of the poor. Agency here is simply foreclosed. It is assumed that perpetrators of violence who are poor are unable to think for themselves; they are said to simply (re-)act as automata to their social condition, and, as a result, their agency is denied, much as the agency of the Santal rebels has been denied by the Santals themselves as well as by their historians. What indeed is the difference between maintaining that ‘God made me do it’ and ‘Poverty made him do it’? None whatsoever as far as the denial of agency is concerned. Yet there is in fact a very important difference, in that the second statement is considered a valid account of politics in modern scientific discourse, while the former is not. Even if a survey were to be conducted showing that a majority of all adults maintained that God was the active agent in xenophobic or ethnic violence today, this would be interpreted as pathological, as an indication of a ‘moral panic’ akin to the belief in ‘supra-terrestrials’, not as a ‘fact’. Yet if the same proportion of respondents stressed the perceived threat to their rights to housing or jobs as the motivation for xenophobic violence – interpreted by scholars as resulting from poverty or unemployment – this is said to constitute a legitimate finding.
The reason for this difference is that social conditions in general and economic forces in particular constitute scientifically legitimate substitutes for agency and the political subjectivities of people today, whereas ‘supernatural’ ones do not; it should be clear that such an account is objectivist and, hence, amounts to a state mode of thinking. Moreover, the emphasis placed on poverty is an inference drawn in the work of scholarly commentators, not necessarily by the perpetrators themselves, who emphasise their citizenship rights (Neocosmos, 2011b). The epistemic reason at work here is clearly apparent. We should also note in passing that ‘supernatural’ factors are not of the same order as ‘religious’ ones; it is quite possible today for the latter to be legitimately included in the list of ‘causes’ in scientific studies of inter-ethnic violence, for example. The most important point is that accounting for violence in terms of poverty (or inequality or even ‘relative deprivation’) is a political discourse of the state today; it is the state that systematically refuses to acknowledge the existence of political subjectivities, reducing them to the socio-economic or to psychology, thus denying agency. On the other hand, to say, as the Santal rebel did, that his god Thakur will do the fighting was a subjectivity totally beyond (external to) colonial state comprehension at the time in that situation (although not necessarily outside precolonial state subjectivity, incidentally), and, as a result, colonial state discourse had to locate it elsewhere: outside ‘the political’, in the domain of the ‘superstitious’ and ‘irrational’.9
What this means is, paradoxically, that the Santal’s statement can be considered as political, in Badiou’s sense of the term, in the context of 19th-century India, as it was expressive of a collective subject and existed well beyond the (scientific) parameters of state thinking, exceeding thereby the subjective configuration of the colonial world and mere agency. On the other hand, the statement about the economic account of violence today is not political, operating as it does within the ambit of state (scientific) thought and thereby simply reproducing the extant and denying subjecthood and the transcending of the extant. An internal analysis of the former in terms of its specific categories and names could possibly have elucidated the singular character of its politics; yet once it had entered the archive, such elucidation became well-nigh impossible, for it was controlled and packaged within a category of ‘atavism’ and ‘irrationality’ by the colonial discourse of power. The conclusion must then be that, whatever the idiom or discourse being investigated, its political character must be established, and it can only be established ‘from within’, as Lazarus (1996, 2001b) maintains, through an analysis of its own statements and categories. It cannot be represented from outside through a sceptical commitment by sympathetic academics. Moreover, it is not the case that just because some statements seem to belong to a realm of ‘the political’, to the ‘public sphere’, to ‘civil society’ or whatever, that because their location labels them as ‘politics’ within modernist (neo-)liberal or Marxist conceptions, they are indeed idioms of agency, whereas others that seem to occur outside such a domain – such as religion – are not to be considered as politics.
The problem here is that of historicism, which, in addition to a notion of time, holds to an idea of totality (e.g. ‘society’, ‘nation’, ‘social formation’) within which political agency is confined to a specific domain of the political. A proliferation of the number of political domains does not, unfortunately, solve the problem of the social reduction of consciousness itself; it is rather the existence of sites which can ‘exist anywhere’, and which have to be ascertained through analysis, that locates politics (Lazarus, 1996). Politics is always singular and located in sites, but is simultaneously irreducible to a social referent. Unfortunately, Guha, Chakrabarty and Subaltern Studies as a whole, because their analyses are firmly situated within the discipline of history, have been unable to move to an irreducible analysis of the purely subjective. This is quite clear from their view of politics as socially located exclusively within two domains, that of the elite and that of the subaltern.
More recently, Chatterjee (2004, 2011) has pursued this argument by noting that two domains of politics exist in contemporary India (and, by extension, in other countries of the Global South) in which the relation of people to the state differs. One, which he refers to as ‘civil society’, is ultimately determined by a relation between people and state founded on ‘sovereignty’; the other is determined by relations between state and people founded on ‘governmentality’, in Foucault’s sense of state classification for welfare and security, which he saw as ‘a particular mentality, a particular manner of governing that is actualized in habits, perceptions and subjectivity’, i.e. as a particular mode of rule as well as a way of being in society (Read, 2009: 34; Foucault, 2000: 201–22). This domain Chatterjee terms ‘political society’. In each domain, he maintains, politics differs: in the former, it is founded on rights, citizenship and administrative technical procedures; in the latter, it is popular and informal and its ‘claims are irreducibly political’ (2004).10 While he rightly recognises that politics does not only exist within the narrow confines of the state but can exist in various realms which themselves originate from the colonial encounter, it is a structural determination, namely that between people and the state established by different modes of state rule, that Chatterjee takes to be the ultimate condition of political subjectivity and that is said to account for the difference between these forms of politics. It is different modes of state rule that determine not only different connections to power but also different subjectivities so that politics are reduced to (social) agency. Popular subjectivities are given no independent effectivity; they possess little choice in effecting these connections themselves and in exceeding the social. We are thus back to considering people simply as bearers of their objective location. A proliferation of state modes of rule, therefore, does not resolve the problem posed by the social determination of subjectivity.11 Politics does not have to be located within a state domain of ‘the political’ for it to be so qualified. This failure is one that leaves no room for a subjective politics beyond the social determinations of the state. Subaltern Studies ultimately misses out on understanding (emancipatory) politics, for it is caught in, and unable to extricate itself from, a statist view of what politics in fact is.
Nevertheless, Subaltern Studies is able to illustrate that there is a seemingly unavoidable limit to historical knowledge established by what Lalu (2009) calls ‘disciplinary reason’. History, as presently constituted, is indeed a state discipline by simple virtue of the fact, as Lazarus (1996) shows, that through a concept of time it objectifies the subjective, thus leaving no room for an understanding of subjective affirmations internally. ‘It is always in the interests of the powerful that history is mistaken for politics, that is that the objective is taken for the subjective’ (Badiou, 1982: 44, translation modified). The current misrecognition by the most progressive Third World historians of the nature of politics is only marginally distinct from the manner in which colonialism saw the actions of the subaltern rebel, as Guha himself makes clear. Indeed, the disciplines of the human sciences as a whole do not currently recognise politics other than as ‘the political’, and control scientifically the thinking of political subjectivity by psychologising it in a similar fashion to the (‘anthropologising’) practice of colonial discourse, for they combine a knowledge system