Version classiqueVersion mobile
OpenEdition Books

Rank and Status in Polynesia and Melanesia

Douglas L. Oliver

Race, Class and Ethnicity : Industrial Relations in the South Pacific with Special Reference to Fiji and Bougainville1

Alexander Mamak et Richard Bedford

Texte intégral

  • 1 The fieldwork conducted by the senior author in Fiji between 1970-72 was made possible by a grant (...)

1The importance of class analysis for explaining contemporary behaviour in the South Pacific is indicated by several developments in the early independence period, namely, the pressures of inflation, increasing rates of urbanization, and the localization of positions formerly held by expatriates. This paper delineates tendencies in class formation and class conflict with special reference to Fiji and Bougainville. Previous accounts of social conflict in this region tended to focus solely on race, mainly because stratification by race and by class tended to coincide. Such an approach precluded an analysis of the underlying class structure that was emerging, and the position of race and ethnicity in that structure. It is suggested that a number of factors, both historical and contemporary, must be taken into account before an attempt can be made to determine the extent to which urban man in the South Pacific is likely to develop into an urban industrial proletariat. A more balanced view of the complex roles of race, ethnicity, and class in contemporary "class" formation is provided so as to enable their functions in the newly-developing urban societies of the South Pacific to be seen with more coherence than is perhaps the case up to date.

2L'importance de l'analyse de classe pour expliquer le comportement contemporain dans le Pacifique du Sud est indiqué par l'évolution de la situation au début de l'indépendance, c'est-à-dire, les pressions in-flationistes, les taux croissants d'urbanisation, et la localisation des postes tenus auparavant par les expatriés. Cet article décrit les tendances dans la formation de classe, et le conflit de classe, avec une référence particulière à Fidji et Bougainville. Les descriptions publiées antérieurement sur les conflits sociaux dans cette région avaient tendance à se concentrer seulement sur la race, en grande partie parce que la stratification par race et par classe avaient tendance à coïncider. Une telle approche empêcha une analyse de la structure fondamentale de la classe qui émergeait et la place des races et ethnies dans cette structure. Il est suggéré qu'un certain nombre de facteurs historiques et contemporains devraient être considérés avant qu'un effort puisse être fait pour déterminer dons quelle mesure l'homme urbain dans le Pacifique du Sud pourrait se transformer en proletariat urbain industriel. Un aperçu plus équilibré des rôles complexes des races, ethnies, et classes dans la formation de "classe" contemporaine est fourni pour permettre que leur fonction dans les sociétés urbaines du Pacifique du Sud en cours de développement puisse être en vue avec plus de cohérence qu'auparavant.


3More than two decades ago Professor Douglas Oliver (1961: 426) was led to ask: "Just how susceptible would the islanders be to the doctrines of Marx?". There was then not much evidence of the acceptance of Marxist ideas, let alone the development of class consciousness, strength, and militancy to act as a driving force of change. But in recent years it would appear from the growing industrial unrest in some urban centres of the South Pacific that a crucial transformation has already taken place.

4The importance of class-analysis for explaining contemporary behaviour is indicated by several developments in the early independence period. Firstly, urban wage-earners in the South Pacific are beginning to represent a large section of the population as a result of increasing urbanization and the continuing displacement of agriculture by urban-based industries as the primary source of employment. Secondly, the advancement of a section of urban labour into skilled and managerial positions as a result of rapid political change and localization has widened the income gap among South Pacific peoples thus increasing the significance of class relations. The pressures of inflation have also encouraged urban workers to become more conscious of their economic roles, and to view themselves as members of distinct economic groups. Mass action by mine workers in Bougainville provides a recent example and is regarded, at least by the Socialist Labour League of Australia (Workers News, May 22, 1975) as part of a world-wide movement against monopoly domination and capitalism.

5While the development of economic corporate groups is a phenomenon common to most societies, and appears to be something new emerging in the urban societies of the South Pacific, the process of class formation and class conflict may be different and much more complex in the South Pacific than in other contemporary societies. In the South Pacific the process involves much more than a simple division between skilled and unskilled local workers, or a struggle between the proletariat and the modern capitalists. What seems to be developing is a complex interaction between economic, racial, and ethnic groups whose boundaries overlap in different ways over time, and leading to an interrelationship between racial, class, and ethnic consciousness. A brief description of a recent strike movement in Bougainville will suffice to show the difficulty involved in finding a proper race-class-ethnic synthesis necessary for explaining contemporary industrial action in the South Pacific.

The Bougainville strike

6On Monday, May 12th, 1975 an inter-ethnic brawl between a trade union official and a company security guard at the Bougainville copper mine soon spread to involve the majority of black mine workers at the site. Close to a thousand men under the leadership of their union began marching on the company's pay office demanding not only that the company re-engage the sacked union official but also wage increases and better working conditions. Women and children, most of them families of white expatriate employees, were hastily evacuated from the mine site. The industry was totally paralyzed during the two days of rioting that followed. Company property was destroyed with damages and production losses estimated at several million Australian dollars. Among the properties damaged was a recreational centre used primarily by white employees. Police action and mass arrests eventually put an end to the strike.

7Despite some evidence to the contrary both editorials in The Australian and the Sydney Morning Herald of May 15 emphasized the absence of anti-white sentiments during the disturbances. Company officials blamed the violence on a breakdown in communication between the trade union and its members, and claimed the incident had developed out of frustrations over differences in pay among black employees. While it is true that black employees are differentiated in terms of income level a close congruity of interest exists as a result of the growing awareness of greater disparity between black and white earnings. Black staff members did not take part in the strike, but a considerable number of high income, skilled tradesmen were directly involved. When a meeting to form a Staff Association was called soon after the strike only about ten of the two hundred or so people who attended were blacks. In brief, it is clear that racial, class, and to some extent ethnic hostility were present in the strike. But which form of conflict, consciousness, and identity is the most fundamental and important explanation not only of the Bougainville strike but of industrial life in the South Pacific in general? To what extent, and in what way is Man in the South Pacific likely to develop into an "industrial proletariat?" Before any attempt can be made to answer these questions it is necessary to provide a background in terms of the intellectual debate which surrounds the concepts of race, class, and ethnicity, and to examine the roots of contemporary movements and proletarian consciousness in the context of these terms.

The treatment of race, class, and ethnicity in the literature

  • 2 According to our usage of the race concept relations between members of a non-white population mho (...)

8The concepts of race, class, and ethnicity have often been used to explore the process and pattern of social relations and organization in developing countries. There is agreement in the literature on the general usage of these terms although some associated characteristics may be left out or emphasized more than others in some instances. As is the case here, these concepts are described in terms of organization and social relations between distinct groups of people. Depending on the level of analysis (e.g., ideological, organizational) they are seen as imposing constraints on interaction and generating conflict, hostility, and antagonism between the various groups. These types of relations and organization imply a sense of group identity and unity, distinct types of membership and inter-group behaviour based on skin colour (race), income level, education, occupation, etc. (class), or cultural variables such as custom, language, and religion (ethnicity). There are of course many situations--as we shall see--where class and race or ethnicity coincide. Finally, the treatment of class in this paper is limited primarily to the description and analysis of economic relations between whites and blacks or between blacks. By contrast, the use of ethnicity is generally confined to relations between nonwhites, while the concept of race is solely used in the context of black-white relations2.

9To date, no attempt has been made to give equal emphasis to all three concepts even though the formulation of one concept generally involves the others in some way. Therefore, in the following brief review of the treatment of race, class, and ethnicity it may be best to examine the emphasis placed on each of these concepts in terms of a race-class continuum with ethnicity taking on varying significance at either end.

The emphasis on race

10Theories of pluralism place a great deal of emphasis on the concept of race. In the plural model conflict is viewed primarily in terms of the relationship to the means of political power rather than to the means of production. Pace is of primary significances because in most colonial societies it is the white minority who has sole access to power.

  • 3 This is contrary to the belief of other race analysts mho argue that some classes are "missing" in (...)

11While plural theorists recognize the probable correlation between economic differentiation and racial stratification they question the primacy often given by Marxists to the economic situation. Economic inequality is viewed in the context of race and is seen as providing only one basis of racial differentiation: "Racial difference... comes to have social significance only as it is elaborated in systems of differential political incorporation, economic stratification, and racial segregation" (Kuper 1971: 595). It is also argued that "physical identifiability in a racially plural society is a more enduring identity than class membership in an industrializing homogeneous society" (Kuper 1971 : 594-595). Furthermore, plural theorists often claim that in most colonial societies low-income workers are too differentiated, both economically and ethnically, to make up a single class3. In light of this there is an absence of working class solidarity. Ethnicity may take on additional significance at the political level because plural theorists believe that with the political ascendancy of blacks at the time of independance there may be a radical transformation of ethnic categories into ethnic blocs.

12Within the range of race analysis is the emphasis given by a number of writers to aspects of race relations, with economic factors taking on added significance. This approach is illustrated by Gail Omvedt (1973) and others who argue that the major cleavage in colonial societies is racial : "The basic social relationship existing at the time of colonialism is racism" (See also Casanova 1965 : 33). Racial antagonism, however, is exacerbated by the correlation between race and distinct classes. Black workers, for example, are regarded as racially inferior and treated alike irrespective of their ethnic or class status. Because these workers are affected in common by racism their solidarity is based primarily on race differentiation and not on class. At the same time, ethnic diversity and minor class differences among the economically poor black workforce are insignificant partly because racial differentiation is more pervasive, and partly because the latter see themselves as members of a single, low-ranking group, in a society where rank is determined on the basis of colour.

13Let us recapitulate the above approach by comparing it with the plural model. In comparison to the plural model much greater emphasis is given in this approach to the economic situation. While primacy is still given to race this emphasis is explained on the basis of racial and class hierarchies coinciding at the wider level. Another important contrast is in the treatment of ethnicity. Plural theorists discuss and compare ethnicity primarily with class and suggest that ethnicity may preclude the development of class consciousness. On the other hand, Omvedt and others discuss and compare ethnicity with race and argue that the pervasiveness of race at the wider level may help to offset the development of ethnicity at the local level.

  • 4 The lack of attention paid to the social system of whites and the interaction between blacks and w (...)

14The flexibility of ethnic boundaries viewed within the wider context of racial dominance is evident in the work of such writers as Amin (1964), Rodney (1973), and Hlophe (1973). It is suggested, for example, that ethnic boundaries are often expanded to include members of the suppressed classes in interaction with whites (as reflected, for instance, in the statement voiced by members of distinct ethnic groups : "We are all blacks"). This theme is also evident in British social anthropology, particularly in the Rhodes Livingston-Manchester approach to African urban studies. According to this approach Africans involved in an urban-industrial environment are very quick to develop new sets of interests. Ethnicity as a category of interaction is sometimes largely irrevelant under urban-industrial conditions. This may be due to the predominance of the racial cleavage at the wider level which in turn helps to reduce ethnic cleavages. As Epstein (1958 : 240) points out in his classical study of the Zambian copperbelt : ".. .in situations involving the total field of Black-White relations the tribal factor tends to be overborne" (See also Gluckman 1961). Surprisingly, however, very little attention is given, especially in later studies, to this "total field" and little use is made of the class concept to explain the racial cleavage4. The focus is on ethnicity, the possible overlap in ethnic and class subsystems and the influence that this might have on other types of urban non-white social relationships. Modern writers give increasing importance to ethnicity and seem to support the plural theorists who emphasize the resurgence of ethnic competition in the development of new power relations during the process of independance (See, for example, Parkin 1969).

15An important recent work that uses the concept of ethnicity in relation to class is P.C. Lloyd's study of the Yoruba of West Africa. According to Lloyd (1974a and 1974b) the Yoruba define people in ethnic rather than class terms. Despite marked inequality in the society the Yoruba both accept and regard inequality as legitimate . Much evidence is provided by Lloyd to show that the Yoruba do not fulfill the stereotype of Western lower classes. Such an approach however suffers from too close a comparison between European class models and social processes in non-European countries. We tend to agree with Wallerstein (1973 : 378) that there are many different forms of class consciousness and class conflict in existence all of which are nevertheless expressions of class interests.

The emphasis on class

16Moving towards the other end of the continuum is the emphasis placed on class. The most important formulation of this approach is found in Marxists theories of class conflict. They provide an interesting contrast to theories proposed by some race analysts. Unlike the plural theorists, for example, the proponents of Marxist analysis emphasize the development of lower working-class solidarity irrespective of whether or not members of this group come from distinct ethnic backgrounds. This solidarity arises from the sharing of common interests in a class situation common to all (Kuper 1971 : 594-595 ; Cox 1948 : 321-352 ; Bauer 1966 : 150). Blacks and whites are primarily seen as members of the working and ruling classes respectively and not as members of distinct racial groups. On economic factors, however, Marxists and plural theorists differ only in their emphasis. Furthermore, the Marxian belief (see, for example, Lenin 1979 : 40) that the class struggle eventually takes on the form of a political struggle directed towards the assumption of political power by the proletariat makes the - Marxian approach almost compatible with the plural approach.

17The primacy given by Marxists to class is often argued on the basis that class factors preceded race in the process of colonial expansion. According to the Marxist view racial antagonism is a relatively late phenomenon. It is associated with the rise of capitalism (most of the leading capitalists are whites) and does not become significant until blacks attempt to assimilate Western economic values and culture. As colonialism develops race may become increasingly important because of its association with other variables such as class. Marxists believe that at the ideological level race functions as a mechanism of control, helping to perpetuate and justify the system of inequality. Such a system may continue even beyond the period of independence as colonial structures and mentality persist and blacks continue to accept white stereotypes of themselves as irresponsible and inferior.

18The foregoing argument provides Marxists with one explanation for the temporary absence of class-based interest groups and class consciousness in some colonial or ex-colonial societies. Contemporary Marxists concede that the industrial proletariat is, and will remain insignificant for a lengthy period as class hatred is often transferred into race hatred, and class factors are often constrained or masked by race aspects. In some cases where working class solidarity is absent Marxists attribute this to the growth of a labour aristocracy. For these reasons it may take some time for the realities of class interests and class consciousness to emerge (See, for example, Cox 1972 : 290 ; 292-295 ; Basham 1975 : 291).

19In brief, the concept of race in Marxist analysis is used primarily in class terms, that is, race is seen as a form of class or status relationship and race conflict as a variant of class conflict. As previously indicated, however, Marxists also concede that racial factors can be so pervasive as to blanket other interests for a lengthy period.

Summary and synthesis

20The pattern of social relations and organization in non-western societies is generally delineated on the basis of three types of differentiation--race, class, or ethnicity. The emphasis on race predominates among race analysts, ethnicity is emphasized by some Bristish social anthropologists, while the importance of class is argued most cogently by Marxist analysts.

21There is agreement among the above writers as to the importance of race throughout colonization or at least at some stage in the process. But while plural theorists focus on race relations almost solely in the context of competition between blacks and whites for political power other race analysts suggest that racial antagonism is largely based on the coincidence of race and class hierarchies. Some British social anthropologists also acknowledge the importance of race differentiation and the overlap of race and class but are often unwilling to use these terms or to examine the race-class relationship. Even Marxists recognize the importance of race relations at a later stage in colonization, and believe that race factors can become so pervasive as to mask the underlying class structure. However, Marxists attribute significance to race primarily because of its association with class. The Marxists view of race is almost completely subordinated to their view of class and the real basis of conflict is seen in class rather than racial terms.

22The concept of ethnicity is also given varied treatment by the above writers althought unlike the race concept there is a greater degree of disagreement about its significance. Plural theorists and some British social anthropologists treat ethnicity almost in the same way as they would approach the concept of race. They argue that with the development of new power alignments at the time of independence race decreases in importance while ethnicity becomes more important. The struggle for political power is now between members of the ethnically heterogeneous black population. According to this viewpoint ethnicity also hampers the development of working-class solidarity and imposes constraints on black interaction even though many of the latter may share the status of a suppressed class. Other race analysts take an opposite point of view by proposing that racial pervasivences can override ethnicity. The earlier studies of the Zambian copperbelt seem to lend credence to the latter viewpoint. From the Marxian perspective ethnicity is also insignificant as it is claimed that differences between the ruling and working classes can override all other types of differentiation. In light of the above it appears that the ambiguous role of ethnicity is still to be resolved.

23The virtual disregard of the importance of class strikes us as being a serious shortcoming in the approach used by some race analysts, especially the plural theorists. By ignoring the association between race and class which Marxists and most other analysts acknowledge, the plural theorists lose sight of an important factor that might help to explain the exacerbation of race or class antagonism. On the other hand, while Marxists and some of the other analysts acknowledge the importance of both race and class they make no attempt to clarify the relationship by combining race with class analysis. Furthermore, their approach is seriously weakened by the scant attention paid to class and/or ethnic differentiation among blacks. Similarly, the emphasis given to ethnicity by some contemporary Bristish social anthropologist and plural theorists precludes an analysis of the relationship between race and class, and ignores the possibility of overlap of race and class and the influence of this overlap on ethnicity.

24We believe the above shortcomings are the result of writers placing undue emphasis on one concept at the expense of others. Although such an approach helps to distinguish between concepts it can also lead to false dichotomies and distract from the true significance of each. It may also lead to ambiguity as when writers, althought agreeing on the importance of race in a colonial situation, differ in their treatment of it. What is required is a synthesis which gives equal theoretical emphasis to race, class, and ethnicity, and which directs attention to their specific relationships and significance over time in a particular geographical area.

25Drawing heavily from the insights and shortcomings of the above writers we propose to look at selected aspects of industrial relations in the South Pacific, particularly in Fiji and Bougainville, Papua New Guinea. Our major aim is to sketch the complex nature of industrial activity in the South Pacific so as to generate a number of propositions which take into account the interrelationship between race, class, and ethnicity. The arguments presented in the following section of this paper, however, must remain largely conjectural until such times as more evidence is available and thoroughly researched.


The early period

  • 5 Unless otherwise indicated the terms "labour", "workers" and "labour force" are used to refer to b (...)

26The history of labour5 in the South Pacific goes back at least to the 1850s when islanders were recruited for employment in cotton and sugar plantations in Australia, Hawaii, and Fiji. With the growth of a white capitalist economy the system of labour recruitement soon extended to other parts of the Pacific. For example, islanders were recruited for mining and plantation work in New Caledonia, phosphate quarrying in Nauru and Ocean Islands, and general plantation work in the Society Islands, the Solomons, Samoa, and Tahiti. The vast majority of Asians began arriving several decades later, mainly as unskilled workers (Oliver 1961 ; Parnaby 1964 ; Brookfield 1972).

  • 6 Much has been written abut the illtreatment andpoor wages accorded to South Pacific labour. See, o (...)

27Despite the low wages and abuses suffered by labour under this system6 organized workers' movements and racial and class consciousness among workers were slow to develop mainly because of instability in the labour force, the low volume of workers, and the development of ethnicity. These factors were in large measure due to, and reinforced by white colonial labour policies and the system of recruitement.

  • 7 In rural Papua New Guinea, for example, the administration mas reluctant to pursue a policy that m (...)

28Labour recruitement was primarily characterized by large-scale movements of labour under contract, sometimes to places several thousands of miles away from their rural homes. This developed out of the general unwillingness of the local population to work for low wages (Rowley 1958 : 546 ; Stanner 1953 : 139 ; Langmore 1973 : 186), the employers assumption that migrant labourers from depressed areas who were separated from kin ties and traditional custom would perform more efficiently (Farrell 1972 : 43), and the attempt by many colonial governments to inhibit, for a variety of reasons the growth of a permanent native labour force7. All of these factors accounted for a high turnover in labour common to most areas of the Pacific.

29The vast majority of islanders were employed seasonally, or on a two to three year contract. With the exception of Asian labour most returned home at the expiration of their contracts. The general shortage of labour enabled a significant number to move from job to job or into new occupations. Many general labourers in the towns, for example, were formerly employed in white plantations. The majority of Indo-Fijian contract workers in Fiji moved into independent cane-farming, while a smaller number gravitated to the towns to seek employment as labourers. The lack of a stable, permanent body of colonial workers in wage-employment thus helped to preclude for a lengthy period the development of working-class consciousness, and organized labour.

30The difficulties of organizing labour and developing class consciousness were increased by the fact that even in some of the major centres of employment the volume of labour in relation to the total population was never very large. Although colonial workers were concentrated in a few industries--plantation, mining, and the public service--they were recruited by a large variety of employers many of whom employed no more than a dozen or so workers at any one time (Parnaby 1972 : 138).

  • 8 Employers however did not hesitate to replace dissident workers despite the labour shortage.

31In the early period there were few devices for protecting the rights of workers. Most were content to depend on the goodwill of employers which was often forthcoming mainly because of the general labour shortage8. In Papua New Guinea, for example, employers were encouraged to maintain fair standards of living and working conditions which gradually came under the general supervision and control of colonial authorities (See, for example, Main 1948 : 117-121, and 144-149). The payment for services in kind (e.g., through housing, rations) meant that labour did not have to depend solely on a money wage for survival. Monetary wages were then considered a luxury and were not generally used for survival purposes or the purchase of basic necessities.

32Race and class consciousness were also precluded to some extent by the low proportion of whites in relation to the total workforce. For example, in June 1939, the earliest date for which figures are available for Papua New Guinea, whites and Niuginians comprised respectively 8 percent and 92 percent of the total workforce of 10,967 employed in the mining industry (Wilkes 1958 : 227). Pacific labour was restricted to employment in unskilled and semiskilled categories and so did not enter in competition with white labour. Distinct class lines did not materialize in the minds of the protagonists as contact between colonial labour and whites was limited, while labour and white management were socially and physically segregated at work and outside the workplace.

33The emphasis on employing migrant labour (New Hebrideans and Solomon Islanders in Fiji, Niuginians in the Solomons and Queensland, etc.) introduced complex problems of communication and cultural differentiation. Many were illiterate and unskilled making it more difficult for labour in the South Pacific to organize collectively than in other countries. The extent of ethnic differentiation increased when locals began to enter the workforce, generally at a much later date and on a more casual basis than migrant labour. Lacking the skills and work experience necessary for promotion most locals were relegated to menial work thus fanning the employer's belief that locally-recruited labour was inferior to imported labour. But while a number of semi-skilled positions were given to migrants from within or outside the area class differentiation among blacks did not figure very prominently since the majority were competing for the same jobs. Employers tended to measure performance standards in terms of ethnic background rather than in terms of class. This is not surprising since the lack of education and training facilities did not permit the growth of a class of skilled tradesman until the modern period (Danks 1956).

  • 9 Some of these features seem to apply to labour bn other colonial areas as well (See, for example, (...)

34From the worker's point of view ethnicity was becoming as much a major form of social differentiation as race. The concentration of a heterogeneous workforce for the first time and in such large numbers led to frequent fighting and inter-ethnic brawling (See, for example, Oliver 1961 : 321). Ethnicity was reinforced by the lack of permanent housing, the absence of family life, insecurity at the place of work, and the temporary nature of the labour force. To prevent inter-ethnic clashes local and migrant labour were separated, eventually leading to the development of ethnic enclaves9 where groups had greater ties to the rural home rather than to workmates and the place of work.

35Finally, to round off this discussion of the early period it may be fitting to indicate how whites and blacks saw their position in the society. To the best of our knowledge no systematic work in this field has ever been attempted, nor do we have much information on the concepts used by both groups to describe their place in the society. What little information we have suggests the predominance of race in the early stages of industrial relations. For example, whites tended for some of the reasons cited above to see their social separation from blacks in terms of race rather than class. The following reaction of a white property owner to increasing Indo-Fijian and Fijian migration to Suva at the beginning of this century is typical :

36There is ample room for dwellings of a better class for Indians and natives should, in course of time, a demand for such spring up in other parts of Suva without permitting the invasion of the red area which Europeans have created for themselves and which they must justly claim to have reserved in the future for their exclusive habitation (Mamak 1973 : 46).

37Furthermore, whites often used such terms as "nigger", "coloured", and "native" in the context of white superiority. Economic differentiation between whites and blacks was also based on race and not on economic factors. As Worsley (1957 : 484) has pointed out : "... the payment of low wages is linked to the white man's belief of his superiority".

38Blacks also tended to see their position in society in terms of race differentiation. While they were no doubt aware of economic deprivation they were not disposed to view exploitation and the economic elements of white-black relationships in class terms. Larson (1970 : 195-209), writing about modern Tikopia in white plantation employment, has provided a reason. The Tikopia, a latecomer to wage employment, uses a model for behaviour based on that of a classless society. "They sensed that the inequity of allocations stemmed largely from differences in skin color and basic capabilities and saw the plantation as a white man's enterprise, a somewhat complicated aspect of his culture which they could not wholly comprehend and would have to accept on management's terms". This view was widely encouraged by management's paternalistic but flexible attitude towards the Tikopia. Although Larson's study is concerned with the modern period, and the Tikopia differ from more experienced workers who view the conflict in the same plantation in both race and class terms, there is no reason to doubt that the Tikopia model for behaviour did not apply to equally inexperienced workers in other places in an earlier period. To see how labour's perception of the industrial system has changed over time it may be worthwhile to examine labour's response to economic differentiation.

The responses of labour

  • 10 The only other strike in Papua New Guinea that is comparable in scope to the Bougainville disturba (...)

39The responses on the part of South Pacific labour to ill-treatment, low wages, and poor working conditions have varied from place to place, and over time. But the outcome of these responses may be classified in terms of a scale of violence and non-violence. Protests accompanied by violence have been rare. Occasional references to killings and murders may be found in the early industrial record, but most examples of excessive violence are restricted to the contemporary period (for example, the recent disturbance in the Bougainville minesite, and the strike by urban Fijian and Indo-Fijian workers in Suva in 1959)10. At the other end of the scale and occurring with greater regularity throughout the history of colonial labour are the silent, less overt and more subtle forms of protest. Rather than complain or openly protest about wages and conditions of service most workers preferred to see through their period of contract, while others simply quit their jobs and moved back to their villages or into other jobs (See, for example, Rowley 1958 : 540-541 ; Wilkes 1958 : 238 ; and Mamak 1973 : 88). Falling within this context of non-violence are the work bans, token strikes, uncooperativeness, desertions, absenteeism and other types of sullen responses.

40A significant variant form of protest, lying somewhere in-between this scale may be seen in several of the Melanesian, so-called cargo cults and millenarian movements. Although these movements offered no violence they were generally feared by employers and the authorities who used repressive measures to counteract them. Rebellious workers, for example, were usually either killed, flogged, imprisoned, or deported.

  • 11 The close association between millenarianism and strike organization of African mineworkers is des (...)
  • 12 To stress the economic component of millenarian movements is not to deny the robe of ideas and rel (...)

41The economic basis of millenarian movements is evident in the literature although most writers tend to emphasize its anti-European nature, or draw distinctions between revivalistic and more secular movements (See, for example, Burridge 1954 : 253-254 ; Inglis 1957 : 261-262 ; and Willis 1970 : 23). On the basis of these studies one can easily get the impression of the predominance of racial hostility in millenarian movements. This is understandable because a major characteristic of these movements is the attempt to invoke the solidarity of its ethnically heterogeneous but black adherents in common struggle for equality with whites. However, it may also be correct to view the cults, or at least some of the later ones, in class terms. The Marching Rule movement in the Solomons (Wilson 1973 : 468 ; Worsley 1957 : 485, 1968 : 193) and the Mbula Tale in Fiji (Mamak 1973), for example, were characterized by an acute awareness of economic deprivation. At least three Fijian protest movements were formed in opposition not only to whites but also to several more privileged and powerful Fijian organizations (Mamak 1973). These movements were a collective reaction to things social, initiated almost exclusively for economic goals, and therefore may be regarded as fore-runners of contemporary workers' organizations11. The presence of class hostility is found even in the records of less secular movements. Many were led by ex-plantation labourers recently returned to their village (Mair 1948 : 68). While a major characteristic of millenarianism is the emphasis on "traditional" elements, a recent study has cautioned against interpreting the reversal to "traditionalism" as a retrograde step. These movements are a response to the new economic order and are an indication that modern economic values had been accepted (Sanford 1974 : 488 ; see also Worsley 1957 : 488)12.

42Although labour was seldom well organized in the early period, and overt class struggles were rare the record of industrial relations in the South Pacific indicates a long history of continuous reaction on the part of labour and the use of various forms of protest, whether accompanied by violence or not. While the response of labour brought little improvement in conditions this was not so much because of labour's passive acceptance of the economic position but because of strong control by employers and the colonial authorities--a control that was often expressed through some form of repression. Labour had nevertheless prepared the ground for actions of greater significance in the modern period.

The modern period

43In the modern period race, class, and ethnicity have taken on new dimensions due to the modification of several factors already described for the early period. The size of a permanent heterogeneous labour force has grown thus increasing the potential for ethnic conflict. Ethnicity tends to be exacerbated when it overlaps the growth of occupational segmentation and class differences among blacks. In most cases however the major occupational categories are ethnically mixed and the new work setting helps to make ethnicity irrelevant. In time ethnicity tends to be superceded by the new class division which in some cases has provided the bases for conflict. In general however class differences among blacks have not yet become predominant for the following reasons : the lack of correlation between ethnicity and class sub-systems ; the small proportion of upper class blacks in relation to the total labour force ; and common interests among blacks arising from the overlap of race and class. While the correlation between race and class was evident in the early period it has become more perceptible today partly because of the presence of a large number of whites in some industries. The complex relationship between race, class, and ethnicity in the modern period is mirrored to a large extent in the growth of trade unionism to be described in detail later.

44Since World War II the size of permanent class of wage-earners has increased steadily due in part to any one or more of the following factors--population growth, the forces of industrialization, the growing desire for a money income, and the gradual disappearance in some areas of land suitable for subsistence production.

  • 13 In 1953 almost 80 percent of the Papua New Guinea labour force mas employed on a casual basis or u (...)

45In most parts of the South Pacific the above trends are associated with increasing rates of urbanization largely because of greater opportunities for deriving a cash income in the towns. In Fiji, for example, the proportion of economically active males occupied in rural industries (mostly Indo-Fijians and Fijians) is gradually decreasing while the volume of migratory movements to Suva and other urban centres which have been in existence for some time is increasing. Permanent Bougainvillean participation in the urban economy also appears to be steadily increasing since commencement of the copper project although there may be significant regional variations to this trend. For example, Bougainvillean employees whose rural homes are close to the mine site seem to be the most unstable group of workers in the company. Local conditions and personal factors combined with easy access to the rural home have so far discouraged these Bougainvilleans from considering anything other than short term employment in the copper mine. Such instability in the labour force is not uncommon in the early stages of urban develoment, however13. As these centres become more established, and with new employer policies which favour the retention of a stable labour force (e.g., the provision of married accomodation for workers and their families) the current instability among local Bougainvillean employees may eventually be reversed.

46In contrast to the general increase in the size of a permanent labour force the proportion of wage-earners in relation to the total population in almost all areas of the South Pacific remains insignificant. This is largely due to the so far low level of industrialization in the South Pacific as compared to many other areas of the world (Report of the 1st conference of South Pacific labour ministers 1974). In the late 1960s only 7 percent of the Papua New Guinea population were engaged in wage labour (Langmore 1969 : 11). In Bougainville the wage-earning workforce actually declined significantly following the completion of the mine construction phase. Another factor which has slowed down the growth of wage-employment in the towns has been the payment of low wages--a trend that has continued from the early period. It is remarkable that only two decades ago the highest paid Niuginian government employee received a cash wage of only A$6 a week (Langmore 1973 : 187). Although wage rates have increased steadily in recent years the majority of workers are still employed at the minimum wage level. One reason why some local Bougainvilleans are only peripherally committed to wage employment is because of their dissatisfaction with the low wages and conditions of service in the mine. In light of the above, it is not surprising that industrial relations and organized labour remain insignificant issues for the majority of the population.

47Growth in size of the wage-earning workforce in comparison to the early period has also increased the potential for ethnic conflict. In Papua New Guinea more than half of the labour force are migrant workers employed from outside their District of origin (Langmore 1969 : 1). Local opposition to mining, coupled with a general unwillingness by local Bougainvilleans to accept anything other than very short term casual employment compelled the mining company initially to look elsewhere for labour. Later, as information (frequently exaggerated) on wages and working conditions offered by the company began to diffuse widely an increasing number of Bougainvilleans and Niuginians were attracted to the mine site in search of jobs. (Bedford and Mamak 1975). Growing competition for jobs and a concentration of large numbers of outsiders in close association with Bougainvilleans for the first time were to produce inevitable ethnic tensions. This led to the emergence of voluntary organizations which recruited members on the basis of ethnicity. In the early stages of mining operations, for example, Bougainvilleans began to organize in such a way as to take advantage of new opportunities, and to protect commonly-threatened interests as a distinct and separate unit. In 1973, a common front for all Bougainvillean workers in the towns was developed and eventually a committee under the name of Panguan Mungkas Association (PMA) was formed. The association was founded on the ethnic distinctiveness of Bougainvillean workers in contradistinction to other ethnic groups employed by the company. Exactly a year later, however, the influence of ethnicity appears to be waning and the association no longer seems to be as active as it was in the past. There are several reasons for this.

48Firstly, the economic interests of Bougainvillean members in the association have become more difficult to integrate as many have gained job promotions within the company. Technical and skilled workers, many of whom are staff employees, enjoy higher incomes and living standards than their compatriots who are employed as labourers and manual workers. In contrast to the distinction between staff and wage employees in the company Bougainvilleans employed by other companies and the government are not so well off in terms of pay, social amenities, and working conditions as Bougainvilleans employed by the mining company. Secondly, the association has suffered from too close competition with non-ethnic associations such as the mine workers' union and the staff association. Bougainvillean staff employees are members of the staff association while most wage employees are members of the mine workers' union. Thirdly, with the advancement of several of the association's executives in the company Bougainvillean consciousness and identity have largely been subordinated to individual and company interests. Finally, the company has responded to Bougainvillean pressure to increase the Bougainvillean component in the labour force thus making the PMA's existence unnecessary (Mamak and Bedford 1974b : 13-17).

49Socio-economic differences which have developed among black workers in Bougainville are not closely associated with ethnicity. In the mining company, for example, there is a balanced distribution of employees from various Districts in both staff and wage categories of the labour force. The new division seems to be based solely on occupational segmentation. There is a significant income gap between black staff and wage employees. In addition to receiving a higher income staff members enjoy such quasimonetary benefits as educational allowances for their children, a better class of accomodation, and superannuation benefits. They also have a better comprehension of the English language and a higher standard of education than most wage employees.

50Some minor occupational categories, however, are ethnically differentiated and in such cases ethnicity tends to be exacerbated. This is a pattern common to all areas of the South Pacific (e.g. ethnic conflict between Tolais and Bougainvilleans in Bougainville ; Highlanders and coastal Papuans in Port Moresby ; Rotumans, Part-Europeans, and Fijians in Fiji ; Solomon Islanders and Chinese in Nauru). As in the earlier period white, employer-created stereotypes tend also to have a negative effect on relations between workers of distinct ethnic backgrounds, especially when employers use ethnic criteria for promotions and evaluating job performance.

51In the main most occupational classifications are ethnically mixed and in such cases ethnicity becomes largely irrelevant. In the Bougainville mining company and in other industries where occupational segmentation does not largely correlate with ethnicity the work setting can provide for the development of common aims which transcend ethnic interests. For example, in Suva, one of the major urban centres in the South Pacific, Indo-Fijians and Fijians who share similar economic and class interests often express egalitarian and friendly attitudes towards each other. There is a similar tendency for mine workers in Bougainville to accept the work group as an important reference group.

52There are two other reasons (besides the lack of correlation between ethnicity and occupational segmentation, and the influence of the work setting) which preclude the development of major cleavages between members of the black workforce. Firstly, the proportion of blacks in higher occupational categories is small in comparison to the proportion of employees in low occupational categories. (This trend is likely to continue for some time given the slow pace of localization. In Bougainville, for example, there were only 97 blacks in staff positions compared with 2,845 employed in wage categories in 1973. Exactly a year later the number of black staff positions has increased by only eighteen). Secondly, class differences among blacks are to a large extent counterbalanced by more significant racial and class divisions between whites and blacks. Whites occupy most of the skilled and professional categories while the vast majority of workers in the South Pacific remain in unskilled and semiskilled positions (For Papua New Guinea see Danks 1956, and Langmore 1969 : 1-2).

  • 14 In contrast to urban industries the pattern of rural occupational activities is marked by ethnic s (...)

53The overlap of race and class is very distinct in Fiji and Bougainville. In the urban-industrial sector of the Fiji economy the great majority of Fijians and Indo-Fijians share low occupational status in contrast to whites who are found in professional and top management positions. Most Fijians and Indo-Fijian unskilled workers are engaged as labourers in local and national government projects, sugar processing, and building and road construction. At the semi-skilled level the following occupations are shared : electrician, carpenter, and the like on government projects and in construction, although there is some evidence of ethnic distribution in the tourist industry and commercial fields (Mamak 1973)14.

  • 15 This pattern of segregation is world-wide and is not limited to the Bougainville copper mining ind (...)
  • 16 For a discussion of the widening economic gap between whites and blacks in PaPua New Guinea see La (...)

54In the Bougainville mining community there are two forms of agreement governing employment and which cut across racial boundaries. The mine workforce is made up of employees who receive fortnightly wages and employees who are recruited as staff members under an annual salary agreement. But within each of these categories wide differences exist between whites and blacks in both pay and conditions of service15. Although black wage rates have increased steadily in recent years the gap between black and white earnings has also increased and is likely to continue on the basis of experience in other parts of the country and elsewhere16. In 1973 the estimated average annual earnings of a black wage employee was A$33.16 per week, or approximately 15 percent of the estimated average earnings of a white wage employee. In that same year the estimated average annual earnings of black staff employees was approximately 27 percent of the estimated average annual earnings of white staff (Bedford and Mamak, in preparation).

55The overlap of race and class has become more noticeable in the modern period partly as the result of a new development--the increase in the size of the white workforce in some industries like mining. In Nauru, for example, over 60 percent of the workforce today is expatriate. In Bougainville the proportion of the white component to the total labour force is comparatively high (25 percent). This factor combined with job promotions for blacks has placed the latter in very close contact with whites for the first time. Under these conditions, how do workers define this new type of relationship ?

56A study of working relationships recently conducted by Mamak (1973) in Suva, Fiji shows that most Indo-Fijian and Fijian workers irrespective of their occupational or class status ranked whites as the most difficult to associate with, and both judged whites in much the same way. No outright opposition to physical or cultural differences such as religion, language, and diet were expressed by either Indo-Fijians or Fijians towards each other, althought colour and class distinctions were sometimes made between themselves and whites.

57Black-white working relationships in the Bougainville mine are more cordial in comparison to the Fiji case, and most of our Bougainvillean and Niuginian informants got along very well with whites in the work place. There are rules governing working relationships which are explicitly stated in the contract of white workers and which provide for immediate termination if negative attitudes towards blacks are maintained or fostered. These rules are accepted by most whites and are regarded as defining appropriate behaviour in the work place. Nevertheless, as in Fiji, most whites regard all blacks as subordinates but react more favourably to blacks in high occupational categories than those in semi-skilled and unskilled positions. Ironically, however, upper class blacks who are in closer contact with whites are also more likely than lower-income blacks to express anti-white sentiments and to compare their earnings and working conditions with whites. The distinction made by black staff in the copper mine is one of class and race rather than class or race alone. Thus, despite a policy of integration close personal contact between whites and blacks is limited and it is rare for members of both groups to associate with each other outside the place of work. This may be due to the fact that in Bougainville as in most other parts of the South Pacific the frame of reference for interaction between blacks and whites is becoming more perceptively one of economic inequality based on race.

58In brief, while occupational segmentation is a new division in the modern period and has helped to preclude working-class solidarity among blacks, the overlap of race and class has also become more distinct, especially in some industries which employ a large white workforce, and has helped to foster a close congruity of interest between black workers irrespective of their ethnic background, occupational or economic status. This complex relationship is also illustrated by the growth of trade unionism in the South Pacific.

The development of trade unionism

59The new norms and values which are imparted in the modern period are most clearly reflected in voluntary organizations. In the industrial field, the development and characteristics of trade unions can provide a clear indicator of the emergence of a social class--a collectivity with common purpose interests which believes these interests can best be served by co-ordinated effort.

60In general, trade unions did not originate directly from outside pressures but emerged spontaneously out of the common needs and interests of South Pacific labour. It has previously been suggested that some of the early protest movements can be conceived as forerunners of contemporary workers' organizations. Although the earlier activities of labour did not always end in victory they were nevertheless effective in making employers increasingly aware of the worker's organizational capacity. Speaking of the 1929 strike in Rabaul (Papua New Guinea), for instance, Healey (1968 : 33) notes : "It would not be too much to say that this incident had terrified the entire European community in the Territory, for it removed the long held illusion that the natives were too stupid and illiterate to be able to combine". Further disturbances in Papua New Guinea prompted the colonial authorities to set up labour departments and introduce new labour legislation and machinery for the regulation of trade unions. Similarly, a Department of Labour was established in Fiji following a wave of industrial unrest dating back to the early part of this century. This opened the way for the formation of unions covering a wide range of activities. Today, approximately two-thirds of Fiji's labour force is active in the trade union movement.

61One of the major forces stimulating the development of an organized base for urban labour has been industrialization. While a number of small unions existed in Bougainville before copper mining operations began the prospects of unionism increased dramatically as a response to rapid industrialization associated with the mining industry. In a few years the Bougainville mine workers' union, formed in 1969, succeeded in establishing itself as the largest union in the District, and one of the largest in the country. Spurred by the presence of a large urban labour force dependent solely on a monetary wage for survival and seeking to better their wages and working conditions the union has been able to gain significant recognition from workers and the company--an achievement which normally takes a decade or more in other parts of the country and the South Pacific where industrialization is relatively stagnant .

62Yet the development of an effective and viable trade union movement has been fraught with difficulties. The mine workers' union, for example, was slow in gaining the support of wage-earners in the early years of the mining operations due in part to the largely transient nature of the labour force. An indication of the importance of stability for increasing union membership is provided by the results of one survey which shows that mine workers who have completed two or more years of work are more likely to join the union that those who have not (Mamak and Bedford, in preparation).

63Other problems associated with the organization of workers is the lack of experience and financial resources of the trade unions ; shortage of full-time officials, the anti-union attitude of many employers ; and ethnic and occupational differentiation among the workforce.

  • 17 A similar case illustrating differences between Polynesian and more experienced Melanesian workers (...)

64Organizational problems based on ethnicity are introduced when members of a single ethnic group are primarily seasonal workers who lack committment to wage-employment. In Bougainville, for example, many local Bougainvilleans see an alternative avenue to mobility and self-betternment in leaving wage employment. Consequently they are not much interested in trade unionism as are the majority of workers recruited from outside the District who are more permanently committed to urban wage employment17. Thus, ethnicity introduced several organizational difficulties for the union in its formative stages (e.g., the need to satisfy the demands of a heterogeneous workforce ; the problem of selecting union leaders ; competition with other organizations formed on the basis of ethnicity). Ethnicity however did not prove as divisive in the Bougainville trade union movements as it did in other places for reasons to be discussed below.

65With more Bougainvilleans recruited into the project in recent years all ethnic groups are proportionately represented in the mineworkers' union. Ethnicity is no longer so predominant in the activities of the union because no ethnic group is large enough to dominate the affairs of the union. As a result neither membership nor squabbles among union officials and members are ethnic based. Union leaders are selected on the basis of their ability to articulate the grievances of the members and not because of their ethnic membership. Care is also taken to see that each department in the company is represented in the union executive.

66Nowadays much of the stimulus for the development of non-ethnic unions has come from employers themselves. In large-scale enterprises such as copper mining the employer is anxious to reduce the disruptive effects of ethnicity which can lead to great financial loss for the company. To prevent disharmony and promote simple but effective channels of communication between management and a large heterogeneous labour force encouragement is given to a policy of integration. Nationalist governments, who also happen to be the largest employer of a country's labour force, are anxious to promote stability and discourage citizens and workers from organizing on the basis of ethnicity. Such was not the case in the initial stages of at least one trade union movement, however.

67In Fiji, the trade union movement started off as a predominantly Indo-Fijian activity with the focus of organizational activity in the sugar industry. But the leaders of the movement soon realized that if organized labour was to function effectively it had to admit all sections of the depressed class to its membership. Union leaders therefore attempted to spread their influence by acting as self-proclaimed spokesmen of the working class. The colonial government and other employers, however, became alarmed at the possibility of working class solidarity cutting across ethnic boundaries and introduced various obstacles to prevent labour from combining. Encouragement was often given to the labour of one section to breakaway from a militant trade union. (Even in some all-Fijian unions provincial committees were used to undermine the influence of the union). Although many unions formed on the basis of ethnic exclusiveness were short-lived the tendency for dual unions persisted for a lengthy period and was reinforced by the division between Indo-Fijians at the political level. Notwithstanding, overt antagonism between ethnic political parties and between ethnic unions was dampened by friction within each organization as the result of leadership struggles and personality clashes. In time, as Indo-Fijians and Fijians in low status positions became more concentrated they began to shed their ethnic separateness and worked together for the betternment of their lot. In 1959, Indo-Fijians and Fijians jointly struck work in Suva in protest against white monopoly of the economy. The trend persisted into the 1970s and today there is little evidence of ethnic exclusiveness in Fiji's trade unions.

68In view of what has been said about the development of class differences and occupational segmentation among blacks it may not be too much to add that class may be a more significant factor than ethnicity in relations between black workers. In the South Pacific, as in most other developing areas, the majority of wage-earners are employed by governments. During the colonial period governments were concerned that the development of a united labour movement could act against them. Today nationalist governments are concerned with the impact of rising wages on the national economy. Workers are encouraged to reduce their economic demands in the interest of nation-building. Although many ex-trade union officials have been recruited into government (e.g., Nauru, Fiji, Papua New Guinea) workers in general feel that these ruling elites have lost touch with the worker's demands. It seems likely therefore that governmental intervention in the trade union movement, and conflict between the new political class and workers will increase in the future.

69A similar pattern of conflict occurring between members of various occupational categories is also becoming evident. The perception of class boundaries between semi-skilled and unskilled workers on the one hand, and skilled and clerical staff on the other are reflected in different rates of participation in trade union activity, membership, and attitudes towards unionism. In Fiji, for example, a large proportion of low-status employees as compared to high and middle status employees are members of trade unions. This is not surprising since trade unions act primarily on behalf of low-income workers. Upper class blacks are more likely to see themselves as part of management, they are more oriented towards occupational advancement, and are not anxious to involve themselves in industrial disputes. On the other hand, most semi-skilled and unskilled workers, irrespective of their ethnic background, share attitudes which suggest the development of working-class consciousness. They express the desire for more intensive involvement in trade unions, a preference for collective rather than individual action, and the need for a strong negotiating body to represent their interests. At the organizational level class-based conflict is beginning to occur between manual workers' organizations and professional type associations, both of which are ethnically-mixed.

70The perception of class boudaries is also beginning to appear in Bougainville. As previously mentioned the basic division among black mineworkers is between staff and wage employees. Since staff members are in closer contact with management they are also more exposed to the paternalistic and anti-union attitude of the employer, and are therefore discouraged from participating in industrial activities. Staff are not permitted to join the mine workers' union, nor to engage in industrial disputes. On the other hand, a significant factor precluding the divisiveness of occupational segmentation and class differences among blacks is the close congruity of interest existing between the trade union and staff members--an interest which seems to have developed out of the growing awareness of greater disparity between black and white earnings.

71An indication of this interest is that in 1974 the union attempted (albeit unsuccessfully due to company pressure) to recruit black staff into the union. Another factor to note is the aim of union officials, some of whom are staff members, to lift the wages of lower income workers so as to reduce existing income differentials.

  • 18 For a description of a parallel situation in the Bulolo goldfleld (Papua New Guinea)in the 1930s s (...)

72By way of contrast, no real interchange has ever existed or exists between white and black workers in the South Pacific, nor are there any common associations for the two groups (See, for example, Report of the 1st conference of South Pacific labour ministers, 1974). This is partly due to the influence of employers. In Bougainville, the mining company is fearful of expatriate influence in the local trade union movement. The subject of union membership is carefully avoided in recruitment and no encouragement is given to whites to join the mine workers' union. The system of dual wages is another factor precluding the development of common interests between white and black employees. For example, none of the three disputes (out of a total of 50 disputes recorded in 1970) in which Niuginian and white employees jointly took part involved a claim for higher wages. Finally, because of their higher earnings and economic status whites tend to regard themselves and are regarded by blacks as members of a privileged group18.

73In brief, while occupational segmentation and class divisions among blacks are a new development in the modern period and have in some cases provided the bases for conflict, there is a tendency for these divisions to be blurred by the persisting overlap of race and class.

Summary and conclusions

74The major aims of this paper have been to indicate the complexities of race, class, and ethnicity in the analysis of industrial relations in the South Pacific, and to suggest the need to examine the relationship between all three patterns of social differentiation over time.

75While there is general agreement in the literature on the importance of race factors and the close association between race and class, to date no attempt has been made to examine the relationship in detail nor to take into account both types of differentiation simultaneously. In this paper we have attempted to show the significance of race in both early and modern periods of industrial relations in the South Pacific, and the close association between race and class. For example, many of the factors (labour policies, stereotypes, paternalism, etc.) which have made the development of class consciousness problematic are racial. On the other hand, the persisting overlap of race and class has exacerbated racial and class antagonisms between blacks and whites, and helps to preclude conflict between blacks on the basis of ethnic and class differentiation. It is therefore clear that any attempt to separate race factors from aspects of class and ethnic relations will fail to explain the significance of either.

76The view of plural theorists that ethnicity precludes lower working-class solidarity is supported only when ethnic and class sub-systems coincide. On the other hand, the Marxist view that racial pervasiveness tends to override ethnicity is given much greater support by data from the South Pacific. In the South Pacific the major occupational categories are ethnically mixed, and the work setting combined with common interests help reduce the salience of ethnicity. Ethnicity in the South Pacific has not reached the significance that it has in other developing countries where the influence of whites is declining, and where ethnic and political and class sub-systems coincide.

77The data that emerges from our discussion of race, class, and ethnicity in the South Pacific lead us to suggest the following propositions as a basis for further research.

78The extent of ethnicity, racial and class antagonisms will depend on a number of factors-- the numerical size of the protagonists, size and stability of the workforce, degree of social control, extent of repression, whether or not major occupational categories are ethnically mixed, etc. These factors may of course be related directly or indirectly to any one of the three types of differentiation.






  • * Ethnic/Class tensions if high status group is ethnically homogeneous ; class tensions of group is (...)
  • If group is ethnically heterogeneous ethnic tensions due to job competition are likely to develop (...)

Fig. 4 CLASS
Heavy lines denote type of cleavage
Note **
Note **

79Race differentiation is predominant throughout the colonial period. The racial cleavage will predominate over all other types of cleavages especially when whites regard and treat all blacks as members of a single inferior race irrespective of the latter's diverse ethnic or class black-grounds, and when blacks see themselves as members of a single, low-ranking group.

80As some blacks begin to experience upward mobility or become more committed to wage-employment they will also tend to become increasingly aware of class interests and will provide the leadership for change. There is some ambiguity in this situation, however, and race will tend to be confused with class. Hostility will be couched in both racial and class terms, and recruitement will be based on both colour and class (Fig. 1).

  • 19 Ethnicity may be emphasized by whites as a diversionary tactic, i.e., to prevent antagonism from b (...)

81On the other hand, the significance of race and black-white class differences may be temporarily reduced by the expansion of black economic differentiation and an increase in the size of a permanent, ethnically heterogeneous black workforce competing for the same jobs (Fig. 2). Ethnicity will tend to be exacerbated when ethnic and class sub-systems overlap and expand, i.e., when ethnic groups become increasingly divided into distinct classes. If blacks who experience occupational mobility are themselves ethnically differentiated, i.e., ethnic and class sub-systems do not coincide (as seems to be the case in the South Pacific), then class divisions among blacks (a situation some employers tend to exploit) will partly overshadow racial antagonism and will counterbalance the potential divisiveness of ethnicity and race19.

82Over time the persisting overlap of racial and class boundaries becomes increasingly more important than ethnicity or class differentiation among blacks, particularly if the upper and lower occupational categories are ethnically mixed and the number of blacks in upper status positions in relation to the total labour force is small. From the point of view of upper class blacks differences in white-black incomes are no longer to be explained in terms of education, skill, etc., but in terms of discriminatory policies based on race. It is also obviously less difficult for them, if they are so inclined, to recruit support from the lower-class on the basis of race. On the other hand race continues to be confused with class because whites tend to become more concerned about the stigma of race prejudice and justify separation from blacks on the basis of class differences. They also regard blacks in the upper class on more equal terms than whose in the lower class (Fig. 3).

83Upper and lower class blacks will join in common struggle against whites until such time when the economic gap between upper class blacks and whites is significantly reduced or when the former constitute a ruling class or labour aristocracy. The growing division between manual workers and a professional class of workers, or between labour and the ruling elites on the basis of class affiliation perhaps signal the next stage of development in industrial relations. As the number of blacks experiencing occupational mobility increases antagonism will be couched almost solely in class terms. The correspondance between race and class becomes increasingly blurred and rank i.s not so much determined on the basis of colour. Whites also begin to recognize the extent of economic differentiation among blacks and no longer regard or treat all blacks as if they were members of a single class or race. Irrespective of whether the conflict is now between blacks or continues between blacks and whites it will be based almost solely on class (Fig. 4).

84The above propositions should be regarded as tentative. More studies of how the protagonists view their place in society and particular conflict situations are needed if we are to determine precisely which type of differentiation is more important and why. Futhermore, although many examples were provided in this paper, it is necessary to specify the relationship between various levels of race, class, and ethnicity.

85In conclusion we may note that the pattern of industrial relations in' the South Pacific today lies somewhere in-between the models represented in Figures 3 and 4. In most cases so far multiple differentiations have prevented antagonism from being directed solely toward any one group. But the trend in Bougainville seems likely to emerge in other parts of the South Pacific now that mining is becoming an important area of activity, and to persist given the cummulative effect of intensive capitalism, continuing inequality between whites and blacks, inflation and the rising cost of living, and a growing permanent class of low-income workers dependent solely on wages for survival. We can look to an intensification of industrial unease based on the old pattern of race-class conflict.



AMIN, Samir. "The class struggle in Africa". Revolution 1, 1964.

BASHAM, Richard. "On the caste system upside down”, Current Anthropology, vol. 16, no. 2, June 1975.

BAUER, Peter T. "Marxism and the underdeveloped countries”, Marxist ideology in the contemporary world--its appeal and paradoxes. M.M. Drachkovitch, ed., New York : Frederik A. Praeger, 1966.

BEDFORD, Richard and Alexander MAMAK. "Bougainvilleans in urban wage employment : some aspects of migrant flows and adaptive strategies", Oceania, 1975 (forthcoming).

--New Towns on Bougainville : copper mining and urban-industrial development in a Melanesian society, (in preparation).

BROOKFIELD, H.C. Colonialism development and independence. London : Cambridge University Press, 1972.

BULMER, M.I.A. "Sociological models of the mining community", Sociological Review, vol. 23, no. 1, February 1975.

BURRIDGE, K.O.L. "Cargo cult activity in Tangu", Oceania, vol. XXIV, no. 4, 1954.

CASANOVA, P.G. "Internal colonialism and national development studies", Comparative international Development, vol. 1, no. 4, 1965.

COX, O. Caste, class and race : a study in social dynamics. New York : Doubleday and Co., 1948.

--"Race, pluralism, and power in the West indies, New Community, vol. 1, no. 4, 1972.

CROCOMBE, Ron. "Anthropologists and other missionaries", a paper presented to the Association for Social Anthropology in Oceania, 26-30 March, 1975.

DANKS, K.H. "Industrial activity in selected areas of the South Pacific", SPC Technical paper, no. 90, 1956.

EPSTEIN, A.L. Politics in an urban African community. London : Manchester University Press, 1958.

FARRELL, Bryan H. "The alien and the land in Oceania", Man in the Pacific Islands, R. Gerard Ward, ed., London : Oxford University Press, 1972.

GLUCKMAN, M. "Anthropological problems arising from the African industrial revolution", Social change in modern Africa, A. Southall, ed., London : Oxford University Press, 1961.

GUTKIND, Peter C.W. Urban anthropology : perspectives in 'Third World' urbanization and urbanism. The Netherlands : Van Gorcum, 1974.

HEALEY, A.M. "Labour and combination in New Guinea to 1942", Labour in the goldfields. Canberra : Australian Society for the study of labour history, 1968.

HLOPHE, Stephen. "The significance of Barth and Geertz' model of ethnicity in the analysis of nationalism in Liberia", Canadian Journal of African Studies, vol. 5, no. 2, 1973.

INGLIS, Judy. "Cargo cults and the problem of explanation", Oceania, vol. XXVII, no.

4, 1957.

KUPER, Leo. "Political change in plural societies : problems in racial pluralism", International Social Science Journal, vol. 23, no. 4, 1971.

LANGMORE, J. "Wage determination in Papua New Guinea", a paper presented to the La Trobe University seminar, May 31, 1969.

--"Public service pay and localization policy", Alternative strategies for Papua New Guinea. Anthony Clunies Ross and John Langmore, eds., London : Oxford University Press, 1973.

LARSON, Eric H. "Neo-colonialism in Oceania : Tikopia plantation labor and company management relations", Oceania, vol. 40, no. 3, March 1970.

LENIN, V.I. On Marxism. Moscow : Novostin Press, 1970.

LLOYD, P.C. Power and independence. London : Routledge and Kegan Paul, 1974a.

--"Class consciousness and the Third World", New Society, November 1974b.

MAGUBANE, B. "The 'Xhosa' in town, revisited urban social anthropology : a failure of method and theory", American Anthropologist, 75, 1973.

MAIR, Lucy. Australia in New Guinea. London : Christophers, 1948.

MAMAK, Alexander. Pluralism and social change in Suva city, Fiji. Michigan : Xerox University Microfilms, 1973.

--and Richard BEDFORD. "Bougainville's students : some expressed feelings towards non-Bougainvilleans, Arawa town, and the copper mining company", New Guinea, vol.

9, no. 1, 1974a.

--and Richard BEDFORD (with Leo Hannett and Moses Havini). Bougainvillean nationalism : aspects of unity and discord. Canterbury : Bougainville Special Publications no. 1, University of Canterbury, 1974b.

--and Richard BEDFORD. Bougainville copper and trade unionism. Canterbury : Bougainville Special Publications no. 4, University of Canterbury, (in preparation).

MEEBELO, Henry S. Reaction to colonialism. London : Manchester University Press, 1971.

OLIVER, Douglas L. The Pacific Islands. New York : The Natural History Library, 1961.

OMVEDT, Gail. "Towards a theory of colonialism", The insurgent Sociologist, vol. 3, n° 3, Spring, 1973.

PARKIN, David. Neighbours and nationals in an African city ward. Berkeley and Los Angeles : University of California Press, 1969.

PARNABY, O.W. Britain and the labour trade in the South Pacific. Durban, N.C. : Duke University Press, 1964.

--"The labour trade", Man in the Pacific Islands, R. Gerard Ward, ed., London : Oxford University Press, 1972.

Report of the 1st conference of South Pacific labour ministers. Canberra : Australian Government Publishing Service, 1974.

RODNEY, Walter. How Europe underdeveloped Africa. London : Bogle-L'ouverture Publications, 1973.

ROWLEY, C. "Labour administration in Papua and New Guinea", South Pacific. March-April, 1958.

SANFORD, Margaret. "Revitilization movements as indicators of completed acculturation", Comparative studies in society and history, vol. 16, no. 4, 1974.

SCARR, D. "Recruits and recruiters : a portrait of the Pacific Islands labour trade", Journal of Pacific History, 2, 1967.

STANNER, W.H.E. The south seas in transition. Sydney : The Australasian Publishing Company, 1953.

WALLERSTEIN, Immanuel. "Class and class-conflict in contemporary Africa", Canadian Journal of African Studies, vol. 7, no. 3, 1973.

WILKES, John. Australia and New Guinea. Sydney : Angus and Robertson, 1958.

WILLIS, Ian. "Rabaul 1929", New Guinea, September-October, 1970.

WILSON, Bryan R. Magic and the millenium : a social study of religions movements of protest among tribal and third world peoples. London : Heinemann, 1973.

WILSON, Francis. Labour in the South African gold mines 1911-1969. London : Cambridge University Press, 1972.

WORSLEY, P.M. "Millenarian movements in Melanesia", South Pacific, September-October, 1957.

--The trumpet shall sound : a study of "cargo" cults in Melanesia. New York : Schocken Books, 1968.


1 The fieldwork conducted by the senior author in Fiji between 1970-72 was made possible by a grant (no. MH49744-01) from the National Institute of Mental Health ; financial assistance provided by the University of Hawaii's Research Corporation to both authors is gratefully acknowledged for field research on Bougainville over the period 1973-74. A modified version of this paper was presented to a symposium on "The Impact of Cultural Exchanges" in the 13th PACIFIC SCIENCE CONGRESS, Vancouver, B.C., 18-29 August, 1975.

2 According to our usage of the race concept relations between members of a non-white population mho come from distinct racial backgrounds (e.g. Indo-Fijians and Fijians) are not race relations because the crucial defining characteristics when made are usually based on cultural criteria and not on colour. An exception to this rube is found in Bougainville where one of the most distinct issues underlying relations between Bougainvilleans and mainland Niuginians is skin colour. Nevertheless, this distinctiveness is also frequently associated with perceived cultural differences (Mamak and Bedford 1974a : 4-9).

3 This is contrary to the belief of other race analysts mho argue that some classes are "missing" in non-western societies such as Africa (See, for exemple, Wallerstein 1973 : 377).

4 The lack of attention paid to the social system of whites and the interaction between blacks and whites in Africa has been criticized by Magubane (1973). For a similar criticism mode in the South Pacific context see Crocombe (1975). Two studies which attempt to fill this gap are : Bedford and Mamak (in preparation) and Mamak (1973).
With reference to the paucity of class analysis bn comparison to studies of ethnicity in Africa see Peter C.W. Gutkind (1974 : 175) mho notes that :
To date we have very few studies which indicate precisely how heterogeneity is manifest in urban life ; how social classes are formed ; whether groups rebate to one another according to ethnic criteria or socio-economic position in the urban system. Because most studies have ignored the conditions and processes of change and modernization, ethnicity has become the focus for analysis, treating it (falsely) os the independant variable. The class factor also remains undefined in the South Pacific context and has largely been ignored to date. Previous accounts of social disorder in the South Pacific tended to focus subjectively on race (see, for example, the early issues of New Guinea and Pacific Islands Monthly). Such an approach precluded an analysis of the economic factors underlying the conflict, the class structure that mas emerging, and the position of race and ethnicity in that structure. Modern studies of the South Pacific have not reversed this trend as emphasis is now placed on such concepts as nation-building and micronationalism. For a similar criticism applied to Africa see Wallerstein (1973).

5 Unless otherwise indicated the terms "labour", "workers" and "labour force" are used to refer to both rural and urban black workers. Similarities in their status and conditions have been noted by Omvedt (1972 : 13) and Bulmer (1975 : 61).

6 Much has been written abut the illtreatment andpoor wages accorded to South Pacific labour. See, on this point, Parnaby (1964 and 1972) ; Scarr (1967) ; and Oliver (1961).

7 In rural Papua New Guinea, for example, the administration mas reluctant to pursue a policy that might "convert peasant proprietors into a landless proletariat" (Healey 1968 : 24-25). The government's motives for discouraging the growth of a permanent urban workforce included the expressed ideal of preserving the traditional may of life, and the fear that a settled native population in the towns would bead to unemployment conditions. An underlying assumption of these policies mas that non-whites did not belong in an urban environment (Cf. Rowley 1958 for Papua New Guinea ; and Mamak 1973 for Fiji).

8 Employers however did not hesitate to replace dissident workers despite the labour shortage.

9 Some of these features seem to apply to labour bn other colonial areas as well (See, for example, Omvedt 1972 : 13-14).

10 The only other strike in Papua New Guinea that is comparable in scope to the Bougainville disturbance occured as far back as 1929 in Rabaul. In the cose of Fiji the only other strike equal in scope to the 1959 strike occurred in 1920 (See also Parnaby 1972 : 140, and Main 1948 : 216).

11 The close association between millenarianism and strike organization of African mineworkers is described in Meebelo (1971 : 256). In Bougainville the predominance of former "cargo cultists" in the Navitu Association, a modern organization concerned with social, economic, and political development provides a good example of this linkage (Mamak and Bedford 1974b). The economic basis of millenarian movements is amply corroborated by the tendency of employers and the authorities to attach a Marxist label to them (See, for example, Worsley 1957 : 490 ; and Mamak 1973).

12 To stress the economic component of millenarian movements is not to deny the robe of ideas and religion in these movements.

13 In 1953 almost 80 percent of the Papua New Guinea labour force mas employed on a casual basis or under agreement (Danks 1956 : 14).

14 In contrast to urban industries the pattern of rural occupational activities is marked by ethnic segregation--the majority of rural Indo-Fijians are engaged in growing sugar while the majority of rural Fijians are engaged in village or specialized agriculture. As previously mentioned, however, on increasing number of Indo-Fijians are moving from rural to urban industries.

15 This pattern of segregation is world-wide and is not limited to the Bougainville copper mining industry.

16 For a discussion of the widening economic gap between whites and blacks in PaPua New Guinea see Langmore (1969) ; increasing income inequality between white and black mine workers in South Africa is described in Wilson (1972). Although black mine workers in Bougainville are comparatively better-off in wages and working conditions than their compatriots employed elsewhere their expectations are also much higher, firstly, because of their involvement in an industry which generates great amounts of wealth, and secondly, because black-white economic differences are very noticeable.

17 A similar case illustrating differences between Polynesian and more experienced Melanesian workers in their attitudes to unionism in the Solomon Islands is described by Larson (1970).

18 For a description of a parallel situation in the Bulolo goldfleld (Papua New Guinea)in the 1930s see Healey (1968 : 32-34).

19 Ethnicity may be emphasized by whites as a diversionary tactic, i.e., to prevent antagonism from being directed solely at them (Figure 2). Conversely, the ethnically homogeneous, upper class blacks may also employ a similar tactic by emphasizing the racial cleavage and/or class differences between blacks and whites (Figure 1).

Notes de fin

* Ethnic/Class tensions if high status group is ethnically homogeneous ; class tensions of group is ethnically heterogeneous.

If group is ethnically heterogeneous ethnic tensions due to job competition are likely to develop but will probably be insignificant due to the low number of blacks in this status position.

Table des illustrations

Légende Fig. 1 RACE/CLASS
Fichier image/jpeg, 89k
Fichier image/jpeg, 100k
Légende Fig. 3 RACE/CLASS
Fichier image/jpeg, 116k
Légende Fig. 4 CLASSHeavy lines denote type of cleavageNote **Note **
Fichier image/jpeg, 97k


University of New South Wales (School of Socioloy)

University of Cantherbury (Department of Geography)

© Société des Océanistes, 1978

Conditions d’utilisation :


Volume papier