#144 Alice Sullivan: Why We Need Accurate Data on Sex in Science
Alice Sullivan
Professor Alice Sullivan joins FiLiA volunteer Sara on the FiLiA podcast to explain why we need accurate data on sex in scientific research (and how it is especially important for women and girls), what happened regarding the sex question on the UK Census in 2021, institutional capture, and the chilling effect in universities on this topic.
Alice Sullivan is a Professor of Sociology and Head of Research at the UCL Social Research Institute and was director of the 1970 British Cohort study. She is outspoken on the importance of sex as a variable in data collection, and concerns over academic freedom on the issue of tensions between sex and gender identity.
Follow Alice on Twitter: @ProfAliceS
Alice’s website is: profalices.co.uk
Listen here:
Transcript:
Sara from FiLiA in conversation with Alice Sullivan
A – I’m a Professor of Sociology at UCL, my work has been on things like the sociology of education, and social stratification. I’m interested in educational and social inequalities and in the last 10 years, 2010 – 2020 I was Director of the 1970 birth cohort studies, a major longitudinal study so I have a lot of experience with things like survey design and data analysis.
S – You will be known for being out spoken particularly about sex as a feminist issue and also from a data integrity perspective. Would you give us your definitions of sex, gender and gender identity?
A – Humans are mammals and we have 2 sexes just like all the other mammal and these are defined by our reproductive classes. There are 2 reproductive classes that are needed to reproduce, to make a new human being and those are male and female. You are a member one of those reproductive classes even if you are too young or too old or otherwise not fertile, you are still a member of one of those classes.
At this point people often bring up intersex, so we’ve all had to become experts on intersex conditions. There is a small minority of people who have these differences of sexual development. The vast majority of those people are readily classified as male or female. There’s a tiny proportion of births each year, around 0.02% where that’s not completely straight forward.
To me as a sociologist it’s amazing that we say there are a tiny number of cases where the categories are problematic so we should abandon the categories. If we did that we would have no categories at all. If we were to apply that to social class or ethnic groups, we would throw up our hands and say we can’t talk about these things at all.
Sex are the cleanest categories we have in the whole of social science and yet these are the ones being challenged. To me that’s fascinating.
Sociologists and feminists and people like Ann Oakley have always talked about gender as a set of expectations, social roles and social norms that is applied to people according to their sex. So it’s not the same thing as your sex but you can’t opt in and opt out. If you could just say – I’d rather not have female gender norms applied to me – patriarchy would be over right away.
Gender is a social category about how other people see you and treat you. Gender identity is an individual psychological category. It’s much less well developed and talked about essentially because there’s been so much silencing. It is a more recent category but also people are not able to have open conversations they need to have to work out what we mean by it.
Having said that, it’s clear that some people feel real discomfort around their biological natal sex. They would say their gender identity is not just different from their sex but it clashes with their sex.
So those are the 3 concepts. People talk about mixing up sex and gender. The bigger problem is people are mixing up gender and gender identity.
S – They seem to have these differences between gender expression, gender identity, gender and maybe something else. I’ll look at the unicorn or the genderbread man with different things about gender. One of the things I wonder about in terms of the eye of the beholder, the perception of someone, you would still get in trouble for mis-gendering someone –
A – And it’s really difficult. It really slows you down in terms of thinking and expressing yourself, maybe that’s that is part of the point of it.
S – I think it’s got to the point where I’ve heard of people saying ‘we can’t describe what we’re seeing in this photograph. We can’t call these people in this staged picture, a man or a woman because we don’t know how they identify’. That’s taking it to an extreme absurd level. In a strange way it logically does make sense.
A – That’s part of the political programme, you can never talk about sex because gender identity is more important.
S – Why is it so important as a feminist issue? Why is it important to count women?
A – If you don’t have data you cannot make assertions about the world. To talk about sexism and sex discrimination when you don’t have pay data is very difficult and then to take the next step as a researcher, to understand, why are women paid less than men? To what extent is it explained by child bearing and to what extent is there a gap even if you take that into account? These kinds of things. The standard research questions that people might have. If you don’t have the data you cannot analyse those things.
Sex is a predictor of every social outcome, whether it’s education, the labour market, crime, sport participation, cultural participation, mental health, physical health. It is really hard to think of anything where it doesn’t matter.
The danger if you don’t have data on sex is that you default to the male as a norm idea. This is something Caroline Criado Perez has documented really well. People start to think well this is okay for men so that will do.
Having data only on men is one thing but having data that is gender neutral with males and females and you don’t know which is which is hugely problematic because they will have different experiences obviously and not just that but if you care about intersectionality there will be interactions between your sex and your ethnic group, your age etc. How does your age affect how you’re discriminated against? Is it going to be the same for men and women? Are black men and women or Muslim men and women treated the same? Obviously not.
So the idea that we don’t need data on sex, to me it’s so obvious that we do, I find it really hard to understand why people don’t get it.
S – Why do you think this happened? The census for example has for 200 years asked relatively straightforwardly what is your sex? Why has that now become such a difficult question to ask?
A – That’s a really good question. How have people who you would think are serious minded, empirically minded at the ONS (Office for National Statistics), how have they been captured by this? They have been sold a post-modernist ideology which seems incredible. How can people who deal with statistics buy in to this idea that sex isn’t real.
I’ve had some extraordinary conversations with ONS where they denied that people in the past knew un-problematically what sex they were. They’ve said things like – we can’t know that – and you’re just thinking – what planet are we on? It’s so bizarre.
I think how it has happened has been well documented by Mary Blackburn McKenzie particularly in the Scottish case and the same has happened all over the country and other places in the world. It’s called policy capture. Essentially you see lobby groups going into the public sector and branches of the civil service which includes ONS and telling them that this is what they have to do. They are paying organisations like Stonewall to tell them what to think.
It’s deeply disturbing in a democracy that Stonewall have been able to do this. Not just the data collection but for the civil service more generally. The British civil service has always prided itself on impartiality and you would think they would see the huge danger of allowing lobbyists to have this huge say to the point where they actually ignore the views of experts.
S – I know you wrote a letter to the ONS with 80 signatures from social scientists saying we need sex data and this is important. Also feminist activists saying this is important.
Why do you think the views of the experts are being ignored? How is it that the scale has been tipped so in favour of one set of views?
A – I find it quite extraordinary. I can give you a partial answer in terms of some of the FOI requests that have been done about the relationship between Stonewall and ONS which is clearly very cosy. ONS have managed to get themselves in a situation where they think that the safe thing to do is to do what Stonewall tells them. Another part of is that my mind just boggles. I can’t really understand how it has happened.
It’s very good to see that the new organisations like Sex Matters are calling for a public inquiry into how Stonewall, while lobbying for change in the law also manage to enact policy change as though the legal change they are lobbying for had happened.
The legal change has not happened. We now know we’re not going to get self-ID; the Government has ruled that out. So why are public bodies acting as though the law has changed when it hasn’t. We really need a proper inquiry into that because I have partial answers as to how it has happened. It is actually mind boggling and fascinating from a sociological point of view. I would like to understand how it’s happened. It’s an extraordinary achievement for an organisation. It would be interesting to know how they have done that and how public bodies have been so vulnerable to that. If Stonewall can do it then why not other organisations too.
It’s quite frightening to think that level of policy capture by lobbyists is possible in a democracy.
S – I wonder about other policy capture that could be harder to spot that would affect women and girls that we don’t even know about yet.
A – In the academic freedom case, if universities won’t stand up for academic freedom on something so simple, how are we going to defend the hard cases where something is genuinely offensive?
I think there are so many people who aren’t saying anything about this. We have to stop and think – hang on a minute, if this can happen then other things can happen that will affect me.
It worries me in particular that people on the left are not speaking up about the issues around academic freedom and freedom of expression and are playing into these culture wars narrative that it’s a right wing hoax or something. They are the ones who ultimately, if our freedom of expression is eroded, people on the left are going to be affected probably more than people on the right.
It’s very disturbing. I think we get culture wars because people are willing to play their allotted roles in those cultural wars and I wish that a lot of people who aren’t really thinking about this would just engage their brain.
S – It seems that the problem is worse in academia than in other places.
A – I think a lot depends on discipline. I feel quite fortunate to be a quantitative social scientist where obviously a lot of people have stood up to be counted and signed their names and that’s really helpful as I know it’s not just me, a voice crying out into the wilderness. I know that the majority of my peers quietly agree with me.
I think it’s very different if you’re in philosophy where things are really toxic for women. It’s really interesting how those unpleasant cultures can develop in particular subject areas.
Academia should be the place where we can have these difficult conversations and open discussions. What is the point of disciplines where that becomes impossible, it’s more like a religion and you’re not allowed to challenge?
S – Can we talk about the latest census and what happened concerning the sex question which was challenged in court by Fair Play for Women. It wasn’t about the question itself but around the guidance to the question. Could you talk us through that and the difference between legal sex and biological sex?
A – What happened was in 2011, the last census, the ONS brought in guidance for the sex question on the instigation of gender identity activists, guidance that said – basically you can answer it how you like according to your identity – that was done on the hoof and they didn’t consult anybody and most people had no idea they had done it. The 2011 census was filled in by the vast majority on paper and the guidance was only available on-line so it really went under the radar and probably wasn’t important at all. The reason the activists wanted it was because it set a precedent to say ‘this is a self-ID question’.
For the 2021 census the discussion went through a number of phases and at one point the ONS were suggesting it shouldn’t be a binary sex question. They were rapidly put on the right track on that point. But the question about the guidance did not go away. Again suggesting self-ID guidance.
I was alerted to this in late 2019 and organised the letter from quantitative social scientists including lots of major figures who run or have run major surveys, a really serious list of people.
I thought it would be more difficult than it was to get people to sign up for it because of all the fear around the subject. I thought, well they’ve got to listen to this group of people, we’ve got this, it’s in the bag.
That wasn’t what happened.
Another thing that happened that was interesting: because I was involved in speaking about this issue, I got de-platformed from a research methods seminar which was organised jointly by National Centre for Social Research who had been working with ONS, and City University. That was a seminar where there would have been a couple of people from ONS speaking. They chose to cancel the event rather than have me to speak on the grounds it would have made people feel unsafe, not just feel unsafe, literally would have made them unsafe to hear that sex is an important variable.
I guess I had been a little bit naïve, okay there are big chunks of academia that are tinged with post modernism but not my field. People who do quantitative science, we don’t read Judith Butler, we just ignore all that nonsense, we don’t have anything to do with it.
It started to dawn on me that the Butlerians and the post modernists had started to land grab. We used to ignore them and they used to ignore us but they’ve stopped ignoring us, they’ve come for our data so we can’t ignore them anymore, we have to stand up for it.
I tried to get the ONS to listen to the experts. Initially I was told by ONS that there would be decisions made in summer last year, 2020. When I first got involved I thought it was very urgent, there was this huge expensive exercise going on. They kind of went on a go-slow. Organised a round table in June which they invited 6 of us who had signed the letter. We were balanced off with 6 academics who were clearly chosen for their particular political affiliations. Completely random disciplines. One was from the Theology department, clearly people who had never used data. Also the usual lobby groups, Stonewall et al.
So that was a really strange event. It really felt like an exercise in window dressing. They know they have to be seen to consult but they don’t actually want to hear what we have to say. I didn’t get the impression they thought – oh fantastic an A-team of really highly esteemed experts on data who want to give us their views, what a fabulous opportunity – that was not their attitude.
So it dragged on and on. It became clear eventually that they were running down the clock and the decision was made at the 11th hour when they probably thought it was too late for legal action.
I was never thinking in terms of legal action. This shouldn’t be a legal decision, it’s a scientific decision. When they finally made their decision, Nic Williams and Fair Play for Women jumped in and did the extraordinary feat of crowd funding £100,000 and taking successful legal action in an incredible compressed amount of time. They (ONS) probably thought they had made it so it was impossible. The decision was that because there is legislation around the census and the compulsory questions they can ask; they cannot redefine the meaning of those questions. For example, they can’t say we’re asking for your sex but what we mean is your income. That would be wrong. Equally, they can’t redefine sex to mean identity because that’s not what it means.
They tried to argue in court that sex is an umbrella term that contains 5 concepts, I can’t remember what the 5 of them were. The judge was unimpressed by that. That was quite nice. There is something to be said for legal actions. You can’t just come out with complete nonsense and expect it to be taken seriously.
The judgement is very specific to the census because sex means legal sex because there’s legislation around the questions so it’s reasonable to say that sex is sex legally defined as opposed to anything else.
The difference between legal sex and biological sex. Until 2004 with (GRA) gender recognition legislation that brought in Gender Recognition Certificates, legal sex and biological sex were the same thing. The GRA legislation uncoupled sex from legal sex and the consequences for data collection were entirely unintended. Hansard records – we’re talking about a tiny group of people here and they just want to get married – this was before same sex marriage legislation.
It’s interesting to read David Lammy and others saying ‘of course marriage is an opposite sex institution and will remain so and this is why we need to do this’. Now we have equal marriage and rightly so.
But we’re living with all sorts of unintended consequences. How we unpack that is really difficult. It’s important that we recognise the census as one form of data collection and there’s lots of other data collection going on. People have got it into their heads that they cannot collect data on biological sex. We need a clear signal from the Government that that is not true. Not just that you can but you should because it’s an important variable.
S - The other thing is you’re not allowed to ask who has a Gender Recognition Certificate and who doesn’t. I heard others say they can’t ask what sex because it would be asking to divulge private information and I might get into legal trouble if I make someone say what they’re biological sex is.
A – That was never the intention and if you look at the vast majority of data collection exercises, they are anonymised anyway. Legislation that was brought in to stop people being doxxed and exposed. The idea that you can’t fill in a form that’s completely anonymous – that was never the intention. It’s been used quite deliberately to try to make it impossible to treat sex as a variable. This is part of the bigger political project to erase the category of sex in law, in data and in language. You can’t talk about sex, you can’t collect data on sex and we can’t treat sex as a protected characteristic and provide single sex services.
S – What would you say to people who say it’s a small minority of people who may identify as something other than their biological sex?
A – It’s the main argument people come out with. It’s a fast moving area, we don’t really know what the numbers are particularly among young people. In the past you had tiny numbers of males, almost exclusively males who not only identify with but transition to being seen as women. In the past we would have talked about transsexuals. Now we talk about trans people and Stonewall has this umbrella which is a much bigger group of people, cross dressers for example. How many cross dressers are there? There are studies that say that’s quite a common thing for men to do. More importantly, among the younger age groups we see a new phenomenon of lots of girls identifying as trans or non-binary. I say lots, we don’t really know how many. It’s increasing at the moment. We need good data on that. That’s one of the reasons we need good data on sex because if we only know whether people are identifying as trans or not, if we don’t know what sex they are then we cannot monitor those kinds of trends.
The other thing that people find difficult and counter intuitive is that small numbers can make a big difference. The reason that small numbers can make a big difference is because we’re not talking about estimating the number of men or women in the population as a whole which we know is roughly 50 50. What we’re interested in is different sub groups of the population so for example if you look at crime figures, particularly violent crime which is almost exclusively male, a tiny proportion of women commit these types of crimes. If you misclassify a proportion of the males who do violent crimes as female, then what it looks like you see is a dramatic increase in females committing sexual assaults or assaults on children for example.
There was a news story recently claiming exactly that that there had been a big rise in women doing these kinds of offences and actually there was no way of finding out. Is that a real social phenomenon? In which case, very interesting and we need to find out why that’s happening. Or is it to do with misclassification of the data?
Without accurate data you cannot test these hypotheses.
We get a lot of claims for example: Transwomen behave just like women in terms of things like crime and other area of social life. The desire not to collect data on sex I think is partly motivated by wanting to make it impossible for those hypotheses to be tested. If we have the data, we can find out. Do men who identify as women behave more like women or more like men.
There’s no reason why the answer to that should be uniform across different areas of life like crime, wages, educational specialisms etc. There’s all sorts of areas where you could look at that and you might get different answers. There’s no reason why it should be the same for transmen and transwomen. You might find that transmen in certain areas are more in line with male norms, say hypothetically, the employment sectors, but males who identify as women are even more male in those terms. Anecdotally we know there’s a lot of transwomen working in IT and even the military which would be really or relatively unusual for women to be working in.
These are all open questions. No judgement. It’s absolutely fine for transwomen to be working in IT. What you don’t want is for it to be unable to track women’s progress in IT or the gender wage gap because you don’t know who is male and who is female.
S – The absence of evidence is being taken as evidence of absence.
A – Exactly, that’s been something that’s been really flagged up by the judicial reviews, that judges have been turning around and saying. - You haven’t got the data on this? - The Tavistock case for example. There were several points where the judge said – excuse me you’re not collecting data on eg, the young people co-presenting with autism or Asperger’s? – just basic things.
If the gender identity lobby really wanted make trans people’s lives better, they would want the data, they would want to understand. You can’t make things better for people without understanding what’s going on.
What has become quite obvious is they don’t want to make things materially better for trans people, what they want to do is affirm a particular ideology and that means making it impossible to collect the data, making it impossible to test particular claims about the important of sex.
S – I’ve never understood how it’s more inclusive to not be accurate.
A – I’ve never heard that claim made about any other variable. Some people might be mortified about being on a low income but you don’t say – oh say the income you’d like to have – it’s not an argument anyone makes in anything else.
Yes, so inclusiveness is just one of those buzz words that seems to disengage people’s brains. They say it’s inclusive and they don’t actually have to unpack that and say – what am I actually saying? What benefit is there in asking people to be inaccurate?
S – How do we move the conversation forward and destigmatise the issue of asking for data on sex. Do you think the ONS ruling will do a lot for that?
A – I think the ONS ruling with make a big difference in terms of making organisations see the risks differently.
We need a post mortem; we need to understand exactly what’s happened. We know the EHRC has been giving dodgy legal advice by telling people they can’t collect data on sex.
I was told a whole lot of contradictory things by the ONS about what their legal thinking was. They were very cagey about it and not willing to share that legal advice. We need to understand – were they given bad legal advice? Or did they ignore their legal advice? What actually happened here?
Other organisations will sit up and take notice and say, ONS did exactly what Stonewall told them to do and it cost them a lot of money and reputational damage. I think it’s a really important line in the sand. Clearly it’s one battle and the war is not over and we need to keep fighting for sex based data collection across the board, whether it’s in major data collection exercises or in your own institution and monitoring and so on. All the little surveys that go around that we fill in. To continue to speak up and say - look sex is a protected characteristic, you can’t just change the meaning of it. You could also collect data on gender re-assignment or another protected characteristic or you might be interested in people’s gender identities in certain contexts and then ask about that.
We’re not telling you not to collect data on anything about trans people. Why can’t we also say we would like to collect data on sex. It’s such an extreme position. I do think the strategy of shining a light on it ultimately will work.
S – Do you have any advice for those who want to shine a light on it but are fearful because of the documented consequences which you have experienced? Do you have actions that our listeners/readers could take?
A – I think for people who are fearful of speaking out, I would say reach out to other people, it’s been really good to see more organisations springing up in universities and particularly student groups like the Cambridge Radical Feminist Network and something similar in Oxford. No-one can do this alone. At UCL we have a UCL Women’s Liberation group. It means that we are visible. For people who are absolutely terrified, get in touch with us sometimes completely anonymously.
At least I am there so people like that can talk to me and they have a place to go. I think that’s really important for networks to get together with other people, women and men who see the nonsense and agree with you and just keep talking about it.
There are fabulous organisations out there. Fair Play for Women who played such a pivotal role in the court case. Woman’s Place UK and MurrayBlackburnMackenzie’s policy analysis had been hugely important in this particular issue of sex based data collection. Follow those groups, reach out to them, volunteer, donate.
I would also say to those people – sometimes you feel like you’re going crazy but you’re not crazy. They are the crazy ones, they’re trying to convince you that 2+2=5 and you know it’s not true. Ultimately in a few years’ time they will look on it and say – this is nuts – and probably most of the people who went along with it will pretend that it never happened.
We are seeing that denialism even as its happening, people saying – oh, no-one has ever said sex isn’t a real thing – I have spoken with people at ONS who are telling me that sex is terribly complicated and difficult to understand. How do we know that people in 1801 knew what sex was –
We are being gas lit on an epic scale but we will overcome.
To contact me:
Follow Alice on Twitter: @ProfAliceS
Alice’s website is: profalices.co.uk