Jul 29, 2005
Diamond, DeLong and Savage Minds
Brad DeLong has been fairly harsh in his response to the Savage Minds bloggers on the subject of Jared Diamond’s Guns, Germs and Steel. I also have problems with the SM bloggers in their reading (or viewing) of Diamond, but there are some legitimate criticisms of Diamond to be made, both problems that are particular to his work and problems that are more general in sociobiological or materialist histories.
I’m going to focus in particular on the chapter of GGS that deals most centrally with Africa.
First, on the contested question of Diamond’s “racism”. It’s a serious mistake to even imply that Diamond is racist, as Henry Farrell properly observes. I would say that he has a stubborn inclination to use racial terms when they don’t serve any empirical or descriptive purpose. It may be that a term like “race” can still serve some useful purpose in describing variations between human populations: I’m not going to make a definitive statement on that subject here. But just to give the example of the Africa chapter, Diamond clings to the term “blacks” as racial category within which to place most pre-1500 sub-Saharan Africans except for Khoisan-speakers and “pygmies”, even as he explicitly acknowledges that it is an extremely poor categorical descriptor of the human groups he is placing in that category. The chapter’s central interest is the migration of Bantu-speakers across the continent, with the argument that iron working and agricultural knowledge were what enabled them to displace autochthonous Khoisan and pygmy societies. This is an uncontroversial argument, but the point is that it doesn’t require a category of “blacks” to function, because the only thing Diamond is interested in is Bantu-speakers and their technological and material capacities. There’s no need for him to enfold the African populations of West Africa, who are not Bantu-speaking: their history isn’t what interests him in the chapter, he doesn’t talk about it save at the beginning. Why not call Bantu-speaking societies what they are? It’s not that much more technical a term than “blacks”. Throughout the book, Diamond seems to me to cling to terms and categories that he doesn’t need, and I’m not really sure why. However, I also think this is a relatively minor technical argument that doesn't demand or deserve any kind of strong rhetoric.
Second, Diamond has a tendency to exclude—not even mention or argue against, but simply bypass—deeply seated causal arguments and evidence that don’t fit his thesis. Let me take the Bantu-speaking migration again. There’s no question that iron working and farming were very important to driving their movements across the central, eastern and southern portions of the African continent, and were the central reason why older populations of pastoralists and hunter-gatherers were either absorbed into Bantu-speaking societies or fled from their advance. But Diamond takes it as a given that iron working and farming are sufficient explanation of the migration itself, that they made the expansion of the Bantu-speakers inevitable. That may be so, but he doesn’t even bother to discuss segmentary kinship as a form of social organization within Bantu-speaking societies, and its possible role in pushing expansion. This is the key explanation that many Bantu-speaking societies offer themselves for their migrations, that when there was at some past point strife or tension among kin, a portion of a lineage would break off behind a charismatic lineage head and move on. That’s obviously not the whole story, but I think it’s part of it. Diamond’s materialism is so confidently asserted and at such a grand scale that he doesn’t even pause to defend it trenchantly the way someone like Marvin Harris does. It’s more at times as if he’s not even aware of other causal arguments. This is especially acute, as many readers of GGS have noted, with his views on Chinese history and the venerable question of why China did not industrialize but the West did. There’s a tremendous weight of evidence that the general political traditions of the Chinese state plus the particular decisions of its political elite at key moments are much more powerfully explanatory of China’s failure to expand or dominate in the post-1500 era than the big-picture materialism that Diamond offers.
Third, he is a bit prone, like many sociobiologists and evolutionary psychologists, to what I think of as “ethnographic tourism”, scooping up all the cases of human practice or culture that fit his assertions about universal patterns or behaviors. I think Diamond indulges in this bad habit more in The Third Chimpanzee than GGS: see for example in TC his discussion of penis size and male display among humans, where he asserts a universal and adaptive explanation but ignores historical and contemporaneous examples that don’t really fit the pattern. You can get away with that when you’re just describing a tendency, but the stronger your claims are couched in terms of universality and adaption, the harder it becomes. Diamond is by no means as egregious in this kind of cherrypicking as some evolutionary psychologists are, but the selectivity of his evidentiary citation grates a bit on anyone who knows the ethnographic literature. (Especially when you know that some of the 1970s and 1980s syntheses that he cites rely on older studies that are dubious or have been challenged since by both empirical work and theoretical critique.)
Fourth, on Yali’s question, I have a few problems. Though Brad DeLong insists that Diamond only means his answer to explain the relative imbalance in material wealth and power between many non-Western societies and the West up to 1500 and not afterwards, I think it’s clear that Diamond thinks that post-1500 events are no more than the icing on the cake, that the fundamental explanation of post-1500 inequalities and disparities in the world derive from the grand arc of pre-1500 development, from the luck of the geographical draw. He’s not alone in that: this is a venerable argument which takes on variant forms among world-systems historians and Marxists. But DeLong is being a bit unfair to insist somehow that the Savage Minds bloggers have in this respect misread Diamond: he clearly argues that the pre-1500 history is crucially determinant of the post-1500 history.
More, Diamond’s arguments about Yali’s question strike me as sometimes being too large in temporal and geographical scale. He goes back too far and enfolds too much for at least some of what he'd like to explain. Because the grand argument of GGS turns on the slow accumulation of geographical advantage to people inhabiting the Eurasian continent, it sometimes ignores much more short-term material explanations which are potentially in and of themselves sufficient explanations. To explain the Atlantic slave trade in materialist terms, for example, you may need nothing more than the relative proximity of Africa and Europe, the trade wind system across the Atlantic, improvements in European nautical capability prior to 1450, and the relative lack of harbors plus poor habitability of the West and Central African coastline. To explain other aspects of Western expansion, you may need little more than the Crusades, the Mongols, and an understanding of long-distance trading patterns circa 1200. That may sound like a lot, but it’s a much more constrained set of factors with a much shorter temporal scale than what Diamond puts into play. And the Atlantic slave trade may itself be a nearly sufficient explanation of the expansion of the West after 1500, given the cascade of effects it unleashed.
Anthropologists and historians interested in non-Western societies and Western colonialism also get a bit uneasy with a big-picture explanation of world history that seems to cancel out or radically de-emphasize the importance of the many small differences and choices after 1500 whose effects many of us study carefully. For example, it seems to me that if you want to answer Yali’s question with regards to Latin America versus the United States, you’ve got to think about the peculiar, particular kinds of political, legal and religious frameworks that differentiated Spanish colonialism in the New World from British and French colonialism, that a Latin American Yali would have to feel a bit dissatisfied with Diamond’s answer.
For me, I also feel a bit at a loss with any big-picture history that isn’t much interested in the importance of accident and serendipidity at the moment of contact between an expanding Europe and non-Western societies around 1500. That seems a part of Cortes’ conquest of Montezuma, or the early beginnings of the Atlantic slave trade, when West African practices of kinship slavery fed quite incidentally into exchange with Portuguese explorers who weren’t there for slaves at all. It may be that such accidents are not the cause of the material disparity that Yali describes, but in many cases, they’re what makes the contemporary world feel the way that it does. It’s not that Diamond argues against such matters, but he doesn’t leave much room for them to matter, either.
I’m going to focus in particular on the chapter of GGS that deals most centrally with Africa.
First, on the contested question of Diamond’s “racism”. It’s a serious mistake to even imply that Diamond is racist, as Henry Farrell properly observes. I would say that he has a stubborn inclination to use racial terms when they don’t serve any empirical or descriptive purpose. It may be that a term like “race” can still serve some useful purpose in describing variations between human populations: I’m not going to make a definitive statement on that subject here. But just to give the example of the Africa chapter, Diamond clings to the term “blacks” as racial category within which to place most pre-1500 sub-Saharan Africans except for Khoisan-speakers and “pygmies”, even as he explicitly acknowledges that it is an extremely poor categorical descriptor of the human groups he is placing in that category. The chapter’s central interest is the migration of Bantu-speakers across the continent, with the argument that iron working and agricultural knowledge were what enabled them to displace autochthonous Khoisan and pygmy societies. This is an uncontroversial argument, but the point is that it doesn’t require a category of “blacks” to function, because the only thing Diamond is interested in is Bantu-speakers and their technological and material capacities. There’s no need for him to enfold the African populations of West Africa, who are not Bantu-speaking: their history isn’t what interests him in the chapter, he doesn’t talk about it save at the beginning. Why not call Bantu-speaking societies what they are? It’s not that much more technical a term than “blacks”. Throughout the book, Diamond seems to me to cling to terms and categories that he doesn’t need, and I’m not really sure why. However, I also think this is a relatively minor technical argument that doesn't demand or deserve any kind of strong rhetoric.
Second, Diamond has a tendency to exclude—not even mention or argue against, but simply bypass—deeply seated causal arguments and evidence that don’t fit his thesis. Let me take the Bantu-speaking migration again. There’s no question that iron working and farming were very important to driving their movements across the central, eastern and southern portions of the African continent, and were the central reason why older populations of pastoralists and hunter-gatherers were either absorbed into Bantu-speaking societies or fled from their advance. But Diamond takes it as a given that iron working and farming are sufficient explanation of the migration itself, that they made the expansion of the Bantu-speakers inevitable. That may be so, but he doesn’t even bother to discuss segmentary kinship as a form of social organization within Bantu-speaking societies, and its possible role in pushing expansion. This is the key explanation that many Bantu-speaking societies offer themselves for their migrations, that when there was at some past point strife or tension among kin, a portion of a lineage would break off behind a charismatic lineage head and move on. That’s obviously not the whole story, but I think it’s part of it. Diamond’s materialism is so confidently asserted and at such a grand scale that he doesn’t even pause to defend it trenchantly the way someone like Marvin Harris does. It’s more at times as if he’s not even aware of other causal arguments. This is especially acute, as many readers of GGS have noted, with his views on Chinese history and the venerable question of why China did not industrialize but the West did. There’s a tremendous weight of evidence that the general political traditions of the Chinese state plus the particular decisions of its political elite at key moments are much more powerfully explanatory of China’s failure to expand or dominate in the post-1500 era than the big-picture materialism that Diamond offers.
Third, he is a bit prone, like many sociobiologists and evolutionary psychologists, to what I think of as “ethnographic tourism”, scooping up all the cases of human practice or culture that fit his assertions about universal patterns or behaviors. I think Diamond indulges in this bad habit more in The Third Chimpanzee than GGS: see for example in TC his discussion of penis size and male display among humans, where he asserts a universal and adaptive explanation but ignores historical and contemporaneous examples that don’t really fit the pattern. You can get away with that when you’re just describing a tendency, but the stronger your claims are couched in terms of universality and adaption, the harder it becomes. Diamond is by no means as egregious in this kind of cherrypicking as some evolutionary psychologists are, but the selectivity of his evidentiary citation grates a bit on anyone who knows the ethnographic literature. (Especially when you know that some of the 1970s and 1980s syntheses that he cites rely on older studies that are dubious or have been challenged since by both empirical work and theoretical critique.)
Fourth, on Yali’s question, I have a few problems. Though Brad DeLong insists that Diamond only means his answer to explain the relative imbalance in material wealth and power between many non-Western societies and the West up to 1500 and not afterwards, I think it’s clear that Diamond thinks that post-1500 events are no more than the icing on the cake, that the fundamental explanation of post-1500 inequalities and disparities in the world derive from the grand arc of pre-1500 development, from the luck of the geographical draw. He’s not alone in that: this is a venerable argument which takes on variant forms among world-systems historians and Marxists. But DeLong is being a bit unfair to insist somehow that the Savage Minds bloggers have in this respect misread Diamond: he clearly argues that the pre-1500 history is crucially determinant of the post-1500 history.
More, Diamond’s arguments about Yali’s question strike me as sometimes being too large in temporal and geographical scale. He goes back too far and enfolds too much for at least some of what he'd like to explain. Because the grand argument of GGS turns on the slow accumulation of geographical advantage to people inhabiting the Eurasian continent, it sometimes ignores much more short-term material explanations which are potentially in and of themselves sufficient explanations. To explain the Atlantic slave trade in materialist terms, for example, you may need nothing more than the relative proximity of Africa and Europe, the trade wind system across the Atlantic, improvements in European nautical capability prior to 1450, and the relative lack of harbors plus poor habitability of the West and Central African coastline. To explain other aspects of Western expansion, you may need little more than the Crusades, the Mongols, and an understanding of long-distance trading patterns circa 1200. That may sound like a lot, but it’s a much more constrained set of factors with a much shorter temporal scale than what Diamond puts into play. And the Atlantic slave trade may itself be a nearly sufficient explanation of the expansion of the West after 1500, given the cascade of effects it unleashed.
Anthropologists and historians interested in non-Western societies and Western colonialism also get a bit uneasy with a big-picture explanation of world history that seems to cancel out or radically de-emphasize the importance of the many small differences and choices after 1500 whose effects many of us study carefully. For example, it seems to me that if you want to answer Yali’s question with regards to Latin America versus the United States, you’ve got to think about the peculiar, particular kinds of political, legal and religious frameworks that differentiated Spanish colonialism in the New World from British and French colonialism, that a Latin American Yali would have to feel a bit dissatisfied with Diamond’s answer.
For me, I also feel a bit at a loss with any big-picture history that isn’t much interested in the importance of accident and serendipidity at the moment of contact between an expanding Europe and non-Western societies around 1500. That seems a part of Cortes’ conquest of Montezuma, or the early beginnings of the Atlantic slave trade, when West African practices of kinship slavery fed quite incidentally into exchange with Portuguese explorers who weren’t there for slaves at all. It may be that such accidents are not the cause of the material disparity that Yali describes, but in many cases, they’re what makes the contemporary world feel the way that it does. It’s not that Diamond argues against such matters, but he doesn’t leave much room for them to matter, either.