Froemming, Steven John; 1999; Rational Choice and Collective Action in an Andean Community. Ph.D. Thesis, University of Washington, Seattle. Excerpt from Summary and Conclusion (pages 825-843).

Summary and Conclusion

Introduction

My research in the Andes was motivated by the question, "How do agropastoralists resolve collective action problems in securing access to resources and producing collective goods?" My working hypotheses were that action in the Andes is primarily instrumental -- that actions are not valued and chosen for themselves, but as more or less efficient means to further ends -- and that conditional cooperation, decentralized sanctions, and communal partitioning of use rights would solve coordination and collective goods problems whenever high levels of certainty about community members' expected behavior could be maintained. To this end, I collected data in a bilingual Quechua-Spanish speaking, agropastoral village of about 225 households in the southern highlands of Peru from mid-1989 to January of 1992.

I conducted my fieldwork in the village of Ccachín, on the eastern slope of the Urubamba mountain range in the District of Lares, Province of Calca, Department of Cusco, Inca Region, Peru. The village encompasses 9,807 hectares, with the village nucleus at 3250 m. and the primary agricultural land varying in altitude from 2733 m. to 3973 m. above sea level. The comunidad campesina, an officially-recognized indigenous peasant community, has four population centers, with the main settlement -- Ccachín -- having about 175 households and 850 people. The settlement pattern is nucleated, but households maintain secondary cabins in outlying production zones. I carried out the bulk of my research in the main settlement and associated fields.

The principle crops of villagers are corn and potatoes; their principle livestock are sheep, cattle, pigs, and horses. Property rights vary from individual to communal by production zone. Close to the village, the land is worked intensively, with high labor and material inputs and seasonal fallow. Further from the village in the surrounding hillsides, the fields are managed through communally-regulated fallow. Households have their own plots, organized into sectors, with each sector fenced to control access by livestock when it comes into production. Every year, one or more of these sectors is cultivated, with the rest used for extensive pasture. There are no significant irrigation works, but the comunidad campesina plays an important role in organizing the use of agricultural and pasture lands. The most intensely-used land close to the settlement has the most private property rights and decision making, while the land furthest away has the most communal.

I focused my research in Ccachín around two goals, the first to explore the role and limits of instrumental rationality in decision making in general, and the second, as specifically applied to collective decision making, to determine the role that instrumental rationality, or alternatively, cooperative ideologies and values, play in determining the outcome of strategic interactions. To achieve these goals, I identified collective situations and problems in the village, determined the group decision making processes, leadership structure, sanctioning system, and other conventions and institutions governing community relations, recorded the pattern of interaction between community members in agropastoral production and public works, and documented the outcomes of these interactions.

My overall research approach is to view collective outcomes in terms of the goals and actions of individuals, with the hope of contributing to the effort to establish coherent microfoundations in the explanation of cooperation in the Andes and elsewhere. A full theory of cooperation requires us to move smoothly between biological, economic, social, and cultural theory, and to identify not only the benefits of behavior to individuals in specific environments, but to link the benefits to evolutionary processes that generate the behavior. In this way, we may be able to avoid the problems that emerged in functionalist and adaptationist accounts of old. The construction of such a comprehensive theory is still in the rudimentary stages and requires the work and cooperation of many people from a variety of disciplines and perspectives in theoretical, experimental, and descriptive realms. My particular contribution is a case study, or a series of case studies -- as the behaviors and traditions discussed in this thesis are varied -- that attempt to determine what would constitute successful explanation for the features I discuss. This is a tall order, and one can't do it all, but I believe that I have carried the torch forward

sufficiently for others to run with it, and to lay the groundwork for my further work. I have geared my discussion to other Andeanists, but it should be of interest to all involved in the investigation of rational choice and collective action.

In the summary that follows, I have organized my principle findings and conclusions according to three topics, rational choice, collective action, and ethnographic findings.

Rational Choice

Writing about his research on Maya corn farmers in Zinacantan, Mexico, Cancian stresses that:

The same can be said of my work in Ccachín. First and foremost, I am interested in the behavior of my hosts, friends, subjects, and acquaintances in Ccachín as exemplary of general human behavior, not as uniquely Andean, Indian ("runa"), or peasant ("campesino") behavior. We do well to bear in mind, for example, as Urbano suggests we do, that "Ayni is not a unique characteristic of the historical experience of the Andes. It's one more example of a fundamental principle that governs all human societies" (Urbano 1991:XXXI). This perspective may be taken for granted by many social scientists, but it bears emphasis in Andean studies, where a distinctly Andean worldview is often highlighted or granted totalizing effect.

I have phrased my working hypothesis in terms of rational choice, but there are many mechanisms that explain human behavior, and social scientists have discovered a repertoire of ways in which things happen. Elster (1989) notes that one of the greatest obstacles to planning is that we lack a metatheory that predicts when one mechanism or another will dominate and how they will work together. As he puts it, "What we have is a toolbox of mechanisms, not a set of laws" (Elster 1989:168). Given this state of affairs, we need to be circumspect about our explanatory claims but vigorous in our efforts to improve upon out theories of human societies. I start with one mechanism here, rational choice, to determine the extent that it successfully explains people's actions, particularly collective action, in the Andes. I start with rational choice not because it is in any sense methodologically prior to other individual-level mechanisms in the explanation of human behavior, but because it is the most theoretically developed, it yields clear predictions, and it is in principle at least tractable, grounded as it is in individuals' intentions and consciousness. Once one determines what a rational individual would do under given circumstances, one can proceed to observe whether this is what the individual actually does. In addition, there are some interesting parallels between evolutionary ecology and rational choice theory, so that we are closer here than with many explanatory mechanisms to being able to join the two together into a more synthetic theory. In the context of collective action, evolutionary and rational choice accounts often converge, so that it is a convenient starting point for building a more comprehensive theory.

Rational choice, as I understand it, is associated with action that is instrumental, self-interested, based on empirically-informed beliefs, and limited to material and social incentives. This rules out actions solely motivated by emotions, adherence to norms, altruistic urges, the pleasure of doing, and expressions of self. It also rules out cognitive fallacies, wishful thinking, and other processes where beliefs are subverted by cognitive limitations and the desires they serve. A process can be rational and fail to reach the truth, however, as it is a relation between a belief and the grounds on which it is held; rationality is relative to the evidence available for a decision and its optimal collection. In that sense, it is both a panhuman mechanism and culturally specific. The concept of rational choice is defined for individuals and not collectivities. There are strategic situations in which all would benefit if all acted irrationally, and my interest in rational choice is in its explanatory power, not its normative role. The province of rational choice theory in social scientific analysis is to explain individual action in terms of individual desires and beliefs. It does not provide a complete theory of social phenomena in itself, and we must turn to causal and functional mechanisms to explain social facts in terms of these actions, and desires and beliefs in terms of ecological and social states.

Often, anthropologists try to pass off the beliefs, dispositions, and cultural conflicts that they identify as being explanatorily relevant without being clear about what, or exactly how much, they are trying to explain. Whether or not we are right in our analysis, the least we can do is be clear about the mechanisms we are invoking in concrete circumstances, more exacting in our descriptions of causal chains.

In this thesis, I take a look at a number of behaviors and cultural traditions where other theorists have implied or suggested that rational choice has little explanatory sway, and by exploring the consistency, logic, and reach of their arguments, along with the features of the phenomena, attempt to show how a rational choice approach can add to their explication.

In Chapter 2, I raise the issue of whether the pattern of production zones in general, and the existence of communally-regulated sectoral fallow in particular, can best be explained in terms of ecological, technological, and sociological factors or by an ideological emphasis on communal values, equal division between all, and sequential rotation. Clearly, Andeans have models and experience from other domains that they can apply to communal land management, and this may reduce the costs of collective decision making and coordination, making the difference between success and failure. But two things must be said. First, property rights vary from individual to communal by item and production zone in most communities, and any attempt to define the Andean ethos in general terms as individualist or communal is at best ad hoc. Attempting to solve the problem by adding the notion of a contradiction between the two as is sometimes done is little better. Second, I conclude that while both practical and symbolic reasons may play a role in maintenance of the sectoral fallow system, they do not do so in the same way. Rational choice theory predicts that when new elements are added to an individual's opportunity set, behavior will change if one of them is seen as better than the formerly top-ranked alternative. In aggregate, this creates pressures for change. Theories of cultural choice and social norms, in the most general sense, do not give rise to the expectation that this will happen. Accordingly, they may help explain why behavior does not immediately and automatically adjust to new opportunities, but they are at a disadvantage in explaining why customary behavior is discarded when new opportunities arise. The communal values model does not do as well at explaining either the pattern or changes in the pattern of property rights and land use in Andean communities as rational-choice and adaptationist explanations do.

In Chapter 5, I discuss theft in the Andes and in Ccachín in particular, and I argue that it is most plausibly explained by rational choice conditioned by seasonal changes in opportunity. This suggestion is not novel, as a host of researchers have recognized that going back at least as far as the Inca era, the chief danger to crops from human theft was in April or May, when the maize was ripe but not ready for harvest. My discussion here shows how explanation can be multidimensional and multicausal, and how the interaction of mechanisms can lead to a situation where theft becomes socially expectable and practically tolerated, though not, as some would say, "institutionalized." At the same time, I do not assign all mechanisms equal weight, nor accept all purported explanations of theft in the Andes as equally plausible. In particular, the culturally-specific but totalizing cultural contradiction between seasonally-patterned dispositions of collective production and private appropriation posited by Gose (1991, 1994) to explain theft and several other features of Andean society is exactly the kind of analysis that troubles me for its lack of specificity about what and how much it is attempting to explain. Crop and animal theft are found in agrarian and pastoral societies throughout the world, and any attempt to explain their prevalence at certain times or places in the Andes in terms of uniquely Andean dispositions or beliefs is noteworthy for its lack of generalizability. The same is true of attempts to explain ayni in terms of a uniquely Andean worldview, as suggested above.

As I discuss in Chapter 6, failure to specify what is being explained is also a problem with Flannery et. al.'s (1989) work on the giving of animals to poor nonkin at animal marking ceremonies, but in a different way. Their work is innovative and important in applying one of Boyd and Richerson's models of cultural inheritance to the spread of a cooperative norm among herders in the Ayacucho area of the Andes. Their research is informed by ecological, evolutionary, and rational choice theory, and on the face of it, appears clear and tenable in identifying the explanatory mechanism involved. However, their effort also points out how difficult it is to adequately delimit the phenomenon to be explained and to model all relevant causal factors when attempting to explain real-life cases. I pick up on one of the suggested but unexplored avenues they leave -- the role that alcohol plays in such rituals and exchanges -- to explore the role that prestige and status play in the sharing of alcohol in other contexts as well. I suggest that the generosity of herd owners at animal fertility rituals, when it exists, may be better explained in terms of the striving for prestige or the signaling of status. A full explanation of suñay-giving, as the term is locally defined in many parts of the Andes, requires reference to kin-selection, tolerated theft, and reciprocal exchange as well, as gifts to nonkin represent a relatively minor proportion of all gifts at such events.

Chapter 6 is likewise important for my discussion of group-level functionalist explanation, as some form of group-selection is often used to explain why a particular behavior does not appear to conform with any of our rational actor models. Most attempts at group-functional explanation in the social sciences have hitherto failed to adequately specify the beneficial consequences involved, the beneficiaries, or the causal feedback mechanism maintaining the behavior at the level of individuals, with the latter proving especially problematic. While conceding Cohen's (1978) point that we may be able to separate functional explanation itself from the more historically contingent explanation of why one of several functional equivalents is the one that actually serves the suggested function, I find that this does little to salvage Flannery et al.'s argument for the group selection of altruistic suñay giving, as the same group-level consequences may emerge in this case as incidental byproducts of more instrumental herd building and herd-management behaviors.

As Winterhalder (1997) indicates in his review of the recent history of biologists' and social scientists' thinking on the mechanisms of selection, our ideas about group selection have evolved through the last forty years. First, there was a period of analytical permissiveness, during which nearly any beneficial effect discovered at any level could be posited as the cause of the origin and maintenance of a trait. The period from the mid-1960s to near-present was much more restrictive in this regard, and the restraint bore fruit with the identification of a variety of mechanisms rendering behaviors that did not previously appear to conform with the assumption of individual self-interest explainable in terms of it. We are now beginning a period in which new biological models of group selection and new sociocultural models of dual inheritance are again expanding the range of mechanisms and circumstances that enable the emergence of social behaviors with group-level benefits, and while I critique Flannery et. al.'s explanation in this particular case, I am inspired by their pioneering work.

In Chapter 12, I turn my focus to boundary marking rituals in the Andes in an effort to sort out the individual interests and collective processes sustaining them. Radcliffe (1989) treats boundary marking as a kind of pilgrimage or procession to express and reinforce collective identity and solidarity. In contrast, I suggest that it arose and has been maintained as a device for signaling group strength and defending property rights. Further, as a kind of collective action, boundary markings are as subject to free-riding as any kind of collective work party, and their gradual disappearance can be linked to the appearance of less costly and more definitive State-sponsored forms of boundary definition and defense. The goal is no longer so much to patrol or capture the land as it is to capture State support for one's territorial claims.

For the most part, I am concerned with strategic action in the Andes. In Chapter 4, however, I consider a domain that is not strategic -- ethnomedical decision making -- for what it can tell us about decision making under uncertainty and about certain nonrational choice mechanisms. I do this a) to contribute an Andean case study to the heuristics and biases research, b) to emphasize that the causal connections that Andeans make between personal misfortune and social and ecological disorders and their concern with supernatural sanctions are givens that potentially affect their calculations of the rewards of their actions, and, c) to acknowledge that because of the placebo effect, it is particularly hard for professional researchers, let alone lay observers, to collect accurate evidence about cause and effect.

The magnitude of the latter is extremely hard to determine for medical scientists because spontaneous remission of a disease cannot be distinguished from the placebo effect unless there is a control group that hasn't received any treatment at all, and such non-treatment controls are uncommon. Andean peasants, as with non-specialists everywhere, do not have such research tools at their disposal. This makes the choice of medical treatment options considerably more complex than choosing what to plant or what to eat, for example. What's intriguing here is that there's an important but limited realm where the belief that something works is the reason that it does so, and that it's an area where rational choice and nonrational kinds of cultural logic become joined. As Harrington (1977) puts it,

Uncertainty plays a role not only in the selection of treatment options in healthcare, but in the action sequences used in the Andes for such things as response to theft. In the latter arena, physical force tends to be used when a thief is caught red-handed, appeals to authority are made when detection is less immediate, and appeals for supernatural detection and sanction are made when the desire for justice are likely to go otherwise unfulfilled.

Gose (1994) takes some of the great minds of anthropology -- Tylor, Frazer, Malinowski, and Evans-Pritchard -- to task for conflating ritual with magic, for assuming that ritual-magic tries to do the same things as instrumental action, and for assuming that it only finds expression when people reach the limits of their technical competence. In his words,

Without challenging his overall conclusion, that labor and ritual do not always try to do the same things, and that they often coincide because they are complementary -- ritual serving to formulate or highlight the ends toward which practical action is directed -- I think that it is important not to throw out the baby with the bath water. Clearly, uncertainty is not an explanatory mechanism as such, but it is an explanatory condition that helps us predict when the choice among actions to take will be governed by one choice mechanism or another. I think that uncertainty helps us explain a lot, and I attempt to restore it to a more central position in my analysis here.

Collective Action

I agree with Elster that:

Referring specifically to the Andes, Urton puts it much the same when he cautions that:

My thesis benefits from and echos these points, and I elaborate in the explanation of a number of practices in the Andes.

The types of social situations I address in Chapters 5 through 14 include spite and revenge (negative rewards to self and others), selfishness (positive rewards to self, negative rewards to others), altruism (positive rewards to others at the expense of oneself), and cooperation (positive rewards to self and others, even when one's dominant strategy might be to free-ride on the cooperation of others). Specifically, I look at the behaviors of theft (+,-), the giving of live animals at animal marking ceremonies ("suñay ," -,+, at least within individual lifetimes), sponsorship of civil-religious cargos and the filling of community positions, taxation and fund raising, communal work parties, the marking of territorial boundaries, and social security and mutual aid (the five last, arguably, cases of +,+). In addition, I consider informal sanctions in the village in general. Expressing disapproval and carrying out threats is costly, and it may be vengeful, altruistic, or cooperative in outcome depending on the payoff structure and context.

Most studies of peasant rationality in the Andes to the present have focused on ritual kinship ("compadrazgo"), festival sponsorship, property rights, risk-reduction, and labor recruitment (one-to-one labor exchange ["ayni"] and labor exchange for food, drink, and festivity ["mink'a"]). While I touch on many of these issues when relevant to my discussion or when I feel that it would add comparative data to the ethnographic database of Andean communities, my particular focus has been on those social situations where there is a temptation for individuals to free-ride on the contributions of others.

The principle cost-sharing rule for community contribution in Ccachín, be it for labor or material contribution, is equal contribution from each family, while benefits tend to sort out naturally according to the material interests that each has in the undertaking. Perhaps the equal cost-sharing rule emerged historically in a time of greater equality of interests and means, but at present, it is probably maintained because of its salience, simplicity, low cost, and enforceability. The pyramidal structure of the traditional civil-religious cargo system insures that while ostensibly all adult men (and through them, their spouses, families, and relations) are equally eligible and expected to participate in the system, only those with sufficient social and economic resources to do so either actively seek or are socially pressured into the most costly positions. This introduces a more progressive element into the system, but it is not the strong leveling-mechanism that some investigators have claimed for other areas of Latin America.

At the same time, the growing importance of common-interest associations in Ccachín assures that such collectively-provided superior goods as education and religious structures, goods that appear to have a proportionally higher demand among the better off in the community, are provided. Such groups not only benefit from their higher than average level of interest and resources, but also from the advantages that smaller group size bring in assuring cooperation among their members in carrying out their aims. Their presence serve as a catalyst for community-wide cooperation through conformity effects, moral persuasion, political entrepreneurship, and other mechanisms.

An important ethnographic finding of my study, however, is that whether we are talking about attendance at community assemblies, participation in communal work parties, or response to fund-raising campaigns and mandatory payment of household assessments, cooperation appears to be far less than universal. Participation rates rarely exceed 60% for any of these, and are often far less. Most often, full participation in these endeavors is taken as given in Andean research, the fruit of a cooperative ideology. I'm not the first investigator to find less than universal participation, but among the first to do it in such detail, and to elaborate on the consequences.

I also establish that there is an elaborate system of formal sanctions -- pawns, fines, suspension of membership rights -- to increase the personal costs of noncompliance, and, as I put it in my working hypothesis, to assure high levels of certainty about community members' expected behavior. As Urton (1992) suggests, participation in village projects by rank-and-file community members seems to be motivated more by the threat of sanctions imposed for failure to participate and by making participation a means and condition for receiving certain community benefits -- most importantly, access to land -- than by communalist values.

Just as importantly, however, the collection of fines for failure to attend assemblies, to participate in communal work parties, or to keep one's animals out of restricted sectors is no more universal than the original actions that they attempt to reinforce. Ensuring compliance with the mandates of the comunidad campesina is itself a collective action problem. It is complicated by the relatively high costs of collection -- the risks of alienating the target individual or provoking retaliation, the time it takes collecting fines door-to-door, ambiguity about ability to pay, and the low personal rewards to the collectors and sanctioners formally delegated by the community in a nonsalaried governance system. In addition, the greater anonymity of noncooperators with the collection process, in contrast to the visibility of people's presence or lack of presence at the original work parties, probably plays a role. While a role call of attendees at every phayna helps bring informal social pressures to bear on consistent free-riders, there is no equivalent public exposure of individuals who have failed to provide expected material contributions to the community. These factors lead to crisis management, the erosion of public confidence in community leadership, and resignation in the face of widespread noncooperation, all of which can have a snowball effect further reducing the levels of cooperation. While the threat of sanctions has an effect on almost all community members and raises the average level of personal contribution, it has a differential effect on members depending on their interest in their reputation and social position, and it probably influences those pursuing resource-gathering strategies and positions that give them community-wide visibility the most.

In a cash-poor economy such as found in Ccachín, the problem of paying tax assessments and fines in a timely manner is always very real. The problem was exacerbated during the period of my study by the declining real value of the Peruvian currency. Since the amounts set for assessments and fines were rarely adjusted to keep pace with the changing value of the money itself, an incentive was created to withhold payment as long as possible. Those who paid their due on time effectively paid more than those who delayed, introducing inequities into the system.

Despite these problems, a lot gets done. On average, male household heads are expected to contribute twenty to twenty-five days to community work projects a year (roughly 10% of their total labor), and even if participation on average is roughly half that, it's still a lot of labor. As a walk through the village can attest, there are new additions and improvements to the infrastructure each year, and the pride that people take in their accomplishments probably has a positive effect on participation rates, particularly among those who feel the greatest stake and who have the most resources to contribute to the production of a particular good.

Unfortunately, while self-interest, along with features of the particular incentive structure of different collective activities in Ccachín probably goes far to explain why cooperation is less than universal, my data is not fine-grained enough to determine the role that self-interest, expressive interests, altruism, and social norms play in explaining the amount of cooperation that actually exists in the village. Likewise, since my thesis is based on a short-term study of a single community, I am not able to control for such variables as group size, commercialization of the local economy, access to alternative labor opportunities, organization and leadership, and other factors that could affect the cooperation rate. Presumably, different communities in the Andes differ in the rates of cooperation they are able to achieve in community projects, and comparative study is required to tease apart these factors. My research needs to be matched by others to create a sufficient database for comparative work.

All too often, communalism as a value is conflated with communalism as an institution and with communalism as an effect in the Andes. The relationship between these is more complicated than that. Cooperation can grow out of self-interest, people can cooperate in the carrying out of anti-social tasks, and unkind attitudes like envy and spite can facilitate sanctioning systems that further the collective good.

Given the extent to which mutualism, reciprocity, and solidity are emphasized by researchers as fundamental features of Andean ideology, it is relatively surprising that so few community-level welfare conventions are documented in the Andes, and that while communities such as Ccachín are relatively strong at conflict resolution, they do not do nearly as well at tending to the personal hardships and disasters suffered by its members. Other than the redistribution of unused land, the comunidad campesina itself does not provide a safety net or social insurance to aid community members who have suffered a family crisis or who have special needs. The burden falls on the afflicted and their families, and to an extent, as much to avoid litigation, retaliation, or social disapprobation as out of a sense of responsibility, on those who have caused others pain. Kin and compadrazgo relations are activated in time of need, but the latter, along with reciprocal relations in general, are harder to establish and maintain for those who have less to offer in exchange.

The lack of community action in these instances belies the assumption that local populations in small-scale societies readily redistribute wealth as a form of collective insurance as an outgrowth of general norms of solidarity. Beyond the limited economic resources of most campesino families, I believe we need to look to the correspondingly limited resources of communal institutions and the high costs of investigation and redistribution that organized social security systems require. However, I present the data I have gathered on the topic in the final chapter to raise the issue rather than to reach any conclusions. I will be content if it inspires others to build upon what I have done here.

As stated at the outset, I think that it is a mistake to assume that one privileged motivation -- such as self-interest -- can explain all instances of cooperation. By the same token, it is a mistake to assume that each instance of cooperation can be explained by one motivation. Cooperation occurs when and because different motivations reinforce each other. This is as true of the Andes as elsewhere. Diamond (1997) recalls the first line of Tolstoy's novel Anna Karenina, that "Happy families are all alike; every unhappy family is unhappy in its own way." This he calls the "Anna Karenina principle," and goes on to suggest that:

The principle can be extended from families to larger organized social groups, and it captures what I am trying to say here about collective action and other forms of cooperation. Whether we are considering the explanatory role of rational choice, social norms, or emotional commitments in cooperation, or the respective weight of kin selection, tolerated theft, or reciprocal exchange, these mechanisms work together and reinforce each other in complex ways, and ethnographically, it is not always easy to tease them apart. The challenge before us is to build a testable theory that models their interaction.

Ethnography

In this final section, I summarize some findings of a more personal or ethnographic nature that don't fit readily into the previous two.

In his study on impartiality as a theory of justice, Barry (1996) discusses how the idea of "informed agreement" is used by some philosophers to exclude superstition and false belief from any agreement among abstract actors, even though such beliefs may be reasonable to hold by the people in question given the evidence available to them. He then goes on to argue that one of the problems with this approach is that one person's superstition is another's belief.

The problem for any anthropologist trying to walk in the shoes -- or, as in my case, the sandals -- of another is to live opposing beliefs at the same time. In Ccachín, I was introduced to a world believed to be inhabited not only by runa in the flesh and blood, but ancient peoples and spirits that have a continuing effect on the present. It's a world where the machula can harm one's children, the Inca can protect their hidden treasures on inaccessible and fog-enshrouded hillsides, and mountain spirits can bring one success and misfortune. It's a world quite unlike the one I am accustomed to, although I make no claim that rural Andeans, or people in small-scale societies in general, are unique in holding the world so. Though differing in detail, homologically-similar beliefs are common in North American and European societies as well. Such beliefs affect the way people evaluate the consequences of their actions, and I have both practical and theoretical reasons for considering them here.

I don't discuss the psychology of cognitive dissonance reduction in the thesis, but I am interested in the tricks that anthropologists have developed for such while doing participant observation, and in the mechanisms of the mind that reduce dissonance at a more unconscious level. One thing that struck me while doing research is how far the very act of imitation -- for example, the blowing of phukuy to the apu (blowing a prayer over the coca leaves to the mountain spirits) -- combined with the desire to fit into a new social setting, the fear of disapproval for doing anything culturally proscribed, and the willful suspension of a skeptical attitude, takes one along the road to internalizing otherwise foreign beliefs.

In the first chapter and again in the fourth, I give readers some glimpses of what this world and these cultural beliefs look like. My discussions of radio uriwa, the use of hak'achu carcasses, and humoral medicine are intended not only to explore psychological and cultural mechanisms at play when empirical evidence is hard to come by or ambiguous and ill-behaved, but to understand the origin and nature of beliefs that I came to take for granted. The first two have received little mention or attention in the literature, and my contribution is to give them some thought and exposure here. The last has been widely written about by Andeanists and other Latin Americanists, and beyond offering an additional case study, my contribution is to focus on what it reveals about belief formation under conditions of uncertainty.

Writing about the Lares region, Angles Vargas (1992:36) says, with reference to the Paytiti-hunters, that "The human mind is so influenceable and so inclined to esoteric attitudes, that . . . it gives life to absurd things and acts." Adopting a dismissive and nonempathetic attitude, however, does little to improve our relations with others or enhance our understanding of universal cognitive processes. Following Shweder (1977), I conclude in Chapter 4 that some of the perceived cultural distance between ourselves and others disappears once one abandons the assumption that they think magically while we do not. The same holds true for abandoning the idea that we think scientifically and they do not. The differences within and between societies do not seem to me to be so much in people's everyday thought as in differential access to specialized institutions capable of organizing information into a format that lends itself to correlational manipulation, and in the degree our training and experiences teach us to rely on and trust in these institutions.

Beyond these brief sorties into the cognitive dimensions of doing fieldwork, in Chapter 3 I attempt to provide a first-hand account of village life in Ccachín through my participation in day-to-day social relations. Among the topics I discuss are taking pictures, villagers' reaction to world events, sending messages through intermediaries, gift-giving, money-lending, ritual-coparenthood, fear of strangers, gold fever, and the coca trade. My intent here is not only to give readers an idea of my modus operandi and the effect of my presence in the village, but to contribute to the broader anthropological discussion of ethnographic practice and the nature of the relationship between researchers and their subjects. Many of my experiences while in Ccachín were timely -- so, for example, I am able to report on how villagers made sense of and responded to radio reports on the Iraq war -- or unique, such as my role in sounding the alarm for a house fire, living in a military emergency zone, and being present for the staging of a motion picture, and I have provided ethnographic reports of these here. More importantly, given my interest in explaining cooperation, I have tried to be forthright about my own experiences in participating in reciprocal exchange networks and communal work parties -- in compadrazgo, ayni and mink'a, and phayna -- to bring self-reflection to bear on the motivations and social pressures involved. Berger and Kellner (1981) have written that "if the human beings to whom a concept is applied can not 'recognize themselves' in it . . . then . . . [we] will be constrained to construct new concepts that will be adequate to the situation in question." If this is to be the case with our subjects, a good place to start is with ourselves.

Several chapters in this thesis provide data of a general ethnographic nature on Ccachín and the Lares area. These include Chapter 1 on community history and cultural identity, Chapter 2 on spatial organization and production cycle, and all of the chapters in Part III describing the organization of the comunidad campesina. Not only does this add additional material from a little-studied region to the corpus of ethnographic studies in the Andes, it will allow other researchers to determine the degree of socioeconomic development in the village and reference it in their comparative works.

There is enough here to pique a variety of interests, and I am anxious to share observations and receive feedback on a variety of features that puzzle me. In Chapter 7 I report on the practice of residencia -- the annual inspection of households -- for example, without being able to say much about its origins or the extent of its dispersion. I have found few reports on it in the ethnographic literature. Kendall (1973:88) reports on what sounds like a similar practice during the Inca era in which "Inspectors saw that the women kept their houses neat, looked after food hygienically, supplied their family with clothing and raised their children properly," but whether the practice is a holdover from such times, emerged during the colonial or republican eras as a feature of hacendado rule, or represents a synthesis of the two, I can't say. Here, as with boundary marking rituals and a number of other cultural features described in this thesis, there are scattered reports, but a lot of systematizing to do.

Conclusion

We have the fortune that two theses have been produced about the same community at the same time, with different focuses, goals, and viewpoints, and readers who have followed me to this point would do well to consult Knox-Seith (1995) if they have not done so already. Reading both works will provide readers with a better-rounded ethnographic perspective.

My focus here has been on rational choice and collective action in an Andean community, and I have tried to provide the framework, data, and analysis to show that a rational choice perspective can make a solid contribution to sorting out many of the issues involved and explaining a variety of behaviors and cultural features in my research community and Andes-wide. It has been said that "Models are to be used, not believed" (H. Theil, source unknown), and I have proceeded with this in mind. Rational choice theory is inherently incomplete, and a variety of mechanisms focusing on different dimensions at proximate and ultimate levels of explanation are required to understand social phenomena. I hope that this study inspires further work on how the intentional, causal, and functional mechanisms that we identify can be fit into a broader theoretical framework.

References Cited

Angles Vargas, Victor
1992 El Paititi No Existe. Cusco-Qosqo: Imprenta Amauta.

Barry, Brian M.
1996 Justice as Impartiality. Oxford: Oxford University Press.

Berger, Peter L. and Hansfried Kellner
1981 Sociology Reinterpreted: An Essay on Method and Vocation. Garden City, NY: Anchor Press/Doubleday.

Boyd, Robert and Peter J. Richerson
1985 Culture and the Evolutionary Process. Chicago: University of Chicago Press.

Cancian, Frank
1972 Change and Uncertainty in a Peasant Economy: The Maya Corn Farmers of Zinacantan. Stanford: Stanford University Press.

Cohen, G. A.
1978 Karl Marx's Theory of History: A Defense. Princeton: Princeton University Press.

Diamond, Jared
1997 Guns, Germs, and Steel: The Fate of Human Societies. New York: W. W. Norton.

Elster, Jon
1989 Nuts and Bolts for the Social Sciences. Cambridge: Cambridge University Press.

Flannery, Kent V. , Joyce Marcus, and Robert G. Reynolds
1989 The Flocks of Wamani: A Study of Llama Herders on the Punas of Ayacucho, Peru. San Diego: Academic Press.

Gose, Peter
1991 House Rethatching in an Andean Annual Cycle: Practice, Meaning, and Contradiction. American Ethnologist 18(1): 39-66.

1994 Deadly Waters and Hungry Mountains: Agrarian Ritual and Class Formation in an Andean Town. Toronto: University of Toronto Press.

Harrington, Anne, ed.
1997 The Placebo Effect: An Interdisciplinary Exploration. Cambridge, MA: Harvard University Press.

Kendall, Ann
1973 Everyday Life of the Incas. New York: Dorset Press.

Knox-Seith, Barbara
1995 Being and Becoming Quechua: Social Relationships and the Dynamics of Control in a Peruvian Andean Village. Ph.D. Dissertation, University of Washington.

Radcliffe, Sarah A.
1989 Marking the Boundaries between the Community, the State, and History in the Andes. Journal of Latin American Studies 22(3): 575-594.

Shweder, Richard A.
1977 Likeness and Likelihood in Everyday Thought: Magical Thinking in Judgements about Personality. Current Anthropology 18(4): 637-658.

Urbano, Henrique
1991 Modernidad en los Andes: un Tema y un Debate. In Modernidad en los Andes. Henrique Urbano, ed. Pp. IX-XXXVII. Cusco - Qosqo: Centro de Estudios Regionales Andinos "Bartolomé de las Casas."

Urton, Gary
1992 Communalism and Differentiation in an Andean Community. In Andean Cosmologies Through Time: Persistence and Emergence. Robert V. H. Dover, Katherine E. Seibold, and John H. McDowell, eds. Pp. 231-266. Bloomington: Indiana University Press.

Winterhalder, Bruce
1997 Gifts Given, Gifts Taken: The Behavioral Ecology of Non-Market, Intra-Group Exchange. The Journal of Archeological Research 5(2):121-168.