I started writing this post as a further contribution to the comment thread sparked by my last post, and in particular to the discussion that Janice Bially Mattern and I were having there. But my reply got too long for Haloscan to process, so I have moved it up a level and made it into a separate post. Plus, in doing so I am able to add this striking graphic, which is, I think, another example of the phenomenon we’re wrestling with. Full disclosure, I found this picture over at FiveThirtyEight. Fuller disclosure, as you might guess, the discussion that Janice and I are having is itself about round 538 or so in an ongoing conversation; this time we’re having the discussion in public and online, however, and I think that’s enough added value to continue the exercise.

Enough preliminaries. I think that our discussion about what to make of expressions like those captured on this video or the above graphic is moving between at least three different concerns or registers. They’re related, but I think it’s helpful to isolate each one so that we can get a clearer picture of the issues at stake.

1) my initial post was a cautionary note about inferring people’s true opinions or motives from their public performances. The fact that people are captured on film or video making racist/sexist/homophobic — or even just downright factually inaccurate — comments, I suggested, does not mean that they “really believe” the things that they are saying. Instead, a whole number of factors might incline a specific person to use such language: yes, they might “really believe” their statement/performance. but they might also be saying/doing something that they think is going to get them the approval of their peer group, or annoy their opponents, or “keep the conversational ball in the air” by recirculating some phrase or gesture that has previously been circulated all around them.

Given this potential micro-level diversity, I argued (and generally argue in situations like this) that we ought not to concern ourselves with the precise motives and beliefs of individuals, but instead ought to craft explanations that allow such inner compulsions towards action to vary, while we instead focus on the social context out of which and into which the action flows. So instead of looking at why a specific individual chose to use a specific piece of language or perform a specific action, we look at the cultural vocabulary available in the situation, and the social transactions with which that vocabulary is entwined — in this case, the vocabulary linking Obama to socialism/communism/terrorism, characterizing him as a Muslim, and impugning the manliness of his supporters, and the social networks of partisan news organizations and long-standing local communities within which signs and symbols can be quickly recirculated. This kind of analysis gives us the conditions of possibility for the performance(s) in question, and it has what I would consider the great advantage of confining itself to the sphere of the empirical instead of making inferences about unobservable motives and beliefs — which is not the same thing as saying that people don’t have motives and beliefs, but is merely a claim that we don’t need to know the precise details of those motives and beliefs in order to account for social action and public performances.

Janice points out, in effect, that this is an incomplete explanation, unless I wanted to conclude that people were compelled by their environment to act in specific ways. If I want to preserve contingency and agency, don’t I have to interpose some kind of process of individual deliberation between the environment and observed performances? Without this, my position looks like a kind of cultural determinism: the circulation of a particular bit of language on Fox News or conservative talk radio leads to its deployment in practice by McCain supporters at campaign events. And I would agree if I were trying to account for individual decisions. But I am not trying to account for particular people’s decisions to use or not to use a given linguistic formulation. My concern is not with why that particular gentleman stood outside of the hall in Toledo where Obama was delivering a speech, holding a grammatically confused but still clearly expressive sign; rather, my concern is with the social and cultural configuration(s) that make such performances possible in the first place.

On that score, I think that we need to look at the interrelation of news and propaganda, as well as the segmented networks through which information flows in present-day America — it is entirely possible to surround oneself with sources that are all slanted in a particular direction, and more or less completely ignore the other sources that are out there in the media landscape, and when this is combined with interpersonal social relations that recirculate those messages, I think we can start to see how performances like these become possible. (In fairness, this works in the opposite direction too: after a hard day at work reading liberal blogs and newspapers, one can come home, attend an Obama house party reception, then catch The Daily Show and The Colbert Report before bed…we may like our own cultural vocabulary-bubble better because it’s ours, but that doesn’t necessarily guarantee that it’s any less subject to the same dynamics as I am highlighting here.)

2) Janice replies with the language of “choice,” and suggests that a) language use in particular always involves choice, and b) an account without choice is insufficient, primarily since it lets people off the hook by not forcing them to take responsibility for their actions and not letting us hold them accountable for their actions. Let me take those in turn.

a) “choice” and “decision” language tends to bother me when applied to social action, and it bothers me both on empirical and on normative grounds. Empirically, “choice” bothers me because the language envisions both the presence of plausible alternative courses of action, and the presence of a more or less conscious process of deliberation between those alternatives, and I am simply not convinced that these two conditions always obtain. What alternative cultural vocabulary is empirically, practically available to ordinary people expressing their dissatisfaction with Obama? I would need to actually see available alternate language in order to buy this — and it’s not sufficient to hypothetically construct an alternative vocabulary that might exist, or to draw on the vocabulary valuable to us as detached observers. What I would need to see is empirical evidence that the speakers in question — the people in the video — actually had alternative modes of expression available to them, legitimated in their local social contexts. Then I would be prepared to concede that there were alternatives available.

But that’s only half the empirical battle, because then there’s this matter of “deliberation” to consider. I’m unsure how one ever knows whether deliberation has actually taken place, since we either have an unobservable internal process or a set of public expressions — the unobservable process can’t be observed, and the public expressions are subject to the same social dynamics I’m discussing here. Someone’s “private” journal is nonetheless “public” in that it uses collectively-constituted vocabularies and arguments and commonplaces — and that looks to me less like “deliberation” (leading to an inward compulsion to act) and more like legitimation (shaping action by negotiating the boundaries of acceptability, and in a sense drawing the action along with it). In the absence of empirical evidence of deliberation, I’m not sure how this presumption helps us.

That said, I can certainly see how it helps normatively, because if we presume deliberation than we can concretely claim that agency intervenes between structured environment and empirical outcome: Fox News didn’t make these people say what they said and do what they did, rather they deliberated and came to the conclusion that saying/doing these things was a good idea. But while this preserves a certain measure of structural indeterminacy, I am unclear that it preserves agency — or, better, I am unsure that it avoids what Talcott Parsons referred to as “the utilitarian’s dilemma.” Parsons pointed out that if we presume that individuals are making deliberate choices, and if we want to explain a social situation while retaining that presumption, we either have to explain their choices in more or less deterministic terms (heredity and environment, he argues) and thus eliminate their freedom to choose, or we have to allow people’s preferences to fluctuate more or less at random and thus give up any hope of explaining the situation but preserve their freedom to choose. Hence: efforts to preserve agency that rely on notions like “decision” strike me as rather paradoxical.

b) I am not convinced that it’s our job as social scientists to hold people accountable for their actions. Choice language — or, rather, an analytic involving individual choice — certainly permits that kind of normative evaluation, but for my part I don’t think that the purpose of analytical tools is to evaluate concrete social action. I think our analytics are about explaining and understanding situations; taking or not taking responsibility is a rather different endeavor than explaining/understanding. And there is nothing in my position that prevents us, or someone else, from holding the speakers and performers accountable for their actions; what my position does do is to deny that act of holding someone responsible the sanction of scientific or even scholarly grounding, and restore it to its (proper, in my view) normative status. Put differently, I don’t think that responsibility and accountability are empirical matters to be settled social-scientifically — I think they’re practical, or practical-moral, matters to be worked out in the course of social and political practice. So whether people are individually responsible for their public performances strikes me as the kind of question that demands a different kind of discussion than a discussion about the how and why of those actions. In the end, this is because there’s no way to social-scientifically determine whether someone is “responsible” for something, because responsibility is a value-orientation rather than a social-scientific conclusion.

[Note that it is, however, entirely possible to social-scientifically analyze how a group or organization comes to regard someone as responsible for something. That’s the kind of question one could, in principle, settle empirically. Whether the person in question was or was not “really” responsible, however? I can’t imagine anything but a social and political resolution to that question.]

3) This leads us to the third issue: whether it’s normatively preferable to use an analytic presuming and thus celebrating individual choice and decision, or to use an analytic that avoids “choice” language, deliberately doesn’t inquire too deeply into motivations, and contents itself with sketching empirically-plausible alternatives and conditions of possibility. I have already signaled my preference here, as has Janice, so there’s not much more to say — except that for my money a situation like the one depicted in the images I’ve linked to here presents an opportunity not for judgment, but for education. Not “education” in the sense of giving these people more information so that they will see that my/our point of view is superior, but “education” in the sense of helping, or forcing, people to confront the implications of what they’ve just said or performed. Holding a mirror up to students is, I find, a very helpful pedagogical technique — reflecting their claims back to them, perhaps slightly sharpened to bring out implications that they might not have considered beforehand. In effect, it’s an opportunity to provoke a crisis. And I would be constrained if I were to presume that performances were the result of conscious deliberation and more or less rational choice — constrained to provoke that crisis, because if the performance emanated from some internal disposition on the speaker then no crisis would result. But in practice, I see all the time — both in the classroom and outside of it — that people do not seem to have fully thought out what they are saying or doing, and that by holding up a mirror a crisis can be provoked. I do not pretend to know precisely why this is so, because that would require me to know a lot more about the internal dispositions of my interlocutors than I think that I can reliably know, but I have observed it happening a lot. The indeterminacy of the crisis, the moment when the ground falls out — that’s when agency happens, that’s when creativity and contingency actually come into play, that’s when I deliberately don’t want to “explain” what is going on because in so doing I’m taking agency away from my interlocutor. I would argue that my non-decisionist analytic supports that practice better than a choice analytic, which is ultimately the justification for using it.

Long enough for now.