sign in sign up
<<
Home , Daryll Forde, Ecological anthropology

The New Ecological Author(s): Conrad P. Kottak Source: American , New Series, Vol. 101, No. 1 (Mar., 1999), pp. 23-35 Published by: Blackwell Publishing on behalf of the American Anthropological Association Stable URL: http://www.jstor.org/stable/683339 Accessed: 13/01/2010 16:28

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=black.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Blackwell Publishing and American Anthropological Association are collaborating with JSTOR to digitize, preserve and extend access to .

http://www.jstor.org CONRADP. KOITAK Departmentof Anthropology University of Michigan Ann Arbor,MI 48109 The New EcologicalAnthropology

Olderecologies have been remiss in thenarrowness of theirspatial and temporal honzons, their functionalist assumptions, andtheir apolitical character. Suspending functionalist assumptions and an emphasisupon (homeo)stasis, "the new eco- logicalanthropology" is located at the intersection of global,national, regional, and local systems, studying the outcome of theinteraction of multiplelevels and multiple factors. It blendstheoretical and empineal research with applied, policy-di- rected,and critical work in whatRappaport called an "engaged"anthropology; and it is otherwiseattuned to thepolitical aspectsand implications of ecologicalprocesses. Carefully laying out a critiqueof previousecologies by wayof announc- ing newerapproaches, the article insists on theneed to recognizethe importance of culturemediations in ecologicalproc- esses ratherthan treating as epiphenomenaland as a mereadaptive tool. It closes with a discussionof the methodologiesappropriate to thenew ecological anthropology. ["the new ecology," , appliedor engaged anthropology,linkages methodology]

Ecological anthropologywas named as such during derstandand devise culturallyinformed solutions to such the 1960s, but it has many ancestors, including problems/issuesas environmentaldegradation, environ- Daryll Forde,Alfred Kroeber,and, especially,Jul- mentalracism, and the role of the media,NGOs, andenvi- ian Steward. Steward's influenced the ronmentalhazards in stimulatingecological awarenessand ecological anthropologyof Roy Rappaportand AndrewP. action. While recognizingthat local and regionalsystems Vayda, but the analyticunit shifted from "culture"to the are permeable,the new ecological anthropologymust be ecologicalpopulation, which was seen as usingculture as a carefulnot to removehumans and theirspecific social and means (the primarymeans) of adaptationto environments. culturalforms from the analyticframework. ColumbiaUniversity can be identifiedas the birthplaceof The following reviews the salient features of the old ecological anthropologyand the relatedcultural material- ecological anthropology,setting the stage for an explora- ism of MarvinHarris, which, however, drew as much on tion of importantaspects of an emergingnew ecological Steward's concern with culture change () and anthropology. culturecore as on his culturalecology. More diachroni- cally and comparativelyonented, cultural The Old EcologicalAnthropology and shared with ecological anthropologyan interest in the Its Unitsof Analysis adaptivefunctions of culturalphenomena, including relig- The ecological anthropologyof the 1960s was known ion (e.g., Rappaport's[1968] focus on ntual in the ecology for its functionalism,, and focus on nega- of a New Guineapeople and Harris's [1966, 1974] analysis tive feedback. examined the role of cul- of the adaptive,conservatory role of the Hindudoctrine of turalpractices and beliefs in enablinghuman ahimsa,with special referenceto the culturalecology of to optimizetheir adaptations to their environmentsand in India'ssacred cattle). maintainingundegraded local and regional . The ecological anthropologyof the 1960swas known Various scholars (for example, Friedman1974) attacked for systems theory arldnegative feedback.Cultural prac- both ecological anthropologyand culturaJmaterialism for tices were seen as optimizinghuman adaptation and main- a series of presumedfaults, includingcircular reasoning, taining undegradedecosystems. Factorsforcing us to re- preoccupationwith stabilityrather than change and simple think old assumptionstoday include populationincrease systems ratherthan complex ones, and Panglossianfunc- and high-tech-mediatedtransnational flows of people, tionalism(the assumptionthat adaptation is optimal-cre- commerce,organizations, and information.The new eco- ating the best of all possible worlds).Rappaport's distinc- logical, or environmental,anthropology blends theoxy with tion betweencognized and operationalmodels was related politicalawareness and policy concerns.It attemptsto un- to ethnoscience,which grew out of linguisticsbut became

AmericanAnthropologist 101(1):23-35. CopyrightC) 1999, AmericanAnthropological Association 24 AMERICANANTHROPOLOGIST * VOL. 101, NO. 1 * MARCH1999

anotherexpression of the ecological anthropologyof the ciety, the role of culture(especially ntual) in local and re- 1960s. Flounshing at Stanford,Yale, Pennsylvania,and gional resourcemanagement, ,and the Berkeley, ethnoscience focused on cognized ratherthan applicationof systemtheory to an anthropologicalpopula- operationalmodels andon classificationrather than action, tion. and it receivedsome of the same criticismsjust mentioned However enlighteningRappaport's analysis may have for ecological anthropology. been for understandingManng adaptation,the limitations The basic units of the ecological anthropologyof the of such an approachfor the study of more complex socie- 1960s were the ecological populationand the , ties were apparenteven in the 1960s. I had to confront treated,at least for analyticalpurposes, as discreteand iso- them as I plannedmy own ecological studyof the Betsileo lable units. The comparableunit for ethnosciencewas the of Madagascar,a muchmore populousgroup with a much ethnosemanticdomain (for example, ,ethno- more complex (chiefdom/state)?ociopolitical organiza- zoology, ethnoforestry).Assumptions of the old ecological tion. In ThePast in the Present:History, Ecology, and Cul- anthropology,now clearly problematic,are apparentin tural Variationin HighlandMadagascar (Kottak 1980), a some of its key definitions most importantlyecological large-scale comparativeand historical study based on populationand ecosystem. fieldworkdone in 1966 and 1967, I aKemptedan ecologi- Rappaportdefines an ecological populationas "an ag- cal analysisof the Betsileo-some 800,000 peopledistrib- gregateof organismshaving in commona set of distinctive uted over a much larger territorythan the Tsembaga means by which they maintaina common set of material Maring.Combining with surveytechniques, I relationswithin the ecosystem in which they participate" evaluatedecological adaptation(of the Betsileo and other (1971a:238).Several elements of this definitionmust now Malagasy)by focusingon associationsor bundlesof inter- be questioned.Given contemporaryflows of people,infor- related material variables (correlationsacross time and mation,and technologyacross cultural and social bounda- space) ratherthan by tryingto def1neand demarcatepre- ries, how distinctiveare the culturaladaptive means em- cise local ecosystems. The categories of materialcondi- ployed by any group? Given the fact and recognitionof tions I (like Rappaport)considered included aspects of the increaseddiversity within populations, how commonis the physicaland biotic environmentsand such regionalfactors set of materialrelations within ecosystems?Nor do most as tradeand warfare,but they also extendedto the role of peopletoday participate in only one ecosystem. stratificationand the state in determiningdifferential ac- Rappaportalso characterizesecological populationsas cess to strategicand socially valuedresources. Clearly, the "groupsexploiting resourcesentirely, or almost entirely, ecological analysisof state-levelsocieties could not be the within certaindemarcated areas from which membersof same as thatof bandsand tribes. other humangroups are excluded."Similarly, he defines Madagascaralso raisedthe complicatedquestion of the ecosystem as "thetotal of living organismsand non-living relationbetween culture (ethnicity), ecology, andthe state. substarscesbound together in materialexchanges within FredrikBarth some demarcatedportion of the biosphere"(1971a:238). (1958, 1969) had postulatedthat, especially Rappaport'scase exampleof a local ecological when thereis niche specializationplus exchange,conver- was the TsembagaMaring, a local territorialgroup com- gence andassimilation of contiguousethnic groups are not prisedof 200 tribalpeople living in colonialNew Guinea. inevitable;ethnic distinctions can be maintainedover time. But in today's world full of rural-urbanand transnational I noted that abruptenvironmental and ethnic shifts have migration,and ensuing remittances, how manygroups sub- been possible in Madagascar.For example, when people sist almostexclusively on local resources?How manyhu- moved to a certainarea of Madagascar'sforested eastern man groups live in precisely demarcatedecosystems that escarpment,they becameTanala, which means"people of are free of intrusionby others?To be sure,Rappaport was the forest.'?(This, by the way, is no longeras clearlytrue.) careful to recognize regional as well as local ecological Here, an ethnic label seems to have correspondedfairly populationsand ecosystems. He noted in 1971 that local closely to an ecologicaldistinction. ecosystems are not sharply bounded and that their dis- But such correspondencewas not generally true in criminationrests to a considerableextent on the aims of a Madagascar,where ethnic labels owed more to the politi- particularanalysis. Thus, "localecological populations. . . cal situationthan to the naturalenvironment. Within terri- participatein regionalexchange systems composed of sev- toriallylarge and populous"ethnic groups" (e.g., Betsileo, eral or many local populationsoccupying a wider geo- Merina,Sakalava), there is considerablevariation in envi- graphicarea" (1971a:251). In fact, the articulationof local ronment,modes of production,and means of adaptation. and regional ecosystems was an importantpart of Rap- Also, the existenceof ecoclines-regions of gradualrather paport'sfamed account of the ritualcycle in the contextof than abruptshifts from one set of ecological variablesto Maringwarfare and land use. His Pigs for the Ancestors: another-makes it difficultto claim a neatcorrespondence in the Ecology of a New GuineaPeople (1968) be- between ethnicityand ecology. Historically,in Madagas- came the classic case studyof humanecology in a tribalso- car as elsewhere,the state has often intervenedreating KOTTAK / THE NEWECOLOGICAL ANTHROPOLOGY 25

ethnic labels and distinctionsthat may or may not have the new ecological,or environmental,anthropology blends muchto do with ecology. theory and analysis with political awarenessand policy It is much more evident today than it was dunng the concerns.Accordingly, new subEieldshave emerged,such 1960s thatthere are no isolatedecosystems and thatall hu- as applied ecological anthropologyand political ecology mansparticipate in a worldsystem. In the contextof popu- (Greenbergand Park 1994). lation increase (more than a doubling since the mid- We cannotbe neutralscientists studying cognized and 1960s), the transnationalspread of information,images, operationalmodels of the environmentand the role of people, commerce, and organizations,and contemporary humansin regulatingits use when local communitiesand high-tech systems of transportationand communication, ecosystemsare increasinglyendangered by externalagents. many of the assumptionsof the old ecological anthropol- Many anthropologistshave witnessed personallya threat ogy need rethinking. to the people they study-commercial logging, environ- For example, Rappaport's "cognized model" (Rap- mentalpollution, radioactivity, and paport1968:237ff.; see Wolf, this issue) requiresmodifica- classism,ecocide, and the impositionof culturallyinsensi- tion. In his formulation,the cognizedmodel refers to native tive external managementsystems on local ecosystems interpretationsof the world, the set of rules and expecta- that the native inhabitantshave managedadequately for tions, orientingprinciples, concepts, meanings, and values centuries.Today's worldis full of neocolonialactions and thatare significantto an individualculture bearer and that attitudes;outsiders claim or seize controlover local eco- accountfor why he or she does things.Contemporary peo- systems, taking actions that long-termresidents may dis- ple still have cognized models, but anthropologistsmust dain.Concerned with proposingand evaluating policy, the increasingly wonder where such models onginate, how new environmentalanthropology attempts not only to un- they are transmitted,and the extent to which they are derstandbut also to devise culturallyinformed and appro- unique and shared.Diffusion may be as importantas en- priate solutions to such problemsand issues as environ- culturationin the contemporarycreation and transmission mentaldegradation? environmental racism, and the role of of cognized models. This would seem to be an issue of as the media,NGOs, and variouskinds of hazardsin trigger- much concernto the new psychologicalanthropology (for ing ecologicalawareness, action, and . example,)as to ecological anthro- Environmentalanthropologists focus on new units of pology. analysis nationaland international,in additionto the lo- The same is trueof his "operationalmodel" (Rappaport cal and regional,as these levels vary and link in time and 1968:237ff.).Rappaport used the termto describethe eth- space. Enteringinto a dialoguewith schools of naturalre- nographer'sabstraction from and analysis of what he or sourcesand the environment,anthropology's comparative she studies:an outsider'saccount of behaviorand its mate- perspectiveadds an internationaldimension to the under- rial determinarlts,context, and results; the trained ob- standingof issues like environmentaljustice aIldecosys- server's interpretationof why people do things; and the tems management,which naturalresource specialists have specificationof the limits that determinewhat individual been studyingfor decades,though mainlywith a U.S. fo- actionsmay be toleratedwithout destroying the systemthat cus. Conversely,anthropologists use methodsand perspec- sustainsthem. Specification of these dimensionsof the op- tives developedin othernations and culturesto shed light erationalmodel would seem to be as importanttoday as it on environmentalissues in the UnitedStates and Canada as was a generationago. The worldhas grownmore complex North Amenca itself becomes an increasinglycommon and probablyless comprehensibleto most natives. Soclal field of study in anthropology.And new methods- from scientistsneed new methods(see below) to studythis com- surveysto satelliteimagery are used to place ecological plexity and the myriadforces, flows, and exchanges that issues in a contextfar larger,deeper, and broaderin space now affect "local"people in their variousimmediate mi- and time thanthe bounded-systemapproach of the 1960s. lieus. Methodologies within the new ecological anthropology mustbe appropnateto the complexlinkages and levels that The New EcologicalAnthropology structurethe modernworld. The changes in ecological anthropologymirror more The differencesbetween the old and the new ecological generalchanges in anthropology:the shift from research anthropologyinvolve policy and value orientation,appli- focusingon a single communityor "culture,"perceived as cation, analyticunit, scale, and method.The studiesin the more or less isolatedand unique,to recognizingpervasive old ecological anthropologypointed out thatnatives did a linkagesand concomitant flows of people,technology, im- reasonablejob of managingtheir resources and preserving ages, andinformation, and to acknowledgingthe impactof their ecosystems (albeit through some rather unsavory differentialpower and statusin the postmodemworld on means, including mortalcombat and female infanticide); local entities. In the new ecological anthropology,every- but those studies, relying on the norm of culturalrelativ- thingis on a largerscale. The focus is no longermainly the ism, generallyaimed at being value-neutral.By contrast, local ecosystem. The "outsiders"who impinge on local 26 AMERICANANTHROPOLOGIST * VOL. 101, NO. 1 * MARCH1999 and regionalecosystems become key playersin the analy- Environmentalismentails a politicaland social concern sis, as contactwith externalagents and agencies (for exam- with the depletion of natural resources (Bramwell ple, migrants,refugees, warnors, tourists, developers) has 1989:34; Douglas andWildavsky 1982:1>16). This con- become commonplace. Ecological anthropologistsmust cernhas arisenwith, andin oppositionto, the expansionof pay attentionto the externalorganizations and forces (for a culturalmodel (developmentalism)shaped by the ideals example, governments,NGOs, businesses) now laying of industrialism,progress, and (over)consumption(Bar- claim to local and regional ecosystems throughoutthe bour 1973; Pepper1984). Environmentalawareness is ris- world. Even in remote places, ecosystem management ing today as local groupsadapt to new circumstancesand now involves multiplelevels. For example,among the An- to the models of developmentalismand . tankaranaof northernMadagascar (Gezon 1997), several Hazardscreated by developmenthave been necessarycon- levels of authorityclaim the rightto use and regulatenatu- ditionsfor the emergenceof new perceptionsof the envi- ronment.Environmental safeguards and conservationof ral resources and local ecosystems. Actual or would-be scarce resources are importantgoals from global, na- regulators there include local communities, traditional tional,long-run, and even local perspectives.Still, amelio- leaders(the king, chief, or mpanjaka),provincial and na- rativestrategies must be implementedin the shortrun and Wide Fundfor tional governments,and WWF (the World in local communities.If traditionalresources and products Nature),which is partlyfunded by USAID. areto be destroyed,removed, or placedoff limits (whether for developmentor conservation),they need to be replaced Issues for the New Ecological Anthropology with culturallyappropriate and satisfactory alternatives. One firm conclusionof the old ecological anthropology A new, possibly mediating,ethnoecological model- in all its guises (for example, the "ecological anthropol- sustainabledevelopment-has emerged from recent en- ogy" of Rappaportand Vayda, the "culturalmaterialism" countersbetween local ethnoecologiesand importedeth- noecologies, respondingto changingcircumstances. Sus- of Hams, and the "ethnoscience"of Berlin, Conklin, tainable development aims at culturally appropriate, Frake,and Goodenough)was thatindigenous groups have ecologically sensitive, self-regeneratingchange. It thus traditionalways of categorizingresources, regulating their mediatesbetween the three models just discussed:tradi- use, and preservingthe environment.An is tionallocal ethnoecology,environmentalism, and develop- any 's traditionalset of environmentalpercep- mentalism."Sustainability" has become a mantrain the tions thatis, its culturalmodel of the environmentand its discourse surroundingthe planning of conservationand relationto people andsociety. Today's world features a de- developmentprojects, but clear cases of successfulsustain- gree of politicaland economic interconnectedness unparal- able developmentare few. leled in global history. Local are being Issues addressedby the new ecological anthropology challenged,transformed, and replaced.Migration, media, ariseat the intersectionof global,national, regional, and lo- and industryspread people, institutions,values, and tech- cal systems, in a world characterizednot only by clashing nologies. Importedvalues andpractices often conflict with culturalmodels but also by failed states,regional wars, and those of natives. In the context of populationgrowth, mi- increasing lawlessness. Local people, their landscapes, gration,commercial expansion, and nationaland interna- theirideas, their values, and theirtraditional management tional incentives to degrade the environment,ethnoe- systems are being attackedfrom all sides. Outsidersat- cological systems that have preservedlocal and regional tempt to remake native landscapesand culturesin their environmentsfor centuriesare increasingly ineffective. own image. The aim of many agriculturaldevelopment projects,for example, seems to be to make the world as EthnoecologicalClashes: Developmentalism and much like Iowa as possible, complete with mechanized Environmentalism farming and nuclear family ownershipespite the fact thatthese models may be inappropriatein settingsoutside Challengingtraditional ethnoecologies are two, origi- the midwesternUnited States. Development projects often nally Euro-American,ethnoecologies: developmentalism fail when they try to replacenative forms with culturally and environmentalism(Kottak and Costa 1993). These alien concepts and productive units (Kottak models enter myriadcultural settings, each of which has 1990). Also problematicis the modernintervention phi- been shaped by particularnational, regional, and local losophy that seeks to impose global ecological morality forces. Because differenthost communitieshave different withoutdue attentionto culturalvariation and autonomy. historiesand traditions,the impactof externalforces is not Countriesand culturesmay resist interventionistphiloso- universalor unidirectional.The spreadof eitherdevelop- phies aimedat eitherdevelopment or globallyoriented en- mentalism or environmentalismis always influenced by vironmentalism. national,regional, and local ethnoecologiesand theirpow- A clash of culturesrelated to environmentalchange may ers of adaptationand resistance. occurwhen developmentthreatens and KOTTAK / THENEW ECOLOG1CAL ANTHROPOLOGY 27 theirenvironments. Native groupslike the Kayap6of Bra- resistancehas been supernatural.The deathof the ranger's zil may be threatenedby regional, national,and interna- young son was attributedto the ombiasa'smagical power. tional developmentplans (such as a dam or commercially After that the rangerwas less vigilantin his enforcement driven deforestation)that would destroy theirhomelands. efforts. A second clash of culturesrelated to environmentalchange occurs when externalregulation threatens indigenous peo- BiodiversityConservation ples. Thus,native groups,such as the Tanosyof southeast- ern Madagascar,may be harmedby regional,national, and Biodiversityconservation has become an issue in politi- internationalenvironmental plans that seek to save their cal ecology, one of the subfieldsof the new ecological an- homelands. Sometimes outsiders expect local people to thropology.Such conservationschemes may expose very give up manyof theircustomary economic andcultural ac- differentnotions about the "rights"and value of plantsand tivities withoutclear substitutes, alternatives, or incentives. animalsversus those of humans.In Madagascar,many in- A traditionalapproach to conservationhas been to restrict tellectualsand officials are botheredthat foreignersseem access to protectedareas, hire parkguards, and punish vio- more concernedabout lemursand other endangeredspe- lators. cies than about Madagascar'speople. As one colleague Problems usually arise when external regulation re- thereremarked, "The next time you come to Madagascar, places the native system.Like developmentprojects, con- there'll be no more Malagasy.All the people will have servationschemes may ask people to change the way they starvedto death,and a lemurwill have to meet you at the have been doing thingsfor generationsto satisfyplanners' airport."Most Malagasy perceive human poverty as a goals ratherthan local goals. In locales as different as morepressing problem than animal and plant survival. Madagascar,Brazil, and the Pacific Northwest of North On the other hand, acceptingthe idea that preserving America,people arebeing asked,told, or forcedto change global biodiversityis a worthwhilegoal, one vexing role or abandonbasic economic activitiesbecause to do so is for applied ecological anthropologyis to devise socially good for "nature"or "the globe."Environmentalists from sensitiveand culturally appropriate strategies for achieving northernnations increasingly preach ecological moralityto biodiversity conservation in the face of unrelenting the rest of the world raisingissues of nationaland local populationgrowth and commercialexpansion. How does autonomy."Good for the globe"doesn't play very well in one get local people to supportbiodiversity conservation Brazil, where the Amazon is a focus of environmentalist measuresthat may, in the shortrun at least, diminishtheir attention.Brazilians complain that Northernerstalk about access to strategicand socially valued resources? global needs and saving the Amazon only after they de- I am one of severalanthropologists who have done so- stroyedtheir own forestsfor FirstWorld economic growth. cial-soundnessanalysis for conservationand development AkbarAhmed (1992) finds the non-Westernworld to be projects.Such projectsaim, in theoryat least, at preserving cynical aboutWestern ecological morality,seeing it as yet naturalresources and biodiversity while promotinghuman anotherimperialist message. "The Chinese have cause to welfarethrough "development." My experiencedesigning snigger at the Westernsuggestion that they forgo the con- the social-soundnesscomponent of the SAVEM project venience of the fridge to save the ozone layer" (Ahmed (Sustainableand Viable EnvironmentalManagement), in- 1992:120). Well-meaningconservation efforts can be as tendedto preservebiodiversity in Madagascar,suggested insensitive as developmentschemes that promoteradical that a gradual,sensitive, and site-specificstrategy is most changes without involving local people in planningand likely to succeed (Kottak1990; Kottakand Costa 1993). carryingout the policies thataffect them.When people are Conservationpolicy can benefit from use of a flexible asked to give up the basis of theirlivelihood, they usually "leamingprocess" model ratherthan a ngid "blueprint" resist. strategy (Korten 1980; see also Kottak 1990). The ap- Considerthe case of a Tanosyman living on the edge of proachI recommendedfor Madagascarinvolves listening the Andohahelareserve of southeasternMadagascar. For to the affectedpeople throughoutthe whole processin or- yearshe has reliedon rice fields andgrazing land inside the der to minimize damage to them. Local people (with at reserve.Now externalagencies are telling him to abandon least some secondaryeducation) were trainedas "para-an- this land for the sake of conservation.This man is a thropologists"to monitorclosely the perceptionsand reac- wealthy ombiasa(traditional sorcerer-healer). With four tions of the indigenouspeople during the charlges. wives, a dozen children,and twenty head of cattle,he is an Like developmentplans in general,the most effective ambitious, hard-working,and productive peasant. With conservation strategies pay attention to the needs and money, social support,and supernaturalauthority, he is wishes of the people living in the targetarea. Conservation mountingeffective resistanceagainst the parkratlger who dependson localcooperation and participation. In ffieTanosy has been tryingto get him to abandonhis fields. The ombi- case mentionedabove, the outsiderguardians of the re- asa claims he has alreadyrelinquished some of his fields, serveneeded to do moreto satisfyaffected people, through but he is waitingfor compensatoryland. His most effective boundaryadjustments, negotiation, and compensation. For 28 AMERICANANTHROPOLOGIST * VOL. 101, NO. 1 * MARCH1999 effective conservation(as for effective development)the riskperception relatedto actions thatcan reducethreats to task is to devise culturallyappropriate strategies. Neither the environmentand to health?(For an Americantake on developmentagencies nor NGOs will succeedif they tryto such questions,see Kemptonet al. 1995.) impose their goals withoutconsidering the practices,cus- A key assumptionunderlying our Brazilianresearch is toms, rules,laws, beliefs, and values of the peopleto be af- as follows: althoughthe presenceof an actualhazard in- fected. creases nsk perception, such perception does not arise Reasons to conserve should be explainedin termsthat inevitablythrough rational cost-benefit analysis of .In- make sense to local people. We found in Madagascarthat stead, risk perceptionemerges (or lags) in cultural,politi- the economicvalue of the forest for agriculture(as an anti- cal, and economic contexts shapedby encountersamong erosion mechanism and reservoir of potential irrigation local ethnoecologies,imported ethnoecologies (often spread water) provideda much more powerful incentiveagainst by the media), and changing circumstances(including forest degradationthan did such global goals as "preserv- populationgrowth, migration, and industrial expansion). ing biodiversity."Most Malagasyhave no idea thatlemurs Environmentalawareness was especially evident in and otherendemic species exist only in Madagascar.Nor Brazil immediatelybefore and afterthe EarthSummit or would such knowledgeprovide much of an incentive for UNCED (the UnitedNations Conference on the Environ- them to conserve the forests if doing so jeopardizedtheir ment and Development),held in Rio de Janeiroin June livelihoods. 1992. Ecologicalawareness has been abettedby the media, In the long run millions of Malagasy stand to benefit particularlytelevision to which Brazil is well-exposed, from forest conservation.This figure includes the urban- with the world's most watchedcommercial television net- ites, who dependon forestedareas for waterand electricity, work,Globo. Brazilian environmentalism began to growin as well as the ruralpeople, whose rice cultivationwill be the mid-1980s,reflecting the returnof publicdebate along hurtby increasederosion and diminishingwatersheds. In with democracybertura, the Brazilianglasnost, after 1990 and 1991 my associatesand I foundthat some villag- two decades of militaryrule. Braziliarlenvironmentalism, ers in northernMadagascar already recognized the linkbe- strongestin cities in the southcentralpart of the country,is tween deforestationand a low watertable. Their ecological a growingpolitical force, but with mainlyurban support. awarenesswas risingslowly. Ruralpeople were startingto There is much less ecological awareness outside the realizethat irrigation water gets scarcerafter nearby forests main cities. A simple illustrationcomes from my own re- arecut. searchin Arembepe(Bahia state), an Atlanticfishing town I have been studyingsince 1962 (Kottak1999). Since the EcologicalAwarenessand EnvironmentalRisk early 1970s, Arembepehas sufferedair and waterpollution Perception from a nearbymultinationally owned titanium dioxide fac- The "applied"("engaged" in Rappaport's[1994] terms) tory. In three decades, Arembepe's municipal seat, role of today's ecological anthropologistmay be as agent Camacari,has growntenfold, from a sleepy ruraltown into or advocate plannerand agent of policies aimed at envi- a majorindustrial (petrochemical) center. Chemical pollu- ronmentalpreservation or amelioration-or advocatefor tion of the region'sstreams, rivers, and coastal waters now local people actuallyor potentiallyat risk throughvarious endangerswildlife andpeople. forces and movements,including developmentalismand Like others in their municipality,Arembepeiros face environmentalism.One research-and-developmentrole for realand immediate hazards- industrialpollution of the air, today's ecological anthropologistis to assess the extent fresh water, and the ocean. Several times, reportersfrom and natureof ecological awarenessand activityin various the nearbymetropolis of Salvadorhave coveredthe chemi- groupsand to harnessparts of nativeethnoecological mod- cal pollutionof Arembepe'scoastal waterand freshwater els to enhance environmentalpreservation and ameliora- lagoons. Most villagers have seen those reportson TV. tion. Still, local awarenessof immediateenvironmental threats With Brazilian colleagues Alberto Costa and Rosane hasn't increased as rapidly as the hazards have. Thus, Prado, I have researchedenvironmental risk perception walking along the beach north of Arembepeone day in and its relationto actionat severalsites in Brazil(Costa et 1985, I passeddead sea gulls every few yards.There were al. 1995; Kottak and Costa 1993). Our assumptionhas hundredsof birdsin all. I watchedthe birdsglide feebly to been that,although people won't act to preservethe envi- the beach, where they set down and soon died. I was ronmentif they perceive no threatsto it, risk perception stunnedand curious, but local peoplepaid little attentionto does not guaranteeaction. Our research sought answers to this matter.When I asked for explanations,people said severalquestions: How awareare people of environmental simply "thebirds are sick.'5Neither Arembepeiros nor sci- hazards?How do, can, and will they respondto them? entists I spoke with in Salvador(who speculatedabout an Why do some people ignore evident hazardswhile other oil spill or mercurypoisoning) could provide a definitive people respondto minordangers with strongfears? How is explanationfor the dead birds. Like other contemporaxy KOTTAK / THENEW ECOLOGICAL ANTHROPOLOGY 29

Brazilians,Arembepeiros seem to pay more attentionto NGOs are generallyviewed as more responsiveto local distantthreats than to local ones. wishes and moreeffective in encouragingcommunity par- In Brazil, nationally publicized environmentalthreats ticipation than are authoritananand totalitariangovern- have included a radioactive-cesiumaccident at Goiania, ments. However, this strategyis being increasinglycriti- the degradationof the Amazon rain forest, the murderof cized, especially in cases (for example, Madagascar)in the ecologically minded labor leader Chico Mendes, and which powerful,expatriate-staffed intemational NGOs are the effects of gold extraction,highway and dam construc- allowedto encroachon the regulatoryauthority of existing tion, and other intrusionsof the world system on native governments.There is a real issue of neocolonialismwhen peoples and their lands. The media have reportedabout it is assumedthat NGOs with headquartersin Europeor posed by mercuryin the rivers, industrialpollution, NorthAmerica are better representatives of the peoplethan andpoor waste disposal. are theirown electedgovernments, although certainly they Although Brazilian environmental awareness has may be. grown, media accounts have followed the international The emergence and internationalspread of "rights" lead by focusingon the Amazon as theecologically threat- movements(human, cultural, animal) is also of interestto ened region. Community-leveldata we have collected at ecological anthropology.The idea of humanrights chal- several sites show that Amazonian deforestationis the lenges the nation-stateby invoking a realm of justice and nonlocalecological issue most familiarto ordinaryBrazil- moralitybeyond and superiorto particularcountnes, cul- ians. When they are asked about "ecology,"most Brazil- tures, and religions.Human nghts are seen as inalienable ians mention the Amazon instead of hazards closer to (nation-statescannot abridge or terminate them) and home. But environmentalthreats with global implications metacultural(larger than and supenorto individualnation- (includingdeforestation) exist in many areasof Brazil be- states). Culturalrights, on the other hand, apply to units sides the Amazon. withinthe state.Cultural rights are vested not in individuals Althoughthe Brazilianmedia have increasedtheir envi- but in identifiablegroups, such as religiousand ethnicmi- ronmentalcoverage, there is little evidence for increased noritiesand indigenoussocieties. Culturalnghts includea ecological awarenessand activity at the local level, espe- group's abilityto preserveits culture,to raiseits childrenin cially among lower-class people. Such activity is more the ways of its forebears,to continueits language,and not likely to be initiated by NGOs and politicians than by to be deprived of its economic base (Greaves 1995:3). threatenedcommunities. My researchin Braziland Mada- Greaves(1995) points out that because culturalnghts are gascarconvinces me thatpeople won't act to preservethe mainly uncodified,their realization must rely on the same environment(regardless of what environmentalistsand mechanisms that create them pressure, publicity, and policymakerstell them to do) if they perceiveno threatto politics. Such rightshave been pushedby a wave of politi- it. They must also have some good reason (for example, cal assertivenessthroughout the world,in which the media preservingirngation water or a tax incentive)for takingac- andNGOs have playeda prominentpart. tion to reducethe environmentalthreat. They also need the The notion of indigenous intellectual propertyrights meansand the powerto do so. Risk perceptionper se does (IPR) has arisenin an attemptto conserve each society's not guaranteeenvironmental organization and action. culturalbase its core beliefs and principles,including its ethnoecology.IPR is claimed as a groupright a cultural NGOsand Rights Movements right, allowing indigenous groups to control who may know and use their collective knowledge and its applica- The worldwide proliferationof nongovernmentalor- tions. Muchtraditional cultural knowledge has commercial ganizationsis a majortrend in late-twentieth-centurypo- value. Examplesinclude (traditional medi- litical organization.This proliferationmerits the attention cal knowledge and techniques), cosmetics, cultivated of the new ecological anthropologybecause so many plants, foods, folklore, arts, crafts, songs, dances, cos- NGOs have arisen aroundenvironmental and "rights"is- tumes,and . According to the IPRconcept, a particu- sues. Over the past decade, the allocationof international lar group may determinehow indigenousknowledge and aid for "development"(including conservation as well as its productsmay be used and distributedand the level of development)has systematicallyincreased the share of compensationrequired. funds awardedto NGOs, which have gatnedprominence as social changeenablers. In the "developmentcommunity" (for example, the EnvironmentalRacism World Bank, USAID, UNDP [United Nations Develop- The issues of interestto the new ecological anthropol- ment Programme]),it is widely assumedthat a strategyof ogy are myriad,but a final one may be mentioned:envi- channelingfunds to NGOs, PVOs (privatevoluntary or- ronmentalracism. This is a form of institutionaldiscrimi- ganizations),and GROs (grass roots organizations)will nation in which programs, policies, and institutional maximize immediate benefits to community residents. arrangementsdeny equal rights and opportunitiesto, or 30 AMERICANANTHROPOLOGIST * VOL. 101 NO. 1 * MARCH1999 differentiallyharm, members of particulargroups. Bunyan ages in relationto researchmethodology and contentwas Bryant and Paul Mohai define environmentalracism as the goal of a working group of anthropologistswho first "the systematic use of institutionally-basedpower by met in 1986.l All of us were concernedwith the impactof whites to formulatepolicy decisions that will lead to the internationaland national forces, includingdevelopment disproportionateburden of environmentalhazards in mi- projects, on our researchlocales. Most membersof the norlty communities"(1991:4). Thus, toxic waste dumps LinkagesGroup (as we called ourselves)had worked more tendto be locatedin areaswith nonwhitepopulations. thanonce in the same region.We knew the advantagesof Environmentalracism is discriminatorybut not always observinghow people respondto differentopportunities intentional. Sometimes toxic wastes are deliberately andperturbations at variousstages of theirlives. dumpedin areasthe residentsof which are consideredun- We recognized the value of research samples (both likely to protest(because they are poor,powerless, "disor- communitiesand mobile individuals)that could be fol- ganized,"or "uneducated").(This is why a pollutingtita- lowed throughtime. Whatkinds of linksdid theyhave with nium dioxide factory was placed near my Brazilianfield others, including external agencies? This line of inquiry site of Arembeperather than in an areahaving more politi- entaileda census approach,a networkapproach (to trace cal clout [see KotLak1999].) In othercases propertyvalues relationshipsassociated with geographicalmobility and fall after toxic waste sites are located in an area. The externalinterventions), plus surveyand ethnographic tech- wealthierpeople move out, and poorerpeople, often mi- niques.The linkagesapproach to change also requiredat- norities,move in, to sufferthe consequencesof living in a tentionto the roles of governmentaland nongovernmental hazardousenvironment. organizations,and of changesin marketing,transportation, andcommunication systems. Methodology in the New Ecological Anthropology One method of linkages researchis to study a site or sites over time. Another is systematic intercommunity The new ecological anthropologycan draw on a series companson, requinng multiple sites that are chosen be- of high-techresearch methods. Satellite imagery (deployed cause they varywith respectto key criteria.These sites can synchronicallyor diachronically)has been used to locate be drawn from the same region, and the data collected ecological hotspots (e.g., areas of deforestationor pollu- would be part of the same study. They can also be from tion), which have then been investigatedon the groundby differentregions (even differentcountries), if anthropolo- multidisciplinaryteams (See Green and Sussman 1990; gists can provide minimumcore data (Epstein 1978:220) Kottaket al. 1994; Sussmanet al. 1994). GIS (geographi- to makecomparison possible. Linkages research extends to cal informationsystems) and other approachesmay be the levels at which policies are workedout, examiningar- used to map variouskinds of dataon humanand environ- chives and official recordsand interviewingplanners, ad- mental features (See Sponsel et al. 1994). Macroscope ministrators,and otherswho impingeon the studypopula- software,developed by J. StephenLansing and others,fa- tion(s).The aim of linkagesmethodology is to linkchanges cilitatesthe mapping-on a computerscreen-of various at the local level to those in regional,national, and world kindsof information,such as yields in Balinesefields in re- systems. lation to pest damage and farrningpractices. Survey data Linkagesresearch is plannedas an ongoingprocess re- can be collected across space and time and compared. quiringteamwork. Time and personnelare neededto fol- However, the availability of such high-tech methods low a dispersingpopulation, to studydifferent sites, to in- should not seduce us away from anthropology'scharac- terview at many levels, to explore archivesand records, teristic focus on people. Ethnographicresearch in varied and to do follow-up studies.Involvement of host country locales helps us discover relevantquestions, which some colleagues,including local assistantsand other community of the techniquesjust mentionedcan help us answer.The residents,is a key to continuity.Thus, linkages also refers new ecological anthropologycan use high-techmethods, to cooperationby people with common researchinterests while takingcare not to let electronicdazzle divertatten- in the effortto generatea fundof data. tion from direct, firsthandethnographic study of people One exampleof linkagesmethodology is the researchI andtheir lives. directed in Brazil on industrializationand commercial Also relevant to the new ecological anthropologyis expansion, focusing on environmentalhazards and risk linkagesmethodology, as elaboratedby Kottakand Colson perception.The investigationproceeded at two levels: (1) (1994). As ElizabethColson and I have pointedout, an- national Brazil as a whole, where the governmentintro- thropologistsare increasinglydeveloping models of their duced a policy of industnalizationin the early 1950s, and subjectmatter that are isomorphicwith the structureof the (2) local across a range of sites differentlyexposed to modern world, including the various regional, national, risks (Costaet al. 1995; Kottakand Costa 1993). The field and internationallinkages withinit. We use the term link- researchdesign was systematicintercommunity compari- ages methodologyto descnbe various recent multilevel, son (basedon quantitativeand qualitative data). This meth- multisite,multitime research projects. A definitionof link- odologyadds an analyticlevel to traditional"risk analysis," KOTTAK / THENEW ECOLOGICAL ANTHROPOLOGY 31

which studies populationsdirectly exposed to environ- Table 1. Linkagesmethodology summarized. mental hazards like nuclear repositories.Given thatre- searchdesign, publicreactions to a threatare inevitably in- * longitudinal terpretedwithin a stimulus-responseframework (a threat * systematicintercommunitycomparison causes certainresponses). By contrast,our design assumed * multiple sample populations thatvariation in environmentalawareness and riskpercep- Irom same reg1on from differentregions tion could be most accuratelyunderstood by studying a from differentcountries range of sites differentiallyexposed to hazards.Compari- * researchextends to levels at which policies are developed son is essential. Any approachlimited to endangered * interviewplanners, administrators, others who impingeon the study groups can't help but see risk perceptionmainly in re- population(s) sponse to an immediatestimulus. (For otherlinkages pro- * examine archives and official records jects, see Kottakand Colson 1994.) * researchplanned as ongoing process The linkagesapproach (summarized in Table 1) accords * requiringteam work with anthropology'straditional interest in culturalchange. * key to continuity involvement of Its roots can be traced to earlier work, including Julian host countrycolleagues Steward's large-scaleevolutionary and comparativepro- local assistants jects (Steward 1950, 1955, 1956), the researchof Max other communityresidents . Gluckmanand others who did "extended-caseanalysis," and world system approachesthat emphasize the em- considerfeedback among local, regional,and nationalin- beddednessof local culturesin largersystems (Comaroff 1982; Mintz 1985; Nash 1981;Roseberry 1988; Schneider stitutions.However, linkages methodologystill requiresa 1977;Wallerstein 1974; Wolf 1982). basisin fieldwork. The linkagesapproach agrees with world system theory thatmuch of whatgoes on in the worldtoday is beyondan- Putting People, and Anthropology, First thropology's established conceptualand methodological While recognizingthat local and regional systems are tools. Traditionalethnography, based on village interviews permeableand that contact and power relationsare key and participant-observation,assumed that informants featuresof ecological adaptation,the new ecological an- knew what was going on in that delimitedspace. Today, thropologymust be carefulnot to removelocal people and can supply all the informa- however, no set of informants their specific social and culturalforms from the analytic tion we seek. Local people may not be helpless victims of framework.We mustpay attentionto the specificsof local the world system, but they cannotfully understandall the cultureand social structureven thoughpeople in many relationshipsand processes affecting them. settings face common problemscaused by world system Not just the old ecological anthropologybut traditional expansion.To illustratethe importanceof local specificity ethnographyin generalalso propagatedthe illusion of iso- and of using a distinctivelyanthropological perspective, I lated, independent,pristine groups. By contrast,the link- I did and rec- ages approachemphasizes the embeddednessof communi- will returnto the social-soundnessanalysis ties in multiple systems of differentscale. Local people ommendationsI made for the USAID SAVEM project take their cues not just from neighborsand kin but also aimed at biodiversityconservation in five areasof Mada- from a multitudeof strangers-either directly or via the gascar.(The Tanosy case describedabove was drawnfrom media. Linkages researchcombines multilevel (interna- this analysis.)To maximizethe likelihoodof success, the tional, national,regional, local) analysis,systematic com- project'ssocial design forchangewas foundedin the tradi- parison,and longitudinalstudy (using modern information tionalsocial formsof each targetarea. technology).Challenging the traditionof the lone ethnog- The largeisland of Madagascarfeatures substantial eco- rapher,linkages methodology develops large-scale, explic- logical andcultural diversity, such that the size andcharac- itly comparativeteam projects (ideally involving interna- teristicsof affectedgroups varied with type of humaneco- tionalresearch collaboration). Ideally research is organized logical adaptation,from region to region and even within so thatas new forcesimpinge on the studyregion, they can the reservesand other protectedareas. The projecthad a be examinedin termsof theirdifferential effects on known site-specificdesign, recognizingthat affected groups ex- researchpopulations. Dealing with social transformation, isted at variouslevels andin differentregions. Members of the linkages perspective considers both the exogenous the project design team visited five protectedareas: the pressurestoward change and the internaldynamic of local AmberMountain complex, Beza Mahafaly,Ranomafana, .Unlike the old ecological anthropology(and tra- Andringitra,and Andohahela.The social characteristicsof ditional socioculturalanthropology in general), linkages each areawere chartedfor incorporationin projectdesign. projects study process, engage with history, considerthe To exemplify, I will describethe differentkinds of social role of political and economic power, and systematically groupsidentified to be involved in the projectfor the four 32 AMERICANANTHROPOLOGIST * VOL. 101, NO. 1 * MARCH1999

protectedareas I actuallyvisited: Ranomafana in the Tanala SurroundingAndringitra were at least 13fokontany (vil- homeland,Andringitra in Betsileo country,Andohahela in lage clusters),having about 10,000 total inhabitants.Each Tanosy-Tandroyterritory, and Amber Mountain in Tanka- fokontanyincluded smaller villages andhamlets, although ranacountry. there was a tendency towardsettlement centralization in the areabecause of the fearof cattlerustlers, who were said Ranomafana to use the forests to hide and dismemberthe cattle they steal. (Peasantsare also said to use the foreststo hide their RanomafanaNational Park is a protectedarea within the cattle from rustlers.)Around Andringitra the Betsileo vil- Tanalahomeland. The Tanalaare not a populousand thriv- lages lie to the north,and the Baravillages lie to the south. ing ,and this has become a transitionalzone I knew the traditionalsocial organizationand economy with considerableethnic diversity.Descendants of nine- of the Betsileo villages aroundAndringitra from my pre- teenth-centuryconquering armies from Imerina still live in vious researchin the 1960s. This is a relativelyrecently the area,along with more recentMerina migrants, includ- settled(nineteenth-century) addition to the Betsileo home- ing merchantsand slave descendantsfrom Antananarivo land. The local economy combines irrigatedrice cultiva- (the nationalcapital). The Betsileo, whose homelandlies tion with cattle pastoralism.Agriculture is less diversified just west, in the southcentralhighlands, have also been ex- here focused moreexclusively on rlce thanin the eastern, pandingand migratingto the Ranomafanaarea, along with central,and northern parts of Betsileo territory.The typical Tambahoakamigrants from the southeastcoast. Betsileo village near Andringitracontained branches of Social issues areproblematic at Ranomafanabecause of several(3 to 5) differentclans. The village foundersin this ethnicdiversity, continuing immigration, land poverty, and sparsely populatedand land-richarea were small family stratificationpatterns. Most of the immigrantshave come migrants from more densely populated Betsileo areas. as land-poorpeople-slaveZescended or free.Villages near They came in searchof land for theirherds and rice culti- the road are socially fragmentaryand disorganized,with vation.After the Frenchconquered Madagascar they were ethnic diversity,multiple unrelated families, and a higher joined by freed slaves from Betsileo countryand Imerina. than usual (for Madagascar)percentage (one third)of fe- All now consider themselvesBetsileo but maintaintheir male-headedhouseholds. Some villages more distantfrom differentclan (foko)affiliations and names. the road are more ethnicallyhomogeneous, offering more It was likely that projectimplementation would be eas- cohesive structuresand organizationsof potentialuse in ier aroundAndringitra than in Ranomafana.Both Betsileo implementingthe project thatis, in gaininglocal support, and Bara have solitarydescent groups,some arrangedin raisingenvironmental awareness, and channeling benefits. largerassociations (phratries). Ties of marriageand blood Given the extent of poverty,stratification, ethnic diver- siblinghoodlinked people in differentvillages and ethnic sity, and social fragmentationaround Ranomafana, project groups.Because irrigationwas traditionaland widespread, implementationneeded to be especially sensitive.The po- inputswould be appreciated.There was room for agricul- tentialfor noncomplianceand resistance was great.Tanala turaldiversification. Agricultural outreach seemed appro- and other horticulturalistswould be hurt by a prohibition priatefor this area.Descent grouplines could also be used on using the forest for slash-and-burncultivation. Land- to enlist support and channel benefits among the Bara poorpeople who used the forestto huntand gatherfor sub- aroundAndringitra. sistenceand sale would also be harmed.And those who re- lied on the forest to graze theircattle and hide them from Andohahela rustlerswould suffer,too. Most likely to benefitwere peo- ple with clear land claims whose fields mightbe improved Andohahelais locatednear Fort Dauphin on the south- by small dams, betterirngation, and otheragncultural in- east coast. Most of the reservelies in the traditionalhome- puts the "development"part of the Conservationand De- land (Anosy) of the Tanosy people. The reservehas two velopmentproject. mainethnic groups:Tanosy (the numericallypredominant group)in the east and Tandroyin the west. The mammoth easternpart of the reserve by far the largest at 63,100 Andringitra ha.-is separatedfrom the western part (12,240 ha.) by The Andnngitramountain area is a long-establishedre- nonreservelands where the Tanosy farm productiveirri- serve in the extremesouth of the Betsileo homeland.The gated rice fields. These f1eldsrely on the Andohahelafor- ethnic diversity aroundAndringitra is of a differentand ests for theirwater supply. Unlike Androy(Tandroy land) less problematicsort than that at Ranomafana.Two ethnic and the rest of the southeastcoast, Anosy is not an areaof groups(Betsileo and Bara)have villages nearAndringitra. strong emigration.Despite some deforestationnear Fort However, each village tends to be ethnicallyhomogene- Dauphin,population pressure on availableresources was ous. Nor are issues of stratificationand land poverty as less obvioushere thanat Ranomafanaor AmberMountain troublingas in Ranomafana. (see below). KOTTAK / THE NEWECOLOGICAL ANTHROPOLOGY 33

Ttle traditionalTanosy economy is diverse,with both the Anjoatsy (a mobile, seagoing group of spiritual-ritual swiddensand irngated rice fields. Roots and tubers (sweet specialists, with traditionalties to an informalversion of potato,taro, manioc) are also cultivated.Cattle is another Islamand ports on the east coast).The Anjoatsyhave spiri- focus of the traditionalTanosy economy and a matterof tualauthority at Ambohitra(Amber Mountain proper). The greatcultural interest, as it is amongthe Tandroyand WWF staff arrangedfor an Anjoatsympijoro (priest) to southernand western Malagasy generally. bless the park in a traditionalceremony. Similarly, at In implementingthis project (or any othercommunity- Ankarana,WWF enlisted the aid of the prince and the level projectin Madagascar),project personnel must un- power of traditionalAnkarana ritual to enhance coopera- derstandthe contrastbetween formal and informal struc- tion with projectagents. tures- betweenstructures and offices of thestate and those There are immigrantsthroughout the Amber Mountain of traditionalsocial organization. The latterwill oftenbe complex area.They includeMerina (still hatedin the area moreuseful than the formerfor projectgoals. Thus, the because of their nineteenth-centuryconquest of the fokontany(village cluster) president, a governmentoffice Tankarana),Betsileo (includingwoodcutters working for a foundthroughout Madagascar, is an electedofficial and commercialfirm that posed a threatto the forest), people administrator.His or (rarely) her authority varies, however, from the southeastcoast (Taimoro,Taisaka, Zafisoro, et fromplace to place.Traditional authority figures are often al.). There are also Sakalava(from the west and northern moreimportant that the fokontany president. In thosefok- coast),Tsimihety (from further south), and Comonans. For ontanywhere one cohesive group predominates, the person generationsthis has been an areaof coastaltrade (extend- chosento standfor election (and the sure winner) is some- ing to the Comorosand the EastAfncan coast),interethnic one with littlereal authority. He is a stoogefor the real contacts,and mixture.The town of Joffrevilleis a micro- powers the descentgroup elders. He is expectedto be cosm of the ethnic diversitythat exists in this region. Al- theiragent, errand boy, and foil in encounterswith the thoughit lackeddescent groups, we did identifysome eth- state. nic, religious,and school associationsthat might be used in Both Tanosy and Tandroyretain powerfuldescent projectimplementation, and people still heededthe ances- groups.Identification of descentgroup heads is vitalin im- tralritual authority of the tompotanyand their priests. plementingthis projectin Andohahela.Descent group Such site-specific analysisand recommendationsfor a headsmust give the project their blessing thusmaximiz- conservation-and-developmentproject illustrate that analysis ing the cooperationof the entiregroup. Descent group of social forms should not be subordinatedto approaches structurecan be usedto channelbenefits and spread infor- that emphasizethe environmentat the expense of society mation.All theethnic groups abutting on Andohahela have and culture,and ecology over anthropology.People must thesekind of structuresand leaders. The National come first. Culturalanthropologists need to rememberthe Departmenthas used them to distributeseedlings and gain primacyof society and culturein theiranalysis and not be cooperationwith its treeplanting programs. dazzled by ecological data.Funding sources that give pri- onty to the hardsciences, fund expensive equipment,and AmberMountain supportsophisticated technology should not lead us away from a focus on culturalspecificity and social and cultural The areaaround Diego Suarezin northernMadagascar variables.Ecological anthropologistsmust put anthropol- is the traditionalhomeland of the Tankarana(Antan- ogy ahead of ecology. Anthropology'scontribution is to karana).Like the Tanosy near Andohahela and the Tanala place people aheadof plants,animals, and soil. nearRanomafana, the Tankaranahave not expanded.The areais one of immigrationrather than emigration. Indeed, the Tankaranaseem to have retractedto theirmountain In Conclusion Romer'sRule homelandat Ankarana,where their prince (mpanjaka) still The paleontologistA. S. Romer (1960) developed the lives, holds court,and headsceremonies at his capital, rule that now bears his name to explain the evolution of Ambilobe.In a countrysuch as Madagascar,where many land-dwellingvertebrates from fish. The ancestorsof land foreignershave been deceivedby the claims of false animals lived in pools of water that dried up seasonally. princes,this is a realand effective prince. The project must Fins evolved into legs to enablethose animalsto get back payattention to him,his assistants,their customs, and their to water when particularpools driedup. Thus, an innova- ceremoniesin implementingthe project.Fortunately the tion (legs) thatlater proved essential to land life originated AmberMountain WWF staff took care (initially at least) to maintainlife in the water.Romer's lesson important to implementthe project in waysthat are culturally appro- for both the old and the new ecological anthropology is priate. that an innovationthat evolves to maintaina system can All areasof Madagascarhave traditional owners, called play a majorrole in changingthat system. Evolution occurs tompotany-mastersof the land.The Tankaranaare the in increments.Systems take a seriesof small steps to main- tompotanyfor the Diego Suarezarea. Also importantare tain themselves, and they graduallychange. Rappaport 34 AMERICANANTHROPOLOGIST * VOL. 101, NO. 1 * MARCH1999

recognizedRomer's lesson in his defitiitionof adaptation: A crucial vehicle for developmentresearch, including study "theprocesses by whichorganisms or groupsof organisms of both spontaneousand plannedsocial change, is the system- maintainhomeostasis in andamong themselves in theface atic integrationof data from longitudinalfield sites. Such sites of both short-termenvironmental fluctuations and long- allow analysis and evaluationof long-termtrends and effects, termchanges in thecomposition and structure of theiren- including cyclical changes relating to humanpopulations and vironments"(Rappaport 1971b:23-24, emphasis added). their ecologies, including the ecology of world systems and Romer'srule can be appliedto development,which, af- networks. terall, is a processof (planned)socioeconomic evolution. ApplyingRomer's rule to development,and here espe- ReferencesCited ciallyto ecologicallyoriented initiatives, we wouldexpect Ahmed, A. S. peopleto resistprojects that require major changes in their 1992 Postmodernismand Islam:Predicament and Promise. dailylives, especiallyones thatinterfere with subsistence New York:Routledge. pursuits.People usually want to changejust enoughto Barbour,Ian, ed. keep what they have. Motives for modifyingbehavior 1973 WesternMan and EnvironmentalEthics: Attitudes to- comefrom the traditional culture and the small concerns of ward Nature and Technology. Reading, MA: Addison- Wesley. ordinarylife. Peasants' values are not such abstract ones as Barth,Fredrik "learninga better way," "increasing technical know-how," 1958 Ecologic Relationships of Ethnic Groups in Swat, "conservingbiodiversity," or "makingthe worldsafe for NorthPakistan. American Anthropologist 60:107-89. democracy."(Those phrases exemplify intervention phi- 1969 Ethnic Groupsand Boundaries:The Social Organiza- losophy.)Instead, their objectives are down-to-earthand tion of CulturalDifference. London: Allen and Unwin. specificones. People want to improveyields in a ricefield, Bramwell, Anna amass resourcesfor a ceremony,get a child through 1989 Ecology in the TwentiethCentury: A History. New school,or be able to pay taxes.The goals andvalues of Haven,CT: Yale UniversityPress. subsistenceproducers may at timesdiffer from those of Bryant,B., and P. Mohai, eds. 1991 EnvironmentalRacism: Issues and Dilemmas. Ann peoplewho producefor cash,just as theydiffer from the Arbor:University of MichiganPress. interventionphilosophy of developmentplanners. Differ- Comaroff,John entvalue systems must be consideredduring planning. 1982 DialecticalSystems? History and Anthropology: Units Thisis onemore way of sayingthat (ecological) anthro- of Study and Questionsof Theory.The Journalof Southern pologistsshould not forget culture and people as theygrap- AfricanStudies 8:143-172. ple withcomplexity, comparison, and change. Change al- Costa, Alberto C. G., ConradP. Kottak,Rosane M. Prado, ways proceedsin the face of priorstructures (a given and John Stiles socioculturalheritage). The direction and nature of change 1995 Ecological Awarenessand Risk Perceptionin Brazil. is alwaysaffected by theorganizational material (sociocul- In Global Ecosystems:Creating Options throughAnthro- turalpatterns) at handwhen the change begins. Thus, cul- pological Perspectives.Pamela J. Putenney,ed. Pp. 71-87. turalways carmotbe regardedas blankchecks on which NAPA Bulletin 15. Washington,DC: AmericanAnthropo- logical Association. the environment,or history,can freelyand mechanically Douglas, Mary, and Aaron Wildavsky wnte. 1982 Risk and Culture:An Essay on the Selection of Tech- nical and EnvironmentalDangers. Berkeley: University of Notes CaliforniaPress. Epstein, T. Scarlett 1. This perspective was formalized at two Wenner-Gren 1978 Mysore Villages Revisited. In Long-Tenn Field Re- supportedconferences organized by Douglas White and held search in .George Foster, et al., eds. in La Jolla, California, in 1986. Participants,who became Pp. 209-226. New York:Academic Press. founding membersof Linkages:The World DevelopmentRe- Friedman,Jonathan searchCouncil, includedLilyan Brudner-White,Michael Bur- 1974 Marxism, ,and Vulgar Materialism. ton, Elizabeth Colson, Scarlett Epstein, Nancie Gonzalez, Man (n.s.) 9:421/1469. David Gregory,Conrad Kottak, Thayer Scudder, and Douglas Gezon, Lisa L. White. 1997 PoliticalEcology andConflict in Ankarana,Madagas- Linkages' goals include assisting in organizingand coordi- car. 36:85-100. nating basic scientific research on development on a world- Greaves, Thomas C. wide basis. This includes formulationof theory, testing of hy- 1995 ProblemsFacing Anthropologists: Cultural Rights and potheses, development of appropriatedatabanks for testing Ethnography.General Anthropology 1 :1, 34. theoretical formulations, monitoring change, establishing Green, Glen M., and RobertW. Sllssman trends, and identifying speciElclinkages or mechanisms in- 1990 DeforestationHistory of the EasternRain Forests of volved in social change, includingdevelopment interventions. Madagascarfrom Satellite Images. Science 248:212-215. KOTTAK / THENEW ECOLOGICAL ANTHROPOLOGY 35

Greenberg,James B., and Thomas K. Park Rappaport,Roy A. 1994 PoliticalEcology.PoliticalEcology 1:1-12. 1968 Pigs for the Ancestors:Ritual in the Ecology of a New Harris,Marvin GuineaPeople. New Haven,CT: Yale UniversityPress. 1966 The CulturalEcology of India's Sacred Cattle. Cur- 1971a Nature, Culture, and Ecological Anthropology.In rentAnthropology 7:51-59. Man, Culture,and Society. H. Shapiro,ed. Pp. 237-268. 1974 Cows, Pigs, Wars, and Witches:The Riddles of Cul- New York:Oxford University Press. ture.New York:Random. 1971b The Sacredin HumanEvolution. Annual Review of Kempton,Willett, James S. Boster, and JenniferA. Hartley Ecology and Systematics2:23X4. 1995 EnvironmentalValues in American Culture. Cam- 1994 Disordersof Our Own: A Conclusion.In Diagnosing bridge,MA: MIT Press. America:Anthropology and Public Engagement.Shepard Korten,D. C. Forman,ed. Pp.235-294. Ann Arbor:University of Michi- 1980 CommunityOrganization and RuralDevelopment: A gan Press. LearningProcess Approach.Public Administration Review Romer, Alfred Sherwood 9-10:480-512. 1960 Man and the Vertebrates,vol. 1. 3rd edition. Har- Kottak,Conrad P. mondsworth,United Kingdom: Penguin 1980 The Past in the Present:History, Ecology, and Cul- Roseberry,William turalVariation in HighlandMadagascar. Ann Arbor:Uni- 1988 Political Economy. Annual Review of versityof MichiganPress. Anthropology 17:161-185. 1990 Cultureand Economic Development.American An- Schneider,Jane thropologist92:723-731. 1977 Was There a Pre-CapitalistWorld System? 1999 Assaulton Paradise:Social Changein a BrazilianVil- Peasant lage.3rd edition.New York:McGraw-Hill. 6:2>28. Kottak,Conrad P., and E. Colson Sponsel, Leslie E., Thomas N. Headland,and RobertC. 1994 Multilevel Linkages: Longitudinaland Comparative Bailey, eds. Studies. In Assessing . Robert 1994 TropicalDeforestation: The HumanDimension. New Borofsky,ed. Pp.39S412. New York:McGraw Hill. York:Columbia University Press. Kottak,Conrad P., and Alberto C. G. Costa Steward,Julian H. 1993 Ecological Awareness,Environmentalist Action, and 1950 Area Research,Theory and Practice.New York: So- InternationalConservation Strategy. cial Science ResearchCouncil. 52:335-343. 1955 Theoryof CultureChange. Urbana: University of Illi- Kottak,Conrad P., Lisa L. Gezon, and Glen M. Green nois Press. 1994 Deforestationand BiodiversityPreservation in Mada- 1956 People of PuertoRico. {Jrbana:University of Illinois gascar:The View from Above and Below. http://infoser- Press. ver.ciesin.org/kiosk/publications/94{)005.txt. Sllssman,Robert W., Glen M. Green, and Linda K. Sussman Mintz, Sidney 1994 Satellite Imagery, , Anthropology, 1985 Sweetness and Power:The Political,Social, and Eco- and Deforestationin Madagascar.Human Ecology 22: nomic Effects of Sugaron the ModernWorld. New York: 333-354. Viking. Wallerstein,Immanuel Nash, June 1974 The Modern World-System: Capitalist Agriculture 1981 EthnographicAspects of the WorldCapitalist System. and the Origins of the EuropeanWorld-Economy in the AnnualReview of Anthropology10:393423. SixteenthCentury. New York:Academic Press. Pepper,David Wolf, Eric R. 1984 The Roots of Modern Environmentalism.London: 1982 Europe and the People without History. Berkeley: CroomHelm. Universityof CaliforniaPress.