Barriers to cognitive diversity

Cognitive diversity is a good thing. We need it if we are to avoid the pitfalls of groupthink, and because – as Hayek said – adequate knowledge of a complex world cannot be known to a single mind. I therefore had sympathy with Dominic Cummings call for “weirdos and misfits” to join Number 10.

And yet one of the first such hires has ended in failure with Andrew Sabisky’s departure following a row about his support for eugenics.

Which poses a question. Why is it so hard to achieve cognitive diversity?

In Cummings’ case one problem is simply the poverty of thought on the right. Intelligent thinking does not come from unaided individuals, but from traditions. And these traditions have withered on the right: free market economics, for example, has pretty much died. Without such moorings, original thinking can be crankery.

A comparison with Thatcher will demonstrate this. Yes, some of her coterie flirted with eugenics: Keith Joseph fretted that “the balance of our human stock is threatened”. But such ideas were sidelined because Thatcherites had other, fish to fry. They were trying to implement an agenda inspired by Friedman, Hayek and Popper. It is a long time since the Tories had influences of similar eminence.

Cummings, though, has another problem. Pretty much nobody of genuine ability would want to take an insecure job in a city that is too expensive to live in. And nobody in their right mind wants a boss who says things like “I’ll see half of you next Friday” or “I’ll bin you within weeks if you don’t fit.” What’s remarkable, then, is not that Cummings should attract the overconfident and unemployable, but that he should have any chance at all of hiring anybody who isn’t.

But, but, but. These are not the only barriers to cognitive diversity. Ours is an age of limited intellectual breadth. Many worthwhile perspectives are rarely heard: free market pessimism; small state Keynesianism, analytical Marxism and so on. There is poverty of thought on the left as well as right – especially among some of those with high profiles.

One reason for this lies with our media. The BBC’s “grid” squeezes out too many voices and nuance, and reduces issues to simple tribal binaries. As an anonymous BBC employee has written:

Broadly speaking, there were only two types of guests for producers to choose from: Brexiteer evangelists or slick and power-hungry London Remainers. That the discussion surrounding Brexit should be conducted from these two extremes has become accepted as a sort of national truth, but it certainly doesn’t reflect the real make-up of Parliament or the country.

The point applies beyond Brexit.

Academia must also take some blame here. A friend recently showed me a rejection letter from reviewer 2, who complained that he couldn’t understand a section of the paper because it was “purely descriptive.” What he meant is that there are rigid rules one must follow if one is to be published, and he was damned if he was going to understand anything outside those rules. Such performative stupidity is a great recipe for groupthink. And the time academics must spend obeying those rules is time they cannot spend in more productive thought. This is one reason (of several) why the UK has suffered the near-death of the public intellectual.

There is, I regret to say, more. Those rightists who whine about the mentality that has driven Sabisky out have a point. Too many of us are sanctimonious narcissists. One of my favourite theories here was suggested by Richard Sennett back in 1970:

The threat of being overwhelmed by difficult social interactions is dealt with by fixing a self-image in advance, by making oneself a fixed object rather than an open person liable to be touched by a social situation…The jarring elements in one’s social life can be purified out…Threatening or painful dissonances are warded off to preserve intact a clear and articulated image of oneself. (The Uses of Disorder p6, 9, 11)

He was writing about the city, but his theory applies to social media too*. And it is rank hypocrisy to think this is found only on the left. The right has its own pitchfork mobs too.

Now, some of me is relaxed about Sabisky’s departure. The problem with eugenic ideas is not that they are offensive but that they are tiresomely unoriginal, unscientific and – for all decent purposes  irrelevant. But another part of me is disquieted, because it draws attention to a big problem – that we lack both the institutions and culture which are necessary for proper cognitive diversity.

* One manifestation of the desire to ward off dissonances is, ironically, eugenicists need to purge undesirable elements from society.

Comments

[AlbertTappman]

Academia is not to blame. Your example is a classic case of the 'availability heuristic'. You are assuming your friends was a good paper, when it might simply have been a weak one, and the reviewer (likely a PhD student on a zero hours contract) was being polite.

Actually, there are many good ideas and theories and analysis out there, but most are crowded out by Whig or Tory intellectual dominance in the press and politics. And you not being aware of them does not mean they don't exist (not everything is on SSRN!)

Cummings is an example of such arch bullshit and sophistry; he says and knows very little yet people who also say and know very little thing he is knowledgeable or 'intellectual'. He was not calling for intellectual diversity, he was calling for the classic Oxbridge BS; say unoriginal and reactionary things in an 'original' fashion. Therefore, a mirror of Cummings himself, a chancer that is in line with current dominant ideology. Sabisky's appointment was not intellectual diversity in the slightest, he was another bullshitter hiding his racism and authoritarianism behind sophistry.

The question you should be asking yourself - if you value intellectual diversity - is why there is zero chance of a Marxist, Critical Theorist or Post-Colonial theorist ever getting anywhere near the government. Answer that question and you might better understand what is going on.

[Mark]

I suspect that diversity has bifurcated: the left likes diversity of identity, the right that of cognition. But the left has more cultural power, so it's outside the Overton window for the right to object if the left's variety of diversity turns up, say, a gay Muslim woman in a position of power, but remains acceptable for the left to object to the right's equivalent to result in the appointment of someone who floats ideas such as eugenics.

[Tim Worstall]

"an insecure job in a city that is too expensive to live in."

Quite so, quite so, that was my reaction to that job ad.

[Scratch]

"And it is rank hypocrisy to think this is found only on the left. The right has its own pitchfork mobs too."

Ignoring the question of whether "left" mobs are actually left, the right mobs tend not to have the backing of capital, the bourgeois, the bourgeois state, the academy, the media and the plod.

There are currently two things that make me laugh about the right, their sincere belief that this orgy of suppression is somehow unwanted by the forces of capital (one presumes that say, the transgender community must have rather more divisions than is immediately obvious for instance) and the peculiar knots they tie themselves in trying to claim that the latest excrescences of feral liberalism are in some way "Marxism."

Funnily enough, I suspect these delusions are the only things standing between them and class consciousness. If there is hope, it might lie with the "deplorables."

[Dipper]

when one person is found to have written some crass comments about race and IQ a few years ago, an someone else says biological sex is not determined at both, and the person who loses their job is the one who made the IQ comments, then we are in trouble.

Holding the view that as a white person you are on average less intelligent than Jews or Chinese is a bit niche, and i don't personally believe this race/IQ /genes stuff. but it seems if society is to change, then it requires that people be allowed to disagree with the current view.

The clear threat here is Labour, who now hold the view that they can espouse a view in complete contradiction to all the available scientific evidence and force people to comply or face complete oblivion.

[georgesdelatour]

We engage in several eugenic practices right now, and most of them probably enjoy broad public support:

1. Genetically testing the foetuses of pregnant women for Down syndrome. Currently 92% of women whose foetuses test positive go on to have an abortion. As the Wikipedia article on Down syndrome notes, “as a result, there are almost no people with Down’s in Iceland and Denmark, where screening is common”.

2. Genetically testing for Tay Sachs disease, which specifically affects Ashkenazi Jews, French Canadians and Cajuns. As the Times of Israel noted in 2017: “Thanks to technology and an aggressive screening campaign, the genetic disease is all but gone today - and carrier couples can have healthy children… Before screening, couples in which both parents were Tay-Sachs carriers “almost always stopped having children after they had one child with Tay-Sachs, for fear of having another,” Ruth Schwartz Cowan wrote it in her book “Heredity and Hope: The Case for Genetic Screening.””

3. Encouraging people intending to marry their first cousin to take a genetic test. As The Guardian’s Ian Sample reported in 2008: “Doctors should offer genetic screening to immigrant communities that encourage marriage between cousins, experts said yesterday, as new research suggested cases of rare inherited brain disorders are eight times higher among Pakistani children born to married cousins than those born to unrelated parents.”

4. Educating would-be older fathers about genetic risks. When men over 40 become fathers, there’s an increased risk to their child’s health, because their sperm usually has more deleterious mutations. The number of children born to older fathers is rising, because people are living longer and they’re starting families later.

5. Screening donors to sperm banks for genetic diseases and chromosomal abnormalities.

[the_last_name_left]

Was ever an idea more dull than racism? Crime? Race! Education? Race! And (really) based on what people look like. Tough to even think of something that could be simpler and less interesting.

[jim]

Cognitive diversity - mmmm. Perhaps what we see here is cognitive arrogance or fraud. Ostensibly Dominic etc are there to provoke whizzy ideas to push GB plc forward into sunny uplands. But therein lies a puzzle, what possible brilliant notions could Dominic et al come up with that are not instantly replicable elsewhere. Then any such brilliant notion would have be sufficiently strong to overcome the few percent GDP barrier created by Brexit. A very surprising and unlikely-to-succeed notion - or a fraud.

Then we have cognitive diversity painted as intellectual honesty. Trawl the think tank and conference circuit and you frequently come across the Cummings and Sabisky type, there is a market for that kind of thinking. Useful fools to put up at a conference to liven debate, but never let them near reality.

Then there is the uncomfortable area surrounding eugenics and intelligence. Maybe there is a very small scintilla of validity here - but it seems of no real use to us. Even if we dragged the topic into daylight for a good look at it I have the feeling such would not be worth the heartache and trouble.

There is a great deal of ruin in the world - we probably could not afford to clean it all up. Suppose in the Rochdale case that all Home Office, Social Service, Police and Council records were complete, honest and available. Suppose further that there were clear punishments all the way to the top for neglect and worse. The Rochdale scandal would not have happened but could we have afforded sorting out the mess? Realistically we pay government to do our dirty work - cheaply.

[Ralph Musgrave]

I do like Chris's claim that "free market economics, for example, has pretty much died." What - so the economy is now run by some sort of central planning committee? If so, where is this committee?

Moreover, according to the two books I read on the USSR so called planned economy, market forces actually played a much bigger role in the USSR economy than the Russian elite were prepared to admit.

[Ralph Musgrave]

The basic argument for eugenics is very reasonable. It’s that most of the random variations in human DNA are no use, and must be weeded out if the quality of the human race is not to deteriorate.

But I appreciate that the mere mention of the word causes lefties to stamp their feet and foam at the mouth nowadays. In contrast, in the 1930s and early 1950s, eugenics was accepted in left of centre circles, as explained here:

https://www.theguardian.com/commentisfree/2012/feb/17/eugenics-skeleton-rattles-loudest-closet-left

All of which goes to show that about 95% of the human race, including self-styled intellectuals, are zombies who can be made to think just about anything.

[Francis]

A highly critical review of Adam Rutherford's new book:

https://www.takimag.com/article/occams-butter-knife/

[the_last_name_left]

Ralph Musgrave: "The basic argument for eugenics is very reasonable. It’s that most of the random variations in human DNA are no use, and must be weeded out if the quality of the human race is not to deteriorate."

-------------

That is absolutely not "the basic argument for eugenics".

Random variations in DNA that is "of no use" can have no consequence, so could never be of concern.

[Ralph Musgrave]

The last name left,

So if a random variation results in someone having a squint or one leg longer than the other, you think that's of no "concern"?

[Dipper]

intelligence isn't like height. There isn't much evidence for a measurable thing called intelligence.

I'm reminded of something a Nobel prize winner said. He said some women seemed to be very keen on being impregnated with Nobel Prize winners sperm so that they can give birth to someone super intelligent. But his father was a carpenter. So if you want to raise a nobel prize winner, better to be impregnated by a carpenter.

[Ralph Musgrave]

Dipper,

So are you saying we can't be sure whether Einstein had any more intelligence than someone who can'd hold down a simple minimum wage job and can't master their two times table?

[Dipper]

Ralph Musgrave

No. Your answer assumes there is such a thing as intelligence and that I am saying Einstein didn't have much. I said there isn't a thing called intelligence.

Einstein had an ability to derive theories that explained observed phenomena that prior theories could not explain. That's it. You can call it whatever you like, but society in general assumes that this is evidence of him having a particular general property called 'intelligence'. I don't see that it is a useful statement to make that inference.

To give an an analogy, let's just assume we could not measure height directly, but instead could infer height from things we observes that we believe only tall people could do. We could measure the ability to score basketball hoops, and the ability to change lightbulbs without using a chair. What if someone could get a very high score on basketball hoops but could not change lightbulbs without using a chair? Would they be tall? In a world where we could observe height directly we might find there are people of average 'height' who can score greatly at basketball but are not tall enough to change lightbulbs, or that there are 'tall' people who can change light bulbs but still cannot score at basketball.

Once we remove a direct measurement of a property and can observe only consequences of that property, we are left wondering whether the property we are inferring is at all relevant. If you wanted someone to change lightbulbs without a chair, would you think that measuring scores of basketball were a good indicator? Why would you not just get them to change a lightbulb without a chair?

Similarly, there are people who score well on 'intelligence tests' but exhibit much stupidity in many other aspects of life, and people who don't score well on intelligence tests but who do other things that correlate well with 'intelligence' indicators but don't score well on 'intelligence' tests.

To be even more specific, I used to recruit mathematicians, and we had a numerical reasoning test under strict time conditions to filter applicants, and we stopped using it because we observed that many very good mathematicians were not scoring highly on the test. It wasn't a good measure of the particular mathematical skills we were looking for.

Do you have to be intelligent to play chess? There are many people who score well on intelligence tests and who are good at chess. There are also people who are average on intelligence tests and happen to be extremely good at chess. So, are members of that latter group 'intelligent'? Does it matter?

[Towerbridge]

Just to add to what Dipper has said and I fully agree with him/her(!)...

When people talk of IQ it is worth knowing where that system originates because it says a lot about its flaws. It was designed to evaluate whether the (then) newly formed French universal education system was performing and designed by a man who thought we could not define intelligence and that even if we could such a quality was malleable. It was then adapted by a prominent eugenicist and thereby designed to show inferiority and superiority between certain groups of people.

For a greater understanding I suggest listening to the podcast Radiolab (it is one of the biggest podcasts out there) and their mini-series "G".

[Ralph Musgrave]

Dipper,

Those who advocate IQ tests have always known that such tests are not good predictors of a person's ability at every single task that faces people in their daily lives: e.g. as you rightly say, some people with high IQs who have "much stupidity in many other aspects of life".

However the evience is that IQ TENDS TO BE RELATED to ability to perform other tasks. Ergo IQ tells us something.

[Dipper]

Ralph Musgrave

Tends to be related? What kind of an excuse is that?

There's an awful lot of brain power being spent on analysing something that 'tends to be related'.

How about trying to define 'athleticism' on a single index and seeing how that works in its predictive ability. Rather than, say people being good at performing specific athletic activities.

[Ralph Musgrave]

Dipper,

What I mean by "tends to be related" should be obvious. But what I meant was that there is a PROVEN relationship between IQ and ability to perform other tasks. The relationship is particularly close when it comes to maths and physics, but also undeniable when it comes to foreign languages, music, mugging up on history and so on.

[Laurence]

The BBC showed a two-part series on Eugenics last year. The first programme is available on YouTube at: https://www.youtube.com/watch?v=paFuyPV8HJU

Interestingly, much support for eugenics these days comes from atheists, which confirms the aphorism that when you stop believing in God, you don't believe in nothing you believe in anything.

Also believers in the meaningfulness of IQ tests need to explain the Flynn effect (an increase in IQ of about 10 points per generation).

[georgesdelatour]

@ Laurence

There’s an important distinction between coercive eugenics and non-coercive eugenics. It’s reasonable to educate people from the UK’s Mirpuri community about the genetic risks associated with first cousin marriage; risks which are especially high for them because their ancestors have been marrying consanguineously for a long time. This doesn’t mean it would be okay to either force-sterilise people or force them to have abortions. And if, in spite of everything, they give birth to children with recessive conditions, we should still give those children all the care we can.

On the whole, I think the Jewish community’s approach to Tay-Sachs has been thoughtful and effective at reducing the incidence of the condition. I don’t see any reason to object to it.

Sperm banks screening donors for Mendelian disorders does not lead society inexorably to the Nazi Aktion T4 program.

[Dipper]

@ georgesdelatour "There’s an important distinction between coercive eugenics and non-coercive eugenics."

well ... for there to be a difference, first you have to have a theory of eugenics.

Toby Young raises this, that working class parents could be offered screening to select higher IQ children. Needless to say, I don't believe in it.

The ability to screen for Tay-Sachs, Huntingdon's, etc shows the debate around IQ for what it is. A debate about nothing. It's just too diffuse a concept. Owes more to nineteenth century phrenology than genetics.

If I was to offer a personal insight, it is that I've come across lots of people in the bottom quartile of society. Not many of them are obviously low intelligence, and quite a few are clearly smart. But people seem to end up in dire straights for two main reasons; addiction, and personalities somewhere on the Aspergers/Autistic spectrum hence not able to understand/do the things that lead to more successful lives in our society. So if people want to make the world a better place through finding genetic explanations for personality traits, they'd be better off finding out whether addiction is hereditary, or whether Aspergers is hereditary, rather than worrying about just how much more intelligent the Chines are compared with Caucasians.

[georgesdelatour]

@Dipper

“… there isn’t a thing called intelligence.”

From the Wikipedia entry on John von Neumann:

“… Von Neumann was a child prodigy. When he was six years old, he could divide two eight-digit numbers in his head and could converse in Ancient Greek… By the age of eight, von Neumann was familiar with differential and integral calculus, but he was particularly interested in history. He read his way through Wilhelm Oncken's 46-volume Allgemeine Geschichte in Einzeldarstellungen… At the age of 15, he began to study advanced calculus under the renowned analyst Gábor Szegő. On their first meeting, Szegő was so astounded with the boy's mathematical talent that he was brought to tears. Some of von Neumann's instant solutions to the problems that Szegő posed in calculus are sketched out on his father's stationery and are still on display at the von Neumann archive in Budapest. By the age of 19, von Neumann had published two major mathematical papers, the second of which gave the modern definition of ordinal numbers, which superseded Georg Cantor's definition…Von Neumann held a lifelong passion for ancient history, being renowned for his prodigious historical knowledge. A professor of Byzantine history at Princeton once said that von Neumann had greater expertise in Byzantine history than he did…”

“He made major contributions to a number of fields, including mathematics (foundations of mathematics, functional analysis, ergodic theory, representation theory, operator algebras, geometry, topology, and numerical analysis), physics (quantum mechanics, hydrodynamics, and quantum statistical mechanics), economics (game theory), computing (Von Neumann architecture, linear programming, self-replicating machines, stochastic computing), and statistics. He was a pioneer of the application of operator theory to quantum mechanics in the development of functional analysis, and a key figure in the development of game theory and the concepts of cellular automata, the universal constructor and the digital computer. He published over 150 papers in his life: about 60 in pure mathematics, 60 in applied mathematics, 20 in physics, and the remainder on special mathematical subjects or non-mathematical ones. His last work, an unfinished manuscript written while he was in hospital, was later published in book form as The Computer and the Brain.”

“His analysis of the structure of self-replication preceded the discovery of the structure of DNA…”

One other point. You say there are plenty of good mathematicians with poor numerical reasoning. That feels like saying there are plenty of good strikers who don’t score any goals. Accounts of the young Gauss and the young Ramanujan portray them both as having prodigious numerical reasoning. Are there any Fields Medalists you could cite as examples of good at maths but bad at reasoning with numbers?

[Dipper]

@ georgesdelatour.

Einstein did hardly any of what Von Neumann did in his early age. He was a notoriously late developer. Clearly Einstein was not intelligent.

Do we know what Von Neumann's recorded IQ was? Assuming it was enormously high, does the fact that some people score highly on this test convey any information on the significance of lower, or more average, scores?

Are you saying the people with PhD's in differential geometry from top university who didn't score particularly well on numerical reasoning tests were not in fact good at maths? My conclusion was that the mathematical skills needed to save deep complex problems are not the same as skills needed to answer a lot of tricky questions quickly.

[Laurence]

@georgesdellatour

I've no argument with education about the downsides to cousin marriage or Tay-Sachs, but this isn't eugenics, because you are not removing the genes from the population only masking their effects. A better example would be pre-implantation genetic testing, which I am mostly* OK with as long as it is a decision made by the parents and not forced on them by the State.

On geniuses, it is generally understood that there are two distinct types: those who would score very high on IQ tests (like Von Neumann) and those whose IQ was tested and was not exceptional (like Feynman).

This is what the mathematician Mark Kac said about Feynman:

"There are two kinds of geniuses: the ‘ordinary’ and the ‘magicians.’ An ordinary genius is a fellow whom you and I would be just as good as, if we were only many times better. There is no mystery as to how his mind works. Once we understand what they’ve done, we feel certain that we, too, could have done it. It is different with the magicians... Feynman is a magician of the highest caliber."

If you focus only on IQ you will miss the 'magicians'.

*Sex selection being one of the less desirable uses of PIGT.