Buổi 82: Reading – Y/N/NG + MC: nói về sự không chắc chắn có làm mất niềm tin?

Admitting what we don’t know

(Bài đọc tự viết; nhà khoa học, cơ quan và tình huống là giả định.)

A When scientists speak to the public, they face a dilemma. If they present their conclusions with complete confidence, they risk being shown to be wrong later, which may damage trust in science as a whole. If, on the other hand, they emphasise the uncertainties, they may be accused of not knowing anything, and their advice may simply be ignored. Many people assume that the second risk is greater – that the public wants clear answers and will lose faith in experts who appear unsure. I think this assumption is largely mistaken.

B Consider a hypothetical case. A public health agency announces that a new infection is “not a serious threat”. Two weeks later, as cases rise, it reverses its advice. The agency’s original statement may well have been reasonable given the evidence available at the time. But because it was expressed with such certainty, the reversal looks like incompetence – or worse, like deception. Had the agency said instead that the threat “currently appears low, although this may change as we learn more”, the later revision would have seemed like a normal response to new information.

C This is not to say that uncertainty should be emphasised at every opportunity. There is a difference between acknowledging what is not known and burying a clear message under qualifications. A weather forecaster who says that rain is “possible but by no means certain” every day is not being honest; they are simply being unhelpful. The skill lies in being precise about uncertainty: saying how confident one is, and why, rather than retreating into vague language.

D One effective way of doing this is to give ranges rather than single figures. A scientist who estimates that a project will cost “between eight and twelve million” conveys more useful information than one who says “ten million” and is later proved wrong by a large margin. Ranges also make it clear that estimates are estimates, which may protect the credibility of the people who provide them.

E Critics argue that ranges and probabilities are difficult for non-specialists to understand, and that they may be exploited by those who wish to cast doubt on well-established findings. There is some truth in both points. Some people do misinterpret probabilities, and groups opposed to particular policies have sometimes seized on scientists’ honest admissions of uncertainty to suggest that nothing is known. But the alternative – presenting uncertain conclusions as certain – carries greater risks. When an overconfident claim is eventually overturned, the damage to trust can be severe and long-lasting.

F It also matters who communicates. People appear to trust experts more when they see them as honest and as having their interests at heart, not merely as knowledgeable. An expert who admits the limits of their knowledge may therefore seem more trustworthy, not less, provided they also explain clearly what is known and what people should do in the meantime.

G None of this will eliminate misinformation, which spreads for many reasons unrelated to how scientists communicate. But it suggests that the conventional wisdom – that the public cannot cope with uncertainty – underestimates ordinary people. Most of us live with uncertainty every day, in decisions about health, money and work. We are capable of understanding that knowledge develops over time, provided that those who speak to us are honest about where it currently stands.