Norms for review times in philosophy compared to other fields?

In our March "how can we help you?" thread, a reader asks:

Can some one explain how the norms for peer review got established in philosophy? I ask because while I'm a philosopher by training, much of my research is very interdisciplinary. As a result, I'm asked to referee for both philosophy journals and journals in the sciences. Science journals typically ask for a review be completed within a week or two, whereas philosophy journals often allow for somewhere around six weeks to hand in a review. For me, refereeing a philosophy paper doesn't take more effort or time than it does to referee a science paper. Both require about the same amount of time/effort, and I generally supply a few pages of notes. This leads me to wonder why philosophy journals give so much more time. Does anyone know the history behind how these norms have been established?

Interesting questions. I don't quite recall where I came across the information, but I seem to recall seeing somewhere that peer-review tends to take longer in a variety of humanities fields compared to in science. I wonder whether there might be a few reasons for this. First, people tend to co-publish with a lot of co-authors in science compared to humanities, where people tend to publish solo--which might leave scientists more time for reviewing papers (indeed, scientists often have well-funded labs with a large number of grad students and junior faculty who collect and analyze data, etc.). Second, perhaps the quality of scientific papers is easier to judge, given the comparative objectivity of scientific methods (e.g., agreement on sound methods of data collection and analysis, etc.) compared to the humanities? Then again, maybe there's some other explanation.

What do you all think? It could be helpful to hear from people who have published in different fields to get their takes!

Comments

[referee .. with a halo]

It does not have to be this way. The last 27 papers I reviewed for Synthese were reviewed between 0 and 14 days upon my agreeing to review the paper (so says EM). The 14 is an outlier, as it is the only paper that took double-digit days. The last 24 papers I reviewed for Philosophy of Science were reviewed within 0 to 3 days (says EM). We have grown accustommed to very bad behaviour.

[Chris]

I'm just guessing here, but another factor might be the urgency of many scientific results: if my paper might contribute to cancer treatment, then it is important that it be peer reviewed in a timely fashion. My commentary on Maimon isn't so urgent. Of course, much of science isn't so applied as curing cancer, but it is often connected to work that is applied (if merely in the sense that ones colleagues who work on more theoretical work are part of the same department as the applied folks, so the faster-referee norms leak over into the more theoretical part of disciplines. I'm sure the greater consensus in science about the good/bad work also helps: scientific journals - even the most prestigious ones - often publish a much higher rate of submissions than "top" philosophy journals. This is likely in part because if I'm a scientist I have a better sense of whether my new cancer paper has a chance of being published in The Lancet or not.

This is merely meant as explanation, rather than justification. As the halo referee above mentions, we can change these norms. Journal editors can ask referees to return reports in 2 weeks rather than 4, referees can do likewise, even if they are given longer times, etc. But it will take time to change these norms, unfortunately.

[Sisyphus]

OP here. I appreciate that some areas of science require more timely publication schedules, like cancer research, however this is certainly not true of the science articles I've refereed. Trust me, no one is asking me to ref cancer research!! In other words, none of the science articles I've reviewed needed to be published any sooner than a given philosophy article.

Though I haven't done near as much refing as Halo, I too think I've gotten all my reports done and submitted within about a week. I'm puzzled why it should take longer than 2 weeks to get a report in unless someone is under unusually demanding circumstances (about to go up for tenure, defend, or whatever), in which case just decline the request. I just get this sense that it has far more to do with disciplinary norms than it has to do with philosophy or humanities papers really requiring more time to ref.

[Laurence B McCullough]

I had a 45-year career as a philosopher medical educator and reviewed for medical and bioethics journals. In academic medicine, being asked to review papers for the leading journals in one's field is valued; for prestigious general journals such as New England Journal of Medicine and JAMA, being asked to review papers is a very important form of recognition and so is highly valued. There is therefore a professional incentive for perform peer reviews. Is this true in Philosophy departments?

With one exception (a bioethics journal) reviews of my submitted work were always helpful and revising in response to reviews strengthened the papers and led to their acceptance. My colleagues had the same experience. Is this true in Philosophy departments?

Then, too, papers at medical and scientific journals are limited to 3000 words or less, so asking for rapid peer review is practical. Papers at bioethics (and Philosophy) journals are much longer (6000+ words). Perhaps academic philosophers are disposed toward prolixity. I was mentored at the beginning of my career by a colleague who taught me: say it once, say it well, next point. When you have only 3000 words you must follow this standard; 6000+ words maybe less so.

[Daniel Weltman]

I know almost nothing about this topic but my impression is that lots of scientists farm the reviews out to their postdocs or grad students or other lab members/underlings/whatever.

As for philosophy, I try to get my reviews in within a week. Very rarely I take more than that, but never more than two weeks. My latest review I think I finished in a day or two. I think it would be better if philosophy reviewing went faster, although I think a large part of the delay is people not saying yes or no to review requests, or saying yes or no but only after a while. (I know very little about this but I think I've seen editors talk about this on this website and other places.) So even if all the reviewers review quickly, philosophy might still take a while if it takes many weeks to find reviewers. I don't know what's up in science such that this is less of a problem (to the extent it's less of a problem). Maybe it's easier for them to say yes immediately because, as noted above, they often farm it out to underlings.

[Michel]

Just a few observations:

1.) I'm told that in math, review times are _much_ longer, because all the work needs to be checked (and that sometimes means learning a bunch of new stuff developed for the paper). I'm told that the situation is much the same in logic; refereeing just takes _a lot_ more work. Generalizing, the standards of scholarship required by some subfields may just take a lot more time to satisfy. It may be that in ancient or other historical subfields, for example, referees are expected/need to return to the source material a lot or check the author's translations; that requires a lot more footwork than contemporary analytic subfields do.

2.) I don't know how people referee science papers, but I _do_ know that when I read them, I don't have to read very carefully. The literature review may or may not be of much interest for my purposes, so I know I can often skip or skim it. The methodology isn't particularly relevant for my purposes, apart perhaps from getting a broad sense of what was done and whether it was appropriate, so I can skim that. The detailed presentation/analysis of results isn't very relevant, so I can skim that. Finally, the discussion component is the most relevant, so I can read that. I imagine that, for referees, there's a good bit that can be skimmed (e.g. they pay closer attention to methodology and analysis/reporting, and less attention to the lit review and discussion). That would take less time, especially if the paper is shorter (which they often are). By contrast, when I read or referee philosophy paper, I have to pay careful attention to much more of the text. (I'm still a relatively quick reviewer, but it still takes more time than if I could skim bits.)

3.) Norms about review times in philosophy have actually shifted _significantly_ in the last fifteen years. From what I gathered online (in places like the Philosophy Smoker, among others), when I started my PhD in 2010, 3 months was considered quite fast, and six months to a year was pretty typical at most journals. It was also pretty commonplace to advise people new to refereeing that they should read _every article/book cited in the paper they were refereeing_. Now, I think that's silly advice, but it was given pretty often and authoritatively at the time; and that would _certainly_ take a lot longer.

4.) Workloads might be a relevant factor, depending on who does most of the refereeing in each discipline. Official workloads in the sciences tend to be smaller, to accommodate running lab sections and things. That might make a difference. What might also make a difference is (1) if referees in the sciences each tend to referee less, individually, than most philosophers do (which is at least plausible, since there are so many more of them), and (2) if most science referees are employed at research institutions whereas many or most philosophers who referee aren't (I don't know how plausible this scenario is, but it's a possibility). In that case, I'd expect one group to just have more time to referee than the other.

Having said all that, I do think we take too long, on average, to referee. I strongly suspect it's mostly just an issue of long deadlines being given by journals, with the end result that referees forget about/postpone their commitment until the last minute and then need more time. Three months seems acceptable but also very generous to me. In fact, most of the requests I receive these days seem to give about four weeks, and I never actually need that much time (though it's good to have that much, since it makes it easier to schedule refereeing around my other obligations).

[historygrrrl]

To answer the original question: It is not clear to me that, at this point, there *are* norms for referee times in philosophy.

When I get asked to referee, there are sometimes requests to do so in a timely fashion - say, 6 weeks or 8 weeks. Other times, there is no real deadline.

When I submit my work to journals, there also seems to be no consistency in how long it takes. I've received a few responses (all rejections, sometimes with comments) within two months, but generally, my work is 'under review' for some period of between 6 and 18 months.

I applaud the journals that are doing something to change this: even 6 weeks is an improvement. I agree with Michel, though, that the workload might be a relevant factor in the longer review times. Somebody who is teaching a 3-3 or 4-4 without grading assistance will probably be swamped with grading during the semester, may have to prioritise their 'free time' in order to get their own research done, and thus, put the referee tasks on hold until the semester is done.

[Chris]

Sisyphus: yes, that's why I added the bit about the work in applied science being done in the same department as the more theoretical science. The applied areas of science - a new version of Cognitive Behavioral therapy, a new cure for a cancer, a new model of how fast the Coronovirus will spread- are being done by people in similar enough disciplines who work on abstract theories of psychology, medicine and evolutionary biology, etc.

[Robin McKenna]

Speaking for myself (relatively senior, usual teaching and service load), the problem is mostly workload, plus I get no obvious immediate benefit from writing a review. When you are busy you prioritise the tasks that need to be done immediately. You can make time for the tasks from which you derive a clear benefit but reviewing isn’t like that. The only time I have for reviewing is the evening, but weeks might pass before I have the energy or desire to spend my evening writing reviews. You might say “we’ll just don’t agree to review if you don’t have the time” but my impression from editors is that they’d rather you return the review at around the deadline date than have you just reject the request because finding someone willing to review the paper at all, never mind within a few weeks, is getting harder.

What’s the solution? No idea but things won’t improve during the current publishing arms race.

[L]

I wonder what the acceptance rate of these scientific journals is – probably a lot lower than in philosophy journals.

I often review formal papers for philosophy journals that are easily 60-80 pages, with proofs etc. How can I review such papers within two weeks, when I have a full workload? I should think that the journals/authors are lucky that I accept to review at all.
Conference papers in computer science, which I also review, are limited to 10-15 pages and – if not sloppier – they are a lot more limited in scope, in my experience. In other words, they don't propose a new grand theory of X that one has to think about for a few weeks.

[L]

I meant 'higher' acceptance rate for scientific journals.

[Emanuele]

It's very simple. When I am asked to referee a paper (I work in phil sc/phil tech), typically I am told to submit my comments within a month or so. I enjoy providing this service to the profession, but I have also other duties, which should be prioritized (teaching; writing my own papers; writing grants; admin). This means that if I can referee papers (before the agreed deadline) during my 9-6 office time, then I gladly do it. If not, I just make sure I do it by the deadline I agreed with the editor. When I am asked to provide comments by one or two weeks, I generally decline (unless it's a revision of a paper I commented) because it takes time to plan ahead.

[Assistant Professor]

I'm at a different career stage from Robin McKenna, but I found their comment to resonate with my own experience. I have to fit reviewing into other obligations (in work and life). If I am asked to review and given a 4 week turnaround, I feel that I can reasonably get to the review within that time frame, even if the review itself likely takes me 1-2 days to complete. I tend to agree in those cases because me saying yes and reviewing in 4 weeks, if I am a good fit for the content, is probably better, all things considered, for the author and the journal than having to find another reviewer.

If a journal asked me to review with a 1 week turnaround, that may simply not fit with what I have going on that week, and I'd be more inclined to say no (I also find the presumption that I can simply drop what I am doing to review that week inconsiderate - maybe if JAMA or NEJM were asking me I'd feel honored, but I also worry about the way prestige institutions take advantage of prestige).

I do a fair amount of interdisciplinary work and I am almost always given a 4 week deadline, in a variety of fields. The only places that ask for a 7-10 day deadline have been journals that I am suspicious of and I decline.

[sisyphus]

I understand much of the reasoning in what many here are saying, however the situation still strikes me as a odd.

What brought this to mind for me, frankly, is that I was relatively recently asked to do a review for Erkenntnis while around the same time asked to do one for PNAS (Proceeding of the National Academy of Sciences, USA). Both papers were of similar length, and in many other ways quite comparable (though on very different topics). Erkenntnis asked for a review back in 5 weeks, PNAS in 2 weeks. I have a relatively full academic schedule, like many here have mentioned. But regardless of whether one would take or not take on such reviews due to the time frames and their availability, I'm just puzzled as to how PNAS seem to function just fine giving a 2 week window while philosophy journals seem to be struggling with givng 4 weeks or more.

[Emanuele]

sisyphus: one explanation could be that certain scientific cultures are just inhumane - the pressure they have for publishing, for getting grants (even tenured professors can be on 'soft money') are just way worse than the pressure we have in philosophy (this is only based on anecdotal experience in working with scientists)

[Necessary Evil]

Perhaps long review times are adaptive for a low acceptance rate field such as philosophy. Shorter review times would imply even more peer review requests, as less time would be lost from submitting an average paper to all the top journals before trying more realistic targets.