Mismeasuring Skulls: New Research Resolves Historical Controversy, Shows Science Resists Bias
Jason E. Lewis, who received his Ph.D. from Anthropology this year (and is now an Assistant Instructor in the Department of Anthropology at Rutgers University), along with former Stanford Anthropology Professor David DeGusta and other colleagues, recently published a paper in the open-access journal PLoS Biology in which they resolve a famous historical controversy in physical anthropology. Samuel George Morton was a 19th century scientist famous for his measurements of human skulls. Stephen Jay Gould, the prominent evolutionary biologist and historian of science, reanalyzed Morton’s data and concluded that Morton’s bias had led him to fudge his results. Gould’s view took hold in social studies of science and Morton became a textbook example of scientific misconduct. The new research shows that Morton’s measurements and results are accurate, and in fact it was Gould who may have fudged the numbers to make his case. For more information, please see an abstract of the article below. To download a copy of the PLoS article, see: http://www.plosbiology.org/article/info%3Adoi%2F10.1371%2Fjournal.pbio.1....
Samuel George Morton (1799-1851) was a 19th century scientist famous for his measurements of human skulls. Stephen Jay Gould (1941-2002), the prominent evolutionary biologist and historian of science, reanalyzed Morton’s data and concluded that Morton’s bias had led him to fudge his results. In his prize-winning 1981 bestseller The Mismeasure of Man, Gould used the Morton case to argue that biased results were common in science because scientists are not immune to prejudice and ambition: “unconscious manipulation of data may be a scientific norm” because “scientists are human beings rooted in cultural contexts, not automatons directed toward external truth,” wrote Gould. Gould’s view took hold in social studies of science and Morton became a textbook example of scientific misconduct.
Now new research shows that Morton’s measurements and results are accurate, and in fact it was Gould who may have fudged the numbers to make his case. A team of scientists led by Stanford doctoral student Jason Lewis remeasured over 300 of the skulls in the Morton collection and, along with former Stanford professor David DeGusta, reexamined both Morton’s data and Gould’s analysis. They found that Morton’s measurements of skull size were correct 98% of the time, and Morton’s errors were not in the direction of his bias. For example, the three skulls Morton overmeasured were two Africans and a Native American.
“Morton was very careful in his measurements and in describing what he did. He published all his raw data, the vast majority of which are accurate,” said Lewis, the lead author of a paper published in PLoS Biology describing the findings.
Furthermore, the team found that Gould made a number of mistakes in his analysis. “Gould messed up – he left out some skulls from his calculations that should have been included according to his own rules. For a few averages, Gould guessed at which skulls were in the sample, and different guesses produce different numbers. All of Gould’s errors were in the direction of Gould’s bias, which was for all populations to have the same skull size,” said Lewis.
The team was surprised to find that Morton’s actual reports differed in some key ways from how Gould had characterized them. “Morton’s books are rare, and so not many people have read for themselves what Morton had to say, but if you do, you notice immediately that there are some important places where Gould gets it backwards,” said Lewis, who added that much of Morton’s work is now freely available online via their website (http://www.stanford.edu/~lewisjas/Morton.html).
That Morton was able to produce reliable data despite his clear prejudices weakens Gould’s argument that biased results are common in science, argues DeGusta. “Good science does not depend on scientists being unbiased robots, it depends on developing methods that shield the results from being influenced by the investigator’s bias,” said DeGusta.
Gould’s analysis of Morton is frequently assigned reading for introductory anthropology courses, and Lewis hopes that their research will set the record straight. “Gould really did a hatchet job on Morton, making him the poster child for scientific misconduct, but that’s an unfair charge – the evidence shows that, in fact, Morton was quite objective and generated a lot of reliable data,” said Lewis.
In addition to Lewis and DeGusta, the other scientists on the project are Marc Meyer (Chaffey College), Janet Monge (University of Pennsylvania), Alan Mann (Princeton University), and Ralph Holloway (Columbia University). For more information and background materials on this study, see http://www.stanford.edu/~lewisjas/Morton.html .
To read what the press and blogosphere is saying about this work, follow the links below: