In return for gifts from the deceased's patri-group, men of the opposite moiety of a dead person would remove the hair from the corpse, make it into feathered ceremonial string, and return it to men of the deceased's patri-group or their affines either for eventual use in the Nga:rra ceremony, to be incorporated into sacred objects, or to be made into a hair-belt. One or two years later the hair-string would be returned to the close kinsmen of the deceased, eventually to be given to his son, if adult, in return for a large presentation to the string's maker of cycad-palm-nut bread, other vegetable foods, and spears. The string would later be cut up to make the feather cords (wana, 'arms') of men's sacred dilly-bags (bathi minydjalpi) (Peterson 1975: 98-9).
The gift of a part of a person's body is one way of initiating an exchange ceremony. The term marradjirri denotes a class of decorated strings and poles, each representing a patri-group totem (madayin) and classified into 'small' and 'large' marradjirri. Formerly the most common use of the small marradjirri was to muster people to mortuary and circumcision ceremonies (Peterson 1975: 104-5). This message function was replaced by the use of two-way radios in the 1980s, and, more recently, by solar-powered telephones. After playing a role in such ceremonies some small marradjirri were used for love magic; following the ceremony a man might loop the string over the hands of a sweetheart or errant wife, who was then impelled to follow him (Peterson 1975: 105). (8)
Marradjirri is also a category of ceremony in which men of one patri-group make and present a pole and an attached string (a large marradjirri) to those of another group, thus performatively establishing and maintaining social relationships with socially and geographically distant peoples (including with people of Canberra; see Wild 1986). The decorated pole and string represent a patri-group totem (madayin) which is presented to a commissioning or requesting group (the djugu'watangu). For example, the Djambarrpuyngu patri-groups possess the Morning Star (banumbirr) marradjirri ceremony, which includes a pole representing the Star. They also possess a small carved head of the Tern (djarrak) minor totem together with a long string, used in public circumcision and mortuary ceremonies (Warner 1937: 423-35).
One immediate trigger of the performance of a marradjirri ceremony is the occasion when an infant makes its first 'gift', that is, picks up an object, such as a pebble, blade of grass, shell, fruit, or small lizard, and gives it to a parent or other close relative. The object is tied into a bundle and sent to a socially and geographically distant relative, who incorporates the object into his own patri-group's marradjirri, later presenting it to the child's parents in a large public ceremony. The parents make a payment to the maker in return. At the end of the ceremony the string is removed from the core around which it was bound and incorporated into arm rings or men's sacred dilly-bags. A child's umbilical cord may also be sent as a gift to instigate an exchange ceremony.
Borsboom (1978) describes a marradjirri ceremony in which the hair of a living girl of the Ngalkbon language group (resident in the south of Arnhem Land) was sent to a Djinang (a western Yolngu language) man to commission a performance in which the hair would be returned. Marradjirri can also be performed as a mortuary rite (Warner 1937: 423-5) or a post-mortuary rite (West 1962). In the latter case the group of the deceased sends the dead man's hair to the performers, who return it in the ceremony, after which the hair is taken to distant brothers-in-law of the deceased, who will be responsible for using it in a madayin revelatory ceremony.
Marradjirri exchange is thus constructed around the intrinsic connection between a person and his or her hair or a child and its umbilical cord or first gift. These objects establish connections with their recipients of a kind that, as it were, attracts a gift back to the originator and his or her close kin (Gregory 1982; Mauss 1954; A. Weiner 1992). A parallel 'binding' of persons takes place in order to avenge a death.
In former times, before the imposition of colonial power, relatives of a dead person presented relics of the dead such as finger-bones, spears, or a hair-belt in order to incite revenge for the death among the younger men to whom they were given. For example, Warner (1937: 178) recounts how a man sent another a hair-belt incorporating the hair of a dead man. The recipient felt obliged to kill a relative of the person accused of killing the man, who was visiting his country. Here the instigation of revenge through the gift of hair was incorporated into the ancestral domain--the hair-belt was madayin.
Gifts of removable but intrinsically connected parts of persons, as well as ancestral doctrines and sorcery beliefs, thus entail the distribution of parts of persons across social and geographical space, spatializing relationships through the imputed powers of the body. It seems that different kinds of relations (patri-group relations, sorcery, gift exchange) are to an extent mediated through different kinds of bodily substance. However, Yolngu do not draw absolute boundaries between domains such as sorcery and the domain of ancestral beings. Sorcery is implicated in the protection of secrecy of ancestral rom ('law', complexes of rituals, songs, designs, and sacred objects), while ancestral powers may be drawn on to attack one's enemies.
I turn now to ways in which ancestral doctrines, sorcery, and gift exchange embed people in the moral-political community.
The moral-political community
Ancestral 'law'
The extension of ancestral persons in space and time forms the 'foundation' (luku, 'feet, footprint', as Yolngu put it) of the regional, communal moral order. How does the spatial distribution of ancestral bodies, powers, and traces relate to this order? Rom (in its sense of ancestral 'law') does not consist simply of a body of rules but of a combination a variety of things. One is explicit rules and precepts governing, for example, the proper way to treat relatives of various kinds or rights in a water hole. Another is the ideal forms of practices (such as rituals, paintings, and the subsection system) and their enactment. A third is the body of rights and relationships including patri-group holdings of land, waters, and sacra as 'inalienable property' (on rom see Morphy 1995; Creighton 2003; see A. Weiner 1992 on inalienable property).
The spatial location of ancestral traces and powers links groups of people to areas of land and waters, and ancestral designs and objects signify these connections. People live and move within a landscape imbued with social identities and relations, all grounded in the presence of ancestral traces. These traces of ancestral persons are, one might say, relational. Such and such a tree is Ba:riya's 'father' and 'father's father' (because the tree species is a totem of his own patri-group), Dja:wa's 'mother' and 'mother's father' (because it belongs to his mother's patri-group), another person's 'mother's mother' (because it belongs to the person's mother's mother's patri-group), and so on. The landscape can also, however, be filled with dangers if one strays into country to which one is not closely related and risks violating secret or restricted ancestral sites or encountering ghosts and mischievous ancestral beings (Biernoff 1978).
In contrast with the living community--transient and relatively mobile on (in former times) their daily and seasonal round of hunting, gathering, and fishing--the community of ancestors to whom a person is related according to the same categories of kin relations as between the living has been more or less bound to places and made permanent. (9) This ancestral community is the source of power, of the reproduction of life, identity, and of the proper order of things.
Relations of authority are in part grounded in terms of the totemic ancestors. Men and women move closer to the wangarr in their later years. Older men and women, the leaders of patri-groups, have control of access to ancestral sacra in the form of knowledge of designs, songs, and rituals; they are the authorities on the 'law' instituted by the ancestors. Younger people appeal to older people for authoritative judgements and, ideally, defer to them, although of course personal dispositions play a major part in relations of power and substance-abuse has now modified these relations at Yirrkala (Creighton 2003: 307-8). The most direct manifestation of authority lies in the subjugation of young men in circumcision initiation, and the disciplines undergone in revelatory ceremonies. In these ceremonies, initiands and other performers identify with the totemic ancestors to whom they are most closely related.
Sorcery and the moral-political community
Sorcery doctrines embed individuals in the moral-political community in a very different way, though drawing ultimately on the same kinds of imputed powers. As well as being more or less constantly in the presence of others in the daily round, everyone is embedded in networks of relations that transcend day-to-day proximity and which exist within a matrix of moral-political relations. Sorcery doctrines imply the ability of a person's neighbours, close or distant, to take secret and aggressive action at a distance using songs, personal names, and images of the person, and by acting on traces a person has left behind. These doctrines express a person's vulnerability to attack, rendering explicable otherwise unexplained misfortune. This vulnerability apparently results (in part at least) from the intrinsic connection between the person and parts of the person, distributed in space.
Sorcery not only has to do with blaming people for causing misfortune, especially grave illness and untimely death, but is also a mode in which persons both threaten harm and take credit for causing deaths. Morever, people invoke sorcery to deflect responsibility for an action away from the self--'Yes, I killed x by throwing a spear, but someone made me throw the spear through sorcery'. (10) Complementing the deflection of responsibility in explaining actions is the extension of the target of an action. When taking redressive action for a perceived wrong, Yolngu found it appropriate to target a person of the same patri-group or another close kinsperson, rather than the wrongdoer him- or herself (e.g. Warner 1937: 177). In a sense, then, the individual is an intrinsic part of a wider kin-network and patrifilial identity, in a relation similar to that between part of the body and the whole (cf. Wagner 1991).
In the gift of parts of the body of living children, of gifts made by children, and of the hair of the living or dead, bodily substance is distributed in social space, extending the person beyond his or her immediate presence. These acts of giving are overt rather than covert, however, and (except for the instigation of revenge) are done without aggressive intent, being tied to the maintenance of relationships (especially between distant communities) and associated with the exchange of goods. Sorcery expresses aggressive aspects of social relationships, reflected in explanations of misforturne and revenge attacks that followed accusations of sorcery. The 'gifts' involved were of parts of body of the dead, intended to initiate revenge or 'pay-back' (Warner 1937: 177). (11)
Conclusions
In this article I have drawn on the literature on partible persons in order to explore Yolngu doctrines about totemic ancestors, sorcery, and gifts, in terms of extensions of persons in time and space, and I have invoked Munn's (1970) account of ancestral transformations in order to relate these extensions of persons to the moral-political domain.
Sorcery and magic, based on the powers of the bodies of the living and the human dead as well on aggressive aspects of wangarr ancestors, are about the day-to-day flux of relationships, notably aggressive aspects, whereas gifts of bodily substance to initiate exchange maintain relations of amity. The exchange of bodily substance can (or could) also instigate acts of revenge for a death. Ancestral doctrines, which relate to the powers of the traces of totemic ancestors (who are imagined as super-persons), are concerned with the timeless foundation of the social order and its reproduction as well as being a source of authority and power for everyday purposes. All are based on the same imputed powers of the body and its parts, and on the intrinsic relation between part and whole, image and object, such that action on the (detached) part affects the whole, and the part is imbued with the power of the whole.
The Durkheimian tradition has been misleading in equating 'religion' with society, and 'magic' with the individual. Ancestral doctrines, sorcery, and gift-giving embed people in moral-political relations in contrasting though inter-related ways. All three domains are aspects of Yolngu sociality with distinct but related bases, and each is a kind of rom ('the proper way' or 'right practice'). The first provides the doctrinal foundations of the moral and political order; the second partakes in the aggressive aspects of personal relations and relates to particular notions of responsibility (which I have heard referred to as 'yolngu rom'--'the way of people'); the third maintains and extends relations of amity. Ancestral powers, like sorcery, can be deployed aggressively, but the threat of sorcery is also a sanction deployed to guard religious secrecy (Keen 1994: 196); the two usages are not antithetical. This article shows that Yolngu ancestral doctrines and practices are, amongst other things, practical means of acting in and on the world, drawing on the imputed powers of ancestors, and having effects on ancestors, people, and the cosmos. Religious doctrines, magic, sorcery, and elements of exchange are constituted from the same stuff.
NOTES
Earlier versions of this article were presented at the 2004 meeting of the Association for Social Anthropology in Oceania, the 2004 San Francisco meeting of the American Anthropological Association (or what remained of it after its postponement due to industrial action at the conference hotel), and at the joint anthropology seminar of the Australian National University in March 2005. I thank participants for their comments, Mark Mosko for his encouragement, and Katie Glaskin, Bentley James, Yasmine Musharbash, and Terry Turner for reading and commenting on earlier versions of the manuscript. The article has also benefited from comments and helpful suggestions from anonymous reviewers, and from Glenn Bowman.
(1) The name 'Yolngu' refers to the indigenous population of northeast Arnhem Land in the Northern Territory of Australia, known in earlier writings as 'Murngin', 'Wulamba', and 'Miwuyt'.
(2) On magic and colonialism see, for example, Comaroff & Comaroff (1993); Geschiere (1988); Kapferer (2002); Shaw (1997).
(3) Gell's (1998) concept of 'distributed person' is rather different from the idea of 'partible person'. He has in mind the connection between a 'primary' agent and a material object that appears to be intrinsic to personhood, such as the relation between a soldier and his or her gun; an object such as this becomes a 'social agent' (Gell 1998: 20-1).
(4) Morton (1987) has extended Munn's analysis, adding a psychoanalytical perspective to look at Arrernte initiation. He argues that Arrernte individuals experienced more creative autonomy than did the people of Munn's account, and Redmond (2001a) concurs in relation to Ngarinyin relations with totemic ancestors, which are reciprocal.
(5) Guykthun denotes the act of making something sacred or taboo (dhuyu) by uttering powerful words, as well as to spit, or spurt out (Zorc 1986).
(6) The following views of Yolngu men have been recorded: the penis forces menstrual blood to flow up into the uterus, where it coagulates to form a foetus (Berndt 1952: 271); repeated copulation is necessary for conception and for the development of the foetus, just as the mother's milk nourishes the infant after birth (Shapiro 1981: 19); the wangarr presents the child as a gift but the semen forms the body, arms, and legs (Keen 1978: 304); or semen is a kind of liquid bone (James n.d.). These views stress the male role in the formation of the foetus.
(7) Indeed, according to Dja:wa, the Daygurrgurr patri-group leader in the mid-1970s, there is a conceptual connection between the madayin (totemic sacra) and the hair; madayin are 'from the head', and hair has power for that reason (Keen 1978: 338).
(8) Other forms of love magic include placing pubic hair or stones steeped in body-fluids in the basket or bed of an intended lover (James n.d.).
(9) See Magowan (2001), Redmond (2001b) and Tamisari (1998) on dynamic aspects of the totemic landscape.
(10) See Keen (2003) and Martin (1993) on the deflection of responsibility. Analogously, a young man, accused of violence, might deflect responsibility for his action, claiming that it was the 'pushers' who made him do what he did (see Warner 1937: 168-9 on 'pushers' in blood-feuds).
(11) Warner (1937: 177) transcribes the Aboriginal English expression 'pay-back' as 'buy-back'; the Yolngu expression is ba:ka-bakmarama, 'take revenge'.
REFERENCES
BELL, D. 1984. Central Australian Aboriginal women's love rituals. In Aboriginal religion in Australia: an anthology (eds) M. Charlesworth, H. Morphy, D. Bell & K. Maddock, 345-69. St Lucia: University of Queensland Press.
BERNDT, R.M. 1952. Djanggawul: an Aboriginal religious cult of north-eastern Arnhem Land. Melbourne: Cheshire.
______ 1974. Australian Aboriginal religion. Leiden: E.J. Brill.
BIERNOFF, D. 1978. Safe and dangerous places. In Australian Aboriginal concepts (ed.) L.R. Hiatt, 93-105. Canberra: Australian Institute of Aboriginal Studies.
BORSBOOM, A.P. 1978. Maradjiri: a modern ritual complex in Arnhem Land, North Australia. Nijmegen: Katholieke Universiteit.
Bos, R. 1979. Report on field work: July 1979. Department of Sociology and Anthropology, University of Queensland.
______ 1988a. Jesus and the Dreaming: religion and social change in Arnhem Land. Ph.D. thesis, University of Queensland.
______ 1988b. The Dreaming and social change in Arnhem Land. In Aboriginal Australians and Christian missions (eds) T. Swain & D.B. Rose, 422-37. Bedford Park, South Australia: The Australian Association for the Study of Religions.
BUSBY, C. 1997. Permeable and partible persons: a comparative analysis of gender and body in South India and Melanesia, Journal of the Royal Anthropological Institute (N.S.) 3, 261-78.
COMAROFF, J. & J.L. COMAROFF 1993. Introduction. In Modernity and its malcontents: ritual and power in postcolonial Africa (eds) J. Comaroff & J.L. Comaroff, xi-xxxvii. Chicago: University Press.
CREIGHTON, S. 2003. The Yolngu Way: an ethnographic account of recent transformations in indigenous education at Yirrkala, northeast Arnhem Land. Ph.D. thesis, Australian National University.
DUMONT, L. 1970. Homo hierarchicus: the caste system and its implications (trans. M. Sainsbury) Chicago: University Press.
DUNLOP, 1. 1989. Djungguwan at Gurka'wuy (videorecording). Produced and directed by Ian Dunlop; documentation and translation Dundiwuy Wanambi. Lindfield, N.S.W.: Film Australia.
DURKHEIM, E. 1915. Elementary forms of the religious life (trans. J.W. Swain). New York: Free Press.
ELKIN, A.P. 1967. Religion and philosophy of the Australian Aborigines. In Essays in honour of Griffithes Wheeler Thatcher 1863-1950 (ed.) E.C.B. McLauren, 19-44. Sydney: University Press.
GELL, A. 1998. Art and agency: an anthropological theory. Oxford: Clarendon Press.
GESCHIERE, P. 1988. Sorcery and the state: popular modes of action among the Maka of southeast Cameroon. Critique of Anthropology 8, 35-63.
GLASKIN, K. 2005. Time, death and being: reflections on material change, relics and ontology in an Aboriginal society. Paper presented at the meeting of the European Society for Oceanists, Marseilles, July.
GREGORY, C.A. 1982. Gifts and commodities. London: Academic Press.
JAMES, B. n.d. Unpublished notes on Yan-nhangu sorcery and magic. Typescript.
KAPFERER, B. 2002. Introduction: outside all reason: magic, sorcery and epistemology in anthropology. Social Analysis 46: 3,1-29.
KEEN, I. 1978. One ceremony, one song: an economy of religious knowledge among the Yolngu of northeast Arnhem Land. Ph.D. thesis, Australian National University.
______ 1994. Knowledge and secrecy in an Aboriginal religion: Yolngu of northeast Arnhem Land. Oxford: Clarendon Press.
______ 2003. Dreams, agency, and traditional authority in northeast Arnhem Land. In Dream travelers (ed.) R.I. Lohmann, 127-48. Basingstoke: Palgrave Macmillan.
______ 2004. Aboriginal economy and society: Australia at the threshold of colonisation. Melbourne: Oxford University Press.
KNAUFT, B. 1984. Bodily images in Melanesia: cultural substances and natural metaphors. In Fragments for a history of the human body (part 3) (ed.) M. Feher, 198-279. New York: Zone.
LIPUMA, E. 1988. The gift of kinship: structure and practice in Maring social organization. Cambridge: University Press.
MAGOWAN, F. 2001. Waves of knowing: polymorphism and co-substantive essences in Yolngu sea cosmology. The Australian Journal of Indigenous Education 29: 1, 22-35.
MARRIOTT, M. 1968. Caste ranking and food transactions: a matrix analysis. In Structure and change in Indian society (eds) M. Singer & B.S. Cohn. Chicago: Aldine.
______ & R. INDEN 1977. Towards an ethnosociology of South Asian caste systems. In The new wind: changing identities in South Asia (ed.) K. David, 227-38. The Hague: Mouton.
MARTIN, D. 1993. Autonomy and relatedness: an ethnography of Wik people of Aurukun, western Cape York Peninsula. Ph.D. thesis, Australian National University.
MAUSS, M. 1954. The gift: forms and functions of exchange in archaic societies (trans. I. Cunnison). Glencoe, Ill.: Free Press.
MORPHY, H. 1977. Too many meanings: an analysis of the artistic system of the Yolngu of northeast Arnhem Land. Ph.D. thesis, Australian National University.
______ 1991. Ancestral connections: art and an Aboriginal system of knowledge. Chicago: University Press.
______ 1995. Landscape and the reproduction of the ancestral past. In The anthropology of landscape (eds.) E. Hirsch & M. O'Hanlon, 184-209. Oxford: Clarendon Press.
MORTON, J. 1987. Singing subjects and sacred objects: more on Munn's transformation of subjects into objects in Central Australian myth. Oceania 58, 100-18.
MOSKO, M. 1992. Motherless sons: divine kings and partible persons in Melanesia and Polynesia. Man (N.S.) 27, 697-717.
______ 2000. Inalienable ethnography: keeping-while-giving and the Trobriand case. Journal of the Royal Anthropological Institute (N.S.) 6, 377-96.
MUNN, N. 1970. The transformation of subjects into objects in Walbiri and Pitjantjatjara myth. In Australian Aboriginal anthropology (ed.) R.M. Berndt, 141-63. Nedlands: University of Western Australia Press.
PETERSON, N. 1975. Mortuary customs of Northeast Arnhem Land: an account compiled from Donald Thomson's field notes. Memoirs of the National Museum of Victoria 37, 97-108.
POVINELLI, E. 1993. Labor's lot: the power, history, and culture of Aboriginal action. Chicago: University Press.
REDMOND, A. 2001a. Rulug wayirri: moving kin and country in the northern Kimberley. Ph.D. thesis, University of Sydney.
______ 2001b. Places that move. In Emplaced myth: space, narrative, and knowledge in Aboriginal Australasia and Papua New Guinea (eds) A. Rumsey & J.F. Weiner, 120-38. Honolulu: University of Hawai'i Press.
REID, J. 1983. Sorcerers and healing spirits. Canberra: Australian National University Press.
SHAPIRO, W. 1981. Miwuyt marriage: the cultural anthropology of affinity in northeast Arnhem Land. Philadelphia: International Society for the Study of Human Issues.
SHAW, R. 1997. The production of witchcraft/withcraft as production: memory, modernity, and the slave trade in Sierra Leone. American Ethnologist 24, 856-76.
SLOTTE, I. 1997. We are family, we are one: an Aboriginal Christian movement in Arnhem Land, Australia. Ph.D. thesis, Australian National University.
STANNER, W.E.H, 1964. On Aboriginal religion. Sydney: University Press.
STRATHERN, M. 1978. The achievement of sex: paradoxes in Hagen gender thinking. In Yearbook of symbolic anthropology 1 (ed.) E. Schwimmer, 171-202. Montreal: McGill/Queens University Press.
______ 1988. The gender of the gift: problems with women and problems with society in Melanesia. Berkeley: University of California Press.
SWAIN, T. 1985. Interpreting Aboriginal religion: an historical account. Bedford Park, South Australia: Australian Association for the Study of Religions.
______ 1993. A place for strangers: towards a history of Australian Aboriginal being. Cambridge: University Press.
TAMISARI, F. 1998. Body, vision and movement: in the footprint of the ancestors. Oceania 68, 249-70.
TAYLOR, L. 1996. Seeing the inside: bark painting in western Arnhem Land. Oxford: Clarendon Press.
THOMSON, D.F. 1948. An Arnhem Land adventure. The National Geographic Magazine 93, 403-30.
______ 1949.Economic structure and the ceremonial exchange cycle in Arnhem Land Melbourne: University Press.
______ 1961. Marrngitmirri and kalka: medicine man and sorcerer in Arnhem Land. Man 61, 97-102.
______ 1975. The concept of 'marr' in Arnhem Land. Mankind 10, 1-10.
WAGNER, R. 1991. The fractal person. In Big men and great men: personifications of power in Melanesia (eds) M. Godelier & M. Strathern, 159-73. Cambridge: University Press.
WARNER, W.L. 1937. A black civilization: a social study of an Australian tribe. (First edition). New York: Harper.
WEBB, T.T. 1936. The making of a marrngit. Oceania 6, 336-41.
WEINER, A. 1992. Inalienable possessions: the paradox of keeping-while-giving. Berkeley: University of California Press.
WEINER, J. 1988. The heart of the pearlshell: the mythological dimension of Foi sociality. Berkeley: University of California Press.
WEST, L.M. 1962. Progress report on the processing of data since return from the field 1st March 1962-1st April 1962. PMS 4424, Australian Institute of Aboriginal Studies, Canberra.
WILD, S. 1986. (ed.) Rom: an Aboriginal ritual of diplomacy. Canberra: Australian Institute of Aboriginal Studies.
ZORC, S. 1986. Yolngu-matha dictionary. Batchelor: School of Australian Linguistics, Darwin Institute of Technology.
Ian Keen studied anthropology at University College London and the Australian National University, and lectured in anthropology at the University of Queensland and the ANU, where he is currently a Visiting Fellow. He is the author of Knowledge and secrecy in Aboriginal religion (Clarendon Press, 1994) and Aboriginal economy and society: Australia at the threshold of colonisation (Oxford University Press, 2004), as well as numerous journal articles and chapters in edited books.
School of Archaeology and Anthropology, The Faculties, Australian National University, Canberra, ACT 0200, Australia.
Ian.Keen@anu.edu.au" onclick="window.open(this.href);return false
IAN KEEN
Australian National University
