The weight of this sad time we must obey, speak what we feel not what we ought to say.
You who've been on the road
Chapter 3: Feeling and Thinking About Sex and Gender
Must have a code that you can live by
And so, become yourselves,
Because the past is just a goodbye.
This chapter reviews psychological research as it relates to transgender experience and expression. Although this chapter intends to show that the establishment clinical perspective lacks scientific grounding, I do not mean to suggest that professional counselors, including psychiatrists and psychologists, have not sometimes been helpful to transgender people. Some guidance for finding good counsel is included in Chapter 7.
I offered the reader several confessions in the first chapter so that my biases might be taken into account. I should also tell you that I'm angry at many establishment clinicians for the shoddy research work they have done, as well as for serious injuries to transgender people that have resulted from the application of their professional opinions. Many have acted with professional hubris when honesty required that they confess that they didn't know what they were doing and that they were trying to find their way. I'm thankful for the writing of Phyllis Burke, Leslie Feinberg, and Daphne Scholinski, who have shown how much damage establishment psychiatry and psychology can do to transgender people, thankful these authors make their anger so palpable and accessible.iii They've written the kind of books you can't pick up and can't put down because the hurting is so close. Chapter 7 of this book includes an extended quote from Burke to provide a taste of this kind of writing. Given their contributions, I feel free to limit the expression of my own anger in this book and to offer a different contribution: an evaluation of the scientific base of establishment psychiatry and psychology with regard to transgender experience and expression.
Although some people in the transgender community are angry at establishment clinicians, others are deeply thankful for the help they have received. Some of the early clinicians -- Harry Benjamin comes to mind -- were humanitarian in intent toward the transsexual people they were trying to help. Benjamin went out on a limb to establish the practice of sexual reconstructive surgery that has been a blessing to many transsexuals. But being humanitarian is not enough. Truth is also important. Humanitarian intent helps, but uninformed good intentions have contributed to the personal hells described by Burke, Feinberg and Scholinski.
This chapter also points out the traditional goals and values that guide clinicians. Heterosexual men and women, well differentiated from each other, are assumed to be normative. Transgender people are treated as gender and sexual transgressors. The "humanitarian" contribution of clinicians is to offer a medical response to such behavior as contrasted to vigilante or police oppression.
In criticizing the damage done by the medical model, I do not mean to suggest that the criminal justice system is better. Even a quick look at Prisoner of Gender by Katherine Johnson and Stephanie Castle reminds us how horrible prisons can be.iv Neither do I mean to suggest that institutional Christianity has a better track record. Joan of Arc should not have been burned to death by the Inquisition.
The first task for developing a better grounded psychology of transgender experience and expression is clearing away clinical misconceptions. The second task is to start reviewing and building on nonclinical psychological research. In later chapters, contributions from learning theory and from social psychology are added.
What You Can't Learn in Clinics
Psychology and psychiatry have been interested in sexual variation since the time of Freud and Jung. Vern Bullough has done an excellent job of tracing the early development of transgender concepts in psychology and psychiatry.v This chapter settles for assessing the scientific base of contemporary positions.
I offer the following true story to highlight the fundamental clinical bias. During the 1970s, when the American Psychiatric Association and the American Psychological Association were changing their collective minds about whether homosexuality was a mental illness, I had occasion to debate with a psychologist on a local television program about whether homosexuals were sick. My credential in that setting was representing a sociological point of view. The following story line captures the gist of the exchange.
The psychologist said, "They're all sick."
I said, "How do you know?"
He said, "Every homosexual who comes to see me is sick."
I said, "You're a clinical psychologist, right?"
He said, "Yes."
I said, "Do you work with heterosexual people too?"
He said, "Yes."
I said, "And do heterosexual people come to you because they are sick?"
He said, "Yes."
I said, "How come you don't conclude that all heterosexual people are sick?"
He said, "You don't understand."
A 1990 study of psychotherapists by Benedict Carey of psychotherapists suggests that they are as prone to gender stereotyping as the general public. The study asked for clinical evaluations of videotapes of a depressed client. The story line was the same for two videotapes except that in one the client was an engineer with a wife at home and in the other his wife was an engineer and he was at home. Those who saw the tape in which the client was at home rated him as far more severely depressed than those who saw the tape with the wife at home.vi Although such biases are important, this chapter focuses instead on four scientific problems commonly found in research done in clinical settings.
Four Common Problems in Clinical Research
The first scientific problem has to do with who is seen and when. Reasoning in clinical studies is based on observations about people who are having problems during the period in their lives when they are having the problems. The selectivity error occurs when the findings are generalized to people who are not having troubles, to people who do not fit the profiles of those who are seen in therapy, and even to the lives of the people with transgender problems at the points in their lives when things are fine.
A second scientific problem concerns what is seen. Clinicians try to help the persons who seek out their services. This leads to the bias of looking for problems rather than to general observation. As a result, when writing about transgender people, clinicians try to explain the problems of transgender people – to explain why they hold a transgender conception of themselves that causes them pain when they are subjected to public ridicule or abuse. Locating the problem within the transgender person rather than in social reactions is a powerful reinforcement of the cultural status quo. Looking on transgender experience as a problem blinds one to an appreciation of all that is going on and misses the point that a lot of transgender people like their lives, like the transgender aspects of their lives.
Data corruption is a third general problem in clinical studies. It's not that clinical studies are unique in having the problem of data corruption. It's that the clinical process by its very nature is corrupting. Not only do clinicians have an interest in seeing some factors while being inattentive to others; clinicians corrupt the observation process by trying to change the people they are seeing.
The fourth general scientific problem is the use of revisionist memory as a source of scientific data. Adults are asked to reconstruct earlier events and feelings in response to leading questions. This is particularly a problem for adults who are asked to remember their childhood. In such an encounter, the clinician and the patient are jointly trying to make sense of why the patient is feeling what the patient is feeling. This is about as far away as one can get from objectivity.
Within the United States, the development of establishment clinical thinking can be traced to two overlapping points of view, represented by Harry Benjamin and George Rekers. In 1966, Benjamin published the first scholarly book reviewing the case records of people seeking sexual reassignment surgery. Although Benjamin opposed the notion that transsexualism was a disease that psychotherapy could heal, he needed a disease diagnosis to justify treatment by hormones and surgery. George Rekers, a Harvard-trained clinical psychologist who led a major program at UCLA, represented the view that transgender experience and expression is some kind of psychological disorder that might be healed with psychological interventions. Rekers began reporting in the early 1970s on his "therapy" with children that he defined as having sexual identity problems. His research is much quoted and is used in some training programs.vii Although there is ongoing debate about the weight of physiological and social causes and about the value of different therapies, Benjamin, Rekers, and the establishment clinicians who have followed them assume there is something wrong that needs correcting and that doctors should be in charge of the correcting.
Rekers' methodology shows three of the common flaws found in clinical research. One example of his research is a case report on a girl named "Becky" who was brought in for therapy by her mother because she refused to wear a dress and act like a girl. Rekers theorized that masculine acting girls grow up to become homosexual, transsexual, neurotic or schizophrenic and are likely to become alcoholics, have bad work records, or commit suicide. Now let's consider what he actually did in his research and what it showed.
Rekers administered several psychological tests, including the IT scale, that presumed to measure Becky's masculinity or femininity. They found out she didn't like to wear dresses and didn't want to act like a girl. Her parents believed this was a serious problem, and so did Rekers. So they tried to change Becky. Becky was subjected to an intense series of clinical interventions in the clinic and at home aimed at trying to get her to change her play habits, her appearance, and her interactional style so that she would appear more feminine. One aspect of the treatment was to make Becky wear a microphone in her ear so that her play could be directed by a psychologist sitting behind a one-way mirror. Then they repeated the psychological measures to see if her scores changed. They did.viii
What the Rekers' case study proves is that if parents join up with a battery of professionals and put an enormous amount of pressure on an 8 year old to act more like a girl they can sometimes force some outward conformity. Or maybe they can. During followup testing 14 weeks after Becky was pronounced cured, she scored at the highest possible level of masculinity on the IT scale. Rekers nonetheless claimed success because he decided that the IT scale, which he had previously affirmed, was probably invalid.ix
In evaluating the case study of Becky, it is easy to see clinical biases. First of all Becky was brought in because her parents didn't like her behavior and they could not make her change. Becky, not her parents, was defined as having the "problem." Instead of noticing that he was studying transgender expression in a child with transphobic parents, Rekers wrote as if he were studying one example of any child who gives off transgender expression.
The case report gives limited information about Becky. It reads more like the presentation of a lawyer building a case against a criminal than like an objective style of reporting. She may well have been a terrific kid. She certainly had ego strength. Most importantly, we are not presented with a theory about why Becky liked to play and dress as she did, so there is no way to evaluate a causal theory in terms of data. We also don't know how she turned out.
Rekers' team and the parents didn't merely observe Becky, they brought enormous pressure to bear on her. It is hard to imagine a more data-corrupting research design. The real research is not about understanding Becky's career as a tomboy, but is rather about whether her parents could team up the clinicians to bully her into acting like a traditional girl.
The Research of Richard Green
If Rekers' work models the worst of clinical research, it is only fair to review the best and most quoted study. The work of Richard Green stands alone in its effort to overcome some of the most scientifically troublesome aspects of clinical research. He published two books on one study of feminine behavior in boys and men. The second was published in 1987 as The "Sissy Boy Syndrome" and the Development of Homosexuality. Richard Green follows in the line of research that includes Harry Benjamin, George Rekers, John Money, and Robert Stoller. Like these other establishment clinicians, Green was favored with big grants from the National Institute of Mental Health and other sources. The work of Green and his mentors is broadly quoted in other sources and is a major basis for the definitions found in the Diagnostic and Statistical Manual of Mental Disorders (DSM) of the American Psychiatric Association.x These DSM standards became the "expert opinion" that is supposed to guide clinical treatment and is relevant for legal actions related to transgender experience and expression.
Green's book deserves its prominence primarily because he followed his research subjects over 15 years and measured outcomes in adulthood. Green's work is also better than other clinical studies for several additional reasons, including a serious effort to combat the effects of bias in subject selection. But Green's work is still clinical research and is reported as summarized and edited clinical reports designed to illustrate various themes and issues. His study focused on 66 "feminine" boys. In addition to the 12 boys who received intensive clinical attention, there were two control groups: (1) the 54 boys who presented some of the same initial behavior but did not receive heavy clinical intervention, and (2) a group of average boys. Unfortunately, the book tells us very little about the control groups and instead mostly follows the 12 who had therapy.
Green obtained the 66 "feminine" boys by advertising in newspapers and the like. This approach repeats the who is seen selectivity problems. The parents brought in their children because at least one of the parents was concerned about, or rejecting of, the child's behavior. The 12 children who received therapy were the 12 whose parents agreed to therapy. Since the book provides very little information about the control group, the selectivity problem is magnified rather than reduced in Green's reporting. For example, I find it fascinating that most of the parents rejected therapy for their boys, and I wish we knew more about that.
As in other clinical studies, Green did not study the everyday childhood behavior of cross-gender play, cross-gender dressing, cross-gender toy selection, or cross-gender identity. He studied clinical relationships with children whose parents responded to a prenamed label – children who had to contend with distressed and disapproving parents.
One of the few bits of information reported for the whole cohort of 66 boys is that, when initially referred, "The ‘feminine' boys had few behavioral problems other than sexual identity conflict."xi The children were happy with their play and otherwise normal. This information is dramatically opposed to a clinical picture of disease or dysphoria. The "problem" that got them into the study was the distress of the parents. This finding, by itself, justifies the wiping away of the concept of Gender Identity Disorder from the DSM. Green was well aware of the significance of this issue when he wrote, "the parent's decision to enter our study may have had a profound impact on the son's sexual identity development. (Parental concern)... may be the most marked ‘intervention' into the boy's emerging sexual identity."xii
It appears that the most potentially helpful parts of Green's research were not reported. What was the gender play and experience of the average boys? All we get is a comment by Green that a lot of cross-gender behavior is shown by boys who are not unhappy being boys.xiii Assuming that this comment is based at least in part on the control group in his study, it makes all the more critical the selectivity processes of who is defined as a feminine boy, and who is selected for clinical attention. These two side comments by Green, that the "feminine" boys were happy and otherwise normal and that "average" boys engage in a lot of cross-gender play, opens more of a window on everyday cross-gender expression in boys than does the 99% of Green's reporting that focuses on the 12 case studies. In terms of overall scientific weight, it should be noted that these side comments are powerful evidence against the establishment clinical perspective.
Green has a lot of problems with his definitions and they affect what he sees and reports. Green's definitions are not merely inexact. They show systematic misunderstanding of transgender experience and expression. Green's concept of sexual identity is the key concept in his book. What he actually means is gender identity, and he slides back and forth in naming the concept.
For Green, there are three components of sexual (gender) identity. The first is core morphologic identity, which amounts to "a person's identity as a male or female." Green is talking about body awareness here. The second is variously named as gender role behavior, or sex-typed behavior, or masculinity and femininity. The terms refer to "the culturally fixed signals that discriminate males from females." Green is talking about the degree to which an individual accepts the typical social expectations for gender role behavior. The third is sexual orientation, or sexual partner preference, or sexual object choice, which amounts to the distinctions between heterosexuality, bisexuality, and homosexuality.xiv
Since it is at the heart of the subject of his book, Green's sloppiness in his use of the language of sex and gender is surprising. It is important to understand that a person's identity is part of the subject matter of psychology. Awareness of being physiologically male or female is only one important reference for gender self-understanding. The challenge to explain is why some (transgender) people have gender experience that is not in accord with cultural expectations attached to physiological signals of being male or female. Such discontinuity often leads transgender people to wonder whether there might be interior physiological realities that are not in accord with what is visible. Additionally, some people have visible physiological features that are easily seen as not fitting the physiological stereotypes of their assigned genders. The first key to unlocking an understanding of transgender experience is simply facing up to the fact that not all females have a simple psychological identity as women, nor do all males as men. As a student of transsexuals, Green is well aware of this point. Nonetheless, in speaking of "core morphologic identity" he makes it clear that he is refusing to take this distinction seriously.
This profound definitional mistake confuses what is to be explained. Green sets out to explain why some "feminine boys" engage in cross-gender expression and then to examine the relevance of such expression for the child's gender career. Although Green emphasizes the significance of "core morphological identity," he does not argue that these "feminine boys" were unaware of, or denying of, their physiology. The definitional failure suggests that Green doesn't appreciate the core psychological truth these boys were actually presenting. They apparently didn't much care that their imagery and play didn't line up with the expectations of others. Maybe they didn't even know they had transgressed. The psychological truth appears to be that there was no problem in the boys from the point of view of the boys. It seems that the purpose of the therapy, for those "lucky" enough to gain its "blessings," was to teach the boys that they should think of themselves as having a problem so that the parents and therapist could correct it.
A second relevant definitional failure by Green is his assertion that the three definitional elements he identified are distinct and vary independently. Green defines a heterosexual as being culturally typical in all three elements of sexual identity. A homosexual is defined as atypical only in sexual partner preference. A transvestite is defined as atypical only in gender role behavior. A transsexual is defined as atypical in all three elements. This sharp differentiation, which is common in the work of establishment clinicians, misses a lot. First of all, it misses homosexuals who are effeminate or who cross-dress. This is a strange failure, since the main logic of Green's book is to reason from cross-gender identity in children to homosexuality in adults. Once again, Green was aware of the issue. He quoted a survey of 1,500 homosexuals that reports high levels of cross-gender behavior.xv Green's definitional approach also misses the critical point that the gender role behavior of a transsexual matches that transsexual's sense of gender identity. Furthermore, Green fails to understand that transsexuals may choose sexual partners of either sex or gender. These several definitional failures point to the enormous distorting power of the clinical perspective which can overwhelm scientific judgment. It also helps us understand why the clinical perspective is so impervious to scientific correction. Green failed to let his own research correct his own perspectives.
The third problem I see in Green's definitions is his confusion of the concepts of roles, behavior, and cultural images, as found in his second definitional component. Roles, behavior, and images may be congruent for a lot of people, which is probably why Green skipped past such distinctions. But such distinctions are critical in transgender studies for deciding who is counted in which category. The distinctions are also critical to the writing of good theory, because so much theorizing about causation differentiates these concepts. As we shall see in the following chapter on sociological research, the number of people who are included in or excluded from the category homosexual, as found in sample surveys, varies enormously, depending on whether one is measuring role identity, feelings, or behavior.
Green's fourth definitional problem lies in the phrase "culturally fixed signals." Research about cultural definitions suggests that they are not fixed or consistent.xvi This definition also misses issues of subculture variation and family variation. For example, Green argues that parents of feminine boys did not provide appropriate signals to their children about masculinity. The rigidity of this definition by Green – who apparently thinks his sense of culture is definitive for everyone – makes him blind to the variability under his clinical nose. In short, Green used a simplistic stereotype of masculine and feminine behavior, called it traditional (culturally fixed), and then weighed his subjects against it. The same data might be reported as showing that the cultural images are not so fixed or consistent and that "feminine boys" might be conforming rather than transgressing when measured against subcultural or family variations. For example, Green refers to the avoidance of "rough-and-tumble play" by his "sissy boys," with the untested presumption that this is a simple violation of culture. But, if a boy grows up in a setting where rough-and-tumble play is restricted by parents, teachers and other adults, a circumstance that may be common, it would be inaccurate to say the boy is going against culturally fixed signals. We need some solid research on the degree to which cultural signals about gender may be ambivalent and ambiguous for many children.
The fifth definitional problem in Green's book is the overall simplicity of his definition as weighed against the complexity that actually shows up in transgender adults. Such complexity was summarized in the first chapter of this book.
Green's study also shares the third general failing of most clinical research, the data-corrupting effects of his clinical interventions. As a clinician he had hoped to cure the feminine boys. He barely reports that his clinical interventions did not have their intended curative effect. Indeed, though it is counterintuitive, Green's therapy may have had the opposite effect. Social psychology labeling theory shows that attaching a label to someone can be a powerful cause of a person's choosing to exhibit behavior in keeping with the assigned label.xvii Phyllis Burke summarizes this point for Green's book by arguing that he was teaching the boys that they were "pre-homosexual."xviii In Chapter 7 I will present a more complete view of how clinical interventions are social control processes that sometimes produce an effect opposite from what is intended.
The fourth common problem in clinical studies is revisionist memory. Green specifically names it as a problem in other clinical studies and correctly points out the value of getting data during childhood, then getting follow-up data during adulthood.xix He further notes that transsexual adults are aware of the rules used by psychiatric and psychological gatekeepers when transsexuals apply for sexual reconstructive surgery (SRS). Green suggests that such applicants are likely to tailor their stories to gain approval for surgery.
Although Green's own study is protected against the critique of relying on revisionist memory, he refers to other studies that are not so protected for some of his strongest argumentation. Green asserts that "This linkage between boyhood femininity and manhood homosexuality has long been suspected. Not only do the retrospective data ...suggest it, but a few prospective studies document it."xx He cites four such prospective studies, one of them a preliminary study he did with 5 "feminine" boys. The other three studies had groups of 13, 16, and 55 boys and were all follow-up clinical studies. The two smaller studies found that the boys were more likely to grow into heterosexual adults, with traditional gender expression, than into homosexual, transsexual or transvestite adults combined. The study of 55 boys had extra methodological problems. For example, 16 of the boys entered the study as adolescents. It found that two-thirds of the boys became homosexuals.
Despite the numerous methodological and theoretical problems in Green's work, it is still interesting to consider several of his research findings. He reports that 9 of the 12 boys who received treatment grew up to have homosexual or bisexual experience as young men. The proportion was similar for the untreated 54 boys, which led him to conclude that therapy with feminine boys was not likely to stop them from growing up to be homosexual.xxi In considering the relevance of Green's study for transgender experience and expression, his finding is devastating for the clinical establishment's theory of gender identity disorder. The "feminine boys" in the untreated sample, with one exception, did not grow up to be transvestites or transsexuals.xxii All the gender identity clinics that have blossomed to treat boys who express feminine behavior are supposed to keep them from growing up to be transvestites or transsexuals. This key research does not justify such an industry, nor does any other. Even if you believe that such clinics are really aimed at preventing homosexuality – which is no longer professionally justified because it isn't seen as an illness – Green's finding is that such therapy is ineffective.
To keep the record clear, for reasons already discussed and for other reasons to follow, I do not believe that Green's research, or any of the other research reviewed in his book, should be counted as scientific evidence for the theory that feminine behavior in boys causes, or is an early indication of, homosexual or bisexual behavior in men. The point made above is merely that the research findings oppose the theory that feminine behavior in boys causes transgender experience or expression in men.
In addition to the common research problems in clinical studies as they relate to Green's work and in addition to the relevance of his research findings in opposition to clinical theories of transgender experience and expression, there are still more major scientific problems found in The "Sissy Boy Syndrome" and the Development of Homosexuality which additionally show the limits of clinical thinking and further show that the clinical label of gender identity disorder is unjustified.
Richard Green's study isn't science for the simple reason that he doesn't actually specify and test a causal theory. What he provides instead is clinical reflection. There are numerous implicit causes in Green's book, but the failures to define them, to theorize there activity ahead of time, and to specify the research measures of them, makes this work non-scientific. In his concluding chapter, Green talks about causes in a way that ducks and dodges any commitments. Green couldn't even find a causal path after reviewing his findings. Instead he repeats the findings of his original (1974) study of these "feminine" boys: "The disparate patterns presented ...make it clear that no consistent etiologic pattern exists for extensive boyhood femininity."xxiii
Although Green notes his theoretical problems, including a comment that his theory may be more artistic than scientific, and although he notes that his findings do not justify a theoretical statement, he nonetheless proposes an eight-point developmental theory – because, he says, "I am not congenial to reducing all behavioral processes to the ebb and flow of neurotransmitters."xxiv His theoretical comments include an unspecified role for predisposing physiological factors, such as hormonal effects in prenatal development, but Green gives primary weight to social factors such as interaction with parents and peers. His strongest emphasis is on what he labels "male affect starvation," which flows, he believes, from alienation of the feminine boy from his father and male peers.xxv But he notes later that male affect starvation could be resolved in many ways other than cross-gender behavior.xxvi
My comment that Green didn't test a theory isn't quite accurate. Although his book is not presented as a test of a theory, he does tell us that he switched theories in the middle of his study. Green reports this switch as if it hardly mattered. He reports that he started out assuming that he was studying pre-transsexuality in children. But when the adult follow-up study was done, he found much more homosexuality than transsexuality. No problem, says Green, because homosexuals are more likely to report cross-gender play, etc., as children. Such slipperiness is easy for Green because he isn't specific about his causal theories in the first place and doesn't seem to care about the distinction between transsexuality and homosexuality, since he says both are "atypical." The bottom line is that Green was so busy observing the "shouldn't" rather than the "is" that he never got around to the question of why the boys liked girl stuff.
Although Green switched theories when his follow-up study didn't fit his original theory, and despite the fact that his findings don't support the theory that feminine behavior in boys causes transsexual or transvestite behavior in men, he still tries to make a case for his original line of thinking based on the research of others. His strongest argument that others have proved his point of view is his summary of Harry Benjamin's reports on hundreds of cases of transsexuality. Of course all of Benjamin's work is based on revisionist memory methodology which Green criticized as a significant problem. Green states that Benjamin found that "Males seeking transsexual surgery recalled feeling like members of the ‘opposite' sex since early childhood."xxvii
Then Green reports on his own interviews with 30 candidates seeking SRS. "My interviews of thirty males requesting medical and surgical reassignment at UCLA confirmed Benjamin's report."xxviii Green continues by reporting on two cases that support Benjamin's view. But, when Green reports his summary statistics for these interviews, it turns out that only 30 % of the 30 candidates reported a childhood self-concept as a girl, while 43 % reported a boy self-concept and 27 percent reported both boy and girl self-concepts.xxix Thus Green's own research on transsexuality is in opposition to the most basic clinical stereotype of transsexuals, a stereotype that shows up repeatedly in "learned opinion.xxx This finding by Green is all the more dramatic because Green had reported that he was aware that candidates for SRS knew they had to fit their stories into a particular story line to get the diagnosis of transsexuality so that they could then receive SRS.xxxi
For all my dissatisfaction with the scientific quality of Green's work, I actually agree with Green about the most reasonable theoretical posture for explaining the causes of transgender behavior. Green posits some possible physiological predispositions and the importance of family and peers. Similarly, Kenneth Zucker, another establishment clinician, who runs one of the largest clinics in the world dealing with gender identity disorder, names three sources of causation: physiological predisposition, learned behavior and family interaction.xxxii In Chapter 6, my own decidedly nonclinical theorizing works with the same kind of variables. There are numerous difference between the way I reason within this theoretical posture and the way they reason, between what I count as data and what they count, but the biggest difference is that instead of looking for a "needle" to explain "abnormal" behavior, I look at the "haystack" of how such causes work in everyday growth and development.
Summary and Critique of the Medical Model
Neither George Rekers, Richard Green, nor any other establishment clinician, has shown that there is any line of causation that justifies a picture of transgender experience as "abnormal," as "disease," as "dysphoric," etc. The review of the work of George Rekers and Richard Green shows that having a prestigious degree and getting big grants from the National Institute of Mental Health doesn't make you a scientist. You have to do work that is consistent with the elementary tenets of science before that work deserves the prestige of having the word science attached to it. Instead of thinking of George Rekers as a scientist, try thinking of him as a cultural bully manipulating a little girl against her will. It is ugly behavior. It was probably damaging to Becky. Although Richard Green's work has scientific problems, at least he had the guts to criticize Rekers in print for his "moralistic" biases, and for calling homosexuality "an unfortunate perversion" and "promiscuous and perverted sexual behavior."xxxiii Green reports that he supported removing homosexuality from the list of disorders in the Diagnostic and Statistical Manual of the American Psychiatric Association (DSM). He has been instrumental in the recent relaxing of some inappropriate standards in the Benjamin guidelines for medical treatment of transsexuals.xxxiv Perhaps he will someday claim the implications of his own research and oppose the concept of gender identity disorder.
The establishment clinical model, whether hostile to transgender experience and expression, as in the work of Rekers, or humanitarian, as in the work of Benjamin and Green, is based on the classic medical model and is employed first of all for medical purposes. The medical model works pretty well for a host of diseases. For example, a particular strain of flu may be identified as having a specific cause (a strain of flu virus), a common course (say headache followed by digestive upset followed by 3 days of weakness), and an expected outcome (say a 99.4 % recovery rate by the fifth day). An accurate medical model can be very helpful for treating an identified disease. Of course the experience of a specific person with this flu may not match the model: An older and weaker person may die. Such small variations can be handled with a few modest qualifications of a specific medical model for a disease.
When Benjamin applied the medical model to justify hormonal and surgical treatment of those who showed up wanting SRS he hypothesized a specific physiological cause, even if it wasn't measured or identified; a traditional course (profound feminine sense of self from a very early age, resulting in feminine expression); and an expected outcome of profound dissatisfaction and distress, often including suicide, if the condition was not relieved by hormones and surgery. This line of thinking was close enough to traditional medical thinking that Benjamin was able to establish medical clinics in the United States for treating transsexuals.
Once established as a theory, the medical model of transsexuality drew criticism from a host of medical and nonmedical sources. Despite some ups and downs in acceptance, SRS is still available and is still being justified on medical model grounds. I strongly support the availability of SRS and believe many people have profited profoundly from the surgery and that this in turn has benefitted our society. But I don't think the medical model of transsexualism is scientifically defensible – for all the reasons presented already and for the following additional reason: certainly some people's stories closely fit the classic transsexual model, initiated by Benjamin, in course and outcome. But a lot of people seeking SRS don't fit the classic story. I've already noted that Richard Green found that many people who were labeled as males at birth and sought SRS had nonclassic elements in their stories. Here I want to point out that in the mid-1970s the concept of secondary transsexualism began to arise.xxxv
The concept of secondary transsexualism is applied to people who had been labeled as males at birth and were older than forty when they began to seek SRS. They reported successful career as boys and men but nonetheless wanted to experience life as women, in keeping with their fantasy experiences of themselves as women. These secondary transsexuals were commonly husbands and fathers and reported themselves as occupationally successful. Since I share a lot of such a story line, even though I do not expect to seek SRS, I don't have any problem recognizing that the story can accurately reflect the experience of some people. But such experience is far from the classic clinical course of transsexualism. The only point of agreement is a strong desire to receive SRS.
Once transsexualism was medically accepted as some kind of disease or dysphoria it has apparently been an easy step to assume there are different kinds of transsexualism. But the modification of the medical model of transsexualism is not a small adjustment to a basic model, like accounting for the variable of age in a medical model of flu. The story lines are so markedly different as to require a completely different approach to theorization. Instead of stretching the medical model of transsexualism, a more reasonable assessment is simply that people who want SRS have vastly different story lines and that the cause and course elements of the medical model, at minimum are not defensible. Of course, those who are committed to medical model thinking can simply decide there are two diseases that require two medical models. Since I've already demonstrated the complexity of transgender experience, I suggest that following this path gets you to a medical model at least as complex as cancer or the common cold - dozens of causes, courses, and outcomes. Those who want to stay with the medical model are going to find themselves saying things that are analogous to, "Most men over 50 have prostate cancer but are non-symptomatic and the cancer is irrelevant to their general health and life-expectancy."
The above critique of the medical model to explain transgender experience and expression, is far more profound than noting a few exceptions to a general rule. Clinicians have not established any of the three components required for a medical model theory for transsexual and transgender expression: neither cause, nor course, nor outcome.
The concept of secondary transsexualism has taken an additional twist with the construction of the concept of autogynephilia, which was initially developed by Ray Blanchard in 1985xxxvi and more recently promoted by Anne Lawrence in 1998xxxvii. Blanchard is a clinical psychologist with the Clarke Institute of Psychiatry in Toronto; Lawrence is an M.D. and a transsexual. Lawrence follows Blanchard in conceiving of autogynephilia as a paraphilia – a love or attachment not directed toward another person. The supposed cause of autogynephilia is masturbation fantasies in which the transgender person imagines being a woman. Lawrence's argument is that only sexual energy is powerful enough to cause a male who is reasonably comfortable in the man's role to be willing to go through SRS. But there are two important lines of evidence which don't fit the concept of paraphilia for so-called secondary transsexuals with autogynephilia. First of all, these supposedly autogynephilic people do report erotic attachment and interaction with other people, even though they sustain erotic fantasies about being a woman. Second,, Lawrence reports that these supposed autogynephiles want SRS even after there is no longer any erotic fantasy. She accounts for this desire as commitment to a pleasant memory.
Beyond the issues of weak evidence, we return once again to core problems with clinical definitions. Masturbation fantasies have long been treated as signs of disease in psychiatric classification manuals and have a cultural history of religious condemnationxxxviii. There is a lot of debate about what fantasies are and about what they mean, but the core idea is that a fantasy is imagery linked to an unfulfilled desire or other strong feeling. Assuming that there are people who have been defined as male and as a man who get a lot of pleasant feelings in masturbation fantasies from imagining themselves as being female and a woman, we should ask what is the unfulfilled desire that is at stake.
Why might a man enjoy imaging himself as a woman? The cultural imagery of a woman in the United States commonly includes such things as the ability to feel one's feelings and the ability to receive love and attention. If a man has been socialized in ways that starve access to his own feelings, pursuing a fantasy that opens up such access might best be seen as a movement toward centeredness and wholeness, toward health rather than disease. Then the question for such a man becomes what he wants to do to pursue access to feelings. For such men the following question becomes what to do to give and receive love more easily. There are lots of ways of reforming one's sense of self to incorporate such insight. One of those paths is to seek SRS. This interpretation is the opposite of paraphilia, the opposite of a medical model that assumes a disease has to be explained. Even when masturbation fantasies are part of the story line, it may be important to give primary weight to factors other that sexual feelings to assess what makes SRS attractive. Even if a person is fully identified by the story line that Blanchard and Lawrence propose, if SRS helps one gain greater access to one's erotic feelings, then it may well become part of learning to share a more complete, more satisfying, and more loving interchange with another person. The proper label would still be health.
Nonclinical Psychological Studies
Vern and Bonnie Bullough have offered an extended nonclinical psychological explanation of transgender experience and expression. Although well aware of clinical writing, the Bulloughs have drawn primarily upon historical and cross-cultural studies for the core of their theory. Taken together, their several recent publications constitute a magnum opus that will influence work on transgender studies for a long time. Their theoretical summary can be found in Chapter 13 of their 1993 book Cross Dressing, Sex and Gender.
The Bulloughs theorize the significance of both physiological and social-psychological causes. Their reflections on physiological studies is brief, assumes an admittedly unmeasured genetic predisposition, and emphasizes the untested theory that prenatal hormone abnormalities create abnormal brain development. They refer positively to Pillard and Weinrich, who develop a detailed theory of prenatal hormone influence to explain transgender and homosexual expression.xxxix However, when the Bulloughs state their own physiological assumptions, they treat physiological causes as vague and general predispositions to gender transgression that are steered by psychosocial events in the child's life. Such generality is in keeping with their historical and cross-cultural studies, which found wide variation in the ways that gender transgressions have been expressed. It is also in keeping with the finding that feminine boys grow up to become heterosexual, homosexual, transvestite, or transsexual.xl In short, it seems to me, they want to include a physiological dimension of cause because transgender expression is found in so many places and times. But they do not integrate physiological causation into their essentially psychosocial explanations.
The Bulloughs present their theory of transgender development as a six-step version of a sociological "career of deviance" theory first presented by H. Taylor Buckner. Buckner collaborated in a survey of transvestites by Bentler and Prince that is discussed in a later chapter.
The first step presumes genetic and physiological factors that produce a boy who is less active and aggressive. This assumption doesn't suggest a needle theory such as prenatal sex hormones, but rather a haystack theory that draws upon complex, interactive, flexible genetic and physiological factors. But this first step has the same problem found in Green's study of feminine boys: boyhood femininity does not lead to adult transgender activity. Furthermore, many transvestites, including me, report an active childhood. Some clinicians, showing their mental "flexibility," call such behavior hypermasculine. Myself, I just enjoyed playing high school varsity basketball and baseball.
The second theoretical step for the Bulloughs lifts up the effects of a wide range of social factors, such as a dominant mother. I agree that many kinds of social causes come into play; I just develop them differently. I would also note that I had a mother who majored in emotional passivity and withdrawal. We seldom touched, but she did at least shake my hand when I went off to basic training in the army at age 17. (Lest anyone think that being in the army was hypermasculinity, I should point out that I was choosing 6 months of active duty training plus reserve duty instead of possibly being drafted for 2 years.)
The third step proposed by the Bulloughs is the discovery of the joys of cross-dressing and masturbation. No one has systematically studied masturbation and cross-dressing, but one main point is that the fantasy content of masturbation is not limited by the clothes one wears. Any good study in this area would have to focus on the fantasy or imagination content rather than on what one is wearing while masturbating. For myself, I think my adolescent fantasies were heavily influenced by wanting to escape the exploitative imagery of men that dominated my subculture. I also remember that I wanted to experience the full range of my feelings and that I wanted to develop a meaningful emotional relationship with a woman. I wanted the girls I dated to know I wasn't like John Wayne, and that helped me have a lot of dates.
The fourth step for the Bulloughs is marriage and the suppression of homosexuality. With regard to this element of the theory, it seems relevant to me that homosexuality is suppressed by our homophobic culture at every point in the life cycle, not just marriage. Furthermore, there are people who cross-dress who are at every point on the continuum between heterosexuality and homosexuality. It seems to me that the problem with the Bullough's theory at this point is that they don't give sufficient weight to their own broad historical and cross-cultural studies but were drawn back toward a clinical perspective that transvestites are overwhelmingly heterosexual.xli
The fifth step that the Bulloughs mention is the response of wives or partners. The Bullough's show their theoretical flexibility by arguing that whether the wife or partner supports, resists or leaves, her action encourages transvestite development. No one has to bother testing this element of the theory since any data will do. I think that partners may support or oppose transgender development but that partner influence is not a primary cause of transgender experience. I also think that many transgender people weigh the costs of transgender expression and that their weighted choices help to shape their careers.
The Bulloughs base the sixth step of their theory on learning theory when they argue that transvestites learn their scripts from transgender support groups before they go public. While I agree with the Bullough's emphasis on learning theory, I need to point out that times have changed. Though transgender support groups might have supported only a single story line in the past, many support groups now include people with great diversity of story lines and social goals.
It should also be pointed out that the Bulloughs focused their theory only on man-to-woman transitions.
Despite my dissatisfaction with their summary theorizing, there is much to affirm in the writing of the Bulloughs. They have made a lot of transgender history and culture available to the transgender community. Most important, they are able to talk about gender transgression without naming it as mental illness. Instead, they write, "The mental illness label ...stigmatizes and lessens human freedom. Since many transsexuals also are very depressed and want sexual reassignment surgery, the illness label, with all its disadvantages, may be reasonable; but in the case of transvestites, who seldom want or need treatment, the label probably only effectively stigmatizes behavior."xlii
Men and Women Are Alike and Different
How far does one trans in transgender experience and expression? How great is the distance between men and women?
Comparing men and women has been a popular activity of learned studies in Western civilization for centuries. Jill Matus lifts up a report by physiologist William Carpenter in 1842 who described a male of African descent, as a "perfect" male with complete genitalia, who had nonetheless developed women's breasts and made his living as a wet nurse."xliii
The feminist revolution, and the antifeminist reaction, have inspired a lot of recent work on differences and similarities between men and women. Some of the scientific work, on both sides, is politicized, which justifies an extra dose of skepticism in relation to scientific value. A lot of the work is simplistic: Pick any psychological attribute. Construct an interview schedule or questionnaire. Give it to a couple of college classes, or print it in a popular magazine, and ask for responses. Publish the results. More substantial work, however, has been done, such as spotting gender bias in standardized tests.
Janet Hyde offers a feminist perspective in a chapter on the differences between men and women in her text, Understanding Human Sexuality, Third Edition (1986). Her conclusion is that, with the exceptions of aggressiveness and self-esteem, "for the vast majority of psychological characteristics, there are no gender differences."xliv I see a more qualified picture of such research that Hyde. I would summarize by saying that similarity outweighs dissimilarity. Part of my caution flows from my judgment that so many of the studies of sexual differentiation are so seriously flawed that the main pictorial quality is muddiness. The most pervasive problem of these studies is that differences are presented as differences in the averages of measures of men and women. These studies commonly provide data that is distributed in the shape of overlapping normal (bell-shaped) curves of the scores of men and women. Such studies, even if they were otherwise methodologically adequate, show large amounts of similarity, even if there is a statistically significant difference between average scores for men and women.
A widely quoted antifeminist book is Brain Sex: The Real Difference Between Men and Women, by Anne Moir and David Jessel.xlv The attractiveness of the book may be that it is written with lots of quotable phrases, which show the writing skills of a television producer and a television journalist. The book is hard to assess on scientific grounds because there are no footnotes or endnotes to link the text to referenced studies. Following are two samples of Moir and Jessel rhetoric: "Men are different from women. ...To maintain that they are the same in aptitude, skill or behavior is to build a society based on a biological and scientific lie. The sexes are different because their brains are different. [the brain] is differently constructed [and]processes information in a different way."xlvi "If women have reason to rage, it is not because science has set at naught their hard-won struggle toward equality; their wrath should rather be directed at those who have sought to misdirect and deny them their very essence."xlvii
Moir and Jessel spend a great deal of their time on reporting measured differences between men and women and then in slopping together the explanations. For example, "The truth is that for virtually our entire tenancy on the planet, we have been a sexist species. Our biology assigned separate functions to the male and female of Homo Sapiens. Our evolution strengthened and refined these differences. Our civilization reflected them. Our religion and education reinforced them."xlviii This is like saying that because women breast feed they must do the cooking and the laundry. For the sake of brevity I will point to only three of their common scientific errors. These same errors show up in even more popularized defenses of sexual differentiation such as He and She: 60 Significant Differences Between Men and Women.xlix These errors are in addition to the already noted problem of comparing averages and suppressing the similarity shown by the overlap of normal distributions.
The first error of Moir and Jessel is their misunderstanding of the methodology of constructing standardized tests. The first evidence they cite for their conclusions is that men typically outscored women in the early Wechsler IQ tests. They then criticize Wechsler and others for substituting male-slanted and female-slanted items to arrive at approximately equal scores. "It is an odd way of conducting a scientific study; if you don't like the result you get from an experiment, you fix the data to produce a more palatable conclusion."l Wechsler was not conducting an experiment in sexual differentiation. He was trying to create a research instrument that accurately measured several kinds of intelligence. Indeed, it is only when a research instrument is free of gender bias in its construction that the results of respondents, differentiated by gender category, can be fairly compared. Biases have been shown in intelligence-testing instruments for a wide range of variables, particularly race, class, and culture. Since there are cultural differences between men and women, it is not surprising that cultural factors would affect scores that are intended to be measures of innate difference. Wechsler was not fixing his test to produce equality, he was fixing his test to reduce cultural biases that were already there, which were called to his attention by the sex differentials. With a more gender-neutral list of questions the tests showed more similar scores. That is, a fairer test showed more equivalent scores. Wechsler did not keep adding questions biased in favor of women until he got to equality.
The second basic scientific error of Moir and Jessel is that they discount the cultural effects of practice. They write, "On measurement of various aptitude tests, the differences between the sexes in average scores on these tests can be as much as 25 percent. A difference of as little as 5 percent has been found to have a marked impact on the occupation or activities at which men or women will, on average, excel."li They note that the biggest of these differences show up in measures of spatial ability. "The fact of the male's superiority in spatial ability is not in dispute. It is confirmed by literally hundreds of studies."lii They then point out that the typical test of spatial ability is the assembly of three-dimensional mechanical apparatus. They offer no comment about cultural differences in experience with assembling mechanical apparatus. For example, when I went to high school I joined all the other boys in a shop class while all the girls took home economics. This second error is about the issue of practice as it affects test scores. If a difference is innate, it should not be affected greatly by practice. Alternatively stated, if boys start out with more practice, on average, in assembling mechanical apparatus, you would expect them to be better at such activity initially. If girls have similar abilities, you would expect their scores after practice to be like those of the boys. For example, in a study of recognizing "embedded figures," boys did better in the initial testing. This advantage disappeared when the girls got to practice with the blocks used to measure "embedded figures.".liii
The third basic scientific error of Moir and Jessel, belying the title of their book, is that they make no effort to show a direct link between specific findings in aptitude differences and underlying structures in the brain. In separate sections Moir and Jessel talk about genetic differences, differences in early physiological development, etc. They assume that physiological differences cause what they count as differences in aptitudes. For example, they make a standard genetic argument in pointing out that males and females have genetic differences that are part of every cell in the body, the difference between the XX and XY versions of the 23rd. chromosome. They conveniently fail to point out that 22 chromosomes are similar and that differences in mathematical or spatial intelligence are probably not related to differences in the 23rd chromosome.liv Similarly, they argue the importance of the difference between male and female hormones and point out that these hormones also affect brains. They then point out that a burst of testosterone sets off puberty in boys while failing to point out that a burst of testosterone sets off puberty in girls. As argued in the previous chapter of this book, the point is not to deny physiological differences between males and females. Moir and Jessel could have argued more effectively for physiological difference than they did. The key question is, what are the differences and what are the effects of the differences? I offered a review of research on brain anatomy that is based on better and more recent research than that quoted by Moir and Jessel and concluded that such research shows far more similarity than dissimilarity between the brains of men and women. More importantly, Moir and Jessel are reasoning in a way no brain anatomist would reason: that a difference in a performance measure necessarily shows a genetically based difference in brain structure.
Moir and Jessel also make the error of generalizing from the study of rat brains to human brains. They report the work of Roger Gorski and others without troubling themselves with the numerous caveats that the scientists made in interpreting their own studies.lv
Moir and Jessel also refer to some intersexual case studies that point to the importance of hormonal activity. But these studies do not point to typical XX versus XY differentiation, nor do they measure the factors Moir and Jessel are considering.lvi
Another book that believes in marked sexual differentiation between men and women is Robert Pool's Eve's Rib: Searching for the Biological Basis of Sex Differences (1994). It includes a number of arguments based on intersexuality and transsexuality.
Pool argues that transsexuality is not learned. His primary argument is the refutation of a (now infamous) study of one boy that was initially widely quoted in favor of the theory that transsexuality is caused by social learning. The boy was born as a standard XY chromosome male whose penis was damaged in an accident during circumcision. The parents decided on sexual reconstructive surgery to mimic a female body and then gave the child female hormones and raised the child as a girl. John Money reported this case while the child was young and claimed that the socialization to being a girl had worked. He reported that the child's mother said she was surprised at how feminine her daughter was.lvii Pool reports from a later review of the case that the child was indeed raised as a girl but wasn't comfortable. When she was told at 18 what had happened, she insisted on getting reverse sexual reconstruction surgery to become a male again.
An even later and more complete review of the story of the boy with the injured penis is offered by John Colapinto. It exposes a story of clinical mishandling and scientific overstatement that puts Money's work in a very bad light.lviii But Pool's assessment is hardly better than the work of Money. Pool doesn't bother to comment that the child was injured at 7 months, that child rearing changes occurred at 17 months, and that the child only went through partial SRS and didn't have a full vagina.lix Furthermore, the child was well aware of being special and different because of repeated visits by Money. The child didn't choose the surgery. The child was aware of not having the body of a biological girl, in part because Money pressed the child to undergo additional surgery to create a more realistic vagina. There was a great big secret floating around in this family, and the child figured it out enough to tell a school psychologist about the injured genitals.lx To assert that there was no social learning is ridiculous. To assert that an example based on a nonchosen accident is equivalent to chosen transsexual SRS is ridiculous. This famous experiment, which propelled Money to sexological stardom, is nearly useless for scientific purposes. If it proves anything, it proves how much human damage can be done by intrusive clinicians. In any case, Pool's error is common for people reviewing scientific work. The fact that one bit of research is inadequate is not evidence in favor of another theory.
Reasoning from the "natural experiment" of doing infant SRS because of a penis damaged during circumcision has taken an additional twist with the publication of a second such study, this one on a Canadian infant. Susan Bradley et al, report that a 27-year-old happily adjusted woman was the subject of SRS at the age of 7 months, a much earlier surgery than in the case reported by Money. They report the case in support of Money's theory that gender is malleable even when there are no indications of intersexuality.lxi Since Bradley and Kenneth Zucker are well known establishment partisans in psychiatric politics, leaders in the creation of the label of gender identity disorder for children, there is reason to greet this report with skepticism. Nonetheless, an argument for malleability is an argument against the theory that gender identity disorder is a physiologically caused pathology.
Pool also makes much of a study of a natural intersexual experiment that took place in the Caribbean and was reported by Julianne Imperato-McGinley in 1974. Because of inbreeding on an island there was a group of 18 people who all had a genetic problem that created a lack of the 5 alpha-reductase enzyme. These people had XY chromosomes but were born with no signs of penis or testes because of the missing enzyme. They were raised as girls. At puberty, undescended testes began to function, and the external genitals became masculine. 17 of the 18 made then made adjustments to being boys. Despite the lack of a penis or visible testes, these people as children had male bodies, and transgender experience helps us understand that many body signals and images other than genitalia are important for the development of self-concepts. They may have been raised as girls, but that doesn't mean that they didn't notice any differences. Furthermore, the male status is valuable in Caribbean culture, as in most other places. Most important, the choice the 18 faced was not whether they wanted to change sexes but whether they wanted to change their gender to match their new penis. This is a very different kind of choice from the choice facing transsexuals. The boys chose an identity that fit. Although they were raised as girls, like all people they had learned both roles. Most people choose one gender role and stick with it. These people chose to be girls when they were small children and then chose to be boys at puberty. Perhaps the amazing part of this story is that one adolescent chose to remain a girl despite learning the truth about her anatomy. In any case, Pool makes the simplest of logical missteps. He didn't study how people do become transsexuals. He studied how people didn't become transsexual.
Paula and Jeremy Caplan offer a general critique of studies purporting to show sharply drawn differences between men and women. Such studies, they write, treat a difference as a bipolar rather than a bimodal difference and/or assume that a discovered difference is biologically based and therefore inevitable or unchangeable.lxii One of their contributions to this literature was to take on the widely held belief that boys are more aggressive than girls. First of all they note that it is not so easy to distinguish aggressiveness from the closely related concepts of violence and assertiveness and that almost all operationally developed hypotheses about aggressiveness can be reassessed and seen quite differently. But their most dramatic point from a review of aggressiveness studies is that in studies in which boys thought they were unobserved they were no different from girls in their aggressiveness, but in studies in which they thought they were observed their behavior showed more aggressiveness than did girls behavior.lxiii
I am not arguing that all cultural images of gender are totally ungrounded in biological or psychological differences. Rather, I am continuing to make the point of emergence and synthesis. Individually and collectively, we face the human task of making sense of the facets of reality that we know through our senses. Although the idea of gender seems fixed and simple to many people, this book has already shown the complexity, flexibility, and interactivity of physiological and psychological factors that affect the creation, transmission, and changing of gender categories and images.
Masculinity, Femininity and Androgyny
Sandra Bem set off an interesting line of studies of the differences between men and women in the 1970s.lxiv She began by asking college students what traits were desirable for men and women. She was operationalizing the concept of gender stereotypes. She selected 20 masculine items (e.g., "aggressiveness"), 20 feminine items (e.g., "loves children") and 20 neutral items (e.g., "happy"). She created a questionnaire with a 7 point scale for each of the 60 items. She then gave the questionnaire to several college classes and rated the respondents on masculinity and femininity. Her Bem's Sex Role Inventory (BSRI) became popular as an element in numerous studies. Her findings regarding college students have been used as a baseline for comparisons with other populations.
Bem reports that in her initial study one-third of the subjects scored as "androgynous" because they had high masculine and high feminine scores.lxv She interpreted her finding of androgyny positively: "androgynous people have the flexibility to exhibit either masculine or feminine behaviors, depending on what the situation calls for" and "those who are androgynous, and thus transcend gender roles, are better able to function effectively in a wider range of situations." For example, she suggested that androgynous males were better able than traditional men to nurture a sad classmate when that was appropriate. Given its sampling and other limitations, I think it is hard to build very much on the Bem foundation. But it does seem worth noting that in at least one kind of nonclinical population a lot of people chose not to fit into cultural stereotypes about men and women.
Hyde joined others in enthusiastic assessments of androgyny. After reviewing several BSRI studies of androgyny, she wrote, "Probably we should not set our expectations for androgyny too high. It probably will not be the cure to all the sexual woes in our society. But it might help."lxvi
June Singer, a Jungian psychoanalyst wrote Androgyny: Toward a New Theory of Sexuality. She argued for a single archetype of androgyny to replace the typical Jungian language of anima and animus. But her work nevertheless comes out sounding a lot like anima and animus, since she asserts that men and women are fundamentally different but that each carries internal imagery of the other. Surprisingly, Singer has little to say about transgender experience. Her book is based, like much of Jung's writing on the subject, on an analysis of cross-cultural and cross-religious themes.
Although Hyde praises androgyny, she is not positive about transgender experience. She discusses transsexuals in her chapter on gender roles. There is little scientific reference in Hyde's textbook with regard to the causation of transsexualism. She favors an early-learning theory of transsexualism based on one clinical study of 17 men and the work of Richard Green.lxvii As we have seen, Richard Green's study does not defend such a theory. However, Hyde does quote an interesting study by Michael Z. Fleming et al, in which 72 presurgery transsexuals took the BSRI.lxviii Fleming reported that the female-to-male preoperative transsexuals had scores that were very similar to Bem's college classes: 35 % scored as masculine, 35 % as androgynous and 30 % as feminine. The male-to-female group scored 60 % feminine, 22 % androgynous and 18 % masculine. The Fleming study can fairly be counted as distinctly counter to clinical establishment theories of transsexualism because it contradicts the theorized course or story line.
Hyde writes about transvestism as a form of fetishism in her chapter on variations in sexual expression. She derives this opinion from Wardell Pomeroy, a clinician who was on the original Kinsey team.lxix However, without noticing the contradiction, she quotes a study of 50 heterosexual transvestites by John Talamini that names four basic motives for men to cross-dress.lxx They include sexual arousal (which is not the same as fetishism), relaxation (escape from the male role), role-playing achievement (success in acting), and adornment (liking pretty things). Perhaps Hyde thinks she is doing transvestites a favor when she dismissively asserts that transvestism is a harmless sexual variation when done in private. It seems that Hyde, Bem, and Singer can praise androgyny only as long as it doesn't mess with appearance norms.
Richard Zuckerberg offers a more explicit challenge to establishment clinicians in his view of androgyny. He writes, "Androgyny speaks to a fundamental issue in all of us. For each of us, it brings up the idea of psychological integration, the notion of being able to explore, develop, bring together, and use all aspects of our humanness, for ourselves and for others. This will lead to a higher level of personal integration, a true joining and a coming together of split-off aspects of ourselves, which will make for a stronger, more resilient structure within."lxxi I like the direction of Zuckerberg's comments but find it unfortunate that Zuckerberg, like Hyde, Bem, and Singer, did not take on issues of transgender expression. Many transgender people have been doing exactly the psychological work that Zuckerberg talks about.
In reviewing these nonclinical studies it is important to understand that they are all studies of attitudes. Any study of attitudes should meet multiple methodological standards to be of much scientific value. Bem's work does not meet all these standards. In a later review of her own work she wrote, "By the late 1970's and early 1980's, however, I had begun to see that the concept of androgyny inevitably focused so much more attention on the individual's being both masculine and feminine than on the culture's having created the concepts of masculinity and femininity in the first place...."lxxii The core of her later book, The Lenses of Gender: Transforming the Debate on Sexual Inequality, is built on calling attention to three cultural lenses about gender that she feels are inherently distorting of underlying human truth. She argues that the lenses affect the everyday processes of learning the roles of men and women and pervade social institutions such as science and the law.
The three cultural lenses that Bem believes distort underlying human reality are androcentrism, gender polarization and biological essentialism. For Bem, androcentrism means the cultural dominance of the masculine perspective. Androcentrism is not merely about social dominance by men. The lens issue is about treating the masculine perspective as normative, or neutral, or natural while treating feminine perspectives as other.
The second distorting lens Bem critiques is gender polarization. She is now unsatisfied with her earlier work, in part, because the problem of false concretization creates gender polarization. It is important to understand this point, because the BSRI is still widely used to measure masculinity, femininity, and androgyny. False concretization is a common problem for social psychological measures of attitudes. In her early work, by asserting that she was measuring masculinity, femininity, and androgyny, she creates the impression that such phenomena are really there and that her BSRI merely points to them. But the "really there" was operationalized by a questionnaire that asked college students to rate items for masculinity, etc. That is, the original measurements were of cultural attitudes, not of some theorized underlying human truth. She reported her finding of androgyny as a new finding that modifies a bipolar picture of gender, when the same data could have been seen as overlap and similarity among people – thus undercutting bipolar conceptualizing.
To understand Bem's reassessment of her work it may help to know that she points to her personal subjectivity about gender as a significant source of her change. She reports this as follows: "Although some of the (very few) individuals to whom I have been attracted during my forty-eight years have been men and some have been women, what those individuals have in common has nothing to do with either their biological sex or mine -- from which I conclude, not that I am attracted to both sexes, but that my sexuality is organized around dimensions other than sex."lxxiii This self-report is another example of confusing choice of sexual partner with the larger category of gender, a concern which will be taken up later in a section on bisexuality.
The issue of whether Bem's concepts of masculine, feminine, and androgynous, are falsely concretized depends on the realm of analysis within which one is working. My answer is that, although they may be falsely concretized as psychological concepts, they are not falsely concretized as cultural concepts. Bem's later concern reflects her awareness that cultural images and stereotypes are not an arithmetic sum of individual experiences. What is going on here is that Bem has become aware that human beings don't always respond to cultural stereotypes in the same way.
Bem criticizes the lens of biological essentialism for undervaluing the creativity shown by individuals and reflected in cultural images. However, although she is able to criticize biological essentialism, Bem shows a psychological essentialism. This psychological essentialism is shown when she claims that the psychological facet of human truth is somehow more basically meaningful than the cultural facet. I simply repeat that scientific truth is relative to the kinds of questions that are being asked. Instead of thinking of cultural images as a sum of individual experience, we can understand it as individuals sharing in cultural creativity. Once created, the cultural images influence the self-understanding of individuals. Attention to the creation and change of gender roles and images is addressed in the following chapter.
I value Bem's research from the 1970's more than she does, because I am not interested in reducing all human truth to individually focused truth. The studies she conducted and stimulated point to the complexity, overlap, and ambiguity of the cultural symbols of masculinity and femininity. She has shown that, even though a lot of the rhetoric about masculinity and femininity is bipolar, that doesn't mean that individuals always take on norms and images in bipolar ways. For example, it may go against cultural standards for men to be nurturant, but sometimes they do it just the same.
Our Freedom Generating and Influential Bodies
One of the important reasons for spending so much time with physiological factors in the second chapter was to lay the groundwork for challenging the misuse of physiological reasoning by some psychologists and psychiatrists. The energy that has gone into studies of physiological and psychological differences between men and women is one measure of how deeply committed our culture is to defending traditional gender relationships. No matter how many studies point to similarity, what continues to be news is another tentative finding of difference.
In the previous chapter I reviewed the studies of an almost microscopic brain aspect (INAH 3) that some think might show a difference between gay and straight males. I pointed out that nobody even knows whether the INAHs are relevant in any way to human behavior. In this chapter, as part of pointing to the difference between the psychological conception of mind and the physiological conception of brain, I point out that many of the studies of this obscure bit of brain anatomy don't mention that the size of the tiny INAH 3 varies by a factor of 10 in both males and females and that there is great overlap.lxxiv The constant, although so far unconvincing, search for differences between the brains of males and females distracts greatly from the common finding of a high degree of similarity and overlap in the mind activities of men and women. The brains of males and females support this similarity and overlap. Similarity is the consistent finding of hundreds, probably thousands, of studies. Findings of psychological similarity, complexity, interactiveness, and flexibility do not prove that there are no brain differences that affect mind activities. Despite repeated failures, it is possible that someone will eventually show some difference between the brains of males and females. I've already argued that any such difference may be the result of practice rather than genetically based structure. Here I want to point out that to assert that the brain is the basis for mind activity is a very different thing from asserting that a difference in brains causes a difference in a specific mental pattern, such as gender identity. This chapter has pointed to the similarity and overlap of many mind activities in men and women. This means that even if someone discovers a brain difference in males and females, that difference has not created differences in a lot of the mind activities of men and women. Therefore it is not scientifically reasonable for psychologists and psychiatrists to pathologize transgender behavior by saying the equivalent of "Oh, that must be because of some brain, or other physiological, defect." It is far more reasonable to argue that complex, flexible, interactive brains give rise to complex, flexible, and interactive mental activities which include the observed capacity to create complex, flexible, interactive mental constructs, which are melded together in complex, flexible, and interactive social and cultural activities to create observable complex, flexible, and interactive gender stereotypes and roles.
Roger Gorski has received a lot of attention for his attempts to find and interpret sexual differences in rat brains. Along the way he has trained people like Laura Allen, who studies human brains looking for sexual differences. Phyllis Burke visited the UCLA lab where Gorski, Allen, and others have been dissecting rat brains for 15 years. One major focus of this lab has been the preoptic nucleus in rats. This is interesting to Gorski and his students because the preoptic nucleus is bigger in male rats than in female rats. It hasn't deterred them that they don't know the function of the preoptic nucleus and it hasn't deterred them that there is no similar structure in human brains.lxxv After being so often denied in their search for the needle, any difference is interesting.
People pursue physiological changes in response to cultural symbols. Some males take estrogen because they want to feel more feminine or look more feminine. After the estrogen has been taken, they are less distinctively male, and the line of reasoning between physiological and psychological factors gains complexity. Hormonal level becomes not only an initial cause but also an intervening cause in explaining, for example, observable breast development. Breast size is in turn a significant social and cultural symbol that can affect attractiveness to potential sexual partners. Related experience may have a significant impact on self-concept. An important difference between a change in breast size and a change in some aspect of the brain is that the breast change is visible and has symbolic cultural value whereas brain changes are hidden. Such interactivity makes it tricky to sort out causal direction.
As a final warning in this chapter to resist the urge to misuse physiological reasoning to fit psychological theories, I refer to the work of Susan Coates, an establishment clinical psychologist who sat on the committee that created the gender identity disorder (GID) label. She was so eager to suggest a physiological base for her opinions about "GID children" that she wrote that feminine boys "may remark on good odors, such as cookies baking in the oven" (showing a heightened sense of smell) and that "Many boys with GID refuse to wear a new shirt unless the tag is cut out (showing a heightened tactile sensitivity)."lxxvi