4 Following

Manny Rayner's book reviews

I love reviewing books - have been doing it at Goodreads, but considering moving here.

Currently reading

The Greatest Show On Earth: The Evidence For Evolution
Richard Dawkins
R in Action
Robert Kabacoff
Fluid Concepts and Creative Analogies
Douglas R. Hofstadter
McGee on Food and Cooking: An Encyclopedia of Kitchen Science, History and Culture
Harold McGee
Epistemic Dimensions of Personhood
Simon Evnine
Pattern Recognition and Machine Learning (Information Science and Statistics)
Christopher M. Bishop
Relativity, Thermodynamics and Cosmology
Richard C. Tolman
The Cambridge Handbook of Second Language Acquisition
Julia Herschensohn, Martha Young-Scholten
The Rising Curve: Long-Term Gains In IQ and Related Measures (Apa Science Volumes) - Ulric Neisser Are people getting smarter or stupider?

This turns out to be a complicated and interesting question, and after finishing The Rising Curve, a collection of papers published in 1997, I understand the issues much better than I did when I started. As is normal for a book of this kind, the individual contributions varied widely. Some were perceptive and well written; some were dull or confused; some people were doing their best to be impartial, and others had obvious axes to grind; many of them were combinations of the above.

If you're used to wading through dull academic papers, you may want to check it out for yourself. Otherwise, let me give you the highlights. There are at least four threads, all of which are related to each other:

The Flynn Effect According to reputable, well-validated studies, IQ rose steadily over most of the 20th century. I'm surprised this isn't better known.

The Dysgenic Hypothesis IQ is largely inherited, and dumb people have more kids than smart people. Hence IQ should in theory fall. Some respectable people have believed this, but the argument is not as obvious as it looks.

Verbal intelligence is declining Although IQ is increasing, verbal intelligence, and in particular vocabulary, seem to be moving in the opposite direction.

Racial differences Black people have substantially lower average IQ than white people. Before you call me a racist, please read the rest of this review.

I'll now consider each theme separately.

Rising IQ: the Flynn Effect

It's not completely clear why, but everyone agrees that IQ scores rose steadily over most of the 20th century; they may have started rising earlier, perhaps as far back as 1850. The effect can be seen in studies carried out in at least 20 countries. The gain may be as much as 15 IQ points, one standard deviation, per generation; this would mean that an average kid ought to be smarter than 85% of his parents' contemporaries. (Didn't you just know that was true?) We don't see it directly because the tests are periodically renormalised to make the average IQ come out as 100, but when you run the same tests on people from different generations it's obvious. The data on Dutch and Israeli army recruits is particularly compelling.

A number of things are credited as possible causes: better nutrition, more years of schooling, lower family sizes and better-educated parents all seem important, and there are strong correlations between these factors and IQ scores. I'll discuss a couple of other factors later on under "Declining verbal intelligence".

The Dysgenic Hypothesis

The argument at first seems clear-cut. Intelligence is inherited, and dumb people have more kids than smart people: hence the average level of intelligence should steadily decline. People as respectable as George Bernard Shaw apparently believed this. The idea was also popularised in a witty series of SF stories by C.M. Kornbluth, of which the best known are The Marching Morons and The Little Black Bag.

But... when you do the math, it turns out that our intuitions are confused! Even if the premises of the argument are completely correct, you don't in fact get a steady decline; instead, you rapidly reach an equilibrium. The point is that intelligence is only partially inherited. Some dumb people will randomly have smart kids, and after a while it evens out. The fact that dumb people are having more kids pushes the equilibrium point a bit towards the dumb end, but not nearly as much as you'd think. The paper by Samuel Preston explaining all this is very elegant and convincing.

Declining Verbal Intelligence

Scores on standardised vocabulary tests show that the average person's vocabulary has been in decline for many decades. School textbooks have been dumbed down by as much as two grade levels since the 1940s.

I was surprised to see so little discussion of the links between this and the Flynn Effect. The psychologists measuring IQ most often use the Raven, where you have to fill in the missing item in a matrix. Here's a typical example:


Now, in theory, these tests are supposed to measure absolute, non-culture-based intelligence. Speaking as a layman, it seems like nonsense to me. A test like the Raven measures a particular kind of skill in finding visual patterns. As several of the contributors pointed out, people over at least the last 70 years have been living in a world where they have progressively received more and more complex visual stimulation (movies, TV, computers, video games, cellphones). You're constantly honing your visual problem-solving skills. The flip side is that you get less complex verbal input - in particular, people read much less. So visual skills improve, and verbal ones decline.

I don't know how to quantify it, but popular movies seem to me a clear illustration of this trend. Movies these days are visually far more sophisticated than they were 40-50 years ago, but verbally I would say they are less demanding. For example, contrast Inception, which looks like being one of this year's biggest hits, with The Sting, one of the most popular movies of the 70s. When we saw Inception the other day, my wife (mid-50s) complained she had trouble keeping up with the lightning-fast visual shifts between the different levels. When we saw The Sting on DVD a few months ago, our son (early 20s) said he couldn't keep track of the story.

Racial differences in IQ

All the authors in this collection agree that black people have, on average, lower IQ scores than white people. They then present evidence to show it's almost impossible that this can be due to genetic differences. IQs among both black and white people have both been rising steadily (the Flynn Effect), and the average IQ for black people is typically about the same as the average IQ for white people a generation earlier. There is no way that genetic factors could work this fast.

Hence, the difference must be due to negative environmental factors, which are still affecting blacks more than whites. There have been many studies carried out in the US to track the differences. The gap between blacks and whites roughly halved between 1960 and 1980, then levelled out. The most obvious explanation is changes in government policies regarding social equality, in particular desegregation and affirmative action.


I had heard about the Dysgenic Hypothesis and racial differences in IQ before, and I was relieved to see that we probably don't need to worry about either one. The Dysgenic Hypothesis appears to be a misreading of the mathematics, and racial differences in IQ are almost certainly environmental and not genetic in nature.

I was very interested to read about the Flynn Effect and the apparent shift from verbal to visual intelligence. This is a trend which could have important consequences for our future. I've just bought Flynn's new book What is Intelligence? and I'm looking forward to reading it.