Should graduate students be encouraged not to publish?

29 Replies to “Should graduate students be encouraged not to publish?”

1. I’ve never heard that, Ralph.
   When I was in grad school, we were not encouraged to (try to) publish, but we were not discouraged from publishing. It was quite unusual and not expected.
   I think now maybe some people at some programs are told that they are better off spending their time polishing a writing sample, finishing their dissertation, preparing a job talk, then getting a paper ready for publication.
   Anyway, no, I don’t think the argument you mention is remotely persuasive either.

2. I worry that this advice might be correct from a one mercenary perspective – it does seem to be true that many junior hires at the top places have published nothing – but I think it should be ignored anyway.
   Why would you spend five or more of the best years of your life studying and writing philosophy, while setting out NOT to contribute to literature? It seems to me that someone who thinks like that – someone who is prepared to sacrifice their activities as a philosopher in order to increase their chances of getting a good job – is over-confident about their chances of getting any job at all (the grad school years may well turn out to be your best chance ever to publish in philosophy), and not the sort of person who should be pursuing an academic career.
   That’s not to say that everyone should publish while at grad school – it depends what you’re into, what material you have, etc – just that no one should decline to publish just so as to improve job prospects.

3. I thought graduate students and others new to the business got their feet wet by writing book reviews and other short pieces that did not carry much weight. The purpose of this “getting into the shallow end of the pool” is to start to learn how to swim.
   Why then would this work be counted against anyone? Shouldn’t the initiative and work done be counted as points for their side?

4. I’ve been given advice along those lines (though not exactly as you put it). One version I’ve heard is that graduate school is a time during which your intellectual development is (hopefully) proceeding very quickly. Because of this, your best work at the end of graduate school (hopefully represented by your writing sample, job talk, and dissertation work) is likely to be significantly better than what you’d be able to produce at earlier stages.
   Basically, the thought is that the average quality of your work that matters to search committees more than the total quality, and that the average quality is maximized by limiting the writing available to them to your best, most polished, late-graduate-school work.

5. Yes, that’s definitely a very bad argument, Ralph. Indeed, it borders on not being an argument at all.
   Another reason for graduate students to not try to publish: so I don’t get asked by journals to referee so many seminar papers. I suspect that editors tend to send such papers to younger junior faculty like me.
   In the great majority of cases, I think graduate students probably shouldn’t be trying to publish a paper, unless one of their professors — or someone else established in the profession — said that it’s worth publishing. The only thing I published in grad school was something Jamie read and said, “Looks right to me.” (Jamie doesn’t say that lightly.)

6. It looks to me like Ralph has offered three different arguments. Arguments 2 and 3 look lame to me, but argument 1 relies on an element of truth. I do think that philosophers tend to suffer from a cognitive bias when it comes to comparing job candidates’ published work to unpublished work, particularly when it is accompanied by over-the-top letters of recommendation. It’s simply easier to forgive some kinds of shortcomings in work in progress, especially when backed up by over-the-top comparisons from people you trust, a fact which people are not good at overcoming. My guess, however, is that at most one or two people a year are in a position to benefit significantly from this kind of advice, and that those one or two people don’t really need its payoff.
   A fourth argument against publishing which I have heard, is that papers published before you begin your tenure-track job may in some cases be discounted or ‘not counted’ toward tenure, so if you held onto the same paper and published it later, it would count toward tenure.
   Again, I’m not impressed by the argument; the more experience you have going through the process of getting your work published, the easier it will be later times around, so unless you are afraid that you simply won’t have enough ideas before you come up for tenure, getting started on learning what it takes to polish something sufficiently for publication, what it takes to make referees (very similar to what it takes to make readers) interested and happy, and how to manage your time and energies while waiting for things which are out under review, the better you will get at these things before your tenure clock puts you in crunch. Moreover, the earlier you get your work published, the more chances it has to be timely and generate a response, and the getting responses to your work is just as important for tenure (and what the whole point is, anyway).
   When I was a grad student, I followed Simon Keller’s example, and tried a lot to publish. I didn’t have any luck at all until after I left grad school, but I don’t think a lack of publications helped nab me my first job. I did, however, learn more from the process than from anything else I did in grad school.

7. Whatever Mark did obviously worked really well!
   Everything about the educative benefits of going through the process definitely seems right. But I think that should be balanced against the burden it places on journals and referees.
   P.S. Mark, do you really think that the whole point of publishing is getting responses (or was it tenure?)?

8. Ralph,
   I don’t find it persuasive either, and I find it hard to believe that graduate students are being dissuaded from publishing. The applications I see for junior positions–from top schools and otherwise–often show remarkable productivity. I don’t get the idea that their papers will be judged with greater severity, since the venues often speak for themselves. I do understand not publishing your work in graduate student journals or in poor journals. But that’s obvious. Maybe a case could also be made for not publishing in specialized journals. The only argument for not publishing that has some traction is that it might make your application too good for departments that focus mainly on teaching. I would add that I do not know of one case where someone has actually been overlooked for that reason, and not publishing for that reason might be too clever by half.

9. I’m intrigued that Jamie has “never heard” of graduate students being discouraged from publishing. (Perhaps this reveals a difference in culture between Princeton, NJ, and Cambridge, MA?)
   Anyway, Mark is right of course that the argument that I discussed involves three sub-arguments. This is what I see as the main flaw in Argument 1. Published work may well be judged by more stringent standards than work-in-progress, but (i) this is probably a very small effect (and it will be counteracted by the way in which the very fact that the article is published may boost the readers’ judgment of it), and (ii) as Mark notes, it is only likely to make a difference in a very small number of cases (“at most one or two people a year” according to Mark’s estimate).
   None of the PEA Soupers is impressed by Argument 2, although I have heard it advanced by at least one distinguished philosopher. But this argument just seems to rest on an empirically false premiss. The published work may well be salient in the minds of the search committee at the time when they are deciding whom to interview. But at that stage the committee will have largely forgotten about the letters of recommendation (because more candidates get wildly enthusiastic letters of recommendation than the committee can possibly interview), and so the committee will then be focused on comparing the candidates’ samples of written work anyway. After the interviews, the salience of the written work will normally be swamped by the mass of information (and sheer noise, as Gil Harman reminds us….) that is generated by the interview.
   As for Argument 3, well, I admit that I was poking fun at it, but it was basically meant to be a version of Daniel’s point. However, I think this argument just misunderstands the nature of philosophy. Many graduate students have quite brilliant original insights (often they’re small points, of the kind that would make an excellent Analysis paper rather than a vast synoptic vision of the whole of philosophy, but they’re none the worse for that), and these ideas are eminently worthy of publication.
   The argument that Mark labels Argument 4 sounds bizarre to me. Do American tenure committees really refuse to consider papers that were published while the candidate was in graduate school? If so, then they’re even more irrational than I had thought…
   Finally, we could consider Argument 5, which is suggested by what Jamie says. Preparing articles for publication takes time, and it might be thought that time is better spent perfecting one’s writing sample and job talk. But the obvious problem with this argument is that the published article can presumably also be the writing sample. So it’s not clear that there is such a big opportunity cost to trying to get articles published.
   A final weakness in these arguments (considered as reasons for advising graduate students not to publish) is that there are also some quite strong counterarguments in favour of graduate students’ trying to get things published. Mark and Simon have articulated these counterarguments quite eloquently, it seems to me.

10. Agreed – it’s not a convincing argument, Ralph.
    Published work should be judged for the quality of its content; this quality is not always determined by how long one has been doing philosophy for.
    Although there may be a connection between practicing a skill, and perfecting that skill, this connection is a qualitative one, not a quantitative one. A carpenter, say, who has trained badly for ten years may produce work of lesser quality than someone who has trained well for three years. People should be judged for how hard & well they’ve work, not how long they’ve work for.
    Philosophy (and thinking in general) as a skill, is no exception.

11. I was actually discouraged by some faculty members when I was in grad school from publishing (or trying to publish, I should say). The stated reason wasn’t anything like the argument(s) you originally gave, though, Ralph. Instead, it was more akin to Jamie’s suggestion: doing so will seriously detract from the limited funded time you’ve got to finish your dissertation (nothing about writing samples or job talks was mentioned).
    While the stated reason had some merit, my fellow grad students (who received similar advice) and I tended to think there was some disingenuousness or misunderstanding involved, for the faculty giving the advice had themselves come from tip-top-notch grad schools and had been heavily recruited with multiple offers just out of grad school with no publications themselves. Thus they seemed to be under the (mistaken) impression that the same would be true of us: just get the dissertation done and all good things will come. This attitude involved a failure to understand (a) the non-tip-top-notch nature of our grad program, and (b) the new realities of the job market. Of course, with respect to (b), there are still those who come out of the very top grad programs who get (tenure-track) jobs with no publications, but I think that’s much rarer than it used to be. For one thing (and this should be the topic of another post), there’s much more mobility on the scene these days, especially among junior faculty. So freshly-minted Ph.D.s are having to compete with many slightly older assistant professors who’ve had a chance to publish since graduating (usually bits and pieces from their dissertation). Unpublished Ph.D.s therefore are at a significant disadvantage in this relatively new environment.

12. David, I think you’re right. I think it depends very much on the university from which you’re graduating. This advice is common at the top schools, less common at middle ranked schools. Students from all-but-the-best schools benefit from publishing. But they don’t need to publish a lot. One paper n a very good journal is all it takes. I do give a version of this advice to my students. I tell them it’s not worth very much (and so not worth the time and effort involved) to publish in not very good journals. Better to go out on the market with the promise of great things, then with certain mediocrity.

13. Samantha’s comment seems right to me. Coming from a non-tip-top-notch university, the advice not to publish sounds insane. All my experience (on both sides of the interview table) indicates that, in a crowded field, those with no publications get their dossiers circular-filed.

14. Most of what is published isn’t worth reading. Graduate students are more likely than others to publish papers that are not worth reading.
    Therefore, there is a reason not to encourage graduate students to publish. Of course, there may be other reasons to encourage them to publish, but this certainly seems like a reason against doing so to me.

15. May I just say that if it’s true that Ph.D.s from non-elite programs need to publish to compete well in the job market, then non-elite Ph.D. programs should really restructure themselves to promote more publishing by their grad students. Otherwise, they’re not well serving their students.
    It’s worse than disappointing if search committees would view publication in less-than-very-good venues as indicating “certain mediocrity.” One would hope that they’d at least read the paper before rendering such a harsh judgment. But I know — that’s probably asking too much.
    (BTW, I understood Samantha to be merely reporting the practice, not endorsing it.)

16. I think David S. is right, too. But Samantha has left out DS’s (b) in her agreement, and I think that’s a huge factor. In 1988, very few grad students had published anything when they went on the market; this year it’s very common.
    Ralph:
    I’m intrigued that Jamie has “never heard” of graduate students being discouraged from publishing. (Perhaps this reveals a difference in culture between Princeton, NJ, and Cambridge, MA?)
    Well, maybe. But to be clear: I think when I was in NJ nobody was discouraging grad students from publishing; it wasn’t really expected and few students did it. So the culture difference between Cambridge and Providence might be more to the point. John, did Jaegwon or Ernie ever discourage you from publishing something?

17. Jamie, no, neither Jaegwon nor Ernie ever did that with me.

18. To answer John’s question, yes, to me the point of publishing is so that people can read your work. Not necessarily to ‘get a response’ as in, to get someone else to write about it, but certainly to tell other people about your ideas and have them notice. I can’t say that I was ever that caught up on the significance of tenure; I chose to come up for tenure early, in fact, so that I could get more people to pay attention to my work sooner – which is one of the perks of both applying for jobs the first time around, and going up for tenure.
    I don’t want to be understood as disagreeing with John or Nishi’s concerns about things being sent out before they are ready, or which aren’t worth publishing. I do think that many graduate students today try too hard to publish something before it is ready, and without guidance. I don’t think there should be a lower bar at which graduate students should try to publish their papers, any more than I think there should be a higher bar. I just think that it’s part of learning how the process works to try and learn what the appropriate bar is. The bar is never that you can get it published, so publication shouldn’t be an end in itself. But if you have work that is promising, part of the process is learning how to turn it into something that is worth publishing. It’s something you have to learn eventually, and waiting until two or three years into your tenure clock doesn’t give you a lot of time.

19. Most of what is published isn’t worth reading. Graduate students are more likely than others to publish papers that are not worth reading. Therefore, there is a reason not to encourage graduate students to publish.
    I can’t see it. I don’t think there is any interesting sense in which most of what is published isn’t worth reading. It is not as though even 10% of philosophers could come to any agreement on what’s worth reading, even if we conceded that every one of them would agree that better than half of the papers aren’t worth the effort. But aside from that, how does the conclusion follow that “this presents a reason for graduate students not to publish”? All that statistic would show (were it true) is the chances of randomly selecting a worthwhile paper is less than 50%. It certainly would not show that any particular person’s chances of writing a worthwhile paper are low. If 60% of dogs in Kansas are terriers, we know that a random selection of a dog would probably yield a terrier. We couldn’t conclude that you shouldn’t move to Kansas because your particular dog would then have a 60% chance of being a terrier. So even if 60% of all published papers are bad, that does not tell us that anything grad student Smith writes has a 60% chance of being bad.

20. Yes, the job market has changed. And most of the people we hire as assistant professors have published. My only point is that they needn’t have published a lot. And quality matters more than quantity. Five papers in five so-so journals does suggest a career path, a trajectory. It’s much better to have 1 paper in a top journal if that’s possible. We also care about the number of publications relative to career stage and so try not to compare newly minted PhDs to people who’ve been in tt posts for a few years.

21. Mike,
    If it is true that not even 10% of philosophers could largely overlap in their judgments about what is worth reading, at least about areas that they have expertise in, that is a very sad fact indeed. (Shouldn’t anyone coming up for tenure be very worried if this is true?) I very much doubt that it is true, but maybe that is wishful thinking.
    Let us assume that there is an interesting sense in which most of what is published is not worth reading, and further, that graduate students in general are less likely than others to produce such work. What follows, I think, is that departments have a reason not to adopt a policy of encouraging their graduate students to publish. Individual advisors may have reason to encourage particular students to publish, but what I am concerned about is creating an atmosphere in which students come to believe that they ought to publish, whoever they are. I do think this is the situation at some graduate programs. But maybe that isn’t due to any general policy that these departments have adopted, I don’t know.

22. If it is true that not even 10% of philosophers could largely overlap in their judgments about what is worth reading, at least about areas that they have expertise in, that is a very sad fact indeed. (Shouldn’t anyone coming up for tenure be very worried if this is true?) I very much doubt that it is true, but maybe that is wishful thinking.
    Well, maybe 10% was an overstatement. Let’s say 25%, and I’d keep my claim the same. But let’s assume–as you say–that there is an interesting sense in which there is broad overlap among lots of philosophers. I think you can conclude the following: for any arbitrarily chosen graduate student P in philosophy, it is likely that P would not do well in trying to publish. But I don’t think we should conclude further that the right advice for P is not to try to publish. Why not say instead, ‘if you’re going to send your work out, send it to PQ or Nous or PPR or PS or AJP . . .”. Something like that. We’ve assumed that most papers are not worth reading, but certainly it’s not true that most papers in those journals are not worth reading. If you can get it placed in one of these (and of course there are many others in roughly the same ballpark), then the chances are, I think, it is worth reading. I guess I’m recommending that we let the journals decide.

23. Samantha,
    “Five papers in five so-so journals does suggest a career path, a trajectory.” For a graduate student?
    I trust that you’re correct that it does in fact suggest such a thing to many SCs, and so it’s probably prudent for your students to heed your well-informed advice.
    But here’s why I think SCs should not think that way.
    Even so-so journals publish some good papers. And such a track record might indicate an unwise publication strategy, rather than any defect in the work itself.
    In my own case, here’s how I’d settle the matter: I’d actually read one or more of the papers and evaluate them before conjecturing that this person does mediocre work. (This is especially true if the person’s writing sample was clearly very good.)
    Again, I recognize that many SCs might not be willing to put in that much time and effort, given their already busy schedules.

24. Mike: If only it were the case that the journals had the capacity to decide. Rather, it’s the overworked editors of those journals that have to do so, based on the comments of overworked referees like you and me.
    The hard question, then, is this: is there a way to help graduate students get a leg up on the marketplace by publishing more (in quality journals) without increasing the burden on editors and reviewers? Answer: not bloody likely.

25. A data point: everyone hired to a TT position at Penn in the last 8 years had at least one publication before being hired. Some had several. Nearly everyone brought in for a job talk had at least one publication. Now, maybe this is just correlation- they might have all had publications because they were good young philosophers and also got the job offer for the same reason but w/o the publications being factors in the offer (or job talk offer) but this seems unlikely to me. Given this, it seems that there is some reason to encourage grad students to publish if their papers are in fact good enough.

26. Should grad students be encouraged not to publish? Yeck yea, how am I gonna get something published if these smart whipper-snappers are crowding me out?

27. This and other current online discussions, combined with many experiences of those without so much as nibbles on this year’s job market, strongly suggest that Weatherson’s advice is no longer good advice:
    “People tend to get hired based on their best papers…As we see every year when looking at junior hires, it doesn’t really matter if that best paper was published in Philosophical Review, the Proceedings of the Philosistan grad conference, or (more likely) the candidate’s own website. What matters is how good it is, or appears. As a rule, spending more time improving your best paper will do more for your professional prospects than sending it off and moving on to another paper.”
    Of course it’s good a great writing sample, but if you want a job, you better have a publication. Trust me, it’s already been a long winter for those of us unpublished whipper-snappers.

28. For those who are not looking at both sites, there is another line of discussion going on at Certain Doubts on this topic:
    http://fleetwood.baylor.edu/certain_doubts/?p=899
    One could get the impression, there, that no publications means no interviews/offers/future. I would not think it a good policy to tell all graduate students to not publish, but I think it is unfortuante if we are telling them they cannot get an interview if they have not published at least one article in one of the so-called best journals.

29. Ralph asks whether American tenure committees really refuse to consider papers published while the candidate was in graduate school. One has to make a distinction between departmental tenure committees and committees (humanities-wide, or arts-and-sciences-wide) evaluating recommendations coming up from departments. These committees, and Deans, will normally consider a tenure candidate’s entire publication record, but will also look hard at what the candidate has done in the years since he or she was hired, or (more accurately) what is in the CV now that wasn’t there when the candidate was hired. That earlier CV may actually be looked at, to make these judgments. There is the idea that the candidate doesn’t get credit ‘twice’ for the same publication. If she was hired on the basis of a fine book, and has had five years since then, the promotion to Associate Professor is based largely on what has gone on since the book. The same is true of promotion to Full Professor — you don’t get promoted to Full Professor on the basis of material that was in your dossier when you were evaluated for Associate Professor. Departments themselves are not likely to think this way about tenure cases: they will ask how good the philosopher is, and how good the whole publication record is. And a strong departmental recommendation from a well-regarded department will overcome objections from people in a Dean’s committee, counting up what a candidate for tenure has done since she arrived. But it is equally true that senior people in departments, who have pushed candidates to unfriendly Dean’s committees, know that they may have a difficult case to make if a candidate’s best publications were the ones that were already in the CV when she was hired. So: bizarre as it may appear, Argument 4 is not ungrounded.