* REPORT : 10

A paper by Arjun Appadurai on the NSDF/Mahila Milan/SPARC alliance in India.

Amsterdam, June 12, 2000

Arjun Appadurai
University of Chicago

Please send comments to: a-appadurai@uchicago.edu

Draft: Not for quotation or reproduction without author's permission

DEEP DEMOCRACY:
URBAN GOVERNMENTALITY AND THE HORIZON OF POLITICS

I. Prelude

What follows is a preliminary analysis of an urban activist movement with global links. The setting is the city of Mumbai, in the state of Maharashtra, in Western India. The movement consists of three partners and as an Alliance, its history goes back to 1987. The three partners have different histories. SPARC is an NGO formed by social work professionals in 1984 to work with problems of urban poverty in Mumbai. NSDF (The National Slum Dweller's Foundation) is a powerful grassroots organization established in 1974 and is a CBO (community-based organization) which also has its historical base in Mumbai. Finally, Mahila Milan is an organization of poor women, set up in 1986, with its base in Mumbai and a network throughout India, which is focused on women's issues in relation to urban poverty, and is especially concerned with local and self-organized savings schemes among the very poor. All three organizations, which refer to themselves collectively as the Alliance, are united in their concerns with gaining secure tenure in land, adequate and durable housing, and access to urban infrastructure, notably to electricity, transport, sanitation and allied services. The Alliance has also strong links to Mumbai's pavement-dwellers and to Mumbai's street children, whom it has also organized into an organization called Sadak Chaap (Street Imprint) which has its own social and political agenda. Of the six or seven non-state organizations working directly with the urban poor in Mumbai, the Alliance has by far the largest constituency, the highest visibility in the eyes of the state, and the most extensive networks in India and elsewhere in the world.

This paper is an effort to understand how this came to be so, by looking at the horizon of politics created by this Alliance and by seeing how it has articulated new relationships to urban governmentality. The paper is a preliminary ethnographic sketch. However, it is also part of a larger on-going study of how grassroots movements are finding new ways to combine local activism with horizontal, global networking.

II. Theoretical Points of Entry

Three theoretical perspectives underlie the presentation of the story of the Alliance in Mumbai.

First, I assume on the basis of my own previous work (Appadurai 1996; 2000; 2001) and that of several others from a variety of disciplinary perspectives, that globalization is producing new geographies of governmentality (Castells; Giddens; Held; Rosenau; Sassen). Specifically, we are witnessing new forms of globally organized power and expertise within the "skin" or casing of existing nation-states (Sassen). One specific expression of these new geographies can be seen in the relationship of "cities and citizenship" (Appadurai and Holston), in which wealthier "world-cities" increasingly operate like city-states in a networked global economy, increasingly independent of regional and national mediation and poorer cities (and the poorer populations among them) seek new ways to claim space and voice. Many large cities like Mumbai express the contradictions between these ideal types and contain high concentrations of wealth (tied to the growth of producer services) and massive concentrations of poverty and disenfranchisement. Movements among the urban poor, such as the one I document in this paper, mobilize and mediate these contradictions. They constitute the effort to reconstitute citizenship in cities. This effort takes the form of what I refer to as deep democracy.

Second, I assume that the nation-state system is undergoing a profound and transformative crisis. I wish to avoid here the sterile terms of the debate about whether or not the nation-state is ending (a debate to which I myself earlier contributed) but want to resolutely affirm that the changes in the system are deep, if as yet not graspable in a simple theory. I suggest that we see the current crisis as a crisis of redundancy rather than, for example, as a crisis of legitimation (as Habermas earlier put it). By using the term "redundancy" I mean to connect several different processes that others have identified with different states and regions, and in different dimensions of governance. Thus, we have an undoubted growth in many parts of the world in various forms of "privatization" of the state, sometimes produced by violent appropriation of the means of violence by non-state groups. In other cases, we can see the virtual take-over of national economies by multilateral agencies such as the World Bank and IMF, sometimes indexed by the voluntary outsourcing of state functions as part of the neo-liberal strategies that have become triumphant world-wide after 1989. In yet other cases, activist NGO's and citizen's movements have appropriated significant parts of the means of governance. And especially in Africa, we see forms of resistance to basic state powers through various forms of what has been called "fiscal incivility" (Roitman; Mbembe).

Aspects of this process blur the lines between privatization and marketization of state functions, but they point in any case to what I refer to as redundancy, that is, to a complex dispersion or distribution of state functions which creates alternatives or replicas of state governmentality in many contexts. This redundancy of state governmentality is matched by aparallel redundancy on the "nation" side of the equation, which has been well-documented in the form of separatisms, diasporic nationalisms, supra-nationalisms (such as the EU, Mercosur etc) and other forms of non-territorial nationalism. In this environment, states have to seek and find new partnerships, alliances and visible functions that can establish the uniqueness, or non-redundancy, of their functions. Urban activist movements exploit this need, as we shall see in the case of Mumbai.

Third, I assume that we are witnessing a notable transformation in the nature of global governance in the explosive growth of NGO's of all scales and varieties in the period since 1945, fuelled by the linked growth of the UN system, the Bretton-Woods institutional order, and especially the growing global circulation and legitimation of the discourses and politics of "human rights". Together, these developments have provided a powerful impetus to non-state democratic claims throughout the world. There is some reason to worry about the extent to which the framework of human rights is now serving as the legal and normative counterpart and facilitator of a neo-liberal, marketized political order. But there is no doubt that the global spread of the discourse of human rights has provided a huge boost to local democratic formations. In addition, the combination of this global efflorescence of non-governmental politics combined with the multiple technological revolutions of the last fifty years have provided much energy to what has been called "cross-border activism", through "transnational advocacy networks" (Keck and Sikkink). These networks provide new horizontal modes for articulating the deep democratic politics of the locality, thus creating hitherto unpredicted groupings of both "issue-based" (environment, child-labor, anti-AIDS) networks and of "identity-based" (women, indigenous peoples, gays, diasporic ) networks. The Mumbai-based movement discussed here is also a site for such cross-border activism.

Together, these three theoretical points of entry allow me to describe the Mumbai Alliance of urban activists as part of an emergent political horizon, global in its scope, whose political vision offers a post-Marxist and post-developmentalist model of how the global and the local can be mutually enabling. I also hope to show that an ethnography of this urban political horizon can offer us a solution to some of our more frustrating academic stereotypes about the global and the local.

III. The Setting: Mumbai in the 1990's

I have recently completed a lengthy examination of the transformation of Mumbai's cultural economy since the 1970's, with a particular emphasis on the brutal ethnic violence of December 1992- January 1993 (Appadurai 2001: forthcoming). That essay contains a relatively detailed analysis of the relationships between the politics of right-wing Hindu nationalism (seen mostly in the activities of India's major urban xenophobic party, the Shiva Sena), the political economy of de-industrialization, and the spectral politics of housing in Mumbai. Here I analyze the steady expansion of anti-Muslim politics of the Shiva Sena, the radical inequality in access to living space in the city, and the transformation of its industrial economy into a service economy. As a result, Mumbai became a perfect site for the violent re-writing of national geography as urban geography, through a paroxysmal effort to eliminate Muslims from its public sphere and its commercial world.

I will not retell that story here but I will review some major facts about Mumbai in the 1990's, which are not widely known. Mumbai is the largest city in a country (India) whose population has just crossed the 1 billion mark (one-sixth of the population of the world). The city's population is at least 12 million (more if we include the growing edges of the city and the population of the twin city which has been built across the Thane Creek). This means a population of 1.2% of one-sixth of the world's population. Not a minor case, even in itself.

By general consensus, here are some facts about housing in Mumbai. About 40% of its population (about 6 million persons) live in slums or other degraded forms of housing. Another 5-10% are pavement-dwellers.Ê Yet, according to one recent estimate, slum-dwellers occupy only 8% of the city's land, which totals about 43,000 hectares. The rest of the city's land is either industrial land, middle and high income housing, or vacant land in the control of the city, the state (regional and federal) or private owners. The bottom line: 5-6 million poor people living in substandard conditions in 8% of the land area of a city no bigger than Manhattan and its near boroughs. In addition, this huge population of the insecurely or poorly housed people has negligible access to essential services, such as running water, electricity, and ration-cards for essential foods.

Equally important, this population, which we may call citizens without a city, are a vital part of the workforce of the city. Some of them occupy the lowest end of white-collar organizations and others the lowest end of industrial and manufacturing industries. But many are engaged in temporary, menial, physically dangerous and socially degrading forms of work. This latter group, which may well constitute 1-2 million people in Mumbai, are best described, in the striking phrase of Sandeep Pendse, as Mumbai's "toilers" rather than as its proletariat, working class or labouring classes, all designations which suggest more stable forms of employment and organization. These toilers, the poorest of the poor in the city of Mumbai, work in such occupations (almost always on a daily or piece-work basis) as cart-pullers, rag-pickers, restaurant-scullions, sex-workers, car-cleaners, mechanic's assistants, petty-vendors, small time criminals, temporary workers in petty industrial jobs requiring dangerous physical work, such as ditch-digging, metal-hammering, truck-loading, and the like. They often sleep in (or on) their places of work, insofar as this work is not wholly transient. While men form the heart of this labor pool, women and children work wherever possible, frequently in ways that exploit their sexual vulnerability. To take just one example, Mumbai's gigantic restaurant and food service economy is almost completely dependent on a vast army of child labor.

Housing is at the very heart of the lives of this army of toilers. Their everyday life is dominated by ever-present forms of risk. Their temporary shacks may bedemolished. Their slumlords may push them out through force or extortion. The torrential monsoons may destroy their fragile shelters and their few personal possessions. Their lack of sanitary facilities will increase their needs for doctors to whom they have poor access. And their inability to document their claims to housing may snowball into a general "invisibility" in urban life, making it impossible for them to claim any rights to such things as rationed foods, municipal health and education facilities, police protection and voting rights.In a city where ration-cards, electricity-bills and rent receipts guarantee other rights to the benefits of citizenship, the inability to secure claims to proper housing and other political handicaps reinforce one another. Housing – and its lack – is the most public drama of disenfranchisement in Mumbai. Thus, the politics of housing can be argued to be the single most critical site of a politics of citizenship in this city. This is the context in which the activists I am working with are making their interventions, mobilizing the poor and generating new forms of politics. The next three sections of this paper are about various dimensions of this politics, of its vision, its vocabularies and its practices.

IV The Politics of Patience

In this section, I give a sketch of the evolving vision of the Alliance of SPARC, Mahila Milan and the National Slumdwellers Federation (NSDF) as they function in the complex politics of space and housing in Mumbai. Here a number of broad features of the Alliance are important to grasp.

First, given the diverse social origins of the three groups that are involved in the Alliance, their politics has negotiation and consensus-buiilding at its heart. SPARC is led by professionals with an Anglophone background, elite connections in Mumbai and beyond and strong ties to global funding sources and networking opportunities. However, SPARC was born in 1984 in the context of highly specific work between its founders (principally a group of women trained in Social Work at the Tata Institute for the Social Sciences) and poor women in the E Ward of Byculla district. This area has a diverse ethnic population and sits in between the wealthiest parts of South Mumbai and the increasingly difficult slum areas of Central and North Mumbai. Among its constituencies was a population of predominantly Muslim ex-workers in the sex trades of Central Mumbai who later became the backbone of another partner in the Alliance, Mahila Milan. The links between these two organizations was made in about 1986, when Mahila Milan was formed, partly through the support of SPARC. The link with the NSDF, an older and broader-based slum-dwellers organization, was also made in the late 1980's. The leadership of these three organizations cuts across the lines between Hindus, Muslims and Christians and is aggressively secularist in outlook. In a very rough way, SPARC brought technical knowledge and elite connections to state authorities and the private sector. NSDF, through its charismatic leader, A..Jockin (a grassroots activist with a slum background himself) brought a radical brand of grassroots political organization (in the form of the "federation" model, to be discussed later in this essay). Mahila Milan brought the strength of poor urban women who had learned, the hard way, how to deal with the police, municipal authorities, slumlords and real-estate developers in the streets of Central Mumbai, but did not previously have a real incentive to organize themselves politically.

These three partners still have distinct styles, strategies and functional characteristics. But they are committed to a partnership based on an explicit ideology of risk, trust, negotiation and learning among their key participants. They have also agreed upon a radical approach to the politicization of the urban poor which is fundamentally populist and anti-expert in strategy and flavor. The Alliance has evolved a style of pro-poor activism that consciously departs from earlier models of social work, welfarism and community organization (akin to those pioneered by Saul Alinsky in the United States). Instead of relying on the model of an outside organizer who teaches local communities how to hold the state to its normative obligations to the poor, the Alliance is committed to what I am calling "deep democracy" where there is a heavy emphasis on methods of organization, mobilization, teaching and learning which build on what the poor themselves know and understand. The first principle of this model of deep democracy is that "no one knows more about how to survive poverty than the poor themselves".

At the same time, this is not a politics of poverty as such. Drawing from the ideas of Paulo Freiere and others, there is a strong sense that the poor often fail because they do not know how to exploit their numbers, their knowledge and their potential for large-scale mobilization as citizens. Here is where the concept of "federating" (discussed below) becomes the key to finding the poor "for themselves" among the large ocean of poor "in themselves". Underlying the federation model is a model of teaching and learning, where the goal is for the poor to "own" as much as possible of the expertise that is necessary to claim, secure and consolidate basic rights in urban housing.

A crucial and controversial feature of this vision is its vision of politics without parties. The strategy of the Alliance is that it will not deliver the poor as a vote-bank to any political party of candidate. This is a tricky business in Mumbai where most grassroots organizations, notably unions, have a long history of direct affiliation with major political parties. Also, in Mumbai, the Shiva Sena, with its violent, street-level control of urban politics, does not easily tolerate neutrality. The Alliance deals with these difficulties by working with whoever is in power, at the level of the central and regional State, the Municipality of Mumbai or even its particular wards (municipal sub-units). Thus, the Alliance has earned a lot of hostility from other activist groups in Mumbai for its willingness to work with the Shiva Sena where this was necessary.Ê But it is resolute about making the Shiva Sena work for its ends and not vice versa. Because of its consistent image for being associated with no particular party, the Alliance has the double advantage of seeming non-political, while also having access to the potential political power of half of Mumbai's population.

Instead of finding safety in affiliation with any single ruling party or coalition in the State Government of Maharashtra or in the Municipal Corporation of Mumbai, the Alliance has developed a complex political affiliation with the various levels and forms of the state bureaucracy. This group includes its national civil servants (who are the bureaucrats who execute state policy at the highest levels in the state of Maharashtra and run the major bodies responsible for housing loans, slum rehabilitation, real estate regulation and the like). The members of the Alliance have also developed complex links to the quasi-autonomous arms of the federal government (such as the Railways, the Port Authority, the Bombay Electric Supply and Transport) and to municipal authorities who control critical aspects of infrastructure, such as regulations concerning illegal structures, water-supply, sanitation and licensing of residential structures. Finally, the Alliance works to maintain a cordial relationship with the Mumbai police and at least a hands-off relationship with the underworld, which is deeply involved in the housing market, slum landlording and extortion, as well as demolition and rebuilding of temporary structures.

From this perspective, the politics of the Alliance is a politics of accommodation, negotiation and long-term pressure rather than of confrontation or threats of political reprisal. This realpolitik makes considerable sense in a city like Mumbai where the supply of scarce urban infrastructure (housing and all its associated entitlements) is embroiled in an immensely complicated set of laws governing slum rehabilitation, housing finance, urban government, legislative precedents and administrative laws which are interpreted differently, enforced unevenly and almost always with an element of corruption in any actual delivery system.Ê

This pragmatic approach is grounded in a complex political vision about means, ends and styles which is not entirely utilitarian or functional. It is based on a series of ideas about the transformation of the conditions of poverty by the poor in the long run. In this sense, the idea of a political horizon implies an idea of patience and of cumulative victories and long-term asset building which is wired into every aspect of the activities of the Alliance. The Allaince believes that the mobilization of the knowledge of the poor into methods driven by the poor and for the poor is a slow and risk-laden process that informs the strong bias of the Alliance against "projects" and "projectization" that underlies almost all official ideas about urban change. Whether it is the World Bank, most Northern donors, the Indian state or other sources of funding, they are strongly biased in favor of the "project" model, in which short-term logics of investment, accounting, reporting and assessment are vital to the process of funding. The Alliance has steadfastly advocated the importance of slow learning and cumulative change against the temporal logics of "the project". Likewise, other strategies and tactics are also geared to long-term capacity-building, slow gaining of knowledge and trust, gradual sifting of more from less reliable partners and so on. This open and long-term temporal horizon is a difficult commitment to retain in the face of the urgency, and even the desperation, that characterize the needs of Mumbai's urban poor. But it is a crucial normative guarantee against the ever-present risk, in all forms of grassroots activism, that the needs of funders will gradually obliterate the needs of the poor themselves.

Patience as a long-term political strategy is especially hard to maintain in view of two major forces. One is the constant barrage of real threats to life and space that frequently land on the urban poor. The most recent such episode is a massive demolition of shacks near the railroad tracks, which produced an intense struggle for survival and political mobilization in virtually impossible circumstances in the period since April 2000, a crisis which is still in progress. In this sense, the strategies of the Alliance, which favor long-term asset building for the poor, run against the same "tyranny of emergency" (Binde: 2000) which characterizes the everyday lives of the urban poor. Thus it is a political vision which has to be maintained against considerable odds.

The other force that makes patience hard to maintain, is the built-in tension within the Alliance about different modes and methods of partnership. Not all members of the Alliance view the state, the market or the donor world in the same way. Thus, every new occasion for funding, every new demand for a Report, every new celebration of a possible partnership, every meeting with a Railway official or an urban bureaucrat can create new sources of debate and anxiety within the Alliance. In the words of one key leader in the Alliance, negotiating these differences, rooted in deep diversities in class, experience and personal style, is "like riding a tiger". It would be a mistake to view the pragmatic way in which all partnerships are approached by the Alliance as a simple politics of utility. It is a politics of patience, constructed against the tyranny of emergency.

To understand how this broad strategic vision is actually played out as a vision of urban governmentality, we need to look a little more closely at some critical practices, discursive and organizational, by which the Alliance has consolidated its standing as a pro-poor movement in Mumbai.

V. Words and Deeds

This section will attempt to give some sense of the embedded and embodied ways in which the Alliance actually deploys the politics of patience in Mumbai, in specific material and social contexts, with particular partners and opponents.Ê

As with all serious movements concerned with change of consciousness and self-mobilization, there is a self-conscious effort to inculcate protocols of speech, style, and organizational form. The Alliance cultivates a highly transparent, non-hierarchical, anti-bureaucratic and anti-technocratic organizational style. A small clerical staff consciously serves the needs of the activists (not vice versa), meetings and discussions are often held with everyone sitting on mats on the floor. Food and drink are shared during meetings, and <br>almost all official business (on the phone or face-to-face) is held in the midst of a tumult of other activities in crowded offices, and in the presence of many others. There is a constant undercurrent of bawdy humor in talking about problems, partners and themselves. Conversation is almost always in Hindi, Marathi or in English interspersed with one of these Indian languages. The leadership is at pains to make its ideas known to each other and to the members of the actual slum communities who are, in effect, the rank and file. Almost no internal request for information about the organization, its funding, its planning etc. is considered out of order. Naturally, there are private conversations, hidden tensions, and realdifferences of personality and strategy at all levels. But these are not recognized or legitimized in bureaucratic protocols or organizational charts.

This style of organization and management produces constant tensions among members of the Alliance and various outside bodies, donors, state institutions and regulators, which frequently demand more formal norms of organization, accounting, and reporting. To a very considerable extent the brunt of this stress is borne by SPARC, which has an office in Central Mumbai where the formal bureaucratic links to the world of law, accountancy, and reporting are largely centralized. This office serves partly to insulate the other two partners, NSDF and Mahila Milan, from the needs of externally mandated book-keeping, fund management, reporting and public legal procedures. The latter two organizations have their own headquarters in the compound of a municipal dispensary in Byculla. This office is in the heart of a slum-world where many of the core members of Mahila Milan actually live, an area in which Muslims are a major presence and the sex-trades, criminal world and petty commerce are highly visible. This office is always filled with the men and women from the communities of slum-dwellers that are the backbone of the Alliance. There is constant movement among key personnel between this office, the SPARC office in Khetwadi and the three or four major projects in outlying new suburbs where the Alliance is building transit facilities or new houses for its members (Dharavi, Mankhurd, and Ghatkopar).

The phones are in constant use, as key members of the Alliance exchange information about breaking crises, plans and news across these various locations in Mumbai and also across India and the world. Every few hours during an average day, a phone will ring at one of these offices and it might turn out to be one of the members of the Alliance checking on or tracking something down from Phnom Penh or Cape Town as likely as from Mankhurd or Byculla (see Section VI below). Because everyday organizational life is filled with meetings with contractors, lawyers, state officials, politicians and each other, spatial fixity is not valued and the organization functions in and through mobility. In this context, the telephoneand e-mail play an increasingly vital role. Almost all the key leaders of the Alliance, with a few significant exceptions, either use e-mail or have access to it through closecolleagues. The phones are constantly ringing. Schedules shift at the drop of a hat as travel plans are adjusted to meet emergent opportunities or to address the presence or absence of key members. The general impression is of a very fast ice-hockey game, with players tumbling in and out of the most active roles constantly in response to shifting needs and game-plans.

Nevertheless, through experiences and discussions evolved over fifteen years (and in some cases, more), there is a steady effort to remember and reproduce certain crucial principles and norms that offset organizational fluidity and the pressures of daily crisis. These norms and practices require a much more detailed discussion than I can give in the current context, but some impression of them is vital to understanding the political horizon of this form of deep democracy.

Possibly the central norm is embodied in a common usage among the members of the Alliance and its partners around the world. It is the idea of federation, used as a noun, and the words "federate", "federated" etc. as verbs. This innocuous term from elementary political science textbooks has a special meaning and magic for the Alliance which I am still in the process of exploring. At it is heart is the idea of individuals and families self-organizing as members of a political collective to pool resources, organize lobbying, provide mutual risk-management devices and confront opponents, when necessary. Members of the Alliance often judge the effectiveness of other NGO's, in India and elsewhere, by reference to whether or not they have learned the virtues of "federating". The National Slum Dwellers Federation is clearly their own actual model of this norm. The critical importance of this image of organization is that it does two things: it emphasizes the importance of political union among already pre-existing collectives (thus, federating, rather than simply "uniting", "joining", "lobbying" etc). As an image, it also mirrors the language of the Indian national state, which is referred to as the Indian Union, but is in fact a "federal" model with remarkable powers for its constituent states.

In the usage of the Alliance, the idea of the "Federation" is a constant reminder that groups (even families) with real existing powers have chosen to combine their political and material forces. The primacy of the principle of federation also serves to remind all members, particularly the trained professionals, that the power of the Alliance lies not in its donors, its technical expertise, or its administration but in the will to federate among poor families and communities. At another level, the reference to the federation is a reference to the primacy of the poor in driving their own politics, however much others may help them to do so. There is a formal property to membership in the federation and members of the Alliance have on-going debates about slum-families, neighborhoods and communities in Mumbai which are not yet part of "the federation". In effect, this means that they cannot be participants in the active politics of housing, resettlement, rehabilitation and the like which are the bread-and-butter of the Alliance.

Savings is another term which means more than what it says in the life of the Alliance. Creating informal savings groups among the poor (now canonized by the donor world as "micro-credit") is a major world-wide technique for improving financial citizenship for the urban and rural poor throughout the world, often building on older ideas of revolving credit and loan facilities managed informally and locally, outside the purview of the state and the banking sector. But in the life of the Alliance, "savings" has a profound ideological, even salvational status. The architect of the philosophy of daily savings is the most charismatic leader of the Alliance, A. Jockin, who might be one of the most important anchors of the Alliance in Mumbai, in India and beyond. He is the President of the National Slum-Dwellers Federation and is the missionary of a specific idea of daily savings among small-scale groups, which he sees as the bedrock of every other activity of the Federation. Indeed, it is not exaggeration to say that in Jockin's organizational exhortations wherever he goes, Federation=Savings. When Jockin and other members of the Alliance speak about daily savings, it becomes evident that they are describing something far deeper than a simple mechanism for meeting daily monetary needs and sharing resources among the poor. They are also speaking about a way of life organized around the importance of daily savings, which is viewed as a moral discipline (in Jockin' words, it is like "breathing") which builds a certain kind of political fortitude and commitment to the collective and creates persons who can manage their affairs in many other ways as well. It is something like a spiritual discipline whose spread Jockin and other leaders see as the building block of the local and global success of the federation model.

It is also important that Mahila Milan, the women's group that is the third partner in the Alliance, is almost entirely pre-occupied with organizing small savings circles. Thus in putting savings at the heart of the moral politics of the Alliance, its leaders place the work ofpoor women at the very foundation of what they do in every other area. In a simple formula: without poor women joining together, there can be no savings. Without savings, there can be no federating. Without federating, there is no way for the poor to drive changes themselves in the arrangements that disempower them. Thus "savings" is an ethical principle which forms the practical and moral core of the politics of patience, since it does not generate large resources quickly. It is also a moral discipline which produces persons who can raise the political force and material commitments most valued by the federation. In my on-going work with the Alliance, I hope to make a much closer cultural study of the discursive and organizational implications of this special ideology of savings as a moral and political practice.

The last key-term that recurs in the writing and speech of the leaders of the Alliance is the idea of "precedent-setting". I am still exploring the full ramifications of this linguistic strategy. What I have learned so far is that, beneath its bland, quasi-legal tone, is a more radical idea. The idea is that the poor need to claim, capture, refine and define certain ways of doing things in spaces they already control and then use these to show donors, city-officials and other activists that these "precedents" are good ones, and encourage other actors to invest further in them. This is a politics of "show and tell", but it is also a philosophy of "do first, talk later". The subversive feature of this principle is that it provides a linguistic device for negotiating between the legalities of urban government and the full force of the "illegal" arrangements that the poor almost always have to make, whether they concern illegal structures, illegal strategies, informal arrangements for water and electricity or anything else that they have succeeded in capturing out of the material resources of the city. When these captured resources are combined with new techniques for accessing food, health services, police protection or work opportunities, the concept of "precedent-setting" moves these practices into a zone of negotiation and quasi-legality (building on the idea of precedent in English common law). This linguistic device shifts the burden for municipal officials and other experts away from the strain of whitewashing illegal activities to building on legitimate precedents. The image and linguistic strategy of "precedent" turns the survival strategies and experiments of the poor into legitimate foundations for policy innovations by the state, by the city, by donors and by other activist organizations. It is a linguistic stratagy that moves the poor into the horizon of legality on their own terms.

But the world does not change through language alone. These key words (and many other linguistic strategies not discussed here) also provide the nervous system of a whole body of broader technical, institutional and representational practices which have become signatures of the politics of the Alliance. Here I briefly discuss three vital organizational strategies that capture the ways in which technical practices are harnessed to the Alliance's political horizon. They are: 1. Self-surveys and enumeration 2. Housing Exhibitions and 3. Toilet Festivals.

As we now know, censuses and various other forms of enumeration were applied to populations by modern states throughout the world after the 17th century, so that it has been observed (particularly by Foucault) that the modern state and the very idea of a countable population were co-productions, tied up with a specifically modern idea of governance, territory and citizenship. Censuses are perhaps the central technique which Foucault identified as lying at the heart of modern governmentality. Tied up by their nature with the state (note the etymological link to statistics), classification and surveillance, censuses remain at the heart of every modern state archive. They are highly politicized processes, whose results are usually available only in highly packaged form and whose procedures are always driven from above, even when many members of the population are enlisted in the actual gathering of data. Against this backdrop, and without any conscious theory of governmentality or opposition to it, the Alliance has adopted a conscious strategy of self-enumeration and self-surveying, by teaching its members a variety of ways of gathering reliable and complete data about households and families in their own communities. They have codified these techniques into a series of practical tips for their members and have thus created a revolutionary system that we may call governmentality from below, a crucial part of what I have called deep democracy.

Not only have they placed self-surveys at the heart of their own archives, the Alliance is deeply aware of the radical power that this kind of knowledge (and ability) gives them in their dealings with local and central State organizations and also with multilateral agencies and other regulatory bodies. This kind of knowledge is a central part of the political capability of the Alliance and is a critical lever for their dealings with formal authorities. The reasons are of particular relevance to places like Mumbai, where a host of local, state-level and federal entities exist with a mandate to rehabilitate or ameliorate slums. But none of them know exactly who the slum-dwellers are, where they live or how they are to be identified. This is a fact of central relevance to the entire politics of knowledge in which the Alliance is perennially engaged. All state-sponsored slum policies have an abstract slum population as their target and no knowledge of its concrete, human components. Since these populations are by definition social, legally and spatially marginal, invisible citizens as it were, they are by definition uncounted and uncountable except in the most general terms. By rendering them statistically visible, the Alliance controls a central piece of any actual policy process, which is the knowledge of exactly who lives where, how they make their livelihood, how long they have lived there and so forth. This information also has vital significance since some of the most crucial pieces of recent legislation affecting security of tenure for slum-dwellers in Mumbai are tied to the date on which they can demonstrate their occupancy of a piece of land or a structure. This information is vital to any official effort to relocate and rehabilitate slum populations.

At the same time, the creation and use of self-surveys is a powerful tool for internal democratic practice, since the major mode of evidence used by the Alliance for claims to actual space needs by slum-dwellers is the testimony of neighbors, rather than other forms of documentation such as rent-receipts, ration-cards, electric meters and other civic insignia of occupancy that can be used by the more securely housed classes in the city. The very absence of these amenities opens the door to radical techniques of mutual identification in the matter of location and legitimacy for slum-dwellers. For, as Alliance leaders are the first to admit, the poor are not exempt from greed, conflict and jealousy and there are always slum-families who are prepared to lie or cheat to advance themselves in the context of crisis or new opportunities. Such problems are resolved by informal mechanisms in which the testimony of neighbors is utterly decisive since the social life of slums is in fact characterized by almost complete lack of privacy. Here social visibility to each other (and invisibility in the eyes of the state) become mutual strengths in deepening mechanisms of self-monitoring, self-enumerating, and self-regulation in the crucial link between family, land and dwelling which is the central negotiable material good in slum life. The Alliance continues to place central emphasis on self-surveys and self-enumeration as part of the on-going political practice of its member families.

Housing exhibitions are the second organized technique through which the structural bias of existing knowledge processes is challenged, even reversed, in the politics of the Alliance. Since the materialities of housing – its cost, its durability, its legality and its design – lie at the very heart of slum life, it is no surprise that this is an area where grassroots creativity has had radical effects. As in other matters, the general philosophy of state agencies, donors and even NGO's concerned with slums has been to assume that the design, construction and financing of houses has to be produced by various forms of expert and professional knowledge ranging from that of engineers and architects, to that of contractors and surveyors. The Alliance has challenged this assumption by a steady effort to appropriate, in a cumulative manner, all the knowledge required to construct new housing for its members. This has involved some extraordinary negotiations in Mumbai, involving private developers and contractors, the formation of legal cooperatives by the poor, innovations in urban law pushed by the Alliance, new types of arrangement between banks, donors and the poor themselves in the realm of housing finance, and direct negotiations over housing materials, costs and building schedules. In effect, in Mumbai, the Alliance has moved into housing development, and the fruits of this remarkable move are to be seen in three or four major sites, in Mankhurd, Dharavi, Ghatkopar. The last of these, the Rajiv- Indira Housing Cooperative, in Dharavi, is a major bulding exercise, which is a decisive demonstration of the ability of the Alliance to put the actual families who will occupy these dwellings at the center of the process where credit, design, budgeting, construction and legality come together.. It is difficult to exaggerate the complexity of such negotiations, which are difficult even for wealthy developers because of the maze of laws, agencies and political interests (including those of the criminal underworld) in any housing enterprise in Mumbai.

Housing exhibitions are a crucial part of this reversal of the standard flows of expertise when it comes to housing for the rehabilitation of slum-dwellers. The idea of housing exhibitions by and for the poor goes back to 1986 in Mumbai and has since been replicated in many cities in India and elsewhere in the world. This idea, which involves large, crowded, open exhibits of housing models built by the poor, is a democratic appropriation of a statist and middle class consumer model which became very popular in India in the 1980's. In this period, there were "exhibitions" (or expos) not only for industrial and high-tech products but also for books, clothing and household goods targeted to the urban middle-classes. These were major venues for demonstrating new kinds of consumer goods (from detergents and washing machines to cookware and cleaning materials). They were occasions for socializing the urban middle-classes into the products and lifestyles of contemporary urban life and for manufacturers to advertise and compete with one another. They were immensely popular, temporary, retail extravaganzas that combined the pleasures of the fair, the tumult of the market and the excitement of circuses and other temporary public events. They were Braudelian scenes of market sociality. They have since been eclipsed by other retail modes (such as door-to-door selling, shopping malls and the like) but remain a vigorous forum for specific products, such as books, clothing and the like, for which urban retail space is inadequate or too costly.

The housing exhibitions organized by the poor since the mid-1980's, through the Alliance and other like-minded groups, are another successful example of the hijacking of an upper-class form for the purposes of the poor. Not only did these exhibitions allow the poor (and especially the women among them) to discuss and debate designs for housing thatsuited their own needs, it also allowed them to enter into conversations with various professionals about housing materials, construction costs, and urban services. Through this process their own ideas of the good life, of adequate space and of realistic costs, were fore-fronted and they began to see that house-building in a professional manner was only a logical extension of their greatest expertise, which was to build adequate housing out of the flimsiest of materials and in the most insecure of circumstances. These poor families were enabled to see that they had always been architects and engineers and could continue to play that role in the building of more secure housing. In this process, many technical and design innovations were made, and continue to be made. More significant, these events were political events where poor families and activists from one city traveled to housing exhibitions in another city, socializing with each other, sharing ideas and simply having fun. They were also events to which state officials were invited, to cut the ceremonial ribbon and to give speeches associating themselves with these grassroots exercises, thus simultaneously gaining points for hobnobbing with "the people" and giving poor families in the locality some legitimacy in the eyes of their neighbors, their civic authorities and themselves.

As with other key practices of the Alliance, housing exhibitions are deep exercises in subverting the existing class cultures of India. By performing their competences in public, by drawing an audience of their peers and of the State, NGO's and sometimes foreign funders, the poor families involved entered a space of public sociality, official recognition and technical legitimation. And they did so with their own creativity as the main exhibit. Thus technical and cultural capital are co-created in these events, creating new levers for further guerrilla exercises in capturing civic space and pieces of the public sphere hitherto denied to them. This is a particular politics of visibility which inverts the harm of the default condition of civic invisibility which characterizes the urban poor.

Running through all this activities is a spirit of transgression and bawdiness, expressed in body-language, speech styles and public address. The men and women of the Alliance are involved in constant banter with each other and even with the official world (though with some care for context). Nowhere does this carnivalesque spirit come out more clearly that in the Toilet Festivals (sandas mela) organized by the Alliance which enact what we may call the politics of shit. Human waste management, as it is euphemistically described in policy circles, is perhaps the key arena where every problem of the urban poor arrives at a single point of extrusion, so to speak. Given the abysmal housing, often with almost no privacy, that most urban slum-dwellers enjoy, shitting in public is a serious humiliation for adults. Children are indifferent up to a certain age, but no adult, male or female, enjoys shitting in broad daylight in public view. In rural India, women go the fields to defecate while it is still dark and men may go later but with some measure of protection from the public eye (with the exception of the gaze of railway passengers inured to the sight of squatting bodies in the fields, and vice versa). Likewise, in rural India, the politics of shitting is spatially managed through a completely different economy of space, water, visibility and custom.

In cities, the problem is ten times as serious. Shitting in the absence of good sewage systems, ventilation and running water (all of which slums, by definition, lack) is a humiliating practice that is intimately connected to the conditions under which water-borne diseases take hold, creating life-threatening disease conditions. One macabre joke among Mumbai's urban poor is that they are the only ones in the city who cannot afford to get diarrhoeia, partly because the lines at the few existing public toilets are so long (often involving waiting times of an hour or more) and of course medical facilities for stemming the condition are also hard to find. So shitting and its management are a central issue of slum life. Living in an ecology of fecal odors, piles and channels, where cooking water, washing water and shit-bearing water are not carefully insulated from one another, adds high risks of disease and morality to the social humiliation of shitting in public view.

The "toilet festivals" (sandas mela) organized by the Alliance in many cities of India are a brilliant effort to turn this humiliating and privatized suffering into scenes of technical innovation, collective celebration and carnivalesque play with officials from the State, from the World Bank and from middle-class officialdom in general. These toilet festivals involve the exhibition and inauguration not of models but of real public toilets, by and for the poor, involving complex systems of collective payment and maintenance, optimal conditions of safety and cleanliness and a collective obligation to sustain these facilities. These facilities are currently small-scale and have not yet been built in anything like the large numbers required for the urban slum populations of India's cities. But they are another performance of competence and innovation, in which the politics of shit is (to mix metaphors) turned on its head, and humiliation and victimization are turned into exercises in technical initiative and self-dignifying. This is a politics of recognition from below. When a World Bank official has to examine the virtues of a public toilet and to discuss the merits of this form of shit-management with the shitters themelves, the materiality of poverty turns from abjectivity to subjectivity. The politics of shit (as Gandhi showed in his own efforts to liberate the untouchables of India from the task of carrying away the shit of their upper-caste superiors) in India is a meeting point of the human body, dignity and technology, which the poor are now redefining with the help of movements like the Alliance. In India, where distance from your own shit is the virtual marker of class distinction, the poor, too long living in their shit, are finding ways to place some distance between their shit and themselves. The toilet exhibitions are a transgressive display of this fecal politics, itself a critical material feature of deep democracy.'

Each of these organized practices sustains each other. Surveys are the basis of claims to new housing and justify the exhibition of models and houses built without attention to toilets and fecal management make no sense. Each of these three major practices (refined over more than a decade), uses the knowledge of the poor to leverage expert knowledge, turns the politics of humiliation into the politics of recognition and enabes the deepening of democracy among the poor themselves. And each of them adds energy and purpose to the others. They provide the public dramas in which the moral injunctions to "federate", to "save" and to "set precedents" are made material, tested, refined and revalidated. Thus key words and deeds shape each other, permitting some leveling of the knowledge field, turning sites of shame into dramas of inclusion, and allowing the poor to work their way into the public sphere and visible citizenship without open confrontation or public violence.

VI The International Horizon

The larger study of which paper is a part is concerned with the way in which transnational advocacy networks, organizations of grassroots NGO's, are in the process of internationalizing themselves, thus creating networks of globalization from below. We have seen such networks mobilized most recently in Seattle and Washington, D.C. but they have been visible for some time now in global struggles over gender issues, environment, human rights, child-labor, indigenous cultures. More recently, there has been a renewed effort to link grassroots activists in areas such as violence against women, refugee and immigrant rights, sweatshop production by multi-national corporations, indigenous rights to intellectual property, popular media, grassroots mediations of major episodes of civil war, and many other issues. The underlying question for many of these movements is: how can they organize transnationally without sacrificing their local projects and when they do build transnational networks, what are their greatest assets and their greatest handicaps? At a deeper political level, can the mobility of capital and the new information technologies be contained by, and made accountable to, the ethos and projects of local democratic projects? Put another way, can there be a new design for global governance which mediates the speed of capital, the powers of states and the profound locality of actually existing democracies?

These are large questions and I can hardly engage them in this paper. But they remind me to note that the Alliance in Mumbai, for more than a decade, has been an active part of a transnational network, concerned with "horizontal learning", sharing and exchanging, which has been given official form as the Slum/Shackdwellers International (SRI) in 1999, with federations in 14 countries in four continents. The process that led to this formalization goes back to the mid-1980's. Links between federations of the poor in South Africa, India and Thailand appear to have been the most vital in the gradual building of these grassroots exchanges, and, to a considerable extent, still are. Key to these exchanges are visits by groups of slum or shack dwellers to each others' settlements in other countries, to share in on-going local projects, get and give advise and reactions, share in social and life experiences and exchange tactics and plans. The model of exchange is based on the idea of "seeing and hearing" rather than teaching and learning, of sharing experiences and knowledge rather than seeking to impose standard practices, with the key words being exposure and pooling. There is by now a large body of practical wisdom about how and when these exchanges work best and this wisdom is being constantly refined. Visits by small groups from one city to another either in their own region or to another region, usually involve immediate immersion in the ongoing projects of the host community, such asscavenging in the Philippines, sewer-digging in Pakistan, women's savings activities in South Africa or housing exhibitions in India. Such exchanges are both internal (within countries and regions) and external.

These horizontal exchanges now function at two levels. At one level, they provide a circulatory counterpart to the building of deep democracies locally. By visiting and hosting other activists concerned with similar problems, communities gain a comparative perspective and provide a measure of external legitimation for local efforts. Thus activist-leaders who may still be struggling for recognition and space in their own localities may find themselves able to gain state and media attention for local struggles in other countries and towns, where there very presence as visitors carries a certain cachet. The fact that they are visiting as members of some sort of International Federation further sharpens this image. Local politicians feel less threatened by visitors than by their own activists and sometimes open themselves to new ideas because they come from the outside.

Second, the horizontal visits arranged by the federations increasingly carry the imprimatur of powerful international organizations and funders, such as the World Bank, state development ministries and private charities from Europe, England, the U.S.A. and Germany, and increasingly include political and philanthropic leaders from other countries as well. These visits become signs to local politicians that the poor themselves have cosmopolitan links, which increases their capital in local political negotiations.

Finally, the occasions that these exchanges provide for face-to-face meetings between key leaders in, for example, Mumbai, Cape Town and Thailand actually allow them to progress rapidly in making more long-term strategic plans for funding, capacity-building and what they call scaling-up, which is now perhaps their central aim. That is, having mastered how to do certain things on a small scale, they are eager to find ways of making a dent on the vast numerical scope of the problem of slum-dwellers in different cities. In parallel, they are also deeply interested in "speeding up", by which they mean shortening the times involved in putting strategies into practice in different national and urban locations. There is some evidence that "speeding-up" through horizontal learning is somewhat easier than "scaling up".

For the latter goal, the core leadership of SDI is thinking of ways to build a large trannsnational funding mechanism which would reduce their dependence on existing multilateral and private funding sources and put even long-term funding in the hands of the SDI, so as to further free them from the agendas of projects, donors, states and other actors whose aims are never quite the same as those of the urban poor. The fund is a grand utopian vision, which forms a crucial part of the global political horizon of this international network. It will require the current leadership of SRI to exercise a demanding mixture of political cooperation, willingness to negotiate and stubbornness of vision, in their dialogues with the major funders of the battle against urban poverty world-wide. This objective to create a world-wide fund controlled by a pro-poor activist network, is the logical extension of a politics of patience combined with a politics of visibility and a politics of self-empowerment. It is directly pitched against the politics of charity, of training and of projectization as solutions to the problems of urban poverty. It is also a large wager about the capacities of the poor to create large-scale, high-speed, reliable mechanisms for the global change of their conditions. It is a new vision for equalizing resources and knowledge at one stroke. For the poor understand that the most important thing about being poor is not having money – and that today specific agencies control lots of money and dole it out at their pleasure, in managed doses. The story of the self-organization of this network as a global network is very much in process and constitutes a major site for a politics that is simultaneous post-Marxist and post-developmentalist. It is this larger possibility to which I turn in my conclusion.Ê

VII. Deep Democracy

In the period after 1989, there seems to have been a world-wide victory of some version of neo-liberalism, backed by the global presence of the United States, and sustained by the openness to market processes of regimes otherwise varied in their political, religious and historical traditions. At the same time, more than 10 years after the fall of the Soviet order, it is clearer than ever that global inequality has widened, intra-national warfare has vastly outpaced international warfare (thus leading some to suggest the image of a Cold Peace), and various forms of violent ethnicization seem to erode the possibilities of sustainable pluralism. All this in a period marked by increased flow of financial capital across national boundaries and of innovations in electronic communication and information storage. The paradoxes abound, and have led to the proliferation of new theories of civizational clash and of global gaps between safe and unsafe physical zones and geographical spheres. Fears of cyber-apartheid mix with hopes for new opportunities for inclusion and participation.

In this confusion, now tempered by the knowledge that neither the most recent innovations in communication nor the defeat of the Soviet Union seem to have created the conditions for global peace or equity, two great paradigms for enlightenment and equity seem to have become discredited. One is the Marxist vision, in all its global variants, which promised some sort of politics of class-based internationalism, premised on class-struggle and the transformation of bourgeois politics by proletarian will. This was an internationalist vision that nevertheless required the architecture of the nation-state as the site of real struggles against capital and its agents. In this sense Marxism was, politically speaking, realist. The other grand vision, certainly after 1945, was the modernization-development vision (with its associated machinery of Western lending, technical expertise and universalist discourses of education, technology-transfer and nationally based electoral democracies). This vision, born in such experiments as the Marshall plan, has been subjected to intense criticism on numerous scores, but its most stunning criticism comes from the fact that more than a half-century after the Bretton-Woods accords, more than half the population of the world lives in severe poverty. Whoever else is to blame, the technocrats of global modernization have publically admitted their share in the blame. Both visions, the Marxist and the devleopmental visions, though opposed on key points, had at least two elements in common. The first was a commitment to social and economic equality on a global basis and the second was a reliance on the nation-state as the principal site and mechanism for the mobilization of the poor.

In this context, a variety of other visions of emancipation and equity now circulate globally, often at odds with the nationalist imagination. Some are culturalist and religious, some diasporic and non-territorial, some bureaucratic and managerial. Almost all of these recognize that non-governmental actors are here to stay and somehow need to be part of new models of global governance and local democracy. The alliances and divisions in this new global political economy are not always easy to predict or understand.

But among the many varieties of grassroots political movements, at least one broad distinction can be made: between those who have fundamentally opted for armed, militarized solutions to their problems of inclusion, recognition and participation and those that have opted for a politics of partnership, that is, partnership between traditionally opposed groups, such as states, corporations, workers etc.'

The Alliance and the transnational network of which it is a part belongs to this latter group, which has consciously decided to opt for partnerships of a variety of sorts with other powerful actors, including the various levels and incarnations of the state, to achieve its goals which are to gain secure housing and urban infrastructure for the urban poor, in Mumbai, in India and beyond.

In opting for the politics of partnership, such movements are taking a conscious risk. The risk is that their partners have at least some moral goals in common with them. They are also taking a risk that the hard-won mobilization of certain groups of the urban poor (as political capital) is best invested in partnership arrangements rather than in the politics of confrontation or mass violence. There is even a larger wager involved in this strategy. And that is the wager that the world of multilateral agencies, Northern funders and Southern ggovernments can be persuaded that the poor are the best managers of solutions for the problems of poverty. This last wager is crucial, because it is the basis for investing a large amount of energy in setting precedents for partnership at all levels, from the ward to the world. The hoped for pay-off is that the poor will prove more capable, once mobilized and empowered by such partnerships, to "scale-up" and "speed-up" their own disappearance as a global category, than either the market, the state or the world of developmentfunders. In the end, this is a political bet, about the relationship between knowledge-building and material equalization and about the best ways to achieve it.

In making these wager, activist groups like the Alliance in Mumbai and its global extensions, are also striving to redefine what governance and governmentality can mean. Especially in regard to the numerous levels of the State (municipal, regional and national), they approach their partners on an ad hoc basis, taking advantage of the dispersed nature of the State to advance their long-term aims, and finding their partners where they can. In a country like India, where poverty reduction is a directive principle of the Indian Constitution, and the tradition of social reform and public service is woven into nationalism itself, the Alliance can play the politics of conscience to considerable effect. But even then, they hedge the bets in their favor through strategies of knowledge-building, sharing and multiplication which increase their hold on public resources.

Deep democracy, in this context, has three meanings. One is that the strategies of the Alliance are based on their constant internal commitment to the direct control of major initiatives by the poor, through the federation model and mechanism, which is grounded in radical internal debate, transparency and inclusion. The second meaning is that they have found a way to convince key actors, especially in the state and local administrations who have the greatest say in their own localities, that working with the poor is not just good in principle but also in practice. The third meaning is that global solidarity among local federations of the urban power can best be found by horizontal linkages across local partnerships, thus leveraging multiple sites of local politics, rather than by a direct attack on global issues and interests, as such. Together, these three meanings of deep democracy constitute a new kind of politics as well as a new order of risk for those whose primary interest is in empowering the poor to end their own poverty.

Acknowledgements

This paper is based on research in Mumbai that I have been conducting periodically since early 1998. Since October 1999, it has been funded by a major grant from the Ford Foundation as part of larger study of grassroots globalization. The paper is entirely reliant on the extraordinary generosity of my friends in the Alliance in Mumbai, notable among whom have been (in alphabetical order): A. Jockin, Sundar Burra, Celine D'Cruz, and Sheela Patel, as well as many other members of SPARC, NSDF and Mahila Milan. I am also grateful to several members of the SDI network for including me in meetings and sharing materials which have been invaluable to me: here Joel Bolnick (South Africa) and Samsook Boonyabancha (Thailand) deserve special mention. Finally, several members of the network of activist-funders also must be thanked for allowing me to listen and learn in several Northern venues: Srilatha Batliwala, Marilijn Wilmink and Ruth McLeod know why they are in this list. This community, and many others not named here, have welcomed me to their meetings, their homes and their list-serves, all of which have helped me immeasurably.Ê I have drawn especially on draft papers by Sheela Patel, Joel Bolnick and Diana Mitlin on the horizontal exchanges; by Sundar Burra on the Indian experiences of the Alliance; and by Ruth McLeod on the insights into the process gained in her work with Homeless International (U.K.). These materials have been supplemented by a large number of informal conversations, interviews and observations in Mumbai and elsewhere. Sundar Burra deserves a special word of thanks, since he trusted me enough to vouch for me to this network of activists, among whom I have found some special friends.

I am also grateful to the Amsterdam School of Social Science Research (University of Amsterdam) and its Dean, Peter van der Veer, for providing me with the support and time to prepare this draft paper, while enjoying the privilege of being a Distinguished Visiting Professor at ASSR in June 2000. ASSR staff provided me with every sort of technical assistance in this context.