Ben Goldacre has posted a fascinating article discussing the thorny issue of retractions in academic journals. Basically, some science papers contain errors which get discovered after publication. These errors can ultimately lead to a retraction of the paper. However, the lack of a clear means to publicise errors and retractions in papers – plus the reluctance of those involved to draw attention to their mistakes – can lead to faulty work remaining in the academic sphere for years. These papers are then cited in numerous other academic works. It is easy to see how such a situation can become an academic ‘virus’ which infects otherwise rigorous work with faulty sources.
To combat this, a wonderful blog has been set up by science journalists Adam Marcus and Ivan Oransky called Retraction Watch. This takes on the tedious but valuable task of following up on announcements of retractions to academic papers. These announcements are usually very terse statements which tell the reader little about the details causing the retraction. Worse, these statements tend to slip out with little fanfare, and no attempt is made to inform those authors who have cited the paper in their subsequent work. So Marcus and Oransky try to get to the bottom of the reasons for the retraction. They resort to good old-fashioned journalism and ring the relevant editors up. Yet the reaction to their enquiries can be less than cordial – as demonstrated when one editor told them that the reasons for one particular retraction was “none of your damn business”. Hmm. Not exactly an advert for scientific good practice is it?
Goldacre likens this reluctance to dwell on errors to the “publication bias” in journalism – the very human tendency to accentuate the positive, and stash the negative away behind the sofa. But is this right for science? The great philosopher Sir Karl Popper would surely not have thought so. His great contribution to the philosophy of science was the idea that scientists should welcome the falsification of their ideas, as this ultimately leads to truth through enhanced insight. Published errors – however embarrassing – are just another form of positive discovery. Conversely, a concealed or suppressed mistake is a damaging subversion of the method.
So did Sir Karl’s idealistic call for the celebration of the “wrong” lead to scientists joyously announcing every refutation and error regarding their theories? Sadly not. Recent studies have shown that the vast majority of scientists are also full-time human beings. Such creatures don’t relish having their life’s work publicly confirmed as wrong. Even if – as in the case of Retraction Watch – full disclosure can be vital for scientific integrity. The great lament of Popperism is that it is a fine scientific ideal that simply doesn’t scale to the messy world of human ego, bias and self-interest.
The problem for science is that it claims to be beyond all this. Its methods are supposed to ensure deliverance from these human biases. I believe that the existence of Retraction Watch – and some of the reactions to it – expose this claim as a worrying error. And the lack of an open means to flush these errors out is lamentable in the extreme.