Kwoma Death
Kwoma Death
Kwoma Death
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.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms
This content downloaded from 1.136.96.22 on Mon, 21 Aug 2017 01:35:41 UTC
All use subject to http://about.jstor.org/terms
KWOMA DEATH PAYMENTS AND
ALLIANCE THEORY
Ross Bowden
La Trobe University
271
This content downloaded from 1.136.96.22 on Mon, 21 Aug 2017 01:35:41 UTC
All use subject to http://about.jstor.org/terms
272 ETHNOLOGY
KWOMA SOCIETY
This content downloaded from 1.136.96.22 on Mon, 21 Aug 2017 01:35:41 UTC
All use subject to http://about.jstor.org/terms
KWOMA DEATH PAYMENTS AND ALLIANCE THEORY 273
This content downloaded from 1.136.96.22 on Mon, 21 Aug 2017 01:35:41 UTC
All use subject to http://about.jstor.org/terms
274 ETHNOLOGY
WIFE-TAKING LINE
WIFE-GIVING LINE
A==a
OBLIGATORY
" A WEALTi
^
RECIPROCAL
SERVICES
111 A <?
FOOD
OPTIONAL
IV A A
NO FURTHER CONNECTION
This content downloaded from 1.136.96.22 on Mon, 21 Aug 2017 01:35:41 UTC
All use subject to http://about.jstor.org/terms
KWOMA DEATH PAYMENTS AND ALLIANCE THEORY 275
This content downloaded from 1.136.96.22 on Mon, 21 Aug 2017 01:35:41 UTC
All use subject to http://about.jstor.org/terms
276 ETHNOLOGY
AA
CA
D ? (AND D WHILE UNMARRIED)'
This content downloaded from 1.136.96.22 on Mon, 21 Aug 2017 01:35:41 UTC
All use subject to http://about.jstor.org/terms
KWOMA DEATH PAYMENTS AND ALLIANCE THEORY 277
This content downloaded from 1.136.96.22 on Mon, 21 Aug 2017 01:35:41 UTC
All use subject to http://about.jstor.org/terms
278 ETHNOLOGY
This content downloaded from 1.136.96.22 on Mon, 21 Aug 2017 01:35:41 UTC
All use subject to http://about.jstor.org/terms
KWOMA DEATH PAYMENTS AND ALLIANCE THEORY 279
DISCUSSION
r "i
DECEASED'S MB
A
(RECIPIENT OF
THE MASIIK)
(FOR C)
This content downloaded from 1.136.96.22 on Mon, 21 Aug 2017 01:35:41 UTC
All use subject to http://about.jstor.org/terms
280 ETHNOLOGY
This content downloaded from 1.136.96.22 on Mon, 21 Aug 2017 01:35:41 UTC
All use subject to http://about.jstor.org/terms
KWOMA DEATH PAYMENTS AND ALLIANCE THEORY 281
RECIPIENT OF
MASI1K
7_*
//
k$
RECIPIENT Of A
SOBATAKEP
= CT ~2*
/ /
RECIPIENT OF A
SOBATAKEP
= G /W
PRIMARY DONORS
/ OTHER DONORS
1-he'house'
The third type of payment made on the occasion of every death is akakep
(payment to the house). The house (aka) in this context refers to the
combined households of the persons for whom sobatakep payments are
made; i.e., the deceased's household and closely-related coresident households
in the same clan. The akakep is actually composed of several separate
payments made independently by the men in other clans who are formally
implicated as members of wife-taking lines in affinal exchange relationships
with members of the "house." That is, they are made by the husbands and
sons of outmarried natal female members of the deceased's household and
closely related coresident households (such as the deceased's Z or FZ). Figure
4 is a schematic representation of potential donors of wealth to an akakep
following the death of a man. In this case, the contributors include the sons
of the deceased's Z, FZ and FFZ, and the husbands of the deceased's (living)
Z and FZ. Although the figure indicates that the mothers' brothers of ego's
This content downloaded from 1.136.96.22 on Mon, 21 Aug 2017 01:35:41 UTC
All use subject to http://about.jstor.org/terms
282 ETHNOLOGY
FZS and FFZS (i.e., ego's F and FF respectively) have predeceased ego, both
men (the FZS and FFZS) are still actively involved in formal exchange
relationships with male members of ego's line, since the duration of these
exchange relationships is determined not by how long individual members of
ego's line live but by how long the FZS and FFZS live. In the case illustrated,
their common exchange partner (including that of ego's ZS following ego's
death) is ego's son, the senior surviving male member of the various mothers'
brothers' line. Similarly, deceased ego's ZH and FZH continue to be actively
involved in formal exchange relationships with (one or more) members of
ego's line, since the duration of these exchange relationships is determined by
how long their wives live, not by how long any individual male member of
ego's line lives. In the case illustrated, their common exchange partner
(following ego's death) is again ego's son.
Like the masiik and sobatakep payments, the separate donations of
wealth that compose the akakep form part of the wider cross-generational
exchange relationships marriages establish between the members of affinally
related lines, for they are all made by members of wife-taking lines to
members of wife-giving lines. In contrast to the masiik and sobatakep,
however, which are made by members of the deceased's "house" to members
of wife-giving lines in other clans, the akakep is made collectively by
members of wife-taking lines in other clans to members of the deceased's
"house."
In addition to the persons identified above, donors to an akakep usually
also include the recipients of the masiik and various sobatakep payments
made on the same occasion (see Figure 4). Kwoma emphasise that the
recipients of these payments are under no formal obligation to contribute to
the akakep, being members of wife-giving lines vis-a-vis the recipients of
the akakep, but do so out of sympathy for the members of the "house" for
having had to lay out so much wealth in order to make the masiik and
various sobatakep payments. Such informal (nonobligatory) contributions to
the akakep represent the only occasion in Kwoma society when prestations of
wealth are made by members of wife-giving lines to members of wife-taking
lines. Normally, wealth flows exclusively in the opposite direction, to wife-
givers.
The masiik, sobatakep and akakep are the three kinds of payments made
obligatorily following every death in Kwoma society. Figures 2-4 illustrate
the donors and recipients of these payments following the death of a man,
but I have indicated that the same range of payments is made following the
death of a woman or unmarried girl. With reference to Figure 3, for
instance, if the deceased had been ego's wife (C) rather than ego himself, the
same range of sobatakep payments are made, except that none go to the
deceased woman's brother, since he is now the recipient of the deceased's
masiik; and a sobatakep is made for the woman's (surviving) husband to his
MB, since the latter is no longer the recipient of the masiik. Similarly, if the
deceased in Figure 3 had been ego's unmarried daughter (E), the same range
of sobatakep is made, except that no payment is made for the girl's surviving
mother to the latter's brother, since this man now becomes the recipient of
the deceased girl's masiik; and a sobatakep is made in place of the masiik
illustrated for the girl's surviving father, to his MB.
This content downloaded from 1.136.96.22 on Mon, 21 Aug 2017 01:35:41 UTC
All use subject to http://about.jstor.org/terms
KWOMA DEATH PAYMENTS AND ALLIANCE THEORY 283
B AND C
A
B
A A
Daribi
??o A
This content downloaded from 1.136.96.22 on Mon, 21 Aug 2017 01:35:41 UTC
All use subject to http://about.jstor.org/terms
284 ETHNOLOGY
This content downloaded from 1.136.96.22 on Mon, 21 Aug 2017 01:35:41 UTC
All use subject to http://about.jstor.org/terms
KWOMA DEATH PAYMENTS AND ALLIANCE THEORY 285
rights in married sisters and their children are ultimately vested not in
individuals but in clans as wholes.
Brown (1961) also rejects an alliance model of Chimbu death payments
because the recipients must have actively assisted the deceased in various
ways during his or her lifetime. The role of helper or assistant, she believes,
is incompatible with their status as affines. But in Kwoma society, members
of wife-giving lines are by definition and of necessity helpers and nurturers
vis-a-vis members of wife-taking lines. I have indicated that the former lose
exchange rights in the latter if they fail actively to help and assist them in
various ways, and their role as helpers and nurturers is also implicit in the
use of the metaphor "mother warchil plant" for wife-giving lines. The fact
that the recipients of death payments must have actively assisted the persons
for whom they are made does not necessarily entail, therefore, that they
cannot be conceptualized as affines.
A second Papua New Guinea society that possesses enduring affinal
alliances and death payments similar in structure to those of the Kwoma is
the Daribi (Wagner 1967; see Figure 5). Wagner indicates that a Daribi
marriage establishes an affinal alliance between a wife-giving and wife-
taking patriline that endures for three generations. As in the case of the
Kwoma and Chimbu, additional marriages may not be contracted between the
same two lines for the duration of the alliance. This entails among other
things that marriage is prohibited between cross-cousins and the children of
cross-cousins (Wagner 1967:117, 127-8). It is only in the third descending
generation, when the relationship between the two lines has become "too
distant, and must be renewed if the alliance is to continue" (Wagner 1967:127)
that another marriage may take place. If one does occur at this level, ideally
it should be contracted in the opposite direction from the earlier one to
balance the exchange of women between the two lines (Wagner 1967:143).
The death payment for a married woman normatively goes to her brother's
line and the death payments for a man and unmarried girl go to their MB's
line (Wagner 1967:69,72).
In contrast to my alliance interpretation of Kwoma death payments,
Wagner (1967) interprets their Daribi equivalents, and the other payments
that are customarily made to members of wife-giving lines at various stages
in an individual's life-cycle, as deriving from the genealogically-based
consanguineal ties that connect intermarrying lines. The exchange relationship
between a MB and ZS, in his view, derives not from an enduring affinal
alliance between the two men's patrilines but from the actual consanguineal
connection between them (Wagner 1967:69, 75; cf. Strathern 1968:451). In
ways that are strongly reminiscent of the writings of Meyer Fortes and other
descent theorists in the 1950s, Wagner even argues that the affinal
relationship between brothers-in-law derives not from the marriage that
connects them but from an anticipated, indirect consanguineal tie via the
children that might result from the marriage (Wagner 1967:76).
Wagner's discussion of death payments focusses on the mother's brother-
sister's child relationship, and in particular on the MB-ZS relationship. A
Daribi child, he argues, is thought to be related consanguineally both to its
mother's and father's clans. The consanguineal tie to the mother's clan,
however, is believed to be stronger than that to the paternal clan, with the
This content downloaded from 1.136.96.22 on Mon, 21 Aug 2017 01:35:41 UTC
All use subject to http://about.jstor.org/terms
286 ETHNOLOGY
This content downloaded from 1.136.96.22 on Mon, 21 Aug 2017 01:35:41 UTC
All use subject to http://about.jstor.org/terms
KWOMA DEATH PAYMENTS AND ALLIANCE THEORY 287
This content downloaded from 1.136.96.22 on Mon, 21 Aug 2017 01:35:41 UTC
All use subject to http://about.jstor.org/terms
288 ETHNOLOGY
CONCLUSION
This paper has two main objectives. The first is to give a description and
interpretation of the structure of Kwoma death payments. In common with
many other Sepik societies Kwoma lack competitive exchange relationships,
and death is the principal occasion on which wealth is redistributed between
individuals and groups. Death payments similar to those of the Kwoma are
found in many Papua New Guinea societies, and I suggest that these
similarities point to some very widespread, but largely neglected,
resemblances in the social structure of societies in this region.
The second aim is to demonstrate the utility of an explicit alliance model
of marriage for the explication of the structure both of Kwoma death
payments and the wider affinal exchange relationships between groups in
which they are embedded. Hitherto, alliance models have been thought to
be applicable only to certain types of societies, notably those in which
positive rules entail the regular exchange of spouses, symmetrically or
asymmetrically, between relatively stable wife-giving and wife-taking groups.
Like most other Papua New Guinea societies, Kwoma lack positive marriage
rules, and marriage is actually prohibited with members of a wide range of
descent lines outside one's own exogamous clan. Papua New Guinea societies,
that is, typically have "Crow/Omaha" rather than "elementary" marriage
systems. Notwithstanding the absence of positive rules, individual Kwoma
marriages can be seen to establish alliances between affinally related lines
that potentially endure for several generations. Death payments, I have
shown, form part of the obligatory asymmetrical exchange relationships that
define these alliances. Since individual marriages in many other Papua New
Guinea societies establish multi-generation affinal alliances similar to those of
the Kwoma, explicit alliance models are suitable to the analysis of these
societies as well. Of more general relevance for kinship theory, the range of
societies to which alliance models apply needs to be widened substantially
from the "elementary" systems that have been the focus of interest for
alliance theorists to include many "Crow/Omaha" societies as well.
NOTES
1. In common with the majority of neighboring Sepik peoples, Kwoma lack com
institutions and men do not (at least overtly) use death payments or other affin
compete for prestige; e.g., by maximizing their size. Rather, men endeavor in all af
to give the minimum that is acceptable so as to conserve as much of their wealth
numerous other obligatory affinal transactions in which they and their clansmen ar
2. A person's "classificatory" brothers are same-generation male members of other c
totemic division (Bowden 1983b).
3. Kwoma use the terms "pay" or "buy" (tokitow) exclusively for giving shell valua
currency) in exchange for something. Since wealth items in affinal alliances conve
wife-givers rather than to wife-takers, the latter "pay" or "buy" the former but not v
4. The exchange relationships that define Chimbu affinal alliances may continue
additional generations, but the exchanges at these levels, as among the Kwoma,
important than those in the first two generations (Brown 1961:91).
This content downloaded from 1.136.96.22 on Mon, 21 Aug 2017 01:35:41 UTC
All use subject to http://about.jstor.org/terms
KOWMA DEATH PAYMENTS AND ALLIANCE THEORY 289
5. Chimbu marriage prohibitions apply to close agnates, mother's sub-clan and other "traceable kin"
(Brown 1964:338), but not to the groups of either "grandmother." Marriage is therefore clearly
permissible with members of FMB's sub-clan; it is not clear whether marriage is permissible with a
member of FMB's line (e.g., FMBSD).
6. The likelihood that death payments in Chimbu society will be made to persons other than the
genealogically-defined relatives referred to above is increased by the fact that young people
commonly spend long periods living away from the natal sub-clans. In Kwoma society such changes
in residence are much less common.
7. That marriage may not be repeated between a wife-giving and wife-taking line for the duration of
a Daribi alliance needs to be qualified by the observation that men are apparently entitled to
contract several marriages simultaneously with members of the same wife-giving clan, including
members of the same wife-giving line (e.g., WZ, WFZ, and WBD) (Wagner 1977:630-633;
1978:116). But Wagner's accounts of this entitlement vary. In one place he (Wagner 1978:116)
states that while men see themselves as having this entitlement, their intended wives, and the
latter's clansmen, may not. Wagner (1977:632) also indicates that sister exchange occasionally
occurs, but that this is not condoned.
8. Wagner's statements about death payments for married women are inconsistent. He (Wagner
1967:69) reports that a married woman's brother is the recipient of her death payment, but
elsewhere (Wagner 1972:53) states that this is passed on to the woman's maternal kin. See
Strathern (1968; and 1971b) and de Lepervanche (1969) for further criticisms and comments on
Wagner's argument.
9. Why further recruitment payments should be made for a woman to her natal clan throughout her
marriage if a bridewealth payment effectively recruits her to her husband's group is one of many
matters that Wagner does not adequately explain. If such payments should be made to anyone they
should presumably be made to the woman's MB, since she is still consanguineally related to him.
But apart from the death payment, about which he is ambiguous, Wagner clearly indicates that this
does not happen.
BIBLIOGRAPHY
Bowden, R. 1983a. Kwoma Terminology and Marriage Alliance: The 'Omaha' Problem Revisited.
Man 18:745-65.
Bulmer, R. N. H. 1960. Political Aspects of the Moka Ceremonial Exchange System Among the
Kyaka. Oceania 31:1-13.
Dumont, L. 1953. The Dravidian Kinship Terminology as an Expression of Marriage. Man 53:34-9.
_1971. The Marriage Alliance. Kinship: Selected Readings, ed. J. Goody. London.
Feil, D. K. 1980. When a Group Takes a Wife: Generalised Exchange and Restricted Marriage in
the New Guinea Highlands. Mankind 12:286-299.
_1984. Ways of Exchange: the Enga Tee of Papua New Guinea. St. Lucia.
Forge, A. 1972. The Golden Fleece. Man 7:527-540.
Keesing, R. M. 1975. Kin Groups and Social Structure. New York.
Lepervanche, M. de. 1969. Review of Wagner The Curse of Souw. American Anthropologist
71:518-520.
This content downloaded from 1.136.96.22 on Mon, 21 Aug 2017 01:35:41 UTC
All use subject to http://about.jstor.org/terms
290 ETHNOLOGY
Levi-Strauss, C. 1966. The Future of Kinship Studies. Proceedings of the Royal Anthropological
Institute for 1965, 13-22.
_ 1969. The Elementary Structures of Kinship (trans. Bell, Sturmer, and Needham).
London.
Ryan, D'A. 1958-59. Clan Formation in the Mendi Valley. Oceania 29:257-89.
Salisbury, R. F. 1962. From Stone to Steel. Melbourne.
Strathern, A. 1968. Review of Wagner The Curse of Souw. Bulletin of the School of Oriental and
African Studies, University of London 31:652-3.
_1971a. The Rope of Moka. Cambridge.
_1971b. Wiru and Daribi Matrilateral Payments. Journal ofthe Polynesian Society 80:449-
62.
Wagner, R. 1967. The Curse of Souw: Principles of Clan Definition and Alliance. Chicago and
London.
_1972. Habu: the Innovation of Meaning in Daribi Religion. Chicago and London.
_1977. Analogic Kinship: a Daribi Example. American Ethnologist 4:623-42.
_1978. Lethal Speech: Daribi Myth as Symbolic Obviation. Ithaca and London.
Weiner, A. B. 1976. Women of Value, Men of Renown: New Perspectives in Trobriand Exchange.
Austin and London.
Whiting, J. W. M., and S. W. Reed. 1938-39. Kwoma Culture: Report on Field Work in the
Mandated Territory of New Guinea. Oceania 9:170-216.
Young, M. W. 1971. Fighting With Food. Cambridge.
This content downloaded from 1.136.96.22 on Mon, 21 Aug 2017 01:35:41 UTC
All use subject to http://about.jstor.org/terms