Chapter 3 - Group Size and Successful Collective Action: A Case Study of Forest Management Institutions in the Indian Himalayas

Arun Agrawal

Introduction

An increasing number of scholars, development practitioners, and environmental activists today forward microinstitutional solutions as the remedy for renewable resource scarcities. Their arguments have helped to shift attention away from market- or state-oriented policies as the only two alternatives to achieve development or environmental conservation (Anderson and Grove, 1987; Ostrom, 1990; Ostrom, Schroeder, and Wynne, 1993). The fresh claims on behalf of the local (Chambers, 1983; Korten, 1986; Uphoff, 1986), the indigenous (Cultural Survival, 1993; Denslow and Padoch,1988; Richards, 1985), and the "little community" (Hecht and Cockburn, 1990; Scott, 1976; Wade, 1994) represent a long overdue move.¹

The growing focus on community institutions and indigenous voices recognizes that national and international environmental trends are the aggregate consequence of the possibly independent concrete actions of millions of users. It accepts the rupture between the interests of local populations, and national governments and/or international institutions. But even more appropriately, the focus on the local marks a shift from the preoccupation with the centralized, overarching solutions of the past decades that failed to reverse, and may indeed have contributed to, environmental problems and attendant social tensions.² Existing state policies may have inflicted violence at multiple levels on everyday relations of existence and livelihood in rural areas (Colburn, 1989; Guha and Spivak, 1988).³

The attention to local spaces and communities, thus, forms a critical move in the conversation on development and conservation. The ensuing study builds upon the insights in this literature by interrogating the relationship between group size and successfully achieving collective action. Contrary to a large literature in the social sciences, I question the presumption that smaller groups are more successful than larger groups.

The study analyzes village van panchayats (forest councils) in Almora district in the Indian Middle Himalayas. These community-level councils help residents utilize and protect forest resources in accordance with rules they themselves craft and attempt to enforce. To meet the objective of the chapter, I first briefly describe the process behind the birth of van panchayats. I then examine the interactions between the interests of the British colonial state, and the actions of local populations, and how these led to outcomes that incorporated the interests of village communities. The story portrays how villagers and local communities are energetic agents rather than passive victims.

The sketch of the birth of van panchayats in the region sets the stage for seeking the solution to a puzzling finding of the research: the observation that larger forest communities find it easier to successfully organize for collective action to protect their forest resources. An enormous literature in the social sciences, inspired by the seminal work of Mancur Olson, has investigated why smaller groups are more successful in organizing collective action. The analysis is convincing. Rational individuals, acting in their self-interest, are unlikely to act in ways that would facilitate the provision of collective goods for a group, even if all group members share the same interests. Hammering this insight home, and in the process disrupting Marxist and pluralist arguments alike, Olson showed how smaller groups are better able to overcome the problem of collective action in comparison to larger groups.

The findings reported in this chapter, however, undermine conventional wisdom. Building on the empirical observation that smaller van panchayats find it more difficult to organize successful collective action, the chapter discusses some significant theoretical reasons why larger groups may be more successful. After describing the basic characteristics of the communities among which research was conducted, I first attempt a local explanation of the success of larger van panchayats. The analysis is then elaborated to provide a more generally applicable theoretical explanation. In examining the relationship between group size and collective action, the study makes two major departures. Much writing on collective action focuses on the internal dynamics of a group. The chapter, rather, looks at the external dynamics-relations of a group with other groups. Second, it draws a distinction between mobilizing a group for collective action and success in meeting the objectives of collective action. Using these two ideas, it constructs an argument about why larger groups may be more successful than smaller ones.

The van panchayats of Kumaon

A multiplicity of institutional forms occupies the terrain of resource management in Almora. Three distinct regimes can be identified: (1) reserved forests controlled by the Forest Department, (2) civil forests managed by the Revenue Department, and (3) community forests managed by van panchayats. The activities of van panchayats are the focus of the investigation.

The history of the van panchayats in the Indian Himalayas can be traced to the intrusions of the colonial British state in the early 1800s. From this period onward, the British government made a number of inroads to curtail progressively the area of forests under the control of local communities (Guha, 1990: 44-45). Between 1910 and 1917 alone, the government transferred an additional 2,500 sq. kms. of forests to the Imperial Forest Department. At the same time, it also enacted elaborate new rules specifying strict restrictions on lopping and grazing rights, prohibited the extension of cultivation, sought to regulate the use of fire that villagers believed led to higher grass production, increased the labor extracted from the villagers, and strengthened the number of official forest guards (Pant, 1922).

The new rules stirred villagers into widespread protest. They simply refused to accept the rules, or the fundamental presumption undergirding them-the state's monopoly over all natural resources it deemed significant. The best efforts of the government officials failed to convince the villagers that the forest belonged to the government (Ballabh and Singh, 1988). The government had hoped that hill residents "would gradually become accustomed to the rules," but "the hill man (proved) impatient of control" (KFGC, 1922: 2). The incessant, often violent, protests by village communities forced the government to appoint the Kumaon Forest Grievances Committee to look into the local disaffection. The Committee examined over 5,000 witnesses from all parts of Kumaon in 1921 to make more than 30 recommendations. On the basis of these recommendations, the government passed the Van Panchayat Act of 1931. This act empowered village communities to create van panchayats and bring under their own control forest lands that were managed by the Revenue Department as Class I and Civil Forests.⁴

Nearly 3,000 van panchayats today formally control 35 percent of the hill forests in Kumaon. Of these, close to 1,700 exist in Almora alone (Agrawal, 1995: 51). The broad parameters that define the management practices of these institutions are laid down in the Van Panchayat Act. More specific content to the day-to-day management of community forests is the result of local action. Rural residents meet frequently, discuss the rules that will govern withdrawal of benefits from forests, and create monitoring, sanctioning, and arbitration devices to resolve the vast majority of management questions at the local level. They elect their leaders from within the community, select guards to enforce rules, fine rule breakers, manage finances, and often deploy earnings for the benefit of the community.⁵

This abbreviated history of the emergence of the van panchayats in Kumaon resonates with some critical issues in the social sciences. It shows-in contrast to much writing on local communities and peasants that treats its subjects as unwitting victims of a power-hungry centralizing state-that in the Kumaon hills, villagers significantly influenced government policies to reflect their subsistence needs. They organized themselves, resisted new state policies, and gained a measure of success in wresting back control over their forest resources. Of course, this is not to say that state-level actors do not seek greater control-as we will see, such objectives are at the heart of some recent modifications in the Van Panchayat Act. The actions of the villagers, rather, show that while macro-level political initiatives can significantly determine micro-level processes, the contours of such initiatives can also often be decisively shaped by organized social action undertaken by villagers.

Resources of the panchayats

The most significant products villagers traditionally harvest from their forests are fodder, fuelwood, animal bedding, organic manure, and construction timber. Figure 3.1 outlines the importance of forests in the hill agricultural and subsistence economy by tracing the links between forest products villagers harvest, and the kind of needs such products fill. It is obvious that forests are the cornerstone of subsistence in the hills, contributing critical inputs to each element of the subsistence economy-the household, agricultural fields, and livestock rearing. In addition, panchayat forests containing chir pine (Pinus roxburghii) also yield resin for turpentine, a commercially valuable product.

Figure 3.1: Forests in the hill subsistence economy

Subsistence products from the community forests are usually available to all residents of the villages in which the van panchayats are located. The cash revenues from the distribution of the forest products are used to monitor and guard the resource, and to meet operational expenses of the panchayat. In some cases, panchayats have also had sufficient surpluses to create communal goods for their villages such as school buildings, or common utensils used to cook food for the community during festive celebrations.

Key actors

The van panchayats are embedded in a web of social and administrative relationships. These relationships presume the patterns of influence laid down in the Van Panchayat Act of 1931, as amended in 1976. While the Act provides for support to the van panchayats from the Revenue and the Forest Departments to facilitate rule enforcement and the maintenance of vegetation in the forests, it grants them only limited authority to enforce rules. Indeed, over the last several decades, the modifications in the Act and the manner of its application have significantly reduced the independence of the villagers. In the quotidian interactions of different actors that influence the performance of the van panchayats, higher-level government officials, especially those in the Revenue Department, have emerged as pivotal in the success of panchayats. That they were assigned supervisory and enforcement powers played a crucial role in the process.

As Appendix 3.1 shows, the powers of the panchayats, especially their enforcement authority, had suffered a substantial decline by 1976 6 The overall framework of rules within which they could operate became far stricter. In addition, new restrictions on day-to-day activities meant they could fine rule breakers only with the consent of the rule-breaker, or once they secured the permission of higher-level government officials. For major disputes they were required either to move the judiciary, or rely on aid from the officials of the Revenue Department.

As a result, those van panchayats that have few local resources at their command have been plagued by rule infractions. Their elected officials, lacking independent means to pursue court cases, and the requisite influence to move the officials of the Revenue Department, have often been helpless to enforce the rules they created. Asked in a meeting to list the four most important problems facing their panchayats, 30 van panchayat chiefs listed problems related to inadequate supervision and local rule-breaking and monitoring 68 percent of the time. In contrast, problems related to low cash incomes of the panchayats were mentioned only 32 percent of the time. ⁷

At the same time, the officials of the Revenue Department who are supposed to help the panchayats must perform a host of other duties, including the maintenance of law and order, collection of taxes, and administration of various development projects. Most Revenue Department officials consider these duties to take priority over the tasks related to van panchayats. For many van panchayats, then, inadequate levels of enforcement and limited local resources are a major problem.

The case studies

Data on five van panchayats forms the basis for the ensuing discussion ⁸ All of them are located in the Dhauladevi Development Block of Almora District. They range in elevation from 1,100 to 2,000 meters; their forests lie between 1,400 and 2,100 meters; and they are all close to motorable roads, and thus more or less equally exposed to market forces. In all, about 25 villages are located in the watershed of the river Jataganga of which 11 possess their own van panchayats. The rest depend on illegal harvests from the forests of their neighbors, and forests owned by the Forest or the Revenue Department. This watershed represents the situation in most of the Kumaon region. Forest resources are scarce, and villages compete for subsistence benefits from forests.

While the selected van panchayats and their settlements are situated within the same ecological and administrative divisions, they differ significantly on their size, organization, age, and resource endowments. As Table 3.1 indicates, Pokhri and Tangnua are very small in area as well as number of households, and have formed their van panchayats only recently. Kotuli and Bhagartola are relatively large. Kana, although it has a large aggregate area, still possesses only a small number of households. In Tangnua, the population has increased in the last two decades, but the number of households has remained more or less stable.

Table 3.1: Basic statistics on the five Dhauladevi van panchayats

The small number of households carries some significant implications for the operations and budgets of the panchayats. The average annual number of meetings for Kana, Pokhri, and Tangnua lies between 2 and 4. For Kotuli and Bhagartola, it ranges between 8 and 12. Data from the meeting records of the first three panchayats indicates that they have also been relatively lax in creating rules to guide user behavior, and ineffective in enforcing the rules they have crafted. Thus, while the meeting records of Bhagartola and Kotuli contain lists of rule-breakers, the dates when guards detected rule infractions, and the amounts levied as fines, the minutes of meetings in Kana, Pokhri, and Tangnua are bereft of such details. By looking at the records one might conclude that no rules were ever broken in Kana, Pokhri, and Tangnua. Yet, in interviews and informal conversations, the members of these three panchayats invariably talked about limited resources and problems they faced in monitoring rule infractions. The absence of rule-breaking in formal records, then, is an indication of lax local supervision and enforcement (see Agrawal, 1994: 277).

In part, these differences among the five panchayats may simply indicate that because the first three are younger their officials as well as members need more experience: in working with government officials, in interacting with each other, and in forming and enforcing rules.

Such an explanation would be simple and attractive. Further examination, however, reveals its invalidity, at least for the selected panchayats. Records for meetings of the Bhagartola and Kotuli panchayats are available for analysis. These records reveal that they met regularly and often, and crafted a variety of rules right from birth. Their current organizational capacity certainly has developed over a period of time, but this cannot be taken to mean that time can be deployed as the explanatory variable for such capacity. A more favorable institutional and political-economic climate in the earlier period that helped establish the authority of the older van panchayats might still be playing a role in their continued survival and success. However, the current macroinstitutional environment has existed at least since 1976 and perhaps since Indian independence in 1950. It is difficult to accept that effects of a supportive environment have lingered on for 20 years or more, when everything around these village panchayats has changed. Further, it is important to understand how the activities and the processes within the panchayats relate to the macroenvironment rather than simply leaving the explanation toundefined historical changes.

A second difference that marks the first three panchayats is their meager budgets (Table 3.2). During the course of their existence, they have seldom been able to raise more than Rs. 750 a year to meet their expenses. Nor has their capacity to raise contributions from villagers increased over the period of their existence. Kotuli and Bhagartola, however, routinely raise between Rs. 2,000 and 4,000. Since all panchayats need money to hire a guard, or must be able to raise volunteer labor from members to substitute for the guard, the level of budget and contributions from members become significant elements in the successful functioning of the panchayats. Higher aggregate and per household contributions from member households increase the overall capacity of the panchayats to hire guards and enforce rules.

Table 3.2: Basic institutional information on van panchayats

To some extent, the ability of households to contribute to the van panchayats relates in a circular fashion to the condition and type of vegetation in the forest itself, making conclusive assertions hazardous. If villagers receive little benefit from the forest, they will have little incentive to contribute to protect the forest. In a vicious cycle, then, the degraded condition of forest will worsen still further, discouraging future contributions. Too much, however, can be made of such a connection. In a condition of generalized poverty in the hills, where few, if any, of the households can be viewed as prosperous or even reasonably well-off, why do we find "institutional robustness" (Ostrom, 1990) in some cases, and miss it in others?

In the case of the van panchayats, the above explanation is simply off the mark. The per capita forest area in the case of all the panchayats is low, but no lower for the first three panchayats than for the Bhagartola and Kotuli, which are more successful. In addition, more than a third of the residents in all five cases, including the less successful first three villages, initiated the process of forming the panchayats; most of the other villagers were willing to experiment. Villagers in all the five cases find significant proportions of their subsistence needs for fuelwood, fodder, and construction timber in the panchayat's forests. Even in the smaller villages, there have been some contributions to the panchayat coffers-all of these indicate that the problem is somewhat different from what the postulation of a "vicious cycle" suggests. It is related more to the inability of small groups of poor households to generate a surplus for protecting commonly owned and managed resources, rather than to their unwillingness .⁹

Implications of the study

The salient features of the situation can be summarized. A number of van panchayats compete with each other to protect, and subsist on, their scarce forest resources. While the per capita endowment of forest resources is similar across the panchayats, the absolute size of the panchayats varies, both in terms of area and households. The rural context remains one of high levels of dependence on forests and low levels of income. Smaller van panchayats have found relatively less success in protecting their resources. This last finding of the study is worth considering at greater length.

The success of the larger panchayats is reflected in the greater number of meetings held each year, the more rules crafted, the larger budgets, the higher levels of monitoring and enforcement, and even, a relatively more dense vegetation cover.

The figures for the "total tree biomass" in Table 3.3 provide some indication that the larger van panchayats have been more successful in protecting their forests.¹⁰

Table 3.3: Tree biomass and diversity in investigated cases

^a Indicates species that comprise more than 10 percent of the total number of the trees in the forest.

According to most writings that explore the relationship between collective action and group size, the probability of collective action becomes progressively bleak as group size increases. The data on five van panchayats indicate, however, that smaller groups may find it too arduous to create viable institutions that will persist over time to encourage collective actions. The larger van panchayats, on the other hand, are more successful in creating and maintaining processes that would organize their members and ensure their contribution to forest protection.

Two reasons can be advanced to explain the success of larger van panchayats. Each relates to protection of forests from unauthorized users and uses. To protect forests successfully in a context of generalized pressure on resources, communities need guards who will enforce rules. But guards who will monitor the condition of forests and prevent rule infringements cannot be hired without a minimum level of surplus. The smaller communities of poor peasants find it difficult to contribute even the relatively modest amounts that are necessary to hire a guard. As group size increases, it becomes easier to organize a surplus and commit it to monitoring and enforcement (Thomson, 1977; Agrawal, 1992).

Second, smaller communities also find it more difficult to prevent residents of other villages from coming and breaking rules related to forest use. In any dispute with residents of other villages they command fewer resources, such as voluntary labor or monetary contributions, that would enable persistence in imposing sanctions on rule-breakers.¹¹ The problem becomes especially acute in the absence of adequate support from the Revenue Department and other higher authorities. If a village community cannot raise sufficient resources to hire a guard to detect and prevent rule infractions, it is unlikely to possess the resources needed either to influence higher-level government officials, or to move the notoriously slow Indian judicial system to resolve disputes. Thus, on both counts-hiring a guard and influencing higher-level enforcement mechanisms-smaller communities are disadvantaged.

The finding that relatively larger groups found it easier to protect their forests successfully permits us also to engage the impressive theoretical literature on the relationship between group size and the probability of collective action. Before Mancur Olson's celebrated The Logic of Collective Action in 1965, Buchanan and Tullock (1962) inquired into the circumstances under which rational individuals would organize themselves to produce collective goods. According to them, as group size increases, the costs of decision-making externalities fall but the costs of coordination rise (1962: 63-64). As a result, medium-sized groups are most likely to organize themselves for collective action. Their discussion, however, assumes well defined and enforced property rights and focuses primarily on the internal dynamics of a group rather than on the results of competition between asymmetrically sized groups. In the situation we consider, it is precisely the delineation of property rights over forests, and their enforcement, that are issues of contention.

Olson's seminal work points to the importance of group size itself in determining whether collective action will be undertaken. According to him, "unless the number of individuals in a group is quite small, or unless there is coercion or some other special device to make individuals act in their common interest, rational self-interested individuals will not act to achieve their common or group interests" (1965: 2, emphasis in original). Focusing on the internal dynamics of groups by examining the motivations of individual members, Olson shows that groups will form to supply collective goods only under restricted conditions-and that these conditions are more likely to be met in small rather than large groups. As he puts it, "the larger the group the farther it will fall short of providing an optimal supply of a collective good" (1965: 48).

In the wake of Olson's work, a number of studies have focused on the impact of group size on collective action. Hardin (1982), for example, summarizes a number of earlier works (Buchanan, 1968; Chamberlin, 1974; Frohlich and Oppenheimer, 1970; Guttman, 1978; Hardin, 1971) to disentangle the effects of the nature of the good, the relation between the costs of collective action and benefits of the collective good per group member, and the likelihood of collective action. A large number of later studies have also tried to relate the possibility of collective action with group size, heterogeneity of member interests, reciprocity and interdependence, and marginal per capita returns from the provision of collective goods (Isaac, Walker, and Williams, 1994; Komorita, Parks, and Hulbert, 1992; Massey, 1994; Oliver and Marwell, 1985, 1988; Rapoport, Bornstein, and Erev, 1989; and Yamagishi and Cook, 1993). These studies have substantially enhanced our understanding of the impact of group size on collective action, and of collective action in general.

The example of the van panchayats in Kumaon, however, highlights some of the significant aspects of the relationship between group size and collective action that merit greater attention. The following discussion builds on existing studies of collective action by making two major points. It calls into focus the external dynamics of a group with other groups; and second, it makes a distinction between the formation of a group and achieving the objective for which the group was formed.

Most existing studies have focused only on the internal dynamics of the group-the relationship among group members. Following Olson's forceful focus on the rational, self-interested individual as the constituent unit of all groups, later studies have also focused primarily on the individual and his/her relation to collective action. In the process, they have ignored the impact of external relationships of a group with other groups. They have seldom considered how in a situation where different groups compete over resources, surely a widespread phenomenon, group size may be positively related to successful collective action. ¹²

The logic is devastatingly simple, almost "tautological," as Hardin (1982: 38) characterizes part of Olson's argument. Most villages in the hills already exist as groups. Individuals are born into these groups. The choice they face, then, is not whether to join a group. Rather, they must choose to not join a group of which they are already members by birth. Their calculus is not about the costs of joining; rather, it is about how expensive it would be to not join. In this situation, where individuals find it costly to leave the group, rather than to join, it should be obvious that larger groups would form more easily, and might even be more successful in protecting and managing their resources.

While villages already existed as informal groups, the Van Panchayat Act in 1931 lowered the cost of constituting the village as a formal-legal group protecting community forests. Government officials from the Revenue and Forest Departments encouraged local residents to create van panchayats. If villagers agreed, they could bring those areas of forests that were under the control of the Revenue Department under their own control. Further, owing to the scarcity of forest resources in the hills, villagers often are forced to harvest forest products in violation of existing rules protecting community forests. In the "drab everyday struggle" (Lenin 1902, rpt. 1976: 93) to protect their resources from others, then, it is not surprising that larger panchayats gain greater success than smaller ones.

Larger groups are more successful in two senses. A group that gains in size as more villagers participate in its activities is better able to raise more resources and expend a greater monitoring and enforcement effort. Two, if there are a number of different groups, some larger than others, the larger groups are more likely to be successful. ¹³ Both propositions in part rely on an added distinction between organizing collective action and success in achieving the objective of collective action.

Most studies on collective action have, by default, assumed that success in organizing a group (or collective action), and success in achieving the objective for which the group (or collective action) is organized, are one and the same thing. Under many conditions, the distinction is unnecessary-perhaps the reason why the obfuscation of this difference has survived for so long. Successfully organizing a march to protest abortion rights is synonymous with succeeding in the objective of organizing a march. But if the objective of the march is to overturn Roe v. Wade, success in organizing the collective action (march) is quite distinct from succeeding in its objective.

In the case of van panchayats, successfully forming a group to protect village forest resources is a very different proposition from succeeding in protecting these resources. And while success in forming a group may come easier to smaller groups, success in protecting resources is easier for larger groups. What we should note is that successful collective action is not just about forming groups, it is as much about being successful in achieving the objective for which the group was formed.

The above distinction is not the same as the difference between initiating and maintaining collective action. To take the example of the van panchayats again, organizing the panchayats is distinct from making sure they are meeting regularly, which in turn is distinct from protection of local forests. The difference between initiating and maintaining collective action necessarily depends upon a temporal disjunction. But the difference between organizing collective action and achieving the objective for which the action was undertaken may or may not possess a temporal dimension. Once this distinction is made, it is easy to see that while smaller groups may find it easier to organize themselves, it is larger groups that will find it easier, in comparison, to succeed in achieving the objective for which they were formed. The logic would also operate at the level of the individual. Villagers, discovering that smaller groups find it harder to protect forests from rule-breakers, may well calculate that it does not make much sense to continue to contribute to an unsuccessful panchayat.

If it is true that as group size increases, at least for some range the like-lihood of successful collective action will also increase, the natural question is: Would continued growth in size lead to lower likelihood of success at some point? It seems unlikely that groups could continue to grow indefinitely, even if continued growth is positively related to greater success in the achievement of objectives. While the studied cases have little to say about the effect of extremely large size on probability of success, ultimately the costs of coordination would increase sufficiently that they would outweigh benefits from increase in size (Buchanan and Tullock, 1962). The exact point at which this would take place, however, is a function of the context in which groups operate. In the context of the uneven topography of the Indian Himalayas, where natural factors such as limited availability of water, arable land, and forests constrain the growth of villages, the costs of coordination in existing villages are unlikely to become extremely high. Most villages comprise less than 200 households. One can then hypothesize the following: In small communities of poor users who use common-pool resources for subsistence, the likelihood of successful collective action to protect local resources increases as group size increases. It may, however, decline as group size becomes very large and creates extremely high costs of coordination.

The latter part of the hypothesis is based on the existing literature on collective action rather than on the data from the studied cases that provide only indirect indication of what would happen to the likelihood of collective action as group size becomes extremely large. It is because the costs of coordination would be very high for groups that are dispersed that smaller villages are unable to join each other to form larger van panchayats. For example, Kana, Pokhri, and Tangnua are more than six kilometers away from each other. They lack incentives to form a joint panchayat.

Conclusion

In conclusion, it may be useful to point to some practical relevance of the research. The findings reported here find significance from the most recent trends in Indian forest policy.¹⁴ In a number of statements issued between 1988 and 1995, the central Indian government and the governments of 15 Indian states have sought to increase local participation in the management of Indian forests (SPWD, 1992). These Joint Forest Management statements constitute a break from the colonial forest policy that had continued in most parts of India, with only a few minor changes, even after independence. Yet, the changes introduced today are far more timid than the British Van Panchayat Act of 1931 examined in this chapter (Appendix 3.1). Most state policy statements allow local populations only a partial share in the benefits from protecting forests and do not permit them a voice in crafting the rules whereby the forests would be managed (SPWD, 1992; GOI, 1992, 1993). Without adequate support from enforcement officials, and without local mechanisms to ensure adequate protection-two provisions that are mostly absent from the pronouncements of the Joint Forest Management policies-prospects of success for the new policy may remain bleak.¹⁵

In addition, the research indicates that where groups are very small and compete for a share in local resources, their performance in protecting resources may improve if government policies create institutional incentives for smaller groups to join together. The attempts of very small groups of the poor to protect local resources on their own may founder because of limited ability to raise a surplus to enable effective local monitoring and enforcement. Finally, it may be kept in mind that if small groups are also highly dispersed, the external conditions might make it very difficult to create institutions through which they would coordinate their resource management and protection activities.

The relevance of the research for India is evident in the context of a declining forest base and changing forest policies. The research is, however, also significant in the context of the emerging international debate over the criticality of local communities and indigenous institutions in managing forests. The example of the forest communities in the Indian Himalayas suggests that autonomy to local communities must be supplemented by arrangements that will help protect local resources by creating user groups that are not too small, and will encourage dispute resolution within the same community, and among users from different communities.

References

Agrawal, Arun. 1992. "Risks, Resources, and Politics: Studies in Institutions and Resource Use from India." Ph.D. diss., Duke University.

1994. "Rules, Rule Making, and Rule Breaking: Examining the Fit between Rule Systems and Resource Use." In Rules, Games, and Common-Pool Resources, ed. Elinor Ostrom, Roy Gardner, and James Walker, 267-82. Ann Arbor: University of Michigan Press.

1995. "Population Pressure = Forest Degradation: An Oversimplistic Equation?" Unasylva 46(2):50-58.

Anderson, D., and R. Grove, eds. 1987. Conservation in Africa: People, Policies and Practice. Cambridge: Cambridge University Press.

Ballabh, Vishwa, and Katar Singh. 1988. "Van (Forest) Panchayats in Uttar Pradesh Hills: A Critical Analysis." Research Paper no. 2. Anand, India: Institute of Rural Management.

Bates, Robert. 1981. Markets and States in Tropical Africa. Berkeley: University of California Press.

1989. Beyond the Miracle of the Market. Cambridge: Cambridge University Press.

Buchanan, James M. 1968. The Demand and Supply of Public Goods. Chicago: Rand McNally.

Buchanan, James M., and Gordon Tullock. 1962. The Calculus of Consent. Ann Arbor: University of Michigan Press.

Chamberlin, John. 1974. "Provision of Collective Goods as a Function of Group Size." American Political Science Review 68(2):707-16.

Chambers, Robert. 1983. Rural Development: Putting the Last First. London: Longman.

Colburn, Forrest D., ed. 1989. Everyday Forms of Peasant Resistance. Armonk, NY: M. E. Sharpe, Inc.

Guttman, Joel M. 1978. "Understanding Collective Action: Matching Behavior." American Economic Review 68(2):251-55.

Hardin, Russell. 1971. "Collective Action as an Agreeable N-Prisoners' Dilemma." Behavioral Science 16(5):472-81.

1982. Collective Action. Baltimore, MD: Johns Hopkins University Press.

Hecht, S., and A. Cockburn. 1990. The Fate of the Forest. New York: Harper Perennial.

Homer-Dixon, T. 1991. "On the Threshold: Environmental Changes as Cause of Acute Conflict." International Security 16(2):76-116.

Isaac, R. Mark, James M. Walker, and Arlington W. Williams. 1994. "Group Size and the Voluntary Provision of Public Goods: Experimental Evidence Utilizing Large Groups." Journal of Public Economics 54(1):1-36.

KFGC (Kumaon Forest Grievances Committee) 1922. "Report of the Forest Grievances Committee for Kumaon." Mimeo.

Komorita, S. S., C. D. Parks, and L. G. Hulbert. 1992. "Reciprocity and the Induction of Cooperation in Social Dilemmas." Journal of Personality and Social Psychology 62(4):607-17.

Korten, D., ed. 1986. Community Management: Asian Experiences and Perspectives. West Hartford, CT: Kumarian.

Lenin, V I. 1902. What Is To Be Done? (rpt. 1976). Peking: Foreign Languages Press.

Marglin, F., and S. Marglin, eds. 1990. Dominating Knowledge: Development, Culture and Resistance. Oxford: Clarendon Press.

Massey, Rachel Ida. 1994. "Impediments to Collective Action in a Small Community." Politics and Society 22(3):421-34.

Oliver, Pamela, and Gerald Marwell. 1985. "A Theory of the Critical Mass, 1: Interdependence, Group Heterogeneity, and the Production of Collective Action." American Journal of Sociology 91(3):522-56.

Oliver, Pamela, and Gerald Marwell. 1988. "The Paradox of Group Size in Collective Action: A Theory of the Critical Mass, II." American Sociological Review 53:1-8.

Olson, Mancur. 1965. The Logic of Collective Action. Cambridge: Harvard University Press.

Ostrom, Elinor. 1990. Governing the Commons: The Evolution of Institutions for Collective Action. New York: Cambridge University Press.

Ostrom, Elinor, Larry Schroeder, and Susan Wynne. 1993. Institutional Incentives and Sustainable Development. Boulder, CO: Westview Press.

Pant, G.B. 1922. The Forest Problem in Kumaon. Nainital: Gyanodaya Prakashan.

Rapoport, Amnon, Gary Bornstein, and Ido Erev. 1989. "Intergroup Competition for Public Goods: Effects of Unequal Resources and Relative Group Size." Journal of Personality and Social Psychology 56(5):748-56.

Redford, Kent, and Steven Sanderson. 1992. "The Brief Barren Marriage of Biodiversity and Sustainability." Bulletin of the Ecological Society of America 73(1):36-39.

Repetto, R., and M. Gillis, eds. 1988. Public Policies and the Misuse of Forest Resources. Cambridge: Cambridge University Press.

Richards, P. 1985. Indigenous Agricultural Revolutions. Boulder, CO: Westview Press.

Scheper-Hughes, N. 1992. Death without Weeping: The Violence of Everyday Life in Brazil. Berkeley: University of California Press.

Scott, James. 1976. The Moral Economy of the Peasant: Rebellion and Subsistence in Southeast Asia. New Haven, CT: Yale University Press.

1985. Weapons of the Weak. New Haven, CT: Yale University Press.

Shiva, Vandana. 1988. Staying Alive. New Delhi: Kali for Women.

Somanathan, E. 1989. "Public Forests and Private Interests." New Delhi: Indian Statistical Institute. Mimeo.

SPWD (Society for the Promotion of Wasteland Development). 1992. Joint Forest Management: Regulations update, 1992. New Delhi: SPWD.

Thomson, J. T. 1977. "Ecological Deterioration: Local Level Rule Making and Enforcement Problems in Niger." In Desertification: Environmental Degradation in and around Arid Lands, ed. M. H. Glantz, 57-79. Boulder, CO: Westview Press.

Trainer, T. 1985. Abandon Affluence! London: Zed.

Uphoff, Norman, ed. 1986. Local Institutional Development: An Analytical Sourcebook with Cases. West Hartford, CT: Kumarian.

Wade, Robert. 1990. Governing the Market. Princeton, NJ: Princeton University Press.

1994. Village Republics: Economic Conditions for Collective Action in South India. San Francisco, CA: ICS Press.

Westing, A., ed. 1986. Global Resources and Environmental Conflict. New York: Oxford.

Wolf, C. 1988. Markets or Governments. Cambridge, MA: MIT Press.

Yamagishi, Toshio, and Karen S. Cook. 1993. "Generalized Exchange and Social Dilemmas." Social Psychology Quarterly 56(4):235-48.

Appendix 3.1 - Changes in the Van Panchayat Act between 1931 and 1976

Subject	1931	1976
Formation/ Dissolution	1. Two or more residents could propose the formation of the van panchayat for a village. 2. The Deputy Commissioner could dissolve a panchayat in case of repeated mismanagement or rule infractions.	Rule 2 remains the same. Modifications: 1. One third of the villagers must propose the formation of the van panchayat.
Membership	1. At least three, and at most nine, members elected to the van panchayat by villagers. 2. Panches select their leader as Sarpanch. 3. Panches could force resignation of individual members by a majority-the empty position could be filled from among right-holders by a majority decision of the panches. 4. All village residents, and others who possessed rights in the forest, could be right-holders in the panchayat forest.	Rules 2, 3, and 4 remain the same. Modifications: 1. Five to nine members to be elected to the van panchayat. 2. The Deputy Commissioner could nominate one member to the panchayat. 3. The Sarpanch could be removed from office by one third of the members, provided this step is approved by two-thirds of the members in a subsequent meeting.
Rules Regarding Resin Extraction	1. The Forest Department to be responsible for harvesting resin from chir pine trees. 2. Profits to be shared between Forest Department and the panchayat in proportions to be determined by the Forest Conservator. 3. Panchayat could harvest resin as long as it is in accordance with rules laid down by the Forest Department; and the resin sold to either the Forest Department or registered buyers. 4. Panchayat members could harvest resin for domestic use.	Rules 1, 3, and 4 remain the same. Further Restrictions: a. See modifications c, d, and e under the subject "Allocation of Income."
Rules Laid Down by Government	1. Panchayat forest land could not be sold, mortgaged, or subdivided. 2. The Products and proceeds from the sale of products of the panchayat forest to be used for the benefit of the community. 3. Panchayat to protect the forest and its trees. (But no explicit restriction on commercial sale of trees or timber.) 4. Panchayat to prevent villagers from cultivating the panchayat forest land. 5. Panchayat to demarcate the forest area. 6. The panchayat to maintain minutes of meetings, records of accounts, and make decisions in regular meetings. 7. Panchayat to follow the instructions of higher revenue officials. 8. Quorum required two-thirds of the members of the committee to be present. 9. All decisions to be made by simple majority.	Further Restrictions: a. All decisions of the panchayat to be made by two-thirds vote. b. Panchayat to meet at least once every three months; proceedings of the meeting to be recorded and copy submitted to the deputy commissioner. c. All extraction of timber beyond one tree requires permission from the Deputy Commissioner, Divisional Forest Officer (DFO), and the Conservator of Forests (CF). Any sales of forest produce must be in accordance to the working plans prepared for the van panchayat by the Forest Department. d. For commercial sale or auction of forest products (fodder, grass, minor forest products, firewood, timber), the permission of the DFO must be obtained. If the value of the auctioned products exceeds Rs. 5,000, the DFO must be present. All auctions above Rs. 5000 must be approved by the Conservator of Forests. e. The panchayat must prepare annual budgets and submit an annual report to the DFO each year. f. Special officers appointed to supervise van panchayats must oversee at least a third of panchayats each year. g. Van panchayat accounts could be audited.
Rights and Powers of Panchayats	In general, similar to forest officials: 1. Fine rule-breakers up to Rs. 5. 2. For offenses where the fine should be higher, the panchayat could file court cases against rule-breakers. 3. Levy fees from users for fodder, grazing, fuelwood, or construction stones. 4. Regulate grazing in the panchayat forest and impound animals that are found in the forest in contravention of rules. 5. Confiscate cutting implements used in contravention of panchayat rules. 6. Restrict/suspend rights of users who break rules regularly. 7. Appoint guards to monitor and enforce rules.	In general, similar to forest officials: Rules 3, 4, 5, and 6 remain the same. Further Restrictions: a. All appointments by the van panchayat require approval of the Deputy Commissioner. b. At least 20 percent of the area of the van panchayat to be set aside from grazing; could lease land for commercial use. c. Could compound fines on individual rule-breakers up to a limit of Rs. 50 with their permission, and up to Rs. 500 with the permission of the Deputy Commissioner; and to file court cases against rule-breakers. d. Could grant no more than one tree to a right-holder-written consent of more than half the panches, and stamp of Sarpanch necessary.
Rule Enforcement	All fines imposed by the panchayat were treated as government dues and recoverable using similar procedures.	Same as before.
Elections	Panchayat officals elected for three years. New elections to be held every three years.	Panchayat officials elected for five years. New elections to be held every five years.
Allocation of Income	1. All income from sale of forest products to right-holders as assigned to the van panchayat. 2. All income from sale of resin to be allocated in accordance with proportions determined by the Conservator of Forests (in practice it went to vanpanchayat). 3. Income from sale of forest products minor forest produce) to non- (such as timber, resin, right-holders was assigned to the van panchayat.	Rule 1 remains the same. Modifications: a. Forest Department to deduct 10 percent from all gross revenues of the van panchayat as its share to meet administrative expenses. b. Net income from commercial sale and auctions to be deposited in a Panchayat Forest Fund, managed by the Deputy Commissioner. c. Twenty percent o f the net income allocated to District Council to meet development costs. d. Forty percent of the net income allocated to the Forest Department to maintain and develop panchayat forests. e. Remaining 40 percent of net income allocated to panchayat to be spent on works of public utility as approved by the Deputy Commissioner.

1. The causes for the emphasis on local institutions may lie in the demonstrated deficiencies of state-directed development and the inability of markets to promote sustainable use of common resources. A large literature documents the vigorous debate on the merits and problems of pursuing development and conservation goals through state- or market-led policies. For useful introductions see Bates (1981, 1989), Repetto and Gillis (1988), Wade (1990), Wolf (1988). For critiques of both the market and the state see Shiva (1988), Escobar (1991, 1992), and Marglin and Marglin (1990).

2. For a discussion of the relationship between renewable resource scarcity and social tensions see Gleick (1989), Homer-Dixon (1991), and Westing (1986).

3. See Escobar (1991, 1992), Redford and Sanderson (1992), Scheper-Hughes (1992), Scott (1985), and Trainer (1985) for some critiques of market- and state-led development and conservation policies that ignore the interests of subaltern groups. The theoretical literature on the necessity of addressing local interplays of power and resistance often finds its inspiration in the works of Michel Foucault (see especially 1978, 1991a, 1991b).

4. According to Somanathan (1989), the Act only formalized the control many hill communities had exercised over their forests before the arrival of the British. Their informal institutions were called lattha panchayats. Lattha means a big stick, and the name evocatively denotes the power the local community held over its members.

5 Thus, they seem to meet many of the design principles that are characteristic of successful community institutions as discussed by Ostrom (1990).

6. Ironically, it is the Indian state after independence that reduced the local authority of the panchayats even more than the colonial British state. Forestry and Revenue Department officials both felt that the Van Panchayat Act devolved too much authority to the villagers, and that the villagers had not been able to manage their forests well. In support of their arguments, they pointed to cases where, they argued, locally powerful individuals had engaged in large-scale felling and had been abetted in some cases by panchayat officials. Their arguments led to the increased restrictions through amendments introduced to the Van Panchayat Act in 1976.

7. The 30 chiefs of panchayats listed a total of 97 problems. Of these, 31(32 percent) related to the low income of their panchayat, 22 (23 percent) to inadequate support from higher-level government officials, and 44 (4S percent) to local-level rule infringements and problems in monitoring and enforcement.

8. The selected panchayats were chosen randomly out of the 11 villages that possess their own community forests and panchayats in the Dhauladevi Development Block

9 It should be obvious that even if the problem relates to lack of incentive to contribute in the smaller communities, the larger argument in this chapter holds-smaller groups find it more difficult to successfully organize collective action.

10. Since the Kana, Pokhri, and Tangnua van panchayats have formed recently, the condition of the vegetation in their forests, unlike the cases of Kotuli and Bhagartola, cannot entirely be attributed to the manner in which the panchayat has functioned. But the relatively lax enforcement of rules in the three panchayats implies there will be little improvement in the condition of the. forest.

11. Voluntary labor and monetary contributions may both be necessary to discourage local rule infractions and resolve disagreements by arbitration or civil suits.

12. Rapoport, Bornstein, and Erev (1989) do consider how differences in group size may affect the probability of collective action when such groups are competing with each other. On the basis of their experimental results, they conclude that group size does not have any effect on provision of collective action. However, the group size for their experiments varies between 3 and 5. It seems hasty to draw the conclusion that group size has no impact on the probability of success, on the basis of such minimal variance in group size.

13. In the second sense, the proposition has also found a defence from such theorists as Dahl and Tufte in their discussion of "system capacity" (1973: 20-21).

14. A number of governments in South Asia, including Nepal and Bhutan, are attempting to craft co-management programs with village communities for more effective forest use and protection.

15. The Indian central and state governments, in formulating the new forest policy statements, seem, thus, to have ignored the lessons that the history of the van panchayats offers.