You are on page 1of 22

University of Pittsburgh- Of the Commonwealth System of Higher Education

Politics, Gender, and Time in Melanesia and Aboriginal Australia


Author(s): Eric Kline Silverman
Source: Ethnology, Vol. 36, No. 2 (Spring, 1997), pp. 101-121
Published by: University of Pittsburgh- Of the Commonwealth System of Higher Education
Stable URL: http://www.jstor.org/stable/3774078 .
Accessed: 06/05/2013 11:21
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

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

University of Pittsburgh- Of the Commonwealth System of Higher Education is collaborating with JSTOR to
digitize, preserve and extend access to Ethnology.

http://www.jstor.org

This content downloaded from 129.64.99.140 on Mon, 6 May 2013 11:21:07 AM


All use subject to JSTOR Terms and Conditions

POLITICS,GENDER,AND TIMEIN MELANESIAAND


s
MORIGINE AUSTIA
FJ b

Tt/

Eric Kline Silverman


DePauw University

This article interprets the symbolism and politics of Iatmul time (Sepik River, Papua New
Guinea). Social life is structured by different forms of time (e.g., totemism, myth,
Omaha terminologies, ritual). Furthermore, mythic history is a mode of ritual politics.
Finally, Iatmul time symbolizes paradoxes of gender. The article concludes by comparing
the temporality and gender of Melanesian cosmology with the Aboriginal dreamtime.
(Time, politicsS gender, Iatmul, Melanesia, Aboriginal Australia)

Anthropologists since Durkheim have argued for the social generation of time (see
Gell 1992: ch. 1). Evans-Pritchard's(1940) analysis of Nuer zstructuraltimen is
particularly striking. Not only did Evans-Pritchard argue for a determinant
relationshipbetween social organizationand time, he implied that Nuer temporality
was radically different from our own Western conceptions. This dual emphasis on
social determinationand cross-culturalvariationhas shapedthe anthropologicalstudy
of time with few exceptions (see Fabian 1983; Gell 1992).
Bourdieu (1977) and Munn (e.g. 1992) challenge the notion that time is simply
a structureof social life that differs aroundthe world. They focus on the phenomenological experience of time (see also Wagner 1986: ch. 5) and the use of time as a
symbolic resource in the pursuit of social strategies. This article draws on their
perspectives to analyze history and temporality in Tambunum an Eastern Iatmul
village along the middle Sepik River in Papua New Guinea.l I offer three related
ethnographicand theoretical arguments.
First, the pace of social life is structuredby multiple forms of history and time.
Each temporal modality corresponds to a particular social context. Second, the
society lacks a static and objective mode of history particularly mythic history7
because Eastern Iatmul often construct their past in accordance with contemporary
politico-ritual strategies. Eastern Iatmul tend not to recollect the past for its own
intrinsic value. To borrow from the language of Sahlins (1985), past zhappenings"
become historical aevents" only when they are socially and politically relevant in the
present. Likewise, zeventst must Elt into one or more culturally specific temporal
frameworks. The third argumentis that local concepts of time symbolize paradoxes
of Iatmul gender and cosmology If, along with Levi-Strauss, we understandculture
to be unresolvableproblems that arise from the imposition of order onto natureSthen
gendered time in the Sepik River is a symbolic response to the problem of a riverine
environment that is locally phrased in a reproductiveidiom.
In developing these themes, I begin with the relationship between totemism,
mythic history, and contemporarypolitics. I next present data on spatiotemporality
and narrative discontinuous time, followed by repetitive and cyclical time and a
101
ETHNOLOGYvol. 36 no. 2, Spring 1997 pp. 101-21.
ETHNOLOGY, c/o Deparment of Anthropology The University of Pittsburgh Piesburgh PA 15260 USA
Copyrighte 1997 The University of Pittsburgh.All rights reserved.

This content downloaded from 129.64.99.140 on Mon, 6 May 2013 11:21:07 AM


All use subject to JSTOR Terms and Conditions

102

ETHNOLOGY

debateon the temporaldimensionsof Omahakinshipterminologies.Subsequent


sectionsanalyzeparallelhistory,chronology,ritualtime, andgenderedtemporality.
The articleconcludeswith a comparisonbetweenEasternIatmulmythichistoryand
the Aboriginalconceptof the dreamtime.This, I believe, is the first attemptin the
literatureat comparingthe temporaldimensionsof thesetwo cosmologies.
TOTEMICNAMESAND THEPOLITICSOF HISTORY
With a populationof about 1,000, Tambunumis the largestIatmul-speaking
village. Patrilinealdescent groups (clans, lineages, and branches)correspondto
cosmologicalcategoriesthataredefinedby hereditarytotemicnames(tsagl). These
namesencodea modeof historyandtemporalitythatis crucialfor villagesocial life
(Bateson1932, 1936;Harrison1990;Wassmann1991; Silverman1996).
In what is one of the few uncontestedelementsof mythichistory,the original
moment,the
stateof the worldwas aquatic(Swadling1989). At an undeterminable
wangu)
(tsagi
pit
waterwas stirredby wind, andlandsurfaced.Therewas a totemic
thatis oftenenvisionedas the centerof the world;it is saidto be locatedin the Sepik
villageof Gaikarobi.Maleancestorsemergedfrom
Plains,nearthe Sawos-speaking
the pit, separatedthe sky from the earthwith forkedbranches,and createdthe
perceptibleworldthroughtoponymyor naming.Historyandtime effectivelybegan
with the migrationsof these ancestors,who, by conferringtotemic names to
phenomena,createdthe zpathsX(yembii)of the world. Each path is a cosmic
categorythatcorrespondsto a descentgroup.As a collectivememory,the physical
referentsof totemicnames,especiallythe landscape,enableEasternIatmulto know
theirdistantpast.2
over
Totemicnamesdetermineritualprerogativeslandrights,andcustodianship
ceremonialobjectssuchas masksandbambooflutes.Forthis reason,the ownership
andknowledgeof namescanbe fiercelycontested.Althoughtotemicspecialists(tsagi
father" (nyait) of eachpatriclanvigilantlysafeguardand
nuffl0a) andthe hereditary
rememberclan names, disputes are common (Bateson 1936:125-28;Silverman
1996:39-41).
analysis,humanactioncan be said
Fromthe etic perspectiveof anthropological
to alter(anddetermine)the totemicsystemand mythichistory.But at the level of
the statusof namesandmythichistoryis
local ideologyor emic self-representation,
less apparent.Sometimestotemiccategoriesor pathsare said to exist immutably
or underlyingtruth)the surface(aiwat) of humanity.Menwill
beneath(attndasfikiit,
cosmology in order to lend their version of history a
timeless
appeal to this
supposedlyrealratherthanpoliticaljustification.Althoughmensaythattheirtotemic
systemis timeless,theyalsoprivatelyagreethatpoliticalactionssuchas debatesand
disputesaltermythichistoriesandtotemicpaths.In this way, all totemicconf1gurations of the pastarecontestable(Harrison1990;Silverman1996:42-45).Indeed,so
concernedare men with the politicsof totemichistorythatthey becomenoticeably
uneasywhenaskedaboutthe namesandmigrationsof rivaldescentgroups.Theydo

This content downloaded from 129.64.99.140 on Mon, 6 May 2013 11:21:07 AM


All use subject to JSTOR Terms and Conditions

POLITICS,GENDER,AND TIMEIN MELANESIA

103

not want to appear guilty of stealing names and trying to restructurethe past in
accordancewith their own politico-ritualstrategies.
CONTESTED SPACE AND HISTORY
In 1988, the village had nearly completed a guest house that stands across the
river on a point of land called Agumurl. Agumoimbange,the hereditaryfather of the
Shui Aimasa patriclan, claims that his ancestors created the location. Along with
other senior men from the clan, Agumoimbangeplaced his name on the guest house
contract, which was cosigned by the Sydney tourist company that fundedthe project.
But Njumwi, a junior man from a different lineage within the Shui Aimasa clan,
protested the agreement since he was neither consulted during the negotiations nor
invited to sign the document. This caused a series of totemic debates inside the cult
house. Njumwi attemptedto assert totemic authorityover Agumurl by arguing that
his lineage ancestors and not those of Agumoimbangenamed and zgave birth" to the
location. As a contemporaryzhappening,"the lodge became an zevent" through the
politics of mythic history and totemism.
Since Njumwi's lineage ls demographicallyand politically minor in the village,
his claim over the guest house's ground was unable to become orthodox history. In
a broad sense, Njumwi's goal was to reorganize the totemic and politico-ritual
hierarchy of the patriclan by challenging the dominant mythic history. His
protestations, in fact, rekindleda series of totemic disputesthat reputedlyextend back
to the mid-nineteenthcentury, prior to European contact. The feud concerned not
only the jural arrangementof the clan but the very legitimacy of lineages. Material
wealth or money was not a primaryconcern, since the guest house employs men and
women from all descent groups as night watchmen, grass cutters, carpenters, house
cleaners, etc. Instead, men were competing for the symbolic power of names and
mythic history.
The totemic name of the disputed ground, Agumurl, is cognate to Agumoimbange's patronymic. Since personal names are totemic names, Eastern Iatmul
intimately identify with totems through zvonsubstantiality" (Harrison 1990:48).
Njumwi denied that the name of the ground was Agumurl; however, he also claimed
that its real name had been forgotten. In this sense, Njumwi's protest was less of a
claim to ownership and more of a thinly veiled threatto Agumoimbange's personal
identity, political authority, and totemic erudition.
This debate illustrates that the Eastern Iatmul past is memorable within the
framework of totemism only in so far as it pertains to contemporarypolitics and
totemic identity. Through his own version of history, Njumwi sought to reorganize
the clan hierarchyand sever Agumoimbange'spersonhoodfrom an importantlocation
in the mythic history of the village. But Njumwi's lineage has few members; most
have migrated to the cities of Papua New Guinea. Furthermore,his political status
has been marginal ever since he became an ardentChristianafter violating a sexual
taboo. For these reasons, gerontocraticconsensus was able to mute Njumwi's version

This content downloaded from 129.64.99.140 on Mon, 6 May 2013 11:21:07 AM


All use subject to JSTOR Terms and Conditions

104

ETHNOLOGY

of history and expunge its fictitioushappeningsfrom the collective and totemic


botharguedfor the
memoryof pastevents. AlthoughNjumwiandAgumoimbange
existenceof an objectivepast that is unalteredby humanpolitics, the practiceof
EasternIatmulhistoryallowsfor no lastingsense of historicalrealism.
Anotherincidentalso reveals the importanceof totemic history for deElning
a totemicspecialistand
presentevents. The lineageancestorsof Gamboromiawan,
one of my key researchassistants,migratedto Tambunumalongthe Korowariand
of theSepik.Severalyearsago, whilehe was
KorosameriRivers,southerntributaries
saw a lake thatwas absentEom the
visiting Korowarivillages, Gamboromiawan
totemicgeographyof his lineage.Thelakeexistedin thephenomenalrealmof human
experiencebut it lackeda totemicname.To solve this dilemma,Gamboromiawan
himself gave the lake a name and thus totemic existence in the EasternIatmul
cosmos. He selecteda Korowariratherthana Iatmulname.Butthis was irrelevant.
The importantissue was simplythat,becauseit existed,the lake requireda name.
In the discourse of totemism, contemporaryhappeningsrequiremythic-historic
justiflcationto becomereal.
MYTHICHISTORYAND CONTINUITY
SPATIOTEMPORAL
Munn(1977, 1986, 1992) and others(Burman1981; Damon 1990) repeatedly
emphasizethe relationshipbetweenculturalconstructionsof time and space in
truefor mythichistory.
Melanesia.In the case of EasternIatmul,this is particularly
Althoughempiricalevidenceoften supportslocal migrationaccounts(Wassmann
1990), I am more interestedin the symbolismof these movements.Duringtheir
plantedtrees,
primordialmigrations,ancestorheroes,amongotheraccomplishments,
to
corresponds
event
rituals.Eachmythic-historic
builtcult houses,andinaugurated
one or moretotemicpaths.Individualnamesreferto pointsin spaceandtime (see
also Wassmann1990, 1991). Takentogether,all the pathsof a descentgroupevoke
ancestral, creativeplace-to-placetravel involving increasingextensionfrom an
originplacet (Munn1992:101).This formof totemictime is spatialanddirectional,
nonrepetitive,successive,andcontinuous.In short,it is a vector.
A chant
Internally,totemic chantsevoke anothersense of spatiotemporality.
beginswith the public,visible, or surfacenamesof a group'stotem,whichreferto
its creationat a speciE1cnode in ancestralspace-time.As the chantproceeds,its
vectors,which
namesz moveXto anotherancestrallocation.Butunlikespatiotemporal
are formedby a series of totemicpaths(a pathof paths, one could say), a single
Consequently,totemicchantsevoke
chantalwaysreturnsto its originalspace-time.3
First, eachnameis a point in space-time.Second,
threetypes of spatiotemporality.
individualpathsshift betweentwo locations,beginningandendingwith the spacetime of a totemitself. Finally,a set of totemicpathsis a vectorthatdeElnesa descent
groupand recalls its primordialmigrations.Totemicforms of time enableEastern
Iatmulto know theirpast. But they also arousedeepemotionsin men andwomen
concerningtheirancestors,recentlydeceasedkin, andtheirchildren'sfuture.

This content downloaded from 129.64.99.140 on Mon, 6 May 2013 11:21:07 AM


All use subject to JSTOR Terms and Conditions

POLITICS,GENDER AND TIMEIN MELANESIA

105

Since totemicnamesare personalnames,the village populationat any moment


can be saidto embodyhistory.This embodimentis also spatialsince EasternIatmul
build their houses in descent-groupwards, which are dispersedthroughoutthe
village. Similarly,as we haveseen the local landscapeis delineatedinto zones that
are claimedby descentgroupson the basis of ancestralnamesand migrations.The
spatial groundingof time also influencesthe local perceptionof seasonality.
Tambunum
patriclansroughlycorrespondto the fourIatmuldirectionsmentionedby
Bateson(1932:254-55;see alsoWassmann1991:10-11, 203). ShuiAimasa(PigClan)
overseestheworldthatlies northof the SepikRiver,whichBatesonspellsmevambutagwi. MboeyNagusamay(SagoClan)claimskubalanggowi-gwi,the worldthatlies
south of the river, includingthe New GuineaHighlands.Woli-agwi,the eastern
worldof the LowerSepikandthe BismarckSea, belongsto Mogua(Fish Clan).The
SepikRiver(Avusett)itselfis thetotemicdomainof theWyngwenjap
patriclan(River
Clan).Finally,west is callediambzen-agwi,
whichrefersto the UpperSepikregio
thatbeginswithYambunvillage, a traditionalterminusof the EasternIatmulworld.
This directionis not, however,claimedby any one clan in Tambunum.
EasternIatmulalso associatewinds (mut)with totemic regions. Shui Aimasa
claimsthepwivu-mut
windthatblowsfromthe northernPrinceAlexanderMountains
duringthe rainseason.MboeyNagusamayownsmabEinjua-mut,
the cool breezethat
dropsfromthe Highlandsin the earlymorning.Mogua'swind is woli-mut,the dryseason wind that flows down the river from BismarckSea. Lastly, Wyngwenjap
claims baralagwa-mut,the northwestmonsoonwind that originatesin the Upper
Sepik.4Overall,then, EasternIatmulhave multipleforms of space-time:names,
paths,sets of pathsthatcorrespondto ancestralmigrations,directions,andwinds.
NARRATIVESOF THE PASTAND DISCONTINUOUS
HISTORY
Time in totemicpathsis continuousandspatiotemporal.
Yet EasternIatmulalso
portraythe past with myths of discontinuous,largelyatemporalevents. Although
theseeventsoccurredduringancestralmigrations,mythicrepresentations
areseparate
fromthe successivespace-timethatorganizestotemicpaths.Thehistoryof the Mboe
NagusamayclanSfor example, is a series of patrilineagemigrationsalong the
Korowari-Korosameri
river system. EasternIatmulcan recall this history as a
continuoustemporalsequenceof pathsandchants.Alternativelys
they can depictit
as disconnectednarrativemythsthatdetailseparateevents such as the creationof
sago anda primordialflood.
Each temporalform zremembersta diifferentsense of history.5WhenEastern
Iatmulwant to emphasizecontinuityand direction,they appealto spatiotemporal
paths. In a totemic dispute, men often recountpaths of space-timein order to
demonstrate
thata contestednamefits intotheirownancestralmigrationanddescentgrouphistory.In fact,the entiretyof the world,fromcosmiccreationto the present?
canbe summarizedwith a few sweepingpathsof totemichistory.However,Eastern
Iatmulreferto mythicnarrativeswhentheywantto recollectin detailspecificevents

This content downloaded from 129.64.99.140 on Mon, 6 May 2013 11:21:07 AM


All use subject to JSTOR Terms and Conditions

106

ETHNOLOGY

and locations.Totemicpathsand myths, we could saySare inverseElguresof the


samehistoricaltrope.Bothcontaintotemicnames.Buttemporalityanddirectionare
in the backgroundof myth,yet in the foregroundof totemicpaths.In this way, the
meaningof myth is largelynarrative,whereasthe meaningof totemismis spatiotemporal.
Bothtypesof historicalthoughtandtemporalityareunrelatedto chronology(see
below). As Bateson(1936:223-24)notes:
lW1hen a Iatmul native is asked about some event in the past, he can as a rule give an immediately
relevant answer to the question and does not require to describe a whole series of chronologically
related events in order to lead up to the event in question. The Iatmulindulge very little in the sort of
chronological rigmarole which . . . is characteristicof those primitive people who have specialised in
rote remembering.

This observationremainstrue today. Bateson(1936:224) also notes that totemic


debatesare handled by the speakersnot as a continuousnarrativebut as a series
of smalldetails. Thisrequiresclarification.Whentotemicdebatesincludechanting,
sequencethat lacks, strictly
men recountthe past as a continuousspatiotemporal
speaking,narrativestructure.But men also recitediscontinuousmythicnarratives.
temporalcontinuityfromnarrativeper
In otherwords, Batesondid not differentiate
se.
time, is political.In fact, my own
Discontinuoushistory, like spatiotemporal
presencein the village promptedthe creationof a potentiallynew myth(whatmay
be termeda zfakemytht6)thatallegedlyaccountsfor the originof ritualartby a pig
spiritof the Shui Aimasapatriclan.Accordingto its author,the fake mythdetailsa
legitimatepastaevent."Butit existsonlyas narrativesincetheeventdoesnotfit into
any spatiotemporalor totemic path. Althoughthe fake myth lies outside the
view of the patriclan'spast,its authorwill soon inheritthe
consensualgerontocratic
position of patriclanzfather, whereuponhe may try to use his new authorityto
anchorthe fake mythinto a totemicpath.It wouldthenbecomean actualmythand
its historical zevent" would be commemoratedin several temporalforms. The
inventionof the fake myth was a clear attemptby its authorto demonstratehis
who was recordingvillagemythology.Thus
totemiceruditionto the anthropologist
it illustratesonce againthatthe pastis shapedby differentmodesof time whichare
shapedby sociopoliticalconcernsandcontexts.
AND
OR CYCLICALTEMPORALITY
REPETITIVE
OMAHATERMINOLOGY
EasternIatmulgrouppersonalnamesandgenealogicallevels intotwo alternating
44lines"(mbapma). A male Ego andhis FF, SS, etc., belongto one line; the other
line includesEgo's S andF (see also Bateson1936:244).Thus a man inheritsthe
namesof his FF, andgives themto his SS, in perpetuity.The systemis identicalfor

This content downloaded from 129.64.99.140 on Mon, 6 May 2013 11:21:07 AM


All use subject to JSTOR Terms and Conditions

POLITICS,GENDER,AND TIMEIN MELANESIA

107

women, who receivenamesfrom FFZ andbequeaththemto BSD. On the basis of


sharednames,personsidentifywiththe living, dead,andfutureagnatesin theirline.
These identiElcations
generatea sense of repetitiveor ziterative"history(M.
Schuster1990; see also Panoff 1969:160-61).A person'sactionsare often said to
replicatethoseof paternalgrandparents
preciselybecausethey havethe samenames.
This is especiallytruefor the FMBSDor iai marriagepreference,whereEgo andhis
wife ideallyhavethe samepatrinamesas theirFF andFFZ respectively(Figure1).
Shouldthe exactnamesdiffer,these alternate-generation
kin are still identiEled
with
each othersince they and their namesbelongto the same line within the lineage.
Throughthe namingsystem, then, the past becomesnoncumulativeand cyclical,
repeatingevery secondgeneration.

A=d
FF

(FFZ)

l
l
A = O
Ego

iai
(ego)

Figure 1: Iai Mamage and Names

Living people may even be held accountablefor the actions of their past
namesakes.Illnessis oftenattributed
to mysticalretribution(vai) thatresultedfrom
a lineal grandparent'ssorcery or ritual transgression.This sense of time is
particularlysalientwhen EasternIatmuldivine the cause of death, sickness, and
misfortune.By choosing or refusingto identifykin in alternategenerations,they
transformcalamityinto a meansof evaluatingandactingon existingsocialtensions.
It is not that EasternIatmulare unawareof temporaldistinctionsbetweengenerations;afterall, temporalsequencesareinherentin mythichistoryandspatiotemporality. Furthermore,circularor repetitivetime adoes not logically exclude 'linear'
sequencingbecauseeach repetitionof a given 'event' necessarilyoccurslaterthan
previousones"(Munn1992:101).I am simplysuggestingthatcirculartime coexists
with lineartime and is a politicalstrategyas well as a structureof social life.
A senseof nonincremental
andcyclicaltemporalityalsoemergesfromthekinship
terminology.A typicalOmahakinshipsystem has some combinationof the lineal
equationsMBD= MZ, MB= MBS, andFZD-ZD . In Tambunum,MBD= MZ, but
thereare also the alternate-generation
equationsMB=MBSSiMBS andFZD=
DD s&ZD.Furthermore,
not all womenin the mother'spatrilinearetermedzmother"
(nyame); only for M, MZ, MBD, and MMZD, but never MM and MMZ (cf.
McKinley1971a:246,n. 4). Finally, EasternIatmuldo not lump all men in the
mother'spatrilineageunderthe sameterm.Instead,theycall themeitherMF or MB
in accordancewiththe two intergenerational
lines. Withthesevariationsin mind,the

This content downloaded from 129.64.99.140 on Mon, 6 May 2013 11:21:07 AM


All use subject to JSTOR Terms and Conditions

108

ETHNOLOGY

EasternIatmulsystemis oftenconsideredto be an Omahaterminology(Ackerman


1976; Williamson1980).
Levi-Strauss(1963:74)suggeststhatCrow/Omaha
systemscontainthreedifferent
ztime continuums.tFirst, thereis intergenerational
time thatis zprogressive,nonreversible"(Levi-Strauss1963:301).Forexample,womenin Tambunum
use separate
termsfor MM(mbanamb),M (nyame),Z (nyanggae),D (nian),andDD (na, kaishelagwu).Second,timeis zstableandreversible"(Levi-Strauss
1963:301).Consequently, EasternIatmulmencall all womenin the father'smotherSs
patrilineiai regardless
of generation. Third, time is aundulating,cyclical reversibleb(Levi-Strauss
1963:301).Hence menuse alternatingtermsfor agnatesthatcorrespondto the two
lines:thetermsfor father(nyait)andson (nian)areinterchangeable,
whereasFF and
SS arebothnggwail.Levi-Strauss'sanalysisthusseemsconsonantwithmy argument
thatdifferentmodesof time structurethe tempoof EasternIatmulsocial life.
McKinley(1971a, 1971b)arguesthatCrow/Omaha
societiesdispersemarriage
alliances,therebypreventingthe regularrenewalof affinallinkstYet they maintain
the value of affinal bonds throughritualexchanges.Crow-Omahaterminologies
resolve this contradictionby perpetuating
the originalmarriagealliancethrougha
terminologythatfreezesthe passingof generational
time (McKinley1971a, 1971b;
see also Sahlins 1985:53). Barnes (1976:392) criticizes McKinley'sfailure zto
distinguishproperly between lines in a terminologyand empirical lineages.
AlthoughEgo in an Omahasocietycalls men in his wife's lineageone kin term, his
son calls them another,and so on. In this manner,accordingto BarnesSOmaha
terminologiesactuallyemphasizeratherthansuppressgenerationalsuccession.But
this is true, I suggest, only fromthe diachronicor intergenerational
perspectiveof
any one lineage reflectingon another.From the perspectiveof a single Ego,
generationalsuccessionappearsfrozendownthe maternalor affinallineage.
This is also the case for the Manambuof Avatip,a non-IatmulSepik society
wherea 4'patrilineage
is knownby thenameof thegenealogicallymostseniorsisterSs
child(preferablya sister'sson) of its seniorlivinggenerationX
(Harrison1984:398).
As perceivedby this sister'schildSsame-sexmembersof the avunculatelineageare
terminologically
equated;the lineageacquirescorporateidentitythroughtheabolition
of generationaldistinctionsandincremental
time. Whenthe nextgenerationascends
to prominence,the lineageis namedafterthenewseniornephew.Intergenerationally,
the lineage is a series of distinctly named stratathat emphasizesincremental
temporality.But to a singlepersonnthe lineageappearsto freezetime.
My pointis thatsocialactionin theseso-calledOmahasocietiesseemsto involve
a dualsense of time, one serialSthe otherstatic.The differenceis merelya matter
of perspective.Analytically,one can try to extracta singularmode of time from
Omahakinshipsystems.However,we canbetterapprehend
the multidimensionality
of cultureand social strategieswhen we view these terminologiesas containing
multipletemporalitiesthatare structurally
significantin differentcontexts.

This content downloaded from 129.64.99.140 on Mon, 6 May 2013 11:21:07 AM


All use subject to JSTOR Terms and Conditions

POLITICS,GENDER,AND TIMEIN MELANESIA

109

PARALLELHISTORY
Iatmulvillagesaredisputatiousandfissile. Totemicdisputesare commonsince,
at some level, each descentgroupboaststo be the '{fonset origo" of the universe
(Bateson 1936:127). Yet it is not the case that these cosmogonicassertionsare
necessarilyopposed,for in manyinstancesthey generatemultiple,almostparallel,
historicaltruths.Thisallowseachgroupto affirmits primacyandautonomy,yet also
to recognizethe legitimacyof adversaries.If it is in theirinterestto do so, menfrom
differentlineages and clans will agree on one totemic cosmology. Yet on other
occasions,men stateexplicitlythat awe have our own history,they have another.
Ours is for us, theirs is for them.- This sense of pluralismis often evokedwhen
people refer to villages that have the same primalancestorsbut differentmythic
historiesof unrelatedeventsandmigrations.
This is not to say thatEasternIatmulenvisiontheirpastas a unitaryframework
withparallelsegments.Rather,therearemultipleor parallelhistories.For example,
the mythof Mendangumeli,
a seniorcrocodilespiritwho floodedtheworld,parallels
the biblical accountof Noah which was introducedinto the Sepik by Catholic
missionariesearlierin this century.However,the two floods are not combinedinto
a single event. They referinsteadto differentyet analogoushistories,each truthful
in its own context.Thereis no contradiction,since this view of the past does not
presupposea unified historicalreality(see also McDowell 1985; cf. G. Schuster
1990;Pomponio1992:51).
Parallelisminfluencesnot only time andhistorybut also a moregeneralview of
the cosmos in which there are separateroads (yembii)or aplanes of existenceX
(Bateson1936:237)inhabitedby humans,witches, tree spirits, and othermystical
beings. Similarly,the totemicsystem consists of public or surface(aiwat)names
which conceala hiddenessence of realitycalled attndasiikiit.Thus truthis always
layeredfor EasternIatmul.
CHRONOLOGICAL
TIME
Chronologicaltime is linear,incremental,andprogressive.It is emphasizedby
Westerninstitutionsandsuch thingsas calendars,government,employers,andthe
ubiquitouswristwatch.EasternIatmulhad two traditionalnotions of incremental
time. One, akinto hours,was determinedfrom the positionof the sun. The other
was delineatedon the basis of lunarphases. The traditionalmonthsor amoonsX
(mbop)were:
1. Mandarimbo:
shortageof food (January).
2. Tshiimbuniimbiit:
H1sh
largelydormant(February).
3. Kuganniimbiit:
fish plentiful(March).
4. Kambowanmbo:
good moonlight,increasedf1sh(April).
>*

r **
vv rr r r s s s
wunaanllm
rllt
llttleTOOa
May*

xs X

This content downloaded from 129.64.99.140 on Mon, 6 May 2013 11:21:07 AM


All use subject to JSTOR Terms and Conditions

110

ETHNOLOGY

6. Kambokwurlniimbiit:partially cloudy nights (June).


7. Tumbokwurlniimbiit:overcast nights (July).
8. Tskinwolimange: moon rises in the early evening (August).
9. Avawolimange: strong moonlight (September).
10. Lananmbo:completely clear nights (October).
11. Lumbanevimange:crescent moon zturns on its side" (November).
12. Pagunimmbiit:cloudy nights (December).
Bateson (1936:254) observes:
Nominally the year consists of twelve moons, of which Elveare moons of high water and Elveare moons
of low water. Between each of these groups of Elvemoons, there is an intermediatemoon. But inasmuch
as the rising and falling of the river is very irregular,my informantswere generally doubtfulas to what
month it was at any given time. The astronomicalyear of 365 days contains of course approximately
thirteenlunar months, so that with their theory of twelve moons in the year, the Iatmulcould never be
precise in the identiElcationof the moons in their calendar.

But chronological precision is not particularlyuseful since the timing of gardening,


ritual, and most other activities is dependent almost entirely on river flooding.
Western months only determinethe school year, Easter, New Year's Eve, elections,
and so forth. Even when Iatmul traditionally scheduled events such as rituals in
advance, they often postponed them until the flood season (Bateson 1936:254-55),
which they expressly try to avoid. I suspect this occurredbecause of village disputes.
Social practices, then and now, often thwartedchronological precision.
Since temporal and seasonal distinctions were traditionally determined on the
basis of a perceived continuum (e.g., the movement of celestial bodies), there was
considerable chronological fluidity. Temporal units were not accorded a numerical
quantity. Like the Tiv (Bohannan1953), EasternIatmultend to aindicate"ratherthan
zmeasure" time, despite an extensive counting system. Their lunar calendardid not
progress numerically, quantitatively,or even cumulatively. Years were unnumbered.
yesterday
The passing of days was organized only referentially:today (mbambra),
day-after-tomorrow
(kinya),
tomorrow
(nangayt),
day-before-yesterday
(naramba),
four days hence (wainya),and Elve days
(ma),three days hence (nangaytmboey),
EasternIatmuloccasionally tie knots, usually five, on a cord of
hence (satngvande).
twine to indicate the numberof days that must elapse until markets. But this, too, is
a relative ratherthan absolute series.
There is magic that can hasten the passage of lunar phases. But idioms such as
zlosing time,- asaving time," and aracing against time" are quite alien. A sense of
temporal scarcity doubtless pre-existed contact. But it is exacerbated today when
persons must occasionally choose between a wide rangeof activities such as repairing
the school water tank, awaiting the arrival of tourists, or traveling to town (see also
Smith 1982). Still, the experience of temporal scarcity is relatively infrequentfor
most villagers. For example, nearly all men carve tourist art. But, to use Western
idioms, nobody has developed a uniformpricing scheme that systematicallyaccounts
for the amount of time spent carving different types of objects.

This content downloaded from 129.64.99.140 on Mon, 6 May 2013 11:21:07 AM


All use subject to JSTOR Terms and Conditions

POLITICS,GENDER,AND TIMEIN MELANESIA

lll

Ironically,the clash betweenEuropeanchronologyand indigenoustemporality


hasgivenEasternIatmula sensefor the socialgenerationof time. I heardstatements
such as zEuropeanseat lunchat 12:00 noonXor drink a cup of whiskeyat 11:00
p.m." They recognizethe Westernpractice(and power) of assigningnearly all
activitiesto a set chronologicalinterval.A moreflexiblesense of time governsthe
paceof indigenoussocial life (Bateson1936:254).Menwill sometimesdrinkbeerin
the morningwithoutfeeling thatthis time shouldbe devotedto otheractivities.A
devotionto time, however,is in theireyes a hallmarkof Westernization,andthus
znativetime" becomesa local meansof resistingthe dominationof Westernwork
andschedules(see Rose 1992:216).But whenthe Westerntemporalgaze becomes
internalized,chronologybecomesa meansof ethnicself-denigration.Time, in this
sense, symbolizespostcolonialtransformations
(Smith1994:ch. 8).
RITUALTIME
AlthoughEasternIatmultimecanbe incrementalandlinear,the namingsystem
andtotemicidentifications
seemto mergethepresentandpast(see Bateson1936:35,
n. 1; Harrison1984:400-01).To somedegree,so doesthe cyclicaltemporalityof the
kinshipsystem. This form of time is also presentin EasternIatmulritualssuch as
curingrites, which often enactprimordialevents as if they were occurringin the
present.7
Personsbecomeill whentheyviolate,directlyor throughkinshipidentifications
(see above), ritualand social taboosthatmaintainthe world and social life as they
werecreatedby mythic-historic
ancestors.Illnessin this sense is a somaticresponse
to the moral violation of a person's ancestralpaths, which are createdanew by
totemicchantingduringa curingrite. By compressingtime, chantsempowertotemic
specialiststo act simultaneouslyin the primordialpast andthe present.
In anotherceremonialcontext,the presentactuallyaltersthe temporalorderof
the past. Cosmiccreationwas followedby mai spirits(Hauser-Schaublin
1983)and
then humanity.Later, accordingto a commonMelanesianmyth, men stole ritual
sacrafromwomen.However,EasternIatmulnearlyreversethis temporalsequence
when they dramatizethese events duringritual in the order of mai, theft, and
creation.The voicesX of the maispirits,menreport,evokesadnessin women,who
recalltheirloss of sacra.The ritualinvertsthe temporalsequenceof the primaltheft
and the mai in orderto cast the entireceremonyin a sentimentthat emphasizes
masculineceremonialprestige.
Umedaritualis relevanthere. For the Umeda,Gell (1975:334-46)argues,time
has two everydayforms. Diachronictime progressessequentiallyas durationand
organicprocess(e.g., aging).Butsynchronictimeis constitutedby staticoppositional
categories(e.g., seniorandjuniorgenerations).The ida rite resolvedthis temporal
contradiction.First, it regenerated
societyby areversing"time (Leach1961),which
was still sequentialor diachronic.Second, the rite mediatedbetweentime as za

This content downloaded from 129.64.99.140 on Mon, 6 May 2013 11:21:07 AM


All use subject to JSTOR Terms and Conditions

112

ETHNOLOGY

continuous process and a synchronic opposition between young and old" (Gell
1975:343)
Recently, Gell (1992: ch. 5) revised his interpretation.Insteadof reversing time
in order to regenerate society, ida
represents a process undoing itself in normal, forwards-runxiingtime, not a process going forwards
[W]hatthe Umedas want to do is regenerate their world in
normally in backwards-runningtlme....
real time, not have it continue inexorably on its degenerative course, but in inverted time. (Gell
1992:51-52)

Furthermore,Gell (1992:52) now argues that synchrony and diachronyare the same
form of time, since synchrony is a classif1catory mechanismwhich arrangesentities
according to the real or putative events in their histories." The ida ritual does not,
therefore, mediate between two different forms of time. Rather, ida reinforces Wthe
Umedas' confidence in the viability of their societyb by mediatingbetween cultural
classifications'' and some external zreality" (Gell 1992:53; see also Werbner
1989:160).
While Gell (1992 passim) is concerned with reducing the cultural variation of
time into basic, universal forms, the EasternIatmuldata suggest that social life in the
Sepik occurs throughmultiple and often contradictoryforms of time. Time is not just
a classif1cationsystem that attemptsto orderreality. It is also a structureof social life
no less a symbolic resource. But time in the Sepik River also has gender, and thus
connotes a fundamentalyet unresolvableculturalparadox, to which I now turn.
HERACLITUS IN THE SEPIK: WATER, TREES, AND GENDERED TIME
Eastern Iatmulhave a genderedconcept of time thatjuxtaposes stasis and change
through idioms of trees and water. Totemic chants commonly substitute the word
tapma(coconut palm) for ngepma(village), and migration accounts often focus on
male ancestors planting trees (mi) and establishing villages (Gell 1975; Tuzin
1992:110-11). In ritual, moreover, coconuts symbolize the heads of homicide
victims, whereas myth associates coconuts with testicles in additionto heads. In the
local discourse of kinship, patrilineagesare like trees that are connectedby vines and
birds, which symbolize women and marriagealliances. Lineages are akinto sprouting
seedlings that droppedfrom an ancestraltree; the meaningof lineage branchesis here
self-evident.
The symbolism of mortuaryceremonies supportsthese associations. Outside the
cult house, men construct a nyaitmi (father tree) from totemic bamboo and flora.
Totemic specialists gather around the father tree and chant totemic paths that are
associated with the primaryancestors and spirits of the sponsoring lineages or clan.
In spite of death and the impermanenceof human life, the father tree signifies the
perduringgrowth of these descent groups and the vitality of their ancestralor totemic
power. Overall, there is an arborealmetaphorin the culture that signifies masculinity, male fertility, social life, patriliny, and permanence.

This content downloaded from 129.64.99.140 on Mon, 6 May 2013 11:21:07 AM


All use subject to JSTOR Terms and Conditions

POLITICS,GENDER,AND TIMEIN MELANESIA

113

Furthermore,
EasternIatmulexpresslyassociatewaterandthe SepikRiverwith
femalefertilityandreproduction,
in particular
thewomb(nian uxt; lit., child'sstring
bag). As Eliade (1958:188-215) and Douglas (1966:161) emphasize, water
symbolizesformlessnessandtimelessness(see also Tuzin 1977). But waterenables
the creationof form, substance,andtemporality,or social life. The SepikRiver is
continuouslychangingandflowing,forevererodingthe terrestrialrealmof treesand
villages. Drawingon the masculineidiomdiscussedabove,trees in this framework
signifythe maledesirefor genealogical,social,andhistoricalpermanencein a world
that is forevertransformingand threateningto returnto the originalstate of the
primordialfemininesea (Schuster1985). As a terrestrialsymbol, the fathertree
contrastsstrikinglywith the finalaquaticimageof the funeraryrite:the expulsionof
recentghosts(wmndumbu)
downthe riveron thebackof crocodilespirits,outto sea,
andto the place of the dead.
Likethe pre-SocraticphilosopherHeraclitus,EasternIatmulrecognizethe flow
of changeandthe possibility,in theircase, thatthe majestic,floodingSepikmight
somedayprevailand encompassthe world. In an idiom of gender,the female-like
riverthreatensto engulfmalehistory(e.g., land,trees, andvillages)into its waters.
EasternIatmuldo not have a tragic eschatologylike the Umeda(Gell 1992:38).
Symbolically,though,genderedimagesof EasternIatmulkinship,time, change,and
stasis hint at a sense of aquaticdestiny, as do men's emotionalresponsesto the
erosionof the village andtheirhouses. As the Sepikflows, so does EasternIatmul
history.Butthe riverflows in two metaphoricdirections:forwardtowardthe future
of society, yet backwardtowarddissolutionandundifferentiation;
thatis to say, the
iFeminine
and aquaticstateof the worldthatexistedpriorto masculineand totemic
creation.All men can do to stave off this wateryfate is to rebuildhouses, stage
rituals,andchanttotemicnames.The masculineillusionof andyearningfor stability
is ultimatelyawashin femalecontingency.
This genderedsense of time is relatedto the humanlifecycle.In local ideology,
conceptionresultsfrom sexualintercourseandthe mixtureof semenandmenstrual
blood. Semen becomes bones that form a stable core for the growth of delicate
organsandflesh, which originatefrom menstrualblood. At birth,the child begins
a lifetimeof exchangeswith matrikinespeciallymother'sbrothers(Bateson1936).
Like flesh, maternalidentityis fragile and thus requiresconstantattention.But
paternalidentity,like bones,is fixed andpermanent.It needslittleor no care.Hence
fathersandsons rarelyinteract,butthe avunculateandhis nephewhavea close and
activerelationship.
At death, maternalflesh decays, leaving only paternalbones. Althoughthe
desiccatedcorpse is male, the ghost or soul of the deceasedtravels under its
matriname
to theplaceof the dead,inhabitedby maternalancestors.Thetermination
of the humanlifecycleis a metaphorfor the relationshipbetween,on the one hand,
men, trees, villages, and land, andon the other,women, water, andthe river. For
the former timebuildsandshapessociallife. Forthe latter,it disruptsanddissolves

This content downloaded from 129.64.99.140 on Mon, 6 May 2013 11:21:07 AM


All use subject to JSTOR Terms and Conditions

114

ETHNOLOGY

it. Both forms of time, like land and water, bones and flesh, are necessaryyet
antithetical.
Trees and villages are masculine;so, too, is dry ground, especially when
comparedto the river. However,we hearin one myththat landis the creationof
maternalancestressesin the guise of floatingislandsor agwi; as seen above, this
word is also the SUfflX for the names of the four cosmic spaces. Furthermore,
domestichouses symbolizethe mythologicalagwi thatform the terrestrialworld,
which men continuouslybuild in orderto assertthe primacyof humansociety as
opposedto riverineerosion.Thesemythologicalagwi are sometimessaidto rest on
the backsof male crocodilespirits.But thesecreaturesfloat atopthe femininesea.
Furthermore,one of my researchassistantsnotedthatin Tambunum,as throughout
all of PapuaNew Guinea,men fight over two things:landandwomen. Thus, he
confided,landis reallyfeminine.As I argueelsewhere(Silverman1996),the gender
of the EasternIatmulcosmos is androgynousin the sense conveyedby Strathern
(1988). Thisandrogynymixesmaleandfemaledispositionsin a mutuallycontestable
way so thatgenderandcosmologyas well as time areultimately
andcontradictory
paradoxicalratherthansolidary.
Genderedtemporalityis found in other PapuaNew Guineansocieties (e.g.,
Strathern1992:199-200)andeven in the West(Macey1987).The Foi havea sexual
axis (Weiner 1984). The male
division of space along an upstream-downstream
sourceof the river is associatedwith pearlshells which move in a downstreamor
femaledirectionduringmarriageexchangesfor women.In a somaticmetaphor,the
front of the body (mouth,eyes, ears) is masculineand west, whereasthe anus is
feminine and east. Foi temporalitymanifeststwo directions:upstreamtoward
masculinelife and sociality, and downstreamtowardfemininedeath.Muchas for
EasternIatmul,Foi time is embodiedtime (Weiner1995:56).
AUSTRALIA
AFFINITIESWITHABORIGINAL
Many characteristicsof EasternIatmultotemism resembletemporalityand
(pre)historyin the dreamtimecosmologies of AustralianAboriginalcultures.
Dreamtimerefers to a prehumanyet atemporalera in which ancestorspirits,
andotherwise,createdthecosmosby shapingthe landscapeduring
anthropomorphic
migrationsandformulatingthe moralrulesthatgovernsocial life. On thesepoints,
the dreamtimeresemblesEasternIatmulcosmogony.Dreamtimetracksor zstrings"
(Rose 1992:52)correspondto the totemicpathsof mythichistory.
boththedreamtimeandEastern
In theiremphasison ancestralspatiotemporality,
Iatmultotemism(tsagi)privilegespatialratherthantemporalorder(Rose 1992:20707; Morphy 1995). Relationsin time are largelyunderstoodin terms of spatial
location.For the dreamtime,however,the only temporalco-ordinatesfromwhich
one could deElnea before or after are major disjunctionsXsuch as the sudden
appearanceof ancestors. Between zdisjunctions,"Rose (1992:106) claims,

This content downloaded from 129.64.99.140 on Mon, 6 May 2013 11:21:07 AM


All use subject to JSTOR Terms and Conditions

POLITICS,GENDER,AND TIMEIN MELANESIA

115

zsynchronyprevails-; a conceptthat is not apparentin the temporalthoughtof


EasternIatmul.
Both EasternIatmuland AustralianAboriginesmemorializethe past through
myth,narrative,andnames.Ritualcommemorates
thesememoriesandrecreatesthe
past,especiallyancestralmovementthroughandcreationof the landscapeandvarious
spatiallyanchoredevents(e.g., Morphy1995:201-03).Peoplein bothregionsbelieve
thatritualinsuresnatural,cosmic, and spiritual renewal"(Berndt1984:198-201).
The dreamtimeandtsagi areassociatedwiththe creativeor generativepowerof
the earlycosmos. Theyconcernbothznoumenal"andzphenomenal"
reality(Myers
1986:49).ButAustralianAborigineshavea moreprofoundsenseof reverencetoward
the manifestationsof dreamtimeancestorsthantheirSepikRivercounterparts.For
EasternIatmul,topographicfeaturesare importantlinksto the past only insofaras
theyhavetotemicnames,whichthemselvesembodyancestralpower.For Australian
Aborigines,the landscapeitself has spiritual charisma"(Berndt1984:177;Rose
1992:46). Since the distantpast for EasternIatmulis signiE1ed
first by namesand
then only secondarilyby the landscape,8
the dreamtimeis more zgrounded"(Rose
1992:57)thanmythichistory.Perhapsthis explainswhy the past is so contestedin
the Sepik River: names are more diffuse than topographicalfeatures.Moreover,
AustralianAboriginesseem to understandthe dreamtimeto be fundamentally
unalterableby humansfor deeplyrooted,almostexistential,reasonsratherthanfor
overtlypoliticalconcerns.
Aboriginesbelieve thatan indestructibleessenceor life force" links a person
with a totem and spirit-child(Stanner1965:232;Tonkinson1978). EasternIatmul
also identiifywith totemic names and their referents.Accordingto linear and
incremental
time, the ancestorslivedin thepast.However,theyarealso activein the
presentthroughthe actionsof living namesakes(as in Tambunum)or peoplewho
identifywith ancestralspaces(as in the dreamtime).Indeed,dreamtimespiritsand
EasternIatmultotemic crocodilesare said to influenceif not determinehuman
pregnancy(see above;Tonkinson1978;Myers 1986:50).Thusthe ancestralpower
of the past in bothregionsremainsa vital force in the worldtoday.Indeed,Layton
(1995:229) suggests that we comparethe landscapein Aboriginalthoughtto an
zabacuson whichpeoplearebeadswhose 'value'(culturalidentity)is determinedby
theirpositionon . . . a boardwhosestructureis in turndeterminedby the placesand
journeysXof the dreamtime.Furthermore,AustralianAboriginalcultures, like
EasternIatmul,compresstime duringritual(Munn1969:199-200).They also tend
to condenseEuropeanreligiousmythologyandtheir own cosmologyinto a unified
temporalscheme(Rose 1992:208),whereasEasternIatmul,we have seen, tend to
envisionparalleltimes.
Dreamtimeancestorsseemto havecreatedtheworldin sucha waythata speciElc
social group is associatedwith a naturalspecies (Bateson1936:244-45;Maddock
1972:28; Berndt 1984:177). This form of cosmologicalessentialismis alien to
EasternIatmul,who believein a pluralisticpastof parallelpathsratherthana single
unifiedpast. Lots of ancestorscreatedgroves of sago, but nonecreatedthe species

This content downloaded from 129.64.99.140 on Mon, 6 May 2013 11:21:07 AM


All use subject to JSTOR Terms and Conditions

116

ETHNOLOGY

in a single act. Therearealwaysotherancestors,fromothergroups,evenwithinthe


same village, who will have createdtheir own ancestralsago. This, too, tends to
resultin a greaterincidencein feudingover the pastthanin AboriginalAustralia.
Many anthropologistshave noted that the dreamtimeimplies a conservative
worldview:
['Islhestructure of the world and life was Elxed once-for-all at a remote time in the past....
possibilities for men's life were determined. (Stanner 1963:253-54)

The

Maddock(1972:129) echoes this view: zThere is among Aboriginesa profound


resistanceto creditingthemselveswith their own culturalachievements....
Aboriginesclaim credit only for f1delityto tradition."Aboriginesbelieve in two
cosmic eras: the atemporaldreamtimeandthepresent(Munn1970:144;Williams
1986:49-51).Thedreamtimeitselflies outsidethebracketsof incrementalandlinear
temporalty.It is a zsingle, unchanging,timeless sourceX(Myers 1986:52) that
surroundsor exists behindor below the presentera of time per se and humanity
(Rose 1992:205).It is, as Stanner(1956:52)so aptlyputsit, zeverywhen."Although
era in the Sepik,
mythic-historicor ancestralpoweralso suffusesthe contemporary
this belief is salientmainlyduringritualratherthaneverydayactivities.
BothEasternIatmulanddreamtimecosmologiesareopaqueaboutfirst causesX
(Stanner1963:266).Butthe Aboriginesbelievein a profoundsenseof cosmological
stasis. Priorto ancestralcreation,the dreamtimecontainedwhatStanner(1960:11318) calls aexistence-classes.Ancestorsmerelyplacedentitiesintothesepre-existing
categories.For EasternIatmul,by contrast,mythic-historicancestorsand spirits
createdall cosmic categoriesthroughnaming.
The timelessera of the dreamtimeconsistedof a featurelesslandscapeuntilthe
andcreativemigrationsof ancestors.Afterthe ancestorscreatedthe land
appearance
the dreamtimefrom
andLaw,therewas a dramaticcosmologicalbreakthatseparated
the present-dayera of historywhereinhumansdiscoverdreamtimecreationsand
follow the Law. EasternIatmulalso believe in a pretemporalera; namely, the
primordialsea. It, too, was followed by a period of ancestralcreation and
migration the era of totemism.Butthe cosmologicalbreakbetweenmythichistory
and today, or ancestorsand humanity,is less distinctand disjunctthan it is in
Aboriginalcosmology. This may explainwhy EasternIatmuldo not assert the
Aboriginalcredothat zto be humanis to reproduceforms"(Maddock1972:129).
personages.
betweenpersonsandmythic-historical
Instead,theyassertidentifications
or
'historicity'
their
Accordingto Munn(1970:144), Aboriginesare afreedfrom
with dreamtimeobjects and landforms;objects that,
'mortality'by integrationX
becausethey exist in boththe pastandpresent,areno longeranchoredto a specific
with totemicphenomenaand
time. EasternIatmuldo likewisethroughidentification
culturesalso organize
Aboriginal
Many
system.
naming
the alternate-generation
andcyclical
alternate-generation
with
namesandkinshiptermsas well as subsections
structures(Meggitt1962; Levi-Strauss1969 [1949];BerndtandBerndt1970, chs.

This content downloaded from 129.64.99.140 on Mon, 6 May 2013 11:21:07 AM


All use subject to JSTOR Terms and Conditions

POLITICS,GENDER,AND TIMEIN MELANESIA

117

4-5; Rose 1992:74-79,111-12).But historicalor temporalcyclicity is rarelynoted


explicitlyin the literature(butsee Stanner1956:60;Williams1986:30).
Both the dreamtimeand EasternIatmultotemismunderstandcosmic creation
throughidioms of genderand reproduction(Munn1969, 1970; Hiatt 1975; Rose
1992:42;Morphy1995:197).However,anthropologists
have not for the most part
elucidatedexplicitlythe gendersof Aboriginaltimes. The MurinbataDreamtime,
Stanner(1963:270)poeticallywrites, acouldnot deny Chronosbut gave Adrasteia
the triumph.-Here, implicitly,masculinetime is an illusionof historicalstasisthat
ultimatelygives way to femininetemporalflux. Similarly,EasternIatmulsignify
dissolutionandchangewith a metaphorof femaleriverineerosion.Politically,men
in Tambunumtry to transcendtemporalityandasserta timelesscosmos in orderto
supportclaimsto totemicentitlements(Silverman1996:43-45).Cosmologically,they
do this in orderto assertmasculineprimacyin a worldof riverinefluidity.Likethe
dreamtime,EasternIatmultotemismzdeniesthe erosionsof time"(Myers1986:52).
But it does so in a widerdiscourseconcerningthe genderof the cosmos.
For Yarralincosmology,dyingis likenedto washing out"(Rose 1992:209).It
is akinto the seasonalrainsthatcleansethe maternalearth, erasing the marksof
ordinarytime . . . preventingthe accumulationof past events"(Rose 1992:217).
Ordinarylife, whichbeginsandends in a short,linear,andincremental
time frame,
is merelytemporalnoisewhencomparedto the eternityof the dreamtimeandits vast
cosmologicalcycles of floods (Rose 1992:217).9
CONCLUSIONS
Ina much-noted
yetcriticizedpassage,Levi-Strauss
(1966:233-34)boldlydivided
theworldinto ahotXandzvold"societies.Hotsocieties,suchas ourown, internalize
historyas the prime mover of socioculturaldevelopment.Cold societies, Eastern
Iatmuland AustralianAboriginesincluded,subsumehistory and time under allencompassing
classifications(Stanner1963:139-48).Accordingto Harrison(1990:7376), Levi-Straussfailed to distinguishbetweenideology and practice.Hence the
coldnessof culture,one couldsay, is opposedby the hotnessof societyandpolitics.
But Levi-Strausswas also awarethatsome so-calledcold societieshavemultiple
formsof time. This understanding,
ratherthanhis hot-colddistinction,is the most
insightfulplace from which to analyzetime and history in the Sepik River and
beyond.It is not the case thatpeople simplylive their lives withina temporaland
historicalframethatadmitsor denieschange.Rather,peopleact in meaningfuland
strategicways; in short,peoplehaveagency.Furthermore,
cultureis not a seamless
web of signif1cations:
it containsparadoxesor problems.The pursuitof solutionsto
these predicamentsis a source of the passion of culture. Time and history are
importantfor anthropological
analysisnot simplyas structures
thatgovernsociallife.
Rather,they are symbolicresourcesandsemioticsystems.
I arguedthat EasternIatmulhave differentforms of time and historythat are
relatedto distinctarenasof social life. Withinthe realmof totemism,there is no

This content downloaded from 129.64.99.140 on Mon, 6 May 2013 11:21:07 AM


All use subject to JSTOR Terms and Conditions

118

ETHNOLOGY

neutralhistory. The presentcannotbe explainedon the basis of an apoliticaland


flxed sequence of mythic-historicevents. Instead, the contemporarypolitical
conElguration
of the village validatesthe totemic and cosmic organizationof the
universe.Historyandtime aresymbolicresourcesusedby agentswho constructand
altertheir world. Yet at the sametime actionis constrainedwithinthe normative
temporalframes of institutionssuch as totemismand kinship. Finally, time and
historyare a semioticsystemthroughwhich EasternIatmulconfrontfundamental
culturalandmoralparadoxesconcerninggender,stasis, andchange.
NOTES
1. Fieldwork in 1988-1990 was funded by a Fulbright Award and the Institute for Intercultural
Studies. A visit in June-August1994 was supportedby the Wenner-GrenPoundationfor Anthropological Research and DePauw University. I am gratefulfor this assistance. Previous versions of this article
were presented at the 1992 and 1993 Annual Meetings of the Association for Social Anthropology in
Oceania. The reference to Heraclitus was offered by G. Ogan, who, along with D Lipset, provided
additional comments. I also thank A. Wohl. But I accept full responsibility for this article.
2. Sometimes latmul will erect stones, which are rare in the Sepik floodplain, to commemorate
historic events such as successful raids on rival villages (e.g., Bateson 1970 [1937]:133; Silverman, In
press; Jorgensen 1990; Kahn 1990).
3. A similar concept is reportedby Lewis (1980:61-62) for song cycles among the Gnau of the West
Sepik.
4. Ecological and seasonal modes of time are common in Melanesia (e.g., Panoff 1969; Scaglion and
Condon 1979; Burman 1981). The common anthropologicalpractice of representingthis form of time
with diagrams is criticized by Goody (1977: ch. 4) and Bourdieu (1977:105-07).
5. This is similar to the concept of a "chronotype"(Bender and Wellbery 1989), which is derived
from Bakhtin's (1981 [19371) "chronotope."
6. Fake myth, like Dundes's (1989) ';fakelore," can be attributedto an identiElablesource and differs,
in the case described here, from Hobsbawm and Ranger's (1983) "invented tradition," which uses
images of historical permanence as ideological justifications for social institutions.
7. Leach (1961), Gell (1975:329-46, 1992: ch. 5), Bloch (1977), Harrison (1982), and Werbner
(1989), among others, discuss the concept of ritual time.
8. As a result, perhaps, EasternIatmulcosmology lacks a sense of ecological unity between humanity
and nature that is associated with the dreamtime. Indeed, recent accounts of Aboriginal cultures often
conclude with a moral plea for Europeans to heed the ecological wisdom of Aborigines (e.g., Rose
1992). Nothing in Eastern Iatmulcosmology, to my knowledge, would engender such a plea, however
much I might wish to do so in this era of unrestrainedlogging.
9. Although some readers might be dissatisfied with my generalizing the Aboriginal case and then
comparing it to Melanesia, this exercise was able to suggest the presence of gendered idioms of time
in both regions, a suggestion that has not been offered or demonstrateduntil now.
BIBLIOGRAPHY
Ackerman, C. 1976 Omaha and "Omaha."American Ethnologist 3:555-72.
Bakhtin, M. 1981 (1937). Forms of Time and the Chronotopein the Novel. The Dialogic Imagination,
ed. M. Holquist, pp. 84-158. Austin.
Barnes, R. H. 1976. Dispersed Alliance and the Prohibition of Marriage: Reconsideration of
McKinley's Explanationof Crow-OmahaTerminologies. Man (n.s.) 11:384-99.
Bateson, G. 1932. Social Structureof the Iatmul People of Sepik River. Oceania 2:245-91, 401-53.

This content downloaded from 129.64.99.140 on Mon, 6 May 2013 11:21:07 AM


All use subject to JSTOR Terms and Conditions

POLITICS,GENDER,AND TIMEIN MELANESIA

119

1936. Naven:A Surveyof the ProblemsSuggestedby a CompositePictureof the Cultureof


a New GuineaTribeDrawnfromThreePointsof View. Cambridge.
1970 (1937).An OldTempleanda New Myth.Traditional
BalineseCulture,ed. J. Belo, pp.
111-36. New York.
Bender,J., and D. E. Wellbery(eds.). 1989. Chronotypes:
The Construction
of Time. Stanford.
Berndt,R. M. 1984. TraditionalMoralityas Expressedthroughthe Mediumof an Australian
AboriginalReligion.Religionin AboriginalAustralia,eds. M. Charlesworth
et al. pp. 175211. St Lucia.
Berndt,R. M., and C. H. Berndt.1970. Man, Landand Mythin NorthAustralia:The Gunwinggu
People. EastLansing.
Bloch, M. 1977^The Pastandthe Presentin the Present.Man(n.s.) 12:278-92.
Bohannan,P. 1953. Conceptsof Time amongthe Tiv of Nigeria. The SouthwesternJournaloi
Anthropology9:251-62.
Bourdieu,P. 1977. Outlineof a Theoryof Practice.Cambridge.
Burman,R. 1981. TimeandSocioeconomicChangeon Simbo,SolomonIslands.Man(n.s.) 16:251-67.
Damon,F. H. 1990. FromMuyuwto the Trobriands:
Transformations
alongthe NorthernSide of the
KulaRing. Tucson.
Douglas,M. 1966. PurityandDanger:An Analysisof the Conceptsof PollutionandTaboo.London.
Dundes,A. 1989. TheFabrication
of Fakelore.PolkloreMatters,ed. A Dundes,pp. 40-56. Knoxville.
Eliade,M. 1958. Patternsin Comparative
Religion.New York.
Evans-Pritchard,
E. E. 1940. The Nuer. Oxiord.
FabianSJ. 1983. Timeandthe Other:How Anthropology
MakesIts Object.New York.
Gell, A. 1975. Metamorphosis
oi the Cassowaries:UmedaSociety,Languageand Ritual.London.
1992. The Anthropologyoi Time:CulturalConstructions
of TemporalMaps and Images.
Oxford.
Goody,J. 1977. LiteracyandClassification:
OnTurningthe Tables.The Domestication
of the Savage
Mind,ed. J. Goody,pp. 52-73. Cambridge.
Harrison,S. 1982. Yamsandthe SymbolicRepresentation
of Timein a SepikRiverVillage. Oceania
53:141-61.
1984. New GuineaHighlandSocialStructurein a LowlandTotemicMythology.Man(n.s.)
19:389-403.
1990.StealingPeople'sNames:HistoryandPoliticsin a SepikRiverCosmology.Cambridge.
Hauser-Schaublin,
B. 1983. The Mai-Masksof the Iatmul,PapuaNew Guinea.OralHistory11:1-53.
Hiatt, L. 1975. Swallowingand Regurgitation
in AustralianMyth and Rite. AustralianAboriginal
Mythology:Essaysin Honourof W. E. H. Stannersed. L. Hiatt,pp. 1-23. Canberra.
Hobsbawm,E., and T. Ranger.1983. Introduction:
InventingTradition.The Inventionof Tradition,
eds. E. HobsbawmandT. Ranger,pp. 1-14. Cambridge.
Jorgensen,D. 1990. Placingthe Pastand Movingthe Present:Mythand Contemporary
Historyin
Telefolmin.Culture10:47-56.
Kahn,M. 1990. Stone-FacedAncestors:The SpatialAnchoringof Myth in Wamira,PapuaNew
Guinea.Ethnology29:51-66.
Layton,R. 1995. Relatingto the Countryin the WesternDesert. The Anthropologyof Landscape:
Perspectiveson PlaceandSpace,eds. E. Hirschand M. O'Hanlonpp. 210-31. Oxford.
Leach, E. R. 1961. Two Essays Concerningthe SymbolicRepresentationof Time. Rethinking
Anthropology,ed. E. R. Leach,pp. 124-36. London.
Levi-Strauss,C. 1963. StructuralAnthropology.New York.
1966. The SavageMind.Chicago.
1969 (1949). The ElementaryStructuresof Kinship.Boston.
Lewis, G. 1980. Day of ShiningRed:An Essayon Understanding
Ritual.Cambri;lge.
Macey,S. L. 1987. Patriarchs
of Time:Dualismin Saturn-Cronus,
FatherTime,theWatchmaker
God,
and FatherChristmas.Athens.

This content downloaded from 129.64.99.140 on Mon, 6 May 2013 11:21:07 AM


All use subject to JSTOR Terms and Conditions

120

ETHNOLOGY

Maddock, K. 1972. The AustralianAborigines: A Portraitof Their Society. London.


McDowell, N. 1985. Past and Puture:The Nature of Episodic Time in Bun. History and Ethnohistory
in Papua New Guinea, eds. D. Gewertz and E. Schieffelin, pp. 26-39. Sydney.
McKinley, R. 1971a. A Critiqueof the ReflectionistTheory of KinshipTerminology:The Crow/Omaha
Case. Man 6:228-47.
1971b. Why Do Crow and OmahaKinshipTerminologies}ixist?A Sociological Interpretation.
Man 6:408-26.
Meggitt, M. J. 1962. Desert People: A Study of the Walbiri Aborigines of CentralAustralia. Chicago.
Morphy, H. 1995. Landscape and the Reproductionof the Ancestral Past. The Anthropology of the
Landscape: Perspectives on Place and Space, eds. E. Hirsch and M. O'Hanlon, pp. 184-209.
Oxford.
Munn, N. 1969. The Effectiveness of Symbols in MurnginRite and Myth. Forms of Symbolic Action,
ed. R. Fo Spencer, pp. 178-207. Seattle.
1970. The Transformation of Subjects into Objects in Walbiri and PitjantjatjaraMyth.
Australian Aboriginal Anthropology, ed. R. M. Berndt, pp. 141-63. Nedlands.
1977. The Spatiotemporal Transformation of Gawa Canoes. Journal de la Societe des
Oceanistes 33:39-51 .
1986. The Fame of Gawa: A Symbolic Study of Value Transformationin a Massim (Papua
New Guinea) Society. Durham.
1992. The CulturalAnthropologyof Time: A CriticalEssay. Annual Review of Anthropology
21 :93-123.
Myers, F. R. 1986. PintupiCountry, PintupiSelf: Sentiment,Place, and Politics among Western Desert
Aborigines. Berkeley.
Panoff, M. 1969. The Notion of Time among the Maenge People of New Britain. Ethnology 8:153-66.
Pompomo, A. 1992. Seagulls Don't Fly into the Bush: CulturalIdentityand Development in Melanesia.
Belmont.
Rose, D. B. 1992. Dingo Makes Us Human: Life and Land in an Aboriginal Australian Culture.
Cambridge.

Sahlins, M. 1985. Islands of History. Chicago.


Scaglion, R. and R. G. Condon. 1979. Abelam Yam Beliefs and Socio-Rhythmicity: A Study in
Chrono-Anthropology.Journal of Biosocial Science 11-17-25.
Schuster, G. 1990. Aibom Genealogy as History. Sepik Heritage: Traditionand Change in Papua New
Guinea, eds. N. Lutkehaus, C. KaufmannsW. Mitchell, D. Newton, L. Osmundsen, and M.
Schuster, pp. 20-22. Durham.
Schuster, M. 1985. The Men's House, Centre and Nodal Point of Art on the Middle Sepik. Art of the
Sepik River, ed. S. Greub, pp. 19-26. Basel.
1990. Aspects of the Aibom Concept of History. Sepik Heritage Traditionand Change in
Papua New Guinea, ed. N. Lutkehaus C. Kaufmann, W. Mitchell, D. Newton, L.
Osmundsen, and M. Schuster, pp. 7-19. Durharn.
Silverman, E. K. 1996. The Gender of the Cosmos: Totemism, Society and Embodimentin the Sepik
River. Oceania 67:30-49.
In press. Traditional Cartography in Papua New Guinea and Melanesia. The History of
Cartography,Vol. 2.3: TraditionalCartographyin African, American, Arctic, Australian,and
Pacific Societies, eds. D. Woodward and G. M. Lewis. Chicago.
Smith, M. F. 1982. Bloody Time and Bloody Scarcity: Capitalism,Authority, and the Tramv,formation
of Temporal Experience in a Papua New Guinea Village. American Ethnologist 9:503-18.
1994. Hard Times on KairiruIsland: Poverty, Development, and Morality in a Papua New
Guinea Village. Honolulu.
Stanner, W. E. H. 1956. The Dreaming. AustralianSignpost: An Anthology Edited for the Canberra
Fellowship of AustralianWriters, ed. T. Hunderford,pp. 51-65. Melbourne.
1960. On Aboriginal Religion. III. Symbolism in the Higher Rites. Oceania 31:100-20.

This content downloaded from 129.64.99.140 on Mon, 6 May 2013 11:21:07 AM


All use subject to JSTOR Terms and Conditions

POLITICS,GENDER,AND TIMEIN MELANESIA

121

1963. On Aboriginal Religion. VI. Cosmos and Society Made Correlative. Oceania 33:23973.
1965. Religion, Totemism and Symbolism. Aboriginal Man in Australia: Essays in Honour
of EmeritusProfessor A. P. Elkin, eds. R. M. Berndtand C. M. Berndt, pp. 207-37. Sydney.
Strathern, M. 1988. The Gender of the Gift: Problems with Women and Problems with Society in
Melanesia. Berkeley.
1992. The Mother's Brother's Child. Shooting the Sun: Ritual and Meaning in West Sepik,
ed. B. Juillerat, pp. 191-205. Washington DC.
Swadling, P. 1989. A Late Quaternary Inland Sea and Early Pottery in Papua New Guinea.
Archaeology in Oceania 24:106-09.
Tonkinson, R. 1978. Semen versus Spirit-Childin a Western Desert Culture. Australian Aboriginal
Concepts, ed. L. R. Hiatt, pp. 81-91. Canberra.
Tuzin, D. 1977. Reflections of Being in Arapesh Water Symbolism. Ethos 5:195-223.
1992. Sago Subsistence and Symbolism among the Ilahita Arapesh. Ethnology 31: 103-14.
Wagner, R. 1986. Symbols that Stand for Themselves. Chicago.
Wassmann, J. 1990. The Nyaura Concepts of Space and Time. Sepik Heritage: Traditionand Change
in Papua New Guinea, eds. N. Lutkehaus, C. Kaufmann, W. Mitchell, D. Newton, L.
Osmundsen, and M. Schuster, pp. 23-35. Durham.
1991. The Song to the Flying Fox: The Public and Esoteric Knowledge of the ImportantMen
of Kandigei about Totemic Songs, Names and Knotted Chords (Middle Sepik, Papua New
Guinea). Boroko PNG.
Weiner, J. F. 1984. Sunset and Flowers: The Sexual Dimension of Foi Spatial Organization. Journal
of Anthropological Research 40:577-88.
1995. The Lost Drum: The Myth of Sexuality in Papua New Guinea and Beyond. Madison.
Werbner, R. P. 1989. Ritual Passage, Sacred Journey: The Process and Organization of Religious
Movement. Washington DC.
Williams, N. M. 1986. The Yolngu and Their Land: A System of Land Tenure and the Fight for Its
Recognition. Stanford.
Williamson, M. H. 1980. Omaha Terminology and Unilateral Marriage on the Sepik. American
Ethnologist 7:530-48.

This content downloaded from 129.64.99.140 on Mon, 6 May 2013 11:21:07 AM


All use subject to JSTOR Terms and Conditions