Nope, sorry, this is not a post about politics, there are no upcoming elections anyway. I am writing in frustration, after checking once again on the web the status of a manuscript that I have submitted for publication over two months ago, to find out that it is still under review, ostensibly in the virtual hands of an unresponsive referee.
I know, I know, I am far from being alone in this predicament. These days, just about every time I speak to a colleague I end up hearing a rant over a manuscript held captive by uncooperative peers, with Editors seemingly unable to do anything to speed up the process.
And really, what could be done ? “People are busy”… “What if someone is traveling ?”… “It is the end of the Spring term, final exam time”… “We cannot expect anyone to drop everything to read someone else’s manuscript”, are some of the standard arguments to which we resort, trying to console one another as we desperately hold on to our panglossian belief that the world in which we live is the best possible, any attempt at making it better being inexorably doomed from the start.
My stand on this subject is known to the few who read my blog. Peer review is a vestige of a time past, and no longer has a place in today’s world. I am by no means the first to say so, much more authoritative sources have already expounded the notion that a world without referees would be preferable.
Besides causing unjustifiable delay in the publication of material that owes much, most of its interest to its timeliness, besides suffering from all the expected shortcomings of any process in which humans face an obvious conflict of interest, besides all of the usual, sacrosanct objections that are raised to peer review, in my opinion the best reasons to do away with it are the following:
1) It is no longer needed, citations providing a far more reliable assessment of the intrinsic, long lasting impact and value of a scientific contribution.
2) All too often it does not do what it purports to do, i.e., it does not weed out effectively incorrect, “crackpot” papers. And my experience actually points to this being especially true for high profile, highly cited journals.
But I do not want to repeat what I (and many others) have written already (here and here). The fact is, peer review has many defenders and apologists, and there is no reason to hope or expect that it will go away any time soon .
But does that mean that we have to accept the idea that we shall be spending a significant fraction of our scientifically active lives waitin’ ‘n bitchin’ ?
Are there no simple, not too disruptive measures that one could take in order to alleviate somewhat the existing problems ?
I hereby do solemnly swear
I think that a reasonable thing to do might be that of allowing for third party (obviously non-anonymous) endorsement of papers submitted for publication. Endorsement of an article is of course not a new idea. Curiously enough, ArXiV already allows that.
This is how I suppose it might work, for articles submitted for publication to regular journals. The author would request one or more respected scientists in the field to recommend publication of the manuscript to the Editor of the journal to which it is submitted. Endorsers would be required to write a review, much like that of an anonymous referee, also stating why, in their opinion, the article should be considered for publication in that journal.
By the time an endorser is ready to take such a step, presumably they have entertained lengthy discussions with authors, expressed to them any objection that they may have over the content of the manuscript, suggested changes to which the author has agreed to implement, and are now genuinely convinced that the work is appropriate for publication. In short, an endorsement is very much a non-anonymous review, one that the Editor would receive and take into account to whatever degree they see fit.
What kind of “improvement” could this bring to the current system ?
I can see an Editor seeking fewer anonymous reports, or none at all, for articles submitted for publication that should be accompanied by one or more authoritative endorsements. Obviously the choice of endorser is crucial to that aim, and I come back to this aspect below. This would speed up the process by reducing or eliminating anonymous peer review (I know, I am such a sneaky bastard), and encourage authors to seek early feedback on their work from competent colleagues before sending it out for publication .
Of course, this is no “magic bullet”, and clearly in order to be effective some obvious precautions must be taken. I can already think of a few objections (remember, there is no fixing what is already perfect, right ?)
I can have a buddy of mine endorse all my papers and endorse all of theirs.
Sure, you can do that, but the Editor is no fool, he only plays one on e-mail — kidding !
The Editor is the one who picks anonymous referees, one of the guiding criteria being precisely their expected impartiality. The Editor will easily be able to tell if the person endorsing the work is a current collaborator, or former mentor of the author, or in any case someone with a vested interest, not in the best position to make an objective assessment on the quality of the article. If that should be deemed to be the case, the endorsement will end up in some virtual dustbin — no weight to it will be ascribed by the Editor.
By contrast, a reliable, cogent endorsement, one that will presumably carry substantial weight in the Editor’s final decision on a submitted manuscript, will be offered by a well-established, experienced scientist — one at “arm’s length” from the authors(s), whom the Editor might have conceivably selected for anonymous review of that very article.
Why would someone not endorse my paper ? They do me a favor, one that I will sooner or later have to reciprocate, and it is no skin off their nose…
I doubt if that would really happen a lot. For one thing, endorsing would mean having to spend time to write a review, and that is not something that people may have the time or the inclination to do. More importantly, no one likes to see their name associated to a lousy piece of work. The names of endorsers would be rendered public, perhaps they could be featured in the published article, together with information such as the date of acceptance.
I can make a secret agreement with a group of other authors who, on paper, look disconnected from me and will endorse all of my papers, in exchange for me endorsing theirs.
I suppose that that is possible in principle, but I do not see it as fundamentally different from much of what already happens in practice. In any case, a possible way to mitigate against any such dynamics might be that of limiting the number of endorsements that anyone can make altogether in a given time period, or of manuscripts written by a single author. This may also reduce the pressure on junior scientists, who may conceivably feel compelled to comply with requests of endorsing work by influential senior colleagues.
Perhaps there are other potential drawbacks of which I am not thinking. However, I would like to stress that endorsement would be in addition to peer review. In the presence of one or two professionally written, competent and convincing endorsements, maybe the Editor may spare the author with the third anonymous review, or be inclined to bring a review to an end earlier. Simple as that. Nothing earth-shattering.
Oh, come on, now, I cannot write earth-shattering stuff all the time…
 Yes, it is here to stay, I am afraid, no matter how increasingly, manifestly irrelevant it is becoming. You can easily get a sense of what a frivolous exercise it is. Just ask yourself whether you are spending more time reading or discussing with colleagues work published in refereed journals or published online (e.g., uploaded on ArXiv), before it has received the “official vetting” of, what… two referees ? A single one, sometimes ?
 Of course, many, perhaps most of us are doing that already. It is good practice.