Highly informative and useful
In our day and age, you can't swing a chimpanzee without hitting some pseudo-intellectual spouting theories about the evolutionary origin of human mental abilities. There are many ways to respond to such nonsense. Stephen Jay Gould thinks that the best one is to learn history. The same ideas that we hear now have been advanced many times before, only in slightly different forms. While previous theories now look ludicrous to us, it's worth remembering how they dominated scientific thinking at the time. Gould takes us on a tour, showing where the bad theories came from and how they were eventually defeated.

Craniometry was all the rage in the mid 1800's, when scientists were certain that intelligence could be measured by skull sizes. Whites had the the biggest skulls, blacks had the smallest, and tada! Out came scientific proof to justify social prejudices. While other historians have looked at the background of the studies, only Gould has actually worked through the data. He finds quite a bit of fakery going on. 19th century scientists would omit some data and magnify others in order to make their point.

Starting in the 20th century, society switched to intelligence tests as a means to evaluate individuals and groups. While these have become ubiquitous over the past hundred years, few people know the history behind them. Binet, the French inventor of IQ tests, explicitly rejected the idea that they numerically measured intelligence, but his warnings were ignored. When the tests were adopted in America, they were used to justify racist policies and immigration restrictions, yet that all occurred merely on the say-so of the test designers. As usual, there was statistical hanky-panky going on behind the scenes.

Beyond that, however, lurks the question of whether measuring "intelligence" at all is legitimate. We take it for granted that every human being has a fixed level of intelligence, which determines their ability to complete any mental task. But is this really true? Gould shows us the various statistical techniques used to analyze data from intelligence tests. The major one, factor analysis, is guaranteed to produce a strong correlation between mental tasks, regardless of whether such a correlation is justified. Alternate approaches to statistics could have provided better data, but that isn't what some of the leading experts really wanted.

"The Mismeasure of Man" is at once a thorough and hard-hitting book, and also a surprisingly clear one. The final chapters on statistics do a wonderful job of making the concepts clear, even to beginners. This is a book that should be used in introductory statistics classes, and in any seminar on the philosophy of science. By showing how unspoken prejudices can underlie seeming rock-solid data, Gould teaches us to look with skepticism on anyone who claims to have the human mind all figured out.

Dishonest
Gould is a talented writer but not a scientifically impartial one. This book has more factual errors per page than any book I've ever read.

Misrepresents the literature
While the nonscientific reviews of The Mismeasure of Man were almost uniformly laudatory, the reviews in the scientific journals were almost all highly critical (Davis, Bernard D. (1983). Neo-Lysenkoism, IQ, and the press. The Public Interest, 74, 41-59).

Overall, Gould provides an interesting history, but he ignores or misrepresents a considerable amount of the literature. Gould's research on the history of craniometry is interesting and possibly valuable for historians of science. His account of the history of mental testing, however, seems biased, and crafted in such a way as to prejudice the general public and even some scientists against almost any research concerning human cognitive abilities.

Gould spends a lot of time attacking old testing methods while studiously avoiding more sophisticated tests that strongly predict academic performance (and that the army still uses). For instance, East Asians (eg Japanese & Chinese tend to do very well on the non-verbal section of the tests, which is consistent with their above average performance in math/science subjects - see Dan Seligman's "A Question of Intelligence").

Gould also says "Thurstone dispersed g as an illusion" but this is horrendously misleading (see John Carroll's review Intelligence 21, 121-134 (1995), (also, Jensen Contemporary Education Review Summer 1982, Volume 1, Number 2, pp. 121- 135.)

David J. Bartholomew, from London School of Economics, who has writtena textbook on factor analysis, also explains in "Measuring Intelligence: Facts and Fallacies" where Gould goes wrong in this area.

Gould also makes some misleading comments about the early performance of Jewish migrants on psychometric tests. Goddard never found that Jews as a group did poorly, and there is no evidence the tests were used in passing the 1924 Immigration Act (see, Franz Samelson (1975, 1982), Snyderman & Herrnstein 1983).

Gould states that Morton "doctored" his collection of results on cranial size, but J. S. Michael (1988) remeasured a random sample of the Morton collection he found that very few errors had been made, and that these were not in the direction that Gould had asserted.

He attacks Cyril Burt for fabricating his twin studies, but books since Gould's first edition came out have vindicated Burt (Joynson (1988) and the other by Ronald Fletcher (1991). Further, twin studies since show average heritability from these studies of 75%, almost the same as Burts supposedly 'faked' heritability of 77%.

So, I guess my message is take Gould's book with a large grain of salt!

Mismeasure of Man
This is a great book and I use it as a textbook in my classes at the university. It is really a refute of the Bell Curve and Gould does a great job in presenting the historical facts that make us question the pervasive uses of IQ testing.

Cogent Analysis on the Misuses of Intelligence Testing
Regarding Stephen Jay Gould's book "The Mismeasure of Man," it seems to me that a recurring theme can be found in many of the negative reviews. The theme is a variation of the claim: "Gould allows his ideology to get in the way of his analysis." Putting aside, for the moment, the question of whether or not that particular criticism has any basis in fact, I find it remarkable that the progenitors of such a claim do not (will not?) consider the possibility that the scientists, scholars and social scientists who hold views antithetical to those of Gould's--e.g. intelligence is largely genetic and heritable; the gap in I.Q. test scores between whites and blacks is due to innate genetic differences--may be guilty of committing the same malfeasance for which Gould is being accused.

Moreover, one has to wonder if the prime reason for all the strong negative criticism is not necessarily a general disagreement with the printed facts, but rather a personal, visceral rejection of the perceived incompatibility between the conclusions Gould recovers, on the one hand, and the chosen ideology of the critical reviewers themselves, on the other.

Given the plausibility of such a scenario, I believe a healthy dose of introspection and sincerity is in order, lest one proceeds to "cast the first stone."