THOlvlAS GREGOR

thal of"ifutisu,"the tefln used by the tribes in the

Xingu system (sce Basso 1973: 12-1'+,.
4. Foliage from this pblltis burnedand theas his Icaehed \Vith water
lo produce potassium ehloride, the sall ofthe Xinguanos As the
villagcrsdescribe it, the Carib tribes' speeialii'ati ol1S in , hell belts
and necklaces should also be consldered labor intensive
monopolies, sineetheyinvoll'elonganddangeroustripstofind the
shell s, and arduous\Vork to craft them into Ihe final produet.
5. AII oftheXingutribeseatfi sh,monkey,andanumberofspeciesof
birds, most game animals.
6. The\'i1lagersarefar moreaeute1yattunedtoodorsthan\Ve are, and
will oftenspitto reducethesensationofabadsmell. In Jeeountsof
interaetionwith both wild Indiansand for es tan imals(with whom
wild Indians are believed to share a great deal in eommon) the
Xinguanos often remark on thefetid smell.
7. Some of tlIc Xinguanos claim to have seen a tribe eall ed "Thc
PeopleWhoLive Inside Trees" somedi stanee totheeastoftheir
reservati on. The tree people purportedly live in tree trunks, eat
uneooked birds , \Vear no clothes, and spea kin hootsand grunts.
Symbolieally, this eulturelesstribefulfill sanani mal-likepotential
whieh other wild Indians onl yapproximat e.
8. In 1972, 1witnessed thekillingofan Xingu Indianwho\Vas haeked
to death \Vith machetesas we tri ed to hold off hi sJssailants in a
room at the Indian Post. .\ftertheexeeuti on, thekillersexhibitcd
an iey calm. Oneapproaehed me, pointed to thecorpse and thc
wa i)ing relatives and said: "You can take photographs now."
9. Thcse findings are based primaril y on Iny work \Vith Tupi and
Arawak spea king informants.
124
6
Raiding, trading and tribal
autonomy in insular Southeast Asia
THOMAS GIBSON
In this paper, warfare in insularSoutheastAsia is examinedthroughthe
compari sonofthreegroups ofhighlandshiftingcultivators .I Thethesis
is that thecurrent social evaluation ofviolence and aggression within
eaeh group is the result of differing historical experi ences wi thin a
loosely integratedregional political economydominated bytheinstitu-
ti ons of slave raiding and coerced trade. Those groups positively
disposed toward bellicosity are those which pl ayed a predominantly
preda toryrole in theregion,while those which are negati velydi sposed
toward violence in anyform were primaril yprc\. There is, however,
nothingdetermini sti cabout the argument , for eachconcreteexample
rcpresents onl )' oneofanumberoflogicallypossibleresponsestoagiven
sequencc ofhistorical events. All three groups still retaill asignifi cant
degreeofautonomyovertheirinternalpoliticalandideol ogical systems,
and each must be seen as creatively responding to a changing se t of
e>..iernal politi cal and economic forces.
In thefi rstpartofthe papcr, abriefoutlineisprovided ofthegeneral
hi storical context in which these three soci eti es havedeveloped. [nthe
secondpart,anovcrviewofeaehsocich is gi ven. III thethirdpart, their
respecti veattitudes toward\'iolcncc, social rankingandindebtednessare
125
THO M A S GIBSON
compared and contrasted. The paper concludes with a look at so me of
the implications of asking questions about human violence in the
contexts of exploitation and of ritual attitudes, as opposed to those of
aggression and competition for scarce natural resources.
SOUTHEAST ASIA AS A LOOSELY
I NT EGRATED R E GION
Southeast Asia is a region in which the sea serves as the major means of
communication and the land forms the major impediment to it. It is
often easier to sail long distances from one island to another, than to
walk across the smal!est island (Reid 1984: 151). Dense rain forest and
rugged mountains provide a natural barrier behind which tribal popula-
bons have been able to preserve a remarkable degree of autonomy into
the modern era. This autonomy should not mislead us into assuming,
on the one hand, a complete lack of contact between tribesmen and sta te
systems. On the other, extensive trade also should not lead us into
treating tribal areas and sta tes as belonging to a single, tightly integrated
system. I t is precisely the importance of trade in the history of the region,
associated with a pattern of predatory raiding, whi ch has al!owed the
complex articulation of fundamental!y diverse cultural and political
systems. Violen ce and commodities are truly inter-cultural
phenomena , which can have very different meanings for neighboring
populations, and at the same time exert decisive influences on the
development of each trading and raiding partner.
Scott (1986) gives an illuminating account ofthe political economy of
the coastal Philippines at the time of Spanish contacto His account mal'
serve as a model of one of the simpler forms of maritime economies in
the region. The Philippines remained peripheral to th e major interna-
tional trade routes until relatively late. Even then it was primarily
Chinese ships which traded into the islands, and not Philippine ships
which sailed to China. As far as domestic Philippine trade is concerned,
however:
Every community traded with other communities, and most of them did it by
boat.. the total impression is one of continual Illovements of rice, camotcs,
bananas, coconuts, \Vine, fish , g:lllle, salt and cloth between coastal barrios - to
say nothi ng of gold , jewelry, porcclain and slaves . . [Raiding] \Vas the
eseemed occupation of the able-bodied maJe who could afford it . .. Although
raiders took booty both ashare and afl oat when they eould, then real objcct was
126
Raiding, trading and tribal autonomy
slavcs . It is to cal! attention to .. the mixed merchandizing of this]
commerce [that it] is here referred to as "trade-raiding." Potential customers for
this trade in the Philippines were legion because the purchase of slaves was an
ordnary means of investing surpJus wealth ... Filipino commllnities supplied
most of their own slave labor by usury and penal aetion, but always preferred
aliens for religious purposes - that is , sacrifice. (Scott 1982: 87-9 1)
Trade in the western parts of the achipelago had long been more
sophisticated. The Sri Vijayan state dominated regional trade between
the seventh century and 1025, when it was destroyed by the Chola
dynasty ofSouth India. Its rulers had been "content to provide a neutral
commercial facility for the exchange of Western, Chinese, and
Southeast Asian products" (Hall 1985: 210). In later times, local rulers
of coastal sta tes began to supplement their position as mere port
custodians, servicing the ships of Chinese or Indian traders, byattempt-
ing to secure local products from their hinterlands for the international
market. Prior to the fourteenth century these products were in large part
gathered in the forests of the interior (items sueh as camphor, sandal-
wood and benzoin), or from the sea (items such as pearls) and had to be
acquired from autonomous groups of shifting cultivators and hunter-
gathercrs (see Reid 1980: 236). After the fal! ofSri Vijaya there fol!owed
a two hundred year period in which Arab, Indian and Chinese traders
attemptcd to trade directly with the producers of the spices and forest
products for which there was a growing demand both in south China
and in Europe. However, according to Hall:
By the 13th century Asia 's internal trade was baek in the hands of Southeast
Asians, as foreign merchants found it once again expedient to deal with
Southeast Asia-based intermediaries at major international entrepots rather
than attempti ng to deal directl y with the people who controlled the so urce s of
supply. (Hall 1985: 24)
One aspect ofthis process insufficiently emphasized by Hall is the fact
that human labor power itselfbecame a "local product" to be acquired
from the hinterland for sale in the entrepots along the coasts.
Before indentured labor was developed in the nincteenth centllry, the move-
ment of captive peoples and slaves was the primary source of labor mobility in
Southeast Asia. Typicall y it too k the for m of transferring people from weak,
poli ti cally fragmellted societ cs to stronger and weal thier (mes. The oldest, and
demographicall y most important movement was the border raiding against
animi st swidden-cultivators and hunter-gathers by the stronger wet-rice culti-
vators of the river valleys ... There secms little doubt that the ma jor i.ty of the
Southeast Asi an urban population prior to about 1820 was rccruited in a captive
127
THOMAS G IBS ON
state . Slavery was an important means whereby animistic peoples \Vere
absorbed into the dominant Islam of the city and coast. (Reid 1983: 28-29.
170- 171 )
Warren (1981) has amply documented the explosive growth of the
Sulu state during the eighteenth and nineteenth century, a growth that
was based upon a system of extracting forest and sea produce, by means
of slave labor, for trade with Cantono The labor was itself"produced" by
raiding the poorl y defended populations of the Bisayan islands and
Luzan. Again, Endicott (1983 ) provides a useful summary of what is
known concerning slave raids against the Orang Asli of the Malay
peninsula. Throughout the area it is clear that slave-raiding has gone
hand in hand with international trade from the earliest times. The
forests and mountains which protected the tribesmen of the inte ri or
from day-to-day control and routinized exploitation by the coastal
states, also left them militarily weak and unable to resist the organizcd
slaving expeditions which were periodically launched against them.
Shifting cultivation in the tropi cs requires a low population density and
is most efncient when the population is evenly dispersed. This is not to
sa)' that shifting cultivators were themselves innocent of slave-trading
and slavc-;IOIding. Among the Ngaju of south Borneo and the Maloh,
Melanau, Ka)'an and Berawan of central Born eo, for example, a class of
aristocrats was freed from subsisten ce production for long- di stan ce trade
by its ownership of a class of hereditar)' slaves. The Iban also took cap-
tives as well as heads from their enemies for ransom, sale or adoptioll.
The relative underpopulation of Southeast Asia also had consc-
quences for lowland societies in the area. One of the chief aims of
warfare for both coastal and inland states was the capture of prisoners to
suppl ement the domestic work force (Reid 1980:243; Tambiah 1976:
120).
The key to Southeast i\sian social systems was the control of meno Land \Vas
assumed to be abundant, and not therefore an index to power (with the partial
exceptions of Java and Central Thailand in relati vely modern times). lt is this
tha! distinguishes traditional Southeast Asian states from feudal Ol1es. Socicty
\Vas held together by verti ca l bonds of obligations between meno (Reid 1983: 5)
These bonds of obligation could take the form of voluntary loyalty to a
lord , indebtedness to a creditor, or outright ownership b), a purchasing
master. Chattel slavery was onl y the endpoint of a continuurn of clyaclic
relations of depenclency.
128
Raiding, trading and tribal autonomy
he prominence of debt and judicial sentence as sou rces of slavery appea rs to
be distmetivc Jll the Southeast .\sian pattern Since inability to paya judi cial
fine rna} be takcn as <3 variaban on inability to pay l debt, \Ve rnay safely regard
debt as the most fundamental source of Southeast Asian slavcry. (ibid: 159)
In lowlancl societies as \\'ell, then, trade goes along with slavcry,
allhough in a rather clifferent wa). In Ihe first place, prosperity clerives so
much from tracle that all social relationships become mercantilizecl, ancl
anv inabilih to maintain one' s creditworthiness soon leacls to one's
a' commoclity oneself. In the second place, centrali zed state
powcr was usually so wcak that semi-autonomous wealthy merchants
were ablc to surrouncl themsclves with virtual armies of slaves to defend
themselves ancl their possessions. Privately owned slaves constitute a
lilll itati on on the power of the monarch and tend to disappear where
central power is greatest (i bid: 18). But in the great trading citi es along
the coast, monarchs tended to be weak, and real power was in the hands
of a merchl nt plutocracy.
Crudcl y speaking, Southeast Asian soeieti es can thus be groupecl into
at least four types as regarcl s their experi enee of slavery. First, there are
the wcak, fragmented societies of the highlands whi eh served primarily
as a source of goods and slaves for the lowlands. Second, there are the
la rger, more organized societies of th e hi ghl ands which maintainecl
their OWIl class of sI aves for ritual purposes and to fr ee a small aristocracy
fOI long-distance trade. Thircl, there are the predatory trade-raicling
soeieties of the coastal arcas who regarded slaves primarily as a COIl1-
ll10ditv to be traclecl on to large urban centers. Ancl fourth , there are
thesc urban centers themselves. There is a fifth type of societ)' represen-
ted bv Bali ancl the Taraja which seems to have generated internally a
large \'olume of sI aves for e:\1lort cluring certain periods (v<1ll der Kraan
198 3; Bi galke 1983). Relatively clense populations dependen t on fi xed-
fiel d, wet-ri ce agriculture \Vere necessar)' for this somewhat un usual
si tuation to arise.
1t is incvitable then, that the highlalld soeieties which managed to
lai nta in their autonomy from lowl and statcs should give a great deal of
prominence to those same lowland sta tes in terms both of practical
rncasures of resistance and of sym bolism. Violent raidi ng, coerced trade
rOl jungle proclucts, ancl clebt bondage are al! probl ema tic areas to which
SOrne sort of response is reguired by their insertion in the regional
politiea} econOll1v.
12
THOMAS GIBSON
RAIOING ANO TRAOING IN SOUTHEAST
ASIAN CULTURES
Theprecedingdiscussion ofthe looselyintegrated regionaleconomyof
insular Southeast Asia has defined the context in which the societies
now to be described have had to operate. In thefol!owing discussion of
three groups of shifting cultivators, I shall focus on the moral and
symbolic values attached to activities in which raiding and tradingare
involved.
The Buid
TheBuid are one ofa numberofhighland groups in Mindoro knovm
col!ectively to tile lowlanders as "Mangyan," a term which has strong
connotations of inferiority, wildness, and even slavery. It should be
pointed out, however, thatthe lowlanders havebegun to use sandugu,
Clone blood,"or"blood brothers,"as aterm ofreferenceand add'ress for
the highlanders. Theimplication of, and reason for using this term, is
that the lowlanders would like the highlanders to feel bound to the
lowland speakers by ties ofartificial kinship. Thesetiesshould obligate
thehighlandersto treatthelowlanderswithaffection andgratitude, but
aboye al! to trade and to pay theirdebts. Thehighland Buid reject the
term as presumptuous, for they view the lowland traders as anecessary
evil, whoseviewofsocietyas madeupofmoralandmaterialdebtorsand
crt:ditors is the direct antithesis oftheir own.
In theBuid valuesystem, violence is onlytheultimateoutcomeofa
wholeseriesofnegativeformsofinteractionwhichbeginswithboastful-
ness (buagun) andquarrelinggarisugan "reciprocal aggression"). Isug,
"aggression,"connotesal! thathas a negative value in Buid sociallife:
the uncontrol!ed expression ofindividual emotion and theassertion of
egotistic desires, the attempt to dominate others, and, ultimately,
physical violence and murder. They regard maisug, "aggrcssive"
behavior, as being typical oftheir lowland Christian antagonists, \\'ho
the)' continual!y hold up as anegative example. This word has a\Vide
distri hutionin thePhilippines. AmongtheTausugMuslims, itrdersto
al! that is positivelyvalued in men: virility, courage, and the ability to
stand upfor onc'shonor(Ki efer 1972: 53).TheBuid, by contrast, ha\'e
no word for "courage," in the sense ofa positively valued aggressi\'e
attitudein theface ofphysicaldanger. Thereare manywordsfor fcarof
130
Raiding, trading and tribal automon y
andAightfrom danger,neitherofwhich is seen as beingreprehensible.
Indced, they are the onl yrational response to danger. It is recognized
that Buid may on occasion act in an aggressive manner, but such
behavior is attributed either to a weak mind, unable to control the
emonons ofits soul and body, orto affliction with afa ngarisuga n "spirit
which causes aggression." Iam aware ofonly onehomicide occurring
within aBuid populationofonethousand in aten yearperiod, and that
was carried out by aman with a fearsome reputation as a sorcerer.
Ratherthan compete in perfonningacts ofcourage, Buid youth are
mostlikely tospendtheirtimelearninglove poemswith which tocourt
young girls, The southern Buid and the neighboring Hanunoo have
preservedascriptwhich was used throughoutthePhilippinesin ancient
times. Themainpurposeofthisscriptis torecordand memorizepoetry.
Thesortofprestigeonecanacquirewithone'sknowledgeofpoetry, and
the sort of"conquests" one can make are the very oppos,ite ofthose
acquired through "bravery" and violence. Seduction is not associated
with aggression among the Buid as it is in other cultures which value
aggressiveness in men. Thepoemsthemselvesarefull ofgentleimagery.
Inaddition to theircategoricalaversion to aggressive behaviorofany
kind, theBuidplacean extremelyhigh valueon individual autonomy.
This is taken to such an extreme that even parents are reluctant to
intcrfere in thebehavioroftheirchildren more than is necessary. ' 1 'hey
striveinsteadto substitutethedependencyofsmal! childrenonspecific
adults with their dependenc\' on the adult community as a whole.
Betwcen adults, there should be no hillt ofdependency, either emo-
tionalormaterial. Suchattitudesaccountin partfortheextremely high
rate ofdivorce among the Buid. While the optimal economic unit in
Buid society consists of one adult maJe and one adult female, and
virtuallyal! adults belong to such a couple at a n ~ one time, there is a
high rateofturnoverin thecompositionofcouples.Adivorce is almost
always caused by one partner deciding to marry a new spouse. The
abandoned partner recei ves a sum ofmutual!y agreed compensation
andthenbeginshis orherown searcbforane\\!spouse, oftensetti ngoff
a chai n reaction. Children soon learn to acceptwhatever step-parents
Come their \Vay. (EIsewhere I have described in dctail Buid attitudes
towardsdyadic relations ofdependenc)' [Gibson 1985J.)
In order to maximize individual autonomy and mobility, some
mechanism must be found which allo\\'s people to detach themsclvcs
from theirprcscntsctofcompanionsandattachthemselvesto anewsct
131
T H OMA S G1BSON
without much trouble. Thi s mccha ni sm is what 1 call "sharing," and
was first clearly identifi ed in the study of hunting and gathering
societi es. It is based on the principIe that "cntitlelllent does not depcnd
in any wa) on donation" (Woodburn 1982: 441 ). Individual s are
obligated to share certain goods and services with the rest of the
community. Among the Buid the most obvious example is again meat,
wh ich is only butcherecl on ritual occasiollS, and whi ch must be sharcd
out in porti ons which are identical in both quantity and quality among
all those prcsent regardless of age or sex. Performing these rituals is no
route to augmented status, for the)' are onl)' carried out in responsc to
illness or c1 eath withi n the sponsoring householcJ. Their/ performancc is
thus a de monstration not of the wealth and vitalty of the sponsors, but of
their misfortune. No one is obliged to receive a portion of meat, and no
one is obliged to repav the individual s whose sacrifices they Ilavc
attended in th e past. T he only obligation is on the giver.
1 have argued elsewhere that th e characteristic manner in which Buid
cOllduct cOl1\CrsatiollS, help each other vvork 011 their fields, and,
indeed, mlrTy and divorce may best be described as the sharing of
speec], labor and sexual intercoursc IGibson 1986: 44-48, 83-84). Tll e
underl)'ing principIe is that strictly d:adi c rclationships are likel y to !cad
to either competition and aggression, or to dominance and indebtcd-
ness. If each indi \'idual sees himself or herself as interacting only witll a
un di fferentiated social group, th e possibility of losing one's personal
autononw is minimized. Sha ring prevents ties of personal depenclcnc"
from ari sing b) l1laking the individual dependent onl)' on the group, ancl
so ensUTCS both autollomy and equality within thc group.
Buie! ritual acti vitv is also carried out on a cooperative basis, and no
individual adopts the role of speciali::t with privileged access to the spirit
\VorJd. 1\11 adult men possess spirit familiars , and m\' stical power derives
from large numbers of men chanting together.
Constant particip;ltion in obligatory coll ectivc ri tual, in whi ch the Icgitirnac\ of
rcligiolTS belicf deri\es f10m personal cxperi ence of the spirit \Vorld, and 110t
from traditioll as interprctccl by elders or speeiali sts, from a sacred te:i!t, or from
the pri\'i1cged cxpclienec of a charislllatie figure , provides t!te Buid \\ith a
shared ce,e of belicf ancl a cl cep conviction of the rca lit)' of invisible p()\\ns
(Cibson 1986 148)
Buid vie\' themselves as the prcy of a host cf evil spirits who fcast 011
them when they die in the same \Va\ that thc Huid feast 011 their mm
pigs. ::meling within thc humall community is bcl ievee! to Tender
1 ~
'-
Raiding, trading and tribal automony
peoplc more vulnerable to attack by these evil spirits. In the religious
spherc, then, aggression is associated with violence, killing, and in the
case of the most dangerous evil spi rit, cannibali sm. The most feared
spiri t is the fangablang, a giant said to resemble a Christi an lowlane!er.
The Buid may be sa ie! to have an economic system which maxi mizcs
a ma rried couple's ability to be virtuall )' self-suffici ent, and a political
system which maximi zes people's ability to form ancl c1issolve social
rclationships. It eliminates competition for pmvcr ~ l l l d prcstige by
restricting ally form of dyadic interaction. 'rhe two institutions which 1
ha\'e saie! constitute thc l!lain forms of interaction between neighboring
societies in Southeast Asia, raiding and trae! ing, are both assigncd
negati ve symbolic values and referred to the exterior of the soci ety. Buie!
acquire lowland trade partners and put up with the c\ ploitation they
suffer bccause they must (cE Gibson 1986: 16). But thc concepts of debt
and dcpendency on which this trade is bascd are not allowed to operate
wi thi n Buid society. Raiding is a matter of suffering predation by nil
and aggressive outsiders in the material and mystical \VorJe!s, but
countcr- ra icl ing is confined to th e imaginary world of spirit mecl ium-
ship. Fli ght, and quick integration into ne\V communities through the
institu ti ons of divorce and sharing, are thc IlIain c1efenses against raiding
in thc material world.
Put more ahstractly, one can characteri ze the Buid and other groups
ofpeop!c like them as having an ideologv of ascribed equality: people are
considered equal no matter how successflll the)' are in various activities.
The llongot
he I1ongot of northern Lllzon resemble thc Buie! in their egalitarian
values ancl ass erti on of indi vi dual autonomy. Thcy differ from the Buid
primarilv in terms of thci r attitudes tO\vare!s violen ce and competition.
Scott (1979), ",ho has \Vorked out a cl assification of "unhispanizcd"
Phili ppine societies acco rdi ng to class structure, includ es the Ilo ngot
with he Mangya n in his "classless" category. Despite thcir fondness for
heacl hunting, Seott resists pl acing them in hi s category of "",arrior
societi es" bccause they do not "practice coup counting whi ch might
produce a warrior elite." Thcy also Iack a class term for accomplished
W<lrriors ' \vhich distinguishes a catcgory of men with shared status,
privilege and responsibility in the community" (1979: 156) \Vhilc
i ncJ udi ng the I1ongot and other \iol ent, class)css socicties wit h peaceful ,
1
-,-,
) )
THOMAS GIBS O N
c1assless societies is perfectl > l,=gitimate given the purposes of Scott's
c1assification, 1 require for my own purposes a c1assification which
privileges attitudes toward violence, aggression and dependency.
The motivation for engaging in violence among !longot youth is
precisely to achieve equality with adult meno The Ilongot do possess a
form of ranking, but it is of the simplest sort: that between youth and age.
Where the Buid are uncomfortable with and play down the dependency
of children on adults, the [Jongot emphasi ze it and play around with the
opposition between the passion of youth and the knowledge of age.
Inside the hou se men tl1 ydek [commandJ women, who in tmn pass their
commands on to children, and children are quick to tl1ydek those who are
younger than themselves. The dynamics of tl1ydek tend ultimately to permit all
adults the ordedy poi se of the platform - while requests for betel, food, tool s,
and water set childrcn in almost continuous motion across the relativelv
unordered Hoor. (M. Rosa ldo 1980: 72) .
The impetus toward violence comes from the fact that yo ung men
must pro ve their passion, or aggressiveness, in order to acquire a wife
and gain the respect of other adult meno
Of coursc, in dail y life, unmarried youths are the subordinates of thcir
established scniors. Though free to move and to resist requests that contradict
their plans and wis hes, young men lack wives and gardens of their own and so
are subject to "commands" (tl1ydek) by seni or kinfolk in whose households they
reside. Adult men may decide to "Iend" the labor of their sons, and thus affinn
their ti es of kinship and concern with their adult fellows. (M. Rosaldo 1980:
155)
With marriage as with beheading, thc youth transforms his social self by
demonstrating that he is an "equal" - dependent on no other man, and
equivalent, in "angry" force, to al!. (Ibid: 165)
Among the !longot, then, courting is associated with aggression. In
line with the cultural idealization of males as warriors is a relati"cly
greater elaboration of gender symbolism, in which tbe sexual division of
labor is more explicit, and the inability of women to scale the same
heights of achievement as men is stressed.
MicheJle Rosaldo (1980) points out that the ideal of autonomy among
the !longot applies less to individual s than to households. Divorce is rare
and a cause for violence (pp. 8, I n). Women are exchanged in
marriage by men, and are always under their authority (p. 84). Ilongot
men must acquirc control over a woman to become autonomous, and
until they do they are dependent on th e men who already control
134
Raiding, trading and tribal al1tomony
women. Thi s notion that one must gain control over women leads to a
rel uctance to relinquish control over them, unless it be in return for
control over another. Thus Ilongot youths reae! aggressively to the
suitors of their sisters.
Because marrying, Iike beheading, provides young men with terms that will
define their adult status, all men of a cohort may find themselves caught up in
fights concerning women - and it is not smpris ing that Ilongots oft en allay the
liget I."pass ion" or "aggression"] sparked by marriage with an "exchange" of
maiden "sisters". . (p. 170)
Tbe notion that marriage is a strenuous achievement, rather than a
natural result of physical maturation, parallels the need to achieve
equali ty with adult men through acts of bra very and violence.
The Ilongot economy is built around a symbolic opposition between
the "eoll ective" activity of men in hunting, meat distribution and
pollardi ng the trees in swiddens, and the "individual" activity of \\lomen
in planting, weeding and harvesting rice in swiddens (R. Rosaldo 1981 ).
While women probabl y cooperate with one another as much as do men,
"women's cooperative labor is seen as casual and informal, deserving
neither cclebratory feasting nor other public note" (M. Rosaldo 1980:
133). Thc llongot hold a "cultural view of rice as private produce, the
fruit of individual effort in a private field" (M. Rosaldo 1980: 133).
Hunting, on the other hand, even though it is often carried out by
individ uals for the private sale of meat in the market, is always bound up
witb the public domain; game is either traded with outsiders or shared
withi n the group: it is never intended for consumption by a single
llOusehold .
And the reluctance ofIl ongots to talk about inequality of skill among hunters is
rooted in a view of game as a collective product, publicly appropriate amI
consumed. (I bid: 118)
Adults both refu sed to boast about how much ga me they had bagged , and
persisted in maintaining thei r doctrine that no man was a more gifted
ht1 ntsman than another ... Men who found themsel ves again and again
compelled to coopera te on short-term projects fe lt they could not enlist the aid
of men who St008 aboye or below them upon any imagined ladder. Onl y
equals, as thev 5aw it, could work toget her in improvi sed and coordinated
hamlOny. (R. Rosaldo 1980: 145-46)
This brings us to what rsee as the peculiaritv of the ll ongot among all
the Southeast Asan societies so far descri bed, and which disposes me to
place them in a class on their own. Thcy place enormous stress on
135
THO M AS G IBSO N
achi evcl11e nt and competiti on, and yet they regul ate these types of
bcha\'i or in such a way that inequalih' cloes not res ult o That is, for thc
Il ongot, ac hi evement is not a matter of "more or less," but of "whethcr
or not ": wheth er or not a man has taken a head , acquired a \Vife,
provid ecl JOeat for the settlement. Having acquired the status of adult-
hood, there are no furth er statuses to be attained, onl y the one to be
maintained. For not onl y must equality be achi eved in th e first
instan ce, it is a fragil e thing whi ch is constantl y being put in question.
:\ dail y \Vorld in whieh autonomous and equaJ adults engage in eoopcrati\'e
interaction is ShO\\'ll , through oratory, to be as fragile as it oft en is ' .. (:-VI.
RosaJdo 1980: 220'1
Oratory is the domaill in which the Il ongot pl ay around with the
notion of dyad ic excha nge and the danger of imbalances between equal s
whi ch may result from it. Oratorical negobati ons are nne of the chi ef
cultural mcchanisms used by the ll ongot to define and crea te equality
bctween men o They are designed to move fr om a situabon in which
neg;, tive reciprocity obtains between fe uding groups, through the
bal anced reciproci ty of negoti ati on to the generali zed reciprocity of the
ki lI group, They requi re two balanced groups to even be initiated: if one
group is much stronger than its enemy, it si mpl y absorbs it or hounds it
out of exi stell ce ,
Formal eovenant appears, in short, to requi re an equaJity of diffcrcllce,
strength, and "anger" . , . Although covenants by th emselves do not neeess itate
further dealings, marri ages over time provide th e groun ds for future visits and
eooperati on; as sueh, they eonstitute a sort of transiti on fr om a state of
di ffe renee and balaneed oppos iti on to one of mu tuality and aekno\\'l edgcd
bonds ([ bid: 210-11)
It is as if Bui d-type rul es of sharing and the repression of aggression are
foll owed within Il ongot berta n (l iterall )' , "groups of peopl e of one kincl ,"
label ed by the Rosaldos' "bilateral descent groups"). But where the Btlid
oppose the operati on of these \ ;]ues wi thin their own society to the
contraIS values operative in lo",land Christi an society, the two \'alue
systems operate in differcnt social si tuati ons within Il ongot society.
Aggression, exchangc, and potcntial debt are Il ot evil per se in Ilongot
society, They are inevitable chardcteri sti cs of d stage in the maJe
I1ongot's li fe cvel e, as he seeks to acquire a \Vi fe , pro\'e his abilih to
defend hi s dutonomy, and cooperate \\' ith adll lt men as an equal. '['bey
are bal anced by the knowledge, mutualih Jll d acknowledged equality of
adlllts,
136
F\a iding, tradi ng and tribal autonomy
The Il ongot have dn econol1l v \vhich stresscs tbe alltonomy of the
household as opposed to the individual or the commlll1ity more strongly
than the Buid, Their politi cal S)'stCIll also binds an indi\ icl ual to hi s
"kin" group h) opposing it to other potent iJl ly hostile "ki n" grollpS,
Whcrc the BlIid see competiti on as leacl ing to the eventual cl ominati on
of one party over th e other, and so necessa ril y evil , thc Il ongot vi ew
compctiti on as a necessary stage in the achi evement of ecuality.
Jt mar be noted that the Ilongot live at thc headwaters of the largest
ri,'cr in Luzon, as far from the coast as one can geL Slave raiding \Vas not
the pers istent threat in thi s area that it \\'as throughollt thc Visavan
islands to the south , The onl y society in nort hern LlIzon whi ch
chattel slavcs was the Bengll et, where thcy \Vere employecl for
mine labor. Raiding tllfoughout the highl ands of northern LlI zon was
alwavs more ritual and political than economi c in moti ve: it was heads
and rcvcnge that warri ors desired, not pri soners to keep or trade as sla ves.
1t is al so possibl e that trade has not pl ayed as important a rol e in Il ongot
hi story as it has in soeieti es cl oscr to the coast. Aside fr om scattered
references to the sale of game, the Rosaldos make littl e menti on of trdde,
or depcncl cncy on tradi ng partners, In sum, 1 would argue that the
lIongot represents a society which bas not cl eveloped the sa me absolute
rejecbon of violence/raidi ng and debt/tradi ng as ha\e the Buicl , and that
thcre is less e\' icl cll ce of the presence of the in thcir symbol ic life,
due tu their relative isolati on from the Ill aritime \Vorld ,
The l ban
The Iba n are vigorous traders and raiders, as their bi story attests (Pringle
1970). As Sahl ins (1972: 224- 226) has argued, they are unusual for
"tri bal" people in keeping an exact account of lahor exchanged bet\\'een
households, and in limiting the sharing of staple food surpluses, This is
because lhe latter must be stored up for tracle \\'i th outsiders, Asicl e from
other fu nctions, the acqui si ti on of prestige goods such as jars and gongs
in traJe al so serves as a buffer <lgai nst a poor ri ce ha rn st:
Each scason, sorne slIeceed in producing a surpJus, whil e othcrs fi nd
Ihernselves wi th a defi cit; farnili es exehange gongs ror ri cel. or [rice! for gongs ,
Ja rs (l a;au), though to a lll ueh lesser (', tent, are ll seJ in the salllc \\'ay, Aga in,
Ill onev - obtaincd from the marketing of rorest produce - is often llScd to
purcha"e rice , , (Freeman 1970: 267)
13:-
THOMAS C IBSON
The Iban method of shifting cultivation is such that the labor of
young men is only reguired for hvo or three months ayear. The
domestic group, or bilek usually contains enough women and older
men to free young men for the acti vihes of tradi ng and raiding, or bejalai
(Freeman 1970: 222- 24).
The Iban live al ong large, navigable rivers and possess a fair degree of
maritime skil!. This meant that they are likel y to have been more
intensively in volved in mari time trade than mountain dwellers, and are
thus somewhat more sophisticated. But it al so gave them the ability to
mobilize much larger numbers of people than shifting cultivators
inhabiting rugged mountain tenain. At the same time, th ey had to
auopt more aggressive tactics in their own defen se. In the nineteenth
century they were able to put thousands of warri ors to sea in war boats
holding 60 to 100 men cacho They are reported to have served as
mercenaries in IIl an un war Aeets, the most fea red slave raiders in al! of
Southeast Asi a, although later th ey beca me their bitter enemies (Pringle
1970: 50-5 3, 76).
They have gone from being the must successful and expansi.onist
warriors of Sarawak, during the nineteenth century, sweeping all other
shifting cultivators before them, to being widel y traveled migrant traders
and laborers in the twenti eth:
For Iban young men then (a nd particul arl y since the cessati on ofheadh unting),
going on journeys is the grca test and most consuming interest life has to offcr . .
. the most powerful incentt ve is the very considerable social prestige which a
much-traveled ma n can commancl . Above all , it is only after having
accompli shed Ill any successful journe)'s that ama n begins the performance of
the seri cs of elaborate rituals (all based on the instituti on ofheadhunting) which
confer prestige in middl e and old age. (Frceman 1970: 222-24)
T radi honall y,
Whcn he had takcn a head, and only then, was an Iba n male entitled to havc
the back ofhis hands tatooed. With this achi eved, hi s pro\Vess was on constant
display . .. A successflll young head-hunter, it is sa id , could have his pi ck oftllc
most desirabl e young women, and \Vas mllch sought after as a husband. In
contrast, aman whu had nc\ cr taken a head, or who \Vas kn own to be rc1uctant
in battl e. woul d bc told b, t:1C \Vomen he courted: ... "FiTSt scale the posts of an
enemy 10ng-holl se; firstbedcck your ha ir as does he who has taken a hcad."
(Freeman 1979: 238)
Vi olence and aggression were directly Iinked to courting, and head-
hunting itself was preliminary to a ritual intended to secure both fe rti lity
138
Raiding, trading and tribal autonomy
and prestige fo r its mal e sponsor. These head-hunting ritual s are
"occasians for the celebrati an of the preoccupati ons and the narcissism
of mcn" ([' recman 1968: 388).
The lbans are known to have takcn their enemies capti ve for sale or
ransom, as \Vell as merely taking their heads. But there was no place in
Iban ~ o i e t y for a class of hercditarv sl aves. Freeman presen ts convine-
ing argu ments that war eapti ves \\'ere not incorporated as a permanent
stratum in Iban society, but were either ransomed or adopted into a
bilek. thc fundamental corporate group, as cqual members, with egual
rights to inheritance.
' nlC Cllstom of rituall y enfranchising and adopting capti ves takcn in war, which
was integral to thc tradi ti onal society of Iban, is a striking expression of their
cgalitaritl Il values. (Freeman 198 1: 47; cf. Rousseau 1980: 59)
Olle must not gel carried away, however, in cxtolling Iban
"egalitarianism" sim ply beca use unequal statuses were not, in
princ ipIe, hereditary. Blek units maintained careful aecounts of labor
crcdits anu debits between each other, and if an individual or hi s bilek
eould not repay a debt, he might be obliged to pay the debt "in part or in
wholc by the labor ofhimself or others ofhis bilek" (Freeman 1981 : 49).
At thc other end of the ladJ er of achi cvcment, \vas the raja brani,
literally, "ri eh and brave man ":
Once he had succeeded in taking a trophy heaJ , a warri or was entitl ed to tatl oo
on the backs of his hands, and hus adorned he could, if he had also amassed
sufficient \\'ealth, come lo be recogni zed as raja brani . This, howc\'Cr , was not
the ene! to an ambiti ous incli vi dual's quest for reknown, for there was <lI so in
Iban culture an ascencling scquence of complex ritual s, or gawai a171 at,
spceifi eall y assoc iated \Vilh thc cult ofh cadhunting, tha! he cou ld, ir he chose,
pcrform. These gawai a171 at (l it. true rituaJs) were al wa)' s perfo rmcc by
indiyiduals . . . Su eh rituals. the more eompl ex of which (lasting for four or five
days) dcmanded exlcnsivc rcsourccs, \Ve rc pcrformed beforc and aft er maj or
raids, \Vith the celebran! t3king the invocat ion (ti171ang) a stagc furthcr on each
occasiOIl unti l, over a span of fort)' or more \'cars, the sequence 'Nas complete.
(fi' recman 1981 : 40)
Even ritual Me. then, is eharacterized hy a hierarchy of achi eved
statuscs. Shamans fall into three classes: "raw" or novi ce, "ripe" or full y
ini tialcd, and "transfo rmed" or transvestite. Frecman describeu the
boasts of onc master shaman, and his denigration of the abilities of all
others (Freeman J967: 320).
Al though skill in rice cultivation \\'3S important to a man's prcsti ge, it
139
GIBSO N
is primaril y success in competitive and violent endeavors away from the
longhouse that is the measure of prestige within it. Internal ranking is
the resu lt of mastering an external realm of activity. The Iban sce
themselves as predators of the outside world: of neighboring tri bes,
whom they raid for heads; ofthe environment, whose virgin forests the\'
cut down; of the regional economy, in which they eam wages for brief
spclls befor e retuming home. They stand in marked contrast to the
BlIid, who see themseh-es as the prey of the olltside world.
In short, while the Bll id got the \Vorst of the regional system of trading
and raiding, the Iban thrived on t. The Iban ma)' be said to ha\'c a
politi cal economy which maximi zes the autonomy of a domestic group
(bilek), fr ees men for raiding and trad ing by assigning most of the
agriclIlturallabor to th e femal e members of th e bilek, and allows for the
temporar)' organiza tion of largc groups of men under charismatic
individual s, \Vho lead their follo\Vers on trading expeditions, migrations
to new river bas ins, and on headhunting expeditions (F' reeman 1981:
35-36). The over-all ethos is one of achie\' ed ranking. Every individual
is expected to prove himsel f, but there is a fin ely graded ladder of
achievement, with the ranking of individuals according to merit the
ultimate goal.
AN O
1'0 sllmmari ze: the Buid rej ect an)' form of violence, aggression, or e\-en
competition. Among the Buid, cIen quarr eling is thought to ha\'(:
mysti call y dan gerous eonsequences for the vitality of both hllmans and
crops. The !l ongot and the Iban stress the competit ive achievement of
status by typically engaging in sporadi c, violent assa ults on members of
neighboring soci eties in order both to demonstrate the virility of their
male members and to acquir e mysti cal vitality by obtaining heads. The
stress on individual achievement also lends a rather c)'clic charactcr to
violen ce in these societies. Raiding parties are organized by ambitious
individual s more or less according to whim 1M. Rosaldo 1980: 139;
Frceman 198 1: 36). As a reslllt, rncrnbers of thesc societies hal'c
historicall y acquired a great deal of notor et)' as headhunters.
AIl these societies represent specialized adaptations to the regional
polti eal economy. Al! lad instituted hierarchy. The cxtreme empllJsi s
on indi vidu al autonomy and rejcction of super-household allt]ority
evident among them must bc seen as a rejcct ion of the poltiC:J1 \alues of
I-H)
Raidlng, trading and tribal autonom y
thcir predatory lowland neighbors Far fr om cOll stituting primordi al
classless societies, they must be seen as political groups whi ch have been
able to mai ntain significant degrees of autonomy only by developi'lg
special social mechani sms for evading control by the lowh: .ds. '1 'he are
neithcr pri stine tribesmen, nor are they, to borrow a phrase from Fox's
argument concerning South 1ndian hunter-ga thers, mere adjuncts of
agrarian sta tes (Fox 1969).
Fo:>. and Gardner (1966) argue that the central features of South Asian
hunter-gatherer societies are attributable to their enclavcment within
agraria n states. These features include highl y migratory indi vidual s, a
stress on individual autonom)', and a va lue system which condemns
aggress ion and competi tion. These features make them clearly compar-
able lo the Buid. Fox and Gardner go on, however, to characte ri ze these
societi es as non-cooperative and as lacking sharing or reciprocity
between fa mil )' groups. This secms to be rather an exaggeration, for as
among most hunter-gatherer groups, even in South India , meat whi ch
is not traded is shared out equally among all members of a camp, while
gathcred vegetable foods are retained within the famil y (Gardner 1972:
41 5; Morris 1977: 231; Ro)' 1925: 69- 88) l'\ evertheless, it remains
possible that the expos ure of the South Asian hunter-gatherers to a
dense, fixed-fi eld agricultural populati on has pl aced them under greater
inter-cultural pressure than the Southeast Asian highlanders exposed
onl)' to the sporad ic interventions of the coastal trade-ori ented states.
Onc result of such pressure might be a necess it)' to further intensify
individual mobility and to further dimini sh the sizc of sharing groups.
Mobility and sharing are mutuall y reinforcing institutions and serve
as a highl y cffective mcafls of evading control by neighbors who cannot
be resi sted mil itaril y. As I ha ve argued above, shar ing is characterized by
a radical disconnection between giver and recipi ent. It is this character-
istic lack of strong dyadic bonds which has led writers like Gardner to
characterize mem bers of societies organized around sharing as "atom-
iSlic." But if one looks at sharing in terms of the moral bonds which
llnite the individual to the group, this characterization appears
inadequate. These bonds are often emphasi zed in collective rituals.
which act out the mystical dependcncv of the individual on the group.
Therc are many ins<lllces of socicties in highland Southeast Asia
whose members are rcpcatcdly dcseribed in the literature as being
systematicall y cheated of the Illarkct pricc of the produce the\ bring to
trade, and as beng perlT1 <Jllcntly incl cbted to lowland trad c.rs. lndeed,
141
THOMAS CISSON
these forest dwellers are often called by a term which means "sIave" or
debt slave, e.f;. Sakai for the Orang Asli, Ata, Aeta or Agta for the
Philippine ),egritos, and .\Jangyan for the highlanders of Mindoro.
Since the lowlanders aIready regard the members of these groups as debt
slaves, it would be fatal for those people themselves to acknowledge the
principIe of debt. These societies thus reject, ofien explicitly, the mast
fundamental principIe of lowland state societies: dyadic dependencv
phrased in an idiom or moral and material indebtedness . .
Individual mobility and a systein of sharing which allows people to
enter into immediate social relations without regard to past transactions
are t\vo essential features ofBuid society. But the rejection of attempts to
acquire prestige through acts of violence or through feast sponsorship
sets the Buid off from the IIongot and Iban as well . Among the Buid, the
symbolic equation of violence and dominance with life threatening
external forces is so complete that the central life generating rituals of
the !longot and Iban are negated: headhunting and the competitivc
sacrificing of large animals. Where animal sacrifice is carried out, it is
done in as non-aggressive a manner as possible.
The Buid share with the Ilongot and Iban a refusal to grant political
power to any individual or office in their respective societies. It has been
remarked again and again that the most "conserva ti ve" cultural groups
are those which lack internal stratification: finding no point of appli-
cation within the enclaved society on which to apply pressure, external
agencies must try to control members of the former one by one.
Colonial states confronted with such a society, tried to create "chiefs"
where none had existed, usually with little success (see, for example, the
appointment of Iban penghulu , "headmen, " by the Brookes [Pringle,
1970: 157]). In the extreme case of the Buid, even boastful behavior is
seen as immora!. It is quite inaccurate to characterize members of such
cultures as "submissive": they do not respond to attcIlIpts to dominate
them with violen ce as do !longot and Iban, but they are just as insistcnt
on the maintenance of their personal autonomy . Buid will ofien resort
to suicide if imprisoned by the state, or even threatened with imprison-
mento They will not voluntarily work for wages, disliking being placed
in a position where they must obey the will of another. They \ViII,
however, at times render labor at absurdly low rates of remuneration as a
sort of tribute to lowland patrons who furnish them sorne protectian
against the more violent elemcllts of the lowland population.
It is when groups ofhighlanders engagcdl in a sulDsistcl'lce systcm sueh
142
Raiding, trading and tribal autonomy
as hunting and gathering or shifting cultivation are brought into
unavoidable confrontation with the members of aggressive state systems
rhal om: is most likely to find Buid-type systems developing. The mode
of subsisten ce must be compatible with a certain degree of mobility, and
the highlanders must inhabit a tenain sufficiently resistent to eas)'
penetration by subjects oflowland states to make this cultural adaptation
feasi ble. It is not unlikely that all the formerIy autonomous societies in
sim ila rIy exposed geographicallocations which had an internally ranked
structure have long since been incorporated into the hierarchical state
systems of the lowlands, and that the autonomous groups which have
survived have done so beca use they possessed an ideology which
rejected any form of dominance.
CONCLUSIONS
The argument of this paper is that any explanation of the role of violence
and domination in the social lives of insular Southeast Asia shifting
cultivators must be sought in terms of their relations to the regional
politica! economy. This relation is inAuenced by their geographical
situati ons, and their consequent degree of exposure to the trading and
raiding economy of the region. But it is impossible to argue that their
idcologies are "caused" in any sense by their own subsistence economies
or by their role in the larger regional system. It is possible to argue,
howevcr, that their continued existence as autonomous political and
cultural groups can be explained in terms of their development of
ideologies appropriate to that end, given their different geographical
situati ons. G roups which failed to develop appropriate ideologies were
either absorbed into lowland states or eliminated. There remains an
urgcn t need to understand the interaction of societies with autonomous
ideologies and value systems in a common regional economy, without
eithcr reducing these ideologies to epiphenomena of the wider s)'stem,
or ignoring the real effects of commodity relations and milita!)' force
Upon the component societies in thc regional system.
'fhe argument in this paper addresses the question of the "causes" of
war and peace in Southeast Asia in a manner very different to the many
of the papers in this volume. In conclusion, 1 would like brieAy to
indicate sorne of the areas of disagreement, and the implications these
have for our understanding of human violen ce and war in general.
Somc writers view war as a natural phenomenon, susceptible of
I-B
THOMAS GIBSO N
explanation in terrn s of universallaws, rather than of parti cular histori-
cal dc\ dopments. Their tendency is to look for pan-human drivcs to
rnaximize either inclusi ve fitness (Chagnon, thi s volurn e), or access to
natural resources (Ferguson, Haas, thi s voJume). Thcse
attempts to treat a socio-hi storical phenomenon in naturali sti c tcrms
have familiar effects: sine: war is assumed to always derive from the
sarne essential cause, it is treated as a single , unitary phenornenon,
easil y identifiabl e across time and space. What is needed, rather, is an
atternpt to differentiate types of inter-group viol ence, and to recogni ze
that each t:pe may require an entirel y different sort of theory to account
for it.
As Ferguson (thi s vol ume) notes, questions about war and viol enee
var)' depending 011 th e broader theoretical context in whi ch they are
asked. If they are asked in the context of individual motivations or
sentiments, such as aggression, there will be a tendency to concentrate
on either biological or psychological issues, whi ch have a timeless
universal quality. If they are rai sed in the context of a struggl e for scarcc
resources , between evenl y balanced opponents, there is a tendeney to
vi ew it as a game betwcen pl aye rs who begin as equ als and onl y end as
"winner" al1 d "Ioser. " There is a reduction of inter-group violence to
timel ess questi ons of ta ctics and strateg)', of"game theory. " Focusi ng on
aggression and focusing on scarcity are both conduci ve to a view of
viol ent conAi ct as somchow pathological , the product of maladaptive
personality or social structures. Violence is both an inevitabl e and an
evil part of the human condition.
In this paper, violence has been di scussed in relation to exploitation
and to ritual. This approach highlights the rol e of vi plence in what is
perhaps its most preva k nt and sigllificant form since the origin of sta tes
and classes: the exploitation of a \\caker group - class, gen der. ethnicit)',
nati on, race- by a stronger group. In thi s view, viol ence and war ar e not
"about" natural scarcity or indi vidual sentiments, but "about" socially-
produced surpluses (for some) and scarciti cs (for the others). Occasional
c1ashes betwecn the ruling elites operating two parall el systems of
exploi tati on, which is the normal subject of in vestigati ons of "war," l11a: '
then be seen for what it is: a subsidiar)' and intermittant form of
violence. Furtherrnore, for the group whi ch profits from the \iolcnce
built into a system, violence will be seen as anything but a ll egati\c
phenomenon. It \Vill , 011 the contrary, be eelcbrated and glorifi ed. This
brings li S to the question of ri tual. Violence in any society \\iJl be
I-H
Raiding, trading and tribal autonomy
assigncd a \ alue, and it \Vill h, ve different values in different situati ons.
Very rarel y is it seen as evil in all situati ons, although the Buid provide a
rare case in which it is. In political economies based on the coe rei ve
exlraction of surplus value from neighboring soci eti es, inter-group
vi olell ce may even be feti shi zed as a product of mysti cal vitality in itself.
' ''hus violence is neit] er a necessary part of social life, nor is it
nccessaril y seen as an evil: there are societi es whi eh systematically
deva lue it, just as there are soci eti es, or ruling groups within societies,
which vi c\\ it as the ultimate good when exerci sed in appropri ate
contcxts against thc right opponents.
Notes
1 Thi s paper is based on researeh earried out among the Buid in the
ficldfrom 1979 to 198 1, andagai n in 1985, andonlibrary research
cOIlJueted in 1985-b6. The second peri od in the fi eld and the
li bra ry rescareh was fu nded by the Harr y Frank Cuggenheim
Foundati on. Different versions ofthi s paper ha ve becll prcsented at
the confcrence 0 11 the Anthropology of War hclJ at the School of
Ameri ca n Researeh, at the Scminar on Eql1ali ty and Incqua li ty
held by the London School of <, conomi cs Department of Social
Anthropology, and in the Coll oquiulll of the Department of
Anthropology of the Uni versity of Roehester. 1 hi s draft has
benefited from corn mellts by parti cipants in a11 these lTleetings, and
from detail ed comments by Wli alTl Henrv Scott, Mauri ee Bl oeh,
and !erome Rousseau. My thanks to all of th em. Whi le 1 have not
al ways foll oweJ their adviee, 1 have allVays found it stimulating.
1-+5