Truth, Hypothesis, Law, and Scientific Theory

Truth, Hypothesis, Law & Theory — Four Words, Four Jobs

Philosophy of Science

“Just a theory.” Hardly.

Few words cause more confusion between the lab and the living room than fact, hypothesis, law, and theory. In everyday speech a theory is a hunch; in science it is the opposite — the strongest knowledge humans possess.

↓ The ladder that isn’t ↓ Skip to misconceptions

These four words are not a ladder of promotion. They are four different jobs in one enterprise: knowing things reliably.

Ali Reza Rashidi7 min readUpdated 6 August 2026

01—The ladder that isn’t

Four jobs, not four ranks#

School diagrams show hypotheses graduating into theories, and theories into laws once the evidence is overwhelming. Real science has no such promotion track. A hypothesis never becomes a theory; a theory never becomes a law. They differ in what they do, not in how much support they have.

Fact / truthA confirmed observation: the world was measured, repeatedly, and this is what it showed.

HypothesisA tentative, testable proposed explanation — built to be proven wrong, and useful either way.

LawA description of what nature does — a pattern so consistent we can write it as an equation.

TheoryAn explanation of why the pattern exists — the broadest, best-tested knowledge science has.

230+ years Newton‘s law of gravitation reigned before Einstein’s theory explained what it describes

4 distinct jobs — fact, hypothesis, law, theory — with zero promotions between them

1 failed prediction is enough to send any hypothesis back to the workshop

0 scientific ideas ever “proven” beyond all doubt — survival under testing is the currency

Same enterprise, different shapes

Fact / truthHypothesisLawTheory

How to read this: a fact is narrow but rock-solid; a theory is broad and rock-solid. That combination — breadth with evidence — is exactly what “just a theory” misses.

Term	Everyday meaning	Scientific meaning	Canonical example
Fact / truth	“I just know it’s true”	An observation repeatedly confirmed under stated conditions	Water boils at 100 °C at sea level
Hypothesis	An educated guess	A testable, falsifiable proposed explanation	“Plants grow better with compost” — then measure it
Law	A rule that can’t be broken	A mathematical description of what nature does	Newton’s law of gravitation; Charles’s gas law
Theory	A hunch, not proven yet	A comprehensive, evidence-backed explanation of why	Evolution; germ theory; general relativity

02—The four words

Each, properly defined#

Strip the schoolbook mythology away and each word turns out to be a precise instrument. Here is what each one actually does for a living.

A fact is an observation confirmed so reliably that we proceed as if it were certain — while staying, in principle, revisable. Better instruments have revised “facts” before, and that humility is a feature, not a bug. What facts never do is explain: a thousand measurements of falling apples explain nothing by themselves.

Facts are

Confirmed observations: measured repeatedly, independently, under stated conditions.
The raw material: hypotheses, laws, and theories are built from — and judged against — facts.

Facts are not

Explanations: a fact tells you what happened, never why.
Immortal: new instruments can revise old facts; science updates without apology.

Facts don’t speak for themselves.They sit in notebooks until a hypothesis gives them a job to do.

A hypothesis is a tentative answer with three non-negotiables: it must be testable (an experiment exists), falsifiable (some outcome would kill it), and specific (it predicts more than it explains away). “This detergent removes grass stains better” earns the title; “good energy helps plants” never will — no test could prove it wrong.

Anatomy of a good one

Testable: names a measurement, not a mood.
Falsifiable: states the result that would refute it.
Predictive: says what you’ll see before you look.
Tentative: held firmly, let go easily.

A hypothesis is a bet on reality.The more precise the bet, the more you learn when it loses. Vagueness is how ideas avoid ever being wrong — and ever being useful.

Laws describe. Newton’s law of universal gravitation tells you exactly how two masses attract — F = Gm₁m₂/r² — and says nothing about what gravity is. Laws are usually born by induction: observe enough cases and the pattern compresses into mathematics. They also have domains: Newton’s law fails near light speed and in strong fields, which didn’t make it “wrong” — it mapped its own boundary.

A law tells you

What happens: gases expand when heated — Charles’s law.
How much: attraction scales with mass, inversely with distance squared.

A law never tells you

Why it happens: that job belongs to theories.
Where it stops: domains are discovered later, often by a theory.

Newton described gravity’s behavior for over two centuries without knowing what gravity is.Description and explanation are different jobs. Einstein’s general relativity — a theory — finally supplied the why: mass curves spacetime.

Theories explain why, and they do it broadly: germ theory unifies hygiene, vaccination, and antibiotics; evolution unifies fossils, DNA, and finch beaks; general relativity unifies falling apples and bending starlight. A theory is not a hypothesis that got promoted — it is a different kind of thing: not one testable claim, but a framework that generates thousands of them.

A theory must

Explain known facts: all of them, not the convenient subset.
Predict new ones: Einstein predicted starlight bending; Eddington measured it during the 1919 eclipse.
Survive attacks: every generation of scientists tries to break it.

A theory never

Becomes a law: explanation and description are different jobs, not ranks.
Claims absolute proof: no such status exists in science.
Gets ‘demoted’: superseded theories still work in their domain — Newton still builds bridges.

“Just a theory” gets it exactly backwards.When someone says evolution is “only a theory,” they are accidentally placing it at the highest rank of human knowledge — alongside gravity and germs.

03—The method

How the words work together#

Science is a loop, not a ladder. Facts trigger questions; questions breed hypotheses; experiments kill the weak ones; and the survivors settle into two kinds of knowledge — descriptions and explanations.

01Observe a factThe world shows you something repeatable. Measure it carefully.

02Ask whyCuriosity turns an observation into a question worth testing.

03Propose a hypothesisA specific, testable, falsifiable candidate answer.

04Try to kill itExperiments designed to break the idea — not to confirm it.

05Publish the survivorsPatterns that hold become laws; explanations that hold become theories.

Why “try to kill it”?Confirmation is cheap — any idea can find supporting evidence if you only look for it. An idea that survives genuine attempts to falsify it has earned something confirmation can never give: tested strength.

04—The misconceptions

Kill these four ideas#

Most confusion about science is really confusion about vocabulary. Four myths do nearly all the damage.

Name that concept

1

“Hypotheses become theories become laws.”

The ladder myth. These are different job descriptions, not ranks — an explanation no more turns into a description than a map turns into a territory.

2

“It’s just a theory.”

In science, theory is the summit — evolution, germ theory, and relativity sit there. The everyday meaning (a hunch) is the exact opposite of the scientific one.

3

“Laws outrank theories.”

Laws describe; theories explain. Neither outranks the other — general relativity explains what Newton’s law describes, and both earn their keep.

4

“Science proves things absolutely.”

Proof belongs to mathematics. Science offers something more honest than proof: ideas that have survived every attempt to falsify them — so far.

One sentence to keep: laws tell you what nature does; theories tell you why; hypotheses are how we find out; facts are what we found.

05—Grounding

Sources#

The definitions here are not my invention — they follow the standard references below, and the framework follows Berkeley’s Understanding Science project.

UC Museum of Paleontology, University of California, Berkeley. Understanding Science: How Science Really Works. The NSF-funded reference this page’s framework follows: science as a dynamic, social, revisable loop — not a linear textbook recipe. undsci.berkeley.edu
National Academy of Sciences and Institute of Medicine (2008). Science, Evolution, and Creationism. Washington, DC: The National Academies Press. The canonical definitions: a fact is a confirmed observation, a hypothesis a tentative testable statement, a law a descriptive generalization, and a theory “a comprehensive explanation of some aspect of nature that is supported by a vast body of evidence.” doi.org/10.17226/11876
Popper, K. (1959). The Logic of Scientific Discovery. Routledge. (First published in German, 1934.) The falsifiability criterion behind the “try to kill it” section: an idea no test could refute explains nothing. en.wikipedia.org/wiki/The_Logic_of_Scientific_Discovery
Kuhn, T. S. (1962). The Structure of Scientific Revolutions. University of Chicago Press. Why science “updates without apology”: normal science, anomalies, and the paradigm shifts that revise yesterday’s facts. en.wikipedia.org/wiki/The_Structure_of_Scientific_Revolutions
Dyson, F. W., Eddington, A. S., & Davidson, C. (1920). “A Determination of the Deflection of Light by the Sun’s Gravitational Field, from Observations Made at the Total Eclipse of May 29, 1919.” Philosophical Transactions of the Royal Society A, 220, 291–333. The expedition that measured starlight bending and put general relativity on the map. en.wikipedia.org/wiki/Eddington_experiment

Ali Reza Rashidi

Ali Reza Rashidi, a Senior Data Scientist-Gen Al | Al Architect | MLOps with over ten years of experience, He is the author of three books that delve into the world of data and management.

Truth, Hypothesis, Law, and Scientific Theory

Skill in Agentic Coding

Step by Step LLM Evaluation

“Just a theory.” Hardly.

Four jobs, not four ranks#