'The social' of Rose's title is described as the 'key zone' of government action (in at least 'Europe and North America' to which we may add Australia) from the middle nineteenth century to the changes of the present. Associated with this conceptualization of the zone of government are the practices that we call the 'welfare state'. The dominant political rationalities of our century, 'socialism, social democracy, social liberalism' agreed on at least this terrain of government; which, and this is crucial, require an expertise of 'the social'. The possible 'death of the social' is in part associated with the passing of the welfare state and the rise of what Rose calls 'advanced liberalism' or 'economic rationalism' (Rose 1996a 30. See also Rose1996b for more detailed discussion). In fact Rose agrees that death is a premature diagnosis for the social, his subject is rather the rise of 'community'. 'Community', Rose notes, has its origins in the language of opposition to what were perceived as the abuses and limitations of government of the social; significantly, its interference with personal autonomy. The language of community was a way of identifying with the governed, the policed. Around the same time authorities began using it in reference to groups of people who resisted being policed. (Rose 1996a 332). Criticism of the government of the social came from across the political spectrum, from 'libertarians of left and right', from 'progressives' concerned that individuals be 'active' in their own government (Rose 1996a 330). Critique of the social thus can be seen as coming out of the heart of liberalism, at least after the manner that liberalism is understood within the governmentality literature. For example Burchell (1993 271) describes both early and modern liberalism as seeking to identify the individual as at once the subject and the voluntary partner of government. This partnership takes different forms both in early liberalism, with its (supposedly) natural market, and in modern liberalism with its consciously artificial markets.*
This may suggest that there is some inherent tension between liberalism and 'the social'; in spite of the fact that it is possible to talk of a 'social liberalism', and despite the claim that 'the social' has been imagined from inside liberalism. Burchell (1993 p270) notes that at least one school of liberalism regarded National Socialism as a logical development of rationalities based on the social. Once one begins to talk of inherent tensions within liberalism it is only a short step towards talking of an internal dynamic within liberalism, tending "naturally" towards elimination of, in this case, the social as a category. The governmentality literature as a whole tends to avoid speculations of this sort. See for example O'Malley (1992) on the subject, arguing that 'technologies of power cannot be hierarchically ranked' from the viewpoint of efficiency and thus those conceptions of an evolution within liberalism towards greater efficiency of government must be rejected. It is as well to make this clear at this point because Rose's argument lends itself easily to such teleological speculations.
As opposed to the unitary domain of the social, community in Rose's sense is an association of individuals, a 'network of allegiances' (p334) on (apparently) any basis that can sustain allegiance - race, residency, sexuality, disability, disease. So we can have for example the HIV-positive community and the Romany community, as well as communities of moral affinity and activist commitment. It is no great leap to identify shared risk as a common factor of these very diverse groupings. Risk management in Simon's (1987) sense may not strictly require communities as a field of operation but the logic of the enterprising individual is common to them both, and they are both associated (whether as responses or simply as correlates) with the decline of the social. Rose (1996a) makes this association, and the closely allied association with situational crime prevention as discussed by O'Malley (1992). Rose does not refer to the strategy of developing Neighbourhood Watch groups, but this springs to mind as an illustration his theme.
A major difference between the social and community is that a community does not necessarily correspond to a literal territory or 'geopolitical space'. Rose (1996 330, 354n) links this trans-spatial feature of community with the modern (real or imagined) globalisation of economic relations and the associated decline of the importance of the social insofar as it is geopolitically defined. Rose notes however the appearance of an international conception of the social through such organisations as the United Nations. Finally we must note the special relationship of the community in the formation of the self, as Rose (p332) puts it 'the very condition of possibility for a community to be imagined is its actual or potential existence as a fulcrum of personal identity'. To govern through the community, then, is to govern through personal allegiances, perhaps to govern the intimate self.
The notion of government rationalities operating in a close way upon the personal identity is explored elsewhere in the governmentality literature. For example Dean's (1995 p567) work on the unemployed and the 'ascetic practices' through which the unemployed are sought to be governed; and Cruikshank's (1995) criticism of the self-esteem movement, summed up in her grimly funny warning not confuse the "empowered" with the powerful. (p338). There is not necessarily anything new about this as a technique of liberal government. Miller and Rose (1990), in the context of discussing liberal government of the economy, had commented on the relationship between liberal 'government at a distance' and the associated attempt to 'regulate the internal world' of (in this case) the worker (p23). However the nature of the self may be conceptualized in different ways. Rose (1996a) associates with the rise of community a more 'superficial' professional interest in human conduct and cognitive functions rather than in 'personal pathologies' that might have underlying social causes. Such a description of this shift in focus involves its own assumptions about the nature of the self.
Community in the present sense can be seen as sharing a fundamental logic with liberalism. Rather as modern liberalism treats the market, so in a similar way Rose's communities are at least in part artefacts, requiring the work of various 'educators' and the like to make us aware of our allegiance to them. (Rose 1996a p332). Expertises of community replace (or appear alongside) the expertises of the social. Furthermore allegiance to a community can be treated as a market-like choice, to the extent that one is imagined as able to vary the behaviours or qualities that bring one into the scope of one community or another (and also to the extent that communities may overlap and "compete"). Rose elsewhere notes (1996b p59) with some distaste the redefinition of a homeless person as a 'roughsleeper'; the former term implying social disadvantage, the latter lifestyle choice. Considered as such the 'roughsleeper' can be treated as belonging to a marginal community and thus brought within the scope of government. Rose (1996a) suggests that here perhaps a new territory of government, the management of the margins, is emerging; to the extent that this management may aim to reduce the perceived dangers of the marginalized it may itself be a form of 'situational crime control'. !
Rose's work raises some interesting questions. According to his account the language of community transformed within a few decades from a language of opposition to current practices of liberal government, to a language associated with the technologies of liberal government. Further insight into these technologies may come from more fine detail on the processes by which this change occurred.
Cruikshank, B 1993 'Revolutions within: self-government and self-esteem' Economy and Society 22(3) 227-244
Dean, W 1995 'Governing the unemployed self in an active society' Economy and Society 24(4): 559-583
Foucault, M 1991 'Governmentality' in Burchell,G Gordon,C and Miller,P (ed) The Foucault Effect: Studies in Governmentality. Hemel Hempstead, England, Harvester Wheatsheaf
O'Malley, P 1992 'Risk Power and Crime Prevention' Economy and Society 21(3):252-275
Rose, N 1996a 'The death of the social? Re-figuring the territory of government' Economy and Society 25(3) 327-356
Rose, N 1996b 'Governing "advanced' liberal democracies' in Barry, A Osbourne, T and Rose, N (ed) Foucault and Political Reason Chicago, Chicago U. P.
Simon, P 1987 'The emergence of a risk society: Insurance, law and the state' Socialist Review