For the next hour and a half or so you will stand up front, a set of prepared powerpoint slides behind you. We will play a game. You will try to talk and we will try to prevent you from talking by asking questions, not waiting for the answer before we say what we think. We won’t know anything about your field and will make mistakes resolved 50 years ago, because we do not need to study biology. Perhaps the most powerful among us will say the most. In fact, really you are just standing in front for our entertainment while we lecture from the rear of the room in question form. If you don’t catch on that this is a contest, then you are a fool.
I asked a friend here at UCSB if he ever went to the KITP talks since they seemed to have so many interesting people come through. No, he said, he did not. The trouble was, he explained, that he was actually often interested in what the person up front might have to say and they were not allowed to actually say it at KITP. So he stopped coming to these exercises that danced between frustration and humiliation. What exactly is going on?
A KITP dinner
Seminars, talks, meetings, coffee hours, and other social events are a major part of how we learn new ideas, then incorporate them into our own thinking. There can be something transcendent about bouncing ideas higher and higher in a vibrant discussion. I suppose that is why we are spending three weeks here in Santa Barbara at the Kavli Institute for Theoretical Physics. We have coffee hours daily. There are two or even more talks every day. We share offices. There are many common areas for informal discussions. Some of us even live in shared housing where conversations can extend to dinner and beyond. Next week there will be a regular meeting with 37 different speakers.
The problem is the convention for interrupting has become extreme. I know that this varies among disciplines, that philosophers often carefully wait until the speaker is done then spend a lot of time with questions, for example. Biologists are typically intermediate, interrupting for clarifications more than for big discussions. KITP is even proud of what they do. You can look at the recordings and see that my initial characterization of the seminars is a bit of a caricature. Here is the link for general talks and for the session I am in. Is it a good idea? Does it make me feel like sharing anything with these people? No. If this is a general pattern for physics and math, it might even contribute to why they have fewer women. Of course causation could go the other way.
I only feel glad that I come from an institution of a very different kind.
Sociobiology 
Very nice talks by Nicolas Desprat (1/10), Gyan Bhanot (1/11), and Michael Doebeli (1/11) were also interrupted frequently. I thought it was interesting to see what people get caught up on… usually definitions and basic approaches. Things that some evolutionists see as trivial, for example using multiple approaches to examine the same problem, or general terms to refer to broad categories of phenomena while at the same time remaining somewhat vague, can be major stumbling points. Several thousand years ago, some “wise men” probably reacted very strongly to the abstract notion of the number “3.” Well, for somebody advocating the use of “3,” the first reaction from the precise mind is “3 what? 3 horses? 3 clouds? How can you possibly talk about 3 without saying what it is you are talking about?!”
The other funny thing is how much the term “anthropomorphic” has arisen when anyone speaks of “altruism” or “cooperation.” How much hubris to we have to say that anything altruistic is a self-imposed reflection of the human condition. We could equally say that the notions of “reproduction” or “variability” are anthropomorphic because humans reproduce and exhibit much variation. Altruism arose in evolution long before we did, and what distinguishes humans from other organisms is not their altruistic nature but their ability to deceive, remember, and punish (and write blogs in response to wrong-doings), which requires much higher cognitive ability. Now, ultimately humans may be capable of certain forms of altruism are unknown to other forms of life on earth, but there is nothing intrinsic to the definition of altruism (as normally defined) that leads to that conclusion.
But this quibbling has made up only a relatively small fraction of the talks. In some cases, the quibbling is only at the end, and this occurs when the talks have been more mathematical or data-focused with some implicit definition or assumption (e.g. Simone Pigolotti on 1/24, and the definition of “neutral” and “cooperation,” or Jeff Gore’s blackboard talk and the definition of “major evolutionary transitions”). A talk that attempts to define terms without really getting into details will of course provoke quibbling much faster than one using them implicitly. I think in evolutionary biology it is better to start with data and theory than definitions, because one has to know how the field works before the definitions begin to make sense. Ultimately the same is found in mathematics as well… once you know how math is used it begins to make sense why you would want abstract numbers to refer to sets of things.
Yeah, this is very much a disciplinary thing in my experience. Mathematicians always interrupt with questions about definitions and assumptions. I’ve always assumed it’s because precise definitions and assumptions are so crucial to mathematics. Whenever I’m invited to give a talk on my modeling work to a mathematical group, I always bring many fewer slides than I otherwise would and don’t put too much effort into prep, since there’d be no point. I’m not going to get through more than a handful of slides or get to do more than sketch the “story” my work tells.
Philosophers are indeed the opposite. As I think I’ve noted in comments here before, the philosophers at Calgary have what I think is a very nice practice. Not only do they never interrupt the speaker, but after the talk, there’s a 10 minute break. This is to allow time for the audience to think about what the speaker has said, chat with one another about the talk, and so come up with good, thoughtful questions.
I guess the question I’ve always had is why mathematicians bother to give talks. If the nature of the subject is such that it just can’t be conveyed with sufficient precision during a talk, why bother to try? In fields like biology and philosophy, talks serve as an efficient and enjoyable way to convey information and insight. In one talk, you can convey the essentials of several papers in less time than it would take someone to read one paper, and do so in a more entertaining way. They also serve as a starting point for productive discussion afterwards. I don’t really see how talks in mathematics serve either of these purposes, or indeed any other purpose, given the constant interruptions.