ABSTRACT
We have held that the order of things to which social institutions belong is built up through collecting together standardized action patterns on the grounds of their aim contents—more specifically, on the grounds of some relation, pragmatic or logical, between the aim contents. Institutions are thus units within that order; we can say at once that they are ‘functional’ units, that is, that they are made up of action patterns which operate in conjunction and whose aims require each other in a practical sense. Here, then, we collect together a series of action patterns on the grounds of their pragmatic relatedness; we also summarize this collection by giving it a name—the name we have for an institution. The institution would thus appear to be once more an action pattern, though magnified and more abstract than any elementary action pattern, embodying a class of aims and a series of ways of behaving rather than a single aim and behaviour cycle. It might also seem that the institution exists or is valid only for the observer, that is, the person carrying out this collection and summary. That this is not so is of course shown by the fact that the people we observe themselves have names for these ‘summaries’ of related action patterns—‘marriage’, ‘family’, ‘chieftainship,’ ‘property’, and the like. All such names, as we shall see, stand for normative concepts; the institution represents, for the actor, a rule or norm, and has that kind of reality, that is, the non-spatial and in a sense timeless validity of concepts. At least, this will always be true for some of the actors; others will at any given moment be concerned with translating the norm into concrete reality or, as I shall put it, with ‘activating’ the institution. Whenever we say that such-and-such institution exists we therefore indicate two things: first, that we can collect together and summarize (‘conceptualize’) certain regularly performed action patterns bearing on a class of aims; and second, that for a certain class of aims a mode of behaviour is laid down. The two viewpoints must converge in our formulation of the regularity or standardization of the behaviour we describe, in the sense that our summary must in some fashion match the conception of a norm which guides the actors.
