March 30th, 2009
The capstone of earning a doctorate is the PhD defense, a final oral exam that is the final hurdle to achieve the degree.
What is a PhD defense really like?
Well, I had my own and have talked with many other PhD holders about theirs over the years, but this past week was my first time (times two!) being on the other side of a defense.
I imagine things may differ a bit from subject to subject, university to university, and country to country, but I think the essentials are similar, although keep in mind I am defaulting to astronomy in the United States specifically. There’s a public presentation of the disseration work followed by a private oral exam by a preselected committee. The committee should ideally be professionals in the field who are experts on some aspect of the PhD topic and who can provide guidance during the project and also evaluate it fairly at the end. Often there is an outside member who provides some quality control to make sure that a given department is being reasonably strict and fair.
I remember, years before my own defense, wondering how I could get through such a thing. I imagined how I could fall asleep the night before, and how nervous I would be.
The truth was that I was with a visiting girlfriend the night before and was very relaxed and confident. (In general, I recommend a long-distance relationship for finishing PhD students.) While some PhD defenses show the candidate to be nervous, quite often this is not the case.
After you spend years becoming an expert on a topic and knowing you know more about it than everyone in the room, except for perhaps your adviser, the confidence comes. In my experience, it isn’t that hard to tell when someone is ready. You can see the changes happen in students as they go from being tentative and mistaken in their scientific assessments to being sure and right most of the time.
And the truth is that it is extremely rare for anyone to fail. Failures usually don’t get to schedule a defense. It does happen once in a long while, and you hear stories. There are people who really don’t know what they’re doing after years, but most people don’t put themselves through the experience.
So, usually it’s a happy time. You plan parties. You buy gifts. It’s just enough of a procedure to make sure the PhD candidate feels like they’ve been through a trial and has had a chance to show off what they know.
I am happy that I had a memorable first question, even though I wasn’t at the time. I was cool and confident, but not so relaxed I could laugh at a joke (and perhaps the delivery wasn’t what it could have been). My dissertation was on quasar emission-line regions, but the first questions was this: “A PhD is a Doctorate of Philosophy, so could you comment on the relationship between Man and Superman and Immanuel Kant?” Very seriously I replied, “I haven’t read Man and Superman, so I can’t respond. Next question.”
Committee sizes vary, but are usually around 5-6, and everyone gets the chance to individually grill the candidate. This means that even when everything goes well and the candidate has done a great job, it can take an hour or an hour and a half. Sometimes it does take more, as many PhD holders like to hear themselves talk (guilty myself on occasion). Usually the worst case scenario is a long session spawned by the discovery of some fundamental misunderstandings, or some flaw in the work and the development of a plan to correct it. It is quite normal for there to be minor problems and for a dissertation to require revision, although it is very rare for a defense to be repeated.
If you haven’t had oral exams, it is a bit difficult to convey the feelings they sometimes invoke. It’s a bit like Socrates asking questions to lead a student to new understanding, at least if the student stumbles over the initial questions. Usually no one is a dick, but not always. Usually the questions are directly related to the PhD work, but not always.
After the questioning, the committee discusses the student’s performance and the written dissertation. Is the project significant enough? Has the field advanced in some way? Can the student go out into the world as a serious professional without embarrassing themselves or the department?
Anyway, I think I’ve talked around some specifics, but every defense is unique. Someone putting 5 years full-time into a project post college has figured out how to be a professional in a field, and the PhD defense is the opportunity to demonstrate that to everyone.