Gender, Property, and the State 247
The relationship between sexual inequality, the emergence of class structures, and the rise of the state have been enduring interests in anthropological studies of gender. The subordination of women appears to emerge as an aspect of state formation. According to Gailey (1987:6), "Institutionalized gender hierarchy ... is created historically with class relations and state formative processes, whether these emerge independently, through colonization, or indirectly through capital penetration." We are led to ask what relationship class and state formation have with the oppression of women and, when gender hierarchy occurs, why women are the dominated gender. How does the state have power to penetrate and reorganize the lives of its members, whether in Sumerian legal codes declaring monogamy for women or in welfare laws in the United States that influence household composition? Eventually, studying state formation may help us understand the origins and interrelationships of class and patriarchy and the social reproduction of inequality.
Much discussion of this subject has centered = on Engels' book The Origin of the Family, Private Property, and the State, a nineteenth-century text in which Engels argues that the emergence of the concept of private property and its owner-: ship by men, as well as the development of a monogamous family, led to the subordination of women. In Engels' scheme, prior to this gender relations were characterized as egalitarian and complementary. All production was for use, and people worked together for the communal household. Thus, changes in gender relations were linked to changes in material conditions because the ownership of productive property (initially domestic animals) was concentrated in the hands of men. This thesis has been influential in many Marxist and feminist analyses of women's subordination. For example, Leacock notes that "there is sufficient evidence at hand to support in its broad outlines Engels' argument that the position of women relative to men deteriorated with the advent of class society" (Lea-cock1973:30).
Following Engels Leacock observes that in early communal society the division of labor between the sexes was reciprocal, and a wife and her children were not dependent on the husband. Further, "the distinction did not exist between a public world of men's work and a private world of women's household service. The |arge collective household was the community, and within it both sexes worked to produce the goods necessary for livelihood" (Leacock 1973:33). In this view the oppression of women was built on the transformation of goods for use into commodities for exchange; the exploitation of workers and of women was generated by this process, which involved the emergence of the individual family as an isolated unit, economically responsible for its members, and of women's labor as a private service in the context of the family. This led to the "world historical defeat of the female sex" (Engels 1973:120; Silverblatt 1988:430).
In a reanalysis of Engels Sacks agrees that women's position declined with the elaboration of social classes but disputes Engels' emphasis on the role of private property in this process (Silverblatt 1988:435). Rather, Sacks links state formation and the decline in the centrality of kinship groups to the deterioration in women's status. She delineates two relationships defining women in noncapitalistsocieties: sisterhood and wifehood. "Sister" refers to women's access to resources based on membership in a kin group. This relation implies autonomy, adulthood, and possible gender symmetry. "Wife," on the other hand, refers to a relationship of dependency on the husband and his kin. Sacks suggests that the development of states undermined women's status by dismantling the kin group corporations that formed the basis for sister relations (Sacks 1982).
However, critics of Sacks' position have argued that the process of state formation may involve uneven and contradictory developments. For example, elite women in the Kingdom of Dahomey, West Africa, challenged state imperatives by means of control of marketing associations. Thus, Silverblatt rebuts Sacks' evolutionary paradigm, suggesting the inevitability of the decline in women's status with state formation; instead, she suggests that we acknowledge the
complex history of the emergence of elite privilege in the rise of the state. In addition, elite women such as the royalty of Dahomey may or may not share the goals of peasant women. They may instead join with male elite in suppressing the authority and power of peasant market women. This illustrates the potential contradictions of gender affiliation on one hand, and class position on the other.
Several anthropologists have observed that Engels lacked reliable ethnographic information and oversimplified the complexities of gender relations in kinship societies and in precapitalist states (Gailey 1987:15). In a critique of Engels Moore argues against his essentialist assumptions that there is a "natural" division of labor in which men are concerned with productive tasks and women with domestic ones. She also disagrees that an inevitable relationship will transpire between property, paternity, and legitimacy in which men "naturally" want to transmit property to genetic offspring (Moore 1988: 47-48).
While Rapp (in this book) agrees with these criticisms, she points out that Engels addresses many current concerns, such as the relationship between women's participation in production and female status and the implications of the separation of the domestic-public domains for women's roles in society (see also Silverblatt 1988:432).
She warns against overgeneralizing when trying to understand the origins of the state and ignoring the history and context in which political formations change. In her view we must examine kinship structures that were supplanted with the rise of the state and replaced by territorial and class specific politics. Kinship domains, formerly autonomous, were subjugated to the demands of emergent elites with repercussions for gender roles and relations.
Among the processes that we need to examine are the politics of kinship, the intensification of military complexes, the impact of trade on social stratification, and the changing content and role of cosmology. For example, in stratified societies the establishment of long-distance trade and tribute systems may affect elite marriage patterns, leading to the emergence of dowry and the exchange of women to cement male
246
248    Gender, Property, and the State
political alliances. Similar alliances are forged in societies that have experienced a rise in militarism. When economies change and demand increases for a traded commodity, exploitation of labor to produce it may also increase, and marriage systems that expand trading relations may be solidified through the exchange of women. Women themselves may be important figures in trading networks, as in West Africa and in Mesoamerica, where women are active in the marketplace. Alternatively, the extension of the state into localized communities can significantly influence gender roles and relations. This occurs, for example, by undermining women's ritual responsibilities (see Mathews, this book} or by generating conflict between government policy and the respective interests of men and women (see Browner, this book). Finally, changes in political hierarchies were legitimated by cosmological explanations in early states such as the Inca and the Maya. For example, as Maya society became increasingly stratified, a category of elite rulers known as ahaw emerged. These rulers were legitimized by myths that established them as mediators between the natural and supernatural worlds and as protectors of the people. State ideology involved the celebration of both male and female forces and the elite contained both men and women. Mothers of kings were always members of the high elite, and this class affiliation was also validated in myth (Scheie and Freidel 1990; Freidel and Scheie, this book).
In this book, Ortner also examines the process of state formation, with particular regard to its effect on gender ideology. She analyzes the widespread ideology that associates the purity of women with the honor and status of their families. This pattern is evident in Latin America and the Mediterranean and in societies of the Middle East, India, and China. Broad similarities exist in these varied societies. Ortner questions why the control of female sexual purity is such a ubiquitous and important phenomenon. She notes that all modern cases of societies concerned with female purity occur in states or systems with highly developed stratification, and they bear the cultural ideologies and religions that were part of the emergence of these states.
She argues that no prestaté societies manifest the pattern linking female virginity and chastity to the social honor of the group. Thus, concern with the purity of women was, in Ortner's view structurally, functionally, and symbolically linked to the historical emergence of state structures.
The rise of the state heralds a radical shift in ideology and practice, with the emergence of the patriarchal extended family in which the senior man has absolute authority over everyone in the household. Women are brought under direct control of men in their natal families and later by their husbands and affinal kin. Ideologically women are thought to be in danger, requiring male protection; they are idealized as mothers and for their purity.
One of the central questions in Ortner's discussion is the role of hypergamy (up-stalus marriage, usually between higher-status men and lower-status women) in state systems. Ortner suggests that one of the significant developments in stratified society involves the transformation through marriage from an essentially equal transaction to a potentially vertical one, where one's sister or daughter could presumably marry into a higher strata (wife of a nobleman, consort of a king). Hypergamy may help to explain the ideal of female purity because concepts of purity and virginity may serve to symbolize the value of a girl for a higher-status spouse. Thus "a virgin is an elite female among females, withheld, untouched, exclusive" (Ort-ner1981:32).
Hypergamous marriages often involve the exchange of significant amounts of property, particularly in the form of dowry. The relationship between dowry, inheritance, and female status has been explored in a number of societies with varying marriage patterns. Dowry has been described as a form of premortem inheritance, parallel to men's rights in property accrued through inheritance after death of parents or other legators. However, McCreery (in this book) argues that considering dowry as a form of inheritance prior to death and as part of a woman's property complex obscures an important difference between the clear legal inheritance rights that men possess and the dowry that women may or may
not receive. Women obtain dowry at the discretion of their parents or brothers, and dowry is not based on the same rights as other forms of inheritance to which men have access. McCreery uses the cases of China, India, and Ceylon (Sri Lanka) to examine women's property rights and how they relate to dowry. Chinese jaw more heavily emphasized the family as a corporate group of men related through patrilineal descent and consequently severely restricted the rights of women. Indian law gave greater weight to individual rights, including those of women. Finally, in Ceylon both men and women could own and dispose of private property.
in general, it seems that dowry is not best understood as a form of premortem inheritance to the bride; in Ceylon and China especially the legal bases of dowry and inheritance are distinct. Dowries appear to serve as a means of social mobility in stratified societies in which rights over women are used in competition for higher status.
While dowry has been viewed as a form of inheritance for women, the dowry system in northern India has taken a pernicious turn as brides are burned to death, poisoned, or otherwise "accidentally" killed by husbands and inlaws who believe that the women have brought inadequate dowries. Registered cases of dowry deaths in India in 1987 numbered 1,786, although women's groups contend that the real number is higher. Legislation banning dowries has been ineffective in preventing these murders, and the dowry system continues to be deeply embedded in local culture. Some feminist critics argue that dowry deaths represent a response to a growing materialist consumerism sweeping India that has stimulated demands for larger and ever-increasing dowries.
While the status of Indian women as reflected in the exchange of property at marriage appears to be deteriorating, in other parts of the world women's legal status and property rights have improved through state-sponsored changes in the judicial system. For example, Starr (in this book) examines how in Turkey the impact of capitalistic agriculture and settled village life profoundly affected gender relations
Gender, Property, and the State 249
and women's access to property. These changes occurred in the cultural context of Islamic notions of male dominance and female submission and legislation giving equal rights to women.
Ottoman family law gave women rights to divorce and some protection against polygamous marriage, but inheritance practices remained constrained by Islamic law, in which women were under the authority of their husbands, had the status of a minor, received half the share of the patrimony obtained by their brothers, and had no rights to children of a marriage. According to Ataturk's secular reforms in the 1920s, women's rights were expanded to full adult status, equal rights to paternal inheritance, protection for widows, and other legislation promoting equality for women. At first these state-initiated rights were incongruent with cultural practices, but by the late 1960s women had begun to use the courts to protect their rights to property and their reputation. Starr points out that while Eng-els emphasized the negative impact of private ownership on women's status, in this case law and culture played a positive counterbalancing role resulting in women's emancipation.
In all the works included in this chapter of the book we see the enduring influence of the issues raised by Engels with regard to the relationship between gender, property rights, and state structures. However, cross-cultural data demonstrate that this relationship is much more complex and varied than Engels' original formulation. Equally, universal evolutionary paradigms that posit a uniform impact of the rise of the state on gender roles cannot do justice to the myriad ways in which specific cultural histories, diverse social hierarchies, and systems of stratification affect gender relations and ideology. Thus, Silverblatt (1988:448) asks, "What of the challenges that women and men, caught in their society's contradictions, bring to the dominant order of chiefs and castes, an order they contour and subvert, even as they are contained by it? And what of the other voices, the voices that chiefs and rulers do not (or cannot or will not) express?" The complex histories of the relationship between any particular state and gender relations in it show that while state formation has contributed to the definition of womanhood,
250
Gender, Property, and the State
women have also contributed to the definition of states (Silverblatt 7 988:452).
REFERENCES
Engels, Frederick. 1973. The Origin of the Family, Private Property and the State. New York: International Publishers.
Gailey, Christine Ward. 1987. Kinship to Kingship: Gender Hierarchy and State Formation in the Tongan Islands. Austin: University of Texas Press.
Leacock, Eleanor Burke. 1973. Introduction. In Frederick Engels (ed.). The Origin of the Family, Private Properly and the State, pp. 7-57. NewYork: International Publishers.
Rayna Rapp
The recent wave of feminism lias prompted many anthropologists concerned with the status of women to return to Engels' work, The Origin of the Family, Private Properly, and the State. While anthropologists vary widely in their assessment of the autonomy of women in prestate societies, there seems to be a general consensus that with the rise of civilization, women as a social category were increasingly subjugated to the male heads of their households. That is, civilizations are properly described as patriarchal. This consensus is based, implicitly or explicitly, on the tradition in which Engels was writing. His schema links the growth of private productive property to the dismantling of a communal kinship base in a prestate society. In this process, marriage
Frnm Dialectical Antlirupology 2:309-316, 1977. Reprinted by permission ol'KUiwer Academic Publishers.
Moore, Henrietta L 1988. Feminism and Anthropology. Minneapolis: University of Minnesota Press.
Ortner, Sherry, 1981. Gender and Sexuality in Hierarchical Societies: The Case of Polynesia and" Some Comparative Implications. In Sherry B. Orty ner and Harriet Whitehead (eds.). Sexual Mean-I ings: The Cultural Construction of Gender and Sex^ uality, pp. 359-410. Cambridge: Cambridge" University Press.
5acks, Karen. 1982. Sisters and Wives: The Past and! Future of Sexuai Equality. Urbana: University of ]|.; linois Press.
Scheie, Linda and David Freidel. 1990. A Forest ort Kings. NewYork: William Morrow. f
Silverblatt, Irene. 1988. Women in States. Annual Review of Anthropology 17:427-461. ..;
An Archaeology of Knowledge Concerning the Origin of the State 251
grows more restrictive, legitimacy of heirs more important, and wires generally become means of reproduction to their husbands. At the same time, reciprocal relations amongst kinsfolk are curtailed, unequal access to strategic productive resources gradually develops, and estates or classes arise out of formerly kin-based social organizations. In this analysis, the creation of a class hierarchy is intimately linked to the creation of the patriarchal family. Restrictions on women's autonomy emerge with class society.
Engels' analysis informs many of the themes concerning women that are currently being investigated, such as the relation between productive contribution, control over distribution ol products, and female status in society; the relation of the mode of reproduction to the mode of production; and the effects of the relative separation or merger of
the domestic and public spheres of activity on the roles available to women.! Consciously or n0t, our questions are often framed within the general territory mapped in Origins. Yet the major problematic of the book—the postulated intertwined origins of class oppression and gender oppression—has barely been examined. Engels was working with a paucity of ethnological, archaeological and historical cases; he lacked data on primär)' or pristine state formation, and knew only minimally of highly stratified prestate societies. We, however, have access to a great deal more information concerning societies in transition between primitivity and civilization with which to examine his theory.J
Twentieth-century archaeologists and social theorists have amassed a substantive and growing body of theory and data concerning state origins to which we can, in principle, turn. Yet when we do, we discover that our question is certainly not theirs. We face a double theoretical problem in using their models. First: their search for "prime mover" explanations severely limits the questions that are being asked. There exists within evolutionär)' anthropology a strongly entrenched tradition which seeks to universalize its explanations. Such theories tend to be extremely reductive, and often condense a multiplicity of processes into unilinear variables. These theories range from a concentration on the extraction of social surplus via increasing division of labor and more productive technology (Childe, 1950, 1952); to a focus on the social-contract necessities of hydraulic societies (Wittfogel, 1955, 1957); to status-limitation, perhaps linked to population pressure (Fried, 1960, 1967); to the effects of population pressure and warfare within circumscribed environments (Carneiro, 1970, Harnei-, 1970). In the search for universal prime movers, the concern of these theorists has been extremely retrospective. They tend to see the state as an inevitable and efficient solution to a particular set of problems. But when the evolution of the state is viewed as a unilincal success story, we lose the specificity of history. A plethora of ranked chiefdoms, proto-states, city-states, feudal domains, empires and even national
states have perished over a span of millenia during which the political apparatus we now identify as "the state" evolved. In over-generalizing, we ignore history, and the context in which political formations change.1 This leads to a second theoretical problem. For political structures in the primitive world both arise from, and encounter resistance within the kinship base that organizes prestate societies.'1 Not uncoincidentally, it is within the kinship domain that women's subordination appears to occur. Yet kinship barely exists in prime mover theories, except as a backdrop to the progress of the growing state. Kinship structures were the great losers in the civilizational process, and they must be examined if we are to understand both incipient class formation, and the changing domains of women. It seems to me that with the rise of state structures, kinbased forms of organization were curtailed, sapped of their legitimacy and autonomy in favor of the evolving sphere of territorial and class-specific politics. Emergent elites needed access to the primary resources of kin-based groups, especially, their labor power. Formerly autonomous kinship domains were domesticated to increased, institutional demands of production and distribution. What was once the realm of total social reproduction got stripped and transformed to underwrite the existence of more powerful, politicized domains. In the process, not only kinship, but women lost out. This is the process we need to examine.
In recent years, many archaeologists have been inclined to reject prime mover explanations for social stratification in favor of more processual, systemic ones. The refutation of the hydraulic imperative and population pressure models on both theoretical and empirical grounds should make us wary of overly-simple explanations which pump causes out of correlations.' The "new archaeology" often uses ethnographic models for hypothesis building.'1 In the process, there are some fascinating hints about the role of kinship structure, and possibly women, in stratification. Some of the processes examined in state formation, such as the politics oi kin-
252    Gender, Property, and the State
ship, the changing content and role of cosmo-logical systems, the intensification of military complexes, and the role of trade in stimulating or increasing social stratification concern our question directly. I will discuss each of these in schematic form, suggesting some of the lines of inquiry they direct us to pursue.
(1) The politics of kinship: The conical clan floats through the literature, bringing with it an increasing tension in alliance, descent, and the transmission of status positions. Adams awards it an ambiguous and critical role in both Mesoamerica and Mesopotamia, and many anthropologists have examined the internal tensions of ranked kinship systems (Kirchoff, 1959; Sahlins, 1958, 1963; Fried, 1967). In highly ranked kinship systems, the role of women is of crucial importance. They not only transmit status, but may be contenders for leadership positions either directly, or through their children. This seems to be the case in Polynesia and in parts of Africa. As Gailey shows for Tonga (1978), the existence of elite, ranked women became more problematic as stratification increased. We need to know more about marriage patterns in such systems. In archaeology, ethnology, and Western history, we find that elite marriages may be implicated in the politics of establishing and maintaining longdistance trade and tribute systems.' Dowry is associated with highly stratified systems and dowered, elite women may appear as pawns in a classic case of male alliances formed via their exchange/ Orlner (1976) suggests that in state-organized systems, marriage may shift from a horizontal to a potentially vertical transaction, with a tendency toward hvpergumy through which elite women accumulate at the top of the system. These structural properties of the marriage system she links with ideologies requiring sexual purity and protection of women (but not of men). Silverblatt, in a study oflnca elite marriages, suggests that as the Incas extended their rule, they used an indigenous, pre-existing system of male/female ritual domains. These were ranked as male-conqueror/female-conquered. Conquered ayllus, identified as female and potential wives of the
Incas, had to send women to Cuzco to senc in the temples, courts, and as noblemen's wives. For the conquered community, this practice represented a loss of autonomy in marriage patterns, and a burden; at the same time, it made possible upward mobility to the specific males who sent sisters and daughters to Cuzco, separating them out in prestige from the collectivity. The women themselves demonstrated the same sort of ambivalence: they gained a great deal of prestige, but lost any autonomy they might have had in arranging their own marriages and living within their natal communities (1976).
(2) Changing Cosmologies: Both Silverblatt and Ortner aid us to focus on the relation of women to cosmological systejns in early states. Along with Wiley (1962, 1971), Eliade (1960), and Adams (1966), they remind us that religious systems were the j>lue that cemented social relations in archaic societies. Such systems were used, as in the Incaic case, to underwrite and justify changes in political hierarchies. They exhibit signs of ideological warfare on the battleground between local and elite pantheons. Tension about female status is often found within them as well. Fliacle claims that the ritual expression of sexual antagonism and the existence ol bisexual and/or androgynous gods accompanies the social organizational changes associated with the neolithic. Female and androgynous gods are often subsumed, or covered up, by ascendant male figures. This cover-up is sometimes performed indigenously, and often the work of later missionaries. This sort of layering over female and/or androgynous figures has been uncovered in Mesoamerica (Nash, 1976), Peru (Silverblatt, 1976) and is, of course, a favorite theme of classicists for early Greek society (Arthur, 1976; Pomeroy, 1974). Pagels (1976), working with 2nd century A.D. Gnostic texts, analyzes the symbol system of sects in which the early Christian god was bisexual, and the Holy Family consisted of mother, father, and son. Such sects were organized into non-hierarchical religious communities, in which offices were rotated, and women participated in both teaching and preaching—a far cry from the
An Archaeology of Knowledge Concerning the Origin of the State 253
ascendant Christian cosmology and practice which became the mainstream tradition. Cosmologies have histories; by peeling away their layers we may learn about how the ascendant estate legitimated itself. Cosmological changes are ideological precipitates of structural tensions; it is clear that their form and content have a great deal to tell us about class and gender.
(3) Intensification of Warfare: As early states became increasingly militaristic, social organization was transformed.''1 We need to know what happened to kinship structure, inheritance and succession in the process. Several case histories lead us to believe that under conditions of intense warfare, men were not only burdened by conscription, but were blessed with increasing power as household heads—a kind of trade-off. Muller (1975) finds this to be the case in feudal Wales. Elite males may gain land, political domains, and alliance-forming wives in the process. Yet the recent sweeping statements about male supremacy and warfare (Harris and Divale, 1976) are overgeneralized. even for state-making groups. Some evidence seems to contradict the association of warfare and an elevation of male status at the expense of female autonomy. I lack data on pristine states, but in the ancient West, the correlation often works the other way: Spartan women held offices, controlled their own property and had a great deal of sexual freedom, allegedly because they kept society functioning while the men were at war. Moreover, producing soldiers was considered as important as training them. In Athens, women's access to public places and roles seems to have increased a great deal during times of warfare, especially the Peloponnesian Wars, during which they left their virtual purdah conditions. In Rome, during the Second Punic War, women gained in inheritance settlements and held public offices formerly closed to them (Pomeroy, 1974). In the medieval Franco-Germanic world, noble women attained approximate parity in politics and property-management during the eras of the most brutal military crisis. McNamara and Wemple argue that their relative autonomy is inversely
linked to the existence of any form of public power structure; to the extent that none exists, women are freer to control their inheritance and their marital destinies (1973). All these examples concern elite women only; we know very little about the effects of warfare on laboring women, who probably suffered then, as they do now. Nonetheless, it is not clear that warfare degrades women's status; the specific context within which military organization and practice occurs must be taken into account. Blood and gore clearly are not uni-versalizable variables in theorizing about women's subordination.
(4) Trade: the role of trade in increasing and/or spreading stratification is an intriguing one. Foreign trade goods may be spread in many ways, by middlemen, via migrations, through central trading centers, and in marriage exchanges, to name but a few. Recent literature tells us that the social relations of a production and distribution which undergircl trade may be relations of incipient stratification between classes and genders (Kohl, 1975; Wiley, 1974; Adams, 1974). We need to know who produces goods, who appropriates them, and who distributes them. As demand for a traded commodity increases, exploitation of labor to produce it may arise. Marriage systems may also get intensified to expand the reproduction of trading alliances. The ethnographic literature runs rampant with examples of polygyny to increase access to goods that wives make. There are also many instances of increasing class division linked to increasing bride-wealth. Sisters and daughters are sometimes portrayed as pawns in alliances that many into expanding trade networks.11 It may be women who are used to "prime the pump" of kinship reciprocity in such circumstances. But appropriation and distribution are not exclusively male functions. In Mesoamerica and the Andes to this day, women are active in the marketplace. Silverblatt (1976) suggests that they were important traders in early Incaic times; Adams records their existence in Mesopotamia as well (1966). And of course, their presence is felt in stratified groups throughout Africa and the Caribbean as well, in contemporary ac-
254    Gender, Property, and the State
An Archaeology of Knowledge Concerning the Origin of the State 255
counts (Mintz, 1971). Under what conditions does long-distance trade pass into the hands of men, and when is it possible for women to continue to perform it? When women are traders, do they constitute an elite, class-stratified group? To the extent that trade is implicated in the intensification of production for exchange, women as producers, reproducers, and traders must be implicated, too.
What I am suggesting in these examples is that structural tensions arise within kinship-based societies and that these need careful examination if we are to understand state-formation. I am making a plea for careful historical reconstructions, informed by a feminist perspective, rather than an overly mechanical and generalizing view which erases certain questions before they can be written.
As we come to identify the factors in state formation as they affect women, we must be careful to think in probablistic rather than deterministic terms. We need to contextual-ize the relative power of kinship and class, the interplay of domestic and extra-domestic economy, the flexibility within cosmological systems, and the relative autonomy or subordination of women, in light of the possibilities open to each society. We should expect to find variations within state-making (and unmaking) societies over time, and between such societies, rather than one simple pattern.
There is another arena to which we can turn in examining how the processes of state formation and penetration a fleet gender relations. The area we now rather euphemistically call the Third World has served as a bloody iaboratoiy in which stratification by culture, class and gender occurred. While obviously differing from pristine state formation, this process of rapid penetration by patriarchal national states allows us to see how the conditions of primitivity are shattered. In examining the colonization of the indigenous societies of the Americas, Asia, Africa, we need to be very cautious. Each area is heir to its own particular history, as is each colonizing power. Specific histories condition the patterns of resistance, acceptance, or modification of the social organization produced by colonization. We cannot subsume such com-
plexity into one simple model of precapitalist penetration.
Nonetheless, certain patterns affecting the ways of life of women can be traced at a general level. Wherever women have been active horticulturalists of collective lands, the imposition of private property, taxation, labor migration and cash cropping has had devastating effects. Their realm of productivity and expertise has been deformed and often destroyed (Boserup, 1970; Blumberg, 1976; Tinker, 1976). Depending upon context, they may become either super-exploited, or underemployed, but always more dependent upon men. The evidence also suggests a general pattern concerning political organization: prior to colonial penetration, indigenous cultures appear to have been organized into gender domains on essentially parallel lines. Men and women both had organizational forms and rituals which were conditioned by gender-linked relations of production and distribution.1- Case histories from Africa, Asia and the Americas suggest that pa-'-, triarchal, colonizing powers rather effectively dismantled native work organizations, political structures, and ritual contexts. The pro-,, cess of demobilizing women occurs when essentially parallel forms are subsumed into one, and that one is male. Leadership and authority are assigned to activities which are male, while female tasks and roles are devaluated, or obliterated. Van .Allen's work (1972) describes the political associations of Igbo women which survived the British freeze, and were effectively used to organize the 1929 tax riots. We have case histories recording the effects of assigning credit, technology and education to men such that traditionally female activities are drastically curtailed, or cease to exist. Female marketing is increasingly circumscribed by the influence of internationalization; its agents prefer to deal with men.'1 Several authors have gone so far as to argue that women thus divested of their social organization and collective roles have become rather like underdeveloped, monocrop regions. Once they lived in a diversified world; now they have been reduced to the role of reproducing and exporting labor power for the needs of the international world economy
(Bossen, 1973; Deere, 197(3; Boulding, 1975). please note: in these processes, class stratification as well as gender stratification is operating- Vast categories of men, too, lose autonomy in the broadest sense. My point is that men and women lose it differently, and their lives are transformed differently.
State formation and penetration is pro-cessual; its form and force are highly variable, both within and between societies. Yet it is important to remember that the processes which began millenia ago are ongoing. Cumulatively, they continue to transform the lives of the masses of people who exist under their structures. It is a long way from the Sumerian law codes declaring monogamy for women to the welfare laws of the United States which affect parental dyads and household structure. But in both cases, the power of the state to penetrate and reorganize the lives of its members is clear. As we seek to understand the complex, stratified societies in which we now operate, we are led to reflect on archaic societies in which the dual and inter-mined processes of hierarchy we now retrospectively label "class" and "patriarchy" took their origins.
NOTES
1. For analyses implicitly or explicitly influenced by Origins, see Brown, 1975; Sacks. 1 97-1; Reiler, 1975; Sanday, 1974; Rubin, 1975: Meillassoux, 1975.
2. Summaries of the state formation literature mav be found in Service, 1975; Flannerv, 1972; Webb, 1975; Kracler, 1968.
3. A less determinant and more processual perspective on stratification and state formation is sei forth in Fiannery, 1972; Tilly, 1975, and SablotT & Laniberg-Karlovsky, 1975. Such thinking also informs the corpus of Marxist historiography.
4. This has been a major thrust in S. Diamond's work (1951, 1974).
5. Adams. 1966, presents the most cogent critique of the hydraulic hypothesis. Population pressure models are rejected for Iran in Wright & Johnson, 1975, and for the Valley of Mexico by Brumfiel, 1976 and Parsons, 1974. See Cowgill, 1975 for a theoretical and methodological critique.
6. For instances of the "new archaeology's" use of ethnographic evidences for hypothesis building, see Fiannery, 1967; Renfrew, 1975, both volumes edited by Sabloff & Lambcrg-Karlovsky, 1974 and 1975. and Adams, 1966 and 1974.
7. Such links are suggested in Fiannery, 1967; Wiley & Shimkin. 1971. and by much of the literature on stratified chiefdonis, fur exam-pie, Sahlins, 1958; Kirclioif, 1959; and Fried, 1967. In European history, wc find instances in the international royal marriage patterns. See McNamara & Weniple, 1973, fur some of the implications of feudal marriage patterns.
8. See Arthur, 1976; Ortiier, 1976; Goody Sc Tambiah, 1973.
9. The temple and militatv-complexes figure in the state formation schemes of Steward, 1955; Adams, 1966, and Wiley, 1971, to name hut a few.
10. The archaeology of trade is discussed in Sabloff and Lamberg-Karlovsky, 1975, and is critically summarized in both Adams, 1974, and Kohl, 1975. See also Wiley, 1974.
[f. Ethnographic examples used lo speculate about ancient trade are provided by Adams (1974) who cites materials on the Cheyenne and Blackfoot, and Fiannery (1967) who uses Tlingit materials. In these cases, marriage alliances were intensified along with trade.
12. Parallel and interarticulating forms of gender social organization are analyzed in Siskind, 1976: Brown, 1970; Silverman. 1976; Van Alien, 1972. and Sacks, 1976.
13. See Mintz, 1971, Boserup, 1970, Tinker. 1976, and Blumberg, 1976 for specific examples. Agricultural schools throughout Africa often closed their doors to women, the traditional horticulturalists. Agricultural machinery for "women's work" often falls into the laps of men as the sole possessors of access to credit.
REFERENCES
Adams, R.M. (1966). The Evolution of Urban Society. Chicago: Aldine.
Adams, R.M. (1974). "Anthropological Perspectives on Ancient Trade," Currant Aiitliropology, Vol. 15.
Arthur, M. (1976). "Liberated Women: The Classical Era," in Bridenthal and Koontz (eds.), Becoming Visible: Women in European Histon: New-York: Houghton and Mifllin.
256    Gender, Property, and the State
The Virgin and The State 257
Blumberg, R. (1976). "Fairy Tales and Facts: Economy, Fertility, Family and the Female.
Boserup, E. (1970). Women's Role in Economic Development. London: George Allen and Unwin.
Bossen, L. (1973). "Women in Modernizing Societies. "American Ethnologist, Vol. 2.
Boulding, E. (1975). "Women, Bread and Babies". University of Colorado, Institute of Behavioral Sciences, Program on Research of General Social and Economic Development.
Brown, J. (1970). "A Note on the Economic Division of Labor by Sex." American Anthropologist, Vol. 72.
Brown, J. (1975). "Iroquois Women," in R. Reiter (ed.). Toward an Anthropology of Women. New York: Monthly Review Press.
Brumfiel, E. (1976). "Regional Growth lit die Valley of Mexico," in Flannery (ed.). The Early Mesoamerican Village, New York: Academic Press.
Carnciro, R. (1970). "A Theory of the Origin of the Svmc," Science, Vol. 169.
Childe, G. (1950). "The Urban Revolution," Town Planning Review, Vol. 21, No. 3.
Childe, G.'(1952). "The Birth of Civilization," Past andPresenl,Vo\.2,No. 1.
Cowgill, G. (1975). "Causes and Consequences of Ancient and Modern Population Changes," American Anthropologist Vol. 77.
Deere, C. (1976). "Rural Women's Subsistence Production in the Capitalist Periphery," Review oj Radical Political Economics, Vol. 8.
Diamond, S. (1951). Dahomey: A Prolostutv in West Africa. Ann Arbor: University Microfilms.
Diamond, S. (1974)./« Search of the Primitive: A Critique of Civilization, New Brunswick, N.J.: Transaction Books.
Eliacle, M. (1960). "Structures and Changes in the History of Religions," in Kraeling and Adams (eds.), City Invincible, Chicago: University of Chicago Press.
Flannery, K. (1972). "The Cultural Evolution of Civilizations, "Annual Review of Ecology and Sys-tematics. Vol. 3.
Planner)'. K. (1974). "The Olmec and the Valley of Oaxaca," reprinted in Sabloff and Lamberg-Karlovsky (eds.). The Rise and Fall of Civilizations. Menlo Park, Ca.: Cummings.
Fried, M. (1960). "On the Evolution of Social Stratification and the State," in S. Diamond (ed.), Culture and Histon. New York: Columbia University Press.
Fried. M. (1967). The Evolution of Pol idea I Society. New York: Random House.
Gailey, C. (1978). "Origins of the State in Tonga:
Gender Hierarchy and Class Formation," Dialectical Anthropology. Vol. 3.
Goody, J. and Tambiah, S. (1973). Bride-wealth and Down. Cambridge: Cambridge University Press. • •
Harner, M. (1970). "Population Pressure and the; Social Evolution of Agriculturalists," Southwestern journal of Anthropology, Vol. 26.
Han-is, M. and Divale, \V. (1976). "Population, Warfare and the Male Supremacist Complex," American Anthropologist, Vol. 78.
Kirchoff, P. (1959). "The Principles of Clanship Human Society," in Fried (ed.), Pentlings in Anthropology, Vol. 2. New York: Crowell.
Kohl, P. (1975). "The Archaeology ofTracle," Dialectical Anthropology, 1:43-50.
Kräder, L. (1968). Formation of the Stale. Engle-wood Cliffs: Prentice-Hall.
McNamara, J. and Wemple, S. (1973). "The Power of Women through the Family in Medieval Europe: 500-1) 00," Feminist Studies, Vol. 1.
Meillassoux, C. (1975). Fem ums, greniers el. caphaux. Paris: Maspero.
Mintz, S. (1971). "Men, Women, and Trade," Com-para live Studies in Society and History, Vol. 12.
Müller, V. (1975). "The Formation of the State and the Oppression of Women: England and Wales," manuscript.
Nash, J. (1976). "Aztec Women," manuscript.
Ortner, S. (1976). "The Virgin and the State," Michigan Papers in Anthropology. Vol. 2.
Pagels, E. (1976). "When did God Make Man in His Image?" Signs, Vol. 2:
Parsons, [. (1974). "The Development ofa Prehistoric Complex Society," fou rnal of Field A rchaeol-ogy. Vol. 1.
Pomeroy, S. (1974). Goddesses, Whores. Wives and. Slaves. New York: Schocken.
Reiter, R. (1975). "Men and Women in the South of France,"in R. Reker (ed.), Toward an Anthropology of Women. New York: Monthly Review Press.
Renfrew, C. (1975). "Trade as Action at a Distance," in Sabloff and Lamberg-Karlovsky (eds.), Ancient Civilizations anil Trade. Albuquerque, N.M.: University of New Mexico Press.
Rubin, G. (1975). "The Traffic in Women," in R. Reiter (ed.), Toward an Anthropology of Women. New York: Monthly Review Press.
Sabloff and Lamberg-Karlovsky (eds.). (1975). Ancient Civilizations and Trade. Albuquerque, N.M.: University of New Mexico Press.
Sacks, K. (1974). "Engels Revisited," in Rosaldo and Lamphere (eels.). Woman. Culture and Society. Stanford: Stanford University Press.
Sacks. K. (1976). "State Bias and Women's Status," American Anthropologist. Vol. 78.
Satiday, P. (1974). "Toward a Theory of the Status of Women," in Rosaldo and Lamphere (eds.), Woman, Culture and Society. Stanford: Stanford University Press.
Sahlins, M. (1958). Social Stratification in Polynesia. Seattle: University ofWashington Press.
Sahlins, M. (1963), "Poor Man. Rich Man, Big-Man, Chief," Comparative Studies in Society and Hillary, Vol. 5.
Service, E. (1975). Origins of the Stale and Civilization. New York: Norton.
Silverblatt, 1. (1976). "Inca Women: Conquered Moon, Conquering Sun," manuscript, Department of An thro po log)', University of Michigan.
Siskind, J- (1976). "Kinship: Relations of Production," manuscript.
Steward, J. (1955). The Theory of Culture Change. Urbana: University of Illinois Press.
Tilly, C. (1 975). Reflections on the History of European State Making," in Tilly (cel.), The Formation of National Stales in Western Europe. Princeton: Princeton University Press.
Tinker, I. (1976). "The Adverse Impact of Development on Women," in Tinker and Bramsen (eds.), Women and World Development. Overseas Development Council.
Sherry B. Ortner
In an extraordinarily wide range of societies in ilie world one Finds a peculiar "complex": ideologically it is held that the purity of the women reflects on the honor and status of their families; and the ideology- is enforced by systematic and often quite severe control of
Tliis article is rtpriiilecl (rum Feminist Studies, volume 4. numbers (October lil78): 19-35, bv permission of the publisher. Feminist Studies, Inc.. c/o Women's Studies Program, University of Mankind, College Park, MD 20742.
Van Allen, J. (1972). "Sitting on a Man," Canadian journal of African Studies, Vol. 6.
Webb, M. (1975). "The Flag Follows Trade," in Sabloff and Lamberg-Karlovsky (eds.). Ancient Civilizations and Trade. Albuquerque, N.M.: University of New Mexico Press.
Wiley, G. (1962). "The Early Great Styles and the Rise of Pre-Columbian Civilizations," American Anthropologist, Vol. 64.
Wiley, G. (1974). "Prccolunibian Urbanism," in Sabloff and Lamberg-Karlovsky (eds.). The Rise and Fall of Civilizations. Menlo Park, Ca.: Cummings.
Wiley, G. (1974). "Commentary on the Emergence <)l'Civilization in the Mara Lowlands," in Sabloff and Lamberg-Karlovsky (eds.), The Rise and Fall of Civilizations. Menlo Park, Ca.: Cummings.
Wiley, G. and Shimkin, D. (1971). "The Collapse of the Classic Maya Civilization in the Southern Lowlands," Southwestern journal of Anthropology, Vol. 27.
Wittfogel. K, (1955). "Oriental Society in Transition," Far Eastern Quarterly, Vol. 14, p. 469—178.
Wittfogel, K. (1957). Oriental Desjmtism. New Haven: Yale University Press.
Wright, H. and Johnson, G. (1975). "Population, Exchange and Early State Formation in Southwestern Iran," American Anthropologist. Vol. 77.
women's social and especially sexual behavior. One sees this pattern manifested among peasant societies in Latin America and around the entire Mediterranean area, among pastoral nomadic tribes of the Middle East and southwest Asia, among the castes of India, and among the elites of China. In extreme cases, such as classical Athens or among Brahmins of India, women were confined to the house for life. In imperial Turkey, the sultan had vast numbers of wives and
258    Gender, Property, and the State
The Virgin and The State 259
daughters "in an elaborately organized harem, or seraglio, with disciplinary and administrative officers, ruled over by [his] mother." Among poorer peasants and nomads, a variety ol" other devices—veils; rules of body-disguising dress and of modest demeanor; restrictions on expression, communication, and movement; all overseen by the family in particular and the gossip of the community in general—serve to restrict women's social and sexual behavior as effectively as if they were locked up.1
Reviewing the variety of cases, one tends to get involved in particular cultural symboliza-tions and practices, and to lose sight of the broad similarities of pattern. Further, the pattern does not seem to be confined to anv particular type of society, or to any consistent stratum: peasants and elites, agriculturalists and pastoral nomads, all seem to embrace some version of the female purity ethic with equal intensity and commitment. It seems difficult to imagine that there might be a single interpretation that would cover, or at least interrelate, all the cases.
In fact, in the anthropological analyses of particular cases, a variety of interpretive frameworks have been used: psychoanalytic, structural, functional, ecological, political, or some mixture of several of these. Let me begin to situate the question of whv the control of female sexual purity is such a widespread and virulent phenomenon, by reviewing briefly some of the major attempts at interpretation in the anthropological literature."
In an early essay from a psychoanalytic perspective, Kathleen Gough analyzed the female initiation rites among the Nayar and other groups of the Malabar region of India. Gough interpreted the rituals as signifying the formal renunciation by the girl's con-sanguineal kinsmen of rights to her sexuality, in a social context in which there is evidently strong incest temptation. Further, during these rites the young girl was actually or symbolically deflowered by a person other than her prospective husband; and Gough interpreted this point as representing male fear of the defloration of virgins in the course of nor-
mal sexuality. Although Nayar women do noi: seem to fit our model in that, following their initiation they had great sexual freedom, We must not forget that their freedom was gained at the expense of Nambudiri Brahmin women, who were subject to virtually total scs elusion and control. Untold numbers of Nambudiri women died virgins, while Nambudiri men mated with the conveniently available Nayar.;l ' :
In a subsequent counteressay, Nur Yalnian challenged (and ridiculed) Cough's interpretation and recast the whole argument in terms of the control of female purity. He stressed that the rites establish the purity of the women in their own castes and serve to define and regulate the women's subsequent choices of mates, who must always be of equal or higher caste status. The issue, he argued, is control of caste purity and status as a whole, which must be maintained by regulation of female sexuality, because regardless of the descent principles operating within intra kinship groupings, caste as opposed to kin af-filiation is always inherited bilaterally. Fur-ther, if caste membership is defined as coming only through one parent, that parent is always the mother. Thus for purposes of sustaining caste purity, the woman's purity in particular must be controlled, protected against pollution by lower-caste mates. Men, on the other hand, arc free to have sexual relations with anyone, "high or low" as Yalman> says, and he then explains this in both cultural and natural terms. Culturally, the Indians distinguish between internal and external pollution. Women are subject to internal pol- • lution in sexual intercourse, which is very:: hard if not impossible to cleanse, but men are subject only to external pollution in intercourse and can be cleansed by a simple ritual ~ bath. Yet Yalman goes further than this oil-; tural point, and relates the ideology to natural factors: "the bond between the genitor . and the child is tenuous; it can always be denied or minimized; the children can always be repudiated by the father."'
More recent studies have tended to get ; away from elusive unconscious factors and symbolic cultural notions of pure and impure,.:
and to stress the brass tacks of economics and politics. Lawrence Watson describes the practices of the Guajiro of Venezuela, among Ivhom the virginity of a girl upon marriage is absolutely demanded, the result being assured by a combination of psychological terrorism and physical punishment, mostly enacted by mothers on their daughters. For a serious offense, "the mother may place the tip of a hot branding iron on the girl's vagina to make the punishment a convincing object lesson." Watson casts his interpretation of this system in terms of the political structure of the society, in which every group is concerned about maintaining its status in a rigid class system; the group's status depends in part upon the quality of the women it can deliver in the marriage alliance process. Especially among the upper classes, influential chiefs overtly use marriage alliances of daughters and sisters as a way of building up fallowings for political and military backing. "If ... a woman causes her father or uncle to lose valuable political allies because of deficiencies in her sexual behavior, she becomes a liability and she can seriously impede her lineage's chances of building up a secure base of political support." In this interpretation, the sadistic control of female purity is simply a form of real poll HI;.''
And finally, Jane Schneider presents an argument in terms of ecological and economic factors, and the politics thereof. In her important paper, "Of Vigilance and Virgins . . . ", she begins with the general point that "honor can be thought of as the ideology of a property holding group which struggles to define, enlarge, and protect its patrimony in a competitive arena." She then goes on to argue that both pastoral and peasant societies tend to be highly socially fragmented and "un-solidary," although the reasons for this social fragmentation are different in the two cases. And according to Schneider, honor is the code that keeps this "centrifugal" situation together: it "helps shore up the identity of a group (a family or a lineage) and commit to it the loyalties of otherwise doubtful members. [It] defines the group's social boundaries, contributing to its defense against the claims
of equivalent competing groups. [It] is also important as a substitute for physical violence in the defense of economic interests. . . . Honor regulates affairs among men." But why the honor of the women ? Why should the women's honor represent the honor of the group as a whole? Because, says Schneider, resolutely practical to the end, among the pastoralists concerned with lineage continuity, female reproductive capacity is valuable, and women are "contested resources much like pastures and water." As for the agrarian peasants, the problem seems primarily to be the potential fragmentation of the family of procreation, with fathers, sons, and brothers set off against one another because of inheritance rules; here the daughters/sisters provide the one shared focus of concern that can hold the group together."
I will restrain my temptation to dissect the circularities and self-contradictions of many of these arguments and will simply say that, with the exception of the psychoanalytic argument, all of them share common functionalist orientations: the purity of women is seen as adaptive for the social coherence, economic viability, or cultural reputation of the group, regardless of whether the group is a caste, lineage, or family. When the theorists trv to explain why women in particular should represent the coherence and integrity of the group, rather than, say, a totemic bird or a sacred flute, the answers are more variable—in terms of women's natural childbearing abilities, women's physical structure (internal pollution), women's function as tokens of alliance, or women's symbolic roles in the family. None of these answers is very satisfactory; all use as explanations the very things that need explaining. We are still left with the paradox that male-defined structures represent themselves and conceptualize their unity and status through the purity of their women.
I would argue then that all of these explanations can be lumped together and that they share a set of common failings. First, all are static functional accounts, and lack time depth. Second, all share the common functionalist fallacy of reifying the unit under study and treating it as closed, exclusive, and
260    Gender, Property, and the State
isolated from a larger social context—the family, lineage, or caste is treated almost as a society in itself. And finally, all explanations, with the exception of the psychoanalytic discussion, take the point of view that the problem is one of male/male relations, in which the women are intermediaries, rather than the problem being, as it at least equally is, a problem of male/female relations. Although each of the arguments contains some useful kernel of truth, none provides a framework for encompassing and accounting for the phenomenon as a whole in cross-cultural, cross-class, and cross-sex perspective.
What I should like to do in this paper, then, is to offer some observations, thoughts, suggestions, and hypotheses for exploring this problem more systematically, and in a way that will illuminate problems of social and cultural process in general, as well as male/female relations in particular.
I would begin by noting that all of the modern cases of societies concerned with female purity are in fact of a certain type, namely, that all are part of, or have historically been part of, states, or at least systems with fairly highly developed stratification.' Thus hyperpure Brahmins and hyperpoor Mediterranean peasants share the status of being part-structures, elements in larger stratified political structures. Even when the larger state structures in which they originally developed are no longer organically intact, all of the modern groups in question bear the cultural ideologies, and particularly the religions, which were part of the organic emergence of their ancestral states in the first place. Most of the societies concerned with female purity are involved in so-called great traditions, especially Christianity, Islam, and Hinduism. And these religions evolved in conjunction with the emergence of states (or "civilizations," or empires) which, although most are no longer intact, nonetheless shaped the societies and cultures of the groups that bear their cultural heritages. And most of the peasants whose ancient states have decayed are now involved with modern states in ways that are structurally similar to their places in the original ones.
It is true that there are pre-state societies which, for example, require the virginity of women at marriage, and probably the majority of human societies expect relative sexual faithfulness of women after marriage. But no pre-state societies, as far as I have been able to ascertain, evince the sort of pattern I am concerned with here—the ideological linkage of female virginity and chastity to the social honor of the group, such chastity being secured by the exertion of direct control over women's mobility to the point of lifetime seclusion, and/or through severe socialization of fear and shame concerning sex." What I am suggesting, then, is that this sort of concern with the purity of women was part of, and somehow structurally, functionally, and symbolically bound up with, the historical emergence of systematically stratified state-type structures, in the evolution of human society.9
Before examining what the purity of women might have to do with the emergence-of states, however, let me sketch very briefly; what I see as the patterns and tendencies of female/male relations in pre-state societies, or rather in contemporary societies that have historically been outside of known state sys-: terns, and that have not themselves evolved the social, political, and economic characteristics of states. I would begin with the point that there is always, even in the most primitive of known societies, some sort of asymmetry between the sexes. Even the most manifestly egalitarian of band societies accords some edge of authority or charisma or status to men, if only on the view that the big game that the men bring home is superior as food to the women's gathered produce. In slightly more complex band societies, it seems that there is always some sacred center or ritual from which women are excluded. And although women may have their sacred ceremonies, from which men are excluded, the male ceremonies are considered to be for the welfare of the group as a whole, while the women's ceremonies are specific to the welfare of women. Finally, in the most complex of the known band societies, primarily in Australia, male authority is asserted through, and reinforced by systematic control of the mar-
riage system, the exchange of women and woods. Control of the marriage system, always in the hands of men, transforms diffuse authority or charisma into the beginnings of real power and control. Nonetheless, being bartered about in a system of marriage exchange is not the same thing as having one's day-today behavior and freedom of movement directly controlled, and in fact women in band societies evidently have a great deal of autonomy of action, as long as they comply with the legal rules of the game. There is also no ideology in these societies about protecting female purity.
If anything, the ideology is just the reverse, and women are often seen as, to some degree, dangerous and polluting. According to Mary Douglas, pollution beliefs are systematically related to cultural category ambiguities and anomalies;11' the danger and pollution of women would seem to derive in large part from the fact that women systematically appear as ambiguous vis-a-vis two very important, and partly related, category distinctions that may be common to all human societies. The first distinction is the nature/culture dichotomy, and I have discussed at length elsewhere woman's ambiguity vis-a-vis this opposition. 11 The second is the structure/ antistructure, or order/disorder dichotomy, in which men and male groups are identified with structure, order, social organization itself. Insofar as women are moved around in marriage, in a social exchange system controlled by, and culturally seen as composed of, structured groups of men, women appear interstitial within the fundamental kinship architecture of society.1" Further, the ambiguity of women would derive not only from a marriage perspective; insofar as there is descent ideology, whether patrilineal or matrilineal, women are seen as "in between" in these sorts of systems as well, for descent groups (such as clans) see themselves as groups of males, with women as their reproductive agents. With respect to either or both of these oppositions— naaire/culture, stmcture/antisu-ucture—women may appear ambiguous, and hence potentially polluting and dangerous. And although none of the simplest hunting/gathering socie-
The Virgin and The State 261
ties manifest the phobia about female pollution and danger that appears among, for example, New Guinea horticulturalists, most have a variety of taboos and avoidance rules that seem concerned with keeping at least some of the boundaries drawn.
Now between hunting/gathering band societies on the one hand, and states on the other, there is obviously a vast range of types of societies, of widely varying structure and complexity. I have yet to find or devise a classification scheme that organizes all of them in some satisfactory evolutionary sequence. I will plow right through all this complexity, however, and simply say that through all the types of pre-state societies, female/male relations stay broadly within the pattern established over the range of band societies—from relatively mutualistic and balanced, to the extreme cases of sex antagonism, with male self-segregation, and strong expressions of fear of women as dangerous. But again the expression of and reaction to fear of women in those extreme cases, and here I am thinking largely of New Guinea and South America, involves exclusion of women, or attempts thereof, rather than systematic domination and control. My image here is the New Guinea or South American village, with the men huddled in men's houses in the center off-limits to the women, and the women strung out around the periphery in their individual huts with their uterine families. Even in North America, a much less extreme area on this score, we find the male sweat lodges, the kivas into which women are not allowed, and so forth. But as long as the women do not trespass on the off-limits areas, they have considerable autonomy of action, and indeed a certain edge of power insofar as they can appropriate and judiciously imply control of some of the powers with which the male culture endows them. As in band societies, the one area in which men do exercise systematic control over women is the marriage system. Again, however, there is no ideology that the women exchanged in marriage must be virginal, sexually naive, and mystically pure; nor is seclusion of women practiced as a means of controlling their sexuality.
262    Gender, Property, and the State
Finally, however, we get to the great divide: the rise of the state. Here there is a radical shift of both ideology and practice. On the ground, we have the emergence of the patriarchal extended family. Indeed here for the first time the term patriarchy becomes applicable, because the structure involves the absolute authority of the father or other senior male over everyone in the household-all junior males and all females. And now women are for the first time brought under direct and systematic control, first by their natal families, and then by their husbands and their affinal kin. Among elites, one has the image of women being rounded up in great numbers and confined in harems and analogous arrangements elsewhere. Among the Brahmins of India, they are locked in great purdah palaces and never emerge into the world. The notion develops that men are directly responsible for the behavior of their women, rendering it part of every man's definition of self and manliness that "his" women never escape his control; his honor, and the honor of his group, are at stake.
At the same time, there is a great shift in the ideology concerning women. Before they were dangerous, but now they are said to be in danger, justifying male protection and guardianship. Before they were polluting, and this had to be defended against, but now they are said to be pure, and to need defending. At the same time, one finds for the first time symbolic idealization of woman in the mother-aspect, rather than in the sexual-reproductive aspect. Eventually, as the symbol system gets itself together in one part of the world with which we are all familiar, the ideal woman emerges as all the best things at once, mother and virgin.
Now the way in which I've described the pattern, and the way in which it might, at first glance, be viewed, is in terms of the domestication of women, a sort of Neolithic of the sexes wherein women, like plants and animals, were brought under control in the service of the race. Actually, however, my thinking is to envision the process in terms of the beginnings of the domestication of men, as part of a larger pattern of systematization of
hierarchy and control in the evolution of state structures. I will return to this point later.
In any case, the whole business is terriblv complex. What I shall do here is simply offer a brief checklist of points that I think would be important to consider in trying systematically to account for the changes in sex-role i ■ -lations and ideology that seem to be associated with the emergence of the state. The checklist consists of the following items, in no particular order: the question of diffusion, the question of changes in the division of labor, changes in religious thought, changes in family structure, and changes in marriage patterns. I shall only be able to say a few words about each, merely pointing in the direction I think investigation should go.
Diffusion, first, is something that cannot be entirely ruled out. It is possible that the p; ■-tern I have described—idealization offemale chastity; ideology of protection, control, and seclusion of women—developed in one area of the ancient old world, and reached other early states through trade and other dif-fusionary mechanisms. Most of the known contemporary societies with this pattern are geographically contiguous, in a broad band from the circum-Mediterranean area, across the Middle East and southwest Asia, across India, and up into China.11 The new world indigenous states would thus have to be investigated for independent evolution of the pattern." Even if indicated, however, we know that diffusion in itself never explains very much, for peoples hear of many peculiar customs practiced by their neighbors, yet those practices will only be adopted if the Su-cial structural and ideological conditions are: ripe for their reception. That is, the diffusion would only have taken hold as independently developing societies evolved the sorts of structures within which such a pattern of sex role relations would be functional and meaningful.
The second point, very briefly, is the question of whether changes in the division of labor may have motivated changes in sex-role relations and ideologies. One standard view has it that, with the rise of plow agriculture and/or systematic irrigation systems associ-
1
ated with the rise of states, women were excluded from major roles in the sphere of production, while their reproductive value in the family was more strongly emphasized. My own reading of the data is quite different. It is probably true that men became associated with plowing as a specific activity, and with the engineering and control of irrigation systems, and both of these points are quite important in the symbolics of male prestige. Nonetheless it seems that women continued to be fully productive, and if anything worked even harder than they did before, in both grain production in the western old world and rice production in the east. The gradual withdrawal of women from production (where it happened) was, I think, a very late development. It will thus not account for the emergence of the female purity pattern, although it will have repercussions in that pattern later.
In the domain of religion, next, I would stress the point that an elaborated notion of purity in general only conies in systematically with the emergence of state structures. In pre-state societies, including the simplest that we know, one of course finds the notion that exceptional purity, often including temporary celibacy, is required for specific important purposes. Generally, it is associated with some major male undertaking—a hunt, a raid, or a ritual—and is conceived in terms of purifying or at least not polluting male energies, so that they will be strong and focused for the big event. Nonetheless, there is no notion that it would be good for some people, female or male, to strive for permanent exceptional purity, including permanent celibacy. Such notions probably come in with permanent standing priesthoods of some kind, and these of course are standard, virtually diagnostic, features of early states.
If the chastity of priests was the first application of the notion of chastity to a social group, its rationale was probably similar to that for the episodic demands for purity in pre-state societies: the priest is charged with protection of sacred objects and activities, and he (or sometimes she) must be in a permanent state of non-pollution for thejob. But it would seem that there is more to it than
The Virgin and The 5tate 263
this. In particular I would suggest that in state religions and cosmologies, what seems to happen is that the whole purity ceiling is raised, so to speak. That is, one finds systematic elaboration of higher realms of purity and sacredness than existed before, with more exacting demands upon the laity for conforming to religious ideals. Thus it may be a matter of the religion postulating higher, more sacred, and more demanding gods (e.g., the Aztec gods who required human sacrifice), and/or a more articulated after-death state (as, for example, in The Egyptian Book of the Dead). Transcendental power, divinity, sacredness, and purity are all more articulated than previously.
It is not at all difficult to account for the emergence of such ideologies in state structures, in relation to the overall increased complexity of society—for example, more complex divine hierarchies may reflect more complex social hierarchies; or more demanding gods may reflect the greater demands of the state and the dominant classes; or more elaborated notions of afterlife may be interpreted as promising the newly emergent masses their rewards later rather than now, and so forth. The situation is of course infinitely more complex than this, but cannot be explored here. My point is simply that one would begin to investigate the elaboration of the notion offemale purity by contextualizing it in the emergence of systematic views of transcendental purity in state cosmologies in general. Purity as something that whole categories of people might intrinsically possess, or might systematically be required to sustain, as itself, I think, a product of state-related religious thought. Mind/body dualisms (reflecting, among other things, new divisions of labor between intellectual/artistic/political elites and producing masses), and the control of sexuality, sensuality, and materialism (part of, among other things, a delaycd-gratifica-tion, reduced-material-expectations, ideology for the masses) would be aspects of this general pattern.
Coming down from the cosmological heights to ground-level social structure, the fourth item covers changes in the structure of
264    Gender, Property, and the State
The Virgin and The State 265
the family. Again I will be very brief, although the problem requires detailed scrutiny. The key point is undoubtedly the emergence of the patriarchal family structure, and probablv ideally the patriarchal extended family. But the way in which I would look at this phenomenon, as I noted earlier, is in terms of the domestication of men, both as husbands/fathers, and as sons. Probably the catalyst around which the whole thing crystallized was the property holding in one form or another, although it was certainly not yet "private" property. Be that as it may, what I think was at issue was the gradual deepening of involvement of individual males in responsibility, as husbands/fathers, for their specific family units—not just economic responsibility, for that was always accepted, but also what might be called political accountability. The family became in a sense an administrative unit, the base unit in the political-economic structure of the state. The husband/father was no longer simply responsible to his family, but also/or his family vis-a-vis the larger system. It became the base, and often the only base, of hisjural status.
Now, judging from contemporary cases, I imagine that such deepening involvement of men in families was accepted only reluctantly, and as part of a tradeoff for patterns of deference and respect from wives and children. The reluctance of males to be involved with their families except on terms of distance, respect, and submission on the part of the other members is still I think to be seen in most of the world today, and the domestication of men is still largely incomplete. Nonetheless, the notion that males are not only economically but also legally and politically responsible for the proper functioning of the family unit seems to be part of the systematic extension of principles of hierarchy, domination, and order in the evolution of states as a whole. Responsible husbands/fathers are more systematically incorporated into the system.
Responsible husbands/lathers, which is to say in this context patriarchal husbands/fathers, in turn keep everyone else in line—the women, of course, but also the sons. Indeed, perhaps the most striking characteristic of the
patriarchal family is the prolongation of dependence and subjugation of sons. This j. such an overdetermined phenomenon that one can hardly begin to sort out its sources and components. However, let it suffice to say that sons are held back from the acquisition of property, wives, and emotional maturity by such a powerful combination of forces, emanating from both father and mother, that it is certainly one of the key changes that we see in family structure, regardless of household composition (that is, regardless of whether /; the family is "extended" or not). Male initiation rites virtually disappear in state societies; and far from fathers and other senior males facilitating, however frighteningly, a young male's passage to adulthood, the young adult male in the patriarchal family remains in a ju-rally dependent status at least until he is married, and often beyond. In many cases marriage itself becomes the only rite of passage, and thus manhood becomes equated with responsibility for wife and children, part of the pattern described above.
This pattern is likely to have certain psy-chodynamic implications. One may wish to go the Freudian route, in terms of deep unconscious factors, and I am not immune to the persuasions of a well-done Freudian analysis. ' • Kathleen Cough's paper on female initiation rites, mentioned earlier, is an excellent and very convincing Freudian discussion of the sorts of fears and ideals of women produced in such family situations in the Indian case. However, one can probably account for a lot without recourse to unconscious factors, through careful symbolic analysis of cultural notions of mothers and wives in such systems. The pivotal point of such analysis would be that men were not only "domesticated" as part of the crystallization of authority structures of the state; they were alsojuvenilized— vis-a-vis women, senior men, and the rulers and overclasses of the system.
Note that I have not tried to postulate motives for either women or men in this process. I have suggested that men were "reluctant" about being domesticated, but I would imagine that women had equally mixed feelings about the greater presence of male authority
in the family unit. I do not think it is useful to view this process in terms of (in Engels* famous phrase) "the world historical defeat of women" by men, or other such motivated formulations. The crystallization of patriarchal familv corporations was doubtless a precipitate of larger political and economic processes. Nonetheless, once it got going, it became a social force in its own right, affecting not only the further evolution of gender relations, but also the economic and political evolution of the larger system itself.
The final item on the present checklist is the question of changes in marriage systems. I noted earlier that concern for the purity of women is found, in contemporary societies, among both elites and lower strata. In southern Europe, the peasants seem much more concerned about the issue than the upper classes; for elites the relative freedom of their women is a symbol of their modernity, or else simply a symbol of their being above the codes. In India or China, on the other hand, the Brahmins and tipper classes were far stricter about the purity of their women than the lower castes and the peasants. In trying to account for the emergence of a code of female purity as part of the emergence of the state, one would perhaps want to begin with the question of which stratum started the whole thing. Thus some might argue that it was probably originally an elite conceit, in that elite women (if not other women) did not need to engage in productive labor, and could be secluded and protected from the pollutions of work and people as a mark of upper-class status. One could argue with equal logic, however, that there are aspects of peasant social life and social structure that would generate a concern for the purity of women, as in jane Schneider's discussion previously noted.
The way out of the puzzle, I would suggest, lies in stressing the stratified nature of the state as a totality, and seeking the dynamics of the process in the interaction between elites and lower strata. In particular, my analytic instinct is to look at patterns of hypergamy (up-status marriage, virtually always between upper men and lower women) in state sys-
tems, and to consider very centrally the possibility that one of the significant developments in stratified societies was the shifting of marriage from an essentially lateral transaction, beUveen essentially equal groups, to at least a potentially vertical transaction, where in one's sister or daughter is potentially a wife or consort of a king or nobleman, or could be dedicated to the temple and the services of the priesthood.
I think it is fairly safe to expect to find patterns and ideals of hypergamy, or what might be called vertical alliance, in stratified societies. Vertical alliance would constitute one of several sorts of paternalistic ties between the strata. But what analytic consequences flow from putting this point at the center of the analysis?
In the first place, as has been noted by others for India, the assumption of systematic hypergamy as an ideal and to some extent as a practice will account for the phenomenon of women accumulating at the top of the system.1"' Because lower-status families are eager to many their women upwards for political reasons, the elites would accumulate wives. At the same time, the elites would often be unable to get rid of their own sisters and daughters, for there is nowhere further tip for those women to go. Thus the emergence harems, purdah palaces, and so forth, would partly be a structural precipitate of the hypergamy system, rather than an indication of (among other possibilities) the extraordinary lust of sultans. At the same time, the accumulation of wives by polygynous royalty and nobility would certainly have value in the symbolics of power, for it would suggest their potency in everything from sex to politics to the fertility of the land. Thus the image of herds of penned-up women that is projected in these sorts of systems ilows partly from the dynamics of the marriage system, and not from men rounding up and controlling women as such.
Second, hypergamy may provide the strongest explanation for the purity ot women ideal and for certain peculiarities of this ideal. The context of hypergamy is a context of orientation toward upward mobility, through manipulation of marriages. We know
266    Gender, Property, and the State
that the economic value of women becomes a focus in these contexts, for it is here that we find the emergence of dowry, enhancing the girl's value for a higher-status spouse.10 In addition, however, there is the question of her mystical or spiritual value, her inner worthiness for such an alliance. The notions of virginity and chastity may be particularly apt for symbolizing such value, rather than, for example, external beauty, because virginity is a symbol of exclusiveness and inaccessibility, nonavailability to the general masses, something, in short, that is elite. A virgin is an elite female among females, withheld, untouched, exclusive.
The assumption of hypergamy would also account for one of the major puzzles of the female purity phenomenon, namely, that the women of a given group are expected to be purer than the men, that upon their higher purity hinges the honor of the group. I would argue that the women are not, contrary to native ideology, representing and maintaining the group's actual status, but are oriented upwards and represent the ideal higher status of the group. One of the problems with the purity literature, I think, has been a failure to get beneath native ideology; the natives justify female purity in terms of maintaining the group's actual status, as a holding action for that status in the system, when in fact it is oriented toward an ideal and generally unattainable status. The unattainability may in turn account for some of the sadism and anger toward women expressed in these purity patterns, for the women are representing the over-classes themselves.
And finally, the hypergamy assumption gives us at least one clue about a girl's (or woman's) motivation for cooperation in her own subordination and control. For if she is a good girl, she has the potential for personal status mobility which in fact exceeds that of most of the men of her group. Here it becomes intelligible that it is often women themselves who actively reproduce the patterns of female purity, socializing their daughters in fear and shame of sex, telling them that it is for their own good (which in a way it is), and spying on and gossiping about
one another's daughters as part of an overall deep internalization of and loyalty to the sy ■ tern. Again the point is the future orientation of the ideology, toward some often quite illusory but nonetheless remotely imaginable st; -tus mobility, which the girl herself internalizes as the "someday my prince will comt theme. It is no wonder too that women later " may resent their husbands as deeply as hu bands resent their wives—not only or eve necessarily because of the husbands' direi domination, but for what their husbands rep- • resent in status terms. For if the husband is of one's own status level, then one has saved all that purity for nothing, while if he is of the ideal higher status level, he is likely to be an undesirable mate who is willing to take a lowerclass wife because of some personal or social defect—some lecherous old Molierian widower, or someone of noble credentials but no money.
A final point about hypergamy leads me to my brief conclusions. Note that, once again, women are crossing boundaries, in this case boundaries separating classes or castes or status groups in vertical stratification systems. Thus the ambiguity of femaleness vis-a-vis social categories remains at the core of the problem, and views of women remain bound in the purity/pollution idiom. Perhaps partly because the boundary crossing is in an upward direction, however, the symbohzation of ambiguity shifts from danger to purity, although the deep structure, if one may use that phrase, remains the same.
At the ideological level, then, one may say that there has been a fairly simple structural transformation, and nothing much has changed in male attitudes toward and mistrust of women. It is clear in contemporary cultures with female purity ideologies, that women are still feared as ambiguous and dangerous creatures. Nonetheless I wish to close on an optimistic note. Levi-Strauss has suggested that there is no reason to assume that women and men would, if left to their own devices, form durable bonds of mutual interdependence.1' The phase of social evolution that I have been discussing may perhaps least depressingly be viewed as a long, painful, and
unfinished moment in the dialectic of the evolution of such bonds.18
NOTES
This paper was written as an informal taik. It was my first stab in thinking about the problem and is highly speculative. It is really designed to generate and orient further thought and research—my own and others. The version printed here is a very slightly (mostly stylistically) revised version of one printed in Michigan Discussions in Anthropology 2 (1976): 1-16. I am grateful to the editors of Feminist Studies for encouraging me to reprint it and to Michigan Discussions in Anthropology for permission to do so. 1. Quote on the seraglio from The Encyclopaedia Britannica, 15th ec!., s.v., "Harem." For a Latin American example, see Lawrence Watson, "Sexual Socialization in Guajiro Society," Ethnology 11 (1972): 150-56. For the Mediterranean, see Joseph K. Campbell, Honour, Family, and Patronage (Oxford, England: Oxford University Press, 1974); j. G. Peristiany, ed., Honour and Shame: The Values of Mediterranean Society (Chicago: University of Chicago Press, 1966); jane Schneider, "Of Vigilance and Virgins: Honor, Shame, and Access to Resources in Mediterranean Society," Ethnology 10, no. 1 (1971): 1-24. For the Middle East and South Asia, see Rose Oldfield Hayes, "Female Genital Mutilation, Fertility Control, Women's Roles, and the Patrilineage in Modern Sudan," American Ethnologist 2 (1975): 617-33; Hannah Papanek, "Purdah: Separate Worlds and Symbolic Shelter," Comparative Studies in Society and History 15 (1973): 298-325; Kathleen Gough, "Female Initiation Rites on the Malabar Coast," Journal of the Royal Anthropological Institute 85 (1955): 45-80; Nur Yalman, "On the Purity of Women in the Castes of Ceylon and Malaber,"_/o/<r?m/ of the Royal Anthropological Institute 93 (1963): 25-58. For classical Athens, see Philip Slater, The GUny of Hera (Boston: Beacon Press, 1968). For China, see Judith Stacey, "When Patriarchy Kowtows: The Significance of the Chinese Family Revolution for Feminist Theory," Feminist Studies 2, no. 2/3 (1975): 64-112. The ethnographic references for this paper are in no way definitive or comprehensive. They are taken from the sources that
The Virgin and The State 267
were at hand during writing. For most areas mentioned in the paper, there is a large body of literature concerning sexual ideology and female sociosexual control.
2. For some general considerations, see also Sig-mund Freud, "The Taboo of Virginity," Standard Edition 11 (London: Hogarth, 1957): 193-208; Ottokar Nemecek, Virginity: Pre-nuptial Rites and Rituals (New York: Philosophical Library, 1958). For more recent overviews, see Louise Lamphere, "Power, Purity, and the Position of Women: The Implication of Sex-role Ideologies for Social Subordination," (unpublished ms., 1975), and Harriet Whitehead, "The Dynamics of Chastity and the Politics of Mutilation," (unpublished ms., n.d.)
3. Gough, "Female Initiation Rites."
4. Yalman, "On the Purity of Women," p. 41.
5. Watson, "Sexual Socialization," pp. 151, 154. For an even more gruesome case, see Hayes, "Female Genital Mutilation," on infibulation, or Pharaonic circumcision, in the Sudan.
6. Schneider, "Of Vigilance and Virgins," pp. 2, 17, 18,21.
7. I am referring primarily to what anthropologists call "primary" (or "archaic") states, manifesting the novel social-evolutionary development of specialized centralized decision making and bureaucratized administration. See Henry Wright, "Recent research on the Origin of the State," Annual Review of Anthropology 6 (1977): 379-97. See also Rayna Rapp, "Gender and Class: An Archaeology of Knowledge Concerning the Origin of the State," Dialectical Anthropology 2 (1977): 309-16.
8. Bradd Shore ("Sexuality and Gender in Samoa: Conceptions and Missed Conceptions," forthcoming in Sexual Meanings, eds. Sherry B. Ortner and Harriet Whitehead) describes a similar pattern for Samoa, which would perhaps be characterized as a "pre-state" society, but not by much: it has a highly developed system of hierarchical social differentiation.
9. Engels of course related the "defeat" of women to the rise of the state (F. Engels, The Origin of the Family, Private Properly and the Slate [New York:" International Publishers, 1972]), but my interpretation of this relationship is quite different from his. Sec also Wilhelm Reich, The Invasion of Compulsory Sex Morality (New York: Farrar, Strauss, and Giroux, 1971) for a rather bizarre, if provocative, discussion of these issues.
268    Gender, Property, and the State
Women's Property Rights and Dowry in China and 5oulh Asia 269
10. Mary Douglas, Purity and Danger (New York and Washington: Pracger, 1966).
1 1. Sheny B. Ortncr, "Is Female to Male as Nature is to Culture?" Feminist Studies 1, no. 2 (Fall 1972): 5-32; reprinted in Woman, Culture, and Society, eds.T M. Rosaldo and L. Lamphere (Stanford: Stanford University Press, 1974).
12. Michelle Z. Rosaldo, "Woman, Culture and Society: A Theoretical Overview," in Rosaldo and Lamphere, Woman, Culture and Society,
13. The pattern is also found throughout most of Latin .America, presumably imported by the Iberian empires.
14. It seems that the Incas had celibate male and female priesthoods (see Victor von Hagen, The Realm of the Incas [New York: New American Library, 1957)], but we do not know whether an ideal of chastity for women was general. As for the Aztecs, Vaillant states that women were supposed to be chaste at marriage, but he does not elaborate on this point (George Vaillant, The Aztecs of Mexico [Baltimore: Penguin. 1950]).
15. See, for example. S. J. Tambiah, "Dowry and Bridewealth and the Property Rights of Women in South Asia," in Bridewealth and Down, cds., ]. Goody and S.J. Tambiah {Cam-
bridge, England: Cambridge University Press 1973).
16. See Jack Goody, "Bridewealth and Dowry in Africa and Eurasia," in Goody and Tambiah, Bridewealth and Dowry.
17. Claude Levi-Strauss, "The Family." in Man,-Culture and Society, ed., H. Shapiro (New York: Oxford University Press, 1960).
IS. I must tack on here one of the more interesting points raised in discussion after I presented this paper as n talk. In many "primitive" societies, sexual activity begins at a very early age (for example, Tiwi men of Australia stretch the vaginas of their child wives with their fingers until the wives are grown enough to engage in full-scale-sexual intercourse). (Sec Jane Goodale, Tiwi Wives [Seattle and London: University of Washington Press, 1971].) It is conceivable then that females in such societies never really develop hymens at ah. As premarital chastity is enforced for women to a later age, however, the hymen would have a chance to grow and harden, and would have to be broken more dramatically at first intercourse. The suggestion is, then, that the hymen itself emerges physiologically with the development of sexual purity codes, and thus presumably with the rise of die state.
John L McCreery
Goody and Tambiah (1973) have provided a major contribution to our understanding of marriage as a social institution. Of particular importance is the way in which these authors draw our attention to the economic aspects of marriage and the ways in which they affect the
From Ethnology 15:16:5-174. 1976. Reprinted with permission of the University of Pittsburgh, Department of Anthropology.
status of women. They have pointed out that bridewealth is associated with lack of social stratification, important economic roles for women, and relatively weak controls over women's sexuality. In contrast dowry is associated with social stratification, restriction of women to the roles of housewife and mother, and emphasis on chastity as a female virtue.
However, in drawing their contrasts between bridewealth and dowry, Goody and
Tambiah—and more especially Goody—have obscured rather than clarified the relationship between women's property rights and dowries: Goody suggests that dowry be treated as a form of inheritance. In fact, I argue, dowry and inheritance are fundamentally different. Where women inherit they exercise rights like those enjoyed by men, but where women receive dowries their rights to inherit are either restricted or non-existent, and their status is markedly lower than that of menfYalman 1967: 172-179).
Goody(1973: 1, 17) has stated that "dowry can he seen as a form of premortem inheritance to the bride" and as part of a woman's property complex. It is, he says, women's ability in many parts of Asia to acquire the same kinds of property as men (and often from men) which he and Tambiah (1973) want to contrast with the situation associated with bridewealth in Africa in which male and female property are sharply segregated.
What Goody seems to be saying is that dowries can only occur where women have rights to the same kinds of property as men. Because of their rights women either inherit shares of their parents" property or receive them as dowries. Inheritance and dowry are only different ways of implementing the same basic rights.
It has been argued, however, that in China, where dowry was customarily a part of marriage, "Chinese women had practically no property rights" (Lang 1946:44). They might legally own property but usually had no right to inherit property. Whatever they owned was received by them as gifts or as wages. Men had rights to inherit, but women did not.
To consider dowiy as a form of pre-mor-tem inheritance and as part of a woman's property complex would, in the case of China, blur an important distinction between the definite legal lights to inherit enjoyed by men and the privilege of receiving a dowry which might or might not be enjoyed by women at their parents' or brothers' discretion. In this case it makes no sense to consider dowry a form of pre-mortem inheritance based on the same set of rights as other forms of inheritance.
The case of China demonstrates that dowries may not be related to rights to inherit, and thus suggests the need for a closer look at women's property rights and how they relate to whether or not women receive dowries in any society where dowries are given. The scope of this paper is, however, more modest. I first describe women's property rights and how they relate to dowiy according to the Ch'ing Code, the statute law of China's last imperial dynasty. I then compare the relevant provisions of the Ch'ing Code to the Indian legal systems described by Tambiah (1973) and the Sinhalese law described by Tambiah (1973) and Yalman (1967). My discussion will be confined to these bodies of law.
In each case we are dealing with legal statutes formulated by members of ruling elites which may differ in some respects from the customs of particular communities. We are also dealing with only a few of the bodies of statute law which might have been considered. These cases suffice, however, to document and to elaborate the critique of Goody's interpretation of down-- which is sketched above.
WOMEN'S PROPERTY RIGHTS AND THE CH'ING CODE
It has been argued that according to the Ch'ing Code a family's1 property belonged to its head and that no other member of the family could use or dispose of its property without his consent. Only a male could occupy the position of family head, and thus women, excluded from becoming family heads, were also excluded from ownership of family property (Ch'u 1961: 29-31, 103-104).
Only in exceptional circumstances were women allowed to inherit. The law decreed that after the death of a family head his property should be divided equally among his sons. If he bad no sons, one of his brother's sons should be appointed his heir. Only if neither sons nor brothers' sons were alive to assert their rights might his daughters inherit. Widows who did not remarry and unmarried
270    Gender, Property, and the State
Women's Property Rights and Dowry in China and South Asia 271
daughters were customarily entitled to maintenance from their husband's or father's estate, but ordinarily they had no right to inherit (Chiu 1966: 1-2).
These arguments are based, however, on a highly schematic conception of Chinese law and the social structure defined by it. Their underlying assumption is that the family was the basic property holding unit, and its property was owned by the family head who was free to dispose of it as he pleased.
In fact, legally as well as customarily, individuals, including individual women, might own personal property distinct from family property. Moreover, only a family head who was also the direct ascendant of all of the other family members was free to dispose of its property as he pleased. If, as was often the case, the family head was either junior to some of the family's members or, if senior, only a collateral relative, his power to dispose of the family property was limited by law as well as by custom.
More sophisticated analyses can be found in the papers by McAIeavy (1955) and Tai (1963) on which the following discussion is based. According to these authors, a woman might become a family head, but only if there were no competent male to occupy the position. Moreover, a female family head's authority was limited. In particular she was not free to dispose of the family's property in the same way as a male family head who was also the direct ascendant of the family's other members.
A family head's authority varied depending upon his relationships with other family members. His authority included three possible components, rights which belonged to any senior member of a clan in relation to his juniors, rights reserved for direct ascendants, and rights attached to the status of family head perse (McAIeavy 1955: 540-541).
A clan was a group of agnates among whom seniority was based, first, on generation and, second, within a generation on relative age. Its membership was defined by patrilineal descent from a common ancestor, and in principle the rights attached to seniority were not dependent on residence or property. The relationship of seniors and juniors
was one of "authority and obedience" (Tai 1963: 5). Seniors had the right to issue commands and demand their juniors' obedience. Their authority was supported by laws which, with a few exceptions, prohibited juniors' taking their seniors to court and punished juniors more severely for offenses against seniors than vice-versa. Except for the specific exceptions discussed below a senior's authority was limited primarily by his own junior position relative to still more senior persons. The law provided no specific regulations, but in principle if two seniors' commands conflicted, the more senior's wishes would have to be obeyed (Tai 1963: 7).
A direct ascendant's authority was virtually absolute. He could not be taken to court by a junior under any circumstances, and offenses against him were the most severely punished. If a family head, a direct ascendant was free to dispose of family property as he saw fit. In contrast a family head who was not a direct ascendant could be taken to court to demand the family property's division or to accuse him of malfeasance in his management or disposal of it. To divide, sell, or pawn the family's property he needed his junior's consent. Thus, while a father's or grandfather's wishes could not be challenged, an uncle's authority was limited (McAIeavy 1955: 540; Tai 1963: 9).
A person's direct ascendants included his father and mother, his paternal grandparents, his paternal grandfather's parents and, potentially, any ancestor related to him through a direct patrilineal line for descent. Consequently a number of different individuals might possess a direct ascendant's authority over him. In contrast, the rights specific to a family head were concentrated in a single individual (McAIeavy 1955: 542).
The family head's principal right and duty was to manage the family's property (McAIeavy 1955: 541). He controlled income and expenditures, decided the division of labor and settled disputes among the family's members (Tai 1963: 6). If the family head were also the most senior direct ascendant of all of the family's other members, his authority was limited only by laws prohibiting willful murder. If, however, the family included persons
senior to him or juniors who were not his direct descendants, the family head's authority-was limited. He would still have to obey the commands of his seniors and, as we have seen, could not dispose of the family's property without the consent of his juniors to whom he was not a direct ascendant.
Concerning the family head's qualifications the law was specific: "Males first, females later; among the males the most senior is appointed" (Tai 1963: 4). A woman could not be family head if any competent male were available (McAIeavy 1955: 9). A widow with a son too young to be family head might act in his stead, but only until he came of age.
As a daughter or wife a woman could not be family head and thus could not freely dispose of family property. A daughter was doubly excluded from the family head's position by being both junior and female. A wife's relationship to her husband and his parents was likened to that of a child to her parents and grandparents (Tai 1963: 6).
If a male family head had died and neither his son nor his grandson were old enough to succeed him, his wife would manage the family property. To pawn or sell it, however, she would have to secure her son or grandson's agreement. She was not free to dispose of the property as her husband had been (Tai 1963: 10).
In any case, if the family head were the owner of the family's property, then women, who ordinarily could not be family heads, would usually not be the owners of family property. Whether or not the family head was the owner of the family property is, however, a debatable question. It has been argued that family property was collectively owned by the family as a corporate group and was not the personal property of the family head. If that were the case then women might have had some property rights as family members, even if they could not become family heads.
The theory that the property belonged to the family head was embodied in Taiwan Shiha, the Japanese government's official compilation of the Chinese law and custom in effect in Taiwan when Japan assumed control of the island after the Sino-Japanese War in 1895. Published in 1905, Taiwan Shiha, has been
one of the basic sources of Japanese scholarship dealing with Chinese law (McAIeavy, 1955: 536). More recently the theory that the family property was the personal property of the family head has been defended by Shiga Shuzo (1967), whose work is cited by Meijer (1971).
To my mind, however, the family head's personal ownership of the family's property has been decisively refuted by Tai (1963: 9-10). Tai points out that the same terminology is applied to division of the family's property both before and after the family head s death. More importantly the theory of the family head's personal ownership fails to explain three points.
First, if a family head's son or grandson entered another family by adoption or by uxori-local marriage he lost the right to share in the division of his natal family's property. If the kinship relation between the son or grandson and his father or grandfather were decisive, then his right to share in the property's division would not have been affected by his leaving one group to enter another. What was decisive was his relationship to the group as a whole, not his relationship to the family head as an individual.
Second, if a family head died and a son or grandson who had entered another family left that family and returned to his natal family before its property was divided, his right to share in the division of its property was restored to him.
Third, even if a family head already had sons and grandsons, if after his death his wife or concubine adopted another son, that adopted son, too, would share in the division of the family's property.
In these two cases, as well, the relationship of the individual to the group as a whole is decisive in determining rights to family property, while his relationship to the family head as an individual is not. If, however, family membership rather than personal relationship to the family head was decisive, then the family property could not have been the family head's personal property in the first place.
Tai also observes that the theory of the family head's personal ownership is based on the ideal case in which the family head is also
272    Gender, Property, and the State
Women's Property Rights and Dowry in China and South Asia 273
the direct ascendant of all of the other family members. In that case his authority to command, admonish and punish his juniors and his own immunity of prosecution would allow him to act arbitrarily in disposing of the family's property. These rights, however, were attached to his status as a clan senior and direct ascendant and not to his status as family head per se. As noted above, his authority was limited if the family members included persons who were either his seniors or not his direct descendants. Then the legal rights of other persons, based on their membership in the family as a corporate group, emerged clearly.
We may now turn to the questions whether or not women had rights to the same kinds of property as men and whether or not they had the same kinds of rights. There was no legal classification of property which restricted the kinds of property to which women might have rights. There were, however, laws which implied that the rights of most women were radically different from the rights of most men.
Family members were divided into two categories, "basic shareholders" whose rights to family property were legally defined and "optional shareholders"" who might or might not receive property at the basic shareholders' discretion. As a general rule the basic shareholders were men, the direct descendants of a common male ancestor, entitled to shares of their family's property because they, and only they, were qualified to continue the ancestral sacrifices. All of a man's legitimate sons, regardless of whether their mothers were his principal wife or his concubine, were entitled to equal portions of their father's share of the family property. Their rights were divided per stirpes, not per capita. Women were not permitted to continue the ancestral sacrifices and thus were excluded from being basic shareholders (Tai, 1963: 12).
These general rules were qualified in certain cases. An illegitimate son, i.e., a son whose mother was neither his father's principal wife nor his concubine, was entitled only to half a basic shareholder's portion. Adopted sons were divided into two distinct categories. Those adopted to continue their father's line
of descent were basic shareholders. Others, whose adoptions were acts of charity, were only optional shareholders (Tai, 1963: 13).
Men who married uxorilocally were, like women, not entitled to continue the ancestral sacrifices of the families into which they married and thus could not be basic shareholders with definite rights to family property. Such rights as they had were established by contracts at the time of their marriages. None were attached to their status per se (Tai 1963-15).
Women were customarily entitled to maintenance and to a trousseau when they married but they had no definite legal rights to them, except for the implicit right implied by the criminal law prohibiting deliberate murder. The law prohibited even parents from willfully starving their daughters to death. When a family's property was divided, the basic shareholders might decide to set aside some part of the property to provide their unmarried sisters with dowries but they were not legally obliged to do so.
Widows of men who had no sons were legally required to adopt sons for them, and only if no possible heirs were available would a widow receive her husband's share of the family property in her own right. A widow who remarried lost her right both to her first husband's share of his family's property and also to the dowry, if any, which she had brought to her first marriage (Tai 1963: 14).
Women, as such, were expected to marry out. If a woman's father had no sons he might, instead of adopting a son, arrange for an in-marrying son-in-law with the hope that his daughter would produce a grandson to continue his line. As we have seen, the uxorilocally marrying man was only an optional shareholder. If her father died before her son was old enough to become family head and assume control of the family property, the woman herself might act as trustee. She was, however, expected to transmit the property intact to her son and was not free to dispose of it however she pleased.
We should also note that when a man died leaving no sons but several daughters, only one of the daughters would marry an in-mar-
rying son-in-law to continue the father's line. The others would many out in the usual way (Tai 1963: 14).
We can now summarize the legal position of women in relation to family property. As unmarried daughters or as widows they were only optional shareholders who legally might be given nothing at all over and above bare subsistence when the men who were basic shareholders divided the family property. In this respect their rights were subordinate even to those of illegitimate sons. As wives or as mothers they had no legal claims on family property in their own rights, unless as widows they had no sons and none might be adopted.
So far, however, we have only considered family property. There is still the question whether or not women, as individuals, could own personal property distinct from family property.
The Li Chi, the classical Book of Rites, had stated the principle that children could not own private property while their parents were alive. In particular, sons and their wives were forbidden to possess their own goods, livestock or implements. As part of the Confucian canon the Li Chi had moral authority throughout much of China's history. Nonetheless, personal property is mentioned in historical records as early as the Han dynasty (206 b.c.-220 a.d.) (Tai 1963: 10). A Tang dynasty (618-906 a.d.) statute implied that rewards for distinguished sendee to the government, even if given to a family's junior members, were not to be included in the property collectively owned by the family as a corporate group (McAleavy 1955: 545-546).
Another Tang statute specifically excluded from family property not only the personal items included in a bride's trousseau but also any goods, including land, brought with her as dowry, and the statutes of later dynasties followed this precedent (McAleavy 1955: 546). Ethically speaking, the management and disposal of a wife's property were supposed to be entrusted to her husband, but the husband's control was not legally necessary (Tai 1963: 10).
A distinction was usually made, however, between the personal items (e.g., clothing
and jewelry) which constituted a bride's trousseau and the land or other property provided as a dowry. The trousseau was always considered the bride's personal property, but generally the dowry was given not to the bride herself, but instead to the branch of her husband's family composed of herself, her husband and their descendants. Only rarely was the bride's personal ownership stipulated in the marriage agreement (McAleavy 1955: 546).
Usually land or other goods given as dowry were kept apart from a family's common property until the family was divided. Then the dowry was merged with the husband's share of his family property and became the collectively owned property of the new family headed by the husband and wife (McAleavy, 1955: 546). It is clear, however, that women could possess private property.
WOMEN'S PROPERTY RIGHTS IN CHINESE AND INDIAN LAW
Tambiah reminds us that there was no single body of traditional Hindu law. Indian law included two main traditions, the Mitakshara law and the Dayabhaga law. The more widespread Mitakshara law was itself divided into four distinct schools. Tambiah (1973: 74) concentrates primarily on the Mitakshara law at the level of general principles which seem to be shared by all four schools and only comments on the Dayabhaga law where it differs significantly from the Mitakshara. In what follows, Indian law refers to the body of genera! principles described by Tambiah, and no attempt is made to challenge his description.
The Hindu joint family, as legally defined, is a corporate group including both male and female members, some of whom are co-parceners (i.e., persons entitled to a share of the family's property) while others are only entitled to maintenance. Tambiah quotes Maine (1883:230) as follows:
When we speak of a Hindu joint family as constituting a coparcenary, we refer not to the entire number of persons who can trace from a
274    Gender, Property, and the State
Women's Property Rights and Dowry in China and South Asia 275
common ancestor and among whom no partition has ever taken place; we include only those persons who, by virtue of relationship, have the right to enjoy and hold thcjoint property, to restrain in the acts of each other in respect of it, to burden it with their debts, and at their pleasure to enforce its partition. Outside this body there is a fringe of persons who possess inferior rights such as that of maintenance, or who may, under certain circumstances, hope to enter the coparcenary.
On the face or it this description would seem to apply to traditional Chinese as well as Hindu families. More closely examined, however, it reveals several subtle but basic differences between them. The phrase "to restrain the acts of each other in respect of it" would apply only to senior or collateral relatives of a Chinese family head whose direct descendants had no right to restrain his use or disposal of family property. The phrase "to burden it with their debts" would only apply to the Chinese family head himself, since other family members were prohibited from giving, selling or pawning family property without his consent. Finally, the phrase "to enforce its partition" would once again only apply to family members who were not the family head's direct descendants.
Tambiah states four rules which govern the rights of co-parceners according to Indian law. First, each co-parcener in an undivided family transmits his share to his sons who receive equal portions of it. This rule is identical to Chinese law. Second, the co-parceners include "the three generations next to the owner in unbroken male descent, i.e., a man, his living sons, grandsons and greatgrand-sons constitute a single co-parcenary with himself (Tambiah 1973: 77). According to this rule a four generation group is theoretically the maximum group, which must be divided when the fifth generation is born, lest the fifth generations members lack any property rights at all. In this respect the Indian law differs from the Chinese law which does not restrict the generational depth of family members' relationships. Third, property acquired by a co-parcener without obligation to his father's estate, i.e., by his own learning or
valor or by gifts from friends, is his personal property and not included in the family's estate. In this respect Chinese and Indian law-are similar. There is, however, potentially room for difference in the qualification "with, out obligation to his father's estate," since the scope of that obligation might be conceived quite differently. Fourth, "any co-parcener may sue for partition, and every co-parcener is entitled to a share upon partition," and, moreover, "the Mitakshara explicitly assert-the son's right to partition even against the father's wish" (Tambiah 1973: 77). Like the Chinese law, the Dayabhaga law prohibits a son's demanding partition before his father's death. The Chinese law, moreover, prohib ited any direct descendant's demanding partition while any direct ascendant was still alive.
Chinese law gave greater stress to the family as a corporate group than the Indian law. The accent of the Indian law was "on the separate existence of each brother, although brothers form the core of the lineage" (Tambiah 1973: 80). Unlike the Indian law, the Chinese law placed no limit on the generational depth of the family and did prohibit sons or grandsons from demanding the family's division while their fathers or mothers, grandfathers or grandmothers were still alive.
Conversely the Indian law seems to have given greater stress to genealogical relations between individuals than the Chinese law. We have observed that in China a son who left his natal family through adoption but returned to it before its property was divided recovered his right to a share of the property. In contrast, at least according to the Mitakshara system, an adopted son's property rights in relation to his natal family "ceased as if he had then died" (Tambiah 1973: 82). It was stated that "the adopted son shall never take the . .. estate of his natural father" (Derrett 1963: 117). This principle was variously interpreted to allow or not to allow the adopted son's right to property already vested in him at the time of adoption, but there is no suggestion that an adopted son could ever recover his right to a share of his natal family's estate.
It is, I suggest, in this context of greater emphasis on the rights of individuals that we should understand the Indian law that "to unmarried daughters a nuptial portion must be given out of the estate of the father; and his own daughter, lawfully begotten, shall take, like a son, the estate of him who leaves no male issue" (Maine, 1883: 501). In contrast the Chinese law, with its greater stress on the family as a corporate group, makes an unmarried daughter's "nuptial portion" dependent on the wishes of her brothers or uncles who are basic shareholders of the family's property.
In Indian law women's property rights belong to different categories depending, first, on the property's origin and, second, on its devolution after a woman's death. According to Tambiah (1973: 86), dowry includes "clothes and jewelry received by the bride from her parents before marriage, betrothal presents from the bridegroom, presents received by her at marriage from her parents and kin, and gifts received after marriage from her husband."
In contrast, there is property received by inheritance or partition as a member of her husband's family, i.e., as wife, widow or mother and property earned by the woman's labor. Gifts of immovable property, e.g., land or houses, received from the husband after marriage seem to belong to the latter category and not to dowry. The dowry itself is subdivided into two components, property received directly from parents or kin from whom it originates and property originally given to the parents as brideprice and then transmitted by them to their daughter.
The non-dowry property acquired by partition, inheritance or labor is controlled by the woman's husband so long as he lives and then is held in trust to be transmitted to his sons or grandsons after his widow's death. The woman herself cannot freely dispose of it. In contrast her dowry is a woman's personal property and she can freely dispose of at least those parts of it which she has received from her parents and kin.
It is the devolution of a woman's property which most strikingly distinguishes the part ol
dowiy received from parents and kin themselves from the part received, albeit indirectly, from the bridegroom who pays it as brideprice. The part received from parents and kin is inherited by a woman's unmarried daughters, her married daughters, or in default of these by their sons. The part which originated as brideprice is inherited by a woman's brothers, her mother or her father in that order.
These generalizations conceal a variety of differences among the different traditional legal systems of India. They suffice, however, to point up a basic contrast between all of these systems and the Ch'ing code. The Chinese law gives no explicit sanction to inheritance of any part of a woman's property by either her daughters or her brothers. The general expectation underlying the Chinese law seems to have been that a woman's property would be merged with her husband's and sooner or later become the family property of the family begun by their marriage. It would then be inherited in the usual way with her daughters' shares, if any, dependent on the wishes of their brothers, the basic shareholders.
In summary we can say that Chinese law more heavily emphasized the family as a corporate group whose core was composed of ag-natically related men. With its greater emphasis on the group of agnates and greater concern for its continuity, the Chinese law more severely restricted the rights of woman than the Indian law. In fact in China women's property rights were generally only customary expectations, permitted by law but not required by law as they were in India. In contrast Indian law gave greater weight to the rights of individuals including, in particular, the rights of individual women.
Goody (1973) has written that he and Tambiah wish to call our attention "to the nature of the property relations between the partners of a marriage and with the parents of each, i.e., to the interrelationship of marriage transactions and the devolution of property." What is striking about this statement is what it omits, the relationships of the bride and groom to their families as groups and rela-
276    Gender, Property, and the State
tionships between the families as groups. If, as seems to have been the case in China, families were conceived as corporate groups whose existence transcended the personal relationships of their members and, indeed, took precedence over their individual members' private interests, then Goody's approach to the study of dowry is lacking a crucial component. This lack is felt even in cases where individual rights are most pronounced.
WOMEN'S PROPERTY RIGHTS AND DOWRY IN CEYLON
Superficially Sinhalese law would seem to offer an ideal case in support of Goody's interpretation of dowry. At least in the Diy Zone of Ceylon, both men and women were individually owners of private property, and their private pro pert}1 might include land as well as personal effects. The owner of property, whether male or female, was free to dispose of it however he or she saw fit, even to give it or to leave it after their deaths to persons other than their legal heirs (Yalman 1967: 130).
If a parent died intestate, sons and daughters shared equally in their parent's property, regardless of their parent's marital history. Women as well as men retained their rights no matter how often they married, divorced, and married again. Yalman (1967: 134) writes: "the conception of marriage as a powerful, almost indissoluble bond between man and wife, in which their separate properties are united and which clearly defined the heirs to the property as thejoint heirs of the couple . . . has no place in the Kandyan way of thinking."
It is against this background that Yalman observes that certain classes of Sinhalese distinguish between virilocal and uxorilocal marriages. Women who many uxorilocally and their unmarried sisters are entitled to inherit equally with their brothers. Women who many virilocally with dowry, however, give up their rights to inherit/'
Here we have a prima facie case for asserting that dowry replaces rights to inherit after a parent's death and thus that dowry can be
seen as a form of premortem inheritance. It can still be argued, however, that dowrv and inheritance are essentially different.
Goody (1973: 17) argues that female inheritance and dowry both reflect "a general interest in preserving the status of offspring oi both sexes." This interpretation might well; apply to the rule of inheritance mentioned above, since daughters inherit equally with sons. There is. however, no rule which requires that a dowiy be equal either to the inheritance received by a son or unman-ied daughter or to the dowries received by the bride's sisters,
Since bride and groom should contribute equally to their marriage and women are not allowed to many down, the size of the dowiy "depends entirely on the status of the son-in-law" (Yalman 1967: 132). The groom's status is a function partly of inherited rank and partly of wealth, and the higher his rank or the greater his wealth the larger the dowiy required to arrange a marriage with him.
In some cases, in order to arrange an especially desirable marriage, "one daughter receives all the cash and even much of the lands owned by her parents" (Yalman 1967: 175). In such a case the daughter's dowiy would probably be much larger than her sister's dowiy or her brother's or sister's inheritance. The difference between her dowry and the shares of her parents' property received by her brothers or sisters cannot be explained by "a general interest in preserving the status of offspring of both sexes." Her dowiy is certainly not simply the share of her parents' property which she would have a right to inherit.
Why dowries are given is a question which the law does not answer. Dowries are legally possible because individuals may dispose of their property as they see fit. They are not, however, legally necessary and thus their causes have to be sought in extralegal circumstances. Yalman (1967: 149) states:
1 lie kick of a down' system is associated with egalitarian and immobile villages. Alternatively, dowiy—wit!) all that it implies in control over women and solidarity among men as ;i mark of
Women's Property Rights and Dowry in China and South Asia 277
status—appears to be a function of diversified communities with at least some mobility.
In China, too, the causes of dowiy are extralegal. According to Freed man (1966: 55), a wealthy family provides a dowiy for their daughter,
not because the girl has any specific economic claims on them (she is not a member of the property-owning unit) but because their own status is at stake; a bride-giving family must, in order to assert itself against the family to which it has lost a woman, send her off in the greatest manner thev can afford. And it is no accident, therefore, that dowry and trousseau are put on open display; they are not private benefactions to the girl but a public demonstration of the means and standing of her natal family.
The Chinese case also fits the social background and implications of dowiy described by Yalman. China lias long been a society of diversified communities within which social mobility was not only possible but highly prized, and in China, as well as Ceylon, dowry lias been associated with control over women and solidarity among men as a mark of status.
CONCLUSIONS
In China, India, and Ceylon dowiy has been associated with women's property rights which have ranged from virtually nonexistent to essentially the same as those enjoyed by men. In Ceylon and China, moreover, rights to inherit have been legally defined, while rights to dowiy have not. It is clear that at least in these two cases the legal bases of dowiy and inheritance were distinct. In neither case were dowiy and inheritance simply alternative ways to implement the same basic rights.
It seems clear then that dowiy cannot be properly understood as a form of pre-mortem inheritance to the bride. Whether or not it is part of a woman's property complex is a moot point.
In all three societies women could own the same kinds of property as men, but this was at
most a necessary but not a sufficient condition for dowiy to be given. At least in Ceylon and in China whether or not a dowiy was given at all, not to mention the size of the dowiy, was up to the men who controlled the family's property.
The conditions which affected whether or not dowries were given were extralegal. Consequently (he laws which we have examined do not tell us what they were. It seems likely, however, that dowries are a means of social mobility in stratified societies where men use rights over women, like other property, to compete for higher status.
NOTES
1. In this paper the terms family, clan and lineage are translations of the Chinese terms cilia, tsidig. and hit respectively. The romamut-lion is Wade-Giles.
2. Basic shareholders and optional shareholders are translations of the Chinese phrases clii-fjeii \ti-fcn jen and clio-lti'i y it-fen jfii respectively. The loimtiiizdlinn is Wade-Giles.
3. To speak of a right to inherit might seem inappropriate since parents were free to write wills and dispose of their property however the)' pleased. A similar objection might be raised for the ideal Chinese case in which a family head was also his heirs' direct ascendant and thus free to dispose of family property as he saw lit. In both the Chinese and Sinhalese cases the rights in question are those which were legally defined when a parent died intestate. In neither case was a right to a dowiy prescribed by law.
REFERENCES
Chiu, V.Y. 1966. Marriage Laws and Customs of China. Hong Kong: Chinese University of Hong Kong. New Asia College, Institute of Advanced Chinese Studies and Research.
Ch'u T'ung-tsu. 1961. Law and Society in Traditional China. Paris and The Hague. Mouton.
Derretl, J. Duncan. 1963. Introduction to Modern Hindu Law. Bombay Indian Branch, Oxford University Press.
278    Gender, Property, and the State
Fvcedman, M. 1966. Chinese Lineage and Society: Fukien and Kwanglung. New York.
Goody, J. 19713. Bridewealth and Dowiy in Africa and Eurasia. Bridewealth and Dowry, ed., j. Goody and S.J. Tambiah. pp. 1-58. Cambridge.
Lang, O. 1946. Chinese Family and Society. New Haven.
Maine, H.S. 1883. Dissertations on Early Law and Custom. London:J. Murray.
McAleavy, H. 1955. Certain Aspects of Chinese Customary Law in the Light of Japanese Scholarship. Bulletin of the School of Oriental and African Studies 17: pp. 535-547.
Meijer, M.J. 1971. Marriage Law and Policy in the Chinese People's Republic. Hong Kong.
Shiga, S. 1967. Chfigoku Kazoktiko no Genri [Principles of Chinese Family Law]. Tokyo.
Taiwan, Governor-General. 1905. Special Taiwanese Customs research group. Taiwan Shihu [Taiwanese Personal Law]. Taipei.
Tai Yen-hui. 1963. Ch'ing Tai Taiwan chih Chia Cliih-tu yti Chia Chan (Taiwan Family System and Estate in the Ch'ing Dynasty). Taiwan Wen Shian. vol. XIV, No. 3, pp. 1-19. Taipei.
Tambiah. S.J. 197:3. Dowry and Bridewealth and the Property Rights of Women in South Asia. Bridewealth and Dowry, cd. J. Goody and S.J. Tambiah, pp. 59-169. Cambridge.
Yalman, N. 1967. Under the Bo Tree: Studies in Caste, Kinship and Marriage in the Interior of Ceylon. Berkeley.
Legal and Social Transformation of Rural Women in Aegean Turkey 279
june Starr
INTRODUCTION
This paper links three independent ideas. First, it provides an alternative model to Engels' provocative theory, expounded in The Origin of the Family, Private Property and the State (first published in 1884).1 Engels suggested that women lose out in the historical process at exactly the point in time that capitalist enterprise develops in each society. Not only do women get squeezed out of the right to claim property for themselves and their children, but as marriage systems change from plural spouses to monogamy, women themselves become a kind of property for men. Monogamous marriage, according to
Reprinted with permission (rum Renee Hirschon (ed.), Women and Properly: Women As Property (New York: Si. Martin's Press. 1984). pp. 92-1 IG.
Engels, makes women dependent on men for economic support. Thus men become dominant and women become subordinate and submissive to protect themselves and their young."
Second, it asserts—contrary to much existing theory—that both written and unwritten law is never neutral on the issue of the relationship between the sexes. When law is silent, it supports the dominant power structure and cultural values of a society. The power structure is almost always controlled by adult males (sometimes with a few token women). When written laws specifically promote norms of equality, however, as in the case of Turkey, they provide a useful option for overturning the cultural bias which favours male dominance.
Third, it builds on the Ardeners' suggestive notion that women's models of the world
jnav be quite different from men's, because the men have generated the norms for the 3renas of reasoned public argument. Women thus may not be as good as men at articulating their unverbalised thoughts because they have not been socialised into modes of'public discourse' which is characteristically male-dominated (S. Ardener, 1975: xi-xvii).
Although Engels foresaw that women would be excluded from owning land as the economic system evolved from transhumance to settled capital intensive agriculture, he foresaw neither how the legal system nor how specific cultural systems would interact with the changing productive and marketing systems. This essay argues that the penetration of capitalist agriculture into the Bodrum region produced a class system which made marriages within a village an advantageous way of consolidating landholdings. Such marriages aid women in two ways. First, it keeps females in close proximity to their mothers, mother's sisters, and own siblings who provide a daily work and supportive group. This prevents young, impressionable brides from being psychologically intimidated into submission by a husband and his kin. Second, a wife's legal right to her share of the patrimony provides a powerful sanction to make a husband treat her well, because the new laws also allow a comparatively easy divorce1 for mistreatment. I argue that the gradual exposure of females to the law system in Bodrum allows them to learn the necessary forms of behaviour to use the law courts to their advantage. Finally, I suggest that laws providing female access to land, in combination with judicial willingness to enforce these laws, is a powerful mechanism for female emancipation in Aegean Turkey.
BODRUM: A CHANGING REGION
Many feminist scholars consider the modernisation process' as always adversely affecting the position of women/' Islam, too, is commonly thought to provide a cultural system in which women for the most part are totally subordinate to men, have few legal rights
and little or no autonomy.1' Turkey thus provides a unique situation in which to study problems of development relating to women because ninety per cent of its population are Moslem, and European legal codes were introduced in the 1920s. Furthermore, it is geographically, culturally and historically diverse. This diversity allows female/male relations to be contrasted across temporal, spatial, and cultural dimensions.
The particular focus of this paper is in the southwestern part of Turkey where the region takes its name from the town of Bodrum. In this essay we examine how male and female relationships are mediated by rights to property. In western countries property is identified with valuable resources such as land, houses, jewels and other highly-valued material goods which can be converted into saleable items on the market. In the Turkish region where I lived, orchards, houses, and productive fields were considered valuable property, as were cows, donkevs, camels, bicycles, jeeps, trucks and boats. Because women did not ride bicycles and were not taught to drive other vehicles, jeeps, trucks, bicycles and boats were owned and used only by men and do not figure further in this discussion.
In addition to material property, in Middle Eastern and Mediterranean cultures there is another valued resource, albeit intangible. This is honour. This essay argues that honour or reputation is also a valued possession, that is worth protecting and that it is as valuable to women as to men.' Furthermore, how a woman behaves affects the honour of her husband if she is married, and always that of her father and her brothers. This gives males social control over females, lessening women's autonomy. Much of female behaviour in the village intensively studied and in nearby Bodrum town only becomes understandable by knowing that honour and shame play a significant part in daily affairs. Like property, honour or reputation can be accumulated and can be lost." Ii is a scarce resource.
Questions raised in the paper are: how does access to property and other resources defined as scarce by the society, affect male/female relationships? Under what con-
280    Gender, Property, and the State
Legal and Social Transformation of Rural Women in Aegean Turkey 281
editions do women begin to assert their legally granted but customarily withheld rights to land, houses and other inheritance? Does a woman's changing relationship to property facilitate her emergence as an independent person with a growing ability to assert control over certain aspects of her life?
Turkey today is a complex nation-state involved economically with the European Common Market, Nato, and with its eastern neighbours." It has a small but growing industrial sector which was hard hit by the oil crisis oT 1973-74. Close to 65 per cent of the country is still agricultural. Poverty and lack of opportunities in rural areas led hundreds of thousands of migrant workers between 1960 and 1974 to seek employment in European countries."1
Turkey is divided into sixty-seven different administrative provinces (//, vilayet). There are strong class divisions, sharp cleavages between urban and rural dwellers (although migrant workers begin to blur these distinctions among the poor) and at least seven historic, cultural and geographic areas with rich distinctiveness." Differences exist between the two religious groups: the dominant Sunni and the minority Alevi (or Shi'ites, some of them remarkably heterodox). Throughout the 1970s tremendous political instability occurred, caused by violence among rival political groups. In September 1980 a military junta took over in a bloodless coup, ostensibly to restore order and to return to the principles of Ataturk.
Answers to questions concerning gender relationships and property need to be regionally, culturally and historically specific. Within the context of a changing social order this essay examines data collected from a village, Mandalinci, (cf. maiidalina, 'tangerines') (population 1,000) and a district town, Bodrum (population 5,200) from December 1966 through August 1968.'-
Bodrum region (kaza) is 66,000 hectares1,1 of which 22,614 hectares or just over one-third is farmed land. Bodrum town1! is the administrative centre for the thirty surrounding villages which vary in size from 293 to 2,000 people. In 1966 tangerines were grown in
walled irrigated fields in Bodrum and thevil-" lages to its west, while animals, tobacco, and5 wheat were cash crops grown in villages orf the Mumcular plateau to Bodrum's east.
Geopolitics and Economics
This section of the essay argues that marriage patterns changed with the changing economic, legal and social order. For centuries the Bodrum coastal region was inhabited by two ethnic groups: Christian Orthodox, Greek-speaking townspeople who inhabited harbour areas and Sunni Moslem, Turkish-speaking Yoriik sheep herders who practised transhumance (cf. Ramsay, 1917, pp. 31, S3). The Greeks farmed coastal valleys around harbours on the Ottoman mainland and were good sailors. The Turkish transhumants migrated between summer pastures near the sea and winter grazing areas further inland on the Bodrum peninsula.
A second Turkish-speaking ethnic group, remnants of the once powerful Turcoman confederacy,1'' occupied an ecological niche on the higher, inland plain commencing about 25 miles east of Bodrum town. These pastoral nomads had migrated over several centuries down to the region from the Anatolian plateau. They gradually settled into eight villages on the Mumcular plain about 125 years ago."' For cultural and religious reasons none of these three groups inter-married.
Between 1900 and 1919 the Greek and Turkish populations were on friendly terms. The Greek population farmed figs, olives, and wheat, and were the craftsmen of the region. They were carpenters, lime-makers, and house builders.1' The area now known as Mandalinci village was a summer camp ground (yayla) for Turkish-speaking Yoriik transhumants. Their winter quarters (kisla) were more protected. The Greek population also was larger in summer than in winter as attested to by ruins of houses and cafes along Mandalinci's deep water harbour (Starr, 1978, pp. 23-4).,s The population in the entire Bodrum region was in 1912, 8,817 Turkish people and 5.060 Greeks (Sotcriadis, 1918, p. 9).1'1
In this period Turkish women from nearby islands were considered the most beautiful and were desired as wives by Turkish-speaking men."11 Thus marriage practices reveal special socio-economic concerns: far-flung networks, embedded in transhumance, provided pastoral households access to diverse pastures, lands, brides, and information. For women, outward stretching networks meant that after marriage at the age of twelve to fourteen a girl was separated from her natal household for much of the year, because the groom was obligated to give labour to his father who had provided the bridewealth for his marriage. This created a virilocal post-marital residence pattern. But, groups moved with flocks between traditional camp grounds, population pressure was not considerable, land was not scarce, and mostly the Greeks owned private farmlands.
The increasing animosity between the Ottoman homeland and Greece from 1919 onward changed the situation. As news of the fighting between Greek and Turkish peoples in the Izmir area spread southward, Greek-speaking families fled from their Mandalinci seaside farmlands and cafes, abandoning the entire area to Turkish transhumant households. During the population exchanges of 1923 between Greece and Turkey, the Turkish government took an interest in Greek landholdings in the village area. Several elite Turkish households were granted farmlands in Mandalinci for their role in the war of 1919-1922 (Starr, 1978, pp. 23-27). Moslem Turkish-speaking people from Crete were moved into 'Rum* ward in Bodrum, now called Kumbahce (in Turkish Rum means Greek).'1
In the early 1930s tangerine agriculture was introduced from Rhodes into Betes, a seaside hamlet near Bodrum town."" The first orchard of tangerines in Mandalinci dates from 1940. To grow tangerines required a capital investment in three year old trees, as well as in a deep water well or overland cement waterways. It meant that soil had to be checked during dry months of summer (from mid-May until mid-October) to determine when the orchard should be flooded with irri-
gation water. Such water is raised from ground wells by mechanical lifts or motorised pumps.
Capital for intensive cultivation could be obtained through a bank loan, but to negotiate one, a person needed a legal title (tajní) and not merely usufruct rights to land. Elite families, of course, already had ties to banking personnel and they obtained much of the best farmland at valley level. Other villagers did obtain legal titles, while others still continued with traditional use rights to grazing lands or fields. There were recognised under village customary law-ways, but they had marginal status under state law until converted through the state legal system into a legally recognised form of ownership."1"1 Households owning a tangerine orchard (the only crop raised on irrigated land) tended to invest profits into building a house at the edge of their orchard.
The transformation to single Turkish occupancy of the region and to cash-cropping agriculture led to settled village life. This made privately-owned orchards a prime resource. Marriage between children of orchard-owning families developed, creating both dense kin ties within each seaside village, and an incipient class structure.
The impact of capitalistic agriculture, settled village life and an emergent class structure had profound effects on female/male relations and on females' access to property. But to comprehend fully these changes, we need to consider the third variable, the cultural system.
The Cultural Framework
In this essay culture has an ideological, institutional and behavioural component. It is viewed as the product of specific historical processes. But, cultural codes of behaviour, developed to cope with particular stresses in a certain historic period, may live on into a new era. Thus they can be viewed as transcending one productive system to emerge side-by-side with more adaptive forms of behaviour exhibited by some members of the group.
The value systems of the Turkish ethnic
282   Gender, Property, and the State
groups were based on male control ofTemales and a rather loose adherence to Islamic religion. Islamic attitudes toward women had co-mingled for centuries with Hellenistic attitudes in the Bodrum region.Jt Moreover, transhumant populations by and large are not known for their religious ardour."1 A daughter was under her fathers control until marriage. After that her husband had strict control and responsibility for her behaviour. Like most transhumant people in Western Turkey, veiling was not practised. Bride-wealth was given by the fiance to the girl's father in the form of sheep, and some gold coins were given to the girl. Lineages were shallow and blood feuding did not develop. The ideology of honour and shame, however, tended to keep males watchful of female actions.
The transition from transhumance to settled village life did not undercut the ideology of honour and shame, despite the pragmatic views of an emerging entrepreneurial class. Thus the Islamic notions of male dominance/female submission became pitted against the secular notions of the Turkish state which had enacted legislation giving equal rights to women.
Three Types of Marriage. The increase of capitalist agriculture and the involvement with the market prompted marriages to be arranged within the village between orchard-owning households. Marriages within the village consolidated property and focused, mobilised and united resources. The effect for women was to forge dense kin networks within a community. Keeping a young bride near for in the same village as) the parental household gave her some protection from an aggressive or cruel husband and some leverage against a demanding domineering mother-in-law.
Villagers distinguish three marriage modes: marriage by engagement negotiations, marriage by connivance and marriage by abduction.
Marriages arranged by negotiation (nikah) are never handled directly by the boy or girl. A mother first makes casual enquiries of her
Legal and Social Transformation of Rural Women in Aegean Turkey 283
relatives and friends as to the whereabouts of a suitable mate for her child. Then a series of negotiations is earned out first by the boy's father or father's brother on behalf of the youth, and at a later stage by the boy's father and mother with the father and mother of the: bride-to-be.
During these negotiations what is discussed is the amount and kind of land and; houses each spouse is due to inherit at the di-= vision of the patrimony. Types of land include a house and lot, irrigated orchards, fields and woodlands or grazing pastures. Discussions also include the amount and kinds of gifts the groom will give his bride in the bridewealth.
In Mandalinci the groom gives the gifts to the bride at the time the actual engagement (nisan) is celebrated. He cannot go with his parents when they cany his gifts to his fiance. The bridewealth for a middle class agricultural family customarily includes four or more gold bracelets, some gold coins for the girl to wear around her neck or forehead, a watch, head scarves, some cloth for dresses for the bride, her mother and her sisters. Shoes are given to the girl's father, and socks and hand- : kerchiefs to her brothers. The cost of such gifts in 1967 ran from seven hundred to three thousand Turkish lira ($700 to S3,000). In one instance, a 28 year old Mandalinci man and his father mortgaged the first good crop of their newly planted tangerine orchard to the man who lent them money to buy the engagement gifts so the youth could marry.
It is normal to wait at least three months between the giving of engagement gifts and the village wedding (diigun). But in Mandalinci the engagement often lasts much longer, because the groom may need to be away for military service, or the girl is not yet ready to leave home, or all the bridewealth has not yet been accumulated and given.
There are two modes of engagement and marriage in Mandalinci—early and later—reflecting economic differences, and especially the difficulties of poor households to accumulate the cash necessary for the bridewealth. The most approved form of marriage (and the only marriage mode available to the wealthy} is for the groom to marry when he is
eighteen or nineteen a female of about fourteen or fifteen. The marriage is celebrated and consummated before the groom leaves the village for two years of compulsory military service. In this case the new bride is brought to five at the parental house of the groom where a room or even a house is provided for the couple (the word for bride, gelin, also means daughter-in-law and is from gel— 'to come').
Most households cannot amass sufficient capital for early male marriages. Youths from impoverished families earn their own bridewealth which means they many much later, around 26 to 30 years of age. They work in the village as day labourers and tenant-farmers or outside the village as more job opportunities occurred with the expanding Turkish road system of the 1950s. The improved transportation also allowed more production of perishables and with it developed a prosperous fishing trade. Jobs were also available in sponge-diving and boat-building industries of Bodrum town.
The breakdown in obligations across the father/son generation in combination with tangerine agriculture has meant that virilocal residence patterns are giving way towards more neolocal households (compare Stirling, 1974). A newly married couple still may begin with virilocal residence (depending on the marriage mode and who provided the bridewealth), but many develop their own home separate from the groom's family. Ritual and emotional ties to both sets of parents are maintained, however.
A girl who did not wish to accept a proposal or whose parents were arranging a marriage not to her liking needed to convince her parents why that union was unacceptable. She might threaten suicide if they persisted. More usually, however, she found a youth she liked better and persuaded him to elope with her. A boy had many more options for avoiding a marriage not to his liking, thus underlining once again the gender asymmetry of rural Turkish society.
The second type of mairiage is by connivance, or elopement, (ktz kacirma).
The advantages of elopement is that the
groom does not have to give any bridewealth. It also allows both males and females to many the person of their own choosing. The girl often is the one to suggest it. The usual pattern is for the couple to flee in the night, have sexual relations and then go to the house of a friendly relative who will plead their cause to the girl's parents.
When the girl's parents notice her absence, they immediately report it to the nearest police station. The police will search for the couple. Once apprehended, they bring them back and formal charges are brought against the youth. Or the couple will reappear on their own and plead with the girl's family that they be allowed to many. Because everyone assumes they have had sexual relations (whether they have or not), the girl is no longer desirable as a local bride, since virginity is a prime requirement. Thus, unless the parents are vindictive they allow them to many. The fate of a girl who has eloped and not married is a worse shame to a family than a less wealthy bridegroom.
Whether they are apprehended or reappear on their own, a criminal case would be opened against the boy by the Public Prosecutor in Bodrum. Charges would be dropped when they produced a marriage licence for the court to see. The girl's active role in marriage by connivance challenges western stereotypes about submissive Turkish women.
The third marriage type is by abduction (zorla luz kacirma). Turkish villagers and Turkish criminal law distinguish between elopement by mutual consent (luz hac.irma) and forcible abduction {zorla luz kacirma). The Bodrum court and written law recognise a number of different actions and degrees of guilt, each canying more severe penalties. Thus, rape of a virgin who is a minor is more severe than rape of a virgin of legal age to many. Kidnapping and rape of a married woman also carried severe penalties.
A girl who has been forcibly abducted, kept against her will, and forced to have sexual intercourse, after a time may agree to many her abductor as the only solution to her future. It is the major way she can be reunited with her family and be re-admitted to local community
284   Gender, Property, and the State
Legal and Social Transformation of Rural Women in Aegean Turkey 285
life, albeit now as a married woman. Because of the norms regarding virginity in a bride (which are supported by the pervasive notions of honour and shame) the girl may realise that if she wishes to marry at all, she must agree to marry her abductor. Here is one victim's story:
I was on my way to school when he came with two friends and put me into his car and carried me off. He was a driver of a jeep between Milas and Bodrum and had noticed me. I was just a small girl. 3 didn't know about men. I didn't think about marriage. I was only twelve. He forced me to have sex with him . . . My father didn't open a court case against him because by the time they found us I was pregnant. ... It was hard at first, but. now 1 am more used to him.-'1
By the summer of 1968, after five years of marriage, they had two small children and had moved to a neighbourhood of Bodrum, near the girl's parents.
In the three year period 1965 through 1967 the Bodrum Middle Criminal Court (As-liye Ceza) processed 29 cases ranging from voluntary elopement to forcible abduction and rape (see Table 1). In 17 of these cases the couple married, so charges against the youth were dropped. We can assume that most of these were voluntary elopements on the girl's part. In three cases, each boy was sentenced to large fines and prison which suggests forcible abduction. (We can draw no conclusions from the five cases which had not finished by the time court records were copied, nor from the three cases where charges were dropped for insufficient evidence.) Thus
TABLE 1. Cases Ranging from Voluntary Elopement to Forcible Abduction and Rape (1965-1967 inclusive)
Married	17
Charges dropped	3
Innocent of Charges	1
Unfinished	5
Fine and Prison Sen-	3
tence	
Total	29
3 of 20 cases or 15 per cent were clear instances of violence against women.
Two of the five cases I witnessed were noteworthy. In one a girl changed her testimony. At the police station she had said she had gone willingly; in court she said it was under duress. Whether the police forced her to say she was willing to protect the lad, or whether her father forced her in court to say it was by force is unclear. In another case a girl said she loved the youth and they had eloped at her suggestion. When asked if they would marry, she looked downcast while her father stated, 'I have already married her to another'."7 There was nothing the judge could do to save the ill-starred romance. Paternal control had overpowered female autonomy and independence, and this father had outwitted the legal norms promoting female rights.
Post-Marital Residence and Social Class. Although Bodrum people still affirm the virilo-cal residence ideal, actual post-marital residence is linked to class, mode of production, resources and bridewealth. By the late 1960s an emerging pattern of class structure had developed, based on intermarriage of landowning households in the village. The strata were:
1. absentee landowning households, which controlled citrus orchards, which were farmed by tenant farmers (orluhc.i);
2. resident orchard-owning households, who did their own farming, hiring day labourers as needed, or had an orlakct;
3. resident field-owning households with no hired labourers;
-1. landless households, whose heads are tenant farmers or clay labourers for others.
With capitalist agriculture came absentee landowners and tenant farming. Tenant farmers were provided a small rent-free cottage for their services, which for women from poorer strata meant their own home, separated from their mother-in-law. Wives of day labourers often worked in the fields or orchards for wages themselves in order to add to the household income. In this stratum newly-married couples lived in whatever ac-
commodation could be provided for them by either family.
It was the wealthier households who could demand virilocal residence, because the father had the resources to build a room for the newly-wedded couple adjoining his house, and to provide the bridewealth which obligated the son to work his farmlands. Yet, some wealthier patriarchs chose to set up their older sons in small businesses in Izmir or Aydin. In two cases they married their younger daughters to village men with whom they established tenant farming arrangements. Such a son-in-law is called an ;'f gtivey (literally the groom who marries in). This allowed a young girl to stay in close contact with her natal household and provided a father with an assistant who may be more docile than his own sons. It also assured the girl's parents that they could mediate to a great extent the ways their daughter's husband behaved toward her.
Female kin living in adjoining households co-operated in food preparation, fieldwork, childcare, and sometimes in gathering vegetables and cutting tangerines. Mutual cooperation among female kin occurred even when sisters and mothers lived in separate parts of the village. Socialisation of village girls and also Bodrum women de-emphasised female rivalry and emphasised warm, mutually supportive relationships. Most girls thought it was a great advantage to remain near their mothers and sisters after marriage. The most adventurous village girls, however, dreamed of being married to a youth in Izmir or at least in Bodrum.
Mothers were also glad not to be separated from their daughters. Even more important, it meant they would be taken care of in their old age.
The Legal Framework
Islamic Law of the Ottoman Empire. During the 19th century several reforms in Ottoman Law affected the legal and social status of women. The major Ottoman innovation, however, its civil code, affected women only
tangentially. Known as the Mecelle, it was simply a modern-looking codification of pure Hanafi law. The committee prepared the code between 1869 and 1888, and published it sequentially between 1870 and 1877. The project was abandoned in 1888 when it proved politically unwise to produce a modern codification of Islamic family law (Onar, 1955, p. 295). Thus it left untouched the Serial, the core of Islamic family law which governed all aspects of family life and personal status including marriage, renunciation of wives, inheritance, and adoption of children.
Other Ottoman reforms affected women directly. For example, the old Ottoman tax on brides (ants resmi)—ot' 60 aspers for girls and 40 or 30 for widows and divorcees—was replaced by a fee for permission to many, given to the local Islamic judge (kadi). The new fee charged 10 piastres for girls and 5 for widows. Under the old Ottoman tax the amount and destination of payment was determined by the status of a bride's father. For widows' remarriage, however, the tax was paid where she resided or married (B. Lewis, I960, p. 679). The significance of the new fee was its implicit recognition of a relationship between an unmarried female and the place she lived. Atatiirk's secular laws continued the payment of a fee for marriage, which became a fee paid to a secular civil servant for a licence to marry.
Ottoman domestic legislation limited the bridewealth to a maximum of 1,000 piastres and specified that no gifts might be exchanged among the relatives of the bride and groom, nor brought by the wedding guests. The bridal dinner was limited to soup, wedding cake, and five other dishes. The bride was to buy her own cosmetics, but the groom was to pay for her use of the public baths (Young, 1905, vol. II, pp. 209-10).-"
Attempts to limit the amount of bridewealth and of gifts exchanged among relatives of the bride and groom again appear in the reformist Ottoman Family Laws of 1915 and 1917. But, the importance of the law of 1917, called the Ottoman Law of Family Rights, was its expanded application of the Serial. It allowed whatever school of Islamic
286    Gender, Property, and the State
Legal and Social Transformation of Rural Women in Aegean Turkey 287
law couples wished to use to be applied. This meant that the most flexible rule of any school, Hanafi, might be used. It also allowed a woman to have written into her marriage contract her right to annulment should her husband take a second wife. This was a major concession to those Europeanised reformers who were pushing for a monogamous marriage law in Turkey,1"1
The Ottoman Family Law of 1917 gave women rights to divorce on grounds such as impotence, insanity or abandonment. If a woman wished to divorce her husband on grounds of extreme cruelty or incompatibility, the law provided that three male family members must first attempt reconciliation of the couple before divorce was possible. Age limits were set, for the first time in Ottoman history, below which females and males could not many.111
But, inheritance practices continued as they had under the Serial. When a man died his widow had the right to one-eighth of his estate, and the remainder was to be divided among all his children (one-quarter if there were no children); each female share was to be half that or a male's. In practice in rural Turkey women rarely obtained their land in Ottoman times, and in Anatolian and Southeastern Turkey even in the early 1950s and 1960s women were denied access to the patrimony (Stirling, 1965, pp. 121-2; Aswad, 1978, p. 475).;u
In conclusion, despite the contractual nature of Moslem marriages, women suffered a number of disabilities under Islamic law. Girls moved from control by their father to the authority of their husbands. They had the status of a minor and could not act as independent persons. Women's share of the patrimony was half that ol their brothers', and the widow's one-eighth share of the husband's estate was not much reward for a lifetime of service. Furthermore, according to the law, a husband could turn a wife out at will by renouncing her in front of three witnesses; he had rights to the children produced by their marriage, she did not. Culture and circumstances may make this right of males under Islam less absolute
than the law provides, but strict application means children always belong to the agnauc line.
Secular Law Reforms of Ataturk. Under Ataturk's revolutionary vision of the early 1920s, women's rights in Turkey were: brought closer to men's. The new Turkish Civil Code, adapted in 1926 from the Swiss Civil Code, abolished male's right to divorce by renunciation and his right to plural wives. Monogamous marriage was established as the only form recognised by the State. A civil certificate, obtainable from a town clerk was the necessary prerequisite for registering a marriage with the state. To obtain a divorce, each party had to apply to the new secular law courts. Polygamy and bigamy were both made punishable by law, and children of polygamous unions were only given legitimate status by a series of separate legislative Acts.32 Women's rights to their paternal inheritance were now legally recognised as equal to their brothers'. A widow's share was increased to one-fourth the estate, and she got the first choice.
Age limits for marriages were again set; this time males were allowed to marry at eighteen, females at seventeen, and in exceptional cases both could marry at fifteen. The Turkish legislature in 1938 reduced the ages of marriage to seventeen for males and fifteen for females, and in exceptional circumstances with a judge's permission to fourteen for females and fifteen for males (Velidedeoglu, 1957, p. 63).
Ataturk's policies strongly opposed social and cultural symbols, such as the male fez and the female veil {carsaf). Turkish friends remember their mothers' stories of walking in city streets in the early 1930s and seeing soldiers rip veils off women. In rural areas of western Turkey, however, agricultural and nomadic women rarefy were veiled as it was a hindrance during work in the fields or with animals.
These new policies, codes and legislation promoting equality Tor women were put in place by a small western-oriented elite. They
did not occur in response to demands of a large segment of society, mobilised for action if rights were not granted. In 1934 this small elite even obtained by legislative act the right of Turkish women to vote in all elections (G. Lewis, personal communication).
Sir Henry Maine once remarked that law is alwavs out of step with society; there is always a gap between the legal rules and the existing social reality (Maine, 1861, p. 69). Thus, it is an empirical matter to establish the extent to which cultural practices changed under the impact of the new legislation. The remainder of the paper uses data from fieldwork in 1966 through 196S to assess the consequences of these new rights for women, especially the ways women's rights to property are changing their relationships to men in rural areas of the Boclrum region.
WOMEN'S ACCESS TO VALUABLES
Earlier we defined property as bridewealth, land, and houses. Reputation is also a valuable resource. It is hoarded in the required virginity of a bride, and in the care with which wives present a modest public demeanour. It is lost through careless and unchaste behaviour of women. A great compliment to a rural Turkish woman is to call her temiz, which means clean, virtuous.
Under the new codes a woman can defend her honour and reputation at court in lawsuits against men who attempt to seduce her, who solicit her favours, or who abduct and rape her. She can bring suit against other women and men for spreading rumours and slander about her, and for bearing false witness. She can oppose her husband's attempt to divorce her to marry a new wife and she can sue him if he takes a common-law wife. In other words, she has become a legal person with full adult status to act on her own behalf with legal rights that no Islamic law system ever gave her.
The extent to which she is using these rights is explored, in a preliminary way, below. But. first we discuss tangible property
because it is an implicit premises of this paper that access to self-sufficiency through owning fields, the means of food production, is an important value.
Through Her Life Cycle
Women's access to valuable property in Mandalinci combines both traditional practices and the growing penetration of the market into everyday life. The significant markers in the female life cycle which mediate her access to property arc:
(1) engagement, when she is given the bridewealth;
(2) marriage at about fifteen years of age when the moveable trousseau which she prepared is transferred to her new home;
(3) the death of her father, when the patrimony is divided:
(4) widowhood, when the patrimony of her husband is divided, and;
(5) old age (which mav occur with (4) above), when she becomes a dependent person in the home usually of a married daughter, but sometimes of a married sun.
From the ages of eleven to fourteen a girl will begin to embroider pillow cases, bed sheets, curtains and hand towels in anticipation of her marriage. The sheeting is bought for her by her parents.
During the bridewealth negotiations prior to marriage, she will have learned how much land, houses and family heirlooms (e.g. old pots, kilims, carpets, and some painted pottery) she will inherit when her father dies and his estate is divided. At the celebration of the engagement (nihait) she will be given the bridewealth, her fiance's gifts to her. In middle strata households these include gold bracelets, gold coins, perhaps a watch and cloth for dresses.
At the celebration of the village wedding, a bride will be given a small amount of cash by her parents and close relatives at the moment she is ready to leave her parent's house and mount the horse to be carried to the groom's house (or enter the hired jeep to travel to a
288   Gender, Property, and the State
Legal and 5ocial Transformation of Rural Women in Aegean Turkey 289
village further away). Her parents also give her a winter coat, guests bring cooking pots, cooking utensils, towels, plates, cups and other useful household items as gifts.
Bndewealth: Broken Engagements and Divorce.''"''' When an engagement is broken, no matter who is at fault all the gold and all the other gifts are returned to the (lance. If the cloth has already been made into dresses then the dresses themselves are returned. The candies and Turkish delights are the only things not returned, because they would no longer be fresh or might already have been eaten.
At divorce, however, none of the gold or presents are returned no matter if the woman is at fault. The bridewealth is considered 'the price of her virginity'. Only if a man married a woman to discover on her wedding night that she is not a virgin can he obtain the return of his bridewealth. But, a provident mother has probably tucked into the bosom of her daughters wedding dress a handkerchief dipped in chicken blood. Not only is the nuptial room a traumatic occasion for the couple, but they must produce a bloodied handkerchief for the ritual benefit of the groom's sisters waiting outside the door. At divorce all the 'things of the house,' the blankets, quilts, bedding, kilim, rugs, cooking utensils, pottery, heirlooms and sewing machine if there is one, belong to the woman. The house itself belongs to the spouse whose family provided it.
Orchards and Houses. Generally, the first opportunity a village woman has to own land is when her father dies. However, I have documented one situation in which a married woman sued her brother for her share of the patrimony and won rights to a house and grounds while her father was still alive (Starr, 1978, pp. 213-23). This was accomplished with her husband's help. He not only acted as her legal representative {vtkil.) in the lawsuit, but quietened her down during the court's visit to the disputed house and house lot in the village, when she began shouting at her brother (Star, 1978, pp. 216-7). As the angry woman and her husband already owned a house in which they lived, this represents a
clear example of accumulation of property by asserting a woman's rights to her father's estate.
This contrasts sharply with Anatolia in the late 1940s where 'a simple division of land be-tween sons seems to have been the normal' customary procedure' (Stirling, 1965, p. 122). It differs also from the Hatay region ofsouth-eastern Turkey in 1965, where 'women are also denied access to land through inheritance unless they are brotherless' (Aswad, 1971; 1978, p. 475). In the Bodrum region by the mid-1960s husbands were realising they could markedly increase their household's prosperity by utilising the wife's legal rights to land. Furthermore, during a year's observation of the Bodrum courts I saw numerous cases of division of patrimony in which female siblings actually appeared in court. Inheritance cases (verascl) and land division cases (tafaim) which represented 'routine' (as opposed to disputed) land cases made up the vast majority of the caseload in the High Court in Bodrum (Asliye Hukuk).M
Land at Divorce. In the three year period 1965 through 1967, 138 divorce cases were heard, and half were brought by wives. The usual grounds were incompatibility and hence most divorces were uncontested. The person who became complainant usually lived nearer the court or desired the divorce more avidly. No stigma was attached to being either defendant or complainant.
Division of land is comparatively easy at divorce, according to the judge interviewed, because each spouse keeps control of her/his orchard during the marriage. If a woman inherited an orchard from her father, the husband merely worked in it. If he bought a motorised pump for the orchard, he could take the pump out of the orchard at divorce. But, if the husband bought the fruit trees while the wife owned the land, the trees belong to the orchard, based on the legal principle 'the person who owns the land owns the trees' (toprak himin, agag onitn). If the husband built a house on her land, and there was no contract, the house belonged to the person who owned the land. If there was a contract, the person who owned the more expensive
thing, would buy the other out. (For a more detailed discussion of divorce cases, cf. Starr, 1983.)
Land at Widowhood. An estate is divided at the death of the patriarch. The legal principle is that the widow gets first choice of one-fourth of the estate, and the remaining parts are divided among all his legitimate heirs equally, regardless of gender. If the wife took the house and remarried, she could live there with her new husband. If the house was given to a child and she remarried , the child could ask her to leave when he/she was of legal age. If a man and woman had no children when he died, the wife still got one-fourth his estate and the rest was divided equally between the dead man's mother and father. But, if there was even one child, the wife still got one-fourth and the child the remainder. His parents then got no part of his estate. Property only went to the deceased's brothers and sisters if he had no children and his parents were not living.
When women were widowed after thirty or forty years of age they may have chosen not to remarry but to live near a married child. They then helped with cooking and childcare, and when they were too old to work they would be looked after by their children till they died.
Women's Use of the Courts
Women's interaction with the Bodrum courts is in part a function of their position in the life cycle and in part a reflection of their growing sophistication in defending their position in society. A young female is slowly socialised into viewing the courts as an institution capable of helping her in a crisis. As a child she may have watched the judge and court recorder come to their village to hold a court hearing to award title to land or to view the place of a serious crime or of a land dispute. Perhaps she had to apply to a judge to have her age raised as her father registered her birth several years after the fact, and without proper age she is too young to many. Or she may need to waive the required fifteen day period between obtaining a state licence to many and celebration of the wedding. If she
has eloped, she will have to appear in court to clear her husband of charges of abduction. If she was abducted, her testimony in court will determine his freedom or prison sentence. Later she may be called as a witness to a crime or to a neighbour's application for land title. She and her brothers may need judicial advice about how to divide their father's property. The largest number of civil cases in the Bodrum court concerns land division (taksim and verasel cases).
Thus in mid-life she is prepared to view the court as a major resource when her husband or brothers fail her."'"' For example, in August 1967, a woman entered the Bodrum judge's office and requested that the judge appoint her as guardian for her mother, who was 'insane'. 'She is giving land to all my brothers and sisters, but not to me' she said. The judge accepted the case and asked that the mother have a doctor's evaluation.11'
In a different case, Ayse, a fifty year old woman went to the Director of Bodrum (Kaymakam) to ask his intervention in a dispute with her neighbour, Hasan. The fifty year old man had built a wall for his orchard on Ayse's field. After viewing the disputed ground, the Director gave a verdict that Hasan must take the wall down from Ayse's land. When he didn't remove the wall, Ayse opened a criminal lawsuit against him in court. She paid for witnesses to appear and eventually she won her case.1'
In a third case, Zehra, a forty-five year old woman asked the kaymakam to prohibit Mehmet's farming of land she had bought. After reviewing the title and the site, the Kavnakam made judgment in her favour. But, Mehmet refused to leave. Zehra then opened a lawsuit claiming. 'I just bought this land a year ago, and I find him still farming it.'
Mehmet answered, 'We have been farming this land for thirty years now.'
The judge asked, 'Who has the title?' Zehra responded that she did. 'But', the defendant said, 'It is not her land. She went to the Kaymakam and he stopped my farming of it.'
Eventually, Zehra won."1H Here we see a woman who not only knows about farm titles but actively opposed a man who had usufruct land rights only. This case demonstrates a sit-
290    Gender, Property, and the State
Legal and Social Transformation of Rural Women in Aegean Turkey 291
uation where usufruct was predominant but is now being replaced by notions of private ownership and a woman is able to take advantage of these changing conceptions of property rights.
Women are using the courts not only to gain and keep their property, but also to protect that intangible valuable, reputation.
For example, Hafise, a widow of sixty-five years, had lived in her house for thirty-seven years, raising eight children there. A civil servant inherited (from his mother) the house next door to Hafise. Although Turkish law specifies that windows cannot be put into a side of a house overlooking someone else's walled courtyard, he had built a window in the back of his house, overlooking Hafise's backyard. In making the extension on his home, he had also cut clown most of Hafise's almond tree. In the summer of 1968 I went to Hafise's neighbourhood to find out how that case had ended. The following is part of an extended interview I had with Hafise:
Hafise said:
The cases are not finished and I have been in court for one year and three months now. I have hired a legal representative (vakil) so that I don't have to go every week to court. The case about cutting down the tree was decided in his favour, but we sent it to Appeal court in Ankara.
I asked if Hafise had opened a case to gain restitution for the destruction of the tree. Hafise answered:
No. I sent the dossier to the Appeal Court. I want to wait and see what they will do First. Maybe they will send him to jail. . . . He wanted to buy my wall. Do you remember that 'viewing' the court had of my garden and wall? Well, that 'viewing' determined the wall was mine. 1 said I'd sell the wall to him for 26,000 Turkish lira, but he offered me only 24,000 T.L. We then opened a different court case to determine the value of the wall. In this 'viewing' the judge only looked at the wall. They established that the wall was worth 1,300 T.L., but they neglected to look at the land it stood on. We are waiting for a viewing of the ground . . . Ill only sell if he pays my price . . . But, I still don't want his window open. Look, 1 wash in that garden. My toilet is
there. I bathe there. I am an old lady. Sometimes I go out in my don (baggy pants). Can he be always looking at me? I saw apartment buildings in Ankara; they are all open, but this isn't Ankara. Bodrum is a small, old town. We are not so modern here. He should have his win-: dow on a street or in his garden. . . . *!<J
Thus Hafise used the court to protect her modesty and her property. She had never been to court before this dispute. Nevertheless she pursued the restitution for her almond tree and the issue of his window through three different cases against her male neighbour, and when too many court hearings and postponements taxed her energies, she hired a local legal expert {vakil) fin-legal advice and court appearances.
In a final example, a twenty-two year old married woman, named Sevcihan, from Saz village (to Bodrum's east, adjoining the government forest) brought criminal charges at court. She accused a forest ranger, Mehmel, of molesting her after drinking with her husband while spending the night in their house. In court she told how two other forest rangers had offered her 500 T.L. to drop charges against their friend. Nevertheless, she pursued her grievance through five separate hearings over a five month time period. Fur each hearing she needed to travel twenty-five miles over rough terrain to court. On at least two occasions she had to bring witnesses and pay their expenses. On three different oc sions she had had to retell the events of that night:
The woman, Sevcihan, in court: That night the forest ranger and my husband came. They had been drinking at the coffee-house. They brought another bottle of raki to our house, and my husband told me to make some food read v. And then my husband became drunk and he fell asleep. I went to my husband's father, and 1 called him. He came to the house for a while and then he left. I went to bed next to my husband. I went to sleep. Someone is touching me and I woke up.
Judge: How is he touching you? Where is he touching you?
Sevcihan: He is stroking my neck, my breasts, my arms, my hands. I ran outside. He followed
me. I came back again. I went to the room where my husband was sleeping and I locked the door. He knocked at the window, saying, 'Come. I am waiting for you.' I went again to bed next to my husband.
Mehmet's version: I went home with her husband. We had been drinking at the coffeehouse, and we had a bottle of raki with us. And then her husband got drunk. He went to sleep. Later his father came and I offered him some raki, but he said, lI do not drink.' When he left I went to the bed they gave me. When I woke up in the morning I went to wash my face. She came to me, and I asked where her husband was. She said, 'He went. He went to the coffeehouse.' (He is the proprietor.) She gave me a cup of water. I drank it and then said. 'Say good-bye to your husband for me.' fudge: Do you go to their house often? Mehmct: This is the first time. Lawyer for Mehmct: That girl and her husband made a plan to get 500 T.L. from my client, judge: Do you have proof? What kind is it? Lawyer: Witnesses. Next time I will give a list of witnesses.
Sevcihan: That's not true. The next day two forest rangers came to me by Jeep. They said if you will give up this court case, we will give you 500 T.L.. But, I didn't want to give up this case.
The testimony of the witnesses can be summarised as follows:
One witness had said, 'I drive a jeep for hire. The day after that event I carried two forest rangers to her village, and they went to her house, but I didn't overhear the conversation there.' Another witness had testified that he saw two forest rangers going to her village, and had seen the defendant and her husband drinking in the village coffee-house the night of the alleged event. A third witness, a twenty-three year old woman, testifying for Sevcihan said:
About a month ago we were stringing tobacco leaves onto thread. This man comes up to Sevcihan and said, 'My dear, Sevcihan, why do you not come to me? Are you angry with me?' She didn't go to him. . . . No, 1 didn't see him touching her.'11'
Eventually, the charges against the forester were dismissed for insufficient evidence, de-
spite the female witness' testimony which implied that Mehmet was aggressively pursuing Sevcihan.
From a different perspective Sevcihan can be seen to have won her goal. Her ardent pursuit of justice through the court made Mehmet's actions public—his co-workers, her villagers, and even her debauched husband must now recognise his lecherous inclinations toward her. After five months of fearing a large fine and jail sentence (which would have meant loss of his job), Mehmet's lusts were probably tempered by prudence. Sevcihan's use of a district level court thus provided her the opportunity to vindicate her honour and safeguard her reputation.
BROADER IMPLICATIONS
In the Bodrum region factors which allowed women to develop more self-sufficient lives were: changing land-use patterns which constructed the daily work routines for both females and males; a change in post-marital residence in response to the emergence of a new class structure, which occurred as a result of the penetration of capitalist agriculture. And third, the law. judges' willingness to enforce legislation promoting norms of equality in union with women's growing knowledge of how to activate the official law system were emancipating mechanisms in western Turkey. In central and eastern Anatolia women's independence is apparently less advanced.'1 I would argue this is due to differences in the agrarian hierarchy, land use patterns, culture, historic conditions, and type of integration into the world market.
An explanation of Bodrum's successful acceptance of the new Civil Code may lie in several directions. Bodrum has a unique geographic and political position as a frontier of Turkish-Greek contact, and it was early pacified by the new Republic of Turkey. With good reason officials in Ankara would have wanted to keep Aegean Turkey pacified, economically productive and indoctrinated into the values of nation-statehood. Bodrum is much too accessible to the Greek Islands to
292   Gender, Property, and the 5tate
let it remain a backwater, illustrating the failure of the nation to maintain a western democratic outlook. Furthermore, it is an area of increasing productivity since the 1940s, and since 1968 Bodrum town has had a huge economic boom in summer months due to tourism. The winter population of the town has doubled between 1965 and 1980.'-
Earlier I suggested that in Aegean Turkey, Hellenistic and Islamic attitudes toward women had long co-existed as Turkish-speaking nomads over a period of several hundred years migrated into and settled in the region now called Bodrum. The ecology of transhumance required far-flung networks for the Yoruk sheep herders to gain access to diverse pastures in this multi-ethnic region. Marrying daughters to Turkish-speaking transhumants of different camps within the same ethnic group cemented pastoral relations. The Islamic ideology concerning male dominance and the required submission of females was supported by the institutions of bridewealth, post-marital virilocal residence, divorce by renunciation and the ideology of honour and shame.
Yet, despite the change from pastoral life to settled capitalist agriculture, cultural institutions such as bridewealth and virilocal post-marital residence remained as 'ideal forms'. Writing about European manners and cultural change in the emerging Renaissance Europe, Elias (1939) demonstrated that two or three hundred years may be necessary for ideas, etiquette, and new cultural practices to diffuse throughout a society. Yet, in the Bodrum region, work routines changed rapidly in response to new crops and a new productive system. Marriage patterns changed, too, as cash-cropping agriculture and the transformation of the class structure made marriages within the village a way to consolidate land holdings for middle and upper strata peasants. And third, the norms of sexual equality promoted by the secular legal system through both written law and judicial decisions, provided a way for females to gain control of resources, especially productive land, which for women {as for anyone), are a bridge to autonomous personage.
Thus, Engels', dynamic theory of the process of change from pre-capitalist social formations to capitalist relations neglected th< positive role law and culture play. In the re gion of Bodrum law balanced the disruptivf effect that capitalist agriculture and the emer gence of private ownership had on women' -lives. With increasing scarcity of land for in tensive agriculture, the norms of equal division of patrimony had become salient to husbands as a way to increase household land holdings. Elsewhere I noted that tangerint cultivation promotes nuclear households be cause two adults and two teenage children can provide all the needed labour from within (see Starr, 1978, pp. 38-42). Mastery of the economic processes behind tangerine mar keting is information easily accessible to any female who keeps her ears open.
Thus, I can unequivocally state that less than fifty years after the introduction of European Civil and Criminal codes in Turkey, women in the Bodrum region were reaping the benefits of laws of equality. They were able to hold titles to land in their own names. Some women successfully opposed husbands' attempts to usurp their economic resources during marriage and at divorce, and many women went to court to protect their landed interests and their reputations.
Some might argue that under the older system women were protected by fathers, husbands, and brothers; that going to court clearly indicates the breakdown of the older protective system."1"1 Didn't women lead better lives, they ask, in a material, social and qualitative sense in the past? I answer, that depends upon your goals for women. Data presented here clearly indicate, I think, that husbands, brothers and fathers do not always look out for a wife's, sister's or daughter's best interests. They may not even know them (even if they wished to) because some women may not be sure what their best interests are, while others may not be able to develop a plan which they can communicate by reasoned argument (cf. S. Ardener, 1975, pp. xi-xviii; E. Ardener, 1975). Given these facts it is better that women have ways to look out for their own interests, that they judge for them-
Legal and Social Transformation of Rural Women in Aegean Turkey 293
selves what these interests are, and that they develop habitual modes of thought and action which allow them to do so.
Therefore, the Bodrum example suggests that several factors need to intersect for women's emergence as more autonomous adults. The implication for policy makers is that legal rights and economic opportunities for women and for men must go hand in hand. The Bodrum study also suggests that we need to take a longer time span than twenty-five years''1 in deciding whether the results of change have improved women's lives. Of course we need to study the processes along the way. But, fifty years after the introduction of new secular codes, and twenty or more years after the gradual emergence of settled village life, we can see ways that women's access to valuables—land, houses, and reputation—are changing their relations to men, allowing them to become fully responsible persons.
NOTES
Acknowledgements. I owe a debt of gratitude to Dr. Geoffrey Lewis, Oriental Institute, Oxford, for taking the time to read critically this manuscript, although I take responsibility for its remaining faults. 1 also acknowledge with pleasure conversations and careful readings of the essay by Helen Callaway, Shirley Ardener, and especially Renee Hirschon.
1. Engels, of course, was not the only nineteenth century anthropologist to discuss woman's position in society in an evolutionary framework. But, precisely because his writing is the culmination of an anthropological perspective beginning with Maine, and developed by Bachofen, McLennan, Lubbock and Morgan, I choose to confront Engels's theories. Two recent critiques of Engels (1884) are Leacock (1981) and Sacks (1974).
2. See Engels (1981, pp. 120-21, 142-44).
3. Divorce is now easier for women and harder for men than it had been under the previous Islamic law system of the 20th century Ottoman Empire. Divorce was now accessible tinder the new Civil Code of 1926 by a spouse applying to the nearest secular district court
on one of the six grounds: adultery, dishonourable life of the spouse, desertion, mental infirmity, or incompatibility (Ansay and Wallace, 1966, p. 122). For a more detailed discussion Starr (1978).
4. Modernisation is here roughly defined as integration of the group into a nation-state. The linkages between the group and the state may, of course, be imperfectly achieved, e.g., the Kurds in Turkey.
5. See Boscrup (1970), Bossen (1975), Papanek (1977), and Nelson (1981).
6. Even those sympathetic to the cultural system of Islam and who advocate reform within Islamic law rather than a complete break, acknowledge Islamic law provides few rights for women and many disabilities when women's rights are compared to men's. See, for example, Couison and Hinchcliffe (1978) and White (1978. pp. 52-3).
7. The organising force of codes of honour and shame in Mediterranean countries has been argued by Campbell (1964), Davis (1977), Peristiany (1965), Schneider (1971), Schneider and Schneider (1976. p. 2) and others.
8. See Stirling (1955, pp. 98, 168, 230-3); Starr (1978, p. 56); Abel (1979).
9. Approximately half of Turkey's trade in 1982 has been with Islamic countries (77»,' Guardian, 12th May, 1982) and in the same year Turkey signed a major Trade Pact with Russia for 600 million lira {New York Times, 20th January, 1982, p. a7).
10. In 1960, 22,700 workers left Turkey (Abactan-Unat, 1981. p. 2). The figure continued to rise each year until the oil crisis of 1973-4. In 1980 the combined figure of Turkish residents in France, Germany, the Netherlands, Sweden and Switzerland was 1,762.9 thousand (SOPEMI, 1981, p. 3).
11. Fisher (1963, pp. 293-338) divides Turkey into five geographic regions, the Anatolian plateau, the Black sea coast, eastern Turkey, the Mediterranean, and the Aegean coast, but I suggest six. European Turkey ought to be separated from Aegean and Mediterranean Turkey at the Meander River.
12. The field research between 1966-68 was financed by a United States National Institute of Mental Health Predoctoral Fellowship and Grant and I gratefully acknowledge this support.
13. A hectare is 100 ares or 2.471 acres.
14. For a very interesting study of Bodrum town, see Mansur (1972).
294    Gender, Property, and the State
Legal and Social Transformation of Rural Women in Aegean Turkey 295
15. Fieldwork revealed cultural differences between villagers on the Bodrum peninsula to Bodrum's west, and those on the Mumcular plain to Bodrum's east which villagers themselves recognise saying 'They are very different from us'. Identifying which villagers were Vuri'ik and which Turcoman was harder and is the topic of current research. But see Field (1881. pp. 62-3), De Pkmhol (1958, pp. 526, 528, 531) and Ramsay (1897, pp. 100-1; 1917, pp. 31,83).
16. The transition from pastoralism to settled village life and the identity of these villages was first suggested to me by Osman Nuri Bilgin, Director of the Primary School in Bodrum, and a historian of the Bodrum region.
17. Interview with the Director of Rural Agriculture, Bodrum.
18. In the later 19th century the Turkish population was losing control of the Aegean areas to Greek-speaking farmers and shepherds. (Ramsay, 1897, pp. 130-31, 133). See also Starr (1978, pp. 23-5).
19. The population from the 1965 Census lists Bodrum town as having 5,136 while the surrounding villages are placed at 20,675 (Genel Niifus Sayimi 1965, p. 483).
20. Most women fifty vears or older also remember Greek, for they came from islands of Kos and Kalimnos as brides.
21. See note 17.
22. Marketing tangerines only became feasible with the completion of a dirt road linking Bodrum to Milas in 1927. because tangerines ripen between December and March, the period of sudden, violent storms on the Aegean. This makes sea transport particularly precarious at this season.
23. The procedure of converting usufruct rights to a state recognised legal title (lafm) involved going to court and applying under Art. 639 of the Turkish Civil Code. 'If the land is not previously registered in the Land Registry and if the person occupies and uses the land as if he were the real owner for 20 years without interruption and dispute, he may request a court to order the registration of the land in his name.' (Letter from Prof. T. Ansay, Dean of Ankara University Law Faculty. 14 December 1980).
24. Cosar (1978, p. 131) astutely observes that as one moves from east to west in Turkey the 'situation of women improves with the general socio-economic situation'.
5. The lack of religious behaviour among Turkish-speaking transhumants has been noted by Barth for the Basscri, (1961, p. 135) and by R.
Tapper for the Shahsevan (1975, pp. 2, 155, 158, 164). But, see Beck for women's religious and ritual practices among the Turkish-speaking Qashqa'i (1978, pp. 363-5).
26. Interview with Ayhan A., concerning Bodrum Court Case, B.C. 62. Bodrum Court Cases File. .Also see Case 11 on film, Adliye: An Ethnography of a Turkish Rural Court,' 1968.
27. Fieldnotes, filed under Kiz Kartrma (Elopement) Cases, 1967.
28. This is interesting as an attempt to use legislation to regulate custom and tradition.
29. Sec .Allen (1935, pp. 137-9), Coulson (1964, p. 184), Starr (1978, ftn.2,pp. 1-2).
30. SeeB. Lewis (1961, pp. 225-6).
31. Maher (1978, p. 102) makes the same point for rural Moroccan women. The male lineage members justify this deprivation of inheritance by saying that if daughters were given their land, it would be transferred to another lineage when they married.
32. The Turkish National Assembly enacted laws legitimising children of irregular marriages in 1932, 1934, 1945, 1950. 1955, 1965 and 1974.
33. The following is a summary of two days of discussions I had in August 1967 with the senior court judge and the Public Prosecutor concerning female and male property rights and bridewealth. at critical times in the life cycle or when engagements were broken or marriages were dissolved.
34. See, Starr and Pool (1974, pp. 552-54) for an analysis of women and men's use of the courts in Bodrum. Also Starr (1983) for women versus men in divorce suits.
35. Additional confirmation of this assertion is available. Female complainants against male defendants as a total of all cases processed in the Bodrum Middle Criminal Court rose from 14 per cent in 1950 to 28 per cent in 1967 (Starr and Pool, 1974, p. 353).
36. Witnessed by mc 7 August 1967. Filed under Conversations with Bodrum judges and Public Prosecutor, p. 33. titled, 'Opening a Case, Suth Hukuk (Lower Civil Court).'
37. Bodrum Court Cases, File. Sulk Ceza (Lower Criminal Court), B.C. 23, 1967.
38. Bodrum Court Cass, File. Sulh Hitluik (Lower Civil Court), B.C. 78.
39. Bodrum Court Cases, File. Asliye Hukuk (Higher Civil Court) B.C. 72 and Asliye Ceza (Middle Criminal Court) B.C. 85.
40. Bodrum Court Cases, File. Snlli Ceza (Lower Criminal Court) B.. 4 IB.
41. For overviews and comparative statements
about women in Turkey, see Abadan-Unat (1963, 1978 and 1981), Cosar (1978), and Kandiyoti (1977, 1980). For ethnographic accounts of women's position, see Aswad (1967, 1974, 1978).
42. Winter population figures for 1980 were: Bodrum town—10,000 people; Bodrum district (kaza) including the town and all the villages had 38,000 people. (Personal communication, Mrs. Emine Cam, Director of Tourism Bureau in Bodrum.
43. Nader(3964, 1965) hypothesized that women in Oaxaca, Mexico used the court only when they did not have a husband, father or brother to protect their interests. In Turkey, however, family structures, law and the market interact so that it is frequently a brother or husband who has usurped the women's resources (see Stirling, 1957. p. 27); or a father may exploit his daughter for his own financial gain (see Stirling, 1957, p. 31). Even in areas of strong kin group control, 'the "protection" of the lineage which had previously been to the economic advantage of the woman is turning increasingly into dominance and exploitation' (Aswad, 1978, p. 475).
44. Twenty-five years was the time period of the evaluative conference in Istanbul, entitled the 'Reception of Foreign Law in Turkey,' which essentially was pessimistic (see Stirling, 1957, and Velidedeoglu, 1957).
REFERENCES
Abadan-Unat, Nermin, 1963, Social Change and Turkish Women. Publication of the Faculty of Political Science. University of Ankara, No. 17.
-.  1978. The Modernization of Turkish
Women. The Middle East Journal 32: 291-306.
-. ed. 1981. Women in Turkish Society. Leiden:
E.J. Brill.
Abel, R. 1979. The Rise of Capitalism and the Transformation of Disputing: From Confrontation over Honor to Competition for Property. UCLA Law Review 27(1): 223-255.
Ansay, T. and Wallace, Jr. 1966. Introduction to Turkish Law. Ankara: Guzei Insianbul Matbaasi.
Ardener. E. 1975. Belief and the Problem of Women. In J, La Fontaine, ed. The Interpretation oj Ritual. London: Tavistock Press.
Ardener, S. 1975. Introductory Essay. In S. -Ardener, ed. Perceiving Women. New York: Halstcd.
Aswad. Barbara. 1967. Key and Peripheral Roles of Noblewomen in a Middle Eastern Plains Vil-lage. Anthropological Quarterly 40: 139-152.
-. 1971. Property Control and Social Strategies: Settlers on a Middle Eastern Plain. Anthropological Papers No. -t-f. Ann Arbor, Michigan: University of Michigan.
-. 1974. Visiting Patterns Among Women of
the Elite in a Small Turkish City. Anthropological Quarterly 47; 9-27.
-. 1978. Women, Class and Power: Examples
from the Hatay, Turkey. In Lois Beck and Nikki Keddie, eds., pp. 473-481. Women in the Muslim World. Cambridge: Harvard University Press.
Boserup, Ester. 1970. Woman's Role in Economic Development. New-York: St. Martin's Press.
Bossen, Laurel. 1975. Women in Modernizing Societies. American Ethnologist 2(4): 587-601.
Campbell, John K. 1964. Honour, Family, and Patronage. Oxford: Clarendon Press.
Cosar, Fatna Mansur. 1978. Women in Turkish Society. In Lois Beck and Nikki Keddie, eds., pp. 124-140. Women in the Muslim World. Cambridge: Harvard University Press.
Coulson, Noel and Dorecn Hinchcfiffe. 1978. Women and Law Reform in Contemporary Islam. In Lois Beck and Nikki Keddie, eds. Women in the Muslim World, pp. 37-52. Clam-bridge: Harvard University Press.
Davis, John. 1977. People oj the. Mediterranean: An Essay in Comparative Social Anthropology. London: Routledge and Kegan Paul.
Elias.N. 1982. The Civilizing Process, Vol 1. The Histoid of Manners. Oxford: Blackwell. Originally published as Uber den Prozess dcr Zivilisation, 1939.
Engels. F. 1981. The Origin of the Family, Private Property and the State. Edited with an introduction by E.B. Leacock. London: Lawrence and Wis hart.
Kandivoti, D. 1977. Sex Roles and Social Change: A Comparative Appraisal of Turkey's Women. Signs 3(1): 57-73.
-. ed. 1980. Major Issues on the Status of
Women in Turkey: Approaches and Priorities. Ankara: Cag Matbaasi.
Leacock, Eleanor B. 1981a. Myths of Male Dominance. London: Monthly Review Press.
Leacock, Eleanor B. 1981b. Introduction to Engels' The Origin of the Family. Private Property and the State. London: Lawrence and Wishart.
Lewis, B. 1961. Arus Resmi. In Encyclojmdia of Islam, Vol. 1, p. 697.
Maine. H.S. 1861. Ancient Laiv: Its Connection with the Early Histoiy of Society and its Relation to Modem Ideas. London: John Murray.
Nelson, N. 1981. Introduction to African Women in the Development Process. Journal of Development Studies 17(3): 1-8.
296    Gender, Property, and the State
Onar, S.S. 1955. The Magalla. Law in the Middle
Ens! \: 292-308. Papanek, Helen. 1977. Development Planning tor
Women. Sign* 31: 14-22. Peristiany, John C. ed. 1965. Honour and Shame: The Values of Mediterranean Society. London: Weidenfeld and Nicolson. Ranisav. W. M. 19f7. The Intermixture of Races in Asia Minor, Same of its Causes and Effects. Proceedings of the British Academy, Vol. 7, pp. 31. 83. London: Oxford University Press. Sacks, Karen. 1974. Engels Revisited: Women, The Organization of Production and Private Property.  In  Michelle Rosaldo and Louise Lamphere, eds. Woman, Culture and Society, pp. 207-220. Stanford: Stanford University Press. Sotenadis, G. 1918. An Ethnological Ma/) Illustrating Hellenism in the Balkans Peninsula and Asia Minor. London: Edward Stanford. Schneider, Jane. 1971. Of Vigilance and Virgins: Honor. Shame and Access to Resources in Mediterranean Societies. Ethnology 10(1): 1-24. Schneider. Jane and Peter Schneider. 1976. Cul-
ture and Politico! Economy in Western Sicily. New York: Academic Press.
Starr. |. 1978. Dispute and Settlement in Rural Turkey; An Ethnography of Law. Leiden: E.J. Brill.
Stirling, Paul. 1957. Land, Marriage and the Law in Turkish Villages. International Social Science Bulletin 9: 21-33.
-. 1965. Turkish Village. Loudon: Weidenfeld
and Nicolson.
-.  1974. Cause. Knowledge and Change;
Turkish Village Revisited. Inj. Davis, eel. Choice and Change: Essays in Honour of Lucy Mnir. London: Athlone Press. Velideclcoglu, H.V. 1957. The Reception of the Swiss Civil Code in Turkey. International Social Science Bulletin 9: 60-65. Voting, G. 1905. Corps de Droit Ottoman. Vol. 1L .
Oxford: Clarendon Press. White. Elizabeth H. 197S. Legal Reform as an Indicator of Women's Status in Muslim Nations. In Lois Beck and Nikki Keddie. eds., Women in the Muslim World, pp. 52-68. Cambridge: Harvard University Press.
The study of kinship has been central to cross-cultural research. Marriage customs, systems of descent, and patterns of residence have been described and compared in a range of societies around the world. At the heart of traditional studies of kinship is the opposition between the domestic domain on the one hand and the public, political, and jural domain on the other. Anthropologists, particularly those working in Africa, studied kinship in this latter domain. They delineated large corporate descent groups called lineages that managed property and resources and that were the basic building blocks of political organization (Fortes 1949, 1953). Marriage, for some kinship theorists, is a political transaction, involving the exchange of women between men who wish to form alliances (Levi-Strauss 1969; see also Ortner, this book). A woman, from this perspective, is a passive pawn with little influence over kinship transactions. She is viewed "in terms of the rights her kin have to her domestic labor, to the property she might acquire, to her children, and to her sexuality"
(Lamphere 1974:98). The dynamic, affective, and even interest-oriented aspects of women's kinship are essentially ignored in an approach that is rooted in androcentric principles: women have the children; men impregnate the women; and men usually exercise control (Fox 1967).
Recent critiques of the traditional study of kinship have pointed out that it is "no longer adequate to view women as bringing to kinship primarily a capacity for bearing children while men bring primarily a capacity for participation in public life" (Collier and Yanagisako 1987:7). A gendered approach to kinship takes a number of different directions but focuses on the status of men and women in different kinship systems and on the power (defined as the ability to make others conform to one's desires and wishes) that accrues to women through their manipulation of social relations. Cross-cultural variations in the status of men and women have been examined in relation to rules of descent and postmarital residence (Martin and Voorhies 1975; Friedl 1975). Among horticulturaiists, for example,
297