I have a confession to make: I enjoyed writing my dissertation.
I started writing it one spring when I realized that the notes I had been taking were really a rough draft. I set a schedule for myself: Between teaching obligations, I aimed to complete three pages each workday. I surprised myself how quickly my dissertation piled up at that rate, but a bigger surprise was how much fun it was. As I immersed myself in my material, ideas took flight in my head.
Memories of that time returned when I read Dian Squire’s published response to an open letter I wrote to journal editors in October. In that column, I suggested that we might make a better world if editors refused to publish scholarly articles by graduate students. Students could focus on their dissertations that way and finish their degrees sooner, and hiring committees could use those same dissertations to evaluate applicants, instead of measuring their publication records. Lots of readers (including a number of journal editors) joined Squire in disagreeing with me.
My main point was that the ever-escalating demands for publication extend the apprenticeship of graduate students longer and longer. Instead of accepting or excusing the status quo, we need to look everywhere we can for ways to fix this "tragedy of the commons," where people acting in their own honest interests wind up maintaining a system that brutalizes its least powerful participants.
Squire suggests that the problem isn’t the articles that graduate students have to write but rather the dissertation that competes with them. He calls it a "singular, extended Medieval hazing ritual."
I’m not so sure about that. Moreover, the dissertation has more important purposes than serving as a barrier to full membership in the scholarly guild. The dissertation isn’t supposed to be just a demonstration of learning. It’s a part of graduate education, meaning that you learn by doing it. I certainly did.
Still, the dissertation is a main ingredient in the stew of problems that plague graduate education in this country right now. Squire asks for a conversation aimed at "meaningful, radical reform of the dissertation process and [the] form itself." In fact, that conversation has been going on for awhile.
Any reform of the dissertation process starts at the intersection of two questions: What should a dissertation be? And how long should it take to write one?
Those questions are interdependent and inseparable. Most historians, for example, agree that a history dissertation should be an extended work of original—typically archival—scholarship. Given how long it takes to produce something like that, it’s understandable that historians have been slow to embrace a reduced time-to-degree (which stands at about nine years in the humanities).
The debate over how long it takes to earn a Ph.D. has some entrenched positions. "The protracted character of doctoral study burns out one’s scholarly interests," says one academic. Another argues that "the article-length dissertation is just common sense and is long overdue." Yet another warns that "it would be a serious error to debase the Ph.D. in the interest of reducing its time."
That back-and-forth ought to be familiar enough, but here’s the rub: Those quotations come from a book published in 1960, Graduate Education in the United States. We’ve been having the same arguments about time-to-degree for more than 60 years. While generations of us have fiddled, our graduate students remain in school for years on end.
The author of that 1960 study, Bernard Berelson, collected some of the first time-to-degree statistics, including the only numbers I’ve ever seen that measure the all-but-dissertation (A.B.D.) period. Back then, the time-to-degree rates that people were complaining about were, ahem, rather lower than what we see today. The median number of years spent "directly working on dissertation" were: 1.7 years in the physical sciences; 1.6 years in biological sciences; 1.1 in social sciences; 1.3 in humanities; 1.2 in engineering; and 0.9 in education.
The higher the Ph.D. unemployment rate, the more unconscionable it is to demand that all graduate students write the kind of dissertation best suited for research-driven academic jobs. Students need a chance to prepare themselves for the types of jobs that they are actually going to get, not just the one that graduate-school culture has deemed the ideal.
We ought to be able to step in and do something. After all, the doctoral dissertation is actually relatively new in graduate education. Introduced in the late 18th and early 19th centuries in Europe, it quickly became central—first over there and then here. The Johns Hopkins University, says the historian Roger L. Geiger in To Advance Knowledge, did more to "standardize" the Ph.D. in the United States than any other institution, not least because it awarded more of them than any other university during the formative decades of the 1870s and 1880s. Hopkins Ph.D.’s in turn became Professor Appleseeds, planting doctoral programs at public and private universities around the country. "By 1893," writes the historian Laurence R. Veysey in his 1965 book, The Emergence of the American University, "some amount of graduate work was required to win a permanent appointment at nearly every prominent institution. At the turn of the century the Ph.D. degree was usually mandatory."
But the requirements for that degree have changed with the circumstances of the academic job market. When there were jobs for everyone, shorter theses abounded. When academic employment got more scarce, so did the quick finishers. The requirements for a doctorate have always been under construction.
Literature departments have proved more willing than historians to consider alternative dissertation formats that might take less time to complete. (However, English departments have stopped short of endorsing Louis Menand’s radically functional suggestion that one peer-reviewed article substitute for the dissertation. Menand’s suggestion is an outlier that serves the salutary purpose of exposing some of the sketchy ethics underlying the current time-to-degree.)
One possibility, advanced by the Harvard literature professor David Damrosch in 1995 and revived recently by current reformers, is to break the traditional monograph into a series of essays. The recent Report of the MLA Task Force on Doctoral Study in Modern Language and Literature offers this and other specific suggestions along those lines: An "expanded repertoire" of dissertation possibilities could include not only a suite of essays but also "Web-based projects," translations (with apparatus), "public humanities projects," and dissertations based in pedagogy.
Such proposals expand what is possible for graduate-student work, but they also question the traditional idea of who a graduate student is. A more flexible view of the dissertation may expand our definition of "scholar" (and "scholarship")—but it might also lead to the position that some graduate students are not scholars at all.
We have to oppose a narrow definition of "scholar." To do otherwise would show a lack of imagination about academic work, to say nothing of a lack of concern for its future prospects. In a landmark 1990 essay, "Scholarship Reconsidered: Priorities of the Professoriate," Ernest L. Boyer distinguished what he called the scholarship of teaching, the scholarship of integration, and the scholarship of application from the traditional scholarship of "discovery."
Professors have operated for too long as though "discovery" were the only kind of scholarship worthy of the name. Even academics who teach heavy courseloads accept the prevailing value system that places the scholarship of discovery (and the publication it produces) at the top of the heap.
Under that value system, everyone should write the same kind of doctoral dissertation. Let’s question that assumption. We don’t need to scrap the dissertation, but we do need to look closely at its purpose. Our graduate students have been waiting—for too many years—for us to figure that out.