Free speech is a fun problem. If we believe in it, then we have to accept that people will sometimes write or say something we think is not just wrong, but that we find offensive.
To what extent is there free speech in the science literature? Obviously there are some limits: we are restricted by the subject matter, of course. And there is also some limit on quality: one of the jobs of peer reviewers is to say when a submission fails in rigour. But there is also a grey area, where the subject is deemed not important enough, or just too wacky (i.e. it’s suggesting ideas very different from what we currently think). The latter sort of paper presents a conundrum: many way-out ideas are flat-out wrong, but a few could actually be right and paradigm-changing, even if this is not apparent. Should we not give these views an airing too?
These questions have been brought into sharp relief by the journal Medical Hypotheses.This journal has, well, a reputation. The editor in chief, Bruce Charlton, has taken a deliberately different approach to the journal. Its aim is set out in its guidelines to authors:
The purpose of Medical Hypotheses is to publish interesting theoretical papers. The journal will consider radical, speculative and non-mainstream scientific ideas provided they are coherently expressed.
The journal has a reputation for publishing some weird stuff, including AIDS denialism. Charlton has a deliberate policy of selecting articles because they raise radical ideas, rather than because they look correct. Obviously, he’s asking for some controversy, and now he’s got it. The Scientist reports: Journal editor facing axe.
Elsevier, Medical Hypotheses‘s publisher, has apparently told Prof. Charlton that he must either implement a series of changes or be let go:
In addition to instituting a peer-review system, an external advisory board assembled by Elsevier also recommends that articles on controversial subjects, such as any that support racism, not be considered for publication.
To be honest, implementing a peer-review system doesn’t sound too bad: it would still allow the editor to have final control over acceptance, but it might help improve some arguments. But telling the editor that he can’t publish articles on controversial subjects surely goes beyond what the publisher should be allowed to do to control a journal1.
Medical Hypotheses raises questions about what we should tolerate in our scientific communication. Much that is published in it is probably largely wrong. But the journal deliberately publishes this sort of material, and its reputation is well known. To change the journal as the publisher wants would be to strike at its raison d’être. It would also mean censorship of the scientific literature. Is it right to do this?
I think for most of us there are two issues at stake: there is the technical quality of the articles that Medical Hypotheses publishes (i.e. whether the arguments are logically sound, or all of the evidence considered). But there is also how radical the ideas are that are presented. I think there is a good argument for allowing radical ideas some leeway in the evidence for them when they are first presented (of course, they still speculative, but a lot of ideas have problems when they are first presented: think Darwin and his blending inheritance). Obviously a balance has to be found, and we might not agree with where Charlton has placed it.
It is clear that some ideas presented in Medical Hypotheses look morally suspect, to say the least. But morals are no reason to reject scientific ideas. And essentially this is what Elsevier are saying should be stopped. I find this wrong – even if I find the moral implications of some of the papers Medical Hypotheses publishes to be distasteful. I think we need freedom to expound on scientific ideas, even if they are wrong, and if they might make some people uncomfortable.
There is another aspect to this – why are Elsevier doing this? They’ll have difficulty claiming the moral high ground: last year it was discovered that they had published six fake journals. I can only assume it’s the bottom line they’re worried about.
1. an exception clearly has to be made if the journal is publishing something that would not be legal, e.g. defamatory material↩.