What is a thought experiment? Testing ideas without a lab

A thought experiment is an imagined scenario that someone follows step by step, in the mind rather than in a lab, to work out what an idea really implies. It keeps the rules of a theory fixed, sets up a situation, and asks what must happen. If the answer contradicts the theory, or contradicts something everyone accepts, the idea is in trouble. Galileo, Einstein and many philosophers changed their fields this way.

In short

• A thought experiment tests an idea by imagining a situation and reasoning out its consequences, with no equipment and no new measurements.
• The strongest ones find a contradiction inside a theory, as Galileo's tied stones did for Aristotle's physics.
• Whether they give genuinely new knowledge about the world, or only draw out what people already knew, is still argued.

How does a thought experiment work?

A thought experiment has the same skeleton as a real one, minus the apparatus. First, it fixes the claim being tested: a law of physics, a definition, a moral rule. Second, it describes a situation, often a deliberately strange one, chosen so that the claim has to say something definite about it. Third, it reasons out what the claim predicts. Finally, it compares that prediction with something else: another part of the same theory, a well-established fact, or a strong shared intuition.

The term itself is older than most people expect. The Danish physicist Hans Christian Ørsted was already using it, as the Danish Tankeexperiment, by 1811, long before the Austrian physicist Ernst Mach, often wrongly credited with coining the German Gedankenexperiment, made it a topic for philosophers in the late 19th century, according to the Stanford Encyclopedia of Philosophy. The practice is older still: ancient philosophers argued this way long before anyone named it.

Not every imagined scenario is a thought experiment. A daydream has no claim at stake. A thought experiment is built to put pressure on a specific idea, and it succeeds when it makes the idea give a clear answer that can be checked, which is why it sits close to evidence and argument rather than to fiction.

How did Galileo beat Aristotle without dropping anything?

Aristotle's physics, taught for nearly 2,000 years, held that heavier bodies fall faster than lighter ones, in proportion to their weight. Galileo Galilei took that rule at face value and followed it into a corner. He sketched the argument in his early writings On Motion and set it out in his 1638 book Two New Sciences, as a dialogue between Salviati, who speaks for Galileo, and Simplicio, who defends Aristotle (Wikipedia).

Walk through it with numbers. Assume the rule, and say a heavy stone falls at 8 units of speed and a light stone at 4.

1. Tie the two stones together with a cord and drop them.
2. The light stone falls more slowly, so it should hold the heavy one back, like a drag. The pair should fall faster than 4 but slower than 8: somewhere in between.
3. But the tied pair is a single object heavier than the heavy stone alone. By the same rule, it should fall faster than 8.
4. The rule now predicts that the pair falls both slower than 8 and faster than 8. That is a contradiction.

The rule refutes itself, and Galileo's way out of the contradiction is that weight makes no difference: all bodies fall at the same rate. No measurement was needed to see that Aristotle's rule, as stated, could not be right.

The simple picture has a limit. Air pushes back on falling objects, and a feather really does fall more slowly than a stone in air. Galileo's conclusion is about bodies falling with no resistance, which is why it took a vacuum to show it cleanly. On 2 August 1971, at the end of the last Apollo 15 moonwalk, the astronaut David Scott dropped a 1.32 kg hammer and a falcon feather together on the Moon, and they hit the ground at the same moment, as NASA records. Whether Galileo ever dropped balls from the Leaning Tower of Pisa is doubtful: the story comes from a biography by his student Vincenzo Viviani, and Galileo left no account of it.

Which thought experiments changed their fields?

The famous cases come from physics, the philosophy of mind, ethics and metaphysics. Each one turns an abstract question into a scene that forces an answer.

Thought experiment	Field	The question it asks	What it showed
Galileo's tied stones (1638)	Physics	Do heavier bodies fall faster?	Aristotle's rule contradicts itself; free fall doesn't depend on weight
Maxwell's demon (1867)	Physics	Could a tiny sorting creature beat the second law of thermodynamics?	Pointed to the link between heat, order and information
Einstein's light beam (about 1895)	Physics	What would a light wave look like if you rode alongside it?	A frozen wave made no sense, a puzzle that led to special relativity in 1905
[[what-is-schrodingers-cat	Schrödinger's cat]] (1935)	Physics	Does quantum superposition reach everyday objects?
The trolley problem (1967)	Ethics	Is it right to divert harm so that one dies instead of five?	Separates the duty not to harm from the duty to help
[[what-is-the-chinese-room-argument	The Chinese room]] (1980)	Philosophy of mind	Does following rules for symbols amount to understanding?
[[what-is-the-ship-of-theseus	The Ship of Theseus]] (ancient)	Metaphysics	Is a thing still itself once every part is replaced?

A few dates need care. Einstein recalled the light-beam puzzle in his Autobiographical Notes, published in 1949, as something he had wondered about at 16. Philippa Foot raised the trolley case in 1967, and Judith Jarvis Thomson named it "the trolley problem" in 1976 (Wikipedia).

What can a thought experiment do, and what can't it?

Thought experiments do three main jobs. Some destroy a theory by finding a contradiction inside it, as Galileo's did. Some build a theory by showing what a new principle would mean: Einstein's light beam and his later imagined elevator worked that way. Others test a concept against intuition, as the trolley problem and the Ship of Theseus do: they don't refute anything, they show where a definition or a moral rule starts to strain.

Their weakness is that they are only as good as their premises. A thought experiment can smuggle in an assumption that doesn't hold in the real world, and an argument about an imagined case can feel decisive while resting on an intuition that other people don't share. When a thought experiment makes a claim about nature, the claim still has to face the world. The strongest scientific ideas say what result would count against them, which is the point of falsifiability. And some imagined scenarios produce outcomes that seem impossible yet follow from reasonable steps, which is where thought experiments shade into paradoxes.

Still argued

Philosophers disagree about how a thought experiment can teach anything about the world without new observations. The philosopher of science John Norton argues that every thought experiment is an argument in disguise: its premises come from experience, and the scene only dresses them up, so it can reveal nothing an explicit argument couldn't. James Robert Brown, in The Laboratory of the Mind (1991), argues that in a few special cases, Galileo's among them, people grasp a law of nature directly, without inferring it from data. A third view, going back to Mach, treats thought experiments as drawing on stored experience that people have never put into words. The Stanford Encyclopedia of Philosophy sets out these positions side by side.

How do you tell a good thought experiment from a bad one?

A short checklist separates the ones worth taking seriously. Is the claim being tested stated exactly? Does the scenario follow the claim's own rules, or does it quietly change them? Is every step forced, or does one rely on an intuition people would dispute? And if the conclusion is about the physical world, can it in principle be checked by a real experiment, the way the hammer and feather checked Galileo?

A thought experiment that passes all four is a powerful tool: cheap, fast and able to expose a flaw before anyone builds a lab. One that fails them is a story with a conclusion attached.

Check yourself

• Without looking back, walk through Galileo's tied-stones argument and say exactly where the contradiction appears.
• What is the difference between a thought experiment that refutes a theory and one that tests a concept against intuition?
• Why does a thought experiment about nature still need a real experiment behind it?

Sources

• Thought Experiments — Stanford Encyclopedia of Philosophy
• Galileo's Leaning Tower of Pisa experiment — Wikipedia
• The Apollo 15 Hammer-Feather Drop — NASA Space Science Data Coordinated Archive
• Thought experiment — Wikipedia
• Trolley problem — Wikipedia