Brandon Keim, Wired Science, June 14, 2011
In the rush to prove bias in a scientist who erroneously used skull-size measurements to demonstrate racial differences, the great historian Stephen Jay Gould may have succumbed to bias himself.
The argument centers on the work of Samuel Morton, who rose to 19th century acclaim by rigorously measuring the volume of human skulls. In those pre-Darwin days, he was looking for evidence that God created the races separately, though his findings that Caucasians had the highest average volume were also interpreted as evidence of their cognitive superiority.
Indeed, it would seem that Gould was guilty of at least one accusation he made against Morton’s methods. He omitted measurement of Native American skulls that would have altered his racial averages in unpalatable ways. As for the charge that Morton, who measured volume by packing skulls with mustard seed or buckshot, packed Caucasian skulls extra-tight, there was no evidence. And once mathematical errors were corrected in Gould’s own cherry-picked dataset, they actually resembled Morton’s supposed racial hierarchy more closely than Morton’s own results.
To be certain, Lewis and DeGusta don’t write that the scientific method can shield interpretations and assumptions from cultural biases. Those are a different matter altogether, and Morton’s work was full of them: that humanity was created in one divine swoop a few thousand years ago, that 19th century racial categories were real and fixed, that between-group cranial differences were more significant than within-group. (That Morton believed cranial volume differences represented cognitive variation is now doubted, but many other researchers did. Such features are now recognized as physiological adaptations to climate, with no cognitive implications.) But as far as the data went, Morton was honest.