Physics as Chess
Feynman's metaphor for how physics is done: imagine being dropped into a chess game with no rulebook. You can observe the pieces moving. By watching enough games, you deduce the rules — pawns move one square forward, bishops on diagonals, the knight's L-shape. You have discovered the rules of chess, not from instruction, but from observation.
Physics is this, at every level of depth. The "rules" are the laws of nature. We don't get to look at a cosmic rulebook; we infer the rules from what we observe. Every time we run an experiment, we are watching moves.
What the Metaphor Captures
The chess analogy does several things at once:
It explains why physics is hard. Chess rules are simple — six kinds of pieces, a handful of rules each — but a master's game is enormously complex. The laws of physics are also relatively few, but their consequences fill the universe. The complexity of the world does not imply complexity in the underlying rules.
It explains the asymmetry of disproof. You cannot prove the rules of chess by watching games; you can only accumulate confidence. But a single illegal move — a bishop moving in a straight line — refutes your inferred rule immediately. This is the logic of the experiment-as-sole-judge principle.
It captures the detective character of science. A physicist studying a new phenomenon is playing detective on a chess game they've never seen. They formulate candidate rules (hypotheses), deduce what those rules predict, and watch to see if the next moves match.
The Limits of the Metaphor
Chess has a fixed, finite rulebook. Physics might too — or it might not. We keep finding that previously inferred rules are approximations of deeper ones. Newton's mechanics looked like the final rulebook for two centuries; it turned out to be the low-speed, weak-field limit of something more fundamental. Feynman suggests there may be no end to this: the game may be "infinite" in the sense that every successful theory reveals a deeper level.
The other disanalogy: in chess, the rules are invented by humans. In physics, we're genuinely discovering something that was true before we arrived.
Connections
- experiment-as-sole-judge — the mechanism of rule-inference in the chess game
- laws-as-approximations — each "rule" turns out to be approximated by a deeper rule
- epistemic-humility — the chess player who forgets they might be wrong about the rules