The correct way to show and tell

Scientists should shun the media until their research has been peer reviewed, argue Brian Heap and Peter Collins.

What do you do if, as a research scientist, you make a discovery that you know will be of huge public interest? If you work for a profit-making company you might tell colleagues whose job it is to turn ideas into money-making products. If you work in a university you might start talking to your PR people. In both cases you would be acting responsibly by alerting the public as quickly as possible to your work.

But you have a prior responsibility. You have to make sure that you are right, that you really have made an important discovery and that you understand it properly. This means talking to people who know as much as you do about science - exposing your work to the criticism of your peers.

The case of Dr Arpad Pusztai is in the news this week. (A mass panic over genetically modified foods was sparked after television and the press reported Pusztai's findings that rats fed GM potatoes suffered impaired growth and a weakened immune system.) The case of Dr Pusztai is a salutary example of what can happen when evaluation by television replaces peer review as the medium for scientific debate. As the Royal Society report criticising the design of Dr Pusztai's experiments concludes, it is vital that scientists expose research results to others able to offer informed criticism before releasing them into the public arena.

If you cut out the step of showing your work to your peers, you run the risk of being wrong. This is bad for your personal reputation - and it may turn out to be even worse for the public.

Of course, even if you show your peers before publishing your work, there are still risks. The reviewers may not spot your errors; they may blunt your creativity by scorning good but unconventional work. All the best scientists have had good papers turned down.

But for all the failings of peer review, every attempt to replace it ends with recognition that the alternatives are worse. One of the strengths of peer review is that reviewers are anonymous. Their critiques are communicated to the researcher but their identities are not. Anonymity is the precondition for frankness.

The keys to ensuring that peer review does not blunt creativity are to make sure that alternative places to publish and fund work exist, and to accept the risks associated with supporting some maverick research.

But the peer-review process should not be side-stepped. Scientists are accountable, both to the public to take all reasonable steps to ensure that what they tell them is right, and to their peer group, not to bring science into disrepute.

Brian Heap is vice-president and Peter Collins director of science policy at the Royal Society. The report is available at www.royalsoc.ac.uk/st_pol54. htm

* How should the academic community deal with ideas that offend mainstream opinion? Email us on soapbox@thes.co.uk