Extension of a philosophical theory of identity

It was a nice surprise to see that philosopher of social science, Daniel Little, has joined the blogosphere. Motivated by (but not responding directly to) his posting “Who has social identity” I offer these thoughts. 


Philosophy has only recently begun to turn its attention to the notion of social identity, territory that sociologists, social anthropologists and historians have long since occupied. Herein lies the problem – social identity falls between two stools. Identity is a canonical philosophical concept with a distinguished tradition of theorizing in substance and personal identity. Other disciplines invoke the notion of identity (albeit social identity) but gloss over elementary logico-metaphysical distinctions that must surely cut across substance, personal and social identity. Typically, the notion of identity that is so often invoked in connection with the in-group/out-group distinction tends to conflate the notions of identical, identify and identity:

identical (1): x = y if all the characteristics of one are possessed by the other and vice versa

identical (2): x = y wherever the totality of what can be said to be true about one something is also the totality of what can be said to be true about the other

identify (1): to establish/recognise what a thing is, or that it is what it is

identify (2): the method of recognising/or re-establishing that x = x or x = y

identity (1): numerical identity

identity (2): synchronic/diachronic identity

identity (3): qualitative identity


So how then can the concept of identity be extended to include social identity? How does one deal with the problem of specifying the identity conditions for social objects? It is my, perhaps idiosyncratic view, that social objects are sufficiently substance-like in that there is a sufficient parallelism for theories of substance identity to illuminate social object identity. I do concede, that even presupposing the primacy of persons as substance, it is not clear how far the problems of substance identity translate across to problems of personal identity. Again, presupposing that social identity and personal identity are constitutive of each other it is not clear how far the problems of personal identity translate across to problems of social object identity. Can the canonical (Leibnizian) philosophical concept of identity be extended to include social identity? How does one deal with the problem of specifying the identity conditions for social objects? An attempt must surely be made.

Historically, Aristotle is the locus classicus of substance theorizing; Locke providing the preeminent “refutation” of Aristotelianism. Locke in turn is the locus classicus of personal identity but personal identity in the Lockean tradition seems little concerned with what socio-cultural identification amounts to. Post-war philosophy of social science has been preoccupied with Methodological Individualism, the doctrine that all statements about social phenomena – their structure and their change – are in principle reducible to statements about or referring only to individuals – their properties, their beliefs, desires, abilities, resources and actions, a debate that has been muddied by ideological considerations.

The phrase “social identity” can refer to the identity of a society – its continuity and persistence. “Social identity” can also refer to the social descriptions that apply to someone (including self-ascribed descriptions). The connection here is rather with personal rather than object identity.

Following Leibniz Wiggins terms social identity as “aggregate” substance (Wiggins 1980: 245); Quinton’s term is “social objects” (1975: 1); Ruben’s term “social entities” (Ruben 1985: 8-9) and Dupre’s “cultural species” (Dupre 1993: 76). Social substance or objects can usefully be usefully distinguished from other notions of the social (Ruben: 8-9):

* Social substances – England; the Catholic Church; the Royal Ballet; the Rolling Stones, the Jews.

* Social types – capitalism; bureaucracy, liberalism.

* Social events – the resignation of Tony Blair; the drowning of Antinuos

* Social processes – the decline of the Roman Empire, the Enlightenment.

* Social states – slavery; class antagonism; sexual division of labour.

To the above list I’d add the notion of social kind since it’s the dominant category in the social identity conceptual landscape. And much the way as there is a essentialist/anti-essentialist divide in substance (object) theorizing so too is there such a divide in the philosophy of social science. By social kinds I suggest that there are those who conceive of group identity as turning on a single highly specific or cluster of collecting (necessary and sufficient) features. For example, there is a strand of feminism that is markedly essentialist; there is a strand of national identity that emphasises the common origins and descent as a defining criterion; there is also the concept of race turning on physical and biological characteristics. “Essentialist talk” animates the politics of identity – race, gender, nationality, class, and so on. But there is an inherent tension in this tendency. One the one hand, there is an appeal to some essence – being black, being Jewish, being an American, being a woman – an essentialism that has a highly disreputable history. On the other hand, these identities are thoroughly constructionist – a fictive and cynical divisiveness generated in the name of “rights-talk” and appeal to (my liberal mind at least) irrelevant collecting features.

There must surely be conceptual space for assimilating the canonical identity conditions of substance to the open or family resemblance concepts intuitively required by social identity.