Anda di halaman 1dari 23

Anthropological Studies of Complex Societies Author(s): S. N. Eisenstadt Source: Current Anthropology, Vol. 2, No. 3 (Jun., 1961), pp.

201-222 Published by: The University of Chicago Press on behalf of Wenner-Gren Foundation for
Anthropological Research

Stable URL: http://www.jstor.org/stable/2739628 . Accessed: 23/01/2014 08:26


Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

.
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.

The University of Chicago Press and Wenner-Gren Foundation for Anthropological Research are collaborating with JSTOR to digitize, preserve and extend access to Current Anthropology.

http://www.jstor.org

This content downloaded from 146.164.3.22 on Thu, 23 Jan 2014 08:26:37 AM All use subject to JSTOR Terms and Conditions

Anthropological Studiesof
Comnplex Societies
by S. N. Eisenstadt
studies on our general undersocial-anthropological of human societies.Out of the standingof theworking THE PURPOSE OF THIS PAPER iS to analyse someof the applicationof thismodel to thestudyof complexsocieproblemsarisingout of the application of the methods ties,several problemshave emergedwhich may be of and approachesdevelopedin social anthropology to the fromthepoint of view of generalanthrogreatinterest studyof more complex societies-whether "historical" pological and sociologicaltheory. or modernsocietiesor parts thereof. We hope to show What, then,is thismodel,and to whatkindsof probboth the contributions and limitations of some of the lems is it applied?2 conceptsand tools of social anthropology in the study is that it combines characteristic Its mostdistinctive of more complexsocieties. We shall not deal here with the descriptionand analysisof social behavior,group general anthropological approaches to such studies structure into one setof intersettings and institutional (Mandelbaum 1956),nor withsuch generalconceptsas relatedconceptsand analyticaltools. "culture" or such general problem areas as "culture level,social anthropologists On thepurelydescriptive and personality,"which have developed mainly in of social behavior usually deal with observedpatterns anthropology. Nor shall we be concernedwith studies ofindividualsin different and in differsocial situations of tribalor peasant communities by anthropologists or ent groups, and with statementsby individuals of otherscientists who have employedconceptsand tech- normswhich would be appropriate to such different niques commonto manybehavioralsciences. situations(Beattie 1959 and, in more detail, Emmett We shall, rather, concernourselvesmainlywith the 1958). However,this descriptionof social behavior is, potentialcontributions to thestudy ofcomplexsocieties at the same time,a studyand analysisof the structure thathave been made by one specific branchof anthro- of groupsand societies. pology,namely,social anthropology as it has been deThis typeof analysisis achieved throughthe great veloped in England (for a general surveysee Evans- emphasison social behavioras relatedto variousnorms Pritchard1951; Beattie 1956),and to a smallerextent which are said to be operativein the social structure. in U.S. and France (Eggan 1950, 1957; Levi-Strauss Most of thesocial-anthropological of social descriptions 1959). These social-anthropological studieshave devel- behaviorare studiesof thewaysin whichmajor norms oped a distincttheoreticalmodel which deals with a found in these societiesare upheld by individuals,of certainorderof problemsor of social phenomena.The norms,and between these different the interrelations developmentof this model, and of specificrigorous of thewaysin whichthesenormsinfluence and regulate offield techniques accountforthegreatimpactof work, groupsin society. the relationsbetweendifferent Thus we find in thesestudies,for instance,a great normsdirectingand reemphasison the institutional N. EISENSTADTis Professorof Sociology at the Hebrew SAMUEL in ofsocial themajor fields behavior individual stricting University, Jerusalem,Israel. Born in 1923, he was educated at the Hebrew University(M.A., 1944; Ph.D., 1947) and at the life-marriageand familylife; economic, political or London School of Economics.In 1955-56he was a Fellow at the and in close connection, Secondly, ritual organizations. Center for Advanced Studies in the Behavioral Sciences, Stanwe findan analysisof the consequencesof upholding ford, Cal., U.S.A. His researchinterestsinclude the absorption of immigrants relationswithina such normsforsocial and intergroup
Gluckman, Prof. 'I am indebted to Dr. J. Ben-David, Prof. WM. D. G. Mandelbaum, Dr. J. Talmon-Garber and Dr. A. Weingrod on it in great for reading the draftof this paper and commenting detail. The workon this paper has been facilitatedby a freegrantin-aid fromthe Behavioral Sciences Division of the Ford Foundation. 2 We shall not deal below with the historical developmentof this model and with the various trendswithinit, but shall concentrate mainly on its basic and salient features (Evans-Pritchard 1951; Beattie 1956).

in Israel; comparative studies of age groups and youth movements; comparative political sociology in historical,modern, and "underdeveloped" societies; and bureaucracy. The present article, submitted to CURRENT ANTHROPOLOGY on February 6, 1960, was sent for CA* treatment to fifty scholars,of whom the followingresponded with writtencomments: Michael Banton, J. A. Barnes and A. L. Epstein, J. H. M. Beattie, Meyer Fortes, J. R. Goody, S. T. Kimball, E. R. Leach, D. Mandelbaum, D. M. Schneider, Laila Shukry El Hamamsy, Ina E. Slamet-Velsink,and P. Worsley. The comments written for publication are printed in full after the author's text,and are followedby a replyfromthe author.

Vol. 2 -No. 3 *June 1961

201

This content downloaded from 146.164.3.22 on Thu, 23 Jan 2014 08:26:37 AM All use subject to JSTOR Terms and Conditions

lation betweenwhatmaybe called "culture"(or rather, on theone hand, and social values) and ritualsymbols, relations,on the otherhand. Values and symbolsare on two levels. An analysed by social anthropologists as analysisof thesevariousbeliefsand culturalsymbols has recently developed(Forde systems in their ownright 1954; Evans-Pritchard 1956). More common,however, in relationto varis the analysisof values and symbols kinship,or politious fieldsof social activity-family, ical activities. While anthropologistshave usually of the recognized the autonomy and distinctiveness sphereof values, theyhave also emphasizedthe direct relevanceof primordialor sacred symbolsand rituals for most typesof social activitiesin the societiesthey and the interlocking of theseactivitiesin a way study, whichassuresthatthey of are given"meaning"in terms and values. thesesymbols by most of these 3. The thirdmechanism,stressed in these sociestudies,is the continuousinterrelation in mostgroups tiesof different typesof social activities and situations.Ritual, jural, contractualand political activities-clearlydistinguishablefrom one another in most (Evans-Pritchard 1955)-are seen as interwoven of the situationsand groupsof thesesocieties,so that each activity with,and directly dependson, articulates Epstein (1954) has put is upheld in terms of,theothers. are thispoint thus: "Most of themajor social functions fulfilled by the same small groups." These threemechanismsare found to regulate the direct interrelation of social behavior to group and in most parts of the societies institutionalstructure, studiedbysocial anthropologists. themost Analytically, is that theyare importantaspect of thesemechanisms II structure of themajor social situaembeddedin thevery This model necessarilyraises the question of the and are not ortions and major groupsin the society, mechanisms through which the interrelationships ganized in special distinct waysor orders.Perhaps this among these variables are maintained. Few examina- characteristic explains why the social-anthropological of its diftions of the functionalaiialysis in anthropology(the model did not explicatetheinterrelationship verytype of analysiswhich stresses interrelationships ferent variables,and could use as analyticaltools conamong different thestructure of variousgroups. partsof societyor culture)have dealt ceptsdescribing explicitlywith this problem (for a partial exception, see Beattie 1959). However, social anthropological III studiescontaincertainpostulates, and descripanalyses, tionsof such mechanisms; and theseanalysesconstitute These assumptions guided the workof social anthroone of thebasic contributions of social anthropology to two decades,even though pologistsforat least the first thestudy and social processes. ofemphasisamongindiofhuman society differences there werefrequent 1. The first such mechanism is the interaction of the vidual anthropologists and among the many trends same personsor groupsin different an inter- whichdevelopedwithinsocial anthropology. situations, action whichmakes theirmutual commitments in one It is beyondthe provinceof thispaper to analyse or situationor group greatlyinfluencetheirbehavior in all the contributions of social anthroeven summarize others.In the societiesstudied by anthropologists, the pologists,or to discuss the many specifichypotheses existenceof such close relationsbetweenthe same peohave developedin the applicationof thismodel to they a very the different ple in different constitutes groupsand situations societieswhich theystudy.But it would importantmechanismregulatingsocial behavior, on some major trends to indicatebriefly or be worthwhile the one hand, and intergrouprelationson the other emphaseswhichhave developed in social anthropology (see Barnes 1959). Gluckmanhas advanced thisanalysis -without attempting to providea full list.All of them by showinghow close interrelations among the same in a way, fromthe basic common ashave stemmed, situationscreate conflicts people in different between sumptionof theirmodel, but each has emphasizeddifthem; and how thoseconflicts, which seem disruptive, ferent aspectsofit. in realityhelp enhance the solidarity and functioning 1. The numerous well-knownstudies-which need of the groupsand the societythroughthe crosscutting not be enumeratedhere-that emphasize the analysis of interests among the same people participatingin of "total" tribal societies,and veryoftenare identical thesevarying situations (Gluckman 1956). with analysis of wide ecological communities (e.g. 1940; Krige 1943). 2. The second mechanismis the specifictypeof re- Evans-Pritchard given society-e.g. how the marriage of a man from relationsbegroupA to a woman fromgroup B affects and thesocial institutions tweenthesegroups.Different of individuals are analysed,not only within activities theirown spheres,but also as links between several groupsin the society, to the conand as contributing tinuity of thesegroups and of the societyin general. Thus, for instance,the rules that governeconomic or politicalactivities are explained in terms of the "needs" of thesegroupsand of thesociety as a whole. (For a very pertinent formulation see Fortes 1953 and Gluckman 1956.)Thus, variousnormsand thepatterns ofbehavior whichuphold themare seen mostly as institutional directivesupholding the interrelations among different groupsin the societyand the continuity of the society as a whole. In this way,most anthropologicalstudies combine in one basic model the analysisof social behavior,institutional norms, groupsand societies.They explain patterns of social behavior through analysisof group-structure, institutions and "total" societies. Neither this model, nor the various studies which have been guided by it, have usually dealt with the problems of the kinds of interrelationships between such different variablesas social behavioror group and institutionalstructure-i.e.,which variables are dominant or influentialin general or in various typical situations.Rather, it has been assumed that all these variables are always verycloselyinterrelated through theircontributions to one another,and to the society as a whole through what are usuallycalled their"functions."(For a partial exception, see Worsley1956.)
202
CURRENT ANTHROPOLOGY

This content downloaded from 146.164.3.22 on Thu, 23 Jan 2014 08:26:37 AM All use subject to JSTOR Terms and Conditions

2. Those studieswhichare also concernedwithtotal societies,but tend to concentrateon the analysis of thesesocieties from thepoint ofviewofone major institutionalsphere-marriage and kinship,political structure,or stratification (e.g. Fortes 1945, 1949b; Kuper 1947). 3. An interesting offshoot of thesecondtrendare the general studies of certain institutionalgroups which are basic constituents of any human society, and some of theirbasic characteristics. The studiesof the domesticgroup,edited by Goody (1958) and brilliantly summarizedby Fortes(1958), are the best recentexamples ofthisdevelopment. We find herean attempt to analyse what may be called the natural life histories and life cyclesof its basic constituents-parents, children,and kin. 4. Studies of a greatvarietyof customsand institutional arrangements in termsof theirrelation to various aspectsof social structure, and to the functioning of the major natural groups of the societyas well as of the societyas a whole. Examples of this approach include the analysis of joking relations and of the "mother's brother" (Radcliffe-Brown1952; Goody 1959); and thestudiesof witchcraft, blood-brotherhood (Evans-Pritchard 1933; Eisenstadt 1956), magical hair (Leach 1958),and manyothertypesof institutional devicesor customs. These studieshave oftentaughtus to look beyond the manifest contentof theseusages,and have sharpenedour understanding of the "meaning" ofvariouscustoms and types ofsocial and human interaction,such as commensuality or even sexual relations, in terms of primordial imagesand sacredsymbols. 5. Various comparativestudies and analyses which have developed withinthe fold of social anthropological studies, the best examples of which are African Political Systems(Fortes and Evans-Pritchard 1940), AfricanSystemsof Kinship and Marriage (RadcliffeBrown and Forde 1950),AfricanWorlds (Forde 1954), and Tribes Without Rulers (Middletonand Tait 1958). The societies studied by social anthropologists evincedgreatvariationin the exact structure of different natural groups and institutions-territorial, kinship,and age groups;and various typesof associations or political groupings. But in spite of thesevariations, anthropologists found that the close interrelationship betweenindividual behavior,group structure, and institutional normswas alwaysmaintainedby the mechanisms outlined above. It is true that anthropologists and sociologists are now beginningto doubt whether all of the social organizationof thesesocietiescan be it has been claimed explained in such terms. Recently, that the absence of historicalrecordsin such societies and the two or threeyears that most anthropologists spent in the field "blinded" them to possibilitiesof change,of differentiation and variability of individual behavior, and of development of new group structures. It has also been claimed that the theoretical nature of their startingpoint may have biased anthropologists to emphasize those situationsin which behavior was regulatedbythemechanisms postulatedby their model. Some of themostrecentdevelopments in social anthropology itself-forinstance,Turner's (1957) study of Ndembu village life-illustratecertain of these shortcomings. (Additional criticism can be found in Fortes
Vol. 2 l No. 3 -June 1961

Eisenstadt:

STUDIES

OF COMPLEX

SOCIETIES

1949a; Gellner 1958 and Gluckman 1959bgive a fuller analysis.) Without doubt, however,this model contributedto the analysisof the functioning of at least some aspectsof the societies studied.

IV
Armed with these basic concepts and approaches, social anthropologists began to study other types of societies,both historicaland contemporary (modern), applyingtheirapproach and the assumptions implicit in theirmodel to understanding the workingof these societies.These studies can, in principle, be divided into several distincttypesor categories, broadly similar to those outlined above for more characteristic anthropologicalstudies. A. The first category comprises studiesof ecological communities-mostly of peasant villages and other units that are similarto, but not identicalwith,those in tribalsocieties.The mostimportant of thesestudies are Arensberg and Kimball (1948) on theIrish peasant; Frankenberg (1956) on footballand politicsin a Welsh village under the impact of industrialization;PittRivers (1954) on the structure of an Andalusianvillage and its place in Spanish society;Barnes' (1954) study of a Norwegian parish; Srinivas' (1952) work on the Coorgsof India; and some of theIndian village studies (e.g. Marriott1955). B. The second categorycomprisesstudies of one main typeof "natural" or institutional group in complex societies, usually examined in relation to the widerinstitutional setting. Naturallyenough,the most frequently studied institutional unit is the familyand kinshipunit; but studiesof such institutional settings as economicand political structure can also be found. Within this category may be distinguished threemain whichoverlap to some extent. typesof studies, 1. The first typeof study, whichdeals withtheinternal structureof such groups, includes, Freedman's (1957) workon Chinesemarriagein Singapore,and the studies of familyand kinshipstructure in modern societyby Young and Willmott (1957), Firth and associates (1956), and Bott (1957). 2. The second typecomprises studieswhich emphasize the interrelationships betweenthesegroupsand the wider institutional of theirsocieties.The best settings examples are Firth's (1946) study of Malay peasant economy; Freedman's (1958) work on the Chinese lineage organization;Miller's (1954) "Caste and territoryin Malabar"; Mayer's (1960) Caste and Kinship in CentralIndia; Bailey's (1957) studyof the interrelation betweenmoderneconomyand caste in an Indian village; and Gullick'sworkon Malayan political structure (1958). These studies aim at understandingthe operation of groups with which anthropologists were traditionallyconcerned-the peasant village, the lineage, the caste,or kinshipunit-in more complex (but notyetmodern)societies:to what extentthey maintain their corporate identity;what functionssuch groups fortheirmembers perform in relation to the widersociety;how themoregeneralpoliticalor economicforces on theseunits,as well as how theseunits are impinlge
203

This content downloaded from 146.164.3.22 on Thu, 23 Jan 2014 08:26:37 AM All use subject to JSTOR Terms and Conditions

integratedwithin this wider institutionalstructure. on the 3. There are otherstudieswhich concentrate analysis of a specific form of "naturalgroup" or institution withina complex society, and attemptto explain itsrelationto variousaspectsof thebroadersocial structure.Smith's(1956) study of theNegrofamily in British Guiana, which attemptsto explain the general social and mainteconditionsthatfacilitatethe development nanceofthematri-focal relations and system ofdomestic householdgroupingin the Caribbean, is the most important workof thistype. It is worthwhile to note an interesting convergence ofthesocial anthropological studiesdiscussed herewith thesomewhat different anthropological tradition representedby Redfield'sstudies(1955, 1956) of the "Great" and "Little" Tradition, and the mechanisms of transmission fromthe formerto the latter. These studies weremostexplicitly focusedon an analysisof themechanismsconnecting such"closed" unitsas village,family, or caste with the more centralinstitutional of system relatively complex societies. C. The third categorycomprisesstudies which attemptto analysethe"total" structure of relatively complex, but not modern,societies.The outstandingexamples are Nadel's (1942) study of the Nupe, and Leach's (1954) Political Systemsof Highland Burma. Although Leach's work, as its title indicates, deals it is reallyconcerned mainlywithpolitical institutions, with the analysis of the concept of "total" societyso oftenused by social anthropologists. A special category of anthropological studiesof "complex" situations, whichcutsacrosstheclassification used above, includesthosedealingwithtribalgroupsor subunitsin situationsof change under the impactof modernconditions. Examples are Gluckman's(1958) "Analysis of a social situation in modern Zululand"; Epstein'swork(1958) on urban politicsin Africa;Little's (1955) studyof voluntary associations;Mitchell's(1956) descriptionof the developmentof a particulardance among urbanizedAfricans;and Watson's (1958) study of the impact of a money economyon two African tribes.Stemming fromearlierresearchon "detribalization," these studieshave graduallyfocusedon understandingthe impact of processesof modernization on tribal structure, and the reorganization of this structurein thenew situations. They have also concentrated on the ways in whichnew typesof groupings, such as urban association trade unions, tend to crystallize in thesenew settings. In addition to these,we may also mentionvarious studies in other fields-especiallyindustrialsociology (Burns 1955; Bradney1957a,b;Gluckman1956)-which deal withthe natureof thepersonaland intergroup relations operativein modern industrialsettings. These were influencedby some of the approaches to social anthropology, even if they do not apply the socialanthropological"model" in the same consistentway as thosecited above. V The ultimate aim of these studies is to show how social behaviorin complex societiesis also determined by the groups and institutions in which individuals
204

to the participate, and how such behavior contributes and of the of different functioning groupsin thesociety societyas a whole. But the factswhich theyhave to explain, and the problemstheyhave to analyse,differ somewhatfromthosewhich are usually dealt with in anthropologicalstudies. It is true that most of these units studiesdeal with typesof groupsor institutional which are, to some extent,similar to those studied in tribalsocieties;yettheseunitsare not as self-contained or closelyoverlappingas in the latter.In the complex into widersettings in ways theunitsinterlock societies, different from those of primitivesocieties. It is this interlocking that createsmany problems for the understandingof patternsof social behavior and their in these structure relation to group and institutional societies. Thus Barnes, in his studyof the Norwegianparish (1954), faced not only the problem of explaining the ways in which the membersof the parish are interrecommittees in the parish lated, or in which different work,and how theyare related to the kinship,workwithinthe parish.He had also ing, and class structure to deal, first, with the problemand the possibility that membersof the same general social category will be roles and participatein differable to choose different of new ent associations;second,with the development under the directimpact groups and typesof activities of outside forces;and third,with the relationsof all these groups to whatevermay be called the over-all Norwegiansociety. Pitt-Rivers(1954) was dealing not only with the of his Andalusian village but also internal structure with the problem of differential impact of the central of different forces on the activities individuals,belonging perhapsto similargroupsin the village. In addition to investigating internalvillage politics, the importanceof football and various ceremoniesin and theirimpacton the theframework of thesepolitics, unity of the village, Frankenberg(1956) is also concerned, even if indirectly, with the impact of wider suchas thoseof industrialization, on theinternal forces, of the village. cohesionand continuity In his studyof Chinese lineage organization, Freednot onlyin the internalstrucman (1958) is interested, ture of the lineages,but also in the ways these interacted with the over-allpolitical structure and with the imperial bureaucracy;how they were able to adjust to this bureaucratic,non-kinship themselves organization; and what functionstheyperformed within that widercontext. In his work on Chinese marriage and kinship in Singapore (1957), Freedman is dealing not only with the internalstructure of the major kinship and marriage groups as relatively "closed" units,but also with theproblemofthebreaking up of thesecorporate units, and of the influence of the "anonymity" of Singapore city life on marriageand kinship behavior,on these units,and on familyand kinshiprelations. Nadel (1942) was the most explicit in focusinghis on the over-allworkingof a complex society interest without concentrating exclusivelyon any small subunit. Throughouthis studyof the Nupe, he was interested in describingand analysinghow different local groups and different typesof orderedsocial activitiesCURRENT ANTHROPOLOGY

This content downloaded from 146.164.3.22 on Thu, 23 Jan 2014 08:26:37 AM All use subject to JSTOR Terms and Conditions

economic, ritual,educational,and political-are related to the working of the over-allNupe society, and how theycreatethe specific Nupe identity. Little(1955),Epstein (1958),Banton (1957), Mitchell (1956),and Watson (1958) deal not onlywith the processesof change and disintegration of tribal units,but also with the ways in which these groups are transin a new setting-howentirely formed new groupings are formed;how conflicts that do not necessarily contribute to the equilibriumof the encompassing society maydevelopbetweenvarioustypesof groups;and how these groups compete for the allegiance of the same persons,or influencetheir behavior in contradictory ways.Thus, theyalso had to explain thewaysin which the people theystudied were drawn into social frameworks and influenced by social forces of processes the major natural ecological groups (e.g. transcending wideroccupational,economic,or political forces)and how the behaviorof people in thesegroups,and their choicesof roles,are relatednot onlyto theirimmediate groupallegiances,but also to wider social forces, some of whichdid not originatewithinthe "old setting." VI Thus, the problemswhich were either implicit or explicitin the data presented by social-anthropological studiesof complexsocietiesare of a somewhat different order from those involved in the study of tribal societies. The main difference could be foundin the factthat patternsof social behavior in these complex societies were influenced and regulatedby manyforces and factorswhich were not embedded in the structure of the major social groups or their interrelations (e.g. they were not regulatedby the interrelations between domestic, kinship,and local groups),and whichwere not borneout bypeople interlocked necessarily in thesame situationsand groups. This difference has, necessarily, posed problemsof the nature of the mechanisms regulatingthe relations betweenvarious typesof social behavior,group structure,and the widerinstitutional of a society. structure How, then, did the social anthropologists approach these problemsand what were theircontributions towards theirsolution? It is here that the most importantcontributions of thesestudies can be found. These contributions were primarilymade throughthe discoveryof several distinctways in which the regulativemechanisms found in tribalsocietiesoperate in more complex societiesas thatin the lattersocietiessuch well; and the discovery mechanismsare not necessarilyalways embedded in concretegroups. A. These studies discovered and analysed various areas of life in complex societiesin which veryclose interrelations exist between people participatingin different situations and groups, and these interrelationsregulatetheirsocial behaviorin different groups. The importance of thistypeof mechanism in corporate groupsand institutional fields withincomplexsocieties is bestexemplified byFreedman's Lineage Organization in South-eastern China (1958) and Miller's work on caste in Malabar (1954). Young's (1957), Firth'sand asVol. 2 *No. 3 *June 1961

Eisenstadt:

STUDIES

OF COMPLEX

SOCIETIES

sociates' (1956), and Bott's (1957) studiesof family and kinship in London, although more limited in scope, also show the importance of theseclose personal interin thestructure of at least some corporate relationships groups and institutional of modern societies. settings Such close relationswere also oftenfound to be operativein manyinformal social relationsnot organizedin corporategroups or institutional frameworks. Barnes(1959),in his reviewofFrankenberg, has aptly of thismechanism summarized the importance in some ecological units of modernsocieties: In Pentrediwaith thesamecoreof individuals interact in many with oneanother different andtheir besituations, in one context is influenced havior commitments bytheir in many others. This is trueof almost all communities in theprimitive anditisa necessary condition for world, using theslowand indirect field to socialantechniques specific thropology (cf.my[i.e.Barnes'] "Classand Committees in a Norwegian HumanRelations, IslandParish," Vol. VII, 1954:44). The ethnographer, withhis traditional distrust ofdirect questions and questionnaires, andhisdesire todo more a correlation, thantest a baldhypothesis or establish is particularly wellqualified to observe these and lengthy devious ofsocialactionand to analyze sequences in them sociological As socialanthropologists terms. we havelong held,either as an article of faith or a matter of academic political thatwe havevaluablecontributions expediency, to makein thestudy of advanced butit has not societies, beenclear, always either tous or toourcolleagues in other disciplines, precisely whatthese specialcontributions are. At one timethe answer might have been "community studies." But someof these, although informative, appear notto differ greatly from thestudies carried out bygeographers and othersocialscientists withbackground and objectives different from thoseof social-anthropologists; other studies underthis aremerely rubric socialsurveys of populations whichmayor maynot form a community. Frankenberg's pioneering studyof Pentrediwaith shows that oneuseful contribution which socialanthropology can maketo theinvestigation of advanced is theobsocieties servation andanalysis ofpolitics round thevillage pump. B. The anthropological studiescited here have also foundthatdirectrelations betweenprimordialsymbols and values and various social relationsare operative in severalareas of complex societies.Such directrelationshave been found,first, as Young's and Willmott's (1957) studies have especially shown, in the field of familyand kinship relations.Second, anthropologists have found such relationsto be veryimportantin an individual'sattitudetowards his community or society. Third, theyhave also been found to be operative in more diffusesocial areas and relations. (For a more general evaluation of the place of primordialimages in social organizationsee Shils 1957.) Thus, variousstudies(Burns 1955; Bradney1957a,b; Lupton and Cunnison 1957) have shownthat in many areas of lifein complexmodernsocietiestheredevelop, betweenpersonsparticipating in some commonformal frameworks, different types of relations, which are regulatedeitherby theirinterrelations in othersituations or by such primordial images and symbolsas "manhood," "human trust,"and "age," or sharingof
common ''fate,'' ''neighborhood,'' or "'goodness.'" Such 205

This content downloaded from 146.164.3.22 on Thu, 23 Jan 2014 08:26:37 AM All use subject to JSTOR Terms and Conditions

relationsand symbolshave been found to cut across hierarchicalrelations-or any other formal relations and groups-and also to influence patterns of friendship and diffuse personalrelations. Such relationsmayoften be more forceful than the formalrole-definitions, and maygreatlyinfluencethe performance of theseformal roles. Similarly,some of these studies indicate that different usages-such as commensuality, certain types of friendship, name-giving,gift-giving, plays and matches-in many of the complex and even modern societies,are often imbued with various primordial meaningsand values (Gluckman1959a). These studies, forinstance, deal withthewaysin whichmanyconflicts and the developbetweentheEuropean industrial forces tribalforces, and howsomecontinuous moretraditional -although not necessarilystable-interrelationsdevelop between theseopposing forces. While some resultsof thesestudiesmay seem identical with those of various researcherson "primary groups,"in facttheyoftengo beyond the latter;or at least complement them. Most research on primary within groupshas pointed out mainlythe importance, of so-calledsolidaryand face-toformalorganizations, face relationships. Social-anthropological studies,however,have gone-potentiallyat least-beyondthesefindnot only ings.They showhow theseformalframeworks but are also greatly are imbued withsolidaryrelations, influencedby the same individual's personal commitand situations, and are invested mentin othersettings whichare priorto the withcertainprimordial meanings as well as to the various solidary formalrole-definition and mayeven,to some extent, relations, regulatethem. C. Several of thosestudiesattemptto apply some of and themore generalprinciplesof social organization, of potentially especiallyof the interlocking conflicting forcesin one "total field,"to more complex situations or societies.Thus, several studies of "colonial" situationsin Africa,or of certainaspectsof industrialstructure(Gluckman 1956,1958),have sharpenedour awareforcesmay sometimes become ness of how conflicting in a single"total field";and of how differinterrelated of behaviorin such a fieldcan be ent concrete patterns explained in termsof the relative strengthof these withintheircommonframework, conflicting forces and of thisframework. of the functioning of all theregulative mechanisms found Significantly, in the social-anthropological model, only one-that of the coexistenceof different types(jural, political,ecoin thesamegroups nomic,ritual,etc.)of social activities or situation-was not found to be of greatimportance in thecomplexsocieties. VII The studieswhichhave shownhow variousmechanismsoperatein complexsocieties have also contributed, explicitlyor implicitly,to the analysis of two other aspectsof thesesocieties. First, the delineation of the areas and relations in complexsocietiesin whichthe typesof relationspostulated by the model of social anthropology still persist pointsmore sharplyto thosestructural pointsat which thesecease to operateand are replacedby moregeneral and differentiated mechanismsof orderingsocial ac206

tivities-even if, as we shall see, specific anthropological concepts do not help very much in the analysis of these mechanisms. To give only one example, we may quote Freedmani on some of the problems of Chinese marriage in Singapore (1957): a link beMarriage in Singapore no longer represents tween two corporategroups,and Chinese in the Colony, while preferring to seek their brides within their own withinnarrower divisions of these dialect-groups (and often only groups),treata new marriageas a matterconcerning the bridal couple and their fellow household-members. Membersof clan associationsmay be invited to wedding in the marbut theyhave littledirectinterest celebrations of Singapore Chinese society The structure riage formed. makes for the individualizationof marriage.This process is also sanctioned by modernistideas which derive their nacentury potencyfromtheirconnectionwith twentieth tionalism. In the disruption of marriagethe inadequacyof Chinese mechanisms to controldomesticdisputesis veryapparent. Since thereare no clearlydefinedlargerkinshipunits,no of aulocal units in urban life, and no formalstructure maritalquarrels tend frein the rural settlements, thority quently to findtheirway to personsand bodies standing itself. It is at thispoint outsidethelimitsof Chinesesociety courts in the shape of the magistrates' that "government," decisively and the Departmentof Social Welfareintervene in Chinese life. Divorce among Chinese in the Colony is essentiallya matter of mutual agreementbetween the spouses. However, just as in the formationof marriage in the theanomymity of Singaporelifeallowsuncertainties statusof men and womenas husbandsand wives,so in the disruptionof marriage there are sometimesambiguities whicharise when wivesleave theirhusbands (or are abandoned by them) and "follow,"as the Chinese say, "other men." Among studies of a somewhat different anthropological tradition which bear on this problem, we should again mention Redfield's (1955, 1956) work on peasant societies. He has described the differentiation of the different orders of social relations in the more complex societies, but at the same time has given us a feeling for the interrelation of these differentorders-even if he did not explain this analytically. Second, these studies can contribute much to the development of comparative studies. This can best be seen in Smith's work on the West Indian Negro family (1956). Smith attempts to give an exact specification of those broader mechanisms of societal differentiation which impinge on and influence the structure and functioning of family organization. Thus, he argues that the husband's role in the matrifocal family is stronglycorrelated with the role of men in the economic system and in the system of social stratification in Guianese society. This specification,when fullyworked out, may serve as a starting point for furthercomparative work on the family, and for parallel attempts in other fields(Eisenstadt 1957). Similarly,Watson's (1958) analysis of the impact of a money economy on tribal society contains very interesting hypotheses about the influence of differentstructural variables (e.g. patrilineal and matrilineal descent, patrilocality and matrilocality) on the process of adaptation of modern forces within this tribal framework,and in termsof the maintenance of this framework.
CURRENT ANTHROPOLOGY

This content downloaded from 146.164.3.22 on Thu, 23 Jan 2014 08:26:37 AM All use subject to JSTOR Terms and Conditions

Finally,thesestudies, especially thoserelatedto problems of change, have necessarilycalled attention to certain problems of the processesof transitionfrom relatively homogeneoussocietiesto more differentiated socialorders, and have pointedout someof theproblems and patterns ofbehaviorand structure whichdevelop in such a process. VIII The significance of all these contributions goes beyond their specificanalyses of the areas enumerated above. Their main significance forsociologicalanalysis lies in theirenlargingour understanding of the complexity and varietyof the social mechanismswhich regulate social behavior and organize the division of labor and the interrelations betweendifferent groups in so-calledcomplex societies.Sociological and anthropolitical analysishas oftentended to classify societies into "types"in terms of some such major mechanisms. Mechanical and organic solidarity, and Gemeinschaft Gesellschaft,"sacred" and "secular" societies, "folk societies,"and often the very terms"primitive"and "complex" societies,have been used to denote these types.And althoughit has become almost a commonplace ofsociologicalqnd anthropological that literature "pure types"of such societiesdo not exist,sociological and anthropological to delve analysisseldom attempts too deeply into the implicationsof this assertionfor the analysisof concretewaysof social organization.It is here that the social-anthropological studiesof complex societieshave made theirmost distinctcontribution to sociologicaltheory. They have clearlyindicated how regulativemechanisms, belonging as it were to one typeof society, operatein some partsof othertypes of societiesand constitute basic componentsof social organizations. Moreover,theyhave shown that in the do not complex societiesthese regulativemechanisms but instead operatein closed,watertight compartments, veryofteninterpenetrate with othermechanisms, permeate the same areas of social life as similarmechanisms,and are not necessarily always embedded in the of the major groupsof a society. structure In thisway, of these studies have contributedto the reassessment the sociological image of modernsocieties.They have shown,at least by implication,that modernsocietyis not a "mechanized"one, in whichatomizedindividuals live in separation,ruled only by impersonal forces. Rather,variouscloselyinterwoven personal and group relations,on the one hand, and relations permeated withsymbolicand primordialmeanings,on the other hand, constitutebasic componentsof even the most differentiated typeof society. Of special significance thispoint of view is the from implicationof theseworksforthe analysisof the place ofprimordial and ritualvalues in theorganizasymbols tion of a society.The importanceof "ritual" or religious activitiesin primitivesocietieshas often been explained mainly in termsof little social differentiaand closenessof social intertion,and of the intensity tction. But anthropological studiesof complexsocieties (as wellas several otherstudies) indicate thatof the sigriifcance of primordial images and of rates interactionand social differentiation mayvaryindependently; thatsymbolicalprimordialimages and values cannot
Vol. 2 *No. 3 *June 1961

Eisenstadt:

STUDIES

OF COMPLEX

SOCIETIES

be reduced to such rates of interaction; and that they constitute an independentrealm whichis autonomous and of great importancein any society,even if the waysin whichthey operatediffer from to society. society

Ix
The importantcontributions of social-anthropological studiesof complexsocieties do not,however, necessarily imply that theyhave solved all the problems which were implicit-and also, veryoften,explicit-in theirmaterials. While thesestudiespointed out some of thoseareas in the social structure of complex societiesin which certain regulativemechanismsmost predominantin ptimitivesocietieswere operative,theywere less successfulin analysingareas of social structure-which are in most of thesecomstill of predominant importance plex societies-withinwhich such mechanismsdo not operate. As we have seen earlier,theirdata showtheexistence, in these societies,of many aspects of social behavior which could not be explained solelyin termsof place in a givengroupor institution, or in terms of contribution to the functioning of these groups in the total society.These data have also indicated that in most the different complexsocieties, typesof social activities -ritual, economic,contractual, political, etc.-are not and interrelated closelyinterwoven in the same situations or groups. In complex societies,each of these into specific of activities oftentendsto crystallize types institutional or ordersof its own. Thereframeworks ritual foreit was moredifficult to pointout theultimate or primordialsanctionsof many contractualtypesof behavior. 1. In moreconcrete thesestudiesand the conterms, ceptsemployedin themdid not explain how behavior was regulatedin situationsin which individualscould choosebetweendifferent rolesand groups,and in which different and contradictory institutionalforces and groups (e.g. tribal and familyloyalties,and modern political or occupational groupings)impinge on the in them,and may oftenhave individualsparticipating made different-even contradictory-demands. 2. The conceptscould not explain thewaysin which of individual behaviorand activity manypatterns (e.g. in leisure time activities, in work-place, and in family groups)were not entirely bound to concretegroupsor embedded in them,but were widely dispersedamong and seemingly different unconnected situations. 3. They did not explain how new patternsof behavior, groups, and situations (e.g. modern working situation,tradeorganization, political affiliation) were continuously emerging, crystallizing, and impinging on different individuals and making different, oftenconflicting, demands. 4. They did not explain the waysin which different typesof social activities-contractual, juridical, political, ritual, etc.-are organized in different "orders" withina society, and how all theseordersimpinge on and regulate various situations,groups,and patterns of behaviorwithinthe society.
207

This content downloaded from 146.164.3.22 on Thu, 23 Jan 2014 08:26:37 AM All use subject to JSTOR Terms and Conditions

5. They did not explain how-if at all-many congroups in these societies (e.g. between different flicts political and economic conflictsbetween Europeans wereregucuttingacrossanysuch society) and Africans lated, or how theycould be said simplyor outrightly of an onto contributeto the continuousintegration goingsociety.This also seems to apply to the analysis of the functionof any one group within the "total" society.These studies have shown that, in the more to delineate the it was more difficult complexsocieties, exact functionof each such group within the "total" which any such or the exact functions social structure, in relationto the basic institutional groupsperformed Hence, (e.g.economicor political) of thesociety. spheres it seems that the whole concept and reality of the "total" societyis here much more problematicthan in the studyof primitivesocieties; and that this "total" society is no longer a relativelywell-circumscribed group with clear ecological, personal and cultural boundaries.3 Thus, most anthropologicalanalysesof complex societieswere unable to analyse fullythe problemsof a The only partial complex divisionof labor in society. exceptionis, again, Nadel's workon the Nupe (1942). of the political,ritual, and economic His descriptions groups within the Nupe State are inlife of different tendednot only to presentan internalanalysisof these of activities types but also to showhow different groups, (e.g. economic,political, ritual) are organized in the the "whole" which affect variousordersof mechanisms of Nupe society.Nadel clearly distinguishespurely activitiesor ecological arrangeeconomic,semi-market mentsfromsuch organizationsas age groups and varactivities (e.g. the Ganni ritual; ious ritual-symbolical forthe transNadel 1949),which serveas mechanisms submission of sacred common symbolsto different While he shows how even ecogroups in the sociery. nomic or political activitiesand mechanismsare directed and influencedby "cultural" or ritual values, he recognizesthat these are not always embedded in of the different the structure groups,but ratherwork mechanisms.He more complicated, throughdifferent, seem to operpointsout someof thewaysin whichthey he also does not fully analysehow each ate,even though is organized,or of thesesocial ordersand mechanisms how they operate togetherin a relativelycomplex society. x met by social-anthropologists The difficulties in explaining all the problemsof the division of labor in complex societies can most clearly be seen, perhaps, an analysisof the conceptsthey employedand through once they coined forthestudyof the complexsocieties, found that their traditionaltechniquesand concepts to deal with all the problemsinherent did not suffice in the material.The most important conceptsevolved forthesepurposesare "network," developed by Barnes (1954) (and also used by Bott 1957); "social field,"initially developed by Fortes and by Gluckman and his
3It is significant to note that some doubts about the validityof the concept of "social structure"or total societywere pointed out by Leach (1954) in his studyof the Burmese highlands.

as developed (1958); and "social organization," students byFirth(1951). is used byBarnes (1954: 43) in the The termnetwork of the social following way to describethe complexity relation to be found in Bremnes: of in touchwitha number is, as it were, Each person with each in touch aredirectly peoplesomeofwhom other has eachperson andsome arenot.Similarly ofwhom other have theirown and thesefriends a number of friends, eachother, know friends someofanyone person's friends; of itconvenient totalk ofa socialfield others do not.I find I haveis ofa setofpoints kindas a network. this The image oftheimage The points ofwhich arejoinedbylines. some and the lines indicate groups, are people,or sometimes We can of course witheach other. which peopleinteract of a network think ofsociallifeas generating ofthewhole I wantto however, purposes, thiskind.For our present thatpartof thetotalnetwork speaking, consider, roughly and thatis leftbehindwhenwe removethe groupings to theterritory which belong strictly chains ofinteraction and industrial systems. The conceptdenotedby theterms social field (which way,by K. Levin was also used, althoughin a different and in sociometry) and social situationhas been developed byFortes (1949b) and thenused byBarnes (1954), Gluckman (1958), and Epstein (1958) to deal with the found,and withthe whichthey new typeof totalsociety between its parts. Barnes (1954: 42-44) interrelations way: uses thisconceptin thefollowing three regions analysis wecanisolate ofthis Thusin terms there is the Firstly orfields in thesocialsystems ofBremnes. of enwitha largenumber socialfield, territorially-based one hierarchically, units, arranged administrative during withanother.... The second socialfield is that bytheindustrial generated of interdependent, Here we havea largenumber system. marunits suchas fishing vessels, autonomous yetformally connected and herring-oil factories, keting cooperatives, rather than yet hierarchically, with eachother functionally in a hierarchy of command.... eachorganized internally it has or boundaries; has no units socialfield The third It is madeup of thetiesof no coordinating organization. up in growing andacquaintance which everyone friendship buildsup for inherits and largely Bremnes society partly A fewof kinsmen. himself. Someof thetiesare between a as between them arebetween peoplewhoarenotequals, he has keptconwith whom manand a former employer who between persons tact.Mostof thetiesare,however, and itis toone another, accord equal status approximately theclass be saidto constitute I think, ties these may which, are ofthissocialfield The elements ofBremnes. ofsystem andold formed newties arecontinually notfixed, for being coldstorage. arebroken orputintoindefinite links definition is given by Epstein A somewhatdifferent 234): (1958: has to be seen as a singlefield "The Copperbelt," then, whichis composedof different sets of social relations forms a distinct Funofrelations, eachofwhich sub-system. is thedominant becleavage damental to this socialsystem andAfricans, and this influences cleavage tween Europeans within eachpartof the behavior and institutional growth At thesametime, eachsub-system a cersocialfield. enjoys they do notreactin thesame tainmeasure of autonomy;
way and at the same time towards the external stimuli makingforsocial change. has been used by Firth The term social organization
CURRENT ANTHROPOLOGY

208

This content downloaded from 146.164.3.22 on Thu, 23 Jan 2014 08:26:37 AM All use subject to JSTOR Terms and Conditions

to denote the difference betweenthe "formal"or normativesocial structure and the patternsof individual behaviorwhich develop within that structure. These concepts, and especiallythatof social network, are veryusefulin that first, theydescribesome of the morecomplexinterrelationships whichtendto develop betweenthe more closed "homogeneous"groups and the wider social setting;and second, theymay make us sensitive to the complex natureof the forceswhich operatewithin any single unit of behavior or group within thesegroupsand societies.But, withthe partial of "network," in the exception theydo not help greatly systematic analysisof the greatvarietyof problemsof social divisionof labor, which are explicit or implicit in the material presented in social-anthropological studiesof complex societies. Thus, while theconceptof social field(or fields)and sub-system does tell us that we have to look for some generaldivisions of society and their interrelations (economic,political, etc.), it does not help us to findexplanationsof the principlesof such interrelationship, or of the ways in which different groups and social situationsare interrelated with respectto these divisions.It does not help in the systematic analysisof the waysin whichthesedivisionsimpingein different ways on different individuals,or explain whysome individuals are drawnmorethanothers into a givenfield.Nor do the uses of this conceptexplain the ways in which possible conflicts betweensuch fieldsdevelop and are, or are not, resolved. Similarly,while the concept of social organization does indicate clearlythe existenceof the problem of differences between individual behavior and institutional structure, it does not, as yet at least, show us what are the major components of thissocial organization,how theyvary,and in what ways theymay affect different aspectsof social structure. It is perhaps only the concept of networkthat to some extentprovidesa potentially new analyticaltool. It clearlydescribes or pointsout the existenceof some differential interrelation betweendifferent people who are not organizedin corporate groups;and it mayhelp in theanalysis oftherelationofdifferent persons, acting in such a network, to different types of social roles and institutional In this way, the concept of frameworks. network does at least point out one way-beyond embedding in the structureof concrete groups-in whichthevariousregulative mechanisms can be organized. However,althoughBotthas analysedtherelations betweencertaintypes of networks and divisionof labor (1957), this concept has not yet been used to explain how different people are drawninto different networks, or how suchnetworks develop. Moreover, theconceptis still limited to that type of social relation in which some of the close interrelationship exists betweenthe same people in different situations. Thus it may be said that, although these concepts describeimportantaspects of social organizationand in thesesocieties, structure mostof them(withthe partial exceptionof network)do not help in an analysis of the systematic variationsof behavior in those types of situationsin which close interrelations betweenindividual behavior, group structureand institutional
Vol. 2 *No. 3 *June 1961

Eisenstadt:

STUDIES

OF COMPLEX

SOCIETIES

orderdo not exist.4 These concepts could notbe applied to the analysis of different mechanismsor regulating social behavior,or of social division of labor, which werenot embeddedin thestructure of concrete groups; or to the problemsinherent in the coexistenceand developmentof different ordersof social relationsin the same society. From thispoint of view it is interesting to note two facts: I. In so far as anthropologists deal with these differentinstitutionalorders, they tend to employ the usual sociological nomenclature,usually without attempting any criticalappraisal of it. 2. It is verysignificant, fromthispoint of view,that mostof theseanthropological studies(with the partial exception of Epstein's) have not made systematic use of the concept of "role"-nor have theycriticallyappraised it. The importanceof this concept,fromthe point of view of our analysis,is that it is the basic sociological concept (even if one of the first to coin it was an anthropologist, Linton) which attempts to link togetherindividual behavior and its social function, not only on the level of concretegroups.Its aim is to show how thebehaviorof individualsin different positions in the societyis influenced, not only by membership in concretegroups,but also by various broader institutional settings and forces, each ofwhichattempts to definecertainaspectsof a social position; to make (veryoftennew) demands on its incumbents;and to employ different sanctionsto ensure their acceptance of such demands. The limitationsof these analysesare perhaps most evidentwhen theydeal with the problemsof change. Social-anthropological studies dealing with this problem do not usually explain how new frameworks of social organization-new"social fields"-have emerged out of theolderones,or how thenew institutional order and normsthatdevelop have become crystallized.5 Nor do they systematically explain the forceswhich influencedifferent individuals to choose between alternatives in the new situation. Rather, most of these studiestake the existenceof some of thesenew frameworksforgranted.Starting off fromthispremise,they tend to investigatethe different groupingsthat exist withinthem. It is veryinteresting to compare,fromthis point of view, Watson's analysisof the impact of moneyeconomyon the Mambwe withotherstudiesof change,for instancethat by Epstein. Watson's main topic focused on the continuous functioning of the tribal groups, and how moneyeconomydid not destroytribal cohesion. Hence, his analysisis muchmoreakin to theusual anthropologicalstudiesand gives a much more "complete" pictureof a still relatively undifferentiated societythandoes Epstein'sanalysesof morecomplexsituations,whichnecessarily focusedon individual behavior
4 The criticism levelled against the Warner school's application of social anthropology stressed thispoint in a pertinent way (Goldschmidt1950). 5 A possible partial exception to this is Worsley's (1957) studyof cargo-cults. But, significantly, thisstudyuses specific anthropological conceptsto a smaller degree.

209

This content downloaded from 146.164.3.22 on Thu, 23 Jan 2014 08:26:37 AM All use subject to JSTOR Terms and Conditions

of groupsin a givensituation,taking and thestructure the developmentof the situation more or less for granted. XI studieshave We see,thus,thatsocial-anthropological to the analysisand undercontribution made a distinct and have provided new standingof complex societies, forsuch analyses.On the otherhand, they perspectives in analysingall the different were not verysuccessful and espein thesesocieties, ofsocial organization aspects cially thoseproblemsof the complex divisionof labor types of different whichwererelatedto theco-existence opor to the simultaneous or ordersof social activities, types of regulative mechanisms eration of different withina given society. general contributionof While the most significant that these studies was their showing systematically within any "type" of social structurethere operate they typesof regulativemechanisms, varied, different implidid not go on to analyse the general theoretical of reguhow thesedifferent types cationof thisfinding: and what operatein thesamesociety, lativemechanisms are the relationsbetween them.It was exactlyat this point that theiranalysisstopped; and theyhave conmainlyon the operationof one (broad) type centrated of such mechanismswithin complex and modern societies. explorationof this implication, However, a further and of the general problemsarisingfromthe anthro-

pological study of complex societies,may contribute much to general sociological theory,and may also help reassess the analysis of tribal societies. Several recent studies of simple societies, such as Turner's (1957), as well as some of Leach's queries about the anthropologicaldefinitionof social structure(Leach 1954; Gellner 1958)-not to mentionLevi-Strauss' analysesof different ordersof social life (1953, 1959)-have pointedout that,in at least some areas of social life in primitivesocieties,there operate types of regulative mechanisms thatare more varied than has been postulated by the social-anthropological model; and that in these societieseven these mechanismsare not always embedded in concretegroups-even if the degree of differentiation betweenthem is less than in the more complexsocieties. The furtherexploration of the insights gained through analysis of the limitations encountered in applyingthe model of social anthropology to complex societies may stimulatesuch reassessments of simple societiesby looking for areas in theirsocial structure where more differentiated regulative mechanismsof social behavioroperate side by side with thosemechanismswhichare embeddedin the structure of concrete groups.In thisway,as well as through sharperconfrontations between the continuouslydeveloping anthropological and sociologicalstudiesof complex societies, the full implicationsof these studies for sociological and analysismay be broughtout. theory

larly in section X. He assertsthat the concept"social field"can be used to describe complex interrelationships, but By J.A. BARNES and A. L. EPSTEIN* not to analyse and explain the way in groupsand social situaEisenstadt's lucid and stimulating whichdifferent tions are interrelated. What does this summary evokes some comments on of judgment,otherson matters mean? For example, Fortes makes use matters of principle. For example, he judges of the social fieldconcept in his discus"social field" to be a concept evolved sion of the spread of matrilateralkin ties (1949b: 286ff.).Is this discussion to grapple with complex societies.Yet description or analysis? Surelyit is both, an early application of this notion, in of the situ- and it is a mistaketo thinkthatthe two the guise of the "structure ation," appeared in Fortes' 1937 ac- processescan be separated in practice. Indeed we would argue thatdescription magic among the Talcount of fishing lensi, where he describes a situation is "merely" analysis using yesterday's satisfyingall the criteria for non- categories,just as analysis is "merely" in terms description of tomorrow's catecomplex tribal conditions set out in gories. sectionII of Eisenstadt'sarticle. We agree thatthe conceptof a social to write But Eisenstadtis not trying does not in of recentsocial an- fieldcontainingsubsystems history a definitive are not itselfhelp to explain the principlesof and such comments thropology, We would not accept thean- division withinthat field,or the interimportant. that relationsbetween them.We would be dichotomy thropologist-sociologist he implies, nor the picture of British verymuch surprisedif it did. Rather, as a brave band we would assert that the adoption of social anthropologists settingout to tackle all so- the "field"conceptin thestudyof comof brothers ciological problems in glorious ig- plex societies,where so much of social norance of the workgoing on in other behaviour is compartmentalized, is a countries and in cognate disciplines; necessaryfirst step towardsthe formubut this also is marginal to his argu- lation of some of those veryproblems that Eisenstadt raises. Their solution, ment. A more seriousmatterof principleis it seems to us, calls, not for an attack and betweendescription the dichotomy on the primaryconcept of the social analysisthat Eisenstadtmakes,particu- field, but for the development of a
210

Comments

body of auxiliary analytical tools, of which"network" mightbe one. One of us has made tentative use of the notion of "feed-back"for exploring the interrelations of the subsystems within the social field,thoughit is clear that this still needs much more carefulworkingout (Epstein 1958: 237; cf. Mitchell 1960). Eisenstadt'santitheticaltreatment of description and analysisappears to have deflectedhim fromtacklingan important and as yet unresolved issue that arises fromthe studyof complex societies. For Radcliffe-Brown, the goal of social anthropologywas the discovery of social laws. One of the reasons he remained lukewarmtowardsplans for tacklingthe studyof complex societies was his expectationthattheregularities in social lifewhichhe believed to exist in tribalsocietywould, of necessity, be absent in complex societies,or at least would be much harder to uncover.Despite Radcliffe-Brown's assertion that few,if any, social laws have yet been established,there are many social anthropologistswho still hope to find them,even in complexsocieties.Others of us are contentto describesocial life as perceptively as we can, withoutworryingtoo much about the discovery of social laws (cf. Barnes 1958: 65-67). Eisenstadtcontinually refers to our DroCURRENT ANTHROPOLOGY

This content downloaded from 146.164.3.22 on Thu, 23 Jan 2014 08:26:37 AM All use subject to JSTOR Terms and Conditions

Eisenstadt: STUDIES OF COMPLEX SOCIETIES social fessionaltaskas "understanding" phenomena;but although he seems to more ciety as a whole has a curiouslyold- hended as totalitiesand that such an implythatthisentails something fashionedring. Of course people, both apprehensionis a valid analytictool. it is never clear to than "describing," individuallyand as groups,have needs means in opus what "understanding" that (thoughit is less plain, pace Durkheim By MEYER FORTES* He holds,we infer, erationalterms. what in precisely it is betterto analyse than to describe, and Radcliffe-Brown, admirableelucidation of Eisenstadt's and triesto help us over this hurdle. sense societieshave them); but I think followed, stuckon thefence that fewmodernsocial anthropologists the postulatesand procedures But he leaveshimself by Britishsocial to explain social rulesor norms often only implicitly, dividingdescriptionfrom social laws, attempt to a is most instructive anthropologists sitting on a softcushionof understand- by referenceto such needs. Thus, one does not seek to explain a particular practitionerof the craft,like myself, ing. of whose instinct is to shy away from marriagerule, forexample,in terms We agree with Eisenstadt that, in His analysis discussions. of social change, sexual need, or the need to procreate; methodological dealingwithsituations takenforgranted. is, in my judgment, correct as far as it theseneeds are surely mostof us have taken for grantedthe existenceof "new networksof organi- Explanation of such a rule is sought goes, though it needs a minor adjustzation," particularlywhen these have for ment of perspectivein one regard. I rather in termsof its significance been such thingsas a colonial adminis- the person who acknowledges it, its do not think that our "models" and conare as systematically a miningcompany,or a Chris- implicationsforothernormsand insti- "mechanisms" tration, tian mission.But this assumptionhas and (wherethere ceptualised or as deliberatelyapplied tutionsin the society, They expositionsuggests. as Eisenstadt's often been made quite explicitly,to Explanation in is evidence) its history. and forfieldresearch providedirectives delimita manageable area of inquiry, termsof some non-biological "social" operarather than analysis, theoretical frame- need-for example, the continuanceof and notbecause theseimpinging workshave been thoughtto be beyond society itself-is hardly more satisfac- tional rules. In Eisenstadt's terminolat least Recently, one of us (Barnes tory,and often amounts to little more ogy thereare, I would suggest, scrutiny. that as such directives. two models serve sally in the 1960) made a preliminary already is than the assertionof what fieldof politics to bring some of these sufficiently obvious: that if a particular In homage to theirclassical exponents, phenomena into our analysis. norm is quite compatible with other one might be called the Malinowski norms in the society, model and the other the Radcliffeinstitutionalized both it and they are likely, all other Brown model, thoughas we well know By J.H. M. BEATTIE* they also owe much to others, e.g. being equal, to endure. things Durkheimand Mauss. The Malinowski Eisenstadt'ssurveyis both able and (3) This leads to the third general model directsthe investigator to begin concerncerMy only comments timely. on which I should venturesome from an empirical isolate of custom, point tain aspects of the theoreticalframecritical comment: Eisenstadt's e.g. the Kula, or the Chisingu(Richards work which he sets out, very compe- mildly that most modern anthro- 1956); the Radcliffe-Brown assertion model diin the first two or threesections tently, of so- rects him to begin by establishingan pological studiesexplain patterns of his paper. of "total" analyticalor paradigmatic analysis cial behaviorthrough isolate of so(1) I feel that Eisenstadt'sformulasocieties.It has been pointed out often cial structure-e.g. a lineage systemdisthe to tends overstate tion slightly enough in recentyears(e.g. by Popper or of ideology (thought,belief,value, tinctionbetween social behaviour and in The Povertyof Historicism)that a etc.)-e.g. a totemic system. Evansbe are said to norms which the various is not a datum foranalysis; Pritchard'sThe Nuer and my own Dysociety I think total operativein thesocial structure. analyse are what social anthropologists namicsof Clanship among the Tallensi would that some social anthropologists and theyattemptto were based fairly explicitly on this institutions, social values,and beliefs(behold thatnorms, demonstrate functional relationships model. But in mostof themonographic tween which Eisenstadt does not disbetween these and other social institu- studiesreferred threesecto in the first tinguish)are betterregardedas aspects tions,as Eisenstadtveryably shows in tions of Eisenstadt's discussion, the beof, ratherthan as related to, social paper. What they cannot do is to models are, to a greateror less degree, account his haviour,so thatno meaningful them to any such entity as a relate mixed; and, of course,some of themost of the lattercan be given withoutreffor thereis no such en- significantethnographicwork of the "total" society, The tendencyto erence to the former. tityto relate themto. years(e.g. Schapera'sHandpast thirty leads to reifythisanalyticaldistinction is well aware book of Tswana Law and Custom),to I am surethatEisenstadt of modof the interests the formulation of the importanceof these considera- which there is no allusion in Eisenas in such terms ernsocial anthropology tions and their implications-indeed, stadt'sarticle,is independentof either "studies of the ways in which major the developmentof his analysis shows model. norms . . . are upheld by individuals," I thinkthatmymajor However, if we consider only these withitssuggestion thathe is. In fact, a formulation which, would concerncertainof the twomodels,we can see thatEisenstadt's criticisms to conformto of a population striving in whichhe setsout the theoreti- mechanism1 (sectionII) has relevance a set of normsapprehended explicitly, terms ratherthan only to the Malinowskimodel. An emforhis study, the actual con- cal framework somewhat misrepresents itself. But these terms pirical isolate is a compositephenomto whom,for the the framework cern of fieldworkers, and especially enon,as Mauss showedin The Gift, quite important, are,perhaps, mostpart,normsare not given as data, withrecentwork the essentialunityamongstits"aspects" but are rather inferred through the forreadersunfamiliar rigor- -to use Malinowski'sterm-lies in the Unlesswe are fairly analysisof social (includingverbal) be- in thesefields. ous, it is all too easy to slip back into factthat all can be seen as activitiesof haviouritself. between"behaviour" the "same people." I do not see any dif(2) Eisenstadt's characterization of theold dichotomy otherthan thatof scale in applyand "ideas," into Malinowskian con- ficulty as modern Britishsocial anthropology an ing this model in a "complex" society. concernedwiththe explanation of-for jecturingabout "needs" as affording example-the economic or political ac- explanatoryprinciple in social anthro- What has hithertohappened is that it tivitiesof social groups in termsof the pology, and into the "holistic" fallacy has been misapplied.It cannot be used "needs" of these groups and of the so- that societies can somehow be appre- in the studyof a village or a defined

Vol. 2* No. 3* June 1961

211

This content downloaded from 146.164.3.22 on Thu, 23 Jan 2014 08:26:37 AM All use subject to JSTOR Terms and Conditions

geographical area, which is not an em piricalisolate in the sense I have given to thisterm.But it could be applied in a study of the Christmasfestival in England,forinstance.This is a national festivalexactly parallel to the Kula; it has economic, religious, political, philosophical,educational, etc. "functions"(sc. "aspects"). If I understandhim correctly, Eisenstadt's mechanisms2 and 3 are more relevantto the Radcliffe-Brown model. But I have a feelingthatEisenstadthas not identifiedthe distinctivefeatures of this model as clearly as he has perceived those of the Malinowski model. In terms of the Radcliffe-Brown model, thereare no "family, kinship,or political activities" (my italics). For such concepts as "family," "kinship," and "political' denote arrangements of social relations,and theseare manifested in all the "activities" thatcan takeplace in a society.They are analytical conto anyone empirical cepts,notreferable isolate. In this model we are not concernedwiththe "same people," or with interrelations of activities,or of "culture" and "social relations." We are with concerned,preciselyand strictly, the matrixof social relations which is analytically prior to-in a sense outside to by of-the values and normsreferred Eisenstadt.If thiswere not the case we could not have such general,comparaas matrilinealdescentor tivecategories segmentary political system.The anhas been to thropologicalcontribution demonstratein the simpler societies that the matrixof social relationsconand multidimensional stitutes a unitary system.Traditionally, the territorial, kinship,and political dimensionshave receivedprimary What recent emphasis. structuraltheoryhas investigatedhas of thesedimenbeen the interrelations sions, as in the studies by Turner and Smith cited by Eisenstadt.Two other to this develimportantcontributions opment are A. W. Southall's Alur Society (1954) and L. A. Fallers' Bantu Bureaucracy (1952). This model, in its pure form,is of to apply in a complex course difficult society.But recent studies of caste (cf. CambridgePapers in Social Anthropology,No. 2, 1960,edited by E. R. Leach) show thatit is as applicable to Oriental civilisationsas to the preliteratesocieties classicallystudied by anthropologists.That it is applicable also, given and guidance,to industrial theresources Westernsocietywas shownover twenty years ago in Lloyd Warner's series on Yankee City. The problem,as I see it, is not to look for "typesor groups or institutional units . . . similarto those studied in tribalsocieties."Where this is done, thereis danger of lapsing into triviality and anecdotage. The task is,
212

mary groups"and "role theory." On the other hand, thereis a great paucityof effort directedto drawingtogether the resultsderivedfromworkin thesevarious fields.Parsons (1937, 1951; Parsons and Shils 1951) has tried to do so on the broadestlevel, and Homans (1951) on a more specificone; Nadel, in his posthumously publishedwork,used material from sociology and social psyBy JACK GOODY* chologyas well as anthropology in his Eisenstadt's reviewraisesa numberof attemptto develop The Theory of Sointeresting issues. The point I wish to cial Structure(1957). But then Nadel, take up is incidental to his treatment unlike many anthropologists, was not of specific material,but myexcuse is its contentmerely to cultivatehis own garimportance,as I see it, for futurede- den; he wanted to see whatwas coming velopments. up in the fieldnext door. In general, Flattering as it is to British social the attemptto develop a social anthroanthropology to be singledout forsuch pological approach-a process inevitconsideration, I have certainmisgivings ably tied up with considerationsof about framing a discussion solely academic status-has often led to an around its achievements, especially in emphasis on "difference" rather than a field that overlaps so definitely with similarity; and this in turnto a restricother social sciences,which share, to a tionof interests whichhas in somecases considerabledegree,the same theoreti- inhibited possible developments. Incal background. In thepresent instance, deed, it has resultedin a limitationof forexample, it would surelyhave been techniques as well as of approaches; an advantageto include some consider- the social anthropologist is rightto ination of such work as Whyte's Street sist upon the advantagesto be derived CornerSociety(1943), Davis and Gard- from the field-work methods he has ner'sDeep South (1941),as well as some customarily used, but to ignore other of the otherstudiesderivingfromWar- techniques is, as Nadel remarks,"obin the U. S. A. But ner's investigations viouslyshortsighted" (1951: 7), and the apart from this concentrationon the sooner he gets over his "traditional work done in one country, it seems to distrusts" (Barnes 1959),thegreater will me in manywaysa mistaketo consider be his contribution to thestudyof man. a ding an sich-an social anthropology Eisenstadt's review prompts this errorto whichmanyof its practitioners rather programmatic because statement, are prone. in some respectsit points to the same The "anthropological model" used by conclusion.For example, to encourage most investigators whose work is dis- the interchangeand development of cussed here is largely derived, either resultsand hypotheses, we need to take or implicitly, from explicitly Durkheim, an intermediary step and put more Simmel,and Weber. These same figures emphasison the sort of codification of have, of course,providedpoints of ori- results that Merton (1957) has so adentation for other social sciencesmirably performedfor some areas of sociology, social psychology,and be- sociology.One formwhich such codifihavioralstudiesin politicalscience.The cation should take is the specification of the mechanismsthat Eisenstadt of widespreadmechanisms first of social life, mentions,"the interactionof the same but in such a way thattheiroperations situa- can be empirically people or groups in different validated. However, tions," appears to be similar to what the analysisof such mechanisms, surely Simmelcalls "multiplegroup-affiliations an essentialpreliminary to the developmentof any sortof "functional"theory withina singlegroup." Simmelgives,as a typicalexample of this,"the compe- (Merton1957; Nagel 1957),should be of kind than thatsuggested tition among persons who show their a niorespecific in otherrespects"(1955: 155). by Eisenstadt.My main point bears on solidarity and The discussion of conflictby Simmel the futureof social anthropology is tangentialto the articleitself.Eisenand othersociologists(Coser 1956) has considerablerelevanceforanthropolog- stadt has attemptedsome drawing toical developments (Gluckman 1954, getherof threadsof the kind the field 1956), just as his work on sociability, requires;on the otherhand, one of the of treatingthe contribution games, coquetry,and conversationhas side-effects recentlyled Homans to speak of Sim- of social anthropologyas a one-way mel as "an ancestorof ... small-group processmay be to add to the growing isolationisttendenciesand thusrestrict research"(1948: 597). In view of the common theoretical the interchange so necessaryto its fursourcesand overlapping. therdevelopment.To paraphrasea reproblems,it is not surprising that many convergences cent remarkof Evans-Pritchard's, most exist.Eisenstadt mentions workon "pri- of our intellectual capital came from
CURRENT ANTHROPOLOGY

quite simply, to see if and how the model can be applied. There may be a closer structuralparallel between the tradeunion system of a European coun tryand the corporatelineage system ol an Africanstate like Ashanti,than betweenthe latterand the so-called"kinship network" made so muchof in some currentwritings.

This content downloaded from 146.164.3.22 on Thu, 23 Jan 2014 08:26:37 AM All use subject to JSTOR Terms and Conditions

whenthose investments; earlier overseas "we are in dangerof fallare exhausted, (1960: 24). ingintomereempiricism"

Eisenstadt: STUDIES was defended on the grounds that a science of social structuredepended upon comparative analysis of institutional formsand mechanismsderived fromthe functionalstudyof whole societies. The firststudies by American adhered to the social anthropologists volume same position.The co-operative edited by Eggan (1937) was preparedas by his stua tributeto Radcliffe-Brown of Chicago, and dentsat the University exhibit the the separate contributions In mark of his intellectualfatherhood. even those who had not been contrast, directlyaffectedby the newer "funcThis tionalism"began to feelits effects. in a volume sponsored by is reflected Linton (1940), which described acculturation among several American Indian tribes. however, It should be remembered, thatby 1930Warner(Warnerand Lunt major study 1941) had initiatedthe first utilizing a social-anthropologicalapproach. The analysis presented in his volume adherescloselyto the crossfirst sectional synchronic approach. But when he (Warnerand Low 1947) faced system the taskof analyzingthe factory of Yankee City,Warner eitherfound it necessaryor deliberatelychose to prein the contextof hissent his findings torical change and of the relation to outside economic forces.He also utilized a singleeventas the point of analyticfocus.It seems reasonable to conclude that,in this instance,the nature of of the data required a modification the synchronic and structurallydeof the positioncharacteristic terministic of Radcliffe-Brown. social anthropology

OF COMPLEX

SOCIETIES

By SOLON T. KIMBALL* A criticalevaluation of Eisenstadt's and brilliantanalysisof comprehensive of social anthropology thecontribution to thestudyof complexsocietiesshould forone or moreof probably be reserved on the his British colleagues.Certainly, basisof the evidencewhichhe adduces and ofmyown morelimitedknowledge of social anthropolof thedevelopment ogy in England, he is justifiedin the to whichhe comes.It would conclusions be improper, dangerous,and erroneous thatthe development however, to infer, of social anthropologyin the United Statesparallels that in England. It is a have curiousfactthatthetwotraditions so widelyin a quarter-century. diverged Perhaps CURRENT ANTHROPOLOGY will someday publish,as a companionpiece to Eisenstadt's paper, a definitivearticle in which this contrast may be drawn. In the meantime,I should like to make some observationson the directionin which some Americansocial havemovedin handling anthropologists problemsconnectedwiththeanalysisof complex societies. developmentof soThe differential in England and the cial anthropology United States is a fact of some significance, and has led to quite different consequencesin the problemsposed by the application of its methods to the studyof complex societies.Quite early in the developmentof social anthropology among Americans,the concept of change became a central concern. Consequently,the theoreticaltools, or "model," differedgreatly from their and acceptance in The rapid growth in England as summarized counterpart anthropologyas the 1930's of applied by Eisenstadt. Emphasis upon norms for anthropologists a activity legitimate and "regulative mechanisms"was rean incalculable had also undoubtedly placed by attention to "systems"and were offered opAnthropologists theirinterdependencies. this effect. Inevitably, in industryand to work portunities led to concern with the "conditions" The Bureau of Indian Afgovernment. withinwhichthesesystems operated,as ofAgriculture and the Department fairs sourcesof modifications which affected were particularlyreceptive to studies the characteristics and operation of a and counselon problemsof innovation system.The techniques of field work in the period immediatelypreceding began to alter with the shiftfromthe of spesearch for functional generalities ex- the war. In addition, a number of so- cial studies,such as food habitsand the pressingthe "directinterrelation were uncial behaviorto group and institutional education of Indian children, dertaken.Basic to all such enterprises structure"to the specification of who and studies was the concept of underdoes whatwithwhomin timeand space standing social and cultural change. (Chapple 1940). dimensions in It seems to me that the difference with Those who concernedthemselves emphasis upon the concept of change theseproblemsof contemporary society the crucial point in explain- were considered to be eithersocial or represents of Britishand Amer- applied anthropologists, ing the divergence and in many ican social anthropology. The insistence instancesthe termswere used synonyby Radcliffe-Brown (1952) upon syn- mously.In effect, responseto influences chronic analysis has apparently been was and demandsoutside anthropology retained as a central tenet by British shaping theory and technique. social anthropologists.This position of thisline of develExemplification Vol.2 *No. 3 *June1961

opment can be found among some simplesowhosestudieswerein strictly cieties. Sharp's (1952) analysis of the consequences of the introduction of steel for stone axes among the Yir Yoront could be considered a classic example of the utilizationof the conand interdependence, cepts of system, change. In fact,the casebook in which it appeared (Spicer 1952) is entirely oriented in this direction. Oliver's feasts (1955) analysis of status-giving the exemplifies among theSiuai further applicabilityof analysiswhichdepends upon recording who-does-what-withas the necessarybasic what-conditions the interconnecdata for determining sysof institutional tions and functions the same technique tems.Substantially and analyzingwas utilized of observing in theBanks WiringRoom studyof the WesternElectricstudy(Roethlisberger and Dickson 1940) in theearlynineteenthirties.A more general but equally utilization,based on examinafruitful tionof thebroad canvasofsocietalcharand change,was achieved by acteristics Coon (1951) in his analysisof theArabic ofNorthAfricaand theMiddle cultures East. Arensberghas demonstratedthe utilityof the approach in several analyses of industrial and market microexbut has also successfully systems, tended it to the characterizationof American settlementpatterns (Arensberg 1955). These referencesby no means exhaust the list of those who approach have found thisa satisfactory of types fordealing witha wide variety of problemsand levels of societal complexity. should also be offered Briefcomment about certain other specificsincluded by Eisenstadt.His examinationof the value of the concept of "social field" leads him to grantthat it does "tell us that we have to look for some general divisions of society and their interrewhere it does numberof problem-areas not help. The conceptof social fieldhas not been utilizedbyAmericansocial or applied anthropologistsand, in my withtheconis incompatible judgment, cept of interdependentvariables and conas a theoretical hence unnecessary structfor purposes of analytical analysis. In contrast,the concept of "network," to which Eisenstadt responds favorably,has been used for nearly three decades and has been given by linkingit to actual specificity greater theutilizaevents(Whyte1943).In fact, tion of "event analysis" (Kimball and Pearsall 1955) as a technique for overinherent comingmanyof thedifficulties in the studyof complex societies may 213
lations . . . ," but he then indicates a whom - in - what - sequence - and, under-

This content downloaded from 146.164.3.22 on Thu, 23 Jan 2014 08:26:37 AM All use subject to JSTOR Terms and Conditions

Machine Age, The Golden Wing,etc.) startedaround 1935, much of it being specifically concernedwith the application of anthropologicaltechniques to and urban thestudyof industrialisation migration. Again, Eisenstadt'sselectionof writings on peasant India is veryarbitrary. My own view would be that the most towards significant recentcontributions the understanding of thiskind of complexity have come from Gough (see in Gough 1959), Dumont bibliography (1957), Dumont and Pocock,and Barth (1960). Eisenstadt mentions none of these. I also complainat thedeceptivereferences to social anthropology in France. L6vi-Strauss is mentionedat the beginning and again at the end in contexts which suggestthat his work is of the same generalkind as thatof the British in thebody authorsdiscussed morefully of the essay.This is untrue in a quite fundamental way. Two further points seem to me misleading. The quotation from Barnes (1959) on page 205 is quite inapplicable to the set of studies mentionedin the previous paragraph. Thus, Freedman field (1958) is not based on firsthand workat all, while Bott (1957) is of especial interest precisely because the methodologyof research ran directly contraryto the scheme of procedures By EDMUND R. LEACH* here indicated. Finally,I findit veryodd thatEisenEisenstadt'sessayseemsto me out-ofbalance. The crucial distinction on stadtshould considerthatsocial anthropage 203 is between"total" or "tribal" pologistshave failed to make a critical societies on the one hand, and "com- appraisal of the concept of "role." plex" societies on the other. The cri- Nadel (1957) is almostwhollypreoccuteria are not clear. No doubt there is pied with the applicabilityof this cona sense in which Londoners and Nor- cept to anthropologicalstudies; and I wegians belong to "more complex" so- should have supposed that most social cietiesthando Nuer or Tallensi "tribes- anthropologists are now veryfullyalert men," but when Eisenstadtimpliesthat both to thevalue and to the limitations Malay peasant society is "complex," of thisconceptas a tool of analysis. while modern Uganda is not, the distinctionis obscure. Social anthropolo- By DAVID G. MANDELBAUM* gistshave been paying close attention to African states for the past thirty survey sugEisenstadt's noteworthy years, but while Eisenstadt mentions gests to me certain implicationswhich Nadel on the Nupe as a studyof "com- deserve mention. It shows how markplexity,"he treatsKuper on the Swazi edly the range of studies in social anas a studyof "total society."He alto- thropologyhas broadened to include getherignoresSchaperaon theTswana, more work than was formerly done on Gluckmanon the Barotse,or Fallers on complex societies.These recentstudies the Soga-to name only three. Since are impressive in number and quality. Gluckman (1955) and Fallers (1956) Eisenstadt legitimately confines this went out of theirway to point up the paper to the workof Britishsocial anparallels between African and Euro- thropologists, althoughhe does mention pean institutions, the omission seems some similarand converging trendsin serious. the work of anthropologists of other The references to Freedman's work affiliation and nationality. Since theconmightbe thoughtto imply that social cepts of "social network"and "social anthropologists have only recently field,"as discussedin thispaper, apply turnedtheirattentionto China, but in also to anthropologists, it would be usefactthe workof Fei Hsiao-tungand his ful in a further surveyof this kind to colleagues (e.g. Peasant Life in China, include the broaderand more complex Earthbound China, China Enters the setting. prove to be a valuable addition to method. as the discipline Social anthropology, and funcconcerned with the structure tioningof society,should supposedly be concerned with the "simultaneous operation of different types of regulative mechanismswithin a given society." Eisenstadt feels that British studieshave not social-anthropological been verysuccessfulin this area. I do not believe that comparable American studieshave been verysuccessfulhere either. Perhaps the real question is whether or not we should be too much concernedwiththisproblem.Physicists have not yetbeen able to encompassthe of the universein one equadynamics tion.Anysuchcomparableachievement in the area of social science similarly appears unlikelyin the immediatefuture. In the meantime,studiesof complex societies can be concerned with moremodestand manageableproblems. In conclusion, Eisenstadt is to be highlycommendedfora most intricate and carefulanalysisof thestateof social anthropology among British anthropologists.If his article has meritin itself,it has double merit in providing the stimulusfor a comparable analysis of social anthropologyin the United States. 214

Implicit in this paper is a contrast between these social-anthropological studies and the studies of complex societies which have long been made by historians,political scientists,sociologists,and others.While the contrastis implicit, the answer to the question which rises fromthe contrastis quite given. Does social anthropolexplicitly and useful conogy have a distinctive tributionto make toward the understanding of complex (i.e., non-tribal) societies?Its contributionis both distinctive and very useful, Eisenstadt concludes. The method of field work and the main model of analyticprocedure are used to produce studieswhich and analysisas no combine description otherstudiesof social behaviordo. Although this method and model were of tribalpeoples, developedin thestudy theyhave been usefullyapplied to the study of civilizational societies,being in showingthe conespeciallyfruitful in thosesocietiesof tinuingimportance kin,and local groups,of basic domestic, images, and of institutionalinterrelatheirresearch tions.Moreover,through in the civilizationalsphere, social anhave become more cogthropologists nizant of thosesocial forcesand bonds whichare not encompassedwithinparticular groups-and may not even be groupof anyspecific integral elements but which can powerfullyaffectbehavior. In showinghow similar social processesoperatein bothtribaland complex societies,these studies provide corrective evidence for the commonlyused, of sociologiand over-rigid, dichotomies can Anthropologists cal classification. take particular note of Eisenstadt's statementin Section VIII that these studies have contributed to the reof the sociological image of assessment modernsocieties.The methodand the are thus model of social anthropology usefulforthe studyof all societies-for all typesof contemporary societies,for evidenceof past societies, peranalyzing haps even (thoughEisenstadtdoes not generalprocsay this)forextrapolating forfuessesto indicate the possibilities turetrends. But, usefulas thisapproach is, it is not yetfullyenough adapted to the studyof complexsocieties. Social anthropology,we may infer the offers fromthis thoughtful survey, advantages and the defectswhich are inherentin its stage of developmentas a discipline. That stage has been achieved by focusing on functioning by tracing groups and clear groupings, the relations among groups and instiof tutions,by noting the contribution individuals and institutions to the maintenance of the groups. This was done in the contextof a total society, taken as relativelyhomogeneous and
CURRENT ANTHROPOLOGY

This content downloaded from 146.164.3.22 on Thu, 23 Jan 2014 08:26:37 AM All use subject to JSTOR Terms and Conditions

relatively isolated. Only passing notice is usuallytaken,at thisstage,of the impact of forces externalto the group being studied,of internal variations,of alternative choices,of situationaldifferencesin structural alignments. This procedureserveswell in formulatingthestructural anatomyof a social entity; crucial articulations of the component parts are shown, and the contributions and of indiof institutions viduals to structuralmaintenance are demonstrated. The resultantanalysisis given living semblance by appropriate examplesfromreal behavior.It can be argued that this simplification was not onlya first stage in the developmentof social anthropology, but also a necessaryfirst step in the analysisof any social entity. Beyondthe first stageor thefirst step, the need for further developmentsof scope and of conceptbecome apparent. Second-stagemodels are in order, not only because of the nature of complex societies,but also because of the rejection (often violent, sometimespremature)of analysesbased on thesimplified social model. In scope, the local comthe unit mostfrequently munity, taken forobservation, is both too narrowand too gross.It is too narrowbecause the largersocial forces whichaffect itsmembers cannot be adequately understood fromthe studyof any one community. It is too grossbecause concentration on social morphologytends to slight the social choices and alternative courses whichare powerful in individual factors behaviorand in social change.Not that thesematters have been ignoredin the first-stage studies; they have not been systematicallyincorporated into the analysis.These problems of scope are illustratedin the paragraph succinctly of Eisenstadt's sectionV whichdiscusses Barnes' study of a Norwegianparish. The social anthropologistcan augment his close field-observations by on relevant drawing studiesin otherdisciplines which do treatof the over-all society. His work, in turn, has augmented and stimulatedstudies in the otherdisciplines. Even moreimportant, as Eisenstadtshows,is the development of conceptswhich will help link individual behavior, group structure, and institutional activities. A main concept to be developed is that of "role." Originallypropounded by Linton, theoretically sharpened by Nadel, further developed by various sociologists, it has not yet been used to full effect in anthropological analysis. We can agree with Eisenstadtthat it is a key concept at thisjuncture;it maynot be too much to say that as the conceptof atom is to molecule,and as gene is to chromosome, so is role to structure. Effective use of the concept of role Vol. 2 -No. 3 -June 1961

Eisenstadt:

STUDIES

OF COMPLEX

SOCIETIES

may well facilitatemore powerfuluse of the newer concepts mentioned by social field, Eisenstadt: social network, and social organization.All should be used to illuminate the choices,contrawhich are at the dictions,and conflicts root of social and culturalchange.And conceptswhichdeal withsocial changes over timewill add notablyto the scientific value of the structuralanalyses whichhave alreadyproved theirusefulnessin thestudyof complexsocieties.

ByDAVID

M. SCHNEIDER*

appraisal of thecontribuEisenstadt's tions of the English social anthropologists whom he lists, and his masterly have of thekindsof workthey summary undertaken,are verywelcome indeed; for such an appraisal and a summary have been very much needed of late. That he is a triflemore generous in givingcreditthanan obsessivedevotion to hard factsmightwarrantgoes without saying. I do not find it possible eitherto add to or to detractfromhis of the subject. treatment I would raise only one point: Eisenstadt begs the question of dealing directly with "complex societies" and what theyare. He does not definethis category;nor does he do more than it from"primitive," without distinguish either.I am puzthatcategory defining zled by his classingthesmallNorwegian hamletstudiedbyJohnBarneswiththe Nupe, and the Nupe with southeast China, London, and Malabar. Are the Tswana "complex" or just numerous? "complex" Is the Murnginsocial system What is or just difficult to understand? between "modern" and the distinction "complex" which is introducedat one point? Does this mean that the contemporaryAfricans whom Gluckman studiedwere "complex" but not "modthe ern," while the Pentrediwaithians, primitiveWelsh group so devoted to are both "complex" and "modfootball, ern"? If this were a minor problem one would hesitate to quibble. But in the presentcontextit is a problemon two counts: First, it is a major problem which social theoryhas clearly failed to cope with,yetone of the mostpressing of theproblemsbeforesocial theory at this time. Second, it is not easy to of appraise properlythe contributions to an unEnglish social anthropology definedentity. treatment of thisproblem Eisenstadt's seems characteristic of of "complexity" the treatment generallyaccorded it today, which consists mainly in simply to "complexity"and then igreferring noringthe matter.One suspectsthat a professional bias migh.tbe involved.

The professionof sociology has been reared in a climate of opinion which acceptsas a factthe idea still implicitly that "modern" "mass" "industrial" "complex" societyis a thing apart, an a radical departurefrom in itself, entity all other formsto be found on earth. of This was undeniablythe assumption Weber, Durkheim, Freud, Spencer, and, indeed, of many if not all 19th20thAlthough anthropologists. century century sociologists have learned to hold theirtongueswhen theyare about to utter such impieties as "primitive state of society," people" or "primitive theyare on the whole the same persons who vehementlyassert that one just cannot comparekinshipin a "modern" "industrial""mass" "urbanized" "complex" societywith kinship among the Eskimo,or the idea of Fate among the Greekswiththe idea of Fate among the Tallensi. Evidence of this bias in Eisenstadt's paper is seen in his surprisethat mechanisms, regulative belongingas it were double to one type of society,[myitalics, operate on 'one typeof society'] emphasis and ofsocieties types ofother in someparts constitutebasic componentsof social

organization.

theoppositebias precisely from Starting at all. I findthisnot surprising myself, doubted theproI shouldhave seriously fessional competence of those who failed to find evidence of such regulativemechanisms. Again, Eisenstadt seems to have that learned fromanthropologists modern society is not a 'mechanized' live in individuals one,in whichatomized forces. byimpersonal ruledonly separation, perRather,various closelyinterwoven on theone hand, relations, sonalandgroup and with symbolic permeated and relations on the otherhand, meanings, primordial of even the basic components constitute typeof society. mostdifferentiated How, in 1961, is it possible foranyone to believe that "modern" or any other society could conceivably be "mechthat be surprised anized," and therefore like all othersocieties "modern"society, of which we have any knowledge,entails closely interwovenpersonal and group relations,symbolizedand meaningful?One mightsuspect a residuum of Herbert Spencer lurkingabout and foulingthe sociologicalatmosphere. problem is that of The fundamental kinds dealingwiththerangeofdifferent of a of societieswithin the framework It is precisely Weber's of society. theory techniqueof puttingone case in a class by itself,and lumping all other cases into anotherclass,whichhas tended to yield, in the hands of others,the patconclusionthatthe one entlyerroneous 215

This content downloaded from 146.164.3.22 on Thu, 23 Jan 2014 08:26:37 AM All use subject to JSTOR Terms and Conditions

class mustbe of a drasticallydifferent orderfromall othersin all relevantrespects.For the problems Weber dealt with,his separation of the one class fromall others may well have been legitimate.But it has had the unintended and highly undesirable consequence of fostering the illegitimate conclusion that in all other respects, too,the one class mustalso be radically fromall othersocieties.Eisendifferent stadt has expanded the one class but withoutredefining it, and, slightly, I believe, has fallen into preciselythis to error. Otherwise I find it difficult explain his designating as "contributions" of English social anthropology what seem to me to be commonplace and reasonable expectationsabout the structure of any society,regardlessof modernity, degreeof inits complexity, dustrialization, urbanization, or the magnitudeof itsmass. By LAILA SHUKRY EL HAMAMSY* Eisenstadt'sreview of the contributo thestudy tion of social anthropology of complex societies comes at a wellchosenmoment. There is no doubt that the time has come for social anthroitselfand question pology to scrutinize its earlier specializationas a discipline concerned with so-called "primitive," "simple,""tribal" societies.World-wide developmentsclearlyindicate that anthropologists would need to look far and wide to findsuch societiestodaysocietieswhich are not themselves deor thatare velopinggreater complexity, untouched by forces originating in more complex societies.The examination of social anthropological approaches to the study of complex societies,therefore, should be of concern not only to a special group of aninterested in complex sothropologists cieties but to the whole field of anthropology. The designation of a tribal group or subgroup for study as if it were an isolated social unit is of course possible, but would, I believe, offend one of the best anthropological traditionsand one of anthropology's most important contributionsto the other fieldsof social science: namely, the insistenceupon placing any social unit in its total cultural context. Inevitably,the total cultural settingof most societies in existence today includes the impact of complex social, economic,and political forces. Eisenstadt has very aptly analyzed and evaluated the special theoretical model-and, in an indirect way, the methodof research-thatsocial anthropology has contributedto the studyof complex societies. He has also clearly pointed out the shortcomings of social through theapanthropology in trying, plication of its theoreticalmodel and 216

related concepts,to analyze and relate the multiplicityof forces at work in thesesocieties. I have little to add to Eisenstadt's analysis,but I should like to present some practical suggestions that may solve some help social anthropologists of the problemsEisenstadtraises.First of colof all, I suggestthe fruitfulness laborationbetweensocial anthropology and other fieldsof social science in an approach to the study interdisciplinary of those problem areas which Eisenstadt has implied would require more refined tools of analysis and a more as sophisticatedtheoreticalframework well as,-perhaps,better techniques of research. Other social sciences have and own methods been developingtheir theoretical modelsforthe studyof comthoughtheyhave usually plex societies, limited their studies to the types of phenomena which fall within their specialized spheres of interest.Social anthropologycould make use of the gleaned by othersocial various insights the use through sciencesand, in return, of its traditionallyfunctionalistand holisticapproach, help integratethese and a set of anainto a body of theory lytical tools that would explain the of a complexsociety. workings In pointingout theareas whichsocial has failed to analyze adeanthropology of complexsocieties, quatelyin itsstudy Eisenstadtis at the same timepointing research. areas forfuture out important One area of researchwhich I submitis in an a fruitful one forthoseinterested reguof the mechanisms understanding of social belating "the interrelations struchavior to group and institutional ture" in complex societies,is the study of power-groupsor "government"in the broadestsense. This would be particularly important in the so-called under-developed areas of the world, where greater centralizationof power a developingnationalism accompanying of tribal and is forcingthe integration This folk groups into state structures. of power has resultedin centralization instithe emergenceof power-wielding tutions and groups which encroach upon, and may dominate,all otherinof the society.These centers stitutions of power regulatebehaviorand induce and functions changesin the structure of existinginstitutions. They also imin culturally heterogepose conformity neous groupsby settingup and enforcing normsof behavior,eitherthrough coercionand legal sanctionsor through persuasion (through propaganda via thecreamassmedia of communication, tion of ritual symbols,and the like). Such power-groups consciously plan changes in behavior and institutional organization,and impose standardsof conduct which may conflictwith pre-

existingculturalvalues. A studyof the internalorganizationof power-groups, of the participants, the value systems thatdomiinfluences and theparticular nate their behavior, as well as the methodstheyuse to influencethe rest foran unare veryimportant of society, derstandingof some of the forces at Some anthroworkin complexsocieties. of the pologistshave studied the effect on subgroups. central power structure Has the time not come to turnour attention, in collaboration with other to studiesof the centers social scientists, of power themselves? By INA E. SLAMET-VELSINK* The problems raised by Eisenstadt are no doubt of crucial importanceto present-daysocial anthropologists.I was directlyconfrontedwith them a couple of yearsago, when I starteddoing researchin a Central Javanesevillage of the Klaten region; and ever since, I have been tryingto elucidate the principal questionsinvolvedin the analysis of local Indonesian cultures from the viewpoint of modern social and methods. sciencetheories AlthoughI thinkEisenstadt'sarticle I should like to is highlystimulating, formulatesome points in a different way, so as to throw,I hope, another lightupon them. I wishto acknowlFirstand foremost edge, with Eisenstadt, the important made to modern anthrocontributions pology by the specificbranch of social anthropologyhe discusses: especially and detailed the patient,conscientious, fieldworkdone by itsmajor representaand refinement tives,and the strictness of the methods used. These achievementshave obviouslyadvanced the developmentof the social sciencesin general to a very considerable extent,at the tendencies thesame timecorrecting toward broad generalizationswithout as well as the negfactualjustification, studyof concrete lect of the systematic which were and specificinterrelations, inherentin most of the older anthropological literature. thesocial anthropologiNevertheless, has cal approach outlinedby Eisenstadt which have always serious limitations, been present but are becoming more in the face of evidentand troublesome complex and/or rapidly changing societies.In my opinion, theseshortcomings are not inherentin social anthropologyas a discipline,but are due to an exclusivelymicrosocial orientation as well as to an almost pathological fear are These assumptions of evolutionism. easilystated,of courseand say nothing original;bulttheirconsequencesshould be fullyrealized, not only because the study of complex societies has been severelyhandicapped by this lack of
CURRENT ANTHROPOLOGY

This content downloaded from 146.164.3.22 on Thu, 23 Jan 2014 08:26:37 AM All use subject to JSTOR Terms and Conditions

perspective,but because it has frequentlyimpaired even studies of relatively smallsocial entities. An intensive studyof the uniformity, as against the variety, of historicaldevelopment,and ofthedialectics ofsocial change(including processesof diffusion), mightwell proveto be a prerequisite forthe study of complexsocieties in particular,and a necessary requirement forunderstanding theexactnatureof,forinstance, the choice that individuals are offered among different and contradictory institutional forces and groups;or a necessaryrequirementfor grasping the full meaningof political and economicconflictsbetween Europeans and Americans on the one hand, and Africans, Asians, and Latin Americans on the other hand; and explaining why these conflicts could never contributeto the continuous integrationof an ongoing in any way. society This does not mean that social anthropologistsshould turn away from the painstakingstudy of the real behavior of human groups and individuals in concrete historical situations, and of the normsthatgovernor fail to governsuch behavior.On the contrary, thesestudies,if combinedwithbroader theoretical views and interests, may be able to bringabout,amongotherthings, thebadlyneeded refinement and sharpening of evolutionary theory and a deepening of insightinto what really happens, on the physicalas well as on the spiritualplane, at all levels and in every part of societyin the course of minoror major social changes or even revolutions. In other words, these studies,if broughtto the point, might singularlyadvance our comprehension of the dialecticsof social change. By some illustrative material,closely related to my own fieldof experience, I shall tryto make clear the practical of theviewsset forth significance above. Leach (1954), in his studyof the political systems of Highland Burma, with the problemsof the instruggled terrelationsand mutual interdependence of threetypesof social and political organization: the equalitarian, segmentary typewithno political units larger than village-republics(gumlaotype); the authoritarian,hierarchical type,characterized of vilby aggregates lages under a commonchief,by a class systemof an almost caste-likerigidity but still contained within the frameworkof a segmentary tribalsociety, by endless feudingand general instability (gumsa-type);and the third type of feudal or feudal-likestates with centralized authority,dominant in the plains and valleys where agriculture is based on wet-rice culturein contrast to the shiftingcultivationstill extensivelypractisedin the hills. Leach, on
Vol. 2 No. 3 -June 1961

Eisenstadt:

STUDIES

OF COMPLEX

SOCIETIES

the one hand, does not neglectto take the ecological, economic, and general historical conditions into account in analysingthe processesof changegoing on in his composite society. On the otherhand, however, he repeatedly presents the situation he depicts as one offering individuals,as well as groups, a choice between different political institutions; and the dynamics of Kachin political organizationas a continuous oscillationbetween the gumsa and the gumlao typeofgovernment. I mentionthe workof Leach fortwo reasons:first, it is one of the fewstudies by a social anthropologist that attacks the problem of complex societiesfrom a ratherbroad angle; second, the subject matterof his book is veryfamiliar to me, because in several aspects it is strikingly similar to Indonesian conditions,so as to make discussionof methods and interpretations more fruitful. I shall limitmyself to onlyone of the several interesting questions raised by Leach, i.e., the oscillation between gumlao and gumsatypeof organization. I admit that fromthe individual point of view the choice betweenthe two traditions concernedmay presentitselfas a simplealternative; but historically we are confronted here with a case of irreversible social evolution, as amply born out by the Indonesian facts.The gumsa type of organization may be called characteristic ofall theDong'sonmegalithicculturesthat still exist, existed until recently, or existed in prehistoric timesin present-day Indonesian In several instances (Batak, territory. Toradja), former equalitarian traditionsare still clearlypreserved in other sectionsof the same ethnicgroupsor in the folklore(Kruytand Adriani 1912; Kruyt1938; Vergouwen 1933),whereas the transition to a full-blown caste systemand a politicalorganization ofpetty feudal kingdoms under Hindu influence is still traceable in the Balinese case (Korn 1932, 1933). Cultures that are directlycomparable to the gumsa typedescribedby Leach are still partly extant in Nias (Schrbder1917), Flores (Arndt 1929, 1931, 1932) and Eastern Sumba (Nooteboom 1940), but thereis evidencethatSouthSumatraand strong at least the greaterpart of Java were politically organized in a similar way beforethe Hindu period and, in some regions,even much later (Thomassen a Theussink van der Hoop 1932, 1938; Heekeren 1958). The most interesting point in this is not thegeneral special case, however, evolutionarytrend froma segmentary tribalorganization to a patriarchal one, becoming increasinglyfeudal in the course of time (cf. Levi-Strauss1949:

301); noreven theinternaland external factorsthat make this trend operative places, at different times,in different forms;but ratherhow and in different Leach came to consider the transition from gumlao to gumsatypeof organization and vice versa as an oscillation. The gumlao revoltshe mentionswere no doubt real, as was-and is-the constantclash betweenthe two contrasting traditions; but the process involved mightbe far more complex and consequential than Leach seems to believe. The gumlao revoltsare clearlypeasant revolts against nascent feudalism, or oppresfeudal-combined-with-colonial sion; and it is, indeed, not uncommon for the goals of such revoltsto be conceived and formulatedin termsof an idealized past; but this does not mean that the process of historicaldevelopment can ever be reversedor that such a reversal is even reallydesiredby those who reject the present.And especially in the case of the gumlao revolt preone dictedbyLeach forthenear future, should like to know much more about the process of reinterpretation of the old values, which has certainlytransformedthe old gumlao ideals in recent situatimes,and about the present-day of the groups tion and the composition supportingthese ideals. Why is it so hard for these gumlao patternsof organization to die? In what sectionsof the society,and in what sort of activities,were theyso entrenchedas to be able to resist all feudal and colonial pressure?How far was their content and what are theirfuture pomodified, ifany? tentialities, As far as my experience goes, these and otherquestionscannotbe answered by scrupulous field work alone, or by knowledge of general evolutionary theory;their solution can be brought neareronlybymicrosocial investigation studyof the combinedwitha thorough and the specific special characteristics evolution of the cultures in the area concerned.Without an understanding of the special nature of slaveryamong the Kachins, most of the Indonesian peoples, and manyother Oriental peoples, as compared to the same institution in classical Europe; without an insight into the particular varietyof feudalismpredominantin regionseconomically dependent on large-scale irrigation; and without an objective evaluation of colonial and semicolonial conditionsand theirimpacton the several kinds of indigenous societies,the of archaic equalitarian trapersistence and theirchangditionsof organization ing role in rural life may become quite incomprehensible. On theotherhand,it seemsthatcom217

This content downloaded from 146.164.3.22 on Thu, 23 Jan 2014 08:26:37 AM All use subject to JSTOR Terms and Conditions

plete familiarity with the sociological and historicalbackground of the culture concerned,and with the general structuraland evolutionaryproblems involved,cannot preventexpertsfrom misinterpreting facts relating to the "little tradition" (in the Redfieldian sense)in a ratherfrightening way.I am thinking,for instance, of Wittfogel's (1957) evaluationof Chinesevillage traditions. General judgments, such as "total submission" and "total terror," mustseem completely inadequate terms for these traditionsto any social anthropologist who has ever made a real effort to understand thecomplexsystem of values and attitudesof a poor peasant population of an Asian village in a "hydraulic key-area"-that is,one who did not contenthimself withsimplyacceptingtheiravowed attitudestowards theiroverlordsas a general norm. In theJavanesecase,a comparison betweenthe conclusionsof Van der Kroef (1951) and Clifford Geertz (1956) may bring out at once the need to supplement studies of writtensources and generalexperience witha givenculture by careful and patient anthroIn a Javanesevillage, pologicalresearch. archaic equalitarian patternsof leadership and mutual aid may be nothing more than glowing embers,hidden in unspectacularand seemingly minor institutions, aftercenturiesof repression and colonialismalike. But by feudalism theyare not dead; and although I do not share all of Geertz'sviews,I fully agree with him about the importance and tenacityof these patterns. They have, indeed, not only been repressed, but also indirectly strengthenedby feudal and colonial exploitation; and theyhave played and are still playing the crucial role of preventing the and even poorerpeasantsfrom starving, -as I was able to observe myselfin some cases-from losing their sense of self-esteem through their inability to meetthecustomary ritualobligationsor coarse treatment-according through to local standards-by kin or neighbours. In this context, no more than the crudestoutlinesof the problem can be sketched.But it may be remembered that the issue is of prime importance fromthe theoretical as well as fromthe practical point of view. I assume that,in China, millenia-old traditions were consciously revived,reand reshaped by the cominterpreted, munistsin order to smash the feudal systemand to undermine the deeprooted patriarchalinstitutions. And althoughin Indonesia thegoals are partly, and the circumstances differmarkedly, ent, the same kind of gumlao institucalled upon to help t:ionsare officially carryout the land reformand to support the movementfornationala recon218

towards led to push his analysisfurther, as a fundaa generalanalysisof conflict mentalprocessin social life. Face to face with the even greater of modernindustrialsodifferentiation ciety, manyanthropologists initiallyreto the studyof cirstrictedthemselves cumscribable milieux, to use Mills' (1959:8) term: the small group, the enclave, the village, the Negro in the city. This patently inadequate definiset up tion-of-the-subject-by-implication a further reaction: anthropologywas, now, chiefly a technique-"microsociology."Both conceptionsare fused in Barnes' phrase "politics round the village pump." In the end, the "technicalists"are perhaps nearer the truth.The subject matteras traditionally known is "melting away with hopeless rapidity."But the conclusionsthatought to be drawn disciplines. of the subject to are not the restriction milieux withinan the limited study of By P. M. WORSLEY* unexamined general social framework, Eisenstadt's article indicates that, nor the reductionof the subject to the whatever the achievements of social an- statusof a mere technique.Instead, we thropology, past and present, serious should be expanding the subject toof its basic postulatesis reconsideration wards its fusionwith sociologyand relong overdueif it is to have any future. lated social sciences,even if thesethemEven in his first great classic,in 1922, selves are also in need of parallel Malinowski ominously began by re- reorientations. It is quite true, of marking(page xv): course, that anthropologists have long in restricted been researching sectors of Ethnologyis in the sadly ludicrous, the "modern" society.The more enternotto saytragic, position, thatat thevery moment whenit begins toputitsworkshop prising anthropological theorists(e.g. Nadel, Bidney,etc.), too, have increasin order . . . the material of its studymelts ingly turned outwards,away fromthe awaywith hopeless rapidity. theoretical inbreeding of the 1930's. Nor is the attemptof CURRENT AN- Sociologists,for their part, are beginTHROPOLOGY to stimulate the subjectby ning to contemplatethe world outside pushing it back 100 years into an un- western Europe and North America, holy alliance with archaeology and and increasingly utilize "anthropologiphysical anthropologylikely to satisfy 1I have attempted to spell this out in the social or cultural anthropologist greater detail in "The Analysis of Rebelwho is increasingly worriedby doubts lions and Revolutions in Modern British about the validityof orthodoxassump- Social Anthropology" (Proceedings of the
t.ions in,

struction. Goals and means are still interpreted ratherdifferently by different sectionsof the population, and the outcome may still be a matterof hot debate, but the importanceof the phenomena is undeniable. So I think that social anthropology still has an importantrole to play in but the task studying complexsocieties, is not an easy one. In the majorityof rural culturesin underdevelopedcountriesat the presentday, the complexity ofvariousplanes of organization, superin a network of posed and intertwined and contradicintricateinterrelations tions withoutclear-cut boundaries; the clash of old values and symbolswith new social conditionsand the manifold reactions ensuing from these antagoradical shifts nisms;and the sometimes in group-alignment if compared to the traditional status system: these phenomena are all but dazzling. and if only because of Nevertheless, the vital issues involved,this situation should not induce us to give up any of the raisonsd'etre of our discipline as a distinct science,whichare: the studyof and developmentof culture the growth and culturesin time; the studyof societies and their culturesas structural and organicwholes,and theircomparison; and the claim to comprehendobjectivelyand to translate accurately the individualand collective life-experience and of human beings acting,thinking, feelingin cultural settingsother than our own. Never before, indeed, has therebeen such an urgentneed to promote a closer integrationand a reciprocal fertilization betweenthe different fields of anthropology, as well as between anthropology and other related

This crisis is theproductof twointerrelatedphenomena:a changein subject matterand a changein theoretical postulates. Whetherwe consider that the classic anthropologistdevoted himself primarilyto "primitive" or to "colonial" social situations, or both,he now mustconsiderminingtownsand political parties as well as kinship systems. Now, "barbarology"-toaccept the jibe valid science; -was and is a perfectly but the assumptionsbuilt up in the closed sostudyof relatively stationary, cieties prove inadequate for the study world. Indeed, it of the contemporary conflictwas thestudy of industrializing, Africawhichstimulated laden southern the first trenchant critique of the of modernstructuralholisticextremes functionalismfrom within social anthropologyitself,by Gluckman (1940,
1942, 1949).1 As a result, Gluckman was

and about,hisscience.

4th World Congressof Sociology).


CURRENT ANTHROPOLOGY

This content downloaded from 146.164.3.22 on Thu, 23 Jan 2014 08:26:37 AM All use subject to JSTOR Terms and Conditions

grew up...." [p.99]; "this practice is often omitted today . . ."; "the postwar generation are taking a very active role in selecting their own spouses" [p.100]; "educational differences are diminishing . . ." [p.106]; "a new distinction is emerging [p.115]; "motorcycles and leather jackets . . . have become the outward symbol of [the moderately wealthy youth]" [p.125]; etc. (emphasis added).

cal" techniques. Unfortunately. the Eisenstadt: STUDIES OF COMPLEX SOCIETIES have by now absorbed the universities anthropology of the 1930's,and regard pushed further and further, searching tion is no longer adequate. Equally, the setting up of new and separate de- for propositionsof higher and higher wherethevalue-systems are not unitary, of anthropologyas a sine levels of generality,towards the very and the "primordial symbols" thempartments qua non in demonstrating their matu- formal sociology he so capably criti- selves are objects of manipulationand in thesocial sciences. cized.3 rity redefinition, orthodox structural-funcEisenstadtis correctin referring to A searchforabsolute propositions of tionalism cannot help us. Current the "distinct"contributionwhich an- thiskind probablyshowsitsweakestas- studiesof "traditional" societiesdemonthropology has broughtto the studyof pect when called upon to help us ana- strate that the examination of power, and conflict is centralto any complex societies, though his list of lyzeprocessesof change that are highly influence, studies is perhaps formidable enough specific to particular types of social comprehension of social processes, even to distractfromhis major conclusions formationanid to particularphases of forthistypeof society.6 What has been achieved by anthrothatthese successesare verylimitedin social development. This is most evisocieties which pologists working in the "complex" province, and that the key concepts dentin rapidlychanging are closely affected bythegrowing inter- field has largely been the product of elaborated "do not help greatly So that while the "technicalists"have dependence of the world: e.g. the sharp empirical observation coupled of politicalpartiesin the "emer- witha largely been realisticabout the change in sub- growth unexaminedad hoc modiof receivedconcepts.The result have not said as clearly gent" countries;the lineamentsof the fication ject matter, they "Organization Man" in neo-corporate is an uneasy compromise, as he has that what anthropologyhas arisingfrom in struggles underlying conceptual unclarity. I had to offer qua technique is verylim- society;the interest-group ited. Whatever mechanismsand con- post-StalinRussia. All this, of course, don't think this state of affairs is any prob- cause forself-congratulation. cepts have been developed (e.g. the implies the settingof the specific three Eisenstadtmentions),and what- lem or situation in the context of its place in human history.4 Neither relaever the value of new emphases upon 6 E.g., Hart and Pilling (1960) on the norformal has anysuch Tiwi; Turner(1957)on theLunda. sociology processand conflict, etc., these are iso- tivism to offer. By avoiding lated from any integrated general generalframework theory of roles-in-change. Oddly the delineation of these generic procenough, Eisenstadt notes this absence esses of change,we end up with either of any systematicrole-theory as the the relativist's culture-bound, overpropmajor deficiencyin current anthro- specific(and oftencontradictory) general statepological theory, yet neglects to men- ositions,or increasingly By S. N. EISENSTADT ofspecificity ("groupshave tion Nadel's (1957) pioneeringwork in mentsempty leaders"). Paradoxically, the point Since a full and satisfactory preciselythisfield. reply to which structural-functionalismlathe wide range of comments Orthodox structural-functionalists would negenerally assumed a statics-dynamics boured so hard-the analysis of the cessitate another article, I shall limit model which postulated a theory of forces making for the integrationof myselfto certain selected points. (1) It was not my intentionto write structure and a separate theory of society-is perhaps the anthropologist's weaknesswhen he turnsto the any kind of history, survey, or analysis change.2It was also relativistic:it in- greatest of social anthropology in general or of formed us, on one page, that ruling complex society.He is nurturedon a in particugroupsfrequently oppose social change; theoreticalmodel which subordinated Britishsocial anthropology and on another, that they often wel- the analysis of power to the study of lar. My main concernwas to analyzethe ' which muted the factsof application of (if I may paraphrase come it. Nadel, virtually alone amongst reciprocity; forceand conflict;and which Fortes)an analyticalconstruct the anthropological theorists, em- interest, or model and in- to the studyof a certain typeof social barked on the needed analysisof role- highlighted complementariness So, in social phenomenon,i.e., "complex societies." change,role incompatibility, the rejec- tegrative value-systems. This model has been widely used by where the key decisions are tion and redefinition of roles, etc. Un- situations, taken outside the milieu; where "com- maniy-although not all-Britcertainly in avoiding relativism, he fortunately, and it was manding heights"exist; where the de- ish social anthropologists, cisions may be taken by remote elites most convenient to draw principally 2 Where this dichotomyleads one is incited ratherthan by men in face-to-face rela- upon theirwork. The references dicated by this sentence in an otherwise tionships;where the decisions may be were mainlyused to illustrate different excellent field-monograph (Willmott imposed and one-sided: in all these aspectsof themodel and itsapplication, 1960), summing up a chapter on "Comorienta- and were never intended to be exsituations, the primary-group munityStructure": haustive.However,I agreethatI should "In this chapter the analysis has been 3E.g. in his criticismsof Chapple and have mentionedFallers' and Southall's almost entirely structural. Very little attempt has been made to discern changes. Coon, Homans, and Simmel el at., in The studies,because theyillustratecertain Theory of Social Structure. . ." (p- 128). interesting applications of this model; 4 Cf. Mills (1959:128): "Classic social scithe publication on caste Yet the chapter includes the followingpasunfortunately, ence . . . neither 'builds up' from microsages: scopic study nor 'deduces down' from recentlyedited by Leach was not yet conceptual elaboration . . . [It takes up] "They tend to lose their regional loyalbefore me, and therefore I could not substantiveproblemson the historicallevel ties . . ."; "the nationalistmovementwhich commenton it. Mention of Warner's

Reply

of reality"(emphasis added). Some anthropologistsare beginning to spell this out in their field studies. For example, Burridge (1960:112) notes: "Behind Tangu as we find them today lie many years of continuous development and change. . . . Indeed, to argue from a presumption of stabilityin Tangu during the last centuryand a half would be most ill-advised." 5 Cf. Gellner 1958.

been in place; but I did not mention the Yankee City studies because to my mind theyconstitute an application of some of the conceptsand researchtools used in social anthropology, rather than of the analyticalmodel in its entirety. Similarly, I did notrefer to Nadel's con219

Black Civilization would certainly have

Vol. 2 *No. 3 -June 1961

This content downloaded from 146.164.3.22 on Thu, 23 Jan 2014 08:26:37 AM All use subject to JSTOR Terms and Conditions

because theory to role-analysis tribution fromthe point it was ratherimmaterial of view of my analysis,which mainly concernedthe extentto which anthrohave used thisconceptin their pologists studies of complex societies. It is, of course,obvious that other typesof anthropological studies of complex societies exist-mostly, though not exclusively, undertaken by American anthropologists-and therefore I entirelyagree with Kimball that an attemptto evaluate thesewould be worthwhile. (2) My main concern was to point out, fromthe perspectiveof a sociologist looking at social anthropological and limthemain contributions studies, itations of studies of "complex" societies thatused thismodel (or perhaps,as Fortes suggests, two interconnected models). I wanted to show that the use of this model had made very specific of to the understanding contributions complex societies,but that these contributionswere limited by the very strengthof the model; and that in order to overcome these limitationsit to contraposethis model was necessary with other models that had been developed withinothersocial sciences. Beyond the analysis of some of the concepts developed in these studies,I the exact ways did not intendto specify in which such limitationscan be overmade by come. Most of the suggestions Barnes and EpEl Hamamsy,Worsley, seem to be stein, and Slamet-Velsink althoughof coursethey verypertinent, exhaust all the posdo not necessarily sibilities.I certainly agree with Goody to look forvarious about the necessity and with Mandelspecificmechanisms, of developbaum about theimportance models." ing "secondary While Schneider's views about the unity of mankind and the basic simihuman societiesand laritiesof different mechanisms of the regulative operating within them are certainlycorrect,the of different modconfrontation specific els which emphasize such various mechanismshas been rare and could be Fortes' suggestionabout very fruitful. the Christmasfestivalin England, for instance,is verystimulatingfrom this (3) The preceding section is closely connectedwiththe important problem, raised by Schneiderand Leach, of the of defining "complex" socidifficulty eties,as distinctfromsimple (or tribal) societies.Certainlyboth terms, to some extent, constituteresidual categories. But it is possible that an explicit analattribysisof the major characteristics uted to each of these types,as well as the analyticalmodels implied in them, may greatlyenhance our understanding of the extentto whichit is usefulor meaningfulto employthese categories. 220
point of view.

in Swat, North Pakistan," stratification (4) It is of course, well known-as in Aspects of Caste in South India, CeyI should have noted-that this model lon and North-WestPakistan (Ed. E. R. was largelyderivedfromDurkheim,alLEACH). Cambridge Papers in Social thoughin itsapplication it goes beyond 2. Cambridge: Cambridge Anthropology, [ERL*] Press. him. However, I do not agree with University Goody that it is also derivedfromSim- BEATTIE, J. H. M. 1956. "Social anthropology," in The new outline of modern mel. It is true that one of its implicit knowledge(ed. A. PRYCE-JONES), pp, 252mechanismswas extensivelyanalyzed 278. New York: Simon and Schuster. by Simmel,but the distinctcharacter--. 1959. Understanding and explanaBritishJourtion in social anthropology. istic of this social-anthropological nal of Sociology 10:45-60. model is the combinationof all the var- BoTT, E. 1957. Family and social network. ious mechanisms.

(5) Barnes and Epstein criticized my distinction between description and analysis. However, my very point was that most of the studies I cited have combineddescription and analysis, and thatonlyby drawing, in so faras thisis betweenthe possible,a clear distinction twocan we arriveat definite hypotheses about "laws" of social behavior that can also help us understand anyspecific situation. (6) I agreewithBeattie'spoint thata total societycan never constitutea legitimatedatum of analysis.However,it seems to me that many of those who have used themodel analyzedherehave thestrong veryoften, through emphasis as an explanation on social integration of phenomena, unwittingly used total societyas such a datum. Consequently, it would certainly be worthwhileto break up this concept into clearlydistinguishablevariables.

References Cited
N., and A. C. KRUYT. 1912. De Bare'e-sprekendeToradja's van MiddenCelebes. Batavia: Landsdrukkerij.[ISV*] ARENSBERG, C. M. 1955. American communities. American Anthropologist57: 1143-62. [STK*] ARENSBERG, C. M., and S. T. KIMBALL. 1941. in Ireland. CamFamily and community bridge: Harvard University Press. ARNDT, P. P. 1929, 1931. Die religion der Nada'a. Anthropos 24:817-61; 26:353405, 697-739. [ISVW] --. 1932. Die megalithenkultur der Nada'a (Flores). Anthropos 27:11-63. [ISV*] BAILEY, F. G. 1957. Caste and the economic frontier. Manchester: Manchester UniversityPress. BANTON, M. 1957. WestAfricancity;a study of tribal life in Freetown. London: Oxford UniversityPress (for the International AfricanInstitute) in BARNES, J. A. 1954. Class and committees a Norwegian parish island. Human Relations 7:39-58. BARNES, J. A. 1958. Social anthropologyin theory and practice. Arts: the ProceedArts Assoings of the Sydney University ciation 1:47-67. [JAB,ALE*] ---. 1959. Politics without parties. Man 59:13-15. 1960. Indigenous politics and colonial administration,with special reference to Australia. Comparative Studies in Society and History 2:133-149. [JAB, ALE*] of social BARTH, FREDRIC. 1960. "The system
ADRIANI,

London: Tavistock Publications. 1957a. The joking relationship Human Relations 10:179-87. in industry. --. 1957b. Quasi-familial relationships in industry.Human Relations 10:27179. BURNS, T. 1955. The referenceof conduct in small groups: cliques and cabals in occupational milieux. Human Relations 8:345486. K. 0. L. 1960. Mambu. London: BURRIDGE, [PW*] Methuen. R. 1958. Les jeux et les hommes. CAILLoIS, Paris: Gallimard. CHAPPLE,E. D. 1940. Measuring human relations: an introduction to the study of the interaction of individuals. Provincetown, Mass.: The Journal Press. [STK*] COON,C. S. 1951. Caravan: the storyof the Middle East. New York: Henry Holt. [STK*] and M. R. GARDNER. A., B. GARDNER, DAVIS, 1941. Deep South. Chicago: University of Chicago Press. [JG*] L. 1957. Une sous-castede L'Inde DUMONT, du Sud: organisationsociale et religion des Pramalai Kallar. Paris: Mouton. [ERL*] (Eds.) ContriL., and D. POCOCK DUMONT, butions to Indian Sociology,vols. 1 to 4. Various dates. The Hague: Mouton and [ERL*] Co. F. (Ed.) 1937. Social anthropology EGGAN, of North American tribes.Chicago: Uni[STK*] versityof Chicago Press. --. 1950. Social organization of Westof Chiern Pueblos. Chicago: University cago Press. --. 1957. "Foreword" to A natural science of society by A. R. RADCLIFFEGlencoe, Illinois: The Free BROWN. Press. S. N. 1956. Ritualized personal EISENSTADT, relationships. Man, No. 96. --. 1957. Review of: The Negro family in British Guiana, by R. T. SMITH.British Journal of Sociology 8:283-84. EMMETT,D. 1958. Function, purpose and powers. London: Macmillan. EPSTEIN,A. L. 1954. Juridical techniques and the judicial process. (The Rhodes LivingstonePapers, No. 23.) Manchester: Press. ManchesterUniversity --. 1958. Politics in an urban African community. Manchester: Manchester Press (forthe Rhodes-LivingUniversity stone Institute). E. E. 1933. Zande blood EVANS-PRITCHARD, brothership.Africa 6:369-401. ---. 1940. The Nuer. Oxford: Clarendon Press. - . 1951. Social anthropology.London: Cohen & West. ---. (Ed.) 1954. Institutionsof primitive society.Oxford: Basil Blackwell. 1956. Nuer religion. Oxford: Clarendon Press. --.1960. "Introduction" in Death and the right hand, by R. HERTZ. London: [JG?] Cohen 8&West. FALLERS, LLOYD A. 1956. Bantu bureaucracy: a study of integrationand conflict in the political institutionsof an East
BRADNEY, P. CURRENT ANTHROPOLOGY

This content downloaded from 146.164.3.22 on Thu, 23 Jan 2014 08:26:37 AM All use subject to JSTOR Terms and Conditions

Africanpeople. Cambridge: Heffer (for East African Institute of Social Research). [MF, ERL*] FIRTH, R. 1946. Malay fishermen: their peasant economy.London: Kegan Paul, Trench,Trubner & Co. ---. 1951.Elements ofsocial organization. London: Watts. ---. 1954. Social organization and social change.Journal of the Royal Anthropological Institute 84:1-20. ---. (Ed.) 1956. Two studies of kinship in Londcon. London School of Economics No. Monographson Social Anthropology 15. London: Athlone Press. FORDE, C. D. (Ed.) 1954. African worlds. Oxford: Oxford UniversityPress. FORTES, M. 1937. Communal fishing and fishingmagic in the Northern Territoriesof the Gold Coast. Journal of the Royal AnthropologicalInstitute67: 13142. [JAB,ALE*] 1945. The dynamics of clanship among the Tallensi. London: Oxford UniversityPress (for the International AfricanInstitute). ---. 1949a. "Time and social structure." In Social structure (ed. M. FORTES) Pp. 54-84. Oxford: Clarendon Press. 1949b. The web of kinship among the Tallensi. London: Oxford University Press (for the International African Institute). ---. 1953. The structureof unilineal descent groups. American Anthropologist 55:17-41. 1958. "Introduction" in The developmental cycle in domestic groups (ed. J. R. GOODY). Cambridge: Cambridge UniversityPress. FORTES, M., and E. E. EVANS-PRITCHARD. (Eds.) 1940. African political systems. London: OxfordUniversity Press (forthe International Institute of African Languages and Cultures). FRANKENBERG, R. 1956. Village on the border. London: Cohen & West. FREEDMAN, M. 1957. Chinese family and marriage in Singapore. London: H. M. Stationery Off. ---. 1958. Lineage organizationin southeastern China. London School of Economics Monographs on Social Anthropology No. 18. London: Athlone Press. GEERTZ, C. 1956. The development of the Javanese economy: A socio-cultural approach. Cambridge: MassachusettsInstitute of Technology. [ISVi] GELLNER, E. 1958. Time and theory in social anthropology.Mind 67:152-203. GLUCKMAN, MAX. 1940. Analysis of a social situation in modern Zululand. Bantu Studies 14. [PW*] 1942. Some Processes of Social Change Illustrated with Zululand Data. AfricanStudies 1. [PW*] ---. 1949. Malinowski's sociological theories. (The Rhodes-Livingstone Papers No. 16.) London: Oxford UniversityPress. [PW*] ---. 1954. Rituals of rebellion in SouthEast Africa. Manchester: Manchester UniversityPress. ---. 1955. The judicial processamong the Barotse of Northern Rhodesia. Manchester: Manchester University Press. [ERL*] 1956. Custom and conflictin Africa. Oxford: Basil Blackwell. ---. 1958.Analysisof a social situation in modern Zululand. (The Rhodes Livingstone Papers, No. 28.) Manchester: Manchester UniversityPress. --.1959a. Football players and the crowd. The Listener, February,1959. --.1959b. The use of ethnographicdata in social anthropological analysis of

Eisenstadt:

STUDIES

OF COMPLEX

SOCIETIES

Great Britain. International Sociological Association(mim.) GOLDScHMIDT, W. 1950. Social class in America-a critical review. American Anthropologist52:483-98. GooDY, J. (Ed.) 1958. The developmental cycle in domestic groups. Cambridge: Cambridge UniversityPress. ---. 1959. The mother'sbrotherand the sister's son in West Africa. Journal of the Royal AnthropologicalInstitute 89: 59-88. GoUGHF, E. KATHLEEN. 1959.The Nayarsand the definitionof marriage. Journal of the Royal AnthropologicalInstitute 89: [ERL*] 23-34. GULLICK, J. M. 1958. Indigenous political systems in Western Malaya. London School of Economics Monographs on Social Anthropology No. 17. London: Athlone Press. HART, C. W. M., and ANTHONYPILLING. 1960. The Tiwi of northern Australia. New York: Holt, Rinehart and Winston. [PW*] HEEKEREN, H. R. VAN. 1958. The BronzeIron Age of Indonesia. Verhandelingen van het Koninklijk Instituutvoor Taal-, Land-, en Volkenkunde. Deel 22. 's-Gravenhage:Martinus Nijhoff.[ISV*] HOMANS, G. C. 1948. Social behaviour as exchange. American Journal of SociolUG*] ogy 63:6. ---. 1950. The Human group. New York: UG*] Harcourt Brace. KIMBALL, S. T., and M. PEARSALL. 1955. Event analysis as an approach to community study. Social Forces 34:58-63. [STK*] KORN, V. E. 1932. Het adatrecht van Bali. 's-Gravenhage:G. Naeff. [ISV*] --. 1933. De dorpsrepubliek Tnganan Pagringsingan.Santpoort: C. A. Mees. [ISV*] KRIGE, E. J.,and D. KRIGE. 1943. The realm of a rain-queen. London: Oxford UniversityPress. KRUYT, A. C. 1938. De West-Toradjas op Midden-Celebes. 5 vols. Verhandelingen der Koninklijke Nederlandsche Akademie van Wetenschappen, Afdeling Letterkunde, Nieuwe Reeks, Deel 15. Amsterdam. [ISV*] KUPER, H. 1947. An African aristocracy. London: OxfordUniversity Press (forthe InternationalAfricanInstitute). LEACH, E. R. 1954. Political systems of highland Burma. London: G. Bell. ---. 1958. Magical hair. Journal of Royal Anthropological Institute 88:147-56. LEACH, E. R. (Ed.) 1960.Aspectsof Caste in South India, Ceylon and North-West Pakistan. Cambridge Papers in Social [MF*] Anthropology, 2. LE'VI-STRAUSS, C. 1949. Les structureselementaires de la parente. Paris: Presses universitaires de France. [ISV*] ---. 1953. "Social Structure,"in Anthropology Today (ed. A. L. KROEBER), Pp. 524-54. Chicago: Universityof Chicago --. 1959. Anthropologie structurale. Paris: Plon. LINTON, R. 1940. Acculturation in seven American Indian tribes. New York: Appleton-Century. [STK*] LUPTON, T., and S. CUNNISON. 1957. The Problem of social contextin the analysis of workshop behaviour. (mim.) Paper presentedat a meetingof the Association of Social Anthropologists. LITTLE, K. 1955. Structural change in the Sierra Leone Protectorate. Africa25:217Press.

don: Oxford UniversityPress (for the International Institute of African Languages and Cultures). ---. 1949. The Ganni Ritual of Nupe: a study in social symbiosis. Africa 19:17786. --. 1951. The foundations of social anthropology.London: Cohen and West. 1957. The theoryof social structure. London: Cohen and West. [JG,ERL, PW*] NAGEL, E. 1957.Logic withoutmetaphysics. Glencoe: The Free Press. [JG*] NOOTEBOOM, C. 1940. Oost-Soemba, een volkenkundige studie. Verhandelingen van het Koninklijk Instituutvoor Taal-, Land-, en Volkenkunde. 's-Gravenhage. [ISV*] OLIVER, D. L. 1955. A Solomon island society. Cambridge: Harvard University
Press. [STK*] T. 1937.Structureof social action. PARSONS, [JG*]

[JAB, ALE*] NADEL, S. F. 1942.A Black Byzantium.Lon-

of the western Pacific. London: Routledge and Kegan Paul. [PW*] MANDELBAUM, D. G. 1956. "The study of complex civilizations" In Current Anthropology (Ed.) W. H. THOMAS, pp. 203-26. Chicago: Universityof Chicago Press. MARRIOTrT, M. (Ed.) 1955. Village India. Chicago: Universityof Chicago Press. MAUSS, M. 1954. The gift;formsand functions of exchange in archaic societies. [MF*] Glencoe: The Free Press. MAYER, A. 1960, Caste and kinship in Central India. London: Routledge & Kegan Paul. MERTON, R. K. 1957. Social theoryand social structure.2nd ed., revised. Glencoe: The Free Press. [JG*] MIDDLETON, J. and D. TAIT (Eds.) 1958. Tribes without rulers. London: Routledge & Kegan Paul. in MILLER, E. J. 1954. Caste and territory Malabar. American Anthropologist 56: 410-20. MILLS, C. WRIGHT. 1959. The sociological imagination.New York: Oxford Univer[PW*] sityPress. MITCHELL, J. C. 1956. The Kalela dance. (The Rhodes Livingstone Papers, No. 27.) Manchester: Manchester University Press. --. 1960. Tribalism and the plural society. London: Oxford UniversityPress.

MALINOWSKI,

BRONISLAW.

1922. Argonauts

Glencoe: The Free Press. [JG*] --. 1951. The social system. Glencoe: The Free Press. [JG*] PARSONS,T., and E. A. SHILS (Eds.) 1951. Toward a general theory of action.
Glencoe: The Free Press. [JG*] PITT-RIVERS, J. A. 1954. The people of the

RADCLIFFE-BROWN, A. R. 1957. Structure

sierra. London: Weidenfeld & Nicolson.

and functionin primitivesociety:Essays and Addresses. London: Cohen & West,


A. R., and D.
FORDE

RADCLIFFE-BROWN,

sity Press. REDFIELD, R. 1955. Societies and culturesas

(Eds.). 1950. African systemsof kinship and marriage. London: Oxford Univer-

natural systems.Journal of the Royal AnthropologicalInstitute 85:19-32. --. 1956. Pleasant society and culture. Chicago: Universityof Chicago Press. RICHARDS,A. I. 1956. Chisungu; a girls' initiation ceremony among the Bemba of NorthernRhodesia. London: Faber and Faber. [MFf]1

Vol. 2 -No. 3* June 1961

221

This content downloaded from 146.164.3.22 on Thu, 23 Jan 2014 08:26:37 AM All use subject to JSTOR Terms and Conditions

F. J.,and W. J. DICKSON. 1940.Management and the worker.Cambridge: Harvard UniversityPress. [STKAj SCHAPERA, I. 1955. A Handbook of Tswana law and custom. (2nd ed.) London: Oxford UniversityPress (for the International AfricanInstitute). [MFI] SCHR6DER, E. E. W. GS. 1917. Nias. Ethnographische, geographische en historische aanteekeningenen studien. Leiden: E. J. Brill. [ISV*j SHARP, L. 1952. "Steel axes for stone age Australians," in Human problems in technological change (ed. E. H. SPICER), pp. 69-81. New York: Russell Sage Foun[STKr] dation. SHILS, E. 1957. Primordial,personal, sacred and civil ties. British Journal of Sociology 8:130-146. SIMMEL,G. 1955. Conflict.Translated by K. H. WOLFF; The web of groupaffiliations.Translated by R. BENDIX. Glencoe: The Free Press. [JG*] SMITH, R. T. 1956. The Negro family in British Guiana. London: Routledge & Kegan Paul.
ROETHLISBERGER,

bridge: Universityof Cambridge Press. [MF*] SPICER, E. H. (Ed.) 1952. Human problems in technologicalchange. New York: The Russell Sage Foundation. [STK*] SRINIVAS,M. N. 1952. Religion and society among the Coorgs of South India. Oxford: Clarendon Press. THOMASSEN A. THEUSSINK VAN DER Hoop, A.N.J. 1932. Megalithische oudheden in Zuid-Sumatra. Zutphen: W. J. Thieme. [ISV*] 1938. "De Praehistorie,"in Geschiedenis van Nederlandsch-Indie (ed. F. W.
STAPEL). Amsterdam: Joost van den VonTURNER, V. W. 1957.Schismand continuity

SOUTHALL, A. W. 1954. Alur society. Cam-

del.

[ISV*]

in an Africansociety.Manchester: Manchester UniversityPress. VAN DER KROEF, JUSTUS M. 1951. Foreign aid and social tradition in Indonesia. Far Eastern Survey. October. [ISV*] VERGOUWEN, J.C. 1933.Het rechtslevender Toba-Bataks. 's-Gravenhage: Martinus Nijhoff. [ISVW]

W. L., and P. S. LUNT. 1941. The social life of a modern community. New Haven: Yale University Press. [STK*] WARNER, W. L., and J. 0. Low. 1947. The social systemof the modernfactory. New Haven: Yale UniversityPress. [STK*] WATSON, W. 1958. Tribal cohesion in a money economy. Manchester: Manchester University Press. WHYTE, W. F. 1953. Street corner society. Chicago: University of Chicago Press. UG, STKr] WILLMOTT, D. E. 1960. The Chinese of Semarang. Ithaca: Cornell University Press. [PWr] WITTFOGEL, K. A. 1957. Oriental despotism; a comparativestudy of total power. New Haven: Yale University [ISVr] Press. WORSLEY, P. M. 1956. The kinship system of the Tallensi: a revaluation.Journal of the Royal AnthropologicalInstitute 86: 37-75. shall sound. Lon---. 1957. The trumpet don: McGibbon & Kee. YOUNG, M. and P. WILLMOTT. 1957. Family and kinship in East London. London: Routledge & Kegan Paul.
WARNER,

222

CURRENT

ANTHROPOLOGY

This content downloaded from 146.164.3.22 on Thu, 23 Jan 2014 08:26:37 AM All use subject to JSTOR Terms and Conditions