Abstract

Our aim is to provide an account of constitutive rules in terms of (1) the acceptance of regulative norms, and (2) a cognitive process we call “symbolization” (in an altogether different sense from what J. R. Searle means by this word). We claim, first, that institutional facts à la Searle boil down to facts concerning the collective acceptance of regulative norms in a given community. This, however, does not exhaust what institutional facts are. There is a residue, symbolization. Symbolization, as we understand it, involves a transfer of cognitive models from one domain to another. We introduce this notion by exploring different sorts of games, taking our cue from games of pretend play. In the context of this exploration, we introduce the idea of the significance—a matter of degree—of symbolization for a given institutional concept. In particular, we claim, symbolization may play, vis‐à‐vis a given institutional concept, a properly constitutive or a merely auxiliary role. We further argue that, in most legal concepts as conceived in our legal culture, symbolization plays a merely auxiliary role. A possible exception is the concept of political representation, at least on some understandings of it.