Sex, lies and the census

StatsNZ, the government’s official statistics agency, apparently thinks that humans can change sex. It tells us in its report on the new questions in the 2023 census that:

Sex is based on a person’s sex characteristics, such as their chromosomes, hormones, and reproductive organs. While typically based upon the sex characteristics observed and recorded at birth or infancy, a person’s sex can change over the course of their lifetime and may differ from their sex recorded at birth.

The glaring error in this statement is of course the claim that a person’s sex can change. We all know humans cannot change sex. Our grandmothers knew it, and their grandmothers before them. I suspect the earliest humans knew this. A standard child development textbook will tell you that human children grasp the immutability of sex roughly by the time they reach primary school (page 535 of the eighth edition, if you are curious). If you are wondering whether ‘the science’ has changed, it has not. You don’t need to take my word for it. Here’s Professor Robert Winston, a specialist in human reproduction with an illustrious career: ‘I will say this categorically that you cannot change your sex.’

The other error by StatsNZ is to reduce sex to ‘sex characteristics.’ Not only is this circular, it opens the door to a misperception that chromosomes simply are sex: XX for female and XY for male. Taking this view would imply that people with nontypical chromosomes – XXX females for example, or XXY males – are some kind of ‘intersex’ or sex other than male or female. This is what StatsNZ gets close to implying, with its new question on ‘variations of sex characteristics.’ In fact, it is more accurate to say that, in humans, specific genes drive sex differentiation into one of two genetic developmental pathways that produce male or female reproductive systems. In particular, the presence of an SRY gene sets an embryo down a male developmental pathway. SRY stands for Sex-determining Region Y; this gene is almost always on the Y chromosome. So sex is a matter of what kind reproductive system an individual has. 

It follows that people with nontypical chromosomes are all either male or female. A man with Klinefelter syndrome has XXY chromosomes. He is most definitely male. Similarly, a girl born with only one X chromosome has Turner syndrome. As Britain’s NHS notes, it is a female-only genetic disorder. Women with Turner syndrome will in almost all cases be infertile and most will need particular healthcare in order to live a healthy life. It is insulting and cruel, as well as inaccurate, to imply that people with such disorders or variations of sexual differentiation (often abbreviated as DSDs) are something other than male or female. 

So, our government’s statistics agency is either lying to us or is inexcusably ignorant. Sure, most people probably get through life without knowing the details of DSDs or even the precise relationship between chromosomes and sex. But we all know that humans cannot change sex. And a government agency that has gone to great lengths to include questions on sex, gender identity and DSDs in its census questionnaire ­– but makes such elementary mistakes – is either grossly incompetent or worse. 

Do the lies matter? Lies and damned statistics

I am going to call obvious false claims ‘lies’, even though probably nobody at StatsNZ consciously intends to deceive. Humans find ways to manage the cognitive dissonance that occurs when there is gap between what they know to be true and what they are actually saying – we can predict they will sustain illusions in order to avoid the negative feelings that come with consciously lying. But right now I am not so interested in how the officials at StatsNZ live with themselves, but rather with what happens when authorities make obviously false statements.

A lying government statistics agency is a problem. It matters, first, for the accuracy and usefulness of the statistics collected and disseminated. StatsNZ is introducing these new census questions as part of its ‘gender first’ reporting policy. This policy means that data on ‘males’ and ‘females’ will ordinarily be based on a person’s subjective gender identity rather than his or her sex. This is more than an irritation for people who don’t have a gender identity – who don’t particularly feel an affinity for gender stereotypes and regard themselves simply as being either male or female, regardless of dress, habits or feelings. I don’t have a gender identity any more than I have a species identity. I am human, regardless of how I think or feel about it. 

The real problem, however, is not that people like me are irritated. The real problem is the loss of integrity in the census data and all the other official sources of data that either use census data or adopt the StatsNZ guidance on how to gather data. That includes data on male-female income gaps, educational achievement and any other type of data you can think of where there is a legitimate reason to report results by sex. As argued by Professor Alice Sullivan, a leading social scientist, conflating sex and gender in official statistics is a bad thing to do. Neglecting to gather sex-specific data, already a problem, is particularly harmful to women.

To be sure, the absolute numbers of those reporting a gender that differs from their sex will be small. In the UK’s most recent census, only 0.2% of the those responding to the gender identity question reported a ‘trans man’ or ‘trans woman’ identity, and a minuscule 0.06% identified as non-binary. But in areas where sex ratios are hugely imbalanced – for example, male-female differences in the prison population or patterns of sexual offending – even small numbers will substantially distort official figures. Even when the numbers are small, deliberately introducing a policy that defines basic categories in misleading ways is still wrong.

StatsNZ might claim that, by asking about sex and gender, it is not conflating the two. But it is. First, by giving an inaccurate and misleading definition of sex, as something that can change over a person’s lifetime. Second, a gender-by-default reporting policy means that data relating to sex will end up mixed in the reported data on gender. Questions answered on basis of sex are going to be reported, in most cases, as referring to gender. StatsNZ will do this through matching census data with administrative data and by imputation, although they are still officially consulting on exactly how they will do this.

When the state lies: the erosion of trust

We lose trust in agencies that lie. We know they are lying and we also know that they must, at some level of consciousness, know they are lying. And the only possible reason for this insistence on lying is a decision to put political expedience or ideology ahead of the truth. 

When public agencies chose to lie in this way, they invite something worse than ridicule: profound mistrust. Once a public authority is known to lie out of expedience or pressures for conformity, all of its claims potentially come under suspicion. Why believe a government agency or a public scientist about climate records, if they can’t even get sex in humans right? Why believe the health ministry about the safety of vaccines, if the same agency claims that men can get pregnant? 

Truth matters. The world’s climate either is or is not changing as a result of rising concentrations of greenhouse gases in the atmosphere. Either vaccines save lives or they don’t. The sum of our knowledge on both of these things remains incomplete, of course, and will be added to and revised over time. And no doubt there are many complexities: room for caveats and nuances. But unless you are willing to abandon the idea of truth entirely, not all claims about climate change and vaccines can be equally true.

And for those who care about truth, trust matters. I think there is overwhelming evidence to conclude that the climate is changing as a result of human activity, but what I believe on this issue is entirely dependent on my trust in public authorities and the credibility of scientists. I am not any kind of atmospheric scientist, so I rely on these people to tell the truth. Just as I rely on the work of research scientists and statisticians in order to form a view on the safety and efficacy of vaccines. 

Credible, trustworthy sources of information and analysis are vital for democracy and good public policy. We’ve seen much handwringing about misinformation and disinformation, about how some people are disastrously ready to believe conspiracy theories and junk science. About how false and misleading information can be put to work to undermine attempts to solve real problems, from public health to climate change. About how extreme and deliberate lies can threaten democratic institutions.

There is no easy fix for these problems. Censoring misinformation is unlikely to work and will often in fact undermine trust, or even play into the hands of those responsible for deliberate disinformation. 

Public agencies could at least avoid making things worse. They could stop lying. 

Cover image: photo by Shot by Cerqueira on Unsplash