The UK could have contained or even prevented disasters such as the BSE crisis if policy-makers had not dismissed expert scientific advice, Hugh Pennington argues.

There is a paradox. The British scientific community delivers discoveries effectively, efficiently and economically. Many of its scientists are in the top rank. Our Nobel tally, impressive by any criterion, continues to rise. Yet our record of disasters is as bad as our performance indicators are good.

Consider BSE and foot-and-mouth disease. They cost the taxpayer billions.

For a while, they made us international pariahs. Policy-makers floundered.

³ÉÈËVRÊÓƵ

They made things worse for themselves by what one official described in his evidence to the BSE inquiry as "leaning into the wind -making more reassuring statements than might ideally have been said".

Everyone remembers the ill-judged media stunt in 1990 when John Gummer, who was then agriculture minister, was filmed with his daughter Cordelia receiving beefburgers, and her peremptory rejection of hers. One of Gummer's successors, Nick Brown, had no better luck when, at the peak of the 2001 foot-and-mouth outbreak, he announced that it was "under control".

³ÉÈËVRÊÓƵ

The number of cases might have started to decline soon after that statement was made, but new ones continued to appear for a long time, reinforcing the perception that the outbreak was running wild. The essence of the paradox is that not only did these diseases break out and go on to discredit policy-makers and experts, they did so in a country better equipped with scientists specialising in them than just about any other.

A first-class state-funded foot-and-mouth disease laboratory was established at Pirbright in Surrey in 1924. It is a World Reference Centre.

A vaccine plant is located next door. One might imagine that when the government prepared its contingency plan for a foot-and-mouth disease importation, the lab and its staff would have played a central role. Not so. When the outbreak came, the defects in the plan meant that within 24 hours it failed.

It was the same for BSE. Alan Dickinson in Edinburgh invented scrapie genetics. He and his team worked out the typing scheme for fingerprinting strains of the agent. It is still the gold standard. Without it, the evidence that variant Creutzfeldt-Jakob disease is the human form of BSE would be entirely circumstantial. In the 1970s, Dickinson told the Medical Research Council that the use of pooled human pituitary glands collected post-mortem to make human growth hormone to treat children carried the risk of transmitting CJD. Sadly, his warnings, which were acted on only in part, came true in 1984. But despite these achievements, Dickinson and his team were not consulted about BSE when it first emerged. Indeed, Scottish scientists, and all workers in university departments, were deliberately frozen out.

It is plain that one explanation why policy-makers made such a mess of BSE and foot-and-mouth is that deliberate decisions had been taken to avoid using the very experts who could help the most. It must be admitted that employing such specialists can sometimes be less than straightforward, and maybe this is why there is a reluctance to use them. Nevertheless, to eschew their use was perverse. It was taking the cult of the amateur to self-destructive lengths. Even if scientists cannot agree, the time for them to have their debate, and to reach a consensus by experimentation and testing, is before a disaster strikes. The arguments about vaccination as a control measure against the pan-Asia foot-and-mouth virus strain in Britain in 2001 should have taken place before the outbreak, not during it. After all, that strain of virus, with its known propensity for crossing national boundaries and invading islands, had been around for a decade; and foot-and-mouth vaccinology is a mature science. Pirbright was monitoring it and knew more about its proclivities than anyone else.

For BSE, there was an additional and more subtle deficiency in the relationship between science and policy-making. The notion that BSE was nothing more than scrapie in cows was a reasonable hypothesis to entertain in the early days; scrapied sheep were not rare. They went into the meat-and-bone meal that was fed to cattle. Even better, humans had been eating them for 200 years and more without harm. But testing the hypothesis was difficult because diseases such as scrapie and BSE have incubation periods measured in years and because there are no simple blood tests.

There is no immune response, so there are no antibodies to measure. All this was well understood. But the scrapie hypothesis gathered strength and took a firm hold before it had been properly examined; in fact, rather little experimentation was done to challenge it, a situation that persists to the present day.

³ÉÈËVRÊÓƵ

As the years went by, evidence came in that should have raised doubts; the Gummers' foray into fast foods was in response to the death of Max, a siamese cat from Bristol. It had been killed by BSE. But scrapie does not pass to cats. Nevertheless, all doubts were swept aside. In spite of The Sun comparing BSE to Black Death, the chief veterinary officer, Keith Meldrum, said the cat was no cause for concern. The notion that BSE and scrapie were in essence the same had become a scientific "fact". It drove a policy of reassurance and "sedating" the public. It was generated by a process described in a book published in 1935 by the Polish microbiologist Ludwik Fleck, The Genesis and Development of a Scientific Fact .

³ÉÈËVRÊÓƵ

Despite being an active and productive scientist, Fleck was a pioneer social constructivist. He said discoveries were influenced by the experiences and assumptions of researchers, as was the way they were presented and received. He placed those with an interest in a particular discovery into a "thought collective". Its structure is like a Russian doll. Specialists are in the centre. He called their work "journal science": provisional, raw and cautious. General experts inhabit the next layer. Their views are "handbook science". It is simpler, more certain and more vivid. The "rounding" process continues as information is transmitted to the outer layer, educated amateurs - policy-makers, civil servants and politicians. Information transfer is two-way. The views of the outer layer influence the inner; the solidarity of the "collective" drives corroboration. It has been said that Fleck's views are of interest only because of his personal circumstances. To be a leftwing Polish Jewish intellectual and to survive the Lwow ghetto, Auschwitz and Buchenwald was remarkable. It happened only because the Germans' fear of typhus - Fleck had developed a vaccine and was forced to work on its development - was greater than their racism.

But the "thought collective" process he described keeps on being rediscovered. It antedated by many years Thomas Kuhn's "paradigms". It applies to technology as well: "Generally, the higher information is transmitted in a hierarchy, the more it gets 'rolled up', abbreviated and simplified. Sometimes information gets lost altogether, as weak signals drop from memos... the same conclusions, repeated over time, can result in problems eventually being deemed non-problems." This was the conclusion of the Columbia Accident Investigation Board in its report, published in August, on the most recent space shuttle disaster.

Fleck said that "heretics who do not share the collective mood... will be burnt at the stake". Remember Richard Lacey, the microbiologist who first raised the possibility of a link between BSE and vCJD?

Those who warned of danger before the space shuttle missions that ended in disaster were sidelined. When the known weakness in the design of the solid rocket boosters was discussed at the Challenger pre-launch conference, one senior manager was unhappy. He was told to "take off his engineering hat and put on his management hat". He did, and the launch proceeded -to disaster. With Columbia , the team examining the effects of the insulating foam that had peeled off the enormous external fuel tank and hit the shuttle at 500mph (creating a hole in the wing that led to the craft's destruction on re-entry) made numerous requests for imagery to be obtained to check for damage. Managers were not interested -such strikes had happened many times before and were classified as not posing a critical threat. Imagery requests were denied. When asked why the team had not pressed their requests harder, the engineers "opined that by raising contrary points of view about shuttle mission safety, they would be singled out for possible ridicule by their peers and managers".

It could even be said that the dossier on Iraq's chemical and biological weapons capability was produced by a "thought collective". We await Lord Hutton's opinion and his verdict on whether David Kelly was deemed a heretic and treated accordingly. Whatever the outcome, it must be admitted that even if there are big deficiencies in how scientists and other experts influence policy, we in Britain make up for the resultant disasters by conducting public inquiries supremely well. The increasing faith put in them by the public, as measured by the number of times they are called for, is inversely proportional to the growing loss of trust in politicians and the government machine. By and large, this faith is justified. In important areas of activity, inquiries penetrate the secrecy that surrounds policy-making. They have been good at dissecting and displaying institutional deficiencies in policy implementation, such as the failings of inspectorates that preceded the 1988 Piper Alpha disaster that killed 165 people -the oil rig had been given a clean bill of health 12 days before it blew up. And the 1996 central Scotland E. coli 0157 outbreak that left 17 dead, where the frequency of inspection of the butcher responsible had been reduced because the inspectors had confidence in his management.

The Aberfan inquiry into the 1966 South Wales mining tip slide that killed 144 people summarised it in these memorable words: "We found that many witnesses, not excluding those who were intelligent and anxious to assist us, had been oblivious to what lay before their eyes. It did not enter into their consciousness. They were like moles being asked about the habits of birds."

It is all very well to eulogise public inquiries. But to need them is an admission of failure. The diagnosis and the remedy are clear. Our problems have not been caused by too much science, but too little. Scientists are cheaper than judges. It is up to policy-makers to use us to the full.

Hugh Pennington is professor of medical microbiology at the University of Aberdeen. His book When Food Kills: BSE, E. coli and Disaster Science is published by Oxford University Press, £25.00.

³ÉÈËVRÊÓƵ