Here is a translation into scientific terms. Chagnon's apparent statistical conclusion linking killers and babies is bogus because of a flaw in the data, which means that it is invalid to derive the conclusion that Chagnon derived. The best he can do is claim impressionistically that he hopes Ferguson is wrong.But that defeats the purpose of pretending to be a scientist and doing statistics in the first place.
Tuesday, February 19, 2013
Meet Joe Science
The flap in the last couple of weeks over Jared Diamond’s publicity for his book, The World Until Yesterday, and Napoleon Chagnon’s publicity for his book, Noble Savages needs a little context, which happens to be a specialty of anthropology.
Anthropology is coming off of a year that was described in the leading science journal in these here United States as an annus horribilis. And you know damn well, that when Americans resort to Latin to describe something, it’s got to be pretty bad. Of course in the background we have the perpetual war against the creationists and the racists, both of whom see anthropology as their enemy.
So we elect a President whose mother was an anthropologist, and you think that might bode well for the public appreciation of our field. But we get Florida governor Rick Scott, upset by his daughter’s choice of a major, declaring it to be not the kind of major we want in Florida. And shortly thereafter, the big business publications – Kiplinger and Forbes – publicly branded it as the worst major. It seems almost as if the Republicans had declared war on anthropology, along with the creationists and the racists.
Then we have the one-two punch from the guys claiming to be Joe Science. Jared Diamond pretends to be an anthropologist, but does things that no competent anthropologist would do, and interprets other peoples as no competent anthropologist does. That is to say, he soft-pedals the historical context of other cultures, and imagines them to be stand-ins for his own ancestors. He tells Stephen Colbert that we don’t call New Guineans “primitive” because it’s politically incorrect, apparently unaware that we don’t call them “primitive” because the term connotes a false, ancestral relationship between “them” and “us”.
Napoleon Chagnon is a sadder story, because he is not a pseudo-anthropologist, but an incompetent anthropologist. Let me be clear about my use of the word “incompetent”. His methods for collecting, analyzing and interpreting his data are outside the range of acceptable anthropological practices. Yes, he saw the Yanomamo doing nasty things. But when he concluded from his observations that the Yanomamo are innately and primordially “fierce” he lost his anthropological credibility, because he had not demonstrated any such thing. He has a right to his views, as creationists and racists have a right to theirs, but the evidence does not support the conclusion, which makes it scientifically incompetent.
And so, after the New York Times runs a puff piece on Chagnon in the Magazine, and a critical review of his book simultaneously in the Book Review section, it seems as though the best we can hope for is a draw. But wait! Into the fray comes their distinguished science reporter Nicholas Wade, with a second puff piece on Chagnon.
WTF? His work isn’t that important to anthropology, except as a methodological counter-example. Why is it so important to the Times?
Again, some context, Nicholas Wade once co-wrote a good book on scientific fraud. Judging from his work since then, I am inclined to attribute its meritorious aspects to his co-author. More recently, Wade has been pushing genetic determinism as hard as he can from his pulpit at the Times. And you know who his antagonists are going to be.
Wade’s 2007 book, Before the Dawn, was marketed as a celebration of anthropological genetics. Except that anthropological genetics means something different to anthropological geneticists than it does to Wade. To anthropological geneticists, it means the mutual illumination of genetics and anthropology, often using the technology of the former to study questions framed by the latter. To Nicholas Wade, however, it means that the Chinese excel at ping-pong because of their genes for it (p. 197).
The book was reviewed in the leading science journals by leading anthropological geneticists. The reviewers in Nature found it to be a work of social darwinism (that phrase is not bandied about as a compliment these days). Rebecca Cann in Science had this to say:
As a graduate student … I often wished that there were science writers energized to follow the new insights from geneticists as closely and rapidly as others reported interpretations of fragmentary fossils. Well, be careful what you wish for.
And she continued,
The book also reveals some unpleasant truths about science writing that currently passes for objective and informed. Only smugness that one’s sources must be correct because they represent a scientific elite group having new and exclusive truths about human evolution makes it possible to write, in 2006, sentences such as “The Australian and New Guinean branch [of our phylogenetic tree] soon settled into a time warp of perpetual stagnation."
In other words, the 19th century idea that in looking at other people we are seeing our ancestors, against the modern anthropological view that we are seeing people with other histories, is what unites Chagnon’s, Diamond’s, and Wade’s books. It is a false and outmoded ideology, one that knowledgeable scholars – or, for the sake of argument, “scientists” - have rejected.
Nicholas Wade, however, is undeterred. In an interview with the Anthropology News in 2007, he told them , "Anyone who’s interested in cultural anthropology should escape as quickly as they can from their cultural anthropology department and go and learn some genetics, which will be the foundation of cultural anthropology in the future." In other words, he wants to return to the pre-modern era of anthropology, before E. B. Tylor separated “culture” from “race” or “nature” – because what we think is cultural is really better studied by geneticists. That was actually a “pro-science” stance a century ago, but it is by no stretch of the imagination “pro-science” today. It is decidedly anti-science. It fact it sounds almost as if he is beginning to believe his own bullshit.
So, let’s see what he has to say about Napoleon Chagnon this week, throwing the weight of his reputation in the New York Times around, to balance their recent Chagnon pieces as 2-1 against anthropology.
He starts off, “What were our early ancestors really like…?” – a good question, but one to which Napoleon Chagnon’s work is irrelevant. Bad start, though, because it means that even now, neither Chagnon nor Wade apparently understands what the Yanomamo actually tell us about anything.
“One of Dr. Chagnon’s discoveries was that warriors who had killed a man in battle sired three times more children than men who had not killed.” Not exactly a discovery, though; more of an assertion, which was published in Science, and shown convincingly to be based on a misinterpretation of the data. Chagnon’s interpretation is that his data have no historical context, and are simply the Yanomamo doing what is natural – not only for them, but for us as well. Brian Ferguson’s interpretation is universally taken to be more insightful then Chagnon’s, because it incorporates politics and history. But neither of the pieces puffing up Chagnon, and publicizing his hatred of his colleagues, even acknowledges the existence of alternative interpretations of Chagnon’s work. The problem, simply put, is that Chagnon's statistics were rubbish, because he neglected to include the children of killers who had themselves been killed.
Chagnon's figures on reproductive success did not include dead unokai. The obvious question, in Ferguson's view, was whether the greater reproductive success of unokai was offset by higher mortality. Responding in American Ethnologist, Chagnon calculated the same figures without the headmen and came up with a correlation similar to, although smaller than, his previous figure. But, Chagnon told Science, he “didn't record at the time the status of unokai men who were killed,” which is necessary to respond to Ferguson's second objection. “But from what I know,” he says, “it looks as though [Ferguson's] hypothesis doesn't hold up.”
So now, the best Wade can come up with is to repeat Chagnon’s claim that he represents science against “the ideology of his fellow anthropologists. The general bias in anthropological theory draws heavily from Marxism, Dr. Chagnon writes.” Kind of makes it sound as if he’s got a list of names that he wants to give to Senator McCarthy. Damn commies. (See my previous post for the commies.) The point is that if the ostensible statistical relationships are mirages, then the only people they are going to be able to convince are other cult members. But science is supposed to be able to convince skeptics, not other cult members.
And finally, explains Wade, the entire field of anthropology is like, totally anti-science, even the American Anthropological Association. “In 2010 the A.A.A. voted to strip the word 'science' from its long-range mission plan and focus instead on ‘public understanding.’ Its distaste for science and its attack on Dr. Chagnon are now an indelible part of its record.”
Sure sounds like the American Anthropological Association is indelibly anti-science. Actually, though, that "distaste for science" is very delible. How delible is it? Well actually, that change (to remove the word “science” as a way to emphasize that anthropology incorporates both scientific and humanistic study, and thus is not limited by the scope of science) was suggested by a committee, and was rejected by the membership of the AAA. Wade is wrong, wrong, wrong. Instrumentally, perniciously, and anti-intellectually.
So on one side you’ve got the creationists, racists, genetic determinists, the Republican governor of Florida, Jared Diamond, and Napoleon Chagnon – and on the other side, you’ve got normative anthropology, and the mother of the President. Which side are you on?