Robert J. Fogelin-Hume's Skeptical Crisis_ a Textual Study-Oxford University Press, USA (2009)

8/18/2019 Robert J. Fogelin-Hume's Skeptical Crisis_ a Textual Study-Oxford University Press, USA (2009)

1/193


2/193

Hume’s Skeptical CrisisZ


3/193

This page intentionally left blank


4/193


A Textual Study

R O B E R T J . F O G E L I N

12009


5/193

1Oxford University Press, Inc., publishes works that furtherOxford University’s objective of excellence

in research, scholarship, and education.

Oxford New York

Auckland Cape Town Dar es Salaam Hong Kong Karachi

Kuala Lumpur Madrid Melbourne Mexico City Nairobi

New Delhi Shanghai Taipei Toronto

With offices in

Argentina Austria Brazil Chile Czech Republic France Greece

Guatemala Hungary Italy Japan Poland Portugal Singapore

South Korea Switzerland Thailand Turkey Ukraine Vietnam

Copyright © 2009 by Oxford University Press, Inc.

Published by Oxford University Press, Inc.

198 Madison Avenue, New York, New York 10016

www.oup.com

Oxford is a registered trademark of Oxford University Press

All rights reserved. No part of this publication may be reproduced,

stored in a retrieval system, or transmitted, in any form or by any means,

electronic, mechanical, photocopying, recording, or otherwise,

without the prior permission of Oxford University Press.

Library of Congress Cataloging-in-Publication Data Fogelin, Robert J.

Hume’s skeptical crisis : a textual study / Robert J. Fogelin.

p. cm.

ISBN 978-0-19-538739-1

1. Hume, David, 1711–1776. Skepticism in the treatise of human nature.

2. Skepticism. 3. Knowledge, theory of. I. Title.

B1489.F638 2009

128—dc22 2008049463

9 8 7 6 5 4 3 2 1

Printed in the United States of America

on acid-free paper

http://www.oup.com/http://www.oup.com/


6/193

For Jane Lincoln Taylor


7/193



8/193

But what have I here said . . . ?


9/193



10/193

PrefaceZ

A few words about the provenance of this work may help explainthe form it takes. It concerns the same material that I examinedin the first eight chapters of Hume’s Skepticism in the Treatise ofHuman Nature, published in 1985. The primary intention of that work was to counter the tendency, common at the time, to playdown or simply ignore the skeptical themes in Hume’s Treatise infavor of a thoroughgoing naturalistic reading in the style of Norman

Kemp Smith. Since the skeptical themes were being played down,to provide a suitable counterweight, I played them up. This workis intended to offer a more balanced account of the relationshipbetween Hume’s naturalism and his skepticism.

In 1990 I was given the opportunity to develop the central ideas ofthat book as a lecturer in a National Endowment for the Humanitiesseminar directed by David Fate Norton and Wade Robison. It con-sisted of five lectures on book 1, part 4 of the Treatise, the locusof Hume’s deepest skeptical reflections. I wrote careful notes andcorrected them in the light of the energetic discussions that greetedthese lectures. My intention at the time was to use these notes asthe basis for a second edition of Hume’s Skepticism in the Treatise.I then became involved in other projects and thought no more ofthese notes until fifteen years later, when I received an invitationfrom Livia Guimaraes to give a series of lectures on Hume at theUniversidade Federal de Minas Geraise in Brazil. I dug out the


11/193

x Preface

NEH notes, reread Hume, and produced an extensive revision of

the original NEH lectures. This work is based on those lectures.Much has changed as a result of these multiple revisions. I have

corrected what I now take to be misreadings of particular texts.I have given more prominence to the naturalistic themes in theTreatise, in particular, to Hume’s attempts to give a naturalisticaccount of the emergence of philosophical positions. On the otherside, I now lay more stress on the claim that Hume’s pursuit of a

science of human nature itself generates a skeptical challenge thatcalls his naturalistic program into question. Hence the title: Hume’s Skeptical Crisis . Rather than an analysis of Hume’s skepticism ofthe kind presented in Hume’s Skepticism in the Treatise of HumanNature , this work presents a narrative account of how this skepti-cal crisis arises as Hume’s investigations penetrate more and moredeeply into the operations of the human understanding.

Over the years a number of books have played an important partin shaping my thinking about Hume’s philosophy, both throughagreement and disagreement. These include Thomas Brown’s Cause and Effect (1822), J. A. Passmore’s Hume’s Intentions , TerrencePenelhum’s Hume , Barry Stroud’s Hume , Annette Baier’s Progress ofSentiments , and Don Garrett’s Cognition and Commitment in Hume’sPhilosophy . I also have learned a great deal from an unpublishedpaper by David Owen titled “Scepticism with regard to reason.”It is available from the departmental website (philosophy.arizona.edu).

The editorial comments in the Selby-Bigge and Nidditch edi-tions of the Treatise of Human Nature and the two Enquiries havebeen of invaluable help, as have the extensive comments in the newDavid Norton and Mary Norton critical edition of the Treatise andthe new Thomas Beauchamp critical edition of the first Enquiry .


12/193

Preface xi

I owe a special debt to the two scholars who served as readers of

the manuscript for Oxford University Press: first for their generoussupport, second for their editorial comments that at times strength-ened my reading of the text, while at other times saved me fromembarrassing errors.

Beyond this, there are the articles I have read, talks I have attended,comments I have received on papers I presented in various venues,and, not of least importance, conversations that I have had at gather-

ings around the world with fellow Hume enthusiasts. I cannot namethem all, but some who have been particularly useful, often callingtexts to my attention either in support or in opposition to views Ihold, include: Annette Baier, Lewis White Beck, Simon Blackburn, Janet Broughton, Thompson Clarke, Willis Doney, Julia Driver,Harry Frankfurt, Don Garrett, Livia Guimaraes, Ted Honderich,Christopher Hookway, Gary Mathews, Peter Millican, David

Norton, David Owen, Richard Popkin, Geoffrey Sayer-McCord, Walter Sinnott-Armstrong, David Stove, Barry Stroud, Meredith Williams, Michael Williams, and Kenneth Winkler. I have surelyforgotten some, and to them I apologize.

I have received institutional support for this project from theNational Endowment for the Humanities, the Rockefeller StudyCenter at Bellagio, The Liguria Study Center for the Arts andHumanities at Bogliasco, the Faculty Research Fund at DartmouthCollege, and a generous Emeritus Grant from the MellonFoundation.

I would also like to thank Peter Ohlin and Linda Donnelly of theOxford University press for their support and editorial help, and, asbefore, thank Florence Fogelin and Jane Taylor for their patienceand skill in dealing with drafts of my work. In gratitude for her helpover many years, I have dedicated this work to Jane Taylor.


13/193



14/193

ContentsZ

Texts and Citations xvii

Introduction: The Interpretive Problem 3

1 Of Knowledge and Probability 11

A Quick Tour of Part 3, Book 1 11

Section 1 . Of knowledge 11

Section 2. Of probability; and of the idea of cause and effect 13

Section 3. Why a cause is always necessary 14

Section 4. Of the component parts of our reasonings concerning causes and effects 15

Section 5 . Of the impressions of the senses and memory 15

Section 6 . Of the inference from the impression to the idea 16

Section 7 . Of the nature of the idea, or belief 18

Section 8 . Of the causes of belief 19

Section 9. Of the effects of other relations, and other habits 20

Section 10. Of the influence of belief 20

Section 11 . Of the probability of chances 21

Section 12

. Of the probability of causes 21

Section 13. Of unphilosophical probability 22


15/193

xiv Contents

Section 14. Of the idea of necessary connexion 22

Section 15 . Rules by which to judge of causes and effects 27Section 16 . Of the reason of animals 28

2 Hume on Unphilosophical Probabilities 29

Unphilosophical as Opposed to Philosophical Probabilities 29

Sources of Unphilosophical Probabilities 30

The effect of the remoteness of the event 30

The effect of the remoteness of the observation 31

Reiterative diminution 31

Prejudice based on general rules 35

Conflicts within the Imagination and Skepticism 37

3 Hume’s Skepticism with Regard to Reason 39

The Turn to Skepticism 39The Basic Argument 40

Reducing knowledge to probability 41

The regression argument 43

The principle of reiterative diminution 44

Hume’s Response to His Skeptical Argument 48

Peritrope 524 Of Skepticism with Regard to the Senses 55

Hume’s Turnabout with Regard to the Senses 55

The Organization of Section 2 57

The Causes of Our Belief in the Existence of Bodies 59

The senses not the source of this belief 60

Reason not the source of this belief 63The Operations of the Imagination in Forming This Belief 65


16/193

Contents xv

Hume’s informal statement of his position 66

Hume’s systematic statement of his position 69The Philosopher’s Double-Existence. Theory of Perception 78

The Pyrrhonian Moment 82

A Concluding Note 83

5 Of the Ancient and Modern Philosophy 85

Reasons for Examining Ancient and Modern Philosophical Systems 85

Of the Ancient Philosophy (Section 3) 85

The false belief in the continued identity of changing objects 87

The fiction of underlying substance, or original first matter 89

The false belief in the simplicity of objects 90

The fiction of a unifying substance93

The incomprehensibility of the peripatetic system 93

Skeptical implications 94

Of the Modern Philosophy (Section 4) 96

Against the distinction between primary and secondary qualities 97

Another Pyrrhonian moment 99

6 The Soul and the Self 101Of the Immateriality of the Soul (Section 5) 101

Setting the dialectical stage 101

The soul as substance 102

The problem of local conjunction 104

Soul–body interaction 107

On proofs of immortality 108Of Personal Identity (Section 6) 109


17/193

xvi Contents

Basic criticisms 110

Account of the fiction of personal identity 112Disputes about identity as merely verbal 116

The reservations in the appendix 117

7 The Conclusion of Book 1 125

A Gloomy Summation of Skeptical Results 125

What Is to Be Done? 130

Being a Philosopher on Skeptical Principles 132

8 Two Openings and Two Closings 139

The Treatise and the Enquiry on Skepticism 139

The Opening of the Treatise 140

The Opening of the Enquiry 140

The Response to Skepticism in the Enquiry 144The Science of Human Nature in the Enquiry 146

The Role of Skeptical Arguments in the Enquiry 149

Skepticism concerning the senses 149

Skepticism concerning reason 151

Skepticism concerning moral reasoning 152

Pyrrhonism and Mitigated Skepticism 155

Notes 159

Works Cited or Mentioned 165

Index 167


18/193

xvii

Texts and CitationsZ

For the reader’s convenience, all citations to A Treatise of HumanNature and to An Enquiry Concerning Human Understanding havea double entry. The first is to the editions of these works that haverecently been published by Oxford University Press: the DavidFate Norton and Mary J. Norton critical edition of the Treatise of Human Nature and Tom L. Beauchamp’s critical edition of theEnquiry Concerning Human Understanding. The second entry is to

Selby-Bigge/Nidditch editions of the Treatise of Human Nature andDavid Hume’s Enquiries . Citations to the Treatise contain only pagenumbers. Citations to the Enquiry concerning Human Understandingare labeled EHU.


19/193



20/193



21/193



22/193

3

How does Hume’s naturalism—his attempt “to introduce the exper-imental method of reasoning into moral subjects”—square with what seem to be his strong skeptical commitments? On the face ofit, these two aspects of his philosophy appear to be at odds with oneanother. For example, in a number of places Hume holds that thereare no rational grounds for believing that the regularities that haveheld in the past will continue to hold in the future; that in turn

seems to show that the inductive inferences he employs in pursuitof his science of human nature are themselves baseless. This conflictis so obvious that even philosophically unsophisticated readers oftenrecognize it. Hume, however, does not seem particularly concernedabout this apparent conflict between his inductive skepticism and hiscommitment to a science of human nature. In the Treatise of HumanNature , a skeptical argument is used to reject reason (in a wide sense)as the source of causal inferences, so that he can replace it with theassociative operations of the imagination. In the Enquiry ConcerningHuman Understanding , Hume celebrates this relocation, telling usthat “it is . . . conformable to the ordinary wisdom of nature to secureso necessary an act of the mind, by some instinct or mechanical ten-dency” (EHU , 45/55). As we shall see, skeptical doubts are the sourceof deep disquietudes that emerge in the concluding section of thefirst book of the Treatise , but his inductive skepticism—though ithas troubled many others—seems not to have troubled Hume.

IntroductionZ The Interpretive Problem


23/193

4 Introduction

A dramatic way of presenting the tension between Hume’s natu-

ralism and his skepticism is to compare two passages: one from theintroduction to the Treatise , the other from the closing section ofbook 1. Hume launches the Treatise with swagger:

’Tis evident, that all the sciences have a relation, greater orless, to human nature; and that however wide any of themmay seem to run from it, they still return back by one pas-

sage or another. Even Mathematics , Natural Philosophy , andNatural Religion, are in some measure dependent on the sci-ence of MAN; since they lie under the cognizance of men,and are judg’d of by their powers and faculties.

Here then is the only expedient, from which we can hopefor success in our philosophical researches, to leave the tediouslingring method, which we have hitherto follow’d, and insteadof taking now and then a castle or village on the frontier, tomarch up directly to the capital or center of these sciences,to human nature itself; which being once masters of, we mayevery where else hope for an easy victory. From this station wemay extend our conquests over all those sciences, which moreintimately concern human life, and may afterwards proceedat leisure to discover more fully those, which are the objectsof pure curiosity. There is no question of importance, whosedecision is not compriz’d in the science of man; and there isnone, which can be decided with any certainty, before webecome acquainted with that science. In pretending there-fore, to explain the principles of human nature, we in effectpropose a compleat system of the sciences, built on a founda-

tion almost entirely new, and the only one upon which theycan stand with any security. (4/xv)


24/193

Introduction 5

In sharp contrast to this brave beginning, in the closing section

of book 1 of the Treatise we find the following forlorn skepticalplaint:

But what have I here said, that reflections very refin’d andmetaphysical have little or no influence upon us? This opin-ion I can scarce forbear retracting, and condemning from mypresent feeling and experience. The intense view of these man-ifold contradictions and imperfections in human reason hasso wrought upon me, and heated my brain, that I am ready toreject all belief and reasoning, and can look upon no opinioneven as more probable or likely than another. Where am I,or what? From what causes do I derive my existence, and to what condition shall I return? Whose favour shall I court, and whose anger must I dread? What beings surround me? and

on whom have I any influence, or who have any influenceon me? I am confounded with all these questions, and beginto fancy myself in the most deplorable condition imaginable,inviron’d with the deepest darkness, and utterly depriv’d ofthe use of every member and faculty. (175/ 268–69)

This study is an attempt to answer two questions: First, how didHume’s pursuit of his “science of man,” so confidently begun at thestart of the Treatise , land him in philosophical despair? Second, howdoes he attempt to extract himself from this melancholy state?

It is important not to write off Hume’s expressions of despair asmere histrionics. As Hume comes to see, his pursuit of a science ofhuman nature has led him into a skepticism that subverts the entireenterprise. Furthermore—though this will have to be shown—he

finds himself completely incapable of presenting arguments that will refute the skeptical challenges he himself has produced. If he


25/193

6 Introduction

is going to find a way out of his intellectual impasse, it will have to

come from the nonrational side of human nature. At the close of book 1 of the Treatise , Hume suggests that a reli-

ance on common (vulgar) opinions can provide a way out of hisdifficulties. As we shall see, his suggestions in the Treatise seem ten-tative and not obviously up to the job he assigns to them. Perhapsreflecting dissatisfaction with his previous treatment of the threatof radical skepticism, Hume returns to this topic in the opening

section of the Enquiry Concerning Human Understanding , where heconfronts the challenge anew. Then, in the closing section of theEnquiry , he offers a remarkable response to it: When the destructivemechanisms of Pyrrhonism—Hume’s label for radical skepticism—are counterbalanced by the mechanisms that produce common(vulgar) belief, then the mind, as a result of a vector of these twoopposing forces, naturally settles into the standpoint of a mitigated

and moderate skepticism. This is Hume’s final response to the radi-cal skepticism that seemed to overwhelm him in part 4, book 1 ofthe Treatise .

Countering radical skepticism by replacing it with a more mod-erate or mitigated skepticism is not, however, cost-free. Hume’snew modesty carries with it a sharp curtailment in the pretensionsof his science of human nature. Where Hume previously spoke ofproducing “a compleat system of the sciences, built on a founda-tion almost entirely new, and the only one upon which they canstand with any security” (4/xv), he now speaks circumspectly ofproducing “reflections of common life, methodized and corrected”(EHU , 121/162)

As this development unfolds, there are four contrasting Humes ,or at least four contrasting voices of Hume, inhabiting Hume’s writ-ing. The first is the confident Hume, projector of a complete scienceof human nature. The second is the melancholy Hume, wracked


26/193

Introduction 7

with Pyrrhonian doubts he seems incapable of shaking. Third, we

have the chastened Hume, modest in his expectations and reason-ably content with his lot. There is also a fourth voice or standpointfound in Hume’s writings, important but easily overlooked. This isthe standpoint of the ordinary people engaged in the affairs of dailylife: the standpoint of the vulgar.

Which standpoint represents the real Hume? As I read the text,all four standpoints are real in representing the way matters strike

Hume when operating at a particular level of reflection. At the startof the Treatise , and well into it, Hume is an enthusiast for the newscience of human nature he is developing. Hume’s standpoint under-goes a radical skeptical transformation in response to the appallingthings his pursuit of the science of human nature reveals to him.This is full-throated skepticism. The third standpoint emerges fromHume’s recognition that radical skepticism cannot be disposed of

by employing arguments against it. When matters are placed onan argumentative basis, the Pyrrhonist always wins. For Hume, theslide into radical skepticism can only be countered by yielding insome measure to our vulgar propensity to believe things that are notbased on sound arguments and, more deeply, even things that runcounter to sound arguments.

The textual study that follows is intended to show the basis forthe broad claims sketched earlier. It largely takes the form of a nar-rative. With very few exceptions, I present matters as they actuallyunfold in the text. Given Hume’s shifts in standpoint, a narra-tive approach seems virtually forced on us. It will operate at twointerrelated levels: one global , the other local . A global interpreta-tion concentrates on major aspects of a philosophical position andindicates how they are interrelated. Dealing with Hume’s writingsinvolves, among other things, keeping track of where Hume is inthe dialectical—dare I use the word?—unfolding of his position.


27/193

8 Introduction

Local interpretation involves a close reading of texts in an effort

to provide a check on the global interpretation, while at the sametime enriching its content. Grotesquely bad readings of a text usu-ally arise from violations of the principle of global interpretation—often made in an effort to square a position under examination with contemporary philosophical fashions. Shallow readings usu-ally result from neglecting the details of the text where, as it is said,God—sometimes the Devil—is said to reside. To borrow phrasing

from Kant, global interpretation without local interpretation isempty, local interpretation without global interpretation is blind.

Because I will be engaged in a close textual study, I will, for themost part, stay within the margins of the text and cite it extensively.I will not be much concerned with the sources of Hume’s ideas andonly occasionally compare Hume’s claims with those made by otherphilosophers. I will make no effort to show that Hume’s writings

are relevant to contemporary philosophical debates. I think theyare, but it seems patronizing to Hume to insist on this. I will notbe much involved with the rich and impressive secondary litera-ture on Hume that has appeared in recent decades. In dealing withthe secondary literature, one is again dealing with texts—texts thatoften refer to other texts and often present interpretive problemsof their own. I have not become deeply involved in the secondaryliterature because I do not see how this can be done in a fair andaccurate way without interrupting the flow of the narrative I ampresenting.

I would, however, suggest three works I find impressive that canbe read profitably in conjunction with this work. Barry Stroud’sHume , among its many virtues, will prove very helpful in fillingout my whirlwind account of book 3 of the Treatise that I pres-ent in Chapter 1. David Owen’s essay “Scepticism with Regard toReason” offers a rich and scholarly examination of the subject that


28/193

Introduction 9

is in some ways similar to mine and in some ways different. Finally,

I recommend Don Garrett’s Cognition and Commitment in Hume’sPhilosophy for a treatment of Hume’s skepticism that differs in fun-damental ways from my own.

In this work I make no claims for originality on particular points.Perhaps there is some novelty in the way I tell the overall story.


29/193



30/193

11

The central claim I will attempt to establish in this work is thatthe skeptical Hume of part 4 of the Treatise emerges naturallyout of Hume’s unfettered pursuit of the naturalistic program ofpart 3. To this end, I will offer a sketch, a précis, or a whirlwindtour of what Hume thinks he has accomplished in part 3. I willcite Hume’s conclusions and exert restraint—though not totalrestraint—in commenting on the strength of the arguments he

presents in their behalf. With this broad sketch in hand, I willthen turn to part 4 and examine the skeptical consequences thatemerge when Hume applies his naturalistic account of belief-formation to philosophical beliefs themselves. Part 3 is titled “Ofknowledge and probability.” One section is dedicated to a dis-cussion of knowledge, the remaining fifteen to probability andrelated topics.

A Quick Tour of Part 3, Book 1

Section 1 . Of knowledge

At the opening of this section, Hume repeats his list of what he callsphilosophical relations that he had originally presented and brieflydiscussed in part 1, section 5. They are

Z

Chapter 1Z Of Knowledge and Probability


31/193

12 Hume’s Skeptical Crisis

resemblance, identity, relations of time and place, proportion

in quantity or number, degrees in any quality, contrariety, andcausation. These relations may be divided into two classes;into such as depend entirely on the ideas, which we comparetogether, and such as may be chang’d without any change inthe ideas. (50/69)

He declares that “only four, which depending solely upon ideas,

can be the objects of knowledge and certainty. These four are resem-blance , contrariety , degrees in any quality , and proportions in quantityor number ” (50/70). The four relations that “depend entirely on theideas, which we compare together,” are the basis for the domain ofknowledge; the remaining three (relations of time and place, iden-tity, and causation), which do not have this feature, are the concernof the domain of probability.

Hume makes quick work of the domain of knowledge. He saysthat three of these relations—resemblance, contrariety, and degreesin any quality—are matters of intuition rather than demonstration. Whether they hold or not can be established by merely inspecting theideas themselves. Arithmetic contains intuitive truths, but containsdemonstrative truths as well. This is in virtue of what has come tobe known as Hume’s Law (one-to-one correlation), which provides

a “precise standard, by which we can judge . . . equality” (51/71). For want of a similarly precise standard, geometry can provide nothingbetter than what we might call damn-near demonstrative truths.

Given its importance for his project as a whole, this discussionof intuitive and demonstrative knowledge is surprisingly brief andunderdeveloped. It is striking that Hume expresses no dissatisfaction with it and gives no forewarning of the assault on both intuitive and

demonstrative knowledge that awaits the reader in the opening sec-tion of book 1, part 4.


32/193

Of Knowledge and Probability 13

Section 2. Of probability; and of the idea

of cause and effect

Hume’s treatment of knowledge is almost peremptory; his treat-ment of probability is elaborate and complex. There are, accordingto Hume, three kinds of relations that do not generate knowledge,that is, intuitive and demonstrative truths. They are identity , rela-tions of time and place , and causation. Each gets a part of book 1

dedicated to it. Relations of time and place are examined in part 2;causation is examined in part 3; and, setting aside its opening andclosing sections, identity is a central topic of part 4.1 I will quicklyrun through the basic moves concerning causation and probabilityas they unfold in book 1, part 3 of the Treatise .

Section 2 raises a question that will not be answered until sec-tion 14: How are we to define the relationship between a cause and

an effect? This section initiates the quest for an answer and movesthrough the following stages:

1. An examination of a particular instance of causation revealsonly two relationships: contiguity and priority.

2. But, according to Hume, these two relations by them-selves do not provide an adequate account of causation:

“Shall we then rest contented with these two relations ofcontiguity and succession, as affording a complete idea ofcausation? . . . There is a necessary connexion to be takeninto consideration; and that relation is of much greaterimportance, than any of the other two above mention’d”(55/77).

3. But an inspection of the objects themselves reveals nothing

corresponding to the idea of a necessary connection, soagain an impasse has been reached.


33/193


4. At this stage of his investigations, Hume seems explicitly

committed to the idea that a necessary connection is anessential component of a causal relation, but so far he hasbeen unable to give an adequate account of this idea. In anattempt to find another way of approaching this problem,Hume proposes to beat about in two neighboring fields,asking:

First , for what reason we pronounce it necessary , that every thing whose existence has a beginning, shou’d also have a cause?

Secondly , why we conclude, that such particular causes mustnecessarily have such particular effects; and what is the natureof that inference we draw from the one to the other, and of thebelief we repose in it? (55/78)

As a result of these two expeditions, Hume will reverse himself,and instead of holding that the idea of a necessary connection isan essential component of our idea of a causal relation, he will holdthat the idea of a necessary connection is an attendant product ofcausal reasoning.

Section 3. Why a cause is always necessary

To state the question more fully, what is the basis for thinking thateverything whose existence has a beginning should also have a causefor this beginning? Hume argues that this cannot be justified as eithera demonstrative or an intuitive truth for the following reason:

As all distinct ideas are separable from each other, and asthe ideas of cause and effect are evidently distinct, ’twill beeasy for us to conceive any object to be non-existent this


34/193


moment, and existent the next, without conjoining to it the

distinct idea of a cause or productive principle. The separa-tion, therefore, of the idea of a cause from that of a begin-ning of existence, is plainly possible for the imagination; andconsequently the actual separation of these objects is so farpossible, that it implies no contradiction nor absurdity; andis therefore incapable of being refuted by any reasoning frommere ideas; without which ’tis impossible to demonstrate the

necessity of a cause. (56/79)

With demonstrative and intuitive reasoning eliminated, it seemsthat the belief in question must have its basis in experience. Butinstead of pursuing this topic directly, Hume turns his attention tothe second neighboring field and asks “why we conclude, that suchparticular causes must necessarily have such particular effects.”

Section 4. Of the component parts of our reasonings concerning causes and effects

Hume argues that causal reasoning must have as its first componentsome initial experience, for without an origin in an initial experi-ence causal reasoning would be no more than hypothetical.

Section 5 . Of the impressions of the senses and memory

Hume lays out the component parts of his program:

Here . . . we have three things to explain, viz. first , The originalimpression. Secondly , The transition to the idea of the con-nected cause or effect. Thirdly , The nature and qualities ofthat idea. (59/84)


35/193


The remainder of the section discusses the difference between

memory and imagination, with some qualifications, drawing thedistinction in terms of vivacity.

Section 6 . Of the inference from the impression to the idea

This section demands special attention because it contains, in a

complex form, a line of reasoning that will later emerge (in the Abstract and the Enquiry Concerning Human Understanding ) asHume’s skeptical argument concerning induction.

Earlier, Hume was stymied in his attempt to discover anythingcorresponding to a necessary connection between causally relatedevents. All he could find of significance are the relations of priorityand contiguity. Now, however, when he turns his attention to the

specific assignment of causes and effects, he hits on the further ideathat events that are causally connected are constantly conjoined interms of priority and contiguity. This newly discovered notion ofconstant conjunction will play a central role in Hume’s first defini-tion of cause, but his initial reaction to it is negative.

To tell the truth, this new-discover’d relation of a constant

conjunction seems to advance us but very little in our way.For it implies no more than this, that like objects have alwaysbeen plac’d in like relations of contiguity and succession; andit seems evident, at least at first sight, that by this means wecan never discover any new idea, and can only multiply, butnot enlarge the objects of our mind. . . . From the mere rep-etition of any past impression, even to infinity, there never

will arise any new original idea, such as that of a necessaryconnexion; and the number of impressions has in this caseno more effect than if we confin’d ourselves to one only.(62/88)


36/193


The expression “at first sight” suggests that Hume is not really

committed to the idea that nothing new can emerge from the mererepetition of the same ideas, and, in fact, he will reject it. Hume alsodrops a hint of even deeper significance for understanding his ultimateposition concerning the relationship between necessary connectionsand causal inferences: “Perhaps ‘twill appear in the end, that thenecessary connexion depends on the inference, instead of the inference’sdepending on the necessary connexion” (62/88). This is just how things

will turn out, but it can only be shown after Hume gives his account ofthe “transition to the idea of the connected cause or effect,” the secondcomponent in the program announced in the previous section.

Hume first attempts to show that causal inferences are not theproduct of our rational faculties. The initial move, and the keymove, is this:

[If reason produced causal inferences] it wou’d proceed uponthat principle, that instances, of which we have had no experience,must resemble those, of which we have had experience, and that thecourse of nature continues always uniformly the same . (62/89)

The question now becomes: How can the principle that the course ofnature continues always uniformly the same itself be justified? Giventhe architecture of Hume’s position, there are only two options:

In order therefore to clear up this matter, let us considerall the arguments upon which such a proposition may besuppos’d to be founded; and as these must be deriv’d eitherfrom knowledge or probability , let us cast our eye on each ofthese degrees of evidence, and see whether they afford any just conclusion of this nature. (62/89)

The first alternative is ruled out by the conceivability argument:

There can be no demonstrative arguments to prove, that thoseinstances, of which we have had no experience, resemble those,


37/193


of which we have had experience . We can at least conceive a

change in the course of nature; which sufficiently proves, thatsuch a change is not absolutely impossible. To form a clearidea of any thing, is an undeniable argument for its possibil-ity, and is alone a refutation of any pretended demonstrationagainst it. (62/89)

The second alternative is eliminated by the argument from circularity:

According to this account of things, which is, I think, inevery point unquestionable, probability is founded on thepresumption of a resemblance betwixt those objects, of which we have had experience, and those, of which we have hadnone; and therefore ’tis impossible this presumption can arisefrom probability. (63/90)

What we have, then, is a skeptical argument intended to showthat there can be no justification of the principle that the course ofnature continues always uniformly the same. For Hume, this pro-vides a sufficient basis for saying that causal inferences are not theproduct of our rational faculties.

When the mind, therefore, passes from the idea or impres-sion of one object to the idea or belief of another, it is notdetermin’d by reason, but by certain principles, which associ-ate together the ideas of these objects, and unite them in theimagination. (64/92)

Section 7 . Of the nature of the idea, or belief

In section 5, Hume lays out a three-part program. The third compo-nent concerns “the nature and qualities of that idea” that is the prod-uct of a causal inference. For his purposes, it is important to draw a


38/193


distinction between merely entertaining an idea and having a belief.

He does so by invoking the notion of vivacity. Beliefs are distinguishedfrom mere ideas in virtue of their superior vivacity. With others,I find Hume’s appeal to vivacity, if taken in any reasonably literal sense,unconvincing, almost unintelligible. I believe that Bismarck is the cap-ital of North Dakota; I can also entertain the idea that Bismarck is thecapital of South Dakota. Is the first thought more vivacious (lively,vivid) than the second? Not to me. Degrees of vivacity might better be

taken as a metaphor for degree of belief instead of its basis.Hume uses the notion of vivacity in an effort to give a causal

account of belief-formation. For Hume, belief-formation is a suit-able topic for his science of human nature—indeed, it is a centraltopic. Beliefs, according to Hume, arise, are strengthened, weak-ened, augmented, revised, rejected, and so on, all in accordance with causal laws. This is an important idea—more important than

his use of vivacity to implement it. In what follows, however, I willsimply accede to Hume’s references to vivacity and treat them asplaceholders for a causal account of belief-formation. Taken as ametaphor, vivacity is not all that bad.

Section 8 . Of the causes of belief

This section presents what we might call Hume’s theory of vivacitytransfer. It is of central importance to Hume’s naturalistic accountof operations of the human mind.

I wou’d willingly establish it as a general maxim in the scienceof human nature, that when any impression becomes present tous, it not only transports the mind to such ideas as are related toit, but likewise communicates to them a share of its force andvivacity . (69/98)


39/193


In a causal inference there is a twofold movement: When an event is

experienced, the mind immediately envisages another event of thekind constantly conjoined with it. But the triggered idea does notsimply pop into one’s mind; it also inherits a portion of the vivacityof the triggering event.

In this section, Hume also invokes the notion of custom as an apt wayof referring to the manner in which causal inferences are grounded:

[From observation] I conclude, that the belief, which attendsthe present impression, and is produc’d by a number of pastimpressions and conjunctions; that this belief, I say, arisesimmediately, without any new operation of the reason orimagination. Of this I can be certain, because I never amconscious of any such operation, and find nothing in the subject, on which it can be founded. Now as we call every thing

custom, which proceeds from a past repetition, without anynew reasoning or conclusion, we may establish it as a cer-tain truth, that all the belief, which follows upon any presentimpression, is deriv’d solely from that origin. (71–72/102)

Section 9. Of the effects of other relations, and other habits

This section examines the lesser influence of the other two prin-ciples of association—resemblance and contiguity—in transferringbelief. It provides an account of why causal associations are, in gen-eral, stronger than the other two modes of association.

Section 10. Of the influence of belief

This complex and interesting section contains a series of reflectionson the following fact:


40/193


Pain and pleasure have two ways of making their appearance

in the mind; of which the one has effects very different fromthe other. They may either appear in impression to the actualfeeling and experience, or only in idea, as at present whenI mention them. ’Tis evident the influence of these upon ouractions is far from being equal. Impressions always actuatethe soul, and that in the highest degree; but ’tis not every idea which has the same effect. (81/118)

Though it contains important material, most of what Hume says herebears more directly on topics discussed in books 2 and 3 of the Treatise than on our present topic, the character of causal reasoning.

Section 11 . Of the probability of chances

In this section Hume uses the notion of dispersed vivacity to explaina priori probabilities. A crude example: We fully believe that, whencast, a six-sided die will come to rest on one of its six sides—not onan edge. We can say that the vivacity level for this claim is effectively 1 .On the assumption that the die is not loaded, there are six equallylikely outcomes from casting the die, so the vivacity level for eachface of the die coming up on top is 1 / 6 . This is the probability we will

naturally assign to the possibility that the die will come up, say, 4.This may not sound particularly plausible, but, as I have suggestedelsewhere, the notion of vivacity transfer can be used to underwritea standard probability calculus (Fogelin 1985, pp. 59–60).2

Section 12. Of the probability of causes

An application of the notion of dispersed vivacity is used to explain what we would call statistical probabilities. Very roughly, the distri-bution of vivacity reflects the relative frequency of the occurrence


41/193


of an event in a given setting. The more often A is followed by B ,

the greater the vivacity handed over by A to B , thus generating agreater degree of subjective probability that B will follow upon anoccurrence of A.

Section 13. Of unphilosophical probability

Having completed his discussion of the probability of chances andthe probability of causes, both of which, Hume tells us, “are receiv’dby philosophers, and allow’d to be reasonable foundations of beliefand opinion” (97/143), he turns his attention to other kinds of prob-ability that have not “had the good fortune to obtain the same sanc-tion” (97/143). This section is important for seeing a fundamentalfeature of Hume’s science of human nature: All beliefs, includingthe ill-formed, the fiction-ridden, and the just plain nutty, fallunder its purview. Hume takes it to be one of the chief strengths ofhis position that it can explain the sources both of beliefs that are“respectable” and of those that are not. Included in the second cat-egory are many beliefs of philosophers—both ancient and modern.This, as we will see, is a central theme of part 4 of book 1. Giventhe importance of this section to this later part of the Treatise , I will

dedicate a separate chapter to it.

Section 14. Of the idea of necessary connexion

After an eleven-section excursion through neighboring fields,Hume returns to the problem posed in section 2 concerning thestatus of a necessary connection. His line of march, as I understand

it, runs as follows. In section 2 Hume takes it for granted that anecessary connection is an essential component of a causal relation.Presumably it serves to connect the cause with its effect and in that


42/193


way underwrites an inference from one to the other. Having shown

(in section 6) that the inference from cause to effect cannot be theproduct of the operations of our rational faculties, the notion of anecessary connection is no longer burdened with the task of under- writing this inference. Hume is therefore free to treat the idea of anecessary connection as a product of this inference rather than as itsbasis . This is precisely what he concludes in the opening paragraphof section 14:

After a frequent repetition, I find, that upon the appearanceof one of the objects, the mind is determin’d by custom toconsider its usual attendant, and to consider it in a strongerlight upon account of its relation to the first object. ’Tis thisimpression, then, or determination, which affords me the ideaof necessity. (105/156)

The text exhibits what has come to be known as a Euthyphro choice:Do we draw inferences from cause to effect because we think theyare necessarily connected, or do we think that they are necessarilyconnected because we draw inferences from one to the other? For what I take to be dialectical reasons, Hume starts out (seemingly)adopting the first option, then ends up adopting the second, moredaring, option. In the Treatise , Hume is fond of such reversals. We will encounter other striking instances of it.

Having settled the status of the idea of a necessary connection,Hume then expresses his concern that his argument, being so obviouslycorrect, might not be appreciated for its depth and importance.

This evidence both in the first principles, and in the deduc-tions, may seduce us unwarily into the conclusion, and make

us imagine it contains nothing extraordinary, nor worthy ofour curiosity. But tho’ such an inadvertence may facilitate thereception of this reasoning, ‘twill make it be the more easily


43/193


forgot; for which reason I think it proper to give warning,

that I have just now examin’d one of the most sublime ques-tions in philosophy. (105/156)

Hume therefore allots another thirteen pages to an expansion ofhis basic argument concerning our idea of a necessary connection.There is some interesting material in these pages, including:

Thus as the necessity, which makes two times two equal to

four, or three angles of a triangle equal to two right ones,lies only in the act of the understanding, by which we con-sider and compare these ideas; in like manner the necessity orpower, which unites causes and effects, lies in the determina-tion of the mind to pass from the one to the other. (112/166)

For Hume, the feeling of necessity that one has in thinking that

2 × 2 = 4 is the same feeling that one has that a heavy object willfall if it is released from a height. By modern standards—perhapsby the standards of Hume’s time—this is a curious notion ofnecessity, but the text shows unambiguously that he is commit-ted to it.

Hume’s additional remarks on necessary connections also con-tain an important summary passage that confirms the interpretation

of the text I have presented:’Tis now time to collect all the different parts of this reason-ing, and by joining them together form an exact definition ofthe relation of cause and effect, which makes the subject ofthe present enquiry. This order wou’d not have been excus-able, of first examining our inference from the relation before we had explain’d the relation itself, had it been possible toproceed in a different method. But as the nature of the rela-tion depends so much on that of the inference, we have been


44/193


oblig’d to advance in this seemingly preposterous manner,

and make use of terms before we were able exactly to definethem, or fix their meaning. We shall now correct this fault bygiving a precise definition of cause and effect. (114/169)

With the idea of a necessary connection now properly put in itsplace, Hume next proceeds to offer two definitions of cause, onetreating causation as a philosophical relation, the other treating it as

a natural relation.There may two definitions be given of this relation, which areonly different, by their presenting a different view of the sameobject, and making us consider it either as a philosophical or asa natural relation; either as a comparison of two ideas, or as anassociation betwixt them. We may define a cause to be “[a]nobject precedent and contiguous to another, and where all

the objects resembling the former are plac’d in like relationsof precedency and contiguity to those objects that resemblethe latter.” If this definition be esteem’d defective, becausedrawn from objects foreign to the cause, we may substitutethis other definition in its place, viz . “[a] cause is an objectprecedent and contiguous to another, and so united with it,that the idea of the one determines the mind to form the ideaof the other, and the impression of the one to form a morelively idea of the other.” (114/169–70)

In case the reader has somehow missed it, Hume adds a paragraphdriving home the point that neither definition makes reference to anecessary connection—thus reversing his original claim that a nec-essary connection is an essential element of a causal relation.

If we define a cause to be, an object precedent and contiguousto another, and where all the objects resembling the former are


45/193


plac’d in a like relation of priority and contiguity to those objects,

that resemble the latter ; we may easily conceive, that there is noabsolute nor metaphysical necessity, that every beginning ofexistence shou’d be attended with such an object. If we definea cause to be, an object precedent and contiguous to another, andso united with it in the imagination, that the idea of the one deter-mines the mind to form the idea of the other, and the impressionof the one to form a more lively idea of the other ; we shall make

still less difficulty of assenting to this opinion. (115–16/172)

These two definitions are not without difficulties. The second, which treats causation as a natural relation, seems circular inemploying the transparently causal term “determines” as part of thedefinition of a cause. This criticism can be avoided by treating thesecond definition (so-called) as a causal statement concerning causal

belief-formation. There is nothing circular in that.The first definition, where causation is treated as a philosophicalrelation, faces a more serious challenge. Some, and not merely a few,have taken Hume to be presenting a regularity definition of a causallaw . On this reading—I’ll call it the causal law reading—Hume’sfirst definition is taken to mean:

If every event of type B is uniquely paired with an event oftype A that is prior to it and contiguous with it, then eventsof type A cause events of type B .

Thomas Reid interpreted Hume in this manner and produced thefollowing counterexample to it:

It follows from this definition of cause, that night is the causeof day, and day the cause of night. For no two things havemore constantly followed each other since the beginning ofthe world. (Reid 334)


46/193


In fact, Hume’s first definition is not correctly represented by the

causal law reading and does not imply it. Hume’s first definitionconcerns a singular causal statement of the form “a is the cause ofb,” and can be represented as follows:

a is the cause of b if and only if a is contiguous and prior to b,and any object resembling a is similarly paired with an objectresembling b, and any object resembling b is paired with an

object resembling a that is contiguous and prior to it.Hume’s first definition, taken in this, the correct, way, is not subject to Reid’s criticism. The point is this: Predicates used to pickout the cause and to pick out the effect need not fix the appro-priate resemblance classes.3 The definition may be subject to othercriticisms—in particular, concerning the specification of the properresemblance classes—but I will not go into this matter here.

Section 15 . Rules by which to judge of causes and effects

Proponents of a naturalistic reading of Hume’s Treatise point to thissection as confirmation of their approach. They are right to do so. It

opens with the following declaration: According to the precedent doctrine, there are no objects, which by the mere survey, without consulting experience, wecan determine to be the causes of any other; and no objects, which we can certainly determine in the same manner notto be the causes. Any thing may produce any thing. . . . Sincetherefore ’tis possible for all objects to become causes or effectsto each other, it may be proper to fix some general rules, by which we may know when they really are so. (116/173)


47/193


He then lays down eight rules that will serve as guides to empirical

inquiry, telling his reader that they are of much more use than theelaborate systems put forward by “our scholastic head-pieces andlogicians [who] show no such superiority above the mere vulgar intheir reason and ability” (117/175). It is hard to see, as naturalisticinterpreters point out, why Hume would find it useful to presentsuch a system of rules for empirical inquiry if he harbored seriousdoubts about the very possibility of such inquiry.

Section 16 . Of the reason of animals

Hume holds that higher animals are capable of reasoning of thesame kind employed by human beings. The dog that “avoids fireand precipices, that shuns strangers, and caresses his master,” hetells us, proceeds “from a reasoning, that is not in itself different, nor

founded on different principles, from that which appears in humannature” (119/177). He thus claims that his theory can account forboth human and animal reasoning and challenges those who rejectit to attempt to do likewise:

Let any philosopher make a trial, and endeavour to explain thatact of the mind, which we call belief, and give an account of the

principles, from which it is deriv’d, independent of the influ-ence of custom on the imagination, and let his hypothesis beequally applicable to beasts as to the human species; and afterhe has done this, I promise to embrace his opinion. (119/178)

Passages like this, and many others of the same kind, seem tosettle the interpretive dispute in favor of naturalism—unless, thatis, we turn pages and enter the skeptical realm of part 4. In fact, asbriefly noted, the skepticism that emerges in part 4 is anticipated inpart 3 itself, in section 13, under the heading “Of unphilosophicalprobability.” This is the subject of the next chapter.


48/193

29

Unphilosophical as Opposed to Philosophical Probabilities

In sections 11 and 12 of part 3, Hume examines what he calls theprobability of chances and the probability of causes. At the start ofsection 13, he tells us that both “are receiv’d by philosophers, and

allow’d to be reasonable foundations of belief and opinion.” He thencontinues: “But there are [other kinds of probability], that are deriv’dfrom the same principles, tho’ they have not had the good fortune toobtain the same sanction” (97/143). Hume dubs these other kinds ofprobability—those not sanctioned by philosophers—unphilosophical probabilities. In sections 11 and 12, Hume argues that philosophicalprobabilities are grounded in the operations of the imagination. Insection 13, he makes the same claim with respect to unphilosophicalprobabilities. The upshot is that both philosophical and unphilo-sophical probabilities spring from the same principles: the opera-tions of the imagination. Furthermore, philosophical probabilitiesand unphilosophical probabilities can come into conflict with oneanother, and that, on Hume’s theory, amounts to saying that theimagination can come into conflict with itself. As we shall see, expos-ing conflicts within the imagination is a central theme of part 4, andtheir discovery is the primary source of Hume’s skeptical jitters.

Z

Chapter 2Z Hume on Unphilosophical Probabilities


49/193


Sources of Unphilosophical Probabilities

Hume examines four sources, or kinds, of unphilosophical prob-ability: the effect of the remoteness of an event, the effect of theremoteness of an observation, reiterative diminution, and prejudicebased on general rules.

The effect of the remoteness of the event

Hume’s discussion of this kind of unphilosophical probability pro-vides his first example of how a clash can arise between the deliver-ances of philosophical and unphilosophical probabilities.

The argument, which we found on any matter of fact we

remember, is more or less convincing, according as the fact isrecent or remote; and tho’ the difference in these degrees ofevidence be not receiv’d by philosophy as solid and legitimate;because in that case an argument must have a different force to-day, from what it shall have a month hence; yet notwithstand-ing the opposition of philosophy, ’tis certain, this circumstancehas a considerable influence on the understanding, and secretly

changes the authority of the same argument, according to thedifferent times, in which it is propos’d to us. (97–98/143)

Here Hume speaks of an argument that appeals to events remote intime and claims that, as a matter of fact, the more remote an eventis, the less authority it will carry. This psychological fact cannot beexplained by appealing to principles of received philosophy, since itruns counter to them. It can, Hume thinks, be explained by thefact that, in general, events remote in time have less impact on theimagination than those that have occurred more recently.

Z


50/193

Hume on Unphilosophical Probabilities 31

It is worth asking why Hume should be interested in what seems

to be an unsound, actually quite stupid, mode of probabilistic rea-soning. Part of the answer is that this phenomenon provides anopportunity for him to show off the power of his own position. Hethinks that on his principles he can give an explanation of why sucha mode of reasoning takes place, whereas other theories cannot.More deeply, in his science of human nature, Hume is attemptingto give a causal account of the ways in which human beings actually

think and act. Beliefs as they are actually formed will play a centralrole in this project. Hume’s scientific investigations will be governedby philosophical rules of probabilistic reasoning (see part 3, section 15),but, as he sees, it would be a mistake to suppose that commonhuman belief-formation is equally governed by such rules.

The effect of the remoteness of the observation

The second kind of unphilosophical probability Hume examines issimilar to the first, except that now it is the remoteness of the obser-vation, and not of the event, that diminishes vivacity. “An experi-ment, that is recent and fresh in the memory,” he tells us, “affects usmore than one that is in some measure obliterated” (98/143).

Reiterative diminution

Because of its close connection with Hume’s skepticism with regardto reason, Hume’s discussion of his third kind of unphilosophicalprobability is worth careful examination.

I add, as a third instance of this kind, that tho’ our reasoningsfrom proofs and from probabilities be considerably differentfrom each other, yet the former species of reasoning oftendegenerates insensibly into the latter, by nothing but the


51/193


multitude of connected arguments. ’Tis certain, that when an

inference is drawn immediately from an object, without anyintermediate cause or effect, the conviction is much stronger,and the perswasion more lively, than when the imaginationis carry’d thro’ a long chain of connected arguments, how-ever infallible the connexion of each link may be esteem’d.(98/144)

The point Hume is making is clear: Long chains of reasoning canlose their force simply in virtue of their length. The reason for thisis that “vivacity must gradually decay in proportion to the distance,and must lose somewhat in each transition.” Here various analogiesspring to mind: The friction involved in each transition rubs offsome of the vivacity’s luster, or, here, as elsewhere, entropy takes itstoll. None of these analogies seems adequate. Then again, Hume’s

talk about vivacity is not without problems of its own. In any case,the important idea here is that “re-iterated diminutions,” to useHume’s phrase (99/145), will reduce the doxastic force of a series ofinferences in direct proportion to its length.

At this point, Hume goes off into a neighboring field to respondto the criticism that the principle of reiterated diminutions, if true, would have the effect of destroying our knowledge of the ancient

past.’Tis evident there is no point of antient history, of which we can have any assurance, but by passing thro’ many mil-lions of causes and effects, and thro’ a chain of argumentsof almost an immeasurable length. Before the knowledge ofthe fact cou’d come to the first historian, it must be convey’dthro’ many mouths; and after it is committed to writing,each new copy is a new object, of which the connexion withthe foregoing is known only by experience and observation.


52/193


Perhaps, therefore, it may be concluded from the precedent

reasoning, that the evidence of all antient history mustnow be lost; or at least, will be lost in time, as the chain ofcauses encreases, and runs on to a greater length. . . . Thismay be consider’d as an objection to the present system.If belief consisted only in a certain vivacity, convey’d froman original impression, it wou’d decay by the length of thetransition, and must at last be utterly extinguish’d: And vice

versa , if belief on some occasions be not capable of such anextinction; it must be something different from that vivacity.(98/144–45)

Hume’s response to this criticism is quite remarkable. In essenceit comes to this: Whether reiterative diminution will destroy ourbeliefs concerning the ancient past depends on the manner in which

we view the sequence of transitions—an idea of central importancein part 4. Looked at in one way, reiterative diminution wouldindeed have the effect of destroying all beliefs handed down to usfrom the past:

If all the long chain of causes and effects, which connect anypast event with any volume of history, were compos’d of parts

different from each other, and which ’twere necessary for themind distinctly to conceive, ’tis impossible we shou’d preserveto the end any belief or evidence. (99/146)

That is, if we look at the matter sequentially, we would recognizethat each time information about the remote past is transmitted,some vivacity (hence assurance) is lost.1 With a sufficient number ofreiterations, belief in the past events would be annihilated. But wedo have strong beliefs concerning the remote past. How, on Hume’stheory, is this possible?


53/193


Hume’s answer is that, in fact, and fortunately, we do not view

the matter sequentially:

Let us consider, that tho’ the links are innumerable, that con-nect any original fact with the present impression, which isthe foundation of belief; yet they are all of the same kind, anddepend on the fidelity of printers and copists. One editionpasses into another, and that into a third, and so on, till we

come to that volume we peruse at present. There is no varia-tion in the steps. After we know one, we know all of them;and after we have made one, we can have no scruple as tothe rest. This circumstance alone preserves the evidence ofhistory. (99/146)

Then this crucial move:

But as most of these proofs [that make up the long chainof inferences] are perfectly resembling, the mind runs easilyalong them, jumps from one part to another with facility, and forms but a confus’d and general notion of each link . (99/146,emphasis added)

What are we to make of Hume’s claim that we have a confused

notion of each link in the chain of inferences? For Hume, havinga confused notion of these links contrasts with having a distinctnotion of them, that is, recognizing them as distinct entities. Thismay seem a bit shifty in its use of the notion of being distinct, butit is just what Hume says:

By this means a long chain of argument, has as little effect indiminishing the original vivacity, as a much shorter wou’d have,if compos’d of parts, which were different from each other, andof which each requir’d a distinct consideration. (99/146)


54/193


It seems, then, that beliefs in remote past events are preserved only

because we become fuddled when we think of how they have beenhanded down to us. This idea that a limitation or weakness in thehuman intellect can protect us from an unanswerable skeptical chal-lenge will reappear in part 4, section 1.

Prejudice based on general rules

Hume calls his fourth species of unphilosophical probability“prejudice”:

A fourth unphilosophical species of probability is that deriv’dfrom general rules , which we rashly form to ourselves, and which are the source of what we properly call prejudice. An Irishman cannot have wit, and a Frenchman cannot have

solidity; for which reason, tho’ the conversation of the formerin any instance be visibly very agreeable, and of the latter very judicious, we have entertain’d such a prejudice against them,that they must be dunces or fops in spite of sense and reason.Human nature is very subject to errors of this kind; and per-haps this nation as much as any other. (99–100/146–47)

With respect to prejudice, the general rules that govern our judgmentare rashly formed and stand in contrast to those that proceed from“more general and authentic operations of the understanding.”

Following the pattern found in his discussion of the first threeforms of unphilosophical probability, Hume proceeds to offer anaccount of why human beings reason in this unreasonable way:

Shou’d it be demanded why men form general rules, andallow them to influence their judgment, even contrary topresent observation and experience, I shou’d reply, that in my


55/193


opinion it proceeds from those very principles, on which all

judgments concerning causes and effects depend. Our judg-ments concerning cause and effect are deriv’d from habit andexperience; and when we have been accustom’d to see oneobject united to another, our imagination passes from thefirst to the second, by a natural transition, which precedesreflection, and which cannot be prevented by it. (100/147)

The text here is not altogether transparent. It seems that Hume isattributing a double capacity to the imagination. After experiencinga customary union of objects of type A with objects of type B , on theappearance of A we are disposed to infer the forthcoming presenceof B , but not only that, we are also disposed to accept a general ruleto the effect that As are sure indicators of the associated B s. Suppose,for example, that Nigel believes that a Frenchman cannot have solid-

ity. He might have absorbed this belief from his cultural heritage,but let us suppose that his belief is based on his own encounters with the French. Having encountered a series of Frenchmen lackingin solidity, his imagination performs two operations. First, on againencountering a Frenchman he will immediately take it for grantedthat he is dealing with someone lacking solidity. Along with this,under the influence of the imagination, he will also adopt a general

rule to the effect that all Frenchmen lack solidity. Now suppose heencounters a Frenchman of unquestionable depth and probity. If herecognizes this, then, from the standpoint of philosophical prob-ability, he should resist automatically assuming that this particularperson lacks solidity. Similarly, he should relax his general rule thatall Frenchmen lack solidity. However, the grip of the general rulecan be tenacious, preventing Nigel from recognizing the depth and

probity of the Frenchman before him. General rules, tenaciouslyheld, can blind Nigel to contrary evidence. He may even see the


56/193


Frenchman’s actions as foppish when they are not. All in all, not a

bad start for an account of the nature of prejudice.

Conflicts within the Imagination and Skepticism

It is easy to read this examination of the nature and sources of preju-

dice as no more than a cautionary tale concerning the dangers ofrash generalizations as opposed to the security provided by general-izations that are broader and more authentic. That, however, is nothow Hume proceeds. Instead, he treats the conflict between generalrules that are grounded in philosophical probabilities and those thatare grounded in unphilosophical probabilities as a challenge to histheory:

According to my system, all reasonings are nothing but theeffects of custom; and custom has no influence, but by enliv-ening the imagination, and giving us a strong conception ofany object. It may, therefore, be concluded, that our judg-ment and imagination can never be contrary, and that cus-tom cannot operate on the latter faculty after such a manner,

as to render it opposite to the former. (101/149)The fact is, however, that general rules derived from unphilosophi-cal probabilities sometimes do clash with those having a philosophi-cally more proper heritage.

Thus our general rules are in a manner set in opposition toeach other. When an object appears, that resembles any cause

in very considerable circumstances, the imagination naturallycarries us to a lively conception of the usual effect, tho’ the

Z


57/193


object be different in the most material and most efficacious

circumstances from that cause. Here is the first influence ofgeneral rules. But when we take a review of this act of themind, and compare it with the more general and authen-tic operations of the understanding, we find it to be of anirregular nature, and destructive of all the most establish’dprinciples of reasoning; which is the cause of our rejecting it.This is a second influence of general rules, and implies the

condemnation of the former. Sometimes the one, sometimesthe other prevails, according to the disposition and characterof the person. The vulgar are commonly guided by the first,and wise men by the second. (101–2/149–50)

It seems, then, that the operations of the associative principles ofthe imagination are not highly discriminate in the beliefs that they

entrench.These reflections lead Hume to make his first reference to skepti-cism in the Treatise :

Meanwhile the sceptics may here have the pleasure of observ-ing a new and signal contradiction in our reason, and of seeingall philosophy ready to be subverted by a principle of humannature, and again sav’d by a new direction of the very sameprinciple. The following of general rules is a very unphilo-sophical species of probability; and yet ’tis only by followingthem that we can correct this, and all other unphilosophicalprobabilities. (102/150)

This passage is a harbinger of things to come in part 4, where con-flicts within the operations of the mind—and the skeptical conse-

quences that flow from them—become a central concern.


58/193

39

The Turn to Skepticism

Part 4 of book 1 of the Treatise represents the outer limit of Hume’scritical reflections on the philosophical enterprise. To borrow aphrase from Immanuel Kant, it represents Hume’s attempt to delin-eate the fate of reason when its demands are pursued in an unre-

stricted manner. The result, as both Hume and Kant portray it, isintellectual disaster. There is, however, an important difference inthe scope of their critiques. Kant’s primary target is the attemptto produce substantive (synthetic) a priori knowledge that reachesbeyond the phenomenal world. Roughly speaking, his critique con-cerns metaphysics as traditionally pursued. Hume’s critique is moreradical and more far-reaching, for it calls into question both theunderstanding and the senses in their modest and natural employ-ments, not just their misemployment in a priori metaphysics. Theopening sections (1 and 2) of part 4 taken together yield the sweep-ing negative conclusion that

’[t]is impossible upon any system to defend either our under-standing or senses; and we but expose them farther when we

endeavour to justify them in that manner. As the scepticaldoubt arises naturally from a profound and intense reflection

Z

Chapter 3Z Hume’s Skepticism with Regard to Reason


59/193


on those subjects, it always encreases, the farther we carry

our reflections, whether in opposition or conformity to it.(144/218)

Hume seems to leave no areas protected from radical doubt except,perhaps, immediate subjective reports.

I think this result comes as a shock to the reader after the con-fident development of Hume’s science of human nature in the first

three parts of book 1. Earlier, in Hume’s discussion of unphilosophi-cal probabilities, there were some indications that the operations ofthe imagination can have problematic results, but nothing preparesthe reader for the radical change of direction that seems to occur when we pass from part 3 to part 4.

The Basic Argument

This much is clear from the text: Section 1, part 4, book 1 of theTreatise opens with a skeptical examination of the demonstrativesciences (Hume’s words), that is, of beliefs based on demonstrationand intuition. Given his own nomenclature, both intuitive anddemonstrative truths are in the domain of knowledge as opposed

to the domain of probability . It is also clear that Hume’s skep-ticism is not limited to the domain of knowledge, but includesthe domain of probability as well. The moral that Hume drawsfrom his skeptical reflections is that our intellectual faculties, whenallowed to follow their own principles without restraint, are whollydestructive of beliefs based on reasoning. So the scope of Hume’sargument is unrestricted: It is not limited to those beliefs that are

the product of demonstrative and intuitive reasoning, though itincludes them.

Z


60/193

Hume’s Skepticism with Regard to Reason 41

It is also clear from the text that Hume’s skepticism concerning

demonstrative reasoning is based on probabilistic reasoning. Thiscontrasts with traditional skeptical attacks on demonstrative reason-ing that usually point to such things as the circularity of demonstra-tive proofs, the threat of an infinite regress, the lack of probity, theappearance of paradoxes, and the like. Hume’s challenge is quitedifferent and, to the best of my knowledge, without precedent. Ineffect, he starts by raising the question: How do we establish the

appropriate level of assurance that we should assign to any piece ofdemonstrative reasoning? That, for Hume, is a matter of assessingthe probability of error.

Reducing knowledge to probability

The first step in Hume’s skeptical argument is an attempt to show

that, under the critical eye of proper reasoning, knowledge degener-ates into probability. The key move involves what Hume will laterrefer to as “a reflex act of the mind” (122/182).

In all demonstrative sciences the rules are certain and infal-lible; but when we apply them, our fallible and uncertain fac-ulties are very apt to depart from them, and fall into error.

We must, therefore, in every reasoning form a new judgment,as a check or controul on our first judgment or belief; andmust enlarge our view to comprehend a kind of history ofall the instances, wherein our understanding has deceiv’d us,compar’d with those, wherein its testimony [emphasis added] was just and true. Our reason must be consider’d as a kind ofcause, of which truth is the natural effect; but such-a-one as

by the irruption of other causes, and by the inconstancy of ourmental powers, may frequently be prevented. By this means


61/193


all knowledge degenerates into probability; and this probabil-

ity is greater or less, according to our experience of the verac-ity or deceitfulness of our understanding, and according tothe simplicity or intricacy of the question. (121/180)

Before getting into details of the skeptical argument concerningreason, we should note that Hume presents his argument using themetaphor of “testimony.” Our reason, he tells us, provides us with

testimony that is sometimes “just and true”; at other times it “deceivesus.” In order to assess the reliability of this testimony, we shouldshift our attention away from the immediate object of our judgmentand examine the reliability of our faculties in dealing with the mat-ter at hand. We should consider the ratio of the testimony that hasdeceived us relative to the testimony that is just and true. Ideally, theveracity should fully outweigh the deceitfulness, leaving us fully (or

at least highly) confident. If not, our level of confidence should beadjusted accordingly. This sounds quite sensible and in line with thediscussion in section 12, “Of the probability of causes.”1

Hume may be wrong in saying “knowledge degenerates intoprobability”—we will look at his arguments for this shortly—but atleast he is saying this: Establishing a claim to demonstrative knowl-edge always involves establishing a prior claim of probability. That

is an arresting idea in itself, and not altogether implausible.2

Hume adds another reason for holding that the evaluation ofdemonstrative reasoning reduces to or is at least dependent on prob-abilistic reasoning.

There is no algebraist nor mathematician so expert in hisscience, as to place entire confidence in any truth immedi-ately upon his discovery of it, or regard it as any thing buta mere probability. Every time he runs over his proofs, hisconfidence increases; but still more by the approbation of his


62/193

Hume’s Skepticism with Regard to Reason 43

friends; and is rais’d to its utmost perfection by the universal

assent and applauses of the learned world. Now, ’tis evident,that this gradual increase of assurance is nothing but the addi-tion of new probabilities, and is derived from the constantunion of causes and effects, according to past experience andobservation. (121/180–81)

Here it will help to contrast a horizontal with a vertical perspective

for evaluating the soundness of a proof. The wise algebraist runsover his proofs a number of times to make sure they are free of error.Similarly

Robert J. Fogelin-Hume's Skeptical Crisis_ a Textual Study-Oxford University Press, USA (2009)

Documents