Intelligence

Feb 6, 2020 15:57 · 1234 words · 6 minute read rants

[//]: # I should make this shorter. Instead of talking about specific skills, talk about broad skills. There was a good section in this article about people who love “facts and logic”.

Another common characteristic of these “logickier than thou” movements is a narrow focus on the type of skill that can be classed as “intelligence.” Affinity for things like social interaction, languages, or the arts (or at least certain types of art) often don’t get a look-in. Everything must be reducible to numbers, hence the typical logic lover’s obsession with IQ.

In The Mismeasure of Man, one of the most well-known critiques of intelligence research, Stephen Jay Gould notes the dangers of scientists’ bias toward reification — the desire to find a definitive thing that is intelligence — and quantification, the desire to slap numbers on stuff. While this is understandable to an extent — things and numbers are easy to understand at-a-glance — Gould warns that this has led to bad science and perverse outcomes in the past, and threatens to mislead us into poor understandings of intelligence, at the expense of nuance and complexity. This is all of little concern to the logic lover, who wishes not to understand, but to use again and again their favorite magic words, as a shield against criticism and as a weapon against others.

Often while talking to people I hear something called intelligence being mentioned. When asked, people explain that intelligence isn’t about being an expert in any particular domain, rather it’s an intrinsic quality that enables people to learn something completely new quickly.

Let’s assume for a second that this is true. There is something that allows people to ramp up quickly in a skill that they previously never had. I made a list of 30 common skills (that at least tens or hundreds of millions can perform well). My claim - most people are good at some of these and struggle with others.

High scores on an IQ test
Learning a new language
Learning a new programming language
Learning badminton
Learning table tennis
Learning chemistry
Learning to cook
Learning a new artistic theory
Learning a new political theory
Learning a new scientific theory
Learning how to motivate yourself
Learning how to motivate others
Learning to explain/communicate ideas in a 1:1 setting
Learning to explain/communcate ideas to a large group
Learning how to tell interesting stories
Learning how to write interesting stories
Learning to paint
Learning a musical instrument
Learning to sing
Learning to dance
Learning to navigate office politics
Learning outdoor survival skills
Learning to provide basic medical care
Learning to hunt with bow+arrow, spear or rifle
Learning to garden
Learning to manage a small farm
Learning to manage a large farm
Learning to operate a small business
Learning to operate a large business
Learning to manage money and balance accounts

I find it difficult to imagine an intelligent (by any measure) person who could pick up all or even most of these skills to a reasonable level of competence. Even skills that appear related are sometimes difficult to pick up. For example, a person could be adept at operating a small business but be stumped by the complexities of a large business and vice versa. Personally I could learn a new programming language in a few days but take 10 years to learn a new language. I have friends who did great at high school chemistry lab but struggle to cook instant noodles. A person could win a Nobel Prize in economics, indicating a deep analytical mind and experience analysing human behaviour and yet fall prey to internet scams.

If true “general” intelligence existed and was widespread, we’d see super humans good at everything. But folks like Lin Manuel Miranda are the exception, not common. And even with LMM, we haven’t seen him try his hand at science or farming.

Of course it is possible to identify a subset of these skills and say that “these are the important skills in 2020. The others are too niche to matter.” and then crown the people who have this subset as the truly intelligent. But that is a product of luck - it so happens that you don’t need to learn how to operate a farm in 2020 because there are relatively few farming jobs. You don’t need to learn outdoor survival skills because we’re sedentary folks living in comfortable homes. But if the same “intelligent” person had been born in the 1600s they would have been branded “unintelligent” because they couldn’t pick up the most basic skills required to succeed at the time. Hunting is a niche hobby in 2020 but being a good hunter would have been the difference between life and death in 1620.

A person’s success in these arenas is therefore unlikely to be tied to a “general intelligence” that is hard to measure in any case. My hypothesis - Their access to opportunities, training, mentorship and past experience doing an identical task coupled with their internal motivation (driven by necessity of current circumstances) and available time are better predictors of success than their “intelligence”.

Scalar thinking

I think there is a temptation to reduce people’s ability to a single measurable quantity. A single number that tells us if they’re intelligent enough to pick up any skill we ask of them. But this temptation isn’t limited only to measuring humans. We’d like to rate everything we see, preferably out of 5 stars - what we eat, what we buy online, where we go, who we vote for. If we need to make a decision, we’d ask around and average the ratings of others.

But I’d argue that in the real world, reducing things to a single number doesn’t give us the whole picture. Suppose we asked if McDonald’s is good or bad, most folks would say that it’s either “good” for society or that it’s “bad” or somewhere in between. But I’d argue that the question itself is wrong. A better approach would be “tell me everything you know about McDonald’s” and figure out ways to influence a change in the parts we don’t like. McDonald’s shares at least some of the blame for the epidemic of obesity, cholesterol and diabetes. And yet McDonald’s is a great place for building community, a haven for low income folks to charge their phones, access Wi-FI and fill up their stomachs for very little money. Therefore rating McDonald’s as “2.5 stars out of 5”. doesn’t make sense. A step in the right direction is keeping the good stuff, and maybe adding a sugar tax.

If we had to apply the same approach to people as we did to McDonald’s just now, we’d realise that people’s existing skills are uncorrelated and there isn’t one single factor that makes them able to do all of those things. People are too complex to reduce to a single number, whether that’s IQ or anything else.

Addendum

The list of skills is loosely inspired by this quote.

A human being should be able to change a diaper, plan an invasion, butcher a hog, conn a ship, design a building, write a sonnet, balance accounts, build a wall, set a bone, comfort the dying, take orders, give orders, cooperate, act alone, solve equations, analyze a new problem, pitch manure, program a computer, cook a tasty meal, fight efficiently, die gallantly. Specialization is for insects. ― Robert A. Heinlein