Subject: plagiarism and dead areas From: "Claude Schochet" Date: Wed, 22 Dec 2004 10:35:38 -0500 To: "'Don Davis'" My old friend Marty Tangora says: >>I know of two egregious cases in algebraic topology. >>One was plagiarism pure and simple. >>The other was a senior professor, A, >>visiting a younger mathematician, B, >>and giving an ultimatum that A must be added as a co-author to a paper by B. In >this case the conditions were actually much worse than just senior over junior, but I >don't want to give In my opinion both cases that Marty brings up are covered by the Guidelines and COPE would have taken both cases and tried (with its very limited powers) to help out. (BTW, when I was chair of COPE I would ask a complainant what he wanted COPE to do if COPE agreed that he had been wronged. Ask yourself for a moment what the AMS could actually do in Marty's second case. It's not so simple. How does one protect the junior author in years to come against poison pen letters of recommendation?) On the general issue - 1. We all agree that plagiarism is wrong. There are two difficulties, though. a) Most plagiarism is not deliberate. We all know examples where person A reproves a result without being aware that it appears as Lemma 2.4b in a paper of person B. (or having forgotten that B proved it.) At worst this is carelessness, but considering the proliferation of math research papers, can any of us seriously claim that we know _every_ result relevant in our area? A related problem is mis-attribution of results. That is, A cites result XYZ, attributing it to B when in fact it is due to C (and perhaps B even says in B's paper that it is due to C.) This is almost always carelessness rather than malicious behavior. b) There is not a lot that we can do about most deliberate plagiarism. Journal editors can try to prevent it and we can denounce it on places like toplist. Having been through some cases while on COPE I assure you that it is really tough to prove it. Even when proved, about the most the AMS can do is to ask the editors of MR to add a supplement to the original MR review of the paper noting the plagiarism. (Do these supplements appear with the original review on mathsci?) Theoretically, the Council of the AMS could write to the dean or president of the offender's university with a view to causing trouble to the offender, but can you imagine how often such action is likely to be taken by Council? [BTW if you think this is a problem in math, how would you like to deal with it in undergrad history term papers???] 2. On dead areas - I agree completely that this is a real problem. Cal Moore and I spent about 80 pages of our book proving two sentences of Connes' proof of the index theorem for foliations. We didn't expect that anybody would say that ours was the first proof and I doubt that a university P/T committee would appreciate the issue. I am dead certain that Connes had the whole proof of those two sentences in his head and just didn't want to take the time to write it down - or perhaps he thought that it was unnecessary to do so. Having said that, I should emphasize that I am not criticizing Connes. He is usually very good at writing up what he claims in a timely way - and usually he writes very well. I admire him greatly. I know of others - not to be named - who are real offenders. The trouble is that generally the person doing it means well - and frequently has no clue of the difficulty that it causes. Certainly it is a societal problem. The only remedy that comes to mind is to have the elder statesmen in the field talk to the offender, explaining to the offender what the problem is. I doubt that would help much. Claude PS: For the record, I don't agree with the definition of ethics given earlier, but that's a theological discussion.