Marxist Anthropology 1975

8/13/2019 Marxist Anthropology 1975

http://slidepdf.com/reader/full/marxist-anthropology-1975 1/31

Marxist Approaches in Anthropology

Bridget O'Laughlin

Annual Review of Anthropology, Vol. 4. (1975), pp. 341-370.

Stable URL:

http://links.jstor.org/sici?sici=0084-6570%281975%292%3A4%3C341%3AMAIA%3E2.0.CO%3B2-0

Annual Review of Anthropology is currently published by Annual Reviews.

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

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained athttp://www.jstor.org/journals/annrevs.html.

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

JSTOR is an independent not-for-profit organization dedicated to and preserving a digital archive of scholarly journals. Formore information regarding JSTOR, please contact [email protected].

http://www.jstor.orgThu Apr 26 11:08:01 2007


http://www.jstor.org/about/terms.html

http://www.jstor.org/journals/annrevs.html

http://www.jstor.org/journals/annrevs.html

http://www.jstor.org/about/terms.html




C o p y r i g h t 1975 All r ights reserved

M A R X IS T A P P R O A C H E S I N

A N T H R O P O L O G Y

Bridget LaughlinDepartment of A nthropology, Stanford University, Stan ford, California 943 5

INTRODUCTION

T h e p u r p o s e o f t hi s p ap e r is t o p r o v id e a n int r od u c t io n to t h e b a s i c co n cep t s an d

me thods of h istor ical m ater ia l ism for an thropologis ts in teres ted in apply ing this

f r am ew o r k in t h e i r o w n w o r k . I t i s n o t p o s s ib l e t o cu t o u t c e r ta in a s p ec t s of

Marxism-such a s the stu dy of cultu re or of pr imitive societies--and assign

them t o an th ropo log ist s ; M arx ' s emphas is on in teg ra ted h is tor ica l ana lys i s ofsocial totali t ies precludes the isolation of anthropology as a dist inct discipline

wi thin the soc ia l sc iences . have in s tead used an th ropo log ica l wo rk , by way of

i l lus t ra t ion and contras t , p r imar i ly to e lucidate Marxis t theoret ical constructs

and methodological pr incip les .

A good deal of th is essay is therefore devoted to cr i t ic ism of convent ional

an thropological theoret ical pract ice . d o not wish to apologize for th is em ph a-

s is . Contrary to some pecul iar ly academic preconcept ions , h is tor ical mater i -

a l ism is no t a fu l ly ar t icu la ted dogmatic gr id (or s t ructural theory) to be

mechanis t ically imp osed o n a ny problem . R ath er i t i s a workin g scient if ic

tradit ion, s truggling to develop theoretical understanding of specif ic histor ical

problem s in a wor ld dom inated by c lass conf lic t . Cr i t ica l the ory , seeking to

penetra te the myst i f ica tions of ou r ow n ideology, is therefore an ess ent ia l asp ect

of th is process of sc ien t i f ic development .

d o , h o w ev e r , w i sh t o s t re s s t h a t m y p r e s en ta t i on d o es n o t en co m p as s t h e f ull

b read th o f per spec tives in the M arx is t tr ad i t ion . Th e v iews expressed here have

deve loped in the g roups wi th wh ich I have w orked , ' bu t the re a r e o ther Marx is t

anthropologists-often wo rkin g in small grou ps and publishing in radical jour-nals-whose progress is no t adequ ate ly rev iew ed. This pa per , th en , should be

tak en as a workin g approximat ion-an a t te m pt to c lar ify cer ta in fundam ental

issues-and a s a very limited in troduct ion to Marxis t con cep ts and method s .

'Critical discussion and research presented by the participants in the Political EconomySem inar at Stanford has been particularly helpful for this review. should especially like tothank Duncan Foley, Jens Christiansen, Michelle Rosaldo, David Howard, Tala1 Asad,James Faris, and Renato Rosaldo for discussion, reading, and criticism of this paper.



34 O'LAUGHLIN

In the first section of the pa pe r, Dialectical Ma terialism, the philosophical

premises which underlie Marxist approaches are summarized and contrastedwith some prevailing epistemological stances in anthropology. The second

section, The Basic Con cepts of Historical Materialism, describes the ways in

which the principles of dialectical materialism have been developed in the

scientific study of human society.

In the third section, Modes of Produc tion, I concentrate on basic meth-

odological pro blem s. This s ection is divided into two par ts: a critical review is

followed by a mo re constructive presentation of the state of the art. Th ese

two parts are interdependent; i t may be necessary to return to the first after

reading the second. The conclusion deals very summarily with the situation of

the Marxist anthropologist working in advanced capitalist society.

D I A L E CT I CA L M A T E RI A LI SM

Historical m aterialism is a sc ience , and like all scien ces, it is based on certain

ontological and epistemological principles that describe what the world is and

how we can k now it. Th e philosophical underpinning of historical materialism is

dialectical materialism, Marx's materialist reconstruction of the Hegelian di-alectic of Mind. In this first section, I will discuss the basic premises of dialec-

tical materialism and then show how its methodological consequences contrast

with some familiar anthropological modes of analysis.

asic Premises

Marx insisted that there are real material structures or regularities in nature,

structures w hich are neither im posed by the human mind nor its epiphenomena1

expression:

Hegel fell into the illusion of conceiving the real as the product of thought con-

centrating itself, probing its own depths, and unfolding itself out of itself,

whereas the method of rising from the abstract to the concrete is only the way

in which thought appropriates the concrete, reproduces it as the concrete in the

mind (43, p. 101).2

Since structures are not simply the prod uct of Mind, then critical theory and

demystification are not enough to change the world. Marx's rejection of

Hegelian idealism in the 844Manuscripts (38) was b ased o n his recognition of

the necessary link between theory and political praxis.Yet in rejecting idealism, Marx did not posit an opposition between the

material and the ideal. He saw that thought could only be the prod uct of real,

sensuou s activity of human subjects and thus that the laws of though t must be

part of the material structures of natu re. In his first thesis o n Feu erba ch, M arx

wro te: The chief defect of all hitherto existing materialism (tha t of Feu erba ch

2Citations of Marx alw ays refer to easily available current ed itions. As will be clear from

the pattern of citation, I take the Grundrisse (43) to be part of the corpu s of M arx 's matu re

work.



M A R X I S T A P P R O A C H ES I N A N TH R O P O LO G Y 343

included) is that th e th ing , real i ty , sensuousness , i s conceived only in the form of

th e object of contemplation, bu t no t as sensuous human activity, practice, notsubjectively (41 , p . 121). Marx abol ished the fa lse d ichotomy b etw een ideal

and mater ia l and thus provided a n epis temological bas is for a mater ial ist sc ience

( thought can a p pr op ri at e t h e c on c r e te ; m a te ri al s t ru c tu re is k n ~ w a b l e ) . ~

Th oug ht is therefore par t of the mater ia l wor ld and governed by t he sa m e law

of d ia lect ical movement that character izes nature . The laws of the mater ia l

wor ld d o no t , howeve r , r educe to the laws o f though t; though t can ne ver

replicate nor appre hend all of m ater ial reali ty . Scientific knowle dge of the wo rld

is apprehending the esse nt ia l determinat ions and suppress ing the unim por tan t .

Sinc e the mater ial world is constantly a nd irreversibly ch angin g, the significanceof par t icu lar theoret ical ca tegor ies and ques t ions wil l change a s wel l .

The s t ructural regular i t ies of the wor ld are mater ia l , bu t they are a lso d i-

a lectical and therefore constan t ly evolving . S ince a ll s t ruct ures are dynam ic

processes-relations betw een being and becoming-the y ca nn ot be kn ow n posi-

t ive ly th rough the i r phenomenal su r face fo rm. S t ruc tu re can on ly be know n

wh en re la t ions are d ia lect ical ly conceptual ized: . . . in general , re lat ions ca n

be es tab l ished a s ex is ting only b y being thought, as d is t inct f rom the sub jects

which ar e in thes e re la tions with ea ch o ther (43 , p . 143). Real i ty canno t be

unders tood o n the sur face of th ings.

nthropological M eth ods o f nalys is

The methodological consequences of d ia lect ical mater ia l ism contras t wi th

mo des o f ana lys is p rac t iced by man y con tem porary an th ropo log is t s. He re I will

d iscuss only three a n thropological approa che s : the 'record ing of pos i t ive fac ts ,

mapp ing on to log ical s t ruc ture s , and synchronic analysis of sys tem s. N on e of

thes e , o f cou rse , is a comp lete methodological f ramew ork , no r ar e these m odes

mutual ly exclus ive . L evi-Strauss is cr i tic ized a t d ispropor t ionate leng th , p r i -mar ily bec ause h i s em phas is o n unconsc ious s t ruc tu re has fo r som e d isgu ised

the bas ic incompat ib i l i ty between s t ructural ism and Marxism.

THE RE CO RD I N G OF POSITIVE F CTS Posi t iv is t approaches that take their

categor ies f rom th e co ncre te i tse l f ar e of tw o types in an thropology: cu l tura l

mater ialis t (etic)s t ra teg ies that observe and measu re fact in th e materia l wor ld ;

an d ethnoscientif ic (emic)methods d es igned to d iscover the cogni tive categor ies

inside of peop les ' hea ds by record ing of essentially l inguistic beha vior .

Whether one acce p ts the p r imacy o f the etnic r ea lm o r tha t of the etic, posit ivismbecomes the on ly a l te rna t ive to t r anscenden t idea l i sm once one accep ts the

ideal imater ia l d ichotomy.

Fr om the v iewpoin t of d ia lectical mater ia l ism , there a re tw o principal er ror s

in pos i tiv ism. Firs t , the con crete m ust be unders tood through conceptual izat ion

of those dialectical relations that determine i t , not through its momentary

surface form. Se con d, thought canno t rep l icate the s t ruc ture of the real wo r ld .

Sinc e knowing cons ists of recognizing and organizing som e facts and not



344 O ' L A U G H L I N

oth ers, we n eve r simply record real ity. I f we can not be theoret ical ly explici t as

to why we hav e noticed some fac ts but not o the rs , asked some ques t ions but nototh ers, then we ar e merel y hiding from ou rselv es the implici t basis of our choic e.

M A P PI N G O N T O L O G I C A L S T R U C T U R E S Like Marx, LCvi-Strams insists that

to understand is to analyze those structural relat ions that underl ie empir ical

reality. But Ma rx's notion of dialectical structur e is materialist and his method

of analysis quite dif ferent f rom Levi-Strauss 's mapping of relat ions onto in-

var iant logical opposi t ions. Levi-Strauss sees history mu ch as Hegel descr ibed

it: the dialectic realizes itself harmoniously in the movement of logical op-

pos ites. Bu t Ma rx finds irreversible dialectical movem ent-discontinuous, un-

e v en d ev el op m en t an d qu alita tiv e ~ h a n g e . ~While Levi-Strauss (34, p. 566) , then , descr ibes structure metaphorically a s a

stable cylinder e xten ding itself infinitely throug h time bu t an alytically isolable

from i t , Marx f inds histor ical s tructures constant ly al ter ing themselves. For

Levi-Strauss al l contradict ions are of the same quali ty; a l l are opposi t ions

reducible in the last analysis to the universal dimensions of the or der of orders .

For Marx, contradict ions vary in qual i ty with their mater ial development in

history, both in antagonism and in significance (in Althusser 's terms some are

overde termined) . Th ere can therefore be no universa l s t ruc tura l opposi tions .

S Y N C H R O N I C A N A L Y S I S OF S YS TEMSIn both Levi-Straussian structural ismand structural- functional ism, analysis of a system requires that a meth-

odological dist inct ion be made between synchronic and diachronic modes of

analysis . N o claim is ma de that system s exist outside of t ime, only that knowing

such systems requires their synchronic descr ipt ion. I f a l l re lat ions are of the

same o rder and if the dimensio ns of the sy stem are specif ied, then th e emergent

pro pertie s of a syst em ca n alwa ys be analyzed synchronically-either as a

rearr ; i rsgement in the elements of the system or as a process of quanti ta t ive

adjustmcrrt (15).

I t Is Iirswever, precisely the assumptions underlying the condit ions of syn-

chronic nna%ysistiaar are challenged by the notion of material contradiction in

Marxist analysis. The gcrresis of slrucaure s indeed-as Lev i-Strau ss (34, p .

560 tsld Piaget in sbnrctrare, bat titilt structure is i tself diachron ic and con-

sianfly evolving. 'Fhe link between the assnmpt ions of s tn~c tura l i s t nd s t ruc-

fural..futactior?ailstanthropaiogy and the telertlslglcal iclealitim of i ts nineteenth

century evoXtrtionist precursors i s exposed by Mno:

I he inetaphysicat or vulgar rvolu~lunistworid outlook sees things as isaiated,

static and one-sided. It regat ds ail things n the universe, their forms and their

species as ctemally i~r7Jstedfrom one ;rr:attier anitl irnnrtatahlc. Such change asthere is can onEy be at? incrra ;c or d c c ~ e ~ ~ s en cgirantity or change of glxc .

86areovcr the cnusc rif cuch an incrcsct or dzcz se or change of place is riot

inside things ut o6rt:-iifs: them, that is their nno force s external B ib , p. 25).

*AltRusser s t2) an;ct hiiztri;cz iV.; 41 s o n ~ r a v b a i ~ d i v e r ~ ~ e t ~ tn lyses o the relatinsra e

tween Marx N I I ~Hrgsl .~ko~a ce cnns~aitedbere.



M RXIST PPRO CHES IN NTHROPOLOGY 345

Thus structuralist readings o f Marx (52) hat accept a methodological opposition

between structural and temporal modes o f analysis and relegate dynamics to a

lower level o f abstraction are false. Structural regularities are always processual

and should be conceptualized as such in understanding any particular concrete

historical situation. do not mean that synchronic description o f a system may

not be a necessary step in analysis. Marx, for instance, heuristically traces out a

synchronic model o f simple reproduction o f capital in order to clarify the

essential determinations o fhis dynamic model o f qualitative change in expandedreproduction.

The latter theory o f expanded reproduction is not, however, less abstract than

the model o f simple reproduction. Rather it is conceptually more complex. Assuch it is the only adequate model o f capitalist development, since the assump-

tions o f the theory o f expanded reproduction (unlike those of simple re-

production) correspond to the material conditions o f he historical reality that it

purports to understand.

Failure to understand this essential aspect o f Marx's method o f analysis in

api tal has led to persistent misinterpretation o fhis formulation of he law of he

falling tendency o f he rate of profit. This law should not be read as an historical

description or prediction, but as a dynamic analysis o f the significant systemic

variables underlying the course o f capitalist development. Similarly, mode o fproduction is a dynamic concept for Marx, not a static type. This digression

into the logic of ~ p t r lshows that the methodological differences between

historical materialism, structuralism, and teleological evolutionism as modes o f

explanation appeareven before one specifically addresses the analysis o f human

society.

In discussing the last two modes o f anthropological analysis, have empha-

sized the idealist elements in Levi-Strauss's work. Ye t structuralist an-

thropologists have consistently refused to accept the charges o f Hegelianismaddressed against their work. Strictly speaking, they are correct. Levi-Strauss's

preoccupation with the logical structures o f the mind does not stem from any

transcendent attachment to Spirit, but from his analysis o f he essential determi-

nant o f human material existence-Man's triumph over Nature through the

symbolic representations o f Culture.

The fact that structuralism looks Hegelian in practice should not disguise the

similarity between Levi-Strauss's basic proposition and discussions o f human

nature by such anti-Hegelian anthropologists as Geertz (22)and White (57),who

emphasize the importance of symbolization in human evolution. The Marxistposition here cannot be clarified at an epistemological level, for the central

question is not the nature o f knowing but the nature o f human society. The

validity o f the assumptions of dialectical materialism is established through the

scientific analysis o f human history.

TH E B S IC CO N C EITS O F H IS TORIC L M TERI L IS M

In this section, will discuss certain basic concepts o fhistorical materialism and



pological viewp oints. T he place to begin is the dialectical relations hip betwe en

people and nature th at Marx saw as fundamental in the process of huma n socialevolution.

Marx began his sea rch for an understanding of the laws of evolution of hum an

society by emphasizing both the unity and opposition between people and

nature . In his often quote d com parison of the worst of architects and the best of

be es, M arx (39, p . 178) recognized the imp ortanc e of huma n imagination. Yet

language and representa tion are themselves derived from peoples' social re-

produ ction of the conditions of their ow n existenc e: Men ca n be distinguished

from animals by consciousness, by religion or anything else you like. They

themselves begin to distinguish them selves from animals a s soon as the y begin to

produ ce their means of subsiste nce (41, p. 42). In production people opposethemselves to na ture by acting on the extern al world and changing it; but at the

sam e time they are dialectically one with na ture, for in changing it they change

themselves as we11 (39, p. 177).The central proposition of historical materialism is, therefore, that social

production and reproduction are the basis of human history. In order to draw ou t

the theoretical meaning of this proposition, I have isolated its two basic prem-

ises: first , that production is a social process; second, that production and

reproduction determine the dynamic structure of human society. These two

premises will be ela borated and c ontras ted with anthropological views of society

in the tw o following sec tions.

roduction is a Social rocess

Historical materialism is concerned with the evolution of people as a social

specie s, not as individual organ ism s. Thu s although preoc cupa tion with Man

the Tool-Maker might seem to be an understandable consequence of Marx 's

emphasis on hum an production of the m eans of subsistence, search for a chim-

panz ee w ho will break straw s to fish ou t termites is , in term s of huma n evolution ,

a meaningless exercise. It is not the intentionality of production that defines

hum an activity, but rath er its necessarily social chara cter. T o be hu ma n is to be

social, for we ca n reproduce ourselves only through coo perative production ofou r means of subsistence . Robinson Cru soe alone on his island was never theless

a prod uct of a particular s ociety (and preoc cupation with his plight the ideolog-

ical product of another).

Since people exist only in society, individuals are never autonomous units.

Quite to the co ntra ry, people ca n individuate themselves only in society, and

each individual is determined by a particular set of social relations. Society

cann ot be un derstoo d as a population o r aggregate of individuals, but only as a

totality of social relations (43, p. 100). It is ther efo re always meth odologica lly

unsound to assume that an ethnic group o r pol it ical unit ( the N uer , United States

society) is an adeq uate unit of analysis. One mu st begin by reco nstructing thesocial relations that d eterm ine particular subjec ts o r group s.

Since people can individuate themselves only in society, there is no natural

opposition betw een individual and society. ,Th e problem of o rde r should not,




therefore , be t he central quest ion of social theo ry , nor should equi libr ium b e

ass um ed as the reprodu ction cond ition of all societies . Structural-functionalistcon cern (of both the Parsonian and Br i t ish vers ions) w ith the maintenance of

systemic compatib i l i ty presupposes inherent a tomist ic tendencies in a l l so-

cieties . As ideology, t he i l lusory individualism of capitalist soc iety is confirmed

through universalization in structural-functional writing (5, 35).

Ma rx , by c on t ras t , exposed the c lass r ela tions tha t under lie the a ppearan ces

of indiv idual f reedom in capita l is t so ciety and showed that societ ies nee d not be

harmonious equi librium syst em s. Precisely bec aus e people a re social ly in ter-

dependent in product ion , societ ies can reproduce themselve s cont inual ly de-

sp i te confl ic t and con t radic t ion . S inc e one canno t assum e tha t an y mov eme ntou t of equi libr ium annihi la tes the sys te m, explanat ions of social facts that res t o nthe m aintenance o f funct ional in tegrat ion provide n o explanat ion a t al l .

In more specif ically anthropological terms, i t is only legitimate to root the

exchange of wo me n o r goo ds in the maintenanc e of social so lidari ty if on e has

f irst show n why the exchanging groups are antagonist ic . Defining exchan ge a s

the basis of social order is therefore e i ther a t bes t an absurd reduct ion of

conceptual vocabulary (all social relations are by definit ion relations of ex-

change) and at wors t a naive acc epta nce of a universal egoist ic psy che .

I ha ve em phasized the implicat ions of M arx s d iscuss ion of the th e social

nature of product ion because of the f requent misrepresentat ion of Marxis t

theory in the d ichotomy between consensus and s t ra in theor ies of society .

His torical m ateria lism is not a s t ra in theo ry; i t doe s not in an y way a ssum e that

conflict , hierarchy and stratif ication are universal in all forms of society. Ex-

ploitation and class conflict are historically specif ic, not general social

p h en o men a .

Uto pian idealization of preindustrial societies was often bo th a n ideological

supp or t for the sup pose d evolut ionary inevitabil i ty of c lasses in technological lycomp lex societ ies and a n ar t i fact of par ticu lar colonial pol ic ies (6 , 55). Thu s

anthropologis ts tended t o over look class dev elopmen t and class conf lict in the

societ ies the y s tudied . Marxis t an thropologis ts hav e t r ied t o redress th is b ias ,

but n ot by assuming that c lass re la tions a re universal.

M arx s ins is tence o n the social def in i t ion of the indiv idual has impor tant

con seq uen ces for theor ies of cul ture as well a s for theor ies of society . Since

there a re no au tonm ous ind iv idua l sub jec ts, there is n o such th ing as the bas ic

natural man, s t r ipped of the accidental acc outre me nts of cul ture and h is tory .

Malinowski s basic self-aggrandizing m an , the universal m axim izer of neo-c lass ica l economic theory , Ardrey s aggresso r , and McC lel land s universal

achievem ent motivat ion are therefore a ll i l lus ions , again m yst if icat ions a ppro -

priate in exp an siv e period s of capitalist socie ty (35) ( though less appe aling whe n

you can not real ly get wha t you wa nt) . People a re ma ter ia l , sent ient beings , of

cou rse, and as su ch are af fected by their own physio logy, but these af fects can

only hav e par t icu lar h is tor ical and social express ions: there is no general psy-

chophysio logical con tent in hum an exis tence .

Confirmation of M arx s insistence o n the im portan ce of flexible social adapta -



348 O LAUGHLIN

t ion in the evolut ion of hum an exis tence see ms to m e to be the s t rongest s ide of

conte mp orary pr imate s tudies ( the i r worst is the search for the universa l ape) .The implicat ions of such s tudies for theor ies of cul ture a re m ost c lear ly dra w n in

the anthropological l i terature by G eert z (22): w e canno t have a theory of cul ture

in general ; we can only in terpre t par t icular cultures . Suc h a s tanc e is m eth-

odological ly possib le only if we ha ve a general theory throug h which t o analyze

the way s tha t cul tura l representa tions are soc ia lly organized and reproduced.

H ere G eer tz and Marx radica lly diverge . Though Ge er tz re jects the not ion of

a universal subject of any inna te or cons tan t nature, he sti ll maintains th at there

ar e universal existential situa tions --co mm on to all societies-that permit inter-

subjec t ive ent ry in to systems of meaning. Such phenomenologica l app roach estha t begin with the existential si tuat ion of the s ubject clash with M arxist meth-

ods o f cul tura l in t e rpre ta t ion . T o show w hy th is is so , w e mu s t re tu rn t o the

re la t ionship be tw een consc iousness and soc ia l exis tence through the Marxis t

con cept of ideology.

Since cul tural representat ions exist only when they are social ly organized,

consc iousness i s a lways subsumed in exis tence , never auton om ous f rom it : I t

is not the consciousness of man which determines his existence, but his exis-

tence which de termines his consc iousness 40a , p . 21). People can and d o hold

all s orts of conflict ing rep resentat ion withou t changing the con dit ions of their

social existence. Cultural representat ion s a re no t logical system s straining to-

ward cons is tency , fo r they ar e ult imately orde red by the i r soc ial contexts , n ot by

the logic of mind.

Yet prec ise ly because exis tence i s not de termined by consc iousness-

though t is part of social real ity but do es not exh aus t it-forms of conscio usnes s

no t only repr ese nt bu t may also systematically m isrepresen t the social relat ions

through which th ey a re forme d. Such misrepresenta t ions const i tu te ideology,

forms of con sc iousness tha t a r e qui te real but non ethe less fa lse . Ideology has nohistory of i ts ow n nor universal co nte nt ; it can only be funct ionally defined in

relat ion to a part icular historical s0ciety. j

Th ere are cer tainly forms of consc iousness tha t a r e not ideologica l . Cul ture

and ideology ar e not the sa me in Marxis t te rms. N ever the less , if consc iousness

does not fa i thful ly represent the soc ia l condi t ions of exis tence , then we can

nev er fully unders tand these condi t ions through the representa t ions of subjec-

t ive consc iousness . T h e in tersubjec tive ent ry of phenomenologica l app roach es

to cul tural interpretat ion is therefore i l lusory, for the subject 's view (whether

consc ious o r unconsciou s) is qui te l ikely to be a myst if icat ion of underlyingsocial relations.

If w e misund erstand the social relations throug h which cul ture is organized,

then w e misinterp ret cul ture as well , for meaning is in its referents. O ne should

therefore begin cul tural interpretat ion w ith an historically formed social s ys tem

and not w ith subjects. Peop le ul timately define them selves thro ugh their social

re la tions , not through th e symbols they use

SAlthusser (3) provides a controversial and theoretically advanced discussion of the



MARXIST APPROACHES IN ANTHROPOLOGY 49

Production and Reproduction etermine the ynamicStructure of Society

Marx s conception o f a social system is that o f a dynamic totality composed o f

relations be tw ee n people and be twe en people and nature. T h e relations o f this

social totality are o f di ffe re nt qua lities; since production and reproduction o f

hum an subsistence constitute the basis o f so ciety , the determinant aspects or

mo me nts are the technical forces o fproduc tion and social relations o f produc-

t ion. Th e productive system , w ith corresponding forms o f consump tion, d is tr i-

butio n, and exchange ( th e ba se ), ult imately determines th e form o f uridical-

political and ideological relations (superstructure).

Marx never claimed that history merely exp resses productive relations; suchec on om ism is antithetical t o his understanding o f the dialectical relationship

bet we en base and superstructure. In the last instance, ho w ev er , the importan ce

o f particular institutions is determined b y the mo de o f production:

the middle ages could not live on Catho licism, nor the ancient world onpolitics. On the contrary, it is the mode in which they gained a livelihood thatexplains why here politics, and there Catholicism, played the chief part. Forthe rest it requires but a slight acquaintance with the history o f the Roman re-

public, for exam ple, to be aware that its secret historys

the history of landedproperty. On the other hand, Don Quixote long ago paid the penalty forwrongly .imagining that knight errantry was compatible with all economic formso f society ( 3 9 , p . 82) .

In positing the determinance o f the ba se, Marx was no t, the refo re,reducing all

social relations t o relations o fproduction; religion , politics, and chivalry are not

primarily econ om ic institutions. Qu ite to the con trary, he wished t o show that

the relations o f the social system were o f d i f fere nt qual it ies , wi th those o f the

base ultimately determining the structure o f the w hole.

T o understand the determinance o f the ba se, it is necessary to comprehend

the importance Marx assigns t o the conc ept o f reprod uction. In social produc-

tion people not on ly produce but also reproduce the conditions o f their ow n

ex iste nc e. Since all production is produ ction w ithin a particular fo rm o fsociety ,

this means reproduction o f labor, reproduction o f the m eans o f production, and

reproduction o f the relations o f production ( 3 ) .

In the dialectical w orking ou t o f he relationship be tw een forces and relations

o f production in time and space, antagonistic contradictions ma y deve lop within

the system-think, for insta nce , o f conflict ov er access to land arising in certainconditions as a swidden s ystem beco me s in t ime a system o f intensive cul-

tivation ( 20 ) .O r, as in cap italism , antago nistic contradictions m ay be contained

within the productive sy ste m from its outset in the essential class contradiction

between labor and capital.

T h e system reproduces i tse l f ,despite these contradictions, through the medi-

ation o f superstructure-juridico-political and ideological relations tha t sup res s,

displace, or misrepresent basic c on flicts. Th es e relations m ay be th em selv es

con trad ictory , and certainly th ey d o not strain toward any necessa ry functional



35 O L A U G H L I N

they mere1 y permit their reprodu ct ion o ften in mo re antagonist ic form s. I t is thus

the conc ept of reproduc t ion tha t es tabl i shes the necessa ry l inks be tween baseand superstructure within the social system.

T he deve lopment of antagonis t ic cont radic tions be tween the forces and re -

la t ions of product ion has historical ly led to cris is and to eventual qual i ta t ive

changes in systems of product ion:

The moment of arrival of such a crisis is disclosed by the depth and breadth at-

tained by the contradictions and antagonisms between the distribution relations,

and thus the specific historical forms of their corresponding production re-

lations, on the one hand, and of the productive forces, the productive powers

and the development of their agencies on the other hand. A conflict then ensues

between the material development of production and its social forms (40, p.883 .

The immediate historical cause of the cris is need not be economic, but i ts

funct ional importan ce is a lways determined by the cond it ions of the base and i ts

resolut ion is a lways economic-a basic change in th e sys tem of produ ct ion.

Th e basic mo vem ent of hum an history is , therefore , th e dialectical develop-

ment of the forces and re lat ions of produc t ion . Marx d id not s ee th is mo vem ent

as an even, progress ive , and harmonious deve lopm ent of the d iv ision of labor ,

but a s unev en, pe r iodized , qua li ta t ive chang e , marked by revolut ionary t rans i-

t ion f rom o ne epoc h of produc t ion t o another . Th e new s ys tem of produc t ion

emerges historical ly from the old, but not as the synthet ic resolut ion of i ts

contradict ion s. Th e dia lectic between f orc es and re la tions of product ion in the

new m ode of produ ct ion differs in its terms from tha t of the preceding epo ch of

produc t ion . T o unders tand h is tory i s the re fore to be able t o define these h is tori -

cally specific terms.

METHODS OF ANALYSIS: MODES OF PRODUCTION

In the preceding sect ion, som e of the basic co nce pts of his torical materia l ism

were discussed and contrasted with prevalent anthropological views. In this

sect ion, I will co nsider in c loser deta i l the metho ds of his torical materia lism, a ndpart icularly the construct of a mode of product ion. In general terms most

Marx ist socia l sc ientis ts agree tha t socie t ies sh ould be analyzed a s socia l for-

mations-re la t ional systems comp osed of superstruc ture and a determin ant

econom ic base which may i tse lf be a com plex art icula t ion of mor e than a s ingle

mode of product ion.

From this conception of a socia l formation fol low certa in basic meth-

odological principles . Firs t , th at in analyzing an y inst itut ion, suc h as kinship or

the family o r the s t a te , on e should break i t do wn to the re la tions that define i t and

l ink these sys temica l ly to th e base .6 Second, the ana lys is of the soc ia l format ion

as a w hole i s pre requis i te to unders tanding; the curren t f ragmenta t ion of thesocial sciences into isolated disciplines is antiscientific.

8Berthoud s (9) critique of Fortes s analysis of kinship includes a good methodological

discussion of this point for anthropologists.



M RXIST PPRO CHES IN NTHROPOLOGY 35 1

Beyond this general accord, however, there is an awesome diversity of

opinion on the implementation of the methodological framework. Beforepresenting a tentative outline o f he method o fanalysis o fa mode of production,

I will discuss some crucial conceptual issues and try to specify errors that have

been made. Most o f the writers discussed consider themselves Marxists, but for

the purpose of clarification I will also refer to the work o f anthropologists who

have been significantly influenced by Marx or whose work has been confused

with Marxist approaches (Goody, Worsley. Harris, Sahlins). The issues ad-

dressed include: 1 the problem o f general vs specific categories o f analysis;2 the dialectical conceptualization o f orces and relations o f production; 3. the

articulation o f modes o f production; 4 concepts o f transition.

Conceptual Issues

G E N E R L vs SPECIFIC C TEGORIES Social production o f the means of sub-

sistence is the basis o f human existence. Thus all epochs o f production have

certain common elements; labor and its means o f production-the object and

instruments of labor (43,p. 85). f we analyze a particular arrangement o f these

traits as a technical process, we describe the forces o f production. I f we analyze

this same arrangement of traits in terms o f relations o f appropriation betweenpersons, we describe the relations o f production. In each instance, the relations

analyzed are both social and material, but we see dif ferent aspects o f social

reality.Viewing production as a labor process, we could thus define a mode o f

production through a rather mechanistic specification o f each o f the common

traits of all production. Here is the rub: production is not a general process, but

the appropriation o f nature on the part o f an individual within and through a

specific form of society (43, p. 87). All systems o f production may haveinvariant elements, but these provide only a tautological framework of analysis

which says nothing of historically specific social forms o f production: T o theextent that the labour-process is solely a process between man and Nature, its

simple elements remain common to all social forms o f development. But each

specific historical form o f this process further develops its material foundations

and social forms (40,p 883). Since the purpose o f theory is to develop those

abstractions through which the concrete (always historically specific) can be

understood, a set o f universal concepts cannot define any particular mode o f

production. Analysis o f a mode o f production must be movement from abstractgeneral determinations to observation and conception at the level o f the con-

crete and then back t o the theoretical articulation o f general and specific

categories.

Distinguishing general and specific categories is necessarily a central problem

in scientific analysis, a problem which cannot be resolved entirely within theory

itself . Marx recognized that it is a common task o f ideology to represent

particular social processes as the expression o fconstant universal laws-natural

social charters. Thus a considerable portion o f Marxist scholarship on non-



352 O'LAUGHLIN

that M arx developed for the analysis of capitalism can o r cannot be applied to all

modes of production.Th e only scientific solution to this problem is the movem ent back fro m theory

to data in order to see how well the concrete can be determined by the ab-

stractions one has developed. This movement is particularly important foranthropologists, whose discipline has so often assuaged contem pt for capitalist

society with the conse rvative as suran ce of cultural universals throu gh which we

see ourselves othe r than as we are. Here I find it particularly central for

Marxist anthropologists to so rt out: (a) the analytical status of value theo ry, andb ) that of the con stru ct of precapitalist m odes of prod uction .

Value th ory As an abs traction from production, the concep t of labor value isapplicable to the analysis of an y mod e of prod uction . But the con cep t of value is

often theoretically articulated with o ther concepts in analytical constructs that

are specific to particular modes of production. In a recent a rticle, for instance,

Jonathan Friedman (21, p. 446) suggests that the social relations of productiondet erm ine the ra te of surplus (slv) and the ra te of profit [sl(c +v)] in all forms of

soc iety, whethe r or not that society has categories corresponding t o s , (surplus-

value), v (variable capital), and c (constant capital). While I agree that theapplication of theoretical categories does not depend on their presence in the

minds of informants, Friedman's extension of concepts M arx developed for the

analysis of capitalism to all modes of production is a fundamental meth-odological error.

In the capitalist mode of prod uction , surplus is app ropriate d as surplu s-valu e

by capital from labor in the process of production. T he value produced by labor

is greate r than the value of those c omm odities that m ake up the wag e-bundle of

worke rs (V) and the value of the means of production used up in production (C):

this surplu s-valu e (S) is app ropriate d by capital. Th e appropria tion of surplus as

surplus-value d epends on particular historical co nditions: the ex istence of wage-labor as a com mo dity; the separation of workers a s a class from their means ofprod uction ; a com plex division of labor reflected in widespread use of machines

that hold only in the capitalist mode of production. Applying the categories C ,V , and S to o ther modes of production can tell you nothing about their specificdynamics.

A second issue in the use of value-c onc epts is the analytical sta tus of the law ofvalue. Unlike the classical econ om ists, M arx neve r argued that the law of value

held (i.e. that articles be ex chan ged in relation to their labor-va1ues)in all mod es

of prod uction . Only when production is organized for exch ang e rath er than use-in petty-c om mo dity prod uction and most fully in capitalist production--do

comm odities e xcha nge in relation to their labor-values (through their prices ofproduction):

Prices are old; exchange also; but the increasing determination of the former bycosts of production, as well as the increasing dominance of the latter over all

relations of production, only develop fully and continue to develop even more

com pletely in bourgeois so cie ty, the society of free competition 43, p. 156).



MAR XIS T AP P R OAC HES IN A NTHR OP OLOGY 3-53

Com modities excha nge in relation t o their labor-values only wh en the law of

value regulates the ir production.Persistent misreading of Marx's analysis of the law of value stems from the

assum ption of the primacy of exchange in bourgeois ideology. This assum ption

is similarly reflected in the explanatory power assigned to the concept ofspheres of value in econo mic anthropology [s ee Barth (8), Boh annan (12),

and Berthoud ( lo ) , the lat ter for a n excellent crit ique of Bohannan]. T he centraltheoretical question should not be why som e goods are traded fo r some goo dsand not for othe rs, but rather why one thinks it natural th at all excha nge

against all. The latter is true only when p eople pro duc e primarily f or exc han ge

rather than for thei r own u se:When production is oriented towards immediate subsistence not every article

can be exchanged for every other one, and a specific activity can be exchanged

only for specific products. The more specialized, manifold and interdependent

the products become, the greater the necessity for a general medium of ex-

change 43, p. 199).

Th e con struc t of sph ere s of value is used t o analyze transition from traditional tomo dern economies-spheres brea k dow n and an all-purpose m oney is intro-

duce d. As Dupre Rey (18) have pointed ou t, how ever, con cepts of transition

restricted to the level of distribu tion ca n provide only tautological explan ation of

the displacem ent of precapitalist modes of production by capitalism and drawattention away from the mechanisms throu gh which basic changes in produc tiverelations are accomplished.

h e s p e c if ic i ty o f p r ec a p i t a l is t m o d e s o f p r o d u c t i o n Difficulties in estab lishingthe analytical range of particular concepts recur in the M arxist characterization

of precapitalist economies. Are there properties common to all precapitalist

ec on om ie s- e .g. natural economies, or economies dominated by personaldependence-such tha t they can be analyzed by a com m on se t of theoreticalco nce pts? O r can w e establish a su bset of precapitalist economies-primitive,

classle ss, cold , or dom inated by kinship-and in doing so estab lish a re-

stricted object of knowledge? These questions are clearly important ones for

M arxist anthropologists con cern ed w ith delimiting th eir discip line, and the re isno unanimity among Marxists in their responses to them.

Throu ghout m ost of his life, M arx was conc erned with the develop me nt of an

analysis of the capitalist mode of production that would provide a theoretical

fram ewo rk for the revolutionary class-struggles of the p roletariat. H e insistedthat this analysis had to be historically specific, but he denied that it wasnecessary to develop historical know ledge of all modes of production in order to

understand the essential relations of capitalism. Quite to the contrary:

Bourgeois society is the most developed and the most complex historic organi-

zation of production. The categories which express its relations, the com-

prehension of its structure, thereby also allow insights into the structure and therelations of production of all the vanished social formations out of whose el-

ements and ruins it built itself up 43, p. 105 .



354 O LAUGHLIN

In the ethnolog ical reading he did just before his de ath , Marx (42) tried to apply

the concepts developed in his analysis of bourgeois economy, but I have notfound it possible to interpret the excerpting , interpolations, and omm issions inhis notebooks in such a way as to yield any analysis of precapitalist modes of

production. When Marx characterizes precapitalist modes of production in his

analytical wo rks, he is almost alway s doing one of two things: a) analyzing thetransition from feudalism to ca pitalism; b establishin g the essential relations of

the capitalist system b y specifying in a dialectical man ner what they are no t.

Th e principal lesson one can learn from Marx's work o n precapitalist modes

of production therefore seem s to be that they a re the same only insofar as they

are not capitalist; there is no positive basis for distinguishing them as a the-

oretical object of knowledge. Given Marx's analysis of the historical specific-ities of capitalism, we should be able to determihich categories should not

be exten ded to precapitalist m odes of production, but we can not thereby estab-lish a set of concepts amenable to the analysis of the specific dynamics of all

precapitalist modes of production. T o say that societies are classless or char-acterized by a low level of deve lopm ent of the prod uctiv e forces tell us wh at they

are not but not what they a re. T o speak of precapitalist eco nom ies as dom inated

by relations of personal dependence tells us nothing of what these particularrelations are.

Som e work that cannot be summ arized here has been done o n the analysis of

particu lar precapitalist m odes of produ ction, particularly in Africa (44, 48, 50,56,5 8). M urra's (46) wo rk on the In ca and G odelier's (25) reanalysis of it are alsovery helpful. In ge ne ral, how ever, any analysis of precapitalist modes of prod uc-

tion mu st be rooted in the broad m ethodological framew ork of historical materi-alism. This brings us to the sec ond and cen tral issue in the analysis of a mod e of

production-the con ceptu alization of the dialectical relationship betw een forces

and relations of production.

THE DIALECTIC OF FORCES AND RELATIONS OF PRODUCTION Analysis ofany problem requires a specification of the mode of produ ction , und erstoo d a s a

dialectical unity of forces and relations of production. Technical relations be-tween people and nature always imply corresponding forms of social relations.

Forces and relations of production cannot therefore be analyzed in isolation

from each other, yet the difference between them must be conceptualized as

well. The basis of this dialectical relationship is the unity of people with natu re

and the opposition of people to nature in production. Production and re-production of human existence requires appropriation of nature by human labor;

Marx criticized those who placed human activity outside the laws of nature:

Lab or is not the sou rce of all wea lth. Natu re is just as much the so urce of use-values and t he se, certainly, form the material elem ents of wealth as labor, which

is itself only the exp ression of a natural fo rce , hum an labor-power (37, 19).This dialectical relationship betw een force s and relations of prod uction is the

key to the unev en, periodized, and nonteleological process of human ev olution.No understand ing of social chan ge can be analytically sep arated from techno -

logical chan ge, for in acting on the ex ternal world and changing it, people at the




same time change their own nature (39, p. 177). Marx s view is therefore quite

different from that of most of the cultural materialists and cultural ecologistswho insist that the ultimate meaning of history lies in technological processes.The problem here is that technological change then becomes an exogenous

independent variable: ways in which social relations of production affect the

development of the productive forces are systematically left unanalyzed.

Leacock (32) has very neatly criticized the reductionist and ultimately idealisttendencies inherent in the cultural materialists rejection of the dialectic-

between forces and relations of production and between consciousness andexistence (see also Friedman 21). Yet the same problems recur in less positiv-

istic approaches to cultural evolution. I will therefore discuss at greater lengththe revisions in evolutionist theorji introduced by Marshall Sahlins, whose

interest in French structural Marxism has done much to resuscitate interest in

Marx among anthropologists.In analyzing the development of political chiefdoms, Sahlins (51, p. 101)

argues that such political structures evolve from a contradiction between forces

and relations of production. This sounds like a Marxist argument, but Sahlinscontradiction is immanent rather than material: there is a difference between

what people could produce and what they actually produce. Political structures

therefore evolve as a means of realizing the surplus that is implicit but un-actualized in the system. In fact, the contradiction that Sahlins finds lies withinthe forces of production, that is, within the technological system itself. Unless

one ahistorically assumes the teleological rationalization of economic efficiency

as the essence of human evolution, there is no way of moving from this supposed

contradiction to the emergence of political hierarchies. In other words, there is a

distinct continuity between Sahlins s early evolutionist work and that of his

recent Marxian stage.

Within British social anthropology, Goody and Worsley have continually

challenged the structural-functionalist assumption (by Radcliffe-Brown andFortes at least) of the autonomy of kinship in primitive societies by arguing thatproductive organization determines the form of kinship relations. As with the

American cultural materialists, however, productive relations tend to reduce to

technological systems. According to Goody 28), for example, it is the absence

of the plow and concomitant aspects of intermediate technology that has large1y

determined the significance of military organization in African state develop-

ment. In a similar fashion, Worsley (59) has argued that the particular forms that

Tale kinship takes are determined by the fixed farm and the need for cooperationin agricultural labor.In the case of Goody and Worsley, it is probably a strong dose of logical

positivism (relations between persons that do not appear on the surface of things

are less real somehow than relations between persons and things), rather than a

well-articulated evolutionist position, which leads to the reduction of relationsof production to technical relations.

From a Marxist perspective, the presence or absence of the plow cannot be anindependent parameter in a productive system, for social relations of production




cooperation and forms o f the family are interdependent in societies where the

household is a basic unit o fproduction. The weaknesses o f echnological reduc-tionism are not particularly apparent in a synchronic analysis, but they aredevastatingly obvious when one tries to analyze the dynamics o f social change.

In both Goody's and Worsley's work the expansion of capitalism is viewed as an

almost automatic process o f echnological change and market involvement. The

necessity o f forcibly displacing noncapitalist relations o f production cannot be

clearly conceptualized within their theoretical framework. And so almost the

entire state apparatus o f colonial rule, including the small functional position ofthe anthropologist, is shrouded from analytical view.

On the other side o f the fractured dialectic stand a number o f Marxist an-

thropologists-preeminently P.- Ph. Rey-who generally maintain the dialec-tical unity between forces and relations of production, but who argue that in the

last instance the social relations o f production must be determinant. Ultimately

the forces that make history lie within people themselves, not outside them, butpeople exist only in dialectical relationship with nature. Therefore, social re-

lations only exist materially in correspondence with technical conditions o f

production. f a particular set o f social relations is viewed as relatively autono-mous, then the material basis for this autonomy must be specified.

Rey's (49 , 50) work on transition and articulation o f modes o f production hasopened important new directions o f analysis, but his failure to specify the basis

o f he autonomy o f social relations somewhat distorts his study o fboth capitalistand non-capitalist modes o f production. In his work (50)on a lineage mode o f

production in Central Africa, for example, he argues that a class o f elders is

able to extract surplus labor from junior men through control over the re-

production or regrouping of productive groups. The basis o f this argument is a

familiar one to Africanists (28,56):shifting cultivation is predominant in Africa,

land is not a scarce resource, land is not owned, therefore it is control overpeople rather than land which is important. The problems here are: I . that

regardless o f property relations, land continues to be the basis o f agriculturalproduction; 2. that scarcity is socially as well as technologically defined. The

autonomy o f the social process o f recruitment to productive groups is illusoryand therefore does not describe the material basis o f exploitation.

The presumed autonomy o f the social relations o f production also appears in

Rey's analysis (50)o f the transition from noncapitalist to capitalist modes of

production. Rey suggests that the determining relation of production under

feudal i sm land ren t4ont inues to play a transitional role during the period ofprimitive accumulation o fcapital. One can therefore expect, he argues, a similar

role to be played by the specific relations o fexploitation (e.g. control o fprestige-

goods) of another mode o f production of transition. This argument depends onthe assumption that capitalism is a system o f social relations that is largely

autonomous from its material referents. This analysis corresponds neither toMarx's view o f the contradictions inherent in the material expansion of capital-

ism nor to any empirical referent in the contemporary articulation of non-capitalist and advanced industrial capitalist modes o f production (13).




Thre e a l te rna t ive posit ions on t he concep tual iza t ion of forces and re la t ions of

product ion have been d iscussed: one can ins is t on the d ia lec t ica l un i ty bu t

d i f ference be tw een forces and re la t ions of p roduct ion , o r o ne ca n f rac ture the

d ia lec t ic and then ins is t on the de terminance of e i ther the forces o r re la t ions of

product ion . A fourth al ternative is logically possible: one can fracture the

d ia lec t ic conceptual ly ye t s t raddle bo th s ides a t the sa m e t ime. This i s in fac t

wha t Maur ice Gode l i er 23, 24, 27) and Jona than Fr i edman 20) have done in

suggesting that th e forces and relat ions of production s hould be conce ptualize d

as tw o separa te s t ruc tures wi th in a funct ional sy s tem . Contrad ic t ion be tween

the two structures is thought of as a kind of l imit ing condit ion, functional

incompatibil i ty between structures within a system.

Godel ier a rgues tha t the evolu t ion of the mode of product ion inc las s soc ie t ies

depe nds on the p lay of two an tagonis t ic contrad ic t ions- -one wi th in the re la t ions

of product ion an d on e be tween th e forces and re la t ions of p roduct ion 25, p .

238). The mater ia l condi t ions for the resolu tion of the c lass contrad ic t ion wi th in

the relat ions of produc tion c an only exist outside of i t since the produ ctive force s

ar e a real i ty that is dist inct f rom th e relat ions of production.

Suc h an in terpre ta tion can lead t o a mechanis t ic theory of revolu tion in which

class s t ruggle ca n be resolved only wh en the a n tagonism betwee n forces and

re la t ions has evolved to a cer ta in level . Godel ier suppresses Ma rx s analysis o f

cyclical cr isis and of the dist inctive coincidence of class and mater ial con-

tradict ion s in capital ism. T he principal thesis of Capital is precisely that in the

c lash be tween capi ta l and labor , c lass conf l ic t expresses the contrad ic t ion

between the forces and re la t ions of p roduct ion : there i s one cen t ra l con-

t rad ic t ion , no t two. This is t rue unde r s lavery as well a s incapi ta l ism 43, p . 463),

henc e the par t icu lar vu lnerab i li ty of these tw o sy stem s, bu t i t does n o t ho ld in a ll

c lass soc ie t ies .

Looking a t p recapita li s t soc ie t ies , Godel ier 24, pp . 364-65; 26) argues tha t

the cen t ra l p roblem of primi t ive soc iet ies i s control l ing ac cess to w om en and

equil ibrat ing their circulat ion. T he im portanc e of this problem derives from th e

centra l ro le of k insh ip s t ruc tures in these soc ie t ies ; k insh ip is a t the sam e t imeboth in f ras t ruc ture and supers t ruc ture 23, pp . 94-95). Th e econom is t can eas ily

d is tinguish product ive forc es in these soc ie t ies ; bu t c anno t i so la te au tonom ous

relat ions of produc tion. R ath er the mult ifunctionali ty of kinship str uct ure s act s

as a l imi t to the developm ent of the product ive forc es and expla ins the general ly

s low rhythm of the i r development . Godel ier s method of analysis therefore

focus es a t ten t ion on the ex ten t to which k insh ip re la tions con st ra in the inde-

pendent ly de termined technological sys te m 23, p . 290).

This mode of analysis i s car r ied to i t s ex t rem e in Fr iedm an s 20, 21) re-

wo rking of Polit ical Syst em s of Highland B urm a. Fr ied ma n very convincingly

shows tha t the mater ia l is t concept ion of i r revers ib le s t ruc tura l evolu tion m oreexactly describes the distr ibution of g u m s a g u m l a o and S han sys tems than

The problem s in Godelier s conception of forces and relations of production as twodistinct systems are discussed by Skve 53); Seve is in turn answered by Godelier 25).




does Leach's Paretian notion of structural oscillation. In his alternative analy-

sis, however, Friedman reverts to a kind of technological determinism quitereminiscent of cultural evolutionism.

Friedman locates the central contradiction of Kachin society in the clash

between the system of exchange of valuables (women and bridewealth) and the

potential productivity of an expansive technological system. Friedman is

thus directly applying Godelier's interpretation of capitalist contradiction to the

Kachin, despite very different levels of development of the productive forces in

the two cases. But Marx (so also Godelier, in principle) is always concerned to

show that the real developement of the forces of production is contingent-

inextricably bound up with the relations of production. Friedman, by contrast,

takes the production function as given, and simply assumes: (a) that populationwill grow, and (b) that the system will expand up to a certain limit with popu-

lation growth. These are assumptions that Durkheim might have permitted, but

never Marx. Thus for both Friedman and Godelier conceptual dissection of the

forces and relations of production has unfortunate analytical consequences.

A R T I C U L A T I O N O MODES OF P R O D U C T I O N I N A SOCIAL FORMATION A par-

ticular conceptualization of a mode of production is analytically useful only as

long as it describes the essential forces and relations of production of the

economic base in a particular form of society. The base, furthermore, is not self-

reproducing; it only can be realized within a social totality. In that sense everymode of production describes not only a base but corresponding forms of

superstructure.

In early Marxist-structuralist work (e.g. 56) there was a tendency to assume

that the concept of a social formation simply described a combination of the

elements of a mode of production, including both base and superstructure. Only

if there is more than a single mode of production will the concept of a social

formation have any analytical meaning in this line of reasoning. Thus Terray (56,

pp. 161-62), in analyzing Guro ideology, tries (unsuccessfully) to identify cer-

tain cults with one mode of production and other rituals with another. There is a

confusion here of concepts and concrete reality. All phenomenal forms have

multiple determinations. If I wanted to interpret the meaning of a Christian

parable told by an African pastor, for example, I would look for allusion to both

capitalist and precapitalist relations of production.

Yet Terray's error is also related to the order of analysis he prescribes-the

analysis of the social formation follows the analysis of modes of production:

In the first place, the various modes of production realized in these formations

must be listed, using as a guide a census of the forms of cooperation in use.

This is what I have tried to do for the Guro. The next step should be to con-

struct the theory of the modes of production identified; each socioeconomic for-

mation would then appear to be composed of such and such modes of

production combined in such fashion that one or other of them is dominant.

Like a chemical m olecule, a socioecono mic formation would then be defined by

its struc ture , by the nature of its component, elem ents as well as by the way

they are organized within the whole 56 p. 179).




This molecular order o f analysis strays. unhappily think, from Marx's method

of investigation.It is sometimes assumed that Marx's analysis of he capitalist mode ofproduc-

tion was to be followed by complementary works on superstructure, since

Capita l contains so few references on the state and class struggle. But the

analysis o f the mode o f production presented by Marx in Capital presupposeshis analysis ofbourgeois society as a whole, an historical analysis which exposesthe essential role o f he state and class conflict in the development o fcapitalism.Thus in analyzing contemporary societies, it is a methodological error to at-

tempt, as Terray does, to isolate the analysis o f a precapitalist mode of produc-

tion from the analysis o f the social formation as a whole. Distinguishing thedeterminations o f capitalist and precapitalist modes o f production is part o f he

process o f identifying a mode o f production analytically, not methodologically

subsequent to it.

THEORIES O F T R N S I T I O N The faulty order o f analysis established by struc-

turalist molecular theories o f social formations has posed severe methodological

problems for the analysis of transition. The concept o f a mode of production is

used both to defineconceptually a particular dialectical (and therefore dynamic)

unity o f forces and relations o fproduction and, at adifferent evel of abstraction,to delimit a period o f history dominated by a particular mode of production. At

this second level we may choose to distinguish certain periods or stages o f

development o f the mode o f production.

When the analysis o f the mode o f production precedes the analysis o f the

social formation, as it does in the molecular theories, then periodization and

transition must be analyzed at the first level o f abstraction as well. Otherwiseone falls back onto a structuralist set o f invariant elements that cannot describe

any historical specificities at all. Thus some structural Marxists (16)search for a

general theory o f transition, applicable to the analysis o f all social formations.They have suggested, for example, that historical transformationscan be under-

stood in terms o f a displacement o f the dominant instance (l6, p. 73).Such general theories, however, are either tautological or misplaced efforts

to make concepts do one's analytical work. Historical processes do not arisefrom the machinations o f a model; rather we use models t o understand histor-

ical processes. There can be no general theory o f transition precisely because

ea ch historical transi t ion is dif fere nt, materially, and therefore concep-tually (7,p. 69). W e do not have to explain historical development, for that is

constant; what we do have to explain are its structural regularities. These can be

understood only by consistently relating the mode o f production to the socialformation in the process o f the analysis.

Alternative Formulations

There is then no general method o f analyzing all modes o fproduction in histori-

cal materialism. W e begin with the m ~ s fgenemJ demenls ofp ~o du cl jo n ndbegin to formulate the concepts that will allow us to describe an historically



own goal-the tracing out o f the relationship between forces and relations o f

production as a dialectic whose boundaries are to be determined and whichdoes not suspend the real difference (43, p. 109).

W e know that forces and relations of production can be conceptually dis-

tinguished, but in analysis there must be a constant dialectical movement be-

tween them. Similarly the conceptualization o f a mode o f production does not

permit a narrow focus on productive process; eventually the entire social totalitymust be encompassed. Since we are analyzing a mode o f social reproduction

rather than a simple labor process, patterns o f consumption, distribution, and

exchange within the base, as well as the role o f superstructure, must always beconsidered in any analysis o f a mode o f production.

There are a number o f conceptual difficulties and methodological problems(demonstrated in the preceding section) that follow from these very generaldirectives as soon as one tries to implement them in analysis. I have alreadydiscussed these questions in a critical way; here I will try to present a more

constructive alternative formulation. There are four texts that seem to me

particularly helpful: Marx's (40a, 43) 1859 Introduction, Balibar's essay on

the basic concepts o f historical materialism in Reading Capital (4), with his

response (7) to Cutler, and Terray's (56) Marxism and Primitive Societies

In presenting the constructive framework, I will consider separately the

conceptualization o f forces and relations of production, then discuss the di-

alectical unity of the two. and finally move to the analysis o f social formations.

FOR ES OF PRODU TION T Oanalyze the forces o f production in a particular

mode o f production, we begin by looking at relations between people and

relations between people and their means o f production in the productive

process. This means describing the various productive units, the tasks per-formed, tools used, the demographic and ecological contexts, etc. Marx as-

sumes that In all states o f society the labour-time that it costs to produce the

means o f subsistence must necessarily be an object o f nterest to mankind (39,

p 71); thus the sequencing and duration o f various productive activities should

also be noted.In describing the forces o f production, it is important for a number o f reasons

not to remain at the level o f immediate production. First, since production is a

social process, there is always some interdependence o f producers which may

be manifest only at the levels o f consumption and distribution. Focusing nar-rowly on productive process, one has a tendency to see such a multiplicity of

modes o f production that it becomes extraordinarily difficult to relate the dy-

namics o f the productive base to the rest o f society. This is more or less whathappened in early attempts to apply the concept o f a mode o f production to

precapitalist societies in Africa (44, 56, 58); hunting became identified with one

mode o f production, agriculture with another. Looking only at immediate pro-duction also makes it difficult to distinguish exploitation from productive co-

operation once one moves to the analysis o f the relations of production. This

problem is particularly crucial in understanding the basis o f sexual asymmetry indifferent forms o f society.



The second reason for not limiting analysis to immediate production is that

human social production requires reproduction--of labor and of the means ofproduction. Tools must be replaced, seed kept for a new planting, land renewed

through fallowing or fertilization, children borne and nourished, new units ofproduction formed. What appears to be surplus from the point of view of

immediate production may in fact be necessary for social reproduction. At the

same time, Rey s (49, pp. 35 ff.) criticism of Terray (56) has shown that a narrowfocus on immediate production may also lead one to overlook the fact that

control ofreproduction may be a mechanism for the extraction of surplus labor.

As the above discussion indicates, the analytical specification of the forces of

production cannot be separated from the analysis of the relations of production.I have emphasized this methodological principle because I have found in my

own work that the absence of landed property in Africa is often taken to be

derived exclusively from technological parameters; changes in land tenure arethen related too narrowly to technological change. Such approaches do not dealsatisfactorily with the importance of control over land in the evolution of

precapitalist modes of production nor with the articulation of these modes ofproduction with capitalism.

In discussing the method of analyzing forces of production, I have not been as

precise as I would like to be. One of the reasons for this vagueness is that there isa paucity of conceptual language for describing the technical relations of non-capitalist modes of production; such concepts are underdeveloped in both the

Marxist and the conventional anthropological literature. We have a language

that tells us a bit about what precapitalist societies are not (minimal division of

labor by sex and age, production for use, simple technology, small-scale econ-omy) but that language is distinctly unhelpful in analyzing what they are. Forms

of kin-groups that have productive functions in noncapitalist societies for in-

stance, are quite variable; it should be important to describe and analyze them.The poverty of concepts for analyzing the forces of production in precapitalistsocieties is of course related to the exclusion of technology from the economy

and the corresponding focus on distribution and allocation in economic anthro-

pology. Excellent criticism of the distributive focus of both the formalists andthe substantivists are developed by Godelier (25), Dupre Rey (18), and

Meillassoux (45), and I expect the work of Marxist anthropologists onproduc-

tion to redress soon this problem of conceptual lag. Meanwhile one can hold onto Marx s (39, p. 180) own advice: in describing the forces of production it is not

a list of articles that is important but the ways in which the articles are made.

R E L T I O N S O PRODUCTION In the fundamental human social process-

appropriation of nature in production-Marx saw a natural unity of labor and its

means of production, the latter consisting both of nature and of its instruments of

production that are themselves embodied labor. This position contrasts withthat of Hegel, for whom all forms of objectification of labor constitute alienation

of the human essence. Marx nevertheless argued that the possibility of true

alienation (as opposed to objectification) is inherent in the development of the



36 O ' L A U G H L I N

from its ow n pr od uc t, an d ultimately-in capitalist production-from itself. In

looking at socia l product ion from the point of view of the social ra ther thantechnical s ide of th e division of labor, w e ar e therefore looking at re la t ions of

appropr ia t ion be tween persons tha t a re based o n the re la t ion of the workers t o

the i r produc t and means of produc t ion .

M ost of Ma rx's analysis of re la tions of produ ct ion deal t with c lass socie t ies .

H e showe d tha t the dynam ics of c lass 'were rooted in the appropr ia t ion of

surplus-labor-as eithe r living o r em bod ied labor-by a class of non -pro duc ers:

The essential difference between the various economic forms of society, betw een, forinstance, a society based on slave- labour, and one based on wage-labour, lies only inthe mode in which this surplus-labour is in each case extracted from the actual

producer, the labourer (39, p. 217 .

Following the wo rk of Bal ibar (4), a num ber of Marxist anthropolog ists , such as

Te r ra y 56 and Rey (49), p icked up o n th i s c ita t ion f rom Marx and applied i t to

their analysis of precapi ta l is t socie t ies . They argued that Engels ' tendency to

define c lass in term s of a property re la t ion-ownership of the means of produ c-

tion-is a disto rtion of the essential dete rm inati on of class , i .e. th e exploitation

of the prod ucer by t he no nproducer . Rey spec if ical ly t r ied to sho w tha t in the

l ineage mod e of product ion in Africa surplus-labor is extracted by e lders

f rom d e pe nde n t jun io r me n .

Re y's emp hasis o n th e category of surplus-labor h as been analyt ical ly frui t ful

for anthropologis t s s ince it cente r s a t tent ion on the d ia lect ica l de te rmina t ion of

surplu s in a Marxist fra me wo rk. W hereas cul tural material is ts and the c lassical

pol i t ical economists insis t that surplus is defined uniquely by technological

param ete rs (produc t iv i ty and the minimum subs is tence requi rements necessa ry

for the biological reproduct ion of labor), Marx, in his analysis of capi ta l ism,

argued that the subsistence bundle of workers is socia l ly, not biological ly,

de te rmined. Both i t s composi t ion and i t s s ize vary wi th the use -va lues tha t

workers dem and, and can the re fore expre ss the outco me of workers ' s t ruggle

for a la rger share of the i r own surplus pr oduc t .

This dia lectical determ inat ion of surplus by t he force s and re la tions of produ c-

t ion holds in precapi ta l is t as well as capi ta lis t modes of product ion, but whe n no

accumulat ion takes place, when nei ther labor ( in s lavery) nor labor-power (in

capita li sm) has becom e a comm odi ty , and w hen wo rkers a r e not separated f rom

their m eans of prod uct ion, surplus is som ew hat difficult to define. I am not fu lly

convinced, for example, that exploi ta t ion best describes the re la t ionship that

Re y analyzes betw een elde rs and junior me n in African l ineages. T he analyt ical

difficul ties here a re re la ted to tho se M arx d iscusses in his analysis of comm unal

rec rui tment of labor for road m aintenance .

This is certainly surplus labour which the individual must perform, whether in the

form of forced labour or in the indirect form of taxes, o ver and above the direct labournecessary for his subsistence. But to the extent that it is necessary for the communeand for each individual as its mem ber what he performs is not surplus labo ur, but apart of his necessary labour, the labour necessary for him to reproduce himself as



commune member and hence to reproduce the community which is itself a general

condition of his productive activity 43, p. 526 .

A number of important methodological directives can be inferred from this

passage and other work of Marx's on precapitalist formations (42). First, the

absence of social surplus, and consequently of social relations which can prop-

erly be called exploitative, does not mean that there are no conflicts and con-

tradictions within the prevailing division of labor-between sexes, between old

and young. between communal head and members. We should in fact expect tofind such conflicts and see them mediated in juridical-political relations and in

ideological representations. Nor should we expect that such contradictions will

never develop into antagonistic class conflicts, for the latter always come fromsomewhere:

It must be kept in mind that the new forces of production and relations of production

do not develop out of nothing, nor drop from the Sky, nor from the womb of the Self-

positing Idea; but from within and in antithesis to the existing development of

production and the inherited traditional relations of property 43, p. 278 .

Another methodological inference that one can draw from Marx's precapitalist

work is that in defining surplus one cannot assume that the individual is an

independent unit of production. Human production is social; surplus must bedefined in relation to a particular social division of labor, not the activities of an

individual subject. Third, surplus must be defined in relation to the necessary

reproduction of the means of production. These latter two inferences are of

course exactly those which were discussed in relation to the definition of the

forces of production, and they have brought us full-circle back to Balibar,

Terray, and Rey's definition of the relations of production.

Recent reconstructions of Engel's work on evolution are based both on a

reading of Marx and a clear look around the empirical world: there are quite

evident forms of exploitation that do not depend on ownership of the means of

production-taxation, pillage, perhaps absorption of surplus-value in exchanges

between systems dominated by different modes of production (43, p. 729). The

BalibariTerrayIRey reformulation leaves us in a logical impasse, however: the

relations of production are defined by the extraction of surplus; the existence of

surplus is dialectically defined by the forces and relations of production.

The solution to this problem is to return to Marx's narrow definition of the

relations of production in terms of relations between persons based on the

alienation of workers from their means of production and/or their product. Theproducts of labor are use-values, either objects of a specific form or living-labor

( services ) appropriated in production. Surplus-labor may be extracted from

people who are not alienated from their means of production, but if so it must bedone by noneconomic means, such as the use of political force. Only when labor

is separated from its means of production are class conflicts subsumed within the

productive base.

The essential difference between the various forms of class-society then does

depend on the mode of extraction of surplus-labor; but the mode of extraction of



364 O ' L A U G H L I N

surplus-labor is determined by the mode of production-the dialectical re-

lationship betw een technical relations and relations of approp riation betweenpersons. This reformulation of Balibar's position see ms to me true to M arx's

con stant conc ern with the relationship of people to their mea ns of production a swell as to their lab or, a con cern illustrated in the following discussion of surplus-

labor.

Th e specific econom ic for m, in which unpaid surplus-la bour is pum ped out of direct

pro du cer s, determ ines the relationship of rulers and ru led , as it grows directly ou t of

production itself and, in turn, reacts upon it as a determining element. Upon this,

however, is founded the entire formation of the econom ic community which grows upou t of the pro duction r elations them selves, thereby simultaneously its specific politi-

cal for m. I t is always the relationship of the owners of the conditions of production tothe direct producers-+ relation always naturally corresp ondin g to a definite stage inthe deve lopm ent of the m ethod s of labour and thereby its social productivity-which

reveals the innerm ost secr et, the hidden basis of the entire social structure and with itthe political form of the relation of sovereignty an d depen dence ; in short, the cor re-

sponding specific forms of the state (40, p. 791).

It should be evident from the preceding discussion that if social relations of

production are defined as relations of appropriation between persons, then it

should be possible to specify relations of produ ction in classless as well as classsocieties. In fact, deciding w hether or not a particular society is a class society

depen ds on a prior specification of both forces and relations of production. T heargument by Deluz Godelier 1 7)that anthropology studies classless societiesin which the social relations of production merge with kinship relations is

therefore m ethodologically unsoun d. T hey a re clearly corr ect in suggesting that

the conc ept of a m ode of extraction of surplus-labor and th e catego ry of a

nonp roducer are not analytically helpful in classless societies w here e veryon e

wo rks, but not in deducing that such societies hav e no relations of produc tion

analyza ble apart from kinship str uctu res. There is thus no analytical basis forisolating primitive societies as a distinct obje ct of know ledge with distinct

methods of analysis.

THE DIALECTICAL U N I T Y O FORCES A N D RELATIONS O PRODUCTION In

outlining methods of specifying forces an d relations of production w ithin a m ode

of production, I have argued that defining a mo de of prod uctio n is necessarily a n

analytical process in which one must move dialectically between forces and

relations of production. There are no all-encompassing typological schemes

through which all possible m odes of produ ction can be described. Th ese ana -lytical difficulties are furthe r com poun ded by tw o mo re conce ptual choice s: a )

it may be n ecessary to describe the articulation of more than a single mode ofprodu ction within the base; (b) one m ay wish to speak of dominant and acces-

sory relations of production.

The art iculat ion of mod es o fproduc t ion Marx (43, pp. 106-7) suggested that inany particular form of society there could be more than one bran ch of produc -




tion, but that in this case one o f these branches would be dominant. This aspect

of he analysis of modes ofproduction was elaborated in the work o fAlthusserBalibar 4), who analyzed all social formations as complex articulations of more

than a single mode of production. The analytical status o f this claim was never

made very clear, but it was used by Terray and others 44, 5 6, 58) to describe

dual modes o f production in precapitalist formations in Africa.

The notion that theremust be more than a single mode o fproduction is based,

think, on a confusion o f concepts and concrete reality. In so far as all concrete

reality is dialectical, then yes, there are always at least two modes o fproduction

-the being and the becoming. But analytically the dialectical concept o fa mode

ofproduction is intended precisely to describe adynamic evolving system. Onlysystems with distinctly different technical relations and relations o f appropri-

ation should be analyzed as separate modes o f production.

The base need not, therefore,contain more than a single mode ofproduction,

but it may do so. Such is the case, for example, when peasants become de-

pendent on the production o f export-crops and the purchase o f manufactured

tools, even though techniques o f production do not resemble those of agrarian

capitalism. In the present historical context capitalism is the dominant mode oi

production on a world scale.

Since modes o f production are historically specific, there can be no general

theory o f articulation o f all modes of production. Looking specifically at the

articulation ofcapitalism with precapitalist modes o fproduction, we can see that

linkages within the sphere o f circulation ofcapital and political domination tend

to precede the development of wage-labor relations. The exact nature of these

linkages will depend, however, both on the historical stage o f capitalist develop-

ment and on the character of the precapitalist mode o f production.

Dominant arzd accesson relatiorzs of production Deciding whether or not thebase of a particular social formation is best described as an articulation of two

modes o f production is an analytical choice: it depends on what it helps you to

understand. An alternative to the formulation o fplural modes o fproduction may

in some cases be provided by the concepts of dominant and accessory relations

o f production, suggested by Pollet Winter 48) in their analysis o f Soninke

slavery. They argued that it would be analytically sterile to describe Soninke

slaves either as a class or as simply incorporated into family groups; rather the

slave relation was an accessory relation in a mode of production defined by the

low level ofproductive forces and access to land for all members o f he society.do not want to deal with the adequacy of Pollet Winter s analysis o f this

particular case, but only to suggest that the concept o f dominantiaccessory

relations may be analytically useful, particularly in the analysis of the place o f

the household in advanced capitalist society.

In a capitalist system o f production, the relations of production are defined in

terms of the appropriation of surplus-value from labor by capital in the process

of production. Forms o f labor which do not correspond to the wage-labor

relation cannot therefore be analytically assimilated to capitalist relations o f



366 O ' LA U G H LI N

product ion. But ne i ther ca n those forms of labor , such as tha t of wom en in the

househ old, tha t prod uce necessary use-va lues be re legated to supers t ruc ture .O ne ana lyt ica l a l te rna t ive i s to descr ibe the household as a non-capita li s t mode

of production, ar t iculated with capital ism.There a re c lear ana lyt ica l advantages to th is formula t ion (30) . For an-

th ropo log is t s , a cohe ren t ana lys is o f wo m en a s a c l a s s f r om whom surp lus l abor

is exploi ted would fac il i ta te a c lean break f ro m the vulgar econom ism of the

Bro wn (14) app roac h tha t a t temp ts to re la te ahis torica lly the s ta tus of wom en to

the impor tance of the i r product ive cont r ibut ions .

Never the less , I suspec t tha t such a formula t ion has unfor tuna te ana lyt ica l

consequences . F i r s t , i t c louds the c lass dynamics of capi ta l i s t soc ie ty: Thi rd

Wor ld worke rs an d wom en m oving ou t o f t he househo ld do no t have the sam elong- te rm posi t ion wi thin the labor - force and should not have the same ana-

lytica l s ta tus ; capi ta l , not m en, app ropr ia tes the surplus- labor of wom en in the

househ old; co nt radic t ions wi thin a p ar t icular sexual d ivis ion of labor need not

have th e s am e an tagonism as c l a s s confl ic t s. Secon d , i t g losse s ov e r t he ro l e t he

househo ld plays in mediat ing the cont radic t ion be tween use-va lue and exchange

value t i thin the capital ist mode of production. Third, i t is consistent with

( though need no t imply) theoret ical constructs-such as the publicipr ivate di-

chotomy-that universal ize exp ression s of wo me n's sub ordin at ion in capital ist

soc ie ty as ex plana tory con cep ts in a ll soc ie t ies .

Thinking of the household as a subordina te mode of product ion does nottherefore seem to be the most ana lyt ica l ly useful way of unders tanding the

re la t ion be tw een ho usehold labor and capi ta l. Th e concept of dom inant iacces-

sory re la tions of product ion cou ld perhaps be an a l te rna t ive . There i s , in fac t , an

ana log fo r such a cons t ruc t i n M y x ' s no t ion o f r em nant f o rm s of p roduc t ion :

. since bourgeois socie ty is i tself only a contrad ictory form of develo pm ent,

relat ions der ive d from ear l ier for m s wil l of ten be fo und w ithin i t only in an

entirely stunte d for m , o r ev en travest ied (43, pp. 105-6).

The order o analysis I hav e discussed the problem of ana lyzing the househ old

in capital ist society a t length. fi rst becau se i t is an im portan t issue for M arxistanthropologis t s , and ' secondly to make a methodologica l point . The re -

construct ion in theory of the relat ions between modes of production within the

base is a p rocess of ana ly t ica l c h o ic e4 e c id in g wha t cons t ruc t s be s t he lp us t o

und erstan d a par t icular co nc rete histor ical s ituation-not an empir icist naming

and typing of c on cre te re la t ions .

S o f o n e s t u d ie s , a s I did, a rura l v il lage in Tchad , then w hat o ne sees i s not the

precapital ist s ect or . but the mater ial expressio n of tw o dyn amic systems-

capi ta l is t and precapita li s t m odes of product ion. T he re is therefore n o inner nor

outer sys tem; radica l anthropology should not consis t of showing how the

exogenous s t ruc tures of the wider capi ta l is t sys tem impinge on i sola tedt radi t iona l com mun i t ies or marginal g rou ps , but ra ther of loca ting these group s

and communit ies within that s tructure i tself .

Moreo ver . in ana lyzing the s t ruc ture of such s ys tems i t is impor tant not to

isola te one ' s ana lysis of the precapi tal i st mode of product ion f rom the dy namics




of capi ta l is t development , les t one ascr ibe to t radi t ion that which is in fact

determ ined by th e ar ticulation of capitalist and precapitalist mo des of produ c-tion. This m ethodological f law, I th ink , is what under l ies Ter ray ' s inadequ ate

discuss ion of the Dioula t rade in h is reconstruct ion of G ur o mode s of produc-

t ion , and R ey 's underes t imat ion of the ef fects of mercan t i le capi ta l o n the

evolut ion of African societ ies .

T H E SOCI L FORM TION The concept of a mode of product ion descr ibes a

s ta tement of sys temic te nde ncy , the d ialectical working out of the re la t ionship

between forces and relat ions of product ion over t ime. At a more h is tor ical ly

precise level of abstract ion , the m ode (o r mode s) of product ion m ust be analyzeda s the base of a social form ation th at includes juridical-polit ical and ideological

relat ions a s well . The elem ents of th is sys tem m ust a lways be conceptual ized a s

relat ions , not as ins t i tu t ions such a s k inship or the s ta te . If we wish to under-

s tand rac ism in con tem porary Am er ican soc ie ty , fo r example . we d o not imme-

diately assign it to the level of ideology, but instead define the relations-

eco nom ic, political an d ideological-that det erm ine it.

Th e preceding m ethodological d iscuss ion of analyzing mo des of product ion

has hopefully a lready ma de clea r that one ca nno t con struc t a social formation in

thought by analyzing modes of production, f inding their contradictions, andmechanical ly assembling the whole. In fact , the mode of product ion can only be

analyzed in re la t ion t o the social formation a s a whole; supers tructural re lat ions

allude to the base and often specify i ts essential contradictions. Since all re-

la tions are both d ialect ical and mater ia l , supers tructural re la t ions may have their

own qual ity and m ovem ent (autonom y) and can not be reduced to re la t ions in the

base. In the las t ins tance, howev er . the re la t ions of the base determ ine the form

of the who le (1 , pp . 200 ff.).

Methodological ly the determinance of the mode of product ion with in the

social formatio n pose s so m e analytical diff iculties: one m ust avoid both re duc-

t ionis t economic determ inism and arb i t rary ass ignment of preeminen t ro les to

supe rs truc ture . I t has therefore bee n suggested that the determinanc e in the

las t ins tance of the base be d is tinguished f rom the domina nce of super-

str uc tur e in a particular historical c on tex t (e.g. 56 p. 147). find M ar x's

d is t inct ion between the order of analys is and the order of presentat ion or

exposi t ion more helpful : in analyzing any problem one must begin with a

specif icat ion of the mo de of product ion; in present ing one 's analys is , howe ver ,

eco nom ic factors need n ot be par t icu larly im por tant , i fan analyt ical groundingfo r the au tonom y f rom the base o f the phenom ena d iscussed is p rov ided .

Kno wing is necessarily organizing an d specifying the essential de term inatio ns

and suppress ing the nonessent ia l . Social formation is a theoret ical construct

designed to help us know. I t should therefore be obvious that the wa y in which

we analyze the social formation depen ds not only on the nature of the real wor ld

but also on the specif ic problem we are po5ing.

The contemporary wor ld can be unders tood as a s ingle social formation ,

dominated by capita l ism in the base , yet determined by o th er mod es of produc-



368 O'LAUGHLIN

however specify all o f these interrelationships nor would we know anything i f

we did. What is important is finding the essential determinants o f he phenomenawe wish to understand. The constructs mode o f production and social formationprovide no automatic discovery procedures; they do specify the direction ofanalysis and help us to organize what we know.

C ONC LUSION TH E ANTHR OPOLOGIST IN ADVANCED

CAPITALIST SOCIETY

In the development o f his paper I have argued that from a Marxist perspective

there can be no autonomous discipline o fanthropology. W e cannot construct ananthropological mirror in which t o find the reflection o f he basic universal Manfor all individuals are determined by particular historical social relations. Nor

can we find theoretical unity by limitingourselves to the study o f precapitalist orprimitive societies for their similarity lies in what they are not rather than in

what they are.

Ye t there is still a further reason for the necessary unification o f anthropology

with history and the other social sciences. I f we are to understand others thenwe must understand ourselves in our social world-as researchers and teachers

producing both science and ideology in the context o f advanced capitalist

society. W e cannot exclude the capitalist mode o f production from our ana-lytical universe; all societies that we find to study must be understood as an

articulation o f capitalist and noncapitalist modes o f production; all knowledge

that we produce must cut through the misrepresentations of our own ideological

mystification.

Many anthropologists today are very critical of the functional connections

between their discipline and repressive state-apparatuses during the colonial

period 29 , 54). What we must come to terms with however. are both theambiguities o f the position of the colonial anthropologists 31 and a politically

relevant analysis o f our own roles as imperialist facilitators and ideological

mystifiers.When Marx wrote o f deology he suggested that we find mythical charters for

contradictory social practice in historical precedent and natural laws. The

quests for origins and for biological universals have both had their place in the

development o f anthropological theory. But we have also elaborated another

ideological genre-explaining historical social conflict in terms o f social univer-sals. Class becomes only a manifestation o f the stratificatory tendency in all

social systems; racism in American society is just a subset o f ethnic boundaryproblems.

Critical theory the unmasking o f ideological mystifications. is therefore an

essential aspect o f a Marxist anthropology and it is being done extraordinarilywell by people such as Tala1 Asad 5,6 ) . Ye t as Marx insisted in the Ger~nc l n

I d eo l o g y , critical theory is not enough to change the world. The sense of aMarxist approach to anthropology emerges only when historical materialism s



l i n k e d t o t h e r e v o l u t io n a r y a i m s o f M a r x i sm - L e n i n i s m . S c ie n ti f ic k n o w l e d g e o f

o u r w o r l d s h o u l d i n fo r m t h o s e p o l i ti c al s t r u g g l e s t h a t t r a n s f o r m i t .

S c i e n c e i s a l w a y s a s k i n g s o m e q u e s ti o n s a n d n o t o t h e r s . I n M a r x i s t te r m s

t h e s e q u e s t i o n s s h o ul d n o t b e d e r i v e d s i m p l y f r o m t h e p l a y o f t h e o r y , b u t r a t h e r

f r o m t h e a p p l i c a t io n o f t h e o r y t o c o n c r e t e a n d c o n s t a n t l y c h a n g i n g h i s to r ic a l

s i t u a t i o n s . T h e c u r r e n t f o c i o f M a r x i s t w o r k i n a n t h r o p o l o g y - i m p e r i a l i s m ,

i d e o l o g y , r a c i s m a n d e t h n i c it y , w o m e n a n d t h e h o u se h o ld - a re a l l a r e a s w h e r e

t h e o r e ti c a l c l ar i fi c at io n i s n e e d e d t o u n d e r s t a n d c o n t e m p o r a r y p o l it ic a l p r o b -

l e m s . o s o m e e x t e n t t h e n M a r x i s t a n t h r o p o l o g y m u s t b e a p p l i ed a n t h r o p o l o g y ,

t h e u n i v e r si t y a n d c l a s s r o o m a l o c u s o f p o l it ic a l s t r u g g l e , a n d p r a x i s a n e s s e n t ia l

a s p e c t o f v e r i f ic a t io n .

Literature Cited

1. Althusser, L. 1970. For Marx. NewYork: Vintage Books

2. Althusser, L. 1970. Su r le Rapp ortd e M a r x a Hege l . In Hegel et laPensbe moderne, ed . J Hyppolite,8 5-1 1 2. P a r i s : P r e s s e s u n i v . d eFrance

3. Al th usser , L . 1971. Ideology andi d e o l o g i c a l s t a t e a p p a r a t u s e s . I nLenin and Philosophy, 127-86.

New York: Monthlv Rev. Press4. Althusser, L. Balibar, E. 1970.Read-ing Capital. Transl. from French byB. Brewster . London: NLB

5. Asa d, T . 1972. Market model, classs t r u c t u r e a n d c o n s e n t : a r e c o n -s i d e r a t i o n o f S w a t p o l i t i c a l o r -ganization. Man 7: 74-94

6. Asad, T. 1973. Two European im-ages of non-European rule. In An-thropology and the Colotiinl Eti-counter, e d . T . A s a d , 1 03 -2 0.London: Ithaca

7. Balib ar, E. 1973. Self criticis m: An

answ er to questions from TheoreticalPractice. Theor. Pracr. 71857-71

8. Barth , F . 1967. Econom ic spheresin Darfur. In Themes in EconomicAnthropology, ed. R. Firth, 149-74.London: Tavistock

9. B erthou d, G . 1969-70. An throp o-logie socia le e t parente . Quelquesreflexions critiques sur Kinship andthe Social Order d e F o r t e s . Arch.Suisses Anthropol. Gen. 34:94-100

10. Ib id . La Val idi te des concepts de"mul t icen t ri c it e " e t d e " sphe resd 'echange" en anthropologie eco-

nomique, 35-6412. Bohannan. P. 1959. The impact of

money on an Afr ican subs is tenceeconomy. J Econ. Hist. 19:491-503

13 . Bo n te . P . 1973. Qu e lqu es p rob-

lemes theor iques de l a reche rchemarxiste en anthropologie. La Peti-s8e 171:86-104B r o w n , J K . 1 9 7 0 . E c o n o m i co r g a n i z a t i o n a n d t h e p o s i t i o n o fwomen among the Iroquois. Ethtio-history 17:15-167C a n c i a n , F . M . 1 96 0. F u n c t i o n ala n a l y s i s o f c h a n g e . An?. Sociol.Rev. 25:818-27

Cutler, A. 1973. "A Letter to Et-ienne Balibar on the Problematic ofH i s t o r i c a l M a t e r i a l i s m a n d R e -sponse ." Theor. Pract. 718:51-55,73-85Deluz , A . , Gode l i e r , M. 1967 . Apropos de deux textes d 'anthropolo-gie economique. I'Hotntne 7:7&91Dupre , G. , Rey, P. -Ph. 1973. Re-flections on the pertinence of a the-o r y o f t h e h i s t o r y o f e x c h a n g e .Ecot~ . oc. 2:131-63Engels, F. 1939. Herr Eugeti Duhr-ing's Revolution in Science, e d .

C. P. Du tt , transl . E. Burns. NewYork: Int . Publ.Fr iedman, J . 1972. System, Strue-ture and contradiction in the evo-lution of Asiati c soc ia l fo rma-tions. PhD thesis. Columbia Univ.,New York, NYFriedman. J . 1974. Marxism, struc-tu ra l i sm and vu lga r ma te r i a l i sm.Man 9 : 4 4 4 6 9Geertz , C. 1973. Interpretation ofCultures. New York: Basic BooksGodelier. M. 1972. Rarionality andirrational it in Economics. Transl.

B . P e a r c e . N e w Y o r k i L o n d o n :Monthly Rev. PressGo del ier . M. 1972. Str uc t ure andcontradiction in capital. In Ideologyin Social Scietlce, e d . R . B l a c k -



37 O ' L A U G H L I N

b u rn , 334-68. N e w Y o r k : V i nt ag e

Books25. Godelier , M. 1973. Horizon, trajets 43.marxistes en anthropologie. Paris: Maspero

26. G o de li e r , M . 1973 . M o de s d e 44.p r od u ct io n , r a pp o rt s d e p a re n te e t s t ru c t ur e s d e m o g ra p h iq u e s . LaPenske 172:7-33

27. G o d e li er . M . 19 74 . O n t he d ef in i- 45.t ion of a social fo rmat ion . CritiqueAnthropol. 1:63-73

28. Goody, J . 1971. Technology, Tradi- 46.tion and the State in Africa. O x-fordUniv. Press

29. G oueh. K . 1968. N ew ~ r o ~ o s a l s 47.ora n t h ~ o p o l o g i s t s . ~ u r r . ~k t h r o p o l .9:403-7H a r r i s o n , J . 1 97 3. T h e p o l i ti c a l econom y of housework . Bull. Conf.Socialist Econ. Winter:35-52James, W. 1973.The an thropologist a s r e lu c t an t i m pe r ia l i st . I n An-thropology and the Colonial En-counter, ed . T . A s a d , 41-70. L o n -don: I thacaL e a c o c k , E . B . 1 972 . " I nt ro d u c- tio n" t o F . E n ge ls . Origin of theFamily Private Property and the

State, 7-67 . New York: In t . Publ .Len in , V . I . 1972. Materialism andEmpirio-Criticism. Peking : Fo re ign LanguagesPressL e v l - S tr a u s s , C . 1 97 1. L'HommeNu. Paris:PlonM a c p h e rs o n , C . B . 1 965 . H o b b e s a nd b ou rg eo is m a n. I n HobbesStudies, e d . K . C . B r o w n, 1 69 -8 4.Oxford:BlackwellM a o T s e - T u n g 1 96 6. Four Essayson Philosophy. P e k in g : F o r ei g nLanguagesPressM a r x , K . 1 96 3. The Gotha Pro-

gram. NewYork Labo r News Marx. K. 1964.Economic and Phil-osophic Manuscripts of 1844 e d . D. J . S t ru ik . NewYork: In t . Publ . M a r x , K . 1 96 7. Capital, V o l . 1 .NewYork: In t . Publ . M a r x , K . 1 96 7. Capital, V o l . 3 . NewYork: In t . Publ . M a r x , K . 1970. A Contribution tothe Critique of Political Economy,ed . M. Dobb, t ransl. S . W. Ryazan-skaya, NewYork: In t . Publ . M a r x, K . 1970. German Ideology,P ar t 1 , e d . C . J . A r th ur . N e w

York: Int . Publ . M a r x , K . 1 972 . The EthnologicalNotebooks of Karl Marx, e d . L .

K r a d er . E s s e n, T h e N e t h er la n d s:

Van GorcumMarx , K. 1973. Grundrisse. Transl.M . N ic o l a us . H a rm o n ds w o rt h, England: PenguinBooksM e i ll a ss o ux , C . 1 96 7. R e c h e r c h e d 'u n n iv ea u d e d e t er m in a ti o n d a n s l a s o c i e te c y n e g e ti q u e . L'HommeSoc. 6:95-106Mei ll as soux . C . 1972. F rom reo ro -duc tion to p roduc t ion . Econ. Soc.1:93-105M u r r a , J . 1956. The economic or-ganization of the Inca state. PhDtheas. Unlv. Chicago, Chicago, I l l .

N i co la u s. M . 1973. F o r e w o rd t othe~rundr isse .SeeRef.43,7-66.P o l l e t , E. W in t e r , G . 1968 .L ' or g a ni s at io n s o c ia l e d u t ra v a il a g ri co le d e s S o n in k e ( D y a h u nu , M a l i ) . Cah. Etud. Afr . 3 2 ( 8 -4 1 :

509-34R e v . P . -P h . 1 97 1. Colonialisme,nkd-colonialisme et transition aucapitalisme: Exetnple de la Comilogdu Congo-Brazzaville. P a r i s : MasperoRey . P .-Ph. 1973. Les alliances declasses. Sur l 'articulation des

modes de production. Suivi du tna-terialisme historique et luttes declasses. Paris: MasperoSahlins,M. D. 1972.Stone Age Eco-nomics. Chicago: Aldine-AthertonSebag , L . 1964. Marxisme et struc-turalisme. Paris: Payot Seve , L . 1970. Reponse a MauriceGodelier.La Penske 149:29-50S tauder , J 1971. The functions offunctionalism. P r es e nt e d t o t he symposium Marx 11: Anthropology a n d I m p e ri a li sm . A m . A n t hr o po l. Assoc. Ann. Meet . , New York

S u r e t - C a n a l e , J . 1 97 3. L e w i s H . Morgane t I'Anthropologie mode me.La PensCe 171:75-85T e rr ay , E . 1972. Marxism and"Primitive" Societies. T r a ns l . M . K lo p p e r . N e w Y or k I L o n d o n : MonthlyRev. PressW h i te , L . 1 94 9. The Science ofCulture. N ew Y o r k : F a r r a r , S t rauss CudahyWillame,J. C. 1971 . Recherches sur l e s m o de s d e p ro du c t io n c y ne -getique et l ignager. L'Homme Soc.19:lOl-19

Worsley , P . 1956.The k insh ip system of theTal lensi : A reevaluat ion .J. Roy. Anthropol. Inst . 86:37-75

Documents

Marxist Anthropology 1975