Contents - Previous - Next

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

The Case Against

Critics of the Narmada projects hold, first of all, that the cost-benefit calculations used to justify the projects are of dubious quality (Alvares and Billorey 1988). For example, the benefit-cost ratio for Narmada Sagar have been revised at least thrice. In the first two sets of calculations (in 1982 and 1984) neither the capital costs of the project nor environmental costs were included. These two calculations generated a benefit-cost ratio of 1.88 and 1.52 respectively as against the Planning Commission's norm of 1.5 for such projects. Thereafter the Government of India's own Department of Environment and Forests estimated the environmental costs as a colossal Rs 30,000 crores. In response to this the Narmada Valley Development Authority generated a new set of benefit figures which are quite dubious since they appear to involve double counting of benefits, as well as questionable calculations of opportunity costs. Similar problems appear to beset the calculations for Sardar Sarovar. Alvares and Billorey (1988) estimate the benefit-cost ratio of Narmada Sagar to be only 0.17 and of Sardar Sarovar to be 0.38, ie well below the figure of unity which would be required for benefits to exceed costs.

Differences on the cost-benefit calculations are, however, a proxy for more fundamental disagreements. Critics of the projects argue along four main lines.

Loss of land and livelihood systems

The loss of land and destruction of the livelihood systems of large numbers of people, many of them tribals, currently living in the proposed catchment areas is the most important of these. According to the Department of Environment and Forests of the Government of India (as quoted in Alvares and Billorey 1988; Tables 2 and 3) the Sardar Sarovar project will submerge 240 villages with a population of 100,000 people, 70 per cent of them living in the state of Madhya Pradesh. The Narmada Sagar project will submerge an additional 254 villages, the homes of
170,000 people, all in Madhya Pradesh. Of these, 60 per cent in the case of Sardar Sarovar and 34 per cent in the case of Narmada Sagar belong to 'scheduled castes and tribes', acknowledged to be the poorest and most socially and economically exploited in the country - the former (and in many instances, current) 'untouchables'. These figures do not include those who may be partially or indirectly adversely affected such as by the backwaters of the Sardar Sarovar dam. The total eventual numbers of displaced persons may be upwards of a million people. The high proportion of 'scheduled tribes' among those whose land will be submerged (as high as 51 per cent in Sardar Sarovar) arises from the fact that the hilly area for the proposed project is heavily populated by tribals who have lived there for long periods of time.

It would be wrong, however, to imagine that current tribal living conditions represent a balance with their environment. 'The romantic image of the "tribal" does not match the reality of adivasi existence as variously experienced in the Narmada valley' (Baviskar 1991). It is all too simple to counterpose an image of tribal livelihoods as deriving from a traditionally harmonious relationship with their environment. But the history of primitive accumulation in the valley goes back at least two hundred years. Tribal resistance to the depletion of forest resources by outsiders (largely from Gujarat) led the British during the colonial period to repress them in a variety of ways including military force, and to encourage non-tribal settlement in the region. Tribal peoples were thus increasingly pushed onto marginal lands in ecologically fragile forest areas. In the following era the exploitation of both timber and non-timber products continued; the growing impoverishment of the tribals led to their increasing transformation into wage-labourers, both within the forests themselves and as migrants to the plains. With their growing involvement in market-oriented production came social and class differentiation among the tribals themselves. They now include both land-owning, settled peasants as well as landless labourers. Some of these peasants, particularly in the districts of Dhar and Khargone, are actually participants in the Green Revolution in the region, using high yielding seeds, fertilizer, pesticides and water. In other districts, tribal labourers are predominant but hardly conform to a romantic image of sustainable livelihoods, even though their survival is heavily dependent on whatever marginal access they continue to have to forest resources. Interestingly, as Baviskar notes, tribal landowners tend to be stronger supporters of the movement against the projects.

As Baviskar (1991) points out, the tribals do not themselves have a romantic conception of 'Nature', the forests or of land. For them, land and forests are a resource, a critical basis of their livelihoods. This is not to deny that their cultural and spiritual beliefs and practices are profoundly rooted in their physical environment; such practices provide the cultural mortar that holds their livelihood systems together. But it does mean that they may not be averse to change, provided the transformation of their livelihood base was in their favour and did not further impoverish them. This includes the dynamic preservation of the cultural practices and social institutions that are the bedrock of their self-image. If economic gain for some among them were linked to the destruction of social institutions and a consequent disorientation and anomie for the majority, then this may well be resisted.

The proposed projects are likely, nevertheless, to overwhelm the existing class differences among the tribals in the region since the scale of proposed land loss is so great.

Poor quality governmental plans

The second source of opposition and controversy surrounding the projects (including the Morse Report) is based on the poor quality of the governments' plans for rehabilitation and resettlement of the proposed oustees. Legal provisions by which the state in India handles the claims of those ousted by development projects are complicated by the federal structure of administration, and by the fact that different states have entirely different laws under which they handle claims. The original law under which the central government addressed the compensation of those affected by its claims on their land was the Land Acquisition Act of 1894. Enacted during the colonial period, the original Act only provided cash compensation (but not alternative land) for those whose land was taken over compulsorily by the state. It was only as late as 1984 that the Act was amended to allow the government to provide alternative land in compensation, but even this was an enabling provision and not a legal requirement. Neither the amendment nor the original Act acknowledged the rights of the landless including labourers and service workers within a community, or of those without formal land-titles but with customary use rights to land such as women. Although the central Ministry of Home Affairs recommended in 1985 that a comprehensive and legally binding national policy should be evolved to include all those displaced by public, private, or joint sector projects, little progress has been made in this direction.

Three state governments, Maharashtra, Madhya Pradesh and Karnataka have, in the meantime, framed laws in this regard. In other states, all that exists are government guidelines which have no legally binding force, and which are arbitrary and often inconsistent from project to project. They typically cover only those with land, excluding not only landless labourers but also artisans and other service workers, as well as adult women, and those without formal titles to land (other than legal tenants). Most provide for cash compensation and a few for alternative land or employment. The Maharashtra Act includes male landless persons
among those considered affected, but the Madhya Pradesh Act does not. Neither Act recognizes the claims of adult women outside the family, but both provide for resettlement and rehabilitation of the entire community of oustees and not just of individuals.

In principle three types of compensation mechanisms are possible - cash, land, or income schemes. Of these, cash is generally recognized as the worst. Typically the land of the oustees tends to be undervalued while the prices of comparable land in resettlement areas tends to be much higher, either because of increased demand (consequent on resettlement itself) or because middlemen take advantage of the ignorance or desperation of the oustees. Alternative land as compensation is in principle much better, and may well be the ideal so far as landed oustees themselves are concerned. In practice, however, it is only marginally better than cash, largely because enough land of comparable quality is rarely available, and because the rights of those without clear land titles remain questionable. Alternative income generation schemes such as jobs training or credit for small enterprises rarely seem to work well in India because of poor programme quality, and also because of the inherent difficulty of converting large numbers of land-oriented people overnight to non-agricultural skills. Compensatory employment probably benefits those who actually obtain jobs especially if these are in the public sector, but few displaced persons actually get such jobs. Nor can public enterprises shoulder the costs involved indefinitely. Thus, no existing compensation scheme reasonably meets the needs of the situation (Mankodi 1992, Viegas 1992). Perhaps the problem has less to do with the type of compensation than with the overall situation which pits unprepared rural or hill people against the vagaries of markets, or the indifference and rigidity of bureaucracies.

In the case of Narmada Sagar and Sardar Sarovar, the Narmada Tribunal's recommendations of land-for-land and resettlement of villages rather than individuals have been observed largely in the breach. There is little likelihood of the Narmada Sagar oustees getting land since it is generally known that little unoccupied land of cultivable quality is available in Madhya Pradesh. Sardar Sarovar oustees have been given the option of either moving to Gujarat to get land or getting only cash if they stay in Madhya Pradesh. But even the land available in Gujarat has been meagre and of very poor quality.

The absence of a coherent plan for resettlement was a major reason for the adverse opinion of the Morse Report to the World Bank. The Bank's own policies require that a resettlement plan be in place prior to the approval of a project. Although Bank staff had been requesting partial or complete plans from the states of Madhya Pradesh, Maharashtra and Gujarat, the response was to produce ad hoc and piecemeal documents. As a consequence the Bank began to lower its sights and settle for whatever it could get, contravening its own stated goals and procedures. Where resettlement has begun, the oustees face considerable conflicts of interest with existing villagers over land titles, as well as problems of water, wood for fuel, and community disintegration. The apathetic response of the Indian state to the growing discontent about the quality of its resettlement programmes was criticized by the Commissioner of Scheduled Castes and Scheduled Tribes of the Government of India in a letter to the then Prime Minister V P Singh (B D Sharma 1991).

Unfortunately, so far as the Indian state's treatment of those displaced by large hydroelectric projects is concerned, past experience with a number of other large dams does not generate optimism. The construction of the Ukai dam which submerged 170 villages (100 of them completely) in Gujarat is a case in point (Mankodi 1992). In that case not only were there the usual problems of bureaucratic insensitivity and lack of a cogent plan for rehabilitation, but compensatory land was made available after extensive and indiscriminate deforestation. Rapid soil erosion resulted in quick deterioration in land quality, adding to the problems of the oustees. The net result of Ukai was that the downstream area, which was already richer, better serviced by transport and communications, and more densely populated, benefited the most from the irrigation and power generated by the dam. The upstream areas were impoverished and ecologically devastated, leading to seasonal distress migration to the downstream areas, which thereby benefited further from the availability of a regular supply of cheap wage labour for new cash crops.

Uneven distribution of costs and benefits

Although each large dam has had its own specific costs and benefits, this kind of lopsided distribution between upstream and downstream areas seems to be a general phenomenon. Uneven distribution of costs and benefits cloaked in the language of national development goals is the third major source of criticism of the projects. In the case of the Narmada projects, particularly Sardar Sarovar, the differential distribution is further complicated by the fact that it crosses state boundaries, with Madhya Pradesh bearing much of the cost, and Gujarat obtaining most of the benefits. Gujarat is a relatively prosperous state, but one with relatively little fresh water resources particularly for intensively irrigated agriculture; Madhya Pradesh on the other hand is a poorer state with a large tribal population. As already discussed earlier, most of the submerged villages and population will belong to Madhya Pradesh. Corresponding losses of biological species and agricultural land will also be high. Under the projects as currently conceived, Madhya Pradesh will only get a share of the power generated, while most of the water and power will go to Gujarat.

R C Singh Deo, former Irrigation Minister of the state of Madhya Pradesh, argues that reducing the height of the Narmada Sagar and Sardar Sarovar dams will not only significantly reduce the numbers of people ousted, but will also make it possible to increase the share of the water to the middle reaches of the river. In this way more water can be used in the dry areas of Madhya Pradesh, but this would reduce water to Gujarat. Singh Deo believes that the Narmada Development Authority has not been interested in such alternatives because of the greater political clout of Gujarat's farmers and industrialists. The political importance of Gujarat to the Congress party (which has been increasingly threatened since the 1980s by the growing power of the Bharatiya Janata Party) and the fact that a former chief minister of Gujarat was inducted as central minister for Planning ensured rapid environmental clearance and planning approval for the current projects.

But Singh Deo also argues that, without significant external assistance, it will be impossible for Madhya Pradesh to complete Narmada Sagar (Singh Deo 1991). Gujarat, on the other hand, is attempting to press ahead with Sardar Sarovar despite the withdrawal of the World Bank, and hopes to tap private capital markets. It has also been constructing a massive canal system to irrigate around six million acres, an amount far in excess of what could be irrigated through its share of the river waters as awarded by the Narmada Tribunal (Singh Deo 1991). But for Gujarat to receive the benefits it hopes for, not only Sardar Sarovar but also Narmada Sagar must be built. Otherwise not enough of a steady discharge of water will be available to allow Sardar Sarovar to operate at its expected capacity. Given the scarcity of resources at the disposal of the Madhya Pradesh government, and given that there is growing opposition to Narmada Sagar within Madhya Pradesh because it threatens to divert those resources (not to mention water) from other smaller projects higher up the river (such as the Bargi reservoir), the end result of all of the expenditure being incurred appears dubious

The irony, according to opponents of the projects, is that the most drought-prone areas of Gujarat - Saurashtra and Kutch - will receive very little of the water from the Narmada projects. Some estimate that only 28 per cent of the state's drought-prone talukas (subdivisions of districts) will receive water from Sardar Sarovar (Paranjpye 1990). The principal beneficiaries are expected to be cash-cropping farmers in the wealthier districts of Gujarat, as well as industry, services and domestic consumers in these areas. The differential distribution of costs and benefits of the two projects is therefore along three axes: upstream oustees versus downstream beneficiaries, the state of Madhya Pradesh versus the state of Gujarat, and differential access among those downstream.

Ecological impact

The fourth important criticism of the projects is their ecological impact. As far back as 1901, the Irrigation Commission appointed by the British Governor-General of India opposed any major irrigation projects in the Narmada valley because of its black-cotton soil which drains poorly and has a tendency towards waterlogging (Paranjpye 1991). Critics of the projects argue that the government of India has overridden the concerns of its own Ministry of Environment and Forests in attempting to go ahead with the projects. As early as 1985, the Ministry had refused to clear Sardar Sarovar because environmental studies had not even begun in some cases and were unfinished in others. But with approval of the project by the World Bank in the same year and under considerable pressure, the Ministry backed down on the understanding that environmental studies would be completed simultaneously with the project itself. This is patently absurd for most of the environmental issues of concern, since project construction would render it impossible to make any other than very minor and marginal changes in design on the basis of the environmental studies. It meant in fact that the project implementers had been given virtual carte blanche.

The absence of serious environmental evaluation and impact studies prior to project implementation is made more serious because almost every aspect of the projects appears fraught with environmental risks. Although many parts of the project area itself are considerably eroded and have already high human and livestock densities, there is still much biodiversity in the area which will be submerged. Of the total area of around 145,000 hectares that will be submerged by the Narmada projects, it has been estimated that about half will be agricultural land and the other half forests. While much of the latter is already under intensive human and animal use, large stands of teak, salad and anjan remain. Wildlife species such as panther, tigers, bears, wolves, antelopes and a number of threatened species are also to be found. Worst of all, no comprehensive study of the region's species has been undertaken, and the actual biodiversity loss of insects, plant strains and genetic resources will be unknown. With submergence of the dam areas, and without proper corridors for movement, there may well be considerable loss of fauna. Submergence will also undeniably increase the pressure on surrounding forests, particularly given the poor quality of rehabilitation plans. Soil conservation plans to cope with problems of silting, waterlogging, salinity and waste-water drainage have also been weak.

Opponents of the projects are also concerned over the earthquake risks in the region which is of moderate seismicity (Alvares and Billorey 1988). Reservoir-induced seismicity has been known to occur in other moderately seismic areas such as Koyna in India with considerable loss of life. Nor have the health risks from new diseases such as malaria, schistosomiasis, and other vector- and non vector-borne diseases that have been experienced in the area of other large dams in India, including the Tawa dam in Madhya Pradesh itself, been assessed.

Are there Alternatives?

Among opponents of the Narmada projects there seem to be two different opinions about alternatives. One starts by asking whether national development goals could be met through other, more humane and environmentally responsible, means. Buch and others argue that reducing the height of the dams would considerably reduce the financial, human and environmental costs of the projects while still meeting the needs they are supposed to fulfil. Supplementing this with methods less wasteful of water such as lift, sprinkler and drip irrigation would further improve the benefits. While this would reduce the power generated, reduction in transmission losses (which are known to be very high) might partially offset this problem.

The other approach questions the validity of the concept of national development itself. In a powerful critique Esteva and Prakash (1991) argue that popular movements that challenge current patterns of development will fail to effect genuine change if they fall prey to the seduction of 'universal' goals:

...No genuine consensus is possible when abstract national plans or dreams replace people's real hopes: always concrete, localized expressions of their cultures... It has taken time for people at the bottom of India's social hierarchy to start perceiving that their real concrete worlds will continue to be sacrificed at the altar of national goals monitored by increasingly faceless bureaucracies, distant from the people they are supposed to protect and defend... In more recent years, people's initiatives... are returning people once again to their roots, to their local spaces, to the regeneration of their concrete hopes, expressed in concrete contexts, intended to change the real life of concrete people...

While acknowledging that the weakness of a concrete focus is precisely that it is local compared to the forces it confronts, Esteva and Prakash argue for flexible coalitions of different local issue-based movements as the answer. In my opinion this approach has three flaws:

1. Its own espousal of the 'local' and 'concrete' appears to be overly universal. In many instances local aspirations may be laudable; but they do not become so simply by virtue of being local. They can become so only on some ethical basis (possibly axiomatic) that serves as the touchstone not only for national goals but equally for local
ones. One cannot avoid this by appealing to the universal legitimacy of local aspirations.

2. Local aspirations are not innocent of gender, class, caste or other biases; this point reinforces the first.

3. This alternative entirely sidesteps any possible principles on the basis of which national objectives can be formulated. While this may not be a problem in many instances, it is unavoidable precisely in the case of 'public goods', ie, those where externalities are involved. Thus, it is precisely the question at issue in riparian disputes; more generally, it is central to many ecological problems where resources, costs and benefits spill over local boundaries. In such cases, concepts of collective welfare (more inclusive and sensitive to distribution than the current paradigm) are essential as a basis for conflict resolution. Otherwise one is left with purely political resolutions which may be quite arbitrary in their outcomes.


It should be clear from the previous section that I believe that the alternatives must be framed on the basis of some principles of collective well-being that centrally address the twin problems of distribution and inclusivity. Our review of the Narmada case shows that all three requisites for a perception of fair process have been violated. Social objectives have been defined without taking account of the concerns of large numbers of those who are likely to be affected. The interests of the weak have been given less weight than those of the powerful: all interests have not had equal access to the domain of public discourse.

I would suggest instead that assessments of projects like Narmada should examine the distribution of benefits, costs and access along various dimensions at the very outset. Second, they should particularly address the concerns of those who are socially or economically without privilege, weighting their concerns about livelihoods and survival higher than the concerns of the more advantaged. Third, the actual formulation of projects should put in place mechanisms to give special voice to the under privileged and draw them into the of public discourse. This may mean in some Instances that even projects with otherwise laudable goals might have to be shelved or rejected altogether. But this will be an essential requirement if the project of national development is to be politically and ethically viable in the end.

In some ways the battle over the Narmada appears to have come to precisely such a juncture. After the withdrawal of World Bank funding, and in the aftermath of the threatened collective drowning by anti-dam activists during the monsoons of 1993, the government of India appointed a special commission of public enquiry to review the issues. As is to be expected, the enquiry has itself become hotly contested; nor is it clear what standing its report will have. But the appointment of the commission itself is a kind of hallmark, being the first time in India that problems of distribution and inclusion have come to the centre stage of the debate over environment and development.


Alvares, C and Billorey, R (1988) Damming the Narmada: India's Greatest Planned Environmental Disaster Third World Network / Asia-Pacific People's

Environment Network, Penang

Baviskar, A (1991) 'The researcher as pilgrim' Lokayan Bulletin 9:3/4 May-August, pp 91-7

Buch, M N (1991) 'And all the boards did shrink' Lokayan Bulletin 9:3/4 May August, pp 47-53

Das, M (1992) 'The internationalization of the Narmada dam: do Western environmental groups have a role in Third World ecology movements?' Harvard Centre for Population and Development Studies Working Paper Series no 1, Cambridge

Esteva, G and Prakash M S (1991) 'Re-routing and re-rooting grassroots initiatives: escaping the impasse of sustainable development for the Narmada' Lokayan Bulletin 9:3/4 May-August, pp 113-25

Lokayan (1991) Akhil Bharati Vidyarthi Parishad Resolufion-Narmada Valley project-plan of development or destruction? Lokayan Bulletin 9:3/4 May-August, pp 139-44

Mankodi, K (1992) 'Resettlement and rehabilitation of dam oustees: a case- study
of Ukai dam' in EG Thukral E G (ed) Big Dams, Displaced People Sage Publishers, New Delhi
Paranjpye, V (1990) High Dams on the Narmada INTACH, New Delhi

Sharma, B D (1991) 'Letter to Prime Minister V.P.Singh' Lokayan Bulletin 9:3/4 May-August, pp 126-33

Singh Deo, R C (1991) 'Venting ire upon the Narmada' Lokayan Bulletin 9:3/4 May-August, pp 60

Thukral, E G (ed) (1992) Big Dams, Displaced People Sage Publishers, New Delhi

Vaswani, K (1992)'Rehabilitation laws and policies: a critical look' in Thukral, E G (ed) Big Darns, Displaced People Sage Publishers, New Delhi, pp 155 68.

Veltrop, J (1991) 'The case for' in 'The debate over large dams' Civil Engineering August, pp 42-8
World Bank (1991) Extract in Lokayan Bulletin 9:3/4 May-August, pp 98-103

Contents - Previous - Next