9 GENDER AND AGE All animals are equal, but some are more equal than others. — George Orwell In all societies, there are differences in power between persons. There is not a single society where all adults have exactly the same influence over every decision, where everyone has exactly the same rights and duties. Social differentiation and inequality are, in other words, universal phenomena. The Romantic ideas of the eighteenth and nineteenth centuries about the 'original primitive society', where all humans supposedly had the same rank and were political equals, was completely devastated when the first professional ethnographers returned from the field. Even among very small groups, and even among peoples with very simple technology, differential rank existed. Unlike what many 'armchair theorists' thought, small-scale societies do not lack internal differentiation. Marx and Engels were thus right in assuming, in the mid-nineteenth century, that there are universal criteria for social differentiation, which are the distinctions between older and younger people, between men and women and between insiders and outsiders, 'us' and 'them' (the latter is dealt with in Chapters 17-18). Not all social differentiation entails unequal access to rank and power, and we may therefore distinguish between vertical and horizontal differentiation. The vertical dimension refers to inequalities in power or rank, while the horizontal dimension encompasses those aspects of social differentiation which do not express unequal rank - differences which may, for example, be expressed through the division of labour. Most forms of social differentiation nevertheless have a vertical or hierarchical aspect, which is frequently contested by people encompassed by these forms. There are enormous and analytically important variations in forms of differentiation. In small-scale societies with a limited division of labour, such as hunter-and-gatherer societies, social differentiation is simple and may in a few cases just be defined by age and gender. In many horticultural and agricultural societies, religious leaders and chiefs have recognised statuses setting them apart from the rest of the people, although those statuses are not necessarily hereditary. In more complex agricultural communities, there tend to be hereditary political offices, often a professional bureaucracy and a professional army, and there may be great differences between the rich and the poor; the powerful and the powerless. In some societies, ascribed statuses 124 Gender and Age 125 (such as caste in India) are decisive in formal social differentiation; in others, achieved statuses may be more important in determining a person's social rank and place in society. There are many theoretical perspectives on such criteria for social differentiation as gender, age, class and caste. Some of the most vigorous debates in anthropology have indeed dealt with issues related to differentiation, power and rank. This and the following chapter, which discuss some of the most widespread criteria for differentiation, are therefore far from exhaustive. GENDER There are two fundamentally different ways of regarding gender differences. On the one hand, there are certain biological differences between men and women; the genitals look different and function differently, women give birth to children, men usually have larger bodies, and so on. At this level, it is customary to use the term 'sex' instead of gender. On the other hand, in practice gender differences are codified and institutionalised socially and culturally, and it is largely this kind of difference anthropologists focus on, which is distinguished from (biological) sex by the term 'gender'. All human societies conceptualise differences between men and women, and all consider such differences to be important in certain regards. However, there are important variations in the ways the relationships between men and women are worked out, and therefore it is difficult to generalise about gender. Gender can best be studied as a relationship - men are defined in relation to women and vice versa - and this relationship is conceived of differently in different societies. Gender was for a long time relatively neglected in anthropological research, which is perhaps surprising, since gender identity may well be the most fundamental basis for personal identity. Malinowski, who has often been praised for his ethnographic detail, is now said to have neglected important women's institutions completely and exaggerated the contribution of men to the reproduction of Trobriand society (Weiner 1988). In many other classic studies too, social actors are more or less seen as equivalent to social men. This kind of perspective may be called androcentric (andros is Greek for man). It seems to be changing, and gender has been a central topic in anthropological research since the 1970s. Some of this research has concerned the notion of the 'original matriarchy': the idea, common in nineteenth-century scholarship, that human society was originally ruled by women. In this regard, it has been argued that such notions, which are widespread among many peoples, are probably myths created by men to justify their own power over women (Bamberger 1974). The relative powerlessness and allegedly universal subjugation of women have also been subjected to a great deal of comparative research, forming the basis of debate both in favour of and against Western notions of 'dis- 126 Small Places, Large Issues crimination' and 'power'. It has been argued that, although women in many societies are deprived of formal political power, they may exert considerable power domestically and indirectly. It has also been suggested that the concepts of anthropologists dealing with discrimination may be ethnocen-trically biased. Even if women in traditional Middle Eastern societies seem discriminated against and powerless from a European perspective, they may perceive their situation otherwise. Some of the most recent anthropological research on women does not primarily deal with the 'position of women', but instead concentrates on different aspects of the male-female relationship in different societies. Several anthropologists have even discovered that not only women but men, too, are gendered (e.g. Herzfeld 1985; Herdt 1987; Gilmore 1989; Archetti 1999). Gender is omnipresent and easily visible; perhaps this is one reason why it has suffered neglect from anthropologists who have regarded it as 'natural'. However, like kinship and ethnicity, gender is not natural and God-given, but socially created. The great cultural variation regarding conceptions and conventions about gender reveals this. Probably all societies hold notions about the 'naturalness' of certain gender differences, but such ideas are themselves cultural constructs and not a part of nature. A contemporary anthropological perspective on gender would thus not try to answer the question 'what is gender really?', but would rather look into 'the social and cultural construction of gender ... [and] the relationship between constructions of maleness and femaleness' (Strathern 1988, p. 69). GENDER IN THE DIVISION OF LABOUR Even in societies with a simple division of labour and little occupational specialisation, women's work is distinguished from men's work. We have already seen examples of this from the Yanomamö and the Fulani. The most typical cases of communities where a division of labour based on gender is the most important are nevertheless hunter-gatherer, or foraging societies, and here it may be worth noting that humanity has consisted exclusively of hunters and gatherers for the greater part of its existence. Some readers may have noted the use of the cumbersome term 'hunters and gatherers' instead of simply 'hunters'. The reason for this is actually an increased understanding of the importance of gender. Among such peoples, men usually hunt and women usually forage. For a long time, anthropologists and other outsiders believed that hunting was the main source of livelihood among these groups. The men of these societies spoke incessantly about hunting, and visitors were given the impression that this was the single most important activity. Detailed research on nutrition among some such groups, notably in Southern Africa, has nonetheless revealed that the most important source of nutrition are tubers, insects, edible plants and small creatures gathered by the women, while the men's hunting activities are Gender and Age 127 irregular, uncertain and form no reliable basis for subsistence. Among a San group studied by Richard Lee (1968), it transpired that the people received two-thirds of their nourishment from what was brought in by gathering. Nonetheless, the men saw their society as a hunting society, and the women's routine work was not given the same symbolic importance accorded to hunting. It seems reasonable to assume, as a hypothesis, that the influence of women is most significant where their economic contribution is important. In a classic work, Ester Boserup (1970) compared the division of labour in a number of agricultural societies with reference to gender. Her main conclusion, which was surprising to many at the time, was that women in many societies, especially in Africa and in communities which do not use draught animals, carry out the main bulk of the agricultural work. It also seems that their influence is generally greater in these societies than in places where they are exclusively responsible for the domestic sphere. Boserup's controversial conclusion is that the political position of women in farming societies is generally weakened when new technology is introduced, since this technology tends to be controlled by men. In addition, land is usually controlled by men. Boserup argues that the position of women is weakened when their economic contribution diminishes; this is nevertheless a thorny issue. For what is an economic contribution? As indicated, foraging may be less spectacular than hunting, and less ritualised and talked about (just like housework in industrial societies), but it is an economic contribution nonetheless. The following example may illustrate the relationship between gender relations, the economy and the realm of ideology and power. The Mundurucu live in the south-eastern Amazon basin (Murphy and Murphy 1985) and traditionally have a way of life comparable to that of several other Amazon forest peoples. They are politically integrated at the level of the village; each village, which has between 50 and 100 inhabitants, is politically independent. They are horticulturalists and grow tubers, fruit and vegetables, apart from fishing and hunting. That is, the men go hunting. The division of labour is based on gender. The men are responsible for clearing the ground, but planting, weeding and harvesting are women's work. The women also gather wild fruits and nuts, while men and women cooperate in fishing. The Murphys write: The men think of themselves essentially as hunters, not as gardeners or fishermen, the religion is oriented towards hunting, and the spirit world is closely associated with the species of game. It is the skillful hunter who is honored, not the industrious tiller of the soil. (Murphy and Murphy 1985, p. 88) In fact the horticultural activities are more important, from a nutritional point of view, than both hunting and fishing, and they also entail much more work. The issue, however, is not purely to do with nutrition: power and prestige are also at stake. If, therefore, it is considered a more valuable thing to shoot a single antelope than to grow a tonne of taro, it does not necessarily 128 Small Places, Large Issues make a difference that the taro is more important to the group's survival than the antelope is. Mundurucú society is politically dominated by men. Like men in many male-dominated societies, they tell stories of an original matriarchal social order when 'everything went wrong', before all was eventually transformed by a mythical cultural hero. Such myths must be considered an important aspect of ideology: together with the disproportionate emphasis put on hunting as a means of livelihood, they contribute to legitimating (or justifying) male supremacy. It may nevertheless be noted that if the Mundurucú regarded female subordination as 'natural', there would be no need for such ideological myths: the stories are told as warnings that the women may rise again unless the men are vigilant. Why, one may still ask, do men have political and economic supremacy in most known societies, when it is not true that men contribute more than women to the physical survival of the group? Let us consider some attempts to account for this more or less universal power discrepancy. THE PRIVATE AND THE PUBLIC Although women may carry out as much or more work than men, they are nearly universally responsible for domestic work- child-raising, cooking and cleaning. Men, on the contrary, tend to be responsible for the household's dealings with the outside world. In hunting and gathering societies, they have to protect the woman and her suckling infants from dangers; in more specialised societies, they tend to hold political and ritual offices. Some anthropologists have seen a principal cause of the subordination of women in the fact, ultimately rooted in biology, of women's lack of physical mobility during pregnancy and suckling. These impediments to free movement serve to link the woman to the home while the man is free to roam the public space. It has thus been suggested (Rosaldo 19 74) that power discrepancies between the genders are related to the distinction between the public and private spheres in society: men control the former and women are confined to the latter. The Murphys' monograph on the Mundurucú suggests that such a distinction may have something to recommend it. In the Mundurucú village, only women and children live in huts; all men live in a men's house. This house has no walls; it is constructed as a large leaning roof supported by long poles. Its open side faces the village, so that the men can at any time follow events there. The woman's place is at home and in the fields, while the man's place is in the public space, where decisions of importance for the whole village are taken. Ortner and Whitehead (1981) have added to this perspective that there is a general tendency to the effect that women are, rightly or wrongly, culturally associated with private and particularistic projects, while men are associated with the public and common good. They argue: Gender and Age 129 [This] relates to a widespread sociological distinction suggested by Rosaldo (1974): Nearly universally, men control the 'public domain', where 'universalistic' interests are expressed and managed, and, nearly universally, women are located in or confined to the 'domestic domain', charged with the welfare of their own families. (1981, p. 7) This distinction has been criticised as ethnocentric - it is said to be meaningful in modern societies but not necessarily in others - but it is nevertheless well established and implicitly assumed in much comparative research on gender. DOMINANCE AND SUBMISSION Is the subordination of women universal - do women everywhere have a lower rank and less power than men? The most common answer is yes, but the question is far from unproblematic. First, there, are very significant variations between gender relations in different societies, ranging from nearly complete equality (as among the Chewong of Malaysia; see Howell 1989), to societies where the women's influence over their own destiny seems very limited. Second, concepts such as 'rank', 'subordination' and 'discrimination against women' are themselves problematic. Perhaps, it has been suggested, the anthropologist's own cultural background creates an unhealthy analytical bias here, in that he or she assumes, as a matter of fact, that equality is desirable, while many of the people studied by anthropologists insist that the genders ought not to be equal but should rather be complementary. Finally, it is not entirely certain that men and women understand the same thing by power and power discrepancies. Perhaps women care little about what the men see as prestige objects? One group frequently mentioned as marked by a high degree of gender equality is traditional Hopi society in what is now the south-western United States. 'When traditional Hopi women are asked "Who are more important, women or men", a common reply is "We are, because we are the mothers", with the qualification that men are important, too, as the messengers to the gods' (Schlegel 1977, p. 186). The Hopis, a North American Indian people famous for having preserved important aspects of their traditional culture, have a matrilineal kinship system. Unlike in most matrilineal societies, a woman's brothers and mother's brother are not strong authorities impeding her freedom of movement and agency. The spouses are considered equally valuable and complementary, and the lineage as well as the household has a female head. On the other hand, men have formal political and religious power at the level of the village, although, as Schlegel remarks, 'women do not hesitate to speak their minds' and, as 'Hopi men readily admit, women usually get their way' (1977, p. 195). At an intermediate level between the household and the village is the clan, which is led by a brother-sister pair. 230 Small Places, Large Issues The division of labour among the Hopi is unexceptional, comparatively speaking. Only the men go hunting, and only the women grind flour; however, they share the agricultural work. Schlegel thus suggests that there may be a cultural, and not an economic, cause for the high level of gender equality among the Hopi: 'Where the ideological focus of a culture is life, and both sexes are believed to be equally necessary to the promotion of life, devaluation of either sex is unlikely' (19 77, p. 205). Could this be adequate as an explanation? It is true that some male-dominated societies regard the woman's role in reproduction as marginal; the male seed is seen to play the crucial part. In many patriarchal or otherwise male-dominated societies (as in the New Guinea highlands), it is believed that women create the soft, perishable parts of the body, while the men create the skull and bones. But could this kind of gender ideology be the cause of inequality between the genders, or is it rather the effect of such an institutional order? No general answer to this question is proposed here; suffice it to note that in actual social process, ideology and practice function together and can only be separated analytically. In other words, it is not easy to point out what it is that determines what. What does it mean that women tend to be 'subordinated'? Even in otherwise egalitarian societies, women rarely, if ever, hold high religious office. Among the Mundurucu, only men are allowed to play and even see the sacred trumpets used in ritual; even among the Hopi, only men can be high priests. On the other hand, it is not obviously true that priesthood is the ultimate expression of power in society. In fact, women who are deprived of public office - be it political or religious - have been known to exert considerable power informally. During my own fieldwork in societies which are apparently strongly male-dominated, I have often met men who sincerely complain that their wives, who control the domestic domain, decide everything in their lives. In an amusing and intriguing study of a Portuguese local community in the Saloio area outside Lisbon, Joyce Riegelhaupt (1967) analyses the relationship between male and female power in a society where male power is officially all-pervasive. At the time of her fieldwork, the subordination of Portuguese women was established by law, in the domestic as well as in the public sphere. Nearly all political offices were held by men. However, Riegelhaupt discovered that women in this community were in practice sometimes more powerful than men. The explanation for the strong position of women in this community seems to lie in the division of labour, which paradoxically allows women a more public role than their husbands. The men are responsible for agricultural work, while female members of household divide their time between housework, child-raising, some agricultural work, marketing and shopping. Since the men work in the fields, only the women stay in the village during the day. They then meet in the shops, where they may exchange information and develop networks. They also travel to Lisbon to sell commodities, and Gender and Age 131 are in this way able to develop networks outside the village as well. The men, on their part, have much less contact with each other, since they work in isolation on their separate plots. As the women market the family's produce, they are central to the domestic finances. They are also important in politics, despite their formally marginal position. Most political issues in the locality concern the maintenance of and improvements to public infrastructure; mending roads, installing water pipelines and so on. It is necessary to have personal contacts with the authorities to achieve this. Thanks to their wide-ranging networks, the women frequently succeed in persuading the wives of local political leaders to talk their husbands into taking the right decisions. There is in other words a great discrepancy between rules and practices in this case. The law and official ideology state that men ought to be in charge of politics and the domestic economy. In practice, the women seem to exert more power than their spouses in both respects. The Salpio example indicates that the subordination of women cannot be assumed a priori, and it also serves as a reminder that even perfect knowledge of the explicit norms of a society does not enable us to predict how people will act. MEN:WOMEN::CULTURE:NATURE? This subheading is shaped like a structuralist formula. It reads like this: 'Does the relationship between men and women correspond to the relationship between culture and nature?' A simple colon refers to a relationship, while a double colon refers to a relationship between sets of relationships (see also Chapter 15). In many societies, women are seen as being closer to nature than men, who are considered more cultivated than women. Like nature, Ortner notes in an article inspired by Lévi-Strauss's structuralism (Ortner 19 74), women are regarded as undomesticated, wild and difficult to control. Therefore they represent danger (to men) and must be domesticated. Ortner mentions three related, universal aspects of female existence which may lend support to this widespread cultural view. (1) The body of the woman and its biological functions (birth, menstruation, suckling) make it necessary for her to spend more time on 'species behaviour' than the man, who is freer to concentrate on 'purely cultural projects'. (2) The body of the woman and its functions place her in social roles which are regarded as 'lower' than those of men (cooking, cleaning, etc.). (3) These two aspects of female existence give her a mental structure which is different from that of men and which is seen as being closer to nature. This point, we should note, is reminiscent of the public/private distinction. It also suggests that women are seen as passive while men are seen as active, which is a very widespread view, whether as regards the sexual act, the transmission of culture or the respective roles of the genders in the upholding of society. 132 Small Places, Large Issues According to this line of thought, biological differences between men and women form the starting-point for gender inequality; for example, her association with small children (who are also considered imperfectly cultured) draws her towards nature. In patrilineal and virilocal societies, it may be added, women also come from 'outside'; they belong to a different kin group from the one in which they live. All of this indicates that - and may contribute to explaining why - women are often regarded as more natural, less cultural, than men. On the other hand, women everywhere obviously have cultural roles; in many societies it is actually the role of women to maintain and transmit tradition, and it is often considered 'natural' that women are more strongly religious than men. In many Mediterranean areas, further, men are considered sexual 'forces of nature' and, for this reason, women are themselves seen as to blame if they are raped - they ought to be cultured enough to protect themselves from the male, who is allegedly unable to contain his sexual drive. Nevertheless, there is a wealth of ethnographic support for Ortner's general assumptions, namely the fact that women give birth, suckle babies and menstruate gives them an ambiguous and sometimes 'dangerous' cultural position between nature and culture. The notion that 'man is to woman what culture is to nature' is controversial, and most scholars agree that the model is simplistic. It has also been pointed out that the cultural distinction between nature and culture may not be as universal as Ortner, drawing on Lévi-Strauss, assumes. It may still be good to think through, however, as an aid to interpreting complex ethnographic material. The general notion that one's enemies or subordinates are somehow closer to nature than oneself is far from uncommon, and it is certainly not restricted to the male-female relationship. According to the racist ideology which served to justify slavery in North America, blacks were closer to nature than whites and were therefore better suited for hard physical work; but simultaneously they were 'like children', unable to handle responsibility and freedom. In this way, the framing of social inequality in a culture-nature dichotomy can be an important ideological instrument legitimating power differences. 'WOMEN'S WORLDS' AND 'MEN'S WORLDS'? If we accept that there are systematic differences between women and men - whether we attribute the differences to biology, socialisation or ideology -we must also admit that men and women may experience the world in different ways (related to what Ortner calls different 'mental structures'). This was the point of departure for Edwin Ardener when he wrote that the 'problem of women has not been solved by social anthropologists' (19 7 7, p. 1). The problem of women is not identical with the problem of the 'position of women', he continues, but rather concerns the methodological and Gender and Age 133 theoretical problems raised by women in relation to anthropological research. Notably, Ardener claims that women in many societies are shy and quiet, and difficult to engage in conversation on topics interesting to the anthropologist. He speaks of them as 'muted' - not literally, but in the sense that women tend to communicate in ways not immediately intelligible to anthropologists (be they male or female), while men more easily talk about their society in ways familiar to anthropological reasoning: the male cultural universe, Ardener intimates, is closer to the anthropological one than the female cultural universe. In his view, it is anthropology as such which has a male bias, not individual anthropologists. This argument, which has been influential in later research on gender, is reminiscent of Ortner's and Whitehead's distinction between the particularistic woman and the univer-salistic man. If Ardener is correct, women's worlds are generally more difficult to explore than men's worlds. Whatever the case may be, it is clear that women and men may experience the world differently in many societies. In research in the Caribbean, it has indeed been argued that women and men have opposing moralities: women try to enhance their respectability while men strive to improve their reputation (P. Wilson 1978), and the two value systems entail two quite different ways of perceiving the world. However, although the two moralities are associated with gender, they are not irretrievably linked with gender. There are 'bacchanal women' - 'femmes fatales' - who look for ways to improve their reputation, and there are also stable and predictable men who rarely touch alcohol and take their children to church every Sunday (Eriksen 1990). This kind of society, marked by strong tensions between the genders, seems paradoxical, since it appears to be based on two, partly irreconcilable value systems. The truth is probably that similar paradoxes are common, if less visible, in other societies as well, and that it may be fruitful to explore these moral contradictions by focusing on gender. This brings us far from the original preoccupations of what is 'really' male and female (the search for essences), and shows that the study of gender is an important dimension in the study of society as such. SEXUALITIES The examples above suggest that if sex is fixed, gender is fluid; and this is indeed the orthodox view in social and cultural anthropology. Since sex is only culturally available as gender (that is, cultural constructions of sex and their accompanying social practices), this seems to indicate that maleness and femaleness may vary indefinitely. In a lively discussion of recent developments in the anthropology of gender, Henrietta Moore remarks that a problem remains in spite of attempts at relativising gender, namely 'the inconvenient fact that people have bodies that are present in a differentiated binary form' (Moore 1999, p. 154; see also Moore 1994). She then goes on 134 Small Places, Large Issues to discuss body mutilation (tattooing, etc.) in contemporary Western societies, transsexualism and homosexuality as practices that seem to pose a serious challenge to this 'binary form' or male/female contrast. It has time and again been shown that gay men are, in many societies, classified as an 'intermediate' gender - neither wholly male nor wholly female, and in many societies (including Mauritius) there is a widespread notion that lesbians are somehow biologically different from heterosexual women. In the emergent field of inquiry known as 'queer theory', this kind of reasoning is developed to great sophistication, in that the relationship between genitals, gender identity, sexual identity and sexual practices is problematised. This is a field where careful attention to the facts is crucial, even though direct observation may for obvious reasons be difficult. During a campaign against AIDS in Norway, the anthropologist Bjorge Andersen thus coined the term 'Men who have sex with men' in order to reach a group of men who did not define themselves as gay. (The term later gave the name to an ill-fated Oslo rock band.) As Moore drily comments, the 'available anthropological data actually suggests that most people do not find their gender identities particularly fluid or open to choice' (1999, p. 158), but she also demonstrates the instability of the sex/gender boundary, paying particular attention to recent developments in the Euro-American part of the world. The uncertainty and reflexivity characteristic of gender relations and gender identities in this kind of society could further be said to form part of a more encompassing phenomenon, namely that of reflexive identity in general (Giddens 1991). Just as it has become difficult to state squarely what it means to be a good man or a good woman, other identities - be they national, ethnic, religious or professional - are also subject to scrutiny and negotiation. Plastic surgery, which in the space of a few years has become very widespread in the affluent Western societies, can frequently be seen as an attempt to change an identity which is usually perceived as no less imperative and absolute than gender, namely age. AGE Like gender, age is a universal principle for social differentiation and classification. Ageing is an inevitable and irreversible biological process but, like gender, age is to some extent socially constructed. In many societies a person's rank rises as he or she becomes older, regardless of gender. Indeed, Holy remarks, with 'the exception of a few hunting and nomadic societies in which survival depends on the physical ability to move around ... the non-industrial societies emerge as distinctly old-age oriented' (1990, p. 167). In modern industrial society, by contrast, old people do not have a particularly high authority by virtue of age: they are no longer achievers and are therefore relatively non-valuable in societies like these, which place a high value on individual achievement. In addition, the rapid pace of cultural Gender and Age 13 5 change in modern societies renders much of the wisdom and cultural competence of old people obsolete. Advanced age is often associated with deep experience, wisdom and a sound sense of judgement. In many societies, old men are the political rulers and old women are perceived as less 'threatening' than younger ones, since they have grown more 'cultivated' and are further removed from nature than younger women are. They no longer menstruate, they no longer have children; they are 'drier' and do not represent a 'threatening' sexual force. Old women may in some societies be more powerful than young men. Societies where the old in the main control the political domain are called gerontocracies. Similarly, children and adolescents are often considered imperfect in the sense that much of their immanent humanness has yet to be realised - while they may also, for the same reason, be considered 'innocent', a condition which is a form of perfection. They have yet to be socialised, and therefore know neither sin nor virtue. In most societies, criteria other than gender and age contribute to distinguishing between categories of people, but there are also societies which only use those criteria, in addition to personal merit. It therefore seems clear that gender and age are more fundamental criteria for social differentiation than, for example, caste, class or ethnicity. AGE GRADES AND AGE GROUPS In some societies where age is an important principle of differentiation, there are several degrees, or institutionalised stages, between youth and old age. The Baktaman of New Guinea thus distinguish between seven age grades among men, and the ritual passing from one grade to the next entails a promotion in relative rank (Barth 1975). Men of the seventh grade possess virtually the entire body of knowledge extant among the Baktaman, including various forms of secret knowledge (which is transferred through consecutive rites of passage), and it therefore seems 'natural' that they should have control over the political domain. In this way, age may function as a vertical principle of differentiation. It may also function horizontally, by dividing the population (usually males only) into peer groups who belong to the same age category. In societies lacking criteria for internal differentiation other than age, gender and kinship, such as the pastoral Nuer and Maasai societies of East Africa, men (and sometimes women) are organised into age sets of people who are not defined as kin, but who were circumcised at the same time. A special kind of solidarity, reminiscent of kinship, exists within these groups, and often age sets have special collective obligations. The formation of such sets contributes to strengthening social cohesion and integration in society, since they cross kin boundaries (see also Chapter 11). In some modern state societies, 136 Small Places, Large Issues Life-stages as an Analytical Category? All societies distinguish between different life-stages. Van Gennep's, Turner's and others' studies of rites of passage indicate how the process of socialisation inevitably creates ruptures in life, whereby persons pass from one stage to another, entailing new statuses, rights and duties. But are such life stages universal? This view has been argued by Colin Turnbull (1985), who presents five life-stages he considers as universal. If he is correct, life-stages may be used as comparative concepts; that is to say, we may expect to find the same stages in all societies. Turnbull's life-stages are as follows. Childhood is marked by dependence on others and fast acquisition of cultural categories. Adolescence is described as that period between childhood and maturity when one develops sexual maturity and is gradually preparing for full social responsibility. Youth is seen as the stage between adolescence and adulthood, which is largely described as a period of higher education. Indeed, Turnbull himself admits that this is not a universal life stage. Adulthood, the fourth phase (or, in many societies, the third one), seems 'more boring', Turnbull claims, than the previous ones; it is full of responsibility, work and routine. The final life stage is old age, a period in which, Turnbull writes, physical and mental defects may set in, but when the 'heart and soul' are more vital than ever before, since old persons have such a great deal of experience. Does ageing and the passage between life stages, then, entail more or less the same thing in different societies, as Turnbull intimates? Probably not. For example, there is little doubt that the European conceptualisation of childhood was developed in the eighteenth century. schoolmates may develop comparable ties of solidarity, not least among the English upper and upper-middle classes, where, for example, old Etonians of the same cohort are expected to support each other. AGEING INTO A GENDERED PERSON Gender and age have both biological and cultural aspects. Age is not necessarily directly correlated with gender. Among the Bakweri of Cameroon, for example, it is nearly impossible for a man to get married before he reaches the age of 40, since he must have property and political influence to find a wife. Women, on the other hand, are married shortly after sexual maturity. Children are often considered relatively sexless, and their socialisation frequently aims at achieving a double end: to turn them into members of Gender and Age 137 society, and to turn them into men and women. For this reason, rites of passage for girls and boys usually differ markedly. Among the people who live near Mount Hagen in highland New Guinea (Strathern 1988), it is a common view that infants are born with both male and female properties. For them to become women and men, they have to go through a long process of learning which culminates in a series of puberty rites. Among the boys, in particular, this ritual is highly demanding. According to the Hagen people themselves, this is due to the fact that the boys have to be worked more than the girls to become properly gendered persons, that is functioning social agents. The girls need to be transformed less through cultivating rites, since they are considered to carry the necessary growth potential in their bodies already. As mentioned earlier, men are often seen as active, women as passive. After such rites of initiation, children are clearly differentiated by gender in most societies. Such rites may take place around the time of sexual maturation or earlier. In some Middle Eastern societies, both boys and girls may go to the public bath with their mothers; after initiation, however, the boys are not even allowed to see naked women. Rites of initiation often entail circumcision of the genitals, tattooing or other visible, physical alterations, making it possible at a glance to distinguish insiders from outsiders and 'complete' humans from 'incomplete' ones. In many societies, the candidates are also subjected to great trials during the period of initiation, frequently physical pain, to give them the opportunity to prove that they are worthy of the responsibility and the rights given them as grown members of society. Among the Nuer, the initiation of boys includes the cutting of six parallel stripes across the brow from ear to ear. The cuts go 'to the bone' (Evans-Pritchard 1940), and the boys are expected not to show evidence of pain during the ordeal. RITES OF PASSAGE The term 'rite of passage' (rite de passage) is associated with the name of Arnold vanGennep, who published the bookLesRites depassagein 1909 (vanGennep 1909,1960). Through such rites, van Gennep wrote, society reproduces itself. People are given new statuses without the social structure changing, and the public character of the ceremonies gives the inhabitants an annual reminder of the fellowship, rights and obligations provided by society. Since then, Victor Turner (1967,1969,19 74) has developed van Gennep's perspective further by looking closely into the phases and levels of meaning provided by these rites among the Ndembu of what is now Zambia. A general point in Turner's studies is that the rites of passage simultaneously function as permitting integration into society and give the participants a mystical experience of oneness with the spiritual world and with the 'societal organism'. Turner follows van Gennep in dividing the rites into three phases: separation, liminality and reintegration. Separation is characterised by the 138 Small Places, Large Issues individual's or group's movement away from a fixed point in social structure towards something unknown. When the breach is completed, the agent enters a liminal phase, an ambiguous stage where he or she is in a certain sense placed outside society, 'betwixt and between' (Turner 1969) two stable conditions. This puts the actor in a dangerous position. Society runs the risk that the actor refuses to be reintegrated and rejects its values and power hierarchies, while the individual for his or her part risks anomie and social homelessness. Turner writes that in nearly every society, a person in a liminal phase is 'structurally if not physically invisible in terms of his culture's standard definitions and classifications' (1974, p. 232). This difficult and dangerous liminality is nevertheless necessary in order to 'clean off the earlier statuses of the individual, to make him or her ready to be reborn as a new category of social person. In a study of the Kaguru, a matrilineal people in Tanzania, T.O. Beidelman (1971) writes that if a boy dies immediately after circumcision he cannot be buried in the ordinary way, since he will have died neither as a child nor as an adult, but as a liminal, indeterminate kind of person, or even non-person. The final phase in rites of passage is reintegration. During this phase, the candidates return as new persons, usually at a higher level of rank. Victor Turner (1920-83) was trained in classical structural-func-tionalism, but his interests gradually developed in the direction of symbolic analysis and psychological anthropology. His first major monograph, Schism and Continuity in an African Society (1957), deals with social integration and fission among the Ndembu in Northern Rhodesia (now Zambia). Several of his later books deal with the Ndembu as well, but his most influential texts focus on ritual and the meaning of symbols. Unlike Lévi-Strauss, Turner emphasises the multivocality or ambiguity of symbols, showing how symbols simultaneously contribute to the maintenance of society and respond to existential problems. Regarding his studies of ritual, Turner is particularly known for his treatment of rites of passage. Here, he has called attention to the liminal phase, which he regards as crucial (for example in The Ritual Process, 1969). Turner has described the transition between different states entailed by liminality as a dynamic interplay between different forms of social integration, namely the formal societas and the informal communitas. To Turner, rituals are condensed expressions of a social form: through studying them in great detail, he has argued, one may make statements about society and individuals in general. Other anthropologists have later tried to apply Turner's insights and concepts about liminality, societas and communitas to contexts other than the purely ritual ones. Gender and Age 139 Beidelman provides detailed descriptions of initiation rites among the Kaguru, and it is easy to identify the three phases analysed by Turner in these. The rites entail both moral education and physical alterations for both genders, but their content differs. The boys, who are ten to twelve years old, are led away from the village into the bush (separation), undressed and shaved. Their foreskins are then removed, and the adult men who have accompanied them then begin to teach them riddles, songs and myths which encapsulate the essence of the Kaguru world-view. After circumcision, the boys have to remain in the temporary camp in the bush for a certain period (liminality), and during this period the adult men arrange various trials for them: they hide in the bushes at night pretending to be lions, tell them that they may die after circumcision, and so on. This prolonged phase is, of course, the liminal one. Finally, the boys are taken back to the village and are given several new names (reintegration). Unlike the boys, Kaguru girls are not regarded as fundamentally transformed after initiation. Whereas the boys are taken out into the bush together, girls are initiated separately and in isolation in huts in the village, but they too are circumcised and given moral education, although to some extent they learn other things than the boys do. The liminal phase gives the boys in Kaguru society (and in many other societies) powerful common experiences; they have gone through trials together and have become adults together. Frequently, such shared experiences create life-lasting ties of solidarity; as previously mentioned, such ties are sometimes institutionalised and serve as a mitigating political factor in societies which are otherwise based on lineage organisation. MARRIAGE AND DEATH Most studies of rites of passage deal with initiation rites. Rites relating to marriage and death are nevertheless also important. They move persons from one status to another and serve as collective reminders - just like initiation rites - of the cohesion of society, its moral values and the legitimacy of authority. In kinship-based societies, marriage gives important opportunities for forging alliances between kin groups and symbolises the continuity of society. The mortuary rite marks the last important rite of passage in the life of any earthling. Among the Kaguru, mortuary rites are associated with two problems. First of all, one must ensure that the deceased is safely transferred from the land of the living to the spiritual world. Recently deceased persons, like adolescents in the bush, are 'betwixt and between' and must therefore be approached with great caution. They cannot be controlled as one controls living people, yet at the same time they are sufficiently close to the living to influence them. They must, in other words, be established as properly dead people as quickly as possible. 140 Small Places, Large Issues A person who has died is shaved all over (like a novice during initiation); men are laid on their right side, women on their left. Both genders are buried with their heads facing left, towards the spiritual world. Dead bodies are buried in great haste since they are considered polluting and dangerous, but the mourning period lasts for at least four days. At the end of this period the second problem associated with death turns up, namely the question of inheritance. In this situation, the issue does not merely concern how to make a good deal for oneself, but also which social ties are to be strengthened and which are to be weakened. There are social obligations, property and social statuses to be redistributed at a death. The challenges faced at a death are perhaps universal. First, death must be given a symbolic, perhaps religious, content to make it possible for the survivors to reconcile themselves with it; and second, one must ensure that life goes on, that is, that society continues to exist more or less unaltered after a death. RITES OF PASSAGE IN MODERN SOCIETIES? Some Western readers may have the impression that rites of passage exist 'out there' and not 'among ourselves'. This is not the case, but it is doubtless true that such rites have a diminished importance in modern societies. In West European societies, four major rites of passage have traditionally been important, although three of them seem to have lost some of their significance since the Second World War. While the funeral is still a social event of great importance, baptism, confirmation (or first communion) and marriage have become both less widespread and, in most European societies, less important. In part, this is due to the fact that these rituals are associated with a religion whose role in the daily life of Europeans has diminished; another partial explanation is that these rituals are no longer socially important for individuals in marking the transition from one status to another. In the Protestant societies of Scandinavia, candidates for confirmation were traditionally obliged to learn by rote quite an amount of biblical knowledge, and failing was a real possibility. At the confirmation party afterwards, young people received 'adult' presents such as a suit and a watch - in parts of Scandinavia, cigarette cases and sets of false teeth (often made of whale ivory) were also common gifts as late as the 1930s - which signified that they were now to be considered grown-ups with full responsibilities. As confirmation takes place when children are about 14, it is easy to understand why its importance has decreased: while they were formerly expected to earn their own living after confirmation, there is no longer a dramatic change in their lives following the rite. Perhaps a similar explanation holds for the reduced importance of the marriage rite in societies where serial monogamy is becoming the norm rather than the exception? Gender and Age j41 An important difference between gender and age as principles for social differentiation is the fact that one more or less automatically changes membership between age groups, while few people change their gender. In gerontocratic but otherwise egalitarian societies, boys may simply await their coming of age to achieve full political rights; they may be eligible to sit on the elders' council and may eventually become powerful ancestral spirits. This is not an option available to women in many societies. As this chapter has stressed, the division of labour in society is fundamental to social differentiation, and the relative complexity of the division of labour may indicate the complexity of social differentiation. It has also been shown that social differentiation cannot be studied independently from politics and ideology. Both gender and age tend to be associated with politics and the division of labour, but they usually function together with other factors. In the next chapter, we consider some such criteria which contribute to the kind of complexity mentioned, namely caste and class. SUGGESTIONS FOR FURTHER READING David Gilmore: Manhood in the Making. New Haven, CT: Yale University Press 1989. Henrietta L. Moore: A Passion for Difference. Cambridge: Polity. Victor W. Turner: The Ritual Process. Chicago: Aldine 1969.