Introduction
The Problem of Persistence—Precolonial Forms in Postcolonial Life

Leaving Lorengau, the capital of Manus Province, you have to travel west for more than two hours on a motorized outrigger before the tallest of Ponam's trees begin to appear above the horizon. By then you have passed Hauai, Pitiluh, and Ahus Islands on the right as you move through the Seeadler Harbour, the deep protected channel that separates the mountainous Manus mainland from the string of flat fringing islands off the north coast. After three hours you have cleared Andra Island, the last before Ponam, and you can see the bay and buildings of the Bundralis Catholic Mission on the mainland opposite Ponam.

In the next hour, spent almost wholly in Ponam's long lagoon, your canoe turns seaward until it crosses to the north of the eastern point of Ponam and then heads west again, traveling more slowly, within a few meters of the northern shore. Though Ponam is only a few hundred meters wide, it is three-and-a-half kilometers long, with the village near the western end. You move past the casuarina trees, scrub bush, and occasional coconut stands that are Koliwar, the place of war, where the United States Navy had an airstrip during World War II. Then one small house appears, built on a concrete foundation left over from the war, then another, and you have reached the edge of the village. Past the church, two new houses, and then Ponam begins in earnest: a succession of hamlets, each a cluster of dwellings centered on a kamal or men's house. Some of the dwellings are built on the ground, some one or two meters up on wooden posts; some are a single story and some are two; some are built of sago palm thatch laid on a frame of roughly shaped tree branches, some are plywood or corrugated iron sheets laid on a frame of commercial timber. And everywhere along the beach are canoes—lying at anchor, pulled above the high water mark, or, the big ones of ten or fifteen meters, resting on posts and rollers above the beach.

By now the villagers have heard and seen your canoe, and as you come to land you see a dozen or more coming to the shoreline to help unload the suitcases, string bags, and mats, the kerosene containers, bales of rice, and cases of tinned fish that lay on the canoe's wooden platform. After the cargo is ashore and the canoe beached, you move off to your host's house, accompanied by children precariously balancing bags or cases on their shoulders. As you walk down the wide main streets of the village, string band and rock music from tape players compete with the disembodied voices of people in their houses calling out greetings.

At your host's house there is a meal ready and, since the sun is going down, a pressure lamp is ready for lighting. You get fresh fish, greens boiled in coconut cream, and fried sago with grated coconut—called "Manus-style" in other parts of the country—and more than anyone not brought up on the diet could possibly eat in one sitting. During the meal there is quiet chat about the flight from Port Moresby to Manus, who you saw in Lorengau, and the canoe journey, broken by sips of sweet tea or fish soup. You in turn ask about the latest births, deaths, and marriages, who has paid and who has received brideprice, whose canoe is finally built and ready for launching, and whose is beyond repair and ready to be stripped for usable planks.

The rest of the evening you spend sitting and chatting with Ponams who come to say hello and receive the messages and small packages that island migrants living in Port Moresby have given you to carry home. They sit for a few minutes, chat, chew betel nut, and then move off into the night, calling out "Good night" as they disappear. Because it has been a long day and canoe rides are so tiring, the pressure lamp is turned off at 9:30.

The next day is Friday, and more importantly sona pene , the day before market. During the day and evening most of the islanders are out fishing, trying to catch the extra that they will need when they go to market Saturday morning. The men and women who have spent the morning or afternoon with their small canoes angling and spear gun fishing in the shallow lagoon bring back their fish, clean and gut them, and set them on wire mesh over smoldering fires to smoke them. That evening, as you sit talking with your host, shapes loom momentarily at the edge of the lamp's light, and you hear the padding of feet moving quickly across the hard sand and the murmur of people speaking quickly and quietly to each

other. The cause of the activity becomes apparent when two young men, half walking and half trotting, go by carrying a wood-framed net. Tonight there is a lawin , a netting expedition in the lagoon, one more try to get fish for tomorrow's market.

Saturday is pene , market, and in the morning the seven or eight large canoes that are going to market are made ready. They are taken down to the water and their masts fitted, except for the one that is powered by outboard motor—quicker and easier than sail, but not for free: fifty toea for the journey to market and back. Once the passengers and their wares are aboard, sails are raised, motor is started, and they set off, some to the market at Tulu to the west of Ponam on the Manus mainland and some to the market at Bundralis.

The main thing villagers want at market is their starch for the next week: mostly bundles of sago flour wrapped in sago leaf, but also taro, sweet potato, and tapioca root. They also buy leaf and fern greens, pumpkin, pineapple, grapefruit, and sugar cane, as well as the rarer fruits and vegetables that their market partners bring from time to time. And of course they want betel nut, leaf, and powdered lime, the ubiquitous accompaniment to much of what they do. Islanders get all of this in trade for their fish or cash, either the money they bring with them or the money that mainlanders have paid them for their seafood.

Market over, islanders sail back home, the canoes mostly lower in the water than when they set off in the morning. They are laden not only with mainland produce, but also with purchases from the trade stores at Tulu and Bundralis, both of which carry more goods at lower prices than do the stores on Ponam, as well as goods that individual Ponams have received from their own kawas or trade partners on the mainland: special vines and leaves for making mats, baskets, and arm bands, and sometimes some sago leaf for re-thatching part of a house. Those who have gone to Bundralis bring back mail for the island, held for them at the Catholic Mission there; they may also bring back one or two islanders who have come home on leave and who have decided to save the K60 it costs to charter a canoe from Lorengau, traveling instead for a few kina on the mission boat Martin and sleeping in one of the two or three houses that Ponams have built on the beach near Bundralis, where they wait for Saturday and a lift to Ponam.

The rest of Saturday and Sunday is spent relaxing. People who

got vines and leaves from their mainland trade partners lay them out to dry in the hot afternoon sun. Saturday afternoon and evening there are desultory discussions of whatever news the mail has brought or the mainlanders have passed on, as well as the usual complaints that there was too little sago at market, that the price was too high, and that, once again, mainlanders would not buy Ponam fish.

Sunday is a day of rest, and work of any sort is forbidden. Most islanders go to church in the morning, overseen by the catechist and usually enlivened by an emotional, but always heavily allusive, speech by Ponam's lay church leader, urging that certain unnamed people resolve their unspecified quarrels. Some Sundays nobody is quite sure who the church leader was talking about. In the afternoon some families go for a picnic in Koliwar or find a shady, breezy spot to chat and doze. Men and sometimes women start up games of "five hundred," a variant of gin rummy, and usually some men will start a penny-stakes game of "lucky" in one of the men's houses. Sunday night is spent as Saturday, sitting on the beach, outside a dwelling or in a men's house, chatting, smoking, and chewing betel nut.

The next morning is Monday, and another week begins.

In late 1978, when we first arrived on Ponam, our impression was that people there were closely involved with the outside world and also involved in a complex round of traditional activities. Islanders had a large number of material possessions: pots and pans, sheets, pillows and pillowslips, watches, radios and tape players, in fact a surprising range of paraphernalia intended to make life easier and more comfortable. To complement this they were knowledgeable about the larger world. In due course people asked us about the success of Margaret Thatcher in becoming prime minister, Soviet activities in Afghanistan, and other newsworthy events. Counterpoised to all this evidence of modernity was flourishing tradition. Ponams maintained markets with mainland agriculturalists, as, they assured us, they had done time out of mind, and they traded with trade partners from other areas of Manus as their parents and grandparents had done. They continued to use fishing techniques that patently harked back to their precolonial past and even continued the tedious and time-consuming manufacture of

the shell money that they still used to make brideprice payments. And in everything they did islanders referred back to the ramified network of kinship relations that bound them to each other, shaped who did what, who they did it with, and how they did it, a network that they celebrated and recreated in the scores of ceremonial exchanges, gifts, and payments that filled their year.

Thirteen months of fieldwork and several later visits ranging from two weeks to two months have not led us to change the initial judgment. They have, however, led us to see both the ways that the prosperity and the tradition are problematic and the ways that they are linked together. This curious juxtaposition of prosperous modernity and flourishing tradition calls for comment, for it shapes much of the argument we put forward in this book.

While islanders did many things that were straight out of introductory anthropology courses, these were not embedded in the sort of society those same courses so often portray, subsistence societies relatively little touched by the outside world. Instead, on Ponam the old ways stood together with the new: complex kinship and exchange coexisted with wage employment and commodity relations; elaborate fishing techniques coexisted with tinned fish; traditional markets and trade partnerships coexisted with trade stores, bank loans, and cash transactions between market partners.

This juxtaposition of old and new posed a difficult question: How were we to think about what we knew of Ponam and its economy and society? How were we to come to some sense of which bits were important and of the ways they fit together? The aspects of the old that we saw in fact were, quite often, traditional, in the sense that islanders saw them as being lineally descended from and very much like ancestral practices. However, these aspects existed in a society closely enmeshed in Papua New Guinea's spreading national economy. How were we to make sense of this?

Had we been doing fieldwork in the 1950s or even the early 1960s the answer would have been fairly simple, for common wisdom then suggested that the old ways were fragments of a precontact order that was being destroyed by colonization and the collision with Western capitalism.^[1] In fact, the power of Western society was so great that even indirect exposure to it could transform the old ways, or at least that was the lesson W. E. H. Stanner drew from Richard Salisbury's study of the effects of steel axes on

precolonial Siane society in what became Simbu Province: "Steel drove out stone; the new cutting edges saved men's working-time; and from those beginnings, the whole design-plan and dynamism of their way of life changed" (in Salisbury 1962, vi). A few years later, C. D. Rowley, a distinguished and longtime observer of colonial Papua New Guinea, came to similar conclusions. From contact, Melanesia became "committed to a permanent process of change . . . from this point, the old order of society is doomed" (Rowley 1965, 94).

The reasons for this change were self-evident, for no one would expect neolithic societies to survive exposure to and incorporation within the collection of British, German, and Australian colonial apparatuses that governed Papua New Guinea. Cosmological upheavals appeared as cargo cults, the vehicles of religious and political reorientation (Lawrence 1964; Worsley 1957). Old trading systems collapsed, pulling down complex sets of economic dependencies and social relationships (Harding 1967; Mead 1968). Tribesmen were turned into peasants and bigmen into entrepreneurs (B. Finney 1973; Meggitt 1971).

This faith in the power of colonization was not limited to Melanesianists, nor was it original with them. Over a century earlier Marx and Engels wrote that "the bourgeoisie . . . compels all nations, on pain of extinction, to adopt the bourgeois mode of production; it compels them . . . to become bourgeois themselves. In one word, it creates a world after its own image" (Marx & Engels 1976, 488). This view, of course, did not merely reflect their assessment of the potency of the productive forces that capitalism unleashed. Rather, it sprang as well from their belief that the falling rate of profit and the periodic crises of overproduction were inherent in capitalism: the one forcing capitalists to search for new and cheaper sources of raw materials and the other forcing them to search for new markets, and both, in other words, forcing capitalism to extend its grasp to ever greater areas of the world.

As the 1970s began to pass, however, it became clear that the penetration of capitalist relations into village life in Papua New Guinea, not to mention the rest of the Third World, was problematic rather than automatic. Tradition, custom, seemed more tenacious than many people expected. So, for example, Harold Brookfield (1973, 127) recanted his earlier belief that Chimbu society

would be transformed by colonization, concluding instead that what was taking place was "simply . . . the partial acceptance of 'modern' innovations into a continuing system whose 'central variables' have not been transformed" (see also Howlett 1973). Likewise, Mervyn Meggitt, speaking of the Mae Engan system of ceremonial exchange, the te, but probably reflecting a more general sentiment about traditional practices, said, "I for one would not assume its demise unless I actually attended the funeral" (Meggitt 1974, 182).

This mood of cautious optimism about the durability of traditional forms of Melanesian life remains an important stream in anthropology. Michael Young, introducing a special issue of The Journal of Pacific History , said of the societies of the Massim area of Milne Bay Province where the kula system of exchange is so important, that "a century of colonial intervention has left them substantially intact. They have managed to absorb new economic relations in ways compatible with traditional values" (Young 1983, 10). And in that same issue, Miriam Kahn (1983) set about explaining how the Wamira people of Milne Bay maintained central elements of their culture in the face of colonization, adapting and adopting colonial introductions that fit traditional Wamira culture and sabotaging the rest.

This concern with persistence has gone hand-in-hand with a tendency to treat village life in isolation from the outside world. This appears strikingly in the long-awaited volume of papers from the first kula conference (Leach & Leach 1983). While many of the papers acknowledge the importance of colonization, the fact that, as C. A. Gregory puts it in his contribution (1983, 103), colonization has served "to integrate the [Milne Bay] province's 108,000 people well and truly into the world economy," generally the importance of colonization is ignored once it is acknowledged, as the authors make no real effort to link the kula, an intensely economic and political system, with the economy and politics of Milne Bay Province. Illustrative of this is the fact that the book's index contains no entries for: government, labor, migration, Milne Bay Province, plantations, provincial government, remittance, or wage.

Important parts of Melanesian ethnography, then, warrant the more general criticism Joel Kahn has made about anthropological studies of peasant societies in general:

Pick up almost any monograph and you will find an introductory statement about the outside society, the total social whole in which the village is embedded, the importance of extra-local ties, etc. And yet I would argue that in most of such cases this holistic view is set aside, and the total social whole ignored in favour of detailed analyses of purely local phenomena. (1980, 2)

Clearly this isolationist approach does not characterize all of Melanesian ethnography; there are anthropologists who have been concerned with the way capitalist colonization has affected village life (e.g., Feil 1982; A. Strathern 1979, 1982). Nonetheless, this is an important aspect of the literature in the region.

Fortunately, Ponam was so closely involved with the outside world that it is impossible for us to restrict ourselves to "detailed analyses of purely local phenomena." The relationship between Ponam and the ever-expanding outside world with which it was related is a central concern of this book.

This relationship is so apparent to us for two reasons. One is our concern to make use of the historical sources that are available for the Admiralty Islands, partial though they are (as we explain in chap. 2). In Papua New Guinea the overwhelming historical fact over the past century has been colonization; therefore, to make sense of Ponam's history, we have to look at the ways this external force impinged upon the island. The second reason is that islanders never approached the sort of self-sufficiency that agriculturalists could achieve. They always traded, for they never grew their own food. This put them among Melanesian trading peoples like the Titan speakers of southeast Manus (e.g., Mead 1930), who were totally landless, and among those from small, infertile islands like Mandok in the Vitiaz Straits between the New Guinea mainland and the island of New Britain (e.g., Harding 1967). Trade being so much a part of Ponam life, it would be impossible to understand the society solely with reference to its internal operations. Instead, we have looked to the regional trade on which Ponam survived and to the ways that colonization and its aftermath disrupted that trade and so disrupted Ponam. In fact, in many ways this book is little more than an extended exercise in describing how islanders coped with this radical alteration of the external world on which they relied for their survival.

Village and Town: The Articulation Model

The view that societies of the Third World were more persistent than the earlier steamroller model of colonization allowed found a timely echo in developments taking place at about the same time among Marxists, particularly developments in their interpretation of the way the encroaching colonial capitalist mode of production articulates with the preexisting traditional modes (see Wolpe 1980). Probably the most influential writer in this area is Claude Meillassoux, especially in his Maidens, Meal and Money (1981). He argued that penetration need not lead to the eradication of precapitalist forms, but rather often results in their encapsulation and perpetuation, not because capitalism is weak, but because the survival of a precapitalist sector is beneficial to capitalism. This is because much of the wage labor in Third World countries is provided by circular migrants, and many of the costs of the reproduction of this labor force are borne by the precapitalist sector, thereby reducing costs to the capitalist sector. Because the circular migrant was reared in the village and will return to the village, his wages (both his direct wage and what Meillassoux calls his "social wage") need cover only the "sustenance of the workers during periods of employment" and perhaps "maintenance during periods of unemployment (due to stoppages, ill-health, etc.)" but not his "replacement by the breeding of offspring" (Meillassoux 1981, 100, emphasis omitted). Indeed, much of the power of capitalism to expand derives from this sort of subsidy of the capitalist sphere by the domestic sphere (e.g., 1981, 138–144; for a description and criticism of this view see J. Kahn, 1980, chap. 10).

Meillassoux's general argument, developed through the consideration of African material, recently has been applied to Papua New Guinea by Peter Fitzpatrick (1980). Much of his analysis of Law and State in Papua New Guinea consists of arguing that colonial and postcolonial government policies served the interests of encroaching metropolitan capital by supporting and solidifying a number of elements of traditional Melanesian life. He points to repeated efforts by colonial administrators to protect village Melanesia from disruption; the result, he says, was the continu-

ation of institutions outside the capitalist-dominated sector which bore the costs of social reproduction and which thus subsidized capitalism by allowing artificially low wages. Likewise, and again following Meillassoux, Fitzpatrick sees the concern for village identity and ethnicity in Papua New Guinea's developing national ideology, exemplified by the legendary diversity of Melanesian societies, as a device that fragments the urban population and hinders the development of class consciousness among the emerging proletariat, again benefiting capital.

This articulation argument, especially as Fitzpatrick has applied it to Papua New Guinea, is macroscopic. It is concerned with the overall articulation of capitalism and the village sector, and Fitzpatrick ignores the sort of detailed analysis of village life that would demonstrate how it is affected by encroaching colonial capitalism. Furthermore, his overall thesis is one of the perpetuation of tradition, explaining just the sort of persistence that anthropologists concerned with Melanesia were beginning to discuss. The upshot of this is that in this model, village and town, tradition and modernity, are separated from each other except at the most macroscopic level. It is not far off the mark to say that while capitalism, according to this model, surrounds and perpetuates traditional life, it does not really impinge upon it; or to repeat what Michael Young said, a century of colonization leaves village life intact.

Not only does the Meillassoux-Fitzpatrick model of articulation warrant a separation of town and village, equally it does not suggest that it is necessary to study village life in its own right. Villages are labor reserves, seats of the domestic mode of production which subsidize encroaching capitalism, and there is little more that we need to know about them. Once capitalism encroaches, villages lose the sort of internal dynamic that was the focus of the whole first half of Meillassoux's Maidens, Meal and Money . Instead, they become passive, locked into their traditionalism and slowly drained of their vitality. Villages do not, then, respond to capitalism in a generative way by becoming something new; they do nothing but decay.

This approach presupposes what we found so curious on our arrival at Ponam and what we found even more curious as we learned more of the society and its past: the nature of its traditional elements. As we shall show, the impact of colonization on Ponam's

trade-based economy led to the modification of some traditional elements and led to the creation of new, traditional-seeming elements. Thus, in suggesting that the survival of tradition is unproblematic, Meillassoux and Fitzpatrick may be misperceiving both the nature of traditionalism and the impact of capitalism.

In a sense, then, Ponams' situation was very much like that of the Dyula of the Ivory Coast, described by Robert Launay (1982). They too survived largely by trade before colonial contact, but they found that colonization destroyed the old economic divisions and relationships on which their trade and their economic survival relied. What Launay has shown is that modern Dyula social organization and practices consist of some elements that developed as a direct response to colonization and some that survived from the precolonial period. "Yet," he points out (1982, 166), "the very features which seem loudest to proclaim the continuity turn out, on closer inspection, to exhibit subtle but hardly inconsequential differences with the way things were." Like Ponam, then, even at the most traditional core of Dyula society lie beliefs and practices that have been shaped by, and thus are linked to, the wider world of which the Dyula have become a part.

This relationship between traditionalism and the impact of colonialism has been highlighted in a striking way by Joel Kahn (1980, chap. 8). He has argued that what previous writers had taken to be traditional social structure in areas of what is now Indonesia, especially the existence of autonomous, self-sufficient communities, was not in fact the persistence of traditional forms. Rather, he says that the existence of these communities was a consequence of Dutch colonial practices in the seventeenth, eighteenth, and nineteenth centuries. These practices depressed local commerce, focusing transactions instead on the Dutch and their agents, a step that made villages more independent of each other than they had been previously. Likewise, the growing Dutch policy of taking agricultural produce, especially pepper, in the form of tribute levied on the local leader, strengthened the local kinship hierarchies, the adat . In conjunction with the depressing of local commerce, this destroyed independent commercial agricultural production and strengthened subsistence production. At its height, under what is called the Culture System, these policies produced, at least in West Sumatra, "a static system of corporate

lineage organization, an increased autonomy of village communities . . . and hence what [Kahn has] called the development of a traditional society" (1980, 163–164).

These studies contain a common message. The persistence of traditional social forms in the face of colonization needs to be approached with caution. These forms may not, in fact, be traditional at all (West Sumatra), and apparently traditional practices may be changed in subtle but important ways by colonization (Dyula).

In addition to showing that seemingly traditional elements may be modern, our analysis of Ponam shows another danger of assuming the persistence of tradition. Some elements of Ponam cultural forms survived colonial impact relatively well. However, because of the radical changes in the situation in which they persisted, the practical activities associated with them also changed radically, with radically different social consequences flowing from them. In such a case, persistence occurs at the level of form, but not at the level of substance. Thus, our description of Ponam and its changes since colonization points to an even greater need for caution in assessing the persistence of traditional forms and indeed raises questions about what "persistence" might mean in the face of rapidly changing circumstance.

For Meillassoux and Fitzpatrick, then, the village is not an object of investigation in its own right. This is because their conceptual orientation is from the national sector or economy looking outward. Though it is a bit of an exaggeration to say so, they treat village societies as black boxes, as undifferentiated (and un-understood) entities conceptualized almost exclusively in terms of their colonial and postcolonial economic function.

Thus, the model ignores the sort of intravillage dynamics and intervillage differences highlighted by, for example, Deborah Gewertz's reanalysis (1981) of Margaret Mead's description of gender relations among the Chambri, of the Sepik area. She shows how those relations, which Mead took to be traditional Chambri culture, were instead a result of colonial Australian intrusion and its effects on political and economic relations between the Chambri and their neighbors. Until studies of articulation are prepared to move beyond their macroscopic focus and recognize and try to explain this sort of change, they cannot come to understand how colonization has affected Papua New Guinea villages.

Moreover, because the Meillassoux—Fitzpatrick model does not attend to the details of village life, it does not tell us how articulation actually takes place, does not, except at the grossest level, identify what the mechanisms are that articulate village and town, precapitalist and capitalist social units. The macroscopic relationship is not traced out in any detail on the ground: there is simply no theoretical need to do so. Consequently, the model does not provide an adequate description of the consequences of articulation for the internal organization of villages.

Some work has been done to illuminate, in fact if perhaps not in name, the nature of articulation from the village viewpoint. A fairly early spate of such studies focused on entrepreneurs, the innovative individuals who are the locus of the articulation of the money economy and village life. These entrepreneurs have been most noticeable in the forward-thrusting Highlands region of the country. The best known of these studies is Ben Finney's Big-Men and Business . For Finney in particular, entrepreneurs are presented as unstable creatures, shuttling back and forth between the two social systems they inhabit, using their gains in one realm to shore up and extend their position in the other. From our point of view, these sorts of studies present two important problems. The first is their methodological individualism. They are concerned primarily with the individual transactions and histories of individual people, rather than with the organization of village societies. The second problem, particularly noticeable in Finney's work, is that this approach tends to be urban-oriented: people are selected for study precisely because they are doing well in the urban sector.

These twin weaknesses become more apparent if we look at two studies that more clearly represent the two tendencies we have described. One is T. K. Moulik's analysis (1973) of cultural factors affecting cash cropping in various regions of the country. Moulik illustrates what we take to be the inherent methodological individualism in entrepreneurial models of the sort popular in the literature around 1970. Moulik sees entrepreneurs in terms of individual strategies and transactions and does not attend to the more broadly structural factors constraining these entrepreneurs, though to be fair we should note that this individualism was ascendent more generally around this time among anthropologists concerned with Papua New Guinea (A. Carrier 1983, esp. 77–83). Thus, while Moulik's orientation focuses on villages rather than the national

economy, his explanation of variation in entrepreneurial innovation reduces to social psychology: where psychological constraints (for example, a fear of sorcery) are strong, individuals will not be motivated to engage in cash cropping.

The second of these studies is Rolf Gerritsen's description of Highlands entrepreneurs, which, though published relatively late (1981), was undertaken in the early 1970s. Gerritsen avoids the individualistic tendency in entrepreneurial studies. Although he is concerned to study a fairly small group of people, his model defines them not as innovative individuals, but as the emerging elements of a new class. Thus, he takes a firm hold of structural factors and processes. However, in doing this he embraces the second problem we mentioned, a strongly urban or national orientation. Gerritsen is much more concerned to establish the existence of an emerging big peasantry in the political economy of the country, to plot its links with the state and its privileged access to state resources, than he is to trace out the social organization of the villages from which these big peasants have emerged, and the way that articulation links these villages to the national economy and affects their structure.

Since the first wave of these studies around 1970, analysts have become better able to relate entrepreneurs or the emerging big peasants, depending upon one's perspective, to village society. However, difficulties remain. The nature of these is illustrated by Lawrence Grossman's recent description of cattle projects in a village in Eastern Highlands Province (1983). Grossman locates the emerging big peasants firmly in their village milieu, by the simple expedient of looking at village enterprises, cattle projects, rather than individual entrepreneurs. Nonetheless, he fails to link these projects more than cursorily to the social organization of the village in which they occur.^[2]

The way that we think he fails to do this illustrates an important shortcoming in much of the recent literature on articulation in Papua New Guinea: the fact that even though the site of the study has moved from town to village, the orientation, the conceptual framework of the study, is still firmly rooted in the national economy. Grossman focuses on village relationships, especially between leaders and participants in the cattle projects. However, his perception of these relationships is shaped by his interest in commodity relations of the sort found in the national economy: he

describes (1983, 67–70) how leaders and participants differ in terms of their investment of labor and money to the projects, and in terms of the distribution of surplus product from the projects. Thus, while Grossman focuses on village relationships, he ends up selectively perceiving and recasting them in terms of the logic of the national economy. The consequence of this, which is almost certainly not what Grossman intended, is to elide all village relationships that cannot be reduced to this economistic calculus, and so obliterate differences between capitalist and precapitalist forms.

While the empirical studies we have described in the last few paragraphs do show an effort to look at articulation from the village end, equally, as we have argued, they are not really satisfactory. Partly this is because these writers did not develop a theoretical framework encompassing both town and village, a framework that is necessary if we are to be able to orient village-level studies systematically to the process of articulation. We think the basis of such a model exists in C. A. Gregory's work.

In his paper "Gifts to Men and Gifts to God" Gregory makes the argument that Western capitalist and Melanesian precapitalist systems are oriented in fundamentally different ways, with the corollary that terms drawn from one are not adequate to describe the other. Gregory characterizes the distinction between these two systems in terms of the distinction between gifts and commodites. He outlines four basic differences between gift—debt and commodity—debt, two of which are important here. One points to fundamental differences between objects and transactors in the two sorts of systems: commodity—debt "is created by the exchange of . . . objects between transactors who are in a state of reciprocal independence, whereas gift—debt is created by an exchange of . . . objects between people in a state of reciprocal dependence" (1980, 640).^[3] Second, while gift—debt on the one hand "must be explained with reference to the social conditions of the reproduction of people ," such as "clan structure and the principles governing kinship organization," commodity—debt on the other hand "must be explained with reference to the social conditions of the reproduction of things ," such as "class structure and the principles governing factory organization" (1980, 641). For the sake of convenience, and borrowing Gregory's terminology, we will refer to these two realms as the realms of gift relations and commodity relations.

We think Gregory has made a point that is no less important

for seemingly being obvious. The cultural logic and sociological order of the encroaching capitalist system revolve around class and the production relationships that are a part of it, the realm of commodity relations. Alternatively, the cultural logic and sociological order of precapitalist Melanesian societies are quite different, revolving around kinship and the exchange relationships that are a part of it, the realm of gift relations.

An example will help illustrate this point. That example is "wok meri," perhaps most briefly defined as a collective women's saving system around Goroka, in the Papua New Guinea Highlands. As described by Lorraine Sexton (1982), "wok meri" was an attempt by village women to adopt and adapt their own version of important Western economic institutions.

The first point we think worth making is that the Western institution involved was banking, an institution that is concerned with exchange rather than production. Second, in adopting banking—in, if you will, remaking banking into a more comprehensible and congenial form—women recast it in terms of "the social conditions of the reproduction of people ," such as "clan structure and the principles governing kinship organization" (Gregory 1980, 641).

Thus, women involved in the different "wok meri" groups were recruited from the in-marrying wives of men's agnatic lineages. The groups themselves were related conceptually as elements of matrilineages: each group was the daughter of an earlier-formed group which was instrumental in founding the new group, and each group was expected to produce its own daughter by helping to form another group. Furthermore, the main transactions involving groups, one near the time of the foundation of the daughter group and one to mark the end of the main phase of the collection (that is, saving) of money, were modeled explicitly on marriage ritual, though Sexton notes (1982, 177) that she detected in the ritual surrounding these transactions elements of all rituals associated with the female life cycle except menarche.

Our description of "wok meri" is cursory, but it is enough to illustrate the point we want to make. Here, village women typical of the communities in the Goroka area produced an economic institution modeled on the introduced Western institution of the bank. These women did not have to transform the institution they adopted; they could have simply set up informal village-based

banks. However, they did transform it, and they did so in a way that fits Gregory's synoptic characterization of Melanesian precapitalist systems.^[4]

This example illustrates the point we drew from Gregory's work. That is, that precapitalist Melanesian systems are dominated by kin relationships and other elements of the "social conditions of the reproduction of people." This has implications for an anthropological approach to articulation, one that recognizes that village systems have their own structures and practices. Even if these villages may be subordinated empirically to encroaching capitalism, the terms and concepts used to describe them should not be subordinated to the conceptual framework of capitalist commodity relations.

Gregory sets out to produce an overall theoretical appreciation of this sort of articulation in his Gifts and Commodities , expecially chapters 6 and 7, "The Transformation of Gifts into Commodities in Colonial Papua New Guinea" and "The Transformation of Commodities into Gifts in Colonial Papua New Guinea." The first of these chapters performs its task admirably. Echoing Meillassoux, it is a systematic description of how gift relations in the village affect commodity relations in the capitalist sector, showing how the domination of gift relations in villages so diffuses the cost of social reproduction that employers in the capitalist sector were able to pay wages well below the actual cost of the reproduction of the labor force, and hence well below the level of wages expected in wholly capitalist systems. Hence, Gregory argues that commodity relations in the capitalist sector of colonial Papua New Guinea were affected by articulation with precapitalist villages and consequently were different from commodity relations in core capitalist areas where this sort of articulation does not take place.

The second of these two chapters is the one that, by its title, bears more directly on the issues we have raised in the preceding few pages. Unfortunately, however, this chapter is unsatisfactory. Certainly it is nothing like the inversion of the preceding chapter that its title suggests: a systematic discussion of the ways that village precapitalist societies have their system of gift relations affected by their link to commodity relations in the capitalist sector. Instead, Gregory contents himself with elaborating an internal classification of gift exchange patterns without reference to com-

modities or commodity relations and with defending the proposition that "the gift economy of PNG has not been destroyed by colonization, but has effloresced. This is reflected in a tendency for European commodities to be transformed into gifts" (1982, 166). The first sentence of the proposition is defended primarily by describing briefly a large number of Melanesian exchange systems; the second sentence, the one with the most interesting implications from our point of view, is left unexplored.

We will attempt such an exploration in this book, showing how the Ponam system of gift relations, the system especially of kinship and exchange, bears the mark of that articulation. In villages like Ponam, systems of kinship and exchange will not be shaped only by factors internal to village life. Instead, they will be shaped as well by their links with the capitalist sector. This point is not, we think, a contentious one.

However much their studies do not always conform to the ideal, anthropologists have long recognized the need to take into account the impact of the outside world in their analyses of village societies. All we are suggesting is that the form of this impact, and thus ways of taking it into account, should be considered problematic. Common sense might suggest that the impact of articulation with the capitalist colonial sector, a sector so oriented to production and commodity relations, would be manifest in the realm of material production and the emergence of commodity relations in the village. Such a view would appear to leave relations of kinship and exchange, especially ceremonial exchange, relatively untouched—barring, of course, the appearance of imported Western commodities as exchange goods. However, the theoretical points we have made here suggest that common sense may be misleading, that kinship and exchange are affected in important ways by articulation, though the form of this impact may not be apparent to those who do not approach it with the sort of perspective we are describing here.

Our point, put most simply, is that a rounded approach would have to try to trace out the articulation between the commodity relations of the national economy and the gift relations of the precapitalist village sector. It is this view of articulation that underlies our objection to the urban and commodity-oriented terminol-

ogy of some village-level studies of articulation. These do not try to demonstrate articulation by looking at the dominant village realm of kinship and exchange, gift relations. Instead they look at what are but the precursors of commodity relations in village life. Failure to look for the signs and processes of articulation in the dominant village realm of gift relations constitutes a failure to see how articulation with the expanding national economy impinges upon the core elements of precapitalist society. Such a limited approach may contain interesting information and insights. However, what it ends up assuming is that the capitalist sector displaces village social organization, not that it articulates with it.

Other anthropologists have tried to identify and investigate the effects of articulation on kinship and exchange. The most obvious example is Deborah Gewertz, whose work on Chambri social relations parallels our interests quite closely, as our earlier discussion of one aspect of her work and our later references to her writing in this book illustrate. Another example is Andrew Strathern, who has dealt with this issue a number of times (most recently 1984), particularly in "Gender, Ideology and Money in Mount Hagen" and "The Division of Labor and Processes of Social Change in Mount Hagen." These papers are partial approaches to articulation, in the sense that they are concerned with events from the village perspective and so do not attempt to analyze changes in the capitalist sector or the ways that these are affected by articulation with villages. However, within the realm that concerns him, Strathern looks at just the sort of issues that, we said, are central both to a theoretically informed view of articulation and to a theoretically informed view of kinship, exchange, and social reproduction.

That is, both of these papers note the way that exchanges, and especially moka (the elaborate regional exchange system), are a crucial aspect of the maintenance and reproduction of social relations among men and between men and women. They then go on to describe the way that the growing colonial and postcolonial impact on Hagen society has affected moka and the relations of which it is a part. Strathern thus demonstrates a sensitivity to the point that exchange and the kin and other social relationships that are bound up with exchange can be understood adequately only if one recognizes that this cluster, crucial to "the social conditions

of the reproduction of people ," in Gregory's terms, is a central aspect of the articulation of capitalist and precapitalist social forms in Melanesia.

We hope that our analysis of Ponam economics, articulation, kinship, and exchange contributes to this understanding.

Our vehicle for describing Ponam's articulation with the developing commodity sector of Papua New Guinea is a presentation of the economic activities and relationships that were the backbone of modern Ponam life: how wealth of different sorts was produced and circulated and, ultimately, consumed. While the economy was the backbone of Ponam, and hence crucial to understanding the society, by itself it was not enough to hold the society up. It needed the muscle and gristle of kinship and descent, religion, ceremony, and all the rest that constituted the tissues that supported, while being supported by, the economic frame. Likewise, even though Ponam is an island, before and since colonization the life of islanders depended on lives beyond the island, and Ponam's way of life was shaped by the way of life of other people in other places. So, in being concerned with Ponam economic activity, we are necessarily concerned with much more than just economics and with much more than just Ponam.

As we have made clear, we do not think it is possible to describe Melanesian societies generally, or Ponam Island in particular, without paying attention to relationships between the society and the outside world, a world that has expanded breathtakingly since colonization. In order that we can deal with this relationship adequately, we divide our description of Ponam economics into three parts: activities revolving around and directed toward local production and circulation, activities revolving around and directed toward overseas production and circulation, and activities revolving around and directed toward the articulation of these two areas of economic life, activities primarily of circulation rather than production. The definitions of "local" and "overseas" are relative, and with the passage of time they have changed a great deal. Because, however, this is a book about Ponam at the time of our fieldwork, generally we construe "local" and "overseas" as they applied at about 1980.

Our distinction between local and overseas is, of course, artificial, because activities in each area affected and reflected the influence of the other, as well as the activities that articulated the two realms. It is not, however, a wholly arbitrary distinction, reflecting as it does the sharp geographical separation of residents and their local production and circulation on the one hand, and on the other the migrants who constituted the dominant direct Ponam involvement with the larger economic order. Moreover, this parallels a distinction that is deeply rooted in anthropological thought as well as in the thought of Ponams themselves, as we shall discuss in chapter 5, the distinction between the traditional and the modern spheres of life.

The first of our spheres is domestic economic activity, production and circulation undertaken by people living on Ponam Island. At the time of our fieldwork this consisted mostly of fishing, the dominant conventional village-based productive activity, and market trade, an important way that islanders secured the necessities they did not produce for themselves. These economic elements deserve to be called "traditional" for two reasons. First, they were central elements of Ponam economic life prior to colonization and so have a historical provenance that many of their other activities lack. Second, they are central elements of the way Ponams saw themselves and their past. Ponam self-conception had many aspects, but one important one was as island people who fish and trade, an identity that they contrasted with mainland Manus people, whom they saw as agriculturalists. At the time of fieldwork, however, these activities were taking place in conditions very different from those that existed before colonization. Thus, even though the physical activities and even many of the social relations that were part of domestic activity survived through time, their social and economic dimensions, invisible but nonetheless real, had changed so much that it is not at all clear that, taken in the round, the fishing and marketing that Ponams did in 1980 were the same things they did in 1880, or even in 1930.

Complementing domestic economic activities were foreign or overseas economic activities. These were foreign because they required a much closer involvement with distant places and people than did domestic economic activities. Thus, we are not referring

to the sort of indirect involvement that comes with being part of an integrated economic system, but direct personal involvement. Foreign economic activities would include migrant labor, cash crop production, and other forms of petty commodity production.

As we have made clear, however, unlike many areas of the country, Ponams grew no crops for sale, manufactured only the smallest quantity of handicraft goods, and sold almost no fish outside their local markets. Thus, for Ponams, as for most of the other societies in Manus Province, the prime foreign economic activity was paid employment away from home, and usually outside the province, primarily in the civil service, the largest employer in the country.

Even though this migration was both relatively recent and alien to what Ponams were accustomed to doing, its consequences had come to shape not only central elements of social life on the island, but also Ponams' self-concept. At the same time that they maintained their traditional identity as island people who fish and trade, they established another, more modern identity, as people who must migrate to seek wealth. And this reliance on migrant labor as the only significant source of cash income for islanders made Ponams very much the circular migrants that the Meillassoux-Fitzpatrick model describes.

Ponam wage employment was not an isolated act or event, and Ponams with paying jobs did not spring into existence their first day at work or cease to exist the moment they quit. Instead, they began to work only after they had been shaped by their childhood on the island and, unlike migrants from many other parts of Papua New Guinea, their years in Ponam's village school, in secondary school, and, increasingly as the 1970s turned into the 1980s, in tertiary institutions. Though they were not so at the beginning, by the 1960s Ponam migrants were thoroughly educated and paid accordingly. If we are to describe Ponam's foreign economic activities, then, it is necessary as well to describe education on Ponam.

These, then, were the two complementary spheres of Ponam economic life in 1980. One was domestic activities, the direct descendents of precolonial economic life. These sprang from the island's and region's endowments of land and sea, sun, wind and rain, and the forms of production and circulation they underlay; they involved relations of exchange and reciprocity; they were

oriented toward subsistence and consumption, and hence ultimately were inward-turning, concerned with the local society and its immediate neighbors rather than the larger world. The other was foreign activities, the consequences of the new relationships islanders entered as a result of colonization. These sprang from the colonial and postcolonial economic order; they involved commodity and wage relations; they were oriented to the national monetary economy, and hence were fairly closely tied to economic relations with Australia, Papua New Guinea's former colonial power, and the international economy generally.

The articulation of the overseas and domestic realms appears in a number of ways in this book. The most explicit discussion, in chapter 6, follows our analysis of these two realms and looks at the way ceremonial exchange mediated between residents and migrants. This mediation tied migrants to home and motivated them to save money and remit it to Ponam, a remittance that produced a number of important results. It subsidized life on Ponam, keeping it pleasant and attractive to migrants, and thus strengthened their commitment to home and their willingness to send money. As part of this subsidy, remittances enabled residents to cope with their deteriorating position relative to their Manus trade and market partners, and so helped maintain that area of local economic activity. Also, it increased the economic significance of ceremonial exchange, making it remunerative to residents in fact as well as in ideology. This seems to have led to an increase both in the frequency of exchange and in the proportion of resident adults participating in it. This had the effect of strengthening the importance of the kin groups that played a central role in this exchange and the kin relationships that shaped people's access to and place in exchange. And in its turn this strengthening helped these groups remain important in the other area of life in which they featured, the ownership of property, particularly property relating to traditional fishing, the main element of domestic economic production.

Somewhat less explicitly we describe other important aspects of this articulation, both as things stood at the time of fieldwork and as they had changed since colonization. What we are concerned with here, however, is the relationship between Ponam social structure and the island's relations with the outside world. As

colonization changed Ponam's significant "outside world" from the rest of Manus to the Papua New Guinea national economy, Ponam's relationship with the outside world changed, and this had important consequences for the ways Ponams related to each other.

Briefly, prior to colonization, when the outside world was the rest of Manus, relations among Ponams were unequal, with the dominant position going to those people having the strongest control over the production and circulation of wealth within the region. However, as the outside world expanded and as the economic importance of Manus itself for Ponam decreased, islanders lost their ability to control significant sources of wealth, and particularly wealth from outside the island, as they became dependent on the emerging national economy. Consequently, no group of villagers was able to control wealth to the exclusion of others, and relations among villagers became more equal. While it is possible that some islanders could develop effective control over significant outside sources of wealth and so assert their dominance over their fellows, by the time of fieldwork this had not occurred, though changes in the organization and operation of provincial government suggested that it could easily occur in the future.

In sum, then, this book uses a description of Ponam economics to address two broader issues. The first of these is the connection between Ponam wealth production and the nature of island social relationships. Our analysis of this connection is made easier by the fact that we are able to trace out changes in both these areas over a substantial portion of the colonial and postcolonial period. Since colonization, and especially since World War II, this connection has been to a large degree the connection between Ponam and the national economy. Thus, we are obliged to address the question of articulation. This is the second of the two broad issues we consider in our study of Ponam society and economy.

Our book is about Ponam Island and Ponam Islanders. However, as our discussion in this introduction has made clear, we think the case we present addresses general anthropological questions and so has a broader applicability to Melanesian societies. We intend our analysis of Ponam to be of more than purely ethnographic interest. We want now to discuss briefly the ways we have attempted to fit our description of Ponam into a larger and more

comprehensive framework and the advantages and dangers inherent in our attempt.

To begin with, we have decided to avoid any extended, explicit comparative discussion, at least once we start our substantive discussion of Ponam in chapter 1. The common practice of a concluding, comparative chapter is useful for those who are presenting a detailed analysis of a society from an area that has been studied extensively. In that sort of circumstance, the weight and detail of ethnographic evidence needs to be deployed before it can be brought to bear in addressing regional ethnographic variations and the issues that those variations allow one to discuss.

Our case is rather different. While Manus may be well known in Melanesian ethnography, in fact it has been studied fairly little. And in any event our purpose is not to address regional patterns or variations. Instead, our aims are less scholarly and more polemical. One aim is to assert that societies in Melanesia are likely to be affected by links with the outside world in a way that has, as we have argued in this introduction, been ignored or slighted by many anthropologists working in the area. A second and related aim is to put forward by example an assertion we have made in the abstract already. That is the assertion that the articulation of a village society with the urban, capitalist sphere is going to have what may be its most interesting anthropological effects in the dominant sphere of village life, that of kinship and exchange.

If we were to pursue a full-blown comparative study in furtherance of these aims we would have to undertake two arduous tasks: an analysis of the ways other anthropologists have ignored the effects of the sort of articulation that interests us, and an analysis of the ways that articulation really is manifest in Melanesian societies. The first task is enough to daunt those much abler than we are, and the results would likely be either highly contentious or soporific. The second task is even more daunting. If our first assertion is correct, and we have already presented enough discussion of Melanesian ethnography to make that at least plausible, then our second task would be almost impossible, for the ethnographies on which we would have to rely would omit just those areas of linkage that we would need to study.

We have, then, made no attempt to present a systematic overview of comparable societies or ethnographies. Part of the reason for this, as our discussion thus far has hinted, is that there has

been little work done in Papua New Guinea or elsewhere in the region that is clearly comparable to our analysis of Ponam: the studies of articulation, migration, and remittance that exist are concerned with very different aspects of these topics. Instead of systematic comparison, we have elected to intersperse comparative comments throughout the book. Further, we have chosen to avoid comparison for its own sake as well as comparison on points of ethnographic interest. Instead, we limit ourselves to comparisons that bear on key aspects of our main arguments.

Made uneasy by the fear that this would limit too much the points of contact between Ponam and other societies in Melanesia, we have sought also to point out the ways that Ponam is unusual. This allows us to indicate points of contact because it necessarily entails comparison with other Papua New Guinea societies, even if only implicitly. When we think it is pertinent, we point out the dimension or variable that concerns us; we place Ponam on that dimension; and we show how the island stands on the dimension compared to other Papua New Guinea societies. This technique allows us to provide a framework of variables important to the issues we address, which readers can use to place Ponam relative to other societies with which they may be more familiar.

We hope this technique allows us to avoid the main disadvantage that Ponam's unusual situation presents, without losing the advantage. That disadvantage is that, because Ponam seems to be extreme on a number of dimensions, often it is not reasonable to generalize from it directly to other parts of Papua New Guinea. Pointing out pertinent dimensions and Ponam's place on them allows us to begin to show in what ways Ponam differs from other societies and what those differences might mean. This in turn allows us to begin to show what Ponam can tell us about other Papua New Guinea societies.

Before concluding this introduction we want to point out the main dimension that concerns us, the degree of village dependence on the urban sphere. Probably the most widely known studies that deal explicitly with the link between villages and the urban sector in Papua New Guinea are Cyril Belshaw's The Great Village , A. L. Epstein's Matupit , and Richard Salisbury's Vunamami , dealing with villages engulfed by, on the edge of, or very close to a city. Ponam looks very different from both the thoroughly urban Hanu-

abada and the more periurban villages of Matupit and Vunamami, villages that maintained significant subsistence and commercial agricultural bases controlled through indigenous systems of land tenure.

Even though Salisbury asserts that Hanuabada and Matupit had been so engulfed that each looked like becoming "a substandard if 'quaint' part of a foreign town" (1970, 174), local custom appears to have remained strong. Thus, for instance, Epstein (1969, 223–224) says that Tolai tambu remained important in affinal exchange, was still controlled by the older generation, and thus still served as a mechanism by which seniors dominated their juniors. As arguments we make in this book imply, we think this survival of indigenous forms of inequality is related to the survival of indigenously controlled productive resources.

Ponam is set off from these villages by its degree of involvement with the urban sphere. Ponam was much more isolated than Hanuabada, Matupit, and Vunamami, but in some ways its links to town were much more significant. This is because the absence of a local productive base of much significance means that Ponam was much more dependent, even though much more isolated, than the other three villages. The very high degree of Ponam dependence makes it possible to see the effects of dependence clearly. And this is one of the prime virtues of studying so extreme a case.

Thus, in its very extremity, Ponam approximates a pure case, which allows us to see with especial clarity the results of certain processes or the relationship between certain variables, without obscuring, extraneous influences. To draw a parallel from elementary physics: feathers falling in vacuum tubes certainly are unusual, but are quite useful if we want to discern the effects of gravity free from the obscuring influence of air resistance.

While neither anthropology nor our approach to Ponam is as rigorous as elementary physics, the parallel is apt. For example, we have mentioned already that Ponam produced almost no significant wealth, but relied on remittances. This is unusual, but also quite helpful if we want to see the relationship between wealth production and social inequality, or the relationship between village social life and dependence upon circular migration.

Having pointed out a key way that Ponam was unusual, we want to dispel the idea that it was so unusual that it can be ignored

by those concerned with the conventional anthropologist's topics, social relations and structure in rural villages. Although, as we noted earlier, we do not want to undertake a substantial, systematic comparison with other Papua New Guinea societies, we do want to make two comparisons here. These will not prove anything. They will, however, buttress our suspicion that the processes we observed so easily in Ponam are there in much of village Papua New Guinea, albeit in a milder form. The cases that interest us are that of the Chambri Lakes people described in Deborah Gewertz's Sepik River Societies and that of the Gende people of upland Madang, described in Laura Zimmer's "The Losing Game."

The people Gewertz studied were in many ways typical Melanesian villagers. They were relatively isolated from the outside world: they were far from towns, plantations, and mines. Equally, they engaged in no significant petty commodity production of their own and were not dependent on remitted wages for their survival in the way Ponams were. Even so, Gewertz notes some important ways village life had been affected by contact with the outside world.

The effect that concerns her most is indirect, brought about by the imposition of state control of the Sepik region. As was the case with Ponam, which we will describe in chapter 2, this led to significant changes in relations between the fish-producing Chambri and their starch-producing trade and market partners. One consequence of this was the deterioration of local trade. This forced people to begin to look outside the local economy for their necessities, and as part of this Chambri women began to travel farther afield, to marry elsewhere, and to become lost to the village. (As Gewertz's description of men's reaction to the Salvinia infestation of the late 1970s shows, this had a significant effect on the way Chambri men saw their society.)

The upshot of even this relatively mild reliance on external sources of wealth was, as Ponam's case will suggest is likely, the deterioration of preexisting relations of inequality. Gewertz says the consequences "have been destructive of the very relationships Chambri elders hoped to maintain, and the texture of life within the three Chambri villages when I was last in the field can only be described as anomic. No one is sure any longer who is beholden to whom, or of how to establish and maintain social relationships" (1983, 195).

The Gende present a similar picture. Even though Zimmer describes flourishing local production, local trade networks decayed with colonization. This made it harder for the Gende to maintain relations with neighboring groups, relations that, significantly, included the giving and getting of women in marriage. The Gende were not, at the time of fieldwork, close to sources of wage labor, and Gende migrants seem to have been less successful and less assiduous about remitting money home than were Ponam migrants, so that survival still depended overwhelmingly on subsistence production. Even so, Zimmer notes that the deterioration of local trade networks led to an increasing demand for cash and cash goods in exchange. Consequently, social success required access to both Western and indigenous sources of wealth.

The result was that Gende aspiring to full and prosperous adult status had to have migrant relatives remitting money. The net effect was that social success became, in terms of earlier Gende values, capricious: virtue and reward no longer marched together, and the moral, as well as the social, bases of Gende inequality was under threat.

Although these are only two cases, they establish one important point: apparently slight degrees of contact and colonization can lead to substantial changes in apparently purely internal village social relationships, such as those of marriage, gender, ceremonial exchange, and inequality more generally. Ponam's situation was extreme to be sure. However, in attenuated form that situation can appear even in seemingly isolated villages and regions without the long period of intense colonial and state penetration that characterized Manus Province. The lessons that Ponam allows us to draw may apply to many more villages than would naively appear to be the case.