An imperfect science

I agree with Sir Peter Medawar (1915-1987), who said "In terms of fulfilment of declared intentions, science is incomparably the most successful enterprise human beings have ever engaged upon". However, despite its great worth, the scientific enterprise is not perfect, a fact often ignored by those who promote it. We should not idealise science.

The function of science is to provide natural explanations of the natural world. Towards this end, it uses the ‘scientific method’, a sort of formalised common sense. The scientist first reviews current knowledge of some as-yet-unexplained phenomenon and formulates a hypothesis (an informed best-guess) as to the explanation. This hypothesis, which must be refutable, is used to make a prediction about the phenomenon, which is then tested by experiment. If the experimental results support the hypothesis, further tests are devised. If all tests continue to support the hypothesis, confidence in the hypothesis grows and eventually the hypothesis is elevated to the status of a theory – science’s best explanation.

A scientific theory is never considered finally proved. It is always revisable in the light of new evidence. Nevertheless, we can have great confidence in the theories of science, eg the theory of evolution.

It is often claimed that science is absolutely objective: absolute objectivity is unattainable by human beings. Nevertheless, scientific knowledge comes closer to objectivity than any other type of knowledge.

One reason why absolute objectivity is unattainable is the ‘background noise’ encountered when doing experiments. Readers who listen to BBC Radio Four on long wave are used to background noise – the hiss that is devoid of information. When signal-to-noise ratio is low, background noise must be screened out to detect the signal, but the extent to which you screen it out is a subjective decision.

But there are several more easily appreciated reasons why the scientific enterprise is less than perfect. Science is a three-legged stool – a method carried out by scientists that must be supported by society. Scientists are prey to all the frailties that beset human beings and the support sources on which research depends each have needs they want science to satisfy.

Mistakes and hoaxes
Scientists make mistakes, mostly minor, but sometimes serious. Science is self-correcting but some mistakes can cause much harm before they are corrected. Eugenics, the improvement of human stock by selective breeding, is an example of a serious mistake that caused much harm, culminating in the catastrophic Nazi policies of racial hygiene. Cold fusion, reported in 1989 by Stanley Pons and Martin Fleischmann, was a mistake that was quickly corrected by science.

The above were professional errors, but scientists also commit deliberate hoaxes and fraud. Fraud has always existed (eg Piltdown Man hoax 1912, Cyril Burt’s studies in the 1950/60s on inheritance of intelligence) but it is now on the sharp increase, particularly data-fabrication. Not only career advancements in science, but also status, reputation and monetary support depend on producing regular scientific publications. Research productivity proceeds at an uneven pace, sometimes going through extended fallow periods despite great efforts. Hence the temptation to fabricate data.

Science is expensive. Funding comes from Government, industry and charitable trusts. Each source has its agenda, but serving such agendas is not always in science’s best interest. For example, Government is permanently interested in using science to generate jobs in the near term. This is understandable but can force science into unproductive cul-de-sacs. And in times of national emergency, Government can dragoon science into weapons projects, such as the Manhattan Project (1942-1946) that developed the atomic bomb. And any industry (eg tobacco or fossil energy) that funds research doesn’t look kindly on outcomes that criticise its products or activities.

Our world is now entirely dependent on science-based technology and thus public understanding of science is vital. There is a temptation amongst scientists (myself included) to present an idealised picture of it as an infallible truth-producing mechanism. This is a mistake. Putting science on a pedestal of purity is not a good idea because a pedestal that wobbles intermittently shatters public confidence.

William Reville is an emeritus professor of biochemistry and public awareness of science officer at UCC. understandingscience.ucc.ie