The Routledge Companion to Thought Experiments

Michael T Stuart, Yiftach Fehige, James Robert Brown, Michael T Stuart, Yiftach Fehige, James Robert Brown

• Publisher: Routledge
• Year: 2017
• Print ISBN: 9780415735087
• eBook ISBN: 9781351705516
• Topic: Philosophy
• Subtopic: Philosophy History & Theory

Thought experiments are a means of imaginative reasoning that lie at the heart of philosophy, from the pre-Socratics to the modern era, and they also play central roles in a range of fields, from physics to politics. The Routledge Companion to Thought Experiments is an invaluable guide and reference source to this multifaceted subject. Comprising over 30 chapters by a team of international contributors, the Companion covers the following important areas:

· the history of thought experiments, from antiquity to the trolley problem and quantum non-locality;

· thought experiments in the humanities, arts, and sciences, including ethics, physics, theology, biology, mathematics, economics, and politics;

· theories about the nature of thought experiments;

· new discussions concerning the impact of experimental philosophy, cross-cultural comparison studies, metaphilosophy, computer simulations, idealization, dialectics, cognitive science, the artistic nature of thought experiments, and metaphysical issues.

This broad ranging Companion goes backwards through history and sideways across disciplines. It also engages with philosophical perspectives from empiricism, rationalism, naturalism, skepticism, pluralism, contextualism, and neo-Kantianism to phenomenology. This volume will be valuable for anyone studying the methods of philosophy or any discipline that employs thought experiments, as well as anyone interested in the power and limits of the mind.

Part I
Selected History of Thought Experiments

1
The Triple Life of Ancient Thought Experiments

Katerina Ierodiakonou

There is no ancient Greek term corresponding to what we nowadays refer to as a thought experiment, and presumably ancient philosophers did not have our modern notion of a thought experiment. But there is no doubt that they did use thought experiments. In fact, they often employed them in ways similar to those of contemporary philosophers, that is, both for defending their own theories as well as for refuting the theories of their opponents. What seems to be particularly intriguing, though, is a third way in which thought experiments were used in antiquity, and particularly in Hellenistic philosophy, namely in order to induce suspension of judgement. The ancient Sceptics, who wanted to avoid being saddled with dogmatic opinions, made abundant use of thought experiments not in order to settle philosophical controversies but to formulate arguments of the same strength in support of contradictory beliefs. Indeed, in some cases the hypothetical scenario of one and the same thought experiment was evoked on both sides of a philosophical dispute. Thus, thought experiments were used by ancient philosophers: first, to support philosophical theories; second, to rebut philosophical theories; and third, to induce suspension of judgement. This is what I call the triple life of ancient thought experiments.

Let me start by presenting three thought experiments from different periods of antiquity and from different areas of ancient philosophical thought: The first thought experiment, which is in fact the first recorded thought experiment ever, is to be found in Simplicius’ commentary on Aristotle’s Physics (467.26–35)¹ and is the thought experiment of the man who stands at the edge of the universe trying to extend his hand or his staff. It is attributed to the Pythagorean Archytas of Tarentum in the first half of the fourth century B.C.E., who first constructed it in order to prove the infinity of the universe, but it subsequently had an illustrious history: It was appropriated by the Epicureans and the Stoics during the Hellenistic period; in late antiquity, the Aristotelian commentators discussed its uses in detail; in the Middle Ages, Thomas Aquinas, followed by a number of fourteenth-century scholars such as John Buridan, Nicholas Oresme and Richard of Middleton, used its Stoic version and at the same time elaborated on it in the light of their Christian beliefs; finally, in early modern times, Pierre Gassendi, Otto von Guericke, John Locke and Isaac Newton also referred to it, each for one’s own reasons. The general structure of Archytas’ thought experiment is the following: Imagine, Archytas says, that the universe is finite, that there is a man at its outermost edge and that he tries to extend his hand or his staff. There are two possibilities: either he can or he cannot extend it. But it is absurd to think that he cannot extend it. And if he can extend it, it means he is not at the edge of the universe; so, the same question will be raised when he moves further out, and this question will continue to be raised ad infinitum. Hence, if we assume that the universe is finite, we either reach an absurdity or conclude that the universe is infinite. Therefore, the universe is infinite.²

The second thought experiment is an equally famous thought experiment from Plato’s Republic (360b3–c8),³ in which Socrates’ interlocutor Glaukon uses a myth, namely the myth of Gyges, to show that people who practice justice do so unwillingly: Imagine, Glaukon says, what would happen if both a just and an unjust man were given a ring, like Gyges’ ring, which has the power to make them invisible and thus allows them to do with impunity what they really want to do. Glaukon suggests that both the just and the unjust man would behave unjustly, which proves his point that just people behave justly unwillingly. However, this seems to go against Socrates’ notion of justice, and Plato devotes the whole of his Republic in order to prove what he stresses at the very end of the dialogue (612a8–b5), namely that the just person would not change his behaviour, but he will continue to behave justly whether or not he has Gyges’ ring.⁴

The third thought experiment, which is admittedly less renowned, comes from a passage in Sextus Empiricus’ Against the Physicists (2.144–147),⁵ in which he attacks the basic Epicurean doctrine of the existence of atoms while discussing the possibility of motion: Imagine, Sextus says, a distance composed of nine partless places with two partless bodies moving through it from each of its end-points at equal speed. Each partless body, i.e., each atom, will evidently go through its first four places in the same time. But what happens next? There are three alternatives: (1) Both stop after occupying four places, but this is implausible since there is nothing to stop them. (2) One stops while the other occupies the remaining place, but this runs counter to the hypothesis that they are travelling at equal speed. (3) They both occupy the halves of the remaining place, but if the places are partless, according to the initial hypothesis, there is no middle for the bodies to meet in, and if the bodies are also partless, it cannot be the case that they occupy a part of the remaining place with a part of themselves. Hence, if motion is possible there are no partless bodies, i.e., atoms do not exist.

So, here are three cases which are meant to either defend or refute an ancient philosophical theory in ethics and in natural philosophy and which would nowadays unreservedly be regarded as thought experiments. But did the ancients themselves think of such cases as thought experiments? There are good reasons to doubt it. As I have argued elsewhere (Ierodiakonou 2005, 125–40), the ancients did not have a technical term for thought experiments; more importantly, it is also questionable whether they had the notion of a thought experiment. For when we currently talk about thought experiments, we rely on some notion of an experiment that we ourselves construct in thought in order to test a theory. However, it is not at all clear whether the ancients had the notion of an experiment, since they did not seem to use a special term to denote the manipulation of experience, or even some artfully planned observations, with the view to test a scientific hypothesis construed as an answer to a problem under study. So, even if we accept that in certain branches of ancient science, for instance medicine, the ancients did perform experiments that could have met the contemporary criteria of scientific experiments,⁶ they seem not to have considered what we now treat as experiments, or for that matter as thought experiments, as belonging to a category of their own.

How, then, did the ancients characterize the three mentioned cases regarded by us as thought experiments? There are ancient texts that refer to typical cases of thought experiments as “paradeigmata,” i.e., as “examples.”⁷ For instance, there is a passage from Plutarch’s Parallel Lives, and in particular from his Life of Theseus, in which he preserves the earliest occurrence of the well-known example of Theseus’ ship and refers to it as a “paradeigma”:

The ship on which he [i.e. Theseus] sailed with the youths and returned in safety, the thirty-oared galley, was preserved by the Athenians down to the time of Demetrius Phalereus. They would take away the old planks, and put new ones in their place and thus make it solid again, so that the ship also has become a controversial example (amphidoxoumenon paradeigma) for the philosophers in the context of the Growing Argument, some declaring that it [i.e. the ship] remains the same, others that it does not.

(Plutarch, Life of Theseus 23.1)

Leaving aside for the moment the intricacies of this thought experiment, let us focus on the term “paradeigma” itself: the preposition “para” indicates that something is being put, placed, thrown beside something else for comparison or juxtaposition, while “deigma” means “sample,” “pattern,” “plan,” “model,” “sketch”; and so “paradeigma” must be understood, in its literal sense, to mean “the sample, pattern, plan, model, sketch that is placed beside something else for comparison or juxtaposition.” Interestingly enough, it seems that the term “paradeigma” was first used to refer to the builder’s plan or model of a building or of a ship (Herodotus 5.62), as well as to the sculptor’s model or the painter’s rough sketches and drawings (Plato, Timaeus 28c6; Republic 500e3). And it is exactly from this sense of the term that we come to get its meaning as an example that serves as a standard by which one can judge something. But if such cases as the above were treated as examples, did the ancients classify them as a particular category of examples? Did they group them in a class of their own?

In our third example of an ancient thought experiment, namely the attack on the Epicurean thesis that atoms exist, Sextus refers to the case which he constructs as hypothetical (ex hupotheseōs/para tēn hupothesin). And it is true that in this case, just as in all such cases, we start from an imaginary or invented assumption, that is, a hypothetical scenario: Imagine a distance composed of nine partless places, with two partless bodies moving through it from each of its end-points at equal speed. Or again, imagine that over a certain period of time every single plank of Theseus’ ship has rotted and has been replaced. Besides, the language used in such examples leaves little doubt: They often start by the phrase “let it be the case that” (estō, hupokeisthō), or they use conditional sentences of the form “if this were the case, then it would make sense to think that.” Also, phrases such as “for the sake of the argument” (theorias heneka, e.g., Philo, De aeternitate mundi 48), “in theory” (logō, e.g., Philoponus, in Phys. 575.6), “in thought” (kat” epinoian, e.g., Philoponus, in Phys. 574.14, 575.8, 575.10, 575.18), and “if we were to conceive” (ei noēsaimen, e.g., Sextus Empiricus M. 3.78, 8.456, 9.431, 11.52), which are sometimes used by our ancient sources in order to introduce or characterize such examples, are clearly indicative of their hypothetical status.

More specifically, the hypothetical state of affairs in such cases could be:

• (i) impossible and counterfactual, for instance in Archytas’ or in Sextus’ thought experiments, or
• (ii) possible; when possible, it means that either
• (ii.a) human beings can bring it about, though there may be no evidence that anyone ever did due perhaps to practical difficulties or moral obstacles, for instance in Plutarch’s thought experiment about Theseus’ ship, or
• (ii.b) human beings cannot bring it about due to our lack of the relevant power, though it may be brought about through the action of some other agent, typically a divine being, for instance in Plato’s thought experiment about Gyges’ ring.

But although ancient philosophers recognized the distinctive role of the hypothetical scenarios in these and similar thought experiments, there is no evidence that they classified examples based on imaginary or invented assumptions in a special category. It is thus reasonable to think that they conceived of thought experiments as examples, though they had no notion of them as a distinct class.

This should not make us suspicious, however, about applying the notion of a thought experiment to certain ancient examples. For it seems that the ancients employed a kind of argumentation that fits in perfectly with the contemporary customary usage of the notion of thought experiments. Besides, although elementary textbooks on thought experiments often begin with, or at least include, a catalogue of Albert Einstein’s famous thought experiments, for instance Einstein’s train, Einstein’s lift, Einstein’s chase after a beam of light, Einstein himself chose as a rule not to use the term “thought experiment” (Gedankenexperiment), which had already been coined by Hans Christian Ørsted in the early nineteenth-century and used extensively by Ernst Mach; he spoke instead of “examples,” “arguments,” “analogies,” “illustrations,” “idealized experiments.” Nevertheless, since Einstein employed specific methods of argumentation that we nowadays designate as thought experiments, we generally have no qualms to make use of the notion of a thought experiment when analysing his works (cf. Roux 2011, esp. 4–19). And we should follow, I think, the same practice concerning the use of the term “thought experiments” in the case of certain ancient examples.

So, assuming that many ancient philosophers used thought experiments and regarded them as examples based on hypothetical scenarios, that is, as some kind of imaginary or invented examples, it is time now to investigate what sort of function they attributed to them. But first, let us briefly consider how contemporary philosophers understand the cognitive function of thought experiments; for instance, how Anna-Sara Malmgren presents in brief outline the standard model of the procedure and function of thought experiments, as currently used in philosophy:

the hypothesis or theory that is under evaluation states or entails some modal claim (typically a necessary bi-conditional or one-way implication) and in a thought experiment we check that modal claim against our intuitive verdict on an imaginary problem case. If the claim conflicts with our intuitive verdict, this is treated as strong evidence against the theory—indeed the theory may be abandoned as a result. We say that we found a counterexample to it. If not, this is treated as at least some evidence in support of the theory. We say that it accommodates our intuitions about the case.

(Malmgren 2011, 264)

In addition, contemporary philosophers may also resort to a thought experiment, especially in popular writings, simply as a striking illustration of a theory that is otherwise well founded, without thus attempting to establish the theory on the basis of this thought experiment. But even if thought experiments may have such an illustrative and didactic function, their main cognitive intention seems to be what Malmgren describes, namely to prove or to disprove a philosophical theory.

Ancient thought experiments, too, were formulated with the same purpose in mind, that is, to prove or to disprove philosophical theories and, in other words, to function as evidence for or as co...

Table of contents

• Cover
• Title
• Copyright
• CONTENTS
• List of contributors
• Thought experiments: state of the art
• PART I Selected history of thought experiments
• PART II Thought experiments and their fields
• PART III Contemporary philosophical approaches to thought experiments
• PART IV Issues, challenges and interactions
• Index of thought experiments discussed, mentioned or invented in this companion
• Author–subject index