Sociology: A different kind of gambler’s fallacy

Pictured: Clip art featuring various anonymous charts and graphs used to illustrate statistics. You’ve probably encountered them before but never noticed.

I’m going to tell you a parable of two friends. At one point, they were both kids. They would hang out, play video games, talk about things, joke, and in doing so, become very good friends.

Then they grew up and got jobs. One of them became a thief, the other became a policeman.

But they stayed friends. Very good friends in fact. Might sound unlikely, but it’s a true story. For you see, I was one of them, and I’m no policeman.

We honestly get along fine, and usually joke about our respective turns in life. It’s not a big deal to us, and because we’re two real people and our rivalry is purely abstract, neither one cares. It’s a lovely thing of the human condition, how uniforms are so bad at separating us.

But I remember once we had a debate. I don’t think it was an argument, we both got kind of passionate but there was no real hostility or animosity. We just argued like romantical people do in southern Europe, where you get a bit loud but you hold no bad feelings about it. It’s all in the spirit of the debate.

The argument was about politics, and I remember one thing that stuck with me, it requires a bit of context, he didn’t say it like he was happy about it, he seemed more troubled. He said “I hate to say it, but statistics don’t lie.”

It was in reference to immigrants and crime. He knows I’m an immigrant, and I think we wanted me to offer the other side of that story, to put him at ease, help him see things differently. Like he didn’t want it to be true, but he was stuck believing in it.

And I thought it was funny, because he’s by no means a liberal. He’s not a conservative either. He’s a Platonic republican I’d say. What some might call a pragmatist or independent in modern jargon.

But what I found funny about it was how this is a very unprogressive notion, and yet it comes from the exact same place as progressive ideas, namely statistics. I kind of hate statistics, and I say that as someone who has not only studied engineering, but also economics for several years. I don’t hate useful statistics of course, I hate the way they’re used in a school of pseudoscience we call sociology.

Because projections, probability, higher numbers, data correlation and all these things are far more of an art than a science, and somewhere along the lines it became the new phrenology. The new pseudoscience of social hierarchies.

And what impresses me is how people don’t just explain with very simple scientific terms how, no, you can’t measure human demographics with statistics in any meaningful sense. Instead we got entire fields of study dedicated to its divination and pseudoscience.

So let’s say you get a statistic in some country like, and again, I don’t believe in this nonsense, I’m just using what is in my opinion a troubling and all too common example:

But let’s say you get a statistic saying that people from some ethnicity is more likely to commit crime. What happens?

Well, suddenly you got a dichotomy of progressive and conservative racists, who both accept the statistic as an a priori truth, and then proceed to conclude it.

“Oh but they live in poverty, they are not given the right opportunities, it’s police bias, etc, etc.”

“They are criminals, they don’t respect our culture, they refuse to assimilate, we must have law and order, etc, etc.”

No one ever bothers saying the right thing, which is “So what? People aren’t their statistics. All you’ve done is draw an arbitrary vector between two data points and pretended it matters. I could do the same to prove dog owners are more likely to commit crime, or that Ikea furniture contributes to house fires.”

Now some of these talking points could be true, I’ve migrated to brutal and uncivilised countries where I certainly had no respect for the law, where they treated their elderly population, and their downtrodden people like trash. Where I am proud to say I would struggle to survive, where I stole food and gave to the needy, where I rebelled.

I have frequently broken laws I consider to be unjust or wrong, usually with the objective in mind to help someone in need. I don’t want to incriminate myself with details, but it happens. So that’s a point for the conservatives.

But the liberals are right too, often I’ve stolen because I’ve been broke, and miserable. But you know what? Most immigrants I knew, even during my gang days, were not thieves or crooks of any kind. Most of them worked twice as hard for half as much, I knew people who were professors in their old country, and who collected welfare in this new place they had to go to.

And if you can’t give a professor a good job then what kind of society are you? How is that civilised? You should be able to find professor jobs in all kinds of places. I knew people who were qualified in all manner of amazing skills, and who could’ve easily used them to live well in the underworld, but who would rather accept honest poverty before unlawful success.

And if you tell the statisticians that, then they say “Oh well that’s anecdotal.”

As opposed to what? Your scryer’s ball? Crime statistics are anecdotal too. It’s just anecdotes that are ratified by an institution. Anecdotes of prosecutors, judges, police, and similar things. And they can absolutely be manipulated.

For instance: Very few of them make distinctions between misdemeanors and what lawyers sometimes call “true crime.” A common crime for immigrants in Europe to commit for instance is violation of fire ordinances. They come from countries with different ordinances, and so they have campfires or cooking fires in public places where it’s not allowed. This is especially common with refugees who, for several months or maybe even years of their lives, have grown accustomed to living without electricity in war torn places.

Usually they get a lecture on how things work in the new country, and they don’t do it again, but it still enters into statistics. You can solve it with a pamphlet, and yet people act as if Hannibal is at the gates.

Another trick is to look at criminal charges rather than criminal convictions. So that the people who say, might not speak the language too well, and might be more likely to get booked on grounds of suspicion but are then acquitted after explaining things with a translator, are more likely to end up on statistics.

Then you got true crime, which is like a crime with a moral implication. Grand larceny, rape, murder, robbery, things that don’t just break rules on paper, but also commonly accepted rules of conscience.

Statistics often conflate these two things. Moreover, it also ignores how policing works, how once you’re known to live in a neighborhood where people don’t understand fire ordinances, then police is going to up activity there to be effective. That’s a vicious cycle.

I could go on forever, but the point is that yes, statistics do lie. Because when you ascribe them to human beings, then you’re basically inventing an imaginary person.

For example, let’s say you run some statistics and calculate things like body mass index and physical proportions in adults, and arm span and so on, and it turns out that statistically speaking, an adult with no professional swimming training will actually win a swimming race against a class of junior swimming athletes because of how their arms and legs can just produce more water pressure quicker.

Now does that mean we would give Stephen Hawking a snorkel? With his education I suspect he would’ve been the first to point out how probability is a bit more complicated.

So you got all these imaginary minorities existing entirely in numbers, who can’t be found anywhere in the world. Supposedly you got devout Muslims, praying 5 times a day at the quibla, read the Qu’ran and the Hadith, honouring Ramadan and living in perfect sobriety, who also just cannot help but steal people’s phones all day.

I assure you: No such people exist. The devout thief is a Christian thing, and our Muslim friends are far too honourable to infringe upon patents like this.

Unless you’re a Salafist I guess, then anything goes. They commit Catholic levels of crime.

Point is, there is no statistical person, and there are no “marginalised” people. There’s oppressed people, there’s downtrodden people, there’s poor people, there’s segregated people, there’s neglected people, there’s suffering people, but each one owes a bit more to their humanity than some accountant’s trickery.

“Marginalised” instead became a trendy way for the haves to exploit the marketable image of the have-nots.

It allows well to do academic types to correlate their identity to some statistic, and tell people who hard they got it, and how they’re a “voice of their community.”

And suddenly Ellen DeGeneres with her Hollywood money, is representing disowned teenagers who sleep in the gutters of San Francisco. Suddenly you got Kuwaiti royalty representing the Iraqi civilians, pleading to the UN to invade their country.

And it gets pretty dehumanising, because it turns anti-racism into the new racism. For instance a lot of Black and Native communities in the US have a disproportionate amount of liquor stores per captia.

Does that mean that the perfect Christmas gift for them is a nice bottle of malt liqueur? That this is a thoughtful gift? That this warrants any recourse besides arranging for someone to die in a misfortunate malt liqueur accident?

No. It just means that the liqueur industry is unambiguously evil. It means they know how to take advantage of people. Doesn’t change how most people have the tremendous strength to walk past those terrible devil’s shops and get on with life.

Nor does it change how statistics define some, but not others. If a German community drinks, then it’s cultural, if a Black community drinks, then it’s social. A German person is being jovial, a Black person is being an alcoholic, and so on. It’s all twisted around and framed in all manner of ways, and none of it is real. It’s just assumptions based on arbitrary framings of numbers.

And the real trick to fighting prejudices is not to reinforce them with pseudoscience and then explain them away, but to dismiss them. To stop looking at people’s identity, whether that’s colour or gender or whatever else, and instead look at their names and faces, how these are people who are only themselves, and that they should be afforded the dignity of only being themselves.

Pictured: Social realist painting of a young black man looking up at something before him. The scene is made up of a metropolitan backdrop that is anonymous and busy, and he appears to be lost in thought. The colour scheme is made up of cold lighting, giving it a sterile and slightly detached feeling.

--

--

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store