Skip to main content icon/video/no-internet

There is no agreed definition of the term thought experiment. However, two important strands of usage may be discerned. In one usage, thought experiment refers to imaginary or hypothetical examples used to bolster, illustrate, or convey a theory. In another, narrower usage, it refers to the use of such examples to test a theory and thus provide evidence for or against it. There is no need to choose between these usages, but it is fruitful to keep them apart. This entry will discuss each notion in separate sections. For brevity, the term will be understood below in the sense indicated by the respective headings.

Thought Experiments as Examples

Sometimes, imaginary examples are used simply to illustrate an abstract theory or concept in a textbook or a classroom setting in economics, jurisprudence, or the sciences. Their utility for this purpose is due to the fact that they are free to abstract from messy, distracting detail and to idealize away nonsalient causal factors that would always be interfering in an actual case.

Thought experiments used for this purpose may draw on analogies or use extrapolation. Albert Einstein and Leopold Infeld illustrated the equivalence of gravitational and inertial mass with the famous case known as Einstein's elevator, in which a person will observe light bending whether the elevator is accelerating in an inertial frame or subject to gravitation. Isaac Newton imagined a cannonball's trajectory gradually widening or extending to the point where it doesn't come down at all, thus illustrating how gravitation keeps the moon in orbit.

Thought experiments may also be used to bolster a theory—without, strictly speaking, testing it—by showing its explanatory potential. Darwin had no access to the historical evidence that would go into a detailed explanation of the evolution of biological variation. But in The Origin of Species, he invites the reader to imagine various scenarios in which selection would lead to what we now observe, thereby establishing his theory as a potential explanation, contrary to creationist objections. Such use of thought experiments to show a theory's explanatory merits is not sharply distinguishable from mere illustration. In both uses, the point of the exercise is to see what would happen in the scenario on the assumption that the deployed theory is true.

Thought Experiments as Tests

Sometimes, however, hypothetical cases are invoked to check whether a theory yields the right result in them. This outcome, in turn, is taken to indicate whether the theory is true. This use of hypothetical cases for providing evidence is very common in philosophy, but it also occurs in the sciences. Galileo asked what would happen if a cannonball and a musket ball were connected by a string and then dropped. In Aristotle's theory, the speed of a falling body is proportional to its weight. Galileo argued that, in his hypothetical scenario, Aristotle's theory implies that the composite body should fall both faster than the cannonball on its own (since it is heavier) and slower (since the small musket ball should act as a drag), which is absurd.

In ordinary testing by experiment—in a wide sense, covering not only laboratory experiments but also observational testing in, for example, astronomy and sociology—a theory is tested by its logical consequences: If the theory is true, certain observations will be made given certain initial conditions. The experiment then proceeds by establishing that the initial conditions hold and checks whether the theory's predictions are right. (This simplifies matters a bit.)

...

  • Loading...
locked icon

Sign in to access this content

Get a 30 day FREE TRIAL

  • Watch videos from a variety of sources bringing classroom topics to life
  • Read modern, diverse business cases
  • Explore hundreds of books and reference titles

Sage Recommends

We found other relevant content for you on other Sage platforms.

Loading