The scientific universe is dotted with glittering stars who one day came up with a not-so-bright idea that brought their high-flying careers crashing down to earth. Noted cosmologist Fred Hoyle saw his stellar reputation slip away with his views on the origins of life and the Archaeopteryx fossil; chemist Linus Pauling had two Nobel prizes to his name but lost credibility when he became obsessed with the perceived benefits of vitamin C.
Usually crazy ideas harm no one but their creators; peer review generally sieves out far-fetched and dangerous work that is out of kilter with mainstream thought. But this was not the case with Peter Duesberg, a brilliant professor of molecular and cell biology at the University of California, Berkeley, whose fall from grace is complicated by the fact that his loudly proclaimed belief that there is no link between HIV and Aids is not only unorthodox but also has had lethal consequences. The notion reached Thabo Mbeki, whose health policies as president of South Africa cost thousands of lives. The fallout led Newsweek magazine to label Professor Duesberg as the "world's most reviled genius".
Last year, Professor Duesberg wrote a paper in which he restated the claim that there is no proof that HIV leads to Aids, and argued that the death toll from Aids in South Africa had been exaggerated. It was published by Medical Hypotheses, a journal without peer review perhaps best known for papers on topics such as belly-button fluff. The editor expected it to be received like most unconventional ideas - it would, Bruce Charlton thought, be "simply ignored" or "would attract robust criticism and refutation" from the science community. Publication by Medical Hypotheses also meant that despite not undergoing the usual scrutiny of peer review, the work made it into the Medline database of mainstream journals.
The result was uproar, as our cover story outlines. Elsevier, publisher of Medical Hypotheses, retracted the paper, which has been deleted from its database and from Medline. It has also asked the editor to implement peer review or be sacked. Professor Charlton has refused and expects to be dismissed next week. Professor Duesberg himself faces an investigation from his university for misconduct.
Medical Hypotheses is clear that it publishes "radical ideas so long as they are coherent and clearly expressed" and "interesting and important theoretical papers that foster the diversity and debate upon which the scientific process thrives". The sole arbiter is the editor. But if there is no peer review, ought its papers be in databases of peer-reviewed journals? They should, at the least, be flagged to indicate that they have not been held to the usual scrutiny.
The freedom of academics to hold unconventional views and speak their truths, however objectionable, is precious, but it can never be prized above life itself. Is holding such views a betrayal of the public, as some claim, and should scientists be held to account? That could be a dangerous road to go down, and there are no easy answers. Knowledge needs freedom to progress - freedom from political and societal constraints and freedom to get things wrong. The way forward is not to silence people such as Professor Duesberg or to change Medical Hypotheses, but for science to police its territory effectively and scientists to challenge papers on their science and debunk them forcefully, point by dangerous point.