Thought Experiment

On the Powers and Limits of Imaginary Cases

Tamar Szabo Gendler

• Publisher: Routledge
• Year: 2014
• ISBN:
  9781135706937
• Edition: 1
• Topic: Philosophy
• Subtopic: Philosophy History & Theory

This book offers a novel analysis of the widely-used but ill-understood technique of thought experiment. The author argues that the powers and limits of this methodology can be traced to the fact that when the contemplation of an imaginary scenario brings us to new knowledge, it does so by forcing us to make sense of exceptional cases.

CHAPTER 1
Introduction

Thinking about imaginary cases can help us learn new things about the world. This simple fact is both a commonplace, and a puzzle. It is a commonplace because it is undeniable that imaginary test cases play a central role in our investigation of the world—in legal reasoning, in linguistic theorizing, in philosophical inquiry, in scientific exploration, and in ordinary conversation. And it is a puzzle because it is prima facie surprising that thinking about what there isn't and how things aren't should help us to learn about what there is and how things are.

The goal of this study is to show that this simple fact is both less of a commonplace and less of a puzzle than it might initially seem. It is less of a commonplace because the failures of this technique are far more prevalent than its enthusiasts would admit. And it is less of a puzzle because what accounts for its success is far less mysterious than its detractors would concede. Moreover, the explanation for both its success and its failure is the same. Both the powers and the limits of the technique can be traced to the fact that when the contemplation of an imaginary scenario brings us to new knowledge, it does so by forcing us to make sense of an exceptional case.

1.1 EXCEPTIONAL CASES

1.1.1 Characterization of “Exceptional Case”

1.1.1.1 Exceptional Cases in Theories with Privileged Characteristics

What I mean by “exceptional case” can be characterized more precisely as follows. (I will give an example of what I mean in the next paragraph.) Suppose that there is some theory which, in general (i) accounts for entities or situations that are identified—by users of the theory—under more than one description or on the basis of more than one characteristic. Now take two of these descriptions such that: (ii) It is not conceptually necessary that these descriptions co-vary, and (iii) one description has privileged status within the theory, such that what the theory purportedly aims to do is to say something about those entities or situations under that description. Let us call this the “privileged description” and call the corresponding characteristic the “privileged characteristic.” Suppose that (iv) in a majority of cases, the entities or situations in question can also be picked out by some proxy characteristic which is generally a concomitant of the privileged characteristic. Then (v) An exceptional case is then one where there is some entity or situation which falls under the purview of the theory, but which lacks the proxy characteristic¹

As an example, take the notion of velocity described in Jean Piaget's famous series of experiments concerning children's conceptions of motion and speed (see Piaget 1964/1970 and discussion in Kuhn 1964). In the experiments, children were instructed to observe a pair of toy cars that were moved uniformly along a straight line towards some pre-specified goal. One of the cars was blue, the other red, and at the completion of each trial, the children were asked which of the cars had moved faster.

The children confronted three sorts of trials. In the first sort, both cars moved the same distance but one took longer to do so. So, for instance, the blue car and the red car might leave the starting line together, but the red car would move at a steady pace of p along the entire course, while the blue car moved at a steady pace of p+1. In the second sort of trial, both cars moved for the same amount of time but one covered a greater distance. So, for instance, the blue car and the red car might leave the starting line together, the blue car moving at a pace of p+1 and the red car at a pace of p, and both would stop after some specified period of time. In the third sort of trial, neither the times nor the distances were the same. So, for instance, the blue and red cars might set off together, with speeds as before, but the blue car might stop after a certain amount of time, while the red car continued on a bit longer. After each trial, Piaget asked the child which car had gone faster, and why².

The children, aged about five, tended to describe as faster the car which reached its goal first; this was so even when they recognized that the “slower” car had traveled a greater distance during the same period of time. Thomas Kuhn suggests that the child's response here manifests what he calls the goal-reaching criterion for the application of faster (Kuhn 1977, 244). That is, the child uses the term “faster” to describe the object that reaches its goal first, regardless of differences in the length of path covered, or the relative rates of motions of the vehicles.³ But in addition to the “goal-reaching” criterion, children of this age also apply a second criterion to determine which of two objects is moving more quickly. Kuhn refers to this criterion as “perceptual blurriness:” One object is “perceptually blurrier” than another if its more rapid motion can be directly observed⁴.

These two criteria—goal-reaching and perceptual blurriness— coexist unproblematically in the majority of cases. In most of the trials presented by Piaget, and in most daily experience, faster-moving objects reach their goals earlier than slower-moving objects⁵. However, as Kuhn notes:

Not very often (or the children could not have preserved the concept for so long) but occasionally nature will present a situation in which a body whose directly preserved speed is lower nonetheless reaches the goal first. In this case the two clues conflict; the child may be led to say that both bodies are “faster” or both “slower” or that the same body is both “faster” and “slower.” That experience of paradox is the one generated by Piaget in the laboratory (Kuhn 1977, 244–245).

That is, in most cases, the object that reaches its goal first is the object that moves most quickly along the designated path. Occasionally, however, the faster-moving object will begin its journey much later than the slower-moving object, or will trace a much longer path. In this circumstance, the object in question presents an exceptional case, which can be characterized as follows: (i) Users of the theory identify the faster participant (that is, the participant that has the greatest average speed) by examining how the participants look when they in are motion (perceptual blurriness), by checking to see which participant reaches the finish line first (goal-reaching), and by calculating the participant's mean speed over the course. (ii) It is neither conceptually nor nomologically necessary that these criteria coincide: the “blurriest” participant will not always reach the finish line first, nor will the blurriest participant or the participant who reaches the finish line first always be the participant who completes a given course in the shortest amount of time. (iii) What the theory really concerns is this last characteristic, namely, average velocity; this is the “privileged characteristic.” (iv) In most cases, the participant that has the greatest average velocity will also be the participant that is “blurriest,” as well as being the participant that reaches the finish line first. Hence either of these characteristics can serve a proxy for the privileged characteristic. (v) An exceptional case is one where the participant that is “blurriest,” or the participant that reaches the goal first is not in fact the participant with the greatest average velocity.

We might also consider, more briefly, a second example. A theory of sex (i) accounts for persons who are identified—by users of the theory—on the basis of more than one characteristic; for instance, women are identified on the basis of reproductive capacities, chromosomal structure, manner of dress, body shape, social role, and emotional tendencies. (ii) It is not conceptually or nomologically necessary that these descriptions co-vary: a person may have a uterus and wear men's bathing trunks, or she may work as a truck driver and have a close and caring emotional relationship with her children. (iii) One description has privileged status within the theory, such that what the theory purportedly aims to do is to say something about persons under that description. In this case, the “privileged description” is the biological description, and the corresponding “privileged characteristics” are, let us say, those that concern chromosomal structure. (iv) In a majority of cases, persons can be picked out by some proxy characteristic that is generally a concomitant of the privileged characteristic. So, for instance, a person wearing a skirt is likely to have the chromosomal structure in question, as is a person who is a nursery school teacher. (v) An exceptional case is then one where some person who falls under the purview of the theory as a result of having the requisite chromosomal structure lacks the proxy characteristics: for instance, someone who is unable to bear children, wears Harley-Davidson T-shirts, is highly muscular, is romantically involved with a number of women, works on an oil rig, and has a violent temper.

1.1.1.2 Exceptional Cases in Theories without Privileged Characteristics

Both of the examples described so far involve theories where one characteristic has privileged status (as required by (iii)). But the term “exceptional case” may also apply to cases within a theory that lacks characteristic (iii), that is, a theory that concerns entities that have some (proper or improper) subset of the cluster of characteristics that together typify the entities that the theory concerns. In the case of such a theory, we might wish to modify (iii) as follows: (iii′) some cluster of properties characterizes entities described by the theory, such that what the theory purportedly aims to do is to say something about those entities or situations that exhibit some significant portion of these characteristics. We will call entities that exhibit the designated properties “paradigm cases⁶/⁷.” In keeping with this adaptation of (iii), we might correspondingly replace (iv) by (iv′)—in a majority of cases, the entities or situations in question can be picked out by many of the characteristics which together characterize paradigm cases—and (v) by (v′)—an exceptional case is one where there is some entity or situation that falls under the purview of the theory, but that cannot be picked out in the way (iv′) requires^8/9.

A theory of gender is a theory of this sort. (iii′) Some cluster of properties characterizes entities described by the theory, such what the theory purportedly aims to do is to say something about those entities or situations that exhibit some significant portion of these characteristics. So, for instance, females are persons who occupy some particular set of social roles, exhibiting a range of characteristics that epitomize being female. Persons who exhibit some (maximal or near-maximal) subset of the designated properties are paradigm instances of females¹⁰. (iv′) in a majority of cases, females can picked out by many of the characteristics that together characterize paradigm cases. (v′) An exceptional case is one where some person falls under the purview of the theory but cannot be picked out in the way (iv′) requires, that is, some case where there is a person who is—according to the theory in question—a female, but who bears few of the attributes associated with paradigm instances of females¹¹.

1.1.2 Ways of Accounting for Exceptional Cases

There are two main ways a theory can account for exceptional cases; or, to speak a bit more precisely, 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 priv...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Table of Contents
• Preface
• 1. Introduction
• 2. Galileo
• 3. Theseus
• 4. Personal Identity
• 5 Conclusion
• 6 Bibliographies
• Index