This has always struck as strange since it (a) reflects some distrust of not just departments but faculties (why can't an Econ prof do this job for a political science dissertation); and (b) indicates a poor understanding of oversight. Regarding the latter, I was in a room chock full of Plant Science people with a dissertation focusing on barley (hence the beer title), and I had no clue whether the person was doing good stuff or not, just that fungus on barley is bad and that resistance varies (is not futile!). So, if they had conspired to fool me about the quality of the work, I would have fallen for it. So, the oversight here is almost entirely symbolic. I found the process interesting but not exactly an efficient use of my time.
McGill has heaps of distrust/oversight issues that seem to provide more trouble than oversight. The Chair attends every dissertation defense in my department. Which means that the chair spends about 16-20 hours a year attending these things. Is that a worthwhile use of the Chair's time? Not so sure. I understand that delegation implies some risks that committees might conspire, that less qualified people might get through the process, but there is more than one way to engage in oversight. Instead of intensive police-patrol style, the university could rely upon fire alarm type oversight (yes, I sucked in this kool-aid while at UCSD despite never taking a class from the folks who pushed this view of stuff), where folks can report if they observe violations. That is, if violations are rare and not that dangerous, do we need a constant effort to monitor? Or can we just set up a system so that the alarm can ring when the rare violation takes place and is observed?
Despite my whining, I did enjoy the experience to see how similar/different the enterprise is in this area when compared with Political Science. Turns out their students pay as much attention to their conclusions as mine do, despite the fact that all students prepare for comprehensive exams by reading the intros and conclusions of other people's work. I was most interested in the big stylistic difference--these folks end their dissertations with their contributions to knowledge--a list of the major insights that are novel. Pretty funky, in a good way.
Anyhow, I am just glad there are PhD students out there working on ways to safeguard our supplies of