Thoughts on the PhD defence


Last Friday there was a PhD defence in our department and Terry’s post about open defences in the USA got me thinking about the different cultures surrounding PhD defences. The first thing that came to mind is how different they can be, from country to country, university to university and even from department to department within universities.

A few axes in which defences can vary:

  • defence versus none
  • an open versus closed defence
  • external examiner(s) versus none
  • student presentation versus none
  • external examiner gives a presentation on  or not
  • official book printed prior to or after your defence
  • who makes the decision (a unbiased committee or one that has been involved throughout your PhD)

So, why so much variation?

Well clearly, some variation might come about from outside sources, such as the law. Much of the variation might simply arise from traditions of the university and culture (we’ve always done it this way…). But this got me thinking about the purpose of a PhD defence. In our teaching it is always better if we have defined goals and learning outcomes of the activities we do and a PhD at its core is fundamentally a learning process. Being a bit new to the other side of the equation, I don’t really have any idea about how much discussion is given to the purpose and expectations of the defence in departments. Are there clear objectives? What is the point? What does it all mean? These questions and many more may drive the form of the defence across universities. Clearly there could be a difference in the form of a defence where the main purpose is to evaluate the quality of the work versus where it is seen as a time point to gather your various projects into a cohesive story (and presumably the evaluation of the work has been done earlier, e.g. in the decision that you’re ready to defend). [Update: scroll down to the comments for a much more detailed examination of the purpose of a PhD defence by Paul Klawinski]

When I started writing this post I realised that I don’t have strong opinions about how a PhD defence ‘should’ be. It seems to me that there are lots of different and equally good ways of awarding PhDs. What constitutes ‘good’ will likely vary a lot based on how the entire program is formulated. But seeing different traditions now in Sweden has opened my eyes to some of the benefits of doing things differently. And thinking more seriously about PhD defences has gotten me thinking about the broader potential impacts of the event beyond being able to call yourself Dr. afterwards.

First maybe I should lay out my own experience on the table so that my biases are in the open. I have a degree from the USA and so my defence went something like this: I handed in my dissertation to the committee that I had throughout my PhD a few weeks before the day, I gave a seminar (50 mins) on my research to the department and answered questions, then I went into a room with my committee and talked with them. They sent me out of the room and talked about who knows what while I waited (the time went on forever…). Then they brought me back and congratulated me (hooray!). I think I might have been told that an open defence was illegal somewhere in the planning but honestly with juggling a baby, an international move and finishing up, that time is a bit hazy for me….

What I liked about the process that I went through is that it gave me a defined goal to work towards for ‘finishing’ writing. In my department you only print bound copies of your dissertation after the defence. That means there is still more to do and you need to incorporate changes that your committee suggests. But the seminar gave me a chance to communicate with my department and let them know what I had managed to do in my time there. So although it was a little stressful, I appreciated having a defence rather than not. I think I benefited from doing mine. It was the first full length seminar on my work, for example. And getting through your defence is definitely something to celebrate.

I’m not sure what it would have been like to have an open defence. The ones I’ve been to so far here in Sweden are much more focused on the details of the papers included in the dissertation. To be honest, I didn’t really feel like I was defending anything in my ‘defence’. In fact, my yearly committee meetings were always much harder and challenging than my defence and that wasn’t a bad thing. It made sure that my progress was going in the best possible direct rather than challenging details after it was too late to change them. So my committee and I talked very little about my dissertation but they focused more on big picture ideas. It was a really a great conversation that got me thinking about my place in science and how I could contribute. I think I’m still learning that but it was a wonderful broadening conversation. I was definitely asked some challenging questions in that closed-door portion of the defence, but I wasn’t actually defending my specific papers as I’ve seen more recently. Even in my former department, I think what constitutes the defence varies a lot between students but I appreciated the form mine took.

One thing I think I might have missed out on with an internal defence with my committee is that I didn’t get a chance to have an in depth conversation about my work with someone from the outside. Watching the defences here in Sweden, I am beginning to appreciate how valuable that can be. I know of a number of people who ended up doing a post-doc with their external reviewer. It seems like a great way to meet and interact with a leader in your field and also gives them a chance to get to know you. I also know of another example of a paper that came out of discussions during the defence. Generally the process seems like a great way to connect with someone and in our department the external examiner also gives a presentation about your work to put it in a broader context. In a way, this gets them to be an advocate of the student and really get to know their work. So even if future collaborations aren’t an outcome, you’ve had someone new think deeply and carefully about your work. However, if I had an external examiner for my own defence, I don’t think I would have had the same kind of interesting conversation as I did. It could have been just as good but likely pretty different.

So overall, I can see benefits to different PhD defence styles but unfortunately you can’t do everything…

What are the traditions at your department? Are there active discussions about what could be broader outcomes of the process of the PhD defence (besides a point where you can pass/fail a student)? And please share more extended outcomes of the PhD defence process! I’m sure I’ve only scratched the surface in this post.

8 thoughts on “Thoughts on the PhD defence

  1. Our system at Northampton is fairly typical of the UK system overall: a closed examination/defence, with usually an internal and an external examiner, plus the candidate, and an independent chair (the latter is becoming more common but is by no means universal). Supervisor can be there if both candidate and supervisor wish it, but that’s rare. Takes anywhere between one and > five hours, but 2.5 is probably average. Can be a very detailed, point-by-point grilling, or a broader overview.

    Having examined in Scandinavia and mainland Europe, as well as the UK, I think that there are advantages and disadvantages to all systems, but on balance the UK system is less certain in its outcome. Some places seem to treat the “defence” as more of a celebration of the candidate and their work, because it’s mostly been published anyway so would be hard to fail or even ask for many changes to the thesis!

    • Jeff, Thanks for your thoughts. I’m not sure a defence needs to be uncertain in its outcome to be effective, so I wouldn’t necessarily want to implement that kind of system. But I can certainly understand the skepticism of those systems that almost always mean a pass for the student. Even though I didn’t ever expect to ‘fail’ my defence it was still certainly a test! :)

  2. I read your and Terry’s posts with interest. I don’t think a thesis defense in the US is illegal the way most universities conduct them. I suspect a lot of schools are like yours in that there is a public presentation of the research with a question and answer section followed by a more private session for the formal evaluation. Of course, no good major professor would let a candidate present who was not assured of succeeding as there should be plenty of evaluation leading up to that point. So we have them and they do vary and you ask a good set of questions – What are the learning outcomes? What is to be gained? Here are some of the purposes that I think the defense serves.

    1. The defense is a rite of passage. There is tradition in academia in which everyone goes through the public airing of their work and a scientist has to be capable of standing up to broad scrutiny of their work.
    2. The defense is a final exam. In some cases the committee grills the student pretty hard and it does begin to seem like drawn out exam.
    3. The defense is a capstone event, celebrating your achievements.
    4. The defense is an opportunity to achieve a form of synthesis that might not have been evident or possible during the course of pursuing the PhD.
    5. The defense is practice for the research talk at future job interviews. In many cases, this is the first time you have actually had to own the stage for a full 50 minutes or hour and that is very different from a 12 minute talk with 3 minutes for questions at a meeting.
    6. The defense is an opportunity for hazing, an cruel initiation into the order of academia (I don’t agree with this but some defenses do feel this way).
    7. The defense is a different kind of test; a test of whether you can communicate effectively in that type of venue.

    There are probably plenty of other “purposes” but this is the beginning of a list. If the defense is a test to see if you are worthy of holding the credential, there are probably more efficient ways of doing this. The committee could look at the list of publications that came out of the dissertation research and if they are plentiful enough and in good enough journals, you pass. I was reading a recent dissertation and was surprised to see that the graduate program allowed the dissertation to have each chapter formatted differently. Each was formatted according to the instructions of the journal it had been published in or the prospective journal to which it would be submitted. I wish this had been the practice when I was a PhD candidate.

    Of the purposes above, I think I like the idea of 1, 3, 4 and 7 although PhD candidates might like 5. I think there is a place for tradition, but it would be nice if tradition had the dual purpose of achieving something beyond the rite of passage. I like the idea of celebration. I think we do not celebrate our science enough. But I also like the idea of using the defense as an incentive to broaden and expand on the research you have devoted 3-5 years of your life to. How well do you think about the next steps, applications to other fields, etc. is a good measure of whether you will make an effective researcher in the future. And we put out plenty of PhDs who cannot communicate to broader audiences. If only we could not pass candidates based on a failure to meet the communication criterion..

  3. I think the system in Finland is rather similar to Sweden. I defended my PhD thesis in the University of Turku in 2009 and I think the system is still the same. First two outside reviewers read my thesis manuscript (published articles, manuscripts and an introduction). After that I had to get a permission to print the thesis from the department board. At the actual defence I had one outside opponent who asked me questions for 2.5 hours (anything between 1-3 hours is normal). There is also always a (silent) chairman sitting in front of the lecture hall. At the defence I first gave a short introduction of my topic in Finnish because it is an open event and there were also my non-biologist relatives in the audience. The actual defence was in English because my opponent was from the Netherlands. He also gave a short summary of what he thought of my thesis was about before starting the actual questions. After the defence the opponent gave an official written statement to the department board and they in their next meeting decided officially if I had passed it. For me everything went fine and my opponent was a cool and nice person. ;-) In Finland the defence is also a very fancy event. You must be dressed in black (men usually in tail coats). And in the evening the party is a long dress event but a very relaxed one. :D Here are some photos of my defence and after-party:

  4. Hi Amy,

    Your post raised a question in my mind that I hope you and your readers will chime in on. What is the probability of failure at a defense? I ask this because I have the feeling, after reading your post, the various comments from your readers and feedback from my colleagues, that for all practical purposes a defense is merely a formality. As such, there is a negligible chance of failure.

    Best regards,
    Shrikant Kalegaonkar

    • I think this varies a lot depending on the institution (as Jeff says above, in the UK the outcome is uncertain). The places that I have been, it is very rare to fail a defence (although there was one I knew of while I was a PhD). But the negligible chance of failure doesn’t mean an instant pass for everyone. Rather, those students have not done sufficient work for a PhD won’t ever get to a defence that they would be doomed to fail. Instead they might leave the program (often their choice) or defend with a masters instead. But generally there is a discussion about whether you are sufficiently ready to defend. So in many places that could mean variation in time it takes to finish, rather than a successful outcome. In these kinds of programs, the problems seem to arise when students decide to go to a defence against the advice of their advisors/committee.
      I think Paul puts together a nice list of reasons to have defences that go beyond a mere formality. I think a lot can be learned from doing one, even if you know the chances of failure are small. It is important to keep in mind that the chances of failure are small precisely because the student takes it seriously and does a good job in preparing for the event. I am not sure that there is anywhere that would be pleased to grant a PhD to a student who blew off their defence and didn’t put an effort into their talk/questions/etc, even if they had a bunch of publications!

      • Thanks, Amy.

        I asked the question because it occurred to me that if the probability of failing a defense is negligible, then it’s really not a defense but a ritual as Paul Klawinski listed. It should instead be referred to by a more appropriate term: Final Presentation, PhD Review/Summary, TED Talk, etc. But it isn’t and I think for good reason.

        I believe that a person that earns a PhD should be knowledgeable in the philosophy of their discipline. Someone with a PhD in Science should be deeply knowledgeable of the scientific method and how science progresses (Karl Popper/Thomas Kuhn). I think they should be able to put their work in the context of the current paradigm; how it supports a given implication of the current theory and what its own implications might be. Or, if it is a new idea, to demonstrate its superiority to currently accepted ideas. They should be well versed in structured experimentation. They should demonstrate a thorough grasp of the tools of their specialty. After all, they are being recognized as original thinkers.

        So a defense should be exactly that, a defense of your idea and its context. You should be able to communicate your thoughts, not just have them. There should be a real likelihood of failure. Yes, you’ve spent X years working on a project; you’re an excellent technician. But, no, you don’t have the other aspects required of a PhD. I have not been impressed by my experience with a majority of PhDs in industry (and some in academics).

        I hope my thoughts aren’t too radical or offensive. I just feel too many PhDs are being handed out to highly skilled technicians.

        Best regards,
        Shrikant Kalegaonkar

        (PS. My own MS thesis defense mirrors your PhD defense. There was no real threat of failure; just corrections and sign-offs.)

Leave a Reply