How Much Should Graduate Students Publish?
A current graduate student writes in with this sense of what is expected nowadays:
I’m under the general impression that I need to get as many publications in top journals as I can before I go on the job market. Considering how slow this process can be, and the fact that you can’t concurrently submit the same paper to more than one journal, it follows that I need to start publishing as early as possible.
And a request for information:
It would be nice to know, in general, how many philosophers have published by year 1, 2, 3, … N of their graduate programs. You’d maybe want to limit it to a certain subset of journals, like the top 20 by the Google Scholar metric or something like that. Ideally, this could all be cross-referenced with job outcomes after graduation (e.g. Adjunct, post-doc, VAP, TT), as well as some way to control for institutional “prestige.” But even just knowing how much people are publishing during their grad careers would be helpful.
If someone out there among the Daily Nous readership wants to construct an easy way to solicit and compile that information—a survey of sorts—go to it and I will happily link to it. In the meanwhile, it may help if readers conveyed their experiences and opinions in the comments here, or took up a discussion of the “general impression” for the need for multiple early publications.
There’s just no answer to this that’s going to be satisfactory, and that’s because getting a TT job isn’t a CV-measuring contest. It’s not about having more publications than another candidate, or more publications in “better” journals. You can have 2 articles with one in a top-20 journal, or 7 in a mix of top-20 journals (and non-top-20) and get tonnes of interviews, or no interviews. Also, lots of people each year get a TT job with no publications. We have some data on this: http://philosopherscocoon.typepad.com/blog/2013/03/some-facts-about-this-years-hires-so-far.html for example calculated that the median number of publications for those hired last year into TT jobs was 1. The mean was 2.2. That means that many were hired without any publications.
Don’t look at publications as either a necessary or sufficient condition for being hired. Rather, view them as one part of a broader narrative about yourself. Articles are instrumental in what they can help show about yourself as a candidate. Articles show that you can publish, and if it’s good work, then that shows that you can do good work. And it’s your ability to do good (published) work that will get you tenure, which is one key think for which hiring departments want some evidence. But one can provide that evidence even without publications.
So it’s good to have an article or two, but it’s neither necessary nor sufficient.
I say this as someone who had two publications (one in a top-20) before I defended, and 7 (with a few top-20) by the time I was on the market post PhD.
I’m sure publications are neither necessary nor sufficient for being placed somewhere, but that advice can be a bit misleading. During the last two hiring cycles we had applicants from many of the highest ranked programs and many of the applicants had several publications in very good places. I was a little surprised to see that, since these students were clearly not relying on the prestige of their program. While it is neither necessary nor sufficient for being placed, I think it’s obvious that your chances for a good placement go up a lot with 3 or so papers in good places.
I teach at a public university with a philosophy B.A. program. We don’t have a graduate program. I can only offer anecdotal evidence of how hiring works at my school. In my experience on three hiring committees, one or more publications, depending on where they are published, can carry a lot of weight. Publications in journals with poor reputations do not help much. (I am not saying whether this is fair or not, it just seems to be how these things work in my limited experience.) Publications in good journals help a lot. Publications in elite journals, at a school like mine, increase one’s change of getting an interview quite significantly. We have still interviewed and hired candidates with no publications.
The number and quality of publications you need to get a job is inversely proportional to the prestige of your pedigree and the power of your advisor. ‘Thanks’ to this, if you’re philosophically well-bred you’ll be able to sit on your work for ages, and so eventually publish something pretty good. Then people will pat you on the back for not being one of those low class rubes who publish lots of meaningless filler, as if they had bills to pay or something.
I think that a lot hangs on the sort of university to which you’re applying, and that it’s not automatically the case that “better” departments will value publications the most. My department is B.A. only, but the number of pubs can matter a fair amount when it comes to tenure. My department, I think, would prefer fewer and better, but at the university committee level departmental expectations are frequently set aside. So you have to be able to publish here, and to do it on a 3-3 load where there’s little intersection between the material you teach and what you’re writing about. We’ve passed on interviewing candidates from terrific schools with strong letters but no publications, because we don’t want to bring someone in and find out that they either have no clue about publishing or that they don’t want to publish anything that’s not going to transform their field. We want to bring in people who can hit the ground running in terms of getting work out. Most people we interview seem to have 2-3 pubs at least, and indeed most of our more recent hires have had post-docs or prior positions.
I don’t want to be taken to be saying that one shouldn’t publish, or that one shouldn’t worry about publishing, while one is a grad student. Quite the opposite! I think it’s very useful to start working through what it takes to publish quality articles while one is still a grad student. That experience will be very helpful if one is fortunate enough to land a TT job where the clock for tenure starts. And certainly, this is something I did myself.
I also don’t want to be taken to say that having (or not having) publications won’t, ceteris paribus, matter at all in one’s application. Of course they matter, but my point is that they’re not all that matter. Moreover, there are considerations that can compensate for not having any publications (even if one is not from a “top” school).
My views on this have changed over time, both being on search committees and working with my own graduate students. I used to think that all publications were good, i.e., any better than nothing. Shows professionalization, for example. But now I think just about the opposite; one thing you do when you start publishing is fix your trajectory, and if you start off with lesser publications (e.g., by venue), then it can be hard to get above that. So now I more think only publish if you can in elite journals and that even second-tier publications can be damaging. It depends, of course, on where you’re applying, but this is just me (at an MA program). It also matters where your Ph.D. is from; if you’re from some top place, committees are more likely to trust that an unpublished person will have the talent to fix it later. If you’re from not a top place (as I was), then I think you have to publish your way out of it.
Fritz, can you please clarify what you mean by “elite” and “second-tier” publications/journals?
Thom Brooks has a ranking of journals that seems pretty widely accepted: http://the-brooks-blog.blogspot.com/2011/09/journal-rankings-for-philosophy_29.html
My guess is that Fritz would say that anything that isn’t in an A or A+ journal in the Brooks rankings is not worth publishing in as a grad student. (Though maybe some of the B journals are.) For what it’s worth, I’ve heard this from others, too.
I am sorry, but I find this unacceptable. Clearly, it is extremely impressive for someone working in phil. of science, or in the history of phil., to publish (respectively) in the British Journal for the History of Science or in the Journal of the History of Philosophy. (Whether or not he/she is a grad student.) Yet these (and other very good) journals are only in the B list of that ranking.
Just to add one further point: both ‘Phronesis’ and ‘Oxford Studies in Ancient Phil.’ are on the C list of this ranking. Yet, it would be foolish for a search committee not to be duly impressed by a candidate working in ancient phil. who has published in these journals.
Graduate students should take these rankings, and more generally talk about ‘elite’ and ‘second tier’ journals, with great caution.
Your papers needn’t be published in an A or near-A journal to be excellent publications. And I’m not saying that good papers get published in so-so journals, though that’s sometimes true. What Thom did was rank journals that are already among the best in the field (how else could Erkenntnis be ranked C?). And he says exactly that in the comments to the ranking. What he did is take the journals that are considered the best/better journals in the field and rank them against each other. This makes much better sense of what is a prima facie somewhat odd ranking.
Thanks for the clarification Mike, that makes more sense. I’d still want to stress that for someone working in History of Ancient Phil. , a publication in Phronesis or Oxford Studies (‘C’ journals) will be very prestigious, indeed just about as prestigious as a publication in most of the ‘A’ or even the ‘A+’ journals (and, when published in the specialist journals the paper will actually get read by the people interested in the field). To that extent, the talk about A, B, C journals still strikes me as potentially misleading and pernicious.
Mike is correct about the rankings on my blog. They are a list of the leading journals found in the Australian journal rankings, the European Research Index for the Humanities, Leiter’s poll at that time and a wider-ranging poll I conducted shortly afterwards. The methodology and links are all given on my blog. The A, B, C are a rough division of the very best journals overall — “B” or “C” are still excellent.
For what it’s worth, much time has passed since and I am planning to update the rankings soon(ish)…
I definitely wouldn’t index my thoughts to the Brooks ratings, for some of the reasons given here. I think “elite” journals can certainly be discipline-specific, which means they tend to not score that well on these generalist ranking lists. So the thought would be to either publish at the top outlets in your field or else consider don’t publishing at all. Some well-regarded specialty journals might be unknown to search committees, but none of the ones in these comments would carry that hazard. Again, though, even that advice needs to be tempered with a cognizance of what sort of schools you’re applying to for jobs; some places just aren’t going to care or be well-situated to know what’s what. One good proxy might be to look at the teaching load which, to some extent (i.e., with many exceptions), inversely tracks the importance of publications.
Fritz Allhof says ‘one thing you do when you start publishing is fix your trajectory, and if you start off with lesser publications (e.g., by venue), then it can be hard to get above that.’
This does not make sense. If you publish a paper in a C or D journal then doing so does not make you less likely to subsequently publish in an A or B journal. How could it? As if choosing to not send off that paper to that C or D journal would make your subsequent work better!
DD – I think the point is that hiring committees want to know that you have the capacity to produce good quality work in decent quantity and get it published. Having a set of paper drafts up on your website that people can quickly skim through can let a hiring committee know that you produce a decent quantity and that the quality is at least decent. (They won’t necessarily judge the distinction between decent and outstanding work in a quick skim.) Having a publication shows also that you at least meet the minimum bar of being able to handle the exchange with a referee and editor. But if that publication is in a C or D journal, then it doesn’t tell the hiring committee anything more about whether you produce decent work or outstanding work. A publication in a higher rank journal does tell them at least something about you having produced work that someone has judged to be outstanding enough for those journals.
My guess would be that for a grad student, having several strong drafts that hiring committees can look at (in later stages of the search process) can be as valuable as several publications in middling journals, but that publications in higher-end journals are the only thing that will convince them (beyond close and detailed reading of the papers themselves, which are unlikely until you get to the very last stage of a search) of the general quality of the work.
Imagine what it would be like to work and study in a profession where the answer to this question was primarily a function of how much the progress and dissemination of knowledge was likely to be furthered by frequent grad student publications.
Of course “better” depts won’t value publications the most: especially for junior hires, they overwhelmingly value the opportunity to sustain the impression of a community of wonderboys (and a few girls) who don’t need to prove themselves if not in each other’s eyes and reference letters.
It seems to me that there should be more faculty mentoring in these matters. I know that some departments place more emphasis on teaching and some more on research, but if a grad student submits a paper or has an idea that a faculty member thinks might be publishable, it would be really helpful to have that kind of encouragement from a more experienced source. I remember when a professor asked me re: a term paper I’d submitted as a grad student, “Have you thought about trying to publish that?” The thought would not have crossed my mind regarding that particular paper, but I submitted it, and it was published. And that experience was incredibly valuable. There might also be something valuable about mentors and grad students publishing papers together. I see it more often these days than I used to, and I don’t think it’s a bad idea, assuming that both sides contribute equally.
I had a few further questions as well: First, how do scholars in the UK/US view publications in edited volumes? Does it depend on the nature of the volume? On the publisher? What about the publication of conference proceedings (as an edited volume)? I’ve been told some institutions don’t even count such publications, which strikes me as odd.
I’m also curious about how much people think grad students should be submitting to conferences. I found that, for me, doing at least 2-3 conferences a year as a grad student really opened up a lot of doors. In a way, I almost think conferences and workshops can prepare one better for the job market than merely publishing. You learn how to present your ideas verbally, to deal with criticism on the spot, to ask good questions, to network with colleagues, etc. And besides, some conferences can lead to publications, whether it be a volume of the conference proceedings are a refinement of the presented paper, which one then submits to a journal.
I guess there are two aspects of the question:
1) what ought your attitude to publications be simply from the point of view of disseminating your ideas, engaging in a conversation, pushing scholarship forward and generally participating in a research community?
2) what ought your attitude to publications be from the point of view of maximising your chance of getting a job?
Or alternatively, (2) could be replaced with
(2′): how should the attitude established by (1) be modified by recognition that what employers want conforms imperfectly to the research-optimal attitude?
I’m going to tentatively suggest that people should mostly focus on (1) and disregard (2)/(2′) – not out of wide-eyed naivete but because answering (2)/(2′) is so fraught, so dependent on inside information about different departments’ expectations that you don’t and can’t get, and leads to such contradictory advice from different people, that I think applicants’ chance of calling it correctly is in practice extremely low – whereas (1) is relatively easy to think about and understand.
(And to be fair, my attitude to (1) is that publishing relatively frequently and not being too perfectionist is good research practice. If your answer to (1) was “publish nothing in the first 20 years” then don’t disregard (2)!)
I wonder if this question could be reframed for narrower categories. I don’t think, for example, distinguishing BA-only programs is enough. There is an enormous difference between degrees of teaching-centeredness even among BA programs. And, more important, there are light years of difference between well-known or high ranking BA programs and unknown, low-reputation BA programs.
Moreover, I think we need to prioritize these categories: which programs will most candidates apply most often to and have the best chance of getting? The likely answer is a teaching-centered, unknown, low-reputation program. So, general advice should be directed there, with qualifications for research/Ph.D. and/or prestige programs.
This might make a huge difference. At my department, I suspect the suggestion that second-tier publications would count against a candidate would be laughable to almost everyone. I suspect there are departments that might even take someone with only top-tier publications as a flight risk, and resent them as too good for us. So the highly conservative publishing advice in many of these comments might not just fail to apply in some cases, it might fail to apply in most. And, further, it might be nearly the opposite of good advice for most cases. Basically telling someone applying for a janitor’s position: don’t bother interviewing if you don’t have Brooks Brothers’ suit to wear.
Two points:
i/ Early year phd students should, if possible, serve on an appointments committee (or participate in an appointments process) to get a look at the kinds of CVs those on shortlists have.
ii/ The rise of the post-doc as an almost standard step in the academic career trajectory reduces, in my view, the importance of coming out of the phd with a fully developed publication record.
A general observation. Departments face a difficult task of selecting one candidate from a group that will usually have several meeting the stated criteria (at least on paper). Publications can be an important proxy providing additional, external verification that a candidate is exceptional among his/her peers. A paper at a weak place is more of an issue if a candidate has only this paper or not much better. (Whereas having one or a few papers, but all at “top” places looks especially good.) After a candidate hits about 10 or so publications, I don’t think some papers in not-the-“top”-places are a serious issue provided there is clear evidence of top quality work on hand.
Now to the UK case. The UK is probably a very different market insofar as there is a national “Research Excellence Framework” exercise whereby research-active staff are expected to produce four publications (subject to certain qualifications) over a 6 or so year period. While there are exceptions galore, most leading UK departments I know will only take seriously candidates with at least some publication track record of 2-3 papers. (Yes, it can be possible that someone hired late in a “REF” cycle might need only one publication for consideration, but departments like to have some choice about which publication that is — so 2 or 3 should be safe and should be aimed at internationally peer reviewed journals.) It is simply a matter of fact that I don’t recall any time where candidates made a shortlist – even for a temporary teaching post – without at least one good publication in hand. Usually, candidates already have at least 2-3. For posts beyond early career, the expectation is usually that you must have at least 4 pieces for review (and not unusual to be asked about top 6 publications for a REF period even if only 4 are required – again, departments like choice).
For what it’s worth, here is a data point. I’m keeping this post anonymous so as not to violate any confidentialities, but I’m at a research-1 school in the middle third of the PGR rankings. [Side note: I suggest we all refer to them as such going forward, and do not refer to them using the proper name of an individual who may not be associated with them into perpetuity.] In the last few years we hired a junior faculty member in an LEM field. We interviewed 8 candidates by Skype. Of those candidates, one was still expecting to receive the PhD and one had just received the PhD. Neither of those 2 candidates had any publications. The other 6 candidates all had multiple publications in highly-respected journals. (They also tended to be a few years out from their PhDs.) Of the 3 candidates we flew out, all were from the latter, published group. We certainly didn’t make our decisions solely on the basis of anyone’s publication record, but the facts are what they are.
I realize that this is not helpful to the student who inquired, but I agree with Ruth Millikan that requiring early publication in philosophy is a bad thing. See her 2012 Dewey Lecture here:
http://philosophy.uconn.edu/wp-content/uploads/sites/365/2014/02/Accidents-APA-John-Dewey-Lecture.pdf
In addition to the arguments she gives there, I’ll add that, as someone who supervises Ph.D. students at a relatively prestigious program , I don’t see how graduate students have the time both to write a really strong thesis with well-worked out ideas which will serve as a basis for future publications down the road, and to do the kind of fine-tuning of one individual piece of work which will result in a publication in a good journal before they go on the job market. So I’m reluctant to encourage people to spend time trying to publish, since it is so often at the expense of their dissertations. I wish the discipline could collectively stand its ground against the increasing pressure to publish as a graduate student. Also, as someone who sits on hiring committees, I regard publications as irrelevant. To a large extent, getting published as opposed to just trying to get published is a matter of luck (who you got as your referee and whether your paper resonated with them). I’d much rather see someone with a really interesting and well-thought-out thesis project and some good chapters.
Also, we all have too much to read anyway! We should be collectively adopting policies which promote fewer and better articles by people who have had the time to work out their ideas on a topic, which is what being a PhD student is supposed to be for.
I think it depends heavily on what structure the thesis takes. If the chapters have a parallel structure to journal articles, then it takes relatively little work to turn a dissertation chapter into an article and vice versa. In my program, rather than comprehensive exams, we did two “research areas” (essentially, sequential 4-12 month one-on-one courses). In some cases, the work students did there became journal articles (for example, 3 became journal articles for one of my RAs, and the work in the other became the basis for another). 2 of these articles (only one published pre-defence) became dissertation chapters with relatively minimal editing to make them flow as part of a single document.
If the dissertation has the “typical” structure of 3-5 gigantic chapters, then of course it’s hard to turn that into a journal article. But if the dissertation is designed more closely to approximate journal article structures (and lengths), then the task is not so difficult.
Perhaps I should stop using myself as an example, but depending on how quickly the student is progressing towards completion, they may have time to carve out 1-2 weeks to work on a side project, then return to the dissertation. Sometimes a project is relevant to, but doesn’t really belong in, the dissertation. This is also something I did (and the result was my first article on luck).
Of course, none of this is to say that I necessarily disagree with Millikan’s point. (Although I do disagree, in some respects.)
I agree completely. (Were you one of my dissertation advisors? If not, I hope you’re on one of the hiring committees reading my applications this year.) But honestly there’s no way the discipline can collectively stand its ground on this without first standing its ground on this: http://dailynous.com/2014/10/07/advertising-exploitative-positions/
For the reasons Thom Brooks gives above – and which have been given pretty much every blog thread about the job market for the few years – there are too many qualified applicants for each position, and hiring committees already have plenty of work to do with their own (often increasing) course loads and various administrative and service tasks. If some candidates have outstanding publications, that obviously makes it very easy for them to stand out early in the sorting process. And if you have enough of those, and the rest of their dossier looks good enough, why kill yourself trying to sort through the other 100+ applicants who don’t have publications?
The problem is precisely that to get any tenure-track job you have to be able to present yourself not as an excellent, well-qualified candidate, but precisely as Thom Brooks says, an “exceptional” one. As long as having secure, professional wage employment continues to be the exception for young philosophers, this credentialing arms race will continue.
I would suggest from my own experience on search committees that graduate students should aim at one or two publications in journals that are considered “major” in their area of specialization. The main objective here is to attract the attention of committee members browsing through 50-80 files in round 2 and make them thoroughly read your writing sample and other material available online. The overall number of publications, and the visibility of a journal within the general profession, will be a factor within the committee as a whole or the department, but not the only one.
The solution is for journals to coordinate on refusing to publish graduate student work, as suggested in this essay in the Chronicle: http://m.chronicle.com/article/An-Open-Letter-to-Journal/149199/ . That would eliminate the competitive disadvantage of not publishing as a graduate student and let students get on with their dissertations.
Wow, that’s a thought-provoking proposal. It would be extremely hard to achieve such co-ordination however – there are so many journals, and I think a substantial proportion of them would feel that the proposal was objectionably paternalistic.
But how would people feel if departments were to forbid their graduate students from publishing? The co-ordination problems there don’t look insuperable – if one department did it its students would be at a disadvantage, but if four or five top departments did it together it is plausible that the benefits to their students would already start to outweigh the costs. (And as more departments signed up the costs would decrease further.) Moreover, the objection from paternalism doesn’t seem so strong in this case, since departments are already in the business of regulating their graduate students’ professional lives for their own advantage.
In fact, it seems plausible that if a group of leading departments agreed that they would merely *discourage* publication, rather than forbidding it, that could still have a huge and beneficial effect in relieving the market-related pressure on their students to publish.
By what mechanisms would departments enforce such a policy? Grad students don’t need their advisor’s permission to submit to a journal.
In any case, the most serious worry isn’t paternalism (forcing someone to do something for their own good), but rather forcing someone to do something that is against their interests for the sake of the greater good (insofar as having publications will provide them with a greater chance of being able to have a career in philosophy).
Well, departments have lots of policies about what their graduate students should and should not do – for example policies to the effect that all graduate students must complete N papers within their first M years. If a department wanted to introduce a rule against publishing it could enforce it in the same general way that it enforces those other rules. However I think that for now it will be more useful if we discuss the idea of departments agreeing to *discourage* publication by graduate students, and set aside the idea of them *forbidding* publication. Mere discouragement would not need to be enforced – it could simply be incorporated in the department’s general advice to students, and explained to prospective employers in a statement accompanying every job application.
‘Paternalism’ may not be the right word, but I definitely agree that there is something prima facie problematic about forcing people to do something against their own interests for the sake of the greater good, or even discouraging the from doing something that is in their own interests for the sake of the greater good. That is why I was focusing on the proposal that a group of departments would co-ordinate on the relevant policy, rather than adopting it individually. I think it is plausible that once the group of departments becomes large enough (I think four or five should do if they were really “top” departments), the adoption of such a policy and its publication to prospective employers would mean that it would no longer be true that publishing while in graduate school was in most students’ best interests. Prospective employers, wanting to leave open the chance of hiring the excellent candidates from those departments, would need to change their approach to hiring in order to give less weight to publications, at least for candidates from those departments. So the cost to the students of not publishing would be reduced, hopefully enough that the costs would no longer outweigh the benefits of being less stressed about publication, having more time to focus on the dissertation, and being able to complete graduate school more quickly.
CD, I think that’s a great idea. (And, David Velleman, thanks for the link to the excellent article.) A number of faculty in my department feel genuinely divided about how to advise our graduate students regarding publication — we think it’s better for them to concentrate on writing a really good dissertation, but we’re worried about their being at a disadvantage on the job market. It would be such a good thing for our students and the profession if some of the “top” departments could just agree on a policy of discouraging publication, and could make it known that this is what they were doing. Enough already with the out-of-control publication arms race!
I’m really amazed by some of these comments. I know things are different in the US, but in the UK if you don’t have publications coming out of your program, you’re going to be unemployed until you get a few. Perceived rankings of journals are important, but there is a quite a lot of disparity in people’s opinions. So, for us the thing to do is not to be afraid to send papers out for review and to aim for a range of top journals according to Leiter and others. Mind, Phil Review, Journal of Phil, Nous, and PPR are generally perceived as being the best places to be published in, but getting an article in one of those places is rather unrealistic in my opinion. Further, except for Nous and PPR these journals are really slow. So, aim one tier down and send off your best work. Then when you get rejected rewrite in light of the comments. Then try again.
Seeing the CVs of those who manage to attain permanent positions or 2-3 year postdocs in the UK, it’s publish or perish!
Interesting, but my understanding was that this is already in practice. Top departments encourage their students not to publish, because it is easier to ride pedigree to a TT than to spend immense amounts of time publishing only to risk an unimpressive result. No reason to show hiring committees you aren’t as amazing as your pedigree halo suggests 😉