Orders of Ordinary Action (Directions in Ethnomethodology and Conversation Analysis)

  • 90 302 10
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up

Orders of Ordinary Action (Directions in Ethnomethodology and Conversation Analysis)

ORDERS OF ORDINARY ACTION Directions in Ethnomethodology and Conversation Analysis Series Editors: Stephen Hester, Uni

1,210 40 2MB

Pages 244 Page size 252 x 378.36 pts Year 2007

Report DMCA / Copyright


Recommend Papers

File loading please wait...
Citation preview


Directions in Ethnomethodology and Conversation Analysis Series Editors: Stephen Hester, University of Wales, UK David Francis, Manchester Metropolitan University, UK Ethnomethodology and Conversation Analysis are cognate approaches to the study of social action that together comprise a major perspective within the contemporary human sciences. This perspective focuses upon naturally occurring talk and interaction and analyses the methods by which social activities are ordered and accomplished. From its origins within sociology, EM/CA has ramified across a wide range of human science disciplines, including anthropology, social psychology, linguistics, communication studies and social studies of technology. Its influence is international, with large and active research communities in many countries, including Japan, Australia, Canada, France, The Netherlands, Denmark and Sweden as well as the UK and USA. The International Institute of Ethnomethodology and Conversation Analysis is the major association of EM/CA researchers worldwide. It was set up in 1978 by Prof. George Psathas to provide a forum for international collaboration between scholars working in the field of studies of social action and to support their work through conferences and publications. It published several books in EM/CA in association with University Press of America. Now reconstituted under the direction of Francis and Hester, supported by an international steering committee, the IIEMCA holds regular conferences and symposia in various countries. This major new book series will present current work in EM/CA, including research monographs, edited collections and theoretical studies. It will be essential reading for specialists in the field as well as those who wish to know more about this major approach to human action. Other titles in this series The Academic Presentation: Situated Talk in Action Johanna Rendle-Short ISBN 978-0-7546-4597-9 Institutional Interaction: Studies of Talk at Work Ilkka Arminen ISBN 978-0-7546-4285-5 Beyond the Black Box: Talk-in-Interaction in the Airline Cockpit Maurice Nevile ISBN 978-0-7546-4240-4

Orders of Ordinary Action Respecifying Sociological Knowledge

Edited by STEPHEN HESTER University of Wales – Bangor, UK DAVID FRANCIS Manchester Metropolitan University, UK

© Stephen Hester and David Francis 2007 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise without the prior permission of the publisher. Stephen Hester and David Francis have asserted their right under the Copyright, Designs and Patents Act, 1988, to be identified as the editors of this work. Published by Ashgate Publishing Limited Gower House Croft Road Aldershot Hampshire GU11 3HR England

Ashgate Publishing Company Suite 420 101 Cherry Street Burlington, VT 05401-4405 USA

Ashgate website: http://www.ashgate.com British Library Cataloguing in Publication Data Orders of ordinary action : respecifying sociological knowledge 1. Sociology 2. Social structure 3. Ethnomethodology 4. Practical reason I. Francis, D. W. II. Hester, Stephen 301 Library of Congress Cataloging-in-Publication Data Orders of ordinary action : respecifying sociological knowledge / edited by Stephen Hester and David Francis. p. cm. -- (Directions in ethnomethodology and conversation analysis) Includes bibliographical references and index. ISBN-13: 978-0-7546-3311-2 (alk. paper) 1. Ethnomethodology. I. Francis, David. II. Hester, Stephen. HM481.O73 2007 305.8001--dc22 2006039296 ISBN: 978-0-7546-3311-2

Printed and bound in Great Britain by Antony Rowe Ltd, Chippenham, Wiltshire.

Contents List of Figures and Tables List of Contributors Preface

vii ix xi


Analysing Orders of Ordinary Action Stephen Hester and David Francis


Four Relations between Literatures of the Social Scientific Movement and their Specific Ethnomethodological Alternates Harold Garfinkel




The Technical Operations of the Levers of Power Wes Sharrock and Graham Button


Operating Together through Videoconference: Members’ Procedures for Accomplishing a Common Space of Action Lorenza Mondada


Doctors’ Practical Management of Knowledge in the Daily Case Conference Nozomi Ikeya and Mitsuhiro Okada


Auspices of Corpus Status: Bibliography* as a Phenomenon of Respecification Andrew P. Carlin







Law Courts as Perspicuous Sites for Ethnomethodological Investigations Michael Lynch Circumstances of Reasoning in the Natural Sciences Eric Livingston








Orders of Ordinary Action

Expert System Technology in Work Practice: A Report on Service Technicians and Machine Diagnosis Erik Vinkhuyzen and Jack Whalen


Thinking as a Public Activity: The Local Order of a Tibetan Philosophical Debate Kenneth Liberman


Cultures of Reading: On Professional Vision and the Lived Work of Mammography Roger Slack, Mark Hartswood, Rob Procter and Mark Rouncefield


The ‘Problem of Dust’: Forensic Investigation as Practical Action Robin Williams

Bibliography Index


211 225

List of Figures and Tables Figures 2.1 2.2 2.3 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 4.10 7.1 7.2

8.1 8.2 8.3 8.4 8.5 8.6 8.7 8.8 8.9 8.10 8.11 8.12 8.13

Steering by the stars from Puluwat to Pikelot Steering by keeping constant the positions of the Southern Cross and Antares Contexture of two dots


Surgeon’s gesture (A) anticipates explanation Surgeon’s demonstration of retractor (gesture D) is synchronized with explanation Introduction of retractor to trocar (gesture F) is synchronized with explanation Surgeon starts gesture, using head to confirm use of optical system for the audience Gesture continues, indicating intended action Gesture is completed, indicating off-camera space to the audience Surgical team’s view of the optical system The expert and the audience in the amphitheatre Interior endoscopic view (via hand-held camera) Exterior view of patient’s abdomen (via fixed camera)


25 27

55 55 56 56 57 57 58 59 59

Latour’s contrast between idealized science and ‘science in action’ Extension of Latour’s dialogue between idealized science and ‘science in action’

111 115

The jigsaw puzzle problem A continuity of detail The gestalt interrelationships of a group of pieces A discovered relationship between separate groups of pieces Discovered detail, part A Discovered detail, part B Specific search criteria occasioned by a puzzle’s semi-finished areas Initial rough grouping of edge pieces Observed gaps occasion the search for missing pieces Details examined from within a specific search A discovered relevance of detail A discovered problem in the relative lengths of the sides A Physics Problem

124 124 124 124 125 125 125 126 126 127 127 127 128


Orders of Ordinary Action

8.14 8.15 8.16 8.17

Method of solving the physics problem First checkers problem Second checkers problem Solution to second checkers problem

129 131 132 132

9.1 9.2 9.3 9.4 9.5 9.6 9.7 9.8 9.9 9.10 9.11 9.12 9.13

Xerox’s Rapid Action Procedure (RAP) for a high-volume copier problem Example of Xerox’s PRIDE interface User instructions from the PRIDE manual Technicians Jeff and Bruce use PRIDE to diagnose a copier fault PRIDE screen prompts inspection of the copier Inspection of the copier’s interior Technicians Al and Grant try to answer PRIDE’s question PRIDE’s offer of possible faults poses interpretation problems PRIDE suggests a course of action Restarted troubleshooting session enables a different decision Example of a EUREKA tip Annotated example of a EUREKA interface Example of a EUREKA tip based on field discovery

139 142 142 143 143 144 144 145 147 148 151 153 155

11.1a 11.1b 11.2 11.3

Mammogram showing right and left oblique views Mammogram with annotations Manipulation of mammograms by readers Screening reporting form, showing annotations by the radiographer

179 179 181 187

Tables 5.1 5.2

Classification of emergency hospitals in Japan Format of case report


Key differences between EUREKA tips and the RAPs

71 73 156

List of Contributors Graham Button, Faculty of Arts, Computing, Engineering and Sciences, Sheffield Hallam University, UK Andrew P. Carlin, Department of Library and Information Studies, University College, Dublin, Ireland David Francis, Department of Sociology, Manchester Metropolitan University, UK Harold Garfinkel, Department of Sociology, University of California, Los Angeles, USA Mark Hartswood, Division of Informatics, University of Edinburgh, UK Stephen Hester, School of Social Science, University of Wales, Bangor, UK Nozomi Ikeya, Palo Alto Research Center, California, USA Kenneth Liberman, Department of Sociology, University of Oregon, USA Eric Livingston, School of Social Science, University of New England, Australia Michael Lynch, Department of Science & Technology Studies, Cornell University, USA Lorenza Mondada, University of Lyon 2, CNRS, France Mitsuhiro Okada, International Christian University, Tokyo, Japan Rob Procter, Division of Informatics, University of Edinburgh, UK Mark Rouncefield, Department of Computing, University of Lancaster, UK Wes Sharrock, Department of Sociology, University of Manchester, UK Roger Slack, Division of Informatics, University of Edinburgh, UK Erik Vinkhuyzen, Palo Alto Research Center, California, USA Jack Whalen, Palo Alto Research Center, California, USA Robin Williams, School of Sociology and Social Policy, University of Durham, UK

This page intentionally left blank

Preface The chapters in this book have their origin in two international conferences of the International Institute of Ethnomethodology and Conversation Analysis (IIEMCA), held at Manchester Metropolitan University in July 2001 and July 2003. The IIEMCA developed from a Summer Institute founded in 1975 by George Psathas and Jeff Coulter at Boston University. Through the late 1970s and 1980s regular meetings were held and out of these came two books, G. Psathas (ed.), Everyday Language: Studies in Ethnomethodology (Irvington Press, 1979) and D. Helm, T. Anderson, J. Meehan and A. Rawls, The Interactional Order: New Directions in the Study of Social Order (Irvington Press, 1989). In 1989 the IIEMCA was formally established under the leadership of George Psathas. Through the 1990s the Institute sponsored a series of conferences, at the University of Amsterdam in 1991, at Bentley College, Boston, in 1992, and at Waseda University, Tokyo, in 1997. In addition to these conferences, the IIEMCA sponsored a book series in conjunction with the University Press of America. This resulted in five books: G. Psathas (ed.), Interaction Competence (1990); E.C. Cuff, Problems of Versions in Everyday Situations (1994); P. ten Have and G. Psathas (eds.), Situated Order: Studies in the Social Organization of Talk and Embodied Activities (1995); S. Hester and P. Eglin (eds), Culture in Action: Studies in Membership Categorization Analysis (1997); and P.L. Jalbert (ed.), Media Studies: Ethnomethodological Approaches (1998). Following the retirement of George Psathas, the IIEMCA was relaunched in 2001 under the new leadership of Stephen Hester (University of Wales, Bangor) and David Francis (Manchester Metropolitan University). They organized the two international conferences referred to above, entitled respectively ‘Orders of Ordinary Action’ and ‘Producing Local Order’. They also established with Ashgate the new book series Directions in Ethnomethodology and Conversation Analysis. Orders of Ordinary Action is the fourth book in this series. It brings together current work of leading figures in the ethnomethodological research community, alongside the most recent statement of the ethnomethodological programme by its founder, Harold Garfinkel. There were many excellent papers delivered at the two Manchester conferences and the task of selecting from these was a difficult one. In addition to sheer quality, amongst our considerations that shaped our selection of the papers to include in this book were a preference for substantive studies rather than theoretical and methodological debate, a reasonable balance between veterans and younger scholars, and diversity with respect to coverage of different substantive phenomena. In our view, the collection testifies convincingly to the continuing vitality of ethnomethodology and to the quality of research and scholarship in its studies.

This page intentionally left blank


This page intentionally left blank

Chapter 1

Analysing Orders of Ordinary Action Stephen Hester and David Francis

Ethnomethodology’s lasting achievement has been to place the investigation of ordinary, practical action at the centre of its studies and reveal, through those studies, the myriad empirical forms that such action takes. Its ‘discovery’ of ordinary action stands in contrast to the prevailing conception in the social sciences. In the latter, the social scientist is trained to regard ordinary action as having no inherent interest but only as a resource for the depiction of (what are deemed to be) theoretically significant phenomena of a more abstract character. Whilst social scientists may indeed make observations of ordinary action it is standard social scientific practice to then render these in theoretical frameworks designed to elucidate and explain general phenomena such as culture, ideology, social structure, deviance, and so forth. Ethnomethodology, in contrast, has uniquely sought to respecify ordinary action as a topic of inquiry in its own right. Its ordinariness lies in its mundane availability for the members of society. The idea that the members of society ‘know what they are doing’ is taken seriously in ethnomethodology. This commitment frequently has been misunderstood. It does not mean that ethnomethodology advocates an individualist and subjectivist theory of action. Rather, it involves recognition that the intelligibility of the myriad actions comprising social life is an accomplishment of those engaged in them. Social ‘order’, meaning the recognizable, intelligible, accountable features of such actions – the features that make them ‘ordinary’ to those engaged in them – is reconceived as endogenously produced. It is part and parcel of the ways that the members of society realize ordinary action rather than being located at the level of theoretical abstractions. Ethnomethodology’s fundamental question, then, is how such mundanely available ordinary action is accomplished as such. In answering this question, ethnomethodology has produced an extensive programme of studies into a wide variety of settings and activities. By even a conservative estimate, ethnomethodology’s studies now number several thousand, ranging over topics as diverse as the interactional competencies involved in telling a ‘dirty joke’ in conversation (Sacks, 1974) to the situated telling of a ‘convict code’ (Wieder, 1974), and from the practices involved in successful control of air traffic (Harper and Hughes, 1992) to those that constitute the reasoned accomplishment of psychiatric assessments (Peyrot, 1995). In this tradition of studies various analytical themes have emerged. In this book, we wish to draw attention to two of these. One is the organization of practical action; the second theme is practical reasoning. Of course, practical action and practical reasoning are never separate: all action involves reasoning and all reasoning consists of action. However, ethnomethodological studies vary to some degree with respect to which


Orders of Ordinary Action

of these two aspects they emphasize. In both cases ethnomethodological studies seek to examine the local situated character of activities. With respect to practical action, ethnomethodological studies reject decontextualized descriptions of ‘social structure’ and instead focus upon how, within a given setting, structures of action are produced and managed. With regard to practical reasoning, ethnomethodology rejects the standard social scientific conception of reasoning as some kind of abstract apparatus, and emphasizes instead the fundamentally situated and contingent character of reasoning as a practical matter. Following this introductory chapter, the studies in this book begin, in Part One, with Harold Garfinkel’s plenary address to the 2001 conference, entitled ‘Four relations between literatures of the social scientific movement and their specific ethnomethodological alternates’. This chapter is unlike those in the rest of the book in two ways. First, it is not a report on a piece of empirical research, but rather a summative statement of ethnomethodology’s programme by its founder. Secondly, since it was given as a talk, and an amusingly informal one at that, we have decided to retain its original spoken form as far as possible. In the talk Garfinkel explicates the rationale of his book, Ethnomethodology’s Program: Working Out Durkheim’s Aphorism (Garfinkel, 2002). After Garfinkel’s chapter the book is organized into two sections in line with the themes discussed above. Part Two consists of four chapters that focus on the organization of practical action in a variety of organizational settings. Part Three consists of six chapters that examine forms of situated practical reasoning. Garfinkel, Literatures and their Ethnomethodological Alternates In his chapter, Garfinkel advocates a deliberate misreading of Durkheim’s aphorism, ‘the objective reality of social facts in sociology’s fundamental phenomenon’. Garfinkel’s argument is that the very possibility of subjecting social facts to statistical manipulation or theoretical explanation, as Durkheim recommends, is premised upon their appearance as objective phenomena in the first place. The possibility of that objectivity, therefore, is not something that a Durkheimian investigation can address since it is presupposed from the outset. The worldwide social science movement, as Garfinkel refers to it, shares the Durkheimian view. It consists of ‘literatures’ in which social phenomena are described and explained in the language of constructivist sociological theorizing. In every case such literatures presuppose the availability of the phenomena they seek to explain. The ‘ethnomethodological alternate’ to the constructivist literature consists in the study of the methodical accomplishment of the phenomena whose availability and intelligibility is otherwise taken for granted. As a literature, the ‘world-wide social science movement’ consists in analyses of real-world activities, analyses that identify the orderlinesses of such activities in terms of social science’s defining concern with accounting for social organization. These orderlinesses are described in various ways. For example, studies in the sociology of medicine have identified various ‘asymmetries’ in doctor–patient interaction. Typically, doctors ask questions, patients supply answers; doctors issue instructions and patients comply with them; patients may offer (when invited) their

Analysing Orders of Ordinary Action


opinions on their illness, whereas doctors provide authoritative judgements. These asymmetries are then explained by some sociologists of medicine by a theory of professional dominance in which, by virtue of a differential distribution of power between the doctor and the patient, the former is able to perform actions that are precluded for the latter. Such is the appeal of this theory that it has been applied to social interaction in a wide variety of institutional settings. The literature of medical sociology, then, is constituted as a literature by the ways in which it ‘finds’, describes and accounts for phenomena such as ‘asymmetries’ in medical relations. In other words, it is the task, and the defining characteristic, of social science as a literature that it involves methods for the production and explanation of relevant orderlinesses in social life, where such orderlinesses may take the form of patterns, rules, methods and principles of social organization, features which can be explained with reference to social science theories. However, literatures are not confined to those of academic social science. Whilst social science takes an interest in the orderliness of activities from its own particular point of view, so also do other kinds of literatures. It is readily apparent that literatures comprising other academic disciplines as well as those pertaining to the arts of practical action describe and analyse the relevant orderlinesses with which they are concerned. Thus, for example, activities such as horticulture, music and surgery are described in their respective literatures in terms of the proper and adequate methods and means of their production. Garfinkel asserts that ethnomethodology stands in relation to such literatures as a radically ‘alternate’ form of inquiry. It needs to be emphasized that this does not mean that ethnomethodology wishes to criticize them. Rather, it means that ethnomethodology’s concern is with the possibility of such literatures, in the sense of investigating that which makes them possible. For a literature to be possible, it must possess a subject matter, an activity or a collection of activities that it analyses. Thus, for there to be a sociology of medicine, there must be medical problems, activities and relations, that is to say, things that are recognizable as such in the real world. Similarly, for there to be a literature on horticulture there must be activities available and recognizable as horticultural activities – such things as planting, pruning, weeding, and so forth. These literatures treat as a premise of their work the availability of the activities that comprise these subject matters. They take that availability for granted. It is the fundamental resource that makes their analyses possible. Thus, the literature does not have to provide for the possibility of its subject matter. It can simply assume that the competent reader will have no problem knowing what this subject matter consists of. It is assumed that the reader will readily know, for example, what ‘going to the doctor’, or ‘describing a pain’ or ‘planting’ and ‘digging’ are as activities in the real world. Where these literatures take for granted the possibility of their subject matters, for ethnomethodology this possibility is the topic of its inquiries. Its question is how constituents of these subject matters are produced such that they are then available as the subject matters they recognizably are. If, for example, the subject matter is medicine, then how is ‘medicine’ recognizably produced as a thing in the world? Similarly, if the subject matter is horticulture, how are the constituent activities comprising this subject matter produced such that horticultural phenomena are


Orders of Ordinary Action

available to be described and analysed? In studying orders of ordinary action, then, ethnomethodology involves an analytical ‘stepping back’ to the activities themselves which provide the taken-for-granted recognizable domains of any literature. Traditionally, social science has been uninterested in these domains except and in so far as they offer starting points for professional theorizing. It is has been assumed that the orderliness of these domains is something that is discoverable via the methods of professional social science. From this point of view, ethnomethodology’s project has frequently been misunderstood as a critique of professional social science. This alleged critique finds the methods of social science as over-generalized, with the result that social science’s descriptions and explanations take insufficient account of the detail of social life. Ethnomethodology’s inquiries, it is argued, are directed at supplying the ‘missing detail’. It is necessary to emphasize here that the requirement in ethnomethodology is not for more detail in the general descriptions offered by the literature. Ethnomethodology is not advocating the pursuit of detail in order to improve the adequacy of general descriptions. General descriptions are not inadequate because they lack detail; there is no ‘deficiency’ to be remedied by describing activities more fully or closely. Consequently, the mere pursuit and provision of more detail may justifiably result in charges of triviality; what purpose is served? Furthermore, if ethnomethodology involved pursuing detail in order to better ground general descriptions of activities it could hardly be claimed to have the ‘alternate’ character that Garfinkel asserts. Rather than pursuing detail, ethnomethodology’s task is quite different. As already indicated, where conventional social science takes for granted the availability of the activities it describes, ethnomethodology seeks to examine their ‘possibility’. In other words, it is interested in the work that has to be done to make the activity in question available or recognizable as that activity in the first place. Indeed, ethnomethodology has no more interest in criticizing social science than it has in criticizing horticulture. It just ‘looks the other way’ to the way in which these and other literatures look at their subject matters. Consequently, its business is not the same as that of the literatures to which it is alternate. In speaking of ethnomethodology as ‘alternate’, then, Garfinkel does not mean that it seeks to offer some alternative theory or explanation of social life that competes with those of conventional sociology. Rather, its focus involves a kind of ‘stepping back’ from the analytical projects of social science, informed as these are by a concern to theorize the activity under study. Where, for social science, the activities in question are already ‘in place’, such that they can then be described and explained, ethnomethodology is concerned with what precedes any social science theorizing. To use the language of phenomenology, it advocates a return to ‘things themselves’ and asks how are they are produced in the first place so as to be recognizable as the things they are. A Musical Example To illustrate these points, we will refer to the activity of music. This example is ours, not Garfinkel’s. There is a social science literature that examines music in terms of its social context and the social organization of musical production. There are other

Analysing Orders of Ordinary Action


academic literatures concerned with music, such as ethnomusicology, which focuses on the cultural variations in musical forms, the history of music, the psychology of music, the economics of music, and so forth. In addition to these academic literatures, there is a vast ‘practical’ literature concerned with such things as (1) how to make music of this or that kind, by means of playing this or that instrument (or the human voice), (2) how to appreciate and understand music of specific kinds, and (3) ‘consumer guidebooks’ to various categories of recorded music. The contrast between these literatures and the ethnomethodological alternate is, with reference to music, something of a ‘cause célèbre’ in ethnomethodology, since it was with reference to ethnographic studies of jazz musicians that the notion of ‘the missing whatness’ of such a literature was identified. Garfinkel has credited Sacks with this insight. By the notion of the ‘missing whatness’, Sacks drew attention to the fact that in the writings of ethnographers of music such as Howard Becker, one is informed about many aspects and organizational features of the activities of musicians – such as the characteristic attitudes of musicians towards their audience, the economic vagaries of the musician’s lifestyle and so on. However, what is noticeably absent from such ethnographic accounts is any discussion, let alone detailed examination, of the ways in which musicians actually make music together. Of course, the business of how to make music is the topic of other kinds of music literatures. As any aspiring musician knows, there are instructional manuals one can consult that offer guidance in how to play a particular instrument, or, in the case of singing, how to develop, train and control the voice. Such manuals present strategies that, if followed correctly and assiduously, should result in the acquisition of a level of instrumental competence. However, as David Sudnow has pointed out, the use of such instructional texts involves the player in finding ways to realize the instructions in and as the activity of playing the instrument. It is these actual ways that are necessarily left unexplicated by the manual. However, without some solution to the problem of what the instruction means as actual, situated musical work, the manual remains simply a text. The task, therefore, for the ethnomethodological alternate to the instrumental handbook is to describe in detail what is involved, in any particular case, in making the textual instructions ‘happen’ as music. Of course, there is a sense in which the paramount musical literature, for most kinds of music at least, is the musical score. It is here that the music to be made is notated. To be a musician is, perhaps first and foremost, to be capable of taking a score and transforming it into the corresponding musical performance. The competencies involved are not to be found in the score, rather they are presumed by it. For example, at a given point in the score the player (singer) may be instructed as to the dynamics and tempi of the music (for example, ‘pianissimo’, andante’, etc). Such an instruction presumes the player’s ability to both understand the meaning of the instruction and produce appropriate musical sounds in response. Such instructions are ‘indexical expressions’, in that what will count as appropriate realization is a local and contextual matter. Thus, understanding what is meant by the score’s instructions involves not simply what a given instruction means in general but what it intends as produced musical sounds at this point in the piece. In this context, then, Garfinkel’s description of ethnomethodology is itself a ‘literature’ that formulates the orderliness of ethnomethodological work, with respect


Orders of Ordinary Action

to which actual ethnomethodological studies stand as alternates. As Garfinkel himself emphasizes, the relationship between a literature and its ethnomethodological alternate does not involve ‘evaluation’ or ‘criticism’. Therefore, it is no criticism of Garfinkel’s general account of ethnomethodology to say that it presumes the accomplished production of the very activities that are thereby formulated. Indeed, Garfinkel offers students of ethnomethodology some ‘advice’ about how to think about the relations between literatures and their ethnomethodological alternates. There are, as he points out, many ways of conceiving these relations. However, in his chapter he states that he will limit himself to the discussion of just four of these. They are: (1) the first rendering theorem; (2) the embodied reflexivity of instructed actions; (3) a comparison of classical properties of instructed actions with phenomenal field properties of instructed actions; and (4) the second rendering theorem. In fact, the plenary talk discusses only the first three. Rather than trying to paraphrase his advice, we will leave it to readers to discover and judge just how they may learn from his arguments. However, the general thrust of Garfinkel’s argument is exemplified by his remarks on the use of the ‘first rendering theorem’ and so we will comment briefly on this. Thus, by the ‘first rendering theorem’ Garfinkel refers to the practice amongst social scientists of transforming ordinary actions into analytical objects or, more precisely, instances of theoretical phenomena. The process of transformation is premised on the position that there is ‘no order in the plenum’. In other words, ordinary ‘concrete’ social life can only properly be understood by being transformed into analytical objects since only the latter enable orderliness to be displayed. In this transformation, the original orderliness of the phenomena, as it was produced by the parties to it such that it was intelligible as the phenomena it was, is lost from view. Part Two: Studies of Practical Action in Organizational Settings The studies in Part Two focus upon the accomplishment of concerted action in various organizational settings. In Chapter 3, on ‘The technical operations of the levers of power’, Wes Sharrock and Graham Button respecify the phenomenon of power as a members’ practical matter. Their data consists of transcripts of middle management meetings in a call centre. The business of these meetings concerns the assessment of staff performance in relation to work targets. The analysis considers the practical reasoning involved in identifying inadequate staff performance. Such identification is a contested matter, in that local managers and their regional superiors offer contrasting versions of the nature of variations in staff performance. The analysis reveals how regional managers superimpose an ‘organizational’ perspective upon the locally oriented accounts provided by staff themselves and their local managers. It is argued that it is in and through such imposition that organizational leverage is brought to bear and power locally manifested. In Chapter 4, Lorenza Mondada’s topic is ‘Operating together through videoconference: Members’ procedures for accomplishing a common space of action’. Mondada’s chapter examines a particular work situation, in which surgeons operating on a patient are connected via a videoconference device. In addition, the

Analysing Orders of Ordinary Action


surgeons’ work is monitored by experts and an audience composed of advanced trainees. The analysis considers how surgeons so connected accomplish together the medical operation at hand. The focus is on how videoconference technology is used to locally shape a common space of action relevant for a precise medical task. It is shown how these spaces are contingently configured and transformed as members go from one task to another, from one sequence to another. Some particular sequences of talk-in-interaction that are involved in this are identified. The analysis shows that technology does not simply provide fixed places where members work or speak but is a resource exploited in activities by which a shared workplace is constituted, and a complex task accomplished. Participation frameworks are shown to be sequentially achieved and to constitute different categories of recipients such as learners, experts, peers and co-operators. In Chapter 5, Nozomi Ikeya and Mitsuhiro Okada analyse ‘Doctors’ practical management of knowledge in the daily case conference’. They show that Schutz’s concept of the social distribution of knowledge can illuminate aspects of work in organizational settings. The setting examined is the daily case conference in a Department of Emergency and Critical Care Medicine of a university hospital. The analysis focuses on how the senior doctor tries to achieve quality control of the treatment of all patients in the department while at the same time endeavouring to be ‘educational’ with respect to junior doctors who directly deal with the patients. There are two modes of the practical management of knowledge displayed by the senior doctor observable in the case conference. The first of these involves their assessment of the team’s treatment of each patient. The senior doctor examines how the team members have come to give such diagnosis to the patient. The second consists of the provision of structures to junior doctors’ particular experiences through their location in a shared medical stock of knowledge. Chapter 6, by Andrew Carlin, on ‘Auspices of corpus status: Bibliography* as a phenomenon of respecification’, reports a project concerned with analysing the constitution and organization of disciplines and interdisciplinary studies, by examining bibliographies as disciplinary artefacts. One aspect of this is taking bibliographies as topics for investigation in their own right. In considering the auspices of bibliographies compiled from available literatures on a topic, the chapter draws upon the writings of Garfinkel, whose recent work provides a perspective on the processes involved in constituting a discipline (or a new branch or topic of it). Garfinkel suggests that bibliographic materials (that is, their corpus status) make the discipline’s phenomena ‘instructably observable’. The analysis shows how the local organization of bibliographies is a phenomenon of ethnomethodological respecification; and that research programmes are available as bibliographic phenomena. Part Three: Studies of Situated Reasoning This part of the book contains a collection of studies that emphasize the situated character of practical reason. The first of these, Chapter 7 by Michael Lynch, discusses legal reasoning around the uses that can be made of ‘science’ in the


Orders of Ordinary Action

courtroom. Such reasoning, Lynch suggests, makes law courts ‘a perspicuous site for ethnomethodological investigations’. As he points out, law courts are sites in which the contrast between ‘science’ and ‘common sense’, rather than being a matter of abstract theorizing, can be a matter of practical concern and significance. Science has been a resource for deciding not only the authority of evidence but also in deciding both ‘what happened’ and how it came to happen. In this way, courtroom personnel have practical, legal interests that in some ways are congruent with the analytic interests of ethnomethodology, hence one sense of their perspicuousness. Interestingly, Lynch also shows how the practical accomplishment of scientific (that is, forensic) investigation, can also be subjected to ‘ethnomethodological’ or ‘ethnographic’ investigation in the courtroom. In Chapter 8, Eric Livingston discusses ‘Circumstances of reasoning in the natural sciences’. Reasoning, and reasoning in the sciences particularly, is usually treated as a universal phenomenon, transcending the immediate material circumstances to which it is applied. In contrast, Livingston examines the discipline-specific character of reasoning in chemistry and physics. In chemistry, problem-solving involves the conjunction of indicial experimental relationships whereas, in physics, parametric models are used to characterize empirical situations. Through a comparison of a problem in chemistry with one in physics, the chapter begins to show the possibilities of a return to the detailed phenomena of reasoning, therein to develop an anthropology of reasoning in the sciences. Livingston’s chapter is followed in Chapter 9 by Erik Vinkhuyzen and Jack Whalen’s study of ‘Expert system technology in work practice’. Despite the controversy during the past thirty or more years over whether computer programs can robustly reproduce – and perhaps significantly improve upon – the judgement and behaviour of human beings who have expert knowledge and experience in a particular field, there have been remarkably few field studies of these expert systems in actual use. In this chapter, the authors report upon such a study, an ethnomethodologically informed, historical ethnography of a system for diagnosing problems with document machines that was designed at Xerox PARC in the early 1990s and then modified by Xerox’s equipment service organization in 1994–95 for use by their service technicians. The investigation of this system’s design, re-design, and operational deployment, including how it was used by technicians in practice, focuses on several issues. Firstly, how is ‘diagnosis’ most commonly organized as a course of practical action in the work of these technicians? Secondly, how does this practical organization compare with the model of diagnosis embodied in the repair manuals, especially the repair analysis procedures, or ‘RAPs’, that technicians are expected to employ? Thirdly, and finally, the analysis considers what it means to support a particular human activity or task, like problem-solving or diagnosis, with or through a computer system. What did (or could) ‘support’ actually look like, with respect to the work practice of Xerox service technicians? In Chapter 10, Kenneth Liberman investigates ‘Thinking as a public activity: The local order of a Tibetan philosophical debate’. In it, Liberman reports a study of some practices of reasoning amongst Tibetan Buddhist monks. The study focuses on the limits and benefits of logical analysis and formal analytic reasoning as courses of practical action. The analysis considers what it is about formal analysis and logic that

Analysing Orders of Ordinary Action


compels Tibetan monks to devote years and decades to its study when their aims are predominantly spiritual ones. The central focus is on the question of what is formal reasoning as an in-situ, empirical matter for ethnomethodological discovery? Chapter 11, by Roger Slack, Mark Hartswood, Rob Procter and Mark Rouncefield, is concerned with ‘Cultures of reading: On professional vision and the lived work of mammography’. Specifically, its focus is on the lived work of reading mammograms. Radiologists and other medical professionals involved in this work are tasked with ‘reading’ mammograms for evidence of pathologies. The work is highly skilled, both technically in the creation of informative X-rays and interpretively in constructing an ‘accountable reading’ of the state of the breast in and as the representations examined. The analysis shows the importance of the use of artefacts, procedures, notes and annotations in the realization of a diagnosis. It shows how medical staff make candidate diagnoses available to each other through annotations and how suspicious objects are subject to ‘worldly interpretation’ in and as a part of the development of diagnoses. The chapter also argues against cognitivist versions of reading and situates reading as a thoroughgoing social practice. Finally, in Chapter 12, Robin Williams considers ‘forensic science’ as practical reasoning. The central role of forensic science in contemporary criminal justice reflects the confidence of law enforcement agencies that a particular body of scientific knowledge and repertoire of technological devices can be used to validate unequivocal and authoritative accounts of criminal actions and criminal identities. The chapter begins with a brief description of a recent study of ‘crime scene examiners’ as central participants in forensic investigations. The second section contrasts textbook efforts to stipulate idealized ratiocinative standards for the cognitive orientation of forensic investigators with what detailed empirical study reveals as two working principles accountably oriented to by investigators in the course of their construction of, and work within, crime scenes. The third section describes how these same principles not only give direction to the production of a range of ‘forensic artefacts’, but are also embedded in the methods of their production. Finally the chapter considers a recent dispute over the forensic examination of a particular domestic burglary, showing how skilled practitioners used their detailed knowledge of the making of such artefacts to raise doubts about the credibility of one particular instance which was central to the prosecution case. The Studies as ‘Alternates’ Each of these chapters can be understood in the terms of Garfinkel’s distinction between literatures and their ethnomethodological alternates. Thus, each one approaches its subject matter with the aim of analysing and explicating the methodical production of that which is given and taken for granted in the manuals, syllabi, endogenous commentaries and critiques, self-evaluations and, of course, constructivist social scientific theorizations that rest and build upon the intelligibility and availability of their phenomena. For each of ‘forensic science and reasoning’, ‘philosophical thinking and debate’, ‘educational instruction in medical practice’, ‘mammography’, ‘document machine diagnosis’, ‘reasoning and problem-solving in


Orders of Ordinary Action

chemistry and physics’, ‘the compilation of bibliographies’, ‘the use of technology in surgical operations’, and ‘the managerial exercise of power in business organizations’, the studies collected here respecify that intelligibility and availability as the accomplishment of local orders of ordinary action. We leave it to the reader to discover the specificities of this accomplishment in each of the chapters that follow Garfinkel’s plenary address.

Chapter 2

Four Relations between Literatures of the Social Scientific Movement and their Specific Ethnomethodological Alternates Harold Garfinkel

In 1985 I made a trip to London to see Peter Hopkins at Routledge. He was preparing a series for Routledge to be edited by me called Studies in Ethnomethodology, to consist of books by students who were working with me for their PhDs. We were discussing their contents when he showed me the jacket blurb he was preparing. The blurb started, ‘Harold Garfinkel, the father of ethnomethodology ...’ ‘I’m the father of ethnomethodology?’ I protested – ‘I despise that blurb.’ Hopkins replied, ‘But Harold, everybody knows you’re the father of ethnomethodology.’ And I said, ‘Yes, and everybody knows the ethnomethodologists are a bunch of bastards. But nobody knows whose bastards they are.’ Now I’m in the position of asking you to tell me some time who your fathers are but not tonight. Tonight I’ve taken as my fatherly job, having been invited to address a familial mix, to remind the company of us that we’re onto something very big. I would like to discuss the book that is currently in press [Garfinkel, 2002] and the corpus paper. I’ll make my talk an invitation. I shall offer remarks by way of calling to your attention what most interested me in the book and I thought should be of interest to others – what the book has as its heart and soul. In preparing today’s talk I elaborated that central concern as points to get our discussion underway. First of all, here’s a book. It’s been late in coming to publication and for that I’ve taken some abuse. ‘What are you sitting on it for? Let’s put it in print.’ Now we have my reply. I grant it’s late but in light of the fact that the company of ethnomethodological authors have come upon something, there’s a proper answer to these complaints. I learned it from a page of Islamic proverbs: ‘Haste is for the devil’. [Audience is silent] You’re a solemn crowd. [Laughter] Durkheim’s Aphorism The title of the book owes its accuracy to Anne Rawls, my colleague and the editor of the book and to my lights and in my offerings she would be a co-author but she refused. The title of the book in its correct sequence is Ethnomethodology’s Program: Working Out Durkheim’s Aphorism. I want to comment on ‘Working Out Durkheim’s Aphorism’. The aphorism is well known to all of us. On our first day of


Orders of Ordinary Action

graduate work we were told to take Durkheim’s aphorism to heart. Put it on our door sills. Seeing it there each next time, bless Durkheim and bless yourself, but learn the mantra. According to Durkheim’s aphorism, ‘The objective reality of social facts is sociology’s fundamental principle.’ Sociology’s fundamental principle?? There’s the rub. Durkheim had it beautifully and originally and right but his particular word ‘principle’ was ill advised. It didn’t help that in another version he wrote ‘The objective reality of social facts is sociology’s fundamental phenomenon’. For ninety years he was understood to have been speaking on behalf of contemporary, technically detailed, formal analytic social sciences. The social sciences owe a professional debt to Anne Rawls’ scholarship [Rawls, 1996, 2001] for showing that for ninety years the aphorism has been systematically misread. For ninety years it has been understood by the social sciences that Durkheim meant by ‘the objective reality of social facts’ what the contemporary worldwide social science movement takes as its own good work in descriptions of order in ordinary society: that the concreteness of social facts of ordinary activities are respecified so that local, actual, concretely detailed, circumstantial workings of immortal ordinary society are absent of orderliness until and only in case concreteness is respecified by the social sciences, displayed in details of orderlinesses of formal analytic methods and generic representational theories. With formal analytic methods, order in, about, as, inside with, as of, and within ordinary society is respecified from original ignorant provisions in prosaic everyday accounts of orderlinesses of daily life, lived in their concreteness, tangled in circumstantiality, and empty of order. In their place, orderlinesses of ordinary social facts, things, are specified with social science methods. Therein, order phenomena are revealed in occasion after occasion, in real case after real case, with the truth of the matter being at hand and of concern under any and every circumstance. That is the objective reality of social facts understood in accordance with established methods of the worldwide social science movement, and the corpus status of its researches. Those respecified orders and orderlinesses of ordinary society, phenomena of order exhibited in the details of social science methods and formal analytic lexical devices of generic representational theories are the objective reality of social facts. In their instructably accountable production and observability of order in ordinary society they are exhibits of sociology’s fundamental principle. No-o-o. For ninety years Durkheim was talking about something else, even though that something else was distinctively and uniquely sociology’s subject matter. Which is what the studies by ethnomethodology’s authors are all about. Ethnomethodology understands Durkheim’s aphorism differently. Ethnomethodology has been respecifying Durkheim’s aphorism so that it reads differently. Yes, his aphorism reads, ‘The objective reality of social facts is sociology’s fundamental phenomenon’. But those words are to be read like this. Ethnomethodology’s Program It is the programmatic task of ethnomethodological studies to specify the naturally accountable work of producing and describing the social facts of immortal, ordinary

Social Science and Ethnomethodology


society. These are the Things of social order – the distinctive order phenomena of ordinary society that Durkheim was actually talking about. Populations, for example. Formal analytic studies focus on aggregated individuals described demographically as enumerated populations. Ethnomethodology proposes instead that the local, endogenous workings of the phenomenon, the Thing, the social fact – for example, freeway traffic jams, walking together, the exhibited order of service in formatted queues, turn-taking in conversation – exhibits among its other details the coherence of its identifying orderliness. The coherence of its identifying orderliness is identical with the population cohort that staffs its accountable production. Ethnomethodological investigations have their origins, aims, directions, policies, methods, the corpus status of their methods and results, their clients, and their consequences, in the worldly and real work of making the Things that Durkheim was talking about discoverable, and making their discovery accountably evident as Things of immortal, ordinary society. ‘Working out Durkheim’s aphorism’ ethnomethodologically is concerned with this work, how it is done, by whom, in various bibliographies, its argued diversity, its center, its consequences, what’s in it for departmentally based sociology, but more consequentially, what’s in it for the endless analytic disciplines and sciences of practical action and practical reason. In its central tasks ethnomethodology is directed to the reform of technical reason, and is doing so with the premier aim of specifying the work of the social sciences and the natural sciences as naturally accountable sciences of practical action and practical reason. Contrary to the commonly accepted interpretations of Durkheim’s work, from the beginning of his writings the objective reality of social facts was Durkheim’s provision for sociology’s subject matter as a distinctive and unique subject matter. Ethnomethodology took up this neglected initiative. According to ethnomethodology’s programmatic understanding the objective reality of social facts was Durkheim’s descriptive proxy for every topic of logic, meaning, reason, rational action, method, truth, and order in intellectual history, specified in any actual case as a congregationally produced and naturally accountable, endogenous order, a production populational cohort’s concertedly witnessable and recognized, intelligible empirical phenomenon of immortal ordinary society. Therein the phenomena of social order consist of lived, immediate, unmediated, congregational practices in their display, in their witnessed recognition, in their intelligibility and their accountability of immortal ordinary society’s phenomena of order, its ordinary social facts, the most ordinary Things in the world. For Durkheim of the neglected argument this was quite deliberately and incommensurably contrary to the canonical teachings of Western philosophy. In my talk today I have assumed the privilege of discussing a version of what Durkheim’s aphorism could possibly mean as it has been explained in the Introduction to [my] book. Chapter 1, the Introduction, is also called ‘The Corpus Status of Ethnomethodological Investigations’. When Steve Hester and Dave Francis invited me to this conference they allowed that I could write a paper for the conference. So I re-titled the Introduction. It just seemed easier to do it that way. Actually, to prepare my talk I read through the ‘Corpus’ paper as an introduction to our work as authors of ethnomethodological studies. When I came to prepare a spoken version of what otherwise could be read in silence my idea was to elaborate


Orders of Ordinary Action

the ‘Corpus’ paper’s argument by offering preliminary remarks that I hoped would encourage us to discuss its relevance for the company of ethnomethodology’s authors coming upon alternate studies of order in ordinary society to those of classical literatures. Ethnomethodology’s alternate investigations describe researches that are unfailingly directed to the immortal ordinary society that Durkheim made available to us, but understood ethnomethodologically, and in that way are taken away from the Durkheim of established ways of understanding what he was talking about. What therein was lost for ninety years of social analysis (in sociology and in many arts and sciences) of practical action and practical reason, has now been come upon and shows his neglected argument. Ethnomethodology’s Phenomena: Literatures and their Alternates Ethnomethodology’s program is concerned to rework Durkheim’s aphorism. It has to have reasons. Therefore, I’ll ask: Where does ethnomethodology get its motives? And where does it get its subject matter? That’s one question. Having asked it, I’ll ask again: What is it about the affairs of ordinary society that answers that question? Having asked those two questions I’ll also answer them: It is the worldwide social science movement; it is the respected, well-known, familiar iron-bound discipline of university departmentally based work addressed to the incessant tasks in every administered society in the world to report truthfully and faithfully with empirical documentation what, just what, do its ordinary affairs consist of in the hands of careful study. Ethnomethodology finds its subject matter alternate to that discipline. Its subject matter in fact does not start with texts. I don’t want to sound like a snot nose or ungrateful to the tradition of godlike authors in moral and social philosophy but the fact is that ethnomethodology finds its subject matter is guaranteed by the output, the results of procedures, the real results of the this-worldly existence of the formal analytic worldwide social science movement. Its literatures are the sources of ethnomethodology’s subject matter. I am speaking literally of literatures of the worldwide social science movement. You can start your studies – in any of an open variety of subjects – with publications, carefully read, written, authored, monitored, peer-reviewed, published – policies, methods, results. All that is to be found not generally speaking but in the specific subjects of every and each and all the arts and sciences of practical action and practical reason. As authors of ethnomethodological studies you don’t have to start your inquiries imagining the world and worldly things. You only need to go to your nearest academic library. Find an area that interests you. This morning we heard about surgery. From a linguist! It was wonderful. Our linguist went searching the literature of surgeons to find out what a laparotomy for them is made to look like in order that it be made to look correctly done. Right? Yes. It is no mystery as to where subjects for research are to be found. You don’t imagine them. You don’t define them to start with. You don’t define them before you start in order to start. You don’t construct them. But instead you go to the library and

Social Science and Ethnomethodology


do some reading. Pretty much of your own choosing. In fact the ethnomethodological emphasis I think should be: Find something that really pesters you. You can’t really be happy or settled without knowing something more about this or that, whatever it is. And then go to the library to find what the established version and treatment of that subject could be made to look like by reason of careful authorship in describing it. As long as it is careful authorship you’re in business because that literature will have its alternate. And that’s where the ethnomethodology is going to live. It’s NOT being said by ethnomethodologists that these literatures are wrong. Not at all! That’s a goofy way to think about what ethnomethodology is up to. Of course if it is going to be charged with a Literature’s interests in errors or making mistakes that’s something else. If ethnomethodology is going to be charged with interests in a Literature’s competences with absurdity, that’s going to be very interesting! Whereas for ethnomethodology being required to say because it is held to serve, or offers to serve, as judge and arbiter, being correct as a matter of a universal observer’s privileges in this-worldly practices and with these asserting judgmentally between disputants who is right and who is wrong and just where is the truth of the matter – is a distraction. Perhaps it will make you happy with your friends but those jobs are not anything of what ethnomethodology is all about. That’s not where your fascination is going to find itself. It is very interesting about catching on to the existence of a Literature and its Literature-specific ethnomethodological alternate that once you are in possession of that distinction and the subject matter, the existence of a Literature that permits you to read it alternately is as Larry Wieder once taught us when he growled, ‘I’ll sell shoes before I’ll do anything different’. Encouraging that kind of fascination has gotten me into trouble with authors of established literatures. That’s too abstract. Let’s just say, I’m a sociologist. I was born in the sociology profession. So that’s that. My companions in the Department, from time to time being antagonistic in their uncertainty that I was a proper guardian of the profession, would wonder: What the hell is this guy doing to our students? He must be a magician. No, he’s not a magician. He delivers interesting lectures to undergraduates about their troubles at home. He has charisma with them. He plays with their heads. He’s a charlatan. He’s leading them in a cult. He’s ruining generations. An occasional defender will argue, ‘No, he’s not a charlatan; it’s that the students have to be stupid.’ No, there’s not a stupid one in the lot. Withal the abiding question is: What gets these people engrossed and is such that, once they catch on they won’t have it differently than that, even in the face of long-lasting opposition – clever, energetic, inventive, incessant, vicious criticism, and political academic adversarial opposition. We can agree: From the beginning of our enterprises the promises of the literatures and their alternates are accompanied by wound stripes. These come with the work. Relations between Literatures and their Alternates In its studies of phenomena of social order ethnomethodology’s central interests and focus are with Pairs made up of Literatures and their specific ethnomethodological


Orders of Ordinary Action

alternates. I want to describe several relations between Literatures of the social science movement and their specific ethnomethodological alternates. These relations are a way to collect your thoughts so as to follow and use what’s going on in the corpus paper. The corpus paper proposes about all the subjects that are worked out and described in the established literatures that they turn up via the work that is seriously done and seriously understood as careful1 investigations by the worldwide social science movement of professional social analysts. In occupational workplace settings social analysts staff the society’s administered affairs as a knowledgeable company who compose for each other work environments of formal analytic colleagues. They are also ethnomethodology’s formal analytic colleagues. In their day’s work the multitudes of formal social analysts are in possession of literatures that exhibit their work in affiliated texts and visuals as a technology. To clarify our exposition we’ll call their work described in affiliated texts and visuals as a technology, a way of working as a discipline. As a way of working, as a discipline, the subjects of their work are collected in Literatures. Literatures of formal, analytic descriptions of order phenomena – of phenomena of social order. Of orders in ordinary society. Of the phenomena of order in, about, as, within, as of immortal workings of the most ordinary things in the world – the ordinary society. And, along with that Literature – that particular Literature, NOT a literature in general, NEVER a literature in general, but always a set of particular readings that you consult to find just what of that competent authored writing the described matter is described as. And how it is justified. You are obliged, in every case, to master that source. You cannot make believe with that source. You cannot start by saying ‘Oh, of such and such, he’s blowing hot air. Who cares?’ If you don’t want to read the materials, don’t read them. But do NOT assume that you are privileged to pass judgment. Now then, you have in hand a Literature. The Literature will have its specific – we’ll call it that – it’s specific ethnomethodological alternate. Those two, Literature and Alternate, make up a Pair. They go together. The Literature is to be read and carefully mastered, because it is Things, methods, and instructed procedures you are looking for. How does the Literature provide that the orderliness – order in and as of the Thing is done. How is it made instructably observable, instructably adequate and instructably evident? How is it done? Just how? With the Literature at hand you work out the alternate that goes with the Literature. About that job I’ll remind you of something you know very well. The alternate doesn’t come etymologically from the Literature. The alternate takes you somewhere else. Which is what I’ll be discussing in my next remarks. The corpus status of ethno-studies is a collection of Literatures and their alternates. Alternates not alternatives. ‘Alternates’ are CERTAIN phenomena that ethnomethodology learned about from the demonstrations of gestalt psychology. ‘CERTAIN’ when it is [thus emphasized] means certainty. It does not mean selectively chosen from alternatives.


Explain ‘careful’. [Editor’s note: This is a reminder by the speaker to himself.]

Social Science and Ethnomethodology


Alternates were borrowed from gestalt psychology, then revised by ethnomethodology and appropriated.2 Ethnomethodology’s Alternates I am speaking most emphatically. In ethnomethodology alternates need always to be examined and elucidated in light of their property that they are affiliated. They are unavoidably affiliated. Not by definition, but empirically. They go together. They have to do with each other. Therein the question is posed: In any actual case what they have to do with each other? Just how are the members of a Pair related? JUST HOW? With that to be claimed it will be the fact of the matter in your life that each particular Pair specifies this question: Just how are the two members of the Pair related? Just what do they have to do with each other? Just what, just how, just in any actual case? When I prepared my talk I examined a bibliography of ethnomethodological hybrid studies of work. I made a list to describe various ways that Literatures and alternates are related. Various ways. Various empirical ways. I started with four old reliables. Ethnomethodologically, they are old reliables. They are reliable ways of describing how these Literatures and their Pairs, their affiliated Pairs, their ethnomethodologically affiliated Pairs go together. What do they have to do with each other? I’m speaking in the fashion of offering good advice to researchers. So, by way of general advisories, here is some advice. Think of this; don’t settle for it, but relations can work like this: Literatures and alternates can be related like this. At times I’ll describe them in a fanciful way, even a seemingly scientific way. Don’t believe me on that score. The relations are not described scientifically. But the relations are descriptive and are meant to be helpful: • • • •

the first rendering theorem the embodied reflexivity of instructed actions a comparison of classical properties of instructed actions with phenomenal field properties of instructed actions in an episode of Polynesian navigation the second rendering theorem

The First Rendering Theorem We frequently speak about the first rendering theorem. Theorem? The first rendering theorem? As I said, talk about a theorem is fancy enough. It proposes a general, recurring procedure, and procedural advice. It is also to be done in and as research. Thus it is also something to be practised, found out about, and learned 2 Insert merely metaphoric examples of alternates with well-known gestalt drawings, diagrams, illusions, etc. (pretty young woman and ugly old woman; rotating trapezoidal window; Franklin Kilpatrick’s targeted balls). [Editor’s note: This is the speaker’s own note reminding himself to find these items.]


Orders of Ordinary Action

in each next actual research. It is this: Your home base will insist, the department from which you come, the national association of your department’s members that certifies that Literature and identifies its authors, will insist, the worldwide social science movement will have it to advise, its members will cut off your heads if you don’t respect that advice, if you don’t follow it, and in your work take it seriously – so that you either incorporate it – that in each, any, every, and for all actual cases you demonstrably and witnessably incorporate it into your inquiries or give good reasons for not doing so – and the policy is this: There is no order in the concreteness of things. If you are in fact, actually and not supposedly, to find anything of order in the ordinary society, then count on it, you’re going to be swamped by the plenum. The plenum is the theorists’ version of concreteness. There is no order in the concreteness of things. There is no order in the plenum. The plenty. The oh-so-much-of-it is beyond specification or articulateness. The plenilunium. None of it has order in it. If you are to do anything by way of finding order then you must model it. You must describe it according to our sketch of it: =( )= ---> ( ) Do you see me? [Audience murmurs, ‘Yes’] Here’s a way we can use a diagram – I’m talking in kind of a cryptic way. These double-ticked brackets on the left mean the holy hellish concreteness of things. The one damned thing along with another and any other and isn’t it first thing in the morning, last thing at night, you’re swamped by the sheer detailing. If you have to tell it, there’s no end to its telling. So let’s say for – whatever it is – you can find it for whatever suits you. =(Formatted queues that exhibit an order of service)= and =(Freeway traffic slowing to a stop)= These examples suit me. Here’s the rest of this device: ---> ( ). The arrow, --->, and now we’re talking ethnomethodologically, the arrow is any method that can be used to respecify the holy hellish concreteness of things. Plain brackets, ( ), refer to the results. What we have in this piece of advice, and the claim on the part of the social science movement, is that in the formal analysis of ordinary society, the concreteness defeats any effort to actually come upon just what is study-able, specifiable – demonstrably the case that makes up the orderlinesses of ordinary society. =( )= ---> ( ). Here we have a way of speaking about that job and it says: There is no order in the plenum. If you are to make a career as a professional analyst you have to master the methods that are available with which to respecify that concreteness. Speaking comprehensively, these ---> ( ) are methods, procedures, policies, workplace practices of generic formal analytic representational theorizing. These --> ( ) will produce as their results the teachable, claimable, recordable, demonstrable, reportable, witnessable product of your efforts to specify the ordinary society brought

Social Science and Ethnomethodology


to instructably witnessable intelligibility in its identifying orderlinesses in material details, adequately and evidently. Now there is the ethnomethodological alternate that we are talking about. What does IT consist of? Well, the alternate also reads this bit of cabalism, which it reads like this: Any ordinary activity addressed in the fashion of its availability in situ as the in vivo work of living in and about and as and as of the activity as what anybody in the world knows consists of in its lived course of things will, if you use the certified methods of the established literatures, respecify that concreteness to exhibit it in terms that then no longer retain what’s so coherent about those activities in the first place. Thus ---> ( ) replaces what is coherent with those activities in the first place with any of the endless devices of, what shall we say, to begin with, with tricks. I mean: Lexical representations; etymological descriptions; given a generic character; by definition; empirical by definition; and with that, as long as you introduce that into the course of the work of analyzing the phenomenon in details, the result found in this part of the procedure ( ) that you will have produced will in its guaranteed systematic way ignore what could possibly be produced and productive and done as the orderliness as represented with our ticked brackets =( )=. So the first rendering theorem proposes: Yes, indeed, there is an order in that concreteness. But this part =( )= of the method of detecting it and bringing it to some intelligibility and handle-ability and empirical witness will be systematically lost with the very methods that are used to respecify that original concreteness in established terms ---> ( ). That will not draw the admiration of your colleagues. If you have only that to say. I mean you are going to have to give them more. The book, [with its subtitle] Working Out Durkheim’s Aphorism, talks on behalf of a company of social analysts. Authors of ethnomethodological studies have been at work respecifying Durkheim’s claim according to the social science movement that begins and ends with Durkheim’s claim that the objective reality of social facts is sociology’s fundamental principle. The orderlinesses in the ticked brackets will have escaped from analysis that administers the premier formal analytic methods of the social science movement. They will have been lost by the very same methods that are administered on these materials =( )= to exhibit the orderliness respecified here ( ) in details of the phenomenon’s generic representations. The phenomenon of order can be described here ( ) exhibited in details as the ideal phenomenon that is used to describe it. Ignoring the inplace practices that are disjunctively accountable of the actual workplace and technological-specific practices that make up the phenomenon’s incessant, watched, witnessable, accountable production. To review: The first rendering theorem proposes: Indeed there are orderlinesses in that concreteness. But this part --->( ) of the methods of detecting and bringing it to intelligibility and handle-ability and empirical witness and managed care, and this just and only in any actual case, and then not as indicators or with an interpretation of indicators but as identifying orderliness, will be deliberately and carefully ignored by classical studies of social order.


Orders of Ordinary Action

Orderlinesses in the concreteness of ordinary society will be witnessable exhibits that can be examined to testify how they are systematically lost with the very same methods whose results are respecifications of order in the original concreteness, exhibiting the respecified order in and as details of methods of formal analysis and generic representational theorizing in established terms. The Embodied Reflexivity of Instructed Actions That and how the Pairs are related is described with a second piece of advice. I’ll give it a flooky name. Let me call it the embodied reflexivity turn of affairs in studying the coherence, the produced coherence of ordinary action. Reflexivity is the key term. With it ethnomethodology has taken an interest in bodies – first of all bodies, which we learned in the first place from Hubert L. Dreyfus and in a way from Merleau-Ponty. But then had to disavow and desert Merleau-Ponty because he had this marvelous love affair with philosophy. It was philosophy’s subject as embodied action that he needed to retain. The result is that he had to depend on the anecdotes and the textual accounts of others to specify what embodied reflexivity could possibly consist of. First of all, it was not going to deal, as he was so beautifully clear-headed about, it was not going to deal with simply the intentionalities of consciousness and the circular, endless, going-over-and-over-again reflections on just those affairs held before consciousness with texts, even the literatures of the gestalt experiments, and the rest, including Gurwitsch’s seminar in Paris on his flight from Germany. Something more was needed. Well, the more got to be known to be needed. It was gradually uncovered by the ethnomethodologists that the reflexivity that had to be incorporated into these inquiries had to do with things, ‘workplace and technological things’ as Lucy Suchman and her colleagues described Durkheimian things. Things of embodied reflexivity were at first unforgettably learned and afterwards known as they were described and insisted on by Dave Sudnow. Sudnow talked about embodied reflexivity almost in the fashion of a confession. He wrote about the torment, as the difficulty of learning to play and to teach the piano in its availability as a jazz instrument played with others. He wrote about it – I’ll give you a short paragraph that will give you some taste of what it was taking out of him – he writes [Sudnow, 1978]: I had come to learn, overhearing and overseeing this jazz as my instructable hands’ ways in a terrain nexus of hands and keyboard whose respective surfaces had become known as the respective surfaces of my tongue and teeth and palate are known to each other, i.e. known in the same way that this jazz music is ways of moving from place to place as singings with my fingers. To define jazz is to define any phenomenon of human action is to describe the body’s ways.

Now, never in this world can you take those words and only by explicating them, no matter at what length, find what he was talking about. You can try. You must try. It will be part of your education to come upon the defeat that the words will lend you if you are going to try to find as a respecification – as rereading the words to find in them the lucidity of what he’s talking about. You will need to get a piano.

Social Science and Ethnomethodology


And you are going to have to get an instructor. Any instructor. And you’re going to have to keep your fingers from getting injured or tired so that at the piano, with just these fingers, in just the time you’re going to have to learn to keep, over the course of bringing out from ways in which the hands will be moving through the work of placing, of leaving, of settling again, in a new place, on these keys, bringing them to simultaneously sounding a chord. Not any old combinations of tuneful soundings. In any case you’re going to be chained to that course of bringing out as the coherent matter embodiedly needed. No other thing can be replacing that. It is in fact the embodied chaining to this keyboard with these fingers doing just what Sudnow in his poetic fashion is truthfully describing. Truthfully describing means he’s teaching it. That means these texts are to be read – you won’t believe it – they are to be read as instructions. And Sudnow’s book, Ways of the Hand, is a book of instructed procedures in how one goes about learning, as a novice, to play jazz piano. It’s not the only way. But the emphasis on embodied reflexivity is, on this curious matter of being – as Sacks spoke of it in his turn ‘tied’ – to the work, being tied to just what is possessed as of equipment’s, or gear’s, or materials’ very own, in the coherent ongoing course of producing the witnessable coherence of the phenomenon. My talk is equally bad. Well, let’s take up several others. I’ve given you two. I have a collection of them that I put together for today’s presentation. I got to fifteen. Phenomenal Field Properties of =(Steering by the Stars in Polynesian Navigation)= Let me give you an idea of the third, even though in every case it’s risky to talk about phenomenal field properties, even without leaving the pages, instead of directly and immediately exhibiting them. Here’s a handout. It’s from a very favorite book of mine by David Lewis, We The Navigators, The Ancient Art of Landfinding in the Pacific [Lewis, 1972]. There are two figures [Figures 2.1 and 2.2]: here is what they are about. Very soon after you begin your graduate studies you’ll encounter repeated reference to what will be described as the phenomenal features of ordinary materials’ things. ‘Phenomenal’ properties of things is a way of alluding to the hopelessly unavoidable relevance of gestalt discoveries and demonstrations. In Literatures of experimental perception psychology there is a respecification of gestalt descriptions as discovered phenomena of experimental psychology. The Literature’s concern is centrally but not exclusively with gestalt phenomena of figural particulars, figural coherence, and figural contexture. To clarify their exposition we’ll collect these phenomena for the time being with figural coherence in its details of materials’ arrangements of their figural contexture. Now what the devil does all that consist of as worldly matters? ‘Figural coherence in its details of materials’ (that’s a possessive) arrangements of details figural contexture?’ Figures 2.1 and 2.2 are taken from a book whose author, David Lewis, was one of many authors in anthropology who were concerned with how the islands of the Pacific ever got settled. The question was: The many islands of Oceania did not have indigenous populations. They were populated. Populations from other places


Figure 2.1

Orders of Ordinary Action

Steering by the stars from Puluwat to Pikelot

Source: Lewis, 1972 Note: From a sketch by the author.

traveled over the ocean in boats of some kind, transported over very long distances of what was called ‘blue water’ – meaning out of sight of any land. Question (and controversy for anthropologists, and a great joy for them to be in valuable dispute with each other) – which was: How did they do it? How? Just how? As a proof! Various theories described how it was done. Carefully exposited, these ranged from random drifting to efforts of crews whose ignorance accompanied them through voyaging’s inevitable contingencies, calling out heroic displays of sustained fortitude as well as skills of many imagined sorts. There were also indigenous navigators, cultural historians, and engineers. Some analysts were scientists. One scientist, Edwin L. Hutchins, in a magnificent book, Cognition in the Wild [Hutchins, 1995] concluded there are some things that navigators had been doing and had to have been doing that are entirely intelligible if they are understood in light of Western navigational science, Western astronomy, and Western cognitive science. He also acknowledged there are things they had to have been doing that are not intelligible. They are intractable to explanation by Western science. A clear case that sustains the claims of Western science respecifies the skill

Social Science and Ethnomethodology


of indigenous navigators by retelling how the navigators did their jobs when they were steering by the stars. But an intractable case for Western scientists was one in which native crews and their navigators would be blown around for weeks at a time in fierce storms. The storms would end and the navigators would orient themselves, [and be] asked by their companions: Where is the home island? Where is the destination island? Navigator and crew would consult with one another and finally they would agree, ‘There’. They would take off in that direction and after a while they would come to the island. Hutchins was fair enough to admit he can’t make scientific head or tail of it. In his book he wrote he doesn’t know how they do it. David Lewis is a close friend of Hutchins. Lewis is a New Zealander; a physician, a world navigator of a [35-foot?] private boat of his own, an amateur lover of Polynesian boats and professionally a much-honored erudite student of navigation and learned

Figure 2.2

Steering by keeping constant the positions of the Southern Cross and Antares

Source: Lewis, 1972 Note: From a sketch by the author.


Orders of Ordinary Action

authority of Polynesian methods of seamanship and navigation. Figures 2.1 and 2.2 are diagrammatic descriptions by Lewis. These drawings are particularly edifying in resembling witnessed episodes. The drawings depict in particulars of boat architecture, seamanship, navigation, and star displays an episode from a night’s embodied work specifying just how steering a course by the stars looked in and as of phenomenal field properties of workplace and technology’s coherent figurational details. Lewis’s drawings render the boat in [the] drawings’ details as an oriented object. Its mast and sails exhibit directional and orientational properties of the course being steered. For clarity in exposition I collect these phenomenal field properties as ‘shopfloor constituents of voyaging’s work’: oriented objects; the proper temporal order in voyaging’s particular episodes of phenomenal field details; the specific ordinariness of details in the passing looks of things; and inner, outer, and temporal horizonal properties of things; the as-yet-unrevealed more there is to the thing than appears. What do these two drawings show? Notice that the drawings are dense with the boat’s directional and orientational properties. Lewis has been particularly careful to draw oriented object properties of the boat in the course of it being navigated during the night to accord embodied reflexively with changing star displays. We see Lewis’s diagrammatic renderings of accountable organizational social facts that the navigator and crew are producing. First of all, the navigator is seated at the back of the boat. So it’s a boat that has properties of ‘the front from the navigator’s fixed seat at the back’, ‘ahead’, ‘left behind’, ‘alongside’, ‘down’, ‘rising up’, ‘coming upon’. These are NOT, however, properties that drew Husserl’s admiration for Euclidean geometry as a natural geometry. They have incommensurably different properties – disjunctively distinctive orientable properties. [Other figures] describe witnessable stars in the sky display changing through the night. Over the night they are made to move across the rigging and across the sails of the boat in specifically unremarkable fashion and that the direction of the boat’s course is therein seen by, it is witnessable by, and it is accountable to the boat’s crew as exhibits of uniform recurrent coherent figurationally substantive details of contexture and therein in their understandable (i.e. their accountable) recognized intelligible generality. What that arcane reference to the issue of details and generality with phenomenal field properties of voyaging’s concrete episode means is this: If you look at Figure 2.1 you see stars. Here is the Pole Star. The Pole Star is held against the post. That’s a strange thing to require of a scientific navigator. But we are now talking about something that involves the produced orderlinesses in their concreteness of holding a uniform course through the night by steering the stars up the mast. In our preference for Western navigational lingo the foregoing phenomenal field properties, which when collected are one of the shopfloor constituents, are extracted from concreteness, abstracted, and reified as ‘steering by the stars’. We are not better off with the work abbreviated ethnomethodologically as ‘practical action and practical reason’. In the rest of Figures 2.1 and 2.2 we have something that, for all the fact that it is strange to the literatures of Western navigational science, is more interesting than strange. It can’t be standardized; it can’t be synchronized; it can’t be calibrated with scientific versions of holding that course; and it can’t be translated. Nevertheless, following it as an exhibited proof of effectively managing episodes of voyaging’s

Social Science and Ethnomethodology


indigenous native accountable course, made good over real-world contingencies, produces an accountable, repeatable, teachable, correctable course of seamanship and navigating actions. Its properties, the properties that we are talking about, were learned and described in a respecification of the gestalt theory of form as it was received from the great gestalt experiments and gestalt demonstrations. The respecification was done by Aron Gurwitsch. What Gurwitsch did in his magisterial book, Field of Consciousness [Gurwitsch, 1964] was to respecify the gestalt findings. We have in this respecification what he called ‘functional significations’ and their ‘contexture’ as an interpretation and a recognition of wherein the coherence consists just and openly and entirely in, as, and as of the specific bearing of any one particular detail upon the particular others. Now what would these ‘functional significations‘ look like? Well, I’ll give you an idea. [The speaker now marks two dots in a video version of a board display.]



Two. Two dots after all? Two dots, in any case? Two pencil dots? Two anything that suits me for the while since I’m showing them to make a point, so we’ll see? Two eyes staring up from the page? Two things waiting for me to draw spectacles in order that I can complete something funny about what they are about but don’t yet show you until I put those circles on them, like this [Figure 2.3]. Oh, yeah. I don’t like to draw spectacles; it spoils my art. So you don’t have to imagine the spectacles; you Figure 2.3 Contexture of two dots don’t infer the spectacles. You see the eyes, each, an eye to the left; and an eye to the right; on the page; to the left of the left hand eye is the page that extends indefinitely to the left; to the right of the right hand eye is the page that extends indefinitely to the right; an interval separates them … Each of these is witnessable as the witnessable properties whose mutual bearing displays the coherent thing – the spectacled eyes looking up from the page – that incorporates the pencilled dots as well. Well, we aren’t setting up a propositional schema. Instead we are offering a reexamination of the coherent looks of witnessable things. Our Dissatisfactions with the Gestalt Discoveries The functional significations and their accord described as ‘contexture’ were faithful to the gestalt experiments. Nevertheless, they were based on line drawings.


Orders of Ordinary Action

Gurwitsch was a mathematician as well as a phenomenologist. He had experience in the labs of the gestalt experimenters (and, I think, the clinics of Gelb and Goldstein), but he was faithful to his teachers, his mentors, his fellow students, his companions, and first of all to his teacher, Edmund Husserl. And he was part of the first incredible cohorts of Husserl’s students. Being faithful to the use of line drawings as well as to published experiments by the gestaltists and to the program of gestalt demonstrations of experimental perception psychology, Gurwitsch had hold of what were the properties of these configured lines and pictured and captioned and textually described displays. Well, for sociologists that wasn’t enough. However, the coherence of objects could be taught as and at a beginning with the use of these flooky diagrams, as I’ve just showed you. Pairs of eyes on a blackboard. With them you could ask: How shall we examine it for its demonstrable properties following Gurwitsch’s description of those properties that he called the functional significations? The big problem is: Here is detail. If this is what detail, could be is it enough? What we needed to have is not only the detail in the generality of the phenomenon found in actual lived service lines and traffic jams as well as in pictured figures. What we needed to have as sociologists and anthropologists and social analysts, who needed to be concerned with every imaginable sort of orderliness in and as of accountably produced social facts of familiar society, were the gestalt properties of social facts, Things, like the proofs of theorems in mathematics as practices making up mathematicians’ day’s ordinary work such that the workplace specific ongoing in-course of proving exhibits as the matter demonstrably proved the gestalt properties of a discovered, reasoned mathematical diagram, or a reasoned course of proving that watchably turns a proof account of a theorem into an orderly, seen, elaborated, methodically witnessed, by mathematicians, or witnessed by those learning the work of mathematics, followable proof that IS the mathematical figure that is demonstrated. But now we are talking about Eric Livingston’s studies of mathematical theorem-proving [Livingston, 1986]. We are talking about his discovery, specification, and demonstration – his proof for mathematicians – in his PhD thesis, an ethnomethodological study of work, that the lived work of proving Godel’s theorem IS Godel’s theorem. Comparably, in her ethnomethodological study of work, Stacy Burns in her PhD thesis [Burns, 1999/2000], a study of judicial mediation of large money damage disputes, reminds us of ‘the ethnomethodological import of how professionals ... competently organize their daily work activities in real time and in detail: … In ethnomethodology procedural means labor of a CERTAIN 3 incarnate methodological sort: (namely) at the workplace, progressively and developingly coming upon the phenomenon via the work in, as, and of the unmediated, immediately and directly observed phenomenal field details of producing it’ [Burns, 1999, 29]. That’s the successful ethnomethodological leap from the coherence of line drawings to the coherence of social facts. Having taken our initiatives from Gurwitsch’s great achievement, ethnomethodology’s authors are now able to talk about hybrid studies of work: Eric Livingston, Lucy Suchman, Stacy Burns, Larry Wieder, Mike Lynch, Doug 3

See earlier section for an explanation of emphasized ‘certain’.

Social Science and Ethnomethodology


Macbeth, Ken Liberman, Jay Meehan, John Heritage, Don Zimmerman, Doug Maynard, Robert Dingwall, Christian Heath, Wes Sharrock, Lorenza Mondada, Rod Watson, Jeff Coulter, Steve Hester, Peter Eglin, Dave Francis, Paul ten Have, George Psathas, Alec McHoul, John Hughes, Jean Widmer and Peter Weekes. Now, I’ve taken up your time. I was allowed to promise you could have your turn. I’m sorry – I said I would give you four relations, I only got through three. I have twelve more.

This page intentionally left blank


This page intentionally left blank

Chapter 3

The Technical Operations of the Levers of Power1 Wes Sharrock and Graham Button

Introduction Ethnomethodology is accused of a range of supposed failings, many of which are so basic and profound that they would spell the death of the ethnomethodological programme if they were correct. In general, however, these accusations are fairly wild and baseless. The absence of an appreciation of the primary role of ‘power relationships’ (the capacity of some parties in society to constrain and direct the activities of others) in social life within ethnomethodological literature is a case in point. The issue is often presented in a way which distinguishes different kinds of sociological analysts from each other, as if ethnomethodologists were unaware of and naïve about certain matters of which other kinds of sociologists are well aware. Thus the accusation made (for example, Contu and Wilmott, 2003) regarding power is that ethnomethodologists fail to appreciate the extent to which social relationships are propelled and animated through differential access to the means of power, something which is appreciated by the kinds of sociologists who are making the accusations. The logic underlying the criticism is the following. Ethnomethodologists, it is alleged, try only to look at social reality ‘through the eyes of the members of society’. No matter how successful ethnomethodologists might be at this, they will inevitably identify only those things that the members of society are aware of. Since the members of society are unaware of many things about society and about their own position in the overall society, then ethnomethodology must also be unaware of these things too. So, with respect to ‘power relations’ then, ethnomethodologists can only be (constitutionally) uncomprehending of power relationships because the members of society themselves are unaware of these. Criticisms of one kind of sociology by another kind, and certainly criticisms of ethnomethodology by other sociologists, is normally a matter of condemning someone for taking the wrong approach, an error which is easily corrected by converting to the right approach. However, the pivotal assumption here is that the observation that power is pervasive but unevenly distributed requires some professional sociological scheme to provide its availability. This is the unrecognized point that is really at issue between ethnomethodologists and other kinds of sociology. It is not, as critics of ethnomethodology pose it, that something is missing from ethnomethodological 1 Thanks to Dave Martin for some thoughtful comments on an earlier draft of this chapter.


Orders of Ordinary Action

descriptions which could be put right by seeing the light. The actual difference between ethnomethodology and the rest of sociology concerns the way in which ‘the member of society’ is conceived. From ethnomethodology’s point of view no approach is necessary or indispensable to enable one to notice that some members of the society act under the directive of others, that they are subject to requirements instituted by others, that some can – legitimately in terms of the society’s order – be commanded by selected others under appropriate circumstances and so forth. In the activities of daily life, members’ sense of what they are doing frequently incorporates an understanding, for example, that they are doing this because they have been told to, or, for other example, that they would rather be doing ‘anything’ other than what they are now doing but that circumstances make the present course of action unavoidable, or even that their present action is the end or by product of doings taking place far, far away or a long time ago. It does not take professional sociology to reveal such matters as these, people simply display in their relevant actions and interactions that they know about such things. For instance, people simply do slow down for traffic cameras when they see them, knowing that if they do not then they will be fined by the appropriate authorities. Only a sociologist denying his or her own experiences as a competent member of society could seriously assert that power relationships require a sociological eye to make them visible and are invisible to ordinary members of society; ‘everyone’ slows down for speed cameras not just sociologists. Our speed camera example might be dismissed as trivial, but the instantiations that are appealed to in sociological literature as evidence of the importance of power for sociological schemes are, of course, ones which, perhaps outside the rarefied atmosphere of professional sociological rumination where there is a curious capacity to be surprised at the unremarkable, are commonplace to all parties involved. Take, for one example, Contu and Wilmott’s (2003) concern to point out that photocopier repairmen are subject to organizational policies which direct them in the preferred method of copier repair. The observations that these things are so are, of course, ones that they have recovered from the report of the ethnographer Julian Orr (1996) that they criticize, but they fail to recognize that such observations must, presumably, be an utterly familiar fact for both those who originate the company’s repair policy and for its repair technicians (not to mention the customers with whom the repair men deal). These facts are available to the ethnographer not because of some refined intellectual skills and sophisticated sociological technical methods but merely because he has kept company with the technicians for whom they are plain facts of life. Contu and Wilmott are equally excited that organizational policies are not always fully compatible with each other and seem to think that this familiar-toalmost-every-employee fact requires us to take up some putatively technical scheme of concepts featuring (for example) ‘contradiction’. Comparably, there are those who criticize conversation analysis on the grounds that its conceptual scheme proposes a falsely egalitarian conception of conversation. However, such an attribution can surely be the consequence of only a partial reading of, for example, ‘A simplest systematics for the organization of turn-taking in conversation’ (Sacks, Schegloff and Jefferson, 1974), where the basic turntaking model is identified as context-free but context-sensitive, which specification

Technical Operations of the Levers of Power


precisely accommodates the fact that conversations can and do proceed in orderly ways and in accord with the standard turn-taking procedures amongst those who are of unequal power – aristocrats and their servants can have conversations, as can adults and children, employers and employees, etc. This fact is provided for in the model, but is not incorporated into it since the specific features of turn-taking are invariant to the model. Within sociology, ‘power’ is not, however, just another member of the discipline’s technical armoury; rather it seems to be widely regarded as occupying a special position within the array of concepts that comprise the various sociological schemes. ‘Power’ is conceived as the fundamental originator of social relations – all social relations ostensibly arise from relations of power, the social order being the product of the directives power exercises on conduct.2 However, from ethnomethodology’s point of view, power is, of course, situated in social settings and relationships. It is easy enough to talk about ‘power’ in the style of yet another abstracted sociological concept, but who is to say, independently of the examination of actual social relationships, what those relationships are, what comprises power, and what kinds of directives with what kinds of effects power might have? As almost invariably in sociological discourse, one needs to go by way of members’ understandings to endow the standard abstractions with empirical reference (as will shortly be illustrated), which is, in sociology’s analytic practice, an overwhelmingly post hoc matter. As the analysis that follows is intended to demonstrate, power is entirely understandable in terms of the examination of practical action, and the availability and efficacy of power is no more and no less subject to the vagaries of practical circumstance than any other matter. The materials we examine are taken from a video recording of a meeting between two middle managers employed by a UK high street bank. Their very designation as ‘middle managers’ displays the fact that their work positions are manifestly located in a hierarchy, that they are themselves subordinates, that they work under the eventual, rather than immediate, supervision of others, and that the power available to them will, in part, be delineated in terms of its assignment to their respective positions. As a matter of fact, one of these two is the subordinate to the other, and their meeting is the occasion for the determination of how to get their own subordinates to do the work that they are ‘supposed’ to be doing. The matter of getting the employees to do the work that ‘management’ requires of them is, it turns out, here a matter of one of the two trying to persuade the other of the necessity to undertake certain measures to give the front-line workers the incentive to intensify their work efforts. We do not know how sociologists might imagine that transactions between managers and relations between managers and workers might proceed within, to use Weber’s old but felicitous term, an imperatively coordinated association. Following our examination of the material we will offer some remarks that address this question. However, to get to that point, we will propose the following characterization of two observable features of the transaction: (1) that it attempts to resolve the difficult 2 The most prominent examples of theorists putting the ubiquity of power at the forefront of their thought are Michel Foucault (see, for example, Foucault 1977) and Anthony Giddens, who insists that power is a dimension of all actions (cf. Giddens 1984, pp. 14–16). The various sociological theories of power are discussed in Giddens and Held (1982).


Orders of Ordinary Action

problematic of what managerial measures to take to achieve the needed intensification of the work effort and (2) that it involves a blend of persuasion and coercion for the superior to encourage the subordinate to adopt a (somewhat) more confrontational approach to relations with his supervisees. Situating the Materials The managers in our data are in the position of having to implement recently formed corporate policies designed to transform the nature of first-line banking work, changing the ways in which personnel orient to customers, characterized (both within the company and in much organizational thinking) as a shift from a ‘service’ to a ‘selling’ culture. Like many other banking organizations the bank in question is attempting to transform itself in response to changes in technology, processes of globalization and the development of the consumer society. These terms are glosses for matters which are familiar enough to bank management in the form of concerns whether and how to place their services online, with respect to the kinds of services that they can interest their customers in accepting, and in positioning themselves in an intensifying and now internationally expanding competition to provide banking services. The transplantation of many of their services from the bank branch to centralized telephone call centres has been one of the ways in which they have sought to address some of these problems, and issues in the application of these policies and their implementation by front-line operatives have been previously detailed (Harper, Randall and Rouncefield, 1999). The bank in question operates regional service call centres to which customer enquiries about their accounts are directed. Of the two people involved in the materials we examine, one, Robert, is responsible for one of the centres and the other, John, is responsible for a large Eastern region, within which Robert’s centre falls. Among the changes that this and other banks have introduced as part of making the transition to a ‘selling culture’, a significant shift involves getting the bank operatives in contact with customers to seek to turn any contact with customers into an opportunity to interest them in, and desirably persuade them to take up, some of the bank’s services that they are not currently using. The operatives are meant to use these customer account enquiries to attempt to integrate a sales pitch into the transaction with the customer, and to attempt to persuade the customer at least to make an appointment to discuss the matter further. The transformation into the selling culture is known to be problematic by management, and there are difficulties in getting the operatives to pursue the policy as resolutely and effectively as the managerial plans for organizational change envisage. The operators’ performance, in terms of targets, is measured along three dimensions. The first is the appointment rate, which is the number of appointments with specialized sales people that they are able to set up; the second is their POP (Personal Overdraft Protector) rate, being the number of insurance policies they can sell; and the final one is their RMS work (filler work, such as recording changes in people’s addresses, that does not contribute directly to selling services). The company has introduced a productivity drive and hitting the performance targets is

Technical Operations of the Levers of Power


important for all concerned, including the operators, Robert and John. Reviewing the mid-year figures for the individual members of the teams, the senior of the two, John, is looking to see, in the first instance, how likely the teams are to achieve their year-end targets. A first finding he makes is that even the best performances are not particularly good, and that there are a number of individuals who are performing well below target. The method of calculation involved in making this determination is to take the annual target, now 120 per cent and to check how many of the operators are achieving, at the mid-year point, half of the designated target. There are some that are on or close to the 60 per cent level but those below represent a problem, and the first thing to be addressed is the individual cases, and an explanation of their performance is sought. Some Initial Observations A first observation we want to make is that John holds Robert ‘accountable’ for the performance figures. John provides for Robert as the one who has to give an account of the activities of various people who are under-performing, and Robert responds by giving such an account. In excerpt 1, John initiates a procedure for reviewing the individual cases, simply by announcing its commencement without allowing for the completion or recognizing any interactional implications of Robert’s utterance. Excerpt 1 77. J 78. R 79. J

is pretty close but we’ve still got one, two, three, four, five, six, seven people that err (.) quite a way below (.) the quarterly target mm (.) yeah I mean (.) the specifics errm starting from the bottom – Jo=

This procedure is thus one in which John announces the name of one of the people who is poorly performing and the identification of her by her first name, ‘Jo’, is a call for Robert to animate the performance figures of the person named by providing specifics of the case, this animation being an account of the poor performance. Excerpt 2a 79. J 80. R

starting from the bottom – Jo= =( ) Or Jo. Jo’s, this is where her term-time contract kicks in erm, she has, she did lose focus for a while but erm her term-time contract has has (.) impacted upon that as well. She’s missed (.) three weeks of the nine

Excerpt 2b 104. J 105. R

Sally (.) p(exhale) (.) errm, I can’t give you a (.) I see, I think Sally(.) Sally’s, there’s an ounce of lack of challenge in there really, I mean she’s obviously (.) she’s still


Orders of Ordinary Action

Excerpt 2c 116. J 117. R 118. J 119. R

What’s happened to Nick DUNNO, ss (.) p( ) a (.) an enigma really [( )] [ errm ] APPOINTMENTS wise he’s never been (.) you know, sparkling, but certainly from a POP’s (.) performance [there’s al]ways a danger, there’s always a

Thus Robert provides an account for several problems of poor performance: in excerpt 2a, Jo’s performance in terms of the fact she only works out of school holidays and so has missed three weeks; in 2b Sally lacks challenge; and in 2c Nick has never been sparkling. A second observation we want to make is that subsequent to Robert’s accounts, John is regularly ‘refusing’ them. His refusals of Robert’s accounts are done in a number of ways during the course of their interaction on these matters. First, as in excerpt 3a, he refuses Robert’s account by ‘toughening it up’: Excerpt 3a 81. R

82. J 83. R 84. J

Judith erm (.) Judith has has overall is is erm would say is has been a disappointment, she certainly hasn’t matched up to the standards I expected to from earlier on and you’ll see it from POP’s as well, it’s a very it’s very inconsistent, I mean her POP’s (.) is sorta hhe its erm (.) actually I’m looking at POP’s (whisper )(.) erm its (.) five, nought, five, five, nought, nought, five, nought, four, nought, so there there is a clear focus issue there erm and the same with appointments. Neil has err an action plan in place with her (cough) erm (.) I mean we’ve not come down too heavily but (.) He certainly needs to be aware that level of performance is not (.) acceptable, forty-five per cent she’s fully trained yeah now so errm

Robert spotlights a number of matters: Judith is a disappointment and hasn’t matched up to standards; there is inconsistency, there’s a focus issue, but while Neil who is the team leader in whose team Judith sits and who reports to Robert, has an action plan in place, they have not come down on her heavily. John however wants to toughen this up, giving Robert, by implication, something to do, which is to make it clear to Neil that more needs to done, stressing that Judith’s performance is way below target and that there is no excuse: she’s fully trained. Further, there is the implication that Judith is not really working as hard as she should be, for he goes on to ask (line 86) if she is ‘too pally with April?’ He may be suggesting that Robert and Neil may not be fully addressing the problem that Judith presents, which is that she is not trying hard enough. A second way in which John refuses Robert’s accounts is to ‘reject’ them: Excerpt 3b 105. R 106. J

p(exhale) (.) errm, I can’t give you a (.) I see, I think Sally(.) Sally’s, there’s an ounce of lack of challenge in there really, I mean she’s obviously (.) she’s still bloody ( ) highly paid for fifty per cent [performance]

Technical Operations of the Levers of Power 107. R

108. J


[OH yes] she is yeah, I don, I don’t dispute that errm, I mean ( ) this is her first full quarter back so (.) if we are gonna allow any leeway it would be yu this, this quarter I guess (.) she’s also got two weeks holiday between now and the end of the quarter so she’s ( ) improve ( )(pause) I think the same applies, I think she needs to be seeing those sort of things, you know, those are the performance levels except ( ) if we reduce the targets in quarter three (.) quarter three – we’re expecting to hit it

In line 106 John undercuts the account that emphasizes Sally’s lack of challenge by making the point about being highly paid for someone who is only turning in a 50 per cent rate, and goes on to further reject Robert’s account of her performance as being affected by her absence by having her know that they expect a certain performance level and she needs to hit any reduced target she’s given. A third way is by ‘undercutting the logic’ of Robert’s account: Excerpt 3c 160. R

161. J

=appointments n-n and vice a versa, but what it does mean is that erm (.) to achieve appointments and to achieve POP’s you gotta have a certain mentality, a certain focus as every call comes through your gonna be looking for the opportunity and and the argument could go that at, that ah the moment they are conscious of, you know, backlogs on on changes of addresses therefore they’re not taking the full- provisionally to ju(???) measure call times in the same period see what the call times co(nsist of) well, I think we should then if that’s the case. Because erm, I mean the reason this stuff went in there was because the number of calls and number of time they were sitting around doing nothing won it. They weren’t actively looking (.) for other things=

Robert is making the point that the rates may be going down because, as they are taking calls, their minds are on the infill work and the need to complete that, therefore they do not concentrate so much on their selling. John, however, attempts to undercut this point by reminding them why they had put in infill work in the first place, which was that in between calls the operators were not doing anything. The fourth way in which John refuses Robert’s account is by ‘upping the ante’: Excerpt 3d 167. R

168. J

That’s only a a a ya know u uh ay part of the equation erm (.) because the actual call numbers aren’t (.) vary hugely (.) probably you know they jump around a little bit and probably on average over the twelve-month comparison then I say the call figure is definitely down but I still think we gotta up the ante, I mean ca-coming back to wha we talked about this morning wi alla this and the question about will we actually deliver the required productivity increase – no, not on its own. We gotta keep pushin’.

John refuses Robert’s account done in terms of averages in making the point that more needs to be done, the operators need to be pushed more. So we have a couple of observations on some of the interactions that make up the episode we have been looking at, which is that there is a cycle of activities, in each of which Robert provides an account that justifies an operator’s performance, following a call for action from John, and then John typically questions Robert’s justifications.


Orders of Ordinary Action

Towards a Deeper Analysis We want to try to deepen our understanding of the way in which the account and the refusal are working, so let us return to the account-giving by Robert and make a third observation, which is that Robert’s accounts are done in an explanatory mode. There are a number of features to the way in which he does this. First, he locates the reasons for the under-performance in the particular circumstances of the nominated individual. Excerpt 4a 126. R 127. J 128. R

[JU JUDITH] just going back to Judith she has obviously had her (.) she has (.) got er heal-f shshe’s she’s a doctor’s For what erm (.) I think its erm (.) oh fur erm some some sosort of bowel disorder iritat-irritated bowel (.) has got stomach problems.

Robert returns to a discussion of Judith’s performance by elaborating his account – Judith has a medical condition for which she is seeing a doctor. This is offered as providing a possible reason for her performance in this personal circumstance. Second, Robert is able, as in this case, to do so in an excusing way. That is, somehow the fact that Judith has an irritable bowel and is under the doctor is to be taken into account in looking at her performance levels. The third way in which his accounts are done in an explanatory mode is that Robert produces his accounts in such a way that says or implies that the problem is in hand, in as much as he is taking or has initiated managerial action. Excerpt 4b 86. J 87. R

Is she too pally with April Yeah we’ve, yeah, Neil Neil’s tried to split them up (.) which hasn’t (…) (.) worked (.) fully, though she did a lot of the infill (…) work now, but when her RMS (….) hundred and seven (.) that call figure’s up (….)

Thus Robert is showing that they have already addressed the point that John is driving at, which is that Judith and April distract each other, because Neil has tried to separate them. Fourth, Robert produces his account in an explanatory mode by raising difficulties in doing anything more. Excerpt 4c 175. R

I mean -I-I-I think it would be fair to say that erm a great deal of-of-of my focus and therefore Neil’s focus, talk about the individuals we talked about ( ), has been on RMS productivity, therefore I mean – I would agree we are cranking up, we are doing that but we’re doing it in terms of RMS productivity as opposed to appointments, and you know I’m not, I’m not saying you can’t have have both, I – I just don’t think you gonna have notable success in both, and I think when we look at RMS (.) you-we would accept we have had some notable success with getting (.) erm, you know, our RE’s ( ) increased and with the individual productivity

Technical Operations of the Levers of Power


So, Robert and Neil have been targeting RMS productivity. If they target something else as well it is not likely to result in success. The fifth point is that Robert places a problematic performance in relationship to good performance in other areas: Excerpt 4d 87. R

(.) which hasn’t (…) (.) worked (.) fully, though she did a lot of the infill (…) work now, but when her RMS (….) hundred and seven (.) that call figures up (….)

Judith’s poor performance is acknowledged by Robert but the fact she has been doing a lot of the ‘fill in’ work and that her call figures are up are invoked. In summary, these features of Robert’s responses to John amount to doing the explanations in an extenuating manner. Robert then is not just producing accounts but is constructing them in an explanatory and extenuating mode. If we turn to John’s refusals we can likewise try to deepen our understanding of these. In this respect we can make a fourth observation which will come in two parts. First, John speaks diagnostically, looking not at the problems of managing the specific cases, with their individual troubles but at having to lift the overall performance of the whole team, and projecting one possible explanation of the problem, which is that the operators are not being treated punitively enough – for, as in excerpt 5a, Christine got no pay rise, and she’s now doing the best. Excerpt 5a 180. J

Are they aware that (.) you know this is a half-year stage and they really need to be doing to make it (.) we got Christine up at the top and the reason she’s there is because she didn’t really get a pay – well she didn’t get a pay rise – yeah , now that’s one of the things we said to her. So where does that leave these

Excerpt 5b 148. J

=I THINK they need to be making the point we’re in the half-year review (.) you know they really ought to be making that extra effort in June to make sure that the (.) ones that are in the sixties get it, because they got the – they’re not far off (.) and you’ll all be having discussions with them now, just up the ante a bit

Second, John also speaks critically. Excerpt 5c 161. J

well, I think we should then if that’s the case. Because erm, I mean the reason this stuff went in there was because the number of calls and number of time they were sitting around doing nothing won it. They weren’t actively looking (.) for other things=

The issue for John seems to be not so much one of inhibiting circumstances, but of not trying hard enough – the suggestion is that the operators are lazy or lack initiative, that they have to be pursued to get them to work more than the minimum.


Orders of Ordinary Action

He’s reminding Robert that the reason for them developing the infill work was that, previously, instead of the operators looking for things they could do when there were no calls for them to answer they were just doing nothing. And, if that is the problem, that they have to be pursued to work, then the solution to the problem is obvious enough: get on their case. John, then, is not just refusing Robert’s accounts done as explanations and attenuations, he is doing so by analysing the problematic and criticizing those involved. Contrasting Perspectives In the analysis so far, we have moved from an initial set of observations concerned with the production of an account and its refusal to a deeper analysis of the production of an explanatory and extenuating account of an individual’s performance and its refusal in a diagnostic and critical analysis of the individual’s performance. We now want to develop this analysis further, because in providing explanatory and extenuating accounts and in providing diagnostic and critical refusals, Robert and John are displaying an orientation to two different methods of understanding the problem and tackling it. It seems to us that the two managers are trying to work out what the problem is, and if there is a solution to be found to it. John is certainly occupied in proposing measures for difficult individual cases, but we understand what he is doing is addressing not merely the individual difficulties, but also addressing Robert’s will to take the measures that John sees as necessary. Thus, we have the overall situation of John proposing that Robert adopt a more punitive management approach, with Robert being somewhat reluctant to do so. What Robert says suggests a preference for dealing with subordinates by recognizing and sympathizing with their personal problems, adopting disciplinary measures only when really called for, and preferring the one that works out the problem with the operator, and generally seeing the point of view of those he supervises, as well as seeing the difficulties that stand in the way of himself and Neil managerially delivering improved results. This contrast between their approaches is exemplified in the following excerpt: Excerpt 6 154. J 155. R 156. J 157. J 158. R

don’t want it seen as an excuse to not [??????? ] [no, no it’s not] but but it’s (.) you know If that had the, if we come back to RMS figures, if we had everyone up to hundred and twenty per cent (.) I’ll (.) okay but you bring a that line across – yeah most are over (pause) It shouldn’t be an excuse but-but it’s no it’s not but you’d accept it’s a (.) I mean it is a distraction, yu-you can’t, th-the two don’t tie in erm (.) tie in together like that, you can’t say well, you know, by by decreasing your (.) erm (.) decreasing the amount of of processing you’re doing you’re automatically increase the erm=

Thus, John is taking the view that having to do the infill work is their, the operators’, own fault and that it is not to be used as an excuse. In response, Robert disclaims allowing it excuse status, but offers an explanation for it in terms of the difficulties

Technical Operations of the Levers of Power


that trying to combine the phone-answering and other ‘infill work’ creates for them. We see a generalized character to John’s responses to Robert’s accounts, which is that Robert is not doing enough, or not doing the right things, to solve the problem, and that he needs to adopt a different frame of mind, a different method, so John is refusing – on specific grounds – Robert’s proposal as to either the nature of the problem or with respect to the adequacy of actions already in place. The consistent theme of his refusals is that Robert needs to adopt a more pro-active, confrontational style of management. John indicates several ways in which this should be done. The first is that Robert needs to ‘up the ante’: Excerpt 7a 168. J

I still think we gotta up the ante …

The second is that he needs to get things said that have not been said: Excerpt 7b 183. J . 185. J . 195. J

[And I think] I would prefer to have these sort of discussions now rather than a shock at the end of the year but we need to be telling them now

The third is that Robert needs to make some threats that the conditions of work will diminish: Excerpt 7c 94. J

This is something that could be happening if you are consistently underachieving on that

The fourth is that there will be no more pay rises: Excerpt 7d 182. J

[we … Paying ] you know if we were looking at pay review now – what sort of pay review would you expect (.) when you’re not delivering on performance (pause)

So, for example, contrast Robert’s and John’s differences over the seller’s pack in excerpt 7e. The seller’s pack is an information pack given to the operators. In it are some warnings of what can happen if sellers do not perform to target. In excerpt 7e, contrast Robert’s having circulated the seller’s pack with John’s desire to have it drawn to their attention: Excerpt 7e 96. J

I’m not saying we are gonna bring it in yet but I don’t see why she shouldn’t see it to say these are the sorts of things that are available

44 97. R 98. J 99. R 100. J

Orders of Ordinary Action mm (pause) [????] ] [ I mean I did circulate] those ah yes, I think that if Neil shows it to her specifically, in the one-to-ones, look we talk about this, this is the sort of thing in the network – how did you, do you react to this game (????) anyway

Robert therefore needs to stop thinking that he has got things in hand and is doing all he can, and to start thinking that he needs to do something and right away. Thus, John emphasizes, in 7f, line 152, that things are getting worse, not better: Excerpt 7f 152. J

There’s no one that’s broke the target completely this time is there. Whereas last quarter we did have several that were way ov=

And in the following excerpts (7g, 7h, 7i) John emphasizes that there is a timeliness for moves to be made. Even those who are closest to the targets may nonetheless need to be pushed: Excerpt 7g 148. J

=I THINK they need to be making the point we’re in the half-year review (.) you know they really ought to be making that extra effort in June to make sure that the (.) ones that are in the sixties get it, because they got the – they’re not far off (.) and you’ll all be having discussions with them now …

Excerpt 7h 180. J

Are they aware that (.) you know this is a half-year stage and they really need to be doing to make it (.) …

Excerpt 7i 181. R 182. J 183. J 184. R 185. J

well a???[???? conversations] [we … Paying ] you know if we were looking at pay review now – what sort of pay review would you expect (.) when you’re not delivering on performance (pause) [And I think] I would prefer to have these sort of discussions now rather than a [??????????] shock at the end of the year

Also while matters may involve putting the pressure on the operators and team leaders, the failure to make the targets implicates John and Robert, for it is, after all, their job to ensure that the operators do make the targets. In sum, for John, the time for waiting to see if they are going to make the targets, whether policies already in place are going to turn out or whether circumstances will improve, is past and things need to be set in motion now. John’s method for viewing the problem thus seems to see the problem as more generalized (that is, as a problem that almost all the operators are underachieving, whatever the variation

Technical Operations of the Levers of Power


in their individual circumstances) than Robert does. Robert’s method is to treat the problem as individual ones. The difficulties that Robert recites – the medical problem of irritable bowel syndrome, having to integrate work with raising the kids, coming back to work and getting up to speed again – clearly exist, but John does not regard these as wholly relevant and perhaps sees things more in terms of a problem with Robert’s style of management. The key issue, for him, is that of introducing a more disciplinary approach to the management of the teams. Robert’s method is more one of attempting to support and sympathize with individual operators by seeing how falling behind targets arises from their specific circumstances; recognizing the limited capacity of managers to control for such personal problems; focusing on the practicalities of managing for success in different aspects of performance; introducing ‘minimal’ disciplinary steps to resolve problems; and compensating for poor aspects of performance by noting more positive features. Thus, Robert can be heard as explaining that he understands what the problem is in a particular case but that he can do nothing about it; that they are already attempting to deal with the problem; that there is a problem but that it is perhaps not so bad as it looks; that they can only deal with one problem at a time, and it is unrealistic to expect more; that Robert (and Neil) are already doing all managerially that they can. However, we understand John’s responses to Robert’s explanations as indicating that, for him, the immediate problem is to get Robert to adopt the necessary policies to intensify the rate of work. John’s diagnosis is in part that Robert and Neil are not effectively – in the sense of ‘forcefully’ – explaining the situation to the individuals in their teams; that they are not managing all the elements that need to be controlled to get the results; that they are not using all the punitive powers that are available to them; and that Robert appears reluctant to take the steps that John deems necessary. Robert and Neil are not appropriately prioritizing the need to meet the targets, regardless. John does not recognize the problems that Robert proposes as relevant; his method is to focus entirely on the fact of under-performance across the board. John’s formulations are meant to highlight that Robert’s method of dealing with the problem is inadequate, ineffective and inappropriate. In part, Robert’s method is to present the problem as one that needs to be tackled in parts, and that needs to be dealt with by the routine supervisory and team management methods (putting them on a plan), but John rejects this picture of the situation. John’s method appears to be that the thing has to be looked at in terms of a whole, in terms of a pattern of performance. Presumably Robert’s idea is that you make progress in solving these problems, that there is quarter-by-quarter improvement. But this is not what is happening – and in some ways this quarter is worse than the one before. For John, there is a need to appreciate how big the problem is; that a significant number of people, seven, are well below target, and that those who are doing better are only close to, rather than exceeding the target; that the time to ensure that the target will be achieved is running out, and for these reasons measures need to be taken now. Also John diagnoses a different problem with respect to the individuals, which is that the individuals are not necessarily trying to achieve the target, and are using difficulties as excuses. Thus, the introduction of infill work provides Robert with an explication for uneven performance, it is not only extra work, but a distraction


Orders of Ordinary Action

from the task in hand, whilst for John it is effectively the team members’ own fault. It needs to be made clear to the team members how serious the problem is, that they are already at the six-month marker and are going to find it difficult to achieve the target – the very best they can hope to do is achieve their target. The consequences of these failings have not been made plain either, that the parties risk either deterioration in work conditions or in pay is the least that will follow. Getting the team leaders to pursue the objectives more single-mindedly – to ‘keep pushing’ – is not simply one more management problem among others, but the central issue: Excerpt 7k 170. J

On this and the team leaders of got to keep pushing to make sure that they deliver not just this but also on the productivity and it is there (??) to keep cranking (???) the handle with it (….)

Of course, it is not merely that the team members need to be brought to realize this, but that Robert and John are also implicated in the situation, that the failure will be their failure. Instruction as a Lever of Organizational Power We have moved from an initial description (that what Robert and John are doing is to, respectively, provide an account and to refuse it) to a more elaborate one (that they are, respectively, providing an explanatory and excusing account and refusing it analytically and critically), to a third, more developed one where we can see operating two different and contrasting methods for viewing the problem and fixing it. These are: (1) the problem lies in the individual cases that need to be fixed by working with the particular individuals on their problems, and (2) the problem is a general one that the teams are under-performing, which needs to be fixed by a general solution which is to put more pressure on everyone and up the general ante. The situation we are considering is a familiar organizational one, that many studies of workplaces attest to, in which the participants are involved in making the organization’s processes and policies work, in finding ways to implement the policies that are incumbent upon them, and which involve them in dealing with materials and issues that are not specifically covered by those policies. Though the achievement of the targets is something prioritized by the organization and already in force for six months, it is clear that whatever the procedures have been they have not been successful, and the problem now is to find what resources are available to extract increased effort from the operatives. From John’s perspective, it is not that the targets are intrinsically too difficult for operators since they are being met by some, though the performance of those who are doing best is not going to be enough, at present rates, to compensate for the under-performance of others and achieve the target collectively. However, there is (as Robert raises) the possibility that the opportunities for success are limited, that the capacity to turn calls into appointments and POPs depends upon the number of calls coming in. The ‘selling’ task is not one that is intrinsically

Technical Operations of the Levers of Power


satisfying to those doing it, nor one that comes easily – the suggestion is that it is something that requires working oneself up to, and sustaining commitment to it is something that is variable. Also, it is perhaps the case that enforcing the prioritization of selling is something that is less than attractive amongst the things that the managers have to do. Therefore chasing up the performance figures, which should perhaps be first amongst the things that Robert and the team leaders have to do, is something that is not consistently done. There is also an issue of ‘compatibility’ amongst the kinds of work that the operators are being called upon to do – the infill work does not necessarily fit with the mental attitude required for the phone selling work, or alternatively doing that can become a substitute for doing the selling – but that is not (as far is John is concerned) a bona fide alternation. Thus, it is clear to John that the operators can do what is needed, and that they are not therefore making the necessary effort. John orients to a feature of the hierarchical make-up of organizations such as this one, which is that although someone may be responsible for some policy or target, they are nevertheless remote from the actual situation in which the policy or target is to bite. Thus, although he is responsible for the achievement of these targets, he is not personally going to be involved with the team members. His method for dealing with the problem is not one where he will be directly intervening with the staff themselves; rather his actions show that he recognizes that he has to follow the organizational channels and proceed via Robert and via the team leaders. He is, then, here operating by ‘remote control’, one step removed from the day-in-day-out work of the operators whose work has resulted in the undershooting of the targets which are his responsibility to bring in. His problem in effecting any change in the work of the operators is, then, that of getting a manifestly less-than-enthusiastic Robert to get the team leaders to do what needs doing, which is just the thing that Robert is reluctant to do, which is to apply more pressure. We also understand John’s proposals as having an instructive character to them, in that these are things that Robert is being effectively told to do. The pressure is being applied, in the first instance, to Robert. But they have another sense of ‘instructive’ in as much as John is showing Robert how to increase the pressure on his subordinates. He is showing Robert how to read the performance data not in terms of, for example, personal details about the operators, such as their troubles with their bowels, but in terms of what it means for the overall target if they are only achieving this percentage at the mid-point. It is a feature of how John is instructing Robert that he seems to be identifying as measures to resolve the problem ones that have a reinforcing character, ways of highlighting things that the operatives themselves might be assumed to be aware of already. The operators can be assumed to know that if they do not reach the targets they may well find that their work conditions are made worse and that they will be given no pay rise. Thus John’s proposals, rather than presenting such matters as revelations, seem to feature ways of bringing these things already known to the operators home to them individually – introducing these topics into the one-to-one sessions and so on. In other words, they must be talked to in the same mode, analytically and critically, that John is currently using to address Robert.


Orders of Ordinary Action

Robert, then, is being instructed in both of these senses to see that the adoption of a somewhat more confrontational style of management is what is needed. He is being invited not only to adopt this style but being instructed in how to pull it off. These instructions consist of the following: • •

recognizing that the problem here is overall performance, and that effecting success with one dimension of it does not solve the problem; that the difficulties which Robert regards as ‘good reasons’ for substandard operator performance are not ‘good reasons’ from the point of view of the problem in hand; that Robert is being unduly considerate in his understanding of operator motives, and that the problem may be not that they have this or that difficulty but that they are not (collectively) trying hard enough; that therefore, Robert himself and his team leaders have not been trying hard enough; they have been satisfied with routine, first-line solutions, such as putting people on a plan, where this, for John, appears to be an insufficient solution; that Robert is waiting for situations and means to work themselves out without applying consistent pressure.

All told, then, Robert himself needs to try harder, and to adopt such means as there are – minor intimidations – in a more consistent and pressurizing way. The hierarchical character of this encounter, then, resides in many parts in the work that John is doing to instruct Robert in how to see and understand the problem and how to address it. It is John’s work to effect a change in the operators through Robert, and he is doing that work in a series of instructional episodes. In part, this displays John’s technical grasp of the organization. In contrast to Robert whose method is to individualize the productivity problems, to account for these by excusing the individuals in terms of their personal circumstances, John generalizes it as a matter of putting pressure on across the board. A consequence of Robert’s method is that to effect improvement each individual’s problem would need to be addressed, whereas all of the operators, even those who are presently ‘doing well’, need to raise their performance to make their collective target and need to have the negative consequences of doing so made plain to them. Casting the problem as a general one of under-performance provides for the applicability of organizational levers being pulled. And in that respect John displays his technical grasp on the ways of the organization and is instructing Robert in those ways. Robert is John’s servomechanism for bringing about change and he is being ‘tooled up’ so that he in turn can work to tune up the operatives. Conclusion Sociological definitions of power typically involve the idea that power is the exercise of constraints on people’s action, often associated with the relative positions people occupy within a social structure with regard to money, position, status, a

Technical Operations of the Levers of Power


deity, authority and suchlike matters. With respect both to the participants in the episodes of interaction we have examined and to the people they are discussing, such definitions, and indeed people’s vulgar responses to hearing the tapes and reading the transcripts might lead us to question why the managers do not simply threaten their employees with dismissal if they do not comply with their requests. After all, they have the power to dismiss them so why not threaten them with this consequence if they do not change. Such a naïve position does not, however, take account of the fact that all social life is conducted within the circumstances of its production, including the exercise of power, and it is the exercise of power to which we have been witnesses. John is exercising his power over both John and their subordinates; he is educating Robert in how he needs to operate and he is influencing the employees’ actions through Robert, and in doing this he displays that he is taking account of certain circumstances. It is not that Robert does not understand that more rigorous enforcement of targets can be attempted, but that he apparently does not prefer this kind of management approach and, from John’s point of view, needs to realize that he really has no choice in the matter – unless an instant improvement is effected, there will be no catching up in pursuit of the targets. In questions at the conference where this paper was initially given, it was asked why the managers did not simply fire the operators. The managers are, however, themselves constrained in what they can do, and they cannot simply and instantly dismiss people – getting rid of staff requires giving them a period of notice to serve, and also requires a prior procedure to establish grounds for dismissal which, itself, can take some time. They cannot instantly dismiss under-performing staff and immediately replace them with well-performing ones. The under-performing ones must continue to work out their notice, and may be less, rather than more motivated to do so, given they are leaving the job. Hiring new staff itself takes time; there is no assurance that, in the long term, the new staff will perform any better than those they replace. Even if they were to, they would have to be trained in the work and have to ‘get up to speed’ with real operating practices, during which period the operation would be falling further behind target. A large part of the one year within which this could be used, as an organizational imperative, has already passed and replacing the existing operators is not a practical option. Thus the circumstances in which John might exercise his power, deriving from his position within the organization relative to Robert and the employees, condition the way in which he exercises this power, and in this respect power is just another circumstantial accomplishment of action and interaction. From ethnomethodology’s point of view, there are only ordinary phenomena, and there is no reason to rate any one of these over any other, and, of course, there is no reason to incorporate into one’s sociology the lay ranking of phenomena as ‘important’. It is not that power relations are excluded from possible consideration but that they comprise only one possible topic amongst a multitude. In respect of the analytic problems they may present, it cannot be assumed that these are in any way distinct from those presented by any of these other numerous topics. Certainly, examining power relations is not of itself a notable occurrence for ethnomethodology but this is not because ethnomethodology ignores it. It is simply an incontestable fact that the existence of power differentials


Orders of Ordinary Action

and imperative coordination of conduct are ordinary, obvious and matter of course features of everyday life in the society and as such familiar and commonplace to members of society. The point at issue with sociology is not whether people in society relate to one another in terms of (among other things) power, hierarchy and the like. The issue has always been from an ethnomethodological point of view the status of claims about these matters. Theoretically minded sociologists suppose that the recognition of the phenomena can only result from the creation of a set of purpose-built professional sociological concepts. Ethnomethodology has maintained, to the contrary, that these concepts are not necessary for the recognition of phenomena that are in fact known by means of socially given ‘commonsense understandings of social structures’. Since for the latter the project is to understand how commonsense understandings of social structures figure in the organization of routine everyday conduct there is no need for ethnomethodology to adopt the lay ranking of the importance of topics in a way that professional sociology characteristically does. The demonstration we have given here, which involves an instance of activity that concerns what is by anyone’s standards a use of power, can be considered in ethnomethodological terms without need for the innovation of any pre-emptive sociological schema. Ethnomethodology’s legitimate interest is in the self-ordering characteristics of social affairs whatever they may be. From this point of view any occasion of social action is as amenable to and worthy of analysis regardless of any supposedly intrinsic importance it may have. This chapter’s topic is, thus, just another topic on a long list of ethnomethodological studies. Engagement with it arises only from our own interest in studies of work and the way in which they are ordered and coordinated and not because it is the prime concept in a sociological schema. Despite the varying definitional differences and understandings that sociologists have of power, there is unanimous agreement as to how power is to be articulated, which is that it is something to be used to account for occurrences in society such as social institutions, tyranny, resistance, exploitation, marginalization, dispossession, lost souls and the like. To ethnomethodology, however, such an approach to power relations presupposes rather than understands the ways in which people actually deploy power and, if it is a means, how it is effected in pursuing the end. Classic definitions of power as involving the means to coerce people to do what they would not otherwise do might sound reasonable enough. Whether or not they are, however, is beside the point because the definition cannot give an indication as to how in any instance one person can get others to do what is wanted of them even over their opposition. This question is a starting point for an ethnomethodological encounter with power, not an explanation of action. What power relationships amount to will be displayed in the particular circumstances of ‘this interaction’ between ‘these people’ for ‘whatever purposes’, and done ‘somehow’, which is what we have been attempting to demonstrate.

Chapter 4

Operating Together through Videoconference: Members’ Procedures for Accomplishing a Common Space of Action Lorenza Mondada

Introduction Medical work, like other professional domains, is increasingly a matter of collaboration between various professionals involved in performing complex and risky tasks. This is particularly the case for surgery, where operations are performed by a team composed of the chief surgeon, his or her assistants, the nurses, the anaesthetists and other possible co-workers. This traditional teamwork has been changed and enlarged by telesurgery projects which allow other colleagues and experts to participate in the operation itself (and not only in its preparation or in the examination of patients’ records for therapeutical advice). Telemedicine projects are therefore creating wider networks, thanks to various technologies, in order to improve collaborative work among surgical teams. In this sense, they become an perspicuous setting for exploring issues tackled by workplace studies (Button, 1993; Luff, Hindmarsh and Heath, 2000; Heath and Luff, 2000) and more generally the ethnomethodological studies of work (Garfinkel, 1986) and talk-at-work studies (Drew and Heritage, 1992), which are interested in the situated, fine-tuned and detailed organization of members’ practices using technologies to accomplish their ordinary work in a coordinated and collaborative manner. Focusing on a particular work situation in surgery, this chapter addresses problems such as the coordination of action in professional settings, the conditions for the collaboration between various professionals, the in situ order of teamwork, the uses of technology mediating the communication between a professional and an expert giving him advice, and the uses of an actual professional situation for teaching purposes. I am therefore interested in the active configuration, in the social accomplishment of the work space and of work relationships within an ongoing collective action. I am interested in the way in which participants organize, shape, adjust the space where they are performing their activities and orient themselves toward its visibility, as well as in the way in which they exploit particular technologies – here the videoconference device – for communicating and for acting together. The point is that this space is not predetermined by the technology, but rather that


Orders of Ordinary Action

technology is a resource available to the participants for shaping the relevant space for their ongoing work. In order to explore those issues, I will focus on two aspects: firstly, I will observe how the images transmitted by videoconference are being both produced and oriented to by the participants, thus reflexively accomplishing a collective visual space for the ongoing action; and secondly, I will observe how a participation space is shaped by the way in which participants organize their talk-in-interaction. I will analyse some resources that shape various spaces of action where different participation modes are made possible and where different membership categories are made relevant also thanks to their specific sequential organization (Watson, 1997). These two aspects contribute to the making of a possible space of collective action. The analysis will focus on the specificities of surgical work – object of a few sociological (see, among others, Hirschauer, 1991; Aanestad and Hanseth, 2000; Fox, 1992; Peneff, 1997) studies, but as yet unexplored by ethnomethodological and conversational studies of work (but see Pilnick and Hindmarsh, 1999, on anaesthetists’ teamwork or Heath, Luff and Svensson, 2002; Koshman et al. 2007, on surgical operations) – as well as on features of teamwork and collaborative activities. The Making of a Visual Space of Action The Video Image as a Constitutive Dimension of the Ongoing Activities The video images which constitute the basis of my analysis were not recorded by myself. This means that they are not the product of a reflection about the best way in which a camera can capture the ongoing activity and its ordered features. Instead, images were produced by the participants and recorded by the videoconferencing system used by them – that is, they were produced by a recording device which was not designed for research purposes (cf. the non-fiction films analysed by Macbeth, 1999; cf. also the kind of records produced by surveillance systems, Ball, 2000). This fact has some fundamental consequences to be explored empirically by the following analysis. First of all, the ways in which video recordings are made reflexively produce the data of the analysis, shape them, give them a particular orderliness and meaning (Mondada, 2006). It is therefore of fundamental importance to integrate into the analysis the practical ways in which recordings are produced, with their local contingencies and for all practical purposes. Secondly, these aspects are neither marginal (so that we could possibly ignore them) nor problematic (so that they would ‘distort’ the phenomena at hand) with regard to the use of the resulting materials for analytic purposes. On the contrary, the very ways of producing these images give us central insights about the organizational features of the recorded practices themselves, revealing their local order and intelligibility as reflexively produced by their display to and for the camera. In other words, the production of the video record is deeply tied to the production of the action; in the case at hand, it is the very condition which allows the surgical procedure to be performed as seen and defined by the participants. This relates to what members are actually doing in this particular setting (Mondada, 2004): first, a team of surgeons is operating on a

Operating Together through Videoconference


patient; second, an expert is collaborating with the team and giving advice to the surgeon; and third, both parties are commenting on the operation for an audience of medical professionals. This means that we have here various activities, and possibly various membership categorization devices, or even standardized pairs (such as ‘colleague’/‘colleague’, ‘expert’ or ‘counsellor’/‘client’, ‘teacher’/‘learner’ ...) (cf. Sacks, 1972). The video image is essential to these activities: •

Laparoscopic surgery (also called ‘keyhole surgery’) is a minimally invasive technique which consists of placing the instruments and an optic system within the patient’s body through small incisions and by means of trocars in which the instruments are inserted. The surgeon operates while looking at a TV monitor where the endoscopic image is made available. Remote expert online advice is made possible by the accessibility of the surgical theatre through videoconference (transmitting both an external and an endoscopic view). Remote learning is also made possible, thanks to the transmission of the image to an audience of trainees in an auditorium (Mondada, 2002a).

Thus the video image is the major link between the various participants and is a central technological tool for accomplishing their activities. Not only do participants orient towards the camera and reflexively organize their conduct in relation to it (as was nicely demonstrated in detailed analyses for other medical settings by Heath, 1986, and by Lomax and Casey, 1998) but, moreover, the video shooting is constitutive of the action they perform (Mondada, 2003). The recording of the image is done with the same videoconferencing device which transmits and produces it, thus integrating and somehow naturalizing the recording itself. A first video excerpt will allow me to explain a few analytic points before I will synthetize some observations. The excerpt is taken from an initial sequence of the operation: the surgeon has already performed the little incisions on the body surface in order to introduce his trocars and he is now introducing the instruments within the trocars. These activities are part of the preparation for the operation itself (transcript conventions are listed at the end of the chapter). Excerpt 1 (1106/K1/DV1/14'/210 V1) 1 AC 2 3 4 5 6 7 8

okay/ . the next step is/ the the instrumentation/ . we put a a retractor of the liver/ .. in the subcostal area .. an atraumatic grasping forceps for the assistant . for michelle/ . on the left side of the patient/ .. another grasping forceps here . for my left hand/ ... and a coagulating hook for my right hand\ . SO/ my head/ ((whistles)) my head\ the optical system/ the pathology/ and the . screen . is . the: different point of the same line\ . and my left/ my left hand and my right hand . is on both sides of the optical system\

This excerpt shows the constitutive role of the optical and the video system within the practical organization of the surgical work. The optical system is a central piece


Orders of Ordinary Action

of the instrumentation in the operation. The optics producing the endoscopic image and the monitor on which it is projected, the surgeon’s body and the pathology are explicitly enumerated (line 6) as different aligned elements of the surgical action. Moreover, the external video camera producing this image is also a central element, since the surgeon’s gestures are clearly oriented towards it and are organized in order to fit the visual field requirements. This orientation toward the camera is manifested in two instances which I will analyse, focusing on the way in which this phenomenon is made available in and by the video record. (In most of the accompanying figures, the external view shows the belly of the patient; at the top right an inset shows the endoscopic view produced by the optics.) This is the very beginning of the operation. When the surgeon, AC, begins to describe the various instruments at hand and shows first the liver retractor, his gesture is not linear, but follows a complex trajectory which can be represented in the following way (the various gestures being indicated by letters): Excerpt 1 (lines 1–2) 1 2

okay/ . the next step is/ the * the instrumentation/* . we pu*t a a *retractor *-------A--------------*---B---- *---C--*---D---of the liver/* .. *in the subcostal area*.. an atraumatic grasping forceps -------------*-E--*--------F------------*

Beginning his description, AC takes the liver retractor and moves it toward the corresponding trocar, on the top of the relevant anatomical space. He goes so far as to reach the trocar (A) (Figure 4.1), as if he would introduce the instrument into it. This gesture is performed before AC mentions the name of the instrument (line 1); it is performed Figure 4.1 Surgeon’s gesture (A) anticipates before the explanation, explanation as a gesture related to the routine of the operation more than to its demonstration. We can therefore say that when this gesture is interrupted, a reorientation of the action is made visible, refocusing from the operation to the demonstration, in a kind of ‘gesture repair’. The instrument is moved away from its trocar, and comes back to the centre of the screen (B). The instrument then goes from the left hand to the right hand (C), in order to be shown to the camera. The right hand can show how the retractor is deployed (D) (Figure 4.2) and it is only after this demonstration that the retractor is

Operating Together through Videoconference


taken again by the left hand and moved towards the trocar (E), being introduced in it (F) (Figure 4.3). If the first gesture (A) pertains to the operation and anticipates the didactical description, the following gestures (B–F) pertain to the demonstration and are synchronized with the explanation. This demonstration of the liver retractor is the Figure 4.2 Surgeon’s demonstration of retractor first gesture in a series (gesture D) is synchronized with which inaugurates the explanation format of the subsequent gestures of showing the instruments one by one (using the clear skin of the belly as a ground on which the instruments can be evidenced as figures) before inserting them in the relevant trocars. Some seconds later, the orientation of the surgeon towards the external camera, at his back, is made manifest by his way of producing the list of the relevant Figure 4.3 Introduction of retractor to trocar elements involved in (gesture F) is synchronized with the procedure and by explanation his way of accomplishing the distribution of these elements in the same space (‘my head\ the optical system/ the pathology/ and the . screen . is . the: different point of the same line\’, lines 6–7) – a space which does not preexist the demonstration but is created by it, a space which is not only the syntactic linear space of the list, but also a linear geometrical axis which is successively performed by his gestures. This axis is accomplished by taking into account the position of the camera, and by displaying the gestures in the field created by the camera. We can try to represent this fragment of action in the following transcript:


Orders of Ordinary Action

Excerpt 1 (lines 5–7) AC


SO/ my head/ ((whistles)) ((moves his head forward and enters the visual field of the camera)) ((laughs)) my head\ the optical sy|stem/ the patholo|gy/ and the . screen| |Fig 4.4 |Fig 4.5 |Fig 4.6 . is . the: different point of the same line\ . and my left/ my left

Every element of the list is accompanied by a gesture pointing to the element, not with the index finger but with the whole length of the hand. In this way, the geometrical space of the ‘point of the same line’ is progressively designed by the position of the head and of the hands (Figures 4.4, 4.5, 4.6). This axis is traced between elements that are inscribed Figure 4.4 Surgeon starts gesture, using head to in a space which is much confirm use of optical system for the larger than the camera audience field and the screen: the practical problem the surgeon has to solve here is how to make visible and available the relevant space of the operating theatre for the camera. He solves this problem by entering his head into the field of the camera, with a whistle vocally marking his movement, thus displacing and bringing into the visual field an element exterior to it, in an exaggerated movement – received by the audience with laughs – underlying the production of its visibility (cf. Heath and Luff, 1993, 50, on similar hyperbolic gestures securing the visibility of the movement Figure 4.5 Gesture continues, indicating intended through a technological interface). action

Operating Together through Videoconference


This first upgraded gesture accomplishes the intelligibility of the subsequent gestures, which can then be performed in a uniform manner (‘the head’ being repeated in this unmarked format). Therefore, AC takes into account in a subtle way the constraints and limitations of the video shooting in order to make accessible and available, thereby Figure 4.6 Gesture is completed, indicating offaccomplishing it, the locamera space to the audience cal ecology of the ongoing operation – comprising the space not only directly filmed by the camera but displaced into it (like the head) or indicated through it (like the screen in Figure 4.6). Thus, the visual space, as it appears on the video screen, is not just a recording documenting a specific event. Participants are actually orienting towards the visual space of action as the common focus of their different activities, even as a condition for the event to be achieved successfully. The availability of the visual image is actively managed by the participants, in various ways. Creating an image is therefore not just the task of the analyst, but a crucial concern of the members. This means that features of the image are reflexively tied to the activity going on and to its intelligibility. Sharing the Image

Figure 4.7

Surgical team’s view of the optical system

The video image transmitted through videoconference is the actual link between all participants to this particular and complex action. We could say – but it would be indeed an abusive generalization – that the ‘same’ image is transmitted to different groups. As a matter of fact, the image is transmitted in very different formats, which differ in


Orders of Ordinary Action

relation to the shape of the image, the screen where the image is projected, the users’ bodily postures towards it and the activities in which it is viewed and used: •

The surgeon looks at the image in the operating room. The screen is situated in an adequate and ergonomic position, allowing the surgeon to organize his/her team around him/herself, to stay close to the patient (for example between his/ her legs) and to look comfortably at the screen (cf. excerpt 1) (Figure 4.7). The remote expert looks at the image in the amphitheatre, while being seated at the podium table. The screen is integrated in the table; it has about the same size as in the operating room. But the expert is not operating, he is sitting in front of the screen, managing at the same time the turn-taking of the participants in the audience (Figure 4.8). The audience looks at the image in the amphitheatre, not on small screens but on a giant screen, as in a movie theatre. The image is magnified, it is looked at not by a small coordinated group watching collectively for the practical purposes of a common task, but by around fifty persons sitting and watching it individually at the same time (Figure 4.8).

Therefore, on the one side, all participants are looking at some kind of screen and the video image of the operation is attracting their attention at the same time – provoking some kind of joint attention or common focus – thereby eventually creating a space for some common action bringing together different participants who do not share the same geographiFigure 4.8 The expert and the audience in the cal space. On the other amphitheatre side, these participants are looking at the screen from the perspective of very different activities, with different relevancies and body orientations. There is indeed a considerable difference between the setting for the transmission of the image and the setting for its projection; however, this very transmission is a matter of concern for the participants: surgeons are concerned whether the audience receives the images and the sound; experts check if the images are correct; members of the audience can complain that they are not seeing anything. Thus the sharing of the images across these settings seems a fundamental feature toward which the members are oriented.

Operating Together through Videoconference

Figure 4.9

Interior endoscopic view (via handheld camera)


Moreover, we should add a fourth type of participant ‘over-hearing’ or ‘over-looking’ the scene, namely the analyst, who looks at these images in another very different setting – looking at the screen of his or her videorecorder in their office, after the videoconference, in a way that allows him or her to re-observe the sequences for an unlimited number of times; and who shows these images to another audience in another amphitheatre for the purposes of an academic presentation. Producing the Image

Various participants are actively engaged in the actual image production. This means that the image is not at all a simple document of what is going on – it is a constitutive feature of what is going on. Images are Figure 4.10 Exterior view of patient’s abdomen (via collectively produced by participants distribfixed camera) uted throughout various geographical locations but accomplishing jointly the common availability of a certain kind of image: •

The interior endoscopic view (Figure 4.9) is produced by the assistant holding the camera during the operation (another, exterior view, is produced by a fixed camera in the operating room – see Figure 4.10). But the assistant does not just hold the camera: he/she follows the instructions of the operating surgeon who (verbally) controls the camera – for instance, by asking the assistant to go back and forth, to approach or to regress (this task can also be done by a robot verbally directed by the surgeon).


Orders of Ordinary Action

• •

The surgeon can also ask the video technicians to switch between the exterior and the interior view. Interestingly, the management of the laparoscopic camera – which is central for accomplishing the operation – is not just done by the operating surgeon but can also be controlled by the expert witnessing the operation. Therefore not only the relevant optic used for the endoscopic image is publicly discussed, but furthermore the manipulation of the optic can be made within the operating room (by the chief surgeon) or outside it (by the expert) (Mondada, 2003).

As for the availability of the image, the ways in which this image is produced, encompassing all the choices to be made in order to produce a relevant image for the task at hand, are also a matter of concern for the participants: routinely, the surgeon asks if the image is good and the expert evaluates the image as ‘excellent’ or ‘beautiful’, assistants can be praised or criticized for maintaining the image or for moving it more or less adequately. These concerns about the quality, these image assessments are not made in general, but for all practical purposes for the task at hand. Moreover, advice is given in order to improve the quality of the image. Producing the Visibility of the Relevant Phenomena It is not only the image that is collectively produced, but also the visible property of the relevant phenomena to be dealt with in the surgical procedure. This visible availability of the phenomena – produced both by the manipulation of the optics (the endoscopic camera) and by the very act of dissecting itself (Hirschauer, 1991) – is also scrutinized by the members constituting their ‘professional vision’ (cf. Goodwin, 1994; Goodwin and Goodwin, 1996). This is accomplished within two intertwined activities: •

on the one hand, the action of operating itself, which exhibits in this way its rationality, orderliness, intelligibility – as it is both managed by the surgeon’s team in order for the procedure to be successfully realized and as it is evaluated by the audience, for whom it is available thanks to the videoconference; on the other hand, the action of demonstrating the operation, showing the procedure explicitly to the audience.

Both are related, but they obey different criteria: the one concerns the efficiency, safety, legitimacy of the procedure; the other concerns the teachability and reproducibility of the same procedure. The visibility of the locally relevant phenomena and of the spatial arrangement is a concern for the participants (cf. Relieu, 1999), since they can interrogate and challenge the way in which the surgeon is satisfied with certain images or looks for a better exposure; they can also discuss the way in which he himself orients towards certain landmarks or relevant reference points of the anatomy. More particularly: •

Discussions about the operative field, how it is prepared, and how it is exposed are central; the good exposure of the operative field is an ongoing

Operating Together through Videoconference


achievement by the surgeon, especially in the cases where the field has to be constantly redisposed because it is unstable – the organs and the anatomical landmarks being not fixed but tending to move away from the position in which they are disposed and arranged by the surgeon for the purposes of his operating gestures. Discussions about the kind of optics used for the endoscopic view are recurrent and important: the lens allows all to see the phenomena or not. The surgeon chooses the lenses to be employed, and he frequently has to justify why he chose this lens and not another (for example a zero-degree instead of a 30degree lens). Therefore choices of the optics are not an individual matter, but are publicly discussed and challenged. Discussions about the exact location of the instruments are also important, implying the recognition of the relevant anatomy (which is often not taken for granted). Anatomical landmarks can be challenged by the expert, thus making problematic the surgical intervention at a specific location.

These ways of accomplishing the visual space of action are relevant both for establishing the conditions of the collaborative work of the experts and professionals engaged in it, and for establishing the phenomenon they are tackling together. The pathology is never just given, nor taken-for-granted; the pathology is defined, identified, visualized by the technologies at hand which produce its relevant image, an image that is compatible with the surgical procedure adopted (cf. Lynch, 1985, 1988). The visual space of action is therefore linked to the accomplishment of the central elements of what is happening. These three last points were focusing on the making of the visual space of action. But this has to be completed by also taking into consideration the participation space more specifically shaped by the organization of talk-in-interaction. The Making of Participation Spaces In these data we have to deal with various complex and intertwined activities, such as demonstrating and operating, where different kinds of participants are involved. We can therefore ask how those groups of participants, those corporate entities, incipient teams or sides are constituted through and for these particular actions. In what follows, I will sketch some resources and devices that allow for the production of such spaces of action. They are the conditio sine qua non for performing common actions such as operating collaboratively without sharing the same geographical space. The analysis will show that although participants are distributed across various geographical places (in the operating room and in the auditorium), this disposition is not to be taken for granted. On the contrary, participants are constantly engaged in maintaining or changing this initial disposition. Participation space is therefore practically and locally accomplished by the participants (Goodwin, 1986). In what follows, I will sketch some resources mobilized for all practical purposes to manage and shape this participation space (Mondada, 2001).


Orders of Ordinary Action

Code-Switching as a Resource for Delimitating Spaces One quite straightforward manner in which those particular activities are organized is by using code-switching as an organizational resource, as in excerpt 2: Excerpt 2 (1106/p23=919/14'40/V4) 1 AC so i i will open . the leicester .. omentum ((whistles)) 2 (3s) ((cough in the audience)) 3 AC mhm ... here\ ... from the lower part here/ ... until i reach/ the 4 extragastric nerve\ . (h) okay/ (4s) and to the lower part/ . 5 °montre-moi marc°/ . until i reach the . (crow foot) here\ (7s) °show me marc°/ . 6 okay\ . now/ . you can SEE .. here/ . the pancreatic tail/ . you see/ 7 °tiens comme ça roberte\° . you see here/ the pancreatic tail/ . you °hold it like that roberte\° 8 see here\ . the right cross ... okay/ 9 X mhm 10 D very well 11 AC and you see here/ the left gastric artery\ . a:nd/ . 12 °un peu plus loin marc/ avant de plonger met toi un peu plus loin° °a little further marc/ before diving into it stay a little further° 13 .. and you see that . i’m in the leicester sack\ . in fact/ . i’m 14 ABOve the leicester sack but i created a window here ((laughs)) 15 D mhm 16 AC when i have a s- small problem because of the vessels/ . so euh .. 17 probably . i- . it’s not necessary to put a stitch but . euh i will 18 i will show you how to do that\ .. okay give me the understitch/ . 19 so you see very well the: all the screen/ . okay\ so i . °roberte 20 tu bouges énormément avec la main gauche hein° ... okay you move enormously with your left hand aren’t you°

In this excerpt, the operating surgeon, AC, is demonstrating a piece of anatomy to the audience, through a kind of interior travel within the body (‘i reach/ the extragastric nerve’, lines 3–4; ‘i reach the . (crow foot) here\’, line 5; ‘i’m in the leicester sack\’, line 13). By this kind of description, the surgeon is projecting himself into the body. This demonstration of the anatomy is recipient designed for the international audience grouped in the auditorium, and consequently it is done in English. But AC uses also French for the coordination of the action with his team – mainly for managing the production of a relevant endoscopic image for the demonstration (lines 4–5, 6–7, 11– 12, 19–20). The instructions to the assistants are coordinated with gestures pointing (with the instrument) to the relevant phenomena made available by the adequate movement of the image. Therefore, the way in which French is inserted into English is less related to the organization of the syntax of the English commentary than to the organization of the action of demonstrating, with its specific sequentiality. French instructions provide for the visibility of the next step in the demonstration. Now, this code-switching between French and English accomplishes the distinction between two recipients, the surgical team and the audience, and thereby the delimitation of two spaces of participation, the operating room and the auditorium. There is thus a raw distinction between the space of the operating room, where French is used by the chief surgeon to coordinate his team, and the space of

Operating Together through Videoconference


the communication with the amphitheatre, where English is used by the surgeon and the expert to make the action visible, accountable, teachable to the audience. The Dynamic Variation of the Work Space Interestingly, those two spaces are not fixed. They can be reshaped, reconfigured for all practical purposes. Code-switching is not obeying a fixed and preexisting functional grammar. It is a resource used by participants to shape the space they are engaged in. In this sense, it is interesting to see that the expert (D), who collaborates with the surgeon for commenting on the operation, generally comments in English, but can also switch to French (AC and EV are the two surgeons; D is the expert): Excerpt 3 (1106/K2DV1/36'/p48/1939) 1 AC 2 3 X 4 D 5 6 7 8 9


10 EV 11 D 12 EV 13 D 14 EV 15 16

so now we’re waiting for the: . the tube\ (4s) xxxxxx ((in the OR)) °oui: . mais maintenant elle est bonne/ mais il y a une minute . °yes: . but now it’s good/ but one minute ago . on était dans le bleu hein/ . je crois que c’est la fum[ée° we were in the blue aren’t we/ . i think it’s the sm[oke° [okay\ there is\ . okay\ . fine/ ((lau[ghs)) [and you can inflate the balloon with twenty-five cc vous insufflez insufflate vingt-cinq cc/ ... tu mets l’aspiration ici roberte twenty-five cc/ ... put here the aspiration roberte °oh oui\ .. c’est tout à fait bon ça° °oh yes\ . that’s totally good° moi j’attends hein i’m waiting mhm in such cases it’s important to avoid any bleeding/ (h) because . we have not a lot of place/ . if you have BLOOD/ . and you have not a lot of luminosity/ so: ehm/ euh .. we try to avoid that\

Excerpt 4 (1106/K2DV1/26'/pl43/1751) 1 AC 2 3 4 EV 5 6 D 7 D 8 EV 9 D 10

you see so/ i have now (h) a vision of my: . my stomach/ here . and i’m going to put now better my liver retractor\ (4s) °i’m going to open as . great as possible° [okay so (°comme ceci°) are you agree claude/ (°like that°) [maintenant c’est très bon [now it’s very good yes/ . very [good/ [so: in a man/ euh there is a lot of baxxxx [fat also [très bien maintenant/ depuis qu’on s’est rapproché c’est parfait . mhm very good now/ since we approached it’s perfect . mhm


Orders of Ordinary Action

A few comments on these two excerpts are appropriate. Switching to French has the effect of restricting the participation space. Not all the audience’s members speak or understand French, thus speaking French means excluding (part of) the audience, or it means making this audience an overhearing audience. Switching to French has another effect: it makes the expert participate in the same (linguistic) space as the surgeon’s. It includes the surgeon in the operating room space. It makes the expert part of the group of the operating team. How is D performing his belonging to the same space as the operating team? In excerpt 3, lines 9–13, we can observe that D integrates his turn in an exchange occurring in French between the operating surgeons AC and EV. This exchange is closed by EV’s public comment in English at line 14. As in lines 4–5, D is uttering his evaluation in a softened voice, which contrasts with his ‘official’ loud voice when he interacts with the audience in English. As in lines 4–5, D is producing a positive and supportive evaluation of the image quality. In excerpt 4 also, the image is the topic of D’s French intervention. So D is actively participating in the accomplishment, by the team, of the adequacy of the endoscopic view for the purpose at hand. In excerpt 4, the position of D’s turns in French is interesting to note. A first evaluation, in French, is given line 6 in overlap with EV’s turn. This evaluation is sequentially related to the first turn construction unit (TCU) of EV’s turn and to the action performed (disposing the liver retractor). A second evaluation is proffered in English (line 7), as a second part of an adjacency pair, responding to EV’s demand (‘are you agree Claude’, line 5). A third evaluation will be provided when the liver retractor has been finally disposed, closing the sequence. Thus D is making different kinds of evaluations: he is self-proposing evaluations that are immediately relevant to the team’s action in French, and he is reacting in English to an official request for ratification. These versions are organizationally different, one aligning with the work of the operating team and the other aligning with the official comment of the surgeon to the audience. Moreover, in both excerpts, D proposes two descriptions of the action performed by the pronoun ‘on’ (‘il y a une minute . on était dans le bleu hein’, in excerpt 5, lines 4–5; ‘depuis qu’on s’est rapproché c’est parfait’, excerpt 6, line 10). ‘On’ in French is a third person pronoun but is often used as an informal equivalent for the first person plural. Here we can observe that this pronoun is used with locations and movement expressions and has the effect of locating the expert and the team in the same space. It thus strengthens what is performed by the code-switching; that is, a participation space where the expert acts as belonging to the team, rather than to the audience. Transforming the Participation Space The previous analysis showed the way in which D is accomplishing his selfcategorization as an expert. However, his interactional activities during the operation are not confined to the expert’s advice-giving. D performs other activities, making relevant other categorization devices and switching between one device and another. These switches are interesting not only from the point of view of his multiple identities and tasks, but also with regard to the differentiated distribution of rights

Operating Together through Videoconference


and obligations to speak, which define the way in which other participants can engage in the activity at hand. In the following excerpt, D is doing the interactional job of distributing the turns at talk, therefore acting as the chairman of the ongoing discussion: Excerpt 5 1 AC 2 3 D 4 5 6 AC 7 A 8 AC 9 D 10 AC 11 D 12 13 AC 14 15

SO/ . well/ let’s start . écarteur de foie . parfait . [xxxx . °non liver retractor . perfect [xxxx . °no celui-ci c’est xxx oui . réducteur . non il faut pas de réducteur xxx° this one is xxx yes . retractor . no we don’t need a retractor xxx° [okay no question about that/ .. position/ . pressure/ . size of the trocar/ [everything is clear [so we’re going to put away a little bit the fat what is the size of the first tro[car/ . what size is the first trocar/ [°xxxx xxxx xxx° ((to the OR)) euh robert/ yes/ a question from the audience/ what size is the FIrst trocar/ could you euh [remind us [the first trocar/ .. so the first trocar here is so a a ten millimetre trocar for my optical system/ . this is a five millimetre trocar/ . euh i put here my my . my hook here/ ((continues))

This excerpt shows another form of participant organization, where members of the audience can ask questions of the operating surgeon. In this case, D is managing, as a chairman, the access to the floor for the audience. D creates an opportunity for questions to be asked at a particular point in the procedure, that is at the start of a new episode (line 1). D then makes visible the relevant sequential places where questions can be inserted – not every moment being adequate for such interventions. D plans not only the position but also the possible content of the askable questions at this place, offering a list of possible topics. He projects therefore the relevance of a possible question, which is actually posed by A (line 7). Even if the access to the floor has been prepared, taking the turn can be difficult, as in this case: AC, the surgeon, in overlap with the introduction of a possible question by D, has already begun the new phase of the operation, therefore occupying the relevant place for the question and modifying the sequential unfolding of the activity. As an effect, A’s question is not understood by AC and has to be reformulated by D (lines 11–12). Although in other places D acts as the spokesperson for the operating room (for example in commenting on the procedure), here D acts as a spokesperson for the audience. Excerpt 5 not only shows the multiple tasks performed by D and the switches of his corresponding categorical identities, but it also reveals another feature of the participation space: whereas the access to the floor is very easy for D or for the surgeon, as they self-select in various positions and do not push a microphone button in order to speak, this is not the case for the audience. The audience has some difficulty in identifying the adequate sequential point and, to self-select, a participant has to push a microphone button before speaking and generally encounters some difficulties in getting the floor, with interactional activities mostly limited to questions.


Orders of Ordinary Action

Therefore the participation space appears hierarchically distributed, with different forms of access facilitated for some members and restricted for others (Mondada, 2002b). This reinforces the specificity of D’s multiple and mobile position, engaged in various activities and locations between the operating room and the audience amphitheatre: he alternates between his collaboration with the team (as a ‘colleague’), his challenges of the work of the surgical team as an exterior expert in a possibly disaffiliating manner, his explanation of the procedure (as a spokesperson of the surgeon and a ‘co-teacher’), his management of the turns at talk of the audience (as a ‘chairman’ and sometimes a spokesperson for the audience). In every one of these actions, D is accomplishing a different job and acts within a different categorical relevancy. Detailed analysis can account for these movements on the basis of the organization and placement of D’s turns within the sequences, of the linguistic resources chosen by D to organize his recipient design, and of the organization of D’s conduct aiming at accomplishing ‘doing being a co-member’ (by producing joint completions of the surgeon’s turns, for example) or ‘doing being an expert’ (by initiating some critical advice, for example). It is through a detailed analysis of this kind that we can decide what is successively the relevant category, from the members’ perspective, for describing D – as an expert, as a co-member, as a co-teacher, as a chairman. The accomplishment of relevant category membership therefore is a central dimension of the making of this particular participation space where D participates not only in the management of the public event, in the production of the image, but also, more radically, in the operation itself. Conclusion This analysis is meant to be a contribution to studies of work and to the way in which a detailed analysis of talk-in-interaction as well as of professional vision and gestures can account for the interactive accomplishment of teamwork involving the use of various technologies. In this chapter I sketched some procedures by which members interactionally define and install an adequate space for the collaborative achievement of tasks and for the exercise of their expertise. I insisted on two interrelated aspects: first of all, the making of a visual space of common action, inscribing the surgeon’s and the expert’s activities in a specific spatial arrangement accomplished for the purposes at hand, allowing collaborative work to be realized across various geographical locations; secondly, the making of a participation space where various accesses to the floor and to the action are defined, differentiating various category-bound activities such as moderating the debate, giving expert advice, teaching the audience or challenging the operation. These fragments of analysis show some of the procedures that make possible teamwork and other collaborative activities: they aim at contributing to the understanding of what Suchman (1993) calls the ‘centres of coordination’, highly complex spaces where distributed activities have to be managed, requiring the coordination of people, technologies and objects scattered throughout distant places and times.

Operating Together through Videoconference


Transcription Conventions [ (2s) / \ ((rire)) par& (cross) * *

overlap . .. ... pauses shorter than 1 second length of pauses in seconds xxx incomprehensible segment rising/falling intonation exTRA accentuated syllable comments °ye:s° softened voice interruption of a word : vowel lengthening continuation of turn < > delimitation of phenomena noted in (( )) uncertain transcription = latching delimitation of the described actions

| (Fig. 4.4) landmark indicating the exact point to which the figure refers. Translation from French is indicative and has the purpose of helping to read the original. Acknowledgements This article has been written within the framework of a research project on the interactive construction of scientific knowledge in plurilingual settings, financed by the Swiss National Foundation (grant no. 1214-051022.97). Without the help and support of the IRCAD (Institut de Recherche pour le Cancer de l’Appareil Digestif) of the Civic Hospital of Strasbourg, fieldwork and analysis of these data would have been impossible. My warmest thanks go to Jacques Marescaux, Didier Mutter and Michel Vix for opening the doors of their labs and operating rooms for me. I am most grateful to Simona Pekarek and William Doehler for revising my English text.

This page intentionally left blank

Chapter 5

Doctors’ Practical Management of Knowledge in the Daily Case Conference Nozomi Ikeya and Mitsuhiro Okada

Introduction Ethnomethodological studies have demonstrated the ways in which (relevant) members orient themselves to the social organization of various settings as part of their conduct of activities (Garfinkel, 1967a; Button, 1991; Watson and Seiler, 1992). Much attention has been paid, in particular, to the contingency of the social organization, as the member experiences it in the developmental course of action in the ‘here and now’. Quite often, theoretical concepts are re-examined or respecified, to use Garfinkel’s term, in the context of actual settings where the member actually treats the problem of power or social distribution of knowledge as their problem, rather than the theorist’s (Garfinkel, 1991). In this chapter we examine the problem of knowledge as the member’s problem, operationalizing Schutz’s notion of ‘social distribution of knowledge’ in an ethnomethodological context. This is the concept that Schutz thought should be central to the sociology of knowledge. He thought that the mechanism of the social distribution of knowledge is, to members, the basis for carrying out any activities. Furthermore, he was not satisfied with the sociology of knowledge at that time, which was mainly concerned with the problem of the social distribution of knowledge only as a social condition, as the basis of ideological basis of the truth, especially in terms of a social effect that economic condition and education have upon a person, or the social role of educated members of the society (Schutz, 1962, 1964). Schutz argued that the problem of social distribution of knowledge was presupposed by the sociologists of knowledge. He considered that the problem of distribution of knowledge should be examined as the central issue in the sociology of knowledge. In arguing this, Schutz treated the mechanism of the social distribution of knowledge as the foundation of social actions; that is, the source of social order. However, in an ethnomethodological context, the problem of social order is not treated as something determined by theorists, but rather as a members’ problem (Sharrock and Button, 1991). In other words, members working within a particular setting take for granted certain stocks of knowledge distributed among members working within the setting; and that knowledge comes to be distributed among members. Therefore, they can carry out the activities in which they are engaged without saying in so many words what they mean by doing certain things (assuming distributed stocks of knowledge among their colleagues), whereas they may encounter situations where they find otherwise, and find themselves having to explain what


Orders of Ordinary Action

they mean so that other participants can understand. In other words, the members attempt to practically manage the problem of knowledge as part and parcel of their activities, hence as part of their practical management of order (Sharrock and Ikeya, 2000; Ikeya, 1997, 2001). Hospital settings offer us many opportunities for studying the practical management of knowledge. Doctors working under division of labour in a hospital are located within a complex and varied distribution of various kinds of stocks of knowledge – of individual patients, of specialized areas, and so forth. Doctors attempt to carry out their activities in collaboration with other doctors with different degrees of skills and expertise in medicine, or with those who have different expertise and roles, such as nurses or technicians. The daily case conference we are going to examine below is an example of activities carried out in such an environment. The daily case conference is one kind of monitoring system in the hospital (Ikeya, Okada and Fujimori, 2004). It is a ‘system’ designed for the senior doctor to learn about each case from the team doctors, so that they can monitor younger doctors’ practice. Younger doctors are the ones who are actually in charge of providing treatments to each patient taken to the hospital where we conducted our study. Furthermore, the conference is also designed for providing training, especially to junior doctors (including interns), as well as to those with a few years of experience, given the fact that they are the ones who must make the presentation of cases to the senior doctor. We will first demonstrate the ways in which the daily case conference is designed so that it allows the senior doctor to monitor their cases, as well as to provide training to junior doctors. Secondly, we will examine how the senior and junior doctors actually carry out their activity – the daily case conference – working from within the system of the case conference which is organizationally located in the hospital, especially focusing on the aspect of practical management of knowledge; that is, how doctors attempt to organize their knowledge as part of reporting the development of each case, of evaluating the case, of giving advice, and so forth. Thus, we attempt to examine how doctors distribute and redistribute their knowledge in accordance with the asymmetrically distributed knowledge among them, as part of their monitoring of each case and their training of junior doctors. This is an attempt to turn the doctors’ practical management of knowledge into a topic, which is often presupposed in the analysis of activities in hospital settings such as doctor–patient or doctor–doctor interaction.1 The Emergency Medical System There are three levels of emergency hospitals, and these levels are established and regulated by Ministry of Health, Labour and Welfare Ministry as well as Fire and 1 See, for example, Anspach (1988), Pomerantz, Ende and Erickson (1997) and Halford (1987). Ten Have, on the other hand, attempts to demonstrate how asymmetries are produced in and through the details of medical encounter, reflecting that the asymmetries have been dealt with as rules or resources in the previous studies (1991). Elsewhere, he also pays attention to doctors’ orientation to various formats in structuring their encounter (ten Have, 1989).

Doctors’ Practical Management of Knowledge


Disaster Management Agency in Japan. As summarized in Table 5.1, requirements for each level of services are as follows: hospitals which are categorized as Level 1 emergency hospitals are to accept patients with minor illnesses or injury; those categorized as the Level 2 emergency hospitals accept patients for those who need hospitalization, and they should be able to provide laparotomy and general anaesthesia if necessary; and lastly, those categorized as the Level 3 emergency hospitals are to accept patients with life-threatening conditions, regardless of their causes, for 24-hour care. The Level 3 hospitals are to accept in principle only patients who are referred by the Command and Control Centre of the Fire Department via their hotline to the hospital. Table 5.1

Classification of emergency hospitals in Japan Level 1

Minor illness or injury

Level 2

Patients who need hospitalization

Level 3

Patients with life-threatening conditions

The clinic where we conducted our study is a part of a university hospital, and is called the Trauma and Critical Centre; it is managed by the Department of Emergency and Critical Care Medicine. The Centre specifically admits patients with life-threatening conditions, regardless of their causes, on a 24-hour basis; that is, the Centre is categorized as an institution that provides Level 3 emergency services. There is another clinic (located within the same building) which provides the Level 1 and 2 emergency services. Doctors from other departments within the hospital, except the Department of Emergency and Critical Care Medicine, manage this Level 1 and 2 clinic. Doctors operate the Trauma and Critical Centre in three teams, each of which consists of approximately twelve doctors (including interns from other departments) who actually accept patients and provide treatments on the daily basis. Three teams take it in turns to be in charge for 24 hours each time: within each team about six doctors take charge for each 24-hour period, ready to accept patients whenever the Control and Command Centre of Fire Department phones in through their hotline requesting acceptance of a patient with a life-threatening condition.2

2 In Japan, it is the Control and Command Centre of the Fire Department located in each region, that accepts emergency calls from the public to request fire brigade services and emergency medical services. And it is from local fire stations geographically distributed within the region that ambulance units as well as fire units are dispatched by the Command Centre of the Fire Department. How the Control and Command Centre provides emergency medical services in collaboration with local fire stations and emergency hospitals is described elsewhere (Ikeya, forthcoming).


Orders of Ordinary Action

The Daily Case Conference Monitoring the Case All doctors from the three teams are supposed to attend the daily conference, which takes place every morning (except Sundays) from 8:45 for about one and a half hours, before the senior doctor makes his/her rounds in the ward. Each team presents to the senior doctor the cases for which they are responsible; there are usually about ten cases from each team. Not only doctors of that team, but also doctors in other teams are supposed to attend the conference, although it is often the case that some attending doctors have to leave the conference to see their patients, being called by nurses via pagers for emergencies, or to accept a new patient who arrives during the conference. The doctor who normally receives reports is the director of the Centre. This doctor is not part of any of the three teams, and is in the position of monitoring team member doctors’ management of patients. That is, it is team member doctors who actually take patients in and provide treatment to the patients. Hence, the senior doctor needs to know on a daily basis what kind of patients their clinic has admitted and what kind of treatments are being provided by the doctors. Thus, the daily case conference is designed to inform the managerial doctors about each case, so that they can keep themselves updated about its development over its course and about its prospects.3 In this sense, the daily case conference can be described as a mechanism for managing the distribution of knowledge among doctors, operating as part of their monitoring of cases. Presentation of each case is organized in such a way that the senior doctor becomes well enough informed to be able to monitor the case, and to give advice to the team members on the basis of the information provided. The senior doctor attempts to scrutinize each case in terms of two points: control of the quality of the practice and control of the allocation of beds in the Centre. In fact, the case report presented in the daily case conference is a formatted formulation consisting of items of information organized in such a way that the senior doctor can monitor each case. By ‘format’ we mean a way of organizing information about a particular patient to present as a case.4 The doctor presenting a case organizes certain items of information of this particular patient into a temporally and causally textured format. The items of 3 The senior doctor also gains information and provides instructions during rounds. This is another interesting setting to examine in terms of the practical management of knowledge. Anspach examines case presentation at bedsides during rounds doctors make (1988). 4 There seem to be different kinds of case presentations carried out for different purposes in different settings: for example, for reporting to doctors during rounds about changes in a patient’s condition at the bedside of a patient; or for reviewing treatment provided to a patient in order to standardize decision-making within the hospital, etc. As Garfinkel argues, ‘any self-reporting system had to be reconciled with the routine ways in which the clinic operated’, and the case presentation we examine seems to be another example (Garfinkel, 1967a: 187). Anspach has listed various kinds of settings in medicine where case presentation takes place (1988).

Doctors’ Practical Management of Knowledge


information are shown in Table 5.2. The doctor organizes the items in a temporal order, in the sense that he or she attempts to provide the senior doctor with the overall picture of (1) how the team has managed the patient retrospectively; and (2) how the team is going to manage the patient prospectively.5 Starting with the first group of items such as the patient’s name, age, gender and the time of arrival at the clinic, the presenter is required to report the patient’s medical history before he or she was brought to the clinic as well as the history of the present case and how the team has managed it so far. Then in the second part of the report, the doctor also reports what kind of prospect the team has about the patient, including the team’s plan for examination and treatment, and their plan of transferring the patient to another department or another hospital. Table 5.2

• • •

Format of case report

Patient’s name, age, gender and time of arrival Medical history History of present case (chief complaint, how the current problem occurred and how the patient was brought to the clinic, condition at the time of arrival, treatment provided, possible causes, changes and present condition) Prospect (plan of the course of treatments, prospective view of the patient’s condition, and plan of transferring the patient)

The format of the case report can be seen as causally textured not only in the sense of listing events and related data in a temporal order but also in the sense of causal explanation in scientific terms – that is, what causes particular symptoms. Thus, the patient’s medical history, for example, is referred to not only in a temporal order with regard to the history of the present case that led to admission, but also in a causal order with regard to the cause of the patient’s present symptoms. Based on the case report provided, the senior doctor evaluates the team’s management of the patient in the past, present and future, and gives some advice, including future courses of action. The senior doctor, while listening to the case report, examines whether the team has made adequate and coherent causal links between the data they have at hand, whether they have provided adequate treatments accordingly, and whether they have an adequate prospect about the patient on the basis of their coherent analysis. Further, the senior doctor tries to draw doctors’ attention to any alternative way of formulating the problem, alternative ways of making links between data. He or she does it not only when there seem to be obvious mistakes in their making causal explanations about the patient’s conditions, but also when it appears that their explanations are plausible, thus making doctors take precautions in case they are wrong, or encouraging them to reconfirm that this is the 5 Berg points out the temporal structure of medical work as ‘all activities are located within a trajectory’s projected “history” and “future,” both of which are continually reconstructed’ (Berg, 1997: 137).


Orders of Ordinary Action

case, and not the other way. Thus, the senior doctor controls the quality of the care provided at the clinic by scrutinizing the teams’ management of their patients. Furthermore, apart from carrying out record-keeping of admission and discharge of each patient, the senior doctor, on the basis of accounts provided by the team, may attempt to encourage the team leaders to start considering discharging specific patients; that is, transferring specific patients to another ward or to another hospital, if the team does not appear to be already considering it. For the senior doctor is often able to foresee the developmental course of the condition of the patient, and may well see that the patient will be out of the life-threatening condition before long if things develop in accord with the account provided, providing no sudden change occurs. In this way, the senior doctor tries to carry out bed control, attempting to make efficient use of the beds that should be allocated specifically to those who have life-threatening conditions. So far we have described how the case report in the daily conference is designed to inform the senior doctor of each case so that he or she can carry out quality control of the doctors’ practice as well as bed control. Moreover, the daily conference is designed to enable doctors to share information about all patients in the clinic, with all the doctors being supposed to attend the conference. This allows the doctors to monitor developments of all the patients, regardless of the team in charge. This is tied to the following things. First, it allows the doctors to prepare themselves for taking necessary actions in case the patient’s condition changes while the doctors in charge are not available. In addition, by listening to the senior doctor’s comments and advice for each case, they come to share the instructions, whereby quality control is ensured. Training the Junior Doctor We have discussed how the daily case conference is designed to provide formatted formulation about each patient as a case to the senior doctor, so that he or she can monitor younger doctors’ practice, thereby carrying out quality control of their practice and bed control. The conference is also designed for providing training to junior doctors, by making them provide accounts of each case to the senior doctor. Within the team, it is the team leader and a few doctors with over four years’ experience in emergency medicine who actually make decisions as they carry out examinations of the patient; hence, it is they who know best about a particular patient. Naturally, if one of the experienced doctors presents the case in the conference, the presentation proceeds smoothly, and at least he or she will not have a problem in providing necessary information about a particular patient in a certain format. However, usually it is not one of the experienced doctors who gives the case presentation. Experienced doctors let one of the junior doctors give the presentation. These are doctors with less than three years’ experience in emergency medicine, or interns from other departments who rotate from one department to the other, staying in each department for three months. In order for the junior doctor to be able to report a case, he or she has to go through pieces of data derived from various kinds of observations and examinations, and also relate various decisions made based on those data. Junior team members are instructed to carry out different manipulations to the patient on site, but usually only parts of each operation. It is often the case that reasoning of each decision made by the senior team members is not necessarily fully

Doctors’ Practical Management of Knowledge


shared by the junior team members. Since doctors are usually allowed only a limited length of time for making decisions about the patient, the senior team members would not say in so many words what is going on to the junior doctors on site; instead, they simply give orders and instructions to junior doctors about what the latter should do. Thus, by asking the junior doctors to report the case in the conference, the senior team members provide a chance to the junior doctor to reconstruct the entire course of development vis-à-vis the patient, from the time the patient was brought into the clinic, relating actions taken and data derived from examinations and observations. In other words, the junior doctor has to organize information as a case. That is to say, the report has to be presented in a certain format that the junior doctor is still in the process of learning. Thus, the daily case conference is designed in such a way that the junior doctor can be trained by going through preparation for presentation and actually giving the presentation to the senior doctor, as well as by being given comments and instructions by the senior doctor during the conference. While the senior doctor may not know in detail about particular patients in the clinic, especially about those he or she has not managed to see – the patient who arrived between the day before and the morning when the conference is held – he or she usually has good understanding of the frontiers of medical knowledge as well as textbook medical knowledge. Besides, he or she also has his or her own experiences of dealing with patients in the past. So even if the senior doctor may not know about this particular patient, he or she is usually able to point out any inconsistencies between the data presented and the team’s decisions. In other words, he or she usually has the well-grounded knowledge of what conditions have to be satisfied to make a patient as a normal case (Sudnow, 1965). Thus, the presenter of a case is not the only person who conveys what is known to others in the conference. In fact, as it will be demonstrated in the following analysis, the senior doctor helps the junior doctor in reconstructing the case, points out inconsistencies and provides alternative ways of viewing this particular patient, referring to the medical knowledge. Thus, training the junior doctor is carried out in accordance with the asymmetrically distributed knowledge, distributing and redistributing what they know to each other. Co-Constructing a Case Report One thing to be noted about the ways in which presentations in the daily case conference are carried out is that the case report, which is supposed to be produced by the presenter, is often found to be co-constructed with the senior doctor. That is, what we find is not necessarily that the case presenter completes the part of case report on his or her own, and all the senior doctor has to do is to give advice to the team on the basis of information provided in the report. Nevertheless, the extent to which the senior doctor gets involved in constructing a case report varies. As we noted earlier, when a person (who has already learned how to construct a ‘story’ based on the information about a patient) presents a case, part of the report may be completed in one turn; and the senior doctor may accept the report without asking for any further clarifications or questions, although some advice may or may not follow. That may happen when the team leader or senior team member, who knows well enough about the case itself as well as having a skill for the presentation, presents a case. Further,


Orders of Ordinary Action

if the case has been already repeatedly reported, and most problems in the case have been already settled, then the presenter would know how to put information together as a case report, and would be able to assume that the senior doctor knows fairly well about the case. In these cases, the extent to which the senior doctor gets involved in constructing a case report is very limited. On the other hand, there are times when the senior doctor needs to get heavily involved in constructing a case report with the presenter. For the senior doctor may not be able to find enough information provided to evaluate the case and the team’s management of it, or in some worse cases, he or she may not find any coherent ‘story’ provided in front of him, except for the enumeration of facts in the initial report. This tends to happen particularly when the case is presented for the first time in the conference, and when it is totally new to the senior doctor who listens to it. What can be found often at the initial stage of treatment in emergency hospitals is that there may be various problems that are unsettled, such as the need for further examinations, and it could be complicated to formulate the report, particularly for the junior doctor who does not have much experience in giving treatment to patients or in producing case reports. In such cases the senior doctor ends up helping a case presenter in constructing a case report, gathering relevant pieces of information to formulate into a story, step by step, as they go along, rather than dismissing the report given to him and asking somebody else to give the report. The senior doctor asks for information primarily of the case presenter, yet other team members senior to the presenter may also provide information as they proceed. In fact, other such members often get involved in constructing a case report as well. It is also noted that the senior doctor seems to process this by means of orientation to the format we described earlier in the chapter. He or she does so, not only in the sense that he or she uses the format to collect a certain set of information about a case, but also in the sense that he or she uses it to order them in particular ways to construct a ‘story’. In the following data, we will examine how the senior doctor and team members attempt to co-construct a case report in the daily case conference. Excerpt 1 (SD = senior doctor; JD = junior doctor; TL1 and 2 = team leaders) 1 JD Uh::, next, Taro Yamada. (4.0) uh:(3.0) hyponatremia and pneumonia 2 {cough}xxxx this patient has {cough} chronicle respiratory failure. (3.0) 3 Probably respiratory failure was caused by pneumonia, CRP is 7.7, leukocyte is 4 12,200, leukocyte has been increasing but CRP has not, right, has not changed 5 much. 6 SD What? 25th means Saturday? 7 TL1 Saturday, right. 8 SD Uh: what brought him here? I mean …

The senior doctor, after having listened to the report to the point where the presenter starts talking about the patient’s conditions, inserts a question to confirm the date of the patient’s arrival, which should have been mentioned by the presenter (line 6). Since the patient was brought in on Saturday, there was no conference for two days before the 27th, Monday. Despite the fact that this is the first time the presentation

Doctors’ Practical Management of Knowledge


for this patient is held, the junior doctor presenting the case does not mention the date of the arrival, which is one of the first necessary items of a case report. Team leader 1 joins in the conversation to answer the senior doctor’s question about the date of arrival (line 7). Then, the senior doctor further asks a question, ‘what brought him here?’(line 8), showing that he cannot see clearly what kind of condition the patient is in, and makes a request for reconstructing the report. The senior doctor sees that the junior doctor has thus far simply listed diagnostic names and data without ordering them to constitute a ‘case’ by following the format. Excerpt 2 6 SD 7 TL1 8 SD 9 JD 10 11 SD 12 TL1 13 JD 14 SD 15 16 JD 17 TL1 18 SD 19 TL2

What? 25th means Saturday? Saturday, right. Uh: what brought him here? I mean … Uh: respiratory failure. xxxxxxxxxxx The main complaint was [respiratory failure [well, he came for the hematemesis = =hematemesis [oh, I’m sorry [Came for the hematemesis? (4.0) Well, then? hematemesis, to Level 1 and 2 clinic? we:ll= =to the Level 3 clinic through hotline for hematemesis as the main complaint [consciousness disorder

Then the senior doctor moves to the next item of information: the chief complaint, the main reason for coming to their clinic. He tries to confirm the chief complaint as respiratory failure based on the junior doctor’s report, but team leader 1 made the correction as hematemesis (lines 11–12). Additionally, the senior doctor asks whether the patient came via the Level 1 and 2 clinic. In the report, junior doctor did not mention how the patient was brought to the Level 3 clinic. Team leader 1 answers that the patient was brought straight into the clinic in an ambulance, after a hotline call from the Command and the Control Centre of Fire Department (line 17). Team leader 2 adds ‘consciousness disorder’ as another item of chief complaint (line 19). Excerpt 3 21 SD 22 JD 23 24 SD 25 JD 26 TL2 27 28 SD 29 TL1 30 SD 31 JD 32 SD

I see:. Then? Right, then, performed the endoscopy and, we found adenoma on esophageal glands= =huh? Where? on the esophagus AGML-like lesion of acute gastric mucosa was found on the esophagus, stomach, and then in the second portion of the duodenum uh huh= =and then regurgitate esophagitis has been worsening. I see (.) then? No active bleeding [was found [xxx


Orders of Ordinary Action

Having confirmed the chief complaint and the route of the patient, the senior doctor then urges the junior doctor to move to another item of information (line 21). The junior doctor explains what the team did for hematemesis, that is, performing the endoscopy, and that they found adenoma on the esophageal glands, but found no active bleeding. Excerpt 4 33 TL1 34 35 SD 36 TL1 37 38 SD

Originally, {the patient} has been suffering from alcoholism, chronic alcoholism uh huh And then, living by himself, well:, (2.0){laughter} he vomited blood or, we:ll, his neighbour found him lying on the bed I see (.) then?

Team leader 1 joins them again in reporting on the patient’s medical history, which is supposed to be reported in relation to the history of present case and the team’s management of the case; that is, performing the endoscopy for hematemesis (lines 33–37). In this way, team leaders attempt to add the missing information to the junior doctor’s report as the senior doctor reconfirms each item of information that is supposed to constitute the report. Excerpt 5 40 SD 41 42 JD 43 SD 44 JD 45 SD 46 JD 47 SD 48 JD 49 SD 50 JD 51 SD 52 JD 53 SD 54 55 JD

ninety-something? Wrong. Sixty years old? We:ll (2.0) and then, uh:= =hematemesis and alcoholic= =yes, [xxxxx [has not been taken care of himself right, right= =yeah. Then vomited blood (2.0) and collapsed right The:n, was found by someone right, we:ll brought by the ambulance right Yeah. The:n, took a peek from the esophagus (.)down(.), found hemorrhagic lesion = =yes

The senior doctor, having gone through some items of information with the team members, starts reorganizing them to formulate a case report. He starts with the patient’s age, medical history (that is, chronic alcoholism), history of present case (that is, he vomited blood and was found by his neighbour), chief complaint, the team’s management, and its result (lines 40–53). In this way, he repeats what has been confirmed so far, by filling in missing elements of the junior doctor’s original report. Further, the senior doctor does so in the form of telling a story in vernacular terms, such as ‘vomited blood and collapsed’ (line 47), ‘took a peek from the esophagus’

Doctors’ Practical Management of Knowledge


(line 53). By doing this, he seems to be highlighting each element and its order to put the elements to formulate a report, so that junior doctor can learn how to talk about a particular patient as a ‘normal case’, as a coherent and recognizable ‘story’ of a patient, instead of simply listing the patient’s data and conditions. Excerpt 6 53 SD 54 55 JD 56 SD

Yeah. The:n, took a peek from the esophagus (.)down(.), found hemorrhagic lesion = =yes What else was found?

Having finished the reformulation of what is confirmed so far, the senior doctor requests for similarly textured reports on the other chief complaints (line 56). The senior doctor already knows from the foregoing discussion that there are other chief complaints, such as respiratory failure and consciousness disorder, apart from hematemesis. For him, only the part of hematemesis has been clarified and confirmed at this stage. But the junior doctor does not follow his instruction, and starts enumerating data as if she were restarting from the place where she was stopped from giving her report by the senior doctor at the beginning of the presentation (line 6 in Excerpt 1 and lines 57–88 in the Appendix). Excerpt 7 86 JD 87 88 89 SD 90 91 TL2

But today, PaO2 has decreased to 45 so (2.0) we:ll we also increased O2 to 7 litre. If that 82 doesn’t bring about any improvement, we have to think about intubation again. Yeah, intubation, well, what (1.0) is this guy anyway? (2.0) Originally came for hematemesis and=

In the end, the senior doctor asks a question, ‘well, what is this guy anyway?’ (line 89 in Excerpt 7). By asking ‘what is this guy anyway?’, the senior doctor is almost repeating the initial question to junior doctor, ‘what brought him here?’ (line 8). By asking this question, he informs the junior doctor that she is not giving him satisfactory information. Then finally team leader 2 intervenes to properly respond to the senior doctor’s request. Excerpt 8 91 TL2 92 SD 93 TL2 94 95 SD

Originally came for hematemesis and= =yes consciousness disorder, but when he came here, uh, as for the hematemesis, there wasn’t that much bleeding= =huh=


Orders of Ordinary Action

For each of three chief complaints – hematemesis, respiratory failure and consciousness disorder – team leader 2 provides an account. He starts with hematemesis, the least problematic item among the three chief complaints (line 93). Excerpt 9 96 TL2 97 98 99 100 SD 101 TL2 102 103 104 105 SD 106 TL2 107 108 109 SD

=the rest was the respiratory failure, uh: when he came, gas was, PaO2 was, uh, 46 or 50 or something like that, and then we immediately intubated him, and respiratory condition had improved, so we extubated, but seems like {the patient} has chronic respiratory failure, emphysema or something like that [uhn…. It was like emphysema and seems like the breathing condition was bad from the beginning, and x-ray film taken at the arrival to the hospital showed a shadow of invasion at the lower part of the right lung, so may be that makes the breathing condition [worse [uh-huh that has improved little bit under the intubation, so we extubated once, but still the breathing condition is not good because he cannot expectorate. And now the breathing condition has been= =hu huh=

As for the second chief complaint, respiratory failure, team leader 2 reports on (1) the condition at the time of arrival, that is, PaO2 around 50 (line 96); (2) the management carried out by the team, that is, intubation (line 97); (3) the consequent change of condition, which was improved, with the subsequent team’s management, that is extubation (lines 98–99); (4) that the possible cause of respiratory failure is considered to be a shadow of invasion in one lung; and (5) their plan for intubating him again because of the breathing condition worsening. Excerpt 10 110 TL2 111 112 113 114 115 116 SD 117 TL2 118

=worsening so we are thinking of intubation. And as for consciousness disorder, hyponatremia, well, hyponatremia and a kind of central DI, 62 at the time of arrival, now it has been up to around 133. Consciousness level was once up to the half of single digit, but now somewhere around 1–3 to 2–10. uh:{cough} one possible reason for this we can think of is that (2.0) there’s something wrong with the brain or= =uh huh?= =or, since it took only 24 hours or so to recover from 112 to 135 this time, and then suddenly paroxysmal consciousness disorder xxxxx.

As for the third chief complaint, consciousness disorder, team leader 2 again reports on (1) the condition at the time of arrival, that is, the value of natrium and the level of consciousness (lines 111–114); (2) any changes over time, that is, the consciousness level has not been fully recovered; (3) candidate diagnosis as hyponatremia and central DI (diabetes insipidus) (lines 112-113); and (4) a candidate cause of consciousness disorder being central DI (that is, it is caused by some brain abnormality). As we will examine in the next section, the senior doctor then moves to the second phase, by first pointing out inconsistency in the team’s reasoning; this is to say, he sees that sufficient information has now been provided by the team for evaluating

Doctors’ Practical Management of Knowledge


the practice and for giving advice, as a consequent of co-constructing the case report together with the junior doctor, with much assistance from the two team leaders. Carrying Out Quality Control As part of each case presentation in the daily case conference we are examining, the senior doctor attempts to provide necessary advice or comments on the team’s practice vis-à-vis the patient, based on his or her assessment. In order for the senior doctor to do so, he or she requires a fairly good understanding of the case, including how the team has managed the patient so far. Thus, the senior doctor pursues to obtain ‘sufficient’ information from the team; that is, information that consists of the right set of items in the right textured format as we have examined in the above excerpts (where the senior doctor insists on gaining missing items of information). For the senior doctor, those items of information he or she requires in the case report are necessary for him or her to understand what kind of patient they are dealing with, to examine whether the right decisions and actions have been taken vis-à-vis the patient, and to examine whether the doctors have the right prospect about the patient especially with regards to the timing they should transfer the patient from the Centre. That the senior doctor does not usually directly get involved in the management of the patient means that he or she might have to rely on the team’s report in understanding the case most of the time. Yet the senior doctor can often point out the team’s possible misunderstandings of the case. He or she does so in the way of pointing out inconsistencies in the team’s report, by bringing the team’s attention to the discrepancy between their reasoning and their data at hand, or to a point that has been left out by the doctors in their reasoning of the case. Excerpt 11 119 SD 120 JD 121 SD 122

yeah, we::ll, if, it’s central DI hu huh? yeah, we::ll, if, it’s central DI and has been polyuria, why {he} became hyponatremia? That doesn’t make sense, does it?

In Excerpt 11, having listened to the team leader’s formulation of the case on three chief complaints, the senior doctor finally sees that the team has provided sufficient information, and then he points out inconsistency in the team’s reasoning that the patient cannot be suffering from central DI which is causing polyuria, at the same time as being hyponatremic (lines 119–120). Let us examine the discussion between the senior doctors and team doctors from the part a little earlier. Excerpt 12 110 TL2 111 112 113

=worsening so we are thinking of intubation. And as for consciousness disorder, hyponatremia, well, hyponatremia and a kind of central DI, 62 at the time of arrival, now it has been up to around 133. Consciousness level was once up to the half of single digit, but now somewhere around 1–3 to

82 114 115 116 SD 117 TL2 118 119 SD 120 JD 121 SD 122

Orders of Ordinary Action 2–10. uh:{cough} one possible reason for this we can think of is that (2.0) there’s something wrong with the brain or= =uh huh?= =or, since it took only 24 hours or so to recover from 112 to 135 this time, and then suddenly paroxysmal consciousness disorder xxxxx. yeah, we::ll, if, it’s central DI hu huh? yeah, we::ll, if, it’s central DI and has been polyuria, why {he} became hyponatremia? . That doesn’t make sense, does it?

What the senior doctor does here is to bring a new diagnostic category, ‘polyuria’, to reformulate the last part of team leader 2’s report on the third chief complaint, consciousness disorder. On the one hand, team leader 2 formulates the cause of consciousness disorder as hyponatremia and central DI (lines 108–113). The senior doctor, on the other hand, reformulates this part, based on the team leader’s reasoning: the cause of polyuria (production of abnormally large volumes of urine) as central DI (that is, polyuria caused by some brain abnormality) (lines 119–120). He narrows down the problem from consciousness disorder to polyuria, so that he can bring doctors’ attention to the issue he wants to focus on – that is, inconsistency in their reasoning. He suggests that the team’s reasoning is contradictory, since if the patient is suffering from central DI, the sodium (natrium) level should be high, that is to say, he should not be suffering from hyponatremia (where the natrium (sodium) level is low). That is to say, the symptom of hyponatremia that the team has identified based on their data should not coincide with central DI since that should show a raised level of sodium as its symptom. This reasoning the senior doctor brings to bear is part of basic medical textbook knowledge, and he in fact suggests team members look into a textbook later in the presentation. By using question formats such as in ‘If it’s central DI and has been polyuria, why does he have hyponatremia?’ or ‘That doesn’t make sense, does it?’, the senior doctor seems to encourage the team doctors to see for themselves the inconsistency in their reasoning in relation to a specific part of basic textbook knowledge. In this way, the quality control of the team’s practice is carried out as part and parcel of case presentation, as the senior doctor pursues not only sufficiency of information provided by the team, but also its consistency. These are the requirements that these team doctors should satisfy in order to achieve their accountability. Carrying Out Training Formulating a case in a standard format itself is a big challenge for junior doctors, as it is demonstrated in various studies (Anspach, 1988; Atkinson, 1988; Halfond, 1987). He or she may have a considerable number of particular facts about this patient as a result of the team’s work – lots of numerical data as well as X-ray and CT films produced by a whole range of examinations – to take into account. Yet formulating them as the team’s present state of knowledge about the patient may not be straightforward, especially for the junior doctor. For it is often the case that he or she has not fully acquired the skill of relating particulars in terms of textbook medical knowledge that is at hand, whilst also using the format of the case report

Doctors’ Practical Management of Knowledge


that he or she has not fully acquired either. Besides, the junior doctor may not know exactly how each decision was made in the course of the team’s work nor how particular parts of the data and of general medical knowledge were used as the bases for such decisions. Thus, presenting a case report is treated as a good occasion for training junior doctors in treating a particular patient as a ‘medical case’, by making him or her reconstruct the entire corpus of knowledge that the team has accumulated so far about the patient. In the conference, for each case, the junior doctor in charge is to give a report to the senior doctor, preferably without any help from other team members. Other team members, in particular members who have more experience than the presenting junior doctor, tend to let the presenter talk on his or her own until he or she completes the report, or until the senior doctor starts asking for clarification even before the completion of the report. They do so because they regard the presentation as part of the training of junior doctors. This does not suggest that other team members simply treat each session as one of training all the way through. They take their turns when they see it is necessary for them to provide information to the senior doctor. For example, as we have seen earlier in Excerpt 7, one team leader replies to the senior doctor’s question, ‘What is this guy anyway?’ after a pause. Team leaders, usually respecting the turn-taking system of the talk between the senior doctor and the presenter, tend to take their turns when there seems to be a space for turn-taking transition; for example, when the presenter seems to have finished his or her turn, or when the presenter seems to be taking time in responding to the senior doctor. Yet, occasionally team leaders may even interrupt the presenter’s talk, particularly when he or she is understood to be not providing adequate information. In fact, team leaders are not only interested in the training of junior doctors, but also in conveying the ‘correct’ information to the senior doctor, so that the senior doctor can assess their practice based on the ‘correct’ information. So a team leader, as we have already examined, may eventually start reporting some parts of the case (or even do the entire report by him/herself), when he or she sees that it is better to do so in terms of economy of the team’s practice, rather than letting the junior doctor to carry on the report with great difficulties (as can be observed for example, in lines 89–118 in the Appendix). Thus, team members do not treat this conference purely as a training session for junior doctors, but as the setting in which the adequacy of their practice is also assessed; that is, their accountability is tested, as examined earlier. It seems the participants understand the importance of training the junior doctors, and in most cases, the senior doctor and other team members let the presenting junior doctor accomplish the task of case reporting by him/herself as much as possible; however, when they see the ‘correct information’ is not conveyed, they cut in and take necessary actions such as the senior doctor asking questions for clarification, team leaders providing adequate and necessary information, etc. As we have pointed out, sometimes team leaders take over the task of case reporting when junior doctors are in trouble. But that does not mean the training is completely taken over by the priority of conveying the correct information, since the team leaders are taken to be displaying an ‘adequate’ or ‘proper’ way of presenting a case to junior doctors, which may be considered as part of the training.


Orders of Ordinary Action

Furthermore, it is not only the junior doctor who is in the position of being trained in the case conference. In fact, the senior doctor may challenge the team’s practice while scrutinizing the information provided by the team, and may lead the team to consider alternative ways of managing the patient. In the following excerpt, the senior doctor starts challenging the team’s reasoning. While it is the junior doctor who actually responds to this (‘hu:huh?’), the senior doctor, in fact, may not be challenging the junior doctor as the individual so much as challenging the team, knowing that it is the team leaders who have constructed this reasoning. Excerpt 13 119 SD 120 JD 121 SD 122

yeah, we::ll, if, it’s central DI hu huh? yeah, we::ll, if, it’s central DI and has been polyuria, why {he} became hyponatremia? . That doesn’t make sense, does it?

As we have noted earlier, here the senior doctor presents the basic textbook knowledge: if the patient has central DI and also polyuria, he cannot be hyponatremic. Yet, by not going into details of this patient’s conditions as well as by formulating this in the interrogative mood, the senior doctor can be seen to be leaving things open, letting the team consider whether they can explain this logical contradiction in this particular case. In other words, the senior doctor leaves room for the team to do ‘self-correction’, rather than imposing his/her observation based on the information provided by the team. In so doing, he/she can maintain him/herself in accord with the principle of medicine not to give diagnosis without observing the patient him/ herself, while respecting the professional autonomy of the team doctors. Thus, in the daily case conference, team doctors, including the team leaders, are put in the position of being carefully trained by the senior doctor.6 Carrying Out Bed Control As part of pursuing the effective use of medical resources, the senior doctor attempts to pay attention to each patient so as not to keep patients in the clinic longer than necessary. First of all, he tries to make sure that beds are ready for potential patients with life-threatening conditions. Doctors in the clinic pay close attention to their patients in terms of planning when to transfer them to an appropriate bed or hospital as well as in terms of providing appropriate treatment. Staying in the Level 3 emergency facility costs between five and ten times as much as staying in the general ward. This can affect the hospital, the patient and the family. So the minimizing of the period of their stay is pursued not only for the sake of potential patients who require this intensive care, but also for the sake of the hospital and the family. 6 Although here we do not go into detailed discussion between team leaders and senior doctor, later in the data team leaders attempt to defend themselves against the senior doctor’s challenge, but in the end the senior doctor asks the team to carry out an examination to see whether the team’s reasoning can be proved against the data.

Doctors’ Practical Management of Knowledge


In order to pursue this principle of effective use of resources, the senior doctor (as an administrator of the clinic) attempts to pay close attention to each patient. As part of this, he or she records essential information about each patient, when the patient was accepted via which route, the chief complaints, and when he or she left the clinic and they went (or sometimes the date he or she died). In fact, the record-keeping is carried out as part of the senior doctor’s authorization of both admission and discharge of each patient, which is a part of bed control. The authorization is carried out in the daily case conference by the senior doctor’s checking of the information provided by the team in the case report – the patient’s name, age, gender and the time of arrival – against those on the form for admission produced by the clerk. Excerpt 14 6 SD 7 TL1 8 SD 9 JD

What? 25th means Saturday? Saturday, right. Uh: what brought him here? I mean … Uh: respiratory failure.

In this excerpt, the senior doctor attempts to clarify the date of the patient’s arrival and the chief complaint (lines 6–9). He then tries to confirm the route by which the patient came to the Level 3 clinic, either straight to Level 3, or via the Level 1 and 2 clinics within the Centre (lines 14–15). These items of information should be recorded by the senior doctor, but are missing from the junior doctor’s initial report. By recording the full set of information, the senior doctor keeps track of each patient. This enables him/her to ask the team doctors in charge to search for possibilities of transferring patients who are well enough, or shortly will be well enough, to leave the clinic. With regards to the decision about when to transfer or discharge the patient, the doctor in charge has to make some negotiations with different parties. This can sometimes be a complicated task for the doctor. There may be a hospital where the patient has stayed before, or a specific hospital the family prefers. But a bed may no longer be available in a previous hospital; in such cases, doctors may have to wait before they transfer the patient, or may attempt to negotiate with the family to settle with an alternative arrangement. Or it may even take time for the family to agree on the very idea of transferring the patient out of the clinic. Thus, bed control carried out in this clinic is a kind of triage in the broad sense. The reason we describe it as ‘triage in the broad sense’ is that the senior doctor does not weigh a particular patient against another, although this may be usual with triage in other settings. For example, in a general ICU (intensive care unit), whenever ICU is contacted by doctors from general wards, simultaneous decisions may have to be made over who should stay, who should leave, and who should be admitted to ICU (Zussman, 1992). In the clinic, the senior doctor does not have to decide who should leave when a new patient arrives, partly because updated information of bed availability is transferred to the Command and Control Centre of the Fire Department, and the Centre would not contact the clinic if the beds were full unless it were really a serious situation (Ikeya, forthcoming). In addition, even if the controller phones


Orders of Ordinary Action

in to check the availability of beds, doctors have the right to reject patients on the ground of availability of beds, and they do not have to consider, for example, moving their patients out of the clinic so that they can accept another new patient. Nevertheless, doctors have to be well ahead of time in initiating the search for a place to which the patient can be transferred. And the senior doctor often presses the doctor in charge to start the search even if there is no sign visible (on the basis of his/her knowledge and experiences) that the patient is, or will shortly be, in a sufficiently stable condition to leave the clinic. Thus, the daily case conference is a significant occasion to monitor the practice within the clinic in terms of carrying out bed control. Conclusion In our attempt to describe the doctors’ competence in conducting the daily conference, we came to describe how much the daily conference is embedded in the activities organizationally located in the hospital. That is to say, the reporting of each patient is organized in such a way that the operation in the emergency clinic can be monitored from certain aspects such as the quality of treatment and the efficient use of beds. These aspects are part of the core principles for operating the emergency medical clinic. Additionally, doctors try to train junior doctors by making them to make presentations. This teaching aspect is another important principle since this hospital belongs to the university. We certainly do not intend to claim that it is possible for us to see all aspects of the emergency clinic by just examining the daily conference. However, it is clear to us that the doctors’ competence in conducting the conference is connected to the practical management of distributed knowledge, not just for carrying out the conference per se, but also for carrying out other operations with principles shared throughout the hospital. And principles that constitute distributed knowledge (such as the necessity for monitoring for quality control and bed control) as well as training junior doctors are probably shared among hospitals as general principles. Here we are reminded of Garfinkel’s characterization of clinical records as one form of doctors’ self-reporting system, which are organized as being tied to their ‘routinized and valued practices’ (1967a). Thus, through our attempt to understand the doctors’ competence in conducting the conference, we can come to gain a good understanding of what kind of practical management of knowledge is involved in being a doctor at the emergency clinic. Clearly we do not intend to claim that we come to understand medical knowledge as doctors do. But we hope to have shown how what can be described as ‘social’ is interwoven with so called ‘medical’ matters when it comes to actual decisionmaking.

Doctors’ Practical Management of Knowledge

Appendix DATA SD = senior doctor, JD = junior doctor, TL1 and 2 = team leaders 1 JD 2 3 4 5 6 SD 7 TL1 8 SD 9 JD 10 ? 11 SD 12 TL1 13 JD 14 SD 15 16 JD 17 TL1 18 SD 19 TL2 20 ? 21 SD 22 JD 23 24 SD 25 JD 26 TL2 27 28 SD 29 TL1 30 SD 31 JD 32 SD 33 TL1 34 35 SD 36 TL1 37 38 SD 39 (4.0) 40 SD 41 (2.0) 42 JD 43 SD 44 JD 45 SD 46 JD 47 SD 48 JD 49 SD

Uh::, next, Taro Yamada. (4.0) uh:(3.0) hyponatremia and pneumonia {cough} xxxx this patient has {cough} chronic respiratory failure. (3.0) Probably respiratory failure was caused by pneumonia, CRP is 7.7, leukocyte is 12,200, leukocyte has been increasing but CRP has not, right, has not changed much What? 25th means Saturday? Saturday, right Uh: what brought him here? I mean … Uh: respiratory failure xxxxxxxxxxx The main complaint was [respiratory failure [well, he came for the hematemesis = =hematemesis [oh, I’m sorry [Came for the hematemesis? (4.0) Well, then? hematemesis, to Level 1 and 2 clinic? we:ll= =to the Level 3 clinic through hotline for hematemesis as the main complaint [consciousness disorder [xxxxxxx I see:. Then? Right, then, performed the endoscopy and, we found adenoma on esophageal glands= =huh? Where? on the esophagus AGML-like lesion of acute gastric mucosa was found on the esophagus, stomach, and then in the second portion of the duodenum uh huh= =and then regurgitate esophagitis has been worsening I see (.) then? No active bleeding [was found [xxx Originally, {the patient} has been suffering from alcoholism, chronic alcoholism uh huh And then, living by himself, well:, (2.0) {laughter} he vomited blood or, we:ll, his neighbour found him lying on the bed I see (.) then? {sound of x-ray films} ninety-something? Wrong. Sixty years old? We:ll and then, uh:= =hematemesis and alcoholic= =yes, [xxxxx [has not taken care of himself right, right= =yeah. Then vomited blood (2.0) and collapsed right The:n, was found by someone


88 50 JD 51 SD 52 JD 53 SD 54 55 JD 56 SD 57 JD 58 SD 59 JD 60 61 SD 62 JD 63 (2.0) 64 SD 65 JD 66 (3.0) 67 SD 68 JD 69 SD 70 JD 71 SD 72 JD 73 74 SD 75 JD 76 SD 77 JD 78 79 SD 80 JD 81 (2.0) 82 SD 83 JD 84 SD 85 TL1 86 JD 87 88 89 SD 90 (2.0) 91 TL2 92 SD 93 TL2 94 95 SD 96 TL2 97 98 99 100 SD 101 TL2 102 103 104

Orders of Ordinary Action right, we:ll brought by the ambulance right Yeah. The:n, took a peek from the esophagus (.)down(.), found hemorrhagic lesion = =yes What else was found? what else, we:ll, leukocyte had increased to 26,500 Hu::::n Also, sodium was 115, potassium was 5.0, colloid was 80(.) and BUN had been increased to 22 How about creatinine? creatinine was 1.0 How was hemoglobin? Hemo[globin [Hemoglobin was 14.4, hematocrit was 42.3 Yap Then, had polyuria polyuria? Yes, urine volume was 600ml per hour How about osmolarity? os, osmolarity was 120, serum osmolarity was 110. Then, performed vasopressin test and urine osmolarity, we:ll, had increased. So [central [How much increase? well {cough}77 before the test and 430 after huh The diagnosed as central diabetes insipidus. (1.0) then, currently continues to use desmopressin through nasal twice per day How about his consciousness? Consciousness, we:ll, has been getting better, I think. And, currently it’s 1-3 Haven’t you intubated him? Intubation, yes, performed extubation, [performed xxxx [well= We had intubated and extubated the day before yesterday But today, PaO2 has decreased to 45 so (2.0) we:ll we also increased O2 to 7 litre. If that doesn’t bring about any improvement, we have to think about intubation again Yeah, intubation, well, what (1.0) is this guy anyway? Originally came for hematemesis and= =yes consciousness disorder but, when he came here, uh, as for the hematemesis, there wasn’t that much bleeding= =huh= =the rest was the respiratory failure, uh: when he came, gas was, PaO2 was, uh, 46 or 50 or something like that, and then we immediately intubated him, and respiratory condition had improved, so we extubated, but seems like {the patient} has chronic respiratory failure, emphysema or something like that [uhn … It was like emphysema and seems like the breathing condition was bad from the beginning, and x-ray film taken at the arrival to the hospital showed a shadow of invasion at the lower part of the right lung, so may be that makes the breathing condition [worse

Doctors’ Practical Management of Knowledge 105 SD 106 TL2 107 108 109 SD 110 TL2 111 112 113 114 115 116 SD 117 TL2 118 119 SD 120 JD 121 SD 122


[uh-huh that has improved little bit under the intubation, so we extubated once, but still the breathing condition is not good because he cannot expectorate. And now the breathing condition has been= =hu huh= =worsening so we are thinking of intubation. And as for consciousness disorder, hyponatremia, well, hyponatremia and a kind of central DI, 62 at the time of arrival, now it has been up to around 133. Consciousness level was once up to the first half of single digit, but now somewhere around 1–3 to 2–10. uh: {cough} one possible reason for this we can think of is that (2.0) there’s something wrong with the brain or= =uh huh?= =or, since it took only 24 hours or so to recover from 112 to 135 this time, and then suddenly paroxysmal consciousness disorder xxxxx yeah, we::ll, if, it’s central DI hu huh? yeah, we::ll, if, it’s central DI and has been polyuria, why {he} became hyponatremia? That doesn’t make sense, does it?

This page intentionally left blank

Chapter 6

Auspices of Corpus Status: Bibliography* as a Phenomenon of Respecification Andrew P. Carlin

Introduction This chapter1 is part of a project2 looking at the constitution and organization of disciplines and interdisciplinary studies, by examining bibliographies*3 and reference lists as disciplinary artefacts.4 The ethnomethodological approach that informs this project involves a focus upon bibliographies* as phenomena of ‘order’, in other words as phenomena to be investigated in terms of their availability to understanding as academic objects constituted by the practical textual reasoning in and through which they are assembled and the practical theorizing involved in their use.5 Bibliographies*, and my commonsense reading of them, form the topic and data of this chapter. In the first part of the chapter some general considerations are presented concerning the status and use of bibliographies* as objects of practical academic theorizing. In the second part, these considerations are illustrated through examination of particular ‘cases’ of bibliographic reasoning – including Garfinkel’s ‘corpus’ and Garfinkel’s own reasoning concerning the ‘ethnomethodological corpus’. In brief, I 1 Presented at the International Institute for Ethnomethodology and Conversation Analysis conference ‘Orders of Ordinary Action’, Manchester Metropolitan University, 9–11 July 2001. I am indebted to Rod Watson for discussing these issues with me, and to Dave Francis for editorial suggestions. I thank Harold Garfinkel for encouragement and permission to use unpublished manuscripts. Ian Cornelius, Chris Hart and Roger Slack gave valuable comments on sections of this chapter. 2 Funded by H.W. Wilson Foundation. 3 Following Garfinkel, I spell bibliography* and bibliographies* with an asterisk to denote that I am referring to a phenomenon of order, to members’ practical achievements. 4 On the feasibility of interdisciplinary work, see Garfinkel (1977: 9–17). His reservations are attuned to preserving the logical coherence of analytic programmes. The strong version of interdisciplinary work is Garfinkel’s notion of ‘hybrid’ disciplines (Watson, 2003). 5 ‘Bibliography’ constitutes an area of knowledge or ‘field’ (Hirst, 1972). However, its phenomenon is not bibliography* but the description of textual materials (Rietz, 1974). This project breaks new ground in studying bibliographies* – how they are assembled and used as ‘academic’ texts – as products of members’ practical reasoning.


Orders of Ordinary Action

shall make reference to Garfinkel’s recent work to show how the local organization of bibliographies* is a phenomenon of ethnomethodological (EM) respecification. Bibliographies* and Citation Analysis Bibliographies* and reference lists have been used as ‘data’ in the service of different research programmes. Analysts have worked to extrapolate quasi-sociological findings from bibliographic lists, or have advocated the possibilities of such extrapolations. Counting bibliographic items is known variously as ‘bibliometry’ and ‘citation analysis’. Bibliometric studies are based on citation indices compiled by commercial enterprises, such as the Institute for Scientific Information. Note that the citation index provides the base statistics from which citation analysis is extracted: for particular fields or disciplines (Case and Higgins, 2000; Prabha, 1983; Shadish et al., 1995);6 or for particular locations (Cano, 1999). Citation analysts use indices to identify types of citation practices; for example, ‘direct citation’ (Garfield, 1955), ‘bibliographic coupling’ (Kessler, 1963), ‘co-citation’ (Small, 1973); to infer predicates such as persuasion (McInnis and Symes, 1988) and rhetoric (Cozzens, 1989) in citation; and to infer ‘motives’ for citation (Garfield, 1977). Citation Analysis as Practical Reasoning Citation-analytic studies investigate research networks (Crane, 1972) as though these are textually incarnate. That is, bibliographic ‘data’ are used to provide access to socially organized activities; in other words, treated as ‘windows on the world’ (Watson, 1997b: 81) of academic or discipline-specific cohorts. There is thus a resemblance between citation analysis and forms of sociological theorizing, in the use of documentary materials to produce ‘sociological reconstructions’ (Sacks, 1999). With varying levels of sophistication, citation-analytic studies are involved in the counting of bibliographic references (citation analysts cannot count what are not in bibliographies*). Citation analysis counts the references recorded in bibliographies*, which are both (1) ‘procedural’7 and (2) ‘published’ outcomes of research. The first relates to members’ work in compiling the bibliography* and how references are implicated in the practical activities8 of research; also to the occasioned nature of references and incorporating readings recommended by colleagues; in other words, the intersubjective and collaborative nature of academic work (Garfinkel and Sudnow, 1972). The second assesses how reviewers’ decisions affect the placement 6 EM has not been excepted from such treatment (Mullins, 1973). 7 Where ‘procedural’ refers to the practical activities of participants (Garfinkel, 1996). On the practical work of compiling bibliographies (that is, how bibliographies constitute Lebenswelt pairs), see Carlin (1999). 8 Such as discussing research projects with colleagues, the collaborative, social accomplishment of studies (Garfinkel and Sudnow, 1972), such as incorporating colleagues’ recommendations and recommended readings.

Bibliography* as a Phenomenon of Respecification


of an article (for example, by suggesting, under the terms of the scope of the journal, that a paper is not suitable for publication in that journal and to seek publication elsewhere). As part of the peer-review process, an anonymous reviewer may suggest that a particular discussion – therefore, a particular reference or set of references – is incorporated or deleted. What readers (including the readership of the journal but also readers at commercial indexing and abstracting services) see, then, is the version that went to press, which includes an adjudicated set of references, rather than the workings of the paper. What is published constitutes ‘the official record of science’ (Cronin, 1984: 13, 20), but the official record incorporates reviewers’ theoretical commitments, idiosyncrasies, preferences, determinations of relevance, assessments of quality and originality, etc. As a consequence, citation-analytic studies neglect the socially produced character of bibliographic ‘data’ – the practical activities of academic work, and reviewers’ decisions apropos publishable work. In that this is the case, bibliographies* can be treated qua official statistics, and citation analyses be subject to the various criticisms of studies that rely upon official statistics (Douglas, 1967). To adapt an argument on the use of official statistics on deviance (Kitsuse and Cicourel, 1963), citation analysis holds that citation indices, which are based on bibliographies*, reflect and reproduce academic or ‘citation behaviour’ rather than incorporating social-organizational, practical decisionmaking activities. That is, the analysis attributes motives to members (for example, why authors select certain references in preference to others) rather than explicate how members make attributions of motive (Watson, 1983). Citation indices, as aggregates of citations from bibliographies*, exclude bibliographic (practical) reasoning, such as authors’ preferences in deleted sections of original manuscripts, rejected submissions, citations to bibliographic items (book chapters, conference papers) not covered by indexing services. However, citation indices incorporate members’ decision-making activities (authors’ and reviewers’ determinations of relevance, bibliographic items covered by indexing services). These are social accomplishments. Respecification of bibliographic methods topicalizes and makes available aspects of social organization to rigorous inquiry. The ‘incompleteness’ of citation indices has not precipitated consideration of foundational issues in the mapping of academic trends, which are circumscribed as methodological troubles. Whilst citation indices are incomplete, for example in terms of the coverage of literature included, they are ‘complete enough’ for the purposes of citation analysis. As one commentator says, the citation indices published by the Institute for Scientific Information ‘include as original entries only those articles published in a selective list of peer-reviewed journals and their accompanying citations’ (Martens, 2001).9 That is, the records upon which citation-analytic studies are based are, ab initio, incomplete. This observation is made not by a critic of citation analysis but by one of its advocates. However, having identified a pitfall of the citation indices or ‘official statistics’, Martens fails to draw out the implications of the compilation of indices for doing citation-analytic studies. Indeed, Martens argues that the selectivity of sources used to compile the citation index heightens 9 ‘Journals covered by the index are chosen by advisory boards of experts in each of the topics represented’ (Garfield, 1970: 669).


Orders of Ordinary Action

the authority of a citation. A limitation of the official record is transformed into the strength of analyses based on the record: ‘Citation indexing operates on the assumption that citations have prima facie equal value … In reality, of course, some citations are ‘more equal’ than others. … However, if the assumption … is conceded, then the benefits flow freely’ (Cronin, 1984: 26–7). The logic of this argument is as follows: whilst the data are flawed from the outset, ‘let it pass’ (Garfinkel, 1967a). Potential pitfalls and caveats are duly footnoted, before returning to the ‘official record’. In response to objections that citation analysis produced particular biases, Garfield maintained that perceived problems were chimerical, or at least trivial; the citation index ‘allows you to work around [the problem] and for all practical purposes to ignore it’ (Garfield, 1977: 282). Problems to which Garfield responded include the identification of future awards, difficulties caused by the ordering of authors’ names on articles and mistakes caused by conflating different authors with similar names into a single author. For Garfield, such pedantry would eventually turn out to be of little consequence in comparison to the power and scope of the cumulative citation index. Garfield could accommodate such matters arising because they do not address foundational issues in the compilation and use of citation indices. These methodological troubles of citation analysis resemble those recognized more widely within the social sciences, for which ‘remedies’ are sought. As against conventional methodological literatures in social science, which attempt to formulate such remedies, EM studies have shown that such ‘troubles’ are constitutive of social science practice. Thus studies that highlight mathematical and procedural difficulties with measurement (among others, Cicourel, 1964;10 Garfinkel and Bittner, 1967a; Rose, 1997) provoke the asking of foundational questions of techniques such as citation analysis and the production of citation indices. What are data, what constitute data in the first place? What is selected as a datum (Garfinkel, 1949)? What does a datum ‘represent’, or, what measures or forms of counting are incorporated within another measure (Sudnow, 1965)? EM studies respecify these methodological problems as practical accomplishments of disciplinary work, and suggest that analysts treat data as glosses rather than indicators, and the compilation of data as ‘glossing practices’ (Garfinkel and Sacks, 1970). Bibliographies* as Accomplishments Bibliographies* may be invoked as proxies for critical literature reviews and bibliographic sets (Dewe, 1995: 150; Glaser and Strauss, 1968: 1), or accorded corpus status (Coulter, 1974: 120; Farrell et al., 2000: 8). Although bibliographies* have been used in the compilation of citation indices, they have not been studied as topics in their own right (Zimmerman and Pollner, 1971). This chapter is a methodological recasting of bibliography*, treating the production of a bibliography* not as a 10 The relationship of this work to the EM programme (that is, corpus status) is regarded as problematic (Coulter, 1974; Lynch, 1991). Suggesting improvements to research methods is not representative of recent EM, which demonstrates that methods produce the phenomenon of investigation (Bjelić, 1995).

Bibliography* as a Phenomenon of Respecification


self-evident but a methodic practice, a member’s achievement. An EM approach to bibliography* involves treating ordinary reference listings as ‘anthropologically strange’ (Garfinkel, 1967a: 9) in order to problematize bibliographic practices.11 To paraphrase Sacks (1984: 27), I happen to have a particular bibliography*, I choose to look at it more carefully, and sometimes find something of interest to say about it. Taking the local organization of bibliographies* as a topic clarifies Garfinkel’s notion of corpus status, which refers to the placement, and the placing, of literature on a topic or within a field. The following section takes Garfinkel’s corpus as an example of practical reasoning in the recognition of a corpus. The Phenomenon of ‘A Literature’ as a Practical Accomplishment The bibliography* that constitutes the Manual (Garfinkel, 1976), that constitutes the EM Catalog (Garfinkel, 1993), that constitutes ‘Ethnomethodology’s Program’ (Garfinkel, 1996), that constitutes ‘The Corpus Status of Ethnomethodological Investigations’ (Garfinkel 2001),12 is perspicuous because Garfinkel explicates the auspices of its assembly. It is not that Garfinkel’s bibliography* is special, or superior to other bibliographies* in any way, that it is subject to analysis. Rather, it is considered here because it makes available the production of bibliographies* as members’ practical actions. It is observable from Garfinkel’s bibliography* that items share a topical coherence (Lynch, Livingston and Garfinkel, 1983: 231) or family resemblance: ‘The accompanying bibliography, Studies of Naturally Organized Ordinary Activities, lists members of a corpus of studies in which the in situ production of an activity’s orderliness was topical’ (Garfinkel and Sudnow, 1975: 1). Garfinkel assembles a bibliography*, a corpus, exhibiting the uniquely referring characteristics of EM studies; EM’s phenomena are instructably observable from the corpus status of works that Garfinkel assembles in the EM programme, such as assigning corpus status of EM studies under the auspices of the ‘unique adequacy requirement’ (Lynch, 1993: 302). Garfinkel, as compiler of the bibliography*, bibliographicizes the corpus that constitutes the alternate literature, which hold corpus status as studies of members’ lived experiences.13 Whilst assembling a bibliography* Garfinkel provides the auspices of bibliographic practices. Likewise, establishing Garfinkel’s corpus as an ‘EM’ corpus is subject to members’ practical reasoning procedures. Bibliographies* of EM studies do not include a ‘complete’ bibliography* of Garfinkel’s work. A short story of a racial incident on a public bus (Garfinkel, 1946) is not itemized (Fehr and Stetson, 1990; 11 In other words, taking compiling bibliographies* as a ‘scholarly activity that requires investigation as a phenomenon in its own right’ (Watson, 1984: 361). 12 This article is a revision of the Introduction to Ethnomethodology’s Program: Working Out Durkheim’s Aphorism, by Harold Garfinkel, edited by A.W. Rawls, Rowman and Littlefield Publishers (2002). 13 Garfinkel (1991: 19) asserts that, rather than his ‘doing cohorting’, ‘the list testifies to a division of labour, and that the list is not a bibliography of all eligible, possible, relevant, or consequential work’.


Orders of Ordinary Action

Jules-Rosette, 1985: 39). Although Garfinkel wrote it, why should it be located within the EM corpus, that is, accorded corpus status within EM?14 The omission of particular items from the bibliography* suggests that, in the practical work of compiling bibliographies*, compilers do not recognize particular items as ‘EM’. Not including items, de facto, denies each item corpus status. Whilst the published account of clinical features of the ‘Agnes’ case (Schwabe et al., 1962) may be disregarded on the grounds that Garfinkel is not listed as a co-author, an early discussion of passing (Stoller et al., 1960) is perhaps more problematic to classify.15 This suggests further that bibliographies* may be used as loci for the accordance of corpus status; in other words, how bibliographies*, and inclusion in bibliographies*, denote corpus status. One example might be assembling a list of ‘works about Goffman that ought to appear in any Goffman bibliography’ (Waksler and Psathas, 1989: 177, emphasis supplied). Instantiations of Corpus Status In his recent work, Garfinkel outlines the auspices of corpus status between formal analytic (FA) studies and ethnomethodological (EM) studies. The auspices of corpus status that he identifies are positions vis-à-vis ordinary action:16 incorporating members’ practical reasoning procedures as resources, versus taking practical reasoning as topics for analysis. These are discrete and alternate bodies of literature. In this way, then, Garfinkel provides an instantiation of corpus status – the position of EM literature vis-à-vis FA literature, the analytic priority given to FA studies over the accounting for ordinary action. Further, the corpus status of the formal-analytic plenum is visually available from formal-analytic studies and attendant bibliographies*. The asymmetry of FA and EM literatures provides a perspicuous example of corpus status. Garfinkel’s practical reasoning about the corpus status of EM provides an opportunity for reflection upon the

14 After all, ‘[f]rom the fact that [both] these works were written by the same man … does not follow that both works belong to the same science …’ (Sorokin, 1931: 27). 15 The work of William Foote Whyte (1955) holds equivocal corpus status, as it is indirectly (and directly) associated with symbolic interactionism and the Chicago School (see Silverman, 1993, passim), when his studies are derived from the Arensberg/Chapple tradition. See Watson (1994) for criticism of interaction analysis in this tradition (Collins and Collins, 1973), demonstrating that studies of conversation are not, ipso facto, ‘conversation analysis’: witness the library-specific co-classification of ‘lectures on conversation’ (Sacks, 1992; Zeldin, 1998). 16 The focus on members’ practical actions is maintained without regard for FA standards of reviewing literature on the purported topic, whereby EM studies are not characterized by literature reviews, which displace the study of locally organized phenomena. Indeed, this is formulated as a study policy: ‘carrying out empirical descriptions of social facts while exercising Ethnomethodological indifference to the corpus status of established literatures’ (Garfinkel, 2001: 4). Maintaining EM indifference involves, among other things, ‘withholding the corpus status of formal analytic descriptive facts’ (Garfinkel and Wieder, 1992: 187).

Bibliography* as a Phenomenon of Respecification


methodic practices in and through which the phenomenon ‘a textual corpus’17 may be accomplished. Among the methods by which corpus status is accomplished, a frequently used one is formulating the ‘paradigm text’: By now it is a truism that John Porter’s The Vertical Mosaic is the magnum opus of Canadian sociology (Heap, 1974: viii). The single most significant piece of work on gaze interaction from this point of view is that of Charles Goodwin’s Conversational Organization (Kendon, 1990: 89). The most illuminating discussion of secrets and secretiveness in the sociologic literature is contained in The Sociology of Georg Simmel … [Kurt Wolff, ed.] (Bittner, 1977: 93). The literature on skid row is voluminous. The classic in the field is Nels Anderson, The Hobo… (Bittner, 1967: 704). The classic account of English conditions up to the 19th century is C.J. RibtonTurner, A History of Vagrants and Vagrancy and Beggars and Begging … (Bittner, 1967: 706).

Another method by which corpus status is allocated to particular items is via inclusion in a ‘reader’; that is, members’ (such as editors’) reasoning allocates corpus status within (for example, Goffman’s) output.18 Asserting that a collection represents Goffman’s ‘most important writings’ (Lemert and Branaman, 1997: x) underscores the received wisdom of Goffman’s sociology being of a piece with the diacritically marked ‘Goffman’ that glosses a recognizable set of concerns (Lemert and Branaman, 1997: ix). So it is a collection of readings of the distinctive style that readers expect – that exhibit the (morally required) ‘Goffman-ness’ of Goffman – which are accorded corpus status in these terms. Accordingly, Goffman’s (1951) paper is mentioned en passant (Lemert and Branaman, 1997: xxi) but no extract from it is reprinted, nor is it listed in bibliographies* of his work (Burns, 1992; Drew and Wootton, 1988: 280–281). Bodies of work are temporally marked but the corpus status of work is not subject to the same temporal organization as the corpus. In these terms, Garfinkel’s corpus is seen to have started with the publication of Studies in Ethnomethodology,19 a collection of papers written over a number of years (Garfinkel, 1967a: ix). Owing to the internal differences between 17 In his attempt to produce a ‘new’ approach to the history of ideas, Foucault (1989) trades upon the existing frameworks of disciplines and concepts whilst seeking to revitalize them, bringing a different set of terms to a problem rather than challenging the bases of the problem. What Foucault offers, then, is not a foundational respecification of the organization of knowledge but a reaffirmation of the corpus status of debates; that is, a ‘new moment’ (Button, 1991a). 18 For the local organization of Goffman’s work ‘as a corpus’, and how bibliographies* are defined, see Carlin (2004). 19 See, for example, Zimmerman’s (1987) contribution to a symposium on Studies.


Orders of Ordinary Action

them, which resulted from the time-span in their production, certain chapters ‘do not exemplify the book’s program and policies with the same clarity as do the sections written later’ (Lynch, 1993: 12). In effect, Lynch is privileging some chapters over others – according some chapters greater corpus status within the EM programme. Another instantiation of corpus status, from a study of aural (rather than textual) materials, explicates members’ practical reasoning outside ‘academic’ contexts. ‘Hearing the Blues’: Corpus Status and Membership Defining a ‘corpus’ and attributing corpus status are members’ practical accomplishments, which are not limited to textual materials. In their membership analysis of musical traditions, Hatch and Watson (1974) do not categorize particular pieces of music or ‘musical items’ into particular genres but remain neutral in regard to the relative merits of individual pieces of music. Whether one musical item is ‘superior’ to another musical item, and in what respects, is not (or should not be) an issue for analysts.20 Rather, the issues involve the definition and recognition of a musical item as a member of a particular tradition or category, by those members/hearers who are competent to recognize and define it as such and as a condition of being accorded such competence by their consociates (Bittner, 1965: 247). Differences between types of blues and folk songs may not be immediately apparent, whereby ‘this potential confusion, among many others, is used by those in the blues community as an “indicator” of competent membership’ (Hatch and Watson, 1974: 163). Members are recognized as competent hearers of blues music through their provision of accountable definitions of musical items; that is, members’ ability to differentiate between musical traditions and allocate pieces of music to ‘recognized’ categories – the categories recognized as appropriate or ‘correct’ categories by other competent hearers. So Hatch and Watson provide loci (items, categories, items that are members of categories, and hearers’ rules for categorization) for the recognition and accordance of corpus status. Competent hearership involves the recognition of categories and the distribution of musical items within categories. The next section takes competent membership as academic practice, namely competent viewing and competent reading as constituent features of research projects. It concerns the phenomenon of ‘a gap in the literature’. ‘Begging’ is used as an example to enable us to examine this phenomenon in its accomplishment. 20 That is, maintaining a position of ‘ethnomethodological indifference’ (Garfinkel and Sacks, 1970) towards the topic of research. Such an analytic procedure distinguishes the ‘texture’ of commentaries on music: compare Hatch and Watson’s analysis with Dettmar’s (1999) appreciation and dismissal of forms of music. His remark that an album ‘is catalogued, by those who devise and believe in such labels, as gothic rock’ (Dettmar, 1999: 41–2) appears to be a complaint about the particular category it is placed in rather than categorizing per se. If nothing else, espousing postmodernism (as Dettmar does) seems to give a writer the leeway to be self-contradictory without reproach.

Bibliography* as a Phenomenon of Respecification


How to Do ‘An Alternate Literature’ In order to examine the methods of practical textual reasoning involved in establishing a corpus, a research project concerned with the topic of ‘begging’ is now discussed. From an EM point of view, studying begging as practical action avoids theorizing begging, classifying begging, seeing begging as an ancillary activity not as a topic in its own right, talking about begging under the rubric of external standards, and conceptualizing begging through a perspective by incongruity. Such procedures, entailing particular bibliographic sets, lose the phenomenon. Yet wider discussions of begging are implicated by the corpus status of materials of discernible relevance. In 1998, I collaborated on an observational study of public space in a large European city.21 Within a relatively short period of time and without any preconceptions of what we might find, we began to focus on a range of ‘publicly available’ phenomena; that is, social phenomena which are ‘there for anyone to see’ (MacAndrew and Garfinkel, 1962: 252). We refer to these phenomena as ‘visible poverty’, how certain people are, indeed make themselves, visually identifiable and recognizable to other people on the street as beggars, ‘doing’ begging (Carlin, Evergeti and Murtagh, 1999). Our observations contradicted the assertion that ‘most [beggars], paradoxically, behave in ways calculated to make themselves as unobtrusive as possible’ (Dean and Gale, 1999: 21). We argue that the procedures used by beggars are orientations and adaptations to the street, whereby this very ‘unobtrusiveness’ is dependent upon and fashioned out of the internal organization of the street. Silent or whispering begging is unobtrusive until it is regarded as an accountable activity within a locally organized setting, which is ongoingly produced and reproduced by passers-by in the street. The importance of this conceptual switch – between ‘plain ordinary theorizing’ (Garfinkel and Wiley, 1980: 9) and the study of practical activities – has a secondary effect on bibliographies* and literature reviews:22 that is, an alternate corpus of literature becomes relevant when moving from ‘studies about’ to ‘studies of’ members’ practical actions. Following our initial period of observation, I embarked upon the task of collecting a body of literature to review. The noticeable feature of doing a literature review on begging was the shortage of material that was available. A search of electronic databases, using the keywords ‘beggar’ and ‘begging’, illustrated this lacuna. Keywords need to be synonyms, such as ‘panhandling’, or broader subject-headings, such as ‘homeless’ or ‘homelessness’. Keyword searches in journal indexes have to be broadened to include supposedly (or intuitively) relevant subject-headings. 21 The other members of the research team were Ged Murtagh and Venetia Evergeti. 22 As Robillard (1999: 64) observes, ‘If I reviewed the literature … I would still be theorizing.’ A note at the beginning of Studies in Ethnomethodology clarifies that for EM it is not the reviewing of research findings that is important but the logical presuppositions and suitability of methods to the topic of inquiry that supersede such reviews. Garfinkel (1967a: xiii–xiv) is aware that during the time between the study and its publication (Garfinkel and Bittner, 1967a), further articles became available. When the need for a respecification of method and measurement has been established Garfinkel does not find it necessary to list further studies that lose the phenomenon.


Orders of Ordinary Action

Obtaining ‘relevant’ material for the project on begging was problematic: not all commentaries, which were purportedly ‘about’ begging, actually looked at begging in its particulars. A further lacuna in the extant literature on beggars and begging is the study of begging as a practical activity. As interaction, begging is a ‘wild’ phenomenon,23 in that its real-world, real-time production is a ‘first-time through’ (Garfinkel et al., 1981) occurrence. The analyst requires means of repeatedly and closely analysing its endogenous structures, for example, video-recording. This epistemological-methodological move was introduced with the development of conversation analysis: The studies by Sacks, Schegloff and their colleagues have demonstrated and settled the matter that where technical order-production phenomena are concerned, no issues of the work of conversational orderliness can be adequately identified by an analysis of reportage, and none can be formulated or solved (Garfinkel, 1976: 63).

For all the literature bases in which beggars and begging were found, the study of begging as a concerted activity by members was absent. In Garfinkel’s words, there was ‘a gap in the literature’. The recognition that there was a gap in the literature derives from the ‘studies of work’ programme. The gap in the literature is the absence of descriptive studies of occupational practices. The specific origin of the gap in the literature was the observation of introductory lectures in chemistry at the University of California, Irvine, in a report (Garfinkel and Sudnow, 1972) detailing lecture-specific work: lecture-specific phenomena that were available for investigation and were observable and describable; that is, accountable. Garfinkel and Sudnow (1972) initiate investigation into the phenomenon of the availability of the constitutive activities of lecturing. In considering this phenomenon, they identify it as unavailable from extant discussions. Noting that accountable, lecture-specific phenomena (that were available for investigation) were not and could not be found as topics for analysis in existing work on occupations, they identified a further phenomenon for investigation: ‘Th[is] phenomenon consists of the combined availability of endlessly numerous constructive analytic studies of occupations, in endless variety of detail, cost, and impact, accompanied by the practical absence of detailed, technically descriptive studies of work’ (Garfinkel and Sudnow, 1972: 20). In the first instance, this was observed to be an ‘eligible phenomenon’. ‘Eligibility’ refers to the nature of ‘the literature’24 as a topic for investigation in its own right; 23 Garfinkel, in Robillard, 1999. Reviewing literature locates a study in relation to other studies. However, literature reviewing in the service of studies based on observations of wild phenomena ‘cannot by any further work of reading and writing about them be made stronger’ (Garfinkel, 1976: 64). 24 ‘The literature’ and ‘in the literature’ are indexical expressions. ‘The literature’ is a gloss for works that are relevant, or have been relevant, to a particular theme or topic. There is an ‘essential vagueness’ about these phrases, which lend coherence to an undefined (and indefinable) body of works, the relations of which are discernible as context-dependent bibliographic items.

Bibliography* as a Phenomenon of Respecification


how ‘the literature’ can be studied as a phenomenon of order; that there is a missing literature; that the missing literature is constituted by the absence of descriptive studies of work (where descriptive studies of work is formulated by Eric Livingston’s distinction (Garfinkel et al., 1981: 132) between studies about work and studies of work); that descriptive studies of work exhibit the uniquely identifying details of that work; that EM studies are ‘in disjunctive contrast to the phenomena of order that are described in the Literatures of the social science movement’ (Garfinkel, 2001: 7). I do not propose a ‘genealogy’ (Lynch, 1998) of EM in these terms, nor to read Garfinkel’s work retrospectively as demonstrations of eligibility; but it is observable that Garfinkel himself would continue to explicate the eligibility of literature in later studies, giving increasing prominence to the issue.25 The gap in the literature is referred to (Heritage, 1984: 298ff.; Lynch, 1993: 270ff.). However, as an axial theme of Garfinkel’s later work it has received comparatively little attention, is treated as an analytic aside, a consequence of the respecification of the social sciences, but not taken as a focal topic. From the point of view of the researchers (Carlin et al., 1999), the ‘gap in the literature’ comprised the fact that begging as an activity and as a subject of study has been concealed (or at least subsumed) under other substantive subject areas. Begging can be found within literature in ‘other fields’ as a sub-phenomenon of other, more focal phenomena, such as alcoholism and substance misuse, homelessness and mental illness, skid rows and vagrancy. The ‘gap’, then, consists in the fact that, in these literatures, begging is seen as a product of particular circumstances, not a topic of study itself. Furthermore, searching discipline-specific literatures renders unavailable potentially relevant materials, which are not necessarily recognized as being relevant because of their topic or discipline-specific nature. An example of this comes from the literature on ‘addiction and substance use’. ‘Cottons’ are absorbent materials – cigarette filters, cotton wool, fragments of sponge – which act as filters in the process of drawing heroin solution from the cooker into the syringe. ‘Cotton bandits’ sustain their heroin addiction by begging for cottons, rather than begging for money (Bourgois, 1998). Although Bourgois’ article is relevant to a project on begging, it does not feature within the so-called ‘literature’ because begging is approached a priori as begging for money. This is a double irony of FA in that practitioners do not account for ordinary action yet ignore other works ‘about’ the phenomenon because these works are not recognized as relevant to the phenomenon; that is, they are not recognized to be ‘part of the literature’. EM as Practical, Bibliographic Reasoning It is of no news to ethnomethodologists, of course, that Harold Garfinkel is a practical actor like everyone else. Nor is it news that EM, like all academic inquiries, is a practical activity. EM does not have any transcendent quality or exemption from 25 As I have shown above, the eligibility of literature is an outcome of the studies of lectures, yet identifying the absence of descriptive studies of discipline-specific work would introduce later studies (Garfinkel et al., 1989).


Orders of Ordinary Action

analysis of practitioners’ practical reasoning. In one of the ‘clinic’ studies, which examines the statistical transformations of the socially organized phenomenon of assembling a cohort, it is observed that The choice of a theory is not only unavoidable but is critical to the task of deciding what will count as a finding. This election of theory furnishes the researcher the grounds for deciding what the results of his statistical evaluations are to stand for as findings (Garfinkel and Bittner, 1967a: 242).

This observation is not limited to mathematical or statistical representations but is pertinent to all forms of social inquiry, including EM investigations. There are no ‘time-outs’ or exemptions for analysts in adopting or subscribing to theoretical and methodological options. Furthermore, this observation may, mutatis mutandis, be brought to the study of bibliographies*. That is, as phenomena of order, the study of bibliographic options is the study of practical reasoning, which applies to all academic inquiries. Conclusion ‘Corpus status’ articulates a thematic concern, which glosses the preservations and transformations in EM investigations from its early studies, the identification of a ‘gap in the literature’ and the studies of work programme, and the programme of respecification. Garfinkel uses the notion of corpus status as an explicative device to clarify distinctions between EM and FA. The auspices of corpus status are exhibited by adherence to methodological principles, such as ethnomethodological indifference and the unique adequacy requirement, and constitute the nature of EM and FA investigations as ‘alternates’. This chapter has provided instantiations of corpus status to exemplify the terms of its use. These include reference to an observational study of public space and the subsequent practices of literature reviewing. The difference in the phenomena of interest – between begging as substantive topic and begging as a situated activity – entailed different sources. A search for references to begging as a course of action revealed a ‘gap in the literature’. Due to a ‘fundamental difference in perspective between most available studies of conduct’ (Sudnow, 1978: 153) and their EM alternates, the programme of respecification implicates alternate literatures. As shown in this chapter, corpus status is exhibited in bibliographies*. Bibliographies* are assemblages of members’ cultural methods – reading,26 writing, discerning the relevance of textual materials, and the practical activities of information retrieval – that underpin research methods as conventionally conceived (Carlin, 1999). In that bibliographic materials are occasioned by and implicated in the practical activities of research, bibliographies* are assembled for ‘organizationally relevant purposes’ (Garfinkel and Bittner, 1967b: 191). 26 The activities of reading include the back-and-forth nature of sections, such as the body of the text and bibliographies*, footnotes (Sharrock and Ikeya, 2000: 278), titles and subtitles (Garfinkel, 1967c), and the mutually elaborative nature of sections.

Bibliography* as a Phenomenon of Respecification


Documents, such as bibliographies*, are not always produced for the analytic purposes for which they are subsequently used: in their study of records at a psychiatric clinic, Garfinkel and Bittner (1967b) encountered the recalcitrant nature of clinic folders to sociological investigation. Rather than suggesting remedies for this problem or treating it as a methodological obstacle, however, they respecify the problem as a ‘normal, natural trouble’ for study in its own right. Bibliographies* in themselves do not constitute useful ‘data’ but, when transformed into records of quantifiable citations (that is, articles published in those journals included in citation indices), afford measurable ‘indicators’ of academic output and discipline-specific cohorts. Citation analysts encounter methodological troubles in transforming bibliographic materials into ‘useable data’ that incorporate the socially organized contingencies of their production. ‘Methodological’ remarks clarify the practical reasoning and ad hoc procedures used by citation analysts to ‘remedy’ these troubles. Approaching bibliographies* from within an EM framework makes available the social organization of discipline-specific inquiries (Watson, 1984), and problematizes the adequacy of textual materials as resources for the reconstruction of disciplinespecific cohorts. In so doing, EM explicates the logics of such endeavours, such as network analysis and citation analysis. The EM approach highlights foundational issues in citation analysis such as the use of textual materials produced for different purposes, and respecifies methodological obstacles as ‘normal, natural troubles’.

This page intentionally left blank


This page intentionally left blank

Chapter 7

Law Courts as Perspicuous Sites for Ethnomethodological Investigations Michael Lynch

Law courts and legal processes have been perspicuous sites for ethnomethodological investigation. This chapter discusses some of the attractions the courts hold for ethnomethodology, and then focuses on court cases in which ‘science’ and ‘scientific methods’ are perspicuous themes for investigation and interrogation. After comparing ethnomethodological and constructionist treatments of courtroom ‘deconstruction’ of expert testimony, the chapter briefly examines materials from interrogation sequences and draws some implications for further study. Perspicuous Sites For ethnomethodologists, a perspicuous site is a researchable organization of activity in which participants raise, address, and locally explicate problems, topics, and conceptual distinctions that have a central place in academic philosophy, social theory, and methodology. Typically, though not always, participants are indifferent to academic interests in the topics they address. A history of scholarship and contemporary academic debates about cognate topics, distinctions, and problems provide motives for investigating a perspicuous site, but an academic interest may or may not be reciprocated by participants. The source of the idea of perspicuous sites for ethnomethodological investigations is Harold Garfinkel, who speaks of ‘perspicuous settings’ for ethnomethodological studies.1 The idea also resonates with Wittgenstein’s remarks about ‘perspicuous representations’.2 Both Garfinkel and Wittgenstein propose that everyday practical reasoning and language-use can and should be investigated. The basis for their proposals is a recognition that philosophical and social theoretical concepts like ‘fact’, ‘reasons’, ‘motive’, ‘method’, ‘knowledge’, ‘meaning’, ‘value’, etc., draw upon, and occasionally rival, familiar everyday uses of such concepts. This recognition of ordinary usage is combined with a particular form of disenchantment with the very idea of ‘concept’. The disenchantment is with the idea that an essential core of meaning accompanies, and gives order to, contextually variable uses of terms and families of terms. When examined by reference to specific cases and contextual uses,

1 2

See Garfinkel and Wieder (1992) and Garfinkel (1991). Wittgenstein (1958), Section 122.


Orders of Ordinary Action

concepts lose their metaphysical aura and take up residence on the streets. Although familiar, and often taken for granted, such ‘ordinary’ or ‘everyday’ concept-use is not necessarily simple, transparent, or readily accessible to reflective and synthetic examination. Indeed, the ethnomethodological proposal is that the organization of everyday language and practical reasoning cannot be reconstructed as an abstract cognitive system, but that it is situated within, and indexed to, particular occasions of use. Given this background, it becomes possible to imagine an aim for investigating what Garfinkel calls ‘the ordinary society’ and its associated activities. This aim radically differs from the more common aims in the social sciences of data gathering, theory testing, or explanatory modeling. Instead of constructing ‘scientific’ models that employ specialized conceptual and mathematical resources – resources that are drawn from literatures independent of the domains of conduct investigated – the idea is to relocate topics of social-theoretical argument and debate within specific organizations of activity. One general rationale for making such a move is corrective – to dissolve the intellectual confusion and sense of hopelessness that can arise from attempts to recover social structures, or, in cognitive science mental processes, by operating on abstract representations. Ethnomethodologists also imagine more positive aims: to ‘discover’ and explicate non-thematic properties of practical reasoning and language-use which are evident, and even obvious, in the midst of embodied actions and situated social interactions. ‘Discover’ is marked with inverted commas in order to denote that, far from involving an effort to uncover something that was not previously known (or even suspected) to exist, it is a matter of disclosing what already was so apparent as to be unworthy of sustained attention and written explication. There must be a point to such ‘discoveries’; otherwise, ethnomethodology would be an endless, boring, and worthless recitation of what everybody knows (an accusation that is not unfamiliar). An important part of such ‘discovery’ is to demonstrate the constitutive role of the ‘discovered’ features of quotidian social life in the production of social order. Law Courts and ‘Science’ In research I have been pursuing for the past several years, I have been increasingly drawn to law courts as sites for investigating ‘science’, ‘scientific methods’, and the science/commonsense distinction. I am certainly not the first ethnomethodologist to be drawn to the courts. Indeed, according to Garfinkel, the very idea of ethnomethodology occurred to him in the course of a study of jury deliberations.3 In a brief, and off-the-cuff, recounting of the insight that led him to coin the term ‘ethnomethodology’, Garfinkel refers to a collaborative study of juries involving a number of prominent social scientists in the early 1950s. The researchers audiotaped the jury’s deliberations in an actual court case, and they sought to analyze the tapes in order to get access to ‘natural’ (non-experimental) reasoning. A legal ruling later prohibited the explicit use of the recorded materials by the research team, but 3

See Hill and Crittenden (1968: 6–7). The excerpt is reprinted in Garfinkel (1974).

Law Courts as Sites for Ethnomethodological Investigations


prior to that the researchers pondered how to analyze the reasoning without reducing it to the social scientists’ categories.4 This was a familiar problem for ethnographers, but Garfinkel appreciated a reflexive depth to the problem, which went well beyond anthropological efforts to represent native classifications of plants, animals, kinship relations and other substantive domains. The problem – or rather the phenomenon – was that the ‘natives’ on the jury were masters of the same natural language used by the investigators. Moreover, the jury members freely, though in a circumscribed and sometimes apologetic way, made use of terms and distinctions that were cognate with the fundamental themes and principles of social science methodology. The jurors engaged in discussion and debate about what was rational, reasonable, logically coherent, and supported by facts in the case at hand.5 In order to signal the need to do justice (so to speak) to the jurors’ discursive reasoning and reflections about possible reasons for decisions, Garfinkel coined the term ‘ethnomethodology’. For the past thirty-plus years, some of the most notable ethnomethodological studies have made use of materials from trial courts, pre-trial negotiations, police interrogations, and other routine sites of legal and criminological practice. These include Sudnow’s study of ‘normal crimes’ in plea negotiations,6 Pollner’s work on mundane reasoning in traffic court,7 Atkinson and Drew on cross-examination sequences,8 Rod Watson on police interrogations,9 and many other studies by Bittner, Pomerantz, Maynard and Manzo, Burns, Meehan, Travers, and others.10 One of the attractions of legal settings – and courtroom hearings specifically – is that they are ‘naturally organized ordinary activities’ in the sense that they are organized in a way that owes little to professional social scientific efforts to investigate and intervene in the workings of their domestic societies. The courts conduct their own inquiries into particular social events, and they deploy typifications and other general understandings of the society in which they live (and sometimes the societies in which they do not live), and they use social science rather selectively, often with considerable skepticism. While court investigations are ‘natural’ in the sense that they recur as routine constitutive activities in the societies social scientists study, the courts also are formal institutions; or rather, they include varying degrees of formality and formalization. They conduct investigations, they examine evidence, they display a concern for proper procedure and objective results, and they make use of vast bodies of documents and scholarly investigations of such documents. The procedures and practices of investigation, deliberation, and resolution in the courts are spontaneously produced in natural language, and they are situated in singular cases requiring ad hoc reasoning, and yet, at the same time, the courts make explicit 4 For an anecdote that nicely encapsulates the issue of how to analyze jury deliberations, as such, and not as an instance of generic social interaction, see the story of ‘Shils’ complaint to Strodtbeck,’ in Garfinkel, Lynch and Livingston (1981). 5 See Garfinkel (1967b). 6 Sudnow (1965). 7 Pollner (1974). 8 Atkinson and Drew (1979). 9 Watson (1990). 10 Bittner (1967); Pomerantz (1987); Manzo and Maynard (1993); Burns (2001); Meehan (1986); Travers (1997).


Orders of Ordinary Action

efforts to articulate and review reasons, to document judgments with scholarship, and to demonstrate formal consistency from one case to another. Courts, as an institution, preceded the so-called scientific revolution, and in fact they provided some of the basis for scientific investigations, tests, facts, and reasoned interpretations. However, present-day law courts in the UK, the US, and virtually all other societies that have courts, organize their investigations in a distinctive way, and while they often make use of ‘scientific’ expertise, they circumscribe such expertise within a distinct order of statutes, precedents, and case-by-case judgments. The courts, especially in the Anglo-American tradition, but also in other traditions, build in specific provision for non-specialized participation, publicly accessible processes, and transparent language. There is a seat in the audience for a non-participating, non-specialized analyst; and, increasingly, the proceedings are recorded, and even televised. In jury trials, specific provision is made to give the jury responsibility for what is called ‘the ultimate issue’: a judgment about the totality of facts in the case and how they bear upon the disputed claims. Stated less grandly, ‘the ultimate issue’ requires jurors to reconstruct a story of singular events culminating in a criminal charge or lawsuit; a story that frames relevant evidence in a way that makes sense for someone familiar with the relevant events and circumstances in the society at large from the courtroom. In principle, experts and specially trained legal specialists have a constrained part in the proceedings, and the collective ‘commonsense’ deliberation of the jury is entrusted with the ‘ultimate issue’. As numerous ethnomethodological studies have shown, the parties to a case inevitably employ ‘common sense’ or mundane reasoning, and formal proceedings are composed by discursive and deliberative practices that are not predicated on formal, legal specifications. In jury trials, such autonomy is itself an explicit property of the design of court proceedings. There can be little doubt that the courts, however well or badly they resolve disputes, constitute social order. While doing so, the courts rely upon an unspecified or underspecified ability of lay participants to comprehend and act competently in court proceedings. A study of how the courts do so is thus a study of how the society organizes itself. Ethnomethodology and Constructionism My particular interest in courts as perspicuous sites for examining ‘science’ arises from a long-standing interest in day-to-day practices in scientific laboratories and field studies. This interest, which came out of the studies of work in the sciences and professions that Garfinkel and his students conducted in the 1970s, coincided with the move toward ethnographic studies of science in social studies of science (studies by Latour and Woolgar, Knorr Cetina and others, which became closely identified with the constructionist movement).11 For those of us in ethnomethodology, the ascendancy of constructionism in science and technology studies, workplace studies, and particular programs of discourse analysis is an unavoidable preoccupation: our 11 Latour and Woolgar (1979); Knorr-Cetina (1981).

Law Courts as Sites for Ethnomethodological Investigations


work often is conflated with constructionism, seen as a precursor to it, or denigrated for being too limited in scope when compared with it. The topic of science has been one, but by no means the only, area of study that has brought to a head the differences between ethnomethodological and constructionist studies. Put very simply, the core lesson of constructionist studies of science is that science in action (science, as actually practised in historical and present-day laboratory and field situations) conflicts with the idealized versions of science that are promoted in textbooks, popular media portrayals, and other public forums. (Though the picture of science is idealized, the version of nature it promotes is ‘realist’.) Whereas idealized, ‘ready-made’ science is said to be objective, consensual, independent of cultural context and personal interest, and demarcated from common sense; actual ‘science in action’ is said to be messy, culturally circumscribed, historically contingent, and fraught with sources of uncertainty that are only partially explored. In his, by now, canonical textbook Science in Action, Bruno Latour represented a series of such contrasts with a Janus-faced figure (Figure 7.1).12

Figure 7.1

Latour’s contrast between idealized science and ‘science in action’

Source: Latour, 1987

We can understand this as an iconic version of the realist–constructivist debate. Properly understood, this is not meant to contrast an idealized (but false) representation of science with an empirically warranted picture of actual science. Instead, the Janus figure represents a dialectical opposition that is both internal to scientific discourse as well as to popular portrayals of science. Latour proposes to break down the opposition by showing that a relentless, only occasionally exposed, set of ‘hybrid’ techno-social practices precedes and constitutes the ‘purified’ (ideal) representations of nature and society associated with the natural sciences and their social science emulators.13 It would be easy to put a familiar constructionist irony in the mouths of Latour’s Janus figure: Where the older, bearded figure states that scientific methods are a matter 12 Latour (1987). 13 Latour speaks of ‘purification’ in several publications, but his most extensive theoretical treatment is in Latour (1993).


Orders of Ordinary Action

of rigorous procedures, guided by rules of method and well-tested protocols, the brash young figure would tell us that methodological canons are routinely ignored, that methods reports are rhetorical after-the-fact constructions, that scientists improvise in ad hoc trial-and-error ways, that they vary standard protocols according to the circumstances, and that methods are only partly rationalized in terms of physicochemical principles. This ironic version has currency among scientists as well as social studies of science, as scientists often acknowledge that ‘real’ scientists do not take methods reports literally.14 So where does ethnomethodology come into this picture? It should hardly be a surprise to anyone in this field that the moment-to-moment production of scientific experiments differs in many important respects from the version of the ‘same’ experiments in a written protocol. The interest is in the question, ‘So, what do they do?’ And the answer does not necessarily take the form of an ironic contrast between actual ‘messy’ practices and idealized versions of scientific methods. In so far as methodological protocols are of interest, the investigative task is to examine the protocols together with their uses, as instances of the more general phenomena of what Garfinkel calls instructed actions,15 or Wittgenstein characterizes as actions-inaccord-with-rules. Accordingly, the ‘incompleteness’ of instructions is not a failing, because it arises from a prior abstraction of the rule from the activities in, and as of which, it provides instruction. The theme of instructed actions invites attention to particular circumstances in which rules (and a whole array of forms of instruction) are incorporated into singular cases of concerted action. In some circumstances, the ‘problem’ of acting in accordance with rules can arise in a mundane and situated way. We are all familiar with the difficulties of reading general instructions for assembling or repairing a mechanical device, adapting a standard recipe to the ingredients we have at hand, and learning to play a game with rules. The programmatic invitation for those of us who take ethnomethodology seriously is to examine, systematically, how rules, directions, programs, and other formatted instructions are used in particular cases. The tension between ethnomethodological and constructionist studies extends to studies of science in the courts. In several areas of contemporary law, the courts perform second-order inquiries in which they trade upon conceptions of scientific method, administer distinctions between science and common sense, and make rulings about what counts as ‘generally accepted’ bodies of scientific fact and reliable procedure. This goes for landmark Supreme Court rulings like Daubert v. Merrell Dow Pharmaceuticals (1993) and Kumho Tire v. Carmichael (1997),16 as well as many local admissibility decisions. Pertinent areas include the admissibility of expert evidence in class-action ‘toxic tort’ cases, such as US class action suits alleging that the drug Bendectin, taken to relieve symptoms of morning sickness, caused birth 14 The most familiar instance of the much-repeated argument that formal scientific accounts misrepresent scientific practice is Medawar (1964). 15 Much of what Garfinkel has said on the topic of instructed actions remains unpublished. To my knowledge, the most extensive treatment on the topic in Garfinkel’s unpublished corpus is Garfinkel, Livingston, Lynch, Macbeth, and Robillard (1989). 16 See end section of Bibliography for full citations for these cases.

Law Courts as Sites for Ethnomethodological Investigations


defects. The courts were faced with deciding which bodies of evidence about the matter counted as ‘scientific’ or not. Legal concerns about, and political exploitations of, so-called ‘junk science’ in the courts led the courts to consider various practical ways of demarcating (good) science from non-science and pseudoscience. Similar issues arise in disputes over the admissibility of psychologists’ testimony about the fallibility of eyewitness testimony. Particular courts have excluded such testimony on the grounds that judgments about the credibility of eyewitness testimony are the province of the jury: a matter of ‘commonsense’ judgment in the context of the singular case, and not a factual conclusion derived from generic social science knowledge. The science/commonsense distinction has come up in appeals cases in the UK, in which the appeal court attempted to stipulate limits to the use of a Bayesian method to assist the jury in weighing all of the evidence in the case. The court ruled that the formula encroached upon the province of the jury and the use of common sense in the circumstances of the particular case to resolve the ultimate issue. The treatment of science in the courts is a distinctive and interesting topic, both for ethnomethodology and social studies of science. The ‘science’ that comes into play is second-order (disengaged from bench practice and shop talk, publicly presented and subject to popularized renderings). It is akin to the ‘science’ in social studies of science, except that it is a subject of specific modes of legal investigation and interrogation. The talk about science in the courts and the presentation of arguably scientific evidence is not idle talk; it is very consequential, and contextually specific. A small, but growing, area of science studies scholarship and examines and criticizes the versions of science featured in the courts and promoted by certain key judicial rulings.17 One source of complaint is that prominent legal scholars and many courts tend to assume an overly idealized picture of scientific method that fails to acknowledge the ‘messiness’ of actual scientific practices, the role of interpretative flexibility in science and technology, and the historically contingent nature of whatever counts as ‘scientific’ at a given time. A number of analysts in the field of science and technology studies (S&TS) have argued that courtroom disputes and government tribunals involving expert testimony can result in a ‘deconstruction’ of specific expert claims, as the public become privy to rival claims and counterclaims about particular facts and methodic procedures.18 When challenged in court, unqualified statements of fact can be shown to be mediated, contingent, interested, and less certain than initially supposed. ‘Deconstruction’ of science in cases in which expert witnesses are cross-examined is said to reveal aspects of the ‘process’ of science emphasized in constructivist studies of science. There is an important qualification about this, 17 See Jasanoff (1996); Smith and Wynne (1989); Edmond and Mercer (1998). 18 The conception of deconstruction used by science studies analysts owes little to Derrida, and more to Latour and Woolgar’s (1979: 179) use of the term to describe arguments among scientists that undermine ‘facts’ that previously had been put forward in an unqualified manner and accepted as reality. Studies that apply this concept to disputes among experts and interrogations of expert witnesses are Jasanoff (1996: 211–214); also Fuchs and Ward (1994).


Orders of Ordinary Action

however: deconstruction is almost invariably partial, limited, and circumscribed. So, for example, when the O.J. Simpson defense attacks the forensic evidence put forward by the prosecution, the defense does not suggest that all scientific evidence is contingent, based on flexible interpretations, and potentially subject to doubt. Instead, the attack is focused on particular procedures, and it casts these against an implied (or sometimes explicit) background of undisputed facts, accepted protocols, and authorized science. A constructivist analyst makes the classic sociology of knowledge move from examining particular challenges in adversary disputes to a more general orientation to the social circumstances and interests associated with expert knowledge. In contrast, ethnomethodology is no more committed to a skeptical picture of science than it is to an unquestioning belief in scientific fact and method. In reference to so-called ‘deconstruction’ in court, an ethnomethodologist has no more or less interest in the validity of an unqualified factual statement made by an expert than in the revelatory value of an interrogator’s ‘deconstruction’ of that statement. A somewhat apocryphal article written by lawyers19 on how to attack expert witnesses in drug cases can give us an appreciation of how deconstruction of expert evidence is not simply a matter of opening up the black box of naturalized science to expose its social construction. Take the following, ideal typical series of questions to be asked of a pharmacologist presenting prosecution evidence: Q: Did you do a quantitative test on the substance? [Few do.] Q: So you don’t know what percentage, if any, of the substance is heroin? [The exact percentage can be determined. It rarely exceeds 20 per cent.] Q: What other substances were present? [Chemists rarely test for other substances.]20

Consider for a moment what this lawyer’s questions anticipate, and what sort of knowledge of pharmacology the questions presuppose. The lawyer – presumably not a chemist by training – makes use of foreknowledge about the routine practices of pharmacological tests in drug cases. The questions in the above series straddle a knowledge gap between the practising chemist and the audience (a judge and/or jury). The questions are designed to elicit admissions to the effect that the procedures actually used differ from those that would have been optimal. In other words, the lawyer seeks to expose an irony akin to the irony exposed by a constructionist analysis of unqualified factual claims. A contrast between ‘ready-made science’ and ‘science in action’ is exploited for particular, local purposes by the lawyer. The particular sequence of questions in this example set up a docile witness whose answers are anticipated in the design of the question and exploited by later questions. Each successive question builds upon the witness’s confirmation (however reluctantly given) to the prior questions, and the series can be seen to be raising doubts and presenting limits that appear to compromise the competence and certainty of the expert’s evidence. A witness who confirms without resistance 19 Oteri, Weinberg and Pinales (1982). 20 Ibid., p. 252.

Law Courts as Sites for Ethnomethodological Investigations


thereby is led step-by-step along the path of the adversary argument, like a lamb to slaughter. Although the logic of the interrogation is canonical for cross-examinations of expert witnesses, witnesses are not always so docile. To go back to Latour’s Janusfaced figure, and the dialectical relation between ready-made science and science in action it illustrates, we can turn these figures around and put them in dialogue with one another (Figure 7.2). Once we do so, we see that we no longer have a dialectic between opposing versions of science but a dialogical organization that is subject to the improvisations and local horizons of interaction.

Figure 7.2

Extension of Latour’s dialogue between idealized science and ‘science in action’

Source: Latour, 1987

An example of a dialogue in which an effort to ‘deconstruct’ an expert witness’s claims encounters resistance occurred during the cross-examination of a key prosecution witness in the O.J. Simpson case. Defense lawyer Peter Neufeld, who specializes in DNA cases, interrogated Robin Cotton, a forensic lab director who earlier had presented the prosecution’s DNA evidence. Neufeld interrogated Cotton over a period of several days, and the particular sequences discussed below occurred well into this period. At the opening of the day’s session, Neufeld complains to the judge of a difficulty he had encountered the day before in ‘impeaching the witness with a learned treatise’. Previously, he laboriously


Orders of Ordinary Action

walked through an authoritative National Academy of Sciences National Research Council report on DNA profiling in criminal justice, inviting Cotton to confirm that the authors of the report were experts, and that the text furnished authoritative standards for laboratory procedures. His complaint now is that Cotton acknowledges the text as a canonical guide to her lab’s practices, but in a selective, and seemingly ad hoc, way that allows her to evade critical implications. He complains that the witness fails clearly to acknowledge the primacy of a formal-literary version of methodology, and he continually searches for an acknowledged standard that he can hold up against the practices of the Los Angeles Police Department and its consulting labs. Whenever he puts Cotton in a position to admit that the lab did not closely adhere to an authoritative standard, she simply asserts that her lab did not agree with the NRC on the specific point. The jury is then left with a judgment about her competence to suspend or override the written recommendations and guidelines in favor of an experience-based situated judgment. This dialogical relationship between setting up a firm standard in order to expose transgressions from it, and a witness’s use of opportunities arising from within the local sequence to claim exemption from the binding terms of the interrogation is visible in the following sequence: Excerpt from People v. Simpson 21 Neufeld: Cotton: Neufeld:

Would you agree, Doctor Cotton, that moisture promotes bacterial growth. Yes. And would you agree, Doctor Cotton, that the bacteria starts eating up the DNA (an’ then) the DNA deteriorates and degrades. Cotton: Over time, yes. Neufeld: And would you agree, Doctor Cotton, that degradation occurs more quickly under the combined effects of moisture and heat. Cotton: Uh yes, I would. Neufeld: In your laboratory, Doctor Cotton, where you are the laboratory director, would it be scientifically acceptable to let wet plastic stains remain in sealed plastic bags inJudge Ito: [ ( ) You said wet plastic stains. (1.0) Neufeld: (’m) sorry. (1.5) Neufeld: ( ) That’s what I have written here. Cotton (and audience): ((Laughter)) Neufeld: At least I can read correctly your honor. ( j’st) can’t write correctly, I’m sorry. Judge Ito: Here. Let me do it. ( ): ( ) Neufeld: Sir- ahm, Doctor Cotton. (1.0) In your laboratory, would it be scientifically acceptable (1.0) to let swatches of wet blood stains remain in sealed plastic bags in the rear of a parked truck un-refrigerated and un-air-conditioned in the middle of June for up to seven hours. Clark: (… argumentative) Judge Ito: Overruled. (2.5)

21 See Bibliography for full reference.

Law Courts as Sites for Ethnomethodological Investigations Cotton:




I don’t think that would be my:: first choice, but please keep in mind that my laboratory doesn’t collect evidence. And so::, we don’t collect it, we don’t have a truck, we- we just receive it from someone else who has already collected it. An’ that’s becau- en as a result of that, Doctor Cotton, there’s no way that you can control for the extent to which thee- ahm, offering agency either degraded those samples or cross contaminated those samples. Of course.

The humorous interlude exposes that Neufeld is reading a series of questions as he proceeds through the interrogation. Though we cannot see what he is reading, his momentary troubles with reading exhibit contingencies that arise in the transition between monological text and dialogical production. I take it that it is recognizable that his questions are (1) sequentially designed and (2) teleologically aimed to elicit Cotton’s confirmation that there are good microbiological reasons not to leave samples in sealed plastic bags inside a vehicle on a hot day (this is connected to the defense’s argument about what the Los Angeles Police Department did with samples collected at the crime scene). Note that the progression of questions moves from the general to the specific; from general principles of biology to the specifics of time and place in this particular case. The questions invite the witness to agree that particular items of evidence in this case are subsumed under good biological reasons that, in turn, provide grounds for doubting the adequacy of the police handling of the legal evidence, and thus of the evidence itself. Cotton’s eventual answer does two things: it relativizes the significance of the practices in question, and it disclaims any direct responsibility for, or knowledge of, the specific details Neufeld cites. A way to conceptualize Neufeld’s difficulty is this: cross-examination is sometimes likened to a powerful machinery for exposing truth in testimony. Neufeld attempts to use a familiar type of ‘machine’ for generating significant discrepancies between (1) authoritative standards governing the collection and processing of evidence, and (2) specific instances supposedly governed by those standards. Expert testimony goes into the machine, contradictions and acknowledged uncertainties come out. In this case, Neufeld’s troubles have to do with securing authoritative standards and getting the witness to acknowledge that those standards govern the evidence in question. The production of testimony, and the witness’s ability to call upon non-standard, localized, competence to make judgments, makes it difficult to ‘screw the machine to the floor’ of the interrogation (I am borrowing Doug Macbeth’s term,22 with some interpretive license, from his wonderful explication of classroom floors.) Conclusion So, what are the implications of this brief analysis for ethnomethodology? One implication is that the courts can be viewed as an ethnomethodological forum, in a particular sense of the word ‘ethnomethodology’. This sense of the word brings into relief the fact that methodological investigations are conducted as part of standing social institutions. Courts conduct what Garfinkel calls ‘formal analysis’ of topics

22 Macbeth (1992).


Orders of Ordinary Action

of order.23 Law courts are, of course, formal organizations, and they are replete with trappings of formality. ‘Formal analysis’ is associated with such trappings, but it refers more specifically to the way the courts explicitly handle the case at hand in terms of a priori resources such as statutes, definitions, documents, precedents, protocols, and procedural standards. Like social scientists, courtroom personnel maintain distance from the lived actuality of the real-worldly events they examine, and they attempt to understand cases systematically by reference to a body of abstractions. The methodology of the court occasionally incorporates scientific and social scientific advice, but it is organized independently of any professional academic discipline (including the academic discipline of Law). Like other instances of formal analysis, the courts’ machinery is embedded in an unspecified array of competencies and organizational contingencies. The problem I have in terms of ‘screwing the machine to the floor’ has to do with the work, and discursive employment, of formal analysis in, of, and as a singular courtroom interrogation. The ‘science’ that is presented and examined in court is a second-order construct that is framed by legal considerations about admissibility and relevance, and is oriented to court-specific conceptions of what ‘science’ might be. Whereas some legal scholars and social analysts offer to correct the idealized, and yet inconsistent, versions of science that tend to predominate in legal scholarship, and to some extent the courts,24 ethnomethodology can give us a different line on it. I would argue that the opposition between ‘ready-made’ (or idealized-naturalized) science and ‘science in action’ has organizational relevance to court-specific science. But it is not an opposition between two incommensurable epistemological positions, nor is this science merely an attribution, image, or play between discursive repertoires. The formal administration of forensic laboratories anticipates the very line of attack that Neufeld pursues, and it does so through what I am inclined to call ‘administrative science’: the work of demonstrating that evidence samples and the chains of custody through which they pass are ‘governed’ by strict, authoritative rules and guidelines. And, in the Simpson case, the attack on those chains and their governance took the form of a powerful ethnography in which defense experts visited the prosecution’s labs and conducted their own, rival, analyses of evidence samples. Consequently, I would not say that ethnomethodology offers a corrective to the court’s ‘epistemology’ of science, but rather that it provides a way of identifying an endogenous ethno-methodology that already takes place in the courts. Consequently, the courts are not only a place in which ethnomethods are enacted, and thus available for ethnomethodological study, but that the courts are a place in which grand methodological problems are investigated and resolved in a pragmatic and circumscribed way. A job for ethnomethodologists who would study such investigations would not be to improve upon them, but to identify the distinctive way in which they are cast. This ‘job’ presents all of us with puzzling implications, because we cannot easily invoke the usual policy-related, normative, or explanatory aims of a social science. Quite the opposite, we may find our own ‘knowledge’ subordinated to the already prevalent policies, normative judgments, and explanations 23 Garfinkel and Wieder (1992: 187). 24 Edmond and Mercer (1998); Jasanoff (1996: 63ff).

Law Courts as Sites for Ethnomethodological Investigations


that are part of the worlds we study. As I see it, there are critical implications when we move, through insights from such studies, back into the usual epistemological debates in philosophy and social science, but beyond that there is the aim to press our ability to understand and articulate the peculiar workings of the social world. In order to deliver upon those aims it is necessary to deliver news, and I would say that ethnomethodological studies of legal practices have delivered news about the way formal systems are embedded in, and dependent upon, unformalized orders of judgment and practice. In the present study, I hope to contribute to that legacy, while also addressing a slightly different line that goes back to Garfinkel’s jury study. And that is, to explicate how the courts themselves produce situated investigations of the relationship between formal analysis and commonplace practices.

This page intentionally left blank

Chapter 8

Circumstances of Reasoning in the Natural Sciences1 Eric Livingston

Introduction: Problem-Solving in Chemistry Why begin a chapter, directed primarily to ethnomethodologists, with the detailed consideration of a problem in first-year chemistry? The chemistry problem has nothing exceptional about it: in the context of first-year chemistry, its solution is relatively simple. This is not the stuff of the history, philosophy, or sociology of science nor, for that matter, is it of obvious interest to ethnomethodologists. The solution takes me to the limits of my understanding of chemistry; it may discourage many of the chapter’s intended readers. At present, I am interested in the ways in which reasoning is peculiar to specific domains of practice. Rather than imagining reasoning as a general type of thing with general properties that are applied and adapted to different circumstances, the idea is to return to the actual phenomena of reasoning, therein to examine how those phenomena are distinctive of, and belong to, the practices of a collectivity. I have come, in this way, to envision an anthropology of reasoning. If anywhere reasoning is treated as a universal phenomenon, it is so treated in the ‘hard sciences’: reasoning in mathematics, qua reasoning, is the same as reasoning in physics, which is the same as reasoning in chemistry. Yet, when I began to study chemistry, I had very little idea of what I was doing. While I may flatter myself by thinking that I have some, albeit small, familiarity with mathematical theoremproving and at least a little appreciation of physics, I had no intuitive understanding of chemistry even at the most elementary level. Thus, the organization of this chapter. By taking up a specific problem, I hope to show, and begin to locate, what might be different and intriguing about the ways of thinking in chemistry without yet being able to formulate precisely what that difference might be. The chemistry problem is the following: A solution of an unknown acid was titrated with base and the equivalence point reached when 36.12 ml of 0.100 M NaOH had been added. Then 18.06 ml of 0.100 M HCl were added to the solution and the measured pH was found with a pH meter to be 4.92. Calculate the dissociation constant of the unknown acid. (Mahan, 1965: 220) 1 I thank Michelle Arens, Peter Forrest, David Francis, Kiyo Fujimori, Steve Hester, Martin Krieger, and Michael Lynch for their careful readings and commentary on this chapter. As indicated in the text, I am indebted to W.W. Norton & Company, Inc. for permission to use a problem and associated illustration from A.P. French, Newtonian Mechanics.


Orders of Ordinary Action

For some readers, even the statement of this problem carries with it the culture shock of being transported to the upper Amazon. The statement is, in fact, extracted from an exotic domain of practice: the statement’s notation, technical terms, and references to experimental procedures – to titrations, equivalence points, and the measurement of solution concentrations and pH – appear to the non-chemist as foreign, distant, strange, and unknown. The solution to the problem does little to clarify the situation; it seems only to make matters worse. Nevertheless, the peculiarity of reasoning in chemistry can only be seen when one tries to engage in that reasoning.2 At one level, the solution to the problem is a straightforward computation. Let Ka be the dissociation constant of the unknown acid. Then pKa is defined as –logKa. For a presumably monoprotic acid (i.e., an acid which, in one model, has one acidic hydrogen atom), the following relationship holds at the midpoint of a titration: pH = pKa From the information given in the problem, we have 4.92 = pKa = –logKa Taking antilogarithms, the result is Ka = 1.2 × 10–5. Such a computation is not very instructive. The idea is this. Let HB represent the unknown acid. NaOH is a strong base which dissociates ‘completely’ in water to give Na+ and OH– ions. The OH– ions will strip the unknown acid of all its acidic protons forming H2O and leaving B– and Na+ ions in solution.3 When half the equivalent weight of the strong acid HCl is then added, this will restore half the protons to the unknown analyte. At equilibrium at this titration midpoint the reaction can be described by the equation HB + H2O

B – + H3O +

The dissociation constant is given by the equation

Ka =

[B – ] [H3O+] , [HB]

where brackets denote the molarity of the solution. Since at the midpoint [B–] = [HB], by definition of pH we have

2 Readers uncomfortable with mathematics or unfamiliar with chemistry should allow themselves to read over this material quickly to get a sense of its technical character; the general argument will develop and become clearer over the course of this chapter. 3 Parenthetically, the solution will not be neutral: the A- ions will interact with water to produce a slightly basic solution.

Circumstances of Reasoning in the Natural Sciences


pH = –log [H3O+] = pKa , hence the solution that was given earlier. Although this is a first-year solution to the problem, nothing about this solution seems to be precise. The molarity of a solution changes with temperature; molal concentrations are more accurate. Theoretically, ‘activities’ should replace concentrations in the expression for the dissociation constant; activities, however, are themselves determined empirically. The formula H3O+ used to describe the hydrogen ion in water is undoubtedly inexact, but the exact structure of the ion (or its rapid mutations) are, as I understand it, matters of research. The computation of the pH at the midpoint neglects the (empirically unimportant) autoprotolysis of water. More practically, confidence limits for the measurements need to be calculated and, thereby, tolerances for the value of Ka. From the perspective of these comments, and undoubtedly many similar ones, it is difficult to find anything that might be technically right about the determination of the dissociation constant for the unknown acid, which is not to say that it is not practically adequate. One solution to this problem – if it is considered a problem – is to propose that the ambiguities surrounding the computation reflect the circumstances of learning chemistry. As the novice progresses, an increasingly complicated but more accurate picture of the titration process emerges. While certainly correct, this suggestion obscures the fact that the same type of ‘ambiguities’ occur at all levels of chemical inquiry: the novice is learning how to work in situations of similar, always-present ‘ambiguities’. The chemist is ‘stuck in the middle’ of laboratory practices and procedures, and ‘stuck in the middle’ of reasoning about such practices and procedures. If we call this situation the ‘condition of reasoning’ in chemistry, the aim of this chapter is to come to an initial, tentative appreciation of this condition and its difference to other forms of inquiry in the sciences. Jigsaw Puzzles In The Structure of Scientific Revolutions, Thomas Kuhn (1970) proposed that normal science is a ‘puzzle-solving’ activity. Kuhn’s idea is that mechanical puzzles have definite solutions; the challenge of working on them is to find such solutions. Similarly, in normal science, practitioners work within a set paradigm which, metaphorically, provides both the puzzle pieces and a sense of what a solution might be. Kuhn’s proposal appears to be an analogy – the everyday practice of normal science is like the work of solving puzzles – but nowhere, so it seems, does he consider the work of solving any particular type of puzzle. The usage is, in fact, metaphorical. Although jigsaw puzzles are interesting objects in themselves, they also serve to illuminate ‘midenic practice’, practice that is hopelessly situated in the middle of what one is doing and in the midst of reasoning about what one


Orders of Ordinary Action

Figure 8.1 The jigsaw puzzle problem

Figure 8.2 A continuity of detail

Figure 8.3 The gestalt interrelationships of a group of pieces

Figure 8.4 A discovered relationship between separate groups of pieces

is doing.4 In this way, jigsaw puzzles can give initial insight into reasoning in chemistry. As illustrated in Figure 8.1, one problem dominates all work on jigsaw puzzles.5 The puzzle piece in Figure 8.1 seems to offer considerable detail, all of it possibly relevant to the solving of the puzzle. Unfortunately, that relevance is not a property of the individual puzzle piece; it is discovered, in the course of working on a puzzle, in endlessly idiosyncratic, detailed ways, in and as discovered relationships between the puzzle pieces. By looking at an individual piece by itself in isolation from the other pieces, one cannot find its place in a puzzle. Instead, as illustrative examples, one finds a continuity of detail between different pieces (Figure 8.2), arrangements of interrelated detail that allow a group of pieces to be fitted together as a whole (Figure 8.3), large and small groups of pieces whose connections are anticipated and discovered as groups (Figure 8.4), features of pieces that can barely been seen until they are placed in their proper position (Figures 8.5 and 8.6), and large and small ‘holes’ whose contours provide important clues (Figure 8.7).

4 Midenic reasoning in puzzle-solving and mathematics is discussed in Livingston, 2005. For beginning work on origami, see Livingston, 2000. 5 The puzzle used for illustrative purposes is ‘Christ Church Cathedral, Fredericon, New Brunswick’, produced by The Canadian Group.

Circumstances of Reasoning in the Natural Sciences


Figure 8.5

Discovered detail, part A

Figure 8.7

Specific search criteria occasioned by a puzzle’s semi-finished areas

Figure 8.6

Discovered detail, part B

This problem – that the relevance of the details of the pieces is a property of the puzzle and not the individual pieces – provides the general condition for working on a jigsaw puzzle. Work on a puzzle is the continual search for ways of finding that relevance. The techniques for doing this are innumerable; they are occasioned by the situated problems that they seek to address. A few examples will give the idea. Most puzzle-solvers initially sort the pieces to find the border components, frequently sorting for general colours of sky, buildings, grass, and the like at the same time. Putting together the border is not, however, automatic. First, the size of the border is unclear, and the puzzle-solver will bring together in the same area pieces of


Orders of Ordinary Action

seeming relevance to each other (Figure 8.8). Gaps in the developing frame (Figure 8.9) occasion searches for pieces which are seen, therein and thereby, to be ‘missing’, where the detail of the missing piece becomes relevant as part of that search (Figures 8.10 and 8.11). Figure 8.12 illustrates a different type of problem: here the opposing sides of the puzzle do not have the same projected length. This circumstance occasioned the examination of the border constructed so far; sections of it were found to have been incorrectly joined together. I had to realize that this was the case, identify the mistakes,

Figure 8.8

Initial rough grouping of edge pieces

Figure 8.9

Observed gaps occasion the search for missing pieces

disassemble parts of the border and reassemble them in different places. The need to do this and the identification of the mistakes were all part of the work of searching (moving, inspecting, arranging, re-arranging, fitting, attempting to fit, etc.) among the pieces and inspecting their developing organization to find how they could be positioned together. These examples serve as an invitation for readers to consider for themselves the actual work of putting together a jigsaw puzzle. Rendering this work as an abstract form of deductive or inductive inference seems pointless: the puzzle pieces cannot be parameterized in any solution-relevant manner, and the features of a puzzle piece

Circumstances of Reasoning in the Natural Sciences

Figure 8.10 Details examined from within a specific search


Figure 8.11 A discovered relevance of detail

or collection of such pieces cannot be characterized independently of all the things that a puzzle-solver is attempting to do to find those features. The situation is extremely fluid, depending on what a puzzle-solver has done, is doing, and is trying to do; on a familiarity with the puzzle pieces and their shapes as they are manipulated and examined for their compatibility; on anticipatory perceptions of the continuity or relationships between various details and patterns; on Figure 8.12 A discovered problem the arrangement of the pieces and their in the relative positioning relative to completed sections lengths of the sides of the puzzle, and all the rest. In some ways, comparing chemistry and jigsaw puzzles is absurd: reasoning in chemistry is not the reasoning of jigsaw puzzles, and chemical practice is hardly the search to fit ‘pieces’ of a puzzle together. The analogy serves only to give a sense of how chemists are situated in the midst of their own practices and procedures, how those practices and procedures give definiteness to their work (like the calculation of the dissociation constant in the chemistry problem), and how that definiteness, for chemists, cannot be extracted from those practices. Consider again the problem introduced at the beginning of this chapter. The problem explicitly concerns the laboratory procedure of titrating a solution to determine its properties. By mentioning a pH meter, it tacitly acknowledges the empirical character of all the measurements that are involved – particularly, the measurement of the equivalence point and the quantity of NaOH and HCl added. ‘Pure’ quantities of NaOH and HCl must be produced first, and solutions of these titrants must be standardized by a titration procedure similar to the one for determining Ka. More to the point, we can ask what the meaning of the dissociation constant is. At first it might appear that it has a clear definition or, if not that, one that is defined


Orders of Ordinary Action

operationally in a precise manner. Operationally, however, we are led into all sorts of refinements concerning temperature, descriptions of concentration, and activity coefficients. Further, the interpretation of the coefficient is essentially comparative (as is the pH of a solution). The coefficient is used in many calculations, but these calculations are themselves reflections of other laboratory procedures. The analogy with jigsaw puzzles, if it has instructive substance, is that the ‘pieces’ consist of a ‘mess’ or ‘morass’ of laboratory procedures and practices; the relevant detail of the ‘patterns’ that emerge from that ‘mess’ are bound to the ways that those procedures and patterns have been fitted together. Chemists, so it would seem, cannot extract their work from the surrounding and continually organized and re-organized procedures of which that work is a part. Nor is this relationship hidden to them. Their reasoning about chemistry is tied to this situation of chemical inquiry; it cannot be disengaged from it, and it is this omnipresent situation of their own work and theorizing that provides the ‘conditions of reasoning’ in chemistry. A Problem in Physics To understand better the peculiarity of reasoning in chemistry, let us consider the following problem (shown in Figure 8.13) from a first-year physics textbook (French: 238–9): A large mass M hangs (stationary) at the end of a string that passes through a smooth tube to a small mass m that whirls around in a circular path of radius l sin θ, where l is the length of the string from m to the top end of the tube (see the figure). Write down the dynamical equations that apply to each mass and show that m must complete one orbit in a time of 2π(lm/ gM)1/2. Consider whether there is any restriction on the value of the angle θ in this motion.

Figure 8.13 A Physics Problem From NEWTONIAN MECHANICS: THE M.I.T. INTRODUCTORY PHYSICS SERIES by A.P. French. Copyright © 1971, 1965 by the Massachusetts Institute of Technology. Used by Permission of W.W. Norton & Company, Inc.

A possible objection to this choice of a physics problem needs to be dismissed immediately. The problem is related to an empirical setting just as the chemistry problem is. In fact, the problem offers a rather amazing idea: if a physical apparatus such as that shown in Figure 8.13 is used, and the ball m is swung faster and faster until the mass M stays stationary, moving neither up nor down, then, to the extent that air resistance can be neglected as well as the weight of the connecting cord, the force exerted downward by M will continue to drive m at a constant speed

Circumstances of Reasoning in the Natural Sciences


in a circle of radius l sin θ. A related, more practical experimental situation is when a ball is connected by a thin, lightweight rod to a spinning shaft driven at a constant angular velocity; the governor of some motors is built on similar principles. On the other hand, the relationship between the empirical setting and the formulation and solution to the problem are different from that relationship in the chemistry problem; that is where our interest lies. Figure 8.14 reflects a method for solving the physics problem. Although the figure looks complicated, it is not if one draws it for oneself and works out the indicated relationships. Two forces act on the heavy block with mass M. One is the gravitational force Mg; since the block is stationary, the other force is the equal and opposite force of the tension Figure 8.14 Method of solving the T1 in the string. Hence, the magnitude physics problem of that force is T1 = Mg. Three forces act on the ball with mass m. One is the tension in the string, which has magnitude T2; one is the gravitational force mg; the third is the centripetal force of uniform circular motion mv2/R. Since the ball moves neither up nor down, the component of the string tension in the vertical direction (of magnitude T2 cos θ) must equal the gravitational force; the horizontal component of the tension (of magnitude T2 sin θ) must equal the centripetal force. Finally, since the string has no upwards or downwards motion, T1 = T2 = Mg. One relationship is expressed by the equation T2 cos θ = Mg cos θ = mg or cos θ = (m/M) This gives the restriction on the angle θ. The other relationship is expressed as T2 sin θ = Mg sin θ = mv2/R Figure 8.14 shows that sin θ = R/l. Moreover, the speed v of the ball equals the distance around the circle it traces divided by the time τ it takes to make one orbit; hence v = 2πR/τ. Making the appropriate substitutions and solving for τ, we get


Orders of Ordinary Action

τ = 2

1m gM

1 2

The achievement of this work is that we have been able to describe the empirical situation in terms of essential features of the objects – their masses, the angle of the string from the vertical, the constant tangential velocity of the ball, the period τ of the orbit. These are essential features of the objects because they are adequate to solving the problem in their own terms by using them to parameterize the empirical situation. Therein, we set up an empirical correspondence between the experimental situation and the mathematical model. The fundamental skill illustrated by this physics problem is the ability to describe the experimental situation parametrically, therein providing an empirical correspondence between the model and the experimental situation. This is different from the chemistry problem. In the chemistry problem, we had, on the one hand, indicators of processes (such as the midpoint and endpoint of the titration, the pH of the solution) and, on the other, laboratory procedures (such as titrations, standardization of solutions, the determination of the endpoint, the measurement of pH). The task was to see how the procedures were being arranged so that the indicators could be used to characterize the properties of an object, a weak acid, itself given through those same indicial procedures. Although we tend to think of procedures as algorithms which, when applied to objects, yield determinate results, they also serve to identify those objects to which the procedures can be applied; the chemist seems to be situated within a procedural world, the arrangement and coordination of those procedures serving to exhibit the identifying properties of objects. Difference and Equivalence The difference between first-year chemistry and physics that I have described is qualitative, not quantitative. In general, it is reflected in the problems given in textbooks – in chemistry, the problems typically involve the calculation of quantities embedded in laboratory procedure; in physics, they involve the mathematical parameterization of physical situations. The contrast is neither strict nor absolute, but seems to reflect a fundamental difference in laboratory practice, at least as that practice is developed in first-year courses. We tend to think of procedural descriptions as algorithms – given an algorithm for finding the least common divisor of two numbers, we put in two numbers, follow the steps of the algorithm, and end up with the least common divisor. More generally, however, procedural descriptions also identify a collection of things on which the procedure can be applied. We may have an algorithm for alphabetizing a list of names, but the algorithm works for those lists of names formatted in a manner suitable to the algorithm. In chemistry, laboratory procedures identify a class of substances or reactions for which they are applicable, and the student adapts these procedures to identify particular features of a particular substance or reaction. In

Circumstances of Reasoning in the Natural Sciences


physics, a mathematical model is compared with data generated from an experiment that, at the same time, might be said to be a representation of the mathematical model. Focus seems to be placed on the comparison with and substantiation of lawlike behaviour, and the fundamental skill involves constructing a physical situation in which such a comparison can be made. One difficulty with this proposition is fairly clear. Once the difference between these situations of inquiry is described in words, those words can be used to show that any such difference is illusory. First-year chemistry just as surely involves ‘models’ as does physics; practical work in the chemistry laboratory is used to substantiate general laws and law-like behaviour the same as in physics. A second problem arises from the logic of explanation. The explanation of a solution to a chemistry or physics problem, and the explanation of the logic and aims of a chemistry or physics experiment, present a problem or experiment as an instance of general principles. Examples are used to illustrate general reasoning applied to particular situations. If, on the other hand, we view these problems and experiments from the point of view of the student rather than the explanation, they seem to require considerably different skills. The type of reasoning that they require, and the circumstances from within which such reasoning arises, are not the same. The idea of this chapter is that if we allow this difference, we can begin to explore reasoning, not as a universal phenomenon, but as something that belongs distinctively to different domains of practice. The Project of an Anthropology of Reasoning

Figure 8.15 First checkers problem

The conceptualization of an anthropology of reasoning arose in a different context. I had become interested in the game of checkers. Faced with problems such as those in Figures 8.15 and 8.16, I did not know how to solve them.6 At the same time, these are considered to be elementary problems; once I began to learn how to solve them, I saw them, if not simple, as relatively easy. The solution to Figure 8.15 is for White to move 15–10. Black is presented with two possible jumps; jumps in checkers are forced. Black has a choice of jumping 24–15 or 14–7. Black must play 14–7; Black 24–15 leads to White 10–26 for the win. On Black’s next turn,

6 The problem in Figure 8.15 appears in Spayth, ca. 1890: 252. The problem in Figure 8.16 is attributed to Joshua Sturges in Gould, 1884: 8. A fuller discussion of reasoning in checkers is found in Livingston, 2006.


Orders of Ordinary Action

Figure 8.16 Second checkers problem

Figure 8.17 Solution to second checkers problem

Black will have to make the second jump 24–25. Between Black’s moves, White plays 6–2. After these moves have been made, White can jump 2–27, completing a ‘three-for-two shot’ where White sacrifices two checkers in order to gain three. The solution to Figure 8.16 involves the perception of the ‘opposition’ of the checkers. White plays 7–10. Black must move 15–19 or else one Black checker will be lost, and the two eventual White Kings will beat one eventual Black King. After Black moves 15–19, White plays 21–17, forcing Black to play 9–13 or lose the checker on 9 by White playing 17–13 next. The situation is shown in Figure 8.17. In Figure 8.17, White’s move is 10–15. Black has to jump 13–22; then White has to jump 15–24. The remaining White and Black checkers are in ‘opposition’ to each other; in this case, the opposition favours White. However Black moves, White will move in the same direction and be able to pin the eventual Black King to the side of the checkerboard. These are, again, elementary problems. Each of Black’s moves is either forced or leads to an obvious loss. In general, what amazed me about these problems is that one had to know how to play checkers in order to be able to reason about checkers. Rather than reasoning being a general type of thing, with universal properties belonging to reasoning itself, reasoning in checkers seemed to belong to the practices of playing checkers. Reasoning in checkers is an autochthonous phenomenon, arising from within the same practices that sustain reasoning-like-that. In a similar manner, whatever and as little background I have in mathematics and physics, I found (and still find) chemistry difficult. The mathematics, at least at the level of my studies of chemistry, is nowhere near as complicated as physics; for the most part, it involves the calculation of quantities. Like checkers, the situation becomes more understandable if one gives up the idea that all reasoning is equivalent

Circumstances of Reasoning in the Natural Sciences


and, instead, begins to see that reasoning in chemistry is reasoning that belongs, distinctively, to chemical practice. These materials offer initial grounds for pursuing an anthropology of reasoning, not only in games like checkers, but in chemistry and physics. They indicate the possibility of returning to the phenomena of reasoning, not only in jigsaw puzzles, but in the sciences therein to learn what reasoning, and what an anthropology of reasoning, might be.

This page intentionally left blank

Chapter 9

Expert System Technology in Work Practice: A Report on Service Technicians and Machine Diagnosis Erik Vinkhuyzen and Jack Whalen

Introduction The history of work is a chronicle of technological innovation, of endeavors to develop tools and machines that augment or supplant human muscles and, more recently and controversially, native human intelligence. With each new technology questions arise about what it can and cannot and especially what it should and should not do; Woolgar (1987: 312) notes that these discussions ‘are the reverse side of the coin to debates on the capacity, ability, and moral entitlements of humans.’ But the emergence of artificial intelligence (AI) as a significant area of both scientific research and industrial application has dramatically shifted the ground, as the idea of substituting machine or ‘artificial’ intelligence for human minds raises fundamental questions about our capabilities as a species; that is, about what it means to be human. Woolgar puts it thus: [W]hen technology predominantly was composed of prosthetic devices (functional additions to the mechanical abilities of the human body), it could always be argued that humans are unique by virtue of their intellectual faculties; no prosthetic device could emulate a human’s ability to reason, know, and understand. The work of AI, however, attempts to develop a technology that emulates the action and performance previously accredited to unique human intellectual abilities (Woolgar, 1987: 313).

The notion of substituting machine intelligence for human minds has perhaps achieved its most apposite rendering in ‘expert system’ applications – computer programs that represent the expert knowledge of some specialist and reason with it to solve problems or give advice (Feigenbaum, McCorduck and Nii, 1988; Jackson, 1990). While the impact of artificial intelligence (AI) had thus far been largely limited to academia, the promise of expert systems was to have genuine impact on commercial enterprises. This promise was that expert systems could replace those human experts in a company who were usually the highest-paid workers, and could make this expertise readily available to non-experts. Expert systems were thus the first steps to make good on the promise of one of the founders of AI, Herbert Simon, who had predicted in 1965 that ‘by 1985 machines will be capable of doing any work a man can do’.


Orders of Ordinary Action

Some two decades later, the success of expert systems has been mixed at best. A notable business consultant (Davenport, 1995) has declared that ‘expert systems are dead’, claiming in his epitaph that many celebrated projects ‘are no longer being developed or used’, that ‘leading expert systems implementers have moved on to other technologies’. Yet, while nobody denies that the aforementioned hype about expert systems is over (many of the expert system companies have in fact gone out of business or radically changed their core business), the idea of capturing knowledge into computer systems remains strong. Curiously, though, throughout the debate over whether computer programs can robustly reproduce and perhaps significantly improve upon the judgment and behavior of human beings who have expert knowledge and experience in a particular domain, there have been few field studies of these systems in actual use at work sites. And this is despite the uncertainty and controversy concerning their ultimate success or failure as viable commercial systems, let alone as systems that are ‘capable of doing any work a man can do’. In order to evaluate existent applications it is imperative that researchers examine not the technical achievements of expert systems per se but rather what Hutchins (1995) terms ‘the system of person-in-interaction-with-technology’. A close examination of how the user and the expert system function together offers us the best opportunity to critically assess the real-world performance of expert systems, to determine how people use these systems in and for their work. That is to say, only field studies that examine the real-time performance of users while using an expert system will allow us to go beyond the current fixation with ‘intelligence’ in the abstract, and to consider the problem of reasoning and action in situ. Those few field studies that have been completed (see especially Suchman, 1987; Hartland, 1993; Whalen, 1995, Vinkhuyzen, 1997, Whalen and Vinkhuyzen, 2000) have demonstrated that crucial and apparently irremediable differences between human action or reasoning and the operation of machine-based intelligence can give rise to particularly vexing difficulties for the users of expert systems. In this chapter, we attempt to extend and deepen the analysis put forward in these studies through an ethnographic and video-based investigation of the design history and practical use of an expert system for diagnosing problems with document machines. This system, code-named PRIDE, was designed at the Xerox Palo Alto Research Center (PARC) in the early 1990s. We will also examine a distinctly different kind of effort at technologically supporting the reasoning of technicians, also originating from PARC’s artificial intelligence research group, aimed at sharing human expertise within the technicians’ work community. Our analysis is motivated by two related questions: What does it mean to truly support a particular human activity or task, like problem-solving or diagnosis, with or through a computer system? And what could effective support actually look like, with respect to the work practice of technicians? By scrutinizing the organizational rationale behind the PRIDE expert system and its design evolution, the commonsense reasoning and competencies that its users brought to bear when trying to employ it in their work, and the merits of an alternate design (and organizational) strategy that essentially turns AI on its head, making the work community itself the ‘expert system’, we should be able to shed some light on the expert systems debate and, by extension, on the larger discussion about artificial intelligence as well.

Expert System Technology in Work Practice


Our data was collected in the field, at customer sites, and in the classroom where technicians were learning how to use PRIDE for diagnosis and repair. We recorded approximately 36 hours of video in classrooms and eight in the field. Our field observations took place periodically over five years, starting in 1995, and involved some 20 different technicians. The field observation included informal interviews with technician informants, as well as group interviews with other technicians at several regional offices. We also collected close to four hours of telephone conversations on a hotline for technicians seeking help with document machine problems from the company’s expert field engineers. These varied data and methods allow us to analyze a particular situated activity – the work of servicing document machines – in both its interactional dimensions (the multiple relationships and dealings between technicians, customers, document machines, and computer programs like PRIDE) and organizational dimensions (the policies, practices, and structure of a large corporation). The chapter is organized as follows. First, we describe the work of Xerox’s service technicians and the job aids that the company developed over time to help technicians in diagnosing machine failures, the culmination of which is the expert system. Second, we present the PRIDE expert system and show how technicians used it to diagnose machine problems, and in doing so identify some troubles with the designers’ underlying assumptions about service technician work and information needs. Third, we describe how the PARC researchers who first built PRIDE came to believe – after doing field tests of an early version and speaking directly with technicians – in a very different model of what expertise is and where it resides, and as a result developed a community knowledge-sharing application they named EUREKA. In the final section of this chapter, we reflect on the evolution of EUREKA as compared to PRIDE and what this means for the design of effective decision support. All material in this chapter is reproduced by permission of Xerox Palo Alto Center. Basic Features of Technicians’ Work and their Repair Manuals Doing Field Service Xerox service technicians spend most of their day at customer sites repairing document machine problems. When customers experience a problem with their machine, they telephone the company to place a service call. The technician assigned to that machine will receive a page with the serial number and a brief description of the problem. The service technicians are part of work groups, the members of which work on similar machines. When technicians are unable to solve a problem or need a part they do not have on hand they can call the other technicians on the radio. The actual work of maintaining document machines involves diagnosis and repair. The diagnostic ‘process’ invariably starts with the customer’s description of the problem. Oftentimes, technicians do not need to repair the document machine but only need to fix the relationship between the customer and the document machine. Orr (1996: 3) explains it straightforwardly: ‘The problems encountered by technicians are most fundamentally breakdowns of the interaction between customers and their


Orders of Ordinary Action

machines, which may or may not include a malfunction or failure of some machine component.’ Most times when customers call for service, of course, there is a genuine problem with the machine. In such cases, technicians must not only diagnose, but also repair the problem. The diagnosis and repair can be very quick, as when the simple replacement of a part fixes the problem, but can be a lengthy process and sometimes take several days. Regardless of whether technicians find a diagnosis instantaneously or after a long search, they are always attempting to understand what is wrong with the particular machine and to construct a diagnosis that ultimately explains why the machine displayed the faulty behavior. In this regard, Orr (1996: 2) maintains, ‘The actual process of diagnosis involves the creation of a coherent account of the troubled state of the machine from available pieces of unintegrated information, and in this respect, diagnosis happens through a narrative process.’ We believe that the term ‘narrative process’ is a bit mystifying and signifies nothing more than the common notion that every machine problem is an accountable phenomenon; a good diagnosis explains why the problem occurred. Diagnosing document machines is an inherently situated activity in which the technicians are trying to discover how the current problem occurred by generating and testing their accounts of the machine’s functioning; the machine is manipulated, its ‘response’ to the manipulation is continuously monitored and accounted for in an interactive, cyclical, process. Service Documentation The service documentation provides technicians with detailed information about the design and operation of the machine. Three decades ago, service documentation contained two parts. The first, called the ‘Principles of Operation’, described how the parts of the machine function individually and in unison to perform the overall functions of the machine. The second part contained schematics, diagrams, and illustrations that made clear how the machine is wired, and the required voltages in every cable. At the time, these two parts of the documentation constituted the entire service manual. The documentation provided no direct help in performing diagnosis; it only described a correctly functioning machine. In the 1980s, the company added another kind of documentation, which was intended to be a more direct help to the service technicians when they were diagnosing a machine. The new ‘directive documentation’ was designed to closely guide technicians through the diagnostic process. To this end, it contained many small decision trees, called Repair Action Procedures (RAPs). At the top of each RAP is a symptom (for example, a machinegenerated fault code); the bottom nodes of the RAP describe repairs that constitute alternative diagnoses for the symptom. The RAP branches into a binary tree through a series of tests and associated questions that require ‘yes’ or ‘no’ answers. Figure 9.1 displays the ‘C2-1 RAP’ for a high-volume copier. As can be seen, the decision tree is laid out on the page as indented text; the ‘no’ branch follows immediately below, and the ‘yes’ branch can be found by tracing the vertical lines down the page, and if necessary to the next column or page.

Figure 9.1

Xerox’s Rapid Action Procedure (RAP) for a high-volume copier problem


Orders of Ordinary Action

RAPs contain two kinds of information. First, it lays out a step-by-step procedure for doing a diagnosis, and second, embedded in the decision tree are diagnostic inferences, conclusions about the implications of test results. As such the RAPs are intended to direct service technicians when doing diagnosis; all they have to do is perform the tests and answer the questions. Since documentation writers author all RAPs, the repair procedures represent their analysis of symptoms. Moreover, as these documentation writers work for the service organization, the documentation also embodies the policies of the service organization. As Orr (1996: 106) remarks, ‘the design of this device for conveying information [the service documentation] is done by one group with information from another group according to policies from yet another group, and the policy input has the effect of changing the service manual from an information device to one that also attempts to determine how the work will be done.’ The directive documentation can be seen as an effort by the corporation to exert some centralized control over the work practices of its geographically dispersed service force. More recently, an electronic hypertext version of the directive documentation (E-doc) was developed. The technicians were supplied with a laptop, on which Edoc was installed. In E-doc, the yes/no branches of the RAPs are hyperlinked to the corresponding section of the decision tree; clicking on a link causes the system to display the appropriate next test. As we will see below, E-doc played a key role in the operation of PRIDE. The PRIDE Expert System How PRIDE Works PRIDE can be viewed as the next generation of electronic documentation for service technicians. The central idea behind the initial version of PRIDE designed at Xerox PARC was to offer technicians much more flexible use of the RAPs. PARC’s plan for expert system diagnosis relied on the text and logic of the RAPs but had a different interface and contained an additional rule-base (based on a model of the document machine) that determined which RAPs would be displayed. The system would also allow technicians to search for solutions in a way that was more commensurate with their particular style or method of reasoning (Bell et al., 1995).1 1 Bobrow (personal communication) explains this design rationale and the resulting system operation as follows: ‘[A technician] could select a possible diagnosis – literally by selecting on the screen – and see what tests you could use to confirm that guess. You could also select a response and find out the corresponding diagnoses. We also designed it so that you could see a skeleton of the tree (for example, test names, branches and diagnoses). Thus, PRIDE supported service engineers in following their own decision making processes, rather than being strictly limited to the decision making structure that was encoded in the static RAPs. For example, if the service engineer’s experience led them to quickly conclude that the problem was with a particular part, they could immediately select that part and get a minimal set of tests that would confirm or exonerate the part as being the problem. The alternative with static RAPs

Expert System Technology in Work Practice


The PARC expert system further extended the capability of the E-doc by utilizing the non-volatile memory (NVM) data stored in the document machine itself (which contains a log of all the fault codes) to determine what part was the most likely cause of trouble. These data are analyzed by a set of rules, and the ultimate result of the analysis is a determination that a certain part is more likely to be at fault than another (see de Kleer and Williams, 1987). Xerox’s service organization very much liked the idea of an intelligent front-end to the E-doc. But they were less enthusiastic about the degree of freedom the PARC prototype allowed the technicians when doing diagnosis. From their view, and in keeping with their own design history, a more directive approach was preferable. A design team working on documentation for a new, multi-function (scan, copy, print, fax) document machine subsequently took the prototype from PARC and started to develop PRIDE; what had been an AI research project was now a corporate program. The version of PRIDE that technicians used in our study was developed for a new multi-function device. The PRIDE interface is divided vertically into two columns, as can be seen in Figure 9.2. On the right side, the system displays the instructions, tests, and questions. At the bottom of that panel are the ‘no’ and ‘yes’ buttons one must use to answer the questions. On the left side, a hierarchy of ‘suspects’ is displayed. The current suspect is highlighted (in this case ‘Printer lead edge registration problem’); a suspect that the system has excluded as a possibility has a faded color. Associated with each suspect is a sequence of tests, and question/answer pairs. The rule-base determines which suspects should be excluded, given the answers to the questions, and which suspects have become more likely. Although the default behavior of the system examines suspects in order from top to bottom, the user also can select a suspect (as long as it has not been excluded) and thus start in the middle. Figure 9.3 shows a partial reproduction of a page from the PRIDE manual that instructs the service technicians on how to use the expert system. As can be seen from the description, technicians are urged to perform the tests and answer the questions posed by PRIDE. The instructions make clear that for the writers of the manual, and probably for the service organization’s PRIDE design team as well, the expert system contains the knowledge to diagnose the document machine and the technician’s task is simply to supply it with correct information. PRIDE in Action The PRIDE manual states that the expert system was designed to help technicians determine a logical path to repair the copier. The technicians are encouraged to follow the expert system’s lead. At times, this can work out very well, especially when was that they would have to find a RAP that had the particular part somewhere in the RAP, then go to the top of the RAP and follow it until reaching the part. This often involved more steps than the minimal set of steps that were provided by PRIDE. Also, PRIDE supported service engineers in entering symptomatic findings in the order they found most convenient, rather than having to enter the findings in the order of a particular RAP. The PRIDE system would maintain the logic of the encoded decision tree, even though symptoms could be entered in different orders.


Orders of Ordinary Action

Figure 9.2

Example of Xerox’s PRIDE interface

Figure 9.3

User instructions from the PRIDE manual

certain events coincide. First, technicians must provide the system initially with just the input that sets it off on the right track to the trouble source, perform the tests that the system proposes correctly, and then supply the system with proper information

Expert System Technology in Work Practice


on the results of those tests. Second, and most obvious, the designers of the expert system must have anticipated the current symptoms and correctly identified the underlying problem as one of the possible causes; in other words, the symptom, diagnosis, and repair must be part of the documentation (cf. Orr 1996: 106–7). We will revisit these issues later, but for now it is clear that if PRIDE is to be a genuine Figure 9.4 Technicians Jeff and Bruce use decision support system for PRIDE to diagnose a copier fault technicians, its performance in the field must meet at least these two criteria. In one of our documented cases, two technicians in the classroom, Jeff and Bruce – in Figure 9.4, Jeff is sitting with his hands behind his head, and Bruce is standing behind the document machine – use PRIDE successfully to diagnose a fault with the machine; they perform the tests it suggests, one after the other, and provide PRIDE with (correct) answers to its Figure 9.5 PRIDE screen prompts inspection yes/no questions. During of the copier this diagnostic effort, PRIDE presents the technicians with a series of tests and questions and, using their answers, works through a long list of suspects until it reaches the point where the screen depicted in Figure 9.5 appears. Excerpt 1 J:

((reading from the screen)) ‘Go to flag three check the wiring for open circuit between P fortyseven and the paper transport m:otor’ ((Jeff walks to the back of the machine and both look inside the machine, as shown in Figure 9.6.)) ((J returns to the front of the machine.)) J: Wiring is good? No, I would say wiring is bad, [right? B: [right (.) yes


Orders of Ordinary Action


So that is a no (..) J: Wiring is good. No. B: Wiring is good? No. B: ‘Corrective action: corrected’ (.) Yeah J: Yes (.) ((Clicks ‘Corrective Action Completed’ button; a window then appears with the text, ‘Did the Action(s) fix the problem?’ with ‘yes/no/cancel’ buttons)) B: (Fix the problem) yeah ((Clicks ‘yes,’))

In this case PRIDE leads the technicians successfully down a ‘logical path’ to a diagnosis and repair. Yet, even in this straightforward case, it is plain that interpreting PRIDE’s questions and answering its questions (examining the wires, measuring connectivity using a volt meter, etc.) require a sophisticated set of skills. Suchman (1987) makes the following observation about following instructions: ‘while Figure 9.6 Inspection of the copier’s interior instructions answer questions about objects and actions, they also pose problems of interpretation that must be solved in and through the very same objects and actions to which the instructions refer’ (Suchman: 142). As an illustration, consider the following excerpt in which two other technicians are trying to answer the question, ‘The lead edge of the sheet in the duplex area is parallel to the duplex nips.’ When this data fragment begins, the technicians, Al and Grant, have opened up the machine (Figure 9.7) and are now looking inside (the screen they are viewing is shown in Figure 9.8). Figure 9.7

Technicians Al and Grant try to answer PRIDE’s question

Expert System Technology in Work Practice

Figure 9.8


PRIDE’s offer of possible faults poses interpretation problems

Excerpt 2 G:

G: A: G: A: G: A:

((reading from the screen)) ‘the lead edge of the sheet is parallel to the duplex nips,’ well we gotta look (....) Is it pretty well (..) I’d call it parallel. Probably Yeah, I would say Yeah, cause it’s=uh (..) yeah, it’s pretty parallel (.) those are the duplex nips up there? Or here? [Okay [yeah I’[d say yeah [say yes

As we can see from this excerpt, the technicians are not sure what the ‘duplex nips’ are and, while looking inside the machine, make a best guess as to whether the paper is parallel to them. Also, the word ‘parallel’ is subject to debate. Grant says that ‘it’s pretty parallel’ and that he would ‘call it parallel probably.’ Al also appears tentative when he says, ‘yeah, I would say’. The technicians need to determine whether the ‘parallelness’ of the paper with the duplex nips warrants, in all reasonableness, agreement with the statement the expert system is presenting, or not. Naturally, the technicians have to make these judgments without further help from PRIDE. Their confusion about the duplex nips is not available to the expert system itself (see Suchman, 1987, and Orr, 1996, for similar observations; see also Whalen and Vinkhuyzen, 2000).


Orders of Ordinary Action

It is quite plain that technicians must make practical sense of PRIDE’s questions and tests in some real-world context. This simple fact has some profound consequences, however. It is the technicians that have sensory access to the document machine and its (mal)functioning. This privileged position visà-vis the expert system gives them the responsibility to perform the tests and provide PRIDE with input. However, having access to the machine also allows them to judge the very relevancy of PRIDE ’s questions in the first place. Although technicians accept that PRIDE was designed to lead them to the logical path to repair the copier, they also know that it can only ‘reason’ with the information they have provided. That is, technicians are aware that their input determines the ‘logical path’ PRIDE takes. Consider, for example, an instance a little while later in the same diagnostic session, when Al and Grant believe they are ‘lost’ in the expert system: Excerpt 3 G: u:hm I’m at a point I need to go back and (..) run a copy or som’thin’ or watch gears turn, or som’thin’ to see what’s not working (..) I guess (....) G: I’m not even sure we’re in the right place (...) G: At all

Grant expresses the desire to see the machine running so that he can get a better understanding of what the ‘actual’ trouble is. He adds that he is ‘not even sure we’re in the right place’. In order to improve his understanding of what ‘actually’ goes on with the machine and to better calibrate what the expert system is saying he wants to observe it while it runs some copies; it is only with an understanding of how the machine works, and just how and when it fails, that technicians are able to make proper sense of the questions and actions suggested by PRIDE. The very appropriateness of the system’s questions can be determined only by examining the document machine at hand; only by diagnosing the machine and understanding what is going on can technicians determine whether PRIDE is on the right track to diagnose the problem. Therefore, when the questions and tests PRIDE proposes no longer make sense to technicians, they do not just doggedly continue following its suggestions, but rather will wonder whether they have answered a question incorrectly. They must have set it off in pursuit of a bad hypothesis. In one case, for example, Jeff muttered, ‘We went down the wrong path!’ when, some time into a diagnostic session, PRIDE presented him with a new screen of ‘initial actions’ that he did not understand. And, consider the following transcript of Al and Grant, who are now looking at a PRIDE screen (Figure 9.9) that suggests that they run ten copies of a test pattern.

Expert System Technology in Work Practice

Figure 9.9


PRIDE suggests a course of action

Excerpt 4 A: G: A: G: A:


Now what are we doin’ here? u::h (..) we need to run=uh the routine oh eight ninety one and run ten prints of the test pattern ( ) (.....) No, that ain’t gonna be right, cause that ain’t gonna run through right? I don’t think so ((deleted lines in which G starts looking for a loose wire in the tray)) You know somethin’ let’s go back and reset the faults (....) ((deleted lines)) A: An’ I gonna say do we wanna assume that that lead edge was parallel or what? (.) [shall we go back [I don’t know

Al and Grant realize that PRIDE’s suggestion to run copies cannot be right because the very problem they are troubleshooting is that the system continually jams in the duplex area. They realize they must have gone down a wrong path; their answer to one of the previous questions must have been incorrect. Al therefore suggests that they go back and reset the faults. This will erase all the answers that they have given thus far and restart the troubleshooting session from the beginning. Later on, when they return to one of the questions they answered before, they decide to answer the question (Figure 9.10) differently ‘this time’.

Orders of Ordinary Action


Figure 9.10 Restarted troubleshooting session enables a different decision Excerpt 5 A: G: G:

You wanna say no, or what? I don’t know (..) I don’t know Let’s try ‘no’ this time (...) (to the) when you do that it’s just like going to the different pages in the manual

These exchanges provide further evidence that whenever PRIDE presents technicians with a test or question that they cannot reconcile with the actual machine trouble, they will surmise whether answers to previous questions may have been incorrect; conversely, whenever the system displays tests and questions that appear to make sense, given the answer to the previous PRIDE question and the technicians’ understanding of the machine at hand, technicians’ confidence in the correctness of their previous answers increases. Notice, then, that the technicians take a stance toward the expert system that is similar to the one they take toward the document machine itself; they are continually accounting for the expert system’s output, attempting to understand its functioning in terms of their input to the system and their understanding of the problematic machine. As is the case more generally with machine diagnosis, whenever they do not understand the PRIDE system, technicians are likely to postpone judgment about whether they have been operating with a correct

Expert System Technology in Work Practice


analysis until they can account for its behavior. Only technicians can be the judge as to whether the intelligent system for diagnosing the machine is behaving intelligently and is going ‘in the right direction’. The intelligent system itself does not have access to the resources that could provide a sound basis for such judgment. Furthermore, consider that one of the ways technicians can judge whether or not the expert system is going down a promising path is to look ahead in the diagnostic chain and examine the consequences of entering one of the answer options. If technicians could appreciate where the expert system would take them if provided with a ‘yes’ or ‘no’ answer, they could better judge what answer to give it now. And this is precisely what technicians routinely do with the paper documentation; they scan an entire RAP to see what diagnoses and repairs it ultimately suggests (and assess the reasonableness of those repairs before even starting to follow a RAP). But scanning ahead is made much more difficult in PRIDE, as it is designed to enforce a step-by-step diagnostic process. Consider Grant’s last turn from the excerpt above ‘let’s try “no” this time (.) It’s just like going to the different pages in the manual’. Grant’s understanding is indeed correct and makes plain one of the major disadvantages of the expert system vis-à-vis the paper documentation – exploring the system for useful bits of information is infinitely more cumbersome. Users cannot easily get a sense of the entire decision tree, as the expert system forces them to go step by step. From Directive Documentation to PRIDE When Xerox decided to produce directive paper documentation for technicians, the goal was to provide technicians with a book that could guide them when they were doing diagnosis on a machine. As we reported above, the company did not want to rely solely on the diagnostic powers of the service technicians; they wanted to enforce proper process when the technicians were doing diagnosis. In practice, however, the service technicians used the (paper) directive documentation as a work aid, not as a prescription for doing diagnosis. That is, they continued to rely on their own ability to service their customers and diagnose machine problems, and would use the documentation only for suggestions and insights when they got stuck.2 Because the directive documentation was paper-based and its logic transparent (the binary tree), technicians could easily explore the documentation and search it for RAPs that they would consider useful given their interpretation of the troubled state of the machine. When the directive documentation had to be converted from paper to electronic form to be used on a laptop computer, the designers of the software were forced to deal with the technical limitations of the laptop screen; only a small fraction of a RAP could be viewed at one time. Whereas the RAPs printed on paper allowed the technicians to track down the branches of the binary tree by following the lines on the page (and, if necessary, using one hand to ‘hold’ their current position in a RAP while using their other hand to leaf forward to examine the 2 It must be noted that in the majority of service calls technicians do not use any documentation.


Orders of Ordinary Action

rest of the tree), this was no longer possible in the electronic version because of its screen size (cf. Yamauchi, Whalen and Bobrow, 2001). Given these limited navigational affordances of the laptop the designers of E-doc made the answers to the questions hyperlinks, so that clicking on an answer option would result in the system displaying only the test under that branch. Clicking on the hyperlink was an alternative way to navigate the RAPs; the technicians also had the option of scrolling down to see the entire RAP. While the PARC researchers designed the PRIDE expert system to be an intelligent front-end to the E-doc that would make it easier to navigate, the actual design had the opposite effect (consistent with the intention behind the directive documentation, namely, to be a step-by-step guide to direct the technicians to the diagnosis). The only way to ‘get to another page of the manual’ was to simply try the different answer options; no longer could technicians scroll to browse the full documentation. From the technicians’ perspective, then, PRIDE’s design denied them the very practice they had always applied to the documentation, namely, to explore it for useful tips to repair the particular problem at hand. Sharing Expertise through EUREKA As stated above, for expert systems to be successful tools in practice, two assumptions must hold. The first assumption – the expert system can only be helpful if it is supplied with the right information – was examined. This assumption is problematic as the only way the system can be supplied with the correct information is when the technicians can correctly account for both the defective machine’s and the expert system’s ‘behavior’. And the logic of the expert system is something that is not easily discerned, as it is so tedious to navigate. In this section we will examine the second assumption, namely that the expert system contains all the relevant diagnoses and repairs. When the PARC AI research group conducted some field observations of technicians they discovered some interesting facts about the utility of the service documentation. Technicians maintained that the problems the documentation addressed were just those problems that were relatively easy to diagnose and fix: that is, the documentation contained only the more or less straightforward diagnoses and repairs – the ones experienced technicians could easily fix themselves without help. What indisputably caused trouble in the field, technicians reported, were the unusual problems, along with especially nagging things like intermittent failures, for which the documentation could offer only meager assistance. The documentation (or expert system based on that documentation) based on standard knowledge and relying on laboratory fault tests was of little use to them with the problems that mattered most for practical purposes. For the most vexing problems, technicians were forced to come up with creative and sometimes quite unorthodox solutions, drawing on their own technical expertise in concert with the experience and inventiveness of their colleagues. The researchers had noticed that technicians typically annotated their repair books with their own ideas, which could include ‘corrections’ of the standard. They also kept cheat sheets that contained not only summaries or truncated versions

Expert System Technology in Work Practice


of procedures but also listed innovative workarounds and other unsanctioned but field-proven remedies to machine troubles (Bell et al., 1997). Technicians shared this sort of knowledge on a regular basis simply by talking about their experiences with machines on almost any occasion when they gathered together, exchanging ‘war stories’ (Orr, 1996) about encounters with curious and difficult problems, and how they diagnosed the problems and came up with new and ingenious repairs. Through cheat sheets and stories, service technicians continuously maintained their knowledge about machines and the most effective methods for diagnosing and repairing. The circulation of such material, however, was typically limited to members of small work groups. Moreover, Xerox had reduced the office space for field service, making face-to-face conversation among work group members more difficult and therefore less frequent. These observations inspired PARC researchers to design a ‘community knowledge sharing system,’ called EUREKA, a database for communal expertise, which would enable technicians to author and then share new solutions for servicing machines – or ‘tips’ as they came to be called – with their colleagues, not only across work groups and districts but national boundaries as well. The tips would be published and distributed in digital form, and therefore storing and searching for information in this electronic database would be much easier and faster than was possible through paper. In the EUREKA process, technicians first write up a tip into a simply structured form with fields for ‘Problem’, ‘Cause’ and ‘Solution’. Then they submit the new tip for validation by other technicians or engineers who are experts on that particular machine. Once a tip is validated it is added to the database and is automatically downloaded by all technicians who connect to the server. An example of a EUREKA tip is shown in Figure 9.11.

Figure 9.11 Example of a EUREKA tip

Orders of Ordinary Action


EUREKA has turned out to be a very successful system for Xerox. Today, there are nearly 20,000 users worldwide. In 2001, over a quarter of a million machine problems were solved using the system (Bobrow and Whalen, 2002). This accomplishment stands in marked contrast to PRIDE, which was never used in the field beyond its first testing on one machine. Most important for our interest is that from the very beginning EUREKA met with strong acceptance, even enthusiasm, in the service community. The reason EUREKA received such a warm reception in the field and PRIDE did not was fairly simple – it worked. Technicians told us that EUREKA’s value came from being thoroughly grounded in field experience, in communal expertise. EUREKA is a bunch of guys sitting together talking about problems they see and solutions they came up with. [It’s] almost like having another technician with you, because you can bounce your ideas or your thought processes off the EUREKA database, and to make sure that, ‘Okay, someone else has seen it, and this is what they did to resolve it,’ or ‘I’m on the right track,’ or ‘I’m not on the right track and I should be looking at a different area.’

Another technician put it bluntly: ‘This is the first time the company has truly listened to the field’ (Bobrow and Whalen, 2002). The differences between EUREKA and PRIDE stemmed from both PRIDE’s form, which embodied its directive approach and ‘knowledge philosophy’, and its information content. The actual content of the tips demonstrates that the technicians’ approach to machine problems and the solutions they propose are usually quite different from those found in the documentation, the standard repair procedures, which served as PRIDE’s knowledge base. As technicians write these tips for each other, they are marvelously revelatory of technicians’ life-world. Furthermore, the software used to organize, search, and read tips is also of interest insofar as it was designed to express an alternative ‘knowledge philosophy’ and a non-directive approach that placed great value on technicians’ practical reasoning when it came to information use. Accordingly, in the following section we give a brief description of the EUREKA software and then provide an analysis of a few of illustrative tips. We will conclude the chapter with some reflections on ‘intelligence’ and expertise in the work of diagnosing and repairing machines, and the role of technology in supporting and enhancing that practice. The EUREKA Software As stated above, tips in the EUREKA database contain descriptions of new solutions to problems, as well as repair workarounds and other practical information deemed valuable for servicing machines. There are three main data fields provided to structure that content: Problem, Cause and Solution. There is no further structuring; that is, the three fields can contain free text. This structure gives the tips an organizing principle, yet allows technicians enough flexibility to include as much detail as they deem necessary. Of course, there is a very practical interest in being able to carry out the actions described; the principal criterion for all tips, then, is reasonable clarity. Figure 9.12 shows the interface for the most recent version of EUREKA. Several features should be noted: a search area (marked A) with search parameters; a ‘hits’

Expert System Technology in Work Practice


Figure 9.12 Annotated example of a EUREKA interface area (B); the view area (C); and, on the tip selected to be viewed, the names of both the technician author and validator (circled). Searching is quick and simple, typing a text string in A brings up a list of tips in B, selecting a tip in B displays the contents of that tip in the view area C. Moving from one tip to another is just a click of a button away, and we often observed technicians scanning through a number of tips, moving rapidly from one to the next, looking for pertinent information. Our field observations and interviews suggest that technicians search for information using a few words of text only; overwhelmingly, they enter fault codes, machine components, part names (such as ‘fuser’, ‘heat roller’), or problem symptoms (for example, ‘jams’). Taken together, these features of interface were intended to afford as much user control as possible. The system does not direct the user to a certain tip; it simply presents a list and allows the technician to choose whatever tip they think might be most helpful to the problem at hand. The design thus attempts to encourage exploration of knowledge resources. Examples of EUREKA Tips There are now several thousand EUREKA tips, written by service technicians from many different countries, covering a wide range of problems and topics. When tips are compared to the directive documentation, to the RAPs, a number of differences stand out. One of the most interesting is that a significant number of tips propose modifications or ‘corrections’ of a machine, addressing what technicians consider


Orders of Ordinary Action

to be design flaws in specific parts; others contain suggestions for handling problems due to worn-out parts. In the following tip, for example, the service technician makes a suggestion to replace a part that is cut by manufacturing, but often cut too short (we display only the relevant text from the tip). In his description of the problem, the technician not only lists a fault code, but also points out that the code can be intermittent and occurs more frequently when the customer is trying to make double-sided copies of double-sided originals (represented in the Problem text by the designation ‘2/2’). The regular service documentation does not list any of these details, yet they are highly significant for diagnosis of this particular problem. The tip also includes a detailed explanation of exactly why it is more common to get this problem with double-sided copies, namely because the sheets ‘are not curled the same way as in the simplex [single-sided copying] mode’. The tip thus demonstrates the deep knowledge technicians possess of a particular machine model, as well as what they judge important enough to share with their colleagues. Note, as well, that the kind of fault that the service technician is describing – and this is typical for EUREKA tips – could not have been known when the product was launched, as it only cropped up later in the product’s life cycle, and not on every machine. Only after considerable time in the field, after some wear and tear, will a problem such as the one described surface. Finally, with respect to the flawed part (the actuator switch), one can imagine that it took a service technician considerable time before he figured out that the switch was cut too short. In all these ways, this tip captures knowledge about problems that only become apparent as a consequence of sustained customer use of the machine, requiring a deep understanding of the machine’s functioning, and sophisticated and persistent root cause analysis by the technician. The fact that document machine parts inevitably wear out over time also means that the specific age of a machine and its repair history are important contingencies that service technicians must take into account when doing diagnosis. In a sense, although machines can be considered to be more or less equal when they leave the manufacturing plant, once they start getting used by customers they develop in idiosyncratic ways based upon their particular context of use and their related service history. For every machine they service, technicians can tell a story that captures how the system is being used; for instance, whether it is being run hard or run by a good operator, what kinds of jobs the customer tends to run through the machine, and what major service work has been done in the past. These stories provide an important backdrop for the way a technician will approach any new problem. But the service documentation is written as if it applies equally to all machines of a certain family and, as this documentation has to be shipped as soon as the product is launched, it cannot contain information about how wear and tear will affect the machine’s functioning. Tips based on field discoveries thus play an important role in disseminating knowledge gained over the course of a product’s life cycle, as shown in Figure 9.13. Here the author suggests that a particular model’s document feeder may become warped due to heat or age, and proposes a repair for that problem. Note how the instructions for this tip are written for an expert reader; complete instructions are

Expert System Technology in Work Practice


Figure 9.13 Example of a EUREKA tip based on field discovery not provided, only some tips and pointers, the assumption clearly being that any competent technician can figure out the rest. An additional way that tips address matters not included in the documentation is to take local contingencies – such as troubles caused by weather extremes in certain regions of North America – into account. Often, the proposed solutions for these sorts of ‘locally induced problems’ require some modification of the standard machine. For example, a technician based in Western Canada authored a tip that addressed a problem due to low humidity, which can cause paper to lose its proper moisture level and then curl; as a result, the paper jams in the copier. Low humidity is common during the cold Alberta winters. The technician recommended using Mylar (polyester film) tape to slightly modify the size of the main feed chute into the copier from the paper tray. ‘This allows the curled corners of the paper which would normally strike the upper chute (resulting in a skewed feed) to be guided smoothly under the upper chute’, he writes, and then adds: ‘I have done this on a number of my machines and in every case the jam rate has been lessened! Perhaps Xerox could come up with a more permanent idea which could then be made into a retrofit! It works, try it.’ Finally, some tips go beyond the standard documentation by presenting new techniques for fixing things; that is, tips are as much methodological techniques as repair remedies. These ‘how to’ instructions thus reveal that for technicians, the sheer mechanical aspect of the job represents as much a practical problem as the problem of diagnosing the machine, but one entirely neglected by the regular service documentation. To be sure, some tips are simply about easing the job, as in ‘To make it easier to adjust [part] M, paint Whiteout [white correction fluid] on the back wall near M’ (a suggestion to help technicians see the precise location of a small component when peering through the dim light inside the copier). The key differences between EUREKA tips and the RAPs are summarized in Table 9.1.

Orders of Ordinary Action


Table 9.1

Key differences between EUREKA tips and the RAPs RAPs


Diagnostic reasoning

Step-wise diagnosis. Binary decision tree through yes/no questions.

Loosely structured by the Problem, Cause, and Solution fields. Comparatively, one ‘step’ from diagnosis to repair.


Written by field engineering or the repair documentation group

Written by technicians


Technicians should follow the documentation; documentation structure enforces proper diagnostic process

Technicians can explore the tip document and determine what parts of the text apply to the current problem

Based on the engineering specifications of the machine

Based on experience with repairing machines and resulting from discoveries (of new, undocumented solutions, workarounds and the like) made in the field

Source of knowledge

Although the Eureka technology is relatively simple (and was designed to be so), it represents a major shift in perspective for the corporation. EUREKA and its success fundamentally challenges the long-standing position that the corporation should be providing the technicians with documentation that forces them to diagnose the machine only in a way sanctioned by experts in product engineering. Even though the company had always suspected that technicians did not use the documentation in the manner intended, this was often attributed to the maverick culture of the service force. Few saw this ‘dissent’ for what it was, namely a rational practice developed by technicians responsible for maintaining machines that often broke in ways unaccounted for by the service documentation. Perhaps the most striking difference between EUREKA tips and the RAPs is that the vast majority of the tips contain detailed causal accounts that explain the problem, but no step-by-step instructions for determining that the tip is indeed the right diagnosis for the problem at hand. EUREKA tips are basically suggestions or recommendations; whether they apply in any particular case is left entirely up to the service technician. Concluding Remarks For the PRIDE expert system to have practical value, at least two assumptions must hold. First, the users of the system must be able to effectively mediate between the system and the real-world environment; that is, they must be able to ‘see’ the real world in the manner that the designers of the system intended it to be seen, and be able to translate what they see into correct input to the system. Second, the

Expert System Technology in Work Practice


designers of the PRIDE must have foreseen (all of) the ways in which the machines can fail. Our analysis of the PRIDE expert system in practice and the experience with demonstrates that both of these assumptions are problematic. Regarding the first assumption, in order to provide the system with the right input, the technicians must continuously reconcile the PRIDE’s output with what they know and can observe about the document machine. Whereas the system was built with the idea that the technicians should only be concerned with the local problem of giving the right answer to PRIDE’s questions – the diagnostic reasoning, after all, would be done by PRIDE – technicians were continually accounting for the expert system’s output and considering whether that output meant they were on a plausible path to diagnose the machine problem at hand. As technicians are the only ones that have the resources to see, feel and hear the troubled machine, they must be the ones that assess what part of the documentation applies in the current case. They treat the system as a mere repository of only potentially useful information – information that they would like to explore. The assumption that they could build an expert system that contained the intelligence led the designers of the system to a design that tried to control the diagnostic process and resulted in a system that technicians had to outwit in order to get it to display information they could use. There is no doubt that the system contained a host of valuable information, but as this information was embedded in opaque diagnostic algorithms it was almost impossible to find. For the technicians the system was excessively cumbersome and became an abject failure. Our ethnographic research and experience with EUREKA reveals that in the expert opinion of the technicians, the second assumption – that the documentation correctly anticipates what the problem with the machine actually is – is equally problematic. The directive documentation only captures the most common and most easily diagnosed problems. For the hard, intermittent, problems that cost them the most time, the documentation usually turns out to be of little value. The success of the EUREKA system provides further evidence for this. The thousands of EUREKA tips all contain solutions to machine problems that are not part of the standard documentation and thus would not have been part of PRIDE. As it turns out, there are myriads of ways in which machines actually fail that were unanticipated by the extensive laboratory tests that formed the foundation for the directive documentation and PRIDE; machines break down differently over time; they inevitably have design flaws that are not discovered until after the machines have been subjected to the wear and tear of sustained use. We end by noting that the assumptions that were the starting point for developing PRIDE are assumptions that are likely to hold for other expert systems efforts as well. The assumptions stem from a view of human intelligence as disembodied and reasoning as the syntactic manipulation of abstract representations. One consequence of this view was the development of expert systems, which came to embody the fallacious idea that human intelligence can be captured in abstract reasoning algorithms representing the authoritative and correct way of reasoning about certain domains in the world. The problem-solving capabilities of experienced practitioners, however, always exceeds this abstract reasoning as they can apply


Orders of Ordinary Action

their knowledge and experience inventively in each new situation. When the PARC AI researchers observed technicians’ practice of diagnosing machines, they realized that the very premises for PRIDE’s design were flawed. Their subsequent design of EUREKA represents not just another system in the evolution of service documentation, it represents a paradigm shift in the way in which human expertise is understood and appreciated.

Chapter 10

Thinking as a Public Activity: The Local Order of a Tibetan Philosophical Debate Kenneth Liberman

Introduction This is a study of some practices of reasoning of Tibetan Buddhist monks. Particularly, I am trying to learn what resources formal analytic reasoning places in the hands of able thinkers. The inquiry was initially motivated by my interest in what may be the limits of logical analysis, but gradually I came to examine more seriously as well what were the benefits that formal analytic reasoning offers to these scholars. My interest in what might be called a ‘philosophical anthropology’ was motivated by my reading of Edmund Husserl’s ‘criticism of reason’ in his Formal and Transcendental Logic and by Emmanuel Levinas’s account of the necessary betrayal by formal philosophy in his Otherwise than Being or Beyond Essence. What is it about formal analysis and logic that compels Tibetan monks to devote years and decades to its study when their aims are predominantly spiritual ones? And how diabolical, and untamable, can the constrictions that formalized thought places upon original thinking be, despite the philosophers’ best intentions? Even more basically, what is formal reasoning as an in-situ, empirical matter for ethnomethodological discovery? By ethnomethodology, I refer to the study of the local practices of congregations of people engaged in developing generic representations of their minutely detailed, social interaction. It is these locally produced generic representations that assist them in making their interaction orderly. Identifying and describing such detailed, intimate phenomena requires more than talking generally. It requires close investigation of the detailed, turn-by-turn dialectics of Tibetan philosophical discussion. Contemporary multimedia software assisted the detailed inspection of digitalized film of actual Tibetan debates. In undertaking this research, I spent three years living at three different Tibetan monastic universities. At these institutions of learning, young Tibetan scholars practise debating as the principal means of their philosophical training, spending from four to six hours daily in the courtyards of the monastery debating vigorously (see Liberman 2004). Buddhist monastic disputation achieved its lasting synthesis in India during the period of Dharmakirti (7th century CE), the great logician who formalized the rules for the debate that had been practised during the preceding half-dozen centuries. Philosophical debate has remained a daily part of Tibetan monastic life since Phyapa Cho-kyi Seng-ge (1109–69) refined and promoted the system in Tibet. This classical South Asian system of philosophical learning and inquiry was preserved on


Orders of Ordinary Action

the Tibetan plateau for nine centuries, long after it disappeared with the destruction of the Buddhist universities in the wake of the Islamic invasions of India. The 20thcentury invasion of Tibet by the Maoist Chinese and the ensuing Tibetan diaspora has afforded Western scholars a rare window on this ancient form of scholastic culture. While the Tibetans’ philosophical disputations are for their philosophical training, the monks also do it for the fun of it, a factor that has no doubt contributed to its longevity. In the courtyard of a Tibetan monastery, a hundred or more debates may be occurring simultaneously: one-on-one, two-on-one, and one-on-two. Because I could not tape-record debates during the white noise of a hundred or more couples debating in Tibetan at the same time, I recorded the doctoral-level scholars’ annual examinations during which there is only one debate at a time. Of the 20 hours of debating I recorded, I selected two-and-a-half hours of debate, consisting of 15 tenminute-plus debates on various epistemological topics. I painstakingly transcribed the debates in the original Tibetan (with the assistance of several Tibetan colleagues, some of them the debaters featured in the videos), and then I translated them into English. The Social Organization of Tibetan Philosophical Debate Briefly summarized, there are three parties in a Tibetan debate: the Clarifier of the reasoning (rTags gsal gtong khan, or ‘T’), the Respondent (rigs Len pa, or ‘L’), and the Audience. The clarifier of the reasoning (sometimes referred to as ‘the challenger’) poses philosophical questions to the respondent or ‘defender of the formal proposition’ (dam bca’ ba). The clarifier of the reasoning is always standing, and when his question takes the form of a formal analytic account of the debate, it will be accompanied by a vigorous hand-clap (indicated in the transcript by the sign >~) which terminates directly above the nose of the respondent, who is seated cross-legged on the floor. The respondent is required to preserve the consistency of his replies, since he is defending one or another philosophical position. But the challenger can approach his inquiries from multiple standpoints, in an effort to capture the defendant in an inconsistency or logical absurdity. It is important to appreciate that Tibetan debaters are not so much eager to win an argument as they are interested in engaging each other in producing an aesthetically pleasing, smoothly functioning, formal analytic examination of a topic. The collaboration is the real story, so it is not simply that there is antagonism. The ‘how’ of doing dialectics with competence outweighs the philosophical product. Our debate commences with the clarifier of the reasoning (T) posing the initial question to the respondent (L) T 6:02:53


One speaks of the realization of emptiness, right? Can you have that realization occur without realizing that an entity lacks a true identity? >~ Can a realization of emptiness occur without realizing that an entity lacks a true identity, huh?

In order to follow this debate intelligently, it is necessary to introduce the debaters’ topic, emptiness. For Mahayana Buddhists, emptiness is not something that exists ‘in

Thinking as a Public Activity


general’ like a phenomenon in the world, to which they pay some sort of obeisance (although many of the interpretations of European commentators in the 18th and 19th centuries treated it that way, leading them to wrongly describe the Tibetans as nihilists). Rather, emptiness is always an emptiness of something specific, such as the emptiness of my pen. My pen does not bear its own existence intrinsically as if its reality was derived from an inherent essence or Platonic idea; instead, it consists of the transitory aggregation of plastics, metal, ink, etc., along with a cultural context and the projection of a label and notion, ‘my pen’. Its pen-ness is therefore due to an adventitious confluence of events and not due to an essence that rests inherent within it, and so it is ‘empty’ of having any independent essence. Similarly, it lacks as part of its intrinsic identity any ‘mineness’, although should someone pocket my pen, I might at that moment feel a strong attachment to it and think of it as though its self-nature did include this ‘mineness’. The Buddhist notion is that we reify our notions and then become obsessed with those reifications, which in turn leads to a strengthening of our desire, which can lead to suffering. Here the debaters are considering the correct way to approach an understanding of emptiness. Since the matter is a profound one, the respondents repeat the question, shaking their heads as they pause to consider the issue. The question suggests another important aspect – one can be a proponent of a view of the emptiness of inherent existence on the basis of a strictly formal (and hence general) understanding of emptiness, without actually having any empirical realization of the more subtle aspects of emptiness. This issue amounts to a consideration of the issue of the relation between the formal analysis of the emptiness of essences and the actual realization of emptiness. The defender and the clarifier of the reasoning each repeat the question to get a better hold of it, and then the clarifier (T) begins to press for an answer, demanding, ‘Now is it yes or no! >~’ The parties come to agree that emptiness must pertain to a specific phenomenon, and they suggest the example of a pot as a phenomenon to consider. Then T poses a second question. (Capitalized phrases here indicate that the phrase is a formal token, which indicates to the parties that the accompanying statement is intended to acquire an abiding place in the emerging formal logic of the debating.) T

:33 L T



Now. When one has realized the emptiness of this pot, since we agreed that one does so on this basis. Truly, oh yes, do we say that IT DOES NOT NECESSARILY FOLLOW that one understands the way in which the pot still exists with some functional efficacy? >~ When one has a realization of emptiness/ /Oh. At the time one has a realization of emptiness. One says it is not necessary. It is necessary isn’t it? Mmm-hmh.

This second question refers to the important fact that the view of emptiness is not a nihilism. It does not deny the ‘functional efficacy’ that phenomena empty of inherent essences can have in the everyday world. Incidentally, this is the import of the famous line from the Heart Sutra, that runs, ‘Emptiness is form, and form is emptiness.’


Orders of Ordinary Action

‘Form’ refers to the fact that something is recognized to exist, so there is no nihilism operating; and ‘emptiness’ refers to the abandonment of any essentialism, and so no positivism or naive realism is operating either. A correct view of emptiness cognizes everyday events while understanding the emptiness of the phenomena that are in play; however, a limited view of emptiness may not sustain such a cognition of mundane events. The center of the debaters’ activity here is that the clarifier is laboring to develop a fully adequate formal account of emptiness, in order to deepen the parties’ understanding. He is fitting together, piece by piece, the essential elements of a competent account of emptiness. Here it includes the observation that emptiness is not an emptiness in general and that the specific phenomenon concerned must retain its functional efficacy. In such a fashion, T is assembling an adequate formal analytic version of emptiness: T L T



When one reflects upon the meaning of interconnectedly dependent arising, on the basis of the pot – – Yes. When undertaking an analysis of the emptiness of the way one customarily apprehends and reifies a referent object, oh. Yes. When one refutes, uhh, a referent object’s inherently true existence, then one posits its emptiness, doesn’t one? >~ Depending upon whether one does or does not realize the correct limits of emptiness [that is, what is to be negated and what is not], right, when one considers and reflects upon what is an acceptable way of formulating the functional conventional existence of the entity, which does have efficacy, without forgetting having negated the previous habit of reifying the entity’s being, there is what we call a realization of emptiness. >~

T is trying to locate the fine line between negating too much and negating too little. One does not want to deny the reality of a beautiful woman, for example, yet it is a truism that beauty rests in the eyes of the beholder. Similarly, although a truck is merely an adventitious collection of parts, such as tires, doors, a chassis, a steering wheel, etc., and so does not bear its own existence as an inherent essence, it can still run over someone, so its functional efficacy must be preserved during any philosophical analysis that focuses upon its emptiness. The defender (L) here is tenacious and will not grant T anything, frustrating T in his attempt to get all of the elements of an adequate formal account of emptiness together so that it can be authorized as a true proposition: 6:04



YOUR REASON IS NOT ESTABLISHED. There may or may not be a realization of emptiness. One who understands only the assertion of emptiness merely posits an emptiness, yes? Are you sure? In The Three Principal Aspects of the Path, right? Oh yes.

This is actually a very clever observation on L’s part, and one which fully recognizes the limitations of formal analytic accounts; i.e., proving that an assertion is correct may or may not be accompanied by a full understanding of its implications. But by being so contrarily clever L begins to anger T, who presses harder and brings in the big guns of an authoritative citation from Tsong Khapa’s The Three Principal Aspects of the Path, a full translation of which is included in the Appendix.

Thinking as a Public Activity


T’s presentation of an authoritative citation clarifies and bolsters the force of his analytic account. Following such an erudite as well as profound citation, L is obliged to accede to T’s position. T repeats the citation a couple of times, flaunting it before the defendants, in an effort to consolidate the position within the debate’s structure. His performance contributes a rhythm to the debating, a rhythm that is marked by hand claps. T [very rapidly] When the previous negation of the establishment of inherently true existence, of the customary mode of reification, is not forgotten, then is 6:04:41 what we call a realization of [a specific] emptiness then needed? >~ L It’s like this. In that case a realization of emptiness is not needed. T First of all, if one consults the text, it goes like this: But when the recognition of functional efficacy and the realization of emptiness do not alternate in one’s consciousness but are cognized simultaneously And one witnesses only the incontrovertible interconnectedly dependent being, then Such ascertainment thoroughly destroys the customary modes of reifying apprehended objects 6:04:56 And at that time we say the analysis of the view is completed. >~ L Yes. T This is what we understand to be the analysis of the view, the analysis of the Middle Way view. The present realization of the emptiness view is due to a 6:05 realization of a Middle Way analysis of the view. >~ L Yes. T This is the Middle Way view. This is how we present the mind that :03 realizes emptiness. >~ L Yes. T Right?

The challenger (T) works to objectivate every commitment, even those with which he may disagree. This provides for the parties clear communication, perhaps better than takes place when two Western scholars debate at an academic conference. By employing such interactive and formal logical mechanisms, Tibetan debaters almost never play the role of two ships passing in the night. The generalized versions of their philosophical inquiries that such objectivation produces may occasionally dumb down an inquiry, but it does provide for a communication in which all parties, including the audience, know what is at stake. This is thinking produced as a public activity. Here the work of validating a universal reason is social and not merely logical; or rather, the logical here is the social. Objectivation → Authorization → Disengagement After an objectivation is secured, T works to authorize it. This means that T must secure public warrant for the truth and correctness of the objectivation, usually in the form of some repeated concurrence on the part of the defender (for example, ‘Yes, that’s it’). Tibetologists bemoan the repetition in ordinary Tibetan debate, and some have suggested that it demonstrates a lack of originality. But the repetition here is playing a social function, not a logical function. Western scholars of Tibetan Buddhism who work only with docile texts and do not examine live debates miss the importance of the repetition for producing a philosophical object that all parties can


Orders of Ordinary Action

have access to in common. Both objectivation and the authorization of what is made a public object rely heavily upon the work of repeats. The authorization converts any discussion, no matter how casual or confused, into a formal analytic representation of itself. The authorization may also include a notable citation that is presented in order to anoint the achievement of the debating parties. Even the disengagement is a concerted activity, done by the parties in a congregational manner. Disengagement permits the parties to remove themselves and the contingency of their congregational products from the authorized, true formal account (in this case, of emptiness) in a fit of what Merleau-Ponty has termed retrospective illusion – the truths are not recognized to be local but are treated as if they have been always there, for all time. Here the adequacy of the philosophical findings is an adequacy whose correctness and whose status as universal truth are identifiably accountable matters. Ethnomethodology tracks the work of parties who employ such accounts. We have here a case of thinking as a public activity, but we have the phenomenon as more than the mere mention it receives in Clifford Geertz’s Interpretation of Cultures (1973: 360), although Geertz is well known for his observation, ‘Human thought is consummately social: social in its origins, social in its functions, social in its forms, social in its applications. At base, thinking is a public activity – its natural habitat is the houseyard, the market place, and the town square.’ In spite of the pregnancy of Geertz’s observation, it is still general talk, and researchers need to provide the observation some specificity. Harold Garfinkel (1977) elaborates upon this idea that thinking is a public activity: Accountable thinking and knowing must be put back into the world. And as soon as one begins to look for such objects in the world, one is simply overwhelmed by them. It is a strange policy which requires that all credible thinking be sought for in the heads. I mean as biologists’ heads, but I mean seriously ‘heads.’ You can bowdlerize the whole enterprise of thinking and knowing by putting it together in a manageable package by speaking of ‘mindful displays,’ ‘thoughtful speaking,’ ‘thoughtful action,’ ‘planful action,’ providing for its observability (since you need to do that) and once you have pinned it down via the devices that are provided for, that are ‘observable,’ then stuffing it into a head somewhere alongside other things in the head like ears, Eustachian tubes, electrical discharges, etc. The question here, however, is why would anyone want to turn his back on the very places where thinking and knowing are first encountered as the most obvious facts of life, namely in the looks of ordinary things … Nothing is available by introspection. We are dealing with the produced looks of things. It is worldly matters we are concerned with.

Edward Rose (1991: 7–8) comments further: We are avoiding the search for reality behind a skull. We are searching for reality out among us, out in the world. It’s clear now that the sensible reality is to be found out in the world, outside of my skull or your skull. We expect to reach that sensible reality only as we and others make sense together … We have to search out in a world where thinking is happening along with the rest that is indeed taking place and is made to happen there among people.

Thinking as a Public Activity


What Geertz addresses in a general way, ethnomethodological analysis attempts to specify, and the respecification of formal reason as a worldly matter requires that one capture thinking – and here, reason – as the ‘produced’ event that it is, that is, as a social activity. But our aim is not some sort of general notion of how reason is a produced achievement of philosophers; rather, our purpose is to identify and describe in its lived details the just-so’s and just-what’s of real and actual reasoning in this empirical setting. The accounts produced by Tibetan debaters are not merely individual offerings, they are through-and-through collaborative achievements. But the term ‘collaborative’ is still too voluntaristic and too Lockean. It deserves the repair that Garfinkel offers in Working Out Durkheim’s Aphorism: Ethnomethodology’s Program (2002) by his frequent use there of the term ‘congregational’. Garfinkel insists that ‘congregation’ is more apt and precise than ‘interactional’, ‘corporate’, ‘concerted’, or ‘collaborative’. ‘Congregational’ indicates that the parties are trapped by the structure and that the building-up of a line of reasoning is not so much a calculated deduction as it is the serendipity of the developing structures; that is, the interactive systems facilitating the dialectics constitute a social order. But let us also remember that the parties are not slaves to the emerging structures; since they have a hand in their production, they have means to manipulate the congregationally developed order or exploit what it reflexively brings to pass. As Eugene Gendlin insists (1992 and personal communication), the parties are always able to subvert the emerging structure, able to say more with whatever means end up being placed at their disposal (cf. Liberman 2007). The third step in this process of producing a universal reason, disengagement, has an embedded form that may be witnessed in the rhythmic way that T and L concert their utterances from 6:04:39 to 6:05:05 (see above), a rhythm that almost becomes mechanical. The rhythm of the debate, once put into play, permits the debate to run as if it was on autopilot, like the inexorable logical machine that the debaters aspire to become. These Tibetan debates occasionally operate like a toy car that a child can wind up with a few passes of the spring-loaded rear wheel, only to allow the car to run on its own for a spell, independently of the child’s hand. In that independence is embodied the debaters’ claim to the very universality of their formal analytic work. One can observe the confidence and pride of T, who stands with his hands folded at 6:05:14, the implication being that the analytic work is now able to operate on its own without requiring any hands-on work: T

6:05:14 L :18




Right? If we speak of when the mind that realizes emptiness is completed, and one comes to cognize the interconnectedly dependent being of an object, there has been a previous refutation of an inherently true essence, right? If the awareness of this negation has not dissipated, and we are able to retain it while also recognizing just how an entity abides with a functional efficacy, at that time we must be cognizant also of what is common to the customary apprehension. >~ [T folds arms] When we realize emptiness we don’t need to pay attention to the customary apprehension. Oh! If we do not, EXPLAIN HOW one analyzes the view. Merely cognizing emptiness and ‘completing an analysis of the view’ are different are they not? Oh. ‘Completing an analysis of “the view” ’, he says. This is extremely wrong-headed! THE THREE PERVASIONS!


Orders of Ordinary Action

L continues to confound T by refusing to go along, which animates T’s dance that always accompanies his presentation of the formal propositions. T presses L for a full explanation at 6:05:18, an interactional move I have termed ‘put-up or shutup’. L offers merely a pedantic reply, which T ridicules before publicly declaring L’s failure (‘THE THREE PERVASIONS’, a term that refers to the poverty of the respondent’s logical analysis). T is frustrated enough with L that he follows his declaration of L’s failure with an argument by absurd consequence, perhaps the favorite logical move of Tibetan debaters: T IT FOLLOWS THAT the cognition of emptiness and the generation of a mental continuum of the Middle Way view are not a completion of the Middle Way 6:05:31 analysis of the view. >~

This proposition is absurd because it is practically the denial of a tautology – a Middle Way practitioner who generates a mental continuum that cognizes emptiness is the definition of the completion of the Middle Way view. Note that the reference here is to an experience and not to a formal assertion. But L has to remain consistent, although he is off balance and stands in the face of further ridicule (which he will soon receive), so he holds firm, placing himself out on a limb, a limb that T moves in to cut off: L T

I will concur. How can that possibly be how it is?!

T proceeds with the work of philosophers, in other words the work of assembling an adequate formal analytic account of (in this instance) emptiness, adding the important distinction between the habitual reification of inherent essence that is conceptually imputed and those reifications or fixations that are innate. Conceptual imputations are projections that humans engage in, especially philosophers, and cover cases where people are victimized by their own intellection. These imputations are acquired ones. Innate reifications are those fixations that all creatures have, including cows whose intellection, presumably, is minimal. The former projections are the easier to abandon, while the latter are nearly intransigent despite diligent efforts to remove them. The distinction is an important component of any adequate formal account of emptiness. T 6:05:55 L





Now, why do you speak of it this way? The text presents an example of the negation of the habitual reification of true existence that is conceptually imputed upon a referent object. In this system, that is not necessarily a realization of emptiness. >~ If one refutes the habitual reification of true existence that is conceptually imputed, eh? Oh, so. In this system, eh? [pregnant pause, as T waits for L to commit one way or the other] The habitual attitude of true existence conceptually imputed upon the referent object is understood to not have any existence. Yes, yes. Now, it would not necessarily be a realization of emptiness, would it? According to this Middle Way Dialectical School. Oh, it’s not necessary.

Thinking as a Public Activity


T’s proposal (at 6:05:55) is competent, but L stalls by calling for identification of which school of philosophical tenets T is considering (:57). This stall is a move in the game, and it gains L time to consider how to handle the newly introduced element. T pushes the matter, and so L goes along, and T repeats the point (at 6:06) to firm up the commitment. An aim of Tibetan debate praxis is to secure clear and firm commitments. At this point, L introduces his own idea, referring to the Syllogism Reasoningbased School (specifically, the Sautrantika Svatantrika Madhyamika), which has a less sophisticated understanding of emptiness. Since the challenger (T) is obligated to ‘pick up’ any matter that a defender introduces, objectivate it as well, and analyze it in a formal way, T carries out an exegesis of L’s topic: L 6:07


The Syllogism Reasoning-based School has its own understanding of the referent object. This is so. Ohh. Their analysis, which proceeds from considering referent objects to be merely a collection of directionless, partless atomic particles, teaches us only a rough sort of emptiness.

Some lower schools of Buddhist philosophical tenets have a physical theory of emptiness that depends upon recognizing that an aggregation of atoms that composes an object, such as a pot, does not involve an inherent identity – only partless atomic particles could be so inherently existing; however, even atomic particles have directional sides (north, east, etc.), so actually they are not ultimate either. This is a discernment of emptiness that is somewhat less sophisticated than an understanding of the interconnected dependence of notions upon each other, with no single idea capable of independent, intrinsic existence. At this point it seems that the thinking of T and L is coming into alignment at long last. But it is an identifying characteristic of Tibetan philosophical debate that just as the parties seem to come to a consensus the challenger will pose an argument by absurd consequence. Accordingly, T carries his exegesis a step further, to the point of an absurd consequence: 6:07



For this reason IT FOLLOWS THAT such a basis for realizing emptiness only permits the realization of a lack of true existence of the conceptual imputations that are projected upon a referent object. CHEE-CHEER ! [‘WHAT IS THE REASON?’ that is, ‘No.’]

L rejects this, since although the Syllogism Reasoning-based School’s understanding is less sophisticated, it is going too far to claim that they are unable to correct any innate fixations. But T keeps pressing. At 6:07:49, L throws T a bone in the direction of T’s contention, but T tries to extend the concession too far by proposing (at 6:07:51) another absurd consequence, which L emphatically rejects. T

6:07:49 L T :51 L

In that way, the Syllogism Reasoning-based School – right – has not identified the referent object that is innately produced. Within the terms of the designation of their philosophical tenets, uh, there is still some sort of habitual reification of an entity’s ‘true’ essence, right? Right. Merely such a refutation does not establish the cognition of a phenomenon’s lack of existing with any true identity. >~ THE REASON IS NOT ESTABLISHED!


Orders of Ordinary Action

L rejects it because the Syllogism Reasoning-based School does have some cognition of such lack of inherent existence, only not enough of it. T responds by extending the consequences of L’s argument to the point of blasphemy (6:57:59): T

If that reason is not established, for this system, right? IT DOES NOT FOLLOW THAT one needs any realization of emptiness in order to refute some sort of true existence of a pot. >~ CHEE-CHEER, CHEE-CHEER!

6:07:59 L

This is a great absurdity. It suggests that if a mere formal assertion about emptiness, without any realization of emptiness, can refute true, inherent self-identity, then a realization of emptiness becomes superfluous. The debate has now become intense. Upon receiving L’s emphatic rejection, T insists that L ‘put up or shut up’: 6:08:10




Oh, PRESENT how one refutes having an imputed conceptualization. The / / The imputed conceptualization of a reified ‘true’ essence, one who is having an imputed conceptualization of a reified, ‘true’ essence can be one who has a realization of emptiness, can he not? That’s right. And one who refutes having an imputed conceptualization of a reified, ‘true’ essence can be one who has not realized emptiness, isn’t it? That’s right.

The conclusion of the discussion at 6:08:12 to 6:08:18 is that a formal refutation of imputed essences may or may not entail a realization of emptiness, depending upon the depth of the reflections of the person concerned. In addition to being an effective and collaborative summary of this portion of their discussion, it also demonstrates that T has been listening to L well. The practical interest of the debaters here is to work their reflections into public formulations, to make their collaborative utterances into a congregational affair. T L 6:08:21 T L T L

Oh, now HOW DOES ONE FORMULATE THIS? According to the analysis of the Syllogism Reasoning-based School / / >~ According to the analysis of the Syllogism Reasoning-based School there is an imputed conceptualization of a reified ‘true’ essence, isn’t there? Oh. There is also an innate proclivity to objectify an entity, but here there is no apprehension that considers well the essentialist reification of an entity’s being that is produced by this innate proclivity to objectify. And such a subtle realization in terms of an awareness of the essentialist reification of an entity’s identity by the innate proclivity to objectify is required.

Although the account L offers is a little weak, he provides in his account an element – the innate proclivity to essentialist reification – that T quickly utilizes to ask a question of considerable probity: if removing the innate proclivity to essentialist reification is the critical task, then why do scholars devote any of their attention to the essentialist reifications that are created by conceptual projections (6:08:30)? This is an example of Tibetan dialectics successfully leading to an interesting question.

Thinking as a Public Activity 6:08:30


:39 L


Oh, TAKE THE TOPIC the analysis of the Syllogism Reasoning-based School regarding the customary mode of reifying an essential nature. When one negates an innate, spontaneously arising object of attachment, right, in this system why then must one refute the object of attachment that is a conceptually imputed reification of a truly existing essence? >~. IT FOLLOWS THAT one must BECAUSE according to their system, there is a reified apprehension of a ‘true’ essence that is a habit of conceptual imputation.

This matter begins a series of entanglements for L, which begins with T’s proposition at 6:09:09, which is a negative formulation that L rejects. Tibetans are so accustomed to negative dialectics that they prefer that the first formulation of a matter be presented in an incorrect and negative form. Then, following its rejection, they can propose a correct and positively worded version of the same notion. This helps to establish a rhythm in the questioning, which they enjoy. However, it also affords T an additional opportunity to catch a defender accepting an incorrect proposition, or (even more diabolically) the challenger can propose a correct thesis in a negatively worded form in the hope of securing the defender’s rejection of a true thesis. Such strategic matters serve to keep defenders very attentive throughout the course of a debate. T 6:09:09 L

Uhh, when the object of attachment produced by an imputed reification is refuted, in actual fact IT FOLLOWS THAT one does not necessarily cognize emptiness. >~ I ACC- WHAT IS THE REASON that it follows that one does not necessarily cognize emptiness? There must be a cognition of emptiness.

At 6:09:09 L adds to his rejection (that is, ‘WHAT IS THE REASON?’) the assertion that there must be a cognition of emptiness when an object of attachment produced by an imputed reification is refuted. The reader should take careful note of this, due to the strategic role this commitment will play in the debating that is to follow. For his part, T picks up L’s assertion and objectivates it by formulating it clearly even though he disagrees with it. This is proper debate protocol. T also gains L’s authorization for it (the ‘Yes’ at 6:09:12): :12


There must be a cognition of emptiness. Yes.

T’s repeat of L’s assertion offers no additional information – it is only the practical work of exhibiting the order. And the order here is logic: part of the formal grammar of logic is used not because it has a logical role but because it orders the talk-indebate. That is, logic is being used as an organizational device. Whatever else logic may be, logic here is a social practice with which the parties can make of their thinking a public activity. The parties here are engaged in the practical, formal, exhibitable work of being philosophers. T once again objectivates-by-repeating the assertion at 6:09:19, making it an abiding acquisition of the debate. 6:09:13 T L

There is what we speak of as the refutation of an object of attachment that is created by an imputed conceptualization of an habitually reified ‘true’ essence, isn’t there? ‘Subsequent to being subjected to the three – the basis, its appearance to awareness, and the path of analyzing and meditating upon it.’


Orders of Ordinary Action :15


:18 :19


Oh, in accord with ‘Subsequent to being subjected to the three – the basis, its appearance to awareness, and the path of analyzing and meditating upon it’, there is a type of refutation of an object of attachment produced by an imputed conceptualization of a reified ‘true’ essence. It’s the same. Exactly, exactly, exactly. And there must be a cognition of emptiness. >~ I ACCEPT that there must be.

It is here that T offers the first of what will be four instances of what I term ‘pulling the rug out from underneath the defender’ (compare, Socrates: ‘tripping them up and turning them upside down, just as someone pulls a stool away when someone else is going to sit down,’ Plato 1961: 392). In the breath that follows his mocking of L’s certainty (‘Oh, because “there must be”?’ [‘O dgos par song tsang’]), T declares just the reverse (‘IT FOLLOWS THAT there must not be!’ [‘mi dgos pa yin par thal’]). This is done so swiftly that it may be likened to a move in aikido during which a party deftly uses an opponent’s own momentum to lay him flat on the floor: L T

:29 L

I ACCEPT that there must be. Oh, because ‘There must be?’ IT FOLLOWS THAT there must not be! Right? BECAUSE ‘Subsequent to being subjected to the three – the basis, its appearance to awareness, and the path of analyzing and meditating upon it’, that is the way that the analyses of the lower tenets establish how entities are apprehended to truly exist by virtue of their own defining characteristics. In this system there are such fixations. >~ Yes, sure.

After T demeans that position as merely being the view of lower schools, the defender meekly goes along. But now T, perhaps still irritated with L’s lack of cooperation earlier in the debate or for the benefit of the examiners, is relentless. He moves in with another first negatively formed, then positively formed sequence, but L accepts the very reasonable initially negatively formed proposition, perhaps in hopes of arriving at some happy accord – one can witness in the video record the gratuitous head-nodding of L in this passage: T L T

6:09:32 L

When one refutes that sort of reified fixation, ‘Subsequent to being subjected to the three – / / That’s right. // // the basis, its appearance to awareness, and the path of analyzing and meditating upon it’, IT DOES NOT FOLLOW THAT it is a refutation of the object of attachment produced by an imputed conceptualization of a reified ‘true’ essence. >~ I ACCEPT.

So just as it appears that at last T and L are coming together, and with L nodding hopefully, T pulls the rug out again, being contentious with his positively formatted version. In order to remain consistent and to preserve some of his dignity, L rejects this positive version (at 6:09:36). T performs more of his work as the clarifier of the reasoning and objectivates L’s position (at 6:09:37), secures L’s authorization (‘I ACCEPT’), and then pulls the rug out once more, contending just the opposite of what T and L have just authorized:

Thinking as a Public Activity T 6:09:35 L T :37 L T :44 L


[Quickly spoken] IT DOES FOLLOW, BECAUSE it is a refutation of the object of attachment generated by the habitual, imputed conceptualization of a reified, ‘true’ essence. >~ THE REASON IS NOT ESTABLISHED. IT FOLLOWS THAT for such a person as this the object of attachment produced by an imputed conceptualization of a reified, ‘true’ essence is not refuted. >~ I ACCEPT. IT FOLLOWS THAT there is a refutation BECAUSE according to the lower schools of tenets’ analyses, methods and commentaries regarding the establishment of true existence, the true existence, uh, a type of reified apprehension is refuted. >~ Yes.

Now the debate is moving extremely quickly, and this is a feature that is characteristic of Tibetan debates when the formal analysis is well in play. The debaters seem to enjoy the very speed their dialectics can gain. T begins to set a trap for L, who (at :44) goes along with the compromise notion that ‘a type’ of reification is refuted, but in his succeeding turn (after :46) L persists almost blindly with his position that the object of attachment is not refuted: T :46 L

[Very quickly spoken] When one refutes this IT DOES NOT FOLLOW that one has refuted the object of attachment generated by the imputed conceptualization of a reified ‘true’ essence. >~ I ACCEPT.

Still speaking quickly, T extends this assertion to include the notion that such a person does not necessarily have a cognition of emptiness. In the face of the speed of the dialectics at this point, L is so engaged with the logical issues that are most proximal that he loses track of his earlier commitment (at 6:09:10–12) and responds that a cognition of emptiness ‘is not logically entailed’, but that response contradicts his commitment that we noted there that ‘there must be a cognition of emptiness’ when an object of attachment produced by an imputed reification is refuted: 6:09:47




[Quickly spoken] Now, when one refutes the object of attachment generated by the imputed conceptualization of a reified ‘true’ essence, is it necessary to have a cognition of emptiness? If you’re speaking of the imputed conceptualization of a reified ‘true’ essence, then that is not logically entailed, is it?

At 6:09:50, T objectivates L’s latest view, making a public exhibit of it, and T then points out L’s inconsistency and ridicules him for it: 6:09:50


6:09:58 L

Oh this, he asserts that there is no logical entailment. Here one has refuted the imputed conceptualization/ of a reified, ‘true’ essence,// / Now here, (- unclear –) // Now above we have already accepted such an entailment. But nevertheless here it is not necessary, is it? Mm. When one has refuted the imputed conceptualization of a reified ‘true’ essence, uh, we say one must have a cognition of emptiness. >~ I ACCEPT.


Orders of Ordinary Action

Once the clarifier of the reasoning publicly recalls the respondent’s earlier commitment, L is obliged to fall into line. At last, it seems that the labored dialectics of these debaters has painfully reached some consensus. But just then T diabolically pulls the rug out from under L one final time, declaring in a single breath, 6:10


Oh. it is necessary, isn’t it? IT FOLLOWS THAT it is not necessary.

At this point L seems to be dangling in the wind, practically ready to say ‘yes’ to ‘no’ and ‘no’ to ‘yes’. T observes that the most profound school of philosophical tenets would not concur with L’s contention, but L meekly tries to hold his ground: T :05 L

In this system, right? In the mental continuum of the Dialectical Middle Way School, uh, the imputed conceptualization of a reified, ‘true’ essence is not an object of concern. >~ If you say because there is no room for the imputed conceptualization of a reified essence for the Dialectical Middle Way School, THE REASON IS NOT ESTABLISHED.

It is noteworthy that a nervous flicker moves across the face of L at this point. It is unclear what are the benefits to philosophical insight of such nervous flickers or what the hermeneutic benefits of public ridicule may be, but L seems to retreat into a sort of meditation with his eyes closed, which in itself may be a recognition (or L’s hope) that there is a truth that resides beyond the truth for formal analytic dialectics. Conclusion What is important for us wanting to assess formal analytic reasoning is that we witness the looks of things from the perspective of L and that our appreciation and understanding of Tibetan debate is not merely that of a remote analyst who works with docile texts at a distance from the real world action. Rather, the horizons of meaning that operate for the parties must be retrieved by means of an ethnomethodologically specified description of the details of the dialectics. That is, what the parties are doing in assembling the orderliness of their inquiry, step by step, must be described from the perspective of the parties, and the public life of the thinking that takes place should be captured in a phenomenologically adequate way. Just as an analyst who observes a cat cornering a rat can hardly capture the horizon of events for the rat from the perspective of the kitchen window but must assess the looks of things and judge any operative creativity from the perspective of the rat, so I have tried to provide an account that recovers the looks of an ordinary Tibetan debate and the work of the debaters organizing its orderliness. Only from the perspective of these Tibetan philosophers, engaged closely with the local details of their dialectical practices, will Western scholars be able to assess fairly what is original or mundane about Tibetan philosophical culture and what are the tangible benefits and limitations of formal reasoning there.

Thinking as a Public Activity


Appendix I have provided in my translation of T’s citation at 6:04:14 the additional preceding four stanzas, in order to set the philosophical context for this citation of Tsong Khapa’s authoritative stanza: T L T

In The Three Principal Aspects of the Path, right? Oh yes. [Without the wisdom that realizes the way things abide, Even though one has cultivated renunciation and the altruistic mind, One will be unable to cut the roots of mundane existence Therefore, one must apply oneself in the methodology for fathoming entities’ interconnectedly dependent being. Whoever views all phenomena To exist only by causal relations, without exception, And destroys all tendencies to concretize them Has then entered the path that is celebrated by Buddhas. One who sees the incontrovertible interconnectedly dependent being of what appears And has an understanding of its emptiness that is not merely dependent upon theorization Yet views these two to be separate events Has still not realized the thought of the Able.]

6:04:14 L

But when the recognition of functional efficacy and the realization of emptiness do not alternate in one’s consciousness but are cognized simultaneously And one witnesses only the incontrovertible interconnectedly dependent being, then Such ascertainment thoroughly destroys the customary modes of reifying apprehended objects And at that time we say the analysis of the view is completed. >~ I ACCEPT.

This page intentionally left blank

Chapter 11

Cultures of Reading: On Professional Vision and the Lived Work of Mammography Roger Slack, Mark Hartswood, Rob Procter and Mark Rouncefield

The relevant unit for the analysis of the intersubjectivity at issue here is thus not these individuals as isolated entities but ... a profession, a community of competent practitioners, most of whom have never met each other but nonetheless expect each other to be able to see and categorize the world in ways that are relevant to the work, tools, and artefacts that constitute their profession (Goodwin, 1994: 615).

Introduction1 In this chapter we provide an ethnomethodologically predicated account of the lived work 2 of visually inspecting, or ‘reading’, mammograms (X-ray films of the breast) for signs of abnormality. It would seem to us that while critiquing cognitivism is ‘old news’ it is important to provide alternates to the impoverished ‘slices of practice’ that such accounts give. We want to move away from accounts that treat of work such as reading mammograms as ‘wholly internal phenomena [whose] reality is ... said to consist in either a purely biological mode of existence ... or in an emergent, “computational” or “functional” set of properties of neural events and processes’ 1 We wish to express our thanks to the staffs of the various UK NHS breast screening clinics and to Rod Watson, Michael Lynch and Andrew Carlin. The research is part of an ongoing engagement with the UK National Health Service Breast Screening Programme. This chapter is based on a study of the work of reading mammograms in order to develop computer-based decision aids for breast screening. Our study involved observation of work practice at six breast screening centres over a period of between two months and one week. In addition, we have been involved in studies of readers reading in conjunction with computerbased detection aids (Hartswood and Procter, 2000). 2 In this sense we follow Livingston’s use of the term. As he says apropos of lived work: ‘We come to the following point: a prover, in the course of working out a proof, extracts from the lived work of that proof, the accountable structure of that work – that is, he extracts the specifically remarkable features of the presented proof and does so against the background of practices that both provide for that structure, and that, simultaneously, that structure makes available. A proof, then, is not the disengaged, material argument, but is always tied to the lived work of that theorem’s particular proving: a proof is this inextricable pairing of the proof and the associated practices of proving’ (1986: 14).


Orders of Ordinary Action

(Button et al., 1995: 51–2). As opposed to taking some experimentally driven slice of work, we propose to treat reading mammograms as lived work and to argue that it is important to appreciate the social nature of the work and the embodied nature of reading and annotation. We are not setting up a straw person in cognitivist accounts of reading but specifying an alternative means of looking at practical actions that does not ‘top-slice’ some phenomenon and thus render it stranger to members involved therein.3 In the radiological imaging literature (see Kundell, Nodine and Toto, 1991) it would seem that there is a picture of the solitary radiologist undertaking what have been formulated as essentially cognitive operations in isolation from colleagues: this chapter is an attempt to re-introduce the sociability of reading practices and to show how readings are intersubjectively, interactionally accomplished objects within a social context as opposed to looking ‘behind the skull’ as Garfinkel (1967a) puts it. Our research in mammography stems from an interest in the ways that technologies can support work in medical settings, and we shall have cause to make some comments on this issue towards the end of the chapter. Here it is our intention to attend to the haecceities of the setting, to show how practical actions such as the arrangement of mammograms, gesturing and pointing to features on mammograms, manipulating mammograms, and annotations are all components of the lived work of doing reading (Livingston, 1986). We will show how what we call ‘breast biographies’ are realized in and as a part of a course of practical actions that constitute the reading process and that these biographies are themselves a part of this process. In other words, there is a reflexive relationship between biographies and the practical actions that constitute them. In addition to the above, we are concerned to develop Goodwin’s (1994) notion of ‘professional vision’, arguing that radiologists’ ‘professional vision’ and their associated (often written) ‘shop talk’ (Lynch, 1996) constitutes mammograms as a part of their craft in a manner that is different from, say a general practitioner or nurse looking at those same mammograms. We will show how the situated practical actions of reading of mammograms constitute the radiologists professional vision as a ‘socially organized way of seeing’ (Goodwin, 1994: 606). Reading as Practical Action Analysis of the methods used by members of a community to build and contest the events that structure their lifeworld contributes to the development of a practice-based theory of knowledge and action (Goodwin, 1994: 606).

Screening’s Work In the UK, women between the ages of 50 and 64 are invited every three years for breast screening. The initial test is by mammography: one or more X-ray films (mammograms) 3 The body is much more than just a makeweight. Practical activities have to be seen as embodied conduct involving ‘work-relevant perceptual structures’ (Goodwin, 1997) as well as the body: thus we attend to talk-in-interaction, written inscriptions and annotations as well as to gesture, manipulations and other embodied actions in and as a part of reading.

Cultures of Reading: Mammography


are taken of each breast by a radiographer and examined for evidence of abnormality by at least one reader.4 Several types of mammogram features are early indicators of breast cancer: micro-calcification clusters are small deposits of calcium visible as tiny bright specks; ill-defined lesions are areas of radiographically dense tissue appearing as a bright patch that might indicate a developing tumour; stellate lesions are visible as a radiating structure with ill-defined borders; architectural distortion may be visible when tissue around the site of a developing tumour contracts; asymmetry between left and right mammograms may be the only visible sign (cf. Garfinkel, Lynch and Livingston, 1981, discussed below) of some features. As a part of the routine procedure when the mammograms are taken, the radiographer will take a short medical history from each woman. This will include questions about whether the woman has felt any lumps in the breast during self-examination, if she has felt any pain and, in some centres, whether she is taking hormone replacement therapy (HRT). This will be written down on the screening form. In addition, the radiographer will examine the breast for moles, ‘skin tags’ and scars; the occurrence and position of these will be noted on the screening forms’ schematic of the breast since they may show up as suspicious objects in the mammogram. The X-ray machine is adjusted for the desired views: mediolateral oblique (known as oblique or MLO) – taken by compressing the breast in a diagonal from the shoulder to the stomach; and craniocaudal (known as CC) – taken by compression of the breast horizontally. Both views will normally be supplied to the reader on the first screen and may also be taken if the woman is recalled. An important component of the radiographer’s skill lies in positioning the woman in relation to the X-ray machine to ensure that as much breast tissue as possible is represented on the mammogram. Although radiographers apply ‘textbook’ techniques, considerable improvisation is often required to ‘position’ women correctly because of variations in physique. Once the mammograms have been developed they are further transformed by the addition of paper labels to indicate the screening round.5 At this point the mammograms might be subject to repair if the labelling is found to be deficient. The mammograms will also be examined by the radiographer for ‘technical quality’ – to determine whether they will meet the accepted criteria for ‘diagnostic mammograms’; that is, to ensure that they will, in fact, be ‘readable’. If the mammograms are deemed to be technically deficient then ‘repeats’ may be taken. The radiographer will also note if the woman was ‘difficult to position’ for some physiological reason, and might then indicate that such mammograms will be the best expectable in these circumstances. The ‘completed’, ‘readable’ mammogram is then made available for reading. Mammograms associated with a contiguous screening session (or clinic) are loaded onto automated viewing boxes and are typically grouped together in ‘batches’, corresponding to all the women who attended a particular contiguous screening session or ‘clinic’. There is often paperwork associated with each of these batches describing, for example, how many cases are in each batch, who loaded the mammograms onto 4 We use the term ‘readers’ here since not all members who read mammograms will be radiologists – some will be breast clinicians and, in some centres, radiographers. 5 The first time a women attends for screening is called the prevalent round. Subsequent screenings are referred to as incident rounds.


Orders of Ordinary Action

the viewer and the date the viewer was loaded. Subsequently, this will include the dates when the mammograms were read, and who performed the reading. A reader will then visually examine each of the mammogram sets6 and identify possible abnormalities (see Figures 11.1a and 11.1b). Their deliberations are recorded on a screening form, which is discussed later in this chapter (the form for this particular case is shown as Figure 11.3). Time pressures mean that readers work quickly and on average the examination of each set of mammograms takes less than one minute. On conclusion of the reading of each set, the reader makes one of the following decisions which is then recorded on the screening reporting form belonging to that set: • •

Return to routine recall: where the reader decides that the case is normal and the woman will be invited to attend for screening after the standard interval. Recall for assessment: here the reader may have detected a possible abnormality and will recommend that the woman attend an assessment clinic for further tests. Technical recall: here the reader decides that the diagnosis may be inaccurate because of imperfect mammography and that repeat mammograms may be required.

As we shall see, the degree of uncertainty about a suspicious object or lesion can vary considerably. Some objects are unequivocally malignant while others may be only mildly suspicious. We should also note at this point that all of the screening centres observed practise double reading where each mammogram is examined independently. Double reading is practised consistently (at least as far as staff resources allow) and is seen to increase cancer detection by 5–15 per cent (cf. Blanks, Wallis and Moss, 1998). The outcomes of double reading vary: in some centres it is the ‘worst opinion’ which generates a recall whereas in others there is discussion of the case or third reader arbitration. The interactional character of such disputes is of interest to us, and we shall discuss the manner in which versions are realized and disputed below. It is interesting to consider the arrangements of mammograms in the realization of what we have called breast biographies. Mammograms are arranged to be viewed in a manner that renders the biography of a particular breast visible.7 Mammograms from previous screenings are juxtaposed with those from the current round. Practically, this enables the radiologist to assess if any changes have taken place and to examine features in a retrospective–prospective manner under the rubric of the documentary method of interpretation (Garfinkel, 1967a). Diagnosis, then, is contingent on examining and providing an account of features that may be on one set of mammograms and not on another – in and through the breast biography readers make a decision as to what it is that they are seeing. In practice the achievement of 6 A mammogram set refers to all the mammograms taken of a specific woman. It will consist of at least two mammograms and usually four. We will attend to ways in which the diagnostic character is achieved through retrospective/current arrangements of mammograms below. 7 See Figure 11.2 for images of the placement of mammograms on viewers.

Cultures of Reading: Mammography


Figure 11.1a Mammogram showing right and left oblique views Together with CC views and mammograms from previous screening rounds, this would be placed on a viewer for reading.

Figure 11.1b Mammogram with annotations 1. Nipple area – the retro-aureole area behind the nipple is one of the ‘danger areas. [Not indicated: 2. Glandular disk. 3. Retro-glandular area (sometimes called ‘milky way’) – also a ‘danger area’. 4. Pectoral muscle.] 5. Potential lesion - ill defines mass (circled).


Orders of Ordinary Action

a biography involves the synchronic reading of diachronic features afforded by their placement and the practical action of looking here and now that readers undertake. This holistic view of reading contrasts markedly with the more cognitivist accounts of reading wherein the act of reading is sliced in a variety of ways in order to account for the ways clinicians make a diagnosis. The artfully achieved historicity of the arrangements adds another dimension to the practice of reading mammograms. An object becomes seeably suspicious in and as a part of this biographical arrangement, not as an isolated object read by an isolated reader. As noted above, the brief medical history taken by the radiographer is a resource on which the reader can draw in the interpretation of objects in the breast that might otherwise appear problematic. This history is, then, an integral component of the breast biography and a vital interpretive resource on which readers can draw: however, as we shall show, this history is, in some cases, problematic in that it introduces too much worldliness, so to speak, into what would otherwise be suspicious (pace Driessen, 1997) objects. In the first (prevalent) screening round, two views of each breast are taken and displayed on the viewer. In subsequent (incident) screening rounds a single view is taken, but the previous equivalent view is also loaded on the viewer so that the reader may judge how the features within the breast have changed over time. So a reading rarely involves the examination of a single pair of mammograms. Instead the reading of the principal view (obliques) is performed comparatively – either with an historical mammogram or with a view showing the breast imaged in a different dimension. Furthermore, single views are not read independently – comparisons are made between the left and the right breasts for asymmetric changes.8 Thus, features that appear equivocal in one view might, in the other view, appear to be more distinctly instances of a particular phenomenon. It is only in looking between the two views that a potentially problematic or equivocal object can be seen for what it is. One reader was heard to say that there was a ‘very large density in top of CC view, nothing equivalent in the oblique view and very spread out’: the case was not recalled because the mass had been accounted for through the use of both mammograms. Where the breast is too large to be imaged on one mammogram, front views are taken to capture the front part of the breast. The overlap can be used to correlate suspicious objects between mammograms – for example, if an object appearing on one view is not present on the other, it is possible to accountably dismiss it as, for example, an artefact of producing the mammogram. One reader noted that ‘front views are quite useful, because if [I am] suspicious of something on one view, then [I] can often dismiss it on the other’ (fieldnotes fr1-A). Thus diagnoses become possible through the provision of and their presentation in a contexture (Gurwitsch, 1964). As Lynch (1993: 127) points out, such elements emerge from the ‘mutually supportive details’ of such contextures. It is only in and through such mutually elaborative artful arrangements that the objects present in the breast become apparent for what they are (in that they are accounted for and they are reported upon, based on through observation). While agreeing with Lynch here, we want to continue to use 8 Since the research on which this chapter is based was undertaken all women screened now have a CC view taken on their first visit.

Cultures of Reading: Mammography


Figure 11.2 Manipulation of mammograms by readers Using hands to move a mammogram out of its carrier; using a magnifying glass to expand potentially suspicious objects; using an opaque mammogram to block off features of the mammogram; using fingers to measure potentially suspicious features on CC view on oblique. The images also show the arrangement of mammograms on the viewer. Photographs: M. Rouncefield

the notion of the documentary method as this appears to us to be a perspicuous way of explicating such phenomenon. Indeed, while the notion of the documentary method may be somewhat out of fashion in current ethnomethodology, we would argue that its explicative power continues to suggest its utility over potential alternates. Readers will use a repertoire of manipulations (see Figure 11.2 for examples) to make certain features ‘more visible’ (again there are important parallels here


Orders of Ordinary Action

with the work of Garfinkel, Lynch and Livingston, 1981). Where, for example, a cluster of calcifications are detected in viewing the mammogram, a reader may use a magnifying glass to assess the shape and texture as well as the arrangement of the calcifications since some arrangements of these phenomena are more suspicious and likely to occasion recall to the screening centre than others. In other cases, such as where the breast is dense, a mammogram may be removed and taken to a separate light box with a stronger source of illumination. Where a reader wishes to attend to a particular segment of the mammogram, an opaque mammogram will be used to blank off a part of it. In other cases where a suspicious feature is seen on one view, a reader may use their fingers or an object such as a pen to measure the distance of the object from some point such as the nipple on one mammogram in order to see if the object falls within the arc on the other view. This provides a means of assessing whether features are artefactual or actual. It is important to note that these techniques are not ad hoc workarounds but repertoires of manipulations that are an integral part of the embodied practice of realizing phenomena as what they accountably are.9 ‘Uncertains’ and Worldly Interpretations An important sub-set of cases is constituted by what we might call ‘uncertains’ – here the reader may see something about which he/she is not sure yet which is suspicious enough to occasion recall despite being an ill-defined object. In the case of ‘uncertains’ recall is achieved in and through the reader’s uncertainty as to what the object actually is. Readers make ‘worldly interpretations’ (Driessen, 1997) of the significance of the object is and whether one can assess it or not. In other words, a feature may be recallable through worldly interpretation as something which is notable enough to occasion recall.10 The parallels between the work of readers and that of the astronomers in Garfinkel et al.’s (1981) paper are striking. We can see that both the readers and the astronomers are engaged in situated practices that are recognizably and accountably constitutive of their disciplines and that their descriptions are predicated on these practices (in other words, that other readers and astronomers would find them acceptable and understandable in and as a part of the work of producing the phenomenon they 9 Of course, following Garfinkel, Lynch and Livingston (1981), the diagnoses have to furnish the irrelevancy of the embodied practices – in other words they have to be read as diagnoses per se as opposed to diagnoses achieved through the application of this or that manipulative repertoire. Insofar as this is the case, these manipulations might be regarded as seen but unnoticed technologies of conferring note to particular features of a mammogram once the features have been intersubjectively validated by second reading (features that might have remained ‘occult’ if these techniques had not been applied). 10 In some cases there are balances of probability – achieved through interrogation of the woman’s biography (if, for example, they are receiving HRT) and through the examination of previous mammograms for the object or some sign of its growth – these may be taken together as components of a breast biography that would or would not occasion recall despite the lack of a clearly recallable object on the mammogram. This is, of course, an integral component of our notion of worldly interpretations.

Cultures of Reading: Mammography


describe). It is, as Garfinkel et al. point out, the ‘intertwining of worldly objects and embodied practices’ (1981: 165) that produces the recognizable and accountable diagnoses and decisions of mammography. As with the ‘potter’s object’ the mammogram is a place to start and the procedures that surround it are means of realizing ‘an increasingly definite thing’ (idem). Like the potter’s object it is realized in and through a hands-on, worked-up character that makes it recognizably what it is. It is ‘real in and as of inquiry’s hands-on occasions’ (1981: 165) and becomes ‘available (to readers) in and as of their primordial embodied practices of findingand-exhibiting-again an IT’s adequately (in this case radiological) … details’ (idem). Further, as we shall see below, this increasingly definite thing has the feature of being seen by a second reader (for another first time) to be (observably, reportably, accountably) that which it is. In addition there are ‘geographies’ of suspicion in the breast (following Sacks (1972) we might equally say that reading occasions ecologies of suspicion). In other words, the positioning of an object in a particular area of the breast renders it more suspicious than if it had been elsewhere. Objects occurring, say, behind the nipple or behind the glandular disk are defined by readers as being suspicious by their very location. These areas are described as ‘danger areas’, ‘review areas’ or, in the case of the space behind the glandular disk, as the Milky Way (so-called because of its appearance) and readers orient to them in their examinations. Apropos of an object in the breast, one reader noted ‘so it’s an asymmetry in the review area therefore it’s a bit more sinister’. Areas where there is a high density of breast tissue are also suspicious since objects are difficult to see in ‘dense breasts’ (being masked by glandular tissue). Readers will, as we have noted, use magnifying glasses or higher-intensity light boxes in the search for objects within the mammogram. In other words, the response of the reader is commensurate with the presentation of the breast as a filmed object and the features within it. Of course not all suspicious objects occasion recall. As we have said, the notion of worldly interpretations is important here and, allied with the concept of breast biographies, mitigates against recalling all potentially suspicious objects. If a woman is on HRT, readers will not recall them if there are ‘patches’ on the mammograms. As one reader noted ‘if you know they are on HRT, for instance, you might accept patches in one (mammogram) where you wouldn’t accept in another’. There are also characteristics of the breast that are a component of these worldly interpretations: if a breast has a large number of lines within it this can produce false features11 such as composite shadows.12 In terms of suspicious objects we can say that only truly lucent breasts (that is, those in which objects can be viewed well) on mammograms that exhibit a high degree of professional competence are unequivocally benign and that the interpretation of other breast types invokes worldly interpretations of suspicious objects: Well, if it’s a completely lucent breast, and it’s been well positioned – a good technique – then you can be almost completely certain. It’s very difficult to say that anything is

11 For a first-class discussion of types of errors see Lynch (1985: Chapter 4). 12 In one case a reader was heard to remark rather ironically, ‘Ah yes, nice dense breasts’.


Orders of Ordinary Action completely normal, and you don’t know for instance if the lesion has been left off the mammograms. It’s really only in the completely lucent breasts you can be as confident as possible (Interview: transcript fr2-F).

Mammograms that have potentially confusing features or which are less than wellcrafted images of lucent breasts are subject to worldly interpretations and readers are forced, in the words of one, to ‘play a probability game’. The concept of ‘undressing’ a suspicious object is an important aspect of the practical constitution of professional vision in reading. Overlapping breast tissue can often mimic the appearance of stellate lesions or distortions in the mammogram. Readers do not necessarily take the ‘spiky’ appearance of an object as being indicative of a recallable object. Rather, they will ‘trace’ the lines through the object: if the lines are ‘drawn to’ the object as opposed to findable through it then this is an instance of a suspicious or malignant object. To get an idea of what we mean here, readers might think of the difference between roads on a map converging on a rotary and those converging but carrying on across a junction – the rotary is the suspicious object because lines radiate to and from there whereas the junction is a series of overlapping phenomena (that is, roads) and thus not suspicious. This ‘undressing’ or ‘picking the lesion to bits’ is a cognate operation to that described by Goodwin (1994) in terms of the relationships of figure–ground: the suspicious object is attended to and either confirmed or denied its suspicion through (1) the divorcing of the object from other components that apparently make it up, and (2) examining if the constituents are found continued relevantly elsewhere in the proximity. There are also occasions where a suspicious object is only seeable on a previous mammogram through recourse to the present mammogram. Readers speak of finding ‘hints’ of suspicious objects on previous mammograms through looking at present ones. As Goodwin (1994) points out, this figure–ground relationship is not something that happens only in the head; it is made accountable in and through natural language13 and in embodied practices such as tracing the lines of suspicious objects with pens and fingers or looking at objects in previous mammograms in the light of current mammograms in a documentary method manner. We should also point out that readers were aware of the potential to both talk themselves in and out of things through ‘undressing’ these suspicious features. Thus there is a reflective dimension in the realization of accounts, which is something that we would like to explore in future work. Suffice to say here that, contra cognitivist notions of the process going on in the head, there is a professional ideology that enjoins readers to reflect on, and to make accountable, their seeing and the objects it produces. Reflecting on Noticing Another aspect of reading is a reader’s perception of their own level of skill. Several readers expressed opinions about aspects of their own general performance 13 For example, one reader notes of a suspicious object ‘bit denser there – all lines seem to go through it rather than be drawn towards it’ (in this case the reader decided that the feature was benign).

Cultures of Reading: Mammography


abilities. Readers are often seen by colleagues and often regard themselves as having skills or deficiencies in noticing particular types of object within mammograms. For example, one reader stated that she was not so good at detecting distortions, suggesting ‘either good at them or not good at them’ (field notes fr2-A). Another suggested that her particular skill lay in detecting ‘patchy asymmetry, distortion, microcalcs’ (field notes fr1-D). A reader from screening centre A suggested that he has ‘a stricter criterion for suspicion’ than his colleagues (field notes fr4-R), whereas a reader from centre B suggested that she tends ‘to bring more cases back than [my] more experienced colleagues’ (field notes fr2-B). In some cases we observed readers being characterized as someone who is good at spotting certain types of ill-defined lesions (a ‘blobby person’) as opposed to colleagues who were capable of spotting other types of objects. This was not a rebuke but a call to ensure that the reader looked beyond their particular competencies in realizing an account of a particular mammogram. Thus it appears that readers have beliefs that they are able to articulate about the level of their skill and abilities relative to those of their colleagues. Readers may also use this knowledge to adapt their reading strategy: ‘My approach tends to be to look [positively?] for things that I know I’m not so good at ... there are certain things that you do have to prompt yourself to look at, one of them being the danger areas’ (Interview: transcript f2-F). This is a similar approach to that taken in other circumstances where there are recognized difficulties; for example, identifying distortion of architecture in dense breasts or, as the reader above suggests, detecting lesions in the ‘danger areas’. One way readers might monitor their day-to-day performance is to compare their recall rate with the frequency at which they might expect to recall.14 Suspicious cases should present randomly, but readers suggest that they are aware that variations in distribution of recalls over short time periods can potentially influence their decision-making: Paranoia can set in if have a large number of mammograms that have passed as normal – might think ‘what have I missed?’ (Comment made while reading: field notes fr1-B). If you get to the end of a session, the end of a pile of reporting and you haven’t recalled anything, then you think ‘this is ..., maybe I’ve missed something’ then in the next bunch you find that you will recall every other one. So it averages out (Interview: transcript fr2-F).

Again we can see that there are ‘worldly interpretations’ of viewing strengths and weaknesses as well as of recall rates. Just because the readers have had a run of cases where they have not found any suspicious objects does not mean that they should change their reading strategies. Thus it is possible to say that they have a worldly interpretation of the ‘typical’ distribution of recalled and non-recalled cases and that a predominance of the latter is not in itself a ‘trouble’. ‘Being noticeable and being deviant seem intimately related’ (Sacks, 1972: 280). 14 Readers might expect to recall 6 per cent on prevalent screen rounds, falling to 3 per cent on incident screening rounds, and 0.6 per cent of all cases seen will have a suspicious object that turns out to be a cancer.


Orders of Ordinary Action

We want to sum up this section of the chapter with some comments about noticing and to draw some parallels between the work of readers and Sacks’ classic account of police assessment of moral character. We do not wish to suggest that the police and readers act in exactly the same way, but that the parallels between their practices are instructive. The development of professional vision as a worldly, lived practice, in both cases, provides for a non-cognitivist explication of terms such as ‘my eye has been drawn to this’ and so on. These terms become available to us as practical, embodied achievements as opposed to some occult event residing in the mind. It is common to hear a reader say that their eye has been drawn to a certain feature within a mammogram. It is interesting to us (as observers not trained in radiology) that this process appears almost effortless. In the majority of cases the readers note that their eye has been drawn to an object that appears to us to be yet another spot on a mammogram. Without resorting to a mystifying (in Garfinkel and Rose’s (1973) sense) cognitivist theory we would argue that in a manner cognate with Sacks’ explication of the ways in which police officers spot criminals attempting to ‘pass’ as normal persons, readers are able to spot abnormalities within the mammograms they read. The concept of territories of normal appearances is important here in that, as Sacks rightly notes, apprentice police officers have to become skilled at detecting deviants attempting to pass as law-abiding citizens, and it would seem to us that readers have to have cognate skills in being able to detect (and to subject to worldly interpretations) candidate suspicious features in mammograms. As we have shown, like the police officer, what is normal (or, more precisely, what will not occasion recall) is contingent on its presence in a particular site and time (hence our stress on breast biographies). In other words, just as there are people who do not belong in a certain part of the city, there are objects that do not belong in a ‘normal’ (that is, non-recallable) mammogram. Annotations in and as Professional Vision Here we want to attend to the ways in which first readers (where double reading is practised) make available candidate diagnoses of benign and potentially recallable objects through the annotation of pro forma schematics of the breast. Figure 11.3 gives an example of a screening reporting form with the radiographer’s annotations. The markings on the schematic at bottom left are annotated as moles. The other comments read: ‘Cyst – R Breast 20 yrs ago. HRT for 2 yrs. Pain L breast and arm – GP says muscular)’. The first reader has marked on the schematic at centre right ‘New’ and the second reader has added BT [breast tissue] I think HRT [hormone replacement therapy] related)’. The first reader has recommended recall for assessment. The second reader recommends return to routine recall. Having given some idea of the manner in which the mammograms are read we want to move on to examine the embodiment of this lived work, and to discuss the ways in which annotations are used. ‘It’s good to know that someone else is seeing the same thing ... for example that something hasn’t changed ... the second reader gets confirmation that they are thinking along the same lines’ (Fieldnotes: no

Figure 11.3 Screening reporting form, showing annotations by the radiographer


Orders of Ordinary Action

date). While it would seem that the work of reading is undertaken in and through the application of individual skill (a sense that perhaps makes it amenable to mentalistic descriptions and their inherent ‘downward conversion’ (Burke, 1935), this chapter shows how reading is collaborative work and how diagnoses or versions of what a particular mammogram contains are intersubjectively constituted achievements. Readers’ activities of annotation and talk around artefacts constitute the mammograms they view as a part of their craft. Their use of mammograms differs from, say, a radiographer or a nurse or a general practitioner in that the readers bring to bear radiological experience and expertise to mammograms. In other words, a mammogram becomes what we would call a reader’s object in and through the discursive practices that take place, among others, in reading, annotating and discussing – this is where professional vision lies, as Goodwin notes they are and make visible ‘categories of relevance to the profession’ (1994: 610). Annotation may be thought of as interaction through artefacts. Because of the practical exigencies of blind independent reading, annotations on screening reporting forms have to be used to convey messages about the mammograms, and about candidate diagnoses of potentially accountable features on mammograms. Annotations, in other words, constitute a primary means of communication and concomitantly a means of arriving at an intersubjectively achieved version of accountable features. Annotations are a means of conferring note on a feature of a mammogram and for making accountable, that which is seen as noteworthy, within the professional vision of the readers. As well as making the features of the mammogram seeable (so to speak) they make the features seeable-as-that-is – marks on the reporting form offer a candidate version of the features to which they orient. Features, once seen and annotated, become accountable as candidate instances of phenomena. A circle on a reporting form with the annotation ‘calc?’ not only renders that feature accountably seen by the reader, but also seen as an instance of a particular phenomenon. Again the notion of figure–ground as an embodied achievement is important here. As Goodwin notes ‘features can be difficult to see’ and annotation ‘establishes a figure in what is quite literally a very amorphous ground’ (1994: 610), drawing attention to the accountable features of the mammogram through practical embodied action.15 Further, because readers work by and large alone, annotations form ways of achieving intersubjective diagnoses at a distance (both spatially and temporally) while still displaying the maxims of professional correctness required by a single reading. We might say that annotations stand proxy for conversations, that they are conversations in and through artefacts, proposing candidate answers that have yet to be intersubjectively validated. It is in such interactions that a mammogram emerges as a reader’s object. That is, only by regarding the mammogram as being embedded in the work-site setting and the interactions that occur within it can we see how it comes to have meaning 15 Interestingly, our studies of computer-based detection aids prompts reveal the computer annotation to be problematic in that its ‘docile’ (cf. Lynch’s (1985: 43) comments) prompts draw attention to figures that have already been seen by readers, who are then forced to account for the prompts before moving on. The machine can be seen to lack the requisite professional vision to become a ratified seer of suspicious objects as it raises too many banal figures (Hartswood and Procter, 2000).

Cultures of Reading: Mammography


within the readers’ professional vision. As Goodwin points out, ‘a relevant object of knowledge ... emerges through the interplay between a domain of scrutiny ... and a set of discursive practices … being deployed within a specific activity’ (1994: 606) (italics in original). The reader’s object emerges from readings made by the reader within the reading room in and as a part of the practical, observable-reportable, accountable practice of carrying out the reading of a mammogram. In short, they make work visible (Suchman, 1995). We can see that the annotation of the feature as a representative of a phenomenon … becomes a public event that can guide the perception of others while further reifying the object that the archaeologist proposes to be visible in the colour patterning of the dirt. The perceptual field provided by the dirt is enhanced in a work-relevant way by the human action on it (Goodwin, 1994: 610–611).

It may be that the first reader has annotated a feature with the comment ‘calc’; this is of the character of a candidate account for an object on the mammogram and it is only when the second reader has made his/her reading that this candidate status and its concomitant suspension ends and the annotation (and that to which it refers) becomes seeable as something.16 This is not to suggest that there is something deeply mysterious about the process or to succumb to residual mentalism by suggesting that the reader ‘cognises’ in this or that manner (whatever this term could be held to mean); rather it is to point to the thoroughgoing sociability of annotations and the work practices that surround them. This leads us into a discussion of the sociability of annotations in that annotations, once seen by both readers, constitute a phenomenon similar in character to that referred to by Sacks (1992) as ‘my mind is with you’ – of course this is not to suggest any mentalistic formulation; rather it is to show how annotations share ‘tellable’ features of the mammogram between readers whose job it is to look past these ‘tellables’ to the phenomenon ‘below’. Annotations on a screening form stand proxy for interactions and employ the selfsame documentary method of interpretation, but do so for readers who, because of the ‘rules’ of reading, do not have contact with each other around these elaborated and elaborating objects in the manner that Goodwin’s archaeologists do. This raises an important point with regard to the nature of reading: with an annotated mammogram, can one read in a manner that the concept of ‘blind reading’ supposes? 16 Apropos of mathematical proofs, Livingston remarks ‘as a prover “writes” a partial or “fragmented” description of a projected proof, the prover simultaneously configures a course of reasoning. The reasoning so embedded in the partial material argument is reasoning that, for provers, is reasoning any competent prover could see in the partial argument. ... The activity of proving builds within itself and maintains its own analysability as mathematical argumentation.’ While we might have misgivings about the notion of configuration (in part because of its place in another analytic vocabulary) there are instructive parallels with the work of provers and readers. The diagnosis of suspicious objects is achieved through procedures which, while not as visible as those of a mathematical proof, are nevertheless followable in the manner described by Livingston; it is this followable nature that in part occasions the defeasibility here and makes the candidate diagnosis of the character – ‘I have seen the object to which you refer but have you seen this information and the implications it might have?’


Orders of Ordinary Action

In other words, what is the status of the annotated mammogram vis-à-vis the unannotated mammogram? The problem turns on the ‘phenomenological intactness’ (Schwartz 1976) of the un-annotated mammogram. This is, however, a chimerical problem because of the work practices of the readers. The professional vision of readers constitutes annotated mammograms as read for another first time. That is to say, rather than supposing that readers do not see the marks made on the reporting form, these marks form part of a candidate vision. That is, the second reader sees a mark on a reporting form, but what it is seeable as is held off until the second reader has made her reading; thus it assumes the character of a ‘candidate version’. As one reader put it: ‘… when someone shows sets of mammograms and they’ll say, [have a look at this], it’s always nice for someone … [not to] point out what they are worried about, because if [they] do, then immediately you [have a] heightened suspicion [it is] because somebody else is suspicious about it’ (Interview transcript). Thus the mark on the form has a moral force because it is an object made observably reportable by someone who is presumed to be a ratified ‘seer’ (that is, someone who shares their professional vision).17 Thus the mark on the form makes something (morally) ‘seeable as’ that thing by the reader’s own account above. Furthermore, if the reader then finds that it is not seeable as that thing for them, this occasions a ‘problem of versions’ (Cuff, 1994). This places readers in a situation where they have to provide an account for the difference and interactionally manage the accounting in a way that does not constitute the first reader as incompetent (for example, he/she may suggest that it is a marginal case, that the task is a difficult one, or that they disagree rather than that the first reader is wrong). For instance, one reader might annotate a feature on a mammogram by circling it and saying ‘new’; the second reader provides an alternative account of the phenomenon saying ‘BT (breast tissue) I think. HRT related’. The important point here is not that the accounts are incommensurable (they do not, for example, formulate something seen by one reader as being nonexistent), rather they are candidate formulations of the phenomenon in question that can be questioned in and as a part of the processes of reading for another first time and thereby assembling another candidate version of what a phenomenon actually is in and through annotations. We should not, however, suppose that there is an infinite number of candidate formulations; rather, the object is defeasible for all practical purposes, whether this involves a third reading or a needle biopsy, it is still defeasible up to the point at which an account-for-all-practical-purposes is accomplished. One of the paradoxes of reading is that readers want to avoid ‘being biased’ – as is suggested by the reader above – in order that they can report the phenomenon as it appears to them, but by the same token it is only through practices that in principle constitute bias that an intersubjectively agreed version of how the phenomenon 17 Presumably it would be different if a layperson had pointed to something on the mammogram. It is this ‘power to speak as a professional’ (Goodwin, 1994: 624) that is a central component of professional vision. It is also professional vision that occasions worldly interpretations – as a matter of logical grammar how could one have a worldly interpretation that was not predicated on professional vision? To have this would surely be something else – perhaps a lay account of a phenomenon. This would be of a different order and perhaps more overtly defeasible by professional vision.

Cultures of Reading: Mammography


appears to them can actually be produced. Through their practices readers appear to subject the notion of ‘bias’ to ‘worldly interpretations’ and thereby to ‘tame’ its contingencies and make it productive. We might thus point to the notion of ‘uniquely adequate’ information; that is, the selfsame information that can be regarded as a ‘trouble’ may well be the information that is required to achieve a version of what a particular phenomenon is (or is not). In other words, anxieties about influence are somewhat chimerical and to be managed in any case in and as a part of the practical action of becoming a competent reader. In some cases this is undertaken through the temporal ordering of access to information; for example, reading of a mammogram before the reading of annotations. In some cases readers orient to the ‘deviance’ inherent in cooperatively produced first readings, noting: I do an interesting thing that probably I should not do; if I’m really dithering about one and I’m not sure whether to bring it back or not, independently of whether I’m first or second reader, I do sometimes take it to [colleague] and see if [they] see it. You know, because we tend to see the same things. I know I shouldn’t do that [laughs]. For example, if I’m reading the mammogram as a second reader and I read the mammogram before I read what say [colleague] has put, and I look down and see that she’s thinking exactly the same thing “that’s benign” and if [they have] marked it in the same area that’s very reassuring, it’s good that is. I like that, we’ve both thought of the same thing. We’ve both seen it but thought it was ok; it’s good, it’s confidence.

This, we think, shows both sides of the paradox. It would appear from our discussions with readers that the paradoxes of bias are highly situated matters. There is a tension between the whole idea of being biased and the need for a second opinion and the imperative to get the job done for all practical purposes.18 Conclusions In the above we have attempted to explicate the lived work of doing reading and to relate this to the notion of professional vision as developed by Goodwin (1994, 1995, 1997). We have also attempted to show the embodied nature of reading; that is, the use of manipulations and measurements to locate a feature. We have also discussed the manner in which breast biographies and worldly interpretations impact on the interpretation of objects and suspicions within a mammogram. We have discussed the ‘sociability’ of reading and its relationship to annotations as a means of accountably conferring note to certain features within a mammogram. Finally, we have attempted to make some comments about the paradoxes of bias in and as a part of the practical work of doing reading. In all this we have, it is hoped, demonstrated the need to look beyond cognitive science’s impoverished accounts towards, as Coulter (1991) 18 Apropos of the notion of bias, it is interesting that viewing a set of mammograms is not in itself generative of bias, but reading a comment on, say, the woman’s being on HRT might be so regarded. What is ‘proper’ ‘professional’ conduct is a situated achievement and bias is managed in and as a part of this: what will be regarded as ‘bias’ is variable and is discursively realized within the practices of a particular clinic or setting.


Orders of Ordinary Action

points out, the ‘praxiological’. To be sure, members speak of seeing, noticing and other topics that are grist to the mentalists’ mill, but we have shown that they do so not in an isolated context (neither behind the skull nor as atomic ‘cognisers’) but in a manner where terms such as ‘seeing’, ‘noticing’ and so on are practical members’ achievements, achieved in and through natural language and embodied conduct. It is for us unfortunate that so many studies of reading and similar tasks have sought the ‘downward conversion’19 of cognitive science. One might regard the manner in which cognitivists seek to attenuate the social dimensions of reading (often couched in terms of eliminating ‘bias’20) as problematic in extremis in that it not only produces an impoverished account of the phenomenon but also leaves a series of ‘missing what’s’ which we would argue are profoundly social in character. The measurement of eye movement, for example, seems to be of great interest to cognitive science in adducing reader performance, yet this produces findings that have a rather familiar character.21 The statement by Kundell, Nodine and Toto (1991) (to the effect that experienced readers see phenomena directly while inexperienced readers have to spend time looking) seems to us to be impoverished (should we just look at eye movements – recall our comments on slicing above – as an index of what?). Also, by the very nature of the experimental conditions under which the experiment is performed, one finds that the attenuated perspective adopted therein actually loses the phenomenon in that the experimental focus slices off fundamental layers of artful practices in which readers are engaged. Cognitivist treatment of the single reader as an ‘atomic object’ also ignores the crucial social dimensions of reading and the interactional constitution of diagnoses inherent in the term ‘professional vision’. While space does not permit here, the interactional bases of diagnoses are important and we have only touched on the surface through our discussion of annotations. It is important to consider more fully the role ‘shop talk’ has in the realization of diagnoses and the seeing-for-what-they-accountably-are of suspicious phenomena in mammograms. Our current research does not fully cover the manner in which such diagnoses are achieved and it is hoped that such data will be available later within the research. However, we hope to have given some preliminary idea of its character as made available through the readers’ annotations. The concept of professional vision and the defeasibility of readings seems to us to capture the lived work of reading as practical embodied action in a descriptively adequate manner, yet there is a sense in which the readers’ worldly interpretations of suspicious objects elude us at present and that is in the interactions that go on around suspicious objects 19 The term comes from the work of Kenneth Burke (1935). In using the term we follow Rod Watson’s illuminating paper in Watson and Seiler (1992). 20 This bias is of a different order from that which readers seek to attenuate through blind single reading, and as such this might be seen to show how such concepts are lived and embodied as opposed to stipulated by experimenters. In a sense, then, we might think of members’ bias as ‘essential’ in Garfinkel’s (1967a) terms. 21 While readers might formulate suspicious features in terms of their eye being drawn to this or that, we would argue that this eye is thoroughgoingly social in character and formulates the feature as suspicious as opposed to describing the behaviour in which the reader is engaged.

Cultures of Reading: Mammography


where there is a difference of opinion. We have pointed to the role of annotations here but consideration of the talk-in-interaction that achieves defeased readings (especially where there is a dispute) remains somewhat opaque at this time. It would also be of interest to consider the notion of apprenticeships with regard to the development of professional vision as Goodwin (1997) has done – how does one come to see objects for what they accountably and reportably are?

This page intentionally left blank

Chapter 12

The ‘Problem of Dust’: Forensic Investigation as Practical Action Robin Williams

It is a recognized fact that in many cases no witnesses are available and all the evidence is circumstantial. In such cases science may help in a way which is beyond the range of any possible evidence depending upon the personal observation of the police or of other witnesses. In some cases it may supply vital evidence against the guilty party, or in others definite proof of innocence where guilt was suspected. It may give certainty amidst apparently conflicting stories from witnesses, and at the very least it may provide a means of reconstructing the crime and thus saving an immense amount of unnecessary work for the investigating officer, by eliminating lines of enquiry which would otherwise require attention and pointing to those which are most likely to be fruitful (Home Office, 1949: 3).

Introduction All police forces in the UK and most police forces in the US increasingly deploy specially trained staff to attend and examine scenes of crimes as one of a series of distinct but linked courses of action that together comprise criminal investigations. The variety of terms used for these staff (‘crime scene examiners’, ‘crime scene investigators’, ‘evidence technicians’, ‘forensic investigators’, ‘scientific support officers’ and ‘scenes of crime officers’) reflects the heterogeneity of the work that they do in different police organizations. However, the common theme used to envelop this heterogeneity and distinguish it from the work of other kinds of criminal investigators is the ‘scientific’ character of their working methods. This chapter seeks to explore the lineaments of this theme. It suggests that more can be learned about its practical significance and organizational consequences by giving close attention to the routine accountable conduct of crime scene examiners than to stipulations of the generic character of ‘scientific reason’ as instantiated in routine and exceptional scene examination forensic practice. It is the argument of the chapter that competent crime scene examination is achieved by its attentiveness to two chronically vague propositions whose epistemic status is uncertain but whose organizational uses are powerful and consequential. It is further argued that reiteration of the propositions provides both for the accountable character of the forensic work of crime scene examiners and the grounds for their exercise of authority over the warrantable designation of persons and places as ‘crime scenes’. Such a designation affords restricted access rights to such scenes and gives crime scene examiners


Orders of Ordinary Action

primary control of movement in and actions at such scenes, including control over the course and methods used to search and examine places and persons, as well as control over the seizure of material objects and property at such scenes and the trajectory of subsequent forensic analysis. The chapter also suggests that competent crime scene examination is largely accomplished in and through routine work using ordinary bodily – especially visual – capacities along with technologies that enhance these capacities in order to make possible the construction of a limited repertoire of ‘forensic artefacts’. Such artefacts may become the subject of further specialist analysis (for example, to match physical traces found at the scene with entries on previous databases, or to infer a course of criminal action) or be used to facilitate the envisioning of events and their traces at a scene of a crime by individuals who were not present at the time of its original and only examination (such as other investigators, prosecutors, advocates, judges and juries). It is the recorded performance and standardized products of these various investigative and prosecutorial actions that provide (or fail to provide) the assurance to participants and observers that accountably rational forensic science has been realized in each instance of its supposed deployment. The chapter begins with a brief description of a recent study of the work of this specialized occupational group before going on to outline the kinds of actions that comprise competent crime scene examination for members of this group and for other investigators with whom they work. A second section contrasts textbook efforts to stipulate idealized ratiocinative standards for the cognitive orientation of forensic investigators with what detailed empirical study reveals as two working principles accountably oriented to by investigators in the course of their construction of, and work within, crime scenes. A third section describes how these same principles not only direction the production of a range of ‘forensic artefacts’, but are also embedded in the methods of their production. A final section describes a recently reported dispute over the forensic examination of a particular domestic burglary to show how skilled practitioners used detailed knowledge of how such artefacts are made to raise doubts about the credibility of one particular example which was central to the prosecution case. The Study The fieldwork on which this chapter is based took place in one particular operational division of a UK police force between October 1999 and May 2000. Throughout these months I spent two full days each week observing the work of individuals who together formed a group of six crime scene examiners on duty for various times during each day. During this time I visited a large number of scenes of crime with the examiners, the majority of these crimes having been reported to the police as domestic burglaries and as thefts of and from motor vehicles. I also visited some scenes involving more serious offences against the person. My participation in these visits was limited to fetching and carrying relevant equipment from the investigators’ vehicles to the scenes attended, laying down stepping plates where it was necessary to protect surfaces from our own and others’ feet, holding light sources to illuminate a

Forensic Investigation as Practical Action


search and talking to victims while the scene examiner was busy with some aspect of their work. In addition I read all the crime scene examination reports written by staff on all burglary scenes during a one-year period, internal force memos and protocols concerned with crime scene examination and the submission of forensic materials for analysis, as well as other relevant handbooks and guides to good practice issued by external authorities and used by the force in question. I also attended routine weekly crime briefing meetings which involved crime scene examiners, along with a small number of exceptional meetings held in relation to the investigation of particular – usually serious – crimes. The examiners that I studied routinely constructed a series of ‘forensic artefacts’ in the course of their crime scene examinations These included scene photographs, footwear marks, casts of footwear impressions, glass samples, casts of toolmarks, and fibre collections. All of these were used in particular investigative contexts to infer the categorical or even unique identity of a criminal suspect as well as to support surmises about the courses of criminal actions that contributed to the visual appearance of the crime scene. However, two additional types of such evidence often provided the main focus of the work of scene examiners, especially in the property and car crime scenes that comprised the majority of the investigations undertaken during my period of fieldwork with them. These were ‘fingermarks’ and ‘crime scene stains’, where the latter (for example, blood, semen, sweat, saliva, skin cells and hair) were expected to yield DNA profiles following laboratory analysis. In fact fingermarks and DNA profiles are the only evidence types which examiners describe as ‘capable of giving a name’ to a potential suspect through the speculative searching of recovered marks and analysed profiles against national databases of fingerprints and DNA profiles previously taken from individuals charged with recordable crimes. The sensitive character of the settings did not allow the audio or videotaping of the conversations or events in which I participated or which I witnessed, although I was able both to pester examiners with questions (such as: Why this now? How does this piece of equipment work? Why are you using this particular powder instead of some other for the development of this particular fingermark or shoemark? Why collect this sample and not another for possible DNA analysis? Why this photograph at this particular distance? etc., etc.) and to make contemporaneous notes of their answers to these questions. However, the lack of technically facilitated and uninterpreted recordings of these events necessarily limits this chapter to the provision of more allusive descriptions than I would prefer. Stipulating Cognitive Authority The warmth of the police force’s embrace of forensic science reflects their confidence that a particular body of scientific knowledge and repertoire of technological devices can be used to assemble a collection of forensic artefacts and their textual correlates capable of supporting unequivocal and authoritative accounts of criminal actions and criminal identities. Such accounts in some cases will be sufficiently strong to withstand the kind of interrogations that exemplify the ‘institutionalized suspicion’


Orders of Ordinary Action

that is a fundamental feature of judicial proceedings. However, not all instances of the construction of these artefacts and accounts will be subjected to these stringent conditions. Many more are used as ‘forensic intelligence’ – as information used further to direct criminal investigations and disruptions rather than as props to support the dramaturgy of a criminal trial. Much of the work of the crime scene examiners that I observed fell into this latter category of ‘intelligence’ so that although the course of their examinations were always oriented to the idealization of ‘working to evidential standards’, only a very small part of their work was used to support criminal prosecutions and therefore became subject to judicial scrutiny. It is perhaps unsurprising that the increasing interest in both prosecutorial and investigative uses of forensic sciences and technologies has been accompanied by attempts to produce generic theoretical accounts of exactly how the authority of such artefacts and accounts is – or may be – assured in and through the course of their construction and deployment. These accounts have been ratiocinative – and stipulative – in character, although inevitably they differ significantly in the cognitive principles that they assert as central to adequate practice. A few instances of such accounts will be sufficient to illustrate their overall approach. For example, Jamieson (2004) begins his advocacy of hypothetico-deductive theorizing as a cognitive model for the process of crime scene investigation by arguing that the searching of crime scenes and the collection of physical trace material is always and necessarily preceded by the existence of some explicit or implicit hypothesis which will explain the events observed in the developing course of the examination. He then goes on to argue for the necessity of assessing the adequacy of a single or a series of such hypotheses by comparing their predictions of what should be observable if they are correct (or at least which may be discarded by the discovery that predicted observations were not forthcoming): ‘Ideally the tests should be devised to differentiate acceptability among the competing hypotheses in the smallest number of tests. These tests include searching for confirmatory or exculpatory evidence at the scene’ (Jamieson, 2004: 4). For others, however, there is little to be learned about adequate crime scene examination practice by the application of a formal model whose competent use rests on the ability of its users to identify relevant physical laws, accurately describe a set of relevant material conditions, formulate logical connections between these two, and establish ways of definitively testing and retesting the adequacy of all of these matters in the relatively uncontrolled and heavily contaminated environments of crime scenes. Partly for this reason, others (such as Nordby, 2000; Eco, 1983; Truzzi, 1983) have argued that the reasoning processes of forensic investigation are best understood as instantiations of Peircian ‘abductive’ inferences previously associated especially with the more ‘conjectural procedures of physicians and historians’ (Eco, 1983: 205). For Peirce, the existence of particular facts can never successfully be explained inductively by inferring a rule for their formation from a collection of individual instances of their occurrence. However, they can successfully be explained ‘abductively’, that is ‘by the provisional entertainment of an explanatory inference, for the sake of further testing’ (Eco, 1983: 206). In Peirce’s own formulation: ‘The surprising fact, C is observed; but if A were true, C would be a matter of course. Hence there is reason to suspect that A is true.’ For Nordby, such ratiocinations

Forensic Investigation as Practical Action


lessen doubt but on their own cannot remove doubt. The codified rule that comprises the hypothesis must still be subjected to further test, although the action of testing encourages further exploration of the details of the case under examination and those that seem similar to it. A whole series of such modified sequences may be necessary before investigators are seemingly able to establish a credible account of the underlying orderliness from which the individual observations in question can be shown to be derived. Achieving Investigative Authority However, whatever the apparent significance of the various ‘principles of forensic science reasoning’ outlined in the previous section, it is important to note that reference to these kinds of ratiocinations did not occur in the shop talk of the scene examiners as they went about their daily work of attending and examining crime scenes during the fieldwork period. Nor were they asserted or contested in the course of formal briefings or informal meetings held between scene examiners and other police investigators. Scene examiners never referred to them or their alternates to mark the distinctiveness of their own approach to crime investigation or to argue for their interpretation of any individual event whose course and meaning was subject to dispute. This is not to say that scene examiners failed to distinguish the particularity of their work from the work of other investigators. In fact they constantly and chronically proposed such a distinction, but they achieved this in verbal and bodily displays in and through which assertions, hints, suggestions, puzzles, complaints, questions, claims, reminders and other actions made use of two basic propositions that shaped both the overall trajectory and the fine detail of their essentially and unavoidably practical work. These propositions can be summarized as: ‘exchange always occurs’ and ‘individuation is always possible’. The recitation of these propositions in a variety of local interactional contexts was the central resource used by examiners not only to direct and account their own ongoing conduct, but also for the control of the conduct of others – witnesses, victims, suspects and other investigators – who were not recognized as appropriate persons to directly participate in or observe crime scene examinations. Finally, there were direct and indirect references to the substance and significance of these propositions throughout the local instructional manuals and in quality assurance guides used by the police force studied. In sum, attentiveness to them provided for the methodical and accountable character of what competent scene examiners did and what they did not do at crime scenes. It will be useful to examine each of these propositions in a little more detail to understand the kind of resource that each provided for the work of this group of crime scene examiners. We can begin to explore the nature of both propositions by means of the simple historical example that is almost always used to introduce them to crime scene examiners – usually in the first few hours of their initial training. The example was originally provided by Edmund Locard, the French forensic investigator whose most important work was done in the early part of the 20th century, although still active well into the 1950s. In a case that he investigated in Lyons during 1912, a bank clerk


Orders of Ordinary Action

had been accused of the murder of his lover and had presented seemingly strong alibi evidence. Having examined the body of the woman who was murdered by manual strangulation, Locard took scrapings from underneath the fingernails of the suspect and claimed to have found not only skin scrapings but also cosmetic powder which Locard further asserted could be matched to the powder used by the victim. According to Locard, when the evidence was put to the suspect a confession was secured. The simple story is regularly used to illustrate the content and implications of the two propositions: that ‘exchange always occurs’ (in this case an transfer of material from the victim to the suspect); and perhaps less obviously that ‘individuation is always possible’ (in this case that the cosmetic powder scraped from the nail of the suspect was identical to the powder known to have been used by the victim). My interest in these propositions is not an assessment of their truth value but rather a consideration of the heterogeneous ways in which they came to be so regularly expressed and discussed in and through the course of the routine and exceptional crime scene examinations that I witnessed and was told and read about in the course of my fieldwork. The first of these two propositions, ‘exchange always happens’, is a compressed expression of the claim that whenever individuals make physical contact with other individuals, or enter and leave defined physical spaces, a (usually unintentional) transfer of varying amounts of material takes place between the parties involved. Some refer to the proposition as ‘Locard’s Law’, ‘Locard’s Truth’ or ‘Locard’s Doctrine’, while others prefer to call it ‘Locard’s Theory’ or even ‘Locard’s Principle’. In fact Locard himself did not originate a proposition with any of the exactitude implied by these designations. The closest he seems to have come to any such formalization occurred in a passage in his L’Enquête criminelle et les methods scientifique (1904) in which he asserted that ‘it is impossible for a criminal to act, especially considering the intensity of a crime, without leaving traces of his presence’. It is noticeable that this version only describes traces left by the individual at a scene and does not refer to traces of the scene taken away by the individual on departure. Many of the other versions of the principle also (but incorrectly) attributed to Locard himself do manage to suggest the complementarity that is an important feature of current usage. Examples include: ‘something is taken, something is left’; ‘with contact between two items, there will be an exchange’; ‘when a crime takes place there is an exchange of “physical information” between the criminal and the crime scene’; and ‘when someone commits a crime they always leave something at the scene which was not there before and they take away with them something which was not on them when they arrived’. However, these literary and historical variations in the expression of the principle are less significant than the investigative authority that its reiteration affords crime scene examiners. Versions of it have been so embedded in routine forensic practice that the terms ‘contact trace material’ or ‘physical trace evidence’ now comprise part of the taken-for-granted background of criminal detection and regularly occur as headings in crime report forms, as rows or columns in police intelligence spreadsheets, and as evidential categories for use in the construction of prosecution cases against individuals. But more importantly than these passive instances of its appearance in

Forensic Investigation as Practical Action


paper and virtual documents is the fact that it is actively voiced by scene examiners as they organize the technical trajectory of their work at crime scenes and as they deal with the social and organizational contingencies of crime scene examination. Despite what I have described as the emphatic character of the proposition ‘exchange always happens’, attendance at a large number of crime scenes showed that many scene examinations were conducted in which no contact trace material was identified or recovered by scene examiners. Interestingly, such instances never occasioned any disputes over the adequacy or accuracy of the principle amongst that staff group that I studied. The seeming incorrigibility of the principle was protected through a number of secondary elaborations in which such failures were seen to result from their own or others’ inability to discover what must have once been there at the crime scene or on the body, clothing or equipment of the suspect. Each such failure then seemed only to provide the opportunity to sustain the incorrigibility and vigour of the principle by detailing what local features of any particular crime scene (or the course of its examination) occasioned this noticeable absence of evidence. Such local features included, for example: the time that had passed since the offence occurred; the weather conditions when the scene was an outdoor one; the obscuring effects of the actions of previous visitors to the scene; the failure to apply appropriate examination techniques; the unavailability of technical resources which would have permitted relevant discoveries; the haste of the examination occasioned by the pressure of work; and in one instance, the assumed forensic awareness of the criminal who had erased what would otherwise have constituted relevant contact trace material. A later section will show the way in which the incorrigibility of the principle can be used to question the adequacy of particular instances of the work of crime scene examiners, but before that it is necessary to consider the ways in which the second principle of crime scene examination gives shape and significance to the detailed focus of attention that characterizes their work at crime scenes. The principle that ‘individuation’ – or the unique identifiability – of any object is always possible is as incorrigible and emphatic as that already described above as ‘exchange’. In this case the principle is protected from falsification by faith in the development of knowledge: despite appearances to the contrary, whenever two objects seem incapable of being differentiated here and now, application of the principle allows the assertion that they are capable of being differentiated or individualized whenever the means become available to do so. It may be necessary to ‘look more carefully’, ‘test more sensitively’ or ‘measure more accurately’. It is tempting to characterize the principle of unique identifiability as ‘protoscientific’, if this term can be used without irony. Certainly it does not seem to be derived from any particular set of disciplinary propositions but instead from an earlier and more widespread assertion that ‘no two things in nature are exactly the same’. But whatever its origins, the assertion is so central to forensic practice that for some authorities, it constitutes its very particularity: for example, ‘criminalistics is the science of individualization’ (Nickell and Fischer, 1999: 3). The favoured illustration remains that of the snowflake, where it is always asserted that two exact copies have never been observed, but the same claim of universal uniqueness is extended from ‘nature’ both to all instances of humanly constructed objects, and also


Orders of Ordinary Action

to the effects on both naturally occurring and humanly constructed objects of their uses and interactions with other objects over time. When examining the differences between fingerprints it can be found in the assertion that ‘no two things that happen by chance ever happen in exactly the same way’ (here being the way in which fingerprints are formed in the uterine environment). When examining a tyre mark left in the soft earth by the side of a burgled house, it figures as the assertion that ‘no two things are ever constructed or manufactured in exactly the same way’ (although in this case an object made from a mould will be highly similar to one another and as a practical matter the technology may not be available to individualize, say, the track of a brand-new tyre). When examining a shoemark left on the kitchen work surface by a burglar who entered via the window it can be formulated as ‘no two things ever wear in exactly the same way’. The shoes on our feet are scarred and damaged and become worn in unique ways. The tool used to force open a door will leave a mark that could not have been left by any tool but that particular one. And finally, when examining two parts of a knife blade, one of which has been recovered from a victim and the other of which remains in the possession of a suspect, the assertion that no two things ever break in exactly the same way is another version of the basic principle. Of course these asserted certainties only offer the potential for the practical realization of the application of the principle of individualization in any specific instance of its application. Two additional issues require consideration: the degree of ‘detail’ required as the grounds on which individualization may be based; and the number of such ‘details’ required to establish uniqueness. Both popular and professional images of forensic investigation have often stressed – and sometimes been surprised by – the extent to which its focus on small and seemingly insignificant detail is an essential feature of its competent practice. Locard’s reference to ‘the problem of dust’ provides a nice exemplification of this feature in that it gives us a sense of both the typical magnitude to which examinations may be oriented (the examination of ‘particulate matter’), and also the seeming irrelevance of the material to those lacking forensic awareness (the ‘unnoticed evidence of presence and action’). It is vital to note that these matters of scale and awareness give rise to the ease with which untrained or unmotivated individuals can contaminate crime scenes with the residue of their own presence and actions, contaminations that may effectively negate the skilled efforts of this group of specialists. Either or both of these elements are also present in Ginzburg’s ‘negligible details’ in his discussion of art history, in Freud’s related search for ‘little details’ throughout his psychoanalytic investigations and Conan Doyle’s description of Holmes’s fictional ‘method of trifles’. Finally, Locard’s analysis of the cosmetic powder referred to earlier in this chapter illustrates all of this nicely. It was said that the composition of the powder found under the suspect’s fingernails comprised ‘rice starch, magnesium state, zinc oxide, bismuth and “Venetian red” pigment in the same proportions as the cosmetic powder used by the victim’; it was the analysis of the composition of this ‘dust’ that made possible the claim that the powder recovered from both sources was ‘identical’ in all relevant respects. Any and every such claim raises the difficult question: how many similarities do we require to be present between an object and its impression to ensure its

Forensic Investigation as Practical Action


individualization?’ This does not permit a single answer independent of the material and context in question. For Tuthill (1994: 21–2) there are rules or conditions which may provide for ‘practically sufficient’ levels of certainty in such instances: ‘the individualization of an impression is established by finding agreement of corresponding individual characteristics by such number and significance as to preclude the possibility (or probability) of their having occurred by mere coincidence, and establishing that there are no differences that cannot be accounted for.’ In some cases it might be possible to obtain or offer probability estimates for the existence of certain events and when this is possible, it is usual to use Newcomb’s rule that the probability of the concurrence of all the whole event having occurred is equal to the continued product of the probabilities of all of the separate events having occurred. Certainly this is the formula used to calculate the random match probability of DNA profiles, but of course the procedure relies on knowing something about the distribution of such details in a larger population, along with confidence in the mathematics deployed. In many other instances of contact trace material, such distributional data are not available and there is no mathematical formula to apply. Nevertheless, accountable scene examination requires knowledge of whatever relevant working standards can be applied to assess the adequacy of particular assertions of individuation, along with the ability to master the techniques necessary for its accomplishment at whatever level of detailed focus is required. Using the Principles It should already be clear from my observations on the incorrigibility of these two central principles that I describe them here not as more or less adequate descriptions of facts of nature or as logical derivations from such facts but as justifications and instructions for certain kinds of consequential actions – in this case the organization of crime scene examinations and the credibility of the outcomes of those examinations. From this perspective then the importance of these principles lies in the heterogeneity of ways in which they are used by crime scene examiners to provide the conditions for, and shape the form of, a series of routine ‘patient and systematic’ actions whose products themselves may be used further to reinforce the weight of the principles as well as to effect the course of a criminal inquiry and contribute to a criminal prosecution. In fact the two principles are used in mutually reinforcing ways to secure both occupational and organizational authority for crime scene examiners. They provide not only incorrigible justification for the overall conduct of search procedure and for detailed actions undertaken by crime scene examiners as part of such searches, but also resources for directly controlling the conduct of others at or near relevant scenes. More than this, they are used to construct an investigative space in which the naturally occurring ‘scene of a crime’ as a place or a person in which a crime had occurred or whose body had become the subject of an assault is transformed into a ‘crime scene’ in which a repertoire of investigative actions are given authoritative priority for an indeterminate but limited time. A crime scene is not a space or a series


Orders of Ordinary Action

of objects encountered by examiners as ‘already there’. The accomplishment of the ‘scene of a crime’ as a crime scene is the product of the methodical work of crime scene examiners. Their actions constitute it as a spatially and temporally bounded region in which specific legal powers of search and seizure make possible what would be otherwise illegal actions. The investigation of some crimes will be reflected in the establishment of several crime scenes, and for many crimes the spatial boundary of a crime scene will be rather indeterminate. In the course of my own fieldwork an orientation to the principles underpinned the requirement that those who examined each of several scenes in a complex murder inquiry were both separated from one another and also separated from any suspect who may need to be examined. In addition, crime scene examiners used them as resources to control and evaluate the actions of those police officers who first attended the scenes of many domestic burglaries that I attended (scene examiners often portray first attending officers as ‘constitutionally incapable’ of acting in accord with the principles). And reference to the principles was always made by those crime scene examiners who were able to contact burglary and car crime victims in advance of their arrival at scenes in order to control both what such victims did whilst awaiting the arrival of the scene examiner and also during the course of the examination itself. Victims were, for example, asked to agree to remain in one part of the house at least until the completion of the initial part of an examination, delay any required repair to broken doors, smashed windows, etc., and to refrain from cleaning and tidying whatever disorder had occurred in the course of the crime, all until the examination had been completed. Forensic Artefacts Competent crime scene examination practice requires constant and immediate decision-making about what forensic examinations and recovery procedures should be undertaken on a variety of sites within such scenes. I have already suggested that the creation of the secure material and temporal context for this work results from the interactional realization by scene examiners of two cognitive propositions. However, the work of crime scene examiners involves direct orientations to these propositions in and through the construction of a repertoire of standardized artefacts which can be understood as the material instantiations of non-laboratory forensic sciences. These ‘forensic artefacts’, understood as deliberately created objects of attention and analysis, are treated by those who encounter them later in the narrative of any particular criminal investigation as the equivalent of, or stand-ins for, the realworld objects from which these artefacts were constructed. Typical such artefacts made by crime scene examiners during the course of the examinations I attended included: photographs taken of places in which crimes had occurred; crime-relevant objects and appearances at such scenes and of the surface appearance of physical injuries to victims; plaster casts of footwear impressions left in soft soil in gardens; ‘lifts’ made from fingermarks left on surfaces which had been first powdered to develop their features before being transferred to adhesive acetate and then backed by more acetate; ‘lifts’ made by using an Electrostatic

Forensic Investigation as Practical Action


Lifting Device which makes it possible to develop and collect footwear marks from hard surfaces; and moulds of toolmarks made by pouring a kind of plastic resin into a section of a door damaged when someone had forced it open with a variety of possible tools. Decisions to make these artefacts are conditioned in each instance of their construction by a complex set of considerations. These include not only by attentiveness to the principles I have already discussed, knowledge of the repertoire of collection techniques, ownership of distinctive technical skills, and the availability of particular technological resources, but also by prospective orientation of the scene examiner to possible investigative and evidential trajectories to which these artefacts may become decisive – or at least relevant. What part of the scene is likely to be the best source of any contact trace material awaiting discovery? And what might any artefact constructed from that material contribute to a prosecution case against a criminal suspect? A fingermark developed and lifted from the outside of a burgled window is worth little. A similar artefact made from material found inside the house may provide circumstantial evidence of the individual’s presence at a more dispositive place within the scene of a crime. A bloodstain on a screwdriver found discarded in the street outside a burgled house may not be worth collecting in order to facilitate the construction of a DNA profile from the blood – at least if the person identified through the matching of that profile admits to having dropped the screwdriver there in the course of an evening walk. The same stain on the same screwdriver discarded in the kitchen of the house has very different evidential value. Alternatively, if these or other kinds of artefacts are not usable for the construction of a case against an actual or potential suspect for this crime, scene examiners still have to consider whether or not they contribute to the collation of police intelligence usable in immediate or future investigations. For example, a shoemark might not reveal sufficient individualizing detail to be matched to a particular shoe, but might it be useful to know that what seems to be pretty much the same shoe has been found at three or four recent burglaries in the area? Equally, in the course of my fieldwork it was discovered by scene examiners that a number of burglaries had recently been carried out using a particular kind of tool (a brace and bit) to gain entry to each house. Whilst it proved impossible to individuate a particular tool through the collection and comparison of toolmarks, the fact that a common – and unusual – device was used was used as grounds to infer a common perpetrator for the series of crimes. Making a Fingermark ‘Lift’ Each kind of artefact is constructed by crime scene examiners through the exercise of particular kinds of bodily skills and with the assistance of particular technical devices. The subsequent examination and interrogation of these artefacts are undertaken by other members of the relevant police organization (and subsequently by external actors and agencies), using knowledge of these skills and devices along with the normal properties of the artefact that their routine application should produce. This


Orders of Ordinary Action

section of the chapter describes some of the skills and devices that make possible the construction of one recurrent and valuable type of crime scene examination artefact – the ‘fingermark lift’. Crime scene examiners make a series of practical decisions about how to conduct their search for fingermarks at any particular scene and also how to deal with the relative invisibility of such marks to the naked (and untrained) eye, and their work in this respect is described by them as an improvised skill: ‘knowing where to look for prints and how to develop them is a fine art learned mostly through experience’ (Fisher, 1995: 158). An understanding of the difficulties of ‘developing’ or ‘visualizing’ fingermarks rests on grasping the properties of the material nature of fingermarks themselves. Nickell and Fischer (1999) describe these as follows: The thin film that comprises a latent fingerprint contains certain substances, predominantly perspiration, secreted by the pores in the friction ridges, and/or bodily oils, acquired by contact with other parts of the body, notably the hair and the face. The perspiration is composed primarily of moisture with the added presence of sodium chloride (salt), amino acids and generally, to a lesser degree, a variety of other organic and inorganic substances. There may also be traces of other materials such as dead epidermal cells and various foreign substances. The decision regarding what treatment should be used to develop a latent print depends largely on the surface on which it lies, experience with the substance in such prints, and the technology available (Nickell and Fischer, 1999: 131).

When examining surfaces and items at burglary and car crime scenes, the scene examiners I observed used a relatively limited range of technologies, being largely concerned to enhance the discovery and recovery of fingermarks by utilizing the following: one or several sources of oblique lighting for the recognition of prints capable of development and capture; a range of powders and brushes to be used for the development of latent prints; and a limited variety of ‘lifting’ materials and methods. In addition to these portable technologies brought to the scene by examiners, some small items that might have contained fingermarks were occasionally retained without being treated in situ and were then submitted to force headquarters for cynoacetate (‘superglue’) fuming treatment. The initial identification of fingermarks relevant for and capable of further examination was usually made using available light or by the deployment of oblique lighting provided by battery-powered torches. Specialist light sources were available to these scene examiners, but their use raises health and safety issues and they are not normally used for the searching of domestic burglaries and car crime scenes unless more serious crimes against the person have been committed at the same time and in the same place. Following any such initial observations of potentially valuable fingermarks, scene examiners began more detailed work by brushing powder very lightly over particular areas of (or objects in) the scene in order to establish the quality of any prints that could potentially be recovered: as soon as such prints became visible, examiners usually tried to identify their contours to that they could brush along the contours of the ridges. This technique results in powder becoming fixed to the oily ridge deposits, with the success of the technique depending on the condition of the original mark, along with the skilled touch of the examiner.

Forensic Investigation as Practical Action


The standard kit carried to scene examinations by the crime scene examiners in the force studied contained a number of fingerprint powders and a range of fingerprint brushes for use in applying such powders in order to develop a discovered mark. Powders in regular use included: aluminium flake (a medium for the development of fingermarks found on non-porous surfaces); a bronze variant of this material (used when marks would not otherwise stand out against an aluminium-coloured background); white fingerprint powder (used on grainy surfaces and normally photographed rather than lifted); and ‘Magneta flake’, a magnetic powder suitable for use on all surfaces, including some wooden surfaces that are heavily grained. Examiners regularly used all of these media, but aluminium flake and Magneta flake powders were normally the first choice of all the examiners. Reasons for the choice between them were not always clear since a large variety of local circumstances seemed to be taken into account in the decisions made. Aluminium flake was used on windows, kitchen work surfaces, painted areas and on many items thought to have been moved or touched by suspects. Magneta flake powder was often used where it was thought capable of producing better results than the non-magnetic powders. It was sometimes asserted that this medium would provide better prints on some shiny surfaces (such as laminated paper or other coated surfaces), and it was often preferred for use on smooth plastic surfaces (including plastic bags) as well as very rough surfaces. A variety of brushes and techniques were used to enhance marks during the course of mark development. Aluminium flake (and other forms of non-magnetic powder) are applied with one of several types of brush, including those made from camel hair or squirrel hair and others based on fibreglass. Brushes are normally loaded with powder by being dipped into the powder jar, but excess powder is normally spun off the brush before applying the brush to the mark. Brushing too heavily, or with too much powder results in the obliteration of the mark through ‘clogging’ and since such excess powder is difficult to remove, powder is always applied with a very light touch. When aluminium powder is used to develop a fingerprint mark, the mark is built up slowly by the successive application of light amounts of powder on each application until the mark is clear. Earlier forms of black and white powder could be deliberately overloaded onto a mark and then brushed away, but although the use of such powders made it difficult to spoil a mark by clogging it, fingerprints developed in this way could not be lifted, but only photographed in situ. In all such cases, of course, the heavy application of a brush to a fingerprint mark can result in the oil and grease that comprises the mark being wiped away with the brush even if the brush contains little powder. When Magneta flake powder was used, it was applied to the mark by a magnetic powder applicator or ‘magnetic brush’. The applicator – which resembles a pen or pencil – has a magnet at one end which is used to pick up the powder from its container and hold it in the shape of a fuzzy globule. The powder is directly applied to the surface containing the fingerprint mark – the applicator itself does not come into contact with the mark since its hard edges will damage the mark. Excess powder can be removed from the areas surrounding the mark by the use of the applicator and when the examiner has finished developing the mark, any remaining powder on the applicator is returned to the powder container. The magnet at the end of the applicator is drawn up and since the powder is no longer


Orders of Ordinary Action

held by magnetic attraction it falls back into the container. Magneta flake marks are especially easy to brush out in cases where the examiner has overloaded the mark with the material. Those fingerprint marks successfully developed by examiners were sometimes photographed in situ, but usually an attempt was made to remove the enhanced print by the use of fingerprint-lifting technologies. Broadly speaking, these technologies are of two kinds, but both work by being pressed onto the powdered fingerprint mark and pushed on to ensure a firm contact with the material in question. In the first case the material applied in this way consists of rubber-gelatin sheets (referred to sometimes as ‘gel’ lifters, sometimes as ‘rubber’ lifters). Rubber-gelatin lifters are used to remove powdered fingerprint marks from surfaces that are rough, or are curved or embellished in some way. For example they can be pushed into wood grain in order to remove a larger surface of the fingermark and thus produce a higherquality print for subsequent examination. They comprise an adhesive fixed to a thin black or white flexible material; the adhesive side of the flexible lifter is placed onto the fingerprint and pulled away. The powdered fingerprint mark will then be covered with a transparent sheet which seals the print onto the lift. More often, however, fingerprint marks are lifted from smooth surfaces by the application of transparent adhesive sheets or ‘tape’. Most often these are pre-cut and applied in the same way as rubber-gelatin lifters. Once the tape is removed from the surface (and if it is seen to have preserved a mark of reasonable quality) it is secured for subsequent examination by being pressed onto another transparent acetate sheet. In cases of both rubber-gelatin and adhesive tape lifts, care has to be taken to record on the backing material exactly from where within the scene the print was taken, along with details of the orientation of the mark itself, who took it and when. Each separately lifted fingerprint mark (or set of adjacent marks) is also numbered by the examiner. This recorded and numbered item is also described in the overall account of the scene examination contained in a separate written report. Scene examiners formulated these locational descriptions in a variety of ways according to the specific circumstances of the case and perhaps also according to individual habit or preference. In some cases, a location description was informed by specific narrative interpretations. Examples of such descriptions include: ‘fingerprint from outer surface of POE’ (point of entry); ‘4 prints from inner surface of POE’; ‘fingerprint from cardboard file in cupboard’ (where the injured party had claimed that objects previously located in this file were now missing), ‘fingerprint lift from inner sill below POE’, ‘fingerprint from CD on kitchen floor’, ‘fingerprint lift from inner LHS of mullion to POE’. In other cases these narrative interpretations were absent (for example, ‘lift taken from inside front door handle’, ‘fingerprint lift from light bulb found on floor of bedroom’, ‘2 prints from kitchen surface’, ‘fingerprint from glass vase’. The Deconstruction of a Forensic Artefact On 5 October 1999, Alan McNamara was arrested by Greater Manchester Police on suspicion of having burgled a house in Rochdale in northwest England and of having

Forensic Investigation as Practical Action


stolen property (including a home music centre and a domestic iron) from the house, together with the houseowner’s car. None of the property or the car was discovered in McNamara’s hands and there were no witnesses who claimed to have seen him at or near the scene (the burgled house). In fact the only evidence available against him was a single thumbprint which had been recovered at the scene by a crime scene examiner and which proved to match fingerprint records of McNamara made 15 years earlier when he had been convicted of a minor offence for which he had been fined £50.00. The match between the scene thumbprint and the fingerprint record has never been disputed by any of the parties in this case. What is disputed is the account given by the crime scene examiner of where and how the crucial artefact – the thumbprint lift – was constructed. It is this dispute that I want to consider in the final section of this chapter since the same principles of ‘exchange’ and ‘individuation’ normally used to provide for the relevance and authority of the work of crime scene examiners were here turned against an individual instance of practice to challenge the credibility of the crucial artefact central to the prosecution case. McNamara claims that he had no involvement in the burglary and was never present in the house. The acetate fingermark lift made by the scene examiner which contains a clear thumbprint of McNamara is marked as having been taken from a jewellery case in a particular room in the house. However, whilst McNamara’s defence did not contest the match, it did contest the claim that the print was lifted from the particular object in question. It referred to the principle of exchange to argue that at least some features of the physical properties of the object from which the thumbprint was taken – along with the residue that comprised the mark left by the thumb which adhered to the powder and subsequently to the acetate lifting medium – should have been transferred to the lift that was made. The fingerprint powder would necessarily have coated parts of the box both within and around the thumbprint, and since the box was a wooden one, the appearance of its grain should have been transferred to the resulting lift. If the quality of the print was sufficient (if the most appropriate powder had been applied in the best possible way, and the lift had been made with sufficient care) it would even have been possible to individuate the surface from which the lift had been made. No grain at all was visible to any of those who viewed the fingerprint lift, let alone the appearance of grain sufficient to individuate the box from which the print had been lifted. Efforts were made by a number of experts to apply material, dust, polish, cosmetic powder and other contaminants, to an identical box before leaving a fingermark on it and lifting the mark. In none of these cases did it prove possible to prevent the grained surface of the box from appearing on the fingerprint lift. Whilst the police accounted for the absence of exchange in this instance by vague reference to the ‘examiner’s technique’, one of the experts used by the defence dismissed their claim by asserting that ‘no expert anywhere in the world is aware of any technique for lifting from a textured surface and dispensing with the texturing showing up in the lift’. The defence case was that the lift was taken from a different surface from the one marked up by the crime scene examiner. Once again their grounds for the assertion rested on applications of exchange and individuation principles, but this time they


Orders of Ordinary Action

were used to explain particular features of the appearance of the lift rather than the absence of other features that should have been visible on it. A second defence expert noted that the lift was ‘wrinkled’. Wrinkling commonly results from examiners having initially allowed small air pockets to occur between parts of the lifting tape and the object against which the tape was initially pressed. When the tape is pressed hard to exclude the air pockets, the tape can wrinkle (in the same way that wallpaper will wrinkle if subjected to the same process) and these wrinkles will be visible on the final lift. However, whilst individual wrinkles cannot be individuated, there are some class differences between kinds of them. In particular, when lifts are made from flat surfaces, the wrinkles are likely to be parallel, whilst wrinkles that occur when lifts are made from curved surfaces are likely to taper towards one or several common points. In the case of this disputed lift, the wrinkles tapered. Accordingly, defence experts argued that the lift had been made from a curved and not a flat surface. In particular it was suggested that there was reason to believe that they were made – as were other lifts in the course of this particular scene examination – from several vases found in different parts of the house. These were vases of a kind that it is not impossible for McNamara to have handled during the course of his business as a shop owner selling a ‘multitude of fancy goods’. It will be obvious that the resolution of the differing accounts briefly presented here is a difficult matter. McNamara was initially convicted of the burglary charge against him despite some of these arguments being heard by the jury, but his case is subject to appeal at the time of writing. There are many more aspects of this case that could be described and discussed. However, I have only briefly introduced them here to show how investigative actions accountable and accounted by reference to two incorrigible principles of crime scene examination may themselves become exhibits in a different kind of investigation – an investigation in which a focus on the material products of crime scene examination is displaced by a focus on the human and social processes involved in their production. It will be clear to readers of this chapter that an orientation to and recitation of these particular formal structures comprises no more than one recurrent instantiation of the complex ways in which many formal structures may be brought into play in the course of crime scene examination. Whilst some of these may have surfaced briefly here, the uses made of other elements of this array of methods, technologies, guidelines, etc., in support of criminal investigations deserve closer and more extensive exploration.

Bibliography Aanestad, M. and Hanseth, O. (2000), ‘Implementing open network technologies in complex work practices: A case from telemedicine’, in Organizational and Social Perspectives on Information Technology, Dordrecht: Kluwer, 355–369. Anspach, R.R. (1988), ‘Notes on the sociology of medical discourse: The language of case presentation’, Journal of Health and Social Behavior 29, 357–75. Atkinson, J.M. and Drew, P. (1979), Order in Court: The Organization of Verbal Interaction in Judicial Settings, London: Macmillan. Atkinson, P. (1988), ‘Discourse, descriptions and diagnoses: Reproducing normal medicine’, in M. Lock and D. Gordon (eds), Biomedicine Examined, Dordrecht: Kluwer, 179–204. Ball, M. (2000), ‘The visual availability and local organization of public surveillance systems: The promotion of social order in public space’, Sociological Research Online , 5 (1). Bell, D.G., Bobrow, D.G., Raiman, O. and Shirley, M. (1997), ‘Dynamic documents and situated processes: Building on local knowledge in field service’, in T. Wakayama, S. Kannapan, C.M. Khoong, S. Navathe and J. Yates (eds), Information and Process Integration in Enterprises: Rethinking Documents, Norwell MA: Kluwer. Bell, D.G., Raiman, O., Shirley, M.H. and Bobrow, D.G. (1995), ‘Community learning and the individual: Valuing diversity of experience and reasoning styles in field service’, Xerox PARC: unpublished paper. Berg, M. (1997), Rationalizing Medical Work: Decision-Support Techniques and Medical Practices, Cambridge MA and London: MIT Press. Bittner, E. (1965), ‘The concept of organization’, Social Research 32, 239–55. Bittner, E. (1967a), ‘The police on skid row: A study of peace keeping’, American Sociological Review 32, 699–715. Bittner, E. (1967b), ‘Police discretion in emergency apprehension of mentally ill persons’, Social Problems 14, 278–92. Bittner, E. (1977), ‘Must we say what we mean?’, in P.F. Ostwald (ed.), Communication and Social Interaction, London: Grune & Stratton; pp. 83–97. Bjelic, D.I. (1995), ‘An ethnomethodological clarification of Husserl’s concepts of “Regressive Inquiry” and “Galilean Physics” by means of discovering praxioms’, Human Studies 18, 189–225. Blanks, R., Wallis, M. and Moss, S. (1998), ‘A comparison of cancer detection rates achieved by breast screening programmes by number of readers, for one and two view mammography: Results from the UK National Health Service breast screening programme’, Journal of Medical Screening 5 (4), 195–201. Bobrow, D.G. and Whalen, J. (2000), ‘Community knowledge sharing in practice: The Eureka story’, Reflections: The SOL Journal 4 (2), 47–59.


Orders of Ordinary Action

Bourgois, P. (1998), ‘The moral economies of homeless heroin addicts: Confronting ethnography, HIV risk, and everyday violence in San Francisco shooting encampments’, Substance Use and Misuse 33, 2323–51. Burke, K. (1935), Permanence and Change, New York: New Republic. Burns, S. (1999/2000), ‘Making settlement work: An examination of the work of judicial mediators’, PhD thesis, Dept. of Sociology, University of California, Los Angeles (published under the same title in 2000 by Dartmouth/Ashgate: Aldershot UK). Burns, S. (2001), ‘ “Think your darkest thoughts and blacken them”: Judicial mediation of large money damage disputes’, Human Studies 24, 227–49. Burns, T. (1992), Erving Goffman, London: Routledge. Button, G. (1991), ‘Ethnomethodology and the foundational respecification of the human sciences’, in G. Button (ed.), Ethnomethodology and the Human Sciences, Cambridge: Cambridge University Press; pp. 1–9. Button, G. (ed.) (1991), Ethnomethodology and the Human Sciences, Cambridge, Cambridge University Press. Button, G. (ed.) (1993), Technology in Working Order: Studies of Work, Interaction and Technology, London: Routledge. Button, G., Coulter, J., Lee, J.R.E. and Sharrock, W. (1995), Computers, Minds and Conduct, Cambridge: Polity Press. Cano, V. (1999), ‘Bibliometric overview of library and information science research in Spain’, Journal of the American Society for Information Science 50, 675–80. Carlin, A.P. (1999), ‘On the linguistic constitution of research practices’, unpublished PhD thesis, Faculty of Human Sciences, University of Stirling. Carlin, A.P. (2001), ‘ “Enquadrando” bibliografias: Reflexividade, relevância e a “Imaginaçao Sociologica” ’, in E. Gastaldo (ed.), Erving Goffman: Desbravador Do Cotidiano, Porto Alegre: Tomo Editorial (forthcoming). Carlin, A.P., Evergeti, V. and Murtagh, G.M. (1999), ‘Visible identities at work: Preliminary observations on categorizations and “street performers” in Cathedral Square’, paper presented at the Annual Conference of the British Sociological Association, University of Glasgow, 6–9 April 1999. Case, D.O. and Higgins, G.M. (2000), ‘How can we investigate citation behavior? A study of reasons for citing literature in communication’, Journal of the American Society for Information Science 51, 635–45. Cicourel, A.V. (1964), Method and Measurement in Sociology, New York: Free Press. Collins, O. and Collins, J.M. (1973), Interaction and Social Structure, The Hague: Mouton. Contu, A. and Wilmott, H. (2003), ‘Re-embedding situatedness: The importance of power relations in learning theory’, Organization Science 14 (3), 283–96. Coulter, J. (1974), ‘The ethnomethodological programme in contemporary sociology’, The Human Context 6, 103–22. Coulter, J. (1991), ‘Cognition; cognition in an ethnomethodological mode’, in G. Button, (ed.), Ethnomethodology and the Human Sciences, Cambridge: Cambridge University Press; pp. 176–95. Cozzens, S.E. (1989), ‘What do citations count? The rhetoric-first model’, Scientometrics 15, 437–47.



Crane, D. (1972), Invisible Colleges: Diffusion of Knowledge in Scientific Communities, Chicago and London: University of Chicago Press. Cronin, B. (1984), The Citation Process, London: Taylor Graham. Cuff, E.C. (1994), Problems of Versions in Everyday Situations, Lanham MD: University Press of America and International Institute for Ethnomethodology and Conversation Analysis. Davenport, T. (1995), ‘Epitaph for expert systems’, Informationweek, June 5, 116. de Kleer, J. and Williams, B.C. (1987), ‘Diagnosing multiple faults’, Artificial Intelligence 32, 97–130. Dean, H. and Gale, K. (1999), ‘Begging and the contradictions of citizenship’, in H. Dean (ed.), Begging Questions: Street-Level Economic Activity and Social Policy Failure, Bristol: Policy Press; pp. 13–26. Dettmar, K.J.H. (1999), ‘There’s always room for cello: Irony and the logics of alt. rock’, Strategies 12, 35–43. Dewe, M. (1995), Planning and Designing Libraries for Children and Young People, London: Library Association Publishing. Douglas, J.D. (1967), The Social Meanings of Suicide, Princeton NJ: Princeton University Press. Drew, P. and Heritage, J. (eds) (1992), Talk at Work, Cambridge: Cambridge University Press. Drew, P. and Wootton, A. (eds) (1988), Erving Goffman: Exploring the Interaction Order, Oxford: Polity Press. Driessen, J. (1997), ‘Worldly interpretations of a suspicious story’, Ethnographic Studies 1 (2), 3–15. Eco, U. (1983), ‘Horns, hooves, insteps: some hypotheses on three types of abduction’, in U. Eco and T.A. Sebeck (eds), The Sign of Three: Dupin, Holmes, Peirce, Bloomington IN: Indiana University Press, pp. 198–220. Edmond, G. and Mercer, D. (1998), ‘Representing the sociology of scientific knowledge and law’, Science Communication 19 (4), 307–27. Farrell, E.P., Führer, E., Ryan, D., Andersson, F., Hüttl, R. and Piussi, P. (2000), ‘European forest ecosystems: Building the future on the legacy of the past’, Forest Ecology and Management 132, 5–20. Fehr, B.J. and Stetson, J. (with Mizukawa, Y.) (1990), ‘A bibliography for ethnomethodology’, in J. Coulter (ed.), Ethnomethodological Sociology, Aldershot: Edward Elgar, 473–559. Feigenbaum, E.A., McCorduck, P. and Nii, H.P. (1988), The Rise of the Expert Company: How Visionary Companies Are Using Artificial Intelligence to Achieve Higher Productivity and Profits, New York: Time Books. Fisher, D. (1995), Hard Evidence, New York: Dell. Foucault, M. (1977), Discipline and Punish: The Birth of the Prison, translated from the French by Alan Sheridan, London: Allen Lane. Foucault, M. (1989), The Archaeology of Knowledge, London: Routledge. Fox, N. J. (1992), The Social Meaning of Surgery, Milton Keynes: Open University Press. French, A.P. (1971), Newtonian Mechanics, New York: W.W. Norton; pp. 238–9. Fuchs, S. and Ward, S. (1994), ‘What is deconstruction, and where and when does it


Orders of Ordinary Action

take place? Making facts in science, building cases in law’, American Sociological Review 59, 481–500. Garfield, E. (1955), ‘Citation indices for science’, Science 122, 109–10. Garfield, E. (1970), ‘Citation indexing for studying science’, Nature 227, 669–71. Garfield, E. (1977), Essays of an Information Scientist, vol. 1, Philadelphia PA: ISI Press. Garfinkel, H. (1946), ‘Color trouble’, in B. Moon (ed.), Primer for White Folks, Garden City NY: Doubleday; pp. 269–86. Garfinkel, H. (1949), ‘Research note on inter- and intra-racial homicides’, Social Forces 27, 369–81. Garfinkel, H. (1967a), Studies in Ethnomethodology, Oxford: Polity Press. Garfinkel, H. (1967b), ‘Some rules of correct decisions that jurors respect’, in Studies in Ethnomethodology, Englewood Cliffs NJ: Prentice Hall; Chapter 4, pp. 104–15. Garfinkel, H. (1967c), ‘Practical sociological reasoning: Some features in the work of the Los Angeles Suicide Prevention Center’, in E.S. Shneidman (ed.), Essays in Self-Destruction, New York: Jason Aronson; pp. 171–87. Garfinkel, H. (1974), ‘On the origins of ethnomethodology’, in R. Turner (ed.), Ethnomethodology, Harmondsworth: Penguin; pp. 15–18. Garfinkel, H. (1976), A Manual for Studies of Naturally Organized Ordinary Activities, unpublished mimeo, Department of Sociology, UCLA. Garfinkel, H. (1977a), ‘When is phenomenology sociological?’, Annals of Phenomenological Sociology 2, 1–40. Garfinkel, H. (1977b), University lectures, University of California, Los Angeles. Garfinkel, H. (1991), ‘Respecification: Evidence for locally produced, naturally accountable phenomena of order, logic, reason, meaning, method, etc., in and as of the essential haecceity of immortal ordinary society, (I) an announcement of studies’, in G. Button (ed.), Ethnomethodology and the Human Sciences, Cambridge: Cambridge University Press; pp. 10–19. Garfinkel, H. (1993), ‘A catalog of investigations with which to respecify topics of logic, order, meaning, method, reason, structure, science, and the rest, in, about, and as the workings of immortal, ordinary society just in any actual case. What did we do? What did we learn?’, unpublished manuscript, Department of Sociology, UCLA (31 October). Garfinkel, H. (1996), ‘Ethnomethodology’s program’, Social Psychology Quarterly 59, 5–21. Garfinkel, H. (2001), ‘The corpus status of ethnomethodological investigations: Four relations between literatures of the social science movement and their specific ethnomethodological alternates’. Plenary address to the International Conference ‘Orders of Ordinary Action’, International Institute for Ethnomethodology and Conversation Analysis, Manchester Metropolitan University, 9–11 July 2001. Garfinkel, H. (2002), Ethnomethodology’s Program: Working Out Durkheim’s Aphorism, Lanham MD: Rowman & Littlefield. Garfinkel, H. (ed.) (1986), Ethnomethodological Studies of Work, New York: Routledge.



Garfinkel, H. and Bittner, E. (1967a), ‘Methodological adequacy in the quantitative study of selection criteria and selection practices in psychiatric outpatient clinics’, in H. Garfinkel, Studies in Ethnomethodology, Oxford: Polity Press; pp. 208–61. Garfinkel, H. and Bittner, E. (1967b), ‘ “Good” organizational reasons for “bad” clinic records’, in H. Garfinkel, Studies in Ethnomethodology, Oxford: Polity Press; pp. 186–207. Garfinkel, H., Livingston, E., Lynch, M., Macbeth, D. and Robillard, A. (1989), ‘Respecifying the natural sciences as discovering sciences of practical action (I and II): Doing so ethnographically by administering a schedule of contingencies in discussions with laboratory scientists and by hanging around their laboratories’, unpublished paper, Department of Sociology, UCLA. Garfinkel, H., Lynch, M. and Livingston, E. (1981), ‘The work of a discovering science construed with materials from the optically discovered pulsar’, Philosophy of the Social Sciences 11, 131–58. Garfinkel, H. and Rose, E. (1973), Mystifying Kinds of Achievement. Conversation (15 July), reprinted 1997 with a new introduction by Edward Rose entitled ‘Rose’s gloss on talk about achievements’, Greeley CO: The Waiting Room Press. Garfinkel, H. and Sacks, H. (1970), ‘On formal structures of practical actions’, in C. McKinney and E.A. Tiryakian (eds), Theoretical Sociology: Perspectives and Developments, New York: Appleton-Century-Crofts; pp. 337–66. Garfinkel, H. and Sudnow, D. (1972), ‘Some feasibility issues in our study of lectures’, Subcommittee on Academic Disciplines Working Paper #3, University of Colorado, Boulder. Garfinkel, H. and Sudnow, D. (1975), ‘A conjecture about an ignored orderliness of lectures as university-specific work’, unpublished manuscript, Department of Sociology, UCLA. Garfinkel, H. and Wieder, D.L. (1992), ‘Two incommensurable, asymmetrically alternate technologies of social analysis’, in G. Watson and R.M. Seiler (eds), Text in Context: Contributions to Ethnomethodology, London: Sage; pp. 175–206. Garfinkel, H. and Wiley, N. (1980), unpublished transcript, 31 March, Department of Sociology, UCLA. Geertz, C. (1973), Interpretation of Cultures, New York: Basic Books. Gendlin, E. (1992), ‘Thinking beyond patterns’, in B. Den Ouden and M. Moen (eds), The Presence of Feeling in Thought, New York: Peter Lang. Giddens, A. (1984), The Constitution of Society, Cambridge, Polity Press. Giddens, A. and Held, D. (1982), Classes, Power and Conflict: Classical and Contemporary Debates, London: Macmillan. Glaser, B.G. and Strauss, A.L. (1968), Time for Dying, Chicago: Aldine. Goffman, E. (1951), ‘Symbols of class status’, British Journal of Sociology 2, 294–304. Goodwin, C. (1986), ‘Audience diversity, participation and interpretation’, Text 6 (3), 283–316. Goodwin, C. (1994), ‘Professional vision’, American Anthropologist 96 (3), 606–33. Goodwin, C. (1995), ‘Seeing in depth’, Social Studies of Science 25, 237–74. Goodwin, C. (1997), ‘The blackness of black: Color categories as situated reasoning’, in L.B. Resnick, R. Saaljo and C. Pontecorvo (eds), Discourse, Tools,


Orders of Ordinary Action

and Reasoning (NATO Asi Series. Series F, Computer and Systems Sciences, vol. 160), Berlin: Springer; pp. 111–140. Goodwin, C. and Goodwin, M.H. (1996), ‘Seeing as a situated activity: Formulating planes’, in D. Middleton and Y. Engestrom (eds), Cognition and Communication at Work, Cambridge: Cambridge University Press. Gould, J. (n.d. [1st edn 1884]), Gould’s Problems, Critical Positions, and Games, 4th edn, London: E. Marlborough, p. 8. Gurwitsch, A. (1964), The Field of Consciousness, Pittsburgh PA: Duquesne University Press. Halford, M.M. (1987), ‘Student clinicians’ difficulties while conducting the summary diagnostic interview’, Text 7 (1), 19–36. Harper, R. and Hughes, J. (1992), ‘ “What a f— system! Send ’em all to the same place and then expect us to stop ’em hitting!”: Making technology work in air traffic control’, in G. Button (ed.), Technology in Working Order: Studies of Work, Interaction and Technology, London, Routledge. Harper, R., Randall, D. and Rouncefield, M. (1999), Organizational Change and Retail Finance: An ethnographic perspective, London and New York: Routledge. Hartland, J. (1993), ‘The use of “intelligent” machines for electrocardiograph interpretation’, in G. Button (ed.), Technology in Working Order: Studies of Work, Interaction and Technology, London: Routledge. Hartswood, M. and Procter, R. (2000), ‘Computer-aided mammography: A case study of coping with fallibility in a skilled decision-making task’, in C. Johnson (ed.), Journal of Topics in Health Information Management, special issue on human error and computer systems, 20 (4), 38–54. Hatch, D.J. and Watson, D.R. (1974), ‘Hearing the blues: An essay in the sociology of music’, Acta Sociologica 17, 162–78. Heap, J.L. (ed.) (1974), Everybody’s Canada: The Vertical Mosaic Reviewed and Re-Examined, Toronto: Burns & MacEachern. Heath, C. (1986), Body Movement and Speech in Medical Interaction, Cambridge: Cambridge University Press. Heath, C. and Luff, P. (1993), ‘Disembodied conduct: Interactional asymmetries in video-mediated communication’, in G. Button (ed.), Technology in Working Order: Studies of Work, Interaction and Technology, London: Routledge. Heath, C. and Luff, P. (2000), Technology in Action, Cambridge: Cambridge University Press. Heath, C., Luff, P. and Sanchez Svensson, M. (2002), ‘Technology and medical practice’, Sociology of Health and Illness, 25, 75–96. Helm, D., Anderson, T., Meehan, J. and Rawls, A. (eds) (1989), The Interactional Order: New Directions in the Study of Social Order, New York: Irvington Press. Heritage, J. (1984), Garfinkel and Ethnomethodology, Oxford: Polity Press. Hester, S. and Eglin, P. (eds) (1997), Culture in Action: Studies in Membership Categorization Analysis, Lanham MD: University Press of America and International Institute for Ethnomethodology and Conversation Analysis. Hill, R.J. and Crittenden, K.S. (1968), Proceedings of the Purdue Symposium on Ethnomethodology, Purdue IN: Institute for the Study of Social Change, Department of Sociology, Purdue University.



Hirschauer, S. (1991), ‘The manufacture of bodies in surgery’, Social Studies of Science 21 (2), 279–319. Hirst, P.H. (1972), ‘Liberal education and the nature of knowledge’, in R.D. Archambault (ed.), Philosophical Analysis and Education, London: Routledge & Kegan Paul; pp. 113–138. Home Office (1949), Scientific Aids to Criminal Investigation, London: HMSO. Husserl, E. (1969), Formal and Transcendental Logic, The Hague: Martinus Nijhoff. Hutchins, E. (1995), Cognition in the Wild, Cambridge MA: MIT Press. Ikeya, N. (1997), The Practical Management of the Social Stock of Knowledge: The Case of an Information-Giving Service, unpublished PhD thesis, University of Manchester. Ikeya, N. (2001), ‘Seikatsu-sekai to joho’ [‘Life world and information’], in S. Tamura (ed.), Joho-tansaku to Joho-riyo [Information Seeking and Information Uses], Tokyo: Keiso Shobo; pp. 41–90. Ikeya, N. (forthcoming), ‘Practical management of mobility: a case of emergency medical system’. Ikeya, N., Okada, M. and Fujimori, Y. (2004), ‘Fieldwork in the medical organization’, in K. Yamazaki (ed.), A Practical Introduction to Ethnomethodology, Tokyo: Yuhikaku. Jackson, P. (1990), Introduction to Expert Systems, 2nd edn, Wokingham (UK): Addison-Wesley. Jalbert, P.L. (ed.) (1998), Media Studies: Ethnomethodological Approaches, Lanham MD: University Press of America and International Institute for Ethnomethodology and Conversation Analysis. Jamieson, A. (2004), ‘A rational approach to the principles and practice of crime scene investigation: 1. Principles’, Science and Justice 44 (1), 3–7. Jasanoff, S. (1996), Science at the Bar: Law, Science and Technology in America, Cambridge MA: Harvard University Press Jules-Rosette, B. (1985), ‘Conversation avec Harold Garfinkel’, Sociétés: Revue des Sciences Humaines et Sociales 1, 35–9. Kendon, A. (1990), Conducting Interaction, Cambridge: Cambridge University Press. Kessler, M.M. (1963), ‘Bibliographic coupling between scientific papers’, American Documentation 14, 10–25. Khapa, T. (1984), Lam gyi gtso bo rnam gsum gyi rtsa ba, Dharamsala, India: Tibetan Cultural Printing Press. (Originally written in the 15th century CE.) Kitsuse, J. and Cicourel, A.V. (1963), ‘A note on the uses of official statistics’, Social Problems 11, 131–9. Knorr-Cetina, K. (1981), The Manufacture of Knowledge: An Essay on the Constructivist and Contextual Nature of Science, Oxford: Pergamon Press. Koschman, T. et al. (2007), ‘Formulating the triangle of doom’, Gesture, 7/1, 97– 118. Kuhn, T.S. (1970), The Structure of Scientific Revolutions, 2nd edn, Chicago: University of Chicago. Kundell, H.L., Nodine, C.F. and Toto, L. (1991), ‘Searching for lung nodules: The guidance of visual scanning’, Investigative Radiology 26 (9), 777–81.


Orders of Ordinary Action

Latour, B. (1987), Science in Action: How to Follow Scientists and Engineers through Society, Cambridge MA: Harvard University Press. Latour, B. (1993), We Have Never Been Modern, Cambridge MA: Harvard University Press. Latour, B. and Woolgar, S. (1979), Laboratory Life: The Social Construction of Scientific Facts (London: Sage; Princeton NJ: Princeton University Press, 1986). Lemert, C. and Branaman, A. (eds) (1997), The Goffman Reader, Oxford: Blackwell. Levinas, Emmanuel (1981), Otherwise than Being or Beyond Essence, The Hague: Martinus Nijhoff. Lewis, D. (1972), We the Navigators, The Ancient Art of Landfinding in the Pacific, Honolulu: University of Hawaii Press. Liberman, K. (1992), ‘Philosophical debate in the Tibetan Academy’, The Tibet Journal 17 (1), 36–67. Liberman, K. (1999), ‘The social praxis of communicating meanings’, Text 19 (1), 272–82. Liberman, K. (2004), Dialectical Practice in Tibetan Philosophical Culture, Lanham, Maryland: Rowman & Littlefield. Liberman, K. (2007), Husserl’s Criticism of Reason, with Ethnomethodological Specifications, New York: Lexington Books. Livingston, E. (1986), The Ethnomethodological Foundations of Mathematics, London: Routledge & Kegan Paul. Livingston, E. (1987), Making Sense of Ethnomethodology, London: Routledge & Kegan Paul. Livingston, E. (1999), ‘Cultures of proving’, Social Studies of Science 29(6), 867– 88. Livingston, E. (2000), ‘The availability of mathematics as an inspectable domain of practice through the use of origami’, in S. Hester and D. Francis (eds), Local Educational Order, Amsterdam: John Benjamins; pp. 245–70. Livingston, E. (2005), ‘Natural reasoning in mathematical theorem proving’, Communication & Cognition 38 (3/4), 319–44. Livingston, E. (2006), ‘Ethnomethodological studies of mediated interaction and mundane expertise’, Sociological Review 54 (3), 405–25. Lomax, H. and Casey, N. (1998), ‘Recording social life: Reflexivity and video methodology’, Sociological Research Online , 3 (2). Luff, P., Hindmarsh, J. and Heath, C. (eds) (2000), Workplace Studies. Recovering Work Practice and Informing System Design, Cambridge: Cambridge University Press. Lynch, M. (1985), ‘Discipline and the material form of images: An analysis of scientific visibility’, Social Studies of Science 15, 37–66. Lynch, M. (1986), Art and Artefact in Laboratory Science, London: Routledge & Kegan Paul. Lynch, M. (1988), ‘The externalized retina: Selection and mathematization in the visual documentation of objects in the life sciences’, Human Studies 11, 201–34. Lynch, M. (1991), ‘Method: measurement – ordinary and scientific measurement as ethnomethodological phenomena’, in G. Button, (ed.), Ethnomethodology and the Human Sciences, Cambridge: Cambridge University Press: pp. 77–108.



Lynch, M. (1993), Scientific Practice and Ordinary Action: Ethnomethodology and Social Studies of Science, Cambridge: Cambridge University Press. Lynch, M. (1996), ‘Detoxifying the “poison pen effect” ’, in A. Ross (ed.), Science Wars, Durham NC: Duke University Press: pp. 238–58. Lynch, M. (1998), ‘Towards a constructivist genealogy of social constructivism’, in I. Velody and R. Williams (eds), The Politics of Constructionism, London, Thousand Oaks CA and New Delhi: Sage; pp. 13–32. Lynch, M., Cole, S., McNally, R. and Jordan, K. (forthcoming), Truth Machine: The Contentious History of DNA Fingerprinting, Chicago: University of Chicago Press. Lynch, M., Livingston, E. and Garfinkel, H. (1983), ‘Temporal order in laboratory work’, in K. Knorr-Cetina and M. Mulkay (eds), Science Observed: Perspectives on the Social Study of Science, London (UK) and Beverly Hills CA: Sage: pp. 205–38. Lynch, M. and McNally, R. (2003), ‘ “Science”, “common sense”, and DNA evidence: A legal controversy about the public understanding of science’, The Public Understanding of Science 12 (1), 83–105. MacAndrew, C. and Garfinkel, H. (1962), ‘A consideration of changes attributed to intoxication as common-sense reasons for getting drunk’, Quarterly Journal of Studies on Alcohol 23, 252–66. Macbeth, D. (1992), ‘Classroom “floors”: Material organizations as a course of affairs’, Qualitative Sociology 15. Macbeth, D. (1999), ‘Glances, trances, and their relevance for a visual sociology’, in P. L. Jalbert (ed.), Media Studies: Ethnomethodological Approaches, Lanham MD: University Press of America and International Institute for Ethnomethodology and Conversation Analysis: pp. 135–70. McInnis, R.G. and Symes, D. (1988), ‘David Riesman and the concept of bibliographic citation’, College and Research Libraries 49, 387–99. Mahan, B.H. (1965), University Chemistry, Palo Alto CA: Addison-Wesley. Manzo, J. and Maynard, D. (1993), ‘On the sociology of justice: Theoretical notes from an actual jury deliberation’, Sociological Theory 11, 171–93. Medawar, P. (1964), ‘Is the scientific paper fraudulent? Yes; it misrepresents scientific thought’, Saturday Review 42–3. Meehan, M.J. (1986), ‘Record-keeping practices in the policing of juveniles’, Urban Life 15 (1), 70–102. Merleau-Ponty, M. (1968), The Visible and the Invisible, Evanston IL: Northwestern University Press. Mondada, L. (2001), ‘Intervenir à distance dans une opération chirurgicale: L’organization interactive d’espaces de participation’, Bulletin Suisse de Linguistique Appliquée, 74. Mondada (2002a), ‘Opérer et enseigner à opérer. Description de l’action et formulation du savoir-faire’, in C. Chauviré and A. Ogien (eds), La régularité. Habitude, disposition et savoir-faire dans l’explication de l’action. Raisons Pratiques 13, 293–318. Mondada, L. (2002b), ‘Processi di categorizzazione e produzione dell’inclusione e dell’esclusione. A proposito di una riunione di lavoro tramite videoconferenza’, in G. Klein and I. Paoletti (eds), IN & OUT. Procedure conversazionali e strategie


Orders of Ordinary Action

comunicative di inclusione e esclusione, Napoli: Edizioni Scientifiche Italiane, pp. 117–175. Mondada, L. (2003), ‘Working with video: How surgeons produce video records of their actions’, Visual Studies 18 (1), 58–72. Mondada, L. (2004), ‘Téléchirurgie et nouvelles pratiques professionnelles: les enjeux interactionnels d’opérations chirurgicales par réalisées par visioconférence’, Sciences Sociales et Santé, 22, 1, 95–126. Mondada, L. (2006), ‘Video Recording as the Preservation of Fundamental Featured for Analysis’, in Knoblauch, H., Raab, J., Soeffner, H.-G. and Schnettler, B. (eds), Video Anaysis, Bern: Lang, 51–68. Mondada, L.( 2007), ‘Bilingualism and the analysis of talk at work: Code-switching as a resource for the organization of action and interaction’, in Heller, M. (ed.), Bilingualism: A Social Approach, New York: Palgrave, 297–318. Mullins, N.C. (1973), ‘The development of specialties in social science: The case of ethnomethodology’, Science Studies 3, 245–73. Nickell, J. and Fischer, J.F. (1999), Crime Science: Methods of Forensic Detection, Lexington KY: University of Kentucky Press. Nordby, J.J. (2000), Dead Reckoning: The Art of Forensic Detection, Boca Raton FL: CRC. Orr, J.E. (1996), Talking About Machines: An Ethnography of a Modern Job, Ithaca NY: Cornell University Press. Oteri, J.S., Weinberg, M.G. and Pinales, M.S. (1982), ‘Cross-examination in drug cases’, in B. Barnes and D. Edge (eds), Science in Context: Readings in the Sociology of Science, Milton Keynes: Open University Press: pp. 250–259. Peneff, J. (1997), ‘Le travail du chirurgien: Les opérations à coeur ouvert’, Sociologie du travail 39 (3), 265–96. Peyrot, M. (1995), ‘Psychological testing and forensic decision-making: The properties-in-use of the MMPI’, Social Problems 42 (4), 574–86. Pilnick, A. and Hindmarsh, J. (1999), ‘When you wake up it’ll be over’: Communication in the anaesthetic room’, Symbolic Interaction 22 (4), 345–60. Plato (1961), ‘Euthydemus’, from E. Hamilton and H. Cairns (eds), The Collected Dialogues of Plato, Princeton NJ: Princeton University Press. Pollner, M. (1974), ‘Mundane reasoning’, Philosophy of the Social Sciences, 4, 35– 54. Pomerantz, A.M. (1987), ‘Descriptions in legal settings’, in G. Button and J.R.E. Lee (eds), Talk and Social Organization, Clevedon (UK): Multilingual Matters; pp. 226–43. Pomerantz, A., Ende, J. and Erickson, F. (1997), ‘Precepting conversations in a general medicine clinic’, in G.H. Morris and R.J. Chenail (eds), The Talk of the Clinic: Explorations in the Analysis of Medical and Therapeutic Discourse, Hillsdate NJ: Lawrence Erlbaum: pp. 151–69. Prabha, C. (1983), ‘Some aspects of citation behaviour: A pilot study in business administration’, Journal of the American Society for Information Science 34, 202–6. Psathas, G. (ed.) (1979), Everyday Language: Studies in Ethnomethodology, New York: Irvington Press.



Psathas, G. (ed.) (1990), Interaction Competence, Lanham MD: University Press of America and International Institute for Ethnomethodology and Conversation Analysis. Rawls, A. (1996), ‘Durkheim’s epistemology: The neglected argument’, American Journal of Sociology 102, 430–82. Rawls, A. (2001), ‘Durkheim’s treatment of practice’, Journal of Classical Sociology 1, 33–68. Relieu, M. (1999), ‘Travaux en public. La dynamique d’une situation problématique’, Raisons Pratiques 10, 95–118. Rietz, R.E. du (1974), ‘What is bibliography?’, Text: Swedish Journal of Bibliography 1, 6–40. Robillard, A.B. (1999), ‘Wild phenomena and disability jokes’, Body and Society 5, 61–5. Rose, E. (1991), ‘A conversation with Harvey Sacks’, Proceedings of the Conference on Current Work in Ethnomethodology and Conversational Analysis, Amsterdam. Rose, E. (1997), ‘The unattached society’, Ethnographic Studies 1, xv–43. Sacks, H. (1972a), ‘An initial investigation of the usability of conversational materials for doing sociology’, in D. Sudnow (ed.), Studies in Social Interaction, New York: Free Press; pp. 31–74. Sacks, H. (1972b), ‘Notes on police assessment of moral character’, in D. Sudnow (ed.), Studies in Social Interaction, New York: Free Press; pp. 280–93. Sacks, H. (1974), ‘An analysis of the course of a joke’s telling in conversation’, in R. Bauman and J. Scherzer (eds), Explorations in the Ethnography of Speaking, London: Cambridge University Press: pp. 337–53. Sacks, H. (1984), ‘Notes on methodology’, in J.M. Atkinson and J.C. Heritage (eds), Structures of Social Action: Studies in Conversation Analysis, Cambridge: Cambridge University Press; pp. 21–7. Sacks, H. (1992), Lectures on Conversation, Oxford: Blackwell. Sacks, H. (1999), ‘Max Weber’s Ancient Judaism’, Theory, Culture and Society 16, 31–9. Sacks, H., Schegloff, E. and Jefferson, G. (1974). ‘A simplest systematics for the organization of turn-taking in conversation’, Language 50, 696–735. Schutz, A. (1962), ‘Common-sense and scientific interpretation of human action’, in A. Schutz, Collected Papers I: The Problem of Social Reality, The Hague, Boston and London: Martinus Nijhoff; pp. 3–47. Schutz, A. (1964), ‘The well-informed citizen’, in A. Schutz, Collected Papers II: Studies in Social Theory, The Hague: Martinus Nijhoff; pp. 120–34. Schwabe, A.D., Solomon, D.H., Stoller, R.J. and Burnham, J.P. (1962), ‘Pubertal feminization in a genetic male with testicular atrophy and normal urinary gonadotropin’, Journal of Clinical Endocrinology and Metabolism 22, 839– 45. Schwartz, H. (1976), ‘On recognizing mistakes: A case of practical reasoning in psychotherapy’, Philosophy of the Social Sciences 6, 55–73. Shadish, W.R., Tolliver, D., Gray, M. and Gupta, S.K.S. (1995), ‘Author judgements about works they cite: Three studies from psychology journals’, Social Studies of Science 25, 477–98.


Orders of Ordinary Action

Sharrock, W. and Button, G. (1991), ‘The social actor: Social action in real time’, in G. Button (ed.), Ethnomethodology and the Human Sciences, Cambridge: Cambridge University Press; pp. 137–75. Sharrock, W. and Ikeya, N. (2000), ‘Instructional matter: Readable properties of an introductory text in matrix algebra’, in S. Hester and D. Francis (eds), Local Educational Order: Ethnomethodological Studies of Knowledge in Action, Amsterdam and Philadelphia PA: John Benjamins. Silverman, D. (1993), Interpreting Qualitative Data, London, Thousand Oaks CA and New Delhi: Sage. Simon, H.A. (1965), The Shape of Automation for Men and Management, New York: Harper Row. Small, H.G. (1973), ‘Co-citation in the scientific literature: A new measure of the relationship between two documents’, Journal of the American Society for Information Science 24, 265–9. Smith, R. and Wynne, B. (eds) (1989), Expert Evidence: Interpreting Science in the Law, London: Routledge. Sorokin, P.A. (1931), ‘Sociology as a science’, Social Forces 10, 20–27. Spayth, H. (n.d., ca. 1890), The American Draught Player, 6th edn, New York: Dick and Fitzgerald, p. 252. Stoller, R.J., Garfinkel, H., and Rosen, A.C. (1960), ‘Passing and the maintenance of sexual identification of an intersexed patient’, Archives of General Psychiatry 2, 379–84. Suchman, L. (1987), Plans and Situated Actions, Cambridge (UK): Cambridge University Press. Suchman, L. (1993), ‘Technologies of accountability: Of lizards and airplanes’, in G. Button (ed.), Technology in Working Order: Studies of Work, Interaction and Technology, London: Routledge; pp. 113–126. Suchman, L. (1995), Making Work Visible. Communications of the ACM 38 (9), 56–64. Sudnow, D. (1965), ‘Normal crimes: Sociological features of the penal code in a public defender office’, Social Problems 12, 255–76. Sudnow, D. (1978), Ways of the Hand, Cambridge MA: Harvard University Press. ten Have, P. (1989), ‘The consultation as a genre’, in B. Torode (ed.), Text and Talk as Social Practice: Discourse Difference and Division in Speech and Writing, Dordrecht and Providence RI: Foris Publications, 115–135. ten Have, P. (1991), ‘Talk and institution: A reconsideration of the “asymmetry” of doctor–patient interaction’, in D. Boden and D. Zimmerman (eds), Talk and Social Structure: Studies in Ethnomethodology and Conversational Analysis, Cambridge: Polity Press; pp. 138–63. ten Have, P. and Psathas, G. (eds) (1995), Situated Order: Studies in the Social Organization of Talk and Embodied Activities, Lanham MD: University Press of America and International Institute for Ethnomethodology and Conversation Analysis. Travers, M. (1997), The Reality of Law: Work and Talk in a Firm of Criminal Lawyers, Aldershot: Ashgate. Truzzi, M. (1983), ‘Sherlock Holmes: applied social psychologist’, in U. Eco and



T.A. Sebeck (eds), The Sign of Three: Dupin, Holmes, Peirce, Bloomington IN: Indiana University Press, pp. 81–118. Tuthill, H. (1994), Individualization: Principles and Procedures in Criminalistics, Salem OR: Lightning Powder. Van der Veer Martens, B. (2001), ‘Do citation systems represent theories of truth?’, Information Research . Vinkhuyzen, E. (1997), ‘Expert systems in practice: Two case studies’, unpublished PhD dissertation, Department of Psychology, University of Zurich. Waksler, F.C. and Psathas, G. (1989), ‘Selected books and articles about Erving Goffman and of related interest’, Human Studies 12, 177–81. Watson, D.R. (1983), ‘The presentation of victim and motive in discourse: The case of police interrogations and interviews’, Victimology 8, 31–52. Watson, D.R. (1990), ‘Some features of the elicitation of confessions in murder investigations’, in G. Psathas (ed.), Interaction Competence, Washington DC: International Institute for Ethnomethodology and Conversation Analysis & University Press of America; pp. 263–95. Watson, D.R. (1992), ‘The understanding of language use in everyday life: Is there a common ground?’, in G. Watson and R. Seiler (eds), Text in Context, London: Sage; pp. 1–19. Watson, D.R. (1994), ‘L’ethnométhodologie, l’analyse conversationelle et la recherche interculturelle’, in C. Labat and G. Vermes (eds), Cultures Ouvertes, Sociétés Interculturelles, Paris: L’Harmattan; pp. 119–128. Watson, D.R. (1997a), ‘Some general reflections on “categorization” and “sequence” in the analysis of conversation’, in S. Hester and P. Eglin (eds), Culture in Action: Studies in Membership Categorization Analysis, Washington DC: University Press of America. Watson, D.R. (1997b), ‘Ethnomethodology and textual analysis’, in D. Silverman (ed.), Qualitative Research: Theory, Method and Practice, London, Thousand Oaks CA and New Delhi: Sage; pp. 80–98. Watson, D.R. (2000), ‘The anthropology of communication: Foundations, futures and the analysis of “constructions” of space’, plenary paper presented at the conference ‘Social Constructions of Space: Territories of the Anthropology of Communication’, University of Liège, 18–19 November 2000. Watson, D.R. (2003), ‘The anthropology of communication: Foundations, futures and the analysis of “constructions” of space’, in T. Lask (ed.), Constructions sociales de l’espace: Les territoires de l’anthropologie de la communication, Liège: Les Editions de l’Université de Liège. Watson, G. (1984), ‘The social construction of boundaries between social and cultural anthropology in Britain and North America’, Journal of Anthropological Research 40, 351–66. Watson, G. and Seiler, R. (eds) (1992), Text in Context: Contributions to Ethnomethodology, Newbury Park CA, London and New Delhi: Sage. Whalen, J. (1995), ‘Expert systems vs. systems for experts’, in P. Thomas (ed.), Social and Interactional Dimensions of Human-Computer Interfaces, Cambridge: Cambridge University Press.


Orders of Ordinary Action

Whalen, J. and Vinkhuyzen, E. (2000), ‘Expert systems in (inter)action: Diagnosing document machine problems over the telephone’, in P. Luff, J. Hindmarsh and C. Heath (eds), Workplace Studies: Recovering Work Practice and Informing System Design, Cambridge (UK): Cambridge University Press. Whyte, W.F. (1955), Street Corner Society, Chicago and London: University of Chicago Press. Wieder, D. L. (1974), Language and Social Reality: The Case of the Convict Code, The Hague: Mouton. Wittgenstein, L. (1958), Philosophical Investigations, Oxford: Blackwell. Woolgar, S. (1987), ‘Reconstructing man and machine: A note on sociological critiques of cognitivism’, in W. Bijker, T. Hughes, and T. Pinch (eds), The Social Construction of Technological Systems, Cambridge MA: MIT Press; pp. 311–28. Yamauchi, Y., Whalen, J. and Bobrow, D.G. (2001), ‘Information use of service technicians in difficult cases’, CHI 2003: New Horizons, Proceedings, Conference on Human Factors in Computing Systems, pp. 81–8. Zeldin, T. (1998), Conversation, London: Harvill. Zimmerman, D.H. (1987), ‘Studies in ethnomethodology: Twenty years later’, Discourse Analysis Research Group Newsletter 3, 20–25. Zimmerman, D.H. and Pollner, M. (1971), ‘The everyday world as phenomenon’, in J.D. Douglas (ed.), Understanding Everyday Life, London: Routledge & Kegan Paul; pp. 80–103. Zussman, R. (1992), ‘The last bed’, in Intensive Care: Medical Ethics and the Medical Profession, Chicago and London: University of Chicago Press; pp. 186–218. Legal Cases Daubert v. Merrell Dow Pharmaceuticals, Inc. (1993) 509U.S.579, 125L.Ed. 2d 469, 113S. Ct.2786. Kumho Tire Co. v. Carmichael (1999), 526U.S.137, 143L.Ed. 2d238, 119S.Ct137. People v. Simpson (1995), No. BA 097211, WL69793 (Cal. Super. Ct. L.A. County), 12 May.

Index Aanestad, M. 52 Activities 3–6, 8–9, 21, 28, 33, 37, 39, 51, 53, 57–8, 60–2, 66, 69–70, 73, 86, 92, 108, 112, 123 collaborative 52, 66 of daily life 34 decision making 93 distributed 66 expert 66 interactional 64–5 of lecturing 100 musical 7 practical 92–3, 99, 102, 176 readers 188 ad hoc practices 103, 112, 116, 182 reasoning 109 adjacency pairs 64 annotation (as interaction through artifacts) 11, 176, 186–93 Anspach, R. 70, 72 anthropology 23 philosophical 159 of reasoning 10, 121, 131, 133 artificial intelligence 135–6 (see ‘expert systems’) Atkinson, J. M. 109 Ball, M. 52 Becker, H. 7 Bell, D. G. 140, 151 Berg, M. 73 Bibliographies (as local orders) 92, 95, 97 Bittner, E. 94, 97–8, 102, 109 Bjelic, D. 94 Blanks, R. 178 Bobrow, D. G. 140, 150, 152 Body 22, 53, 58, 62, 135, 176, 200–1, 203, 206 of literature 99 of scientific knowledge 11, 197 surgeon’s 54 Bourgois, P. 101 Branaman, A. 97

Burke, K. 188, 192 Burns, S. 28, 109 Burns, T. 97 Button, G. 8, 51, 69, 175 Cano, V. 92 Carlin, A. 8, 92, 99, 102, 175 Case, D. O. 92 Casey, N. 53 Case reports 72–7 Categories 7, 9, 98, 109, 188, 200 (see membership categories) Causal texture 72–3 accounts 156 checkers, problems 131–3 chemistry, problems 121–3, 127–8 Cicourel, A. 93–4 Co-construction (of accounts) 75–81 Code switching 62–4 cognitive models 198 operations, 176 orientations 11, 196 principles, 198 propositions 204 science 24, 108, 191–2 Collins, J. M. 96 Collins, O 96 commonsense knowledge, understandings 50, 91, 110, 113, 136 commonsense/science distinction 108, 113 communication 51, 63, 163, 188 community 98, 136–7, 151–2, 175–6 computer 149 programmes 10, 135–7, 175, 188 systems 10, 136 Conference 163 case conference (in hospital) 9, 70, 72–6, 81, 84–6 (see also ‘videoconference’) constructivist theorizing 4, 11, 100, 114 debate with realism 111 studies of science 113 contexture 21, 23, 26–7, 180


Orders of Ordinary Action

Contu, A. 33–4 Conversation analysis 34, 96, 100 Coulter, J. 29, 94, 191 Courtrooms 10, 107, 109–10, 113, 118 Cozzens, S. E. 92 Crane, D. 92 Crime scene examiners 195 – 206 Criminal investigations 195–9 Crittenden, K. S. 108 Cronin, B. 93–4 Cuff, E. C. 190 Culture 3, 156 of reading 11 scholastic 160 ‘selling culture’ in organizations 36 shock 122 Davenport, T. 136 Dean, H. 99 Deconstruction of expert/scientific testimony 107, 113–4 of forensic artifacts 208 Derrida, J. 113 Dettmar, K. 98 Deviance 3, 93, 191 Dewe, M. 94 Diagnosis machine repair 10, 135–41, 144, 148–50, 154, 156 medical 9, 11, 80, 84, 178, 180 as situated action 192 Dingwall, R. 29 Discourse analysis 110 Division of labour in organizations 70 DNA profiles 197 Documentary method 178, 181, 184, 189 Documents 102, 109, 118, 201 Douglas, J. 93 Drew, P. 51, 97, 109 Dreyfus, H. 22 Driessen, J. 180, 182 Durkheim, E 4, 14–6 ‘Durkheim’s aphorism’ 4, 13–6, 21, 165 Eco, U. 198 Edmond, G. 113, 118 Eglin, P. 29 Embodied action/practices 108, 176, 182–4, 188, 191–2 embodied reflexivity of instructed actions 8, 19, 22–3, 26

Ende, J. 70 Erickson, F. 70 Ethnography 10, 118 Ethnomethodology 3–8, 10, 13–19, 22, 28, 33–5, 49–50, 107–10, 112–14, 117–8, 159, 164, 181 ‘ethnomethodology’s programme’ 4, 13, 95, 165 Evergeti, V. 99 Evidence 10–11, 34, 109–10, 112–3, 115, 117–8, 157, 177, 195, 197–8, 200–1, 205, 209 expert 114 forensic/ physical trace 114–5, 200 prosecution 114 scientific 113–4 evidence technicians 195 experience 9–10, 28, 34, 69–70, 74–5, 83, 86, 95, 116, 136–7, 150–1, 156–8, 166, 185, 192, 206 Experiments 10, 112, 122, 128, 130–1, 176, 192 conditions 192 experimental psychology 23 gestalt 22, 27–8 Expert systems 10, 135–7, 140 PRIDE system 140–50, 156–8 Facts 34, 76, 82, 109–10, 113–4, 150, 164, 198, 203 social facts 4, 14–15, 21, 26, 28, 96 family 84–5 of machines 154 resemblance 95 Farrell, E. P. 94 Fehr, B. J. 95 Feigenbaum, E. A. 135 Fingerprinting methods 197, 202, 205–9 First rendering theorem 8, Fischer, J. F. 201, 206 Forensic investigations 118, 195–6, 204 artefacts 11, 196–7, 204–5, science 11, 118, 196–8 formal analytic methods 14–5, 21 reasoning/theorizing 10, 20, 159, 172 sciences 14–5, 96 accounts/representations 162, 164, 166 Formulations 45, 72, 74, 79, 81, 129, 168–9, 189–90, 198 Foucault, M. 55, 97 Foundations 69, 97, 157 foundational issues 93–4, 103

Index Fox, N. J. 52 Francis, D. 29 Fuchs, S. 113 Fujimori, Y. 70 Gale, K. 99 Garfield, E. 92–3 Garfinkel, H. 4–9, 51, 69, 72, 86, 91–2, 94–5, 98–101, 107, 112, 117–9, 164, 176–8, 182, 186, 192 Geertz, C. 164–5 Gendlin, E. 165 Gender 73, 85 Gestalt phenomena 23, 27–8 (see also ‘contexture’) figure-ground relationship 184 psychology 18–9 Giddens, A. 35 Glaser, B. 94 Glossing practices 94 Goffman, E. 96–7 Goodwin, C. 60–1, 175–6, 184, 188–91, 193 Gould, J. 131 Grammar 63, 169, 190 Grommes, P. 52 Group interviews 137 Groups 57–8, 61, 64, 156, 202 of crime scene examiners 199 of jigsaw pieces 124–6 occupational 196 PARC AI research group 136, 150 workgroups 137, 140, 151, 201 Gurwitsch, A. 22, 27–8, 180 Halford, M. 70 Hanseth, O. 52 Harper, R. 3, 36 Hartland, J. 136 Hartswood, M. 11, 188 Hatch, D. 98 Heap, J. 97 Heath, C. 29, 51, 53, 56 Held, D. 35 Heritage, J. 29, 59, 101 Hester, S. 29 Hierarchy 35, 50, 141 Higgins, G. M. 92 Hill, R. J. 108 Hindmarsh, J. 51–2 Hirschauer, S. 52, 60


Hirst, P. H. 91 History 15, 107, 121, 135–6, 141, 154 of art 222 of ideas 97 medical (of patients) 73, 78, 177, 180 of music 7 Home Office reports 195 Hughes, J. 3, 29 Husserl, E. 26, 28, 159 Hutchins, E. 24, 136 Ideology 3, 184 Ikeya, N. 8, 70–1, 85 Incommensurability 118, 190 Indexical expressions 7, 100 Inferences 140 abductive 198 inductive 126 Instructions 4, 7, 23, 48, 59, 62, 72, 74–5, 112, 138–49, 154–6, 203 Interaction (social) (see also ‘Talk-ininteraction’) 5, 34, 37–9, 49–50, 52, 66, 97, 109, 115, 159, 178, 190, 202, 204 begging as 100 doctor–doctor 70 doctor–patient 4 through artifacts 188–9 with technology 136–7, 176 Interests 16–7, 107, 114 analytic 10 practical 10 Interrogation 107, 109, 113, 115–8, 197, 205 Interviews 136–7 Investigation 3–4, 9–11, 15–6, 18, 94–5, 100, 102, 107, 109–10, 113, 117–9., 159, 195, 197, 205, 210 crime scene 198 criminological 195, 198–9, 210 forensic 202 scientific 136 video-based 136 Jackson, P. 135 Jamieson, A. 198 Jasanoff, S. 113, 118 Jefferson, G 34 Jigsaw puzzles 123–8, 133 Judgements 5, 10 normative 118 Jules-Rosette, B. 96


Orders of Ordinary Action

Jury deliberations 108–9, 113–4, 119, 210 Trials 110 Kendon, A. 97 Kessler, M. M. 92 Kitsuse, J. 93 Kleer, J. de 141 Knorr-Cetina, K. 110 Knowledge (see also ‘textbook knowledge’) 9–11, 70, 86, 97, 107, 136, 150–1, 185, 189, 196, 201, 203, 205 commonsense 113 expert 114, 117, 135–6, 141 medical 75, 82–4, 86 practical management of 70, 72, 86 scientific 67, 197 sociology of 69, 114 stock of 70 Knowledge sharing system (Eureka) 137, 151–8 Kuhn, T. 123 Kundell, H. 176, 192 Language 110 everyday 107–8 natural 109, 184, 192 of phenomenology 6 of theorizing 4 Latour, B. 110–11, 113, 115 Lawyer’s questions 114–7 Lemert, C. 97 Levinas, E. 159 Lewis, D. 23–6 Liberman, K. 10, 29 Literatures 4–9, 11–12, 93, 96–9, 100–2, 176 of beggars and begging 100–1 and their EM alternates 4, 16–19, 95–6, 99, 102, 108 of formal analytic science 18 of gestalt psychology 22–3 of music 6–9 of sociology/social science 34, 94–5, 101 of western navigational science 26 Literature reviews 99 ‘a gap in the literature’ as a phenomenon 100–2 Livingston, E. 9–10, 28, 95, 100–1, 112, 124, 175–7, 182, 189 Local orders 3, 9–10, 12, 14–5, 52, 60, 96, 99, 159, 164, 207 locally accomplished meaning 7, 107 locally organized activities 4, 9, 61, 172

local ecology 57 local features of crime scenes 201 local manifestations of power 8 local problems 155, 157 Locard, E. 199–200, 202 Lomax, H. 53 Lynch, M. 9, 28, 61, 95–6, 98, 100–1, 112, 175–7, 180, 182–3, 188 MacAndrew, C. 99 Macbeth, D. 29, 52, 117 McCorduck, P. 135 McHoul, A. 29 McInnis, R. G. 92 Mahan, B. H. 121 Mammography 175–84 Managers 35 managers’ work 35–46 Manuals 11 of instruction 7, 199 repair 10, 137–40, 142, 148–50 Manzo, J. 109 Mathematics 28, 121–2, 124, 132, 203 Maynard, D. 29, 109 Medawar, P. 112 Meehan, J. 29, 109 Mental attitude 47 illness 101 processes 108 Mentalism 189, 192 mentalistic descriptions 188 Mercer, D. 113, 119 Merleau-Ponty, M. 22 Methods (members) 5, 11, 16, 18, 20–1, 42, 46, 97, 102, 137, 171, 176, 196–7, 206, 210 of bibliography 93, 102 for diagnosing machine problems 151 formal analytic 14, 21–2 of practical tyextual reasoning 99 of seamanship 26 of social science 6, 14, 34, 94, 109, 116 of team management 45 Missing whatness 7, 192 Misunderstandings 81 Mondada, L. 8, 29, 52–3, 66 Moral character 186 entitlements 135 force 190 philosophy 16 requirements 97

Index Moss, S. 178 Murtagh, G. 99 Naïve realism 162 (sse also realism) Natural language 109, 184, 192 Naturally occurring phenomena 202–3 Navigation (Polynesian) 23–7 western science of 24 Newcomb’s rule 203 Nickell, J. 201, 206 Nii, H. P. 135 Nordby, J. 198 Nordine, C. F. 176, 192 Normal appearances 186 case 75, 79, 178 crimes 109 persons 186 properties 205 science 123 Normal, natural trouble 103 Observation 33, 37–8, 40–1, 84, 93, 99–100, 102, 137, 144, 162, 164, 175, 180, 195 Okada, M 8, 70 Operations cognitive 176 surgical 12, 51 technical 8 Ordinary action (see ‘practical action’) Ordinary society 14–16, 18, 20, 22, 108 Organization of bibliographies 92, 95 business organizations 12 of disciplines 91 of everyday language 108 of musical production 6–7 of practical action 3–4 service organizations 10 of surgical work of talk-in-interaction 61, 66 Organizational work 8 perspectives 8 settings 8–9 Orr, J. 34, 137–8, 140, 143, 145, 151 Oteri, J. S. 114 Practical action/work 3–5, 8, 10, 15–6, 26, 35, 61, 92–3, 95, 99, 169, 176–80, 188–9, 191, 202 ethnomethodology as practical action 101


Practical reasoning 3–4, 8–12, 91–9, 101–3, 107–10, 152 Peneff, J. 52 Peyrot, M. 3 Phenomenology 6 Philosophical analysis 162 concepts 107 debate 10–11, 159–73 inquiries 163 tenets 167, 172 Philosophy 15–6, 22, 107, 119, 121, 152, 156, 159 Pilnick, A. 52 Plenum 20, 96 Pollner, M. 94, 109, 121, 232, 236 Pomerantz, A. 70, 109 Power 33–5, 48–50 instruction as a lever of 46–8 organizational 46–8 Prabha, C. 92 Procter, R. 11, 188 Professional autonomy 84 competence 183 dominance 5 ideology 184 theorizing 6, 20 vision 11, 66, 176, 184–6, 188–93 work 51 Psathas, G. 29, 96 Questions 16, 49, 94, 135–6 in courtrooms 114, 117 of/by doctors 4, 65, 75, 83, 177 of/by forensic examiners 197, 199 in machine repair 138, 140–1, 143–8, 156–7, 160 Randall, D. 36 Rawls, A. 13–4, 95 Realist/constructivist debate 111 Reality (of social facts) 4, 14–5, 21 social 33 as topic of philosophical debate 161–2, 164 Reasoning (see also ‘practical reasoning’) 74, 81–2, 121, 136, 140, 157, 189 an anthropology of 131–3 bibliographic 91 in checkers 131–3 in forensic science 198–9 formal, abstract 4, 157, 159, 172


Orders of Ordinary Action

in natural science 121–31 in Tibetan philosophical debate 159–73 Recipient design 62, 66 Record keeping 74 Relevance 93, 99, 118, 124–6, 188, 209 Relieu, M. 60 Rendering theorem first 8, 19, 21 second 8, 19 Renderings 26, 113, 135 Representations 11, 21, 102, 107–8, 111, 157, 159 Reviewers 92–3 Rietz, R. E. 91 Robillard, A. B. 99–100, 112 Rose, E. 164, 186 Rosen, A. C. 96 Rouncefield, M. 11, 36 Rules 5, 70, 98, 112, 118, 141, 159, 189, 203 Sacks, H. 3, 22, 34, 53, 92, 94–6, 98, 183, 185, 189 Schegloff, E. 34 Schutz, A. 69 Schwabe, A. 96 Schwartz, H. 190 Science and technology studies 110 Scientific discourse 35, 111 methods 107–8, 111–2, 195–8 reasoning 10, 121–31 in chemistry 121–3, 127–8 in physics 128–31 Secondary elaborations 201 Seiler, R. 69 Shaddish, W. R. 92 Sharrock, W. W. 8, 29, 69–70 ‘Shop talk’ 176, 192, 199 Silverman, D. 96 Simon, H. 135 Simpson, O. J. court case 114–7 Slack, R. 11 Smith, R. 113 Social constructionism (see also ‘realist/ constructivist debate’) 110–14 Social distribution of knowledge 69 Social science (see also ‘methods’) 5–6 Social structure 3–4, 48 Society 3, 18, 28, 33–4, 50, 109–10 consumer 36 members of 34, 69 ordinary 14–16, 18, 20, 22, 108

Spayth, H. 131 Stetson, J. 95 Stoller, R. 96 Strauss, A. 69 Studies of work (in EM) 19, 28, 50–1, 66, 100–2, 110 Suchman, L. 22, 28, 66, 136 Sudnow, D. 7, 22–3, 75, 92, 94, 100, 102, 109 Surgery 5, 16, 51, 53 Symbolic interactionism 96 Symes, D. 92 Talk-at-work studies (see also ‘shop talk’) 51–2, 66 Talk-in-interaction 9, 52, 61, 88, 176 Teamwork 51–2, 66 Technology 9–10, 12, 18, 26, 36, 51–2, 110, 113, 135–6, 152, 156, 202 Temporality 72, 97, 204 properties 26 temporal order 26, 73, 188, 191 Ten Have, P. 29, 70 Testing 138, 152, 198–9 of theory 108 Texts 7, 16, 18, 22–3, 72–3, 81, 91, 97, 116–7, 140, 152–4, 156, 163 textual accounts 11, 22, 28, 79 textual corpus 97 textual materials 98, 102–3 textual reasoning 91, 99 Textbook knowledge 75, 82, 84, 128 Textually incarnate networks 92 Theorizing 4, 6, 20, 22, 91–2, 99, 127 Toto, L. 198 Tradition 3, 6, 16, 51, 96, 98, 110 Transcription conventions 67 Travers, M. 109 Truzzi, M 198 Tuthill, H. 203 Typifications 109 Understandings (members) 7, 34–5, 42, 48, 50, 75, 81, 91, 109, 121, 146, 148–9, 154, 161–2, 167, 172, 206 analytic 40–1, 50, 66, 86 Unique adequacy 95, 102, 191, 201 Universal observer 17 phenomena 10, 121, 131 reason 132, 163, 165 truths 164 uniqueness (of snowflakes) 201

Index Versions (problem of) 190 Video recordings 35, 52, 100, 136–7, 160, 170, 197 conferencing 8–9, 51–66 video-based investigation 136 Vinkhuyzen, E. 10, 136, 145 Visual appearances 196–7 images 52–60 visual space of action 52–61, 66 Visually available phenomena 60–1, 96, 99, 175, 178 Waksler, F. 96 Wallis, M. 178 Ward, S. 113 Watson, D. R. 29, 52, 91–3, 98, 109, 175, 192 Watson, G. 69 Weekes, P. 29 Weinberg, M. 114 Whalen, J. 10, 136, 145, 150, 152 Whyte, W. F. 96 Widmer, J. 29 Wieder, D. L. 3, 17, 28, 96, 107, 118 Wiley, N. 99 Williams, B. 141 Williams, R. 11 Wilmott, H. 33–4 Wittgenstein, L. 107, 112 Woolgar, S. 110, 113, 135 Wootton, A. 97 Work academic 92–4 bibliographic 92, 96–7 of crime scene examiners 11, 195–9, 201, 204, 206, 209


ethnomethodological 7, 14, 17 of everyday life 21 interdisciplinary 91 legal 118 managerial 35 mathematical 28 medical 7, 11, 51–2, 63, 73, 83–3, 175–8, 182–4, 188–9, 191–2 surgeons’ 9, 53, 66 organizational 8–9, 36, 39, 41–3, 45–9 of philosophers 166, 172 of playing the piano 23 professional 61 of professional social science 16, 18, 92 of Service Technicians 10, 135–40, 142–52, 154–8 of steering a course 26 of system designers 141–2 Worldly events 118 interpretations 11, 182–6 matters 23, 164–5 phenomena 16, 183 practices 17 Writers of manuals 140–1 Wynne, B. 113 Xerox Corporation 10, 137, 141, 149, 151–2, 155 Palo Alto Research Centre (PARC) 10, 136–7, 140 Yamauchi, Y. 150 Zeldin, T. 96 Zimmerman, D. 29, 94, 97 Zussman, T. 96