Go on, make your refutation

We now have journals of negative results, but will Bruce Charlton's dream turn into a monster?

If it is every columnist's dream to make a difference "out there in the real world", then I have just achieved my dream.

In 1987, I wrote a piece for New Scientist advocating a "Journal of Negative Results". It had the desired effect - it just took 18 years. But now, the Journal of Negative Results in Ecology and Evolutionary Biology has come into being - and I have been elected honorary patron.

Once the glow of ego-gratification had subsided, I considered why my wishes had come true, why it had taken so long and what might happen next.

In 1987, I suggested that the understandable desire of journal editors to publish "positive" results that confirm or suggest new things was having an unfortunate effect on the conduct of science.

Negative results, on the other hand, tend not to make it into print. But a rigorous failure to replicate findings is potentially of great importance because, as the philosopher Karl Popper famously suggested, positive results are only provisionally true, and most will eventually be refuted.

And if results are wrong, the sooner they are refuted the better. Negative results minimise the time and energy wasted in trying to pursue avenues of research that have already been shown to be blind alleys.

It is not hard to see why there is a problem. There is the perception that negative results are dull, and to some extent this is justified.

A bigger obstacle to publication is their regicidal potential - they usually imply that some high-profile scientist has got it wrong. What are the chances of your paper being published if the people who have the most to lose are exactly the people most likely to referee it?

The loss is felt by all. In the past, personal networks informally disseminated news of negative results that might not have been published.

But the scale of modern science is so vast that networks of gossip have lost their effectiveness. And anyway they always discriminated against less prestigious researchers who were out of the loop.

Most crucially, information carried on the grapevine was soon forgotten and lost for ever. It is clearly preferable that negative results be permanently archived in a form that is widely accessible.

Despite all these good reasons for publishing, the fact remains that scientists whose research apparently disproved the validity of earlier methods or results frequently had a hard job communicating their findings to their peers.

After all, they had to be confident that their failure to replicate findings was not due to their own incompetence. If you can't perform a technique properly, it won't work.

Iconoclasts also need to demonstrate that they have gone to great lengths to give their research a chance to yield positive results. Consequently, papers reporting negative results will tend to be long and detailed.

This probably explains why it is only now that web-based journals of negative results are beginning to emerge in several disciplines. Internet journals are not limited for space in the way that constrains paper publications.

So JNR-EEB is only one of a new wave of negative result web-zines spanning fields as diverse as speech, psychology, biomedicine and computer science.

But if the idea of publishing negative results catches on and becomes even more fashionable, it may lead to a situation in science akin to that in history - where "radical" historians concentrate on revolutions and social change while "revisionists" stress stability and continuity.

Perhaps one faction of somewhat credulous "positive" scientists will spend their time erecting speculative theories and generating spurious results while an army of "negative" revisionists will gleefully demolish their pretensions. The thought strikes me that I - like Dr Frankenstein - may have created a monster.

The cult of negative results might trigger scientific schism. And, as a radical and speculative theoretician, I might even find myself targeted by that new breed of revisionists that I helped to foster.

Bruce Charlton is a reader in evolutionary psychiatry at Newcastle University.