SEARCH

SEARCH BY CITATION

Keywords:

  • Motu-Koita;
  • sorcery;
  • magic;
  • death;
  • translation

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. The Dog's Death
  5. Ontological Lacunae
  6. Conceiving Vada
  7. The Meaning of Death
  8. Souls and Shadows
  9. Conclusion
  10. References

This article is driven by the equivocal possibility of doing analytic justice to the cosmo-ontology of the Motu-Koita, of Papua New Guinea, as it was when early missionaries and colonial officers credited south-east-coastal indigenes only with unsystematized beliefs and superstitions about invisible forces. It focuses on an incident in which traditional ‘sorcerers’ were put to the test by the early colonial administration, which was trying to destroy local beliefs in sorcery. By interrogating the discursive stereotypes brought to this episode by the administration, and problematizing the translation and understanding of some Motu-Koita terminology, it attempts some first steps toward a more nuanced understanding of the pre-European-contact lifeworld of the Motu-Koita.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. The Dog's Death
  5. Ontological Lacunae
  6. Conceiving Vada
  7. The Meaning of Death
  8. Souls and Shadows
  9. Conclusion
  10. References

The Motu-Koita are the traditional landholders of the area on which Port Moresby, the capital city of Papua New Guinea, has grown. This article is part of a work in progress: an attempt to do analytic justice to the cosmo-ontology of the Motu-Koita as it was when early missionaries and colonial officers denigrated it as superstition. In this instance I am also challenged by a comment by Mimica (2010:220) that in many present-day anthropological writings ‘even the would-be most epistemologically hyper-reflexive authors … depend for their thinking on the most basic categories inherited from the age-old Western cosmo-ontological traditions’. Given the necessity of using Western theoretical concepts in the absence of theoretical elaborations by the people we research, the unreflective application of those theoretical concepts, as well as the lack of rigour in the application of seemingly mundane Euro-American language terms become particularly problematic in Mimica's analysis. Accordingly, I attempt here to maintain an intrinsic acknowledgement that ‘there is no knowing without an N number of correlative modes of un/knowing’ (2010:204), where ‘un/knowing’ indicates the ‘internal coexistence of knowledge and motley ignorance’ (2010:221n1) in anthropological knowledge-making.

In the distant past the Motu and Koita were distinctly different from each other, as linguistic and archaeological evidence along with oral histories shows.1 The Motu spoke an Austronesian language. They built their houses on or near the shoreline, and were marine-oriented. Motu women specialized in the making of clay pots which the men traded along the coast, especially to the distant west from where they obtained sago (Dutton 1982b). The Koita spoke a non-Austronesian language, inhabited the coastal plains and were gardeners and hunters. They are thought to have split in the distant past from hinterland people known as Koiari (Dutton 1969). The original encounter between the Motu and Koita and the terms of their alliance are the subject of innumerable oral histories (see for example Oram 1981) and no conclusive account exists. They were, however, intermarried by the time Europeans arrived and there were few remaining differences between their sociality and ontology (see for example Lawes 1879).

Their earliest experience of Europeans was the arrival of exploring sailors and London Missionary Society (LMS) missionaries in the 1870s and, soon afterwards, a colonial administration. While descriptions of the indigenes by these Europeans were probably reliable so far as material culture was concerned, their representation of the ontology of the coastal societies was less so. The Motu-Koita and neighbouring people were credited only with unsystematized beliefs and superstitions about invisible forces. European ethnocentricity was compounded by the use of Christian terminology to represent many aspects of indigenous lifeworlds, even though the subject people were judged to have ‘no religious observance’ (Turner 1878:492). Moreover, as the Motu-Koita absorbed the English language, they came to use this and other analytically inappropriate terminology to represent their own ‘traditional’ ontology. The influential anthropologist Seligman (1910) described the Koita (whose material culture and social organization appeared by then to be more or less indistinguishable from the Motu) before the demise of social evolutionism and the advent of participant observation. Seligman's account of the material culture and ‘customs’ avoided the ascription of superstition, but showed no inclination to see what he called ‘magic’, ‘sorcery’, ‘religion’ and ‘eschatology’ as constituents of a systematic cosmo-ontology. The government anthropologist of the interbellum period, F.E. Williams, preoccupied himself with societies further afield, and thus the Motu-Koita received no further serious anthropological attention until after the Second World War (see, for example, Belshaw 1957; Groves 1954, 1957), by which time very significant social changes had taken place.

From this period onwards the investigation of the cosmo-ontology informing the everyday and ritual activities of the Motu-Koita in earlier times became a salvage enterprise made increasingly difficult with the passing of generations who had participated in them. Interviews and queries among these, moreover, indicated that many of them had since incorporated Christian language and thinking into their memories.2 Research elsewhere in Melanesia had indicated a complex, coherent system of thought among traditional Austronesian and other societies articulating personhood, mortality and mortuary procedure ontically, and thereby inviting ontological reflection.3 In the case of the Motu-Koita, however, nineteenth-century European contact and subsequent documentation in a Christian and colonial climate precluded similar reflection. In what follows I problematize the translation and interpretation of some Motu-Koita concepts and terminology, making some hesitant first steps toward a more nuanced understanding of the Motu-Koita lifeworld of the early colonial period than that achieved by Europeans of the time. My point of entry is the demise of a dog in an incident interpreted by the colonial administration as putting traditional ‘sorcerers’ to the test in its efforts to destroy local belief in sorcery. Interrogating the discursive stereotypes brought to this episode by the administration, I show how subscription to an image of naïve credulity, superstition and existential confusion among Papuans during several decades of colonial rule had impeded European understanding of their cosmo-ontology.

The Dog's Death

  1. Top of page
  2. Abstract
  3. Introduction
  4. The Dog's Death
  5. Ontological Lacunae
  6. Conceiving Vada
  7. The Meaning of Death
  8. Souls and Shadows
  9. Conclusion
  10. References

There is no doubt that the dog died on Saturday, 12 September 1931, at Port Moresby. European officials ensured its absolute lifelessness. The alleged sorcerers or ‘vada’, said to be from the inland Koiari tribe, had pummelled the dog until it lay inert and were preparing to demonstrate their skill at restoring it to its former vigour when the officials intervened. On the basis of a stethoscopic examination they declared that the vada had not killed the dog as they claimed, and it was consequently beaten further by the Europeans, to clinical death. Little else of the episode has proved indubitable with passing time however. European testimony agreed that the vada had subsequently been unable to reanimate the dog (Laloki 1931:21; Lett 1942:163, 1949:250; Murray 1970:139; Papuan Villager 1931:32). On the other hand a Melanesian version of events given to an anthropologist two decades later represented the vada as having killed the dog to the satisfaction of the officials and then successfully reanimated it (Belshaw 1957:193). Moreover, in the undiscerning European discourse of non-scientistic agency, concepts such as ‘sorcery’, ‘magic’, and even ‘wizardry’ were used interchangebly and were assumed to adequately translate local language terms such as ‘vada’. This linguistic licence served the pejorative perspective Europeans brought to the episode but did little to facilitate a precise appreciation of the indigenous interpretation of it. We do, however, know a little of the background to the event.

The dog's death was fated by the aftermath of a public meeting ten weeks previously, attended by village representatives of the Motu-Koita, on whose territory the colonial administration had installed its headquarters in 1884. A major topic of discussion was recent significant misfortunes among the indigenes which most of those present blamed on vada, described by the government report for the year as ‘sorcery’ and ‘black magic’. Two indigenous men disagreed with the sorcery diagnosis. One of these was Ahuia Ova, a Koita man whose service to the colonial administration had brought him local prominence (see Shelley 1978; Williams 1939). He declared his own belief in sorcery, but argued that the misfortunes were actually due to human shortcomings such as laziness and carelessness. The other dissenter, a Deacon in the Christian Church that had been introduced half a century earlier by the LMS, declared his disbelief in sorcery. Neither was persuasive of the majority (TPAR 1933:15). Following the meeting the Governor of Papua, Hubert Murray, was approached by a Motu-Koita ‘deputation’ with a man from the hinterland who was said to be a former vada now converted to Christianity. It was proposed that this man would bring some practising vada from his village to kill a ‘cat or dog’ and bring it back to life. Murray was told the vada would need time for preparation, and they would have to come by land, because if they came by sea, travelling over salt water, they would lose their power (Murray 1970:136).

The European view of the events of 12 September, then, was that Koiari vada had intended to demonstrate their abilities by killing a dog and bringing it back to life. The colonial administration had been attempting to eliminate ‘sorcery’ for more than four decades, and believed its attempts to disabuse Melanesians of their belief in the practice would be aided by the visible outcome of the demonstration. The government anthropologist, F.E. Williams, reported the incident in the Papuan Villager, his newspaper for literate Melanesians, concluding ‘We hope that the failure of the vada men at Port Moresby will help to open your eyes’ (Papuan Villager 1931:82). Governor Murray believed, naïvely, that sorcery had already ceased in the Gulf district to the west (Murray 1970:137) and hoped the rest of the territory's native population could be persuaded to the same end. Errol Flynn, then an occasional correspondent to the Bulletin magazine in Australia using the pseudonym ‘Laloki’, reported with sarcasm on what he said should have been a moral victory for Murray:

But the Papuan sorcerer is an astute opportunist. The chief pourri-pourri4 expert suddenly remembered that the medical officer, invited to satisfy himself that the dog was dead, had touched it with his stethoscope … Every reasonable native knew at once that the medical officer had applied a powerful counter pourri-pourri. It is unfortunately true that the magicians were given presents of rice and flour – a not unusual instance of official stupidity. If these natives did not return to their village and exhibit the rice and flour as proof that the Government was much impressed with their prowess, then they're not the pourri-pourri men I take them for. (Laloki 1931:21)

Less derisive of the officials was Murray's private secretary Lewis Lett who provided, in the 1940s, two reminiscences – slightly differing from one another – of the course of events (Lett 1942:163, 1949:249–51). In one of these Lett conceded that the exercise had not had the immediate effect the administration would have liked, despite the vada admitting ‘defeat’:

But of course there was a reason. Interviewed afterwards, they explained that the performance had been spoiled first by the intervention of the white doctor, and secondly by the fact that the dog had been ‘killed too much’; that the last scrap of vitality had been taken from it, and nothing left for them to work on.

Inevitably, among a superstitious people, the explanation given by the sorcerers carried more weight than the mere fact of their failure; and the reputation of the Koita [sic] sorcerers remained very much where it had been before. (Lett 1942:163)

Williams, the government anthropologist, attempted a rationalist dismissal of the vada protestations that they had not meant to ‘kill the dog altogether’ but ‘wanted to leave a bit of life in him so that when they used their medicines he would get up again’ (Papuan Villager 1931:82). ‘Surely’, he scoffed, ‘there is nothing in that! … All you need to do would be to leave the dog alone and by and by he would get up himself’ (1931:82). Governor Murray reported the argument of a sorcerer who subsequently asked him to kill three more dogs so the sorcerer could demonstrate an infallible method by which the animal would be restored to life. Murray declined to batter more dogs, with a comment that if he did kill one he would shoot it, which the sorcerer conceded would render the dog too lifeless for him to revive.

I said it was like a man being knocked insensible – he was dead, apparently, and then got up and walked away. But the sorcerer said No – for two reasons (i) the dog, if left to himself would never get up – the life would soon fade away and (ii) the dog would only live as long as the sorcerer appointed – perhaps one day, perhaps two, and perhaps more – but no longer. (Murray 1970:139)

So much, then, for the Europeans' perceptions and interpretations of the dog's death, of the intentions and failings of the men they called vada, and of the explanations the latter offered. In all respects, the adequacy of those perceptions and interpretations was equivocal, informed as they were by a cosmo-ontology which could not accommodate the lifeworld of the Motu-Koita except in the form of an untidy knot of existential confusion, causal misconception and superstition.

Ontological Lacunae

  1. Top of page
  2. Abstract
  3. Introduction
  4. The Dog's Death
  5. Ontological Lacunae
  6. Conceiving Vada
  7. The Meaning of Death
  8. Souls and Shadows
  9. Conclusion
  10. References

The burden of primitivism weighed heavily on the Administration's understanding of the case of the dead dog, as it had on anthropological research on the Motu and Koita during the previous four decades which might otherwise have left it better informed about local existential attitudes and more able to place them in an appropriate ontological context. The first scholarly publication on the Motu, by the missionary ethnologist W.Y. Turner in 1878, was no portent of the immersive style of fieldwork which Gunnar Landtman and Bronislaw Malinowski would introduce three decades later.5 The superficiality of Turner's observations on Motu material life was manifest in a littering of misinformation and inaccurate speculation, and his institutional religious conditioning inclined him to the view that as well as having ‘no god or gods’ the Motu had ‘no religious observance, and no sacrifices or offerings’ (Turner 1878:492). He found the absence of religion remarkable ‘as they believe in the immortality of the soul: hence in spirits’ (1878:492). The non-material aspects of the Motu lifeworld were accordingly addressed under more or less discrete categories such as superstition (1878:491–2) or beliefs about the destination of the ‘soul’ after death (1878:485–6), rather than being viewed as implications of a cohesive mode of thinking about their lifeworld. Turner's views were shared by the better-informed missionary W.G. Lawes, who had at least studied the Motu language in some depth (see Lawes 1896), but whose published short notes on the Motu, Koita and Koiari (Lawes 1879) offered little correction to the understanding that these groups were ruled by superstition.

The perception that Papuans had no religion was not, in its day, informed by the kind of reflection on the nature of religion that generated, for example, Lévi-Strauss's argument that ‘There is no religion without magic any more than there is magic without at least a trace of religion’ (1966:221). The measure applied to Papuans was the institutional and teleological character of Christianity and other monotheistic systems. By that measure an unremarked, commonsense, distinction prevailed between the incantatory rituals of Christianity and those of the Motu-Koita. The former were ‘religion’, the latter ‘magic’. This distinction precluded the ability of Europeans to recognize incantatory rituals among Papuans as cosmo-ontologically integral, deserving of comparison for example to the sacrament administered by a Catholic priest, the Communion conducted by a Protestant minister, or the private prayer of a parishioner, which were perceived as integral to the relationship between Christians and God. The designation ‘magic’, carrying all the incredulous connotions of European usage, excluded Motu-Koita practice from consideration as either religion or science and juxtaposed it to superstition.

When the anthropologist C.G. Seligman arrived at the beginning of the twentieth century he attempted to concentrate descriptively on the Koita, rather than the Motu (Seligman 1910). However the Motu and Koita had been intermarrying for some time, Motu villages had Koita sections within them, and there had been extensive cultural sharing between them. Depending mostly on a single Koita interlocutor – the above-mentioned Ahuia Ova – Seligman described the material life of the Koita in some depth, although he was unable to significantly distinguish them from the Motu apart from occasionally giving comparative Motu and Koita language terms for the same practice or concept. Seligman was less inclined than his ethnological predecessors to use the label ‘superstition’. Nevertheless he treated ‘magic’ and ‘sorcery’ together in a single chapter (1910:167–82), death and mourning practices in another (ibid.:159–66) and, in a chapter entitled ‘Religion’ (ibid.:183–93), catalogued the ‘spirits’ of the Koita lifeworld and described the Koita ‘soul’ and its activity after death under the heading ‘Eschatology’ (ibid.:189–193. The veneration of celestial objects was indicated in a single paragraph headed ‘Cult of the Heavenly Bodies’ (ibid.:193).

Malinowski, briefly in Port Moresby before his Trobriand fieldwork, used Ahuia Ova as his interlocutor like Seligman and traversed some of the same topics (Malinowski n.d.) without adding any implications that the Motu and Koita might have an ordered view of the cosmos worthy of considering with the ontological rigour which the Greco-European philosophical tradition has applied to its own lifeworld. Rather, a fragmentary and misrepresentative translation of decontextualized processes and entities created ontological lacunae. Integral elements of the cosmomorphic experience of Papuans were lost to European understanding, replaced by discursive stereotypes which were more easily digested and sustained a comfortable image of naïve credulity, superstition and existential confusion. The stereotypes included the sorcerer and his spurious claims to control human mortality. I will first critique the European construction of the Papuan ‘sorcerer’, and then consider the Papuan view of mortality.

Conceiving Vada

  1. Top of page
  2. Abstract
  3. Introduction
  4. The Dog's Death
  5. Ontological Lacunae
  6. Conceiving Vada
  7. The Meaning of Death
  8. Souls and Shadows
  9. Conclusion
  10. References

The anthropology of sorcery has been described as ‘an important area where anthropologists have attempted to overturn colonialist and other forms of dominance’ (Kapferer 1997:9). The historical trajectory of the endeavour is exemplified by Evans-Pritchard's 1937 argument for the rationality of Azande views of sorcery (Evans-Pritchard 1976), and in more recent times by Kapferer's phenomenological consideration of Sinhalese sorcery ‘as disclosing dimensions of human action which may be more obscured in apparently different or similar practices elsewhere’ (Kapferer 1997:11). Evans-Pritchard's careful description (which distinguished between ‘sorcery’ and ‘witchcraft’6) has become a classic text, not least because it was ‘a positive attempt to separate other systems from the glare of a post-Enlightenment prejudice against all things that smacked of superstition’ (Kapferer 1997:12). In the early colonial encounter in Melanesia, however, terms like ‘sorcery’ and ‘magic’ were still prejudicial, connoting irrationality and primitive thought. Governor Murray lamented that indigenes lived in ‘constant terror of witchcraft’ (1925:67), to which he accorded no credibility: ‘We know of course that sorcery is all rubbish’ (1926:12).

The Motu-Koita themselves learned the exotic term ‘sorcery’ from Europeans, whose usage was imprecise and referred to any human endeavour to damage or improve the health or well-being of others by magical means. Melanesian interpretation refracted the European generalization through their cosmomorphic understanding of a lifeworld in which most of a person's encounters with human and non-human entities could be modified through ritual action and incantation and the entreated or uninvited intervention of ancestors and a variety of spirits. Consequently, Motu-Koita interlocutors responded to occasional European attempts to collect a catalogue of sorcery by providing examples of their use of incantatory rituals to various ends. The colonizers, guided by European moral values, dichotomized these as ‘good’ and ‘bad’ sorcery. The ‘good’ included rituals to ensure successful fishing or gardening and good health. The ‘bad’ ranged in malevolence from rendering someone's fishing trip unproductive to homicide (see, e.g., O'Malley 1912:99). Of these, missionaries and the colonial administration were most preoccupied with the kinds more threatening to human mortality, for which the local language term ‘vada’ became the favoured word both for the agency and for the person capable of it.

Vada had been defined in various ways by Europeans in the half-century before the 1931 dog killing. Turner had described them in 1878 as evil bush-dwelling spirits to which the Motu were susceptible (1878:483). Lawes added that Koita were thought to have power over vada, the ‘prince of evil spirits’ (1879:374), and could negate their power if approached with gifts by the Motu. In his published grammar and vocabulary of the Motu language he rendered the term vada as a substantive pronounced ‘vata’,7 added a reduplicative form, ‘vatavata’, and offered a translation and definition: ‘ghosts; an unknown spirit supposed to have the power of killing whom he will’ (Lawes 1896:136). Human agency was implied in another entry in the same volume, ‘Vatari tauna or hahinena, a sorcerer or sorceress’ (ibid.): this introduced an apparently adjectival usage with tau and hahine (‘man’ and ‘woman’ respectively, /-na/ is a possessive suffix).

An early colonial deputy-commissioner suspected that Koita pretended to be ‘vata-vata’ by creating appropriate noise at night in order to extort gifts from Motu seeking protection (Romilly 1889:28). Seligman, also preferring the reduplicative form ‘vadavada’, said it referred to ‘predatory bushmen, who might fall upon and kill a wandering Koita, as well as to certain non-human beings stated to kill men with an egg-shaped stone club, make medicine of the body, and then bring the victim to life again’ (1910:187). He remarked upon the variations of local descriptions, some of which implied the vadavada were human Koiari, and reported that his principal interlocutor – the Koita Ahuia Ova – said vadavada were human ‘sorcerers’ (1910:187). Ahuia Ova, suspected to be a sorcerer himself (Shelley 1978), told Malinowski the same thing in 1914, identifying vada as having traditionally been Koiari (Malinowski n.d.:604–5). Seligman occasionally ‘quoted’ Ahuia in his book (for example 1910:129, 171), as did Malinowski in his 1914 fieldnotes (n.d.:passim). These references suggest that Ahuia spoke broken, rather than conversationally good, English at that time:8 both authors' representations of indigenous concepts therefore need to be treated with caution. Malinowski was also told by a European cleric that vada was a generic name for all evil spirits and influences, though was a personal agency (n.d.:85).

Meanwhile Resident Magistrate J.T. O'Malley, apparently depending on Koita interviewees, reported that vada were human, and traditionally Koiari, though some Koita had learned the skill from the inland group (O'Malley 1912:99). His discussion mentioned vadari, possibly a modification of Lawes' adjectival vatari (and see note 7 this article), though he implied that it was a noun by defining it as ‘jealousy sorcery’ used to impoverish the rich and successful (ibid.). In 1924 Lister-Turner, a missionary, described vada as a form of witchcraft general to the Central Division (the 200-mile stretch of coast and its hinterland proximate to Port Moresby) though ‘not practised by the coast natives’ (1924:117). The victim was knocked unconscious and then either had poison (from plants or a worm species) introduced into his body or was extensively pummelled on the abdomen before he recovered consciousness. The attacker (for whom Lister-Turner used the term ‘magician’, not ‘sorcerer’) would utter spells during the process. The poisons or the internal damage from the pummelling would cause the victim's death within a a day or two (1924:118–19).

In 1931 (the year of the dog experiment) Lister-Turner co-authored a dictionary of Motu in which Lawes's translation and definition of vata (now conventionalized as ‘vada’) was reproduced with an addition, thus: ‘ghost; unknown spirit supposed to have the power of killing whom he will; human prowlers at night to kill whom they find’ (Lister-Turner and Clark 1949:123). In the same publication the sorcerer was gendered as masculine only – ‘Vadari tauna, c.s. a sorcerer’ (‘c.s.’ is an abbreviation of ‘compound substantive’) – and a verb form was added: ‘Vadari-a, v. to cause evil by sorcery’ (ibid.). It can be seen from the various definitions above that during a half-century of European presence a transition had taken place in their use of the term vada (or vata). Where it had originally been used of bush spirits it was, by the time of the dog experiment, being used of magical agency and human agents. This development can be partly explained by means of some linguistic investigation.

Motu is an Austronesian language, and cognates for vada could be found in early colonial times among other Austronesian languages along the south-east coast of Papua New Guinea. To the near-west of the Motu, among the people now known as Nara, the term was oada. To the east, among the Sinaugoro, Balawaia, Vula'a (commonly called Hula), and Keapara, the term was wara (see, e.g., Guise 1891:109).9 The languages of these groups are all subsumed under what has come to be known as the Central Papuan family of Austronesian languages. The Central family is a subgroup of the Papuan Tip Cluster, which is itself a subgroup of the Western Oceanic group of Oceanic languages.10 Many languages of the Central family have a high cognatic correlation with Motu. Some other Austronesian-language-speaking neighbours west of the Motu have lower cognatic correlations, and used other terms than vada to refer to bush spirits. These terms included paipai among the Roro (Monsell-Davis 1981:239–41) and – a cognate of the latter – faifai among the Mekeo (Hau'ofa 1981:37; Stephen 1995:55) and the inland Kuni (Gostin 1986:16).

Ritual action involving incantations and the use of a variety of objects (plants, stones, bones, etc) was endemic among local peoples. While ‘magic’ may appear to be an appropriate description of this resource it was not regarded as magical by Melanesians, and all adults used it to some extent in much daily life (see, e.g., Monsell-Davis 1981:241–2; Seligman 1910: passim), for instance in gardens simply to encourage cultivants to grow, or in the bush to attract hunted animals or ward off malevolent spirits. Incantations in general (I am avoiding the more favoured English term ‘spells’) were called mea, mega or meamea among most of the Austronesian speaking groups mentioned above.11 Mea was, unfortunately, lexically defined by Lawes as ‘superhuman power, as possessed by sorcerers’ (1896:119), exaggerating its nature and implying that its use was solely malevolent. He also offered a misleading definition of mea tauna as ‘a sorcerer who has cursing power’ (1896:120). In fact knowledge of incantatory ritual varied among individuals, depending on factors such as inherited knowledge or apprenticeship to specialists, and the term mea tauna (‘person of incantations’) could be applied, without necessarily implying evil intent, to people who were known, or suspected, to have greater powers than others to affect or control the course of events. At the same time, some people had the ability to persuade or force bush spirits to aid them in their own endeavours. This was particularly so among the Mekeo, whose powerful and deadly specialists were known as ugauga12 (Hau'ofa 1981:220–22; Stephen 1995:22–3 and passim). Among the Roro, on the other hand, adepts were called by different terms according to their particular sphere of activity (Monsell-Davis 1981:255).

Interestingly, knowledge of death ‘magic’ was not recorded among the Motu by missionaries or early colonial writers. However, the Motu were reported to fear the Koita, with whom they were socially intimate but whom they considered to have power over vada (Lawes 1879:374). Koita, in turn, responded to European queries by attributing knowledge about vada primarily to the hinterland Koiari (the people from whom the Koita appear to have split in earlier times13). An interlocutor from the Koita village of Kila Kila suggested, though, that Koiari may have given vada knowledge to the people of another Koita village, Baruni, on the basis of marriage relations with that village (O'Malley 1912:99). By way of contrast, interlocutors from the Kuni group (inland from Mekeo) told an ethnographer many years later that ‘sorcery’ (that is, use of incantatory ritual to kill people) was not widespread among themselves before European contact, when tensions had been addressed through raiding and warfare (the implication is that these activities had been outlawed by missionaries and colonial authorities). The Kuni had consequently obtained oada knowledge from Nara and Kabadi (the neighbours of the Nara), and mino (sympathetic and contagious ‘magic’) from an inland group, the Fuyughe, to fight their enemies by other means (Gostin 1986:74).

Most Europeans of the early colonial period were not competent speakers of the various native languages of British New Guinea. Missionaries and some colonial officers made an effort to learn some Motu, but there is evidence that most of these were taught a simplified form by the Motu, who did not encourage outsiders to thoroughly learn the ‘true’ form of their language (Taylor 1978:1328, 1340–41). In general, Europeans communicated with non-English speakers through the lingua franca known as ‘Police Motu’,14 a pidgin Motu with a restricted lexicon and simplified grammar. In this form of communication it was not possible to precisely identify or discuss, in respect of each cultural group, the nature and abode of spirits, the degree and type of control over them that local specialists might have, or the manner in which this was achieved. Consequently ‘vada’ and ‘vada tauna’ (vada man), derived from Motu, became all-purpose ‘native’ terms in colonial discourse for deadly magical agency and the people who were specialized in it. The nominals ‘tau’ and tauna' were frequently dropped, so that ‘vada’ alone was used of deadly magical specialists. Malinowski (not a Motu speaker, and only briefly in Port Moresby) adopted the term to refer to human agents (‘sorcerers’) in his enquiries. His notes of an explanation by his Koita interlocutor, Ahuia Ova, include the statements ‘The Vada … goes bodily over to the victim. When he goes away he is absolutely and bodily absent from the place where he has slept. Any person present would notice his absence. All the Koiari were Vadas in olden days, so were the Kabadi, Nara, Waima and also the Purari’ (Malinowski n.d.:604–605). As well as being completely human in this description, vada were historically dispersed far along the south coast: ‘Kabadi’ is immediately west of Motu-Koita territory, ‘Waima’ is some distance further west and ‘Purari’ is in the more distant Papua Gulf.

Whether it was completely accurate or not, Malinowski's interpretation of Ahuia's comments, a far-flung historical geography of human vada, contains within its extravagance an important implication which was present in the earliest records of Motu attitudes: the habitat of vada was beyond their village area. Koita were suspected to be implicated in vada activity, but the territory of the more distant inland group, the Koiari, was seen as their provenance. The social distance between the Motu and the distinctly alien Koiari, combined with their geographical location, facilitated a conceptual blurring between the bush-dwelling spirits and the bush-dwelling Koiari as their human operators. The Koita had a far greater degree of interaction with the Koiari and were trading intermediaries between the forest dwellers and the coastal Motu. Their oral histories also suggested an even closer relationship to the Koiari in the distant past (Dutton 1969:102–04; Gadiki 1971; Oram 1981). Understandably, Motu who suspected they were under attack from vada approached Koita for assistance.

According to most versions of the story of the dog killing, the vada who were brought to Port Moresby to demonstrate their skills in 1931 were Koiari. Yet even this appellation appears ambiguous when the European versions of the dog killing are examined. The most precise description of where the ‘vada’ came from is in Governor Murray's correspondence to a relative, which refers to ‘some mountain place at the back of Rigo’ (Murray 1970:136). Rigo is a village some 40 miles east of Port Moresby, and was also the name of an administrative patrol sub-district in Murray's day. The area was and is inhabited by Sinaugoro, an Austronesian-speaking group whose cultural territory stretches considerably further inland than other Austronesian groups in the region (see Kolia 1976:25–28) and part of the territory further inland (that is, northward) was actually beyond the western boundary of the Koiari language group. Murray also commented, before the arrival of the ‘vada’, that he was told they would be coming by land, because if they travelled by sea they would lose their power (ibid.). This implies that wherever their ‘place’ was, its inhabitants might normally have come to Port Moresby by sea. The Koiari, whose inland territory was large, would on the other hand invariably come to Port Moresby overland through Koita territory. We cannot, therefore, even be sure that these ‘vada’ were Koiari, despite the popular reference.15

In this section I have indicated that the discursive encompassment of linguistically and culturally diverse local groups by Europeans particularly through the medium of Police Motu contributed to an elision between a variety of conceptions of dangerous spirit entities and a generalized class of death-dealing ritual specialists, increasingly commonly called ‘vada’ by Europeans, and glossed in English as ‘sorcerers’. This was brought to the colonial officials' understanding of the dog-killing incident in 1931. Bracketing for the moment the question of whether the conscripted ‘Koiari vadas’ really were death-dealing magicians by any measure, I will now turn to the matter of their apparent failure to kill the dog in the first place and then to bring it back to life, and their protestations that the dog had been killed too much.

The Meaning of Death

  1. Top of page
  2. Abstract
  3. Introduction
  4. The Dog's Death
  5. Ontological Lacunae
  6. Conceiving Vada
  7. The Meaning of Death
  8. Souls and Shadows
  9. Conclusion
  10. References

Missionary agendas such as Bible translation and Christian education drove the first efforts to translate the languages of the south-east coast (see Taylor 1977, 1978:1335–38). The communicative needs of the colonial administration reinforced these projects. As the administration headquarters were in Port Moresby, it was inevitable that the Motu language, or versions of it, became a vehicle for communication in the area. Motu was more extensively represented in published grammars and vocabularies16 than most of the nearby languages by the 1930s. However, the phenomenological sensitivity of the linguistic theories of the time was relatively undeveloped. In particular, European grammatical convention provided the interpretive context of translation (see, e.g., Lawes 1896; Lister-Turner and Clark 1931a). These texts continue to be the major source material on the Motu language but are relatively naïve in their comprehension, compared to more sophisticated linguistic approaches that have been applied to Oceanic languages in recent decades. The latter attempt more reflectively to address the imperative that in a scientific description of a language ‘the end product … should mirror the unconscious “theory” of native speakers rather than that of the linguist or any school of linguistics’ (Jones 1998:25–6).

Motu shares with other members of the Central family of Austronesian languages some significant characteristics which contribute to a considerable degree of formal indeterminacy. Conversational and narrative reference can be largely implicit, and meanings are thereby negotiated, rather than fixed. Several aspects of the Mekeo language (which is also in the Central family) addressed in an innovative lexicogrammatical study by Jones (1998) are shared by the Motu language. Drawing from Jones's Mekeo description, a number of qualities pertinent to the current argument can be identified in Motu. First, like Mekeo, Motu is a verb-final language and a head-marking17 language (see Jones 1998:2). Second, almost any given root can function nominally or verbally as the occasion demands, so essentially there are no nouns or verbs as such, only multifunctional bases (1998:33). Third, it does not use switch reference systems and has no system of grammatical gender, and so it relies heavily on exophoric reference and the pragmatic knowledge of the hearer (1998:37–8). Fourth, ‘despite the use of extralinguistic props the language manages to retain an extraordinary level of intrinsic and … functional indeterminacy, especially as regards the participant roles in an event (or process)’ (1998:38). A more general point made by Jones – which is of particular relevance to consideration of the dog-killing episode – is that ‘some languages are more implicit than others, leaving far more interpretive responsibility to the hearer, in terms of the pragmatic reconstruction of a prototypical scene’ (1998:550). Bearing these characteristics in mind, and recalling that the principal indigenous language used in relation to colonial administration was Motu, we can critically consider some of the concepts which would have been translated, perhaps by Motu-Koita and even Koiari interlocutors, in the European representation of the intentions and actions of the ‘vada’ in 1931: in particular, terms related to killing and death.

The term mase is represented in the standard dictionary of Motu (Lister-Turner and Clark 1931, 1949) as a verb, ‘to die’ (1954:104), and separately as an adverb ‘of intensity’. In the latter entry the authors use the untranslated example ‘tahua mase’ (ibid.): tahua is translated elsewhere in the dictionary as a verb, ‘to seek: to examine’ (1954:118). In the English-Motu vocabulary section of the accompanying grammar of the language (Lister-Turner and Clark 1931a) mase is used as a translation of ‘death’ and also ‘dead’ and ‘die’ (in both of the latter entries preceded by the morpheme /e/ – a ‘verb particle’ in the authors' terminology – thus, e mase) (1931a:55). If we treat mase as a multifunctional base instead of identifying it variously and discretely as a noun or verb, we can avoid the apparently disjunctive identification of it also as an adverb ‘of intensity’. To clarify this, I will introduce the term (or base) botai, ‘hit’. Botai can function as a verb when marked by the suffix /-a/; thus botaia. Botaia mase can be translated ‘beat to death’, as in the phrase ‘tatau ese sisia e botaia mase’ (the men/tatau beat the dog/sisia to death). Revisiting the example given by Lister-Turner and Clark, (if we comprehend tahu-a as a verb function of the base tahu) tahua mase can be translated as ‘investigate (or seek or examine) to death’, of which the idiomatic sense is not difficult to grasp.

In relation to the killing of people or animals, a number of terms could apply in Motu. For example the base ala functions verbally as alaia, a term which could be translated ‘kill’, and ala takes a reduplicative form, alala (that is, a contraction of ala-ala) to function as a noun for ‘war’. Depending on context, the verb form might be combined with mase, thus alaia mase. It can also be used in the compound alaia ore, where ore is ‘completion’ and the compound sense is ‘kill off’ or ‘exterminate’. More specific methods of deliberate killing include beating to death (botaia mase) and spearing/stabbing to death (gwadaia mase). A less specific sense is conveyed by hamase, where /ha/ is a causative prefix: thus ‘tatau ese sisia e hamase’ could be translated ‘the men caused the dog to die’. A simple but important point about what can be lost in translation can be illustrated with this last example. In free translation into English the clause ‘tatau ese sisia e hamase’ could also roughly be rendered ‘the men (tatau) killed the dog (sisia)’, as could the clause ‘tatau ese sisia e alaia’. However, in Motu the use of alaia implies a definite violent intent, whereas hamase allows for a range of interpretations: the death could have been intended, but also could have been an accident, or due to neglect, a careless act, lack of foresight or other indirect means. Without contextualizing information, the simple translation of hamase as ‘killed’ is inadequate.

I have translated the term mase thus far as ‘death’ and ‘die’, which has been adequate for the discussion of terminology to do with ‘killing’. However mase needs investigation in itself, for the English-language terms do not fully represent its use and meaning for Melanesians and a greater understanding of it is crucial to an interrogation of the European's interpretation of the 1931 dog-killing exercise. Missionary linguists and the anthropologists Seligman and Malinowski did not consider the meaning of mase beyond its simple translation as ‘death’ and ‘die’. A colonial report in 1908 stated that a magistrate was ‘staggered’ by the response of a native who was being ridiculed for his belief in sorcery. The response was reported as being:

I saw your doctor kill a man by putting puri-puri (sorcery medicine) to his face. He then cut him open with a knife, and there was plenty of blood. Then he brought the man back to life again. If your doctor can do these things, why do you not believe that the native sorcerers can do the same? (PAR 1908:63, my italics).

Leaving aside questions of whether this statement was accurately reported, quoted or translated, the possibility that such an utterance might represent an alternative understanding, rather than a misunderstanding, of the nature of ‘death’ does not appear to have been entertained by the magistrate.

The first European publication to interrogate the conventional translation of mase as ‘dead’ was by W.M. Strong in 1919. Strong was the chief medical officer during the period when Seligman and Malinowski visited Port Moresby. He also acted as an unofficial government anthropologist. Malinowski was not entirely impressed by Strong during his visit, partly because of the latter's limited grasp of the Motu language (Malinowski 1989:76). However, in his publication of 1919, Strong demonstrated some thoughtful insight. He argued that mase meant ‘to be unable to be made to move’ (1919:303) rather than ‘dead’ in the English sense, and included fainting and unconsciousness among its range of meaning (ibid.) He noted that the anaesthetic chloroform was called mase muramura (mase ‘medicine’) by Motuans, as well as that a watch or clock which was run down was said by them to be mase (ibid.:304). He wrote : ‘Our idea of “dead” can be translated by mase wadain’ (ibid.:303), and explained wadain as adding ‘a sense of permanence to the word it is used with’ (ibid.). Strong interpreted well but heard incorrectly here: ‘wadain’ was vadaeni, a term indicating completeness: mase vadaeni can be translated as ‘completely dead/died’, or ‘dead/died altogether’.

Strong's observations fell on infertile ground. Colonial interpretations of Melanesian responses to ‘death’ did not allow for the possibility that the Motu-Koita might have a nuanced view of the existential transitions that took place when the human body was incapacitated, and of the interventions that might be possible. Lister-Turner's description (above) of a ‘magician’ rendering a victim unconscious and poisoning or damaging their internal organs to bring about their future ‘death’ (1924:118–19) was also worthy of more extensive consideration. In combination, observations such as those of Strong and Lister-Turner contained potential for valuable insights into the Melanesian lifeworld. But they would have required a heuristic suspension of the European belief that death was simply the moment when life ended and any extant soul or spirit became irretrievably separated from the world of the living. The colonial account of indigenous responses to death could only accommodate the recording of mourning and mortuary rituals, and occasional apparent failures to understand the simple facts of life and death.

For example, unsuccessful attempts to reanimate the dead by men of equivocal knowledge of the necessary techniques had been reported before the 1931 incident, including the case of a Koita man who in 1906 killed a European, Richard Weaver, apparently to gain the insignia and indigenous status of a man-killer. Mindful of the colonizers' disapproval of the killing of white men, he then attempted to restore Weaver to life under the guidance of a companion said to be ‘skilled in charms’. Under subsequent questioning, according to the colonial report, he claimed to have successfully re-animated the European's legs but said his progress up the body was eventually defeated by a ‘ghastly’ wound in Weaver's chest. He was hanged for murder (PAR 1907:11–12). An administration report in 1908 related an incident in which a village constable said to be a former sorcerer was asked to ‘prove that the dead could be brought back to life’ (PAR 1908:63). Consenting reluctantly under administration pressure, and pleading many years' lack of practice, he beat a lizard to death with a stick and tried unsuccessfully for half an hour to reanimate it. He was said to attribute his failure to insufficient self-preparation and the destruction of his powers by his connection with ‘the Government’ (ibid.).

The complement of European-witnessed failures at reanimation was reportage of indigenous eye-witness accounts of sorcery, interpreted as evidence of the power of belief which fuelled native credulity and imagination. For example a report of murder ‘many miles inland from Port Moresby’ (ibid.) quoted the ‘invariable’ testimony of several indigenous eye-witnesses compositely thus:

‘I saw the sorcerer go up to A—– and speak to him. Then he lifted his club and struck A—– a heavy blow on the head. A—– fell to the ground, and the sorcerer struck him on the head again and again. A—–'s head was broken open, and he was covered with blood and quite dead. The ground where he fell was also covered in blood. Then the sorcerer called two other sorcerers and they worked charms over A—–, and he came to life again and stood up. His head was no longer broken, but perfectly cured, and there was no sign of any blood left on him or on the ground. A—– went home to his house and was quite well, and that night he went with us to another village and danced all night. In the morning he went home to his house and died.’ (ibid.)

The European incredulity accompanying these accounts was partly fuelled by the facile translation of the term mase to mean ‘death’ or ‘dead’ in the clinical sense preferred by Europeans. However, like cognate terms in other Austronesian languages, mase was applied by the Motu to a spectrum of conditions which, in English glosses, ranged from involuntary unconsciousness (but not sleep, which is mahuta) to clinical death, as Strong had intuited. The phrase botaia mase, which I previously translated simply as ‘beat to death’, can thus be translated also as ‘beat into insensibility’. Relating this to the battering of the dog we should bear in mind Jones's point above that some languages are more implicit than others, ‘leaving far more interpretive responsibility to the hearer’ (Jones 1998:550). Had they adopted a more nuanced understanding of the Motu language, colonial officials might have given more thought to what was intended by a proposition to ‘kill’ a dog, and what was meant by a protestation that the dog had been ‘killed too much’.

Souls and Shadows

  1. Top of page
  2. Abstract
  3. Introduction
  4. The Dog's Death
  5. Ontological Lacunae
  6. Conceiving Vada
  7. The Meaning of Death
  8. Souls and Shadows
  9. Conclusion
  10. References

Different interpretations might also have been encouraged had the observations of Strong and Lister-Turner been related to the notes of missionary linguists and anthropologists on Motu-Koita terms for the lauma or ‘soul’ of a person (Lawes 1879:370, 1896:86, 116; Malinowski n.d.:605; Seligman 1910:189–93). They had also described sets of rituals conducted for a lengthy period after a person's death (Seligman 1910:159–66; Turner 1878:484–6). In combination these observations recorded characteristics shared by many societies in Melanesia and beyond. They resonated with the material considered by the Belgian scholar Robert Hertz in his 1907 essay on the collective representation of death, based on comparative data from Borneo and elsewhere (Hertz 1907, 1960). Hertz was intrigued by the extended mortuary processes which he conceptualized as ‘double burial’ (1960:28 and passim). He used this term to refer to the relationship between an initial ritual response to a death and a second and final ritual, often conducted many months later. Between the two rituals an active ‘soul’ was understood to be proximate to the corpse, and the community needed to be attentive to this presence in various ways. Hertz gave some consideration to the relationship between the ‘double burial’ and the nature and behaviour of what he called the ‘soul’, which seemed in some societies to have two forms, or perhaps to split into two after death (1960:34, 46–7).

The English term ‘soul’ is redolent especially with Christian meaning (and was certainly so in early colonial Papua) and may not be an appropriate translation of the Motu word lauma. An early definition of lauma was offered by Lawes, who called it ‘a spirit: formerly used only of ghosts of those killed, who appeared in terrible form’ (1896:116). This was largely reproduced by the later Lister-Turner and Clark dictionary: ‘spirit, ghost appearing at night. Formerly used only of ghosts of those killed, who appeared in terrible form’ (1931a:100). Seligman atttempted to described the Koita sua, which he presented as the equivalent of the Motu lauma, and identified as a ‘ghost’, ‘shade’, ‘soul’, ‘life’, and ‘vital essence’ (1910:189). This wide variety of labels was made more complicated by some slippage between the Motu and Koita terms during Seligman's descriptive passage, and some confusion in his attempt to contrast the sua/lauma with another entity represented in Motu as laulau and in Koita as variva. In the course of this some typographic confusion occurs, where both laulau and lauma become ‘laulauma’, to the detriment of an attempt (using the Motu language terms) to explain that some animals have no lauma (ibid.). His further discussion of the actions of sua/lauma after death represents them as living in a parallel world to that of humans, with the ability to affect the latter (mostly by punishing them, or appearing as ghosts), before finally disappearing altogether, perhaps when their names are finally forgotten by the living (1910:189–193).

Seligman reported being told about the danger of the sua/lauma leaving the body of a living person, its vulnerability to possession by malevolent spirits, and the role of a specialist, the babalau, in restoring the sua/lauma to the body. (1910:167–71, 191–2). But in addition to explanations by Ahuia and perhaps other interlocutors, his information was drawn from the writings of Hugh Romilly, who was at one time Acting Special Commissioner in British New Guinea and wrote a memoir of his experiences (Romilly 1889). Romilly learned some Motu but was a story-teller rather than a serious scholar, and Seligman was obliged to acknowledge ‘a certain vagueness’ in his ‘accounts of spirit invocation’ (Seligman 1910:169). Seligman further failed to question Romilly's reference to the ‘papalau’ (babalau) as a ‘sorceress’ (Romilly 1889:91–6): a babalau is in fact a diviner.

The laulau to which Seligman attempted to compare the lauma was agreed in most early literature to be a shadow, reflection, or (in the colonial situation) a photographic likeness of a person (Lawes 1896:116; Seligman 1910:189). However Malinowski's notes from discussions with Ahuia refer at one point to ‘the spirits of the dead (lauma) or living (laulau)’ (n.d.:605) as if the lauma was a variation or development of the laulau after death. He goes on to say that ‘Lauma is defined as the person coming in dreams to a man … Laulau is defined as the man's shadow … The Lauma is invisible and it goes to some unknown land. The Laulau goes about in the night, but when one wants to seize it it vanishes’ (ibid.). The babalau is described as being capable of interacting with various kinds of spirits, including lauma and bush spirits known as devase (ibid., cf. Romilly 1889:91–6, Seligman 1910:186–7) who ‘come like Laulau’. Malinowski further noted (from Ahuia): ‘A babalau sees the lauma and knows their signs. The babalau is the biaguna [Motu: equivalent to “manager”, “co-ordinator”] of all the lauma’ (n.d.:611).

These descriptions are inconsistent and confusing. Their shortcomings are not lessened by comparison with discussions of human ‘spirits’ among the neighbours of the Motu-Koita. As with vada, cognates for the Motu laulau, for example, can be found in nearby Austronesian languages, and latter-day ethnographers have been cautious in their interpretation of these. Among the Mekeo for example the term is lalauga, for which Stephen has argued ‘spirit’ and ‘soul’ are inappropriate translations, even though these have been adopted (from Catholic missionary use) by the Mekeo themselves (Stephen 1995:116). Among the Vula'a the term is avuavu, and Van Heekeren's account of it describes qualities which have been attributed to both the laulau and lauma of the Motu (2012:93–7). Among the Roro the cognate term is auba, and Monsell-Davis's account not only describes auba before and after death, but also refers to other terms used of spirits after death, aiha and biriua, which depend on the circumstances of death (1981:235–8).

A satisfactory exegesis of laulau and lauma would require more space than can be permitted in this article, but the disparate references of early colonial writers alert us at the least to the possibility of the lauma's detachment from the live body, its vulnerability to appropriation by forces external to the person, the deleterious effects of this on the health of the body, and the communicative relationship between specialists such as the babalau and spirits of various kinds. Combined with the extended sense of the term mase (that is, encompassing European notions of unconsciousness and clinical death) and the extended mortuary processes of the Motu-Koita and nearby groups which conform with Hertz's ‘double burial’ model,18 the confused representation of the lauma indicates the complex nature of Motu-Koita cosmomorphism and the need for cautious reflection on the meaning of a translated proposition to kill and reanimate a dog.

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. The Dog's Death
  5. Ontological Lacunae
  6. Conceiving Vada
  7. The Meaning of Death
  8. Souls and Shadows
  9. Conclusion
  10. References

Two decades after the 1931 episode, the anthropologist Cyril Belshaw reported being given a Motu-Koita version of events in which the dog was killed to the satisfaction of the European officials and then successfully brought back to life. He commented ‘The significance of the incident is that belief in sorcery is so strong as to create a myth accepted as historical evidence, yet based on an incorrect version of events’ (1957:194). In contrast I believe there could be no correct or incorrect version, given the degree to which the nature, motives and intentions of the indigenous participants can be questioned. And here I should remove the brackets I applied earlier to the question of whether the indigenous participants were death-dealing ‘sorcerers’ at all. Given the difficulty and rigours of gaining expertise in magical death-dealing and the jealousy with which it was guarded (see for example Monsell-Davis 1981:242–4; Stephen 1995:177–85; Van Heekeren 2007:417–21), the colonial authorities might have paused to question whether the men who presented themselves in 1931 to publicly reveal the actions and incantations involved were properly qualified in their own society to begin with.

I have interrogated here the assumption by Europeans in the early colonial period that the Motu-Koita were ruled by superstition, I have attempted to critique the European construction of sorcery and sorcerers from poorly understood examples of the use of incantatory rituals in the indigenes' encounters with human and non-human entities, and I have suggested that the eschatological assumptions of Europeans impeded their understanding of indigenous concepts such as mase and lauma. These errors and shortcomings are nicely exemplified in the dog-killing episode of 1931 and the naïve hope of the colonial administration and the government anthropologist that indigenous belief in sorcery would be disturbed by the failure to revive the dog. I believe that the significance of the radical disparity between the European view of the outcome and that presented to Belshaw is not that it is evidence of an ‘incorrect version of events’ on the part of the Motu-Koita, but that it is evidence of a failure on the part of the Europeans to properly understand the cosmo-ontology that informed the indigenous participants.

More than three-quarters of a century later, I cannot presume to achieve substantial insight by comparison. My contemporary Motu-Koita interlocutors can tell me little, for a century and a quarter of Christian influence, medical-scientific intervention and the sanitization of mortuary processes has taken its hermeneutic toll. Even the elderly, unaware of the archival material, complain humorously that I seem to know more about their distant past than they are supposed to. Yet ‘sorcery’ as the Motu-Koita have learned to call it, is still practised, and villagers obliquely indicate to me people whom they understand to be practitioners. We cannot know whether it is the same as the ill-defined ‘sorcery’ which preoccupied Murray's administration. The Motu-Koita learned an important lesson from that era, not – as Murray had hoped – that ‘sorcery is all rubbish’ (1926:12), but that circumspection is wise when foreigners enquire about it. Unless one is prepared to become an apprentice, as Michele Stephen was in fieldwork among the nearby Mekeo people (Stephen 1995), people are reluctant to elucidate its procedure in precise detail, let alone admit to being adept in it.

Knowledge about contemporary sorcery practices, however, is not my objective here. This article, as I stated in its introduction, is part of a project to do analytic justice to the cosmo-ontology of the Motu-Koita as it was when the early missionaries and colonial officers denigrated it as superstition. A comparative and phenomenologically sensitive historical anthropology drawing on the findings of archaeology and linguistics holds the possibility of some redress for those interpretive shortcomings, as I have indicated here in revisiting the events of 1931. Other documented encounters of the early contact period offer the promise of insight through critical revisitation. Perhaps, then, the dog did not die in vain.

Notes
  1. 1

    See for example Chatterton (1968), Dutton (1969, 1994), Oram (1981), Swadling (1977, 1981). The Motu call the Koita ‘Koitabu’: both forms are used in lay reference. I will use Koita, the traditional self-referential term of the group.

  2. 2

    See for example Goava (1979), Groves (1957), Gwilliam (1982), Nou (1975), Oram (1963–69), Price (1975).

  3. 3

    A historically important example is Do Kamo, published in 1947 by the French missionary Maurice Leenhardt (Leenhardt 1998), which displayed some evolutionism but made a serious attempt to grasp New Caledonian Kanak ontology, indicating the potential of a phenomenological turn which took some decades to eventuate in anglophone anthropology.

  4. 4

    ‘Pourri-Pourri’ or ‘puripuri’ was a quasi-indigenous term for ‘sorcery’ or ‘magic spell’ that gained some currency among Europeans in the early twentieth century. Its origin is uncertain. The government anthropologist F.E. Williams identified it as a Suau (a Milne Bay society) language term (Papuan Villager 1929:1) and used it repeatedly during the 1930s in his monthly villager-directed newspaper The Papuan Villager, which was widely distributed around the south-east mainland. The newspaper may have been instrumental in spreading the use of the term (Flynn was in the area from 1930 till 1933). The spelling ‘puripuri’ gradually became conventionalized as a pidgin reference, and is today popularly regarded as a Motu language term. Its occasional use continues, among English-speaking south-coast Papua New Guineans, as a nominal and adjectival generalization in English conversation for malevolent ‘sorcery’, e.g. ‘There's a lot of puripuri going on’ or ‘I'm afraid of being puripuried’.

  5. 5

    In his major ethnography of 1922, based on fieldwork in the Trobriand Islands beginning in 1915, Malinowski laid considerable emphasis on having done participant observation (Malinowski 1922:passim), to the effect that he is popularly thought of as the first to conduct fieldwork in such a manner. Landtman (1927) did not describe his own, earlier, research among the Kiwai in the same terms, but there is now little doubt that he conducted fieldwork in the fashion whose provenance has been credited to Malinowski (see Lawrence 2010).

  6. 6

    ‘To [the Azande], the difference between a sorcerer and a witch is that the former uses the technique of magic and derives his power from medicines, while the latter acts without rites and spells and uses hereditary psycho-physical powers to attain his ends’ (Evans-Pritchard 1976:177).

  7. 7

    Early colonial representations of Motu and nearby languages were phonologically inconsistent. It is not clear whether this was the result of difficulty in discerning indigenous phonetics, or of a variety in dialect among indigenous interlocutors, or of organic processes of language change (on the latter see for example Dutton 1982a; Ross 1994). The inconsistency is particularly noticeable in respect of labial and alveolar stops: for example the name of the main Motu-Koita village cluster adjacent to Port Moresby was variously written ‘Anapata’, ‘Hanuapata’, and ‘Hanuabada’, until the last of these spellings became conventional. Lawes demonstrated greater linguistic precision than many of his contemporaries. His representation of Motu distinguished, for example, between the verbal particle vada, (Lawes 1896:12,134) and the noun vata. On the other hand lexical evidence on the prehistory of local Austronesian groups, together with an examination of consonant correspondences among them (I follow Ross 1994) incline me to the view that the verbal particle and the noun were homonyms, both pronounced vada. Either way, Lawes' distinction is nowadays redundant in usage, as within a few decades the spelling and pronunciation of vata had become conventionalized as vada and remains thus today.

  8. 8

    Seligman's representations are possibly ‘impressionistic’ rather than rigorous, but there is no reason to doubt that they portray Ahuia's grasp of English fairly. Hubert Murray also ‘quoted’ Ahuia in correspondence in the 1930s (for example Murray 1970:135–6) representing his English as relatively fluent. It suggests an improvement in the latter's English over two decades, unless Murray was grammatically correcting Ahuia's utterances.

  9. 9

    In distinguishing groups by name here I am following indigenous cultural preferences, rather than current linguistic conventions. For example, linguists have come to regard ‘Hula’ (spoken by the Vula'a) as a dialect of Keapara, and Balawaia as a dialect of Sinaugoro (see Ross, Pawley and Osmond 2007:347). The Vula'a, however, regard themselves as historically distinct from the Keapara (see Van Heekeren 2010:48–9) and the Balawaia represented their relationship to Sinaugoro ambiguously to their principal ethnographer John Kolia (1976:20). Further ambiguity can be found in names of indigenous groups before they became conventionalized during the colonial period: the most confusing perhaps being the group now known as Nara, who were once also known as Lala, Ala'ala, and Pokau, depending on which of their neighbouring groups Europeans happened to be talking to (and for other complications in the vicinity see Jones 1998:9,10n15). As Kolia observed while wrestling with language and cultural distinctions, the first foreigners arriving in the area ‘saw tribes and chiefs where there were coalitions of clans and clan leaders. Because these terms were not defined we cannot be sure of whom the early ethnographers were speaking. This ambiguity is reflected in the way people see themselves’ (1976:19–20).

  10. 10

    For examples of the way linguists currently schematize Austronesian languages see Ross (1988) and cf Ross, Pawley and Osmond (2007).

  11. 11

    For example mea in Motu, Roro, mega in Mekeo, Sinaugoro. The term is a root which can take either a noun or verb form. Reduplication often (but not always: see Taylor 1970) indicates a qualitative intensification; meamea can be a ‘whispered’ incantation. While Hula (the language of the Vula'a) has a high cognatic correlation with Motu a different term, inaina, is used for incantation and – interestingly – also for intimate conversation (D. Van Heekeren, pers. comm.).

  12. 12

    Hau'ofa (1981) renders this as ungaunga. In using ugauga I follow Stephen (1995), who treats the topic in more depth and with closer attention to linguistic nuance.

  13. 13

    Linguistic evidence strongly suggests that Koita broke away from the Koiari in the distant past and moved towards the coast (Dutton 1969, 1994). The same movement is suggested by oral history (Dutton 1969; Gadiki 1971; Oram 1981).

  14. 14

    At the end of the colonial period this was renamed ‘Hiri Motu’. The change was encouraged by romantic arguments that it was originally developed and spread along the coast by way of the traditional Motu trading voyages known as hiri, during which Motu-made pots and shell ornaments were exchanged for sago. Dutton has convincingly shown, however, that the trading language used on the voyages was different from the lingua franca spread mostly via the police force of the early colonial administration (Dutton 1985). In Pari, the Motu village where I have conducted fieldwork, the lingua franca is still referred to as Police Motu by elderly villagers, who use it only when necessary to communicate with outsiders – if English cannot be substituted.

  15. 15

    I am inclined to treat Lett's reference to ‘Koita’ instead of ‘Koiari’ in the 1942 recollection mentioned above (1942:163), as a slip of the pen. It should also be noted that the Port Moresby-based Europeans of Murray's time had only partially explored the inland area, and would not have mapped language boundaries: ‘Koiari’ was a generalization for mountain dwellers north and east of Port Moresby.

  16. 16

    See for example Lawes 1885, 1896; Lister-Turner and Clark 1930, 1931a, 1931b.

  17. 17

    Here I follow a seminal discussion by Nichols (1986), where a head is defined as ‘the word which governs, or is subcategorized for – or otherwise determines the category of – the other word. It determines the character of its phrase’ (1986:57). A phrase (for example within a longer sentence) can thus be viewed as having a ‘head’ and a ‘dependent’. For example in the phrase ‘The man's dog. …’, ‘dog’ is the head. The rest of the phrase, telling us that the dog is possessed by the man, syntactically ‘depends’ on the head. In this English-language phrase the possessive relationship is ‘marked’ on the noun of the dependent, thus /'s/ is added to ‘man’. In Motu, where tau is ‘man’ and sisia is ‘dog’, the equivalent phrase would be ‘tau sisiana. …’. Here we see that the possessive relationship is marked not on the dependent (tau) but on the head noun sisia, by the addition of /na/ (a 3rd-person-singular possessive suffix). Thus where English is characterized by ‘dependent-marking’, Motu is characterized grammatically by ‘head-marking’.

  18. 18

    See Hau'ofa (1981:169–76), Monsell-Davis (1981:136–47), Seligman (1910:159–66), Stephen (1995:32–3), Van Heekeren (2007:414), and cf. Hertz (1960).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. The Dog's Death
  5. Ontological Lacunae
  6. Conceiving Vada
  7. The Meaning of Death
  8. Souls and Shadows
  9. Conclusion
  10. References
  • Belshaw, C. 1957. The Great Village. London: Routledge and Kegan Paul.
  • Chatterton, P. 1968. The story of a migration. The Journal of the Papua and New Guinea Society 2(2): 9295.
  • Dutton, T. 1969. The Peopling of Central Papua. Pacific Linguistics Series D–9. Canberra: Australian National University.
  • Dutton, T. 1982a. Towards a history of the hiri: some beginning linguistic observations. In T. Dutton (ed.) The Hiri in History: Further aspects of long distance Motu trade in Central Papua, pp. 6598. Canberra: Australian National University.
  • Dutton, T. 1985. Police Motu: Iena Sivarai. Port Moresby: University of Papua New Guinea Press.
  • Dutton, T. 1994. Motu-Koiarian contact in Papua New Guinea. In T. Dutton and D. Tryon (eds) Language Contact and Change in the Austronesian World, pp. 181282. Berlin: Mouton de Gruyter.
  • Dutton, T. (ed.) 1982b. The Hiri in History: Further aspects of long distance Motu trade in Central Papua. Canberra: Australian National University.
  • Evans-Pritchard, E. 1976. Witchcraft, Orcales, and Magic among the Azande. Oxford: Clarendon Press.
  • Gadiki, H. 1971. The Pure Koitabu or Rokurokuna. Oral History 1(1): 2932.
  • Goava, S. 1979. Kori Taboro ena sivarai. Oral History 7(3): 95123.
  • Gostin, O. 1986. Cash Cropping, Catholicism and Change: Resettlement among the Kuni of Papua. Pacific Research Monograph No 14, Australian National University National Centre for Development Studies.
  • Groves, M. 1954. Dancing in Poreporena. Journal of the Royal Anthropological Institute of Great Britain and Ireland 84: 7590.
  • Groves, M. 1957. Sacred past and profane present in Papua. Quadrant 1(3): 3947.
  • Guise, R. 1891. Vocabulary of Bula'a dialect, spoken in the Central District of British New Guinea, by some coast tribes. British New Guinea Annual Report, 1 July 1890 to 30 June 1891.
  • Gwilliam, J. 1982. Some religious aspects of the Hiri. In T. Dutton (ed.) The Hiri in History: Further Aspects of Long Distance Motu Trade in Central Papua, pp. 3563. Canberra: Australian National University.
  • Hau'ofa, E. 1981. Mekeo: Inequality and ambivalence in a village society. Canberra: Australian National University Press.
  • Hertz, R. 1907. Contribution à une étude sur la représentation collective de la mort. Année Sociologique 10: 48137.
  • Hertz, R. 1960. Death and the Right Hand. London: Cohen and West.
  • Jones, A. 1998. Towards a Lexicogrammar of Mekeo (An Austronesian Language of Western Central Papua). Pacific Linguistics Series C–138. Canberra: RSPAS, Australian National University.
  • Kapferer, B. 1997. The Feast of the Sorcerer: Practices of Consciousness and Power. Chicago: University of Chicago Press.
  • Kolia, J. 1976. The History of the Balawaia. Port Moresby: Institute of Papua New Guinea Studies.
  • Laloki [Errol Flynn]. 1931. ‘The case of the dead dog’, The Bulletin, weekly magazine, 2 Dec, p. 21.
  • Landtman, G. 1927. The Kiwai Papuans Of British New Guinea. London: Macmillan and Co.
  • Lawes, W. 1879. Ethnological notes on the Motu, Koitapu and Koiari tribes of New Guinea. Journal of the Anthropological Institute of Great Britain and Ireland, 8: 369377.
  • Lawes, W. 1885. Grammar and Vocabulary of Language spoken by Motu Tribe (New Guinea). Sydney: Government Printer.
  • Lawes, W. 1896. Grammar and Vocabulary of Language spoken by Motu Tribe (New Guinea). Third and Enlarged Edition. Sydney: Government Printer.
  • Lawrence, D. 2010. Gunnar Landtman in Papua 1910 to 1912. Canberra: ANU EPress.
  • Leenhardt, M. 1998. Do Kamo: La Personne et le Mythe dans le Monde Mélanésien. Paris: Gallimard.
  • Lett, L. 1942. The Papuan Achievement. Melbourne: Melbourne University Press.
  • Lett, L. 1949. Sir Hubert Murray of Papua. London: Collins.
  • Lévi-Strauss, C. 1966. The Savage Mind. London: Weidenfeld and Nicolson.
  • Lister-Turner, R. 1924. ‘Malignant witchcraft in Papua and the use of poisons therein’, Man 24: 117119.
  • Lister-Turner, R. and J. Clark. 1930. Revised Motu Grammar and Vocabulary. Port Moresby: Government Printer.
  • Lister-Turner, R. and J. Clark 1931a. A Dictionary of the Motu Language of Papua. Sydney: Government Printer.
  • Lister-Turner, R. and J. Clark 1931b. A Grammar of the Motu Language of Papua. Sydney: Government Printer.
  • Lister-Turner, R. and J. Clark 1954. A Dictionary of the Motu Language of Papua, Second Edition. Sydney: Government Printer.
  • Malinowski, B. 1922. Argonauts of the Western Pacific: An account of Native Enterprise and Adventure in the Archipelagoes of Melanesian New Guinea. London: Routledge.
  • Malinowski, B. 1989. A Diary in the Strict Sense of the Term. Stanford: Stanford University Press.
  • Malinowski, B. n.d.Fieldnotes’. Photocopy in Nigel Oram Papers 1963–69, Box 5, Folder 32. National Library of Australia, Canberra.
  • Mimica, J. 2010. Un/knowing and the practice of ethnography: A reflection on some Western cosmo-ontological notions and their anthropological application. Anthropological Theory 10: 203228.
  • Monsell-Davis, M. 1981. Nabuapaka: Social Change in a Roro Community. Unpublished PhD thesis. Macquarie University, New South Wales.
  • Murray, H. 1925. Papua of Today, London: P.S. King and Son.
  • Murray, H. 1926. Native custom and the government of primitive races with especial reference to Papua. Paper read to 3rd Pan-Pacific Science Congress, Tokyo. Port Moresby: Govt Printer.
  • Murray, H. 1970. Selected letters of Hubert Murray, ed. F. West . Melbourne: Oxford University Press.
  • Nichols, J. 1986. Head-marking and dependent-marking grammar. Language 62(1): 56119.
  • Nou, I. 1975. Pari hanua edia sene sivaraidia. Port Moresby: Institute of Papua New Guinea Studies.
  • O'Malley, J. 1912. Native affairs. Papua Annual Report, Year ending 30 June 1912, pp. 98105. Canberra: Government Printer.
  • Oram, N. 1963–69. Nigel Oram Papers, MS9436. Canberra, Australian National Library.
  • Oram, N. 1981. The History of the Motu-speaking and Koitabu-speaking Peoples according to their own Traditions. In D. Denoon and R. Lacey (eds) Oral Tradition in Melanesia, pp.207229. Port Moresby: Moresby: University of Papua New Guinea and Institute of Papua New Guinea Studies.
  • Papua Annual Report [PAR]. 1907. Annual Report for the year ending 30th June 1907. Commonwealth of Australia, Government Printer, Canberra.
  • Papua Annual Report [PAR] 1908. Annual Report for the year ending 30th June 1908. Commonwealth of Australia, Government Printer, Canberra.
  • Papuan Villager, The. 1929. Magic. The Papuan Villager, monthly newspaper, 1(12): 12.
  • Papuan Villager, The 1931. Sorcery. The Papuan Villager, monthly newspaper, 3(11): 8182.
  • Price, A. (ed.) 1975. Traditional Motu Customs. Port Moresby: Institute of Papua New Guinea Studies.
  • Romilly, H. 1889. From My Verandah In New Guinea: Sketches and Traditions. London: David Nutt.
  • Ross, M. 1988. Proto Oceanic and the Austronesian Languages of Western Melanesia. Pacific Linguistics, Series C–98. Canberra: RSPAS, Australian National University.
  • Ross, M. 1994. Central Papuan culture history: some lexical evidence. In A. K. Pawley and M. D. Ross (eds) Austronesian Terminologies: Continuity and Change, pp.389479. Pacific Linguistics Series C–127. Canberra: RSPAS, Australian National University.
  • Ross, M., A. Pawley and M. Osmond. 2007. The Lexicon of Proto Oceanic: The Culture and Environment of Ancestral Oceanic Society. Vol 2 The Physical Environment. Canberra: ANU EPress.
  • Seligman, C. 1910. The Melanesians of British New Guinea. Cambridge: Cambridge University Press.
  • Shelley, R. 1978. Ahuia Ova: new insights into the life of a prominent Papuan. Oceania 48(3): 202206.
  • Stephen, M. 1995. A'aisa's Gifts: A study of Magic and the Self. Berkeley: University of California Press.
  • Strong, W. 1919. Some personal experiences in British New Guinea. The Journal of the Royal Anthropological Institute of Great Britain and Ireland 49: 292308.
  • Swadling, P. 1977. A review of the traditional and archaeological evidence for early Motu, Koita and Koiari settlement along the central south Papuan coast. Oral History 5(2): 3743.
  • Swadling, P. 1981. The settlement history of the Motu & Koita speaking people of the Central Province, Papua New Guinea. In D. Denoon and R. Lacey (eds) Oral Tradition in Melanesia, pp. 240251. Port Moresby: University of Papua New Guinea and Institute of Papua New Guinea Studies.
  • Taylor, A. 1970. Reduplication in Motu. In S. A. Wurm and D. C. Laycock (eds) Pacific Linguistic Studies in Honour of Arthur Capell, pp. 12351244. Pacific Linguistics Series C–13. Canberra: RSPS, Australian National University.
  • Taylor, A. 1977. Missionary lingue franche: Motu. In S. Wurm (ed.) New Guinea Area Languages and Language Study Vol. 3: Language, Culture, Society, and the Modern World, Fascicle 2, pp. 881891. Pacific Linguistics Series C–40. Canberra: RSPS, Australian National University.
  • Taylor, A. 1978. Evidence of a pidgin Motu in the earliest written Motu materials. In S. Wurm and L. Carrington (eds) Second International Conference on Austronesian Linguistics: Proceedings, pp. 13251350. Pacific Linguistics Series C–61. Canberra: RSPS, Australian National University.
  • Territory of Papua Annual Report [TPAR]. 1933. Report for the year 1931–1932. Commonwealth of Australia, Canberra: Government Printer.
  • Turner, W. 1878. The ethnology of the Motu. Journal of the Anthropological Institute of Great Britain and Ireland 7: 470499.
  • Van Heekeren, D. 2007. The essence of Vula'a historical consciousness. History and Anthropology 18(4): 405426.
  • Van Heekeren, D. 2010. Cracks in the road: trying the moral limits of capitalism in a Hula village. In M. Goddard (ed.) Villagers and the City: Melanesian Experiences of Port Moresby, Papua New Guinea, pp. 4774. Wantage: Sean Kingston Publishing.
  • Van Heekeren, D. 2012. The Shark Warrior of Alewai: A Phenomenology of Melanesian Identity. Wantage: Sean Kingston Publishing.
  • Williams, F. 1939. The reminiscences of Ahuia Ova. Journal of the Royal Anthropological Institute of Great Britain and Ireland 69(1): 1144.