THE VARIETIEOSF REFERENCE GareEtvha ns EDITED BY JOHN McDOWELL CLARENDPORNE S·OS X FORD OXFORUDN IVERSPIRTEYS· NS E WY ORK O,iford University Press, Walton Street, O,iford OX2 6DP London Glasgow New York Toronto Delhi Bombay Calcutta Madras Karachi Kuala Lumpur Singapore Hong Kong Tokyo Nairobi Dar es Salaam Cape Town Melbourne Auckland and associates in Beirut Berlin Ibadan Mexico City Nicosia Published in the United States by Oxford University Press, New York © Antonia Phillips 1y82 All rights reserved. No part of this publication may be reproduced, stored in a retrieval .rystem, or transmitted, in a'!Y form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior permission of Oxford University Press British Library Cataloguing in Publication Data Evans, Gareth The varieties ofreference. 1. Reference (Philosophy) I. Title II. McDowell,John 149'.946 B105.R25 ISBN o-19-824685-4 ISBN o-19-824686-2 Pbk Library of Congress Cataloging in Publication Data Evans, Gareth. The varieties of reference. Bibliography: p. Includes index. 1. Reference (Philosophy) I. McDowell,John Henry. II. Title. B105.R25E9 1982 160 82-8052 ISBN o-19-824685-4 AACR2 ISBN o-19-824686-2 (pbk.) Typeset by Butler & Tanner Ltd, Frome and London Printed in Great Britain at the University Press, Oxford by Eric Buckl ey Printer to the University Editor's Preface Gareth Evans, who died, aged 34, in August 1980, had been working for years on a book about reference. But he was con­ stantly rethinking both his ideas on the subject and his strategies of exposition; each successive draft is not so much a polishing of its predecessors as a first version of a substantially new work. In his last months he undertook an attempt to prepare his book for publication, and he managed to write or dictate new versions of the Introduction and chapters 1, 2, and (in part) 3. For the rest of the book he left drafts from various earlier dates (very occa­ sionally only in note form); these were more or less heavily annotated, with indications of footnotes, intended additions, and criticisms of the material as it stood. In the case of chapter 7, and, to a lesser extent, chapter 6, the later material is more substantial: in the Trinity Term of 1980 Evans offered a course of graduate classes on self-identification and self-reference, and in his preparation for this he arrived at improved formulations of many of the arguments of those chapters, and some thoughts that were altogether new. It is clear that he intended a radical revision of those two chapters. It would have been possible simply to transcribe Evans's words as they stood. But even with a quite extensive commen­ tary, the result would have been very difficult reading, except perhaps for people already partly familiar with Evans's ideas through hearing his lectures. It seemed clear that the overriding aim of this publication should be to make his thoughts as ac­ cessible as possible, and that this aim would not be best served by an excessively reverent approach to the draft. Accordingly, where his intention seems clearly expressed in notes, I have worked them up into prose. Similarly, where there is a sub­ sequent expression of dissatisfaction with the original draft, and amendment is possible without disruption to the flow of the argument, I have simply recast the material to reflect the later VI Editor's Preface view. In chapters 6 and 7, I have added extensively to the basic drafts, and sometimes replaced parts of them, with material taken from. the lectures. (Much of 4.3 is from this source too, replacing a version Evans was dissatisfied with.) Throughout­ bearing in mind that what we have is essentially a first draft­ I have recast some sentences and paragraphs in the interest of clarity. When it seemed to help the exposition, I have incorpor­ ated material from lecture-notes or earlier drafts into the text or the footnotes. I have corrected anything that seemed to be a mere slip. In all this, it seemed best to effect changes without remarking on them, rather than to burden the text with a complex apparatus of brackets and editorial footnotes. The Appendices contain material which for various reasons resisted this silent incorporation into the book. The Appendix to chapter 3 stands in for a planned final section for which there are only brief notes; although the material is all taken from writings by Evans, its conformity to his plan for the section is to some extent a matter of guesswork, and it seemed best not to pretend otherwise by including it in the body of the chapter. In assimilating material into chapters 6 and 7, I have adhered (as everywhere) to Evans's own articulation of the chapters into sections: the Appendices to those chapters preserve some rele­ vant material which does not readily fit into those frames. Elsewhere in the Appendices, I have tried to convey the charac­ ter of some doubts and afterthoughts which could not be dealt with by minor emendations of the chapters. The substance of all the Appendices, and most of the writing, are due to Evans. But in some places he clearly has to figure in the third person rather than the first; and I have adopted the convention that that is so throughout the Appendices. Square brackets round footnotes and parts of footnotes, elsewhere in the book, mark off either references to the Appendices or comments of mine on the arguments. The latest draft contains no indication of an intended title. However, in notes for a lecture course on the theory ofreference, Evans remarked that whereas some years previously he would have been tempted to call such a course 'The Essence of Refer­ ence', now he would prefer to call it 'The Varieties of Reference'; and this idea appears also in some notes for a Preface to an earlier version of the book. What he meant in the lectures was Editor's Preface vu probably connected with his having become convinced that 'descriptive names' are a perfectly good category of referring expressions. Earlier he would have insisted that al} genuine singular reference is, in the terminology of this book, Russellian. Now that struck him as unwarrantedly essentialistic: a theoreti­ cally well founded conception of genuine singular terms could embrace both Russellian and non-Russellian varieties. But the title is appropriate to this book in another way as well. It fits the conviction-manifested especially in chapters 6, 7, and 8- that a simple general theory even of one of those two varieties, the Russellian, is no substitute for a detailed and specific inves­ tigation of how each of its sub-varieties functions. In his final attempt on the book, Evans contracted its plan, in the hope of being able to work through the whole in the time that was left to him. Earlier plans had involved a more complex treatment of the different kinds of demonstrative expressions, to be interwoven with the discussions of different modes of identi­ fication which form the bulk of the present Part Two. And he had intended to reinforce the chapter on proper names with a partly parallel chapter on natural-kind terms; and to discuss testimony at much greater length. Even discounting the con­ traction, moreover, this volume obviously constitutes at best a remote approximation to what would have resulted from a complete revision by its author. Nevertheless, I believe that the brilliance and depth of his thinking about reference emerge from these pages with sufficient clarity to make generally ac­ cessible a lively appreciation of how much philosophy has lost by his early death. Acknowledgements I am very grateful to Antonia Phillips for entrusting to me the task of preparing this book for publication, and for much help and encouragement in its execution. I have to thank the Master and Fellows of University College, Oxford, for granting me a leave of absence to work on the book. I am most grateful to the British Academy for a grant, originally to Evans and then to me, which made the work possible; thanks are due to Richard Wollheim for his good offices in securing this beneficence. I have J. had help and advice from Ronald Dworkin, A. Gray, I. L. Humberstone, Christopher Peacocke, Galen Strawson, P. F. Strawson, and David Wiggins. Andrea McDowell prepared a beautiful typescript from what became a very messy original, and helped in many other ways too. I should like to acknow­ ledge also the helpfulness of the staff of the Oxford University Press. Chapter I draws on Evans's article 'Understanding Demon­ stratives', in Herman Parret and Jacques Bouveresse, eds., Meaning and Understanding (De Gruyter, Berlin and New York, 1981 ), pp. 280-303. He seems also to have intended some of this material to appear in chapter 6, and I have incorporated the relevant part of the article, slightly amended, into the Appendix to that chapter. Chapter 2 draws on Evans's article 'Reference and Contingency', The Monist lxii (1979), 161-89. Thanks are due to the original editors and publishers for permission to use this material. J.McD. Many people have helped to bring this book to publication, with advice, practical assistance and encouragement. I want to take this opportunity to express gratitude to all of them, on Gareth's behalf as well as my own: Katherine Backhouse, Ar­ nold Cragg, Ronald Dworkin, Andrea McDowell, David Pears, X Acknowledgements Galen Strawson, Sir Peter Strawson, David Wiggins and Richard Wollheim; Peter Brown and the British Academy for their speedy and flexible support; University College, Oxford, for the use of its photocopying facilities and for releasing John McDowell from his teaching duties; and, of course, John McDowell himself. Antonia Phillips Contents Introduction PART ONE: Historical Preliminaries 5 Chapter 1: FREGE 7 1.1 Introductory 7 1.2 Meaning (Bedeutung) 8 1.3 Empty singular terms: preliminary remarks 10 I .4 Sense: preliminary remarks 14 1 .5 Sense and thought 18 1.6 Empty singular terms: sense without Meaning? 22 1. 7 Empty singular terms: sense without referent 30 1.8 lnterpretational semantics and truth theories 33 1. g Conclusions 38 Chapter 2: RUSSELL 42 2.1 Introductory: Russell's criterion 42 2.2 Radical reference-failure 44 2.3 Russellian singular terms and descriptive names 46 2.4 Definite descriptions 51 2.5 'Rigid designation' and Fregean sense 60 Chapter 3: RECENT WORK 64 3.1 Going beyond Russell: singular thoughts 64 3.2 Russellian sayings: the two strategies 67 3.3 Kripke: singular thought without discriminating knowledge? 73 3.4 The Photograph Model 76 Appendix 80 Xll Contents PART TWO: Thought Chapter 4: RUSSELL'S PRINCIPLE 89 4. 1 Its meaning and importance 89 4.2 Verificationism and ideal verificationism 93 4.3 The Generality Constraint 100 4.4 The fundamental level of thought 105 4.5 Comparison with verificationism 112 4.6 The counter-examples 1 14 Chapter 5: INFORMATION, BELIEF, AND THOUGHT 121 5. 1 Information-based thoughts: introductory 121 5.2 The informational system 122 5.3 Interpretation and psychological attributions 129 5.4 The risk ofill-groundedness 132 5.5 Preview 135 Appendix 138 Chapter 6: DEMONSTRATIVE IDENTIFICATION 143 6. 1 Demonstrative identification and perception 143 6.2 Information-links are not sufficient 145 6.3 Egocentric spatial thinking: 'here' 151 6.4 Demonstrative identification of material objects 170 6.5 Some consequences 176 6.6 Immunity to error through misidentification 179 Appendix 192 Chapter 7: SELF-IDENTIFICATION 205 7.1 Introductory 205 7.2 Immunity to error through misidentification 215 7.3 Bodily self-ascription 220 7 .4 Mental self-ascription 224 7.5 Memory 225 7 .6 The possibility of reference-failure 249 7. 7 Conclusions 2 55 Appendix 258 Chapter 8: RECOGNITION-BASED IDENTIFICA- TION 267 8. 1 Introductory 267