Table Of ContentTHOUGHT EXPERIMENT
ON THE POWERS AND LIMITS
OF IMAGINARY CASES
Tamar Sza b6 Gendler
GARLAND PUBLISHING, INC.
A MEMBER OF THE TAYLOR & FRANCIS GROUP
NEW YORK & LONDON/2000
Published in 2000 by
G:~rland Publishing, Inc.
A member of the Taylor & Francis Group
29 West 35th Street
New York, NY 10001
Copyrigln © 2000 by Tamar Szabo Gendler
All rights reseved. No p:lrt of this book may he reprinted or repro
duced or utilized in any form or by any electronic, mechanical, or
other means, now known or hereafter invented, including photo·
copying and recording, or in any information storage or retrieval
system, without wrirten permission from the publishers.
10 9 8 7 6 5 4 3 2 l
Library of Congress Catalogi,g-iu-Publicatioll Data
Gendler, Tamar.
Thought experiment : on the power' nnd limits of imasi
nary cases I Tam:~r Szabo Gendler.
p. em.-(Srudies in philosophy)
Includes bibliographic::~! references ::~nd index.
ISBN 0-81 53-3656-X (alk. paper)
1. Thought experiments. I. 1itle. II.Srudies in philosophy
(New York, N.Y.)
80265 .G46 2000
128'. 3-dc21
00-034074
Printed on acid-free, 250 year-life paper
Maufactured in the United States of America
Contents
0. Preface IX
0.1 Revisions .Made X
0.2 Summary of Coments xi
0.2.1 Chapter 1-lntroduction xii
0.2.2 Chapter 2-Galileo (First Case Study) xiii
0.2.3 Chapr~r 3-The Ship of Theseus
(Second Case Swdy) xiii
0.2.4 Chapter 4- l>ersonal Identity
(Third Case Study) xiv
0.2.5 Chapter 5-Conclusion xiv
0.3 Acknowledgments xvi
1. Introduction 1
1.1 Exceptional Cases 2
1.1.1 Characterization of "Exceptional Case" 2
1.1.1.1 Exceptional Cases in Theories with
Privileged Characteristics 2
1.1.1.2 Exceptional Cases in Theories without
Privileged Characteristics 6
1.1.2 Ways of Accounting for Exceptional
Cases 8
1.'1.3 Patterns of Accounting 9
1.1.4 Application to the Question at Hand 12
v
vi Contents
1.2 Imaginary Cases 14
1.3 Thought Experiments 18
1.3.1 What is a Thought Experiment? 18
1.3.2 The Tripartite Structure of
Thought Experiments 21
1.3.3 Three Sorts of Thought Experiments 25
1.4 Appendix to Chapter 1 28
2. Galileo 33
2.1 Argumentative Reconstruction 33
2.1.1 The Elimination Thesis 34
2.1.2 Clarification of Terminology 34
2.1.3 The Negative Argument and
the Positive Argument 36
2.1.4 The Dispensability Thesis and
the Derivativity Thesis 37
2.2 Galileo's Thought Experiment
and its Reconstruction 39
2.2.1 Galileo's Thought Experiment 39
2.2.2 Reconstruction of the Galileo Case 41
2.2.3 Four Ways out for the Aristotelian 42
2.2.4 What the Reconstruction Misses 45
2.3 Denying the Dispensability and
Derivativity Theses 47
2.3.1 Rejecting Reconstruction: What the
Thought Experiment Does 47
2.3.2 Rejecting the Positive Argument:
What Makes these Beliefs New? 49
2.3.3 Rejecting the Negative Argument:
What Makes these Beliefs Knowledge? 53
2.3.4 Constructivism and the Contrast
with Norton and Brown 56
2.4 Conclusion 62
3. Theseus 65
3.1 Conceptual Thought Experiments 65
3.2 The Story 67
3.3 The Puzzle 68
3.4 Is the Ship of Theseus an
Exceptional Case? 70
Contents vii
3.4.1 Automatic and Specially-Secured
Identity 70
3.4.2 Organisms, Artifacts, and
Exceptional Cases 72
3.5 Attempts to Dissolve the Problem 79
3.5.1 Van Inwagen 79
3.5.1.1 Identity Under a Sortal 82
3.5.1.2 Summary 84
3.5.2 Parfit 85
3.6 Attempts to Solve the Problem 88
3.6.1 A Traditional Solution: Hirsch 88
3.6.2 A Meta-solution: Nozick 91
3.7 The Proposed Diagnosis 97
3.7.1 Some Very General Candidate
Principles 99
3.7.2 Remarks on the Candidates 101
3.7.3 A Messier Puzzle 104
3.7.4 The Proposed Diagnosis 106
4. Personal Identity 111
4.1 Introduction: The Facts of Life 111
4.2 Setting the Stage 115
4.2.1 A Context for Parfit's Argument 115
4.2.2 What Fission Might Show 116
4.3 The Argument and its Crucial
Assumptions 120
4.3.1 Parfit's Fission Argument 120
4.3.2 Four Crucial Distinctions 125
4.3.3 Comments on these Distinctions 127
4.3.4 The Intrinsicness Premise 129
4.3.5 Summary 129
4.4 Two Unsuccessful Strategies 130
4.4.1 An Unsuccessful Attack on the
Intrinsicness Premise 130
4.4.2 An Unsuccessful Defense of the
Intrinsicness Premise 131
uiii Contents
4.5 Why is the Fission Argument
so Compelling? 134
4.5.1 The Casewise Explanatory
Difference Principle 134
4.5.2 The Casewise Explanatory Principle
and the Method of Agreement 136
4.5.3 Fission and the Method of
Agreement 136
4.6 How Absent Features can be
Explanatory 138
4.6.1 Human Bodies and Borrowed Luster 138
4.6.2 Explaining Valuation 139
4.6.3 Exceptions, Norms and Local
Adaptation 142
4.6.4 Prudential Concern in a World
of Fission 143
4.7 Conclusion 146
4.7.1 Summary 146
4.7.2 Larger Lessons 147
5 Conclusion 149
5.1 Factive Thought Experiments: Galileo 150
5.2 Conceptual Thought Experiments:
The Ship of Theseus 154
5.3 Valuational Thought Experiments:
Personal Identity 156
6 Bibliographies 161
6.1 Bibliography of Works Cited 161
6.2 Bibliography on Galileo, Experiment
and Thought Experiment 189
6.3 Bibliography on Personal Identity
and Identity 194
6.4 Bibliography on Thought Experiment
and Experiment 229
Index 251
Preface
0.1 REVISIONS MADE
The pages that follow represent a slightly revised version of my dis
sertation (completed in 1996 under the supervision of Robert
Nozick, Derek Parfit and Hilary Putnam). Although I have not
made major structural revisions, much of the text that appears
below differs from what originally appeared in the dissertation.
Some of the changes have been made in the course of preparing
chapters for publication as articles; others have been made specif
ically with this manuscript in mind.
The changes made to chapter 1 are minor: examples have been
slightly modified, terminology has been standardized, a few foot
notes have been added or omitted, references have been updated,
and material from (what appeared in the dissertation as) Appendix
A has been incorporated into the text. (Appendices A, B, and C,
which appeared in the dissertation, have been omitted from the
published version of this work.) Elements of Chapter 1 have been
incorporated into section IV of my "Exceptional Persons: On the
Limits of Imaginary Cases." Journal of Consciousness Studies, Vol.
5, Nos. 5-6 (1998): 592-610. The remaining revisions are specific
to this volume.
The changes made to chapter 2 are rather more extensive: sec
tion 2.2.4 has been significantly rewritten, section 2.3.4 is entirely
new, and a number of arguments have been refined throughout.
The text that appears in this volume is almost identical to that
published as "Galileo and the Indispensability of Scientific
Preface
X
Thought Experiment" in British Journal for the Philosophy of
Science, Vol. 49, No.3 (September 1998): 397-424.
Sections 3.1 through 3.6 of Chapter 3 have been edited slightly
and section 3.7 has been significantly rewritten, all specifically for
the purpose of this volume. Even with these changes, the chapter
remains unsatisfactory: although the critical sections (3.5 and 3.6)
strike me as quite sensible, the positive proposals (especially those
in 3.4.2 and 3. 7) neglect a number of important problems, and the
text as a whole fails to take account of central metaphysical issues,
particularly those raised in recent discussions of the problem of
material constitution.
A number of arguments in Chapter 4 have been considerably
revised in preparation for publication: Most of sections 4.3 and
4.5 have been extensively rewritten, and the rest of the chapter has
been revised accordingly. Chapter 5 is virtually unchanged: A few
references have been added, cross-references have been updated,
terminology has been standardized, and stylistic infelicities have
been corrected.
The volume also contains four bibliographies. The first,
"Bibliography of Works Cited" (6.1) provides references for mate
rials cited in the text itself. The three others, prepared with the
research assistance of Dave Bzdak, offer comprehensive listings of
post-1970 English-language books and articles (and, less compre
hensively, earlier works) on the topics of: Galileo, Experiment and
Thought Experiment (6.2); Personal Identity and Identity (6.3);
and Thought Experiment and Experiment (6.4)1
•
0.2 SUMMARY OF CONTENTS
This study offers a novel analysis of the widely-used but ill-under
stood technique of thought experiment. Thought experiment,
which is a process of reasoning carried out within the context of a
well-articulated imaginary scenario in order to answer a specific
question about a non-imaginary situation, is extensively employed
in fields as diverse as physics, computer science, law, and philoso
phy. What I argue is that both the powers and limits of this
methodology can be traced to the fact that when the contempla-
1 I thank the Allan and Anita Sutton Fund of Syracuse University for providing me
with a book subvention in support of this assistance.
Preface xi
tion of an imaginary scenario brings us to new knowledge, it does
so by forcing us to make sense of exceptional cases.
The study has five chapters: an introduction, three case studies
and a conclusion. My main contention is that certain patterns of
features which coincide only fortuitously may nonetheless play a
central role in the organization of our concepts, and that to the
extent that imaginary scenarios involve disruptions of these pat
terns, our first-order judgments about them are often distorted or
even inverted.
0.2.1 Chapter 1-Introduction
I begin the introductory chapter by identifying a certain class of
exceptional cases with which the volume is primarily concerned.
These are cases where some entity which falls within the purview
of a theory lacks the characteristics (which I call the proxy char
acteristics) by which entities described by that theory are ordinar
ily identified2 I contrast these sorts of theories
•
(privileged-characteristic theories) with a second sort of theory
(prototype- or paradigm-characteristic theories) where there is no
such privileged characteristic3
•
With this distinction in place, I suggest that there are two main
sorts of strategies users of a theory can employ in accounting for
exceptional cases. The first strategy is to use exceptional cases as a
way of progressively narrowing the range of privileged character
istics, thereby ascertaining the theory's exception-driven norms;
the second strategy is to use exceptional cases as evidence for the
strength of the theory's core, thereby ascertaining the theory's
norm-driven exceptions.
' So, for instance, according to some theory of sex, a person might fail to have the
clothing, body type, social role, and emotional characteristics by which women are
generally identified, but she might nonetheless be a (exceptional) woman because
she has the requisite privileged characteristics, namely those which concern repro
ductive organs and chromosomal structure.
3 A theory of gender is a theory of this second sort. What it is to he an exemplary
female is to exhibit some (maximal or near-maximal) subset of the characteristics
which together characterize paradigm cases of femaleness-but no single character
istic is privileged such that possession of it alone is sufficient to make someone
female. Correspondingly, an exceptional female is a female who exhibits few of
these characteristics.
xii Preface
In later chapters, I argue that appeal to exceptional cases is
indispensable if we wish to avoid mistaking accidental regularities
for regularities which reflect a deeper truth about the world. But I
show how the failure to attend to the distinction between theories
with norm-driven exceptions and theories with exception-driven
norms can lead to radically misguided conclusions.
0.2.2 Chapter 2: Galileo (First Case Study)
In chapter 2, I discuss the thought experiment with which Galileo
is supposed to have refuted the Aristotelian view that heavier bod
ies fall faster than lighter ones4 Through a careful examination of
•
this factive thought experiment5 I try to show that the guided con
,
templation of an imaginary scenario can provide us with new sci
entific knowledge in a way that argument alone cannot.
In the first part of the chapter, I explain the philosophical moti
vations for adopting the view that any scientific thought experi
ment can be reconstructed as a non-thought-experimental
argument without loss of demonstrative force, and distinguish two
versions of the position. The first-the dispensability thesis-con
cerns the replaceability of thought experiments; the second-the
derivativity thesis-concerns their justificatory force. In the
remainder of the chapter, I try to refute both of these theses.
Through a detailed discussion of Galileo's thought experiment, I
show that the standard argumentative reconstruction of the case
fails to capture its justificatory power, and I suggest reasons to
think that any other argumentative reconstruction would fail in
similar ways. I then argue that even if one were to provide an argu
mentative reconstruction that did almost perfectly capture the
thought experiment's demonstrative force, this would not show
that the reason the thought experiment is successful is because
deep down-it is nothing more than an argument in disguise. I sug-
4 Imagine that a heavy and a light body are strapped together and dropped from a
significant height. What would the Aristotelian expect to be the natural speed of
their combination?
5 A (active thought experiment is one where the question at hand is: "What would
happen?" Such thought experiments can be contrasted with what I call conceptual
thought experiments, where the question is: "How, given an answer to the factive
question, should we describe what would happen?" and with valuational thought
experiments, where the question is: "How, given an answer to the conceptual ques
tion, should we evaluate what would happen?"
