On the process of writing a dissertation – Today I submitted a draft of the Dis (i.e., “dissertation,” for those of you joining us mid-program) to my committee members. I have been using the metaphor of hurdles for each of the steps along the way to getting my doctorate, but today my advisor used the phrase “to tie up a knot.” I like that. A hurdle is just another obstacle along the way, but tying a knot makes it sound as if I’ve completed something. And I guess I have. There are still more hurdles ahead—or more knots to tie, as the case may be—but finishing and submitting the “evaluation copy” is a big one.
I have a rough idea of what happens next. My evaluation (or defense, if you will) will take place on 23 June. The final deadline, when the final copy must be submitted to the department, is 20 July. You’ll notice that there is less than one month between those dates. What exactly this means, I don’t know, but I do know that I will not have as much time to make corrections as I had after the presentation. Thankfully, the semester will be over by then, so I will be able to focus (almost) all of my energy on tying up the last few knots.
I’ll post more about the details when I get them, but today I wanted to take a quick breather and step back a little. I said at one point that I would write about the process of writing the Dis, and I had originally thought that I would do this after everything was done. In a way, this makes sense—I will have the benefit of hindsight and no pressure. But I have realized that when the Dis is finally done, I’m probably not going to want to spend too much time writing about it. There is one entry I’d like to write about what makes a successful Ph.D. candidate (basically a response to “3 qualities of successful Ph.D. students,” a link that I posted last time), but I don’t know if I’ll be able to manage that before I start itching to move on to other things. So if I’m ever going to write about the process, the best time to do that is probably right now.
Although I am not yet finished, I have learned a few important lessons about what it means to write a doctoral dissertation. Obviously this is coming from my own personal experience, so not all of this will apply to every dissertation. I do think that there are some fundamental and possibly universal ideas here, though.
The first thing I struggled with was “topic creep.” Programmers and designers deal with something called “feature creep,” which is essentially a project that becomes bloated as feature after feature gets added. Even though I do not do much programming, I’ve dealt with this first-hand—namely when helping my brother create a website for Promethean Candle Boutique, his online candle shop. (To be precise, what I did for his site was scripting as opposed to programming, but the principle is the same.) We started out with a pretty good idea of what we wanted to do, but as time went on we started thinking of new features and functionality to add to the site. It got so bad that I developed a nervous twitch that acted up every time my brother said, “I’ve got an idea.”
When it comes to a dissertation, topic creep is similar. You start out with a fairly clear idea of what you want to do, but as you progress you begin to think of new areas to explore and new directions to take. After all, this is not just some paper to be published in a journal. This is your dissertation, your magnum opus, your trumpet blast from the top of the ivory tower. So it’s only natural that you want to make it as impressive as possible.
This is all well and good, but I learned very quickly that I had to hold back on these ideas and fight against topic creep. It would have been very easy to allow the Dis to bloat beyond my ability to handle it, and in the early stages the biggest challenge was paring things back to the essentials. This meant giving up on a lot of great ideas, but that became a lot easier when I realized I wasn’t actually “giving up” on these ideas. I’ve been working on the Dis for so long that it had become difficult to see past it. It was my own personal event horizon. But gradually, and through talks with my advisor (this was my former advisor, not my current advisor), I realized that the Dis was not the end all be all of my academic existence. It was, in fact, just the beginning. Trying to do everything in the Dis was not only impossible, it was also foolish—it would be like an up-and-coming singer trying to contain the entirety of her artistic potential in her debut single. Realizing this made it a lot easier to say no to topic creep.
The second thing I struggled with is something that may be more common in humanities research. I’m not too familiar with the sciences, and it may indeed be a problem there as well, but it strikes me as something almost endemic to the humanities—namely, the problem of categorization. Categories are something of a paradox. We create them so that we can better see the naturally-occurring relationships between things, but the categories themselves are artificial and never fit perfectly. As anyone who has ever tried to categorize anything will know, there are always exceptions or anomalies. The tighter you make your categories, the more anomalies you have, and the more anomalies you have the more difficult it is to achieve meaningful results based on these categories. But if you make your categories loose enough to encompass all exceptions and anomalies, these categories become useless for research purposes.
My Dis deals primarily with 267 tales featuring three different protagonists. In order to deal with them effectively, I had to categorize them. The first step was creating a catalog of “episodes” found in the tales. Many of the tales consisted of a single episode, but there were a few that consisted of more than one episode (the longer tales had up to ten episodes). After going through each of these 267 tales with a fine-toothed comb, I ended up with 379 episodes, and these could each be classified as one of 129 episode types. This took a very long time, as you can imagine, but it was critical to my research—the tales themselves were of arbitrary length (sometimes the researchers would break them up, sometimes they wouldn’t), so the only viable unit of meaning was the episode.
Once I had finished the classification, I was then faced with the task of categorizing these episode types. I ended up with ten categories that were then grouped into four super categories, with two categories left over. I won’t bore you with the details, but how I categorized these episodes would ultimately determine the structure of the Dis, so it was very important.
As I mentioned above, the problem with categories is that they are always going to be imperfect in some way. To make matters worse, your categories are going to be very visible, since they will be reflected in the structure (read: table of contents) of your dissertation. This makes them low-hanging fruit when it comes to criticism, and every step along the way—whenever I presented my research for criticism—my categories were called into question. This continued all the way up to the most recent presentation, and each time I tweaked the categories. This last time I actually tweaked one category significantly, but even though it was quite painful and required rejiggering a full twenty or so pages of text, it turned out to be for the better.
I’ve tried my best to make these categories as airtight as possible, and I think I’ve made significant improvements from my initial categories, but I fully expect them to be called into question once again during the evaluation. The difference now is that there is no more room for retreating. I have made my adjustments and modifications, some drastic and some less so, and I am confident enough in what I have that I can say it’s time to stand and defend my choices. Which is not to say, of course, that there might not be some tweaking in store, but if I am forced to make major changes I will have failed (and probably will not be able to submit the Dis this semester).
I could probably go on, but it’s getting a little late (or at least what passes for late these days) and I’m getting tired. I can feel my brain starting to shut down, so I think it’s time to wrap this up. I wouldn’t call today’s entry an exhaustive discussion of the process, but hopefully it did manage to provide some insight into the Dis and, in the process, fulfill the promise I made earlier. I suspect that I will not revisit this—what you see here will have to do, at least for now.