This is yet another look at misgendering, in which I take a rather less robust view than Naomi – referencing her post here – as to when it might be reasonable to misgender a colleague in the workplace. (I am not looking at it outside that context, because the Equality Act does not govern how people relate to one another in the course of interpersonal relationships.)
I will use the same characters from Naomi’s blog: Jen and Liz. But in my version, Jen is not transitioning – rather, having attended an Alpha course, she has become a practising Christian, while Liz is an avowed atheist. In each case, the situation arises after a casual discussion about their respective plans for the weekend. Jen has told her colleagues that she will be baptised. It leads on to a more general discussion about religion in which, having been asked directly what she thinks, Liz makes clear that she views any religious belief as “delusional,” and refers dismissively to “sky fairies,” “science-deniers” and “medieval superstition.” (In a social conversation in which she was expressly asked, she is entitled to answer.)
In this scenario, Liz does not repeat her views on religion to Jen, and Jen does not say anything more about her faith to Liz. They continue working together, albeit with some awkward silences.
This is clearly acceptable.
In this scenario, Liz makes a point of repeating her views on religion to Jen whenever they are together. Liz asks Jen whether she also believes in the tooth fairy, and demands to know why she is wearing a polycotton blouse when there is an edict in Leviticus 19 against wearing clothes made from more than one fabric. In the canteen, she highlights news stories about child abuse in the church or religious wars whenever Jen is nearby. When Jen asks her to tone it down, she points out in a way Jen finds belligerent that her lack of belief is just as much a protected characteristic as Jen’s belief, and suggests sarcastically that Jen practise turning the other cheek.
Liz is clearly harassing Jen.
Knowing that Liz is an atheist, Jen persistently tries to convert Liz despite Liz’s clear lack of interest, offering to pray with her, and giving her Bible study leaflets. After Liz’s picture is in the papers showing her attending a pro-choice march, Jen tries to interest her in materials including a “post abortion course” run by her church and focusing on forgiveness, which Liz finds particularly offensive because, unknown to Jen, Liz had an abortion aged 15 after being date-raped.
Jen is clearly harassing Liz, even if she was unaware of Liz’s personal history.
In this scenario, Liz does not target Jen in that she does not seek her out to denigrate her views and she does not repeat her own views to Jen directly. However, every time she mentions Jen’s name, she references their opposing views – with sentences such as “I’ll have to ask Jen, whose belief in a deity I do not share, if she can make 20 copies of that” or “Can we check if Paul, Amy, and Jen (whose belief in a deity I do not share) can make the meeting?”
Putting aside for a moment that this scenario is necessarily artificial, would it be harassment? She is not saying it directly to Jen. Her own (lack of) belief is indeed protected. There is no evidence that she is treating Jen badly; she just doesn’t share her beliefs. And yet I think that most people would agree that this is indeed harassment, because in every single interaction Liz has about Jen, if not to Jen, she uses a phrase that reminds her colleagues that Liz thinks that Jen’s religious beliefs are nonsense. Would it make a difference if she had used less forthright language about her own atheism at the outset? I think probably very little, if any.
I cannot see any way in which a person could ‘misgender’ a colleague behind their back, even while avoiding using any pronoun but “you / your” to their face, without falling into the same error. Every interaction about the person serves only to reiterate and reinforce the disputed issue, and to remind everyone present of the subject’s biological sex. On that basis I tentatively disagree with Naomi’s view that refusing to use someone’s preferred pronouns will “almost never” amount to harassment.
This of course raises the question as to whether requiring an employee to use preferred pronouns stumbles into the same trap. Using the same, admittedly imperfect, analogy, would it amount to requiring Liz to refer to Jen as “Jen (whose faith I share)” at all times? On balance I don’t think that it would. This is partly because employees do say – may even be required to say – things at work which they would not say outside work – such as “have a nice day” to a customer upon whom they wish nothing but leeches and misery, or “our products are the best” when their own preference is the product of a competitor.
But it is also because, socially, pronouns have a little more fluidity than the GC view tends to admit. ‘Passing’ trans people, however few in number, have been referred to in their preferred pronouns since Lili Elbe in the 1930s, through to Christine Goodwin in the 1990s and on to today. The use of feminine pronouns among effeminate gay men has been both reality and the subject of comedy, sometimes self-deprecating comedy, for decades. Are we really to believe that a neurotypical employee who could understand that “ooh, get her” might refer to a male person would still find referring to John / Jen as “her” impossible? And is it really coherent to say that using a socially feminine-coded name – a proper noun – is a reasonable request for a male person but using a socially feminine-coded pronoun – a substitute for that proper noun – is not?
While it is of little immediate help to employees, HR departments, or those with the protected characteristics of gender reassignment or gender critical belief, the historical tendency of the English language to develop along the path of least difficulty may resolve this over the next century or so. In 2121, the use of the singular “he/she” may be as archaic as the 2021 use of the singular “thou/thee,” retained only in local dialect or historical language, or as completely obsolete as the 1021 use of the dual pronoun (wit – we two, git – you two). Equally plausibly, in 2121, linguistics students may be studying as a sociolinguistic phenomenon the brief historic revival of the Chaucerian singular ‘they’ in the 2010-2030 period. Who knows?
Meanwhile, at least until there is a reasonable body of case law on the subject, I think that ‘misgendering’ a specific colleague may very arguably amount to harassment. Either Naomi or I will be wrong – or, given how case law develops, we will both be right and wrong on different points and at different times.
Legal Feminist tweeted a short thread starting like this the other day:
It seems worth elaborating briefly in a blog, so here goes.
The first point to make is that the allegation made by @MotherCecily is unverified: I don’t know who she is, or who her husband is, and I haven’t seen the email or the agenda. But it will serve anyway as an example of the kind of thing that an employer might do.
It’s an extraordinarily bad idea. Any HR director tempted to organise training with this kind of content needs to catch up with the implications of the judgment of the Employment Appeal Tribunal in Forstater. Gender critical beliefs are capable of being protected under the Equality Act: that means that someone with gender critical beliefs is entitled not to suffer discrimination on grounds of those beliefs, or harassment related to them. That protection works in the same way as protection from discrimination on grounds of other protected characteristics: sex, race, disability etc. If you want to make this real – well, run the thought experiment, substituting in groups defined by other protected characteristics for “TERF” in “Be less TERF.” It looks pretty bad, doesn’t it?
The memo doesn’t seem to have circulated very far yet. Anecdotally, it seems that large numbers of gender critical employees are suffering various kinds of discrimination and harassment at work because of these beliefs, or even being disciplined by regulators and professional associations for expressing them. A rash of employment tribunal claims following in the wake of Forstater seems inevitable.
But harassing your gender critical staff through the medium of your diversity training is taking things to another level. It has various snazzy features as compared to common-or-garden workplace harassment.
First, it’s exceptionally efficient. You don’t have to bother to harass your gender critical staff individually. Instead, with a single document or training event, you can harass all your gender critical employees at once – even including those you don’t know about (yet). Bearing in mind the prevalence of active harassment of those who express gender critical views, there may be quite a few.
Secondly, it’s likely to be pretty bullet-proof. If you try to discriminate against staff members who express their views, there may turn out to have been something in the manner in which they did so that gives you a defence. But if you harass them at large, irrespective of whether they have said anything at all, there’s no possibility of running a defence of that kind.
Finally, connoisseurs of such things will admire the irony. If employment tribunals awarded points for style, being found liable for discrimination contained in your diversity training ought to get full marks. But if you’re an HR manager who’d rather not be awarded points for style (which an employment tribunal might possibly call “aggravated damages”), you should be careful not to expose staff to training of this nature.
The example given above is an extreme case, but employers should think seriously even about what may seem to them to be innocuous exhortations to “allyship,” like encouraging staff to wear a rainbow lanyard, or give their pronouns at the start of meetings or in their email sign-off, etc. The problem, in a nutshell, with pronouns and similar observances is that they are a public profession of belief. If you “encourage” your staff to profess a belief, you are in effect forcing them either to say a creed they may not believe (and which some may find profoundly menacing; for more on that, read this powerful blog), or else to decline to say it, and thus to confess their unbelief in an environment where unbelievers may be unpopular.
The judgment of the EAT in the Forstater v CGD Europe & ors UKEAT/0105/20/JOJ is prefaced – quite unusually – with a list of things that it does not mean. There had been hyperbolic predictions from some quarters (including the Respondent’s counsel) about the dire consequences of a ruling in Ms Forstater’s favour, so the disclaimers weren’t wholly misplaced. But they were ripe for parody, and Twitter and Mumsnet didn’t shirk the task. The Guardian writer Oliver Burkeman started it: “It’s important to emphasize that the ruling does NOT give Maya Forstater the right to come round and steal your plasma screen tv,” and presently there was a long and helpful list of all the things the ruling didn’t give Maya Forstater the right to do, from tipping her seat back on a short-haul flight to Düsseldorf to wearing armour in the Houses of Parliament.
The list given by the EAT is shorter, running to only four items, and more prosaic. To summarise:
The EAT isn’t taking a position on “the transgender debate”.
The judgment doesn’t mean anyone can “misgender” trans persons with impunity.
It doesn’t mean trans persons aren’t protected from harassment and discrimination under the EqA.
It doesn’t mean employers and service providers won’t be able to provide a safe environment for trans persons.
The first item is self-evident: the EAT was not asked to give its own view on the merits of Ms Forstater’s belief, and it would have been irrelevant to its task (and very surprising) if it had done so. The third item on the list is scarcely less obvious: of course trans persons retain the protection of the EqA from discrimination and harassment, just like everyone else.
The fourth item is that the judgment doesn’t mean that employers and service providers will be unable to provide a safe environment for trans persons. This is closely related to the third, and scarcely less obvious: trans persons are no different from anyone else in that they are protected from unlawful discrimination and harassment on grounds of any protected characteristic – that is the mechanism by which employers and service providers are required to provide them with a safe environment. Harassment for the purposes of the EqA is defined as conduct that has the purpose or effect of violating a person’s dignity or creating an intimidating, hostile, degrading, humiliating or offensive environment for him. When a tribunal considers whether a claim of harassment is made out, it must take into account both the subjective perception of the person who feels harassed, and the objective question whether it is reasonable for him to feel that way; as well as “the other circumstances of the case”.
It is the second item on the list I want to take a closer look at:
This judgment does not mean that those with gender-critical beliefs can ‘misgender’ trans persons with impunity. The Claimant, like everyone else, will continue to be subject to the prohibitions on discrimination and harassment that apply to everyone else. Whether or not conduct in a given situation does amount to harassment or discrimination within the meaning of EqA will be for a tribunal to determine in a given case
There were plenty of hot takes on Twitter to the effect that the EAT had ruled that “misgendering” was unlawful harassment; or that even if it hadn’t, that it was possible to infer from the judgment that “misgendering” in the workplace would amount to unlawful harassment in almost all imaginable circumstances. I dealt with one of the latter here.
What the judgment actually says is just that it doesn’tsay anything about the circumstances in which “misgendering” will amount to harassment. The EAT sets that out at a bit more length at ¶104:
That does not mean that in the absence of such a restriction the Claimant could go about indiscriminately “misgendering” trans persons with impunity. She cannot. The Claimant is subject to same prohibitions on discrimination, victimisation and harassment under the EqA as the rest of society. Should it be found that her misgendering on a particular occasion, because of its gratuitous nature or otherwise, amounted to harassment of a trans person (or of anyone else for that matter), then she could be liable for such conduct under the EqA. The fact that the act of misgendering was a manifestation of a belief falling with s.10, EqA would not operate automatically to shield her from such liability. The Tribunal correctly acknowledged, at para 87 of the Judgment, that calling a trans woman a man “may” be unlawful harassment. However, it erred in concluding that that possibility deprived her of the right to do so in any situation.
That’s worth some unpacking.
The Claimant [cannot] go about indiscriminately “misgendering” trans persons with impunity.
That’s the bit that looks most like an assertion that “misgendering” is prohibited. But it needs to be read together with the next sentence:
The Claimant is subject to the same prohibitions on discrimination, victimisation and harassment under the EqA as the rest of society.
The first thing to note is that those prohibitions are quite limited and specific. The EqA does not place a general obligation on all of us not to discriminate against – or even victimise or harass – others on grounds of protected characteristics in our daily lives. It operates in defined spheres: the workplace; provision of goods, services and public functions; education; and associations. So if your friend asks you to use zie/zir to refer to him from now on, and you decline, you may lose your friend, but he’s not entitled to sue you under the EqA for any variety of discrimination for “misgendering” him. If a celebrity who is obviously male announces publicly that he wishes to be referred to as a woman from now on, and you write about him using grammatically correct pronouns on Facebook or on your blog or in a comment piece in a national newspaper, he doesn’t have a claim against you under the EqA either: you’re not his employer, or providing him with a service, or running an educational establishment at which he is a student or an association he belongs to or wants to join.
If your friend asks his employer to require all his colleagues to use his neo-pronouns, and it says no, that may be another matter: your friend’s employer is bound by the EqA in its dealings with him, so he could at any rate frame an intelligible claim against it. And if you work for the same employer as your friend, and you refuse to use his neo-pronouns in the workplace, you could be personally liable under the EqA if a tribunal decided that your conduct amounted to harassment.
The rest of the EAT’s ¶104 just says that “misgendering” may sometimes be harassment, but that whether or not it is in any given case will depend on the surrounding circumstances.
I want to provide some pointers to the circumstances in which I think that “misgendering” might – and might not – be regarded as harassment under the EqA. I’m going to do that by examining a series of scenarios (some of which appeared without analysis in my previous blog on the subject), and saying briefly which side of the line I think they fall, and why. But before I do that, a short observation about the word “misgender”, and the manner in which the EAT uses it in its judgment in Forstater.
Quotation marks in the EAT’s judgment
The word (including “misgendered” and “misgendering”) appears 14 times in the judgment (leaving aside its appearance in direct quotes from the employment tribunal’s judgment), in the following distribution:
“misgender” (double quotation marks): 5
‘misgender’ (single quotation marks): 2
misgender (no quotation marks): 7
There are also several occasions – notably at ¶90 – where instead of speaking of “misgendering”, the EAT refers more neutrally to a failure to use preferred pronouns.
“Misgender” means “to gender wrongly”; its use to refer to a refusal to bend the rules of grammar on the request of a trans person is tendentious, to put it mildly. I infer from the EAT’s use of quotation marks that – whether instinctively or as a matter of deliberate calculation I cannot guess – it was disinclined to accept that tendentious implication uncritically. That may be a straw in the wind as to the EAT’s future treatment of complaints about pronouns.
Is it reasonable to insist your colleagues use your preferred pronouns?
One final preliminary point. The EAT in Forstater deliberately limited what could be inferred from its judgment, preferring to leave wider questions about “misgendering” for another day. In particular, it did not express a view on how reasonable it was – or in what circumstances it might be reasonable – for an employee to demand that his colleagues use language in referring to him that is both grammatically incorrect and psychologically unnatural.
My view on this is that such a demand will rarely, if ever, be reasonable.
I want to pause here, because what I have just written may strike some as shocking or heretical. So let me say it again, with greater emphasis. I think it is an astonishing and audacious power-grab to announce your (ungrammatical) pronouns and expect others to use them. I don’t think anyone is entitled to exercise that kind of detailed control over other people’s speech, or make that kind of incursion into other people’s freedom of expression. I think it is truly amazing that we have arrived at a point where pointing this out may be widely regarded as a sign of bigotry. And yet, there is no natural limit to the extent of this power-grab, if once we accede to it. Some of the examples that follow demonstrate that.
I’m going to recycle some of the scenarios from my previous post on misgendering, as well as adding a few more. The purpose of the previous post was to demonstrate that it was too simplistic to claim that “misgendering” a colleague in the workplace would always be harassment, so in some cases I just offered them without analysis as examples of situations in which the answer wasn’t obvious. This time I’ll say what I think the answer is in each case.
In each case John/Jen (referred to as “J”) is the trans employee, and Liz (L) is his colleague. J, who is married, makes his announcement on his first day at the office – the sales department of Zeitghost plc, an IT firm – at the staff meeting at which he is introduced to his colleagues. He’s in smart-casual masculine dress that day, but he explains that from tomorrow he will be consistently wearing women’s clothing, and hopes to embark on a process of medical transition over the coming months. He wants to be known as Jen. He mentions that his marriage is still happy, and his wife is supportive. A male colleague who has always had a friendly, jokey relationship with J asks, “Does this mean you’re a lesbian?” and J says “I suppose I must be.”
L is a Quaker. She says her commitment to the truth as she understands it is central to her belief, and although she is perfectly content to use J’s new name, she is not able in conscience to use grammatically inaccurate pronouns. She says she will do her best to accommodate J by rephrasing anything she says about him to avoid using pronouns at all where she reasonably can, but she warns that this will be easier in writing than in speech. J complains that by refusing to use his preferred pronouns, L is harassing him.
L is entitled not to suffer discrimination on grounds of her Quaker beliefs. J is entitled not to suffer conduct by colleagues that has the purpose or effect of violating his dignity or creating an intimidating, hostile, degrading, humiliating or offensive environment for him. When a tribunal considers whether conduct amounts to harassment of J, it must take into account both J’s perception and whether it is reasonable for the conduct in question to have the effect of violating his dignity (etc.). L’s entitlement not to suffer discrimination on grounds of her beliefs must be relevant to the analysis of whether it is reasonable for her conduct to have that effect.
My view is that J’s expectation – that his preference to be referred to using female pronouns should trump L’s right not to be forced to use language in a way she regards as untruthful – is unreasonable. He may, subjectively, feel harassed; but I think the extent of his proposed incursion into L’s rights means that the answer to the question whether it is reasonable for him to feel harassed is considered should be an unequivocal “no”. Note, though, that although this is my confident view of the correct interpretation of the EqA, it can’t be assumed that an employment tribunal would necessarily agree. On balance, I think on these facts L would probably prevail in the end, but it could well require an appeal.
L has gender-critical views, but she doesn’t feel confident to express them openly in the workplace. She says nothing when J makes his announcement, but in the months that follow, she avoids using any pronouns to refer to him. Mostly, she manages that quite smoothly, but occasionally it makes her sound a bit stilted.
After a few months, J notices that L is avoiding using any pronouns to refer to him. He raises a grievance, saying that this shows that she doesn’t accept him as a woman. He says this has the effect of creating a degrading and humiliating environment for him.
In this scenario, it is not enough for J that L avoids using masculine pronouns for him: he is aggrieved by her refusal to use feminine pronouns.
This scenario seems to me the one most likely to arise in real life. Most people with gender-critical views will not be seeking to pick a fight with a trans-identifying colleague; but they may feel quite strongly about their own entitlement to draw a line short of active assent to a belief system which they reject. It may feel to them as if adherents to the dominant belief system in their workplace are demanding from them a humiliating gesture of submission.
I think facts similar to these are likely to give rise to bitterly fought discrimination cases in the employment tribunals in the months and years to come. I can’t offer HR departments much comfort, either: if they back L, J may sue; but if they back J, L may sue. J may have the enthusiastic support of his trade union, which L will probably lack; then again, L, lacking union support, may be driven to crowd-fund for her legal fees, adding a lot of adverse publicity to the employer’s woes. On balance, backing L is probably the more prudent course for employers, as well as being the right thing to do.
L has gender-critical views, which are well known to her colleagues. When J makes his announcement, she says “I have no wish to offend you, and I’m happy to call you Jen if that’s what you would like. But I am not prepared to refer to you using female pronouns, because I don’t want to signify assent to a belief system I don’t accept.”
My view is that L is within her rights in this scenario, too, but I don’t feel any confidence that a tribunal would agree. This, too, is the stuff of test cases.
L is on the autistic spectrum. She is confused and upset by J’s insistence that he is now a woman called Jen, and being required to use what she thinks are the wrong name and pronouns for J causes her intense distress.
I think this case is clearer. The analysis is very similar to the case where L is a Quaker. J’s demand is unreasonable, and L’s inability or refusal to use his preferred pronouns cannot reasonably be characterised as harassment. If L is disabled within the meaning of the EqA, any attempt to force her to comply with J’s demands is likely to be disability discrimination.
L is a child-abuse survivor. When she was ten, her abuser, who was in his mid-20s, groomed her by saying that he was really a teenager in his heart – he’d always been lonely as a child and just wanted another child to play with. L believed him, and at first she liked him and felt a bit sorry for him. He was obsessed with Harry Potter, just like her, and they’d played make-believe games together. L is a lesbian.
On hearing J say that he supposes he is a lesbian, L suffers a severe PTSD reaction. She goes off sick for a couple of weeks. Her fit note just says “stress”, and when she returns to work she conducts herself as in variation 2: she ducks the whole issue in J’s presence, but refers to him by grammatically accurate pronouns in his absence, and it gets back to him.
J complains of harassment, and HR calls L in for a meeting to explain herself. L breaks down in tears and explains what lay behind her reaction to J’s announcement. She says she has been horrified by make-believe games ever since being abused as a child. She says that she has no wish to upset J, and she would never describe his transition to his face as “make-believe,” but in truth that is how she experiences it. She says if the employer insists she has to refer to J using female pronouns, she will have no option but to resign.
This is a somewhat more difficult situation for HR to deal with, because although J’s demand is grossly unreasonable as applied to L, they can’t explain to J why that is so without disclosing highly sensitive confidential information about L.
My advice to Zeitghost in this situation would be that they should apologise to L, and tell J that he is at liberty to think of himself and express himself how he chooses, but he is not entitled to require his colleagues to use his preferred pronouns. If J brings an employment tribunal claim and they want to explain the full circumstances that led to their decision, they will need to ask the tribunal for an anonymity order to protect L’s privacy.
This time, J has announced that he is non-binary, and his pronouns are zie and zir.
L says she’s busy at work and in her personal life, and she has no intention of learning a load of made-up grammar in order to refer to J.
I think J’s demand is unreasonable, and L’s response – even if the grammar isn’t actually terribly complicated, and “zie” and “zir” are just to be swapped in for “he” and “him” – is forgivably short. Again, though, I am not confident that in the current climate a tribunal would necessarily agree.
This time, J has announced that he has a complex non-binary identity. He says his pronouns are are “zoi, zer, zin, zim” in the vocative, nominative, accusative and dative cases, respectively; and his possessive adjective is “zein/zoiner” in the third person and “zoir” when addressing him. He passes a short handout around explaining the grammar. (Some of his colleagues are relieved to learn that his possessive adjectives are required to agree only in number, but not also in gender, with the noun to which it refers.)
L’s response is as above.
If you didn’t agree with me on the zie/zir scenario, what about J’s more complicated demands in this one? Do you think it’s ok for zin to require zoiner colleagues to grapple with zein invented grammar? And if not, where exactly do you draw the line?
When J makes his announcement, L says that she holds gender-critical beliefs, and is not prepared to pander to his delusions. She makes a point of calling him “John,” and referring to him using male pronouns when referring to him in meetings, whether in his presence or not, and in emails to the team. She says things like “Just like a man!” any time he does anything that she regards as stereotypically male behaviour, and frequently talks of his “male privilege.”
This is what harassment looks like. L is going out of her way to cause J distress and humiliation. Her employer must put a stop to her behaviour at once.
The Protection From Harassment Act 1997
Finally, it’s worth noting that in Forstater, the EAT is referring only to harassment as a form of discrimination under the EqA. There is also an offence, and a civil wrong, of harassment under the Protection From Harassment Act 1997. No doubt “misgendering” could be performed in a manner that would give rise to liability under the PFHA. Detailed comment on what that would involve is a matter for a separate blog; for now it’s sufficient to comment that the threshold is high: the ordinary annoyances, affronts and upsets of everyday life will not cross it.
“Misgendering” is a concept that offers the employers of trans-identifying people nothing but trouble, from all sides. Pronouns are a part of language that we normally use almost entirely unconsciously and automatically. Putting them on permanent manual override imposes a cognitive cost – as is obvious from the regularity with which even committed allies stumble when trying to comply. It demands that attention be paid to something that we can normally do with no attention at all. I suggest above that the demand for ungrammatical pronouns is a power-grab, so perhaps the difficulty and the call on conscious attention is part of the point.
“Neo-pronouns” are the perfect reductio ad absurdum: if a trans-identifying male is entitled to “she/her,” why isn’t a non-binary person entitled to “they/them”? And if “they/them”, why not “zie/zir” or “xe/xem/xyr”? And if a non-binary person is entitled to neo-pronouns that substitute one-for-one for English pronouns, what possible justification could there be for saying that they can’t borrow the more complex grammar of another language – or invent their own? What rational limit could there ever be to their entitlement to hijack their colleagues’ attention with awkward and unfamiliar grammar?
Far from accepting that failure to use a trans-identifying individual’s preferred pronouns will always or normally amount to harassment, my view is that – unless done aggressively and with intent to harass – it almost never will. The very concept of “misgendering” is a menace: it should be carefully wrapped in quotation marks, and disposed of as hazardous waste.
BBC employees are being “encouraged” to put pronouns at the end of their emails and we look at the possible issues here. Is that a kindness that only a misanthrope could oppose, or is there more to it?
The first issue is that of compelled speech. Pronouns are not neutral. The move towards declaration of pronouns presupposes that everyone “has pronouns”; which is to say that everyone has an inner gender identity, and being described by the pronouns he / him, she / her, they / them, zie / zem, or something else is an expression of that identity. It also suggests that there may be repercussions for failing to remember a colleague’s preferred pronouns.
This is a highly political position. At the moment, the law recognises two sexes (male and female) through s.212 Equality Act 2010, and that a person can change their legal sex from one to the other by operation of the GRA 2004. There is also established case law which recognises that a person’s gender can be central to their private life protected by Article 8. The law does not lay down that a) everyone has a gender or b) that gender is innate.
The concept of gender identity entered the legal lexicon with the Yogyakarta Principles. These Principles do not carry legal force, but have often been adopted as a convenient shorthand. They were drafted in response to global discrimination and persecution of LGBT people. The definition given of gender identity is this:
We can see two things from this: first, that it assumes that each person does have a deeply felt internal and individual experience of gender. Secondly, that it rather correlates to the Equality Act definition of gender reassignment, envisaging a process which may include medical modifications, rather than a simple declaration.
But the more commonly used definition in the UK is that provided by Stonewall through their training. You can see that Stonewall depart from the idea of reassignment altogether (it is described as “a term of contention” in their glossary). Here are their definitions of gender and of gender identity:
What does this mean? Three things: a) that everyone has an “innate” sense of gender; b) that “culturally determined” masculinity and femininity is innate to males and females; and c) that those who reject their culturally determined gender are at odds with their sex, while those who embrace it are aligned with their sex, and are “cis.”
This is a political, and controversial, perspective. There are many people, male and female, across the political spectrum, and across sexual orientations, who regard it as problematic. It is a particular issue for those women who reject culturally determined femininity as oppressive and sexist, and for whom the idea that it is innate to most women – and by extension, that for those to whom it is not innate are not fully women – is nothing more than reinforcement of harmful stereotypes.
It is from the belief that gender is innate that the drive to announce one’s pronouns stems, because pronouns then become an expression of individual gender rather than a convenient linguistic replacement for a proper noun.
Insisting that employees put pronouns into their signature therefore leaves women who do not accept innate gender theory in a dilemma. They must either comply, aligning themselves with a political position they disagree with; or else reveal their political views in the workplace, which carries a risk of adverse consequences. We know that the popularity of innate gender theory means that those who take the contrary view may be visited with vile abuse, reported to their regulator, complained about to their employer, or even fired – so a woman who opposes innate gender theory may nevertheless feel obliged to comply through fear of losing her employment or being socially ostracised.
Some will suggest that this is acceptable – that to reject the notion of innate gender is so repugnant that those who do so must expect to face adverse consequences. They may point to EJ Tayler’s judgment in Maya Forstater’s case that gender critical beliefs did not qualify as a protected characteristic under the Equality Act 2010 for that reason. There are two answers to that. The first is that a first instance employment tribunal judgment has no weight as precedent, and this particular judgment is under appeal, and seems likely to be overturned. The second is that there is a great difference between disciplining an employee or treating them adversely because they voluntarily express opinions that they know to be controversial on the one hand, and forcing employees to sign up in public with a political statement that they may find profoundly objectionable.
A belief that gender identity is innate may also be quasi-religious; the concept that each of us has an inner being, a soul, which is gendered, contained inside the mortal flesh which has a reproductive sex that may not match that gender. As the MP Layla Moran said, “I believe that women are women…. I see someone in their soul and as a person. I do not really care whether they have a male body.”
It has long been held that the freedom to believe is matched by the freedom to disbelieve, not just for outright atheists but also ‘sceptics and the unconcerned;’ as per §31 of Kokkinakis v Greece (1994)17 EHRR 397:
“As enshrined in article 9, freedom of thought, conscience and religion is one of the foundations of a ‘democratic society’ within the meaning of the Convention. It is, in its religious dimension, one of the most vital elements that go to make up the identity of believers and their conception of life, but it is also a precious asset for atheists, agnostics, sceptics and the unconcerned. The pluralism indissociable from a democratic society, which has been dearly won over the centuries, depends on it.”
Freedom to disbelieve in the context of political, not just religious, scepticism was considered in RT (Zimbabwe) & Ors v Secretary of State for the Home Department  UKSC 38. The Court commented that “As regards the point of principle, it is the badge of a truly democratic society that individuals should be free not to hold opinions. They should not be required to hold any particular religious or political beliefs…. One of the hallmarks of totalitarian regimes is their insistence on controlling people’s thoughts as well as their behaviours.” The Appellants, who were politically indifferent, were protected as they could not be expected to assert loyalty to the Zanu-PF regime in Zimbabwe against their true views.
RT (Zimbabwe) was cited in the more recent case of Lee v Ashers Baking Company Ltd & Ors (Northern Ireland) (Rev 1)  UKSC 49 (often described as “the gay cake case.”) The bakery could not be compelled to ice a message “with which they profoundly disagreed” onto a cake. It is difficult to see how an employee could be compelled to align themselves with a perspective with which they profoundly disagree in their email signature.
The next issue is whether a female employee encouraged or compelled to declare pronouns could legitimately argue that this discriminates against her, directly or indirectly, because of her sex.
We know that sexism in the workplace is far from over. Conscious or unconscious bias operates against women. This example, from 2017, illustrates the point: when Nicole and Martin swapped email signatures, they learned that “Nicole” would be perceived as far less competent than “Martin” by clients. Without that sexism, if Nicole truly had been less competent, she would still have been regarded as such when signing off as Martin – and yet that is not what happened.
In 2019, the Royal Society of Chemistry undertook an analysis of gender bias publishing in the chemical sciences. It recognised that biases were “subtle” and could be “inadvertent.” Women were invited to review less often, their work was more harshly received, their initial submissions more frequently rejected. These “small biases” led to a “significant cumulative effect.”
The RSC are not the only organisation to have done such research. Others have found similar results, and of course there are numerousarticlesspanning the last decade or more which find that CVs with a female name get poorer results than the same CV carrying a male name. CV writing services recommend against including gender on the CV – a practice which used to be common and is now recognised as archaic. Race is also a factor – although for now at least, nobody is suggesting we declare our race at the bottom of email signatures.
And it is not just the recipient of the email who may be unconsciously biased against a female sender. The female sender herself may be subject to ‘stereotype threat.’ This is where a person is reminded of membership of their group and then under-performs; for example, women who were told that women do worse than men in maths tests then really did perform significantly worse in a maths test than women who were told there was no difference in performance (Cordelia Fine, Delusions of Gender, p.32-33). Even being reminded of one’s own sex at the beginning of a test can have the same effect (ibid, p32).
It would seem that women who are compelled to declare female pronouns in their signatures may be vulnerable to stereotype threat and also to unconscious bias on the part of the recipient of the email, thereby entrenching those biases further.
This does not mean that any woman whose workplace initiates a pronoun policy has an automatic, unassailable, claim. An employer might defend the claim by arguing that they are aware of the negative impact on women but that it is justified as a proportionate means of achieving its legitimate aim of trying to create an inclusive work environment. Argument and evidence would then centre on balancing the potential harms and benefits of the policy. It would be relevant, for example, if the employer was already struggling with recruitment and retention of women, or if there were a male / female disparity in sales commission. The ‘compelled speech’ aspect of the policy would also be relevant to this balancing exercise.
It should go without saying that if an employee is beginning gender reassignment and wishes their colleagues to use a different pronoun for them, they should be supported to do so. There are rightly prohibitions on victimisation and harassment on the grounds of gender reassignment. That person may wish to send an email round-robin with their news, or may wish to have their pronouns in their email; an employer should not prevent that. However, when the BBC’s guidance suggested that all employees should put pronouns in their signature, and said “It’s really simple,” that was, we suggest, premature, and may be experienced as coercive. And when they speak of “creating a culture where everyone feels comfortable introducing themselves with pronouns” they should also consider whether they might be inadvertently creating a culture where those with the she/her pronouns experience discrimination as a result of their sex.