The secret connexion: causation, realism, and David Hume

OUP CORRECTED PROOF – FINAL, 12/3/2014, SPi

The Secret Connexion

OUP CORRECTED PROOF – FINAL, 12/3/2014, SPi

The Secret Connexion

Causation, Realism, and David Hume

REVISED EDITION

Galen Strawson

I am not such a sceptic as you may, perhaps, imagine.

(Hume, letter to Stewart, February 1754)

1

OUP CORRECTED PROOF – FINAL, 12/3/2014, SPi

3

Great Clarendon Street, Oxford, OX2 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

# in this edition Galen Strawson 2014

The moral rights of the author have been asserted

First published 1989

First issued in paperback with corrections 1992

This edition published 2014

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: 2014933805

ISBN 978–0–19–960584–2 (Hbk.)

978–0–19–960585–9 (Pbk.)

As printed and bound by

CPI Group (UK) Ltd, Croydon, CR0 4YY

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.

OUP CORRECTED PROOF – FINAL, 12/3/2014, SPi

[N]ature has kept us at a great distance from all her secrets, and has afforded us

only the knowledge of a few superficial qualities of objects; while she conceals

from us those powers and principles on which the influence of these objects

entirely depends.

[W]e are ignorant of those powers and forces, on which [the] regular course and

succession of objects totally depends.

[E]xperience only teaches us, how one event constantly follows another; without

instructing us in the secret connexion, which binds them together, and renders

them inseparable.

(Enquiry 32–3/4.16, 55/5.22, 66/7.13)

OUP CORRECTED PROOF – FINAL, 12/3/2014, SPi

OUP CORRECTED PROOF – FINAL, 12/3/2014, SPi

Preface to the First Edition

Does Hume endorse a regularity theory of causation? The view that he does is very

widely held—it seems that it is still the standard view. But there seems to be no

evidence for it in the first Enquiry—whatever one thinks about the Treatise.

My doubts began when I started teaching the first Enquiry at an elementary level at

Oxford in 1980. This discussion of Hume grew directly from that teaching. I finished

the first draft convinced that I was heroically alone in my doubts, but I was quite

wrong. Three works deserve particular mention. Craig’s The Mind of God and the

Works of Man (1987) refutes the commonly accepted conception of Hume’s philo￾sophical position. Wright, in The Sceptical Realism of David Hume (1983), and

Livingston, in Hume’s Philosophy of Common Life (1984), also argue powerfully

against it. As far as the question of causation is concerned, they argue not only that

Hume is not a regularity theorist, but also, as here, that he believes in the existence of

something like natural necessity—in the existence of causal power conceived of in

some essentially non-regularity-theory way.1

The fact remains: the view that Hume held a regularity theory of causation is still

the standard view.2 It is still worth arguing against. As Beauchamp and Rosenberg

say, ‘[al]most all recent writers on causation believe that Hume holds a pure

regularity theory of causation’ (1981: 31). As Blackburn remarks, Hume has ‘been

shamefully abused by many commentators and their victims’ (1984: 211).

This is not just a book about Hume. It also argues directly against the regularity

theory of causation, which has taken on a life of its own. Chapters 5, 8, and 22.2 form

a discussion of the regularity theory of causation which is independent of the material

on Hume. Chapter 23, on the meaning of ‘cause’, is independent both of this

discussion of the regularity theory of causation and of the main discussion of

Hume. Chapter 7 on the notion of the ultimate nature of reality, together with its

Appendix, can also be read independently of the discussion of Hume.

1 Throughout this book I use the expression ‘causal power’ to mean ‘causal power conceived of in some

essentially non-regularity-theory way’. 2 For recent expressions of the view, see e.g. Woolhouse: ‘Hume’s conclusion [is] that so far as the

external objects which are causes and effects are concerned there is only constant conjunction’ (1988: 149);

so far as the ‘operations of natural bodies’ are concerned, ‘regularity and constant conjunction are all that

exist’ (p. 150). See also O’Hear: Humeans hold that there is not ‘any more to causality than ‘regularity of

succession’’ (1985: 60). ‘The Humean attitude to causality is . . . that there is nothing in the cause . . . that

mean[s] the effect has to follow’ (p. 61). See also Kripke: ‘If Hume is right, . . . even if God were to look at

[two causally related] events, he would discern nothing relating them other than that one succeeds the

other’ (1982: 67). Hamlyn, by contrast, scrupulously avoids this error, in his History of Western Philosophy

(1987). So does Schacht (who restricts his discussion to the Enquiry) in his Classical Modern Philosophers:

Descartes to Kant (1984).

OUP CORRECTED PROOF – FINAL, 12/3/2014, SPi

I also discuss a number of other questions about meaning, understanding, know￾ledge, and existence. These are questions which are naturally and vividly raised by the

study of Hume—questions about the relations between semantics, epistemology, and

ontology and metaphysics. I defend—where I do not assume—a straightforwardly

realist position on these issues (see especially Chapters 7 and 12). Virtually all the

worthwhile features of the current debate about them are present in the seventeenth￾and eighteenth-century debate.3

Someone who believed that the regularity theory of causation was true, and came

to the conclusion that Hume never held it, would probably not much enjoy writing a

book that tried to show that this was so. But if one thinks that the regularity theory of

causation is (in its standard version) metaphysically fantastic one sees things differ￾ently. One’s task is to clear a great philosopher of a damaging charge, and to show

that his attitude to the question of causation is in most respects right. This is how

I see it.

The foundations of the case for saying that Hume believes in causal power or

natural necessity can be established without any close consideration of what he says

about causation, and I try to do this (among other things) in Part 1. One problem

with this approach is that the more detailed analyses of Hume’s discussion of

causation in Parts 2 and 3 (which consider the Treatise and the Enquiry respectively)

may then come as an anticlimax. I have accordingly signalled short cuts at various

points throughout the book for those who want to get on to Part 2 or Part 3 quickly.

A period of illness in the spring of 1984 gave me time to reread Hume and organize

the quotations used in this book. The main text was written up in the spring and

summer of 1987, and for comments on what I wrongly took to be the final version

I am most grateful to John Cottingham, Edward Craig, William Jordan, Paul

Snowdon, and P. F. Strawson. For some helpful conversations, and for contributions

to a class I gave on Hume at Oxford in the Hilary Term of 1988, I would like to thank

Thomas Baldwin, Bob Hargrave, Dan Isaacson, Martha Klein, John Roe, Helen

Steward, Luigi Turco, and David Womersley. For comments of a non-philosophical

kind I am grateful to Angela Blackburn, Redmond O’Hanlon, Andrew Rosenheim,

and finally to Jose´ Strawson.

3 This book contains relatively little discussion of the intellectual context in which Hume wrote (Kant,

rather than any of Hume’s contemporaries or immediate predecessors, provides the other main source of

historical reference). For a good account of Malebranche’s influence on Hume’s thought about causation,

see McCracken 1983: 257–69.

OUP CORRECTED PROOF – FINAL, 12/3/2014, SPi

viii PREFACE TO THE FIRST EDITION

Preface to the Paperback Edition

I wrote this book very fast. The speed of writing shows in the tone and in the

excessively numerous and sometimes intrusive footnotes. I have not, however,

done any major rewriting for this paperback edition. Instead I have made a few

corrections, a few small cuts, and a few small additions. The most substantial changes

occur in the paragraph on pp. 56–7, on page 201 in note 5, on page 203 which has a

new note 8, and on page 210 which has a new note 8. (These page references are to

the first edition.)

As remarked in the original Preface, Chapters 5, 8, and 22.2 of this book present an

attack on the ‘regularity’ theory of causation which is independent of the discussion

of Hume. One objection to the attack is so common that it is worth mentioning here.

Briefly, it claims that one must in the end admit that the regularity of the world’s

behaviour is a ‘brute’ fact, and hence admit that the regularity theory of causation is

true. For if one tries to deny that the regularity of the world is a brute fact, and

proposes that there is something x which is other than the regularity of the world,

and is the reason for the regularity of the world, then one will need an account of why

x is itself regular in its underwriting of the regularity of the world. And either one will

have to say that x’s regularity of operation is itself a brute fact, or one will have to

embark on a regress which can be stopped only by saying that there is something

whose regularity of operation is a brute fact. Hence the regularity theory of

causation—or something like it—must be true.

I tried to answer this sort of objection on pp. 91–2 of this book. I’m not sure what

‘brute’ means, but I’m not worried by the brutishness of the regularity of the world so

long as it does not have the consequence that the regularity of the world is, as it

continues from moment to moment, and from year to year, a continuous fluke or

chance matter. With hindsight, I think that the best statement of the correct

alternative to the regularity theory that the book contains is on pp. 225–6 below).

I made a third brief attempt to say what is wrong with the regularity theory in a paper

called ‘The Contingent Reality of Natural Necessity’ [1991; included here as an

Appendix to Chapter 22].

The 1996 and 2003 paperback impressions incorporate a few minor corrections, a

more substantial change to note 30 on page 50, and a new Index of Passages from Hume.

OUP CORRECTED PROOF – FINAL, 12/3/2014, SPi

Preface to the Revised Edition

In this new edition of The Secret Connexion I’ve made almost no substantive

alterations to the book’s original content, because I haven’t read anything that has

given me reason to doubt its main arguments and conclusions. I have, however,

thinned the text throughout, reworked Chapters 14 and 15 on the Treatise, and added

quite a number of further thoughts and quotations. The original Appendix A is now

an appendix to Chapter 6. Appendix B is now an appendix to Chapter 7 (reproduced

essentially unchanged, although it later grew into a separate paper (Strawson 2002)).

The original Appendix C, a summary of the overall position, has been dropped. It

was superseded in 2000 by a paper, ‘David Hume: Objects and Power’, from which

I’ve here incorporated some considerations about the relative importance of the

Treatise and the Enquiry in assessing Hume’s considered view (pp. 10–13). I’ve also

included as an appendix to Chapter 22 a reply to a helpful objection to The Secret

Connexion raised by Nicholas Everitt in 1991.

Many now talk of ‘Old Hume’ as opposed to ‘New Hume’.

1 According to the

supporters of Old Hume, Hume holds that ‘even if God were to look at [two causally

related] events, he would discern nothing relating them other than that one succeeds

the other’.

2 On this view, Hume holds a metaphysical regularity theory of causation,

according to which things in reality succeed each other in a regular fashion, but

nothing ever really causally influences anything else in any way at all. All supporters of

New Hume deny that he holds this view. Some go further and follow John Wright, as

I do, in holding that Hume is a ‘sceptical realist’.

The question of what it means to say that Hume is a realist about causal power is

part of the subject of this book, but one thing that is clear is that to be a realist about

causal power is to hold that there really is such a thing as causal influence. That is,

there is something more in reality, causally speaking, than one thing’s just following

another, even if all we can ever actually empirically detect, strictly speaking, is one

thing regularly following another.

According to the sceptical realist view of Hume, he’s not only a sceptical realist

about causal power, but also about physical objects like tables and chairs. He’s not

concerned to deny categorically that things that we ordinarily suppose to exist don’t

exist. Real sceptics don’t do that. His general, sceptical, empiricist, philosophical

1 The terms derive from Kenneth Winkler’s paper ‘The New Hume’ (1991). 2 This is Kripke’s accurate report (1982: 67) of the view of Hume that prevailed thirty years ago. Jackson

in 1977 invites us to ‘consider . . . the possible world where every particular fact is as it is in our world, but

there are no causes or effects at all. Every regular conjunction is an accidental one, not a causal one. Call it,

for obvious reasons, the Hume world’ (1977: 5). See also p. vii n.2, p. 13 n.8, and p. 22 n.2. Some who take

themselves to be supporters of ‘old Hume’ now deny that this used to be the standard view.

OUP CORRECTED PROOF – FINAL, 12/3/2014, SPi

point is simply that we can form no empirically respectable, clear, and distinct

conception of the ‘real nature’ of concrete reality (other than the concrete reality

that consists in the existence of our experiences). We can’t hope to have insight into

its ‘real nature and operations’ (63, 638/1.2.5.25, 1.2.5.26), its ‘ultimate original

qualities’ (xvii/Int}8), its ‘ultimate principles’ (xviii/Int}10). We can’t hope to have

insight into the ‘internal structure or operating principle of objects’ (169/1.3.14.29),

‘the nature of bodies’ (64/1.2.5.26), ‘the essence and construction of bodies’ (660/

Abs}32). And ‘the essence of the mind [is] equally unknown to us with that of

external bodies’ (xvii/Int}8).

The terms ‘Old’ and ‘New’ are unfortunate, for ‘New Hume’ is simply Hume—but

also because the ‘New Hume’ interpretation isn’t new. As Helen Beebee observes,

Norman Kemp Smith classifies as a supporter of ‘New Hume’ in his 1941 book The

Philosophy of David Hume (Beebee 2006: 173). Kant was equally clear in 1783 that

Hume didn’t deny the existence of causal power (Kant 1783: Preface).

Even if this were not so, the terms would be obsolescent. Don Garrett has recently

argued that so far as the textual evidence is concerned, there’s nothing to favour one

of the views decisively over the other (Garrett 2009). It may be that supporters of the

old Hume are now in the minority (at least outside Canada). It’s hard to be sure, for

many who take themselves to support ‘Old Hume’ have shifted their ground con￾siderably over the last thirty years.

One of the ironies of the debate has been that some ‘Old Hume’ supporters have

suggested that ‘New Hume’ supporters, being themselves realists about things like

objects and causation, have been motivated to interpret Hume in such a way that he

agrees with them. The irony is that almost all those who have argued in this way have

tended to be ‘anti-realists’ of some stripe or other, determined to maintain an

interpretation of Hume that agrees with their anti-realism.

‘Old Humeans’ have also charged that ‘New Humeans’ (like myself ) rely on

isolated quotations to support their view. Again this seems back to front, although

it’s true that the three quotations on p. v that form the epigraph to this book suffice,

in effect, to establish the ‘New Hume’ case. The ‘Old Hume’ interpretation relies

principally on overliteral readings of a relatively small number of passages from

the Treatise, whose publication Hume later regretted (see p. 10); the deep drift of

Hume’s epistemology and metaphysics is ‘New Humean’ to the core.

I’m very grateful to Peter Momtchiloff for encouraging me to produce this new

edition, and to Keith Turausky for reading through the whole text with care and insight.

OUP CORRECTED PROOF – FINAL, 12/3/2014, SPi

PREFACE TO THE REVISED EDITION xi

OUP CORRECTED PROOF – FINAL, 12/3/2014, SPi

Contents

Abbreviations and Conventions xvi

Part 1. Meaning, Scepticism, and Reality

1. Introduction 3

2. The ‘Humean’ view of causation; and an exegetical principle 9

3. A summary of the argument 14

4. ‘Objects’: preliminaries 19

5. The untenability of the realist regularity theory of causation 22

6. ‘Objects’: complications 32

6.1 Strict idealism 32

6.2 Perception-constituted objects and

perception-content-constituted objects 35

6.3 A viable regularity theory of causation 40

6.4 Hume uncommitted 41

6.5 Supposing and conceiving 44

6.6 Basic realism 52

6.7 Bundles and fiction 57

6.8 Hume in metaphysical space 58

6.9 Writing as a realist 59

Appendix Cartoon-film causation: idealism and the regularity theory

of causation 60

7. The notion of the ultimate nature of reality 65

Appendix Reality and truth 75

8. ‘Causation’ 87

9. Hume’s strict scepticism 95

10. Hume’s theory of ideas as applied to the idea of causation 102

11. The ‘AP’ property 108

11.1 The curious idea of a priori causal inference 108

11.2 An objection 110

11.3 The objection varied 112

12. The problem of meaning 115

12.1 The ‘Meaning Tension’ 115

12.2 Experience-transcendent reference: E-intelligibility and

R-intelligibility 120

OUP CORRECTED PROOF – FINAL, 12/3/2014, SPi

12.3 Example: Hume on the mind 123

12.4 Conclusion 125

13. ‘External objects’ and Causation 128

13.1 The parallel 128

13.2 A possible disanalogy 129

13.3 An objection 131

Part 2. Causation in the Treatise

14. Causation in the Treatise: 1 137

14.1 Introduction 137

14.2 Referring uses of Causation terms 138

15. Causation in the Treatise: 2 142

15.1 Three stratagems 142

15.2 Ignorance, irony, and reality 144

15.3 Hume’s global subjectivism about necessity 147

15.4 The ‘necessity, which we ascribe’; the ‘necessity, which

we conceive’ 150

15.5 ‘So far as we have any notion of it’ 153

15.6 Conclusion 158

Part 3. Causation in the Enquiry

16. Enquiry Section 4: the question of irony 165

17. Enquiry Section 4: Causation and inductive scepticism 169

18. Enquiry Sections 5–6: undiscovered and undiscoverable 171

19. Enquiry Section 7: Causation and human beings 176

19.1 Will and force: a last look at irony 176

19.2 Resemblance, solidity, and force 180

19.3 A rhetorical question 181

20. Enquiry Section 7: the Occasionalists 183

21. Enquiry Section 7: the two definitions of cause 188

21.1 Extraordinary ignorance 188

21.2 The two definitions 190

21.3 Conclusion 197

Part 4. Reason, Reality, and Regularity

22. Reason, Reality, and Regularity 201

22.1 A summary of Hume’s position 201

22.2 The general form of the argument for Causation 203

Appendix The Contingent Reality of Natural Necessity 210

OUP CORRECTED PROOF – FINAL, 12/3/2014, SPi

xiv CONTENTS

23. The meaning of ‘cause’ 215

23.1 Content: experience and concepts 215

23.2 The ‘Anscombean’ approach 220

23.3 The wisdom of nature 223

23.4 Causation: a non-sensory property 230

References 236

Index 241

Index of Passages from Hume 244

OUP CORRECTED PROOF – FINAL, 12/3/2014, SPi

CONTENTS xv