Huli War and Compensation

by Dr. Laurence R. Goldman

“Brothers, can you give me some pigs so I can close the wound?”
(Aumai.D.9:4 32-433)

In the preceding section I was concerned with certain self-help mechanisms as a unified field of activity and meaning associated with negative imposition – the inducement of states of illness and pain. That discussion relates directly to the subjects of warfare and compensation to which I now turn. Both of these have been given considerable ethnographic and analytic attention by Glasse (cf.1959b,1968), and it is therefore not my intention to here duplicate his admirable accounts. Rather, I focus on particular facets of physical violence and indemnity that impinge upon the general themes pursued herein concerning the pre-eminence of “talk” in Huli society. Glasse’s analyses were addressed to the problem of how social control could be maintained in a society lacking centralised authority. He concluded that the resolution to this enigma resided in the mechanisms of revenge and redress that inhibit the formation of enduring coercive factions:

“Revenge and redress ally and oppose different combinations of groups over a period of time and thereby promote an uneasy integration and a kind of balance within the wider social order”(1959b:274).

Revenge in Huli, according to Glasse accounts, is obligatory and defines units while the processes of redress engender internal divisions within these same groups. The parameters of the system are articulated by the principle that revenge seeks always to inflict a greater injury, or demand a higher compensation, for real or imagined injury. The jural axioms that both constitute and regulate settlement procedures are founded on the premise that for every confrontation, whether verbal dispute or warfare, there exists two “sources” (tene). In the former instance they are the source of the ‘talk’ (bi tene), and in the latter the source of the ‘war’ (wai tene); on these people devolves responsibility for indemnity. In war, compensation is paid occasionally to enemies with whom close relations exist, and invariably to injured allies. Significantly, Glasse states that the Huli phrase the above in individual rather than group terms, through, on account of obligations to assist section members, the “burden of death compensation during war falls upon the two local descent groups” (1959b:286).

To some Melanesian researchers the supposition of an order into which disruption is introduced is deemed to be analytically constricting. It is, they argue, a perspective that engenders a false hiatus between social reality – embracing breaches – and ethnographic reconstruction. The indigenous model of the Huli encapsulates both viewpoints: disputes are essentially a deviance from the straight(order), a “knot”. However, settlement is merely a phase in a more enduring and cyclic pattern of hostilities -talk never dies because there is a recurrent oscillation between states of order and disorder. The model of war/dispute “initiators” is consistent with the nature of Huli antagonism as an adversary system; as such warfare is not seen as discontinuous behaviour from other manifestations of conflict. It is structurally similar to U Wai(“song-war” for brides in Dawe), lai (verbal argument), and the competitive facets of Haroli and Mali. The paired terms gungu and wai, ¹ which for convenience I gloss as ‘fighting’ and ‘warfare’ respectively, are not defined by types of weapon used but rather by the perceived situational context of the conflict. As situational descriptions the former is applied to ‘domestic’ involvements, while the latter is most usually reserved for ‘unit’ opposition. The type of fighting, duration or numbers involved do not determine indigenous applications of these terms which are perhaps better interpreted as signifying simply ‘conflict’. This is an important observation since both gungu and wai may be waged with verbal or non-verbal weapons; words and arrows are interchangeable terms in the “talk about conflict”. The basis for their application may be located along the semantic dimension of control (cf.Chapter 3) as realisations of ‘disorder’. They are reprehensible, in either their verbal or non-verbal manifestations, because they represent an individualistic appropriation of action contrary to the norm that “talk should be shared”. It is a contrast represented by the opposing notions of Iba Tiri – a person without real ‘talk’ – and Manali – the person with real ‘talk’. In the Huli schema of reality there is a cyclic transition between the activities of war / fighting (no-talk) and talk; this is succinctly stated by Kumuria in D.9:141-144: “When the pigs have been stolen, the fight done, the houses burnt then we used to collect all the men and talk. We never used to do these things by ourselves. Did you ever make decorations by yourselves (kau wialu mbira biagabe)?” The Huli conceive the transition to be from the domain of egocentric assertion to the sphere of social exchange. At this juncture we can begin to appreciate how the discussion of semantic dimensions (in Chapter 3) and the analysis of boundary discrimination (Chapter 4) coalesce. The former, I argue, provide the terms and continua that are used to regulate sanitization processes; they define what behaviour is to be “quarantined” and what may be allowed to transfer from anda (home / inside) to hama (outside). These axes fire thus interstitial within the framework depicted by Figure 7; the resultant model is one that I contend offers real interpretative and explanatory insights into most forms of social action in Huli.

It reveals and orders the behavioural rationalities behind pig production, warfare and dispute, and decoration while enabling us to comprehend the bases for evaluation, definition and negotiation of any situation. It may be that further dimensions can be elicited for other fields such as illness, but the fundamental distinctions and polarities remain central to Huli thought and language. Both deviance and conformity can be dealt with by application of the interpretative concepts used to a assess and relate disparate modes of action. I have been concerned in this thesis with the stated interrelations between talk and decoration, but an equally detailed analysis of economic or ritual behaviour would, I believe, demonstrate a parallel ordering of phenomena along similar parameters. The social process in Huli is at some stage an involvement with “talk” and as such is determined by the configuration of meanings
presented in Figure 7 and the previously elicited semantic dimensions. With regard to the following discussion of warfare, compensation and production I have restated the terms of debate in Figure 9. As the quotation above shows, war and debate are opposed individual (revenge) and social (redress) action. The former is regulated, like so many other aspects of anda, asymmetrical because initiative emanates from individual will, it is both covert and futile. It partakes of the negative import associated with women – anger and lack of control. The transition to debate, to the zone of hama, necessitates that the covert be made ‘open’. The patterned expectation of deceit and duplicity manifests the understanding that all ‘anda’ behaviour is not ‘publicly shared’. Individuals here act towards each other as they act towards Dama; it is not countenanced but it is presumed. Figure 7 also illustrates that situations are controlled through the medium of talk, meaning is conveyed by speech through both the regulation and change of vocabulary. My observations are ethnographically descriptive not analytic, they translate what is explicit and implicit in the texts I have presented as well as the nomenclature of speech. It is not simply people that the Huli seek to control by talk, but events and outcomes. The prevalence of Gamu is a ritualistic adaptation of concepts of action to process, and represents the quintessential ‘power’ of the verbal component. The communicative importance attached to talk in Huli renders it the most suitable window through which to view the culture.

The paramountcy attached by the Huli to the individual in war, is an emphasis abstracted and remarked upon by Melanesian anthropologists in their comparative studies. While I must disagree, on the basis of data presented in Chapter 4, with Berndt’s (1964: 203) conclusion that marriage and warfare are incompatible actions in the Huli, it is nevertheless the case, as Strathern(1972:74) intimates, that the Huli system contrasts with Highland models based on complex enemy-ally relations, paired fighting units, and interconnected structures of war and feasting. In presenting the results of my own research I had in mind two principal aims: (1) to assess the extent to which the prevailing model of Huli warfare – founded on Glasse’s ethnography of the Tari region – might be qualified for communities such as Ialuba; and (2) to enhance our understanding of redress by augmenting structural-functionalist approaches with data gleaned through attention to semantic morphologies. With regard to a subject such as compensation, the former approach – historically “traditionalist” – concerns itself with questions pertaining to the range of social relationships, types of issue, degree of effectiveness, and strains in the execution and arrangement of indemnity. While these are certainly important concerns, they do not of themselves explain what compensation represents to the people themselves. We need to ask how the nomenclature and talk about compensation provide indices of “meaning”, and how this enlightens us concerning why “compensation” is given at all. Furthermore, what clues does the taxonomy of indemnity mediums provide with respect to the import of all transactions in the social network? The tenor of the argument which follows is that compensation is an extension of the system of meanings inherent in the social reality of Huli talk.

In order to comprehend the kind of transfer compensation represents we need to remind ourselves of the predominant observation made in Chapter three. I there argued that an indigenously stated structural homology existed between the pathology and treatment of sickness and speech; “talk about talk’ manifested the range of shared meanings between the two domains. Illness is a source of metaphor used to express fundamental truths about talk. Pain and knotting are the basis of the Huli con c ept of negative attributes not only in speech evaluation, but in behaviour concerned with negative imputation (e.g. sorcery). In this respect one might reasonably infer that compensation in Huli is a symbolic application of “medicine”, a “healing” rite metaphorically consistent with the perceived an analogies between sickness and dispute. How far this might apply to other Highland systems is difficult to gauge at this stage, but it indicates the need to examine types of interrelation between action and metaphor in cross-cultural a analyses of dispute and dispute-settlement. It is a level of explanation sadly neglected by ethnographers of compensation in Papua New Guinea. To what extent then do the data support and reflect the above? The term for any kind of inde unity in Huli is abi which I contend to be etymologically related to dabi meaning “to heal / recover / get well”; the transfer of abi is the giving of something to relieve pain. The term is most usually compounded in a format of adjectival qualifier+abi, the former constituent denoting the type of breach for which payment is made. For example: ‘taga abi (‘shame compensation’:ref.D.3:3), keba abi( ‘anger compensation’:ref.D.3), mege abi (‘insult compensation’:ref.D.3 and D.4, part 8), and agali abi (‘man compensation’:refers to payments for the death of an ally or enemy). Whatever the internal specification or qualification of ill, the payment is made – and my previous discussion of anger and shame as illness states implies this – to “rub on the sore” (D.3:192-193). The speculative nature of the hypothesis concerning abi and dabi is identical to the unequivocal evidence contained in “compensation talk”. A secondary mode of reference to the type of indemnity issue makes use of a terminological repertoire denoting the pigs involved. I have set these out, as well as other closely related forms of pig transaction – in Table 26. These expressions, like kin terms, are shorthand normative formulations that indicate who should give and receive abi; they both describe the “topic” of payment as well as restating what function the pigs fulfill. The terms thus encode the significant meanings relevant to interaction in this field. The phrase Nogo Nigi (Table 26) thus denotes a situation of ally indemnity, as well as terminologically reflecting the figurative equivalence of pigs and healing leaves (nigi). The obligation to compe nsate is a corollary of the normative prohibition on causing sickness within the parish. The existential basis of the nomenclature given in Table 26 is the cultural validity of the above statement. The categories of abi can be conceived as a class of impersonal directives that implicitly recognises the reprehensible nature of illness, and the need to provide “relief/cure”.

The command to indemnify is the logical equivalent of the prohibition against forbearing compensation. The logical interrelations between these act-categories have been long ago pointed out by deontic philosophers concerned with the concept of a norm; the prescriptive forms presuppose that the subject of the directive can (ability) do the enjoined or permitted thing. In regard to illness, the meaning universe stipulates that states of pain rarely develop “of themselves”, they are explained by the causative influence of man or Dama. The possibility thus exists that an actor can produce changes in states of health either positively – by doing, verbally or non-verbally, x – or negatively defined, by forbearing to do x. The indemnity system is thus underpinned by these moral postulates concerning the production of illness. The fundamental directive category of obligation -the ought, must, should, duty and right of our own normative discourse – is expressed and encapsulated in Huli by the application of abi categorisations; their ontological criteria are thus located at the level of linguistic reality. The taxonomic terms are the legal and moral modalities of the compensation system, expressing pragmatic(producing effect) and semantic (bearing meaning) considerations. The reference systems of abi and nogo are to a certain degree interchangeable, and one may denote ally compensation by either the phrase timu abi (‘arrow indemnity’) or Nogo Nigi.

A second important feature of the terms given in Table 26 is that they appear also to apply to Duna payments:
“Whenever an ally kills an enemy or enemy ally, the killer’s ‘hand’ is ‘backed’ (cf.Nogo Gima Table 26) for each injury sustained by an ally, the wei tse must kill pigs as kaka. Each death of an ally must finally be reciprocated by Damba (Modjeska 1977:266).

Damba pigs are thus central to both the Huli and Duna indemnity systems, but for the former we are able to perceive a consistency in meaning behind the various contextual applications of the term. Damba means essentially to “close, close over”, so that when bride- price pigs are collectively denominated Nogo Damba (ref,Chapter:1) the transaction is conceived as being thereby terminated. The parties do not enter into any further exchanges on the basis of that transfer. This meaning is manifested by all applications of the phrase Nogo Damba to economic exchanges. Precisely the same concept inheres in the speech genre Damba Bi (ref. Table 5) used traditionally on occasions of war or compensation ( and thus paralleling the Melpa form elik) his oratory effectively “closed” all talk on any issue because control of speech resided largely with headmen. In Chapter three I illustrated how one may use Damba Bi in a censorial fashion for any talk that “forecloses” on another person’s verbal contribution (ref.D.5:157-160).

Some indication of the stylistic properties of this genre can be gained from a reading of the example given in Appendix 9. Assertion of status as a manali (polymath) is ensured by reference to Dindi Pongo mythology (lines 1-8) and the distinctive decoration markers of Huli and Duna (lines 49-64). Characteristically, deference is absent as the speaker egotistically affirms himself as a man whose “mind is straight” (lines 65-66) and one who used to perform Ega Hangaga for Tege (lines 67-68). The speech is particularly relevant to the discussion of compensation. The men who have been killed are talked of as marsupials (lines 9-16) and the audience is enjoined to collect “bamboo containers” (i.e. pigs). The closing of the indemnity issue -is metaphorically developed as a “closing” of the bamboo containers with leaves (i.e.pigs:lines 17-30). Settlement phraseology reflects the notion of termination of hostility phases. The ideological clarity of the system breaks down in praxis on account of the workings of concepts of responsibility and liability – the interpretation of tene examined in chapter three. The sources of any war or dispute are the tene, like agnates (tene) they are the “origin”. The process of resolution is one concerned to locate issues and persons as ‘first causes’. However, as previously noted, these are continually redefined in any debate according to the level of information made available (see discussion following D.3). Consequent upon the changing chronological position of issues in the causal chain, so culpability is continually reallocated to different sets of persons. Thus Walumbu in D.3 attempts to shift blame to others by citing their laughter as the cause of her brother’s anger. The search for tene encompasses the history of relations between the disputants so that with an increasing time depth to the events mediators realign their perceptions of ’cause’ – “This talk now is the source of it all”(D.3:113). The boundaries of the process are set by the past relationships between the parties, and the nature and amount of issues left unsettled. We can expect a contrast in the kinds of resolution patterns prevalent in different types of dispute, and it seems expedient at this juncture to elicit the main statistical trends suggested by the data presented in Appendix 6. Within the given geographic and temporal parameters, the Tables nevertheless indicate what part compensation plays in the overall picture of settlement – and I use this term without any “once and for all” connotation – procedures. With the presentation of types of case I have related observations to statistical facets for that class, so that in what follows I focus attention more directly on the discriminated resolution forms – namely, pending, compensated and renounced claims.

Table 26: The Semantics of Compensation

To be continued…

(An extract from Talk Never Dies: An Analysis of Disputes Among the Huli. A thesis submitted by Laurence R. Goldman for the degree of Doctor of Philosophy, University College, London. February, 1982., p. 340-351.)

1. This may be semantically related to the tonally separate wai (seed) – the seed of future strife [↩]