Dr. Karl Galinsky
Department of Classics
(Floyd Cailloux Centennial Professor of Classics, University Distinguished Teaching Professor)
Foreword: given my experience mostly with dissertations in the arts and humanities, my advice is directed primarily at students in these areas though it may be selectively useful to others, too:
1) Keep it in perspective. To appropriate the title of a current dating service for busy professionals ("It's Just Lunch"): it's just a dissertation, and not a book. I'm saying this for several reasons. To begin with, American universities are a hybrid of the British model at the undergraduate level, and the German model at the graduate level. In Germany, it was--and to some extent still is--obligatory to have the dissertation published. Alas, dissertations read, well, like dissertations. And do not expect university presses to beat down your door eagerly to partake in the fruits of your research; in fact, most American university presses have cut down drastically on publishing books that, even with the professed "revisions" still quack like dissertations.
This does not mean that a good book can't originate with a dissertation. Some do, but that's a different project. You write the dissertation to validate your credentials of being an effective and knowledgeable member of your profession. It's a springboard for further exploration of the field, a beginning rather than a culminating achievement. The extraordinary length to which graduate study and dissertation writing have ballooned have kind of anchored the latter notion. That's just as false as the equally entrenched notion that tenure is the pinnacle instead of being a promissory note for further productive activity.
2) A dissertation is a work in progress. It's the process of discovery that matters. Yes, you should have a well thought-out conceptual framework and not simply write as you go along. But be wary of the pitfalls of a thesis. Thesis too often can mean that you're trying to straighten out the messiness of any kind of human experience and make it fit into tidy academic compartments. Whatever may contradict the thesis then tends to be given short shrift. True to human life, of course, most topics are more complex. So don't be afraid to change course along the way. The voyage is more rewarding than the destination.
3) To combine (1) and (2): no, your dissertation will not be the final word. As early as two weeks after handing it in, you may already have second thoughts about some of your conclusions. Good. Our perspectives evolve both per se and in the light of new evidence. Some of the best scholars revisit topics on which they have written and that's a sign of maturity, and not of flip-flopping.
4) Don't be afraid to take on a big topic to begin with. As you proceed, you'll have to concentrate on one aspect or the other, but the bigger picture should always be in sight.
5) The real choice is not what to put in, but what to leave out. As I said earlier, a dissertation is an essay in the original sense of the word, i.e. an attempt or enterprise rather than an encyclopedic, alpha-to-omega package. Once you reach page 240 and have demonstrated some good insights, stop and finish up. The material you are not using may exceed the material you are using: here is the material for your next articles, or, if you must, the expansion of the dissertation into a book for an audience that is larger than your dissertation committee.
6) As for the committee, stay in frequent touch with all of its members. Make sure to make this into a good experience for all concerned, including yourself: a shared dialogue and shared excitement at new discoveries and original perspectives.
7) Move along at a good pace. Dissertations don't always get better the longer you work on them. In fact, you may run out of gas and that's a drag for everyone involved. Keep the topic manageable and don't worry about not being able to say it all. Quality matters, not quantity, and you can always point out how your particular approach will be applicable to topics and areas not covered in your dissertation.