Concept

Is Being Gay Biological

Intro

Short answer: biology plays a real part in who people are attracted to, but there is no "gay gene," and biology does not decide it all by itself. The largest gene study ever done found many tiny genetic effects, not one big one, and said you cannot use a person's DNA to predict whether they will have same-sex partners. Identical twins share the same genes, but when one twin is gay the other usually is not. Things that happen before birth also seem to matter, and so do factors no one has pinned down yet.

Two honest points cut in different directions. First, most gay and lesbian people say they never chose their attractions, and the research backs them up on that. Telling someone "you just decided to feel this way" is not supported by the evidence. Second, the science does not show that orientation is fixed by genes from birth. Attractions are fairly stable for most people, but they shift over time for some, more often for women and for people attracted to both sexes.

The biggest point is this: however the science turns out, it cannot answer the moral question. Where a desire comes from and whether it is right to act on it are two different questions, and facts about genes or hormones only speak to the first one.

A word on tone: what follows examines claims and arguments, not people. Every person bears the image of God and is loved by him.

◎ See this page's graph and discover linked topics →

In full

This page is the codex's science reference on the causes of sexual orientation. It aims to state each finding at its real strength, so that neither "born this way, case closed" nor "it is just a choice" survives contact with the research. The debate replies that rely on it live on two companion pages: Born This Way Objection Defeater handles the argument "it is inborn, so acting on it is fine," and Epigenetic Homosexuality Model Objection Defeater takes apart one specific 2012 epigenetic model. The Christian position on the ethics itself is set out on Homosexuality.

The best current summary, drawn from the major review by Bailey and colleagues (2016) and the largest genetic study (Ganna and colleagues, 2019):

• Genes contribute. Twin studies drawn from whole populations attribute roughly a third of the variation in men's same-sex behavior to genes, and less than that in women. Genome-wide studies confirm a real but diffuse genetic signal spread across many variants of tiny effect.
• Genes do not determine. Most of the variation is not genetic, identical twins usually differ, and DNA cannot predict an individual's behavior.
• Prenatal biology is the best-supported non-genetic influence: in men, the fraternal birth order effect; in women, the raised rates of same-sex attraction seen after unusual prenatal androgen exposure.
• Social explanations have fared poorly. Bailey and colleagues concluded there is "considerably more evidence supporting nonsocial causes of sexual orientation than social causes," and found the evidence for recruitment, disordered parenting, or the influence of gay parents "generally weak in magnitude."
• Orientation is largely unchosen as an experience, and largely stable, but not perfectly fixed. Change over time is documented, mostly among women and bisexual people, and deliberate efforts to change it have weak evidence of success.
• None of this settles the ethics. Bailey and colleagues say so themselves, and the reason is logical rather than empirical: an "is" does not by itself yield an "ought" (Is-Ought Fallacy).

Key terms used below. Heritability is a population statistic: the share of the variation in a trait, within a particular population, that is statistically associated with genetic differences. It does not mean "this percent of a person's orientation comes from genes," and it can change when the population or environment changes. Concordance is the chance that a twin shares a trait given that the co-twin has it. Shared environment is whatever makes twins raised together alike apart from genes (the same home, parents, schools). Unique or nonshared environment is everything that makes them different, including measurement error.

What "choice" means

Much of the public argument goes nowhere because the word "choice" is used for three different things.

1. Choosing the attraction. Did the person decide to be attracted to the same sex? The evidence says, for most people, no. In a U.S. probability sample of 662 lesbian, gay, and bisexual adults, 88 percent of gay men and 68 percent of lesbians said they had "no choice at all" about their sexual orientation (Herek and colleagues, 2010). Bisexual respondents reported more sense of choice: about 39 percent of bisexual men and 44 percent of bisexual women said they had a fair amount or a great deal of it.
2. Choosing the action. Does the person decide whether to act on the attraction? Yes, in the same way every adult decides whether to act on any sexual desire. This is the level at which the historic Christian ethic speaks, and it applies to everyone's desires, not to one group's.
3. Choosing the identity. Does the person decide how to describe themselves, whether "gay," "same-sex attracted," "bisexual," or no label at all? Partly, yes. Identity labels shift more often than attractions do, and the American Psychological Association's 2009 task force concluded that sexual orientation identity, rather than sexual orientation itself, appears to change through therapy, support groups, and life events.

Why debates equivocate. "Being gay is not a choice" is true in sense 1 for most people. It is false in sense 2 for everyone, and partly false in sense 3. A defender of the slogan can win the argument about attraction and then quietly carry the win over to action, which is a different question. A critic can make the opposite mistake, arguing about action and implying something false about attraction. Holding the three senses apart keeps both sides honest.

Genetics

The largest study: Ganna and colleagues, 2019

Andrea Ganna and colleagues published the largest genome-wide association study (GWAS) of the trait in Science (2019), using 477,522 people from the UK Biobank and 23andMe, with replication in 15,142 people from the United States and Sweden. A GWAS scans hundreds of thousands of common DNA variants to see which are statistically associated with a trait.

• What was measured. Same-sex sexual behavior, defined as whether a person had ever had sex with someone of the same sex. That is not the same as orientation, attraction, or identity.
• Five loci. Five locations in the genome reached genome-wide significance. Each had a very small effect.
• 8 to 25 percent. In the abstract's words, "all tested genetic variants accounted for 8 to 25% of variation in same-sex sexual behavior." By comparison, the family-based heritability estimate in the same UK Biobank data was 32.4 percent (95 percent confidence interval 10.6 to 54.3).
• No individual prediction. The variants "do not allow meaningful prediction of an individual's sexual behavior." Polygenic scores built from them explained under 1 percent of the variance in the test samples, and the authors state that such scores cannot accurately predict an individual's behavior.
• Different in men and women. The genetic effects "only partially overlapped between males and females."
• No single continuum. Comparing the genetics of ever having a same-sex partner with the genetics of the proportion of same-sex partners suggested "there is no single continuum from opposite-sex to same-sex sexual behavior."
• No X-chromosome signal. Unlike the earlier linkage studies, the authors found no excess of signal and no genome-wide significant locus on the X chromosome, which counts against the older Xq28 findings below.

What this means. There is no "gay gene." There is a real, polygenic genetic contribution, of the kind found for nearly every complex human behavior. A 2025 review in Trends in Genetics (Felesina and Zietsch) summarizes the field the same way: the trait "is influenced by many common genetic variants, each with tiny but cumulative effects," and the evolutionary explanation for why such variants persist "remains uncertain." A 2024 article in the American Journal of Human Genetics (Ventresca and colleagues) surveys the methodological and ethical concerns this line of research raises.

Earlier linkage studies: Xq28 and its replications

Before GWAS, researchers used linkage studies, which track whether brothers who are both gay share stretches of chromosome more often than chance predicts.

• Hamer and colleagues, 1993 (Science). In 114 families of gay men, they reported more gay relatives on the mother's side, and in a selected group of 40 families with two gay brothers, linkage to markers at Xq28 on the X chromosome (multipoint LOD score 4.0).
• Hu and colleagues, 1995 (Nature Genetics), from the same lab, reported linkage at Xq28 in new families of gay brothers but not in families of lesbian sisters.
• Rice and colleagues, 1999 (Science). In 52 pairs of gay brothers from Canadian families, allele sharing at Xq28 was "not increased over expectation," and the authors concluded the results "do not support an X-linked gene underlying male homosexuality."
• Mustanski and colleagues, 2005 (Human Genetics), a genome-wide scan of 456 individuals from 146 families with Hamer as a co-author, did not find linkage to Xq28 in the full sample.
• Sanders and colleagues, 2015 (Psychological Medicine), the largest linkage study, with 409 independent pairs of gay brothers, reported linkage signals on chromosome 8 (maximum two-point LOD 4.08) and at Xq28 (maximum two-point LOD 2.99).
• Sanders and colleagues, 2017 (Scientific Reports), a GWAS of 1,077 gay and 1,231 heterosexual men, reported suggestive peaks on chromosomes 13 and 14 and one association on chromosome 8 that reached genome-wide significance only when combined with an earlier result.

Fair reading. The Xq28 story is contested, not simply debunked: one large linkage study gave it partial support, while independent replications and the far larger Ganna GWAS did not. What no study has shown is a single gene that makes a person gay.

Twin studies

Twin studies compare identical twins (who share essentially all their DNA) with fraternal twins (who share about half). If identical twins resemble each other more, genes are implicated.

Volunteer samples versus population samples

Volunteer samples ran high. Bailey and Pillard (1991, Archives of General Psychiatry) recruited gay men with twins through gay publications. Of the co-twins whose orientation could be rated, 52 percent (29 of 56) of identical co-twins, 22 percent (12 of 54) of fraternal co-twins, and 11 percent (6 of 57) of adoptive brothers were gay. The weakness is recruitment: a gay man whose twin is also gay may be more likely to answer an advertisement about gay twins, which inflates concordance.

Population samples run lower. When twins are drawn from registries rather than recruited by advertisement, concordance drops sharply.

• Bailey, Dunne, and Martin, 2000 (Journal of Personality and Social Psychology), drawing 4,901 twins from the Australian Twin Registry, found identical-twin concordance of 20 percent for men and 24 percent for women on the strict criterion, against 52 percent for men in the 1991 volunteer study. The authors attributed the earlier, higher figures to recruitment bias.
• Kendler and colleagues, 2000 (American Journal of Psychiatry), using a U.S. national sample of twin and nontwin sibling pairs, found identical twins more alike than fraternal twins and concluded that familial factors "which are at least partly genetic" influence orientation, while cautioning about "low statistical power and the use of a single item" to measure it.
• Långström and colleagues, 2010 (Archives of Sexual Behavior), the largest population twin study of the trait, used 3,826 same-sex twin pairs from a survey of all Swedish adult twins aged 20 to 47. Measuring same-sex sexual behavior (any, and total number of lifetime same-sex partners), they estimated:
  • Men: genes 34 to 39 percent, shared environment 0 percent, unique environment 61 to 66 percent.
  • Women: genes 18 to 19 percent, shared environment 16 to 17 percent, unique environment 64 to 66 percent.
  • The authors note that "wide confidence intervals suggest cautious interpretation."
• Burri and colleagues, 2011 (PLoS ONE), with 4,426 British female twins, estimated a "modest" genetic influence of 25 percent on sexual attraction.

What "unique environment" does and does not mean

The large unique-environment share in these studies is often misread in both directions.

• It is not parenting. Parenting style, family religion, and household attitudes are shared by twins raised together, so they belong to the shared environment, which was estimated at zero for Swedish men. The twin data give no support to the idea that particular parenting produces gay sons.
• It is not choice. "Unique environment" is a statistical remainder, whatever is left after genes and shared family factors are accounted for. It includes chance differences in development (including before birth), different individual experiences, illness, and measurement error. Långström and colleagues describe it as "social and biological." Nothing in the statistic identifies a decision.
• It is not nothing either. The remainder is the largest single component in every population estimate above. Whatever causes orientation, most of it is not captured by shared genes.

Identical-twin discordance

Most identical twins with a gay co-twin are not gay themselves. In the Australian registry, concordance was about one in five for men and one in four for women. That single fact is enough to rule out strict genetic determinism: two people with the same genome usually differ. It does not show that the difference was chosen. It shows that something other than DNA sequence, or something random in development, matters a great deal.

Prenatal factors

The fraternal birth order effect

The most consistent non-genetic finding concerns men with older brothers.

• The finding. Blanchard and Bogaert (1996, American Journal of Psychiatry), comparing 302 gay men with 302 heterosexual men matched on year of birth, found that "each additional older brother increased the odds of homosexuality by 33%," while older sisters, younger siblings, and parental age had no effect. The older-brother pattern also appeared in Kinsey Institute interview data from 1938 to 1963 (Blanchard and Bogaert, 1996, Archives of Sexual Behavior, 4,948 men).
• The meta-analysis. Blanchard (2018, Archives of Sexual Behavior) pooled 30 gay and 30 heterosexual groups from 26 studies (7,140 gay and 12,837 heterosexual men). The pooled "Older Brothers Odds Ratio" was 1.47. The ratio was significantly above 1 in 20 groups, above 1 but not significantly in 9, and slightly below 1 (not significantly) in 1.
• Biological, not social. Bogaert (2006, PNAS), across four samples totaling 944 men, including men raised with step-brothers, half-brothers, or as adoptees, found that "only biological older brothers," not brothers they were reared with, predicted orientation, "regardless of the amount of time reared with these siblings." That points to something that happens in the mother's body, not in the household.
• A proposed mechanism. Bogaert and colleagues (2018, PNAS) tested the idea that some mothers develop an immune response to a male-specific protein with each son. Comparing 54 mothers of gay sons with 72 mothers of heterosexual sons, plus 16 women with no sons and 12 men, they found that mothers of gay sons, "particularly those with older brothers," had higher antibody levels against the Y-linked protein NLGN4Y. This is one small study and a hypothesis under test, not an established mechanism.

How big is the effect? A 33 percent increase in odds sounds large, but it multiplies a small starting rate. As an illustration only (not a figure from these studies): if the base rate were about 2 percent, one older brother would raise it to roughly 2.6 percent. Cantor and colleagues (2002, Archives of Sexual Behavior) estimated that "roughly 1 gay man in 7 owes his sexual orientation to the fraternal birth order effect." The effect is documented in men. So it is real and robust, but it explains a minority of cases, and many gay men have no older brothers at all.

Prenatal androgens and congenital adrenal hyperplasia

Congenital adrenal hyperplasia (CAH) is a genetic condition in which the adrenal glands produce excess androgens (male-typical hormones) before birth. It acts as a natural experiment on prenatal hormone exposure.

• Meyer-Bahlburg and colleagues, 2008 (Archives of Sexual Behavior) studied 40 salt-wasting, 21 simple-virilizing, and 82 non-classical CAH women and 24 unaffected female relatives. "Most women were heterosexual," but rates of bisexual and homosexual orientation were raised in the CAH groups and "correlated with the degree of prenatal androgenization."
• Hines, Brook, and Conway, 2004 (Journal of Sex Research), with 16 women and 9 men with CAH and unaffected relatives as controls, found less heterosexual interest in women with CAH and no difference in men with CAH.
• Reiner, 2004, reporting on 29 genetic males with cloacal exstrophy (a severe birth defect) who were assigned and raised female, found that all showed "a marked male-typical shift" in psychosexual development and nearly half later declared themselves male. Bailey and colleagues (2016) cite such cases as evidence against social causes: when infant boys are surgically and socially "changed" into girls, "they remain sexually attracted to females."

Fair reading. Prenatal hormones shift the odds, especially in women. Even large hormonal differences leave most affected women heterosexual, so hormones influence orientation without dictating it.

Brain-structure studies

• LeVay, 1991 (Science) measured four small cell groups in the hypothalamus in brains obtained after death: 19 gay men, all of whom had died of AIDS, 16 presumed heterosexual men (6 of whom had died of AIDS), and 6 presumed heterosexual women. He reported that one group, INAH3, was more than twice as large in the heterosexual men as in the gay men.
• Byne and colleagues, 2001 (Hormones and Behavior), with 34 presumed heterosexual men, 34 presumed heterosexual women, and 14 HIV-positive gay men, confirmed the sex difference in INAH3 but found only "a trend" toward smaller volume in gay men and "no difference in the number of neurons" by orientation.

Limits. The samples were small, orientation was presumed from records rather than self-reported, AIDS was a confound, and individual measurements overlapped across groups. A difference in adult brains also cannot show which way causation runs. These studies suggest a biological correlate; they do not show a cause.

Epigenetics

Epigenetics refers to chemical tags on DNA that change how genes are switched on without changing the DNA sequence.

• The 2012 hypothesis. Rice, Friberg, and Gavrilets (Quarterly Review of Biology, 2012) proposed that sex-specific epigenetic marks, which normally buffer a fetus against too much or too little androgen, sometimes escape erasure and pass from a parent to an opposite-sex child, shifting partner preference. It was offered to explain why orientation runs in families yet identical twins are often discordant. It is a model built from existing data, not a test of it: it identifies no specific mark, gene, or transmission event. It also predicted that "the molecular feature underlying most homosexuality is not DNA polymorphism(s)," a prediction the 2019 Ganna results cut against.
• The 2015 conference claim. At the American Society of Human Genetics meeting in October 2015, Tuck Ngun of UCLA presented an algorithm using DNA methylation at nine regions, measured in saliva from 37 identical male twin pairs discordant for orientation and 10 concordant gay pairs, claiming about 70 percent accuracy in predicting orientation. The headlines did not survive scrutiny. The work was unpublished when presented, the sample was very small, and as reported at the time the model was far better at classifying gay men (about 83 percent) than straight men (about 50 percent). J. Michael Bailey said it "certainly" needed replication in larger samples, Peng Jin warned that methylation in saliva may say little about the brain, and Ed Yong's critique in The Atlantic was titled "No, Scientists Have Not Found the 'Gay Gene'."

The full critique of the 2012 model, including why it cannot be cited as proof in either direction, is in Epigenetic Homosexuality Model Objection Defeater.

Fluidity and change over time

Lisa Diamond's longitudinal work on women. Diamond (2008, Developmental Psychology) followed 79 lesbian, bisexual, and "unlabeled" women over 10 years. Two thirds changed the identity label they had claimed at the start, and one third changed labels two or more times. But the changes did not fit a simple "phase" story: more women adopted bisexual or unlabeled identities than gave them up, and bisexual and unlabeled women "had stable overall distributions of same-sex/other-sex attractions" with larger short-term fluctuations than lesbians. Diamond concluded that the lesbian and bisexual distinction is "a matter of degree rather than kind."

Sex differences. Kinnish, Strassberg, and Turner (2005, Archives of Sexual Behavior) asked 762 adults aged 36 to 60 to rate their fantasy, romantic attraction, and behavior across their lives. Gay and heterosexual women reported more change over time than gay and heterosexual men; bisexual men and women did not differ. Bailey and colleagues (2016) likewise report that men's orientation is more tightly tied to their pattern of sexual arousal than women's.

Large samples.

• Ott and colleagues, 2011 (Archives of Sexual Behavior), following 13,840 U.S. youth aged 12 to 25 across up to four survey waves, found low change in identity for the cohort as a whole, more change in girls than boys overall, and much more among sexual-minority youth, where the difference between the sexes "appeared to be eliminated."
• Savin-Williams, Joyner, and Rieger, 2012 (Archives of Sexual Behavior), using the Add Health study (12,287 young adults, two waves about six years apart), found that stability was more common than change. Men and heterosexual-identified people were the most stable. People who identified as 100 percent homosexual were nearly as stable as people who identified as 100 percent heterosexual. Bisexual identity was the least stable, with about a quarter keeping it at the second wave.
• Mock and Eibach, 2012 (Archives of Sexual Behavior), using the national MIDUS survey of 2,560 adults (average age about 47) over 10 years, found that "somewhat more than 2%" reported a different identity at the second wave. Heterosexual identity was the most stable. Among women, lesbian and bisexual identities were equally less stable; among men, gay and heterosexual identities were both relatively stable, and bisexual identity was "particularly unstable."

Fair reading. Change in identity, and for some people in attraction, is real and measurable, and it occurs mostly among women and bisexual people. For most adults, especially men with exclusive attractions, orientation is quite stable. Diamond and Rosky (2016, Journal of Sex Research), writing in support of legal protections for sexual minorities, argued from the longitudinal data that arguments based on the immutability of orientation are "unscientific," and also that those rights should not depend on the science either way. Natural change over time is not the same thing as change produced by deliberate effort, which is the next question.

Orientation-change efforts

The APA 2009 task force. The American Psychological Association's Task Force on Appropriate Therapeutic Responses to Sexual Orientation reviewed the peer-reviewed research on sexual orientation change efforts (SOCE). Its findings, stated accurately:

• Efficacy. "The results of scientifically valid research indicate that it is unlikely that individuals will be able to reduce same-sex sexual attractions or increase other-sex attractions through SOCE." The few rigorous studies, mostly of older aversive treatments, showed that "enduring change to an individual's sexual orientation is uncommon."
• Harm. The task force found "some evidence to indicate that individuals experienced harm from SOCE," especially from early aversive methods, but also stated that "there are no scientifically rigorous studies of recent SOCE that would enable us to make a definitive statement about whether recent SOCE is safe or harmful and for whom."
• Mixed reports from participants. "Some individuals perceived that they had benefited from SOCE, while others perceived that they had been harmed." Some participants reported going on to lead "outwardly heterosexual lives."
• Identity versus orientation. The task force concluded that sexual orientation identity, as distinct from sexual orientation itself, appears to change through psychotherapy, support groups, and life events.
• Religious clients. It acknowledged that many religious people "desired to live their lives in a manner consistent with their values," which it called "telic congruence," and described a therapeutic approach to a client's choice of celibacy that is "neither rejecting nor promoting" it.

Studies on the other side, and their limits.

• Spitzer, 2003 (Archives of Sexual Behavior) interviewed 200 self-selected people who reported change after reparative therapy and judged their reports credible. In 2012 Spitzer published a letter in the same journal reassessing the study: he wrote that he now judged the major critiques "largely correct," chiefly that there was no way to verify the participants' accounts, and apologized to the gay community. It should not be cited as evidence of change.
• Jones and Yarhouse, 2011 (Journal of Sex and Marital Therapy) followed 72 men and 26 women in Exodus-affiliated Christian ministries over six to seven years. They reported that "change of homosexual orientation appears possible for some" and that psychological distress did not increase on average. Among the roughly 60 participants remaining at the final assessment, about 23 percent fell in a "conversion" outcome category and about 30 percent in a "chastity" category. Limits: the study was funded by Exodus (disclosed by the authors), relied on self-report, had no comparison group, retained about two thirds of the original sample, and, as the authors acknowledge, may reflect "sexual identity change or adjustment" rather than change in attraction.
• Sullins, 2022 (Archives of Sexual Behavior) reanalyzed the Blosnich and colleagues (2020, American Journal of Public Health) national sample of 1,518 sexual-minority adults, which had linked SOCE to nearly twice the odds of lifetime suicidal ideation. Sullins argued that once suicidality occurring before SOCE is accounted for, the positive association disappears. Blosnich and colleagues published a rebuttal in 2023. That dispute is unresolved and concerns harm, not efficacy.
• Sullins, 2024 (Cureus), a convenience sample of 72 highly religious men who had completed SOCE, reported that about a third (23) described a shift from mainly homosexual to mainly heterosexual attraction, with change in behavior much stronger than change in attraction.

Fair reading. The evidence that deliberate efforts reliably change attraction is weak. Most supportive studies rely on self-selected, self-reported samples without controls. Evidence about harm from recent, non-aversive efforts is also contested. What is better supported is that behavior and identity can change, which matches the distinction in "What 'choice' means" above.

What the major reviews conclude

Bailey and colleagues, 2016. "Sexual Orientation, Controversy, and Science," in Psychological Science in the Public Interest, by J. Michael Bailey, Paul Vasey, Lisa Diamond, S. Marc Breedlove, Eric Vilain, and Marc Epprecht, is the most cited comprehensive review. Its main conclusions:

• There is "considerably more evidence supporting nonsocial causes of sexual orientation than social causes," citing "moderate genetic influences demonstrated in well-sampled twin studies," the "cross-culturally robust fraternal-birth-order effect on male sexual orientation," and the cloacal exstrophy cases.
• Evidence for the most commonly proposed social causes, "sexual recruitment by homosexual adults, patterns of disordered parenting, or the influence of homosexual parents," is "generally weak in magnitude."
• In the publisher's summary, biological factors such as prenatal hormones and genetic profiles "are likely to contribute to sexual orientation, though they are not the sole cause," and there is little evidence that non-heterosexual orientations become more common as social tolerance increases.
• Most relevant here, the abstract states that "the actual relevance of these issues to social, political, and ethical decisions is often poorly justified." The review's own authors say the causal science does not decide the moral question.

Felesina and Zietsch, 2025. "Emerging insights into the genetics and evolution of human same-sex sexual behavior," in Trends in Genetics, confirms that twin studies show "a substantial heritable component" and that GWAS show many variants "each with tiny but cumulative effects," while stressing how much remains unknown.

How gay and lesbian people describe their attractions

The best probability-sample data on this question is Herek and colleagues (2010, Sexuality Research and Social Policy), a U.S. national sample of 662 self-identified lesbian, gay, and bisexual adults. Asked "How much choice do you feel you had about being" gay, lesbian, or bisexual:

The authors' summary: "Most respondents reported experiencing little or no choice about their sexual orientation." Self-report describes experience, not cause, but on the narrow question of whether attraction is felt as a decision, it is strong evidence that for most gay men and lesbians it is not.

What the science can and cannot settle

What it can settle. The science can tell us how much of the variation in a population tracks genes, which prenatal factors shift the odds, how stable orientation is over time, and whether particular interventions work. On those questions the answers above are the current best estimates, and they will be revised as data improve.

What it cannot settle. No finding about where a desire comes from can, on its own, show whether acting on it is right or wrong. "This desire is partly inborn" is a description. "Acting on it is permissible" is a moral judgment. Getting from the first to the second requires a moral premise that biology does not supply. This is the is-ought gap (Is-Ought Fallacy), and it is set out as the decisive move in Premise 2 of Born This Way Objection Defeater.

The heritable-trait comparison. Other dispositions with comparable or higher heritability show the point without any moral comparison between the traits themselves:

• Alcohol use disorder is "approximately 50% heritable": a meta-analysis of 12 twin and 5 adoption studies put it at 0.49, with a 95 percent confidence interval of 0.43 to 0.53 (Verhulst, Neale, and Kendler, 2015, Psychological Medicine).
• Aggression: a meta-analysis of 24 genetically informative studies found a genetic effect that "may account for up to 50% of the variance," though observational laboratory ratings showed no heritability (Miles and Carey, 1997, Journal of Personality and Social Psychology).
• Antisocial behavior: a meta-analysis of 51 twin and adoption studies estimated additive genetic influences at 0.32 and nonadditive at 0.09, about 41 percent in total (Rhee and Waldman, 2002, Psychological Bulletin).

No one concludes from these figures that heavy drinking or assault is therefore right. The comparison tests the form of an inference, from heritable to permissible. It does not equate the traits.

The gap cuts both ways. Fairness requires saying that the same logic binds the other side. "Orientation can change for some people" does not show that anyone ought to try to change it, and "it is partly environmental" does not show that it is wrong. The Christian ethic does not rest on orientation being chosen, unstable, or caused by upbringing; it rests on a claim about the design and purpose of sexuality and on the authority behind that design. The science neither proves nor refutes that claim. Homosexuality states the position, and Moral Argument takes up where moral obligation comes from at all.

How Scripture frames desire. Briefly, since this is the science page: the biblical picture already expects people to find desires in themselves that they did not choose (Psalm 51:5), and it places the moral question at the level of the heart and the will. Jesus taught that "from within, out of the heart of men, evil thoughts proceed" (Mark 7:21, ASV), and called every disciple, whatever his desires, to "deny himself" (Luke 9:23, ASV). The historic Christian sexual ethic calls everyone to chastity outside the marriage of a man and a woman, so it singles out no group.

Evidence at a glance

See also

• Homosexuality, the anchor hub: the orientation, desire, and act distinction and the Christian position
• Born This Way Objection Defeater, the debate-prep reply to "it is inborn and natural, so acting on it is fine," which relies on this page for its evidence
• Epigenetic Homosexuality Model Objection Defeater, why the 2012 epigenetic model is not proof of biological causation in either direction
• Biblical Sexual Ethics Objection Defeater, the reply to the biblical-text objections
• Is-Ought Fallacy, why a biological cause would not settle the moral question
• Moral Argument, the grounding question the debate ultimately turns on