Write for us!
Ed Davey, Leader of the Liberal Democrats, and Marie Goldman, the Liberal Democrat Spokesperson for Women and Equalities, have recently written to Bridget Phillipson protesting the publication of the EHRC’s new Code of Practice for the Equality Act 2010 and declaring it not fit for purpose. Quite how balanced the evidence base was in reaching this conclusion is unclear.
Nevertheless, their letter calls for the need for post-legislative scrutiny of the Gender Recognition Act and the Equality Act by a cross-party committee, “Taking evidence from all communities who have been impacted.” It suggests that the purpose of this is to propose amendments or new legislation to ensure “existing rights are protected.” It does seem clear at this stage, given the April 2025 Supreme Court ruling in For Women Scotland, that a root and branch legislative review is required. The law has become a confused mess, though the outcome of a review may not be quite what the Liberal Democrat leadership envisions.
One Swallow Does Not Make a Summer
Categorisation is a universal behaviour in animals. It is intrinsic to the neural networks that make up our brains and it has evolved because it has utility. Patently, the identification of prey from predator has a survival value. The hunter, staring through the forest, needs to decide whether it is a deer or a tiger through the leaves, and the sooner the better. Life depends on the processing of imperfect information in order to make dichotomous decisions. And yet, reliance on any one criterion is fraught with peril. One swallow does not make a summer, and at the same time one doesn’t need a weatherman to know which way the wind blows.
Ludwig Wittgenstein, in discussing the challenges of categorisation, introduced the concept of “family resemblance,” his example being the concept of “games.” Wittgenstein noted that the word game signified a commonality, suggesting that they all had something in common. The trouble was, identifying any one commonality was impossible. Rather, he argued, we are left with a complicated network of similarities, large and small, that overlap and intersect with one another. Wittgenstein acknowledged that we do not know the boundaries of any such family resemblance until such a time as we draw a line for a special purpose, though even then that boundary is useful only for the purpose it was drawn for.
Seventy-three years on and Wittgenstein’s point remains effectively unchallenged; not all members of a class need share the same attribute and no one attribute need be unique. An attribute can be neither necessary, nor sufficient and yet remain important for a class. Any perusal of a modern psychiatric diagnostic manual will show Wittgenstein’s insight in action.
Categorisation and gender
J.K. Rowling has suggested that a woman is such because she belongs to the sex class that produces large gametes, though she does accept that some women might not produce large gametes. The production of large gametes, by her own argument, is a sufficient but not necessary criterion. Undoubtedly Rowling will have a conceptualisation of other attributes that make a woman, but what of the person who produces large gametes, yet feels that they are male? Can that feeling be discounted?
In many online forums such feelings and attributes are bandied about as the “debates” inevitably degenerate. “No True Scotsman” fallacies come thick and fast as repeated counterexamples are presented, causing the original class to be redefined until all challenges are negated in the minds of the opposing camps. Only in this way are the arguments deemed won, and yet peace is never obtained. Rather, opponents are left shouting at walls whilst clutching their singular criterion of choice. This misses Wittgenstein’s point entirely. The families of men and women are broad churches which cannot, and should not, be reduced to single constructs.
The Negation of Sex
The UK legal position on the classification of sex, prior to the Gender Recognition Act 2004, was based on case law, specifically Corbett v Corbett (otherwise Ashley) 1970. This judgment set out five medical criteria for assessing the “sexual condition” of an individual:
- Chromosomal factors.
- Gonadal factors (i.e. presence or absence of testes or ovaries).
- Genital factors (including internal sex organs).
- Psychological factors.
- Hormonal factors or secondary sexual characteristics.
In Corbett v Corbett there were differences of opinion between the medical experts, such that factors 4 and 5 were seen to be of lesser importance by some. That is, they were neither necessary nor sufficient. Indeed 4 was held to be immaterial if 1 to 3 were congruous. This was reinforced in the case of Bellinger v Bellinger 2003, where it was held that where 1 to 3 were originally congruous, then these determined the sex, irrespective of 3 being altered by surgery and totally irrespective of 4.
The European Court of Human Rights judgment in Goodwin v The United Kingdom in 2002 took a different human-rights approach. Where 3 had been satisfied by surgery, and 5 obtained through administration of hormones effecting change, then, irrespective of 1 and 2, a person’s self-perception could be considered relevant. That is, 4 became an accepted criterion, although on its own it was insufficient.
The GRA extended this concept further, with consequences that may themselves now invite human-rights challenge. By introducing an eligibility to be recognised as the opposite sex that did not require 1 or 2 at all, or 3 (even through surgery) or 5 (through hormonal treatment), the GRA effectively reduced the categorisation of sex to 4 alone. In effect, a person’s self-perception of their gender became the only necessary and sufficient criterion, irrespective of the Act’s procedural safeguards.
Now, much as I would like to give validation to everyone’s self-perception, I have seen enough mental illness in my time as a psychiatrist to know how much that can also be an unreliable measure. It is neither necessary, nor sufficient, and it should never have been taken in isolation. And yet through the Gender Recognition Act 2004, an enormous weight has been put on something very fragile. Yes, there are other safeguards in the Act, but the recent drive towards Self-ID in Scotland was an attempt to negate even these and reduce sex as a category to one atomic consideration.
The difficulty with this approach is that by moving a single criterion to one that has hegemony over all others, the original category no longer effectively exists. That distortion is not then limited to one aspect of statute, as the last twenty-two years has demonstrated. It would be as well not to call it sex at all, because it no longer resembles the historical multifactorial conceptualisation in any material way. It has become something with a life of its own. That is a potential medical nightmare, and its impact in other areas of life, where demonstrable sex differences have always existed, could be equally detrimental.
The Negation of Reality
The Yogyakarta Principles, adopted by the International Commission of Jurists in 2007, relate to the application of international human rights law in relation to sexual orientation and gender identity. The 31st Principle of the 2017 revision of the Yogyakarta document calls for the ending of all sex or gender-based registration on official documents. By extension, this is a call for the ending of sex and even gender as constructs which can be legally defended.
There are also models of mental illness which completely reject categorisation. There is a movement to de-pathologise psychosis by arguing that the hearing of voices is a normal phenomenon. I have certainly met people who have hallucinated whilst living perfectly normal and healthy lives. However, hearing voices is often a symptom of a serious mental disorder when other symptoms of mental disorder are present. And dependent on the clustering of symptoms, the pathologies behind such presentations can be quite different; their treatments specific.
Even if you take a symptomatic approach and focus on treating voices alone, you start to find that the symptom grows into a collection of conditions of its own accord. The psychopathology gets parsed more and more. The boundaries of the presentation ebb and flow, and we are back to where we started. Nothing is reducible to simple atomic facts. Wittgenstein and his families always lurk around the corner.
A Legal Folly
Doubtless judges and lawmakers had positive reasons for considering and revising the law as they did over two decades ago, much as there were good intentions behind the prohibition movement in the US over eight decades before that. Unfortunately, when laws are ill-considered, and their likely consequences ignored, well-meaning changes can evolve in a legal vacuum in ways that were never imagined.
When the GRA was debated, the discussion was about a person changing their registered sex. At no point was a non-binary gender even considered, never mind any of the many other genders that have subsequently emerged. Further, neither the GRA, nor this cornucopia of genders, appear to have had any serious consideration when the Equality Act 2010 was debated. Ed Davey, an MP during both sets of debates, now appears to accept that these issues were not properly examined. Even after the April 2025 Supreme Court judgment, the travails over the subsequent update of the EHRC guidance noted in the joint letter demonstrate that the matter is far from settled.
Does it matter? Well, yes it does. Whenever decisions need to be made in life that require distinct outcomes, then categorisation is required to assist this process. We have to draw the lines which Wittgenstein recognised the need for, where there is a special purpose. Perhaps we need more than one line in matters related to gender and sex, as has been recognised in other countries. Wherever such lines are needed they should be well thought out and justified.
In the end, any judgments made or laws drafted need to reflect the reality and complexity of human existence. They need to accept that any groups so defined are families and not reduce them to caricatures of their original selves. And that work needs to be done before long existing rights are removed. Such laws should not be drafted at a pace, in the hubris of a large Government majority, to score political points. I agree with Davey and Goldman in their apparent call for a collaborative and deeply considered review of the legislation. However, they may be opening a door to the very scrutiny the Liberal Democrats have previously resisted. I would suggest that the outcome might not be quite what they expect. Categories matter. They are not reducible. In the end, smashing one’s teapot and expecting to make tea with the lid is just folly.
Dr Richard Hopkins is a retired general adult consultant psychiatrist and former honorary senior lecturer in psychiatry. He is a Liberal Democrat Party member.
