Coming to Terms with Navajo "nádleehí": A Critique of "berdache," "Gay," "Alternate

Gender," and "Two-Spirit"

Author(s): Carolyn Epple
Source: American Ethnologist, Vol. 25, No. 2 (May, 1998), pp. 267-290
Published by: Wiley on behalf of the American Anthropological Association
Stable URL: http://www.jstor.org/stable/646695 .
Accessed: 21/09/2013 15:25

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.

Wiley and American Anthropological Association are collaborating with JSTOR to digitize, preserve and
extend access to American Ethnologist.

http://www.jstor.org

This content downloaded from 158.143.192.135 on Sat, 21 Sep 2013 15:25:20 PM

All use subject to JSTOR Terms and Conditions
 coming to terms with Navajo nadleehi: a critique of
berdache, "gay,""alternategender," and "two-spirit"

CAROLYN EPPLE-Dine College

Inthe course of the lastseven years, I have been learningfromand about Navajo persons known
as nadleehi, or, as Hilldescribesthem, "hermaphrodites. . orthose who pretendto be" (1935:275).1
Traditionally,2nadleehi have practiced certain behaviors-such as attire, occupation, and, on
occasion, sexual partnerpreference-in a manner associated with the gender of the opposite sex.
Thus, a male nadleehi may (to varying degrees) wear women's clothing; participatein activities
associated with women, such as cooking and washing; and have sexual relationswith other men.
In general this configurationof "other-gender"behaviors has been treatedas a single phenomenon
across culturesand such individualshave been termed berdache.
Butwith the greaterfocus on gender and sexuality in anthropologyand heightened sensitivityto
Native American voices and categories, the term berdache has been criticized (e.g., Jacobs and
Thomas 1994:7). As a result alternate terms and categories such as gay, alternate gender, and
two-spirithave arisen. Most agree that berdache, with its Frenchmeaning of "male prostitute,"is
problematic;but in creating new categories, many rely on berdache's emphasis on other-gender
behaviors, same-sex sexual practices, and unique roles, thereby creating slight variationson the
same themes. The use of alternategendersuggests that in mixed-genderbehaviorsthere is evidence
for an altogetherdifferentgender (Jacobs 1983; Jacobs and Cromwell 1992); gay emphasizes the
"multi-dimensionalsocial role," and, less overtly, sexual practicesof nadleehi and others (Roscoe
1988); and two-spiritdescribes those with "botha male and a female spirit"in terms of gender and
sexual practices(Tafoya1992:256; Jacobs and Thomas 1994:7).3
As I argue, however, none of these categories is adequate. I follow instead Halperin'sapproach,
developed fromthe historicalstudy of sexual practices. He redirectsattentionaway fromthe cultural
uses of terms (e.g., do the Iroquoishave berdache?)toward their culturalmeanings. He describes
"the real issue"as twofold: "Firstof all [it is] how to recover the terms in which the experiences of
individualsbelonging to past societies were actually constituted,and second, [itis] how to measure
and assess the ones we currentlyemploy" (1990:28-29). To these pointsIrespondwith two questions
of my own: First,what are the premises or epistemological assumptionsof the four categories?And
second, are the premises of the categories consistent with a Navajo worldview (at least with the
version I am beginning to learn) and thus with how many Navajos understandnadleehf? Unlike
other studies,then, my inquiryextends beyond determining the absence or presence of alternate

In this article I assess what premises underlie the categories berdache, "alternate
gender," "gay," and "two-spirit";and whether these premises are relevant to the
ways in which many Navajos construct the "alternategender" of those known as
nadleehi. Proponents of these categories often extricate traits from their contexts
and perceive male and female as mutually opposed, absolute values. Many
Navajos, however, describe traitsas inseparable from the universe and view male
and female as situational values. [Native American, Navajo, gender, sexuality,
worldview]

American Ethnologist 25(2):267-290. Copyright ? 1998, American Anthropological Association.

coming to terms with Navajo nadleehi 267

This content downloaded from 158.143.192.135 on Sat, 21 Sep 2013 15:25:20 PM

All use subject to JSTOR Terms and Conditions
genders, gays, and so forth, or debating the relative merits of one category over another. I rely
on Navajo frameworks to learn "the terms by which the experiences are constituted" and so
"measure and assess" the four categories (Halperin 1990:28-29).
My privileging of certain Navajo frameworks is rathersimple. The synthesis of nadleehi and
others into a single category has often ignored the variability across Native American cultures
and left unexamined the relevance of gender and sexuality. Itseems only logical that frameworks
arising from the cultures in which nadleehi developed (and cultures on which nadleehf act)
would reveal far more (and would do so more soundly) about who and how nadleehi are than
would superimposed categories and concepts.4 Also, by looking at nadleehi in certain Navajo
frameworks,we can see concepts thought to be implicit in the terms berdache, alternategender,
gay, and two-spirit(such as gender and sexual practices) as constructs;consequently, their usage
can be understood in some cases as responses to political and historic events.
Butthis endeavor requiresqualification. Ido not purportto representsome "trueemic"-there
are many Navajo emics, and my discussion of Navajo ways of knowing reflects some of my
own Euro-Americanbiases. What I present is a superficial learning of one approach to Navajo
knowledge among many. My reliance on a particularNavajo worldview is based in part on its
analytical power-it brings clarity to Navajo understandingsthat are otherwise inscrutable. For
example, nadleehf define themselves in terms of both male and female, as interconnected with
their surroundings,and as varying with specific context.5 Currentanalytical concepts simply do
not accommodate the simultaneous distinctness (identity as nadleehf) and fluidity (identity as
context-dependent) of nadleehi's self-descriptions.
Neither is the article an exegesis of a single, alternative nadleehi definition; indeed, I doubt
that one would be possible. Nadleehi express themselves differently with different groups of
people, Navajo definitions of nadleehi vary widely, and the particularNavajo worldview I am
learning seems to place more emphasis on situation-based definitions than on fixed categories.
Instead I point out the flaws in the (putatively)cross-cultural and ahistorical categories, while
calling for a shift back to culturally relative and specific understandings.
Readers should understandthat I use such key terms as gay, berdache, and related categories
as approximations. It will be apparent when I use the term gay with its Euro-American
connotations and when I use it as a gloss on Navajo concepts. The phrase "nadleehi and others
who may be similar,"or some variant, underscores that nadleehi are my point of reference for
understanding other so-called berdaches and that cross-cultural comparability with nadleehi
should not be assumed. I refer to my nadleehf teachers, all of whom are males, with feminine
pronouns, in keeping with the way in which they often referto themselves and other nadleehf.
My comments do not necessarily apply to all Navajo nadleehf. My nadleehi teachers were
between 20 and 40 years old, male, lived on or close to the reservation, retained varying degrees
of Navajo traditional practices, spoke Navajo and English, and had extensive interaction with
Euro-American cultures. My observations may not have relevance for deeply traditional
nadleehf, nadleehi in different age groups, female nadleehi, or various other groups.
In what follows I focus on three areas, beginning with an analysis of the classificatory
principles underlying the four categories. Second, I examine each category, pointing out its
major limitations and strengths, and several of its epistemological premises. Third, I compare
the categories' classificatory premises with various Navajos' constructions of nadleehi, illustrat-
ing the inadequacy of the categories. In consequence I argue for a "deep" cultural contextuali-
zation to understand nadleehi as well as those who are assumed to be similar to them.

classificatory principles
Foundedon positive,ascertainable,and objectivebehavioralphenomena-on the factsof who hadsex
withwhom-the new sexualtaxonomy[homo-versusheterosexuality] could lay claimto a descriptive
validity.Andso it crossedthe "threshold
transhistorical and was enshrinedas a working
of scientificity"
conceptin the social and physicalsciences. [Halperin1990:16]6

268 american ethnologist

This content downloaded from 158.143.192.135 on Sat, 21 Sep 2013 15:25:20 PM

All use subject to JSTOR Terms and Conditions
 Halperin notes how a set of phenomena is identified (in this case, sexual partner's sex),
elevated to universal salience, and assumed to be meaningful across time and cultures. In other
words, one may focus on same-sex sexual practices in, say, ancient Greece, and, having decided
that this behavior exists beyond the constraints of time and culture as something called
"homosexuality," can then pronounce other same-sex unions as instances of "homosexuality."
Using much the same process, many researchers, both Native and non-Native, have classified
nadleehi (as well as their assumed counterparts in other cultures) as berdache, gay, alternate
gender, or two-spirit.7They assume that certain aspects exist outside the influence of culture,
and thus can be applied to all Native American cultures to determine whether such entities as
berdache, alternate gender, and so forth exist.8 Weston's cogent description identifies the
"behavioral phenomena" generally considered transhistorically valid: "Berdache is another
catch-all term that ethnographers have used to describe males (or, less often, females) who take
on at least some of the garments, occupations, and/or sexual partnersculturally prescribed for
what Anglo-Europeans might call the opposite sex" (1993b:351-352). Although there are
qualifications-individuals possess different degrees of the aspects, no single aspect is neces-
sarily present in all berdaches, and so on-traits remain the primary means for identifying
nadleehi (and other Native peoples thought to be similar).
But the presence of the same behavior (such as same-sex sex) in two different time periods
or societies does not mean that the behavior is equivalent in meaning. As Padgug notes, that
different societies "share general sexual forms [does] not make the contents and meaning of
these impulses and forms identical or undifferentiated. They must be carefully distinguished
and separately understood, since their inner structures and social meanings and articulations
are very different"(1979:1 1). The same holds true for attire, occupation, and the other alleged
berdache traits-categorizing individuals across cultures and times on the basis of a handful of
supposedly transcultural and historical features misses the dynamic and rich contexts from
which these behaviors derive meaning.9 Indeed, one need often look no furtherthan within a
single society or to an individual to realize that a trait changes meaning with time, place, and
situation.10
In relying on traits such as "male in women's attire doing women's work" to identify a
nadleehi, those who use these categories make rather bold assumptions. These include the
following: that all cultures define and use the concept "trait"in a similar manner; traits are
equally relevant within and across cultures; individuals define themselves as the categories
imply (that is, as an aggregate of traits);and constructions of self are similar enough to permit
comparison of a nadleehi with other so-called alternate genders, such as a Lakota winkte or a
Tewa kwidb. The categories also require that an individual's identity be separable from other
cultural aspects; otherwise, that identity is not reducible to specific traits. In other words, the
categories cast the individual's relationship with the universe as a discontinuous one. The
following suggests that these and other assumptions are erroneous.

specific categories

gay The use of the term gay to describe nadleehi (and others with perceived commonalities)
involves many issues, of which same-sex sexual practices is perhaps the most contentious.
Roscoe and Williams, both proponents of "gay,"associate the "berdache"or "alternategender"
with same-sex sexual practices in their terminology and categorical constructs. Although they
note that occupation and religious roles are important features of a Native American gay or
queen, (Roscoe's [1988:65] and Williams's [1986:126] terms, respectively), they nonetheless
retain a strong emphasis on same-sex sexual practices. Williams explains that, except for an
occasional opposite-sex relationship, "homosexual attraction is an important aspect of the
berdache character" (1986:125-126). And Roscoe suggests that "alternategender/sexualities"

coming to terms with Navajo nadleehi 269

This content downloaded from 158.143.192.135 on Sat, 21 Sep 2013 15:25:20 PM

All use subject to JSTOR Terms and Conditions
are the trunk of a family tree, from which "berdache/alternate gender" and various forms of
"homosexuality" branch (1991:21 1).1
This reliance on sexuality to classify individuals, however, is a relatively new phenomenon,
and one open to a variety of interpretations.While Boswell (1990) may concur with Roscoe's
idea (1988) of a "gay history,"transcending time and cultural places,'2 others do not. To some
(e.g., Foucault 1978), the construct of "sexuality"is a product, at least in part,of modern power
relations; D'Emilio, for example, contends that "homosexuality"was made possible by capital-
ism (1993:469). Sexuality is not, as Boswell argues (1990:22), a given.
With the development of sexuality, classification shifted away from passive versus active
roles, masculine versus feminine styles, and so forthto a "sexual identity"based on "the binary
play of sameness and difference in the sexes of sexual partners"(Halperin 1990:16). Along with
this "new taxonomy" came the notion that "human beings are individuated at the level of their
sexuality . . . and, indeed, belong to differenttypes or kinds of being by virtue of their sexuality"
(Halperin 1993:417). Padgug describes this as an "enshrinement of contemporary sexual
categories as universal, static, and permanent, suitable for the analysis of all human beings and
all societies" (1979:8).13
It appears that Roscoe, Williams, and others have frequented the shrine of The Perpetual
Homosexual and, in so doing, not only have overlooked the cultural boundedness of sexuality
as a concept but also subsume nadleehi (and possibly others with similar characteristics) under
the principles of present-day sexuality classification-an unfounded inclusion.'4 And yet, as
Besnier notes, an identification with nadleehi and others who may be similar is "understandable
in the context of lesbians' and gays' struggle for a political voice in postindustrial societies";
thus they attempt to "demonstrate that preindustrial societies are more 'tolerant' ... or
'accommodating' of erotic diversity and gender variation than 'the West' " (1994:316). The
benefits of identifying with "preindustrial"societies are many: thus, for example, Williams looks
to "the American Indian concept of spiritualityto break out of the deviancy model to reunite
families and to offer special benefits to society as a whole" (1986:207). And Roscoe adds, "I
have no difficulty imagining the rationale and rewards of specializing in a work otherwise
considered female. My own consciousness has thus absorbed the berdache" (1984:48, as
quoted in Williams 1986:207).'5
The differentiation of humans by sexuality contrasts sharply with the ethnographic record.
Jacobs and Cromwell explain that the Tewa kwid6 engage not only in same-sex sexual practices
but can have sex with men, women, or other kwid6 (1992:56). Several male nadleehi cultural
teachers described sexual relationships with women, and some nadleehi (PK,PA, EB)were (and
are) celibate. Neither are nadleehi the only ones who have same-sex sex. Nadleehi's partners
(sometimes called "straights")generally retain their status as masculine men and are not
considered "homosexuals." Fromthe limited data I have, it also appears that sexual unions with
nadleehi in earlier times16were determined more by issues of convenience and situation than
by one's sexuality.17EB,a nadleehi, explains, "Thenadleehi would help keep heterosexual men
happy, and alleviate their sexual frustrations.... Ifall of a man's wives were pregnant, he could
go to a nadleehi."
But gay is not entirely without its merits: it encourages awareness of both Native Americans'
identities and their terms of self-reference. Williams notes that some Native American individu-
als, having "been exposed to urban gay communities ... see themselves as gay, not berdache"
(1986:215). In identifying strongly with the Euro-Americangay community, "they want to be
treated as men ... not like a halfman-halfwoman" (Williams 1986:215). Moreover, in selecting
the terms gay and queen to discuss nadleehi (and others with possible similarities), Williams
(1986:225) and Roscoe (1988:65) use the terms by which those of so-called alternate genders
are frequently mentioned and by which they refer to themselves. Furthermore,both note that

270 american ethnologist

This content downloaded from 158.143.192.135 on Sat, 21 Sep 2013 15:25:20 PM

All use subject to JSTOR Terms and Conditions
Native American and Euro-Americanmeanings of gay or queen are not necessarily similar to
each other, an observation consistent with my own and others' research.'8
The interplay between the use of gay and nadleehf gives insight into various Navajo ways of
understanding nadleehi, and how the understandings are changing. WA, a nadleehi in her
mid-twenties, appears to emphasize sexuality in noting that nadleehi originally "founded a
homosexual society [in which] they were looked up to." In contrast, HA, a traditional Din6
scholar,19 highlights other-gender behavior when he defines "gay" as "a boy that's acting like
a girl [or vice versa]." CB's focus is different again; she states, "Nadleehi's a holy word and
should not be used to referto gays." And CY suggests such distinctions may be less important
to Navajo elders: "Iwonder what the elders would think of all this. They'd probably just laugh."
While the concept of nadleehi as "holy" is confirmed by Hill (1935), and may be alluded to
in the Navajo origin teachings (e.g., Haile 1981; O'Bryan 1956), this perception was not
necessarily universal. HA notes that "in earlier times" certain clans let hermaphroditic infants
die. The association of nadleehi with hermaphroditismalso varies (GB):while some uphold the
distinction between "real" (i.e., hermaphroditic) and "pretend" (i.e., nonhermaphroditic)
nadleehi, others do not. HA, for example, notes that the concept of nadleehi may have started
with hermaphrodites and was later generalized to the "pretend ones," but the distinction was
not necessarily important-both "real"and "pretend" nadleehi were often accorded respect.
Another Navajo culturalteacher explained that people usually did not talk about genitalia;thus,
who was and was not a "true"hermaphrodite may not have been shared beyond immediate
family members.20
Other historical contexts also need to be considered in Navajo meanings of "gay."With the
attempts to Christianize and assimilate Navajos, many nadleehi have encountered increasing
Navajo intolerance and reduced recognition and support of their-variously defined-identities
and roles.21As a result, nadleehi are often faced with forging new identities that draw on both
Navajo and Western cultures. RN explains, "We tend to get our ideas for what it means to be
gay from Western ideas, so you see [Navajo] gays calling themselves 'queens,' and doing a lot
of the camp that you would see in the Western gay community." 22 But they also retain
identification with "traditional"cultural values of nadleehi: many, for example, still eschew
sexual relationships with other nadleehi and find explanations for their existence in Navajo
teachings.23
Although "gay" can provide important insights into some Navajos' (and possibly others')
classificatory systems, its use as a generic term for nadleehi (or for those Native Americans with
presumed commonalities) remains unwarranted. By placing nadleehi in the schema of sexuality,
proponents of "gay" imply that sexuality, despite its distinctly modern origins, is not culturally
or historically bounded; furthermore,they make the same assumption about many Navajos'
classifications-that is, that definitions of nadleehi are constant; are removed from other social,
cultural, and historical phenomena; and are thus unchanging regardless of time, clan member-
ship, degree of traditionality, or exposure to Christianity. In suggesting that "gay" or other
Euro-American gay-derived terms can function as a generic label, Williams and Roscoe
overstate gay's significance and seem to imply a Native American heritage for Euro-American
gay and lesbian identities. But the ethnographic data are clear: Euro-Americangay and lesbian
roots are not awaiting "discovery"in Native American soils.24

alternate gender and berdache The category alternate gender challenges both "berdache"
and sexuality-based definitions of nadleehi (and others who may be similar).25By suggesting
that more than two genders could exist, alternate gender theorists question the universality of
a binary-gender system,26and with it, the formulation of a "berdache" as simply a "mixer"or
"crosser"of man and woman (Devereux 1937:501; Whitehead 1981:85). As a result, "alternate

coming to terms with Navajo nadleehi 271

This content downloaded from 158.143.192.135 on Sat, 21 Sep 2013 15:25:20 PM

All use subject to JSTOR Terms and Conditions
gender" proponents conceive of nadleehi (or others with possibly shared commonalities) as not
merely derivative of the two genders but as a gender in its own right.
Among its strengths, the category "alternate gender" enables us to distinguish among the
various phenomena lumped together under the category "berdache." Forexample, ceremonial
transvestism (wearing the attire of another gender for ceremonial purposes) is frequently cited
as an instance of "berdache,"since it is a somewhat regular social practice associated with the
gender of the opposite sex. But as Parsons notes, a Native American male dance society may
take on the dress of women for ceremonial practices without being considered "men-women,"
or "berdaches" (1920:28). "Alternategender" accommodates these situations by requiringthat
the trait(such as attire)be associated with a shift in gender status.Thus, if the other-gendertraitis
not associated with a differentsocial role, the individualdoes not belong to an "alternategender."27
The idea of "multiple genders" (that is, more than man and woman) also addresses the
relevance of sexuality-based definitions. Thomas (1993) describes four Navajo genders: female-
bodied women, male-bodied men, female-bodied nadleehi, and male-bodied nadleehi. In this
system male nadleehi same-sex sexual practices are not equivalent to many present-day
Euro-Americanhomosexual or gay practices, since nadleehi partnersare of a differentgender
(usually male-bodied men) than nadleehi, while present-day Western gays and their partners
are of the same gender (Thomas 1993:4-5). By positing sexual practices as secondary expres-
sions of the overall gender system-claiming that gender status occurs first and that sexual
practices are simply expressions of one's gender role-this classification challenges those who
would locate sexual practices in some fundamental, all-determining aspect of the individual
such as sexual orientation (Thomas 1993:5).
Ethnographicdata are initially consistent with the idea of "alternategenders." First,"alternate
genders" are often designated by specific terms such as Zuni Ihamana, Lakotawinkte, and so
forth.28Second, nadleehi (and other Native Americans who may be comparable) have charac-
teristics different from men or women, such as exceptional physical abilities, specific attire,
enhanced "spirituality,"a highly respected status, and association with wealth (Callender and
Kochems 1983). "Alternategender" affords a heuristic benefit, as well, as Blackwood explains:
To call the berdache"role"a separategendergives itthe fullstatusof a "gender"... equivalentto thatof
"man"or "woman".... Theberdachegender,then, is not a deviantrole ... nor is it an alternativerole
behaviorfornontraditional individualswho arestillconsideredmen or women. [1988:171]

Given their unique labels and roles, nadleehi (and possibly those of similar identity) seem to
possess a clearly defined status and are not merely opposite-gender wanna-bes.
Finally, the idea of genders beyond man or woman addresses feminist theoretical, political,
and social goals. Forone thing, categories such as nadleehi and winkte illustratethe point that
gender is not a naturalgiven. As Martin and Voorhies conclude from a review of "berdache"
and others, "biological sex is used by societies to build social categories, and human ingenuity
permits more elaboration on the biological theme than at firstseems possible" (1975:107). Once
we view gender as distinct from sex, several absolute distinctions begin to crumble: a female
(or a male) is not bound by biology to a set role, societal dictates and repression along gender
lines can no longer find justification in an assumed state of nature, and so forth. Additionally,
if gender is a social construct, the assumed ubiquity of a binary gender system falls apart, and
with it the hierarchical dualism between man and woman and (to some) the dualistic order of
modernity itself (Fox-Genovese 1993:246).
More recently, sex has also become a "construct,"a by-product of a hegemonic system that
established sex, gender, and desire as "naturallycongruent" (hence the assumption that a male
becomes a man and has sex only with women [Butler 1990:151, n. 6]). The individual whose
sex does not correspond with gender or desire, such as the male who dresses in women's
clothing or has sex with other males, is seen as disrupting this "heterosexual matrix" (Butler
1990:151, n. 6). In so doing, the individual reveals that gender is a performance, a compilation

272 american ethnologist

This content downloaded from 158.143.192.135 on Sat, 21 Sep 2013 15:25:20 PM

All use subject to JSTOR Terms and Conditions
of various actions prescribed and "compelled by the regulatorypractices of gender coherence,"
giving the "illusion of an abiding gendered self" (Butler 1990:24, 140). Obviously the "discov-
ery" of persons such as Navajo nadleehi or Zuni Ihamana did not singlehandedly bring about
feminism, the push for gender equality, the downfall of modernism, or the upsurge of perfor-
mance theory. Butthe possibility of recognizing more than two genders and of ceasing to assume
that gender identity was necessarily derived from sex is clearly central to many social goals
(deliverance from biology as destiny) and political agendas (disruption of the masculine,
heterosexist hegemony). It is little wonder that with such momentous issues at hand, interest in
Native American gender practices has been high.
As promising as "alternate gender" may appear, it has some significant shortcomings. For
one, it does not adequately address the question of what constitutes "gender." Weston notes
that its classifications do not specify when a role "stop[s] being an instance of gendered
ambiguity, or a variant of masculinity or femininity, and start[s]becoming a gender in its own
right"(1993b:354). Besnier, drawing on Polynesian data, notes that "gender liminals" contrast
sharply with man and woman in "the grammar of kinship" and in the relative "porosity"of the
categories; thus, "gender liminals" may not warrant a full-fledged gender status (1994:319).29
Other problems with "alternategender" include its "uncanny"resonance with "the 19th-cen-
tury categorization of homosexuals as members of a third sex" (Weston 1993b:354).30 Murray
(1994:60) also questions the specificity of "alternate-gender" traits and points out that a
"basketballsuperstar"could belong to an "alternategender":31each category is a specific status
grounded in a shared range of diagnostic features (such as distinctive dress, occupation, and
association with wealth), is linguistically designated, and combines masculine and feminine
behaviors.32 Additionally, "gender" and "sexuality" are assumed to be relevant criteria in
cultures' classifications of persons such as nadleehi and in universally agreed constructs (Murray
1994:59; Weston 1993b:346-347).
Of final concern is the political saliency of "alternate gender." As I have mentioned above,
the construct meshes well with current feminist political, social, and theoretical agendas, at
least initially. But is it not possible that the existence of Navajo nadleehi is not in fact evidence
of third genders but is instead constructed as such, given current theoretical interests?And yet
the configuration of persons such as nadleehi as belonging to "alternate genders" is under-
standable-the marginalization of women, as well as "gender blenders" and other so-called
gender nonconformists, has been fierce (Devor 1989).33
It would consequently appear that the role of nadleehi-and, possibly, of others who share
assumed commonalities-is not one of gender at all. Ironically, casting them as such does not
subvert but reifies-indeed is based upon-the very system it is intended to dismantle: the binary
gender system and its assumed natural coherence among sex, gender, and desire. In setting up
nadleehi (and presumably similar others) as belonging to a "third(or fourth, fifth, and so on)
gender," theorists reify Man and Woman as binary opposites, using them as standards by which
to identify "alternates."Furthermore,the relegation of nadleehi (or winkte, kwid6, and so forth)
to an additional gender status sets gender incongruence apart, keeps the meanings of "man"
and "woman" safe from its disruptive influences, and thus forecloses the opportunity for truly
radical reformulationsof gender. As long as there remains an "alternate,"Man and Woman can
be nothing other than a male who became a man and has sex with a female who became a
woman; anythingelse will be "other,""third,""alternate,""deviant.""Alternategender" theorists
and others seem to have undermined their own efforts: instead of challenging binary classifica-
tion, they have made it requisite to their schemas; and instead of dispensing with the "Othering"
of Woman, they have created a new "Othering"-that of the Native American, Melanesian,
Indian, or other Alternate Gender.34

two-spirit Two-spiritwas formulated by several Native Americans (who are "self-identified

'alternative'sex and gendered" (sic) [Jacobsand Thomas 1994:7]) and differsfrom the preceding

coming to terms with Navajo nadleehi 273

This content downloaded from 158.143.192.135 on Sat, 21 Sep 2013 15:25:20 PM

All use subject to JSTOR Terms and Conditions
three terms in many (butnot all) assumptions about gender. The term is "acontested compromise
to move forwardthe debate in eliminating culturally inappropriateterms,"and includes a wide
variety of Native persons: "cross-dressers,transvestites, lesbian, gay, transgendered, or [those]
otherwise 'marked'as 'alternativelygendered' within tribes, bands, and nations where multiple
gender concepts occur" (Jacobs and Thomas 1994:7).35
The wide range of two-spirits, or persons "who combine both masculine and feminine
personality traits," results in part from the contexts in which the term was proposed. Lang
(1993:13) and Tafoya (1992:256) note that contemporary Native peoples lack identification
with Euro-Americancategories such as gay, lesbian, or bisexual; furthermore, within Euro-
American gay cultures, racism against Native peoples can occur. Affiliation with the "Indian
community" remains important;two-spirits frequently emphasize "Native American identity"
over sexual identities and use Pan-Indian ceremonial features (Lang 1993:13-14).36
The concept's inclusivity also stems from what many view as "Native American gender"
(Tafoya 1992:256; cf. Allen 1981), in which spiritual or social identity is emphasized over
psychosexual. Tafoya suggests that "tribalconcepts" emphasize "relationships, contexts, and
interactions," not the "individuality"of "Euro-Americanconcepts" (1992:256). To him, " 'Gay'
can be seen as a noun, but 'Two-Spirit' is a verb. (This is meant as a metaphoric statement,
where a noun is a place, person or thing, where a verb deals with actions and interactions)"
(Tafoya 1992:256). Two-spirited individuals, then, are acting and interacting to define and
redefine themselves. This dynamic self-defining, according to Tafoya, is based on "Native
tradition" that "emphasizes transformation and change, and the idea that an individual is
expected to go through many changes in a lifetime" (1992:257). Given the constant flux of one's
identity, it is little wonder that a Native American-generated category, such as "two-spirit,"
requires only that one be both male and female, and Native American. The sexual, gender, or
other manifestationsof one's two-spiritednessare understoodto varyas widely as humanityitself.
The perception of self as male (or man) and female (or woman) is also a feature of "berdache"
and "alternategender": in both categories participation in men's and women's activities is a
criterial attribute. But "alternate gender," "berdache," and "two-spirit"do not converge in
respect to the specific gender constructions. Tafoya, for example, notes that Euro-Americans
define gender rigidly, while Native Americans' classifications "range from appropriateness to
inappropriatenessdepending on the context of a situation"(1992:254). Inconsequence, "Native
individuals may be quite comfortable with their presented identity shifting its emphasis on
so-called 'masculine/feminine' behavior, depending upon social context and the behavior/iden-
tity of a partner"(Tafoya 1992:258).
Two commonalities between "two-spirit"and the preceding three categories require brief
mention. First,"two-spirit"also relies on traits, in that this category examines gender-associated
behaviors in order to assess the individual's degree of male- and female-spiritedness. Second,
although "two-spirit"implies that both masculine and feminine can be expressed by the same
person and that gender delineation is somewhat fuzzy, it presupposes that what is masculine
or feminine is for all intents and purposes known and stable across situations and cultures. Such
an assumption is essential to the categories: if male or female were situationally defined, then
what was male in one instance could laterbe defined as female. The "male-female"person would
be unknowable, varying among "female," "male," and "male-female" as the interpretationof
masculinity and femininity changed. By maintaining that masculine and feminine are not
situational values, however, the proponents of the "two-spirit"category ensure identification of
"one-" versus "two-spirited"individuals, regardless of circumstance.
While the term two-spirit offers many benefits, such as its emphasis on Native American
prioritiesand gender constructs, its adoption by academia as a generic label should be carefully
evaluated. The decision to apply any construct or category to many cultures should arise from
sound research, which involves firstdetermining whether the category's criterial attributesand

274 american ethnologist

This content downloaded from 158.143.192.135 on Sat, 21 Sep 2013 15:25:20 PM

All use subject to JSTOR Terms and Conditions
classificatory principles are relevant to other cultures. In that "two-spirit"arose out of concerns
for empowerment and identity in the face of derogatory labels and racism, it may be less precise
as a research tool. As I point out in more detail below, the term two-spirit is not applicable to
many nadleehi and, I suggest, lacks significance for other Native peoples; its usefulness as a
generic term may therefore be limited. Students of Native peoples certainly must listen to those
whom they professto describe. Ifwe impose a term that may be irrelevantto many of our cultural
teachers, however, we are once again re-creatingtheir world to suit our own intentions.37
The conflation of diverse Native cultures in the category "two-spirit"rests on its emphasis on
Pan-Indian identity; in the other three categories, the conflation is not so easily dismissed. Each
of those categories implies that nadleehi, their possible counterparts, and the cultures in which
they live share the same emphasis on gender and possibly other features; prioritize and
interrelate attire, occupation, sexual practices, and so on according to only the categorical
specification of "berdache," "gay," or "alternate gender"; and, through an assumption of
transhistoric or transculturaltraits, class individuals in mutually exclusive categories of man,
woman, and "berdache,""gay,"or "alternategender." Each also suggests that attention to a few
traits suffices to account for the dynamic and diverse processes with which cultures shape (and
are shaped by) their participants. In overemphasizing gender and sexuality, then, the cultural
contexts in which the "alternategenders," "gays," "berdaches,"and "two-spirits"emerged have
been forgotten. While the motivations behind such overemphases may be understandable and
even laudable-such as seeking societies more "tolerant" of diverse gender and sexual
practices; contesting the mandated congruence of sex, gender, desire; undermining rigid,
hierarchical dualisms; uniting against racism; and subverting masculinist hegemonies-the
classifications have subsisted at the expense of deeper ethnographic understanding. An explo-
ration of a specific cultural contextualization begins to reveal what depths have been over-
looked.

Navajo ethnographic findings

These four categories are at odds with the way in which many Navajos understandsuch things
as male and female, the meanings of traits, and the role of traits in defining the individual, and
in which they thereby call into question the analytic utility of the categories for apprehending
the diverse Navajo meanings of the term nadleehi. Ibegin with a version of Navajo worldview,38
relying on it in part because of the diffuse nature of definitions of nadleehi. Forexample, three
nadleehi describe themselves in these terms:
BH: Well,likein mycase, I am differentfroma "straight" male,a heterosexualmale,andfrom
a homosexualmale. I mean with homosexualmales I am uncomfortable, but I'mmore
comfortablewith "straight" males, even thoughthereare hasslessometimesand stuff.I
also connectwith a heterosexualmale,sometimes.
[Later]I am a completewoman,justwithoutthe vagina.I'mgay and I'mcomfortable
with that, but sometimesI wish I were different,you know, like have more womanly
qualities,even the parts.
[Andlateryet in the interview]Likeif I had to choose betweena queen and a woman
[asa sexualpartner],I'dratherhave a womanthana queen. I meanIdo havesome kind
of maleness.
PA: That'sthe funny thing about it, there's an in-betweentype of person. It's hard to
understand.Not being a dragqueen and not being a woman.I thinkit's all psychologi-
cal-I don'twantto be a dragqueen, and I don'twantto be a girl.
PK: A queen is identifiedwith a female.You know,you call hergirl,bitch,witch, and things
likethat.Butforme, I don'tconsidermyselfa girl.I'ma manand am attractedto men.
In the above definitions, nadleehi illustrate several importantpremises. First,an individual is
understood in terms of her interconnections, and as both male and female. Thus, BH and PK
describe themselves in relation to men and women, and PA does so in terms of "dragqueens"

coming to terms with Navajo nadleehi 275

This content downloaded from 158.143.192.135 on Sat, 21 Sep 2013 15:25:20 PM

All use subject to JSTOR Terms and Conditions
and women. Second, no individual's definition is fixed; all vary according to the situation. BH,
for example, changes definitions depending on discussion topic and our increased familiarity.
Finally, while many nadleehf agree on how the definition is structured (namely, as intercon-
nected, situational, and as both male and female), they do not necessarily agree on its content.
The description of nadleehi, then, requires a framework capable of accommodating these
constructs of the individual; the frameworkwithin which my Navajo culturalteachers explained
nadleehi is known as Sa'Rh Naaghaf Bik'eh HozhQ, or "the natural order."
The use of Sa'ah Naaghai Bik'eh Hozh(, however, presents an intriguing tension. An
approach from cultural relativismsuch as the one I attempt necessitates the acceptance of Sa'ah
Naaghai Bik'eh H6zho largely on its own terms-that is, as in DB's definition, "a set of facts or
principles detailing how the world is organized and works." This generates an essentialist
position: Sa'ah Naaghai Bik'eh H6zho appears to exist outside the contexts of time, personal
situation, and larger societal issues, all aspects that I have criticized others' approaches for
overemphasizing. But Sa'ah Naaghai Bik'eh H6zho remains a relevant explanatory framework
of nadleehi for a variety of reasons.
First, essentialisms cannot be entirely avoided; even the most antiessentialist approach
requires some measure of precision. For example, if I were to describe Sa'ah Naaghai Bik'eh
H6zho solely as it relates to issues of power, age, gender, and the like, and so demonstrate that
it was tendentious, I would reduce it to the equally tendentious truths of antiessentialism
(namely, that nothing exists outside context, any supposed truth or constant essence being
derivative). I would assert that cultural teachers err when they describe Sa'ah Naaghai Bik'eh
H6zho as valid and real beyond these contexts. If, however, I choose to take Sa'ah Naaghai
Bik'eh H6zho as factual, I again rely on a set of truths;in this view, Sa'ah Naaghai Bik'eh H6zhp
provides an accurate depiction of the universe and is not entirely reducible to the contexts from
which it was derived. Either interpretationrequires some set of assumptions and by the very
nature of assumptions, an abiding core of reality-a fundamental essence-is present (even if
that reality is the absence of realness). Since essentialism is ultimately unavoidable, it seems far
better to privilege the essentialisms of the cultures in which nadleehi have developed and on
which they act than to privilege the essentialisms of antiessentialism.39
Second, Sa'ahNaaghai Bik'eh H6zho is a system of knowledge containing basic observations
of the universe that, to many, serve as truths.At another level, however, it is also variable: while
such principles as situationality and interconnectedness may remain constant, their interpreta-
tion does not.40 In a sense, it is both essentialist and context-dependent. For example, when
Sa'ah Naaghai Bik'eh H6zhO is used in education, the emphases vary with the educators'
needs.41 It is also cited to support any number of different positions, including who is nadleehi
and which of their behaviors are appropriate;thus its usage and interpretationare malleable in
response to the speaker's intent.42And while many Navajos make sense of the universe in terms
of Sa'ah Naaghai Bik'eh H6zho, whatever its interpretation,others do not.
Finally, this is not a definitive interpretation of nadleehi. I explore Sa'ah Naaghaf Bik'eh
Hozhp because its basic principles significantly challenge the four gender- and sexual prac-
tice-based categories, and so afford insights into nadleehf not otherwise available.43While the
interpretationI favor is by no means universally applicable, it is internally consistent with (and
relies on) the principles of Sa'ah Naaghaf Bik'eh Hozhp, and is one with which I have some
familiarity.
I focus primarilyon five principles of Sa'ah Naaghaf Bik'eh HozhM as DB explained them to
me: it is male and female and organizes everything as male and female; it is a living cycle and
organizes everything as a cycle; it interconnects everything; through that interconnectedness it
cycles everything into everything; and it is an ongoing cycle, since each male or female has the
other (i.e., female or male, respectively) into which it can cycle.

276 american ethnologist

This content downloaded from 158.143.192.135 on Sat, 21 Sep 2013 15:25:20 PM

All use subject to JSTOR Terms and Conditions
 According to DB, Sa'ah Naaghaf Bik'eh H6zh( can be understood in male aspects, or Sa'ah
Naaghaf, which include protection, aggressiveness, and building up one's defenses; and female
aspects, or Bik'eh H6zh6, which include fruitfulness, creativity, and nurturance (cf. Farella
1984). Everythingexists in terms of this arrangement: humans, air, and water as well as less
tangible things like thought or emotions. All males and females are themselves both male and
female, a condition that DB explains in terms of the Sun, the "male" aspect in the Sun-Moon
pair:
The Sun gives you strengthand grows the food you eat. Fromthatyou grow and can do all sortsof
productive,creativethings[i.e.,the femaleaspect].Butyou need protectionfromtoo muchSunas well.
Rememberthatthe Sunsaid,"Thosenay6e [problemsthatarisefroman imbalance]aremy creation."44
The forcesandthe energythata personuses to producenay6earefromthe sun, the air,the water,as is
the energyto protectfromthose nayee.

Thus the male aspect-the Sun-is both male and female by virtue of providing the energy for
protection from nayee (male) and for growth and creative aspects (female).45
The second aspect of Sa'ah Naaghai Bik'eh H6zho, its existence as a living cycle, can be
seen in the passing of seasons or in the individual's life cycle of childhood, maturity,and with
death, the passing back to the soil, air, and other natural elements of that individual's basic
components. DB explains this more fully in terms of the diurnal cycle:
Ifyou look at the timesof day, it is this naturalprocessof sunwise [clockwise]movement.At the east,
where the sun firstappears,there is the white air, the color rightat dawn. As you move in a sunwise
directiontowardthe south,thatis middaywiththe blue air,which is the colorwhenthe sun is rightabove
you. Eveningtwilightis at the west, wherethe sun sets and the airis yellow.Andin the north,it is dark
air,the colorof night.As the sun continuesitsmovement,it goes on to the eastagainandthe cycle starts
all over. Everythingcan be talkedaboutlikethis, as thissunwisecyclicalmovement.

The passing of dawn into midday is a good example of the third and fourth aspects of Sa'ah
Naaghai Bik'eh H6zho, namely, that all things interconnect, and that, through this intercon-
nectedness, all things cycle into all others. The times of day, for example, are not discontinuous,
but are literally in a state of becoming each other, as one time cycles into another. The individual
is no less "well-connected"-but is inseparable from, and cycles into and back from, the
mountains, earth, air, and all other things. DB explains these aspects more fully in the following
example:
Everyaircycle is connectedto allof the other[cycles]andis connectedto yourmotherandfather.During
the nine monthsof gestation,you were raisedin the daily cycles of the white, blue, yellow, and black
airs,and in the 12 airsthatyourmotherwas breathing.Yourfatheralso was raisedwith breathingthose
12 airs,so hisoffspringandhisfluidcamefromthose aircycles. Youwereraisedwiththoseairsallduring
the nine months[ofgestation]and afteryou were born.

As dawn interrelates with the sun, earth, and other times of day, so too are humans deeply
integrated with everything around them. They are literally the air cycles (cycles of the sun, the
earth, or other naturalelements) in which they live.
Sa'ah Naaghai Bik'eh H6zhp is also the continuous cycling of male and female into each
other. Consider, for example, the inherent balance in daily activities: the protectiveness (or
maleness) of a night's restcycles into the fruitfulness(or femaleness) of a productive day's labor.
As DB notes, if one were to work incessantly, exhaustion would force one to replenish one's
strength through sleep. Likewise, rest without labor would result in a lack of food, shelter, and
fulfillment, without which the individual would eventually, as DB also explained, return to
productive (female aspect) activities. And this cyclical alternation between male and female is
inherent to the individual, as HA explains:
We all possessbothmasculineandfemininecharacteristics in ourbody.Forexample,in a tragicsituation,
beinga man [sic],the femalecharacteristics
areelicited.Thesame is trueforthe woman.Somesituation
happensand it elicitsthe male, and she acts like a man. Forthatreason,becauseyou need both, you
have male andfemalecharacteristics.

coming to terms with Navajo nadleehi 277

This content downloaded from 158.143.192.135 on Sat, 21 Sep 2013 15:25:20 PM

All use subject to JSTOR Terms and Conditions
Similarly, DB notes that because both male and female exist, each has the other into which it
can cycle and by which it can be balanced, and the cyclical movement of Sa'th Naaghai Bik'eh
Hozhd can continue uninterrupted.
The implications of the above for trait-based definitions of nadleehi are significant. As I noted
earlier, proponents of the four categories generally focus on a few isolated features of nadleehi
(and of those who may be similar), such as dress and occupation, to designate who is or is not
a berdache, person of alternate gender, gay, or two-spirit. Consequently, the categories
consistently presuppose that both the trait and the aggregate of the traits-the individual-are
not interconnected with the universe.46
But such a conceptualization of trait and individual is at odds with many Navajos' under-
standings.47Reichard has previously alluded to this, noting that in general Navajo categories
are not based on the "unique, the distinctive," as are many Euro-Americanclassifications, but
"on the inclusive" (1944:4).48 In discussing berdache traits DB clarifies this inclusiveness:
So with nadleehi,likethe clothingandstuff,that is so artificial,so why makea big stinkaboutit?Ifyou
to him,
wereto look at thatperson,at all the naturalprocesses[suchas the air,sun,etc.] interconnecting
thatalonewouldfillup booksandbooks.Thenyougetto thisone part,thisartificialpartabouthisclothing.
Ina drawingof himas allof his interconnections, you'dhaveto magnifythatartificialparta milliontimes
even to see it.

Thus, an understanding of nadleehi, if limited to the categories' selected traits, is very

incomplete. To extend DB's analogy, one or two commas have been mistaken for the entire
library. By removing the traits from the cultural contexts, one cannot assess a trait's specific
meaning, the various meanings of the term trait(if it is culturally relevant at all), or the different
ways in which people construe the individual in relation to traits. As DB explained it to me, the
individual is seen as "the unique configuration of all natural processes coming into her or him,"
not as a handful of traits.49
The categories also err in their determinations of maleness (or masculinity) and femaleness
(or femininity). Each category assumes that masculine and feminine are constants; thus one can
say with certainty that a male with a woman's mannerisms is exhibiting "feminine"behavior.
The existence of everything as both male and female, with male and female cycling into each
other, however, suggests that masculine and feminine are not as completely separate or mutually
exclusive as is usually assumed. As DB explained, the Sun's energy gives the strengthto develop
protection (male) and to undertake creative, fruitfulactivity (female). The Sun is not so much a
distinctly male or female natural process as it is both. Given that both male and female are
ever-present, a gender valuation of masculine versus feminine will generally reflect the
perspective of the observer ratherthan some absolute value. HA furtherelaborates on this:
Male and female-it dependson the situationof which one it is thatyou are talkingabout.Thereare
thingsthatwomendo and men do, butthatdepends.... I had threesistersandone brotherand I'mthe
oldest.Ourmotherdied when I was nine, and my babysisterwas ten monthsold. I hadto takecareof
her,scroungeforcloth [fordiapers],and hadto learnto wash a baby. I prettymuchhadto takecareof
the kidslikeI was theirmother... . WhenI got married... my wife askedme to do otherthingsaround
the house,so I did them.... I neverthoughtof these thingsas woman'swork.Itwas survival.

In the above, HA performed what for some are women's tasks by cooking, cleaning, and
washing. Yet each of these is also Sa'ah Naaghai-that is, the male principle of maintaining
one's defenses. The food HA cooked builds one's immune system and protects against infection.
And because male and female are intricately connected, the protection cycles into health and
fruitfulactivity. Were HA's activities seen in acultural terms, as "a male doing female behavior,"
HA would be considered a male nadleehi. By seeking meaning in the interconnected situ-
ations-and specifically that of HA's family's need to survive-we learn that he does not belong
to that category.
Because everything exists as both male and female, gender valuation to many Navajos is
largely situational,50even when it appears in combination with seemingly fixed attributessuch

278 american ethnologist

This content downloaded from 158.143.192.135 on Sat, 21 Sep 2013 15:25:20 PM

All use subject to JSTOR Terms and Conditions
as genitalia. While anatomy is often the basis for female or male social, familial, and kinship
roles, from another perspective each sex's genitalia also belongs to the opposite sex. A Navajo
cultural teacher elaborates on this point:

Ifyou were to look at all of us, we arethe balance[of maleandfemale],andwith nadleehithey arethat
balancetoo. Eventhe organs-at the tipof the penisis a littlevagina[ureter],while on the vulvais a little
penis [clitoris].Thatis how it is saidin Navajo.So you see bothpenisandvagina,butonly one functions.
Itis likethatin the hermaphrodite, too-you see bothbutonly one functions.

This situationality of male and female challenges a key assumption made in the use of
terminology like berdache, person of alternate gender, gay, and two-spirit. Since what is male
or what is female may not be definite, there is no basis for determining whether the individual
has the personality aspects, occupations, attire,and other features of only one gender or of both.
As such, the "two"cannot be delineated from the "one spirit"or the person of alternate gender,
berdache, or gay from everyone else.
Sa'ah Naaghfi Bik'eh H6zhQ also has implications for the assumption that male and female
exist separately from each other, with each the exclusive converse of the other. Because the
categories presuppose that male is distinct from female, the conflation of male with female must
indicate a new class, and this has generated the proliferation of generalized terms such as
berdache and the others. As we have seen, however, to many Navajos all things exist as both
male and female; the mixing of genders is therefore not a criterionof demarcation but an attribute
of similarity. According to DB, that one exists as both male and female aspects holds as true for
nadleehi as it does for a mountain, a tree, a woman, or a man. He went on to note rhetorically,
"Everythingis two, so how can you have this as a third?You don't have man, woman, and another."
Other ethnographic data also support the understanding of nadleehi in terms of male and
female rather than as an additional gender. Hill (1935), for example, observed that a male
nadleehi's murder incurred the same blood payment as a woman's. Kinship terminology for
nadleehi relies on the distinction between male and female, and a male nadleehi is referredto
as "daughter-in-law"when introduced to her partner'sparents (PK,BH). Terms of endearment
between nadleehi and their partners are based on female and male associations, respectively,
and not on association with a third gender. And my nadleehi cultural teachers refer to each
other using female pronouns, not neuter pronouns or pronouns specific to nadleehi.
These are, of course, only general outlines; Nadleehi expressions of male and female are
subject to context. For example, the use of female pronouns for male nadleehi and feminine
behaviors are often restricted to nadleehi-only or nadleehi-sympathetic settings; otherwise
nadleehi use masculine pronouns to referto themselves. WA and TYexplained that other-gender
attire may be limited to Halloween or to "passable" androgynous clothing. And yet this, too,
varies, according to PK:some do engage in such so-called overtly feminine behavior as wearing
makeup and "being more flamboyant" in public situations. Attire and mannerisms are some-
times also influenced by normative standards-to some Navajos, certain body parts should not
be exposed. In keeping with this, EB, a nadleehi, explained that she altered her dress and
demeanor when visiting her grandmother.
Given this degree of variability, let alone major epistemological differences, the four catego-
ries are clearly insufficientto accommodate many Navajos' constructs. Although many Navajos
may describe nadleehi in terms similar to "alternate gender" or "berdache" definitions, the
construction, meanings, and interrelations of these traits and of their gender associations may
arise from a very different set of understandings. While the four categories often allow scholars
to isolate traitsfor analytical purposes, many Navajos understandtraitsonly in the largercontext.
And while the four terms classify gender according to fixed-definition, noncontinuous catego-
ries, many Navajos know of masculine and feminine as a dynamic cycling of male into female,
with its valuation dependent on the setting. PKneatly summarized the salient issue: "Out here,
you don't have the groups you have in the Anglo [cultures]. Time and events and classification

coming to terms with Navajo nadleehi 279

This content downloaded from 158.143.192.135 on Sat, 21 Sep 2013 15:25:20 PM

All use subject to JSTOR Terms and Conditions
and categories, that's how you Anglos try to put everything. You get so caught up, you don't
see people as humans responding to situations."

conclusions

The terms berdache, alternategender, gay, and two-spiritare clearly inadequate, a shortcom-
ing, I suspect, stemming from the continued uncritical reliance on unexamined assumptions
about the centrality of gender and sexual practices. Sexual practices, for example, are often
described in terms of sexuality, a modern invention; thus one could legitimately question the
significance of sexual practices so construed for cultures with different historical constructions
of identity and the individual. The concept of gender practices is no less problematic: the
supposed discovery of third, fourth,and even more genders reifies the binary opposition of man
and woman ratherthan disrupting it, and imposes gender constructs that may be inapplicable.
And these notions of gender and sexual practices employ classificatory procedures at odds with
many Navajos' understandings:both assume rigid, preset meanings for masculine and feminine,
and both confer "identity"on a person on the basis of only a few traits.
And yet the current privileging of gender and sexual practices does not rest entirely with late
20th-century theorists and ethnographers.The proponents of the categories are heirs to a history
of casting and recasting gender and sexual practices as the central concerns. Berdache was used
by explorers and missionaries to point out the supposedly perverted gender behaviors ("men
debased to women's occupations") and unnatural sexual practices ("sodomy and nefarious
practices") (Katz 1976:288-291). Such notions of berdache deviance have tainted modern
descriptions as well, as in Devereux's depiction (1937) of the Mojave alyha and hwame.
Most recently, gender and sexual practices once considered unnatural or debased have
become a source of identity and community. Indeed, many involved in berdache studies identify
themselves as lesbian, gay, bisexual, transgender, or gender-radical. I, for example, choose
women as sexual partnersand play at androgyny. Thus, the very concepts by which nadleehi
and others have been described historically are often personally salient to those researching
berdache today. Littlewonder, then, that gender and sexuality have been reissued in the latest
categories-"gay" works well for those seeking recourse against homophobia; "alternate
gender" resonates deeply for those who find the binary gender system rigid and oppressive; and
in challenging colonialist discourse, "two-spirit"gives voice to those whose sexual and gender
practices have been vilified or appropriated,and perhaps to others seeking their own particular
voices.
In conclusion, gender and sexual practices have occupied our attention for so long that they
have come to seem natural in the berdache studies landscape. The relationship of these
constructs to historical and personal contingency and their influence on the ways in which a
nadleehi is seen, recorded, and depicted have become invisible. This, I contend, has reduced
the value of the categories for research and has obscured the epistemologies of other cultures.
While part of nadleehi identity may indeed lie in gender and sexual practices, it is time to ask
what else may be salient. Future work should begin with attention to the history of specific
cultures, the exploration of multiple systems of meanings (as in other interpretationsof Sa'ah
Naaghai Bik'eh Hozh( and nadleehi), and the identification of culturally specific and relevant
constructs. Such a particularistic focus does not preclude cross-cultural research; indeed, it
should enhance intercultural comparison by ensuring that research proceeds from culturally
valid classifications. With locally salient meanings finally reinserted, new ways to organize the
discourse can emerge, ways that take the analysis beyond gender and sexual practices and
redefine the discourse itself.

280 american ethnologist

This content downloaded from 158.143.192.135 on Sat, 21 Sep 2013 15:25:20 PM

All use subject to JSTOR Terms and Conditions
 notes

Acknowledgments. Funding for the research came in part from a National Science Foundation (NSF)
Dissertation Improvement Grant, an NSF Facilitation Award, a Sigma Xi Grant-in-Aid of Research, a
Northwestern University Alumnae Fellowship, an Eleanor Association Fellowship, and a Northwestern
University Dissertation Year Grant. I was greatly aided by the reviews and comments of Ann Wright, Laura
Helper, Oswald Werner, Derek Milne, Niko Besnier, and by the comments of three anonymous AE
reviewers. I remain responsible for any errors and oversights that may remain.
1. I use the term Navajo rather than Dine (a Navajo word for the people) for several reasons. First,the
Navajo Nation refersto itself as Navajo. Second, the term Dine is less well-known. Third, when I asked one
of my cultural teachers about the preferredterm, I came away with a sense that Dine might have a slightly
more restrictedmeaning in some contexts, and that it could refer only to men.
2. My use of the term traditionalmay be objectionable, and can wrongly suggest that Navajos are divisible
into some sort of static, homogeneous "traditional"group as opposed to one which has changed over time,
adopting various aspects from other cultures. I do not wish to convey this idea; rather,many of my Navajo
culturalteachers use the term traditionalto signify times perceived as less influenced by Euro-Americanand
other cultures. Also, since Navajo plurals are generally indicated by context or the associated verb, I do not
pluralize nadleehi.
3. I limit the critique to the four most frequently encountered concepts. Berdache has long been used to
describe nadleehi in cultural trait lists, anthropological writings, and elsewhere. Alternate gender has
garnered increasing support in a variety of scholars' works. Gaywas made more widely known by Williams
(1986). The fourth is the most recent to gain acclaim. Following two Wenner-Gren conferences on berdache,
many of the attendees called for abandoning the term in favor of two-spirit (Jacobs and Thomas 1994).
Other classifications exist as well. Forexample, Money considers nadleehf to be a kind of "gynemimetic,"
or an individual who "change[s] from he to she," but avoids genital surgery so as to live "full-time as a lady
with a penis" (1988:98). Greenberg uses berdache to describe "transgenderal homosexuality," or an
individual who "relinquishes the gender . . . ordinarily associated with his or her anatomical sex and lays
claim to the gender associated with the opposite sex" (1988:40). Transgenderal homosexuality is not
restrictedto North American Natives-Greenberg includes examples from Central and South America, the
East Indies, Asiatic peoples, and so forth (1988:57-62). Herdt (1991 a, 1991 b) also relies on a concept of
"transgenderedhomosexuality" in his critique of Devereux's description (1937) of the Mohave berdache,
known as alyha (males) and hwame (females). Herdt, however, moves beyond gender shifting, adding, "I
suspect that the erotic is the royal road to understanding cultural ontology in many times and places....
To view the erotic in its social tradition is to see its part in the whole, that is, to recognize that homosexuality
is not reducible to sex acts or to cultural meanings" (1991 a:487-488). On the basis of this approach, Herdt
suggests that "these characteristics [specific sexual practices] support an ontological theory of the berdache
as represented primarilyby its erotic nature. For the Mohave, this erotic nature could be expressed in and
only in gender transformationof social role and sexual nature"(1991 a:497).
Many of the criticisms presented in the essay apply to these categories as well. Reliance on broad
cross-cultural categories overlooks the culture-specific meanings; the focus on gender assumes that gender
is conceptualized, enacted, and partitioned in the same ways universally; the various cultures are lumped
together on the basis of a few traits;generally, Western categorical constructs such as erotic, gender, and
the like are privileged over indigenous cultural constructs and epistemologies without attention to their
saliency. Both Herdt (1991 a, 1991 b) and Greenberg (1988) acknowledge several of these limitations of
cross-cultural research.
4. My critique of the work of Thomas, himself a Navajo, could be interpreted as overlooking an emic
perspective, and in a sense this may be the case. But his use of "gender"without attention to its cross-cultural
relevance suggests a reliance on categories and concepts that may be locally irrelevant.
5. The reliance on Native explanatory models has had a curious trajectoryin the hands of postmodernists
and others. Initiallyit was celebrated as a means to address the "crisisin representation"(Marcus and Fisher
1986) in which "grand theories" and "the hegemony of objectivity" became suspect and were seen as
constrained and determined by "a tendentiousness that is ... based in any major division of power, any
axis along which power differentials are organized and distributed, such as race and gender" (de Lauretis
1990:128). But Native voices were not heard for long. Weston (1993a:6) notes that postmodernist theories
describe the individual as both a passive consumer of power (as de Lauretisindicates) and as resisting such
power axes (e.g., Butler's 1990 description of drag "gender performance"as undermining the coherence of
anatomy, gender, and desire). Ineither case, the "native's"experiences and explanations are often subsumed
by the author's concern with power relations and their effects. Abu Lughod's method for "unsettling the
culture concept and subverting the 'othering' process" illustratesthis particularlywell, which she explains
as the "crafting,reconfiguring, and juxtaposing [of] these women's and men's stories to make them speak
particularlyto my concerns and those of my audience" (1993:13, 16). She rationalizes her use of others'
words for her own purposes by noting it is only "a false belief in the possibility of a nonsituated story (or
"objectivity") [that] could make one ask that stories reflect the way things, over there, 'really' are"
(Abu-Lughod 1993:1 7). It would seem that since an "objective" story cannot occur, one need not attempt
even the slightest approximations, permitting researchers to exploit the natives' words for analytical ends
without attention to the speakers' original intent. Or, as Mascia-Lees et al. point out, precisely when women

coming to terms with Navajo nadleehi 281

This content downloaded from 158.143.192.135 on Sat, 21 Sep 2013 15:25:20 PM

All use subject to JSTOR Terms and Conditions
and non-Western peoples are beginning to "claim themselves as subject[s]," and so demand recognition,
the "ontological status of their subjecthood"-recognition that their realities are "reallythere" and can be,
at least to a certain degree, ascertained and acted on-is refuted (1989:15). Indigenous and other voices are
thus once again muted in favor of theoretical orientations.
In terms of my position, I see neither objectivity nor relativismas absolutes or as entirely attainable; rather,
both are necessary, each providing a balance to the other (cf. Knauft1994). The inability to achieve either
completely should not excuse researchers from attempting description of others, or from considering at
length the individual, societal, and larger issues (such as racism, poverty, and historical events) impinging
on ethnographic assessment, analysis, and representation. As I explain later, that the particular Navajo
worldview I describe is significantly context-sensitive does not make it entirely unknowable or render it
invalid as an explanatory model.
6. In using this quotation, I do not wish to misrepresent my position on scientific inquiry and method. I
see in Halperin's quote a critique more of "scientism" than of science itself. I concur with Paul Roscoe's
assessment that the "scientistically inclined" conflate science with positivism, and, in so doing, erroneously
suggest that science seeks "truth"(as opposed to validation); posit "an Order of Nature ... in which entities
are wholly governed by a transspatial,transtemporalorder of forces or relationships"(as opposed to being
governed by quantum physics, in which "no law can ever apply to all space and time"); and ignore the
constructed nature of categories (as opposed to science's frequent recognition of its own constructions)
(Roscoe 1995:496-501, 505, n. 15).
Although I emphasize throughout notions often associated with "anti-science"-such as the processual
nature of identity, the need for contextual understandings, and a deconstruction of accepted categories-I
arrivedat this perspective using scientific method, seeking as faras possible an approximationof the various
meanings and experiences of nadleehi. I asked nadleehi first about the relevance of "homosexual" to their
identities, and found it in many ways invalid; subsequent inquiries about "alternategender," "berdache,"
and "two-spirit"were equally problematic, requiring new guesses at the meanings of nadleehi, exploration
of alternative hypotheses (such as Navajo Creation teachings, semantic domains, and a specific worldview),
and "modification of interpretations in terms of subjectively perceived consistency and problematicity"
(Roscoe 1995:496). But the use of scientific methods to arrive at ratherrelativistic conclusions should not
be surprising. As Roscoe points out, scientific method is no different from interpretive anthropological
methods: both begin "with an initial guess about the meaning ... move to procedures for 'validation,' and
finally move back to a more developed understanding..." (Roscoe 1995:496, 497).
7. See, for example, Blackwood 1988, Carrier1980, Devereux 1937, Forgey 1975, Jacobs 1968, Katz
1976, Kessler and McKenna 1978, Martinand Voorhies 1975, and Whitehead 1981.
8. The reliance on supraculturaltraits is similar to componential analytical methods, in which specific
atomic components (such as generational level or sex in kinship studies) comprise a metalanguage. The
terms of this metalanguage (much like the berdache traits) supposedly can be applied to any culture,
regardlessof a linguistic equivalent in the culture(s) under study (Goodenough 1972:229; Lyons 1977:318).
9. Halperin's work is particularlyinstructive on this point. He states, "Butredescribing same-sex sexual
contact as homosexuality ... effectively obliterates the many different ways of organizing sexual contacts
and articulating sexual roles that are indigenous to human societies-as if one were to claim that, because
feudal peasants work with their hands and factory laborerswork with their hands, feudal peasantry was the
form that proletarianismtook before the rise of industrialcapitalism" (Halperin 1990:46). The same could
be said of the other berdache traitsand those associated with the other categories considered here. Similar
behaviors do not automatically imply similar meanings.
10. De Lauretis (1990:133-134), among others, notes how the interplay of race, gender, and class
variably define an individual's identity, so that these concepts must logically be bound by the individual's
specific experiences. Forexample, a Euro-Americanwoman may have a very differentsense of what it means
to be a woman than, say, an African American woman. See also Weston's insightfuldescription (1993a) of
gendering practices among lesbians, illustratinghow the larger contexts of ethnicity, age, physical build,
socioeconomic status, bigotry, and so on confer differentmeanings on a single feature. Weston also explains
as well that an individual's own "gender" meanings can vary: "However carefully crafted a lesbian's
presentation may be, once she brings it out into the street, that presentation can be jarred into a different
interpretiveframework by the teenager who throws a rock at her and calls her 'dyke' " (1993a:14).
11. Halperin provides an insightfulcritique of this approach:
Forso long as sexuality, like sex, was thought to be rooted in nature, historiansand anthropologistsguided
by that assumption were bound to unearth merely different "attitudes"to or "expressions" of sexual-
ity-historically or culturally variant responses to the universal "fact"of sexuality, local improvisations
on nature's unchanging theme; that theme, moreover, regularly turned out on inspection to be a
remarkably familiar one, uncannily recapitulating (and thereby reaffirming)traditional categories and
experiences. [1990:7]
12. Roscoe partlydefends his Iinkingof Euro-Americangay practices with earlier(and present-day)Native
American practices by explaining that
Symbols derive meaning from social contexts, but the play of meaning is not limited to the context in
which the symbol was first created. A thought, a concept, any signifier can escape its context and
disseminate through references and citations across endless texts and historical documents to be given
meaning again-rethought-by another individual in another context. [1988:15]

282 american ethnologist

This content downloaded from 158.143.192.135 on Sat, 21 Sep 2013 15:25:20 PM

All use subject to JSTOR Terms and Conditions
Thus "gay" need not be restricted to the cultural and historical contexts from which it has emerged; but
without the cultural and historical contexts of its first occurrence it is difficult (or perhaps impossible) to
ascertain whether it retains its original meaning in subsequent contexts.
Boswell errs in a similar fashion, stating,
Even if societies formulate or create "sexualities"that are highly particularin some ways, it might happen
that different societies would construct similar ones, as they often construct political or class structures
similar enough to be subsumed under the same rubric (democracy, oligarchy, proletariat, aristocracy,
etc.). [1990:35]
The advantages of being shorn of cultural and historical contexts become apparent:just as another culture's
social organization can be labeled according to such present-day classifications as "democratic,"so people's
sexual practices can be recast in current terms such as homosexual or gay.
13. Although the discussion has focused on the implications of sexuality contained in the term gay, it
would be wrong to suggest that this term is solely about sexuality. Herdt and Boxer provide a concise
summary of the other connotations contained in the term:
"Gay," in other words, represents more than a sexual act.... It signifies identity and role, of course, but
also a distinctive system of rules, norms, attitudes, and, yes, beliefs from which the culture of gay men is
made, a culture that sustains the social relations of same-sex desire. [1992:5]
14. Other problems are implicit in the reliance on sexuality. Padgug notes that recent sexuality
classifications treat "sexuality [as] a separate category of existence," one that operates independently of
"other spheres of reality.... Such a view necessitates the location of sexuality within the individual as a
fixed essence, leading to a classic division of individual and society" (1979:8). Although I do not describe
the matter at length, DB points out that many traditional Navajos understand the universe as exquisitely
interconnected, with the individual literally inseparable from the mountains, the sky, the earth, and all other
natural phenomena. To ground the definition of nadleehi in sexuality presupposes that the individual is
independent of these interconnections and overlooks the highly integrated nature of the individual's
existence.
Present-daysexuality classifications also rely on a split between subject (such as nadleehi) and object (the
subject's partner)(Herdt 1991a:501). Additionally, the subject's "deep nature" can be discerned through
knowledge of the object's anatomical sex-that is, whether this is the same as or different from that of the
subject (Herdt 1991 a:501). The same assumption underlies the incorporation of nadleehi into sexuality: in
looking at male nadleehi (subject), the researcher learns that their partners (objects) are also male, and
thereby grants nadleehi "homosexual" status. Here again, however, many Navajos' ideas of interconnect-
edness and cycling suggest otherwise. Elsewhere I describe how nadleehi and "straights"are simply the
ongoing cycling of the universe, in which each cycles into the other (Epple 1994). And, as will be discussed
shortly, the individual is a composite of these interconnections, not determined by one aspect (such as sex
of the sexual partner).
15. Although both authors acknowledge differences between Euro-American and Native American
meanings of gay, they clearly conflate the meanings for their political and personal purposes. It is little
wonder that Jaimes, a Native American woman, objects to such perspectives: "Particularlyoffensive have
been non-Indian effortsto convert the indigenous custom of treatinghomosexuals (often termed 'berdache'
by anthropologists) as persons endowed with special spiritual powers into a polemic for mass organizing
within the dominant society" (1992:333).
16. I use the phrase "earlier times" to refer to a time when Navajo cultures were less influenced by
Euro-Americanaspects. Many of my cultural teachers spoke about time in this manner. The category is
fraughtwith problems such as an assumed homogeneity of practices in these so-called earlier times, but it
does avoid the Eurocentricdelineation of pre- versus postcontact.
17. I concur with Halperin's assessment of the prominence afforded sexuality:
If there is a lesson that we should draw from this picture of ancient sexual attitudes and behaviors, it is
that we need to decenter sexuality from the focus of the cultural interpretationof sexual experience. Just
because modern bourgeois Westerners are so obsessed with sexuality, so convinced that it holds the key
to the hermeneutics of the self (and hence to the social psychology as an object of historical study), we
ought not therefore to conclude that everyone has always considered sexuality a basic and irreducible
element in, or a central feature of, human life. [1990:242]
18. For example, BH and PA were adamant that they did not identify with "Anglo homosexuals";
nevertheless, they referredto themselves and peers as "gays."PKalso noted, "Anglo gays screw each other,
but for two nadleehi, that relationship is forbidden"; she nonetheless referredto herself and other nadleehi
as "gay"or "queen."
19. The phrase "traditional Dine scholar" refers to Navajo (Dine) individuals who are particularly
knowledgeable about Navajo worldviews. I have borrowed the term from a Navajo Community College
project in which hataatff(sometimes glossed as singer or medicine person), traditional and contemporary
Dine scholars, and others developed the "Dine philosophy of learning," an extensive exegesis of Navajo
knowledge systems (Navajo Community College 1992). I was fortunateto work with three participantsfrom
the project, MB, DB, and HA.
20. The meanings Navajo attributeto gay may arise from the distinction between "real"and "pretend"
nadleehi, with gay or even queer sometimes used by non-nadleehi Navajos to refer to "pretend nadleehi."

coming to terms with Navajo nadleehi 283

This content downloaded from 158.143.192.135 on Sat, 21 Sep 2013 15:25:20 PM

All use subject to JSTOR Terms and Conditions
 21. Two nadleehi's experiences when they were growing up illustratethis further.PK'sfamily is rather
"traditional"and gives her extensive support:when young, she slept with the female children and was never
discouraged from pursuing gender activities associated with females. Her partnerswere treated as sons-in-
law by her parents, her father told her about Navajo traditional teachings on nadleehi, and so forth. In
contrast, the devout Christian parents of another nadleehf kicked her out of their home when they found
out she was "gay."The tensions were so severe that when a newspaper article dealing with AIDSawareness
mentioned her name, she spent several days in another town to limit family contact.
22. The sources for Western ideas of gay identity are numerous. The nadleehf I know, for example, watch
a great deal of television and frequently rent gay and lesbian videos. In fact, RN keeps me apprised of the
best places to get films, which ones are worth viewing, and so forth. Additionally, PKnoted that the queens
travel frequently to nearby metropolitan areas such as Albuquerque, Phoenix, and Denver to take in gay
bars, pick up sexual partners,and socialize. In doing so, they encounter and adopt aspects associated with
Euro-Americangay and gender practices. For example, the so-called berdache trait of cross-dressing has
expanded beyond wearing women's clothing to include the use of makeup, fingernail polish, feminine
hairstyles, and so forth, particularlyin settings sympathetic to nadleehi.
23. The avoidance of other nadleehi appears to be somewhat akin to incest taboos. With less cultural
support for nadleehf, however, some nadleehi overlook this restriction. This, combined with several other
developments (such as borrowing from Western gay practices and increasing intolerance), may lead to the
adoption of sexuality-based identities by more nadleehi. Some may increasingly coalesce around the very
things for which they are being oppressed-sexual practices and other-gender behaviors-as have many
modern Western gays and lesbians.
"Traditional"cultural values of nadleehi vary, as is illustrated by different clans' treatment of nadleehi.
Where nadleehi were accepted, there seems to be some agreement on what their range of behaviors might
include; examples given by PK, PA, and BH include mediation between men and women, performance of
men's and women's work, some cross-dressing (although this varied by individual), and certain ceremonial
dances (see also Hill 1935).
Thomas, relyingon a shared set of "traditional"values (1993:6-8), has developed a continuum of variation
in nadleehi, and provides furtherinsight into the different ways nadleehi express themselves. To Thomas,
traditional nadleehi live on the Navajo Reservation, live with or near female relatives, are deeply involved
with their extended families, perform ceremonial and domestic activities as an occupation, are rich in
cultural terms, participate in ceremonials, possess extensive cultural knowledge, were educated on the
Reservation,and identifywith traditionalnadleehf. Transitionalnadleehf live both on and off the Reservation;
maintain strong family ties; are secretaries, nurses, and teachers; have less involvement in ceremonial
practices but retain a fair degree of cultural knowledge; are more Western-educated; and identify with
Western gays and traditional nadleehi. Contemporary nadleehi have a small degree of Reservationcontact
and ceremonial and cultural knowledge and identify with Western gays and "two-spirits."Acculturated
nadleehf, as the term implies, have lived off the Reservationfor several generations, have littleor no cultural
knowledge, and identify exclusively with Western gays and "two-spirits."
24. I use the metaphor of soil and roots as a satirical comment on Roscoe's family tree of alternategender
and sexual practices, as well as on Columbus's so-called discovery. It is not meant to imply some single set
of roots common to all lesbians and gays or that all Native Americans are grounded in the same "soil" of
identity, experiences, or ways of knowing.
25. I combine the discussion on "alternategender" and "berdache," since each places an emphasis on
gender roles, and the term alternategender is in some ways a refinement of berdache.
26. Martin and Voorhies (1975) were among the first to use a concept of "alternategenders," or what
they termed "supernumerarysexes."
27. A comparison between the Lagunadancers and the Navajo nadleehf's participationin the Nine Night
Ceremonial illustratesthis further.Inboth situations, the dancers wear other-gender clothing, but the Laguna
do not view their dance as partof a largersocial role. For the Navajo nadleehi, however, the attire is a part
of their alternate gender status, above and beyond the ceremonial functions.
28. Jacobs and Cromwell explain, "Differentgroups may have differentcriteriafor gender determination;
it is through the examination of labels attached to organized criteria that we can locate the range of human
gender and sexual variation"(1992:49).
29. Although Besnier's research was among Polynesians, adequate parallels warrantuse of his data. He
notes that Polynesian kinship "is structured on the basis of a fundamental opposition and asymmetrical
complementarity between male and female entities, which leaves no room for an 'in-between' category...
neither does one find any overt reference to gender liminality in praxis" (1994:319). The same is true for
male nadleehi, as I explain later;if they are accepted by their families, they are frequentlytreated as females
in kinship reference and practices. In terms of the porosity of the gender liminal category, Besnier explains
that "particularmen may retreatout of gender liminality in the course of their lives [by marryingwomen],"
an occurrence that has "no counterpart in the grammar of gender: no boundary crossing ever takes place
between men and women [i.e., a man does not become a woman or vice versa]" (1994:319). A similar
observation can be made of Navajo practices. PK, BH, and PA described how adolescent "queens" are
encouraged by parentsto "settledown and have a family." Some of those who heed their parentsare known
as "less masculine" men or "straights"and, although married and sexually active with women, may later
become the sexual partnersof male nadleehi. Thatthese individuals are no longer considered to be nadleehi
is indicated by the change in descriptor (i.e., from "queen" to "straight")and by the generic view that a

284 american ethnologist

This content downloaded from 158.143.192.135 on Sat, 21 Sep 2013 15:25:20 PM

All use subject to JSTOR Terms and Conditions
relationship between a nadleehi and a "straight"is not incestuous. (In contrast, a relationship between two
nadleehi is considered incestuous [Hill 1935].) If nadleehi in fact constituted a "gender" class, one would
expect a similar phenomenon in the other genders-that is, that a Navajo man could become a Navajo
woman or vice versa-but this is not the case. Although, as I explain later, "man" and "woman" are both
comprised of male and female, with male and female cycling into each other, the social category of "man"
is nonetheless distinct from that of "woman." (See Epple 1994 for a more thorough explanation of this
apparent paradox.)
30. The characterization of nadleehi as a third gender emphasizes many of the same features character-
istic of "fairies,"or effeminate men, at the turn of the century. As Chauncey explains, among working-class
men in New YorkCity, "the fundamental division of male sexual actors. .. was not between 'heterosexual'
and 'homosexual' men, but between conventionally masculine males, who were regarded as men, and
effeminate males, known as fairies or pansies, who were regarded as ... members of a 'third sex' that
combined elements of the male and female" (1994:48). Inboth "alternategender" and "fairy"classifications,
gender is given primacy-the fairy was a "she-man"-with sexuality of secondary importance (Chauncey
1994:47, 80). Like nadleehi, fairies generally did not engage in sexual practices with each other but took
on relationships with masculine men, who also engaged in sexual practices with women. Furthermore,the
sexual partnersof both nadleehf and fairies were (and in the case of nadleehi, are) not considered "different"
because of their sexual relationships with the "alternategender" or "thirdsex"; rather, they retained their
status as masculine men (Chauncey 1994:66).
Butthe classifications ultimately reston differentepistemologies. The gender schemas among the working
class differentiatedsharply between men and women, treated male and female as absolute values, seemed
to view the "fairy"as a gender distinct from man and woman, and so on. For these and other reasons,
nadleehi and "fairy"are only superficially comparable categories.
31. As pointed out by a reviewer, "basketball superstar"often brings to mind an African American,
thereby interjectingethnicity into the gender analysis. As I have shown above, cultures (and by extension,
ethnicities) figure prominently in the ways in which an individual constructs gender. To some the use of
"basketball superstar"could suggest a glossing of African American gender constructs, an interpretation
Murraydid not intend.
32. Murray explains that the superstar's feminine behaviors include "display/commodification of a
skimpily dressed and widely fetishized body [and] frequent public embracing of each other" (1994:60).
33. Indeed, my own research was initially intended to rebut the homophobia of "the West" and, by
looking to an alternate gender, to challenge binary gender systems.
34. Besnier's comments on the "othering"of "gender liminals" is particularly incisive. He notes, "So
characterizing gender liminality as something other than a third gender is not simply the result of the
naturalizationby Western ethnographers of gender as a uniquely dichotomous phenomenon, as is popularly
maintained in many gender- and gay-studies circles. Indeed, the insistence on viewing liminal individuals
as forming a third-gender category can be equally criticized as a Western romantic construction of the
'Other' as 'different' from a reified 'Western' view of sex and gender, which itself is in need of critical
clarification"(Besnier 1994:320). Forexamples of additional "thirdgenders," see Herdt's collection (1994).
35. The absence of queer from the list is intriguing.It is possible that the authors decided to avoid it given
its frequently derogatory use. In my experiences among the Navajo, queer was not generally used by
nadleehi, and, when other Navajos used it, the term had connotations of deviance.
36. Both Lang and Tafoya speak in generalizations-"Native American identity," "Native American
gender," "tribalconcepts," and so on. These characterizations may have arisen from their close work with
"two-spirits."In many ways, "two-spirit"is a term of identity politics, in which differences are glossed so as
to form a community for empowerment, social action, and acceptance, for which a unifying ethnic and
gender identity would be essential.
37. This is not an easy issue and deserves fuller discussion. Perhaps those of us who engage in this
terminological debate should be willing to countenance many contexts, and, within those contexts, different
terms. Among some Native peoples, two-spirit is, at least for the present, one of the best options. Forthose
who argue the finer points in journals and conference halls, a differentset of terms will be necessary. Knauft
offers an important caution, applicable to all who would undertake such debates: "Exploring cultural
richness easily leads to endless relativity. It can also lead to blind empathy or uncritical sympathy"
(1994:123).
The use of two-spirit as a generic term should also be seen in its larger context. As I mentioned earlier,
the term was proposed as a replacement for berdache during a conference including Euro-Americanand
Native American academics, as well as Native American "two-spirits,""gays," "winkte," "nadleehi," and
others. The conference is to be lauded, as it clearly sought to explore issues of representation, "othering,"
and "authority."And it was out of these urgent concerns that two-spirit was proposed and accepted. But
there must be ways of accomplishing these goals without losing analytical precision.
38. Given my primaryNavajo cultural teachers-DB, HA, and MB, the last an Anglo woman well-versed
in Navajo understandings-the presentation is a kind of pristine knowledge, one that traditional Navajo
scholars like DB and HA have spent numerous years exploring and learning. Other Navajos have different
ways of understanding and explaining the universe, and many are ignorant of traditional Navajo forms of
knowledge.
39. It is inevitable that individuals have some essentialistic notions of themselves. Weston's description
(1993a) of lesbian gender practices suggests as much, as did nadleehi (PK, BH, EB)on occasion. Should

coming to terms with Navajo nadleehi 285

This content downloaded from 158.143.192.135 on Sat, 21 Sep 2013 15:25:20 PM

All use subject to JSTOR Terms and Conditions
these findings be ignored simply because they are essentialistic, thereby mandating that all descriptions of
reality be subjugated to the most current standard, that of nonessentializing? Not only would that preclude
learning of other ways of knowing, but it imposes, yet again, a Euro-American,academically derived order
on the world. See also Mascia-Lees et al. 1989 for a similar discussion on related issues.
40. Sa'ah Naaghaf Bik'eh H6zhp definitions can vary widely. Witherspoon (1975:75), following Haile
(1938:31, n. 69), describes it in terms of inner (Sa'ah Naaghaf) and outer (Bik'eh H6zhQ,) forms; Farella
(1984:181) discusses it as a kind of completeness; Reichard (1990:45) defines it as the "synthesisof beliefs"
about the naturalworld and humankind;and Wyman (1970), among others, frequentlyglosses it as "long-life
happiness." And yet, even a cursory look at Navajo teachings on Sa'ah Naaghai Bik'eh H6zhp reveals that,
despite different interpretations,the underlying principles are often the same: all things are male and female,
the world is a cyclical process, and all humans are interconnected with their surroundings. For example,
Farellanotes that "Everything,as any Navajo will tell you, can be divided into male and female.... [R]ecall
that s'a naghai [sic] is equated with ... the male component within all of us (bik'e hozho being the female)"
(1984:133, 170). Reichard learned from Singer of Rain Ceremony that Sa'ah Naaghai Bik'eh H6zhp is a
male-female pair (1990:310); Wyman describes it as "long-life boy" and "happiness girl" (1970:28); and
Kluckhohn points out thatchants, rivers,plants, and other items are arrangedas male and female (1968:681).
Matthews makes a similar observation: "Thereare many instances in Navaho language and legend where,
when two things somewhat resemble each other, but one is the coarser, the stronger, or the more violent it
is spoken of as male or associated with male; while the finer, weaker, or more gentle is spoken of as female,
or associated with the female" (1897:235).
The cyclicity of Sa'ahNaaghai Bik'eh H6zhp is apparent in the passing of seasons, as male (winter)cycles
into female (summer).The Navajo hataalif Slim Curly explained it thus: "Therebythe earth, when vegetation
appears in the spring, becomes as a young woman clothed in a new dress, whereas harvest in the fall lets
her appear as a declining old woman. White Shell Woman is, in reality, the earth which changes in summer
and becomes young again, then relaxes or dies off in winter" (Slim Curly as cited in Wyman 1970:28).
According to Wyman, Slim Curly "called the earth 'Changing Woman Happiness' [female aspect] for
summer and 'ChangingWoman Long Life'[male aspect] in winter"(1970:28). (ChangingWoman and White
Shell Woman are sometimes the male and female aspects, respectively, of the same natural process,
according to DB.) Reichard (1990:47), Farella (1984:149) and Matthews (1897:34) heard and reported
similar descriptions.
Interconnectedness is also evident in a range of Sa'ah Naaghai Bik'eh HozhQ descriptions. Briefly, DB
explained that the individual is inseparable from nature,from which that person receives guidance, thought,
and all things necessary for life. Forexample, Wyman writes of Changing Woman Long Life and Changing
Woman Happiness (i.e., Sa'ah Naagh6i and Bik'eh H6zhp), "They can make conditions blessed before,
behind, below, above, all around one everywhere" (1970:30). Reichard explains that in Navajo philosophy
a human relies on "all beings in the sky, on the earth, in the waters, under the earth, and in the subterranean
waters" to "make [her or him] one with the universe" (1990:49). The idea of being one with the universe
was also identified by Kluckhohn,who stated that "everything in the universe is interrelated"(1968:679);
by Farella,who described the view that such things as the Sun, Moon, and the times of day "transmitthoughts
and feelings to a person" (1984:108); and by Witherspoon, who observed that an individual must
"harmonize and unify"with Sa'ah Naaghai Bik'eh H6zhQ (1974:56).
Thus, despite how they are interpreted,the same principles underlie many descriptions of Sa'4h Naaghai
Bik'eh H6zhp. Inother words, certain principles have remained fairlyconstant over time. Given their usage
in explaining the individual's place in the universe, these principles may affordsome particularlyimportant
insights into how many Navajos understand nadleehi.
41. A very clear example of how Sa'4h Naaghai Bik'eh H6zhp interpretationsvary is the Dine (Navajo)
Educational Philosophy (DEP).While all versions of the philosophy have been based on Sa'ah Naaghai
Bik'eh H6zhp, the content has varied over the years. Initially, the DEP divided educational concepts and
student "goals for living"into four partsof a cycle, corresponding with the four directions. MB and DB noted
that while some found it easy to apply, others felt it was "too simplistic" and that it lacked the more profound
elements of Navajo philosophies. Some also contended that the interpretation reflected the author's
particularreligious affiliation.
In response, a more abstract and less formulaic set of ideas was developed, dealing with the "natural
process" of cyclicity and its inherent male and female aspects. (My research is derived from this version.)
Most recently, the focus has shifted again, this time to emphasize t'ia h6 ajit'eegoo, or, roughly, "Itis up to
the individual to determine one's own life course." The current DEPversion restson the abstractknowledge
explored earlier, while seeking a model more easily understood and implemented by Navajo and non-
Navajo faculty. Additionally, each version is affected by the College's mandates and goals, personalities,
and so on.
Or consider the Navajo Diyin Dine'e (glossed as Holy People), which DB explains are known to some
Navajos as "gods" while others describe them as "naturalprocesses." And yet, as anyone who has spent
much time on the Navajo Reservationknows, such variability is accepted: no single version of Sa'4hNaaghai
Bik'eh H6zhp is necessarily considered the only acceptable version. Indeed, Navajo ceremonial songs often
end with the statement "they say," which I interpret as meaning, "This is one way of saying it." DB, for
example, described his knowledge as coming from many places, such as the sun, the earth, and so forth.
As such, it is a knowledge that has been around for a long time, a knowledge that is capable of giving rise
to many different things, including interpretations.

286 american ethnologist

This content downloaded from 158.143.192.135 on Sat, 21 Sep 2013 15:25:20 PM

All use subject to JSTOR Terms and Conditions
 42. For instance, DB and PK explain nadleehf as the universe's balance of male and female. As such a
male nadleehi can have a sexual union with a male or a female, since in either case it will be a male and a
female coming together (the maleness of a male nadleehi with a female sexual partneror the femaleness of
a male nadleehi with a male sexual partner).DC and others contend that a sexual union of a male nadleehi
with another male is a male-male union and is counter to the universe's balance of male and female. When
I asked one nadleehi about this latter interpretation,she suggested that the speaker was biased because he
disliked several of his son's friends who were queens. Clearly, Sa'ah Naaghai Bik'eh Hozh6 can be used as
an authoritative "truth"according to context. The intriguing aspect, however, is that the terms of discus-
sion-the universal balancing of male and female-remain the same.
There are many other examples of the invoking of Sa'ah Naaghai Bik'eh H6zho to interpretphenomena.
When a local college was struck by lightning near the administrativeoffices, some interpreted it as a sign
of imbalance specific to administrators;others saw it as an indication thatthe college overall needed balance;
still others suggested that because another campus of the college was also struck,the two campuses needed
to work toward greater harmony between them.
43. In relying on these principles (although they are presented through a Euro-Americaninterpretation),
I concur with humanistic approaches and no doubt open myself up to much criticism. Yet, other ways of
knowing do have something to say (see Faris 1990; Turner 1996). Reducing them to religion (as has been
done in the past) or to discourses of power (a more recent strategy) serves to promote Euro-American
arrogance, leaves unexamined a highly valuable and insightfulexplanatory model, and ignores important
informationabout how the world is understood to work.
44. Of the many examples of nayee DB provided, two are the hole in the ozone, created by pollution in
the desire for industrialprofits,and high cholesterol, which individualsbringon themselves by "eating greasy
foods all of the time." In several ceremonials and teachings the Sun claims at least one nay6e as his son
(Haile 1938:97, 1981:188; Matthews 1897:113; O'Bryan 1956:81). Other nayee are said to arise from the
masturbationthat occurs during the Separation of the Sexes (O'Bryan 1956:7). While many translate nay6e
as "monsters,"I preferthe definition "problems that arise from an imbalance" (Navajo Community College
1992:27).
45. An example perhaps more familiarto some is human chromosomes (a comparison suggested by DB).
For example, on the initial level, there is a male individual who is comprised of genes from a male and
female (i.e., his parents). In terms of the genes each parent contributes to him, these too arose from male
and female, in that each parent is also comprised of the respective maternal and paternal genes. And so it
goes through the generations-every individual is comprised of male- and female-contributed genes, with
every female- (or male-) contributed gene itself derived from male and female. Everythingin the universe is
similarly comprised of multiple layers of male and female.
46. The understanding of the individual as interconnected varies among Navajos. It is more prevalent
among those who live in more remote areas of the Reservation, those who spent less time in boarding
schools, and older individuals.
47. And yet, for some Navajos, traitscan play an importantpartthat encompasses gender aspects. During
a field trip to an off-reservationcity, for example, three Navajo students were alarmed to see a male in drag.
To them, the drag queen was clearly just a compilation of traits, a person who was not following set
expectations of how a man should look. Their ideas of gender, at least at that moment, reflected a certain
fixity of what is meant by masculine and feminine. When I suggested they see the individual as simply a
balancing of male and female, however, they agreed that my interpretationwas a valid one. One student
explained, "He was just right there, right there going down the street. Itwas just a shock and I guess that's
why I made such a big deal about it."
48. This is actually only one side of the matter.As I have explained elsewhere (Epple 1994), many Navajos
view the individual as both inseparable and distinct. But in terms of the current discussion the individual's
interconnectedness clearly challenges the atomistic nature of trait-baseddefinitions.
49. Weston's look at the features of "butch"versus "femme" lesbians also illustratesthe difficulties with
trait-basedanalysis. Forexample, she describes a butch (a lesbian generally associated with more masculine
demeanor or dress) who is fond of bubble baths, and a femme (a lesbian generally associated with more
feminine mannerisms and dress) who hates high heels and finds men's clothing more comfortable and
convenient to wear (Weston 1995:241-240). She concludes, "Traffickingin the abstractions of gender traits
artificially isolates gender from class, race, sexuality, ethnicity, and context.... To grasp what's going on
here, you need something more than gendered terminology, because forced dichotomies like butch versus
femme do not accommodate multiple lines of identification"(Weston 1995:241-242).
50. Butthe very fact that male and female are both recognized suggests they are also distinct. Forexample,
an elder Navajo advised one of my cultural teachers that washing certain women's garments was not
appropriatefor a man. Thus there is something known as work appropriateto women and men, respectively.
The distinction, however, does not negate the proposition that each kind of work is also male and female
at deeper levels, much as one could say an individual with an XXchromosomal arrangement is female, but
at a deeper level is comprised of male and female (i.e., a father's and a mother's) chromosomes.

references cited
Abu-Lughod, Lila
1993 Writing Women's Worlds. Berkeley: University of California Press.

coming to terms with Navajo nadleehi 287

This content downloaded from 158.143.192.135 on Sat, 21 Sep 2013 15:25:20 PM

All use subject to JSTOR Terms and Conditions
Allen, Paula Gunn
1981 Lesbians in American Indian Cultures. Conditions 7:67-87.
Besnier, Niko
1994 Polynesian Gender LiminalitythroughTime and Space. InThirdSex, ThirdGender: Beyond Sexual
Dimorphism in Culture and History. Gilbert H. Herdt, ed. Pp. 285-328. New York:Zone Books.
Blackwood, Evelyn
1988 Review of The Spirit and the Flesh: Sexual Diversity in American Indian Culture. Journal of
Homosexuality 15(3-4):165-1 76.
Boswell, John
1990 Revolutions, Universals, and Sexual Categories. In Hidden from History:Reclaiming the Gay and
Lesbian Past. Martin Duberman, Martha Vicinus, and George Chauncey, eds. Pp. 17-36. New York:
Penguin Books.
Butler,Judith
1990 Gender Trouble: Feminism and the Subversion of Identity. New York:Routledge.
Callender, Charles, and Lee M. Kochems
1983 The North American Berdache. CurrentAnthropology 14:443-470.
Carrier,Judd
1980 Homosexual Behavior in Cross-CulturalPerspective. In Homosexual Behavior: A Modern Reap-
praisal. Judd Marmor,ed. Pp. 100-122. New York:Basic Books.
Chauncey, George
1994 Gay New York:Gender, Urban Culture, and the Making of the Modern Gay World 1890-1940.
New York:Basic Books.
de Lauretis,Teresa
1990 Eccentric Subjects: Feminist Theory and Historical Consciousness. Feminist Studies 16(1):
115-150.
D'Emilio, John
1993 Capitalism and Gay Identity. In The Lesbian and Gay Studies Reader. Henry Abelove, Michele
Aina Barale, and David M. Alperin, eds. Pp. 467-476. New York:Routledge.
Devereux, George
1937 InstitutionalizedHomosexuality of the Mohave Indians. Human Biology 9:498-527.
Devor, Holly
1989 Gender Blending: Confronting the Limitsof Duality. Bloomington: Indiana University Press.
Epple, Carolyn
1994 Inseparableand Distinct: An Understanding of Navajo Nadleehi in a Traditional Navajo World-
view. Ph.D. dissertation, Northwestern University.
Farella,John
1984 The Main Stalk:A Synthesis of Navajo Philosophy. Tucson: University of Arizona Press.
Faris,James C.
1990 The Nightway: A History and a Historyof Documentation of a Navajo Ceremonial. Albuquerque:
University of New Mexico Press.
Forgey, Donald G.
1975 The Institutionof Berdache among the North American Plains Indians.The Journalof Sex Research
11:1-15.
Foucault, Michel
1978 The History of Sexuality, Volume I: An Introduction. Robert Hurley, trans. New York: Random
House.
Fox-Genovese, Elizabeth
1993 FromSeparate Spheres to Dangerous Streets:PostmodernistFeminism and the Problem of Order.
Social Research 60(2):235-254.
Goodenough, Ward H.
1972 Componential Analysis. In Culture and Cognition: Rules, Maps, and Plans. James P. Spradley, ed.
San Francisco:Chandler Publishing.
Greenberg, David
1988 The Construction of Homosexuality. Chicago: University of Chicago Press.
Haile, FatherBerard
1938 Origin Legend of the Navajo Enemyway. Yale University Publications in Anthropology, 17. New
Haven, CT:Yale University Press.
1981 Upward Moving and Emergence Way. American Tribal Religions, 7. KarlLuckert,ed. Lincoln:
Universityof Nebraska Press.
Halperin, David M.
1990 One Hundred Years of Homosexuality and Other Essayson Greek Love. New York:Routledge.
1993 Is There a History of Sexuality? In The Lesbian and Gay Studies Reader. Henry Abelove, Michele
Aina Barale, and David M. Halperin, eds. Pp. 416-431. New York:Routledge.
Herdt, Gilbert H.
1991 a Representationsof Homosexuality in Traditional Societies: An Essayon Cultural Ontology and
Historical Comparison, Part 1. Journalof the History of Sexuality 1(3):481-504.

288 american ethnologist

This content downloaded from 158.143.192.135 on Sat, 21 Sep 2013 15:25:20 PM

All use subject to JSTOR Terms and Conditions
 1991 b Representationsof Homosexuality in Traditional Societies: An Essayon Cultural Ontology and
Historical Comparison, Part2. Journalof the History of Sexuality 1(4):603-632.
1994 Third Sex, Third Gender: Beyond Sexual Dimorphism in Culture and History. New York:Zone
Books.
Herdt, Gilbert H., and Andrew Boxer
1992 Introduction:Culture, History, and Life Course of Gay Men. In Gay Culture in America: Essays
from the Field. Gilbert H. Herdt, ed. Pp. 1-28. Boston: Beacon Press.
Hill, W. W.
1935 The Status of the Hermaphrodite and Transvestite in Navajo Culture. American Anthropologist
37:273-279.
Jacobs, Sue-Ellen
1968 Berdache: A Brief Review of the Literature.Colorado Anthropology 1:25-40.
1983 Reply to the North American Berdache. CurrentAnthropology 24:459-460.
Jacobs, Sue-Ellen, and Jason Cromwell
1992 Visions and Revisions of Reality: Reflections on Sex, Sexuality, Gender, and Gender Variance.
Journalof Homosexuality 23(4):43-69.
Jacobs, Sue-Ellen, and Wesley Thomas
1994 Native American Two-Spirits. Anthropology Newsletter 35(8):7.
Jaimes, M. Annette
1992 American IndianWomen: At the Center of Indigenous Resistance in ContemporaryNorth America.
In The State of Native America: Genocide, Colonization, and Resistance. M. Annette Jaimes, ed. Pp.
311-344. Boston: South End Press.
Katz, Jonathan
1976 Gay American History: Lesbians and Gay Men in the USA. New York:Thomas Y. Crowell.
Kessler, Suzanne J., and Wendy McKenna
1978 Gender: An Ethnomethodological Approach. New York:Wiley-lnterscience.
Kluckhohn, Clyde
1968 The Philosophy of the Navaho Indians. In Readings in Anthropology. Morton H. Fried, ed. Pp.
674-699. New York:Crowell.
Knauft,Bruce
1994 Pushing Anthropology Pastthe Posts:Critical Notes on CulturalAnthropology and Cu tural Studies
as Influenced by Postmodernism and Existentialism.Critique of Anthropology 14(2):1 17-152.
Lang, Sabine
1993 Masculine Women, Feminine Men: Gender Variance and the Creation of Gay Identities among
ContemporaryNorth American Indians. Paper presented at the 92nd Annual Meeting of the American
Anthropological Association, Washington, DC, November.
Lyons, John
1977 Semantics, 1. New York:Cambridge University Press.
Marcus, George, and Michael J. Fisher
1986 Anthropology as Cultural Critique: An ExperimentalMoment in the Human Sciences. Chicago:
University of Chicago Press.
Martin, M. Kay, and BarbaraVoorhies
1975 Female of the Species. New York:Columbia University Press.
Mascia-Lees, Frances, Patricia Sharpe, and Colleen B. Cohen
1989 The PostmodernistTurn in Anthropology: Cautions from a FeministPerspective. Signs 1 5(1 ):7-33.
Matthews, Washington
1897 Navajo Legends. Memoirs of the American Folklore Society 5. New York:Houghton and Mifflin.
Money, John
1988 Gay, Straight,and In-Between: The Sexology of EroticOrientation. New York:Oxford University
Press.
Murray,Stephen
1994 Subordinating Native American Cosmologies to the Empire of Gender. Current Anthropology
35(1):59-61.
Navajo Community College
1992 Dine Philosophy of Learning.Tsaile, AZ: Navajo Community College.
O'Bryan, Aileen
1956 The Dine: Origin Myths of the Navajo Indians. Smithsonian Institution Bureau of American
Ethnology, 163. Washington, DC: U.S. Government PrintingOffice.
Padgug, Robert
1979 Sexual Matters:On Conceptualizing Sexuality in History. Radical History Review 20:3-23.
Parsons, Elsie Clews
1920 Notes on Ceremonialism at Laguna. Anthropological Papers of the American Museum of Natural
History, 19. New York:The Trustees.
Reichard, Gladys
1944 Prayer:The Compulsive Word. Monographs of the American Ethnological Society, 7. Seattle:
University of Washington Press.
1990[1950] Navajo Religion: A Study of Symbolism. Tucson: University of Arizona Press.

coming to terms with Navajo nadleehi 289

This content downloaded from 158.143.192.135 on Sat, 21 Sep 2013 15:25:20 PM

All use subject to JSTOR Terms and Conditions
Roscoe, Paul
1995 The Perils of "Positivism"in CulturalAnthropology. American Anthropologist 97(3):492-504.
Roscoe, Will
1984 Making History:The Challenge of Gay Studies. Unpublished MS.
1988 The Zuni Man-Woman. Outlook 2:56-67.
1991 The Zuni Man-Woman. Albuquerque: University of New Mexico Press.
Tafoya, Terry
1992 Native Gay and Lesbian Issues:The Two-Spirited. In Positively Gay: New Approaches to Gay and
Lesbian Life. Betty Berzon, ed. Pp. 253-259. Berkeley, CA: Celestial Arts.
Thomas, Wesley
1993 A Traditional Navajo's Perspective on the Navajo Cultural Construction of Gender. Paper
presented at 92nd Annual Meeting of the American Anthropological Association, Washington, DC,
November.
Turner, Edith
1996 Commentary: "A Delicate Science" Revisited. Anthropology and Humanism 21(2):187-188.
Weston, Kath
1993a Do Clothes Make the Woman?: Gender, Performance Theory, and Lesbian Eroticism.Genders
17:1-21
1993b Lesbian/Gay Studies in the House of Anthropology. Annual Reviews of Anthropology 22:
339-367.
1995 Guessing Games. In Dyke Life: A Celebration of the Lesbian Experience. Karla Jay, ed. Pp.
221-246. New York:Basic Books.
Whitehead, Harriet
1981 The Bow and the Burden Strap:A New Lookat InstitutionalizedHomosexuality in Native America.
In Sexual Meanings: The CulturalConstructionof Gender and Sexuality. Sherry B. Ortner and Harriet
Whitehead, eds. Pp. 80-115. New York:Cambridge University Press.
Williams, Walter L.
1986 The Spiritand the Flesh. Boston: Beacon Press.
Witherspoon, Gary
1974 The Central Concepts of Navajo Worldview (I). Linguistics 119:41-59.
1975 The Central Concepts of Navajo World View (II).Linguistics 161:69-88.
Wyman, Leland
1970 The Blessingway. Tucson: Universityof Arizona Press.

submitted September 25, 1995

revised version submitted July 16, 1996
revised version submitted February11, 1997
accepted March 13, 1997

290 american ethnologist

This content downloaded from 158.143.192.135 on Sat, 21 Sep 2013 15:25:20 PM

All use subject to JSTOR Terms and Conditions