Larry Summers, 2013

Larry Summers, 2013

A friend once said to me “once you’re at a certain level of intelligence, most people you meet are either about as smart as you, or stupider.”

I’m at that level of intelligence, I suspect many of my readers are as well. If I go into a 10,000 person organization which doesn’t select primarily for intelligence, I expect to either be the smartest person in the room, or as smart as the smartest person in the room. In an org that does select for intelligence, I still expect to be able to keep up, and to be smarter than most, even if they know more about the subject than I do. (Plus, lots of very high IQ people have terrible intellectual judgment).

Divide intelligence into three parts, (yes, you can divide other ways):

1) processing power and pattern recognition (measured pretty well by IQ)

2) Creativity

3) Judgment

A lot of people only have the first, they are very smart ordinary people, they will get to the same solution a modestly bright person would, just a heck of a lot faster. The folks who put up their hands first in class, whose self-worth is based around .


(I am fundraising to determine how much I’ll write this year.  If you value my writing, and want more of it, please consider donating.)


High IQ people without 2 or 3 and preferably both, are extraordinarily dangerous if the problem isn’t straightforward. They are the brilliant people who can completely fuck things up. Think Larry Summers — he really is VERY high IQ, I know people who know him. Brad DeLong has very little of either, though he’s very very smart. (He’s very good when his emotions aren’t involved, his historical economic work is excellent). is much more common than and .

It isn’t primarily intelligence based, but empathy also has a multiplicative effect in certain circumstances.

Around about 4 standard deviations IQ starts to go really off tracks without , because at that IQ level people can make connections between almost anything, the pattern recognition is in overdrive.

To use a metaphor, think of processing power and pattern recognition as the engine of a motorcycle.  Think of creativity and judgment as the rider.  In a straightaway, powering down the highway, no other vehicles on the road, what matters it the engine.  As long as the rider can stay on the bike, the guy with the highest IQ will win any race.

But the more difficult the road conditions, or when you go off road, the more the rider matters.  The guy with the big motor, faced with erratic drivers and lousy weather is likely to get himself, and possibly others, killed.  The good rider will make it thru.

Learning how to think is, in many ways, more important than raw processing power.  The raw processing power will hold you back (to an extent, there are accounts of people raising their IQ by over a standard deviation thru concentrated intellectual effort), but too much power and too little judgment will get you killed, and too much processing power and no creativity will just get you where everyone else would have gone, but faster. Better hope that’s the best place to go.

(Adapted from a comment from 2013.)