09 March 2007

A cautionary tale

It came as something as a shock to me to find that respected government scientists can lie, that they can falsify evidence and break all the rules of science – and that government officials will quite deliberately seek to cover-up the misdeeds.

If that is something of a naïve statement, so be it. But my background as a technician rather than a front-line scientist puts me closer in touch with the bulk of ordinary people, and my own personal experience of scientific fraud has important lessons for the current global warming controversy.

The fraud of which I write arose in the context of a then famous outbreak of salmonella food poisoning in 1998, claimed to be "caused" by infected eggs.

And while, at first sight, there would appear to be little in common between this case – and what became the famous Salmonella-in-eggs scare of 1988 (which is now a distant memory on the minds of most people)- and global warming, the dynamics are exactly the same. We are dealing with examples of what sociologists call the "moral panic" but which, actually, is a very specific phenomenon called the "scare".

Booker and I are writing a book on this subject (actually he is doing the writing), due out later this year. It is called Scared to death, and it charts the rise of the phenomenon in the 20th and 21st Centuries. Amongst our findings are that "scares" are man-made constructs and part of the thesis we explore in the book is that, for a scare to succeed, it must have a very rigid and predictable structure, and follow very specific rules.

In that respect, the 1988 salmonella in eggs scare and the current global warming scare are identical, not least in being triggered by flawed science promulgated by dishonest scientists.

That brings me neatly back to my own experience where I was called upon to re-investigate an outbreak investigation in an East Yorkshire farm – oddly enough at a fund-raising event for the Conservative Party, with the food made by the farmer’s wife and other Party members. And, at the centre of the outbreak has been home-made vanilla ice-cream, for which fresh eggs had been used.

The investigation had been carried out by a respected consultant epidemiologist who, at the time, had worked for an organisation within the (then) Public Health Laboratory Service (PHLS) called the Communicable Disease Surveillance Centre. So hugely respected was it that its findings were generally accepted without question, especially by people like myself, who were working in the field on preventative hygiene.

In fact, I recall in November 1988, before the scare broke, being asked by a chef whether there was a problem with eggs. I did not know, personally, but the fact that the PHLS was saying there was one was enough for me. That was the information I conveyed back.

Anyhow, the outbreak in question, which occurred in April 1988, was one of the key events which sustained that claim that we were being confronted with a new and deadly form of salmonella, which had managed to invade eggs and was causing a massive upsurge in poisoning cases.

When he arrived at the farm to conduct a field investigation, however, the epidemiologist had told the farmer's wife that, "I'm almost a hundred percent sure it was caused by eggs". He then proceeded to "prove" just that.

But there was a slight problem with his thesis. The event had actually been styled as a "paté party" and several home-made patés have been made by villagers for the event, one of which was made with chicken livers from a broiler flock. This, subsequently, had been found to be contaminated with exactly the same type of salmonella as had infected the guests.

Thus, when the epidemiologist carried out a statistical analysis of the data, using a technique called a "cohort study", three of seventeen people who had been ill had not eaten the ice-cream but had eaten a paté. In a private letter to the farmer, he was forced to conclude that the source of the infection "was impossible to prove epidemiologically".

The trouble was that it did not end there. Some months later, an outline of the investigation appeared in the authoritative Communicable Disease Report, attributing the source of infection to eggs, details of which were subsequently published in the equally authoritative Epidemiology and Infection and the British Medical Journal, later forming part of the official government evidence.

In the published account, however, there were important changes to the data. The epidemiologist had removed any reference to the sufferers who had not eaten the ice-cream and had re-worked the calculations, which now demonstrated a very strong statistical association between the consumption of a product made with raw shell-eggs and the illness.

On that basis, the outbreak was submitted as official government evidence to the House of Commons select committee investigating the affair and became subsumed, unchallenged, as part of the core evidence "proving" the link between salmonella and eggs.

It was some time later that I was able to assemble the evidence of the fraud and, thus armed, was invited to deliver my evidence to a major conference in London.

I was the last speaker of the day and my session chairman was a senior official from the Department of Health. He let the penultimate speaker run grossly over time so that, when my turn came to speak, I was told to speak to the timetable so that the conference finished on time. Thus, just as I was coming to the critical details, the chairman stood up and attempted to close the conference, precipitating a blazing row. But he prevailed, and the evidence was never heard from a public platform.

The epidemiologist went on to head the Scottish Centre for Infection and Environmental Health, where he remains a respected scientist, on the back of whose work an estimated 9,000 poultry keepers went out of business and inestimable lives were ruined, at a cost to the economy of hundreds of millions of pounds.

An important part of the scare mechanism at the time, however, was the media, which accepted uncritically what amounted to scientific fraud. Equally, politicians of all sides were only too keen to accept a body of evidence which, even when it was first produced, never stood up (with another outbreak, for instance, attributed to "eggs", where the illness had started three days before the implicated eggs had been consumed).

Thus it is with global warming, where – as the Channel 4 documentary yesterday shows – the evidence simply does not stack up. More to the point, the failure of the global warming advocates to provide any good evidence of their claims is so transparent that, if it was not so serious, the whole affair would be laughable.

The problem though is not scientific. When apparently authoritative scientists stand up and make claims, supported by a rent-seeking media, people tend to believe them. Moreover, because such claims invariably support the interventionalist tendencies of governments and politicians, there is a natural bias towards accepting that which legitimises the intervention. This is the "beneficial crisis" to which we have referred so often.

With no countervailing force, we get the build-up of the scare dynamic which then dominates public policy, even (or especially) where the scientific foundation is hopelessly flawed.

In the fullness of time, the scare will dissipate – scares always do – leaving a trail of wreckage behind it. Looking back, we will view the claims of pending Armageddon with amused puzzlement, wondering how people could have been so stupid as to have accepted such crazy alarums.

By then, of course, we will all have moved on to yet another scare, and another, each of which will have seemed every bit as plausible and rational as did global warming at the time. And each time we will have forgotten how easily we were gulled by that which we now deride.