Suppose you’ve been asked to referee a paper for a prestigious journal. When looking over the paper you come to realize that you’ve refereed this paper for the another prestigious journal. You have a few options. They all seem bad to me.
Option One: Decline to Referee the Paper (Give no reason)
This seems bad. You don’t want to go burning bridges with journal editors, especially if you’re a junior professor. You also don’t want to appear to be someone who is not pulling their weight. For junior philosophers, chances are you’re not pulling your weight when it comes to refereeing journal articles. I don’t think this is their fault…I just suspect that they don’t get asked very often. (See Clayton Littlejohn’s post on this here.)
Option Two: Referee the Paper
This also seems bad. I’ve heard a few anecdotes about people getting identical referee reports from two different journals (I’m one of them). In conversations with philosophers, the general consensus seemed to be that this was poor form to even agree to referee the paper at all.
Option Three: Decline to Referee the Paper (and disclose that you’ve refereed it for another journal)
This seems like a good, noble option for a person. You’re not leaving the editor hanging without good reason, and you’re not engaging in the poor form that Option Two exhibits. For awhile, I thought this would be my policy too – but now I’m not so sure. If you say that you’ve refereed the paper before, then you disclose something about the paper’s history that the author of the paper might prefer not to have disclosed. You disclose to the editor that this paper has been rejected from another journal.
So…what do you do?
I think I have a fourth option, but I’m curious to see what other people think first.
Why not option 3.5: decline to referee the paper, and give as your reason the fact you are familiar with the paper already and couldn’t be objective about it. This certainly is (at least part of) the reason why you shouldn’t referee it at all, and the circumstances of *how* you came to be familiar with the paper are beside the point, which is that your familiarity with it makes you unable to referee effectively.
I actually discussed this with one of my classes before class was starting, and that’s pretty much the option we came up with. If I’m ever in this kind of situation, that’s probably what I’ll end up doing.
Here’s a related puzzle. When I discuss anecdotes about referees that have pretty obviously taken option 2, there are two different kinds of stories I get that seem to yield different intuitions among philosophers.
Case One: The referee rejected the paper elsewhere, and decides to referee it again.
Case Two: The referee enthusiastically recommended acceptance. For some reason (unknown to the referee) the journal doesn’t accept the paper. Much to the referee’s dismay they are asked by another journal. The referee decides to referee the paper again, and enthusiastically recommends acceptance.
I get strong negative reactions from philosophers regarding Case One. I don’t get strong negative reactions from philosophers regarding Case Two.
I’m inclined to play things safe and just bow out regardless of what my first verdict was, but the difference in reactions to Cases One and Two are curious. I’m wondering:
(A) Does this motivate the idea that whether you should referee a paper again depends on what you said the first time around?
(B) If the answer to (A) above is “Yes” – What explains it. Why is OK to ref again if you recommended acceptance, but not OK to ref again if you recommended rejection?
I think that what the potential referee should do will depend on further facts: (i) what was the previous verdict he/she gave on the paper; (2) is the paper changed and, if so, how much?; (3) do the two journals involved have different standards for publication?
Now the trick is that at the time of the request, the potential referee probably won’t know the answer to some of these questions. So it would seem to me that he/she should apprise the editor of the general facts and ask for more information.
Option 1 isn’t so terrible provided (a) you do it quickly and (b) you suggest some other plausible names for referees; preferably ones the editor wouldn’t have thought of themselves. Lots of people decline to review papers.