Wesleyan University

Is There a History of Sexuality?

Author(s): David M. Halperin
Reviewed work(s):
Source: History and Theory, Vol. 28, No. 3 (Oct., 1989), pp. 257-274
Published by: Wiley for Wesleyan University
Stable URL: http://www.jstor.org/stable/2505179 .
Accessed: 02/01/2013 10:23
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 [email protected].

Wiley and Wesleyan University are collaborating with JSTOR to digitize, preserve and extend access to History
and Theory.

http://www.jstor.org

This content downloaded on Wed, 2 Jan 2013 10:23:00 AM

All use subject to JSTOR Terms and Conditions

IS THERE A HISTORY OF SEXUALITY?*

DAVID M. HALPERIN

Sexhas no history.1 It is a naturalfact, groundedin the functioningof the body,

and, as such, it lies outside of historyand culture.Sexuality,by contrast,does
not properlyreferto some aspector attributeof bodies. Unlike sex, sexuality
is a culturalproduction:it representsthe appropriationof the humanbody and
of its physiologicalcapacitiesby an ideologicaldiscourse.'Sexualityis not a somatic fact; it is a culturaleffect. Sexuality,then, does have a history-though
(as I shall argue)not a very long one.
To say that, of course,is not to statethe obvious-despite the tone of assurance with which I just said it-but to advance a controversial,suspiciously
fashionable,and, perhaps,a stronglycounter-intuitiveclaim. The plausibility
of sucha claimmightseemto reston nothingmoresubstantialthanthe prestige
of the brilliant,pioneering,but largelytheoreticalworkof the late FrenchphilosopherMichel Foucault.3Accordingto Foucault,sexualityis not a thing, a
naturalfact, a fixedand immovableelementin the eternalgrammarof human
subjectivity,but that "set of effectsproducedin bodies, behaviors,and social
relationsby a certaindeployment"of "acomplexpoliticaltechnology."4
"Sexuality,"Foucaultinsists in anotherpassage,

* Mostof the materialcontainedin thisarticleappears,in slightlydifferentform,in the titleessay

of my collection,One HundredYearsof Homosexualityand OtherEssays on GreekLove (New
York:Routledge,1989),or in the Editors'Introductionto BeforeSexuality:The Constructionof
EroticExperiencein theAncientGreekWorld,ed. DavidM. Halperin,John J. Winkler,and Froma
I. Zeitlin(Princeton,1990).
1. Or,if it does, that historyis a matterfor the evolutionarybiologist,not for the historian;see
LynnMargulisand Dorion Sagan, The Originsof Sex (New Haven, 1985).
2. I adaptthis formulationfroma passagein LouisAdrianMontrose,"'ShapingFantasies':Figurationsof Genderand Powerin ElizabethanCulture,"Representations
2 (1983),61-94 (passageon
p. 62), whichdescribesin turnthe conceptof the "sex/gendersystem"introducedby GayleRubin,
"TheTrafficin Women:Notes on the 'PoliticalEconomy'of Sex,"in Towardan Anthropologyof
Women,ed. RaynaR. Reiter(New York,1975), 157-210.
3. VolumesTwoandThreeof Foucault'sHistoryof Sexuality,publishedshortlybeforehis death,
departsignificantlyfromthe theoreticalorientationof his earlierworkin favorof a moreconcrete
interpretative
practice;see my remarksin "TwoViewsof GreekLove:HaraldPatzerand Michel
Foucault,"One HundredYearsof Homosexuality,62-71, esp. 64.
4. MichelFoucault,TheHistoryof Sexuality,VolumeI: An Introduction,transl.RobertHurley
(New York,1978),127. See Teresade Lauretis,Technologiesof Gender:Essayson Theory,Film,
andFiction(Bloomington,1987),1-30,esp.3, whoextendsFoucault'scritiqueof sexualityto gender.

This content downloaded on Wed, 2 Jan 2013 10:23:00 AM

All use subject to JSTOR Terms and Conditions

258

DAVID M. HALPERIN

must not be thoughtof as a kind of naturalgiven whichpowertries to hold in check,

or as an obscuredomainwhichknowledgetriesgraduallyto uncover.It is the namethat
can be givento a historicalconstruct[dispositifj:not a furtiverealitythat is difficultto
grasp,but a greatsurfacenetworkin whichthe stimulationof bodies,the intensification
of pleasures,the incitementto discourse,the formationof specialknowledges,thestrengthening of controlsand resistances,are linkedto one another,in accordancewith a few
majorstrategiesof knowledgeand power.'
Is Foucault right? I believe he is, but I also believe that more is requiredto establish the historicity of sexuality than the mere weight of Foucault's authority. To
be sure, a great deal of work, both conceptual and empirical, has already been
done to sustain Foucault's central insights and to carry forward the historicist
project that he did so much to advance. But much more needs to be accomplished if we are to fill in the outlines of the picture that Foucault had time only
to sketch - hastily and inadequately, as he was the first to admit' - and if we are
to demonstrate that sexuality is indeed, as he claimed, a uniquely modern
production.
The study of classical antiquity has a special role to play in this historical enterprise.The sheer interval of time separatingthe ancient from the modern world
spans cultural changes of such magnitude that the contrasts to which they give
rise cannot fail to strike anyone who is on the lookout for them. The student
of classical antiquity is inevitably confronted in the ancient record by a radically
unfamiliar set of values, behaviors, and social practices, by ways of organizing
and articulating experience that challenge modern notions about what life is like,
and that call into question the supposed universality of "human nature" as we
currently understand it. Not only does this historical distance permit us to view
ancient social and sexual conventions with particular sharpness; it also enables
us to bring more clearly into focus the ideological dimension -the purely conventional and arbitrary character-of our own social and sexual experiences.8

5. Foucault, The History of Sexuality, 105-106.

6. Of special relevance are: Robert A. Padgug, "Sexual Matters: On Conceptualizing Sexuality
in History,"Radical History Review 20 (1979), 3-23; George Chauncey, Jr., "From Sexual Inversion
to Homosexuality: Medicine and the Changing Conceptualization of Female Deviance,"in Homosexuality:Sacrilege,Vision,Politics, ed. Robert Boyers and George Steiner = Salmagundi58-59
(1982-1983), 114-146; Arnold I. Davidson, "Sex and the Emergence of Sexuality," Critical Inquiry
14 (1987-1988), 16-48. See also The Cultural Construction of Sexuality, ed. Pat Caplan (London,
1987);T. Dunbar Moodie, "Migrancyand Male Sexuality on the South African Gold Mines,"Journal
of Southern African Studies 14 (1987-1988), 228-256; George Chauncey, Jr., "Christian Brotherhood or Sexual Perversion? Homosexual Identities and the Construction of Sexual Boundaries in
the World War One Era," Journal of Social History 19 (1985-1986), 189-211.
7. E.g., Michel Foucault, The Use of Pleasure, The History of Sexuality,Volume Two,transl. Robert
Hurley (New York, 1985), 92, 253.
8. In applying the term "ideological" to sexual experience, I have been influenced by the formulation of Stuart Hall, "Culture,the Media, and the 'Ideological Effect,"' in Mass Communication and
Society, ed. James Curran, Michael Gurevitch, Janet Woolacott, et al. (London, 1977), 315-348,
esp. 330: "ideology as a social practice consists of the 'subject' positioning himself in the specific
complex, the objectivated field of discourses and codes which are available to him in language and
culture at a particular historical conjuncture" (quoted by Ken Tucker and Andrew Treno, "The Culture of Narcissism and the Critical Tradition:An InterpretativeEssay,"Berkeley Journal of Sociology

This content downloaded on Wed, 2 Jan 2013 10:23:00 AM

All use subject to JSTOR Terms and Conditions

IS THERE A HISTORY OF SEXUALITY?

259

One of the currently unquestioned assumptions about sexual experience which

the study of antiquity calls into question is the assumption that sexual behavior
reflects or expresses an individual's "sexuality."
Now that would seem to be a relatively harmless and unproblematic assumption to make, empty of all ideological content, but what exactly do we have in
mind when we make it? What, in particular, do we understand by our concept
of "sexuality"?I think we understand "sexuality"to refer to a positive, distinct,
and constitutive feature of the human personality, to the characterological seat
within the individualof sexual acts, desires,and pleasures-the determinatesource
from which all sexual expression proceeds. "Sexuality"in this sense is not a purely
descriptive term, a neutral representation of some objective state of affairs or
a simple recognition of some familiar facts about us; rather, it is a distinctive
way of constructing, organizing, and interpreting those "facts," and it performs
quite a lot of conceptual work.
First of all, sexuality defines itself as a separate,sexual domain within the larger
field of human psychophysical nature. Second, sexuality effects the conceptual
demarcation and isolation of that domain from other areas of personal and social life that have traditionally cut across it, such as carnality, venery,libertinism,
virility, passion, amorousness, eroticism, intimacy, love, affection, appetite, and
desire-to name but a few of the older claimants to territoriesmore recentlystaked
out by sexuality. Finally, sexuality generates sexual identity: it endows each of
us with an individual sexual nature, with a personal essence defined (at least in
part) in specifically sexual terms; it implies that human beings are individuated
at the level of their sexuality, that they differ from one another in their sexuality
and, indeed, belong to differenttypes or kinds of being by virtue of their sexuality.
These, at least, appear to me to be some of the significant ramifications of
"sexuality,"as it is currently conceptualized. I shall argue that the outlook it
representsis alien to the recorded experience of the ancients. Two themes, in particular, that seem intrinsic to the modern conceptualization of sexuality but that
hardly find an echo in ancient sources will provide the focus of my investigation:
the autonomy of sexuality as a separate sphere of existence (deeply implicated
in other areas of life, to be sure, but distinct from them and capable of acting
on them at least as much as it is acted on by them), and the function of sexuality
as a principle of individuation in human natures. In what follows, I shall take
up each theme in turn, attempting to document in this fashion the extent of the
divergence between ancient and modern varieties of sexual experience.
First, the autonomy of sexuality as a separate sphere of existence. The basic
point I should like to make has already been made for me by Robert Padgug
in a now-classic essay on conceptualizing sexuality in history. Padgug argues that
whatweconsider"sexuality"
was,in the pre-bourgeoisworld,a groupof actsand institutions not necessarilylinkedto one another,or, if they werelinked, combinedin ways
25 [1980], 341-355 [quotation on p. 351]); see also Hall's trenchant discussion of the constitutive
role of ideology in "Deviance, Politics, and the Media," in Deviance and Social Control, ed. Paul
Rock and Mary McIntosh, Explorations in Sociology 3 (London, 1974), 261-305.

This content downloaded on Wed, 2 Jan 2013 10:23:00 AM

All use subject to JSTOR Terms and Conditions

260

DAVID M. HALPERIN

verydifferentfrom our own. Intercourse,kinship,and the family,and gender,did not

formanythinglikea "field"of sexuality.Rather,eachgroupof sexualacts wasconnected
directlyor indirectly- that is, formedpartof- institutionsand thoughtpatternswhich
wetendto viewas political,economic,or socialin nature,andthe connectionscut across
our idea of sexualityas a thing, detachablefrom otherthings,and as a separatesphere
of privateexistence.9
The ancient evidence amply supports Padgug's claim. In classical Athens, for
example, sex did not express inward dispositions or inclinations so much as it
served to position social actors in the places assigned to them, by virtue of their
political standing, in the hierarchical structure of the Athenian polity. Let me
expand this formulation.
In classical Athens a relatively small group made up of the adult male citizens
held a virtual monopoly of social power and constituted a clearly defined elite
within the political and social life of the city-state. The predominant feature of
the social landscape of classical Athens was the great divide in status between
this superordinate group, composed of citizens, and a subordinate group, composed of women, children, foreigners, and slaves -all of whom lacked full civil
rights (though they were not all equally subordinate). Sexual relations not only
respected that divide but were strictly polarized in conformity with it.
Sex is portrayed in Athenian documents not as a mutual enterprise in which
two or more persons jointly engage but as an action performed by a social superior upon a social inferior. Consisting as it was held to do in an asymmetrical
gesture- the penetration of the body of one person by the body (and, specifically,
by the phallus)10of another- sex effectivelydivided and distributedits participants
into radically distinct and incommensurable categories ("penetrator" versus
"penetrated"),categories which in turn were wholly congruent with superordinate
and subordinatesocial categories. For sexual penetrationwas thematized as domination: the relation between the insertive and the receptive sexual partner was
taken to be the same kind of relation as that obtaining between social superior
and social inferior." Insertive and receptive sexual roles were therefore necessarily isomorphic with superordinateand subordinatesocial status; an adult, male
citizen of Athens could have legitimate sexual relations only with statutory minors
(his inferiors not in age but in social and political status): the proper targets of

9. Padgug, 16.
10. I say "phallus" rather than "penis" because (1) what qualifies as a phallus in this discursive
system does not always turn out to be a penis (see note 29, below) and (2) even when phallus and
penis have the same extension, or reference, they still do not have the same intension, or meaning:
"phallus" betokens not a specific item of the male anatomy simpliciter but that same item taken
under the description of a cultural signifier; (3) hence, the meaning of "phallus"is ultimately determined by its function in the larger sociosexual discourse; i.e., it is that which penetrates, that which
enables its possessor to play an "active" sexual role, and so forth: see Rubin, 190-192.
11. Foucault, The Use of Pleasure, 215, puts it very well: "sexual relations - always conceived in
terms of the model act of penetration, assuming a polarity that opposed activity and passivitywere seen as being of the same type as the relationship between a superior and a subordinate, an
individual who dominates and one who is dominated, one who commands and one who complies,
one who vanquishes and one who is vanquished."

This content downloaded on Wed, 2 Jan 2013 10:23:00 AM

All use subject to JSTOR Terms and Conditions

IS THERE A HISTORY OF SEXUALITY?

261

his sexualdesireincluded,specifically,women of any age, free males past the

age of pubertywho werenot yet old enoughto be citizens(I'llcall them"boys,"
for short), as well as foreignersand slavesof either sex.12
Moreover,the physicalact of sex betweena citizenand a statutoryminorwas
stylizedin such a way as to mirrorin the minutedetailsof its hierarchicalarrangementthe relationof structuredinequalitythat governedthe wider social
interactionof the two lovers.What an Atheniandid in bed was determinedby
the differentialin status that distinguishedhim or her from his or her sexual
partner;the (male)citizen'ssuperiorprestigeand authorityexpressedthemselves
in his sexualprecedence-in his powerto initiatea sexualact, his rightto obtain
pleasurefromit, andhis assumptionof an insertiveratherthana receptivesexual
role.Differentsocial actorshad differentsexualroles:to assimilateboth the superordinateand the subordinatememberof a sexual relationshipto the same
wouldhavebeenas bizarre,in Athenianeyes,as classifyinga burglar
"sexuality"
as an "activecriminal,"his victimas a "passivecriminal,"and the two of them
alikeas partnersin crime-it would havebeen to confuse what, in reality,were
supposedlyseparateand distinctidentities.13Each act of sex was no doubt an
expressionof real,personaldesireon the partof the sexualactorsinvolved,but
theirverydesireshad alreadybeen shapedby the sharedculturaldefinitionof
sexas an activitythatgenerallyoccurredonlybetweena citizenanda non-citizen,
betweena personinvestedwith full civil status and a statutoryminor.
The "sexuality"of the classicalAthenians,then, far from being independent
of "politics"(eachconstruedas an autonomoussphere)was constitutedby the
veryprincipleson which Athenianpublic life was organized.In fact, the correspondencesin classicalAthensbetweensexualnormsand socialpracticeswere

12. In orderto avoidmisunderstanding,

I shouldemphasizethatby callingall personsbelonging
to thesefourgroups"statutoryminors,"I do not wish eitherto suggestthat theyenjoyedthe same
statusas one anotheror to obscurethe manydifferences
in statusthatcouldobtainbetweenmembers
of a singlegroup- e.g., betweena wife and a courtesan- differencesthat may not havebeen perfectlyisomorphicwiththe legitimatemodesof theirsexualuse. Nonetheless,whatis strikingabout
Atheniansocial usage is the tendencyto collapsesuch distinctionsas did indeedobtainbetween
differentcategoriesof social subordinatesand to createa singleoppositionbetweenthem all, en
masse,andtheclassof adultmalecitizens:on thispoint,seeMarkGolden,"Pais,'Child'and'Slave,"'
LAntiquite' classique 54 (1985), 91-104, esp. 101 and 102, n. 38.

13. I haveborrowedthis analogyfrom Arno Schmitt,who uses it to conveywhat the modern

sexualcategorieswouldlook like froma traditionalIslamicperspective:see GianniDe Martinoand
Arno Schmitt,KleineSchriftenzu zwischenmdnnlicher
SexualitatundErotikin der muslimischen
Gesellschaft(Berlin,1985),19. Note that even the categoryof anatomicalsex, definedin such a
wayas to includebothmenandwomen,seemsto be absentfromGreekthoughtfor similarreasons:
thecomplementarity
of menandwomenas sexualpartnersimpliesto the Greeksa polarity,a difference in species,too extremeto be bridgedby a single sexualconceptequallyapplicableto each.
In Greekmedicalwritings,therefore,"thenotionof sex nevergets formalizedas a functionalidentity of male and female,but is expressedsolely throughthe representation
of asymmetryand of
complementarity
betweenmaleandfemale,indicatedconstantlybyabstractadjectives(to thely['the
feminine'],to arren['themasculine']),"
accordingto PaolaManuli,"Donnemascoline,femminesterili,
verginiperpetue:La ginecologiagrecatra Ippocratee Sorano,"in SilviaCampese,Paola Manuli,
and GiuliaSissa,Madremateria:Sociologiae biologiadelta donnagreca(Turin,1983),147-192,
esp. 151and 201n.

This content downloaded on Wed, 2 Jan 2013 10:23:00 AM

All use subject to JSTOR Terms and Conditions

262

DAVID M. HALPERIN

so strictthat an inquiryinto Athenian"sexuality"

per se wouldbe nonsensical:
such an inquirycould only obscurethe phenomenonit was intendedto elucidate, for it wouldconcealthe sole contextin whichthe sexualprotocolsof the
classicalAtheniansmakeanysense- namely,the structureof theAthenianpolity.
Thesocialarticulationof sexualdesirein classicalAthensthusfurnishesa telling
illustrationof the interdependencein cultureof social practicesand subjective
experiences.Indeed,the classicalGreekrecordstronglysupportsthe conclusion
drawn(from a quite differentbody of evidence)by the Frenchanthropologist
MauriceGodelier:"it is not sexualitywhich hauntssociety, but society which
hauntsthe body's sexuality.""4
Forthose inhabitantsof the ancientworldaboutwhomit is possibleto generalize,sexualitydid not hold the keyto the secretsof the humanpersonality.(In
fact, the very concept of and set of practicescenteringon "the human personality"- the physicaland social sciencesof the blank individual- belong to
a muchlaterera and bespeakthe modernsocial and economicconditionsthat
accompaniedtheirrise.)In the Hellenicworld,bycontrast,the measureof a free
malewasmostoften takenbyobservinghowhe faredwhentestedin publiccompetitionagainstotherfreemales,not by scrutinizinghis sexualconstitution.War
(and otheragonisticcontests),not love,servedto revealthe innerman, the stuff
a freeGreekmalewasmadeof. 15A strikinginstanceof this emphasison public
life as the primarylocusof significationcanbe foundin the workof Artemidorus,
a masterdream-interpreter
who lived and wrote in the second centuryof our
era and whose testimony,thereis good reasonto believe,accuratelyrepresents
the sexual normsof ancientMediterraneanculture.16 Artemidorussaw public
life, not eroticlife, as the principaltenorof dreams.Evensexualdreams,in Artemidorus'ssystem,are seldomreallyabout sex: rather,they are about the rise
and fall of the dreamer'spublic fortunes, the vicissitudesof his domestic
economy.17If a mandreamsof havingsexwithhis mother,for example,hisdream
signifiesto Artemidorusnothingin particularabout the dreamer'sown sexual
psychology,his fantasylife, or the historyof his relationswith his parents;it's
a verycommondream,and so it's a bit trickyto interpretprecisely,but basically
it's a lucky dream:it may signify-depending on the family'scircumstancesat

14. Maurice Godelier, "The Origins of Male Domination," New Left Review 127 (May-June, 1981),
3-17 (quotation on p. 17); cf. Maurice Godelier, "Le sexe comme fondement ultime de l'ordre social
et cosmique chez les Baruya de Nouvelle-Guinee. Mythe et reality,"in SexualitMet pouvoir, ed. Armando Verdiglione (Paris, 1976), 268-306, esp. 295-296.
15. I am indebted for this observation to Professor Peter M. Smith of the University of North
Carolina at Chapel Hill, who notes that Sappho and Plato are the chief exceptions to this generalrule.
16. See John J. Winkler, "Unnatural Acts: Erotic Protocols in Artemidoros' Dream Analysis,"
Constraints of Desire: The Anthropology of Sex and Gender in Ancient Greece (New York, 1989),
17-44, 221-224.
17. S. R. F. Price, "The Future of Dreams: From Freud to Artemidorus," Past and Present 113
(November, 1986), 3-37, abridged in Before Sexuality: The Construction of Erotic Experience in
the Ancient Greek World,ed. David M. Halperin, John J. Winkler, and Froma I. Zeitlin (Princeton,
1990), 365-387; see also Michel Foucault, The Care of the Self, The History of Sexuality, Volume
Three, transl. Robert Hurley (New York, 1986), 3-36, esp. 26-34.

This content downloaded on Wed, 2 Jan 2013 10:23:00 AM

All use subject to JSTOR Terms and Conditions

IS THERE A HISTORY OF SEXUALITY?

263

the time,the posturesof the partnersin the dream,andthe modeof penetrationthat the dreamerwill be successfulin politics("success"meaning,evidently,the
powerto screwone's country),that he will go into exile or returnfrom exile,
thathe willwinhislawsuit,obtaina richharvestfromhislands,or changeprofessions,amongmanyotherthings(1.79).Artemidorus'ssystemof dreaminterpretation resemblesthe indigenousdream-loreof certainAmazoniantribes who,
despitetheirquitedifferentsociosexualsystems,sharewith the ancientGreeks
a belief in the predictivevalueof dreams.Like Artemidorus,these Amazonian
peoplesreversewhat modernbourgeoisWesternerstake to be the naturalflow
of significationin dreams(from images of public and social eventsto private
and sexualmeanings):in both Kagwahivand Mehinakuculture,for example,
dreamingaboutthe femalegenitaliaportendsa wound(and so a man who has
such a dreamis especiallycarefulwhen he handlesaxes or other sharpinstruments the next day);dreamtwoundsdo not symbolizethe femalegenitalia.18
Boththeseancientand moderndream-interpreters,
then, areinnocentof "sexuality":whatis fundamentalto theirexperienceof sex is not anythingwe would
regardas essentiallysexual;"9
it is insteadsomethingessentiallyoutward,public,
and social. "Sexuality,"for culturesnot shapedby some very recentEuropean
and Americanbourgeoisdevelopments,is not a cause but an effect.The social
body precedesthe sexual body.
I now come to the secondof my two themes-namely, the individuatingfunction of sexuality,its role in generatingindividualsexualidentities.The connection betweenthe moderninterpretationof sexualityas an autonomousdomain
and the modernconstructionof individualsexualidentitieshas been well analyzed, once again, by RobertPadgug:
the most commonlyheldtwentieth-century
assumptionsabout sexualityimplythat it is
18. See Waud H. Kracke, "Dreaming in Kagwahiv: Dream Beliefs and Their Psychic Uses in an
Amazonian Indian Culture," The Psychoanalytic Study of Society 8 (1979), 119-171, esp. 130-132,
163 (on the predictive value of dreams) and 130-131, 142-145, 163-164, 168 (on the reversal of the
Freudiandirection of signification - which Kracketakes to be a culturally constituted defense mechanism and which he accordingly undervalues); Thomas Gregor, "'Far, Far Away My Shadow Wandered. . .': The Dream Symbolism and Dream Theories of the Mehinaku Indians of Brazil,"American Ethnologist 8 (1981), 709-720, esp. 712-713 (on predictive value) and 714 (on the reversal of
signification), largely recapitulated in Thomas Gregor, Anxious Pleasures: The Sexual Lives of an
Amazonian People (Chicago, 1985), 152-161, esp. 153. Foucault's comments on Artemidorus, in The
Care of the Self, 35-36, are relevant here: "The movement of analysis and the procedures of valuation do not go from the act to a domain such as sexuality or the flesh, a domain whose divine, civil,
or natural laws would delineate the permitted forms; they go from the subject as a sexual actor to
the other areas of life in which he pursues his [familial, social, and economic] activity. And it is
in the relationship between these different forms of activity that the principles of evaluation of a
sexual behavior are essentially, but not exclusively, situated."
19. Note that even the human genitals themselves do not necessarily figure as sexual signifiers
in all culturalor representationalcontexts: for example, Caroline WalkerBynum, "The Body of Christ
in the Later Middle Ages: A Reply to Leo Steinberg," Renaissance Quarterly 39 (1986), 399-439,
argues in considerable detail that there is "reason to think that medieval people saw Christ's penis
not primarily as a sexual organ but as the object of circumcision and therefore as the wounded,
bleeding flesh with which it was associated in painting and in text" (p. 407).

This content downloaded on Wed, 2 Jan 2013 10:23:00 AM

All use subject to JSTOR Terms and Conditions

264

DAVID M. HALPERIN

a separatecategoryof existence(like"theeconomy,"or "thestate,"othersupposedlyindependentspheresof reality),almostidenticalwiththe sphereof privatelife. Sucha view

necessitatesthe locationof sexualitywithinthe individualas a fixedessence,leadingto
a classicdivisionof individualandsocietyandto a varietyof psychologicaldeterminism,
and, often enough,to a full-blownbiologicaldeterminismas well. Thesein turninvolve
the enshrinementof contemporarysexualcategoriesas universal,static,and permanent,
suitablefor the analysisof all humanbeings and all societies.20
The study of ancient Mediterranean societies clearly exposes the defects in any
such essentialist conceptualization of sexuality. Because, as we have seen in the
case of classical Athens, erotic desires and sexual object-choices in antiquity were
generallynot determined by a typology of anatomical sexes (male versus female),
but ratherby the social articulation of power (superordinateversus subordinate),
the currentlyfashionable distinction between homosexuality and heterosexuality
(and, similarly, between "homosexuals" and "heterosexuals"as individual types)
had no meaning for the classical Athenians: there were not, so far as they knew,
two differentkinds of "sexuality,"two differently structured psychosexual states
or modes of affective orientation, but a single form of sexual experience which
all free adult males shared-making due allowance for variations in individual
tastes, as one might make for individual palates.21
Thus, in the Third Dithyramb by the classical poet Bacchylides, the Athenian
hero Theseus, voyaging to Crete among the seven youths and seven maidens destined for the Minotaur and defending one of the maidens from the advances
of the libidinous Cretan commander, warns him vehemently against molesting
any one of the Athenian youths (tin' ethe6n: 43)- that is, any girl or boy. Conversely,the antiquarian litterateurAthenaeus, writing six or seven hundred years
later, is amazed that Polycrates, the tyrant of Samos in the sixth century B.C.,
did not send for any boys or women along with the other luxury articles he imported to Samos for his personal use during his reign, "despite his passion for
relations with males" (12.540c-e).22Now both the notion that an act of heterosexual aggression in itself makes the aggressor suspect of homosexual tendencies and the mirror-opposite notion that a person with marked homosexual tendencies is bound to hanker after heterosexual contacts are nonsensical to us,
associating as we do sexual object-choice with a determinate kind of "sexuality,"
a fixed sexual nature, but it would be a monumental task indeed to enumerate
all the ancient documents in which the alternative "boy or woman" occurs with
20. Padgug, 8.
21. Paul Veyne, in "La famille et l'amour sous le Haut-Empire romain," Annales (E. S. C.) 33
(1978), 35-63, remarks (p. 50) that Seneca's Phaedra is the earliest text to associate homosexual inclinations with a distinct type of subjectivity. The question is more complex than that, however, and
a thorough exploration of it would requirescrutinizingmore closely the ancient figure of the kinaidos,
a now-defunct sexual life-form: for details, see Maud W. Gleason, "The Semiotics of Gender: Physiognomy and Self-Fashioning in the Second Century C.E.," in Before Sexuality, 389-415; John J.
Winkler, "Laying Down the Law: The Oversight of Men's Sexual Behavior in Classical Athens,"

Constraintsof Desire,45-70, 224-226.

22. See Padgug, 3, who mistakenly ascribes Athenaeus's comment to Alexis of Samos (Jacoby,

Fragmenteder griechischenHistoriker539, fr. 2).

This content downloaded on Wed, 2 Jan 2013 10:23:00 AM

All use subject to JSTOR Terms and Conditions

IS THERE A HISTORY OF SEXUALITY?

265

perfectnonchalancein an eroticcontext, as if the two werefunctionallyinterchangeable.23

A particularlystrikingtestimonyto the imaginableextentof maleindifference

fromHellenistic
to the sexof sexualobjectscan be foundin a marriage-contract
Egyptdatingto 92 B.C.24This not untypicaldocumentstipulatesthat "it shall
not be lawfulfor Philiscus[theprospectivehusband]to bringhomeanotherwife
in additionto Apolloniaor to havea concubineor boy-lover...."25 The possibilitythat one'shusbandmighttake it into his headat some point duringone's
marriageto set up anotherhouseholdwithhis boyfriendevidentlyfiguredamong
the variouspotentialdomesticdisastersthat a prudentfianceewouldbe sureto
anticipateand to indemnifyherselfagainst.A somewhatsimilarexpectationis
articulatedin an entirelydifferentcontextby Dio Chrysostom,a moralizingGreek
oratorfromthe latefirstcenturyA.D. In a speechdenouncingthe corruptmorals
of citylife,Dio assertsthatevenrespectable
womenareso easyto seducenowadays
thatmenwillsoon tireof themandwillturntheirattentionto boysinstead-just
as addictsprogressinexorablyfrom wine to harddrugs(7.150-152).According
to Dio, then, paederastyis not simplya secondbest;it is not "caused,"as many
modernhistoriansof the ancientMediterraneanappearto believe,by the supposedseclusionof women,bythe practice(it wasmorelikelyan ideal)of locking
themawayin the innerroomsof theirfathers'or husbands'housesand thereby
preventingthem from servingas sexualtargetsfor adultmen. In Dio's fantasy,
at least,paederastyspringsnot fromthe insufficientbut fromthe superabundant
supplyof sexuallyavailablewomen;the easierit is to havesex with women,on
his view,the less desirablesex with womenbecomes,and the more likely men
areto seek sexualpleasurewith boys. Scholarssometimesdescribethe cultural
formationunderlyingthis apparentrefusalby Greekmalesto discriminatecategoricallyamong sexualobjects on the basis of anatomicalsex as a bisexuality
of penetration26
or- evenmoreintriguingly- as a heterosexualityindifferentto

23. See K. J. Dover, Greek Homosexuality (London, 1978), 63-67, for an extensive, but admittedly partial, list; also, Robert Parker, Miasma: Pollution and Purification in Early Greek Religion
(Oxford, 1983), 94. For some Roman examples, see T. Wade Richardson, "Homosexuality in the
Satyricon," Classica et Mediaevalia 35 (1984), 105-127, esp. 111.
24. I wish to emphasize that I am not claiming that all Greek men must have felt such indifference:
on the contrary, plenty of ancient evidence testifies to the strength of individual preferences for a
sexual object of one sex rather than another (see note 42, below). But many ancient documents bear
witness to a certain constitutional reluctance on the part of the Greeks to predict, in any given instance, the sex of another man's beloved merely on the basis of that man's past sexual behavior or
previous pattern of sexual object-choice.
25. P. Tebtunis I 104, translated by A. S. Hunt and C. C. Edgar, in Women'sLife in Greece and
Rome, ed. Mary Lefkowitz and Maureen B. Fant (Baltimore, 1982), 59-60; another translation is
provided, along with a helpful discussion of the document and its typicality, by Sarah B. Pomeroy,
Women in Hellenistic Egypt from Alexander to Cleopatra (New York, 1984), 87-89.
26. "Une bisexuality de sabrage":Veyne, 50-55; see the critique by Ramsay MacMullen, "Roman
Attitudes to Greek Love," Historia 32 (1983), 484-502, esp. 491-497. Other scholars who describe
the ancient behavioral phenomenon as "bisexuality"include Luc Brisson, "BisexualitUet mediation
en Grece ancienne," Nouvelle revue de psychanalyse 7 (1973), 27-48; Alain Schnapp, "Une autre
image de l'homosexualit6 en Grece ancienne," Le Detbat 10 (1981), 107-117, esp. 116-117;Lawrence

This content downloaded on Wed, 2 Jan 2013 10:23:00 AM

All use subject to JSTOR Terms and Conditions

266

DAVID M. HALPERIN

its object,27but I think it would be advisablenot to speak of it as a sexuality

at all but to describeit, rather,as a more generalizedethos of penetrationand
a sociosexualdiscoursestructuredby the presenceor absenceof
domination,28
its centralterm:the phallus.29It maybe worthpausingnow to examineone text
in particularwhichclearlyindicateshow thoroughlyancientcultureswereable
to dispensewith the notion of sexual identity.
The documentin questionis the ninthchapterin the FourthBook of the De
A.D. Latintranslationand adaptationby
morbischronicis,a mid-fifth-century
the AfricanwriterCaeliusAurelianusof a now largelylost workon chronicdiseasesby the GreekphysicianSoranus,who practicedand taughtin Romeduring
the early part of the second century A.D. Caelius'swork is not much read
nowadays,andit is almostentirelyneglectedby modernhistoriansof "sexuality";30
its date is late, its text is corrupt,and, far from being a self-consciousliterary
artifact,it belongsto the despisedgenreof Romantechnicalwriting.But, deStone, "Sex in the West," The New Republic (July 8, 1985), 25-37, esp. 30-32 (with doubts). Contra,
Padgug, 13: "to speak, as is common, of the Greeks, as 'bisexual' is illegitimate as well, since that
merely adds a new, intermediate category, whereas it was precisely the categories themselves which
had no meaning in antiquity."
27. T. M. Robinson, [Review of Dover's Greek Homosexuality], Phoenix 35 (1981), 160-163, esp.
162:"the reason why a heterosexual majority might have looked with a tolerant eye on 'active' homosexual practice among the minority, and even in some measure within their own group [!], ... is
predictablya sexist one: to the heterosexualmajority, to whom (in a man's universe)the 'good' woman
is kata physin [i.e., naturally] passive, obedient, and submissive, the 'role' of the 'active' homosexual
will be tolerable precisely because his goings-on can, without too much difficulty, be equated with
the 'role' of the male heterosexual, i.e., to dominate and subdue; what the two have in common is
greater than what divides them." But this seems to me to beg the very question that the distinction
between heterosexuality and homosexuality is supposedly designed to solve.
28. An excellentanalysis of the contemporaryMediterraneanversionof this ethos has been provided
by David Gilmore, "Introduction: The Shame of Dishonor," in Honor and Shame and the Unity
of the Mediterranean, ed. Gilmore, Special Publication of the American Anthropological Association, 22 (Washington, D.C., 1987), 2-21, esp. 8-16.
29. By "phallus" I mean a culturally constructed signifier of social power: for the terminology,
see note 10, above. I call Greek sexual discourse phallic because (1) sexual contacts are polarized
around phallic action-i.e., they are defined by who has the phallus and by what is done with it;
(2) sexual pleasures other than phallic pleasures do not count in categorizing sexual contacts; (3)
in order for a contact to qualify as sexual, one-and no more than one-of the two partners is required to have a phallus (boys are treated in paederastic contexts as essentially un-phallused [see
Martial, 11.22;but cf. Palatine Anthology 12.3, 7, 197, 207, 216, 222, 242] and tend to be assimilated
to women; in the case of sex between women, one partner- the "tribad"- is assumed to possess
a phallus-equivalent [an over-developed clitoris] and to penetrate the other: sources for the ancient
conceptualization of the tribad- no complete modern study of this fascinating and long-lived fictional
type, which survived into the early decades of the twentieth century, is known to me -have been
assembled by Friedrich Karl Forberg, Manual of Classical Erotology, transl. Julian Smithson [Manchester, 1884; repr. New York, 1966], II, 108-167; Paul Brandt [pseud. "Hans Licht"], Sexual Life
in Ancient Greece, transl. J. H. Freese, ed. Lawrence H. Dawson [London, 1932], 316-328; Gaston
Vorberg, Glossarium eroticum [Hanau, 1965], 654-655; and WernerA. Krenkel, "Masturbation in
der Antike," WissenschaftlicheZeitschrift der Wilhelm-Pieck-UniversitatRostock 28 [19791,159-178,
esp. 171. For a recent discussion, see Judith P. Hallett, "Female Homoeroticism and the Denial of
Roman Reality in Latin Literature," Yale Journal of Criticism 3.1 [19891, forthcoming).
30. Exceptions include Vern L. Bullough, Homosexuality: A History (New York, 1979), 3-5, and
John Boswell, Christianity, Social Tolerance, and Homosexuality: Gay People in WesternEurope
from the Beginning of the Christian Era to the Fourteenth Century (Chicago, 1980), 53n., 75n.

This content downloaded on Wed, 2 Jan 2013 10:23:00 AM

All use subject to JSTOR Terms and Conditions

IS THEREA HISTORYOF SEXUALITY?

267

spite all these drawbacks, it repays close attention, and I have chosen to discuss
it here partly in order to show what can be learned about the ancient world from
works that lie outside the received canon of classical authors.
The topic of this passage is molles (malthakoi in Greek) - that is, "soft" or
unmasculine men, men who depart from the cultural norm of manliness insofar
as they actively desire to be subjected by other men to a "feminine"(that is, receptive) role in sexual intercourse.31Caelius begins with an implicit defense of his
own unimpeachable masculinity by noting how difficult it is to believe that such
people actually exist;32he then goes on to observe that the cause of their affliction is not natural(that is, organic)but is rathertheir own excessivedesire,which in a desperate and foredoomed attempt to satisfy itself- drives out their sense
of shame and forcibly converts parts of their bodies to sexual uses not intended
by nature. These men willingly adopt the dress, gait, and other characteristics
of women, therebyconfirming that they suffernot from a bodily disease but from
a mental (or moral) defect. After some furtherargumentsin support of that point,
Caelius draws an interesting comparison: "Forjust as the women called tribades
[in Greek], because they practice both kinds of sex, are more eager to have sexual
intercourse with women than with men and pursue women with an almost masculine jealousy ... so they too [i.e., the molles] are afflicted by a mental disease"
(132-133). The mental disease in question, which strikes both men and women
alike and seems to be defined as a perversion of sexual desire, would certainly
appear to be nothing other than homosexuality as it is often understood today.
Several considerations combine to prohibit that interpretation, however. First
of all, what Caelius treats as a pathological phenomenon is not the desire on
the part of either men or women for sexual contact with a person of the same
sex; quite the contrary:elsewhere,in discussing the treatment of satyriasis (a state
of abnormally elevated sexual desire accompanied by itching or tension in the
genitals), he issues the following advice to those who suffer from it (De morbis

acutis, 3.18.180-18
1).33
Do not admitvisitorsand particularlyyoungwomenand boys. Forthe attractivenessof
suchvisitorswouldagainkindlethe feelingof desirein the patient.Indeed,even healthy
persons,seeingthem, would in manycases seek sexualgratification,stimulatedby the
tensionproducedin the parts [i.e., in their own genitals].34
There is nothing medically problematical, then, about a desire on the part of
males to obtain sexual pleasure from contact with males - so long as the proper
phallocentric protocols are observed; what is of concern to Caelius,35as well as
31. For an earlieruse of mollis in this almost technicalsense, see Juvenal,9.38.
32. See P. H. Schrijvers, Eine medizinische Erkldrung der mannlichen Homosexualitat aus der
Antike (Caelius Aurelianus DE MORBIS CHRONICIS IV 9) (Amsterdam, 1985), 11.

33. 1haveborrowedthis entireargumentfromSchrijvers,7-8; the samepoint aboutthe passage

fromDe morbisacutishadbeenmadeearlier- unbeknownstto Schrijvers,apparently- by Boswell,
Christianity, Social Tolerance, and Homosexuality, 53, n. 33; 75, n. 67.

34. Translation(withmy emphasisand amplification)by 1. E. Drabkin,ed. and transl.,Caelius

Aurelianus: ON ACUTE DISEASES and ON CHRONIC DISEASES (Chicago, 1950), 413.

35. As his chaptertitle, "De mollibussive subactis,"implies.

This content downloaded on Wed, 2 Jan 2013 10:23:00 AM

All use subject to JSTOR Terms and Conditions

268

DAVID M. HALPERIN

to other ancient moralists,36is the male desire to be sexually penetrated by males,

for such a desirerepresentsa voluntaryabandonment of the culturallyconstructed
masculine identity in favor of the culturally constructed feminine one. It is sexrole reversal, or gender-deviance, that is problematized here and that also furnishes part of the basis for Caelius's comparison of unmasculine men to masculine women, who assume a supposedly masculine role in their relations with other
women and actively "pursue women with an almost masculine jealousy."
Moreover,the ground of the similitude between these male and female genderdeviants is not that they are both homosexual but rather that they are both bisexual (in our terms), although in that respect at least they do not depart from
the ancient sexual norm. The tribads "are [relatively] more eager to have sexual
intercourse with women than with men" and "practice both kinds of sex"-that
is, they have sex with both men and women.37As for the molles, Caelius's earlier
remarks about their extraordinarily intense sexual desire implies that they turn
to receptivesex because, although they try, they are not able to satisfy themselves
by means of more conventionally masculine sorts of sexual activity, including
insertive sex with women.38 Far from having desires that are structured differently from those of normal folk, these gender-deviants desire sexual pleasure
just as most people do, but they have such strong and intense desires that they
are driven to devise some unusual and disreputable (though ultimately futile)
ways of gratifying them. This diagnosis becomes explicit at the conclusion of
the chapter when Caelius explains why the disease responsible for turning men
into molles is the only chronic disease that becomes stronger as the body grows
older.
For in other yearswhen the body is still strongand can performthe normalfunctions
of love, the sexualdesire[of these persons]assumesa dual aspect, in whichthe soul is
excitedsometimeswhile playinga passiveand sometimeswhile playingan active role.
Butin the caseof old menwho havelost theirvirilepowers,all theirsexualdesireis turned
in the opposite directionand consequentlyexertsa strongerdemandfor the feminine
role in love. In fact, manyinfer that this is the reasonwhy boys too are victimsof this
affliction.For,like old men, they do not possessvirilepowers;that is, they havenot yet
attainedthose powerswhich have alreadydesertedthe aged [137].39

36. See especially the pseudo-Aristotelian Problemata 4.26, well discussed by Dover, 168-170, and
by Winkler, "Laying Down the Law,"67-69; generally, Boswell, Christianity, Social Tolerance,and

Homosexuality,53; Foucault,The Use of Pleasure,204-214.

37. The Latin phrase quod utranque Veneremexerceant is so interpreted by both Drabkin, 901n.,
and Schrijvers,32-33, who securesthis readingby citing Ovid, Metamorphoses 3.323, where Teiresias,
who had been both a man and a woman, is described as being learned in the field of Venusutraque.
Compare Petronius, Satyricon 43.8: omnis minervae homo.
38. I follow, once again, the insightful commentary by Schrijvers, 15.
39. I quote from the translation by Drabkin, 905, which is based on his plausible, but nonetheless
speculative, reconstruction (accepted by Schrijvers, 50) of a desperately corrupt text. For the notion
expressed in it, compare Marcel Proust, A la recherche du tempsperdu, ed. Pierre Clarac and Andre
Ferr6(Paris, 1954), III, 204, 212; Remembrance of Things Past, transl. C. K. Scott Moncrieff and
TerenceKilmartin (New York, 1981), III, 203, 209; discussion by Eve Kosof sky Sedgwick, "Epistemology of the Closet (II)," Raritan 8 (Summer, 1988), 102-130.

This content downloaded on Wed, 2 Jan 2013 10:23:00 AM

All use subject to JSTOR Terms and Conditions

IS THERE A HISTORY OF SEXUALITY?

269

"Soft"or unmasculinemen, far from beinga fixedand determinatesexualspecies witha specificallysexualidentity,areevidentlyeithermen who once experiencedan orthodoxlymasculinesexualdesirein the past or who will eventually
experiencesuch a desirein the future.They may well be men with a constitutional tendencyto gender-deviance,accordingto Caelius, but they are not
homosexuals:beinga womanishman, or a mannishwoman,afterall, is not the
all the otherancienttextsknown
samethingas beinga homosexual.Furthermore,
to me, which assimilateboth males who enjoy sexual contact with males and
femaleswho enjoysexualcontactwith femalesto the same category,do so -in
conformitywiththe two taxonomicstrategiesemployedby CaeliusAurelianuseitherbecausesuch males and femalesboth reversetheir propersex-rolesand
adopt the sexualstyles, postures,and modes of copulationconventionallyassociatedwith the opposite gender,or becausethey both alternatebetweenthe
and sexualpracticesproper,respectively,to men and to
personalcharacteristics
women.40

Caelius'stestimonymakesan importanthistoricalpoint. Beforethe scientific

constructionof "sexuality"as a positive,distinct,and constitutivefeatureof individualhumanbeings- an autonomoussystemwithinthe physiologicalandpsychologicaleconomyof the humanorganism- certainkindsof sexualacts could
be individuallyevaluatedand categorized,and so could certainsexualtastesor
inclinations,but therewas no conceptualapparatusavailablefor identifyinga
person'sfixedand determinatesexualorientation,much less for assessingand
classifyingit.41That humanbeingsdiffer,often markedly,from one anotherin
theirsexualtastesin a greatvarietyof ways(includingsexualobject-choice),is
an unexceptionableand, indeed,an ancientobservation42:
Plato'sAristophanes

40. Anon., De physiognomonia 85 (vol. ii, p. 114.5-14 Fdrster); Vettius Valens, 2.16 (p. 76.3-8
Kroll); Clement of Alexandria, Paedagogus 3.21.3; Firmicus Maternus, Mathesis 6.30.15-16 and
7.25.3-23 (esp. 7.25.5).
41. See Foucault, The History of Sexuality, 43: "As defined by the ancient civil or canonical codes,
sodomy was a category of forbidden acts; their perpetrator was nothing more than the juridical
subject of them. The nineteenth-century homosexual became a personage, a past, a case history,
and a childhood, in addition to being a type of life, a life form, and a morphology, with an indiscreet
anatomy and possibly a mysterious physiology. Nothing that went into his total composition was
unaffected by his sexuality. It was everywhere present in him: at the root of all his actions because
it was their insidious and indefinitely active principle; written immodestly on his face and body because it was a secret that always gave itself away. It was consubstantial with him, less as a habitual
sin than as a singular nature." See also Randolph Trumbach, "London's Sodomites: Homosexual
Behavior and Western Culture in the 18th Century,"Journal of Social History 11 (1977), 1-33, esp.
9; Richard Sennett, The Fall of Public Man (New York, 1977), 6-8; Padgug, 13-14; Jean-Claude
F1ray,"Une histoire critique du mot homosexuality, [IV]," Arcadie 28, no. 328 (1981), 246-258, esp.
246-247; Schnapp (note 26, above), 116(speaking of Attic vase-paintings): "One does not paint acts
that characterize persons so much as behaviors that distinguish groups"; Pierre J. Payer, Sex and
the Penitentials: The Development of a Sexual Code 550-1150 (Toronto, 1984), 40-44, esp. 40-41:
"thereis no word in general usage in the penitentials for homosexuality as a category.... Furthermore, the distinction between homosexual acts and people who might be called homosexuals does
not seem to be operativein these manuals"(also, pp. 14-15, 140-153);Bynum, "TheBody of Christ,"406.
42. For attestations to the strength of individual preferences (even to the point of exclusivity) on
the part of Greek males for a sexual partner of one sex ratherthan another, see Theognis, 1367-1368;

This content downloaded on Wed, 2 Jan 2013 10:23:00 AM

All use subject to JSTOR Terms and Conditions

270

DAVID M. HALPERIN

invents a myth to explain why some men like women, why some men like boys,
why some women like men, and why some women like women (Symposium 189c193d). But it is not immediately evident that patterns of sexual object-choice
are by their very nature more revealing about the temperament of individual
human beings, more significant determinants of personal identity, than, for example, patterns of dietary object-choice.43And yet, it would never occur to us
to refer a person's dietary preference to some innate, characterological disposition,44to see in his or her strongly expressed and even unvarying preference for
the white meat of chicken the symptom of a profound psychophysical orientation, leading us to identify him or her in contexts quite removed from that of
the eating of food as, say, a "pectoriphage" or a "stethovore";nor would we be
likely to inquire further, making nicer discriminations according to whether an
individual's predilection for chicken breasts expressed itself in a tendency to eat
them quickly or slowly, seldom or often, alone or in company, under normal
circumstancesor only in periods of great stress, with a clear or a guilty conscience
("ego-dystonic pectoriphagia"), beginning in earliest childhood or originating
with a gastronomic trauma suffered in adolescence. If such questions did occur
to us, moreover, I very much doubt whether we would turn to the academic disciplines of anatomy, neurology, clinical psychology, genetics, or sociobiology in
the hope of obtaining a clear causal solution to them. That is because (1) we
Euripides, Cyclops 583-584; Xenophon, Anabasis 7.4.7-8; Aeschines, 1.41, 195; the Life of Zeno
by Antigonus of Carystus,cited by Athenaeus, 13.563e;the fragment of Seleucus quoted by Athenaeus,
15.697de (= Collectanea Alexandrina, ed. J. U. Powell [Oxford, 19251, 176); an anonymous dramatic fragment cited by Plutarch, Moralia 766f-767a (= Tragicorum Graecorum Fragmenta, ed.
August Nauck, 2d ed. [Leipzig, 1926], 906, #355; also in Theodor Kock, Comicorum Atticorum
Fragmenta [Leipzig, 1880-1888], III, 467, #360); Athenaeus, 12.540e, 13.601eand ff.; Achilles Tatius,
2.35.2-3; pseudo-Lucian, Er6tes 9-10; Firmicus Maternus, Mathesis 7.15.1-2; and a number of epigrams by various hands contained in the Palatine Anthology: 5.19, 65, 116, 208, 277, 278; 11.216;
12.7, 17, 41, 87, 145, 192, 198, and passim (cf. P. G. Maxwell-Stuart, "Strato and the Musa Puerilis,"
Hermes 100 [1972], 215-240). See, generally, Dover, 62-63; John Boswell, "Revolutions, Universals
and Sexual Categories," in Homosexuality: Sacrilege, Vision, Politics (note 6, above), 89-113, esp.
98-101; Winkler, "Laying Down the Law"; and, for a list of passages, Claude Courouve, Tableau
synoptique de references a l'amourmasculin: Auteurs grecs et latins (Paris, 1986).
43. Hilary Putnam, in Reason, TruthandHistory (Cambridge, Eng., 1981), 150-155, in the course
of analyzing the various criteria by which we judge matters of taste to be "subjective,"implies that
we are right to consider sexual preferences more thoroughly constitutive of the human personality
than dietary preferences, but his argument remains circumscribed, as Putnam himself points out,
by highly culture-specific assumptions about sex, food, and personhood.
44. Foucault, The Use of Pleasure, 51-52, remarks that it would be interesting to determine exactly when in the evolving course of Western cultural history sex became more morally problematic
than eating; he seems to think that sex won out only at the turn of the eighteenth century, after
a long period of relative equilibrium during the middle ages: see also The Use of Pleasure, 10; The
Care of the Self, 143; "On the Genealogy of Ethics: An Overview of Work in Progress," in Hubert
L. Dreyfus and Paul Rabinow, Michel Foucault: Beyond Structuralism and Hermeneutics, 2d ed.
(Chicago, 1983), 229-252, esp. 229. The evidence lately assembled by Stephen Nissenbaum, Sex,
Diet, and Debility in Jacksonian America: Sylvester Graham and Health Reform, Contributions
in Medical History, 4 (Westport, Conn., 1980), and by Caroline Walker Bynum, Holy Feast and
Holy Fast: The Religious Significance of Food to Medieval Women (Berkeley, 1987), suggests that
moral evolution may not have been quite such a continuously linear affair as Foucault appears to
imagine.

This content downloaded on Wed, 2 Jan 2013 10:23:00 AM

All use subject to JSTOR Terms and Conditions

IS THERE A HISTORY OF SEXUALITY?

271

regard the liking for certain foods as a matter of taste; (2) we currently lack a
theory of taste; and (3) in the absence of a theory we do not normally subject
our behavior to intense, scientific or aetiological, scrutiny.
In the same way, it never occurred to the ancients to ascribe a person's sexual
tastes to some positive, structural, or constitutive sexual feature of his or her
personality. Just as we tend to assume that human beings are not individuated
at the level of dietary preference and that we all, despite many pronounced and
frankly acknowledged differences from one another in dietary habits, share the
same fundamental set of alimentary appetites, and hence the same "dieticity"
or "edility,"so most premodern and non-Westerncultures, despite an awareness
of the range of possible variations in human sexual behavior, refuse to individuate human beings at the level of sexual preference and assume, instead, that
we all share the same fundamental set of sexual appetites, the same "sexuality."
For most of the world's inhabitants, in other words, "sexuality" is no more a
"fact of life" than "dieticity."Far from being a necessary or intrinsic constituent
of human life, "sexuality"seems indeed to be a uniquely modern, Western, even
bourgeois production -one of those cultural fictions which in every society give
human beings access to themselves as meaningful actors in their world, and which
are thereby objectivated.
If there is a lesson that we should draw from this picture of ancient sexual
attitudes and behaviors, it is that we need to de-center sexuality from the focus
of the cultural interpretation of sexual experience -and not only ancient varieties of sexual experience. Just because modern bourgeois Westernersare so obsessed with sexuality, so convinced that it holds the key to the hermeneutics of
the self (and hence to social psychology as an object of historical study), we ought
not therefore to conclude that everyone has always considered sexuality a basic
and irreducibleelement in, or a central feature of, human life. Indeed, there are
even sectors of our own societies to which the ideology of "sexuality"has failed
to penetrate.A sociosexual system that coincides with the Greek system, insofar
as it features a rigid hierarchy of sexual roles based on a set of socially articulated power-relations, has been documented in contemporary America by Jack
Abbott, in one of his infamous letters written to Norman Mailer from a federal
penitentiary; because the text is now quite inaccessible (it was not reprinted in
Abbott's book), and stunningly apropos, I have decided to quote it here at length.
It reallywas years,manyyears,before I beganto actuallyrealizethat the womenin
mylife-the prostitutesas wellas thesoft, prettygirlswhogiggledandteasedmeso much,
my severalwivesand those of my friends-it was yearsbeforeI realizedthat they were
not women,but men;yearsbeforeI assimilatedthe notionthat this wasunnatural.I still
only knowthis intellectually,for the most part-but for the small part that remainsto
my ken, I knowit is like a hammerblow to my templeand the shameI feel is profound.
Not becauseof the thing itself, the sexuallove I haveenjoyedwith thesewomen(some
so devoted it aches to recall it), but because of shame - and anger - that the world could

so intimatelybetrayme; so profoundlytouch and moveme-and then laughat me and

accusemy soul of a sickness,when that sicknesshas rescuedme from mentalderangementand despairsso blackas to cast this night that surroundsus in prisoninto day. I
do not meanto say I neverknewthe physicaldifference-no one but an imbecilecould

This content downloaded on Wed, 2 Jan 2013 10:23:00 AM

All use subject to JSTOR Terms and Conditions

272

DAVID M. HALPERIN

makesuch a claim. I took it, withoutreflectionor the slightestdoubt, that this was a
naturalsex that emergedwithinthe societyof men, with attributesthat naturallycomplementedmasculineattributes.I thoughtit was a naturalphenomenonin the society
of womenas well.Theattributeswerefeminineandso thereseemedno grossmisrepresentation of facts to call them (amongus men) "women.". . . Manyof my "women"had
of handsome,extremelyneat,andpoliteyoungmen.I havelearned,
merelytheappearance
analyzingmy feelingstoday,that those attributesI called femininea momentago were
not femininein any wayas it appearsin the realfemalesex. Theseattributesseem now
merelya tendencyto need, to dependon anotherman;to need neverto becomea rival
or to competewithothermen in the pursuitsmen, amongthemselves,engagein. It was,
it occursto me now, almost boyish-not reallyfeminineat all.
This is the way it alwayswas, even in the State IndustrialSchool for Boys-a penal
- whereI servedfiveyears,fromagetwelveto ageseveninstitutionforjuveniledelinquents
teen. Theywerethe possessionand sign of manhoodand it neveroccurredto any of us
thatthiswasstrangeandunnatural.It is howI grewup-a naturalpartof mylifein prison.
It was difficultfor me to graspthe definitionof the clinicalterm"homosexual"
-and
when I finallydid it devastatedme, as I said.45
Abbott's society surpasses classical Athenian society in the extent to which power
relations are gendered. Instead of the Greek system which preserves the distinction between males and females but overrides it when articulating categories of
the desirable and undesirable in favor of a distinction between dominant and
submissive persons, the system described by Abbott wholly assimilates categories of sociosexual identity to categories of gender identity - in order, no doubt,
to preservethe association in Abbott's world between "masculinity"and the love
of "women."What determines gender, for Abbott, is not anatomical sex but social status and personal style. "Men" are defined as those who "compete with
other men in the pursuits men, among themselves, engage in," whereas "women"
are characterizedby the possession of "attributesthat naturallycomplement masculine attributes"- namely, a "tendency to need, to depend on another man"
for the various benefits won by the victors in "male" competition. In this way
"a natural sex emerges within the society of men" and qualifies, by virtue of its
exclusion from the domain of "male"precedence and autonomy, as a legitimate
target of "male" desire.
The salient features of Abbott's society are uncannily reminiscent of those features of classicalAthenian society with which we are alreadyfamiliar.Most notable
is the division of the society into superordinate and subordinate groups and the
production of desire for members of the subordinate group in members of the
superordinate one. Desire is sparked in this system, as in classical Athens, only

45. Jack H. Abbott, "On 'Women,' " New YorkReview of Books 28:10 (June 11, 1981), 17. It
should perhaps be pointed out that this lyrical confession is somewhat at odds with the more gritty
account contained in the edited excerpts from Abbott's letters that were published a year earlier
in the New YorkReview of Books 27:11 (June 26, 1980), 34-37. (One might compare Abbott's statement with some remarks uttered by Bernard Boursicot in a similarly apologetic context and quoted
by Richard Bernstein, "France Jails Two in a Bizarre Case of Espionage," New York Times [May
11, 1986]: "I was shattered to learn that he [Boursicot's lover of twenty years] is a man, but my conviction remains unshakable that for me at that time he was really a woman and was the first love
of my life.")

This content downloaded on Wed, 2 Jan 2013 10:23:00 AM

All use subject to JSTOR Terms and Conditions

IS THERE A HISTORY OF SEXUALITY?

273

whenit arcsacrossthe politicaldivide,only whenit traversesthe boundarythat

marksout the limitsof intramuralcompetitionamongthe eliteandthat thereby
distinguishessubjectsfrom objectsof sexualdesire.Sex between"men"-and,
remainsunthinkable
in Abbott'ssociety(eventhough
therefore,
"homosexuality"between
anatomical
males
is
an
and
intrinsicpart of the system),
sex
accepted
just as sex betweencitizens,betweenmembersof the empoweredsocial caste,
is practicallyinconceivablein classicalAtheniansociety.Similarly,sex between
"men"and"women"in Abbott'sworldis not a privateexperiencein whichsocial
identitiesarelost or submerged;rather,in Abbott'ssocietyas in classicalAthens,
the act of sex- instead of implicatingboth sexual partnersin a common
- helpsto articulate,
in status
to define,andto actualizethe differences
"sexuality"
betweenthem.
To discoverand to writethe historyof sexualityhas long seemedto manya
sufficientlyradicalundertakingin itself, inasmuchas its effect(if not alwaysthe
intentionbehindit) is to call into questionthe verynaturalnessof what we currentlytake to be essentialto our individualnatures.But in the course of implementingthat ostensiblyradicalprojectmanyhistoriansof sexualityseemto
havereversed- perhapsunwittingly- its radicaldesign:by preserving"sexuality"
as a stablecategoryof historicalanalysisnot only havethey not denaturalized
it but, on the contrary,they havenewlyidealizedit.46Tothe extent,in fact, that
historiesof "sexuality"succeedin concerningthemselveswith sexuality,to just
that extentarethey doomedto fail as histories(Foucaulthimselftaughtus that
much), unlessthey also includeas an essentialpart of their properenterprise
the taskof demonstrating
the historicity,conditionsof emergence,modesof construction,and ideologicalcontingenciesof the verycategoriesof analysisthat
undergirdtheirown practice. Insteadof concentratingour attentionspecifically
on the historyof sexuality,then, we needto defineand refinea new,and radical,
historicalsociologyof psychology,an intellectualdisciplinedesignedto analyze
theculturalpoeticsof desire,bywhichI meanthe processeswherebysexualdesires
are constructed,mass-produced,and distributedamong the variousmembers
of humanliving-groups.48
Wemust trainourselvesto recognizeconventionsof
feelingas well as conventionsof behaviorand to interpretthe intricatetexture
of personallife as an artifact,as the determinateoutcome, of a complexand
arbitrary
constellationof culturalprocesses.Wemust,in short,be willingto admit
46. See Davidson (note 6, above), 16.
47. I wish to thank Kostas Demelis for helping me with this formulation. Compare Padgug, 5:
"In any approach that takes as predetermined and universal the categories of sexuality, real history
disappears."
48. Stephen Greenblatt, "Fiction and Friction," in Reconstructing Individualism: Autonomy, Individuality, and the Self in WesternThought, ed. Thomas C. Heller, Morton Sosna, and David E.
Wellbery,with Arnold I. Davidson, Ann Swidler, and Ian Watt (Stanford, 1986), 30-52, 329-332,
esp. 34, makes a similar point; arguing that "a culture'ssexual discourse plays a critical role in shaping
individuality," he goes on to say, "It does so by helping to implant in each person an internalized
set of dispositions and orientations that governs individual improvisations." See also Padgug; generally, Julian Henriques,WendyHolloway, Cathy Urwin, Venn Couze, and ValerieWalkerdine,Changing

the Subject:Psychology,Social Regulationand Subjectivity(London,1984).

This content downloaded on Wed, 2 Jan 2013 10:23:00 AM

All use subject to JSTOR Terms and Conditions

274

DAVID M. HALPERIN

that what seem to be our most inward,authentic,and privateexperiencesare

actually,in AdrienneRich's admirablephrase,"shared,unnecessary/andpolitical."49
A little less than fifty yearsago W. H. Auden asked, in the openinglines of
a canzone,"Whenshallwe learn,whatshouldbe clearas day,Wecannotchoose
It is a characteristically
whatwe arefreeto love?"50
judiciousformulation:love,
if it is to be love,mustbe a freeact, but it is also inscribedwithina largercircle
of constraint,withinconditionsthatmakepossiblethe exerciseof that"freedom."
The task of distinguishingfreedomfromconstraintin love, of learningto trace
the shiftingand uncertainboundariesbetweenthe self and the world, is a dizzying and, indeed,an endlessundertaking.If I havenot significantlyadvanced
this projecthere,I hope at least to haveencouragedothers not to abandonit.
MassachusettsInstitute of Technology

49. "Translations"
(1972),lines32-33, in AdrienneRich,Divingintothe Wreck:
Poems1971-1972
(New York,1973),40-41 (quotationon p. 41).
50. "Canzone"(1942),lines 1-2, in W.H. Auden,CollectedPoems,ed. EdwardMendelson(New
York,1976),256-257 (quotationon p. 256).

This content downloaded on Wed, 2 Jan 2013 10:23:00 AM

All use subject to JSTOR Terms and Conditions