26 SOCIAL SCIENTIST should contain the expansion of the state (because people will oppose the invasion of 'their9 space by the state), and will make for 'good government5 (that is, 'democratic9, meaning responsive, accountable and transparent government). It is expected, too, that in the context of such a strong civil society people will be broadly supportive of the market-led orientation of economic policy (which respects their rights to make 'choices'). Some statements of what constitutes 'good government" actually say that the pursuit of policies which are supportive of markets are as much a part of what it means as 'responsiveness', 'accountability' and 'transparency'. In these formulations of it, therefore, 'good government' actually transcends democracy. 'Decentralisation' of government should facilitate all of this, partly because it is at local levels that citizens' or community action is most effective. The whole set of ideas is pitched specifically against 'top-down' or dirigiste development, which is seen as having failed. These ideas are often painted as being 'progressive' and they are deceptively attractive. They are attractive because they imply respect for the ideas and the needs and the aspirations of the common people. Can one possibly be 'against' participation and empowerment? But these ideas are deceptive because they are used to veil the nature and the effects of power, and hold out the prospects of democracy (in civil society) without the inconveniences of contestational politics and of the conflicts of ideas and interests which are an essential part of democracy. It is most significant that the concept of 'civil society' as is used in this discourse, excludes 'political society'; and that the sorts of 'voluntary local associations' which are endorsed are not political organisations. That civil society exists in a field of power -or that there are differences of power within civil society - hardly seems to cross the minds of those who wish to see the space of civil society expanded, and that of the state (and of the market, maybe) reduced. The discourse is in fact quite deliberately a-political, in a way which is ultimately supportive of neo-liberal orthodoxy. Here I subject it to critique through an examination in particular of the idea of 'social capital' - because in the ways in which this notion has come to be used (as for example in the work on 'the missing link in development') it more or less subsumes all the other key ideas in the discourse. Social capital is, after all, held to be the key to 'health, wealth, wisdom and happiness' (according to the work of the great populariser of the idea, Professor Robert Putnam of Harvard University); and we have the authority of Mike Edwards, sometime