Richard Van Noorden in Nature reports that 120-odd published conference proceedings are to be withdrawn, after it was pointed out that they were products of ingenious drivel-generator SCIgen.

So what, right?

Well, the hook this time is that the egg is on the faces of publishers you've actually heard of, namely Springer and IEEE, rather than the predatory types found at the villainous end of the Open Access world.

Director of corporate communications Monika Stickel is quoted as saying that IEEE have:

refined our processes to prevent papers not meeting our standards from being published in the future

Refinements that presumably include reading submitted papers.

Interestingly, Springer's Ruth Francis apparently confirmed to Nature that the offending papers were peer-reviewed. When a predatory OA journal gets stung like this, their motive for publishing anything and everything is clear—author fees. I've always assumed that those publishers don't bother with peer review at all, or that a fraction of papers are reviewed to give some veneer of respectability. However, it's difficult to see what subscription-based publishers like IEEE and Springer would gain from uncritically accepting garbage, and, assuming that these articles were reviewed, I find it hard to believe that so many referees could be so incompetent.

Working for a scientific publisher of some repute, I've seen many referee reports that leave a distinct impression of ‘nodding a paper through’ without a proper evaluation, but never the lack-of-giving-a-fuck that would be required to give a SCIgen paper the green light. Even if, like me, you're almost entirely unversed in computer science research, reading the SCIgen articles reveals them as incoherent nonsense pretty quickly. It's surely unthinkable that an expert referee would be fooled.

So what's going on? Are Springer lying when they say the papers were peer-reviewed? If so, why would they gamble their credibility for no apparent benefit? Or are there referees who, confronted with work supposedly in their field that they couldn't make head-nor-tail of, didn't have the confidence to call it out, and instead assumed they were missing something and gave a vague approval? Could that be compounded by a less-than-perfect grasp of English? I struggle to imagine being confronted with the (generally) grammatically correct yet logically challenged SCIgen text as a non-native English speaker.

If a chemistry SCIgen-equivalent existed, I'd bet the farm on its faux-papers being laughed out by the referees that I've worked with, if they ever got that far.

That said, I'm now tempted to test that theory.

Update Feb 27th, 2014: Another Nature News article covering the publishers' responses. Springer are removing the fake articles, but leaving a note on the page explaining what used to be there.

IEEE, on the other hand, are apparently just purging them, leaving nothing but a 404 behind. Not cool.