Social categories and theoretical descriptions
Two kinds of social category
Few social researchers, including sociolinguists, would dispute the claim that the social categories used in research cannot be taken as presented to us by the social world, and are therefore dependent on theories about it. Our argument will start by drawing a distinction between two kinds of social category: those constituted by involuntaristic characteristics, and those characterized by some degree of choice on the part of the people who belong to them.
We should note here that the distinction we are drawing is not the same as that often made in social psychology between ascribed and achieved- or chosen- identities. This theory derives from the structural functionalism of Parsons, which is concerned with patterns of norms into which people are socialized, through the occupation of particular social roles. In pre-industrial societies, according to this approach, social roles are largely ascribed, that is, fixed and largely unalterable, whereas in modern industrial societies, social roles are achieved. The distinction drawn here is less concerned with 'roles', but is, rather, about the way in which researchers' descriptions of the social world are theoretically informed. Our distinction is thus between those forms of social category which are based on objective features which people happen to have in common, and those which presume some common understanding on the part of those grouped within the category.
Following Greenwood's useful distinction (1994), we shall refer to the first kind of category as 'social aggregates'. These are groups such as the poor, the unemployed, women over 50- whose only common feature is the property identified as salient by whoever is employing the category- such as those responsible for benefits policy or for selling insurance to over-50s. Aggregates do not imply shared conventions and norms to which people can be party. Of course, it is undeniably the case that being poor or unemployed may induce some common habits, forms of recreation, states of health and so on, but this cannot be taken as a commitment to a shared culture of being poor or unemployed. And again, although the feature in question has an objective quality- lack of a job or lack of income or having lived for a certain number of years- the grouping of people having this feature in common is a consequence of a particular description, one that realizes some purpose in aggregating people in this particular way. It is important to note that 'aggregate' in this sense is not equivalent to 'average': it is used in the mathematical sense of 'a collection of elements having a common property that identifies the collection' (Chambers Dictionary).
The extent to which the poor or the unemployed or those over 50 wish to identify with such a definition of themselves, the extent to which belonging to such an aggregate is seen as expressive of what they are, cannot be determined from membership of the category itself. There are no shared rules and conventions about how to belong to the culture of the poor or the unemployed. Nevertheless, the objective features of people's situations that are the basis of their ascription to the category (low income or no job) may seriously constrain potential efforts to be described in terms of other categories (as 'upper class', say, or as 'wealthy').
The second approach to categorizing groups of people identified by Greenwood is with reference to their membership of 'social collectives'. These are descriptions of those groups in which members must be party to sets of conventions and norms in order to belong to the group in the first place. There are two implications that follow from the use of the term 'party to'.
Firstly, it signals that membership of such groups is indicated by an awareness of, and therefore some kind of commitment to, the conventions that constitute the group in the first place. A social institution such as marriage entails an understanding of what it is in order to undertake it (or to reject it). Marriage is, properly speaking, a conventional arrangement. It is used by Greenwood (1994) as an example of a collective category, whose membership involves understanding certain rules about it, performing a recognized ceremony, having a spouse and so on. Being unemployed or over 50, by contrast, are not conventional arrangements, although there may well be all manner of conventional ways of responding to the fact of being unemployed or over 50.
Secondly, the term 'social collective' avoids what Archer (1988) has called 'the myth of cultural integration'. We explore the significance of this later, but we may note here that itis possible to employ the social category 'married people' without reading back from it, as it were, that all those falling into this category share an identical commitment to the meaning of 'being married'. This conflationary impulse is particularly marked in the discussion of groups based on notions of race or ethnicity (Carter 2000) as we shall see.
The distinction between the social categories of aggregates and collectives that we are advocating is one grounded squarely in the notion of emergence. Social reality, we have noted, is made up of more than the sum of the individuals who comprise it. This 'more' consists of those features of social reality, such as legal and political institutions, property relations, belief systems and cultural mores, that are the emergent product of (arise out of and are irreducible to) social inter action. Some of these products of social interaction possess emergent properties of their own and so are able to exert systematic causal effects on subsequent interaction. Our distinction between social aggregates and social collectives seeks to reflect the emergent nature of the social world. Social aggregates, as Greenwood has noted, have no structural features (social or otherwise) that ensure their maintenance throughout changes in membership, and they are not constituted by joint commitments by members to arrangements, conventions and agreements. Social collectives, on the other hand, as emergent entities, possess the conceptual and practical resources for the formation and maintenance of identity over time.
