The negotiators

There exist, in India today, at least two ways of resolving water-related conflicts. One is the way the conflict over river Cauvery is being managed. The river of woes for Tamil Nadu chief minister J Jayalalitha and her Karnataka counterpart S M Krishna, not a single day has passed in the last six months when both haven't worried over their rights. For the states, it is an intense three-decade old battle. The problem is as natural as the flow of the river: Tamil Nadu, downstream of the river, wants water use regulated in upstream Karnataka. Karnataka, in turn, refuses to do so, citing its primacy as the upstream user. The conflict is so overpowering that the prime minister has already called six meetings in the last six months. "Can we solve it anyway?" he asked in a meeting in August. (Good question, for at any point of time, as an official in the water resources ministry informs, about 300 government officials are managing the conflict.) Replied an official present in the meeting: "We can't reposition the states in rotation from upstream to downstream."

The other way -- far less prevalent today, but normative until the 1950s -- can be found not very far from Chennai. In Madainipatty village, Madurai district, stays A Solai, now 85 years old. For more than 50 years, Solai has been the village's water manager. He laughs at the vexed Cauvery conflict and asks: "Along a river there will be people downstream and upstream. There must be some problem with the policy makers." He doesn't remember a conflict over the tank water, shared by farmers of three villages. "If we have any problem, people have given me enough power and freedom to solve it."

In South India
In Karnataka, Tamil Nadu and Andhra Pradesh, the water manager is called a neerkatti . They manage traditional tanks. "Their knowledge of the terrain, drainage and irrigation needs is much beyond the imagination of the present day irrigation engineers," says Ramachandre Reddy, professor of history at Sri Venkatesware University in Madurai, Tamil Nadu. The British first codified their work in Tamil Nadu, testifying to a vibrant tradition. But today, only the Karnataka government recognises this practice and usually allots the management of tanks bigger than 40 hectares to neerkattis.

A neerkatti's role starts much before the monsoon. The tank, being a common property, requires collective action to maintain it. Neerkatti s decide the date on which residents help desilt the tank and clean the catchment. They size up the work required and divide up labour among the tank's beneficiaries. With the first shower they take stock of the water available and so decide per capita allocation as well as the kind of crops to be taken. Thus farmers are thwarted from irrigating fields at their own will; and the neerkatti ensures supply to every field on a rotational basis.

In village Pungamma Cheruvu, in Chittoor district of Andhra Pradesh, neerkatti N R Peddamiah first distributes water to fields farthest from the village tank. "The lands located near the tank will get the benefits of tank seepage water. Those located at the tail end will not, and so I give preference to the tail-end fields," she says. The duration for which a field receives water depends on the crop being grown. If a crop begins to dry, she even has the right to divert water to the dried fields by closing the diversion to all other fields.

Neerkattis like Peddamiah don't enjoy political power in the village, but are given immense administrative power by the Gram Sabha temporarily. In general a neerkatti doesn't own land in the command area of the village tank thus making them neutral to the job. "The neerkatti is to the tank what a priest is to a temple," says R Addhinarayanan, regional co-ordinator of Dhan Foundation, a non-governmental organisation (ngo) working to revive this practice. Besides managing irrigation and practical advice on crops, overall crop management like pest attacks and local remedies also forms a part of the mandate. "They are the unwritten local knowledge base," adds Addhinarayanan.

In Uttaranchal
In Uttaranchal water managers oversee irrigation through traditional water channels called guhls (unlined diversion channels from mountain streams or springs to farmlands). They are called kollalu s in Garhwal and chowkidars or thekedars in the Kumaon hills. Their primary responsibility is to ensure that a second rotation does not begin without the last farmer having got his turn. Here too, the gram sabha appoints chowkidar s and kollalu s as per the seasonal requirement of crops, usually in villages with more than 2000 residents and irrigated area more than 5 hectares. Says Umesh Singh Kira, the head of Beyala Khalsa village in Almora district, "There was no formal system of keeping a chowkidar 25 years ago; but due to constant intra-village fights over water, we now have a formal system."

In Maharashtra
Water managers called havaldar, jagliyas or patkaris manage Maharashtra's co-operative phad irrigation system. Cultivator committees appoint managers in the last week of April. The post is hereditary, to ensure loyalty to local practice. Usually landless residents are chosen. Their roles and responsibilities are formally recorded in a panchnama (a kind of 'minutes').

Each kind of manager has a different role. Usually one or two havaldar s are appointed in each village. Their job is to look after, regulate and maintain flow of water from bhandara s (diversion structures) into the main canals. Patkari s oversee water channels from the main canal to the distributory channels; they guide jagliya s, who see to the field channels and protect crops. All are appointed just before the cropping season. Generally four to eight jagliya s are selected in a village (it depends on how much land is to be irrigated). If the water flow is less, patkari s can supply water on a rotation basis to cultivators.

Scarcity of water nowadays ensures patkari s play a major role in conflict resolution, due to their knowledge of water flow in different canals. In fact in a few villages patkari s are appointed just to take care of conflicts. "Amid conflicts our job is a big challenge, but we always win," says patkari Sayeed Isak Kasam of Neer village in Dhule district.

In Ladakh, the churpun or water manager is an imposing convention. "No body can touch the canal without the permission of the churpun ," says Sonam Loldan, sarpanch of village Saboo near Leh. Perhaps the only place in India where the water manager is fully in charge, Ladakh has a system to ensure everyone learns the art.

Here, churpun s are appointed in rotation every year. Every villager must have hands-on experience; for the younger generation, being a churpun is mandatory. In Saboo, for instance, of the four churpuns appointed every year, two are from the younger generation and two are senior residents with experience in irrigation management. Indeed, even an outsider settling in the village has to be a churpun once.

The major source of irrigation in Ladakh is meltwater from glaciers ( kangri s) and snow. This is complemented by water from springs ( chumik ) and marshes ( nyema ). This water is diverted to individual fields through an intricate network of earthen channels. Prior to the annexation of Ladakh to the kingdom of Jammu, the king's edicts had defined the rights and rules governing the allocation of water within and between villages. These records were replaced by the Riwaz-i-abpashi . The Riwaz-i-abpashi , similar to the mamulnamas of the tanks in South India, outlines the rules for water distribution. Amazingly, it exists only in oral form. Churpun s use these rules even today.

The cultivation period is limited to 3-4 months of the summer, when it is warm enough to grow crops. Then the churpun swings into action, regulating water in each and every canal. In case of dried-up fields with standing crops, it is totally the churpun 's discretion to divert water and save the standing crop. "At times he will also bypass the irrigation of other fields. So his knowledge is crucial for averting crop damages," says Loldan. Wheat and barley are the two major crops grown every alternate year and any change in cropping pattern is also decided by the churpun .

Once a churpun is selected, a document called the kamgya is drawn up. It is like a contract between the newly appointed churpun and the village community, outlining responsibilities and stating his moral obligations, in that he must be impartial. The kamgya also mentions how much the churpun will receive for services rendered. In this way, a certain formality defines the process which both villagers and churpun adhere to.

Impartiality is built into the rules for allocating water within a village: a kind of lottery system is used. By this, each household's turn for receiving water is determined. This is followed most stringently, when there is scarcity of water at the beginning of the season.

Apart from supervising maintenance, the churpun mediates major disputes. According to the people, the likelihood of a dispute occurring is inversely proportional to the abilities of a churpun . Disputes are usually petty, and resolved with the imposition of minor penalties by villagers themselves. The symbolic power of disrepute is enough to prevent people from stealing water or disrupting the system. For disputes of a more serious nature, the village head also mediates. In case the problem is really serious, people refer to the tehsildar , who takes recourse to the Riwaz-i-abpashi to clarify the existing norms.

Jairam Sravan, a staunch patkari in Jaikheda village of Nasik district in Maharashtra, is now redefining his life. For 11 years now, he has had no patkari work to do. First, the irrigation department took over. Then, the bhandara broke down and the main canal got clogged and blocked up. Farmers sold off their lands. "I have to look for an alternative livelihood," he says. Says S G Bhogle, faculty of social science, Water and Land Management Institute (WALMI), Aurangabad, "Due to huge water interception upstream [by dams], the phad system as well as the water managers are in a state of collapse."

The report of the irrigation department at Sakri, Dhule district, is symptomatic of the decay. Just before Independence, there were 66 bhandaras on the Panjhra river in Dhule districts, with a command area of 5569 hectares (ha). A study of the Panjhra project in 1964 recorded about 45 functional bhandaras. When the Panjhra Medium Irrigation Project came up in 1971, most bhandaras downstream became non-functional. Today, the patkari or jagliya tends to a mere 1500 ha. Except in a few villages, a management practice first evolved in the 17th century has ceased to exist.

On March 10, 2003 the Dhan Foundation organised a convention of water managers in Madurai, Tamil Nadu. 1000-odd neerkattis gathered to tell a story which contained the same message: neerkattis were on the verge of extinction.

In 1950, in Tamil Nadu, Andhra Pradesh and Karnataka, neerkattis helped irrigate 4.15 million ha of land (18.6 per cent of India's net irrigated area at that time) By 1990, the area had shrunk to 2.9 million ha. Neerkattis used to manage 1,60,000 tanks in the three states. But now, as the Dhan Foundation's estimate shows, they look after a paltry 1000. Says A Vaidyanathan, former director of Madras Institute of Development Studies, "The intervention of government in irrigation is the sole reason for the decline of the neerkattis."

Post-Independence, India's hydraulic environment underwent a major change. Large dams, canals, and irrigation departments replaced the water managers. Overall, there was a great expansion of government control over irrigation. Says S M Ratnevel, retired chief engineer of the Public Works Department, Tamil Nadu, "In the process government forgot the traditional managers and they are not serious in rehabilitating them."

A major reason for this decline, especially in the three southern Indian states, was the shift in government policy of focussing on groundwater rather than surface water. The emergence of well-driven irrigation ensured farmers lost interest in tanks, and their management. A well is a private resource; a tank, on the other hand, is a common property. Moreover, well irrigation is more stable and reliable. So farmers preferred individually owned wells rather than community owned tanks.

Consider the scenario in tank-irrigated Andhra Pradesh. In 1955-56, when the state was formed, there were 58,527 tanks irrigating about 10.68 lakh ha of land, or two-fifths of the state's total irrigated area. Since then, the number of tanks have increased (70,000 in 1993-1994), but - amazingly - the area under tank irrigation has declined to 8.62 lakh ha. Why is this so? It is so because, in the same period, the area irrigated under wells increased from 2.84 per cent to 9.03 per cent, and the area irrigated under tube wells increased from 0.66 lakh ha in 1970-71 to 7.73 lakh ha in 1998-99.

Such is the impact of groundwater use that in Karnataka, most existing water managers have abandoned their profession. In Pavagada taluk - with bore wells and open wells in every field - a handful of farmers depend on tank irrigation. Rues Ramanjanayappa, neerkatti of Doddachelluru village, Chitradurga district, "I cannot give up my traditional profession because it gives me something to eat."

A similar shift has caused water managers to virtually vanish in Uttaranchal. The first water legislation related to control, maintenance and management of irrigation channels or guhls was passed in 1917. This was the Kumaon Water Rule, which transferred ownership of water resources to the state. The rule stated any guhl to be constructed by an individual or community Had to have the prior permission of the irrigation department. This system persisted after independence. Says Raghubir Singh Negi, the head of village Maletha in the Tehri-Garhwal district, "Since the government has taken over the management and maintenance of guhls and levy water tax on people for the same, water managers no more find acceptance at the village level." Government collects a water tax of Rs 2,300 from Maletha residents. The kollalus are appointed on a temporary basis to only distribute water in summer. The task of maintaining guhls is now in the hands of the irrigation department chowkidar. Now, the number of guhls in the Almora district alone has reduced significantly: from 2,539 in 1998-99, to about 1,265 in 1999-2000 to 1,136 today. This bears testimony to the fact that these farmer managed irrigation systems have died out.

Other developments have also affected the water manager: urbanization, and the rise of non-farming employment. Says Ravi Chopra of People Science Institure, a Dehradun-based NGO working to revive traditional management mores, "How will we cultivate or irrigate small terraced fields when we don't have labour in the rural areas?" Since the chowkidar gets meagrely paid for his work, he looks to other sources of income, such as daily wage labour. Lalit Pande, director of Uttarakhand Seva Nidhi, an Almora-based NGO, says, "With government control, the very basis of local management has been destroyed. People are now taught to think that management is best done by the irrigation department."

No space even now
Andhra Pradesh and Tamil Nadu have recently enacted legislation on participatory irrigation management to tide over growing conflicts over irrigation and revive the traditional tanks. Even here, there isn't much space for water managers. The Andhra Pradesh Farmers Management of Irrigation Systems Act 1997 - which created about 70,000 water users associations at the village level to control local irrigation - doesn't include neerkattis in the scheme. So with the Tamil Nadu Farmers Management and Irrigation Systems Act, 2000.

Even Ladakh has been shaken up. After the 1962 war with China, a huge military build-up occurred in the region. In 1974, the region was opened to tourists. Such new connections with the wider world means that villages like Saboo, closer to Leh town, are today growing potatoes on nearly 50 per cent of the land, completely changing the cropping pattern. In such places, the churpun is now paid in cash, and not in kind. The prestige accorded to the churpun has also declined. People are reluctant to take up the responsibility, for it is not commensurate with earnings. "In Leh, people refused to take up the job of churpun, hence the government has to intervene by paying salary for the churpun," says Sonam Dawa, Director, Ladakh Ecological Development Group.

Is it time for Union minister of water resources Arjun Charan Sethi to pay heed to traditional methods to resolve disputes? Out of India's 18 major rivers, 17 are inter-state rivers. All are embroiled in conflicts over sharing of water. In no case is a solution visible. Sethi admits that the search for solutions to such disputes has exhausted all sources. "If we can't solve the water disputes among states the central government will create a centralised agency that will take over all disputes," he recently said.

Alternatively, he could turn to water managers. "As the literature on international water negotiations continues to grow, one resource of expertise remains untapped - that of indigenous population," says Aaron T Wolf, a professor of geography in the department of geosciences at Oregon State University, USA, and has been working on indigenous ways of conflict resolution and their relevance to contemporary conflicts. Videh Upadhyay, a Delhi-based lawyer, explains how the expertise should be tapped: "These customary practices and unwritten codes of dispute resolution have to be validated into formal law, so that we can draw lessons from these."

What water managers know, and practice, possesses four principles that could inject much-needed energy into the ongoing, urgent - even desperate - work of resolving water-related conflicts.

• Don't allocate by volume of water. Allocate time. This method allows for local management of a fluctuating supply without even creating a storage structure - such as a dam - at the local level. Existing water agreements between states go by quantity of water. It creates a difficult situation, when the flow of water is less but states downstream place a demand for the agreed amount.

• Plan the priority of local uses before embarking on water distribution. This is very helpful in times of scarcity. In the traditional practices, in a time of scarcity, traditional managers quantified water availability and the demands. Accordingly they suggested a cropping pattern change and per capita water availability in the gram sabha. In inter-state water disputes, there is no such provision; there is no way to institutionally respond to such circumstances

• Strong mechanism for equitable downstream and minority rights. This is the most important. Depending on local topography and needs, water distribution always starts from the downstream to the upstream, but without any inequality in water allocation. The existing agreements don't have provision for such kind of resolution, again because quantity of water is the basis for water distribution. Also, no contemporary assessment of water needs has been done to guide water distribution in an equitable way

• Alternative dispute resolution mechanisms. In the Indian federal structure, water disputes can only by resolved by tribunals, after parliamentary approval. Tribunals take time and usually operate in fixed terms, leaving little scope for innovation.

But in India today
Rivers are managed in terms of colonial agreements. The codes laid down between the British government and princely states were dictated more by political and military needs, making for inflexible laws. These laws - and so the entire institutional edifice - fail to respond to the circumstantial needs of the present. Independence hasn't led to institutional innovation. In India no state has been given an entirely free hand in respect of a common source of water, such as an inter-state river. Under the Government of India Act, 1935 even though water supplies, irrigation canals, drainage and embankments come in the state list, Section 130 to 132 of the Act imposes certain limitations on the provinces in the use of inter-state river waters. These limitations still apply, via a constitutional provision that debars a co-riparian state from developing an inter-state river, regardless of the harm to other co-riparian states. In addition, Article 262 of the Constitution empowers Parliament to provide for "the adjudication of any dispute or complaint with respect to the use, distribution or control of the waters of, or in, any inter-state river or river valley". Article 262 (2) of the Constitution lays down that all inter-state water disputes are to be resolved through special tribunals through Inter State Water Disputes Act of 1956.

Problem is, these tribunals have hardly stood the test of time. They don't deal with community needs. They are bad at dispute resolution. For example the Krishna water dispute, or the Cauvery spat of 13 years, have not been resolved, as also the Punjab-Haryana conflict. Indeed, disputes stretching back to imperial times are still on.

Cauvery, for example
Consider the legal cauldron the Cauvery dispute is still cooking in. The Cauvery treaty is governed by an imperial treaty signed in 1924. It entitled Tamil Nadu to get 60 per cent share over the river's waters. At that time Tamil Nadu was under colonial rule and Karnataka, a princely state. The latter had to honour this unfair treaty, which still governs Cauvery water use. So every year, with irrigation demand growing at about five per cent in upstream Karnataka, the states fight over water share.

Isn't it is an irony that India has not paid heed to its many decentralised forms of water dispute resolution? "If water management is entrusted to the state there will be no community control but political manipulative biases and endemic unfairness. We have, therefore, not just to look at distribution of water between states but at people's own community based systems, which has a lot to teach us," says lawyer Rajiv Dhawan. This could be why the Cauvery or the Krishna have become major political issues.

Three doctrines
Three doctrines govern inter-state water disputes worldwide. The first gives the upper riparian state the right to use water as it feels within its territory. In the Cauvery dispute, Karnataka, though an upper riparian state, does not have this right, thanks to agreements made in 1892 and 1924. However, all parties concerned ask for this right. The second entitles the lower riparian state to receive the natural flow, and prohibits the upper riparian from stopping water altogether except for drinking purposes. Tamil Nadu demands this (and to some extent also receives, by agreement). The third doctrine is called 'equitable apportionment'. Both Karnataka and Tamil Nadu have to balance their respective demands without affecting the river's flow.

India adopts the 'equitable apportionment' formula to resolve inter-state conflicts. This formula is motored by the principle of equitability, and not equality. It is the most widely used principle in the US and UK. It gives the maximum benefit accruing to all the riparian states of the river, keeping in view the economic and social needs of different riparian states. The convention on the non-navigational uses of international watercourses, adopted in 1997 by the UN General Assembly, also highlights this as one of the important principles for joint management of such watercourses.

But...
A problem with this principle is that it is not dynamic and practical (see box: How effective...?). It is difficult to evolve a way to determine the equitable share of each riparian state, in a way that applies to all cases or situations. In the Cauvery dispute, for example, Tamil Nadu got a bigger share of water because in colonial times it was a leading irrigated state. In Karnataka, irrigation developed later. Now when it demands more water, there is no way the right can be settled.

It is more than 30 years that governments, including the Union government, have been trying to quantify the water demand of the two states. Last year, a bad monsoon resulted in less flow in the river. The dispute took a violent turn. Farmers committed suicide in Karnataka, protesting against release of water to Tamil Nadu. In all this while, villages with strong traditional managers smoothly sailed over the water scarcity without any conflicts.

The problems of each state and river are unique and a solution in one case may not be feasible in another. Working out an equitable share of each basin state requires an analysis of complex technical and economic data, and the judicious balancing of conflicting claims on, and uses of, the river. Further, the demand for water in states occurs differentially - the cropping season, or the kind of crops grown, may be different. The quantum of drinking water needs, too, may be different. The present mechanism, and so the agencies evolved to solve such disputes, is unable to address this circumstance.

Traditional practices, on the other hand, are responsive to precisely such situations. It is inherent in the institutional character of such practices. The gram sabha prescribes strict water use, on advice from the water manager. Being a decision enforced through consensus, everybody adheres to it.

"Equitable apportionment based on 'macro-needs' as applied to inter-state basins is a different concept. It is distinct from equitable allocation of water between and within villages. But both these concepts must intersect to incorporate justice. Water must be looked at as a community asset," says Dhawan.

This is precisely where traditional practices become relevant. "They are unique and give power to the powerless. Based on equity principles, they give priority to the disadvantaged communities. These community systems are genuinely 'sarvajanik'," says Dhawan. "If these institutions are revived and an attempt is made to build them up - starting from the minor head level, to the distributary level, and on to project basin level - water disputes can be easily resolved. It is important to build up this organisational chain," says Bhogle of WALMI.

In fact the Dhan Foundation is organising association of the traditional managers from village to district level. From the March 10 convention of the traditional managers, many associations have been formed in Tamil Nadu and Andhra Pradesh. "The most important lesson is to incorporate the principles of these practices into inter-state disputes," Addhinarayanan says.