It is alright that a journal rejects a paper. There are thousands of papers rejected everyday, this is the basis of peer review.
What is strange is that a paper is "disappeared"!
What is strange is that a paper is "disappeared"!
It doesn't matter how silly the paper is, the problem is that they are treating their readers as idiots.
Who was that friend of his? Rivin, the editor who contacted him after learning about the article from his co-author, Steinberger, the editor-in-chief and founder of the journal who decided to publish the paper, or the unnamed referee who had a very positive opinion?
Still, if it should be so obvious to any casual observer that the paper doesn’t belong in that journal: how did that escape the journal’s founder and editor-in-chief? (Maybe Rivin was also impersonating Steinberger!)
I don't know exactly where I got the impression that Rivin was the person who got this paper fast-tracked, but I think probably somewhere on Joel Fish's Twitter timeline.
As I understand it: NYJM is an online-only journal that continually adds papers to each "volume", so it's not like a typical refereed journal where there's a cohort of submissions and a calendar during which the whole peer review committee downselects submissions. You couldn't, for instance, easily fast-track a Usenix paper into a conference proceedings without blowing up the program committee.
And, again: you can just look at NYJM and then at this paper and see immediately that it didn't go through peer review. It is unlikely any paper I've found in NYJM for several years back. There are practicing mathematicians on this thread saying the same thing; Tim Gowers also says as much in the blog post we're commenting on.
What do you mean? Do you expect the whole editorial board to review every paper? I think it’s usual for acceptance to be decided by one editor, based on the report from referees (in the case of the NYJM, one referee). Peer review is provided by the external referee(s), not by the journal editors (but they can also act as reviewers for submissions handled by another editor, I guess it depends on the journal’s policy).
(It is, for whatever it's worth to you, also essentially the practice at Black Hat USA, which gets thousands of submissions every year --- albeit the review board is separated out into tracks with their own covering reviewers. But BH Cryptography, for instance, has 10 cryptographer reviewers, and they reviewed everything on the track).
Can I ask, from where did you get the impression that it was typical for an individual member of a review board to get assigned a paper and pass/fail it, or for any one review board member to have the authority to accept a paper?
Editors normally invite more than one reviewer to evaluate a scientific manuscript.
Even if there were two referees instead of one, they are usually external and not members of the editorial board. And you could still believe that no peer review was done or suspect that both referees were the handling editor himself.
Apparently in this case the final decision is taken by the editor-in-chief. Other academic journals may have other policies and maybe in some cases the whole editorial board is involved in every publication but I don’t think it’s usual.
Would you argue that Social Text would, in that situation, still have some sort of duty to publish the paper? Would you argue they'd have a responsibility to give it a prominent, permanent, place in their archives to avoid accusations of having "disappeared" it? Would you argue that they'd need to air that dirty laundry in public? If they did air it all publicly and honestly, would you treat that as a form of bullying of Sokal and his accomplice(s)? Remember: that would be the powerful journal leveling accusations of deliberate fraud against an academic who submitted the paper, and accusations of, at the very least, dishonesty and violation of review processes against one of their own.
Yes, airing the problem as publicly as possible would be the right thing to do, out of respect for the readers of the journal.
Removing the paper from the publication record is acceptable, as long as it is done very openly, as a kind of "erratum".