In volume 83(3) of the Modern Law Review, Professor Alex Sharpe asks whether gender self-declaration will undermine women’s rights and lead to an increase in harms. The question is posed in the context of recent proposals, which may have stalled in in England and Wales, but which appear set to go ahead in Scotland, to amend the Gender Recognition Act 2004 (henceforth, GRA), so an individual may change gender (which in official documents means sex) by self-declaration alone. In brief, Professor Sharpe argues:
- Reforming the GRA 2004 will not undermine the sex exceptions in the Equality Act 2010 (henceforth, EA); and
- Even if the reform will affect those rights, this will not increase harms for women [italics by the authors].
Here we advance the case against GRA reform (and, incidentally, against any redefinition of the word ‘woman’ in law), focusing on how self-ID impacts on ‘woman,’ examining ‘harm,’ and providing a rejoinder to Professor Sharpe’s three-pronged rebuttal of the gender critical argument against reform of the GRA. Our argument is that GRA reform indirectly undermines women’s rights to single sex spaces (amongst other single sex rights currently afforded to women on the basis of sex) and that this will lead to harm. Before we do that, we must first examine the present state of the law.
The GRA Reform and the Sex Exceptions in the EA
Although the GRA was introduced to allow legal sex change for both sexes, we adopt Professor Sharpe’s focus on women’s rights only. Due to the GRA, once a man acquires a Gender Recognition Certificate (henceforth, GRC) he legally becomes a woman, subject to general and specific limitations, evidencing that the recognition is a legal fiction with no bearing on biological status. This is underlined by the subsequent wording of both the EA’s sex exceptions themselves and the later Gender Representation on Public Boards (Scotland) Act 2018.
The sex exceptions in Schedule 3, Part 7 of the Equality Act allow single sex and separate sex services, and, by virtue of section 28, the exclusion of people in the protected category of gender reassignment, provided this is done as a ‘proportionate means of achieving a legitimate aim.’ Professor Sharpe disagrees with gender critical feminists on how the proportionality and legitimacy tests should be applied.
Legal Definitions and Language
In order to interrogate whether women’s legal rights are affected, there must be an exploration of the distinction between biological and state-recognised sex. It is unhelpful to be forced, as Professor Sharpe‘s approach would require us, to use unwieldy terms such as ‘non-trans’ for men and women who are not trans. This language forces men and women into a sub-category of their own sex class, defined by failure to belong to another, much smaller, class whom the author centres.
We find Professor Sharpe’s lexicon elides the distinction between woman and transwoman, making it difficult clearly to examine women’s rights as distinct from transwomen’s rights. Confusing the two groups undermines the examination and also recognition of the potential impacts of the reform on women.
In contrast to Professor Sharpe, we maintain that transwomen are males who are trans, and transmen are females who are trans; transwomen are a subset of the male sex class and transmen are a subset of the female sex class. As the terms transwoman and transman are not legally defined, the closest legal category is gender reassignment, as per section 7 of the EA:
A person has the protected characteristic of gender reassignment if the person is proposing to undergo, is undergoing or has undergone a process (or part of a process) for the purpose of reassigning the person’s sex by changing physiological or other attributes of sex.
According to section 9(1) of the GRA, an individual not in possession of a GRC remains the sex recorded at birth. Further, biological sex is unaffected by possession of a GRC.
We regard it as a fundamental human right that a private female citizen, when confronted by a demand to share intimate services or spaces made by an individual whom she perceives as a male, is able to assert her right to withhold consent and not be victimised for doing so. This is a fundamental sex-based demand, on human rights and feminist grounds.
We refute Professor Sharpe’s footnoted affirmation that ‘all trans women are women’ as neither legally accurate nor biologically correct nor logical. We propose that the ordinary meaning of section 212 of the EA, which defines woman as ‘a female of any age,’ references the biological sex class, subject as aforementioned to whether a GRC has been obtained. We suggest that Professor Sharpe’s statement, that biology need not determine sex, is inaccurate. The legal system admits biological sex and State-recognised sex but not self-declared sex. Additionally, if it currently already did this, the GRA would not need reforming to introduce self-ID.
The Fallacy of the Harm Argument: A Response
We turn now to Professor Sharpe’s presentation of the gender critical argument that women’s rights would be harmed by self-ID. The entire argument is predicated on the equivalence between harm and sexual harm. Harm is further quantitively restricted to a ‘significant increase’ of harms. This focus overlooks our – preferred – wider definition of harm, which includes any effect on the rights of women originating from a redefinition of the legal category of woman to include individuals not possessing the necessary criteria to belong to the category. Specifically, gender critical feminists believe self-ID will undermine women’s rights in many ways, including its effects on agency and self-determination, the rights of religious women, the accuracy of statistical data, medical and health issues, criminal justice, education, sports, the rights of lesbians, and communication of female health issues.
Accepting Professor Sharpe’s focus on sexual assault as the main or only component of serious harm invites reflection on what feminists consider sexual harm, and of female mitigation strategies for those harms. Many feminists consider sexual objectification, harassment, and abuse, sometimes called ‘micro-aggressions,’ which are experienced by women from a very early age, to be sexual harm. Their occurrence is much higher than police-logged sexual assaults and each individual instance is experienced as having a lesser impact. Few women bother to report sexual harassment incidents including flashing, obscene language, unwanted touching, etc. Their cumulative effect however is such that most women’s behaviour is shaped by these micro-aggressions and their perceived threat is pervasive. Women and girls suffer elevated psychosocial stress as a response to the risk of sexual violence and voyeurism. We do not believe that Professor Sharpe’s argument addresses this wider understanding of sexual harms.
Professor Sharpe begins by describing gender critical feminists’ equivalence between ‘trans women’ and ‘non-trans men’ as false, adding:
the claim trans women pose some kind of special risk to non-trans women is empirically without merit. It is based not on credible evidence of harm but on different patterns of offending between men and women and the assumption trans women’s offending patterns are likely to mirror those of non-trans men.
Behind this statement is the assertion that transwomen already use female spaces in great numbers and that the EA sex exceptions are not relied upon in practice, due to the perception of a high bar militating against their application. This is, however, inconsistent with Sharpe’s statement that transwomen ‘self-exclude’ from female-only spaces, because of the chilling effect of the sex exceptions and the toxic effect of the debate on the GRA reform.
In Professor Sharpe’s view, the risk of harm to women from transwomen is not comparable to the risk of harm to women from men; this view is, however, not supported by any empirical data. Professor Sharpe goes on to assert that transwomen are in fact ‘especially vulnerable from male violence.’ While transwomen may be vulnerable to male violence, this does not mean that they are not a threat to women; nor does a comparison of the impact of male violence on each group have any bearing on that threat. A vulnerable cohort of males (such as transwomen) may still pose a significant threat to women.
Crime data for transwoman is difficult to collect due to data recording which captures gender rather than sex. However relevant sources of information do exist. The first source is prison and crime statistics disaggregating data on the basis of sex and gender (to control for transgender individuals). A Freedom of Information request by ‘Fair Play for Women,’ published in October 2017, revealed that 60 out of 125 convicted transgender individuals (male to female) were sex offenders (this number may be too low as it does not include individuals with a GRC, or on short sentences, or who did not declare any transgender status). Recent data provided by the UK Government show that seven out of 124 sexual assaults in the female estate since 2011 were committed by transprisoners. When considering the proportion of the female prison estate made up of transgender inmates, the figures (which are admittedly low) show that transgender prisoners carried out five times more sexual assaults then other inmates. Furthermore, several expert bodies provided evidence to Parliament, during GRA reform review, of issues arising from housing transgender prisoners in female estates. These include male prisoners opportunistically identifying as transgender in order to be moved to female estates.
The second source is the 2011 Swedish study of 324 sex-reassigned persons (191 male-to-females, 133 female-to-males), ‘conducted to estimate mortality, morbidity, and criminal rate after surgical sex reassignment of transsexual persons; the findings on criminal offending rate were that male-to-female transgender individuals maintain male patterns of offending’. A fortiori, it could be hypothesised that transwomen who have undergone no significant procedure to approximate the female sex would present patterns of offending typical of their sex.
Professor Sharpe relies on US data to assess whether allowing transwomen in female spaces has resulted in increased harms for women, according to the definition of harm adopted in the article. There are several issues with this approach:
- The legal framework is different, both between the UK and the US, and within US states.
- Data collected on basis of self-ID makes sex disaggregation impossible. This issue also arises for the UK of course, but the different legal framework means, for example, that while transgender individuals are specifically protected under gender reassignment in the UK, there is no equivalent clear statutory protection in the US, making disaggregating data more difficult.
- We hypothesise that certain women, for example survivors of sexual assault or religious women are likely to be self-excluding from gender neutral spaces or avoid female toilets open to transwomen.
In the study relied upon by Professor Sharpe, conducted in Massachusetts, rates of sexual violence in single sex toilets and in recently introduced gender neutral toilets were compared. It is unclear whether the City Ordinances concerned the substitution of female and male toilets with gender neutral ones, or whether transgender individuals were allowed to choose which toilet to use. The researchers queried local police forces on the incidence of sexual crimes in toilets. This narrow query excluded, naturally, all those lesser incidents which went unreported.
The study shows no change in the rate of recorded sexual violence in toilets, possibly showing that there is no specific advantage in introducing gender neutral toilets on the limited grounds of reducing harms for transwomen. The study further acknowledges several limitations:
- Sex crimes are notoriously under-reported and lesser sexual offences even more so. This is particularly troubling if these lesser offences take place in spaces traditionally considered relatively safe.
- The police forces contacted for the study admitted recording crime without distinguishing between sex and gender, both for the offender and the victim.
- The quality of data is uneven, including because of the ways crimes were recorded by different police forces. The authors admit these crimes are normally rare in toilets, seemingly unable to draw the obvious conclusion that they are rare because toilets are normally segregated by sex.
Notwithstanding the limitations of this study, Professor Sharpe cites it prominently to prove that reducing or eliminating female-only spaces has no effects on the safety of women, while noting the absence of similar studies in the UK.
Professor Sharpe states that there is no empirical merit in considering sex-differentiated offending patterns but provides no grounds for this argument. Concurrently Professor Sharpe argues that the (admittedly scant) evidence of increased harms to women due to opening up single-sex spaces to transwomen is insufficient to prove increased risk, without acknowledging that untaken steps cannot be measured. We point out that once data are only recorded on the basis of self-ID rather than biological sex (as seems to now be the case in the UK) then we will not have sex disaggregated data, and so we will not be able to measure whether removing safeguards results in increased assaults by males.
Professor Sharpe also considers the gender critical argument that men might exploit self-ID to access female-only spaces. The argument is dismissed summarily: ‘It is also somewhat far-fetched to imagine non trans men with bad motives would employ this tactic when access to women’s bodies in our society is ubiquitous.’ This statement ignores the potential for self-ID to facilitate sexual assault.
We argue that self-ID creates a presumption of access to female spaces, instead of a presumption of exclusion, even for evidently male individuals. It creates an unverifiable legal mechanism impervious to objective interrogation dependent upon an individual’s subjective apprehension of their gender identity, removing objective scrutiny. This creates a legal loophole exploitable by predators, fetishists, exhibitionists and voyeurs. Determining how the temporary assumption of false identity as well as longer-term false declaration can be assessed is crucial to the prevention of structural harm to women’s rights.
We are unconvinced that criminal sanctions will form the deterrent which Professor Sharpe optimistically avers. We point out that where motives for a false declaration are themselves criminal, a criminal sanction for the declaration is an unlikely deterrent, particularly when considering the UK’s very low rape conviction rates. Professor Sharpe’s argument, that it is both unlikely that a male predator would opportunistically attempt to access female spaces and that, even if he did, this is not a problem, because adequate safeguards are in place is naive at best. Professor Sharpe seems to consider that the example of the Irish GRA 2015, which makes it an offence to ‘knowingly or recklessly provide information to the Minister that is false or misleading in a material respect,’ should assuage concerns; we are underwhelmed. In our view the system of criminal sanctions for male violence is already ineffective, so we believe the confidence in it misplaced.
We question whether it is possible to create practical, adequate safeguards, especially in urgent situations, such as in domestic violence or rape shelters. The objective justification test has already been interpreted by bodies such as the EHRC in such a way that front-line staff are burdened with cumbersome risk-assessments, rather than making a sex-based judgment.
Assessing whether a declaration is false is relevant to any criminal sanction. There are three concerns: facts contradicting a declaration of changed gender; ‘detransitioning’ i.e. changing one’s mind on the gender reassignment process and expressing the wish to return to one’s original sex; and thirdly displaying behaviour inconsistent with one’s new ‘gender.’ All of these elements are problematic: the first potentially involves third-party evidence, as the criteria for ‘living in the acquired gender’ include changing one’s name socially or ‘presenting as’ the acquired gender; the second includes people who change their mind in good faith and should not be penalised for doing so, but helped to revert to their original sex without having to go to the process of gender reassignment again; and the third relies on consensus regarding sex-specific actions. For example, does rape render a declaration false or are we unable to interrogate the declaration? If acquired female legal status cannot be questioned, then are criminal sanctions moot?
Professor Sharpe cannot avoid tackling the infamous case of ‘Karen White,’ a convicted rapist without a GRC and with no hormonal or surgery background who was housed in a female estate on his request and committed sexual offences against female inmates while in detention there. Professor Sharpe’s argues that while men do pose a danger to women in female only spaces, there is no evidence that men would pretend to be women to access female spaces, going on to suggest that predatory access is considered the social privilege of transwomen. This argument is muddled and misses the concern that men (not transwomen) would pretend to be women. It is interesting that Professor Sharpe frames predatory access as a privilege. The logic of the argument would instead place ‘privilege’ on having been born a woman with a guaranteed right of access.
Professor Sharpe concludes by stating that excluding ‘a whole class of women [transwomen] from women-only spaces is not justified in public policy terms by a handful of cases […]’. We believe that Professor Sharpe’s apprehension of how the sex exceptions function is flawed. It is not absurd to exclude a whole class of people, as this is precisely what the sex exceptions are for, as long as the exclusion is justified. We already exclude men as a class from women’s spaces. The only question is whether transwomen belong to the male class, as we aver, or to the female class.
We turn back now to the proportionality/necessity discussion on which there is also a clear difference of opinion between Professor Sharpe and ourselves. Professor Sharpe does not appear to appreciate that the proportionality of the objective justification required for legitimate discrimination relates to the mitigation measures deployed by the service provider, not to the discomfort which women may experience when their consent is not requested or their privacy and dignity ignored. It is legally incorrect to focus on the service user and argue that her discomfort needs to ‘be shown to meet the threshold of “proportionality”’ before the sex exceptions can be applied. The question instead should be whether avoiding her discomfort is a legitimate aim. A proportionality test is applied to the measures adopted rather than to her reaction to a mixed-sex space.
On Reform of the GRA and Women’s Rights: A Response
Professor Sharpe addresses reform of the GRA and women’s rights by separating the arguments advanced by gender critical feminists against change into three strands. Before tackling them, it is important to note that the topic is not just the reform of the GRA, but the interaction of the proposed reformed GRA with the sex exceptions contained in the EA. This is important, because effects of reform must be assessed not just generically in relation to ‘harm,’ but also specifically in relation to how they impact on EA sex-based rights compared to gender reassignment rights. The three strands in which Professor Sharpe grounds the arguments raised by gender critical feminists against reform of the GRA, are as follows, with our comments set out for each:
Argument One: sex-based exceptions cannot be invoked against transwomen who hold a GRC.
We agree with Professor Sharpe that a GRC is irrelevant to the application of the sex-based exceptions in the EA, by virtue of s 9(3) GRA. In short, whether an individual is or is not in possession of a GRC is not relevant to whether that same individual is considered to belong to the protected category of gender reassignment in the EA, subject to the exceptions contained in the accompanying Schedules, and this is in contrast to the argument put forward by some gender critical feminists, that a GRC grants additional rights of access to female spaces and services to the individuals having one (and secondarily, that simplifying the procedure to obtain a GRC, will increase the number of individuals being granted these additional rights). We disagree with the argument in support of this position taken by Professor Sharpe, to refer to the Code of Practice of the EA, the Equality and Human Rights Commission statements and case law. We believe this matter can be disposed of as a simple instance of statutory interpretation.
Argument Two: the appropriate ‘comparator’ for a transwoman without a GRC in a claim for gender reassignment discrimination is a man.
It is evident that, regardless of how one assesses the effect of the GRC on the application of the sex exceptions in the EA, the very limited number of GRC holders in the UK, combined with the ‘open’ category of individuals capable of raising a claim of discrimination under the protected category of gender reassignment, evidences that the real debate falls under this rubric. However, the GRA remains relevant, because an individual not in possession of a GRC remains the sex recorded at birth; therefore, a transwoman without a GRC is still legally male, holding the same rights with regards to the sex exceptions in the EA as men.
Professor Sharpe disagrees with gender critical feminists that the comparator established for a male having the protected characteristic of gender reassignment is a male without the protected characteristic of gender reassignment, as established in 2013 (post-EA) in R (Green) v Secretary of State for Justice, where the claimant, a transwoman in prison, claimed discrimination for having been denied access to clothing and accessories normally designated for women’s use. Unlike Professor Sharpe, we consider that Green is the legal precedent here and cannot have its authority undermined by a case from a lower court, namely the unreported County Court claim of Brook v Tasker, as averred by Professor Sharpe.
We do not believe that this renders sections of the EA unworkable, as transwomen are entitled to rely upon the protected characteristic of gender reassignment in order to prevent less favourable treatment and harassment. We have seen no evidence in Hansard to show that Parliament addressed the issue of women-only spaces in relation to whether males with the protected characteristic of gender reassignment should be granted access to female-only spaces. There is in short, no evidence, neither as a matter of statutory interpretation, nor of parliamentary record, that the EA was designed to give as the ‘default position’ transwomen right to access to female only spaces.
Argument Three: it would be impossible for women’s organisations wanting to rely on the sex exceptions in the EA to establish the birth sex of a GRC holder, and this problem would be exacerbated by the proposed reforms, which liberalise and simplify obtaining a GRC.
As summarised by Professor Sharpe, feminists have argued that it would be impossible for women-only organisation to ascertain legally the birth sex of a GRC-holder. Once again, it has to be noted, and we agree here again with Professor Sharpe, that a GRC does not make a difference to the application of the sex-based exceptions in the EA. The problem, rather, is that organisations already pressured to open up female-only spaces on the basis of misleading guidance, might feel an increased amount of pressure once the access is under the guise of a document granting a legal change of sex. But here Professor Sharpe’s narrow focus on trans status fails to engage with the fact that the issues here are centred on biological sex. Gender critical concerns relate to the impact of biological sex on religious rights, previous trauma, privacy, dignity and sports irrespective of legal sex status; these issues remain unexplored and unacknowledged. Professor Sharpe’s comparison of female discomfort in the presence of male bodies with discomfort in the presence of lesbians in female spaces relies on the misconception that sexual violence is the result of sexual attraction, rather than male violence against females using sex. Conflating same-sex female sexual orientation with male violence is, we suggest, a category error.
Professor Sharpe states that transwomen are ‘gender non-conforming’ or ‘morphologically atypical,’ shifting focus to women policing gender non-conforming women. We accept that transwomen are gender non-conforming males and would have liked to have seen acknowledgment that women have the right to request exclusion of morphologically typical males (and legal males, in the overwhelming majority of transwomen) from female spaces. This is a point of absolute disagreement with Professor Sharpe where, we argue, biological sex is still a legally significant category under the EA, while Professor Sharpe’s position appears to be that it is not. The burden of proof that gender identity should trump sex rests with the proponents of this change, certainly not on the category mostly affected by it, women.
We suggest that Professor Sharpe’s refutation of how self-ID undermines women’s rights is faulty, as it is based on the assertion that transwomen are women, which we have shown to be neither legally, biologically nor logically accurate. We also question the interpretation of the sex exceptions and we are concerned about the direction of travel here, which we believe is towards direct, de facto self-ID.
Service providers are increasingly removing single sex spaces, replacing them with ‘gender neutral’ ones. We are concerned that the implications of this shift are unexplored. Professor Sharpe does not engage with the structural inequality of these spaces becoming the norm. The distinction between transwomen (already) accessing female-only spaces and an express policy to provide shared spaces in order to accommodate transgender and non-binary individuals must be examined; it may seem inconsequential, but conceptually it is important. If, for example, as it is often claimed, transwomen are at increased risk of violence in places used by males, gender neutral toilets would pose an unacceptable risk and, as such, would be rejected by transwomen as well as by women. In practice this does not happen, quite the contrary in fact.
If, on the other hand, Professor Sharpe wishes to argue that transwomen should be granted access to female-only spaces, there needs to be a more cogent articulation of the objective justification for the choice of disapplying the EA sex exceptions. Arguing for the elimination of sex-based exceptions does not mitigate against the risk of male violence; it simply removes all security measures, as it eliminates female-only spaces, exposing both women and transwomen to the same risk of male violence, aggression that they claim to want to escape by using female-only spaces.
Our foremost concern, however, is that the enforcement of single sex services is already compromised and that the sex exceptions are, in practice, not being invoked against transwomen with or without GRC, because the official guidance is misleading and, we suggest incorrect. Advice from the Government, the Equality and Human Rights Commission and also that of third parties such as Stonewall has echoed this incorrect guidance with the result that the sex exceptions are often incorrectly applied. Service providers have stated that they are confused as to whether transwomen have the right to access female-only spaces. Professor Sharpe appears to contribute to this confusion, when suggesting that prison risk assessment policies for transwomen (designed to assess whether legitimate discrimination is objectively justified) are applied due to transphobia rather than the public sector equality duty to assess whether the sex exceptions should be applied.
Gender critical feminists further argue that single sex spaces help prevent and/or mitigate sexual harms. Removal of risk-mitigation measures against male sexual assault therefore, increases harm and undermines women’s rights. Risk management assesses the probability of assault against the impact of that assault and then determines whether the risk should be accepted or mitigated. In 2017 in the UK there were only 2.1 recorded offences of sexual assault per 1000 population, rendering the statistical probability of sexual assault quite low. The consequences of an assault for the victim can, however, be life-changing and so we have single sex facilities for women as risk mitigation. This measure meets the requirement of the EA 2010 sex exceptions such that excluding men is a proportionate means of achieving a legitimate aim. As single sex spaces are normally not policed, we rely upon mutual respect between the sexes, coupled with the power for women to object to and challenge male presence. We assert that self-ID further decouples legal gender reassignment from any significant physical change, and even the requirement to live in the acquired gender for a significant extension of time, and we hypothesise this renders the enforcement of single sex spaces problematic. Opening spaces to those who self-declare their sex and who are perceived as males undermines women’s empowerment to challenge all male-bodied entrants and, we posit, will embolden male opportunists to enter single sex spaces, reducing their risk-mitigation role.
It is evident that Professor Sharpe has not considered this issue: we note the throw-away comment that ever-present concerns about male opportunists are concerns about ‘non trans bogeymen.’ Bodily integrity and autonomy necessarily require that a woman may choose to exclude a male (regardless of gender identity) from female-only spaces, including communal changing rooms, rape refuges, prisons, gynaecological care, nursing and medical care. Restricting the argument to toilets, which are single-use and more easily modified to allow privacy, misses the bigger picture. Dismissing concerns about removal of safeguards as conjuring up a bogeyman reveals a profound misunderstanding of the impact of male presence upon women and girls. The threat of male sexual assault is not a bogeyman but a reality. Removing women’s ability to withhold consent or to voluntarily accept the risk of assault strips her of autonomy and holds her real fears of sexual assault in contempt. We, like Kathleen Stock object to letting ‘even small numbers of females be the automatic collateral in sweeping social changes such as those proposed.’