Early on, a Foreign Language Steering Committee, comprised of Char Valley personnel and representatives from the University of Pittsburgh and Carnegie Mellon, was established. You have been actively involved with that Committee. From your perspective, what was the major role (or purpose) of the Committee? …how did you see your role on this Committee? Can you compare the discussions and the activities surrounding the implementation of this foreign language program to the implementation of any other curricular innovation in Char Valley? If so, have the processes, the discussions, the actions taken, been similar or different? Is there anything distinctive about the procedures that were followed with the foreign language program? …is that good, bad or irrelevant? 3. All: The foreign language program seems to have been successfully implemented (at least to date). Is this inference correct? What do you think were the biggest obstacles that had to be overcome to put the program in place? …what were the major strengths on which the district was building when it decided to put the program in place? …what lessons do you think have been learned in terms of implementing future innovative programs? 4. All: What do you think is the long term prognosis for the program? …do you have any concerns about the sustainability of the program over the long run? …are there issues, in particular, that you think need to be attended to in order to maximize the likelihood of successful full implementation from kindergarten through grade 12? 5. Superintendent, Director of Curriculum and School Board Member: What do you see as the major benefits (e.g., to the pupils, to the participating teachers, to the School District) of successful program implementation? …what do you see as the major costs of program implementation?



258

G. Richard Tucker, Richard Donato and Kimmaree Murday

6. All: If you were faced with the question of implementing such a program tomorrow, with the benefit of hindsight, what questions would you raise? …what advice would you give to planners? …what cautions would you have for those involved? 7. All: Are there any other comments that you’d like to make about the program?

Appendix B Interview Respondents Dr. Bernard Sulkowski Dr. Katherine Gori Ms. Ellen Shar Mr. Jerry Koziak Mr. Kevin Kuhn Dr. Sarah Tambucci Mrs. Betty Kripp Mrs. Cassie Kuzniewski Ms. Lisa Bischoff Mrs. Helen Lucatorto Ms. Bridget Kelly

Superintendent Director of Instruction Foreign Language Department Head Former Primary School Principal Primary School Principal Intermediate School Principal Intermediate School Department Head Spanish Teacher Spanish Teacher Primary School Department Head School Board Member

Notes * Tucker is Professor and Head of the Department of Modern Languages at Carnegie Mellon University. Donato is Associate Professor in the Department of Instruction and Learning at the University of Pittsburgh and Coordinator of the Foreign Language Education specialization. Murday is a doctoral candidate in the Second Language Acquisition program of the Department of Modern Languages at Carnegie Mellon University. 1. It is interesting to note that these desired outcomes are quite similar to the rank-ordering of goals expressed by parents in our longitudinal study of a Japanese FLES program (Donato, Tucker and Antonek 1996; Tucker, Donato and Antonek 1996; Donato, Antonek and Tucker 1994).

References Baranick, W. and Markham, D. 1986. “Attitudes of elementary school principals toward foreign language instruction”. Foreign Language Annals 19: 481–489.



The genesis of a district-wide Spanish FLES program 259

Branaman, L. E. and Rhodes, N. C. 1997. A National Survey of Foreign Language Instruction in Elementary and Secondary Schools 1997. Washington, DC: Center for Applied Linguistics. Carroll, J. B. 1967. The Foreign Language Attainments of Language Majors in the Senior Year: A Survey Conducted in U. S. Colleges and Universities. Cambridge, MA: Harvard University Graduate School of Education. Dickson, P. and Cumming, A. (eds). 1996. Profiles of Language Education in 25 Countries. Slough, Great Britain: National Foundation for Educational Research. Donato, R. and Terry, R. M. (eds). 1995. Foreign Language Learning: The Journey of a Lifetime. Lincolnwood, IL: National Textbook Company. Donato, R., Antonek, J. L. and Tucker, G. R. 1994. “A multiple perspective analysis of a Japanese FLES program”. Foreign Language Annals 27: 365–378. Donato, R., Tucker, G. R. and Antonek, J. 1996. “Monitoring and assessing a Japanese FLES program: Ambiance and achievement”. Language Learning 46 (3): 497–528. Eddy, P. A. 1980. “Foreign languages in the USA: A national survey of American attitudes and experience”. Modern Language Journal 64 (1): 58–59. Holliday, A. 1994. Appropriate Methodology and Social Context. New York: Cambridge University Press. Markee, N. 1997. Managing Curricular Innovation. New York: Cambridge University Press. National Standards in Foreign Language Education Project. 1996. Standards for foreign language learning: Preparing for the 21st century. Yonkers, NY: ACTFL. Rosenbusch, M. 1995. “Language learning in the elementary school: Investing in the future”. In Foreign Language Learning: The Journey of a Lifetime, R. Donato and R. M. Terry (eds), 1–36. Lincolnwood, IL: National Textbook Company. Pesola, C. A. 1992. “Notes from the president”. FLES News 6 (1): 3. Tucker, G. R., Donato, R. and Antonek, J. L. 1996. “Documenting growth in a Japanese FLES program”. Foreign Language Annals 29 (4): 539–550.



The distinctiveness of Applied Linguistics in Australia A historical perspective Tim McNamara and Joseph Lo Bianco University of Melbourne / Language Australia

Introduction In this paper an attempt is made to identify the distinctive character of Applied Linguistics in Australia. The origins and character of Applied Linguistics in Australia are interestingly different from those of the main traditions of Applied Linguistics, those of the United States and the United Kingdom. In Australia, the founders of the Applied Linguistics Association of Australia were in the main university teachers of French and other modern languages, who were interested in researching and promoting the learning of modern languages in Universities and schools. English language teaching, in other words, was not the primary context of origin. This fact has had significant consequences for ongoing developments of the field in Australia, particularly the appearance and implementation of the National Policy on Languages. Through a series of extracts from interviews with Dr Terry Quinn, a leading figure in the early history of Australian Applied Linguistics, the unique character of Australian Applied Linguistics is revealed.

The British tradition of Applied Linguistics As a point of reference, we will begin our discussion by briefly tracing the history of Applied Linguistics in Britain, whose origins go back to 1957, when the first academic program in Applied Linguistics was started at the University of Edinburgh. The establishment of a postgraduate Diploma in Applied Linguistics was in response to overtures from the British Council and the

262 Tim McNamara and Joseph Lo Bianco

Foreign Office, and built on Edinburgh’s existing strengths in linguistics and phonetics. One of the key figures in the establishment of the initiative at Edinburgh was the phonetician David Abercrombie, who had taught English as a Foreign Language in the Middle East; the first Director of the Applied Linguistics program (Ian Catford) was a phonetician from Abercrombie’s department. The origins of Applied Linguistics at Edinburgh drew, according to Davies (1993), on the “tradition of descriptive language studies and of anglicist research, in particular the work on lexicography and in phonetics as a means of teaching English to foreigners”. This intellectual resource represents, for Davies, one aspect of the beginnings of Applied Linguistics in the UK, the other being the political pressure of the post-colonial period where the teaching of English as a second language emerged as an important issue. Davies quotes from the University of Edinburgh Calendar of 1957 to define the role and scope of Applied Linguistics at Edinburgh as it was understood at that time: The primary object … is to provide experienced overseas teachers, particularly those who are actively concerned with the control of English language teaching policy and the training of teachers, with an intensive training in the disciplines which are relevant to language teaching and in the methods of research in this field.

He also points out that the creation of an independent Department of Applied Linguistics at Edinburgh from 1964 was presided over by Pit Corder, “who, in the characteristic pattern of that generation of British applied linguists, had long experience of overseas language teaching, in this case as a British Council officer”. Elsewhere (Davies 1992) has written: Corder’s career started as was normal in the British tradition of Applied Linguistics with the hands-on, practical work of the British Council’s English teaching operations in non English speaking countries where he worked for a number of years on the practicalities of classroom teaching, materials preparation and syllabus design.

Davies explains that changes in the 40 years since that time have included “a growing realisation in other sectors of the language teaching profession that Applied Linguistics has a part to play in the education of teachers of languages other than English”. Given the strong influence of British academic traditions in Australia, it might have been expected that Australian Applied Linguistics would have mirrored the British model. In fact, something quite different happened. If Pit Corder is emblematic of the British tradition of Applied Linguistics, then Terry

Applied Linguistics in Australia 263

Quinn, former Director of the Horwood Language Centre at the University of Melbourne, represents the characteristic essence of the very different Australian tradition.

Applied Linguistics in Australia Applied Linguistics in Australia originated in an entirely different way. Uniquely, its origins lie in the teaching of modern languages in the Universities. Key figures in its development were Terry Quinn, Wilga Rivers and Keith Horwood. The Applied Linguistics Association of Australia (ALAA) was formed in 1976; it grew directly out of a Language Laboratory Workshop, established over a decade earlier, held regularly as part of the annual meeting of the Australasian Universities Languages and Literatures Association (AULLA), the peak body for university modern language teachers. Terry Quinn, arguably the most distinctively Australian applied linguist, was Director of the Horwood Language Centre at the University of Melbourne from 1975 to 1989 and the founder of the graduate Applied Linguistics program at Melbourne. (The Centre was named for his predecessor, Keith Horwood, who had died suddenly in 1974.) Quinn was a high school teacher of French in Melbourne in the early 1960s when he read an article by Albert Valdman entitled “Vers l’application de la linguistique à l’enseignement du français parlé” (Le Français dans le Monde 1962, 7: 1).1 This marked the beginnings of Quinn’s interest in Applied Linguistics. Quinn recalls it in this way:2 I got really interested in the whole conception of Applied Linguistics because of language laboratories and because of one article by Albert Valdman … Valdman wrote an article about structural linguistics and language laboratories, and the structural linguistics really turned me on. What he was doing with that article was analysing … if you did a grammar of spoken French from the point of view of structural linguistics you’d end up with a very different grammar from what is traditionally taught. I could see all that and it lit so many light bulbs in my head. And he related it to language laboratories and said ‘If you’re going to do language practice of the kind I’m outlining here the language laboratory is a good way to do it’ — mimicry memorization methods and so on. This turned me on. And this was the beginning of my genuine interest in linguistics and Applied Linguistics.

264 Tim McNamara and Joseph Lo Bianco

At the time, language laboratories were being sold to University language departments but the expertise to exploit them was often lacking. According to Quinn: They had all the cachet of modern science and technology and so on — miracle cures for language learning deficiencies and all that sort of nonsense — and of course they were being pushed by American companies. Nobody knew what to do with them. … And out of that need, within AULLA, the Australasian Universities Languages and Literatures Association, a sub-group was formed called ‘The Language Laboratory Workshop’ and at each meeting of AULLA there would be probably a day or half day a day something like that devoted to this Language Laboratory Workshop.

Quinn himself, who in response to Valdman’s article had created a home-made language laboratory for use at his school, was invited to speak at one Workshop: I had a certain amount of modest notoriety as a secondary school teacher who was doing these things and I can remember … being invited to one of these AULLA Language Laboratory Workshops to talk about what I did with the language laboratory.

According to Quinn, the existence of the AULLA group was one of two factors leading to the formal establishment of Applied Linguistics in Australia; the other was the presence and influence of Wilga Rivers. Rivers was a former high school teacher of French in Melbourne who, faced with redundancy in her 40s, had gone on to do a Masters degree and then to the United States to do a Ph.D. With extraordinary timing her thesis delivered the critique of audiolingualism the field had been waiting for. The other one I would undoubtedly have to say was Wilga Rivers. … She must have come back to Australia about 1965 I think. … Wilga … came on the scene as the great prophet of the new approach to language teaching and professionalism in language teaching. Those two things made this AULLA Language Laboratory Workshop quite a vigorous, attractive [forum] … it attracted a lot of people. And it was out of that that ALAA grew.

(Quinn notes in passing that ironically, the ‘new approach’ to language teaching which Wilga Rivers was espousing was the very audiolingualism that her doctoral thesis had critiqued: It was very surprising that Wilga rocketed to fame in the United States because of her doctoral thesis which was a critique of audiolingualism. And yet she came back to Australia pushing audiolingualism, pushing modern methods, pushing new methods which were basic audiolingualism … The French

Applied Linguistics in Australia 265

program that she implemented at Monash [University, in Melbourne] … was basically straightforward audiolingualism.)

The Applied Linguistics Association of Australia (ALAA) was formed in 1976 at a conference at the University of Newcastle organized by Gay Reeves, a lecturer in French there. Gay was the first Secretary; Ross Steele, a lecturer in French at the University of Sydney, was its first President. Quinn became editor of the Association’s journal, the Australian Review of Applied Linguistics; he was also editor of Babel, the journal of the Australian Federation of Modern Language Teachers Associations. In the early 1970s Quinn had gone to Ohio State University to be trained in its doctoral program; when he returned to Australia he was in a position to exert great influence on the course of Australian Applied Linguistics. Quinn recalls: I was incredibly lucky to have had that extremely thorough professional training from my doctoral studies at Ohio State. And I was an extremely rare bird in Australia. I was probably the only one at that time — I mean we’re talking about the early 70s — a person thoroughly well trained at the doctoral level in all aspects of language teaching and language planning and policy and all that — very rare indeed.

(Rivers by this time had earned a Professorship in the United States; she completed her career at Harvard). Quinn was appointed Director of the Language Centre at the University of Melbourne in 1975, following the untimely death in 1974 of its first director, Keith Horwood (after whom it was the named). Horwood had a background in German, and in the 1960s had been appointed lecturer in Science German at the University to help Faculty of Science students meet the requirement to complete the compulsory reading course in a foreign language (one or more of German, French and Russian). Horwood eventually became head of the science languages section in the Faculty. In Quinn’s words: His [Horwood’s] interest was in efficient language teaching. His students were science students who really had a very pragmatic need, who knew exactly what they needed and wanted in their language courses, and so Keith was interested in effective language teaching for that reason. It must have been about the same time that I was getting turned on by structural linguistics and language laboratories Keith must have been too because he got interested in modern linguistics and what was happening in [Britain] … Keith’s orientation was entirely British by the way, whereas mine was American.

266 Tim McNamara and Joseph Lo Bianco

Horwood was one, perhaps the most impressive, of a number of similar figures emerging at that time: In each University as the impact of this language laboratory phenomenon — ‘update the language teaching’ process — began, one figure emerged. … Often they were people that knew how to use language laboratories and often that meant they were from Point Cook,3 they were ex-Air Force people from Point Cook — Robert Hooke at La Trobe [University, in Melbourne], for instance.

The influence of modern language training in military language schools is a notable feature of the origins of Australian Applied Linguistics. Quinn’s new position at Melbourne gave him time to think and write, and he chose to work to influence policy. He was effective in this by virtue of his training, and because of his reputation as a thinker, achieved through his influential editorials in Babel (Ozolins 1995 has traced their powerful influence on language policy in Australia): It was a very deliberate decision on my part to make Babel editorials say something … my editorials were a clearly designed series of statements about key issues in language policy. And I can remember I did make that decision that I was going to be able to have an impact by writing editorials that said something quite sensible and strong and cogent and gave the evidence and stirred people up and I think they did to some extent. … I regularly got involved and invited onto government committees and I’m sure it was basically because of that.

Quinn’s work on such committees involved the administration of policy, and the formulation of policy advice in language fields, particularly in the area of immigrant languages in an era of official multiculturalism. For example, he chaired the Victorian Advisory Committee on Migrant and Multicultural Education (VACMME) from 1978 to 1982. It was in the context of this committee that in the early 1980s he met Joseph Lo Bianco. Lo Bianco completed a Masters thesis in Applied Linguistics with Quinn as supervisor on an ItalianEnglish bilingual education program in Melbourne (Quinn was fluent in Italian from his years in a religious order in Rome). Lo Bianco subsequently both wrote and secured the adoption of the enormously influential National Policy on Languages (Lo Bianco 1987). This policy and its associated funding decisively influenced the direction of research in Applied Linguistics in Australia for a decade. Characteristically, the bulk of this was in relation to languages other than English (LOTE), although now it was not the traditional European

Applied Linguistics in Australia 267

languages of University study but the languages of immigrant communities, and of Australia’s new trading partners in Asia. When Quinn made the key appointment to staff the new graduate program in Applied Linguistics at Melbourne in the late 1980s, immediately prior to his premature retirement through ill health, he favoured the choice of Tim McNamara for the position. Significantly, McNamara, an Australian, had studied Applied Linguistics at Birkbeck College at the University of London, perhaps the least typical of the British programs in Applied Linguistics. It was staffed by two psychologists and two former University teachers of French; the course specialised in the study of bilingualism; none of the staff had training or experience in English language teaching, the staple of most British applied linguistics courses. Quinn had visited Britain to examine its graduate training in the area in the early 1980s, and had chosen Birkbeck as the most likely model for the new Melbourne program. This choice was entirely characteristic if his orientation to Applied Linguistics, and of the now well established Australian tradition.

The emergence of English When work in the Applied Linguistics of English did emerge as a powerful force in Australian Applied Linguistics, it did so again in distinctive ways. McNamara: That’s something very, very distinctive about Applied Linguistics in Australia isn’t it, … if it came out of modern language teaching? Applied Linguistics in America and the UK came out of English language teaching. Quinn: Not at all in Australia. Not at all. They were Johnny-come-latelys. … Here was this enormously important ESL industry in Australia and it was quite untouched by and uncontributing to Applied Linguistics in the early days. It was not until very, very much later that the whole AMEP4 ESL industry became part of Applied Linguistics. It was unquestionably tertiary modern language teaching where the origins … lay.

One of the crucial influences on the emergence of work on English in Australian Applied Linguistics was Michael Halliday. Halliday moved to Australia to take up a Chair in Linguistics at the University of Sydney in the late 1970s. He had been one of the original figures in the Edinburgh program, although there his contribution was the development of a new linguistic theory and he was not an EFL specialist — in fact, he had done significant work on Chinese. According to Davies (1992):

268 Tim McNamara and Joseph Lo Bianco

Michael Halliday, a lecturer in General Linguistics in McIntosh’s department, made a major contribution to lecture courses in the School [of Applied Linguistics]. At that time Halliday was developing his systemic grammar and it could therefore be claimed that systemic linguistics was invented in his courses in this new School of Applied Linguistics.

At Sydney, Halliday’s work inspired applications in the field of mother tongue education before it had any significant influence on ESL/EFL. The Applied Linguistics of English as a second language was developed in the context not of English has an international language but of the teaching of English to adult immigrants, initially in the work of Ingram, Nunan, Burton, Brindley and others in the early 1980s. This work was greatly strengthened by the establishment of the National Centre for English Language Teaching and Research (NCELTR) at Macquarie University in the late 1980s, funded by the Department of Immigration and Ethnic Affairs. (It was headed by another British applied linguist, Chris Candlin.) Thus the most distinctive Australian work in the Applied Linguistics of English has been significantly different from the mainstream British traditions.

Conclusion This paper has argued for the distinctiveness of Australian Applied Linguistics, and has investigated this in particular by examining the role of Terry Quinn. Much more work needs to be done to understand the Australian tradition fully, and to carry out similar analyses of traditions elsewhere, particularly in Canada, the United States and the U.K. This paper explores a theme that Bernard Spolsky has long emphasized in his research, namely, the historical and contextual situatedness of intellectual developments in our field, epitomized in his marvellous Measured Words (1994). We have taken our inspiration for this paper from him.

Notes 1. We have drawn on Steele (1989) for many of the details of Quinn’s career. 2. This and subsequent quotes from Terry Quinn are drawn from transcripts of interviews conducted with him by the first author in Melbourne during 1998.



Applied Linguistics in Australia 269

3. The Royal Australian Air Force School of Languages at Point Cook Air Base, near Melbourne, the Australian equivalent of the Defense Language Institute in Monterey, California. 4. AMEP = Adult Migrant English Program, an extensively funded Australian Government program of English language teaching for adult immigrants.

References Davies, A. 1991. “British applied linguistics: The contribution of S. Pit Corder”. In Festschrift to Claus Faerch, R. Phillipson et al. (eds), 52–60. Clevedon: Multilingual Matters. Davies, A. 1993. Real Language Norms: Description, Prescription and their Critics. Chair of Applied Linguistics Inaugural Lecture, University of Edinburgh, November. Lo Bianco, J. 1987. National Policy on Languages. Canberra: Australian Government Printing Service. Ozolins, U. 1995. Language Policy in Australia. Cambridge: Cambridge University Press. Spolsky, B. 1994. Measured Words. Oxford: Oxford University Press. Steele, R. 1989. “Some reminiscences”. In Language, learning and community: Festschrift in honour of Terry Quinn, C. N. Candlin and T. F. McNamara (eds), 223–225. Sydney, NSW: National Centre for English Language Teaching and Research, Macquarie University.



Educational linguistics as a field A view from Penn’s Program as it approaches its 25th anniversary Nancy H. Hornberger University of Pennsylvania

Introduction In 1972, Bernard Spolsky proposed the title “educational linguistics” for a discipline whose primary task would be “to offer information relevant to the formulation of language education policy and to its implementation” (1974c: 554). He affirmed that it “should be a problem-oriented discipline, focusing on the needs of practice and drawing from available theories and principles of many relevant fields including many of the subfields of linguistics” (1975: 347). Shortly thereafter, two doctoral programs in Educational Linguistics were inaugurated at U.S. universities — one at the University of New Mexico, directed by Spolsky and closely linked to the Navajo Reading Study being carried out there and one at the University of Pennsylvania’s Graduate School of Education, inaugurated under the aegis of Dell Hymes and the direction of Nessa Wolfson.1 Educational Linguistics at the University of Pennsylvania traces its beginnings to 1976, when Hymes appointed Wolfson lecturer in education and assigned her the task of creating the Educational Linguistics program, which would come to encompass not only the Ph.D. specialization but also a master’s specialization in Teaching English to Speakers of Other Languages (TESOL), and soon thereafter (1978), another master’s specialization in Intercultural Communication.2 In the ensuing years, the program took on additional faculty; Teresa Pica, one of the first graduates of the program (Pica 1982), was appointed assistant professor and director in 1983,3 I joined in 1985, and after Wolfson’s untimely passing in 1989, Rebecca Freeman was recruited as third member in 1992. As Educational Linguistics at Penn approaches its 25th

272

Nancy H. Hornberger

anniversary, and in honor of Professor Spolsky who provided both a title and a programmatic description for the field, it seems appropriate to take a retrospective and prospective look at this one program’s approach to the practice of educational linguistics. In keeping with Spolsky’s initial formulation that educational linguistics should take the practice of education as its starting point, I will begin from the practice of educational linguistics in the Penn program, moving from there to implications for the field as a whole (rather than the reverse). In the sections which follow, I take up various aspects of the practice of educational linguistics at Penn and discuss them in relation to issues that have been raised in the literature about the definition, nature, and scope of the field. I conclude with a brief comment on the relationship of educational linguistics to applied linguistics and other disciplines. It is my hope that this exploration of a particular set of practices might contribute to the advancement of the field of educational linguistics.4

Educational linguistics defined The Educational Linguistics Ph.D. specialization at Penn is one of nine doctoral specializations offered at the Graduate School of Education. The Educational Linguistics handbook introduces the doctoral specialization in the following way:5 The Ph.D. program in Educational Linguistics integrates scholarship, training, and research in linguistics as they relate to theory, practice, and policy in education. The program maintains a perspective on issues in linguistic and cultural diversity and approaches to language learning and teaching that embraces local, national, and international interests. Research interests of Ph.D. candidates currently enrolled in the program include: second language acquisition; language choice, maintenance and shift; language and ethnicity; descriptive analysis of speech acts and discourse; educational implications of linguistic diversity; language planning; bilingual education; spoken interaction in professional settings; and biliteracy. Graduates can expect to find teaching, administrative, and research positions in colleges and universities, and administrative, research and advisory posts in government, community and private organizations. All students enrolled in this program are expected to gain a solid foundation in linguistics. For this purpose, students take courses in the Department of Linguistics as well as in the Graduate School of Education. (Educational Linguistics Handbook 1997–98: 21).

Educational linguistics: A view from Penn’s Program 273

The above introduction offers a brief definition of educational linguistics, as well as a suggestion of its scope and relationship to linguistics. The handbook goes on to outline a 20 course curriculum, including seven core courses, four distribution courses (two in linguistics and two in education), and two research methods courses, as well as inquiry skills, candidacy, comprehensive examination, and dissertation requirements. Beginning with the first Educational Linguistics Ph.D. in 1981 (Zentella 1981), the Faculty of the Graduate Group in Education has approved more than 70 candidates for the Ph.D. degree with specialization in Educational Linguistics, approximately two-thirds of them women and one-half international.6 Consistent with the professional positions outlined in the above introduction to the program, graduates have gone on to hold academic, research, and administrative posts in institutions of higher education across the country and around the world, in departments of education, linguistics, applied linguistics, English, English as a second language, foreign language education, multilingualmulticultural studies, anthropology, Japanese language and literature, and Black and Puerto Rican studies, and in international and intensive English language programs, among others.7 The conception of educational linguistics enunciated in the program handbook, with its emphasis on the integration of linguistics and education, the close relationships among research, theory, policy, and practice, and on language learning and teaching as the core focus, is consistent with the field as it has been both explicitly and implicitly defined in the literature. Spolsky’s definition, above, explicitly specifies that the discipline should focus on language education policy and implementation and that it should take a problem- and practice-oriented approach, and these are the crucial characteristics he returns to again and again in his writings. In introducing the section on educational linguistics in Current Trends in Linguistics, he writes that he and the contributors set out “to show how linguistics and its various fields can help define and solve problems that reflect the centrality of language in the educational process” (1974a: 2024), again alluding to the focus on language in education, the problem-solving orientation, and the link to linguistics. In his volume entitled Educational Linguistics, he goes on to write that “the field of language education must depend on a wise, soundly-based, but modest set of principles and practices derived from the relevant theoretical and empirical disciplines. It is the primary task of the field I call educational linguistics to provide such a basis” (1978: 175). Here again, he takes the practice of language education as a starting point and looks to educational linguistics to draw from

274 Nancy H. Hornberger

relevant related disciplines to provide needed principles to guide that practice. Like Spolsky, Shuy also sees an important role for educational linguistics in relating linguistics and its subfields (e.g., sociolinguistics and psycholinguistics) to relevant teaching and learning (1981: 460), and Pica sees it as a problem- and practice-based field “whose research questions, theoretical structures, and contributions of service are focused on issues and concerns in education” (1994: 265). Others define educational linguistics implicitly by what is or is not included in their discussion under that title. Stubbs’ volume entitled Educational Linguistics (1986) is perhaps the most elaborate of these (we will return to his view of educational linguistics below), but others have also approached it this way. Under the title of educational linguistics, Smitherman (1979) addresses herself to black linguistics, Suardiaz and Domínguez (1987) to mother tongue teaching, Myers (1994) to second language teaching, and Freeman (1994) and Brumfit (1997) to language teaching. None of them explicitly defines what they mean in using the term, but by implication include their particular topic within the scope of the field; moreover, their discussions advocate the same emphases on the integration of linguistics and education, close relationships among research, theory, policy, and practice, and a focus on language learning and teaching, as articulated in the more explicit definitions of the field above. Van Lier, on the other hand, at the 1994 Georgetown University Round Table on Educational Linguistics, Crosscultural Communication, and Global Interdependence (Alatis, 1994), approaches the definition of educational linguistics explicitly in terms of its substantive content, but posits that in fact it does not exist “as an academic field, subfield, profession or discipline” (1994: 200). Following Bourdieu (1990), he defines a field as a “historically constituted area of activity with its specific institutions and its own language of functioning” (1994: 203) and suggests that for a field of educational linguistics to exist, there would have to be departments, programs, (doctoral) degrees, courses, textbooks, materials, and insights proper to it (1994: 207). At the same conference (but speaking in reference to the field of language testing), Spolsky suggests that “to be considered a profession, a calling needs to have a number of attributes, such as professional associations, textbooks, training programs, journals, conferences, and certification” (1994: 88). I suggest here that, based on a practice approaching 25 years and by criteria such as those proposed above, educational linguistics has indeed earned the right to be considered an academic field. We will consider the nature and scope of the field in terms of the three emphases alluded to above: the integration of linguistics and education (“the

Educational linguistics: A view from Penn’s Program 275

relevance of linguistics for education and the reverse”); the close relationships among research, theory, policy, and practice (“a problem-oriented discipline”); and the focus on language learning and teaching (“scope with depth”); after which, we will conclude with a consideration of educational linguistics as a discipline among other disciplines (“birds on a wire”).

The relevance of linguistics for education and the reverse Anthropological linguist Dell Hymes agreed to become Dean of the Graduate School of Education because he “believe[d] profoundly in the need for change in the way we understand language, and in what we do with language in schools” (1980: 139). It is not too surprising, then, that one of the first things he did was to inaugurate Educational Linguistics with the appointment of Wolfson. Educational Linguistics was housed in one of four newly created Divisions in the school, the Language in Education Division (LED), along with several existing programs which eventually became unified as the Reading / Writing / Literacy program. This alignment has enabled Educational Linguistics students to benefit from language education fields such as Reading and Language Arts, Children’s Literature, Rhetoric, and Adult Literacy, fields that have traditionally been kept separate from language teaching. Situating the Educational Linguistics and Reading / Writing / Literacy programs in close proximity and under one administrative head has over the years led to greater coherence and complementarity between them, to their mutual strengthening; indeed, as the literacy field has evolved to take social, cultural, political and historical context into account (e.g., Street 1984, 1993), it has become increasingly difficult and undesirable to separate the study of language and literacy practices in any setting, in any event. Evidence of this merging of interests can be seen, for example, in the student-edited Working Papers in Educational Linguistics, which announce all of the programs in the Language in Education Division and highlight a selection of divisional courses encompassing reading/writing/literacy as well as educational linguistics offerings, as follows: Sociolinguistics, TESOL Methodology, Structure of English, Educational Linguistics, Second Language Acquisition, Language Diversity in Education, Multicultural Issues in Education, Classroom Discourse and Interaction, Language Planning and Language Policy, Social and Historical Perspectives on Literacy, Teaching Reading to Second Language Learners, and Forming and Reforming the Reading and Language Arts Curriculum.8

276 Nancy H. Hornberger

From its beginnings, Educational Linguistics also sought to maintain ties with the Department of Linguistics in the School of Arts and Sciences.9 To this end, there are a number of institutionalized reciprocities. As mentioned above, Educational Linguistics requires that all its students take a minimum of two courses in Linguistics; for their part, the Linguistics department offers Educational Linguistics as one of eleven areas of study from which their students choose four for their qualifying examinations. Two Educational Linguistics faculty have been appointed members of the Graduate Group in Linguistics; and reciprocally, there have been secondary appointments of Linguistics faculty in the Graduate School of Education. Faculty of both programs serve as needed on dissertation committees or faculty review committees in the other program. Research articles by faculty in one program may appear on course syllabuses in the other and vice versa. There is informal interaction as well, including occasional invitations to present brown bag talks and participate in locally organized conferences or working papers series. Of course, as with any other innovation, and continuing program, there have been differences of opinion, and tensions, across programs both within and outside of GSE. On the whole though, there is acceptance of autonomy within programs and at the same time a mutual recognition of the relevance of one field to the other. The proposal for a field of educational linguistics was premised on the mutual relevance of linguistics and education (Spolsky 1974a: 2021). Van Lier spells this out, arguing on the one hand for the relevance of education for linguistics in terms of (1) the way in which linguistics is taught to future and current teachers and (2) classroom interaction data for linguistic theories; and on the other for the relevance of linguistics for education in terms of (3) language and content teaching in first language/second language classrooms, (4) language across the curriculum, (5) school-community information flow and discourses, (6) school to work discourse transitions, (7) critical linguistics, power and control in classrooms and schools, and (8) classroom interaction, this last which he sees as the core of the educational process and of educational linguistics research (1994: 204–207). As van Lier points out, though, the argument for the relevance of linguistics for education has on the whole been more readily apparent and accepted than the inverse relationship (1994: 204). In his text Educational Linguistics (1986), Stubbs takes a strong stance for the relevance of linguistics to education, arguing, in terms reminiscent of Spolsky’s call for a problem-oriented discipline, for the value of a research paradigm in Kuhn’s (1962) sense, which “tackl[es] a well-articulated set of problems in well-defined ways, with agreed standards of solution and explanation, and

Educational linguistics: A view from Penn’s Program 277

drawing on a consensus of theory” (1986: 233–234). He points out that it is important to distinguish between language in education and linguistics in education (1986: 34), referring to the need to study language “in its own terms” (1986: 232), as a discourse system, rather than treating “language at the level of isolated surface features, ignoring its abstract, underlying, sequential and hierarchic organization” (1986: 243); and he suggests that discourse analysis can be applied to education in direct teaching about discourse and communication and in the study of classroom discourse (1986: 31, cf. van Lier’s 1 and 2 above). Stubbs’ text addresses what educators (educationalists, in his usage) need to know about linguistics and how to teach it to them, taking up in turn such English language education problems as the teaching of vocabulary, reading, and writing. While there has been a general consensus on the relevance of linguistics for education (and far less attention to the relevance of education for linguistics), there is less clarity and perhaps a certain wariness as to the nature of the relationship between them: is it application, implication, interpretation, or mediation? Coexistence, collaboration, complementarity, or compatibility (Pica 1997)? In his early programmatic statements, Spolsky argued that linguistics has applications to and implications for education (1974b: 2034), both directly through language descriptions and secondarily through linguistic subfields like sociolinguistics and psycholinguistics (1978: 2–3; cf. Shuy 1981: 460). Yet, he was careful to note that while linguistics can contribute language descriptions to inform language teaching, a description is not a prescription for teaching (1978: 2–3) and he urged “steering clear of excessive claims that have caused so much damage to all concerned” (1974a: 2021); he emphasized that educational linguistics “should not be, as it often seems, the application of the latest linguistic theory to any available problem” (1975: 347), but rather a problemoriented discipline focused on the needs of practice (see the next section below). While the metaphor of “application” will probably be with us for a long time,10 scholars have recently argued that a view of educational linguistics as applying — or even mediating or interpreting — linguistic theory for the practice of education suggests an inappropriately hierarchical view of the knowledge base of language teaching (Freeman 1997: 194). Stubbs traces a shift in views on the relation between theoretical and applied linguistics in similar terms, from a view of applied linguistics as mediating (interpreting) theory for teachers, to the view that applied linguistics should develop its own model of language. He argues that analysts’ (linguists’) and users’ (teachers’) models are radically different and indeed must be so due to differing aims; analysts need

278

Nancy H. Hornberger

precision for validity and users need a degree of imprecision in order to communicate effectively (Stubbs 1986: 249). “Educational linguistics cannot just be linguistic theory applied to educational practice, … rather the relationship must be reciprocal and dynamic” (van Lier 1994: 203; see also 1997: 97, 101). Pica (1997) celebrates the growing number of relationships available to teachers and researchers, among them coexistence of activities, collaboration of efforts, complementarity of contributions, and compatibility of interests — an egalitarian reciprocity which may well serve as a model for theory and practice in the whole of the field of educational linguistics. In the same vein, Bartels (1999) expresses dissatisfaction with mediation as the role of linguistics in language teacher education (LiLTE) and suggests new roles for applied linguists in LTE, calling for more actual research on the role of LiLTE, for development of pedagogical practices and materials for teaching LiLTE, and for doctoral apprenticeships into teacher education as well as into research.

A problem-oriented discipline A sampling of feature articles in Penn’s Language in Education Division Newsletter of the past few years gives an indication of the practice-based interests and activities of Educational Linguistics faculty and students: a series of short courses on indigenous and intercultural bilingual education taught in South America (LED News, September 1998); an action research collaborative which studies and supports the implementation of a dual language education program at a Philadelphia middle school (LED News, July 1997); and a five-year teacher enhancement program that seeks to effect broad based and long range improvements in the development of children’s science and math literacy skills at the elementary school level (LED News, Spring 1996). Educational linguistics, as practiced at Penn, is not only situated within a school of education, but is also grounded in schools and communities (local and non-local) and geared toward professional practice. Coursework in the program consistently requires students to be in Philadelphia public or private schools or in adult English language/literacy teaching programs, whether for short-term observation or longer-term research projects, and program faculty maintain ongoing contacts and collaborations with teachers and staff of these institutions. The master’s programs in Intercultural Communication or TESOL (which many Educational Linguistics students

Educational linguistics: A view from Penn’s Program 279

complete en route to the doctorate) require internship and service outreach, respectively, as part of the comprehensive examination requirement. Most of the students enrolled in the doctoral program bring with them pressing (or incidental) questions of practice gleaned from their prior or ongoing teaching experience, questions which provide a focus and a prod for inquiry in the classroom and in their studies. Dissertation topics range from ethnographies of bilingual language and literacy practices at home and in school in both immigrant and indigenous language minority communities, to investigations of the acquisition of communicative competence in specific speech acts in ESL, to studies of language and culture learning in language immersion camps or foreign exchange programs, to explorations of the eVects of particular tasks or interaction patterns on second language learning, to interpretive studies of the implementation and impact of language policy on language use and language teaching. Pica notes that educational linguistics research has shed light primarily in two domains of practice: design and implementation of learner-centered, communicative curricula (LCCC) and professionalization of the classroom teacher as decision-making educator. With respect to the former, Pica points out, educational linguists have identified and recorded language used in the professional, vocational and academic contexts toward which learners aim and they have also built on theories of communicative competence (1994: 265). With respect to the teacher as decision-maker, educational linguists have carried out research which teachers can draw on to answer questions such as: (1) “how should a LCCC be organized with respect to classroom content and activities?” (1994: 269), (2) “which types of classroom organization are effective in providing a social and linguistic environment for L2 learning?” (1994: 274), and (3) “how can a LCCC be adjusted and enhanced when exchange of message meaning is not su cient for L2 mastery?” (1994: 276). Consistent with Spolsky’s early formulations which opened the present paper, Pica emphasizes that educational linguistics o ers no prescriptions, but rather a source of information that teachers can apply as they make decisions (1994: 280). More recently, Brumfit (1997) suggests that there is a need for more research into teachers’ (both language teachers’ and other teachers’) explicit beliefs about, and understandings of, language in order to enable us to understand teachers’ central role as educational linguists, that is, as conscious analysts of linguistic processes. The view from Penn’s Educational Linguistics program, as enunciated so clearly by Pica above, is one based in the practice of language learning and teaching. It is a view consistent with Spolsky’s suggestion that “a more productive approach is to start with a specific problem and then look to linguistics and

280 Nancy H. Hornberger

other relevant disciplines for their contribution to its solution” (1978: 2). The problem areas Pica identifies are curriculum and teacher decision-making, problems also identified by Spolsky (along with materials development) in his description of an educational linguistics approach to the Navajo Reading Study begun in 1969 (1975: 349–351; also 1974c). The Navajo Reading Study was itself an attempt to address an even more fundamental problem — the language barrier to education, i.e. the instance where a child acquires a vernacular language informally and is required by the educational system to acquire a different, standard language (Spolsky 1974b: 2029), a problem which recurs for millions of children daily, weekly, and yearly all over the world and is, as Spolsky suggested, a perennial pursuit for educational linguistics. Spolsky’s book, Educational Linguistics (1978), is in effect addressed precisely to the range of issues which require attention in order to address the language barrier, as revealed in his chapter titles, ranging from sociolinguistic issues such as multilingualism, language situations and policies, language, society, and education, and speech communities and schools — to psycholinguistic considerations such as the nature of language, acquisition of language, what it means to know a language, and language, the individual, and education. As with the general consensus on the relevance of linguistics for education mentioned in the preceding section, there is also a general consensus that educational linguistics must take (language) educational practice (rather than linguistics) as starting point. Suardiaz and Domínguez begin from the contact between linguistics and a particular (language) educational problematic, that of mother tongue teaching (1987: 162). Myers, “in an attempt to define more clearly what the subject of inquiry should be in the field of Educational Linguistics, … invited practitioners and students to voice their concerns around major classroom questions related to language … ” (1994: 193). Pakir, writing about multilingual Singapore, makes a plea for educational linguists to “worry less about educational and psychological perspectives and look for the larger goal of achieving success in bilingual education for the community” (1994: 371), by which she appears to be calling for more attention to educational practice and less to linguistic theory. Even Stubbs, who on the whole begins from linguistics rather than education, introduces a number of the linguistic topics he takes up throughout the book on the basis that they (actually or potentially) present problems to educationalists: e.g., the teaching of vocabulary (1986: 112), the model of language underlying the concept of oracy (ability in spoken language) (1986: 142), the diagnosis of semantic pragmatic disorder (1986: 174), and

Educational linguistics: A view from Penn’s Program 281

problematic characteristics of the English writing system (1986: 224); and devotes his final chapter to “ways in which linguistic theory should take more account of practice” (1986: 246). In sum, educational linguistics takes as its starting point the practice of (language) education, addressing educational problems and challenges with a holistic approach which integrates theory and practice, research and policy. Stubbs recognizes this when he outlines description, theory, and practice as the three ways in which any linguistic topic of interest to educationalists must be approached (Stubbs 1986: 7). Smitherman (1979) enunciated it clearly and early on with respect to formulating an adequate theory of pedagogy for AfricanAmerican children (many of whom arrive at school speaking a vernacular language different from the standard), calling for a holistic approach to language that would encompass theory and research within a paradigm that allows for the analysis of speech and language systems in their socio-cultural reality, policy and planning that would put the study of black speech in school, address testing issues, and push for national policy affirming all languages and dialects, and implementation and practice that would adopt a theme of pedagogy and knowledge for liberation for the community, establish training in language and culture of blacks for all teachers, and promote recognition that “everybody needs communicative competence” (1979: 210).

Scope with depth Communicative competence, first proposed by Hymes in 1966 (1972) in reaction to Chomsky’s (1965) use of the term competence in a much narrower sense, describes the knowledge and ability of individuals for appropriate language use in the communicative events in which they find themselves in any particular speech community. This competence is by definition variable within individuals (from event to event), across individuals, and across speech communities, and includes rules of use as well as rules of grammar. Hymes’ functional and multiple conception of language ability and use in communicative context gave impetus to the development of not only a whole branch of sociolinguistics (the ethnography of communication) but also a language teaching movement (communicative language teaching), both of which have endured to the present. The influence of these ideas on Penn’s Educational Linguistics program is readily evident, perhaps most noticeably in the inclusive, sociocultural ap-

282

Nancy H. Hornberger

proach to language education practiced in the program, an approach which, among other things, emphasizes the learning and teaching not only of linguistically defined grammatical knowledge (rules of grammar) but also of culturally embedded ways of speaking (rules of use); acknowledges the role of not only the immediate interactional context but also the historical, sociocultural, economic, and policy context surrounding language learning and teaching; recognizes the value of learning and teaching not just one standard language variety, but multiple varieties and patterns of language use; and perhaps most importantly, addresses not just language learning and teaching per se, but also the role of language in the construction and negotiation of both academic knowledge and social identity. Hereinafter, I will signal this last triple emphasis with the phrase “(the role of) language (in) learning and teaching”.11 At Hymes’ very first meeting with the Faculty of Education in the spring of 1975 (before his actual appointment as dean), he announced his intention to develop two academic emphases under his deanship, namely educational linguistics and the anthropology (or ethnography) of education (Erling Boe, personal communication, 9 September 1998).12 In the ensuing years, there emerged “an environment favorable to interests in language and anthropology/ ethnography, involving a variety of people, some there only for a while” (Hymes, personal communication, 26 October 1998). We have mentioned above the inauguration of the Educational Linguistics program and the Language in Education Division as a reflection of his emphasis on educational linguistics; similarly, Hymes’ goals with respect to the anthropology/ ethnography of education were infused into the Education, Culture, and Society Division (now a program within the Educational Leadership Division). In both cases, the one emphasis has informed the other, and, indeed, a sociocultural approach to language and a linguistically informed understanding of sociocultural context have permeated into other parts of the school as well.13 In keeping with this inclusive, sociocultural view of language in education, the comprehensive examination in Educational Linguistics identifies six areas of coverage (of which students choose three to be examined on): microsociolinguistics; macrosociolinguistics; language teaching methods and program design/evaluation; language planning and policy/educational policy; language acquisition: first and second; and interdisciplinary perspectives on educational linguistics. Similarly, the Working Papers in Educational Linguistics, fifteen volumes of which have been published under student editorial direction since 1984, solicit papers on topics “ranging from speech act analysis and classroom

Educational linguistics: A view from Penn’s Program 283

discourse to language planning and second language acquisition” (1997, 13 (2), inside back cover). This inclusive, sociocultural view of communicative competence and communicative contexts is also reflected in the range of dissertation topics pursued in Penn’s program, including (1) descriptions of native speaker and/or non-native speaker communicative competence for various speech acts and social networks and identities: Speech acts: Benander 1993 on positive evaluations, Billmyer 1990 on compliments, Boxer 1991 on indirect complaints, D’Amico-Reisner 1985 on disapprovals, Goldschmidt 1993 on favor-asking, Meyer 1996 on disagreements, Rabinowitz 1993 on offers, and Walter-Goldberg 1985 on jury summation. Social networks and identities: Adams 1998 on gesture in foreigner talk, Citron 1996 on study abroad students in Spain, Jakar 1995 on Hebrew language immersion camp, Kabongo-Mianda 1990 on two Zairean children acquiring English, Kubota 1999 on ESL learners’ construction and display of social identities in the language learning classroom, Morgan 1989 on three generations of black women in Chicago, and Newman 1993 on the Russian Jewish immigrant community in Philadelphia.

and (2) investigations of communicative contexts for second language, foreign language, and bilingual learning and teaching in language and content classrooms at elementary, secondary, and adult or higher education levels, including: Interaction and negotiation in second language learning and teaching: Berg 1998 on peer response in adult ESL writing classes, Dessner 1991 on teacher feedback in ESL, Doughty 1988 on effect of instruction in acquisition of relativization in ESL, Futaba 1994 on learner interaction in second language learning, Holliday 1995 on native speaker-nonnative speaker negotiations in second language learning of syntax, Linnell 1995 on negotiation as a context for second language learning, Tseng 1992 on peer participation in second language learning, and Wanmansor 1998 on the MOO (multi-object oriented synchronous computer communication) as an environment for ESL; Processes and models in second language learning and teaching: Boatman 1989 on perception and production in adult second language learning, Hwang 1989 on bidirectional transfer and markedness in Korean and English second language learning, Keenan 1993 on a connectionist model of second language acquisition, Kim 1992 on typological and transformational perspectives on ESL learning, Nimmrichter 1997 on the role of universal grammar in German second language acquisition, Washburn 1992 with a Vygotskian perspective on fossilization in second language learning, and Young 1989 on variation in interlanguage morphology in Chinese speakers learning ESL;

284 Nancy H. Hornberger

Foreign language learning and teaching: Chen 1996 on corrective feedback in foreign language learning (Chinese), Freire 1989 on teachers’ theoretical framework and classroom practice in foreign language teaching (Portuguese), and Kanagy 1991 on developmental sequences in foreign language learning (Japanese); Content learning and teaching: Berducci 1995 on science mentoring in the elementary school, Kaplan 1992 on the mutual influence of speaking and writing in classroom interaction, Kipers 1993 on the role of gender in teacher interaction, Sotillo 1991 on facilitating input comprehension in transitional college courses for ESL students, and Tanner 1991 on questioning as a teaching strategy among international teaching assistants.

The inclusive, sociocultural view of communicative competence and contexts has also been the impetus for the program’s research and practice initiatives on bilingualism and biliteracy in Philadelphia’s diverse urban schools and communities, as well as in multilingual settings all over the world. The program has had a steady record of involvement since the mid-1980s with the Puerto Rican community in North Philadelphia (e.g., Hornberger 1989, 1990, 1991, 1992; Hornberger and Micheau 1993; Micheau 1990; Rubio 1994; Freeman forthcoming a and b, Varghese 2000) and with Asian immigrant and Southeast Asian refugee communities in West and South Philadelphia (e.g., Weinstein-Shr 1986, 1993, 1994 on Hmong literacy; Chen 1992 on language maintenance and shift in the Chinese community; Hardman 1994; Hornberger and Hardman 1994; Hornberger 1996, and Skilton-Sylvester 1997 on language and literacy in the Cambodian community; Gordon in progress on language learning and identity in the Lao community). In the early 1980s while she was an educational linguistics student, Gail Weinstein-Shr founded Project LEIF (Learning English through Intergenerational Friendship), an ESL tutoring service with particular outreach to older refugees, and it has since been staffed and directed by numerous other educational linguistics students.14 These efforts in Philadelphia’s diverse urban sphere have flourished under and also furthered overall Graduate School of Education (GSE) priorities, as exemplified in the Center for Urban Ethnography and the Ethnography in Education Forum, both founded under Hymes’ leadership, and a more recent Spencer Training Grant (launched under current Dean Susan Fuhrman) with its focus on urban education research. Educational linguistics faculty have not only carried out research internationally, but have also consulted, lectured and taught in Bolivia, Brazil, Ecuador, El Salvador, Japan, Peru, and South Africa. Likewise, educational linguistics

Educational linguistics: A view from Penn’s Program 285

students have carried out research on the same wide range of topics noted above, and in many corners of the globe, for example: Botswana: Language planning and education policy (Nyati-Ramahobo 1991) Brazil: Peers as resource for language learning in foreign language context (Assis 1995) Britain: Mainstreaming as language policy and classroom practice (Creese 1997) Cyprus: Turkish language reform in a language planning framework (Dogancay 1993) Ecuador: Language revitalization in the Andes (King 1997) Israel: The social world of a preadolescent school class (Spiegel 2000) Japan: Gaijinization of Japanese language and culture in contact situations (Iino 1996); honorific use in everyday speech of Japanese women (Okushi 1997) Kazakhstan: Identity planning and language orientation planning (DeLorme 1999) Malaysia: Learning ESL: Reinforcing and suppressing factors in two communities (Razali 1992); Acquisition planning for English in tertiary education (Zakaria 1997) Pakistan: Literacy practices in a rural community (Farah 1992) Singapore: Production principles in non-native institutionalized varieties of English (Williams 1987) Yugoslavia: Self-management and classroom interaction (Rosenfeld 1986).

These efforts have provided impetus for and are congruent with recent GSE priorities in international education, with the inauguration in 1993 of GSE’s Office of International Programs and its Six Nation Education Research Project (under then Dean Marvin Lazerson and Acting Dean Nancy Hornberger); and the opening in 1994 of GSE’s UNESCO-sponsored International Literacy Institute (opened under Prof. Daniel Wagner and Active Dean Nancy Hornberger, currently directed by Drs. Wagner and Mohamed Maamouri). “The scope of the field of educational linguistics”, Spolsky wrote, “is defined by the interaction of language and formal instruction” (1975: 347), or, in a slightly more elaborated phrase, “the intersection of linguistics and related language sciences with formal and informal education” (1978: 2). Like Hymes and Penn’s Educational Linguistics programs, Spolsky’s concept of educational linguistics begins from the concept of communicative competence: Educational linguistics starts with the assessment of a child’s communicative competence on entering school and throughout his or her career, includes the analysis of societal goals for communicative competence, and embraces the

286 Nancy H. Hornberger

whole range of activities undertaken by an educational system to bring its pupils’ linguistic repertoires into closer accord with those expected by society. It thus is concerned with the processes used to bring about change, whether to suppress, enrich, alter the use of, or add, one or more styles, dialects, varieties, or languages (1978: viii).

He mapped out on various occasions some of the areas encompassed within this scope, including child language acquisition, first (mother tongue) and second (or additional or foreign) language teaching, the teaching of reading/writing/ literacy, bilingual education, the teaching of literature, and testing (1974a: 2023; 1974b: 2034; 1975: 347; 1978: 175; 1994: 88). All of these have remained remarkably consistent topics in the field, as evidenced for example in the 1994 Georgetown University Round Table (Alatis 1994), which included, in addition to most of these, explicit attention to second language acquisition, language minority education, English as a world language, and the study of speech acts and discourse; and in the 1997 Encyclopedia of Language and Education (Corson 1997), whose eight volumes are devoted to language policy and political issues, literacy, oral discourse, second language education, bilingual education, knowledge about language, language testing and assessment, and research methods. The scope of educational linguistics is, then, remarkably wide, but does not therefore sacrifice depth. Because of the functional and multiple conception of language in use which underlies it, there is constant attention to the possibilities of different meanings, different implications, different choices of language in particular contexts.15 Hymes himself warned against the pitfall of scope without depth, noting that simple models of rational actors and participants in discourse, while seeming to clarify experience, actually may obscure and mystify it … Rational choice, propositional clarity, clear turn taking, and the like are not models from which to predict the movement of participant-particles, but half of a dialectic between convention and choice (1986: 87–88).

Birds on a wire Such a dialectic is also at play in the interaction among academic disciplines. Since there is no one “conventional” choice for professional affiliation for educational linguists, Penn’s Educational Linguistics faculty and students participate in a wide array of professional associations. Faculty and students are

Educational linguistics: A view from Penn’s Program 287

members and regular presenters at, for example, the American Anthropological Association (AAA), the American Association for Applied Linguistics (AAAL), the American Educational Research Association (AERA), the Linguistic Society of America (LSA), the National Association for Bilingual Education (NABE), New Ways of Analyzing Variation in English (NWAVE), the Second Language Research Forum (SLRF), the Sociolinguistics Symposium, and Teachers of English to Speakers of Other Languages (TESOL), among others. Similarly, they subscribe to and publish in myriad professional journals such as Annual Review of Applied Linguistics, Anthropology and Education Quarterly, Applied Linguistics, International Journal of the Sociology of Language, Journal of Multilingual and Multicultural Development, Language and Education, Language in Society, Language Learning, Language Problems and Language Planning, Linguistics and Education, and TESOL Quarterly, to name only a few of the most frequently consulted. These multiple affiliations provide for constant interchange among the many disciplines which inform the field of educational linguistics, primary among them linguistics, sociolinguistics, psycholinguistics, applied linguistics, anthropology, and education; and they also provide multiple forums in which the voices and concerns of educational linguists are heard. Van Lier uses the very apt metaphor of birds on a wire to characterize the shifting and repositioning that goes on among academic disciplines when a new one joins their midst; he also observes that “if they refuse to budge, the newcomer will have to fly off again” (1994: 203–204). The foregoing sections of this paper suggest, I think, that educational linguistics has indeed found a place on the wire amidst its peer disciplines.16 In his proposal for the field of educational linguistics, Spolsky had suggested that it would constitute a subfield of applied linguistics, the latter inclusively defined as “the cluster of fields embracing all studies of language intended to be directly and immediately relevant to some social, educational, political, literary, or commercial goal” (1974c: 553). He also identified two reasons why he preferred the title educational linguistics to applied linguistics: one, that applied linguistics includes some topics that may be outside of educational linguistics, such as “translation, lexicography, language planning”; and two, that the assumption that linguistics can be directly applied to education is problematic (1978: 1). Shuy, too, wanted to differentiate educational linguistics from applied linguistics; indeed, he wanted to substitute the former term for the latter, because he felt that applied linguistics had become misunderstood to mean only ESL, the teaching of English as a Second Language (1981: 458). Finally, van Lier was concerned that the 1980s had seen the development of a serious rift in the

288 Nancy H. Hornberger

field of applied linguistics, between SLA researchers who distanced themselves from practice and teacher researchers who emphasized a strong pedagogical focus (1994: 202). From today’s perspective, none of these concerns seems any longer relevant. A glance at the topic areas for recent AAAL conferences reveals a list very similar to, if somewhat more elaborated than, the ones presented earlier for educational linguistics, to wit: assessment/evaluation, critical linguistics, discourse analysis, first language acquisition, immersion/bilingual education, interlingual/cross-cultural pragmatics, language and culture, language and the deaf, language and gender, language and socialization, language and sociocultural theory, language and technology, language for specific purposes, language policy/planning, literacy, mother tongue pedagogy, psycholinguistics, rhetoric and stylistics, second/foreign language pedagogy, second language acquisition, sociolinguistics, translation and interpretation. It appears that applied linguistics is no longer solely identified with ESL, nor is it split between theory and practice. Neither is it necessarily wider in scope than educational linguistics;17 even applications of linguistics for social, political, literary, or commercial ends (following Spolsky above) may ultimately relate to education in one way or another. The core differences between applied linguistics and educational linguistics, and they are not negligible ones, are the focus and starting point for the discipline. In educational linguistics, the starting point is always the practice of education and the focus is squarely on (the role of) language (in) learning and teaching. It is on those important differences that the argument for educational linguistics as a separate field rests, and it is in addressing those important challenges that the field of educational linguistics has its work cut out for many years to come.

Notes 1. More recently, the Monterey Institute of International Studies changed the name of the Department of Language Studies to Graduate School of Languages and Educational Linguistics (Leo van Lier, personal communication, 7 November 1998); the School offers advanced language courses (usually content-based in several disciplines) and masters’ degrees in TESOL and TFL (Teaching Foreign Language), but no doctoral degrees. At the University of New Mexico, Professor Leroy Ortiz, student of Bernard Spolsky, currently directs the Division of Language, Literacy and Sociocultural Studies which houses the Educational Linguistics program there. So far as I know, these, with the Penn program, constitute the only three Educational Linguistics programs to date, although a number of related programs

Educational linguistics: A view from Penn’s Program 289

in Language and Literacy, Language in Education, or some variation thereof, emerged in the 1980s (more on that below). 2. In the budget climate at Penn at that time, creating masters specializations alongside the doctoral specialization was seen as a wise strategic move, since the masters students could provide tuition dollars that would help support the doctoral specialization (Hymes, personal communication, 26 October 1998). All three specializations continue to operate to the present. While all form an integral part of the Educational Linguistics program, the focus here will be on the doctoral specialization only. 3. Teresa Pica, now Professor and Chair of the Language in Education Division, is the single person with the longest affiliation to the program, having begun there as student shortly after its establishment and continued on as professor until the present. I would like to acknowledge here the profound influence Pica has had on the development of the program, on the professional development of its students, and indeed on my own professional career. 4. I write from the perspective of one affiliated with the program since 1985 as professor and for several years as director. It is in many respects a personal, and undoubtedly biased, perspective, but it also affords the benefits of insider knowledge. 5. A note on terminology: The Graduate School of Education is organized into four divisions, which in turn house programs; programs may encompass one or more specializations. In common usage, the term program is often used to refer to specialization, and that usage is reflected in the Handbook excerpt quoted here. 6. International doctoral graduates have been from: Botswana, Brazil, England, Germany, Indonesia, Israel, Japan, Korea, Malaysia, Pakistan, Peru, South Africa, Sweden, Taiwan, Turkey, and Zaire. 7. Graduates hold tenure-track or tenured faculty positions at, for example, the following universities nationally and internationally: Georgetown, New York University, University of Florida, University of Illinois, University of Puerto Rico, University of Wisconsin, LaTrobe (Australia), University of Botswana, University of Rio de Janeiro (Brasil), Obirin (Japan), Suk Myung (Korea), Universiti Teknologi (Malaysia), Aga Khan (Pakistan), Donghwa (Taiwan), and Bogaziçi (Turkey). 8. Similarly, a number of Language and Literacy graduate programs/departments/divisions/ centers have emerged in schools of education at various U.S. and international universities during the 1980s, including University of California at Berkeley (Language and Literacy), University of Arizona (Language, Reading, and Culture), and an increasing number of programs in Language, Literacy, and Culture, e.g., at Stanford, University of Colorado — Denver, and University of Maryland — Baltimore County, among others, in the U.S. International examples include the Centre for Language and Literacy at the University of Technology in Sydney, Australia and the Centre for Language in Education at the University of Southampton, UK. Of course, there are myriad programs in applied linguistics or TESOL or bi(multi)lingualism-bi(multi)culturalism or literacy, but I am highlighting here specifically those programs that unite language education and literacy education concerns under one institutional umbrella. One set of programs that appear to have developed along very similar

290 Nancy H. Hornberger

lines to the Language in Education Division programs at Penn are the Literacy and Bilingualism Research Groups and the Centre for Language in Social Life at the Department of Linguistics and Modern English Language, University of Lancaster, UK. 9. It is worth noting that Hymes was a faculty member in Linguistics and Wolfson had earned her doctorate there. 10. For one recent example, see Kachru 1994: 19 on the application of educational linguistics for exploring the cross-cultural dimensions of world Englishes. 11. Wodak (1997: xii) makes a similar tripartite emphasis in describing language as central in the socialisation process and in schools, as “L1 which determines the identity and the intellectual and cognitive development of individuals; as mode for transfer of knowledge and for interaction between teacher and student; [and] as object of knowledge and critical reflection in both L1 and L2 education”. 12. As Associate Dean under Hymes, Professor Boe worked closely with Wolfson in strategizing on the fiscal and administrative dimensions involved in building Educational Linguistics; he also chaired the search committee that recommended the appointment of Wolfson as Assistant Professor of Education at GSE in 1978. 13. The Education, Culture, and Society division/program has been another site for linguistics at GSE, numbering among its faculty over the years such distinguished linguistic anthropologists as Shirley Brice Heath, Bambi Schieffelin, Katherine Woolard, Frederick Erickson and, currently, Stanton Wortham. The infusion of interests in language and ethnography extends beyond these two divisions as well, as seen for example in the Psychology in Education Division, in the work of former faculty such as Michele Fine and Brian Sutton-Smith, as well as current faculty such as Howard Stevenson, Margaret Spencer, and Daniel Wagner. Wagner directs the Literacy Research Center and its affiliated National Center on Adult Literacy and International Literacy Institute, in which Educational Linguistics and Language in Education faculty and students collaborate heavily. 14. Project LEIF is “a model program developed at Temple University Institute on Aging’s Center for Intergenerational Learning [through which] over 1,000 college-age volunteers have been trained to tutor English as a second language (ESL) to elder refugees and immigrants at community centers throughout the city; these include a Cambodian Buddhist temple, a Chinese community center, a Latino senior center, and a multicultural neighborhood center” (Weinstein-Shr 1994: 120). 15. This emphasis on meanings and choice is consistent with a plea issued by Christie (1994) for an educational linguistics which would operate with a model of language in terms of a resource for meaning (following Hallidayan systemic functional linguistics), rather than in terms of ‘rules’ (1994: 122); Penn Educational Linguistics emphases on meanings and choice are also consistent with, indeed as I argue here, informed by both Hymes’ notion of communicative competence and Spolsky’s concept of educational linguistics, Christie’s reservations notwithstanding (1994: 100, 106). 16. The growing visibility of educational linguistics as a field is illustrated by two tenuretrack faculty search advertisements posted in the 1998–1999 academic year, one by the University of California at Berkeley for an Assistant Professor in Educational Linguistics, and another by the University of New Mexico for an Assistant Professor in Native American

Educational linguistics: A view from Penn’s Program 291

Educational Sociolinguistics, whose responsibilities would include “mentoring of Native American and other students of educational linguistics”. Another index is the emergence of electronic listserves using the name educational linguistics, such as edling@education. leeds.ac.uk or the former [email protected]. 17. Van Lier sees educational linguistics as a sub-classification of applied linguistics, in turn a sub-classification of linguistics, while acknowledging that this hierarchical nomenclature may not satisfy all of his colleagues (1997: 95).

References Adams, T. 1998. Gesture in Foreigner Talk. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Alatis, J. E. (ed.). 1994. Educational Linguistics, Crosscultural Communication, and Global Interdependence. Washington, D. C.: Georgetown University Press. Assis, A. A. 1995. Peers as a Resource for Language Learning in the Foreign Language Context: Insights from an Interaction-based Study. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Bartels, N. 1999. Beyond Mediation: Applied Linguistics and Language Teacher Education. Unpublished manuscript. Benander, R. 1993. Communicative Competence in Second Language Acquisition: Noticing and Choosing to Perform Positive Evaluation Speech Acts. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Berducci, D. 1995. Internalization and Structure of Science Mentoring: An Analysis Following Vygotsky and Pike. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Berg, C. 1997. The Effects of Trained Peer Response on Writing Quality, Revision Strategies, and Peer Talk about ESL Texts. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Billmyer, K. 1990. The Effects of Formal Instruction on the Development of Sociolinguistic Competence: The Performance of Compliments. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Boatman, D. 1989. An Investigation of Adult Second Language Perception and Production: Evidence for and against a Predictive Account. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Bourdieu, P. 1990. The Logic of Practice. Stanford: Stanford University Press. Boxer, D. 1991. A Descriptive Analysis of Indirect Complaint Sequences among Speakers of American English. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Brumfit, C. 1997. “The teacher as educational linguist”. In Knowledge about Language, L. van Lier and D. Corson (eds), 163–172. In Encyclopedia of Language and Education 6, Dordrecht: Kluwer. Chen, H. H. 1996. A Study of the Effect of Corrective Feedback on Foreign Language Learning: American Students Learning Chinese Classifiers. Unpublished Ph.D., University of Pennsylvania, Philadelphia.

292 Nancy H. Hornberger

Chen, S. 1992. Language Maintenance and Shift in the Chinese Community of Greater Philadelphia. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Chomsky, N. 1965. Aspects of the Theory of Syntax. Cambridge: MIT Press. Christie, F. 1994. “Developing an educational linguistics for English language teaching: A systemic functional linguistic perspective”. Functions of Language 1 (1): 95–127. Citron, J. 1996. The Cross-cultural Re-entry Experiences of Short-term Study Abroad Students from the U. S. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Corson, D. (ed.). 1997. Encyclopedia of Language and Education. Dordrecht: Kluwer Academic Publishers. Creese, A. 1997. Mainstreaming as Language Policy and Classroom Practice: An Interpretive Study of Partner Teachers’ Roles, Relationships, and Talk in Multilingual British Secondary Schools. Unpublished Ph.D., University of Pennsylvania, Philadelphia. D’Amico-Reisner, L. 1985. An Ethnolinguistic Study of Disapproval Exchanges. Unpublished Ph.D., University of Pennsylvania, Philadelphia. DeLorme, S. 1999. Government and School construction of Identity and Language Planning Mother Tongue, Mother’s Touch: Kazakhstan. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Dessner, L. E. 1991. English-as-a-second-language College Writers’ Revision Responses to Teacher Written Comments. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Dogançay, S. 1993. Turkish Language Reform in a Language Planning Framework: Its Impact on Language Use of Turkish Cypriot High School Students. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Dogançay-Aktuna, S. 1995. “An evaluation of the Turkish language reform after 60 years”. Language Problems and Language Planning 19 (3): 221–249. Doughty, C. 1988. The Effect of Instruction on the Acquisition of Relativization in English as a Second Language. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Educational Linguistics Handbook 1997–98. Language in Education Division. Philadelphia: University of Pennsylvania Graduate School of Education. Farah, I. 1992. Literacy Practices in a Rural Community in Pakistan. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Freeman, D. 1994. “Educational linguistics and the knowledge base of language teaching”. In Georgetown University Round Table on Languages and Linguistics 1994, J. E. Alatis (ed.), 180–196. Washington D. C.: Georgetown University Press. Freeman, R. Forthcoming a. “Contextual challenges to Dual-language Education: A Case Study of a Developing Middle-school Program”. Anthropology and education Quarterly. Freeman, R. Forthcoming b. Legitimizing Spanish and Spanish Speakers in North Philadelphia: A Microethnographic Analysis in a Dual-language Classroom. Submitted for publication. Freire, A. F. 1989. Communicative Competence as the Goal of Foreign Language Teaching: Teachers’ Theoretical Framework and Classroom Practice. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Futaba, T. 1994. Second Language Acquisition through Negotiation: A Case of Non-native Speakers who Share the Same First Language. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Goldschmidt, M. 1993. For the Favor of Asking: A Sociolinguistic Analysis. Unpublished Ph.D., University of Pennsylvania, Philadelphia.

Educational linguistics: A view from Penn’s Program 293

Gordon, D. In progress. “Why America Destroys my Country?”: Investment and Resistance to Learning English among Laotian Refugees in Philadelphia. Unpublished Ph.D. dissertation proposal, University of Pennsylvania. Hardman, J. 1994. Language and Literacy Development in a Cambodian Community in Philadelphia”. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Holliday, L. 1995. Native Speaker Syntactic Modifications in Native Speaker — Nonnative Speaker Negotiation as Input Data for Second Language Acquisition of Syntax. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Hornberger, N. H. 1989. “Continua of biliteracy”. Review of Educational Research 59 (3): 271–296. Hornberger, N. H. 1990. “Creating successful learning contexts for bilingual literacy”. Teachers College Record 92 (2): 212–229. Hornberger, N. H. 1991. “Extending enrichment bilingual education: Revisiting typologies and redirecting policy”. In Bilingual Education: Focusschrift in Honor of Joshua A. Fishman on the Occasion of his 65th Birthday, O. García (ed.), 215–234. Philadelphia: John Benjamins. Hornberger, N. H. 1992. “Biliteracy contexts, continua, and contrasts: Policy and curriculum for Cambodian and Puerto Rican students in Philadelphia”. Education and Urban Society 24 (2): 196–211. Hornberger, N. H. 1996. “Mother tongue literacy in the Cambodian community of Philadelphia”. International Journal of the Sociology of Language 119: 69–86. Hornberger, N. H. and Hardman, J. 1994. “Literacy as cultural practice and cognitive skill: Biliteracy in a Cambodian adult ESL class and a Puerto Rican GED program”. In Adult Biliteracy in the United States, D. Spener (ed.), 147–169. Washington D. C.: Center for Applied Linguistics. Hornberger, N. H. and Micheau, C. 1993. “‘Getting far enough to like it’: Biliteracy in the middle school”. Peabody Journal of Education 69 (1) 30–53. Hwang, S. 1989. Bidirectional Transfer and Markedness: Acquisition of Korean and English as L2. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Hymes, D. H. 1972. “On communicative competence”. In Sociolinguistics: Selected Readings, J. B. Pride and J. Holmes (eds), 269–293. Harmondsworth: Penguin Books. Hymes, D. H. 1980. Language in Education: Ethnolinguistic Essays. Washington D. C.: Center for Applied Linguistics. Hymes, D. H. 1986. “Discourse: Scope without depth”. International Journal of the Sociology of Language 57: 49–89. Iino, M. 1996. “Excellent foreigner!”: Gaijinization of Japanese Language and Culture in Contact Situations: An Ethnographic Study of Dinner Table Conversations between Japanese Host Families and American Students. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Jakar, V. 1995. A Society Contained, a Culture Maintained: An Ethnography of Second Language Acquisition in Informal Education. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Kabongo-Mianda, K. 1991. Child Interlanguage Discourse: The Role of Peer Interaction. Unpublished Ph.D., University of Pennsylvania, Philadelphia.

294 Nancy H. Hornberger

Kachru, B. B. 1994. “The speaking tree: A medium of plural canons”. In Georgetown University Round Table on Languages and Linguistics 1994, J. E. Alatis (ed.), 6–22. Washington D. C.: Georgetown University Press. Kanagy, R. 1991. Developmental Sequences in the Acquisition of Japanese as a Foreign Language: The Case of Negation. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Kaplan, M. 1992. Writing as Listening: The Mutual Influence of Speaking and Writing in Classroom Interaction. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Keenan, T. 1993. A Connectionist Model of Second Language Acquisition. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Kim, K. 1991. Relative Clause Realization, Categorization, and Acquisition in English as a Second Language: The Case of Genitive Relative Clause. Unpublished Ph.D., University of Pennsylvania, Philadelphia. King, K. A. 1997. Language Revitalization in the Andes: Quichua Instruction, Use, and Identity in Saraguro, Ecuador. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Kipers, P. 1993. The Role of Gender in Conversation: Some Insights from a Community of Teachers. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Kubota, M. 1999. Choice of Performed Identities: Negotiating, Constructing and Displaying Identities in an Adult ESL Classroom for Teaching and Learning Communicative Competence. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Kuhn, T. S. 1962. The Structure of Scientific Revolutions. Chicago, IL: University of Chicago Press. LED News (1993–1998). The Language in Education Division Newsletter. Philadelphia: University of Pennsylvania Graduate School of Education. Linnell, J. 1995. Negotiation as a Context for Learning Syntax in a Second Language. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Meyer, T. W. 1996. Language and Power in Disagreements: Analyzing the Discourse of Male, Female, and Male/Female Couples. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Micheau, C. 1990. Ethnic Identity and Ethnic Maintenance in the Puerto Rican Community of Philadelphia. Unpublished Ph.D., University of Pennsylvania. Morgan, M. 1989. From Down South to up South: The Language Behavior of Three Generations of Black Women Residing in Chicago. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Myers, M. J. 1994. “Various perspectives on educational linguistics gleaned from a collaborative project on the use of dictionaries”. Language Awareness 3 (3–4): 193–200. Newman, J. 1993. Ethnography, Classrooms, and Social Networks in the Russian Jewish Immigrant Community of Northeast Philadelphia. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Nimmrichter, S. 1997. The Role of Universal Grammar in Second Language Acquisition: Explaining Variability of Verb Placement in L2 German. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Nyati-Ramahobo, L. 1991. Language Planning and Education Policy in Botswana. Unpublished Ph.D., University of Pennsylvania, Philadelphia.

Educational linguistics: A view from Penn’s Program 295

Okushi, Y. 1997. Patterns of Honorific Use in the Everyday Speech of Four Japanese Women. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Pakir, A. 1994. “Educational linguistics: Looking to the East”. In Georgetown University Round Table on Languages and Linguistics 1994, J. E. Alatis (ed.), 370–383. Washington D. C.: Georgetown University Press. Pica, T. 1982. Second Language Acquisition in Different Language Contexts. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Pica, T. 1994. “The language educator at work in the learner-centered classroom: Communicate, decision-make, and remember to apply the (educational) linguistics”. In Georgetown University Round Table on Languages and Linguistics 1994, J. E. Alatis (ed.), 264–288. Washington D. C.: Georgetown University Press. Pica, T. 1997. “Second language teaching and research relationships: a North American view”. Language Teaching Research 1 (1): 48–72. Pousada, A. 1984. Community Participation in Bilingual Education: The Puerto Rican Community of East Harlem. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Rabinowitz, J. 1993. A Descriptive Study of the Offer as a Speech Behavior in American English. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Razali, N. 1992. Learning ESL in Malaysia: A Study of Reinforcing and Suppressing Factors in Two Communities. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Rosenfeld, M. 1986. Yugoslav Self-management and Classroom Interaction. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Rubio, O. 1994. “Una buena educacion”: a study of parental values, beliefs, and aspirations in a dual-language school. Unpublished Ph.D., University of Pennsylvania. Shuy, R. W. 1981. Educational Linguistics. Neueren Sprachen 80 (5): 455–468. Skilton-Sylvester, E. 1997. Inside, Outside, and In-between: Identities, Literacies, and Educational Policies in the Lives of Cambodian Women and Girls in Philadelphia. Unpublished Ph.D., University of Pennsylvania. Smitherman, G. 1979. “Toward educational linguistics for the first world”. College English 41 (2): 202–211. Sotillo, S. M. 1991. Input, Interaction, Content and Language Learning in the Bridge Classroom. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Spiegel, J. 2000. “It’s My Life”: Independence, Cohesion, and Tensions in the Social World of an Israeli Preadolescent School Class. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Spolsky, B. 1974. “Linguistics and education: An overview”. In Current Trends in Linguistics, T. A. Sebeok (ed.), 12: 2021–2026. The Hague: Mouton. Spolsky, B. 1974. “Linguistics and the language barrier to education”. In Current Trends in Linguistics, T. A. Sebeok (ed.), 12: 2027–2038. The Hague: Mouton. Spolsky, B. 1974. “The Navajo reading study: An illustration of the scope and nature of educational linguistics”. Applied Linguistics: Problems and Solutions, J. Qvistgaard (ed.), 3: 553–565. Heidelberg: Julius Groos Verlag. Spolsky, B. 1975. “Linguistics in practice: The Navajo reading study”. Theory into Practice 14 (5): 347–352. Spolsky, B. 1978. Educational Linguistics: An Introduction. Rowley, MA: Newbury House.

296 Nancy H. Hornberger

Spolsky, B. 1994. “The beginning of language testing as a profession”. In Georgetown University Round Table on Languages and Linguistics 1994, J. E. Alatis (ed.), 88–101. Washington D. C.: Georgetown University Press. Street, B. 1984. Literacy in Theory and Practice. New York: Cambridge University Press. Street, B. V. (ed.). 1993. Cross-Cultural Approaches to Literacy. Cambridge: Cambridge University Press. Stubbs, M. 1986. Educational Linguistics. New York: Basil Blackwell. Suardiaz, D. E. and Domínguez-Colavita, F. 1987. “Aportes a la lingüística educacional: La lengua materna en la educación elemental”. Revista Argentina de Lingüística 3 (2): 161–169. Tanner, M. W. 1991. NNSTA-student Interaction: An Analysis of TAs’ Questions and Students’ Response in a Laboratory Setting. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Tseng, S. D. 1992. Adding a Third Language: An In-depth Analysis of the Teacher Questioning Processes in a Chinese Classroom. Unpublished Ph.D., University of Pennsylvania, Philadelphia. van Lier, L. 1991. “Language awareness: The common ground between linguist and language teacher”. In Georgetown University Round Table on Languages and Linguistics 1991, J. E. Alatis (ed.), 528–546. Washington D. C.: Georgetown University Press. van Lier, L. 1994. “Educational linguistics: Field and project”. In Georgetown University Round Table on Languages and Linguistics 1994, J. E. Alatis (ed.), 197–209. Washington D. C.: Georgetown University Press. van Lier, L. 1997. “Apply within, apply without?” International Journal of Applied Linguistics 7 (1): 95–105. Varghese, M. 2000. Bilingual Teachers in the Making: Advocates, Classroom Teachers, and Transients. Unpublished Ph.D. dissertation proposal, University of Pennsylvania. Walter-Goldberg, B. 1985. Jury Summation as Speech Genre: An Ethnographic Study of What it Means to Those Who Use It. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Wanmansor, F. 1998. Interaction on the Network: A Case of PennMOO among ESL Students. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Washburn, G. 1991. Fossilization in Second Language Acquisition: A Vygotskian Perspective. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Weinstein-Shr, G. 1986. From Mountaintops to City Streets: An Ethnographic Investigation of Literacy and Social Process among the Hmong of Philadelphia. Unpublished Ph.D., University of Pennsylvania. Weinstein-Shr, G. 1993. “Literacy and social process: A community in transition”. In Crosscultural Approaches to Literacy, B. Street, 272–293. Cambridge: Cambridge University Press. Weinstein-Shr, G. 1994. “Literacy and second language learners: A family agenda”. In Adult Biliteracy in the United States, D. Spener (ed.), 111–122. McHenry, IL: Delta Systems. Williams, J. 1987. Production Principles in Non-native Institutionalized Varieties of English. Unpublished Ph.D., University of Pennsylvania, Philadelphia.



Educational linguistics: A view from Penn’s Program 297

Wodak, Ruth 1997. “Introduction”. In Language Policy and Political Issues in Education, R. Wodak and D. Corson (eds), xi–xiv. In Encyclopedia of Language and Education 8, Dordrecht: Kluwer. Working Papers in Educational Linguistics (1984 to present). Vol. 1 (1)-13 (2). Philadelphia: University of Pennsylvania Graduate School of Education. Young, R. 1989. Variation in Interlanguage Morphology: (s) Plural Marking in the Speech of Chinese Learners of English. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Zakaria, M. H. 1997. Acquisition Planning for English: A Case in Malaysian Tertiary Education. Unpublished Ph.D., University of Pennsylvania, Philadelphia. Zentella, A. C. 1981. “Hablamos los dos. We Speak Both”: Growing up Bilingual in El Barrio. Unpublished Ph.D., University of Pennsylvania, Philadelphia.



Index

A AAA 287 AAAL 110, 287–288 Aboriginal language 220, 222 Accent 75, 91, 106 Adult education 178, 183, 204 Adult foreign language education 172, 180 Adult language education 179 Adult language learning opportunities 179 Adult use of foreign languages 173, 176, 181, 191 AERA 121, 287 Affect 31, 38, 49–51, 55, 68–69, 73–74, 84, 126 Africa 16, 167–168, 171, 188, 190, 193, 284, 289 ALAA 263–265 Alphabetic writing systems 30 ALS 214 American 51, 78–79, 83, 87–89, 93, 105, 107, 111, 143, 150–151, 154, 160, 169, 171, 188, 192, 194–196, 208, 218, 235–238, 246, 259, 264–265, 281, 287, 290–291, 293, 295 Applied Linguistics vi, 2–3, 17–18, 60, 87–88, 111, 121, 259, 261–269, 272–273, 277, 287–289, 291, 293, 295–296 Arabic 3, 28, 134, 137–138, 140, 155–156, 159–169, 190 Aramaic 133–136, 138–141, 143 Articulation 184, 197–200, 202–204, 207, 239, 252, 255

ASLPR 91 Assessment 2, 59, 63–74, 76, 78–79, 82–90, 108, 111, 118, 121, 129–130, 155, 177, 180, 193, 213, 218–219, 240–241, 243–244, 285–286, 288 ASTP 8 Attitudes 46, 123, 164, 167, 170, 241, 251, 254, 258–259 AULLA 263–264 Australia vi, 1, 3, 74, 176, 179, 185, 192–194, 211–215, 221, 224–226, 229, 233–234, 261–267, 269, 289 Australian National Plan 171 Authenticity 13, 51, 60, 64–65, 70–74, 83, 86–87, 90, 148, 196 Authority 8, 10, 12, 14, 58, 113–114, 118, 218 Autochthonous 187 B Basic user 205 Behavior 31, 39, 56, 60, 108, 112–113, 118, 120–121, 124, 151, 163, 294–295 Bilingual v, 1–2, 7–9, 12–13, 16–17, 20, 22–25, 27–33, 35, 37, 41, 44, 56, 59–60, 93, 134, 138, 140–141, 178, 186–187, 219, 238, 254, 266, 272, 278–280, 283, 286–288, 293, 295–297 Bilingual compounding 9, 12 Bilingual pedagogy 16 Bilingual processing v, 35, 37, 56, 59 Bilingualism 7–8, 11, 20, 24–26, 30–32, 136, 138–141, 186, 267, 284, 290

300 Index

Bilingualization 8, 11, 13, 16–17 British 3, 14–15, 17–18, 31, 87, 93, 95, 101, 107, 159, 165, 175, 177, 184, 188, 261–262, 265, 267–269, 292 British Council 14, 17–18, 87, 261–262 Business and trade 176 Business skills 177 Business use 177, 179 C Cantonese 218 Catalan 186 Childhood acquisition 93–94 Chinese 22–23, 25–27, 57, 99, 193, 218, 267, 283–284, 290–292, 296–297 Code of Ethics 124–125, 128 Code of Practice 125, 130 Code-switching 143 Cognitive development 30–32, 290 Cognitive processes 41, 66, 86 Common European Framework of Reference 204, 207–208 Communication 13–15, 47, 75, 92, 119, 124, 147, 156, 160, 173, 179, 181–182, 185, 188–190, 194, 201, 204, 213–217, 221, 223–224, 226, 228–229, 238, 248–249, 271, 274, 277–278, 281–283, 288–289, 291 Communicative competence 2, 13, 86, 104, 279, 281, 283–285, 290–294 Competence 2, 8, 13, 63, 67, 75–77, 86, 88, 90, 95–96, 104, 111, 162, 174, 177, 199, 212, 214, 236, 256, 279, 281, 283–285, 290–294 Concepts of print 21, 24, 30–31 Construct validity v, 63, 65–66, 72–74, 78–79, 85–87 Content learning and teaching 284 Contrastive analysis 9, 16 Control 36–37, 46–48, 55, 88–89, 94, 97, 105–106, 114, 155, 158, 165, 168–169, 178, 181, 202, 208, 262, 276

Corpus planning 194 Corpus policy 171 Council of Europe 173, 175, 178, 184, 192, 196, 199, 204, 208 Cross-cultural competence 256 Cross-language communication 181 Cultural diversity 215–216, 272 Curriculum 1–3, 9, 49, 51–53, 55, 57–58, 87–88, 155, 165–166, 168, 191–192, 200–201, 218, 226, 235–243, 249–253, 255–257, 273, 275–276, 280, 293 D Dialect 111, 142–143, 159, 167, 187 Different kinds of writing systems 23 Diffusion 143, 168–169, 181 Diglossia vi, 133–134, 136–143, 161–163, 168–169 Direct method 8 Discourse 12–13, 28, 31, 36–37, 45, 67, 73, 84, 88, 94, 117, 119, 213, 218, 220, 232, 272, 275–277, 283, 286, 288, 293–294 Diversification 107, 168, 185, 192–193 Diversity 137, 186, 215–216, 226, 272, 275 Domain 20, 37, 65–67, 70–74, 82–86, 158–159, 171–172, 176, 183, 186 Domestic language 3, 14–15, 171–172, 185, 188, 190 Dutch 99, 171, 173–174, 176, 179, 185, 189, 194, 196, 198, 201, 203, 205–208, 219 Dutch National Foreign Language Plan 201 E Eastern Europe 135–136, 143, 146, 148, 151, 190, 193 Education vi, 1–4, 17, 32, 36, 56, 60, 77, 88, 89, 104, 111–113, 116, 120, 122, 128–130, 134–135, 140–141,

Index 301

155–156, 159–170, 172, 174, 176–188, 190–191, 194–208, 211–213, 215–217, 219–223, 225, 231–232, 236–237, 239–241, 245–246, 250–251, 253–256, 258–259, 262, 266, 268, 271–278, 280–295, 297 Educational linguistics v-vi, 1–4, 20, 31, 59, 209, 271–282, 284–292, 294–297 Educational system 173, 175, 180–181, 186, 188, 193, 198–200, 202, 280, 286 English v, 1, 3, 7, 11–18, 22–23, 25, 28–29, 37, 40, 42–51, 57, 60, 74, 78–79, 82–83, 87–89, 93, 95, 98, 100–102, 104–108, 110–112, 139, 142, 147, 155–156, 159–166, 168–169, 174–175, 180, 185–192, 194–196, 198, 204–207, 211–215, 217–220, 222, 224–225, 227–231, 236, 238, 255, 261–262, 266–269, 271, 273, 275, 277–278, 281, 283–287, 290–297 Ethics 2, 90, 114, 120–121, 124–125, 127–128, 130 European Community 182, 192 European Union 174 Evaluation 9, 72, 88, 111, 121–122, 130, 150, 161–162, 167, 169, 191, 208, 218, 237, 243–244, 248, 256, 282, 288, 291–292 Exposure 8, 30, 51, 55, 94, 164, 166, 180–181, 219, 246 F Finnish 195 FLES vi, 199, 235–237, 239–240, 242, 245–246, 251–252, 254–256, 258–259 FLIP 38–49, 51–59, 136 FLT 200, 203–204 Fluency 29, 44, 50, 55–56, 59, 181, 199 Foreign accent 91

Foreign language vi, 1, 3, 7–8, 11–13, 15, 17, 31, 36, 38–39, 42, 57, 59–60, 87–89, 107, 130, 160, 171–205, 208, 212, 217–218, 236–239, 241–242, 245–250, 252–255, 257–259, 262, 265, 273, 283–285, 288, 291–292, 294 Foreign language competencies 179 Foreign language education establishment 172 Foreign language enablers 181, 184 Foreign language instruction 172, 176, 179–181, 184, 191, 193–194, 197, 204, 208, 236, 247, 253–254, 258–259 Foreign language learning 3, 59, 173, 183, 252, 254, 259, 284, 291 Foreign language needs 182, 194, 196 Foreign language planning 181, 208 Foreign language policy vi, 171–172, 177–180, 182, 184–185, 193–196, 203 Foreign language programs 236, 242, 245 Foreign language teachers 198, 201 Foreign language teaching system 197, 202 Foreign Service Institute 91 France 142, 174–175, 186, 189, 191 French 3, 22–23, 29, 32, 35, 38–41, 44–53, 55, 58–59, 95–96, 137–138, 140, 142, 147, 150, 155, 171, 174–175, 185–186, 188, 190–192, 198, 204–207, 218, 225, 238–239, 261, 263–265, 267 FSI 91 Functional 43, 67, 90, 138–139, 143, 146–147, 152–153, 167, 199, 236, 242, 254, 281, 286, 290, 292 G Galician 186 German 35, 38–50, 52, 54, 58, 99, 103, 138, 145–147, 149, 152, 174–175,

302 Index

185, 190–192, 201, 205, 218, 225, 238–239, 265, 283, 294 Globalization 175, 212, 225 Government 115, 129, 147, 157–158, 178, 180, 187, 193, 195, 200, 211–212, 215, 219, 225–226, 232–234, 266, 269, 272, 292 Grammar 9, 16, 42, 94, 96–97, 107, 162, 263, 268, 281–283, 294 Grammaticality judgements 98, 104, 108 Greek 99, 133–134, 136, 138–141, 143, 186, 192, 212, 225, 230 H Hebrew iii, vi, 1–3, 22–23, 99, 133–136, 138–141, 143, 145, 150, 155–156, 159–161, 163–164, 166–170, 201, 283 Hegemony 16–17, 190 Higher education 113, 160, 163–164, 203, 225, 273, 283 Hindi 187–188, 230 Historical language 134, 190 I ILR 91 ILTA 124–125, 128 Immersion v, 1, 32, 35–46, 48–51, 53–60, 214, 218, 222, 236, 245, 279, 283, 288 Immigrant 1–2, 150, 186, 266–267, 279, 283–284, 294 Implementation 155, 172, 195–196, 214–215, 227, 229–230, 233, 235–239, 241, 243, 245, 247, 249, 251, 257, 261, 271, 273, 278–279, 281 Independent user 205 India 16, 171, 187–188, 190 Indonesian 107–108, 187, 219, 225 Information technology 223 Instruments 118 Interaction 38, 55, 57, 69, 72, 81, 205,

272, 275–276, 279, 283–286, 290–291, 293–296 Interlanguage 8–11, 16, 37, 60, 88, 283, 293, 297 Interpretation 56, 70, 72, 79–81, 87–88, 96, 106, 182, 221, 231, 277, 288 Interview 40–41, 45, 51, 57, 75–76, 88, 237, 243–246, 252–253, 255, 257–258 Israel 1, 3–4, 136, 140, 149, 152, 155–160, 162–170, 173, 285, 289 Italian 25, 107, 138, 174, 201, 212, 225, 266 J Japan 189, 191, 284–285, 289 Japanese 28, 99, 193, 200, 219, 225, 237–239, 258–259, 273, 284–285, 293–295 Jewish 4, 133–134, 136–137, 139–140, 143, 145–162, 164–167, 169–170, 283, 294 Jewish identity 147 Jewish languages 4, 134, 143, 145, 153 Job 75–76, 86, 109–110, 117, 177, 183, 201 K Khmer 216, 219, 230 Kindergarten 235, 239–244, 247, 250, 252, 254–255, 257 Knowledge 2, 10–11, 14, 20, 27–28, 32, 60, 64, 67–69, 72, 74–76, 78–81, 83–85, 87, 92, 96–97, 101, 113, 117–118, 127–129, 136, 164, 174, 183, 191, 201, 203, 205, 227, 239, 247, 252, 256, 277, 281–282, 286, 289–292 Korean 99, 152, 216, 219, 225, 231, 283, 293 L Language 12, 20, 27–28, 36, 41–42, 93,

Index 303

95, 140, 146, 152, 160, 164, 167, 200, 217, 220, 222, 232, 235–236, 239, 265, 277, 280, 282, 287–288, 290 Language ability v, 63–70, 72, 74–76, 78–81, 83–85, 108, 112, 128, 174, 281 Language acquisition 2, 17, 19, 30, 32, 37, 60, 87–88, 95, 97–98, 111–112, 169, 183, 220, 258, 272, 275, 282–283, 286, 288, 291–296 Language attitudes 170 Language change 111, 137, 142 Language choice vi, 3, 37, 171–172, 184–188, 190–193, 272 Language competence 174, 177, 236 Language contact 7, 10, 169, 196 language Immersion 1, 35, 37–39, 57, 59, 279, 283 Language instruction 135, 155–156, 172, 176, 179–185, 188, 191–194, 197–198, 204, 208, 236, 247, 253–254, 258–259 Language knowledge 67, 74, 92, 183 Language learning v, 2–5, 7, 9, 11, 17–18, 36, 38, 57, 59–60, 66, 71, 87–88, 112, 129, 173–174, 178–181, 183–184, 194, 208, 237, 239, 245, 247, 249, 252–255, 259, 264, 272–275, 279, 282–285, 287, 291, 295 Language maintenance 136, 187, 284, 292 Language modality 35, 38, 46, 55, 59 Language planners 146, 189 Language planning vi, 2–3, 111, 131, 146, 169, 181, 194–196, 202, 208, 214, 234, 254, 265, 272, 275, 282–283, 285, 287, 292, 294 Language policy i, iii, vi, 1, 3, 168, 171–172, 177–180, 182–185, 187–190, 192–197, 199, 203, 208, 211–214, 220–224, 226–227, 233–234, 266, 269, 275, 279, 285–286, 288, 292, 297

Language proficiency 27, 37, 39–40, 69–70, 87–88, 91, 107, 129, 163, 200, 204, 208, 213, 237–239, 254–256 Language shift 133 Language spread 143 Language teaching v-vi, 2, 5, 8, 10–12, 14–18, 36, 87–89, 93, 108, 173, 178–181, 192, 194–195, 197–199, 202, 220, 228, 234, 261–262, 264–269, 274–275, 277, 279, 281–282, 284, 286, 292, 295 Language tester 115, 126 Language testing v, 2, 59–61, 63–65, 67–68, 70–71, 73, 78, 81, 84–90, 111, 114, 121, 123–130, 274, 286, 296 Language training 178, 181, 266 Language use 2–3, 35–37, 40–41, 51, 55–56, 64–68, 70–72, 74, 82–83, 86, 91–92, 134–136, 142, 173–174, 176–178, 182–183, 185, 193, 195, 206, 216, 279, 281–282, 292 Language use tasks 66, 70 Latin 39, 57, 138, 140, 143, 192, 201, 230 LCCC 279 LCTL 57 Learning 2, 10, 237, 239, 274, 284 Learning context 52 Learning methods 36 Learning styles 49 LiLTE 278 LINGUA 133, 141, 173–174, 176–178, 187–189, 196 Lingua franca 133, 141, 173–174, 187–189 Listening 35, 38, 40–41, 45, 47–49, 51, 55, 78, 162, 201, 243, 294 Listening comprehension 78, 243 Listening skills 201 Literacy v, 2, 5, 19–21, 24, 27, 30, 32, 143, 146, 212, 215, 233, 238, 256, 275, 278–279, 284–286, 288–290, 292–293, 296 Literature 16, 30, 36, 39, 67, 70, 83, 86,

304 Index

127, 136, 141, 145, 147, 164, 166, 171–172, 184, 189, 205–206, 220, 228, 255, 272–273, 275, 286 LOTE 220, 231, 233, 266 LSA 287 LTE 278 Luxembourg 174–175 M Machine translation 182 Maintenance 136, 139, 187, 193, 212, 272, 284, 292, 294 Mandarin 218–219, 225 Manipulating 114 Market competition-based system 180 Mental dialog 37–38, 40–42, 46–48, 54–56 Metacognitive strategies 67, 74, 86 Micro language policy vi, 211, 233 Ministry of Education 155, 161–162, 165, 167, 169, 201, 208 Minority language 1, 160 Mobility 181, 188, 225 Models in second language learning and teaching 283 Modernization 146–147, 151–153, 158, 164, 168 Monolingual pedagogy 9, 12–13, 17 Monolingual teaching v, 7–10, 13, 15–16 Morality 114, 116, 120–121 Mosaic societies 173 Mother-tongue 146 Motivation 1, 40, 83, 166, 183, 203, 255 Multicultural education 266 Multiculturalism 212, 266 Multilingualism vi, 2–4, 17, 131, 133, 135, 137, 142–143, 280 Multinational data collection 174 N NABE 287 National Foreign Language Center 89, 130, 195, 208

National foreign language policy 178–179, 182 National language policy 199, 211–212, 233 National need 175–176 Nationalism 140, 148, 150, 155, 168, 196 Native v, 2, 12–15, 28–29, 35–37, 39, 44, 47, 53, 82–83, 91–97, 99–101, 104–108, 111–112, 129, 140, 165, 180–181, 189, 200–201, 205, 236, 256, 283, 285, 290–293, 296 Native language 36, 44, 47, 189, 236, 256 Native speaker v, 2, 12–13, 53, 91–97, 101, 105–108, 111–112, 129, 200, 205, 283, 293 Netherlands 3, 173–174, 176, 189–191, 194, 197–198, 200–202, 204, 208 NNS 99–105, 108 Non-native 82–83, 93–96, 99–101, 104–106, 108, 111, 129, 189, 200, 205, 283, 285, 292, 296 Non-native speakers 82–83, 94–96, 99–100, 104–106, 108, 111, 189, 200, 292 NS 100–106, 108, 207 NWAVE 287 O Official language 156, 159–160, 167, 174, 186–187, 224 Official language status 186 P Palestine vi, 2, 133–137, 139–141, 143, 150, 158 Penn’s Program vi, 271, 283 Performance v, 23, 26, 28, 31, 33, 40–41, 57, 63–76, 78–79, 81–82, 84–90, 94, 96, 102, 107–108, 160, 217, 227, 291 Performance assessment 63–66, 69–72, 74, 79, 84–86, 89 Performance test 63 Phonological awareness 24–27, 30–32

Index 305

Pidgin 187 Polish 3 Polyglot 188 Portuguese 284 Power 89, 114, 118, 129–130, 134, 140, 152, 158–159, 211, 224, 226, 233, 276, 294 Pragmatic competence 104 Pragmatics 288 Pre-university education 197–198 Primary education 199 Proficiency 19–20, 27–28, 37, 39–40, 69–70, 78, 87–88, 90–91, 93, 107–108, 111, 127–129, 140–141, 162–163, 173, 185, 197–201, 203–208, 213, 220, 236–239, 254–256 Proficiency scales 204 Promotion 14, 16, 129, 178 Pronunciation 91, 204, 244 Q Questionnaire 35, 39–40, 100, 104, 109, 174, 204–206 R Reading 2, 19–21, 23–25, 27–33, 35–36, 38, 40–41, 43–44, 47, 49–51, 60, 69, 78, 87–88, 128, 136, 161–162, 201, 205, 207–208, 247, 265, 271, 275, 277, 280, 286, 289, 295 Reading skills 201, 207 Reliability 87, 115, 120 Responsibility v, 108, 113, 115–120, 124–126, 128, 130, 155, 215, 223, 228, 248, 250–251, 256 Roman 23 Russian 1–2, 145–149, 153, 190, 193, 230, 265, 283, 294 S Sami 187 Sanskrit 28, 190

Science 31, 38, 57, 88, 114, 127, 142, 160, 179–180, 189, 194–195, 208, 227, 236, 264–265, 278, 284, 291 Scores v, 54, 63, 72, 74, 79, 85–86, 102–103, 113, 120, 204, 243 Second language v, 2, 4, 7–8, 10–12, 17–20, 27–29, 31–33, 36, 51, 59–60, 69–70, 85, 87–91, 93–97, 106–108, 111–112, 163–164, 189, 212, 218, 220, 227, 249, 256, 258, 262, 268, 272–276, 279, 283, 286–288, 290–296 Second language learner 59, 93–95, 97, 107–108 Second language learning 2, 4, 7, 11, 17–18, 112, 249, 279, 283 Secondary education 197–198, 201–206 Secondary schools 179–180, 189–191, 193, 197, 225–226, 259, 292 Self-observation 35, 39 SLA 9–10, 16, 60, 87, 288 SLRF 287 Social v, 1, 3, 35–38, 47, 51, 53–55, 58, 97, 107, 111, 113–116, 118–119, 121, 124, 126, 130, 135–137, 141–143, 151, 155, 162–164, 168–169, 174, 188, 194–196, 202, 208, 211–212, 223, 225, 232, 256, 259, 275, 279, 282–283, 285, 287–288, 290, 294–296 Social context v, 35, 53, 58, 259 Social ecology 35, 37–38, 51, 55 Social networks 283, 294 Sociolinguistics vi, 2, 4, 7, 18, 133–135, 137, 139, 141–143, 274–275, 277, 281, 287–288, 291, 293 Software 215, 218, 222, 229 Spanish vi, 3, 25–28, 35, 37–42, 44–48, 52–55, 58, 99, 103, 138–139, 171, 174, 191–192, 235, 237–246, 248, 250–253, 255, 258, 292 Spanish FLES program vi, 235, 239, 242, 245

306 Index

Speaking 12–13, 35, 37–38, 40–42, 46–48, 50, 52–53, 55, 59, 86, 88, 98, 135, 138, 140–141, 146, 158, 160, 162–166, 175, 191, 197, 201, 213–215, 217, 238, 250, 252, 262, 274, 281–282, 284, 294 Speaking skills 201 Speech 20, 32, 45, 68, 71, 82, 91, 98, 107, 111, 138–139, 152, 272, 279–283, 285–286, 291, 295–297 Speech acts 272, 279, 283, 286, 291 Standard 36, 43, 49, 75, 78, 93–98, 104, 107–108, 117, 123, 156, 158–160, 162–163, 167, 177, 188, 236, 280–282 Status policy 171, 185 Swahili 188 Sweden 135, 189, 289 Switzerland 137, 139, 173, 186 T Target language use 35, 37, 40, 55–56, 66, 70, 72, 74, 82–83 Tasks 22, 25–26, 28, 35–37, 40–41, 47–48, 55–56, 59, 64–66, 69–76, 78–79, 81–85, 106–107, 115, 176, 205, 243–244, 248, 255, 279 Teacher 1, 8, 60, 90, 107–108, 111, 115, 130, 160, 164–165, 180, 214, 220, 231, 235, 237, 239–244, 247–255, 258, 263–264, 278–280, 283–284, 288, 290–292, 296 Teacher preparation 180, 237 Teacher qualification 164 Teacher training 160, 214, 220 Teaching v, vi, 2, 5, 7–18, 36, 87–89, 93, 98, 108, 126, 129, 155, 164–169, 171, 173, 175, 178–181, 186, 188, 190, 192–195, 197–203, 208, 211–212, 217, 219–221, 226–228, 230–232, 234, 236, 239, 244, 249, 252–253, 261–269, 271–284, 286–288, 292, 294–295

Territorialism 2, 145, 149–151 Tertiary education 197, 199, 200, 202–204, 232, 285, 297 TESOL 2, 4, 17–18, 88, 90, 231, 271, 275, 278, 287–289 Testing v, 2, 9, 59–61, 63–65, 67–68, 70–71, 73, 78, 81, 84–91, 111, 114–130, 161–162, 167, 174, 178, 208, 244, 274, 281, 286, 296 Testing methods 88–89, 118, 122–123, 130 Thai 216, 219, 230 TLU 70–73, 82 TOEFL 107, 217, 231 Topical content 64, 68–69, 72, 82, 84–85 Topical knowledge 64, 67–69, 74, 79–81, 83–84 Trade and industry 176, 196 Transition 145, 187–188, 251, 296 Translation 8, 11, 16–17, 35–38, 40–49, 54–56, 59–60, 139, 142, 182, 287–288 Turkish 186, 285, 292 U United Kingdom 172, 191, 261 United States 135, 148, 150, 154, 172–173, 175, 177–180, 182–183, 187, 190–193, 235–236, 250, 261, 264–265, 268, 293, 296 University iii, vi, 1, 3–4, 7, 17–19, 31–32, 35–39, 43, 51, 55–60, 63, 87–91, 111–113, 130, 133, 141–143, 145, 155, 157, 164–165, 167–171, 175, 178, 184, 194–198, 200, 202–208, 211–224, 226–227, 229–235, 237–238, 240–241, 243, 245, 247–250, 257–259, 261–269, 271, 274, 286, 288–297 University language policy vi, 211, 214, 222–223, 233



Index 307

V Validity v, 2, 9, 63–66, 72–74, 76, 78–79, 82–83, 85–90, 115–116, 120, 130, 278 Verbal report 35, 39–42, 47–49, 51, 55, 57 Vernacular 146, 188, 280–281 Vietnamese 212, 216, 219, 225, 230 Vocabulary 42–43, 45, 48, 119, 182, 241–243, 277, 280 Vocational language education 178 Vocational language use 178 W Wales 14, 191 Wider communication 185, 194

Word awareness 24 World communication 188, 190 World language 66, 188, 286 Writing 20–23, 25, 28, 30–33, 35–36, 38, 40–43, 45–51, 55, 59–60, 78, 87, 98, 107, 115, 121, 133, 137, 150, 161–163, 168, 208, 211, 227, 229, 266, 275, 277, 280–281, 283–284, 286, 291, 294 Writing systems 20, 22–23, 30, 163 Y Yiddish 2, 145–154 Z Zionism 2, 145–150