One of my graduate students e-mailed me with this question:
"Yesterday I got an email asking me to be a discussant at ISA. I've never done this before and I want to do a good job. Do you have any tips on being a good discussant?"
With the student's permission I am posting my reply here in the hopes that it may prove interesting or informative. [ISA = International Studies Association, the major professional association for IR scholars; their annual meeting is where most of us go to present work in progress and/or to hang out with colleagues from other universities and colleges/]
"Indeed I do.
First of all, I'm not sure that it's a good idea for graduate students to serve as discussant in the first place. The presenters-and-discussant(s) format lends itself to the posing of thorny questions by the discussant directed at the presenters, and this might lead to some role strain if the discussant is a graduate student and the presenters are established scholars. Far better, in my view, is an arrangement in which the presenters can run the gamut from graduate students to senior folks, and the discussant is at least a tenure-tracked professor someplace or has a comparable level of job security. I have no problem with a panel where all the presenters are the same rank, even if they are all graduate students, but I generally think that discussants should be a bit more established. So proceed at your own risk.
Second, a discussant in the traditional presenters-and-discussant(s) format has two distinct tasks: to discuss the papers, and to help to foment a discussion among the panelists and perhaps even members of the audience. Many people make a serious mistake and overemphasize the former task to the detriment of the latter. This is generally a mistake in the conference format because you cannot presume that the audience has read the papers in advance; if you could presume this, then commenting on the papers would be a good way of starting a discussion. But otherwise, comments on specific passages from the papers is likely to just confuse or bore the audience. In my view such feedback (which is in fact one of the tasks of a discussant) is better handled through e-mail or in some other more interpersonal and private setting.
The most boring discussants I've ever seen are those who proceed step-by-step through each of the papers on the panel, making suggestions that are generally of interest only to the author(s) of the individual papers, and then sit back as though they have completed their job. They have not. A panel is not, in my opinion, a kind of feedback session to which the audience has been invited as spectators; it's not a "fishbowl" situation in which the audience is simply observing. Rather, a panel is -- or can and should be -- something of a conversation, a discussion, a clash, a debate: at any rate, something more active and participatory.
It is the second task of the discussant to jump-start that conversation. There are better and worse ways to go about doing this. Often the worst way is to try as hard as possible to find some common thread running through all of the papers, and to display that for the audience regardless of how strained and awkward it is -- as though the point of a panel was for people to agree! I think this is largely silly. A panel is a public forum for disagreement, not agreement; it is contentious, not conciliatory. And it's a lot easier for the audience to participate in a debate than it is in a long train of agreement, because in a debate speakers can take positions -- even if those positions are sometimes "I agree with you about X but disagree with you about Y." The goal here is not to divide people into camps, but to give people an opportunity to articulate stances and to have those stances challenged -- and then give them an opportunity to defend them.
In my experience, a discussant manages this best by articulating opposing points of view -- opposed to the views presented by the panelists. Sometimes this involves setting the panelists against one another by drawing out their differences and disagreements; sometimes it involves explicitly mentioning the elephant in the middle of the room, the implicit Other against which everyone is arguing; sometimes it involves setting the papers in a broader disciplinary context so as to invite other parts of the discipline into the discussion; and yes, sometimes it just involves going to town against a paper and demolishing its absurdity. (This last possibility is also a large part of why I don't think that graduate students should be discussants, because they are most likely going to be more restrained in commenting on the papers of more senior scholars. There's also another caveat here, which is that I think it basically unconscionable to publicly flay a graduate student presenter; you can press them, you can raise objections to their argument, but you have to be somewhat more restrained because a graduate student is still learning the ropes. If the presenter is an established scholar, then they know better and in my opinion you can do things like bluntly point out that they're making no bloody sense -- and do with impunity, and with a clear conscience.)
Hence: I'd say that the job of a panel discussant is to serve as the living exponent of Weber's "uncomfortable facts," to explicitly introduce those lines of inquiry that raise problems and challenges for the paper presenters and their arguments. And then sit back and let the discussion proceed -- unless you're also the chair, you are not responsible for shaping the discussion, just for kick-starting it. Precisely what this involves is going to vary tactically from panel to panel: if the papers are all close together, play the role of the alternative point of view; if the papers are implicitly disagreeing, fan that into a full-blown clash; if there are gaps and silences, call attention to them.
Finally, remember that the panel is not about you. A discussant should not seize the opportunity of being a discussant to present a fully-formed alternative perspective or paper of her or his own. You can allude to it, reference components of it, but do not fall into the trap of actually presenting another quasi-paper of your own. This is both unfair to the participants (neither the other panelists, nor the members of the audience, will have had any opportunity to read your "paper" in advance -- hence they have to react to your arguments on the spot, which not everyone is particularly good at doing) and something of a betrayal of the format. Save your own presentation for another occasion.
Of course, all of this only applies in my opinion to a traditionally-organized panel in the traditional format. The funkier one gets in panel organization, the less these rules apply. I have seen more collaborative panel discussions, in which what transpires is less of a debate and more of a genuine exchange of ideas; I have seen and participated in autobiographical roundtables where people share their stories in preference to making and defending claims; and we all know about "moosehead" panels where the Usual (senior scholar) Suspects show up and give their established song-and-dance routines, and the megawattage of the star-power in the room effectively squelches anything like a real debate or discussion. (That last one is a particularly difficult place to play discussant, I think, unless one is already shining as an established disciplinary star; otherwise, the potential consequences are likely to prevent one from raising anything particularly challenging or difficult. But interestingly, one senior moosehead on a panel with other less luminous scholars is a great opportunity for a discussant; that's an environment where many such scholars are much more comfortable being pressed and challenged. And it's a lot of fun to be able to do that as a relatively junior scholar, believe me.) So I suppose that my last piece of advice is to figure out what kind of panel you are serving on, and adjust accordingly."
[cross-posted at the Duck]
