Saturday, 2 July 2011

Arroseur arrosé?

Stephen Jay Gould in a 1997 Simpsons episode (Pagepulp). Gould enjoys an almost iconic status in American culture.

Who was the greatest evolutionary scientist of recent times? Most people would answer “Stephen J. Gould,” at least on this side of the Atlantic. With the possible exception of L.L. Cavalli-Sforza, he was the one best known to non-biologists, partly because he wrote well but also because he said the sorts of things that people in the humanities and social sciences wished to hear. My anthropology department was no exception. When I presented my dissertation proposal, one committee member launched into a criticism that he supported with a quote from one of Gould’s works. I didn’t understand the relevance of the quote—other than the banal point that many scholars are unconscious liars. But it hit home among the other people present.

And why not? When Gould died, in 2002, his reputation was unshakable. You might have disagreed with his conclusions, but his methodology seemed sound. This was particularly so with his 1978 Science paper on 19th-century physical anthropologist Samuel George Morton, which showed how a reputedly objective scientist had unconsciously fudged his data to make Europeans look larger-brained than sub-Saharan Africans.

These findings were later carried over into The Mismeasure of Man (1981), a bestseller and now required reading in many undergrad social science courses. Gould also brought up his 1978 paper in public lectures, making it a centerpiece of his attacks on the “myth of scientific objectivity.” In reviewing a posthumous Gould anthology, Richard Lewontin, underlined this point as “the one that is most important to the public understanding of science”:


Despite the myth of detached objectivity that scientists propagate, their motivations are as messy as everyone else's. In particular, they have political, social, and personal concerns that may influence what they do, how they do it, and what they say about it. Putting aside deliberate fraud, of which we have an embarrassment of examples, the gathering of data, their statistical representation, and their interpretation offer many opportunities for unconscious bias toward conclusions that we already "knew" to be true. (Lewontin, 2008)

By the time of his death, Gould had become an icon of popular culture:

Dr. Gould achieved a fame unprecedented among modern evolutionary biologists. The closest thing to a household name in the field, he became part of mainstream iconography when he was depicted in cartoon form on "The Simpsons." Renovations of his SoHo loft in Manhattan were featured in a glowing article in Architectural Digest. (Yoon, 2002)

The Simpsons episode aired in 1997 and is worth summarizing:

Lisa Simpson wants to stop a huge mall development from proceeding at "Sabertooth Ravine" because the ravine is a fossil site. As a compromise, the mall developers decide to let Lisa dig for fossils while they continue to build the mall. While digging, Lisa finds an almost human fossil. Almost, but not quite: in place of arms the fossil has wings. "It's an angel" declare the naive and religiously motivated townfolk. Lisa, who plays the scientific naturalist, will have none of it. She therefore enlists Gould to prove that the fossil is nothing of the sort. Gould claims that the DNA tests he performed proved inconclusive.

[…] In the closing scene, Lisa asks Gould why his test failed to detect that the angel fossil was a fraud. Gould (and mind you, this was Gould's actual voice--he is listed explicitly in the credits) admits that in fact he never did perform the test--even though he claimed he did earlier.
(Dembski, 1997)

The Simpsons episode raised a few eyebrows. Was Gould guilty of the sort of shenanigans he had accused others of doing?

Arroseur arrosé

Yes indeed. A team of physical anthropologists recently located half of the skulls that Samuel George Morton had measured more than a century and a half ago. When they remeasured the skulls they found very few errors in Morton’s measurements. More to the point, the errors were distributed randomly. There was, in fact, a non-significant tendency to overestimate African skull size (Lewis et al., 2011).

It is also doubtful whether Morton considered Africans to be less “evolved” than Europeans. His Crania Americana was published in 1839, long before the first attempts to apply evolutionary theory to human races. Morton was in fact a devout Christian who wished to find out whether different human populations were separate species resulting from multiple divine creations or a single species created but once. He had little if any interest in research on human intelligence, which anyhow was embryonic at the time.

Needless to say, Gould never remeasured any of Morton’s skulls. His paper was at best a clumsy re-analysis of Morton’s published data. I say “at best” because Gould bolstered his argument by creating facts out of thin air. It is a wonder that he managed to get published in a first-tier journal like Science, which as a rule publishes only original data.

Further thoughts

There is another disturbing element in this affair. Many of the flaws in Gould’s paper had already been pointed out … twenty-three years ago (Michael, 1988). And they were pointed out in a first-tier journal (Current Anthropology). Yet that other paper was studiously ignored. Gould owed his reputation not so much to the quality of his work as to an academic milieu that covered for him, acting more as cheerleaders than as responsible critics. He was shielded by a personality cult. Without it, he would have been just another biology professor.

What now? Academia will likely enter a long and painful process of “de-Gouldization.” Long, because many other academics were in on the collective lying. Painful, because the lies were far from trivial.

References

Dembski, W.A. (1997). An Analysis of Homer Simpson and Stephen Jay Gould, Access Research Network.
http://www.arn.org/docs/dembski1129.htm

Gould S.J. (1981). The mismeasure of man. New York: W. W. Norton and Company.

Gould , S.J. (1978). Morton’s ranking of races by cranial capacity: unconscious manipulation of data may be a scientific norm, Science, 200, 503–509.

Lewis, J.E., D. DeGusta, M.R. Meyer, J.M. Monge, A.E. Mann, R.L. Holloway. (2011). The Mismeasure of Science: Stephen Jay Gould versus Samuel George Morton on Skulls and Bias, PLoS Biology, 9(6) e1001071

Lewontin, R.C. (2008). The Triumph of Stephen Jay Gould, New York Review of Books, 55(2), 39-41, February 14, 2008.

Michael, J.S. (1988). A new look at Morton’s craniological research, Current Anthropology, 29, 349–354.

Pagepulp (2011). The literary world of the Simpsons, April 24
http://www.pagepulp.com/176/the-literary-world-of-the-simpsons/

Yoon, C.K. (2002). Stephen Jay Gould, Biologist and Theorist on Evolution, Dies at 60, New York Times, May 20, 2002,
http://www.nytimes.com/2002/05/20/obituaries/20CND-GOULD.html

20 comments:

Anonymous said...

There is another disturbing element in this affair. Many of the flaws in Gould’s paper had already been pointed out … twenty-three years ago (Michael, 1988). And they were pointed out in a first-tier journal (Current Anthropology). Yet that other paper was studiously ignored.

Lewis, et al. (2011) claims that Michael's study had "multiple significant flaws rendering it uninformative".

B.B.

Anonymous said...

The media complex boosting one of their own.

I suspect that the same happened with that other Freud, er, fraud.

James said...

Bernard D. Davis quoted, approvingly, the reviewer in Nature commenting on Gould's book:

"... a book which exemplifies its own thesis. It is a masterpiece of propaganda, researched in the service of a point of view rather than written from a fund of knowledge."

... and ...

"(Gould) has nothing to say which is both accurate and at issue when it comes to substantive or methodological points"

... and (my personal favorite) ...

many of his assertions "have the routine flavor of Radio Moscow ..."

Source: Storm over Biology by Bernard D. Davis, Prometheus Books 1986.

I've proposed elsewhere a phrase I use when I encounter any enthusiasm for this charlatan's so-called ideas: gouldcrap.

With his fellow Marxist Lewontin about to retire as head of Harvard's Genetics Dept it might be worthwhile asking oneself if he ever approved of hiring any "scientist" who did not share his ideology.

Anonymous said...

It's hard to know just how corrupt are the "academics" of the West--in every damn field.

Makes me sick.

Ben10 said...

I suspected Gould to work for the agenda of his buddies, consciously or not, and your text shows that it was consciously. I am disapointed, but not surprised. Thanks, Peter, for this 'mise au point'.

But still, I read some of Gould for his famous punctuated equilibrium, I thought that here, he had a point, and a strong one. Nobody can be completely wrong, even Gould, so, were the data overblown or even faked by Gould?

There is a passage I can't find, from memory he said more or less something like this:

"species seem to go trough environmental changes without any physical modification then suddenly new species appear completely formed. In the strata series, new variants of a species that are mere fluctuations around the common type, appear and disappear, perhaps as adaptation to environmental change. But the new variants disappear and the oldest type of the species, at the bottom of the strata, is often the only one to survive until the top of the serie...
"

Does anybody remember he said that and is there any truth in it?

Kiwiguy said...

Peter,

The author of the 1988 Current Anthropology paper, John Michael, comments on the paper:

"I am John S. Michael who as an undergraduate measured the Morton skulls. I am tickled to see that my work appears to be vindicated. I subsequently went onto a career in environmental planning, so I have little to say that is substantive regarding whether or not I used the proper statistics.

But what I can say is that as a 20-year-who was interested in attending graduate school, I was advised not to be too critical of Gould lest I jeopardize my future career. He was a big shot, and I was just a kid. Thus I held back on what I felt was his unfair and overly simplistic treatment of Morton, and more importantly Gould’s assertion that all races were equal, when there was ample evidence that the traditional concept of race has no biological basis.

I am pleased that this latter point was presented prominently in the above article. For me this point is, and always has been, the most important fact, which transcends academic arguments and could have a real impact on public policy and the dream of a more equitable world...

http://www.plosbiology.org/article/comments/info%3Adoi%2F10.1371%2Fjournal.pbio.1001071;jsessionid=D66C3DABD85747409EC7A5FFE9E71FD7.ambra02

Kiwiguy said...

cont'd

"I will also note that soon after my initial research, I mailed Gould my preliminary results and some months later spoke with him briefly when he was visiting in St. Paul. He refuted my arguments, and we left agreeing to disagree.

I would also like to comment on Kitcher and others who have described my work as being of “questionable value.” It is true that my work was done by an undergrad who did not have the resources of the six full time professors who wrote this paper, and who ultimately came up with the same conclusions as an undergrad with one semester’s worth of time to undertake his research. And to these six professors, I would like to say, thank you.

If Kitcher and others had chosen to, they could have read my paper and declared that I had a valid point, and given me some credit for having taken the time to measure the skulls. (I forget whether measuring the skulls was my idea or Janet Monge’s, but clearly the both of us were willing to give it a shot, which no one else was.) I was put in a position where I had to critique the biggest name in evolutionary biology and that was not a popular stance. So instead of being given credit as a creative rough-around-the-edges young man who was willing to do the leg work, I was dismissed based on technicalities, which I openly admit were quite valid.

Now that twenty years have passed, during which time I had assumed my work had been entirely forgotten, I have gained some perspective. Indeed my technique was imperfect, as was Gould’s and Morton’s. We are all complex human beings just like everyone else.

The anthropological community had an opportunity to say that my work was somewhat flawed but worthy of considering, but instead it chose to say it was somewhat flawed, and so should be discounted. It could have gone either way. And now we are left to ask why things went the way they did. As an outsider, I am wholly unqualified to answer that question.

Stephen Gould should be celebrated as an innovative scientist whose significant work on punctuated evolution is way beyond me. However, my sense is that he wanted to be a philosopher and to that end tried to show that science could serve a moral function by proving all men are equal. But science is a tool, so like a primitive stone hammer, which can be used for good or ill. We should not view science as moral or immoral, but rather as amoral. That was Gould’s mistake.

I wanted to write these words 20 years ago, and thanks to your web site, I now can. Again, thanks to Janet and the gang for publishing this.

John S. Michael, West Chester, PA"

Kiwiguy said...

My comment appears to have vanished, but JS Michael's comment on the new paper features here.

sykes.1 said...

An indication of Gould's honesty is on pp. 109-110 of "The Mismeasure of Man."
There he castigates a researcher for using the standard error of the means to determine whether the differences in the means of skull sizes between professions here is significant. Gould asserts the researcher should have used the standard deviations of the samples. Since standard deviations are always larger than standard errors, Gould says there is no statistical difference in skull sizes.

Now, every Stat 101 course proves that the appropriate statistic to be used in comparing means is the standard error. That is unequivocal; it is a strict mathematical proof.

And Gould routinely used statistics in his own research on shell fish.

So, either Gould's own research is by and large nonsense, or he lied on pp. 109-110.

Gould's academic and scientific career is contemptible.

Tod said...

The committee member was not trying to be relevant, he mentioned Stephen Jay Gould by way of a shibboleth.

People like Gould become public intellectuals because the criterion for a evolutionary theorist being hyped into a public intellectual is that they support the political implications of the dominant political worldview in respect of human biology. (John Gray on why the left is in flight from 'human nature'.)

As that worldview is quite wrong about human nature(s) the work they choose to promote is necessarily quite unscientific. Hence the heavily promoted part of Gould's work is impervious to scientific falsification, it exists now as part of the first principles with which the dominant worldview of the West constructs its picture of reality.

Academia will likely enter a long and painful process of “de-Gouldization.” Long, because many other academics were in on the collective lying. Painful, because the lies were far from trivial.

Like it did with the lies of Boas?

Reader said...

I don't think we have to worry too much about Gould and Lewontin these days, because the field of social science has generally unravelled and been widely discredited over the past decade. I'm talking about general trends, which are influencing public opinion.

Gould and Lewontin harken back to the days of behaviorists like B.F.Skinner and the like, and no one's taking that seriously anymore, post-'90s. We're in a new era.

Instead, there has been a recent explosion of interest in Human Biodiversity (HBD) in the new millenium, led by scientists from David Buss to Satoshi Kanazawa to, of course, Peter Frost, with many others chiming in and pointing out the undeniable effects of biology and genetics, rather than spurious cultural notions and the falsehood that "we are all the same."

Even among laypeople, who don't read scientific journals, the politically correct ideas of social science are being rejected, perhaps the only exception being the readership of the New York Times. But in the blogosphere, influential posts by the likes of Steve Sailer, Dalrock, and many others are gradually leading a revolution in our perception of human behavior. The message from that revolution is that behavior is shaped by biology/genetics much more than by the environment, and that no, we are not all the same.

Kiwiguy said...

***owed his reputation not so much to the quality of his work as to an academic milieu that covered for him, acting more as cheerleaders than as responsible critics. He was shielded by a personality cult. ***

From the comment above by JS Michael, author of the Current Anthropology paper:

But what I can say is that as a 20-year-who was interested in attending graduate school, I was advised not to be too critical of Gould lest I jeopardize my future career. He was a big shot, and I was just a kid. Thus I held back on what I felt was his unfair and overly simplistic treatment of Morton...

I will also note that soon after my initial research, I mailed Gould my preliminary results and some months later spoke with him briefly when he was visiting in St. Paul. He refuted my arguments, and we left agreeing to disagree.

I would also like to comment on Kitcher and others who have described my work as being of “questionable value.” It is true that my work was done by an undergrad who did not have the resources of the six full time professors who wrote this paper, and who ultimately came up with the same conclusions as an undergrad with one semester’s worth of time to undertake his research. And to these six professors, I would like to say, thank you.

If Kitcher and others had chosen to, they could have read my paper and declared that I had a valid point, and given me some credit for having taken the time to measure the skulls. (I forget whether measuring the skulls was my idea or Janet Monge’s, but clearly the both of us were willing to give it a shot, which no one else was.) I was put in a position where I had to critique the biggest name in evolutionary biology and that was not a popular stance. So instead of being given credit as a creative rough-around-the-edges young man who was willing to do the leg work, I was dismissed based on technicalities, which I openly admit were quite valid...

Now that twenty years have passed, during which time I had assumed my work had been entirely forgotten, I have gained some perspective. Indeed my technique was imperfect, as was Gould’s and Morton’s. We are all complex human beings just like everyone else.

The anthropological community had an opportunity to say that my work was somewhat flawed but worthy of considering, but instead it chose to say it was somewhat flawed, and so should be discounted. It could have gone either way. And now we are left to ask why things went the way they did. As an outsider, I am wholly unqualified to answer that question.

Stephen Gould should be celebrated as an innovative scientist whose significant work on punctuated evolution is way beyond me. However, my sense is that he wanted to be a philosopher and to that end tried to show that science could serve a moral function by proving all men are equal. But science is a tool, so like a primitive stone hammer, which can be used for good or ill. We should not view science as moral or immoral, but rather as amoral. That was Gould’s mistake."

Kiwiguy said...

Regarding Gould being protected, JS Michael's comments:

But what I can say is that as a 20-year-who was interested in attending graduate school, I was advised not to be too critical of Gould lest I jeopardize my future career. He was a big shot, and I was just a kid. Thus I held back on what I felt was his unfair and overly simplistic treatment of Morton...

I will also note that soon after my initial research, I mailed Gould my preliminary results and some months later spoke with him briefly when he was visiting in St. Paul. He refuted my arguments, and we left agreeing to disagree.

I would also like to comment on Kitcher and others who have described my work as being of “questionable value.” It is true that my work was done by an undergrad who did not have the resources of the six full time professors who wrote this paper, and who ultimately came up with the same conclusions as an undergrad with one semester’s worth of time to undertake his research. And to these six professors, I would like to say, thank you.

If Kitcher and others had chosen to, they could have read my paper and declared that I had a valid point, and given me some credit for having taken the time to measure the skulls. (I forget whether measuring the skulls was my idea or Janet Monge’s, but clearly the both of us were willing to give it a shot, which no one else was.) I was put in a position where I had to critique the biggest name in evolutionary biology and that was not a popular stance. So instead of being given credit as a creative rough-around-the-edges young man who was willing to do the leg work, I was dismissed based on technicalities, which I openly admit were quite valid.

Now that twenty years have passed, during which time I had assumed my work had been entirely forgotten, I have gained some perspective. Indeed my technique was imperfect, as was Gould’s and Morton’s. We are all complex human beings just like everyone else.

The anthropological community had an opportunity to say that my work was somewhat flawed but worthy of considering, but instead it chose to say it was somewhat flawed, and so should be discounted. It could have gone either way. And now we are left to ask why things went the way they did. As an outsider, I am wholly unqualified to answer that question.

Kiwiguy said...

***the politically correct ideas of social science are being rejected, perhaps the only exception being the readership of the New York Times.***

Not really, Nicolas Wade covered this paper showing up Gould and quoted Ralph Holloway saying he was a charlatan.

Kiwiguy said...

JS Michael, author of the Current Anthropology paper, on the Lewis paper (see link above):

"But what I can say is that as a 20-year-who was interested in attending graduate school, I was advised not to be too critical of Gould lest I jeopardize my future career. He was a big shot, and I was just a kid. Thus I held back on what I felt was his unfair and overly simplistic treatment of Morton...

I will also note that soon after my initial research, I mailed Gould my preliminary results and some months later spoke with him briefly when he was visiting in St. Paul. He refuted my arguments, and we left agreeing to disagree.

I would also like to comment on Kitcher and others who have described my work as being of “questionable value.” It is true that my work was done by an undergrad who did not have the resources of the six full time professors who wrote this paper, and who ultimately came up with the same conclusions as an undergrad with one semester’s worth of time to undertake his research. And to these six professors, I would like to say, thank you.

If Kitcher and others had chosen to, they could have read my paper and declared that I had a valid point, and given me some credit for having taken the time to measure the skulls. (I forget whether measuring the skulls was my idea or Janet Monge’s, but clearly the both of us were willing to give it a shot, which no one else was.) I was put in a position where I had to critique the biggest name in evolutionary biology and that was not a popular stance. So instead of being given credit as a creative rough-around-the-edges young man who was willing to do the leg work, I was dismissed based on technicalities, which I openly admit were quite valid.

Now that twenty years have passed, during which time I had assumed my work had been entirely forgotten, I have gained some perspective. Indeed my technique was imperfect, as was Gould’s and Morton’s. We are all complex human beings just like everyone else.

The anthropological community had an opportunity to say that my work was somewhat flawed but worthy of considering, but instead it chose to say it was somewhat flawed, and so should be discounted. It could have gone either way. And now we are left to ask why things went the way they did. As an outsider, I am wholly unqualified to answer that question.

Peter Frost said...

Anon,

The word "uninformative" seems to me unjustified. I suspect it was inserted to appease one of the reviewers.

James,

Gould and Lewontin were ideologically very different from most American biologists. But they were good showmen. They knew how to sell their viewpoints, and shout down alternate viewpoints.

Anon,

Most academics are honest men and women. Unfortunately, they also prefer to do what they're supposed to do, i.e., teach and do research.

Ben10,

Yes, he did good academic work ... when he didn't have an axe to grind. I particularly enjoyed reading "Ontogeny and Phylogeny."

As for punctuated equilibrium, I have to disagree with you. To me at least, he seemed to be setting up a straw man, in typical Gouldian fashion. No one ever claimed that evolution takes place at a constant uniform rate. It has long been accepted that evolution accelerates whenever a new niche opens up.

Sykes,

Yes, there are many other cases of "creative statistics" that could be cited.

Tod,

Boas, like Gould, has long been a teflon man. Perhaps the teflon is losing its effect? Who knows.

Reader,

Thanks for the compliment, but I'm a very minor player in all of this. As for David Buss, you may know something I don't know. What has he said about HBD?

Kiwiguy,

Good post! I think Michael's study is a bit like Mendel's a century ago. It is only now getting its just appreciation.

Reader said...

David Buss is mostly known for his work on human mating behaviors. I consider him part of the HBD sphere because he emphasizes biological factors in shaping mating behavior, rather than culture or upbringing. He takes exception to culturalists, gender denialists (feminists), and others who use social science to explain male/female differences.

He is not nearly as abrasive as Kanazawa, of course, and he also sometimes does what's necessary to get published, but overall he leans in the direction of biology.

Peter Frost said...

Reader,

I admire David Buss. He was one of the few people who helped put evolutionary psychology back on its feet after the witch hunts of the 1980s.

But I've never heard him express any interest in HBD. His official stance is that the genetic underpinnings of human behavior have not significantly changed since the Pleistocene.

Perhaps he thinks otherwise in private. But does it matter? What is the real-world impact of privately held opinions?

Anonymous said...

The blank slate fallacy was designed as a weapon in the cultural warfare that took place over the 1924 and the 1965 immigration acts. Maintaining the fallacy has the same, sole purpose - mass immigration and the dispossession and eventual extinction of white Americans.

Anonymous said...

The blank slate fallacy was designed as a weapon in the cultural warfare that took place over the 1924 and the 1965 immigration acts.

Perhaps (with respect to the form the BS fallacy took at that time, though other forms predated it). But, if so, it wasn't the opening shot. It was a reaction to the implications of the hereditarian anthropology of the day.

What were those implications? In a nutshell, that greatness is the only thing that matters. Anything less than greatness is rank inferiority, wholly undesirable and (we'll see to it!) wholly unnecessary.

In truth, such implications hardly necessarily flowed from the study of "HBD," but combating the perception that they did, in the cultural milieu of the early 20th century, was deemed to require something more potent than merely pointing that out.

Maintaining the fallacy has the same, sole purpose - mass immigration and the dispossession and eventual extinction of white Americans.

It certainly can be (and is, eg Amitai Etzion) used as a weapon in this regard, but there's much more to it than merely that.

As a thought experiment, imagine a world populated by humans of only one, "completely homogeneous" race (the sneer quotes are necessary because the notion of complete homogeneity largely misunderstands what race is). The BS fallacy would still be seen as politically useful in preventing the powerful from inflicting suffering on the powerless as well as advancing the cause of progress, both social/societal and individual.

If I had to choose, I would nominate the last as the most important. Life's events have a tendency to grind most of us down, such that we typically operate well below whatever our natural capacity may be. While we may not be completely malleable, the promotion of the BS fallacy is inherently (heh) more ennobling and enabling than its antithesis, and accepting it can reasonably be thought to increase the likelihood that we will take such steps as necessary to operate at our full capacity -- ie, we'll believe it's possible (often accurately) for us to improve ourselves . Accepting the tenets of HDB, on the other hand, is much more likely to cause us to throw our hands up in resignation ("What do you expect? I was BORN this way!") and even to depress us ("...and it SUCKS!").

So it's not as straightforward as WNs (which I'm assuming you are) would have it. And frankly, untangling where one motive ends and another begins is fraught with difficulty. I mean, has any Nazi ever hated any Jew as much as some communists (like Robert Lindsay, with whom you might be familiar) hate reality? It can be an interesting question to ponder.

- Silver