Is it good for a journal, no matter how prestigious, to reject excellent scientific contributions on the grounds that it gets substantially more first-rate submissions than it is able to accept? In TCS, I am aware that JACM has such a policy, and I wonder whether it is backfiring badly. (My, possibly wrong, impression is that several authors decide not to submit top-notch papers to that journal because of that policy. I myself have always some trouble in deciding whether a paper is among the best papers of the year in its area since this seems to involve some crystal-ball gazing.)
A recent, remarkable instance of this kind of rejection is mentioned in a letter in the latest issue of the Notices of the AMS. (See here, on page 2 of the file. The letter is co-signed by Vaughan Pratt, one of my favourite theoretical computer scientists.) Apparently, the editorial board of the Journal of the AMS, which is the flagship journal of the American Math Society, has declined to publish a 14-page paper reporting on Friedrich Wehrung's solution to Dilworth's Congruence Lattice Problem for its lack of “interaction with other areas of mathematics”. The problem had been open for about fifty years, and drove the development of lattice theory during that time. See this web page for more information.
I am sure that the author will rapidly publish the paper in a top-notch journal, given that it had glowing referee reports. What I am not sure of is how many, apparently superb papers, a journal can decline to publish before authors stop submitting to it.
I guess that, as usual, the great judge will be Time.
Addendum 12 June 2007: A look at Friedrich Wehrung's publications page indicates that the aforementioned paper of his is going to appear in Advances in Mathematics.
2 comments:
I think it's good that we have some journals that are trying to maintain very high standards, accept only the very best papers and are very selective. I don't see anything wrong with it. On the other hand, I hate to see a journal with excellent reputation that accepts crap papers. The point is that someone must judge which papers are "excellent" and which are not. And people do mistakes, sometimes. Still, if it's really a very strong paper than it will be accepted in some other very strong journal, and hopefully the mistakes are rare. So, mistakes happen and let's try to do our job as editors and referees to make sure that this won't happen to often. (And also let's try to avoid pushing weak papers of our friends.)
A
Thanks for your thoughtful comments.
I think it's good that we have some journals that are trying to maintain very high standards, accept only the very best papers and are very selective. I don't see anything wrong with it.
Neither do I, and I never said otherwise. However, I wonder what "high standards" can be used to justify rejecting a paper that solves a 50-year-old open problem and that is considered by the cognoscenti to be a major achievement in a thriving research area.
So, mistakes happen and let's try to do our job as editors and referees to make sure that this won't happen to often. (And also let's try to avoid pushing weak papers of our friends.)
Again, I agree. Being an editor or a referee gives power, and with power comes a fair amount of responsibility. Scientists are also human beings, and it is at times difficult not to let issues like friendship interfere with scientific judgement. I am not naive enough to believe otherwise.
I am always amazed, however, by the overall fairness of the decisions that are made by editors and PCs, bearing in mind that PC decisions are always subjective in the grey area where many submitted papers lie.
I believe that the scientific evaluation system works well in general and that mistakes are relatively rare.
Post a Comment