Science and Society

Science can tell us how a disease spreads, how fast a climate is warming, or what a policy is likely to cost. It cannot, by itself, tell us how much risk to accept, whose interests to prioritise, or what to trade against what. Those are questions of value, and treating them as if data alone settled them is a common and costly mistake.

When we blur the two, debate goes wrong in both directions. Some people reject sound evidence because they dislike a policy attached to it; others present a contested value choice as though disagreement were simply scientific ignorance. Separating the two lets you argue honestly about the part that is actually in dispute.

Facts and values are different questions

The philosopher David Hume observed that you cannot derive an "ought" purely from an "is." A measurement describes the world; a decision adds a judgement about what matters. "The bridge will fail under this load" is a factual claim; "we should close it" depends on the cost of closing it and how much risk we are willing to tolerate. Good policy needs both, kept distinct.

Expertise, trust, and consensus

No one can verify everything first-hand, so we rely on expertise. Warranted trust is not blind: it rests on whether a field has a track record, transparent methods, and mechanisms for catching its own errors. Scientific consensus is the convergence of many independent lines of evidence, not a show of hands — which is why a lone dissenter is not automatically a suppressed genius, and why manufactured doubt (amplifying a fringe view to imply real division) is a recognisable tactic.

Uncertainty is not ignorance

Policy usually cannot wait for certainty. Expert bodies therefore communicate calibrated confidence — stating how strong the evidence is and how likely an outcome is — so decisions can be made under uncertainty rather than treating "not yet certain" as "unknown."

Suppose evidence shows a chemical raises a health risk at high doses. The science can estimate the size of that risk and its uncertainty. Whether to ban the chemical, restrict it, or merely label it is a further question involving costs, benefits, available alternatives, and how much precaution a society wants. Two people can accept the identical evidence and still reasonably favour different policies. Recognising that keeps the argument on its real footing.

"The science says we must do X." Usually it doesn't. The science may say X would achieve a stated goal, but "must" smuggles in a value — that the goal outranks everything traded against it. Watch for the word: it often marks the exact point where evidence quietly hands off to preference dressed as fact.

In 1974 Mario Molina and F. Sherwood Rowland published in Nature the hypothesis that chlorofluorocarbons (CFCs) — then common in aerosols and refrigeration — could destroy stratospheric ozone, the layer that shields the Earth from ultraviolet radiation. In 1985 scientists from the British Antarctic Survey (Joseph Farman, Brian Gardiner and Jonathan Shanklin) reported a large seasonal ozone loss over Antarctica, also in Nature.

The evidence identified a risk; acting on it was a societal choice. In 1987 governments agreed the Montreal Protocol to phase out ozone-depleting substances, and it entered into force in 1989. Molina, Rowland and Paul Crutzen shared the 1995 Nobel Prize in Chemistry for the underlying atmospheric science. The episode is often cited as science informing policy: research established the facts and their uncertainty, while the decision to act, how fast, and at what cost was made through political agreement — the fact and value roles working as they should.

• "If we agree on the facts, we'll agree on the policy." Value differences remain even under identical evidence.
• "Trusting experts means never questioning them." Warranted trust rests on process and track record, and stays open to new evidence.
• "Consensus is just conformity." It is the convergence of independent evidence, not a vote.
• "Uncertainty means we know nothing." Calibrated uncertainty is quantified knowledge, not ignorance.

Take a public claim ("we should ban X," "the data prove we must act") and split it into its factual component and its value component. Then name which part is actually being disputed.

Think Like a Maester: Before arguing, ask whether the disagreement is about what is true or about what we should value — the two need entirely different arguments.

1. What is the difference between a factual question and a value question?
2. Why can two people accept the same evidence yet favour different policies?
3. What makes trust in expertise warranted rather than blind?
4. What does scientific consensus actually represent?
5. In the ozone case, which parts were scientific findings and which were political choices?

Science establishes what is likely true and how sure we are; societies decide what to do about it. Keeping facts and values distinct, understanding warranted trust and consensus, and treating uncertainty as quantified knowledge lets public debate stay honest about what is really in dispute.

• Scientific Thinking in Everyday Life
• Communicating and Consuming Science