chevron-down Created with Sketch Beta.
November 12, 2021 Feature

Exploring Identity

By Marie-Amélie George

Custody disputes between parents who disagree as to how to address their child’s gender identity have become the newest battleground in the country’s culture wars. This Article provides the first comprehensive analysis of custody cases involving “gender expansive children”—an umbrella term this Article uses to refer to transgender and nonbinary children, as well as children exploring nontraditional gender identities. By explaining the medical debates over treatment for gender expansive children and connecting these disputes to related precedent, this Article makes three distinct contributions.

First, this Article demonstrates that courts are focusing on the wrong question when adjudicating these cases involving pre-adolescent children. Judges have been trying to determine which parent is correct as to the child’s gender identity. However, the science of gender identity development indicates that only adolescent children’s gender identity is stable. The adult gender identity of pre-adolescent children, on the other hand, may be unknown until after a period of exploration. As a result, courts should be focused on which parent is best suited to help the child with the exploratory process, rather than its outcome.

Second, the Article reveals that courts have recognized children’s interests in exploring their identities as part of the “best interests of the child” analysis. By conducting an analysis of more than 90 custody opinions involving children who explore their sexual orientation, racial identity, and religious affiliation, this Article demonstrates that gender expansive child custody cases are the latest iteration of a long line of family court precedent involving judicially recognized fundamental identity characteristics. Just as judges have endorsed exploration in these analogous situations, so too should they recognize pre-adolescent children’s need to explore their gender identity.

Third, this Article provides decisionmakers with the practical information they need to resolve their cases. The “best interests of the child” standard is largely discretionary, which provides family courts with flexibility to promote children’s well-being. However, to reframe their approaches, family courts must have the relevant scientific data and precedent. Additionally, this Article argues that legislators should enact statutes that reinforce gender identity exploration, which they are more likely to do when armed with the scientific evidence and precedent that this Article provides. As a result, this Article’s analysis of medical literature and fundamental identity cases is more than a framework for understanding the issue—it is also part of the solution to the problem.


A tweet turned Luna Younger’s personal struggle into a national controversy. Using 148 characters, Texas Governor Greg Abbott announced that the Texas Attorney General’s Office and the Texas Department of Family and Protective Services would be investigating the seven-year-old’s family.1 Prompting his declaration was a jury’s award of custody to Luna’s mother, Anne Georgulas, a pediatrician who supported Luna’s gender transition.2 A year before the case made its way into court, Luna had asked her parents to call her Luna, rather than her (traditionally male) legal name,3 to reflect her4 gender identity.5 That same year, a therapist diagnosed Luna with gender dysphoria, which is distress from the mismatch between a person’s assigned sex at birth and their gender identity.6 As a result, medical professionals recommended that Luna be referred to as “she” and be allowed to wear the feminine clothing and keep the long hair that she preferred.7 Luna’s father, Jeffrey Younger, registered his objection to Luna’s gender identity by shaving her head, even as he allowed Luna’s twin brother to maintain his locks.8 Georgulas petitioned for an order prohibiting her ex-husband from “engaging in non-affirming behavior and/or taking Luna outside the home as [her birth name], or allowing others to do so.”9 Jeffrey Younger counterclaimed for sole legal custody.10

Luna’s case quickly became a symbol for broader debates over transgender rights and identity. Senator Ted Cruz and Donald Trump Jr. jumped into the fray, with the former describing Luna “as a pawn in a left-wing political agenda,” and the latter characterizing the situation as “child abuse.”11 As a hashtag referencing Luna’s birth name trended, legislators in six states introduced bills that would limit medical professionals’ ability to provide hormone treatments to gender expansive children, eliminate a court’s ability to consider a parent’s stance on gender transition in adjudicating the best interests of the child, or do both.12 The laws’ proponents accused parents of subjecting their gender expansive children to irreversible medical procedures, a claim far removed from medical reality.13 A pre-adolescent child’s gender transition is social, marked by clothing, hairstyle, pronouns, and name, rather than a physical intervention brought about by puberty blockers, hormones, or surgery.14

Even in cases where hyperbolic accusations do not take the place of facts, the ultimate question of what is in the best interests of a pre-adolescent gender expansive child—an umbrella term this Article uses to refer to transgender and nonbinary children, as well as children who are exploring nontraditional gender identities—may seem impossible to determine.15 For parents and courts alike, how much credence they should give to a child’s expressed gender is complicated, given how mercurial children can be. Scientists are unable to provide a definitive answer to this question, as studies have shown that, although adolescent children’s gender identity is stable, there is no way to predict whether a pre-adolescent child’s gender variance is fleeting or permanent.16 Exacerbating this scientific uncertainty as to how pre-adolescent children will identify as adults are political debates over transgender rights and identity, which together have left parents wondering whether they should affirm or oppose their children’s gender expansive identity. Notably, neutrality is not an option, as any response puts weight on one side of the scales.

This Article provides the first comprehensive analysis of custody cases between parents who take different approaches concerning their child’s gender identity.17 These cases are likely to be litigated in greater numbers as adults increasingly recognize that children articulate their non-normative gender identities at early ages.18 In analyzing these issues, this Article makes three distinct contributions, which it presents in three Parts.

First, this Article argues that courts adjudicating pre-adolescent gender expansive child custody cases are focusing on the wrong question—namely what is the “real” gender of the child. This analysis, presented in Part I, is based on a dozen gender expansive custody cases, identified through published court opinions and newspaper accounts. To provide a comprehensive account of gender expansive child concerns, this Part supplements the custody cases with related court battles, which involve foster care placements and petitions to change gender expansive children’s names.

Parents in these custody disputes typically ask the court to determine whether the child is transgender, and consequently to adjudicate which parent is correct as to the child’s ultimate gender identity. However, the science of gender identity development indicates that only adolescent children—defined as minors who have begun puberty—have a stable gender identity.19 The adult gender identity of pre-adolescent children, on the other hand, may be impossible to determine.20 According to scientists, what many of these children need is an opportunity to explore and come to terms with their gender identity.21 This part of the custody inquiry should consequently center on which parent is best suited to support the child’s gender identity exploration. By affirming pre-adolescent children’s right to explore their gender identities, courts can avoid an impossible question and come to the result that serves the children’s best interests.

The custody disputes involving gender expansive children that this Article discusses all arose in the context of petitions to modify custody, rather than initial custody decisions that determine what arrangement best meets the child’s interests. However, these cases demonstrate that parental support for exploration should be an important factor that courts weigh regardless of the case’s procedural posture. Legislative approaches to the best interests of the child standard can vary significantly across jurisdictions, with some states requiring courts to consider specific factors and others permitting judges to use their complete discretion. Regardless of which approach to the best interests standard any given state uses, one of the primary considerations that shapes courts’ decisions is the child’s physical and mental health.22 As a result, scientific conclusions regarding what promotes children’s health and welfare are integral to courts’ custody decisions more generally. This is, of course, not to say that parental support for identity exploration will outweigh other factors, but simply that it is an important element that courts should consider. Additionally, this Article does not resolve the question of how parents should support their children’s identity exploration. Rather, its argument is that courts should impose a supportive floor.

Second, this Article reveals that courts have recognized children’s need to explore their identities in related doctrinal areas.23 As Part II explains, although what constitutes the best interests of gender expansive children is a new question for courts to resolve, it is also the latest iteration in a long line of family court cases that form a framework for addressing these disputes.24 By compiling and analyzing more than 90 cases involving sexual orientation, race, and religion, this Article demonstrates that courts have recognized identity exploration as essential to children’s well-being.25 The relevant issue in these cases is the child’s need to explore these identities, as opposed to the way in which the child will ultimately identify.26 As a result, children’s exploration is an inherent element of the best interests analysis.27

Cases in which children’s sexual orientation, race, and religion are at issue all involve, in a general sense, situations involving children’s identity explorations.28 The sexual orientation cases involve children who identify as gay, lesbian, or bisexual, and whose parents either reject them because of their sexual orientation or attempt to change their sexual orientation. Sexual orientation as well as sexual identity may develop over time.29 Individuals also experience a period of social transition as they express this identity, as the phrase “coming out” reflects. Biracial and multiracial children’s races do not change, but their identification might, as they may come to see themselves as one, multiple, or a combination of races. As for cases of religious conversion, they typically appear in court because one parent has begun to follow different religious practices or ceased to follow previous customs,30 raising questions as to how much the adult may expose the child to new or different religious doctrines and experiences.31 Religion seems categorically different from sexual orientation and race because it is not an immutable characteristic.32 And yet the three share a constitutional space precisely because they are so constitutive of identity. All of these traits are considered either innate or so important that no one should be asked to change them, and therefore receive heightened protection under the U.S. Constitution.33

The cases are not perfectly analogous, but they nevertheless demonstrate a judicial recognition of children’s need to engage in identity exploration. In the sexual orientation cases, children are often identifying as gay, lesbian, or bisexual, rather than questioning their sexual identity, and therefore the conflict typically arises after the children’s exploration process is complete. These cases are still important because they highlight that parents must tolerate the end result of the exploratory process. The race cases are more akin to the gender expansive disputes in that the children are coming to terms with their identity. These parents do not assume that their children will identify as any specific race, which is a particularly important premise for the gender identity context.34 Moreover, courts typically award custody to the parent best positioned to support the child’s identity exploration, a commitment judges have not undertaken for gender expansive children. As for the religion cases, they involve parents who disagree over religious practices, rather than the children’s religious identity. However, they are nevertheless instructive because courts in these disputes focus on the effect that exposure to different religions has on the children.

This Article’s argument that exploring identity serves children’s best interests thus shifts the focus of the analysis from the fact of transition to the exploration that any transition requires, as well as the role of parents during the transition process. Together, these cases emphasize the potential for fluidity and change among fundamental identity characteristics, as well as delineate parents’ obligations to tolerate, if not affirm, their children’s identity development. These cases are consistent with the broader best interests of the child framework, which evaluates parents’ ability to provide children with the nurture, love, affection, and guidance they need to develop into self-sufficient adults.35

By clarifying the scientific points of contention in the gender identity disputes and presenting the related precedent, this Article makes a third, more practical, contribution: It provides litigants, courts, and other decision-makers with the information they need to better serve children’s interests. Family courts rarely enforce formal rules, but rather are guided by discretion, which allows them to respond to the unique circumstances of each family.36 As Part III explains, because of this discretion, family court judges are able to pivot when they receive new information about what children need. Historical examples involving gay and lesbian parents, as well as parents who engage in bondage, dominance, and sadomasochism (BDSM), suggest that family law judges will reframe their inquiries based on the social science evidence that parents introduce.

However, history also indicates that not all judges will respond to the evidence they receive. As a result, Part III argues that legislatures should amend the best interests statutes to explicitly recognize children’s need to explore their gender identity. In addition to reframing how courts adjudicate custody disputes, such a change would benefit families with gender expansive children more generally. That is because, in addition to prescribing conduct, laws convey expressive messages that, in this circumstance, would emphasize that all parents—not just those disputing custody—need to support their children’s gender identity explorations.37

Legislative change will promote uniformity across the jurisdiction, but it can be slower than judicial action. Judicial decisions have the benefit of speed, but lack the scope of statutory change. As a result, gender expansive children need both judges and legislators to act. These decision-makers are more likely to respond when armed with the relevant evidence and precedent. As a result, this Article’s analysis of medical literature and the fundamental identity cases is more than a framework for understanding the issue—it is also part of the solution to the problem.

I. Exploring Gender Identity

The main reason that gender expansive child custody cases are so complicated and fraught is that courts are attempting to adjudicate a question with an unknowable answer: whether a pre-adolescent gender expansive child is male or female. This framing is flawed, however, because scientific research indicates that only an adolescent child’s gender identity is stable.38 The gender identity of pre-adolescent children, on the other hand, may be impossible to determine until they are older.39 As a result, what many children need is an opportunity to safely explore their gender identity. Courts’ best interests of the child inquiries should therefore be focused on which parent will best be able to support the child’s gender identity exploration.

A. The Framing Problem

Parents who ask courts to identify gender expansive children as their sex assigned at birth typically make one of two arguments. Some claim that their children are not transgender because they do not conform to all of the stereotypes associated with their expressed gender. Others insist that their children’s gender expansive presentation is not authentic, but rather a response to outside pressure. Although these two claims are categorically different, they are problematic for the same reason: They put courts in the position of predicting the child’s adult gender identity. Notably, parents typically do not make explicitly antitransgender arguments, although political, social, and religious perspectives may be why they contest their children’s gender expansive identity.40

1. Gender Nonconformity

Gender nonconformity is both a central argument in these types of cases as well as a catalyst for these disputes. Gender conformity plays an outsize role because social markers of gender—such as clothing, toys, friends, and hairstyle—are often how children express their gender identity.41 How a child dresses, the toys they play with, and the gender of their friends all become tools to forecast the child’s future gender identity.42 Moreover, gender is a significant organizing principle in children’s lives.43 Children’s merchandise, from toothbrushes to bedding to snacks, are often gendered, with pink the default color for girls’ items and blue for boys’ products.44 In 2015, retailer giant Target began phasing out gender-based signage after consumers complained, but continued to label toys as “girls” and “boys” online because shoppers searched for merchandise by gender.45 The products that children consume indicate just how much gender serves as an organizing principle for children, a phenomenon that begins prebirth.46 Pregnancy announcements are often followed by a question as to the baby’s sex, if not a gender reveal party.47

In the custody cases, when a child does not exhibit all of the gender stereotypes of their claimed identity, parents often argue that this establishes that the child is not transgender or gender expansive.48 In Williams v. Frymire, a 2012 case from Kentucky, the father sought custody of the five-year-old child after the mother said the child was transgender and “she would begin transitioning [the child’s] gender from girl to boy.”49 The father objected, arguing that the child was female because they conformed to feminine gender norms. At trial, the father called his sister to testify, who noted that the child “played like a girl,” even though she conceded they were not a “‘girly’ girl.”50 The court, which agreed that the mother had “over-respond[ed]” to the child’s gender nonconformity, ultimately modified custody and made the father the primary residential parent.51 Likewise, in the Younger case, Luna’s father argued that Luna did not have a female gender identity because she was comfortable dressing “as a boy” in her father’s home.52 The judge concurred and reaffirmed the joint custody order.53 In these and other cases, the children’s gender conformity served as evidence that they were not transgender.

There are two fundamental problems with parents’ arguments concerning their children’s gender conformity. First, what undergirds these arguments is a flawed conception of gender, one that is rooted in biological essentialism. According to this formulation, sex is a dimorphic set of biological traits, based on reproductive organs, chromosomes, and hormones, and gender necessarily aligns with that biology.54 Much of the work of feminist rights advocacy was to disaggregate sex and gender, with gender defined as the social implication of sex.55 Identifying gender as a set of social, cultural, and political assumptions that are separate from biological difference—even as they are associated with physiology—formed the basis for sex equality gains.56

The line between gender nonconformity—which indicates a potential transgender identity—and gender role nonconformity—which feminists have challenged for decades—is an important one for parents in these cases. A girl who prefers short hair to long locks and track pants to skirts may be transgender, but she also may be a “tomboy.”57 The same parents who object to their children’s gender nonconformity might not consider gender role nonconformity a problem.58 For example, Bradley’s mother was untroubled when he played with dolls, had female friends, and identified with female characters—until she thought Bradley might be transgender.59 At that point, she consulted with a psychologist at Toronto’s Child Youth and Family Gender Identity Clinic.60 At the clinician’s urging, she removed Bradley’s “girl” toys to help him identify as male.61 In these cases, parents are often insisting that their children not cross the boundary between gender nonconformity and gender role nonconformity.

The second reason that these parents’ arguments about their gender expansive children are problematic is because they fail to recognize that gender exists beyond male and female. Gender is a spectrum, with identities that may range between male and female.62 Individuals may also be nonbinary, agender, or have some other type of gender expansive identity.63 Approximately 1.2 million LGBTQ adults in the United States identify as nonbinary—a number akin to the population of Dallas, Texas.64 In recognition of nonbinary individuals, a growing number of states have amended their identity documents to include an “X” option in addition to male or female.65

Parents’ incorrect understanding of gender identity reflects the extent to which nonbinary identity is invisible in society, politics, and law. That, in turn, is partly a function of how transgender rights advocacy evolved.66 Trans rights advocates initially framed transgender identity as a matter of being “born in the wrong body,” requiring surgery to make individuals’ physical presentation match their psychological self and to alleviate the psychological distress the incongruence produced.67 This “mismatch” explanation reflected some individuals’ experiences, but it was also a means to an end, with medical providers refusing to provide hormones and surgeries to trans individuals who did not fit this convention.68 To access the treatments they so desperately needed, trans individuals told doctors accounts of their lives that matched the diagnostic models, excluding any deviations from their narratives.69 Early trans advocacy legal efforts were based on that same medical model, which assumed transgender identity involved transitioning from one gender to another.70 As a result, arguments around trans rights tended to ignore gender fluid and nonbinary identities.71 For example, advocacy around gender identity markers on government documents began with states allowing individuals to modify the gender listed on their birth certificates and driver’s licenses only after undergoing gender confirmation surgery.72 After decades of lobbying, states began eliminating the surgical requirement, although they continued to require proof of clinical treatment.73 Despite their longstanding work in this area, advocates only recently began including nonbinary gender markers in their reform efforts.74

As their work around gender markers indicates, transgender rights advocates are increasingly highlighting the existence and needs of nonbinary individuals. However, nonbinary individuals are often left out of conversations around transgender rights.75 Many nonbinary individuals let others assume they are men or women, rather than explain nonbinary gender identification, and most avoid asking employers to use their correct pronouns to hide their gender identity at work.76 Nonbinary youth suffer disproportionately from anxiety, depression, and low self-esteem,77 which may be linked to their social and political invisibility.78

Parents’ arguments about their children’s gender conformity are thus rooted in two distinct misconceptions about gender identity, but both reflect that gender is an especially salient trait for parents. Parents face incredible social pressures to identify their child as male or female, as the experiences of parents of intersex children demonstrate.79 Intersex individuals are people born with variations in sex characteristics, such that sex indicators like genitals, gonads, or chromosomes do not clearly establish them as male or female.80 Parents of intersex children often consent to irreversible medical procedures on their infants, a decision influenced in part by their intense need to know whether they are raising a daughter or a son.81 It can be difficult for parents to bond with a child they refer to as “baby,” rather than “he” or “she,” let alone explain to relatives, friends, and caregivers that they do not know whether they have a girl or a boy.82 Gender’s prominence in the lives of parents is thus a factor that undergirds the custody disputes, leading them to center their arguments around flawed understandings of gender identity.

2. External Pressure

In addition to highlighting their children’s gender conformity, parents also often make a second argument: that the children’s gender expansive identity is a response to outside pressure, rather than an authentic expression of self. In a study of custody litigation involving transgender and gender nonconforming children, all 10 of the mothers who affirmed their children’s cross-gender identities explained that their ex-partners blamed them for “causing the children to assert transgender identities.”83 In one of these cases, a community member called child protective services to investigate another mother, claiming that the affirmation of the child’s transgender identity constituted “child abuse.”84

Published decisions reinforce the trend that this study of transgender child custody cases identified. In a 2007 case from Ohio, the father filed for full custody after learning the mother planned to enroll their son in a new school under the name of Christine.85 The father believed that his child’s behavior “was an attempt to gain attention and to win approval of his mother,” a conclusion the trial court ratified when it ordered that the child be “disassociated with [the transgender] lifestyle.”86 Although most cases involve claims of parental pressure, courts are also attuned to the potential effects of peers. When 15-year-old H.C.W. sought to change his name to conform with his gender identity, the probate court asked H.C.W. whether he had any transgender friends to satisfy itself that H.C.W.’s expressed gender identity was authentic, rather than a product of his peers’ influence.87

Parents’ and courts’ concerns about the effect of authority figures in the child’s life reflect the fact that identity formation is both personal and social.88 There are psychological and personal components to identity formation, insofar as individuals must come to recognize their identity as well as express it.89 At the same time, identity is a social phenomenon; identity both shapes community roles and expectations and is contoured by those same external forces.90 Because identity solidifies through day-to-day interactions, it evolves over time.

Parents are an essential part of a child’s social world and therefore have a significant influence on a child’s identity. Psychological literature is replete with studies of how parents help their children negotiate their identity from childhood to adolescence and adulthood.91 Parents facilitate their children’s identity development by communicating their views and values, which occurs directly through instruction and indirectly via the role models they provide.92 As Anne Daily and Laura Rosenbury have argued, “[a] child’s sense of self and belonging emerges early on and both strengthens and changes with age. This evolving sense of identity is first and continuously forged through children’s relationships with their parents, but children’s identities are also always to some degree independent of parental influence.”93 Parental guidance in the realm of gender identity formation is thus another dimension of parents’ broader work of raising children, which family courts regularly address.

Parents’ strenuous arguments that their child is not transgender—regardless of whether their claims are rooted in the children’s nonconformity to gender stereotypes or because they blame parental pressure—may be linked to the effect of the uncertainty or change on the parents themselves. As a general matter, “a transition for one family member affects the entire family system.”94 Changes to a child’s gender identity therefore may produce a concomitant period of uncertainty for how their parents self-identify. A mother with a son might become a mother with a daughter, or the mother of a nonbinary or agender child. Researchers have consequently argued that “families with transgender youth are all transitioning.”95 For that reason, courts have sometimes required the parents of gender expansive children to participate in therapy.96

It is possible, and indeed likely, that some parents oppose their children’s gender exploration because they are against gender variance for social, political, or religious reasons.97 In Kristen L. v. Benjamin W., for example, the mother told their older child that “it’s sick for [the younger child] to play with girl toys.98 The father successfully sought sole custody after seeing bruises on the child’s back that the mother had inflicted.99 Parental abuse in response to children’s expression of identities with which parents disagree is all too common.100 However, parents in these cases do not tend to make explicitly antitransgender arguments,101 nor do they assert religious liberty claims.102 Parents also do not frame their objections as a concern that children will suffer stigma as a result of their gender identity, which might be a relevant issue in adjudicating what is in a child’s best interests.103 Instead, parents’ disputes center around whether the child’s gender nonconformity is authentic—and will thus persist.

B. An Unknowable Answer

Although parents are asking courts to adjudicate which parent is correct regarding the child’s ultimate gender identity, scientific research has shown that only adolescent children’s adult gender identities are stable. Pre-adolescent children’s future gender identities, however, may be impossible to determine. Medical providers’ recommended approach consequently focuses on giving these children the time and opportunity they need to explore their identity. Given that children’s health and welfare is a critical part of the best interests of the child inquiry, this science should inform how courts adjudicate these cases.104

Scientific research distinguishes between pre-adolescent and adolescent children, with the onset of puberty dividing the groups.105 Public health organizations typically define adolescents as minors between the ages of 10 and 19, but children will reach puberty—and therefore become adolescents—at different ages.106 Because studies have shown that adolescent youth have gender identities that are likely to be permanent and stable, the treatment for adolescents involves affirming their gender identity. The first step is social, allowing the adolescent to present themselves according to their gender identity through clothing, hairstyle, name, and gender pronouns.107 The next three stages of treatment all involve medical interventions. The first type, which is entirely reversible, consists of hormone suppressants to delay the onset of puberty, which gives the adolescent more time to have a clear and definite sense of their gender identity.108 Because some gender expansive adolescents experience their body’s masculinization or feminization as excruciating, puberty suppression can be key in preserving these adolescents’ mental health.109

The next phases of medical treatment for adolescents are more consequential because they may have permanent effects. After they turn 16, adolescents may move to the second stage, the administration of cross-hormones, which initiates puberty in the gender in which they identify.110 Some of the physiological changes are reversible, while others—like a deepening of the voice as a result of testosterone—are not.111 The third stage, surgery, is generally reserved for those who are at least 18 years old, though doctors may offer mastectomies earlier.112 Some transgender individuals choose to undergo hormone therapy and surgery, but many do not want any or all of the medical treatments available.113

Scientific consensus concerning treatment for pre-adolescent children, typically children under the age of 10, is much less precise because the children’s ultimate gender identity is often unknown.114 There are three main approaches to gender expansive pre-adolescent children: reparative, affirmative, and watchful waiting. The reparative method identifies gender expansive identity as pathological and undesirable; it consequently attempts to have the child’s gender expression match their assigned sex at birth.115 To tamp down on any kind of cross-gender expression, parents must police children’s everyday activities, from the toys a child plays with to the gender of their friends to the colors they use to draw pictures.116

Scientists have resoundingly rejected the reparative approach. The World Professional Association for Transgender Health Standards of Care, for example, identifies treatment “aimed at trying to change a person’s gender identity and expression to become more congruent with sex assigned at birth” as unethical.117 The American Psychiatric Association and the American Academy of Pediatrics have likewise criticized efforts to change transgender individuals’ gender identity.118 Because of its documented harm to patients’ mental health, many states prohibit licensed mental health professionals from offering reparative therapy, which is also called “conversion therapy.”119

The other two methods—the affirmative and watchful waiting approaches—are both supportive of children’s gender identity explorations. Where they differ is in the extent to which children should express their gender identities. The affirmative approach facilitates children’s expression of their gender identity in all aspects of their lives. The watchful waiting approach, on the other hand, delays having the child live as their expressed gender until puberty.120 Under watchful waiting, which is also called the Dutch Protocol because it was developed in the Netherlands, children will not change their names or gender pronouns. However, children may engage in other types of gender expansive behavior that the reparative approach prohibits, such as dressing according to their gender identity at home.121 The Dutch Protocol is not neutral, as “watchfully waiting” while a child dresses according to their gender identity effectively reinforces the behavior.122 Unlike the affirmative approach, however, a child is not permitted to express their gender identity in all circumstances. Notably, neither the Dutch Protocol nor the affirmative approach permits pre-adolescent children to undergo any medical treatment. Thus, when conservative news outlets claimed that Luna’s mother was seeking to “chemically castrate” Luna, they mischaracterized how medical professionals approach pre-adolescent gender expansive children.123 At the time, Luna’s mother was not exploring permanent changes; the main issues that Luna’s parents were debating were how Luna would dress and the name she would use.124

Medical providers have questioned whether watchful waiting permits as much exploration in the United States as it does in the Netherlands. Dutch culture is known to permit greater latitude in gender expression than the United States, making it possible for children to explore their gender identities without a full social transition.125 Indeed, Americans’ social discomfort with a child engaging in a period of gender exploration, and therefore not necessarily identifying as male or female, means that parents attempting to follow watchful waiting may veer into reparative behavior.126 Additionally, parents who oppose gender expansive identities for personal, social, or religious reasons have claimed they are following watchful waiting because reparative is no longer an accepted option. For example, although Luna Younger’s father said he endorsed the Dutch Protocol, his actions were far from watchful waiting.127 He cut Luna’s long hair and required her to wear boys’ clothes, despite her protestations.128 The label that parents put on their approach is therefore less important than what it means in practice.

The Dutch Protocol restricts children’s abilities to express their identities because research has shown that gender dysphoria in childhood often does not persist through adolescence.129 In other words, gender expansive pre-adolescent children may not grow up to be gender expansive adults. According to the studies, the vast majority—between 73% and 94%—of pre-adolescent children with gender dysphoria did not later identify as transgender adults.130 Researchers emphasized that early transition for children whose gender dysphoria did not persist could be harmful because these children might delay a reversion to their sex assigned at birth for fear of being teased or shamed.131 The division between medical providers who endorse the affirmative approach as opposed to the Dutch Protocol stems from the question of whether having children fully express their gender identity outweighs the possible distress the children might later face if they ultimately transition back to a cisgender identity.132

What complicates the debate is that the studies documenting desistance have significant methodological flaws that undermine their conclusions.133 The researchers characterized every individual who did not return to their clinic for follow-up care as “desisters,” ignoring that the attrition may have been linked to other factors, like seeking care from another provider.134 Additionally, the research studies did not differentiate between gender conforming transgender children, meaning those who identified as male or female, and children with other types of gender expansive identities. Instead, the scientists simply subsumed all of the children within the broader category of transgender.135 Because the outcome of the gender exploration process may not be male or female, but rather a nonbinary identity, the desisting patients may simply have been mislabeled in the first place.136 Responding to these criticisms, the authors reevaluated the data and indicated that more children persisted than they originally thought.137

Research on the positive effects of gender identity exploration, even for those who ultimately desisted in their cross-gender identification, has called into question whether providers should even be concerned about desistance. A period of social transition—during which a child changes their name, pronouns, and appearance—can be psychologically beneficial, even if the child later identifies with their sex assigned at birth or adopts a different gender identity.138 Thus, a child such as “Alex,” who transitioned from male to female to nonbinary and uses they/them pronouns, explained the process was “necessary to fully solidify their gender identity.”139 Moreover, preliminary studies have shown that gender expansive youth who socially transition have lower levels of anxiety and depression than those who do not.140 For these reasons, some researchers emphasize that desistance does not necessarily mean that the child’s exploration was inappropriate or harmful.141 Instead, exploration can and does serve the children’s best interests because it allows children to be confident in their gender identity.

Despite debates over which supportive approach is best, medical providers agree that how a pre-adolescent child ultimately identifies is less important than relieving a child’s dysphoric distress.142 The aim of both the affirmative and watchful waiting approaches is therefore to ease the child’s anguish over their gender identity. Rather than restricting youth to the expression of any specific identity, health care providers seek to offer opportunities for youth to express themselves.143 What this indicates is that courts would better serve children’s interests by focusing on which parent best meets the pre-adolescent child’s need for exploration, rather than attempting to identify the child’s ultimate gender identity. Of course, reformulating the initial inquiry does not resolve the thorny questions of how or how much a parent should support a child’s gender identity exploration, but it does provide courts with the correct framework from which to evaluate these issues.

C. Custody Confusion

Courts’ efforts to answer a question with an unknowable answer has produced stratified results that are often not in the child’s best interests. They tend to follow scientific principles when adjudicating cases involving adolescents, allowing them to transition, but they do not consistently adhere to medical views when addressing pre-adolescents. Courts’ decisions thus reflect one part of the scientific consensus, that the gender identities of older children are relatively constant. However, most courts are ignoring scientists’ other conclusion, that gender expansive pre-adolescent children need an opportunity to explore their identities.

Courts tend to award custody of gender expansive teenagers to their affirming parent, demonstrating how scientific principles inform the best interests analysis in these cases.144 In an Ohio case from 2019, for example, a court awarded temporary custody of 14-year-old Skyler to his affirming mother.145 Likewise, a New York court granted a 17-year-old the right to wear “female attire” in the state’s “all-male” foster care facility, despite its policy prohibiting residents from wearing skirts or dresses.146 Courts have also approved the name change petitions of several transgender teenagers in decisions that recognize the “anxiety, embarrassment or discomfort that may result from the child having a name [that the child] believes does not match his or her outward appearance and gender identity.”147 As children become adolescents, the incongruity between their name and appearance may become more pronounced, particularly because transgender teenagers may pursue hormonal treatments. With transgender teens applying for driver’s licenses and college admittance, the implications of their legal names also become more significant.148

Courts have reached similar results even in cases that involved potentially irreversible medical treatments for the gender expansive teenagers. In 2018, an Ohio case made headlines when a judge removed a gender expansive teen from his parent’s care and placed him with his gender affirming grandparents.149 Doctors had advised the court that medical treatments were necessary to reduce the child’s suicide risk.150 The teen had emailed a crisis hotline after one of his parents “told him to kill himself and refused to let him get therapy ‘unless it was Christian-based.’”151 His parents’ attorney argued that the teen “was not ‘even close to being able to make such a life-altering decision at this time,’” although the parents’ religious objections were what seemed to drive their opposition to the medical treatments.152 In response to the case, Ohio Representative Thomas Brinkman introduced H.B. 658, which provided that parents had the right to determine what was in the best interests of gender dysphoric children, to “strengthen parental rights” and avoid judicial “overreach.”153 The proposal, which did not make it out of committee, also would have required schools to immediately inform parents if a child “demonstrate[d] a desire to be treated in a manner opposite of the child’s biological sex.”154

Cases involving pre-adolescent gender expansive children, on the other hand, have produced disjointed results. That fact reflects that the courts are attempting to predict the child’s ultimate gender identity, despite scientific evidence that this is often impossible to do. As Table 1 indicates, judges have granted custody to parents who follow the affirmative, watchful waiting, and reparative methods. Courts often express their frustration with the lack of clarity around which treatment model is best, with one judge stating that he was “humbled” by his “lack of qualification to resolve the gender issues” before him.155 Another blamed medical providers, stating that the medical profession “let [the child] down” as a result of the contradictory and inconsistent information that the child’s medical providers presented.156 What that court failed to recognize was that no doctor could have accurately predicted the child’s ultimate gender identity.

Table 1: Gender Expansive Child Custody Cases

[Table 1 is a PDF Download]

The cases in Table 1 reveal that, when parents disagree as to whether they should support their child’s gender identity exploration, courts often grant the parents joint legal custody. That may appear to be a neutral solution, but it instead gives the unsupportive parent a great deal of influence on the child’s ability to express their gender identity.157 This was the ultimate result of Paul E. v. Courtney F., which involved a child named “L.”158 L. was biologically male but began showing interest in toys and clothing stereotypically associated with girls at an early age.159 The mother sent five-year-old L. to school dressed in a skirt and provided L.’s teacher with a “Princess Boy” book to read to the class.160 The father disagreed with this approach, and as the mother became more insistent that they affirm L.’s identity, he petitioned the court for an order removing “girl-oriented toys” from the mother’s home and prohibiting the mother from “dressing L. in clothing generally worn by girls, referring to L. with feminine pronouns, and discussing gender-related issues with L. and the other children.”161 The trial court granted the father’s petition on a temporary basis, then awarded him sole legal custody two years later.162 In the permanent order, the court mandated that the father take L. to a specific psychologist for therapy and extended the restrictions on the parents’ ability to discuss gender issues with their child.163 The appellate courts vacated these limitations as unlawful restrictions on the father’s decision-making authority, thereby allowing the father to continue taking reparative actions.164

The trial court’s decision in Paul E. is not an outlier, as Table 1 indicates. Other courts have permitted parents to pursue reparative treatment,165 which is both outmoded and discredited.166 Moreover, in a study of custody challenges involving 10 affirming mothers of gender expansive children, many of the surveyed parents with pre-adolescent children lost custody to their unsupportive co-parents.167 Likewise, in the Younger case, the trial court overturned the jury’s award of sole custody to Georgulas and instead reaffirmed the joint custody arrangement.168 In doing so, the court not only ignored the father’s reparative actions, but also overlooked additional concerning behavior on Younger’s part.169 Georgulas had obtained an annulment from Younger after the family court found that Younger had lied to his ex-wife about his former marriages, education, and prior income, debt, and earnings.170 Younger also subjected Georgulas’s daughters to such extreme psychological abuse that one developed a suicide plan, while the other began cutting herself.171 Given the uncertainty around how courts will respond when the case involves pre-adolescent children, specialists in this field counsel parents with a partner who disagrees as to their child’s treatment to “proceed cautiously” before affirming their children’s gender expansive identity.172

The pre-adolescent gender expansive child custody cases have yielded unpredictable results, with courts granting custody to parents who follow any of the three approaches to gender expansive identity. These disparate outcomes reflect the fact that courts are framing their inquiries incorrectly, asking what pre-adolescent children’s adult gender identities will be, rather than recognizing that medical providers cannot forecast this with certainty. Instead, these children are best served when they have an opportunity to explore their gender identity. What may help courts in reframing their focus in the gender expansive child custody cases is that family courts have ratified identity exploration in related areas.

II. Exploring Other Fundamental Identities

Although gender expansive children’s identity exploration may be a new problem for courts, these cases are a variation on a longer line of precedent involving constitutionally protected identities: sexual orientation, race, and religion. Each of these identity categories is considered fixed, or at least so fundamental that no one should be asked to change it, yet they all may require some type of exploration to come to terms with the identity. Therefore, much like the gender expansive custody cases, courts are in the position of regulating parents’ approaches to their children’s identity explorations. The cases on sexual orientation emphasize that parents are obligated to tolerate their children’s identity, even if they disagree with it, while the race cases highlight how parents must affirmatively act to help their children in their identity explorations. The religion cases indicate that parents may not impose their views onto their children if doing so would cause the children harm. Taken together, what these cases demonstrate is children have a judicially recognized interest in exploring their identities because doing so best serves their needs.

A. Sexual Orientation

The sexual orientation cases highlight that parents must, at minimum, tolerate their children’s expression of identity. These cases, which are admittedly few in number, typically do not distinguish between children who are outwardly expressing their known sexual orientation and those who are coming to identify as nonheterosexual, even though the coming out process can involve both.173 This precedent thus does not directly address identity exploration, yet the cases nevertheless reinforce the science around gender expansive adolescents by establishing a floor of tolerance.

Describing sexual orientation as an identity that needs to be explored may seem odd given longstanding LGBTQ movement arguments as to the innate and unchanging nature of sexual orientation and gender identity.174 However, simply because a trait is something a person explores before understanding it does not mean that either sexual orientation or gender identity is mutable. There is a difference between requiring time, thought, and experience to comprehend an identity and altering that identity.175 Additionally, that many people are confident of their sexual orientation does not reduce others’ need to come to terms with who they are.

As Table 2 indicates, most published cases in which decision-makers addressed a child’s sexual orientation did not involve custody disputes. Indeed, because a search of legal databases, secondary sources, and newspaper reports revealed so few custody cases, Table 2 supplements those data points with other family law decisions involving children’s sexual orientation. More often, the conflicts arose in the context of child support. In Catherine W. v. Robert F., for example, the father sought to condition his support obligations on visitation with his children after the children refused to see him.176 The court noted his son, Robbie, had good reason to avoid his father; the father had told Robbie “you act like a queer” after learning that Robbie refused to strike a boy who had physically attacked him.177 Similarly, in Dzierson v. Dzierson, the father sought relief from his obligation to pay for college because of his son’s “abandonment.”178 What precipitated the son’s refusal to see his father was an exchange they had after the son had told his father he was gay. The father had replied that “he didn’t know how he would be part of his [s]on’s life” and that “he couldn’t accept that lifestyle.”179 The family court refused to modify the father’s support obligation.

Table 2: Sexual Orientation Family Court Cases

[Table 2 is a PDF Download]

Also more typical than custody battles are termination of parental rights and dependency proceedings, which again emphasize adults’ need to tolerate the children’s identity expressions. In In re Lori M., a mother petitioned to have her 15-year-old daughter adjudicated as a person in need of supervision because her daughter was bisexual and was “associating with a twenty-one year old lesbian.”180 The court denied the mother’s petition, concluding that the teenager was “free to choose her sexual orientation without interference from this Court.”181 The court did note that the 15-year-old could not have a sexual relationship with her girlfriend given the penal code’s definition of statutory rape, and therefore admonished the teenager “to avoid engaging in any conduct violative of the Penal Law of this State.”182 Thus, the court recognized the adolescent’s right to explore and express her identity, so long as doing so did not violate the state’s criminal law.183 In a dependency case, the Los Angeles County Department of Children and Family Services removed 15-year-old C.O. after her mother scratched her face and threatened to have “a man rape C.O. to cure her of [her] bisexuality.”184 That decision primarily focused on the abuse that the child suffered, but the end result was to ratify the teen’s ability to express her identity.185

That parental disapproval of same-sex sexuality results in so many dependency cases, rather than custody disputes, is perhaps unsurprising given the disproportionate number of LGBTQ youth who are in foster care or homeless. LGBTQ youth are overrepresented in foster care, typically becoming wards of the state after their families reject them because of their sexual orientation, gender identity, or gender expression.186 Although only approximately seven percent of youth in the United States are LGBTQ, a study of Los Angeles foster care revealed that 19% of the youth living in foster care were LGBTQ.187 The data on youth homelessness are even more striking. LGBTQ youth comprise 40% of homeless youth, and LGBTQ youth are 120% more likely to experience homelessness than their non-LGBTQ peers.188

Much like the parents in the dependency cases, the state is often obligated to permit the youth in their care to express their sexual orientation. Twenty-seven states and the District of Columbia prohibit discrimination based on sexual orientation and/or gender identity in foster care;189 21 states and the District of Columbia also do so in their juvenile justice systems.190 Additionally, New Jersey has “safe zones” for gay youth, and several states have hired dedicated staffers to assist LGBTQ youth.191 That does not prevent experiences like that of Kenneth Jones, whose foster father threw him out of the house after Jones brought his boyfriend home one afternoon.192 More typically, foster families refuse to house a gay child, leaving child welfare agencies to put LGBTQ teens in group homes.193 Despite the impediments to finding accepting and loving homes for gay youth, the fact remains that most state agencies are mandated to do so—or at least place these youth in residences that allow them to express their sexual orientation.194

In addition to helping to identify children’s interests in identity exploration, the sexual orientation cases reinforce that reparative approaches are harmful to gender expansive children. Because gender expression and sexual orientation are so often conflated, the sexual orientation cases tend to arise when children exhibit gender nonconforming behavior. In Buxton v. Storm, for example, the father opposed his male child wearing earrings and sporting nail polish because this might indicate a nonheterosexual orientation.195 Likewise, in In re Shane T., the father repeatedly taunted his 14-year-old, Shane, with the slurs “fag,” “faggot,” and “queer,” justifying the verbal abuse as a means of curing Shane of “girlie” behavior.196 The father further explained that “it would be embarrassing to him if Shane were ‘queer.’”197 Because the fathers in both cases linked effeminacy and same-sex sexuality, they punished the children’s gender nonconformity. In Shane’s case, the court determined that the father’s conduct rose to the level of abuse and that it was in Shane’s best interests to be removed from his parents’ care.198 Similarly, another court removed D.M. from his grandmother’s home in part due to the taunts, name-calling, and shaming he experienced in her care.199 He repeatedly heard his grandmother tell him that he “is gay and should wear a dress and high heels.”200

The fact that parents view gender nonconformity as an indicator of a nonheterosexual orientation is logical, given that the most visible marker of same-sex sexual attraction—and the impetus for discrimination—is often men’s effeminacy or women’s masculinity.201 Indeed, that shared basis for discrimination is what led lesbian and gay rights activists to incorporate trans rights issues into their agendas in the late 1990s, giving rise to the LGBT movement.202 There are several other points of connection between sexual orientation and gender identity: Both are considered immutable, become evident as a child develops, and are often the target of social disapproval.203 This condemnation is often based on moral objections, which may lead parents to censure their children’s expression of their sexual orientation or gender identity.

Reinforcing the link between the two identity traits is that gender nonconformity in childhood is often an indication of a later nonheterosexual sexual orientation, rather than a marker of transgender identity. The same research studies on the desistance of childhood gender expansive identity found that most of the desisting children identified as gay or lesbian in adulthood.204 Gay and lesbian adults often report being shamed for their gender expansive expressions in childhood, an experience that has long-lasting adverse effects on their mental health.205 What this indicates is that the vast majority of gender expansive children will be harmed by parents’ negative responses to their gender presentation and efforts to restrict it, even if their gender dysphoria desists after adolescence.206

Because gender expansive children will overwhelmingly grow up to identify as members of the LGBTQ community, attempting to restrict children’s cross-gender expression communicates that there is something wrong being LGBTQ, no matter how these children ultimately identify. If parents have an obligation to tolerate their gay, lesbian, and bisexual children’s identity expression, the parents of gender expansive children logically should as well. These cases thus bolster the science of gender identity development, which indicates that parents should affirm their adolescents’ gender identities and support their pre-adolescent children’s gender identity explorations.

B. Race

Whereas the sexual orientation precedent indicates that parents must tolerate children’s expressions of their identity, the race cases impose an affirmative duty on parents to assist their children’s identity explorations. In custody disputes involving multiracial children, the relevant question is which parent can best help the child negotiate their racial identity, whatever that ultimately ends up being. Thus, this line of cases reveals another dimension to children’s interest in identity exploration. They shift the focus away from how parents should respond to the ultimate result of the exploratory process, instead highlighting adults’ role in helping children develop that identity.

Like sexual orientation and gender identity, race is constitutionally protected as an immutable characteristic.207 Presenting race as a matter of exploration may seem odd or inappropriate because children are not changing their skin color.208 However, race is both a matter of personal identification and social attribution, and thus racial identity is a trait that children explore and embody over time. Racial categories are social designations, rather than biological markers, such that the social significance of race is what renders the child’s identification so fraught.209 Historically, biracial individuals with a White parent identified with their minority heritage, but they are now increasingly identifying as biracial or multiracial.210 In doing so, these children may have a racial identity that differs from that of their parents, much like nonbinary or agender children. They may also defy social expectations that they “should” identify as one racial identity or another, which likewise renders their situation similar to that of gender expansive children.211

Until the mid-1970s, courts adjudicating custody cases between Black and White parents often granted the Black parent custody on the assumption that, because the child would identify and be identified as Black, the Black parent would be better able to help the child negotiate their “place in soci[e]ty.”212 This racial default was problematic for several reasons, with one scholar explaining that “it cannot be reliably assumed that a parent of the same . . . race as the child will be better able to model that identity than . . . the other parent.”213 In other words, just as a mother does not necessarily serve as a better model of behaviors and attitudes for her daughter than the father, courts cannot categorically assume that a White woman has less understanding of her biracial child than the child’s Black father.214 Additionally, the racial default reflected how society would view the children, rather than how the children would understand themselves.215

Courts have since eliminated the racial default, although decision-makers can and do consider race when adjudicating custody cases.216 The Supreme Court struck down interracial marriage restrictions in Loving v. Virginia217 in 1967, but race-based considerations continue to permeate family law.218 In the foster care and adoption context, although courts insist that placements cannot be based solely on race, social workers may consider race to weigh in favor of placing a mixed-race child with a non-White family.219 Thus, when the Millers, a White family, sought to adopt their biracial foster son named Shane, the social services agency denied their petition in favor of a Black couple.220 The Pennsylvania Supreme Court approved, explaining that placement with the Millers would “not seem to be very conducive to the inculcation of a sense of racial identity in Shane.”221 Whether agencies should pursue transracial adoptions has been the subject of intense debate among scholars, social welfare professionals, and activists alike.222

In the late 1970s and early 1980s, courts began recognizing that biracial children might have a racial identity that differed from that of either parent. In Farmer v. Farmer, the court noted that “a child of a racially mixed union potentially faces special problems unknown to a child whose parents are both [W]hite or both [B]lack.”223 For that reason, custody “ought to reflect . . . which prospective custodian is best able to guide the child, not only for the present, but when she confronts them in the future.”224 The appellate court in Raysor v. Gabbey came to a similar conclusion.225 It consequently remanded the case back to the lower court, which had granted custody to the child’s White grandparents over the objection of the Black father, for factfinding on how the grandparents would help the child integrate “not only with the [grandparents’] [W]hite community but with [B]lacks sharing her ancestry.”226

The courts were correct that multiracial individuals face unique challenges unlike those that monoracial individuals experience.227 Several studies have documented that multiracial individuals suffer from increased social exclusion and discrimination, as well as face disapproval from extended family members.228 Multiracial individuals report a constant social pressure to “choose” one of their racial groups, rather than identify as multiracial, even though multiracial individuals who identify as multiracial rather than with only one racial group often have higher levels of self-esteem and psychological well-being.229 Multiracial people also report being able to easily switch between or identify with multiple racial identities at any given moment, a fact that reinforces that racial identity may be fluid rather than fixed.230

As Table 3 indicates, when courts began recognizing that multiracial children might have a different racial identity than that of their parents, the relevant question became whether the custodial parent would support the child’s identity exploration. Initially, these decisions favored White parents.231 In Davis v. Davis, the court awarded the White mother custody after determining that “the child, although biracial, could just as easily be raised in the [mother]’s all-white household as he could in [the father]’s biracial household,” so long as the mother “was loving and nurturing.”232 Similarly, in Olivier A. v. Christina A., the court granted custody to the White father after explaining that, while it was important for the child to maintain his Nigerian cultural heritage, the father would not deny the child his cultural identity.233

Courts that framed the question slightly differently, by asking about the extent of the opportunities a child would have to explore their identity, rather than simply whether the parent would support the child in their exploration, were more likely to grant custody to the parent of color. In Jones v. Jones, the South Dakota Supreme Court reasoned that courts could properly “consider the matter of race as it relates to a child’s ethnic heritage and which parent is more prepared to expose the child to it.”234 As a result, it upheld the trial court’s custody award to a Native American father.235 Similarly, in In re Marriage of Gambla, the trial court granted custody to the Black mother in part because she would be able to provide the child with a “breadth of cultural knowledge and experience” from her own life as an African American woman that the White father could not.236

Table 3: Bi/Multiracial Child Custody Cases

[Table 3 is a PDF Download]

In making these custody awards, courts considered how parents addressed race in the home, as well as the opportunities that the children would have to learn about their racial, ethnic, and cultural heritage in their communities. Thus, in Hamilton v. Hamilton, a case in which the parents lived in different cities, the court noted that, although the children would live primarily with their White father, he showed “awareness and concern for the problems the children may face” in “fitting in” at the school where he lived.237 In Ebirim v. Ebirim, the court weighed the racial composition of the cities where the parents lived because this would affect the children’s ability to be in daily contact with racially diverse individuals.238 Although the White mother lived in a district with no Black children, she affirmatively introduced her son to biracial and minority children in the area.239

That the biracial and multiracial custody cases focused on which parent would most effectively promote a child’s racial identity development reflects that race and cultural heritage are learned, despite being immutable characteristics. Whereas the sexual orientation cases indicated that parents had to tolerate their lesbian, gay, and bisexual children’s identity expression, the race cases highlight that parents may have a positive duty to help their children explore their identities. This difference is particularly important because tolerance and affirmation are not the same things. Studies have indicated that there is a palpable difference between how children and young adults experience neutrality and enthusiasm.240 LGBTQ youth who experience moderate acceptance from their families report lower rates of self-esteem, higher rates of depression, and poorer health overall than children whose families exhibit high rates of acceptance.241 Children may, therefore, experience neutrality as disapproval, even if it is technically not.

What helps explain the gap between courts’ approaches to these two types of disputes is that courts understand that the biracial or multiracial children may be undergoing a period of identity exploration, while in the sexual orientation disputes, courts generally assume that the children have formed their identities. As a result, sexual orientation in these cases appears as a fixed status, while biracial or multiracial identity appears as a trait with which children must come to terms. Taken together, these cases suggest that, while parents must tolerate the ultimate expression of a child’s identity, they must also affirmatively support its development.

Ultimately, the precedent on children’s racial identity makes explicit what the sexual orientation cases imply: that children may need to come to terms with their fundamental identities and that parents are an integral part of that process. Consequently, this precedent reinforces the science that underscores pre-adolescent gender expansive children’s need for exploration.

C. Religion

If precedent on multiracial children demonstrates that parents must support their children’s identity explorations, the cases on religion illustrate how courts balance the children’s interests with their parents’ constitutionally protected rights. These cases center on parents’ religious affiliations, rather than those of the children. They are nevertheless instructive because courts focus on how the parents’ actions affect their children’s religious identity. In these cases, courts must draw a line between parents’ and children’s freedom of religion. The need to explore identity mediates the two, with courts holding that the balance tips when there is harm to the child. Courts may restrict parents’ constitutional right to raise their children in a particular religion, or even require a parent to support their children’s conflicting religious principles, to promote the children’s best interests.242

Custody case law is replete with conflicts over religion, which typically arise after one parent converts and wishes to inculcate their new religious beliefs in their child, rather than when a child converts against the wishes of their parents.243 As a general matter, courts will not enforce premarital agreements to raise a child in a specific religion, although they may give effect to religion-related clauses in separation agreements.244 Children often observe the religion they were raised with, but some adopt the religion of their converted parent. Because “[t]he freedom to choose any religion necessarily comprehends the freedom to change religions,” courts respect the religious conversion of both parents and children.245 The fact that individuals can explore multiple religions, or change religions, is a particularly important point for the gender expansive child custody cases, given that gender may be as fluid.

While the children are learning about their parents’ religions, courts may identify them as being of neither religion. In Schwarzman v. Schwarzman, when the father sought an injunction prohibiting the mother from converting their children from Judaism to Catholicism, the court held that the children were “neither Jewish nor Roman Catholic.”246 Courts have yet to adopt a similar perspective on gender exploration, meaning courts have not recognized that a child may identify as neither male nor female.

Courts often credit children’s statements concerning their religious beliefs, a fact that stands in stark contrast to the skepticism many judges exhibit when youth discuss their gender identity.247 For example, in Bienenfeld v. Bennett-White, the court considered the 12-year-old’s unwillingness to be bar mitzvahed in making its determination.248 Some courts have also allowed younger children to make religious decisions, though these are fewer and further between. When a trial court ordered that an eight-year-old child could not be baptized in the religion of her choice until she was 13, the appellate court reversed, allowing the baptism to proceed over the father’s objection.249 The appellate court noted the lack of evidence as to a substantial threat of present or future physical or emotional harm to the child. A few courts have additionally weighed pre-adolescents’ preferences when their parents disagreed over religious decisions with irreversible physical consequences. In In re Marriage of Boldt, the mother sought custody modification or an injunction to prevent the father from having their nine-year-old child circumcised so the child could convert to Judaism.250 The appellate court ruled that the trial court should determine the then-12-year-old child’s preference before ruling on the motion.251 However, there are few of these types of cases where a child has made a decision to adopt one religion as opposed to another, as compared to disputes in which a parent has converted. As a result, it is not possible to use this line of cases to draw broad conclusions for the gender expansive child cases.

When parents’ and children’s religious views differ, courts have ordered parents to support their children’s religious upbringing. In McCown v. McCown, the parents had raised the children as Protestants, but the mother converted to Judaism after the separation and converted her daughters over the father’s objection.252 At the time of the conversion, the children were three and six years old; when the trial court addressed the dispute, the children had become active in their Jewish community.253 The trial court prohibited the father from enrolling the children in a Christian Sunday School or other similar religious educational program, as well as mandated that the children could follow their religious practices when they were with their father, which included observing the Sabbath on Friday nights and attending synagogue on Saturdays.254

When one parent converts after the couple has been raising the child in a particular religion, the parent who has converted can instruct the child in the new religious teachings, so long as doing so does not harm the child.255 Table 4 indicates just how clearly harm serves as the limiting principle in these cases. As one court explained, judges “should adhere to a policy of impartiality between religions, and should intervene in this sensitive and constitutionally protected area only where there is a clear and affirmative showing of harm to the children.”256 In re Marriage of Reyes illustrates this principle.257 In that case, the father had the child baptized after the parents divorced, even though they agreed to raise their child in the Jewish faith.258 The trial court refused to enjoin the father from taking the child to church or engaging in “other actions inconsistent with [the child’s] Jewish upbringing” and ruled the father could take the child to church services during his visitation.259 The appellate court affirmed the decision, noting that the mother had not demonstrated that the order would endanger the child’s welfare.260

Table 4: Religious Conversion Child Custody Cases

[Table 4 is a PDF Download]

Because courts must protect the converting parents’ First Amendment right to instruct their children on religious matters, the question becomes how much harm must befall a child before a court will intervene. A mere fear of future harm, or speculation that the exposure to conflicting religious beliefs will cause harm, is insufficient, as is confusion on the part of the child.261 Thus, in CLB v. DLO, the court denied the father’s petition to modify custody based solely on the father’s belief that “the child may face ‘opposition’ if he follows the mother’s religion.”262 Courts have also held instruction in two belief systems is not, in and of itself, harmful to children. In Greene v. Greene, the mother converted to Judaism prior to the marriage, but the family celebrated both Christian and Jewish holidays.263 After the divorce, the mother enrolled the children in Catholic Catechism classes, which the father sought to enjoin.264 The court denied his petition, reasoning that the classes were a means of exposing the children to their Italian Catholic heritage, rather than denying them their Jewish identity, and that this dual instruction would not cause the children harm.265

On the other hand, where a child begins exhibiting anxiety or where there is generally an adverse effect on the child’s welfare and happiness, courts have upheld restrictions on a parent’s religious activities.266 Many of the cases are vague on the deleterious effects that children have suffered, though some identify extreme consequences, including the children suffering from sleep hyperactivity, bed wetting, and fear of leaving their primary custodial parent.267 Courts have also identified less severe repercussions as cognizable harms, such as religious instruction that strains the relationship between either parent and child. In Baalla v. Baalla, the court granted the mother sole legal custody in part because the father had been pressuring the child to adopt Muslim practices, which adversely affected the child’s relationship with him.268 In Homonnay v. Homonnay, the mother’s conversion, and her efforts to introduce her children to her religion, had a negative effect on the children’s relationship with their father.269 As a result, the court ordered the mother to refrain from taking the children to any activities associated with her church or its members.270

Where there is harm, the religion precedent indicates that courts may prevent parents from inculcating their religious viewpoints, a limitation that demonstrates the importance of a child’s emotional and psychological well-being in the best interests of the child analysis.271 In LeDoux v. LeDoux, the court directed the father, who had become a Jehovah’s Witness, to not expose the child to any religious practices or teachings that were inconsistent with his Catholic upbringing because of the substantial stress the child experienced as a result of his indoctrination in two religions.272 In Homonnay, the court ordered the parent to stop attempting to instruct her children in her religion entirely.273

Courts have also imposed restrictions on indirect religious instruction. In Bienenfeld v. Bennett-White, the parents’ dispute centered on where to enroll their child; the mother, who had converted to Orthodox Judaism before the divorce, favored a Jewish school, while the father, who was Episcopalian, wanted their child to attend public school.274 The child was anxious about attending the religious school because he did not want to be bar mitzvahed and feared his peers would ostracize him.275 The court granted the father sole legal and physical custody, thereby giving him decision-making authority over which school the child would attend.276

Because of parents’ constitutionally protected freedom of religion, courts are careful to narrowly tailor their limitations. Thus, in Kirchner v. Caughey, the appellate court reversed a lower court order that prohibited a Baptist father from having his child participate in “any church or church-related activity” during his visitation periods.277 Although the child exhibited anxiety as a result of the father’s efforts to pursue his “fundamentalist beliefs in his conversations and dealings with the child,”278 the appellate court noted that events like church picnics were unlikely to cause the child harm.279 Moreover, the restriction did not prohibit the conversations giving rise to the child’s distress.280 Because the order was both too broad and too narrow, the appellate court remanded for further consideration.281 In the gender expansive child context, courts do not need to engage in such tailored line-drawing because parents do not have a constitutional interest in raising their child in a particular gender.

Courts similarly may require parents to take affirmative actions in support of their children’s religious identities. In Weisberger v. Weisberger, the case turned on the extent to which the mother, who had left the family’s ultra-orthodox Jewish community and adopted a secular Jewish practice, had to raise the children according to Hasidic practices.282 The father petitioned for sole legal and residential custody because the children did not dress according to Hasidic custom, speak Yiddish, or maintain the Sabbath while in their mother’s care.283 The conflict between the parents led the children to be “upset and confused” after spending time with their father, who disapproved of his ex-wife’s decisions.284 The appellate court ruled that it was in the children’s best interests for the mother to serve as the primary residential parent, as long as she abided by certain restrictions. The court ordered her to “make all reasonable efforts to ensure that the children’s appearance and conduct comply with the Hasidic religious requirements . . . while the children are in the physical custody of their father or their respective schools.”285 However, the court otherwise permitted the mother to exercise her discretion as to the children’s upbringing.286

Courts’ line-drawing at harm to children has particular resonance in the gender expansive child custody context because of the high potential for harm in those cases. Because gender expansive youth often endure rejection by their families, ostracism from their communities, and high rates of bullying and physical violence, they suffer disproportionately from depression, anxiety, and suicidality.287 Recent studies have revealed that more than 40% of adolescents who identified as not exclusively male or female, as well as more than half of adolescents who have transitioned from female to male, have attempted suicide.288 Thus, the 2018 case from Ohio, in which doctors recommended hormone therapy to reduce the teen’s suicide risk, is not atypical.289 Neither was a 2019 case from Ohio, in which 14-year-old Skylar was admitted to psychiatric hospitals on three separate occasions, the last one due to self-harm.290 The line-drawing at harm thus reinforces that courts should prohibit parents from adopting reparative methods, which medical professionals have identified as inherently harmful.291

These statistics, of course, do not mean that any particular gender expansive child will suffer from these harms. However, these data should lead courts to tread carefully before they permit parents to expose children to a discouraging approach. Applying the religious precedent would likely have led to a different result in some of the gender expansive child custody cases. For example, five-year-old L. expressed anger at being denied “girl-oriented clothes” and blamed her father for her inability to dress in clothing typically associated with girls.292 Likewise, Luna Younger expressed deep sorrow at not being able to express her gender identity and explicitly communicated that she feared her father.293 Were these results the product of a parent’s religious instruction, as opposed to a response to the child’s gender identity, courts would have identified them as cognizable harms and restricted the parents’ reparative actions.

The harm principle is also important for gender expansive child custody cases should parents argue that their religious beliefs require them to tamp down on their child’s gender identity exploration. As Part I.A noted, parents typically do not assert that their religious beliefs conflict with their children’s gender expansive identity. However, parents might claim that supporting their children’s gender identity exploration makes them complicit in the sinful conduct of their children, thereby violating their rights under the First Amendment or a Religious Freedom Restoration Act.294 In those circumstances, harm to a child would serve as a sufficiently compelling countervailing state interest to justify restricting the parents’ freedom of religion, as the cases on religious instruction indicate.

* * * * *

The precedent on religion, when combined with the race and sexual orientation cases, provides a helpful framework for the gender expansive child custody disputes. The sexual orientation decisions emphasize that parents must tolerate the expressions of an identity with which they might disagree, while the race cases indicate that parents may have an affirmative duty to support their children’s identity explorations.295 The religion cases identify harm as the dividing line between parents’ views and those of their children.296 Together, they establish that focusing on gender identity exploration, rather than the child’s ultimate gender identity, is what promotes children’s best interests.

III. Promoting Exploration

Given that judges have recognized that identity exploration is in children’s best interests, reframing pre-adolescent gender expansive child custody disputes to focus on exploration does not require any changes to the law that family courts already apply. If history is any indication, parents can remedy the situation by providing judges with the relevant scientific data and doctrinal precedent, which they will find in this Article. However, history also indicates that some judges will not respond to the information they receive and thus that legislators must also act. A statutory change has an additional benefit, in that laws have expressive effects that extend beyond the conduct they directly regulate.297 As a result, such a statute would express to all citizens—not just litigants in custody disputes—that it is essential to support a child’s gender identity exploration. These legislative and judicial options are not alternatives to one another, but rather should work in tandem to promote the needs of gender expansive children. While elected officials debate and consider legislation, judges can make decisions based on the science and precedent that promotes identity exploration.

A. Reframing the Best Interests Inquiry

Judges should be able to shift their analysis without a substantive change to existing standards. Family courts are institutionally designed to address fact-specific, evolving, and changing situations, given that they work with discretionary standards and have continuing jurisdiction over cases. Indeed, there is historical evidence to suggest that family courts will respond to scientific evidence and reframe how they approach gender expansive child custody cases.

One instructive historical example comes from another set of LGBTQ-related custody contests. Until the 1970s, courts refused to consider lesbian mothers’ and gay fathers’ custody claims, assuming that their sexual orientation was innately harmful to their children.298 Even after courts stopped automatically disqualifying gays and lesbians, judicial bias routinely resulted in these parents losing their custody cases.299

To counter judges’ concern that the adults’ sexual orientation would influence that of their children, litigants introduced scientific evidence showing there was no connection between parental sexual orientation and the sexual identity of children.300 This evidence eliminated one of the main bases for denying custody, and, as a result, judges around the country began granting lesbian mothers and gay fathers custody and visitation.301 In 1980, for example, a New Jersey superior court judge granted Rosemary Dempsey custody of her 10- and 11-year-old daughters, relying on psychiatrist testimony and research studies about the future sexual orientation of children with gay and lesbian parents.302 The Massachusetts Appeals Court likewise ruled in favor of a lesbian mother in 1983, citing the testimony of four unidentified psychologists, as did the Oklahoma Supreme Court in 1982, which granted a gay father custody of his twin 11-year-old sons in a decision that relied heavily on psychiatric expert testimony about the children’s future sexual orientation.303

Gay rights organizations and movement lawyers recognized the crucial role of this psychiatric testimony in custody disputes, circulating information about scientific studies and experts to lesbian and gay parents in informational packets.304 The attorneys at the Lesbian Mothers National Defense Fund and Lesbian Rights Project also drafted manuals for lesbian mothers that emphasized the role of expert testimony and highlighted the importance of psychiatric studies in custody determinations.305

The information that gay rights advocates distributed was useful for both litigants and their experts. In a 1980 lesbian mother custody case, an Arizona court required the attorney general’s office to introduce expert testimony on gay and lesbian lives.306 The psychologist that the office selected based his testimony in large part on the information in the National Gay Task Force’s packet.307 He later wrote to the organization to explain how useful the information had been.308 His argument appeared to convince the court, which awarded the mother custody.309

Like these cases involving gay and lesbian parents, gender expansive child custody disputes typically turn on expert testimony and advice. In Smith v. Smith, for example, each party called two expert witnesses to discuss gender dysphoria and treatment methods.310 In Johnson v. Johnson, the gender expansive child’s mother involved the child’s therapist and a psychologist in developing a plan for the court.311 Cases involving gender expansive children are replete with references to psychiatrists, psychologists, and psychotherapists, whose views inform courts’ decisions.312 Litigants in the gender expansive child custody disputes, like gay and lesbian parents, may be able to reframe how both experts and courts address the cases by providing them with the studies and scientific information concerning pre-adolescent children’s gender identity development.

Lending support to this solution is another, more recent example of sexual minority parents who have succeeded in attaining widespread change by altering the views of judges one courthouse at a time. For many decades, parents who engaged in BDSM typically lost custody and visitation rights as a result of their sexual relationships and practices.313 Until 2010, the American Psychiatric Association (APA) identified BDSM and kink as paraphilias, or atypical sex practices, that indicated mental illness in their diagnostic and statistical manual (DSM).314

Because the medical profession defined their sexual activities as psychopathic, BDSM practices were powerful weapons that ex-spouses wielded in custody battles.315 In 2008 and 2009, almost 300 people contacted the National Coalition for Sexual Freedom (NCSF), an advocacy group on behalf of consenting adults in the BDSM, leather, fetish, swing, and polyamory communities, for help with divorce and child custody issues that arose because of their sexual practices.316 NCSF estimates that between 1997 and 2010, 80% of parents who sought legal assistance from the NCSF lost their custody battles.317

After the APA revised its diagnostic code in 2010 to clarify that engaging in atypical sexual practices was not a mental illness, however, practitioners of BDSM were equipped to successfully petition for custody of their children. The percent of parents seeking legal assistance from the NCSF who lost their custody cases dropped from 80% to less than 10%.318

A custody case from 2010 demonstrates the direct legal effect of introducing evidence of the diagnostic change. The caseworker for the Department of Social Services (DSS) questioned whether the mother, who engaged in BDSM, could separate her “alternative lifestyle . . . from her parenting” given the DSM-IV-TR’s definition of sexual sadism as a mental illness.319 At the final hearing on the case, the mother’s lawyer provided the judge with the proposed DSM-5 revisions, asking the court to reevaluate the case.320 The next month, the judge awarded the mother custody.321 He also rebuked DSS for its failure to stay apprised of new diagnostic criteria.322

Just as the history of sexual minority parents provides a reason to expect that courts will adapt their perspectives based on new expert evidence, so too does it provide a warning as to the limits of case-by-case reform. In the gay and lesbian parenting context, scientific evidence did not sway all judges, particularly those in the South and Midwest. There, courts expressed concerns about the parents’ “immoral” behavior and concluded it was inappropriate to allow “children to be placed in a home where the felony of sodomy is committed at least twice a week.”323 Similarly, courts continued to consider BDSM a negative factor in weighing custody determinations despite the diagnostic change. In 2012, an Illinois court denied a father’s petition for custody after finding that the father’s BDSM practices meant that the “mental-health factor” favored the mother.324

Despite these cautionary examples, the slow process of change that sexual minority parents have attained provides a model for the parents of gender expansive children, who may be able to succeed in much the same way by drawing on the studies and cases that this Article has detailed. The fact that judges have recognized the children’s need to explore other identities should make family courts more willing to adopt the new framework. Because the scientific evidence and precedent are integral to the best interests analysis, the only legal change that should be necessary is for judges to deploy their discretion differently. Expert testimony and other types of evidence may thus change the outcome for many litigants, even though it will not likely change the minds of decision-makers who have entrenched views.325

B. Statutory Reform

As the historical examples illustrate, the same institutional features that will likely make family court judges responsive to information on gender expansive children are also those that will permit bias to impede reform. The result may therefore be a patchwork of judicial decisions that does not promote the interests of all gender expansive children. For those reasons, legislators should also enact statutes that emphasize the importance of supporting gender identity exploration.

Custody disputes are especially unpredictable and subject to judges’ biases for several reasons. First, custody determinations are often not a decision as to whether a parent will have custody or not, but rather an inquiry into how much time parents will get to spend with their children and how much decision-making authority each will have.326 Judges consequently are making a large array of decisions in each case.327 Second, judges’ custody decisions are subject to change. Unlike other types of civil litigation, in which a resolution terminates the case and closes the docket, family law judges retain continuing and exclusive jurisdiction over custody and may modify their orders upon one party’s petition.328 For that reason, judges are able to introduce their viewpoints more frequently. Third, these family law decisions are subject to an abuse of discretion standard on appeal, such that reviewing courts are unlikely to overturn the rulings.329 As a result, the likelihood of meaningful judicial review is much lower in this context. Finally, critics have charged that the “best interests of the child” standard is often so vague that it allows judges to “impose idiosyncratic value judgments on families.”330

Family law courts are not the only ones with this potential for bias, which is a vexing problem that plagues many judicial decisions. Scholars of law and politics, law and economics, and social and cognitive psychology have repeatedly demonstrated that, in the absence of controlling precedent, “when facts give rise to legal indeterminacy” such that the applicable law does not dictate a clear outcome, a judge’s preexisting attitudes will influence their analysis of applicable law.331 As cognitive psychologists explain, it is not that judges are overtly imposing their ideological preferences, but rather that, when deciding a difficult case, a judge’s “cognitive system imposes coherence on the arguments so that the subset of arguments that support[] one outcome becomes more appealing to the judge [than] the opposite subset . . . .”332 Judges thus “may sincerely believe that their rulings are grounded in an analysis of law alone, even though their predilections influence their rulings.”333

There are several mechanisms within the law to reduce judicial partiality, but these options have limited salience in the family law context because they aim at reducing overt bias, rather than the intuition-laden decisions that psychologists have identified.334 Under the Due Process Clause, litigants are entitled to fair trials, which prevents decision-makers with a conflict of interest from presiding over a proceeding. The constitutional provision is an outer boundary, or a backstop, for disqualification procedures, which Congress and state legislators have enacted.335 The codes of judicial conduct provide for disqualification where a judge has a conflict of interest, as well as where a judge has “a personal bias or prejudice concerning a party or a party’s lawyer . . . .”336 Judges enjoy a presumption of impartiality, which constrains disqualification’s utility as a means of policing bias beyond overt prejudice.337 That disqualification rarely happens does not mean that it never transpires; in the Younger case, the Dallas County court ultimately removed the judge for posting about the dispute on her official Facebook page, despite having entered a gag order on the parents because of the publicity’s negative affect on Luna and her brother.338

Given the potential for judicial bias, elected officials should step in and enact statutory reform. There is good reason to think that elected officials may act—namely, they have promoted gender expansive children’s right to exploration in another context. Almost half of states have enacted bans on conversion therapy, a practice intended to reduce or eliminate a person’s same-sex sexual attraction or transgender identity.339 In 2012, California became the first state to prohibit licensed mental health professionals from providing conversion therapy to minors.340 States and cities around the country quickly did the same.341 In 2020, the Eleventh Circuit struck down a municipal law prohibiting conversion therapy, ruling that the ordinance violated therapists’ freedom of speech, but other circuit courts reviewing these types of laws have upheld them.342

The gap between conversion therapy bans, which prohibit licensed mental health professionals from providing reparative therapy, and a statute that defines reparative behavior as being against a child’s best interests, is minimal. The former regulates the behavior of medical professionals, while the latter is a factor in custody decisions, yet both limit what treatments the state identifies as appropriate for children. It is not unusual for states to impose restrictions on children’s medical care. Although parents have broad authority over their children’s welfare, the state can and does intervene when parental decisions are harmful to their children.343 As a result, moving elected officials from conversion therapy bans to codifying children’s right to exploration may be easier than other types of custody reforms. Given that many states have already codified that courts must weigh children’s emotional needs and well-being, adding gender identity development would only expand the existing framework slightly.344

Moreover, a statutory change would send a message to nonlitigants who have gender expansive children in their lives. Legal scholars have repeatedly noted that laws have both material and expressive effects, such that “law matters for what it says in addition to what it does.”345 In other words, laws shape behavior by changing the social meaning of an action.346 The law contains a normative commitment that affects how citizens behave, which is why laws that do not directly target a specific individual’s conduct may still have an impact on that person’s actions.347 Precisely because laws shape social norms, interest groups have pursued symbolic laws on a range of topics, such as repealing unenforceable prohibitions on miscegenation and flag-burning.348

Legislators can expect statutes endorsing children’s gender identity exploration to influence the views and actions of individuals who are not litigating custody disputes. From the time that legislators first consider introducing a bill, they are signaling how they believe individuals should act, a message that becomes more powerful when elected officials ratify the law.349 The media attention around legislative actions helps disseminate that message, thereby informing the general public of the norms the law promotes.350 Statutes that emphasize gender expansive children’s need to explore their identities will thus help all constituents better understand that children may be gender expansive and therefore require an opportunity to explore their gender identity. Moreover, such a law communicates that adults should support children in that exploratory process. These laws would thus help parents who have gender expansive children, but who are not embroiled in custody disputes, better meet their children’s needs.

Statutory reform would promote gender expansive children’s needs in many ways, but legislators seeking to enact these laws face an uphill battle. Advocates have repeatedly sought legislative reform to remedy family law’s open-ended standards and judges’ largely unreviewable discretion in custody disputes. These efforts have ranged from introducing bright-line rules on visitation, to presumptions on custody, to guidance that offers advice when a totality of the circumstances test is inconclusive.351 In 2002, the American Law Institute promulgated principles that included presumptions for custody and property division to limit judicial discretion in these areas.352 Statutory reform projects have thus addressed all parts of the divorce process, not just custody.

However, elected officials have resoundingly rejected these calls for reform.353 In part, their objections have been substantive. Because presumptions for joint custody favor men, while the presumption for primary caretakers favors women, legislators understandably resisted the changes.354 Even where elected officials agreed with the reform proposal, however, some failed to act because they were reluctant to wade into controversial waters.355 Yet others’ inaction was the product of political stalemate. As Professor Sean Williams has noted, “because it is generally easier to defeat legislation than pass it, the groups opposing reforms have the upper hand” in these legal battles.356 Reformers have thus faced significant political impediments in their efforts to introduce clear rules into family courts.

Moreover, despite success with conversion therapy bans, the halls of the legislature are often quite perilous for LGBTQ families.357 When elected officials have attempted to address gender expansive child custody issues, they have typically sought to circumscribe LGBTQ rights, rather than expand them. In response to the Luna Younger case, lawmakers from six states proposed bills to regulate gender transition.358 In Texas, five representatives sponsored a law that would prevent judges from considering a parent’s stance on gender transition in its best interests of the child analysis,359 while another promised supporters that the first bill he would introduce in 2021 would make “transitioning of a minor” child abuse under state law.360 In contrast, no legislature has yet attempted to codify children’s right to gender identity exploration. Yet the difficulty of enacting this change does not lessen its importance.

A statutory change that codifies the right to gender identity exploration would help judges adjudicate the cases in ways that promote the children’s best interests, as well as send an important expressive message. In the meantime, judges do not need a new law to accomplish this same goal because the best interests standard that they already apply should be sufficient. Family law judges, who are often called upon to exercise discretion in determining what is in the best interests of the child, have both the expertise and institutional capacity to promote children’s gender identity exploration. They consequently may be in a position to instantiate a more accurate understanding of what it means to be gender expansive, a change that legislators should also promote.


Custody cases involving gender expansive children are in many ways less complicated than they initially appear. Courts should award custody to the parent who supports the child’s gender identity exploration regardless of whether the child’s gender variance persists past adolescence. Such an approach is also consistent with judicial precedent on other types of identity exploration, namely sexual orientation, race, and religion. Although all are considered fundamental, if not immutable, characteristics, they are identities that children can and do explore.

The assertion that courts should reframe the best interests analysis to focus on a child’s need to explore their gender identity is not meant to suggest that these types of cases are easy. Far from it. The uncertainty around what gender expansive identification means for any given child renders every decision complex and extremely fact dependent. Revising the framework for these cases, however, does make the cases easier in important ways, as it provides courts and litigants with principles from which to reason and structure their actions.

Educating judges case-by-case, by repeatedly presenting studies, expert reports, and briefs with related precedent, may seem like fighting a battle with a chisel rather than a sword. Yet presenting judges with scientific evidence is a small-scale solution that has worked in the past. Equally important is statutory reform, which would produce more widespread change and would have the additional benefit of sending an expressive message to all. By promoting support for gender identity exploration, both solutions would make a significant difference in the lives of gender expansive children.


Citations for Cases in Tables

The citations below are for court decisions appearing in case reporters or electronic databases. The cases are organized by chart, and listed in reverse chronological order.

Table 1: Gender Expansive Child Custody Cases

  • In re JA.D.Y. and JU.D.Y., DF-15-09887-S (Tex. Dist. Ct. Jan. 29, 2020).
  • Paul E. v. Courtney F., 439 P.3d 1169 (Ariz. 2019).
  • In re Churchill/Belinski, No. 337790, 2018 WL 1343986 (Mich. Ct. App. Mar. 15, 2018), vacated by joint motion of the parties, 923 N.W.2d 885 (Mich. 2019).
  • In re R.A., No. CO74489, 2014 WL 989190 (Cal. Ct. App. Mar. 14, 2014).
  • Kristen L. v. Benjamin W., No. S-15302, 2014 WL 2716842 (Alaska June 11, 2014).
  • Williams v. Frymire, 377 S.W.3d 579 (Ky. Ct. App. 2012).
  • Morrison v. Cochrane, No. FBTFA040410982, 2009 WL 2357621 (Conn. Super. Ct. June 2, 2009).
  • Johnson v. Johnson, No. RF09-463371 (Cal. Super. Ct. Alameda Cnty. Oct. 10. 2009).
  • Smith v. Smith, No. 05 JE 42, 2007 WL 901599 (Ohio Ct. App. Mar. 23, 2007).
  • Shrader v. Spain, No. 05-95-01649-CV, 1998 WL 40632 (Tex. App. Feb. 4, 1998).

Table 2: Sexual Orientation Family Court Cases

  • Buxton v. Storm, 238 P.3d 30 (Or. Ct. App. 2010).
  • In re C.O., No. B206425, 2008 WL 4670513 (Cal. Ct. App. Oct. 23, 2008).
  • Acevedo v. Williams, 985 So. 2d 669 (Fla. Dist. Ct. App. 2008).
  • In re Zion J., No. A111895, 2006 WL 2709831 (Cal. Ct. App. Sept. 22, 2006).
  • Dzierson v. Dzierson, 661 N.Y.S.2d 779 (Fam. Ct. 1997).
  • In re Lori M., 496 N.Y.S.2d 940 (Fam. Ct. 1985).
  • In re Shane T., 453 N.Y.S.2d 590 (Fam. Ct. 1982).
  • Catherine W. v. Robert F., 455 N.Y.S.2d 519 (Fam. Ct. 1982).

Table 3: Bi/Multiracial Child Custody Cases

  • In re Marriage of Gambla, 853 N.E.2d 847 (Ill. App. Ct. 2006).
  • Olivier A. v. Christina A., No. 30779/2002, 2005 WL 2171176 (N.Y. Sup. Ct. 2005).
  • Hamilton v. Hamilton, 42 P.3d 1107 (Alaska 2002).
  • Ebirim v. Ebirim, 620 N.W.2d 117 (Neb. Ct. App. 2000).
  • Davis v. Davis, 658 N.Y.S.2d 548 (App. Div. 1997).
  • Jones v. Jones, 542 N.W.2d 119 (S.D. 1996).
  • Rooney v. Rooney, 914 P.2d 212 (Alaska 1996).
  • In re Marriage of Kleist, 538 N.W.2d 273 (Iowa 1995).
  • Schultz v. Elremmash, 615 So. 2d 396 (La. Ct. App. 1993).
  • Lee v. Halayko, 590 N.Y.S.2d 647 (App. Div. 1992).
  • In re Marriage of Brown, 480 N.E.2d 246 (Ind. Ct. App. 1985).
  • Farmer v. Farmer, 439 N.Y.S.2d 584 (Sup. Ct. 1981).
  • Raysor v. Gabbey, 395 N.Y.S.2d 290 (App. Div. 1997).
  • Beazley v. Davis, 545 P.2d 206 (Nev. 1976).
  • Tucker v. Tucker, 542 P.2d 789 (Wash. Ct. App. 1975).
  • Fountaine v. Fountaine, 133 N.E.2d 532 (Ill. App. Ct. 1956).
  • Ward v. Ward, 216 P.2d 755 (Wash. 1950).

Table 4: Religious Conversion Child Custody Cases

  • Baalla v. Baalla, 71 N.Y.S.3d 138 (App. Div. 2018).
  • Nunez v. Gray, No. 2210, 2017 WL 3381132 (Md. Ct. Spec. App. Aug. 7, 2017).
  • Weisberger v. Weisberger, 60 N.Y.S.3d 265 (App. Div. 2017).
  • Pierson v. Pierson, 143 So. 3d 1201 (Fla. Dist. Ct. App. 2014).
  • Harrison v. Tauheed, 256 P.3d 851 (Kan. 2011).
  • In re Marriage of Reyes, No. 1-10-1289, 2011 WL 10068626 (Ill. Ct. App. Feb. 16, 2011).
  • Pace v. Farr, No. 1 CA-CV 09-0575, 2010 WL 5030870 (Ariz. Ct. App. Oct. 26, 2010).
  • CLB v. DLO, 61 So. 3d 325 (Ala. Civ. App. 2010).
  • In re Marriage of Boldt, 176 P.3d 388 (Or. 2008).
  • Hicks v. Hicks, 868 A.2d 1245 (Pa. Super. Ct. 2005).
  • Homonnay v. Homonnay, No. FA030401627S, 2004 WL 1245636 (Conn. Super. Ct. May 19, 2004).
  • Frank v. Frank, 833 A.2d 194 (Pa. Super. Ct. 2003).
  • Hamilton v. Hamilton, 42 P.3d 1107 (Alaska 2002).
  • Meyer v. Meyer, 789 A.2d 921 (Vt. 2001).
  • Wood v. DeHahn, 571 N.W.2d 186 (Wis. Ct. App. 1997).
  • Baker v. Baker, No. 03A01-9704-GS-00115, 1997 WL 731939 (Tenn. Ct. App. Nov. 25, 1997).
  • Hoedebeck v. Hoedebeck, 948 P.2d 1240 (Okla. Civ. App. 1997).
  • In re Marriage of Whalen, 569 N.W.2d 626 (Iowa Ct. App. 1997).
  • In re Marriage of Jensen-Branch, 899 P.2d 803 (Wash. Ct. App. 1995).
  • Abbo v. Briskin, 660 So. 2d 1157 (Fla. Dist. Ct. App. 1995).
  • McCown v. McCown, 649 A.2d 418 (N.J. Super. Ct. App. Div. 1994).
  • Lange v. Lange, 502 N.W.2d 143 (Wis. Ct. App. 1993).
  • Greene v. Greene, No. FA 91-0399669, 1992 WL 329286 (Conn. Super. Ct. Oct. 28, 1992).
  • Pater v. Pater, 588 N.E.2d 794 (Ohio 1992).
  • Kirchner v. Caughey, 606 A.2d 257 (Md. 1992).
  • Zummo v. Zummo, 574 A.2d 1130 (Pa. Super. Ct. 1990).
  • LeDoux v. LeDoux, 452 N.W.2d 1 (Neb. 1990).
  • Von Tersch v. Von Tersch, 455 N.W.2d 130 (Neb. 1990).
  • Khalsa v. Khalsa, 751 P.2d 715 (N.M. Ct. App. 1988).
  • Hanson v. Hanson, 404 N.W.2d 460 (N.D. 1987).
  • Funk v. Ossman, 724 P.2d 1247 (Ariz. Ct. App. 1986).
  • Gould v. Gould, 342 N.W.2d 426 (Wis. 1984).
  • Siegel v. Siegel, 472 N.Y.S.2d 272 (Sup. Ct. 1984).
  • In re Marriage of Mentry, 190 Cal. Rptr. 843 (Ct. App. 1983).
  • Burnham v. Burnham, 304 N.W.2d 58 (Neb. 1981).
  • Felton v. Felton, 418 N.E.2d 606 (Mass. 1981).
  • Morris v. Morris, 412 A.2d 139 (Pa. Super. Ct. 1979).
  • Stevens v. Stevens, No. CA-1678, 1978 WL 217380 (Ohio Ct. App. Feb. 3, 1978).
  • Robertson v. Robertson, 575 P.2d 1092 (Wash. Ct. App. 1978).
  • Asch v. Asch, 397 A.2d 352 (N.J. Super. Ct. App. Div. 1978).
  • Jacobs v. Jacobs, 323 N.E.2d 21 (Ill. App. Ct. 1974).
  • Munoz v. Munoz, 489 P.2d 1133 (Wash. 1971).
  • Goodman v. Goodman, 141 N.W.2d 445 (Neb. 1966).
  • Gluckstern v. Gluckstern, 220 N.Y.S.2d 623 (Sup. Ct. 1961).
  • Gottlieb v. Gottlieb, 175 N.E.2d 619 (Ill. App. Ct. 1961).
  • Jackson v. Jackson, 309 P.2d 705 (Kan. 1957).
  • Boerger v. Boerger, 97 A.2d 419 (N.J. Super. Ct. Ch. Div. 1953).


1. Greg Abbott (@GregAbbott_TX), Twitter (Oct. 23, 2019, 7:58 PM),

2. Katelyn Burns, What the Battle Over a 7-Year-Old Trans Girl Could Mean for Families Nationwide, Vox (Nov. 11, 2019),

3. Although Luna’s legal name is a matter of public record and became widely known through trending tweets, this Article intentionally refrains from using it because of the damage that the use of birth names does to trans and nonbinary people. See, e.g., Lauren Freeman, Micro Interactions, Macro Harms: Some Thoughts on Improving Health Care for Transgender and Gender Nonbinary Folks, 11 Int’l J. Feminist Approaches Bioethics 157, 160 (2018).

4. This Article uses the gender pronouns with which a child identifies. When a child’s preferred pronouns are unknown, the Article uses the gender-neutral pronouns “they,” “them,” and “theirs.”

5. Burns, supra note 2.

6. Id.; see also Am. Psychiatric Ass’n, Diagnostic and Statistical Manual of Mental Disorders: DSM-5 452 (2013).

7. Burns, supra note 2.

8. Id.

9. First Amended Petition to Modify the Parent-Child Relationship at 3, In re JA.D.Y. and JU.D.Y., DF-15-09887-S (Tex. Dist. Ct. July 2, 2018) (on file with author).

10. Burns, supra note 2.

11. Ted Cruz (@SenTedCruz), Twitter (Oct. 23, 2019, 8:01 PM),; Donald Trump Jr. (@DonaldJTrumpJr), Twitter (Oct. 24, 2019, 7:44 AM),

12. H.B. 303, 2020 Leg., Reg. Sess. (Ala. 2020); H.B. 1365, 2020 Leg., Reg. Sess. (Fla. 2020); GA H.B. 1060, 2019–2020 Leg., Reg. Sess. (Ga. 2020); H.B. 321, 2020 Leg., Reg. Sess. (Ky. 2020); H.B. 1-57, 2020 Leg., Reg. Sess. (S.D. 2020); TX H.B. 1910, 86th Leg., Reg. Sess. (Tex. 2019). Ohio had introduced a similar bill the previous year. H.B. 658, 132d Leg., Reg. Sess. (Ohio 2018).

13. Burns, supra note 2.

14. For a discussion of the types of medical treatments for gender expansive children, see infra Part I.B.

15. Such an umbrella term is both useful and necessary as judges and parents in these cases often assume a child is transgender, when they may be nonbinary or have some other non-normative gender identity. This Article takes the term from a medical article on gender expansive youth. Diane Ehrensaft et al., Prepubertal Social Gender Transitions: What We Know; What We Can Learn—A View from a Gender Affirmative Lens, Int’l J. Transgenderism 1 (2018). For a discussion of nonbinary and other types of non-normative gender identities in law, see Jessica Clarke, They, Them, and Theirs, 132 Harv. L. Rev. 894, 906–08 (2019) (describing nonbinary gender identities); Marie-Amélie George, The LGBT Disconnect, 2018 Wis. L. Rev. 503, 571–72 (detailing the variation within transgender identities).

16. Florence Ashley, Gender (De)Transitioning Before Puberty?: A Response to Steensma and Cohen-Kettenis (2011), 48 Arch. Sexual Behavior 679, 679–80 (2018); Jack L. Turban, Transgender Youth: The Building Evidence Base for Early Social Transition, 56 J. Am. Acad. Child & Adolescent Psychiatry 101 (2017); Jack Drescher & Jack Pula, Ethical Issues Raised by the Treatment of Gender-Variant Prepubescent Children, in LGBT Bioethics: Visibility, Disparities, and Dialogue S18, S18–20 (2014); World Pro. Ass’n for Transgender Health (WPATH), Standards of Care for the Health of Transsexual, Transgender, and Gender-Nonconforming People, Version 7, 11 (2012); Jesse Singal, How the Fight Over Transgender Kids Got a Leading Sex Researcher Fired, N.Y. Mag. (Feb. 7, 2016),; Thomas D. Steensma & Peggy T. Cohen-Kettenis, Gender Transitioning Before Puberty?, 40 Arch. Sexual Behavior 649, 649 (2011).

17. Several works have analyzed a limited subset of transgender child custody cases or have focused on transgender minors’ right to transition. Angie Martell, Legal Issues Facing Transgender and Gender-Expansive Youth, 96 Mich. B.J. 30, 31 (2017); Kasia Szczerbinski, I Am Whoever You Say I Am: How the Custodial Decisions of Parents Can Affect and Limit a Transgender Child’s Freedom and State of Mind, 36 Child. Legal Rts. J. 177, 178 (2016); Jaime B. Margolis, Two Divorced Parents, One Transgender Child, Many Voices, 15 Whittier J. Child. & Fam. Advoc. 125, 125, 126, 128 (2016); Erika Skougard, Note, The Best Interests of Transgender Children, 2011 Utah L. Rev. 1161. Scholars have also recently turned their attention to custody disputes involving a transgender parent. Sonia K. Katyal & Ilona M. Turner, Transparenthood, 117 Mich. L. Rev. 1593, 1600, 1622 (2019).

18. See Selin Gülgöz et al., Similarity in Transgender and Cisgender Children’s Gender Development, 116 Procs. Nat’l Acad. Scis. 24480 (2019).

19. Susan M. Sawyer et al., The Age of Adolescence, 2 Lancet Child & Adolescent Health 223 (2018); see infra Part I.B.

20. See infra Part I.B.

21. See id.

22. For examples of state statutes requiring courts to consider the children’s emotional, developmental, or psychological needs and well-being, see Alaska Stat. § 25.24.150(c)(1); Ariz. Rev. Stat. Ann. § 25-403(A); Colo. Rev. Stat. Ann. § 14-10-124(1.5) (West); Haw. Rev. Stat. § 571-46(b)(7); 750 Ill. Comp. Stat. Ann. 5/602.7(b) (West); Iowa Code Ann. § 598.41 (West); Kan. Stat. Ann. § 23-3203(a)(5); Me. Stat. tit. 19-a, § 1653(3); Minn. Stat. § 518.17(1); Nev. Rev. Stat. Ann. § 125C.0035(4)(g) (West); N.D. Cent. Code § 14-09-06.2(1); 23 Pa. Cons. Stat. § 5328(a); Tenn. Code Ann. § 36-6-106(a); Tex. Fam. Code Ann. § 153.131(a) (West); Utah Code Ann. § 30-3-10(2) (West); Va. Code Ann. § 20-124.3; Vt. Stat. Ann. tit. 15, § 665(b); Wash. Rev. Code § 26.09.002.

23. Disparate literatures on LGBTQ children, biracial and multiracial children, and religious conversion all point to this common thread, though scholars have yet to connect them. For articles on biracial and multiracial children, see Solangel Maldonado, Bias in the Family: Race, Ethnicity, and Culture in Custody Disputes, 55 Fam. Ct. Rev. 213 (2017); Katie Eyer, Constitutional Colorblindness and the Family, 162 U. Pa. L. Rev. 537 (2014). For articles on religious conversion, see Donald L. Beschle, God Bless the Child?: The Use of Religion as a Factor in Child Custody and Adoption Proceedings, 58 Fordham L. Rev. 383 (1989); James G. Dwyer, Parents’ Religion and Children’s Welfare: Debunking the Doctrine of Parents’ Rights, 82 Cal. L. Rev. 1371 (1994). For scholarship on sexual minority youth, see Jordan Blair Woods, Religious Exemptions and LGBTQ Child Welfare, 103 Minn. L. Rev. 2343 (2019); Orly Rachmilovitz, Family Assimilation Demands and Sexual Minority Youth, 98 Minn. L. Rev. 1374 (2014); Clifford J. Rosky, Fear of the Queer Child, 61 Buff. L. Rev. 607 (2013).

24. Some courts have questioned children’s right to self-expression, ruling that schools may censor children’s political expression and also upholding grooming codes. However, others have recognized that children have an interest in expressing their political views and identities. Also relevant to this Article is that the Supreme Court has recognized that children’s identities change rapidly and are more fluid than those of adults. Anne C. Dailey & Laura A. Rosenbury, The New Law of the Child, 127 Yale L.J. 1448, 1498–1500 (2018).

25. In presenting this argument, this Article demonstrates that courts have partially endorsed Anne Dailey and Laura Rosenbury’s argument that children should have the right to explore their identities. Id. at 1497–98. Whether courts’ recognition of children’s need to explore their identities is the same as having a right to do so is a different question, and one that is beyond the scope of this Article.

26. Exploration may mean different things depending on the identity category, as children may be asking who they are, who they want to be, or both. Alan S. Waterman, What Does It Mean to Engage in Identity Exploration and to Hold Identity Commitments?: A Methodological Critique of Multidimensional Measures for the Study of Identity Processes, 15 Identity 309, 312 (2015).

27. Cases in which children change, explore, or inhabit the liminal space between these fundamental characteristics render parental conflict over them different from disputes over other important issues. Disputes over such topics as which schools a child should attend or how much visitation to allow with extended family do not implicate constitutionally protected identity traits. See Dailey & Rosenbury, supra note 24, at 1496–98 (discussing children’s general right to self-expression).

28. The transition from childhood to adulthood is in large part a psychosocial process of developing one’s identity. Social scientists, including psychologists and sociologists, have described the process of identity formation, which includes subjective, behavioral, and social dimensions. For a discussion of this literature, see James E. Côté, Identity Formation, Agency, and Culture: A Social Psychological Synthesis 6–9 (2002).

29. Steven E. Mock & Richard P. Eibach, Stability and Change in Sexual Orientation Identity Over a 10-Year Period in Adulthood, 41 Archives Sexual Behav. 641 (2012); Miles Q. Ott et al., Stability and Change in Self-Reported Sexual Orientation Identity in Young People: Application of Mobility Metrics, 40 Archives Sexual Behav. 519 (2011).

30. There are also cases involving parents of different faiths who marry, then disagree as to how they should raise their child upon divorce. This Article limits its discussion to parents who convert during or after the marriage because conversion introduces a new, unexpected element into the parents’ relationship with one another—and their children. As a result, the latter subset of religion cases is more analogous to the gender expansive child custody disputes than the former.

31. Religious conversion is not uncommon, although it can be extremely controversial. Massimo Leone, Conversion and Controversy, in Controversies and Subjectivity 91 (Pierluigi Barrotta & Marcelo Dascal eds., 2005).

32. Courts—and society more generally—once identified children’s religion as an immutable characteristic. See, e.g., In re Vardinakis, 289 N.Y.S. 355, 359 (Dom. Rel. Ct. 1936) (“A custom has grown up that where a child is once baptized or entered in any prescribed manner into a church, that the child is to be treated as belonging to that church so long as he is a minor.”). For a discussion of how this perspective resulted in religious “matching” in adoptions, see Elizabeth D. Katz, “Racial and Religious Democracy”: Identity and Equality in Midcentury Courts, 72 Stan. L. Rev. 1467, 1509–14 (2020).

33. Jessica A. Clarke, Against Immutability, 125 Yale L.J. 2, 27 (2015); Kenji Yoshino, The New Equal Protection, 124 Harv. L. Rev. 747 (2011). The Supreme Court has never declared sexual orientation to be a suspect class but has described sexual orientation as an immutable characteristic. Obergefell v. Hodges, 576 U.S. 644, 661 (2015). In overturning same-sex marriage bans, several jurisdictions applied heightened scrutiny after determining that sexual orientation was an immutable characteristic. Baskin v. Bogan, 766 F.3d 648, 657 (7th Cir. 2014); Kerrigan v. Comm’r of Pub. Health, 957 A.2d 407, 438–39 (Conn. 2008); Varnum v. Brien, 763 N.W.2d 862, 893–94 (Iowa 2009); Griego v. Oliver, 316 P.3d 865, 884 (N.M. 2013).

34. Parents do not currently assume that the children will identify in a particular way, though decisions from decades past indicate that litigants expected children to identify with the race of the racial minority parent. See discussion infra Part II.B.

35. See Joel Feinberg, The Child’s Right to an Open Future, in Whose Child?: Children’s Rights, Parental Authority, and State Power 124 (William Aiken & Hugh LaFollette eds. 1980) (discussing parents’ obligation to keep children’s future options open, rather than direct children toward a specific goal).

36. One exception to this is the shift from fault-based divorce to no-fault divorce, which moved divorce from a fact-intensive analysis to a bright-line rule of irretrievable breakdown and living separate and apart for a statutory period. Barbara B. Woodhouse, Sex, Lies, and Dissipation: The Discourse of Fault in a No-Fault Era, 82 Geo. L.J. 2525 (1994).

37. For a discussion of the expressive function of law, see Richard H. McAdams, The Expressive Powers of Law: Theories and Limits 165–67 (2015).

38. See infra Part I.B.

39. See id.

40. Legislators responding to these cases have made explicit antitransgender arguments. For example, when Georgia State Rep. Ginny Ehrhart proposed the Vulnerable Child Protection Act, S.B. 1864, 2020 Leg. Reg. Sess. (Ga. 2020), which would have banned medical treatment for gender expressive minors, she described the increase of gender expansive children as an “epidemic.” Juwan J. Holms, Another Republican Wants to Ban Gender-Affirming Medical Treatment for Youth, LGBTQ Nation (Nov. 3, 2019), A year earlier, she had derided gender identity antidiscrimination protections, explaining that “[w]hether a person identifies as a man or a moose, he does not have the right to demand special treatment.” Id. For a discussion of other legislative attempts to restrict the rights of gender expansive youth, see infra notes 358–60 and accompanying text.

41. Jason Rafferty, Gender Identity Development in Children, Am. Acad. of Pediatrics (Sept. 18, 2018),

42. See discussion infra Part I.B.

43. This is not to say that gender is irrelevant for adults, or that sex segregation is simply a social phenomenon. There are spaces that remain gender-segregated as a matter of law, including prisons, detention centers, and homeless shelters. Marie-Amélie George, Framing Trans Rights, 114 Nw. U. L. Rev. 555, 613–14 (2019). The flashpoints for controversies over transgender rights have been restrooms, locker rooms, and changing facilities, all of which may be sex-segregated by law. Id. at 614. Additionally, when patrons call upon law enforcement to regulate access to the spaces, the state becomes involved in policing gender. Id.

44. Hiroko Tabuchi, Sweeping Away Gender-Specific Toys and Labels, N.Y. Times (Oct. 27, 2015),

45. Id.

46. Jennifer Finney Boylan, Why Are Gender Reveal Parties a Thing?, N.Y. Times (Dec. 3, 2018),

47. The retailer Party City has sold a million items a year for gender reveal parties. Id.

48. See Singal, supra note 16 (discussing how clinicians diagnose children as gender dysphoric based on how closely the children adhere to gender roles). Notably, these same parents likely would not expect a gender conforming child to strictly adhere to gender norms. See id.

49. 377 S.W.3d 579, 580 (Ky. Ct. App. 2012).

50. Id. at 582.

51. Id. at 582, 591.

52. Burns, supra note 2.

53. Id.

54. Talia Mae Bettcher, Full-Frontal Morality: The Naked Truth About Gender, 27 Hypatia 319, 320 (2011); see also Robin Dembroff, Beyond Binary: Genderqueer as Critical Gender Kind, 20 Philosophers’ Imprint 1, 15 (2020) (identifying the four axes of dominant Western gender ideology: binary, biology, teleology, and hierarchy).

55. Sally Haslanger, Resisting Reality: Social Construction and Social Critique 39–42 (2012); Simone de Beauvoir, The Second Sex 39–42 (Constance Borde & Sheila Malovany-Chevalier trans., Vintage Books 2011) (1949).

56. See Haslanger, supra note 55, at 42–47.

57. Lisa Selin Davis, My Daughter Is Not Transgender. She’s a Tomboy, N.Y. Times (Apr. 18, 2017),

58. See, e.g., Singal, supra note 16.

59. Id.

60. Alix Spiegel, Two Families Grapple with Sons’ Gender Identity, NPR (May 7, 2008),

61. Singal, supra note 16.

62. Clarke, supra note 15, at 90608.

63. Nat’l Ctr. for Transgender Equal., The Report of the 2015 U.S. Transgender Survey 44–45 (2016) [hereinafter U.S. Transgender Survey].

64. Bianca D.M. Wilson & Ilan H. Meyer,The Williams Inst., Nonbinary LGBTQ Adults in the United States 2 (2021),; Dallas, Texas Population 2021, World Population Rev.,

65. California, Colorado, Connecticut, Illinois, Maine, Nevada, New Mexico, New York, Oregon, Rhode Island, Utah, and Washington offer nonbinary options on both birth certificates and driver’s licenses. Arkansas, Hawaii, Maryland, Massachusetts, Minnesota, New Hampshire, Pennsylvania, Vermont, Virginia, and Washington, D.C., offer nonbinary designations on driver’s licenses, while New Jersey offers nonbinary options on birth certificates. Indiana’s Bureau of Motor Vehicles (BMV) attempted to introduce a nonbinary gender designation option, but the state’s attorney general determined that the policy exceeded the BMV’s authority. See S.B. 179, 2017–2018 Leg., Reg. Sess. (Cal. 2017) (enacted); Colo. Rev. Stat. Ann. § 25-2-113.8 (West); Colo. Code Regs. § 204-32:1 (2018); D.C. Code § 50-1401.1; H.B. 1165, 30th Leg., Reg. Sess. (Haw. 2019) (enacted); 2018 H.B. 3534, 101st Gen. Assemb., Reg. Sess. (Ill. 2019) (enacted); S.B. 196, 2019 Leg., Reg. Sess. (Md. 2019) (enacted); H.B. 669, 2019 Leg., Reg. Sess. (N.H. 2019) (enacted); N.J. Stat. Ann. § 26:8-40.12 (West 2019); 2021 S.B. 4402B, 244th Leg., 2021 Reg. Sess. (N.Y. 2021) (enacted); S.B. 247, Va. Code Ann. § 46.2-323 (2020); Wash. Admin. Code § 246-490-075 (2018); Teresa Boeckel, “X” Will Become an Option for Gender on Pa. Driver’s Licenses, USA Today (July 31, 2019, 2:42 PM),; Jordan Fenster, DMV to Allow Non-binary Gender Option Starting Monday, Stamford Advoc. (Jan. 25, 2020, 6:58 PM),; Nico Lang & Kate Sosin, Utah Among Growing Number of States Issuing Gender-Neutral IDs, NBC News (Mar. 18, 2019, 12:37 PM),; Mary Emily O’Hara, Oregon Becomes First State to Add Third Gender to Driver’s Licenses, NBC News (June 15, 2017, 5:57 PM),; Steve LeBlanc, Massachusetts Senate OKs Bill to Allow Gender “X” Option, Associated Press (June 28, 2018),; Cynthia Silva, Following Trans Teen’s Suit, Minors Can Now Change New York Birth Certificates, NBC News (Mar. 12, 2020, 3:22 PM),; Paul Walsh, Minnesota Now Offers “X” for Gender Option on Driver’s Licenses, Star Trib. (Oct. 3, 2018, 8:34 AM),; WJAR Staff, RI Will Offer Third Gender Option for Licenses, Birth Certificates, News Channel 9 (Aug. 9, 2019),; Curtis M. Wong, Arkansas Has Been Offering a Nonbinary Gender Option on State IDs for Years, Huffington Post (Oct. 17, 2018, 6:50 PM),; Staff, Curtis Hill: BMV Doesn’t Have Authority to Allow Nonbinary Gender on Licenses, WTHR (Mar. 11, 2020),,female%20on%20their%20driver’s%20licenses.&text=The%20BMV%20announced%20it%20was,issued%20IDs%20in%20March%202019; Press Release, Dep’t of Sec’y of State, State of Maine, Maine BMV to Offer Non-binary Gender Designation on Driver’s Licenses, ID Cards (June 11, 2018),; Press Release, New Mexico Dep’t of Health, New Mexico Becomes Fourth State to Allow Gender-Neutral Sex Designation on Birth Certificates (Oct. 29, 2019),; Press Release, Nev. DMV, Nevada Implements Gender-Neutral IDs (Apr. 22, 2019),; Self-Designated Descriptors, Vt. Dep’t Motor Vehicles,

66. George, supra note 15, at 53640.

67. Robin Dembroff, Moving Beyond Mismatch, 19 Am. J. of Bioethics 60 (2019); Bettcher, supra note 54, at 330.

68. Joanne Meyerowitz, How Sex Changed: A History of Transsexuality in the United States 161 (2002).

69. Id.

70. See Sonia K. Katyal, The Numerus Clausus of Sex, 84 U. Chi. L. Rev. 389, 410 (2017).

71. See Elizabeth M. Glazer & Zachary A. Kramer, Transitional Discrimination, 18 Temp. Pol. & C.R. L. Rev. 651, 664–65 (2009).

72. See Meyerowitz, supra note 68, at 242–53; Paisley Currah & Lisa Jean Moore, “We Won’t Know Who You Are”: Contesting Sex Designations in New York City Birth Certificates, 24 Hypatia 113 (2009); Dean Spade, Documenting Gender, 59 Hastings L.J. 731, 767–69 (2008); Press Release, Transgender Equality, Breaking News: New York State Modernizes Requirements for Birth Certificate Gender Markers (June 5, 2014),

73. See Movement Advancement Project, Identity Document Laws and Policies: Driver’s License (2021),; Movement Advancement Project, Identity Document Laws and Policies: Birth Certificates (2021),

74. See sources cited supra note 65.

75. George, supra note 43, at 607–13.

76. U.S. Transgender Survey, supra note 63, at 49, 154.

77. Nat Thorne et al., A Comparison of Mental Health Symptomatology and Levels of Social Support in Young Treatment Seeking Transgender Individuals Who Identify as Binary and Non-binary, 20 Int’l J. Transgenderism 241, 246 (2019).

78. Cf. Kenji Yoshino, The Epistemic Contract of Bisexual Erasure, 52 Stan. L. Rev. 353, 388 (2000) (discussing the forced invisibility of bisexuals).

79. Georgiann Davis, Contesting Intersex: The Dubious Diagnosis 82, 123, 125 (2015).

80. Maayan Sudai, Revisiting the Limits of Professional Autonomy: The Intersex Rights Movement’s Path to De-medicalization, 41 Harv. J. L. & Gender 1, 6 (2018).

81. Katrina Karkazis, Fixing Sex: Intersex, Medical Authority, and Lived Experience 96 (2008); see also Davis, supra note 79, at 82; Ellen K. Feder, Making Sense of Intersex: Changing Ethical Perspectives in Biomedicine 133, 141 (2014).

82. Feder, supra note 81, at 141.

83. Katherine A. Kuvalanka et al., An Exploratory Study of Custody Challenges Experienced by Affirming Mothers of Transgender and Gender Non-conforming Children, 57 Fam. Ct. Rev. 54, 61 (2019).

84. Id. at 58.

85. Smith v. Smith, No. 05 JE 42, 2007 WL 901599, at *1 (Ohio Ct. App. Mar. 23, 2007).

86. Id. at *3, *5.

87. In re Change of Name of H.C.W., 123 N.E.3d 1048, 1050 (Ohio Ct. App. 2019).

88. Côté, supra note 28, at 14.

89. Erik H. Erikson, Identity: Youth and Crisis 22–23 (1968).

90. Id.; see also Sune Qvotrup Jensen, Othering, Identity Formation and Agency, 2 Qual. Studs. 63 (2011) (discussing how “othering” influences identity formation among ethnic minorities).

91. Wim Meeus, The Study of Adolescent Identity Formation 2000–2010: A Review of Longitudinal Research, 21 J. Rsch. on Adolescence 75 (2011).

92. Diane Breshears, Coming Out with Our Children: Turning Points Facilitating Lesbian Parent Discourse with Their Children About Family Identity, 23 Communication Rpts. 78 (2010).

93. Dailey & Rosenbury, supra note 24, at 1496.

94. Sabra L. Katz-Wise et al., Transactional Pathways of Transgender Identity Development in Transgender and Gender-Nonconforming Youth and Caregiver Perspectives from the Trans Youth Family Study, 18 Int’l J. Transgenderism 243 (2017).

95. Id.

96. See, e.g., Johnson v. Johnson, No. RF09-463371, at 1 (Calif. Super. Ct. Alameda Cnty. Oct. 10, 2009), cited in Skougard, supra note 17, at 1163 n.14, 1197; see also In re Churchill/Belinski, No. 337790, 2018 WL 1343986, at *1 (Mich. Ct. App. Mar. 15, 2018), vacated by joint motion of the parties, 923 N.W.2d 885 (Mich. 2019) (directing parent to participate in psychological therapy).

97. Although many religious denominations welcome and affirm transgender individuals, some churches press their transgender members to embrace their sex assigned at birth and discourage any attempts to physically transition. Aleksandra Sandstrom, Religious Groups’ Policies on Transgender Members Vary Widely, Pew Rsch. Ctr. (Dec. 2, 2015), Churches that do not accept transgender identity include the Assemblies of God, Church of Jesus-Christ of Latter-day Saints (the Mormon Church), Lutheran Church–Missouri Synod, Southern Baptist Convention, and Roman Catholic Church. Id.

98. No. S-15302, 2014 WL 2716842, at *1 (Alaska June 11, 2014).

99. Id. at *2–3.

100. Comm. on Adolescent Health Care, Am. College of Obstetricians & Gynecologists, Committee Op. No. 685, at 2 (Jan. 2017).

101. Although parents in contemporary cases do not make this claim, they did in disputes from more than a decade ago, and courts permitted them to pursue a reparative approach. Shrader v. Spain, No. 05-95-01649-CV, 1998 WL 40632, at *2 (Tex. App. Feb. 4, 1998); Smith v. Smith, No. 05 JE 42, 2007 WL 901599, at *1 (Ohio Ct. App. Mar. 31, 2007).

102. Parents’ freedom of religion may be curtailed if their exercise of that religion harms their child. See discussion infra Part II.C.

103. That parents have not focused on this topic is perhaps because of Supreme Court precedent that undercuts courts’ ability to consider this issue. In Palmore v. Sidoti, the Supreme Court reversed the Florida court’s decision to grant custody to a White father after his White ex-wife began living with a Black man. It ruled that “[p]rivate biases may be outside the reach of the law, but the law cannot, directly or indirectly, give them effect,” thereby eliminating social stigma that attaches to parents as a permissible consideration in custody cases. 466 U.S. 429, 433 (1984).

104. Relying on science for legal change is a fraught endeavor, especially because scientific principles may reflect political and social biases. See, e.g., Wendy E. Wagner, The Science Charade in Toxic Risk Regulation, 95 Colum. L. Rev. 1613, 1640 (1995). At the same time, movement advocates often work with researchers because scientific evidence can be a crucial element to legal change. Suzanne Goldberg, Constitutional Tipping Points: Civil Rights, Social Change, and Fact-Based Adjudication, 106 Colum. L. Rev. 1955, 1991 (2006).

105. See Sawyer et al., supra note 19, at 223.

106. Am. Psych. Ass’n, Guidelines for Psychological Practice with Transgender and Gender Nonconforming People, 70 Am. Psych. 832, 842 (2015).

107. Jean Malpas, Between Pink and Blue: A Multi-dimensional Family Approach to Gender Nonconforming Children and Their Families, 50 Fam. Process 453, 458 (2011).

108. Id. at 456; Am. Psych. Ass’n, supra note 106, at 842.

109. See Malpas, supra note 107, at 456.

110. Comm. on Adolescent Health Care, supra note 100, at 3; Malpas, supra note 107, at 456.

111. Comm. on Adolescent Health Care, supra note 100, at 4.

112. Id.

113. U.S. Transgender Survey, supra note 63, at 99–101.

114. Am. Psych. Ass’n, supra note 106, at 842.

115. Ehrensaft et al., supra note 15, at 8. Reparative therapy to forestall same-sex sexual attraction has been the subject of legislative prohibitions. Marie-Amélie George, Expressive Ends: Understanding Conversion Therapy Bans, 68 Ala. L. Rev. 793, 794–95 (2017).

116. Spiegel, supra note 60.

117. WPATH, supra note 16, at 16.

118. Am. Psychiatric Ass’n, Position Statement on Issues Related to Sexual Orientation and Gender Minority Status (2020),; Am. Acad. of Pediatrics, Ensuring Comprehensive Care and Support for Transgender and Gender-Diverse Children and Adolescents, 142 Pediatrics e20173737 (2018).

119. George, supra note 115; see Colo. Rev. Stat. § 12-245-224 (2019); Conn. Gen. Stat. § 19a-907a (2017); Del. Code Ann. tit. 24, § 3510 (2018); D.C. Code § 7-1231.14a (2014); Haw. Rev. Stat. § 453J-1 (2019); 405 Ill. Comp. Stat. 48/20 (2015); Me. Rev. Stat. tit. 32, § 13800-B (2019); Md. Code. Ann., Health Occ. § 1-212.1 (2018); Mass. Gen. Laws ch. 112, § 275 (2019); Nev. Rev. Stat. § 629.600 (2017); N.H. Rev. Stat. § 332-L:2 (2018); N.J. Stat. § 45:1-54 (2013); N.M. Stat. Ann. § 61-1-3.3 (2017); N.Y. Educ. Law § 6531-a (McKinney 2019); Or. Rev. Stat. § 675.850 (2015); 23 R.I. Gen. Laws § 23-94-3 (2017); Vt. Stat. Ann. tit. 18, § 8352 (2016); Va. Code Ann. § 54.1-2409.5 (2020); Wash. Rev. Code § 18.130.020 (2018).

120. Ehrensaft et al., supra note 15, at 9. Under the Dutch Protocol, physicians may administer hormone suppressants when the child is 12 years old. Annelou L.C. de Vries & Peggy T. Cohen-Kettenis, Clinical Management of Gender Dysphoria in Children and Adolescents: The Dutch Approach, 59 J. Homosexuality 301, 311 (2012).

121. de Vries & Cohen-Kettenis, supra note 120, at 308.

122. Id. at 308–09; Singal, supra note 16. In their description of the Dutch Protocol, researchers explained that the key was for children to grow up safely, which can include imposing limits on a child’s gender expression. Thus, “if a young boy likes to wear dresses in a neighborhood in which aggression can be expected, [the family] could come to an understanding with their son that he only wears dresses at home.” The researchers emphasized that the parents should “give their child a clear explanation of why they have made their choices” and emphasize “that this does not mean that they themselves do not accept the cross-dressing.” de Vries & Cohen-Kettenis, supra note 120, at 308–09.

123. Burns, supra note 2. Some media outlets asserted that the hormone medications were linked to “thousands” of deaths but did not acknowledge that the drugs at issue have multiple uses, including as palliative care for terminally ill cancer patients. The death toll is thus unrelated to gender identity treatment. Tim Fitzsimons, A Viral Fake News Story Linked Trans Health Care to “Thousands” of Deaths, NBC News (Sept. 27, 2019, 6:16 PM),; María Méndez, Could Transgender Kids’ Care Be Next “Bathroom Bill” for Texas Republicans?, Dallas Morning News (Oct. 25, 2019, 11:57 AM),

124. Burns, supra note 2.

125. Ehrensaft et al., supra note 15, at 6.

126. Id. at 7.

127. Burns, supra note 2. Georgulas championed the affirmative approach. Id.

128. Teo Armus, A Texas Man Says His 7-Year-Old Isn’t Transgender, Wash. Post (Oct. 24, 2019, 7:13 a.m.),

129. Drescher & Pula, supra note 16, at S17; WPATH, supra note 16, at 11.

130. Drescher & Pula, supra note 16, at S17. Most of the children in these studies whose gender dysphoria desisted later identified as gay or lesbian. Id. at S18; WPATH, supra note 16, at 11.

131. Steensma & Cohen-Kettenis, supra note 16, at 649.

132. Turban, supra note 16, at 102; Drescher & Pula, supra note 16, at S18–S20; WPATH, supra note 16, at 17. Practitioners who attempt to keep the child in his or her sex assigned at birth typically want to forestall the child’s later difficulty of a second transition. Drescher & Pula, supra note 16, at S20; WPATH, supra note 16, at 17. Other proponents of this approach have identified adult transgender identity as an undesirable outcome due to the social stigma associated with transgender identity and the invasive medical procedures that transgender individuals often undertake so their physical bodies will align with their gender identities. Drescher & Pula, supra note 16, at S18; Kenneth J. Zucker et al., A Developmental, Biopsychosocial Model for the Treatment of Children with Gender Identity Disorder, 59 J. Homosexuality 369, 391 (2012).

133. Ehrensaft et al., supra note 15, at 3–4, 7 (discussing methodological limitations of desistance data and presenting research that demonstrates that “transgender children know their gender as clearly and consistently as cisgender children of the same gender”). One set of researchers that has published studies on the high rates of desistance has also noted that “a complete remittance of gender dysphoria did not occur in all desisters.” Thomas D. Steensma et al., Desisting and Persisting Gender Dysphoria After Childhood: A Qualitative Follow-Up Study, Clinical Child Psych. & Psychiatry, at 15 (2010), (also at 16 Clinical Child Psych. & Psychiatry 499 (2011)).

134. Ehrensaft et al., supra note 15, at 4.

135. Ashley, supra note 16, at 679–80.

136. Jack L. Turban et al., Understanding Pediatric Patients Who Discontinue Gender-Affirming Hormonal Interventions, JAMA Pediatrics, Oct. 2018, at 904.

137. Thomas D. Steensma et al., Gender Variance in Childhood and Sexual Orientation in Adulthood: A Prospective Study, 10 J. Sexual Med. 2723, 2730 (2013).

138. Turban et al., supra note 136, at 904 (“Some adolescents who endorse a clear desire for gender affirming hormones may change their mind after receiving them. While this may be uncomfortable for prescribing clinicians, we have observed that some gender-questioning adolescents benefit from a period of exploration that includes hormonal intervention in order to consolidate their gender identity.”).

139. Jack Turban, It’s Okay to Let Your Transgender Kid Transition—Even if They Might Change Their Mind in the Future, Vox (Oct. 22, 2018),

140. Stephen Russell et al., Chosen Name Use Is Linked to Reduced Depressive Symptoms, Suicidal Ideation, and Suicidal Behavior Among Transgender Youth, 63 J. Adolescent Health 503 (2018); Kristina Olson et al., Mental Health of Transgender Children Who Are Supported in Their Identities, 137 Pediatrics 1, 5 (2016).

141. Turban et al., supra note 136, at 904.

142. Id.

143. Ehrensaft et al., supra note 15, at 4.

144. The decisions highlight how parents are addressing the child’s gender identity, rather than focusing on other potentially salient factors in the best interests analysis. It is possible that judges are swayed by other considerations, such as income, health, or other elements that they regularly weigh in these decisions. However, their analysis turns on how the parents are addressing the child’s gender identity.

145. Lucas Holtvluwer, Ohio Father Loses Custody of 14-Year-Old Transgender Child, Could Lose Right to Object to Injections or Surgery, Post Millennial: Am. News (May 2, 2019),

146. Doe v. Bell, 754 N.Y.S.2d 846, 848–49, 853, 856 (Sup. Ct. 2003).

147. Sacklow v. Betts, 163 A.3d 367, 369 (N.J. Super. Ct. Ch. Div. 2017); see also In re Petition for Change of Name of S.E.M., No. 2019-00715, 2019 WL 6037823, at *3, *7 (Pa. Ct. Common Pleas Oct. 23, 2019) (adopting the Sacklow test). But see Tim Cwiek, Judge Denies Name Change for Trans Youth, Phila. Gay News (Mar. 2, 2016, 4:00 PM),

148. See In re Change of Name of S.E.M., 2019 WL 6037823, at *1 (noting the desire to apply for a driver’s license with a gender-conforming name).

149. Jen Christensen, Judge Gives Grandparents Custody of Ohio Transgender Teen, CNN (Feb. 16, 2018),

150. Id.

151. Kevin Grasha, Prosecutor: Parents’ Refusal of Transgender Treatment Made Teen Suicidal, Cincinnati Enquirer (Jan. 26, 2018),

152. Christensen, supra note 149.

153. Marilyn Icsman, Transgender Students: Ohio Bill Would Require Teachers to Report Children Who May Be Transgender to Parents, Cincinnati Enquirer (June 27, 2018),

154. H.B. 658, 132d Leg., Reg. Sess. (Ohio 2018).

155. Skougard, supra note 17, at 1195–96 (quoting Johnson v. Johnson, No. RF09-463371, at 1 (Cal. Super. Ct. Alameda Cnty. Oct. 10, 2009)).

156. Williams v. Frymire, 377 S.W.3d 579, 590–91 (Ky. Ct. App. 2012).

157. See, e.g., Morrison v. Cochrane, No. FBTFA040410982, 2009 WL 2357621 (Conn. Super. Ct. June 2, 2009).

158. 439 P.3d 1169, 1178–79 (Ariz. 2019).

159. Id. at 1171.

160. Id.; Paul E. v. Courtney F., 418 P.3d 413, 415 (Ariz. Ct. App. 2018), vacated in part, 439 P.3d 1169 (Ariz. 2019). The book was likely Cheryl Kilodavis, My Princess Boy (2009).

161. Paul E., 439 P.3d at 1171.

162. Paul E., 418 P.3d at 416–19.

163. Id. at 418–19.

164. Id. at 425; Paul E., 439 P.3d at 1177–79.

165. See Shrader v. Spain, No. 05-95-01649-CV, 1998 WL 40632 (Tex. Ct. App. Feb. 4, 1998); Smith v. Smith, No. 05 JE 42, 2007 WL 901599 (Ohio Ct. App. Mar. 23, 2007).

166. Am. Psychiatric Ass’n, Position Statement on Conversion Therapy and LGBTQ Patients (2018); Am. Acad. of Pediatrics, supra note 118.

167. Kuvalanka et al., supra note 83, at 59 tbl.1.

168. Final Order in Suit to Modify Parent-Child Relationship at 2, In re JA.D.Y. and JU.D.Y., DF-15-09887-S (Tex. Dist. Ct. Jan. 29, 2020) (on file with author).

169. Id.

170. Findings of Facts and Conclusions of Law at 1, In re JA.D.Y, and JU.D.Y. No. 05-16-01412-CV (Tex. Ct. App. Feb. 27, 2017) (on file with author).

171. Id. at 3.

172. See Nat’l Ctr. for Lesbian Rights, Transgender Family Law in the U.S.: A Fact Sheet for Transgender Spouses, Partners, Parents, and Youth 6 (2015),

173. Victor G. Carrion & James Lock, The Coming Out Process: Developmental Stages for Sexual Minority Youth, 2 Clinical Child Psych. & Psychiatry 369, 370 (1997).

174. George, supra note 115; Noa Ben-Asher, Conferring Dignity: The Metamorphosis of the Legal Homosexual, 37 Harv. J.L. & Gender 243, 247–53 (2014).

175. Scientific research on sexual fluidity has called into question whether sexual orientation is in fact immutable, but the law treats it as such. See, e.g., Lisa M. Diamond, Sexual Fluidity in Male and Females, 8 Current Sexual Health Rpts. 249 (2016).

176. 455 N.Y.S.2d 519 (Fam. Ct. 1982).

177. Id. at 521.

178. 661 N.Y.S.2d 779, 780 (Fam. Ct. 1997).

179. Id.

180. 496 N.Y.S.2d 940, 940 (Fam. Ct. 1985).

181. Id. at 942.

182. Id. at 943. In Brayman v. Deloach, a Georgia court granted partial summary judgment in a mother’s lawsuit against the county after the daughter had a “lesbian sexual relationship” with a county employee she met through a county-sponsored recreational program. 439 S.E.2d 709, 710 (Ga. Ct. App. 1993).

183. In re Lori M., 496 N.Y.S.2d at 943.

184. In re C.O., No. B206425, 2008 WL 4670513, at *1 (Cal. Ct. App. Oct. 23, 2008).

185. Id. at *7–12.

186. LGBTQ Youth in the Foster Care System, Human Rights Campaign, (last visited July 15, 2020).

187. Bianca D.M. Wilson et al., The Williams Inst., Sexual and Gender Minority Youth in Los Angeles Foster Care: Executive Summary 2–3 (Aug. 2014),; Our Issue, True Colors United, (last visited July 15, 2020).

188. Our Issue, supra note 187; New Report on Youth Homeless Affirms That LGBTQ Youth Disproportionately Experience Homelessness, Human Rights Campaign (Nov. 15, 2017),

189. See Lambda Legal, Safe Havens: Closing the Gap Between Recommended Practice and Reality for Transgender and Gender-Expansive Youth in Out-of-Home Care 4, 13–14 (2017), Thirteen states and D.C. protect against discrimination based on both sexual orientation and gender identity by statute or regulation. Cal. Welf. & Inst. Code § 16001.9 (2020); Colo. Rev. Stat. § 19-7-101 (2019); D.C. Code §§ 2-1402.73 (2019), 4-1303.72 (2013); Haw. Rev. Stat. Ann. § 587A-3.1 (West 2018); 20 Ill. Comp. Stat. 521/5 (2018); 18-6-1 Miss. Code R. § A II XIV (2016); Nev. Rev. Stat. Ann. § 432.525 (2017); N.H. Rev. Stat. Ann. § 170-G:21 (2018); N.J. Admin. Code § 3A:11-1.4 (2020); N.M. Code R. §§, (LexisNexis 2020); N.Y. Exec. Law §§ 292(35), 296 (2020); N.Y. Comp. Codes R. & Regs. tit. 18, § 441.24 (2020); Ohio Admin. Code § 5101:2-7-09 (2020); 42 R.I. Gen. Laws § 42-72-15 (2020); Wash. Admin. Code § 110-147-1595 (2020). Another nine states have statutes that protect against discrimination based on sexual orientation. La. Admin. Code. tit. 67, pt. V, § 7115(A)(5) (2020); Md. Code Regs. § (2020); 110 Mass. Code Regs. § 1.09 (2020); Minn. R. 2960.0050 (2020); N.D. Admin. Code 75-03-36-35 (2020); 55 Pa. Code § 3800.32 (2020); Utah Admin. Code R501-12-13 (2020); Wis. Admin. Code DCF § 56.09 (2020); Wyo. Code R. 049.0029.3 § 23. Other states, including Connecticut, Idaho, Indiana, Iowa, Maine, Michigan, Oregon, South Dakota, Tennessee, and Vermont, have departmental-based policies that protect against discrimination based on sexual orientation and gender identity. See Lambda Legal, supra, at 13.

190. Lambda Legal, supra note 189, at 4, 15–16.

191. Jason Cherkis, Queer and Loathing: Does the Foster Care System Bully Gay Kids?, Mother Jones (Dec. 2010),

192. Id.

193. Id.

194. LGBTQ youth nevertheless suffer from verbal and physical harassment in state-run facilities, including foster care. See Joel A. v. Giuliani, 218 F.3d 132 (2d Cir. 2000); Judge Blasts Hawai’i Juvenile Detention Facility for Pervasive Harassment of Gay and Transgender Youth, Am. Civ. Liberties Union (Feb. 8, 2006),

195. 238 P.3d 30, 36 (Or. Ct. App. 2010).

196. 453 N.Y.S.2d 590, 591–92 (Fam. Ct. 1982).

197. Id. at 593.

198. Id. at 593–94.

199. In re Zion J., No. A111895, 2006 WL 2709831 (Cal. Ct. App. Sept. 22, 2006).

200. Id. at *1.

201. George, supra note 15.

202. Id.

203. George, supra note 115. Same-sex sexual attraction is now more widely accepted in the United States than transgender identity, but both groups are discriminated against. Compare Gay and Lesbian Rights, Gallup, (last visited July 15, 2020) (63% of American respondents in 2017 believed that gay and lesbian relationships are morally acceptable), with Global Attitudes Towards Transgender People, Ipsos (Jan. 29, 2018), (51% of American respondents in 2017 said they would like to see greater protections for transgender people).

204. Drescher & Pula, supra note 16, at S18; WPATH, supra note 16, at 172.

205. Ehrensaft et al., supra note 15, at 14–15.

206. Id.

207. See sources cited supra note 33.

208. The history of racial passing makes this argument particularly fraught. See, e.g., Daniel J. Sharfstein, The Invisible Line: A Secret History of Race in America (2011).

209. Karen Fields & Barbara Fields, Racecraft: The Soul of Inequality in American Life (2012); Charles W. Mills, The Racial Contract (1997).

210. Biracial often conjures the image of a Black and a White parent, and indeed most of the published custody cases fit this pattern. However, the paradigm applies to a wide range of racial combinations, including Black and Latinx, White and Asian, and Latinx and American Indian, to name a few. Twila L. Perry, Race, Color, and the Adoption of Biracial Children, 17 J. Gender, Race & Just. 73, 85–86 (2014).

211. Id. at 85; Sarah S.M. Townsend, Being Mixed: Who Claims a Biracial Identity?, 18 Cultural Diversity & Ethnic Minority Psych. 91 (2012).

212. Ward v. Ward, 216 P.2d 755, 756 (Wash. 1950); Fountaine v. Fountaine, 133 N.E.2d 532, 534–35 (Ill. App. Ct. 1956); Beazley v. Davis, 545 P.2d 206, 207–08 (Nev. 1976) (overturning the lower court’s award of custody to the Black father based solely on race).

213. Katharine T. Bartlett, Comparing Race and Sex Discrimination in Custody Cases, 28 Hofstra L. Rev. 877, 891 (2000).

214. Id.

215. Id.

216. They may constitutionally do so as long as race is only one factor that they weigh, rather than being determinative. Tucker v. Tucker, 542 P.2d 789, 791 (Wash. Ct. App. 1975); Beazley, 545 P.2d at 208; Savage v. Cota, 885 N.Y.S.2d 798, 799 (App. Div. 2009).

217. 388 U.S. 1 (1967).

218. Eyer, supra note 23, at 541.

219. See, e.g., Tallman v. Tabor, 859 F. Supp. 1078 (E.D. Mich. 1994); In re Adoption of Haven, No. C-780343, 1979 Ohio App. LEXIS 9744, at *8 (Aug. 1, 1979).

220. In re Davis, 465 A.2d 614, 622 (Pa. 1983).

221. Id. at 628.

222. See, e.g., Elizabeth Bartholet, Where Do Black Children Belong? The Politics of Race Matching in Adoption, 139 U. Pa. L. Rev. 1163 (1991); compare Twila L. Perry, The Transracial Adoption Controversy: An Analysis of Discourse and Subordination, 21 N.Y.U. Rev. L. & Soc. Change 33 (1993–1994).

223. 439 N.Y.S.2d 584, 588 (Sup. Ct. 1981) (citation omitted).

224. Id. (citation omitted).

225. 395 N.Y.S.2d 290 (App. Div. 1997).

226. Id. at 295.

227. Maldonado, supra note 23, at 216.

228. Sarah E. Gaither, Mixed Results: Multiracial Research and Identity Explorations, 24 Current Directions in Psychol. Sci. 114, 114 (2015) (discussing studies).

229. Id. at 114–15.

230. Id. at 115.

231. Raysor, 395 N.Y.S.2d at 295; Farmer v. Farmer, 439 N.Y.S.2d 584, 589–90 (Sup. Ct. 1981); In re Marriage of Brown, 480 N.E.2d 246, 248 (Ind. Ct. App. 1985); Lee v. Halayko, 590 N.Y.S.2d 647, 648 (App. Div. 1992); Schultz v. Elremmash, 615 So. 2d 396, 400 (La. Ct. App. 1993); Rooney v. Rooney, 914 P.2d 212, 218 (Alaska 1996); Davis v. Davis, 658 N.Y.S.2d 548, 550–51 (App. Div. 1997).

232. 658 N.Y.S.2d at 549–50.

233. No. 30779/2002, 2005 WL 2171176, at *30 (Sup. Ct. 2005).

234. 542 N.W.2d 119, 123–24 (S.D. 1996).

235. Id. at 122.

236. 853 N.E.2d 847, 862, 868 (Ill. App. Ct. 2006).

237. 42 P.3d 1107, 1118 (Alaska 2002).

238. 620 N.W.2d 117, 121–22 (Neb. Ct. App. 2000).

239. Id.

240. See Jennifer Hefner & Daniel Eisenberg, Social Support and Mental Health Among College Students, 4 Am. J. Orthopsychiatry 491 (2009) (discussing ranges of social support and its effect on depression and anxiety); Gregory D. Zimet at al., The Multidimensional Scale of Perceived Social Support, 52 J. Personality Assessment 30 (1988) (same); see also Elizabeth A. McConnell et al., Families Matter: Social Support and Mental Health Trajectories Among Lesbian, Gay, Bisexual, and Transgender Youth, 59 J. Adolescent Health 674 (2016) (discussing how levels of family support affect LGBT youth mental health).

241. Caitlin Ryan et al., Family Acceptance in Adolescence and the Health of LGBT Young Adults, 23 J. Child & Adolescent Psych. Nursing 205, 208 tbl.1, 209 (2010).

242. The religion cases simultaneously appear particularly applicable and completely inapposite. On the one hand, religion’s mutability makes it seem especially analogous to gender expression, which changes over time. On the other, individuals in this context are choosing which religion to adopt, whereas for gender expansive children, the question is what their gender identity actually is.

243. A parent with sole legal custody has the associated legal right to make decisions as to the child’s religious upbringing, such that some courts have resolved parental conflicts by defaulting to the views of the legal custodian. See, e.g., Boerger v. Boerger, 97 A.2d 419, 427 (N.J. Super. Ct. Ch. Div. 1953); Siegel v. Siegel, 472 N.Y.S.2d 272, 273 (Sup. Ct. 1984); Von Tersch v. Von Tersch, 455 N.W.2d 130, 136 (Neb. 1990). Other decision-makers have permitted parents to expose their children to conflicting religious views, even where one has sole custody, indicating that the general legal principle is not always determinative. Munoz v. Munoz, 489 P.2d 1133, 1135 (Wash. 1971); Schwarzman v. Schwarzman, 388 N.Y.S.2d 993, 999 (Sup. Ct. 1976); Khalsa v. Khalsa, 751 P.2d 715, 720 (N.M. Ct. App. 1988); In re Marriage of Reyes, No. 1-10-1289, 2011 WL 10068626, at *5 (Ill. Ct. App. Feb. 16, 2011). Still others have modified joint custody when there is a conflict over religion, giving the objecting parent sole legal custody and then enforcing that parent’s right to make religious decisions for the child. Nunez v. Gray, No. 2210, 2017 WL 3381132, at *1, *11 (Md. Ct. Spec. App. Aug. 7, 2017).

244. See, e.g., Asch v. Asch, 397 A.2d 352, 355 (N.J. Super. Ct. App. Div. 1978) (examining separation agreement to make determination); Gluckstern v. Gluckstern, 220 N.Y.S.2d 623, 624 (Sup. Ct. 1961) (enforcing separation agreement’s religious education provision); In re Marriage of Nuechterlein, 587 N.E.2d 21, 22 (Ill. App. Ct. 1992); In re Marriage of Bonds, 5 P.3d 815, 830 (Cal. 2000). The data set on which this Article is based excludes disputes that center on the enforceability of these clauses.

245. Abbo v. Briskin, 660 So. 2d 1157, 1159 (Fla. Dist. Ct. App. 1995).

246. 388 N.Y.S.2d at 999.

247. Other courts, however, have declined to defer to the religious viewpoints of similarly aged children. Frank v. Frank, 833 A.2d 194 (Pa. Super. Ct. 2003); Gluckstern, 220 N.Y.S.2d at 623, 624.

248. 605 A.2d 172, 177 (Md. Ct. Spec. App. 1992).

249. Hicks v. Hicks, 868 A.2d 1245, 1252 (Pa. Super. Ct. 2005).

250. 176 P.3d 388, 389 (Or. 2008).

251. Id. at 394.

252. 649 A.2d 419, 419–20 (N.J. Super. Ct. App. Div. 1994). This conversion involved the issuance of a decree or statement of a decision by a rabbinical court. Id. at 420 n.4.

253. Id. at 420.

254. Id. at 420–21.

255. Pater v. Pater, 588 N.E.2d 794, 794 (Ohio 1992); Pierson v. Pierson, 143 So. 3d 1201 (Fla. Dist. Ct. App. 2014); Hanson v. Hanson, 404 N.W.2d 460, 464 (N.D. 1987).

256. Khalsa v. Khalsa, 751 P.2d 715, 721 (N.M. Ct. App. 1988).

257. No. 1-10-1289, 2011 WL 10068626 (Ill. App. Ct. Feb. 16, 2011).

258. Id. at *6.

259. Id. at *1.

260. Id. at *6–7. The appellate court found the trial court acted within its statutory authority, but the factual record was “inadequate for appellate review” as the parties had not provided their testimony as part of the record. Id.

261. Gould v. Gould, 342 N.W.2d 426, 502 (Wis. 1984); Felton v. Felton, 418 N.E.2d 606, 608 (Mass. 1981); Munoz v. Munoz, 489 P.2d 1133, 1136 (Wash. 1971).

262. 61 So. 3d 325, 332 (Ala. Civ. App. 2010).

263. No. FA 91-0399669, 1992 WL 329286, at *1 (Conn. Super. Ct. Oct. 28, 1992).

264. Id.

265. Id.

266. Kirchner v. Caughey, 606 A.2d 257, 264 (Md. 1992); Pace v. Farr, No. 1 CA-CV 09-0575, 2010 WL 5030870, at *6 (Ariz. Ct. App. Oct. 26, 2010).

267. Stevens v. Stevens, No. CA-1678, 1978 WL 217380, at *1 (Ohio Ct. App. Feb. 3, 1978).

268. 71 N.Y.S.3d 138 (App. Div. 2018).

269. No. FA030401627S, 2004 WL 1245636, at *7, *11–12 (Conn. Super. Ct. May 19, 2004).

270. Id. at *15.

271. See Ledoux v. Ledoux, 452 N.W.2d 1 (Neb. 1990); Meyer v. Meyer, 789 A.2d 921 (Vt. 2001); Burnham v. Burnham, 304 N.W.2d 58 (Neb. 1981).

272. 452 N.W.2d at 5–6; see also Meyer, 789 A.2d at 924–25; Burnham, 304 N.W.2d at 62.

273. 2004 WL 1245636, at *15.

274. 605 A.2d 172, 176 (Md. Ct. Spec. App. 1992).

275. Id. at 177.

276. Id. at 179.

277. 606 A.2d 257, 259 (Md. Ct. App. 1992).

278. Id. at 260–61.

279. Id. at 262.

280. Id.

281. Id.

282. 60 N.Y.S.3d 265 (App. Div. 2017).

283. Id. at 270–71, 276–76.

284. Id. at 272.

285. Id. at 276.

286. Id.

287. Comm. on Adolescent Health Care, supra note 100, at 2.

288. Russell B. Toomey et al., Transgender Adolescent Suicide Behavior, 142 Pediatrics e20174218, at 1 (2018).

289. Christensen, supra note 149.

290. Holtvluwer, supra note 145.

291. See George, supra note 115, at 809–10.

292. Paul E. v. Courtney F., 439 P.3d 1169, 1172 (Ariz. 2019).

293. Burns, supra note 2. The religion cases may indicate that courts should not weigh concerns that children will suffer should they have to re-transition, insofar as this is a speculative future harm.

294. See Douglas NeJaime & Reva B. Siegel, Conscience Wars: Complicity-Based Conscience Claims in Religion and Politics, 124 Yale L. J. 2516, 2516 (2015).

295. See discussion supra Parts II.A–B.

296. See discussion supra Part II.C.

297. See generally McAdams, supra note 37 (analyzing the expressive effects of laws).

298. Marie-Amélie George, The Custody Crucible: The Development of Scientific Authority About Gay and Lesbian Parents, 34 L. & Hist. Rev. 487, 487–88 (2016).

299. Julie Shapiro, Custody and Conduct: How the Law Fails Lesbian and Gay Parents and Their Children, 76 Ind. L.J. 623, 627 (1996). Transgender parents faced similar obstacles, with courts shifting from a per se presumption of unfitness to a neutral standard. Katyal & Turner, supra note 17, at 1600.

300. George, supra note 298, at 490–91.

301. Id. at 504.

302. Id.

303. Id.

304. Id.

305. Id. at 506.

306. Id at 505–06.

307. Id at 506.

308. Id. at 505–06.

309. Id.

310. No. 05 JE 42, 2007 WL 901599, at *3 (Ohio Ct. App. Mar. 23, 2007).

311. No. RF09-463371 (Cal. Super. Ct. Alameda Cnty. Oct. 10, 2009), cited in Skougard, supra note 17, at 1196.

312. In re R.A., CO74489, 2014 WL 989190, at *2 (Cal. Ct. App. Mar. 14, 2014); Williams v. Frymire, 377 S.W.3d 579, 582, 583–86 (Ky. Ct. App. 2012); Doe v. Bell, 754 N.Y.S.2d 846, 848 (Sup. Ct. 2003); Brian L. v. Admin. for Children’s Servs., 859 N.Y.S.2d 8, 11–12 (Sup. Ct. App. Div. 2008); Shrader v. Spain, No. 05-95-01649-CV, 1998 WL 40632, at *2–3 (Tex. App. Feb. 4, 1998); Holtvluwer, supra note 145; Name Change Petition for Transgender Teen, ACLU of Pa., (last visited Apr. 28, 2021).

313. See Marie-Amélie George, Queering Reproductive Justice, 54 Rich. L. Rev. 671, 682–83 (2020); Margo Kaplan, Sex Positive Law, 89 N.Y.U. L. Rev. 89, 116 (2014).

314. George, supra note 313, at 682.

315. Id.

316. Susan Wright, Depathologizing Consensual Sexual Sadism, Sexual Masochism, Transvestic Fetishism, and Fetishism, 39 Arch. Sex. Behav. 1229, 1229–30 (2012).

317. George, supra note 313, at 682; Merissa Nathan Gerson, BDSM Versus the DSM, Atlantic (Jan. 13, 2015),

318. Gerson, supra note 317.

319. Wright, supra note 316, at 1229.

320. Id. at 1230.

321. Id. The mother received custody of three of the four children; the father retained custody of the fourth due to issues concerning health insurance coverage. Id.

322. Id.

323. George, supra note 298, at 499 (citation omitted).

324. In re Grose, No. 4-12-0005, 2012 WL 7050410, at *6 (Ill. App. Ct. May 24, 2012).

325. Moreover, introducing expert evidence in cases can be prohibitively expensive for many parents. George, supra note 298, at 506.

326. Sean Hannon Williams, Divorce All the Way Down: Local Voice and Family Law’s Democratic Deficit, 98 B.U. L. Rev. 579, 588 (2018).

327. See Daniel A. Krauss & Bruce D. Sales, Legal Standards, Expertise, and Experts in the Resolution of Contested Child Custody Cases, 6 Psych. Pub. Pol’y & L. 843, 848 (2000) (discussing the variety of factors that courts weigh when determining a child’s best interests).

328. Unif. Marriage and Divorce Act §§ 408–09 (1974); see also Unif. Child Custody Jurisdiction Act § 202 (1997); Parental Kidnapping Prevention Act, 28 U.S.C. § 1738A(f) (1980).

329. Steven N. Peskind, Determining the Undeterminable: The Best Interest of the Child Standard as an Imperfect but Necessary Guidepost to Determine Child Custody, 25 N. Ill. U. L. Rev. 449, 461–62 (2005).

330. Williams, supra note 326, at 580–81.

331. Evan R. Seamone, Judicial Mindfulness, 70 U. Cin. L. Rev. 1023, 1030 (2002); Lawrence S. Wrightsman, Judicial Decision Making: Is Psychology Relevant? 55 (1999); Charles Gardner Geyh, Can the Rule of Law Survive Judicial Politics?, 97 Cornell L. Rev. 191, 213 (2012).

332. Dan Simon, A Psychological Model of Judicial Decision-Making, 30 Rutgers L.J. 1, 20 (1998).

333. Geyh, supra note 331, at 213–14.

334. See Charles Gardner Geyh, The Dimensions of Judicial Impartiality, 65 Fla. L. Rev. 493 (2014) (discussing mechanisms for ensuring judicial impartiality).

335. U.S. Const. amend. XIV, § 1; Aetna Life Ins. Co. v. Lavoie, 475 U.S. 813, 828 (1986) (“The Due Process Clause demarks only the outer boundaries of judicial disqualifications.”).

336. Model Code of Judicial Conduct r. 2.11(A)(1) (Am. Bar. Ass’n 2020).

337. Geyh, supra note 334, at 516, 519.

338. John Riley, Texas Judge Removed from Trans Child Custody Case Due to Facebook Posting, Metro Weekly (Dec. 13, 2019),; Burns, supra note 2.

339. See generally George, supra note 115 (discussing the evolution of conversion therapy laws).

340. Cal. Bus. & Prof. Code § 865.1.

341. See sources cited supra note 119.

342. Otto v. Boca Raton, 981 F.3d 854, 872 (11th Cir. 2020); see also King v. Governor of New Jersey, 767 F.3d 216, 241 (3d Cir. 2014); Pickup v. Brown, 740 F.3d 1208, 1230–31 (9th Cir. 2013).

343. See Jessica Dixon Weaver, The Principle of Subsidiarity Applied: Reforming the Legal Framework to Capture the Psychological Abuse of Children, 18 Va. J. Soc. Pol’y & L. 247, 269 (2011); see also Shawn Frances Peters, When Prayer Fails: Faith Healing, Children, and the Law (2008) (discussing cases that test the line between parental and state authority over child welfare).

344. See sources cited supra note 22.

345. Richard H. McAdams, An Attitudinal Theory of Expressive Law, 79 Or. L. Rev. 339, 373 (2000); see also Richard H. Pildes, Why Rights Are Not Trumps: Social Meanings, Expressive Harms, and Constitutionalism, 27 J. Legal Stud. 725, 755 (1998).

346. McAdams, supra note 37, at 166.

347. Elizabeth S. Anderson & Richard H. Pildes, Expressive Theories of Law: A General Restatement, 148 U. Pa. L. Rev. 1503, 1504 (2000); see Robert Cooter, Expressive Law and Economics, 27 J. Legal Stud. 585, 607–08 (1998); Richard H. Pildes, The Unintended Cultural Consequences of Public Policy: A Comment on the Symposium, 89 Mich. L. Rev. 936, 962 (1991); Milton C. Regan Jr., How Does Law Matter?, 1 Green Bag 2d 265, 271 (1998).

348. McAdams, supra note 37, at 379–80; see also George, supra note 115, at 825–27.

349. Richard H. Pildes & Cass R. Sunstein, Reinventing the Regulatory State, 62 U. Chi. L. Rev. 1, 66 (1995).

350. George, supra note 115, at 827–28. Indeed, conversion therapy bans do little to limit the practice of conversion therapy because they only regulate licensed mental health professionals, few of whom offer conversion therapy. Id. The laws are nevertheless important because they send a message to the general public that conversion therapy is harmful. Id. at 811–12, 827–28.

351. Williams, supra note 326, at 585, 593, 599. For examples of proposals, see, for example, Joseph Goldstein, Anna Freud & Albert J. Solnit, Beyond the Best Interests of the Child 38 (1973) (bright-line rule on visitation); Nancy D. Polikoff, Why Are Mothers Losing: A Brief Analysis of Criteria Used in Child Custody Determinations, 7 Women’s Rts. L. Rep. 235, 237 (1982) (primary caretaker presumption).

352. Principles of the Law of Family Dissolution: Analysis and Recommendations §§ 2.08, 3.05, 4.09, 7.05 (Am. Law Inst. 2002).

353. Elizabeth S. Scott & Robert E. Emery, Gender Politics and Child Custody: The Puzzling Persistence of the Best-Interests Standard, 77 Law. & Contemp. Probs. 69, 82–83, 91–92 (2014); Michael R. Clisham & Robin Fretwell Wilson, American Law Institute’s Principles of the Law of Family Dissolution, Eight Years After Adoption: Guiding Principles or Obligatory Footnote?, 42 Fam. L.Q. 573, 576 (2008) (discussing the Principles’ limited impact).

354. Scott & Emery, supra note 353, at 76.

355. Williams, supra note 326, at 594.

356. Id.

357. Legislators have proposed adoption and foster care restrictions based on sexual orientation, same-sex marriage bans, and “bathroom bills” to limit access to sex-segregated facilities based on assigned gender at birth. George, supra note 43, at 580–82. Elected representatives have also promoted LGBTQ rights through antidiscrimination bills, but these legislative successes have often been repealed by voters through ballot measures. Id. at 580–90.

358. See sources cited supra note 12.

359. Texas H.B. 1910, 86th Leg., Reg. Sess. (Tex. 2019).

360. Steve Toth (@Toth_4_Texas), Twitter (Oct. 26, 2019, 12:08 PM),

The material in all ABA publications is copyrighted and may be reprinted by permission only. Request reprint permission here.

By Marie-Amélie George

Marie-Amélie George is an associate professor at Wake Forest University School of Law. In August of 2021, this Article was awarded the 2021 Haub Law Emerging Scholar Award in Gender & Law, which is presented annually by the Pace University Elisabeth Haub School of Law in recognition of excellent published legal scholarship related to women, gender, and the law. Professor George is grateful for the insightful comments and feedback of Alyse Bertenthal, Meghan Boone, Michael Curtis, Mark Hall, Andrew Verstein, Jessica Dixon Weaver, and Ron Wright; the faculties of the University of Georgia and Washington University law schools; and the participants at the 2020 Family Law Scholars and Teachers Conference, 2021 Richard B. Atkinson LGBTQ Law & Policy Colloquium, and 2021 Families Unbound Conference. She also thanks Seth Barry-Hinton, Hayley Degnan, Olivia Doss, and Noah Hock, whose research assistance made this Article possible, and Lisa Grumet and the New York Law School student editors of the Family Law Quarterly, who carefully edited the Article.