HOW INEXORABLE IS THE "TRAGEDY OF THE COMMONS”?

INSTITUTIONAL ARRANGEMENTS FOR CHANGING

THE STRUCTURE OF SOCIAL DILEMMAS

By Elinor Ostrom

 

Since Garrett Hardin's captivating article in Science (1968), the expression, "the tragedy of the commons," has come to symbolize the degradation of the environment to be expected whenever many individuals own a scarce resource in common. Godwin and Shepard (1979: 265) refer to Hardin's article as “…the dominant framework within which social scientists portray environmental and resource issues.” To illustrate the logical structure of his theory, Hardin asks the reader to envision a pasture "open to all."' He then examines the structure of this situation from the perspective of a rational herdsman. Each herdsman receives a direct return from selling his own animals and suffers delayed costs from the deterioration of the commons when he and others overgraze. The herdsman is motivated to add more and more animals because he receives the proceeds from the sale of his own animals and bears only a share of the costs resulting from over‑grazing. Hardin concludes:

 

Therein is the tragedy. Each man is locked into a system that compels him to increase his herd without limit – in a world that is limited. Ruin is the destination toward which all men rush, each pursuing his own best interest in a society that believes in the freedom of the commons (Hardin, 1968: 1,244).

 

Hardin was not the first to notice the tragedy of the commons. Aristotle had long ago observed that "what is common to the greatest number has the least care bestowed upon it. Everyone thinks chiefly of his own, hardly at all of the common interest" (Politics, Book 11, Ch. 3). Over 150 years ago, William Forster Lloyd (1833; reprinted 1977) sketched a theory of the commons, which predicted improvident use for property owned in common. More than a decade before Hardin's article, H. Scott Gordon, of our own faculty, clearly expounded "The Economic Theory of a Common‑Property Resource: The Fishery" (1954), in what has become a classic of its own.

 

If the only "'commons" of importance were a few grazing areas or fisheries, the "tragedy of the commons" would be of little general interest. This is not the case. Hardin himself used the grazing commons as a metaphor for the general problem of overpopulation. The tragedy of the commons has been used to describe such diverse problems as the Sahelian famine of the 1970s (Picardi and Seifert, 1977), the problem of acid rain (R. Wilson, 1985), the organization of the Mormon Church (Bullock and Baden, 1977), the inability of the U.S. Congress to limit its own capacities to overspend (Shepsle and Weingast, 1984), urban crime (Neher, 1978), public sector/private sector relationships in modem economies (Scharpf, 1985), the problems of international cooperation (Snidal, 1985), and communal conflict in Cyprus (Lumsden, 1973).

 

Analytically, the theory that Hardin sketched has been formalized as an N‑Person, Commons Dilemma Game (Dawes, 1973; 1975). When the stark features of the formal representation are examined, the decision facing the herdsman in an open access commons has the same underlying structure as the decision facing each prisoner in the so‑called Prisoner's Dilemma (PD) game. For each of the players in this dilemma, the "don't cooperate" strategy strictly dominates the "cooperate" strat­egy. The equilibrium resulting from each player selecting his "best" individual strategy is, however, not the best joint outcome. Each player seeking to obtain the best result (the temptation payoff) and to avoid the worst result (the sucker's payoff) ends up with a third‑rate outcome.

 

The normal form to represent the structure of a PD game is:

 

                                    Player 2

Cooperate                   Don't Cooperate

Player 1

Cooperate                   Second best                 Worst result for 1

                                    result for both             Best result for 2

 

Don't Cooperate         Best Result for 1         Third best result

                                    Worst result for 2       for both

 

The Prisoner's Dilemma game has fascinated scholars in many fields. The paradox that individually rational strategies lead to collectively irrational outcomes seems to challenge a fundamental faith that rational human beings can achieve rational results. In the introduction to a recently published book, Paradoxes of Rationality and Cooperation, Richmond Campbell explains the "deep attraction" of the dilemma.

 

Quite simply, these paradoxes cast in doubt our understanding of rationality and, in the case of the Prisoner's Dilemma, suggest that it is impossible for rational creatures to cooperate. Thus, they bear directly on fundamental issues in ethics and political philosophy and threaten the foundations of the social sciences. It is the scope of these consequences that explains why these paradoxes have drawn so much attention and why they command a central place in philosophical discussion (Campbell, 1985: 3).

 

The deep attraction of the dilemma is also illustrated by the number of articles written on it. At latest count – a decade ago – more than 2,000 papers had been devoted to the Prisoner's Dilemma game (Grofman and Pool, 1975). 

 

When viewed as a situation that will be repeated for a finite and known number of rounds, most theorists predict that players will continue to select their dominant strategy in each round yielding a deficient equilibrium over the series of plays. Several predictions are made for an iterated Commons Dilemma game when the number of iterations is unknown. Many philosophers and formal theorists still argue that a rational player should play the "don't cooperate" strategy in every round (see, for example, Sobel, 1985). Others argue that rational players facing one another for an indefinite future could use contingent strategies to "teach" one another the benefits of selecting cooperative strategies (see, for example, Braybrooke, 1985; R. Hardin, 1982) or Bayesian estimates of subjective probabilities to resolve the dilemma (see J. Wilson, 1986). Still other models assume that resolute players can use strong threats of permanent retaliation, instead of cooperative moves and forgiveness, to develop models of repetitious games which predict the selection of cooperative strategies by all (see Lewis and Cowens, 1983; Cae, 1984; Bendor and Mookherjee, 1985). Taking an evolutionary approach, Axelrod (1981; 1984) has examined which strategies may be collectively stable under varying conditions of long‑term play.

 

With few exceptions (R. Hardin, 1982; Braybrooke, 1985; Orbell and Wilson, 1978), analyses of the Commons Dilemma have focused on the structure of the game as given.  The "solution" is viewed as immutable. From within the game, participants are trapped in an eternal struggle of tragic proportions. Even when analysts have examined situations which would extend for infinite periods of time, the presumption is usually made that the participants themselves have no control over the structure of the situation in which they find themselves. The prisoners in the story on which the PD game is based were indeed trapped. The physical constraints of separate cells in a prison and resolute District Attorney impose an immutable structure upon them. Is this immutability a temporary conceptual and methodological constraint or a deeper substantive necessity? Scholars and government officials presume that all participants in situations with the structure of a PD game are necessarily trapped in the structure of the situation; as prisoners are trapped in their cells, participants are themselves trapped in their own mental apparatus. I shall argue that the structure is conceptually and methodologically necessary for analysis, but not an empirical necessity. The inability of participants to change the structure may be an empirical reality in some situations. It is not an empirical reality in many situations, however.

 

All analysis is based on assumptions which keep some conditions constant and allow others to vary. Without considering some variables as exogenous to the situation under analysis, it is not possible to analyze that situation. Taking the structure of a Commons or Prisoner's Dilemma as given allows the analyst to derive the likely results that would occur if individuals were to find themselves in a situation that meets the conditions of the model.

 

In the on‑going, complex, multi‑level world of action, what is exogenous at one level of analysis may be endogenous at another level of analysis. This fixation on the rigidity of analytical constraints has had unfortunate consequences when scholars have turned from analysis to prescription. The grim predictions generated by many analysts about the Commons Dilemma has led to policy recommendations of an equally grim character. Ophuls (1973: 228) has, for example, argued that "because of the tragedy of the commons, environmental problems cannot be solved through cooperation ... and the rationale for government with major coercive powers is overwhelming…” Ophuls concludes that "even if we avoid the tragedy of the commons, it will only be by recourse to the tragic necessity of Leviathan" (1973: 229).

 

Garrett Hardin himself argued a decade after his earlier article that we are enveloped in a "cloud of ignorance" about "the true nature of the fundamental political systems and the effect of each on the preservation of the environment" (1978: 310). The "cloud of ignorance" did not, however, prevent him from presuming that the only alternatives to the Commons Dilemma are what he calls "a private enterprise system" on the one hand or "socialism" on the other (1978: 314). With the assurance of someone convinced that "the alternative of the commons is too horrifying to contemplate" (1968: 1,247), Hardin indicates that change must be instituted and with "whatever force may be required to make the change stick" (1978: 314). In other words, "if ruin is to be avoided in a crowded world, people must be responsive to a coercive force outside their individual psyches, a 'Leviathan,' to use Hobbes's term" (1978: 314).

 

The presumption that Leviathan is necessary to avoid tragedies of the commons leads some analysts to recommend central government control of most natural resource systems. Robert L. Heilbroner (1974) opined that "iron governments," perhaps military governments, are necessary to achieve control over ecological problems. In a somewhat less Draconian view, Ehrenfeld (1972: 322) suggests that if "'private interests cannot be expected to protect the public domain then external regulation by public agencies, governments, or international authorities is needed" (see also Carruthers and Stoner, 1981). Peter Stillman (1975: 13) points out that those who see "a strong central government or a strong ruler"' as a solution, implicitly assume that "the ruler will be a wise and ecologically aware altruist . . ." even though these same theorists presume that the users of common‑pool resources will be myopic, self‑interested, and ecologically unaware hedonists.

 

In contrast, other analysts call, in equally strong terms, for the imposition of private property rights whenever resources are owned communally (Demsetz, 1967; Welch, 1983). "Both the economic analysis of common property resources and Hardin's treatment of the tragedy of the commons"' leads Robert J. Smith (1981: 467) to suggest that "the only way to avoid the tragedy of the commons in natural resources and wildlife is to end the common‑property system by creating a system of private property rights” (my emphasis). I have no quarrel with the argument that dividing a commons and assigning individual property rights enhances efficiency in many situations (see, for example, Feeny, 1982). Similarly, I have no quarrel with the argument that administering some resources through central‑government authority may avoid the tragedy of overuse in other situations. I do take issue with the presumption that either central‑government administration or private property rights is "the only way to avoid the tragedy of the commons."

 

What is perplexing, as well as dangerous, is that scholars are willing to propose the imposition of sweeping institutional changes without a rigorous analysis of how different combinations of institutional arrangements work in practice. The current lack of sophisticated analyses of alternative institutional arrangements is a major weakness of contemporary work in the social sciences. Limiting institutional prescriptions to either "the market"" or "the state," means that the social‑scientific "medicine cabinet"" contains only two nostrums.

 

Game theory has provided powerful tools for analyzing certain situations. The tools exist to analyze far more complicated situations than the simple game that is repeatedly used to "illustrate"' the Commons Situation. Social scientists and philosophers have allowed themselves to be "hung up"' on simple, one‑level, paradoxical situations.  Immense scholarly energy has been devoted to trying to prove that individual rationality in a perverse situation will somehow avoid an irrational outcome. Why should we expect perfectly rational individuals placed in highly irrational structures, with no opportunity to change the structure, to achieve collective rationality? What is a more irrational way to structure any enduring situation than that represented by the PD game: no communication among the participants, no previous ties among them, no anticipation of future interactions, and no capacity to promise, threaten, cajole, or retaliate.

 

Would reasonable humans, trying to order their own long‑term relationships in a productive manner, structure a situation in such a perverse way? Reasonable humans may, of course, structure situations in this manner when they wish to prevent the participants in a situation from cooperating with one another. Some cooperation among participants may lead to harms externalized on others, such as in criminal conspiracies or economic cartels. Thus, cooperation is not an unambiguous good in all situations (see Ullmann‑Margalit, 1977). Is the only "choice" available to rational human beings a "choice" within the constraints of an externally imposed structure? Once we accept this limited view of choice, we are doomed to accept the imposition of structure by external authorities as the only way out of perverse situations such as the Commons Dilemma. I do not accept such a limited view of choice and want now to turn from this critique to a more positive approach to the study of Commons Dilemmas.

 

In the next section I will briefly describe four commons situations which have not resulted in tragedy. If we are to understand how individuals can escape from tragedy, we need to study such "success" stories carefully. These are particularly interesting success stories because none of them relies on central control or market mechanisms as its primary mode of management. Empirical cases provide the grist for further theoretical development. Once I have presented these four cases, I will then turn to several substantive and methodological lessons to be learned from the analysis of these success stories.

 

Successful Efforts to Cope with the Commons

 

West Basin, California

 

Given the arid conditions of Southern California, the development and use of water resources has been crucial to the growth of the area during the twentieth century. Metropolitan Los Angeles has the fortuitous circumstance that it overlies a complex set of interrelated groundwater basins. In addition to the construction of several major aqueducts to bring water from the Owens Valley, from the Colorado River, and finally from Northern California, water producers in Southern California have been dependent upon underground basins for storage as well as for the flow of fresh water. Building surface structures or towers to store water for peaking purposes is prohibitively costly. Since groundwater basins can provide some of this peaking capacity at low cost, the value of the groundwater basins for their storage potential (as contrasted to their flow of water) has become their most important function in a complex, conjunctive use system. The loss of one of the groundwater basins underlying metropolitan Los Angeles would be a major economic disaster.

 

The incentives facing producers of water from an underground basin depend, in part, on the type of property rights system in force at a particular point in time. When groundwater resources were first developed in Southern California, legal relationships were governed by a riparian‑like doctrine. One could not purchase groundwater rights without purchasing land. Once land had been purchased, however, overlying landowners had the rights to put as much water to beneficial use as they could withdraw. As long as the demand for groundwater did not exceed the average, long‑term supply, no problems resulted from the open access to all landowners.

 

As population and industry increased during the 1930s and 1940s, however, demand for water also increased. An annual overdraft (more water being withdrawn than was being replaced) occurred each year. Several of the basins were located immediately adjacent to the Pacific Ocean. Overdraft in these basins meant not only less water in storage, but continued overdraft could destroy the groundwater basin itself through salt water intrusion. Each producer faced a perplexing problem. As water levels fell, each producer was tempted to increase production in order to establish a proportionately larger claim to future pumping rights. Water that was not used simply flowed to the lowest water levels in the basin – allowing sea water to flow into the basin. The short run incentive was to pump as much as possible before disaster hit.

 

However, other possibilities existed. If all, or even most, of the pumpers would cut back on production, they could jointly benefit from the prevention of salt water intrusion. A substantial common good could be achieved if most producers halted their accelerated use. Hundreds of water producers pumped from each basin. No mechanism existed for them to come to agreement concerning joint strategies. No governmental authority had boundaries coinciding with any of the groundwater basins. Portions of 11 cities lay over West Basin, the most exposed basin in the series. The County of Los Angeles contained many of the basins within it, but was larger than any one of them.

 

By the end of the second World War, the problem facing water producers in West Basin can be clearly represented as a Commons Dilemma. Given the large number of participants and the absence of any ways to communicate and develop enforceable joint production strategies, one would predict from the theory of the commons that the basin would be destroyed by salt water intrusion within a few years.

 

But this is a success story. Today, West Basin is in better condition than it was forty years ago. Local water producers found a way to reduce their production from the basin and to create several special districts that now enable them to manage West Basin in a productive manner. This "success" was not imposed on West Basin by the State of California or the U.S. government. The initiatives to cope with their water problems came from the producers themselves and from local governmental officials.

 

How did this success come about? First, the users established a voluntary private association – the West Basin Water Association – to provide a forum for face‑to‑face discussion about their common problems. The producers used this forum to obtain the best available evidence about the current conditions of the basin and to discuss alternative joint production strategies. The association was supported by voluntary dues paid by producers based on the volume of water produced. A decision was made within the Association to use equity court procedures in helping to solve the problems they faced. Through discovery and reference procedures, the producers were able to obtain reliable information on past and current supply and demand conditions.

 

In the shadow of the court, producers were able to negotiate a contingent contract. This contractual device enabled a producer to agree to limit production if, and only if, 80 percent of the other producers also agreed to limit their production. A contingent contract effectively eliminates being played for a sucker while others pursue temptation strategies. The choice for each producer in deciding whether to sign a contingent contract is between: (1) cooperating in a situation where most others are also cooperating or (2) not cooperating in a situation where most others are also not cooperating. The contingent contract operated as an interim court decree for several years before it became the final court decree and was imposed on nonsignators, as well as those who had signed the original agreement. In addition, the court decision assigned each producer defined rights to the flow of water for the basin which could be purchased independently of land. Once rights were so defined, a market for groundwater could and did develop. A water master was appointed to continue to monitor production and ensure that producers remained within agreed limits.

 

While utilizing court procedures, West Basin producers also initiated proceedings to create several special districts. The first special district enabled producers to supplement the underground supply with a surface supply. The second district enabled producers to tax themselves on the amount of water they produced and to use that tax revenue to engage in replenishment efforts along the coast as well as inland.  Replenishment efforts involved several more local public districts which agreed to cooperate in a series of contractual arrangements (see E. Ostrom, 1965, for details; and Blomquist and E. Ostrom, 1985, for a recent analysis). 

 

The participants themselves, in the West Basin Commons Dilemma, were the major actors in designing a series of institutional arrangements to meet their particular needs. Cost‑sharing arrangements were developed for each step in the resolution process. The use of proportionate cost‑sharing began with the voluntary association. Dues were assessed based on the amount of groundwater extracted (thereby creating an incentive to understate use) while votes on association matters were also based on groundwater extraction (an offsetting incentive to overstate use). In the court case, costs of investigation and litigation were proportioned to the benefit obtained in the judgment, i.e., the prescribed rights to water. The cost of monitoring compliance is again proportioned to rights, with a portion being borne by the State of California. The state has an interest in accurate information about groundwater conditions and the provision of facilities to help avoid the tragedy of the commons.

 

Alanya, Turkey

 

Our second case stands in marked contrast to the highly modernized political economy of the Los Angeles metropolitan area. The inshore fishery of Alanya in Turkey is a relatively small operation. Fikret Berkes (1985c), a human ecologist at Brock University in Ontario, Canada, has provided an excellent description of the fishery and its institutional arrangements. Many of the 100 local fishermen operate in 2- or 3‑man boats using various types of nets. Half the fishermen belong to a local producers cooperative and half do not. The economic viability of the fishery in Alanya was threatened in the early 1970s by two factors. First, unrestrained use of the fishery created conflict among the users. Secondly, competition among fishermen for the better fishing spots greatly increased production costs and uncertainty regarding the harvest potential for any particular team of fishermen.

 

Fifteen years ago, members of the local cooperative began to discuss and implement a rather ingenious system for allotting fishing sites to the local fishermen. Fikret Berkes has described the system in the following words:

 

a.         Each September, a list of eligible fishermen is prepared, consisting of all licensed fishermen  Alanya, regardless of co‑op membership.

 

b.         Within the area normally used by Alanya fishermen, all usable fishing locations are named and listed. These spots are spaced so that the net set in one does not block the fish that should be available at the adjacent spot.

 

c.         These named fishing locations are in effect from September to May…

 

d.         In September, the eligible fishermen draw lots and are assigned to named fishing locations.

 

e.         From September to January, each day, each fisherman moves to the new location to the east. After January, the fishermen move west. This gives each fisherman an equal opportunity at the stocks which migrate east to west between September and January, and reverse their migration from January to May through the area (Berkes, 1985c: 14‑15).

 

Each year the list of fishing sites is endorsed by each fisherman and deposited with the mayor and local gendarme. The few infractions which incur are "'dealt with by the fishing community at large, in the coffee house. Violators may come under social pressure and, on occasion, threats of violence" (Berkes, 1985c: 15). If needed, the local gendarme is prepared to help in the enforcement of the agreement. Enforcement has, however, not been a major problem because the system is supported by most of the fishermen themselves. The system helps to allocate the best fishing sites to all fishermen on an equitable basis and has severely reduced conflict as well as production costs.

 

Toerbel, Switzerland

 

The third case is of Toerbel, a village of about 600 people located in the Vispertal of the Upper Valais region of Switzerland. Netting (1976: 133) identifies the most significant features of the environment as: "(1) the steepness of its slope and the wide range of microclimates demarcated by altitude, (2) the prevailing paucity of precipitation, and (3) the exposure to sunlight." For centuries, Swiss peasants have planted their privately owned plots with bread grains, garden vegetables and fruit trees, and hay for winter fodder. Cheeses produced by a small group of herdsmen, who tended village cattle pastured on the communally owned alpine meadows during the summer months, have been an important part of the local economy.

 

Written legal documents dating back to 1224 provide information regarding the types of land tenure and transfers that have occurred in the village and the rules used by the villagers to regulate the five types of communally owned property: the alpine grazing meadows, the forests, the "waste" lands, the irrigation systems, and the paths and roads connecting privately and communally owned properties. On February 1, 1483, Toerbel residents signed articles formally establishing an association to achieve a better level of regulation over the use of the alp, the forests, and the waste lands.

 

The law specifically forbade a foreigner (Fremde) who bought or oth­erwise occupied land in Toerbel from acquiring any right in the communal alp, common lands, or grazing places, or permission to fell timber.  Ownership of a piece of land did not automatically confer any communal right (genossenschaftliches Recht). The inhabitants currently possessing land and water rights reserved the power to decide whether an outsider should be admitted to community membership (Netting, 1976: 139).

 

The boundaries of the communally owned lands were well established long ago as indicated in a 1507 inventory document. 

 

Not only was access to well defined common property strictly limited to citizens, who were specifically extended communal rights, but written regulations specified in 1517 that "no citizen could send more cows to the alp than he could feed during the winter…" (Netting, 1976: 139). This regulation, which Netting reports is still enforced, imposed severe fines for any attempt by villagers to appropriate a larger share of grazing rights. The rules regulating the use of irrigation water involved an intricate rotation system based on sun and shadow movements on the surrounding mountains. Timber for construction and wood for heating were marked by village officials and assigned by lot to groups of households who then were authorized to enter the forests and harvest the marked trees.

 

Regulations also stated the obligations of those with use rights to provide labor inputs related to the cleaning of springs, the maintenance of an extensive irrigation system, the construction and maintenance of roads and paths, rebuilding avalanche‑damaged fences, and redistributing manure on common pasture lands. A codification of these regulations signed in 1531 included 24 separate articles regulating such diverse activities as: "immigration to or emigration from the community, hunting on the alp, stock damage to private plots, the spread of cattle disease, dispute settlement, participation in village government, alp pasturate rights, and compulsory communal building" (Netting, 1976: 139‑140).

 

In addition to a detailed system of communal rights, private rights to land are also well developed in Toerbel and other Swiss villages. Most of the meadows, gardens, grainfields, and vineyards in Toerbel were owned by separate individuals, but complex condominium‑like agreements were also worked out for the fractional shares that siblings and relatives may have in barns, granaries, or multi‑storied housing units.

 

Hirano, Nagaike, and Yamanoka Villages in Japan

 

The last case study involves several villages located in a mountainous region of Japan. For centuries in that country, extensive common lands have existed and been regulated primarily by local villagers. In an important study of traditional common lands in Japan, Margaret A. McKean (1984) estimates that about 12 million hectares of forests and uncultivated mountain meadows were held and managed in common by thousands of rural villages during the Tokugawa period (1600‑1867) and that about 3 million hectares are so managed today. While many villages have sold or divided their common lands in recent times, McKean (1984: 2) indicates that she has "not yet turned up an example of a commons that suffered ecological destruction while it was still a commons."

 

Three Japanese villages – Hirano, Nagaike, and Yamanoka – are similar in many respects to Toerbel. The villages are also established on steep mountains where many micro‑climates can be distinguished. Peasants cultivated their own private lands raising rice, garden vegetables, and draft animals. The common lands in Japan produce a wide variety of forest products of value to local peasants including timber, thatch for roofing and weaving, animal fodder of various kinds, and plant and forest residue for fertilizer, firewood, and charcoal.

 

Each village in earlier times was governed by an assembly. The assembly was usually composed of the heads of each of the households which had political standing in the village. The basis for political status varied substantially by village. In some villages the standing of households was based on cultivation rights in land, some on taxpaying obligations, and some on ownership rights in land. In some villages almost all households had political standing and rights to the use of the commons. In others, these rights were more narrowly held (McKean, 1984: 26).

 

Each village assembly established a relatively complex set of rules regulating both the use and enhancement of the commons owned by the village. Boundary rules clearly demarked which lands were held in common and which in private ownership. Entry rules unambiguously specified who was authorized to use the communally owned land. Ownership of the uncultivated lands near a village devolved from the imperial court to the villages through several intermediate stages involving land stewards and locally based warriors. National cadastral surveys were conducted late in the sixteenth century at a time of land reform that assigned "most of the rights to arable land that we today consider to be 'ownership' to peasants who lived on and cultivated that land" (McKean, 1984: 6). In the earlier systems the owners of large estates had employed agents in each village and authorized these agents to regulate access to the uncultivated lands. As villages asserted their own rights to these lands, they shared a clear image of which lands were private and which were held in common, and of how those lands held in common needed management in order to serve the long‑term interests of the peasants dependent upon them.

 

In traditional Japanese villages, the household was the smallest unit of account. Each village contained a carefully recorded, defined number of households. Households could not divide into multiple households without permission from the village. Rights of access to the communally held lands were accorded on a pro rata basis to each household. Consequently, households with many members had no advantage, and considerable disadvantages, in their access to the commons. Population growth was extremely low (.025 percent for the period between 1721‑1846) and ownership patterns within villages were stable (McKean, 1984: 29).

 

In addition to delimiting the ownership status of all lands' village assemblies also established detailed partitioning rules (Oakerson, 1984) specifying in various ways how much of each valued product a household could harvest from the commons.

 

Different villages arrived at different arrangements for guaranteeing an adequate supply of each of these products. For items that were needed regularly and that the commons Yielded in abundance, a village might allow co‑owners free and open entry as long as they abided by certain rules to make sure that a self‑sustaining population of mature plants or animals was left behind. To enter the commons, one might need to go to village authorities to obtain an entry permit, carved on a little wooden ticket and marked 'entrance permit for one person! The rules would probably restrict the villagers' choice of cutting tools or the size of the sack or container used to collect plants. Everyone would be expected to abide by the village headman's instructions 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.

 

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 only in adequate, not abundant, amount, villagers usually set aside closed reserves.... The village headman would be responsible for determining when the time had come to harvest thatch or winter fodder or other products, and would schedule the event... (McKean, 1984: 33).

 

The tailoring of village rules to the specific needs of each village and the ecological condition of a particular commons also required input from the villages to enhance and maintain the yield of the commons.

 

For collective work to maintain the commons‑to conduct the annual burning…, to report to harvest on mountain‑opening days, or to do a specific cutting of timber or thatch – there were written rules about the obligation of each household to contribute a share to this effort. Accounts were kept about who contributed what to make sure that no household evaded its responsibilities unnoticed....[and] if there was no acceptable excuse, punishment was in order (McKean, 1984: 39).

 

McKean's study is also strong testimony that it is possible for local communities to devise effective rules for managing their own common property resources. The establishment of the rules, the monitoring of behavior, the monitoring of the conditions in the commons, and the assignment of punishment were all conducted primarily in the village. McKean concludes that the long‑term success of these locally designed rules systems indicates "that it is not necessary for regulation of the commons to be imposed coercively from the outside" (McKean, 1984: 56).

 

What Can We Learn From These Cases?

 

Let me turn to the substantive lessons that can be learned from these four success stories. I will follow this with a discussion of the methodological lessons.

 

The Substantive Significance

 

The most important substantive lesson to be learned from these four cases is that it is possible for individuals facing a Commons Dilemma in natural settings to design their own institutional arrangements that change the very structure of the situation in which they find themselves. A self‑conscious process of instructional change occurred in West Basin and in Alanya. The participants designed new structures for themselves that have enabled them to use common‑pool resources in a productive manner. In West Basin, a rich supporting institutional structure enabled participants to enter into contingent contracts, to agree to create new political jurisdictions with specific powers to tax, and to engage in a creative form of public entrepreneurship to manage the commons. In Alanya, relatively poor fishermen, living in marginal circumstances, were able to extricate themselves from a deteriorating Commons Situation by inventing an ingenious set of rules for rotating fishing sites, enabling everyone to have a fair opportunity to obtain the catch.

 

These are not unique cases. In Southern California, participants in other groundwater basins have developed similar institutional arrangements to those of West Basin (Weschler, 1968; Rolph, 1982). While the designs are similar, each is tailored to meet particular circumstances. Besides Alanya, Berkes describes two other inshore fisheries owned communally where local fishermen have developed effective institutions for regulating use. The rules used in these other fisheries are different from those used in Alanya. Swiss peasants living in other alpine villages besides Toerbel have evolved their own systems for allocating the use of common grazing land (Wiegandt, 1977). Many other success stories are recorded in the literature (Siy, 1982; Wade, 1985; Cruz, 1985; Berkes, 1985a; 1985b; Uphoff, 1985; McCay, 1980; Berkes and Pocock, 1981; Acheson, 1975).

 

Success is, of course, not the only outcome. In Northern California, Arizona, and New Mexico, many groundwater basins are currently being threatened with excessive depletion (Knapp and Vaux, 1982). On the Turkish coast several inshore fisheries, not far from Alanya, face resource depletion and severe user‑group conflicts (Berkes, 1985c). Establishing a possibility is not the same as establishing necessity.

 

West Basin and Alanya illustrate how individuals can engage in self‑conscious design to change patterns of behavior within a relatively short period of time. These cases illustrate what I think Giddons (1979: 56‑57) means by the reflexive monitoring of action. Giddons considered this reflexive monitoring to be related not only to the actions taken in a situation but also to the "monitoring of the setting of interaction" (his emphasis). The Swiss and Japanese villages illustrate how institutions which evolved in the distant past can be well adapted to particular environmental and cultural circumstances. That the inhabitants of these extremely fragile mountain environments have been able to use them intensively for centuries, while harvesting a rich variety of forest and forage products, is strong testimony to the possibility of long‑term, stable outcomes that are not the tragedies posited in theory.

 

Another lesson to be learned from these four cases is that it is essential for those interested in proposing ways of improving human welfare to recognize the difference between the methodological requirement of exogenous variables and the mutability or immutability of these same variables. To analyze an action situation, the theorist must assume some variables are beyond the control of individuals in the situation. This methodological device of placing some variables outside short‑term control enables the analyst to understand the logic of the short‑term situation and to predict likely patterns of behavior in short‑term situations of a similar structure. Something must be taken as given in any analysis. The variables which are considered exogenous may or may not be under the long‑term control of participants. In some political systems, users of common‑pool resources cannot affect the institutional structure of their common resource and are like prisoners trapped in their cells. But our cases illustrate that, when users of a common‑pool resource do have long‑term capacities to change their own institutional arrangements, it is possible for them to create new arrangements well tailored to their particular circumstances. Consequently, no logical or empirical necessity traps users of a common‑pool resource into an inevitable tragedy.

 

Instead of the users of a commons being inextricably trapped in a tragedy, it is scholars who have allowed their assumptions to trap them into a presumption that short‑run tendencies will necessarily prevail in the long run. As long as scholars are enmeshed in a conceptual trap, they fail to address how participants in a Commons Dilemma can restructure their relationships to achieve mutually productive ways of relating to one another. Further, presuming the helplessness of participants leads to recommendations to impose solutions upon them. If we had strong logical and empirical reasons to believe that there is only "one best way" to avoid Commons Dilemmas, then we might urge the imposition of a single institutional arrangement for efficiency's sake.

 

But another lesson obtained from these cases is the futility of presuming there is "one best way." None of the four institutional arrangements which successfully overcame a Commons Dilemma is either a strict market arrangement or a central‑government arrangement. While the West Basin "solution" does involve creating firm property rights to the flow of water in West Basin, the basin itself is not privately owned. A "market" for water rights emerged subsequent to the court decree allocating rights to water. But that is not all that emerged. Water producers went on to create several local, public jurisdictions with regulatory and taxing authority to supplement their efforts to control the basin through the assignment of fixed rights to water. A complex series of private and public agencies jointly manage this sensitively balanced system. Nor is the polycentric, locally governed system, involving both private and public enterprises, a central‑government­ solution.

 

In none of the other systems do the rights to use even approach fully marketable rights. In Alanya, one must be a registered, local fisherman to qualify for the annual lottery. One fisherman cannot sell his annual schedule of fishing spots to another. Rights to the Toerbel Commons are individually inherited, but an outsider cannot buy rights to use the commons like an outsider can buy water rights in West Basin. Rights to use of the Japanese village commons are assigned to family units and remain with family units from one generation to the next.

 

None of the four systems resembles a central‑government solution either. The participants themselves decided which rules are to apply for allocating use. The administrative structure in all four cases is minimal. The users of the commons are also the governors of the commons.

 

The primary substantive lesson from these cases is that it is possible for humans to break out of the logic of the commons and to restructure the situation itself. Thus, it is important for policy analysts to recognize the difference between assuming a set of givens during an analysis and presuming these givens are really unchangeable. We have also observed that there may not be "one best way" of organizing for the management of natural resource systems. At least we are now aware of several systems which work quite well without being either pure market or pure central control systems.

 

The Methodological Significance

 

These cases do not only teach us substantive lessons; they also raise important methodological issues about how to study institutions and institutional change. In my earlier discussion, I referred to current analyses of Commons Dilemmas as single‑level analyses. The analysis is completely contained within the structure of a given situation.  The problem in understanding institutions is that one must use multiple levels of analysis. Several ways exist to identify levels. One method is to separate levels of operational choice, collective choice, and constitutional choice (see Kiser and E. Ostrom, 1982).  The typical way of modeling a Commons Dilemma is at an operational choice level. Ana­lyzing how individuals might change the rules of an operational choice situation is at a collective choice level. And, further, analysis of the rules for making rules is at a constitutional choice level.  When we move from an analysis at one level to a prescription for changing the rules used by people to structure that level, we must self‑consciously use multiple levels of analysis.

 

I cannot emphasize too strongly the importance of recognizing the necessity for self‑conscious and rigorous analysis of multi‑level systems if social scientists are to contribute to the improvement of social welfare. An assertion that there are multiple levels of analysis in the social sciences is not revolutionary. But until we begin to develop rigorous methods for separating the various levels, we will continue to thrash about in our own confusion (see V Ostrom 1985a; 1985b). This is a central theme in Richard Hofstadter's Goedel, Escher, and Bach (1979). Hofstadter distinguishes between systems in which the levels are well separated in time, in space, and in the language used to describe them, from systems which are not well separated.

 

In multi‑level systems which are well separated in time and space, we do not need to keep more than one level of analysis in mind at an one time. Our minds can effectively jump from one level to another depending upon the context of discourse and thought.  In many physical systems, for example, various levels of analysis are effectively separated by large gaps in space and/or time. In a kinetic model of a chemical reaction, the differential equations used to represent the chemical reaction rely on an assumption that the process under analysis can be isolated from its environment. P.J. Courtois has described these multilevel, chemical systems in the following way:

 

On the one hand, the environment is supposed to remain unaffected and is held constant; it is represented by a few parameters with fixed values. On the other hand, underlying processes, at finer scales in time or space, are hidden. Their dynamics are completely ignored. They are supposed to be in a state of equilibrium.…The success and the ac­curacy of these isolated analyses are, of course, to a great extent due to the large values of the differences in the time and size scales of the structures involved (Courtois, 1985: 591).