% fortune -ae paul murphy

So where are they?

According to legend Enrico Fermi asked, in a discussion about extra terrestrials with a group of colleagues from the Manhattan projects, "So? Where is everybody?" Meaning, of course, that if there really are millions of intelligent species in the galaxy some of them should have stopped by to chat - prime directive or no prime directive.

The high IQ societies base their categorisations on how many standard deviations members are from the mean on measures relating to intelligence. It's easy to argue that intelligence itself hasn't been defined and therefore that the tests don't measure a defined quantity, but they do measure something that lets the members categorise themselves and others.

In terms of the traditional IQ scores used in schools the entry point for these societies is somewhere around a test score of 135 on any of the widely accepted tests. Roughly 2.3% of the population meets this criterion but things get a lot weirder as you try to go up the scale.

First of all, the scale itself is a matter of hot dispute and so are all of the proposed measures. It's not obvious, in fact, that either tail of the normal curve actually fits anything in this context, but there's a rough rule of thumb non participants like you and me can use. Basically, you can assume that roughly 0.3% of the population would hit the next to top rung on any sufficiently scalable test and roughly one person in a million would max out such a test.

That's around 300 Americans in the top bracket, along with nearly 9,000 at the level just below them. Since those numbers are percentage based, the absolute numbers are much larger in countries with bigger populations. China, for example, should have at least 1,300 super geniuses and nearly 40,000 "ordinary" geniuses.

So where are they?

I've met a lot of people in technology, but exactly one whom I'd guess belongs to that one in a million category and perhaps a dozen or so who I'd guess would test in the four sigma group. Look around you, do you see 9,300 people whose obvious intelligence makes you think they're at least four sigmas off the mean? I don't, and yet, what we use as a measure of intelligence is relative to population norms, not absolute, and so those people must exist.

If there were a thousand fields of human intellectual striving, less than one third should be dominated by Americans - but the real number, although unknown, is almost certainly considerably higher than that.

Worldwide there should be about 6,000 people in the super genius category -about six per field if there were a thousand fields and the distribution was purely random. There aren't a thousand, and the distribution is't random, but even heavily overweighted fields like physics and mathematics don't make their numbers: there are two or three widely recognized top people, and perhaps thirty to forty juniors out of whom another two or three might make eventually make the top rung. Since that's not remotely enough, and they're certainly not working in technolgy either, where are they?


Paul Murphy wrote and published The Unix Guide to Defenestration. Murphy is a 25-year veteran of the I.T. consulting industry, specialising in Unix and Unix-related management issues.