Sexuality and Gender in Certain Native American Tribes: The Case of Cross-Gender Females

Author(s): Evelyn Blackwood

Source: Signs, Vol. 10, No. 1 (Autumn, 1984), pp. 27-42
Published by: The University of Chicago Press
Stable URL: http://www.jstor.org/stable/3174235 .
Accessed: 09/05/2014 06:39

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.

The University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend access to Signs.

http://www.jstor.org

This content downloaded from 161.116.148.143 on Fri, 9 May 2014 06:39:41 AM

All use subject to JSTOR Terms and Conditions
 Sexualityand Gender
in CertainNative AmericanTribes:
The Case of Cross-GenderFemales

Evelyn Blackwood

Ideological conceptsof gender and sexualityarise fromculturalconstruc-

tions and varyfromculture to culture.The female cross-genderrole in
certainNative American tribesconstitutedan opportunityforwomen to
assume the male role permanentlyand to marrywomen.' Its existence
challenges Western assumptionsabout gender roles. Some feministan-
thropologistsassume thatit is in the nature of sex and gender systemsto
create asymmetryin the formof male dominance and female subservi-
ence and to enforce correspondingformsof sexual behavior.2Because
kinship and marriage are closely tied to gender systems,these social
structuresare implicatedin the subordinationof women. The existence

I am particularlygratefulto Naomi Katz, Mina Caulfield,and CarolynClark fortheir

encouragement,support,and suggestionsduring the developmentof thisarticle.I would
also like to thankGilbertHerdt, Paula Gunn Allen, Sue-EllenJacobs,WalterWilliams,Luis
Kemnitzer,and Ruby Rohrlich for theirinsightfulcommentson an earlier version.
1. The term"berdache" is the more common termassociated withthe cross-gender
role. It was originallyapplied by Europeans to Native American men who assumed the
female role, and was derived fromthe Arabic bardaj,meaning a boy slave kept for sexual
purposes. I preferthe term"cross-gender,"firstused byJ. M. Carrier,particularlyforthe
female role. See J. M. Carrier, "Homosexual Behavior in Cross-CulturalPerspective,"in
HomosexualBehavior:A ModernReappraisal,ed. Judd Marmor (New York: Basic Books,
1980), pp. 100-122.
2. Sherry B. Ortner and Harriet Whitehead, eds., Sexual Meanings: The Cultural
Construction ofGenderand Sexuality(Cambridge: Cambridge UniversityPress, 1981); Gayle
Rubin, "The Trafficin Women: Notes on the 'Political Economy' of Sex," in Towardan
Anthropology of Women,ed. Rayna R. Reiter (New York: MonthlyReview Press, 1975), pp.
157-210.

[Signs:Journalof Womenin Cultureand Society1984, vol. 10, no. 1]

( 1984 by The Universityof Chicago. All rightsreserved. 0097-9740/85/1001-0001$01.00

27

This content downloaded from 161.116.148.143 on Fri, 9 May 2014 06:39:41 AM

All use subject to JSTOR Terms and Conditions
28 Blackwood Cross-Gender
Females

of the female cross-genderrole, however,points to the inadequacies of

such a view and helps to clarifythe nature of sex and gender systems.
This studycloselyexaminesthe femalecross-genderrole as itexisted
historicallyin severalNative Americantribes,primarilyin westernNorth
Americaand the Plains. It focuseson westerntribesthatshared a basically
egalitarian mode of production in precolonial times,3and for which
sufficientdata on the female role exist. Although there were cultural
differencesamong these groups, priorto the colonial period theyall had
subsistence-leveleconomies thathad not developed significantformsof
wealth or rank. These tribesinclude the Kaska of the Yukon Territory,
the Klamath of southernOregon, and the Mohave, Maricopa, and Co-
copa of the Colorado River area in the Southwest.The Plains tribes,by
contrast,are noteworthyfor the relative absence of the female cross-
gender role. Conditions affectingthe tribesof the Plains varied from
those of the westerntribes,and thus analysisof historical-culturalcon-
textswillserve to illuminatethe differingconstraintson sex and gender
systemsin these two areas.
Ethnographicliteraturehas perpetuatedsome misconceptionsabout
the cross-genderrole. Informantsfrequentlydescribe the institutionin
negative terms,statingthat berdache were despised and ridiculed. But
ethnographerscollected much of the data in thiscentury;it is based on
informants'memories of the mid- to late 1800s. During this period
the cross-genderinstitutionwas disappearing rapidly.Thus, twentieth-
centuryinformantsdo not accuratelyrepresent the institutionin the
precontact period. Alfred Kroeber found that "while the [berdache]
institutionwas in full bloom, the Caucasian attitudewas one of repug-
nance and condemnation.This attitude. . . made subsequentpersonality
inquirydifficult, thelaterberdache leading repressedor disguisedlives."4
Informants'statementsto later ethnographersor hostilewhiteofficials
were fardifferentfromthe actual attitudetowardthe role thatprevailed
in the precolonial period. An analysis of the cross-genderrole in its

3. Much feministdebate has focused on whethermale dominance is universal,or

whethersocietieswithegalitarianrelationsexist. For a more comprehensivediscussionof
egalitariansocieties,see Mina Davis Caulfield,"Equality,Sex and Mode of Production,"in
and Developmental
Social Inequality:Comparative Approaches,ed. Gerald D. Berreman (New
York: Academic Press, 1981), pp. 201-19; Mona Etienneand Eleanor Leacock, eds., Women
and Colonization:Anthropological (New York: J. F. Bergin, 1980); Eleanor Burke
Perspectives
Leacock, MythsofMale Dominance:CollectedArticleson WomenCross-Culturally (New York:
MonthlyReview Press, 1981); Karen Sacks, Sistersand Wives:ThePast and FutureofSexual
Inequality(Westport,Conn.: Greenwood Press, 1979); Rayna R. Reiter, ed., Towardan
Anthropology of Women(New York: Monthly Review Press, 1975); and Eleanor Burke
Leacock and Nancy 0. Lurie, eds., NorthAmerican IndiansinHistorical
Perspective (New York:
Random House, 1971).
4. AlfredL. Kroeber,"Psychosisor Social Sanction,"Charactetand Personality 8, no. 3
(1940): 204-15, quote on p. 209.

This content downloaded from 161.116.148.143 on Fri, 9 May 2014 06:39:41 AM

All use subject to JSTOR Terms and Conditions
Signs Autumn1984 29

proper historicalcontextbrings to lightthe integralnature of its rela-

tionshipto the larger community.

oftheFemaleCross-Gender
CulturalSignificance Role

Most anthropologicalworkon the cross-genderrole has focused on

the male berdache, with little recognitiongiven to the female cross-
gender role. Part of the problem has been the much smaller data base
available for a studyof the female role. Yet anthropologistshave over-
looked even the available data. This oversighthas led to the current
misconceptionthat the cross-genderrole was not feasible for women.
Harriet Whitehead, in a comprehensivearticleon the berdache, states
that, given the small number of cross-gender females, "the gender-
crossed status was more fullyinstitutedfor males than for females."5
Charles Callender and Lee Kochems, in a well-researchedarticle,base
theiranalysisof the role predominantlyon the male berdache.6Evidence
fromthirty-three Native Americantribesindicatesthatthe cross-gender
role forwomen was as viablean institution as was themale berdache role.7
The Native American cross-genderrole confounded Westerncon-
cepts of gender. Cross-gender individuals typicallyacted, sat, dressed,
talked like, and did the work of the other sex. Early Westernobservers
described the berdache as half male and half female,but such a descrip-
tion attestsonly to theirinabilityto accept a male in a female role or vice
versa. In the great majorityof reportedcases of berdache, theyassumed
the social role of the other sex, not of both sexes.8Contemporarytheo-
rists,such as Callender and Kochems and Whitehead,resistthe idea of a
complete social role reclassificationbecause they equate gender with
biological sex. Native gender categoriescontradictsuch definitions.

5. HarrietWhitehead,"The Bow and the Burden Strap: A New Look at Institutional-

ized Homosexualityin Native NorthAmerica,"in Ortnerand Whitehead,eds. (n. 2 above),
pp. 80-115, quote on p. 86.
6. Charles Callender and Lee M. Kochems,"The NorthAmericanBerdache," Current
Anthropology 24, no. 4 (1983): 443-56.
7. These tribesby area are as follows:Subarctic-Ingalik, Kaska; Northwest-Bella
Coola, Haisla, Lillooet, Nootka, Okanagon, Queets, Quinault; California/Oregon-Acho-
mawi, Atsugewi,Klamath, Shasta, Wintu,Wiyot,Yokuts, Yuki; Southwest-Apache, Co-
copa, Maricopa, Mohave, Navajo, Papago, Pima, Yuma; Great Basin-Ute, Southern Ute,
Shoshoni, Southern Paiute, NorthernPaiute; Plains-Blackfoot, Crow, Kutenai.
8. See S. C. Simms,"Crow Indian Hermaphrodites,"American 5, no. 3
Anthropologist
(1903): 580-81; Alfred L. Kroeber, "The Arapaho," AmericanMuseumof NaturalHistory
Bulletin18, no. 1 (1902): 1-150; Royal B. Hassrick,TheSioux:Lifeand Customsofa Warrior
Society(Norman: Universityof Oklahoma Press, 1964); Ronald L. Olson, The Quinault
Indians (Seattle: Universityof Washington Press, 1936); Ruth Murray Underhill, Social
Organizationof thePapago Indians (1939; reprint,New York: AMS Press, 1969).

This content downloaded from 161.116.148.143 on Fri, 9 May 2014 06:39:41 AM

All use subject to JSTOR Terms and Conditions
30 Blackwood Cross-Gender
Females

Althoughthedetailsof thecross-gendertemales'livesare scantin the

ethnographicliterature,a basic patternemerges from the data on the
westerntribes.Recognitionand culturalvalidationof the female cross-
gender role varied slightlyfromtribeto tribe,althoughthesocial role was
the same. Among the Southwesterntribes,dream experience was an
important ritual aspect of life and provided success, leadership, and
special skillsforthosewho soughtit.All cross-genderindividualsin these
tribesdreamed about theirrole change. The Mohave hwamedreamed of
becomingcross-genderwhilestillin the womb.9The Maricopa kwiraxame
dreamed too much as a child and so changed her sex.10No informationis
available forthe developmentof the femalecross-genderrole (tw!nnaek)
among the Klamath. It was most likelysimilar to the male adolescent
transformativeexperience, which was accomplished through fastingor
diving." Dreaming provided an avenue to special powers and also pro-
vided sanctionfortheuse of thosepowers.In thesame way,dreams about
the cross-gender role provided impetus and communitysanction for
assumption of the role.
The femalecandidate forcross-genderstatusdisplayedan interestin
the male role during childhood. A girl avoided learning female tasks.
Instead, as in the case of the Cocopa warrhameh, she played withboysand
made bows and arrows with which to hunt birds and rabbits.'2The
Mohave hwame"[threw]away theirdolls and metates,and [refused] to
shred bark or performother femininetasks."'3Adults, acknowledging
the interestsof such girls,taughtthem the same skillsthe boys learned.
Among the Kaska, a familythathad all femalechildrenand desired a son
to hunt forthemwould selecta daughter (probablythe one who showed
the most inclination)to be "like a man." When she was five,the parents
tied the dried ovaries of a bear to her belt to wear forlife as protection
againstconception.'4Though in differenttribesthe socializingprocesses
varied, girls achieved the cross-genderrole in each instance through
accepted cultural channels.
Upon reaching puberty,the timewhen girlswere considered ready
formarriage,thecross-genderfemalewas unable to fulfillher obligations

9. George Devereux, "InstitutionalizedHomosexualityof the Mohave Indians," Hu-

manBiology9, no. 4 (1937): 498-527.
10. Leslie Spier, YumanTribesoftheGila River(Chicago: Universityof Chicago Press,
1933).
Universityof CaliforniaPublicationsin Amer-
11. Leslie Spier, KlamathEthnography,
ican Archaeologyand Ethnology,vol. 30 (Berkeley: Universityof CaliforniaPress, 1930).
12. E. W. Gifford,The Cocopa, Universityof California Publications in American
Archaeologyand Ethnology,vol. 31, no. 5 (Berkeley: Universityof CaliforniaPress, 1933).
13. Devereux (n. 9 above), p. 503.
14. JohnJ. Honigmann, TheKaska Indians:An Ethnographic Yale Uni-
Reconstruction,
versityPublicationsin Anthropology,no. 51 (New Haven, Conn.: Yale UniversityPress,
1954), p. 130.

This content downloaded from 161.116.148.143 on Fri, 9 May 2014 06:39:41 AM

All use subject to JSTOR Terms and Conditions
Signs Autumn1984 31

and duties as a woman in marriage,havinglearned the tasksassigned to

men. Nonmarriageable statuscould have presenteda disadvantage both
to herselfand to her kin,who would be called upon to supporther in her
later years. But a role transferallowed her to enterthe marriagemarket
fora wifewithwhom she could establisha household. The Mohave pub-
liclyacknowledged the new statusof the woman by performingan initia-
tion ceremony.Followingthisceremonyshe assumed a name befittinga
person of the male sex and was given marriagerights.'5At puberty,the
Cocopa warrhameh dressed her hair in the male styleand had her nose
pierced like themen,insteadof receivinga chintattoolikeotherwomen.16
These public ritesvalidated the cross-genderidentity,signifyingto the
communitythat the woman was to be treated as a man.
In adult lifecross-genderfemalesperformedthe duties of the male
gender role. Their tasks included hunting,trapping,cultivatingcrops,
and fightingin battles. For example, the Cocopa warrhameh established
households like men and fought in battle.'7The Kaska cross-gender
female "dressed in masculine attire,did male allocated tasks,often de-
veloping great strengthand usuallybecoming an outstandinghunter."'8
The Mohave hwamewere knownas excellentproviders,huntingformeat,
workingin the fields,and caring forthe childrenof theirwives.'9 Cross-
gender females also adhered to male ritual obligations. A Klamath
tw!nnaekobserved the usual mourningwhen her long-timefemale part-
ner died, wearinga barkbeltas did a man.20Mohave hwameweresaid to be
powerful shamans, in this case especially good at curing venereal dis-
ease.2' Many other cross-genderfemaleswere considered powerfulspir-
itually,but mostwere not shamans,even in the Southwest.Cross-gender
femalesdid not bear childrenonce theytook up the male role. Their kin
considered them nonreproductiveand accepted the loss of theirchild-
bearing potential, placing a woman's individual interestsand abilities
above her value as a reproducer.22
In most cases ethnographersdo not discuss the abilityof cross-gen-
der femalesto maintainthe fictionof theirmaleness.Whiteheadsuggests
thatwomen were barred fromcrossingover unless theywere,or at least
pretended to be, deficientphysically.23 However, despite some reports
that cross-genderwomen in the Southwesthad muscular builds, unde-

15. Devereux (n. 9 above), pp. 508-9.

16. Gifford(n. 12 above).
17. Ibid., p. 294.
18. Honigmann (n. 14 above), p. 130.
19. Devereux (n. 9 above).
20. Spier, KlamathEthnography (n. 11 above), p. 53.
21. Devereux (n. 9 above).
22. Ibid.; Gifford(n. 12 above); Honigmann (n. 14 above).
23. Whitehead (n. 5 above), pp. 92-93.

This content downloaded from 161.116.148.143 on Fri, 9 May 2014 06:39:41 AM

All use subject to JSTOR Terms and Conditions
32 Blackwood Cross-Gender
Females

veloped secondary sexual characteristics,and sporadic or absent

menstruation,24 convincing physical evidence is noticeablylacking. In
fact, the Mohave hwamekept a husband's taboos with regard to her
menstruatingor pregnantwifeand ignored her own menses.25 That such
mayhave been the case in othertribesas well is borne out bythe practice
of the Ingalik cross-gender female. Among the Alaskan Ingalik, the
kashimwas thecenterof men's activitiesand the place formale-onlysweat
baths.The cross-genderfemaleparticipatedin theactivitiesof thekashim,
and the men were said not to perceive her true sex.26Cornelius Osgood
suggeststhat she was able to hide her sex, but, as withthe Mohave, the
people probably ignored her physicalsex in favor of her chosen role.
Through this social fiction,then, cross-genderfemales dismissed the
physiologicalfunctionsof women and claimed an identitybased on their
performanceof a social role.

GenderEquality

Women's abilityto assume the cross-genderrole arose from the

particularconditionsof kinshipand gender in these tribes.The egalitar-
ian relationsof the sexes were predicated on the cooperation of auton-
omous individualswho had controlof theirproductiveactivities.In these
tribeswomen owned and distributedthearticlestheyproduced, and they
had equal voice in matters affectingkin and community.Economic
strategiesdepended on collectiveactivity.Lineages or individualshad no
formalauthority;the whole group made decisionsby consensus. People
of both sexes could achieve positionsof leadershipthroughskill,wisdom,
and spiritualpower. Ultimately,neitherwomen nor men had an inferior
role but ratherhad powerin thosespheresof activity specificto theirsex.27

24. C. DaryllForde, Ethnography oftheYumaIndians,Universityof CaliforniaPublica-

tions in American Archaeology and Ethnology,vol. 28, no. 4 (Berkeley: Universityof
CaliforniaPress, 1931), p. 157; Gifford(n. 12 above), p. 294; Devereux (n. 9 above), p. 510.
25. Devereux (n. 9 above), p. 515.
26. Cornelius Osgood, IngalikSocial Culture,Yale UniversityPublicationsin Anthro-
pology, no. 53 (New Haven, Conn.: Yale UniversityPress, 1958).
27. Based on ethnographicdata in Honigmann (n. 14 above); Gifford(n. 12 above);
Leslie Spier, CulturalRelationsoftheGila and ColoradoRiverTribes,Yale UniversityPublica-
tions in Anthropology,no. 3 (New Haven, Conn.: Yale UniversityPress, 1936), Klamath
Ethnography (n. 11 above), and YumanTribes(n. 10 above); Theodore Stern,TheKlamathTribe
(Seattle: Universityof WashingtonPress, 1966); AlfredL. Kroeber,MohaveIndians:Report
onAboriginal Territoryand Occupancy oftheMohaveTribe,ed. David Horr (New York: Garland
Publishing,1974), and HandbookoftheIndiansofCalifornia,Bureau of AmericanEthnology
Bulletin no. 78 (Washington,D.C.: GovernmentPrintingOffice,1925); William H. Kelly,
CocopaEthnography, AnthropologicalPapers of the Universityof Arizona, no. 29 (Tucson:
Universityof Arizona Press, 1977); Lorraine M. Sherer, The Clan System oftheFortMohave
Indians (Los Angeles: HistoricalSocietyof Southern California, 1965).

This content downloaded from 161.116.148.143 on Fri, 9 May 2014 06:39:41 AM

All use subject to JSTOR Terms and Conditions
Signs Autumn1984 33

Among these tribes,gender roles involved the performanceof a

particular set of duties. Most occupations necessaryto the functioning
of the group were defined as eithermale or female tasks.A typicaldivi-
sion of labor allocated responsibilitiesfor gathering,food preparation,
child rearing,basket weaving,and makingclothesto women,while men
hunted, made weapons, and built canoes and houses. The allocation of
separate tasksto each sex establisheda systemof reciprocitythatassured
the interdependenceof the sexes. Because neithersetof taskswas valued
more highlythan the other, neithersex predominated.
Gender-assignedtasksoverlapped considerablyamong thesepeople.
Many individuals engaged in activitiesthatwere also performedby the
othersex withoutincurringdisfavor.The smallgame and fishthatKaska
and Klamath women hunted on a regular basis were an importantcon-
tributionto the survivalof theband. Some Klamathwomen made canoes,
usually a man's task, and older men helped women with food prep-
aration.28In the Colorado Riverarea, bothmen and womencollectedtule
pollen.29Engaging in such activitiesdid not make a woman masculineor a
man femininebecause, although distinctspheres of male and female
production existed,a wide range of taskswas acceptable forboth sexes.
Because there was no need to maintain gender inequalities,notions of
power and prestigedid not circumscribethe roles. Withoutstrictgender
definitions,itwas thenpossible forsome NativeAmericanwomen to take
up the male role permanentlywithoutthreateningthe gender system.
Anotherfactorin creatingthe possibilityof thecross-genderrole for
women was the nature of the kinshipsystem.Kinship was not based on
hierarchicalrelationsbetween men and women; it was organized in the
interestof both sexes. Each sex had somethingto gain byformingkinties
throughmarriage,30 because of the mutualassistanceand economic secu-
ritymarital relations provided.31Marriage also created an alliance be-
tween two families,therebybroadening the networkof kin on whom an
individualcould rely.Thus, marriagepromotedsecurityin a subsistence-
level economy.
The marriagecustomsof thesetribesreflectedtheegalitariannature
of their kinship system.Since status and propertywere unimportant,
marriage arrangementsdid not involve any transferof wealth or rank

28. Julie Cruikshank,AthapaskanWomen:Lives and Legends(Ottawa: National Mu-

seums of Canada, 1979); Spier, KlamathEthnography (n. 11 above).
29. Gifford(n. 12 above).
30. The fivetribesdiscussed here variedin formsof kinship,but thisvariationdid not
have a significanteffecton the relationsbetweenthe sexes. Lacking rankor wealth,kinship
groups were not the focus of power or authority,hence whethera tribewas matrilinealor
patrilinealwas not as importantas the overall relationshipwithkin on either side.
31. JohnJ. Honigmann,CultureandEthosofKaskaSociety, Yale UniversityPublications
in Anthropology,no. 40 (New Haven, Conn.: Yale UniversityPress,1949), and KaskaIndians
(n. 14 above).

This content downloaded from 161.116.148.143 on Fri, 9 May 2014 06:39:41 AM

All use subject to JSTOR Terms and Conditions
34 Blackwood Females
Cross-Gender

throughthe female.The smallmarriagegiftsthatwereexchanged served

as tokens of the woman's worth in the marriage relationship.32Fur-
thermore,because of the unimportanceof propertyor rank,individuals
oftenhad a seriesof marriages,ratherthanone permanentrelationship;
divorce was relativelyeasy and frequent for both women and men.33
Marriagesin thesetribesbecame more permanentonlywhencouples had
children. Women were not forced to remain in a marriage,and either
partner had the right to dissolve an unhappy or unproductive rela-
tionship.
This egalitariankinshipsystemhad importantramificationsfor the
cross-genderfemale.A daughter'smarriagewas not essentialformainte-
nance of familyrank; thatis, a woman's familydid not lose wealthif she
abandoned her role as daughter.As a social male, she had marriagerights
through which she could establish a household and contributeto the
subsistenceof the group. Additionally,because of the frequencyof di-
vorce,itwas possible fora marriedcross-genderfemaleto raise children.
Evidence of cross-gender females caring for their wives' offspringis
available only for the Mohave hwame.Women in other tribes,however,
could also have broughtchildreninto a cross-gendermarriage,since at
least younger offspringtypicallywent withthe motherin a divorce.34A
cross-genderwoman mightacquire childrenthroughmarriageto a preg-
nant woman, or possiblythrough her wife'sextramaritalrelationships
withmen. Cross-gendercouples probablyalso adopted children,a prac-
tice common among heterosexual couples in many tribes.
Details fromthe Mohave help to illuminatethe cross-genderparent/
child relationship.The Mohave believed thatthe paternityof an unborn
child changed ifthe pregnantwoman had sex withanotherpartner;thus,
the cross-genderfemale claimed any child her wife mightbe carrying
when theymarried. George Devereux statesthatsuch childrenretained
the clan affiliationof the previous father.35
But the clan structureof the
Mohave was not stronglyorganized and possessed no formalauthorityor
ceremonial functions.36The significantrelationships were those de-
veloped through residence with kin. Thus, children raised in a cross-
gender household establishedstrongtieswiththose parents.The invest-
ment of parental care was reciprocated when these children became
adults. In this way the cross-genderfemale remained a part of the net-
work of kin through marriage.

32. Spier, KlamathEthnography (n. 11 above); J. A. Teit, "Field Notes on the Tahltan
and Kaska Indians: 1912-15," Anthropologica 3, no. 1 (1956): 39-171; Kroeber,Handbook(n.
27 above); Gifford(n. 12 above).
33. Kelly (n. 27 above); Spier, KlamathEthnography (n. 11 above).
34. Kelly (n. 27 above).
35. Devereux (n. 9 above), p. 514.
36. Kelly (n. 27 above); Forde (n. 24 above).

This content downloaded from 161.116.148.143 on Fri, 9 May 2014 06:39:41 AM

All use subject to JSTOR Terms and Conditions
Signs Autumn1984 35

Sexual Relationsin theCross-Gender

Role

Sexual behavior was part of the relationshipbetween cross-gender

female and the women theymarried.Althoughthe cross-genderfemale
was a social male, NativeAmericansdid notconsiderher sexual activityan
imitationof heterosexual behavior. Her sexual behavior was recognized
as lesbian-that is,as femalehomosexuality.The Mohave were aware of a
range of sexual activitiesbetweenthe cross-genderfemale and her part-
ner-activities thatwere possibleonlybetweentwophysiologicalfemales.
Devereux recorded a Mohave termthatreferredspecificallyto thelesbian
love-making of the hwameand her partner.37The Native American
acceptance of lesbian behavior among cross-genderfemalesdid not de-
pend on the presence of a male role-playingperson; their acceptance
derived instead fromtheirconcept of sexuality.
Native Americanbeliefsabout sexualityare reflectedin themarriage
system.Theoristssuch as Gayle Rubin have implicatedmarriageas one of
the mechanismsthatenforceand definewomen'ssexuality.Accordingto
Rubin, the division of labor "can ... be seen as a taboo against sexual
arrangements other than those containing at least one man and one
woman,therebyenjoiningheterosexualmarriage."38 Yet in certainNative
American tribes other sexual behavior, both heterosexual and homo-
sexual, was available and permissiblewithinand outside of marriage.
Homosexual behavior occurred in contextswithinwhich neither indi-
vidual was cross-gender nor were such individuals seen as expressing
cross-gender behavior.39Premarital and extramaritalsexual relations
were also permissible.40Furthermore,through the cross-gender role,
women could marryone another. Sexualityclearlywas not restrictedby
the institutionof marriage.
Native American ideology disassociated sexual behavior fromcon-
cepts of male and female gender roles and was not concerned withthe
identityof the sexual partner. The status of the cross-genderfemale's
partneris tellingin thisrespect.She was alwaysa traditionalfemale; that
is, two cross-genderfemalesdid not marry.Thus, a woman could follow
the traditionalfemalegender role,yetmarryand make love withanother
woman withoutbeing stigmatizedbysuch behavior.Even thoughshe was
the partnerof a cross-genderfemale,she was notconsidered homosexual
or cross-gender.If the relationshipended in divorce,heterosexualmar-
riage was stillan option fortheexwife.The traditionalfemalegender role
did not restricther choice of marital/sexualpartners. Consequently,

37. Devereux (n. 9 above), pp. 514-15.

38. Rubin (n. 2 above), p. 178.
39. See Forde (n. 24 above), p. 157; Honigmann,Kaska Indians(n. 14 above), p. 127.
40. Spier,KlamathEthnography (n. 11 above), and YumanTribes(n. 10 above); Kroeber,
Handbook(n. 27 above).

This content downloaded from 161.116.148.143 on Fri, 9 May 2014 06:39:41 AM

All use subject to JSTOR Terms and Conditions
36 Blackwood Cross-Gender
Females

individualspossessed a gender identity,but not a correspondingsexual

identity,and thuswere allowed severalsexual options.Sexualityitselfwas
not embedded in Native American gender ideology.

Womenon thePlains

The conditionsthatsupported thedevelopmentand continuationof

the cross-genderrole among certainwesterntribeswere not replicated
among the Plains tribes.Evidence of cross-genderfemalesthereis scant
while reportsof male berdache are numerous. Whitehead suggeststhat
the absence of cross-genderfemales resulted fromthe weakness of the
cross-genderinstitutionfor women.41A more plausible explanation in-
volves the particular historicalconditions that differentiatethe Plains
tribes from the westerntribes.Yet it is preciselythese conditions that
make accurate interpretationof women's roles and the female cross-
gender role much more difficultfor the Plains tribes.
The Plains Indian cultureof nomadic buffalohuntingand frequent
warfaredid not develop until the late eighteenthand early nineteenth
centuriesas tribesmoved westin response to theexpansion and develop-
ment of colonial America. The new mode of life representedfor many
tribesa tremendousshiftfroman originallysettledand horticulturalor
huntingand gatheringlife-style. Withthe introductionof the horse and
gun, the growthof the fur trade, and pressure fromwestward-moving
white settlers,tribesfrom the east and north were displaced onto the
Plains in the late 1700s.2 As the importanceof hide trade with Euro-
Americansincreasedin theearly 1800s,italteredthemode of production
among Plains tribes. Increased wealth and authoritywere accessible
throughtrade and warfare.Individual males were able to achieve greater
dominance while women's social and economic autonomy declined.43
Withthe growingimportanceof hides fortrade,men who weresuccessful
huntersrequired additionalwivesto handle thetanning.Their increasing
loss of controlin thisproductivesphere downgraded woman's statusand
tied her to maritaldemands. Recent workon the Plains tribes,however,
indicatesthatthisprocesswas notconsistent;womenmaintaineda degree
of autonomyand power not previouslyacknowledged.44

41. Whitehead (n. 5 above), p. 86.

42. Gene Weltfish,"The Plains Indians: Their Continuityin History and Their
Indian Identity,"in Leacock and Lurie, eds. (n. 3 above).
43. Leacock and Lurie, eds. (n. 3 above); Alan Klein, "The Political-Economyof
Gender: A 19th Century Plains Indian Case Study,"in The HiddenHalf: Studiesof Plains
Indian Women,ed. Patricia Albers and Beatrice Medicine (Washington,D.C.: University
Press of America, 1983), pp. 143-73.
44. See Albers and Medicine, eds.

This content downloaded from 161.116.148.143 on Fri, 9 May 2014 06:39:41 AM

All use subject to JSTOR Terms and Conditions
Signs Autumn1984 37

Earlyethnographicdescriptionsof Plains Indian women were based

on a Western gender ideology that was contradictedby actual female
behavior. Although traditionalPlains culture valued quiet, productive,
nonpromiscuous women, thiswas only one side of the coin. There was
actuallya variabilityin femalerolesthatcan onlybe attributedto women's
continued autonomy.Beatrice Medicine providesan excellentdiscussion
of the various roles open to women among the Blackfootand Lakota.
Such roles included the "manly-heartedwoman," the "crazy woman"
(who was sexually promiscuous), the Sun Dance woman, and the chief
woman or favoritewife.45According to Ruth Landes, Lakota women
served in tribalgovernmentand were sometimesappointed marshallsto
handle problemsamong women. Most Plains tribeshad women warriors
who accompanied war partiesforlimitedpurposes on certainoccasions,
such as avenging the death of kin,and who received warriorhonors for
their deeds.46 As Medicine states, "These varied role categories...
suggest that the idealized behavior of women was not as rigidlydefined
and followed as has been supposed."47
The presence of a varietyof sociallyapproved rolesalso suggeststhat
these were normativepatternsof behavior for women thatneed not be
construedas "contrary"to theirgender role. Warriorwomen were not a
counterpart of the male berdache, nor were they considered cross-
gender.48Ethnographers' attributionsof masculinityto such behavior
seem to be a productof Westernbeliefsabout therigiddichotomizationof
gender roles and the nature of suitable pursuitsfor women. That men
simplyaccepted females as warriorsand were not threatened by such
behavior contradictsthe notion thatsuch women were even temporarily
assumingthe male role.49The men's acceptance was based on recognition
of the women warriors'capabilitiesas women.
There were individual Plains women in the nineteenth century
whose behavior throughouttheirlives exemplifieda cross-genderrole.
They did not always cross-dress,but, like Woman Chief of the Crow,
neitherdid theyparticipatein femaleactivities.They tookwivesto handle
their households and were highly successful in hunting and raiding
activities.They were also considered verypowerful.Of thesewomen,the
Kutenai cross-genderwoman always dressed in male attireand was re-

45. Beatrice Medicine, "'Warrior Women'-Sex Role AlternativesforPlains Indian

Women," in Albers and Medicine, eds., pp. 267-80; see also Oscar Lewis, "Manly-Hearted
Women among the North Piegan," AmericanAnthropologist 43, no. 2 (1941): 173-87.
46. Ruth Landes, The MysticLake Sioux (Madison: Universityof Wisconsin Press,
1968).
47. Medicine, p. 272.
48. Sue-Ellen Jacobs, "The Berdache," in CulturalDiversity and Homosexuality,ed.
Stephen Murray (New York: IrvingtonPress, in press); Medicine, p. 269.
49. On male acceptance of women warriors,see Landes.

This content downloaded from 161.116.148.143 on Fri, 9 May 2014 06:39:41 AM

All use subject to JSTOR Terms and Conditions
38 Blackwood Females
Cross-Gender

nowned for her exploits as warriorand mediator and guide for white
traders.Running Eagle of the Blackfootlived as a warriorand marrieda
young widow. Woman Chiefbecame the head of her father'slodge when
he died and achieved the thirdhighestrank among the Crow. She took
fourwives.50Particularlysince no recordsof earliercross-genderwomen
have been found,thesefewexamples seem to constituteindividualexcep-
tions. What then was the statusof the female cross-genderrole among
Plains tribes?
Part of the difficultywith answering this question stems from the
nature of the data itself.Nineteenth-century observersrarelyrecorded
informationon Plains Indian women,"consideringthemtoo insignificant
to meritspecial treatment."5'These observersknewfewwomen and only
the more successfulmales. "Those who did become knownwere women
who had acted as go-betweensforthe whitesand Indians,"52 such as the
Kutenai cross-genderfemale. Running Eagle and Woman Chief were
also exceptional enough to be noticed by white traders.Except for the
Kutenai woman, none of the women are identifiedas berdache in
nineteenth-century reports,although all were cross-gender.Observers
seem to have been unable to recognize the female cross-genderrole.
Indeed, no nineteenth-century reports mention cross-genderfemales
among even the western tribes, although later ethnographers found
ample evidence of the role.
Ethnographershad no solid evidenceof thefemalecross-genderrole
among Plains Indians. Several factorsmay help to explain this discrep-
ancy. WhitecontactwithPlains tribescame earlierthan withthe western
tribesand was more disruptive.The last cross-genderfemales seem to
have disappeared among Plains tribesby the mid-nineteenthcentury,
while in the Southwestthis did not occur until the end of the century,
much closer to the time when ethnographersbegan to collectdata. Dis-
crepancies also arise in informants'stories. The Kutenai denied the
existence of cross-gender females among them, in contradictionwith
earlier evidence, and yetwillinglyclaimed thatsuch women lived among
the Flathead and Blackfoot.33 The Arapaho told AlfredKroeber thatthe

50. Edwin Thompson Denig, Of theCrowNation,ed. John C. Ewers, Smithsonian

Institution,Bureau of American Ethnology,Bulletinno. 151, AnthropologyPapers no. 33
(Washington,D.C.: GovernmentPrintingOffice,1953), and Five Indian TribesoftheUpper
Missouri,ed. John C. Ewers (Norman: Universityof Oklahoma Press, 1961); Claude E.
Schaeffer,"The Kutenai Female Berdache: Courier, Guide, Prophetess, and Warrior,"
Ethnohistory12, no. 3 (1965): 193-236.
51. Patricia Albers, "Introduction: New Perspectiveson Plains Indian Women," in
Albers and Medicine, eds. (n. 43 above), pp. 1-26, quote on p. 3.
52. Katherine Weist, "Beasts of Burden and Menial Slaves: Nineteenth Century
Observations of Northern Plains Indian Women," in Albers and Medicine, eds. (n. 43
above), pp. 29-52, quote on p. 39.
53. Harry H. Turnev-High, Ethnography of theKutenai,Memoirs of the American
AnthropologicalAssociation,no. 56 (1941; reprint,New York: Kraus Reprint,1969), and

This content downloaded from 161.116.148.143 on Fri, 9 May 2014 06:39:41 AM

All use subject to JSTOR Terms and Conditions
Signs Autumn1984 39

Lakota had femaleberdache,but thereis no corroboratingevidence from

the Lakota themselves.54Informantswere clearlyreticentor unwillingto
discuss cross-genderwomen. In her articleon Native Americanlesbians,
Paula Gunn Allen suggeststhatsuch informationwas suppressed by the
elders of the tribes.55Most informationon Plains Indian women was
transmittedfromelder tribesmento whitemale ethnographers.But men
were excluded fromknowledge of much of women's behavior;56in this
way much of the data on cross-genderfemales may have been lost.
The record of Plains cross-genderfemalesremainslimited.Certain
social conditions may have contributedto the small number of women
who assumed the role in the nineteenthcentury.During the 1800s the
practice of taking additional wives increased with the men's need for
female labor. This phenomenon may have limited women's choice of
occupation. The pressuresto marrymayhave barred women froma role
thatrequired success in male tasksonly.The practiceof sororal polygyny
particularlywould have put subtle pressures on familiesto assure that
each daughter learned the traditionalfemale role. Indeed, there were
said to be no unmarriedwomen among the Lakota.57Furthermore,given
the constantstateof warfareand loss of able-bodied men, the tribeswere
under pressure merelyto survive.Such conditionsin the 1800s discour-
aged women fromabandoning theirreproductiveabilitiesthrough the
cross-genderrole. In fact,among the Lakota, women who insisted on
leading men's liveswere ostracizedfromthe group and forcedto wander
by themselves.58Knowledge of the female cross-genderrole may have
persisted,but those fewwho actuallylivedout the role were exceptionsin
a changing environment.

The DemiseoftheCross-Gender
Role

By the late nineteenthcenturythe female cross-genderrole had all

but disappeared among Native Americans.Its finaldemise was related to
a change in the constructionof sexualityand gender in these tribes.The
dominant ideology of Western culture, with its belief in the inferior

TheFlatheadIndiansofMontana,Memoirsof the AmericanAnthropologicalAssociation,no.

48 (1937; reprint,New York: Kraus Reprint, 1969).
54. Kroeber, "The Arapaho" (n. 8 above), p. 19.
55. Paula Gunn Allen, "Beloved Women: Lesbians in American Indian Cultures,"
Conditions:Seven 3, no. 1 (1981): 67-87.
56. Alice Kehoe, "The Shackles of Tradition," in Albers and Medicine, eds. (n. 43
above), pp. 53-73.
57. Hassrick (n. 8 above).
58. Jeannette Mirsky,"The Dakota," in Cooperation and Competition
amongPrimitive
Peoples,ed. Margaret Mead (Boston: Beacon Press, 1961), p. 417.

This content downloaded from 161.116.148.143 on Fri, 9 May 2014 06:39:41 AM

All use subject to JSTOR Terms and Conditions
40 Blackwood Cross-Gender
Females

nature of the female role and itsinsistenceon heterosexuality,began to

replace traditionalNative American gender systems.
Ideological pressuresof whitecultureencouraged Native American
peoples to reject the validityof the cross-genderrole and to invoke
notionsof "proper" sexualitythatsupported men's possession of sexual
rightsto women. Communitiesexpressed disapproval by berating the
cross-genderfemale for not being a "real man" and not being properly
equipped to satisfyher wife sexually. In effect,variationsin sexual be-
haviorthathad previouslybeen acceptable were now repudiated in favor
of heterosexualpractices.Furthermore,theidentityof thesexual partner
became an importantaspect of sexual behavior.
The lifeof the lastcross-genderfemaleamong the Mohave, Sahayk-
wisa, provides a clear example of this process. According to Devereux,
"Sahaykwisa... was born toward the middle of the last centuryand
killed ... at the age of 45. Sahaykwisahad at a certaintimea verypretty
wife.Other men desired the woman and triedto lure her away fromthe
hwame."The men teased Sahaykwisain a derogatorymanner,suggesting
thather love-makingwas unsatisfactory to her wifein comparisonto that
of a "real man." They ridiculed her wifeand said, "Why do you want a
transvestitefor your husband who has no penis and pokes you withthe
finger?"59 Such derision went beyond usual joking behavior untilfinally
Sahaykwisawas raped bya man who was angered because hiswifelefthim
for Sahaykwisa. The communityno longer validated the cross-gender
role, and Sahaykwisaherselfeventuallyabandoned it, only to be killed
later as a witch. By accusing the cross-genderfemale of sexual inade-
quacy, men of the tribe claimed in effectthat they had sole rightsto
women's sexuality,and thatsexualitywas appropriateonlybetweenmen
and women.

Conclusion

In attemptingto fit the Native American cross-gender role into

Westerncategories,anthropologistshave disregarded the waysin which
the institutionrepresentsnative categories of behavior. Western inter-
pretationsdichotomizethe gender roles foreach sex, whichresultsfrom
erroneous assumptionsabout, first,the connectionbetweenbiologyand
gender,and, second, the natureof gender roles. Callender and Kochems
state,"The transformationof a berdache was not a complete shiftfrom
his or her biologicalgender to the opposite one, but ratheran approxima-
tionof the latterin some of itssocial aspects."6bThey implythatanatomy

59. Devereux (n. 9 above), p. 523.

60. Callender and Kochems (n. 6 above), p. 453 (italicsmine).

This content downloaded from 161.116.148.143 on Fri, 9 May 2014 06:39:41 AM

All use subject to JSTOR Terms and Conditions
Signs Autumn1984 41

circumscribedthe berdache's abilityto functionin the gender role of the

other sex. Whitehead findsthe anatomical factorparticularlytellingfor
women, who were supposedly unable to succeed in the male role unless
deficientphysicallyas females.61These theorists,by claiming a mixed
gender status for the berdache, confuse a social role with a physical
identitythat remained unchanged for the cross-genderindividual.
Knowing the true sex of the berdache, Native Americansaccepted
them on the basis of their social attributes;physiologicalsex was not
relevantto the gender role. The Mohave, forexample, did not focus on
thebiologicalsex of theberdache. Nonberdache were said to "feeltoward
theirpossible transvestitemate as theywould feel towarda true woman,
[or] man."62In response to a newlyinitiatedberdache, the Yuma "began
to feel toward him as to a woman."63These tribesconcurredin the social
fictionof the cross-genderrole despite the obvious physicaldifferences,
indicatingthe unimportanceof biological sex to the gender role.64
Assumptionsregardingthe hierarchicalnature of Native American
gender relations have created serious problems in the analysis of the
female cross-genderrole. Whitehead claimsthatfewfemalescould have
been cross-genderbecause she assumes the asymmetricalnature of gen-
der relations.65In cultures with an egalitarian mode of production,
however,gender does not create an imbalance betweenthe sexes. In the
westernNorthAmericantribesdiscussed above, neithergender roles nor
sexualitywere associated withan ideology of male dominance. Women
were not barred fromthe cross-genderrole by rigid gender definitions;
instead,theyfilledthe role successfully.Althoughcross-genderroles are
not limitedto egalitariansocieties,the historicalconditionsof nonegali-
tarian societies,in which increasingrestrictionsare placed on women's
productive and reproductiveactivities,stronglydiscourage them from
takingon the cross-genderrole.
Anthropologists'classificationof gender roles as dichotomous has
served to obscure the nature of the Native American cross-genderrole.
For Whitehead, the male berdache is "less than a full man" but "more
than a mere woman,"66suggestinga mixed gender role combiningele-
mentsof boththemale and thefemale.Similarly,Callender and Kochems

61. Whitehead (n. 5 above), p. 92.

62. Devereux (n. 9 above), p. 501.
63. Forde (n. 24 above), p. 157.
64. Data on the Navajo nadle are not included in this article because the Navajo
conceptionof the berdache was atypical.The nadlewas considered a hermaphroditebythe
Navajo-i.e., of both sexes physically-and thereforedid not actuallyexemplifya cross-
gender role. See W. W. Hill, "The Statusof the Hermaphroditeand Transvestitein Navaho
Culture," AmericanAnthropologist 37, no. 2 (1935): 273-79.
65. Whitehead (n. 5 above), p. 86.
66. Ibid., p. 89.

This content downloaded from 161.116.148.143 on Fri, 9 May 2014 06:39:41 AM

All use subject to JSTOR Terms and Conditions
42 Blackwood Cross-Gender
Females

suggestthattheberdache formedan intermediategender status.67 Native

of
conceptualizations gender,particularly in the egalitariantribes,do not
contain an invariableopposition of two roles. The Westernideology of
feminineand masculine traitsactuallyhas littlein common with these
Native American gender systems,withinwhichexistlarge areas of over-
lapping tasks.
The idea of a mixed gender role is particularlygeared to the male
berdache and assumes the existenceof a limitedtraditionalfemale role.
Such a concept does not account forthe wide range of behaviorspossible
for both the male and female gender roles. By contrastthe termcross-
gender definesthe role as a setof behaviorstypifying the attributesof the
othersex, but not limitedto an exact duplication of either role. Attributes
of the male berdache thatare nottypicalof thefemalerole-for example,
certainritualactivities-do not indicatea mixed gender category.These
activitiesare specialized tasks thatarise fromthe spiritualpower of the
cross-genderindividual.
The term"cross-gender,"however,is not withoutitsproblems.Sue-
EllenJacobs suggeststhata person who frombirthor earlychildhood fills
thisvariantrole maynotbe "crossing"a genderboundary.She prefersthe
term"thirdgender" because, as among the Tewa, the berdache role may
not fiteithera male or femalegender categorybut is conceived insteadas
another gender.68Kay Martin and Barbara Voorheis also explore the
possibilityof more than two genders.69Certainlythe last word has not
been spoken about a role thathas confounded researchersforat leastone
hundred years. But it is imperativeto develop an analysis of variant
gender roles based on the historicalconditionsthatfaced particulartribes
since gender systemsvaryin differentculturesand change as modes of
production change.

Department ofAnthropology
San FranciscoStateUniversity

67. Callender and Kochems (n. 6 above), p. 454.

68. Sue-Ellen Jacobs, personal communication,1983, and "Comment on Callender
and Kochems," CurrentAnthropology 24, no. 4 (1983): 459-60.
69. M. Kay Martinand Barbara Voorheis,FemaleoftheSpecies(New York: Columbia
UniversityPress, 1975).

This content downloaded from 161.116.148.143 on Fri, 9 May 2014 06:39:41 AM

All use subject to JSTOR Terms and Conditions