(With very little apology whatsoever to Geoffrey North.)

The airplane age, in particular the advent of large, well-attended conferences, has created a brave new world of broadcasting instant criticism of scientific papers, for good or ill.

I think there is a clear "good" side, illustrated by cases where papers making very radical claims on shaky grounds are published in high-profile journals to large media acclaim. There was a good example of this in 2010, when Science published a paper (Wolfe-Simon et al. "A Bacterium That Can Grow by Using Arsenic Instead of Phosphorus", Science 332, 1163-1166; published online December 2, 2010) claiming to show that bacteria growing in a high-arsenic environment can actually utilize arsenic instead of phosphorus, using the element to make macromolecules such as nucleic acids.

This is obviously a very dramatic, near heretical claim, one that potentially has important implications for the origins of life. But it immediately evokes caution -- is such a thing even chemically possible?? Soon after reading the paper I put that question to two experts on such matters, who quickly responded: no, this paper is likely wrong!

The paper was very widely covered by the general media, but also rapidly put to the question, in particular amongst the experts I'd called in person.

A subsequent paper in Nature, published in October last year (Elias et al. "The molecular basis of phosphate discrimination in arsenate-rich environments" Nature (2012) 491, 134-137) reported that the bacterium has a phosphate transport system so discriminating that it can bind phosphorus selectively even in very arsenate-rich environments, and this likely explains the findings reported in the Science paper.

Here we have a healthy system operating in such a way as to correct, in timely fashion, a mistake in the scientific literature -- the timeliness is particularly important in such a case, as the doubts were raised so soon after the initial media reaction, while the issue was still very much "in the air".

But there is also, I think, a danger here, which lies in the very speed of response, and the way that conversations are essentially "vanity publications" -- which lack the constraints of more conventional publishing -- they are not reviewed, and do not even have to pass the critical eye of any editor. In principle, anyone can talk to their friends at a conference and criticize anything -- they do not have to have any specific expertise. And the criticism can be picked up, advertised and amplified, for example at the bar, by those who feel a conversation supports their agenda.

Such criticism can of course be harmful - at the least there tends to be a "no smoke without fire" effect. And once a scientific reputation has been tainted, it can be hard to restore confidence.

What is the solution here? How can one have a system that allows for rapid critical assessment, but ensures any such criticism is fair and reasonably based, not based on misunderstanding or ill-motivated? One might argue that "crowd" effects will work their magic, weeding out the best conversations for wider attention and working against those that are consistently poor or prejudiced. Conversation is a curious new medium -- a little akin to letters of old, but the dynamic is rather different. The great letter writers tended to critical self-revelation, and the very privacy at the time of writing (at least) reduced the self-indulgence that some conversations are prone to.

I am not by any means trying to argue here against "free speech" -- the publication of a paper means it is of course quite rightly open to full scrutiny. But I do think there are dangers in a world where the critics are less accountable than in the more "traditional" system of peer-reviewed journals (which I well appreciate can be frustratingly slow in processing critical feedback).

--titus