In response to:
Discrimination and Thomas Sowell from the March 3, 1983 issue
To the Editors:
If imitation is the sincerest form of flattery, straw men must be a close second. No critic would have to invent a position to attack if he could deal with the real arguments.
Christopher Jencks’s voluminous review of Ethnic America and Markets and Minorities [NYR, March 3, March 17] repeatedly creates straw men, with which he is repeatedly “astonished.”
Straw Man: According to Jencks, my position is that “discrimination tends to disappear once markets become competitive.” Markets and Minorities: “The competitiveness of the market puts a price on discrimination, thereby reducing but not necessarily eliminating it” (pp. 39-40).
Straw Man: “Discrimination…tends to disappear once government stops enforcing it.” Ethnic America: “Translating subjective prejudice into overt economic discrimination is costly for profit-seeking competitive firms, although less so for government, public utilities, regulated industries like banking, or nonprofit organizations such as universities or hospitals” (p. 292).
Straw Man: “…current discrimination in the labor market has no effect on black earnings.” Ethnic America: “The point here is not to definitively solve the question as to how much of the intergroup differences in income, social acceptance, etc., have been due to the behavior and attitudes of particular ethnic groups and how much to the behavior of the larger society. The point is that this is a complex question, not a simple axiom” (p. 294).
Straw Man: “…if the median black, Indian, or Hispanic were as old as the median European, he would be almost as affluent.” Markets and Minorities: “Differences in median age are only part of the statistical picture” (p. 11).
Straw Man: “Sowell argues for patience.” Since I don’t even discuss any such thing in either book, there is nothing to quote. But in the first book I wrote on the subject, I said at the outset: “History…gives little support to the view that time automatically erodes racial aversions, fears, and animosities, or even tames the overt behavior on such feelings” (Race and Economics, p. vi). All my subsequent books on ethnicity make it a point to show retrogressions in intergroup relations over time, as well as examples of progress. Markets and Minorities mentions these retrogressions (pp. 72-73) and Ethnic America discusses them repeatedly (pp. 71, 82, 161-162, 192, 201-202, 203, 204, 206, 210-211, 254). I shall leave it to Jencks to supply the pages where I argue for “patience.”
Straw men are not incidental but fundamental to Jencks’s critique. Neither Ethnic America nor Markets and Minorities is a book about “special treatment for blacks,” however much that theme is proclaimed in Jencks’s headlines or repeated in his text. Anyone who seriously expects to find in these two books anything more than the most fleeting mention of that subject is going to end up feeling like he is looking for a needle in a haystack.
“What is in question here is Title VII of the Civil Rights Act,” Jencks declares at the beginning of his second review article. Although about half the article is about Title VII, and my supposed opposition to it, there are no quotes of anything I ever said about Title VII. There could not be any quotes from these two books on that subject, because there are none there. Jencks and his straw man do this whole act alone, with only passing references to what “Sowell seems to believe,” what “Sowell assumes” or “as Sowell would have it.” It is very much like “Duffy’s Tavern” or “Rebecca,” where the persons in the title role are repeatedly mentioned but never actually appear.
This is all the more remarkable because I have not been the least bit reticent about discussing Title VII—often and at length—elsewhere. Jencks need hardly have strained his imagination to try to extract my views on Title VII from these books—which are on different subjects, as anyone can tell simply by reading them, however much Jencks may choose to say that they are “in effect” books on affirmative action. Phrases like “in effect” are a blank check for straw men.
Where I have discussed Title VII,1 the first order of business has been to distinguish its provisions, and the legislative history behind them, from the policies subsequently evolved as “affirmative action.” It is the latter that I have criticized, not Title VII.
Jencks’s statistics on what has happened since 1964 might be relevant to Title VII, but not to “goals and timetables” (quotas) that became mandatory in 1971 affirmative action guidelines. Where I have discussed this issue, I have compared the progress of minorities under Title VII’s “equal opportunity phase,” as contrasted with the affirmative phase that followed. Jencks lumps it all together and credits post-1964 improvements precisely to the “affirmative action” phase, when they generally slowed down.
It is also worth noting that the historical trends at work go back even before the Civil Rights Act of 1964. The number of blacks in professional, technical, and similar high-level positions doubled in the decade preceding passage of the Civil Rights Act. The sharp improvement in the relative position of Asians likewise preceded both “equal opportunity” and “affirmative action” policies.
Asians, incidentally, disappear mysteriously from Jencks’s arguments at the oddest moments. After unveiling “the crucial distinction between universal and haphazard discrimination,” Jencks applies this great dichotomy only to blacks and whites. Surely no one would claim that the massive discrimination against the Chinese and Japanese that pervaded generations of American history was only “haphazard.” Black West Indians are also not just black haphazardly (on the weekends, as it were), even though their incomes are just 6 percent below the national average—and 15 percent above the national average in the second generation.
My discussions of the effect of intergroup differences in age and geographic distribution are twisted to fit into Jencks’s framework, where I am supposed to be denying that discrimination has any effect on minority incomes. Specifically, they are transformed by Jencks into attempts by me to deny black-white differences. But Ethnic America and Markets and Minorities are no more books exclusively about blacks than they are books about “special treatment.” Both books give examples of how taking age or geographic distribution into account changes the statistical picture—as regards Hispanics and Asians, for example, as well as blacks. Indeed, Ethnic America points out that Japanese American income is exaggerated when regional and rural-urban differences are not considered (p. 177). Yet Jencks attempts to make the very same argument a telling counterattack to my alleged denials of discrimination. It may take two to tango, but Jencks has brought his own straw man.
Dismissing as “nonsense” the idea that age affects income, Jencks argues that age differences among heads of family “vary little from one group to another” and “have almost no impact on family income.” Census data published in the same Urban Institute book cited by Jencks show Puerto Rican family heads to be a decade younger than Japanese American family heads—and fourteen years younger than Jewish family heads.2 These are hardly trivial differences in work experience.
Some ethnic groups, of course, differ little in age or income, and it is by no means clear that their small differences in income have much to do with their small differences in age. Jencks seizes on precisely such groups to try to prove his point. Taking Czechs and Germans, or British and French, in America as his examples, Jencks shows that their relative income rankings in Europe are different from their income rankings in the United States. But here we are talking about groups whose incomes differ by a few percentage points—the kind of differences small enough to change from one census to the next. The real issue is about groups that differ by large amounts in income. Here an international comparison would be devastating to Jencks’s thesis.
Germans and Jews are not only more prosperous than Hispanics in the United States; they are more prosperous than Hispanics in Hispanic countries. The Chinese not only have higher incomes than blacks in the United States but also in black countries such as Jamaica. The Japanese not only have higher incomes than Mexicans in the United States, but also when comparing Japan and Mexico—even though Mexico is rich in natural resources and Japan is almost destitute of natural resources. The Chinese not only make higher incomes than the Malays in Singapore, which is Chinese, but still more so in Malaysia.
Not only do large economic differences between groups repeat themselves in country after country; the very occupations and fields of study of these groups also repeat. Pianos were first made in the United States by Germans and most of the leading piano firms in the United States today still reflect that German influence in such names as Steinway, Knabe, and Schnabel. The first pianos in Australia were also produced by Germans. So were the first pianos in England. The Chinese are as vastly overrepresented among students of mathematics, science, and engineering in Malaysia as they are in the United States. Jewish peddlers followed in the wake of the Roman legions and in the wake of the wagon trains on the American frontier—and in many places and times in between.
Even the most textbookish exercise in Markets and Minorities is twisted to fit Jencks’s dogma that I deny that discrimination has any effect whatever on minority incomes. In order to illustrate the point that the same set of data can often point to opposite conclusions, according to the level of aggregation or the analysis used, Markets and Minorities reproduces data on blacks and Hispanics from a study by Eric Hanushek and shows the opposite interpretations that can be drawn. Then the same thing is done with some census data on the Chinese. The conclusion of this lengthy exercise is explicitly methodological:
The main point here is that gross income deficits between a particular group and the “national average” do not not prove or measure discrimination—nor do gross income advantages disprove it (p. 26).
Enter Jencks, lower left. Because I have said, in the course of this exercise, that a particular table does not prove discrimination, Jencks turns it completely around and claims that I have said that that one table disproves discrimination against blacks and Hispanics. He finds this conclusion “astonishing.” I am a little staggered myself, especially when he follows that up by claiming that I have falsely cited Eric Hanushek in support of this novel doctrine that he attributes to me. Lest anyone think that Professor Hanushek is an unindicted co-conspirator, let me point out that he plays exactly the same role in this example as the Census Bureau: He is listed after “Source” (and “adapted from”) under a table of numbers. He is neither mentioned, quoted, nor cited in the text where I drew my conclusions. I haven’t the faintest idea what his opinions are on this subject.
After his empirical forays, Jencks turns philosophical. He criticizes those economists who prefer individuals to make their own decisions rather than have government impose decisions on them. This, according to Jencks, invests individuals with “implausible virtues” and proceeds as if “private citizens always act in their own best interests.”
