II. CONSTRUCTION AND ROLES
The notion of a social role is familiar enough that we can introduce it summarily. Roughly, a social role involves a patterned set of norms defining rights, responsibilities, and more generally expectations regarding behavior by the role-holder as well as by others toward her.6 We can distinguish in a role two aspects: a formal aspect, consisting in the norms that form the role’s script; and a material aspect, consisting in the actual patterned behavior that conforms to the script and realizes it. Relatedly, roles can be connected so as to form relatively stable and recognizable clusters in two ways, formally and materially. A formal connection among a cluster of roles exists when the scripts of the member roles systematically make reciprocal references to one another. A material connection among roles exists when the actual role-performances are systematically connected. There are accordingly two ways in which roles can be synthesized or unified into composites. A formally integrated cluster of roles forms a collective entity (or a collectivity for short); a materially integrated cluster of roles forms a self.
Starting from the latter, consider an example. I am a parent and a law professor. What makes both of these roles mine? A natural answer (pun intended) points to the spatiotemporal trajectory of a particular living organism. A further question, however, arises: which one? Well, mine, of course. The difficulty is that, notoriously, it is no easier to account for the mineness of the organism than for the mineness of my roles. In both cases, in order for the answer to lead to a self we need to understand the question as in the first place posed by one, that is as posed in the first person by someone whose awareness of some bodily movements as actions realizing certain meanings by comporting with some norms—hence as potentially amounting to the performance of certain roles—is part of what constitutes her as the self that she is. Thus, the “mine” in question in regard to the roles, no less than in regard to the body engaged in their performance, is not one of possession but one of identity: the two roles I mentioned, parent and professor, are mine in the sense that they are among the constitutive elements of my identity on a par with my body, rather than their being mine due to their relationship to a particular body.
Consider now the formal synthesis of roles to form collectivities. Notice first that many roles, such as those of painter and freelance writer, do not participate in forming any collectivity at all. They are defined exclusively in terms of some task to be accomplished or some objective to be realized. Call them functional roles. But other roles, such as those of parent and professor, are collective: in addition to tasks to be performed they include an affiliation, such as to a family in the one case and to a university in the other. What does my affiliation as a parent or professor consist in? The answer lies in the scripts of the respective roles. The script of my parental role makes systematic reference to my children and my spouse, just as the scripts of their roles reciprocally refer to mine. These coordinated formal connections form the D-C family. Similarly, the script of my role as law professor includes reciprocal references to the dean, colleagues, students, and staff, thus forming together the coordinated formal cluster known as Berkeley Law. What guides the coordination among a bunch of roles so as to constitute a collectivity? When are roles merely functional and when do they become collective? The answer points to some normative orientation—call it a mission—in light of which the interconnections among the roles are defined. The mission may but need not be goal-oriented; the various roles may instead be expressive of some values or other meanings. Nor need the mission be explicit. Though in many cases, such as a state’s constitution or a corporate charter, the mission is contained in a formal document, in other cases it is not.
The resulting picture amounts to a kind of role atomism, in which roles are the common building blocks for both individuals and collectivities. Or, shifting the metaphor to better suit a conception of human life as a skein of meanings akin to a text, we can see roles as providing the vocabulary of the self, a vocabulary that captures both the distributive and the joint employment of the we. So understood, role-atomism invites reflection on human self-creation in terms of modular units of signification and meaning by which all normative subjects, individual and collective, are constituted.
Before we proceed any further, two cardinal issues must be addressed. The first concerns the scope of the role-atomistic approach: how much of human life can be credibly associated with one role or another? In one sense, the answer is: as much as we want. Role is a theoretical term whose content depends on its place within the theory we construct. All human goings-on can be seen as falling within some role, and whether they should be so seen is a matter of the theory’s overall adequacy and fruitfulness. Still, we do come to the construction of a theory with some pretheoretical expectations, not only regarding the theory’s overall shape but also regarding the content of its main terms. At least rough compliance with such expectations is what gives a theory its plausibility, itself an important theoretical desideratum. Seen in this light, the claim that all of human life can be reduced to the performance of one role or another may seem implausible. Surely there are things one does, such as scratching one’s back, that are performed outside the perimeter of any social role. But reflection on this trivial example actually teaches the opposite lesson. Numerous roles include provisions regarding the propriety and the permissible methods, if any, of scratching. The role of an audience at a lecture, for example, is rather liberal about the action, whereas that of lecturer is more restrictive, permitting only a discreet scratch in extremis. Not even this degree of latitude is given, however, to a Buckingham Palace guard, nor for that matter to Her Majesty the Queen herself when in ceremonial circumstances; neither would be allowed to perform a back scratch no matter what. Methods of scratching are equally prescribed. A permissible discreet scratch at a dinner party must be administered manually—using an implement, such as a fork, is decidedly out of the question—whereas the opposite instruction applies to an Arapesh firsttime father, who is allowed to scratch himself only by using an implement, a stick in this case.7 But while many roles regulate scratching, not all scratching, it might appear, falls within any role, so that there would still remain cases of lawless scratching after all, viz., the scratching performed in the privacy of one’s room. However, even here social norms impinge, allowing the role-theorist to subsume the conduct in question under some role, familiar or ad hoc. Though using an implement such as a ruler would be in order here, one can use one’s cat only on pains of appearing bizarre. Appearing to whom? Well, to oneself, and counterfactually to everyone else. In this sense, and for better or worse, one is never alone.
The inflationary use of roles I propose amplifies, however, the tension I’ve already mentioned between self-constitution and social construction, thus giving rise to the second cardinal issue we need to address. Roles are social units. And although society, as already indicated, is a vague term ranging from the companionship between two people to humanity as a whole, it always pertains to the intersubjective, and so is supra-individual. Isn’t then subsuming all of human life within one role or another to stifle or nullify individual constructive freedom? In addressing this question, it should be made clear at the start that roles (at least as I understand them) consist in bundles of norms, to the exclusion of any coercion that often “backs up” various norms. Appending to norms coercive threats obviously complicates the picture and compromises freedom.8 But the normative guidance the roles themselves provide must be seen as appealing in some sense to the individual’s own will and as mediated by it. Accordingly, roles leave substantial room for constructive individualism of two kinds: both in regard to which roles one occupies and as to how one performs one’s roles. But though both freedoms can be in principle quite extensive, they are not unlimited. On the view I expound, individual self-constitution is indeed bounded by exigencies of intelligibility, which are intersubjective and thus supra-individual. Whether these exigencies are better seen as constraints on human freedom or as one of its enabling conditions is a much debated question; but either way, these exigencies must be acknowledged and specifically their influence over the two kinds of constructive freedom just mentioned must be recognized. I will illustrate each.
As to freedom regarding the roles one occupies, consider an impressionable young man who is inspired by reading Don Quixote to emulate the legendary figure. But though he can buy a scrawny horse, get the proper attire and a spear, and roam the land in search of adventure and injustice, he would not be a knight-errant for all that. Corresponding exigencies of intelligibility pertain to the degrees of freedom in enacting a role. There is obviously a large variation in latitude that different putative roles leave their bearers. But ascertaining this variation and assessing it requires that we (and this includes the role-bearer) appeal to intersubjective and so supra-individual factors. Recall the role of back-scratcher and compare it to, say, that of a physician. Intuitively, the former leaves its bearer much less elbow room, so to speak, than the latter. But pondering the example raises doubts: doesn’t the scratcher, who appears to be confined to a single role-related action, have in fact indefinitely many ways of performing it, each way consisting in a slightly different configuration of fingers, patterns of moving them, and so on? The example of back- scratching is of course artificial and contrived. But now think in this connection of a musician, say a violinist, whose behavior is tightly controlled by a highly prescriptive, detailed script, consisting in large part in the musical score. The discretion left to the performer by the score consists in the minutest variations in the placing of fingers and the pace of moving them within an infinitesimally narrow range, similar in kind to the range of options we have just lampooned in regard to scratching one’s back. And yet for the violinist the discretion turns out to be all-important, as it is entirely responsible for the gradation of musicians all the way from the virtuoso to the hack.
-  In particular, which clusters of norms are designated as a role (or, put differently, how dowe individuate a role) is part of a more specific role theory, one I don’t purport to provide.
-  Freedom regarding which roles one occupies is not limited to existing roles but must allowfor the opportunity to invent new ones. This raises a range of issues I don’t enter here.