MIKE LEVIEN

A

Satyagraha and the Government’s Betrayal

Sardar Sarovar returned to claim national attention again last April with a month long dharna by the Narmada Bachao Andolan in Delhi, including a 20 day fast by Medha Patkar, Jamsingh Nargave and Bhagwatibai Patidar. Forcibly hospitalised, famished, Patkar and the NBA resolutely demanded that the government halt dam construction until rehabilitation is completed as required by law. The escalation of tactics was a reflection of the tremendous stakes that hung on the decision of whether to raise the dam height, affecting the homes and livelihoods of approximately 35,000 families who would be in the submergence zone at the 122m dam height. Support poured in from all quarters, and solidarity protests were held in cities across India and embassies across the world. Coming after 20 years of persistent agitation and struggle, innumerable fasts, protests, and monsoon satyagrahas, the dharna was a tremendous display of courage and persistence, and a demonstration of the incredibly strong solidarity the NBA has inspired in people across the country.

Nonetheless, and in the face of overwhelming evidence that rehabilitation in the Narmada Valley is incomplete, the Supreme Court and the prime minister failed in their responsibility to uphold the law and stop dam construction. In Madhya Pradesh’s own submission to the court, it admitted that construction on 11 out of 86 rehabilitation sites was still incomplete (itself an underestimation). The group of ministers (GoM) team, appointed by the prime minister to investigate rehabilitation in the valley, found upon their visit that rehabilitation was far from accomplished. The ministers, lead by water resource minister Saiffudin Soz, released a report on April 17 in which they noted “extreme shortcomings” in the resettlement and rehabilitation of oustees and estimated that it would take at least another year for rehabilitation to be completed if the government showed the political will to accomplish this “gigantic task.” In the NBA’s own brief to the court, it thoroughly documented the total absence of land-based rehabilitation in Madhya Pradesh, the exclusion of thousands of people from the official project affected families (PAF) lists, and the faulty and incomplete nature of even the 75 rehabilitation sites that the government claims to have constructed. It is beyond dispute that rehabilitation in the Narmada Valley is incomplete.

It is also indisputable that rehabilitation must be completed before construction on the dam can legally continue. According to clause XI of the NWDTA, rehabilitation of all PAFs in all three affected states (Madhya Pradesh, Gujarat, and Maharashtra) must be provided for one year before each successive increase in the dam height. The Supreme Court has itself upheld this position in its decisions of 2000 and 2005. The court’s 2005 decision clearly states, “In terms of NWDT awards the irrigable lands and house sites were required to be made available to PAFs one year advance to submergence, and requisite amenities were also to be provided. Further, the notices for vacation of the lands are to be given after completion of R&R of PAFs on or before December 31, i e, six months before actual submergence, (likely on July 1 of the next year). In terms of these stipulations, raising of the dam

Economic and Political Weekly August 19, 2006 which would cause submergence would not be permitted unless rehabilitation programme is carried out.” The judgment thus clearly and unambiguously states that no construction can be allowed on the dam unless rehabilitation sites are completed one year before submergence and PAFs shifted no later than six months before submergence. Yet, the court has inexplicably failed to uphold even its own judgment.

In its 2005 decision, the court also upheld the principle laid down in the NWDTA that rehabilitation must be land-based. Cash compensation, which the Madhya Pradesh government (GoMP) has been distributing in lieu of land, is explicitly prohibited by the NWDTA, which mandates that all PAFs be given two hectares of cultivable and irrigable land either in the command area of the dam or in their own state (according to the oustees choice). Land-based rehabilitation is also mandated by ILO Convention 107 on indigenous people, to which the Indian government is a signatory. The convention clearly states that in the exceptional circumstances in which indigenous people are to be forcibly removed from their lands, they must be given replacement land that is as good as, if it not better than that from which they are removed. Thus, cash compensation, as per the NWDTA, the Supreme Court’s own decision, and the ILO, is completely illegal for the SSP. Nonetheless, GoMP refuses to recognise this fact and continues to illegally distribute cash instead of land. Amazingly, in Madhya Pradesh, where 193 of the 245 villages affected by the SSP are located, almost no PAFs have been given replacement cultivable land!

Thus, the petition filed in the Supreme Court by 48 dam-affected villagers from Madhya Pradesh was air-tight. The law is indisputable that rehabilitation must be totally completed and all villagers totally shifted six months before submergence. The facts were clear that rehabilitation was incomplete, even by GoMP’s own admission, not to mention the much graver shortcomings identified by the GoM and the affected people. Thus, both the law and the facts were clear, with the one standing in indisputable contradiction to the other. Unfortunately, this is not always enough. The highest court in the land refused to enforce its own previous decisions and enforce the law. The PM refused to step in and halt the construction by executive order, which is his prerogative as per the Supreme Court decision of 2000. Instead, he appointed, and the Supreme Court endorsed, an Oversight Group headed by V K Shunglu to investigate the status of rehabilitation in the valley. The Oversight Group, with the help of the National Sample Survey Organisation (NSSO) was to survey damaffected villages in MP and report back to the court and prime minister. In the meantime, construction on the dam was allowed to continue.

On July 11, two hours before the court would hear the matter, the Shunglu Committee finally released its report to the public along with the PM’s announcement that he accepted its findings as “fairly accurate” and did not find any reason to stop dam construction. Among the committee’s specific findings were that the state’s estimation of PAFs was largely accurate; that out of 86 rehabilitation sites, 24 were poor and 25 average; that there were significant shortcomings in civic amenities and construction at these sites; and that there were 4,000 grievances pending by PAFs with the Grievance Redressal Authority (GRA). The committee asserted that cash compensation was an adequate substitute for land and that PAFs were accepting it voluntarily. Further, it recommended that clearance be given for construction up to the full dam height. Essentially, the Shunglu Committee’s report, as will be shown, drastically underestimates the extent to which rehabilitation is incomplete, justifies illegal rehabilitation policies such as cash compensation, and refuses to acknowledge that even the shortcomings they did find in themselves make construction illegal. Before engaging the specifics of the report, however, it is important to put it into context.

The entire premise of the Oversight Group (or Shunglu Committee) was flawed from the beginning. The idea of allowing dam construction to continue while investigation of rehabilitation in the valley was underway, is entirely contrary to the law. In the face of overwhelming evidence that rehabilitation was incomplete, the Supreme Court and the PM were legally required to halt any further construction. Instead, they created the Shunglu Committee and gave it a mandate to report back to the PM on its findings by June 30. The PM then would have six days to consider the report and issue a recommendation to the Supreme Court, which would make a judgment by July 10. But, by this time, construction of the Sardar Sarovar up to 122 metres would be a fait accompli. The rains would have begun, and no matter what the committee found, up to 35,000 families would be facing the possibility of submergence in August-September, with or without rehabilitation. This, of course, is entirely illegal and a violation of the rights of the dam-affected families. And as past experience has shown, without the threat of halting dam construction, the respective governments have no incentive to actually provide rehabilitation to affected families. Any person can travel to the Narmada Valley, and the various resettlement sites and see thousands of families affected at 110 metres and below who have yet to receive land-based rehabilitation. But with no power to prevent submergence even if it did recognise this reality, the Shunglu Committee appeared to be merely a diversion from the start. One strongly suspects that its creation was simply a stalling tactic, allowing construction on the dam to proceed without prior rehabilitation.

Besides this fundamental flaw, the Shunglu Committee’s methodology for conducting surveys in the Narmada Valley was itself highly problematic. This begins with the terms of reference given to it by the PM’s office. First, the investigation only looked into R&R in Madhya Pradesh where 175 villages are affected at the 122 metre dam height, and not Maharashtra and Gujarat where 33 and 19 villages are affected respectively. While the governments of both states claim rehabilitation to be complete, this is far from the case with thousands of PAFs at 110m and below yet to receive cultivable, irrigable land and many still languishing without rehabilitation in the valley. But the Shunglu Committee did not even visit the villages in these states. Further, the committee only investigated those displaced at 110 and 122 metres, even though there are thousands affected beneath 110 metres in all three states who have yet to receive rehabilitation.

Further, the committee’s investigation was hamstrung by its insistence on only using the Madhya Pradesh government’s own data on PAFs, provided in its Action Taken Report (ATR). Thus, for the most part, the committee was only verifying whether the people on the government’s existing PAF list were rehabilitated or not. While in some villages they took the names of those who claimed to be affected but were undeclared, people from multiple villages report that this was not done in their village. There was no comprehensive attempt to do complete village surveys and ascertain who in fact would be affected and was not on the PAF lists. Yet, one of the biggest problems with rehabilitation in the

Economic and Political Weekly August 19, 2006

valley is that thousands of people have been excluded from the PAF rolls altogether. Many people are left off these lists because of faulty surveys, the failure of the government to recognise the traditional but untitled landholdings of adivasis, and the government’s unwillingness to properly identify major sons of landholders (who are also entitled to land). In addition, over the past several years, the MP government has resorted to large-scale statistical manipulation in order to simply delete families from the PAF list whom it finds inconvenient to rehabilitate. An analysis of Narmada Control Authority (NCA) documents shows that GoMP has made terribly inconsistent estimates of PAFs, erasing thousands with the stroke of a pen from one statement to the next.

For example, at two meetings of the NCA’s R&R Sub-Group (February and November 2002), GoMP testified that there were a total of 12,600 families from 135 villages in the state affected by the dam at 110m. This was the number it had consistently given at meetings over the course of several years. Nevertheless, in April 2003 it made the claim that now only 8,406 families were affected at 110m, thus reducing the number of PAFs it recognised by a third. The GoMP justified these reductions by making a novel distinction between “temporarily” and “permanently” affected people, arguing that only those affected permanently at the present height (meaning those families whose agricultural lands will be submerged for the entire year or whose house plot is affected) needed to be rehabilitated before a height increase. This is nonsensical since even if fields are submerged temporarily, the year’s crops will be destroyed. Further, this distinction is contrary to the NWDTA and was found illegal by the Supreme Court in its 2005 decision. Nonetheless, these deleted PAFs are yet to be accounted for. So, by only verifying the government’s existing ATR report, and not inquiring any further, the Shunglu Committee largely missed one of the biggest problems with rehabilitation in the Narmada Valley.

There are also problems with how the NSSO teams were actually asking the questions of those it chose to interview. The “surveys” (really just verifications of the ATR) were being conducted with standardised questions, demanding simple yes/no answers. With no open-ended questions, the teams could not delve into the complicated reality that people faced with respect to displacement and rehabilitation. This created numerous absurdities. For example, in one village an interviewee reported that he had not received compensation for his soon-to-be submerged house. Instead of noting this, the surveyor asked him if he had received a houseplot. The interviewee responded that he had on paper, but had not received compensation for his house and thus could not build a new one, and that moreover there were no civic amenities like electricity at the site. But the surveyor simply recorded that the person had received a houseplot! In another example, surveyors refused to listen to villagers who were trying to tell them that their land would become “tapu” (surrounded by water) because they did not have a column for that on their survey. Without open eyes and ears, how could the survey team understand the actual situation in the valley?

Further, reports from many villagers suggest inappropriate behaviour by government surveyors, which call into question their “independence” and impartialness. While surveyors initially stated that they would not be accompanied by government personnel, in some cases even NVDA officials were seen with the teams. This suggests improper influence by an agency that has a vested interest in continued dam construction and, as a part of the MP government, is itself a respondent in the Supreme Court case. Meanwhile, villagers not being directly surveyed and the people’s representative organisation (the NBA) were not allowed to talk to surveyors. Even more disturbing, surveyors “propagandised” and even made speeches to villagers about the dam project. Some surveyors encouraged villagers to accept the meagre (and illegal) cash compensation that the government of Madhya Pradesh was offering instead of demanding the land they are legally entitled to. In another village, one of the surveyors (who happened to be from Gujarat) actually gave a speech to PAFs about how the dam was in the national interest and that they should make a sacrifice for it. Another told them that they should move to Gujarat. Thus, the NSSO teams did not at all appear to be conducting an impartial survey. Further, numerous reports of surveyors recording responses in pencil, and the refusal of the NSSO teams to distribute questionnaires to the people, have further undermined confidence in the survey.

Omissions and Inadequacies

Given these premises and methodology, the Shunglu Committee’s report is suspect from the beginning. But, the report’s specific findings and recommendations are also highly problematic, containing numerous underestimations and contradictions, and yet ironically still proving grave illegalities in rehabilitation. To begin, the report claims that the survey found “no deviations” between the GoMP’s estimation of affected families and the survey data. However, it also reports that 6,485 families who are not on the PAF list claimed to be affected. But it then dismisses these claims on two grounds. First, it suggests ignoring 4,000 claims because they have cases pending before the GRA. This is nonsensical. These petitions to the GRA reflect the large-scale lack of rehabilitation in the valley; how can they be excluded without verification and, the conclusion reached that the government’s PAF lists are correct? Second, the report claims that most of these petitions are bogus and made by people who have come to the villages and constructed houses for the purpose of receiving compensation, basing its argument on what it claims are dramatic changes in demographic census data. Yet clear absurdities show this data, and the allegations on which they are based, to be groundless. For example, the report’s tables show the village of Picchodi and Anwli, among others, as having no population in 1991 and suddenly developing human habitation by 2001. Both are decades-old villages with approximately 800 and 600 families respectively, all PAFs. Meanwhile, the survey lists other villages such as Khujawa as having a population in 1991 and no population in 2001. Anyone can go to the village of Khujawa and see that there are approximately 125 houses with a population of roughly 900. Most glaringly, the Shunglu report itself shows that Khujawa villagers’ houses were acquired in 2003. Whose houses were acquired if there was no population in the village? Thus, these few examples show the Shunglu Committee’s “demographic changes,” by which it is actually levying the charge of “cheating” on the people in the valley, are based on incorrect data and thus cannot be used to dismiss the over 6,000 people claiming to be affected.

Moreover, people in many villages such as Semalda of Dhar district, Kukra of Badwani district, and almost all of the 26 villages in Jhabua district claim that schedule II (the part of the survey that is supposed to enumerate people not on the official lists) was not even administered. This might be one of the reasons why the Shunglu report did not include a villagewise breakdown of the number of people claiming to be PAFs. In many of these villages, those present report that the

Economic and Political Weekly August 19, 2006 surveyors spent minimal time in the villages, and in some cases didn’t even get off their boats. In only a few villages, due to the presence of outsiders, did they conduct house-to-house surveys. This brings up a larger issue with the Shunglu report, which is how it is possible to do any kind of comprehensive survey of 177 villages in only 18 days. The joint task force conducted in Maharashtra, which surveyed only 33 villages affected by the dam, took a year and a half to complete. The Shunglu report itself admits that it was not possible to evaluate judiciously the petitions made by people claiming to be affected.

One of the most egregious aspects of the Shunglu report is that, in addition to simply verifying compliance with the law, it attempts to push its own interpretation of the law itself. Most problematically, it tries to advance the idea that cash compensation (known as the Special Rehabilitation Package (SRP)) is a legal substitute for land. The report states that, “SRP, in the opinion of GRA, is a legitimate substitute for providing land for land as stipulated in the NWDT Award.” First of all, the GRA has no power to overrule the NWDTA and two Supreme Court decisions, which clearly mandate that rehabilitation must be land-based. The committee justifies its stance by saying that the acceptance of cash compensation was voluntary. But this is belied by its own finding that the land bank of Madhya Pradesh contains totally uncultivable and unirrigable land. Many of those who accepted cash compensation did so because the government refused to offer them good land, and they feared that they would get nothing else. Few have been able to purchase alternative land with the compensation they have received. The SRP is simply an illegal strategy for evading the responsibility to provide cultivable and irrigable land to every eligible family. Thus, the Shunglu Committee’s advocacy and acceptance of cash compensation instead of land is unjustifiable and has no basis in law whatsoever.

Further, in its findings on the status of rehabilitation sites, the Shunglu report itself acknowledges their inadequacy and incompleteness. The report admits that 24 rehabilitation sites are “poor” and another 25 are “average.” Leaving aside what these 25 “average” sites look like, this means that rehabilitation is not complete in at least 24 sites, which on average, are supposed to accommodate at least 200 families each. This means that even by the Shunglu Committee’s generous count, approximately 5,000 families do not have a suitable rehabilitation site in which to reconstruct their lives and homes. In its description of these sites, the committee reports that, “Around 50 per cent of the plots demarcated on government land appeared to be in an underdeveloped condition. These plots are demarcated either on the rocky areas or on slopes of hillocks and/or require substantial filling.” Further, the report states that, “The water storage system at a majority of the sites was not satisfactory” and that “there is a need to ensure continuous supply of electricity.” With regard to civic amenities, while some structures are built, “In almost all cases, these facilities are not functioning. These are just bare buildings without any provision of the staff and support structure for their functioning. Many of these buildings were constructed a few years ago and hence require repairs and proper maintenance. The absence of ponds and children parks is rather glaring.” The report concludes that, “There are gaps between the R&R claims lodged by NVDA and those discovered.” How