Gould, Science, Mistakes and Fraud
15 Aug 1996 09:58:12 -0600
Joel's been mentioning fraud in science from time to time. The NSF
publishes a list of folks caught doing that, every year. Here's an
interesting case that will never make that list. It involves Stephen Jay
Gould, Ivy Tower resident at Harvard.
In 1978 and 1981, Gould challenged the early use of morphometrics by
intelligence researchers in the last century. Much of what Gould said
seems reasonable, and on the whole he was well received. However, in the
1978 paper (page 503, ref below), Gould accused Morton (1849) of
"finagling" and "juggling" data. Morton's bias, Gould suggested, need
not have been conscious:
"Morton, measuring [cranial capacity] by seed, picks up a
threateningly large black skull, fills it lightly...Next, he
takes a distressingly small caucasian skull, shakes hard, & pushes
mightily at the foramen magnum."
In the book _Not In Our Genes_, (1984) Lewontin and friends quoted this
passage authoritatively, claiming that Gould had "exposed" intentional,
"systematic distortion of the evidence" by nineteenth century
But in 1988, a re-measurement of the skulls used in Morton's original
study found that the few, small errors Morton made were *not* in the
direction Gould had claimed. Michael (1988) concludes that Morton had
conducted himself with scientific integrity, and that Gould, charitably,
was "mistaken" in his attack.
In his book, Gould admitted to "embarrassing errors" in his 1978
calculations (Gould 1981, p. 66). Nobody used this as fodder for
discrediting Gould, as he himself would surely have done if, say, Rushton
had made the same mistakes. (It looks like the supposedly raging
political war in academia may be largely one-sided. Heh. That happens
when you fight staw men instead of real ideas, as Gould often does.)
More typically (Gould is rarely caught out and out fudging), in his
essay, "Male Nipples and Clitoral Ripples," Gould complains about
adaptationists' insufficient testing of their evolutionary hypotheses
about the origins of the female orgasm. In the same essay, Gould himself
declares without evidence (and without presenting a single testable
prediction) that "the real answer" is that female orgasm is analogous to
a male nipple: functionless baggage from early, sexually undifferentiated
embryonic development. (Evidence published since strongly suggests that
Gould was wrong.)
These examples well enough illustrate the points I wished to make:
1. hypothesis testing is not "arrogant," as has been suggested by some;
indeed, as Gould found, it can be humbling! Propagandistic prose will
never replace hypothesis testing science as an effective way to
2. Science is self-correcting. Gould screwed up, but the mistake was
caught by Michael.
Gould, 1978. Morton's ranking of races by cranial capacity. Science, 200:
Gould, 1981. The Mismeasure of Man. New York: Norton.
Michael, 1988. A new look at Morton's craniological research. Current
Anthropology, 29: 349-254.
Morton, 1849. Observations on the size of the brain in various races and
families of man. Proc. Acad. Natural Sci., Philadelphia, 4: 221-224.