Experimental Economics in Anthropology.pdf

Please download to get full document.

View again

of 12
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
MICHAEL CHIBNIK University of Iowa Review essay Experimental economics in anthropology: A critical assessment Foundations of Human Sociality: Economic Experiments ipants have a fixed amount of money (the stake) that must an
  Foundations of Human Sociality: Economic Experimentsand Ethnographic Evidence from Fifteen Small-ScaleSocieties.  Joseph Henrich, Robert Boyd, Samuel Bowles,Colin Camerer, Ernst Fehr,  and  Herbert Gintis,  eds. New  York: Oxford University Press, 2004. xix + 451 pp., map,appendix, bibliography, index. A B S T R A C T In recent years, well-funded cross-cultural economic experimentshave received considerable publicity in science journals. Theanthropologists conducting these experiments have made ambitiousclaims about the value of their research, saying that their findings‘‘illuminate the nature of human nature’’ and the ‘‘appropriatenessof the assumption of self-interest that underpins much of social science.’’ The publication of a new book,  Foundations of HumanSociality,  provides an opportunity to assess these claims. Thefindings reported in the book are consistent with long-accepted ideasin sociocultural anthropology. The book’s contributors are ofteninsufficiently critical of theoretical assumptions in economics,psychology, and biology and pay insufficient attention to relevanttheory in economic anthropology. In particular, the evolutionarypsychology paradigm accepted by many anthropologists conductingeconomic experiments does not aid in understanding their findings.[ experimental economics, economic anthropology, reciprocal exchange,decision making, anthropological theory  ] I n the summer of 1996, members of an indigenousgroup in the Peruvian Amazon took part in an un-usual economic experiment. Under the supervisionof Joseph Henrich, Machiguenga men and womenplayed the Ultimatum Game, in which two partic-ipants have a fixed amount of money (the stake) that mustbe divided between them. One participant, the ‘‘proposer,’’suggests how the stake should be divided. The other partic-ipant,the‘‘responder,’’mayacceptorrejectthisoffer.Iftheoffer is accepted, both participants receive the share sug-gested by the proposer. If the offer is rejected, neitherparticipant receives any money.This brief piece of fieldwork by a graduate student inanthropology at the University of California, Los Angeles,has had an extraordinary impact on funding opportunitiesand scholarship in diverse fields. The MacArthur Founda-tion subsequently gave grants to anthropologists to study the Ultimatum and related games in 15 small-scale soci-eties. The theoretical orientation of many of these re-searchers came from the controversial field of evolutionary psychology. The results of the MacArthur-funded research were discussed in a cover story in  Scientific American (Sigmund et al. 2002) and were reported in other widely read publications such as  Nature   (Fehr and Gachter 2002), Science News   (Bower 2002), and the  Wall Street Journal  (Wessel 2002). In 2002, the National Science Founda-tion’s (NSF) Cultural Anthropology Program gave its larg-est grant ($463,425 for three years) for further researchon this topic to many of the same anthropologists whohad been funded by the MacArthur Foundation. StuartPlattner, head of the NSF Cultural Anthropology Pro-gram, said at the time that ‘‘this new research is cut-ting edge stuff   . . .  [that] will advance economic theory’’(Bower 2002:106).The detailed results of the cross-cultural experimentsfunded by the MacArthur Foundation have now beenbrought together in  Foundations of Human Sociality  . Although the book describes diverse outcomes of the MICHAEL CHIBNIK University of Iowa Review essay Experimental economics in anthropology:A critical assessment  AMERICAN ETHNOLOGIST  , Vol. 32, No. 2, pp. 198–209, ISSN 0094-0496, electronic ISSN 1548-1425. A  2005 by the American Anthropological Association. All rights reserved. Please direct all requests for permission to photocopy or reproducearticle content through the University of California Press’s Rights and Permissions website, www.ucpress.edu/journals/rights.htm.  experiments, the principal and most-publicized finding of the project is a significant positive correlation between thedegreeofmarketintegration ofasocietyandtheproportionof the stake offered to the responder in the UltimatumGame. This correlation has been widely interpreted (e.g.,Ensminger 2002a:60; Surowiecki 2004:125) as an indica-tion that the development of markets is associated withgreater fair-mindedness and trust of strangers. The sevenauthors (Joseph Henrich, Robert Boyd, Samuel Bowles,Colin Camerer, Ernst Fehr, Herbert Gintis, and RichardMcElreath) of the overview and synthesis chapter of   Foun-dations of Human Sociality   confidently assert that thisand other project findings ‘‘illuminate the nature of humannature, the potential importance of culture, and the appro-priateness of the assumption of self-interest that underpinsmuch of social science’’ (p. 10).Few projects in sociocultural anthropology havereceived the publicity and funding given to these cross-cultural experiments. The publication of   Foundations of Human Sociality   provides an opportunity to assess theambitious claims that have been made about the valueof this research. The book consists of 14 chapters—a brief introduction, an overview and synthesis of the projectresults, a guide to experimental games aimed at measur-ing social preferences and norms, and 11 case studies of experimental games played in different parts of the world. Although Henrich and Boyd are the only anthropologistsamong the editors, the great majority of the case studiesare written by anthropologists about field sites where they have long carried out research.I find much to admire about the goals and activitiesof the anthropologists carrying out economic experi-ments. Their methodological rigor in the conduct of experiments, willingness to engage in dialogue with schol-ars in biological and social sciences, and commitment tocross-cultural comparisons are refreshing in a time whenmany anthropologists question any attempts to measurevariables and test hypotheses. Nonetheless,  Foundations of Human Sociality   promises more than it delivers. Intheir efforts to participate in interdisciplinary conversa-tions, the book’s contributors are often insufficiently critical of theoretical assumptions in economics, psychol-ogy, and biology and insufficiently attentive to relevanttheory in economic anthropology. In particular, the evo-lutionary psychology paradigm accepted by many of theauthors of the case studies does not help at all in under-standing their results.The findings in  Foundations in Human Sociality,  Iargue, are consistent with long-accepted ideas in socio-cultural anthropology. Why, then, is this research of suchinterest to scholars in other disciplines? To begin to answerthis question, I examine both the history of experimentaleconomics and theories about ‘‘selfishness’’ in economics,evolutionary psychology, and biology. Experimental economics, ‘‘selfishness,’’ andthe Ultimatum Game Some years ago Edward Hastings Chamberlin noted thateconomicshasnothistoricallybeenanexperimentalscience:It is a commonplace that, in its choice of method,economics is limited by the fact that resort cannot bemade to the laboratory techniques of the naturalsciences. On the one hand, the data of real life arenecessarily the product of many influences other thanthose which it is desired to isolate—a difficulty whichthe most refined statistical methods can overcome in a small part. On the other hand, the unwanted variablescannot be held constant or eliminated in an economic‘‘laboratory’’ because the real world of human beings,firms, markets, and governments cannot be repro-duced artificially and controlled. [1948:95]Nonetheless, in the past 60 years or so an increasing number of economists, including Chamberlin himself,have conducted economic experiments in laboratories with university students as participants. The diverse topicsthat have been examined include public goods, bargaining,industrial organization, auctions, and individual decisionmaking (Roth 1995).The Ultimatum Game has been described as ‘‘perhapsthe most well-known experiment in behavioral eco-nomics’’ (Surowiecki 2004:112). The game is designed tomeasure the extent to which responders will sacrifice theirown money to punish a proposer who has been unfair. A key feature of the Ultimatum Game is that it is a one-shot affair played between anonymous participants; one’sbehavior in the game (unlike much interaction in real life)has no influence on future socioeconomic relations. Thegame, therefore, is supposed to provide some indication of abstract society-wide values concerning fairness in socio-economic exchange between strangers. The first Ultima-tum Game experiment took place two decades ago (Gu¨thet al. 1982); since then, hundreds have been carried out(see Oosterbeek et al. 2004).Social scientists have shown so much interest in theUltimatum Game because participants’ behavior seemsto violate the principles of rational choice. Economistsgenerally assume that decision makers seek to eithermaximize gains or minimize losses of some type. In many situations, figuring out which behavior is ‘‘most rational’’is far from obvious, and analysts cannot easily say if indi-viduals are acting as economic theory would predict. 1 Thebest strategy for the responder in the Ultimatum Game,however, seems straightforward—accept whatever the pro-poser offers. Otherwise, the responder gets nothing. Thebest game-theoretic strategy for the proposer is only a bitharder to grasp. Because the responder should accept any offer, the proposer should suggest the division of the stake Review essay  n American Ethnologist 199  thatleavestheresponderwiththesmallestpossibleamountof money. But in experiment after experiment with univer-sity students, many respondents have been willing to re- ject what they regard as unfair offers. Proposers, apparently awareofthepossibilityofrejection,typicallymakeoffersthatleave respondents with a substantial portion of the stake. Accounts of these results often regard rejections of low offers in the Ultimatum Game as a puzzle to be explained. Why should participants care about fairness in this con-text? Furthermore, even university students in differentcountries vary considerably in the average size of theiroffers and their responses to similar-sized offers. Do inter-cultural differences in ‘‘selfishness’’ exist? If so, what arethe causes of these differences?Prior to the MacArthur-funded anthropological ex-periments, the results of Ultimatum Games were inter-preted in two principal ways. Some scholars (e.g., Fehr andGachter 2002; Nowak et al. 2000) emphasized similaritiesin results, seeing them as evidence that contemporary humans’ hunter-gatherer ancestors evolved an emotionalapparatus favoring generosity in situations in which shar-ing was necessary for survival. Others (e.g., Roth et al. 1991)emphasized the variability in results, seeing it as evidencethat cultures differ in their emphases on the importanceof sharing and fairness.The behavior of the Machiguenga in Henrich’s experi-ment in the summer of 1996 differed considerably fromthat of the university students who had previously playedthe Ultimatum Game. Students often rejected offers; pro-posers typically suggested almost equal divisions of thestake. Among the Machiguenga, an offer was only oncerejected; the proposers’ average offer was only 15 percentof the stake. These differences cannot be facilely explainedby the greater importance of money for the Machiguenga.The stake in the Machiguenga experiment was the equiva-lent of a day’s wages, the same as that offered in mostgames played by university students.Henrich’s experiment attracted attention because thelarge differences between how the Machiguenga and uni-versity students played the game suggested that cultureinfluenced notions of fairness considerably more than hadpreviously been thought. Moreover, ideas about the in-nate generosity of hunter-gatherers might be questionedbecause the Machiguenga (although primarily shifting cultivators, rather than foragers) made such low offers.Funding agencies were therefore amenable when an-thropologists sought support for comparative studies of economic experiments in settings very different from thosein which the great majority of Ultimatum Games hadpreviously been played.Some scholars think that the Ultimatum Game andrelated economic experiments can improve understanding of certain issues in evolutionary psychology and biology.Many classic problems in these fields concern altruisticbehavior, when individuals act in ways that are good fortheir group but appear to reduce their own chances of passing their genes on to future generations. The usualattempt to resolve such paradoxes is to argue that a particular example of altruistic behavior actually helps topass on the individual’s genes by increasing the probability of survival and reproduction of closely related kin. Muchaltruistic behavior involves cooperation between differentindividuals. Scholars sympathetic to the approach of evo-lutionary psychology often assume that human geneticpropensities to cooperate and trust one another aroseduring hunter-gatherer times when people helping oneanother were likely to be closely related.Karl Sigmund, Ernest Fehr, and Martin Nowak presenta typical argument about the relationship of economicexperiments to issues in evolutionary psychology:Ethical standards and moral standards differ fromculture to culture, but we may presume that they arebasedonouruniversal,biologicallyrootedcapabilities,in the same way that thousands of different languagesare based on a universal language instinct. Hume andRousseau would hardly be surprised. Today we havereached a stage where we can formalize their ideasinto game-theory models that can be analyzed mathe-matically and tested experimentally. [2002: 87]These authors assert that ‘‘our emotional apparatus’’ wasshaped by years of living in foraging groups, in whichkeeping secrets is hard: ‘‘If others know that I am content with a small share, they are likely to make me low offers[presumably of food]: if I am known to become angry  when facing a low offer and to reject the deal, others havean incentive to make me high offers. Consequently, evo-lution should have favored emotional responses to low offers’’ (Sigmund et al. 2002: 85).Other advocates of evolutionary psychology inter-pret ideas about reciprocal fairness in the context of groupselection. They argue that the groups that developed themost successful guidelines for pursuing fair interactions were evolutionarily successful. These groups developedgenetic traits that helped work out equitable exchanges(Bower 2002:106). Such claims are made even thoughbiologists are ordinarily skeptical about group selectionarguments to explain altruism in most species:There may be unusual characteristics of   HomoSapiens   and our close ancestors allowing biologicalgroup selection to work with greater force. Among these distinct characteristics is the superior ability of  Homo Sapiens   to maintain group membership char-acteristics and practice exclusion thus reducing thelevel of inter group mobility and enhancing the forceof group selections. [Bowles et al. 1997:9] American Ethnologist  n Volume 32 Number 2 May 2005 200  Critiques of assumptions of selfishness  A focus on participants in the Ultimatum Game acting contrary to their self-interest seems misguided from a purely economic perspective. For some years, social sci-entists (e.g., Simon 1976:262) have emphasized the multi-plicity of goals of decision makers and have noted thatprofit-maximization models are often incomplete guidesto actual behavior even in contexts in which they might beexpected to be most useful. Many economic decisionmakers, for example, engage in diverse activities aimed atreducing risk as well as increasing income. They may alsosimultaneously pursue economic activities that provideshort-term income (a college student waiting tables) andactivities aimed at providing income in the long term(the same student taking premed courses). Economic ex-changes involving cooperation, reciprocity, and trust en-able decision makers to reduce risk, diversify food andincome sources, and plan for both the short and long term. Why should anyone be surprised that players of the Ultimatum Game would think about the strategicadvantages of cooperation even in a one-shot situationin which profit maximization might appear to be the only reasonable goal?Evolutionary psychology explanations of the results of Ultimatum Games in different societies must deal with twofundamental problems. Because such approaches empha-size shared genetic traits among different human groups,they seem unable to provide insights into the reasons forcross-cultural variation. Moreover, the relative lack of generosity in the Ultimatum Game among foragers wouldseem to contradict the usual explanations for the evolutionof supposed genetic traits related to emotional reactionsto selfishness.Most sociocultural anthropologists are well aware thatindividuals rarely act solely on grounds of self-interest. Anthropologists (along with sociologists and political sci-entists) have long emphasized how societies are organizedto prevent individuals from behaving in ways—including the excessive pursuit of self-interest—that are harmful totheir group. We teach in our introductory courses how mores against selfishness in unstratified societies areenforced via ridicule, gossip, and the threat of banishment; we also point out how institutions in state societies, suchas laws, courts, police, and jails, prevent the untrammeledpursuit of self-interest. Such legal and social sanctionsmight with difficulty be fit into theories emphasizing selfishness by arguing that individuals follow their self-interest given the constraints of the particular society they live in. But this begs the question of how these legal andsocial sanctions evolved.The more important point that most anthropologistsemphasize in our introductory courses, however, is thatmuch of culture acts to make individuals want to behavemuch of the time in ways that are useful to society (i.e.,morally) even if this involves deferred gratification andsacrifice. The cultural morality that is transmitted fromgeneration to generation may well involve what the eco-nomic experimenters call ‘‘ideas about fair play.’’From thisperspective, most anthropologists would hardly be sur-prised by a finding that cultural ideas about sharing andcooperation prevent participants in economic experimentsfrom acting in their narrowly defined self-interest. Ultimatum Games and the world outsidethe laboratory The Ultimatum Game experiment that Henrich ran among the Machiguenga in 1996 was a surprising thing for ananthropologist to do. A significant tradition of formalmethods exists within sociocultural anthropology (seeBernard 1998, 2002), in which ‘‘informants’’ are asked torespond to psychological tests, ethnosemantic queries,and economic surveys. What Henrich did, however, seemsquite different from these nonexperimental quantitativemethods; his activities were more akin to what social psy-chologists do in their laboratories.Proponents of experiments in the social sciences em-phasize rigorous measurement and hypothesis testing.The confounding effects of extraneous factors can be con-trolled for through systematic selection of participants andthe manipulation of independent variables. Replications of experiments allow comparisons of results in different set-tings.Variablescanbeoperationallydefinedandmeasured.Some types of experiments are regarded as morerigorous than others (Bernard 2002:105–140; Johnson1998:146–151). True experiments, in which participantsare randomly assigned to groups, are more rigorous thanquasi experiments, in which the assignment of membersto groups is nonrandom. Still less-rigorous natural experi-ments are not conducted by researchers; instead, hypo-theses are tested by observing different groups whoseexperiences vary. A distinction is sometimes made (e.g.,Bernard 2002:106) between laboratory experiments, in which conditions are carefully controlled, and field ex-periments, which take place in situations that allow lesscontrol of extraneous variables. Of these varied researchmethods, only natural field experiments are commonly carried out by anthropologists. The projects funded by theMacArthur Foundation and the NSF’s Cultural Anthro-pology Program are best described as quasi experimentstaking place in (atypical) laboratories. As such, they repre-sent a significant methodological innovation in our field. 2 Sociocultural anthropologists have avoided true ex-periments and quasi experiments for good reasons. Therandom assignment of subjects to groups required in trueexperiments is seldom possible. Moreover, field situationsrarely allow the intentional manipulation of independent Review essay  n American Ethnologist 201
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks