Clinical Observations and Systematic
Studies of ~ u t o ~ p e ~ h i l i a
The term autogynephilia denotes a male's paraphilic tendency to be
sexually aroused by the thought or image of himself as a woman. This
term subsumes transvestism as well as erotic ideas or situations in which
women's garments per se play a small role or none at all. This review
article presents clinical examples of the lesser known types of autogy-
nephilia (i.e., those in which the element of cross-dressing is secondary
or entirely absent), sbtches earlier attempts to label and conceptualize
these phenomena, summarizes recent quantitative studies exploring the
relationships between autogynephiliu and other psychosexual variables
(e.g., heterosexual attraction), and speculates on the etiology of autogy-
nephilia and its relationship to transsexualism. It is concluded that the
concept of autogynephilia is needed to Jill a gap in our current butte9
of concepts and categories for thinking about gender identity disorders.
All contemporary clinicians are familiar with the phenomenon known as
tramvestism, that is, recurrent cross-dressing in heterosexual males that,
at least in puberty or adolescence, is associated with sexual arousal. Less
well known, however, is the wide range of other cross-gender behaviors
and fantasies that are sexually arousing to subgroups of men who cross-
dress, engage in other symbolically feminine activities, or habitually imag-
ine themselves as females.
In a previous article,' I coined the term autogynephiliu to refer to the
full gamut of erotically arousing cross-gender behaviors and fantasies.
This term was intended to subsume transvestism as well as erotic ideas
or situations in which women's garments per se play a small role or
none at all. The word autogynephilia was constructed from Greek roots
meaning "love of oneself as a woman" and was formally defined as a
male's propensity to be sexually aroused by the thought or image of
himself as a female.
Address reprint requests to Ray Blanchard, Ph.D., Gender Identity Clinic, Clarke Institute of
Psychiatry, 250 College Street, Toronto, Ontario, Canada M5T 1R8.
1 wish to thank Dr. Friedemann Pfifflin for researching the first appearance in print of the word
aulomonusexualism, which I have in previous articles attributed to the wrong edition of Rohleder's
Journal of Sex & Marital Therapy, Vol. 17, No. 4, Winter 1991 O BrunnerlMazel, Inc.
Jou17zal of Sex & I~Iaritd Therapy, Vol. 17, NO. 4, Winto 1991
In this article, I review most of the clinical and all' of the research data
currently available on this erotic phenomenon. In the first section, I
present clinical examples of the lesser known types of autogynephilia,
that is, those in which the element of cross-dressing is secondary or en-
tirely absent. In the second, I summarize earlier attempts to label and
conceptualize these phenomena. In the third, I review recent quantitative
studies exploring the relationships between autogynephilia and other
psychosexual variables, for example, heterosexual attraction. Finally, in
the fourth section, I speculate on the etiology of autogynephilia and its
relationships to gender dysphoria (discontent with one's biological sex, the
desire to possess the body of-the opposite sex, and the desire to be
regarded by others as a member of the opposite sex) and transsexualism
(the extreme form of gender dysphoria).
The purpose of this review is to highlight, for clinicians and researchers
alike, a class of clinically significant cross-gender behaviors and fantasies
that has largely been overshadowed by its most conspicuous exemplar,
namely, transvestism. I hope to show, among other things, that the
broader concept of autogynephilia explains the correlation of various,
seemingly dissimilar sexual behaviors in cross-gendered male popula-
tions, and also that this concept is useful in understanding the develop-
ment of transsexualism in nonhomosexual men.
DESCRIPTION OF THE PHENOMENA
Autogynephilic fantasies and behaviors may focus on the idea of exhib-
iting female physiologic functions, of engaging in stereotypically femi-
nine behavior, of possessing female anatomic structures, or of dressing
in women's apparel. The last-mentioned class of fantasies and behaviors
represents the familiar form of autogynephilia, transvestism. All four
types of autogynephilia tend to occur in combination with other types
rather than alone.
The first of the above-listed types may be designated physiologic autogy-
nephilia. Prime examples of this variety are those occasional males, called
"pregnancy transvestites" by Hirschfeld,* who masturbate with the fan-
tasy of being a pregnant woman or of giving birth. There are, in a similar
vein, men whose favorite masturbation fantasy is that they are lactating or
breast-feeding3 and others whose favorite fantasy is that of menstruating.
These physiologic functions may also be simulated during masturbation
with the aid of appropriate props.
The second type, behavioral autogynephilia, involves the thought or per-
formance of activities that symbolize femininity to the affected male. For
example, one individual, whom I have described elsewhere,* reported
that his early masturbation fantasies included the thought that he was
helping the maid clean the house or that he was sitting in a girls' class
at school. Another 33-year-old patient, whom I recently interviewed,
reported that his current masturbation fantasies were knitting in the
company of other women and being at the hairdresser's with other
Obseruations and Stwly o f Auiogynephilia
The most common behavioral fantasies of adult autogynephilic men
involve the thought of themselves, as women, engaging in sexual inter-
course or other erotic activities. Male patients at the author's gender
identity clinic, for example, commonly report inserting dildos or similar
objects into their rectum while masturbating; this behavior is accompa-
nied by the fantasy that the* =us is a vagina. The same fantasy may be
inferred in other male popuhtions: Blanchard and ~ u c k e r ~
significant correlation, in 1 17 fatal cases of autoerotic asphyxia, between
the presence of dildos at the death scene and feminine attire on the
corpse, suggesting that anal self-stimulation had a symbolic cross-gender
meaning within that group of men as well.
Some autogynephiles fuse the idea of being a woman with their sexual
attra~t:ons toward real women in sexual fantasies in which they are lesbi-
ans engaging in lesbian interactions."-" The occasional man will find a
wife or girlfriend who is prepared to participate in this fantasy to some
extent, stimulating his nipples durin intercourse as if she were fondling
another women's breasts, and so on.
The erotic idea of interpersonal sexuality in the cross-gender role may
also find expression in the fantasy of having intercourse, as a woman,
with a man. The male partner represented in these fantasies is usually a
vague, anonymous figure rather than a real person and probably has
little excitatory function beyond that of completing the fantasy of vaginal
intercourse in the female role. Fantasies of this class sometimes lead to
actual sexual intercourse with men, particularly with the affected individ-
ual in cross-dress or otherwise performing in some role he conceives as
feminine. lo- l2 The effective erotic stimulus in such interactions, however,
is not the male physique of the partner, as it is in true homosexual
attraction, but rather the thought of being a woman, which is symbolized
in the fantasy of being penetrated by a
The fantasy of vaginal intercourse in the female role may be enacted
with a female rather than a male partner, and even without the partner's
knowledge. In many cases, the autogynephile prefers to have intercourse
with his wife in the female superior position. He then fantasizes that his
wife-imagined as a man-is penetrating him-a
go on for years without the individual's wife ever realizing why her hus-
band prefers that position for intercourse or how she is transformed in
The third type might be described as anatomic autogynephilia. Anatomic
autogynephilia, in its purest form, is repre<ented by rather static fanta-
sies--one might call them rather images or icons-consisting of little
more than the idea of having a woman's body. These may focus on female
anatomic structures such as the breasts or the vulva or on typical but
acquired characteristics such as hairless legs. A patient seen by one of my
colleagues, for example, was sexually aroused by shaving his legs and
then contemplating the result.
The fourth major type of autogynephilia, as I have defined it, is
transvestic autogynephilia (or simply, transvestism). The rationale for sub-
suming transvestism under the heading of autogynephilia is that the
woman.15-l8 This may
Journal of' Sex & Marital Therap?, Vol. 17, No. 4, Winter 1991
transvestite's excitement results from making himself, in some sense,
more like a woman, whatever his conscious thoughts during that act. In
fact, most transvestites do fantasize themselves as females when they
are cross-dressing and may also act this out in their behavior. Other-
individuals, who have no explicit thoughts of femininity, spend consider-
able amounts of time admiring their appearance in the mirror and are
sexually aroused by the image of themselves as w~men.".'~.~~
In all the above-mentioned types of autogynephilia, the relationship
between the cross-gender stimulus and sexual excitement is probabilistic
rather than inevitable. An autogynephile does not necessarily become
sexually aroused every time-he pictures himself as a female or engages
in feminine behavior, any more than a heterosexual man automatically
gets an erection whenever he sees an attractive woman. Thus, the concept
of autogynephilia-like that of heterosexuality, homosexuality, or pedo-
philia-refers to a potential for sexual excitation.
The first three types of autogynephilia-physiologic,
anatomic-are usually found in association with cross-dressing. This .is
probably one reason why clinicians and researchers have tended to re-
gard these phenomena, when they have noted them at all, as outgrowths
or extensions of transvestism. The questionability of this assumption is
illustrated by the following clinical vignette of an adult male outpatient
with a lifelong history of anatomic and behavioral autogynephilia to-
gether with a virtual absence of transvestism.
Philip was a 38-year-old MBA referred to our gender identity clinic for
assessment. His presenting complaint was chronic gender dysphoria,
which had led, on occasion, to episodes of depression severe enough to
disrupt his professional life. He presented as an intelligent and coopera-
tive individual, unremarkably masculine in appearance and in manner.
He had never been married and he had no children.
Philip was the third of four children in a happy and prosperous mid-
dle-class home. Most of his childhood friends were other boys, but he
also got along well with girls. He did well in grammar and high' school,
obtaining good grades and excelling in sports. He was popular with his
classmates and enjoyed harmonious relations with his family. This pat-
tern of social competence continued into adult life, although, in later
years, he exhibited a preference for friendship with women.
His first specific recollection of wanting to be a female dated back to
age 6. Encountering a wishing well for the first time, he begged a penny
from his father to throw in. His wish was that God would listen to his
prayers and let him change into a girl. At about the same age, he cross-
dressed for the first and only time in his life. This consisted of trying on
a dress belonging to an older cousin. When questioned why he did not
cross-dress at present-he lived alone and there was nothing to prevent
him-he indicated that he simply did not feel strongly impelled to do so.
Obsmatzom and Study o f Auto~ynephilia
He began masturbating-at puberty, which occurred at age 12 or 13.
The earliest sexual fantasy he could recall was that of having a woman's
body. When he masturbated, he would imagine that he was a nude
woman lying alone in her bed. His mental imagery would focus on his
breasts, his vagina, the softness of his skin, and so on-all
tic features of the female physique. This remained his favorite sexual
fantasy throughout life. His other masturbatory fantasies were less fre-
quent and much less powerful. One of these was the idea of dressing as a
woman; another was the fantasy of himself, as a woman, being penetrated
vaginally by a man. The latter thought began to occur to him in his
thirties; by the time he presented to us, it was arising in one-third to one-
half of masturbatory sessions. The imagined partner remained vague .
in outline, however: a nameless, faceless abstraction rather than a real
acquaintance or remembered stranger.
Philip's first heterosexual intercourse was at age 18. This was a one-
night stand with a woman he met at a party. He experienced no erectile
difficulties. During the next two decades, he had intercourse with nine
different women, of which six were one-night stands. In common with
most heterosexual male gender dysphorics, he tended to employ cross-
gender ideation as an aid during coitus. He preferred to have intercourse
with the woman on top, and he would fantasize that he was the woman
and his partner was the man. At the time he presented, he had not had
intercourse for over a year. His last attempt had been unsuccessful be-
cause of erectile difficulties.
His only long-term heterosexual relationship was with a fellow student,
Elisa, in graduate school. This was in his early twenties and lasted about
two years. The couple had intercourse only during their first few months
of living together. The frequency of coitus then tapered off to zero,
although they continued sleeping in the same bed.
Philip never had a homosexual experience. This was from lack of
interest rather than lack of opportunities. Elisa had been interested in
the arts, and Philip had known a number of gay individuals through her.
This vignette illustrates that when a patient's primary sexual object is
the thought of himself with a woman's body, there may be little overt
paraphilic behavior. This does not, of course, mean that the deviant
interest has no clinical significance. In the present case, for example, it
was clearly related to the patient's gender dysphoria and to his inability
to form long-term relationships with women. -
PREVIOUS CLINICAL FORMULATIONS
Many clinical observers have noted behaviors similar to those described
above. They have often, however, attributed them to motivations other
than autogynephilia. The fo!lowing quote from ~ a r ~ m a n ~ l
If a married man insists in his relations with his wife in occu-
pying the succubus position and at the same time demands of
Journal of Sex & Marital Therapy, Vol. 17, No. 4, Winter 1991
her that she massage his breasts, this cah hardly be interpreted
as anything else but an expression of unconscious or latent
homosexuality. (p. 293)
It was self-evident to Karpman that such behavior betokened a sexual
interest in men; he never even- considered that the fundamental and
irreducible sexual stimulus was the idea of being a woman. Other writers,
however, both before and after Karpman, have perceived such behavior
in terms resembling my own notion of autogynephilia.
Probably the first such writer was ~irschfeld.~
idea of being a woman irr-a group of cross-dressing males whom he
described as automonosexuals:
He identified the erotic
We are almost tempted to believe that we are here faced with
a splitting of the personality in the sense that the masculine
component in the psyche of these men is sexually stimulated
by the feminine component and that they feel attracted not by
the women outside them, but by the woman inside them.22 (p.
Hirschfeld borrowed the term "automonosexualism" from ~ohleder.~
Rohleder, however, had used the term to denote a kind of pathological
narcissism in which the individual is excited by his own body in its real
(i.e., male) form; whereas Hirschfeld's automonosexual cross-dresser is
aroused by the fantasy that his body is that of a woman.
Havelock Ellis, a contemporary of Hirschfeld's, had similar percep-
tions, although he couched them in somewhat different language. Ellis
used the term Eonism, usually in regard to nonhomosexual males, to
designate overt cross-gender behavior as well as subjective feelings; he
sometimes used an alternative term, sexo-aesthetic inversion, for the same
thing. In his view:
The Eonist is embodying, in an extreme degree, the aesthetic
attribute of imitation of, and identification with, the admired
object. It is normal for a man to identify himself with the
woman he loves. The Eonist carries that identification too far.24
In other writings, Ellis reiterated his opinion that "Eonism" and normal
heterosexual interest have some common point of origin: "Psychologi-
cally speaking, it seems to me that we must regard sexo-aesthetic inver-
sion as really a modification of normal hetero-~exuality"'~ (p. 103). This
point will be taken up again in the next section.
Feni~hel,~~ writing on transvestism, also noted autogynephilic phenom-
ena in terms not dissimilar from Hirschfeld and Ellis. He did not, how-
ever, dwell long at the descriptive level:
Love for the subject's own self-phantasies that the masculine
element in his nature can have intercourse with the feminine
Observations and Study o f Autogymphilia
(i.e., with himself) are not uncommon. Love for the phallic
mother is often transformed into love for the ego in which a
change has been wrought by identification with her. This is a
feature in the psychic picture which has struck even non-analyt-
ical writers, who have described a narcissistic type of transves-
tist. (p. 214)
Although Fenichel noted the same fantasies as Hirschfeld and Ellis, he
also, in a sense, denied their importance. In Fenichel's view, the transves-
tite's driving fantasy was not the conscious thought of himself as a woman
with a vulva but rather the unconscious thought of himself as a woman
with a penis.
B~ckner*~ advanced an elaborate theory of the developmental events
leading to transvestism. In his theory, the future transvestite begins with
fetishistic masturbation, but then
begins to build in fantasy a more complete masturbation im-
age. . . . Through a process of identification and fantastic social-
ization he takes the gratificatory object into himself. . . . [The
next step] involves this elaboration of masturbation fantasies
into the development of a feminine self (pp. 383-384)
. . . [which is] gratifying in both sexual and social ways. When
it becomes fixed in his identity, he begins to relate toward him-
self in some particulars as if he were his own wife. (p. 387)
Thus, Buckner also recognized the erotic idea of being a woman, al-
though it is debatable whether he located it correctly in the develop-
The foregoing examples show that a variety of writers have specifically
noted autogynephilic ideation, although their formulations of it have
been differently colored by their theoretical views. One final comment
on the phenomenology of autogynephilia is necessary. Some of the fore-
going clinical quotes might give the impression that the autogynephile
experiences his feminine persona as a sort of imaginary twin, somehow
distinct or detached from himself. Although this might be true in some
cases, it is my sense that, most often, the individual simply experiences
himself, in his own body, as a woman.
FINDINGS OF QUANTITATIVE STUDIES
Although clinical observations of autogynephilia go back several decades,
the only quantitative studies I know that bear directly on this topic are
four recent investigations of my own. These studies suggest three main
conclusions: 1) Autogynephilia is a misdirected type of heterosexual im-
pulse, which arises in association with normal heterosexuality but also
competes with it. 2) The majority of men who acknowledge anatomic
autogynephilia also report some history of transvestism, and a substantial
proportion also report some history of attraction to specific garments or
Journal o f Sex &? Marital Tlterapy, Vol. 17, No. 4, Winter 1991
materials. 3) Anatomic autogynephilia is more closely associated with
gender dysphoria than is transvestism. The research program generating
these conclusions is described in the remainder of this section.
Autogynephilia and Heterosexuality
Hirschfeld's observation that autogynephilic men "feel attracted not by
the women outside them, but by the woman inside them" actually con-
tains two separate ideas. The first is that autogynephilia may be conceived
as a kind of misdirected heterosexuality. The second is that autogynephi-
lia competes with normal heterosexual attraction.
The first hypothesis suggested by Hirschfeld's observation-that auto-
gynephilia is a misdirected type of heterosexual impulse-predicts
one should find higher levels of autogynephilia in heterosexual--or at
least nonhomosexual-men than in comparable homosexual men. This
prediction has been supported by the results of a study by ~lanchard.~~
The subjects in this study were 212 adult male-to-female transsexuals.
These were divided into four groups: one homosexual (attracted to other
males) and three nonhomosexual (attracted to females, to both sexes, or
to neither sex).
The measure of autogynephilia used in this study was called the Core
or CAS for short. Most of the items in this multi-
ple-choice questionnaire measure ask whether the respondent has ever
become sexually aroused while picturing himself with various features of
the female anatomy (e.g., breasts). Therefore the CAS is primarily a
measure of anatomic autogynephilia.
The four transsexual groups were compared on the CAS (and on
several other psychosexual variables that are not immediately relevant).
As predicted, all three categories o'f nonhomosexual males were more
likely to report sexual arousal in association with fantasies of womanhood
than the homosexual males. This finding supports the view that autogy-
nephilia is, as Ellis put it, "really a modification of normal hetero-sexu-
A subsequent studyz8 examined the second hypothesis suggested by
Hirschfeld's observation, namely, that autogynephilia and normally di-
rected heterosexual interest are competing drives (or behaviors, or orien-
tations). The subjects were 427 adult male outpatients who reported
histories of dressing in women's garments, of feeling like women, or both.
These were selected without regard to sexual orientation, and the sample
included men reporting all degrees of sexual attraction to adult women.
For a sample thus composed, Hirschfeld's first hypothesis predicts that
men reporting little interest in (real) women will also report little autogy-
nephilia; as one moves along the continuum from subjects with low levels
of heterosexual attraction toward subjects with intermediate levels, the
amount of observed autogynephilia should increase. Tho second hypoth-
esis concerns the middle-to-high range of the heterosexual interest con-
tinuum. The notion of intrinsic competition implies that high degrees of
interest in the one type of sexual object preclude high degrees of interest
Observutions and Study of Autogynephilia 243
in the other. Thus, as one moves further along the continuum from
subjects with intermediate levels of heterosexual attraction to subjects
with high levels, the amount of observed autogynephilia should reverse
direction and begin to decrease again. In operational terms, the com-
bined hypotheses imply that a plotted function relating measures of auto-
gynephilia and heterosexual interest should take the form of an inverted
U; this was the prediction tested in the study.
As in the previous study, autogynephilia was measured with the CAS.
Heterosexual interest was measured with the Modified Gynephilia
Scale,ls or MGS, a measure of erotic attraction to physically mature
women specifically developed for the assessment of adult male gender
patients. A third variable of present relevance, transvestism, was mea-
sured with the Cross-Gender Fetishism S ~ a l e , ~ ~ , ~ "
is a measure (for males) of the erotic arousal value of putting on women's
clothes, perfume, and make-up, and shaving the legs. The items in the
CGFS focus on the act of cross-dressing rather than subjective feelings
As predicted, the highest levels of autogynephilia were observed at
intermediate rather than high levels of heterosexual interest; that is, the
function relating the CAS and the MGS did take the form of an inverted
U. A different result was obtained for transvestism; the CGFS curve
appeared to level off rather than reverse direction at the highest degrees
of heterosexual interest. The former finding supports the view that auto-
gynephilia is a misdirected type of heterosexual impulse, which arises in
association with normal heterosexuality but also competes with it. The
latter further suggests that some types of autogynephilia (e.g., anatomic
autogynephilia) may compete more strongly with normal heterosexual
attraction than other types (e.g., transvestism).
These findings are reinforced by other kinds of evidence that also
suggest autogynephilia and normal heterosexual attraction are compet-
ing phenomena. The foregoing study was entirely cross-sectional in na-
ture, the autogynephilic and heterosexual tendencies of each subject hav-
ing been measured concurrently and only once. Clinical experience,
however, suggests that this competition may also be observed longitudi-
nally. It is not rare for a heterosexual male cross-dresser or gender dys-
phoric to report that, when he first met a woman and fell in love, his
desires to cross-dress or engage in other cross-gender behaviors dimin-
ished or disappeared, sometimes for longer than a year. When, however,
the intensity of passionate love resobed into the mellower comforts of
married life, his desires to dress or Iive as a female reasserted themselves.
The results of the above study bear upon, and erha s illuminate, one
further clinical observation. Person and Ovesey' (p. 307) remarked that
66. interpersonal sexuality is almost always attenuated" in transvestism, and
other clinical authors have made similar statements. My findings suggest
that it may not be transvestism per se that competes with normal hetero-
sexual attraction so much as the anatomic autogynephilia that often ac-
The conclusion of this section requires one additional point. In some
autogynephilic men, the loss of sexual attraction to women is offset by a
or CGFS for short. This
Journal of Sex U Marital Therub, Vol. 17, No. 4, Winter 1991
kind of secondary erotic interest in men, alreaay described in a previous
section. Thus, the net impact of autogynephilia on erotic interest in other
persons may be rather small. This was confirmed by ~lanchard,~~
found a near-zero correlation hetween the CAS and an Alloeroticism
Scale, which was developed in .that study to measure sexual interest in
other persons irrespective of their sex.
Anatomic Autogynephilia, Transvestism, and Fetishism
I have already stated the opinion that anatomic autogynephilia is usually
found in association with-transvestism rather than alone. There is no
previously published research to support this assertion, however. In pre-
paring the present article, I therefore carried out a small study to demon-
strate this overlap. I also tried to estimate the proportion of autogyne-
philes with some evidence of fetishistic traits.
This investi ation used the same resources as my other studies of auto-
gynephilia.27.2 3' The on-line database of the Clarke Institute of Psychia-
try's Research Section of Behavioural Sexology includes questionnaire
data on 3,500 male patients who have presented either at that department
or at the Institute's Gender Identity Clinic since September 1980-the
date when the last questionnaire items used in this study were added to
the test battery. These cases were searched for all subjects who satisfied
both of the following criteria: 1) The subject obtained a score less than
10 on the Modified Androphilia-Gynephilia ~ n d e x , ~ ~
nonhomosexual partner preference; and 2) he obtained a score of 3 or
higher on the CAS, clearly acknowledging some history of autogynephilic
A total of 210 cases was retrieved for this study. The mean age of the
sample was 33.4 years (range = 18-67 years). The mean, median, and
modal educational level were all Grade 12.
The selected subjects were then dichotomously classified as transvestic
or not-transvestic, fetishistic or not-fetishistic. A subject was classified as
transvestic if he endorsed any item on the CGFS. A subject was classified
as fetishistic if he responded positively to the individual questionnaire
item, Do you think that certain inanimate objects (velvet, silk, leather, rubber,
shoes, female underwear, etc.) have a stronger sexual attraction for you than for
most other people?
The results showed that 90% of these autogynephilic men acknowl-
edged some history of transvestism: 35% acknowledged transvestism
alone, and 55% acknowledged fetishism as well as transvestism. The re-
maining 10% denied both. It should be noted that this last group did
not necessarily deny cross-dressing; they merely denied being sexually
aroused by cross-dressing.
The above percentages should be regarded as "ball park" estimates
rather than precise figures. The item used to diagnose fetishism does not
distinguish as clearly as one would like between transvestism and fetish-
ism (although its connotations seem clear enough, particularly for gender
thus indicating a
Obsen,ution.s and Study o f Autogynepldia
patients, who rarely are very naive about such matters). We do not, more-
over, know whether subjects are equally willing to acknowledge autogy-
nephilia, transvestism, and fetishism; or how a greater reluctance to re-
port one of these paraphilias might affect our results. I believe, however,
that the above study is adequate to indicate the general trend: The major-
ity of men who acknowledge anatomic autogynephilia also report some
history of transvestism, and a substantial proportion also report some
history of attraction to specific garments or materials.
Autogynephilia and Gender Dysphoria
The different types of autogynephilia, alone or in various combinations,
tend to occur in association with another phenomenon of equal clinical
significance, namely, gender dysphoria. Analyzing the relationship be-
tween autogynephilia and gender dysphoria is likely to prove a long-term
undertaking, requiring multiple studies. A reasonable place to start on
this task is determining which types of autogynephilia are most closely
associated with gender dysphoria. One would think, on clinical or even
commonsense grounds, that different types of autogynephilia are likely
to differ in the strength of such associations. It would seem, in particular,
that the man whose principal sexual fantasy is that of having a woman's
body would be closer to requesting vaginoplasty than the man whose
principal fantasy is that of wearing women's clothes. This logic leads
to the general prediction that anatomic autogynephiles should be more
gender dysphoric than transvestites.
The above prediction was tested in the fourth quantitative
subjects were 238 nonhomosexual male outpatients with some history of
autogynephilic behavior. These were divided into three groups: those
most aroused sexually by images of themselves as nude women; those
most aroused by images of themselves as women in underwear; and those
most aroused by images of themselves as fully clothed women. Thus, in
the terminology used in the present article, the first, or Nude, group
were primarily anatomic autogynephiles; the Underwear and Clothed
groups were primarily transvestic autogynephiles (or simply, transves-
These subjects were then compared on questionnaire measures of gen-
der dysphoria as well as other psychosexual variables. As predicted, the
Nude group was significantly more gender
Clothed or the Underwear group.
The above finding suggests the following, rudimentary theory relating
an individual's type of autogynephilia to the presence or absence of
transsexual wishes: Autogynephilia takes a variety of forms. Some men
are most aroused sexually by the idea of wearing women's clothes, and
they are primarily interested in wearing women's clothes. Some men are
most aroused sexually by the idea of having a woman's body, and they
are most interested in acquiring aewoman's body. Viewed in this light,
the desire for sex reassignment surgery of the latter group appears as
logical as the desire of heterosexual men to marry wives, the desire of
- dysphoric than either the
JounznL @'Sex @ Mantcd Theruf?y, VoL. 17, No. 4, Winter 1991
homosexual men to establish permanent relationships with male part-
ners, and perhaps the desire of other paraphilic men to bond with their
paraphilic objects in ways no one has thought to observe. I will return to
this idea in a later discussion.
An unanticipated result of this study was the pattern of differences
between the Clothed and Underwear groups. The Underwear group was
reliably less gender dysphoric, reliably more sadistic and masochistic.
This suggests that an adequate taxonomy of autogynephilia may require
more than a breakdown into anatomic, physiologic, behavioral, and
transvestic types. One may also find clinically significant distinctions by
subdividing (primarily) trawestic cases according to their preferred arti-
cles of women's apparel.
Another secondary finding was that the Nude group was the same age
as the Underwear group and significantly younger than the Clothed
group. This outcome makes it unlikely that erotic fantasies of having a
woman's body are the end result of some progression that necessarily
begins with erotic fantasies of wearing women's clothes. This finding,.
therefore, reinforces the point made in introducing the clinical vignette
of "Philip," namely, that it is questionable whether anatomic autogyneph-
ilia should be regarded as an outgrowth or extensior? of transvestism.
This section discusses hypotheses of autogynephilia and autogynephilic
gender dysphoria suggested by the objective data reviewed above. These
are intended as working hypothesesfor future studies rather than final
conclusions on the research completed to date.
Etiology of Autogynephilia
The finding that anatomic autogynephilia tends to be accompanied by
transvestism and fetishism is another example of the well-known ten-
dency for multiple paraphilias to occur in the same indi~idual.~.~-" 1 I
Specific clusters previously described include one comprising vo eurism,
exhibitionism, toucheurism-frotteurism, and preferential ra e3z-40 and
another comprising shoe or foot fetishism and masochism.'
Ban~roft~~ observed that the tendency of paraphilias to occur together
suggests that the conditions necessary for the development of one type
of paraphilia may facilitate the development of others. He conjectured
that this potential might stem from some characteristic of the individual's
nervous system that underlies sexual learning. Bancroft's notion may be
extended a bit further: The fact that there is more than one type of
paraphilic cluster suggests that there may be more than one type of
What kind of defect in a male's capacity for sexual learning could
produce anatomic autogynephilia, transvestism, and fetishism, singly and
in various combinations? Common to all these phenomena is a kind of
error in locating heterosexual targets in the environment. In fetishism,
Observations and Study o f Autogmephiliu
the individual orients toward a particular garment (e.g., panties, bras-
sieres) rather than those parts of the female body the garment usually
covers. In transvestism, the individual is aroused by the appearance of
an attractively clad woman, but he locates this image on himself rather
than another person. In anatomic autogynephilia, the individual is ori-
ented toward the characteristic features of the feminine physique (e.g.,
breasts), but he attempts, in some way, to locate these features on his own
The above analysis suggests the failure of some developmental process
that, in normal males, keeps heterosexual learning "on track," perhaps
by biasing erotic response toward external rather than internal stimuli,
and inherent rather than variable features of the female appearance. '
This putative defect allows the development of various misdirected-but
still recognizably heterosexual-behaviors, and makes it possible, if not
probable, that more than one misplaced interest will appear in the same
This etiological hypothesis explains the tendency for anatomic autogy-
nephilia to be accompanied by transvestism and fetishism without as-
serting that any one of these paraphilias is a direct consequence of an-
other. This is critical because, although these paraphilias do tend to
cluster, each of them can and does also occur alone. There are, on the
other hand, some types of autogynephilic behavior and fantasy that might
be explained as secondary elaborations of more basic forms.
Autogynephilic fantasies of vaginal intercourse with men, with the sub-
ject imagining himself in the female role, are a case in point. I have
already presented my view that the effective erotic stimulus in such inter-
actions is not the male physique of the partner, as it is in true homosexual
attraction, but rather the thought of being a woman, which is incorpo-
rated in the fantasy of being penetrated by a man. This analysis prompts
the question: Why should the autogynephile fantasize vaginal intercourse
when he could simply fantasize himself with a vagina? The answer may
lie with a general characteristic of heterosexual men. The most common
activity in pornographic videotapes made for this population is hetero-
sexual intercourse (including fellatio and cunnilingus); scenes of solitary
women masturbating themselves are considerably less frequent. This sug-
gests that heterosexual observers are more aroused by a man and woman
together than by a woman alone, even though the male actoi- may have
no value as a sexual object in his own right. In autogynephilic men "ob-
serving" their own fantasies, this same propensity may make images of
vaginal intercourse more exciting than simple images of themselves with
vaginas. This hypothesis illustrates that the relatively complex and active
fantasies of behavioral autogynephilia may be the result of anatomic
autogynephilia interacting with certain components of normal heterosex-
In summary, the quantitative data27 together with the qualitative analy-
ses presented in this section support Ellis's contention that autogynephilia
is "really a modification of norma! hetero-sexuality." The nature of this
JOUW~ o f Sex &f Marital Tlmapy, Vol. 17, No. 4, Wilzter 1991
"modification" is, at least initially, one of direction, with behaviors sug:
gesting heterosexual intent directed at objects other than real-life fe-
Etiology of Transsexualism in Nonhomosexual Men
The last quantitative study clarified one aspect of the relationship be-
tween autogynephilia and transsexualism: Within the population of adult
nonhomosexual men with recurrent cross-dressing or cross-gender ide-
ation, the desire to be female is more closely associated with anatomic
autogynephilia than with t~ansvestism.~'
able study, however, elucidates the developmental interrelationships of
autogynephilia and transsexualism.
Any viable theory relating the etiologies of autogynephilia and trans-
sexualism must explain the following well-established observation: Gen-
der dysphoria, in young nonhomosexual males, usually appears along
with, or subsequent to, autogynephilia; in later years, however, autogy-
nephilic sexual arousal may diminish or disa ear, while the transsexual
wish remains or grows even s t r ~ n ~ e r . ' ~ . ' ~ . ~ ~ . ~
produced by gender-dysphoric patients, but one does not have to rely on
self-report to accept that the transsexual motive may attain, or inherently
possess, some independence from autogynephilia. The same conclusion
is suggested by the fact that surgical castration and estrogen treat-
ment-which decrease libido in gender dysphorics as in other men-usu-
ally have no effect on the desire to live as a female or the resolve to
remain in that role.
One may speculate that the above developmental sequence reflects
the operation, in autogynephilic men, of certain normal heterosexual
behaviors. Many men, after years of marriage, are less excited by their
wives than they were initially but continue to be deeply attached to them;
in other words, pair-bonding, once established, is not necessarily depen-
dent on the continuation of high levels of sexual attraction. It is therefore
feasible that the continuing desire to have a female body, after the disap-
pearance of sexual response to that thought, has some analog in the
permanent love-bond that may remain between two people after their
initial strong sexual attraction has largely disappeared.
It must be emphasized that the foregoing discussion pertains entirely
' to transsexualism in nonhomosexual males. I have previously argued, on
the basis of both formal research and clinical evidence, that the type of
transsexualism that develops in this population is gualitatively different
from the type that develops in homosexual males ~
type that develops in nonhomosexual fema1es.l~~~
Neither this nor any other avail-
45 Such histories are often
and from the
The concept of autogynephilia-or something very like it-is
fill a gap in our current battery of concepts and categories for thinking
about gender identity disorders. There is no other term for designating
Observations and Stud?, of Autogynephilia
erotic arousal in men associated with the thought or image of themselves
as women; there is no other conceptual basis for classifying erotic fanta-
sies as diverse in form-but similar in meaning-as menstruating, breast-
feeding, making love to a lesbian as a gay woman, making love to a
man as a straight woman, sitting in a girls' class at school, knitting, and
possessing shaved legs. The concept of transvestism does not meet this
need. Transvestism is now "officially" and quite narrowly defined in the
DSM-111-R as "recurrent intense sexual urges and sexually arousing fan-
tasies involving cross-dressing"48 (p. 289). Broadening the definition of
transvestism to capture behaviors and fantasies like those mentioned
above would, at this point, create more confusion than simply adding
another term to the nomenclature; moreover, the term transvestism would
be a very poor descriptor of the behavioral class in question.
The notion of autogynephilia points toward an unexplored, multiform
array of cross-gender behaviors that must eventually be explained by
any comprehensive theory of gender identity disorders. The process of
generating and testing such theories might even force refinements, if
not substantial revisions, in current explanations of the familiar form of
autogynephilia, transvestism. Not only the etiological theory of gender
identity disorders but also the clinical management of these conditions
could profit from adding the notion of autogynephilia to our conceptual
armamentarium. The research completed to date on nonhomosexual
male gender patients indicates that anatomic autogynephilia rather than
transvestism is the main correlate of transsexual tendencies and also of
diminished capacities for heterosexual relations and pair-bond forma-
tion. These findings suggest that clinicians' prognostic judgments might
improve if they question their patients about the less visible forms of .
autogynephilia as well as the outward forms such as transvestism.
I have, at a few places in this article, speculated that some aspect of
autogynephilia might be understood in relation to characteristic features
of heterosexual behavior. If I am correct, then progress in explaining
autogynephilia is linked to progress in explaining normal heterosexual
development. It is conceivable that concepts and theories might flow in
both directions, with research on autogynephilic behavior elucidating
aspects of ordinary heterosexuality. It is, on the other hand, certain that
a full account of normal psychosexual development is essential to any
comprehensive theory of autogynephilia, and that we cannot, in the final
analysis, understand how one man comes to love himself as a woman
until we understand how another man comes to love a woman.
Blanchard R: The classification and labeling of nonhomosexual gender dysphorias.
Arch Sex Behuv 18:315-334, 1989. -
Hirschfeld M: Sexwllpathologie [Sexual pathology] (Vol. 2). Bonn, Marcus & Weber,
Hirschfeld M: Transvestites (MA Lombardi-Nash, trans). Buffalo, NY, Prometheus,
1991. (Original work published 19 10).
Ioi~muI of' SIX &f Murital T/mufi, Vol. 17, No. 4, Winter I991
4. Blauchard K: Gender identity disord'ers in adult men. In R Blanchard, BW Steiner
(eds), Clillical rnallcpr,~,lanl ofge?tder identitv diwrders in children and adults. Washington,
DC, American Psychiatric Press, 1990, pp. 47-76.
-5. Blanchard K, Hucker S.1: Age, transvestism, bondage, and concurrent paraphilic
activities in 117 fatal cases of autoerotic asphyxia. Br J Psychiatry, in press.
6. Bancroft J: The relationship between gender identity and sexual behaviour: Some
clinical aspects. In C. Ounsted, DC Taylor (eds), Gender differences: Their ontogenv and
sipiJiconce. London, Churchill Livingstone, 1972, pp. 57-72.
7. Beigel HG, Feldman R: The male transvestite's motivation in fiction, research, and
reality. In HG Beigel (ed), Advances in sex research. New York, Harper & Row, 1963,
8. Ovesey L, Person E: Transvestsm: A disorder of the sense of self. Int J Psychoanal
Psvclwther 5:2 19-235, 1976.
9. Newman LE, Stoller RJ: Nontranssexual men who seek sex reassignment. Am J
P.yc/iiut 131:437--441, 1974.
10. Benjamin H: Transvestism and transsexualism in the male and female. J Sex Res
3: 107-127, 1967.
11. CroughanJL, Saghir M, Cohen R, Robins, E: A comparison of treated and untreated
male cross-dressers. Arch Sex Behav 10:515-528, 198 1.
12. Person E, Ovesey L: Transvestism: New perspectives. J Am Acad Psychounul
13. Blanchard R: Typology of male-to-female transsexualism. Arch Sex Beluw
14. Ellis H: Studies in the psychology of sex (Vol. 7). Philadelphia, FA Davis, 1928.
15. Benjamin H: The transsexual phenomenon. New York, Julian, 1966.
16. Lothstein LM: The aging gender dysphoria (transsexual) patient. Arch Sex Buluw
8:43 1444, 1979.
17. Lukianowicz N: Survey of various aspects of transvestism in the light of our present
knowledge. J Nen, Ment Dis 128:3&64, 1959.
18. van Kammen DP, Money J: Erotic imagery and self-castration in transvestism/trans-
sexualism: A case report. J Homosex 2:359-366, 1977.
19. Buhrich N, McConaghy N: The discrete syndromes of transvestism and transsexu-
alism. Arch Sex Behav 6:483495, 1977.
20. Gutheil EA: The psychologic background of transsexualism and transvestism. Am J
Pqchother 8:231-239, 1954.
21. Karpman B: Dream life in a case of transvestism with particular attention to the
problem of latent homosexuality. J Nen, Ment Dis 106:292-337, 1947.
22. Hirschfeld M: Sexual anomalies. New York, Emerson, 1948.
23. Rohleder H: Vorlesungen iiber Geschlechtstrieb und gesamtes Geschleschtsleben des Memchen
[Lectures on sexual drive and the complete sexual life of man] (Vol. 2). Berlin,
Fischers medizinische BuchhandlungtH Kornfeld, 1907.
24. Ellis H: Psychology of sex: A manual for students. New York, Emerson, 1935.
25. Fenichel 0:
The psychology of transvestism. Int J Psychoanal 11:211-227, 1930.
26. Buckner HT: The transvestic career path. Psychiatry 33% 1-389, 1970.
27. Blanchard R: The concept of autogynephilia and the typology of male gender dys-
phoria. J N m Ment DG 177:616-623, 1989.
28. Blanchard R: The nonmonotonic relationship of autogynephilia and heterosexual
attraction. J Abnorm Psychol, in press.
29. Blanchard R: Research methods for the typological study of gender disorders in
males. In BW Steiner (ed), Geruler dysphoria: Deuelopment, research, managemmt. New
York, Plenum, 1985, pp. 227-257.
Observations and St* of Autogynephilia Download full-text
30. Blanchard R: Cross-Gender Fetishism Scale. In CM Davis, WL Yarber, SL Davis
(eds), Sexuality-related meawres: A compendium. Lake Mills, I A, Graphic, 1988, pp.
31. Blanchard R: Varieties of autogynephilia and their relationship to gender dysphoria.
Arch Sex Behau, in press.
32. Abel GG, Becker JV, Cunningham-Rathner -1, Mittelman M, Kouleau J-L: Multiple
paraphilic diagnoses among sex offenders. Bull Am Acad Pqchiatr?, Law 16: 153-1 68,
33. Buhrich N, Beaumont T: Comparison of transvestism in Australia and America. Arch
Sex Behav 103269-279, 1981.
34. Lang RA, Langevin R, Checkley KL, Pugh G: Genital exhibitionism: Courtship disor-
der or narcissism? Can J Befmu Sci 19:216-232, 1987.
35. Macdonald JM: Indecent exposure. Springfield, IL, Charles C Thomas, 1973. .
36. Rooth G: Exhibitionism, sexual violence and paedophilia. Br J Psychiat 122:705-710,
37. Wilson GD, Gosselin C: Personality characteristics of fetishists, transvestites and sado-
masochists. Personal Ind DifJ' 1 :289-295, 1980.
38. Freund K: Diagnosis and treatment of forensically significant anomalous erotic pref-
erences. Can J Criminol Cowec 18:181-189, 1976.
39. Freund K, Scher H, Hucker S: The courtship disorders. Arch Sex Behuu 12:36'3-879,
40. Freund K, Scher H, Racansky IG, Campbell K, Heasman G: Males disposed to
commit rape. Arch Sex Behuv 15:23-35, 1986.
4 1. Krafft-Ebing R, von: Psychopathia sexuulir (FS Klaf, trans). New York, Scarborough,
1978. (Original work published 1903).
42. Bancroft J: Human sexuality and its problem (2nd ed). London, Churchill Livingstone,
43. Buhrich N, McConaghy N: The clinical syndromes of femmiphilic transvestism. Arch
Sex Behav 6:397-412, 1977.
44. Levine SB, Lothstein L: Transsexualism or the gender dysphoria syndromes. J Sex
Marital Ther 7:85-113, 1981.
45. Wise TN, Meyer JK: The border area between transvestism and gender dysphoria:
Transvestitic applicants for sex reassignment. Arch Sex Behuv 9:327-342, 1980.
46. Blanchard R: onh homosexual gender dysphoria. J Sex Res 24: 188-193, 1988.
47. Blanchard R: Gender identity disorders in adult women. In R Blanchard, BW Steiner
(eds), Clinical manugement of gender identity disorders in children and adults. Washington,
DC, American Psychiatric Press, 1990, pp. 77-91.
48. American Psychiatric Association: Diagnostic and statzrtzcal manual of mental dirorden
(3rd ed, rev). Washington, DC, APA, 1987.