Halloween is my favorite holiday: you aren’t strong armed into unhealthy levels of conspicuous consumption, the costumes and pumpkins are creative and fun, the autumn colors are fantastic, and the weather is typically quite pleasant (or at least it was in pre-climate change/hurricane Sandy days.) You don’t even have to travel at all! So in honor of Halloween, I’m going to tell a (true) horror story about…
Back in August, I was asked to referee a paper for a certain prestigious physics journal. The paper had already been reviewed by two referees, and while one referee was fairly clear that the paper should not be published, the other gave a rather weak rejection. The authors replied seeking the opinion of a third referee, and that’s when the editors contacted me.
I immediately noticed that something was amiss: the title of the paper was nearly identical to a paper that my co-authors and I had published in that same journal a couple of years earlier. In fact, out of 12 words in the title, the first 9 were taken verbatim. I’m sorry to say, but it further raised my hackles that the authors and their universities were unknown to me and came from a country with a reputation for rampant plagiarism. Proceeding to the abstract, I found that the authors had copied entire sentences, merely substituting some of the nouns and verbs as if it were a Mad Lib. Scrolling further, the authors copied an entire theorem, taking the equations in the proof symbol-by-symbol and line-by-line!
I told all of this to the editor and he of course rejected the paper, providing also an explanation of why and what constitutes plagiarism. A strange twist is that my original paper was actually cited by the copy. Perhaps the authors thought that if they cited my paper, then the copying wasn’t plagiarism? They had even mentioned my paper directly in their response to the original reports as supporting evidence that their paper should be published. (“You published this other paper which is nearly identical, so why wouldn’t you publish ours?”) Thus, at this point I was thinking that it’s possible they simply didn’t understand that their actions constituted plagiarism, and I was grateful that the editor had enlightened them.
Fast forward to today.
I receive another email from a different journal asking to referee a paper… the same paper. They had changed the title, but the abstract and copied theorem were still there. Bizarrely, they even added a fourth author. The zombie paper is back, and it wants to be published!
Of course, I can also raise my previous objections, and re-kill this zombie paper. And I’m considering directly contacting the authors. This clearly isn’t a scalable strategy, however.
It got me thinking. Is there a better way to combat plagiarism of academic papers? One thing that often works in changing people’s behavior is shame. My idea is, perhaps if we build a website where we publicly post the names and affiliations of offenders, then this will cause enough embarrassment to stem the tide. Sort of like the P vs. NP site for erroneous proofs.
What’s your best idea for how to deal with this problem?
O brave new quantum world!
Journals should adopt plagiarism detection software. For this to work they need to be able to text-mine the literature, so this is yet another argument for open access. In our field it would work fine with the arXiv, since most papers are there and it gives access to the full TeX source. There will always be unscrupulous journals that will publish anything, but at least the most prestigious journals could do this.
I would be really shocked if somebody plagiarized a paper of mine…as they are already quite useless in their original version! so congratulations to you, Steve:)
In general I’d think it’s kind of rare in physics anyway, but otherwise the website idea sounds fun (unless you get into legal troubles). It’s very popular instead self-plagiarism, to produce n papers out of one…but at last arxiv seems to have a new nice feature to unmask that.
ciao!
I don’t these people are worth the mental effort to even remember their names, let alone naming and shaming them.
If you see your work plagiarised in a journal, inform the editor and the editor will (should!) revoke it. If you see it on the arxiv inform the admin (they have acted against plagiarism in the past).
But otherwise just ignore these creeps.
Ignoring them makes it a no lose proposition: they either get away with it or they get ignored. This way, the problem likely will get worse.
Naming and shaming please. Starting with the above. Why the kid gloves Steve?
I wanted to test the waters first. I still think that these people don’t really understand that they are doing something wrong since they make no attempt to hide the source. This is just something they were presumably taught to do by their mentors.
And honestly, I was hoping that maybe someone would start a “hall of shame” website, as spontaneously happened against Elsevier, where I could then put these people at the top of the list and start a movement. 🙂
Indeed arXiv does it already:
http://arxiv.org/find/grp_physics/1/AND+co:+AND+text+overlap+co:+AND+authors+AND+by+other/0/1/0/all/0/1
Interestingly if you remove “by other authors” from the search, you pass from 162 results to more than a thousand:
http://arxiv.org/find/grp_physics/1/co:+AND+text+overlap/0/1/0/all/0/1
They also seem to have posted some public shame page in the past:
http://arxiv.org/new/withdrawals.aug.07.html
I had this happen to me back in ’04. I was google searching my paper by title and up popped a paper with the exact same title minus two words. Large chunks of text and equations were lifted but the two authors never posted on the arXiv so it went under the radar. What was strange is that they cited a different paper of mine (I suppose to deflect criticism from a casual referee who might investigate but only to affirm they cited prior workers). I was working at a government lab at the time and brought it up with the admin there. Lab lawyers got into the mix and the journal was contacted but the process seemed to have stalled, partially because the publishing company was based in another country, and they never actually withdrew the paper. I was bored with the entire process but I wish I had followed up because it is way too easy to get away with plagiarism. The good news those two dopes don’t seem to have anything else in quantum since.
Relating “radarâ€: I’ve already seen twice recently withdrawing from arXiv after acceptance in a journal (once with direct pointing to demand of editorial). If someone has an information or ideas relating the practice?
Can you point out an example?
This withdrawing presumably wouldn’t remove old versions from the arxiv, so it seems a strange policy.
One example may be found directly via search form
http://arxiv.org/find/all/1/all:+AND+withdrawn+journal/0/1/0/all/0/1
Pretty striking!
Well, I for one am going to stop donating my free refereeing services to the Journal of Siberian Federal University until they stop this offensive practice.
I caught someone recently. I was a referee, and noticed that one of their pictures looked very familiar. A bit of googling showed it as one of the first hits, on a different groups homepage. Anyway, I told the editor, who informed the author, who then resubmitted with the figure replaced (and who didn’t seem to be to amused to be bothered about something as trivial as plagiarism). Anyway, the paper was pretty bad and hard to read, so I ended up having to read some of the references to understand what the author was trying to say. It was only then I realised that they had also copied large chunks of the introduction from their references. Argh!