Abstract
It is only recently that research on Indian groundwater has considered a perspective in terms of commons. ATCHA, an interdisciplinary project that includes among others hydrology, crop modelling and remote sensing analysis, includes such a lens in its study of the Berambadi watershed, Karnataka, India. Participant observation, semi-structured interviews and focus groups have shown that the local aquifers are not managed as a commons, and brought into light several factors hindering collective action. In this paper, these factors are reconsidered, in particular through Ostrom’s criteria. The national policy is currently trying to define a new legal framework for more sustainable management of the resource, but this new law is not known to users and it seems difficult to implement because it calls into question too many vested interests. We argue for aquifer management committees, which could be an intermediary between national policy orientations and users who are (rationally) not endorsing collective action.
Keywords
Instead of bemoaning the tragedy of the commons, however, pragmatic policy-makers would do well to embrace it and use it to organize their thinking for at least the short to medium terms – the pitch to a farmer for saving water cannot be a vague appeal to the common good, but a demonstration that doing so is likely to carry real economic benefits, in the shape of better produce, fewer pests or savings in other inputs. (Hasanain et al. 2013, p. 22).
Introduction
Commons’ has become a trendy word for the last two decades that has almost replaced the ‘common pool/property resources’ concept in the dominant academic narratives. Paradoxically, whereas most of the research results highlight the decline or the disappearing of commons because of globalisation, privatization, land-use change, and so on, the notion is more and more used. What is this word that has become so trendy for describing something that is so threatened? Can we still speak of ‘commons’ if there are no more common management and shared representation of it? How does a good become a commons?
In Hardin’s famous article (1968), any village grassland seems to be a ‘commons’ by nature, and so is the air we breathe or the forests where we go for hunting. It is only because of population growth that their management should stop being ‘common’ and be replaced either by a control by the state, or—rather—by privatisation. Indeed, it is often implicitly assumed that resources such as air, water or forests have specific traits, which naturally lead them to be considered less ownership-prone than, for example, cars or houses. Yet, is there really something consubstantial, inherent or inborn in air, water or biodiversity that makes them be obviously considered as a commons? The answer is negative. It is even more so if we question their situation as ‘a common’, that we propose to write without final ‘s’ in order to emphasise the importance of collective action and social construction in this notion. This contrasts with ‘the commons’, as used by Hardin, which is a word coming from ancient English and designating the common land of a community, that is, a physical, material piece of land. In this paper, ‘commons’ shall refer to a good more or less commonly appropriated, while ‘common’ shall refer to the social construction that manages and reflects it 2
(See also the distinction in the conclusion of this paper between substantial and normative commons). In French, there exists the same distinction between “communaux” (village common lands) and “le commun” (a more abstract word referring to collective action). Unfortunately, most of the French academics who read Ostrom and others lazily translated commons as “communs” instead of keeping “communaux”. This was all the more pitiful since “les communs”, in good French, means the subaltern buildings attached to a castle or manor (stables, etc.). Only recently “le commun” has been born and disseminated with a very voluntary, leftist, often revolutionary meaning (Dardot and Laval 2014).
If there exist no commons and even less common by essence, we can understand more easily why in India (as elsewhere) the status of water or forests are sources of conflicts. There is nothing in the basic nature of water that makes it fundamentally prone to be a commons. It needs a social collective agreement for it to be managed ‘commonly’ and in a sustainable way. The example of groundwater in Chamrajanagar district, Karnataka, is a good illustration of this statement. Our collective research in the [the ATCHA project] 1
This Indo-French project brings together biophysical scientists (hydrology, agronomy, geosciences, remote sensing, and so on), modelling economists and social scientists. The ATCHA project (2017–20) is the follow up of AICHA project which already was assessing and modelling the relationships between climate, water availability, cropping systems, and decision making in Berambadi watershed, Gundlupet taluk, Karnataka. It is associated to the SUJALA project of the Government of Karnataka-NBSS. This work was funded by the French national agency ANR (ANR-16-CE03-0006). The Berambadi watershed belongs to the Environmental Research Observatory BVET (
We, as scientists and citizens, are alarmed about the preservation of what we consider (obviously) as a commons, and first as an endangered heritage 3
Indeed, groundwater can be considered as a commons in different perspectives. For example, through the analysis of 5 years publications in The Hindu, one of the leading newspapers in India, we identified four typical qualifications of groundwater associated with best management measures: (a) endangered heritage whose access must be regulated, (b) limited resource that must be optimized, (c) issue of survival whose access must be ensured, and (d) source of emancipation that must be acknowledged (Richard-Ferroudji 2017).
The first part of this article provides a brief review of the literature on the subject. The second section describes the agrarian ‘tragedy’ of a region where groundwater decline has been dramatic. The third part explains the lack of collective action both by the type of agricultural development and diversification for the last decades and by an analysis inspired by Ostrom (1990). The present legislation and State interventions are also responsible for this situation, the fourth section argues, even though rays of hope are appearing on that matter. The conclusive section focuses on the role of intermediaries to foster collective action.
Groundwater and Commons in India: A Brief Review
What is new in this research? Commons are sometimes considered through what could be called ‘ex-post studies’, analysing the past of the commons and what happened to them. For India, the case of forests and the dispossession of the local populations by the colonial or post-colonial state (e.g., Guha 1989; Sundar et al. 2011), or the case of irrigation tanks (e.g., Aubriot 2013; Mosse 2003) are well known. More recently, the rise of a political wave perhaps to be linked to the free software movements and a new ethic based on ‘the commons of knowledge’ (Dardot and Laval 2014) has been combined with the extension of neoliberal globalisation for further strengthening criticism against private grabbing of resources (Dahou et al. 2013). Research on forest or tanks often refers to a past when these resources were considered as commons—by sometimes falling into the trap of romanticising a golden age that had gone. It is not the case with the study of groundwater that in India has never been managed as a common, with some exceptions (Birkenholtz 2009; Hardiman 1998, Janakarajan 2006). In this case, there is no past to refer to for helping conceive new present modes of property and management. Groundwater nevertheless reacts physically as a collective, since the actions of each user have a direct impact on the resource used by the other users. It is this paradox that the article wants to explore: groundwater has everything to be managed as a common, and yet it is not.
The research debate on groundwater management in India is more than three decades old.
– Either it largely overlapped with studies on irrigation in general, to possibly underline the differences between surface water and groundwater: in particular, the individual use of groundwater is often opposed to the above all collective use of surface water. The institutional and legal division in the management of these two types of water tends to aggravate the local crises (Aubriot 2013). The researchers’ effort can then focus on an institutional analysis aiming to apprehend the two types of resource in the same paradigm (Saleth 2004). It should be noted that there are very few studies on groundwater supply for urban areas, which is, therefore, a ‘blind spot in urban water planning’ (Shah and Kulkarni 2015).
– Either it focused on the aquifer water. Many studies analyse the harmful consequences of the multiplication of tube/borewells (Bhatia 1992; Janakarajan and Moench 2006; Prakash 2005), even though underground irrigation can, in the first instance, or if it is well managed, greatly reduce poverty (Moench 2003). How to shift from groundwater management to the more holistic concept of groundwater governance (Mukherji and Shah 2005)? Much work has been done on water markets, which are most often denounced for their environmental and social harmfulness (Dubash 2002; Shah 2005; Saleth 1994). In many cases, the ‘spatial justice’ induced by the development of borewells since the Green Revolution in areas that until then had no surface irrigation because of their distance from rivers and canals, has often resulted in ‘social injustice’ when falling water tables and the commodification of water have reduced the access opportunities of the poorest (Landy and Varrel 2015; Srinivasan and Kulkarni 2014). But these markets hardly exist in the region studied in our article. This paper, therefore, makes it possible to go beyond the ‘commons vs. commodities’ opposition (the one explained by Moench 1998, for example), since it analyses a case study that does not correspond to either of these terms (water is certainly commoditised through the sale of irrigated crops, but only indirectly).
– It is only recently that Indian groundwater has been considered from a commons perspective. Shah (2005) is critical of this approach. Firstly, in order to combat the ‘anarchy’ of underground irrigation (Shah 1993, 2003), this researcher advocates rather indirect policy instruments, such as electricity pricing (Dinesh Kumar et al. 2013) or a change in the procurement policy of the Food Corporation of India. Secondly, many of the cases he cites show that the villagers know how to get along, but to manage supply rather than demand, to recharge the water tables (water harvesting, and so on) rather than to reduce the use of borewells and overdraft. This is what interests us here, which few authors have dealt with so far. At best, the ethics of customary use rights have been highlighted by historians or anthropologists, but mostly regarding drinking water (hospitality obliges to give a glass of water to every passer-by), and more rarely for irrigation: the fact that some situations of sharing existed in India in the past spurs Moench (1998) to argue that water shares could be implemented successfully with maximum cropped areas as a proxy.
– Moreover, until now, little research has listened to farmers’ narratives and taken into account their knowledge on groundwater (Aubriot 2011, 2013), listened to farmer’s resistance to water policies (Birkenholtz 2009) or considered the ambivalence between groundwater preservation and development (Richard-Ferroudji 2019). In India, groundwater research is mainly carried out by economists or hydrologists, much less by anthropologists, sociologists, or geographers (Richard-Ferroudji 2017).
A Tragedy of Non-Commons: Groundwater Decline in South Karnataka
In the next 20 years, a good deal of what can be done with India’s groundwater resources will have been done—by some 30 million private well owners. If this process of appropriation is to be made orderly, equitable and sustainable, the time for action is now. (T. Shah, 1993)
At an elevation of about 850 m, the twin villages of Gopalpura and Kunagahalli (further: Gopalpura) are located in Gundlupet taluk, Chamarajnagar district, at the extremity of the Maidan plateau, a few kilometres away of the Bandipur National Park and the first heights of Western Ghats (Figure 1). According to our survey, only 20 per cent of the landowners own more than 5 acres. Half of the farmers have borewells for irrigating at least part of their land. The other half practices rainfed agriculture only, with more than 800 mm of rainfall in yearly average. Sixty-seven per cent of the 1925 inhabitants are Lingayats or upper castes, while 23 per cent are Scheduled Castes and 10 per cent Scheduled Tribes.
Since the Deccan Plateau is composed of hard crystalline bedrock with low hydraulic conductivity, water storage is poor. Groundwater transmissivity and borewell yields decrease with groundwater table depth. In consequence continuous pumping causing groundwater table drawdown leads to a disproportionate decrease of the amount of groundwater available for irrigation (Sekhar et al. 2016). This feedback loop makes predefined land-use scenarios difficult to determine, since farmers are continuously adapting their actions according to groundwater availability (Robert et al. 2017).
Gopalpura is far from being the only area in peninsular India suffering from such a decline (Shankar et al. 2011). Because of a spectacular and steady increase of groundwater irrigation since the late 1960s, India has become the biggest consumer of groundwater in the world, with the highest number of borewells (Figure 2). Such a trend benefited the national agricultural production, not to speak of the access to drinking water, but in many regions it was achieved at the expense of groundwater resources. In Gopalpura, groundwater pumping for irrigation started only in the 2000s, and the groundwater table dropped from a depth of about 5–10 meters in 2010 to about 25–35metres in early 2018 (Figure 3).
Such a trend is not due to the decline of annual rainfall, contrary to what many local farmers argue (cf. Aubriot 2011 for similar narratives in the Pondicherry region). As proven by Figure 3, the level of groundwater is declining in the irrigated area whereas in the neighbouring Bandipur National Park (Mule Hole) its level remains more or less constant. Rainfall data in the neighbouring Gundlupet town do not show a visible trend (Figure 4). Many farmers also argue that they are forced to dig borewells because of an increase in the seasonal variability of rainfall (rains delated, and so on). Long term statistics are missing for confirming this. Whatever this variability, we hypothesise that the increasing uncertainty of the dates and amounts of rainfall has just accelerated (not generated) the proliferation of borewells, which is due mostly to the socioeconomic factors to be reviewed below.
A specific process is visible at the regional level. Areas in the eastern side of the Berambadi watershed suffered earlier from the drying up of aquifers, while the western side, where Gopalpura is located, has been concerned later. This is because the rise of groundwater use started later in the west, where rainfall is slightly higher (Sharma et al. 2018). In Puttanapura, a surveyed village in the centre-east of the watershed, the situation is spectacular. About one thousand borewells were dug since the 2000s in a desperate race for water. In 2018, only 15 were still yielding water. Mr. N, a rich Lingayat farmer owning 40 acres, had drilled 10 borewells which were all dry in 2018. Today he grows rainfed sunflower on 20 acres, the rest being sorghum and coconuts or let fallow. For this type of farm, admittedly the process has been very damaging (less annual income and non-profitable investment in borewells) but investment in education and in some (rare) non-agricultural activities provides possible resilience. For small farms, this is another matter. Rainfed crops are insufficiently profitable for making viable the holdings with less than 2 ha.
As in many other Indian regions, a spectacular process is taking place (Landy 2018): the reversal of the centuries-old process of agricultural intensification(more inputs per hectare providing more output per hectare), that in South India had started with ancient diversions of the Kaveri river, the medieval creation of irrigation tanks, and continued with the green revolution. What famines and wars had generated in history, groundwater depletion is generating today: a process of abandoning irrigation for less intensive, rainfed farming systems. Today the trend is a U-turn process of ‘agricultural extensification’, with fewer inputs (notably labour) and less annual gross income per hectare. In Tamil Nadu or Kerala, the process is rather positively triggered, since the main factors are the rising cost of labourers’ salaries and the diversification of the rural economy which leads to abandoning labour-demanding crops such as paddy, to the advantage of coconuts or casuarina. Yet, in this region of Karnataka, many farmers have to survive with only one crop a year, and cease the cultivation of irrigated garlic, banana or vegetables. Some new rainfed crops, like sunflower or marigold, bring good profit; but often the income per worker remains below the ‘reproduction line’ (about 45,000 Rs. per annum) given the small size of holdings and the low level of local off-farm activities (Fischer 2016). Investing in dairy cows is not easy for smallholders who cannot grow a lot of fodder. Hence, small farmers and agricultural labourers try to compensate for their loss of income by migrating seasonally to neighbouring Kerala—or to other villages with remaining underground water. This is the case of Puttanapura labourers, who are hired in villages like Gopalpura, where indeed some groundwater is left.
Is this one more example of the tragedy of the commons? At first glance, it is. In India, unlike surface water that is under the control of the state with a few exceptions, groundwater belongs to the owners of the land. Farmers have started digging borewells in their owned land without collective concentration on the reserve of the aquifer. When their borewells dried up, either they tried to deepen them, or they dug another well nearby. Short term individual interest seems to have been the major engine in this race for underground water. The Figure 3 above clearly shows that all the borewells react similarly, whatever the individual irrigation practices of the farmers: this proves that de facto the water table functions like a commons; all the borewells are interdependent and doomed to a common future. Nevertheless this is not sufficient for making the water table managed as a commons; and even less for making it perceived as a common.
Yet it remains to be proven that, in the past, water has ever been seen as a ‘commons’ for the villagers. Before the borewells, except a few open wells, the only source of irrigation was the village tank, which was somehow managed collectively (Fischer 2016). However, for irrigation purpose, it mostly allowed only a few fortunate farmers having land downstream of the tank (ayacutdars) to cultivate water intensive crops by buffering the variability of rainfall and expanding the cropping season by a few weeks. Furthermore, these tanks had suffered from the takeover by the Government, which even reduced the ‘common’ nature of their management. As for the use of groundwater, it used to be limited to open wells, of which the ‘area of influence’ was very weak (Figure 5). The situation was the type of those described by Shah: ‘Well interference was non-existent and only a fraction of the utilisable groundwater potential was used’ (Shah 1993, 129). Our case study does not present a situation of current degradation vis-à-vis a past period when collective action would have properly managed and protected a natural resource, before population increase or takeover by the Government eventually ruin the long term, common management. On the contrary, from the beginning, the resource was not considered as a commons, even less a ‘common’ in the sense of a resource seen to be managed collectively. There is no ‘golden age’ to regret or to recreate. In the absence of such reference, constructing commons is all the more difficult. Just consider what a participant to a game organised during a participatory workshop (July 2018) told us with a somewhat disillusioned tone. ‘In India, to discuss together is not possible: we don’t share (sic). If his neighbour has water, the farmer wants to abstract water as well, because he can see there is a resource’.
Why No Collective Action Regarding Groundwater in Such Situation?
…the key unknowns in the groundwater game have less to do with water and how it behaves than with people and how they behave under different situations. (T. Shah, 2005, p. 9)
A first explanation for the absence of concentration and collective action is provided through a historical approach, in terms of the evolution of agrarian systems. A second, complementary explanation is provided through political economy perspective, inspired by the works of Ostrom.
The History of the Agrarian System
The recent agrarian history of India, at least since the green revolution, has increased individualism and broken social patterns that would ‘cushion’ the vagaries of rains and prices: marketisation, reduction of the family size, individual indebtedness, collective irrigation replaced by holding’s borewells, and so on lead to an ‘individualisation of risks’ that may largely explain the growth of suicides (Vasavi 2009) and the present so-called ‘agrarian crisis’ (Reddy and Mishra 2009). The green revolution brought a very individualistic approach to agricultural activity. Before it, farmers used to help each other on rainfed fields during sowing or harvesting times. With the mushrooming of borewells, the demand of work in the irrigated fields increased and farmers could no longer find time for mutual help between neighbours or relatives. Borewell agriculture is a private and commercial activity. In the past in Karnataka, negative village karma, such as the non-performance of some rituals by the collective, was believed to entail possibly a drought due to ‘linkages between human moral order and larger ecological conditions’ (Vasavi 1999, 56). Such beliefs are disappearing with the advent of the new agricultural technologies: today getting water depends on the level of the water table in one’s borewell. Water is just a private resource, used as an input in the agricultural process of production. It is so precious that one farmer told us: ‘Nobody can ask not to dig a borewell, not even the government’. Such a position was shared by most interviewed farmers.
Such trend is to be seen in our study area. The history of the local agrarian systems (Fischer 2016), attests that up until the end of the 1960s, rainfed agriculture dominated in the studied Berambadi watershed, even though tanks provided additional water to prolong the growing season of paddy, and a few open wells also existed. The green revolution brought borewells and with it sugarcane, and later irrigated turmeric. Crop diversification impacted also rainfed agriculture: the bimodal precipitation regime allows for two harvests a year (in the past, typically millets followed by legumes), but soon rainfed sunflower took more and more space, followed later by rainfed marigold that is grown under contract with industrial firms. Such a dramatic crop diversification is explained partially by the geographic situation of the region which is at a crossroads between the states of Kerala and Tamil Nadu, just 20 km away. In these states, entrepreneurs are quite dynamic and some of them set up irrigation-dependent, commercial estates in the Berambadi watershed to cultivate ginger, banana, mango or coconut trees.
At the end of the 1980s, nation-wide policies led to the development of village-level dairies and hybrid cows became the norm. Since the early 2000s, subsidies for sprinklers and drip irrigation favour farmers with the means to invest in these techniques. The private agri-business players expanded their grip, that accelerated the spread of bore-well based commercial cultivation. The changes in irrigation methods did not reduce the decline of water tables, far from it: even though sprinklers appeared, followed by drip irrigation that became the general norm, water is hardly saved. These technologies allow to irrigate more lands; drip irrigation is often practiced with furrows in an almost flood irrigation system; and the recent farm ponds (encouraged by government subsidies) are far to alleviate the pressure on aquifers since farmers pump to fill the ponds and irrigate from them. ‘Despite the boom of irrigation and the democratisation of wells, inequalities persist. Farmers of the lowest socio-economic classes had to indebt themselves to access irrigation and they are the first to lose, financially speaking, when water levels drop’ (Fischer 2016, 188).
Despite the dead-end it may represent, the situation at study is relatively simple compared to other regions in India where water markets exist. Many studies have focused on these regions (Shah and Chowdhuri 2017), where dense relations exist between farmers due to the sale of water. In contrast, in the study area, the formal or informal exchange of water is rare. True, it exists a rare system of water sharing (vaara): if a farmer has a borewell with sufficient water, he can decide to share the water with his neighbour. All the expenditures (inputs, labour … ) are shared, as well as the benefit of the sale. Yet this type of water sharing tends to disappear because of groundwater depletion. In any case, its goal is not to save water or avoid digging a new borewell, but to maximise profit by maximising the use of water. One interviewed farmer even said he was doing vaara in order to make money for digging his own borewell. The abstraction and use of groundwater is individual. This may contribute to the absence of collective vision for its management.
Vasavi (1999) had found in rural North Karnataka that ‘though there is a constant emphasis and reiteration of the continuity and identity between nature and society and the need for collective addressing of the larger cosmic order, drought-related stress, especially food scarcity, is experienced on a non-collective basis’ (p. 68). This is even truer today: in case of crisis the villagers accuse rainfall, or the market prices, or the governmental policy, but they hardly change their practices. The rise of commercial agriculture, with inputs and outputs bought and sold outside the village, made them ‘move away not only from cosmogonic understandings of loss but also often from the support of fellow kin, caste and village residents’ (p. 119). The history of the region is one of diversification of agricultural income which was partially based on individual access to groundwater. This diversification took place without pooling of the modes of production (land, capital, money), nor marketisation (with the exception of milk). The political structures of decentralisation (village panchayat), which developed very early on in Karnataka, do not seem to have played an important role in the agricultural development of the region. Needless to say, nothing favoured collective management of underground water.
An Ostromian Approach
An analysis in institutional economics can enrich this explanation by focusing on the present socio-political situation. It is well known that Ostrom’s first book (1990) starts introducing three models: Hardin’s tragedy of the commons, the prisoners’ dilemma game, and Mancur Olson’s logic of collective action. These three models recommend actions by external governmental authorities since they highlight the alleged weakness of collective action in the situation of incomplete information, population growth, and so on. It is less known that their criticism by Ostrom is full of nuances: ‘Instead of being wrong, these are special models that utilise extreme assumptions rather than general theories (…). They are useful for predicting behaviour on large-scale CPRs [common-pool resources] in which no one communicates, everyone acts independently, no attention is paid to the effects of one’s actions, and the costs of trying to change the structure of the situation are high’ (p. 183). In other words, models such as those of Hardin or the prisoners’ dilemma could be relevant for explaining our case study, if only the studied village were characterised by such a situation. Now, except for the last point (‘everyone acts independently’), none of these traits is present in Gopalpura. On the contrary, in this small farming community, discussion and interaction are vibrant, kinship links are intense, and though as we shall see enmity and caste/class inequalities may prevent a fully collective action, the setting is far from larger-scale situations such as the Antartic blue whale fishery hinted at by Ostrom. Yet, groundwater is declining in Gopalpura, and there is no consensus or even discussion about mitigating the process. How could this be explained?
According to Ostrom (1990, 211), stakeholders shall improve operational rules if the following criteria are met:
Most appropriators share a common judgement that they will be harmed if they do not adapt to alternative rules.
Most appropriators will be affected in similar ways by the proposed rule changes.
Most appropriators highly value the continuation activities from this CPR; in other words, they have a low discount rate.
Appropriators face relatively low information costs, transformation, and enforcement costs.
Most appropriators share generalised norms of reciprocity and trust that can be used in initial social capital.
The group appropriating from the CPR is relatively small and stable’.
In our case study, only criteria No. 3 (about the recognised importance of groundwater) is validated. Yet it might be possible that for farmers above 40 years, who have lived the period without borewells, a return to rainfed agriculture is not so alarming. As for the other criteria, the situation is much less positive.
– Criteria 1 (they ‘will be harmed if they do not adopt an alternative rule’) is not fully valid, since most of people consider poor rains, and not borewells, as the responsible of the decline of groundwater. More precisely, everyone knows by empirical knowledge the links between the excessive number of borewells and groundwater depletion; but it is commonly stated that if so many borewells are dug, it is because of the failure of rains. There is no alternative; if they reduce their pumping, then they will lose an income. Irrigation may bring 10 and even 15 times more income than rainfed agriculture (Fischer 2016) Unless the macro-economic environment changes, there is no way to convince farmers to ‘stay rainfed’. So, one might as well keep pumping and we will see what happens, they think. What would be the point of rationing ourselves now, and changing our crops, or livelihoods? When this was suggested during a workshop these authors organised, a farmer exclaimed: ‘But I have to do three crops a year to pay back my loan!’ The farmers also feel that since good rains allow some rise of the water table, the resource is resilient. Some of them hope that after some years the aquifer will be sufficiently recharged to allow borewells to work again. That is what eventually happened in 2018, when due to heavy rains the table rose to optimal level (Figure 6).
As Aubriot (2011) noticed, this reasoning might be a legacy of the times of tank irrigation or open wells, when farmers used to draft water till the annual stock gets exhausted. Since the water table is not fossil but recharged every year, the similarity with tanks is striking and farmers’ behaviour is quite rational. As Shah (2009a) noticed, generally the more the pre-monsoon level of water table is low, the more it is recharged: ‘monsoonal recovery of water levels tends to increase with pre-monsoon depth to the water level’ (p. 8). Yet, in our case study, the farmers have not to wait for a few months, but for several years! Who among small farmers has enough resources to wait for five, six or seven years before eventually aquifers are replenished? The alternative logic to which the majority of external observers subscribe is that farmers would have an individual interest to act collectively to protect groundwater. However, this does not make much sense for the principal stakeholders. Most of them do not, in fact, have a personal interest in protecting groundwater. Why protect a resource which would not serve them, all the more while they could use this resource to increase their profits! Protecting the resource for their children does not make sense either; their offspring will be faced with the same situation. Might as well use the resource entirely if need be and turn toward non-agricultural livelihoods afterwards (migration to neighbouring Kerala, or education in the long run).
In these times dominated by the mainstream paradigm of ‘sustainable development’ for ‘future generations’, such farmers’ representations might appear somewhat deviant for some readers. The farmers are not in an environmentalist vision where a natural resource has to be protected at any cost; water is a means of production and most of the farmers have no naturalist, emotional feeling for groundwater (Aubriot 2006, 2011). This proves Ostrom right, who remains in a perspective of classical economy in terms of resources, without plunging into questioning about the ‘nature of nature’ (Morin 1992) and whether nature is to be protected as a resource or for itself. According to Ostrom and many authors, the definition of the commons is summarised by the term ‘common-pool resource’. This definition had been criticised for its utilitarian and anthropocentric character. Must biodiversity, water, forests, and so on, necessarily be considered as resources in order to be protected? Can they not have an intrinsic value? Can’t nature be protected for nature’s sake? (Newsham and Bhagwat 2016). We see here the limits of these critiques: all things natural are not systematically worthy of conservation in the eyes of stakeholders—here, in the eyes of villagers. In order to rigorously analyse this tragedy of the non-commons, we must forget our assumptions that there is an evident necessity to protect natural resources with no necessary or further justification 4
This applies not only to the farmers of Gopalpura, but also to much of public opinion in India and beyond. While it is often asked to “keep the rivers alive”, it is rarely requested to “keep the groundwater alive” (Richard-Ferroudji 2017).
–Criteria 2 (the farmers will be affected in similar ways by the proposed institutional changes) is not valid either. Farmers know quite well that their existing borewells are impacted by the digging of a new one. Yet, not all villagers understand the complexity of groundwater circulation, in particular of horizontal flows which makes farms interconnected by their pumping actions. Must a commons be visible for it to be considered a common? Groundwater is mostly hidden from view. It can be seen in broad daylight only when it gushes out from a borewell or when it lies at the bottom of an open well. Making groundwater visible is a challenge (Richard-Ferroudji and Lassaube 2020). Another reason why the criteria is not valid is that, since power relationships are very hierarchical in most Indian villages, dominated people may fear, rightly or wrongly, that they shall lose more than the dominant groups in case of change. What is clear in any case is that the present situation—groundwater declining for several years, before possibly rising again during very good rains—is not bearable for small holders, and even less so for agricultural labourers. Our surveys show that when the groundwater level drops, the lower social strata have no other resource than to migrate even more than they already did. Even their milk cattle, which is often an essential income, suffer from drought because fodder is scarce and they may have difficulty watering them if the borewells are dry. The higher social strata, on the other hand, have income in the city or irrigated land owned or rented in other villages: they can wait. In the end, this status quo in terms of groundwater management is not a status quo in terms of social differentiation. The absence of common management is to the detriment of the poor.
–Criteria 4 (farmers face relatively low information and implementation costs) is hardly validated either: admittedly, villagers, men and women, do understand the issue. They stated during workshops and focus groups organised by the ATCHA scientists that there was no need to explain them the situation; they just need a solution. Yet communication between different kinships and communities is not very dense. Disinformation and gossips are another major hindrance. There is also a lack of dialogue between extension services and farmers, between science and society. That is also why criteria 5 on reciprocity and trust is partly validated only. Furthermore, transforming and implementing a system for rationing groundwater, whatever this system could be, has a not negligible short term cost since it means that many farmers shall have to renounce to the profitable irrigated crops.
–Criteria 6, about the stability of the small group of stakeholders, is more or less validated since almost all the water users are from the same village. Nevertheless, a good part of the relationships between people are led by political interests or private conflicts among farmers, in addition to the strong caste hierarchy. Furthermore, some farmers own, or at least cultivate, land in other neighbouring villages; reciprocally, some outside farmers cultivate village lands. As elaborated below, the scale of the village is not the ideal level of reasoning since aquifers ignore administrative boundaries. Lastly, estates, whose size may reach 60 ha and who make about 7 per cent of the cultivated area in the studied region, are owned by people alien to this region, sometimes even to the State (Tamil Nadu and Kerala): they are big water users even if they are not numerous (Hooge 2019). These outsiders tend to break the ‘stability’ of the group. There is little possibility for them to attend village meetings.
At the first glance, the region did not appear far from ‘the ideal logic of common management: one resource, one type of user, one circumscribed territory’ (Dahou et al. 2011, 16). Yet, after deeper analysis, this situation appears not sufficient for generating such a management. During the game we organised 5
The game is played by a group from 5 to 8 players. Players sit in a circle; in the middle there is a bowl with candy in it. They do not know how many pieces of candy there are. The objective of the game is for each person to get as much candy as they can but also for each member of the group to get candy. There are 3 years (with different conditions) and 2 rounds each year.
Year 1: players cannot talk to each other and they do not know how many pieces of candy there are;
Year 2: players can talk to each other but still they do not know how many pieces of candy there are (collective action possible); and
Year 3: players cannot talk to each other but they know the amount of candy (knowledge of the “resource’’).
Year 1 is a regular year: this means there are enough sweets in the middle for each person. The second year there is less: only one per person. And the third year there is a lot: at least 3 per person.
Each person writes on a little piece of paper, anonymously, how many pieces they would like. Then they reveal to the group how much they asked for. If the total asked for exceeds the amount in the bowl, no one gets anything. If the total is equal to or less than the quantity in the middle, then everyone gets some candy. (Roma Hooge, personal communication).
The Groundwater Regulations: A Promising Step Forward?
As long as water remains an open access resource and questions of water rights remain unresolved, there shall persist an incentive problem that will interfere with efficient management of water. To resolve the equity issue further, there is a need to separate rights in groundwater from rights in land and to legally confer them upon local communities at the level of a watershed. (…) This implies a move from groundwater being an open access resource to a common property resource. (Narain 1998, p. 362)
Another explanation of the lack of collective action is the type of policies designed and implemented by the state. The Government has not facilitated any regulation of water use. For the sake of agricultural and rural development, since the 1970s borewells have been subsidised. So has been irrigation electricity. It is free in Karnataka since 1992. Rules are imposing a minimal distance between borewells, but they are not followed, not even by the banks that grant the loans. The only regulation by the State is the rationing of power (between 3 and 5 hours a day—with a target of 7 hours announced in 2019), but less for limiting water use than because in Karnataka electricity supply is insufficient 6
In her similar analysis on groundwater overuse in Tamil Nadu, Aubriot (2006) mentions the dangerous incentive provided by the procurement price of paddy, a water-demanding crop, guaranteed by the Government. This is not the case here since most of irrigated crops (garlic, bananas … ) are not under the support price scheme.
The groundwater legislation is another crucial factor explaining the lack of collective action.
A Common Law that does not Encourage Commons
In India, the present legal regime for groundwater ‘is based on the common law notion that groundwater is part and parcel of the land lying above the resource (Water Governance Facility 2013). Groundwater law in India follows this common law logic and gives uncontrolled power to landowners to use groundwater extracted from their land’ (Cullet 2017, 6).The Indian Easements Act, 1882 is still valid today whereas it had been written following a colonial model, more or less adapted to Great Britain, with its humid climate, little use of groundwater and much larger holdings than the atomised Indian farms of today, that are so many to abstract from the same aquifer. Furthermore, the systemic relations between surface and ground water were even less well known than today (Cullet 2014).
Hence, given that this doctrine requires water to be available to everyone equitably, the land-groundwater nexus completely contradicts the public trust doctrine which is theoretically relevant for water (at least for surface water).The stakes are increasingly serious. Due to the demand but also to a deficient management, in India 29 per cent of districts are in a ‘critical’ or ‘overexploited’ situation (Central Ground Water Board). The situation is critical both socially and ecologically.
Since water is constitutionally a State subject, there is a strong suggestion that the States should follow a model bill defined by the central government. In the 21st century, a first bill was published in 2005, but groundwater continued to appear distinct from surface water, and the nexus lining land and water was thus maintained. The law passed in Karnataka reflects this shyness. The Karnataka Groundwater (Regulation and Control of Development and Management) Act, 2011 7
In 2015, Gundlupet taluk is not listed either among the 12 ‘overexploited’ taluks by the Water Resources Department, G.o.K. (cf.
Yet, this recentralisation process is far to be clearly seen on the ground. During our investigation in the studied villages, we were told about the rule prohibiting borewells with less than 200 m between them, but nobody respects it. At the end of 2016, the Government of Karnataka announced that new borewells would be banned after January 2017: instantly, in the whole State, new borewells mushroomed in order to get in ahead of the ban. The Karnataka Electricity Board accepted to connect them. It is even said that some people bored wells in January and argued that they had started in December but financial hurdles had slowed down the work.
Hence, the local management of water is not a kind of ‘anarchy’ where the village sphere would be cut off from the government (Shah 2009b): many governmental provisions (electricity, subsidies…) exist, and contacts with administration are permanent. Yet, it is even less, at the other extreme, a management by a Leviathan-like, powerful State. It is rather a ‘hybridisation’ mixing government regulations with the informal practices of the villagers (Fofack et al. 2015, about Morocco). Most of farmers do not ignore the rules that favour them (subsidies for irrigation). Yet, firstly the latter are not always accessible (bribes may be demanded), secondly the farmers bypass the rules restricting access to water, with the support of officials themselves. Hence must be considered ‘both the local scale with its various (informal) processes for access to groundwater, and the regional scale where policies governing the agricultural development and groundwater management are implemented. Rather than a stacking of rules, we observed a hybridisation of these rules’ emanating from various actors and various levels (Fofack et al. 2015).
Such is the situation on the field: ‘The exclusive right of landowners, as recognised by the existing groundwater law, contravenes established legal principles such as the public trust doctrine, the human right to water and sustainable development, because these principles do not approve a system of private appropriation of groundwater on the basis of land rights. It is a visible contradiction in the Indian legal system that while these principles are well recognised, one of the critical sources of freshwater continues to be available only to landowners’ (Cullet et al. 2017, 650). Our study shows that the situation is even more tragic, since, in the studied villages, even the landowners have a precarious access to groundwater!
A Step Forward
Yet, is not a lot of improvement to be expected? In 2016, the Model Groundwater (Sustainable Management) Act was passed in New Delhi. This time, the suggestions to the States are much more original (Cullet et al. 2017). The preliminary considerations of the act are remarkable : ‘Whereas water is unitary in nature, requiring the integration of surface water and groundwater, has integral links to land and vegetation and has an intricate relationship with rainwater (through natural recharge); And whereas groundwater in its natural state is a common pool resource and the Supreme Court of India has applied the public trust doctrine to groundwater, in recognition that private property rights in groundwater are inappropriate given the emerging status, conflicts and dynamics of groundwater…’ (underlined by authors). This satisfies long-standing recommendations, who considered that the state, considered a trustee of the national water resources, should distribute individual and collective water rights with ceilings on water draft (Saleth 1996).
The Section 9. Legal status of groundwater also includes: ‘1. Groundwater, as a common pool resource, is the common heritage of the people held in public trust, for the use of all, subject to reasonable restrictions to protect the fundamental right to water for life. In its natural state, groundwater is not amenable to ownership by the state, communities or persons. 2. The state at all levels is the public trustee of groundwater’. This implies that the Panchayat Raj Institutions (PRI) is involved: at the level of Gram Panchayats as well as at the Block and District Panchayats, Groundwater Committees shall be formed; they should elaborate a Groundwater Security Plan in the threatened areas and shall recommend the granting of authorisations.
Yet, have we really entered a straight ‘aquifer-based approach’ (Cullet et al. 2017)?
Aquifers are very often mentioned in the Act. ‘The State Groundwater Agency shall, in consultation with the appropriate government and local communities, demarcate natural or artificial recharge areas of an aquifer or aquifers as groundwater protection zones based on the categories—safe, semi-critical, critical and overexploited—with reference to the latest assessment by CGWB’. Yet, because the act wants to give an important role to the PRIs, following what was long ago suggested by Narain (1998), it tends to see the latter as the basic bricks of the institutional construction, to the expense of aquifers that have no committee planned at their scales. What to do when an aquifer does not match with one only jurisdiction, be it a village or a district? Admittedly, in hard rock settings such as the studied village the surface of water tables follows more or less the topography (when they are not overexploited!). In sedimentary settings, the situation might be more complex.
The act answers by emphasising spatial integration and coordination of scales: for example, ‘the functions of the District Groundwater Council shall include (…) coordination of the preparation of groundwater security plans between panchayats and wards sharing aquifers where their boundaries do not correspond with the boundaries of a single panchayat or ward, as the case may be’ [underlined by authors]. Yet the fact is there is no institutionalisation at the scale of aquifers. ‘As the case maybe’, the act says: this phrase does not sound very appropriate, since by definition aquifers ignore administrative boundaries. Remaining locked within the grid panchayat/block/district/state allows a simplification of the very complex reality, in order to ease the management as well as for valid political reasons 9
“Panchayats/municipalities may not always be demarcated in an ideal manner from the point of view of water resources but they have the advantage of being clearly demarcated in law and have the power to regulate all natural resources together, including in particular land and water. In addition, they have the advantage of being democratically elected and permanent, something that project-based institutions cannot achieve.” (Cullet 2014, 70).
Beside the PRIs’ power, social audits to be organised by the gram/ward sabhas are provided in the act. Public hearings are also planned for environmental and social assessments, in case of ‘projects’. But it has been chosen not to take into account the multiplicity of other stakeholders (associations, firms, users representatives …). Scientific experts are present at the state level only, in the State Advisory Council, whereas they could have been expected at the district level as well (Indian districts have in average 1.8 million inhabitants).
As for our study area, the case of estates does not seem to be addressed in the act. It states that ‘every person is entitled to use groundwater for meeting livelihood needs, so long as such use does not adversely impact the right to water for life of any other person’ 10
Note also that this phrasing says nothing on the needs of future generations.
Also to be noted: “The authorisation holder shall be prohibited from selling, by whatever name or form, groundwater extracted under the authorisation to someone else for commercial use and/or gain.”
It remains, of course, for states, and particularly Karnataka, to legislate according to this model. Then, once the law shall have been drafted and passed, it will remain to be implemented, in the generally unfavourable social and political context of the Indian countryside (caste and class hierarchy, politicisation, corruption … ). In particular, enacting and implementing a regulation affecting farmers on the short run (e.g., banning new borewells) might have electoral consequences that prevent populist politicians to advocate such measures. The aquifers of Chamrajanagar district are probably not close to be regulated in a sustainable way.
Conclusion: Focus on Intermediaries to Foster Collective Action
Our article started from the observation that the farmers of Gopalpura did not manage groundwater in a ‘common’ or even a collective way that could have prevented the aquifers from being depleted. It was understood why by mentioning several factors, including the criteria for common property defined by Ostrom. The national policy is currently trying to define a new legal framework for a more sustainable management of the resource, but it seems difficult to implement. Finally, what prospects should therefore be proposed?
Two research gaps had been identified at the beginning of this paper:
Our article makes it possible to go beyond the ‘commons vs. commodities’ opposition, since it analyses a case study where water is neither a commons nor a commodity. However, the result is comparable to those of research studies on water markets: in the end, the poor are the most penalised.
Few researchers have listened to farmers’ narratives and taken their perceptions into account. Our article shows that the notions of water scarcity are very relative. The rich farmers in the area are enough resilient to bear it—they only have to wait for a year with very good rains for the water table to recharge. But poor farmers, as well as ATCHA project researchers, talk about water scarcity and water crisis.
As for governmental narratives, the new Model Act, 2016, states that groundwater is a common heritage of the people and that the state at all levels is its public trustee. But it is allowed to remain sceptic on the power of performativity narrative. Stating that groundwater ‘in its natural state is a common pool resource’, as the Act says, has little meaning since we have seen how much constructed and socially defined are existing commons. They have very little ‘natural state’.
How to build a common in such conditions? Probably by focusing not only on the natural resource concerned, in this case groundwater, but also by seeing more broadly the modalities of collective action. Brédif and Christin (2009) thus distinguish the ‘substantial common’ (grazing land, biodiversity … what we called the commons, with a final s) and the ‘normative common’, the more abstract result of collective action (a common work, a life together … ). Sometimes the common ‘is approached through the status of the environmental good in its relations with the property; sometimes it is viewed from the perspective of the human collective’s ability to produce rules (p. 6)’. In the first case, the Model Act should theoretically make it possible to break the land-groundwater nexus and create a ‘substantial common’. But it remains to create a complementary ‘normative common’, otherwise the substantial common would have little life expectancy. What is the point of creating a groundwater committee if its members cannot collaborate for lack of democracy, sociability, or shared rules?
This is why we would like to argue for aquifer management committees, which could be an intermediary between national policy orientations and users who are (rationally) not endorsing collective action. In this situation, the aquifer may be a good management unit (Kulkarni and Shankar 2009).There are two reasons for this: a hydrological reason, since the aquifers do not know the administrative boundaries, as we have seen; and a participatory objective, since ignoring stakeholders other than local communities may seem restrictive. Some groundwater and environmental professional and women’s associations (the gender dimension of groundwater is essential, see Solomon and Rao 2018) may be active on the spot: they should therefore be invited to meetings and included as much as possible in participatory and decision-making processes. Under ATCHA project, we are conducting participatory workshops and focus groups in Gopalpura: we gave up our initial ambition to have general meetings bringing together all the classes, castes and genders and we accepted to match the social stratification of the local community by inviting only peers for each meeting. The results are encouraging, in particular for hearing the voice of women and understanding their role in water management.
In France, under some specific conditions, ‘water table contracts’ are established by ‘local water committees’ who define for 5 years an action program. This program is in agreement with the Planning Scheme for Water Management (SAGE) which, at the watershed scale, establishes a ‘common project for water’ with more or less strict rules. These water table contracts are signed by the local bodies concerned, from the village to the district, and by the Water Authority at the regional basin level 12
In France there exist democratic institutions both at the regional basin and at the watershed levels.
Similarly, the SAGEs are drawn up by a local committee including government officials (25%), local bodies (50%), but also users (25%) such as representatives of consumers, local population, fishermen, environmentalists, etc.
This would be to expose oneself to the criticism of Shah (2014) who points out in a comparative study that measures that have worked in the United States or in other developed countries cannot be replicated blindly in India, where the state has fewer resources, where water users are much more numerous, often poor and more dependent on groundwater for their livelihoods. Each country, and often even each region, has to define its own policies according to its specificities. Even though France is one of the developed countries that has maintained a rural culture for the longest period of time, even though its population and government are very attached to maintaining agricultural landscapes for cultural, environmental as well tourist reasons, the fact remains that it has only three per cent of farmers in its working population, while India still has half of its one. The French rural society suffers also from less hierarchy and unequal power relationships than the Indian one.
Whatever the degree of participation and decentralisation chosen, an institutional framework provided by the Government is a necessary support for facilitating the management of common-pool resources—even more when these resources have never been considered as commons and when everything remains to build. In Chamrajanagar district, the ATCHA project is organising participatory workshops with the villagers in order to intensify discussions and generate more reflection among the participants, in the hope that some collective action will take shape. But this shall not be enough if a new legislation is not enacted in Karnataka for delinking the land-groundwater nexus. In a second stage, this legislation should also be implemented on the ground efficiently, sustainably and democratically. Environmental education programs should also improve the knowledge on groundwater of all stakeholders. As already mentioned, you can barely see an aquifer. An aquifer should be made ‘visible’ through better communication (Richard-Ferroudji and Lassaube 2020) before a committee can sit around it. Spokespersons for the aquifers must be recognised and promoted: scientists, among others can ‘voice groundwater’ and make aquifer issues visible at the local level. The task remains immense, and a lot of (ground) water shall have passed under the bridge before its achievement.
