Global Ethnography Author(s): Zsuzsa Gille and Sen Riain Source: Annual Review of Sociology, Vol. 28 (2002), pp.

. 271-295 Published by: Annual Reviews Stable URL: http://www.jstor.org/stable/3069243 . Accessed: 22/05/2011 15:04
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=annrevs. . 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 support@jstor.org.

Annual Reviews is collaborating with JSTOR to digitize, preserve and extend access to Annual Review of Sociology.

http://www.jstor.org

Copyright(

Annu. Rev. Sociol. 2002. 28:271-95 doi: 10.1146/annurev.soc.28.110601.140945 2002 by AnnualReviews. All rightsreserved

ETHNOGRAPHY GLOBAL
ZsuzsaGillel and Sean 0 Riain2
IDepartment Sociology, Universityof Illinois, Champaign-Urbana, of Illinois 61801; e-mail: gille@uic.edu 2Department Sociology, Universityof California,Davis, California95616; of e-mail: sporiain@ucdavis.edu

relations transnationalism, Key Words globalization, local-global * Abstract Globalization poses a challengeto existing social scientificmethods of inquiryandunitsof analysisby destabilizing embeddedness socialrelations the of in particular communities places.Ethnographic are globalizedby meansof and sites variousexternalconnectionsacrossmultiplespatialscales andporousandcontested Globalethnographers boundaries. must begin their analysisby seeking out "placethatseek to definenewkindsof places,withnew definitions social of makingprojects" relationsandtheirboundaries. studiesof globalprocessestend Existingethnographic to clusterunderone of threeslices of globalization-global forces, connections,or kindof place-making The imaginations-eachdefinedby a different project. extension of thesitein timeandspaceposespractical conceptual and forethnographers, problems butalso politicalones.Nonetheless, locatingthemselves by firmlywithinthetime and the can howglobalprocesses spaceof socialactors "living global," ethnographers reveal arecollectivelyandpoliticallyconstructed, the of demonstrating variety waysin which is grounded the local. in globalization

INTRODUCTION
Globalizationposes a challengeto existing social scientificmethodsof inquiryand units of analysis by destabilizingthe embeddednessof social relationsin particular communitiesand places. By locating themselves firmly within the time and can space of social actors"livingthe global,"ethnographers revealthe socioscapes thatpeople collectively constructof global processes (Albrow 1997), thusdemonstratinghow globalizationis groundedin the local (Burawoyet al. 2000). At the same time, globalizationalso poses problemsfor ethnography. The potentialand uneven delinking of the spatial and the social under conditions of globalization claim to understand social relationsby being there and thus upsets ethnography's demandsthat we rethinkthe character global ethnography.1 of
how can we undertake (1995) poses a similarquestionto anthropology: ethnog1Appadurai raphyin a world where locality is contestedand shifting. 0360-0572/02/0811-0271$14.00 271

272

GILLE

6 RIAIN

Globalizationhas exploded onto the sociological agendain the past 10 to 15 years.2The first generationof globalizationstudies was concerned with how to historicalcontinudefineglobalization;which aspectsof globalizationrepresented ity and discontinuity;and how to theorizethe relationshipbetween globalization and modernity,postmodernity,and postcoloniality.As such, these studies were the concerned primarilywith understanding characterof globalization as a social phenomenon(Beck et al. 1994, Featherstone& Lash 1999, Giddens 1991, Harvey 1990, Robertson 1992-see Guillen (2001), Lemert (2002), and Waters (1995) for reviews of this generationof studies).Morerecently,however,scholars have begun to ask what implicationsthese sociohistoricalchanges may have for social science itself, andthey haveaddressedthe metatheoretical, epistemological, and political implicationsof thatolder body of literature (Abell & Reyniers2000, Albrow 1995, Gane 2001, Hargittai& Centeno 2001, Kilminster1997, Pieterse 2000, Tsing 2000). A few studieshave even startedtheorizingthe natureof global ethnography(Amit 2000, Burawoy 2000a, Ethnography2001). Our goal in this review is to analyze, evaluate, and synthesize the achievementsof these newer responsesto the challengesglobalizationposes to sociological inquiry,especially as they apply to qualitativemethodsand fieldwork. In the first section of the paper we analyze existing attemptsto re-define the character social relationsin an era of globalization.We pay special attentionto of betweensociety andspaceis theorizedandtheimplicationsfor how therelationship wherethese theorists of sites. Whereis the "there" ourunderstanding ethnographic should be? Some advocate replacing ethnographic imply global ethnographers place-basedsites with locations within networksand flows, within transnational social formations,or at the bordersof places where difference is produced.We argue that place still provides a foundationfor global ethnographers,but as a can location from within which ethnographers explore the sociopolitical projects thatare remakingsocial relationsand places. Here we'll differThe second section reviews a series of global ethnographies. the entiateamongthreeperspectivesof globalization: globalas forces,connections, and imaginations.The thirdsection turnsto methodologicalissues raised by our We sites as politically constructed. explore the impliredefinitionof ethnographic cations of working in sites that extend across multiple places and spatial scales, thatextend in time, and the boundariesof which are deeply contested.

2Accordingto the CambridgeSociological Abstracts,between 1985 and 1990 twenty-nine of the studiesabstracted assignedglobalizationas a keyword,in contrastto the 14 abstracted between 1965 and 1984. There were 240 such works in the first half of the 1990s, 410 in 1995 alone, anda whopping985 in 1998. Todaythe cumulativetotalof sociological studies (includingreviews) addressingglobalizationis 4,876. While these dataof course can only be suggestive, the trend is clear: From 1990 there is a sudden increase in sociological efforts studyingglobalization,culminatingin close to a thousandannualpublicationsand conferencepapersby the new millennium.

GLOBALETHNOGRAPHY

273

REIMAGINING THE SOCIAL IN GLOBALETHNOGRAPHY

is Ethnography uniquelywell placed to deal with the challengesof studyingsocial life underglobalizationbecause it does not rely on fixed and comparableunits of research.However,it also faces significant analysis, as do surveyand comparative challenges in reconfiguringitself for a global era-ethnography explicitly seeks to analyze the social by locating the researcherin the space of the social relations access the socialby going to the being analyzed,andthisabilityto straightforwardly local becomes problematicunderconditionsof globalization.In the following we will surveyrecentattemptsin sociology to redefinethe social underglobalization.

Disembeddingthe Social?
The conventional postwarsocial science view assumesthatthe nationis a container for everythingwithin it, while international relationsare assumedto accountfor all relationsoutside of the national.Even in world-systemstheory,the subunitsof the system are almost always nations,whose relationshipto each otheris ordered tends to accept by capitalistdevelopmentandinterstatecompetition.Ethnography these categories-either, as in sociology, generalizingto the national society or, as in anthropology, takingthe local as the site of culture,which is often analyzed in terms of its relationshipto the world of nations (colonialism, nation-building, etc.). However, thematic approachesto globalization identify a new empirical or phenomenonthathas undermined, at least destabilized,these establishedhierarchiesof the local, national,and international. Even as debaterages bitterlyover the precise meaningand extent of globalization,includingthe destabilizationand transformation the nation state, the decenteringof nationalsocieties is increasof ingly widely accepted.In ourdefinition,buildingon Mato's(1997) typology,globalizationsignifies the increasingsignificanceof trans-localrelations,local-global relations.3 relations,andglobal-globalrelationsat the expenseof national-national Some scholarsclaim thatglobalizationfundamentally reordersthe classical relationshipbetween self and the other, society and knowledge (Beck 1992, Beck et al. 1994), local and global (Dickens 1992, Hall 1991a, 1991b), and most
3Matoclassifies local, national,transnational, global agents,basedon the impactof their and whose social activity (1997:170-71). Local agents are those individualsand organizations practices are mainly concentratedin the same locality in which they are based, although from time to time they develop these practicesbeyond this locality and maintainrelations with social agents from abroad.Depending on the level of analysis this locality may be regardedas a small town or community,a system of towns or communities, a state or province, or a subnationalregion. National agents are those whose practicesare regularly developed at nationallevels. Transnational agents are those whose practices are regularly borders.Globalagentsare a subclassof transnational developedacrossinternational agents whose practicesare regularlydevelopednot just transnationally at worldwidelevels. but

274

GILLE 6 RIAIN between space and society. Accordingto these authors,we mustredeimportantly fine the concept of the social itself. Giddens (1991) arguesthat underconditions of globalizationsocial relationsare disembeddedfrom the local and can operate in contexts where space no longer mattersbecause shared systems of symbols andknowledgecirculateglobally.While disagreeingwith Giddensaboutthe modernistnatureof globalization,Albrow(1995) latertakesup this termto arguemore broadlythatsocial relationsaredisembeddedfromspace;Albrowarguesfor a conas that cept of "globality" a new level of organization has no organizingagent,thus freeing sociality from state control. Altvater& Mahnkopf(1997), arguingmore strictlyfrom an economic perspective,define globalizationas the culminationof the disembeddingof economy from society (the originalmeaningof the word in Polanyi), leading to a world marketunbound. While we find that these social theoristshave a propensityto exaggeratethe extent of this disembeddingprocess and to ignore its unevenness, we grant the claim thatglobalizationbreaksthe one-on-onemappingof the local onto the social. This in turn, makes the classic ethnographicstrategyof being there much more Before we addressthis issue, let us see how sociologists haverewoven problematic. the brokenthreadconnectingthe spatialand the social.

The Social as Flow or Network

Lash & Urry (1994) argue that this disembedded"social" is increasingly constituted by flows of people, information,goods, and particularlysigns or cultural symbols. For them the startingpoint of a "sociology after societies" (Urry For 2000) is these "mobilities,"replacingthe hallowed concept of "community." the (1990), the entitiesthat"flow"around worldare"scapes"or cultural Appadurai formationsaroundfinance,media, ideologies, technologies, andpeoples. Yet anothergroup of scholars relies on the networkas their central concept. Hannerz(1992) sees society as constitutedby "networksof networks,"down to networksamongindividuals.ForCastells(1997) the networksarebetweenplaces, and a space of flows is being superimposed upon, andreplacing,a space of places. Those places left outsidethe space of flows areprofoundlydisadvantaged their by structural exclusion. WhereasCastellsmay not have fully elaboratedthe implications of the networkmetaphorand did generallyignore the literature networks on in sociology (Abell & Reyniers2000), a newer set of studies consciously borrows the networkconcept from economic sociology and talks of a new geographyand the need to drawnew maps (Am.Behav. Sci. 2001). For all theirdifferences,each of these approachesdisconnectsthe social from any particularplace-seeing contemporarysocial relations as characteristically tend to reify these networks,flows, and stretchingacross places. Such approaches othermobilities (Urry 2000) as themselves defining society. Despite recognizing thatnetworkscanbe exclusionary, these approaches providelittle analysisof power relationswithinnetworksandthereforefindit difficultto explainreproduction and in networks. Such explanationsrequirethat place-based resources and change

GLOBALETHNOGRAPHY

275

these approachesneglect the processes be includedin the analysis. Furthermore, of actors and their sense-makingactivities as forces in shapingthe flows agency themselves. While the network is at least activatedby and even defined by the connectionsamongactors,the conceptof flowsposits a worldof disembodiedflows of information,signs, finance,and otherresources-it is the actor'sconnectionto these flows that defines the actor, not how they activate connections as in the In networkmetaphor. such a view, places disappear entirely.In orderto go beyond the oppositionof "thespaceof flows versusthe spaceof places"(Castells 1997), we of mustdevelopourunderstanding how places andnetworksconstituteone another, rather thanseeing themas opposingprinciplesof social life. This conceptualization of the social as fluidity,mobility,andconnectednessdooms the local to a stoppage in such flows, perceivesit as a staticplace, rendering methodsfocusing on concrete places irrelevant.With Sassen (2000), we see globalization as a repattemingof fluidities and mobilities on the one hand and stoppages and fixities on the other, ratherthan an all-encompassingworld of fluidity.4

The Social as Transnational

Othershave sought to retainthe insight that cross-nationalnetworksare increassocial relations ingly significantwhile still providingan analysis of the structured withinthose networks(Keamey 1998). Schiller(1997) definestransnational studies as the study of varioustypes of border-crossings people, texts, discourses,and by at representations various geographicallevels. Portes et al. (1999) limit transnationalism to activities of immigrants,migrants,transnational and entrepreneurs, culturalgroups from the sending countryregularlytravelingto entertainemigre communitiesin recipientcountries.The authors a transnationportray thus-defined alism as the antithesisof globalization,which is understoodas a projectinitiated and by and for the benefit of multinational corporations by nation states. For these authors(Porteset al. 1999, Schiller 1997, Smith & Guamizo 1998), the emphasisis on studyingglobalizationor transnational social relationsas a new to subjectmatter,andas sucha considerableamountof conceptualrigoris required delineatewhat might be properlydeemed a qualitativelynew social development or institutionworthyof a novel term.Porteset al. (1999) define transnationalism, a narrower term than the subjectmatterof Schiller's transnational studies would suggest, as an emergentresearchfield, in fact a measurableobject of inquiry. The conceptof transnationalism providesmanyinsightsinto the strategicaction of social actorsdevelopinga new scale of social activitybetween the nationaland the global. However, this comes at the expense of a historical view that would reveal how fields of activities (including stages of production,the distributionof culture,regulatorymoves, etc.) are transformed a new division of laboramong by differentscales. Some activities that were once the legitimateprovince of actors
4Thisview also contrastswith Harvey's (1990) much referencedidea of a universaltimespace compression.

276

GILLE a 6 RIAIN

at the nationallevel may now have shiftedto the scope of authorityand expertise of global actors, while other activities may now have moved from the global to level of local levels, and so on.5 On the one hand, the focus on the transnational analysis is a necessary conditionof these often revealingstudies and, admirably, Smith & Guamizo (1998) providea clear analyticaland methodologicalbasis for their preferredresearchprogramof comparativetransnationalisms (pp. 24-29). the social formation However,while the focus is on understanding transnational how the relationship itself, it is difficult to move beyond this site to understand scales of social activityis being reorganized. among multiple

The Social as BorderZone

Otherauthorsare more concernedwith social relationsat the bordersand boundaries of social orders. Marcus & Fischer (1986) opposed the imagery of global versus local with a view of still distinctculturalworlds increasinglyin communias Their"anthropology cultural cationwith one another. soughtto explore critique" therecombinant, hybridformsof culturallife thatwere emergingat these boundary points of culturesin contact with one anotherand enhancingthe possibilities for othersocieties to provideus with tools for culturalcritiqueof our own society. However,conceiving of the social as a borderzone and emphasizingconnections and contacts means that the cultural worlds that come into contact with worldswith readily each otherarestill conceptualizedas self-contained,territorial identifiabledifferencesthat then clash. Conceiving of the social as a borderzone are often implies thatboundary-localities liminal, hybrid,syncretic,and fluid, an assumptionthatcan only hold if we abstractaway from the powers thatcreateand of maintainboundaries.AlejandroLugo's (2000) ethnography the United StatesMexico border,for example,promotesa view of bordersnot as spaces of mobility and hybriditybut as places producedby increasedsurveillanceat the borderand Berdahl's(1999) Wherethe WorldEnded is also discipline in the maquiladoras.6 that an ethnographyof an actual borderland: dividing the formerEast and West of Unlike MarcusandFischer,she calls for anunderstanding boundaries Germany. not as escaping but ratheras deeply enmeshedin existing social and power relais tions. Her borderland also less a place of diversity,fluidity,and hybriditythan a "placeof intense and inflexible lucidity,"a site in which variousfeaturesof the economic, cultural,andpolitical regime on one side of the borderare crystallized and manifestin theirpurestform. Both the East Germans'and Mexican workers' experiences of being stuck on one side of the bordercall forth other imagined boundaries,racial, ethnic, and both internaland external.Based on these ethnographies,instead of the assumptionof liminality and hybridity(Kearney 1998),
5Sassen(2001) for exampletalks aboutthe productionof new legal subjectsas local actors treatiesandappealto supranational areincreasinglyableto applyinternational organizations or in their strugglesagainsta corporation a nation state. 6Lugoexplicitly takes issue with Morales' applicationof Foucault'snotion of heterotopia to the experienceof living in borderzones.

GLOBAL ETHNOGRAPHY 277 we call for a focus on "theproductionof local differences,"the political processes throughwhich places are produced,when studying social action at the margins (Gupta& Ferguson 1997).

The Social as Place-MakingProjects

Each of these approachesto redefiningthe relationshipbetweenthe social andthe local is limited, but each offers a differentavenue for future conceptualization. From the discussion of flows, we see the need to redefine place in light of the multiple connections cutting across places. From the study of transnationalism, we see the criticalimportanceof the emergenceof new scales of social action and the reconfiguringof relationshipsamong the multiple scales within which places are embedded. Finally, from the study of borders,we see the vital importance of seeing place as politically producedand contested. Togetherwe can combine these various threadsinto a concept of the social as increasingly embroiled in place-makingprojects7that seek to redefinethe connections, scales, borders,and characterof particular social orders.These projects are the places and particular critical sites throughwhich global ethnographers interrogatesocial relations can in an era of globalization. Wehavearguedthatplace continuesto be centralto globalethnography, albeitin a reconceptualized form. Our startingpoint in tacklingthese challenges therefore is to build a revised definition of place that builds in large part on geographer Doreen Massey's concept of a global sense of place. Massey proposes a concept of locality based on fourkey arguments: places arenot static, (b) places do not (a) have the kind of boundariesthat warranta simple counterposition the outside, to (c) the identity of a place is not homogenous, yet (d) places are unique and their specificity resides in the distinct mixture of local and wider social relations. In short,the locality-the site-is historicallyproducedin interactionwith a variety of external connections, and this process also produces distinctive patternsof inequalityinternalto the locality.Together,these propositionsform the basis of a global sense of place (Massey 1994). Albrow (1997), similarly to Massey-and quite in opposition to the authors advocatingthe idea of the social as networkand flows that imply the fixity of the local-argues that communities in globalized places are fluid and scape-like as well. People may live in the same neighborhoodor town, but their meaningful social lives may reach beyond that locality to a highly uneven degree. These reaches or networks constitute what he calls sociospheres, "distinctpatternsof social activitiesbelongingto networksof social relationsof very differentintensity, spanning widely different territorialextents, from a few to many thousands of miles"(Albrow1997,pp. 51). Whilepriorto the contemporary of globalization, era these sociospheres usually intersectedin the locality, new intersectionsare now
7Theconcept of place-makingappearsin Gupta& Ferguson(1992) and Tsing (2000), but Appadurai's(1995) notion of the productionof the local conveys a similaridea.

278

GILLE* 6 RIAIN forming that he calls socioscapes. Socioscapes are fluid imaginationsof spatial belonging and of the social formationscreatedby and makingpossible the reach of social relations beyond the locality. Taken together,Massey's and Albrow's is contribution to advancea new sense of place that,while it is relevantto localities before this era of globalization,is all the more salienttoday. Such concepts of place treatall places as if they occupied the same level in a However,as we have seen, at differentperiodsof history,social spatialhierarchy. actionis legitimatedat differentscales. Neil Brenner(1999, 2000) arguesthatany of understanding space as socially constructedrequiresthat we take into account externalconnectionsandbordersof a place butalso how it is locatedin not only the the "politicsof scale"-the negotiationof the hierarchyandlegitimacyof different era scales of social action.Brennerarguesthatthe contemporary of globalization not simply of a shift of power and of social interactionupwardfrom the consists nationalto the global but of a destabilizationof the existing hierarchiesof spatial scales. While creating a crisis in national social formations,this also opens up opportunitiesfor social actors to develop new combinationsof local, national, and transnational, global social relations. in This spatialanalysis is echoed in recent discussions of social structure socienterdirectlyinto dialogue.In contrastto the two literatures rarely ology, although as treatinghigh levels of social structure compositionistviews of social structure, lowerlevels of social the aggregateof lower levels, andhierarchical views, treating structure the effects of higherlevels, Kontopoulos(1993) arguesfor a view of as in with differentlevels integrating differentways as social structures heterarchical, but in a structured unevenmanner. Given the densely intertwinedmultiplelevels of analysis, how can we hope to from particular social relationsand social structures places? We argue interrogate that the key to this is to recognize that these shifting socio-spatialrelationsbetween levels of analysisarethemselvessocially andpoliticallycontested,not only in academia,but also in the real world. This interestin the political contestation interestin the productionof differencebetween of place extends anthropologists' & Ferguson 1992, Marcus& Fischer 1986). Tsing (2000) argues, places (Gupta for example,thatscholarsassume a global level of analysis at theirperil andmust begin their analysis by seeking out place-makingprojectsthat seek to define new Places kindsof places, withnew definitionsof social relationsandtheirboundaries. matterbecause it is in places that we find the ongoing creation, institutionalization, and contestationof global networks,connections, and borders.Insteadof a comprehensiveaccount of a self-containedset of social relations,the ethnographer now uses her location to interrogatea variety of intersectingplace-making spatiallocation.Reflectingchanges projectsas they are manifestedin a particular the in the worlditself, locationin place is crucialfor understanding social relations is thatextend beyond it. The ethnographer less a chroniclerof self-evidentplaces of than an interrogator a varietyof place-makingprojects.How the ethnographer analyzes the intersectionof scales depends in parton her position in a particular enables us to set of place-makingprojects.Such a concept of global ethnography

GLOBALETHNOGRAPHY

279

make sense of the variety of ethnographiesdealing with global processes and to classify them according to how they identify their subjects' relations to certain place-makingprojects.

GLOBAL ETHNOGRAPHY
is Ethnography anespecially suitablemethodologywith which to investigatesocial structuresthat are constituted across multiple scales and sites. Even the most sophisticatedstatisticalmethods tend to rely on a nested hierarchyof scales and units of analysis, whereas ethnographycan strategicallylocate itself at critical points of intersectionof scales and units of analysis and can directlyexamine the negotiationof interconnectedsocial actors across multiple scales. In this section we outlinethis contribution reviewinga varietyof globalethnographies by through the lens of three slices of globalization-global forces, global connections, and global imaginations(Burawoyet al. 2000)8. In studiesof global forces the social actorsandplaces being studiedare caught in a place-makingprojectconstitutedwell beyondtheirinfluencethatcan hardly up be shaped by them-although they may develop complex forms of adaptation, avoidance, and survival. Studies of global connections show how certain social actorsare able to take advantageof the destabilizationof sociospatialhierarchies to centeredon the nation-state buildnew translocalandtransnational connections. The social actorswho constructglobal imaginations-a less easily classifiedgroup than those in the other two slices-are most explicitly engaged in place-making, contesting definitions of local, national, and global scales and of the relations among them. None of these three "slices" of globalization can be understoodin terms of a simple shift upwardin power, a relocation of power from the national to the global, buteach reflectsone way in which the global orderas complex multiscalar social place-makingprojectsis experienced.For this reason, studies of particular phenomenatend to cluster togetherundera particularslice-studies of modernization underforces, of migrantsunderconnections,and of political initiativesof environmental protection,regionaldevelopment,andespeciallyculturalconsumption and productionunderimaginations.
8Thereare perhapsonly a handful of 'global ethnographies,'narrowlydefined.However, we review a wide rangeof studieshere in orderto illustratesome of the potentialdirections which global ethnographymight follow. Many of the studies are anthropologicaldue to the importanceof ethnographywithin the discipline and the rethinkingof anthropological concepts which has been promptedin particularby post-colonial thoughtand which has focused anthropologicaltheories on globalizationfor quite some time now. A relatively small portion of sociological ethnographyaddressesthe kind of macro processes that we discuss in this review but we have included some studies (such as KatherineDudley's The End of the Line) in order to illustratehow they are shaped by particularaspects of globalization,even if the study as a whole could hardlybe called 'global ethnography.'

280

GILLE? ( RIAIN

Global Forces
Most ethnographicanalyses of global forces begin with a constructionof an externalforce or overarchingstructure-capitalism, modernity,science-which is then examined at work within the site(s) being studied. These ethnographiesat theirbest reveal not just the impactof an impersonalforce but also how localities are made penetrableby forces, how localities assimilate these forces into their own socioscapes, and how forces are resisted, accommodatedto, and fled from. These forces are in reality also place-makingprojects but are analyzed in these ethnographiesfrom the point of view of those caught up in these projects with little ability to influencethem. The globalizationprojectis itself boundup in manyways with the expansionof capitalistsocial relations(McMichael1996). Extensiveresearchhas exploredhow global capitalism shapes working conditions and politics, focusing particularly on women workers. Initial research documentedthe integrationof third world women into the global factory and revealed the lived experience of these new industrialworkers(FernandezKelly 1983, Mies 1986, Nash 1979, Nash & Kelly 1983). Globalizationof productionmight be seen as unambiguousexploitationof has women. However,ethnography revealeda muchmorecomplex disempowered Controlstrategiesvary,resistanceis widespreadand takes many formspicture. even withinthe same locality (Ong 1987, Salzinger2000) or at the core (Cho 1985, Hossfeld 1990). The effects of globalizationareoften contradictory-many young women, exploited at the factory,were still able to use their new-foundincome to relationsin the household(Wolf 1992) and to build avoid the worst of patriarchal selves basedaroundconsumptionpatterns alternative 2000, Mills 1999). (Freeman of the closure of a Chrysler car factory in KathrynDudley's (1994) study Kenosha,Wisconsin,revealshow, throughforces beyondtheircontrol,auto workers not only lose theirjobs but also their place in the world as they come to be looked down upon by the newly confidentprofessionalsof the town as dinosaurs who can't adaptto a global informationeconomy. Otherforms of globalizationhave only complicatedthe effects of global capiand talism. Female migrantworkersfindnew lifestyle opportunities access to the into moder pleasuresof life even as they are incorporated undoubtedlyexploitative relationsas workers(Constable 1997, Mills 1999). Even global sex workers, often situatedat the intersectionof theirown migration,abjectpoverty,andglobal of researchto strategizewithin patterns tourism,arerevealedby close ethnographic the overwhelmingconstraintsof their situation(Kempadoo& Doezema 1998). effects of global forces is theirintersectionwith One sourceof the contradictory division of laborbringsnot only capitalistexploitaotherforces-the international tion but also modernistconsumption.Modernityis itself a projectof an emergent in world polity (Boli 1997). But the modem too can be appropriated different ways in differentcontexts (Freeman2000, Miller 1995, Mills 1999). This is not only linked to privateconsumption-workers of all kinds seek to professionalize their work on the global stage (Constable 1997, Freeman2000, Salzinger 1991).

GLOBAL ETHNOGRAPHY 281 Modernitybrings with it also the scientizationof everydaylife-Scheper-Hughes (1992) provides a graphicaccount of how the medicalizationof the condition of povertyin Brazil serves to obscurethe sourcesof thatpoverty(1992). The studies of in Whiteford& Manderson(2000) documentboth the universalizingcharacter healthpolicy and the diversityand inequalityof its local applications. global The moder also serves as a powerfuldiscourse throughwhich local concerns or are expressed, appropriated, erased. Donham (1999) showed how the Marxist revolutionin Ethiopiawas shapedby local strugglesbetween Ethiopianmilitary and intelligentsiaover who would lead the countryinto the progressive,modern future.Ferguson (1994) studied a very differentincarnationof modernityin the form of a World Bank plan in Lesotho. His ethnographicstudy of the failure of this plan revealshow the use of the developmentdiscoursedepoliticizes a variety of local issues and strengthensthe state. In a more recent book, Ferguson(1999) documentsthe collapse of the faith in developmentand how it is replacedby an aggressive neoliberalism,claiming to be no less modern, but reconfiguringthe relationshipbetween state, population,and global economy. These studies reveal a numberof aspects of forces that structural analyses do not. Any global force relies on enablinglocal conditionsin orderto take hold in a of location-Haney (2000) documentsthecriticalimportance Hungarian particular welfare neoliberalreformsof the Hungarian social scientistsin the IMF-prompted Otheranalyses show how the global fails to reshapethe local in intended system. ways yet createsnew forms of domination(Ferguson1994). Identitiesassociated with prioreras of global or nationalpower may be critical to the experience and globalization,as revealedin Gowan's(2000) studyof the politics of contemporary of homelessnessin SanFrancisco.Ethnographic analysesof forces can experience be particularly revealingregardingthe mechanismsthroughwhich global forces operate, offering a view of reality that goes beyond a simple dichotomy of a powerfulforce confrontedonly by the weapons of the weak.

Connections
In contrastto analystsof globalforces,writerson globalconnectionshavetypically focused on the agencyof social actors-in fact, as we have seen, writerson transnational connections explicitly position themselves against the overly determinist point of analyses analyses of globalizationtheorists(M. Smith2001). The starting of connectionsis typically a type of strategicaction or a groupthatexhibits, or is even definedby, strategicbehavior.Migrantsareperhapsthe prototypical case, but traders,social movements,tourists,technicalandotheroccupationalcommunities also figureprominentlyas criticalfiguresin makingconnectionsacross social and political borders(Kyle 2000). The strategiesof these variousactorsare explicitly tied to the makingof new places butlargelythroughthe strategicactionof individuals or networks,ratherthanthe collective politicizationof global imaginations. Whereassome studiesof transnational migrantcommunitieshavereformulated such as communityat a new spatialscale (Portes traditional sociological concepts

282

GILLE 6O RIAIN et al. 1999), others have suggested that these transnational networks require a rethinkingof the categories of community and center-periphery (Rouse 1991). The collection of studies in Smith & Guamizo (1998) explore the construction of economic and political identities in transnational networks, the effect of the localities on local and nationalpolitics, and the posemergence of transnational culture.Ong (1999, 1996) describes the sibility of a cosmopolitantransnational productionof a new and rathermobile group of citizens acting on behalf of a distinctAsian mode of globalization.Otherstudies explore the reconstruction of ethnic and racialidentities (Ethn.Racial Stud. 1999), the emergenceof a variety of tradingand migrationbrokers(Kyle 2000), the genderingof migrationcircuits (Hondagneu-Sotelo1994), and the continuingsignificanceof ties to the sending region in shapingthe experiencein the host country(George2000). Thereareoftenclose linksbetweentheliterature migrationandthaton traders on and occupationalcommunities.Where factoryworkersare often seen in termsof forces, artisansare often seen in termsof theirtradingconnections,implying that the latterhavemore agency,althoughthey arenot necessarilybetteroff financially. Craftsmen (Cook 1998), artists,andmusicians(Kyle 2000, Marcus& Myers 1995, & Steiner 1999) capitalizeon local cultureand the taste for the exotic in Phillips global markets,while challengingexisting ideas of ethnic or nationalcultureand authenticity. The emphasis in the connections literatureon the emergence of new social on spaces is perhapstakento its logical conclusion in the literature virtualworlds can & Centeno2001). This is a uniquekind of place whose participants (Hargittai choose to log in or out relativelyfreely and can adopta wide varietyof identities communitiescontinueto be structured (Turkle1995). Even so, internet-based by local and nationalinstitutions(Hakken2000) in ways that may be obscuredby a focus on the communicationin cyberspacealone. This may narrowethnographic researchsolely on serve as a more generalwarningagainstfocusing ethnographic the relationsinternalto networks. Althoughoften respondingto global forces, the agency of social actorsin their responses is typically seen in terms of global connections.This is clearestin the literatureon transnational social movements.Thayer (2000) traces how feminist theoriestravelthroughtransnational social movementspace and are transformed in thatprocess,ratherthansimply impactinguponlocalities. Elsewhereshe argues that transnational feminism appearsmuch more like a connection than a force: the bureaucratic reachof fundingagencies and NGOs may widen in the "Though of creatingtransnational relations,local movementsare also empowered, process as well as discursiveand materialresources, and honing their gaining solidarity feminist public"(2001, p. 268). negotiatingskills in the transnational While it would be most appropriate, relatively few studies examine global commodity chains through the lens of connections. An exception is Parrefias (2001) who documentshow the transnational inequalitiesin both workandquality of care that underpinwhat Ginsburg& Rapp (1998) call stratifiedreproduction send migrantFilipina mothersto work as childcareand domestic workersin the

GLOBAL ETHNOGRAPHY 283 United States while creatinga demandfor childcarefor their own childrenin the Phillipines. of Whereasthepreviousethnographies globalconnectionstendedto concentrate studiesof on agency frombelow, thereis a greatdeal of potentialfor ethnographic from above, thatis, of how global connectionsproduceglobal forces. Such agency studies could cast light on the place-makingprojectsof forces such as capitalism andmodernity, only fromthe perspectiveof those who makethese projects,rather than from the perspectiveof those who are made by them. Over time, as world polity institutionssuch as the United Nations and the WorldTradeOrganization of proliferate,new potentialsites emerge for the ethnographer global connections and forces who is willing to spend time with the corporateand political elites. of thereis very little ethnography these agentsandthe social relations Interestingly withinwhich they areembedded.In some ways, however,these sites arerelatively for unproblematic globalethnography-these arenew locales operatingon a global boundaries arerelativelyeasily and scale, buttheyhaverelativelyclearinstitutional identifiable.This is potentiallya very rich vein of research-andone that is often more open to sociologists than we might think, given the huge numbersof social science consultantswithin these world polity institutions(see Falk Moore 2001 and Goodman2001 for good examples of building a global ethnographyaround such a consultingrole). Neither does the analysis of global connections exclude an ethnographyof Chinatowninvestigatesthe transformation a of place. Lin's (1998) Reconstructing place by globalizationthroughthe lens of global connections(see M. Smith2001, pp. 191-92, for an insightful discussion of Lin's book as an example of global ethnography).The book reveals that Chinatownin New York,often assumed to be internallyhomogenous,is riven by internalconflicts often rooted in historical connections beyond the enclave. But this is not simply a case of historical ties likely to wane as assimilation occurs. In fact, Chinatown'sfuture is negotiated throughnew strategiesfor connectingto the worldoutside-through tourism,new business relations,and connectionsto the state.

Imaginations
In studiesof global imaginations,the local activelyparticipates public discourse in about what globalizationmight look like. Local and nationalpolitics in many instances turns into a battle of competing visions of the global. The construction of a global vision has tangible implications for the outcome of a conflict. First of all, references to global ideas and actors today provide an entranceticket to in participating public discourse,and those unwilling or unableto formulatetheir claims in global termsoften find themselves invisible. Second, when local actors wage their battles with claims about the global, to acquiremore credibilitythey themselves build connectionsto outside actors and enterglobally circulatingdiscourses. This not only sends an importantsignal that the concrete local meaning of globalizationis up for grabs,but it also strongly shapes the circle of potential

284

GILLEI 6 RIAIN allies andenemies. In short,explicit place-makingthroughthe politicizationof the global scale is increasinglycentralto a multitudeof differentpolitical projects, whetherthey conceive of themselves in these termsor not. Lopez (2000) provides an example of what happens when a movement fails to develop a global imagination,as Pittsburgh'sservice workers' union proved incapableof counteringthe local goverment'svision of Pittsburghas a deregulated global city. Klawiter(2000) shows how a new social movementemergedto challenge a globally circulatingmedical discourse aboutbreastcancer.The most radicalwing of thismovementposes a challengenot only to stigmatization also but to the explanationof what causes breastcancer and how it should be dealt with. In the process it generatesthe powerful concept of a global incineratorindustry spewing toxic pollutantsinto the environment,into food and thus into women's bodies. In Gille's (2000) case study of a siting controversyarounda Westernincineratorplanned in a small village in southernHungary,both the pro- and antiincinerator of campswaged an intensebattlewith competingunderstandings what Europe and joining the EuropeanUnion mean and therefore what they imply for how local wastelands,both figurativeand literal, should be cleaned up. De Soto's (2000) ethnographyof a battle aroundthe recultivationof a heavily polluted industrialregion located in the middle of formerEast Germanyalso identified two competing visions-a modernistvision, supportedby various funds of the EuropeanUnion, that would create a high and supposedly clean-tech industrialparkinterspersed with green parksandlakes, andthe vision supported by the Bauhaus/green group,which would reachback to local historicaltraditionsof environmental renewalbut would also preservethe monumentsof socialist industrialization.De Soto arguesthatneithervision capturesthe needs of the residents living there, aptly symbolizing how debates about the nature of the East/West dichotomy can take place above the heads of citizens. Global imaginationsmay constructplaces thereforein ways that are quite differentfrom global forces and connections,even if at othertimes these threeelementsof globalizationaredeeply intertwined. Interestingly,in these studies the scales being contested were the European and local scales-not the global. Indeed, the EuropeanUnion, as an increasingly muscularsupranational actoracting on an increasinglysignificantscale of social has beenthe subjectof manyrecentstudies.Geographer JohnAgnew (2001) action, that the productionof the Padaniaregion in NorthernItaly, as influenced argues with the European Union, also relies on a selective collage by legal harmonization of local/regionalmyths, memories,and symbols. In each of these studieswe see the importanceof the politics of scale, in which actors contest the privileged scale of social life (Brenner2000, M. Smith 2001, Tsing 2000). As powerful actors' claims to exclusively representthe global are challenged,the choice of geographicalscale at which the battle should be waged becomes a crucial strategic issue for social movements, which seem a natural target and subject of global ethnography.But it is equally importantto grasp

GLOBAL ETHNOGRAPHY 285 how elites producetheirimaginations,where these imaginations ethnographically derivetheirpower,andhow they findresonancewithimaginationsforgedat various geographicalscales. of Goldman's(2001) ethnography the WorldBankprovidesan exampleof such a process.Goldmanfollowed consultantsto Laos andVietnamto observehow they decided whatconservationandothergreenprogramsthe WorldBankshouldfund. Their constructionof what we would call eco-scapes producedactualecological conditions,as they importconservation projectsfromelsewhere.Theirdivisionof a zones andfishingzones led to dislocation country'slandscapeinto rice-cultivating and a profoundchange not just of people's relationshipto and reliance on nature but also of existing ecological and agricultural boundaries. This vision of globalimaginationscould shednew light on issues of nationaland racialidentitiesandplaces constructed fromimagination. Darian-Smith Eve (1999) used the ChannelTunnel,which links EnglandandFranceandits environs,as both site (she lived neartheTunnelin England)andsymbolforexploringtherelationship between Britishnationalidentityand Europeanidentity.Berdahl's(1999) already mentionedWherethe WorldEnded is also an ethnographyof place-makingin an She demonstrates actualborderland. how formerEast Germans,the Ossies, keep remakingtheir place and their identity as necessitatedby their experience of the changingpolitical and economic relationswith formerWest Germany. Whetherthroughstudies of social movements,elites, or borders,the study of global imaginationsreveals how taken for grantedsocial hierarchiesof spatial scales are being actively contested and reconfigured.Most obvious in studies of the explicit politicizationof the national,Europeanor global, global imaginations of provide a vital addeddimensionto the more widespreadunderstanding globalization as constitutedby forces and connections.

ISSUESIN GLOBAL ETHNOGRAPHY

An ethnographicapproachto globalizationrequiresthe understanding locally, of socially, and culturally specific ways in which people understandthe place of their locality in the global scheme of things, and the actions they take to shape that place. These understandingsand actions are deeply political, as we have seen, and the very definitionof the ethnographer's topic and site is shapedby the site place-makingprojectswithinwhich anyparticular is embedded.Globalization involves the contesting of the boundariesof places and negotiationsconcerning which geographicalscale is best suited for action. As a result, the choice of site also becomes political. What sites does the ethnographer choose to study?Where does she draw the boundariesof her site? Which events and processes at which These questions bear geographicalscales will shape the ethnographicnarrative? on the global ethnographer. heavily In this section we trace some of the majorchallenges facing global ethnograacrossmultiplesites and scales; the phersin light of the extension of ethnography

286

GILLE? 6 RIAIN of in extension of ethnography time throughthe incorporation historicalanalysis; of the global ethnographer. and the changing subjectposition

AcrossSites and Scales Extendingin Space:Ethnography

In his 1998 volume Ethnography ThroughThickand Thin,George Marcusargues thatthe studyof an increasinglyglobalizedworldrequiresmultisitedethnography or at least a multisitedresearchimaginary.He proposes to extend fieldworkto multiplesites, extendingthe researchsite in space, in orderto study localities tied to the outside world in complex and consequentialways. Multisitedresearchis chains,paths,threads,conjunctions,orjuxtapositionsof locations designedaround establishes some form of literal, physical presence, in which the ethnographer with an explicit, posited logic of associationor connectionamong sites thatin fact of defines the argument ethnography. The extension of ethnographyto multiple sites is a seemingly excellent way to meet the challenges posed by globalizationto place-based studies. However, while finding connections is certainlynot difficult, deciding which of them are a worth pursuing seems somewhat arbitrary, feature of multisited ethnography when he calls it a kind of constructivism thatMarcusacknowledges (1998, p. 90). Marcusidentifiesthe methodsof such construction: connectingsites by following plots, stories,allegories;following conflicts people;following objects,metaphors, discovWhatties togetherfieldworklocationsis the ethnographer's andbiography. of tracesand clues, herlogic of association.The methodologicalimperativeof ery being thereis replacedby that of chasing things around,things that are identified interestsprior to enteringthe field than by the field more by the ethnographer's Marcusprovideslittle guidanceanda ratherscantylist of models thatwould itself. allow us to answerpressingquestionssuch as how to identify all the relevantsites or what weight each of the sites should carry. This dilemma is only aggravatedif we recognize that ethnographicsites are networkof multiplesites of always and everywhereembroiledin an intermeshing social action, operatingacross multiple spatial scales and levels of social structure.Historically,sociology has been largely guilty of what, afterDoreen Massey (1994), we may call the confusion of the level of analysis and geographicalscale theoretical Even theoristswho have made important with the level of abstraction. of to contributions the sociological understandings globalizationtendto equatethe (Harvey1990, Robertson globalwith the universal,andthe local with the particular researchsites we cannota priori 1995, Wallerstein1991). However,in multiscalar identify a dominantlevel of analysis. How do we identify the limits of a community we are studyingwhen the communityis constitutedacrossa varietyof spatial scales (local, national,global, transnational etc.)? This interweavingof ethnographicsites across a range of spatial scales, and thereforeunits of analysis, poses serious challenges to establishedethnographic practice-challenges thathave barelybeen raisedwithin sociology as a whole let in While we do not deny the place of constructionism alone amongethnographers.

GLOBAL ETHNOGRAPHY 287 research,sociologists are likely to find Marcus'definitionof a multisitedresearch imaginaryto be wantingprimarilybecause social relationsamong sites can never be reducedto the connectionsforgedby the ethnographer's imaginationand logic of association. Conceiving of ethnographicsites as internallyheterogenousand connectedto otherplaces by a myriadof social relationsrequiresthatthe extension of fieldworkto severalsites be dictatednot by the logic of the ethnographer by but the character these social relationsthemselves, both within and between sites. of we Among the ethnographies havereviewed,only a smallnumberpursuea multisited approach.The ethnographiesof transnationalism discussed above provide useful examples of the multisitedapproachand show where it works well-when there are clearly defined patternsof connection between relatively highly concentratedsites, e.g., in the case of studyingmigrants.Donham's (1999) study of in Marxistmodernization Ethiopiainvolvestwo sites, butthese aresites thatcan be understoodas clearly located within a well-establishedhierarchyof scales-the national capital, which is the center of mobilization of modernizationprojects, and a village where they are implementedand contested. Such situationswhere clearly definedrelationsexist between two sites or acrosstwo scales arerelatively rare,however,and do not addressthe dilemmasof investigatingthe multiple sets of externalconnections, bordersand scales within which global places are often located. Eade and his collaborators(1997) take a differentapproach-fixing their ethnographicsite in London but using a variety of studies carriedout by a large researchteam to explore the multiple connectionsof their researchsite to many partsof the world. The need to pursue actors throughspace and time in order to explore placemaking projects seems likely to increase our use of interviews, history, tracing networks, and so on and to decrease our time spent simply being on site. The classic model of extended stays in a site extends to multisitedethnographyquite poorly-for the reasonswe havediscussedandalso for the purelypracticalreasons of time and other resources(see V. Smith 2001 for a revealingdiscussion of the demandsof the "workof ethnography" the "ethnography work").Becoming in of partof a site remainsa criticalpartof ethnography-the issue of gainingentry-but the very natureof thatmembershipchanges for the ethnographer it changes for as those aroundher or him. Place becomes a launchingpad outwardinto networks, backwardinto history and ultimatelyinto the politics of place itself.

Extending in Time: From Context to History

Social relations change, and with them, the sites in which they are embedded. Marcus'multisitedethnography, however,ignores the dynamicprocess by which sites aretransformed theirexternalconnections.Furthermore, multisitedethnogby raphydoes not allow us a sufficientamountof criticalattentionto political efforts to naturalizethe local community because it provides no space from which to notice that such constructionoccurs. It takes places for granted and leaves no room for accountingfor the productionand transformation sites. In short, in of

288

GILLE? O RIAIN ratherthan a factor that multisitedethnography, history remainsan afterthought has implicationsfor what can be seen as a site (Gille 2001). Because we focus on dynamic social relationsratherthan static sites and see our to localities as politicallyandhistoricallyconstructed, approach globalethnographyrequiresthe historicizationof the locality and of local and extralocalsocial relations.Social relationsare,by definition,dynamic,and as such they affect how the issues we study manifest themselves at a particulartime. In sociology, historical analysis has too often implied unilineardevelopment,as spatialvariation is translatedinto temporalvariationor stages in social change (see Dove's 1996 on critiqueof world systems analysis). In the literature globalization,this unilinin a model of a (non-Western) local simply reactingto the earity expresses itself (Western)global-while the global remains apparentlyunaffectedby the localglobal relation(e.g., Miller 1995). Wonderfulcorrectivesto this view of history and local-global relations are aboutwho encompasseswhom MarshallSahlins's(2000) upsettingethnographies, in the process of development, and the work of the Comaroffs (1992). Their practice of historical ethnographydefies the imposition of the microsociology/ macrosociology dichotomy on the local/global one. "Even macrohistoricalprocesses-the buildingof states, the makingof revolutions,the extension of global capitalism-have their feet on the ground.Being rooted in the meaningfulpractices of people, greatandsmall,they arein short,suitablecases for anthropological this treatment" (1992, p. 33) Furthermore, ethnographichistorytreatsthe agents of such events (missionaries,settlers,etc.) as a complex collectivity with its own history and practiceswith which they attribute meaning to their participationin events. Mary Des Chene finds that this one or anotherof these macrohistorical historical ethnographyis still too bound to a particularlocality. As she argues, re"spatialcontiguityis not essential to every kind of historicalanthropological search"and "the field may not be a place at all, but a periodof time or a series of to events, the study of which will take a researcher many places"(1997, p. 71). Both Des Cheneandthe Comaroffsacceptarchivalresearchas partof historical However,both redefineits role in a novel way-the formertreating ethnography. the archivesthemselvesas a site, while the lattertreatit as a sourceof textualtraces can of a field of arguments,which the ethnographer reconstructwithout having record.While much of comparativeto accept archivesas objectivedocumentary historical sociology has relied on historical documents, letters, diaries, official record,etc., this approachrequiresthat sociologists reevaluatearchivalresearch. of researchcan be a powerful entranceinto such reinterpretation Ethnographic stories archival materials,oftenaidedby analysesof social memory.Contemporary regardinghistoricalevents can be measuredagainst the archivalrecordto reveal how historical events are reconstructedand contested as part of contemporary can culture. While ethnographicinterpretation be aided by historical materials, the detailed knowledge of the contemporarysite can cast new light on archival materialsand the intendedand unintendedconsequencesof historicalactions. Of course, history can be incorporatedmore directly into ethnographicresearch,albeit over a more limited time span. Donham's(1999) study of Marxism

GLOBAL ETHNOGRAPHY 289 in Ethiopiaprovidesa fascinatingexampleof a historicalethnography-only this in case incorporating revisits to sites of earlierfieldwork.9 While such revisits to the site of an earlierethnography very demandingof individualethnographers, are one researchis to revisit the to introducea historicalcomponentto ethnographic way sites (or similar sites) studied by other ethnographers. Burawoy (2000b) argues that while revisits typically seek to debunkthe original study,they may be better as deployedto analyzehistoricalchange-using the earlierethnography historical data, as occurredthroughgood fortuneto Burawoyhimself throughhis research on a Chicago factorypreviouslystudiedby Donald Roy.

Transformations Ethnographers: of How Do Relationswith Those We Study Change?

The extension of the ethnographicsite in space and time sharpensone's sensibilities to the political consequencesof defininga site or sites. The subjectposition of the ethnographer global ethnography fraughtwith difficulties.Whatmight in is be the implicationsof the methodologicaldilemmasdiscussed above for the political position of the ethnographer?Does the classical problematic of gaining entry and maintainingtrustworthiness change in this new type of ethnography, in which the field is composed of multiple sites and the social relationsin which the sites are enmeshed?Marcusprovides an interestingevaluationof the shifting assumedidentitiesof the observer.He claims that doing ethnography multiple in sites amountsto activism,rather thanjust to mereethnography. the example Using of Emily Martin's(1994) Flexible Bodies, he arguesthatin a kind of ethnography in which one is a medical studentin one site, an AIDS volunteerin another,and a traineein the third,the ethnographer findherselfsometimesworking will corporate with the subjects she observes, sometimes againstthem. These conflicts, according to Marcus, are resolved "not by refuge in being a detached anthropological scholar,but in being a sort of ethnographer-activist" 98). This circumstantial (p. activist figure appearsto be superiorto the "traditional self-defined activist role claimed by left-liberalscholars"(p. 98) perhapsbecause it activelyparticipates in and negotiatesamong differentsets of subjects(Gupta& Ferguson 1997). However,this mobile activism can be expected when the sites are linked only by the connections forged in the researcher'simagination.In Martin'scase, for example, the link is a metaphoricone-that is, she follows the metaphorof the immune system through various locales, but the sets of subjects she connects are not necessarily in a concrete conflict with each other (though they may have
gDonham spenttime in Addis Adaba,the Ethiopiancapital,andMaale, an areain Southern Ethiopia. With major visits in 1974-75, as the revolution began, and in 1983-84, after the revolutionarystate had consolidated itself, Donham spent a total of three years in Ethiopia,spreadacrosstwenty-five.Partlythroughgood luck, arrivingin Maalejust before the revolution,andpartlythroughdedicatedandtime-consumingfieldwork,Donhamis able to give concreteform to the historicaltransformation not only Ethiopiansociety but to of the meaning of Marxismand modernityitself within Ethiopia.

290

GILLE 6 RIAIN differing perspectives on the particularissue) and may not even know of each other.In this case, maintainingthe position of an AIDS activist is relativelyeasy, as long as the targetsof AIDS activism do not know aboutthe commitmentsand in presence of the ethnographer the other locations. But how could one maintain any meaningfulsense of activism when the sites are in actual conflict with each other,when one is visibly affiliatedwith certainof the partiesin conflict andwhen a consequentlyopposingpartiesare more likely to considerthe ethnographer spy for theiropponents?Gille (2000, 2001) foundherselfin such a situation.She went into the field expecting to conduct her researchon the politics of location of a waste incineratorin an open and reciprocalmanner.In contrast,she found the conflict aroundthe incineratorhad become so bitterthat she had to revise many of her expectationsregardingher position as an ethnographer-relatively mundane mattersas where to live, compensatinginterviewees,discussing local affairs,and so on were deeply politicized. Even if not located in such a precariouspolitical spot, the ethnographer who is following connections througha network will almost inevitably find herself also makingconnections.For to be in a networkis also to practicenetworkingas such is the mode of existence in decentralizednetworks.The ethnographer, someonewith manyconnectionsacrossmultiplecliques, is also likely to become a potentiallyvaluableresourcefor otherswithinthe network-as well as a potentially dangerousfigurewho knows too much. when the The relationbetween ethnographer subjectsis also transformed and is ethnographer much more accessible to many of their subjects,even when they returnto their computerin the university.The relationof the anthropologist who from the far off village, and of the sociologist who returnsfrom the inner returns that city,to peoplein theirsites is verydifferentfromtherelationship exists between in transnational social movements and the researcherwho studies them activists (see Bickhamet al. 2001 for a discussion of similarissues in a programdesigned to create dialogue between academicsand activists). We are left then with a vision of the global ethnographer extendingherself in time and space to attemptto uncoverthe multilevelsocial processes at workin her site. But these extensions take a toll-they must be negotiatedand relationships or must be managedthatare often contradictory even directlyconflictual.Thereis no easy answerto the dilemmas of power in global ethnography-if anything,it is less clear for whom the ethnographer should speakunderthese conditionsthan it has seemed in the past.

CONCLUSION
Our travels towardglobal ethnographyhave taken us across many terrainsand betweensociety andspace,the importance connections-the changingrelationship of locatingthe site in relationto place-makingprojects,the varyingways in which have sliced into the study of globalization, and the dilemmas of ethnographers undertakingethnographyunder these conditions. We have argued for a global

GLOBAL ETHNOGRAPHY 291 ethnographythat still locates itself firmlyin places but which conceives of those places as themselves globalized with multiple external connections, porous and contested boundaries,and social relations that are constructedacross multiple spatial scales. The place-boundsite becomes a platformfrom which a variety of place-makingprojectscan be investigated.These projectsprovide a way in to the investigationof heterarchicalsocial structuresand deeply intertwinedscales of social life. We have suggested that global ethnographers have investigatedsuch sites as homes to particularplace-makingprojects, leading to the experience of Whichof these globalizationsas eitherglobal forces,connections,or imaginations. in a given ethnographic has much to do with the topic experiencesprevails project being studiedandthe social locationof those social actorsbeing studied.Finallywe well suitedto uncovering have arguedthatwhile ethnography may be particularly in theirmultipleandoverlappingcontexts, significantchallenges global processes The exist for the would-be global ethnographer. extension of the site in time and andconceptualproblemsfor ethnographers also political but space poses practical ones. Groundingglobalizationthroughethnographywill present challenges and dilemmasbut also rich rewards. The Annual Reviewof Sociologyis online at http://soc.annualreviews.org
LITERATURE CITED Abell P, Reyniers M. 2000. On the failure of social theory. Br. J. Sociol. 51(4):73950 Agnew J. 2001. [No title.] Presentationat session on 'Citizenship and Identity' Annu. Meet. Am. Sociol. Assoc., Aug. 17-20 Albrow M. 1995. The Global Age: State and Society beyondModernity.Cambridge,UK: Polity Albrow M. 1997. Travellingbeyond local cultures:socioscapes in a global city. See Eade 1997, pp. 20-36 AltvaterE, MahnkopfB. 1997. The worldmarket unbound.In TheLimitsofGlobalization: Cases andArguments,ed. A. Scott, pp. 30626. London/New York:Routledge American Behavioral Scientist. 2001. Special issue on Globalization.44(10) Amit V, ed. 2000. Constructingthe Field. London/ New York:Routledge A. Appadurai 1995. The productionof locality. In Counterworks: ManagingtheDiversity of ed. R Fardon,pp. 204-25. LonKnowledge, don/New York:Routledge A. Appadurai 1990. Disjunctureanddifference in the global culturaleconomy. Public Cult. 2(2):1-24 Beck U. 1992. Risk Society: Towardsa New London:Sage Modernity. Beck U, Giddens A, Lash S. 1994. Reflexive Modernization.Stanford:StanfordUniv. Press BerdahlD. 1999. Wherethe WorldEnded:ReUnificationand Identityin the GermanBorderland.Berkeley:Univ. Calif. Press BickhamMendezJ, Wolf D. 2001. Wherefeminist theory meets feminist practice:bordercrossing in a transnationalacademic feminist organization. Organization (4):72350 Boli J. 1997. Worldculturein the world polity: a centuryof international non-governmental organization.Am. Sociol. Rev. 62(2):17190 BrennerN. 2000. The urbanquestionas a scale question:reflectionson Henri Lefebvre, urban theory and the politics of scale. Int. J. UrbanReg. Res. 24(2):361-78

292

GILLE

RIAIN Dove MR. 1996. Rice-eating rubber and people-eatinggovernments:peasantsversus statecritiquesof rubberdevelopment coloin nial Borneo. Ethnohistory 43(1):32-63 Dudley KM. 1994. The End of the Line: Lost Jobs, New Lives in PostindustrialAmerica. Chicago:Univ. ChicagoPress EadeJ, ed. 1997. Livingthe Global City:Globalizationas Local Process. New York:Routledge Ethnic and Racial Studies. 1999. Special issue on Transnationalism. 22(2) Ethnography.2001. Special issue on Global 2(2) Ethnography. Falk Moore S. 2001. The international production of authoritative knowledge: the case of West Africa. Ethnography drought-stricken 2 (2):161-90 FeatherstoneM, Lash S, eds. 1999. Spaces of Culture: City, Nation, World. London/ ThousandOaks, CA: Sage FergusonJ. 1999. Expectationsof Modernity. Berkeley:Univ. Calif. Press FergusonJ. 1994. The Anti-PoliticsMachine: 'Development,' Depoliticization, and Bureaucratic Power in Lesotho. Minneapolis: Univ. Minn.Press Fernandez Kelly P. 1983.For WeAreSold,I and and MyPeople: Women Industryin Mexico's Frontier.Albany:State Univ. NY Press FreemanC. 2000. High Tech and High Heels in the Global Economy: Women,Workand Pink-Collar Identities in the Caribbean. Duke Univ. Press Durham/London: Gane N. 2001. Chasing the 'runawayworld': the politics of recent globalization theory. Acta Sociol. 44(1):81-89 George S. 2000. "Dirtynurses"and "menwho miplay:"gender and class in transnational gration. See Burawoy et al. 2000, pp. 14474 GiddensA. 1991. TheConsequencesof Modernity.Cambridge: CambridgeUniv. Press in Gille Z. 2001. Criticalethnography the time of globalization: toward a new concept of Methodol.1(3):319-34 site. Cult.Stud.-Crit. in Gille Z. 2000. Cognitive cartography a European wasteland:multinationalcapital and

Brenner N. 1999. Beyond state-centrism? and geographicalscale Space, territoriality, in globalizationstudies. TheorySoc. 28:3978 BurawoyM. 2000a. Introduction: reachingfor the global. See Burawoy,et al. 2000. pp. 1-40 BurawoyM. 2000b. The ethnographyof revisits. PresentedatUniv.Chicago,May 11, 2000 Burawoy M, Blum JA, George S, Gille Z, Gowan T, et al. 2000. Global Ethnography: Forces, Connectionsand Imaginations in a Postmodern World. Berkeley: Univ. Calif. Press BurawoyM, BurtonA, FergusonAA, Fox KJ, Gamson J, et al. 1991. Ethnography Unbound.Berkeley:Univ. Calif. Press CastellsM. 1997. TheInformation Age. 3 vols. Oxford,UK: Blackwell Cho SK. 1985. Laborprocess and capital modibility: the limits of the new international vision of labor.Polit. Soc. 14(2):185-222 Comaroff J, Comaroff J. 1992. Ethnography and the Historical Imagination. Boulder, CO: Westview Constable N. 1997. Maid to Order in Hong Kong: Stories of Filipina Workers.Ithaca/ London:CornellUniv. Press Cook S. 1998. Mexican Brick Culture in the Building of Texas, 1800s-1980s. College Station:Texas A&M Univ. Press Darian-SmithE. 1999. Bridging Divides: the Channel Tunneland English Legal Identity in the New Europe. Berkeley: Univ. Calif. Press Des Chene M. 1997. Locating the past. See Gupta& Ferguson1997, pp. 66-85 De Soto HG. 2000. Contested landscapes: reconstructing environment and memory in postsocialist Saxony-Anhalt.In Altering States: Ethnographiesof Transitionin EasternEuropeand theFormerSoviet Union,ed. D Berdahl,M Bunzl, M Lampland,pp. 96113. Ann Arbor:Univ. Mich. Press Dickens P. 1992. Society and Nature:Towards a GreenSocial Theory.Philadelphia: Temple Univ. Press Donham D. 1999. MarxistModern. Berkeley: Univ. Calif. Press

GLOBALETHNOGRAPHY greens vie for village allegiance. See Burawoy et al. 2000, pp. 240-67 GinsburgF, RappR, eds. 1998. Conceivingthe New World Order:TheGlobalPolitics of Reproduction.Berkeley:Univ. Calif. Press Goldman M. 2001. The birth of a discipline: producing authoritativegreen knowledge, World Bank-style. Ethnography 2(2):191218 Gowan T. 2000. Excavating "globalization" from streetlevel: homeless men recycle their pasts. See Burawoy et al. 2000, pp. 74105 GuillenM. 2001. Is globalizationcivilizing, destructiveor feeble? A critiqueof five key debates in the social science literature.Annu. Rev. Sociol. 27:235-60 GuptaA, FergusonJ, eds. 1997. Anthropological Locations. Berkeley:Univ. Calif. Press GuptaA, FergusonJ. 1992. Beyond 'culture': space, identity,andthe politics of difference. Cult.Anthropol.7(1):6-23 HakkenD. 2000. Cyborgs@ Cyberspace?An Ethnographer Looks to the Future. New York/London: Routledge Hall S. 1991a. The local and the global: globalization and ethnicity. See King 1991, pp. 19-40 Hall S. 1991b. Old and new identities,old and new ethnicities. See King 1991, pp. 41-68 Haney L. 2000. Global discourses of need: mythologizing and pathologizingwelfare in Hungary.See Burawoyet al. 2000, pp. 48-73 HannerzU. 1992. The global ecumeneas a network of networks.In ConceptualizingSociety, ed. A Kuper, pp. 34-58. London/New York:Routledge HargittaiE, Centeno MA. 2001. Defining a global geography.Am. Behav. Sci. 44(10): 1545-60 Harvey D. 1990. The Conditionof Postmodernity:An Inquiryinto the Origins of Cultural Change. Oxford:Blackwell Hondagneu-SoteloP. 1994. Gendered Transitions: MexicanExperiencesof Immigration. Berkeley:Univ. Calif. Press Hossfeld K. 1990c. 'Theirlogic againstthem': contradictions in sex, race, and class in

293

Silicon Valley. In Women Workers and Global Restructuring, K Ward,pp. 149ed. 78. Ithaca,NY: CornellUniv. Press in KearneyM. 1998. Transnationalism California andMexico at the end of empire.In Border Identities:Nation and State at International Frontiers,ed. TM Wilson, H Donnan, pp. 117-41. Cambridge:Cambridge Univ. Press Kempadoo K, Doezema J, eds. 1998. Global Sex Workers: Rights,Resistanceand Redefinition.New York/London: Routledge KilminsterR. 1997. Globalizationas an emergent concept.In TheLimitsof Globalization: Cases and Arguments, A. Scott, pp. 257ed. 83. London/NewYork:Routledge King AD, ed. 1991. Culture,Globalizationand the World Conditions System:Contemporary for the Representation Identity.Binghamof ton: StateUniv. NY, BinghamtonPress Klawiter M. 2000. From private stigma to the global assembly:transforming terrainof breastcancer. See Burawoy et al. 2000, pp. 299-334 KontopoulosKM. 1993. The Logics of Social Structure. York: Cambridge/New Cambridge Univ. Press Kyle D. 2000. TransnationalPeasants. Baltimore:JohnsHopkinsUniv. Press Lash S, Urry J. 1994. Economies of Signs and Space. London:Sage LemertC. 2002. Social Things:An Introduction to Sociological Life. New York:Rowman& Littlefield Lin J. 1998. Reconstructing Chinatown: Ethnic Enclave,Global Change.Minneapolis:Univ. Minn. Press Lopez S. 2000. Contestingthe global city:Pittsburgh'spublicserviceunionsconfronta neoliberalagenda.See Burawoyet al. 2000, pp. 268-98 Lugo A. 2000. Theorizingborderinspections. Cult.Dyn. 12(3):353-73 MarcusGE. 1998. Ethnography ThroughThick and Thin. Princeton, NJ: Princeton Univ. Press Marcus GE, Fischer MJ. 1986. Anthropology as Cultural Critique: an Experimental

294

GILLE a 6 RAIN Essays on Women and Globalization, ed. A Hochschild, B. Ehrenreich.New York: Nation Books. In press PhillipsRB, SteinerCB. 1999. UnpackingCulture: Art and Commodityin Colonial and Postcolonial Worlds.Berkeley: Univ. Calif. Press Pieterse JN, ed. 2000. Global Futures: Shaping Globalization. London/New York: Zed Books Portes A, Guanizo L, EdwardsL, Landolt P. 1999. Introduction: pitfallsandpromiseof an emergentresearchfield. EthnicRacial Stud. 22(2):217-37 Robertson R. 1992. Globalization. London: Sage RobertsonR. 1995. Globalization:time-space In and homogeneity-heterogeneity. Global ed. M Featherstone, S Lash, Modernities, R Robertson,pp. 25-44. London/Thousand Oaks, CA: Sage Rouse R. 1991. Mexican migration and the social space of postmodernism.Diaspora 1(1):8-23 SahlinsM. 2000. Culturein Practice: Selected Essays. New York:Zone Books SalzingerL. 1991. A maid by any othername: of work"by Central thetransformation "dirty American immigrants. See Burawoy et al. 1991, 139-60 sexual subSalzingerL. 2000. Manufacturing desireanddisciplineon a jects: 'harassment', Maquiladorashopfloor.Ethnography 1(1): 67-92 of SassenS. 2000. Spatialitiesandtemporalities the global: elements for a theorization.Public Cult. 12(1):215-32 Sassen S. 2001. [No title.] Presentationat Session on 'Citizenship and Identity' Annu. Meet. Am. Sociol. Assoc., August 17-20 N. WeepScheper-Hughes 1992.Death Without The Violenceof EverydayLife in Brazil. ing: Berkeley:Univ. Calif. Press Schiller NG. 1997. The situation of transnational studies.Identities4(2):155-66 Smith M. 2001. TransnationalUrbanism:Locating Globalization. Oxford, UK: Blackwell

Moment in the Human Sciences. Chicago: Univ. ChicagoPress MarcusGE, Myers FR, eds. 1995. The Traffic Art in Culture: Refiguring andAnthropology. Berkeley:Univ. Calif. Press Martin E. 1994. Flexible Bodies: Tracing Immunity in American Culture From the Days of Polio to the Age of AIDS. Boston: Beacon Massey D. 1994. Space, Place and Gender. Minneapolis:Univ. Minn. Press Mato D. 1997. On global and local agents and the social making of transnationalidentiAmerica. ties andrelatedagendasin "Latin" Identities4(2):167-212 McMichaelP. 1996. Globalization:myths and realities.Rural Sociol. 61(1):25-55 on andAccumulation Mies M. 1986. Patriarchy a WorldScale: Womenin the International Division of Labor.London:Zed Books Miller D. 1995. Worlds Apart: Modernity the Through Prismof theLocal. London/New York:Routledge Mills MB. 1999. Thai Womenin the Global LaborForce:Consuming Desires, Contested Selves. New Brunswick, NJ: Rutgers Univ. Press Nash J, 1979. WeEat the Mines and the Mines Eat Us: Dependencyand Exploitationin Bolivian TinMines. New York:ColumbiaUniv. Press Nash J, Kelly P, eds. 1983. Women,Men and the New International Division of Labor Albany:State Univ. NY Press Ong A. 1987. Spiritsof Resistanceand Capitalist Discipline: Factory Womenin Malaysia. Albany:StateUniv. New YorkPress ChinaandmoderOng A. 1996. Anthropology, nities: the geopolitics of culturalknowledge. In The Future of AnthropologicalKnowledge, ed. H L Moore, pp. 60-92. London/ New York:Routledge Ong A. 1999. Flexible Citizenship: The Cultural Logics of Transnationality.Durham: Duke Univ. Press RS. Parrenias 2002. The carecrisis in the Philipfamilies in pines: childrenand transnational the new global economy. In Global Woman:

GLOBALETHNOGRAPHY Smith MP, Guamizo L, eds. 1998. Transnationalismfrom Below. New Brunswick, NJ: Transaction Smith V. 2001. Ethnographiesof work and In the work of ethnographers. Handbookon Ethnography,ed. P Atkinson, A Coffey, S Delamont,L Lofland,J Lofland,pp. 220-33. Los Angeles: Sage Thayer M. 2000. Traveling feminisms: from embodied women to gendered citizenship. See Burawoy et al. 2000, pp. 20334 feminism:readThayerM. 2001. Transnational ing JoanScottin the Braziliansertao.Ethnography2(2): 243-72 Tsing A. 2000. The global situation.Cult.Anthropol. 15(3):327-60

295

TurkleS. 1995. Life on the Screen. New York: Schuster Touchstone/Simon UrryJ. 2000. SociologyBeyondSocieties. London/New York:Routledge Wallerstein 1991.The nationalandtheuniverI. sal. In Culture,Globalizationand the World System, ed. A King, pp. 91-105. London: Macmillan WatersM. 1995. Globalization. London/New York:Routledge Manderson eds. 2000. Global WhitefordLM, L, HealthPolicy,Local Realities: the Fallacy of the Level Playing Field. Boulder,CO:Lynne Rienner Wolf DL. 1992. Factory Daughters: Gender, Household Dynamics and Rural Industrialization in Java. Berkeley:Univ. Calif. Press