- v.r-....,a. .....I.. . A '7‘ \9'| l "W / UNIVERSITY LIBRARIES IIHIHHIHHIIIH HWH I \l il 3 1293 00567 754 This is to certify that the dissertation entitled WOMEN, POWER, AND GENDER: A CRITICAL ANALYSIS OF FEMINIST PERSPECTIVES IN ANTHROPOLOGY presented by Karen Collamore Sullivan has been accepted towards fulfillment of the requirements for w/flaK/fl/L] (NA) Major professord Date " ~30 "9361 MSU is an Affirmative Action/Equal Opportunity Institution 0-12771 LIBRARY Michigan State University l E SE MSU LIBRARIES RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date stamped below. f» . “I 'h a ‘- .‘J : ' I ’ . If -¢-de J 1.5... .C AR 13 2000 8,71% zgoj U u' w t) WOMEN, POWER, AND GENDER : A CRITICAL ANALYSIS OF FEMINIST PERSPECTIVES IN ANTHROPOLOGY BY Karen Collamore Sullivan A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Anthropology 1989 ABSTRACT WOMEN, POWER, AND GENDER : A CRITICAL ANALYSIS OF FEMINIST PERSPECTIVES IN ANTHROPOLOGY BY Karen Collamore Sullivan The purpose of this study is to present a critique of feminist perspectives in anthropology. Focusing on the topical issue of women and power, the critique addresses efforts to construct an 'anthropology of women'. The dis- cussion argues that feminist perspectives in anthropology are built around assumptions of Western feminist thinking rather than anthropological understanding. The first part of the study identifies three general themes underlying feminist anthropology. The first theme concerns the association of female status with female biology and the idea that women’s 'biological superiority' is everywhere translated into social inferiority. The second theme con- cerns the use of male bias both as a form of explanation for anthroplogy's lack of attention to women, and as a premise guiding women-focused research. The third theme centers on the idea that manifestations of women’s power and their social value as individuals are to be found in specifically female domains of social life, in what is frequently referred to as ’women's culture’. The discussion shows how each of these themes combine to create a general picture of 'women' that can accomodate Karen Collamore Sullivan intellectual commitments to Western feminist ideology but cannot, at the same time, accomodate the ethnographic prob- lematics with which anthropology must deal. Drawing on ethnographic data from the culture area of Melanesia, the latter part of the discussion looks to recent anthropologi- cal frameworks for the study of gender construction as an alternative to feminist frameworks for the study of women. Copyright by KAREN COLLAMORE SULLIVAN 1989 iv TO Kay and Sully ACKNOWLEDGEMENTS I would like to express my gratitude for the intellectual guidance and support that I received from my doctoral committee : Dr. Charles Morrison, Dr. Robert McKinley, Dr. John Hinnant, and Dr. Harry Raulet. I am also grate- ful to Dr. Diane Levande for her helpful comments and suggestions. In addition, special thanks go to John Hinnant for his generous computer assistance, and to Richard Burke for his patience and good humor. vi CHAPTER ONE CHAPTER TWO CHAPTER THREE CHAPTER FOUR CHAPTER FIVE CHAPTER SIX APPENDIX A APPENDIX B BIBLIOGRAPHY TABLE OF CONTENTS Introduction Female Status and Female Biology Male Bias and the Anthropology of Women Politics and Domains of Female Power Feminist Anthropology and the Cultural Construction of Gender Conclusions Map of Melanesia Map of Papua New Guinea vii 28 46 64 111 173 194 195 196 CHAPTER ONE INTRODUCTION The existence of what Western intellectuals recognize as female subordination is documented throughout the world. That 'one half of humanity' is consistently accorded what many refer to as 'secondary status' presents a troubling fact. Motivated in part by Western feminist concerns regarding political and economic equity for women, anthro- pologists over the last decade have given considerable attention to the position of women in non-Western societies. If, through new interpretations or new data, anthropologists cannot revise the picture and show perhaps that women are not as subordinate or powerless as previous studies suggest, they can at least attempt to explain why the structure of human relations in so many of the world's societies consistently places constraints on the opportuni- ties and participation of women in their society. Anthropological efforts to explain women’s powerlessness, and conversely, to ascertain indicators suggesting women’s power in a particular socio-cultural context, reflect a varied range of analytic approaches.1 1 Yet, a common thread throughout most of these discussions is the recognition of an inadequate representation of women in the ethnographic record. Explanations ranging from overt male bias to an uncritical acceptance of ’male- generated’ models for ethnographic anlysis. have been posited for the historical paucity of data detailing women's lives. We have been unable to identify the existence and nature of women’s power, it is argued, be- cause the theoretical models available for analysis of power relations have been developed by male anthropologists whose studies have focused on male perspectives. In an effort to counteract this imbalance a great deal of ethnographic attention has been directed toward collecting data on women that reflects women's perspectives. In the process, some anthropologists have gone back to older ethnographic accounts in an attempt to piece together the 'neglected' data on women and thus provide a re- interpretation (Weiner 1976). Others have turned their attention to evolutionary questions on the origins of gen- der inequality (Leacock 1983), of male dominance (Sanday 1981), of the exploitation of women (Moore 1977), as well as attempting to identify universals concerning the posi- tion or status of women and patterns of female subordina- tion in human societies (Sanday 1974, Whyte 1978). Embedded within this literature are two assumptions. The first is that in order to understand the power women do or do not have, analyses are needed that take women as the primary focus. The second assumption is that there exists something that can be called "women's power". While recent critiques of women-focused research have noted the limita- tions such single-gender approaches have for our understanding of social relations,2 few have seriously questioned the theoretical basis upon which notions such as ’women's power' rest. The purpose of such research is to raise such questions and to present a critique of feminist anthropology. The argument throughout is that the genre of women-focused anthropology (and within that, the issue of women and power) is based on Western feminist thinking rather than on anthropological understanding. Instead of expanding our knowledge about the nature of human social relations, or about male/female relations in particular, the current trend of research on women limits us to a very superficial and simplistic view of women and women's lives. From a feminist point of view this may seem to be suffi- cient in that such research is intertwined with ideological efforts to redress a perceived historic neglect and misrepresentation of women in sociological analysis, and to facilitate structural change in the position of women worldwide. From an anthropological perspective, however, such an approach misreads the integrity of indigenous structures of knowledge and what we know, as anthropolo- gists, to be the complexity of social relations cross- culturally. By focusing only on women, we in fact learn very little about women or about the societies in which they live. It should be emphasized that the critique presented here is not meant to discount the relevance of feminist issues and discourse in Western society, but to question the ease with which these issues are transposed onto other cultural systems. The focus here is on efforts to con- struct an ’anthropology of women’. One of the results of these efforts has been a heigtened awareness of the impact of cultural influences on nodes of analysis. Few would argue against the value of insights gained through the self-reflection of an intellectual discipline, and part of the value of feminist research rests on the reflexive stance it often takes regarding anthropological discourse. Yet at the same time, feminist anthropology remains curiously exempt from such scrutiny. In reviewing anthropological research concerning women one is struck by a preoccupation with female biology, with perceived linkages between female biology and female status; with the search for empirical indicators and 'measurements of such things as male dominance and female subordination; with assumptions about the existence and distinctiveness of ’women’s culture; and with an abundance of criticism aimed at male-generated analytic models, yet little critique of the models themselves. By taking a closer look at this literature we begin to see how popular beliefs about the universality of male dominance, and about the historical prevalence of male bias, combine with the assumption that intrinsic qualities of ’femaleness’ render women not only biologically distinct from men but psychologically and culturally unique as well. In other words, we begin to see the construction of ’women’ as an analytic category separate from men. As this critique will show, the formulation of feminist perspectives on women (and on the ’anthropology of women’) requires a revision of the anthropological record and assumes that such revisions will provide a significantly different picture of women. That this is often the case seems to lend support to the idea that previous male-biased studies ignored and devalued the participation of women in society, and that the picture of women that emerges from feminist studies is more authentic. The discussion here questions this assumption of authenticity by arguing that the significantly different picture of women that seems to result from women-focused research already exists in the frameworks used to study 'women; and that these frameworks are derived from feminist objectives in our own culture rather than a clearer or less biased reading of the ethnographic data. While addressing the ’anthropology of women’ in general, there are several reasons for selecting women and power as a focal point. First, there is the practical necessity of narrowing down what is a large and continually expanding amount of anthropological literature focused on women. Second, studies of women, regardless of their ex- plicit focus, are often about power (and powerlessness). This is because feminist perspectives in anthropology incorporate components of the wider field of Western feminism which is, after all, about power relations. The fact that topics such as women’s status, or women’s roles, or women’s productive (and reproductive) activities are frequently framed in terms of contrast to men, in terms of exclusion from ’male’ activities, and in terms of structural features seen as representing power hierarchies bounded by gender, is a reflection of the relationship between the anthropology of women and the wider world of Western feminist thinking. In part the issue of ’women’s power’ has been merged with recognition of an historic neglect of women on the part of anthropologists. Over the last decade attempts to account for why women have been ignored in anthropological analysis pointed to the need for (and at times heralded the arrival of) new theoretical perspectives and analytic models designed specifically to accomodate the study of women (Reiter 1975, Rosaldo and Lamphere 1974). From such research we are frequently reminded that women don’t just give birth, raise children, and tend to domestic chores. Women do have wider social concerns and interests and they pursue these interests with as much intent and savvy as men. The difference, however, between the power women may have and that of men is that women are often not part of the formal (male) power structure, being instead, confined to the domestic domain. Therefore, it is argued, women’s skills at achieving their goals must be measured by their ability to circumvent the formal power structure. The discovery that women often do have power, or at least are not as powerless in their own societies as is frequently thought, has been credited to new perspectives and interpretive frameworks generated by feminist anthropologists. Unlike earlier male-biased, male-oriented frameworks that paid scant attention to women in the societies studied, these new analytic frameworks focus on women as ethnographic persons, as individual contributors to social life who have a social identity in their own right and are not just appendages to the male world. Close inspection reveals, however, that aside from their focus on women, these new analytic frameworks are not really new at all, but are essentially the same models used in the ’male- oriented’ studies. Thus, while it is true that such work has enabled us to better situate our knowledge of the nature and range of women’s participation in their societies, the feminist call for re-thinking anthropological models has not lived up to its earlier enthusiasm. Instead of re-thinking the models, anthropologists studying women have, for the most part, merely shifted the focus to women. In this manner, focusing on women has become confused with and sometimes misrepresented as re-thinking analytic models. This is particularly apparent in studies of women and power where it is assumed that there is such a thing as ’women’s power’ (as distinct from ’men’s power’) and that women’s power is derived from a uniquely female repertoire of social action. The discussion that follows falls roughly into two parts. In the first part (chapters two, three, and four) three general themes are identified as forming the basis of feminist perspectives in anthropology. Throughout the course of the discussion it will be shown how each of these themes overlap and combine to create a general picture of ’women’ that can accomodate intellectual commitments to Western feminist ideology but cannot, at the same time, accomodate the ethnographic problematics of cultural differences with which anthropology must deal. The first theme concerns the feminist association of female status with female biology and the idea that women’s ’biological superiority’ is everywhere translated into social inferiority. Chapter Two will show how arguments attempting to explain male dominance and female subordina- tion take as their starting point the assumption that the ’secondary status of women’ is a cultural universal derived from an association (assumed to be equally universal) of women with nature and men with culture. Based on the idea that biology is the primary determining factor in cultural perceptions of ’femaleness’, the male/female : culture/nature association forms the basis of a range of logically equivalent analytic dichotomies employed by anthropologists to ascertain the status, position, power, and ultimately, the cultural value of women. The second theme to be discussed concerns the use of male bias both as a form of explanation for anthropology’s lack of attention to women and as a premise guiding women- focused research. Because it has become commonplace now to account for an absence of data on women by invoking the existence of male bias it is easy to miss the deeper implications of the way this issue is used in feminist perspectives. Chapter Three explores these implications by looking at the thematic role of male bias in the development of an anthropology of women. As this chapter will show, continued interest in the accumulation of ’evidence’ of male bias among anthropologists has led to the rather easy dismissal of earlier ethnographic accounts and suggests that what constitutes a ’new interpretation’ need only be a singular focus on women. Here the feminist 10 preoccupation with the biological facts of ’femaleness’ takes on an added dimension -- biology becomes a measure of ethnographic competence and insight as feminist discussion of male bias asserts that studies of women by women are somehow more authentic. The third theme centers on the idea that manifesta- tions of women’s power and their social value as individuals are to be found in specifically female domains of social life, in what is frequently referred to as "women’s culture". Chapter Four focuses on analyses of women’s power. In examining feminist approaches to the study of women and power one model in particular -- referred to throughout the discussion as the "political action model" -- stands out both for its impact on anthro- pology in general and for the pervasive influence it has had (and continues to have) on the study of women. The salient features of this model include : the analytic focus on individuals (as opposed to groups); the conceptual expansion of the notion of politics such that ’politics’ is seen to inform behavioral dynamics at all levels of interaction; and the assumption that power, derived from politics, is thus an aspect of all social relations and is accessible (though in widely varying degrees) to all mem- bers of society. 11 This model has been widely criticized on the grounds that it implicitly incorporates Western convictions about individual free will and the competitive marketplace, equating the latter, by analogy, to society in general (Ahmed 1976, Asad 1972). As employed in anthropological studies of power, this model imputes a calculated rationality to individual decision-making suggestive of a further analogy to Western economic theory : the cost/benefit analysis. Faced with choices, individual actors make decisions based on personal self-interest, and the behavioral dynamics of individual decision-making are frequently discussed in terms of "action strategies”. According to this framework the acquisition of power is tied to the individual’s skill in manipulating the actions of others and the cultural resources available to them. Despite the theoretical flaws and the conspicuous ethnocentrism identified by other anthropologists, the political action model has been widely adopted for studies of women and power. While feminist critique of this frame- work does exist, it is consistently focused on the issue of male bias and the assumption made by anthropologists that power and politics are exclusively male pursuits. But why use an analytic model characterized as male-biased to study women? As the discussion will show, for feminist anthropology the problem was not the model but the male perspective it generated; and the resolution to this prob- 12 lem was sought not in the development of new theoretical frameworks but in the application of this same model to studies focused on women. Within much of feminist anthro- pology the question is not "do women have power?" -- for in fact the political action model already assumes that, to some degree at least, they do. Rather, the question asked is "what is the nature of women’s power?”. The nature of women’s power, as it turns out, is directly linked to distinctively female styles of action occurring in female domains of social life. To understand the popularity of this model for studies of women we must look to the themes underlying feminist anthropology. On the one hand this model supports conten- tions about male bias and the belief that female ethnographers have a special insight into their (female) subjects. On the other hand, the model accomodates assumptions that ’women’s power’ exists and, because it derives in part from the unique qualities of ’femaleness’, it is best understood by focusing on women as a separate analytic category. As Chapter Four will show, the politi- cal action model provides an ideologically appealing frame- work for the ’anthropology of women’ because it facilitates the feminist depiction of women as individuals and actors in their own right. In the process it grants a degree of social and cognitive autonomy to women in societies which appear to structurally deny them both of these qualities. 13 It should be emphasized that the purpose of this critique is not to suggest that women may or may not be subordinate, oppressed, powerless, or exploited, but rather, the intent is to question this as a starting point for comprehending cultural differences in the treatment of women (or ’position’ of women) in human societies. What do we learn about women when we view them as separate from men; when the analytic models we use not only categorically separate women and men in pre-defined domains of social life, but implicitly rank these domains as well? And what do we learn about male/female relations when we begin by assuming an imbalance exists, or that questions concerning sexual parity and equality are relevant cross-culturally? The answer to this might be that we learn pretty much what we already suspected -- that for the most part throughout the world an imbalance exists in male/female relations with women playing a subordinate role vis-a-vis men. Contrary to much of the women-focused literature in anthropology, this discussion argues that if we are compre- hend the ’position’ of women we must do so by first comprehending the culture of which they are a part, and that the topic of ’women’ or ’women’s power’ is not a particularly relevant starting point for accomplishing this. It is not relevant because the analytic models used cannot accomodate cultural context in a manner that would allow for an understanding of native meanings as these l4 pertain to men and women alike. What these models do accomodate is the creation of an artificial picture of ’women’ and ’women’s culture’, the origins of which can be found in the problematics surrounding women’s roles in Western industrial society. How, then, do we come to understand the place women hold in their societies? While the first part of this discussion focuses on the identification of problems with feminist approaches to the cross-cultural study of women, the latter part looks to the anthropological examination of cultural notions of gender and social personhood as providing a substantive alternative. Chapter Five presents the view that cultural theories of gender are the conceptual key to understanding the social organization of relations between the sexes, and that this must be the starting point for examining the position, status, or power of women. In other words, if women’s power is a relevant issue, it is so only in the wider context of culturally constructed theories about gender identity and gender rela- tions. On a more general level, the argument here is that the depth of our understanding of women’s place in society is a direct reflection of the depth of our understanding of cultural diversity and variation in the way different societies comprehend and reckon with their own worlds. So while feminist writers might argue that this understanding can only be achieved by focusing on women, by presenting a 15 female perspective, or by looking at women’s activities, this research suggests that a different perspective is needed, one which takes into account the fundamental impor- tance of cultural context and cultural theories of maleness and femaleness. Before proceeding with a description of Chapter Five, it should be noted that over the last six or seven years ’gender’ has become an increasingly popular topical focus within anthropology in general and feminist anthropology in particular. For feminist anthropology this interest in gender suggests a shift in orientation away from women- focused research toward a more balanced examination of the cultural configurations of male/female relations. But this shift in focus is somewhat deceptive insofar as feminist research tends to view ’gender systems’ as the ideological basis supporting structural arrangements that define women as inferior. As Chapter Five points out, the assumptions embedded in feminist perspectives -- about male dominance, about the secondary status of women, and about the biologi- cal basis of cultural perceptions of ’femaleness’ -- remain essentially intact, and examining gender becomes a contemporary approach to verifying them. Gender and sexual (physiological) identity tend to be treated as synonomous, and. ’gender constructs’ (which frame a culture’s ideas about what constitutes male and female) are viewed as the cultural translation of biological differences rather than 16 the cultural construction of differences that transcend the biological facts. It is on this latter point -- how the cultural construction of gender differences transcend biology -- that Chapter Five focuses. The ethnographic focus of much of the overall discussion, and specifically Chapter Five, is the culture area of Melanesia. The ethnography of Melanesia presents us with numerous examples of why women-focused anthropology can provide only a superficial and simplistic picture of women and the societies in which they live. In the process the data re-directs our attention away from issues concerning power per se, toward the complexities of gender ideologies and the ways such ideologies are manifest in relations between men and women. Anthropologists working in Melanesia have historically had to deal with the diversity, the contrast, and the complexity of cultural patterns that characterize this region. Partly because of this, research on this culture area reflects a rich cross- section of anthropological thinking and re-thinking of key issues concerning kinship, political organization, ritual and religion, economics, and most recently, gender. Indeed, several of the examples cited in the earlier chapters will illustrate how anthropologists, drawing on data from Melanesian societies, have disputed the relevance of the nature/culture dichotomy and the domestic/public distinction, and have questioned feminist emphasis on 17 looking at ’women’s culture’. In addition, recent analyses of male-focused ritual behaviors found throughout this region highlight efforts to recast anthropological perspec- tives on gender in order to theoretically accomodate the complex cultural formulations upon which individual societies construct their social worlds. Thus it can be said that both on the level of indigenous culture theory as well as on the level of anthropological interpretation, the ethnography of Melanesia provides a compelling counterpoint to feminist perspectives in anthropology. Located in the Western Pacific, the major land areas that define Melanesia geographically include : Fiji, New Caledonia, Vanuatu (formerly New Hebrides), Loyalty Islands, Solomon Islands, New Britain, New Ireland, Admirality Islands, Trobriand Islands, and the island of New Guinea3. (See map in Appendix A.) Estimates of the number of distinct cultural groups within Melanesia range between 700 to 1000, and the number of indigenous languages is etimated at over 2000 (Herdt 1984). Such estimates suggest a striking range of cultural difference, and ethnographic reports on particular groups continually bear this out. While broad patterns and similarities can be identified regionally, one can simultaneously identify exceptions and distinct local variations specific to individual societies. sn su ra va ti sm. ki] ti< alt rec (Ti. an ant div Vag Var. and fans aSsc ante ritu Virt (Pol 18 On a very general level, however, traditional Melanesian societies can be characterized as predominantly small-scale technologically simple settled communities supported by a combination of subsistence gardening, pig- raising, and hunting (though each of these may receive vastly different regional emphases). Political organiza- tion throughout the area suggests a continuum ranging from small localized polities where politics is enmeshed in kinship, to those exhibiting increased scale, specializa- tion, differentiation, and hierarchy (Allen 1984a). And although patrilineality is the predominant mode of descent reckoning, cognatic and matrilineal systems also exist.4 (The nature of group structure in Melanesia has long been an issue of considerable interest and debate among anthropologists.5) Despite its immense cultural and environmental diversity, Melanesia displays a fairly consistent and per- vasive theme of sexual polarity which manifests itself in a variety of pronounced and vivid ways through myth, ritual, and daily life. Cultural expressions of this polarity range from beliefs concerning female pollution and dangers associated with sexual contact, overt displays of antagonism and hostility between men and women, dramatic rituals of male initiation and secret male cults, to. the ‘virtual exclusion of women from decision-making contexts (political, economic, and ritual), and general cultural 19 perceptions which denigrate the capabilities of women and the value of their labor. While variation exists in the degree of intensity and elaboration of this polarity, the ethnographic literature on Melanesian societies provides numerous striking illustrations of an overarching cosmology which emphasizes male dominance and control over women, over social relations, and over material resources. Prominent cultural patterns such as patrilineal descent, warfare, and ceremonial exchange systems provide much of the structural support of this cosmology, while variations on symbolic themes concerning life processes, the body, gender, work, marriage, and personhood give shape and meaning to social relations. On the level of generalized ethnographic description of Melanesian societies it is fairly easy to view women’s position as subservient to men, the relationship between men and women antagonistic, tension-ridden, and at times overtly hostile, and from this to assume that women lack power and control over their own lives. Yet social rela- tions are rarely so simple or so unidimensional as this characterization implies. Data from Melanesia, and in particular Papua New Guinea, continues to draw our atten- tion to the complexity inherent in worldviews vastly different from our own. 20 Chapter Five takes a closer look at ethnographic data from Melanesian societies to illustrate the point that there are more fundamental questions concerning native worldview that must be asked prior to trying to assess the behavioral evidence of female subordination or powerlessness, or, conversely, the existence of female power. One must first look to the ideological constructs which give shape and meaning to cultural notions of personhood and gender (and hence give meaning to social roles and behavior). Indeed, topics such as female subordination and female power tend to fade into the background as one begins to look more closely at cultural patterns of gender definition in Melanesian societies. Anthropologists have long drawn our attention to the ways gender distinctions seem to permeate Melanesian societies. Throughout the region culture theories of procreation, human development, health and illness, as well as success and failure in daily pursuits invoke a powerful imagery of the interaction of human substances such as blood, milk, and semen with the natural environment and with the social affairs of human beings. This is particularly apparent in beliefs concerning the polluting qualities of women, and restrictions surrounding the handling and consumption of certain foods. For the most part, the practices ensuing from these beliefs are perceived to protect men from women, and it has been widely 21 noted that such beliefs serve to reinforce and institutionalize women’s subordinate status. In several respects, then, Melanesia enables an easy application of a feminist framework insofar as the most visible features -- residential segregation of the sexes, male fear of female pollution and avoidence of prolonged contact with women, exclusion of women from ritual and political affairs, devaluation of women’s productive labor and in some cases women’s reproductive role as well -- stand as evidence of male dominance and female subordina- tion. In addition, the cultural emphasis on the influence of human substances (particularly female substances) on the physiological and social progress of an individual’s life is frequently taken as evidence of how biology is culturally translated into structural relations of superiority and inferiority. Thus it has been argued that despite the overt separation of the sexes in daily life, men cannot symbolically separate themselves from the biological and regenerative powers women possess; that through their control over social and ceremonial affairs men are able to publicly deny what they privately know to be the ’natural’ superiority of women (Meigs 1984, Weiner 1976). The problems identified for feminist perspectives come into sharp relief when juxtoposed against recent studies of 22 gender acquisition in Melanesian societies. As the earlier chapters explore why feminist perspectives in anthropology fail, this chapter will illustrate hgg they fail by re- examining a recent analysis of gender among the Hua (Papua New Guinea) by Anna Meigs (1976,1984) in light of other ethnographic discussions of gender construction in Melanesia. Meigs’ work is interesting because it incorporates the general premises of a feminist perspective : first, that physiological differences between men and women are the determinants of cultural perceptions of gender, and second, that the apparent male dominance in Hua society derives from male recognition and envy of female reproductive superiority. At the same time she tries to frame her analysis in terms of the cultural construction of gender. That she fails in this has to do with her reliance on a theoretical model informed by Western feminist theory and Western models of human physiological and psycho-social development. According to Meigs’ analysis, gender constructs in Hua society are extremely restrictive. Both sexes feel con- strained by gender roles and both sexes, at different times and in different ways, "imitate" the opposite sex. Although men do so more frequently than women, Meigs con- tends that all such "imitative" practices constitute con- scious efforts to break through or "blur" the cultural boundaries of gender. The evidence for Hua imitation of 23 the opposite sex is located in specific behaviors identi- fied by Meigs as : male menstruation, male pregnancy, periodic and semi-secret consumption of (prohibited) foods identified with the opposite sex, and gender reversals occurring in old age. The ’imitative’ behaviors cited by Meigs are not unique to Hua society but occur (with varying frequency and elaboration) elsewhere in Melanesia. Drawing on recent anthropological discussions of similar practices in the region, it will be argued that rather than fighting against the rigidity of gender boundaries, Hua are collectively involved in creating and re-creating these boundaries: that the feelings of constraint Meigs attributes to Hua (and in particular Hua males), are perhaps more appropriate viewed as expressions of amibivalence about their control over processes through which social identity is forged. In the course of the discussion it will be shown how the data Meigs presents actually supports this re-interpretation, but the analytic framework she employs prevents her from situating the cultural behavior she describes within the larger context of Hua cosmology. In considering how and why Meigs’ analysis fails to achieve the interpretive depth that her data actually suggests, we must return again to the problems inherent in feminist-oriented models in anthropology. One of the 24 points to be made in this part of the discussion is that feminist analysis of gender, like the earlier analysis of female status and women’s power, rest on a superficial reading of cultural behavior and context. Evidence for the structure of gender differences in social life, and cultural perceptions of these differences, are taken from the most obvious levels of social interaction. Drawing on alternative frameworks for the study of gender, this chapter represents an effort to explore how ethnographic data might be ’read’ differently without the assumptions embedded in feminist perspectives. From the standpoint of the overall discussion Chapter Five may seem to move somewhat afield of the earlier focus. But this in fact is one of the points the chapter tries to make. Once one begins to look closely at native cosmology in Melanesian societies it becomes increasingly difficulty to analytically sustain such narrowly conceived topics as those addressed by feminist perspectives in anthropology. The final part of the discussion, Chapter Six, weaves together the various arguments presented throughout this critique. While summary in nature, this chapter will also highlight some of the tangential issues regarding feminist analysis in anthropology that emerged throughout the discussion. In particular this chapter will address some of the ironies and contradictions that arise from efforts 25 to develop an anthropological perspective that incorporates Western feminist ideology. 26 NOTES TO CHAPTER ONE See for examples E. Ardener 1972, S. Ardener 1977, Atkinson 1982, Collier 1974, Leacock 1983, Milton 1979, Rosaldo 1980, Sandy 1974, Weiner 1976. Notably, M. Strathern 1984, Lindenbaum 1984, Rosaldo 1980. The island of New Guinea is politically divided in two : the eastern half is the independent nation of Papua New Guinea; the western half, Irian Jaya,(formerly Dutch New Guinea), is currently under the political control of Indonesia. (See map in Appendix A.) In this discussion references to New Guinea are drawn from data on cultural groups located in Papua New Guinea. While much less common than patilineal or cognatic descent systems (particularly on the island of New Guinea), there is a fairly widespread occurrence of matrilineality in Melanesia. In his recent discussion of the political implications of group structure in Melanesia, Allen (1984a) states that matrilineal descent systems are found extensively in areas of the Huon Gulf, most of New Britain and New Ireland, the Massim arche- pelago (Trobriand Islands), Bougainville, parts of the Solomon Islands, and a large portion of north and central Vanuatu. On a general level, politics in these areas still conform to the Melanesian pattern : polities are small in scale and there is a consistent absence of overarching structures of a bureaucratic or administra- tive kind. On a more specific level, however, differ- ences do exist and Allen notes that the Melanesian communities where matrilineal principles are accorded importance also exhibit the most elaborate and complex forms of political association reflected in such features as : ranked descent groups, hereditary titular systems (often combined with increased political significance of territorial organization),voluntary male associations (ranging in form from the secular clubhouse found in the Huon Gulf area to discrete secret societies found in New Britain and adjacent areas), and elaborate male status hierarchies based primarily on achievement (Allen 1984a:26). .sr' 27 5. For an overview of some of the main issues involved in anthropological discussion of group structure in Melanesia, see Barnes 1962, Langness 1964, Pouwer 1964, Kayberry 1967, de Lepervanche 1967, 1968, Kelly 1976, A. Strathern 1972. CHAPTER TWO FEMALE BIOLOGY AND FEMALE STATUS In looking for explanations of male dominance, several anthropologists provide evolutionary perspectives highlighting a biological basis for cultural roles attri- buted to women. Because of women’s ’biological superiority’ (i.e. the ability of women to give birth and the physical dependence of young children on the mother) it has been hypothesized that women are perceived by men as a threat; men envy women for their reproductive abilities and thus proclaim dominance over them in the one area where they can exhibit control -- social life (Ortner 1974). Along these lines, Raphael (1975) argues that the biological demands of bearing and raising children and providing for the daily of needs of a domestic household leave little time for women to enter the political sphere in a significant way. Likewise, the time required for dealing with political and economic concerns leaves men with little time for becoming involved with caretaking and domestic functions. According to Raphael, however, this does not mean women are powerless, for in fact women’s role in "determining the outcome of each new generation" holds 28 29 an "intrinsic power unparalleled elsewhere in society" (1975:11). The division of labor and the ensuing structure of relations between men and women is thus attributed to the biological facts of human reproduction. According to this view, cultural evolution provides us with a record of the cultural variation in, elaboration upon, and further embedding of what is essentially a ’natural’ (biologically based) system of male/female relations in human society. A similar yet somewhat more sophisticated version of this explanation is presented in the work of Sanday (1973, 1974) who attempts to develop an empirical model of the evolution of female status. She suggests that throughout cultural evolution social survival has depended upon different energy expenditures by males and females in three major activity areas : reproduction, defense, and subsis- tence (1974:189). Similar to Raphael, she maintains that the constraints of reproductive activities limit the amount of energy women can direct to other activity areas, and these other activity areas, in turn, place greater demands on the energy of men. It is through this increased energy expenditure that men gain access to and control over cultural resources. Since reproductive activity falls to the female, a constraint is imposed on the proportion of total female energy to be utilized in other activities. Such a constraint in turn increases the proba- 30 bility that the other two tasks draw more on the energy of males, thus placing men in a strategic position to gain control of resources. (1974:189) A characteristic feature of arguments attempting to explain male dominance and female subordination from an evolutionary perspective is the assumption that the ’secon- dary status’ of women is a cultural universal. Sherry Ortner (1974) calls women’s secondary status one of the true universals -- "a pan-cultural fact" (1974:67). Ortner’s work explores the universal devaluation of women in terms of the biological fact of female reproductive abilities and universal cultural perceptions (drawn from the biological facts) of women as being closer to nature. As a theory it is wrought with assumptions that remain problematic. In the formula posited by Ortner (culture: nature::male:female) the association of nature with women and culture with men assumes the superiority of men as a category over women as a category. But it also depends on a further association between the dyad men/women and another opposition, public/domestic. Given the assumption that culture is everywhere perceived to be superior to and dominates nature, so too the male public world of politics and ritual dominates the female world of domestic labor and childbearing. In this sense, the opposition nature and culture is seen as mediating the facts of biology and the organization of society, and providing a basis for legiti- 31 mizing and institutionalizing women’s social location with- in the domestic sphere and their subordination to men. Another problematic feature of this argument is the way in which Ortner moves from the cultural recognition of a distinction between nature and culture to the cultural assertion of a hierarchical relationship with nature. Thus culture (i.e. every culture) at some level of awareness asserts itself to be not only distinct from but superior to nature... (1974:73) This notion of hierarchy is essential to her argument. If women are everywhere devalued vis-a-vis men, and if women are culturally associated with, or perceived of as closer to nature, then nature, too, must be culturally devalued vis-a-vis society. In other words, in order for the equa- tion -- cultureznature::male:female -- to hold up, it must be logically consistent. Culturally recognized distinction (as between nature and culture) alone is not enough because while all cultures distinguish between male and female, they do so (according to Ortner) within a framework of superiority and inferiority. Therefore, a universal distinction between nature and culture must also reflect elements of hierarchy, with culture deemed to be not only different from but superior to nature. 32 Part of the appeal of the culture:nature::male:female formula stems from the acceptance of physiology as the primary determining factor in cultural perceptions of ’femaleness’. Those who use this model have already located ’femaleness’ in biology and ’maleness’ in the social domain. Yet the meanings culturally attributed to male and female are as arbitrary as the meanings attributed to culture and nature. Ortner’s argument has been characterized as "remarkably ethnocentric” in its sweeping assumptions about cultural perceptions of male and female and about nature and culture (MacCormack 1980). As MacCormack asks, "is there anything more intrinsically natural about women’s physiology than men’s?" (1980:16). She notes that although categories exist within a culture that anthropologists may wish to label ’culture’ and ’nature’, male and female do not categorically constitute a metaphoric transformation of culture and nature. Others have argued that the opposition is complementary rather than hierarchical; that culture is different from nature in a manner such that questions of inferiority and superiority are simply not relevant, nor easily superimposed onto male/female relations (Forge 1972, M. Strathern 1980). This line of criticism is echoed by several recent studies on gender in Melanesian societies which dispute the ethno- graphic relevance of a universal culture/nature, male/female equation.1 For example, Goodale (1980) notes that among the Kaulong of southwest New Britain the 33 distinction made between married/unmarried holds greater significance than the distinction Kaulong make between male/female. Married people constitute a category most closely associated with pollution (through heterosexual contact) and most similar to animals (through the act of copulation). For the Kaulong, it is married persons, not females, who are viewed as marginal, ambiguous, and ’closer to nature’. Similarly, Gillison (1980) notes that the beliefs of the Gimi people of the Eastern Highlands in Papua New Guinea do not conform to the nature:culture::male:female perspective. One cannot assume Gimi women are symbolically associated with the surrounding rainforest because, according to Gillison, the rainforest is considered a male refuge from women and ordinary life in the settlement. Gimi men’s intense fear of female pollu- tion and their efforts to avoid contact with women are ritually expressed as a desire to escape from women ”into nature”, into a non-human world which is seen to revitalize their masculinity (1980:146). Thus, if one were to apply the nature/culture model to Gimi society, the formula would have to be reversed, associating men with nature and women with culture. Assuming a universal cultural association of women with nature lends support to the prior assumption that cultural perceptions of ’femaleness’ are universally drawn from the facts of female biology. These two notions are 34 further linked through projections of inferiority and superiority on elements of each dyad : because of female biology women are everywhere perceived as ’closer to nature’ and nature is everywhere seen as inferior (and threatening) to ordered social life. What results is a speculative and superficial view of cultural uniformity in the structure of relations between men and women and a re- affirmation (framed in evolutionary terms) of the power- lessness or subordination of women. The imputation of a hierarchical framework structuring male/female relations in pre-industrial societies has drawn criticism from anthropologists who contend that non-class societies were (are) in fact characterized by a ’relative’ sexual egalitarianism. For example, Leacock (1981, 1983) cautions against using hierarchically informed analytic frameworks to assess the ’role of women’ or ’women’s status’ in egalitarian societies and argues that female subordination is an historical, not a natural (biologically based), phenomenon. According to Leacock, the origins of gender hierarchy ”are inextricably meshed with the origins of exploitation and class stratification" (1983:269). Sacks (1976, 1979) poses a similar argument when she criti- cizes the assummption that a sexual division of labor implies asymmetrical relations between men and women. .Addressing the assumption that women’s status in society is dictated by their reproductive functions, she calls to task 35 other anthropologists for failing to recognize that the production of children and the production of culture are neither incompatible nor mutually exclusive. Rather, she notes that certain forms of social relations, particularly those instigated by the rise of industrial capitalism, have made them so. As both authors argue, the perceived sexual asymmetry among pre-industrial (non-class) societies can be explained either by historical transformations resulting from contact with and incorporation into the world market system -- a process by which symmetrical systems are transformed into asymmetrical systems; or, by Western observers who are often conditioned to see hierarchy every- where and who simply assume the existence of asymmetry. Stating that sexual inequality is not a given (non-class societies have ’relative’ equality among the sexes), they argue that anthropologists have been blind to sexual equality in non-class societies. Both Leacock and Sacks have been criticized on the grounds that while their arguments concerning Western intellectual preoccupation with hierarchy may be sound, they are unable to indicate what a sexually egalitarian society might be (Atkinson 1982). Furthermore, they fail to recognize the fact that notions of equality are as much a part of Western consciousness as those of hierarchy. Strong in their criticism against assumptions of universal 'male dominance, and in their objections to the contention 36 that female reproductive capabilities play a determining role in female subordination, they presume the existence of a rather uni-dimensional historical process leading to the subordination of women as well as all other forms of social inequality. Again we have a view of cultural uniformity, this time augmented by uniform historical process. Along with the issues discussed so far, how one de- fines ’female status’ is a perennial problem in cross- cultural research on the position of women in society. Few writers, however, provide discussion on what is meant by female status. The work of Sanday is particularly interesting in this regard because her efforts to construct an operational definition actually highlights the conceptual limitations of the notion of female status. According to Sanday (1974), any operational definition of female status first requires a distinction between the public and domestic domains of social life. Having drawn this distinction, she suggests that a general definition might be framed in terms of : 1) "the degree to which females have authority and/or power in the domestic and/or public domains": 2) "the degree to which females are accorded deferential treatment and are respected and revered in the domestic and/or public domains" (1974:191). Because of the difficulties in empirically measuring ’deferential treatment’ and respect, this latter aspect is excluded from her own analysis. And, because of her 37 insistence that the domestic and public domains remain analytically separate (treating them together would ”con- fuse the analysis"), Sanday’s treatment of female status concentrates exclusively on the degree of female power and authority in the public domain. While this seems to narrow things down considerably, it does not clarify the diffuse quality of ’female status’ -- a concept which is essentially dependent upon cultural context to give it meaning. What we can see here is a fairly prominent tendency among feminist anthropologists to assume that indications of female status are found by first identifying those activities of women that link them (however briefly or minimally) to the public domain. The extension of the male/female culture/nature association to include public/domestic continues the logical symmetry and reinforces the notions of superiority and inferiority already implicit in each of these dichotomies. Female status is seen as dependent upon the degree of women’s involvement in what is assumed to be a male domain. Using the degree of female participation in a male domain as the primary gauge for weighing female status (insofar as greater participation is seen to indicate higher status) supports the prior assumption that the secondary status of women is, in fact, a cultural universal. More importantly, it reflects our own culture’s evaluative sentiments 38 regarding ’work’ as opposed to ’domesticity’ and through this the value of men as opposed to women. (This will be discussed at greater length in chapter four.) Sanday’s work typifies the way in which feminist discussion of women’s status is skewed toward establishing evidence for what is widely accepted as women’s subordinate position relative to men. Instead of dealing with the problematic qualities of female status as a concept, Sanday focuses on a very limited (and limiting) range of measurable criteria. One needn’t be surprised then, that Sanday’s analysis leads her to conclude that : There is no doubt from the data examined that there is a wide range of variation in female public status cross-culturally. (1974:205) 2 Other anthropologists have noted methodological diffi- culties in trying to establish criteria for assessing the status of women and the cultural variation that results from such cross-cultural research. While she maintains that women’s secondary status is a "pan-cultural fact", Ortner nonetheless acknowledges the immensely diverse ways it it culturally manifested. ...within this universal fact, specific conceptions and symbolizations of woman are extraordinarily diverse and even mutually contradictory -- actual treat- ment of women and their relative power 39 and contribution vary enormously from one culture to the next and over different periods of history. (1974:67) In his study of pre-industrial societies, Whyte (1978) reaches a conclusion similar to Sanday’s as he notes that no pattern of universal male dominance emerged from his study. Instead, the study indicated ”much variation from culture to culture in virtually all aspects of the position of women relative to men" (1978:167). Yet, unlike Sanday, Whyte does not accept cultural variation as indicating only that there exists no society where males are totally domi- nant over females. Rather, it prompts him to question the assumption of universal male dominance, and to characterize such cross-cultural studies on the status of women as "an unproductive enterprise" (1978:168). Drawing from his own work he states that one can no longer assume there is such a thing a the status of women cross-culturally and sugg- ests that future research must begin with a very different assumption : That there is no coherent concept of the status of women that can be identified cross-culturally, and when we think that we are looking at aspects or indicators of the status of women we are dealing with essentially unrelated things. (1978:170) 40 similar sentiments are voiced by Rosaldo (1980). Reflecting on efforts over the last decade to ascertain women’s status cross-culturally, she notes that she has come to realize that ’women’s status’ is not one but many things. The failure of attempts to rank societies in terms of ’women’s place’ or to explain apparent variations in the amounts of privilege women elsewhere may enjoy (in terms consistent with cross-cultural data) suggest that we have been pursuing some- thing of a ghost -- or rather, that an investigator who asks if women’s status here or there ought to be reckoned high or low is probably conceptually misguided. (1980:401) The issue of cultural variation, while frequently noted, is rarely dealt with by anthropologists attempting to construct general (evolutionary) statements about female status in human society. Yet the immense variation that does occur from such studies should be a signal that not only are new questions needed, but new frameworks of inquiry as well. Little critique has been given to the overall feminist emphasis on searching for origins and evolutionary patterns, but the fact that anthropologists outside the feminist category rarely concern themselves with such questions is significant. In reviewing the literature, it could be argued that much of the writing on the status of women over the last 41 decade reflects Western feminists’ ambivalence toward and preoccupation with the biological process of childbirth and the changing structure of female roles in Western society. Many feminist writers, in fact, state explicitly the programmatic nature of their interest in exploring women cross-culturally in that such research will advance our understanding of the position of women in Western society.3 But the focus on the physiological fact of female reproduc- tive capabilities and its association with female status (and, by implication, powerlessness) nonetheless reflects a striking form of biological determinism which, curiously, is used to explain both how women are powerless and how they are powerful. As was stated earlier, to some the constraints imposed by childbearing and childrearing, which prevent women from becoming active participants in the political and economic spheres, are overshadowed by the "instrinsic" power women have as mothers and thus creators of whole new generations (Edholm, et. al. 1977, Raphael 1975). To others, these same constraints are seen as the ultimate basis for women’s secondary status and the insti- tutionalization of female subordination (Sanday 1974). Ortner’s (1974) argument, that the secondary status of women results from a universal association of women with nature, is interesting in that it implies that women are culturally perceived as possessing a kind of power (somewhat mystical) which men feel threatened by and are compelled to exert control over -- in much the same way as 42 they are (ostensibly) compelled to exert control over nature. In a sense then, women have power and lack power simultaneously. Both the link between women and nature, and the source of female power is female biology. That women are capable of giving birth is a fact acknowledged by all human societies, but this fact can provide only the most superficial of starting points for understanding the range of social roles and expectations of women. The inevitabilities of biology amount to the ’fact’ that women give birth and men don’t. Anything beyond that is open to cultural interpretation. On a theoretical level, the emphasis placed on female biology in studies of female status blurs the complex relation between sexual identity and the allocation of social roles within society. In much of the literature men are seen as freed from the confines of the domestic group and less constrained by an identity derived from male physiology. While women are defined by childbearing and lactation, the possibility that men are equally defined by their reproductive capacity is not usually considered. Rarely is reference given to the male role in socialization processes, or in procreation for that matter. Yet nowhere are women the sole participants in biological reproduction, nor are they ever culturally perceived as such. As LaFontaine (1978) notes, in most societies men are culturally recognized as begetters of children and their role in biological reproduction is given 43 comparable symbolic significance. Male potency is the conceptual counterpart of female pregnancy and receives symbolic emphasis in ritual contexts in many societies. (1978:9) LaFontaine argues that much of the literature on the social and symbolic significance of the division between the sexes is premised on the notion that physiological differences (’natural’ differences) are universally transformed into cultural inequality. The ’universal asymmetry of the sexes’ is seen as a means by which women are classified as inferior and excluded from the exercise of power in society. The study of sexual differentiation has often been confused with the study of women with the result that perceptions of the inferiority of women have colored both discussions of the symbolism of sexual differentiation (only women are perceived as defined in ’biological’ terms) and the relationship between such symbolism and the allocating of social roles (all men are seen as dominating women). (1978:6) Along with the conceptual vagueness and the inability to deal analytically with cultural variation, what this literature seems to be missing most is a sense of cultural context and, with this, a sense of the diversity and depth of cultural meanings. How much do we learn about female 44 status when we view women primarily in terms of their biological role; when we assume that every culture views women in terms of biology? How much do we learn about women when we view them as analytically separate from men, indeed, separable from culture itself? Finally, what do we really learn about male/female relations when the analysis begins with apriori assumptions about universal male dominance or female subordination? As this chapter has tried to show, the literature on female status and the origins of sexual asymmetry consti- tutes one thematic segment of women-focused studies in anthropology. Underlying these studies is the belief that prior neglect of women in anthropological research has prevented us from fully comprehending women’s lives and the nature of their contribution to society. Assumptions about male bias are a pervasive feature in feminist anthropology and have clearly influenced the direction that research on women takes. These assumptions also constitute a justifi- cation for focusing only on women. The following chapter looks at the issue of male bias as the second thematic segment of women-focused research. 1. 3. 45 NOTES TO CHAPTER TWO See in particular discussions by Biersack 1985, Poole 1981, 1984, and M. Strathern 1980, 1984. In her later work, Sanday (1981) shifts away from the concept of ’female status’ to a focus on female power and male dominance -- equally ill-defined concepts. Influenced by Ruth Benedict’s (1934) Battern§_2f_£ulture and Margaret Mead’s (1935) Sex eng Tempermehh in Thgee Erimihive Societies, she suggests that each culture must select a "sex-role plan”, a ”template for the organiza- tion of sex-role expectations" (1981:3). Accordingly she argues that the power of women may be diminished as new metaphors for sexual identities replace old (as new ”sex-role plans" replace old sex-role plans) and men gain advantages from increased access to strategic re- sources. This is essentially the same thesis she pre- sents in her work on female status, and while the model she employs for her cross-cultural survey of "sex- role plans" appears more refined than her earlier work, she nonetheless ends up with the same kind of nebulous conclusion : Power is accorded to whichever sex is thought to embody or be in touch with the forces upon which people depend for their perceived needs. Concerning power in this way, one can say that in some societies women have more power, or men have more, or both sexes have an approx- imately equal amount. (1981:11) See for example, Leacock 1981, 1983, Ortner 1974, Sacks 1976, 1979, Sanday 1973, 1974, Rosaldo 1974, 1980, and Weiner 1976, 1979, 1980. CHAPTER THREE MALE BIAS AND THE ANTHROPOLOGY OF WOMEN As has already been noted, the issue of male bias is a pervasive theme in much of the women-focused anthropology. Yet it is not always clear what is being referred to, as anthropologists writing about male bias move back and forth between accusations of male bias within anthropology (i.e. the male anthropologist’s perspective), and general statements assessing the degree of male bias culturally expressed toward women in a particular society.1 As with notions such as ’female status’ and ’male dominance’, male bias is particularly difficult to define cross-culturally and tends to rely on the same kind of broadly based criteria leading to similar "some societies are, some socieities aren’t, some are more so than others" conclusions.2 Accompanying this conceptual vaguery is an assumption that there exists a special (somewhat mystical) bond shared by female anthropologists and the women they study. This common bond of womanhood is often invoked as a measure of ethnographic credibility.3 As M. Strathern (1981) notes in her critique of women-focused anthropology, the gender of 46 47 the male ethnographer proves to be a liability in that it is seen to contribute to a biased, culture-bound perspec- tive - The gender of the female ethnographer, however, enhances her work by providing her with a unique ’natural’ insight into her subject -- a "double-consciousness” (M. Strathern 1981:670) . Hence the assumption that women constitute a distinct analytic category that can be compre- hended only by taking up a woman’s point of view; and that when women study women the interpretation is somehow more ' authentic' . One of the earliest (and one of the most influential) formulations of this approach to the study of women can be to\ll'ici in the work of Edwin Ardener (1972) . Addressing the q“'Iestion of ’male bias’ in ethnographic reporting, he aITQWJed that at issue is not really the status or position <3 :6 women but the "problem" women present to anthropolo- gists. He suggests that both male and female ethnographers get‘lerally accept and interpret male models of the societies they study ”because the men consistently tend, when bl‘essed, to give a bounded model of society such as chnographers are attracted to" (1972:2) . The frequency of the ethnographer’s reliance on male informants, pat‘ticularly for translation and linguistic assistance, influences the ethnographer’s interpretation of the data, find ultimately the understanding of the society. T0 Ardener, this is a technical problem : the difficulty 48 i n ethnographically dealing with women, which leads to an analytic problem : if the models of a society that anthro- pologists present are derived from the male portion of the society, how does the ’other half’, the female portion of the society, perceive, interpret, and express the world in which they live? He suggests that women’s models of their society are different from those held by men; that women do 1101: perceive society as bounded from nature (as presummably men do) . Because their models are not organized along the same lines as men’s, women’s models are less acceptable to etit‘l'ltngraphers. In effect, Ardener states, women lack the I 'meta-laguage" to discuss their society. The idea that women may be culturally inarticulate, in t he sense that they lack the skills possessed by men to deseribe their society and express themselves, has farecguently been invoked as a partial explanation for diffi- Q‘lzlties anthropologists may have in eliciting information tron women in the field. But other cultural factors are involved as well, not least of which is the gender of the anthropologists and the nature of the information being ac"Light. To move from what are some of the inherent §Qhiplexities of conducting field research in a cross- Q“llltural context, to the general claim that men and women thd categorically different models of their society igl'xores these factors and provides a rather superficial justification for the further analytic isolation of women. 49 The suggestion that women actually perceive their world differently from men gained considerable ground in the 19708 with feminist- oriented anthropologists.4 Indeed, Ardener himself added a further refinement to this idea when, several years after first suggesting that women lacked the meta-language to discuss their social worlds, he cox-1c luded that women constitute a relatively prominent eXample of a "muted group" which, like other marginal groups within society, lack access and skills to manipulate the symbols of the dominant culture (E. Ardener 1977) . While Ardener’s argument is intriguing, it presents a 1:‘a'tller simplistic interpretation of the relationship bet‘ween the ethnographic process and the cultural context ‘1th which this process is superimposed. Furthermore, by taking gender as the primary factor such an approach Iueglects the numerous other social facts that enter into the formulation of individual’s perceptions of the world atI<>und them. There remain questions as to whether one “Quan’s model of her society would agree or be consistent V“'jL‘th that of another woman from the same society. Who are the people being asked to comment on their society? SQ1.1citing a model of society from a New Guinea big-man -- thse reputation is based on oratorical skill and ability ‘z‘3 command attention and influence over people both within hr1d outside his own society -- and soliciting a model of the same society from a woman who is neither the wife nor 50 the daughter of a big-man could easily produce two very dissimilar interpretations of a shared social world. The Same holds for old and young (of either sex), as well as those who are perceived by others in their society as successful and those considered failures. The reflections of Roger Keesing (1985) on his earlier failure and later success in obtaining life history material on Kwaio (Solomon Islands) women provides an interesting perspective on this problem; one which points more to the nature of the anthropological life history than ‘10 assumptions about contrasting ways women and men may Perceive their social worlds. Having recorded the life hj-s‘U—Tcary of a leading Kwaio feastgiver (see Keesing 1978), ReeSing then tried to obtain a parallel account from a Kwaio woman (of middle age, who also happened to be the da“lgliter of another important Kwaio' feastgiver) . As he states, "little came of it", the interview sessions were brief and the woman was easily distracted, frequently it1"3-‘Lting men to join the discussion and provide their own aczcc>unts of events. The experience led Keesing to conclude that perhaps Kwaio women, like other women throughout the tribal world, are "relatively mute about themselves and t11%er place in their cultural tradition" (1985:30) . liq"Wever, several years later, accompanied by a female eQI league, he succeeds in obtaining numerous richly ti eta iled self-accounts of Kwaio women . While he 51 acknowledges that his long-time friendship with his informants, along with the presence of his female Colleague, certainly helped in the process of interviewing the women, he suggests that this alone is not enough to account for the earlier failure and later success. Rather, he argues that when dealing with personal accounts of the lives of non-Western peoples we must constantly remind ourselves that they are likely to reflect very different folk models of self and person than those we take for granted. According to Keesing, more emphasis should be Placed on what he terms ”the context of elicitation" (1985:31) which goes beyond the immediate context of ac=‘t111ally obtaining life history material from a native informant to include the historical context of the society itSelf. Profound change continued to confront Kwaio s'l’chety in the intervening years between his first and se':3-<>nd attempts at gathering female life histories. He cites the Kwaio struggle for autonomy and the "elevation of 'culture’ as a political symbol” (1985:37) as fundamentally atIfecnzing Kwaio models of themselves and the world around the“. He suggests that the accounts Kwaio women gave of the ir culture can only be understood in the historical Qthext of colonial domination and the dramatic changes that occurred in Kwaio culture in general and women’s lives in particular. Related to this, he also suggests that pQt‘haps the many cultural accounts constructed for e‘bl'lriographers by male informants should similarly be viewed 52 as "artifacts of the historical context of colonial domination" (1985:37) .5 To Keesing, then, the question is not whether women constitute a ’muted group’, or whether women's view of their culture is less ’global' than men’s, or even whether the colonial experience widened women's perspectives (and thus their expressiveness) by enlarging the range of their Participation in things both within and outside their societies. Rather, he asks, what is it about Kwaio society that made it possible at this later date to obtain numerous detailed self-accounts of Kwaio women? And what, then, do these accounts tell us about Kwaio society? The point to be made here is that self-accounts are by hatitslznre subjective and individuals from the same society are likely to differ in their personal perceptions; men may dif fer from other men, women may differ from other women, and certainly men may differ from women. The idea that individuals from the same society might hold significantly d“'Lfrferent cognitive interpretations of their social world is not new anthropology: attention has long been given to the ways that an individual's social and temporal location within the economic, political, and ritual structures of a 3°<-‘-:i.ety influence their sense of place vis-a-vis other Iuglluaersfi Focusing only on the apparent contrast between are:Lative expressive skills of men and women too easily 53 suggests that such differences are reflective of more flmdamental differences within society and that these differences are indigenously framed in terms of gender. Within such a perspective, the ’inability’ of female informants to express themselves and reflect on their culture in a manner comparable to male informants is interpreted as further evidence of their subordination. Although few anthropologists have pursued Ardener’s notion that women constitute a ’muted group’ , the idea that “Cullen constitute a distinct and separate analytic category rema ins a prevalent theme, and many of the assumptions al¢><311t ’women’s models’ continue to inform both theoretical Works and ethnographic studies focused on women. Ardener’s idea that women do not perceive their social world as ho‘-ll'1