Are you normal? What even is normal?

Gus

Welcome to The Tech Pill, a podcast that looks at how technology is reshaping our lives every day and exploring the different ways that governments and companies use tech to increase their power. My name is Gus Hosein and I'm the Executive Director at Privacy International.

Caitlin

And I'm Caitlin and I'm PI's Campaigns Coordinator. Hi. This week we're talking about what it means to be normal and how that defines and sets the boundaries on the way the technologies and stuff we interact with are developed.

Gus

So Caitlin has wanted this edition for a while, so I'm terribly excited that we finally get to it. As you'll hear, we are going to find out that we nearly lost Caitlin to a PhD, but I'm grateful that we did not lose her. But what we're essentially doing is we're exploring her journey through normal and how normal intertwines with power and gets embedded into our technologies and policies.

So it's very on brand for this podcast. Only a topic like this lets Caitlin help us explore the theories of Greek philosophers, mathematicians, urine samples, and of course, eugenicists. And you'll hear me struggle with how to behave on public transport, at airports, with body mass index, and my bitter, bitter failure at being a statistician.

But if you really want to know what this is about, it's our rage about how abnormal behaviour arises in public transport, notably London's underground rail, the infamous Tube. Without further ado, let's get going.

Caitlin

Part of the reason we're doing this, right, is because I suggested two years ago, and part of the reason I suggested it is my undergrad dissertation was on panopticons and surveillance and the panopticon relationship and whether the modern internet counts as a panopticon or whether it doesn't.

Gus

Damn. Oh, so this job is actually cool for you. Like this is something you might have actually wanted to do for a career.

Caitlin

Oh, yeah. By the way, the modern internet is not a panopticon. It's very annoying when people describe it as such.

It might be for specific people, but as a general rule, no, it is not. My postgraduate dissertation was on the life cycle of norms, using the example of women shaving their underarms as a reference point. And so this is a topic, both normality and norms, was the beginnings of an aborted application for a PhD.

Back when I had like excess mental energy, I was like, I'll do a PhD at the same time I was working. And then I got a job that was like mentally engaging. And then I was like, maybe I won't do that. Maybe that's insane.

Gus

Yeah. PhDs generally are insane.

Caitlin

Yes. This is what I've heard. But yeah, if I ever did do a PhD, if I ever wrote a book, it'd be on this, which is why how loosely organised the following information is is going to be deeply embarrassing.

So, if you are a PhD examiner, just bear in mind, this isn't what it would look like. If my name ever comes up in an application, this is unrelated. Right.

Yeah. So what does it mean to be normal? Largely, I think what people mean when they say it is either typical or average, which is to say, like common or standard, which when I did the first draft of this PhD application, I sent it in to my dissertation supervisor and I said, Oh, does this make sense?

She said, Oh, you should really go away and read Ian Hacking and Mark Foster before you.

Gus

Oh, great.

Caitlin

Which is who you'll know from maths.

Gus

Yes. Yes.

Caitlin

That took me like two years. And by that point, I was like, she's forgotten who I am and I have to abandon this. Anyway.

So Ian Hacking charted what he described as the history of the concept of normal. And obviously he's one guy. This is just his opinion.

It's not, you know, in depth. Well, it is quite in depth, but like, it's not the definitive history, but the way he describes it is a normal replaces human nature in the history of philosophy. So enlightenment thinkers think of human nature, whereas modern thinkers tend to think about like concepts of normality and it comes largely from the medical.

So it starts in quite a tight scientific sense of being normal or abnormal, which is to stay normal or pathological, like it's fine or it's something we need to deal with. So it becomes, rather than normal or pathological, normal or damaged, normal or outside of a standard and therefore unhealthy.

Gus

Okay. So as somebody who just went for a ECG test this morning, I was relieved when the very annoyed clinician said everything looks normal, but then started to describe a bunch of other things about me that she wasn't terribly impressed with.

Caitlin

But that's what it means, right? It's like, here is the normal range within which we're not concerned about it. And if you're outside of the normal range, then we believe there to be a problem.

And that's what it means largely in medicine. Then he has a braces example.

Gus

We're talking about braces and teeth.

Caitlin

Yeah. Yeah. Yeah.

Gus

And he's English, right?

Caitlin

I'm pretty sure Ian Hacking's English.

Gus

Okay. Because you know, the rest of the world does laugh at the English teeth.

Caitlin

What's funny about that is, and this is a good reason that braces are a good example. On average, English teeth are healthier or they used to be, but they look worse, whereas American teeth tend to be less healthy, but look quote unquote better or quote unquote normal because of the standard.

Gus

And I'm basing all this on the Simpsons where they had the great book of British smiles and on the front cover was now King Charles, then Prince Charles with a jagged smile, basically commenting on how the English do not like braces.

Caitlin

Which in the UK is very normal.

Gus

Yes.

Caitlin

But where he's trying to defend himself saying that this concept of normal is fairly recent as in 19th century, 1800s onwards, and he's saying, of course, like the root words exist in Latin, it's normal, which is related to a T square, but in Greek it's also as in related to orthodoxy, which takes us to orthodontists who put children's teeth right as in they correct them. And it is a really funny description. It goes, they force the teeth to be like those of normal children, a purely descriptive dental structure, but of course the way children's mouths ought to be.

The middle class has spent a fortune on braces and their children endure years of a kind of oral clitorodectomy or circumcision all in honour of the great God of normal, which I'm not sure that will make it in, but yeah, they put children's teeth right. They make them normal. They make them orthodox.

They make them conform to a particular made up standard. And it comes back to there's the medical aspect of normal, which sounds like definitive, right? When it comes to medicine, there's requiring intervention, not requiring intervention, normal, not normal, normal, abnormal, normal, pathological, but of course those standards are somewhat determined socially or culturally because what counts as abnormal or what counts as wrong or what counts as pathological varies culture to culture.

If you think that children's teeth should look a certain way in order to be normal, then that might require intervention, even where in other cultures what children's teeth should look like is defined in a different way. But the other thing he says is normal always refers to a category of thing, to like a noun. It's always a normal child or a normal adult, a normal heart rate, a normal something.

You can't just have a normal without reference to a thing. And that partly speaks to the word normal referencing typical, like as in representing a type of thing or in reference to a category of thing. But the two definitions become really clear and he picks them out when it comes to a medical investigation.

So some really old study was trying to look at like urine samples and what kind of normal urine would look like. And one set of investigators said, Oh, if you want normal urine, go to the Gare de Lyon and take urine from any of the local pissoirs and get a range of like normal urine, just standard people's urine. And the other set said, no, we need to find people that are typical and take urine from those people whom we know to be typical.

And they're both variations on an understanding of what normal urine would look like. Is it from a huge range of people and therefore it's a kind of distributed set of normal people in society who are coming to and from the Gare de Lyon or is it a set of people who are themselves like typical or good representations of a type of person?

Gus

So in this example, I guess the urine at the train station is likely to consist of a little bit more alcohol, a little bit more drugs and a little bit more cholesterol, let's say, than the urine of somebody who has a regular appointment with their doctor, who monitors their urine samples to see whether they're healthy or not.

Caitlin

Well, I think it's even more like oblique than that. It's like, do you consider normal to be a cross-section of a large proportion of society or do you consider normal to be, we found very normal people who are like the most normal, you know, insert profession here. They're like the platonic ideal local butcher or whatever.

And therefore they're the most normal. We found peak normal of that type.

Gus

Yes.

Caitlin

They just fit under slightly different versions of the definition, because unfortunately normal means lots of things to lots of people.

Gus

Yes.

Caitlin

And I think when people say it, they often mean one or the other, but it's not clarified in our useful understanding of normal. So there's that kind of common understanding, there's where it starts, which is medical, and then it moves into the statistical and we get the curve of errors, which is the kind of bell curve that you normally see if you're, or you commonly see normal, where you commonly see when investigating natural phenomena, right? It's also called the Gaussian curve, but it's that bell curve of things kind of cluster towards generally around a particularly statistically significant, I guess, point, and then tend to drift around a range that can largely be described by a bell curve.

Gus

To be clear, I'm nodding my head as if I'm understanding, but the only course I ever failed at university was stats. And that's why to this day, I still play the lottery and expect to win. But what you're describing, I do remember the trauma of learning the bell curve.

Yes. Okay. And then normal is what fits within the bulge of that curve. Yes.

Caitlin

Well, not even that. So the curve itself is largely describing like natural phenomena. Height, for example, is a good one.

Some people are very, very short, some people are very, very tall, but people tend to cluster around an average. Very few people are really, really short and very few people are really, really, really tall. They tend to sit with a natural range.

And so you get a peak in the middle and then it curves out towards the edge. And the problem with this is that loads of eugenicists popped up. And so you get Galton and you get other people who are taking what is a natural phenomenon, which is seen in lots of different observations of the world and then trying to apply it to things.

And so you get things like the IQ test and how the IQ test is designed is it's designed such that it produces on purpose this bell curve.

Gus

Oh.

Caitlin

Yeah. And so you get this really weird relationship where they've gone, that's how things are in nature. We're going to devise a test which produces that result that almost validates the existence of the test, because by producing that result, it looks legit.

And so we can define what averages and standard is, and then we can use it in our weird eugenicist theories of the world and understandings of normal.

Gus

Yeah. Generally, whenever somebody starts talking about what is seen in nature, I start getting very nervous.

Caitlin

Yes. Which I think is the right response.

Gus

I feel the same about when they talk about culture too. It's usually the very powerful and often the dominant part of society saying what is and what is not.

Caitlin

And what Galton and his kind of group of eugenicists do is as well, they look at what becomes the normal curve and they're the people that call it that. It wasn't particularly called that before them. And they also start thinking of it as a curve towards mediocrity.

So you've got the normal as rather than the healthy or the preferred state of being, you've got the mediocre. Below which is a problem and above which is where you should be, I guess. Which is, I guess, how you get things like Mensa, where you're saying above average or above normal IQ is preferred because you're not mediocre, you're not common, you're not normal, you're better than in some way.

And so now we have loads of warring different versions of normal. There's normal as the healthy and that you can see in lots of different things. You can see it in medical tests, but you can also see it in things like normalising relationships.

So when we are trying to ensure relations between states are peaceful or settled, we often describe it as normalising our relationship. As if the normal state of the world is peace. I like that idea.

I don't think it's necessarily a perfectly descriptive idea of the world, but we tend to call that normalising. And we tend to think of health under that umbrella. And that gets you the abnormal.

And then you get the statistical normal, which gives you the common, but also gives you the deviation, which also gives you the strange and the outlier. And then you get the normal as in the sense of mediocre, as in the sense of not interesting or special, which comes from eugenicists. Already, you've got like 50 million understandings of normal, some of which are pathologising, some of which are nice, but still limiting.

And some of which are mean, but also limiting.

Gus

And when you have it in the context of normal as mediocre, I always feel like that's about intelligence and judgement making. But when you talk about normal in the context of natural, I start thinking of behaviour. You know, it's like somebody's going to find what is it to be a normal human being or normal behaviour or normal response.

And then you use the word abnormal, which immediately I apply to, okay, I've stepped out of a norm. And if you step out of a norm, you deserve to be punished.

Caitlin

Yeah. And it takes us to this concept of normal as in obeying norms and norms, which are described by someone called Cristina Bicchieri, which is, I think, a butchering pronunciation of a last name. She's got this book called The Grammar of Society, which I think is honestly such a brilliant title and like a perfect description of what a norm is.

Because a norm is largely those subconscious cultural things that define for you what acceptable social behaviour is. And it's really, it can be really small stuff. It can be really big stuff.

Like I would say in the UK, it's a norm that, you know, two people accidentally bump into each other. They say, sorry, even if it wasn't their fault. Most people do it.

If someone doesn't do it, it feels a bit weird or rude. It's just kind of like a standard of behaviour. Not all norms are good.

Not all norms are bad. Not all norms are anything in particular, but you'll find them in lots of different units of thing, right? So across a city, the norm might be, well, polite behaviour might look like ignoring other people, getting through interactions quickly because there's a big queue normally.

It can look like not talking to people on the tube in London. You know, it can look like those things. And then if you go into a more rural area or a countryside area, a smaller place or a place with different norms, it can look like saying hello to the bus driver and having a little chat.

Whereas in London, if you tried to have a little chat with the bus driver while people were trying to get on, everyone would think that you were rude. Those things aren't good or bad, but you can get norms that are more onerous or problematic or concerning that over time change, hopefully. But nonetheless, that's kind of what they are.

And so you kind of end up at a version of normal, which is in compliance with those norms. This person is behaving normally because they're complying with how I expect them to behave according to the norms of our kind of cultural understanding. That means that if someone is behaving normally for themselves, a silly example, but say they're from Birmingham and they come to London and they're trying to talk to people, they're behaving normally for where they're from.

They're behaving normally for themselves. They're behaving normally for their community. They're not behaving normally in London and people are going to think they're weird and they're not going to want to sit near them on the tube if they keep trying to talk to them.

Gus

Yes.

Caitlin

And so normal in that sense is extremely socially constructed and varies dramatically based on context. And it brings us to normal as visible, right? Like the things you see all the time are the things you largely think of as normal, which is how you get culture shock, right?

Like if you go to someone that's very different and everything is very visibly different, it can be quite disconcerting because what is normal to you is what is visible all the time. And that's also why we talk about normalising X by having more representation, for example, on telly. You know, like when we try and normalise, you know, gay relationships or disability, we often do that by trying to make them more visible through things like TV, film and other visual mediums where, you know, inclusion becomes a way of expanding our understanding of what normal looks like.

Of course, potentially statistically, it has always been normal to be disabled. Most people in their life at some point will be disabled, whether that's through age, whether that's through an accident, whether it's temporary, whether it's permanent, it's incredibly normal to have some kind of disability. But of course, that's not how it looks often.

And often that's not what people think of. And that's how we get to, even though statistically it's normal, traditionally people haven't always thought of it as normal in the sense of something they need to think about or engage with or wrap into their worldview or expect of other people, right? Like it's how people say incredibly insensitive things to, you know, someone who might be blind or deaf because they didn't expect them to be those things, even though those things are pretty common.

Because whilst they're common, they're not necessarily perceived as quote unquote normal. Unfortunately, normal is a very loaded term, right? Normal implies all those things we said before about the pathological, about the abnormal, about, you know, these things which carry negative meaning and connotation, something to not be normal, then it has to be outside of the normal.

And that might mean bad. It might mean unhealthy. It might mean pathological.

It might mean et cetera, not failing to meet a standard, any number of things. And so again, the word normal encapsulates a huge amount of things, but it also implies a certain level of judgement, even where what people are trying to mean is average, common, statistically quote, normal, because that's the language we've gotten to given it.

Gus

But it could also be expected, required. You know, everybody's behaving to a type because that type is expected at that moment in time. And that's power might've created that.

Local norms might've created that. Like I often think, using your example of the bus or the tube in London, I feel like there's almost a subculture of behaviour on the tube in London that is neither English or London or anything. It is just the behaviour in the tube.

And the moment you get out of the tube, you no longer behave according to that norm. And if you get on a bus, it's a completely different, like the people who would breach a norm in a bus, but abide by the norm in the tube and vice versa. It's fascinating.

But I also imagine there's some elderly Londoner tutting everybody who's not abiding by the norms of the 1950s and 1960s behaviours on the tube and whatnot.

Caitlin

Yeah. And it's one of those things. Average is a very good at population level stuff.

Like I think the example generally used as BMI, BMI can be very good at capturing the health of the population, but it's useless for individuals.

Gus

Body mass index?

Caitlin

Yes. Which is like your height and your weight. That's the ratio between the height or something like that.

Gus

I never fit that one. Yeah. I'm always, whatever, in new math or old math, I always come out as being obese and it's infuriating.

Caitlin

It doesn't tell you much about an individual person's health. It doesn't tell you much about their body composition. It doesn't, like an incredibly healthy, kind of very well muscled person will always be problematically overweight according to the scale because it doesn't account for muscle versus fat.

It doesn't account for any number of things. Like if you're disabled and you have an amputation, then your BMI is always going to be off. But at a population level, apparently it can be quite good as a statistical way of like kind of broad strokes, how healthy is a society?

And that's largely true about most things. Most people are not average in every single aspect. No one in and of themselves in every aspect is quote unquote normal.

Like the most average person, statistically, the old jokes, the most average person has like 2.5 children. You know, they have an impossible series of things because it's averaging every single thing to find the perfect middle, which doesn't exist. But anyway, most people in and of themselves are not wholly quote unquote normal.

Everyone has something about them, which arguably depending on where they're standing and who they're talking to is not average, common, et cetera. Now that we have an increasingly usefully confused understanding of what normal is, it takes us to what I think is really interesting about the concept, which is how it applies day to day. Increasingly, everything you interact with has been designed by a person who is trying to account for your way of using it, right?

And that's true of web interfaces. It's true of physical interfaces. It's true of AI systems.

It's true of et cetera, et cetera, et cetera. It's true of the things that are trying to capture you in databases or in, well, largely in databases. And so what normal looks like for that person that's designing those systems and for those people that are designing those systems is really, really, really key.

You know, if you're trying to design an ID database, so listen here to our podcast with Edgar Whitley, which is about designing ID databases basically, or ID systems. What your understanding of what a normal person or a normal experience is, is central to your kind of understanding of what you need to capture. And that's true of loads of things.

And I think a really good example is the proctoring software that schools tried to roll out during the pandemic. So if you were trying to take an exam in the pandemic and now, but it's kind of fallen off a little bit, particularly because of AI, or like consumer access to AI, you were likely engaging with a system like Proctorio’s or like, you know, any other company, which forced you to have your camera on, that tried to work out if your behaviour was suspicious. And it did that using lots of different inputs, including how much you moved, where your eyes moved to, you know, and other different things.

But of course, what counts as suspicious depends on what your understanding of normal behaviour is and what your normal range of behaviour is. And if you have any number of conditions that are very common, you know, where you maybe don't have full physical control or for whatever reason, you might move around a lot more. If you have a condition like ADHD, you might move around a lot more.

And those things, which in kind of a classroom setting would be normal for you because you have a relationship with your teacher and they understand kind of how you generally are. So normal in that sense of like common or average, the system is not designed with a knowledge of you and your history. It's designed of that kind of overarching population level standard of what normal is.

So it's trying to capture that almost centre of what it believes the bell curve to look like. And that can mean they've just not thought about the way that other people move or how much they shift around, whether that's because they have, I think it's called dystonic or dystonia where, you know, your muscular control is different and you're twitchier or any number of other things. Either you haven't thought about it necessarily because it's outside of your conception of what normal will look like, or you have, but you didn't think it would matter because it would be a small number.

But either way, it's not who you were designing the system for.

Gus

What also that I mentioned that annoys me about that example is it's a example of a situation where coercion and control is already being asserted. There is a normal and expected behaviour of a student under an exam setting, and you shall not deviate from that. So it's not even how does somebody normally behave in an educational setting?

It is how do you behave in a highly coercive controlled environment of an exam? And then the system developer saying, well, I have an idea of what that is, and therefore the system shall observe for that. And then you shall not deviate.

Caitlin

It also takes us again to that version of normal as the standard and as meeting the standard and anything that doesn't as abnormal or falling outside of that standard. And that's if the system lets you in to begin with, because a lot of people were finding, particularly black kids, were finding that unless they let themselves really aggressively, the camera just wasn't recognising there was a person there because the software wasn't recognising there was a person there. Which takes us to who do you think is normally going to be taking the exam?

Gus

Oh, Jesus.

Caitlin

Yeah. And so increasingly, what developers understanding of normal is, is becoming increasingly the governing system for how we interact with any number of other systems. And that's even worse when you think about the fact that so much is culturally constructed.

Another really good example is emotion AI. So yes, there's this concept of emotion recognition, which I think is largely agreed to be nonsense. It makes no sense and it doesn't really exist.

Gus

Except for anybody who's in the AI industry.

Caitlin

But except anyone who's crawled up. I probably can't say crawled up their own ass and died, except anyone who wants to believe it exists, if that makes sense. I don't think scientifically it doesn't exist, but like some people really, really want it to exist.

And it feels like to them, like it should exist. But the way that you express emotion is largely culturally constructed. So, you know, in the UK, in America, in lots of different places, it's quote unquote normal that when you smile, you're happy.

Of course, that's not inherently true. Some people, you know, you might smile because you're trying to be polite, because you're trying to convince the other person that you're happy when you're very sad. Like there are any number of things that they can mean.

But the idea is that largely emotions translate to reliable physical expressions that can be recorded and understood and therefore perceived by a computer. Of course, this isn't true.

Gus

No, no. And for people who haven't heard me rant about this, one of the best books I've ever read is called How Emotions Are Made by Lisa Feldman Barrett. We will include a link to it in the description because it's about six years ago she wrote this book and it just tears apart this notion that there is a thing called an emotion and that emotion is visibly verifiable.

It's just neither of those things are true necessarily.

Caitlin

Yeah. And it comes back to, well, normally, but when people are happy, they smile. Well, statistically, that's not inherently true.

Scientifically, that's not inherently true. Like mathematically, that's not true every time. What you mean when you say normally is not any of those things because it's so culturally constructed, because it's based on your understanding of the world, because, because, because we end up with people trying to sell emotion recognition, which can't exist and doesn't exist already.

Now people sell these systems, even though there's nothing to suggest that they would be real.

Gus

Yeah. And the environment where it's often heralded is at airports and at borders, where the idea is that there's a normal behaviour and normal emotions to feel.

Gus

Yeah. And the environment where it's often heralded is at airports and at borders. Where the idea is that there's a normal behaviour and normal emotions to feel at an airport, and there's those who are abnormal, and those are the ones who are clearly a threat.

And as somebody who gets very nervous in queues, as somebody who gets nervous about behaving the right way at the right moment in time, about what goes into the bin at the airport security, what doesn't go in the bin, do my belt come off, do my shoes come, my brain is working so hard. I might be smiling because I am freaking the hell out doing a billion computations at once and afraid of getting yelled at. So of course I look suspicious.

I don't look like the family is about to go on vacation. They might be on the same flight as us, but they don't look like me because they're not stressing.

Caitlin

You know? Again, like what is normal in that context, but what is normal for the people in the queue is not the same thing as what is normal for the airport. Like, you know, someone who dresses in religious dress all the time, and that's their normal way of being in the world.

It's very common in their community. It's very normal. Might get to an airport and suddenly, like, for the people who are judging them, it's not quote unquote normal.

And it takes us back to what do we mean? Do we mean statistically across our personal society? Do we mean statistically across the people we interact with, the people we see all the time?

What do we mean? And normal is both a word that carries a huge amount of meaning and very little meaning all at the same time. And it's such a frustratingly slippery word, but it defines just an increasing volume of the things that we interact with or engage with, or are designed for us on the basis of what normal looks like for the users of that system.

And we can fall down a million rabbit holes about how normal is both something that people perceive about the world, but it's also something people enforce on the world. Which takes us back to the panopticon and Foucault and all that stuff, which is to say that we socially enforce our understanding of normal on other people. And that might be the older woman tutting at the people not following the tube norms that she understands.

Or it might be the same older woman, I'm stereotyping, it's not inherent to older women, but tutting at the gay couple kissing or whatever it is that is a social enforcement of her understanding of normal, which is the same thing that developers are doing when they create these systems. They're socially enforcing their understanding of what normal is. And for people who are outside of that normal, who are abnormal or strange to them or fall below the standard that they're expecting or outside of the standard that they're expecting to account for, they're enforcing that they're different and that they're enforcing that they're strange.

And if you're trying to log on to your exam and it's not letting you unless you light yourself up incredibly brightly, it's enforcing that you weren't expected to be the kind of person to take that exam. And that in and of itself is a harm and is damaging and is a problem. And so when you only account for the normal, the average, what you believe to be the common user, which may well be a fault of your imagination, then you're telling people who you believe will be the normal or the common or the standard user of your system.

And you're telling people who fall outside of that, I think there's a problem with you. Like this system that you're interacting with believes you shouldn't be here and that it's strange that you're here and it becomes its own social punishment. You have to jump through more loopholes, except it's the computer enforcing those loopholes now.

It's not just, but not individual people, it's the entire technological system that you're being forced to engage with. And so what normal looks like, it becomes just deeply damaging.

Gus

It causes like a billion and one calculations. Like I'm a good immigrant and with all the horrible things that goes along with being a good immigrant, but I'm a good immigrant and Canadian society. I'm a good immigrant and American society.

I'm a good immigrant and British society, and each are very, very different. And then you add London society to that, and that's a different, and I have to at any moment in time, jump through the hoops. So I was in New York two weeks ago, and I knew how to behave as a non white American in a setting like New York city or at the border or on the subway or at a tourist destination.

There's all these calculations, but I know how to behave. And then I land back in London and I'm on the tube. I know how to behave.

And yet I'm a different type of immigrant there. Like to be a Brown person in the UK is different than being a Brown person in America. To give an example, after 9-11, in order to make sure that everybody around me did not know that I was an immigrant that they had to worry about and treat differently and stop and search, I would wear a baseball cap.

But to wear a baseball cap over in Britain at that time was just, well, that's terribly odd. And it didn't help that in 2005, during the London bombings, one of the failed bombers was wearing a New York Yankees hat. And it's like, you just robbed me of the one bit of camouflage I used to look normal.

And that was considered, he was threatening New York at that point. So then all of a sudden, Brown and Black people wearing hats became symbolic of you being. But then when we add your lens to this, your additional lens of systems being designed with a definition of normal, the student showing up at the exam doesn't know who designed the system.

And I wouldn't be able to ask, is it the American version of being Brown and normal? Is it the English version? Is it the London version?

And that makes, as I say, a thousand and one additional calculation you have to do.

Caitlin

And it comes back as well to, if I'm not in an English speaking country and the system was designed in an English speaking country, well, what's the understanding of what normal in my country is in a different country? If it's France, is it that French is the only language that's kind of spoken? French is incredibly common, but it's not the only language.

I mean, if you're in Wales, there are plenty of people in the UK, in Wales, who English is not their first language. Is that captured with American education technologies? No.

In the same way that, you know, if you're deploying a technology in a country like Mexico, which has a huge number of indigenous languages, are you making it available only in Spanish? Because that's your understanding of what normal looks like in Mexico, even if it doesn't meet the needs of everyone that lives there or has to interact with that system. And so it gets, you know, even more complicated when you add in all the different normals that exist globally.

But yeah, I think, and obviously it interacts with a lot of different things, right? Like racism, ableism, you know, homophobia, class, all these things both play into this understanding of normal, but also are their own things. It's like intersectional, right?

It might be that disability falls outside of someone's understanding of what normal is, it's an and, not an or, if that makes sense. It's not that they don't exist separate to this understanding of normal or not normal, but it's that they have a relationship. That makes sense.

But I just think it's such a fascinating way of looking at this problem of design and unconscious bias and any number of other things, which comes back to this very normal way of talking and thinking about the world, because it comes up all the time. Once you start looking for the word normal, it's everywhere, people use it constantly. And you see it in people, I was Googling this as a thread on Reddit, where someone was saying, what do people mean when they ask, is it normal for someone to do X?

Because it's a question that's seen all the time. Is it normal for me to do this? Is it normal for me to do that?

And they were getting different answers. And what those people are looking for is reassurance, right? They're not asking, is it statistically likely that given my background, I would do this.

They're asking, is it okay? Because that's often what people mean when they're talking about normal. Is it okay that I do this?

Is it okay that I want to do this? Is it normal? Am I strange?

Do I fall outside the standard?

Gus

I feel like you're drawing an X, Y axis with your definitions of what the X and the Y are to navigate normality, where I have a slightly different lens I'm applying to it, which is, periodically, society goes back and forth between defining normality through mathematical means or through behavioural means. When big data came along as this idea, that was the idea that with enough data, you can establish correlation, which is close enough to causation. And so therefore, we can define the world mathematically.

So that's where BMI was having its heyday, or now it's like, is your testosterone count? And all these numbers that we're using. But then there's a behavioural approach, which is what you described, which is like, how are we supposed to behave?

What is the normal way of behaviour? But I feel like with this AI moment, where AI is both statistical, but also trained on vast amounts of data, and we can be honest that a lot of it converges, that the data is a lot of normal data. But I would also imagine the data is very varied.

But we're in this moment where actually the two might come together, the statistical normal and the behavioural normal, and you have these systems deciding what is normal and defining in a way that maybe the human brain has never been able to navigate. So am I in normal health when it comes to my physical and health? And I would say my BMI is this, my weight is this, and so there's that numerical, no, I'm abnormal.

Then there's the new metrics of like testosterone counts and cholesterol counts and all of that. I was just in a conversation with people about it and how they're struggling as men to understand what they should be doing, blah, blah, blah. And then there's like, as a man, how are you supposed to behave?

The Reddit conversation you're talking about. And then how am I supposed to show my strength? How am I supposed to show my health?

And that's when you can describe most of the toxicity on the internet with just that one example.

Caitlin

I think there's an interesting point about correlation, which is if you do this, it's normal for you to do this. A normal person that runs a lot talks about running a lot. And so it's weird if you go running a lot and you don't talk about it.

That's a really sketchy example.

Gus

It's a great example. It's like Fraser would always say, you would always know somebody went to Harvard because they would mention it within the first 30 seconds of getting there.

Caitlin

Well, and when it comes to the healthy range, and I think increasingly what we're seeing is goes back to the mediocrity curve. It's like, well, here's the healthy range. Here's what a doctor would tell you not to worry about.

And here's what you could do. And here's the ways that you could push yourself past normal into special. And here's all the things that you should be doing if you cared.

And I think that's particularly true of things like testosterone and the very, very online health trends, where it's not about being normal. Or if it is, it's not about being medically normal, as in within the healthy range that a doctor would tell you not to worry about. It's about being more than.

It's about being special. It's surely more is better.

Gus

It's to be real. A real man has this. A real soldier has testosterone of a certain level.

Caitlin

The platonic ideal of it is this. And it is the only normal thing for you to be doing is to care about that platonic ideal and reach out for it. And if you don't do that, what is wrong with you?

You're strange. It's an artificial understanding of it. But that's, I think, what the marketing is.

And that actually, to come back to my master's dissertation, to bring us back to the beginning, is shaving started as a marketing campaign. It started with Gillette creating a marketing campaign functionally saying, you know, normal people shave. If you don't, you're stinky and weird.

It was a little bit more elegant than that, but largely that's what it is. It's like, this is normal. This is what the standard is.

If you don't do it, you're failing to meet the standard and you're strange and you probably smell and none of your friends want to tell you. And you don't necessarily think about marketing campaigns as having that kind of society level normative impact, whereby, you know, for a long time, not shaving was seen as very strange. You know, you get and you get the stereotype, you know, in the eighties and the nineties of kind of unwashed hippie lesbians with lots and lots of armpit hair.

Now, armpit hair is obviously incredibly normal in the sense that everyone has it. Like there is arguably nothing more normal unless you have alopecia, which is a condition which is also very common and normal. But the standard was, if you don't shave it off, you're strange.

You're not normal because normal behaviour was shaving it off. And it's a life cycle which you wouldn't have thought you would be able to trace back to something that feels fairly, not innocuous, but like normal and small and like a standard thing you interact with day to day. But actually that kind of beginning and that moneymaking impulse and that marketing campaign and that kind of bandwagon drove a lot of behaviour for a very long time and arguably still drives a lot of behaviour for a lot of people.

And it's a microcosm of that normalising trend, which is socially constructed and is XYZ. But you will still get people telling you that if you don't shave your underarm hair, that's not healthy. It's smelly.

It's, you know, got the potential for bacteria. It's any number of things. Of course, that's somehow magically only true if you're, a woman and you present as a woman, you know, like as far as I know, men generally don't get that feedback and, but it nonetheless feeds into this idea of normal as healthy versus an unhealthy, even when that's not a medical opinion.

And so, yeah, I think it's just something out of interest that I think it would be fun for people to look out for and kind of make a mental note going, Oh, that's the word normal. Which version do I think that they mean?

Gus

And at any moment in time, what version of normal am I being asked to endorse or perform? And how many calculations do I have to make as to who's defining it? What their definition of normal is?

What the people around me's definition of normal is? And then going back to the airport experience, there's the normal in the eyes of the airport security guard. There's the normal in the eyes of the person travelling beside you.

There's the normal in the eyes of the people travelling with you.

Caitlin

There's the normal in the eyes of the scanner.

Gus

The scanner of the CCTV. And yes, all of those things.

Caitlin

You know, it doesn't have to be normal in the sense of people. It can be normal in the sense of any number of things.

Gus

And I think like the, what the end game of this entire conversation is, is like, I think there's the one hope that with emergent intelligent systems, they will develop an attitude that is beyond what humans have done throughout time. And they will be able to identify a toy gun, a drum, or any of the things that a wide swath of society would bring through an airport scanner. And it would say, it's okay.

Whereas an individual who is looking through the bag, bringing their own sense of what's normal in the setting of what's normal at London Heathrow airport versus Edinburgh airport, and all those things that my brain has to compute 1001 times, maybe emergent intelligent systems will combine both the normative and the behavioural and the scientific, and then everything will be all right. But that's just, that's this imagined impossible imagined future versus the future you're worried about, which is we're going to have to constantly try to figure out what normal is being applied in any setting with technology without technology. And we're kind of, unless we're able to do what you're asking us to do, which is the constantly question, what is the definition and the perspective of normal?

Caitlin

Is this idea that computers will be able to see things as they are and describe the world exactly one-to-one as it is, as if there is a completely objective version of reality that computers are able to engage with in a way that humans aren't, which I just don't think is true as long as computers are built by people on the basis of data collected by people, on the basis of sensors, or, you know, any number of other things designed by people, like the proctoring software, not recognising, you know, black students goes back to a very long history of camera equipment, failing to properly capture black people back to, you know, we mentioned a film and those, the sensor that is the camera sensor exists on that historical trajectory and isn't objectively separate from it, which is very difficult, I think, to truly absorb.

And it's a known cognitive bias, which is functionally the computer says yes. And so I believe the computer, the computer says no. And so I believe the computer, but yes, that was a very normal podcast from, from Tech Film and the Privacy International.

Gus

I'm glad we finally got to this. I know you've wanted to discuss it for a very long time.

Caitlin

I've wanted to want to discuss it, which isn't the same thing. It's a second order want.

Gus

But it's a second order thing too.

Caitlin

Yeah.

Gus

Because you're not just talking about classification. You're not just talking about right and wrong. You're not just talking about racism, sexism, and all the other isms.

You're talking about the very mindset and the questionable frames around the various mindsets that apply to the definition of something that dominates our societies and dominates our minds. And as I say, dominates the calculations I have to do at any moment in time. And if anything, that's the oppression.

Caitlin

Yeah.

Gus

And that's why, as I, as I said, there's this hope that either emergent systems will erase all of that or compute it in a way that we don't have to worry, or we will become so good at navigating these things. No, actually there's, there's no, there's no positive ending really.

Caitlin

Well, our conception of normal has expanded. I think that's fair to say, like people's access to such a wide global variety of media, of imagery, of conversation, because of things like the internet, I think has expanded most people's definition of normal, which has changed over time. That comes with a certain homogenising kind of flattening, you know, but it also means that things are more visible to people.

It would be hard to argue that gay relationships, for example, are seen to be less normal now than they used to be. I think they are considered to be a lot more quote unquote normal. And within that comes a lot of freedom because it's a permissive thing, right?

Like it's normal and therefore it's permitted. And there's nothing to say that our understanding of what normal looks like won't continue to expand. And it's also, I think, I'm not saying it's morally neutral because it isn't, but at the same time, it's kind of okay to have an understanding of what's rude or not rude in your local area, as long as you're not being a dick about it.

Like most societies operate with some level of like social lube of understanding, of shorthand that allows you to quickly move throughout your day fairly frictionlessly. It's a question of ensuring that you yourself aren't enforcing your understanding of normal on other people in ways which are mean or cruel, you know, and taking a step back and being like, when I say that that's weird or strange or abnormal, what do I mean? What definition am I using?

I'm noticing it when other people do it. Because it's a really common tactic, I think, to dismiss things as abnormal or strange or weird rather than engage with what they are and how common they might be and what they actually look like in practise. And it's a conversation.

Oh, that's weird. Okay, great. You don't have to investigate it any further or think about it or engage with it.

You can go, that's weird and dismiss it. But I think if you can recognise the instinct, go, that's weird, that's strange, that's outside my understanding and dismiss it. And you can recognise it in other people.

It gives you a clue as to, okay, maybe that's an area where you need to think about or engage with or consider and not allow other people to end the conversation with, well, I think that's weird. Well, okay. Well, why do I care about what you think is weird?

What is your understanding of normal? And why is it something that I need to prescriptively comply with? I think it's just normal.

It's just a warning. You know, someone's saying something's normal or not normal. It's just a little, hmm, that's interesting.

But yeah.

Gus

Caitlin, it's rare that I get through a conversation like this with a different understanding of when I went into it. This is wonderful because I get entirely what you're saying. It isn't just a progressive attitude of like, we should be more inviting to normal and understand that there are as many normals.

It's actually to say, we're navigating the world at any moment in time with a conceptualisation of normal, and we should be understanding that there are others. And even if for a moment you tut-tut at an abnormality, you should just recognise that your sense of an abnormality is an abnormality in its own right. And that these things are just, as you say, the social lube of getting on in life is to understand and navigate all of these things, which almost takes me back to what I consider to curse of having to navigate other people's expectations of normal.

But actually what you're asking for is that everybody should have that in mind as they navigate an increasingly diverse life that we have, which we're fortunate to have.

Caitlin

Also, sometimes people are just rude and that's fine. If someone is being rude on the tube or they're not moving down the carriage when they should be, that can both be annoying, weird, rude, and rude. They need to move down the carriage.

You don't have to pathologise it to then go, move, move. More people can fit on this train.

Gus

There's an urgency of the moment. There's a certain mechanics.

Caitlin

Not all things are created equal. You know what I'm saying? And sometimes the norm is the norm because it allows for the most people possible to fit on the central line in rush hour.

It's horrible. Everyone hates it, but you've got to move down the carriage so people can fit on the train. It's the norm for a good reason.

But also if you're in London and you're on the tube and it's busy, take your backpack off and put it on the ground and move it down inside the carriage. If I get hit in the face with your backpack because you're tall, I'm going to be unhappy. I'm not sure that is a norm, but I think that it should be.

Gus

I keep on waiting for it to become a norm. It drives me nuts every day.

Caitlin

You can tell who's on the tube frequently and who's not based on where their backpack is. It's a weird hill to die on, but I'll stand by it.

Gus

On that note, thank you for listening. You can sign up to be the first to learn more about our work at pvcy.org/podsignup and we'll include some links to relevant articles and information including some of the books that Caitlin read on her undergraduate and master's degree in the description wherever you're listening or on our website at pvcy.org/techpill. Don't forget to rate and subscribe to the podcast on whichever platform you use.

The music is courtesy of Sepia. This podcast was produced by Max Burnell for Privacy International.