Agents —> Actors: Latour on Bourdieu Pt. 2

In my last post I talked about habitus, and about how Latour tries to refashion the concept in a less structural way: it's not a "structuring structure" any more, it's now a collection of "plug-ins" acquired (and lost or phased out) over time. While this is a Bourdieuvian concept Latour seems happy to keep, there are plenty of others he criticizes or simply jettisons: the idea of "fields," for instance, and of "capital," to name two of the most famous. There are of course large practical consequences that follow from these theoretical refusals: Latour takes the concept of habitus very far from its roots as a refinement of Althusser's "ideology" (this is my interpretation, not necessarily sanctioned by Bourdieu) and, thus, seemingly, from the political sphere. Because one of the reasons Bourdieu makes use of habitus so extensively is that he's committed to the idea of a critical sociology that can effect epistemological breaks, and that can thus produce the conditions for both personal autonomy and collective political action. And doing away with "capital," similarly, makes it quite quite a bit harder to talk about "capitalism" (though it should be noted that Bourdieu himself rarely used this word either).

So what about Latour and Bourdieu and politics? Latour's not silent on this matter; in fact much of the last third of Reassembling the Social is devoted to defending ANT against accusations of political complicity and quietism, and showing how sociology can still be politically efficacious while refusing a "critical edge." But to stick for now to his critique of Bourdieu, a passage from the little Socratic dialogue Latour includes as an "interlude" in RTS should serve to indicate what he thinks is wrong with Bourdieu's politics:

[Professor]: Now you have to tell me what is so politically great about transforming those you have studied into hapless, "actless" placeholders for hidden functions that you, and you only, can see and detect?
[Student]: Hmm, you have a way of turning things upside down. Now I am not so sure. If actors become aware of what is imposed on them, if they become more conscious, more reflexive, then is their consciousness not raised somewhat? They can now take their fate into their own hands. They become more enlightened, no? If so, I would say that now, and in part thanks to me, they are more active now, more complete actors.
P: Bravo, bravissimo! So an actor for you is some fully determined agent, plus a placeholder for a function, plus a bit of perturbation, plus some consciousness provided by enlightened social scientists? Horrible, simply horrible. And you want to apply ANT to these people! After you have reduced them from actors to placeholders, you want to add insult to injury and generously bring these poor blokes the reflexivity they had before and that you have taken away by treating them in a structuralist way! Magnificent! They were actors before you come in with your 'explanation.' Don't tell me that it's your study that might make them so. Great job, Student! Bourdieu could not have done better.
S: You might not like Bourdieu very much, but at least he was a real scientist, and even better, he was politically relevant. As far as I can tell your ANT is neither. (154-155)

The student has put his finger on two important, but separate, issues: first, whether Bourdieu is more of a "real scientist" than Latour; and second, whether he is more politically relevant. I think Latour has a very convincing argument about the first; I'm much less sure about the second. I'll try to do justice to both charges in what follows, though if I focus more on the former it's because I feel less conflicted about it myself.

In a way the critique of Bourdieu that Latour advances is a pretty old, familiar one, often made against structuralism (and Marxism) of various kinds. For Bourdieu and his student disciple in the passage above (Latour claims), the agent is just an effect of structure: the agent/object is just a hollowed out, empty place where something happens. Where Bourdieu differs from straight structuralism — and particularly from Althusser — is in his insistence that, through reflexivity, through careful and scientific description of structural determinations, one can widen and extend this empty place of action (or at least make its real dimensions known, and he says somewhere that it is "not that large"), and thus make genuine autonomy — political, moral, aesthetic, whatever — possible. In other words, we are not absolutely determined by the economy (which for Bourdieu, of course, includes the "symbolic" as well as the material economy), but we are much much more determined than we think: and our only hope of establishing the parameters of this determination, and breaking out of them, is reflexive social science. Thus, one might say there is a certain disinterest in what the object actually is in Bourdieu: he's much more interested in articulating baroque contextual frames than in pinning down exactly what's happening in the center. (See also the remark by Hélène Mialet, in her review of Science of Science, that Bourdieu repeatedly conflates the "empirical" with the "statistical": "Bourdieu, it seems, bases his arguments on his métier as a sociologist, which enables him to refine established concepts and to apply them to another 'field.' This experience is what seems to count for him as empirical," 617-618.)

For Latour, on the other hand, a proper scientific object is full, well-defined, and unpredictable, much more so than the structures that surround it: the object is not what is given by structures but what makes networks (which are called structures if we zoom out far enough) by virtue of its activities. He totally rejects the predictive and generalizing aspects of Bourdieu's sociology: the idea that we could extrapolate from data to make large a priori claims about the social world is anathema to him. (It's significant, I think, that Latour almost never uses statistics in his work, though he talks about them sometimes as forms of metrology.) Instead, he wants to be as specific as possible about what occurs in that tiny space that Bourdieu left for action; and, contra Bourdieu, he insists it can be filled in pretty exactly, without need for any margin of freedom or undecidability (I'm beginning to think that there's no real interest in moral autonomy in Latour's thought, despite his occasional rhetorical invocations of it).

We shouldn't think, then, that Latour refuses the bad news of Bourdieu's quasi-structuralism for a return to the old autonomous moral subject. But he also refuses to get all that depressed about it: an advantage of his anti-Kantianism is he doesn't have to share the eternal disillusionment of sociologists like Bourdieu (and his disciple Luc Boltanski) who have to run up against the obstacles that society places in the way of real moral freedom over and over and over. Action, for Latour, has never been a heroic overcoming of determination but is always a "slight surprise," which is why it has to be empirically, not theoretically, accounted for. This is why I don't think we should be too convinced by the moralizing tone of passages like the one I quoted above, where Latour seems to display a (quasi-ethnomethodological) ethical qualm about "critical sociology." It might bother Latour that sociologists like Bourdieu think they know better about the social than the people they study, but his objection to the method doesn't come down to only this: he also just thinks it's bad science! He "respects" his actors, certainly, but not for their moral autonomy: he respects them because one simply cannot predict what they'll do without following them; that is, he respects them for their empirical singularity. He respects them, that is, until he's managed to adequately describe them: after that, they're on their own.

It seems to me that many of the relevant differences here might be registered by the shift from Bourdieu's "agent" to Latour's "actor." An agent, as the legal sense of the word implies, is given the power to act by someone or something else: the important thing is it's been invested with power. An actor, on the other hand, plays a part for a while, interacts with other actors. The emphasis thus shifts, when we move from talking about "agents" to talking about "actors," from talking about whence the objects get their power to what they do with it. But we're also talking temporally: Latour's actors are temporary in a way Bourdieu's agents aren't.

Thus it seems to me Latour has very nearly reversed Bourdieu's theory of action: we've passed from a tiny space of indeterminacy and freedom in the center of an enormous determining social structure to a tiny point of empirical traceability in the center of a huge unformatted mess. For Latour — and he makes this fully clear only in Reassembling the Social, I think — the object or actor is not only affected by other actors it is directly engaging with but also impinged upon from the outside, from things that emerge from what he calls the "plasma," or the vast areas in between networks.

I'll stop there, for now… Next and last, I'll consider Bourdieu's critique of Latour in Science of Science and Reflexivity.