Failing to manage CPRs

Back to contents - Previous file - Next file

One thing that comes out of the above presentation is that sharing arrangements devised for regulating access to a scarce resource in a more or less equitable manner (at least within the local community) are more likely to be observed when the physical and technical attributes of the resource tend to create many opportunities for conflict. This is particularly evident in the case of surface irrigation systems since they pose an inherent conflict between farmers upstream and those downstream: if upstreamers take too much water, downstreamers may end up with agricultural failure. The case of inland fishermen using weirs crossing over the entire width of a river is basically similar in that it also creates opportunities for serious externalities now imposed by downstreamers on upstreamers. Conflicts are likewise very probable when resource users have to compete for scarce locations, whether they be fishing spots, pastures, or woodlands, in which cases assignment problems unavoidably arise. In all these circumstances, rules are badly needed not only to reduce transaction costs but also to avoid repeated outbursts of sheer violence. In a study based on thirty case-studies of fisheries, Schlager has thus found that the existence of assignment externalities is significantly related to whether or not fishermen have adopted systematic procedures for fishing operations (Schlager, 1990). In the case of irrigation in India, a high correlation has been found between the degree of water scarcity on the one hand and the level of activity of informal water users' organizations and the incidence of rules for tank-water rotation on the other hand (Easter and Palanisami, 1986).

What needs to be recalled at this stage is that rules solving assignment problems do not, by themselves, help to conserve resources: as a matter of fact, management of a resource implies restraint in its use and there is nothing in sharing rules to ensure that users control the extent of pressure on the natural environment or the mode of exploiting it. It could be argued that exclusionary arrangements directed against outsiders have the (unintended) effect of limiting the pressure brought to bear on a resource, thereby helping to maintain its viability. This, however, misses the main point in so far as tightening the access to a resource by outsiders only postpones the problem if this is not accompanied by measures of restraint impinging on the insiders themselves. If the latter kind of measures are not enforced, the threat on the resource posed by outsiders is likely to be just replaced by the increasing pressure exercised by the insiders, because they grow in numbers and/or they are more and more willing to respond to market opportunities. Admittedly, it may be the case that a scarce resource is not liable to be depleted as a result of the users' behaviour, in which circumstances distributive measures are really the only ones required. Beachseine fishing (a technique that has no destructive potential) is a good illustration of such a possibility.

In the first part of this section, we have suggested that traditional societies may possibly fail to adopt rules other than those answering the rather immediate need to regulate access to the local CPR, and this failure may possibly take place even when vital resources are threatened with depletion or degradation. In other words, such societies do not necessarily take steps to conserve resources although the latter are obviously finite. To understand why it is so, the following fact must be borne in mind: rural people, particularly so in hunter-gatherer societies, are traditionally inclined to view the flow product of a resource system as a given. When an excessive number of claims are laid to this given flow product—so that the number of resource units harvested per unit of time by individual claimants tends to diminish—the need is felt to sort them out and, as we have seen, this is always done by referring to the basic principle according to which priority access must be reserved for members of the local community. What is therefore important to realize is that rural dwellers are not used to see the flow of harvested natural products as yielded by a resource system or stock on which they may themselves impinge through their own harvesting behaviour. If they do not adopt conservation measures, it is because they do not perceive the relationship that exists between the stock and the flow of a resource nor the causal link between their own actions and the level of this stock. Consequently, they are not aware that they can increase the availability of appropriable resource units (to use Ostrom's language) in the future by harvesting in a restrained and selective fashion in the present. In other words, they are not aware of the existence of the sustainable yield curve since they do not clearly perceive the fact that today's choices may constrain the set of future choice possibilities. This comes down to saying that the concept of resource management is alien to them.

How is it that members of traditional societies consider natural products as a given on which they have no hold? The answer lies in the fact that they share a magical pre-rationalist view of the world. They thus conceive of surrounding resources as being provided by some supernatural agencies (deities or cosmic forces) which are in charge of catering to human needs. If at all resources have to be conserved, it is a task for the gods, not for the humans. Humans must only be careful to please the gods so that the latter do not stop caring for them by supplying regular sources of livelihood for day-to-day subsistence. Thus, for example, in the minds of the Mbuti pygmy tribe of Equatorial Africa, 'it is not the hunters who catch the game, it is the Forest which gives the game to the hunters so that they can catch it and subsist and reproduce by consuming it'. In this worldview, the forest is seen as 'the collection of all the animate and inanimate beings that live in it'. It is a personalized deity which is figured out fat once as the father, the mother, the friend, and even the lover' of the Mbuti (Godelier, 1974: 340-1).

Among North American Indians, there was a 'probably universal' belief in active, watchful, and potentially vengeful animal spirits. 'The Indians had to use them carefully and propitiate them for their sacrifice if they were to rely on their continued abundance. Animism was the way in which the Indians, like most hunter-gatherers, expressed their awareness of the fact that their lives and those of their food resources were ecologically intertwined' (McEvoy, 1986: 31).

This kind of belief is still alive today, at least among the older generations: to take just one example, the Bakusu in Zaire remain convinced that animals and certain types of food are released or supplied to the living by the ancestors who act as the custodians of the food resources upon which man's livelihood crucially depends (personal communication of Mr Masaki). Likewise, artisanal fishermen of Kerala (India) view 'Mother ocean as a life-giving system rather than a hunting ground' (Kurien, 1991: 3).

In West African inland fisheries, local water spirits are thought to be the sources which activate or withdraw fish. In Nepal, according to Messerschmidt, there is a nearly universal belief that certain resources have supernatural characteristics. Thus, various gods and godlings 'are commonly believed to dwell in or be otherwise closely associated with water sources, forests, pasture sites, and other resources or natural objects' (Messerschmidt, 1986: 468). At a village called Muktinath in the Upper Kali Gandaki, for example, 'the springs that serve as a principal source of irrigation and drinking water are elaborately enshrined near the source within a sacred forest of considerable antiquity. This water and forest place is, furthermore, considered a source of spiritual power and authority.' (ibid.).

The idea that natural resources are under the responsibility, or are even the concrete manifestation, of supernatural agents which continuously and actively intervene in human affairs so as to allow men to survive is also a typical feature of traditional agricultural societies. Thus, it is in the hope that the bush will recover from a period of (shifting) cultivation that Mende tribesmen in the Gola Forest (Sierra Leone), when they are about to fell trees to make a farm, or burn dried vegetation, invoke the patient understanding of the ancestors and spirits of the land for the necessary damage they must inflict on the bush (Richards, forthcoming: 150). Speaking about the systems of belief among traditional peasantries of Europe, Badouin has noted that land is conceived of as an active agent that 'carries the crops much in the same way as it serves as a basis for all sorts of buildings' (Bedouin, 1971: 25, our translation). Another French rural sociologist, Barthez, has similarly remarked that the peasant traditionally perceives himself as a collaborator or a partner of nature. The isomorphism between man and land can be taken very far: like man, land must be fed to be capable of producing (Barthez, 1982: 32-5; see also Bourdieu, 1980: 197-9). As aptly summarized by Bourdieu: 'Convinced as he is that he does not have any effective means of influencing his own future and the level of his future production, the peasant does feel responsible only for the actions undertaken, not for their failure or success which depend on natural and supernatural powers' (Bourdieu, 1977: 39, our translation).

Now, the point has often been made that, in so far as they personify and deify nature in the above-described way, people in traditional rural societies are naturally inclined to show a deep and abiding respect for their environment, which starkly contrasts with the Promethean attitude of people in modern, rationalist societies. Furthermore, since the cult of ancestors—which is tightly enmeshed in, or associated with, the cult of nature—manifests an unremitting concern for the continuity of generations, past, present, and future, they are naturally predisposed to take account of the interests of succeeding generations while making their decisions (their discount rate of the future is close to zero). For example, Guha is deeply impressed by the attitude of deep respect displayed by villagers in Uttarakhand (Uttar Pradesh, India) towards their local forests, as well as by their concern for future generations, two intermingled attitudes to which he tends to ascribe much of their presumed success in managing forest resources. According to him, indeed, many of the myths and legends emphasize the deep sense of identity of tribal people with the forest: 'In tribal forest areas, not only did the forests have a tremendous influence in moulding religious and spiritual life, but the inhabitants exhibited a deep love of vegetation, often acting entirely from a sense of responsibility towards future generations by planting species whose span of maturity exceeded a human lifetime' (Guha, 1985: 1940). The same kind of analysis is made by Messerschmidt when he writes:

Belief systems in which nature is sanctified often function to hold resource abuse in check through some combination of respect and fear that disturbance or neglect of the supernatural may cause more harm than good to the resource and to the people associated with it. At various levels, these beliefs serve to remind people of the miraculous (hence fragile) nature of the resource and of the people's own responsibilities to manage it for the sustained public good. (Messerschmidt, 1986: 469)

It is no doubt true that people in traditional societies are acutely aware of the immense fragility of the ecological systems in which they live, and that they are consequently very much willing to adopt the kind of behaviour required to maintain the long-term viability of these systems. None the less, this is not sufficient to make them effective resource managers. In effect, it is not because people are both respectful of nature and concerned with the well-being of future generations that they are able to view resource systems as partly man-made and to have a clear analytical perception of the relationship between the intensity of exploitation and the level of yields. Thus, Richards notes that, although living from the forest, Mende farmers of Sierra Leone 'do not see themselves as standing over it, either to exploit or to conserve it. In local eyes the relationship is the other way round—the community is under the protection of the forest' (Richards, forthcoming: 139). In actual fact, the adaptive attitude of people vis-à-vis nature in traditional societies may lead them, in a perfectly consistent manner, even to accept the disappearance of a resource on the simple ground that this is the will of some supernatural force. Worse still, they may actually behave in a resource-destructive way if this type of behaviour accords well with their idea as to how resource systems reproduce themselves. It is useful, at this stage, to illustrate these possibilities and to gain precise understanding of the underlying ecological conceptions.

To begin with, we may refer to a fascinating article by Carrier (1987) in which the author attempts to elucidate the ecological practices and beliefs of the Ponams fishermen of Papua New Guinea. According to Carrier, the Ponams believed that 'substantial changes in the marine environment such as the extinction of species were not necessarily undesirable, though they may be inconvenient in practice' (Carrier, 1987: 153). Thus, even though sea turtles were extremely important in Ponam life (they played a central role in weddings and other feasts and were enjoyed as a regular part of the diet), Ponams actually objected to the government's plan aimed at the conservation of this resource because they felt that the cost of the restraint required from them could not be possibly balanced by any real benefit. After all, 'if sea turtles were going to die out, as the government said, it would be because God wanted them to. A halt on hunting sea turtles would not change God's mind but would be very costly to the Ponams' (ibid. 154). Interestingly, Ponams islanders were quite aware that areas could be overfished: they actually said that 'this had happened on some of the small islands off the northeast Manus coast' where rapacious fishermen 'were said to have fished their lagoons so much that it became much more difficult to catch fish than before'. Yet they refused to ascribe the declining catches to a decrease in fish population and they instead held that catches fell because fish became wary. 'Fish, many Ponams explained, are like people: confronted with danger, they move.' Thus, at this pragmatic level of knowledge, 'fish themselves are the agents of ecological change and the cause of decreased catches'. Islanders used this explanation to account for the presence of Japanese and Korean fishing vessels in their waters: 'the vessels were following fish that had fled waters closer to Japan and Korea' (ibid. 154-5). To take a last example, although Ponams could agree, as a matter of principle, that too much shell gathering would lead to failure of the beds—a failure which they themselves experienced around the middle of this century—they typically located the ultimate cause of shell depletion in the supernatural sphere: growing disrespect among islanders for the group that controlled the production of shell money led God or perhaps the ancestors to move the beds elsewhere (ibid. 156).

One characteristic feature of the ecological model adhered to by the islanders of Papua New Guinea is therefore that men are not the crucial agents of environmental change even though they may possibly be instrumental in bringing it about. The latter part of this proposition explains the fact noted by Carrier that some conceptions of the environment held by these people had functional aspects which meshed with Western understanding. Yet the former part makes it plainly clear that 'the fundamental way they viewed the environment was alien to Western thought' (Carrier, 1987: 156). The crux of the matter is that 'they did not organize the facts that they knew about marine life into the same models that Westerners have produced': as a result, their practical conclusions were derived from a completely different 'cultural superstructure' (ibid. 150, 156). Contrary to the Western view which tends to see human action as the principal agency of ecological disruption, their view was based on the belief that external agencies are at the heart of environmental phenomena. To quote Carrier again: 'Contrary to the materialist view of a self-ordering and self-sustaining natural environment, Ponams held that the ordering and operation of the world was dictated by God. God and fish themselves were the critical agents of change in natural resources' (ibid. 153). Revealingly, when they converted to Catholicism in the 1920s, they interpreted their new religion in a syncretic way in which their traditional core belief that God (the supernatural powers) acts continually in the world could be maintained. They also ensured the continuity between the old and new world-views by seeing God as an ultimate ancestor (ibid.).

Let us now turn our attention to a second illustration taken from another original study, this time by Brightman ( 1987), concerning the Boreal forest Algonquians. What makes the case of these Indian hunting tribes particularly worth exploring is the existence of a considerable body of evidence pointing to the lack of conservation and intentional management among them as well as to 'a proclivity to kill animals indiscriminately in numbers beyond what were needed for exchange or domestic use' (Brightman, 1987: 123). For instance, by the late 1700s or early 1800s, 'the Crees in the

Northern Department had overhunted beaver and big game to the point that reliance on hare and fish, migration south onto the plains, and starvation became humiliating and tragic consequences' (ibid. 125). Still more puzzling are reports about the apparently purposeless slaughters of certain species such as the deer or the beaver, and reports about non-selective killings of all animals, great and small, male and female, belonging to given species.

How can one account not only for the absence of resource management but also for the aforementioned acts of indiscriminate and non-selective hunting among American Indians? The first thing to note is that, like islanders in Papua New Guinea, Indian hunters of North America did not understand their own role as determinants of ecological change in general and of resource depletion in particular. Consequently, they were unaware that animals can be managed, that is they were unaware of their ability to influence the availability of animals in the future by harvesting in a restrained and selective fashion in the present: 'the conviction that hunters can reliably manipulate the quantities of animals subsequently available to them cannot . . . be ascribed to prehistoric and early-contact-period Algonquians' (Brightman, 1987: 132). Again like with Ponams, although Algonquians—or at least some of them - could recognize their proximate role as agents in the destruction of animals (most notably, the beaver), they would never admit final responsibility, the ultimate cause of any ecological change being necessarily located in decisions of supernatural beings: if the beaver disappears, it is ultimately because the creator and the trickster-transformer intended to destroy it (ibid. 133). Yet the idea that hunting pressure could reduce species populations in the long term and on a large sea/e was all the more absent in Algonquian culture that game animals killed by hunters were thought to spontaneously regenerate after death or reincarnate as foetal animals (ibid. 131). Thus, Cree trappers could well imagine that 'an adult, trapped animal was "the same one" that had been killed the previous winter', and modern Crees followed ritual procedures in which animal bones and blood were disposed in such a way as to prefigure and influence animal regeneration and reincarnation (ibid.). This last belief points to a central element of the Algonquian culture, namely that their environment was one of primordial abundance. This ecological optimism was actually reinforced by the feeling that 'game could not be destroyed but only temporarily displaced' (in the same way as depleted shell beds were thought by Ponams to have been relocated elsewhere), and that 'animals were "given" to hunters when they were needed' (ibid. 132).

The above ecological model can no doubt account for the absence of effective management among Boreal forest Indians, yet it is probably not sufficient to explain the kind of overkilling that has been reported in the literature dealing with the old hunting practices of these people. Acts of indiscriminate killing, at least among certain American Indian tribes, may be understood by reference to a specific belief concerning the mode of operation of their resource systems. This belief—widespread among the Iroquoians in the Great Lakes area—is the idea that animals are able to communicate with each other and to share their experiences. More precisely, Iroquoians believed that they had to kill all the animals which they encountered because, if they did not do it, the animals would go and tell the others how they had been hunted. The result, the Iroquoians thought, was that they would not find any more animals in times of want (Brightman, 1987: 125).

In another study, Berkes has described the life of Cree Indian fishermen in the James Bay. He reached the same conclusion as Brightman and Carrier, namely that those fishermen believed, at least until recently, that 'fish is an inexhaustible resource, and that the numbers available are independent of the size of the previous harvest' (Berkes, 1987: 84). Berkes's contribution is especially interesting because he deals with the ecological attitudes and practices of a presently living traditional group. It is therefore worth quoting him at some length:

Cree fishermen use the most effective gear available to them and the mesh sizes that give the highest return in a given situation, . . . the Cree are not conservationists because, among other things, they do not know and do not seem to care about animal population numbers, they never throw back undersized fish, and they concentrate their efforts on vulnerable aggregations of fish . . . Cree practices violate nearly every conservation-oriented, indirect-effort control measure in the repertory of contemporary scientific fisheries management. Furthermore, the Cree abhor catch-and-release practices currently fashionable in North American sport fishery management, disapprove of population surveys, and oppose tagging experiments. . . many groups of northern Indians and . . . Canadian Inuit (Eskimo) think that animals will go away if you count them or otherwise show disrespect. In many of these groups, as with the Cree, it is the animals who are considered to be making the decisions; hunters are passive. Any management system claiming to maximize productivity by manipulating the animals is considered arrogant. Thus, not only does the Cree ideology differ from Western scientific management, but it also opposes it. (Berkes, 1987: 85-6, emphasis added)

Much the same can be said of the observation made by Martin regarding the cod fishermen of Fermeuse, Newfoundland (Canada):

Until very recently, Fermeuse fishermen have seldom thought of anyone's fishing activities, perhaps least of all their own, as having an appreciable effect upon fish populations. 'Queer things' happen, as in years when fish does not appear, but this is explained in terms of natural factors (e.g. a change in water temperature) over which man has no control. (Martin, 1979: 2X5; cited from Schlager, 1994: 247, emphasis added)

In the same vein, Wilson and Thompson partly trace present overgrazing in Mexican pastoral ejidos to the ejidatorios' limited technical understanding 'regarding the complex interdependence of individual grazing decisions and the impact these choices have on the range resource and, ultimately, on livestock productivity and human welfare'. In particular, 'local understanding of elementary soil-water-plant-animal relationships is rudimentary' (Wilson and Thompson, 1993: 314).

Strict access rules and conservation: the case of type-III societies

A few illustrations
In the previous sections, we have referred to traditional societies in which conservation rules are conspicuously absent either because resources are abundant (an objective factor which makes these rules useless) or because the people concerned do not perceive themselves as important determinants of ecological change when resources are liable to depletion/degradation (a subjective factor that prevents people from adopting the 'right' kind of behaviour). We are now considering the ease of societies in which competition over a scarce CPR that is at the same time liable to depletion/ degradation has led the corporate custodian community to adopt not only strict access rules but also conservation procedures or management mechanisms. (Bear in mind that a scarce CPR is one that has become subtraetible due to pressure of human exploitation.) There is actually plenty of evidence that these type-III societies may exist or have existed both in the Third World and in today's advanced countries. To make things clearer, it is worth illustrating this new possibility by quoting a number of striking examples from the literature.

Let us first have a look at the fishing sector. It is not difficult to find instances where management rules have been effectively enforced together with strict access rules with the apparent effect of ensuring the long-term viability of local fisheries once threatened with depletion/degradation. For example, in feudal Japan during the Edo Period (1603-1867), 'strict regulations were drawn-up by the Uwajima fief to ensure the conservation of resources. The use of bottom gill nets for the benthos catch was prohibited, and fishing with a small mesh trawl net ("with a mesh like a mosquito net") was forbidden. Pesticides were not permitted and night fishing with torches was regulated by season and sometimes prohibited. Seaweed harvesting was also banned during the spawning season because the eggs of some species of fish attach to the weeds. Similar regulations obtained in other fiefs, such as Fukuoka' (Ruddle, 1987: 21).

Turning now to a more egalitarian type of society, we may refer to local systems of fisheries management in Sahelian West Africa. Thus, among the three main fishing groups operating in the Niger River delta (the Bozo, Somono, and Sorko), regulation was based on village control over local fish resources. The timing of fishing, the spots where fishing was allowed or prohibited, the size of specimens kept, and the technology used were regulated by a local 'water master' who was also in charge of enforcing exclusionary rules against outsiders (Lawry, 1989a: 4; Fay, 1990a: 170-4, Verdeaux, 1990: 209 n. 8; Jeay, 1984). Since the 'water master' always symbolized a treaty of alliance with the spirit(s) believed to inhabit the local waters ('the water and the fish belong only to the spirits'), his authority was great among the fishermen and it was actually supported by a series of myths describing in a typically symbolic way the various rules and agreements in force within the local group as well as vis-à-vis neighbouring communities. The role of 'water-masters' was especially important in connection with the partitioning of waters among various groups of claimants or with the succession of fishing operations by different groups using different techniques, not a surprising feet given the high conflict potential involved in these situations. 'The conflicts presented by those [Bozo] myths often lead to conclusions backed by the strength of the water spirits: such spirits emasculate the Somono enemy, cause the appearance of hordes of crocodiles in order to repel bambara fishermen, or it is the water master himself who is a spirit, knows how to spoil the fishing party, and can transform himself into an offensive fish' (Fay, 1990a: 169, our translation). Yet, ideological intimidation was also a powerful means of sanctioning breaches against conservation rules. In the laguna Aby in the Ivory Coast, the fisherman who operated during the closed season exposed himself to the reprisals of the local spirit which could inoculate him with one or several diseases or even make him mad (Perrot, 1990: 184).

In the Pacific islands, a wide range of restrictions on fishing have been used for centuries which, according to Ruddle, 'were clearly designed to conserve stocks' (Ruddle, 1988: 81). This is confirmed by Johannes, a biologist, who goes so far as saying that Pacific islanders 'devised and practiced almost every basic form of modern marine fisheries conservation measure centuries ago, long before the need for marine conservation was even recognized in Western countries' (Johannes, 1982: 259, cited from Acheson, 1989: 362; see also Hviding and Baines, 1994: 25). This wide range of conservation measures included: the use of closed seasons (particularly during spawning), size limitations (bans on taking small fish), bans on destroying eggs, taboos on certain fishing areas, instructions to allow some proportion of the fish (or turtles) to escape so as to maintain the breeding stock, and, more recently, limitations on numbers and amount of gear that could be used.

In the Molucca Islands (in the eastern part of Indonesia), the people have inherited a system of traditional regulation called sasi. Sass literally means 'prohibition'. It is a customary law that prevents the people from degrading their natural resources according to a communal agreement. Sasi actually applies both on the mainland and in coastal areas. On the mainland, it regulates the protection of forests and staple foods like sago plants. At sea, it regulates the harvesting of certain kinds of marine resources such as sea cucumber, pearl shells (in Aru islands), a sardinelike fish (in Haruku islands), etc. Closing sasi means that for a certain period determined by communal agreement people are prohibited from taking out certain natural resources (Marut, 1994: 6).

What needs to be stressed, however, is that if communities of inshore fishermen are frequently found to have regulations regarding fishing gear, fishing seasons, and the rotation or assignment of favoured fishing locations, they typically fail to regulate the total pressure on the stock of fish by restricting harvesting efforts in the open season (Schlager, 1990; Scott, 1993). This failure can be ascribed to both informational problems (it is easier to monitor a vessel's gear and location or whether a closed-season regulation is violated than to monitor the individual catches of many vessels in ordinary times) and to the heterogeneity of fishermen communities regarding endowments (see above, Chapter 8, Sect. 2 and below, Chapter 12, Sect. 3).

One of the most widely cited example of self-managed CPR is that of a collectively-run system of meadows and forests in the Swiss Alps, described by Netting (1972, 1976, 1981) on the basis of the case-study of Törbel village. This study is instructive because the system emerged at a time when agrarian pressure on the alpine ecosystem was much higher than it now appears under low fertility and net population exodus. Indeed, from the sixteenth to the late nineteenth centuries, 'the Swiss highlands resembled very much the overcrowded mountain ecosystems of the developing countries today'. More precisely, new crops and cultivation techniques had permitted denser settlement, as a result of which both humans and livestock overstressed the natural environment. Ecological consequences are well known: 'The increase in the number of goats, and the greater fuel demands of the larger population caused serious degradation of the mountain forests' with consequent floods, avalanches, and topsoil loss (Pfister, 1983: 293-4, 297, quoted from McNicoll, 1989: 150).

It is in the last quarter of the fifteenth century that the inhabitants of Törbel signed a convention establishing an association to achieve collective regulation over the use of the alp, the forests, and the waste lands. According to the articles stipulated in that convention, access to well-defined common property was strictly limited to citizens. Moreover, rules regulating access to CPRs within the village community and intensity of use for the whole village, as well as rules providing for the protection of these resources from destruction, were carefully designed and effectively enforced. For example, it was decided that no citizen could send more cows to the summer grazing grounds than he could feed during the winter. Substantial fines were (and, apparently, still are) imposed on any villager found guilty of appropriating a larger share of grazing rights. This so-called 'wintering' rule is used by many other Swiss villages as a means for allocating appropriation rights to the commons. An alp association is specifically in charge of arranging for distribution of manure on the summer pastures, and organizing the annual maintenance work, such as building and maintaining roads and paths to and on the alp, and rebuilding avalanche-damaged corrals or huts. As for trees that provide timber for construction and wood for heating, they are marked by village officials and assigned by lot to groups of households, whose members are then authorized to enter the forests and harvest the allotted trees. The various measures taken— particularly the tight controls aimed at preventing overgrazing and the investments in weeding and manuring the summer grazing areas—have allowed Törbel villagers not only to maintain but even to enhance the productivity of their CPRs for many centuries. And it would seem that Netting's major findings are consistent with experience in many Swiss locations (Ostrom, 1990: 64).

We are also very fortunate to have at our disposal a remarkably well-detailed study by McKean (1986) on the experience of Japan in the management of traditional common lands. The author estimates that about 125 million hectares of forests and uncultivated mountain meadows were held and managed in common by rural villages during the Tokugawa period (1600-1867)—this represented approximately half of the surface of this category of land, the other half being under imperial or private property—and that more than 3 million hectares are still managed in this way today (McKean, 1986: 538). According to her, thousands of Japanese villages have actually succeeded in developing 'management techniques to protect their common lands for centuries without experiencing the tragedy of the commons'. Even as late as the 1950s, the remaining common lands continued to be managed on a sound ecological basis (ibid. 534).

McKean collected materials on three villages—Hirano, Nagaike, and Yamanaka— with a view to assessing the way commons were regulated on a micro-basis in Tokugawa Japan. From her account, it is plainly evident that Japanese villages used extremely detailed rules of access and conservation procedures in connection with these locallevel commons. For example, in order to prevent the kaya—a grass grown to produce thatch for roofs—from being cut at an immature stage for horse fodder, villagers usually designated an area with kaya as "closed' during the growing season. On the other hand, to ensure that daily cutting of fresh fodder for draught animals and pack-horses did not deplete the supply available for winter, villagers of Hirano designated one open area for daily cutting of fresh grass and another closed area as a source of grass to be dried into fodder for the winter' (McKean, 1986: 553-4).

Village forests were essentially divided into two zones: open patches of forest and closed reserves. Villagers could enter the first zone at any time 'as long as they obeyed rules about taking fallen wood first, cutting only certain kinds of trees and then only those that were smaller than a certain diameter, and only with cutting tools of limited strength (to guarantee that no tree of really substantial size could be cut)': or 'about leaving so much height on a cut plant so that it could regenerate, or taking only a certain portion of a cluster of similar plants to make sure the parent plant could propagate itself, or collecting a certain species only after flowering and fruiting, and so on'. Also, to limit the quantity of plants collected, village authorities could prescribe the size of the sack or container used towards that purpose. To control access to the first zone in a tighter way, the same authorities could also issue entry permits 'carved on a little wooden ticket and marked "entrance permit for one person"' (McKean, 1986: 554-5).

As for the closed reserves, they were set aside 'for items that had to be left undisturbed until maturity and harvested all at once at just the right time, or that the commons supplied in only adequate, not abundant, amounts'. The time for collection and the rules to be followed by each collector were decided by the village headman. For example, if the supply of a given natural product was limited, 'the reserve might be declared open for a brief period (two or three days) and households allowed to send in only one able-bodied adult to collect only what could be cut in that time'. Precise rules for harvesting varied from product to product and from village to village, yet, as a matter of principle, they 'appeared to be a judicious combination that rewarded strength and hard work but also severely limited the circumstances in which cutting was allowed, which ensured that the total supply was not threatened and no extreme inequality appeared among households in a given year or among kumi [groups of households] over time' (McKean, 1986: 555-6). The latter requirement, as we have seen in the previous section, could drive the local authorities to devise fixed rotational sequences so that each household or group of households had access to patches of varying quality.

Finally, it is important to note that there were written rules about the obligation of each household to contribute a share to collective work intended for maintaining the commons, such as systematic programmes of harvesting and weeding of certain plants in a particular sequence to increase the natural production of the plants they wanted; or the burning of the common meadow lands which was conducted each year to burn off hard and woody grasses and thorny plants (and kill pests), and which involved 'cutting nine-foot firebreaks ahead of time, carefully monitoring the blaze, and occasional fire-fighting when the flames jumped the firebreak' (McKean, 1986: 558-9).

There is solid evidence of effective traditional forest management systems pertaining to Third World countries as well. In Table 10.2, taken from Arnold and Campbell (1986), various types of rules and procedures used to conserve forest resources in traditional Asian village societies have been listed. In their study of traditional management systems of hill forests in Nepal, these authors have noted that there was a tendency for these systems to be very conservative, allowing access only to a few products: if the amount of a resource was too small to be adequately shared, or if it was difficult to control an open harvest, communities preferred to stop collection of the resource altogether. Thus, for example, in an oak forest managed for leaf litter, all fuelwood cutting was banned, even though some trees were overmature or unproductive' (Arnold and Campbell, 1986: 438). As a matter of principle, the length of time during which the forest was open to member villagers for specified product collection, the number of times in the year when collection was permitted, and the timing of the periods of access were determined in such a way that the plants exploited could grow before they were again harvested (ibid. 436). A study conducted in two districts along the upper Kali Gandaki river watershed in north-central Nepal confirms the existence of traditional community forest management systems in this country (Messerschmidt, 1986). It however brings out the fact that these systems may be of quite recent origin, dating back to only, a few generations or even a few decades. For instance, in Ghasa (Lete panchayat) forest access to sheep and goats has been strictly forbidden since 1974 (yet cattle, water buffalos, horses, and pack-mules are allowed to graze). In this village, moreover, cutting fuelwood and building materials by individuals is prohibited, although cutting poles and timber for public use is allowed on request. Also, every winter, each Ghasa household is required to collect debris and litter within the forest in order to reduce the risk of forest fires. Note that, if the district forest controller's staff intervenes to regulate permits for thinning the forest, the system basically remains under local control: access is controlled and the forest is patrolled by members of a panchayat forest committee following an old-style dating to prepanchayat times (ibid. 461-2).

Traditional community forest management systems have also been reported for India. Thus, in the hill districts of Uttar Pradesh, in a vast area known as Uttarakhand, informal practices used to regulate the utilization of forest produce by local communities. We are thus told that old customary restrictions on the use of the forests operated 'over large areas': the grass cut by each family was strictly regulated and branches or trees could be cut only at specified times, and with the approval of the entire village community. Successful management of the village forests 'was facilitated by the near-total control exercised by villages over their forest habitat' (Guha, 1985: 1940). A recent survey of six villages in the Almora district (Agrawal, 1994) shows that, in most of them, there still exist rules prescribing how fodder is to be extracted from the village forest (e.g. when cutting leaves from trees for fodder, villagers must leave behind at least two-thirds of the leaf cover on the tree) and prohibiting animal grazing for most of the year. In some cases, also, in ways very similar to those observed in Japan, it is specified how much fodder can be extracted from the village forest by each (resident) rights-holder and how many animals can be grazed, and these quantities can vary from year to year depending on an assessment of the state of the resource made by panchayat officials who inspect it before opening it to villagers. Simple measures are often used to meter the amount of grass withdrawn from the forest area. For example, 'passes entitle holders to cut a specified number of fodder bundles' and 'users are provided with a rope that they must use to make a bundle out of the grass they have cut' (ibid. 272). In two of the six villages surveyed, the grass in the village forest is sold primarily through auctions. The auction winner is free to cut grass from that section of the community forest for which he has successfully bid during the time set for that purpose (ibid.).

TABLE 10.2. Control systems used in traditional forest management

Source: Arnold and Campbell 1986: 437).

Our last example is taken from South America. The Huastec Indians of northeastern Mexico manage their forests in an indigenous system based on communal or individually owned forest plots called te 'lom (Alcorn, 1984). In these forest patches, arboriculture and natural forest management are integrated to produce elements of primary and secondary vegetation as well as introduced species. A great diversity of species—typically over 300 species—are found in a te'lom plot, some used for construction purposes, others for utilization in and around the farmstead, still others as medicinal plants, as fruits and leaves fed to domestic livestock, or as human foods. Most of the te'lom are managed in a way that causes little disturbance of the forest ecosystems. 'Useful plants that arise naturally are selectively allowed to mature to productive size. If a desired species does not, or cannot volunteer, or does not volunteer where it is desired, a person will plant or transplant it into the te'lom. Thus, the number of useful species in a te'lom increases slowly as some are planted or transplanted and others arise spontaneously and are spared. Unwanted plants [are] irregularly removed.' (ibid. 394).

Continue