TRUTH IN VIRTUE OF MEANING This page intentionally left blank Truth in Virtue of Meaning GILLIAN RUSSELL 1 1 Great Clarendon Street, Oxford Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide in Oxford New York Auckland Cape Town Dar es Salaam Hong Kong Karachi Kuala Lumpur Madrid Melbourne Mexico City Nairobi New Delhi Shanghai Taipei Toronto With offices in Argentina Austria Brazil Chile Czech Republic France Greece Guatemala Hungary Italy Japan Poland Portugal Singapore South Korea Switzerland Thailand Turkey Ukraine Vietnam Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries Published in the United States by Oxford University Press Inc., New York Gillian Russell 2008 The moral rights of the author have been asserted Database right Oxford University Press (maker) First published 2008 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, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this book in any other binding or cover and you must impose the same condition on any acquirer British Library Cataloguing in Publication Data Data available Library of Congress Cataloging in Publication Data Data available Typeset by Laserwords Private Limited, Chennai, India Printed in Great Britain on acid-free paper by Biddles Ltd., King’s Lynn, Norfolk ISBN 978–0–19–923219–2 13579108642 For my teachers This page intentionally left blank Contents Preface ix Acknowledgements xiii Introduction 1 I. THE POSITIVE VIEW 1. The ‘in virtue of’ Relation 29 1.1 The Two-Factor Argument 29 1.2 Disambiguating ‘in virtue of’ 32 1.3 Collapse into Necessity? 37 2. Meaning 43 2.1 The Language Myth 43 2.2 Kripke and Kaplan 47 2.3 Truth in Virtue of Reference Determiner 52 2.4 Examples of Analytic Truths 57 2.5 Two Objections and a Serious Problem 66 3. Beyond Modality 71 3.1 The Problem 71 3.2 Semantics and Modality 72 3.3 Strict Truth in Virtue of Meaning 82 3.4 The Definition of Analyticity 99 3.5 Examples 104 Appendix A. The Formal System 109 A.1 Preliminaries 109 A.2 The Language 117 A.3 Semantics 119 A.4 Content, Validity and Reference Determiners 121 A.5 Some Theorems 122 viii Contents II. A DEFENSE 4. The Spectre of ‘‘Two Dogmas’’ 129 4.1 The Circularity Objection 129 4.2 The Argument from Confirmation Holism 135 5. Definitions 143 5.1 The Transience of Definition 144 5.2 Rethinking Definitions 146 5.3 Definitions as Postulates 157 5.4 Conventions of Notational Abbreviation 159 6. More arguments against analyticity 163 6.1 The Regress Argument 163 6.2 The Indeterminacy of Translation 170 6.3 Two Arguments from Externalism 175 6.4 An Argument from Vagueness 177 6.5 Blue Gold, Robot Cats 180 III. WORK FOR EPISTEMOLOGISTS 7. Analytic Justification 195 7.1 A Priori Justification 196 7.2 Analytic Justification 198 7.3 Theory 1: Naive Analytic Justification 199 7.4 Theory 2: Nihilism about Analytic Justification 202 7.5 The Problem of Semantic Competence 203 7.6 An Alternative Basis for Analytic Justification 207 7.7 Theory 3: Analytic Justification 209 7.8 Some Consequences 215 Bibliography 223 Index 229 Preface Sometimes it seems as if the debate over the analytic/synthetic distinction didn’t get resolved, so much as left behind. At its zenith, the tussle was between the Carnapians and the Quineans. For those in the Carnapian camp, the distinction was a consequence of some apparently obvious facts about language; it seemed amazing that the Quineans couldn’t see that. For Quine and his followers, the disappearance of the distinction was a consequence of the drive to import scientific rigour into philosophy; some things that had seemed fine—or even obvious—to the naked eyes of Kant, Bolzano and even Frege, dissolved under the microscope of precision; the world, at base, wasn’t quite as it seemed, and among the mere illusions were the ‘‘obvious facts” supposed to support the analytic/synthetic distinction. Of course, this is too simple a sketch, but I think it will do for the preface. The view of language that makes the existence of analytic sentences seem inevitable is a very intuitive one: expressions (both sentences and subsentential expressions like words), have meanings and an expression’s meaning plays three roles; (i) it is what a speaker has to grasp or know in order to count as understanding the expression; (ii) it determines what objects in the world the expression applies to (less colloquially: it determines a function from possible worlds to extensions); (iii) it is what the expression says or what it contributes to what is said (a proposition) by a sentence which contains it. That’s all but, with this picture in the background, the analyt- ic/synthetic distinction emerges very naturally. Suppose we stipulate that a new word tove is to mean the same as a word already in our language, say, cluster. Given the assumptions above, the sentence all toves are clusters seems to have some special properties. First, tove and cluster, having the same meaning, must pick out the same objects in the world, by (ii), which, given the meanings of the rest of the expressions in the sentence, is enough to guarantee its truth—we might call it true in virtue of meaning. Now by (iii) the sentence’s meaning is the proposition it expresses. Since that is such that it cannot be false, the sentence expresses a necessary proposition. Finally, since its truth follows from the mean- ings of the expressions it contains, and by (i) a competent speaker has to know those meanings, we might expect a competent speaker to be in a x Preface position to work out that the sentence is true, even if he has no particular experience of the world. Sentences with these three properties—truth in virtue of meaning, necessity and something like a priority—seem special, and it isn’t unnatural to mark that specialness by calling them analytic. The Quinean camp raised a lot of problems for this picture of analyticity but in the meantime, the ‘‘obvious’’ picture of meaning that supported it started to slip for relatively independent reasons. In three astonishingly influential pieces of philosophical writing, Putnam (1985[1973]) argued that meaning couldn’t be both what a speaker grasped and what determined extension, Kaplan (1989b)arguedthat what determines extension (character) and what got contributed to what a sentence said (content) came apart in the cases of indexicals and demonstratives, and Kripke (1980) argued that what determined the extension of a name or natural kind term need not be known in order for a speaker to understand the expression, nor was it what was contributed to the proposition expressed by a sentence containing one. Each was suggesting that the roles attributed to a single thing—the expression’s meaning —in the picture above, can be played by distinct things. If that is right, then the expression meaning requires some disam- biguation, and in this book I use the following terminology for that purpose: • character: what a speaker must know in order to understand an expression • content: what an expression contributes to the proposition semanti- cally expressed by a sentence containing it • reference determiner: a condition which an object must meet in order to be the referent of, or fall in the extension of, an expression With this terminology it is still possible to express the old view, on which character, content and reference determiner are all names for the same thing, but it is harder to pass that view off as inevitable, since it is easier to express views that run counter to it. Given that the old view provided intuitive support for the traditional conception of analyticity, perhaps it is unsurprising that Putnam, Kaplan and Kripke each produced examples of sentences that aren’t easily classified as either analytic or synthetic on the old conception: Putnam can be construed as arguing that the putatively analytic all cats are animals is true in virtue of meaning if it means what we think it means, but (since we don’t know for sure that it means that) it is not a priori and not necessarily true, and hence not analytic. Kaplan’s Iamherenow Preface xi seems analytic, but it is not necessary, and Hesperus is Phosphorus seems true in virtue of meaning in some sense (and so not synthetic), but not a priori (and so not analytic). With the obviousness of the picture supporting the distinction gone, and the extension of the distinction newly unclear, the analytic/synthetic distinction might seem to be based on substantial—and questionable—assumptions after all. One might even be tempted to think that it was an artifact of a view of language that we have left behind. My aim with this book is to tempt you with something else. As I like to think of it, our old view of analyticity was based on a naive theory of meaning, and some Quinean challenges to it were basically right. But our new theories of meaning will support a new picture of analyticity,one which—being based on a better theory of meaning—admits of a more robust defence.