Browse the book with Google Preview »
The changing of gender roles—the adoption, for the most varied reasons, of the culturally defined social role of the opposite biological sex—has been reported worldwide for numerous cultures and for all historic periods. In Greek mythology, Thetis hides her son Achilles, whose death before Troy has been prophesied, by disguising him as a girl at the court of King Lycomedes on the island of Scyrus. But in order to win him for the military campaign, Odysseus resorts to a ruse which, in modified form, was also used by North American Indians in order to determine whether a boy really was a "berdache": he has a spear and a shield brought into the women's room, and orders the war trumpet blown. Achilles reaches for the weapons and thus gives himself away as a man (Ranke-Graves 1960, 2:271; Schwab 1986:354).
The Amazons of antiquity were described as an entire nation of women who pursued male occupations (Hammes 1981; Herodotus 1971:292ff.). Change of gender roles for a lifetime was found among some men of the Scythians, and according to Herodotus, the young women of the neighboring Sarmatians led the life of warriors until they had killed an enemy; only then were they permitted to marry. But even as married women, they went on the hunt dressed as men, whether alone or in the company of their husbands, and they accompanied the men in war (see Herodotus 1971:294; Hammond 188z:339ff., 1891:107ff.). Gender role change also occurred in ancient Rome (Green 1974:4ff.), and the female pope Joan has gone down in the history of the European Middle Ages as a spectacular, if isolated, case (Fiocchetto 1988:212; Green 1974:6).
In later centuries, reports are encountered over and over again concerning persons in Europe (see Dekker and Van de Pol 1990; Green 1974:7ff.) and in the United States (Katz 1985) who lived successfully in the social role of the opposite sex, and who frequently went unrecognized for a long time.
With the broadening of horizons during the Age of Discovery, it became apparent that gender role change was also to be found in numerous cultures outside of Europe, often even in institutionalized form (Bullough 1976; Greenberg 1988; Karsch-Haack 1911; Williams 1986b:252ff.) The hijras of India (Nanda 1986), the Samoan faafafine (Munroe and Munroe 1987:60), the Tahitian mahus (Williams 1986b:255ff.), and the Siberian "soft men" (Bogoraz 1907:449ff.; Williams 1986b:252ff.) are additional examples of this.
In the ethnographic literature, however, the best-known example is that of the North American "berdaches" who constitute the subject of this book. A listing of sources within the framework of the GAI (Gay American Indians) history project reveals "berdaches" and other alternative gender roles in I33 Indian groups (Gay American Indians and Roscoe 1988:217ff.; Roscoe 1987).
In our own culture, gender role change appears in the form of transsexualism. Transsexuals aspire not only to the role but also to the gender status of the opposite sex. Usually they desire a new classification of their persons not as ambivalent (transsexual) but rather as unambiguous and definite. In these cases hormonal, surgical, and legal measures such as a change of personal status and of name are intended to facilitate the adjustment of the physically actual to the subjectively felt sex (see Benjamin 1966; Burchard 1961, Green 1974; Green and Money 1969; Hamburger 1954a, 1954b; Kessler and McKenna 1977; Pauly 1974; Schicketanz et al. 1989).
Not least because of their appearance in numerous groups and because of the comparatively rich and comprehensive source material do the North American "berdaches" seem particularly appropriate for the investigation of gender role change in extra-European cultures. Regionally, the present work embraces the entire North American subcontinent from the Mexican border southwest of the United States to Alaska in the northernmost part of the hemisphere, and along with the Indian groups, also includes the Inuit (see Map 1).
The North American "Berdaches"
The designation "berdache" originally comes from the Arabic-speaking region, where bardaj or barah meant "kept boy," "male prostitute," "catamite" (Angelino and Shedd 1955:121). "Berdache" is the French adaptation of these terms and was first used by eighteenth-century French travelers, who mainly applied the word to supposedly "passive homosexual" Native American males who were transvestites and who fulfilled the culturally defined role of a woman (see Angelino and Shedd 1955:121f.).
In the course of time, very different phenomena came to be referred to by the designation "berdache": transvestism (cross-dressing), effeminacy, the carrying out of female tasks and activities, entering into homosexual relationships, and intersexuality (Angelino and Shedd 1955: 122ff.). Around the beginning of this century, the term was extended to females who had taken on the social role of a man. Emphasis on the choice of sexual partners was particularly long-lasting, so that the designations "berdache" and "homosexual" frequently appear as synonymous, but the terms "berdache" and "transvestite" often occur as synonymous as well (see Angelino and Shedd 1955: 122ff.).
However, so many variant expressions of gender role change exist that such equations turn out to make little sense: males in the social role of a woman enter into marriages with women; males in a masculine role have sexual relations with non-"berdache" men; and transvestism (cross-dressing) proves to be as readily dispensable a component of any gender role change as homosexual behavior (Angelino and Shedd 1955:122ff.). With this in mind, Angelino and Shedd suggested a definition of "berdache" which is based on such persons' social (gender) role:
In view of the data we propose that berdache be characterized as an individual of a definite physiological sex (male or female) who assumes the role and status of the opposite sex, and who is viewed by the community as being of one sex physiologically but as having assumed the role and status of the opposite sex. (1955:125)
Thirty years later, in light of more recent investigations and the results of his own fieldwork, Williams proposed another definition: "Briefly, a berdache can be defined as a morphological male who does not fill a society's standard male role, who has a nonmasculine character. This type of person is often stereotyped as effeminate, but a more accurate characterization is androgyny" (1986b:2). At the basis of this definition lies the observation, also attested to both by numerous primary sources and by Native American terms for women-men and men-women, of "gender mixing." Gender mixing represents a combination of masculine and feminine gender statuses instead of a genuine change from "one" gender into "the" other. Consequently, gender mixing is often expressed by a combination of the feminine gender role with the masculine gender role (see Callender and Kochems 1986). As will be shown in what follows, this combination is not uniform, and comprises a spectrum from quasi-masculine to quasi-feminine role makeup and organization (Ausgestaltung).
The entire adoption of the feminine gender role by males, which lies at the basis of Angelino and Shedd's (1955) definition, is found here as well, but it no more leads to the ascription of a feminine gender status (ie., the gender status of "woman") than does a "mixed" gender role.
In this connection, the distinction between gender category and gender status is significant: the gender category is based on the biological distinction between males and females, whereas the gender status, as culturally defined, can be either masculine (man) or feminine (woman), but can also entail and hold open other possibilities:
At the gender-category level, classification as not-male necessarily means being classified as female, and vice versa. At the level of gender status, however, definition as not-man is not equivalent to identification as woman... because status depends on the cultural construction of gender. While women are, by definition, not-men, other social groups within a society may consist of males whose gender status is that of not-men but who are also defined as not-women. Gender-mixing statuses... are an example. (Callender and Kochems 1986:166)
This view, shared by Whitehead (1981), Kessler and McKenna (1977), and Martin and Voorhies (1975), assumes at the outset that, beyond a purely masculine or feminine socially defined gender membership, other, "supernumerary genders" (Martin and Voorhies) can also exist which can be characterized as gender mixing.
Within the framework of this view, investigations of nonmasculine or nonfeminine genders in other cultures no longer need to be limited by the polarization of sex/gender membership so characteristic of Western culture—a polarization that in fact shapes gender role change in Western culture. Because the category "gender mixing" does not exist in the West, transsexuals strive for complete identification with the opposite sex, even to the extent of a new, surgically constructed physical definition of their sex (see Benjamin 1966; Green 1974; Green and Money 1969; Kessler and McKenna 1977:26).
In Western culture, the social pressures to conform to a nonambivalent gender status are great. Gender status is not defined as in Callender's and Kochems' cross-cultural notion: if you are a man but do not so identify yourself, you must try to be accepted as a woman; likewise for a woman who does not identify herself as such. Within this pattern of thinking, there is no place for the ambivalence of androgyny.
Prevailing uses of the term "berdache" reflect this inflexibility as well as the distortions that can result from imposing gender polarity on non-Western cultures. Williams pointed out, for example, that the application of the term "berdache" to females who effect a change in gender roles is inappropriate because the word originally derives from a term for male prostitutes, as well as because females in the role of hunters and warriors would also be conceptualized in Native American cultures as being different from male "berdaches." He therefore suggested the term "Amazons" (1986b:11). This does not appear to be the optimal solution, however, because historically, Amazons are distinctly associated with the warlike element of the masculine role: the ancient sources emphasize above all the Amazons' warlike fighting spirit (see Hammes 1981). In Native North America, however, this warlike component appears in connection with women who go to war without either giving up the feminine role or transforming their feminine gender status into an ambivalent one (see Chapter 17).
What terminology, then, is appropriate? As noted in the Preface, Native Americans and anthropologists alike have rejected the term "berdache" as inappropriate for persons of either sex; in the following, it will be retained only in references to earlier sources. For reasons also outlined in the Preface, those individuals who were referred to as male "berdaches" in the earlier anthropological writings will usually be termed "women-men" in this book; females in a masculine role who occupy a nonfeminine gender status in their respective tribes will be termed "men-women." And, as noted, I will also use the Native American terms for these individuals.
The designation "gender role change" will serve as a higher level generic term for various kinds of crossing over culturally defined gender roles, usually including the entrance into an ambivalent gender status. At the same time, I assume along with Callender and Kochems (1986:166) that a change of gender roles does not imply a change of gender status from "man" to "woman" or vice versa. (For the definition of "status" and of "role" used in this study, see Chapter 4.) Following Callender and Kochems (1986:168), I will define gender role change using four features which in turn form the basis for defining an alternative gender status ("gender mixing"). However, unlike Callender and Kochems, I will refer to persons of both sexes. Furthermore, I will call ambivalent gender status what they designate as "gender-mixing status" (see Chapter 4).
Gender role change may include the following characteristic features:
- partial or complete transvestism (cross-dressing);
- the expression of culturally defined characteristic behavior patterns of the opposite sex;
- the carrying out of activities that are culturally assigned to the opposite sex;
- no sexual relations with persons who occupy the same gender status as the person in question.
Transvestism may not appear at all, and would seem to be the most dispensable component of gender role change and ambivalent gender status (see Callender and Kochems 1986:168). Further, modes of behavior and occupations that belong to the standard gender role of one's own sex can be retained, along with the practice of behavioral modes and activities of the opposite sex. The most important characteristic of gender role change is the preference for the occupations and activities of the opposite sex (Callender and Kochems 1986:176; Whitehead 1981: 85f.). The following definition of women-men and men-women (as opposed to those who cross gender role boundaries without having a special assigned status in their own culture) combines the elements of gender role change and gender status change: A woman-man or a man-woman is a person of usually physically unambiguous sex who voluntarily and permanently takes on the culturally defined activities and occupations of the opposite sex, and who has a special (ambivalent) gender status assigned to him or her by his or her culture. The following are distinguished in Native American cultures from women-men and men-women ("berdaches"):
- persons who have sexual relations with members of their own sex without carrying out a change of gender roles;
- persons who wear the clothing of the opposite sex (transvestites) without carrying out a change of gender role (e.g., in certain ritual situations);
- feminine men and masculine women who retain their gender status;
- "warrior women" and other persons who cross gender role boundaries without, by so doing, exchanging their gender status for an ambivalent one; and
- men who because of a failure in the warrior role are forced as a humiliation to wear women's clothing, and occasionally also to take on the womanly role (although these men actually want the masculine role and occasionally rehabilitate themselves by undertaking some daredevil act of war on their own; see Bossu's example from the Illinois, 1962:82).
Hermaphrodites and intersexuals were usually integrated into the woman-man/man-woman statuses because their physical ambivalence was regarded as comparable to the ambivalence (due to the discrepancy between physical sex and lived-out gender role) manifested by nonintersexual women-men and men-women (see Chapter 8 ). The above list of exceptions should not be taken as absolute. The distinction between gender role change and gender role crossing is not qualitative, but rather quantitative. When sources give details concerning persons to whom an ambivalent gender status has been assigned by the members of an ethnic group, such "emic" classifications and their culturally specific backgrounds must be considered. The definition suggested here has the advantage of taking into account the wide range of statuses formerly referred to as "berdache." It seems reasonable to include examples of gender role crossing in the present study. Gender role crossing refers to people who take up some elements of the culturally defined role of the other sex but at the same time largely retain the standard gender role culturally assigned to their own sex. Unlike the majority of women-men or men-women, they do not take up the gender role of the other sex more or less completely. In many places, gender roles are more flexible than has long been assumed (see Blackwood 1984). Yet there is scarcely any information available concerning the degree or point in gender role crossing beyond which a person was classified as a woman-man or man-woman (see Parsons 1939h: 38 and 38 n. on the delimitation of gender role crossing and woman-man/man-woman statuses among the Pueblos). A comparison of women-men and men-women with persons who cross over into the role of the opposite sex without being so classified can help to illuminate the statuses of women-men and men-women within a broader cultural context.
Focus of This Book
Although "berdaches" (primarily women-men) have been the subject of in-depth investigations by Callender and Kochems (1983), Trexler (1995), and Williams (1986b), a thorough, detailed, and systematic reappraisal of the available published sources of information is still lacking. The present study is intended to fill this gap and constitutes an expansion of the above-named investigations. At the same time, however, it is also intended as a close reexamination of statements and of theoretical reflections encountered in these and other works on the subject.
The primary emphasis of the present study is on the traditional status (i.e., the status as held in the tribal societies and in the early reservation period) of women-men and men-women in Native American cultures in North America. For a discussion of more recent and to some extent homosexual/gay self-definitions by members of widely differing Native American groups, the reader should consult Williams (1986b). In his detailed investigation of the "berdache" phenomenon, Williams dealt mainly with those groups in which women-men enjoyed high standing as recipients of especially potent latent spiritual power. He also analyzed the changes that the "berdache" roles and statuses underwent within processes of acculturation. However, he heavily emphasized the sexual aspect of the institution as a culturally approved opportunity for same sex relationships. By contrast, the present study takes into consideration groups from which a "berdache" status has (to my knowledge) ever been reported, irrespective of the prevailing attitude toward women-men and men-women within the respective groups, and whether or not a spiritual component existed (although when present, the latter has naturally been taken into account).
The study which in its approach most closely resembles my own is that of Callender and Kochems (1983), who also adduced the entire corpus of available information in order to arrive at definitive statements concerning the "berdache" status. However, because that study is in essay form, many questions of detail had to remain unanswered. Among other things, the present study is intended to address some of these questions, so that, against the background of a broader cultural context, more differentiated statements can be made. In the other most recent work on this topic (Kenny 1988; Midnight Sun 1988; Trexler 1995; Whitehead 1981), the body of data has been applied selectively, partly in order to support preconceived theoretical standpoints. Because of this, the authors mentioned have not done justice to the complexity of the topic. For example, Whitehead's argument (see Chapter 3) is acceptable only if one accepts her equation:
male: female :: culture: nature
and shares her view that a universal gender hierarchy necessarily results from it.
Finally, investigations of gender role change among females and of female homosexuality have been scarce up to now (but see Allen 1981, 1986; Blackwood 1984; Whitehead 1981; Williams 1986b:233ff.). The present volume, by using the entire corpus of data on men-women and women-men, will contribute to this relatively unexplored area of study.
Primarily, this investigation focuses on working out the relationship of gender role, as it is lived out and experienced by the woman-man or man-woman, to the sexual role differentiation customary in each of the respective groups for which data have been reported. The goal is to determine the kind and degree of gender role change and at the same time to pursue the question of the gender status of the women-men/menwomen. In connection with this, my point of departure is recent theoretical approaches to problems of gender status which depart from the concept of two polarized gender statuses. Instead, these approaches demonstrate that a change of gender status is more likely to result in an ambivalent status than in the status of the opposite sex: it will be shown in the case of the women-men/men-women and related phenomena that "gender mixing', takes place, not "gender crossing" (Callender and Kochems 1986; Kessler and McKenna 1977; Martin and Voorhies 1975; Whitehead 1981; Williams 1986b). This viewpoint will also make it possible to recognize the alternative gender status constructions that exist in non-European societies, and, by examining the experiences of persons who live in such mixed roles, to come to grips with these constructs. A two-part preliminary question, therefore, is this: which North American Indian cultures developed an ambivalent gender status? For the groups where it existed, how was/is the content structure of this status actually worked out and enacted by individuals making use of the gender roles? It is also necessary to determine whether persons described in the literature as "berdaches" actually were/are women-men or men-women, according to the definition provided above. If they are not, it is important to determine whether they were classified as "berdaches" in the sources by anthropologists or other writers on the basis of the sex/gender concepts of the ethnographer or of the consultant. In addition, it is crucial to consider how men-women and women-men might have been associated with particular social phenomena. For example, the Navajo connected nadle (women-men and men-women; in more recent literature also spelled nádleehé, see Thomas, 1997) with intersexuality, whereas some Shoshoni-speaking groups associated them with infertility.
Women-men's and men-women's relationships with partners as an expression of gender role and gender status also require examination. Among other things, it is necessary to investigate the extent to which classifications such as "homosexual" or "heterosexual" are applicable to women-men and men-women, or whether such classifications instead represent projections of Western categories upon non-Western cultural phenomena.
The motivation for gender role change constitutes an additional topic of investigation, and it is important to distinguish between what consultants said about this topic and what outside observers reported. All too often, ethnographers and other observers speculated about etiology in terms of unsubstantiated ideas based on prevailing Western psychiatric and psychological models (such as overprotective mothers, weak masculine identity, latent homosexuality, etc.).
Another question that needs to be addressed is the nature of the special spiritual powers sometimes attributed to women-men (and, though to a lesser extent, to men-women) on account of their gender ambivalence. What special tasks were assigned to them? Are these tasks associated with the feminine or masculine role domain? Furthermore, given that these powers have by no means been universally granted to the women-men, in which cultural context(s) were they regarded as being spiritually gifted?
Because the sources have focused almost exclusively on women-men, much of the present discussion focuses on them. Part 3 deals with the available source materials relating to females in a masculine role. Gender role crossing, which has been observed much more frequently among women than among men (e.g., in the form of "warrior women," manly-hearted women, etc.), will also be discussed, even though gender role crossing is, as noted, not the same as gender role change.
The present book, therefore, first of all examines the componential organization of gender role as executed by women-men and men-women: what feminine and masculine role components, respectively, are lived out? Are there differences between transvestite and nontransvestite women-men and men-women? Apart from everyday and specialized activities, what behavior patterns or attributes could possibly signal that the person who adopted these has or will effect a change of gender role and gender status? What is the relationship of women-men and men-women, who are ambivalent because of a discrepancy between biological sex and behavior, to hermaphrodites or intersexuals, who manifest a sexual ambivalence that is purely biological?
This book also addresses the consequences of gender role change. To what specialized occupations or ritual activities were women-men and/or men-women entitled on the basis of their special gender status? On the other hand, which of these activities or occupations were, by contrast, seen as constituent aspects of a feminine or masculine role? These questions are especially instructive for drawing conclusions regarding the gender status of women-men and men-women, for example, in cases in which they carry out occupations otherwise reserved for members of their own biological sex. In other cases, it is necessary to ask whether an activity possibly devolved upon the woman-man or man-woman less because of any inherent special powers than because the activity in question was chosen by the woman-man or man-woman as an additional component of his or her chosen opposite-sex role.
Relationships of marriage or of partnership with persons of the same biological sex can also be an expression of gender role change. These relationships, as well as relationships with partners of the opposite sex are discussed in Chapter 11. The central question is, to what extent were relationships with partners of the same biological sex, but not of the same gender, actually comparable to those relationships classified in Western culture as homosexual, as some authors have asserted (see Chapter 3)? A further issue for investigation here is the degree to which sexual relations between partners of the same biological sex and of the same gender might have corresponded to the definition of homosexuality currently valid in Western culture. In addition, the cultural background of the promiscuity frequently attributed to women-men will also be discussed.
The actual process of carrying out a change of gender role is likewise a topic of the present inquiry. What motives lie at the basis of a person's taking on the role of the opposite biological sex? How was entry into a new gender status legitimized and, if necessary, culturally standardized? Did visions provide occasions in an individual's life for an abrupt turning point from one gender status to another, or did they only legitimate a process of change that had been under way for a long time? How did the social environment react to signs of a gender role change, or to those persons who entered into an ambivalent gender status?
Another important goal of the present study is to determine how it becomes possible for individual persons in the Native American cultures investigated to cross over or even to abandon completely the standard gender role of their sex in a culturally acceptable and often even institutionalized form. One factor is surely the construction of alternative gender statuses in addition to the standard masculine and feminine ones. Western culture does not admit entry into gender ambivalence; it recognizes only a polarization into male and female, masculine and feminine. Ambivalent individuals consequently have to choose one or the other of these poles. Furthermore, in extreme cases (e.g., among transsexuals), this leads to a situation in which individuals actually feel out of place in their own bodies, and consequently attempt by surgical means to adjust their "wrong" bodies to their "right" sexual identity. This constitutes a gender crossing in the sense of Callender and Kochems (1986), as well as a change in gender status and probably even in terms of gender categories. Transsexuals in Western culture do not wish for a "berdache" status for themselves, but rather a nonambivalent gender status: "There're only two alternatives in society. You're either a man or a woman. If I don't feel like a woman then it's got to be the other way" (female-to-male transsexual, in Kessler and McKenna 1977:112). In other cultures, quite obviously, there are more than two alternatives. The present study is intended to show how gender status alternatives can in fact be formulated culturally.
A comparison of research results pertaining to women-men as opposed to men-women also makes it possible to arrive at distinctions about the cultural acceptance or institutionalization of gender role change and gender role crossing. In particular, the biologistic explanatory pronouncement of Whitehead (1981; and see below, Chapter 3) requires close examination: were female "berdaches" essentially rarer than male ones because the female bodily functions made access to male areas of life impossible? This explanatory statement is not acceptable if one does not agree with the structuralist equation, first formulated by Ortner (1974) and later asserted by Whitehead, that female is to male as nature is to culture, together with the assumption deriving from it of a universal dominance of man (Culture) over woman (Nature) (see Ortner 1974: 67ff.). The discussion in the literature regarding the existence of male dominance in Native American cultures has not yet been concluded. However, several authors have justifiably come out against the sweeping assumption of such a dominance (see Albers and Medicine 1983; M. N. Powers 1986; Weist 1980). Above and beyond this, it can be shown that crossing beyond the boundaries of their gender role was easier for women than it was for men, and that in such cases women did not have to take on an alternative gender status permanently (see Chapter 17). The present study likewise seeks explanations for this differential flexibility with regard to gender roles. For example, could the higher incidence of women-men possibly be connected with the fact that there existed more gender role alternatives for women than for men? (see, e.g., Kehoe 1983:66, on Blackfoot women). And how does this variability relate to a possible gender hierarchy?
Before proceeding to investigate male and female gender role change, I first need to discuss several topic-specific problems regarding the source materials and then present a critical account of the contemporary state of research as well as previous theoretical approaches to the topic. After these preliminaries, I will describe the mutual interconnection of gender identity, gender role, and gender status. These relationships are crucial to the approach taken in the present work, which constitutes a study of women-men and men-women against the background of culturally defined gender roles and gender statuses, with special reference to aspects of gender role change among members of North American Indian nations.