I noticed this post over at A Scientist's Life on some recent instances of retracted papers and scientific fraud. Those two phenomena aren't linked in every case, but they're often seen in each other's company. People do tend to think they're a couple.

The papers were from Science and Cell, two of the really top-shelf journals (links are at the blog above.) I suppose that makes sense, because there's really no point in faking your way into the Transactions of the Ruritanian Academy or something. It would be like counterfeiting nickles. Lou, the biology post-doc blogger, rightly says (with reference to the recent Jan Hendrik Schon case at Bell Labs):

"I've never thought about falsifying data. That goes against my education and belief as a scientist. Naive as it may be, I thought the whole point of science was to look for answers in what is already there - how nature works. I was about to state that it must be biological sciences, but physicists do it too...hey, where are the chemists then?"

Well, clearly, it's because we're too, ahem, upstanding and - haaarrgh - intrinsically honest to do anything of th-aaaackh. . .sorry, couldn't make it all the way to the end of that one. I second that thought about never thinking about faking data, of course, but much as it pains me, I can give some examples of fraud in the chemistry world. To start with, I'll reprise a post from a couple of years ago on my old site, which pointed readers to an article in the journal Synthesis (p. 29, 2002). That paper belongs to the select group of those whose sole purpose is to demolish another one.

The original, now discredited paper was in the same journal over a year before, presenting an interesting reaction that I thought we could make use of in my lab. We actually tried the chemistry out, but it flopped cleanly and completely, giving exactly the wrong product. I chalked it up to the weirdness of our compounds, which was not to be underestimated. Some things worked on them, and some just didn't. We poured the reaction into the red waste can and did something else, which is almost always an option in medicinal chemistry.

But the author of that reference I cite had the same thing happen to him, and he didn't take it as quietly. Going back over the original examples, he showed that the first published work wouldn't, didn't, and couldn't go the way it was reported. Some of the discrepencies could have been put in the "honest mistake" category, subheading "really sloppy honest mistakes," but it seems to me, in the end, that some of it couldn't. The editors of Synthesis seem to have agreed, and to their credit the criticism found its way quickly into print. The (single) author of the non-reproducible work is still listed, though, as a faculty member at the institution he published his paper from - if there have been any consequences of this affair, I haven't heard of them.

I'll dredge up some other examples in future posts. Many of them involve self-deception, at least at first. But as time goes on, the deception becomes contagious, as the originator of the suspect work realizes just how far out from shore he is. "Stepp'd in so far that, should I wade no more, Returning were as tedious as go o'er. . .", as it's put.