Contents - Previous - Next

This is the old United Nations University website. Visit the new site at

Chapter Ten:National Development and Local Environmental Action the Case of the River Narmada

Gita Sen

No development project, however laudable, can possibly justify impoverishment of large sections of people and their utter destitution.

Supreme Court of India in Lalchand Mahto and Ors vs Coal India Ltd Civil Original Jurisdiction, MP No 16331 of 1982; quoted in Vaswani (1992)

No trauma could be more painful for a family than to get uprooted from a place where it has lived for generations... Yet the uprooting has to be done. Because the land occupied by the family is required for a development project which holds promise of progress and prosperity for the country and the people in general. The family getting displaced thus makes a sacrifice for the sake of the community. It undergoes hardship and distress and faces an uncertain future so that others may live in happiness and be economically better off... If not handled properly, these human beings could even impede the progress of project building. They could constitute pockets of protest, unrest and dissatisfaction....

S C Varma, former Chairman, Narmada Valley Development Agency; quoted in Alvares and Billorey (1988)

All those directly affected by a proposed project should be involved in developing the project concepts. Water resource planning and management are not battles to be won by a particular
interest group, but are a matter of mutual problem solving.

Jan Veltrop, President, International Commission on Large Dams; in Civil Engineering, (August 1991)


Few problems of Third World development are as sharply contested as the displacement of local people by the building of large infrastructure projects. Primitive accumulation, ie, the separation of groups of people from their traditional means of livelihood by others more powerful than they, has occurred in history both gradually and in sharp, concentrated episodes. Infrastructure projects such as the building of large power or irrigation facilities are typical of the latter, though the effects (both positive and negative) are usually felt over many generations. As a result, the ensuing conflicts of interest also tend to be sharply focused, and frame the problems of national development in high relief.

This paper uses the controversy over large dams to explore the contradictions of defining and implementing national development in the face of competing local concerns and needs. The exploration is along three dimensions - national economic objectives, local livelihoods and ecological sustainability. The paper uses the current debate on the projects along the River Narmada, as well as the history of earlier dams built in India to illustrate its arguments.

The conflicts generated by a large hydroelectric project are multifaceted. At one level the issue appears to be the classic dilemma of environment versus development, pitting ecological concerns about the damage to local ecosystems as well as more global issues of warming and loss of carbon dioxide sinks against the drive for higher economic growth. 'Sustainable development', defined by the Brundtland Report as that pattern and pace of development which does not buy the welfare of the present generation at the expense of future generations, provides the solution in principle, however difficult to implement in practice.

In a Third World context, however, the problem is more complex. Ranged against environmentalist concerns are problems of poverty removal, the need to increase agricultural production, the growing energy and water needs of a number of productive sectors as well as domestic consumption. Economic growth has a qualitatively different premise and promise in such a case. The distribution problem is one within the current generation, and is not addressed even in principle by the Brundtland definition of sustainable development which focuses on trade-offs across generations.

How can or should such conflicts of interest be resolved? On the one hand are concerns that fall within the ambit of what has typically been understood as national development objectives; on the other are the fears of eroding livelihoods, loss of community, and the destruction of social institutions. In a Third World context, the environmental problem is in many ways subsumed under more fundamental disagreements over development itself.) What should the project of national development include, and who should define it? Certainly in the Indian case the major ecology movements, eg, the Chipko movement, the Narmada Bachao

Andolan, the Western Chats movement, the current struggle by farmers against multinational seed companies such as Cargill, have all posed environmental questions within a broader challenge to the pattern of national development itself. This paper examines the usefulness of the main framework used for this class of problems, and argues for alternatives.


In the public debate of most nation states, national development has generally been an unchallenged goal whose legitimacy in the general imagination overrides the conflicting claims of particular interests. Its dominant paradigm is one in which private interests are subordinate to and hence must be made compatible with the public good. Its analytical underpinnings include a combination of the economic theory of public goods/externalities and the political science concept of eminent domain. The underlying premise is that private interests and behaviour may generate externalities that are inimical to social welfare; the behaviour of individual agents has therefore to be modified so that their particularistic claims are induced to fall in line with social objectives.

The corresponding problem for policy is how to induce such changes in private behaviour. Under the principle of eminent domain, the state has the right not only to use and dispose of all commonly held resources, but to acquire even resources held privately under clear ownership titles upon payment of compensation, if such acquisition is deemed to be in the larger social interest. The assumption is that the state can compensate losers to induce them to accept the primacy of social goals.

Whether the actual operation of the dominant paradigm is generally perceived to be 'fair' or not in a particular case depends on the following:

This does not deny that some particular interests may actually gain from the state's actions while others lose. But the implicit assumption is that, while some interests might win in this process, this is only because they happen to be in line with the objective social interest, and not because they can exercise special economic, political or social clout. Furthermore, the system must contain democratic checks that will ensure that all those affected can participate in the process of project formulation and decision making, or at least can obtain an impartial hearing of their concerns.

This issue of 'fairness' comes into sharper focus if we consider three sets of questions that challenge both the concept and the practice of national development:

1. What constitutes the public interest? All public policy is rife with divergences between public interest and private concerns. In most cases private interests can be defined relatively unambiguously even though there may be a multiplicity of them. But the definition of what constitutes the public interest is more tricky, since in theory it involves some amalgamation of private interests, and in practice this is in the hands of governments. The critical question when public goals are seen as being considerably at odds with at least some private interests is whether government can be trusted to be an impartial definer and arbiter of social welfare. In principle a democratic polity with separation of powers, and checks and balances on the functioning of the executive branch, may be the key to ensuring that governments are impartial in defining social goals. In practice, the developing world is full of post-colonial governments and state apparatuses that are far from democratic. Even in political democracies such as India, the actual functioning of the system's checks and balances leave much to be desired.

2. Whose interests are taken into account when social goals are defined by governments? (Or what is national about national development?) There is considerable potential for bias when there are wide social, economic and gender disparities between government functionaries and the private agents whose interests are seen as being at cross-purposes, between different groups of agents, and within each sub-group itself. In the context of a large hydroelectric project, those whose land is likely to be submerged are often tribals, while those who are likely to consume the increased water or power are generally non-tribals - either farmers, industrialists or domestic consumers from other regions. Both economically and socially the losers and gainers are very different. Similar and wide differences are likely to separate government functionaries from those whose interests are at odds with the project. There may even be divergent interests within a particular group, ie gainers or losers, with differential perceptions of gain and loss, and different capacities to articulate interests and engage in public debate.

Such divergences are important not only as pointers to the presence of multiple and competing interests, but also because they affect the very formulation of social objectives or national goals. It is highly likely in current political economic contexts that the concerns of those at the lower end of the economic and social scale will not weigh in very heavily in the formulation of social objectives. Their losses will be weighted lightly relative to the gains of others higher in the social order and closer to the formulators of social goals. The practice of national development in a context of wide disparity needs, therefore, to be examined since the distribution of its benefits and costs can be highly unequal. To what extent does it include the concerns and aspirations of those likely to be dispossessed? Or is national development a convenient abstraction for the domination of a subset of particular interests distinguished from the others by class, caste or regional identity? In particular, how does the concept address the security of livelihood systems of the very poor or those who are socially marginal? This brings us to a third critical set of issues.

There are, however, two fundamental differences between the first case and the latter two. In the first case, the regulation of individual behaviour is likely to benefit all those whose behaviour is being regulated by creating a safer and more orderly traffic environment. In the latter two cases, neither the sidewalk vendors nor the oustees are likely to share in the benefits from regulation. More importantly, what is being given up in the first case is not so fundamental a requirement of the individual's subsistence (at least in a Third World or Indian context) as in the other cases. Should the three cases then be treated on par? More specifically, should not the costs in (2) and (3) be weighted higher than otherwise? Secondly, should the compensation provided take into account not only the nominal value of what is lost (in the case of the pavement vendor such costs might be assessed at zero since the pavement did not belong to her/him in the first place, and no compensation may be offered) but a recognition of the disruption of livelihoods systems and security?

3. What if the private interests that are viewed as conflicting with social objectives are driven not by some general perception of private advantage, but by basic survival needs which, furthermore, are not accorded much weight in social goals? The implications of this question become clearer by considering the following. In what way is the problem of regulating the traffic behaviour of automobile owners different from the problem of regulating sidewalk hawkers who make a living from their sales, or from the problem of inducing the potential oustees of a large dam to give up their lands and systems of living? One might view all three cases as being equivalent on the grounds that all three involve the subordination of individual interests to the common good.

The next section uses the case of the hydroelectric projects along the River Narmada to illustrate some of the issues raised thus far.


The proposed hydroelectric projects on the Narmada river in central and western India are among the most hotly debated environment development issues in India today. As we shall see below, the debate over the projects is by no means new. But while the 1950s and 1960s were a period of public optimism and faith in their economic and social value, this is no longer the case. Within the last decade a powerful movement of opposition to the projects (the Narmada Bachao Andolan) has marshalled arguments, evidence and support. Its relative success is evident from the fact that first Japanese foreign assistance, and more recently the World Bank's funding have been withdrawn. The World Bank did not actually withdraw funds but was near enough to doing so, following the adverse report submitted by its Morse Committee, so that the government itself pulled back. Under pressure from those who stand to gain from the projects, and given that considerable financial capital has already been invested, the government of Gujarat is trying to go ahead on its own. Bonds of the Sardar Sarovar Nigam (one of the two major projects) have been floated recently and the public response appears to have been positive since the issue has been oversubscribed. Whether this means that the government will actually go ahead and complete the projects is not clear, since finance is only one of the issues in the debate.

The three sets of issues raised in the previous section are central to the current debate over the hydroelectric projects on the River Narmada. Interestingly, ecosystem considerations, while important, are subsumed under fundamental questions relating to the meaning of development itself. Environmental effects have certainly been discussed extensively, particularly in the context of the benefit-cost analysis used for project evaluation. But although critics of the projects have argued that they are not ecologically sustainable, this is secondary to the principal concern, namely, the impact on the livelihoods of the large numbers of people whose land base will be submerged. Indeed, one might argue that a defining feature of many 'environmental' movements in the Third World is that ecological damage is evaluated through the lens of people's livelihoods. Human activity is viewed as part of the ecosystem rather than as an external force that alters the ecosystem. But human activity is not monolithic in that the activities of some can affect the livelihoods of others. Sustainability in this context means the ability of the system, including its livelihood aspects, to be reproduced over time.

The River Narmada is one of the largest Indian rivers and one of the few that flows from east to west with a basin area of about 100,000 square kilometres. The river originates in Shahdol district in the state of Madhya Pradesh and runs for a total of over 1300 kms before flowing into the Arabian Sea. The first 1000 kms are in Madhya Pradesh, after which the river forms the border between Madhya Pradesh and Maharashtra, and then between Madbya Pradesh and Gujarat, before flowing its last 160 kms through Gujarat to the sea. About 40 tributaries join its waters along the way, and some 20 million people, including large numbers of tribals such as the Bhils and the Gonds, cultivate the land in its basin.

Harnessing river waters through major hydroelectric schemes has been an important feature of India's development plans from their earliest phase. In the mid-1960s the government decided that the waters of the Narmada would be shared between the states of Madhya Pradesh, Maharashtra, Gujarat and the arid western border state of Rajasthan. In 1969 the Narmada Water Dispute Tribunal was set up. Its constitution itself set off a competitive spate of project formulation by each state attempting to maximize its share of the river's waters (Alvares and

Billorey 1988). The two largest and most important of these projects are the Sardar Sarovar and Narmada Sagar dams which the Indian government cleared for implementation in 1987.

The Narmada Sagar dam is to be built in East Nimar district of Madhya Pradesh to irrigate 123,000 hectares of land and generate initially 223 megawatts of electricity (which will later fall to 118MW). The Sardar Sarovar dam, currently under construction in Bharuch district of Gujarat, is expected to irrigate 1,792,000 hectares and generate 300 megawatts (dropping later to 150MW) (Alvares and Billorey 1988; Table 1).

The Case for the Projects

The most consistent case for the projects has been made by the Agricultural Division of the World Bank which, until the Morse Commission report of 1992, was a major funder for Sardar Sarovar. Its argument was that the benefits from the projects considerably outweighed the costs. The benefits included, according to the Bank, significant potential increases in agricultural output, particularly food production, improvements in power availability during peak loads and for agricultural pump sets, and multiplier effects in increased investment and employment. It estimated the ratio of beneficiaries to those adversely affected as being 100:1 even without including the multiplier effects, and argued that the project has the potential to feed 20 million and to employ about 1 million people (World Bank 1991). The justification for these estimates was based particularly on the urgency of water needs in Gujarat, with food and domestic water scarcity, and the cost of relief efforts seen as a major and growing problem.

These people (ie those facing escalating shortages) can be seen as the resettlers of the future who will have to move, perhaps into city slums, to find food and employment, in the event that the irrigation does not come. They do not hold public meetings now, but in the year 2020, if you deprived them of their Sardar Sarovar dam, they would certainly be out on the streets.

(World Bank 1991)

Furthermore, proponents of the projects argue that the costs have been exaggerated in a number of ways. First, they hold that the number of persons adversely affected by dam construction will be lower than the estimates of opponents, since many of the villages and much of the land classified as 'affected' will only partially be affected, not permanently inundated. Second, they argue that while there may be some negative environmental impacts due to loss of forests, wildlife habitat, consequent increased pressure of men and animals on surrounding areas, and loss of religious and/or cultural sites, these costs will be offset by environmental gains as well. The latter will include better micro-climates due to increased evaporation, more habitats for waterfowl, and possibilities for fisheries and more photosynthesis in the irrigated areas. Third, proponents hold that, while problems of resettlement and compensation are important, it would be wrong to romanticize the current living conditions of the tribals. Many of those whose lands are likely to be affected do not live currently in pristine forests, but in poor quality and highly degraded lands.


Contents - Previous - Next