Women-only spaces and the right to exclude

'The woman-only meeting is a fundamental challenge to the structure of power. It is always the privilege of the master to enter the slave's hut. The slave who decides to exclude the master from her hut is declaring herself not a slave' (Frye 1983, p. 104).

I. ‘Sex’ in the law

In August 2019, a bill was passed in Victoria, Australia, making it possible for people to change their official record of sex in the birth register by making a statutory declaration that they believe their sex to be as nominated. From May 2020, any person observed male at birth will be able to change their legal sex to ‘female’, and any person observed female at birth will be able to change their legal sex to ‘male’. Similar bills have been considered in other countries (New Zealand, the UK), and have already passed into law in other states of Australia (Tasmania) and in other countries (Ireland, Malta, Norway, Argentina, Portugal, and Belgium). (1) (2)

Such bills have implications for access to single-sex spaces, services, and provisions (see e.g. Gilligan 2019; Alves 2018; Patrick 2016; Murphy 2018). If any biologically male person can change his legal sex to female, what implications does this have for women’s sports,(3) women’s scholarships, girls’ schools, rape and domestic violence shelters, female prisons, and more? There are rising numbers of trans people,(4) so this is no longer an issue of tiny numbers of people needing to be accommodated within existing legal and social categories (which it arguably was in the 1960s when the conflicts between radical feminists and transwomen first gained some prominence—see Goldberg 2014). Most transwomen go through a male puberty (which is relevant to women’s sports) and increasing numbers are choosing not to have sex reassignment surgery (which is relevant to spaces involving full or partial nudity).

It is common for media and political debate to present the stakeholders in the legal change as being trans groups only. But there are obviously (at least) two stakeholder groups: the transwomen who wish to acquire the legal sex ‘female’/‘woman’, and those who are as a matter of biological fact female/women. Legal recognition of a new category, like ‘nonbinary’, is a different matter from legal recognition as a member of an already existing category. The first need involve no one but the members of the new category while the second invariably involves those already in the category. It is a principle of liberal democracy that people have a say in matters that affect them. Changing the definition of legal sex in a way that affects access to single-sex services, spaces, and provisions will obviously affect women.

Some would respond to this by pointing to the way that transwomen are marginalised. But marginalisation cannot function as a trump here, because women are also marginalised, which is one reason why women-only spaces, services, and provisions exist. It needs to be worked out what interests women-only spaces serve, and whether we have adequate reason to maintain them as women-only, even given facts about transwomen’s marginalisation. The central question of the paper is: do women have the right to exclude transwomen from women-only spaces? (5)

In Section II I’ll argue that biological sex matters politically, and should be protected legally—at least until such a time as there is no longer sex discrimination. In Section III I’ll turn to the rationales for women-only spaces, arguing that there are eight independent rationales that together overdetermine the moral justification for maintaining particular spaces as women-only. I address a package of spaces, including prisons, changing rooms, fitting rooms, bathrooms, shelters, rape and domestic violence refuges, gyms, spas, sports, schools, accommodations, shortlists, prizes, quotas, political groups, clubs, events, festivals, and terms. I do this instead of taking each space in turn, for two reasons. The first is that these spaces have something in common, namely that they were established in response to women’s marginalisation under patriarchy, the second is that women themselves have demanded—and in many cases themselves established—these spaces, services, and provisions as a way to advance their own political interests. For example, as women entered the workforce and faced harassment from men in men’s facilities, they pushed for women’s bathrooms (Burlette-Carter 2018); women in the second-wave of feminism established, funded, and volunteered at women’s refuges and shelters (Jeffreys 2018, pp. 57-60). The arguments of these two sections taken together make a strong case against self-identification as the basis for legal sex. In Section IV I turn to the objection that I’ve obtained my conclusion through linguistic sleight of hand, and answer it by saying that choices about naming don’t affect the underlying questions about the basis for inclusion or exclusion.

Around the world, women are subject to sex-specific forms of violence, including Female Genital Mutilation, female infanticide, child brides, forced marriage, rape, domestic violence, intimate partner strangulation, ‘accidental’ killing during sex, prostitution, pornography, and forced surrogacy. Women are presented across the media as sexual objects for men’s gratification. Women in poorer countries (and poorer parts of rich countries) experience period poverty. The ‘default male’ assumption in medicine and product design puts female people at greater mortality risk (Criado Perez 2019). Women are under-represented in politics, and in some employment areas, as well as at upper-levels in all employment areas. Women still don’t have full reproductive rights in many countries, and are still subject to pregnancy and breastfeeding discrimination in the workplace and in access to public life. Women undertake a disproportionate share of unpaid labour, e.g. in childcare and the running of households.

In International Law, the Convention on the Elimination of All Forms of Discrimination against Women (1979) (CEDAW) notes that despite various legal instruments existing to uphold the equality of men and women, ‘discrimination against women continues to exist’ (CEDAW, p. 1). (6) It is clear from the wording of the Convention that it means ‘women’ to be understood as synonymous with female sex, for example ‘Bearing in mind the great contribution of women to the welfare of the family and to the development of society, so far not fully recognized, the social significance of maternity and the role of both parents in the family and in the upbringing of children, and aware that the role of women in procreation should not be a basis for discrimination but that the upbringing of children requires a sharing of responsibility between men and women and society as a whole’ (ibid, p. 2 [my emphasis]).

It defines discrimination against women as ‘distinction, exclusion or restriction made on the basis of sex’ which affects the ‘enjoyment or exercise by women… of human rights and fundamental freedoms’ (ibid [my emphasis]). Article 5 talks about prejudices based on ideas about the inferiority or superiority of either of the sexes, and about stereotyped roles for men and women; Article 6 talks about measures to suppress the traffic of women and the exploitation of women through prostitution; Article 11 talks about prevention of discrimination against women in employment, particularly relating to pregnancy and maternity leave; and Article 12 talks about pregnancy and lactation.

In Australia specifically, women first got the right to vote in South Australia in 1895. The first women were elected to the House of Representatives and Senate in 1943. It wasn’t until 1956 that the Marriage Bar (which prohibited women from continuing to work after marriage), was lifted for women in education. It wasn’t until 1961 that women could acquire the contraceptive pill (and even then, only with a husband). Indigenous women didn’t get the right to vote in Federal elections until 1962. The Marriage Bar for women working in the Commonwealth Public Service was only lifted in 1966. The first abortion rights came in 1966.

In 1972, Australian women were granted the right to equal pay. In 1974, the minimum wage was extended to female workers. In 1975, the first women’s refuges received government funding. In 1975, women became able to file for no-fault divorce. Rape in marriage was outlawed in South Australia in 1976. In 1977 the Victorian Equal Opportunity Act outlawed discrimination on the grounds of gender or marital status. In 1979 women who had been employed for 12 months or more became entitled to 52 weeks of unpaid maternity leave. Sexual harassment was outlawed by an amendment to the Victorian Equal Opportunity Act in 1995. The first female Prime Minister was elected in 2010. (7)

All of this is fairly recent history. There are Australian women alive today who will have been forced to give up their jobs for marriage, who will have had unwanted pregnancies, who will have been paid less than a man for doing the same work, who will have been prevented from voting in elections that affected her, who had no legal recourse when forced into sex within their marriages. Most Australian women spent most of their lives without seeing a woman occupy the highest political office in the country. Older Australian women lived through a time when they were subordinated to men (as wives and homemakers lacking in social and political rights) and excluded from work, from political life, from public life, and from sport.

This sex-based marginalisation has not been fully mitigated, and is not yet over (Australia has not achieved women’s liberation). Some of the way that we can mitigate women’s historical exclusion and ongoing underrepresentation is by maintaining single-sex spaces, services, and provisions. For example, women are underrepresented in politics, and we can partly remedy this by having women-only shortlists; women have been historically excluded from sport in a way that has long-lasting effects, and we can remedy that by having women’s sports, and pushing for them to be equally funded and publicized. Until such a time as sex doesn’t make such a difference to how people’s lives go, it should be protected legally. Women-only spaces, services, and provisions are part of this protection.

III. Moral rationales for women-only spaces

Using the terminology of ‘women-only spaces’ makes it seem like there’s one specific kind of space that serves one specific purpose for one specific group. But that couldn’t be further from the truth. The term is used to refer to a diverse range of things, some of which aren’t really ‘spaces’ at all in any strict sense. When radical and gender critical feminists talk about the importance of keeping women-only spaces women-only, they might be referring to any or all of:

Prisons | Changing rooms | Fitting rooms | Bathrooms | Shelters | Rape and domestic violence refuges| Gyms| Spas| Sports| Schools| Accommodations| Shortlists| Prizes| Quotas| Political groups | Clubs | Events | Festivals | Terms (8) (9)

Sheila Jeffreys writes in The Lesbian Revolution (2018) about the various kinds of women-only spaces available to—and the role they played in the feminist organizing of—the women of the second-wave of feminism. These include feminist bookstores (Jeffreys 2018, pp. 39-41), women’s squats (ibid, pp. 42-44), women’s discos (ibid, pp. 47-48), women’s and lesbian theatre (ibid, pp. 48-50), and feminist conferences (ibid, p. 50). But the moral justification of having these spaces seems to have been more or less taken as obvious by feminists. My project here is to articulate those underlying justifications.

There are at least eight distinct moral rationales for women-only spaces, or so I suggest, and elaborate upon below. These are:

1. safety;

2. privacy/dignity/comfort;

3. justice/fairness;

4. respite;

5. likelihood of shared bodily experience;

6. intimate association;

7. self-determination; and

8. intent of the creators.

The first two are regularly suggested by radical and gender critical feminists in discussions over spaces like prisons, bathrooms, changing rooms, fitting rooms, bathrooms, and toilets. The others are less commonly invoked (although fairness looms large in the discussion about transwomen competing in women’s sporting categories), but are no less important. I take each of these in turn below, ending each sub-section with the list of spaces, where those not struck out have been rationalised by the discussion in that subsection.

1. Safety

Women have an interest in being safe, from physical assault and sexual violence. However, we know that the risk of physical assault and sexual violence is highest for women in their own homes, and from men they know; not from men in public spaces. The risk of the latter is not zero, but it may be fairly low. The risk also goes up as the egregiousness of the offence goes down (stranger rape is fairly rare, while voyeurism is less so). What risk should we accept? Given the harms that are at stake, and their long-lasting effects, it seems that we’re justified in deploying a precautionary approach: exclude all biologically male people, until there is evidence establishing that a particular group of biologically male people are no more of a threat to women than other women are. (And in doing this, be sensitive to the differences between transwomen mentioned already: transwomen who have passed as women and been treated as women for a substantial period of time will have more in common with women than those more recently transitioned and do not pass).

The safety rationale seems to justify prisons—there have already been problems of sexual assault from housing transwomen in the female estate in the UK, (10) not to mention that the incentives for gynephilic males (i.e. males attracted to females) to transition and gain access to women would be significant. Research out of the UK in 2018 also found that women were at higher risk of sexual assault in unisex changing rooms, with 90% of complaints about sexual assault, voyeurism, and harassment coming from unisex facilities (Hosie 2018). (The extent to which this is relevant depends on whether this same outcome is likely in only partially unisex— including only transwomen rather than all males—rather than fully unisex changing rooms. There is no reason to think it isn’t, at least if we’re including all biologically male people who claim to be trans today, because even if the experience of gender dysphoria can be expected to have some inhibiting effect on socialization (although there is no evidence that this is true), those whose transitioning is motivated by queer theory or social contagion did not experience gender dysphoria, so their inclusion is on a par with the inclusion of any other biologically male person, and so makes spaces unisex in a similar way, just to a lesser extent). There have also been assaults in bathrooms, of women by transwomen (again, to make the point that the risk is not zero) (Corbishley 2019). Arguably, then, in any spaces where women are vulnerable—so those involving full or partial nudity, or intimate physical contact, or sleeping—there is a safety rationale for exclusion. Of the spaces in our list, the safety rationale gets us prisons, changing rooms, fitting rooms, bathrooms, shelters, rape and domestic violence refuges, [boarding] schools, and accommodations (shelters, refuges, and boarding schools because they centrally involve accommodations).

So we have:

Prisons | Changing rooms | Fitting rooms | Bathrooms | Shelters | Rape and domestic violence refuges | Gyms | Spas | Sports | Schools | Accommodations | Shortlists | Prizes | Quotas | Political groups | Clubs | Events | Festivals | Terms

2. Privacy / dignity / comfort

Women have an interest in privacy, dignity, and comfort, which requires protection from voyeurism (a risk in mixed sex spaces involving full or partial nudity) and the male gaze, and in some cases, the mere presence of male people (particular important for sexual and domestic assault survivors and some religious women). (11)

Women don’t want to be leered at, or sexually objectified, or filmed, or photographed, when they’re naked or partly naked. Many women want to be served by people of the same sex in intimate situations, like for some medical exams, or for airport body searches (the former made the news in 2017 when a woman in the UK went to the NHS for a smear test, and despite having requested a female doctor, was assigned to a biologically male nurse (who identified as a woman) (Paterson 2017)). Some trauma survivors feel in danger around men, particularly in intimate situations, and more relaxed and comfortable around women only. This fear for safety and inability to relax around men can be coded as a comfort-based reason to justify having some female-only spaces, with shelters and refuges being of particular importance. In 2018 Kristi Hanna filed a Human Rights suit against the Jean Tweed Centre, a women’s shelter in Toronto, after she was forced to share a room with a male (Chart 2018).

This gets us:

Prisons | Changing rooms | Fitting rooms | Bathrooms | Shelters | Rape and domestic violence refuges | Gyms | Spas | Sports | Schools | Accommodations | Shortlists | Prizes | Quotas | Political groups | Clubs | Events | Festivals | Terms

3. Justice/fairness

Women also have an interest in justice and fairness, justice understood as the mitigation of historical exclusion, and fairness understood as fair terms of competition. Justice arguably require categories that guarantee women’s representation in male-dominated public arenas or employment areas, for example in politics (women-only political party shortlists, such as exist in the UK), or in STEM (women-only hiring shortlists), or in recognition of overlooked accomplishments in the arts (e.g. in women’s fiction writing prizes). It may justify schools, if evidence can be provided that these result in better educational outcomes for girls. It also arguably justifies the terms necessary for women to self-refer and act politically to advance their own interests, which is why the appropriation of the terms ‘woman’, ‘female’, and ‘lesbian’ by trans rights activists have been so fiercely contested by gender critical feminists. Fairness arguably requires e.g. sporting categories in which women compete against each other, rather than against male people (give that the latter have a significant physical advantage). (12) Note that even if a case can be made for some biologically male people competing on fair terms with women—e.g. a biologically male person who took pubertyblockers and so did not go through a male puberty, and so whose body has not benefited from the historical effects of testosterone—that case will not extend to merely self-identified women who did go through a male puberty.

We end up with:

Prisons | Changing rooms | Fitting rooms | Bathrooms | Shelters | Rape and domestic violence refuges | Gyms | Spas | Sports | Schools | Accommodations | Shortlists | Prizes | Quotas | Political groups | Clubs | Events | Festivals | Terms

4. Respite

Women also have an interest in respite. By ‘respite’, I mean a break from male attitudes, expectations, and behaviours. Women and men are socialized very differently in most societies, and this leads to differences in the average behaviour of women and men, which colours their interactions. Women can experience men as both imposing and exhausting. Men take up more physical space; in conversations they interrupt more, explain things to women that women know more about, and take up more time (many of us will have firsthand experience of this from running philosophy tutorials); they may look at or speak to women in sexualized ways; they may make clear in their attitudes, comments, or behaviours that they have certain expectations about how women ought to present themselves, or speak, or act.

All of this can (obviously) be tiresome for women to deal with, and they will be forced to deal with it in all mixed-sex spaces. So women-only spaces can provide an important respite from this. (For example, many university campuses have women’s rooms for roughly this reason. If transwomen will bring some or all of these attitudes, expectations, and behaviours into these spaces, then that is a prima facie reason to exclude transwomen from those spaces). Note that this rationale does not depend on it being the case that a particular male person would in fact bring male-socialized behaviours into a space. It’s enough that women have to manage this possibility when men are around. Women-only spaces provide respite not just from certain sorts of behaviours, but from having to make contingency plans about what to do in light of certain sorts of behaviours, from being on guard.

There’s also a much more serious kind of respite, which applies to women who have experienced severe or sustained male violence (rape, assault, domestic violence, intimate partner strangulation, etc.). For some women, it’s difficult or impossible to feel fully comfortable around male people, and women-only spaces provide an important place to escape such discomfort.

This rationale justifies shelters and refuges (at the extremes; male violence), and things like campus women’s rooms, or women’s political groups, events for women such as those during university orientation weeks or as part of corporate events, festivals, dance-parties, etc.

Of our initial list, we get:

Prisons | Changing rooms | Fitting rooms | Bathrooms | Shelters | Rape and domestic violence refuges | Gyms | Spas | Sports | Schools | Accommodations | Shortlists | Prizes | Quotas | Political groups | Clubs | Events | Festivals | Terms

5. High likelihood of shared bodily experience

There is also an interest that women have in a high likelihood of shared bodily experience in particular spaces.(13) Prisons, changing rooms, bathrooms, shelters, and refuges may all involve shared facilities where there is full or partial nudity. Women are embodied in a particular way, which they have in common with each other and which they don’t have in common with men. In bathrooms in particular, women may deal with a range of embodied experiences ranging from annoying through embarrassing through distressing (for example, not having tampons when you need them; bleeding through underwear and clothing while menstruating and attempting to deal with the mess; going through a miscarriage. There’s something comforting and reassuring about being in a space with other people who are very likely to understand those experience, and have the resources to help when necessary (e.g. to give tampons, relate over mess, provide comfort for distress). Obviously not all women will be interested in providing support and not all women will need it, but for those who are and do, having spaces where there is a high likelihood of shared experience is enormously important. (14)

This gives us:

Prisons | Changing rooms | Fitting rooms | Bathrooms | Shelters | Rape and domestic violence refuges | Gyms | Spas | Sports | Schools | Accommodations | Shortlists | Prizes | Quotas | Political groups | Clubs | Events | Festivals | Terms

6. Intimate association

This rationale is borrowed from the literature on the ethics of immigration. In the debate over the right states have to exclude would-be migrants, Christopher Wellman has argued in defence of the right to exclude by appealing first to the importance of self-determination, and second to the connection between self-determination and freedom of association (Wellman 2008). If one is free to associate then one is also free to disassociate; so from self-determination we get the right to exclude. Wellman’s initial argument depended on an analogy with marriage. He argued that it wouldn’t matter how important it was to a potential suitor that he be accepted as a partner, the person faced with the option of accepting him has an absolute right to refuse. And then he argued that this is true of the state:

‘…just as an individual has a right to determine who (if anyone) he or she would like to marry, a group of fellow-citizens has a right to determine whom (if anyone) it would like to invite into its political community. And just as an individual’s freedom of association entitles one to remain single, a state’s freedom of association entitles it to exclude all foreigners from its political community’ (Wellman 2008, pp. 110-111).

He thinks the same is true of religious self-determination: ‘[i]f I elect to explore my religious nature in community with others, I have no duty to do so with anyone in particular, and I have no right to force others to allow me to join them in worship’ (p. 110).

Wellman considers an objection that would block the generalization from marriages and religious association to citizenship, namely that ‘the intimacy of marriage makes freedom of association immeasurably more important in the marital context than in the political realm. After all, in the vast majority of cases, fellow citizens will never even meet one another’ (p. 113). Perhaps we should respect freedom of association in groups where there are intimate attachments, but not otherwise. Wellman responds to this by granting that freedom of association is more important in intimate associations, but saying that it doesn’t follow that it’s unimportant in other kinds of associations. It’s also less important when it comes to religious association, but still important (ibid, p, 113).

We can either side with Wellman, and think that all kinds of associations—intimate, expressive, and political—have a right to exclude, or we can side with the objection he entertains, and think that there is a right to exclude only in intimate and expressive associations. (15)

If we do the latter, groups of women would have an absolute right to exclude only if they counted as an intimate associations. Even without taking a stand on the second two, we can agree that there is an absolute right to exclude in intimate associations, which close friendships will count as. If we do the former, then they just have to be some kind of association. Are they? The groups of women who use women-only prisons, changing rooms, bathrooms, toilets, fitting rooms, shelters, refuges, gyms, saunas, spas, or sports, don’t seem to be intimate associations—at least not in the technical sense that romantic relationships and friendships are, of course there’s intimacy in the sense of close proximity and full or partial undress—so there’s no quick path to an absolute right to exclude there. The groups of women who make up political groups, e.g. local radical feminist activist groups, might be intimate associations if the women in them are friends; or they might be expressive associations. The groups of women who attend women’s marches, and other kinds of women’s events and festivals, likely count as expressive associations. Groups of women will have common or at least loosely-aligned political goals, and will interact on a semi-regular basis.

The groups that use the remaining spaces—prisons, changing rooms, bathrooms, toilets, fitting rooms, shelters, refuges, gyms, saunas, spas, and sports—don’t seem to be expressive or political associations either, at least not in the sense that the state is a political association. There are no formal institutions coordinating these women into mutual benefit relationships. It’s not clear that women, as a class, are any association at all. Rather, they are simply a group of people who have certain interests in common, and which are served by their being grouped together. Women are physically vulnerable to men. Sex-segregated prisons protect women from sexual violence by men. Sex-segregated bathrooms, toilets, and fitting rooms protect women from sexual violence, voyeurism, and harassment by men, and provide a space where those they interact with are likely to empathise with certain body-related predicaments. Sex-segregated shelters and refuges give women safe haven from the class of people who have caused them harm in the past. And so on.

From thinking about types of association, then, we see that if it’s only intimate associations that justify exclusion, then women have an absolute right to exclude from political groups and clubs (where they are likely to have friendships) at best. If there’s an absolute right to exclude from expressive associations too, then women have an absolute right to exclude from political groups, women’s marches, and other kinds of women’s events and festivals. (I’ll assume it’s both, but readers who disagree should make the relevant adjustments).

So we get:

Prisons | Changing rooms | Fitting rooms | Bathrooms | Shelters | Rape and domestic violence refuges | Gyms | Spas | Sports | Schools | Accommodations | Shortlists | Prizes | Quotas | Political groups | Clubs | Events | Festivals | Terms

7. Self-determination

Women’s oppression involves a particular feature: what it means to be a woman has been disproportionately determined—at least historically—by men, and women are still trying to shake that off and decide for themselves what it means to be a woman. Women of the past were determined by men to be decorative objects, to be wives and homemakers, mothers, people who serviced men’s needs (think of even the working woman being expected to pour the tea). Women have a lot more liberation now, but the effects of being denied a vote, or denied the capacity to drink in a public bar, or forced to quit one’s job once married, and unable to access an abortion when wanted, and so on will have long-lasting effects, even if only through social impacts on other women. Women are particularly badly off when it comes to self-determination as a social group, because they have been other-determined, so they do not even really have the ‘self’ to get started in thinking about a form of collective-oriented self-determination. Here’s second-wave feminist Kate Millett ([1971] 1977):

“Under patriarchy the female did not herself develop the symbols by which she is described. […] The image of women as we know it is an image created by men and fashioned to suit their needs” (p. 46).

“As the history of patriarchal culture and the representations of herself within all levels of its cultural media, past and present, have a devastating effect upon her self image, she is customarily deprived of any but the most trivial sources of dignity or self-respect” (p. 55).

For this reason it’s particularly important that women, without men—the ‘other’ who did all the previous determining—decide what it means to be a woman (if anything), what needs to change about the ways women and men relate to each other, and what feminism should look like going forward. This justifies women having some spaces to be together, e.g. for consciousness-raising about feminism, for talking about shared experiences, for building feminist movements. (For the case against compelled association on grounds of freedom of thought and expression see Shiffrin 2005; for a consideration of whether women should be considered as a minority group [considering that there are legal protections in many countries protecting minority groups’ right to exclude] see Hacker 1951). Given that the content of this self-determination is likely to extend to an understanding of what it means to be a woman, a lesbian, a mother (etc.), this rationale will also extend to terms.

The justification just given was instrumental: it said, women need spaces of their own in order to self-determine. But it is also possible to rationalise the creation / maintenance of women-only spaces as an act of self-determination. Part of what men determined women to be is ‘accessible to men’, so in creating space in which to be inaccessible, women are exercising self-determination. Marilyn Frye puts it like this: “When women separate (withdraw, break out, regroup, transcend, shove aside, step outside, migrate, say no), we are simultaneously controlling access and defining. We are doubly insubordinate, since neither of these is permitted. And access and definition are fundamental ingredients in the alchemy of power, so we are doubly, and radically, insubordinate” (Frye 1978, p. 38).

We end up with:

Prisons | Changing rooms | Fitting rooms | Bathrooms | Shelters | Rape and domestic violence refuges | Gyms | Spas | Sports | Schools | Accommodations | Shortlists | Prizes | Quotas | Political groups | Clubs | Events | Festivals | Terms

Self-determination (and relatedly, freedom of association, and freedom of thought and expression) justifies exclusion directly in spaces where there is the chance to talk and raise consciousness, e.g. in political groups, feminist events, and women’s festivals. But it might also justify some, or even the rest of, the spaces indirectly too. Self-determination justifies there being ‘some’ women-only spaces rather than none; it doesn’t really matter which these are. But given that the ‘some’ we have already include those contested at the moment in the debate between gender critical feminists and trans rights activists, we might as well think that those are the spaces that matter. Unlike other dominated social groups, most women live in intimate relationships with members of the dominator social group (e.g. husbands, fathers, sons) (Beauvoir 1949). This means there are fewer opportunities for developing resistance. Once women have some spaces, whatever they are, they can make them their own, and can co-opt them for political purposes (consider e.g. feminist graffiti on the insides of bathroom stalls) (see also Scott 1992).

The indirect version of the rationale then gives us:

Prisons | Changing rooms | Fitting rooms | Bathrooms | Shelters | Rape and domestic violence refuges | Gyms | Spas | Sports | Schools | Accommodations | Shortlists | Prizes | Quotas | Political groups | Clubs | Events | Festivals | Terms

8. Intent of the creators

Finally, it’s relevant that for at least some women-only spaces, in particular women’s shelters and refuges, women established these themselves. Women funded, volunteered at, and ran these spaces as services by women for women. Some of those women are have passed on the running of such spaces to the next generation of women, but are still around to watch how their legacies are handled. The fact that other marginalised groups may benefit from access to such spaces does not give them an automatic right to be included; perhaps such groups should do the same work that women during the second wave of feminism did and establish such spaces for themselves. This rationale applies most clearly to shelters, refuges, clubs, political groups, events, and festivals (although may also apply in some cases to prizes).

So we end up with:

Prisons | Changing rooms | Fitting rooms | Bathrooms | Shelters | Rape and domestic violence refuges | Gyms | Spas | Sports | Schools | Accommodations | Shortlists | Prizes | Quotas | Political groups | Clubs | Events | Festivals | Terms

I have given eight distinct rationales for a package of women-only spaces, services and provisions. These rationales are not mutually exclusive. For example, discussion around transwomen’s inclusion in women’s bathrooms tends to centre around the question of whether the former pose a threat of (sexual) violence to women and therefore compromise women’s safety. If safety was the exclusive rationale for women-only bathrooms, and it was shown that there is no such threat, then there would seem to be no further justification for women-only bathrooms. But although they get less airtime, other rationales apply to bathrooms too, like privacy/dignity/comfort, respite, and (less obviously) self-determination (which, as discussed above, justifies there being some women-only spaces, where it’s contingent rather than necessary that bathrooms are among these). These rationales together overdetermine the moral justification of women’s retaining this package of spaces. If including some males in those spaces were not undermining of those rationales, then there would be a reason to at least start asking how the relevant interests between the two groups trade off against each other. But including transwomen—especially recently-transitioned (or transitioned in name and pronouns only) non-passing transwomen— in women-only spaces is undermining of all these rationales. If any male person willing to declare himself a woman is legally a woman and cannot be excluded, then safety is undermined (at least to the degree it would be by allowing the same numbers of non-trans men); privacy/dignity/comfort is compromised because some women will feel uncomfortable (e.g. trauma survivors in refuges); fairness is undermined because women are forced to compete with male-bodied people who have a physical advantage in sports, and because measures implemented to include more women in areas where they are underrepresented are used instead to include even more male people; likelihood of shared bodily experience is undermined because the likelihood is affected (the higher the numbers of transwomen using the space, the more the likelihood is affected); intimate association is compromised because women are compelled to associate with people they are not interested in associating with; self-determination is compromised because the ‘other’ who has been part of the story of women’s being ‘other-determined’ is given a say in what it means to be a woman; and intent of the creators is undermined because the spaces women fought for, for women, are being given to male people.

In some cases, there’s even a direct contradiction, rather than just an ‘undermining’. For example, if ‘lesbian’ means female homosexual, then it doesn’t mean male heterosexual; so including transwomen in lesbian speed-dating events, or other types of events aimed at creating intimate association between lesbians is directly antithetical to what it means to be a lesbian (this issue received public attention very recently when producers of the UK show First Dates sent a lesbian on a date with a transwoman, see The Guardian 2019b).

IV. Women-only spaces: sex or gender identity?

It is easy enough to anticipate the objection from trans rights activists, which would go something like this.

Objection. You’ve been asking whether women have the right to exclude transwomen from women-only spaces. But your stipulation that you’ll use ‘woman’ as a sex term, and ‘female’ synonymously, and your (related) choice to use ‘transwoman’ (no space) instead of ‘trans woman’ (space) begs the question. If you’d instead called them ‘trans women’, and even better if you’d referred to women as ‘non-trans women’, it would be clear that you were actually talking about two groups of women, and whether one has the right to exclude the other. And that’s actually the right way to pose the question, because trans women are women! So your question is: do non-trans women have the right to exclude trans women from women-only spaces?

This does reframe the question, and in doing so invites the rejoinder: why should one group of women have the right to exclude another group of women from spaces that belong to both of them? Just as straight women wouldn’t have the right to exclude lesbian women, or middle-class women wouldn’t have the right to exclude working-class women, neither would non-trans women have the right to exclude trans women. Women-only spaces are for women, so no women should be excluded. (Notice that if you accept transwomen are women then you get to position transwomen as subordinate to non-trans women, in virtue of being trans; whereas if you reject transwomen are women you also get to position transwomen as dominant compared to non-trans women, in virtue of being male. This surely accounts for quite a bit of the disagreement between trans rights activists and gender critical feminists, because it sends people who agree on the relevance of marginalised-status hierarchies in two different directions).

The problem with this objection is that it simply begs the question. Even if there’s some conception of gender (e.g. gender as identity) that transwomen and women share, and even if it would be best all-things-considered to use the terms ‘female’/‘women’ to refer to that gender class rather than to biological sex class, it remains true that some of the people in that class are male while others are female (in the original biological sense), and that sex class differences matter morally and politically, at least at the moment. At best, this objection simply establishes that there’s a sex class that might have an entitlement to women-only spaces, and there’s a gender [identity] class that might have an entitlement to women-only spaces, and the composition of these groups is different. Then we’d have to go through and figure out, for each of the women-only spaces in question, whether their justification is on the basis of sex, or gender identity. (Note that this also means dropping the package approach to spaces in favour of a piecemeal approach).

For all the spaces whose justification is safety, privacy/comfort/dignity, fairness, and likelihood of shared experience, the justification is sex class. Women face a risk of violence from male people. There is no evidence that this risk is from male people except for those with ‘female’/‘woman’ gender identities. Until such a time as there is such evidence, there is no reason for women to make exceptions for that sub-group of males. (Note also that it’s likely that other sub-groups are exceptions, such as gay males, but that we don’t think this means they should have access to women-only spaces). Women are subject to sexual objectification and harassment by male people. Male people—at least, people who have gone through male puberty—have a physical advantage over women when it comes to sport. Male people do not have an embodiment in common with female people, and so lack shared embodied experience. So for all these spaces, it is sex class membership, not gender [identity] class membership, that rationalizes inclusion (and justifies exclusion).

What about respite, and self-determination? These give a little more latitude for inclusion. Respite is premised on a break from male-socialized attitudes, expectations, and behaviours. Again, there’s nothing about being a transwoman that guarantees that these traits won’t be brought into women-only spaces. But if the transwoman has severe dysphoria (which may lead to her rejecting at least those behaviours over which she has conscious control), transitioned early rather than late in life, and has passed as a woman and therefore been treated as a woman for a significant period of time, then it is at least less likely that she will bring these traits into women’s spaces. (16) Similarly, self-determination is premised on women being able to decide for themselves what and how they want to be. The whole point is to be able to do this free of males, who have historically played a massive and invasive role in deciding what and how women will be. So where transwomen bring male attitudes, expectations, and behaviours into women-only spaces rationalized by self-determination, they will be particularly damaging. Again, some transwomen can be expected to do this to a greater degree than others. But it remains a problem that they are not a member of the constituency of those who have been other-determined in the way that female people have, so even if they won’t make a negative difference to women’s self-determination, it still seems to matter that they are given any say at all.

It’s not clear that it’s really possible to throw off all male socialization, so it’s not clear that it’s justifiable to include any males in such spaces. Unfortunately, the policy on exclusion or inclusion is unlikely to be able to track the nuance of which transwomen are actually likely to bring male socialized behaviours into women-only spaces. Policies will have to be trans- (women)-inclusive or trans-(women)-exclusive. So we have to make a more sophisticated tradeoff. If we were to include all transwomen in spaces rationalized by self-determination, how likely is it that women’s freedom of thought and expression in self-creating (self-determining as women) would be seriously compromised? If we were to exclude all transwomen from those spaces, how likely is it that we’d be perpetrating the harms of exclusion (even if these are only harms to the transwoman’s feelings, or the small harm of a frustrated desire) against people who would not have compromised the spaces? Because gendered socialization is pervasive and all-encompassing, I think there’s a serious risk of compromise to women’s self-determination, and so a strong reason to exclude all transwomen from these spaces. Thus, even if we make a sex / gender identity distinction and grant that transwomen might in this specific sense be women, this doesn’t settle in advance that they cannot be excluded from women-only spaces. This question is not settled merely by choices about naming. Almost all women-only spaces are justified on the basis of sex (in particular the politicization of sex differences under patriarchy), not gender identity; so almost all womenonly spaces are such that women may exclude transwomen from them. Spaces for respite, like women’s rooms on university campuses, may be the only exception to this general rule, at least if it is possible to make a more nuanced distinction between the types of transwomen who are less likely to bring male socialized behaviours into the space (e.g. those with dysphoria) and those who are more likely (e.g. those without dysphoria). In conclusion, changing the definition of legal sex in a way that affects access to singlesex services, spaces, and provisions will obviously affect women. Women have a moral right to exclude all males, including transwomen, from those spaces. Changes to the legal definition of sex would remove that right, so changes to the legal definition of sex should be resisted or rolled back.

Acknowledgements

I’m grateful to audiences at the University of Sydney, Murdoch University, the University of Western Australia, the University of Reading, and the Australasian Association of Philosophy Conference 2019; and to Luara Ferracioli, Ryan Cox, Lachlan Umbers, Nin Kirkham, Graham Brown, Sam Baron, Stephanie Collins, Colin Klein, Kim Sterelny, Kieran Oberman, Kate Phelan, and Rosa Freedman—as well as a number of anonymous/pseudonymous radical feminists—for helpful comments and discussion.

Footnotes

(1) The earliest change was Argentina in 2012, followed by Ireland and Malta in 2015, Norway in 2016, and Portugal in 2018. So there hasn’t been a lot of time, in most cases, to assess out the social and legal implications of these changes (Greaves & Hudson 2018).

(2) For a survey of law relating to legal sex, see Grenfell & Hewitt 2012.

(3) In this paper I use ‘woman’/‘women’ as a sex term, synonymous with ‘female’/‘females’, following the arguments in Byrne 2019. I leave open, for now, whether there is some conception of gender such that at least some transwomen and women might share it. Whatever this conception is, some other term will have to be used to pick it out. I also use ‘transwoman’ instead of ‘trans woman’, because I have just said that I take ‘woman’ to be synonymous with ‘female’, so the latter—which suggests that trans women are one type of female— would be misleading.

(4) For changes in both the numbers and the categories being identified into, see Fair Play For Women 2018; for discussion of the rise in young girls transitioning to live as boys—which has increased 4,400% in less than eight years—see Rayner 2018.

(5) Another way to put this, which shifts the burden of proof, is ‘do women have an obligation to include transwomen in women-only spaces?’ On this framing, the presumption is with exclusion, and we’re asking about the moral reasons for inclusion. This framing would come with the added benefit of avoiding the language of rights, which is ambiguous between legal, political, and moral claims. But because there’s such unreflective social support for the idea of ‘inclusion’ at the moment (arguably ‘diversity’ and ‘inclusion’ are buzzwords), I think that framing may make the discussion even more difficult. Thanks to Graham Brown for discussion on this point.

(6) Some countries believe that CEDAW includes transwomen as women and so include data about transwomen in their reporting. New Zealand is one such country. I have been advised by an International Human Rights Law academic that that New Zealand’s approach is idiosyncratic.

(7) All of the facts in this paragraph come from the ‘Gender Equality Milestones’ page of the Victorian Women’s Trust. Online at https://www.vwt.org.au/gender-equality-timeline-australia/ accessed 30th June 2019.

(8) I’ve included terms as conceptual spaces, for instance ‘lesbian’, ‘female’, ‘woman’. Women have an interest not only in sharing certain physical spaces with other women, but in having the capacity to label themselves together with other women (given that they share biological interests e.g. when it comes to healthcare, and political interests, e.g. when it comes to reproductive rights). If trans rights activists force a revision of the term ‘lesbian’ then female homosexuals lose this capacity, and the same point might be made about the terms ‘female’ and ‘woman’.

(9) Jane Clare Jones makes a conceptual distinction between physical spaces (e.g. toilets, changing rooms), virtual/social spaces (e.g. shortlists, participation in feminism), and conceptual spaces (e.g. the definition of ‘woman’), which I find helpful (Jones 2019).

(10) On the well-known case of transgender prisoner Karen White in the UK see e.g. Parveen (2018) & The Guardian (2019a).

(11) Privacy has recently become an issue in US high schools, with the question being whether girls have a right to 'visual privacy', i.e. not being seen naked by a person of the opposite sex. See discussion in Gerstmann 2019.

(12) It might be possible to organize some sports such that there are categories providing fair competition that are not delineated by sex. That would mean taking seriously all the average physical differences between men and women that are relevant to sporting performance (bone size and density, myonuclei, testosterone, muscle mass, height, grip strength, speed, Q-angle…) and figuring out a way to separate people into leagues on the basis of these. Thanks to Sam Baron for discussion on this point.

(13) I’m grateful to Nin Kirkham for pushing me to think about this rationale.

(14) W. Burlette Carter (2018) shows that two popular theories about how intimate spaces like toilets, locker rooms, and showers became sex-separated are false. The first theory is that it was a sign of ‘upper-class gentility and elitism’ (p. 227; see pp. 254-268), the second that ‘authorities overreacted to the notion of the intermingling of the sexes as women entered the workplace during the Industrial Revolution’ (p. 227; see pp. 268-279). Sex separation dates back to ancient times (ibid, p. 287-288), and as a public policy is ‘rooted primarily in safety and privacy concerns’ (p. 227 & pp. 287-290). Legal requirements for sex-separated facilities in the 19th century were among the first anti- sexual harassment laws (see pp. 281-287). It’s worth quoting Burlette Carter at length here: ‘[t]hose who push narratives that contort women’s history, even when they do it in the interest of aiding transgender people, oppress women. While supporting the rights of everyone to be treated fairly, women and supporters of women have the right to push back against […] these narratives. And push back they should’ (ibid, pp. 289-290).

(15) An example of an intimate association is a marriage. The parties to an intimate association generally have regular face-to-face contact and relate in ways that are deeply personal to those involved. An example of an expressive association is a church. Members of the same church have something that is important to them in common, and generally interact on a semi-regular basis. Finally, an example of a political association is a state. It’s not clear that members have anything in common (although they might), and the scale of the association is usually such as to prevent face-to-face interaction between all members. They may nonetheless be coordinated in some way, including into mutual benefit relationships (we all pay taxes, and we all receive state services).

(16) However, it may be relevant whether or not she passes as female, because if she doesn’t, then the benefits of respite that come from simply not having to think about managing male expectations and behaviours may not be generated for the women in the space.