Sitting in front of a row of MPs this month was Sir Walter Bodmer, a geneticist who is director of the Imperial Cancer Research Fund and chairman of the Committee on the Public Understanding of Science. He had been summoned to tell them about the new genetics, but he objected to their questions about some of its more frightening parts.

"This is a totally premature discussion," he said. "There's a lot of public confusion about these issues. The public understanding of science is absolutely the key issue."

The MPs looked baffled at his implication that a set of thorny moral problems could best be tackled by distributing basic scientific information. Sir Walter believes that an increase in public understanding of science will lead to an increase in public trust in science.

Other scientists are making similar calls to be trusted. Geneticist Ruth McKernan wrote in the Independent this month: "The message that we need to get over is that science is well looked after in the hands of responsible and trustworthy individuals."

ÖйúAƬ

But do scientists deserve the trust they are angling for? No. Scientists think the fact that most of them are pleasant, well-meaning people is a good reason to trust them, but trust is not earned by being nice or well-meaning or by regularly wearing woolly jumpers, as scientists who appear in public are sometimes told to do.

Such people are unsuspicious about the world and therefore easily exploited. Scientists are often naive about the world outside science, perhaps because they stop learning about its social side at the age of 16.

ÖйúAƬ

Brian Wynne, of the Centre for the Study of Environmental Change at Lancaster University, has demonstrated the "naive sociology" that scientists use when they assess how their proposals will operate within the wider world. They make "naive assumptions of an ideal world of operation, inspection, management or maintenance," he says. They think the system will work perfectly, taking no account of the disasters caused, for example, by human error or sabotage.

But there is a second, more fundamental reason why scientists cannot be trusted. Scientists demand, as a condition of their work, freedom of thought and expression and the licence to explore anywhere that the pursuit of knowledge may take them. This may lead them to conclusions that upset moral codes or inadvertently help groups trying to oppress parts of society.

If scientists demand this freedom they cannot also expect to be trusted. Being mistrusted is the price paid by all innovators. The people who are trusted in their working roles are doctors, priests, counsellors, and so on, who have given up their rights to freedom and fenced themselves into a smaller area of operation. They have underwritten their work with a code that puts the good of the individuals they serve above aims such as increasing knowledge.

Dr McKernan protested that doctors are more trusted by the public than scientists are. She said that this was because breakthroughs in medicine are attributed to those who adminster it rather than to the scientists behind it. This is not true. Doctors are trusted not because of the knowledge they possess but because they are regarded as safe repositories of that knowledge. They are committed to using the knowledge only for benevolent, or at least harmless, ends.

ÖйúAƬ

Some scientists have argued that they should set themselves a code of ethics in order to increase trust. And Lewis Wolpert, professor of biology as applied to medicine at University College London, has explored the limits of scientific responsibility by arguing that scientists must always tell the public about their discoveries.

This is at once a little and a lot to ask. It absolves them from duties such as refraining from certain research because it is likely to be used for damaging purposes. There are strong arguments that scientists should not commit themselves to such restraints. But without such constraints they cannot demand that they should be trusted.

Professor Wolpert's suggestion must presumably require scientists to break confidentiality contracts agreed with industrial funders. To expect university scientists to tell you what they are up to when they have signed secrecy clauses is to expect a radically new code of ethics. Again, there are strong arguments that scientists cannot expect to reveal to the public what they are doing. But, unless they do, they cannot expect the public to trust them.

SET95, the national science, engineering and technology week, is currently taking science to the people. Scientific literacy is useful to the country but for Sir Walter to assert that it is "absolutely the key issue" for tackling public mistrust of science-based controversies is unbalanced. Scientists should not be indignant that they are not trusted. It is a patronising response to a public that often has a much better understanding than they do of the way in which scientific discoveries are handled. For scientists really to merit the trust they are asking for they would have to change what they do to the extent that it would no longer be science.

ÖйúAƬ