Karl W. Butzer, Dickson Centennial Professor of Liberal Arts
I don't have any special points to make on the basic procedures of dissertation writing, but I do have some notions about writing a good dissertation. Two of my Ph.D.'s have received UT best dissertation awards in section A, and I have served on that university committee twice.
My ideas can be summarized as follows:
1. The choice of dissertation
supervisor. Ideally someone who nurtures
the student's ideas, rather than his/her
own; someone who is willing to take the
time to work with the student; someone who
is interested in ideas more than playing
the role of a taskmaster. Don't pick someone
solely because he/she has a national reputation,
or even worse, because he/she is 'easy'.
A good dissertation commonly takes years,
and afterwards the supervisor must write
countless letters; so be sure that the 'chemistry'
is right, and also that the supervisor can
be relied on to help in placement.
2. The choice of a committee.
At least one of the intradepartmental members
should have an interest in the dissertation
topic or in your education, and be sure
to develop a productive relationship with
that committee member. That does not mean
sporadic, impromptu demands on that professor's
time: a relationship is a two-way street
that involves consulting and discussion,
after taking some coursework to allow you
to understand his/her viewpoints and methods.
It also implies presenting a reasonably
finalized draft at least two months prior
to any scheduled defense. Suggestions offered
for revision should be taken seriously,
even when you disagree on something. Disagreement
can and should lead to productive results,
not hostility. If at all possible, the outside
reader should be chosen according to similar
criteria, and treated in a similar manner;
and there are obvious advantages to selecting
someone on-campus rather than out-of-town.
Remember that most committee members will
need to write letters for you for years
after the degree is awarded.
3. A real introduction.
A dissertation should not be written only
to address the interests of a narrowly-defined,
subdisciplinary constituency. The first
half or so of the first chapter should be
intelligible to faculty in at least several
departments. That can be done by explaining
the issues to be addressed and why they
are interesting or important, using a minimum
of jargon (or plays on words). That is beneficial
to the dissertation, it generates interest
for other committee members, it helps greatly
in subsequent placement, and more often
than not it is critical for publication
as a book. That should all be obvious, but
most students wear blinders in their subdisciplinary
introspection. Half of the dissertations
recommended by departments for a University
Dissertation Award don't make the first
cut because the writer has failed to articulate
his/her intellectual quest.
4. A real conclusion. A
dissertation in the social sciences or humanities
should consist of more than a set of chapters,
terminated by a summary labeled as 'Conclusions'.
That happens all too easily, because people
do run out of steam. All too many dissertations
fall flat at the end. After a three-week
break, you should rewrite your concluding
chapter from scratch, making sure that it
integrates all of the threads developed
in the various chapters, and focuses them
on the 'thesis' being presented and defended.
The concluding chapter should represent
the cumulative body of insights that you
have touched on or dealt with in depth.
This is the best opportunity to articulate,
develop, and expand on ideas, to discuss
what it may mean, and what one might attempt
to follow up, in order to achieve greater
understanding in the future. Keep in mind
that this is the best occasion to identify
who you are and what you have to offer.
5. Reformulating your introduction. Finally, after completing that really good concluding discussion, go back and rewrite the introduction, because by now you can do it much better and focus it more strongly.