Meaning, Truth, and the Limits of Analysis: Ten Studies 9780198726173

Table of contents :
Cover
Title Page
Contents
Sources
Introduction
1. Meaning and Truth-Conditions: From Frege’s Grand Design to Davidson’s
2. Concept and Copula
3. Donald Davidson’s Account of Semantic Interpretation. How Comprehensive is it? ‘All’, ‘Some’, and ‘Most’
4. Names, Existence, and Contingency
5. Modes of Grammatical Combination, Adverbs, and the Case of Action-sentences
6. Three Moments in the Theory of Definition or Analysis: Its Possibility, Its Aim or Aims, and Its Limit or Terminus
7. Locke: ‘The Great Conduit’
8. Languages as Things in their Own Right
9. Peirce: Reflections on Inquiry and Truth arising from his Method for the Fixation of Belief
10. An Indefinibilist-cum-Substantivist Account of Truth and the Marks of Truth
Bibliography
Index

Citation preview

Meaning, Truth, and the Limits of Analysis

David Wiggins

MEANING, LIMITS

TRUTH, OF

AND

ANALYSIS

THE

MEANING, TRUTH, THE OF

AND

LIMITS

ANALYSIS TEN

DAVID

STUDIES

WIGGINS

UNIVERSITY

PRESS

OXFORD UNIVERSITY

PRESS

Great Clarendon Street, Oxford, 0X2 6Dp,

United Kingdom 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. Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries © David Wiggins 2022 The moral rights of the author have been asserted First Edition published in 2022 Impression: 1 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, by licence 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 work in any other form and you must impose this same condition on any acquirer Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY

10016, United States

of America

British Library Cataloguing in Publication Data Data available Library of Congress Control Number: 2021949449 ISBN DOI:

978—0—19—872617—3

10.1093/0s0/9780198726173.001.0001

Printed and bound in Great Britain by Clays Ltd, Elcograf S.p.A. Links to third party websites are provided by Oxford in good faith and for information only. Oxford disclaims any responsibility for the materials contained in any third party website referenced in this work.

Contents

Vil

Sources Introduction

Meaning and Truth-Conditions: From Frege’s Grand Design to Davidson’s Concept and Copula . Donald Davidson’s Account of Semantic Interpretation. How Comprehensive is it? ‘All’, ‘Some’, and ‘Most’

Names, Existence, and Contingency Modes of Grammatical

Combination,

Adverbs,

and the Case of Action-sentences

Three Moments in the Theory of Definition or Analysis: Its Possibility, Its Aim or Aims, and Its Limit or Terminus Locke: ‘The Great Conduit’

. Languages as Things in their Own Right Peirce: Reflections on Inquiry and Truth arising from his Method for the Fixation of Belief I10.

I18

129

An Indefinibilist-cum-Substantivist Account of Truth and the Marks of Truth

164

Bibliography

177

Index

189

Sources

Chapter 1 was first published in B. Hale and C. Wright eds Blackwell Companion to the Philosophy of Language, Oxford: Blackwell, 1994. Chapter 2 draws on ‘The Sense and Reference of Predicates: A Running Repair to Frege’s Doctrine and a Plea for the Copula’, Philosophical Quarterly

34 (1984), 311—28. Chapter 3 is based on ‘“Most” and “All”: Some Comments on a Familiar Programme, and on the Logical Form of Quantified Sentences’ in M. Platts ed. Reference, Truth and Reality, London: Routledge & Kegan Paul, 1980. Chapter 4 draws upon material from ‘Contingency, Identity, and de re and de dicto Necessity’ in Jonathan Dancy ed. Papers on Language and Logic: Proceedings of the 1979 Keele Conference on Language and Logic (Keele: Keele University Library, 1980); ‘“The Sense and Reference of Predicates: A Running Repair to Frege’s Doctrine and a Plea for the Copula’, Philosophical Quarterly 34 (1984), 311—28; The Kant—Frege—RussellView of Existence: A Rehabilitation of the Second Level View’ in W. S. Armstrong, D. Raffman, and N. Asher eds Modality, Morality and Belief: Essays for Ruth Barcan Marcus, Cambridge: Cambridge

University

Press,

1994;

and

‘Names,

Fictional

Names

and

“Really” Proceedings of the Aristotelian Society, Suppl.Vol. 73 (1999), 271-86. Chapter 6 draws upon material from ‘Three Moments in the Theory of Definition or Analysis: its Possibility, its Aim or Aims and its Limit or Terminus’, Proceedings of the Aristotelian Society, 107 (2006/7), 73—109.

Chapter 7 is based upon ‘Language: The Supplement, 12 April 1996, 15.

Great Conduit’, Times Literary

Chapter 8 is based upon ‘Languages as Social Objects’, Philosophy 72 (1997),

499—524. Chapter 9 1s based upon ‘Reflections on Enquiry and Truth arising from Peirce’s Method for the Fixation of Belief’ in Cheryl Misak ed. The

Viii

Cambridge

SOURCES

Companion

to Peirce, Cambridge

and New York: Cambridge

University Press, 2004.

Chapter 10 draws upon Chapter IV of Needs, Values, Truth (amended third edition Oxford University Press, 2002) and ‘An Indefinibilist-cumNormative View of Truth and the Marks of Truth’in R. Schantz ed. What is Truth?, Berlin and New York: Walter de Gruyter, 2002. Permission from the original publishers to draw upon these publications is gratefully acknowledged.

Introduction

This volume draws together work on topics to do with language, meaning, truth, and the limit of semantic analysis, written from 1980 to 2020. Each chapter draws upon previously published material, but that material has been revised, sometimes significantly, for republication here. The book opens with a summary but selective account of a century’s work in the philosophy of meaning, starting from a paragraph of Frege’s Grundgesetze (1893) as construed and transposed by Wittgenstein in his Tractatus Logico-Philosophicus. The chapter advances thence to the Wittgenstein of the 1920s, the Vienna Circle and verificationism, meaning as use (Wittgenstein, Blue and Brown Books), then to the proposals of H. P. Grice,]. L. Austin, Donald Davidson, and (refining Davidson) John McDowell and Richard Grandy. Against that background, the book engages first with the nuts and bolts of sentence construction: predicates and the copula, quantifiers, names, existence treated as a second-level predicate, adverbial modification and the

difficulties of Davidson’s account of adverbial modification. On a more general level, the last five chapters then treat of definition and (as dreamt of by Leibniz and others) the terminus of semantic analysis; the

idea of natural languages as real things with a history; the idea of truth conceived as correlative with inquiry (C. S. Peirce) and, finally, the properties we look for in truth itself—the marks, as Frege or Leibniz might have said, of the concept true. There follow abstracts of the chapters. Chapter 1,‘Meaning and Truth-Conditions: From Frege’s Grand Design to Davidson’s’, is a selective history of the philosophy of language onwards from Frege’s Grundgesetze der Arithmetik (1892), and a defence of the idea propounded there that to understand a sentence is to know the conditions under which it i1s true. This claim is distinguished from others which succeeded it—that to understand a statement is to know the method for its verification

Meaning, Truth, and the Limits of Analysis: Ten Studies. David Wiggins. Oxford University Press (2022). © David Wiggins. DOI: 10.1093/0s0/9780198726173.001.0001

X

INTRODUCTION

(Vienna Circle), to know its use (Wittgenstein in the Blue and Brown Books), or to know what belief it is intended to induce in a hearer (H. P. Grice). Not just any true equivalence in the form ‘s is true if and only if p’ will suffice for s’s meaning

that p. We

need

an equivalence

that is properly

founded in the semantic powers of the components of the sentence s. Such can be given either in Frege’s way or else in Davidson’s way, in reliance upon his post-Quinean idea of ‘radical interpretation’. That later approach was refined by John McDowell and Richard Grandy. [A reprint of 1996a—see Bibliography at the end of the book] Chapter 2, ‘Concept and Copula’, begins from the diagram that Frege drew for Husserl to show his (Frege’s) doctrines of sense and reference.The chapter explains and further develops the doctrines that the diagram illustrates, demonstrating their continuing usefulness and applicability, and suggests how a fair number of contemporary doctrines can be accommodated within Frege’s picture. Then the chapter seeks to show how the doubts and difhculties that Frege himself expressed concerning concept and object can be answered by a reading of the copula which assimilates it to what Alonzo Church calls an ‘improper symbol’. The chapter ends with a suggestion, following on from Frege’s idea of a concept, about the best way of understanding the idea of a property—not least its use in evolutionary theory as Elliott Sober, among others, construes that.

[A reprint of what remains of 1984, but reaching different conclusions.] Chapter 3,°‘Donald Davidson’s Account of Semantic Interpretation. How Comprehensive is it? “All”,“Some”, and “Most’’. In a remarkable series of

papers Donald Davidson offered a would-be general account of how the sentences of a language can have the values of truth or falsity and thereby attain meaning. Some of these sentences, moreover, can be singled out (he says) as ‘structurally valid’. One major part of this account relates to the quantifiers ‘some’ and ‘all’. Unluckily for any larger aspirations Davidson may have had to generality, the treatment he proposes canot be applied to other quantifiers such as ‘most’ and ‘few’. These invite another approach, but that presupposes a basic

semantical form—a binary form—entirely different from that upon which Davidson based his own proposals. Inevitably questions arise here about the very idea of a universal grammar applicable somehow to all human thought or speech.

INTRODUCTION

X1

[Based on 19802, but detached now from a larger exposition, no longer necessary, of Davidson’s semantical programme.] Chapter 4, ‘Names, Existence, and Contingency’. Timothy Williamson

has offered a proof that any object that exists exists necessarily. Williamson’s conclusion depended crucially—as the Ontological Argument also depended—on the supposition that ‘exists’ stands for a property of objects. According to Frege and Russell, ordinary grammar misleads us here. What we need in order to understand existence is not a first-level notion such as eat or growl but the second-level notion of the instantiation of first-level concepts. Among first-level concepts are individual concepts. Deploying these, the chapter shows how proper names can be fitted within the Russell/Frege account. It also offers a treatment of the problems presented by true sentences such as “Vulcan does not exist’ or ‘Nausicaa does not exist’. These problems are best solved by deploying Frege’s idea that empty names depend for their sense upon the as if, and by reformulating in entirely natural ways the logical principles of existential generalization and universal 1nstantiation. [Based now upon 1980c¢, 1984, 19942, 1999b, and 2003a.] Chapter s, ‘Modes of Grammatical

Combination,

Adverbs, and the Case

of Action-sentences’. Davidson proposed that we should see a sentence such as ‘Shem kicked Shaun on the shin with a hobnail boot’ as having a logicogrammatical form tantamount to this: (17A) ‘there 1s an event e that consisted of Shem kicking Shaun e and e

was with a hobnail boot and e was on the shin’. But how can an event be with a hobnail boot or on the shin? I argue now for a less problematic alternative: namely, with Jennifer Hornsby, to see action-sentences as relating to acts—the act of kicking, say, or kicking with a hobnail boot. On these terms, an action is something there is when an act has been done—in this way or that. The chapter concludes with a discussion of the claim that ‘surface structure’ can easily mislead us. So often what philosophers want to call surface structure represents nothing more than a mistaken reading of the sentence in front of us. [Setting out from a paper of 1985—6, the chapter now arrives at a completely different conclusion.]

X11

INTRODUCTION

Chapter 6, ‘“Three Moments in the Theory of Definition or Analysis: Its Possibility, Its Aim

or Aims, and Its Limit or Terminus’. The first moment

discussed in this chapter i1s marked by Frege’s (1894) review of Husserl’s Philosophy of Arithmetic (vol. 1), discussions of the paradox of analysis, and Frege’s division—reaching beyond his distinction of Sinn/sense from Bedeutung/reference—between two conceptions of definition. Six familiar and different ideas of definition are then set out. The second moment is Leibniz’s still serviceable demarcation between the clarity (practical effectiveness) of ideas, and their distinctness, our explicit understanding of the various marks (components) of the concept that the idea relates to. Within Leibniz’s scheme, the pursuit of distinctness leads to a third moment: his demand for adequacy (the enumeration of the marks of a concept, and the marks of these marks...) and Leibniz’s speculations concerning the limit or terminus of analysis, beyond which there can be no greater or more complete distinctness. These problematic speculations are not entirely alien to the Zeitgeist of our own epoch. But so often the understanding of a concept depends not so much upon the unearthing of hidden constituents as upon grasping the concept’s overall shape, the shape that organizes its constituents.

[An extension and improvement of 2006—7.] Chapter 7, ‘Locke: “The Great Conduit”’. John Locke claimed that to abuse language is comparable to breaking or stopping the pipes whereby clean water is distributed to the public. Cultural conservatives will applaud Locke’s comparison. But linguists, regarding themselves as empirical scientists, will ridicule the snobbish prescriptions and proscriptions by which reactionaries seek to ‘fossilize’ the parlance of yesteryear. How well do the parties to this contest understand one another? Conservatives despise or ignore the linguist’s interest in the sprawling actualities of everyday usage and speech-patterns. More reasonably, though, conservatives also criticize the linguists’ particular brand of empiricism for refusing to recognize speaking and writing as an art—the art in which any language user can find a role for themself. [An enlargement of 1996.]

Chapter 8, ‘Languages as Things in their Own Right’. Chomsky has said that the notion of a public language is unknown to empirical inquiry and

INTRODUCTION

X111

‘raises what seem to be irresolvable problems’. This chapter takes an opposing view—namely the common-sense view which sees a language as something with a history, and the common possession as often as not of far-flung communities. It seeks to disarm arguments against this view. [1997a, with minor alterations.]

Chapter 9, ‘Peirce: Reflections on Inquiry and Truth arising from his Method for the Fixation of Beliet’, begins from Peirce’s essay ‘On the fixation of belief’ and shows how it leads outwards into a large body of Peircean claims concerning inquiry, the ‘secondness’ of experience, abduction, hypothesis, perception, and truth. It stresses the importance of Peirce’s insistence that, when any belief of ours is found to be determined by a circumstance extraneous to the facts, that belief is weakened or abandoned.

Only abduction, as Peirce characterized it, can expand our understanding from the known to the unknown. In conclusion the chapter offers a Peircean critique of Hume’s understanding of ‘the problem of induction’. Another conclusion is that Peircean pragmatism, properly construed, is consistent with a full-blooded realism. That which is distinctive in Peircean pragmatism is rather Peirce’s insistence

that truth and inquiry are correlative notions, intelligible only as correlative with the idea of experience. [A reprint and extension of 2004a] Chapter 10, ‘An Indefinibilist-cum-Substantivist Account of Truth and the Marks of Truth’. Philosophers and logicians have expended enormous effort upon the determination of the extension of the concept of truth. Remarkable discoveries have resulted.A different question is this: what are the marks of that concept? What is truth like? ‘Deflationists’ and their various allies will regard any such approach to truth as mistaken—and Tarski’s famous paper itself, some say, illustrates precisely this. Here, however, the deflationists misread Tarski, who saw himself as furthering the substantial account of truth offered by Kotarbinski. The chapter seeks then to enumerate the marks of truth. Truth is the first and foremost dimension of assessment for beliefs, statements, or utterances.

Secondly, if s is true then s will under certain convergence, not least among Thirdly, if s is true then s has content Fourth (a mark which is wide open

favourable circumstances command a those who want or need to know. which can be accepted or contested. to misunderstanding), every truth is

X1V

INTRODUCTION

true in virtue of something. Fifth, if two sentences are true then their conjunction will be true (as will be the disjunction). Finally, from Frege, ‘truth is the goal’—a claim pregnant surely with practical, even political import, or so history suggests. Witness the disasters that have been attendant upon its neglect or its loss. [Draws upon Chapter IV of Needs, Values, Truth, and 2002a.]

Bibliographies have been retained for Chapters 1, 6, and 10 in the hope that these will be useful to readers.

I Meaning and TruthConditions: From Frege’s

Grand Design to Davidson’s* Summary This chapter is a selective history of the philosophy of language onwards from Frege’s Grundgesetze der Arithmetik (1892), and a defence of the idea propounded there at I.32 and in Wittgenstein’s Tracatus Logico-Philosophicus that to understand a sentence is to know the conditions under which it is

true. This claim is distinguished from others which succeeded it—that to understand a statement is to know the method for its Circle), to know its use (Wittgenstein in the Blue and know what belief it is intended to induce in a hearer Not just any true equivalence in the form “s is true

verification (Vienna Brown Books), or to (H. P. Grice). if and only if p” will

suffice for s’s meaning that p. We need an equivalence that is properly founded in the semantic powers of the components of the sentence s. Such can be given either in Frege’s way or else in Davidson’s way, in reliance upon Davidson’s post-Quinean idea of ‘radical interpretation’. That later approach was refined by John McDowell and Richard Grandy, and several others.

1. However close it may have lain beneath the surface of some earlier speculations about language, the idea that to understand a sentence is to have grasped its truth-condition was first made explicit by Frege, for whom it was simply an unemphasized consequence of his general approach to questions of meaning. In the transition from logical positivism to modern analytical philosophy, the idea came near to being mislaid entirely. It was brought back into a new prominence in the late 1960s by Donald Davidson.

* From A Companion to the Philosophy of Language, Second Edition (2017). Edited by Bob Hale, Crispin Wright, and Alexander Miller.

Meaning, Truth, and the Limits of Analysis: Ten Studies. David Wiggins. Oxford University Press (2022). © David Wiggins. DOI: 10.1093/0s0/9780198726173.003.0001

2

MEANING

AND

TRUTH-CONDITIONS

Having rediscovered the idea for himself and in his own way, Davidson pressed its claims as a principle in the philosophy of mind and meaning, and as the only proper basis on which to conduct serious semantic investigations. In advance of considering these and more recent claims about meaning, it will be useful to mark certain moments in the formulation and reformulation of the original insight of the truth-conditional theory. In a historical framework, even the bare skeleton of one furnished here, truth-conditional

notions may be expected to transcend our more immediate sources of theory or doctrine concerning them as well as our more ephemeral disputations. 2. What i1s it for a declarative sentence to mean something, or have a sense? For Frege, to answer such a question was not, as it was later for Carnap or his inheritors, an all-important end in itself. Nor was answering that question a part of a comprehensive effort to arrive at a philosophical account of the relation of language to mind, as it 1s for Davidson and his inheritors. For Frege, it was a means, a propaedeutic for the understanding of the specific thing whose status and nature centrally concerned him, namely arithmetical judgment. Nevertheless, despite the special character of this original interest, Frege saw the question of the meaning of a declarative sentence as a general question, requiring not so much the introduction of a calculus ratiocinator (he said) as the creation of something more resembling a Leibnizian lingua characteristica. (“My intention was not to represent an abstract logic in formulas, but to express a content through written signs in a more precise and clear way.”') What Frege took the answer to his question to require was a general notion of meaning that could be correlative with the general idea of the understanding of a sentence. The conception he formed was of the Sinn or sense of a sentence that was to be understood thus or so, the sentence itself being seen as something built up by iterable modes of combination from component words, each of which had its own contributory sense. The senses of part and whole were to be such that the latter could be determined from the former (given an account of the

I. Frege,‘Uber den Zweck der Begriffsschrift’, lecture delivered at the meeting of 277 January, 1882 of the Jena Society for Medicine and Science, Jenaische Zeitschrift fiir Natunvissenschaft, 16/ Supplement (1882/1883), pp. 1-10. See ‘On the object of my concept-writing’ in T. W. Bynum (ed. and trans.), Conceptual Notation and Related Articles, Oxford: Clarendon Press, 1972. For the concept-writing itself, see Begriffsschrift, eine der arithmetischen nachgebildete Formelsprache des reinen Denkens. Halle:Verlag von Louis Nebert (1879), republished and translated inT. W. Bynum op.cit.

MEANING

AND

TRUTH-CONDITIONS

3

modes of grammatical combination involved in the construction of the sentence).

The culmination of Frege’s efforts under this head may be found in Volume 1, §32 of the Grundgesetze der Arithmetik,” where he declares that there is both sense and reference for every sentence of his ‘concept-writing’ or ‘ideography, his Begriffsschrift or concept-script. The Begriffsschrift is the constructed language whose operations are to shadow the workings of natural language and, in matters of difficulty such as the foundations of arithmetic, to regulate or supplant natural language. The reference of a sentence of Begriffsschrift is its truth-value, and the sense of the sentence is the thought that the sentence expresses. But how exactly does a thought attach to a sentence? And what is a thought? Well, which thought it is that a sentence expresses and how the thought attaches to the sentence will depend upon nothing other than this: under what conditions i1s the sentence to count as true? Or, as

Frege describes the matter for the artificial language he has just finished constructing: It is determined through our stipulations [for the linguistic expressions and devices comprising the language of Begriffsschrift] under what conditions |any sentence of Begriffsschrift] stands for the True. The sense of this name of a truthvalue, that is the thought, is the sense or thought that these conditions are fulfilled...The names [expressions|, whether simple or composite, of which the [sentence or| name of a truth-value is constituted contribute to the expres-

sion of a thought, and this contribution [of each constituent] is its sense. If a name [expression] is part of the name of a truth-value [i.e. is part of a sentence|, then the sense of the former, the name [expression], is part of the thought expressed by the latter [the sentence].

This statement comes at the end of Frege’s detailed explanations of Begriffsschrift. But its import is potentially perfectly general, and the stipulations of sense for the expressions of his invented language simulate what it is for the expressions of a natural language to have a given or actual (not merely stipulated) sense. The institution of the Begriffsschrift—the project Frege had begun in preparation for his books on the foundations of arithmetic

(1884; 1893) and published in part in 1879,” but then resumed and

2. See Grundgesetze der Arithmetik (1893). In this work, §32 and the preceding sections consolidate, codify, and complete the doctrines of (direct) sense and reference explored and expounded in ‘Uber Sinn und Bedeutung’, pp. 25—50. 3. See Begriffsschrift (1879, n. 1).

4

substantially

MEANING

corrected

AND

in the

TRUTH-CONDITIONS

work

of 1893, from

which

we

have just

quoted—at once illuminates natural language, albeit only in microcosm, and extends it. It illuminates it by displaying clearly the workings of a distinct language abstracted from natural language, namely the concept-script in which Frege hoped to make newly perspicuous all questions of “inferential sequence”. The purposes this serves are akin to the practical and theoretical purposes that the construction of an artificial hand with a specialized function might have for a community of beings whose normal members had natural hands with less specialized functions. 3. Given Frege’s concern with “a formula-language for pure [i.e., nonempirical] thought”, it is unsurprising that, as he said, he “confined [him]|self for the time being to expressing [within it] relations that are independent of the particular characteristics of objects” (Begriffsschrift, 1879, preface). Properties and relations that were not so independent registered in the Begriffsschrift only in the form of generality-indicating letters such as ® or W that prescinded from all particular content.* Nevertheless, Frege did envisage successive relaxations of this ordinance, and he spoke of possible extensions of his formula language to embrace the sciences of geometry, motion, mechanics, and so on. Given the universality and generality of the insights that originate with Frege, what we now have to envisage is a further extension of Begriffsschrift, namely the extension which, for purposes rather different from Frege’s, will even furnish it with the counterpart of such ordinary sentences as “the sun is behind cloud” (say). In the very long run, the extended Begriffsschrift (Bgt) might itself be modified yet further, to approximate even more closely to the state of some natural language. In the interim, however, in the transition from Frege’s to our own purposes, it stands as an illustrative model

of that which will In an extension behind cloud” will particular thought sentence (for it to

be much more complicated. such as we are to imagine, a sentence like “the sun is have a sense if and only if it expresses a thought. For the that the sun is behind cloud to attach to this English attach to such a social artefact as this, produced and held

fast in its temporal, historic, and social setting, Frege need not forbid us to say) will be for the sentence

to be so placed in its total (historical and

customary-cum-linguistic) context that it stands [in some situation] for the

4. In view of the confusion surrounding this mathematical term, Frege did not call them ‘variables’.

MEANING

AND

TRUTH-CONDITIONS

True just in case [in that situation]

the sun is behind

5

cloud. Putting the

matter in a way that is not Frege’s, one is tempted to say that he who understands the sentence is party to a practice that makes this the circumstance under which the sentence counts as true. What mystery remains about what a thought is? The thought expressed by a sentence is expressed by it in virtue of ordinary linguistic practices (the practices that we have imagined will be encapsulated in the definitions or elucidations of the empirical terms to be introduced into the extended Begriffsschrift), which expose the sentence to reality, and its author to the hazard of being wrong, in one way rather than another way. Once you know that, you know what the thought is that the sentence expresses. 4. The truth-conditional thesis, so seen, can be detached from more ques-

tionable features of Frege’s semantical doctrine, such as the idea that a sentence is a complex sign standing for objects called the True or the False or is a name of a truth-value. Wittgenstein does detach it (an act of retrieval for which he is too rarely commended) in Tractatus Logico-Philosophicus (1921):° 4.022

A sentence in use (Satz) shows how things stand if it is true. And it says that

4.024 4.002

To understand a sentence in use means to know what is the case if it is true. A sentence in use is true if we use it to say that things stand in a certain way,

they do so stand.

and they do.

These are striking formulations, more general than Frege’s and not radically dependent upon Wittgenstein’s picture theory of meaning. But now it seems we must attend to a problem that neither Frege nor Wittgenstein

s. Tractatus Logico-Philosophicus (1921). 1 translate Satz here not as ‘proposition’ but as ‘sentence in use’, in order to mark and preserve the continuity (as well as the discontinuity) with Frege, who

always used Satz to mean what we now mean by ‘sentence’. I think Wittgenstein effectively answers the complaint that Frege has nothing to say about what 1t is to understand a sentence or grasp a thought. For this complaint—justifiable enough, perhaps, when directed against such traditional accounts as the one given in Church (see §ogq of Introduction to Mathematical Logic, 1956)—see, e.g., Dennett, The Intentional Stance (1987, p. 123): “Frege does not tell us anything about

what

grasping

a thought

consists

in...”

In fact, it would

be

much

fairer to complain

against him (if one thinks this a matter for complaint) that, by introducing the thought as that which one grasps by virtue of grasping the acceptance/rejection conditions of something linguistic, Frege

may

seem

poised

to

acquiesce,

not

in a vacuous

Platonism

of noeta, but

in a

potentially highly controversial quasi-linguistic view of thinking as the soul’s internal dialogue with itself. Interestingly, this view really 1s Platonic: “The soul when it thinks is simply conversing with 1tself, asking itself questions and answering, affirming and denying...So I define one’s thinking as one’s speaking—and one’s thought as speech that one has had—not with someone else or aloud but 1n silence with oneself” (Plato, Theactetus 189F—190*). On matters, see now

this and cognate

Dummett, ‘The philosophy of thought and the philosophy of language’ (1986).

6

MEANING

AND

TRUTH-CONDITIONS

addressed explicitly. It is the problem (which still excites controversy in connection with Donald Davidson’s version of the truth-conditional view of meaning®) that not just any true equivalence in the form [s is true if and only if p] can sufhice to show that s actually means that p. Suppose that the sentence “the sun is behind cloud” is now true.Then all sorts of other things have now (as matters stand) to be the case. It is daytime, the sun has risen, it is not dark, more people are awake than asleep, and millions of automobiles are emitting smoke into the atmosphere, and so on— all this in addition to the sun’s being behind cloud. For these are the accompaniments, in the world as it is, of its being daytime and the sun’s having actually risen (to be obnubilated or not obnubilated). It is only to be expected, then, that, where s makes such a particular historical statement as

it does, in a manner dependent upon some historical context, any of these extra things may in that context be added salva veritate to the right-hand side of the biconditional “s is true if and only if the sun is behind cloud and...” (It 1s certain that, without detriment to the truth of s, any necessary truth or natural law can be added so.) It is only by virtue of knowing already what s means that one would pick on the “sun is behind cloud” conjunct, from out of the mass of things that also hold when the “sun is behind cloud” is true, to be the clause to give the proper truth-condition for s. It follows that, to put down what a given utterance of a sentence s means and impart its meaning to someone, we need to be in a position to signal some ‘intended’ or ‘privileged’ or ‘designated’ condition on which its truth depends. Only where ‘s is true iff p’ signals on its right-hand side an intended, privileged, or designated condition can we conclude from this biconditional’s obtaining that the utterance of s actually means that p. Look again at Wittgenstein’s 4.024. His “what is the case if it [the sentence] is true” presupposes the identification of that intended or designated condition. 5. One way to advance might be to recast Frege’s and Wittgenstein’s thesis as follows: Sentence s has as its use to say that p—or s means that p—just if whether s is

true or not depends specifically upon whether or not p.

But this is not really the end of the difficulty. For one of the things that the truth of “the sun is behind cloud” (as said at a given particular time and 6. For Davidson’s version, see ‘Truth and meaning’ (1967). And see below §18. See also P. E Strawson, ‘Meaning and truth’, inaugural lecture (1969); and Foster, ‘Meaning and truth theory’ (1976). See also Davidson’s ‘Reply to Foster’ (1976), on which see n. 29.

MEANING

AND

TRUTH-CONDITIONS

7

place) depends specifically upon, in one ordinary and standard sense of “depend”, might perhaps (at that time and place) be low atmospheric pressure plus the obtaining of other meteorological conditions. None of this, however, 1s what the sentence actually says. And for the same reason we cannot improve the formulation just given by ruling that the truth of the sentence has to depend only upon the designated condition. It cannot depend ‘only’ on that condition, in the ordinary sense of “depend”. For it will have to depend (in that ordinary sense) on everything that the satisfaction of the intended condition itself depends upon. 6. Consider now what Frege could have said in reply to this sort of difhiculty, pointing to things already done in Grundgesetze. Suppose (as before), that the language of his Begriffsschrift has been formally expanded to enable one to say “the sun is behind cloud” and all sorts of similar empirical things. Let us call the imaginary extension (Bg+). Each new primitive expression (‘sun’, ‘cloud’, etc.) will have had a reference stipulated for it in accordance with an empiricized extension of Frege’s canon for definitions (see Grundgesetze, 1893, I, §33). In each case, the sense of the new primitive expression will

consist in the fact that its reference is stipulated thus or so.” By virtue of this, it will have been contrived that the sense of any complex expression can be determined from its structure and from the referential stipulations governing each constituent expression. But now, in the light of all this, Frege is entitled to insist that, if we stick scrupulously to what actually flows from the full and appointed referential stipulations for all the individual expressions and devices of the extended Begriffsschrifft—let us call the set that consists of these stipulations ®@(Bg+)—, then we shall never arrive at an unwanted biconditional such as ‘the sentence “the sun is behind cloud on 25 June 1993”7 is true if and only if on 25 June 1993 the sun is behind cloud

and the sun has risen and there is low pressure and more people are awake than asleep and... (or its counterpart in Bg+). For the stipulations for the extended Begriffsschrift furnish no way to derive such a biconditional. The intended condition will be the particular condition that the appointed stipulations deliver. Not only that. In concert, these stipulations, which license nothing about low pressure as part of the truth-condition for s, will spell out the specific particular dependence that had to be at issue in the restatement of the Frege—Wittgenstein thesis. 7. Here I borrow an expository idea from Michael Dummett. See his Frege: Philosoply of Language, pp- 227-8.

8

MEANING

AND

TRUTH-CONDITIONS

No wonder (Frege’s philosophical champion may want to say) that we can hear ‘“the sun is behind cloud” is true if and only if the sun is behind cloud’ as more or less equivalent to “The truth of “the sun is behind cloud” semantically depends upon whether or not the sun is behind cloud’. For we can hear the biconditional ‘ “The sun is behind cloud” is true if and only if the sun is behind cloud’ as something delivered to us by whatever plays the part for English that the Fregean stipulations ®@(Bg+) will play for the extended Begriffsschrift. Where the turnstile “F" adjoined to ‘@ (Bg+)’ signifies that the Bg+ stipulations ® suffice to derive that which follows, what we are saying is, in effect, this: (X3

|s means in Bg+ that p| is equivalent to

(Bg+)

[True s if and only if p|.

There is nothing strange or scandalous in the suggestion that we hear the conditional as enclosed or nested in this way within a ‘turnstile’ operator “I_O(Bg+)” whose presence has to be understood. Countless conditionals we utter are intended by us to be understood as presupposing some norm or tendency that we could roughly identify but do not attempt to describe in the form of an explicit generalization. In so far as some residue of a philosophical problem still persists, the place to which it escapes is the characterization of the turnstile “”" and the general idea of a set of specifically referential specifications that imply this or that equivalence in the form [True s if and only if p]. The point that is left over, which we shall have to attend to in due course, is that, even though @(Bg+) would exemplify such a set, @(Bg+) could scarcely stand in for a general characterization of what a referential specification is. We need to understand “F- oL " for variable L. 7. The residual problem is philosophical—and serious. There is no immediate solution.We will return to the matter in {18 following, at a point decades later in our narrative. Meanwhile, let us consolidate the position now arrived at and pause here to show—if not in Frege’s symbolism (which continues to daunt typesetters and readers equally) or in strict accordance with every particular of Frege’s own view of predication®—how, more exactly and in more detail, the claim might be made good that Frege can pick out the particular sort of dependence which he needs to secure between the truth of a particular sentence s of a language L and the obtaining of some condition that p. Let us do so by giving the referential specification 8. For some discussion of these 1ssues, see my ‘On the sense and reference of predicate expressions’ (now Chapter 2 in this volume) with references there to V. Dudman and P. Sen.

MEANING

AND

TRUTH-CONDITIONS

)

of the semantics of a tiny sub-language L(1) of English that might be the counterpart of some small fragment of the extended Begriffsschrift. Suppose the constituent strings of L(1) are simply the following: (1)

The sun is behind cloud

(2)

Not (the sun is behind cloud), [which is said aloud as follows: the sun

is not behind cloud] (3)

The moon is behind cloud

(4)

Not

(the moon

is behind cloud), [which is said aloud as follows: the

moon is not behind cloud], together with all possible conjunctions

of (1), (2), (3), and (4).

Then we

can determine the sense of an arbitrary string of L(1) by the following provisions: Terms:

T(1) T(2)

“The sun” is a term and stands for the sun. “The moon”is a term and stands for the moon.

Predicates:

P(1)

“Behind cloud” is a predicate and stands for being behind

Connectives:

C(1)

“Not” is a unary connective: where A is a string of L, “not” + A is true if and only if A is not true.

C(2)

“And”is a binary connective: [A + “and” + B] is true if

cloud.

and only if A is true and B is true. Syncategorematic

“Is” is a syncategorematic expression, whose role it is to

expressions: Rule of truth:

signal the fundamental mode of combination R(1)

exemplified in R(1) below. A sentence that is of the form [t+ “is” + F], i.e., a sentence consisting of a term t, such as “the sun” or “the moon”,

followed by the syncategorematic expression “is”, followed by a predicate expression F such as “behind cloud”, is true if and only if what t stands for has what

F stands for.”

Now let us put these rules together and note their effect. Given the sentence [“the moon” + “is” + “behind cloud”] = [The moon is behind cloud], we can agree, by R(1), that the sentence is true if and only if what “the moon’ stands for is or has what “behind cloud” stands for, which last we can show to be true (see T(2) and P(1)) if and only if the moon is behind cloud.

That does not make news—no

more than news is made when, having

multiplied 13 by 25 and got 325, you then divide 325 by 13 and get 25. But

9. For the use of the relative pronoun ‘what’ in connection with the references of predicates, see Frege Posthumous Writings, 1979, p. 122.See also Dummett, Frege: Philosophy of Language, pp. 211—17.

10

MEANING

AND

TRUTH-CONDITIONS

it verifies something. Similarly, as we could point out here on Frege’s behalf, our semantic derivation helps verify something, namely that, so far as they go,T(1),T(2),P(1), C(1), C(2),and R(1) represent a correct reckoning of the semantic resources of L(1). What is achieved would have looked more impressive, no doubt, if L(1)

had been a fragment of Chinese or Arabic while our referential specification had been done in English. Such a specification is something we can more easily imagine someone’s failing to get right. There is no question, however, of a specification of this sort’s looking impressive—or its needing to do so (unless it solves neatly and correctly a known grammatical difficulty or casts some light, however indirect, on a real obscurity in the workings of the language under analysis). The specification simply leaves nothing to chance in the idea that, where s is an L(1) sentence, s means in L(1) that p if and only

if the biconditional [True s if and only if p] flows from @“("). Thus, for this particular sentence ‘s’ and that particular p, s = p. So s means that p in the context of Frege’s own particular purposes in the Grundgesetze der Arithmetik, let this serve as a model for the defence of what Frege wanted to say there about sentence-sense. For all he needed to claim at 1.32 of that work was his

complete control over the sense of a sentence in concept-script. There is no relevant doubt, either theoretical or practical, of that grasp. 8. In Tractatus 4.024 Wittgenstein is heir to Frege’s idea of sentence-sense,

and he tries to prescind from the particularities of Begriffsschrift in order to make a general claim. Nearby, at 4.002, he points to the need to bring real,

live speakers into the picture. Once we take their presence seriously, however, we shall notice for ourselves a new kind of difficulty—the first of several that will come in due course to occupy us. Consider the Latin sentence alea jacta est. Like its standard translation into English, the die is cast, the sentence is true if and only if a die [the die] has been thrown. This requires, among other things, that there be a real die and someone who has thrown it. But it is safe to say that what speakers have normally used the Latin or the English sentence to state or to intimate—to say in the full and ordinary sense of ‘say’—is nothing of that sort. The normal use of the sentence is to say the sort of thing that Julius Caesar said by alea jacta est when he broke the laws of the Roman Republic and, instead of disbanding his troops, led them towards Rome across the boundary marked by the river Rubicon.We who follow Caesar use the English sentence to assert that, in doing some act or other such as crossing that stream, we have committed ourselves irrevocably.

MEANING

AND

TRUTH-CONDITIONS

II

What 1s the difficulty here? The difficulty this creates for the Frege— Wittgenstein characterization of sense is that it shows that there is no simple route between the ordinary or normal use of a sentence such as “the die is cast” or “alea jacta est’—or from what people usually say by uttering it— and its strictly or narrowly linguistic meaning. The proper response to this problem is to concede something. We must adjust the Frege—Wittgenstein thesis to read as follows: Sentence s has as its use in L(i) to say literally (to say in the thinnest possible acceptation of ‘say’) that p—thus s means that p in the narrowest strictest sense of ‘means’—if and only if the referential specifications specific to the language L(i) [e.g., the sorts of specification given in 7] rule that whether s is true or not depends upon whether or not p.

This reformulation simply spells out an intention that Frege or Wittgenstein could have voiced. But what it suggests is that, in order to implement that intention, we have to embed our new formulation in some larger, more comprehensive theory, the sort of theory for which we have to look forward to the work of J. L. Austin.'® This can persevere in the Fregean explication of the literal meaning of a sentence as consisting in its sense or truth-condition. But the fuller kind of saying that we find in the the die is cast example is something that the comprehensive theory will have to explain by building upwards and outwards from literal meaning as characterized after the fashion of provisions like T(1), T(2), P(1), C(1), C(2), and

R(1). A neo-Austinian theory may suggest that, by doing the rhetic act of uttering something which has as its sense (and means literally) in language L(i) that the die has been thrown, and by performing thus the locutionary act of saying ‘the die is thrown’, a speaker can perform a further speech act, namely an illocutionary act, tantamount in force to the declaration or intimation that he is irrevocably committed. By saying one thing then (here a false thing) Caesar conveys something else, which proves to have been a true thing. 9. There is more to say about this, but here it only needs to be shown how one might place in a single focus the Frege—Wittgenstein conception of 10. For J. L. Austin’s theory of locutionary, illocutionary, and perlocutionary acts, see How to Do Things with Words (1962).A rhetic act 1s an act of using vocables with a contextually determinate sense and reference and 1n such a way that one can be reported as saying that...For the connection between the locutionary and the rhetic, for the connection between Austin’s researches and post-Austinian developments, and for much else besides that belongs in the areas I have so roughly blocked in, see Hornsby, “Things done with words’.

12

MEANING

AND

TRUTH-CONDITIONS

sense, in the condition in which it was available by 1921, and the different researches of J. L. Austin. These were undertaken some thirty years after the Tractatus, in a framework of theoretical expectations both at odds with the concerns of Grundgesetze and Tractatus and uninformed by attention to very much that these works had in common. But the reason to mention the difficulty here is that, unless we are prepared to use Austin’s work to delimit the area within which Frege, Russell, and the early Wittgenstein wanted to operate, their theories will be plagued with irrelevant objections.All will be well, however, provided that the theory of literal sense can be fitted into a larger framework that embraces among other things both the non-literal use of declarative sentences and the literal use of ordinary non-declarative sentences. '’ At this point, let us attend to a passage too rarely heeded as already expressive of Wittgenstein’s constant awareness of the importance of such a framework and of the wider frameworks that must contain this one. It occurs at Tractatus 4.002: Man possesses the ability to construct languages capable of expressing every sense, without having any idea how each word has meaning or what its meaning is—just as people speak without knowing how the individual sounds are produced. Everyday language is a part of the human organism and is no less complicated than it. It is not humanly possible to gather immediately from it what the logic of language is. Language disguises thought. So much so that, from the outward form of clothing it is impossible to infer the form of the thought beneath it, because the outward form of the clothing is not designed to reveal the form of the body, but for different purposes. The tacit conventions on which the understanding of everyday language depends are enormously complicated.

11. [f the inner core of a theory of sense for a given language 1s stated truth-conditionally, then the immediately adjacent next outer portion of that larger theory comprises the theory of the other linguistic moods of L(1). This will identify linguistic acts as acts of specifically asserting that [the sun 1s behind cloud, say], asking whether [the sun 1s behind cloud|, or enjoining (again in the thinnest

possible

sense, and

however

vaingloriously

in this particular

case)

that

[the

sun

be

behind cloud]. Cp. McDowell, “Truth conditions, bivalence and verificationism’, p. 44, who

assigns this task to a “theory of force”. For the reasons why one might hive this task oft from a theory of force in Austin’s more general sense, see Davidson, ‘Moods pp- 109—21. See also Hornsby, “Things done with words’, op. cit.

and performances’,

MEANING

AND

TRUTH-CONDITIONS

I3

10. 4.002 presages some of Wittgenstein’s later dissatisfaction with his own Tractarian philosophy—and anticipates many of the preoccupations of later twentieth-century philosophy. But the thing that must immediately have troubled him about what he had written at 4.024 was the non-operational

character of the conceptions of sense and truth that he had espoused in the Tractatus. By the time of Philosophische Bemerkungen, what he prefers to say is this: To understand the sense of a Satz means to know how the issue of its truth or falsity is to be decided. (Philosophische Bemerkungen, IV. 43). (Wittgenstein, 1975)

This new formulation looks backwards one decade at the doctrine of Tractatus, and sideways perhaps at the work of the Dutch mathematician L. E. J. Brouwer. But it is no less recognizable as the antecedent of the infamous claim advanced by the logical positivists of the 1930s—the claim that dominated the 1930s and 1940s and had an even longer period of influ-

ence in the philosophy of science—namely that the sense or meaning of a sentence is nothing more nor less than the method of its verification. In the next phase of his thinking about linguistic meaning, Wittgenstein came to advance a different claim, namely, that (“for a large class of cases”)

to understand a linguistic expression is simply to grasp “its use in the language”. (See the Blue and Brown Books and see the two decades’ worth of

philosophy books by other philosophers who were influenced by this formulation.) As verificationism fell out of favour, this doctrine rushed in to fill

the vacuum that was left by its disappearance.'® Then, as the limitations came to be perceived of the doctrine of meaning as use, ous contribution to the philosophy of meaning was H. the meaning of a declarative utterance was a function of to use that sentence to induce (by the recognition of

the next conspicuP. Grice’s idea that speakers’ intentions that intention) this

or that belief. (The trouble with that, one might think, was that such an

12. My recollection from being an undergraduate at Oxford during the 1950s at the time when Austin was giving the lectures he then called Words and Deeds (1954—5), but before the appearance of Grice’s article ‘Meaning’ (1957), 1s that in that period the doctrine then current about the meaning of words and sentences was simply a generalization of the Wittgensteinian thesis that meaning was use. There was no audible trace of the idea that to know the meaning of a sentence was to know what it would take for 1t to be true. To judge by my experience three years later in the Princeton philosophy department, the situation was very much the same in North America.

14

MEANING

AND

TRUTH-CONDITIONS

intention was really the intention to tell someone something. It seemed to presuppose rather than to explicate the notion of saying that thing. It left too much to do under that head.)

The Fregean idea was destined to be rediscovered for philosophy and accorded an attention it had never previously enjoyed, but scarcely immediately.”” For English speakers, it remained more or less buried until 1959, when

Michael

Dummett’s

article “Truth’ (1959)

disinterred it and put it

back into circulation.'* This limited circulation was yet further narrowed by the fact that Dummett expressed reservations of his own, not dissimilar to those we find in Wittgenstein, about the acceptability of the Fregean equivalence between sense and truth-conditions.

11. So much then for the shift that Wittgenstein himself seems to have prompted away from the doctrine of Tractatus 4.024, and so much for the philosophy of language that had worked itself out over the period between 1921 and the 1960s, downstream of Frege, Russell, and early Wittgenstein,

before Davidson’s philosophy of language first became visible. But now let us go back to the point in the argument that we had reached at the end of §7. In §7, having expounded Grundgesetze 1.32, we claimed that Frege or Wittgenstein would have been well placed to defend the truth-conditional thesis against the objections mentioned in §{4 by formulating it as follows: in Begriffsschrift extended (Bg+), s can be used to say literally that p if and only if the equivalence [True s if and only if p] flows from the referential stipulations for the language Bg+.The difficulty that this left over was this: that the most that this positive doctrine will ever enable us to put on the page is an account of what it is for a sentence to say-in-the-language-of-Bg+ that p, or 13. It 1s true that in the 1950s Frege’s writings were being translated. But neither The Foundations of Apritlunetic nor ‘On sense and reference’ (the one paper which Carnap, Quine, Feigl, and Sellars had

made familiar to all professional philosophers) explained what the sense of a sentence was to be. Nor did any of Geach’s and Black’s other Selections. It is true, too, that Tractatus 4.024 was legible enough. But, by its apparent archaism, the picture theoretical framework obscured the doctrine. 14.

Dummett,

‘Truth’.

It 1s noteworthy

that

in

the

several

decades

here

under

consideration,

Wittgenstein’s 1s the one clear philosophically salient formulation of the connection that Frege discerned between sense and truth-condition. Frege’s doctrine on this point is conspicuous by its absence from expositions where we might have expected to find it, such as those of Alonzo Church at §o4 of his introduction to Introduction to Logic (1956) and Rudolf Carnap at §33 of Der Logische Aufbau der Welt (1928). (For Carnap’s own nsufhiciently remarked final return to a Fregean position, without explicit acknowledgment to Frege, see Introduction to Semantics, 1944, p-22.) | have wondered whether it 1s something connected with the blind spot I seek to explain in the text that accounts for the strange neglect of Richard L. Cartwright’s definitive improvement (1954) of Quine’s criterion of ontological commitment, namely his reformulation of this in terms of rules of truth. See Cartwright’s ‘Ontology and the theory of meaning’, an article that rehearses and resolves difficulties that were still under active discussion a whole decade later.

MEANING

AND

TRUTH-CONDITIONS

I§

s can be used to say-literally-in-Bg+ that p—and s means-literally-in-Bg+ that p—if and only if it is derivable from the referential-stipulations-for-Bg+, specified thus.. ., that s is a true-Bg+-sentence if and only if p.

This points at something general, namely the thing that Wittgenstein gets across in 4.024. But how can we articulate this general thing? How can we

extricate “mean literally”, “say literally”, or “referential stipulation” from these hyphenations with “Bg+""? 12. One manageable objective we might set ourselves is this: to arrive at the generalization we need by satisfying all the necessary conditions to supplant the constant “Bg+"” with a variable “L(1)”. If we proceed in this way, how can we make explicit the thing that the Bg+-relative condition only shows? Looking back at what we then have to generalize and free from relativity to Bg+, it may appear that the chief obligation we now incur is to dispense with the reference to particular stipulations such as T(1),T(2), P(1),..., and

so on. Instead, we have to say explicitly what sort of thing a referential stipuJation is. And perhaps the most natural first suggestion will be that we should advance on the following basis: s means that p in L(i) if and only if there is a theory © for L(i), namely @0, that associates each expression of L(i) with its proper value, and this @® implies that s is true if and only if p.

Such a proposal will resonate in multiple ways with a common theme variety of semantical traditions. (Davidson calls it the building-block posal.) The only trouble is that, in practice, it has never been brought vincingly to life. There is nothing both general and foundational to be

in a proconsaid,

simply in terms of reference, about how “and”,“not”,“Caesar”, and “behind

cloud” all have their meaning.We cannot dispense in semantics with something like the idea of reference. Equally, however, we cannot make out of

the idea of reference the whole basis for the semantics of the sentence. From a standing start, we cannot even explain in such terms what distinguishes a sentence from a mere list. Frege himself never at any point dispensed with the idea of reference. But he also insisted, in the preface to The Foundations of Arithmetic, that “only in the context of a sentence does a word mean or stand for anything”. Somewhere near the beginning of our account we have to render it more intelligible than this first suggestion will that sentences can be used not merely to list items of reference but to say things. Rather than start with the idea of reference, ought we not to try to start at the other end, with truth itself and the contributions that the constituents of a sentence

16

MEANING

AND

TRUTH-CONDITIONS

make to the conditions for its truth? On these terms, let the meanings of words be just as various as their contributions. 13. Noting that truth and meaning are symmetrically relativized to language in the elucidation of meaning we offered at the end of §11, we shall find a

different suggestion we can explore. Not only did O

in §7 state the

meanings of each sentence of the language L(1) and the meaning of [each] constituent of L(1).As a by-product of doing that, it fixed systematically and non-accidentally correctly the extension of the predicate “true” as restricted to L(1) sentences.

Thus we have it that “The sun is behind cloud” is true if

and only if the sun is behind cloud, “The moon is behind cloud” is true if and only if the moon is behind cloud, and so on. (Such biconditionals are sometimes called partial definitions of ‘true sentence of L(1).) We need not

know which sentences are the true ones or constitute the actual extension of ‘true-in-L(1)’. But we do have a systematic way to state the principle on which that extension is assembled and, in that however strange or philosophically unwonted sense, we have a ‘definition’ of ‘true-in-L(1)’. We have

arrived (that is to say) at a way of giving a condition for its truth to each possible sentence of L(1). (Compare Chapter 6, Section s, for this variety of

‘defining’.) So the new thought is this: rather than arrive at an account of meaning by trying to generalize from the idea of designation, why not underwrite the Tractatus 4.024 generalization by saying the following? for any s, s can be used to say literally in L(i) that p—s means literally in L(i) that p—if and only if it is derivable from the definition of true sentence of L(i) that s is true if and only if p. 14. Having had recourse, in this last transposition, to the idea of a definition of truth in L(i)—compare Section 7 and the treatment there of L(i)—,the time has come to turn our attention away from the main trend of semantic

speculation in analytical philosophy, and away from Jena, Vienna, and Cambridge towards Lwow, Warsaw, and the study that Tarski called the “methodology of the deductive sciences”, which was one part of Tarski’s contribution to the prewar development of mathematical logic.'

15. By “methodology of the deductive sciences” was meant, inter alia, the systematic study of such notions as sentence, consequence, definition, deductive system, equivalence, axiom system independence, consistency, and completeness.

MEANING

AND

TRUTH-CONDITIONS

17

The change of orientation is at first surprising. We are inclined but not necessitated in this direction by the formal shape of the problem we have been considering, which relates only to the conceptual lacuna that divides Grundgesetze 1.32 from

Tractatus 4.024. Other directions are thinkable. Yet,

given the actual influences that have formed the semantical speculations of nowadays (Davidson’s and others’), there is no helpful alternative—however

oblique Tarski’s concerns are to Davidson’s, and however indirect our progress towards a general theory of linguistic meaning may appear. Let us begin by asking the question how it can have come about, if the theory of Fregean sense was in no way Tarski’s preoccupation, that Tarski should have been interested in identifying a set of axioms for a language L (1) that delivered theorems given in the form [s is true in L(1) if and only if p]. Why was Tarski interested in axioms delivering the theorems of which philosophers of language such as Davidson and his followers were going to say that they determined the sense or contribution of each of the expressions of L(1)? The answer is that, even though Tarski was not interested in

meaning as such, he was interested, and interested in a special way, in truth.' He was interested in the idea of truth neither after the fashion of the traditional logic—truth simply as the thing that valid inference preserves—nor after the fashion of philosophers who are exercised by the more mysterious and perennial questions about truth. The sort of thing Tarski was interested in doing was finding ways to compare and contrast the class of true formulas of a given formal language L(i) with the class of formulas that the rules and axioms make provable there. Embarking on inquiries of this kind, the thing that Tarski needed was a systematic account of what determined the extension of the concept true.'” (Such a systematic account, given in what I have invited the reader to see as a modernization of the method of Frege’s Grundgesetze, is what Tarski and many others since have called a ‘definition’.) But that was not everything he needed. He also needed to find assurance that his account of truth would not be undermined and discredited in the eyes of the community of mathematicians by the ancient paradoxes that

16. See Tarski, “The concept of truth in formalized languages’; also ‘“The semantic conception of truth’ and ‘Truth and proof”. 17. Having determined the extension of these concepts, of the true and the provable, he could then inquire whether they coincided. Tarski went on to show that the metalinguistic definition of ‘provable 1n L(1)’—a purely syntactical notion—could be given within L(1); but that, for any L(1) of suflicient expressive power, the semantic paradoxes would obstruct the definition of ‘true within L(1)” in L(1).

18

MEANING

AND

TRUTH-CONDITIONS

exploited that idea, Epimenides’ paradox, for instance (cp.Tarski, 1931, p. 110; 1930, p. 252).

Let us take the second of these problems first. Tarski’s analysis of the liar paradox and its variants suggested to him that the best way to safeguard the construction he had in mind was to begin with some particular object language that was itself free from all semantic notions. Once this object language itself was made determinate, semantic concepts'® such as satisfaction, truth, and designation,'” as restricted to that object language,

could be introduced into the metalanguage for that object language by defining each deliberately, with full formal correctness, in terms drawn from the object language (or translations of the same into the metalanguage), from elementary set theory and from the formal morphology of the object language, as given in the metalanguage.®® On these terms, one could assure oneself that, if the object language was immune from paradox, then the metalanguage that contained the object language would be immune

too.

15. So far so good. But on what principle was a restricted, paradox-free notion of truth, the concept true sentence of L(i), to be positively characterized? What was the philosophical or intuitive substance of the idea? For his thoughts about this, Tarski turned (by his own account®') to his teacher Tadeusz Kotarbinski’s book Elementy Teorji Poznania (1929), where we find the following passage (itself reminiscent of Tractatus 4.061): Let us pass to the classical doctrine and ask what is [to be] understood

by

“[a sentence’s or thought’s] accordance with reality”. The point is not that a true thought should be a good copy or [fac]simile of the thing of which we are thinking, as a printed copy or photograph is. Brief reflection sufhces to recognize the metaphorical nature of such a comparison.A different interpretation of “accordance with reality” is required. We shall confine ourselves to the

18. That 1s, as Tarski puts it, “concepts which, roughly speaking, express certain connections

—

between the expressions of a language and the objects and states of affairs referred to by those expressions’. 19. The extensionally defined counterpart of reference 1s the valuation or asterisk function as it 1s defined for each L(1). For the importance of not beginning by calling this function that of ‘reference’, see McDowell, ‘Physicalism and primitive denotation’. 20. The metalanguage 1s the language in which one may speak of whatever the object language speaks of and also of the expressions of the object language in their relation to that which the object language speaks of. 21. See the Bibliography to Tarski, ‘“The concept of truth’.

MEANING

AND

TRUTH-CONDITIONS

109

following explanation: “John judges truly if and only if things are thus and so: and things are in fact thus and so”.*

Spelling out this explanation for the case of some particular sentence, we have John judges truly in saying “snow is white” if and only if (1) John is right in saying “snow is white” if and only if snow is white (2) snow i1s indeed white.

But then it seems we can have, more simply* “Snow is white” is true if and only if snow is white. The

chief thing that it seems the definition of “true in L(1)” must do in

order to conform to Kotarbinski’s requirements is to imply one such equivalence in respect of each sentence of L(i).** But now, having come this far, we shall be moved to ask: how otherwise

can the definition of truth in L(i) furnish the thing Kotarbinski required than by doing the sort of thing we have seen that O"() did? This is how Tarski’s path comes to cross the path we have seen Frege’s and Wittgenstein’s thoughts as marking out. The parties are moving in different directions, but at the intersection there is one common thing each party needs in order to arrive where it is headed. Each party needs to involve itself, for any language that comes into consideration, in the generalization of something like the exercise conducted in §7.

16. In the light of this, how is the problem to be solved of saying what a referential specification is? Well, if there is this convergence, then Tarski

must have the same problem under a different name if he is to say what a definition of truth is. Tarski has to say what such a definition must be like in order to be adequate. The problem is solved as follows: 22. Elementy Teorji Poznania, pp. 106—7 in the English translation. Note that neither Kotarbinski nor Tarski takes this schema to be the recipe for a redundancy, deflationist, or (as Tarski says) nihilistic theory of truth. Indeed, Tarski sometimes claimed to be coming to the rescue of the

correspondence theory—though this claim must be taken with a pinch of salt. (Nothing in Tarski’s theory sentences and will vindicate 23. For the claim addition

can vindicate the 1dea that truth is to be defined in terms of a relation between states of affairs. Nor 1s there anything essential to the Tarskian construction that the classical conception of truth as bivalent. Such questions could remain open.) about Tarski and Kotarbinski, see Wiggins, Needs, Values, Truth, pp. 333—4. (In

to making

general

reference

to Kotarbinski’s

book, Tarski

refers also

to lectures

in

Warsaw by Lesniewski. But the main burden of that acknowledgment seems to relate to the semantic paradoxes.) 24. For the failure of proposals to deliver this result by the method (which 1s not Tarski’s official

method) of simply conjoining ‘partial definitions’, see Milne (1999).

20

MEANING

AND

TRUTH-CONDITIONS

A formally correct definition OO of the predicate “true” as applied to L(i) sentences is materially adequate if and only if, for every sentence s of L(i), ® implies a biconditional (or so-called T-sentence) in the form [True s if and only if p], where ‘p’ holds a place for a translation of s into the metalanguage ML(i).

Tarski calls this provision—which is evidently not itself statable at any level lower than the meta-metalanguage—Convention T.*® It is simply the gen-

eralization of Kotarbinski’s desideratum.”® Similarly then, each referential specification for L(i) assigns a value to every expression in L(i); and a set of such assignments is materially adequate under the very same condition as Tarski gives. It must yield a T-sentence for each sentence of L(i).

And each

T-sentence must in the same way be translational. This is to say that, in each case, ‘p’ must hold a place for a translation of s into the metalanguage. 17. Does this represent any progress? For Tarski, it is progress, because Tarski’s

only objective is to arrive at a non-accidentally and recognizably correct definition of true sentence of L(i). The word “translation” is not being used

here in a manner that offends notions. It occurs only in the fully say) in Kotarbinski’s and it presupposes only this: that given on the right-hand side

against Tarski’s professed attitude to semantic meta-metalanguage, or (as one might fanciTarski’s philosophy of truth. Occurring there, a logician can recognize when the sentence of a T-equivalence is faithful to the meaning

of the sentence mentioned on the left. Nevertheless, because Convention T

includes within it a semantical term coordinate with the ideas of meaning, definition, and the rest, anyone who is concerned with the idea of meaning

for its own sake still faces the same old question. How can we eliminate the semantical term “translation” from Convention T? Or how can we analyse or dismantle it there? Here at last we can resume the story that we have already carried up to 1959, which was the moment, as we saw, when Michael Dummett put the

truth-conditional insight back into circulation. If anybody had been concerned with the question of how to make Wittgenstein’s generalization 4.024 work, then Tarski’s construction would have served him perfectly—

unless he had had such an obsessive concern with the nature of meaning itself that it was not sufficient to trace and explore the small circle that joins 25. Cp. ‘The concept of truth’, p. 187. The words in the text above are an application of Tarski’s doctrine, not a quotation.

26. Material adequacy 1s adequacy to the subject matter, which 1s truth. It therefore entails nonaccidental fidelity to the extension of the predicate.To think here of the “material” conditional /biconditional will excite precisely the wrong associations. See Tarski (1931), p. 129.

MEANING

AND

TRUTH-CONDITIONS

the ideas of truth, meaning, and translation. Here

though

21

is the trouble—:

that, perfectly properly, ex officio, and by its nature, philosophy is imprisoned within that obsession.

18.To understand Donald Davidson’s revival of the general idea of meaning as given by truth-conditions and the distinctive advance that this made possible, it helps to appreciate the immediate background of his speculations. This was not any concern on Davidson’s part with the theory common to early Wittgenstein (to whom Davidson rarely, if ever, referred) and Frege (to whose doctrines Davidson evidently regarded Alonzo Church as the complete guide, even though this guide completely omitted all mention of Frege’s truth-conditional insight). The background was more topical, namely

Davidson’s doubts about Carnap’s methods of extension and intension,*” his considered rejection of the answer to the question of linguistic meaning provided by H. P. Grice’s reduction of semantic notions to psychological ones such as belief and intention,*® and Davidson’s attachment to the speculative framework furnished byW. V. Quine’s book Word and Object (1960)—

most especially the question of what a thinker from outside a community of speakers would need to avail himself of if he were to try to make sense of utterances in their unknown language. What Davidson wanted was to retain Quine’s naturalistic approach to such questions, to align himself with Quine’s objection to all “museum myths” of meaning, but to do so without commitment to Quine’s talk of ocular irradiation, neural impacts upon subjects, and the rest. According to Davidson, the thing that impinges on subjects had better be the world itself, the world that is common to interpreter and subjects alike. Seeking for some framework within which to give a systematic account of the information (or putative information) that an interpreter would need to amass and draw upon in order to frame his hypotheses about the meanings of his subjects’ uttered sentences, and seeking at the same time to sweep away the supposed obscurity of ‘s means that p’, the construction Davidson found himself reaching for was in effect none other than Tarski’s. Davidson writes: Let us try treating the position occupied by ‘p’ [in ‘s means that p’] extensionally: to implement this, sweep away the obscure ‘means that, provide the

27.

Davidson, ‘Carnap’s methods of intension and extension’.

28.

See again Grice,‘Meaning’.

22

MEANING

AND

TRUTH-CONDITIONS

sentence that replaces ‘p” with a proper sentential connective, and supply the description that replaces s with its own predicate. The plausible result is (T) sis

T if and only if p.

It is worth emphasizing that the concept of truth played no ostensible role in stating our original problem [the problem of a theory of meaning for a given language]. That problem upon refinement led to the view that an adequate theory of meaning [for the language spoken by the interpreter’s subjects] must characterize a predicate meeting certain conditions. It was in the nature of a

discovery that such a predicate would apply exactly to the true sentences...A Tarski-type truth definition supplies all we have asked so far of a theory of meaning. (Davidson, 1967)

The discovery is of course a rediscovery—the rediscovery of the thing that Frege and Wittgenstein had articulated and that Davidson failed to credit to Frege or to Wittgenstein. If Frege’s original insight had not been correct, there could have been no such discovery. Working within Quine’s framework, however, the attitude Davidson

had towards Tarski was as follows.

Taking translation for granted (or taking “means in L(1)” for granted), Tarski

had defined “true sentence of L(i)”. Conversely, then, why

should not

Davidson take truth in L(1) for granted, in order to define “means in L(1)"?

The residual problem was then to dispense with Tarski’s use of the word “translation”’.

19. Davidson’s first thought about that problem seems to have been that he could secure everything he needed if he were simply to omit the requirement that the T-sentences generated by @@ in the form [s is true if and only if p] should provide “translations” on the right-hand side of the L(j) sentence s mentioned on the left. Could he not stipulate instead that absolutely all the T-sentences that ®@'® generated should be true? But it is now

pretty clear that the condition is not sufficient.*” From the beginning of all Davidson’s speculations, shaped as they were by Quine’s Word and Object, the real solution to this problem was always at hand. Perhaps Davidson’s best account of this solution is the one given in 29. It can be proved that, if there 1s of the language L(1), then there ations to be read off the second Evans and McDowell’s editorial superseded by the footnote that meaning’ (p. 26, n. 10) however

one theory that provides a true T-sentence for each sentence will automatically be a second such theory, and the interprettheory will be different from those to be read off the first. See introduction to Truth and Meaning (1976). Their finding 1s not Davidson added in 1982 to the Inquiries reprint of ‘“Truth and illuminating the footnote might be in other ways.

MEANING

AND

TRUTH-CONDITIONS

23

his “Radical interpretation” (1973). But there is a real point in giving a Davidsonian solution in a variant that is not open to the objections that so many critics have urged against the particularities of Davidson’s own for-

mulation.” The distinctive features of the variant presentation are chiefly due to Richard Grandy and John McDowell.*' If the interpreter of the utterance of a sentence is to say what it means, then he has to find out under what conditions the sentence, being the sentence it is, counts as true.To say so much is to say little more than Frege said. But the next thought one will have reaches beyond Frege. It is that linguistic behaviour is a proper part of behaviour. But, if so, there ought to be some other than purely, simply semantic way of specifying what it is for a radical interpreter to succeed in interpreting alien subjects. Surely the interpreter’s linguistic efforts are part of the larger effort to interact successfully with subjects, to coordinate his/her (the interpreter’s) practical doings with those of subjects...in a word, to make sense of subjects and have them make sense of him/her. If we enlarge in this way, in terms which are not specifically semantical, upon what interpretation is attempting to achieve and to explain—and if we count the interpretation of speech as one proper part of the larger undertaking—then here at last we arrive at the substantive non-semantic constraint upon @-® that we have been looking for: A definition of truth in L(i) will be materially adequate if it generates a T-sentence for each sentence s of L(i) and collectively the T-sentences that the definition implies, when experimentally applied to individual utterance by the speakers of L(1), advance unimprovably the effort to make total sense of the speakers of L(i). The notion of total sense is not a semantic notion, but it subsumes one. One

person’s making sense of another is a matter of their participative interaction in a shareable form of life, of their homing upon the same objects and/or the same sorts of object, of their being in a position ceteris paribus to

30. Objections have mostly related to Davidson’s freewheeling use of the i1dea of an interpreter’s needing to find what sentences a subject lolds true. It must be noted, however, that Davidson has persisted in this part of his original presentation, and has developed it further in his Dewey lectures, Journal of Philosophy (1990). 31. See Grandy, ‘Reference, meaning and belief’; Evans and McDowell, editorial introduction to Truth and Meaning; McDowell, ‘Truth conditions, bivalence and verificationism’, §1; and McDowell,‘On

the sense and reference of a proper name’.

24

MEANING

AND

TRUTH-CONDITIONS

succeed in joint enterprises, and so on. In so far as we make sense of others, we deploy a mode of understanding that can be redescribed, however artificially, as follows. There is a store of everyday predicates of human subjects, of features of the environments that impinge on subjects, and of the events that are counted as the actions or conduct of such subjects. When we seek to attain participative understanding, we seek in response to circumstances, including the speech or conduct of subjects, to distribute predicates of these and other kinds across features of reality, mental states, and actions in such a

way that: (1) the propositional attitudes we ascribe to subjects, specifying the content of these attitudes, are intelligible singly and jointly in the light of the reality to which we take subjects (or their informants, or their informants’ informants.

..) to have been exposed; and (2) the actions (and actions

of speaking) that we ascribe to subjects are intelligible in the light of the propositional attitudes we ascribe to them.> In the form in which we now have it, the new elucidation of meaning finally bridges the gap between Frege’s stipulations for his concept-script and Wittgenstein’s bold generalization of Frege’s idea. Of course, it inherits all the well-known difficulties of the ideas of understanding, explaining, making intelligible, imaginative projection, or identification. But these difficulties are there anyway. The proposal not only depends upon these ideas. [t assists us by helping to trace their interrelations. 20.The conclusion to which we have been drawn is that what it is for a sentence to mean that the sun is behind cloud and to be available to enable us to say that the sun is behind cloud, is as complicated as this. It involves a biconditional, ‘“The sun is behind cloud” is true if and only if the sun is behind cloud’, which is imbedded within the scope of an operator whose presence indicates that this biconditional is derivable from the whole system by which we make sense to one another and make sense of one another. What we have here is the idea of a significant language as a system that correlates strings of repeatable expressions with the states of affairs that the strings can draw attention to or get across, this system itself being a subsystem of the larger system by which social beings participate in their shared life. There is nothing abstruse in that. It is because we grasp it so readily (I think), both in philosophy and before philosophy, that we can hear a T-sentence given in the form “s is true if and only if p” as the output of such a system.

32. See again note 31 above.

MEANING

AND

TRUTH-CONDITIONS

25

When we grasp that, it is tantamount to our grasping something intuitively similar to the turnstile = (Bg+) . that played the part we described in the Fregean elucidation of the meaning of Begriffsschrift-extended sentences. But the turnstile I is no longer relativized to any particular language. 21. Objection may be made because, in the formulation I have set down here, s can only have it as its literal use to say that p if all suitably constrained theories imply that s is true if and only if p. What reason is there to suppose that this condition is non-vacuously satisfiable? The objection is a good one, because the formulation does seem to foreclose a matter that ought to have been left open. It seems better on reflection to postpone such questions until we have a fuller account of what it is to make sense of the shared life and conduct of L-speakers. This is a question of the indeterminacy of interpretation—or of translation, as Quine says. In the interim, perhaps we should rule that it is sufficient for s to mean that p that some putatively unimprovable theory that meets all the constraints should entail the biconditional [s is true if and only if p]. 22. [t may be objected that the idea of translation, which our final proposal purported to remove, returns surreptitiously with the idea of an interpreters ‘making sense’ of other people. But to this the theorist of truthconditions must reply by simply reiterating his claim that the idea of making sense that we find here is a much wider one than the idea of linguistic interpretation. The presence or absence of this more general thing can be demonstrated non-linguistically. The ideas of making sense of and being made sense of embrace and subsume the ideas of saying and the interpretation of saying, and they involve them illuminatingly with coeval, collateral ideas of explanation and understanding—even (as you may say, if you are as convinced as I am of the indispensability of these further things to the full story) with the idea of participation by interpreter and subjects in a shared form of life, and the idea of explanation as Verstehen. 23. A third objection might take the following form. After all the changes and emendations consequential upon earlier objections, should not all residues of the idea of compositionality itself have been expelled from the final formulation? “Truth itself is unduly emphasized in your construction”, the objector may say.“One might accept this for argument’s sake as the result of your foolish concentration upon declarative utterances. But, even in the cases where truth really does belong, it is surely not necessary to insist that the interpretive biconditional should be generated by the recursively or

26

MEANING

AND

TRUTH-CONDITIONS

compositionally generated definition of truth that you envisage for the language L(i). If we are simply helping ourselves now to the idea of what it requires to ‘make total sense‘ of speakers, Verstehen and the rest, why cling to this residue of Fregean compositionality?” To this I would reply that the meaning we are interested in understanding is linguistic meaning, the non-natural meaning possessed by sentences

that will be further saturated by context of utterance (etc.). It is the meaning with which sentences of what we recognize as languages are invested. Generally speaking, what makes interpretation possible is the fact that the language to which the sentences belong can be treated as pre-existing any particular speaker or hearer and any particular act of communication. It is something that speakers and hearers need to know about already. The compositionality that theories of L(i)-sense or definitions of ‘true sentence of L(1)" have to reflect is a property of the language L(i) itself, L(1) and its prop-

erties being something irreducible to any psychological, social, or prelinguistic fact or facts about individual speakers or individual situations of communication.” (See Chapter 8.) 24. In opposition to such claims as the one just entered, many have tried to see the clauses of the definition of ‘true sentence of L(i1)’ as answerable, in the

last analysis, to psychological or neurolinguistic facts about speakers. After further reflection, some among those who are tempted by such an approach have shied away from the manifest embarrassments of getting involved in all that. And, backing off, they have preferred to say (as John Foster and Donald Davidson have more or less agreed in saying™) that the “theory” corresponding to the definition of true sentence of L(i) “explicitly states something knowledge of which would suffice for interpreting utterances by speakers of the language”.”® There are doubts about this kind of formulation. My own view would be that the question it answers should never have been permitted

33. That 1s to say that | propose

that one see language

as a social object with a past, a present, and

a future, something that is for each generation of speakers an objet trouve, with words and modes of combination possessed contingently of this, that, or the other meaning. Languages are not, on this conception, abstract objects defined by their syntax or semantics. (As Nietzsche remarks, nothing with a history can be defined.) What the syntax and semantics (as of f) are answerable to 1s the state of this language (as of f), not the states of the speakers who aspire to speak that language. 34. See their respective contributions to Evans and McDowell, Truth and Meaning. 35. That is to say that they shy away from claiming that this is the theory that speakers actually use. Davidson, however

(who

has so much

to lose from misunderstanding

here), has not, when

he

has spoken of speakers and interpreter’s ‘theories’, exercised all the caution I should have counselled on this matter. See, for one instance among several, his ‘A nice derangement of epitaphs’.

MEANING

AND

TRUTH-CONDITIONS

27

to arrive at the point where it could exact either this or any remotely similar answer. The thing the definition of truth for L(1) 1s answerable to is how things

are with the social object that is the language L(i)—mnot how things are with the speakers past and present in virtue of whose existence that language is extant. The question for anyone who would define truth for L(i) is this: how have we to see L(1)—how

must we parse it and segment

it—in

order to

understand why its sentences mean this or that? How do we have to see L(3)

in order to get principles by which we work out what its more complicated or obscure L(i) utterances mean? Again, why do L(i) sentences have to be translated into foreign tongues on this principle rather than that principle in order to arrive at a passable version of what was originally said? In so far as purposes such as working out what sentences mean and discerning principles of translation do not force us into one sort of grammatical description rather than another, there may be indeterminacy about the properties of L(i). But that is nothing new. Nor does it render it indeterminate which object the language L(i) is. L(i) is a historically given thing, changeable no doubt, and always in process, but a persisting social object nevertheless. It is not in any reprehensible sense an indeterminate or mythical object. (See Chapter 8.) 25. One last question.

What, then, after all these twists and turns, was the

advantage of going by the Tarskian route to our final destination? One alternative might have been to reflect that we never really define or reduce anything in philosophy. So someone might ask: Why not gloss the notion of meaning in a freewheeling fashion by simply using it and involving it with all the collateral notions that are imported by the idea of interpretation?* Such, after all, is the method of philosophical elucidation—the method we have learned not to hope to improve upon. There is much to agree with in this objection (the Davidsonian account is an exercise in elucidation too) except that the one principal contention seems wrong. It seems wrong to suggest that we should deny truth its foundational place in the elucidation of meaning. For there is a real advantage in going by the Fregean and Tarskian way. Yes, Tarski’s construction, which consolidates Frege’s, is conditioned in the first instance by Tarski’s deep suspicion of primitive semantic notions, and this is a suspicion one may not share. But suspicion of the semantical as such is not the only possible reason one might

36. See, e.g., the approach to meaning of Sainsbury, ‘Understanding and theories of meaning’, pp- 127—44; and of Martin Davies, Meaning, Quantification, and Necessity.

28

MEANING

AND

TRUTH-CONDITIONS

have to applaud the fact that Tarski gives his construction in terms of simple truth (not truth in a structure/model),” that he introduces semantic notions

deliberately and in a measured fashion, and defines notions such as satisfaction and the valuation function by fixing their extension. One may applaud all this not because one thinks semantic notions really are suspect, but because an account of meaning that builds on Tarski’s construction helps to show how meaning is possible. By seeing the definition of ‘true sentence of L(i), for any language L(i), as needing to be built up in this careful and austere fashion, while the output of the definition is constrained in a manner that is irreducibly non-austere (as messy and anarchical as the social always will be), we can understand something about how it is possible for there to be such a thing as the semantical, and on what conditions it is possible, namely the existence of both the compositional (in the small) and the social (in the large).

References Austin,J. L. 1962 How to Do Things with Words, edited by J. O. Urmson. Oxford: Clarendon Press. Ayer,A. J. 1953 “Truth’, Revue Internationale de Philosophie 25, 183—200. Carnap, R. 1928 Der Logische Aufbau der Welt, Berlin: Weltkreis-Verlag. Carnap, R. 1944 Introduction to Semantics, Cambridge, MA: Harvard University Press.

Cartwright, R. L. 1954 ‘Ontology and the theory of meaning’, Philosophy of Science 21 (4), 316—25.

Church,

A.

1956

Introduction

to Mathematical

Logic, Princeton,

NJ:

Princeton

University Press.

Davidson, D. 1963 ‘Carnap’s methods of intension and extension’ in P. A. Schilpp ed. The Philosophy of Rudolf Carnap, La Salle, IL, and London: Cambridge University Press. Davidson, D. 1967 “Truth and meaning’, Synthese 17, 304—23. Reprinted in Inquiries into Truth and Interpretation, 1984, Oxford: Oxford University Press. Davidson, D. 1973 ‘Radical interpretation’, Dialectica 27, 313—28. Reprinted in Inquiries into Truth and Interpretation, 1984, Oxford: Oxford University Press. Davidson, D. 1976 ‘Reply to Foster’ in Evans and McDowell 1976, pp. 33—41.

Davidson, D. 1979 ‘Moods and performances’ in A. Margalit ed. Meaning and Use, Dordrecht: Reidel. Reprinted in Inquiries into Truth and Interpretation, 1984, Oxford: Oxford University Press. Davidson, D. 1986 ‘A nice derangement of epitaphs’ in E. Le Pore ed. Tiuth and Interpretation, Oxford: Blackwell. 37. On this point, see again Milne (1999).

MEANING

AND

TRUTH-CONDITIONS

209

Davidson, D. 1990 ‘The structure and content of truth’, Journal of Philosophy 87 (6), 279—328. Davies, M. K. 1981 Meaning, Quantification, Necessity, London: Routledge. Dennett, D. 1987 The Intentional Stance, Cambridge, MA: Bradford Books.

Dummett, M. 1959 “Truth’, Proceedings of the Aristotelian Society $9, 141—62. Dummett, M. 1973 Frege: Philosophy of Language, London: Duckworth. Dummett, M. 1986 ‘The philosophy of thought and the philosophy of language’ in Meérites et limites des methodes logiques en philosophie, ed. J. Vuillemin, Paris:Vrin et Fondation Singer Polignac. Evans, G. and J. McDowell eds 1976 Truth and Meaning: Essays in Semantics, Oxford: Oxford University Press. Foster,J. 1976 ‘Meaning and truth theory’ in Evans and McDowell 1976,pp. 1—32. Frege, G. 1879 Begriffsschrift, eine der arithmetischen nachgebildete Formelsprache des reinen Denkens, Halle: Verlag von Louis Nebert. Republished and translated in Terrell Ward Bynum ed. Conceptual Notation and Related Articles, Oxford: Clarendon Press, 1972.

Frege, G. 1882 ‘Uber den Zweck der Begriffsschrift’, lecture delivered at the 27 January, 1882 meeting of the Jena Society for Medicine and Science, Jenaische Zeitschrift fiir Naturwissenschaft, 16 / Suppl. 1882/1883, pp. 1—10. See ‘On the object of my concept-writing’ in T. W. Bynum ed. and trans. Conceptual Notation and Related Articles, Oxford: Clarendon Press, 1972.

Frege, G. 1884 Die Grundlagen der Arithmetik, eine logisch-mathematische Untersuchung tiber den Begriff der Zahl, Hamburg: E Meiner. Translated as The Foundations of Arithmetic, by J. L. Austin, Oxford: Blackwell, 1950. Frege, G. 1892 ‘Ueber Sinn und Bedeutung’, Zeitschrift fiir Philosophie und philosophische Kritik 100, 25—50. Republished and translated in H. Feigl and W. Sellars eds Readings in Philosophical Analysis, New York: Appleton-Century-Crofts, 1949. Also republished in Translations from the Philosophical Writings of Gottlob Frege, edited and translated by P. T. Geach and M. Black, Oxford: Blackwell, 1952.

Frege, G. 1893 Grundgesetze der Arithmetik, begriffsschriftlich abgeleitet, vol. 1, Jena: Verlag Hermann Pohle. Grandy, R. 1973 ‘Reference, meaning and beliet’, Journal of Philosophy 70 (14),439—52. Grice, H. P. 1957 ‘Meaning’, Philosophical Review 66 (3), 377—88.

Hornsby,J. 1988 “Things done with words’ in Jonathan Dancy, J. M. E. Moravcsik, and C. C. W. Taylor eds Human Agency: Language and Duty: essays for J. O. Urmson, Stanford, CA: Stanford University Press. Kotarbinski,T. 1929 Elementy Teorji Poznania. Republished in 1966 in G. Bidwell and C. Pinder eds Gnoseology: the Scientific Approach to the Theory of Knowledge, translated from the second Polish edition by Olgierd Wojasiewicz, Oxford and New York: Pergamon Press. McDowell, J. 1976 ‘“Truth conditions, bivalence and verificationism’ in Evans and

McDowell (1976), pp. 42—66. McDowell, J. 1977 ‘On the sense and reference of a proper name’, Mind 86, 159—85.

30

MEANING

AND

TRUTH-CONDITIONS

McDowell, J. 1978 ‘Physicalism and primitive denotation’, Erkenntnis 13, 131—52.

Milne, P. 1999 ‘Tarski, truth and model theory, Proceedings of the Aristotelian Society 90, 141—07. Quine,W. V. O. 1960 Word and Object, Cambridge, MA: MIT Press.

Sainsbury, R. M. 1979/1980 ‘Understanding and theories of meaning’, Proceedings of the Aristotelian Society 80,127—44.

Strawson, P. E 1971 Logico-Linguistic Papers, London: Methuen. From ‘Meaning and truth’, inaugural lecture, Oxford 1969. Tarski, A. 1931 ‘On definable sets of real numbers’

Metamathematics,

translated

by ]J. H.

Woodger,.

in his Logic,

Oxford:

Semantics,

Oxford

and

University

Press, 1955.

Tarski,A. 1936 “The concept of truth in formalized languages’ in his Logic, Semantics, and Metamathematics, translated by J. H. Woodger, Oxford: Oxford University Press, 1955.

Tarski,A. 1944 ‘The semantic conception of truth’, Philosophy and Phenomenological Research 4 (3), 341—76. Tarski,A. 1967 ‘“Truth and proof’, Scientific American 220, 63—77.

Wiggins, D. 1980 Sameness and Substance, Oxford: Blackwell. Wiggins, D. 1984 ‘On the sense and reference of predicate expressions.’, Philosophical Quarterly 34 (136), 311—28. Wiggins, D. 1991 Needs, Values, Tiuth, second edition, Oxford: Blackwell.

Wiggins, D. 1996 ‘Meaning and Truth-Conditions: From Frege’s Grand Design to Davidson’s’ in Bob Hale and Crispin Wright eds A Companion to the Philosophy of Language, Oxford: Blackwell, 1997, pp. 3—38.. Wiggins, D. 1997 ‘Languages as social objects’,

Philosophy

72

(282),

499—534,

reprinted here as Chapter 8. Wittgenstein, L. 1921 ‘Logisch-philosophische Abhandlung’ in his Annalen der Naturphilosophie. Republished and translated as Tractatus Logico-Philosophicus, 1922. Wittgenstein, L. 1958 Blue and Brown Books, Oxford: Blackwell; New York and London: Kegan Paul, Trench, Trubner. Wittgenstein, L. 1975 Philosophical Remarks, edited by R. Rhees, trans. R. Hargreaves and R. White, Oxford: Blackwell.

2 Concept and Copula’ Summary This chapter begins from the diagram that Frege drew for Husserl to show his (Frege’s) doctrines of sense and reference. The chapter explains and further develops the doctrines that the diagram illustrates, demonstrating their continuing usefulness and applicability. It suggests how a fair number of contemporary doctrines can be accommodated within Frege’s picture. Then the chapter seeks to show how the doubts and difficulties that Frege

himself expressed concerning concept and object can be answered by a reading of the copula which assimilates it to what Alonzo Church calls an “improper symbol”. The chapter ends with a suggestion, following on from Frege’s idea of a concept, about the best way of understanding the idea of a property—not least its use in evolutionary theory as Elliott Sober among others construes that.

Sentence

Singular term

Sense of the sentence

Sense of the singular term

Concept-word

Sense of the concept word

(Thought)

Reference of

Reference of

the sentence, a truth value

the singular term, an object

Reference of \ —

Object(s) that

the concept word, a

fall under the concept

concept

1. A much revised version, or rather a sequel, of “The Sense and Reference of Predicates: A Running Repair to Frege’s Doctrine

and a Plea for the Copula’,

1984.

Meaning, Truth, and the Limits of Analysis: Ten Studies. David Wiggins. Oxford University Press (2022). © David Wiggins. DOI: 10.1093/0s0/9780198726173.003.0002

32

CONCEPT

AND

COPULA

1. In a letter to Husserl dated 24 May 1891, Frege included an explanatory diagram to illustrate both the extent and the limitations of the parallelism he saw between the sense-reference relation for singular terms and the sense-reference relation for predicates:® Four points about this diagram deserve particular mention. First, in showing (as other evidence also shows) that Frege firmly intended predicates to have both sense and reference, the diagram vindicates the once nearly solitary stand taken up by Michael Dummett against numerous scholars who wanted to deny Frege’s intention to ascribe both Sinn and Bedeutung to predicates.® Secondly, in showing that the concept is never a sense but always a reference, it discredits a myth about Frege that prevailed in departments of philosophy influenced (it has been said) by memories of Rudolf Carnap’s lectures on semantics or else by Alonzo Church’s variant but professedly non-Fregean use of the word “concept” (see his Introduction to Mathematical Logic p. 6 n. 17). Thirdly, it vindicates the correctness of the general insistence (M. Dummett’s, M. Furth’s and others’)

that, for Frege, the analogies between possession of sense and reference by names and possession of sense and reference by predicates are no less important than the acknowledged differences. Fourthly, it serves as a warning of how misleading it is to try to equate the sense-reference distinction with Mill’s connotation-denotation distinction or divers other intension-extension distinctions which can still cloud the exposition of Frege (and of Leibniz). These are all salutary reminders. But how shall we state or explain the doctrine, now that we have recovered Frege’s picture of it? And what light can it cast upon the actual functioning of predicate expressions? If the diagram is where we start, and we treat it as a supplement to the doctrines of ‘On Sense and Reference’, we may want to enlarge upon the doctrine of columns 2 and 3 as follows. Just as we give the sense of a singular term and explain its contribution to truth-conditions by saying what object it stands for, so we may give the sense of a predicate and explain its contribution to truth-conditions by saying what concept the predicate stands for.* 2. See page 63 of Gottlob Frege, Philosophical and Mathematical Correspondence ed. and trans. Brian McGuinness and Hans Kaal (Oxford: Basil Blackwell, 1980). 3. See Dummett ‘Frege on Functions: A Reply’, Philosophical Review 64 (1955). See also Frege’s ‘Function and Concept’ in Geach and Black’s Frege: Translations (Blackwell, 1994). 4. To say that 1s to say what function, what mapping from objects to T or F, the predicate stands for. Where predicates with different sense import the same concept, their difference in sense is correlative with a difference in the way in which the mapping to T or F has to be conceived.

CONCEPT

AND

COPULA

33

Moreover, just as in the case of a singular term we can show one particular sense by using or exploiting one mode of presentation rather than another in explaining which object this is—preferring the particular body of information that sustains the particular mode of presentation that pertains to some particular name of that object®—so, in the case of a predicate, we may show this sense in preference over that sense, e.g., the sense of ‘horse’ rather

than the sense of (say) ‘Equus caballus’, by preferring some everyday conception over any scientific-cum-zoological conception of a horse by which

to show the sense intended.® 2. Such a view of the sense-reference distinction as it applies to predicates, whatever its ultimate difficulties as doctrine or as interpretation, would cohere very well with Frege’s insistence in the letter to Husserl and elsewhere (in the posthumous fragment ‘Ausfiihrungen tber Sinn und Bedeutung’,” for instance) that, just as singular terms without reference are unfitted to figure in the expression of a judgment that enables us to move forward® to a truth-value, and just as any name capable of figuring in the expression of a judgment that can constitute knowledge must have reference, so must any predicate that aspires to this status. It too must have reference. But many predicates essential to the expression of good (e.g. negative)

5. This formulation (moving beyond Frege’s own, but following on from the mode of presentation 1dea) draws upon several sources, notably G. E. M. Anscombe, Introduction to Wittgenstein’s Tractatus (London, 1959), p. 42; John McDowell, ‘On the Sense and Reference of a Proper Name’, Mind 96 (1977); Michael Dummett, FPL, p. 227 (“Even when Frege 1s purporting to give the sense of a word or symbol what he actually states 1s what its reference 1s...In a case 1n which we are concerned to convey or stipulate the sense of the expression, we shall choose that

means of stating what the reference is which displays the sense. We might borrow here a famous pair of terms from the Tractatus and claim that, for Frege, we say what the reference of the word 15, and thereby show what its sense 1s”); Evans, Varieties of Reference, Chapter 1. For the 1dea of sense as contribution to truth-conditions, see Grundgezetze, 1.32. A d

6. In my usage, conceptions of something are ways of conceiving or thinking of this (object, concept or whatever). In Sameness and Substance (Oxford, 1980), pp. 78—9, I have maintained that the Fregean apparatus 1s extensible to the special case (see e.g., Putnam, ‘Is Semantics Possible?’, Metaphilosophy 3 (1970)) of predicates whose sense is extension-involving (see pp. 10—11), and where (as with ‘horse’) the full grasp of the reference 1s coeval with some mastery of some part

of the extension (i.e., some knowledge of which objects ‘horse’ is true of).A rough and ready equivalence then becomes possible between knowledge of what Putnam calls the stereotype and having a de re conception of what a horse 1s—in a word, being party to a body of de re knowledge of horses. A version of these claims is given in my Continuants: their Activity, their Being and their Identity, Chapter 8. 7. In citations from that essay I have drawn upon PW, p. 118 ff. But, here as everywhere else, I shall remove the rendering ‘meaning’ that these translators prefer for ‘Bedeutung’ and replace it by ‘reference’. Or else | import the German word. 8. Cf. PW, p. 122: “The step from thought to truth-value, more generally the step from Sinn to Bedeuitung, has to be taken”.

34

CONCEPT

AND

COPULA

information do not have any object they are true of. It follows that their reference cannot be the object or objects they are true of.” The reference is to the concept. Thus, as Frege says: With a concept word it takes one more step to reach the object than with a proper name, and this last step may be missing—i.e., the concept may be empty—without the concept-word’s ceasing to be scientifically useful. I have drawn the last step from concept to object horizontally in order to indicate that it takes place on the same level, that objects and concepts have the same objectivity (see my Foundations of Arithmetic §47). In literary use it is sufficient if everything has a sense; in scientific use there must also be Bedeutungen."

This and the ensuing remark about Husserl’s rival or alternative scheme depend for their point on our taking seriously a however incomplete parallelism or analogy between singular term-reference and predicate-reference. (Parallelism, not assimilation.)

3. If one already had reason to want to do this, one might now embrace the doctrine of column 3. Consequentially upon doing so, one might then assign to a one-place predicate such as ‘horse’ what Dummett describes as “a property defined over every object” (one for which “it is in some manner specified [or specifiable] for each object that that object has, or lacks, that property”)."" This will furnish a concept for every eventuality. But at this point, more may be demanded about why predicates must have a Bedeutung— and how they can do so." 4.1 begin with the must. If we are to explain what it is for two concepts to be equinumerate—and surely we do have to be able to explain this—then it is (again) impossible to deny reference to predicates. For we need to be able to explain ‘equinumerate’ by saying “there exists a relation ¢ which correlates one-to-one the objects falling under the concept F with the objects falling under the concept G” (Grundlagen §72). And once we are moved by this example and its compulsory deployment of a two-place predicate ¢, we must be ready to advance beyond the claim that predicates

9. Thus 1n the Fregean scheme every significant predicate has a sense that i1s concept-involving or reference-involving. That 1s their normal condition. In note 6, I mention Putnam’s claims about the case of predicates whose sense and reference are extension-involving. This claim of Putnam’ and the general claim Frege makes about reference-involvingness are importantly distinct claims. Reference and extension for predicates are not the same. 10. Letter to Husserl dated 24 May 1891, op. cit. n. 2. 11. See Dummett FPL,p.89. Here I have noted also the sceptical comments ofVictor Dudman, 1976. 12. Cp. Dudman, op. cit., p. 73.

CONCEPT have

a sense

which

contributes

AND

COPULA

to sentence-sense. We

35 need

to insist that

what is at issue is a relation between two Bedeutungen. But how can a predicate have a Bedeutung? And what sort of thing is this? Well, one who understands a predicate will have to have some conception or comprehension of how a thing must be, what sort of thing it must be, or what it 1s like, if the predicate applies to it. And, if so, then the reference imported by that predicate has to be something general, something that is exemplifiable, something such that our comprehension of it can carry us from an object or objects that constitute the subject of a sentence to a truthvalue T or E" According to Frege’s general account of predication there is an analogy or parallel between the way in which ‘horse (...)" or‘...is a horse’ is seen as essentially incomplete and inviting completion by a singular term and the way in which the functor or function-frame ‘() squared’ or ‘( )*

stands in need of completion with some numeral or variable. Whenever we have a concept serving in this way as a function from objects to truth-values, we have the special case [where| the value is always a truth-value. That is to say, if we complete the name of a concept with a proper name, we obtain a sentence whose sense is a thought; and this sentence has a truth value as its reference. To acknowledge this reference as that of the True—to take the reference as the True—is to judge that the object which is taken as the argument does indeed fall under the concept. What in the case of a function is called unsaturatedness, we may in the case of a concept call its predicative nature (ibid.).

Frege calls this a special case. But it is the key to one of the most fundamental questions there are about language itself, namely how an assortment of words, over and above picking out or pointing to an object, can enable us to say something true or false about that object and what it is or what it’s like. s. At this point, however, Frege confronts what he sees as a serious obstruction—an obstruction that may seem to endanger the whole idea of a predicate’s both having a reference and needing completion by a singular term. Because every position in a significant sentence comes to be seen as essen-

tially “cut out” either for a singular term, saturated and standing for something complete, or else for a predicate, unsaturated and standing for a concept which is incomplete—incomplete, that is, without a singular term or a variable:

13. See Geach and Black Frege: Translations, pp. 24, $4—5, 113; compare PW p. 119 and Correspondence p. 5.

36

CONCEPT

AND

COPULA

there is now a great obstacle in the way of expressing ourselves correctly and making ourselves understood. If [ want to speak of a concept, language with an almost irresistible force compels me to use an inappropriate expression which obscures—I might almost say falsifies—the thought. One would assume, on the basis of its analogy with other expressions, that, if I say ‘the concept equilateral triangle’, 1 am designating a concept, just as I am of course naming a planet if [ say ‘the planet Neptune’. But this is not the case; for we do not have anything here with a predicative nature. Hence the Bedeutung of the expression ‘the concept equilateral triangle’ (if there is one in this case) is an object.We cannot avoid words like ‘the concept’, but where we use them we must always bear their inappropriateness in mind. From what we have said it follows that objects and concepts are fundamentally different and cannot stand in for one another. And the same goes for the corresponding words or signs (loc. cit., PW, pp. 119—20).

Or, as Frege says in ‘On Concept and Object’, the three words ‘the concept “horse”” do designate an object, but for that very reason they do not designate a concept (PW, p. 94 column 2) b A

and then again later: In logical discussions one quite often needs to assert something about a concept, and to express this in the grammatical form usual for such statements, so that what is asserted becomes the content of the grammatical predicate. Consequently, one would expect the concept to be the content of the grammatical subject; but the concept as such cannot play this part, in view of its predicative nature. It must first be converted into an object, or,

speaking more precisely: an object that is connected with it in accordance with a rule must be substituted for it, and it is this object we designate by an expression of the form ‘the concept x’. (Cf. p. x of my Grundlagen.) So the phrase ‘the concept horse’ must be regarded as a proper name which can no more be used predicatively than can, say, ‘Berlin’ or ‘Vesuvius’ (loc.cit., PW,

pp- 97-8 col. 2).

6. In response to all these apparent difficulties and in the hope of simply avoiding them, let us begin by distinguishing at the outset between (I) the bare noun “horse”, seen as standing for the reference of a concept-word “horse” as in Frege’s diagram; (II) “is a horse”, the essentially predicative ordinary-language form which we are to try to understand here in Fregean terms; and (III) the formal-cum-explanatory counterpart of (II) namely “horse ()", understood as analogous to “( ) squared” or “()"”. Even where the parenthesis “( )" is omitted, it is to be understood as present. (No doubt 29

”»

this controls what can be said of concepts.)

CONCEPT

AND

COPULA

37

When “horse ()" is supplied with a singular term such as “Victor” or “Shergar”, what we then obtain is a sentence which scores either the True or the False. But in the framework (III) and when it is adjoined to the brackets “( )”, what is the reference of “horse”? Well, see again the third

column of Frege’s diagram for Husserl.As it stands there and as it stands in the context “horse ()”,“horse” stands for something that has its place in the same ontological realm as the objects that fall under it. We ought though to refuse to categorize the concept horse (in the way in which Frege does in one of the passages just quoted) as an object. For the work of “horse” is to subsume such objects as Victor or Shergar.'* In that role, already signalled by the parenthesis, “horse” stands for something categorically different from any object. It stands for what Victor or Shergar are.” 7. What then of the copula? It is simply the counterpart of the bracketing conventions that appear indispensable to any readily imaginable notation for functions. Thus one likens the bracket to what Alonzo Church calls an improper symbol.'® Such symbols serve a purely grammatical purpose. They structure the sentence that expresses the thought. They show how the word thus bracketed is to be construed. They dictate its role. Could any other proposal have pleased Frege better or more eftectively discredited the assimilation of the copula to a relational term? For Frege’s aversion to that assimilation see (for instance) the following: In the sentence “Two is a prime’ we find a relation designated, that of subsumption... This creates the impression that the relation of subsumption is a third element supervenient upon the object and the concept. This isn’t the case . .. the object engages immediately with the concept without need of special cement. Object and concept are fundamentally made for one another, and in subsumption we have their fundamental union. (PW, p. 178, italics added.)

Faced with this or similar statements, I can imagine that a friend of the copula might reply by saying “Frege begins by calling subsumption a relation”. Yes, | say, but then he denies that any ‘special cement’ is needed to secure the ‘fundamental union’ of subject and predicate. Here the friend of the copula may 14. For the same reason we should refuse to count “equilateral triangle” as the name of an object. 15. Cp. Dummett FPL p. 213. 16. See Introduction to Mathematical Logic, vol. 1, p. 32.*“In addition to proper symbols there must also occur symbols which are improper. .. which have no meaning in 1solation but which combine with proper symbols...to form expressions that do have meaning in isolation. Conspicuous among improper symbols are brackets of various kinds. . . employed to show the way in which the parts of [such] expressions are associated.” (Contrast Russell’s quite different use of the words “improper symbol™.)

38

CONCEPT

AND

COPULA

object:“But how does it help—what can it achieve here—to mention a complete thing and an incomplete thing and place the two mentions together ‘as a predication’? And what help is it—what difference can it make—to insist that the one mention is put into a gap within the other mention?” The right reply to such doubts is not (I think) to deny but to insist that, when I utter “Arkle is a horse” or “Horse (Arkle)”, the subject-word and the predicate-word, with their senses, are indeed, as a result of my effort, placed in

a certain relation—the relation well enough signalled by the parenthesizing schema. But, when the words are placed so, as in our explanatory scheme (III),

for instance, the result is not for the speaker to say that“Arkle” and “horse” stand in a relation. Rather, invested as each word is with the wherewithal to make its

own distinctive contribution, what the subject and the predicate (these 1 in their relation) then combine to say is that Horse (Arkle), i.e. that Arkle is a

horse. That is what the speaker contrives for the complete thing and the incomplete thing to combine to help him or her to say. It is the semantics of the language which provides that in virtue of which such words can combine to do this. The parenthesizing schema, signalling the way in which the parts are to be associated within the whole, is one of the things that make this possible. 8. Perhaps there is something we can still learn here from one of the earliest accounts of these matters, namely that given in Plato’s Sophistes at the point where, having precisely distinguished a sentence from a list, Plato writes: “When verbs are mingled with nouns, then the words fit together and the simplest combination of them constitutes language . . .When anyone says/utters ‘[such and such a] man learns’. .. he not only names but also achieves something (ti perainei) by connecting verb (rhema) with noun (onoma)” Sophistes, 262C-D.

In order to achieve the saying of something we have to do something else, namely place verb and noun in a recognizable array—the array we can formalize as (predicate(subject))

or ((subject)

relation-word

(subject))...The

array

itself needs no semantics.All that array does is to display in their respective roles the words that have semantics. It designates them, and shows them at work on

behalf of him or her who places a name within the predicative frame."’ 17. Some may complain that, in so far as this defence achieves anything, it covers only the simple present or timeless usage of the copula and ignores the fact that in natural language the copula or the termination of a verb 1s the locus of tense, aspect, and mood.To this I shall reply—programmatically and provisionally—that, once we understand the subsumption of a singular zero-level subject within a level-one predicate, we can seek to understand other uses as syntactical modifications or adaptations of a fundamental time- and mood-unspecified mode of combination. (Compare A. N. Prior’s way of building past and future upon a timeless, only presumptively present, mode of predication.) Aspect, by the way, 1s a further matter. It impinges on the shape of the verb 1tself.

CONCEPT

AND

COPULA

39

9. Implicit in almost everything already said is a general principle: if we apply a predicate to some subject (object, concept or whatever), there must be some sort or kind within which the subject in question can indeed belong. Thus, when I say that Arkle is a horse, “Arkle” will stand for a levelzero thing that everyone is happy to call an object and horse must be a levelone thing-kind—better, a concept—that the object Arkle belongs within or instantiates it.'® Applying an analogous principle to the concept horse where the subject, the concept horse, is first-level, the only concepts that horse itself (this concept

itself)

can instantiate

are second-level

concepts.

On

these

terms, horse can indeed be said to be a concept. But the further concept which will come into play when one says that horse is a concept is the concept of a concept. Or so it must seem. Consolidating such a multi-level approach, one must distinguish among possible Bedeutungen, between (1) things which cannot be predicated at all, the zero-level things that Frege was happiest to call objects; (2) things we can talk about and predicate of zero-level particulars, namely first—level concepts; and (3) second-level concepts, which first-level concepts (horse etc.) can instantiate or exemplify or fall within. When we say that the first— level concept horse is a first—level concept, it seems that we have to call upon a second-level concept— the concept of a concept—to say this. In its intensional form, Russell’s paradox suggests exactly the same. (Is non-selfpredicable non-self-predicable or not? If it 1s, it isn’t. And if it isn’t, then it

is!. . The question itself must be ill-formed.)

10. Before we conclude, there 1s another quite different sort of question to consider. For the implications of our exploration impinge upon metaphysics and the philosophy of science. [t is often said that a thing’s properties explain its behaviour. What is the relation of this familiar use of the idea of a property to the Fregean idea of a concept? The matter may seem obscure.We speak of the property of being strong or clever or capable...We do not speak of the concept of being strong or clever or capable. We speak of having or possessing these properties. We do not speak of having or possessing such (Fregean) concepts. But rather than engage with these niceties, let us begin with a real example. Elliot Sober argues that evolutionary theory as we now have it generalizes over properties, claiming that properties as construed by that theory in 18. We

have already seen that Frege

tion the concept

himself sometimes

deploys the converse idea that in predica-

must be such as to subsume the subject. If horses exemplify or instantiate the

concept forse, then the concept subsumes or comprehends each and any horse.

40

CONCEPT

AND

COPULA

its present condition cannot be redefined in terms of physical objects or sets of material objects.”” He then declares: What we now need to ask in our discussion of evolutionary theory is this: Are the property generalizations which one finds there mere placeholders— admissions of ignorance—which will be eliminated as we learn more about evolutionary processes? Or are they part of the substance of evolutionary theory, which further information may elaborate, but never leave behind? My judgment about this question is that evolutionary theory as it matures will embrace such generalizations in greater numbers. The development of the theory to date exhibits a proliferation of property generalizations, and this encourages the view that it is part of the point of evolution theory to codify generalizations about what kinds of properties will be selectively advantageous in what kinds of environments. (p. 169)

In the light of this claim let us consider just one evolutionary generalization of the kind Sober considers. In his formulation it goes: for any property P, if there is selection for greater offspring variance-in-P, and if sexual individuals are more likely [than non-sexual ones] to produce offspring with greater variance-in-P, then sexual individuals will be favoured by selection. (p. 168) Sober represents this as a quantification over properties. Suppose that, in the

same spirit, we say that sexual animals have an evolutionary advantage because they are more likely than asexual animals to produce offspring that vary in size (or colour or shape or strength or...).What must Sober mean by a

property? And what sort of thing is an object’s size (or colour or shape...)? Well, size (or colour or shape...) is not a Fregean concept. But the claim is clear as soon as we can understand it like this: sexual creatures have an evolutionary advantage over asexual creatures because they are more likely than any others to produce offspring that vary in respect of how big they are. The general form for the sort of explanation Sober has in mind seems then to be: Sexual individuals are favoured by selection because, for all ¢ such that there is selection in favour of parents whose offspring vary in respect of how ¢ they are, sexual individuals are more likely to produce offspring with greater variance in respect of how ¢ they are.

Here in this paraphrase let us read the generalization ‘for all ¢’ as a quantification over Fregean concepts. As usual, the predicative array is present in

19. ‘Evolutionary Theory and the Ontological Status of Properties’, Philosophical Studies 40 (1981).

CONCEPT

AND

COPULA

41

the open sentence quantified upon. It is ready and waiting to make an unsaturated and predicable expression from the values of the variable. In so far as properties seem obscure, we can understand them by seeing them as abstractions deriving from nominalizations based upon families of ordinary predicates that stand for common-or-garden Fregean concepts.We may not want to take seriously the generalizations over properties such as size or shape or colour. .. that serve Sober’s purpose. But, if we do want to dispense with such abstractions or we find ourselves puzzling how to replicate for colour or shape or...the easy transition from “size” to “big”, then a good thing we can do (or so I aver) is to look always for their grounding in the ordinary predicates that stand, in the ways we have set forth, for commonor-garden Fregean concepts.

3 Donald Davidson’s Account

of Semantic

Interpretation. How

Comprehensive 1s it? ‘All’, ‘Some’, and ‘Most’ Summary In a remarkable series of papers Donald Davidson oftered a would-be general account of how the sentences of a language can have the values of truth or falsity and thereby attain meaning. Some of these sentences, moreover, can be singled out (he says) as “structurally valid”. One major part of this account relates to the quantifiers “some” and “all”. Unluckily for any larger aspirations Davidson may have had to generality, the treatment he proposes cannot be applied to other quantifiers such as “most” and “few”. These invite another approach, but that presupposes a basic semantical form—a binary form—entirely different from that upon which Davidson based his own proposals. Inevitably questions arise

here about the very idea of a universal grammar applicable somehow to all human thought or speech.

According to Donald Davidson’s conception, the proper understanding of any expression that counts as one of the words of a language L may be registered by an axiom that is part of a theory with the following properties.

Meaning, Truth, and the Limits of Analysis: Ten Studies. David Wiggins. Oxford University Press (2022). © David Wiggins. DOI: 10.1093/0s0/9780198726173.003.0003


Acts or doables are 1. Even if this thought receives no assistance from the symbolism of modern formal logic, 1t has worn reasonably well and received due recognition in our own times from P. E Strawson 1in Part Two of Individuals, where sortal universals and characterizing universals are distinguished, and from W. V. Quine at §15 of Word and Object. Both treatments are anticipated by Frege at Foundations of Arithmetic §s4. 2. See Donald Davidson, ‘The Logical Form of Action Sentences’ in N. Rescher (ed.), The Logic of Decision and Action (University of Pittsburgh 1967). Davidson cites an earlier treatment of these subjects due to Reichenbach (Elements of Symbolic Logic, Macmillan, New York 1947). For criticism of Davidson’s theory see P. E Strawson, ‘On Understanding the Structure of One’s Language’, in Gareth Evans & John McDowell (eds), Truth and Meaning: Essays in Semantics (Oxford: Oxford University Press, 1970). 3. This term was introduced into the theory of action by Jennifer Hornsby. For the distinction of act and action, see her Actions (London: Routledge & Kegan Paul 1980) especially Chapter 1, Section 2.4;‘Action and Ability’ in R. Haller (ed.) Sprache, Logik und Philosophie: Proceedings of the

66

GRAMMATICAL

COMBINATION

the level-one references of ordinary verbs of agency.® They are not levelzero things such as events.

II In our epigraph, why does Davidson want to see “Shem kicked Shaun” as an existentially quantified proposition? To understand his reasons we may perhaps begin from E P. Ramsey’s claim that the proposition “Brutus kissed Caesar”’ needs to be understood as asserting the existence of an occasion or event of a certain sort.” It is necessary and sufficient for the truth of such a sentence that there should have been some occasion or event that consisted in Brutus’ kissing Caesar. How then should we think of verbs such as “kissed” or “kicked” or “buttered”? Should we think of them as carrying along with them a slot or argument-place (or several such) reserved for a time or a time and a place, or a time and a place and a manner, or...? One potential difhiculty with such suggestions is that it may be hard to know where to stop supplying extra argument-places.As Anthony Kenny says: If we cast our net widely enough, we can make “Brutus killed Caesar” into a sentence which describes, with a certain lack of specification, the whole history of the world. (Anthony Kenny, Action, Freedom and Will p.160)

But the Davidsonian response to a challenge such as Kenny’ is to steer a middle course and, given a simple sentence with subject x, object y,and verb V, to assign argument-places within the frame ‘V/...° to its subject and its object or objects (if any), and then supply one other argument-place—this is Davidson’s innovation—for a variable that ranges over actions/events. See (17), as displayed in the epigraph.There is then the option to conjoin further separate

conjoined

clauses

relating

to

time,

place,

instrument,

manner...,

where each of these clauses lies within the scope of the aforesaid action/ event variable. Here is the rationale that Davidson gives for this approach. If I say I bought a house downtown that has four bedrooms, two fireplaces, and a glass chandelier in the kitchen, it’s obvious that I can go on forever adding Fourth International Wittgenstein Symposium at Kirchberg 1979 (Vienna: Holder-Pichler-Tempsky 1980) and see her later exposition in Tsohatzidis S. L. (ed.), Foundations of Speech Act Theory: Philosophical and Linguistic Perspectives (London: Routledge 1994). I take the term “doable” from these and other works of Hornsby'’s. 4. As for the 1dea of agency itself, its philosophical elucidation belongs elsewhere. But see for instance Davidson’s Chapter 3 in the same collection of essays referenced in note 2. s. See E P. Ramsey, Foundations of Mathematics (London: Routledge & Kegan Paul, 1931) pp. 138 f.

GRAMMATICAL

COMBINATION

67

details. Yet the logical form of the sentences I use presents no problem (in this respect). It is something like “There is a house such that I bought it, it is downtown, it has four bedrooms...” and so forth.We can tack on a new clause at will

because the iterated relative pronoun will carry reference back to the same entity as often as desired...Much of our talk of action suggests the same idea: that there are such things as actions, and that a sentence like ‘Jones buttered the toast in the bathroom with a knife at midnight’ describes the action in a number of ways. ‘Jones did it with a knife’ ‘Please tell me some more about it’. The ‘it here doesn't refer to Jones or the knife, but to what Jones did — or so it seems. Austin writes . . .1t is in principle always open to us, along various lines, to describe or refer to “what I did” in so many different ways’. Austin is obviously leery of the apparent singular term, which he puts in scare-quotes; yet the grammar of his sentence requires a singular term. (Donald Davidson, Essays on Actions and Events, pp. 108—9, quoting from J. L. Austin, ‘A Plea for Excuses’

’

Philosophical Papers (Oxford: Clarendon Press (1967) p. 148).)

The first of these two paragraphs (we shall come to the second later and in due course, see VI) proposes or shows forth a parallel between the way in which Davidson’s house downtown can collect all sorts of different attributions and the way in which one and the same dateable action or doing— Shem’s kicking Shaun, for instance—can collect all sorts of attributions, temporal, instrumental, or other. By way of illustration, consider (17) and then amplify and extend it as follows: (17A)

(dx) | (kicked Shem, Shaun, x) & (with a hobnail boot, x) & (on the shin, x) ].

This spells out conjunctively and one by one the various adverbial qualifications which come into play in this action-sentence. There will be room for plenty more just as required. (17A) says in the Davidsonian way that Shem kicked Shaun on the shin, with a hobnail boot.

[t is noteworthy that, simply by virtue of propositional logic, (17A) will imply (17).This is a point which Davidson and his successors have been apt to lay some emphasis upon. How better, they ask, could it be explained why ‘Shem kicked Shaun, with a hobnailed boot, on the shin’ logically implies ‘Shem kicked Shaun’? (But Davidson’s way of managing this need not be the only way—or so we shall find in Part VII.)

II1 To judge by the extant literature, a majority of philosophers and philosophically inclined linguists have wanted to accede in one way or another to

68

GRAMMATICAL

COMBINATION

Davidson’s general approach. But not all have. In his essay “On understanding the structure of one’s language”, P. E Strawson strongly opposed Davidson’s theory, saying that the title of ‘plausibility, simplicity and realism’ ought rather to be awarded immediately to ‘the Adverbial Account [of adverbs, which is] to be set over against the Davidsonian Account’ (p. 189).

Strawson says of the simple Adverbial theory that he will pretend that we have it to hand. Later he says (pp. 196—7): By hypothesis, we have available an alternative account (the [properly] Adverbial) of the elements and modes of combination involved, according to which these elements and combinations are not at all unperspicuously represented by the formal arrangements of surface structure sentences as they stand.

There 1s something right here. An account of adverbs should not advance immediately into the symbolism of logical or linguistic theory. Rather, it should clarify the modes of combination that we find in natural languages by the explication and amplification of existing surface structures, allowing itself, wherever that may be necessary, “a more complicated basic syntax than a Davidsonian theorist is prepared to allow for” (p. 189). First and foremost, though, the thing that really matters is “what are these familiar words doing here?”. This is Davidson’s question.” But Strawson can compete with Davidson to answer it. [t will appear later what is at issue in the choice between Davidson’s and Strawson’s accounts of adverbial modification. Meanwhile I remark that neither Strawson nor Davidson is in a position to postulate just one single, so to speak “basic”, form of adverbial modification.® Indeed, even where we set aside adverbial constructions that specifically involve the verbs of agency and we focus upon examples that might seem to belong exclusively on Strawson’s side of the fence, we shall find a rich plurality of modes of combination—modes of combination which in some cases make ontological demands analogous to those that Davidson and his adherents have postulated for the case of action-sentences.

6. In Gareth Evans and John McDowell (eds), Truth and Meaning: Essays in Semantics (Oxford: Oxford University Press 1970). 7. See, for instance, his ‘On saying that’, Synthese 17, pp.130—46. 8. If there were some one (so to say) ‘standard mode of adverbial combination’, what would it be? In search ofit one might look first to the scheme x is [ADVERB] such that it s PREDICATE. (For the versatility and colourlessness of ‘such that’, seeW. V. Quine at Ways of Paradox pp. 275-82.) This works well enough for ‘scarcely’, ‘necessarily’, ‘often’, ‘usually’, ‘nearly’ but not for the examples (b), (c), (d) that I shall enumerate, and still less for the adverbial cases to which we shall return in due course, ‘with a knife’, ‘in the bathroom’, or ‘greedily’.

GRAMMATICAL

COMBINATION

69

(a) Consider first ‘James was wonderfully kind’. Does this sentence require for its truth anything besides James? Does it make implicit reference to the way in which James exemplifies the property of kindness? Can the mode of combination itself force such a conclusion upon us? Maybe not. Perhaps we should try something else. Trying and then rejecting the doubtfully satisfactory proposal ‘James was wonderfully such as to be kind’, we are pointed immediately in the direction of the obvious fact that what the ‘wonderfully’ does in ‘wonderfully kind’ (contrast the unsatisfactory proposal) is to apply to how kind James was. ‘It was wonderful how kind James was’, we may say. And this, if we think we are committed independently and in any case to degrees, we may perhaps parse as ‘It was wonderful to what degree James was kind’. Compare for similar treatment ‘Her house was perilously close to party H. Q., which seems to mean that it was perilous (for someone or other) how close

her house was to party H. Q. (b) Consider now ‘The cat was fast asleep’” or “The ship was fast aground’. At least for one inclined to go along with Davidson’s readiness to invoke new ontology, there is nothing particularly outrageous in the idea that we should see such an adverbial coupling as saying something about the fixity of a state of being asleep—or aground—attributed to some subject of discourse.” Do such forms presuppose an ontology of states? I am not

sure.

(c) Consider the sentence ‘At this point in the conversation, where laughter threatened to break upon the scene, Dean Swift remained ironically silent’. How do the words work in this combination? Perhaps what is signified is that Swift’s silence or grave demeanour was feigned, or purposely aftected. Taking fright though at the threat of an ontology of demeanours, and avoiding the unilluminating ‘Swift was silent in an ironical manner’ (however exactly that would work), someone may now say that what it comes to is that Swift purposely affected to be grave. If that is right, then in a longer enquiry we should need to look for the general recipe for making such a transposition.We should need a principle and some more examples to test it by. 9. The OED says that in the phrases fast aground, fast ashore, fast asleep, ‘fast seems to have been originally the grammatical predicate; now it 1s usually apprehended as an adverb qualifying aground, ashore, asleep’. If I understand this, it suggests that originally ‘fast asleep’” was apprehended as meaning something like fixed asleep. Compare “hammered flat” (=? so hammered as to be flat, hammered in such a way that it 1s flat).

70

GRAMMATICAL

COMBINATION

(d) Consider the predicative combinations ‘grossly unjust’, ‘coarsely kind’, ‘humbly fearful’, ‘numinously wise’. If there is a common pattern here, then what we seem to have to say in order to explain it is that in each case a property—of injustice, kindness, fearfulness, wisdom (or the state or disposition of having some property)—is specified, and then it is described, or else some particular way of having the property (or some determination of it) is singled out. Thus the injustice of x is gross (‘of conspicuous magnitude, palpable, striking’ as the Oxford Dictionary puts it); x’s fearfulness is humble (? expressive of humility, or informed or sustained by humility); his kindness is coarse (or admixed with coarseness); his wisdom

1s numinous

(i.e. not narrow or restrictive or

myopic, but possessed of a vast universality). This is not yet, I realize, an argument for e.g.‘he was wise’ fout court requiring for its truth the existence of wisdom. But does not the intelligibility of the combination with the adverb seem on the face of it to suggest the availability of an item such as wisdom lurking within even the simplest predications? Finally we may consider colour predication, where nominalists have taken themselves to be in the securest position of all. But, in considering modes of combination such as that which we encounter in “The sea was ultramarine blue’, let us remind ourselves that (as Jonathan Westphal has insisted'’) we predicate colours not only of objects but also of colours themselves. Ultramarine is blue; but scarcely in the way in which the sea is blue. But this isn’t to say that ‘blue’ is ambiguous. It plainly isn'’t. Maybe “ultramarine is blue” means that ultramarine is a shade of blue. But how is this claim, in which ‘blue’ appears to be the name of a colour, to be brought into some intelligible relation with ‘The sea is blue’, where ‘blue’ is predicative, and then with “The sea is ultramarine blue’, where ‘blue’ not only combines with ‘ultramarine’ to designate an ultramarine shade of blue (or so it appears) but is also predicative? Only one systematic solution suggests itself. “The sea is blue’ means that the sea is of blue, or is of some shade of blue, where ‘blue’ is still a name of a colour: and then ‘The sea is ultramarine blue’ means that the sea is of an ultramarine shade of blue. Pending the discovery of another approach to the problem, it appears that, contrary to habitual philosophical expectations, what we are speaking of when we say how things are coloured is colours 10. See Jonathan Westphal, Oxford 1991.

Colour,

A Philosophical

Introduction

(2nd

edition), Basil

Blackwell:

GRAMMATICAL

COMBINATION

71

and shades themselves. Let the nominalist show how to reduce the commitment we seem to have incurred to recognize the actual existence of colours and their shades.

What do these examples achieve? They show forth the variety of constructions and the variety of ontological questions with which seemingly ordinary adverbs can confront us—not only for the case of the adverbs that figure in the particular verb and adverb constructions to which Davidson has pointed (we shall come to these) but even within the province that Strawson wanted to regard as straightforward and illustrative of how adverbial modification ought quite generally (he suggested) to be understood. On this basis I venture to say that, pace Strawson, there need be nothing wrong or misguided in principle with a theory of action-sentences such as Davidson’s and nothing wrong with the claim that even the simplest actionsentence involves quantification over events. The question is whether they really must (or can) involve that. Pace Davidson however, it will appear in the end how doubtful it is that they can.

IV In order to advance any further we need to ask how well we understand Davidson’s sentence (17). (17) says that (3x)( kicked (Shem, Shaun, x) ). It

says that Shem kicked Shaun something. But what can this x, this something, be? An event/action? But how can Shem kick an event or an action? [t was Shaun that Shem kicked. Here is what Davidson says in order to explain (17): The basic idea is that verbs of action — verbs that say ‘what someone did’ — should be construed as containing a place for singular terms or variables that they do not appear to contain. For example, we would normally suppose that ‘Shem kicked Shaun’ consisted in two names and a two-place predicate. I suggest though that we think of ‘kicked’ as a three-place predicate and that the sentence will be given in the form (17) (3x)( kicked (Shem, Shaun, x) ). If we try for an English sentence that directly reflects this form, we run into difhculties. “There is an event x such that x is a kicking of Shaun by Shem’ is about the best I can do, but we must remember that ‘a kicking’ is not a singular term. (op. cit. p. 110)

But proposition (17A), which was introduced at the beginning of Part II, brings out a difficulty in this proposal. How can an event or even an action

72

GRAMMATICAL

COMBINATION

be ‘on/in the shin’ or ‘with a hobnail boot’? It doesn’t make sense for Shaun to say to Shem “Your action (what you did) was on the [my] shin”. Davidson does not say enough about this sort of question or offer any general way of alleviating the discomfort that it represents. Indeed even his paraphrase of (17) doesn’t force us to see the verb “kick” as a three-place predicate. But we can cast around for help from other English locutions that may seem to bear upon this matter. There is at least one that deserves note. Begin with ‘There is something Shem kicked Shaun’ and go back to the sentence

(17). If (17) is true then perhaps we

can say that Shem

kicked

Shaun a kick. If Shem kicked Shaun viciously, then he kicked him a vicious kick. Shem may have kicked Shaun in the shin. If so, then the kick was in (or on) the shin. Shem may have kicked Shaun with a hobnail boot. If so, the kick was with a hobnail boot.This helps to vindicate (17A), does it not?

For now at last we have something that makes sense. How can we generalize from this success? Verbal nouns such as “a kick” occurring in this way are familiar enough. They were treated by traditional grammarians under the heading of the “internal—or cognate—accusative”. Once we search, it may be suggested, we shall find a whole host of similar (however various) examples: ‘dance a dance’, ‘die a [happy/miserable/wretched] death’, ‘swear an oath’, ‘fight a war’, ‘laugh a loud laugh’, ‘win a [Pyrrhic/resounding/expensive| victory’, ‘plough a ploughing’, ‘battle a battle’, ‘dream a dream’, ‘walk it [walk a walk, walk a journey such as others have walked before] all the way from Cambridge to London’."" Even more conspicuously and in closer parallel with “Shem kicked Shaun a kick”, we have “Clytemnestra struck Agamemnon a mortal blow [with such and such weapon]”. Where nothing very like any of these is at all obviously available or familiar, should we try to supply the lack? Take the verb “butter”, for instance. Can we not hear the cognate accusative at work in “the agent butters (or buttered) the toast a [good] buttering”? Cannot the connection between agent and action be provided in this very way by the verb “butter” itself? Maybe we can generalize this strategy. If we can, then the internal accusative construction provides a ground-leve] (level-zero) item to play a role analogous to that of Davidson’s town-house.

Shem’s

kick,

Clytemnestra’s

blow,

Graeme’s

walk,

Scipio’s

11. Where verb and noun matched, the traditional name was “cognate accusative”. On these matters, see E. A. Sonnenschein, The Soul of Grammar (Cambridge: Cambridge University Press, 1927) §23. See also Gildersleeve and Lodge, Latin Grammar (third edition, Macmillan, London 1895) §332—4 or Goodwin, A Greek Grammar (Macmillan, London 1894), p. 223.

GRAMMATICAL

COMBINATION

73

dream, Carr’s catch...can these not be the sorts of thing that Davidson’s variable x ranges over? Aren’t these the sorts of thing that can be ‘with a hobnailed boot’,‘in the shin’... ‘with a knife’, ‘at midnight’, ‘greedy’. .. In so far as the cognate/internal accusative is needed in order to secure our understanding of (17), (17A), and the like, it seems strange that Davidson

never invoked it or even mentioned this construction. If he had laid any store by it, though, then it might have troubled him that, in English as it stands, it is not always available—or not always nameable. It wants reflection how much this matters. But it really must matter. For sentences such as (17) or (17A)—and any generalization we formulate on this pattern—can hold true

only where there is a value of x and a corresponding accusative for that value of x such that all the various conjuncts involving x will work in the natural and intelligible way in which ‘in the shin, x’ and ‘with a hobnail boot, x’ work

for kick. They must all make sense. Only in the presence, actual or correctly assumed, of such a cognate accusative can propositions such as (17A) have their intended meaning.'” Davidson’s proposal comes at a high price.

V We shall return in the end to a Strawsonian alternative to Davidson’s whole approach. First though let us attend to one or two further questions which seem central to Davidson’s concerns. Davidson writes constantly of actions as events. This is integral to his philosophy of mind and action. Suppose then that we go along with the ‘butter a buttering’ model for framing a cognate accusative wherever one is needed. That suggests that, when Jones buttered his toast greedily, hungrily, angrily, the buttering he buttered was a greedy, hungry, angry action. But was it therefore an irritable, greedy or hungry event?” Actions are indeed events but greedy actions are not greedy events. If this strikes us as a difhculty for Davidson, though, then compare the way in which we could say of a solicitor that she was learned, well-read, and generous with her time

without committing ourself to say that she was a learned, well-read substance, 12. These are formidable requirements. Could their force be tempered by calling upon some general theory of prepositions? The strength of the motive to explore this possibility depends perhaps on whether an altogether different approach to action-sentences, a real alternative to Davidson’s, 1s available. For one such alternative, see VI below. 13. See here Helen Steward in ‘Actions as Processes’, Philosophical Perspectives 26 (Philosophy of Mind), 2012, p. 376, where

she points to the absurdity of the very idea of a greedy event.

74

GRAMMATICAL

COMBINATION

generous with her (its?) time. Persons are indeed substances, but ‘substance’,

being a determinable and standing as it does for a formal concept, consorts ill with ordinary/everyday predication in the category of (specific) quality. [t would be good to have a more complete understanding of such matters. The paralle]l suffices however—or so I submit—to protect the claim that, just as persons are substances—and in just the same way as persons are substances—so, as Davidson says, actions are events.'* So much for the first (and

provisional, indeed temporary) defence of the Davidsonian approach. But the time is come

to consider alternatives.

VI What is the alternative would-be Strawsonian account of the actionsentence, with and without adverbs? The scheme which most naturally suggests itself is as follows: [subject + active verb + inflection or copula (such as “do”) + object + adverbial qualification].

Quantification can be built up from this familiar scheme but no quantification is hidden within the basic or simple scheme. For the case of an action-sentence 14. The point that emerges here needs to be collated with a point about the phenomenon of attributivity. Suppose two burglars who are planning to break into a house on a hillside have agreed that

one of them will keep watch from the road while the other enters the house, and the signal for danger and immediate escape will be that the sentinel should instantly walk uphill, up the road. Otherwise he i1s to stand still, binoculars 1n hand, as if interested in the wildlife he can observe there. Suppose

the burglars

are about

to be surprised

and, seeing this, the one

keeping

watch

makes the agreed signal. He gives the other burglar a warning by walking uphill up the road. His action of walking uphill 1s one and the same as his action of warning his partner of the need to escape. (In “the sentinel’s action of walking uphill”, the preposition “of " is to be construed as it 1s 1n “the city of Paris”. The city of Paris 1s simply Paris.) So far so good. How though 1s Davidson to prevent a critic from exploiting this identity to obtain the strange assertion He walked a warning that was uphill? And what 1s there then to prevent us from exploiting the facts of the case and i1dentity of the walking and the warning to obtain the surely nonsensical He warned a warning that was uphill. or the even more

nonsensical He

warned

uphill?

All this we can escape, however, by nsisting that the verb and its adverb cling together in the

characterization of the event that consists of walking uphill. Then we can say that the sentinel walked a walking uphill and the walking uphill was a warning, yet refuse to infer that it was a

warning uphill. See Jennifer Hornsby, Actions pp. 6—10 for further discussion of this sort of case.

GRAMMATICAL

COMBINATION

75

in English, either there will be an inflection of the verb (an inflection sometimes invisible or null, where the English infinitive and the English plural form happen to be the same word), or else the copula “do” will serve the same purpose as inflection. (“Rough winds do [or did] shake the darling

buds of May.”)" According to this proposal, things are just as they seemed to be before Davidson came along proposing to read “what Jones did” and the “it” that we find in “please tell me some more about it” as referring to an action that Jones did = an event Jones was involved in, Jones’s buttering of the toast, for instance. Davidson’s

alternative, however, once

spelled

out in full, cannot help but lead him into a complication that he never contemplated and might not have welcomed—the cognate accusative, I mean. (See Part IV.)

The alternative favoured by an advocate (such as I am) of the simpler scheme is to find the answer to the question “what did x do?” as specifying the act that the verb stands for within the simple or basic scheme. Davidson himself would have had little use for the simple or basic scheme and even less for the idea that “that which I did” refers not to an event or action but to a doable such as [to] kick or [to] butter.'® But, having seen the difficulties of generalizing the cognitive accusative approach, let us persist.

At this point the importance becomes newly apparent of the disagreement that Davidson finds between himself and J. L. Austin. For convenience let us Jook again at the second half of the long citation from Davidson that was given in Part II: ...|A] sentence like ‘Jones buttered the toast in the bathroom with a knife at

midnight’ describes the action in a number of ways. ‘Jones did it with a knife’ ‘Please tell me some more about it’. The ‘it’ here doesn’t refer to Jones or the knife, but to what Jones did—or so it seems. Austin writes .. .1t is in principle always open to us, along various lines, to describe or refer to “what I did” in so many different ways’. Austin is obviously leery of the apparent singular term, which he puts in scare-quotes; yet the grammar of his sentence requires a singular term. 15. The verb “do” does, of course, have other roles besides that of serving as a copula. Consider “what did you do then?”—where the second occurrence of “do” holds a place for an ordinary plain verb, itself inflectable or open to copula. What I did was walk. 16. See Davidson’s response to Jennifer Hornsby in E. L. Hahn (ed.), The Philosoply of Donald Davidson, The Library of Living Philosophers, volume XXVII. Peru, IL: Open Court Publishing Company (1999).

76

GRAMMATICAL

COMBINATION

Let us begin here with ‘singular term’. If “what he did” is a singular term, must it stand, as Davidson supposes, for some action or event? Or can it stand for what Hornsby calls a doable? [t 1s not easy to know for sure what Austin himself would have said. But consider the not implausible claim “Before he was murdered, what Caesar was intending to do was to rewrite his will”. Here there is a ‘what’, but it cannot stand for any particular dateable doing such as Davidson had in mind. For one who makes that claim about what Caesar was intending to do knows perfectly well that, in Caesar’s case, there wasn’t ever going to be such a rewriting. But the speaker’s claim is not thereby made false or void for lack of reference. The thing there is to talk about for the speaker (or for us) 1s (I submit) the act that Caesar was thinking to do. There was such an

act for Caesar to intend, even though he never did it. When we refer to this act or talk about it, we can employ either the infinitive ‘to I’ or the bare verb, or a gerund (the kind of gerund that stands not for a historic happening but for something doable)."”” Whichever way we decide to talk, why shouldn’t these ways of referring to it give us a singular term, a term that

stands for one thing—if not a zero-level thing then a doable?™ The general distinction between an action and an act is no less applicable, of course, where the agent is not thwarted in his or her purpose and actually does the act that is in question. There too we have the act and the doing of the act. There is “kick Shaun” and there is Shem’s doing just that—his doing that act. There are also more specific acts. There is kick Shaun in the shin and there is kick Shaun with a hobnail boot. There is butter the toast and there is butter the toast at midnight and there is Jones’s doing that act. There are acts of buttering the toast at midnight, of buttering the toast in the bathroom, of buttering the toast with a knife, etc. In the Jones narrative, the action there and then of doing any one those acts there and then may come to be counted as one and the same action as the action of doing some of the others there and then. Thus, for actions, we may want to accept Davidson’s claim of identity. The same cannot, however, apply with the acts just enumerated. The acts are different, no matter how much doing one act may involve doing other acts.

17. See note 3. 18. Is that thing a universal? Against that it may be objected that an act may be highly specific. But, in itself, that is no obstacle. ‘Specific’ and ‘universal’ are not contraries. See here R. M. Hare, Freedom and Reason (Oxford University Press 1963) p. 39.

GRAMMATICAL

COMBINATION

77

So far so good. But so far the act/action distinction leaves us short of understanding Austin’s thought that “it is in principle always open to us, along various lines, to describe or refer to “what I did” in so many different ways’ . To clear the ground in order to make sense of this claim, the new and first thing we need to notice is that “what”, in both its relative and interrogative occurrences, embraces both the singular and the plural. “She asked me what I had done when I got home. I said I had swept the floor, cleaned the stove and lit the oven. That was what I had done. I had done precisely what she had asked me to do.”"” Here “what she had asked me to do” imported several things or acts and all of these I had done, together of course with the further things that doing each of these components may have involved.When I tell my story I can emphasize this or that component act. The more interesting the story, moreover, the more point there may be in making the kind of selection and emphasis that Austin seems to envisage in the passage Davidson quotes.

At this point, if only in a farewell gesture of goodwill towards Davidson, let us look back at Part II, and try in new terms to reinstate an analogy of sorts between

the doable, what Jones did, and Davidson’s

town-house. But

here let us construe “what Jones did” not as an action but as the act he did. Recapitulating then, (1) there 1s the agent and (2) there is what the agent did (in Austin’s and Hornsby’s sense) and (in a standard case) set out to do—to butter the toast, for instance (butter the toast being a doable). Next (3) there

is the agent’s actual doing of the act or acts in question—the agent’s dateable action or effort (as one might say). Finally (4) there is the how, when, where, etc. the agent did what he did—with a knife, at midnight, in the bathroom,

etc. The suggestion we can now consider is that under (4) we may, if we wish, enumerate all the component acts the doing of which came to be involved in Jones’s doing of the act (2) of buttering the toast in the ways in which he actually did that act—in the bathroom, with a knife, etc. In how-

ever faint and abstract analogy with the way in which Davidson’s townhouse coinstantiates the mutually consistent, not necessarily independent 19. Here as everywhere I reserve the word an action that is the doing of that act. to this way of using “do”. If the action let us try to avoid altogether speaking

“do” for the case of doing an act, never for the case of On pain of confusion, [ have tried throughout to stick consists of my doing an act, then, for the sake of clarity, also in the same breath of our doing an action.

This 1s bare, shameless legislation, the reader may say. Speakers

of English frequently speak

of doing an action as well as of doing an act. [ reply: yes, that 1s a fact. But the double usage of “do” is confusing and one of the several sources (I tentatively assert) of the troubled state of the philosophy of action. Here at least, I have done my best to avoid it myself.

78

GRAMMATICAL

COMBINATION

properties of having four bedrooms, having two fireplaces and the rest, so the whole complex act that Jones did when he buttered the toast involved the following co-practicable, distinct albeit interdependent, acts: [to] butter the toast with a knife, [to] butter the toast in the bathroom, [to] butter the toast at midnight...So much we can say without any commitment either to Davidson’s conjunctive approach—to his (17), to (17A) or to his attempt

to find an event-variable at work within “Shem kicked Shaun”. About this Strawson was surely right. Strawson was right also to insist that the implications that Davidson wanted to vindicate didn’t need to be vindicated within the propositional calculus.

VII The notion of logical form as exhibited and exemplified in first-order (and second-order?) logic struck Davidson as a uniquely perspicuous model or paradigm for understanding the idea of logico-grammatical structure. The work that has resulted has cast a strong if variegated light upon a significant number of constructions and idioms of ordinary language. But, pace Davidson and his attachment to a technical and somewhat restricted conception of logical form (derived from first-order predicate logic), a theory of adverbs does not have to deploy Davidson’s conjunctive strategy in order to vindicate the obvious entailments. It is transparently obvious that one who does the complex act that Jones did must have done each and all of the acts that Jones needed or chose to engage in and all the further actcomponents of those acts. On this basis, it is open to us (or so I conclude)

to abandon altogether the Davidsonian approach to adverbs of agency, and open to us to leave aside all the problems we encountered in the context of (17) and (17A) with making

sense of “in the shin x” or “with a hobnail

boot x”. If we prefer a would-be Austinian way of seeing action-sentences over Davidson’s way, then we

can dispense with Davidson’s

(17) and dispense

with the conjunctive analyses into which that leads.We shall say rather that in “Shem kicked Shaun” we have a subject and an object and the verb stands for something doable, namely the act of kicking. For the verb stands for this doable and its presence, suitably inflected (or coupled with the quasi-copula “do”), creates the option for the utterer of “Shem kicked Shaun”, thus understood, to mean something by uttering the words “Shem kicked

GRAMMATICAL

COMBINATION

79

Shaun”. Adverbial qualifications qualify the verb and enable the utterer to say when, where, how, etc. the agent did the act. If there is a preferred order

in which these qualifications should be applied indicating how, where, when, for instance, the act was done—they can be applied in that order. (Contrast Davidson’s simply conjunctive approach.) The act is done and the actuality of the doing of the act is indicated by the inflection etc. of the verb that introduces some simple or complex doable.Where someone says “please tell me some more about it”’, what they want to know about is the act or acts that were involved in doing the act or acts that the agent is being said to have done. Contrast the Davidsonian theory.And compare the Strawsonian account, as sustained by Jennifer Hornsby’s conception of the doable.

VIII Between two different theories here of action-sentences and two theories of adverbs that qualify verbs of action, we have in the end preferred a quasitraditional one over Davidson’s. But nothing that has been said in this chapter will count against Davidson’s altogether admirable general concern to say “what these familiar words are doing (here, there, or wherever it may

be)”. Nor have we encountered any disadvantage obstructing the general method of explication in schematic English, as expanded whenever that serves some simply clarificatory purpose, by the use of logical notation. But about that a word or two more. Bertrand Russell wrote, on page 168 of Introduction to Mathematical Philosophy: Misled by grammar, the great majority of those logicians who have dealt with [the] question [of descriptions| have dealt with it on mistaken lines. They have

regarded grammatical form as a surer guide in analysis they have not known what differences in grammatical met Jones’ and ‘I met a man’ would count traditionally same form, but in actual fact they are of quite different

than in fact it is. And form are important. ‘I as propositions of the forms.

What we should notice here is Russell’s unjustified slide, copied since by countless philosophers, between logico-grammatical form on the one hand and, on the other, grammar and grammatical form as mistakenly conceived. Traditional grammar was mistaken in its own terms in so far as it assimilated ‘a man’ to ‘Jones’. There is no need for the notation of Principia Mathematica to explain this distinction. It is enough to appeal to ideas we have already about reference. To say something about Jones—that he is bilingual, for

80

GRAMMATICAL

COMBINATION

instance—we have to be able to make it clear which person has to be bilingual for something true to have been said. Contrast a sentence only deploying “some man”, where there is no such requirement. It is true that for purposes of this explanation we need a pre-theoretical conception of reference. But formal logic itself will not really dispense with this conception either. Rather than imitate Russell’s slide and then follow his pursuit of the rainbow all the way to the idea of a logically perfect language in which apparent and real form

coincide

(or in which

surface structure, as some

now

say,

simply coincides with deep structure or logical form), let us resolve instead to arrive at a new and improved understanding both of grammar itself and of surface structure of natural language. Once we achieve that, we are less likely to be “misled” by surface structure. (There is a temptation to say that the surface can only mislead those who misperceive or misread it.) How much you can see in a place depends on what you know or can find out. Once we take this thought seriously, we shall find ways to direct Davidson’s good question “What are these familiar words doing?’ onto what is potentially or actually visible in language—uvisible to a knowing eye, that is. Once this possibility is envisaged, and once we see that there is really no theoretically neutral account of surface structure (unless as a bare enumeration of words

in their order of occurrence)—and

once “deep

structure”

ceases to be seen by contrast as the exclusive preserve of a mysterious or implicit (speaker) knowledge—there will be a chance to resolve the ruinous conflict that arose in analytical philosophy between those who stressed the perspicuity and naturalness of natural language and its amenability to one sort of philosophy and those who set store by the theoretical insights—supposedly of an altogether different sort—that were gained in the effort to describe the extension of the predicate “true” in the manner of Tarski for formal transpositions of manageable fragments of this or that natural language. But can we not consistently side with both parties? Rather, do we really understand well enough what ‘keeping closer to the surface structure’ amounts to for us to see a clear issue? So often what philosophers want to call ‘surface structure’ stands

only for the reification of a misleading appearance.*** 20. Since first writing this sentence, I have discovered in Stanley Cavell’s

The Claim of Reason

(New York: Oxford University Press, 1979) the following passage: “Nor do I want to go into, nor will [ ever mean to rely on, [Wittgenstein’s] distinction between ‘surface grammar’ and ‘depth grammar’. In such contrasts, ‘surface’ mostly means, so far as I can see, ‘incomplete’ or ‘hasty’ and Wittgenstein uses the notion as part of a diagnosis of philosophical confusion.” 21. In revising and reconstructing this paper, I have benefitted from suggestions made to me by Gareth Jenkins and Stephen Williams—as, above all, from the writings of Jennifer Hornsby:.

6 Three Moments in the Theory of Definition or Analysis: Its

Possibility, [ts Aim or Aims, and [ts Limit or Terminus Summary The first moment

discussed in this chapter is marked by Frege’s (1894)

review of Husserl’s Philosophy of Arithmetic (vol. 1), discussions of the paradox of analysis, and Frege’s division—reaching beyond his distinction of Sinn/sense from Bedeutung/reference—between two conceptions of definition. Six familiar and different ideas of definition are then set out. The second moment is Leibniz’s still serviceable demarcation between the clarity (practical effectiveness) of ideas, and their distinctness, our explicit understanding of the various marks (components) of the concept that the idea relates to. Within Leibniz’s scheme, the pursuit of distinctness leads to a third moment: his demand for adequacy (the enumeration of the marks of a concept, and the marks of these marks...) and Leibniz’s speculations concern-

ing the limit or terminus of analysis, beyond which there can be no greater or more complete distinctness. These problematic speculations are not entirely alien to the Zeitgeist of our own epoch. But so often our understanding of a concept depends not so much upon the unearthing of constituents as upon grasping the concept’s overall shape, the shape that organizes its constituents.

The reflections rehearsed in this chapter arise from three moments in the theory of definition and of conceptual analysis. Moment (I) is marked by Frege’s (1894) review of Husserl’s Philosophy of Arithmetic (vol. 1), the discus-

sion of the paradox of analysis, and the division that Frege proposes there (reaching beyond his distinction of Sinn/sense from Bedeutung/reference), between two different conceptions of definition. That division—suggesting

Meaning, Truth, and the Limits of Analysis: Ten Studies. David Wiggins. Oxford University Press (2022). © David Wiggins. DOI: 10.1093/0s0/9780198726173.003.0006

82

THREE

MOMENTS

as it does the division between lexical definition and real definition— prompts the author to place these two kinds, lexical and real, alongside four

other and different kinds of definition or analysis. The second moment (II) is Leibniz’s still serviceable account (1684, 1704)

of his demarcation between the clarity and the distinctness of ideas or conceptions. This demarcation prompts the suggestion that the guiding purpose of lexical definition is to promote Leibnizian clarity (the practical effectiveness of our understanding of a given concept) whereas that of real definition (understood more or less as Aristotle understood it) becomes inseparable from the pursuit of something else, namely what Leibniz calls distinctness, the explicit understanding, that is, of the various marks (components) of the concept that the idea relates to. Within Leibniz’s scheme, we see the pursuit of distinctness of ideas or conceptions Jead into something well beyond the demands of either lexical or real definition, namely adequacy. Adequacy requires the enumeration of the marks of the concept that any given word stands for and the marks of these marks...One determines the marks of a concept C by asking what properties something has (what other concepts it falls under) by virtue of falling under C. At this third moment (III), so soon as adequacy enters the scene, the search

for distinctness leads into Leibniz’s speculations

(1679)

concerning the limit or terminus of analysis beyond which there can be no greater or more complete distinctness. The problematic nature of these speculations, as pursued by Leibniz, will perhaps cast doubt upon aspirations which are not yet entirely alien to the Zeitgeist of our own epoch. Ought we to reconfigure our own understanding of the business of inquiry into the formation and constitution of concepts? So often the understanding of a concept depends not so much upon the unearthing of yet further constituents as upon the grasp of the shape that organizes constituents.

I 1. In his review

The Aim of Husserl’s

Zeitschrift fiir Philosophie

or Aims

of Definition

Philosophie der Arithmetik

und philosophische

Kritik,

103,

I, published

in

1894, pp. 313—32,

Frege laments Husserl’s psychologism. In the course of developing one of his criticisms, he sets out a problem which (unfairly perhaps, but that is not my concern) he thinks Husserl has made for himself concerning definition. On Husserl’s terms,

THREE

MOMENTS

83

a definition is...incapable of analysing a sense, for the [sense which results from the process of analysing] is not the original sense. In using the word to be defined, I either think clearly everything that I think [already] when I use the defining expression; we then have an ‘obvious circle’ [it being ‘pointless to equate definiendum and definiens by means of such a definition’]; or else the defining expression has a more richly articulated sense, in which case I do not think the same thing in using it as I do in using the word to be explained, and the definition is wrong.

At this point in the exchange between Frege and Husser]l we catch sight of the anticipation of a problem which came later to trouble C. H. Langford and G. E. Moore. In Langford’s words: Analysis states an appropriate relationship of equivalence between the analysandum and the analysans. And the paradox of analysis is to the effect that, if the verbal expression representing the analysandum has the same meaning as the verbal expression representing the analysans, the analysis states a bare identity and is trivial; but if the two verbal expressions do not have the same meaning the analysis is incorrect.

One is tempted to say there must be some appropriate sense of ‘meaning’ in which the two verbal expressions do have the same meaning and some other

appropriate sense in which they do not.’ This reaction of Langford’s and the distinction he makes between two conceptions of meaning might pass muster as a report of a more simply Fregean

response than that which Frege himself gives here to Husserl’s quandary: if we are to avoid the paradox of analysis, then we must distinguish sense from reference. We must distinguish that which is meant by a word—the reference, that is, or thing meant—from the meaning of the word understood as

the signifying power of that word, its Sinn, and the way in which the word secures its reference. Exact equivalence with respect to this second thing— complete synonymy that is between different words—is something rare. That is not however a difficulty for ordinary definition. For it is open to Langford to say (truly enough) that almost all ordinary defining falls well short of achieving (or even aiming for) total interchangeability of definiendum and definiens. The ordinary business of defining involves finding the conceptions of a word’s reference that enable or animate its ordinary use. Consider “an ophthalmologist is an eye-doctor”. This tells normal enquirers all they want or need to know. It gives the ordinary way of thinking of an

1.

See P A. Schilpp (ed.), The Philosophy of G. E. Moore (Evanston, IL: Northwestern, 1942), p. 323.

84

THREE

MOMENTS

ophthalmologist. For that purpose “eye-doctor” serves well enough. Never mind exact or precise equivalence. Never mind total interchangeability. The possibility of such reflections owes something to Frege and his paper ‘On Sense and Reference’.? But at this moment, in this review from 1894,

the thing that Frege himself wants to say to Husserl lies well outside the confines of that celebrated paper. What Husserl’s perplexity reveals, Frege suggests, 1s something else. It reveals a split between the psychological logicians and the mathematicians. What matters to the former is the sense of the words along with the [subjective or imagistic] ideas which they fail [so culpably] to distinguish from the sense; whereas what matters to the mathematicians is the thing itself: the reference or Bedeutung of the words. ..

Enlarging on the second of these concerns— concern distinct from either Husserl’s or Langford’s concerns—Frege then continued: For the mathematician

it is no more

right and no more wrong to define a

conic as the line of intersection of a plane with the surface of a circular cone than to define it as a plane curve with an equation of the second degree in parallel coordinates. His choice of one or the other of these expressions...is made irrespective of the fact that the expressions have neither the same sense nor evoke the same ideas. I do not mean by this that a concept [the Bedeutung or reference, that is, of an ordinary first-level predicate] and the extension of

the concept are one and the same, but that coincidence in extension is a necessary and sufficient criterion for the occurrence between concepts of the relation that corresponds to identity between objects.’

2. The doctrines of that paper, concerned as it was with the designation of objects, are readily transferred to the case of the sense and reference of predicates—concepts that i1s. From that paper we have fully enough to dissolve the paradox of analysis. (The paradox of synonymy 1s another matter altogether.) 3. Pp. 319—20 1n Zeitschrift, op. cit. Frege’s review is translated in Gottlob Frege, Collected Papers, ed. Brian McGuinness (Oxford: Blackwell, 1984), pp. 199—200. For Frege’s account of sense and reference/Bedeutung as 1t applies to predicates, see Frege’s letter to Husserl of 24.5.1891, published (with a diagram from Frege’s own hand) in Gabriel et al. (eds), Wissenschaftlicher Bricfivechsel (Hamburg: Felix Meiner, 1976). See pp. 96—7.The diagram itself1s reproduced in the second chapter of the present volume. NB Just as it stands, the criterion that Frege gives at the end of our citation for the 1dentity

of concepts (in his and my use of the word ‘concept’) is well enough suited to his subject matter. For that subject matter 1s inherently a priori. Once we move to other subject matters, however, nothing much less than necessary coincidence of extensions will sufhice for concept-identity. | see this last stipulation as a kind of completion of Frege’s criterion, a completion not in conflict

with his decisions on cases. For the Leibmizian criterion for identity of concepts, see Hide Ishiguro Leibniz’s Philosophy of Logic and Language (London: Duckworth

Cambridge: Cambridge University Press, 1990), Chapter 2.

1972; second edition,

THREE

MOMENTS

85

2. Something was lost—or rather passed over—in Frege’s response to Husser], but something new was gained. As for the loss, easily recovered, an analytical philosopher or a lexicographer interested in Sinn might reasonably complain against Frege that we need not be falling into the vice of psychologism if we try to catch the thoughts that sustain the public use of a given word (indeed this may even have some bearing upon its exact Bedeutung). Nor need we ourselves be falling into psychologism if we see some utility or point in the sort of analysis that Moore was describing when he wrote as follows in his response to Langford: If you are to ‘give an analysis’ of [a] given concept, which is the analysandum, you must mention, as your analysans, a concept such that (a) nobody can know that the analysandum applies without knowing that the analysans applies to it, (b) nobody can verify that the analysandum applies without verifying that the analysans applies, (c) any expression which expresses the analysandum must be synonymous with any expression which expresses the analysans.* Not all is well here.We have seen already that (c) is wrong. But there is rela-

tively little that is harmfully psychologistic about the Moorean stipulation.® Indeed, on

other occasions, Frege

himself took

some

care to clarify or

improve the ordinary understanding of arithmetical assertions. He sought, that is, to improve the understanding of their Sinn—if only in order to position himself to determine better the proper Bedeutung of the numberwords and other terms in which they were expressed. He could scarcely disclaim all interest in sense and the clarification of sense.® Something needs to be clawed back then from the criticism of Husserl. For Frege did not need to forbid us to contemplate the possibility of a less defamatory and contemptuous understanding than his of the word ‘idea’. Where a philosopher of language speaks (as we shall find that Leibniz does) of the ordinary ideas or conceptions by which we exercise our ordinary grasp of the signification of familiar words, an idea of x may be nothing

4. See The Philosophy of G. E Moore, op. cit., p. 663. 5. The better ground for reservation would be that, by the standard of Moore’s own practice, not all analysis or definition can be constrained by so strict a requirement. Not even all of Moore’s own analytical efforts could have been. Consider the analytical proposals which resulted from Moore’s lifelong struggle with the philosophical problem of perception. Cf. Thomas Baldwin, G.E. Moore (London and New York: Routledge, 1990), p. 222 ff. By Moore’s strict criterion (c),

few if any of these ‘give an analysis’. 6. Cp. Michael Dummett, Frege and Other Philosophers (Oxford: Clarendon Press, 1991), Essay 2.

86

THREE

MOMENTS

fugitive, private, or subjective, but some perfectly communicable conception of x.

3. So much for what was set aside. But what was gained in Frege’s response to Husserl? The gain was for Frege to have drawn attention to a less narrowly constrained kind of definition or analysis than any that Moore or his successors were eager to contemplate. In the sentence last quoted from his review of Husser], Frege’s concern is with the special case where a mathematical discovery or a new approach to some subject matter prompts a radical redefinition of something which is already familiar in another way. In effect, moreover, he also draws attention to something else, namely the quite general possibility (if one may put the matter in terms that Frege does not employ here) of looking for real or substantial definitions. For there is a kind of defining, all of a piece with what Frege called ‘fruitful definition’, where, concentrating analytical effort upon Bedeutung, we advance well beyond the given sense of a predicate, beyond the commonly accepted account of its extension and beyond the knowledge that can be demanded of an ordinary speaker, in order to discriminate and articulate further and previously perhaps unnoticed marks of the concept that the predicate stands for.” The existence of such marks is independent of how much or little ordinary users of the predicate standing for the concept understand about the composition of the concept. One more point about the concept. In order to read aright the sentences that Frege addresses to Husserl, it is important to know that,according to Frege, just as a singular term (by virtue of its sense, by virtue, that is, of its particular mode of determining what object it brings into question) stands for an object that is its reference or Bedeutung, so a predicate— by virtue of its sense, and by virtue of its particular mode of determining how something needs to be in order to satisty the predicate—stands for a concept or Begriff which is its Bedeutung.® Frege thinks of our grasp of any particular concept as our grasp of some particular function from 7. The Fregean marks of a given concept are the concepts (as already explained) that everything falls under which falls under that given concept. (Thus the concept neigh 1s one of the marks of the concept horse.) Why should the word ‘property’ not replace the word ‘concept’ throughout the present chapter? An advantage of the Fregean ‘concept’ 1s that it 1s available to Leibniz, who will be essential to Parts II and III of the chapter. 8. The concept may or may not have objects that fall under it. Where objects do fall under it, they are not the reference but the extension of the concept. See again the last sentence of the last citation in section I. See also the diagram from a letter of Frege’s to Husserl. This is reproduced

in Chapter 2.

THREE

MOMENTS

87

objects to truth-values (T, F).” (See Chapter 2 and the diagram of Frege’s

reproduced there.) 4. Substantial

definition, definition

that takes

off from

Bedeutung,

will

figure large in that which is to follow in Parts II and III. In the interim it will be a pity not to mine Frege’s review of Husserl and the LangfordMoorean exchange, now we have all this in front of us, for certain insights

about that which is involved, perfectly generally, in giving a definition or analysis—whether lexical or real or whatever. In his reply to Langford, Moore wrote: A man might point out to me two expressions...and tell me they had the same meaning without telling me what they meant. So far as he [is] merely telling me they [have] the same meaning..., | know what is meant...But if this were all he was doing, he would not have told me anything at all about any concept [conception, DW would say, to avoid confusion with the Fregean usage| or idea which either of the expressions expressed and he would therefore certainly not have been giving me an analysis of any concept [or idea]...[or] giving me an analysis of any expression.

This seems indisputably correct. But what follows? Surely this. That, despite the appearance created by bilingual dictionaries or simple glossaries, the universal and invariable purpose of definition had better not be the bare provision of translations, synonyms, quasi-synonyms or variants. One who defines—whether his chief intention is to delineate the sense of a predicate,

to arrive spirit of commit cate and

at some further analysis of the predicate’s Bedeutung, or else, in the consolidating some discipline or theory, to do both—mneeds to himself to an account of the kind of thing that satisfies the predifalls under the concept that the expression stands for. In that cause,

9. Does that mean that, in order to understand a predicate, speakers have to know what a function from objects to truth-values 1s? Surely not. Let ¢ be the predicate that some ordinary speaker has learned to apply. Then the thing he has caught on to is a rule or standard which awards the verdict T to an object that meets the standard and, along with T, awards the predicate ¢. (Otherwise F 1s awarded and the predicate 1s withheld.) But the speaker/thinker can catch on to such a rule without knowing any of this. For to learn such a rule 1s simply to acquire a conception or 1dea of a particular property of objects. The conception amounts to just one way of grasping the Bedeutung of ¢. By whatever route someone arrives at that grasp, however, there may be another predicate { and another way of framing some rule or standard that an object needs to pass in order to secure the verdict T and the appellation {; and this other rule may determine the very same concept or Bedeutung, give access to the same function from objects to truth-values, and make the concept’s extension (its satisfiers) the same. For 1dentity of concepts, see again

the second

paragraph

of note

3.The

rule for { and the

rule for ¢ will then yield the same inverse image of T under the said function. But here again

an ordinary competent speaker need not be aware of any of this.

88

THREE

MOMENTS

let one who defines put both analysandum and analysans to work. Let him link them in a declaration of his own. In that way, let him give voice (in a spirit of prescription, of record, or of invitation to agree) to a shared or shareable conception or idea of the Bedeutung or concept that the analysan-

dum word brings into question.'’ There may be other ways to engage with Moore’s point. But the proposal just made does meet it. Furthermore, there is at least one dictionary which follows the procedure just described, explicitly and always, namely Collins

Cobuild English Dictionary (1987, edited by Sinclair and Hanks).'' At random, see p. 980:‘Lobster, lobsters.A lobster is a sea creature that has a hard shell, two

large claws, and eight legs..” The layout or form that we find here answers admirably to our need, even if, in the case of this particular entry, the ordinary give and take of ordinary thought and talk about the definiendum calls for a slightly richer conception. The old Oxford English Dictionary (1884—1928) adds more information. When that is transposed to the Cobuild format, we have: ‘[a lobster is] a large marine stalk-eyed ten footed long-tailed crustacean of the genus Homarus, much used for food; it is greenish or bluish black when raw, and of a brilliant red when boiled; the first pair of feet are very large and form the characteristic claws.” Some of that information is certainly needed within a strictly lexical definition. Other portions trespass beyond the sparer conception which is all you need to be party to if you are to know well enough what creature it is that you are referring to by

10. In this chapter (unless in citations where others besides Frege or Leibniz are speaking) the concept

1s what

a predicate

stands for. For the purposes

of this chapter, nothing will be lost

and clarity may be gained if we confine attention to first-level predicates. We examples in the same way. If we

want

to call the Bedeutung

of a predicate

shall choose

the Begriff or concept, then one who

i1s

ready to use the word ‘conception’ in the way [ propose will need to be ready to speak of the definer’s giving voice to a conception of the concept that the words stands for. Despite the assonance, why not speak like that? Even if it be ugly, it will be clear. To ensure that it be clear, one more explication. Consider the sentence: ‘Cantor invented the concept of transfinite number’. If this sentence 1s true, the wording is at variance with our usage. If the sentence were following our usage, it would have to be nonsense.To make sense of it, we can replace ‘invent’ by ‘discover’. Or else we can recast as follows: ‘Cantor invented (was the first to formulate) the i1dea (or conception) of [the concept| [of] transfinite number’. Note that, in this version, the first of means

‘concerning’

or ‘relating to’. But

the second

[square-

bracketed] ‘of” must be construed on the model of the ‘of” in ‘the city of Paris’. It follows that we can simplify and say: ‘Cantor invented (was the first to formulate) the i1dea (or conception) of transfinite number. But here (let it be plain) the last two words will have as their Bedeutung a concept. 11. See also Patrick Hanks, ‘Definitions and Explanations’, in J. M. Sinclair (ed.), Looking Up:An Account of the Cobuild Project in Lexical Computing (London and Glasgow: Collins, 1987), pp- 116—30.

THREE

MOMENTS

89

‘lobster’."* In so far as the OED definition goes beyond that need and supplies marks of the concept lobster exceeding that which anyone and everyone who properly understands the word must be expected to know, it is headed beyond the strictly lexical as if in the direction of real or substantial definition. More immediately it is headed towards the kind of definition Leibniz calls nominal (not an ideally good name, but there you are): a definition that gives everything that bears upon telling a lobster from another kind of thing. Never mind whether all ordinary speakers know these things or not. 5. Where one who offers a lexical definition scrupulously limits his defining to setting out that which can count as common knowledge among competent language users—and even where he catches very nicely the minimal conception that animates the public use of a word in its application to the extension of its Bedeutung—is there any reason to claim that his putative analysans ought really (or ideally) to be synonymous—truly synonymous— with the original analysandum? And can the analysans replace the analysandum in a sentence without in any way at all affecting the thought expressed? To both questions our answer is ‘no’. Before taking leave of dictionary definitions, there is at least one other thing we should notice. Strictly speaking, even if we take the full OED definition as giving the whole conception that animates the common use of the substantive “lobster”, an object’s satisfying the OED specification is neither necessary nor fully sufficient just as it stands for the object’s being a lobster or satisfying the predicate ‘is a lobster’. A genuine lobster could be deformed. Moreover, something that wasn'’t really a lobster might yet answer to the OED description." For there is more to the conception or idea that 12. Not all English-speakers know that lobsters are blue. (Some may assume that they are pink. Hence, perhaps, the OED explanation about their colour.) Never mind. There 1s a temptation to suppose that it 1s a matter of degree how well a person knows what he says by the words that he utters. NB

In this chapter, a lexical definition of t will be a form

of words

that sets out—so

far as

this 1s possible in a dictionary entry—all and only the information one needs, over and above the rest of one’s knowledge of the language, in order to understand sentences involving t. Such a definition may fall well short of nominal definition in Leibniz’s sense. See section 6, para 3. 13. The difficulty 1s well known. The best present-day account of it 1s given in Hilary Putnam’s article ‘Is Semantics Possible?’, Metaphilosophy, 3, 1970, reprinted in his Mind, Language and Reality: Philosophical Papers,Vol. 2 (Cambridge: Cambridge University Press, 1975). But 1t was familiar to Leibmiz. Compare New Essays in Samtliche Schriften und Briefe (Berlin: AkademieVerlag, 1923—),VL.6, p. 255: “When we think we have thoroughly described a plant, someone

may bring from the Indies a plant which exactly fits everything we have put into our description

and

follow

which

Bennett’s

nevertheless and

can

Remnant’s

be seen translation

This uses the Akademie pagination.

to belong

to a different species.” Here, as below,

(Cambridge:

Cambridge

University

Press,

1981).

|

90

THREE

MOMENTS

sustains our ordinary use of the substantive ‘lobster’ than can be reproduced by the verbal specification given in a dictionary. This is to say that that sustaining conception, in so far as it is doing its proper work, is in part recognitional. This surely requires that, among users of such a substantive, there should be some who know what it takes to recognize a lobster when they see one. Such persons may stand in readiness to help to promote others’ understanding of the word by pointing and saying ‘that is a lobster’.'* There is room for more than one opinion about how exactly the phenomena of deixis and the synchronic and diachronic division of identificatory labour between experts and lay persons should be characterized. For present purposes, the thing that matters is that, however these things are managed, a recognitional conception such as this, attaching the predicate to instances of the concept that it stands for, will never be transposed without residue into an analytical equivalent for the definiendum. Dictionaries such as the old Larousse which, in lieu of pointing, included line-drawings of many of its definienda, were precisely avowing the persisting impotence in this regard of bare words, however numerous or explicit.”” They

avow

the existence of needs which, in the end, can only be

satisfied by real definition. 6. Out of the great variety of purposes that someone who defines might have, let me enumerate just six: (1) There is the practical business of meeting the needs of language users. These are, of course, a heterogeneous group. Mostly though, in so far as they look for definitions—and receiving a definition is no more than a special case of having something explained to one or having it recorded or inventoried for one—the thing that users will normally want is to consolidate or

14. Once a substantive 1s explicated in this way, the possibility looms of open-ended further enquiry (e.g., as here, by marine biologists or other inheritors of Aristotle) into the given nature of things that answer to the substantive. In this way 1t will appear that ordinary nomenclature for natural things and stuffs will have to subject itself, however indirectly, to the authority of experts, present or future. Compare Leibniz, New Essays, Ak. pp. 354, 399—402. 15. In so far as the lexical effort seeks to transcend 1ts limitations and encapsulate recognitional criteria, 1t may resort to an idea given in Putnam (op. cit.) but anticipated in Leibniz: x 1s an f just if 1t 1s relevantly similar to this, that or these [demonstrated] exemplars. This gets beyond bare words, but ofters nothing we can substitute for ‘lobster’. It trespasses also beyond the limits

of an entry for a dictionary. It exceeds the lexical. See note 11. For more about Putnam’s proposal and related matters, see David Wiggins, Sameness and Substance Renewed (Cambridge: Cambridge University Press, 2001), p. 79, note 2, and p. 83, notes 2 and 3. For the provisional/experimental character of the choice of such exemplars, see

New Essays, Ak. pp. 322-3.

THREE

MOMENTS

o1

to extend their own powers of utterance or of comprehension.'® This is the chief work

of lexical definition. Sometimes

that work

can be turned back

into some sort of equivalent of the definiendum, but it need not do so. Transparent or manifest equivalence of reference and/or extension is the

most that needs to be looked for."’ (2) In contrast with type (1) purposes, but in parallel with Frege’s purposes in his converse with Husserl, there are other purposes dear to Aristotle and to other scientifically minded philosophers, of real definition: It is useful to ascertain what a [given kind of] thing is, if our purpose is to understand why that [kind of] thing’s attributes flow from its nature or essence...Conversely, these attributes of the thing do themselves contribute greatly to our knowledge of the what that [kind of] thing is. It is [only] when we can enumerate all or most of these attributes as they appear to us that we shall say properly what its nature or essence is. Always the starting point of explanatory demonstration is the what the thing in question is. (Aristotle, De Anima, 402 b16—402 b2s)'® 10.

Sometimes they will want to understand better the things that used to be said in the past, to perpetuate some of them, or to contribute to contestations that were begun by dead authors. Such speakers acquiesce or rejoice in the thought that a language needs to hold on to certain

precious resources whose preservation and perpetuation partly depend on the existence of dictionaries themselves. Some who consult dictionaries are actively looking for prescriptions,

prescriptions general compliance with which will contribute to that preservation and perpetuation. Who shall say, and on what good grounds, that they are wrong to have this motivation? (See Chapter 8.) . These exclusions give no reason, however, not to call lexical definition ‘definition’. Contrast

Dwight Bolinger’s use of the word ‘define’ in the following: ‘Dictionaries do not exist to define, but to help people grasp meanings, and for this purpose their main task 1s to supply a series of hints and associations that will relate the unknown to something known’ (‘The Atomization of Meaning’, in Language, vol. 41. (4), 1965, 572). There 1s something to agree with here, despite the needless subjectivism. But I should insist

that an insistence on a would-be rigoristic use of ‘define’ often rests on misconceptions we have already criticized. . In A Map of Metaphysics Zeta (Pittsburgh, PA: Mathesis Publications, 2001), Myles Burnyeat

offers the following commentary on this passage: Imagine coming across what we take to be a new, hitherto unstudied species of fish. Provisionally, we may call them Zetafish...[Suppose| there 1s a unity to the phenonema that we observe, which entitles us to say that a single kind 1s responsible for them all. .. [Then] the next task 1s to collect the properties that are common to all Zetafish [the marks, that is, of the concept Zetafish] and decide which of them are distinctive (differentiae) of the new species.

All these will be necessary to being a Zetafish, but only some of them will belong essence... To find the essence of a kind 1s to find the cause or explanation of the properties that necessarily belong to members of the kind in virtue of their being that thing. In Aristotle’s language, the explananda are the per se attributes, [and]| the essence explanans. Such a cause will take some finding. In De Anima 402b 21—-26 Aristotle piece

of heuristic

advice. Familiarize

yourself with

the full range

of the attributes

to the various kind of [is] the offers a to be

explained, because it is [only] when you can give a provisional account of all or most of these [attributes] that you will be best placed to determine the essence which is the starting point of explanatory demonstration [e.g. of the life and being of the Zetafish].

02

THREE

MOMENTS

Let this be our paradigm of real definition. It is a paradigm that other philosophers have gone on to interpret or specify further and in their own ways. In Aristotle, as we see, the real definition subsumes

not only that which

Leibniz calls ‘nominal definition’—see section 4, ad fin.—but everything else besides that is needed in order to explain why the things in question have the attributes that they do. As regards that aim, which is usually dependent upon empirical enquiry, Leibniz will want to construe real definition as requiring (in theory) (i) that, in due course, the concept answering to the nominal definition be further analysed and (ii) that the marks and the

marks of the marks...be enumerated to the point where the concept’s soundness and non-self-contradictoriness are finally evident. (See below Parts II and III.)

(3) Next let us mention the remedial or prescriptive role that definition and redefinition can play in furthering or refining the aims of some particular a priori or a posteriori pursuit or discipline. Here lexical and type (2) considerations will interact. The particular technical or scientific purposes of the discipline in question may require the pre-existing understanding of a given

term to be modified or supplanted in the light of some useful discovery which makes it desirable or even imperative to stipulate a new meaning for an old term." (4) Next

there is the definition of new

terms

(or the stipulating of new

acceptations for existing terms) in a developing discipline. The associated hazards have been well documented in theory, even if they are all too cheer-

fully ignored in practice.* 19. Here 1s an example that I owe to Patrick Hanks—see his article ‘Definition’, in Keith Brown ed., Encyclopaedia of Language and Linguistics (Oxford: Elsevier, 2005).A second 1s usually defined as a sixtieth part of a minute, where a minute 1s defined as a sixtieth of an hour and an hour 1s defined as the twenty-fourth part of a civil day. But in 1967 the international committee charged with determining the meaning of international scientific units (SI units) defined a second as the duration of 9,192,631,770 cycles of radiation corresponding to the transition between two hyperfine levels of the ground state of an atom of caesium-133. Here a discovery

about the special regularity of cycles of radiation made it possible—or so an engineering colleague tells me—to build the new definition of second into ordinary workaday instrumentation which will serve certain highly non-arbitrary scientific or technological purposes. In this sense, the new

to type (3). 20. See Patrick

definition has become

Suppes,

Introduction

mandatory. Indeed it may

to Logic

(Princeton,

NJ: Van

in some

Nostrand,

cases move

us closer

1957), Chapter

8.

Elsewhere, I have said my own piece about the hazards of creative definition within theoretical

enquiries such as philosophy or the social sciences and wherever else it 1s unclear how

an

empirical or conceptual discovery will vindicate or attest a Bedeutung for a term introduced either creatively or as if ‘on spec’. See ETL (2000), especially pp. 28, 127, 357; and S&ESR 2001), pp- 170, 190, 212.

THREE

(5) One

MOMENTS

other kind of defining deserves mention

apart. Here that which words (new or old) but duced for discussion, an of sentences (e.g. those

03

here—if only to set it

defines is some particular expression or form of the thing defined will be something newly introobject (putatively) or a class of objects, say, or a set true in some language L), rather than a word or a

concept: (a) ‘Consider the least ordinal not definable in a finite number of

words: this ordinal is itself defined in a finite number of words’ (Berry’s paradox, 1906); (b) ‘Apart from the universal class and the empty class, no

class of individuals can be defined by purely logical means’ (Alfred Tarski, Logic, Semantics and Metamathematics, trans. Woodger, Oxford: Oxford University Press, 1956, p. 386).

Finally (6), there is defining as we have it in present-day philosophy where philosophy is relatively rarely concerned to find either a lexical or a real definition. Loosely and frequently, we speak of philosophical analysis. But that which goes under this name will relatively rarely attain the traditional ideal of informative non-circular necessary and sufficient conditions for the presence of the analysandum. By way of eminent instance, consider here the labours of ‘connective

analysis’ (as he came

later to call it) that

P. E Strawson devotes in Chapter 3 of his Individuals (London: Methuen, 1959) to the concept of person. In this kind of work, even where it singles

out certain key marks of a concept, we see how philosophical explication as we now conceive it can and often must come apart from definition, either lexical or real.?!

II

Leibniz’s Difterentiation between Clarity and Distinctness

There is room for a fuller conspectus of the varieties of definition. But at this point, and following on from the distinction we have developed out of Frege’s reaction to Husserl, let us focus upon type (1) and type (2), defining and bring them into relation with the division that Leibniz proposes between the clarity of an idea or conception, which is a primarily operational or practical merit, in line with the aims of lexical definition and the competence to which ordinary thought and speech have to aspire, and the 21. For connective analysis, see P. E Strawson, Analysis and Metaphysics (Oxford: Oxford University Press, 1992), Chapter 2.

04

THREE

MOMENTS

distinctness of an idea or conception. Distinctness is an incipiently theoretical or speculative achievement requiring a further and better grasp of the marks of this or that concept. Distinctness needs clarity but, in the cause of theoretical understanding, the Leibnizian distinctness of an idea demands much more than clarity. In line with this division is the division between (a) the

part that lexical (Sinn-based) definition can play (or help to play)—as with defining of type (1)—in supplanting one idea with another idea that is surer, more certain, or more effective in application to instances of a concept, and (b) the potentially different role that substantial or would-be real definition

can play—as in this case of defining of type (2). For real definition not only looks for a securer foundation for the clarity of ideas. In the further cause of distinctness, it can place alongside a given idea a supplementary idea that identifies more of the marks of the concept and better reveals the nature of

whatever falls under it.** A Leibnizian idea is a disposition of mind that makes possible the thought of a certain kind of object or concept. It furnishes a conception of a thing or a kind of thing, a differentiated way of responding to experience, a sensibility specifically directed de re at things or kinds of things to be found in reality. Throughout the present paper, ‘idea’ has this same Leibnizian usage.” Such ideas can be clear (simply) or else clear and distinct. What is the difference? And what more can be asked beyond distinctness? 22. Here 1s Leibniz’s explanation of his intentions in calling such an 1dea ‘distincte’: Ainsi, quoyque selon nous les 1dées distinctes distinguent I’objet d’un autre, neantmoins

comme les claires mais confuses en elles mémes le font aussi, nous nommons distinctes non pas toutes celles qui sont bien distinguantes ou qui distinguent les objets, mais celles qui sont bien

distinguées, c’est a dire qui sont distinctes en elles mémes

et distinguent

dans |'objet les marques qui le font connoistre, ce qui en donne I'analyse ou définition; autrement nous les appellons confuses. Et dans ce sens la confusion qui regne dans les 1dées pourra estre exemte de blame, estant une imperfection de nostre nature. (New Essays 1. xxix, 4, Ak. p. 2550, 1talics added) For the connection between distinctness and real definition, see also the preface to the same work: Et toutes les fois qu’on trouve quelque qualité dans un sujet, on doit croire que si on entendoit la nature de ce sujet et de cette qualité on concevroit comment cette qualité en peut resulter...Ce qui est nature] doit pouvoir devenir concevable distinctement si I'on estoit admis dans les secrets des choses. (Ak. p. 66, italics added) 23.

To attain more precision here, it would be desirable to distinguish between

two related mean-

ings of the words “idea” and “conception”: (A) the mental disposition or state of one who conceives of a thing or kind of thing in a certain way; and (B) the way in which the thing so conceived is taken to be, and the properties it 1s taken to have, by the possessor of the said 1dea

or conception umproved;

or

(A). Conceptions else

supplanted

by

in sense some

(A)

better

may

be

correctly

conception. As

said to be

a thinker’s

enhanced

or

understanding

(or

conception 1n sense (A)) improves, s/he may exchange one idea or conception (in sense (B)) for another and better 1dea or conception (in sense (B)).

THREE

MOMENTS

95

7. Leibniz set out his distinction between clarity and distinctness in ‘Meditations on Knowledge, Truth and Ideas’, which appeared in November

1684 in the Leipzig journal Acta Eruditorum.>* A cognition, conception, or idea is clear (the opposite of obscure) just to the extent that a possessor of this idea can be relied upon to apply the predicate associated with the idea to things that fall under the concept which the idea is an idea of. Someone’s cognition or idea can be clear without their having any very articulate account of the satisfaction conditions of the predicate that introduces the concept. It is a profound and important fact about human thought and practice that, so far from being unusual, clear reliable cognition without further, articulated knowledge of marks is normal. That is to say (in Leibniz’s way) that clear confused cognition is the ordinary or habitual case. In Leibniz’s usage, “clear and confused” is not a contradiction. It precisely signals that ordinary case, while making room for the further possibility of distinctness. What then is distinctness? A cognition, conception or idea is distinct just to the extent that the conception or idea enumerates and spells out more explicitly the properties that 24. See Gerhardt (ed.), Die philosophischen Schriften (Berlin, 1875—90),Vol. 1V, p. 422, which 1s translated at Loembker (ed./trans.), Leibniz: Philosophical Writings (Dordrecht: Reidel, 1969), p. 291. NB Descartes and Locke each speak of clear and distinct 1deas, but Leibnmiz’s use of these terms, which is all that concerns us, 1s not exactly the same as either of theirs.

In the ‘Meditations’ paper Leibniz makes extensive use of the Latin terms cognitio and notio. It would be nice if cognitio could be paired off tidily with idea or conception (as 1 have been fol-

lowing Leibniz in using these words) and notio could be paired oft with concept as Frege used that term. Unluckily though, notio seems in Leibniz’s paper (as in other places) to hang uneasily between

that which

was intended by idea or conception (as I have used these words), some-

thing on the side of sense, and the as-if Fregean concept on the side of reference. So 1t is fortunate for our purposes, which do require that Frege and Leibniz be comparable, that, when Leibniz returned to the matter in the dialogue that he constructed in New Essays

between Philalethes/Locke and Theophilus/Leibniz, he employed idea in a less special sense than he had in 1684 (details not immediately relevant) and in more or less the way in which

he used cognitio in 1684, and fortunate again that this is the same as the way in which (throughout this chapter)

we

have been

using idea and conception.

Thus, at New

Essays, II, xxix, Leibniz

treats ideas as having as their objects the thing-kinds of sensible things, animals, colours, or stuffs (gold etc.). (See New Essays, Ak. pp. 254—63. See also pp. 52—3, 80, 96.) Another helpful

thing is the frequency with which Leibniz uses concept on the side of reference rather than sense and uses 1t with a meaning more or less coincident with Frege’s. See Ishiguro op. cit p.2s. See also footnote 18 of my contribution, ‘The Concept of the Subject contains the Concept of the Predicate’, to J. J. Thomson (ed.) On Being and Saying: Essays for Richard L. Cartwright (Cambridge, MA: MIT Press, Bradford Books, 1987).That paper 1s reprinted in my Continuants: their activity, their being and their identity Oxford, OUP, 2016. To measure the congruity of Leibniz’s usage of the word ‘concept’ with Frege’s and with talk of the properties of things, see for instance G. Grua (ed.), Textes inédits (Paris: Presses Universitaires de France, 1948), p. 383:“Of the essence of a thing 1s what belongs to 1t necessarily and perpetually. Of a concept of a singular thing, however, is also what belongs to it contin-

gently or by accident, or what God sees in it when he has perfectly understood it.”’ [Italics added. |

96

THREE

MOMENTS

are needed in differentiating between instances of the thing-kind that answer to the conception or idea and instances of other kinds. This is to say that a distinct idea articulates the recognitional marks of the concept that the cognition, conception, or idea conceives or cognizes.” (See section 4, ad fin.) Suppose we say that it is the work of lexical definition as focused on Sinn /sense to begin the work of making our ideas clearer (more distinct, yes, but only to the extent that clarity itself demands that) and we say that it is the work of real definition, as focused on Bedeutung, to make our ideas distinct enough for us to satisfy certain further demands that we have seen Aristotle place upon explanatory demonstration /comprehension (see De Anima 402 b16 following, cited in section 6). There is much to be said for this finding, but, once we accept it and we turn to Leibniz, we have to confront the further expectation according to which lexical definition understudies nominal definition and nominal definition awaits its proper validation in real definition, which awaits its further completion in a kind of adequacy that lies way beyond any Aristotelian conception of explanatory understanding. “Everything rests upon distinctions which eventually go back to the ideas from which all others are derived” (New Essays,Ak p.352). Here, with the thought, that is, of primitive ideas, we enter upon another moment

in the theory and practice of defining.

III

A Limit or Terminus of Defining?

8. In the ‘Meditations’ paper of 1684, Leibniz mentions two more virtues, besides clarity and distinctness, to which our cognitions, ideas, or conceptions can (just imaginably) strive to attain, namely adequacy and intuitiveness. Through the consideration of these we shall approach our third theme— the would-be terminus (supposedly) of definitional analysis. It is one thing for a state of cognition

to strive to attain distinctness,

where (as we have seen) the subject starts upon the business of identifying the marks of a given concept. At the next stage, Leibniz places cognition that is adequate. One’s cognition or idea of a composite concept (and according to Leibniz most or all of the concepts familiar to human beings are 25. In most of its applications, Leibniz’s use of ‘mark’ coincides well enough with Frege’s. Where it holds universally that, for all x, if x if F then x 1s G, then the concept G is a mark of the concept E It follows, of course, that every concept 1s a mark of itself.

THREE

MOMENTS

97

composite) will count as adequate just to the extent to which, over and above furnishing a distinct idea of the recognitional marks of a concept, it also furnishes a distinct idea of the marks of these marks and the marks of the marks of these marks...all the way (it might be hoped) down to the primitive or irresoluble concepts that compose it, namely concepts that Leibniz describes as ‘their own marks’, each of them being index sui.*® A little later in the 1684 paper, he describes such ‘irresolvable notions or concepts as coming to the same as the absolute attributes of God’. 9. In a matter so mysterious, where the idea of distinctness culminates so swiftly in the altogether new ideal of adequacy, it is worthwhile to have parallel testimony. There are similar thoughts in Leibniz’s (1679) ‘Introduction to a Secret Encylopaedia’ (Couturat (ed.), pp. s11—15). But here too I must

paraphrase or expand whenever it may be necessary if we are to uphold the distinction, to which Leibniz is fully entitled, between conception or idea,

on the side of sense, and concept, on the side of reference: A [conception or idea of a] concept is distinct when [ can consider separately and distinguish between the marks I have for recognizing a thing [that falls under the concept]...The conception [of a concept] is adequate when it is so distinct that it contains nothing indistinct—whether the marks themselves of the concept be known by distinct conceptions or known [directly, by the direct cognition

of]

simple

or primitive

concepts...A

concept

is primitive

when it cannot be analysed into others; that is when the concept has no marks but is its own sign, an index sui. But it can be doubted whether any concept of this kind appears distinctly to human beings, namely in such a way that they know they have it [i.e. know they have a conception or idea of it and know that that is what their conception or idea is of]. Indeed such a conception can only be of the thing which is conceived through itself, namely the supreme substance, that is God. (Trans. Morris and Parkinson.)

Leibniz then adds, “We can have no conceptions of derivative concepts except by the aid of primitive concepts’. In so far as this amounts to more than the contention that complex concepts are constituted from more primitive ones, we need to understand why Leibniz thought that there had to be simple primitive concepts and how he thought the theoretical availability of these must ‘aid’—that is, support or sustain—our ordinary conceptions. We also need to know Leibniz’s account of cognitions or 26. Compare Gerhardt (ed.), Vol. VII, p. 292, where Leibniz speaks of an ‘alphabet of human thoughts, the catalogue of summa genera, out of whose combination derived concepts would be formed’.

98

THREE

MOMENTS

conceptions that are intuitive. For the latter I revert to the paper of 1684 (paraphrasing at some points, but only on the same basis as before): “We do not usually grasp the entire nature of a thing all at once...but in place of things themselves we make use of signs...[We] use words whose sense may only appear obscurely and imperfectly to the mind in the place of the ideas [we] have of these things...I usually call such thinking, which is found both in algebra and arithmetic and indeed almost everywhere, blind or symbolic. For, when a concept is very complex, we cannot consider all of its component concepts at the same time. When we can, or indeed in so far as we can, | call our cognition intuitive.All distinct knowledge of a primitive concept is intuitive, just as most of our thinking about composite concepts is symbolic. In so far as we can attain intuitive knowledge of something, we conceive it through itself’ 10. That there are indeed primitive concepts is something Leibniz tries to prove as follows.>” Whatever is cognized is either conceived through itself or conceived through another thing. If something a were conceived through something else b which was conceived through something else ¢...in infinitum, then (Leibniz says) things would be like this: ‘I give you a hundred crowns to be received from Titus: Titus will send you to Caius, Caius to Maevius: but if you are perpetually sent on in this way, you will never be said to have received anything’. Putting Leibniz’s conclusion in terms evocative of the Axiom of Regularity or Fundierung in the theory of sets, we have to conclude that no cognition that is a thought of something can rest upon a non-terminating downward chain of things each of which is conceived only through the conceiving of something else. Applying this argument more closely to concepts, it then appears that, if any concept is to be conceived at all, then some concepts must be conceived in themselves. It follows—or so Leibniz supposes—that some concepts must be simple. And if so, then in the end some terms must be indefinable, each being secured to its own feature of reality, either by convention or by the help of other mechanisms—causal, depictive, or mimetic mechanisms

(onomato-

poeia, for instance). In some cases, these mechanisms are the subject of

27. See Couturat (ed.), Opusales et fragments inédits de Leibniz (Paris: Alcan, 1903), p. 429, ‘Of an Organon or Ars Magna of thinking’.

THREE

Leibnizian

speculations

almost

MOMENTS

as strange

99

as those

offered

in Plato’s

Cratylus.* [t may seem curious that such groundedness should be a condition for the significance of our concept-terms when it is so hard for practised speakers to identify a single primitive concept. It may seem strange that that which we commit ourselves to in our everyday utterance should be answerable to a structure of underlying primitive concepts which so artfully conceal themselves. (Cp. L. Wittgenstein, Philosophical Investigations, Oxford: Blackwell, 1953, section 91.) Yes. But, when we engage in symbolic thought

or pensée sourde (in Leibniz’s Latin, caeca cogitatio) and when we exercise our clear but confused/indistinct ideas, maybe the confidence that we have in the significance of what we say and think—and our trust in the general effectiveness of our recognition of thing-kinds—should be likened to the act of faith by which citizens of a sound economy will accept coinage or credit in exchange for goods or services without knowing anything at all of the real assets that underwrite the currency. But what else can one do? 11. This is the moment to mention the way in which Leibniz’s characteristica universalis was intended to lend to ordinary thought the power to reduce all good reasoning to calculation. This characteristic was to be a logically and semantically transparent language supplied with all the terms we shall need in order to set out everything we can find out about the universe. Leibniz saw human

knowledge as best extended, improved, and reformu-

lated pari passu with a process of successive approximations to a universal characteristic, the approximations themselves being made from the side of natural science in progress and the more perspicuous portions of our extant language.” Trailing after the general encyclopaedia of the sciences that trails after the active inquirer, the exponent of the characteristic is to lay out in primitive terms,”® or in terms standing only so many abbreviatory steps 28. Compare Leibniz, New Essays, Akademie, pp. 278, 281—3, and Plato, Cratylus, 422°=27¢, 433 Couturat writes in La Logique de Leibniz, d’aprés des documents inédits (Paris: Alcan, 1901), ‘Une telle nomenclature, ou le nom de chaque chose (ou idée) en serait le symbole adéquat et transparent, et pour ainsi dire le signalement ou le portrait logique, constituerait évidemment une sorte de langue naturelle, comme celle que Platon révait dans le Cratyle; ce serait la langue d’Adam ... (p. 77). 29. Cp. L. Couturat, La Logique de Leibniz, p .28;William and Martha Kneale, The Development of Logic (Oxford: Oxford University Press, 1962), p. 329. 30. ‘Cette encyclopaedie estant fait selon 'ordre queje me propose, la caracteristique serait quasi toute faite’ See p. 97 in Eduard Bodemann (ed.), Die Leibniz Handschriften der Kéniglichen Offentflich Bibliothek zu Hannover (Hanover: Hahn, 1895).

100

THREE

MOMENTS

away from primitive terms, all the marks of each ordinary concept that a word of the universal language stands for. ‘By means |of the art of combinations], all composite notions in the whole world are reduced to a few simple ones as their alphabet; and by the combination of such an alphabet, a way is made of finding in time, by an ordered method, all things, with their theorems and whatever it is possible to inquire into

concerning

them’

(Letter to the Duke

of Brunswick,

Gerhardt

(ed.),

Vol. I, pp. 57-8).

Here in sum is the picture of language that results. In speaking an ordinary natural language, we use simple signs to represent things and natures that are not at all simple in reality. But in the universal characteristic—in a calculus which will at once validate and regulate the serious or scientific employment of natural Janguage and underpin its very significance—the sign for a complex thing will either be the definition of the concept of that thing or else be understood as deputizing for that definition. (Could Aristotle have dreamt that that was where the search for the what it is or the essence would lead?) An apt summation of the spirit of the enterprise is given in a manuscript described and excerpted by Bodemann (op. cit., pp. 80—1). Its significance was clearly perceived by William Kneale:*! The art of the characteristic is the art of forming and ordering characters in such a way that they should recapitulate thoughts—which is to say that they should have the same relation to one another as thoughts have to one another. An

expression

is an aggregate

of characters

representing

the

thing

that is

expressed. This is the law for expressions: that the expression for a given thing should be put together from the characters for those ideas from which is composed the idea of the thing that is to be expressed.

In order to align this passage with our policy of distinguishing cognition/ idea/conception from concept, I follow the paraphrase given by Benson Mates (The Philosophy of Leibniz, Cambridge: Cambridge University Press, 1986, p. 187): Just as a complex concept is composed of simpler ones, so the linguistic expression representing that concept is to be composed of expressions representing its components.

Leibniz connected this whole plan in various ways with theoretical aspirations that we should subsume nowadays under the headings of 31. See William Kneale, ‘Leibniz and the Picture Theory Philosophie, 20, 1966, pp. 204—15.

of Language’, Revue Internationale de

THREE

MOMENTS

101

philosophical grammar and semantics, deductive logic, and the fundamental principles of science, including its methodology and heuristics. Once the whole was achieved, Leibniz predicted, then not only should we have the thread of Ariadne to guide us in discovery, but: When disagreements arise, there will be no more need for disputation between two philosophers than there is between two arithmeticians or accountants. It will be enough to pick up pens, to sit at the abacus and say to one another (with or without

calling a friend), let us calculate. (Gerhardt

(ed.), op. cit.,

Vol. VII, p. 200)

12. What then are the ‘letters’ or characters of this alphabet? In the same ‘Introduction to a Secret Encyclopaedia’, to which we have already referred, Leibniz writes:

Categories: i.e. a catalogue of concepts set out in order and of conceivable things i.e. of simple terms. The concepts are: possible; entity; substance; accident or adjunct; absolute substance; limited substance, or that which

can be

passive; living substance, which has in itself a principle of operation, or soul; thinking substance which acts on itself—this is also called mind. (Couturat, Opuscules, p. $14)

It would appear that, in this list, Leibniz intended the various qualifications of substance which make up the last four items to be read as diverse specifications of some yet more primitive concept, something underlying the complex concept, itself of a simple or unitary substance. This brings us to the margin of the same manuscript. For in the margin, Leibniz has entered a different list, one intended perhaps to push the same process one step farther back in the direction of that which is more primitive. It reads: possible; entity; existent; potent; knowing; willing; enduring; what is changed; suftering; perceiving; having location; extended; bounded; shaped; touching; close; distant.

Was it Leibniz’s intention, one wonders, to try to capture the notion of substance or that of ultimate subject (which is the notion he adverts to at the

beginning of the fragment “Termini Simpliciores’ reproduced in part in Grua

(ed.)

1948, pp. 542 ft.) and to capture its various differentiations by

judicious use of the terms entity, potent, knowing, willing, enduring, perceiving?

Imagine that some such thing were seriously proposed and we were working our way towards the simple concepts that Leibniz is in search of,

102

THREE

MOMENTS

namely concepts that have no marks (beside themselves). Then with respect

to each of these candidates we should need to ask whether every determination of each of these determinables is independent of each and every determination of each of the others. It is hard however to believe that this is always or usually possible. How can anything think in the way that is involved in knowing, perceiving, or willing unless the thinking thing endures through time? If (as seems obvious) the answer to this question is that it can’t and thinking entails enduring, and if know, perceive, and will all

have this mark, then they are not primitive. And why should we suppose that matters will become any easier at some succeeding stage in the push towards the simple, primitive, or indefinable? More generally, why should we suppose that, as a condition of its very significance, every meaningful term can be exhaustively dismantled into ultimate components each of which is index sui? Why moreover should we suppose that any and every combination of such simples will amount to a concept of which there can be some intelligible conception or idea? 13. We are not alone in our perplexities here. Among the remarkable instances of Leibniz’s own openness and creativity in the exploration of difficulty is a letter from the same period and the same phase of his thought, which Leibniz wrote to his correspondent Vagetius on 12 December, 1679:? Protonoemata secundum quid or concepts that are primitive relative to something, and protonoemata simpliciter or concepts that are simple absolutely, are rightly distinguished, and this is a distinction I have made long since, albeit in terminology other than Jungius’... As regards protonoemata simpliciter—absolute indefinables—or things [concepts] that are conceived through themselves, these are something I have often thought about. Even if it is difficult for us human beings to articulate such concepts distinctly enough, we may still think about them in principle by imagining that we have enumerated some absolute indefinables or simples. At this point, various questions can arise: (1) whether there are any protonoemata at all—or is there division in infinitum, as there is in some other cases of division?—and (2), supposing that there are protonoemata (for it seems that nothing is truly conceived of unless some things are conceived of through themselves), whether there is only one or more than one. If there is only one, 32. Samtliche Schriften und Briefe, Akademie 2.1.497. Couturat, p. s13. 1 owe the incitement to study this letter to Benson Mates’s declaration that he does not know what to make of it; see The Philosophy of Leibniz, p. 60, note 51.

THREE

MOMENTS

103

how can it give rise to so many composite concepts? If there are several, then they are bound to have something in common, such as possibility. And there will be relations between them, otherwise they will not be able to combine to make composite concepts. How then can there be simple notions? Whichever way we turn, we run into difficulties. Yet the settlement of these questions is of great moment for the establishment of the true principles of the sciences.

The difficulty Leibniz sees in there being only one protonoema is straightforward enough. But the difficulty he sees in there being several protonoemata needs a serious answer. What will strike one here is how doubtful it must be whether there is a useful understanding of ‘mark’ according to which possibility—exemplifiability he must mean—is itself a (Leibnizian) mark of a concept. Can we not help Leibniz out of the difficulty he mentions in the letter by supplying to him the strictly Fregean account of marks as it was set out at the end of the first paragraph of our section 3? For neither exemplification nor exemplifiability can be such a mark of a Fregean concept. ‘Neighs’ is a first-level predicate of Victor, Arkle, or Shergar. Neither ‘exemplifiable’ nor ‘exemplified’ can be that sort of predicate. This helps perhaps to secure the multiplicity of simplest concepts against the difficulty that Leibniz apprehends in his letter to Vagetius. He can insist maybe that each primitive sign points to its very own—exclusively its own—feature of reality. That is what makes the primitive concepts compossible, independent, and multiple. That (he can say) is what grounds the universal characteristic. Such a picture might explain the close linkage in Leibniz’s exposition between a concept’s being primitive and its having no marks beyond itself or being an index sui. We might ask though how every genuinely significant term of which a language needs to avail itself can be underwritten by primitive terms whose senses are fixed by signs that show their own proprietary and exclusive Bedeutung, as if by pointing? Can we really arrange to segment reality itself in such a way as to furnish each and every putatively primitive term with its very own designation unshared? 14. Once challenged upon this point, I think Leibniz himself might have postponed all reformulation or refinement of his theory of simples or ultimate concepts and tried to advance along another front, inviting his critics to join him in considering the special claims of concepts such as gold for inclusion in each and any alphabet for any universal characteristic.

104

THREE

MOMENTS

When one thinks of gold as yellow, fusible and resistant to cupellation,® its specific essence reveals itself to one albeit confusedly, through the qualities, yellow, fusible. New Essays, p. 405 There are hundreds of truths we can be certain of concerning gold, the body that is whose inner essence reveals itself through the greatest weight known here on earth and the greatest ductility. New Essays p.400

Less impressively perhaps, this success could surely have been repeated reassuringly many times. Advancing from there, moreover, Leibniz certainly hoped to discover something similar within the biological sciences: The more deeply we study how species are generated, and the more thoroughly our rankings follow the necessary conditions of generation, the nearer we shall come to the natural order...If we ...knew things well enough, perhaps we would find for each species a fixed set of attributes which were common to all the individuals of that species and which a single living organism always retained no matter what changes or metamorphoses it might go though. (New Essays, Ak. p. 310; cp. p. 325) Here, however, with this particular sort of sort, Leibniz would have had to

be disappointed. No workable theory of species and their origination can possibly make room for a fixed set of attributes common to all organisms that belong within one species.>® It is anybody’s guess how Leibniz would have responded to this setback. But leaving his whole venture now in the care of those better equipped to develop and then defend it, let us now advance towards something else that Leibniz, given his starting point with the universal characteristic, would have needed to attend to. If we follow his

theory there will be indefinitely many possible combinations of mutually independent primitive concepts, but it seems impossible that all or even most of them should signify something. It seems that, faced with any combination at all, one who wanted to find something signified would have to look for some organizing principle, some larger scheme (where there is one) which would shape and coordinate these components.

33. Peter Remnant and Jonathan Bennett explain cupellation as follows (xxxv1): a process for separating out gold and silver from lead and other base metals. When heated 1n a cupel—a container moulded out of wood-ash and bone-ash—the base metals oxidize, the lead oxide

absorbs the oxides of the other base metals and 1s in turn absorbed by the bone-ash. A button of gold or silver or their alloy 1s left in the cupel. In the process of parting, the silver is separated

oft by being dissolved in aqua fortis—usually nitric acid. For more on gold see New Essays, pp. 267, 204, 312, 324, 338, 348, 354, 410—-12... 34. See for instance Elliott Sober, ‘Evolution, Population Thinking and Essentialism’, Philosophy of Science, 47, 1980, pp. 350—83.

THREE

MOMENTS

10§

In our actual practice, the problem just mentioned need never arise. For in our actual practice, we start out not from a collocation of primitive ideas but from a single complex idea—human being, say, or person or know or believe or remember or whatever—and

a matching (however incomplete)

concep-

tion of that which is signified, in this context or that...by whatever word or conception. That conception will not normally, however, have the form or content of a would-be analytical definition. Our understanding of “know” for instance will have neither the form nor the content of (say)

“justified true belief”. There would be a sort of justice in saying that that famously defective proposal belongs within the same Leibnizian tradition which has dominated linguistic philosophy for so long. But that is the tradition that is here under question. Suppose, for instance, that instead of trying to understand knowledge, seen no longer as justified true belief or in any similarly dissective way, we conceive knowledge, with Timothy Williamson, as one fundamental kind of mental state, the state sensitive in a whole variety of specifiable ways to the knower’s environment.”> Suppose similarly that we conceive experiential memory not as a complex of simpler ideas—such as recall, experience, identity etc.—but as a whole capacity by which we retain information about our own past doings and experiences enabling us to retrace them, if or when we wish, all the way up to the present.” Once we take ourselves to be possessed of larger conceptions such as these, we can speculate if we wish what needs or interests of ours they subserve. Indeed we can speculate what would have happened to us if we had never had such and such a concept. In that way, we can begin to appreciate the concept for what it is, with the shape that it has. But that is precisely not to appreciate it as a conjunction or disjunction of simpler or underlying elements. Their contributions will be consequential upon that shape. 15. Pursuing this line of opposition to Leibniz, what should we ourselves say is the terminus of definition by analysis? Loosely speaking, I should say there is no terminus. Despite the fact that we

cannot

without

remainder

dismantle

a concept

into simpler concepts,

the lexicographer can still write an entry for the term; a philosopher can still try to enlarge our comprehension of its meaning and try to understand 35. See his Knowledge and its Limits 36. Compare Gareth Evans, The Chapter 7, especially pp. 190—1; Continnants: Their Activity, Their

Oxford, OUP, 2000. Varieties of Reference (Oxford: Oxford University Press 1982), David Wiggins, Sameness and Substance Renewed, pp. 211—25, and Being, and Their Identity pp. 27—9, 93—4.

106

THREE

MOMENTS

better the point or interest with which we apply it. Philosophy’s work here is to elucidate.”” As for philology, the fascinating and wonderful work of this science is to trace, inter alia, the transmission from language to language of certain roots or morphemes

or certain modes of construction or combin-

ation. Philosophy, comparative philology, linguistics. . .are all in some sense analytical ventures, but neither in and of themselves nor by their achievements can they confirm Leibniz’s account of semantic meaning as underlain by simple concepts of completed distinctness. On this basis, one may want to say that analysis does have a terminus, but the terminus is the host of ordinary terms (e.g. ‘know’, ‘remember’,...) for

whose applicability we can rarely or never offer illuminating, strictly noncircular necessary and sufficient conditions. In a sense, such terms are primitive, but, pace Leibniz, they have numerous marks. If I know that the sun is

shining, then it is true that the sun is shining. I am right. If I remember (personally remember) seeing the Venus de Milo for the first time in 1949, then it is I who did indeed do just that in 1949. But from these implications, it doesn’t follow that, for some X, knowledge consists of X plus truth. It doesn’t follow that, for some Y, x’s remembering doing-whatever-it-was consists of Y plus x’s having been the person who did whatever-it-was.>* We have no assurance that there is any such concept as that of knowing less (minus) truth or of one’s remembering one’s V-ing less (minus) one’s being

the very person who did whatever-it-was.We learned the terms ‘know’ or [experientially] ‘remember’. . by coming to see the place they occupy and the office they fulfil in the world of signs and significance (Sinn) that comes into being as thinkers such as we are combine to explore and then colonize the world. For that reason, philosophical elucidation of such terms will succeed best in its aims by identifying the whole interest with which they are applied. It is not something to take for granted that, if the analysandum demands the satisfaction of these or those apparently simpler concepts, then the analysandum less (minus) this or that requirement will make sense too. 37. See section 6, para. 6. For the early history of the 1dea of elucidation, see first B. A. W. Russell and A. N. Whitehead, Principia Mathematica (Cambridge: Cambridge University Press, 1910), Part [, section A, *1:‘The primitive i1deas are explained by means of descriptions intended to point out to the reader what is meant; but the explanations do not constitute definitions, because they really involve the ideas they explain’; see also L. Wittgenstein, Tractatus LogicoPhilosophicus (London: Routledge & Kegan Paul, 1922), 3.263. For recent discussion of elucidation, see my Needs, Values, Truth (Oxford: Oxford University Press, 1987, 1991, 1998, 2002), pPp- 142, 314. See also P. E Strawson, Analysis and Metaphysics, op. cit. 38. For the claim that Q-memory as defended by Derek Parfit and Sydney Shoemaker is 1ll-made and 1ll-defined, see Evans as cited in note 36 and David Wiggins as cited in that note.

THREE

MOMENTS

107

Only a Leibnizian theory of simpler and simpler concepts can deliver that questionable result or insist that the process of cancelling components to a given complex of requirements is bound to produce something that still makes sense or means something.> Is this not the moment then to abandon altogether the thought, not yet defunct, that it is the existence of semantic primitives that undergirds the business of lexical definition? Decomposition must indeed further the process of lexical defining, but that is not to say that giving definitions in the most primitive terms one can find will promote the purpose of extending or consolidating speakers’ powers of utterance or comprehension. See the enumeration of kinds of definition given in Part I, section 6. Decomposition is not central or essential for clarity in Leibniz’s sense. Of course, language as we know it depends on a mass of elementary meaningful units, lexemes, morphemes, roots, and other devices which (as Plato might have said) weave

in and out or fly here, there, and everywhere. The interest of these things is not however that they are a starting point for the Leibnizian hunt after the semantically simpler and then the utterly simple. They are interesting because they are there in the language and play a special role there. They are interesting because they promote—and help to explain—the productivity of natural language, both in general and in detail, but not because they can illustrate or confirm the Leibnizian account of semantical meaning.*’ REFERENCES

Aristotle De Anima. Baldwin,

Thomas

1990 G. E. Moore, London and

New York: Routledge.

Bodemann, Eduard ed. 1895 Die Leibniz Handschriften der Koniglichen

Offentflich

Bibliothek zu Hannover, Hanover: Hahn. Bolinger, Dwight 1965 ‘The Atomization of Meaning’, Language 41 (4), §55—73-

Burnyeat, Couturat, Couturat, Dummett,

Myles 2001 A Map of Metaphysics Zeta, Pittsburgh, PA: Mathesis. L. 1901 La Logique de Leibniz, d’aprés des documents inédits, Paris: Alcan. L. ed. 1903 Opuscules et fragments inédits de Leibniz, Paris: Alcan. Michael 1991 Frege and Other Philosophers Oxford: Clarendon Press.

Frege,Gottlob 1891 Letter to Husserl (24.5.1891),in Gabriel et al.eds Wissenschaftlicher Briefwechsel, Hamburg: Felix Meiner, 1976. 39. For a fresh neo-Leibnizian stand, see Anna Wierzbicka at p. 140 of her ‘Replies to Discussants’ in Dictionaries: Journal of the Dictionary Society of North America,Volume 14, 1993. 40. In this chapter I am indebted in various ways to Wilfrid Hodges, Patrick Hanks, Ian Rumfitt, and Peter Hacker for things they have said to me over the years. As the chapter proceeds, readers of Hidé Ishiguro’s Leibniz’s Philosophy of Logic and Language will swiftly recognize the nature and extent of everything that I owe here to this remarkable work.

108

THREE

MOMENTS

Frege, Gottlob 1892 ‘On Sense and Reference’ in Max Black and P. J. Geach Translations from the Philosophical Writing of Gottlob Frege, Oxford: Blackwell, 1970. Frege, Gottlob 1894 Review of Husserl’s Philosophie der Arithmetik I, in Zeitschrift fiir Philosophie und philosophische Kritik 103, 313—32;1in Brian McGuinness ed. Collected

Papers, Oxford: Blackwell, 1984. Gerhardt, C. 1. ed. 1875—90 Die philosophischen Schriften von G.

W. Leibniz, seven

volumes, Berlin: Weidmann. Grua, G. ed. 1948 Textes inédits de Leibniz, Paris: Presses Universitaires de France.

Hanks, Patrick 2005 ‘Definition’ in Keith Brown ed. Encyclopaedia of Language and Linguistics, Oxford: Elsevier. Hanks, Patrick 1987 ‘Definitions and Explanations’ in J. M. Sinclair ed. Looking Up: An Account of the Cobuild Project in Lexical Computing, London and Glasgow: Collins, pp. 116—306.

Hanks, Patrick and Judy Pearsall eds 1998 New Oxford Dictionary of English, Oxford: Oxtford University Press. Ishiguro, Hidé 1972 Leibniz’s Philosophy of Logic and Language, London: Duckworth; second edition, Cambridge: Cambridge University Press, 1990. Kneale, William 1966 ‘Leibniz and the Picture Theory of Language’, Revue Internationale de Philosophie, 20. Kneale, William and Martha Kneale 1962 The Development of Logic, Oxford: Oxford University Press.

Langford C. H. 1942 ‘The Notion of Analysis in Moore’s Philosophy’ in Schilpp ed. 1942.

Leibniz, G. W. von ed. 1903. Leibniz, G. W. von I1, vol. 1, p. 497.

1679a ‘Introduction to a Secret Encylopaedia’ in Couturat 1679b Letter to Vagetius (12.12.1679), in Leibniz

1923—, Series

Leibniz, G. W. von 1684 ‘Meditations on Knowledge, Truth and Ideas’ in Gerhardt ed. 1875—90, vol. IV, pp. 422—6. Leibniz, G. W. von 1704 Nouveaux essais sur 'entendement humain, in Leibniz 1923—, Series VI, vol. 6. Leibniz, G. W. von 1875—90 Letter to the Duke of Brunswick, in Gerhardt ed. vol. I, pp. s7-8.

Leibniz, G. W. von 1903 ‘Of an Organon or Ars Magna of Thinking’ in Couturat ed. 1903.

Leibniz, G. W. von 1923— Samtliche Schriften und Briefe, Berlin: Akademie-Verlag. Leibniz, G. W. von 1948 “Termini Simpliciores’, reproduced in part in Grua ed. 1948. Leibniz, G. W. von 1973 Philosophical Wrkitings, trans. Mary Morris and G. H. R. Parkinson, London: Everyman.

Leibniz, G. W. von 1981 [1704] New Essays on Human Understanding, trans. Jonathan Bennett and Peter Remnant, Cambridge: Cambridge University Press. Loembker ed. 1969 Leibniz: Philosophical Writings, Dordrecht: Reidel.

THREE

McDowell, John 86, 159—85.

1977 ‘On

the

Sense

MOMENTS

and

109

Reference

of a Proper

Name’,

Mind

Mates, Benson 1986 The Philosophy of Leibniz, Cambridge: Cambridge University Press. Moore, G. E. 1942 ‘A Reply to My Ceritics’ in Schilpp ed. 1942. Plato Cratylus. Plato Sophistes. Putnam, Hilary 1970 ‘Is Semantics Possible?’, Metaphilosophy 3, reprinted in his Mind, Language and Reality: Philosophical Papers, vol. 2, Cambridge: Cambridge University Press, 1975.

Rumfitt, Ian 1994 ‘Frege’s Theory of Predication:An Elaboration and Defense, with Some New Applications’, Philosophical Review 103, §99—637. Rumfitt, lan 2003 ‘Contingent Existents’, Philosophy 78, 461-81.

Rundle,

Bede

2001

‘Objects

and

Attitudes’,

Language

and

Communication

21, 143—56.

Russell, B. A. W. and A. N. Whitehead 1910 Principia Mathematica, Cambridge: Cambridge University Press. Schilpp, P. A. ed. 1942 The Philosophy of G. E. Moore, Evanston, IL: Northwestern. Sinclair, John and Patrick Hanks eds 1987 Collins Cobuild English Dictionary, London and Glasgow: Collins. Sober, Elliott 1980 ‘Evolution, Population Thinking and Essentialism’, Philosophy of Science 47, 350—83. Strawson, P. E 1959 Individuals, London: Methuen.

Strawson, P. E 1992 Analysis and Metaphysics, Oxford: Oxford University Press. Suppes, Patrick 1957 Introduction to Logic, Princeton, NJ:Van Nostrand.

Tarski, Alfred 1956 Logic, Semantics and Metamathematics, trans. J. H. Woodger, Oxford: Oxford University Press. Wierzbicka, Anna 1993 ‘Replies to Discussants’, Dictionaries: Journal of the Dictionary Society of North America 14.

Wiggins, David 1987 ‘“The Concept of the Subject contains the Concept of the Predicate’in J. J. Thomson ed. On Being and Saying: Essays for Richard L. Cartwright, Cambridge, MA: MIT Press, Bradford Books.

Wiggins, David 1992 ‘Remembering Directly’ in J. Hopkins and A. B. Savile eds Psychoanalysis, Mind and Art, Oxford: Blackwell. Wiggins, David 1980 Sameness and Substance, Oxford, Blackwell. Wiggins, David 2001 Sameness and Substance Renewed, Cambridge: Cambridge University Press.

Wiggins, David 2002 Needs, Values, Truth, amended third edition, Oxford: Oxford University Press.

Wiggins, David 2006 Ethics: Tivelve Lectures on the Philosophy of Morality, London: Penguin.

Wiggins, David 2016 Continuants: Their Activity, their Being, and their Identity, Oxford: Oxford University Press.

110

THREE

MOMENTS

Williamson, Timothy 2000 Knowledge and Its Limits, Oxford: Oxford University Press. Wittgenstein, Ludwig 1922 Tractatus Logico-Philosophicus, London: Routledge & Kegan Paul. Wittgenstein, Ludwig 1953 Philosophical Investigations, Oxford: Blackwell.

7 Locke: ‘The Great Conduit’ Summary John Locke claimed that to abuse language is comparable to breaking or stopping

the

pipes

whereby

clean

Cultural conservatives will applaud regarding

themselves

as empirical

water

is distributed

to

the

public.

Locke’s comparison. But linguists, scientists, may

prefer to ridicule

the

snobbish prescriptions and proscriptions by which reactionaries seek to “fossilize” the parlance of yesteryear. How well do the parties to this contest understand one another? Conservatives despise or ignore the linguist’s interest in the sprawling

actualities of everyday usage and speech-patterns. More reasonably, though, conservatives will also criticize the linguists’ particular brand of empiricism for refusing to recognize speaking and writing as an art—the art in which any language user can find a role for themself.

1. In a passage of his Essay Concerning Human

Understanding (I11. 11) from

which Jean Aitchison quotes at the beginning of her second Reith Lecture, John Locke writes: “Language being the great Conduit whereby Men convey their Discoveries, Reasonings and Knowledge, from one to another, he that makes an ill use of it...does as much as in him lies, break or stop the Pipes whereby it is distributed to the publick use and advantage of Mankind”. In this review, I defend Locke’s analogy against Professor Aitchison and then extend it. Meanwhile, en route, I identify the great Conduit itself with a

waterway which is still there to be found in North London. It is still useful. Professor Aitchison’s verdict on Locke’s idea is that it is “misleading”. She does not say this, as others might, on the grounds that, just as it stands, Locke’s sentence understates the role of language in the very framing of thoughts. Reading the sentence that she quotes as Locke’s definition of language—rather than as simply conveying one large and important truth about it—she finds what Locke says “misleading”. It is misleading (she says) because it makes no mention of the non-cognitive uses of language, such as

Meaning, Truth, and the Limits of Analysis: Ten Studies. David Wiggins. Oxford University Press (2022). © David Wiggins. DOI: 10.1093/0s0/9780198726173.003.0007

112

LOCKE:

‘THE

GREAT

CONDUIT’

“human grooming talk”,“caretaker speech”, or “solidarity calls”. (Compare, she suggests, chimpanzee pant-hoots.) That is not all that she finds wanting in Locke’s statement. She makes a further point: that language is singularly ill-adapted to convey how to tie a clove hitch or a bowline, for instance. “This patchwork of efficiency and inefficiency is fairly typical of behaviour that is biologically programmed.” The phrasing of this criticism of Locke is significant. The modern linguist sees herself not as a humanist, not as a mere grammarian, not as a participating member of the republic of letters (in a linguist that would be unprofessional perhaps), but as a scientist for whom language is one, however singular, aspect of human biology and psychology. A linguist’s proper role is to describe and explain the language-function in the manner of an empirical scientist. It is in this capacity that she asks: Is British English really changing for the worse, as some people argue? Of course it isn’t. Over a hundred years ago, linguists, those who worked in linguistics, the study of language, realized that different styles of language suit different occasions, but that no part of language is ever deformed or bad. People who dispute this are like cranks, who argue that the world is flat. Yet flat-earth views about language are still widespread...On inspection, the web of worries surrounding change turns out to be largely traditional, somewhat like the worries each generation has about its oftspring. Laments about language go back for centuries.

2. Conservatives pay little or no attention to the Reith Lecturer’s account of language-science. They accuse her of preaching that, in language matters, anything goes. They accuse her of encouraging ordinary people to forgo a large part of their precious birthright in the English language. She makes heavy weather of the question of what Standard English is. But the simplest way for her critics to circumnavigate trouble of this particular kind may be for them to say that the language they are concerned with is the language in which Aitchison herself gave her lectures—namely ordinary English. The points that they want to insist upon do not depend at all on the definition of “Standard English”. Their concern is with the speaking or writing of English as such. Few if any of these critics’ actual concerns are at all sensitive to dialect. (However incongruously, some of their claims could be illustrated by reference to Estuary English.) Conservatives, however, for their part, ask too rarely why the claims they themselves make provoke such hostility among so many people. But seeking to understand better the passions aroused here, we need look no further than the first sentence of Locke’s Book Three: “God having designed Man

LOCKE:

‘THE GREAT

CONDUIT’

113

for a sociable Creature, made him not only with an inclination, and under a necessity to have fellowship with those of his kind, but furnished him with Language, which was to be the great Instrument and common Tye of Society”. Had Locke foreseen the conflict in which the Reith Lectures represent one engagement—had he foreseen its connection with the incipient political conviction (clearly presaged in the Lectures) that every citizen must have his or her fair share of almost everything that is worth having, and of absolutely every quality that is to be publicly recognized as a moral, artistic, or civic virtue—then he could as well have written “the great Instrument, the common Tye of Society and its great Bone of Contention”. Here, in the background, is something to which conservatives may need to pay more attention.

3. If her critics have an imperfect grasp of the Reith Lecturer’s scientific concerns and her other concerns, then how well does she understand what

concerns her critics? She insists upon seeing the people she opposes as harking back constantly to some long-past moment in the perfection of the English language, as upholding the snobbish prescriptions of a pointless and obsolete linguistic etiquette, and repeating hackneyed complaints whose perennial familiarity ought long since to have prompted us all to see how groundless they are. [t is not to be denied that linguistic conservatives have sometimes looked back to an ideal state of the Janguage—as Samuel Johnson looked back at pre-Restoration writings as “wells of English undefiled”. But here let it be noted that Johnson himself, whom she sees as an ally, is the very same witness whom Aitchison herself calls upon to affirm also that “to enchain syllables and lash the wind are equally the undertakings of pride”. If Professor Aitchison

had wanted

to understand

the conservative

outlook, she would

have had to ask herself how Johnson could have been party to both of these attitudes. She makes things much too easy for herself if she supposes that linguistic conservatism is not only committed to the possibility of comparing the beauty and effectiveness of a language at different epochs—a possibility she seems to scorn but does not disprove—but indissolubly linked also with nostalgia for some imaginary epoch of timeless perfection. Are conservatives really so committed? Is that really what they are concerned with? Suppose that the conservatives preferred to say something quite else: that the way in which people think is all of a piece with the way in which they talk. It is a measure of the thoughtfulness or thoughtlessness of a time how many or how few human beings there are at that time who strive to make

114

LOCKE:

‘THE GREAT

CONDUIT’

that speaking or writing be a sort of thinking, or seek to make their every word or phrase be the choice to say this rather than that. Simply to announce such a position is of course to take speaking or writing for an art, and an art that promotes a strictly non-natural purpose. In the face of this statement of conservatism, the linguists will probably say that it is not for a professional linguist to approve such a purpose or to second any of the value-judgments that the conservatives make on this basis.To this, however, the conservative

reply must be that it is not for linguists to disapprove either. It is just a fact that there are people who embrace this non-natural purpose and see it as integral to human life. It is just a fact that such people tend (however inegalitarian this may appear) to think ill of those who have no time for such a purpose or aspiration. But linguistic conservatives, so concerned, do not and cannot try to ground either their outlook or their purposes in the psychoneural nature of language—any more than those who strive to make a kingdom of ends seek to ground their outlook in some proposition of sociology. They ground it in their aspirations—among which are aspirations for language as a vehicle of thought and its apt or graceful expression. It simply does not matter one way or the other whether these aspirations count for anything at all among those that have no share in them. 4. The conservatives’ next point must be this: if language is to subserve the non-natural purposes of talking or writing considered as an art, then language needs (inter alia) both power and precision; and power and precision depend everywhere not only upon the presentation of that which is clear and useful (let us resist the replacement of “as if”” by the ubiquitous “like”) but also on the husbanding of differences. (Consider the differences, say, between “may” and “might”,’ or “can” and “could”). Everywhere, conservatism says, the enemy is entropy. And among the several ways in which a language can husband differences is to respect and preserve certain norms. These norms are not simply regularities that lie open to the unblinking stare of the sociolinguist, but practical claims upon speakers that arise from what this or that expression means in English—means in the English to which the speaker holds answerable all his or her own speaking or writing. That language is changeable no doubt. It is a social artefact, an object or thing dependent upon constant use and constant renewal. But the thing that is special about it for the speaker who seeks to make use of it is not that it is 1. Consider the difference between “If Beckham had taken the corner, then a goal might have been scored” and “If Beckham had taken the corner, then a goal may have been scored”. (Was 1t?)

LOCKE:

‘THE GREAT

CONDUIT’

11§

ideal, but that, as it stands at the time of speaking or writing, this language is his or her language—an indefinitely divisible birthright by the acquisition and possession of which he or she can either draw upon or add to that which has up to that point been thought, discovered, or felt. s. Finally then, what about the “laments about language that go back for centuries. ..somewhat like the worries each generation has about its offspring’—the laments whose simple recurrence Professor Aitchison seems to think invalidates them? To understand better the real nature of such lamentations, let us go back to Locke himself, Locke the inveterate inquirer into engineering and mechanical devices. The great Conduit that he speaks of carries something wholesome to the public, and it carries it by pipes. It seems certain (I do not know whether this has been pointed out before) that the conduit Locke had in mind was the New River, completed in 1613. This was not a river but a channel for the conveyance of fresh water. The promoter of this channel was Sir Hugh Myddelton who, with the subvention of James I, brought the channel all the way from a chalk outcrop in Hertfordshire to an open basin he had lined with stone slabs near Sadlers Wells in Islington. The channel itself, ten foot wide and lined at various points with clay, wood, or other materials, was dug along a contour with a gentle fall of only eighteen feet from start to finish along its meandering forty-mile route. The basin at Islington supplied storage cisterns which distributed the clean fresh water to City subscribers by means of wooden pipes hollowed from elm trunks. Such a structure was in conception and execution a triumph of engineer-

ing. But at every point in its entire length it was liable to leakage, to collapse, and to contamination. On pain of its not conveying that which it was designed to convey, the New River stood in constant need of constant vigilance and repair, both major and minor. Among the major repairs, some were effected by the engineers of later times, who cut out troublesome sections by building aqueducts, eventually reducing the length of the river to twenty-seven miles. If there is to be a comparison between language and the conduit that was the New River, then the Lockean metaphor suggests that one of the things that language must need is unremitting maintenance and protection— maintenance subsuming improvement or enrichment. But this is only to suggest that the thing language needs at any point is for the norms to be upheld and sustained on which its effectiveness depends at that point. And how else can such norms be upheld, preserved, or maintained if we renounce all efforts of prescription or exhortation?

116

LOCKE:

‘THE

GREAT

CONDUIT’

6.To see the claim that in speech or writing not just anything goes—or to see the prescriptions that conservatives support as turning upon the timelessly ideal—is as silly, however, as insisting that, when a farmer keeps trespassers from his wheat field, he will refer to an ideal moment of land tenure, or claim that the field always has been and always ought to be his. The truth is that exhortation or advice about what means what in English says nothing much either way for or against change. In the here and now, it simply sustains prescriptions for the here and now which need not even mention change, let alone forbid it. Such prescriptions (to quote Johnson) serve in practice to “retard what we can not repel” and “palliate what we cannot cure”. Their effect is simply to make (or keep) room for the artistic, intellectual, and other purposes that are pursued in language. It is no surprise, moreover, if each age has backed its prescriptions with lamentation—just as it is no surprise if the process of education by one generation of the next gives rise to the expression of acute anxiety. Such anxiety is not always unjustified. Or is it to be supposed that the recurrence of parental anxieties proves that there is simply nothing about which parents or educators really need to worry or to concern themselves? Surely not. Such anxieties may be at once constructive and timely.

7. How then—once the nature of the dispute is clarified in this way and we see prescription as subserving the retention and renewal of whatever has served so far the divers purposes of speaking and writing—ought the suit to be decided between the conservative or prescriptive conception and Aitchison’s purely descriptive conception of language as a psychological function? If might is right, or if the issue is between the cheery populist tones of permissive descriptivism and the inky coughs and scrannel protestations of conservatives who counter insult with contumely, then the verdict will in practice go to Aitchison and her countless allies in the fields of education, politics, and communications. I think we can know these people by their works and products—among which I number each new generation of students for whom it is more difficult than it was for the last to read (‘in the

original’ as one might say) John Locke, Samuel Johnson, Adam even J. S. Mill. Nobody needs to take my word for this, however. It would be persons of good will to take neither side as the best authority on other side stands for. I would only plead that, in considering

Smith, or better for what the the Reith

LOCKE:

‘THE GREAT

CONDUIT’

117

Lectures, we should distinguish between descriptivism itself as the scientific programme within linguistics to which Professor Aitchison herself contributes (I have nothing to say about that) and the permissivism she has fallaciously deduced from that programme. Better still, having made that distinction, let any open-minded arbiter discover his or her preference between two careful English versions of some text less familiar and less contentious than the King James Bible—this one, for instance: When carrying the tablet of jade [the symbol of the rulers’ feudal investiture] he seems to double up, as though borne down by its weight. He holds it at the highest as though he were making a bow, at the lowest, as though he were profering a gift. His expression, too, changes to one of dread and his feet seem to recoil, as though he were avoiding something. When presenting ritualpresents, his expression is placid.At the private audience his attitude is gay and animated. When he held the jade tablet, he gathered himself in as if he would not cope with it. At the highest he held it as if he were bowing, and at the lowest as if he were making a presentation. His expression became serious and apprehen-

sive, and his feet were constricted as if his progress were hindered.

When he

presented ritual gifts, he wore a complaisant expression.At private audiences,

he seemed at ease.

The second of these translations from Confucius comes with the cachet of a great university press (1994).

The first comes from Arthur Waley

(1937).

Anyone who prefers this earlier version must ask what it was and what attitude to words and writing or speaking it was that made Waley’s incomparable version of the Analects possible. 8. Professor Aitchison’s Reith Lectures seem to perpetuate the division of science from letters—and they do so at a moment when it has begun to dawn upon many people, helped perhaps by some of the marvellous pieces of prose collected in John Carey’s Faber Book of Science, how many natural scientists are fully enfranchised, non-spectatorial participants in the republic of letters. In its new-found populism and political (would-be) correctness, I believe that the BBC has given its blessing to one of the least interesting and least constructive of the alternatives open to us in our practical attitudes towards language, science, and the arts of writing and speaking.

3 Languages as Things

in their Own Right’ Summary Chomsky has said that the notion of a public language is unknown to empirical inquiry and “raises what seem to be irresolvable problems”. This chapter takes an opposing view—namely the common-sense view which sees a language as something with a history, and the common possession as often as not of far-flung communities. It seeks to disarm arguments against this view.

1. There is a tendency nowadays for linguists, philosophers, and other theorists of language to dismiss the notion of an object such as the English language or the Polish language as simply mythical and of no interest to any serious science of language. Some deny that there are such things as languages (in the plural). “This notion [of a public language] is unknown to empirical inquiry and raises what seem to be irresolvable problems’, Chomsky said in a public lecture that I attended in London in 1994.”

My purpose here is to speak for an opposing view, which I shall call the common-sense view, to extend it a little, and to gesture towards some of its

theoretical implications for mind and language. This is the view—the ordinary 1. This chapter was written at the time of Chomsky’s Jacobsen Lecture in London in 1994. It was written mainly for the sake of discussion with colleagues in London. The publication in Mind 1995 of a printed version of what Chomsky had said in 1994 then encouraged me to revise and submit my essay to Philosophy vol. 72, 1997. For Chomsky see his ‘Language and Nature’, Mind, vol. 104, 413, Jan. 1995. Chomsky provides the starting point. But the aim here is not to try to assess Chomskian linguistics. It is rather to comment upon one Chomskian contention concerning natural languages considered as objects. For friendly reference to the present paper by one of its opponents, see Richard G. Heck ‘Idiolects’ in Judith Jarvis Thompson and Alex Byrne (eds) Content and Modality: Themes from the philosoply of Robert Stalnaker, Oxford: OUP, 2008. 2. ‘Language as a natural object’, Jacobsen Lecture, 23 May 1994, Part 1 of ‘Language and Nature’, op. cit. In this Mind version, the passage I quote occurs at page 24. See also pages 13, 29 (where Chomsky quotes Davidson to similar effect), 41, 48—9.

Meaning, Truth, and the Limits of Analysis: Ten Studies. David Wiggins. Oxford University Press (2022). © David Wiggins. DOI: 10.1093/0s0/9780198726173.003.0008

LANGUAGES

AS

THINGS

IN

THEIR

OWN

RIGHT

119

pretheoretical view—that a language such as English or Polish is one particular artefact, a public thing with attributes irreducible to the psychology of its speakers. A language (I say) is an instrument by which speakers may frame their thoughts—something not concrete, yet ‘out there’ and wide open to be encountered in the human world.A language is an object with an origin, a past, a present, and a future, not a thing essentially defined by a particular set of syntactic or semantic axioms, syntactic and semantic structures, or what-

ever. Rather it is a thing historically given, changing or changeable and possessed of a multitude of sentences that have, as a matter of history, such and such meaning at this, that, or the other time and place.3 Let it be clear that the

belief in things such as natural languages neither implies nor excludes the existence of the sort of thing that Chomsky prefers to recognize, namely a universal grammar given to us in the shape of innate syntactic principles. If, pace Chomsky, there are such things as the English or Polish language, then how are they singled out as objects of reference? Well, English is a language arising, under the influence of Norman

Germanic

French, from the West

language, Anglo-Saxon; it is the language possessed of many

forms and dialects spoken in the British Isles, North America, Australasia,

India, etc.; it is the language in which you can say ‘Fear no more the heat of the sun’, ‘Unto those who have shall be given’, or ‘Never say die’. That is a

definite description of the English language. If you disjoined statements, each having the form ‘at time t sentence s means you might perhaps arrive at a complex description that could apply to the English language. But any such claim will fall far essentialism that is espoused by yet other theorists who prefer

enough true that p’, then not help but short of the to see a lan-

guage as corresponding to a function from expressions to intensions which

can combine to define various possible worlds. On the view to be defended here, there will be little or nothing to be said for the denial that we share the language of English with Dr Johnson, the first lexicographer of that language. Even more arresting than that denial 1s Chomsky’s positive claim that really and truly there is just ‘one human language, with minor variants’. 2. What else seems immediately apparent about languages conceived in the broadly historicist fashion here advocated? On this common-sense view, a 3. Languages do not have to change. Unless you count as change the regularization of spelling or the sumple acceptance of new words, Old Norse (Icelandic) has barely changed in two millennia. But languages are by their nature changeable.

120

LANGUAGES

AS

THINGS

IN

THEIR

OWN

RIGHT

particular Janguage with its past and present semantic attributes, with all its achievements and latent resources, is something that influences normatively,

by its palpable presence in the social world, the linguistic strivings of children, adults, foreigners, poets, writers, politicians, and the rest. The judg-

ments that these and other speakers of English make of significance, appropriateness, accuracy and taste would need (if they were fully spelled out) to make irreducible reference to this very thing—this very social object or artefact. As it passes from one generation to the next, it can change of course—just as any object can.

It is a further and interesting question whether we should see English as something properly exemplified in the totality of its forms and dialects— even as the determinable red is exemplified in all its determinations or shades—or whether some ‘outlying’ dialects should be seen as deviations from more central forms (in which case the language is unequally well exemplified in its dialects). If the common-sense view sought to close such questions, it would be extending itself beyond its strength. In advance of a more comprehensive theory, the view does not even exclude the possibility that two distinct languages (Russian and Ukrainian, say, or Provengal and Catalan) might each count among their own past forms some one language.® Nor again is it any part of the ordinary view that all speakers of a language, no matter when or where they live, should be accessible to one another. The ordinary view is free to acknowledge, moreover, that writing

has long since been one part of the account of how speakers gain access to one another’s meanings, and an indispensable part of the process by which languages are propagated and perpetuated. One of the roles of dictionaries—English dictionaries, say—is to put a larger and larger semantic store—and a huge store of English utterances written or spoken—at the common disposal of speakers. 3. Among those who deny the existence of any such object and insist that the thing to talk about is not this or that language, but rather language taken as a mode of psychological functioning shared among human beings, many 4. To conceive that one variant could be a variant of two distinct languages is to conceive that the individuation and differentiation of L and L, need not depend upon some locally evaluable discontinuity among variants of languages L and L . Compare the differentiation of the determinables red, orange and yellow. For determinables and their determinations or determinates, seeW. E. Johnson, Logic volume I, pp-

174 ff. For the ontology

of colours

and

their hues, see my ‘Verbs

other Modes of Grammatical Combination’, Aristotelian 1985—0, revised and absorbed here as Chapter 5.

Society

and Adverbs, and

Proceedings,

vol

some

LXXXVI

LANGUAGES

AS

THINGS

IN

THEIR

OWN

RIGHT

121

will emphasize the quantity and quality of the specialized neural and cerebral provision that underlies all human language aptitude. What they say about that provision itself is both arresting and interesting. Another interesting contention is that humans have an instinctive propensity for language which renders speech ‘contagious’ (as some say) among human beings. But none of this will show that human speech can be properly or adequately described otherwise than by reference to some public language, or show that the capacity for speech can be adequately conceived otherwise than as realized in some particular version or dialect of some particular language. 4. Nobody denies that human languages, once we take them seriously and in the plural, confront us with formidable problems of identity and individuation. But so do all sorts of other artefact. Languages are not concrete. They are the common possession of far-flung communities of speakers. They form part of a larger compact between the dead, the living, and the as yet unborn. But each of these features can be

matched mutatis mutandis by other human artefacts or institutions. On the common-sense view, there is no more reason to allow the problems of identity and individuation of distinct languages to dissuade us of the reality of natural languages than there is to allow the problems of identity and individuation affecting pictures, canals, ideograms, codes, laws, deliberative

assemblies, symphonies, texts, sewing machines, bicycles, or combineharvesters to persuade us that there are no such things. 5. Now another matter. Some linguists like to echo Weinrich’s saying that a language is simply a dialect with an army and a navy. The common-sense view, so far from shrinking from what this claim actually says, will be positively eager to accommodate the fact that armies and navies and other instruments of domination loom large in the histories of particular languages and confer a special role upon certain dialects. For common sense is just as eager to accommodate historical contingency as it is to do justice to

norms of ordinary language use. What its upholder will rather protest against is another suggestion, supposedly following from Weinreich’s, namely that, by parity of reasoning, a dialect is simply an idiolect with a goad or a gun (in a sense of ‘idiolect’ to be defined by the theorist) or the further suggestion some theorists make, that even a whole language such as English, if we deign to recognize it at all, is best understood bottom up as a vast aggregation of personal idiolects. The commonsensicalist will protest how difficult it would be to rewrite every fact about English—e.g. (at random) what

122

LANGUAGES

AS

THINGS

IN

THEIR

OWN

RIGHT

‘precarious’ means in English and why it means that—as a fact about individuals trying to influence other individuals. Rather he will insist that the pressures that strive for linguistic domination reflect rival but complementary perceptions of what is in fact a social and public thing, namely the language that is in question, English or Spanish or Mandarin or whatever it is.’

6. So much for what the common-sense conception seems to amount to and the claims by which it makes itself almost invisible to Chomskian theory. Those who reject this common conception may now ask what else there is to be said positively in favour of it. But here let us propose a counterquestion. On the view that dispenses with languages as social objects, what are the efforts of ordinary speakers aimed at? A helmsman steers for port. A doctor aims to cure. What does a speaker aim for? He aims to get across a certain thing. He aims, if possible, to say this thing and to be understood as saying it. If there were no such thing as a public language, how would we mark the difference between (a) someone’s simply uttering words in their ideolect and (b) their being understood by an audience as having said something and gone on record to a certain effect? Simply to point to this difference is scarcely conclusive for the commonsense view. But here let us pass on to something more immediately tractable, namely how benignly (despite the individuative responsibilities that it incurs) the common-sense conception of natural languages as social objects can advance all sorts of vexed questions—questions about speakers’ knowledge, the role of semantic compositionality, and the nature of the lexical reality to which a grammatical account of English needs to be answerable. The answer that the common-sense conception offers to such questions is this. The reality to which theories concerning a particular language L have to be made answerable is of course the language L itself. Grammars of L, like systematic determinations of the extension of ‘true sentence of L', do not have to aim to recapitulate the ‘implicit knowledge’ or subpersonal linguistic routines that a theorist might attribute to speakers and/or hearers of L. That to which grammars, truth-definitions and the nascent practical abilities of L-speakers have all, in their different ways, to be answerable (at this time or that) are the public facts about what means what in L and how and why. These facts float free from the neural-cum-cognitive processes that

s. This 1s to say that the common-sense view is not only anti-essentialist. [t is also historicist.

LANGUAGES

AS

THINGS

IN

THEIR

OWN

RIGHT

123

underlie those abilities. They float free of idiolects. They float free of theories of the speaker-knowledge that is implicit in linguistic competence. 7.1 conclude in further defence of one central contention of the autonomy theorist. The contention is that, if we omit from the account of linguistic communication all mention of the public language in which speech is conducted—if we take this piece away from the scene where utterances occur and things are said—then we leave too little room for normative considerations or informed judgments to the effect that uttering sentence s qualified the utterer there and then as saying that p and going on record to that effect. How could any locus remain for judgments of error/ineptness/aptness/ excellence in an utterance? Linguists and other social scientists who feel that they have to oppose or obstruct the claims of the autonomy theorist run the risk of ending up—whether they wish to or not—with nothing much to say except that ‘provided enough people will now go along with it, anything goes’. On these terms however there will be no question of saying seriously and objectively that such and such an utterance is apt or inept, well-made or ill-made, not in good English, or just plain wrong. There are theorists—I am not referring here to Chomsky—who seem to believe this. But nobody who actively and energetically and carefully speaks a common language could possibly accept it. What the common-sense conception says—and what the autonomy conception and common sense say—is that, even to an audience that knows no better, you cannot utter the sentence ‘the sonatas are just the anecdote to a gloomy summer’s day’ and count as having stated that the sonatas are an antidote to such a day, or utter the sentence ‘a wind machine is used to emulate breathing at the start and finish of the work’, and count as having stated that a wind machine is used there to imitate or simulate breathing. That will only become possible when the day dawns on which so many people use these words in this way, and they use them thus so deliberately, and they are so multiply and densely defended

and reinforced in these choices

of words, that that is what

these

words do mean in the sort of English in which they are speaking and

writing.® 8. My restatement of the autonomy view is complete—except in so far as the preliminary defence of the view has incurred certain further responsibilities. One who measures up to these responsibilities, which will be 6. For the examples, see K. Amis, ‘Getting it Wrong’, in L. Michaels and C. B. Ricks (eds) The State of the Language, (University of California Press, 1980).

124

discharged graph 1 of viduation important express by

LANGUAGES

AS

THINGS

IN

THEIR

OWN

RIGHT

here only in strict subordination to the aims specified in parasection I, ought probably to say something more about the indiof artefacts such as languages. But at this point it is far more for us to address special doubts that the Chomskians sometimes saying such things as this:

‘Italian’ has no real world denotation in the technical sense [cp. 48—s1].

At this point, as if in anticipation of the general line I am preparing to take about reference and identity, Chomskians and other sceptics in this matter may however momentarily join forces. They may offer general reasons to discourage one from proceeding on the kind of assumption one might otherwise make, to the effect that words for non-natural objects, words such as ‘London’ or ‘Jerusalem’ or ‘Chinese’, belong to some public language [in which]... we might try to sharpen meaning and ideas for conditions under which the presuppositions of normal use do not hold [21].

In the same lecture from which I quoted at the outset, Chomsky said Even the status of (nameable) thing, perhaps the most elementary concept we have, depends crucially on such intricate matters as acts of human will, again something understood |by human subjects] without relevant experience determined by intrinsic properties of the language faculty and others. A collection of sticks in the ground could be a (discontinuous) thing—say, a picket fence, a barrier, a work of art. But the same sticks in the ground are not a thing if left there by a forest fire [22].

Chomsky goes on to remark that ‘space-time continuity has no particular relevance to these issues, contrary to what is sometimes assumed’. (For that which seems right in this remark and that which seems wrong, see my Continuants 2016.) But for Chomsky it is part of his build-up to further and increasingly demanding conclusions. Let me mention four of his contentions: (1) that ‘it is hard to extricate [from the investigation of how it can be

that speakers agree about how to understand substantives and names] very much that might be subjected to naturalistic inquiry’ [23]; (2) that we should ‘drop the empirical assumption that words pick out things apart from particular usages’ [24]; and (3) that we should not ‘assume that expressions pick out things intrinsically’. This last is something a conceptualist of my persuasion can either agree with, or at least—for lack of explanation of what this ‘intrinsically’ means—simply let pass. Then however there is a further conclusion (4), which Chomsky puts like this:

LANGUAGES

AS

THINGS

IN

THEIR

OWN

RIGHT

125§

A good part of contemporary philosophy of language is concerned with analysing alleged relations between expressions and things, after exploring intuitions about the technical notions ‘denote’, ‘refers’, ‘true of’, etc, said to hold

between expressions and something else. But there can be no intuitions about these notions, just as there can be none about ‘angular velocity’ or ‘protein’. These are technical terms of philosophical discourse with a stipulated sense that has no counterpart in ordinary language—which is why Frege has to provide a new technical meaning for ‘Bedeutung’, for example. If we rerun the thought experiments with ordinary terms, judgments seem to collapse, or rather, to become so interest-relative as to yield no meaningful results...It is sometimes maintained that such technical notions are required to account for communication. The belief is groundless [however] ... ‘Chinese is the language of Beijing and Hong Kong but not Melbourne’. . . is true but [the term] ‘Chinese’ has no real world denotatum [24—s5 italics not in the original].’

At this point it is becoming clear that, by the standard which Chomsky proposes, a passable relation of reference of the kind envisaged in philosophical semantics between substantive and thing—e.g. ‘English’ and English—would have to consist of the word’s picking the thing out ‘intrinsically’. But what does he mean by ‘intrinsically’? It seems that, according to Chomsky, it is a necessary condition of an intrinsic relation’s obtaining that the kind of thing in question should be (or belong to a kind) singled out by one of the serious ongoing sciences (i.e., presumably, some science methodologically continuous with physics, biology, physiology or biochemistry . . .) It seems that, according to Chomsky, it is the interest-relativity—and the “non-intrinsic”’ character—of the ordinary concepts that we have for singling out such things as public languages—or cities, dwellings, tools, stuffs, or artefacts—that debars the conceptions that generate these concepts from counting as objects of genuine reference. Only a seriously ‘naturalistic’ or scientific interest can validate that. Or so Chomsky would seek to persuade us. 9. Such is the background to the Chomskian denials which I am concerned to counter. (See again section 1.) I can agree with him in finding many of the claims of philosophical semantics overblown.We can agree in treating ‘Bedeutung’ as a technical term that needs to establish its credentials as a technical term. (But by its successes, it has gathered those credentials, I believe.) 7. The reference to Chinese 1s a trap for the unwary. As Chomsky well knows, Mandarin is not, on any responsible account, the same language as Cantonese. No language 1s the common language of both Beijing and Hong Kong. But Mandarin (Beyjing) is widespread and a real world

denotatium—according

to the common-sense

or autonomy

view.

126

LANGUAGES

AS

THINGS

IN

THEIR

OWN

RIGHT

We can agree in finding something special and impressive in the ontological commitments that we incur by the pursuit of science, and agree in refusing to apply tests of ontological commitment to sentences other than those that we are seriously and justifiably content to be committed to affirm.® Yes. How then, agreeing about so much else, can we have come to such utterly different conclusions? We chiefly diverge by taking different attitudes towards concepts that are interest-relative (that is relative to some non-*scientific” interest) and/or by cultivating different attitudes towards conceptions from which ‘it is hard to extricate anything that might be subjected to naturalistic inquiry’.A theoretical defender of the common-sense view will say here that the fact that such concepts and conceptions are interest-relative (and in some cases rela-

tive to acts of collective human will—as our conceptions of artefacts frequently are) does not imply that it is a simple matter of human choice or will whether a given object does or does not answer to (or fall under)

a

given concept. It is surely up to us collectively how to conceive of an eggtimer or a house or a statutory instrument or perjury or contempt of court or legal possession. But, once that conception becomes sufhiciently clear for us to raise the question whether or not a given x falls under the concept in question, it is not up to us whether the answer is yes or no. Interest-relativity affects sense in that way. It need not in that same way affect reference

or truth-value. That, in a word, seems

to be the chief

point at issue. | think that, if I could convince Chomsky or his followers of this, then there would be some hope of convincing them further that, in order for reference and satisfaction to do what reference and satisfaction need to do in philosophical semantics, they never needed to be what they call ‘intrinsic’ relations. What is more, reference and satisfaction do not need to be programmatically confined to those developing sciences by which we are to discern the features of natural reality. The sciences themselves start from pre-scientific conceptions many of which they leave substantially intact and undisturbed. Consider for instance a concept equally vital to ordinary life and ordinary biological science, namely human being:

8. This 1s not to say that the common-sense conception will let pass the question of what the recipe must be for dispensing with the sentence “Chinese 1s the language of Beijing but not Melbourne”. Surely we cannot allow this or similar facts simply to disappear without trace from an objective world-view.

LANGUAGES

AS

THINGS

IN

THEIR

OWN

RIGHT

127

If we abstract from the perspective provided by natural language, which appears to have no names in the logical sense..., intuitions collapse: Nixon would be a different entity I suppose, if his hair were combed differently.”

That won’t do. Our intuitions do not collapse with respect to the individuation of persons—uwhatever Nixon does with his hair. The ordinary conception of person—either practically or theoretically considered—remains. Often too the biological sciences build outwards from the pre-scientific conception.At this point I shall venture a counter-claim to Chomsky’s; any discourse that can make a non-arbitrary distinction between truth and falsity—any kind of inquiry that can lay claim over and over again to put us into a position to come to believe that p and do so precisely because p—can aspire to discern some particular aspect of some reality. There is nothing outré or strange in this. Over and over again, scientific naturalism begins from ordinary notions that it needs not to despise or abandon but simply to develop. 10. So much for the position I need to advocate. So much for the position I need to oppose, and so much for the Chomskian rejection of any serious ontology of human artefacts, social institutions etc. But what more needs to be said about languages in particular? Particular languages, [ claim, influence normatively by their presence in the social world the communicative efforts of speakers. He who speaks or writes goes on record (we have said) in a public language as saying this or that. But what language? Which language? Well, it is normal, or was normal until recently, for speakers and writers to have rather precise intentions with respect to their choice of language. They will aim to speak or write not just English (say) or twentieth-century English, but that particular sublanguage of their contemporary English that meets certain extra conditions— colloquial English, nautical English, diplomatic English, or whatever. Consider too the special position and intentions of Dante in the development of the Italian language. Consider further Mark Twain’s ‘explanatory

9. Noam Chomsky in Philosophical Topics 20.1 (1992), page 226. If | had known earlier of this exchange between Chomsky and Putnam, it would have expedited my opposition to Chomsky’s conception. In the event, there i1s a pleasing complementarity and/or convergence between Putnam’s and my reactions to his position. Another complementarity is the view that [ have long held in common with Putnam, namely the distinction between ordinary realism and what Putnam calls “metaphysical realism” according to which all objects of genuine reference individuate themselves (so to speak) independently of the intelligence of the thinker. See my Continuants op.cit, pages X111-Xvil, 199-200, 218-219.

128

LANGUAGES

AS

THINGS

IN

THEIR

OWN

RIGHT

note’ for Huckleberry Finn, which was a part of his literary endeavour to get as far as he could from the ‘showiest kind of book talk’: In this book a number of dialects are used, to wit: the Missouri negro dialect;

the extreme form of the back-woods South-Western dialect; the ordinary ‘Pike County’ dialect; and four modified varieties of this last. The shadings have not been done in a haphazard fashion, or by guess-work; but painstakingly, and with the trustworthy guidance and support of personal familiarity with these several forms of speech.

In this passage we are reminded that, in addition to any philosophical or theoretical problems there may be about language individuation, there are more particular questions about which sentences count or do not count as belonging to the sublanguage that the speaker or author has in mind.We are also reminded that there are recognized ways of determining the answer to the question. But to understand either the sophisticated use of unsophisticated language or sophisticated language itself (which deserves no less attention, surely, than unsophisticated), it seems we need to recognize both the

determinable and the determinate or determination, both English and the Pike County dialect. The determinable is a stable enduring albeit changeable thing. It exists in its determinations. But its determinations exist as determinations of that determinable. Here, in pacificatory spirit, let me enter one concession: that to find one-

self needing to speak of sublanguages of a language (both in connection with languages and their dialects and in connection with choice of special colorations) is to make it evident that neither languages nor their sublanguages can purport to be things like substances. But that is no reason to deny that languages are artefacts—things that human beings make, remake and refer to over and over again.

9 Peirce: Reflections on Inquiry and Truth arising from his Method for the Fixation of Belief Summary This chapter begins from Peirce’s essay ‘On the fixation of belief” and shows how it leads outwards into a large body of Peircean claims concerning inquiry, the ‘secondness’ of experience, abduction, hypothesis, perception, and truth. It stresses the importance

of Peirce’s insistence that, when

any belief of ours is found to be determined by a circumstance extraneous to

the

facts, that belief is weakened

or abandoned.

Only

abduction,

as

Peirce characterized it, can expand our understanding from the known to the unknown.

In conclusion the chapter offers a Peircean critique of Hume’s understanding of “the problem of induction”. Another conclusion is that Peircean pragmatism, properly construed, is consistent with a full-blooded realism. That which is distinctive in Peircean pragmatism is rather Peirce’s insist-

ence that truth and inquiry are correlative notions, intelligible only as correlative with experience.

My

paper of November

1877, setting out from the proposition that the

agitation of a question ceases when satisfaction is attained with the settlement of belief...goes on to consider how the conception of truth gradually develops from that principle under the action of experience, beginning with willful belief or self-mendacity, the most degraded of all mental conditions; thence arising to the imposition of beliefs by the authority of organized

society; then to the idea of settlement of opinion as the result of fermentation of ideas; and finally reaching the idea of truth as overwhelmingly forced upon the mind in experiences as the effect of an independent reality. CP 5.504, ‘Basis of Pragmatism’ 1906. [italics not in original]

Meaning, Truth, and the Limits of Analysis: Ten Studies. David Wiggins. Oxford University Press (2022). © David Wiggins. DOI: 10.1093/0s0/9780198726173.003.0009

130

PEIRCE

The third philosophical stratagem for cutting off inquiry consists in maintaining that this, that, or the other element of science is basic, ultimate, independent

of aught else, and utterly inexplicable—not so much from any defect in our knowing as because there is nothing beneath it to know. The only type of reasoning by which such a conclusion could possibly be reached is retroduction. Now nothing justifies a retroductive inference except its affording an explanation of the facts. It is, however, no explanation at all of a fact to pronounce it

inexplicable. That, therefore, is a conclusion which no reasoning can ever justify or excuse. CP

1.139 “The First Rule of Logic’ 1899.

Abduction consists in studying facts and devising a theory to explain them. Its only justification is that, if we are ever to understand things at all, it must be in that way. CP 5.145 ‘Harvard Lectures on Pragmatism’ 1903.

[Scientific procedure]| will at times find a high probability established by a single confimatory instance, while at others it will dismiss a thousand as almost worthless. Frege (1884) Foundations of Arithmetic, trans.J. L. Austin: 16

‘The Fixation of Belief” was published in 1877 as a popular essay. But Peirce must have attributed to it not simply the literary felicity that we find in it, but high philosophical importance. For in the ensuing decades he constantly returned to this paper as a focus for the clarification of his thoughts, either entering corrections and amplifications or else adapting it to new philosophical initiatives. Some of the amendments were designed to adjust the essay to the projects of “The Grand Logic’and “The Search for a Method’. Our chief concern here will be with the essay as Peirce came to reread and rewrite it, rather than with the essay in its exact original condition.The first of our epigraphs, which is dated 1906, is surely the product of one of these rereadings. Not only does ‘Fixation” appear at least as important as Peirce supposed. There radiate from it some of the grandest themes of modern philosophy— the nature

of truth, for instance, and the relation truth has to meaning

when meaning is operationally or pragmatically conceived; inquiry and the ethics of belief; the epistemic status of perceptual experience;and the proper aspiration or aspirations of hypothesis. Once ‘Fixation’ is seen in proper

PEIRCE

I31

conjunction with other Peircean claims into which it leads, the essay will even promise a line of response to Hume’s much celebrated doubts about the rational basis of our efforts to argue from the known to the unknown. What has prevented philosophers from investing the paper with the sort of importance in connection with truth that we find Peirce attributing to it in our first epigraph? Maybe the tendency to read the paper itself as a phase in a one-issue philosophical campaign to demystify the idea of truth by redefining it as the eventual, if not predestinate, opinion of those who openendedly and resolutely pursue the business of inquiry. But only for a small minority who still espouse some sort of verificationism or “propepositivism” (Peirce’s term) could such a campaign of redefinition be either convincing or interesting.

Recently, other Peircean scholars' have pointed to the implausibility of attributing to the exponent of a theory of signs and signification as special as that of Peirce the project of offering an analytical decomposition of the concept of truth (or of any other basic concept). In the light of this doubt, it will no longer do to suppose that, at any time (let alone in 1906, by which point he had recanted the worst exaggerations of ‘How to Make Our Ideas Clear’ (1878)), Peirce would have positively approved a report, given in our language as used by us, to the effect that Charles Sanders Peirce thought that an opinion’s being true and its being the eventual opinion were simply, analytically, or necessarily one and the same thing. Not only is that a questionable report. Unless some quite peculiar sense is attached to “the eventual opinion”, it appears inconsistent with that which we read towards the end of “The Fixation of Belief” and read again in Peirce’s subsequent reports of its content (e.g., that which I have cited from 1906). It is time then to supersede the form of words that Peirce took the risk of using when he wrote so rashly: “The opinion which is fated to be ultimately agreed by all who investigate is what we mean by truth, and the object represented in this opinion is the real” (CP 5.407, 1877). If Peirce’s ideas are to reach again into the bloodstream of philosophy, then we need not only fresh studies of his texts but speculative 1. See Cheryl Misak Truth and the End of Inquiry: a Peircean Account of Truth (Oxford: Oxford University Press, 1991), whose Chapter 1, for instance, alerts us to the consistently and strictly pragmatic signification of Peirce’s use of words such as “mean”. For good or ill, my essay prescinds entirely from Peirce’s theory of signs and signification, transposing discussion to the familiar language of sense, reference, proposition etc.

132

PEIRCE

transpositions of these ideas—transpositions recognized as speculative but given in language that can be understood without any reference to special Peircean terminology. So, to the extent that our chosen vehicle of expression is modern philosophical language, let it be our policy to confine ourselves to such portions of that as are securely cantilevered from the everyday language of those who are wont to listen to what they are saying. Under the transposition I shall propose in this paper, the exact meanings of “ultimately” and “end of inquiry” will no longer be so important, and the relation between truth and inquiry will prove slightly (but only slightly) more indirect than Peirce was accustomed to allow. There is another Peircean preoccupation we shall abandon. This is Peirce’s thought that proper conduct of inquiry not only comprises the motive and means for correcting its own conclusions but is bound in the long run to iron out every error. This perilous claim will be no part of the core Peirceanism that is here to be consolidated and defended.

II Peirce says in ‘Fixation’ that, with respect to any question that concerns us, belief or opinion is the state we seek to attain, whereas doubt (not knowing what to think about this or that) is the disquieted, dissatisfied state that we seek to end. The essay reviews four different but developing ways of tempering that disquiet: the method of dogmatism or tenacity, the method of authority, the a priori method, and the method of experience, which Peirce himself approves and commends to his reader. This last method embraces logic, in the broad nineteenth-century sense of the term: the “distinction between good and bad investigation...is the subject of the study of logic....Logic is the doctrine of truth, its nature and the manner in which it 1s to be discovered” (CP 7.320-1, 1873). In Peircean usage, logic is the general art of reasoning—nothing less than everything which ‘“The Fixation of Beliet’ opens out into—and it subsumes the art of making inferences from the known to the unknown. Logic embraces not only deduction, not only induction, which is the testing of hypotheses, but also abduction, which is the framing of explanatory hypotheses. “Reasoning is good if it be such as to give a true conclusion from true premisses and not otherwise [good]”, Peirce wrote (CP 5.36s, 1877). Later, he amended this sentence to

PEIRCE

133

say “Reasoning is good if it be dominated by such a habit as generally to give a true conclusion from true premises” (CP 2.11, 1902). Nineteenth- and twentieth-century conceptions of the province of logic are interestingly different. Each conception insists, however, on the incompleteness of the canon that we shall recognize at any point for good reasoning. Twentieth-century logicians have stressed the essential incompleteness of the provable, an incompleteness demonstrated for purposes of strictly deductive logic by meta-mathematical investigations initiated in the nineteen-thirties. For Peirce, the incompleteness of logic in his broader sense is made manifest in constant extensions of the methods of scientific argument and concomitant renewal of the abductive initiative of the community of inquirers. “Each chief step in science has been a lesson in logic” (Fixation, W3, 243, 1877).7 In so far as a Peircean philosophy of truth will elucidate truth by reference to inquiry itself and inquiry by reference to a struggle against doubt (= the state of not knowing what to think about this or that)—the struggle, that is, which finds its final fruition in the dispassion of a pure science of unbounded aspiration—it is not to be expected that these efforts will have the effect of circumscribing truth itself or limiting truth to that which is discoverable by any particular research method or aggregation of research methods. The insatiability of the inquiring mentality, like the ordinary discomfort (from which the scientific outlook originates) of not knowing what to think about some particular question, is one part of the background for the very idea of belief or opinion. But so too is the calmness and satisfactoriness of knowing what to believe. Christopher Hookway> has been troubled that Peirce should at once have condemned psychologism in logic and intruded psychological facts into his account of inquiry. But in so far as we see these background facts as conditioning the emergence of fully fledged opinion or belief—and in so far as we see the concern for truth as latent already within the nature of opinion and belief themselves—I hope we can exempt Peirce from the charge that he allows mere contingencies of human psychology to corrupt his conceptions of logic and truth themselves. These are not mere

2. See T. L. Short ‘Peirce on the Aim of Inquiry’, Transactions of the Charles S. Peirce Society 35 (1) (2000), p. 2, and n3. Another citation given by Short:“The method of science 1s itself a scientific result” (CP 6.428, 1893). 3. Hookway Peirce (Routledge & Kegan Paul, London, 1985) 52 f.

134

PEIRCE

contingencies. They are the enabling facts for the very existence of reasonable (however tentative and fallible) belief.

II1 [t might be questioned how exactly and faithfully, dating from 1906, our first epigraph reflects Peirce’s intentions of 1877.* But for present purposes it matters far more—and matters especially for the Peircean conception of truth—how Peirce himself, in his full maturity, wanted to read or reread or rewrite ‘Fixation’ and what place he came to want it to occupy within the context of his mature position. If ‘Fixation’ read with the emphases Peirce suggests in 1906 can help us to see the conception of truth “gradually” and “under the action of experience” emerge from the abandonment of dogmatism and authoritarianism, and help us see that conception emerge thence by virtue of the workings of the principle that the agitation of a question ceases when satisfaction is attained with the proper settlement of belief, then the task for the philosopher of inquiry will be to speculate what it is about the notions of truth and belief that fits them to cohere and consist with one another in this way. For beliefs, truth must be the first dimension of assessment of their goodness and badness (of their eligibility, so to speak)—even as true opinion must be our preeminent aspiration if we ask

“What shall I believe about such and such or so and so?”

IV Our first epigraph recapitulates ‘Fixation’. It shapes my commentary. That commentary starts out from Peirce’s claim that belief or opinion is the state we seek to attain and doubt the state of irritation we seek to end.The complexity of this simple-seeming declaration may be brought out by an analogy. Suppose that someone has appealed to me for my help, I make an excuse, and then I feel ashamed of letting them down. Finding it hard to live with 4. The importance of this citation 1s pointed out by Skagestad The Road of Inquiry, Charles Sanders Peirce’s Pragmatic Realism, New York: Columbia University Press (1981), 141.

PEIRCE

135

this failure, [ try at first to forget all about the matter. (After all, I didn’t owe the person any help, rather the reverse perhaps. And there are all sorts of other people they could have appealed to.) Suppose however that, within my own mind, this doesn’t work and I have a simpler thought: oughtn’t I to go back to the person and see if there is anything left that I still can do? Suppose that, in pursuing that very thought, I forget my own disquiet and minister directly to its intentional object (the other person’s need). Suppose

that, like a cloud, the disquiet itself then disappears. Mutatis mutandis, compare now the disquiet/dissatisfaction of not being sure what to think about whether...Should such disquiet be remedied by attention to the state of mind

itself, vexatious

as it is, or to the intentional

object of the state? If Peirce had been asked this question, I think his reply ought to have been this: the first two of his methods of combating the discomfort of not knowing seek to work directly upon the state; the third method is transitional; the fourth works directly upon the question itself which disquiets me. As we review the four methods, let the reader verify this for himself or herself. Concerning the first method, of tenacity or dogmatism, Peirce says that the social impulse, which comprises the inner compulsion to pay anxious heed wherever others think differently from oneself, practically guarantees the total ineffectiveness of this method to implant or maintain conviction or to forestall the disquiet of not knowing what to believe. The second method is the method of authority, consisting of dogmatism supported by the repression of social impulses that unsettle prescribed opinion. Here Peirce’s prescription is this: Let [men’s] passions be enlisted, so that they may regard private and unusual opinions with hatred and horror. Then, let all men who reject the established belief be terrified into silence. (W 3, 250 ‘Fixation’, 1877)

[L]et it be known that you seriously hold a tabooed belief, and you may be perfectly sure of being treated with a cruelty less brutal but more refined than hunting you like a wolf. (W 3, 256 ‘Fixation’, 1877)

Even though the method of authority holds better promise for the end of doubt than the first,and its past triumphs are manifest, Peirce then declares— in passages which seem in the light of the collapse of communism in Eastern Europe at once prophetic and illustrative of the long-term futility of the method of authority:

136

PEIRCE

[N]o institution can undertake to regulate opinions upon every subject. Only the most important ones can be attended to, and on the rest men’s minds must be left to the action of natural causes. (W 3, 251 ‘Fixation’, 1877)

For that reason, Peirce holds, once some people are led by unregulated convictions to question that which is officially prescribed for general belief, more and more other people will come to suspect that their own adherence to this or that approved opinion may be owed to “the mere accident of [their] having been taught as [they] have”. Evidently then, a new method of settling opinions must be adopted, which shall not only produce an impulse to believe, but shall also decide what proposition it is which is to be believed. Let the action of natural preferences be unimpeded, then, and

under their influence let men, conversing together and regarding matters in different lights, gradually develop beliefs in harmony with natural causes. (W 3, 252 ‘Fixation’, 1877)

Peirce calls this third method, the a priori method, new. But all that “new”

needs to mean here is that in his enumeration of remedies for doxastic disquiet he has put it next after tenacity and submission to authority. Then he says of the a priori method “so long as no better method can be applied, it ought to be followed” because “it is the expression of instinct, which must be the ultimate cause of belief in all cases”. Under this aspect, the a priori method is only a resumption of other proto-rational methods of enquiry. Self-evidently, then, even when dignified as a method for “the fermentation of ideas” (CP 5.564, 1906), the method can only restore the state where we

were before we turned to these other expedients. It is not surprising then if (as Peirce claims) its failure has been the most manifest. It makes of inquiry something similar to the development of taste; but taste, unfortunately, is always more or less a mat-

ter of fashion...[And] I cannot help seeing that...sentiments in their development will be very greatly determined by accidental causes. Now, there are some people, among whom I must suppose that my reader is to be found, who, when they see that any belief of theirs is determined by any circumstance extraneous to the facts, will from that moment not merely admit in words that that belief is doubtful, but will experience a real doubt of it, so that it ceases to be a belief. (W 3, 253, ‘Fixation’, 1877, my italics)

The last sentence is one of the most important sentences in Peirce’s whole essay. It suggests inter alia that those who practise the first or second methods

PEIRCE

137

or even the third have misunderstood the nature of the disquiet or irritation of not knowing. Once they understand this better, they will return to the intentional object of their disquiet, namely the particular question they are unsure about. I would that Peirce had put a more general emphasis upon this sentence of his. For the last sentence, here italicized, is the point of tran-

sition to Peirce’s fourth method of countering our disquiet at not knowing or not knowing for sure: To satisfy our doubts, therefore, it is necessary that a method should be found

by which our beliefs may be caused by nothing human, but by some external permanency—by something upon which our thinking has no effect...[That external permanency| must be something which affects, or might affect, every man. And, though these affections are necessarily as various as are individual conditions, yet the method must be such that the ultimate conclusion of every man shall be the same. Such is the method of science. Its fundamental hypothesis, restated in more familiar language, is this: There are real things, whose characters are entirely independent of our opinions about them; those realities affect our senses according to regular laws, and, though our sensations are as different as are our relations to the objects, yet, by taking advantage of the laws of perception, we can ascertain by reasoning how things really and truly are, and any man, if he have sufficient experience and reason enough about it, will be led to the one true conclusion. The new conception here involved is that of reality. (W 3, 253—4 ‘Fixation’, 1877)

Here, in so far as we are influenced by Peirce’s gloss of 1906, we shall understand him to say something like this: in your inquiry hasten beyond the management of states of doubt or disquiet to the kinds of question that can be exposed to some “permanency”’ upon which our thinking has no effect. Let any opinion or belief you arrive at to the effect that p be determined by circumstances that are “not extraneous to the fact that p”. Meanwhile, reflecting on the aim that any ordinary inquirer will have if they feel the dissatisfaction of not knowing what to believe, the philosopher of inquiry must be tempted to think that this anxiety of the inquirer’s for their belief to be determined in just such a way is exactly what is needed for us to begin to understand the idea of “a reality” to which the belief that p is answerable. Let us distinguish deliberately here between the roles of inquirer and of philosopher of inquiry. Normally, when we engage as inquirers in some investigation, we

do

not think, in the abstract, about

methodology. The

Peircean philosopher of inquiry knows that. But if, even as inquirers themselves submit to experience, they do reflect abstractly about their procedures

138

PEIRCE

and the rationale of what they do, then, according to Peirce, the things inquirers themselves will find they have discovered are the ideas of truth, of fact, and of a reality or (as Peirce rewrote some passages of ‘Fixation’ to say) “a Real”. The corresponding task of the philosopher of inquiry is to make the however inexplicit working ideas of first-order inquirers more explicit. In this way let the philosophy of inquiry harvest a conception of the full potential of the meagre resources that the Peircean methodology of inquiry begins with. Amid this harvest, once it is properly examined, and along with the ideas of truth, fact, and reality (Real), the theory or philosophy of the practice of inquiry will find Peirce’s “fundamental hypothesis”. This speaks of our taking advantage of our perceptions and the “laws of perception”, in order to ascertain “by reasoning how things really and truly are”.”> At this point, moreover, the consideration of the fundamental hypothesis will force upon a philosopher of inquiry the task of arriving at a proper conception of experience. For experience is that by which we can expose our minds to realities/Reals and make our beliefs answerable to realities/Reals. The forcible element in our experience is what Peirce calls “secondness”.” “It may be asked”, Peirce notes, “how I know that there are any realities [Reals]”.To this question Peirce gives four replies, of which the most striking, interesting, and conclusive is this one: 6

¢

The feeling which gives rise to any method of fixing belief is a dissatisfaction at two repugnant propositions. But here already is a vague concession that there is some one thing to which a proposition should conform. Nobody, therefore, can really doubt that there are realities [Reals], or, if he did, doubt

would not be a source of dissatisfaction. The hypothesis, therefore, is one which every mind admits. So the social impulse does not cause me to doubt it. (W 3, 254 ‘Fixation’, 1877)

A

s. It 1s worth collating the indispensability of this hypothesis with one of the several roles of Leibnizian Sufhicient Reason. For all these roles, see Wiggins (1996e) ‘Sufficient Reason: a Principle in Diverse Guises both Ancient and Modern’, Acta Philosophica Fennica 61: 117-32. 6. Here there 1s rich collateral evidence of Peirce’s intentions. Especially, perhaps, we should take note of a manuscript of 1893—5 that Cheryl Misak draws to our attention: As for the experience

under

the influence

of which

beliefs are formed, what

1s 1t? It 1s

nothing but the forceful element in the course of life. Whatever it 1s. .. 1n our history that wears out our attempts to resist it, that 1s experience.... The maxim that we ought to be “guided” by experience means that we had better submit at once to that to which we must submit at last. “Guided” 1s not the

word; “governed”

1893—5, quoted in Misak (Truth and the End of Inquiry, p. 83)

should be said. MS

408, p. 147,

PEIRCE

139

There 1s more to say about realities/Reals (see section IX) and more to be said also about the nature and extent of Peirce’s commitment to causal realism (see section VI), but such in bare outline is Peirce’s doctrine. Before we

convert any of this into a distinctively Peircean contribution to the philosophy of truth, however, we must attend to Peirce’s conception of belief, to

the abductive coloration that he gives to the idea of experience, and to his idea of abduction itself. We shall attend to each of these things, in sectionsV,

VI, and VII, beginning with belief.

V At the outset, Peirce says that belief in a particular proposition is a calm and satisfactory state. It is a state “we do not wish to avoid, or to change to a

belief in anything else. On the contrary, we cling tenaciously not merely to believing but to believing just what we do believe”. (CP 5.372, 1902—3) In the same tenor he writes, With the doubt...the struggle begins, and with the cessation of doubt it ends. Hence, the sole object of inquiry is the settlement of opinion.We may fancy that this is not enough for us, and that we seek, not merely an opinion, but a true opinion. But put this fancy to the test, and it proves groundless; for as soon as a firm belief is reached we are entirely satisfied, whether the belief be true or false. And it is clear that nothing out of the sphere of our knowledge can be our object, for nothing which does not affect the mind can be the motive for a mental effort. The most that can be maintained is that we seek for a belief that we shall think to be true...and, indeed, it is a mere tautology to say so. (W 3, 248 ‘Fixation’, 1877)

From this conclusion about the thoughts that first-order inquirers can have Peirce derives attractive corollaries, corollaries that maintain the impossibility cum pointlessness, in the absence of real and living doubt, of any general project or plan of exposing everything to question. (Contrast Descartes.) Yes, but one ought not to permit the claim that the sole object of inquiry is the fixation of belief to escape all criticism just because it delivers conclusions that we have other reasons to find attractive. The claim needs more extended comment. There is a real mistake here. The directive, “seek a true belief”, Peirce says, has no more practical con-

tent than “seek a belief you think true”. And then he continues, “we think each one of our beliefs is true. It is a mere tautology to say so”. If doubt

140

PEIRCE

irritates us, Peirce seems to say, his advice to us is “seek something you can

really believe”, not “search for a true belief™. This does not look right. Similar claims used to be made to the effect that there is no practical difference between the directives “do your duty” and “do what you think is your duty”. But makers of such philosophical claims always paid too little attention to the fact that it is not without consequence which of these directives you second. One who fails to think hard what his duty is but does what he takes to be his duty obeys the directive to do what he thinks is his duty, but he does not necessarily do his duty. In Peirce’s advice, there is a closely parallel oversight. One wishes he had not said what he says. It will miss the point, though, to insist too much upon it. For it becomes clear slightly later in Peirce’s essay—in the transition from the third to the fourth method—that he is deeply impressed by a particular and special point about belief and the conditions that are constitutive of belief. That point is that the belief that p, once

challenged, is a state which

needs, on

pain of extinction, to be able to see itself as a state not “determined by circumstances extraneous to the facts [concerning whether or not p].” By its nature, belief as Peirce needs to conceive it is a touchy, uncomplacent condition of the mind, a disposition which will not and cannot stay around on just any old terms. This is one of the things that bring the normative science of logic into being. Reinforcing Peirce’s insistence that the sole object of inquiry is the settlement of opinion, let us suggest that he say instead: “Believe what you will—end the irritation of doubt however you like—but only provided that the belief with which you conquer doubt will stick, provided it really will conquer the particular doubt”.” The prescription suggests that, given the exigence that Peirce will find latent in the state of belief and given the object-directedness of the disquiet of not knowing, the injunction to get oneself a belief in order to end the irritation of doubt can never be satisfied by possessing oneself of just any opinion or just any substitute for the belief that gives dissatisfaction or disquiet. Once a question arises that one cares about, one can only be satisfied by an answer that one can take

7. An analogy. Augustine wrote “Dilige [deum| et quod vis fac”. The exhortation “Love God and do what thou wilt” may seem to be utterly permussive. It seems so until you reflect that such an injunction requires you to desire nothing God would not wish you to desire (or nothing you think he would not wish you to desire). It does not entail that you should do whatever you will. There 1s no doubt how Augustine’s double direction is to be understood. (I note that such double commands seem to discredit any logic of imperatives that authorizes ‘conjunction elimination’. Do they discredit then the very idea of imperative logic?)

PEIRCE

141

oneself to have acquired in a manner that is proper to the content of both the question and the answer.” What then ought Peirce to have said was the whole aim of inquiry? It would have been better, and much less open to misunderstanding, if he had said that the whole aim of inquiry was to end the irritation of not knowing (whether/who/when/what/ ...) by bringing into being the proper conditions for the settlement of opinion with respect to the matter that is in question. The aim must be to secure everything that it takes to obtain this settlement. Once we say that, however, there is scarcely any temptation at all to say that there is no difference between seeking an opinion and seeking a true opinion. The notion of truth lurks already within the notion of the “proper condition for the settlement of opinion”.

V1 So

much

for belief. Now

for Peirce’s

fourth

method,

the fundamental

hypothesis, “realities (Reals)”, and the “external permanency upon which our thinking has no eftfect” and which will under the right conditions prompt beliefs to us. Where Peirce speaks of “realities/Reals affect[ing] our senses according to regular laws” or speaks of anyone with sufficient experience and willingness to reason “taking advantage of the laws of perception” in order to “ascertain by reasoning how things really and truly are”, the cases that come to mind first as illustrations of this strange-sounding doctrine are singular empirical judgments relating to the past or present as treated by the causal theories of memory and perception. It would be a pity if these were the only cases that were tractable by Peirce’s theory of inquiry. But let us start in the area that is plainest for the doctrine and inquire what kind of reasoning it is that leads there to such “ascertaining”. In the most straightforward perceptual case, how does Peirce envisage its workings?

8. In something he wrote before ‘The Fixation of Belief’, Peirce had already noted that there 1s an important difference between the settlement of opinion which results from investigation and every other such settlement. Investigation “will not fix one answer to a question as well as another, but on the contrary it tends to unsettle opinions at first, to change them and to con-

firm a certain opinion which depends only on the nature of investigation 1itself” (CP 7.317, 1873). By the time someone for this.

has reached for the fourth method, they will be fully prepared

142

PEIRCE

The answer to the question is that the reasoning Peirce has in mind is abductive or retroductive, though in a special way. Even for the normal case, of perception or memory, Peirce offers no systematic account of the relation between perception and abduction or the relation between memory and abduction. We do, however, get some indications of the link he saw between

hypothesis or abduction and states such as memory or perception: I once landed at a seaport in a Turkish province;and as I was walking up to the house which I was to visit, [ met a man upon horseback, surrounded by four horsemen holding a canopy over his head.As the governor of the province was the only personage I could think of who would be so greatly honoured, I inferred that this was he. This was an hypothesis. Fossils are found; say, remains like those of fishes, but far in the interior of

the country. To explain the phenomenon, we suppose the sea once washed over this land. This is another hypothesis. Numberless documents and monuments refer to a conqueror called Napoleon Bonaparte. Though we have not seen the man, yet we cannot explain what we have seen, namely, all these documents and monuments, without supposing that he really existed. Hypothesis again. As a general rule, hypothesis is a weak kind of argument. It often inclines our judgment so slightly toward its conclusion that we cannot say that we believe the latter to be true; we only surmise that it may be so. But there is no difference except one of degree between such an inference and that by which we are led to believe that we remember the occurrences of yesterday from our feeling as if we did so. (W 3, 326—7, ‘Deduction, Induction, and Hypothesis’, 1878, my italics.)

This last is the case of memory. For the case of perception, we have the following: ...abductive inference shades into perceptual judgment without any sharp line of demarcation between them; or, in other words, our first premisses, the per-

ceptual judgments, are to be regarded as an extreme case of abductive inferences, from which they differ in being absolutely beyond criticism. The abductive suggestion comes to us like a flash. It is an act of insight, though of extremely fallible insight. (CP §.181, 1903)

This is a strange idea. What could Peirce have had in mind when he claimed that, like memory, perception too was abductive? For the case of perception our answer needs to cohere with two other Peircean doctrines (the first not,

in the light of our earlier mention of “secondness”, unexpected): ...this direct consciousness of hitting and getting hit enters into all cognition and serves to make it mean something real (CP 8.41, c.1885)

PEIRCE

143

and, second, The chair I appear to see makes no professions of any kind, essentially embodies no intentions of any kind, does not stand for anything. It obtrudes itself upon my gaze; but not as a deputy for anything else nor “as” anything. (CP 7.619, 1903)

A full reconstruction of Peirce’s doctrine would have to cohere with these clues and with his fallibilism (which suggests that what is “beyond criticism” must be not a perceptual belief but the perceptual state itself, which may or may not sustain a belief). It would need to cohere also with Peirce’s numerous but sketchy hints concerning the distinct roles in perception of percept, percipuum, and perceptual judgment. In lieu of such a reconstruction, I offer an interim statement. It is not given in Peircean language, and the result will seem strange. Suppose object and perceiver encounter one another in perception.Then

independently of will or reason, the perceiver may be moved to report what he sees by uttering the words “Six windows obtrude, it seems, upon my gaze”. No abduction yet. But for the perceiver to take what he is confronted with for six windows just is—whether he knows it or not—for him to take it that the best explanation of his perceptual state is that there are indeed six windows there. Mutatis mutandis it must have been roughly similar for the remembering case. From one’s remembering—or its being as if one remembers—the messenger giving one a letter yesterday, one can convince oneself that the messenger did indeed give one a letter yesterday. People do sometimes need to convince themselves like that. But, for the ordinary case, let us prefer to say (more weakly and still somewhat strangely) that in so far as one takes oneself to remember or perceive, one is committed (whether one knows it or not) to an abductive argument. Of course, this is a third-person remark about the legitimacy of what the inquirer does. For the ordinary case it is not a reconstruction of his own thoughts or reasoning processes. Can we generalize this? Well, it seems, according to this doctrine, that the

relation of experience and belief must be this: that the experience creates, by its nature as experience, a fallible presumption that what we are moved to report that we see or remember is that which accounts for our being so moved to report. But rather than attribute thoughts of this kind to ordinary percipients or intellectualize that which needs not to be intellectualized, one might prefer to say that the acceptability of abduction is quietly and tacitly institutionalized in our exercise of our faculties—in our practice and in the title that perceivers could claim that the use of senses or memory affords for

144

PEIRCE

them to make empirical claims. Echoing a formulation that appealed at one time to A. J. Ayer, the philosopher of inquiry can say that it is the outcome of the proper exercise of these faculties which, with respect to certain indispensable judgments, gives ordinary inquirers in ordinary circumstances the right to be sure. However strangely, the Peircean contribution to that thought must then be this: such normative claims are undergirded by non-normative laws whose dependability will be accounted for only by appeal to abduction.

VII Here ends the interpretation and explication of “The Fixation of Belief” (at least with respect to judgments conforming to the easiest empirical paradigm). Indeed, in one way, we are well beyond the end of the paper itself. But we are not yet at the end of expounding the fourth method, the method of experience. This is still in the condition of a programme needing to be worked out. The fourth method depends on abduction, not only in the (limiting) perceptual case but for almost everything else that we can then build upon perception. In order to enlarge upon the method, we now have to set out certain details that Peirce gives in other writings that he devoted to logic and his theory of inference. Peirce classifies inferences as deductive/analytic/explicative and as synthetic/ampliative. And the synthetic/ampliative he subdivides into (1) abduction, hypothesis, or retroduction (these terms are close to synonymous in Peirce) and (2) induction, which is very different indeed.

Let us begin with induction: Induction is where we generalize from a number of cases of which something is true, and infer that the same thing is true of a whole class. Or, where we find a certain thing to be true of a certain proportion of cases and infer that it is true

of the

same

proportion

of the

whole

class.

(W

3, 326, ‘Deduction,

Induction and Hypothesis’, 1878)

Hypothesis, on the other hand, is where we find some very curious circumstances, which would be explained by the supposition that it was a case of a certain general rule, and thereupon adopt that supposition. Or, where we find that in certain respects two objects have a strong resemblance, and infer that they resemble one another strongly in other respects. (W 3, 326, 1878)

PEIRCE

145

Or as Peirce describes abductive thought elsewhere: The first stating of a hypothesis and the entertaining of it, whether as a simple interrogation or with any degree of confidence, is an inferential step which I propose to call abduction. This will include a preference for any one hypothesis over others which would equally explain the facts, as long as this preference is not based upon any previous knowledge bearing upon the truth of the hypotheses, nor on any testing of any of the hypotheses, after having admitted them on probation. I call all such inference by the peculiar name, abduction ... (CP 6.525 ‘Hume on Miracles’, 1901)

Here the restrictions we see Peirce start to draft may need very careful statement. For we shall also need to prevent this form of inference from allowing into the place of a hypothesis—into the place marked by “A” in our next citation—suppositions that are contrary to things in the reasoner’s

evidential background or that are gratuitous relative to that background. The thought that some such preclusion is needed becomes even more evident when abduction is set out as starkly as it is here: The hypothesis cannot be admitted, even as an hypothesis, unless it be supposed that it would account for the facts or some of them. The form of inference, therefore, is this: 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.

Thus, A cannot be abductively...conjectured until its entire content is already present in the premise,‘If A were true, C would be a matter of course’.

(CP 5.189, 1903) When the form of this reasoning is set out in this way, the question that takes shape is whether (subject to the restrictions Peirce gives in 6.525, cited) just any supposition, any supposition at all which would make “C” a matter of course, should be permitted to count as a hypothesis, and as something ready to move up to the next stage of being subjected to confirmation/disconfirmation.” Ought there not to be criteria for the interrogation and selection of things that shall count as hypotheses?’® And where do they spring from? Do they entirely spring from the need to stabilize belief on 9. Peirce sometimes talks like this:“Abduction commits us to nothing. It merely causes a hypothesis to be set down upon our docket of cases to be tried” (CP 5.602, 1903). Elsewhere, he 1s troubled by the fact that “it 1s well within bounds to reckon that there are a billion hypotheses that a fantastic being might guess would account for a given phenomenon”. See Peirce (1929, pp- 269—83). For just one response to this difhiculty, see the sentences from ‘Guessing’ cited at the end of the paragraph to which the present note attaches. 10. See Misak, Truth and the End of Inquiry, p. 99.

146

PEIRCE

belief’s own terms, etc.? How much can it help to reflect with Peirce that, in hypothesizing, “man divines something of the secret principles of the universe, because his mind has developed as a part of the universe and under the influence of these same secret principles”? I shall not attempt these questions.

According to Peirce’s doctrine, retroduction or abduction (however we enlarge upon it) is a distinctive mode of thinking. It is reducible neither to deduction, whose role is the ancillary one of drawing out the consequences of hypotheses, nor yet to induction, to which Peirce assigns the special role of testing (refuting or supporting) the hypotheses that are submitted to it by abduction. Induction itself, as Peirce sees it (and note that Peirce does not

deny that there is any such thing as reasonable induction), can help to support generalizations. But, pace Nicod, it can never license us in or of itself to go from positive instances of an arbitrary putative generalization towards the assertion of that generalization. Before that can happen, the generalization has to enjoy the status of a hypothesis. It can only attain that status if, in the right way, it renders less surprising something else that has seemed surprising or wanted explaining. From this it follows that no methodological paradoxes such as Hempel’s (of the ravens, etc.)'' or Goodman’s

(of “grue”, etc.)

can gain any purchase on the Peircean account of inquiry. For there is nothing in the Peircean account that corresponds to Nicod’s postulate or depends upon it. If a white shoe really did confirm to some degree that “all nonblack things are nonravens”—this would be the effect of Nicod’s postulate—then it would have to confirm to the same degree its contrapositive equivalent “all ravens are black”. And that, in any normal setup, is evidently absurd. In Peirce’s conception of inquiry, apportioning work in the way it does between induction and abduction, there is simply no place for Nicod’s postulate.

VIII How does a putative subject matter need to be if Peirce is to allow that it constitutes a proper field of genuine inquiry? A similar (or equivalent?) question: to what standard must a putative subject matter attain, and what 11. On Nicod’s postulate, see Carl pp- 1-26.

Hempel, ‘Studies in Logic and Confirmation’, Mind

s4 (1945),

PEIRCE

147

must be its condition, for the judgments it throws up to count as properly answerable to Peircean realities/Reals? Suppose there is a mode of thinking, neither purely perceptual nor relating only to what is remembered, that is well enough manageable for the following to hold: if you engage in the form of thinking in question, then “secondness jabs you perpetually in the ribs” (CP 6.95, 1903). Suppose that,

practising this mode of thinking, you can reach by patient labour a complex and many-layered state of readiness and then arrive, when jabbed, at a belief. This is to say that, at some crucial point in your thoughts or explorations, something that is not up to you but is of the right sort to do this can bring it about that you are convinced, fallibly but fully. Suppose that in this field you can arrive at a belief (as Leibniz would say) malgré vous. Then, whatever the distance at which this form of thinking lies from the direct perceptual case or the memory case, your search cannot help but count as a genuine form of inquiry—a form within which the judgment that you arrive at can be answerable for its correctness to some reality/Real. Or so it seems. If Peirce’s accounts of the fourth method and of secondness have any generality at all, then the only doubt there can be concerning whether there is any such reality/Real is a doubt relating to the credentials themselves of the form of thinking that purports to invoke the Real in question. Such is the distance that it appears one can put between Peirce’s theory of inquiry and any plainer, more simply causal theory of cognition/inquiry. Consider here Peirce’s philosophy of mathematics, where Peirce describes the sort of secondness that can arise from experimenting by pencil and paper with a representative diagram, running through all possible cases and finding (say) that some apparent plurality of alternatives reduces to one case. See, for instance, CP 4.530 (1905), 3.516 (1896). Now

here, it may be

said, there is an objection. Consider Peirce’s own phrase in ‘Fixation’, “determined by circumstances not extraneous to the facts”.What can these words

mean, it will be asked, unless Reals are items with a distinctively

causal role? If this objection is right, then either we must abandon every kind of thinking that trespasses outside the paradigm illustrated by the causal theories of memory and perception (as arithmetical thinking surely does) or else we must try to unpack the phrase “determined by circumstances not extraneous to the facts”. The second response seems more promising. Nor are we the first to think this. In the course of one of his own rereadings of ‘Fixation’, Peirce made an annotation against the words (already quoted in section IV)

148

PEIRCE

“To satisfy our doubts, it is necessary that a method be found by which our beliefs may be caused by nothing human, but. . by something upon which our thinking has no effect”. Peirce’s annotation requires the word “caused” to be replaced by the word “determined”.” It suggests that he wanted to construe “[beliefs or opinions] determined by circumstances not extraneous to the facts” in a way that allowed but did not require such determination to be simple causal determination. In order to set out some of the options that this creates for the different kinds of case that Peirce needs to accommodate here (they are far too many for comfort, but let us see whether anything at all can be said at this level of generality), we must begin with a concession to causality. Opinions arise from thoughts and thoughts are produced by earlier thoughts.“If we mount the stream of thought instead of descending it, we see each thought caused by a previous thought” (W3, 34, 1872).Taking our cue from this dictum and

tracing the sequence from later to earlier, let us conviction that a thinker reaches at the end will some secondness experience (as one might say). earlier point, the secondness experience itself proper ancestry in some reality that it presents.

accept that the be the product Let us allow too must be traced These ancestries

opinion or or effect of that, at the back to its will come

in different varieties, however.

In an ordinary causal case, there is a causal-cum-perceptual transaction between (say) the Cathedral at Chartres and a conscious, properly recipient subject S; and then, on the strength of this event, S believes justifiably and correctly that the Cathedral at Chartres has two spires. Here it is by virtue of the causal perceptual transaction that the Real consisting in the cathedral’s having two spires determines S’s belief that the cathedral has two spires. (In a fuller treatment one would attend separately of course to the cases of seeing x and of seeing that x is ¢.) That is the familiar case. But now, in a less familiar case, suppose that the initiator of belief was not perception but some “elaborative process of thought” (W 3, 42), one leading into a gradual accumulation of reasons that culminated at the moment of secondness in the thinker’s finding nothing else to think but that (...). Here the thing which brought the thinker to the point of conviction was not just any old causal effectiveness. Still less need the thinker’s finding that there was nothing else to think but that (...) be the

12.

See Short, Peirce on the Aim

of Inquiry, n. 9.

PEIRCE

149

outcome of some reality’s (or Real’s) causally effective agency."” Rather the reason why the thinker was unable to find anything else to think but that (...) was that there is here nothing else to think. If there is nothing else to think, no wonder the thinker thought that! You can say, if you wish, that some reasonable being’s finding himself unable to discover anything else to think causally explains his finally arriving at the opinion that (...). But at the temporally first link in the chain, the reasonableness of the thinker and the reasoned character of his thought are essential to the explanation. It is in this essentially normative way that we satisty Peirce’s requirement that the inquirer’s opinion that (...) should be determined by a circumstance not extraneous to the facts. It is satisfied because the circumstance of there being nothing else to think but that (...) is not something extraneous to the facts. Rather, one may want to say that this circumstance bears a (so to speak) constitutive relation to the reality/Real that consists in the fact that (...)."

IX [ hope that the proposal just offered remains within the spirit of Peirce’s annotation and correction. It shows how the purely causal case need only

be one among many others. Elsewhere,'® I have tried to illustrate the formal pattern given in the previous section. I shall give again here two examples, doubtful though it is how content Peirce would have been with the second. (A) Peter believes that 7 + § = 12. He has learned this neither by rote nor

yet by reading that famous passage of Kant’s Critique of Pure Reason where 7 + § = 12 serves as an example. Why then does he believe it? Well, the explanation begins with the fact that all the other answers to the question “what is the sum of 7 and §?” are blocked or excluded. In a full version of

the explanation, this exclusion could be proved by reference to the calculating 13. Here one 1s eager to allow on Peirce’s behalf for the full force of a remark that he made 1n 1902—I owe the reference to Skagestad, The Road of Inquiry, p. 30—"“In reasoning, we have the singular phenomenon of a physiological function which is open to approval and disapproval” (CP 2.152). 14. Compare “[T]he truth of the pure mathematical proposition is constituted by the impossibility of ever finding a case in which 1t fails” (CP 5.567, 1901, 1talics added).We shall supersede 1n the next section the apparently relational mode of discourse adopted in the sentence to which the footnote is annexed. 15. See for instance Chapter 12 of my Ethics: Tivelve Lectures on the Philosophy of Morality (ETL)

150

PEIRCE

rules. In a maximal version, one would also rehearse the irresoluble difficul-

ties attaching to proposals for different rules. Once so much was set down, the explanation might continue as follows. Peter knows those calculating rules. Moreover, in espousing the answer 12, Peter is going by the rules. So no wonder it is his opinion that 7 + § = 12. So, in this case, Peter’s belief that 7 + § = 12 is determined (as Peirce requires) by a circumstance not extraneous to the fact that 7 + § = 12. His reasoning summarily recapitulates the

very reason (expandable at will) why seven plus five is twelve. Indeed the full explanation of Peter’s belief precisely vindicates Peter’s belief. (B) Paul believes, let us suppose, that slavery is unjust and insupportable. Suppose that, in seeking to explain why Paul believes this, we inquire into his reasons for thinking this, and suppose we then look for further amplifications and elucidations of those reasons, drawing on the whole ethical background that we share with Paul. This will take a long time, but suppose that, as we proceed, it appears more and more clearly that the only way to think anything at variance with the insupportability and injustice of slavery is to opt out altogether from any moral viewpoint that can make sense of asking the very question “What is one to think of the supportability or justice of slavery?” For suppose that at some point, in heaping consideration on to consideration, we find we have enough and it becomes apparent that there is simply no room in which to form another opinion. No doubt there will be many ethical cases where we do not reach this point and we do not know how to close off every avenue. But, in the case where we really can see Paul’s belief as downwind of reasons such as the convincing ones that we have imagined someone’s eventually rehearsing about slavery, we can surely say “No wonder Paul believes what he believes! On this matter there is nothing else to think”. In other words, Paul’s belief about slavery is determined by circumstances (namely the considerations that we are supposing could have been rehearsed and impinged on one who understood the moral question) not extraneous to the fact that slavery is unjust and insupportable. For Paul’s reasons for thinking what he thinks do summarily recapitulate that in virtue of which slavery is wrong and insupportable. Peirce might have been sceptical, I fear, whether our example (B) could

be worked out in the way I have imagined. He might wonder whether the larger ethical background can (as I suppose) shape the very sense of the terms “‘unjust” and “insupportable”. But he could not object in principle to the idea that a mass of considerations can culminate in conviction. It is noteworthy how he speaks in other connections of reasons “not form[ing]

PEIRCE

IS1

a chain which is no stronger than its weakest link but a cable whose fibers may be ever so slender provided they are sufhiciently numerous and intimately connected” (W 2, 213, 1868).

X It is evident—and a search for more examples would make it even more evident—that the generality to which Peirce aspires in his theory of inquiry involves us in a bewildering and indefinite variety of different ways in which thinkers in different areas of concern can satisfy the Peircean requirement on which we have laid such stress. It is no less evident, though, that in

so far as we want to persist at that level of generality, the answer we give to the question proposed at the beginning of section VIII will have to be as follows: the thing that is minimally required in order to secure pragmatic content to a subject matter is this: that there, in that subject matter, a belief

to the effect that p can be determined by circumstances not extraneous to the fact that p. More generally, the conclusion to which we are drawn is that, for any genuine belief, whether true or false, there has to be something it is answerable to and sensitive to. This last may as well be called a Real. But instead of rushing into a new ontology of Reals, let us look carefully to the status of our familiar form: whoever sincerely inquires whether p seeks to ensure that any belief of theirs to the effect that p be determined by circumstances not extraneous to the fact that p. This 1s only a schema. Reals are not here objects, still less objects quantified over. The sentence letter “p”, being not a variable, functions by holding a place for a sentence in use. On these terms, the minimal claim about the formation of the belief that p is a notionally simultaneous assertion of all instances of the italicized sentence form with all possible sentential fillings for the letter “p”. But in putting forward this schema, we gesture at something entirely general. The secondness requirement, the non-extraneousness condition, and the other requirements on the determination of the inquirer’s belief are to be understood in the divers ways that are appropriate to different examples. If what Peirce says about Reals is interpreted or elucidated in this way, then the philosophical effect is that the schema is grammatically and philosophically filled out according to the subject matter—all in the light of whatever Peirce’s logic can add to his characterization of the fourth method.

152

PEIRCE

X1 If the fourth method as now explained and enlarged upon is the only method of satisfactorily settling opinion (albeit fallibly, always fallibly), then what conception of truth do theorists of inquiry have to see as animating and constraining the epistemic efforts of those who practise the method? And how are theorists further to elaborate or elucidate this conception? Let us begin with some of the materials of ‘Fixation’ itself. In a footnote to a passage that I quoted in sectionV, and continuing that passage into an

afterthought dated 1903, Peirce writes: 1. CP

5.375: [T]ruth

is neither more

nor less than that character of a propos-

ition which consists in this, that belief in the proposition would, with sufficient experience and reflection, lead us to such conduct as would tend to satisfy the desires we should then have.To say that truth means more than this is to say that it has no meaning at all.

This 1s the kind of statement that has given pragmatism such a bad name. Peirce speaks here of what truth is. But if we take the characterization in passage (1) as intended to bring out what is so good about truth, then not only do we find ourselves attributing to him an instrumentalism that is alien to his actions, his character, and his expressed views of science and life itself.

We interpolate into his theories something that is alien to the later sections of ‘Fixation’. It is true that, in other places, Peirce gives a pragmatic reinterpretation of the notions of “reality”/“Real”/“external permanency” that play such an important role in ‘Fixation’. But (as I read it) that reinterpretation is precisely not intended to blunt the force of what is said in the later sections of ‘Fixation’. The intention (whether successtul or unsuccessful) is

rather to explicate these notions—in the spirit of “look[ing] to the upshot of our concepts in order rightly to apprehend them™ (CP 5.3, 1901). We need perhaps to consider some more Peircean testimony, testimony

beyond that already displayed in section IV, about the idea of truth:' 2. CP 2.135, 1902: You certainly opine that there is such a thing as Truth. Otherwise reasoning and thought would be without a purpose. What do you mean by there being such a thing as Truth? You mean that something is 16. Useful collations of sources on truth will be found in S. Haack *“We Pragmatists...”; Peirce and Rorty in Conversation’, Partisan Review 64 (1997) 91—107 (which sets out some wicked, curious, and 1nstructive contrasts between her two subjects, Peirce and Rorty) and in Migotti ‘Peirce’s double aspect theory of truth’ in C. Misak (ed.) Pragmatism: Canadian Journal of Philosoply Suppl.Vol. 24 (1999): 75—108.

PEIRCE

153

SO...whether you, or I, or anybody thinks it is so or not. ... The essence of the opinion is that there is something that is SO, no matter if there be an overwhelming vote against it. 3. CP 5.56s, 1901 (‘Truth and Falsity and Error’): Truth

is that concordance of an

abstract statement with the ideal limit towards which endless investigation would tend to bring scientific belief. . .. Reality is that mode of being by virtue of which the real thing is as it is, irrespectively of what any mind or any definite collection of minds may represent it to be.The truth of the proposition that Caesar crossed the Rubicon consists in the fact that the further we push our archaeological and other studies, the more strongly will that conclusion force itself on our minds forever—or would do so, if study were to go on forever. An idealist metaphysician may hold that therein also lies the whole reality behind the proposition; for though men may for a time persuade themselves that Caesar did not cross the Rubicon, and may contrive to render this belief universal for any number of generations, yet ultimately research—if it be persisted in—must bring back the contrary belief. But in holding that doctrine, the idealist necessarily draws the distinction between truth and reality. 4. CP 5.416, 1905: [A truth is] that to which belief would tend if it were to tend

indefinitely to absolute fixity ... Let us begin

with

(2). (2) effectively reinforces

the manifest

purport

of

Peirce’s rationale for the fourth method. It reinforces the concluding message of ‘Fixation’ but it does not carry us beyond. The proposals (3) and (4) carry us back to formulations we took from ‘How to Make Our Ideas Clear’ and found wanting in section I. I note however that these proposals were intended by Peirce to give the effective or pragmatic meaning of the “merely nominal” conceptions expressed in proposals (2) and (3). In (3), such pragmatic proposals, even the idealist ones, are defended from the charge of losing the distinction between truth and reality. When read literally, however, these proposals all seem to depend upon the supposition that no information of the kind that would be needed to test plausible guesses already made or to discover truths as yet unknown (e.g., concerning that which is past or is presently hidden) ever perishes or becomes lost or unavailable to inquiry. For if, always and constantly, such information is being lost, then it is neither here nor there that inquiry can be constantly renewed, constantly corrected, and open-endedly prolonged. And the trouble is that such perishing, as Hilary Putnam points out, is not

only a fact but a fact that is implied by modern physics."’ 17. See Putnam ‘Pragmatism’, Proceedings of the Aristotelian Society 95 (1995): 272—306.

154

PEIRCE

Peirceans may respond to this objection by reading proposals (3) or (4) less literally. But then the construal will need to lean heavily on our understanding of that which (3) and (4) purport to define/explain/elucidate. It is

also worth remarking that, once the reference of “that to which inquiry would tend,” etc., is sufficiently carefully distinguished from any particular set of propositions that has been redacted or will have been redacted at any particular point in the future, the phrase “that to which inquiry would tend”, so far from distilling an effective or pragmatic meaning from the truisms that figure in (2), is a form of words that stands in radical need (as

radical a need as any expression ever could) of pragmatic elucidation.

XII How much is lost then so far as truth is concerned? Can it be that truth waits in the wings, is latent in the inquirer’s project of deciding what to think (see again our first epigraph), can be clearly seen emerging in the thoughts of someone who moves through the first, second, and third methods into the method of experience, abduction, and the rest—and yet is a character that defies all identification or elucidation? Are things as bad as that? Surely we can find for Peirce some form of words that fastens down and promises in due course to help elucidate, in terms that essentially involve the business of inquiry and the method of “experience”, the nature of that property, namely truth, which (unless we are complete strangers to opinion or doubt) is already familiar to any or all of us. Once we allow ourselves to speak of a property that is already known to us, and once we dissociate ourselves as we have from attempts to arrive at the property from simple pragmatist would-be determinations of the extension of “true’, several suitable forms of words stare us in the face:

s. Truth is the property that it is the aim of inquiry as such to find beliefs possessed of. 6. Truth is the character which, if only we follow the fourth method of inquiry, we may justifiably hope will be enjoyed by beliefs that survive however long or far inquiry is pursued or prolonged. 7. Truth is the property that anyone will want for his or her beliefs who sincerely inquires whether p (or not) and who seeks to ensure that any belief of his or hers to the effect that p (or not p) should be determined by circumstances not extraneous to the fact that p.

PEIRCE

1SS

Such formulations might or might not have appealed instantly to Peirce, but they hold a place for a view to which he could lay claim if he wanted. In so far as they arise from Peirce, however, is there something both purely and specifically pragmatic about them? Surely not. They are available to anyone regardless of their sympathies with Peirce.

XIII Suppose that, in the cause of further elucidating'® the property of truth, we were to deploy the identities given in (5), (6), and (7) and we were to elab-

orate upon the plurality of linkages holding between truth, on the one side, and inquiry, experience, secondness, hypothesis,..., on the other. Suppose that, proceeding in this way, we were to present our findings as the marks, in Frege’s sense,'” of the concept true, and suppose that, in the same effort,

we tried to explore the logical properties of the concept of truth (ascertain what properties the property of truth implies, excludes, etc. in a thing thought or said). Then what would follow from the fact that the whole basis on which this elucidatory exercise was conducted was a link between a notion of truth awaiting further specification and the notion of inquiry that is already developed (cp. IV,V,VI) and is partially definitive of pragmaticism as a philosophical position? Would the result be a specifically pragmatic conception of truth? I suspect not. Suppose that, proceeding in the way indicated and adducing our understanding of inquiry, we look in this spirit of pragmaticism “to the upshot of our conception [of truth] in order rightly to apprehend [it]”. Will that have the effect of subverting the ordinary (“realist”) presumption that the truth is perfectly independent of us (except, of course, in so far as some judgment that is in question relates to doings of 18. In order to elucidate a predicate (without necessarily defining it or giving necessary and sufhcient conditions for its application) one deploys the predicate and puts to use the concept that it introduces in ways that exhibit the character of the concept and reveal its connection with other concepts that are established, coeval, or collateral with it, and already intelligible in their own right. (For the pedigree of the term elucidation, see Wittgenstein 1921: 3.263, 4.020, 4.112.) 19. For various attempts of my own, independently of pragmatism, to pursue this line of inquiry, see my 2002a (Chapter 10 in this volume). Marks. The marks of the (first-level) concept horse are the (first-level) properties possessed by all things that fall under the first-level concept horse. Thus we arrive at the marks of the concept horse by asking, of things

that share the property

of being a horse, what

properties

they

have. The answer in this case will be the properties of having a head, four legs, a solid hoof, a flowing mane and tail, a voice that is a neigh....Similarly then, what properties do things (propositions, sentences, or whatever) have that possess the property of truth?

150

PEIRCE

our very own, or relates to the effects of such doings)? And will the pragmaticist outlook have the effect of undermining the ordinary idea— compare citation (2)—that the truth is “there anyway”, definitely and determinately?®® I think not. Might then Peirce’s emphasis upon the idea of inquiry have the effect instead of moving us towards an anti-realist position such as is developed by Michael Dummett under the influence of mathematical intuitionism and the emphasis that intuitionism places on the relation between understanding a proposition

and grasping a proof of that proposition? Again

I am

doubtful. The anti-realist whose position Dummett develops is one who affirms the laws of non-contradiction (no statement is true and false) and of

tertium non datur (no statement is neither true nor false) while withholding assent from the principle of bivalence (every statement is either true or false). Such assent is withheld by virtue of the absence of any assurance that, with regard to every well-formed assertion, either it or its negation can be established to be true. Things seem very different with Peirce. In his logical explorations, his doubts and hesitations relate to the law of double-negation elimination. Since double-negation elimination elides the subtle difference between tertium non datur and bivalence, committing anyone who accepts the former to the latter, it seems that Peirce’s does not anticipate any antirealism such as Dummett’s. It is true that, in a more philosophical context, Peirce writes (using the name ‘excluded middle’ where modern antirealists might prefer to speak of bivalence) that “Logic requires us, with reference to EACH question we have in hand, to hope some definitive answer to it may be true. That HOPE

with reference to EACH

case as it comes up is, by a SALTUS,

stated by logicians as a law concerning ALL CASES, namely the law of excluded middle. This law amounts to saying that the inverse has a perfect

reality”.>' The appeal to hope is significant, but the hope Peirce speaks of here relates to truth, not to proof or verification; and the substance of the hope surely relates to truth as ordinarily conceived. The confidence that Peirce speaks of as presupposed by the logical principle in question attests better to Peirce’s confidence in the significance of declarative sentences that are properly answerable to experience or experiment than it attests to any faith of Peirce’s in all such declarative sentences or their negations having 20. On these matters see C. Misak, Verificationism: its History and Prospects, London: Routledge (1995), pp- 121, 125, 127. 21 NE 4.x11, undated, emphasis as cited in Misak, Truth and the End of Inquiry, p.157.

PEIRCE

157

proof or verification. So, as I say, Peirce’s concerns seem to be different from

those of an anti-realist. In Peirce, the key to a proposition’s having significance or sense (and to our grasping that significance or sense) rests not with the prospect of proof/disproof or verification/falsification but in his apparent confidence in thinkers’ concerted and practical engagement with the business of inquiry—his faith in their competence to reach for verification or falsification. If (as I venture to think) pragmaticism leaves truth just as it was, what then is the real purport of pragmaticism in its connection with truth and meaning? What is the intended import of such dicta as these?— There is no conception so lofty and elevated that it cannot be fully defined in terms of the conceptions of our homely, instinctive everyday life. (MS 313, p. 29, quoted in Misak, Truth and the End of Inquiry, p. 119) or the familiar foundational claim

Consider what effects that might conceivably have practical bearings we conceive the object of our conception to have. Then our conception of these

effects is the whole of our conception of the object.”* (CP s.402, ‘How To Make Our Ideas Clear’, 1877)

According to the transposition of Peirce’s thoughts that I offer in these reflections, the real purport is relatively simple. Even though Peirce is a realist about truth, he is an operationalist about meaning. There is no specifically pragmaticist conception of truth,* but there is a pragmaticist conception of sense/significance/Sinn. A Peircean pragmatist, a pragmaticist (as Peirce

was led to say when he wanted to insist upon the differences between William James and himself), will scarcely think it worth saying that there is more to reality than could ever be put into however many propositions. Nor is it his special role to point out that, truth being what it is, there are all sorts of truths we shall never formulate and never could. The pragmatist’s key concern is with the propositions we shall ourselves arrive at, express, affirm, or believe. It is with the real purport or meaning of our actual 22. Cp. “We must look to the upshot of our concepts in order rightly to apprehend them” (CP 5.3, already cited). 23. As we saw 1n the first epigraph, Peirce thinks that the concept of truth 1s identifiable by reference to the concept of inquiry. In section XII, I have exploited that very thought. But truth 1s ot for that reason an epistemological or inquiry-based notion. It 1s a misunderstanding of the nature of elucidation to suppose that the concept of truth had to be epistemological just because one elucidatory route to truth was through the concept of inquiry. Even if truth is identifiable by reference to the idea of inquiry, it 1s not definable thereby.

158

PEIRCE

utterances, and the illusions we so easily fall into about what we can mean by what we say. The pragmaticist’s chief contribution to these questions, and the source of his critique of “vagabond ideas that tramp the public roads without any human habitation” (CP 8.112, 1900), lies where the citation last given indicates that it does. It lies in Peirce’s account of the grades of clarity that can be attained in our understanding of the terms that enter into meaningful sentences and in our grasp of the concepts that enter into the propositions that such sentences express. For Peirce and where conceptterms are concerned, we have the first grade of clarity if we can apply the term to things in our experience.We have the second grade when we can produce the kinds of explanation that pass muster as dictionary definitions or the like. At the third grade, where we attain to that, our recognitional capacity has been elaborated into a further and better state of practical readiness, a fully operational state that engages with inquiry, experience, secondness, guessing, retroduction...with these things as they are or can be in life. See again the citations displayed at the beginning of the present paragraph. The practical conceptions mentioned in the second relate to habits of action. They also involve a rather specific orientation towards possible or actual future experience. (Compare CP 8.194.) It is at the third grade (presupposing and not superseding the first and the second) that the grasp of sense/significance/Sinn of a symbol has to be made complete. The third grade of clarity is only attained where, independently of any particular person’s efforts, there awaits any sufficiently intelligent thinker who seeks to grasp the meaning of a given term some publicly completed or completable meaning for it. Peirce offers no unitary or full answer to the question what it is that completes this meaning or saturates our understanding. (See Misak, Tiuth and the End of Inquiry, pp. 12—35). Supposing, though, that somehow this saturation has been achieved or is in the process of being achieved, we may expect the proposition expressed by a sentence comprising such symbols to depend on the sense of its constituent parts. Thinkers’ corresponding grasp of the proposition and its truth-condition, arrived at through their understanding of a sentence’s mode of composition, may or may not put them into a position to verify or falsify the proposition expressed by the sentence. Where thinkers can verify or falsify, it needs to be no accident that that is so. But being placed to verify or falsify is not the general form of the kind of readiness which Peirce is concerned with.

PEIRCE

159

Pragmatism leaves the concept of truth where it is. Its special contribution lies at the level of sense.** In so far as Peircean pragmatism is revolutionary, its chief purport relates to the question of what it takes to achieve the operational clarity and effectiveness of real understanding. The Peircean conception of truth itself is more or less conservative. It is the theory of enquiry that is ground-breaking.

APPENDIX Peirce and Hume on arguing from the known to the unknown Inquiry conducted along the lines of Peirce’s fourth method, inheriting as it does the merits of various predecessors, is a process that gathers rational strength as it gathers force and gathers force as it gathers rational strength. On the proper understanding of this process, we have said, truth is conceived as the property that we can hope to steer our inquiry to home upon. The beliefs that inquiry furnishes to us are beliefs that it is rational for us, however fallibly, to persist in until specific grounds for doubt present themselves. And the method of inquiry makes room for any or all modes of research or criticism, whether commonsensical or scientific, that promise to steer us to beliefs along routes not extraneous to the facts they are concerned with. One who conducts himself on these principles will be no more eager to define “rational” than he is to circumscribe legitimate methods of exploration and discovery. But such a person will surely insist that the method of inquiry is a fully rational way of arguing from the known to the unknown. It is a paragon of rationality, he can say. Such an attitude will appear to conflict with something commonly regarded as one of the great insights of David Hume. Hume points out in Enquiry Concerning Human Understanding that all reasonings concerning matters of fact are founded in the relations of cause and effect, and the foun-

dation of our understanding of these is experience. But here he claims to

find a problem.” How may I rationally infer from past bread-eatings’ having 24. Here and everywhere in this exposition of Peirce, the word “sense” is transplanted from the current philosophical usage followed throughout this book. (The same goes for “concept” and “conception”.) Peirce’s own usage has separated his efforts too long from readers of the present day. 25. A problem often known as the problem of induction. But, as we have seen, this 1s not Peirce’s use of the word.

160

PEIRCE

nourished me that similar eatings will nourish me? If there is such an inference it is not intuitive (knowable without demonstration); nor yet is it demonstrative. What is it then? It is experimental, Hume imagines your saying. But to this he replies that experimental reasonings already presuppose that the future will resemble the past. How then, presupposing this, can they show or even suggest that the future will resemble the past? Hume infers that it is not reasoning that engages us to suppose the future will resemble the past. It is habit, not reason. What ought Peirce to say? Peirce would begin by agreeing that inference from the known to the unknown is a matter of habits and is not demonstrative. But habits, he would insist, can be good or bad. And good habits can exemplify a distinctive form of reasonableness. (See section I1.) After all, we

need to argue from the known to the unknown. If we need to, then it is reasonable for us to do so (intuitively rational you can say, if you wish) and it is irrational for us not to do so—provided that we do not entrust ourselves to a particular policy that there is reason for us to regard as reckless (as exposing our vital needs to risks there is no necessity for us to incur) or as ill-calculated to bring us to beliefs we shall accept for reasons non-extraneous to the facts. [f Hume

wants to make a point about habit, Peirce can say, let

him make it as a point about the relevance of habits to his science of logic. [t 1s a good point, and Peirce would second it. But it is no excuse for an assault on reason as such—unless Hume’s aim is to put himself at the centre of a long-running controversy. It is easy to imagine that, if he were allowed a response, Hume would still

press upon the question how Peirce can argue non-question-beggingly from past nourishings by bread to future nourishings by bread, if this presupposes the general claim that the future will resemble the past—which is something yet harder to establish than future nourishings by bread. To this Peirce would surely reply (here anticipating Karl Popper) that good arguments from the known to the unknown had better not presuppose that the future will resemble the past. For it is not even true that it will! Nature is not regular....It is true that the special laws and regularities are innumerable; but nobody thinks of the irregularities, which are infinitely more frequent. (W 2, 264)

Moreover, when we argue from past nourishings by bread to future nourishings by bread, we are not, according to Peirce, simply extrapolating a past regularity. That is never, in his view, a valid procedure. If that was what

PEIRCE

161

Hume was attacking, then Hume was right, Peirce will say, but far short of

the conclusion that Hume was aiming for. When we extrapolate a regularity, there has to be another reason to do so beside the fact that the regularity has held so far. Even in the special case of the “particular methods” beloved of the inductive reliabilists, Peirce would say, it would be futile to argue from the mere past success of a method to its future success. With any method, there has to have been something else to commend it. And here is the role of abductive thought. (Compare section VII.) Let us distinguish here two cases. The first is that of the ordinary person with an ordinary need not to starve, who wants to prolong life and needs some determinate way, here and now, of sorting the nourishing from the non-nourishing. Any such sorting must either deploy existing categoriza-

tions such as “bread” or deploy improvements upon the categorizations the person already has. There is nowhere else for him to work from. In so far as “bread” is one of the categorizations on which the person habitually relies and on which he acts, he is committed to think that there is something about bread—a substance that he can identify where necessary with some precautionary care—which would explain why it nourishes. Under interrogation he would appear to be committed to think there is some generalization about bread and nourishing (one he may not know how on demand to formulate very carefully or articulately) which would not, if it were formulated and tested, be falsified. (Compare our discussions of perception and memory at section VI.) If the question were raised why, once formulated or reformulated, any such generalization should be relied upon, the person might reply first that faith in this is a much more reasonable faith than faith in the future’s resembling the past; and second that some such generalizations have to be relied upon if life is to go on. Criticize that basis and he will look for something better, for something that is adequate for the matter in hand. But the only point of departure in the search for something better is the place where we are. Cp. Plato Phaedo

101d (ad fin.); Aristotle

Nicomachean Ethics 1095°2.

As one makes more and more explicit that which an ordinary person might say in defence of his habit of taking bread to be nourishing, one converges on the case where a more theoretical answer is to be given. That theoretical answer is not, according to the pragmatist, essentially different, only more discursive. It begins in the same place. If we are to do what we are naturally committed to do and argue in this case from the known to the unknown, then we must begin by trying to understand the thing that is

162

PEIRCE

known. So the thing we must understand better is bread. The problem of understanding or singling out this particular kind is not, however, one we need to solve all on its own, or without any reference to the state of our inquiries into other empirical questions. We can only approach it from where we are at any given point in our inquiry. Looking at things from where he is, the inquirer notices the remarkable phenomenon that some have nothing to eat and starve and die while others who eat, and eat bread among other things, sustain their life. (See the third epigraph.) If bread nourished then it would be a matter of course that those who ate it sustained life. So it seems, according to the abductive hypothesis, that bread nourishes. This is a generalization worthy to test; and in the interim it is one to live by, pending any refinement or refutation that it may suffer. Hume or his followers will notice that the Peircean strategy leans here upon the fundamental hypothesis. So they are bound to inquire what grounds the fundamental hypothesis itself. One tempting answer is: “Nothing holds or is so or obtains but that there is some reason why it is so”’. Readers of

Leibniz will recognize the thought.” It is true that the claim is quite as general as the claim that the future will resemble the past, but it is a far better candidate to be the regulative assumption of inquiry. At least it suggests nothing that is manifestly false. Still better, it scarcely needs to be thought of as an empirical generalization about reality. It proposes rather a certain attitude towards reality—an attitude that it would be unreasonable for us not to share in if we are to do that which we shall almost certainly perish by not doing. What then is the connection between Sufficient Reason and the twofold procedure that Peirce commends to us? Suppose our methodological stand is that nothing holds unless there is a reason why it should. Then we are committed to think that, if some phenomenon C obtains, something must be true which explains why C obtains. But then it must be possible for us to argue backwards, against the current of deductive sequence, and to infer from C’s obtaining whatever best explains why C obtains. But here we come back to abduction, which supplies selected materials to induction. (See VII end.) If stuffs like bread nourish, there must be something or other about them in virtue of which they do that.... Of course “bread” may be the wrong basis for an abduction and ensuing generalization. But this is a question that we can only attempt from the midst of a large background, already 26. Leibmz puts the claim to a theological use. Indeed, he sometimes tries to prove by its means the existence of God. But Sufficient Reason itselfis neither theological nor teleological in its original purport. For more on some of these matters, see note §.

PEIRCE

163

given, of collateral beliefs, nonarbitrary suspicions, conjectures, questions,

and the rest. The label “bread” is our provisional place-holder for some stuff or other that makes a difference to life’s being sustained. (Cp. CP 4.234, 1902) “Bread” provides us with the materials for a hypothesis that can be tested, qualified, reformulated, tested again, and so on. In practice and so far,

some hypotheses have stood up.When they fail, we will start repairing them. It would be arbitrary to proceed in any other way and worse than arbitrary not to proceed in this one. Of this much we can be intuitively certain. None of this proves that bread will continue to nourish. Such a proof was not what Hume took himself to be entitled to ask for. What he asked was what kind of reasonable inference it is that gives the conclusion (however fallible) that bread will nourish. The answer to his question is that it is a fallible extrapolation, which we should be practically irrational not to attempt, from an abductive hypothesis that we should have been practically irrational not to try to formulate and test, an abductive hypothesis arrived at from wherever we actually are, and made in accordance with the branch of thinking that the nineteenth century called logic. Except in so far as it subsumes the science of deduction, it is not the business of such logic, and it does not need to be its business, to furnish infallible directions by which to argue from the known to the unknown*—only directions that it would be unreasonable not to employ. Let those who are expert in the classification of forms of reasonableness now classify the various elements of this response to Hume’s challenge and let them assign them variously to the intuitive, the demonstrative, and the experimental.*® REFERENCES

TO

WRITINGS

OF

PEIRCE

(CP) 1931—58 The Collected Paper of Charles Sanders Peirce, 8 vols, ed. C. Hartshorne and P Weiss (vols 1—6) and A. Burks (vols 7-8) Cambridge, MA: Harvard University Press.

(W) 1982— Writings of Charles S. Peirce: A Chronological Edition, 6 vols, ed. M. Fisch, C. Kloesel, and N. Houser Bloomington, IN: Indiana University Press. (NE) 1976 The New Elements of Mathematics, 4 vols, ed. C. Eisele, The Hague Paris: Mouton; Atlantic Highlands, NJ: Humanities Press.

and

27. Or even to furnish procedures that “will, if persisted in long enough, assuredly correct any error concerning future experience into which [they|] may temporarily lead us” (CP 2.769, 1905). Peirce does make such claims, but they are inessential to his contribution to the socalled “problem of induction”. (On the status of these Peircean claims, see Misak, Truth and the End of Inquiry, pp. 111, 115). See also sections I, XI. 28. Peirce refers to Hume rather infrequently. But see CP 6.500, 1906 6.605, 1891 5.505, 1905.

10 An Indefinibilist-cumSubstantivist Account of Truth

and the Marks of Truth' Philosophers and logicians have expended enormous effort upon the determination of the extension of the concept of truth. Remarkable discoveries have resulted. A different question is this: what are the marks of that concept? What is truth like? “Deflationists” and their various allies will regard any such approach to truth as sadly mistaken—and Tarski’s famous paper itself, some say, illustrates precisely this. Here, however, the deflationists misread Tarski, who

saw himself as furthering the substantial account of

truth offered by Kotarbinski. The chapter seeks then to enumerate the marks of truth. Truth is the first and foremost dimension of assessment for beliefs, statements or utterances. Secondly, if s is true then s will under favourable circumstances com-

mand a certain convergence, not least among those who want or need to know. Thirdly, if s is true

then

s has content

which

can

be

accepted

or

contested. Fourth (a mark which is wide open to misunderstanding) every truth is true in virtue of something.

Fifth, if two

sentences are true, then

their conjunction will be true (as will be the disjunction). Finally, from

Frege, “truth is the goal”—a claim pregnant surely with practical, even political, import, or so history suggests. Witness the disasters that have been

attendant upon its neglect or its loss.

1. Suppose a whole cluster of fundamental terms came into being more or less simultaneously and, onwards from the dawn of thought, the uses of these terms evolved simultaneously in some sort of mutual reciprocity. If that 1s how things began, and if the idea of truth depends not so much on an inner content as on what it can combine with other concepts to enable us to say, then there will be connections to explore and elucidations to look 1. Where 1t holds universally that, for all x, if x 1s F then x 1s G, then the concept G 1s a mark of the concept E

Meaning, Truth, and the Limits of Analysis:Ten Studies. David Wiggins. Oxford University Press (2022). © David Wiggins. DOI: 10.1093/0s0/9780198726173.003.0010

TRUTH

165

for.” The idea of truth leads easily into the idea of coherence, for instance— but coherence with that which is given or treated as frue. Correspondence too seems to have something to do with truth. But the only form of correspondence that will serve the intended purpose is one that it seems impossible to delineate without reference to the idea of truth itself. We need to begin from a whole configuration. 2. In the course of the present chapter, these claims are taken a little further. But at the outset, before we attempt anything else, it may be sensible to attend to an altogether different philosophical position, namely deflationism about truth. Deflationists would have us look for nothing else by way of a philosophy of truth beyond a simple generalization on the pattern of the T-sentence form “ ‘snow is white’ is true if and only if snow is white”. The deflationist doctrine is attractive to philosophers who are eager to fumigate the apparatus of ordinary human thought—all for the supposed benefit of the scientific world-view.They derive a satisfaction I have never properly understood from the idea that the philosophy of truth makes too much of the idea of €¢ ¢

truth. “ ‘Snow 1is white’ is true” is nothing more (I heard one of them say)

than a fluffed-up version of ‘snow is white’. 3. Any such view is seriously mistaken. Deflationists refer habitually to the much-discussed

so-called

Convention

T, which

Tarski

announces

in

his

famous paper ‘“The concept of truth in formalized languages’ (1935). In practice the deflationists treat Tarski’s Convention T itself as a would-be general schema—the schema whose most famous instance is the T-sentence “Snow is white” is true just if snow is white. ConventionT as Tarski gives it affords no such schema. Convention T is an ordinary, closed sentence whose announced concern is with definitions of truth that are materially adequate. Tarski states that a definition in the metalanguage of truth-in-L for the object language L will be materially adequate if and only if (a) the definition is formally correct (that is to say, inter alia, that it makes use only of terms drawn from L or from the morphology of L or from elementary set theory) and (b) the definition implies, for each object-

language sentence s of L, a biconditional of the form ‘s is true if and only if p’, where ‘p’ holds a place for the use of sentence s itself or of some transla-

2. On elucidation, see Wittgenstein Tractatus 3.263, 4.11, 4.026. For further references to the idea of elucidation, see my ‘Replies’ (1996), page 283, note 45.

166

TRUTH

tion of s into the metalanguage. Thus stated, Convention T is entirely consistent with the denial—Tarski’s own denial—that there is any correct and complete generalization to be made upon the pattern “ ‘Snow is white’ is true if and only if snow is white”. Any such generalization—Tarski says— would inevitably lead to contradiction and paradox.’ In Tarski’s methodology of deductive science, truth—substantive truth— goes along with technical concepts such as definability, provability, conse€¢ ¢

quence, consistency, independence, which

form

the subject matter of so

many of his inquiries. Truth itself, however, is not a technical notion. Any

account of it will need to be materially adequate, he says. What does he mean by that? There is no need to speculate. In another paper from the same era (a paper of 1931) Tarski writes: “the question arises whether the definitions

he has just constructed are also adequate materially.. . Do they in fact grasp the current meaning of the notion as it is known intuitively? Properly understood, this question contains no problems of a purely mathematical nature, but it is nevertheless of capital importance for our considerations” (129 [italics added]). Evidently then “material adequacy” as it figures in Tarski’s account of ‘truth-definition’ is plenary adequacy. Material adequacy condenses not only freedom from antinomy* but also fidelity.” Truth itself is then a substantial given, a given to which every acceptable determination for specified L of the extension of the concept true-in-L needs to be faithful. It follows that, if the determination of the extension of true-in-L is to pin down something with the right sort of interest, then it must determine the extension of some restriction to L-sentences of the plenary concept of truth—the ordinary truth which is the goal of all serious investigation into the reality of anything. Suitable determinations of such an extension will not only keep out paradox. In the name of material adequacy they will also respect the philosophical idea, as Kotarbinski puts it in the book which Tarski says he (Tarski)

“consulted repeatedly”, of “accordance with reality”. What did that involve? Well, not a revival of the ideas of copy or facsimile, Kotarbinski insisted, but

at least this: that an adequate definition of true-in-L should do justice to 3. The liar paradox, for instance, More generally see, for instance, the survey thatVan McGee offers in the Routledge Encyclopedia of Philosophy vol. IV, in his article ‘Semantic paradoxes and theories of truth’. 4. That 1s to say that any full explanation of formal correctness and the rest would involve or presuppose the word by word account of the vocabulary and constructions of a given language and Tarski’s principled segregation of object language and metalanguage, his finding against languages that are ‘semantically closed’, and his diagnosis (motivating the finding against semantic closure) of the semantic paradoxes. 5. See especially Tarski (1931), page 129, quoted above.

TRUTH

167

that which is conveyed by the Kotarbinskian schema (itself reminiscent of Wittgenstein’s

Tractatus Logico-Philosophicus, see 4.022

4.024

4.062): ‘John

judges truly if and only if John judges that things are thus and so; and things are in fact thus and so’.° But it is of precisely this that we shall be assured if the determination of the extension of ‘true in L’ implies such a T-sentence for each sentence of L. Without substantial loss, Tarski’s Convention T sim-

plifies and purifies Kotarbinski’s requirement. But it keeps faith at every point with the spirit of his teacher’s discussion. 4. What can all this show? It shows that an upholder of Tarski’s scientific semantics and his methodology of the deductive sciences would have a perfect liberty to direct an entirely disengaged benevolence towardrs the theoretical concerns of one who wants even more from a philosophy of truth than Tarski himself was going to offer. What mattered for his own philosophy of truth was only that, for numerous kinds of object language, Tarski could show what it would take, axiom by axiom, to construct faithful,

serviceable truth-definitions for them. ConventionT did not summarize or condense these truth-definitions. Nor did it give a general recipe for making them. It constrained recipes otherwise devised. It prepared the ground also for Tarski’s own proof that truth (ordinary truth) and provability can never exactly coincide. (See his 1935, pp. 186, 198.)

5. So far so good. But deflationists may not be satisfied. Perhaps something else needs to be done in order to defuse the very inclination or impulse to declare that “it’s true that snow is white” is just a fluffed-up way of saying that snow is white. Under the influence of Ramsey’s first and less satisfactory work on truth, A. J. Ayer wrote as follows in “The Criterion of Truth’ in Analysis, 3, 1, 1935 (compare Language, Truth and Logic, Chapter 5): ‘...to ask “What is truth?’ is tantamount to asking “What is the analysis of the sentence ‘p is true’?’ where the values of p are propositions. But it is evident that in a sentence of the form “p is true” the reference to truth never adds anything to the sense. If I say that it is true that Shakespeare wrote Hamlet,

6. See Kotarbinski (1929), pp. 126 ff.. See pp. 106—7 in the English 1966 translation, Gnoseology. Tarski’s remark about his repeatedly consulting Kotarbinski’s (1929) will be found in the footnote on page 153 ofTarski’s (1935). For Kotarbinski’s account of truth as expounded in the book cited in the original Bibliography for Tarski (1935) and the part that it played in the genesis of Tarski’s own account of truth, see also my NV'T pages 333—5. I note in passing how the almost forgotten resonance of Wittgenstein’s Tractatus echoing Frege’s Grundgesetze 1.32 has left numerous other traces: e.g.W. H. Auden,“A sentence uttered makes a world appear where all things happen as 1t says they do”.

168

TRUTH

I am saying no more than that Shakespeare wrote Hamlet. ... The words ‘true’ and ‘false’ are not used to stand for anything, but function in the sentence merely as assertion and negation signs. That is to say truth and falsehood are not genuine concepts. Consequently there can be no logical problem concerning the nature of truth.

Since Ayer offers an argument, let us try to set it out. Let square brackets around a sentence designate the sense of the sentence. Let mean ¢ ~ ~

“amounts to the same as”. Let “+”" stand for concatenation. Then

the argu-

ment proceeds through a sequence of would-be equivalences-: [It is true that Shakespeare wrote Hamlet] =~ [Shakespeare wrote Hamlet]

~ [E] + [Shakespeare wrote Hamlet|, where & is null, = [It’s true that] + [Shakespeare wrote Hamlet]

Therefore [It’s true that] = [E]

Once the argument is set out in this way, it can be appraised. And here, to begin with, is a similar argument made on a similar principle: [Socrates killed Socrates] = [Socrates was killed by Socrates] Furthermore

[Socrates] = [Socrates]. Therefore

[killed] = [was killed by]

Something is wrong. Nor do we need to look very far to find what it is. Geach has trodden this ground: We cannot infer that, if two propositions verbally differ precisely in that one contains the expression E , and the other expression E , then if the total force of the two propositions is the same, we may cancel out the identical parts and say that E here means the same as E .’

6. So much for one futile and almost forgotten attempt to show that and falsehood are not the genuine concepts we take them for. Since there have been numerous other attempts. If present-day deflationism any grounds for itself, would the arguments offered for it be any 7. P. T. Geach, Reference and Generality, (1962, Cornell), p. 61.

truth then, gave more

TRUTH

169

persuasive than the argument that is given by Ayer? Most likely, they would derive from a reinvention of it. Or is the deflationist idea something slightly different, namely the idea that, when anyone learns the sense of the word “true”, they have to learn it by learning some disquotational schema and learning that that is all there is for them to learn? If that is the thought, however, I say that only a cramp of the imagination could prompt such a suggestion. A second comment is that there had better be some other possibilities. Not only does the disquotational idea fall far short of illuminating what Dummett would call the point of the attribution of truth.® Not only does disquotationalism prescind from the normative force of “true” and cause this to vanish in a way radically destructive of that point. If the only general thing that could be said about “true” were that it was a device of disquotation, then that would come close to building into the proposed sense of “true” the one thing that can be depended upon to create paradox.’ 7. Tarski’s own finding has prompted all sorts of technical questions. (See note 6 and the reference there to McGee.) But these questions ought not to be allowed to take up the whole sky. For Tarski’s work clears the ground for further, more philosophical reflections. From this point, let us think of this word “true” as having the content that human beings grasp by immersion in practices of ordinary speech and thought that make ordinary use of the concept, and think of these practices are shaped by our need to find out enough about the world at large in order to act seriously and effectively there. 8. How much more is there to say then about the content of “true’? Ordinary philosophical intuition, further sustained by Frege and Wittgenstein—see the claims of theirs reported in Chapter 1—suggests that there is a notion of sense, significance, or conveyance of literal meaning for

which the following holds: (*) where s 1s a declarative sentence, s signifies or conveys that p if and only if whether s counts as true or not depends upon whether p. What can we derive from this about truth? (*) looks more or less correct

just as it stands. But, as it figures here, the word “depends” has to connote semantical dependence. Can we give an account of the semantical that is independent of the idea of truth? Surely not. But the possibility can still 8. Michael Dummett “Truth’: Proceedings of the Aristotelian Society, 1959. For further discussion of this passage of Dummett’s article, see section 12. 9. Compare Tarski (1935) p. 156 footnote.

170

TRUTH

suggest itself that, within a properly simultaneous elaboration of truth and meaning, we can try to answer the question: what is truth like? What sort of attribute is it? 9. Among the answers to this question [ have been drawn to are the following: (1) If s 1s true then s passes muster in that dimension of assessment in which

s demands, qua s, to be assessed. Normally then (play apart) truth is the first dimension of assessment for beliefs, statements, utterances; (2) If s 1s true, then s will under favourable circumstances command a con-

vergence of opinion among those properly placed to judge the matter in question and the best explanation of the existence of this convergence, driven as it 1s by our need, will require the actual truth of s;

If so much can be established, little or nothing else is needed to extract from the ideas of agreement and disagreement an idea of the content on which there is or is not convergence: (3) For any s, if s is true, then s has content; and, if s has content, then s says something about something; s is open to agreement or disagreement;

In earlier writings I have been tempted to add here, simply on the basis of a familiar intuition about truth: (4) Every true belief (every truth) is true in virtue of something;

There 1s something right about this too. But if we accept it, does that commit us to find an ontology of “truth-makers™? It 1s often said by critics of that ontology that (4) works fairly well for ordinary subject-predicate sentences—where it is meant to be obvious what the “truth-maker” is to be—but becomes more problematic beyond these. But that critique itself deserves further scrutiny. In the well-worn case of “The cat is sitting on the mat”, for instance, what or where is the truth-

maker? Is it the very animal which is sitting on the mat at the time t in question? So it has appeared to some defenders of “correspondence”. But any truth-maker for “the cat is sitting on the mat” would need to include something about the cat and its sitting on a mat. Shall we follow through then by ruling that the cat-sitting-on-the-mat-at-t should be the truthmaker? But what exactly is this? Is it a temporal slice of the animal? Was a temporal slice really what we wanted to talk about when we said the cat was sitting on the mat? Would it not be better then to clear things up by replacing

TRUTH

171

the cat-sitting-on-the-mat, where “sitting” is simply a participle, and making the truth-maker be the cat’s sitting on the mat? Note however that, after this shift, which is surely benign (indeed compulsory), the word “sitting” is no longer a participle. It is a sort of noun. It is a gerund.'® The gerund is a kind of expression that we are even more quickly forced into in cases where our report reaches beyond the simple subject-predicate form. Consider now “nobody came to the party”’. Here the truth-maker must be nobody’s coming to the party. And here too what we have is a gerundial, not a participial construction. It is in virtue of nobody’s coming or having come to the party that “nobody came to the party’ is true. Reverting to the earlier example, I claim then that, because a truth-maker

theorist is compelled to have recourse to the gerund “sitting” or some other verbal noun, and because the result is no more than a grammatical transformation of “the cat sat on the mat”, truth-maker theory appears empty. Can (4) amount to much more than our saying that, if it is true that the cat is on the mat, then the cat must

indeed be on the mat? But that dispenses

altogether with the ontology of truth-makers. Is that the final put-down for the correspondence theory? For the truthmaker version, yes, I think so. Not yet however for the simple idea of correspondence, if we are prepared to understand truth as correspondence with the findings of the best (most accurate, most responsible, most careful) firstorder answer to the questions that the world in which we live prompts us to ask about that world. On these terms, everyone can cheerfully accept the notion of correspondence as correspondence to something’s obtaining (gerund) in the actual world.

Finally, capitalizing upon the role of truth within logic, we may enlarge the scope of the philosophical account of truth once more by adding further afthirmations such as (s) If s is true and s_is true, then their conjunction is true. 10. Only marks (1) and (2) require further discussion. For mark (3) will col-

lect everything it needs from the arguments to be consolidated here for marks (1) and (2).

10. The gerund for any given verb functions as a noun and stands for the act, action or activity or state ... that the verb stands for.“One escapes by running” introduces the act.“She thinks now about her eating that orange last night” introduces an action. That action was her doing (gerund) of an act of eating (another gerund). (For this “of”’, consider ‘the city of Athens’. That is just Athens.)

172

TRUTH Mark (1). When someone believes that p, we may predicate truth of the

object of their belief, i.e. the proposition or judgment; or we may predicate truth of the sentence which (in context) expresses that belief. Taking these

usages as interdependent, I argue for the first mark of truth as follows." Suppose truth were not the primary dimension of assessment for beliefs, so that falsehood was not accounted a cardinal defect in them. Then there would be no reason to suppose (as the business of interpretation requires) that it is a constitutive norm for beliefs as such to be aimed at truth. But if the belief that p did not have, as such, to carry with it a sensitivity to the question whether indeed p, then it would not be an option for us to interpret one another’s beliefs by asking what it is that the other person who has said that p (or done this or that) must, it seems, believe.

Suppose that truth were not a primary dimension of assessment for the declarative utterance of a sentence. To conceive of that is to conceive of there being nothing in the practice of speakers of sentences to sustain the idea that it is a constitutive norm for such speakers to aim at truth in the sentences they utter declaratively. But, in that case, the utterance of a sentence would give no ground, however defeasible, for the expectation that a speaker believed the thing he was saying. And then, without the supposition that truth is the primary dimension of assessment for declarative utterances, there would be nothing left to connect utterances of the sentence with the one thing that makes interpretation possible, namely what it would be reasonable for anyone to believe if they were placed as this subject is placed and liable to be impinged upon by whatever impinges upon this subject. There are three assumptions here: (a) that there can be no such thing as

meaning unless there is the possibility of the interpretation of subjects by interpreters—and of interpreters by subjects; (b) that interpreting subjects have to see one another as party to some shared norm of speech and thought departures from which stand in prima facie need of explanation; and (c) that the idea of such a norm imports the idea of information, where the discrimination of good from bad information has its rational culmination in belief. Holding these assumptions to be indispensable, however, and taking them as integral to the very existence of thinking or believing or of one person’s interpreting another, let us simply avow our acceptance of (a), (b), and (c). So much for the first mark of truth. 11. The arguments I have used go back to Needs, Values, Tnith Chapter IV. The most recent discussion I have attempted of the second mark 1s in my (19964). See also my ETL (2006). The latter embraces also the less problematic third mark.

TRUTH

173

11. Mark (2). If we confine attention to that which i1s both knowable and

true then there will be at least two cases to consider. The first is the case where a question—*‘what is six times seven?’ (for instance)—leaves room for

only one possible answer. Then there is the case where the evidence in front of us all points to one conclusion. The question prescribes the method of finding the answer. Here too there may well be absolutely nothing else to think. If so, it is no wonder that that is what people do think. Even where they consult one another, there was no real need for them to do so—still less any need for them to arrange to “sing from the same hymn-sheet”."” They can sing from their own text and their own score. When each has their own text but in substance these texts all agree, how can that help but count as an interesting or real consensus or convergence?"’

How close does there being nothing else to think but that p come to its being true that p? Are we within sight here of some sort of full explanatory definition? Not really. For how could we apply this to the case of unknowable truths? And, waiving those, what happens if we try to think harder about “think that..””? Can we pin down the kind of thinking we need without leaning heavily upon the idea of truth or else pinning it down by borrowing Davidson’s verb-phrase “hold true”? Hardly. 12. If “true” lacks an (ordinary) definition, and has to be characterized in

these indirect ways, then how do I say that we catch on to the idea? Rather than attempt an utterly conjectural history or genealogy of the idea, I say again, taking thought, speech, and civilization for granted, that the idea of truth is something we grasp by being gradually drawn into the whole set of practices, practices that culminate in speaking, understanding, and thinking— and in trying to get something right. The ideas of content and truth are immanent in these practices and the norms that govern these practices.To predicate

truth of what

someone

says is, moreover, to ascribe to it and its

speaker a certain sort of success. (See again Dummett (1958/9).) Something

that has the property of truth has the property that it must normally (if it is 12.

For more

on this matter, see my ETL

(2006), p. 362.

13. If there is enough here to begin the case for (2), the second mark of truth, then consider now the case where a subject matter fails to produce the kind of convergence that we are concerned with. Then anyone who wants to defend the objectivity of that particular subject matter will have something to explain. But that is what you would expect. One such explanation is that— as 1n the case of some ethical predications—there 1s more to know, more to experience or see or suffer, before our collective understanding of the ethical property in question can approach

completion. Meanwhile understanding.

we

must

use

1t in full knowledge

of the

incompleteness

of our

174

TRUTH

ever to be a candidate for truth) purport that it has. By virtue of its particular content (see mark (3)), it has the property that a belief, declaration or utter-

ance ought to have. It answers to out need. If the scientific world-view is seriously incomplete or defective without the idea of truth and our idea of truth is an inescapably normative idea, then the philosophers who have tried to bury this fact may need to re-examine their conception of the scientific world-view—dispensing at the same time with second-hand

notions of the normative and the valuational.

13. Let us conclude with one more sentence from Frege: “the word ‘true’ specifies the goal” (Logic, 1897 (PW, 126)). Here let us acknowledge one more

truth about truth—one

more mark, if you wish, of the concept of

truth. Frege’s claim here is pregnant (or so history may suggest) with reproach against those who have not stopped to think, to find out, or to follow through upon that which proper inquiry might have shown. If around here no mishap or disaster ensued upon a mistake they could have avoided, they were luckier than they deserved to be. We cannot leave it to chance whether what we believe is true.' BIBLIOGRAPHY

OF

WORKS

CONSULTED

OR

REFERRED

TO

Davidson, Donald 1967 “Truth and Meaning’, Synthese 17. Dummett, Michael

1958/9 “Truth’, Proceedings of the Aristotelian Society, vol. 9.

Frege, Gottlob [1979]| Posthumous Writings. Grandy, Richard 1973 ‘Reference, Meaning and Belief’, Journal of Philosophy. Kotarbinski,T. 1929 Elementy Teorji poznania, logiki formalnej i metodologji nauk, Lwow, translated by O. Wojtasiewicz as Gnoseology, Pergamon Press, 1966.

Mackie,J. L. 1957 Ethics: Inventing Right and Wrong, Harmondsworth: Penguin. McDowell, John 1976 ‘Truth-Conditions, Bivalence and Verificationism’in G. Evans

and J. McDowell

eds Tiuth and Meaning: Essays in Semantics, Oxford: Oxford

University Press. McDowell, John 1977 ‘On the Sense and Reference of a Proper Name’, Mind 86.

McGee, Van ‘Semantic paradoxes and theories of truth’, Routledge Encyclopedia of Philosophy (vol. IV) 14. This chapter reassembles materials from various papers and chapters dated 1980 onwards. In the light of Ramsey’s second and improved account of truth (1991) 1t will no longer quite do to say (as [ did) that truth cannot be defined. See E P. Ramsey On Truth, ed. N. Rescher and D. Majer, Dordrecht, Kluwer (1991). In a plain English version proposed by lan Rumfitt (see The Boundary Stones of Thought, Oxford, Clarendon Press 2015, p. 304), Ramsey’s revised view may be stated (with the help of A. N. Prior) as follows: a statement 1s true if and only if there 1s a way it says things are and they are thus. But about truth itself there 1s surely something more. Why 1s truth worth striving for? Chapter 10 tries to say.

TRUTH

175

Peirce, C. S. 1877 “The Fixation of Belief’, Popular Science Monthly XII, 1—15.

Putnam, Hilary 1978 Meaning and the Moral Sciences, London: Routledge & Kegan Paul. Quine,W. V. 1952 ‘On an Application of Tarski’s Theory of Truth’, reprinted in his Selected Logic Papers, Random House, 1966. Quine,W. V. 1960 Word and Object, Cambridge, MA: Harvard University Press. Quine, W. V. 1992 Pursuit of Truth, revised edition, Cambridge, MA: Harvard University Press. Tarski, Alfred 1931 ‘On definable sets of real numbers’, in Logic, Semantics and

Metamathematics, trans.J. H. Woodger, Oxford: Oxford University Press, 1955. Tarski, Alfred 1935 “The concept of truth in formalized languages’, in Logic, Semantics and Metamathematics, trans.J. H. Woodger, Oxford: Oxford University Press, 1955. Often referred to as Wahrheitsbegriff. Tarski, Alfred 1944 ‘“The semantic conception of truth’, Philosophy and Phenomenological Research 4, 341—76. Tarski, Alfred 1967 “Truth and proof’, Scientific American 220, 63—77.

Wiggins, David 1976 ‘“Truth, Invention and the Meaning of Life’, Proceedings of the British Academy 62, pages 331—78. (Reprinted amended in G. Sayre-McCord ed. Essays on Moral Realism. Ithaca, NY: Cornell University Press 1998, and in Needs, Values, Truth, 1987, 1998 below.)

Wiggins, David 1980a “Truth and Interpretation’ in W. Leinfellner, R. Haller, A. Hubner, and P. Weingartner eds Proceedings of the 4th International Wittgenstein Symposium, Kirchberg 1979. Vienna: Holder-Pichler-Tempsky:. Wiggins, David 1980b Sameness and Substance, first edition: Blackwell, Oxford. Wiggins, David 198oc “What would be a Substantial Theory of Truth?’ in Zak van Straaten ed. Philosophical Subjects: Essays Presented to P E Strawson, Oxford: Clarendon Press. Wiggins, David 1987 “Truth As Predicated of Moral Judgements’, in Needs, Values, Truth (1987, 1991, revised 1998).

Wiggins, David 1995 ‘Objective and Subjective in Ethics, with two Postscripts about Truth’, Ratio VIII (3). (This appeared in another form in the third edition of M. Canto-Sperber ed. Dictionnaire de philosophie morale. Paris: Presses Universitaires de France, 2001.)

Wiggins, David 1996a ‘Meaning and Truth Conditions: from Frege’s Grand Design to Davidson’s’in B. Hale and C. Wright eds Blackwell Companion to the Philosophy of Language, Oxford: Blackwell (second edition 2018). See Chapter 1 in the present volume. Wiggins, David 1996b ‘Replies’ in S. Lovibond and Williams eds Identity, Truth and Value: Essays for David Wiggins, Blackwell/ Aristotelian Society, pp. 219—84. See especially pages 265—79. Wiggins, David 1998—2002 Needs, Values, Truth, originally Blackwell 1987. In a second edition (Oxford: Oxford University Press 1991), substantial corrections and additions were made. In 1998, there was a third edition, with some corrections

specifically to the chapter cited above as 1987. An amended third edition was

176

TRUTH

published in 2002, with an extra chapter, by Oxford University Press. Unluckily, when

the 1998 edition was exhausted, the 1998 text was reissued. Matters have

now been put to rights, but bad copies still circulate. Wiggins, David 2001 Sameness and Substance Renewed, Cambridge University Press. (A second edition of 1980b.)

Wiggins, David 2004 ‘Reflections on Inquiry and Truth Arising from Peirce’s Method for the Fixation of Belief” in C. Misak ed. The Cambridge Companion to Peirce, Cambridge: Cambridge University Press. Wiggins, David 2006 Ethics: Tivelve Lectures on the Philosophy of Morality, Chapter 12, Cambridge, MA: Harvard and Harmondsworth: Penguin Books.

Bibliography

WRITINGS

OF

DAVID

WIGGINS

Books

ISTC

Identity and Spatio-Temporal Continuity (Oxford: Blackwell, 1967).

S&ES

Sameness and Substance (Cambridge, MA: Harvard University Press and Oxford: Blackwell, 1980; a second impression appeared in 1981 with errata and addenda.)

Needs, Values, Truth: Blackwell, 1987).

NVT

Essays

in

the

Philosophy

of Value

(Oxford:

NVT

Needs, Values, Truth, second edition, revised (Oxford: Blackwell, 1991).

NVT

Needs, Values, Truth, third edition (Oxford: Clarendon Press, 1998).

with

extra

chapter

(cp.

1998a)

Needs, Values, Truth,amended third edition (Oxford: Oxford University

NVT

Press, 2002; unluckily and by mistake, a subsequent reprint of the book

followed

the

1998

text instead

of this amended

impression

of 2002.) Sameness and Substance Renewed (Cambridge: Cambridge University

S&ESR

Press, 2001; an extended and heavily revised edition of S&S). ETL

Ethics: Tivelve Lectures on the Philosophy of Morality (Harmondsworth: Penguin and Cambridge, MA: Harvard University Press, 2000).

Continuants

Continuants: Their Activity, Their Being, and Their Identity (Oxford: Oxford University Press, 2016).

Other writings 1963

‘The Individuation of Things and Places’, Proceedings of the Aristotelian Society, suppl.Vol. 37, 177—202. Reprinted in corrected form, in the first edition only of M. Loux (ed.), Universals and Particulars (New York: Doubleday 1967).

1965

‘Identity-Statements’ in R. J. Butler (ed.), Analytical Philosophy, second series (Oxford: Blackwell). Reprinted in German as ‘Identititsaussagen’, trans. Matthias

Schirn, in Kuno

Lorenz

(ed.) Identitit und Individuation. Band

1:

Logische Probleme in historischem Aufrifs (Stuttgart: Frommann-Holzboog, 1982). (The chief contention and error of this paper was recanted at ISTC note 7, and 1971a, 17 note b.)

178 19068

BIBLIOGRAPHY

‘On Being in the Same Place at the Same Time’, Philosophical Review, 77:1, pp- 90—s5. (See also 1997c.)

1969

‘Reply to Mr Chandler’, Analysis 29:5, pp. 175—6.

1970

‘Freedom, Knowledge, Belief, and Causality’ in G. Vesey (ed.), Knowledge and Necessity (London: Macmillan), pp. 132—54.

1971a

‘On Sentence Sense,

Word Sense, and Difference of Word Sense:

Towards

a Philosophical Theory of Dictionaries’ and ‘Reply to Mr Alston’ in D. Steinberg and L. Jakobovits (eds), Semantics:An Interdisciplinary Reader in Philosophy, Linguistics and Psycholinguistics (Cambridge: Cambridge University Press). Translated into Greek in Deucalion: A Quarterly Review, 31 (1980), pp. 35—52. 1971b 1973

‘Sentence Meaning, Negation, and Plato’s Problem Gregory Vlastos (ed.) Plato I (New York: Doubleday).

of Non-Being’ in

“Towards a Reasonable Libertarianism’ in Ted Honderich (ed.), Essays on Freedom of Action (London: Routledge & Kegan Paul), pp. 270—302. (Also in amended edition 1979; rewritten for NIV'T and rewritten again for 2003b.)

1974 1975

‘Identity, Designation, Essentialism and Physicalism’, Philosophia, s:1—2, pp. 1—30.

1975—06

‘Deliberation and Practical Reason’, Proceeding of the Aristotelian Society, 76,

‘Essentialism, Continuity, and Identity’, Synthese, 28:3—4, pp. 321—59.

pp- 29—sI.

(Rewritten

for NVT.

An

excerpt

was

rewritten

for

1978c.

Rewritten again for 2008b.) 19762

‘Truth, Invention and the Meaning of Life’, Proceedings of the British Academy, 62, pp. 331—78. Reprinted in G. Sayre-McCord (ed.), Essays on Moral Realism (Ithaca, NY: Cornell University Press, 1988). (Rewritten for

1976b

1976¢C

NV'T.)

‘The De Re “Must”: A Note on the Logical Form of Essentialist Claims’ in Gareth Evans and John McDowell (eds), Truth and Meaning: Essays in Semantics (Oxford: Clarendon Press). ‘Locke, Butler and the Stream of Consciousness: and Men

as a Natural

Kind’, Philosophy, §1:196, pp. 131—58. Reprinted in A. O. Rorty (ed.), The Identities of Persons (Berkeley, CA: University of California Press, 1976). Also reprinted in H. Noonan (ed.), Personal Identity (Aldershot: Dartmouth Publishing, 1993). 1976d

‘Frege’s Problem of the Morning Star and the Evening Star’ in M. Schirn (ed.), Studies on Frege 11: Logic and Philosophy of Language (Stuttgart: FrommannHolzboog).

1976¢

‘Identity, Necessity and Physicalism’ and ‘Reply to Professor Marcus and Mr Hacking’ in Stephan Korner (ed.), Philosophy of Logic (Oxford: Blackwell).

1978a

‘Are the Criteria of Identity that hold for a Work of Art in the Different Arts Aesthetically Relevant? Reply to Richard Wollheim’, Ratio, 20, pp. §2—68.

BIBLIOGRAPHY 1978b

1978¢

1978—9

‘Aurel Thomas

179

Kolnai (1900—1973)’, Introduction (written with Bernard

Williams)

to A. T. Kolnai, Ethics, Value and Reality, ed. B. Klug

FE Dunlop

(London:

Athlone

and

Press).

‘Deliberation and Practical Reason’ in J. Raz (Oxford: Blackwell).

(ed.), Practical Reasoning

“Weakness of Will, Commensurability, and the Objects of Deliberation and Desire’, Proceedings of the Aristotelian Society, 79, pp. 251—77. (Rewritten

and corrected for

NVT.)

19794

‘Ayer on Monism, Pluralism and Essence’ in G. E MacDonald (ed.), Perception and Identity: Essays Presented to A. J. Ayer (Ithaca, NY: Cornell University Press). (Badly mistaken with respect to Leibniz. Superseded entirely by 1987c.)

1979b

‘The Concern to Survive’, Midwest Studies in Philosophy, IV:1, pp. 417—22. (Rewritten for NV'T.)

1979¢

‘On Knowing, Knowing that One Knows and Consciousness’ in E. Saarinen (ed.), Essays in Honour of Jaako Hintikka (Dordrecht: Reidel).

1979d

‘Mereological Essentialism, Asymmetrical Essential Dependence, and the Nature of Continuants’, Grazer Philosophische Studien, 7/8 (Festschrift for

Roderick Chisholm), pp. 297—315. Also in E. Sosa (ed.), Essays in the Philosophy of Roderick M. Chisholm (Amsterdam: Rodopi, 1979; in this place, Chisholm makes a response. See his ‘Objects and Persons: Revisions and Replies’ at pp. 317—88). 1980a

““Most” and “All”: Some Comments on a Familiar Programme, and on the Logical Form of Quantified Sentences’ in M. Platts (ed.), Reference, Truth, and Reality (London: Routledge and Kegan Paul).

1980b

“What would be a Substantial Theory of Truth?’ in Zak van Straaten (ed.) Philosophical ~Subjects: Essays DPresented to P FE Strawson (Oxford: Clarendon Press).

1980c

‘Contingency, Identity, and de re and de dicto Necessity’ in Jonathan Dancy (ed.), Papers on Language and Logic: Proceedings of the 1979 Keele Conference on Language and Logic (Keele: Keele University Library).

1980d

“Truth and Interpretation’ in W.

Leinfellner, R. Haller, A. Hubner, and

P. Weingartner (eds), Proceedings of the 4th International Wittgenstein Symposium, Kirchberg 1979 (Vienna: Holder-Pichler-Tempsky). 1081 1982

‘Public Rationality, Needs, and What Needs are Relative to’ in P. Hall and D. Bannister (eds), Transport and Public Policy (London: Mansell). ‘Heraclitus’

Conceptions

of Flux,

Fire,

and

Material

Persistence’

in

M. Nussbaum and M. Schofield (eds), Language and Logos: Essays for G. E. L. Owen (Cambridge: Cambridge University Press), pp. 1-12. 1984

‘The Sense and Reference of Predicates: Doctrine

and

a Plea

for

the

Copula’,

A Running Repair to Frege’s Philosophical

Quarterly,

34:136,

180

BIBLIOGRAPHY

pp- 311—28. Also in Crispin Wright (Oxtord: Blackwell, 1984).

1985 1985—6

(ed.), Frege: Tradition and Influence

‘Claims of Need’ in Ted Honderich (ed.), Morality and Objectivity: A Tribute to J. L. Mackie (London: Routledge and Kegan Paul). (Rewritten for NVT.) ‘Verbs and Adverbs and some other Modes of Grammatical Combination’,

Proceeding of the Aristotelian Society, 86, pp. 273—304. 1986a

‘On Singling Out an Object Determinately’ in Philip Pettit and John McDowell

(eds), Subject, Thought and Context (Oxford: Clarendon Press).

(Corrected on one important point by the reply to Williamson among ‘Replies’ in Lovibond and Williams, 1996d.) See also S&SR, Chapter 6. 1986b

‘Teleology and the Good Philosophy, 4, pp. 1—18.

1987a

‘The Person as Object of Science, as Subject of Experience, and as Locus of Value’ in A. Peacocke and G. Gillett (eds), Persons and Personality (Oxford: Blackwell).

1987b

in Plato’s Phaedo’, Oxford Studies in Ancient

‘Needs, Need, Needing’ (with Sira Dermen), Journal of Medical Ethics, 13:2, pp. 61-8.

1987¢

‘“The Concept of the Subject contains the Concept of the Predicate: Leibniz on Reason, Truth and Contingency’, in J. J. Thompson (ed.), On Being and Saying: Essays for Richard Cartwright (Cambridge, MA: MIT Press). (Reprinted in Roger Woolhouse (ed.), Leibniz: Critical Assessments, vol Il (London: Routledge and Kegan Paul, 1994), pp. 141—63.)

1990—1

‘Moral Cognitivism, Moral Relativism and Motivating Moral Beliefs’, Proceedings of the Aristotelian Society, 91, pp. 61—85. Reprinted in Archivio di Filosofia Analitica (Roma: Lithos, 1993).

1991a

‘Categorical Requirements: Kant and Hume

on the Idea of Duty’, The

Monist, 74:1, pp. 83—106. 1991b

1991cC

‘Ayer’s Ethical Theory: Emotivism or Subjectivism?’, Philosophy, 30 (Supplement), pp. 181—96. (This is an abbreviation of 1992a. The same Supplement of Philosophy was published as A. Phillips Grifhiths (ed.), A. J. Ayer: Memorial Essays (Cambridge: Cambridge University Press, 1991.) “Temporal Necessity, Time and Ability: A Philosophical Commentary on Diodorus Cronus’ Master Argument as given in the interpretation of

JulesVuillemin’ in G. C. Brittan,Jr (ed.), Causality, Method, and Modality: A Festschrift for Jules Vuillemin (Dordrecht: Kluwer). (Superseded

by 1993c.) 1991d

‘Pourquoi

la notion

de substance parait-elle si difficile’, Philosophie, 30,

pp- 77-89. 1992a

‘Ayer on Morality and Feeling: From Subjectivism to Emotivism and Back?’ in L. E. Hahn (ed.), The Philosophy of A. J. Ayer (LaSalle, IL: Open Court).

1992b

‘L’éthique et la raison’, Studia Philosophica, 51, pp. 75—87.

BIBLIOGRAPHY

1992C

181

‘Meaning, Truth-Conditions, Proposition: Frege’s Doctrine of Sense Retrieved, Resumed and Redeployed in the Light of Certain Recent Criticisms’, Dialectica, 46, pp. 61—90.

1992d 19932

‘Remembering Directly’ in J. Hopkins and A. Savile (eds), Psychoanalysis, Mind, and Art: Essays for Richard Wollheim (Oxtord: Blackwell). ‘Sinn, Bedeutung, et les mots d’espece’, Revue de Theologie et de Philosophie, 24, pp- 225—37. (Translated and substantially rewritten as 1993d.)

1993b

‘Cognitivism, Naturalism, and Normativity’, and ‘A Neglected Position?’

in J. Haldane and C. Wright (eds), Reality, Representation, and Projection (Oxtord: Oxford University Press). 1993¢

‘Time, Ability, and Real Choice’ in P. Horak (ed.), Responsibility (Actes de I'l. I. P. a Prague, 1993; some misprints and mis-edits).

1993d

‘Putnam’s Doctrine of Natural Kind Words and Frege’s Doctrines of Sense, Reference and Extension. Can they Cohere?’in A. W. Moore (ed.), Meaning and Reference (Oxford: Oxford University Press), pp. 192—207. Also (with an answer by Putnam) in P. Clark and B. Hale (eds), Reading Putnam (Oxford: Blackwell, 1994); and in J. Biro and P. Kotatko (eds), Sense and Reference: One Hundred Years Later (Dordrecht: Kluwer 1995).

19942

‘The Kant—Frege—Russell View of Existence: A Rehabilitation of the Second Level View’ in W. S. Armstrong, D. Raftman, and N. Asher (eds), Modality, Morality, and Belief: Essays for Ruth Barcan Marcus (Cambridge: Cambridge University Press).

1994b

‘Vérité et morale’, trans. Monique Canto-Sperber from 1990—1, in M. CantoSperber (ed.), La philosophie morale britannique (Paris: Presses Universitaires de France).

19954

Preface for The Utopian Mind, a posthumous work by Aurel Kolnai, ed. E Dunlop (London: Athlone Press).

1995b

‘Eudaimonism and Realism in Aristotle’s Ethics:A Reply to John McDowell’ in Robert Heinaman (ed.), Aristotle and Moral Realism (London: UCL Press).

1995¢

‘Categorical Requirements’ in R. Hursthouse, G. Lawrence, and W. Quinn (eds), Virtues and Reasons (Oxford: Oxford University Press). (The longer version of 1991a, corrected.)

1995d 1995¢

‘Substance’ in A. C. Grayling (ed.), Philosophy: A Guide through the Subject (Oxtord: Oxford University Press), pp. 214—49. ‘Objective and Subjective in Ethics, with Two Postscripts about Truth’, Ratio, N.S.

8:3, pp. 243—58,

and

in Brad

Hooker

(ed.),

Truth

in Ethics

(Oxford: Blackwell). There is another version in M. Canto-Sperber (ed.), Dictionnaire de philosophie morale (Paris: Presses Universitaires de France). See 2001a. 1996a

‘Meaning and Truth Conditions: From Frege’s Grand Design to Davidson’s’ in B. Hale and C. Wright (eds), Blackwell Companion to the Philosophy of Language (Oxford: Blackwell).

182 1996b

BIBLIOGRAPHY

‘Natural and Artificial Virtues: A Vindication of Hume’s Theoretical Scheme’ in Roger Crisp (ed.), How Should One Live? Essays on the Virtues (Oxford: Clarendon Press).

190906C

‘Language: The Great Conduit’, Times Literary Supplement, 12 April.

1996d

‘Replies’ in Sabina Lovibond and S. G. Williams (eds), Identity, Truth, and Value: Essays for David Wiggins (Oxford: Blackwell/ Aristotelian Society).

1996¢e

‘Sufficient Reason: A Principle in Diverse Guises Ancient and Modern’, Acta Philosophica Fennica, 61, pp. 117—32.

1996f

‘From Piety to a Cosmic Supplement, 4 October.

19974

1997b 1997¢

Order’

(review),

Times

Higher

Education

‘Sortal Concepts:A Reply to Fei Xu’, Mind and Language, 12:3—4,pp- 413—21. ‘Natural Languages as Social Objects’, Philosophy, 72, pp. 499—524. ‘On Being in the Same Place at the Same Time’, a reprint with corrections and postscript of (1968), in M. C. Rea (ed.), Material Constitution: A Reader (Lanham, MD: Rowman & Littlefield).

1997d

‘Editorial Introduction’ for Conditionals by Michael Woods, with a commentary by Dorothy Edgington (Oxford: Clarendon Press).

1998a

‘Incommensurability: Four Proposals’ in Ruth Chang (ed.), Incommensurability, Comparability, and Practical Reasoning (Cambridge, MA: Harvard University Press).

1998b

‘In a Subjectivist Framework, Categorical Imperatives and Real Practical Reasons’in C. Fehige and U. Wessels (eds), Preferences (Berlin: De Gruyter).

1998c

‘Railways, Settlement and Access’ in Anthony Barnett and Roger Scruton (eds), Town and Country (London: Jonathan Cape).

1998d

‘The Right and the Good:and W. D. Ross’s Critique of Consequentialism’, Utilitas, 10:3, pp. 3—24. Reprinted in Royal Institute of Philosophy, Suppl. vol. 47 (September 2000), pp. 175—95.

19098¢

“What is the force of the claim that one needs something?’ in Gillian Brock (ed.), Necessary Goods: Our Responsibilities to Meet Others’ Needs

1999a

‘C. S. Peirce: Belief, Truth, and Going from the Known to the Unknown’,

(Rowman & Littlefield). (An edited version of 1985.)

Canadian Journal of Philosophy, Suppl. vol. XXIV. (Superseded by 2004a.) 1999b

‘Names, Fictional Names and “Really”’, Proceedings of the Aristotelian Society, Suppl. vol. 73, pp. 271—86.

1999~ 2000

‘Nature, Respect for Nature, and the Human Scale of Values’, Proceedings of the Aristotelian Society, 100, pp. 1—32. (The correct text of this paper was given only in the bound Proceedings of the Society.)

2000

‘Sameness, Substance,and the Human Animal’,an interview, The Philosophers’ Magazine, 12, pp. $0-3.

2001a

French version of 1995e in M. Canto-Sperber (ed.), Dictionnaire de Philosophie Morale, third edition (Paris: Presses Universitaires de France).

BIBLIOGRAPHY

183

2001b

‘Valediction for a Subfaculty’, Oxford Magazine, 191, fourth week, Trinity.

20022

‘An Indefinibilist-cum-Normative View of Truth and the Marks of Truth’ in R. Schantz (ed.), What is Truth? (Berlin and

New York: de Gruyter).

2002b

‘Identity and Supervenience’ in A. Bottani. M. Carrara, and P. Giretta (eds), Individuals, Essence, and Identity (Dordrecht: Kluwer), pp. 247—66.

2003a

‘Existence and Contingency: A Note’, Philosophy, 78:4, pp. 483—95.

2003b

‘“Towards a Reasonable Libertarianism’, a corrected and extended version

of (1973), in Gary Watsoed, Free Will, third edition University Press).

(Oxford: Oxford

2004a

‘Reflections on Enquiry and Truth arising from Peirce’s Method for the Fixation of Belief’ in Cheryl Misak (ed.), The Cambridge Companion to Peirce (Cambridge and New York: Cambridge University Press). (Misprints and the like will be corrected in a collection forthcoming from Oxford University Press.)

2004b

‘Aurel Kolnai on Utopia’ in Francis Dunlop and Zoltin Balazs (eds), Exploring the World of Human Practice: Readings in and about the Philosophy of A. I Kolnai (Budapest: Central European University Press), pp- 219—30.

2004C

“That which is inherently practical: some brief reflections’ in Fatima Evora, Paulo Faria, Andrea Loparic, Luiz Henrique Lopez dos Santos, and Marco Zingano (eds), Logica e Ontologia: Ensaios em Homenagem a Balthazar

Barbosa Filho (Sio Paulo: Discurso Editorial), pp. 461—71. 2004d

‘Reply to Shoemaker’ and ‘Reply to Shoemaker’s Reply’, The Monist, 87:4, pPp- $94—609 and 614—15.

2004e€

“Wittgenstein on Ethics and the Riddle of Life’, Philosophy, 79, pp. 363—91.

2004f

‘Neo-Aristotelian Reflections on Justice’, Mind, 113:451, pp. 477—512.

2005a

‘Objectivity

in

Ethics; Two

Difhiculties, Two

Responses’,

Ratio,

18:1,

pp. 1—27. 2005b

‘An Idea we cannot do without: what difference will it make (e.g. to moral, political, and environmental philosophy) to recognize and put to use a substantial conception of need?’ in Soran Reader (ed.), The Philosophy of Need (Cambridge: Cambridge University Press).

2005§C

‘Précis of Sameness and Substance Renewed’ and ‘Replies (to Book Symposium contributors)’, Philosophy and Phenomenological Research, 71:2, pp- 442—8 and 470—6.

2006—7

‘Three Moments in the Theory of Definition or Analysis: Its Possibility, its Aim or Aims, and its Limit or Terminus’, Proceedings of the Aristotelian Society, 107, pp- 73—109.

2008a

‘Deux conceptions d’identité’ in Edgardo D. Carosella, Bertrand SaintSernin, Philippe Capella, and S. E. Marcelo Sancho Sorondo (eds), L’identit¢ changeante de I’individu: la constante construction du Soi (Paris: LU’Harmatton), pp. 17—28.

184 2008b

BIBLIOGRAPHY

‘Deliberagio e Razio Prictica’ in Marco Zingano (ed.), Sobre a Etica Nicomagqueia de Aristoteles (Odysseus), pp. 155—82. (A translation from a newly revised English version of 19756, ‘Deliberation and Practical Reason’.)

2008—9

‘Solidarity and the Root of the Ethical’, Lindley Lecture published by the Philosophy Department, University of Kansas: then corrected and expanded in a version published in Tijdschrift voor Filosofie, 71, pp. 239—69.

2000a

“What is the Order among the Varieties of Goodness? A Question posed by von Wright and a Conjecture made by Aristotle’, Philosophy, 84:2, pp. 175—200.

2009b

‘Knowing How To and Knowing That’ in Hans-Johann Glock and John Hyman (eds), Wittgenstein and Analytic Philosophy: Essays for P M. S. Hacker (Oxford: Oxford University Press).

2010

‘The Paradox of Analysis and the Paradox of Synonymy’ in Gilles-Maurice de Schryver (ed.), A Way with Words: Recent Advances in Lexical Theory and Analysis. A Festschrift for Patrick Hanks (Kampala: Menha Publishers).

20I1a

2011b

20I2a

2012b

‘Platonism and the Argument from Causality’ in Grazer Philosophischen Studien, 82, pp. 315—28. Reprinted in Benjamin Schneider and Mortiz Schultz (eds), Themes from Early Analytic Philosophy (Amsterdam, New York: Rodopi). ‘A Reasonable Frugality’ in A. O’Hear (ed.), Philosophy Environment, Royal Institute of Philosophy Supplement, (Cambridge: Cambridge University Press), pp. 175—200.

and vol.

the 69

‘Identity, Individuation and Substance’, European Journal of Philosophy, 20:1, pp. I—25. ‘Practical Knowledge: Knowing

How To and Knowing That’, Mind,

121:481, pp. 97—130. 20I12C

‘From A Humanities Perspective: New Public Management and the Universities” in Jenny Manson (ed.), Public Service on the Brink (Exeter: Imprint Academic).

2013

‘Truth, Pragmatism and Morality’, Philosophy, 88:3, pp. 351—68.

2013—14

‘Hilary Putnam’s Philosophy of Morals and Values’, an extended version of 2013, with a reply by Putnam, in M. Frauchiger (ed.), Themes from Putnam, Lauener Library of Analytical Philosophy (edsW. K. Essler and M. Frauchiger), vol. 5 (Berlin, Boston, MA, Beijing: de Gruyter).

2014

‘Work, its moral meaning or import’, Philosophy, 210:4, and in Virginia Mantouvalou (ed.), The Right to Work: Legal and Philosophical Perspectives (London: Hart Publishing). OTHER

WRITINGS

REFERRED

TO

Altham,]. E. J., and Tennant, Neil W.,‘Sortal Quantification’in Edward L. Keenan (ed.),

Formal Semantics of Natural Language (Cambridge: Cambridge University Press, 1975). Amis, Kingsley, ‘Getting it Wrong’ in Leonard Michaels and Christopher Ricks (eds), The State of the Language (Berkeley, CA: University of California Press, 1980).

BIBLIOGRAPHY Anscombe,

G.

E.

M.,

An

Introduction

185

to Wittgenstein’s Tractatus: Themes

in

the

Philosophy of Wittgenstein (London: Hutchinson University Library, 1959). Austin,J. L., Sense and Sensibilia, ed. G. J. Warnock

(Oxford & New York: Oxford

University Press, 1962). Cavell, Stanley, The Claim of Reason: Wittgenstein, Skepticism, Morality, and Tragedy (New York: Oxford University Press, 1979).

Chomsky, Noam, ‘Explaining Language Use’, Philosophical Topics, 20:1 (1992), pp. 205—31. Church, Alonzo, Introduction to Mathematical Logic (Princeton, NJ: Princeton University Press, 19506). Davidson, Donald, ‘The Logical Form

of Action

Sentences’ in Nicholas Rescher

(ed.), The Logic of Decision and Action (Pittsburgh, PA: University of Pittsburgh Press, 1967), pp- 81-95. Davidson, Davidson,

Donald, “Truth and Meaning’, Synthese, 17:3 (1967), pp. 304—23. Donald, ‘On Saying That’, Synthese, 19:1—2 (1968), pp. 130—46.

Davidson, Donald, Essays on Actions and Events (Oxford: Clarendon Press, 1970). Davidson, Donald, ‘In Defence of Convention T’ in Hugues Syntax and Modality (Amsterdam: North-Holland, 1973).

Leblanc

(ed.), Truth,

Davidson, Donald, ‘Semantics for Natural Languages’, 1970, reprinted in Donald Davidson, Inquities into Truth and Interpretation (Oxford: Oxford University Press, 2001). Dudman,Victor 1976 ‘Bedeutung for Predicates’ in Matthias Schirn (ed.) Studien zu Frege, ii1 (Stuttgart: Frommann Holzboog, 1976). Dummett, Michael, “Truth’, Proceedings of the Aristotelian Society, §9:1 (1959), pp. 141—62. Evans, Gareth, The Varieties of Reference, ed. John McDowell (Oxford: Clarendon Press, 1982).

Evans, Gareth, ‘Semantic Structure and Logical Form’ in Gareth Evans and John McDowell (eds), Truth and Meaning (Oxtord: Oxford University Press, 1976), pp. 49—75. Foster, John, ‘Meaning and Truth Theory’ in Gareth Evans and John McDowell (eds), Truth and Meaning (Oxford: Oxford University Press, 1976). Frege, Gottlob, ‘On Sense and Reference’ [‘Uber Sinn und Bedeutung’], Zeitschrift fiir Philosophie und philosophische Kritik, 100 (1892), pp. 25—50. Frege, Gottlob, Die Grundlagen der Arithmetik, eine logisch-mathematische Untersuchung tiber den Begriff der Zahl (Hamburg: E Meiner, 1884); trans. J. L. Austin as Foundations of Arithmetic (Oxford: Blackwell, 1959).

Frege, Gottlob, ‘Uber die Grundlagen der Geometrie’, Jahresbericht der Deutschen Mathematiker Vereinigung, 12 (1903), 319—24, 368—75; trans. E.-H. W. Kluge in On

the Foundations of Geometry and Formal Theories of Arithmetic (New Haven, CT:Yale University Press, 1971). Frege, Gottlob, ‘Compound Thoughts’, (1923), in Brian McGuinness (ed.), Collected

Papers on Mathematics, Logic and Philosophy (Oxford: Blackwell, 1984). Geach, Peter Thomas, Reference and Generality:An Examination of some Medieval and Modern Theories (Ithaca, NY and London: Cornell University Press, 1962). Geach, Peter, and Black, M. (eds), Translations from the Philosophical Writings of Gottlob Frege (Oxford: Basil Blackwell, 1952).

186

BIBLIOGRAPHY

Gildersleeve, Basil L., and Lodge, Gonzalez, (London: Macmillan, 1895).

Gildersleeve’s Latin Grammar, 3rd edn.

Gillespie, Charles (ed.), The Dictionary of Scientific Biography, 16 vols (New York: Charles Scribner’s Sons, 1973), vii, ‘Le Verrier’. Goodwin,William W., A Greek Grammar (London and New York: Macmillan, 1894). Haack, Susan, ‘ “We Pragmatists...” Peirce and Rorty in Conversation’, Partisan Review, 64:1 (1997), pp. 91—107.

Hahn, Lewis Edwin (ed.), The Philosophy of Donald Davidson (Chicago, IL: Open Court, 1999), The Library of Living Philosophers, Volume XXVII. Hare, R. M., Freedom and Reason (Oxford: Oxford University Press, 1963). Heck, Richard G., ‘Idiolects’ in Judith Jarvis Thompson and Alex Byrne (eds), Content and Modality: Themes from the Philosophy of Robert Stalnaker (Oxford: Oxford University Press, 2006). Hempel, Carl,‘Studies in the Logic of Confirmation’, Mind, §4:213 (1945), pp. 1—26. Hermes, H., E Kambartel, and E Kaulbach (eds), Gottlob Frege: Posthumous Writings,

trans. Peter Long and Roger White (Oxford: Basil Blackwell, 1979). Hookway, Christopher, Peirce (London: Routledge and Kegan Paul, 1985). Hornsby, Jennifer, Actions (London: Routledge and Kegan Paul, 1980). Hornsby, Jennifer, ‘Actions and Abilities’ in R. Haller and W. Grassl (eds), Language, Logic and Philosophy: Proceedings of the 4th International Wittgenstein Symposium, 25th August to 2nd September 1979, Kirchberg/Wechsel,

Austria (Vienna: Holder-Pichler-

Tempsky, 1980), pp. 387—91. Hornsby, Jennifer, ‘Illocution and its Significance’ in S. L. Tsohatzidis (ed.), Foundations of Speech Act Theory: Philosophical and Linguistic Perspectives (London: Routledge, 1995), pp. 187—207.

Johnson,W. E., Logic, Part 1 (Cambridge: Cambridge University Press, 1921). Kneale,W. and G. E. Moore,‘Is Existence a Predicate?’, Proceedings of the Aristotelian Society, Supplementary Volume XV (1936), pp. 154—88. Kripke, Saul A., Naming and Necessity (Cambridge, MA: Harvard University Press, 1972). Leibniz, G. W, ‘Calculus Consequentiarum’ in Louis Couturat (ed.) Opuscules et fragments inedits de Leibniz (Paris, 1903).

Marcus, Ruth Barcan, ‘Modalities and Intensional Languages’, Synthese, 13:4 (1961), pp- 303—22. McDowell, John, ‘On the Sense and Reference of a Proper Name’, Mind, 86:342

(1977), pp- 159-8s. McGee, Van, ‘Semantic Paradoxes and Theories of Truth’ in Edward Craig (ed.), Routledge Encyclopedia of Philosophy (London and New York: Routledge, 1998), viii, pp- 642—8.

Migotti, Mark, ‘Peirce’s Double-Aspect Theory of Truth’, Canadian Journal of Philosophy, 28:Supp.1 (1998), pp. 75—108. Misak, C. J., Truth and the End of Inquiry: A Peircean Account of Truth (Oxtford: Clarendon Press, 1991).

Misak, C. J., Verificationism: Its History and Prospects (London: Routledge, 1995).

BIBLIOGRAPHY Moore,

187

Adrian, ‘On There Being Nothing Else to Think, or Want, or Do’ in Sabina

Lovibond and S. G. Williams (eds), Essays for David Wiggins: Identity, Truth, and Value (Oxford, Basil Blackwell, 1996).

Putnam, Hilary,‘Is Semantics Possible?’, Metaphilosophy, 1:3 (1970), pp. 187—201. Putnam, Hilary, ‘Pragmatism’, Proceedings of the Aristotelian Society, 95 (1995), pp. 272—3060.

Quine W. V., The Ways of Paradox and Other Essays (New York: Random House, 1966). Ramsey, E P, The Foundations of Mathematics and Other Logical Essays (London: Routledge and Kegan Paul, 1931). Ramsey, E P., On Tiuth: Original Manuscript Materials

(1927—1929) From

the Ramsey

Collection at the University of Pittsburgh, ed. N. Rescher and D. Majer (Dordrecht, Netherland: Springer, 1990). Reichenbach, Hans, Elements of Symbolic Logic (London: Dover Publications, 1947). Rescher, Nicholas, ‘Plurality Quantification’, Journal of Symbolic Logic, 27:3 (1962), Pp. 372—4. Rumfitt, Ian, The Boundary Stones of Thought: An Essay in the Philosophy of Logic (Oxford: Oxford University Press, 2015). Russell, Bertrand, Introduction to Mathematical Philosophy (London: George Allen & Unwin, 1919).

Sheard, Michael, ‘A Guide to Truth Predicates in the Modern Era’, The Journal of Symbolic Logic, 59:3 (1994), pp- 1032—54-.

Short, T. L., ‘Peirce on the Aim of Inquiry: Another Reading of “Fixation™’, Transactions of the Charles S. Peirce Society, 36:1 (2000), pp. 1—23. Skagestad, Peter, The Road of Inquiry: Charles Peirce’s Pragmatic Realism (New York: Columbia University Press, 1981).

Sober, Elliott, ‘Evolutionary Theory

and the Ontological

Status of Properties’,

Philosophical Studies, 40:2 (1981), pp. 147—76.

Sonnenschein, E. A., The Soul of Grammar (Cambridge: Cambridge University Press, 1927).

Strawson, P. E, Individuals (London: Methuen, 1959).

Tarski, Alfred, “The Concept of Truth in Formalized Languages’ in A. Tarski, Logi, Semantics, and Metamathematics, trans.J. H. Woodger (Oxford: Oxford University Press, 1950).

Westphal, Jonathan, Colour: A Philosophical Introduction (Oxford: Blackwell, 1993).

Wittgenstein, Ludwig, Tractatus Logico-Philosophicus (London: Routledge and Kegan Paul, 1922).

Wright, Crispin, ‘Theories of Meaning and Speaker’s Knowledge’, in his Realism, Meaning, and Truth (Oxford: Blackwell, 1987).

Index

abduction

xiii, 130, 132—3, 1425,

act/action distinction actions

76

0S, 74,74 n. 14

acts/do-ables 65—6, 65 n. 3 adequacy 20, 20 n. 26, 81—2, 967, 167 Adverbial theory (Strawson) 64, 67—9, 68 n. 8, 71, 74,79

adverbs

68,68 n. 8,69, 69 n. 9, 78

agency

04—0, 60 n. 4, 08,78, 149

Aitchison, Jean

111—13, 115—17

alea jacta est 10—11 analysandum

83, 85, 88—9, 93, 106

analysis paradox xii, 81—4 Anscombe, G.E.M 33, 56 anti-realism 156—7 a priori method 92, 132, 136 Aristotle 64—s, 82,91—2,91 n. 18, 96, 100, 101

artificial language

3

Augustine 140, 140 n. 7 Austin,J. L. 11-12, 11 n. 10, 13 n. 12, 54,

67,75, 77

Chomsky, Noam xii—xiii, 118-19, 118 n. 1, 122—7, 127 n. 9

Church, Alonzo x, 32, 37 clarity, of ideas (Leibniz) 81-82, 93—0, 107, 158—9 cognition 95—8, 100, 142

coherence 165 colours 70—-1 common-sense view 118—22, 120, 126 n. 8 conception 82, 86, 90, 93—0, 157

concepts: different attitudes towards 126 first-level 39, 47-8, 54, 57, 66 and Frege 32, 36, 39,806,806 n. 7 of individuals §6—7 and logic 133 97, 101—7

second-level

134

39, 47-8, 54

concept-writing, see Begriffsschrift

authority, the method of

conduit

132, 135—6

111—17

connective analysis 93

autonomy theory 122—3 Axiom of Regularity 98 Ayer,A. .

‘caretaker speech’ 112 Carey, John 117 Carnap, Rudolph 2, 21, 32

primitive

authoritarianism (Peirce)

Brouwer, L. E. J. 13

Burnyeat, Myles 91,91 n. 18

145 n. 9, 154, 161—3

conservatism (in linguistics) contingency $2—03

144, 168—90

ConventionT

1668, 166 n. 6, 167 n. 8, 170

Bedeutung (reference) 32, 34-s5, 39, 85—9, 86 n. 8, 906, 123; see also Frege

copula 1x-X, 31—41, 38 n. 17 cupellation 104, 104 n. 33

Begriff 86, 88 n. 10; see also concepts;

Dante

Frege

Begriffsschrift 3—4, 7—10, 14, 245, 86

113—14

127

Davidson, Donald: idea of truth 174

belief 132, 136—41, 147-8 bivalence 156

modes of grammatical combination

Bouvard,

semantic interpretation 42—5I, 43 n. I truth conditions 1-2, 0, 14, 17, 21—3

Alexis

04—9, 05 n. 2,71—9, 73 n. 12,74 n. 14

§2

bread, nourishment by

159—63

190

INDEX

decomposition 107, 131 deduction 132, 142, 144, 140, 163 definition: aims

82—93,91

n. 17,92 nn.

‘fruitful definition’ ‘F then G’ 47-8

Fundierung 98

19—20

Galle,]. G. 53, 58—9, 61 Geach,P. T. 169 gerunds 171-2, 171 n. 14 God 97, 140, 140 n. 7

and concept 94, 96 deflationism xiii, 165—6, 169

descriptivism

86

116—17

designation 18, 45, 84, 84 n. 2, 103 dialects 112, 119—21, 128

Goodman, Henry Nelson

146

dictionaries 88, 90 disquiet 132, 135—7, 140 disquotation 167, 170 distinctness 82, 93—7, 106 ‘doable’ (Hornsby) 76—7, 77 n. 19

grammar 43, 49—5I, 79, 80 n. 20, 101 Grandy, Richard x, 23 Grice, H. P. 1x—x, 13, 21

dogmatism/tenacity (the method of,

‘homophony’ 48, 51 Hookway, Christopher 133 Hornsby, Jennifer xi, 64, 76—7, 79 ‘horse’ (the concept horse, Frege) 33—,

Peirce)

Hempel, Carl

132—s5, 136

double-negation elimination Dummett, Michael

156

14, 14 n. 14, 20, 32,

38—9, 39 n. 18

34, 150, 170

elucidation

g, 16, 24—7, 106, 106 n. 37,

155, 15§ n. 18, 165, 165 n. 3

English language

112-14, 116—17, 118—23,

114

Epimenides’ paradox

18

Evans, Gareth s0, so n. 11, 60—1 evolution/evolutionary theory 39—40 exemplification 54, 103 existence

‘How to Make Our Ideas Clear’ (Peirce) 131 human beings 96—7, 102, 1201, 126—8, 170

‘human grooming talk’ 112

125, 1278

entropy

146

27-8, §2—03, 60, 70—1, 80, 90,

107, 119—20 existential generalization (EG)

58—9, 63

Hume, David 131, 15963 Husserl, Edmund x—xii, 32—3, 37, 81—7,91,93

hypothesis

142, 144—6

idiolect 121, 123 Immortal 47-8

‘improper symbol’ (Church)

x, 37,

37 n. 10

fixation 1301, 147, 152 Frege, G.: account of marks 103 and Bedeutung 125 concepts

induction 144, 146, 159, 159 n. 2§, 162 inference 17, 43, S0, $8, 130, 132, 142,

1445, 163 instantiation $4—