Eenglish-tests.com

C2 Text 21 · Essay · 767 words

Uncertainty in Science

An essay arguing that uncertainty is not a weakness of science but the source of its strength, and that learning to live with 'probably' is a valuable habit.

There is a peculiar kind of disappointment that greets the news that scientists have revised a finding. 'So they didn't know after all,' runs the familiar complaint, as though a discipline that corrects itself had been caught out in a lie. The irony is hard to miss: the very feature that makes science trustworthy is routinely mistaken for evidence that it cannot be trusted. Uncertainty is not a blemish on the scientific method; it is the method's beating heart.

Consider the humble weather forecast. When a presenter says there is a seventy per cent chance of rain tomorrow, she is not hedging out of cowardice. She is giving a precise statement about imprecision: of all the days that look, to the models, much like tomorrow, roughly seven in ten will turn out wet. A forecaster who promised certainty would be less honest, not more knowledgeable. The same logic lies behind the error bars that adorn graphs in research papers, and behind the ranges, rather than single figures, that accompany projections of sea-level rise. A range is not an admission of ignorance so much as a map of how far our knowledge reaches, with its edges clearly drawn.

Philosophers of science have long argued that this provisional quality is built into the enterprise itself. Karl Popper, writing in the twentieth century, proposed that what distinguishes a scientific claim is not that it can be proved but that it could, in principle, be shown to be false. A theory earns our confidence by surviving repeated attempts to refute it, yet no number of successful tests can guarantee that the next one will not fail. On this view, even our best-established ideas are held, as it were, on probation.

History offers a sobering example. For more than two centuries, Isaac Newton's account of gravity seemed beyond question; it predicted the motion of the planets with astonishing accuracy. Then, in the early twentieth century, Albert Einstein's general theory of relativity showed that Newton's picture broke down in extreme conditions, such as near very massive objects. Yet it would be a mistake to say that Newton was simply 'wrong'. His equations are still used to plan the paths of many spacecraft, because within their proper limits they work superbly. Scientific progress, in other words, is less a matter of demolition than of discovering where the old maps stop being reliable.

Why, then, does uncertainty unsettle the public so? Part of the answer lies in how science is communicated. Research papers are cautious, peppered with phrases such as 'suggests', 'is associated with' and 'further work is needed'. By the time a finding reaches a headline, these qualifications have often been stripped away, so that 'may slightly reduce the risk' becomes 'prevents'. When a later study tempers the claim, it looks like a U-turn, although the original research never said what the headline did.

There is a subtler problem, too. Uncertainty can be exploited. Anyone wishing to delay action on an inconvenient finding need only point out that 'the science isn't settled' — a phrase that is technically true of almost everything and therefore, on its own, says almost nothing. The question that matters is never whether doubt exists, but how much, and whether it is the kind that ought to change what we do. A doctor who is ninety-five per cent sure that a treatment will help does not wait for the remaining five per cent before prescribing it.

Dr Helen Marsh, a statistician who teaches a course on risk at a university in Leeds, puts it bluntly. 'We don't ask people to be certain before they cross the road,' she says. 'We ask them to look carefully and act on the best information they have. Science is the same, only with more paperwork.' Her students, she adds, often arrive believing that a good scientist is someone who is always right. They leave, she hopes, understanding that a good scientist is someone who knows precisely how wrong they might be.

None of this means that every claim is equally uncertain, or that scientists should be immune from criticism. Some conclusions rest on such a vast and varied body of evidence that doubting them would require overturning half of what we know. Others are genuinely tentative. The skill, for researchers and readers alike, is to tell the difference. Learning to live comfortably with 'probably' may be one of the most valuable intellectual habits a society can cultivate. It is, admittedly, less satisfying than a confident yes or no. But confidence that outruns the evidence is not knowledge; it is merely noise with good posture.

Words in colour are explained in the glossary below.

Glossary

WordMeaning
blemish/ˈblemɪʃ/a small mark or fault that spoils how something looks
imprecision/ˌɪmprɪˈsɪʒn/the state of not being exact
error bars/ˈerə bɑːz/small lines on a graph that show how far a result might be from the true value
adorn/əˈdɔːn/to decorate; here, simply to appear on
provisional/prəˈvɪʒənəl/accepted for now, but able to be changed later
refute/rɪˈfjuːt/to prove that an idea or statement is wrong
on probation/ˌɒn prəˈbeɪʃən/allowed to continue for now, but still being watched and tested
sobering/ˈsəʊbərɪŋ/making you think seriously about something
demolition/ˌdeməˈlɪʃən/the complete knocking down or destroying of something
peppered with/ˈpepəd wɪð/containing many examples of something, spread all through it
qualifications/ˌkwɒlɪfɪˈkeɪʃənz/words or details that make a statement less strong or less general
U-turn/ˈjuː tɜːn/a complete change of opinion or plan
exploited/ɪkˈsplɔɪtɪd/used unfairly to get an advantage
bluntly/ˈblʌntli/in a very direct way, without trying to be polite or soft
overturning/ˌəʊvəˈtɜːnɪŋ/changing an accepted decision or idea completely
tentative/ˈtentətɪv/not yet sure or fixed; possible but uncertain

Uncertainty in Science: tests

Found this text useful? Share it with a friend or your class.