Volume 99 1990 > Volume 99, No. 2 > Competitive equality in Melanesia: An exploratory essay, by N. McDowell, p 179-204
COMPETITIVE EQUALITY IN MELANESIA: AN EXPLORATORY ESSAY 1
Anthropologists are beginning to explore seriously the range of variation in traditional Melanesian political systems (see, for example, Berndt and Lawrence 1971, Burridge 1975, Chowning 1979, Godelier 1986, Keesing 1985, Lawrence 1969, Lindstrom 1981). 2 It is no longer appropriate to gloss a particular ethnographic case as an example of “the Big Man system” without significant qualification and elaboration. One of the themes emerging from this exploration is the idea that one cannot assume some kind of egalitarian relations as given and then go on to explore how hierarchy evolves from basic equity. As James Flanagan (personal communication) writes, “egalitarian bases of social action (including group formation)… are frequently taken for granted, a ‘natural condition’, a residual category, requiring no socio-cultural explanation either of their origin or continuity”. But equality, however defined, is problematic and cannot be taken for granted: one must understand the processes that generate equitable relations and not assume that humanity's natural state is one of egalitarianism out of which inequality evolves. Josephides (1985:1) persuasively argues that all systems contain within them the “seeds of inequality”, and anthropologists need to ask why these seeds do not grow— what is it about egalitarian systems that makes and maintains them as egalitarian? Lederman (1986a:5) questions whether egalitarian societies are necessarily more “traditional” than hierarchical societies, and echoes Josephides and Flanagan when she says that:
we have a tendency to take “egalitarian” organisation for granted. They always come first in our prehistorical reconstructions, as if they are closer to a state of nature; hierarchy is more often conceived as difficult to organize, a cultural and political achievement. But it may be more productive to assume that both kinds of polity require active cultural attention since each contains the possibility of the other within it as a threat …(1986a:24; see also Lederman 1986b).
We cannot take egalitarianism as a natural quality; we must explore the - 180 processes involved in generating both egalitarian and hierarchical relations. The goal of this essay is simply to explore some of the factors that contribute to the establishment and maintenance of egalitarian relations in Melanesia. 3
Because there are different kinds as well as amounts of equality, it is necessary to place systems that manifest it in a larger context, and I therefore begin with a brief discussion of hierarchy and equality in general. After this preliminary context-setting, I move on to my main concern: an exploration of the nature of systems of competitive equality, i.e., systems in which people must compete to remain equal. I discuss the role that cultural and social constructions of exchange have in the generation of the particulars in these societies and examine the works of several anthropologists to illustrate how various factors, including exchange, might contribute to the establishment and continuation of egalitarian processes. It must be noted that this is a preliminary exploration only; I hope to encourage others to examine their material with some of the ideas presented in mind.
HIERARCHY AND EQUALITY: CONTINUUM AND DOMAINS
Although it would be difficult (but not impossible) analytically to label the actors in any particular relationship equal or unequal, that is, one probably could generate acceptable criteria to define such terms as egalitarian, rank, hierarchy, equality, etc., 4 it is a far more difficult task to so label anything more complex than a simple dyadic relationship. Whole systems never seem to be as clean as that. Hierarchy and equality are not totally separable categories but different positions in a field in which a whole range of possibilities exists. There are, simply, more or less hierarchical systems (or relationships) and more or less egalitarian ones.
One range of variation is the extent to which status, and the equalities and inequalities associated with it, is based on achievement or ascription. One of the interesting things about many Melanesian systems is the extent to which these two mix within the same society. Some peoples rely heavily on ascription and its incumbent hierarchy (e.g., Manam [Lutkehaus 1987], the Trobriands [Malinowski 1961, Weiner 1976], Mekeo [Hau‘ofa 1971]; see also Chowning 1979). Keesing (1985:238) argues that the existence of hereditary titleholders we can legitimately call “chiefs” runs counter to the Big Man stereotype, but we now know that chiefdoms were widely distributed in Melanesia. … Indeed, it now seems very likely that the Oceanic Austronesian-speaking peoples who colonised seaboard Melanesia had hereditary chiefs of the Western Polynesian type, and that the Big Man systems of the ethnographic present reflect political devolution in the past two millennia as regional trade systems disintegrated. … Other systems contain - 181 only a small element of ascription, if any at all. Among the Baruya, for example, only some men inherit magic that allows them to initiate others, thus perhaps giving them a slight advantage in achieving the status of “great man” (Godelier 1986). Many Melanesian systems allow no place for ascription at all (e.g., Bun [McDowell 1978, 1980], Wape [Mitchell 1978, 1988], Tangu [Burridge 1971, 1975]).
We are dealing with fluid systems and processes here, not necessarily static structures. A significant possibility is that any particular relationship and/or system need not statically remain one or the other; like Burmese gumsa and gumlao (Leach 1954), vacillation between inequity and equity is a potential in political process. Alternating equality with inequality is clearly a common event in the context of a particular dyadic relationship between two individuals. The alternation can depend on a variety of factors, such as whose turn it is to give and/or receive or the position of the dyad in a developmental cycle of transaction. Such alternating disequilibrium might also characterise the relationship between exchanging groups or networks. Other questions need exploration as well: to what extent are both present in any system? What role does the tension between hierarchy and egalitarianism have in the sociopolitical dialectic?
The fact that hierarchy and egalitarianism can coexist in the same society but in different realms has caused analytical difficulties. We must be able to specify which domain is characterised by either hierarchy or equity, for human systems are composed of a variety of domains organised in different ways. It is common in Melanesia, for example, for relations among adult men to be at least predominantly egalitarian while those between men and women, between generations, or between senior and junior of one generation, to be unequal and clearly hierarchical (see Godelier 1982, Josephides 1985, Lederman 1986b, Modjeska 1982, and A. Strathern 1982 for examples of analyses that incorporate recognition of this; also see M. Strathern 1987 for a series of essays directly addressing issues pertaining to male-female inequality). The Baruya seem to stress equality more among themselves but inequality with others (Godelier 1986), a not uncommon occurrence. It is probable that different kin relations manifest different qualities: brothers may be equal but brothers-in-law or indebted affines unequal — wife-givers may simply be superior to wife-takers while agnates are defined as equal. Among the Chambri, for example, one can simply never be equal to an affine without becoming unequal to someone else (Gewertz 1983). 5 Furthermore, the nature of the equity or hierarchy in different domains need not be the same; that is, one kind of hierarchical arrangement may obtain in one domain while another kind is manifest in another. It is common, for example, that the inequality - 182 between female and male takes a significantly different shape and texture from the inequality between ritually or generationally senior and junior men. 6
One of the most compelling questions in this area has to do with the nature of the articulation of these domains and how they influence one another, for it is by understanding how domains articulate that we will come to know how systems generate and reproduce and perhaps transform themselves. Significant and suggestive work has already been done on how male-female inequality interfaces with other domains (see especially Godelier 1982 and 1986, Josephides 1985, Lindenbaum 1984, and Modjeska 1982; also Lederman 1986b, Ortner and Whitehead 1981, Read 1984, and M. Strathern 1987).
Understanding these issues of the shading between hierarchy and egalitarianism, of the coexistence of and alternation between the two, of the tension between them, of the differential manifestation of them in a single society, will go a long way in helping understand political process in the Pacific. Although Sahlins articulated only the extreme ends of the continuum and neglected to identify the complexities, these are some of the issues he was grappling with in his classic comparison of Polynesian and Melanesian leaders (1963). Although the general characterisation of Melanesians as egalitarian and Polynesians as hierarchical is too simple, the recognition that perhaps certain systems emphasise hierarchy more than others is correct, and it reminds us that we need to investigate the mix. Detailed investigations of how egalitarian relations and hierarchical ones play themselves out in the same sociopolitical dialectic are beginning to appear for Pacific societies generally. For example, Borofsky (1983:9-10) articulates the coexistence of hierarchy and egalitarianism on Polynesian Pukapuka. Although the atoll society is typically Polynesian in its concern with deference and hierarchy, a strong thread of egalitarianism contributes to the rivalry for status and position. Peterson (1985) has recently argued that Micronesian Ponape is not nearly as hierarchical as it was once portrayed, and Poyer (1987) examines related issues in nearby Sapuahfik.
COMPETITIVE EQUALITY: GENERAL CONSIDERATIONS
If one scans the literature on Melanesia with an eye towards domains, the general impression is that inequality characterises male-female relations and that it is more likely to appear in intergenerational relations and junior-senior relations than equity is. If anything approaching egalitarianism exists, it is to be found among same-sex adults. Here I focus on societies in which equity of one sort or another characterises the relations among adult men. 7 Remembering that all such systems are not alike and that there is diversity in the ways - 183 in which and the extent to which egalitarianism is manifested and generated in particular societies, I specifically want to focus on what Woodburn (1982) calls systems of competitive equality because these seem to be especially prominent in Melanesia.
Woodburn makes a critical distinction between those egalitarian societies that he calls noncompetitive and competitive. As examples of the former, he discusses several hunter-gatherers (such as the African !Kung and Hadza) and argues that within these groups are found the most egalitarian relationships we know about. He contrasts these with other egalitarian societies, those in which some seem to be more equal than others, those — many of them Melanesian — more aptly described as systems of competitive equality. Woodburn describes the differences between the two in this way:
there [among noncompetitive peoples] equality does not have to be earned or displayed, in fact should not be displayed, but is intrinsically present as an entitlement of all men. There are no casualties of the principle of equality among the Hadza or the !Kung, none of whose moral worth is destroyed by poor economic performance or lack of personal competitiveness. Egalitarianism is asserted as an automatic entitlement which does not have to be validated (1982:446).
As he describes it, although they may occur, systems of noncompetitive equality seem to be rare in Melanesia. His depiction of noncompetitive equality almost necessarily locates it in hunting and gathering societies, and these are not common in the area. Far more typical are horticulturally-based systems of competitive equality, where equality is not simply presumed to exist but must be validated. In a seminal article Forge (1972:533-4) argues that the so-called egalitarianism of most Melanesian societies is of a competitive type in which
… to be equal and stay equal is an extremely onerous task, requiring continual vigilance and effort. Keeping up with the Joneses may be hard work, but keeping up with all other adult males of a community is incomparably harder.
Although it would probably not be wise to draw totally opposing pictures of these two systems, there certainly are differences between them. In one, people assume equity, and cultural constructions undergird social processes which generate it. However, in the other, different assumptions obtain. Although he does not explicitly recognise that systems of equality can be noncompetitive because his concern is to describe only one kind of system, - 184 Bailey (1971:20) identifies a major component of competitive equality when he writes that
equality … is in fact the product of everyone's belief that everyone else is striving to be more than equal. Equality comes about through the mutual cancellation of supposed efforts to be unequal.
A system such as the one Bailey describes involves an inherent competition and must be distinguished from the noncompetitive !Kung and Hadza. In the former, there is a cultural notion of superior/inferior that does not get played out; in the latter, there is no such conception (or at least it is not a significant factor in political motivation). Although competitive equality may result in equity of some sort, it is sometimes if not usually based on cultural notions of inequity and is the result of processes profoundly different from those in noncompetitive egalitarian societies.
Although most Melanesian societies are those in which competitive equality is the norm, there remains much diversity in the means by which, and the extent to which, equality is achieved in social action. It is necessary to make distinctions among these systems; individual political strategies as well as cultural assumptions must be examined in order to tease apart some of the causes of this diversity. Are political actors striving, as in Bailey's example, to be better but must settle for being equal? Or are they, knowing what the world is like, merely striving to remain equal as the best possible alternative, trying to prove that they are as good as any other, as in Forge's keeping up with everyone? Or are they striving to be superior and recognising in the process that a few actually do achieve some higher position, however one defines higher?
These three cultural strategies (and the structural forms which must necessarily be associated with them) are substantially different and to some extent underlie the diversity in Melanesian political systems. This is not to deny that all three may be present in the motivations of individual actors in any particular society; indeed, some people must settle for less than dominance if others are to achieve even a temporary superiority. It is for this reason that it is impossible to delineate anything approaching ideal types here. And because the first two strategies are closely related to each other and exist in different proportions in the motivations of individual actors in every society, it is more difficult to tease them apart than it is the third one. Although systems differ in terms of the centrality of one strategy (and its concomitant structural forms) over another in contributing to the generation of the overall sociopolitical system, only the third — in which some do achieve recognised ascendancy if - 185 only temporarily — is at all easily differentiated from the others.
Societies which exemplify this third strategy, that competition leads some to assume positions of at least temporary superiority, represent the classic Big Man systems of many areas such as Hagen and Enga (Strathern 1971, Meggitt 1967). Here, the systems are egalitarian in that in theory all adult men have the opportunity to compete to become Big Men. They are hierarchical in that they usually include at least some recognised status positions: Big Man, ordinary man, rubbish man. Many include shadings within each category as well: not only can some be more equal than others, but also some can be bigger than others. This is certainly true of the Lak of New Ireland, where some are measurably bigger than others (Albert 1987).
Systems in which the competition is to remain equal, to assert one's equality with all others, are less recognisably hierarchical and seem to be relatively rare, although individuals in various societies must settle for being equal rather than struggling for more. These are systems which tend not to have the classic Big Man and in which leadership is attenuated, and although competition exists, it is only to stay equal. These societies fall closer to the noncompetitive end of any continuum, and hierarchical relations are generally muted among adult men. Of course, larger social structural considerations are as important as individual motivation, and structures exist within the broader context that foster the maintenance of equality. Mitchell (1978, 1988), for example, describes a variety of levelling mechanisms, particularly gambling, that facilitate Wape equality and support their fiercely egalitarian ethos. Cultural notions of equivalence are not enough to dampen individuals’ desire for additional wealth, and social mechanisms such as gambling facilitate “the defeat of hierarchy” (Mitchell 1988). This is not a system of noncompetitive equality: cultural elements recognise the potential for inequality — adult men are not defined and presumed equal by assumption.
In these systems of competitive equality, it is important to consider who competes with whom. As Chowning (1979:84) astutely notes, “there is no glory in defeating a nobody”. It is not honourable to crush an inferior opponent (Bourdieu 1966:197, 199). Care must be taken to challenge and compete with those who one believes are appropriate adversaries. In challenging those who are worthy, one thus acknowledges their potential or temporary equality.
The egalitarian Tangu seem to come close to this kind of system, for among them there is a great deal of competition within notions of equivalence. All exchanges among the Tangu must be equal; returning too much is as undesirable as giving back too little (Burridge 1957a, 1957c, 1959). This is not, however, a system of noncompetitive equality, for competition does play - 186 a significant part in the system, but the result of the competition should be balance and equity:
all claims are subject to the overriding ideal of equivalence, a prime moral value which is also the critical norm governing all social and interpersonal relationships. All exchanges should be equivalent: if an exchange is not equivalent — and because exchanges are in foodstuffs “faulting” is relatively simple — it leads to troubles and disputes in which neither party is considered guiltless, and which can only be resolved by a further series of festing [sic] exchanges in which equivalence is made explicit, or by realigning the members of co-operative groups so that equivalence emerges from the ensuing exchanges. As influential men can only become so by demonstrating superior productive ability in feasting exchanges — where the exchanges must be equivalent — it follows that they must exchange with equally competent men, or equally competent groups managed by men who are as industrious and able as they are themselves. Their influence depends not on dominance, but on maintaining equivalence. … Equivalence implies — even demands — that no dominant group shall emerge, and it is entirely consistent that the structure of a community should be built upon the feasting exchanges between two approximately equivalent groups of households (Burridge 1957b:97-98).
But there are tensions, and some clearly are not satisfied with equality. And it is difficult to execute precisely equivalent exchanges; there remains ample room for dispute, disappointment, anger, and suspicion that the other may just be trying to dominate. Burridge (1971:99) describes the tensions: “but while Tangu watch one another closely, insisting that others should conform to normative expectations, each individual covets that reserve of power and ability which will enable him to be rather more than equivalent to others”. The Tangu overwhelmingly compete to remain equal, but the cultural recognition of dominance and inequity is, unlike systems of noncompetitive equality, ever present in the shadows.
I would expect those systems in which the dominant competitive strategy is to achieve some sort of superiority but in which most competitors settle for equality to fall between the classic Big Man and the more egalitarian Wape or Tangu, although in most cases such societies would be closer to full egalitarianism. The stakes are higher, and at least temporary ascendancy is possible. Often the resulting social forms may resemble those described above, but the sociopolitical dynamics as well as the cultural underpinnings are considerably different. The Mundugumor provide an apt example (Mead 1963, McDowell MS (a)). Adult men compete to remain equal with the hope - 187 of coming out more equal than others, which some inevitably did. Mead (1963) describes the ideal that most men sought as one of having many wives and adherents as being a powerful warrior and spokesman. Men strived for this goal, but few achieved it; most settled for just being equal to others.
There are a variety of other questions relevant to the delineation of these processes: Who is competing with whom? Is the primary competition between individuals or groups? How do individuals’ transactions articulate with group transactions? For what are they competing (honour, status, wealth, knowledge, power)? Does the nature of the prize affect the process (e.g., is competition for intangible honour different from competition for material wealth)? What is the nature of the competition? How are power and knowledge manifested? What basic values are concerned with the process (autonomy, strength, control, power, honour)? Is closure achieved in the process, and, if so, how? How is the tension between the ideology, whether it be egalitarian or allow for potential hierarchy, and the reality of political relations dealt with?
EXCHANGE AND COMPETITIVE EQUALITY: GENERAL CONSIDERATIONS
If people are competing to stay equal or be superior but end up equal, how do they do so? In some areas of the world, personal qualities such as strength, integrity, or honour are central values, and it is by manifesting or achieving more of these qualities that people assert their equality or superiority. 8 There are societies in Melanesia in which the competition is played out in predominantly these ways. And rare is the leader who does not exhibit special qualities such as wisdom, strength, fierceness in battle, or oratorical ability; some possess ritual or special knowledge while others gain advantage because of kinship position. There is a variety of ways in which one can win renown, make one's name known (see Lindstrom 1987). However, in Melanesia it is impossible to ignore the central role that exchange — or competing with things — plays in the processes that generate equity. Exchange need not refer only to the transfer of material items in the context of a competitive or ritual prestation; it may encompass marital exchanges of people or goods, or it may mean the exchange of intangible qualities such as hostility (see Mitchell 1988 for a discussion of so-called negative reciprocity). 9
That exchange is deeply embedded in the processes that foster equity is apparent in several contemporary discussions. One of Mitchell's (1988) main points is that exchange works against hierarchy among the egalitarian Wape: “… exchange is organised to prevent the accumulation of wealth or the concentration of power…” (p.638). With the introduction of colonialism and the potential for capital accumulation, gambling becomes the way in which - 188 potential accumulations of cash are quickly dispersed. Lederman (1987:2), too, stresses the need to examine “indigenous exchange practices” in order to understand egalitarian relations in the Highlands: “they are at once the motivating and organising framework for production and a central medium of political action.” She goes on to discuss different kinds of transactions and their impacts on various relationships; particularly she illustrates how individual network transactions differ from and articulate with bounded-group sponsored exchanges. 10 Although she argues (1986b:208) that we must examine processes in context, particularly historical context, she notes a possible association of an emphasis on network over group ties with more egalitarian relations.
These and other recent treatments of exchange (e.g. Feil 1984, Josephides 1985, Schwimmer 1973, Sillitoe 1979, Weiner 1976) indicate that anthropological recognition of the role of transaction in the establishment of political relationships has progressed since Malinowski's (1961) early analysis of kula exchange. I do not mean to assert that exchange is the only or even necessarily the main factor everywhere (there can be no competition without rewards); we cannot understand equality (or inequality) without examining production, relations of production, and so on. The most complete analysis is one in which both production and exchange are considered, and in examining specific ethnographic cases, the two cannot be easily disentangled. Furthermore, it is the cultural construction of these concepts and their location in historical process that give form and meaning to both exchange and production. Nevertheless, systems of competitive equality in Melanesia give prominence to the role of exchange, and it is the ways in which exchange influences and shapes the sociopolitical processs that I want to explore here.
The idea that the form exchange takes impacts profoundly on relations of hierarchy or equity (and the mix between the two) requires more systematic attention than we have given it. That particular patterns of exchange have significant political consequences is not a new idea. Certainly it is one of Polanyi's (1957, 1968) main contributions to note that one of the differences between reciprocity and redistribution is that the economic centralisation permitted in the latter gives rise to potential political centralisation and the congealing of power in the hands of a few. Equality and balance are far more likely to occur in societies in which the predominant mode of transaction is reciprocity than in redistribution or market societies. And, of course, it is Lévi-Strauss (1969) who delineates the political consequences of marriage transactions. Direct exchange, that is brother-sister exchange marriage, fosters egalitarian relationships among affines and adult men in general, whereas indirect exchange introduces the possibility of hierarchical arrange- - 189 ments between affines and, more importantly, between and among groups related affinally. These general observations from Polanyi and Lévi-Strauss hold true in Melanesia; systems based on reciprocity and direct exchange do tend to be more egalitarian than those based on redistribution and indirect exchange. The form that exchange takes and its cultural underpinnings influence power constellations and political relationships in most if not all societies. 11 But the ways in which equity is accomplished and the extent to which it is achieved differ even among systems based on reciprocity and direct exchange. 12
Anthropologists have contrasted two primary forms of transaction and associated one with equality and the other with inequality. On a simple level, one might ask, as Godelier (1986:172) does: Do exchanges between groups and individuals depend on principles of equivalence and on mechanisms designed to restore rapidly the equilibrium between the exchangers…? Or do they depend rather on the quest for nonequivalence in exchanges, on the principle of calculated disequilibrium or unequal exchange? To continue the quote begun above from Forge's article (1972:534) is revealing of the general point here: the principal mechanism by which equality is maintained is equal exchange of things of the same class or of identical things. Basically all presentations of this type are challenges to prove equality. The basic idea goes back to Bateson's (1958) classic analysis in Naven, where he described two schizmogenetic processes, symmetrical and complementary. Forge (1972:539) draws the conclusion: Bateson's dualities [symmetrical versus complementary schizmogenesis, direct versus diagonal dualism] and their expression in either equality searching symmetrical behaviour or in inequality emphasising complementary behaviour, point directly at the two principles of exchange that form the majority of New Guinea social structures and dominate their workings. Thus, direct, symmetrical behaviours — and by extension exchanges — are “equality searching” while diagonal, complementary ones are “inequality emphasising” ones. Although the delineation of these two types of transaction is too simple for real life, the contrasts are significant and worth exploring further. If the aim of a transaction is to return the same amount that one receives, equality is asserted and probably maintained. This is especially likely to be true for situations of immediate exchange. It is possible (but not necessary) that a delayed return — that is, one removed in time from the initial prestation, even if equal, may introduce an element of hierarchy because until the return is complete, inequity exists. Even after the return is made, the perception may exist that the time lapse indicates a temporary inability to repay and thereby reveals inferiority. Thus, systems stipulating immediate return of precise equivalents will logically (but - 190 again not necessarily) engender more egalitarian relationships.
As Lévi-Strauss (1969) demonstrates, direct exchange allows for more egalitarian relations than indirect exchange. If A gives to B and B is required to return something to A, there is at least a potential for equity, even if B’s return is not necessarily equivalent or symmetrical. But if A gives to B and B has no obligation to return to A, there seems to be an inherent inequality. Whether B perceives that which is given by A to be justly due tribute and, therefore, a marker of his own superiority, or whether B perceives the resulting debt to A to be an indication of own inferiority, does not matter: there exists inequity in either case — its direction will depend on the cultural interpretation attached.
Indigenous categories of transaction are also significant. In societies in which different manifestations of the tension between egalitarianism and hierarchy exist in different domains, such distinctions are likely to be expressed by different transactions. A cultural map of transaction — such as that provided for the Trobriands (Malinowski 1961, Weiner 1976) — goes some distance in helping to understand the nature of the relationships characterised by the transactions. The distinction between sharing and more formal exchanging is important in some groups. It is central in the Sepik village of Bun (McDowell 1978, 1980, 1984a, 1984b), and it is relevant in understanding some aspects of political process in Duna as well (Modjeska 1982). In differentiating between gift and commodity, Gregory (1980) reminds us that not only are indigenous cultural constructions relevant, but also that context is important; whether something is a gift or a commodity depends on social context, and this has significant political implications.
The nature of the items exchanged in material transactions may also be relevant. Systems in which the same items — pigs for pigs, yams for yams, shells for shells — are exchanged for one another have more egalitarian potential than ones in which different goods move in opposition to one another. 13 Of course, in one way an exchange even of pig for pig can never be equal because no two pigs are ever precisely alike; Burridge (1957a), for example, argues that, despite the egalitarian ethic of the Tangu, the fact that there is always room to dispute equivalence in a transaction introduces conflict in egalitarian exchanges.
Perhaps the transfer of disparate items stresses the separate identities of the transactors, or maybe it reflects differential access to significant areas of production. Exchange can, after all, perform two functions that may be relevant here: it can unite in alliance or it can differentiate into other (see McDowell 1980 and 1984a for brief discussions of this; see also Burridge 1975). If the consequence is to unite, it may be that same or similar items be - 191 exchanged; if it is to differentiate, using different items might be more appropriate and effective. And to the extent that hierarchy requires separation, as surely it must, the transfer of disparate items would allow for more hierarchy while, all other things being equal, the exchange of same items could (but again not necessarily does) foster identity and therefore more equity.
In some situations the requirement to exchange the same items inhibits the production of specific forms of wealth by discouraging, disallowing, or preventing conversions from one economic sphere to another. This is certainly the case in some kinds of marital exchanges: if, for example, brother-sister exchange is the process by which people marry, then no matter how many pigs a man may have, it does not help him acquire more wives, something for which he needs more sisters. Unless there are additional ways of acquiring wives, such as taking them from neighbouring groups with whom an ethic of reciprocity does not obtain, as among the Mundugumor (Mead 1963, McDowell MS (a)), a man with only one sister is at a serious disadvantage. 14 Conversions from one economic sphere to another allow more fluidity in the manipulation of power through exchanges, and whatever discourages or prevents such conversions must necessarily inhibit potential hierarchy.
Even in systems based on direct exchange and balanced reciprocity, inequalities of various sorts can creep in. If one is required to exchange precise equivalents, one can still rise above the norm by participating in more exchanges of equivalents than others do. That is, one need not increase the amount given to any particular partner—one need only increase the number of partners with whom one exchanges. An ordinary Trobriander has a few kula partners, a chief many (Malinowski 1961). Alternatively, one could exchange equivalent amounts only with people capable of producing and exchanging large quantities, thus refusing to transact with unworthy adversaries. In some cases, cultural norms exist that dampen the potential generation of equality through reciprocity: for example, among the Lak of southern New Ireland, Albert (1987) describes a particular feast in which the recipient is not allowed to reciprocate equally. Josephides (1985:107-9) also ponders how reciprocity can create hierarchical relations. 15 Thus, even in a system which seemingly requires equivalence, there are means of manipulating transactions to achieve some semblance of hierarchy.
EXCHANGE AND COMPETITIVE EQUALITY: WOODBURN, GODELIER, MODJESKA
These tendencies, primarily based on common sense, are meant to be - 192 suggestions for thinking and research only. They are too simple for the living world of ethnographic reality and the complexity of functioning sociocultural systems. But a cursory perusal of the literature suggests that there is something in them worth further elaboration. In the rest of this essay, I reiterate portions of the arguments made by Woodburn (1982), Godelier (1982, 1986) and Modjeska (1982) to illustrate how these considerations might be explored. These examples indicate how, despite the complexities, we might begin to examine systems of competitive equality. 16
Woodburn's distinction between noncompetitive and competitive equality provides a beginning. He argues that egalitarian hunting and gathering societies (to which his discussion is primarily restricted) are of two types: immediate-return systems and delayed-return systems. It is important to note that he uses these phrases, immediate-return and delayed-return, in a particular way; he is referring to return on labour invested, not on prestations made. He asserts that all immediate-return systems are profoundly egalitarian whereas a great deal of variability is evident among delayed-return systems and that a delayed-return system is never as egalitarian as an immediate-return one. He describes (1982:432) the nature of immediate-return systems in this way: people obtain a direct and immediate return from their labour. They go out hunting or gathering and eat the food obtained the same day or casually over the days that follow. Food is neither elaborately processed nor stored. They use relatively simple, portable, utilitarian, easily acquired, replaceable tools and weapons made with real skill but not involving a great deal of labour.
He contrasts delayed-return systems: people hold rights over valued assets of some sort, which either represent a yield, a return for labour applied over time or, if not, are held and managed in a way which resembles and has similar social implications to delayed yields on labour (p.432). He notes that all systems of farming are probably delayed-return ones (and thus recognises that these are predominant in the world). What is important is that there are implications for social organisation: [delayed return systems] depend for their effective operation on a set of ordered, differentiated, jurally-defined relationships through which crucial goods and services are transmitted. They imply binding commitments and dependencies between people (p.433). In the discussion of the characteristics of immediate-return systems, what is striking is the depiction of the circulation of items among individuals. He discusses, in particular, the Hadza, among whom circulation of material items among adult men occurs primarily in the context of gambling. 17This circulation is accomplished, then, not through some form of exchange which would bind participants to one another in potentially unequal relationships of kinship or - 193 contract. The transactions are neutralised and depersonalised by being passed through the game. Even close kin and affines gamble with each other and the game acts against any development of one-way flows and dependency in relationships between them (p.443).
Noncompetitive equality and immediate-return systems seem to entail a disengagement both from property and from the requirement to participate in transactions of property with specified others (Woodburn 1982:445). Competitive equality and delayed-return systems, then, entail an obligation to enter into designated relationships with specific requirements to transact. Josephides (1985:152) argues that among the Kewa one source of inequality relates to the fact that “…the labour of certain categories of men [ordinary as opposed to Big Men], as well as the labour of women, is separated from its product through the mechanism/strategy of exchange”.
Woodburn's major discussion concerns immediate-return systems and noncompetitive equality, and he recognises that, because of the vast diversity within the category of delayed-return systems, his treatment of this second kind of system is cursory. He does use particular Melanesian examples as exemplars and cites Forge (1972) and Burridge (1975) on systems in which people strive mightily to remain equal. I think it no accident that he uses lowland examples to illustrate relatively egalitarian competitive systems, those in which the goal is to keep up with others rather than achieve, if only fleetingly, an ascendancy; the classic Highlands Big Man is further towards the nonegalitarian end of the continuum. The main point here, however, is Woodburn's suggestion that required transactions between specific individuals (as opposed to a general ethic of sharing and generalised reciprocity) are essential in the generation of systems of competitive equality. Modjeska (1982) has a similar idea (discussed in more detail below).
Godelier's (1982, 1986) focus is on somewhat less egalitarian societies. He distinguishes carefully between two forms of leadership (and corresponding social orders) found in the Papua New Guinea Highlands: the Big Man versus what he refers to as the “great man.” He uses as his picture of the Big Man Sahlins’ classic description of a man for whom “…it is exchange and competition in wealth between tribes which become the basis and privileged channel of establishing his renown” (Godelier 1982:31). The Baruya great men, in contrast, “…perform, to an exceptional degree, specific roles, among which warriorhood and shamanism are the most important” (1982:3). Great men are an interesting mix in that, for some, the status is based on a position of hereditary ritual leadership, whereas for others, it is based on competition among individuals in the arenas of warfare, shamanism, and hunting. Although there are definite domains in which inequality prevails (male-female, - 194 senior-junior), relations among adult men seem to be of the egalitarian competitive equality kind. Godelier concludes by contrasting the social systems which seem to go along with the Big Man and the Great Man. He attributes the differences to the existing kinship structures, and a close reading of his conclusion (pp.31-2) illuminates the significance of Woodburn's dichotomy and as well as exchange in this context:
The Baruya practise restricted exchange; most big-men societies practise generalized exchange of women for wealth. Among the Baruya, fundamentally, a woman is equivalent to another woman and only the gift of a woman can annul or, better, counterbalance one gift against another. It seems to me that the principle of direct exchange of women simultaneously entails a whole series of social effects which determine and comprise a sort of global social logic…. The fact that a woman is worth a woman has the direct consequence that one has no need to accumulate wealth in order to reproduce life and kinship relations. There is thus no internal articulation or direct connection between material production and the reproduction of kinship relations. In this logic, the production of pigs, like subsistence activities, remains of relatively low value…. By contrast, the big-man appears to emerge in a situation where restricted exchange of women is not practised as the dominant principle, where the generalised exchange of a woman for wealth is carried out and where the accumulation of wealth becomes a direct condition for the reproduction of kinship relations…. The production of pigs and the intensification of agriculture…prevail over the hunt and upon the role of game in exchange between the sexes, generations and kinship groups. Finally, in relations between tribes, in war or in trade, complex systems of ceremonial exchange are added. 18
There are several suggestions here that are relevant. The first is that the more hierarchical (less egalitarian) Big Man system tends to be the result of generalised exchange and bride-wealth, whereas less hierarchical systems are a consequence of direct exchange. From this basis and these premises, logic generates other associations. In restricted exchange societies practising direct exchange, there is no need to accumulate wealth to reproduce kinship relations, a relatively low value attached to the production of pigs, and few large intergroup ceremonial events and exchanges. The contrast with Big Man societies is sharp, for among these the accumulation of wealth (in the form of bride-wealth at least) is essential in reproducing kinship relations, the production of pigs takes on an especially high value, agriculture is intensified, and there are large intertribal ceremonial events and exchanges. Outlined here are two significantly different systems of competitive equality and a brief - 195 sketch of the differential role exchange plays in each. Godelier seems to attribute some kind of determination to the kinship system, particularly the marriage system, but even if we are to do so, it is still basically a question of exchange and its cultural construction — woman for woman or woman for wealth — that comprises the primary distinction.
Godelier is not the only contemporary analyst to make suggestions about the role of marriage patterns in the generation of egalitarian political relations in Melanesia. Gregory (1980:641-2) contrasts gift-debt with commodity-debt and suggests, for example, that
… different types of gift-debt must in turn be explained in terms of the different types of kinship systems. For example, marriage systems based on the exchange of ‘sisters’ cannot support a big-man system with incremental gift exchange. A marriage system involving bridewealth is a necessary, but not sufficient, condition for incremental gift exchange, that is incremental gift exchange implies bridewealth but the opposite is not necessarily true (see also Gregory 1982).
Here again we see the association of more hierarchical systems (the Big Man system) with indirect exchange and bride-wealth in contrast with more egalitarian systems with direct exchange and brother-sister exchange marriage. He adds the idea that incremental gift exchange is involved. If one is to have an elaborate exchange system, such as the ceremonial ones of the Highlands like the moka and tee, there must be the notion of incremental gift exchange, a notion that presupposes the existence of bride-wealth.
Modjeska (1982) has also expressed several of these and related ideas. Although his overall model is too complex to discuss here, and although his primary focus is on production rather than exchange, he makes a variety of related points. He contrasts what he calls high-production societies with low-production ones and expands the distinction by elaborating on the Duna of the Highlands, a society which falls in the mid-range of his hypothetical production continuum. There is considerable overlap between Modjeska's high-production society and Godelier's Big Man society, but the fit between the low production and Great Man societies, although substantial, is not as complete. In general, according to Modjeska (1982:52), high-production societies are characterised by a large linguistic group, high population density, intensified horticulture, high pig production, and elaborate ceremonial exchange cycles that involve incremental increases in items given. These are also societies in which inegalitarian relations between adult men are more typical than in low-production societies, which are of smaller - 196 size and lesser population density, have less horticultural intensification and are more likely to rely on extensive swidden gardens, have low pig production and attentuated if any ceremonial exchange systems that lack the notion of increment. Although Modjeska couches it in an evolutionary context, a general contrast also obtains in leadership:
leadership, and the inequality implied by it, among Enga and Hageners [high production societies] is based upon pre-eminence in the ceremonial exchange of tee or moka. Among the Ok and Papuan Plateau peoples [low production societies] leadership is based upon participation in hierarchies of ritual initiation, or upon shamanistic spirit-mediumship …. An obvious point emerges: lacking pigs, these latter peoples have not developed forms of domination and influence based upon the production and circulation of material wealth (1982:86).
He describes the midway Duna in this way:
leadership was…based upon pre-eminence in several practices, none of which constitute an unequivocal cultural focus. Positioned at a mid-point between low- and high-production systems, Duna society seems also balanced between ritual and political-economic modes of domination (p.86). High individual productivity is necessary for domination in high production societies and less so in low production ones (pp.101-2).
The importance of pig production in all its varieties is clear in the Highlands. Modjeska (1982:51-2, 55-6, 62) stresses the pig's exchangeability as well as its reproducibility and argues (pp.57-8) that it is pig exchange that creates social and political power. Shell exchange perhaps allows men to control a sphere all their own, separate from the productivity of women (who are prominent in the production of pigs) (p.95), and perhaps it also facilitates an increasing bigness of Big Men (pp.67-8), but pigs remain the critical factor (not in any simple sense and not universally throughout the production continuum). Citing A. Strathern's (1969) distinction between production and finance (two strategies of participating in exchange systems; see also A. Strathern 1978), Modjeska argues that
inequality among Duna men cannot then be reduced to differential pig production, although production is probably more a determinant of inequality than would be the case among Hageners and Enga where finance predominates. The - 197 absence of strategic tokens of exchange (pearl shells) and the absence of a sphere of strategic manipulation (reciprocal incremental exchange, enchained exchange) condition the pattern of domination in the direction of equality. But although pre-eminent Duna men are not big-men transactors, there are none the less inequalities beyond differences in production alone (1982:101).
What does determine the generation of inequalities among the Duna? The pig producer is not the most powerful man since he lacks the temperament and rhetorical skill of the “man with talk”. It is the latter who appropriates pigs by his persuasiveness. And it is he as well who precipitates wars, which end by focusing attention and prestige not upon the producer but upon the public distributor, the “man with talk” in his capacity as … fight leader. Production remains a necessary condition of domination (in a way that it is not in neighbouring low-production societies), but the capacity to produce pigs is more general (and hence conducive to equality) than the capacity to “produce” domination through persuasive speech (1982:101-2).
The relative importance of generalised reciprocity in societies such as the Duna has an additional effect on the generation of hierarchy among adult men in that it dampens status differences on the lower end of the scale as well as the higher end. “Ideals of generosity rather than competition are emphasised in exchange, and unambitious, retiring men are granted a ready tolerance” (p.90).
Modjeska argues that incremental ceremonial exchange is a primary factor in the generation of bigger Big Men: “Enga and Hagen big-men are a product … of the social environment in which they are found, a social environment profoundly structured by the totalising activity of massive ceremonial exchange” (p.166). Various forms of exchange process play a critical role in Modjeska's argument about why incremental ceremonial exchange generates relatively greater inequality, and he echoes some of the ideas of Woodburn (1982) and Josephides (1985) concerning the separation of producer from product:
my argument is that the diffuse balanced reciprocity among kin and affines in Duna “helping” relationships, and the separation of this form of circulation from non-reciprocal damba [compensation] and marriage payments to non-kin, precludes the development of a transactional sphere in which financed pigs become differentiated from domestically produced pigs. This can be seen most clearly in enchained exchanges.… The separation of pigs from their specific contexts in domestic production permits big-men to manipulate alienated - 198 values independently of the claims of producers.… The existence of exchange patterns differentiating financed pigs from home-produced pigs constructs the sphere of transaction that allows Enga and Hagen big-men to function as true transactors (pp.95-6).
Lévi-Strauss would remind us that kinship and marriage are especially relevant, of course, as the expectation for generalised reciprocity in certain societies might hint. 19 Modjeska does not detail consequences of sister exchange as opposed to bride-wealth, but he does address the issue in his general discussion of social organisation. Citing Craig (1969), he notes that
… bride-exchange ensures the status quo of groups while bridewealth serves the different function of defining rights in women and children. It would seem an indication of the demographic success (as well as the success in pig production) of the large groups east of the Strickland river that men are there able to forgo the direct exchange of women (p.67).
Modjeska also suggests that “relations of siblingship are perhaps always more compatible with egalitarian assumptions than are relations by descent” (p.65). Descent is only a weak organising principle in low-production societies, but it becomes a stronger force (and more firmly patrilineal) as productive capacities increase (p.66). It is no accident that many of the examples cited here to illustrate the most egalitarian of societies with competitive equality — e.g., Bun (McDowell 1978, 1980, 1984b), Tangu (Burridge 1957b, 1959), Mundugumor (Mead 1963, McDowell MS (a))— do indeed stress the significance of the brother-sister relationship and considerably downplay the importance of descent.
There were, in Duna society, a variety of avenues to achieving some influence and prestige. The truly great “men with names” displayed some prowess in all categories: pig production, oratory, exchange, wealth, warfare. Pig production and, one assumes, exchange were necessary but not sufficient: oratorical skill was required as well. This consideration leads to a recognition that there are various domains of power within a single society and that different societies possess or conceptualise different constellations of power. Modjeska cogently argues that
these multiple aspects of pre-eminence as recognised in Duna society, an ethno-catalogue of dimensions of inequality among men, lend themselves to a conceptualization of power such as proposed by Foucault: power as a multifaceted, shifting, strategic resource, with dominance and inequality developing - 199 according to the points of power available for the manipulation of on-going social arrangements— here ritual mystification, there commodity production, at another point naked coercive force (1982:92).
If equality and inequality have anything to do with power, as surely they must, we need to comprehend the ways conceived by a people of grasping that power. It is in this context, especially, that cultural concerns must enter the analysis. “Practical reason”, as Sahlins (1976) argues, is not sufficient.
Egalitarian relationships are not necessarily humanity's natural condition but must be culturally and socioeconomically produced and maintained just as hierarchical ones must. The aim of this essay is simply to begin to explore the nature of competitive equality, that is, systems in which individual actors must strive to maintain their equality with others. This very preliminary exploration reveals that, although these systems resemble one another in important ways, they also exhibit significant diversity.
It is necessary to examine in some detail the nature of these competitive systems: for what are people competing, why do they compete for it, and what are the rules by which they do so? One must necessarily look first to the conceptual frame within which action takes place. What are the appropriate cultural goals and rules? What invests meaning into an actor's actions? The means of achieving power and personal autonomy vary with the way power is conceptualised and construed to be available. To what extent does transaction and its cultural construction — material, marital, immaterial — affect the process? Exchange plays a far more central role in some situations and contexts than in others: where knowledge or ritual dominance are predominant, not only are processes of material exchange somewhat less important but they are also of a substantially different nature. All of these factors need to be examined in order to understand the nature of Melanesian egalitarianism.
- 200 Page of endnotes- 201
1 Portions of this paper were read at the annual meeting of the Association for Social Anthropology in Oceania in Monterey, California, in 1987. I should like to thank those who attended or participated in the session on hierarchy and equality for their helpful comments, particularly Steven Albert, Fred Errington, James Flanagan, Deborah Gewertz, Rena Lederman, Margaret Rodman, and Ted Schwartz.
2 For classic discussion of Melanesian political systems, see Fortune 1947, Malinowski 1926, Oliver 1955, Pospisil 1958, Read 1950, and Reay 1959.
3 I also agree with Flanagan (personal communication) that we ought to be talking here about egalitarian relationships, not egalitarian societies; we barely know what the former may mean and certainly are unclear about the latter. However, various authors cited here do use the phrase “egalitarian society” or “egalitarian system”, and I follow their lead out of convenience and as a shorthand device. Just what the problematic label “egalitarian” may mean is an underlying theme of this essay.
4 Indeed, some of this analytic work is already being done. Rodman (personal communication) notes that hierarchy implies ranking and that hierarchy is not necessary for inequality. And Brown (1987:88) asserts that … [Fried's] category of rank societies “… in which hierarchy without political authority or highly developed restrictions on access to resources is evident fits our New Guinea highlands cases better than the egalitarian or stratified types,”. This kind of conceptual tool sharpening is long overdue.
5 Although the example is a non-Melanesian one, Bern (1987) argues that, in general, egalitarian relations within lower classes are dependent upon embeddedness in a wider, and presumably dominant, society. And Howes (1984) Provides an excellent example from Eastern Indonesia of the way in which brothers are conceived to be equal while affines are by definition in an asymmetrical relationship.
6 Modjeska's (1982:92) discussion of power is relevant here; see the discussion later in this paper for a consideration of it.
7 If sufficient data were available about women's realms, it would be possible to explore the extent to which some of these issues obtain among women as well.
8 It would be interesting to make a detailed comparison of Melanesian competitive equality with that found in the Mediterranean area, but it is beyond the scope of this essay. The game is similar, but it is played out with different prizes and within a different semantic field, particularly that of honour and shame. For a consideration of shame in Melanesia, see Epstein 1984. For an especially relevant discussion of honour, see Bourdieu 1966; see also Bailey 1971.
9 The role of sorcery and relations with ancestors and/or spiritual beings should be considered as well. Modjeska (1982), for example, notes that sacrifice is a kind of exchange with the spirits. The place of knowledge can be especially relevant too; see Lindstrom 1984.
10 See Sillitoe (1979), who argues that we must focus far more on individuals and less on groups if we are to understand exchange in Melanesian societies.
11 Exchange can occupy different functional niches, each of which may impact on considerations of equality and hierarchy. Exchange acts clearly communicate a variety of kinds of information (see Bourdieu 1966, Ernst 1978, Weiner 1976). But exchanges go beyond mere communication: they create. Ernst's 1978 article includes a good discussion of how the exchange of material items can both communicate and create.
12 Burridge (1960, 1969, 1975 and 1979) has explored the relationship between exchange and morality in Melanesia and suggests that “… the Melanesian concern for translating a mystical, emotional, or intellectual experience into a moral relationship expressible in an exchange of material stuffs comes uppermost” (1975:104).
13 Impersonal trade relationships are excluded here.
14 This is, of course, an oversimplification. See the discussion below for some indication of the real-world complexities involved.
15 The direction can go the other way. Lederman (1986b:151ff) discusses the way in which unequal exchanges do not necessarily result in inequality.
16 Portions of this section are included in McDowell MS(b).
17 It is interesting to note that gambling is a significant levelling mechanism among the Wape as well (Mitchell 1978, 1988).
18 Godelier incorporates various aspects of male domination over women into this model, but for simplicity I have omitted them here. See his full discussion.
19 Lindenbaum's (1984) discussion of ritualised homosexuality touches on many of the issues raised here, and she begins with kinship and marriage patterns as well. She argues that “sister exchange and the absence of bride-price emerge as crucial features of those cultural systems with ritualised homosexuality, which attempt to sustain egalitarian relations among affinal groups” (p.339). She (pp.349-50) sets up an evolutionary transition in which the relative equity of sister exchange gives way to relations dominance and bride-price, ceremonial exchange, and Big Men. Throughout her analysis, the interpenetration of various domains is critical.