Science Puzzle
Two Explanations, One Comet
Two theories both predict the comet's path perfectly. Theory A explains it with gravity alone, a law already tested independently in a thousand other settings. Theory B explains it with gravity plus an unseen force, an invisible body, a shielding field and a hidden constant, none of which has any independent support.
No observation so far contradicts either. Why should a scientist prefer Theory A, and what would it be a mistake to conclude from that preference?
The Answer
Because every extra assumption in Theory B is doing no work. It buys no additional prediction, explains nothing that gravity alone did not already explain, and yet each one is a fresh opportunity to be wrong. Theory B is not richer than Theory A; it is Theory A carrying four unexamined passengers.
This is Occam's razor, usually stated as: do not multiply entities beyond necessity. The crucial word is necessity. The razor does not say "simple is better". It says that a component which adds no explanatory or predictive power should be cut, because it costs something (another untested claim) and returns nothing.
Now the part that is genuinely important, and routinely got wrong. The razor is not evidence. It does not show that Theory A is true or that Theory B is false. Nature is under no obligation to be simple, and plenty of correct science is baroque: the standard model of particle physics is not tidy, and epicycles were simpler than general relativity. If an observation turns up that gravity alone cannot account for, the razor offers no defence at all, and Theory A must give way.
What the razor actually is, is a rule for allocating effort under uncertainty. Faced with two theories that fit the evidence equally, adopt the one making fewer unsupported commitments, because it is the one that can be tested and broken most efficiently. It is a tool for deciding what to do next, not a verdict on what is true.
The practical tell is to ask of every component: what would change if I deleted this? If the answer is nothing, it is not part of the explanation. It is decoration, and decoration that cannot be tested is where pseudoscience does most of its living.
The principle: Occam's razor. Prefer the explanation with fewer unsupported assumptions, because assumptions that add no predictive power add only ways to be wrong. It is a rule for choosing what to test, not evidence of truth.