What’s the point of arguing with an idiot?

The question I had was: who disagrees with that? What serious, intelligent person would say that they want to keep cultures entirely separate? Are you, when you stop to think about it, really arguing against the strongest case your opponents could make? That was the first thing I was taught, nearly two decades ago, as part of my philosophy undergraduate degree: look for your opponents’ strongest arguments, anticipate their objections. Make their case for them, so that when you knock it down, it stays down.

I don’t want to pick on Young and Harris; what they were doing was completely standard behaviour for all of us. We all tend to assume our opponents are stupid or malign; we tend to pick on the easiest things to attack rather than the hardest. It’s not deliberate, it’s just how we’re built. It’s easy to assume someone worried about gender self-identification is simply bigoted; harder to talk about how we’re going to make society easier for trans people while also protecting female-only spaces such as women’s prisons. It’s easy to say that Tories cut spending because they hate the poor; harder to talk about how we’re going to pay for health and social care in an ageing society without raising taxes on already struggling people.

But we should try to be better. We should try to understand what we’re arguing against before we argue against it.

Alan Turing, the pioneer of computing, once proposed a test. He had, it seems, grown understandably bored of debates about whether it was possible for a ‘machine’ to ‘think’. Instead of the interminable philosophising, he said, which gets bogged down so rapidly in the definitions of ‘machine’ and ‘think’, let’s set up a test. The test was: put a human and a computer in separate rooms. Get another human in another room to ask them both questions, which they answer by text. If the interrogator can’t reliably tell which is which, then to all intents and purposes, the computer can think. Leave the “yes, but can they think?” debate to stoned undergraduates.

Bryan Caplan, an American economist, proposed a similar test in 2011. His idea was an ideological Turing test.

Caplan was responding to the New York Times writer and economist Paul Krugman, who argued that liberal economists understand what conservative and libertarian economists are proposing, but not vice versa.

Well, is that true, wondered Caplan. How about you test it? Take some liberal economics PhDs, and some conservative (or libertarian) economics PhDs. Get the liberal ones to pretend to be conservative and the conservative ones to pretend to be liberal, have some observers ask them all questions, and see which group does better at fooling the judges.

Just as with Turing, it sidesteps the philosophical “but do they understand?” question. If some neutral observers, a set of pre-appointed judges, cannot easily tell that you are faking it, then – to all intents and purposes – you can be said to understand what you’re saying.

I don’t think you can realistically set up a controlled trial before every comment piece is published. But something like that, as a thought experiment, would be really useful.

Go back to the cultural appropriation thing. If Young or Harris stopped to think about it, would they really think that anyone is seriously suggesting that cultures should be sacrosanct, or “copyrighted”? If they’d run a sort of internal ITT and imagined trying to convince people that they actually believed what they were writing, would they have really thought that this was something someone might actually say?

What offended people, as I understand it, about Jamie Oliver’s rice was that it used the name (and therefore the appeal) of an iconic part of a culture, without respecting its origins.

Stephen Bush, in his Statesman piece, drew a parallel with the ‘cultural appropriation’ by white people of the feathered war bonnets worn by Plains Indian tribesmen, which by tradition could only be worn if you had won the respect of the tribe, perhaps in combat. It is, said Bush, not unlike wearing an unearned Victoria Cross as a fashion accessory; most of us would find that distasteful.

Having then found a strong form of argument against cultural appropriation, as per the ITT, you can then try to knock that down. And Bush does, by suggesting that the term itself is too broad and loaded, and that you should instead focus on the specifics involved: that the ‘jerk rice’ thing is just, basically, rude, but borrowing from other cultures is perfectly fine if it’s done respectfully.

Bush’s article, although also to an extent arguing, like Young and Harris, against the use of the term ‘cultural appropriation’, would pass an ideological Turing test: he argues against something that I can imagine an intelligent person actually agreeing with.

There’s a problem here, of course, which is that there are billions of people in the world, and some of them really are stupid and/or malign. If I wanted to, I could easily find examples – as both Young and Harris do in their respective articles – of real people believing ridiculous or awful things, and then argue against them, instead of choosing the best and cleverest. The ideological Turing test falls down if you set the bar for your ideological AI too low. If Caplan hadn’t limited himself to economics PhDs, he could easily have found lots of people who don’t even understand their own ‘side’s’ arguments, let alone their opponents’.

I’m in danger of a no-true-Scotsman logical fallacy here: I can keep defining away bad arguments by saying “that person is too stupid, they don’t count”. And sometimes you do want to point out prominent dimwits saying dimwitted things.

But on the whole, the way to advance the national discourse and increase the sum total of human understanding is to argue against the most compelling arguments, not the least. And a good way of checking whether you’re doing that is to run a little subroutine in your mind: imagine someone who disagreed with you, someone clever and moral and thoughtful and well-informed but who has a different view of the world, arguing in favour of what you’re arguing against. It’s really hard. I am a staunch Remainer. I can’t help but assume that Brexiters are almost all stupid or self-interested. And yet I know several who are not.

The temptation, for me, is to assume that the “real reasons” people support Brexit are political ambition – as in Boris Johnson – or because they have been duped by such political charlatans. But that’s not true for everyone.

If I try to write a pro-Brexit argument that would pass an ideological Turing test, it might go something like this: the EU has become so entangled in British affairs that it is almost impossible to leave it even now. Our criminal justice system, our agricultural system, our trade and customs laws: almost no aspect of British life is untouched. If we want to gain more direct, local, democratic control over those things, we have to leave, and given that the EU, since Maastricht, has become more and more integrated, it will only get harder to leave. So it has to be now, or never; it will come at a cost, and will be complex and protracted and painful, but it will be worth it in the end.

I disagree with that. I think these gains in intangible sovereignty are not worth the economic damage leaving the EU will cause, and I think that our loss of influence means that the gains in sovereignty are largely illusory anyway. But I can imagine someone intelligent sincerely arguing it, and that gives me something worth arguing against. I think – I hope – that it would pass an ITT. It is an opponent worthy of the name.

Alternatively, we can keep arguing against half-imagined idiots and the sweepings of Twitter, because it’s easy and fun and it gets our side cheering. Furiously disagreeing with a caricature, though, rarely gets us anywhere.

---

Tom Chivers is a science writer. His second book, How to Read Numbers, is out now.

TomChivers