Being epistemologically complacent, billiard balls, and polarised politically obnoxious culture

The best way I know of for ruining a conversation with my wife or close friends is to talk about epistemology. Few things seem as boring and inconsequential to the well-educated middle-classes as the question of whether the chair you are sitting on can be proven to be real.

To the philosophers, the common-sense assumption that there is such a thing as objective reality that you're looking at is called naive realism. I used to find this phrasing objectionable. But once you grasp the essential details of this thinking, it is only sensible for you to be worried about how certain you can be about the nature of reality. Especially when you realise it has more influence on modern politics than almost anything else.

An example of how easy it is to undermine the idea of objective reality: is what you call green actually green when you know that other creatures see on visual spectrums we don't have? What about the colour blind? Isn't it more true to say that you see a version of the green object, the insect sees another version, the colour blind person sees a third version?

So, there is some gap between the thing you look at and what you see.

(Remember when your physics teacher taught you that when you sit on a chair you don't actually touch it, but your atoms and the chair's atoms exist in a state of massive resistance with an infinitesimal gap between them?)

That almost non-existent gap is also real in your broader perceptions. We cannot claim 100% absolute certainty of what we know. Maybe it's as high as 99.9*%, but the gap is there. As Bertrand Russell said, "There can be no doubt that some of our beliefs are erroneous; thus we are led to inquire what certainty we can ever have that such and such a belief is not erroneous. In other words, can we ever know anything at all, or do we merely sometimes by good luck believe what is true?"

And if you follow a gap like that for long enough it makes a big difference. A miss by a millimetre is as good as a miss by a mile. Newton was almost right about the nature of the universe but Einstein had to correct him. Newton's measurements only show small gaps, but the gaps were enough. Reality either is real or not quite real. The extent of the not quite is somewhat secondary.

This problem matters because it is the basis of postmodernism and the radical subjectivity that is at the heart of the new version of politics we are all watching and suffering through. This is not just Trump and Fake News. The arguments about cancel culture, group identity, free speech, whether or not children should be allowed to have sex change surgery and the whole host of associated issues is all traced back to the philosophical idea that reality is not objective and cannot be known by reason.

This is about the idea that the Enlightenment ('Dare to know!' in Kant's words) was wrong because reality isn't really there. It's about whether we need reason and logic and science in a world without certain knowledge of reality.

Once you abandon reality, radical subjectivity takes over. And we are reaching the point where the idea of reality is completely irrelevant in political culture.

As Andrew Sullivan said, describing post-modernism, "The idea of objective truth—even if it is viewed as always somewhat beyond our reach—is abandoned. All we have are narratives, stories, whose meaning is entirely provisional, and can in turn be subverted or problematized." That is not (just) a political description: Sullivan was reviewing a new book by two academics.

In his excellent book Explaining Postmodernism Stephen Hicks traces this line of thought from Hume through to modern thinkers. Hume showed up the problem of causation. It is not certain that because the sun rose today it will rise tomorrow. When we see one billiard ball strike another one the most we can be certain of is that the two events happened together, not that one caused the other. From there it was a hop, skip and a jump for Kant to reject reason as a means of knowing the world, and then swiftly on to a century of German thinkers from Hegel to Nietzche slowly dismantling reason to be replaced with power, instinct, and identity. Heidegger would then write that we must reject the assumption "that 'logic' is the highest court of appeal."

And now we have a world where everyone's version of reality is legitimate, where reason and science are replaceable with power and identity.

This sounds like a bogey man, an exaggeration. And in political terms it probably is a bit. But the philosophy that underpins these developments is clear: the objective world does not exist and reason is not needed, is in fact pernicious. When there is no Truth there are truths, and you can have your own version of the truth. Radical subjectivity takes over. Reason itself becomes subjective.

It matters what you think about the objectivity of the chair for more reasons than whether you think your racist parents ought to be more woke. Suddenly you realise that not caring about whether the chair is real is politically important in a big, long-term way. It's about the fundamental beliefs of your civilisation. Are we pursuing the truth or not?

Denying universality is about so much more than being sensitive to the existence of other cultures and experiences. This is a political mission that comes out of two hundred years of anti-Enlightenment thinking. Plenty of intelligent people are signing on to the noble idea that we should reject the lazy assumption that the white male experience is universal, and then getting pulled along on the tide of the latest phase of the Counter-Enlightenment because they are epistemologically complacent.

Here's the thing. I don't think Hume was wrong. I am not a "naive realist". And it's important to say here that this doesn't oblige me to vote Trump and support everything about the radical subjectivity of modern politics. I am marginally on the woke side of the Conservative-Woke axis here.

Just like Hume was sceptical about how well we know reality, but not part of the Counter Enlightenment, you too can be a bit woke and a bit non-woke, and accept that naive realism is probably not an option.

But I don't think the inevitable outcome of that is the trail that Western Philosophy went down, leading up to the point where Bertrand Russell was pacing up and down a hotel room in agonies because he had no way of proving to Wittgenstein that even a piece of paper with three dots on it could be proved to exist in the world.

Philosophers have taken an all or nothing approach to this question, hence the Enlightenment/Counter Enlightenment division. If we cannot know reality with certainty the philosophers said, then reason and logic are out of the window.

But that isn't how reality works. Yes, truth is greyscale not black and white, but that's more a measure of our inability to find it and measure it. What amazing certainty we have had that we were right. Just because we cannot be sure (per Hume) we know the world exactly as it is, doesn't mean we have to deny it.

We need a new epistemology based on the probability of reality. As Scott Alexander says about Bayesian epistemology, "it concedes that you are not certain of any of your beliefs. But it also concedes that you are not in a position of global doubt, and that you can update your beliefs using evidence."

This is Bayesian epistemology. Thomas Bayes worked out his theory of probability by thinking about billiard balls. Imagine standing with your back to a billiard table and throwing a ball over your shoulder. How do you know where it lands?

The postmodernists, following Kant, Hegel and all those other crones, say you don't know. And that's the end of it. Bayes says, if you repeat the exercise many many times and record where the ball falls you will be able to approximate with some certainty where the original ball landed.

Bayes' idea is that you can take a prior piece of knowledge and update how accurate you think that is based on new information. This is the way Turing broke enigma and Amazon created the algorithm that recommends books to you. It's also the basis of spam filters in email.

Roughly speaking, it works like this. The spam filter assumes it's 50/50 whether an email with words like enlargement or discount in are spam. But then it sees more emails. The ones with those words keep getting marked as spam. And so it moves to 60/40. And then more emails come in with those words... and so on. Eventually you train the filter to get to a point where almost no spam arrives in the inbox and no real mail arrives in the spam. It can be astonishingly accurate.

The real problem with spam filters is false positives. Real emails that get called spam. It's a small scale example of how well we can know the world. If the system cannot recognise the difference between discount enlargement and the an special offer for bigger shrubs, reality is a long way away. (For more on spam filters, read Paul Graham.)

False positives are a big issue. Let's say you get a test for cancer and it comes back positive. Do you have cancer? Well, it depends on how often the test reports a false positive i.e. sometimes the test says it is cancer, when it is not cancer.

Look at this example, from Less Wrong:

100 out of 10,000 women at age forty who participate in routine screening have breast cancer. 80 of every 100 women with breast cancer will get a positive mammogram. 950 out of 9,900 women without breast cancer will also get a positive mammogram. If 10,000 women in this age group undergo a routine screening, about what fraction of women with positive mammograms will actually have breast cancer?

The correct answer is 7.8%, obtained as follows: Out of 10,000 women, 100 have breast cancer; 80 of those 100 have positive mammograms. From the same 10,000 women, 9,900 will not have breast cancer and of those 9,900 women, 950 will also get positive mammograms. This makes the total number of women with positive mammograms 950 + 80 or 1,030. Of those 1,030 women with positive mammograms, 80 will have cancer. Expressed as a proportion, this is 80/1,030 or 0.07767 or 7.8%.

This is Bayesian thinking because we have the prior information about how many people have cancer, how the tests work etc and then we update that to answer the question, 'What is the chance she has cancer give she has had a positive test.' It's the same as throwing balls behind us onto a table only with better data.

As they say at Less Wrong, 'the final answer always depends on the original fraction of women with breast cancer.' i.e. the answer will change as the data changes. This is about the chance of something being true. The test doesn't prove anything for certain. It just changes the chances of it being true. It updates the probability.

Here's a more anecdotal example from Scott Alexander...

I remember a rotation in medical school. I and a few other students were in a psychiatric hospital, discussing with a senior psychiatrist whether to involuntarily commit a man who had made some comments which sort of kind of sounded maybe suicidal. I took the opposing position: “In context, he’s upset but clearly not at any immediate risk of killing himself.” One of the other students took the opposite side: “If there’s any chance he might shoot himself, it would be irresponsible to leave him untreated.” This annoyed me. “There’s “some chance” you might shoot yourself. Where do we draw the line?” The other student just laughed. “No, we’re being serious here, and if you’re not totally certain the guy is safe, he needs to be committed.”

... I should add that the senior psychiatrist then stopped the discussion, backed me up, and explained the basics of probability theory.

Really, most of our decisions are like this. Something with one in a million odds will happen to you once every three years. Trump won the election with a probability of winning of 29%. Shit happens, as they say.

But there's the key lesson. Just because things are not certain does not mean they are unknowable. The Enlightenment project should not be abandoned, just updated. Newton wasn't correct in the final estimate, but he was close enough that his calculations can be used to launch rockets into space. We cannot be certain about reality, but that doesn't mean we needed postmodernism and all its attendant horrors.

This sort of Bayesian epistemology is a well developed idea among modern academic philosophers. Those of you with more smarts than me will be able to understand this summary. But in Tim Lewens overview of the Philosophy of Science, it is not mentioned. A basic understanding of probability eludes most graduates, including very often doctors who are asked to work out the probabilities involved in the cancer example above.

During COVID it was clear that most people simply didn't understand the idea of probability. COVID wasn't real to them until, very suddenly it was. Similarly, lots of otherwise supposedly intelligent people mocked 538 for giving Trump a 30% chance of winning the morning after the election.

This has an impact on politics. And not just in the obvious ways in which post-modernism has screwed up our ability to deal with the world as we find it. We get the politicians we deserve. And right now we are all involved in a political culture where everyone is appalled about the fact that people only want to deal with their version of reality, but none of us will do the hard thinking about how the epistemological problem of how much we can know about the world.

Rather than toss billiard balls behind our back and make empirical decisions about the world we stick to pro-woke or anti-woke positions. We are part of the smart people who are against Fake News and Brexit or we are racists morons who were duped. Journalists think it is acceptable to draw their salary to be pro- or anti-mask rather than try and write articles that tell us what we can know right now.

It's astonishing how well all of this is understood by people in tech, blogging, certain parts of Twitter, former libertarians, some economists, and just how much it is ignored by many journalists, writers, 'thinkers', and other elites.

If you object to the overly polarised, obnoxious political culture, spend more time thinking about why the chair you are sitting on might not be real. It will probably do us all some good.