Can science silence the doubters on David Kelly?

 If you want to repudiate the group of people who claim that the death of Dr David Kelly has not satisfactorily been proven as suicide, despite last week's publication of the post-mortem report, the easiest and cheapest way to do so is simply to label them conspiracy theorists. Without addressing any of the evidence, the mere use of those two words belittles and dismisses more effectively than any forensic examination of their case ever could.

But be careful. If it is a fatal flaw to see any inconsistency as a sign of a sinister conspiracy, then it is just as mistaken to see any suggestion of a conspiracy as a sign of irrational paranoia. Being sceptical of official explanations, especially when they concern national security, is the duty of every rational citizen. And it is vital for a healthy democracy that some people take the time to examine them thoroughly, even obsessively.

Let us not also forget that some conspiracy theories turn out to be true. Watergate was no figment of the imagination; Ronald Reagan's administration did sell arms to Iran, its avowed enemy; even the demonised Senator Joseph McCarthy turned out to be right to think that many Soviet spies, sympathisers and informants had infiltrated the highest echelons of the US government.

It would be wonderful if we had some way of distinguishing between good and bad conspiracy theories from the outset, and there are indeed some general indicators. The worst kind do not examine all the evidence as carefully as possible, but simply attempt to accumulate as much evidence as possible on one side. This flagrant bias is actually what makes them so appealing. Confronted with a mountain of evidence that questions the official accounts of 9/11, for example, it is natural to think that something must be wrong. But the persuasiveness is just an illusion caused by the failure to balance the case with the even taller mountain of evidence on the other side.

In bad conspiracy theories, weak evidence in favour is given undue weight while strong evidence against is always explained away. Take Loose Change, the 89-minute 9/11 conspiracy film, viewed by millions online. The first fact it points out is that in 1962, the chairman of the joint chiefs of staff suggested staging terrorist attacks in and around Guantánamo Bay as a pretext for an invasion of Cuba, a suggestion that was rejected. If this adds any weight at all to the case that 39 years later, a vaguely similar plan was actually carried out, it is barely measurable. Yet the fact that George W Bush ran away while Mayor Rudy Giuliani stood on the rubble – quite strong evidence that the Washington administration was clueless – is not even mentioned. Bring it up with a believer, and you'll be told that not even Bush was in on the plot, or that he behaved as he did precisely to make it look as though he wasn't prepared.

Such one-sidedness is a sign that the conspiracy theorists have ceased – if they ever started – to conduct a sound investigation and have ended up becoming committed to a certain version of events, come what may. Often the personal or ideological motivation behind this is pretty clear. However, it would be wrong to make too much of this. The fact that someone has a strong personal motivation to believe a certain account of events will indeed distort their judgment, but it might also drive an obsessive investigation that might indeed unearth vital facts. Sometimes the truth is indeed as we always hoped it would be. That is why even the one-sidedness of such theories does not give us sufficient reason to automatically dismiss them.

Given all that, what should we make of the campaign for a formal inquest into Dr Kelly's death? Those who dismissed all the main players in this as conspiracy theorists owe at least some of them an apology. Professor Julian Bion, one of several signatories to a letter in the Times calling for an inquest, has been perfectly willing to change his mind in the light of the new evidence, saying he found the "contemporaneous report convincing". Even medical experts who are not yet satisfied seem genuinely to be concerned by failings in process, rather than being convinced of a cover-up. As Dr Margaret Bloom said: "My concerns would be allayed if due process – which I think Dr Kelly is entitled to – is carried out."

If only all those concerned about Kelly would proportion their beliefs to evidence so carefully. Human psychology being as it is, however, it seems you can't have only the best, balanced kind of sceptical investigations: some, even most, are always going to tip over into the paranoid, prematurely convinced second kind. Baseless conspiracy theories are the inevitable side-effect of the kind of sceptical inquiry we really do need. But it's a side-effect whose worst excesses are avoidable, as the good doctors in the Kelly case have shown.