Were I to summarize the essence of my job at your favourite weekly professional science journal beginning with N in a single word, it would be ‘making decisions’. (Well, that’s two words, actually, but, hey, I’m a palaeontologist, so sue me). Almost always my decision is ‘no’.
I reject at least four in every five manuscripts straight off the bat, before review; most of the rest perish in review (Note added later: some people are shocked to hear this, perhaps assuming that we send out for review everything we receive). In the end, barely one in twenty new submissions makes it through to publication.
One reason for maintaining a very high rejection rate is to ensure that the quality of the material we publish is always high (which I maintain it is, even when one accounts for the fact that editors are only human and prone to err). This creates a feedback – if the quality is high, and is perceived to be high, people will tend to submit their best stuff to us. Were we to loosen our belts, lowering the bar (and increasing pagination) the quality would lessen, people would send us their just-about-okay stuff as well as their best stuff, and, well, it would be the end of civilization as we know it. In a crowded market, it’s always been a good strategy to offer a high-end product.
The operative word here is crowded. For I have recently discovered yet another reason why editors at your favourite weekly professional science magazine beginning with N, and for all I know more ordinary journals, should be even more discerning.
That reason is the increasingly limited time and patience of referees.
I have been at your favourite weekly etcetera etcetera and so on and so forth for almost a quarter of a century. Back in the day, we used to get reports by mailing (remember that?) manuscripts to potential referees without even asking them first, and be reasonably sure that we’d get a result. These days we send email requests, and, in general, referees are happy to oblige. This is, in fact, remarkable, given that the referees do not get paid for this; they accrue no academic credit for it; and it takes the bulk of a working day to referee a manuscript properly. In many cases, the referee will see the same manuscript in several iterations before a final decision is made.
The system is maintained because today’s referee is tomorrow’s author, and authors will, naturally and reasonably, expect their manuscripts to be treated with the same care, courtesy and expedition as they would accord anyone else’s manuscripts. To be sure, the system is open to abuse by what we in the evolutionary game-theory biz call ‘defectors’, but on the whole it’s what we in the same biz call an ‘evolutionarily stable strategy’.
During that time I’ve been an editor at your favourite weekly panjandrum I have seen a secular growth in the number of manuscripts, and the number of journals in which they might be published. In the mid-1990s, for example, I headed up Nature‘s popular science writing arm (a very distant ancestor of what eventually became this), while at the same time handling all manuscripts in organismal biology as a part-time job. Today, the manuscript load in organismal biology has more than tripled, and it takes at least two of us, working full time, to keep up with the Red Queen, who in our case is anything but hypothetical (NB – this is an ecological in-joke). Year on year, I have ever tougher targets: the percentage of manuscripts I can afford to accept gets smaller and smaller. Over the same period, NPG has spun off a large number of top-tier specialty research journals that didn’t exist when I joined, and other publishers have done the same.
The number of editors has increased over the same period, but hasn’t kept up with the volume of manuscripts. There is a good reason for this – more editors would handle more papers and increase the demand on page space in the journal, which, as I have said, benefits from limiting its pagination – Nature‘s pagination hasn’t increased markedly in a generation: where new pages are added they tend to be for news and magazine-style features rather than primary research (and – cynical aside – publishing more research papers does dreadful things to one’s journal impact factor).
But I digress.
I suspect that whereas and notwithstanding the number of potential referees has also increased over the same period, as with editors, it hasn’t kept pace with the increased numbers of manuscripts. For referees are authors, and because research excellence is measured in coffee spoons publications, scientists (I suspect) churn out more manuscripts than they used to. This increases demands on the space of top-tier journals which (for the reasons I have explained, hey, keep up there at the back) reject more, creating a market for specialty spin-offs. Frank can no doubt explain the impact of this on research libraries and their budgets – but what does it mean for authors and referees? It seems pretty obvious that as authors are submitting more, they are also being asked to review more. Not only that, they are being asked to serve on editorial boards of journals more, if not to serve as editors, in which they have the task of finding suitable people in a limited pool to assess an ever-increasing tide of manuscripts. All uncredited. All unpaid. None of which accrues brownie points.
In the journals the scientists come and go, towards absurdam, reductio.
Now, where was I?
Ah, yes. Now, I’m not sure if it’s a blip, but over the past couple of months I have seen an increase in the incidences of referees declining to review manuscripts because they are too busy. They will, increasingly, cite the reasons for being too busy – serving on editorial boards, being asked to review too many manuscripts and so on. One referee complained – in not so many words – that the authors of a particularly recalcitrant paper weren’t taking his comments seriously, which, given the time he was spending on it, seemed unfair. This hit home, given that this particular referee has been reviewing manuscripts for your favourite weekly oompa-loompa for a long time, and is valued for his informative and conscientious reports – reports we use to make informed decisions.
What, then, is to be done?
I know what I would do – and what I am already doing, and always have done, only this time more consciously than before. Given that we at your favourite weekly brouhaha absolutely depend on the goodwill of referees, my task will be to use their services even more sparingly than I do already, and reject manuscripts even more ferociously at the first cut. If the job of an editor is to make decisions, what is the nature of such decisions? In short – that there is (or is not) an ironclad editorial case for publishing a manuscript, irrespective of its technical merits.
It is the referee’s job to judge those technical merits, once an editorial decision has been made. There is a tendency for editors, in a state of indecision, to float manuscripts off to referees in the hope that these (unpaid) referees will do their (paid) editorial job for them. This was and is always bad practice (at least in my personal opinion), but matters now are more acute than ever, given that the time of referees is so valuable. If we, as editors, send manuscripts out to review if – and only if – we think there’s a pretty good chance they’ll be good enough – then the job of referees will be easier and more pleasurable (long ago, I remember a scientist telling me he enjoyed reviewing manuscripts from your favourite poobah because, in his words, “you don’t send me crap”). Referees will know that, whatever else might happen, a manuscript from your favourite WPSMBWN will be pretty decent, and make time to review it.
Having kept our referees happy, what then? Editors will make more decisions based on their own recognizance or after consultation with colleagues, and reject more manuscripts without review. This will put more demands on the instincts, experience and judgment of editors, who, believe it or not, are only human.
I can’t see what’s wrong with this, and indeed cannot see any other future, so long as authors, referees (and, let it not be forgotten, tenure committees) value the imprimatur of a top-shelf top-tier publication. It’s possible that schemes such as post-publication peer review might offer some relief, but – and it’s only a guess – in some perhaps rather specific subject areas, rather than for the scientific endeavour more generally.
Phew. And after all that, a survey showed that eight out of ten referees preferred their cats.