how homosexuality left the DSM
for a hundred and four years, from a Berlin journal article in 1869 to a board vote in December 1973, being gay was a diagnosis. it was not removed because a study disproved it; it was removed because two data sets found the same thing in people who were not patients, a generation of activists made the profession look at them, and a committee chair concluded that the argument was about values, not facts. here is the record in order, with the sampling problems and the politics left in.
the record, in order
Carl Westphal, in the Archiv für Psychiatrie, publishes case reports under the title “contrary sexual feeling, symptom of a neuropathic (psychopathic) condition.” the phrase that would become “inversion” and then “homosexuality” enters medicine as a symptom.[1]
Krafft-Ebing’s Psychopathia Sexualis makes it a diagnosis with a cause: a person might be born with the predisposition, and that made it a congenital disease. the book went through many editions and was the reference text for a generation of forensic psychiatry.[2][8]
Kinsey publishes histories from about 5,300 white American men. 37 percent had at least one homosexual experience to orgasm; 10 percent were “more or less exclusively homosexual” for at least three years; 4 percent were exclusively so for life. these were not patients.[3]
the American Statistical Association’s committee (Cochran, Mosteller, Tukey) publishes its review of Kinsey’s sampling. Kinsey’s own text says the histories came “more particularly from the northeastern quarter of the country” and included prison populations.[3][4]
Evelyn Hooker gives projective tests to 30 gay men and 30 heterosexual men, none of them patients, and finds no more disturbance in the gay group. it was the first controlled comparison; every earlier “finding” had come from men who were in treatment or in court.[5][8]
activists disrupt the APA’s 1970 and 1971 meetings. a 1971 panel, “Gay is Good,” with Frank Kameny and Barbara Gittings; in 1972 a psychiatrist in a mask, “Dr. H Anonymous,” tells the hall what gay psychiatrists face inside their own profession.[8]
the APA holds a symposium, “Should homosexuality be in the APA nomenclature?”, printed in the American Journal of Psychiatry that November: Stoller, Marmor, Bieber, Gold, Socarides, Green, Spitzer. both sides, in print.[6]
the Board of Trustees votes to remove homosexuality from DSM-II. it is replaced, not deleted: “Sexual Orientation Disturbance,” which Spitzer later called “a compromise between the view that preferential homosexuality is invariably a mental disorder and the view that it is merely a normal sexual variant.”[7][8]
opponents petition for a referendum of the whole membership. the removal is upheld by 58 percent of about 10,000 voting members. a diagnosis had been put to a vote and survived it by a margin, not a landslide.[8]
DSM-III replaces Sexual Orientation Disturbance with “Ego-dystonic Homosexuality.” Spitzer, who chaired the process, writes in 1981 that the dispute is “a value judgment about heterosexuality, rather than a factual dispute about homosexuality.”[7][8]
DSM-III-R drops Ego-dystonic Homosexuality. fourteen years after the vote, the manual has nothing left under the word.[8]
the WHO’s ICD-11 working group recommends deleting ICD-10’s remaining sexual-orientation categories. what the literature studies now is the excess of depression and anxiety in gay, lesbian and bisexual people, and the model it uses for it is minority stress, not orientation.[9][10]
five questions, answered from the sources
what did the science actually show?[3][5][8]
two things, from two data sets. Kinsey: homosexual behaviour was far more common in the general population than the clinic suggested, 37 percent of men with at least one experience to orgasm, 4 percent exclusively homosexual for life. Hooker: when you test gay men who are not patients against straight men who are not patients, expert judges cannot find more pathology in the gay group. neither study proved anything about cause. together they removed the empirical ground under the diagnosis, which had always rested on the men psychiatrists happened to see.
what were the problems with that science?[3][4][5][8]
Kinsey’s sample was not random and he said so: histories came mostly from the northeastern United States and included prison populations, and the American Statistical Association published a 44-page review of the sampling in 1953. Drescher writes that the famous 10 percent “is today believed to be closer to 1%–4%.” Hooker’s study had thirty men a side, and men who are not patients are, by construction, the men who were doing well enough not to be. a small study of volunteers cannot show that no gay man is disturbed; it could only show that the clinic’s sample was not the population, and that is what it showed.
so was 1973 a scientific decision or a political one?[6][7][8]
both, and the record says both. activists disrupted two annual meetings and got a hearing; a nomenclature committee reviewed the definition of a mental disorder and found homosexuality did not meet it; the board voted; the losing side forced a referendum and lost it 58 to 42. the chair of the committee wrote in 1981 that the question was a value judgment, not a factual dispute; the other side had argued its case in the same journal that May. a vote does not settle a fact. the committee’s point was that there had never been a fact to settle.
what replaced it?[7][8]
not nothing. Sexual Orientation Disturbance in the 1974 printing of DSM-II, then Ego-dystonic Homosexuality in DSM-III in 1980, for people distressed by their orientation. that category went in 1987. the removal took fourteen years to finish, and for those years the manual still had a way to file a gay patient as ill if he said he wished he were not.
what does the literature say now?[10]
it studies the excess, not the orientation. Meyer’s 2003 review found higher prevalence of mental disorders in lesbian, gay and bisexual people and framed it as minority stress: stigma, prejudice and discrimination as the exposure. the federal numbers behind that are on the LGBTQ+ data page, with each construct named. documenting a disparity is not diagnosing an identity, and the page that holds the numbers says so.
what the diagnosis rested on
from Westphal to the 1950s, every clinical claim about homosexuality came from a sample nobody chose: the people who ended up in a consulting room or a courtroom.[1][2] Drescher’s history is blunt about the mid-century analysts whose “cures” shaped DSM-I and DSM-II: their claims “were broadly accepted by their professional community although never verified in any meaningful or empirical way.”[8] Kinsey and Hooker did not refute a theory; they changed the sample. once you counted men who were not patients, the clinic’s picture stopped describing the population, and a diagnosis that had only ever been a description of the clinic had nothing left to stand on.[3][5]
the other direction
the case against the 1973 decision, stated fairly, is that it was made under pressure and by ballot. the meetings had been disrupted, the board voted, and when the losing side demanded a referendum the diagnosis survived by 58 to 42, which is not the margin science usually claims for a settled question.[8] Kinsey’s numbers were drawn from a sample he himself described as regional and partly institutional, a statistical committee spent 44 pages on the problems, and the 10 percent figure that entered the culture is, by Drescher’s account, probably one to four.[3][4][8] Hooker’s thirty men were volunteers who were not patients, so the study could not speak to the men who were.[5] and the removal was itself a compromise: for fourteen more years the manual kept a category for people distressed by being gay, which the chair of the process later described as a value judgment dressed as a diagnosis.[7] all of that is true. none of it restores the diagnosis, because the diagnosis never had a controlled data set behind it in the first place; what the vote settled was not whether homosexuality is an illness but whether a profession would keep saying so without evidence.
what this page does not say
it does not say what causes sexual orientation; none of the sources here do, and Kinsey and Hooker were explicit that they were describing, not explaining. it does not reproduce anything from the Westphal, Krafft-Ebing, Hooker or Cochran texts beyond what the source list says was read. and it says nothing about treatment of any kind: the excess of depression and anxiety that the modern literature documents in gay, lesbian and bisexual people is on the LGBTQ+ data page, with every construct named and the minority-stress model quoted rather than asserted.[10]
questions people ask
When was homosexuality removed from the DSM?
The American Psychiatric Association’s Board of Trustees voted to remove it from DSM-II in December 1973. A referendum of the membership upheld the decision in 1974 by a 58 percent majority of about 10,000 voting members. The replacement category, Sexual Orientation Disturbance, became Ego-dystonic Homosexuality in DSM-III (1980) and was removed in DSM-III-R (1987).
What was the earliest medical description of homosexuality?
Carl Westphal’s 1869 article in the Archiv für Psychiatrie und Nervenkrankheiten, titled “contrary sexual feeling, symptom of a neuropathic (psychopathic) condition.” Krafft-Ebing’s Psychopathia Sexualis (1886) then treated it as a congenital disease. The word “homosexual” itself was coined in the same year, 1869, by the Hungarian journalist Kertbeny.
What did Kinsey find in 1948?
From about 5,300 white American men’s histories: 37 percent had at least some overt homosexual experience to orgasm between adolescence and old age; 10 percent were more or less exclusively homosexual for at least three years between ages 16 and 55; 4 percent were exclusively homosexual throughout their lives. The sample was not random; it was drawn mostly from the northeastern United States and included prison populations, and the American Statistical Association reviewed its problems in 1953.
What did Evelyn Hooker’s 1957 study show?
She compared psychological test results of 30 gay men and 30 heterosexual men, none of whom were psychiatric patients, and found no more signs of psychological disturbance in the gay group. It was the first controlled comparison; earlier claims had come from clinical and forensic samples. The study was small, and its subjects were, by design, men who were not in treatment.
Does this page give medical advice?
No. It is a history of a diagnostic category, sourced to the journals and the original texts, with the limits of each source stated. It says nothing about any treatment.
sources
- Westphal C. Die conträre Sexualempfindung, Symptom eines neuropathischen (psychopathischen) Zustandes [Contrary sexual feeling, symptom of a neuropathic (psychopathic) condition]. Archiv für Psychiatrie und Nervenkrankheiten 1869/70;2(1):73-108. The title is the claim. Bibliographic record from Crossref (Springer dates the issue February 1870; the volume is conventionally cited as 1869). The German text was not read for this page; nothing beyond the title is taken from it. https://doi.org/10.1007/BF01796143
- Krafft-Ebing R von. Psychopathia Sexualis: eine klinisch-forensische Studie. Stuttgart: Enke, 1886. Digitised first edition (Wellcome Collection, on the Internet Archive). Not read for this page; its position is taken from Drescher 2015 (source 8): “Krafft-Ebing believed that although one might be born with a homosexual predisposition, such inclinations should be considered a congenital disease.” https://archive.org/details/b20421746
- Kinsey AC, Pomeroy WB, Martin CE. Sexual Behavior in the Human Male. Philadelphia: W. B. Saunders, 1948. Digitised copy read (pp. 650-651): “37 per cent of the total male population has at least some overt homosexual experience to the point of orgasm between adolescence and old age”; “10 per cent of the males are more or less exclusively homosexual (i.e., rate 5 or 6) for at least three years between the ages of 16 and 55”; “4 per cent of the white males are exclusively homosexual throughout their lives, after the onset of adolescence.” On the sample, in the book’s own words: “about 12,000 persons have contributed histories”; “about 5300 of these are the white males who have provided the data for the present publication”; histories came “more particularly from the northeastern quarter of the country”; “these prison populations have augmented our understanding of economically and educationally lower social levels.” https://archive.org/details/in.ernet.dli.2015.187552
- Cochran WG, Mosteller F, Tukey JW. Statistical Problems of the Kinsey Report. Journal of the American Statistical Association 1953;48(264):673-716. The American Statistical Association’s review of Kinsey’s sampling. Bibliographic record only (Crossref); not on PubMed and not read for this page, so no finding from it is stated here beyond its existence. https://doi.org/10.1080/01621459.1953.10501194
- Hooker E. The adjustment of the male overt homosexual. Journal of Projective Techniques 1957;21(1):18-31. PMID 13417147. PubMed carries no abstract and the paper is paywalled; it was not read for this page. Its design is taken from Drescher 2015 (source 8): “she compared psychological test results of 30 gay men with 30 heterosexual controls, none of whom were psychiatric patients. Her study found no more signs of psychological disturbances in the gay male group.” https://doi.org/10.1080/08853126.1957.10380742
- Stoller RJ, Marmor J, Bieber I, Gold R, Socarides CW, Green R, Spitzer RL. A symposium: Should homosexuality be in the APA nomenclature? American Journal of Psychiatry 1973;130(11):1207-1216. PMID 4784866. The printed record of the May 1973 debate, both sides. No abstract on PubMed; not read for this page beyond the record. https://doi.org/10.1176/ajp.130.11.1207
- Spitzer RL. The diagnostic status of homosexuality in DSM-III: a reformulation of the issues. American Journal of Psychiatry 1981;138(2):210-215. PMID 7457641. Abstract, verbatim: “In 1973 homosexuality per se was removed from the DSM-II classification of mental disorders and replaced by the category Sexual Orientation Disturbance. This represented a compromise between the view that preferential homosexuality is invariably a mental disorder and the view that it is merely a normal sexual variant”; “He argues that what is at issue is a value judgment about heterosexuality, rather than a factual dispute about homosexuality.” https://doi.org/10.1176/ajp.138.2.210
- Drescher J. Out of DSM: Depathologizing Homosexuality. Behavioral Sciences 2015;5(4):565-575. PMID 26690228. Open access; full text read. “In the wake of the 1969 Stonewall riots… gay and lesbian activists, believing psychiatric theories to be a major contributor to anti-homosexual social stigma, disrupted the 1970 and 1971 annual meetings of the APA”; “in December 1973, APA’s Board of Trustees (BOT) voted to remove homosexuality from the DSM”; “The decision to remove was upheld by a 58% majority of 10,000 voting members”; Rado-era analysts’ “claims of psychoanalytic ‘cures’ of homosexuality were broadly accepted by their professional community although never verified in any meaningful or empirical way”; “his now-famous ‘10%’ statistic is today believed to be closer to 1%–4%”; “ego-dystonic homosexuality was removed from the next revision, DSM-III-R, in 1987.” https://doi.org/10.3390/bs5040565
- Reed GM, Drescher J, Krueger RB, et al. Disorders related to sexuality and gender identity in the ICD-11. World Psychiatry 2016;15(3):205-221. PMID 27717275. Abstract: “ICD-10 categories related to sexual orientation have been recommended for deletion from the ICD-11.” https://doi.org/10.1002/wps.20354
- Meyer IH. Prejudice, social stress, and mental health in lesbian, gay, and bisexual populations: conceptual issues and research evidence. Psychological Bulletin 2003;129(5):674-697. PMID 12956539. Abstract: “LGBs have a higher prevalence of mental disorders than heterosexuals”; the excess is explained “in terms of minority stress—explaining that stigma, prejudice, and discrimination create a hostile and stressful social environment that causes mental health problems.” https://doi.org/10.1037/0033-2909.129.5.674
related: how transgender left the mental-disorders chapter · LGBTQ+ mental health in the data · the DSM
adam