What to do when I think a published paper displays academic misconduct?

Question

In a paper published recently in a math journal, the main result is presented as being new, but in fact it is not new, and I'm sure that the authors knew this before submitting the paper. I have informed the journal editor who handled this paper, who told me he would forward my message to the managing editor. I have not heard anything since then. What else can (or should) I do?

For what it's worth, I am a tenured professor.

In more detail, there was a previous paper [A] which proved a certain result (1). The proof given in [A] actually proves a stronger result (2) which has a messy statement, and notes that (2) implies (1) (where (1) has a simple statement). The main result (3) of the recent paper [B] implies (1), but is implied by (2). Moreover, the proof in [B] is identical to the proof in [A] (and the authors of [B] say this in their paper), except that the proof in [B] uses a stronger auxiliary result at one step. So the authors of [B] must have been aware of the stronger result (2) proved in [A], but they do not mention (2) in their paper, and certainly do not mention that their main result follows immediately from (2). To me, this shows that the authors of [B] published their paper under false pretenses, by intentionally misrepresenting prior work. In addition, the auxiliary result mentioned above is presented as a new result in [B], even though it has been published several times and also the authors of [B] were fully aware of this fact, since they had written previous papers citing an essentially identical result from a paper [C]. Even beyond this, the paper [C] contains a result which is stronger than (1), (2), or (3), although the authors of [B] do not mention this.

If people think I am overreacting, I would be happy to hear this. It's possible that my views of proper academic behavior are not standard ones. I don't have a fully articulated philosophy about these things, I just feel like the authors did something wrong and should not benefit from doing it.

Added later: The situation has been resolved. The journal's editor informed me that, after inspecting the documentation I sent him, he has decided to require the authors to withdraw their paper from the journal. I don't know the logistics of this, but to me it seems like the appropriate outcome.

Answer 1

Let me congratulate you for seeing a publication practice that you suspect is fishy and taking the trouble to inquire about it. This is something that contemporary academia really needs: we are self-policing, so when we let these things slide it is as though we are sitting back and watching a robbery take place from the comfort of our squad car.

Here are my reactions to your story.

1. You make a very compelling argument that the authors of [B] should have known better than to try to publish their paper. The details you provided do not yet compel me to believe that the authors did so deliberately or with malicious intent. Math papers are hard to read: just because you cite one theorem from a paper does not necessarily mean that you are aware of the other theorems in the paper (you "should be", whatever that means, but that's not the same thing at all) and just because you come across an equivalent version of a theorem does not necessarily mean that you see that it's equivalent. Rather recently I got sent a draft of a paper that contained three theorems. They looked interesting, so I read more deeply. It turned out that two of the three theorems were not new. How did I find that out? By reading the papers cited in the bibliography! But one of the papers was written in a manner where it was a little puzzle to decode every single result (and the author did not shy away from stating more general theorems with more complicated statements; I can't argue with that, but it certainly does make the paper harder to read). The point of the story is that the people who sent me this paper were not bozos: the third result was very nice, and I have just written a paper with them that builds on it. Or here is an even more personal example: in my PhD thesis I combined work of Ogg with more recent work (of Ribet and others) in order to determine whether certain Shimura curves had points over Q_p. I didn't write these results up right away, and when I looked back at them a couple of years later I had to reread the paper of Ogg to refresh myself on what he had done. I was surprised to see that one of the theorems from my thesis was already proven in Ogg's paper! My proof was not the same as Ogg's, but neither was it "better" in any clear way. This may not be that different from what you're describing. These things do happen.

2. Divining the authors' true motivations does not seem to be your job as a self-deputized academic policeman (if it is, you have an awfully hard job). What matters is that a paper got published that doesn't do anything new and that in fact cites the paper that majorizes it. That's bad no matter what the authors were thinking...but when you look at it that way, doesn't a lot of the blame rest with the journal's handling of the paper? It seems like a referee who was a true expert in the literature or honest and conscientious enough to track down the citation should certainly have caught this issue and prevented the publication of the paper. The journal looks bad just as the authors do. If they are really not being responsive to this, they look even worse.

It seems more likely that the journal is looking into this but doing so slowly (this involves a lot of delicate conversations, including with the authors of the paper), and in the meantime they are not doing a good job of assuring you that they are actually doing something. You say that the matter got referred to the managing editor. If you have waited what you think is a reasonable amount of time and not gotten any further response, by all means take it up directly with the managing editor. It would be pretty surprising if you didn't get at least an acknowledgment of the issue from him/her within a few days. If you really hear nothing back, then waiting a week or so and sending a followup email conveying the fact that a total lack of response is only going to make you have to do more yourself to deal with the situation seems in order.

That "more" would involve talking publicly about the situation: e.g. bringing it up to other mathematicians you know and, depending upon the importance, raising the issue on a site like MathOverflow, putting something about this on your own webpage, writing a letter to the editor to some other journal, and so forth.

Finally, let me say that proper police work here may require giving up your anonymity. I think you should be willing to do so if it comes to that. In your comment you say that you are "[s]eriously" worried about getting a mailbomb from the authors. Seriously seriously? You think that some ethically questionable mathematicians are likely to commit a violent felony?? I find that hard to believe, but if it's really true you need to contact the actual police.