Version classiqueVersion mobile

Pathways of change

Anton Ploeg

Chapter 3. Hubula

Texte intégral

1By discussing the Hubula prior to the Damal and the Lani, I depart from the west-east sequence that I started with my discussion of the Me and the Moni. I do so because there is reason to regard the way of life of the Lani as an offshoot of that of the Hubula. And as regards the Damal, it seems likely that they adopted several elements of the Lani way of life. Hence I proceed by discussing the Hubula first, then the Lani and then the Damal.

2The Grand Valley of the Baliem was, with the Paniai area, a focus of colonial interest. As I mentioned in the Introduction, ever since 1938, when Europeans had become aware of the existence of the valley, its dense population and the wetland agriculture practised there, the area had been a source of wonder. The name “Grand” Valley is fitting, since the centre of the area is a relatively flat valley floor, oblong, but not regularly shaped due to mountain spurs and side valleys of tributary rivers. The altitude of the valley floor is about 1650 meters (Peters 1965: 13, 1975: 2). It is about 60 kilometres long and 15 kilometres wide (Haberle, Hope & de Fretes 1991: 26). Downstream the valley narrows into a gorge through which the Baliem flows into the southern lowlands. At least part of the gorge is populated by Hubula.

Pict. 5. Wetland agriculture in the Grand Valley

Pict. 5. Wetland agriculture in the Grand Valley

Pierre Pétrequin took this picture in 1985, near the airport of Wamena, in mid valley. Sweet potato cultivation on the beds is in different stages. The picture was taken decades after first contact, but I have no reason to suppose that the Hubula had greatly changed their cultivation methods.

Source: Pierre Pétrequin & Anne-Marie Pétrequin.

3An important motive for the post-war exploration of the highlands had been to reach the Grand Valley (Van Baal 1989: 390-391). After the exploratory party led by Meyer Ranneft had attained this goal in 1953, the valley soon became a centre of missionary activity and, in the early 1960s, a centre for anthropological research. I refer especially to the Harvard-Peabody Museum expedition to the Grand Valley, led by film-maker Robert Gardner, in 1961. The team included another film-maker Eliot Elisofon, the photographer Samuel Putnam, the naturalist Peter Matthiessen, and a team of anthropologists, Karl Heider and Michael Rockefeller, as well as Jan Broekhuyse, who had spent the previous year in the valley as a colonial administrator. His superiors had seconded him to the team. At the time administrative control of the Hubula was more tenuous than the colonial authorities were prepared to admit. Accordingly, when it became known to the outside world that the expedition members had been easily able to film scenes of war, and of its aftermath, such as treatment of the wounded, the death ceremonies for a killed boy and the victory dances, the colonial administration was greatly embarrassed (Broekhuyse 1996: 141-142).

4The immense appeal of the ethnographic film Dead Birds (1963), coupled with other publications resulting from the expedition, such as Gardens of War (1969) by Gardner & Heider, and Matthiessen’s Under the Mountain Wall. A Chronicle of Two Seasons in the Stone Age (1962), introduced the Hubula to both academics and the wider public. Heider and Broekhuyse published doctoral dissertations on their Hubula research (Broekhuyse 1967, Heider 1970a). Heider further published a series of papers and, like Pospisil, a monograph in the series Case Studies in Cultural Anthropology; the latter went into a third edition in 1997. In addition the missionaries Myron Bromley, a linguist, and Herman Peters ofm, a trained anthropologist, produced scholarly publications. In my account of the Hubula way of life in this book, I draw primarily on the ethnographies of Peters, Broekhuyse and Heider, since they come closest to being “first-contact” ethnographers. To what extent their data are representative of all Hubula remains unclear to me. I return to the subject below. How many people lived in the Grand Valley in the early 1960s was not known with any certainty. Both Heider and Peters mentioned a figure of 50,000. But in the Dutch original of his dissertation, Peters called it a “rough estimate” (1965: 16, 1975: 5), and according to Heider it may have been wide off the mark (1970a: 60).

5The number of researchers working among Dani speakers, enabled the staging, in August 1962, of a conference to discuss their way or ways of life. At the time, the distinction between Hubula and Lani was not yet made. The conference was held in the Grand Valley, in Wamena, then the centre of the colonial administration in the highlands, and the papers and the ensuing discussions were published as the “Working Papers in Dani Ethnology” no. 1 (1962). Participants included administrators, anthropologists, linguists, and both Protestant and Catholic missionaries. Of the eleven participants, four proceeded to write a Ph. D. dissertation dealing with the Hubula, three with the Lani and one with the Damal, easterly

6neighbours of the Lani. Of these dissertations, five were in anthropology, two in linguistics and one in missiology.

7Having discussed the Me and the Moni in the preceding two chapters, I turn in this chapter, with the discussion of the Hubula, to people speaking a language belonging to what Silzer & Heikkinen Clouse called (1991: 25) the Dani sub-family of the Great Dani language family (see Table 2). I continue the discussion of speakers of that language sub-family with the Lani in chapter 4. The language of the Yali, discussed in chapter 6, belongs to the Ngalik-Nduga sub-family of the Great Dani family. Table 2 shows how Silzer & Heikkinen Clouse (1991, Map 8) subdivided it. A slight majority of the highlanders of western New Guinea spoke languages belonging to that family. In Silzer’s & Heikkinen Clouse’s view the Hubula spoke a single language consisting of a chain of three dialects, upper, mid and lower Grand Valley (1991: 25; see also Bromley 1973: 6). That the chain of dialects followed the southeast-northwest axis of the Grand Valley is in line with migrations along that axis. Such a hypothesis is supported by oral traditions of the Hubula that traced “the first appearance of mankind” most frequently to “Seima… at the top of the gorge through which [the Baliem] drains the Grand Valley” (Bromley 1990: 1-2; see also Broekhuyse 1967: 21, Heider 1970a: 140, 300, and Peters 1965: 20, 1975: 10) for accounts that largely but not completely agree with Bromley’s statement. In contrast to Silzer & Heikkinen Clouse, the editors of Ethnologue, 14th ed. vol. 2 (Grimes 2000: v, 472) have in a later publication divided the Grand Valley communalects into three languages. Not being a linguist, I am unable to decide in favour of either view. What matters to me is whether, and to what extent, the linguistic differences were mirrored by cultural ones. The extant ethnography is, unfortunately, inconclusive in this regard.

Table 2

Great Dani family of languages
Dani sub-family Ngalik–Nduga sub-family Wano, sub-family level isolate iii
Lani Hubula Hupla Walak Nggem i Yali Silimo ii Nduga
i. Hupla is spoken in the gorge area south of the Grand Valley, Walak and Nggem to its north. Hupla and Nggem have far fewer speakers than both Hubula and Lani (Silzer & Heikkinen Clouse 1991: 50, 69). The number of Walak speakers was then unknown, but it was presumably modest. ii. Silimo and Nduga are spoken south of the main range, and south of the Western Dani area. Also these languages have far fewer speakers than Hubula and Lani (Silzer and Heikkinen Clouse 1991: 69, 77). iii. Wano is spoken on the northern edge of the highlands, north of the Lani area. Silzer & Heikkinen Clouse estimated the number of speakers at 3,000 to 3,500 (1991: 83).

8Heider was not consistent about cultural homogeneity among Hubula in the early 1960s. The Harvard-Peabody expedition worked northeast of Wamena, on the edge of the valley, among people Heider called the Dugum Dani. In his dissertation as published, he wrote that “culturally and linguistically we may describe” the Dugum Dani “as Central or mid-valley Dani” (1970a: 8-9). But in his later case study, Peaceful Warriors, he was more outspoken and wrote “with some trepidation, I am calling this study Grand Valley Dani”, since “most of the important things which I am saying here are valid for the entire Grand Valley” (1997: 29). The titles of Peters’ and Broekhuyse’s doctoral dissertations (1965 and 1967 respectively) also suggest that each of these books concerns a single group only, yet their contents suggest great cultural homogeneity among the Hubula (similarly Broekhuyse 1996). Both Broekhuyse and Peters had previously worked in several parts of the valley: Broekhuyse southeast of Wamena (Broekhuyse 1996: 133-134) and Peters near Wamena and in several locations east of Wamena (1965: 12, 16-20, 1975: 1, 5-10). These locations are all in mid valley. Incidental remarks single out the people living in the “southern” part of the Grand Valley as culturally different, for instance with regard to the occurrence of anthropophagy and also to the sanctions applied to sexual relations with a member of the same moiety. In the south, offenders were to be executed (Broekhuyse 1967: 21). It has, however, remained unclear to me whether “southern”, as mentioned here, coincided with the area where, according to Ethnologue, south Grand Valley Dani was spoken.

9As appears below, peoples in the eastern highlands, including the Yali and the Eipo, ate the corpses or part of the corpses of slain enemies. By contrast, the occurrence of this practise, or of other forms of anthropophagy, in the western highlands is a matter of controversy. In an early publication discussing the work of orthodox Protestant missionaries, Russell Hitt characterised the Grand Valley as a “cannibal valley” (1962), on hearsay evidence. Peters rejected that name, since colonial newcomers had not observed the Hubula actually eating human flesh (1965: 105, 1975: 107). Nevertheless he concluded that the people may have practised anthropophagy in the pre-colonial past (1965: 97, 1975: 97). Largely in agreement, Heider commented that it may have been practised in the southern part of the Grand Valley in the 1950s, but that, in the mid valley where he had worked, people had practised it but had abandoned it before first contact (1970a: 128, 140-141, 1997: 126).

1. Hubula economy

  • 1 As number one, Brookfield ranked both the dryland and wetland cultivation in New Caledonia; he ran (...)

10The Hubula also impressed European observers at first contact by the ecosystem that they had modelled. The inhabitants practised wetland agriculture on the wide valley floor, using both irrigation and drainage. Brookfield visited the valley in 1959 and reported that the Hubula controlled “the water level through lateral dams in the system of ditches, and by systems of regional drainage carrying the outflow of whole drainage districts out to the river at controlled points” (Brookfield & Hart 1971: 114, Brookfield 1964: 22). Szalay summed up the process as “the manipulation and careful control of water by means of an elaborate system of ditches, barrages, levees and sluice gates” (1999: 29; also Soenarto & Rumawas 1997). The Hubula also cultivated the slopes surrounding the valley floor, in part by building stone retaining walls. The valley, including side valleys of a number of tributaries, was densely populated. Brookfield has ranked Grand Valley agriculture second in his comparison of the intensity of 44 agricultural systems in Melanesia (Brookfield & Hart 1971: 106, 114-115; also Golson & Gardner 1990: 409).1

11Hubula agriculture was productive. Frits Veldkamp, the administrator who established the first patrol post in the Grand Valley, in late 1956, and who had in that same year worked in the Paniai area (Van der Schoot 1996: 617-618), commented favourably: the “extensive gardens planted with sweet potato made a fertile impression, in contrast to the smallish, poor and stony fields that I knew from the [Paniai] lakes” (1996: 91). The yield from their gardens allowed the Hubula to maintain large herds of pigs which they used to foster a rich ritual life. Peters, who worked from 1957 to 1959 among Amungme, or Damal, south of the main divide, and from 1959 in the Grand Valley, has written that, among the Amungme, the Grand Valley was “pictured as a kind of paradise, where pigs abound” (1965: 75, 1975: 72). The early ethnographers Heider and Broekhuyse did not report food shortages occurring at the time of their research, and Peters commented on the very large numbers of pigs killed during the pig feasts that he attended in the early 1960s (1965: 147-148, 1975: 156-157). The productivity of their agriculture may account in part for the vigour of Hubula life, and the satisfaction the people derived from it at the time and the place the researchers observed it (Heider 1970a: 15). Yet Peters reported that food shortages did occur (1965: 71, 1975: 66-67; also Heider 1970a: 9). For instance, in 1964 serious flooding took place in parts of the valley resulting in emergency harvesting. Like the Me, the Hubula were inclined to blame the colonial administration for such misfortunes, in their case because it had vetoed warfare (Peters 1965: 71, 1975: 67).

12Hubula lived in palisaded hamlets with one men’s house and, I suppose, most often several houses for women and children. There were separate buildings for cooking and for penning pigs (Heider 1970a: 252-251; Peters 1965: 49-50, 1975: 43-44). Whereas houses among Me and Moni were rectangular and often were internally sub-divided into separate rooms, Hubula houses were round and had a horizontal sub-division with a single room at ground level topped by a sleeping loft. Often hamlets were joint, with, I understand, always separate entrances (Broekhuyse 1967: 29). When they contained sacred objects such as used in ancestor worship, men’s houses were out of bounds for females (Peters 1965: 52, 1975: 46). The absence or presence of such sacred objects signified a hierarchy among men’s houses, a topic I return to below.

13Since, in the course of the generations, the Hubula had deforested the valley floor and much of the surrounding slopes, they had to trade for forest products. In the eastern part of the valley, where the Harvard-Peabody expedition worked, trading relations were oriented towards the east, towards the Yali, as discussed below in chapter 6 (Heider 1970a: 26-27). Hubula also had to import their stone artefacts. Their provenance and distribution have been researched by the archaeologists P. & A-M. Pétrequin. I mention here salient features of their work, since Hubula derived many of their stone artefacts from the same quarries as did the Lani, the Damal and the Yali, all peoples discussed below. The Pétrequins’ focus was on stone artefacts primarily as used by the highlanders of western New Guinea, and hence not on a single ethnic group.

  • 2 Pétrequin noted one exception. But it took place post-1970, so long after first contact.

14The site of these quarries lies north of the main ranges of the central highlands, northwest of the Grand Valley, and north of the Yamo Valley currently inhabited by Lani. The area in which these quarries are located was sparsely populated by people known as the Wano (see Table 2, Pétrequin 1993: 88, 118-119; Temple 2002: ch. 5). From 1984 when the Pétrequins started their field research (2006: 11-13), they observed and documented quarrying and also the reduction of quarried stones to portable, but further unformed roughouts. Since the site is close to the territories of the Lani, they were in a privileged position to distribute the stone objects farther afield (Pétrequin 1993: 155). The Hubula had to rely on trade to acquire them (Pétrequin pers. comm. 2018).2 They had three categories of stone artefacts: firstly tools, such as blades for adzes and axes, secondly ye stones, used in exchanges of wealth, as in marriage payments, and lastly stones used in ritual. The ritual stones, habo, were inalienable (Heider 1970a: 287). Other wealth items put to use were pigs; long, narrow, tightly looped bands set with cowries; and long rectangular nets, probably looped by women (as implied by Broekhuyse 1967: 82). Unlike the Me and the Moni, the Hubula at the time apparently did not grade cowries. Broekhuyse commented that they accorded ye stones “small differences in value”, based on “color, size and form” (1967: 137). They used the different categories of wealth on different occasions and for different purposes: the looped shell bands were part of marriage and mortuary payments (Heider 1970a: 151-152, plates 21 & 22; 1972: 180; Broekhuyse 1967: 155-156). So were the ye stones. But my reading of the ethnography is that ritual wealth items formed a separate category, used primarily, if not exclusively, in ancestor worship. Pigs and pork were used in exchanges, in ritual, and also as welcoming gifts (Lemonnier 1993: 37; Ploeg 2001: 35-36). In addition, pork appears to have been a means to upgrade a party or a ritual. Lemonnier argued that such a categorisation of wealth, I suppose both as to kind and to use, contrasted with what obtained in a Big Man “logic”, with its “single sphere of exchange” in which “overwhelmingly… a single item of wealth” figured (1999: 326-327, 2002: 118-119). As appears below, his argument fits the Hubula way of life in which, in my view, Big Men were absent.

2. Hubula polities and leaders

15At the time of the early researchers, the Hubula were sub-divided over a number of “alliances”, in Peters’ expression “war alliances”, reportedly varying greatly in size (1965: 59, 1975: 53). Bromley, one of the first missionaries to settle in the valley, reported the existence of fifteen alliances in “the upper gorge and Grand Valley”, with membership ranging “from fewer than a thousand to two or three thousand or perhaps even five thousand” (1960: 242). Peters arrived at a different number of alliances. According to him there were at the time ten such groups on an estimated total population of 50,000. However, judging from the sketch map that he provided in the Dutch version of his book, he did not include the upper gorge in his account (1965: 58). He put the current upper limit of alliance membership at about 7,000 people (1965: 58-59, 1975: 53-54). Heider mentioned that the alliance among whose members he did fieldwork counted about 5,000 people (1997: 29). Broekhuyse estimated that this alliance could field from 1,000 to 1,200 warriors (1967: 60-61). He did not provide total membership numbers. In comparison with other highlands ethnic groups, both in western New Guinea and in Papua New Guinea, these were large groups. Alliances were composed of several “confederations” and occupied contiguous blocks of land that their members defended against hostile neighbours. Together they covered the entire Grand Valley, both the valley floor and the surrounding slopes, with strips of no-man’s-land between them. When confederations fought wars on their part of the border with the enemy, members of other confederations of that same alliance used to join in (Heider 1970a: 80).

16The fighting shown in Dead Birds was according to Heider “relatively low-key”. He called it “ritual warfare” which he contrasted sharply to the massive surprise attacks (1970a: 105). The latter were infrequent, but led to scores of casualties. Moreover they might entail the break-up of an alliance and the subsequent regrouping of confederations into new alliances (Heider 1970a: 119-121). Broekhuyse provided several instances of such changes occurring in the period 1935-1950 (1967: 68-69). The colonial administration had unwittingly built its post in the Grand Valley on land that had become vacant after its inhabitants had been routed in a war that Veldkamp, on the basis of oral information, described as “apocalyptic” (1996: 92; see also Peters 1965: 16-18, 1975: 6-7). Broekhuyse pointed out that fighting was almost permanent among the Hubula (1967: 217). According to Heider that was “likely” so (1970a: 122). It seems to me that mutual military support and the successful, joint conduct of war were major factors in the continued existence of alliances. Feil questioned the large scale of Hubula political organisation by pointing to its instability (1995: 39). Yet the constituent parts of a broken-up alliance apparently attempted to realign into newly constituted alliances. Of the three examples Heider provided of non-ritual warfare, two consisted of an attack on a confederation that could be attacked from several sides (1970a: 121-122). About the third instance, Heider lacked detailed information.

17Within each alliance, one man tended to be most prominent. In my view the ethnography has not made it completely clear why others perceived this man to be so prominent. Heider named four factors: skill in warfare, in the manipulation of wealth, the advantage of being the son of a leader and “charisma”, as perceived in a mild-tempered man (1970a: 92-94). Oratory did not count. In his case study, entitled Peaceful Warriors, he added: “important men have more to do with the sacred”, referred to as wusa (1997: 70). Broekhuyse argued that many leaders were sons of former leaders, but that also success in politics and war made for leadership (1967: 86-88). It betokened support by the ancestors, believed to be necessary for prosperity, fertility and success in dealing with enemies. Many such men had married a sizeable number of wives (Broekhuyse 1967: 123; also Peters 1965: 34-35, 1975: 27-28). In 1961, the alliance that then hosted the Harvard-Peabody team was led by a man called Gutelu, in Heider’s (1970a: 78) and Peters’ (1965: 114, 1975: 117) spelling. According to Broekhuyse, Gutelu was at the time in the latter days of his leadership (1967: 93-94). Earlier he had been the war leader of his confederation, while he had advanced his leadership by effecting a large increase in the membership of that confederation.

18Heider revisited the Grand Valley in 1970. At that time he attended the pig feast of which he has given the account (1972), to which I refer, in section 5 of this chapter. Since the early 1960s, the time of his major field research, the alliance then headed by Gutelu had violently split during an episode of secular war (Heider 1970a: 118-119). It was instigated, against Gutelu’s counsel, by a younger leader. In 1970 this man was the top leader of the reconstituted alliance. By then, however, the political situation had undergone a radical change occasioned by the forced incorporation of the Hubula into the Indonesian state. Gutelu had met the president of Indonesia, in Jakarta, and derived prestige from this outside recognition (Heider 1972: 175-176).

19I take it that Gutelu’s early stratagem to enlarge his confederation counted, in Paula Brown’s terms, as a “bold imaginative programme… for group achievement”, for her one of the hallmarks of an “extremely successful leader” (1978: 195). She coined the term “Huge Men” to refer to them. In 1961 Gutelu was indeed one of the outstanding leaders among the Hubula. The ethnographers related that there were only four to six such men in the entire Grand Valley, a number smaller than that of the alliances (R.J. Gonsalves 1960; C.J. Schneider 1962; Broekhuyse 1967: 94-95; Gonsalves and Schneider were successively Officer in Charge of the colonial administration in the Grand Valley in the period 1958-1962). Truly exceptional leaders were mummified (Broekhuyse 1967: 90, Heider 1970a: 166), unlike all other Hubula, who were cremated. The desiccated bodies were kept in a men’s house in an effort to hold on to the dead man’s capacities.

20The occurrence of Huge Men in highlands political life is to be expected given that men achieved their leadership positions to varying levels due to personal gifts and efforts. Moreover, since “hugeness” was a cultural phenomenon, it could take different forms in different cultural formations. It could also, for that reason, become manifest in the varying leadership types identified by Godelier and Lemonnier, such as for instance the tyranny practised among the Tairora by a man like Matoto (Watson 1967), the “despotism” that, according to Salisbury, was a recurrent phenomenon in many New Guinean societies (1964: 225-226) and the mediatorship of Kondom Agaundo, in the 1960s the first Simbu parliamentarian (Brown 1995: 249-251). Among the Hubula, “hugeness” seems to have been prompted also by the prominence of ancestor worship and the notion that successful men enjoyed ancestral favour. That, I suppose, is the reason that, in the presence of these most prominent men, other Hubula were deeply afraid (Schneider, pers. comm. 1994). When discussing the Lani, in the next chapter, I mention two other Huge Men. The ethnography, however, leads me to suppose that, among the Hubula, Huge Men were a permanent feature, in contrast to what obtained among the Lani.

  • 3 Peters used the spelling gain. But in this book, I adopt the spelling used by Broekhuyse and Heide (...)

21According to Broekhuyse (1967: 74-75) and Peters (1965: 82, 1975: 80), the Hubula used the term kain for their leaders. In partial disagreement, Heider has written that the term is used only in the southern part of the valley.3 The term referred both to a leader and leaders and to the behaviour they displayed. It is noteworthy that the term does not appear to have been used among the neighbouring Lani and Yahuli Yali, although they too were speakers of Dani languages. Nor does it occur in Dale’s provisional Lani dictionary (1985). That contrast seems to me to be another hint that the three ethnic groups were on divergent pathways of change.

22Confederations were more tightly organised than were alliances, and the early ethnographers did not agree on their constitution. They did agree that patrilineal clans or sub-clans were their main constituents, though. Whereas the sub-clans of a single clan lived in many cases spread over the valley, the clans or sub-clans constituting a confederation were localised. Against Heider, both Broekhuyse and Peters reported that the two main patrilineal clans or sub-clans of a confederation were united, paired, by multiple cases of delayed exchange of sisters and classificatory sisters (Broekhuyse 1967: 24-26, 31-32, 34-35, 46; Peters 1965: 29-30, 1975: 20-21). Heider (1970a: 71) seemed to deny the existence of a preferential marriage rule. I favour Broekhuyse’s and Peters’ reports, for two reasons. The first is that they, unlike Heider, provided numerical data on marriage patterns. The second is that other Western highlands ethnic groups, too, followed this preference. Of the groups discussed in this book I mention the Lani, the Damal and the Yali. In the perception of the Damal, Ellenberger wrote, “there are two different kinds of everything, and… it takes both kinds to make a viable whole” (1996: 99). The moiety division of the Hubula, most Lani and the Damal seems a case in point. Both Peters (1965: 163, 1975: 178) and Hayward (1980: 41) reported that, for Dani speakers, the moiety division had existed since the emergence of their ancestors from underground.

23Ellenberger noted in addition (1996: 100) that a tendency towards pairing has been reported among various ethnic groups in various parts of the highlands of Papua New Guinea. Notably Francesca Merlan & Alan Rumsey made it a major theme in their ethnography of the inhabitants of the Nebilyer Valley, west of Mount Hagen. Commenting on Lévi-Strauss, they argued that for him dual organisation followed from reciprocity, while for the Nebilyer the reverse was the case since for them “pairing [Merlan & Rumsey’s emphasis] is the elementary principle” (1991: 209), a statement that fits with Ellenberger´s comment on what obtained among the Damal. As was the case among Hubula, Lani and Damal, pairing among the Nebilyer occurred with regard to “named, quasi-agnatic identities” (1991: 34), strongly resembling the (sub)-clans that were basic to Hubula, Lani and Damal social organisation. However, while, among these western New Guinea ethnic groups, important, if not the most important, instruments of pairing were moiety organisation and/or preferential marriage, among the Nebilyer this was not the case. On the contrary, for them those “quasi-agnatic identities” that they considered to be or to have become paired were also too close to intermarry (1991: 39, 82-83). It is noteworthy that anthropologists have reported such a tendency for pairing, although expressed differently, among ethnic groups that were not in direct contact with one another and that to all appearances had only in common that they spoke remotely related languages belonging to the Trans New Guinea phylum. Moreover, in his comparative study of seven South Coast New Guinea peoples, Bruce Knauft points to the “elaborate dual organisation” (1993: 220) that these peoples had developed. Most of them, too, spoke languages belonging to the Trans New Guinea phylum.

24The ethnography offers only incidental remarks about the historical depth of the Hubula practise of exchanging sisters and classificatory sisters between confederated sub-clans. Broekhuyse reported (1967: 86) long-existing marriage relationships between two sub-clans. And Peters mentioned, by implication, (1965: 106, 1975: 108), the long-standing existence of two such pairs. So it seems only likely to me that the practice has existed for several generations, as happened among neighbouring Lani, and that as a result the two paired sub-clans were linked by a dense net of consanguineal kin ties. The Hubula combined a wedding with a series of marriage payments (Heider 1972: 180) and ultimately a mortuary payment. Marriages with members of other confederations belonging to the same alliance did occur. Marriages with members of other alliances occurred also, but less frequently. The ensuing payments consisted of the main wealth items of the Hubula: pigs, looped bands set with cowry shells, ye stones, oblong, smooth, slatey stones (Heider 1972: 179-180) and long looped nets. As appears below, the Lani used the same set of wealth items. Intriguingly, and also against Peters’ reports, Broekhuyse mentioned a preference for direct sister exchange. The reasoning he reported contrasts markedly with what Hylkema argued with regard to the Me (see above, ch. 1 section 3).

Marriages by [sister] exchange are seen as unions which create an extremely solid relationship between the lineages and the exchange partners since the kin relationships are as it were doubled. The exchange partners are among others the mothers’ brothers of their respective children (Broekhuyse 1967: 34).

25However, he did not mention the frequency of this practice. Yet, the Hubula seem to have viewed sister exchange as an attractive option.

3. Hubula warfare

26The capacity of Hubula confederations for common action mattered greatly given the frequency of war among the Hubula. Just as the Me had caught the imagination of Europeans by the eagerness with which they engaged in cowry transactions, the Hubula did so by their fierce fighting. The two ethnic groups appeared to display two contrasting cultural orientations that at first overshadowed what became known about other highlands groups, even though field research among these other groups had started in the late 1950s and was well under way at the beginning of Indonesian rule in 1963.

27Throughout the fifteen months of his stay in the valley, from 1958 to 1960, the Officer in Charge, Veldkamp, was confronted with war and violence. In retrospect he wrote:

I remember the Valley as a place full of tension. Often a breeze started blowing around noon. Then, the rattling of the aluminium roofing of our sheds sometimes mingled with the yelling and war chants that came to us across the open plain. Very occasionally, small groups of warriors scurried along our camp, with their redoubtable, three meter long spears casually on their shoulders, en route to the field of battle. Always there were rumours of armed conflicts or raids which, in our view, showed little heroism. The reports about the number of people killed or wounded during these fights were hard to verify. Victory dances after a successful battle lasted for days; the refrains that were repeated for hours on end, got finally on our nerves. […] Of course, the continuous tension, the ignorance of what the next day might bring, threatened to influence the mood among our own group. Fortunately, virtually all its members were impervious to panicking. One policeman […] asked for, and got, a transfer (1996: 93).

28At the end of his tenure, Veldkamp advised his superiors, in his Handover Report, to be cautious with the suppression of warfare. They went against that advice, which given the international conflict over the administration of west New Guinea was inexpedient, but Veldkamp’s successors were unsuccessful, notwithstanding sometimes harsh measures. Warfare continued (Peters 1965: 113, 1975: 116). Lemonnier concluded, in my view correctly, that the Hubula had created a “warrior society” (1991: 25).

29For the Hubula conducting war was a religious duty. In Heider’s pithy expression, drawing on Von Clausewitz, it was “the continuation of religion by other means” (1970a: 130). And by religion Heider meant primarily ancestor worship. As mentioned, he distinguished between ritual and non-ritual warfare (1970a: 105-114). By the former term, he was referring to the regularly occurring battles on special battle grounds, as shown in Dead Birds, and with “secular warfare” to the massive surprise attacks, even though he did not demonstrate that these attacks were not related to religion. Recently Roscoe has pointed out that also during “ritual warfare” warriors did attempt to kill each other, and sometimes succeeded in doing so (2011: 57). For Broekhuyse, Hubula warfare was a sacred, ancestral assignment that people had to carry out lest they lose ancestral support and suffer personal and economic misfortunes (1967: 217; similarly Peters 1965: 112, 1975: 115-116). Killings had to be answered by killings and failure to do so resulted in despondency.

30In the course of the expedition, the hosts of the Harvard-Peabody team, the Wiligiman, engaged in a series of fights with the neighbouring Wittaia. Broekhuyse has provided (1967: 232-267) a lengthy account of the hostilities that covers the period from late April until early September 1961. In the first phase the Wiligiman had the advantage, since several Wittaia had been killed. But in the course of the weeks, the Wiligiman suffered two losses and, worse, were unsuccessful in revenging themselves.

31By the second half of August, hence in a short period of time, Broekhuyse wrote:

the urge to kill an enemy had grown to a collective obsession. The Wiligiman had to kill, and without delay. On the one hand they felt that the ghosts of the ancestors were withdrawing their support, and that they had to fend for themselves and likely were thwarted. That frightened them and paralysed their urge to fight the enemy. On the other hand two of them had been killed and it was the ancestors’ wish that these killings were revenged (1967: 265-266).

32Broekhuyse argued further that an enemy killed signified ancestral approval. It was not merely an act of revenge on the enemy (similarly Heider 1970a: 130).

33In their history of the Enga in the western highlands of Papua New Guinea, Polly Wiessner & Akii Tumu discussed the “great ceremonial wars”, which occurred in central Enga five generations before their research in the late 1980s and 1990s and which had been discontinued by the time of first contact (1998: ch. 10). “These spectacular, semi-ritualised ‘tournament’ wars were fought repeatedly over generations… at ten to thirty years intervals” (1998: 265, 267). They were waged especially in times of prosperity when people had the means to entertain and in a later stage reward their allies. So the wars entailed extensive exchanges, socialising and courtship leading to marriages and so to further exchanges. Wiessner & Tumu argued that Enga instituted the Great Wars to escape from the “vicious wars of central Enga after the introduction of the sweet potato [that] had disrupted relations to land, balance of power, and trade routes.” The Wars rallied larger coalitions, held together by exchanges, and aimed at display of force rather than defeating the enemy (1998: 266-267).

34Like the Enga Great Wars, the Hubula alliances contained fighting, at least fighting between their constituent confederations. In other respects, the two institutions differed. Among the Hubula the recurrent fights between neighbouring, hostile confederations belonging to different alliances displayed force, the resolve to defend resources, and the readiness of friendly confederations to assist each other; but, unlike the Enga Great Wars, they were not large-scale hostilities. Another point of difference seems to be that they also aimed at maintaining or restoring an even death tally and so provided the opportunity to achieve that goal by killing one or more opponents. Alternatively, the members of an alliance celebrated their union by means of the ebe akho, an institution that consisted of life-cycle rituals. These rituals provided the context for mutual transactions, including compensation for slain allies.

4. Ancestor worship

35Just as the colonial and post-colonial administrations were long unsuccessful in bringing Hubula warfare to an end, the missions were equally unsuccessful in converting them to Christianity. Orthodox Protestant missionaries had started work in the Grand Valley in 1954. They were followed in 1958 by Catholic priests from the Franciscan order (Sloot 2010: 240). Over the decades, the Franciscans became the more successful proselytisers, in part because they allowed Hubula to keep practising parts of their pagan religion, in particular ancestor worship (Butt 1998: 101; Sugandi 2014: 33). Leslie Butt did fieldwork in the mid-1990s and was told by an “informant” (she did not make his/her identity clear): “One hundred percent of people in this valley still believe in ancestors, no matter what religion they say they belong to” (1998: 169). And Yulia Sugandi was able to describe ancestor worship on the basis of her 2007 field research.

36Peters wrote (1965: 167, 1975: 182), and Broekhuyse (1967: 97, 289) and Heider (1970a: 134) agreed, that Hubula religion focused on the ghosts of the ancestors. Both Peters (1965: 88-89, 107-108, ch. 6, 1975: 87-88, 109-110) and Broekhuyse (1967: ch. 6, 231) devoted sizeable sections of their books to describing the rituals that formed part of ancestor worship. And Heider claims that “all their ceremonial activity and all their pigs are used in placating the ghosts” (1970a: 138-139). Also if this statement is an exaggeration, it still points to the predominant use of pigs in worship, instead of in exchanges between humans. As mentioned above, the ethnographers made it clear, too, that conducting war followed directly from such worship. Moreover, according to Broekhuyse (1967: 289), the ghosts of the ancestors participated in the daily life of the Hubula. They assisted them in all their activities. The people had also to deal with the ghosts of dead enemies who attempted to harm them (1967: 213).

37A central concept in ancestor worship was wesa, which referred, following Broekhuyse’s definition to the “sacred, everything that is related to the spirits” (1967: 298). For Heider wusa (his spelling), corresponds to taboo and mana (1970a: 136-137). And Bromley, based in the south of the valley, wrote: “‘That’s tabu (wesa); aren’t you afraid?’ So the observer hears children continually warned in the Grand Valley” (1960: 237). However I mention the concept especially since, among the Lani, it appears to have been either absent, or not prominent. Among the Bokondini Lani, the word for taboo was maga. Just as the word kain, wesa, or wusa, does not appear in Dale’s 1985 Lani dictionary, whereas maga does, with the meanings: “forbidden, don’t” (1985: 79). As appears below, the Yahuli Yali do employ the term. The use, and nonuse, of this term also suggests the divergent paths of change that these ethnic groups were following.

38Hubula transactions with ancestors were mediated by wealth items, especially the sacred stones, and also a few other sacred objects, that Hubula men kept in closets in their men’s houses. Given the predominance of the stones among these artefacts, I refer to them below as “stones”. These were similar to the ye stones that were part of bride wealth and they represented individual ancestors (Peters 1965: 111-112, 1975: 115; Broekhuyse: 1967: 98-99, 211; Heider 1970a: 286-287). Apparently there were two categories, stored in separate men’s houses and figuring in separate rituals. The first category was kept in a special men’s house, the “war house”, of a sub-clan. According to Broekhuyse (1967: 123), these implements were associated with deceased great, successful war leaders. Slightly differently, Peters opined (1965: 159-160, 1975: 173-174) that they were there “for” ancestors killed in warfare. The stones of the other category were stored in other men’s houses. They were to promote general welfare (Broekhuyse 1967: 127). The stones were objects of worship, and Hubula men cleansed and invigorated them, the latter by rubbing them with pig fat. This treatment took place at intervals of about six months. The rituals were observed by Peters (1965: 88-89, 107-108, 1975: 87-88, 109-110) and Broekhuyse (1967: 104-105, 125-126). They lasted two days and were attended by men only. In his account Peters wrote that, on the third day, the land of the sub-clan was taboo and not to be worked on until late in the afternoon. Pigs were not to rout it. The ethnographers did not mention how many men attended the two rituals. Broekhuyse was of the opinion (1967: 127) that they strongly resembled each other. The main difference seems to be that the rituals dealing with stones kept in the “war house” concerned a larger group of people.

39In the above I briefly mentioned exchanges, or “transactions”, in the sense used by the contributors to the volume “Great Men and Big Men”. I refer to the killings in warfare and the marriage payments. In both cases the Hubula had encompassed them in institutions serving a common interest, namely ancestor worship and the continued existence of the confederations, in tandem with the alliances the most important groups protecting their members. In my perception the exchanges were subordinated to these institutions. Since Broekhuyse argues that a killing signified ancestral approval, I conclude that acts of violence in warfare were not primarily transactions between the warring parties. Warfare in the Grand Valley, in Broekhuyse’s terms, “was not about territorial conquest, nor about killing as many enemies as possible; it was about maintaining the intense relationship between the living and the dead” (Broekhuyse, 1996: 144, see also 1967: 231, 289).

5. Other rituals; funerals and ebe akho

40Other Grand Valley rituals, too, were encompassed in institutions serving common interests, and in the following sections I briefly present two of them: funerals (Heider 1970a: 146-167) and the ebe akho, the “pig feasts” (Peters 1965 and 1975: ch. 5; Heider 1970a: 162-163, 1972). Hubula had arranged funerals in four stages. The first immediately followed death and took four to five days (Heider 1970a: 147-158). On the first day the corpse was cremated in the presence of a sizeable crowd – Heider did not give an estimate. People brought the main wealth items as gifts, pigs, shell bands and ye stones, the same as those used in marriage payments. Heider was informed that the “basic pattern” was that members of the deceased’s moiety brought pigs and those of the other moiety “shell goods.” He reported that this pattern was, however, not always followed. Those present ate the pigs and those who had contributed them received some of the shell goods. Narrow, looped bands set with cowries were distributed to individuals; Heider does not mention why they received them. He pointed out that these gifts were “contribution[s]… towards placating the ghost of the dead person, and ghosts in general” (1970a: 150). Broekhuyse commented (1967: 152-154) on the large size of these payments, also compared to marriage payments (see also Heider 1972: 180). In the course of the following days, the remains of the corpse were collected and disposed of; finger phalanges of several girls were chopped off, large string bags were distributed – a women’s affair –, a small, looped bag set with the tails of the butchered pigs was hung on the inside wall of the men’s house, and the ghost of the deceased was driven away. Heider stressed that the events had the “explicit function of placating the ghost of the dead person” (1970a: 158).

41The second stage of a funeral took place four to six weeks later and lasted three to four days. On the second day pigs were, again, given and eaten during a ritual meal. The ethnographers did not report who contributed the pigs. The third stage combined the rituals for a number of persons deceased since the last pig feast. Heider observed one such ritual. It was celebrated by three confederations together. Again the attendants brought gifts, and Heider noted that they seemed “more extensive” than at the earlier phases. He supposed that the goods brought for exchange during the two earlier stages were initiatory gifts for the major contributions of the third stage. All these gifts were assembled in bundles in the settlement of the most prominent man of the three confederations. He inspected what must have been a major outlay of wealth, shouted the names of the deceased, each represented by a bundle, and thanked the contributors formally, after which they returned, with the goods they had brought, to their own settlements. Heider did not say what people did with them afterwards, but presumably they were distributed, locally.

42The fourth stage of a funeral formed part of the ebe akho, the “pig feast” that I turn to now. Heider observed the performative phase of the feast in 1970, during his brief revisit. Earlier, Peters had observed, in the same alliance, the 1964 predecessor of the 1970 feast. In the course of his research, from late 1962 to early 1964, he was present during three such feasts, celebrated by the three alliances in which he had done fieldwork (1965: 147-148, 1975: 156). He included a composite account in his dissertation (1965, 1975: ch. 5). In most respects, his account concurs with that provided by Heider. Heider characterised the feast as a “rite of passage and of intensification.” The dual characterisation concerns a distinction between, one, ritual serving the passage of an individual from one stage in his life to the following, as conceptualised by Arnold van Gennep, and, two, ritual emphasising “the needs of society”, as conceptualised by Eliot Chapple & Carleton Coon (Heider 1972: 194).

43The ebe akho was celebrated by the entire alliance every three to six years (Heider 1972: 171; Broekhuyse 1967: 127; Peters 1965: 114, 1975: 117). It took place during a time of peace, most likely a truce, declared by a taboo on pig killings announced by the most prominent man of the entire alliance. Apparently he had the power to time the announcement at his pleasure (Heider 1970a: 163-164; Peters 1965: 114, 1975: 117). The rituals were preceded by a transfer of wealth items compensating the kin of allies killed in fights that took place in the period between consecutive pig feasts (Peters 1965: 107-108, 1975: 109-110). Broekhuyse confirmed (1967: 134-135) Peters’ account in almost every respect. However he added that these transfers did not constitute compensation, but merely excused the protagonists in the fights during which one or more of their allies had been killed. Since the announcement of an ebe akho ushered in a period of peace, the protagonists were not in a position to revenge the killings of their allies. Hence, according to Broekhuyse, the transfers may have only postponed revenge killings. Heider did not mention this ceremony in his ethnography. As appears below, he was of the opinion that compensation for a killed ally was part of his mortuary payment.

44As is the case among other highlands groups (Lemonnier 1991: 10), the ebe akho combined several ceremonies, in this case three life-cycle rituals: the wedding, the final stage of the funeral, and initiation, of a sort, for boys. They occurred in the sequence mentioned, but overlapped in time. In this case as well, two of the three rituals lasted more than one day. Weddings started with clothing the brides in their wedding skirts. The process took place all over the alliance territory, in the settlements where the girls were living. Simultaneously wealth transactions took place, with items similar to the ones in use during funerals. Four days later, the brides were fetched by a party of women from the groom’s settlement. Apparently almost all weddings took place only during the pig feasts (Heider 1970a: 71; Peters 1965: 33, 1975: 25).

45The second ritual was the concluding part of the funeral rites. Again a large number of pigs was killed, butchered and cooked. The “skins” were brought over to the settlement of the most prominent man of the alliance. There they were laid out so as to memorialise the “most important people who died since the last pig feast”. The ritual drew an “immense crowd.” Heider referred to estimates of five to eight thousand people (1972: 183). The figure suggests that people from other alliances also attended. Special attention went to the people killed by the enemy during the previous years. For Heider (1972: 182) it was the most important ritual of the feast. He adduced two reasons: firstly it concerned the entire alliance, and secondly, it aimed at placating the ghosts, as did all major Hubula rituals.

46The third and final part of the feast was an initiation of boys. This ritual made apparent the existence of “moieties”, the term used by Heider although he, with reason, wondered about its applicability (1970a: 62-63). The ritual concerned primarily sons of men belonging to the Waia moiety (Heider 1972: 184-185). Boys belonging to the Wida moiety were not initiated. By going through the initiation, the boys “became” Waia. So apparently they were not born into that moiety. But that is not the only reason I wonder about the applicability of the term moiety. Another is that at least a number of Hubula acknowledged the existence of a third category of men, half Wida, half Waia (similarly Peters 1965: 134, 1975: 140), who were also initiated. Heider wrote (1970a: 65) that Wida men could choose to become part Waia. Hence I conclude that Lemonnier’s opinion (1990: 85) that, in the New Guinea Highlands, initiation led to the creation of groups of co-initiates and of warriors did not apply to the Hubula, given that the ritual concerned the Waia, and excluded the Wida from all but one of the proceedings. Nevertheless, Wida men could become great warriors and even leaders (Peters 1965: 146, 1975: 154).

47The initiation took far longer than the other two rituals that formed part of the ebe akho, and included numerous episodes of which I mention the following. Away from the settlements, a sacred compound had been built where the boys had to spend several days. While there, they were ritually cooled and cleansed. Adult men of the two moieties fought each other in a mock, but nevertheless rough battle. This was the only occasion on which the Wida had a part to play. On the day of their return to their everyday settlements, the boys were first ritually heated. The proceedings clearly fitted Van Gennep’s tripartite scheme for rites of passage. Heider’s account includes the text of a song sung during the proceedings (1972: 189 n.). The language used was different from that spoken in the Grand Valley and unknown to him. I mention this song, since its text included the term wid, Heider’s spelling, which was a term for initiation among a number of ethnic groups in the highlands of western New Guinea. The term also occurred in the name of one of the constituent episodes of the ritual (Heider 1970a: 65, 1972: 186). Given that initiation was practised by eastern highlands groups, among them the Yali, the Eipo and the Ngalum, while in the western highlands that was the case only among Hubula and several Lani groups, I wondered whether the local Hubula had adopted the language used in the initiation song from easterly neighbours. But neither Siegfried Zöllner (pers. comm. 2015), who worked among the Yali, nor Volker Heeschen (pers. comm. 2015), who worked among the Mek-speaking Eipo and Yalenang, was acquainted with the words used.

48The ethnography, based on research carried out in the 1960s, did not mention compensation for killings. As far as enemies killed were concerned, it is not surprising, since compensation seems hardly “compatible” (Lemonnier 1990: 146, 1991: 27) with the prolonged state of hostility between enemy groups. However, as regards people killed while supporting their allies, Heider merely mentioned that allies might make a disproportionally large contribution to the death payment (1970a: 113), while Broekhuyse and Peters reported, as mentioned, payments to allies during the months preceding the pig feast, payments which, according to Broekhuyse, should not be seen as compensation.

49Comparing the Hubula ebe akho with the Me yuwo, it appears that an important component of the latter was concerned with the manipulation of both home-produced and imported wealth items. The Dani feast was built on what Wendy James, in her The Ceremonial Animal. A New Portrait of Anthropology, has called the “trajectories of sex/gender and of age” (2003: 156), including age beyond death. Both feasts were large scale, but participants in yuwo appear to have come from different polities, while the Hubula feast was celebrated by a single polity and confirmed its existence. These differences in orientation also suggest the different pathways of change that the members of the two ethnic groups were following.

50In their reconstruction of the recent prehistory, up to first contact, of the Enga in the western highlands of Papua New Guinea, Wiessner & Tumu (1998) noticed a sharp contrast between the “most significant social events” as they occurred generations ago and those of the era just before contact. The former were dances during which “cosmetic oil, shell ornaments, plumes, net cloth, and ceremonial axes”, all nonagricultural products, were displayed. During the latter “sacred and secular wealth distributions engineered by big men, involving massive pork feasts and the presentation and distribution of live pigs… overtook the dances in importance” (1998: 367). The authors ascribed the inclusion of agricultural produce among the festive goods to the enlarged scope for agricultural production brought about by the adoption of sweet potato cultivation (1998: 111-113). The Hubula ebe akho, just after contact, with their massive pork distribution and consumption, appear to show the outcome of an analogous development, with the important proviso that these exchanges had remained subordinated to life-cycle events. It is also unclear to me to what extent Hubula leaders “engineered” the distributions.

51The programmes of the two Hubula rituals, funerals and ebe akho, discussed above, also support Heider’s conclusion that the primary focus in the ritual life of these people was securing the benevolence of their ancestors. I concur with that conclusion and I would like to place it in the perspective opened up by, first, Godelier with his distinction between Great Men and Big Men, and later by Lemonnier with his identification of Leaders. Their work contained a historical – prehistorical – perspective, suggesting the flourishing of Big Men societies following on and helped along by the sweet potato revolution. However in the case of the Hubula, it appears to me that, after the adoption of the sweet potato, they did not pursue Bigmanisation, pace Roscoe (2011: 58). That supposition is in accordance with Lemonnier’s observation (1990: 149) that Big Men cannot emerge in a cultural context in which “durable peace is non-existent” and “ceremonial exchanges between enemies do not occur” (cf. Godelier 1991: 300). It seems more plausible that the Hubula used their increased harvests to intensify an ancestor worship that they were already practising (see also Godelier 1991: 276). Furthermore they did not pursue a “Great Man logic” either: they lived in large-scale groups, transacted marriage payments and did not show a marked compartmentalisation of leadership. Following his tripartition, Lemonnier referred to Hubula leaders as Leaders (1990: 129, 1991: 20f). However I am disinclined to do so if using that term implies that Bigmanisation among the Hubula was taking place, and that their leaders were, to use Godelier’s expression, small-scale Big Men. In my view the Hubula were taking a separate path of change in their way of life, one not covered in Godelier’s and Lemonnier’s analyses.


1 As number one, Brookfield ranked both the dryland and wetland cultivation in New Caledonia; he ranked Me wetland cultivation number four (Brookfield & Hart 1971: 106, 113-115).

2 Pétrequin noted one exception. But it took place post-1970, so long after first contact.

3 Peters used the spelling gain. But in this book, I adopt the spelling used by Broekhuyse and Heider.

Table des illustrations

Titre Pict. 5. Wetland agriculture in the Grand Valley
Légende Pierre Pétrequin took this picture in 1985, near the airport of Wamena, in mid valley. Sweet potato cultivation on the beds is in different stages. The picture was taken decades after first contact, but I have no reason to suppose that the Hubula had greatly changed their cultivation methods.
Crédits Source: Pierre Pétrequin & Anne-Marie Pétrequin.
Fichier image/jpeg, 887k

© pacific-credo Publications, 2020

Conditions d’utilisation :


Volume papier
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search