Wednesday, June 26, 2002

My answer to the question we are addressing today is simple: a particular idea of civil society has become a fad of development agencies and in the process it has been sought to be made universal. My argument is that these are steps that are fraught with difficulties that actually threaten the objectives that the development agencies have espoused. This is both because the idea of civil society that is being universalised often maps very poorly onto the social realities of the societies that are the object of development; and because it is a part of a package of related ideas that have the effect of depoliticising development. In large part the aim of my talk today will be to explain why I believe that ‘depoliticising development’ is damaging.

The idea of civil society that is being universalised by development agencies is that which is defined by the World Bank in the following way:

"Civil society consists of groups and organisations, both formal and informal, which act independently of the state and market to promote diverse interests in society. Social capital [the Bank goes on to say, directly], the informal relations and trust which people bring together to take action, is crucial to the success of any non-governmental organisation because it provides opportunities for participation and gives voice to those who may be locked out of more formal avenues to affect change"

These two sentences neatly bring together the package of ideas to which I referred: ‘civil society’ is bracketed together with ‘social capital’ – indeed Mike Edwards has told us that the terms are used interchangeably in BankSpeak - and both terms are linked in turn with ‘non-governmental organisation’, with ‘participation’ and with ‘empowerment’ (in the sense of ‘those locked out’ acquiring ‘voice’). The first sentence also reflects the particular idea of ‘civil society’ in the tradition of European political thought associated with Alexis de Tocqueville, and arising from his celebration of associational life in early C19 America. It is not my purpose in this short talk to enter into a lengthy discussion of the competing and contested definitions of ‘civil society’ in European thought. Suffice it to say that de Tocqueville’s is only one understanding of what civil society is, and that there are others which rather stress the notion of civil society as an arena in which the powerful seek to establish their ideological hegemony, while these hegemonizing intentions are contested by others. In other words whereas the Bank’s definition of civil society suggests solidarity over ‘the promotion of diverse interests in society’ - implying homogeneity – there are other understandings which recognise that civil society as an arena is cross-cut by power differences. It will perhaps be immediately clear that my objection to the ways in which the ideas of civil society and those that relate so closely to it – social capital, participation and empowerment – are used is that it is in such a way as to obscure the problems of power, and of inequalities of entitlements. This connecting set of ideas seems to promise the possibility of development, and democracy – an idea of people, as legally equal citizens, making decisions and managing their affairs for themselves - but without the contestatory politics that are a necessary part of any functioning democracy, and are inherent within development

The Bank’s definition probably reflects the influence of Robert Putnam’s ideas, for his notion of ‘civic engagement’, later referred to as ‘social capital’, is an idea of horizontal, soldaristic association between people that is independent of state and market; and of course in his famous work on Italy he seemed to have demonstrated that variations in such associational life (=’civil society’) very substantially determined variations in both economic development and the performance of government. His arguments certainly lent credence to the view that a robust civil society (in the sense of associations independent of state and market) is instrumental in creating a virtuous circle: freely associating individuals, through their combination, are able positively to influence governments (this is the popular idea of civil society as a counterbalance to the state), and thereby to give rise to sound policies and human development - and the security of those very civil rights and liberties of individuals that are essential for civil society to exist in the first place. Thus there has come about a ‘society-centric’ view of the development process: it is on the basis of a strong civil society that both democracy and economic development can best be built. It should be salutary, therefore, to recognise that Robert Putnam’s critics amongst his fellow Italian specialists, argue that the historical evidence is at least as good for the contrary view: that associational life celebrated by Putnam is a consequence not a cause of political mobilisation and organisation. And historians of America argue that Putnam has it wrong for that society as well: they argue that the most significant associational life in the US has followed from rather than having given rise to political organising and interventions by the state. ‘Bowling Alone’ is neither here nor there in relation to political forces; and it is of course striking that Putnam himself shows the strong correlation both temporally and spatially between the incidence of ‘social capital’ as he defines it for the USA, and wealth and income inequality in that country.

I have said that the concept of civil society remains contested, and there is no one definition that commands complete authority. But there is a good deal of agreement at least on the view that it refers to institutions of modern associational life, orginating in Western societies, that are based on such principles as those of equality, autonomy, freedom of entry and exit, contract, deliberative procedures of decision-making, and recognised rights and responsibilities of members. There is perhaps more controversy over the implications of such associational life, than over this statement of basic principles. .Anyway, it seems to me to be helpful analytically to hold to this conception of civil society when starting to analyse other societies, or we slip into relativism. Then I think that we may well conclude that the extent of civil society, say in India, or in most of Africa, is actually very restricted. In India certainly, and probably in much of Africa, too, civil society in this sense was sought to be built up by nationalist elites in the struggle for independence, and with some success. Ashutosh Varshney in his recent analysis of the causes for communal conflict in some Indian cities whilst others with comparable characteristics have remained by and large peaceful has shown that the difference has to do with variations in the extent and the intensity of inter-communal relationships. There is much less propensity to ethnic violence in those cities where there is rich inter-communal civic life – and this very largely goes back, he argues, to the period of the Gandhian political and social mobilisations of the 1920s and 1930s. Here, note, is a case against Putnam. To be sure, the relative lack of ethnic violence has to do with the existence of ‘bridging capital’ between religious communities, but the existence or not of such bridging capital depends upon prior political action. It has not sprung from everyday social relationships or the equivalent of bowling leagues. But my main point here is that the kind of organised civic life that was got moving by the nationalist elite remained very restricted in its extent. Indeed, after independence was achieved, those same elites that had previously sought to construct civil society against the colonial state, now that they held power themselves, no longer had the same interests at all, but rather the complete reverse. And the most important forms of social organisation that may be seen as lying ‘outside the state’ in post-colonial societies in South Asia and I suspect in Africa too, are rather of a ‘community’ type, based on caste or on other forms of ethnicity – and not therefore on those principles of the equality and autonomy (of citizens), and of freedom of entry and exit that are a fundamental aspect of civil society. In the Indian case I find persuasive Partha Chatterjee’s argument that the main sphere of mediation between the state and the population is what may be called ‘political society’, with its main instrumental form being ‘welfare’. In political society many mobilisations are actually based on violations of the rule of law in such actions as encroaching ing on government land, or refusing to pay electricity dues, or defaulting on the payment of taxes or repayments of loans. In these actions, however, groups of people are often demanding welfare as a matter of collective rights, and they are treated in turn by agencies of the state, or by NGOs even, not as citizens belonging to a lawfully constituted civil society, but as population groups deserving welfare.

But even if you accept my argument and agree that ‘civil society’ is not much developed in many parts of the developing world, you might still want to say that this doesn’t matter, for – surely, for the reasons having to do with the virtuous circle that I described earlier – the construction of civil society is a worthy objective for policy and can be encouraged by donors?

Those who believe that it can be should pause for a moment and reflect upon the failures of earlier attempts to impose political concepts and institutions emanating from the history of the West onto other parts of the world. Earlier efforts to build modern democratic states were not notably successful. Indeed an important part of the attraction, at the moment, of the idea of civil society is as a response to those earlier failures. ‘Civil society’ is supposed to be the answer to authoritarianism, unresponsiveness on the part of regimes and to poorly functioning bureaucracies. But what reason have we to believe that ‘we’, outsiders, will do better this time, and realise all these ambitious objectives?

With Rob Jenkins, one of the earlier speakers in this series, I don’t think the auguries are terribly good. Jenkins has pointed out the confusions that have beset the interventions of USAID, which has discovered the hard way that those organisations that it has supported as agents of democratic transition, because they were opposed to existing authoritarian regimes, have often turned out to be no less anti-democratic in their own practices. Where, as is often the case according to my argument today, organisations outside the state are based on notions of community rather than being fully voluntary associations of a truly civil society, donors tread on dangerous ground. It is also true, as Jenkins has shown so very well in his research on the politics of India’s economic reforms, that ‘strong associations in civil society’ are not necessarily supportive of that ‘sound economic management’ which is supposed to be the cornerstone and the objective of ‘good government’, in the prescriptions of the World Bank. They may well be forceful opponents of economic liberalisation. Indeed the reality may well be much closer to Olson’s prognoses in his work on the rise and decline of nations, when he argues that well organised groups have no incentive to work towards the common good of society and every incentive to engage in socially costly and inefficient, but privately profitable rent seeking.

Well-intentioned efforts to ‘build local organisation’ – never mind anything as ambitious as ‘civil society’ -can so easily go wrong, as Sanjay Kumar and Stuart Corbridge show in a forthcoming article in the JDS, about the East Indian Rainfed Farming Project. Their independent evaluations of this project show that it has had a fair degree of success as a farming systems programme. But the logic of the project is that sustainable livelihoods are to be constructed through collective action in local organisations. Consequently the project has devoted much effort to forming local development organisations. Unsurprisingly to anybody who knows anything at all about the extent of inequality and of social hierarchy in this broad region of India, and who is not gulled by the misleading stereotype of the ‘egalitarian tribal society’, these development organisations that have been built up with such effort by project staff are very substantially dominated by the more affluent and powerful members of the society, because they know that membership in the organisations is the key to accessing valuable project inputs. Maybe this would not matter much were it not for the fact that these same more powerful members of village societies have at the same time been withdrawing from collective arrangements, village grain-banks and reciprocal labour arrangements, that previously supplied some guarantees of livelihood to the poorest people.

And very often local NGOs, too, fall into line with the ‘anti-politics machine’ of development. They do not tackle the social relationships that give rise to the poverty with which they are concerned – the relationships, say, between adivasis and landlords in many of the remoter parts of India - but rather pose the problems with which they seek to deal in terms of ‘inputs’ , or more generally (as Sangeeta Kamat argues in her book Development Hegemony) in terms of ‘needs’ and ‘absences’ without reference to the class relationships that give rise to those ‘needs’ and ‘absences’. In this way principled NGO activists may contribute to the reproduction of the problems they intend to solve.

Of course there are positive cases. My colleague David Lewis cites the struggle in Mozambique to recognise collective land rights in the development of the 1997 Land Law, "and the subsequent civic movement in support of the rights which it enshrines, [that] has involved a pragmatic combination of mobilization and advocacy work by local NGOs, international NGOs and some donors". But such positive cases draw, I suspect, either on substantial political analysis, or on initial political mobilisation in the country or region concerned. [My case …]

[through the language and the practices of civil society, social capital and empowerment ..] is precisely that they stand in the way of such political analysis and potential support for political movements by promising, as I have put it, democracy without the inconveniences of democratic politics, and certainly without the dangers of supporting radical political action. It is a mistake to view civil society as some kind of an entity apart from the state - civil society is really unthinkable except in relation to the modern state. I question the causality that is explicit or implicit in arguments like Putnam’s, which suggest that any old ‘horizontal associations’ outside the state can generate social capital such as to transform it. Let me take the case of India again (just to flash out a new trick). I think there can be little doubt that, as Jean Dreze and Amartya Sen have observed in their commentary on the extent of ‘public action’ in that state, Kerala has the most developed civil society in India, probably followed by Tamil Nadu and Maharashtra. This is confirmed indeed in the most rigorous attempts to date to construct an index of social capital – sens Putnam – for the major Indian states, by Peter Mayer at the University of Adelaide. Mayer’s index shows Kerala way out in front, followed by Tamil Nadu, then Maharashtra, Karnataka, West Bengal and Gujarat. But what accounts for the sharp differences between Kerala, Tamil Nadu and these other states, on the one hand, and the northern states of the Hindi Heartland, at the bottom of Mayer’s scale, on the other.? I suggest that the difference cannot be understood outside the context of political action. The states with more social capital and, we may presume more in the way of civil society, are all states in which lower castes/classes were mobilised politically even before the end of the C19, in a way that has only just begun tio happen in Bihar or UP and still not at all in Orissa or Madhya Pradesh. The state with most public action, most social capital and a more developed civil society, by far, is Kerala. And on what are these achievements based if not on the history of popular mobilisation by the Communist Party?