The Scientific Process

Physics is not a list of facts to be memorised. It is a method for deciding which explanations to trust, and the method matters as much as the results.

Theory — a proposed explanation for something we observe. A theory is not a guess; it is an attempt to say why something happens, and it is judged by how well it survives testing.

Model — a deliberately simplified version of a real situation, used because the real thing is too complicated to handle. Treating a person as a cylinder to estimate their volume, or a gas particle as a perfectly elastic sphere, both work this way.

Hypothesis — a specific, testable prediction that follows from a theory. This is the part you can actually check. "Light is a wave" is a theory; "light passing through two narrow slits will produce bright and dark fringes" is a hypothesis, because an experiment can disagree with it.

Example: a hypothesis must be capable of being wrong. "There is an invisible force that has no measurable effect" predicts nothing, so no experiment can test it — which makes it useless as science, however true it might feel.

Peer review — before publication, a report is sent to other scientists working in the same field, who check that the methods are sound and the conclusions follow from the data. It filters out obvious errors and reduces the effect of personal bias. What it does not do is prove the work correct: reviewers see the report, not the laboratory.

Validation — the real test comes afterwards, when other scientists reproduce the work or use the theory to predict something new and check it. A result that nobody else can reproduce does not stand, however respectable its source.

Example: in 1989 two chemists announced they had achieved nuclear fusion at room temperature. The claim was published and widely reported. Laboratories around the world tried to repeat it and could not, and the claim collapsed — not because of who made it, but because it failed reproduction.

Why scientific knowledge is always provisional — a theory that has survived every test so far is treated as correct for now, and used with confidence. It is never proved beyond revision, because the next experiment might disagree. Newton's laws were not wrong for two centuries and then suddenly wrong; they were, and remain, an excellent description within their range, and relativity showed where that range ends.

Evidence inside and outside the laboratory — a laboratory lets you hold every variable fixed except the one under test, which is what makes a clean conclusion possible. Many important questions cannot be studied that way. There, the best available approach is a well-designed study comparing groups matched as closely as possible, accepting that some differences will remain uncontrolled.

Science and decision-making — scientists establish what is the case; societies decide what to do about it, and those decisions weigh more than the physics. Acting on a finding may be too expensive to be practical, may demand changes people are unwilling to make, or may carry environmental costs of its own. Disagreement about what to do is not the same as disagreement about the evidence, and confusing the two makes for bad arguments.