Thomas Sowell: Race and Culture

I finally got around to reading Thomas Sowell's classic book, Race and Culture.

Often a provocative and difficult/controversial topic, I appreciate the scholarly approach Sowell brought to the project, looking at various angles through separate chapters on culture (through migration and conquest) and race (and its connections to economics, politics, intelligence, slavery and history). He opens by poking at a predominant "social science" model that largely (and ironically) ignores various social factors (choosing to focus on a few, simple, favored hypotheses).
In our time, a similar irony is that we're often told about the importance of individual biology-- while we're told that aggregating individual biology to groups is somehow automatically troublesome.

One of the huge over-arching questions in play is "environment vs. culture". Sowell notes, "While we can all agree on the influence of 'environment' in some very general sense, there is a vast difference between (1) regarding groups as being shaped by immediate circumstances, including the people and institutions around them; and (2) regarding groups as having their own internal cultural patterns, antedating the environment in which they currently find themselves." (x). The value of international comparisons in this context? When we see patterns which recur across many countries, then we can more easily see what's internal vs. external (x). As such, the purpose of the book "is to demonstrate the reality, persistence, and consequences of cultural differences (xii).

As one might expect/hope, Sowell provides a ton of examples on differences (e.g., p. 2-4) and why culture matters (e.g., p. 85). On the former, see: the prevalence of Jews in the apparel industry, Germans with pianos, Italian fishermen and architects, India's entrepreneurial Gujaratis, Scots' medical knowledge, and the difference between particular subgroups of Scots who settled in Appalachia. On the latter, Sowell notes the differences in Jewish and Christian Sabbath practices and the implications for operating a factory while avoiding/engaging in "discrimination".

Sowell even wrestles with "cultural superiority". One should be careful-- and judgments might be contingent-- but at times, it should be safe to say that X is superior to Y. "Arabic numerals are not merely different from Roman numerals; they are superior..." (5) The punchline, as opposed to cultural relativism: "It is not necessary to claim that a particular people or a particular culture is superior in all things or for all times...but neither is it necessary to deny the greater effectiveness of particular cultures for particular things at particular times and places..." (6a, 225).

On "conquest", Sowell cautions against a simplistic view of conquest for purely economic/financial motives (73-75). Certainly, those motives are in play at times, but Sowell notes the importance of political motives, even at the expense of economics/finance. "The reductionist notion that economic motives can be automatically inferred behind conquests of the modern capitalist era is ironically applied to a period-- the 19th and 20th centuries-- when non-economic influences were especially strong, particularly in the case of much of sub-Saharan Africa. European officials responsible for the public treasury were often opposed to the development of a colonial empire in Africa, which they correctly saw as having little capacity to repay the cost of conquest..." (74) In terms of outcomes, Sowell notes the abuses of the colonists, but also that exports and imports grew dramatically during the colonial era.

Sowell also discusses "statistical discrimination"-- a term used by economists to describe the inevitable/universal stereotyping of people and situations, given highly-imperfect and costly-to-obtain information. For better and for worse, members of the "good" (bad) group will tend to be viewed (un)favorably-- at least until more accurate information emerges about the individual.

Such decision-making is regrettable at some level, but unavoidable given the limited info at hand. As a result, all of us discriminate in this way. We rely on generalizations that are, generally, true-- to help us make relatively effective decisions in a low-information world. For example, we tend to have more respect for the intellect of a Purdue grad in Engineering with a 3.5 GPA than an Indiana State grad in a social science with a 2.5. And so on.

On occasion, these measurable/proxies can be gender or race-- when that attribute correlates with success or failure. Sowell provides examples-- e.g., 19th-century Irish with alcohol problems, saying it was common for employers to routinely and openly discriminate against the Irish based on this generalization (89). And he quotes WEB Dubois on the same thing with respect to blacks in late 19th century America-- that the group's low productivity on average was hurting the perception of productive individuals (90).

On the minimum wage, Sowell notes the theory and empirical work of its impact on unskilled labor, which has been disproportionately minority. In 1950, the minimum wage jumped up considerably and has been much higher since then (in real terms) with more extensive coverage. Sowell observes that the black and black teen unemployment rates were comparable with whites until then-- and diverged sharply after that (94-95). Of course, there are many reasons why workers are unskilled-- from education to family structure/stability. (And there are many causes of unemployment that might vary by race-- at least in theory.) But increasing the cost of unskilled labor should be expected to decrease the number of unskilled hired-- and thus, to disproportionately cause unemployment among groups with more unskilled workers.

On the politics of race, Sowell makes three prescient observations: First, it's ironic that we claim to want to see individuals as individuals-- but we measure success by group outcomes. Second, democracy alone is not nearly enough (124-125). Third, discrimination-- even when a secondary factor-- will often be an attractive hypothesis politically, in comparison to the hard work of dealing with thornier and more sensitive primary causes. "Factors such as inter-group differences in demographic characteristics, geographical distribution, skill levels, or cultural values tend to be ignored, however demonstrably important they may be in a cause-and-effect sense. Thus, while the black population in late 20th-century America suffered greatly from soaring rates of violent crime, from having much of its newborn generation raised by [single] teenage mothers, and from widespread drug addiction, its political leaders have focused their efforts on correcting the failings (real and presumed) of the white population..." (139-140)

On race and intelligence, Sowell opens by noting the difficulty in discussing the topic dispassionately and in measuring race or intelligence (156). His discussion was helpful although surely out-of-date on at least the data. I did find one thing quite helpful: moving away from incendiary references about general intelligence (e.g., an IQ test) to narrower examples: the Chinese over Jews and blacks on "spatial conception" (162, 169).

On slavery, Sowell is helpful in promoting a broader understanding-- in particular, that's it not simply (or even primarily) a Western or U.S. institution. Arguably, it's most interesting in the West, given our supposed political values! But "the irony of our times is that the destruction of slavery around the world, which some once considered the supreme moral act in history, is less well-known and less-discussed among intellectuals in either Western or non-Western countries, while the enslavement of Africans by Europeans is treated as unique-- and due to unique moral deficiencies in the West." (222)

Moreover, focusing on slavery solely in the West will almost certainly lead one into various errors (219). For example, one is more likely to imagine that contemporary problems for African-Americans are largely caused by the "legacy of slavery". But this doesn't stand up to empirical analysis or anything beyond simplistic, univariate analysis (220).

Sowell notes that slavery was long viewed by "both the secular and religious moralists of societies around the world ...as something requiring no special moral justification" (186). It was done in Africa-- and by Africans trading with Europeans (188). North Africans and Middle Eastern nations dominated the trade for centuries before the Europeans got involved (189). There are contexts where families preferred "slavery" for their daughters rather than a life of poverty but "freedom" (208). (This is reminiscent of the mass immigration to South Africa by blacks, even with Apartheid.) He describes the end of slavery (in the West!) as the result of "moral revulsion" by those in the West, particularly Christians (210-211) and military might (212-214), with the English imposing their moral standards on others, particularly the Ottomans (222, 250)

Finally, Sowell also has an interesting section on the impact of geography (235-240)-- on everything from resource availability (as a secondary cause of prosperity, at best), transaction costs (e.g., the wide-open U.S. vs. countries geographically-bound by mountains or land-locked by political boundaries). Although he does not tease out many particulars, he notes that it would not be surprising if such constraints and opportunities did not shape individuals and societies-- in terms of culture, work ethic, etc. "It would take an almost miraculous coincidence for all these factors to balance out..." (240). If so, then we would expect to find cultural differences.


Other related work on the complicated relationship between biology, culture, etc.:

1.) WSJ article on autism by ethnicity and birth nation

2.) Sue Shellenbarger in the WSJ on the interplay between language (part of culture) and math (comparing English and Chinese, in particular).

3.) Nicholas Wade in the WSJ and a longer essay in Time on race and biology-- as excerpted from his book, A Troublesome Inheritance. (Try googling the article's title, "Race has a biological basis" to get the full WSJ article.) Wade argues that evolutionary theory requires race to be a consideration-- if not a significant factor. (Here's a snarky reply with two deeper cites from PZ Myers; h/t: Chris L. Maybe it's proper to dismiss it, but it doesn't seem like it can be this easy. And if it is that easy, why is that not a problem for the comprehensive-Evolution narrative folks?)