Printer Friendly

Kabit, rimbu and opa: modalities of Anganen sacrifice.


Once the topic of grand theorists such as Robertson-Smith (1899) or Hubert and Mauss (1964), sacrifice can no longer be comprehended as a unitary phenomenon (Valeri 1985). As van Baal (1975) notes, one reason for this is that such arguments did not consider sacrifice among small, tribal societies such as the Anganen of the Southern Highlands Province of Papua New Guinea. Here I discuss the two most important forms of male exclusive sacrifice, kabit and rimbu, the Anganen practised at the time of Australian colonial control and their abandonment soon after due to missionary pressure. The present argument builds upon a series of previous discussions (Nihill 1988a, 1988b, 1996a, 1996b, 2000a, 2001) which explore the theme that Anganen forms of social practice do not simply possess their own intrinsic meaning, but gain further significance in relation to other forms of practice. This may be partly through similarity, but mostly through contrast. It is the dynamic interrelations between forms of practice that gives vitality to the expression of key aspects of Anganen culture and social structure. This is especially the case when a temporal perspective is adopted as rimbu was periodically abandoned and the houses defiled, while kabit was not, the houses only being rebuilt if beyond repair.

I have already considered some aspects of rimbu and kabit in comparison with each other and with gift exchange (Nihill 1996a), focusing on similarities and differences in the forms of social organisation. Here I wish to consider aspects tangential to that argument with the focus on animal sacrifice and the venues in which it takes place, though it will be necessary to reiterate certain points of the earlier discussion. Once this is done, I wish to consider the impact, especially upon men, of the abandonment of these cults with conversion to Christianity. Part of this discussion will also consider what the Anganen consider to be a similarity between their pre-Christian cults and aspects of Christian practice, despite the radical difference between the two. This is known in Tok Pisin as opa, literally, 'offer/offering', involving the donation by the practitioners of money, food and labour to the church and church projects.

The critique of singular approaches to sacrifice is perhaps most severe with Victor Turner's (1977) likening sacrifice to another concept which has had long anthropological interest, totemism. In doing so he invokes the argument of Levi-Strauss (1963) that, like totemism, sacrifice as a phenomenon in and of itself does not exist, and therefore no single theory could be a satisfactory explanation of those practices categorised by anthropologists under this label (also see Bloch 1992:25). Moreover, even using the term 'sacrifice' is problematic if the definition of Hubert and Mauss (1964) that the offering is consecrated is adhered to. That is, it undergoes a rite of sacralisation to be followed by one of de-sacralisation. While I will note certain elements consistent with rites of passage, that the offering is specifically sanctified does not apply in Anganen. In the Highland context, this point leads Meggitt (1965:120) to reject sacrifice in favour of the more general term 'offering'. Certainly this observation is consistent with the Anganen recognition of similarity between kabit, rimbu and opa despite their at times profound differences. However, I shall use the term sacrifice here for a number of reasons. In part, as will be subsequently commented upon, the sacrificial offering as consecrated assumes the Durkheimian legacy of the sacred (and its opposition to the profane). More specifically, I use the term in part through convenience and in part as a heuristic device to explore these forms of practice.


The Anganen live in the southern Lai-Nembi Valley region of the Southern Highlands Province of Papua New Guinea, being part of the some 200,000 Mendi-Pole language speakers (Franklin 1968). Their economy is based on subsistence horticulture featuring sweet potato as the staple and pig husbandry. For the vast amount of households, the principle source of cash is the sale of coffee from small holder plots. Land is held by patrilineal clans, but non-agnates can achieve full ownership rights (Nihill 1986:63-83), and the local group is always comprised of a number of named patrilineal groups, something important for the organisation and meaning of the rimbu spirit cults in particular. This is the main theme of Nihill (1996a), and a point returned to below.

The Anganen were brought under Australian colonial control in the early nineteen sixties. The Det Roman Catholic mission was established in 1964 and, apart from a small Protestant mission, namely the Christian Union Mission based at Farata, Roman Catholics dominate religious affiliation. By 1980 when I first commenced fieldwork research, everyone I met claimed to be a Christian belonging to one or the other of these churches. In addition to the church and mission quarters at Det, there are a number of village churches scattered throughout the region, with the Roman Catholic mission also providing the local primary school, a health centre that also doubles for nurse aide training, and they have undertaken a myriad of projects aimed at economic development. Even so, the Anganen remain poor by Highland standards. The Australian government priority was to achieve so called 'pacification' as a prerequisite for economic, legal and political development, whereas the missions' first aim was conversion, something that required the cessation of animal sacrifice to the spirits. The abolition of warfare (despite occasional outbreaks of armed hostilities) and the abandonment of spirit houses had a profound effect, especially upon men who were warriors and animal sacrificers, an issue I will return to when discussing opa once the cults are considered.


There are two kinds of spirits, remo, those of dead humans and those which have never had human existence. Sokele dominate the latter. These are associated with places of danger such as the forest, rivers, sinkholes and so on. They are totally malicious and associated with drowning and hunting accidents. They are completely beyond human influence and must be avoided. The only exception is a spirit called yeki. It is felt that it can bring good fortune, but it is also thought to be the cause of leprosy. Lepers horrify the Anganen, and in the past they said that not only did they throw their corpses into the river they may have thrown bound live persons in too. In men's houses, kapanda, a pig may be slaughtered and the smell of its singed bristles and internal organs such as the liver or spleen (collectively pu) given in offering. For instance, yeki is also thought a cause of diarrhoea, and if there is widespread occurrences or an individual is severely affected then men undertake such offerings. Yeki is associated with the sky, especially with clouds, lightning and thunder.

Of most importance are the spirits of dead humans. The spiritual aspects of humans have two named components: the remo, or what is now understood as the soul, and the wesa, or animating force. The latter does not survive death, but the former does. Eventually the individuality of the deceased disappears and the ancestors, sumbal, are seen as a collective. Ghosts are seen as particularly malevolent, seeking revenge on their kin due to their anger at having been let die. They are also reluctant to leave the places of their residence and are thought to attach themselves to small marsupials sometimes referred to as house pigs, that enter the village and houses seeking food. Should such an animal enter the village it must be handled with great care so as not to enrage the ghost further.

Ancestors may assist the living. For instance, via dreams they may communicate certain future events. For the most part, though, they are malevolent. They are said to 'strike', ta, or 'eat', na, the living, causing serious afflictions in the process. (2) No specific ailments are associated with their attack. Instead, spirits eating the living is the explanation for the great, if not life threatening, severity of the illness. While widespread conflict in the community or acts of incest trigger their attacks, (3) for the most part motive is unknown. However, in addition to the human quality of anger, they are also thought to experience hunger and are open to influence by the living through the offering of the aroma of cooking meat which they crave. In the most general sense, all the aromas of cooking meat are seen as an offering, something some support by noticing that aromas fade. It is this that the Anganen feel is why the ancestors did not wreak havoc after the abandonment of spirit houses and their associated rituals.

The Anganen remember eight types of spirit houses, some imported new, one, the banana rimbu (epele rimbu) was thought to be largely invented locally, while others are modifications of previous rituals. All had the commonality of the offering of the aroma of cooking meat. Here I will concentrate upon the kabit and rimbu.


Kabit is said to be a much older ritual than rimbu. (4) No one could say with any certainty when it was imported, but many think it directly succeeded the 'bone [head] houses', aesumba anda, where pigs' blood would be offered by rubbing it onto the skulls of dead, patrilineal relatives to 'feed' them. Until colonisation, the Anganen did not bury the dead, preferring to place them on rocky ledges or in caves to decay. Bones such as skulls or finger bones would be appropriated. For example, finger bones were thought to make auspicious arrow heads.

While only men and older boys could enter, women knew of their location as they were built very close to villages. Women and children were also given meat to eat once the ritual was over. Those males in attendance were usually members of the singular subclan (yamongiki) of the afflicted individual. An older man would act as the leader, kabit aesumba. Save the necessity to keep the procedure from unknowing women and children, necessitating the exclusion of younger boys, there was nothing approaching what could be glossed as initiation for males when first attending the ritual.

The kabit house, kabit anda, resembles the men's house, being built of identical materials, though no one saw the similarities between the names--kapanda for men's house and kabit anda--as significant. The main differences were that kabit anda were shorter and a little higher. They had a single door for entry and exit, with a central pole (pinki). There were benches to sit on and a fire toward the back middle. The other feature was a small house in one comer, also called kabit anda or onge kabit anda, 'little kabit house'. It was built out of the same materials as the bigger house and likewise was covered with a roof. This contained stones, kabit, thought of as representatives of the dead. These stones were selected for some association with the deceased, say being found in one of their gardens. They were also chosen for their particular qualities such as smoothness and roundness, but mostly because of perceived anthropomorphic qualities. Of most note here are holes which look like mouth protrusions and are seen to resemble breasts or buttocks, or length which resembles the nose and tongue. Despite certain phallic and uterine images, the Anganen deny any sexual characters. This is so even though the spirits of dead men are honoured with longer stones, but this fits into a gendered schema where greater height is masculine. For instance, men's houses are taller than women's. While they were stored together, the longer stones were placed vertically into the house, while the smaller were laid down. This was said to 'avoid them fighting'. Occasionally fighting is a metaphor for sexual intercourse but again no significance was placed upon this.

The stones were placed on a special edible green called kapibio to 'sleep', and red paint was regularly applied to them. There is an interesting comparison with the 'bone houses' they replaced where red blood was applied to skulls. The kabit procedure seems to be based on the same meaning but in more abstract form. This paint was also thought to feed the stones and make them happy, with the red ochre pleasing them just like humans when they decorate. The image of the red stones on green leaves also parallels the way red pearl shells, which also have this ochre applied to their 'skin', were placed against a green back drop for display. This is another aesthetically pleasing sight (at least until the substantial devaluation, especially by younger men in the latter part of the nineteen eighties).

Spirits appealed to in kabit could be dead persons of either gender, but there was ambiguity over married women. As a rule of thumb, it depended on the length of residence a women had in her husband's territory. If short, it is most likely the spirit would return to the woman's natal place to be with her brothers: if long, and especially with the birth of sons, it may be the case it remains in her husband's territory. The only way to know is if a person falls seriously ill soon after an in-married woman dies.

Sacrifice in kabit virtually always followed a member of the subclan or an in-married woman falling seriously ill, with the potency of sacrifice related to patrilineal descent proximity and concentration; indeed often the ghost of a recently dead person would be regarded as the afflicting spirit. As such, a single ancestor would be approached. The procedure for the offering was simple. While the pig was cooking, the leader would ask the afflicting spirit to stop eating the sick person and consume the smell and smoke of the pork instead, even though it was thought that all the spirits represented by the stones would also partake thereby expanding possible positive influence. This usually happened in late afternoon and, once done, the participants would return home to consume the meat with their families. The afflicted person should then recover.


Rimbu is an interesting word, and seems related to terms that other Southern Highlanders use in reference to spirit cults (e.g., see Clark 2000 for the Wiru and Ryan 1961 for the Mendi). In Anganen rimbu may have similar characteristics to those Burridge (1969:152) notes for divine, namely having pronounced positive but also pronounced negative connotations. The Anganen version of the infamous 'rubbish man' of the Highlands, a man who is a political nonentity, often unmarried and dependent upon others, and perhaps with physical or mental handicap, is termed a rimbu, a being the generic term for the male person. Rimbu also refers to this most powerful spirit cult which is deemed to have widespread positive benefits for people, pigs, gardens and wealth. As a term, rimbu suggests the exceptional or extraordinary. The linguists the Franklins (1978:442) maintain rimbu means 'sacred or feared' in the MendiPole language sub-family. While apprehension is an important aspect of this cult at certain stages of its performance, I am hesitant to adopt the term 'sacred'. It seems to assume the universality of the Durkheimian sacred-profane dichotomy. In line with certain assumptions based on Judeo-Christian thought which may have undermined the effectiveness of certain arguments about sacrifice in general (Bell 1997: 112; Bowie 2000:139), to deploy it may be misleading when dealing with peoples such as the Anganen. As such, terms like exceptional or extraordinary, although of far less impact than that of 'sacred', appear best to designate rimbu, with the contrast not on the profane but the mundaneness of the everyday as should become clear.

The social dimensions of rimbu are in distinct contrast with kabit. Its social organisation is based on the local group. While in nearly all cases a named patrilineal clan is seen as the original inhabitant of a territory, a substantial amount of the local groups' permanent residents are not agnates of the settler clan, with most in fact seen as having full ownership privileges. Moreover, for any given performance of rimbu, men from outside the local group may also be invited to attend. The spectrum of spirits open to influence is also certainly much greater than the ghost of a recently deceased person as is possible in kabit. Unlike the neighbouring Kewa (see Jebens, 2005; Josephides 1985:76; MacDonald, 1991), yeki, the powerful, never human, pre-Christian figure but with the term adopted for God in the vernacular, may also be included into what may be called a spirit pantheon.

There are two rimbu houses, house sites, and associated rituals, the solu rimbu, 'long rimbu', and the rundu rimbu, 'short rimbu'. (5) Although both rimbu were purchased from Kewa men to the south, they arrived in Anganen at different times. The long rimbu was originally just termed rimbu until the importation of its short counterpart and is said to have been practised for some time. As far as I could ascertain, it appears to have been first practised in the early stages of the twentieth century. The short rimbu came later, probably in the late nineteen thirties. At any rate, it was adopted prior to the Anganen hearing the engines of Japanese aircraft in World War Two. It directly replaced the epele rimbu ('banana-base rimbu'), a cult thought to have originated in Anganen but abandoned as it was deemed too powerful. For instance its inventors, the Aramuri clan, were initially blessed with male child after male child. However, they came to realise that this was a situation that could not continue, given that bridewealth paid for a woman to enter the clan often came from the bridewealth paid for an outgoing sister upon her marriage.

The two houses shared many important features, but they also differed in some respects, most notably in location and, to a degree, conceived purpose. Both houses were constructed near rocky outcrops which often may have spiritual association. They were also built in marginal social space such as copses on the edges of villages or abandoned garden sites. Such sites, if on high ground, are also favoured for present day cemeteries. Spirit houses were never built in the forest, the domain of spirits, but sites such as these are often associated with spirits who have temporarily left the forest for whatever reason. Inexplicable light, especially red in colour, is the most common indication of this occurring. This spatial marginality is important, as it mediates the human world of the village to what is labelled the 'wild' (kera) of the forest. Whatever their location, they were built further from the village than kabit anda.

The location of rimbu houses was known to all, but the houses and their surrounds were shielded behind trees or substantial secondary growth. It was vital that their ceremonies were shrouded in secrecy. Females and uninitiated males could not even enter the compound surrounding the houses. Initiation does not formally feature in Anganen, except for rimbu, though it does not have the social significance of the pronounced length of male initiation in other parts of Melanesia, such as in cultures of the Eastern and fringe Eastern Highlands Province. Beyond the cult itself, it had no impact on an individual's social standing. However, it did possess some features common to initiation in these other regions. Initiation revealed the power of the spirits and the avenues available to channel it to human advantage. Initiation also instilled the need for secrecy, with death through angry spirits or through angry men certain to follow if anything to do with rimbu were to be revealed to the uninitiated. This coercion was extended to men telling the boys that rimbu may attack at any time if the boys were deemed unworthy of membership. Indeed, the induction of boys takes place after others have entered the house and the mysterious flutes are played with the boys told they are the rimbu inside. The only exception to this secrecy was that men conveyed to the uninitiated the overall power and importance of the cult, the true impact of a secret being its partial revelation. Doing so, helped cement the power of adult men through generating the dependence of women and children on male performance. As Bloch (1992) argues, the power dimensions of ritual, sacrifice included, are integral to its significance.

To sustain initiated male exclusivity, at dusk on the day of the ritual, men would go to the rimbu compound. Two men would play bamboo flutes, one long (solu) and one short (rundu). Collectively the flutes produced rimbu anga, literally 'the talk of the spirits'. This was an eerie whistling sound that is said to terrify women and children if they heard it, while also alerting them to stay well clear of the area. This noise as the 'talk of the spirits' is interesting as it is freely acknowledged that men play the flutes and that the flutes carry no great significance in and of themselves. They are roughly constructed by their players, there is nothing that could be construed as consecration, the players need not have any social prominence, and it does not confer any significant prestige upon them. The only rule governing who may be a flute player is that no two agnates could simultaneously play both the short and long flutes. In most cases, a man of the clan thought the original settlers would play one, while the other was played by a land owning descendant of an immigrant.

The flute players are one aspect of a dualism or binarism that was a central theme in my earlier discussion of a number of the social dimensions of rimbu (Nihill 1996a). In a thematic and not statistical sense, rimbu social organisation is very much premised on marriage. On the one hand, Anganen local groups, despite having agnates and non-agnates as land owning residents, tend to progress toward what D'arcy Ryan (1969) termed 'antigamy' in the neighbouring Mendi. The consequence of marriage prohibitions is a tendency for local groups to progress toward exogamy. Any one sharing the same clan name is not a possible spouse. Moreover, any one marriage between two subclans prohibits all further unions. In tandem with the preference to marry those close by, quickly the accumulated sum of marriages between non-agnatic subclans means that increasingly co-residents in any local group will be unlikely marriage partners. This is never achieved statistically, as subclans may fission and new immigrants may enter, meaning that new possible spouses may occur, but it does give a sense of unity, despite non-agnatic connections, to the local group which is usually the basis for rimbu organisation. This is also played out in the setting and design of the rimbu houses themselves.

There is a single gate, paruwa, by which men enter the area surrounding the house. This too is regarded as rimbu and is also highly auspicious. Men enter in two lines, one the original settlers (and their invited guests if present), the other those of immigrant groups (and their invited guests, if any). Facing them are two entrances to the house itself. Two rimbu leaders (rimbu aesumba), again one from the original settlers and one from the other side, open the doors. These are always men of renown, and in the early years of each rimbu operation, were likely to be those that negotiated and contributed most to the purchase of cults. They open the door in a ceremony kept secret from all others present. The leaders employ a special argot, namongo, that must be repeated in sequence each time rimbu is staged and is repeated when the sacrificial animals are slaughtered. (6) This takes place before the rest of the men enter the house. These words are also repeated once those present have left the house following the ritual's completion and the men have left the house by two doors at the back of the house which always faces towards, if not being adjacent to, the forest. Leaving the houses usually occurred just before dawn. These occasions are considered the most dangerous and fit well with the schema first proposed by van Gennep (1960) and expanded upon by Turner (1967, 1969). The killing of the sacrificial beast is the passage from life to death; dusk and dawn, together with doors are excellent representations of limen or threshold and pending transformation, and denote the beginning and end of a ritual. (7) Moreover, as Turner among others has noted, notions of danger and heightened emotion do not simply accompany but are constituted through and are constitutive of ritual intensity.

Dualism is repeated once more inside the house, as men from the original settler clan sit on one side and immigrants on the other. Excluding any invited guests, from the original settler clan perspective, the opposite side is thematically premised on marriage, as these men are from groups from which they may have obtained wives and into which their clan sisters have married. Once again, this is thematic, as it is probable that intermarriage occurs among nonagnatic immigrant groups, but the themes of marriage are important. As the prohibition on further marriage between any two subclans shows, sister exchange is not possible, and Anganen social structure is premised on generalised exchange. The image of local group closure on the one hand and a dualism reflective of a restricted type of exchange thereby gives rise to a image of the social in rimbu in stark contrast to that of conventional Anganen social structure. This is further magnified by a male exclusivity simply not possible in reality.

Two further points may be taken from this contrast between rimbu and conventional social organisation. First, although not in an evolutionary sense, this contrast is similar to that which Levi-Strauss (1969) peruses in The elementary structures of kinship. In my discussion of rimbu and kabit in relation to gift exchange (Nihill 1996a), I argue this contrast not only intensifies the meaning of each form of sacrifice and different exchange practices, their multiple interconnection significantly generates much of the dynamics of Anganen social life. Second, as with the male cults of the Eastern Highlands for instance, there is a male only, but nonetheless potent, force. Despite the somewhat oblique reference to the 'initiation' of boys into rimbu discussed previously, this is not, to use Marilyn Strathern's (1984) term a 'parthenogenetic' making of men in their ritually guided transition from boys, but a male only creativity through which men act upon their social and biological worlds. Underpinning the sense of male exclusivity is that water which is conventionally fetched by women is prohibited from the house, with sugar cane, a 'male' plant in Anganen classification quenching any thirst. Perhaps even more notable is the taboo on sweet potato, the staple of the Anganen diet but very much associated with women as its producers. Not only do these prohibitions help intensify the maleness, they are important markers of the distinction of rimbu performance from this conventional world thus highlighting its extraordinary character.

Here the contrast of rimbu with kabit is important, and this too lends gravity to the importance of rimbu. Demographically, rimbu is much greater even though both cults are exclusively male, and this also suggests the greater potency of rimbu. Marriage and the symbolic importance of any notions of reproduction can never feature in kabit which always must reflect only patrilineal descent and a narrow focus on descent as such. The double dualism of the rimbu doors in contrast to the single door of kabit is also significant. The singular door likewise may symbolise descent while the duality of the doors of rimbu reflects this broader social thematic articulation. There is no ritualisation of the kabit door's opening and closure. This contrast also lends weight to the overall importance and power of rimbu in comparison with kabit.

Another event, though not a feature of all rimbu performances is a dance called pol-keri (the words apparently have no meaning beyond the name of the dance). Beyond it being a good thing to do, no one would venture an explanation of its meaning apart from a degree of emphasis that it was best undertaken when new members were inaugurated into the cult. The dance, by necessity if not design given the confines of the house, was stilted with men remaining very much on the one spot, and only accompanied by a hissing sound, 'sis-sis-sis' (although speculative, it may have some resonance with the subsequent discussion which returns to the flutes). Pairs were formed, one dancer from each side of the house, with some emphasis, especially for new members, on partnership of men of high social esteem and knowledge and those their juniors. The partners would hold a long section of sugar cane and, hissing, jump up and down. The dance could continue for some time, and a number of older men said they were quite exhausted when it finished. Once the ritual leaders had brought the dance to a halt, each pair would tussle to get the sugar cane in a manner reminiscent of a tug-of-war. If there were substantial discrepancies in age, the younger male was likely to prevail.

As noted, sugar cane is categorised as a 'male' plant in Anganen, and as in Kewa it may have phallic connotations, being 'sometimes a metaphor for the penis ... the locus of male generative powers' (LeRoy 1985:216). It may thus be tempting to see this as some symbolic or analogical equivalent of so called ritualised homosexuality practised as part of male initiation elsewhere in the Highlands. However, I think this would be severely over-reading the matter. Certainly no man gave any hint of an erotic character to pol-keri. I am more swayed by part of the argument Schwimmer (1993) puts forward for the importance of male pairings in Melanesia when discussing their common presence in dancing. 'While devoid of overt sexual content, dancing expresses sensual relations of symbolic supremacy, submission and balance ... in male gender identity.' (Schwimmer 1993:253). I would also add the dimension of competition. To me, as already noted, male dancing pairs and the centrality of sugar cane express most immediately male creativity and potency, just what the staging of rimbu is endeavouring to achieve while reinforcing the significance of dualism once more.

In due course, I will elaborate on how this potency is achieved, but first I wish to return to the structure of the houses themselves, and eventually the major contrast in the physical manifestation of the long and short rimbu houses. The size of the rimbu houses is much greater than both the men's house and kabit house design. Obviously there is a simple pragmatic to this, as many more men go inside, but the size too has its symbolic importance for the potency of rimbu. Unlike the single upright pole of kabit, once again there are two large pinki poles, one at each end. Again practicalities rule this to a degree, given the far greater size and weight of the roof the poles support. Nonetheless, the issue is more than this, as the banana base of the rimbu house was also large but only featured a single upright pole. Once again, the singularity/descent and dualism/marriage themes are given architectural expression.

Moreover, while nowhere near as marked as, say, Umeda culture featuring in Gell's (1975) Metamorphosis of the cassowaries, space and descent/affinity do feature here. Men's houses and kabit are rectangular, while the architectural emphasis of a 'banana based' rimbu house is on narrowness, height and verticality, while rimbu houses are tall and roughly square. The analogy male:female::tall:short is pervasive in Anganen. For instance, the Anganen utilise trees as metaphors of descent groups; the roots and base are the ancestors and source of group strength, the branches are subclans, and the smaller branches and leaves at the top, the 'growing parts' as they are called, the living male members who will generate 'growing parts' of their own in the future. The 'banana-base' rimbu and the kabit houses reflect this, albeit through transformation. Both are relatively narrow but extended, one vertically, one longitudinally. In contrast, marriage is represented horizontally as women and for that matter crops such as sweet potato vines which lack the strength of trees, move across space, say between different local groups. That rimbu is about both descent and marriage organisationally and thematically may be seen in the tall but broad, the virtual and the horizontal, orientations of the houses.

This, I suggest, is related to a number of juxtaposing elements concerning gender. If the male/female and descent/marriage dimensions reflect how space may be conventionally gendered, the preferred but contrasting locations of the long and short rimbu do not. The short rimbu houses were constructed on higher ground than the long rimbu houses thus inverting the high:low::male:female dyads. (8)

There is one stark contrast between the design of the long and short rimbu enclosures. There is only one house with the long rimbu but two houses comprise the short rimbu. One house is called and engi 'house mother' and the other is iki and, the 'little or son house'. Pigs were exclusively sacrificed in the long rimbu and the 'house mother', while marsupials were offered in the little house. (9) The Anganen say that the name of the house reflects its size, both relative to the 'house mother', and in and of itself as its ritual leader would be stooped and cramped while in it. Onge is the standard term for small or little, although it is often emphasised by the suffix iki, hence ongeiki. Iki can also be used to denote 'son' (on occasion, 'child'). Little and son/child are compatible here, with the filial relationship of small child and parent an apt illustration of the physical contrast of the two short rimbu houses. The iki anda is thus the 'little house of the mother'. While I do not wish to fall into the trap of a linguistically based argument alone, and feel that the broader argument averts this, the names of the houses are interesting, especially as they too are semantic 'plays' on gender.

While there are no ostensible physical differences between the two main rimbu houses, linguistically they are opposed in terms of gender imagery, and this goes beyond tall being 'male' and short 'female' in Anganen classification established above. (10) And engi appears to be an inversion of engi anda, the birth/menstrual house where women were secluded during labour and their periods. Indeed the size and shape of the iki anda is strikingly similar to that of the birth/menstrual hut. Furthermore, rimbu houses and menstrual/birth huts are similarly located in what I have referred to as marginal space, neither of the village nor of the forest. However, the locations must be kept well apart, due to fears of pollution that would anger the spirits. A similar opposition occurs in the behaviour of men and women. Just as men would never enter a birth/menstrual house, fearing illness through pollution, women do not enter the space of the and engi, fearing attack by angry spirits or men. This is an expression of male power which augments the earlier argument about male exclusivity and prowess. Both gender and kinship idioms are being manipulated here. Featuring often in male only ritual, for example in the Melpa male and female spirit cults (A.Strathern 1979) (11) or in the Hua manipulation of gendered food categories in male initiation (Meigs 1984), these idioms are deployed to underpin the sense of male achievement: there are female images but no women, with the mother-son/child relation a most powerful representation of creativity.

I do not mean this in any psychoanalytic sense that has been proposed to explain the appropriation of female capacities in male exclusive ritual (e.g., see Feil 1987:172-3 for an overview). Men as much as women did and continue to most value adult women as mothers, in addition to being gardeners and pig raisers. Nor am I convinced by the Marxist interpretation put forward by Lindenbaum (1993) when contrasting the big man societies of the Western Highlands with initiation and male ritualised homosexual societies to their east and south. For Lindenbaum, the premise for making of big men is the mystification of female production, while the making of men through ritual features the mystification of female reproduction. I would concede that there is idiomatic appropriation of attributes conventionally associated with women, but not mystification any more than the pol-keri dance by male couples is a reference to homosexuality rather than to what Schwimmer (1993:253) terms 'homomorphism'. For Schwimmer, whatever form it takes, homomorphism gains its meaning from 'the complex system of symbolic transactions between partners', although I would add that this must be seen as only part of the overall rimbu complex when seeking where its meaning ultimately lies. Rather than resort to such explanations, more immediately, kinship and gender are not fundamental social principles, they are cultural and cognitive categories, being 'good to think'.

No explanation was forthcoming for why the short rimbu were constructed on higher ground than the long rimbu houses. Some suggested that the proximity of rivers flowing on the valley floors was suitable for the long but antithetical to the short. This may be related to the conceived remedial purposes of each house. The long rimbu was primarily concerned with disorders with which blood and flesh were afflicted such as malaria, festering sores, and perhaps leprosy. These are afflictions that are conventionally treated, in part, by fluids, such as the rubbing of pig's blood or plant excretions onto the skin. There thus may be a consistency between flowing water and the type of afflictions to be cured in the long rimbu, whereas deep seated afflictions of the internal organs or bones for which balms and salves have little or no influence were treated in the short rimbu. Unlike in Mendi where there was a progression from short rimbu active to long rimbu active (Ryan 1961), both houses were operational together. However, given that the Anganen, apart from leprosy, do not usually regard skin complaints as life threatening in contrast to the potential of organ afflictions, the short rimbu house has more telling implications in this regard.

This should not be taken as two distinct therapeutic procedures as their conceived functions overlap: the act of offering was tar more important than any connection between symptom and house. Indeed, in contrast to the likelihood that an individual would be the focus of kabit, the reasons for undertaking rimbu were not only as a response to widespread afflictions--not only to humans but to animals as well--or to failing gardens, there was a strong sense that such performances had manifold and intense positive effects more generally. In a manner similar to that which Jean Comaroff (1985) notes for the Tshidi, there is a quasi-mathematical procedure operating here: if affliction is a negative that sacrifice returns to normalcy, then rimbu can increase that conventionally (12) there. At best kabit, beyond the individual's recovery and the manifest nutritional effects of pork consumption, only has a broader consequence of limiting the likelihood of a further, immediate spirit attack (though to call it aprotropaic would be over-stating the issue).

One final aspect of the house design links these various points on space and gender to the broader issue of sacrifice. The 'house mother' and long rimbu house both feature a crudely constructed structure of woven rope called the rimbu-a, literally the 'rimbu man'. Remnants of previous sacrifices like pigs' tails would be attached to this structure. It is suspended directly above the wem cooking pit. The immediate linguistic identification of the structure, its construction, and its location are important. The roughness of construction befits disarray of the spirits' main domain, the forest, in contrast to the orderliness of villages and gardens. There is also a metonymic identification with human qualities of volition and openness to influence. Eating is the most apparent of these traits. The Anganen say that spirits eat the 'smell and smoke' of cooking flesh and blood, the latter here soaked up in a species of wild tern called engi. The spirits thus directly get what they desire while the ritual experts ask them to assist the living while the meat is cooked. Great things will come if this is successful. All meat must then be consumed by the men present, in contrast to kabit where the meat is taken from the house once cooked and the spirits appealed to. and then is taken home to be shared with females and boys too young to attend the ritual. Unlike standard pork distributions where men who host the occasion may actually forgo eating pork in order for others to receive large portions, all men must receive an equal share in rimbu. However. each single cut is small. with emphasis placed on numerous exchanges across the chamber before the men eat. This also emphasises male community, here one also based on consumption, thereby highlighting the intensified masculinity of rimbu over kabit despite both cults featuring male exclusivity.

The phrase 'good to think' is, of course, that which Levi-Strauss (1966) proposed for comprehending what motivates human interest in animal classification, and this point holds true for the use of animals in rimbu. Before returning to the significance of marsupial sacrifice during the performance of the short rimbu, I will discuss the role of pigs as they feature in both rimbu houses as well as in kabit. In exchanges such as bridewealth or mortuary compensation, pigs substitute for, and are symbolic of, people. Sacrifice is premised likewise on substitution: the smell and smoke of cooking pork for the afflicted's animating spirit. Moreover, if pigs replace cattle in the following quotation of de Heusch (1986:200), it is as pertinent to Anganen as offering among the Nuer: 'The sacrificial debt paid by the Nuer is taken from the very treasure which is supposed to perpetuate life: cattle are in fact bride wealth par excellence.' In Anganen, a principal reason bride wealth is given is to incorporate the waluma or 'womb blood', maternal substance that, in combination with paternal semen, is thought to create human life. Both bridewealth and pig sacrifice have the potential for life, but through differing avenues which accord with the status of the receiver: live pigs to living persons, dead pigs to the spirits, the most important being the ancestors, that is once living people.

To a degree, this may also help explain why marsupials have a role in the short rimbu, although the difference is important. Marsupials are said to be the 'pigs of the ancestors' in part due to their forest habitat. For the most part they are not widely deployed except in a prestation called yapu ta kala, 'marsupial hit give' (see Nihill 2002). Husbands are said to experience shame when confronted by the wife's close kin such as mothers and fathers. This is especially the case in situations that, in whatever way, suggest they are having sexual relations with this woman. To avoid this, especially when a woman falls pregnant with her first child, her husband may hunt and kill (but not skin, butcher, or cook) marsupials which the pregnant woman will present to her parents by way of announcement through yapu ta kala. There is a resonance of themes permeating bridewealth, yapu ta kala and sacrifice. Pigs belonging to humans feature in bridewealth which in part is to incorporate the life creating waluma of a woman from another patrilineal group. The marsupials of yapu ta kala are also about the continuity of the patrilineal group. The sacrifice and cooking of pigs and the 'pigs of the ancestors' in rimbu should also have life giving results. Thus, be it through sacrifice, bridewealth or to announce a woman's pregnancy, these are premised on similar themes of the substitution of animals for the benefit of those that offer them.

The preferred sequence for the short rimbu was for a ritual leader (who could not also be the short rimbu aesumba) to kill and cook the marsupials near, but not in the 'small/son' house. Then the sequence would start in the 'mother house' and the cooking of the pigs would begin, upon which the men would return outside, eat the marsupial meat, return inside to await the cooked pork for consumption. It seems that oscillation between ritual themes, even a sequence of the ritual being metonymic of the whole features in many rituals. (13) The progression goes from marsupial focus to pig focus and back again and again. (14) However, this is never a complete reversal. The sequence and its juxtapositions are premised on the binary opposition between raw and cooked and what I (Nihill 2001) have termed non-food and food (or nonedibility/edibility) elsewhere. The latter dichotomy notably features in certain Anganen exchanges such as the most prestigious gifts of pork in the yasolu ceremonial exchange being given in huge amounts and noticeably undercooked. To be edible, the recipients have to take them home, recook and then divide them to be fit for human consumption. In turn, this can be seen as part of a wider thematic dichotomy that pervades much of Anganen culture (see Nihill 1999, 2002), namely the opposition between tame, kumapi, and wild, kera, and their interconnection. This dichotomy and the shifts from wild to tame seems a major issue that frames rimbu and indeed sacrifice in Anganen more generally.

The best illustration of this interconnection is the socialisation of children, but parallel arguments could be put forward for the taming of pigs and dogs, the only two types of animals with which the Anganen are willing to share domestic and village space. For the Anganen, children are born 'wild', being dominated by their bodily functions and being incapable of truly acceptable social interaction. The taming process is achieved through three main avenues, those of talk, solid food consumption, and the capacity for proper reciprocity. For instance, it is a major milestone when a father feeds his child masticated banana, the first solid food children are given. While weaning does not occur for many years, the gradual introduction of more and more foods adults consume is seen as important. Similarly is the acquisition of language and the ability to engage in proper verbal and moral interaction.

This process is mirrored in the use of animals, most notably in the sequence of the short rimbu. First, and most generally, inedible flesh is transformed into desired meat for consumption. More specific is this shift in focus from marsupials to pigs. Marsupials are of the wild forest and are known as the 'pigs of the ancestors'. Most spirits likewise reside in the forest, except on these occasional forays into villages and their surrounds, and are regarded as kera. The shift from marsupial to pig in the short rimbu thus parallels socialisation, from wild to tame. Indeed this is more strongly emphasised in transition from live/wild to slaughtered/cooking marsupial, to slaughtered/cooking pig, to cooked, edible marsupial meat, to cooked, edible pork. Not only is this a shift towards tameness in the sequence, the identification of pigs with humans, here men, underscores the sacrificers' empowerment. However, one point tempers this to a degree. While domestic pigs featured most as sacrificial animals, in rimbu there was some emphasis on the suitability of wild boars, pigs that had for some reason gone feral, large and thus often difficult to control pigs, even the ambiguous, rare five toed pigs which continue to have ritual and exchange importance. Large or feral pigs, especially if they have undergone the inversion of taming, going from the village to the wild, add to this point made earlier on oscillation, but overall, sacrifice in the short rimbu highlights the shift from wild to tame, and this too, in this case, reinforces the point on male achievement. This concurs with the observation of Levi-Strauss (1966:225) that one object of sacrifice is to establish a viable relation of contiguity between opposites through mediation by intermediaries. It begins with the spatial location of the rimbu houses as between the 'tameness' of the village and the 'wildness' of the forest, and continues processually with the specific deployment of the animals. The wild-tame opposition of spirits-men is thus, despite these oscillations, one of an overall shift towards taming without ever dissolving the spirits-men opposition.

This process of mediation has additional significance in events where sacrificers are not motivated by affliction caused by spirits attacking people, that is, those that could be labelled increase rituals. Sacrifice in rituals of affliction is different due to the different status of the spirit-human relation. Increase rituals are motivated to enlist spirit support, that is, bring them closer to human affairs and thereby temporally lessen without negating the spirit-human opposition. By contrast rituals of affliction, be they for kabit or rimbu, take place when there is a lack of proper opposition between spirit and person which sacrifice aims to restore. Or to paraphrase Evans-Pritchard (1956:275) when discussing the Nuer God: 'It [sacrifice] is made to separate spirits and man, not to unite them. As such, Robertson-Smith's concept of sacrifice as communion is not relevant in Anganen. As Bell (1997:112) states: 'Communion implies that at a critical moment in the rite there is a union of the human and divine worlds: the offerer, the recipient, and the offering are understood to become together in some way, however briefly.' Even though they are thought to eat the same animals, they eat differentially: men the flesh, spirits the smell and smoke, with consumption ultimately underpinning opposition and not unity. In this sense, it is highly reminiscent of the yasolu pork exchange in Anganen, where givers and receivers cannot eat the same portion of an animal. The recipients are given the body and legs, while the givers are left to consume only the head with their families and perhaps close agnates after the completion of the event. Similarly, affines cannot consume meat of the same marsupial, as in the prestation called yapu ta kala, 'marsupial hit give' discussed above. To do so would render all parties ill and seriously threaten the health of the foetus it is intended to announce to a man's affines. This gift of yasolu pork and yapu ta kala marsupials is thus like sacrifice in articulating a relationship premised on profound difference.

Overall these three main principles for taming, namely talk, consumption, and reciprocity, albeit differing in ways of achieving socialisation in humans, are integral to how this is achieved, and to how sacrifice operates no matter what motivates men to undertake it. Here we can return to the flutes played in rimbu rituals primarily introduced earlier to preface the brief treatment of the cults' social organisation. The playing of the flutes is a mandatory aspect of the ritual and is much more than an act of deception. Rather, it is one feature of human-spirit communication. Firstly, sound is an important communicative medium in Anganen, and here I do not mean just speech itself. For instance, Anganen spirit mediums are able to interpret the 'whistles', kompres, of the spirits while in trance, a state likened to dreams which are also an important means whereby spirits communicate with the living. (15)

Dreams, whistles and the eerie sounds of the flutes are all, for lack of a better term, ethereal, perceivable but without substance. These stir the senses by conveying meaning, placing them on a par with the spirits themselves. Indeed, spirit attack is also on this dimension, as when spirits 'eat' or 'strike' the living, they attack the victim's animating spirit. In this regard, the non-substantial but nonetheless profoundly real dimensions have a number of manifestations in human-spirit interrelations and communication and this features in sacrifice as well. Unlike other Highlanders (e.g., the Kuma, Reay 1959:134), the Anganen do not think that the spirits eat the spirit of the sacrificial beast despite the obvious loss of its animating spirit, wesa, being a necessary prerequisite to cooking the animal carcass. (16) For the Anganen, the spirits consume the smell and smoke of cooking meat. That these dissipate while the meat remains for eventual human consumption is evidence of this. Detienne (discussed in Bloch 1992:29) argues the significance of cooking and eating in sacrifice, but in Anganen this is not simply for the human consumers of the carcass. In the most general of terms, cooking is, as Levi-Strauss (1975) argues, a fundamental cultural act, and thus adds a further dimension to the importance of taming in Anganen. Critically, the transformation cooking achieves is integral to the overall structure of sacrifice as a series of transformations-sick to well, increase, satiated spirits and so on.

More specifically, how the spirits eat and thus are influenced by the living augments these ethereal ways in which communication may be achieved. Writ most simply: the premise underpinning all sacrifice which is a response to affliction in Anganen is a substitution of the ethereal, the smell and smoke of cooking flesh, for the spirits to consume rather than the spirit of the living. However, like the quasi-mathematical basis of the efficacy of sacrifice noted earlier, this is the premise of all Anganen sacrifice, the ethereal dimensions of cooking meat, in order for the spirits to confer their power for the benefit of living persons, animals or gardens. While it is true that the Anganen have no term denoting health per se, at best inj yena, literally 'no sickness' prevails. However, rimbu is not simply about maintaining or restoring the status quo, it may be about increase as well.

Overall, then, sacrifice, be it in kabit or rimbu, broadly conforms to the arguments concerning communication or the gift put forward by Hubert and Mauss (1964) and Mauss (1969). It achieves its goals through a kind of reciprocity, the smell and smoke of the sacrificial beast for positive intervention or removal of improper spirit intervention in the affairs of the living. It is a process in which communication, be it through the sounds of the flutes, the opening and closing of the doors to the rimbu houses or direct appeal, figures. In his reassessment of the Maussian notion of 'gifts to gods', Godelier (1999:30, 193) argues that the equivalents of Anganen spirits are free to respond or not; or, in other words, there can never be the equivalence which is at least possible with gift giving between humans. Moreover, the spirits remain most powerful, whatever the outcome of any sacrifice. Perhaps the termination of rimbu is a reminder of this. Men have entered from the direction of the village, but leave to confront the forest and all it connotes as the principal domain of the spirits.

All this said, however, there are important differences between, and in the motivation for undertaking, kabit and rimbu. Kabit is always a response to the negative consequences of spirit power, and is only staged when serious affliction follows spirit attack. Rimbu in fact underscores spirit power even more if it is held following widespread illness to humans or pigs or serious problems with gardens. Failure to redress a negative situation, irrespective of the cult concerned, emphasises the power of the spirits over the living even further. But when successful, rimbu simply highlights that its potency is superior to kabit. If rimbu is motivated by widespread disaster and succeeds, it is on a scale far greater than success in kabit where one person recovers from illness. Moreover, kabit involves only one animal in contrast to the sacrifice of a number of animals in rimbit. Gi ven the pervasiveness of the logic of gift exchange in Anganen where economic quality underpins the articulation of power, this too underpins the greater potency of rimbu which, furthermore, engages a far greater spirit pantheon than the narrow patrilineal focus of kabit. Rimbu is simply more powerful, and thus empowers its leaders by creating a dependency of others, including women and children who have no knowledge of its procedures save that it is a crucial way of acting upon the world.

This contrast between the two spirit cults helps explain the strong impression I developed when talking with men about rimbu. Certainly they regarded the overall phenomenon with great reverence dedicated to the power of the spirits. Indeed, as with the opening of the doors of the rimbu houses, this matter is experienced with great trepidation if not outright fear. For the most part, however, it seems men acted within the rimbu house without showing anything that could be interpreted as piety. For the most part it appears men enjoyed their time in the house, some describing it is as ran, 'fun'. They are away from women and children, at least potentially may feel a sense of potency, augmented by the consumption of highly desired but rarely eaten meat.

In the lead up to the yasolu pig kill and pork distribution, both rimbu houses were abandoned, while kabit was not. This suggests that the members of the sub-clan are still at risk of attack but the reasons for undertaking rimbu have lost their previous importance. The leaders of the cults would terminate the houses' viability with namongo and men would drag both the pinki poles which supported the roof and the rimbu-a to be buried near the banks of rivers. Their decay would in turn invigorate the land as a special adjunct to the fact that the Anganen believe that spirits are responsible for fertility generally. Once this is complete, women, girls and uninitiated boys were free to enter the rimbu compound whereas previously this meant death. A festival lasting three days called rimbu res'u literally 'rimbu remove' would take place, before the pig kill would occur. Rimbu would recommence once the pork distribution was over.

Yasolu is said to be a time of 'health and happiness' which would suggest one reason that rimbu is redundant- no widespread sickness or garden failure. I suggest this rimbu active/rimbu inactive progression is indicative of a more common sense of pcriodicy found in the Highlands. The most famous instance of this is the distinct war and peace phases associated with the principal Maring pork festival, the kaiko (Rappaport 1984). However, what Andrew Strathern (1974:243) notes can be extended further when he states: 'Explorers, government officers and missionaries all note the presence of a duality in Hagen life: there was endemic warfare, yet also there were frequent festivals, in which pigs and shell valuables passed between individuals and groups...' For yasolu to succeed, widespread cooperation is necessary. It is organised on a multi-local group level meaning peace is a strong motivation. Even those outside any coordinating unit would in all likelihood concur as they are probable pork recipients and would rely on those of the coordinating unit to cooperate when it was their turn to stage the event. It seems a similar situation also takes place between spirits and people, thus rendering rimbu temporarily redundant. Yasolu is regarded as the principal single form of social action in Anganen (Nihill 1988a. 1988b, 1996b). As I have argued above, rimbu may be seen to be a form of male self-empowerment without ever leading to men dominating spirits. This is far more emphatic with yasolu. That rimbu will need to be reinstated is testimony of the power or tne spirits over tne living, out its abandonment sign--is tne niatus in Anganen male achievement.


Missionisation and the significance of Christianity among the Anganen is a complex issue. Here I largely limit the discussion of these concerns to their relation to sacrifice. The more things-change-the-more-they-stay-the-same thesis is somewhat banal and not altogether illuminating when considering the mission impact in Highlands communities (cf Clark 2000). I certainly do not wish to give any impression of some seamless, unproblematic continuity between Anganen beliefs and related practices concerning spirit cults like kabit or rimbu and Christianity. In fact, missionisation was fundamental to the overall dissonance colonialism wrought. However, especially for Anganen understandings of Christianity as a set of meanings and practices, it must be seen from the pre-existing cultural orientation its advent confronted, as well as from the aspect of its own unique powers.

As I have discussed at some length elsewhere (Nihill 1994, 1999), colonisation was particularly emasculating for Anganen men, and here I can add that mission pressure for them to abandon their sacrificial cults was part of their dilemma. In the local idiom men say they were rendered 'like women'. While they say the spirits still consume the smell and smoke of cooking meat, especially during feasts, older men especially retain grave reservations of spirit attack in the wake of cult abandonment. Apart from one small Protestant mission, the Christian Union Mission at Pareta, the entire region from the south of Mendi township to Nipa, three valleys away, was dominated by the Roman Catholic church until relatively recently. Certainly everyone in the vicinity of my field site in the mountains to the east of the Det mission station in the Nembi Valley claimed to adhere to Roman Catholicism. To a degree local men have taken advantage of the mission by utilising the male bias in the church itself, such as becoming catechists (even though a number surrendered these positions once polygynous). Nonetheless, the sense of disempowerment to which cult abandonment was central is crucial. Moreover, that women, previously precluded from major forms of sacrifice, have access to church equal to that of men makes this identification even more the case.

In contrast to the large number of small, fundamentalist and perhaps evangelical missions that feature in the Southern Highlands Province, the impact of the Roman Catholics appears at least relatively benign. Indeed, the mission quite intentionally encouraged dimensions of local culture such as supporting the practice of bridewealth which is perceived as legitimating marital unions and encouraging their persistence. However, I was told that if men, once baptised, were to even enter a spirit house, that God would render them immediately insane.

Both Protestants (particularly through the Summer Institute of Linguistics) and Roman Catholics endeavoured to translate the Bible (or at least key concepts) into local languages. The existing term Yeki was chosen for God. Clearly this was based on perceptions of similarity, most notably in Anganen thinking that yeki never had human existence. This figure is associated with the sky, the place of Heaven for the Anganen. Unwittingly, the missions' adoption of the term also resonates with the Anganen understanding of what God is and does. As noted previously, yeki could bring good fortune but equally it has negative connotations such as being thought responsible for leprosy. Despite the churches' emphasis on the positive attributes of God, the Anganen are most inclined to highlight His negative, indeed violent, nature. They offer little comprehension of the New Testament beyond what they call the 'big fight' between God/Jesus and Satan, but can relate in great detail stories from the Old Testament where God wreaks vengeance upon the living. Indeed, while no doubt underpinning the significance of God's own sacrifice of His son, the patrilineal Anganen find it somewhat incredulous that any father would knowingly do so. As a final example, God is also sometimes thought to cause the deaths of people such as the young child of a polygynous man or a person thought misappropriating funds from a mission supported trade store.

There is the expectation that all adults should volunteer opa, 'offerings', be it money, food (especially to help support the small local hospital run by Roman Catholic nuns), or wok sol, work for the good of one's soul rather than for monetary reward. The term opa for gifts of money, food and labour was, of course, the lingua franca Tok Pisin word imported by the missionaries themselves. However, I suggest it retains salience due to the Anganen linking it to pre-Christian spirit cults despite the obvious differences and consequences for the Anganen of these practices. Although giving offerings is seen as pleasing the principal Christian figures, the Virgin Mary is particularly appealed to by women such as when their children fall ill. Alternatively, men primarily see church attendance and opa as vital means of avoiding God's wrath.

All of the means of giving opa are open to women as well as men. However, men are most likely to have greater access to cash as the owners of garden land in which coffee is grown for sale. They are more likely to be wage labourers, and dominate bisnis (business) such as operating trade stores or passenger vehicles, and are recipients of money through gift exchange. Women do receive a small share of coffee receipts due to their labour inputs in picking and processing the beans for sale, but overall have far less direct access to money. As such, one way men endeavour to empower themselves vis-a-vis women in terms of the church is to give larger monetary opa. However, that women have access to, and the potential to influence, Christian spiritual figures cannot be denied, and in a manner beyond the emulation of women, men also seek to undertake alternative forms of empowerment. These do not amount to the power men gained through increase rituals in rimbu nor could they abandon Christianity in the way men did with this cult leading up to the yasolu pig kill, but they have their similarities. This is especially so in terms of the male experience of their actions.

I will start with the relation of the church to yasolu first. Yasolu is like other large ceremonial exchanges in Melanesia. It is the site for male self-aggrandisation, by big men in particular. As yasolu is the single most prestigious event staged in Anganen (Nihill 1988, 1996b), big men and those aspiring to be so, do not simply display and distribute wealth that others, including other male participants cannot emulate, but do so ostentatiously. One way they do this is to give shells or money to others, including their wives, while the pigs are tethered in the middle of the ceremonial ground before slaughter. While this is a public acknowledgement of the labour input of others, I (Nihill 1988b) argue elsewhere it projects a distance from producers so that men's achievements are seen to be all their own. Big men are likely to do the same with others these days such as invited politicians and, most importantly here. missionaries. This latter instance is once again termed opa, but its significance is not simply through the size of the presentation, it is also due to its radical contrast to the conventional opa of the Sunday service, say hundreds of kina in contrast to ten toea coins or vegetables.

The other event associated with Christianity, missionaries and the church that men may strategically deploy for their own empowerment is the opening of a new church. Apart from the central church on Det station, each Roman Catholic mission has a number of satellite churches. Catechists mostly perform the service in these, though occasionally the resident priest will do so. These satellite churches are primarily constructed using bush materials like conventional houses, eventually decaying and requiring replacement. Be it a church in a new location or one that has been rebuilt, alter its consecration, a feast is held in celebration. In contrast to the individual focus of yasolu, men such as subclan brothers may jointly purchase a cow to be slaughtered, cooked and its meat distributed to those in attendance, missionaries included. This too confers substantial prestige upon men. Scale is once again vital here, as is the fact that women and most men cannot emulate their achievements. Similar to yasolu, this is a time of 'health and happiness', and this form of opa is thought to bring widespread positive effects for the living, including those missionaries in attendance. It is also thought to be viewed favourably by God, The Virgin Mary, and Jesus as is the opa presented at yasolu. The single term of opa thus obscures the variation in motivation for its presentation, its perceived effects in terms of empowerment and relations with spiritual figures. In certain respects, then, opa, has its similarities to the dual reason for staging rimbu and its contrast with kabit. Kabit and rimbu held following widespread misfortune is men's reaction to the world acting upon the living, whereas rimbu undertaken when there is not widespread misfortune is men acting upon the world. While all are forms of male empowerment in relation to non-participating women and children who are dependent upon men's actions, rimbu staged for increase is exceptionally so. For convenience, I will only deal with these contrasting opa in their relations with God here. Conventional opa donated in the church or for wok sol is significantly motivated through fear of a vengeful God; opa given at church openings and at yasolu is, like rimbu may be, a form of male empowerment both politically in terms of prestige and spiritually in inducing favourable reaction rather than staving off the negativity which largely typifies Yeki.

When considering Christianity and Anganen culture, it is crucial to understand that there was no seamless contiguity with pre-existing beliefs and practices. Quite the opposite is in fact the case: as a key aspect of colonisation and its subsequent effects, missionisation caused great upheaval, indeed at times great confusion and dismay. However, one general point emerges from this brief consideration of Christianity and its possible similarities to the pre-Christian spirit cults that mission force abandoned. To an important degree, and despite mission hegemony, the Anganen, and here men in particular, reacted to these novel circumstances in terms meaningful unto themselves. Thus, it is important to stress that Anganen notions of the spirit world and how men deal with it substantially form how Christianity is understood and experienced. Depending on context, this may be both as the trepidation of powerful spiritual agencies and as possible forms of empowerment. One result is that, despite it being central to the feelings of the profound emasculation men endured with colonisation and beyond, they endeavour to use Christianity for their own empowerment in ways that are similar to, and also clearly distinct from, their actions in spirit house cults.


Earlier I noted that the term 'rimbu' featured in a number of Southern Highlands cultures such as the Mendi or Wiru. Moreover, Ryan's (1961:269-72) discussion of the adoption of rimbu in Mendi indicates that Anganen and Mendi share a historical link, both importing the cult from the south. What is interesting, however, is the at times pronounced differences in their respective version of this cult. Just the sequential dimension of the short and long houses in Mendi that is lacking in Anganen mentioned above will suffice as illustration. What this suggests is that, not only were these Southern Highlands cultures disposed to adopt novel, exotic cultural forms, they did so through their own innovation: not only were these cult forms adopted, they were adapted. That is, unlike the image evoked by concepts such as the Levi-Straussian 'cold society', cultures such as the Anganen were significantly predisposed to change. Oral historical accounts often feature their importing beliefs and practices from elsewhere, but they did so through an articulation with their own particular cultural orientations and preferences. Doing so produced syncretic forms that have their own dynamics, and thus help account for the differences between Anganen and Mendi that developed over a relatively short period of time. Despite the weight of colonial hegemony of which missionisation and the conversion to Roman Catholicism is a vital part, this point holds true for the historical production of the Anganen consciousness of what Christianity specifically means to them and their practices as katolik subjects.

Here I have considered only a few aspects of the relation of pre-colonial spirit cults and Christianity. Anganen domination by spirit agencies features in both the cults and Christianity. Equally, however, both present the opportunity for Anganen men in particular to exercise agency in ways that may empower them. Sacrifice and opa demonstrate this contrasting situation. On the one hand, kabit and rimbu are reactions to the power of the ancestors to cause misfortune to the living. Similarly, the giving of conventional opa of small amounts of money, garden produce and wok sol are largely contingent on the Anganen understanding of the Old Testament stories in which God wreaks havoc on the living that these gifts are thought to avoid. On the other hand, sacrifice for increase in rimbu or these ostentatious displays for great prestige, as with the donation of large amounts of money to Christian personnel at yasolu or the killing of a cow for consumption to celebrate the opening of a church are means of empowerment for Anganen men. This holds true even in the context where they were emasculated by colonialism and all it entails. Yeki as the Christian God and yeki as part of the Anganen spirit world that causes leprosy but may bring good fortune metonymically illustrates this contrast. Overall, both the pre-Christian spirit cults and contemporary context give rise to an ultimately Anganen-specific sense of being human in a world they may act upon but which also may act upon them, of power and powerlessness simultaneously.


BAAL, J. Van. 1975. Offering, sacrifice and gift. Numen 23:161-78.

BELL, C. 1997. Ritual: perspectives and dimensions. Oxford: Oxford University Press.

BLOCH, M. 1992. Prey into hunter: the politics of religious experience. Cambridge University Press.

BOWIE, F. 2000. The anthropology of religion. Oxford: Blackwell.

BURRIDGE, K. 1969. Tangu traditions. Oxford: Oxford University Press.

CLARK, J. 2000. From steel to stone: a chronicle of colonialism in the Southern Highlands of Papua New Guinea. Oxford: Oxford University Press.

COMAROFF, J. 1985. Body of power, spirit of resistance: the culture and history of a South African people. Chicago: Chicago University Press.

EVANS-PRITCHARD, E. 1956. Nuer religion. Oxford: Oxford University Press.

FEIL, D. 1987. The Evolution of Highland Papua New Guinea Societies. Cambridge: Cambridge University Press.

FRANKLIN, K. 1968. Dialects of Kewa. Pacific Linguistic Series C, number 5.Canberra: Australian National University Press.

FRANKLIN, K and J. 1978. Kewa dictionary. Pacific Linguistic Series B, number 10. Canberra: Australian National University Press.

GELL, A. 1975. Metamorphosis of the cassowaries: Umeda society, language and ritual. New Jersey: The Athlone Press.

GODELIER, M. 1999. The enigma of the gift. (Transalted by N.Scott). Cambridge: Polity Press.

HAWKINS, M. 1984. Gender symbolism and power relations: a reassessment of the Arab Kor cult of Mt.Hagen. Mankind 14:217-224.

De HEUSCH, L. 1986. Sacrifice in Africa. Manchester: Manchester University Press.

HUBERT, H. and M.MAUSS. 1964. Sacrifice: its nature and functions. London: Cohen and West.

JEBENS, H. 2005. Pathways to heaven: contesting mainline and fundamentalist Christianity in Papua New Guinea. New York: Berghahn.

JOSEPHIDES, L. 1985. The praduction of inequality: gender and exchange among the Kewa. London: Tavistock.

LeROY J. 1985. Fabricated world: an interpretation of Kewa tales. Vancouver: The University of British Columbia Press.

LEVI-STRAUSS, C. 1963. Totemism. (Translated by R.Needham). Boston: Beacon Press.

1966. The savage mind. (Translated by G.Weidenfeld and Nicolson Ltd.). London: Weidenfeld and Nicolson.

1969. The elementary structures of kinship. (Translated by J.Bell, J.von Sturmer and R.Needam). Boston: Beacon Press

1975. The raw and the cooked: introduction to a science of mythology. (Translated by J. and D. Weightman.) New York: Harper Colophon Books.

LINDENBAUM, S. 1993. Variations on a sociosexual theme in Melanesia. In G. Herdt (ed.), Ritualized homosexuality in Melanesia, pp:337-61. Berkeley: University of California Press.

MACDONALD, M. N. 1991. Mararoko:a study in Melanesian religion. New York: Lang.

MAUSS, M. 1969. The gift: forms and functions of exchange in archaic societies. Thetford: Routledge and Kegan Paul.

MEGGITT, M. 1965. The Mae Enga of the Western Highlands. In P.Lawrence and M.Meggitt (eds.), Gods, ghosts and men in Melanesia: some religions of Australian New Guinea and the New Hebrides, pp; 105-31. Oxford: Oxford University Press

MEIGS, A. 1984, Food, sex and pollution, a New Guinea religion. New Brunswick. Rutgers University Press.

NIHILL, M. 1986. Roads of presence: exchange and social relatedness in Anganen. Unpublished Ph.D thesis, The University of Adelaide.

1988a. Worlds at war with themselves: notions of the 'antistructure' in Anganen ceremonial exchange. Oceania 58:255-74.

1988b. The 'two men' of Anganen exchange: structure and concepts of the individual in the Highlands of Papua New Guinea. Mankind 18:146-60.

1994. New women and wildmen: 'development', changing sexual practice and gender in Highland Papua New Guinea. Canberra anthropology 17:48-72.

1996a. Alternating ontologies: exchange and ritual in Anganen history. The Australian Journal of Anthropology 7:275-298.

1996b. Beyond bodies: aspects of the politicisation and development of exchange in the south-west Highlands of Papua New Guinea. Oceania 67:107-26.

1999. Time and the red other: myth and the paradoxes of power in Anganen. Canberra Anthropology 21:66-87.

2000. Exchange as sensory experience among the Anganen. Oceania 71 : 110-28

2001. Not feasting with friends: the meaning of meat in Anganen. Oceania 71:265-78.

2002. Dangerous visions: the cassowary as good to think and good to remember among the Anganen. Oceania 72:258-74.

RAPPAPORT, R. 1984. Pigs for the ancestors: ritual in the ecology of a New Guinea people. [New, enlarged edition.] New Haven: Yale University Press.

REAY, M. 1959. The Kuma. Melbourne: Melbourne University Press.

REITHOFER, H. 2006. The python spirit and the cross: becoming Christian in a Highland community of Papua New Guinea. (Gottingen Studien zur Ethnologie Bd 16). Berlin: Lit Verlag.

ROBERTSON-SMITH, W. 1899. Lectures on the religion of the Semites. Edinburgh: Black.

RYAN, D.1961. Gift exchange in the Mendi Valley: an examination of the socio-political implications of the ceremonial exchange of wealth among the people of the Mendi Valley. Unpublished Ph.D thesis, The University of Sydney.

1969. Marriage in Mendi. In R.M.Glasse and M.J.Meggitt (eds.), Pigs, pearlshells and women: marriage in the New Guinea Highlands, pp: 159-75. New Jersey: Prentice-Hall.

SCHWIMMER, E. 1993. Male couples in New Guinea. In G.Herdt (ed.), Ritualized homosexuality in Melanesia, pp:248-91. Berkeley: University of California Press.

STRATHERN, A. 1974. When dispute procedures fail. In A. Epstein (ed.), Contention and dispute: aspects of law and social control in Melanesia, pp:240-70. Canberra: Australian National University Press.

1979. Men's house, women's house: the efficacy of oppositions, reversals and pairing in the Melpa Amb Kor cult. Journal of the Polynesian society 88:37-51.

STRATHERN and STEWART, P. 1999. The spirit is coming! A photographic-textual exposition of the female spirit cult performance in Mount Hagen. Ritual Studies Monograph Series, Monograph No 1. Pitsburgh.

STRATHERN, M. 1984. Subject or object? Women and the circulation of valuables in Highlands New Guinea. In R. Hirchon (ed.), Women and property, women as property, pp:271-300London: Croom Helm.

STEWART, P. and STRATHERN, A. 1999a Female spirit cults as a window on gender relations. Journal of the Royal Anthropological Institute, 5(3):345-60.

1999b. Politics and poetics mirrored in indigenous stone objects from Papua New Guinea. Journal of the Polynesian Society 108(1):69-90.

2001. Timbu wara figures from Pangia, Papua New Guinea. Records of the South Australian Museum, 34(2):65-77.

TURNER, V. 1967. The forest of symbols: aspects of Ndembu ritual. New York: Cornell University Press. 1969. The ritual process: structure and anti-structure. New York: Cornell University Press.

1977. Sacrifice as quintessential process: prophylaxis or abandonment?. History of religion 16:189-215.

VALERI, V. 1985. Kingship and sacrifice: ritual and society' in ancient Hawaii. Chicago: University of Chicago Press.

Van GENNEP, A. 1960. The rites of passage. (Translated by M.Vicedom and G.Caffee). Chicago: University of Chicago Press.

Michael Nihill

A posthumous paper


(1.) The ethnography of this paper was collected over twenty five months and generously assisted by The University of Adelaide.

(Oceania editor's note: The author had passed away in 2008 before he could resubmit a revised version of this paper. The editor, Jadran Mimica, prepared the original manuscript for publication by following selectively the comments made by three reviewers. Virtually all additions to and modifications of the original manuscript are confined to the footnotes. Accordingly, the paper published here is fundamentally the author's original version. I am grateful to Professor Andrew Strathern, Dr Pamela Stewart, and Dr Hans Reithofer for their comments on the original manuscript).

(2.) As such, sacrifice is not truly prophylaxis, only remedial.

(3.) Even if this is the case, Anganen sacrifice is not about expiation or as 'piacular' as Turner (1977:200) terms it. That is, it is not undertaken to amend prior immorality.

(4.) (Ed) Rimbu and kabit correspond to timbu and tapa in Wiru (see Stewart and Strathern, 2001).

(5.) (Ed) In Wiru this is 'toronea' in the context of timbu and 'lundu' means 'long'.

(6.) This seems the closest to a rite of consecration in rimbu. Certainly, the Anganen say that were these words not spoken on these occasions, then the ritual would fail: far from pleasing the spirits, it would anger them. However, once again I would err on the side of caution when trying to utilise ideas germane to religious thinking in the West, and treat rimbu first and foremost on its own terms.

(7.) As with many practices that may be labelled under the term ritual, kabit is far less clear cut in terms of the van Gennep tripartite structure Turner uses for his various arguments concerning ritual. That is, there is no formal, marked separation or aggregation. In certain respects it reminds me of the 'traditional' church Christian wedding: just when is the bride liminal? E.g., when wearing the dress, as all would label her bride, when she enters the church veiled as a symbolic statement of her lack of fixed identity, or when referred to only by her first name given that one potential mark of aggregation could be the conferring of her husband's surname? This, and kabit show the practice problems of utilising an abstract (and therefore simplified) theoretical model when interpreting ethnography.

(8.) There is no necessary historical continuity here given that the short rimbu replaced the 'banana-base rimbu'. The latter need not be built on higher ground, with height articulated through the architecture of this house itself.

(9.) However, there does seem some symbolic merging of the pig-marsupial then in the long rimbu as it features powerful stones called yapu, the generic name for marsupials. No reason for naming these stones yapu was forthcoming. There are no physical resemblances, unlike the stones in kabit for instance. Nor are the stones found in marsupial habitats, with most being located in gardens or even women's houses. This may be related to the argument concerning gender in the main text.

(10.) Once again, there is no simple historical continuity from the 'banana-based' rimbu to the short rimbu that the Anganen say replaced it. Bananas are important 'male' plants in Anganen thinking.

(11.) (Ed) See also Stewart and Strathern, 1999a, b; Strathern and Stewart, 1999. For a Marxist feminist interpretation of these cults, see Hawkins, 1984.

(12.) The editor finds the use of 'conventionally' puzzling. One may replace it with 'exponentially' perhaps.

(13.) One such instance is in the film Masai manhood concerning male initiation. Two aspects of the ritual concern respect and the shift in notions of authority in the sequence between the 'mothers' and the moran initiates. The narrator, Llewellyn-Davies, points out that respect is a vital dimension of Masai social relations including those between sons and mothers, yet the sequence is defined by licentiousness such as the suggestions of mother-son sexual intercourse, whereas the idea of violating the incest taboo is absolutely repugnant to the Masai. A classic joking relation, the sequence is repeated twice. In the first staging the 'mothers' are thought to win, while in the second the sons triumph. As the narrator notes, one of the important consequences of this rite of passage is the shift in authority of mother over son prior to it, to son having authority over mother once the ritual is complete. For instance, a man's mother will then look after his cattle herd. Symbolically, the 'mothers' winning denotes previous maternal authority, the sons winning as heralding the change the ritual will achieve.

(14.) Fell (1987) suggests that such things may point to earlier modes of production in which hunting featured far more than in the contemporary Anganen context in which animal husbandry dominates. A similar point was made by Nicholas Modjeska when commenting on my doctoral dissertation (Nihill 1986). While I cannot go into it in any depth here, it does seem that many aspects of rimbu entered Kewa from further to the south where hunting is a major male productive activity.

(15.) While some individuals are noted for dreaming more than others, unlike spirit mediums, any relatively mature person can have divination dreams (polu upa) and endeavouring to interpret dreams is a regular social activity. Sometimes the communication is direct such as a forewarning of sickness which should be met with preventative measures which are thought to 'tie' the animating spirit wesa to the body, others far more non-literal. An instance of the latter is to be covered in copious faeces may be the sign of wealth to come one's way.

(16.) (Ed) It should be noted that among the Karinj people further north (Reithofer, 2006:102), the officiant of a sacrifice may explicitly refer to the spirit's taking the pig's wesa when he offers some 'sacred' portions to it: As you have taken the [pig's] spirit [wesa], I am giving you the kidneys now. The reference of wesa is to the pig's blood poured previously into the oven pit (comments on Nihill's paper December 2007). As for the Anganen, they are not certain that pigs may have souls but are adamant that other animals such as marsupials or eels which also featured in sacrifice on occasion do not.

(17.) Here I follow the mission terminology of using the capitalised version to refer to the Christian God as opposed to yeki as the spirit previously discussed when considering illness and sacrifice.

(18.) (Ed) An observation by Hans Reithofer should be recorded here. One may question 'whether the Anganen actually associate the term sol with the English 'soul'. Among the Karinj, the expression wok sol is also used among the Catholics for what they call wok penans in Tok Pisin, but it derives from the Aklal Heneng term sol for pay, wages, income (sol pi = pay). Wok sol then is understood as 'the work payable/due to the church', which accords very well with the Anganen understanding of it as an 'offering' (comments on Nihill's paper, December, 2007)
COPYRIGHT 2010 University of Sydney
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2010 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Nihill, Michael
Article Type:Report
Geographic Code:8PAPU
Date:Nov 1, 2010
Previous Article:Marriage, love magic, and adultery: Warlpiri relationships as seen by three generations of anthropologists.
Next Article:The carnival of custom: land dives, millenarian parades and other spectacular ritualizations in Vanuatu.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |