The emergence of these mass democracies produced new challenges, new problems of governance. A crucial development through the 20th century was the emergence of an idea of governance where it was understood that you could not have one or two very simple straightforward policies which would apply equally to all citizens. There was an increasing differentiation of sections of the population. It was understood that different sections of the population required different things, that different policies needed to be targeted at specific population groups. For instance, you could say at a very simple level that men and women required different kinds of benefits from the government; even among women, different age groups would have different requirements. All of this was elaborated in Western countries in terms of what was called the "welfare state". Even in that welfare state, it was understood that in order to be properly responsive to the needs of different groups of people, the state must be flexible in its policies; you could not have one simple blanket policy for everybody.

These techniques were then transferred and adapted to postcolonial countries, where for these policies to be effective - in health and education, in the basic needs sector, for instance - you needed similarly flexible policies. Even in terms of delivering food, it was understood that urban populations and rural populations would have to be treated differently. Among rural populations, children must be treated differently from adults, and so on. There are numerous ways in which specific policies began to be formulated for very specific groups of people.

So these are the new governmental technologies that I say were developed and made available. The idea behind this, which I think is a late 20th century development, was that no matter what the form of government, all governments needed to perform certain basic services and provide certain basic goods. At the global level, for instance, the UN is an international institution which provides certain services to anyone if the local government fails. If there is a famine in Ethiopia, one cannot simply say that the government in Ethiopia has failed, therefore nothing can be done. The idea is that certain basic services have to be provided to the people of Ethiopia and if the local government cannot do it, then other people must. And how does one provide these services? That is where I am suggesting that certain technologies have now been developed to ensure that particular services and resources are allocated to particular populations groups, no matter where they are.

What this means is that there are certain expectations which are now more or less universal; everybody, everywhere in the world, expects that a government ought to provide certain kinds of services at the very minimum. The interesting question now is how this technology of governance and the forms through which these services are provided will be combined with or related to political mobilization and ideology. This is where some interesting and perhaps quite basic changes have in fact emerged, starting with different forms of political mobilization. Specifically, I mean the ways in which population groups get classified and divided in the context of precisely these governmental services. Governmental classification could take numerous forms, for instance rural-urban, or could employ cultural or religious or ethnic categories, or some combination of all of the above. The point is that there is a whole range of ways in which populations might be classified for purposes of providing these services.

Very often, one would see that mobilizations would then occur around those governmental classifications. In other words, mobilization would take place precisely to make demands of the government on the basis of the category of which one was a part. The argument is roughly that the government has categorized people in this way and has guaranteed the provision of certain services to those belonging to that category. What you then get is a form of organization of political groups around classifications which are determined by the government. This can easily result in forms of mobilization which have nothing to do with the older ways in which people conceived of their identities. A lot of the identity politics that seems to have exploded in the last 20 years or so is probably conditioned by the ways in which people expect governments to provide them with certain kinds of basic services - welfare of different kinds, particular claims on things like education or employment, and so on.

The interesting question is how many of these categories - which have no moral foundation or any kind of ethical claim at all but are completely empirical descriptions of particular population groups - have actually managed to acquire a certain moral content through these political mobilizations. The claim is made that a group formed by a government classification is actually a community, there is a kind of solidarity, there is some moral identity of these people. This is the really interesting aspect of many of these political mobilizations: how do they manage to give themselves the form of a community when in fact there is no necessarily primordial or any other basis for this?

There are very interesting examples: in India, for instance, there is a category called 'BPL' which means "Below the Poverty Line." This is obviously an administrative demarcation. But there are policies of the government which say that if you belong to the BPL category, you are entitled to certain things and you actually carry a card which is called a BPL card. In many places, there are organizations or associations of BPL people. People below the poverty line could be from many different communities, many different caste groups, but suddenly the fact that all these people are classified as a group, as the target of a particular kind of policy, produces the ground on which these people can mobilize as a community. I think the interesting question would be: what is the moral character of this group? How is this local contextual community invented?

This is how the question of governmental technologies is connected with new forms of political mobilization, much of which is simply described as identity politics or sometimes as ethnic politics. Unlike the old anthropological understanding that most of this ethnic or identity politics has some kind of basis in primordial loyalties, a lot of it is probably simply a product of the way in which the new governmental technologies actually categorize people for purposes of administering policy.