Levi has a post (http://larvalsubjects.wordpress.com/2010/11/23/a-brief-note-on-subjects-collectives-and-groups/) up that touches on groups as the very stuff of the political:
There are, in my view, a number of advantages to speaking about groups rather than subjects. First, of course, we can seek the conditions under which groups emerge out of collectives or come into being. Sartre has an exquisite analysis of all of the conditions that led to the storming of the Bastille and the invention of a new unity and the broader conditions that occasioned that invention. However, the concept of group rather than subject brings into relief other valuable questions. With the concept of group we get the question of how praxis is coordinated among diverse individuals. In other words, we get all sorts of practical questions pertaining to how individuals invent a group or themselves or mediate their differences. Likewise, we are afforded with the opportunity to analyze the microfacisms that emerge within groups, how groups ossify and become collectives, and how invention becomes self-defeating and alienating dogma. The question of political theory is not a question of how subjects are possible, but of how groups are possible. The problem with the concept of subject in political theory is that it illicitly unifies that which must be produced or unified, foreclosing these sorts of questions. As a consequence, we should eradicate this term from our vocabulary when doing political theory, instead asking how a group is possible. It is groups, not individual subjects, that are the subject of political theory.
I agree, and I think Levi and I have gone back and forth on this on the blogs some months ago. In fact, I should collect various writings on Sartre I have for a mini-volume arguing for forms of group praxis. On the one hand, of course, someone reading this might think it ridiculous that discussing groups is an advance. What about the polis? What about Mitsein? What all the forms of solidarity that have long been discussed in political theory? But groups as such seem to be a vanishing mediator between the state functions and the subject. On this point, Sartre (as well as Peter Hallward and Nina Power more recently) have much to say, especially about the formation of groups involved in revolutionary praxis. I think Rousseau is also an excellent resource here.
The time has certainly ended when all forms of groups or communities or solidarity as such are labeled as bad faith forms of communalism and forced adhesion to a center. The strange aspect of groups, as Rousseau recognized, is how they have a directedness of their own beyond its parts, not by some mystical incantation, as is often suggested. The example I use in courses is banal: think of a party (easy enough for students). One adheres to a certain dynamic, following an immanent code, while no individual is the party and no one has to rule over the group. Yes, there can be revolutionary terror (as Sartre notes), but a party (in the non-political sense) is not formed by bashing others over the head as Schmittians would suggest (and as poststructuralist critiques would worry over), but has its own dynamic that has its only end the party itself. (This is analogous to Rousseau: the end of the general will is perfectly circular, since it wants nothing other than that which a general will can enact itself.)