Dissertations

Here’s an odd fact that about philosophy and linguistics, one that I probably should have noticed a while ago.

In linguistics, or at least in semantics which is the only area I’m particularly familiar with, it is quite common to see PhD dissertations cited in research articles. This is true even when the dissertations have been turned into books. (Which they often are, and which are often widely cited.) To take one prominent example, I think the canonical work on negative polarity items is still William Ladusaw’s 1980 PhD dissertation, which is (or at least should be) cited in every paper on negative polarity.

In philosophy this kind of thing is very rare, at least in the areas in which I work. I can’t remember the last time I saw a dissertation cited that wasn’t written by one of the authors of the citing paper. (Perhaps there were some were the dissertation was by a student of the citer, but I can’t even remember one of those.) And this isn’t because dissertations are published so the books that come out of them are cited. In the areas I work in, many if not most people do not publish their dissertation as a book, and those that do are often much less widely cited than the journal articles by the same authors. (There’s one prominent exception.)

(UPDATE: Gil Harman pointed out several counterexamples to my claim that philosophy dissertations don’t get cited. Historically, there have been lots of dissertations turned into important books – Lewis’s Convention, Nagel’s The Possibility of Altruism, and Katz’s The Problem of Induction and Its Solution are particular prominent examples. I had forgotten a couple of those, but these weren’t the cases I was really worried about, because I suspected that things had changed significantly in the last three decades. But he also quickly noted examples of citations of recent dissertations by Maria Merritt, Peter Turney, John Doris and Sarah McGrath, which really do constitute counterexamples to the thesis I was advancing here. One may well conclude this undermines my argument for some of the conclusions below.)

I think there’s an obvious conclusion to be drawn from this for philosophy graduate students. Don’t worry about your dissertation too much. If you go on the job market with (a) one good paper in a good journal, (b) another very good paper for a job talk and (c) some roughly worked out papers that you can polish into publications after you get a job so you quickly look like you have a research output, you’ll be doing fine. Getting (a), (b) and (c) down well will do much more for your career than worrying about the precise formulation of condition D” on page 197 of your (little-read) dissertation. And in any case, if you want to write a big book on the topic of your dissertation, you’ll write a better big book with a few years experience behind you than you will fresh out of graduate classes.

And there’s another conclusion to be drawn from this. You can complete (a), (b) and (c) well within the 5 years of a normal PhD program. Which means you shouldn’t take any longer than that to complete. Which means you should choose graduate schools who make a point of getting their students educated and placed on the job market with all due haste. In philosophy at least one can make a very accurate prediction of how long someone will take to complete their PhD knowing virtually nothing about them except where they go to grad school. I used to think it was a virtue of certain schools (Princeton and MIT being prominent examples) that their students frequently completed on time. On further reflection I think I was underestimating how valuable this is.

I have no idea whether other disciplines are more like philosophy or more like linguistics in terms of how frequently PhD dissertations get cited. It would be interesting to know whether there are any patterns to the practices in different disciplines.