Mac Donald says some good things about the real reasons for “disparate impact,” then throws it all away: an object lesson in the futility of speaking half-truths about racial differences in intellectual ability

Heather Mac Donald’s article, “Ricci and the Skills Gap,” published at City Journal and copied at FrontPage Magazine, is a useful and informative, though ultimately disappointing, critique of the “disparate impact” issue that is at the heart of the Ricci case. The entire article is below, followed by my response.

Mac Donald writes:

The Supreme Court tweaked the edges of discrimination law in its New Haven firefighters decision last week, but otherwise left the evasions and euphemisms of that hoary edifice largely intact. This is probably as it should be. It is not for the Court to deconstruct our official legal discourse about race unless it is explicitly asked to do so. But the political branches need not be so constrained. Judge Sonia Sotomayor’s upcoming confirmation hearings are the perfect occasion to question the assumptions that underlie the race industry, since Sotomayor, with her history of launching racism accusations, has been, and promises to remain, an active participant in it.

The main function of the race industry today is to repackage problems of black underachievement as instances of white racism. For decades, the vast majority of alleged discrimination violations have been manifestations of the black-white performance gap, whether in academic achievement, crime rates, or poverty-producing behaviors like illegitimacy and dropping out of school. The race industry cloaks such problems in the language of rights and racism—pushing the achievement gap offstage, keeping alive the phantom of ubiquitous white bias, and generating jobs in the race industry. Thus, employment and educational standards that no one would otherwise think twice about are suddenly viewed as legally suspicious, without any reason to think them flawed except that blacks do not meet them at equal rates.

The New Haven firefighters case, Ricci v. DeStefano, provides a perfect example of how the race industry converts innocuous job practices into markers of hidden bias. As is well known by now, New Haven canceled the results of a 2003 promotional exam for its fire department after no black firefighters scored high enough to be promoted. The successful candidates, 17 whites and one Hispanic, sued the city, claiming that it had discriminated against them on the basis of race. New Haven argued in court that had it promoted the high-scorers, it would probably have been sued by the black test takers for discriminating against them with an apparently biased test. Thus, it was justified in nullifying the entire process to avoid liability for discriminating against blacks, the city maintained.

Ricci raises questions about the use of race in decision-making—especially in government decision-making—that have simmered within discrimination law for years. Does a race-conscious remedy for alleged discrimination against one group, for example, discriminate against another? Does the effort to craft a selection method that produces a racially proportionate result constitute impermissible discrimination? When does the effort to avoid allegedly discriminating against one group become discrimination against another?

Such questions occupy constitutional-law theorists endlessly, and for good reason, since they implicate the core constitutional value of color-blind equal protection before the law. In Ricci, the majority predictably finessed them with the familiar judicial tactic of crafting an evidentiary standard, holding that an employer could take a race-conscious action that discriminates against one group—in this case, the successful candidates for fire department promotion—only if it finds a “strong basis in evidence” to believe that it would face liability from another group if it failed to take that action. But the Court avoided the more basic question raised by the case: Why was the exam being discussed in terms of race and discrimination at all? The New York Times’s Supreme Court reporter, Linda Greenhouse, wrote in an op-ed that the exam “appeared to favor white test-takers.” It did nothing of the sort. It merely favored those who had studied hard and prepared themselves to become captains and lieutenants. But we have been conditioned by the decades-long reign of disparate-impact theory, which the black firefighters would have used in a potential suit against New Haven, to discuss neutral employment practices in racial terms and to entertain the idea that the expectation of moderate cognitive performance is an unfair imposition on blacks.

Arguments offered by New Haven against its own promotional test—embraced, disturbingly, by nearly half of the Supreme Court—demonstrate how desperate the search for bias has become. New Haven pointed out that the exam asked test takers to show “how they would address a particular problem [by] verbally saying it or identifying the correct option on a written test,” in the words of a rival test developer; the city’s implication was that any expectation that someone be able to use language to express ideas would be unfair to blacks. New Haven also objected that the test’s final results were based on a 60-40 weighting of a written and an oral exam, a formula that the department had used for two decades. This objection is purely ad hoc; had the weighting been 40-60, the race advocates would have been just as adamant that it was unfair. Nothing in the record suggests that a different weighting would have changed the pass rate. New Haven’s general counsel further complained that the exam tested the “ability to memorize textbooks but not necessarily to identify solutions to real problems on the fire ground.” But memorizing information is a key mechanism of learning, without which little knowledge is possible. To the extent that the lieutenant and captains’ jobs require the acquisition of additional knowledge, conveying and testing that knowledge in written form is perfectly reasonable. None of these features of the exam would have been considered the least bit problematic if blacks had not performed poorly.

The final charge against the exam is the most desperate of all. Janet Helms, a professor of clinical psychology at Boston College and author of such works as A Race Is a Nice Thing to Have: A Guide to Being A White Person or Understanding the White Persons in Your Life speculated to the city’s civil-service board that black firefighters in New Haven may have developed their own unique, black way of fighting fires in response to unequal opportunity on the job. The portrait of firefighting that the test developer used in drawing up the test might have been biased toward the “white” way of firefighting in New Haven, Helms continued, since two-thirds of the firefighters who submitted analyses of their jobs to the test developer were white. Helms had no knowledge of firefighting, had refused to review the promotional exam, and had sought no exposure to the New Haven Fire Department. She provided not one shred of evidence to support any plank of her theory—which makes her superlatively representative of the race industry.

In light of such fantastical speculations, it is worth reviewing New Haven’s efforts to ensure that the test measured the requisite supervisory skills and did not contain any possible hypothetical bias. The test developer interviewed and rode alone with the department’s chiefs, lieutenants, and captains to identify the knowledge and abilities essential for the positions of captain and lieutenant. At every point, the test developer oversampled minority firefighters while building his portrait of the job. The 100 questions on the written exam were pitched below a tenth-grade reading level. Each of the three-man assessment panels for the oral exam was composed of one white, one black, and one Hispanic firefighter. Third-party reviewers confirmed that the oral exam accurately tested the real-world situations confronting supervisors.

Perhaps in implicit recognition that the case for bias in the New Haven exam was laughably weak, Justice Ruth Bader Ginsburg pulled out the usual race-industry trump card in her dissent: historical discrimination. The majority opinion, she said, “leaves out important parts of the story”—to wit, that firefighting “is a profession in which the legacy of racial discrimination casts an especially long shadow.” It’s no surprise that Ginsburg never bothered explicating her metaphor, because she provided no evidence, apart from the allegedly biased test itself, of how the “long shadow” affects the New Haven Fire Department today. We are just supposed to assume that because firefighting was inhospitable to blacks four decades ago, it remains so now, and that such inhospitality prevents blacks from succeeding on promotional exams. (Past discrimination against Jews by the Ivy Leagues may also be said to cast “an especially long shadow,” yet it has had no noticeable effect on their performance today.)

The Court’s majority and minority opinions are assiduously silent about the only reason why anyone views the New Haven firefighters’ exam through the lens of race at all: the gap in cognitive attainment between blacks and whites. As Stuart Taylor pointed out in his National Journal blog, the average black high school senior possesses the academic skills and knowledge of an average white eighth-grader. For decades, blacks have scored 200 points below whites and Asians on the verbal and math SATs. This skills gap means that it is virtually impossible to devise any job test that measures mastery of a body of knowledge and cognitive expertise that will not have a disparate racial result. To discuss black underperformance on any given task without mentioning the skills deficit is like having a discussion about rain and drought without mentioning cloud formation. As long as black underachievement remains unacknowledged, the old standby—racism—will always rush into the vacuum as an explanation for disparate impact.

This ongoing taboo, in judicial discourse and elsewhere, suggests certain questions for Sotomayor during her confirmation hearings. Does she believe that it is unfair to expect minorities to display knowledge through the use of language, and if so, why? In light of the black-white achievement gap, why should disparate results on a cognitive test create any presumption of bias? What kinds of exams would be fair to blacks? Does it lie within the power of minorities to start achieving at higher rates by staying in school, studying harder, and paying more attention in class? We have heard repeatedly about the studying marathon engaged in by the lead plaintiff in the firefighters case, Frank Ricci. None of the many advocates charging bias in the New Haven firefighters’ exam has produced any examples of black firefighters’ expending anywhere near Ricci’s eight to 13 hours a day of preparation. Presumably, if such cases existed, we would have heard about them by now. Is it possible that if more black firefighters had devoted as much effort to studying as the successful candidates did, they would have done better on the exam?

It will take more than one judicial confirmation hearing to overcome our evasions about today’s most pressing racial problems. But those problems won’t get solved until we break the monopoly on explanations for black underperformance that the race industry currently enjoys.

[end of Mac Donald article]

LA writes:

Near the end of her article, Mac Donald finally touches on “the gap in cognitive attainment between blacks and whites,” on the “skills gap,” on the four year difference in reading abilities between blacks and whites, and on the fact that blacks score 200 points below whites and Asians on the verbal and math SATs. That’s a rich dose of racial truth for City Journal and FrontPage Magazine, and it gets close to the edge of saying that the reason for the racially disparate impact of various tests is that blacks are on average less intelligent than whites. But it doesn’t cross over the line to the forbidden realm, because the reference is still to “cognitive attainment” and “skills,” rather than to innate intelligence as such. However, having gotten reasonably close to the truth, Mac Donald then slips further from away from the truth, by speaking of “black underachievement.” As though it were blacks’ fault. As though the problem were that black haven’t tried hard enough. That’s not the case, and it’s not fair to blacks to say that it is. The simple truth is that blacks are consistently far behind whites in all measures of cognitive attainments and skills because blacks are on average significantly less intelligent than whites. MacDonald, while criticizing the taboo on speaking the truth about racial differences in intellectual abilities, won’t speak it herself. I can’t blame her for this, since, even if she had wanted to name the problem in plain language, her editors at City Journal, not to mention David Horowitz at FrontPage Magazine, would not have allowed her to do so, just as Horowitz did not allow me to do so in my 2005 article, “Guilty Whites.” He permitted me only to speak of blacks’ failure to perform as well as whites, without going into the reason for that failure.

However, that limitation on the argument is not necessarily fatal. Mac Donald has presented an effective case that conservatives could make without reference to intrinsic racial differences in ability. All they would have to do is say that the demonstrated racial gap in achievement, not white bigotry, is the reason for “disparate impact.”

Thus Mac Donald writes:

This ongoing taboo, in judicial discourse and elsewhere, suggests certain questions for Sotomayor during her confirmation hearings. Does she believe that it is unfair to expect minorities to display knowledge through the use of language, and if so, why? In light of the black-white achievement gap, why should disparate results on a cognitive test create any presumption of bias? What kinds of exams would be fair to blacks?

That’s good. The problem, of course, is that mainstream conservatives will not ask the questions Mac Donald suggests they ask. They will not refer to the black-white achievement gap, let alone to a black-white intelligence gap. Which is why I always say that we should go for the whole truth, because ONLY the whole truth can break through to people’s minds and free them from the egalitarian fiction by which they are at present mesmerized. The achievement gap is not the truth that will make them free.

But then, as if in proof of what I just said about the futility of half-truths, Mac Donald blows it herself. She continues:

What kinds of exams would be fair to blacks? Does it lie within the power of minorities to start achieving at higher rates by staying in school, studying harder, and paying more attention in class? We have heard repeatedly about the studying marathon engaged in by the lead plaintiff in the firefighters case, Frank Ricci. None of the many advocates charging bias in the New Haven firefighters’ exam has produced any examples of black firefighters’ expending anywhere near Ricci’s eight to 13 hours a day of preparation. Presumably, if such cases existed, we would have heard about them by now. Is it possible that if more black firefighters had devoted as much effort to studying as the successful candidates did, they would have done better on the exam?

Mac Donald’s argument is based on a possibility that is a pure chimera. Yes, if blacks worked much harder, their performance would improve somewhat, but their performance is not going to become equal—or anywhere near equal—to that of whites. And therefore the disparate impact will remain, and the blaming of the disparate impact on white bias will remain. Mac Donald knows as well as I that the large black-white gap in SAT and reading scores is a proxy for a large black-white gap in IQ, meaning a large black-white gap in innate intelligence. Yet she goes out of her way to construct a scenario suggesting that if blacks studied like fireman Ricci, month after month, on weekends, they could eliminate the racial gap in intellectual achievement. That delusory hope, spread by Mac Donald, will help fuel yet more futile charter schools, more fraudulent “No Child Left Behind” laws, more dishonest books by Stephan and Abigail Thernstrom claiming excitedly that as a result of improved educational techniques the “racial gap is being closed,” when what they actually mean is that the racial gap is being “closed” from an unimaginably huge four year gap in reading abilities to a merely inconceivably huge three year gap in reading abilities. And then, when these improvers of mankind finally admit that the gap has not been closed, they will come up with yet more schemes to close it, all the while saying that it is whites’ responsibility to make this happen, and their moral guilt if they fail to make it happen.

The only way to avoid the kind of outright fraudulent suggestion that Mac Donald unfortunately falls into at the end of her article is to stop speaking half truths, because half truths are half lies and inevitably morph into full lies, just as Mac Donald’s language morphs from the half-true “cognitive attainment” gap into the completely false “hard work” gap. We must instead speak the whole truth. Someday, it will become possible to do so in mainstream society. And even if that remains impossible for the time being, by speaking the truth now, we prepare for the day when it will become possible.

* * *

(See my discussion of the Thernstroms’ dishonest treatment of the racial “gap” here, here, and here.)

- end of initial entry -

David Levin writes:

WOW!! Excellent critique on Ms. MacDonald’s piece. I have to admit I haven’t read Heather MacDonald’s work since reading her study of illegal alien crime in Los Angeles County. I thought what she wrote an exceptional piece. But that’s because I’m not sold on the idea that blacks are intellectually inferior to whites. [LA replies: I do not speak of blacks or any group as being “inferior” to another group. It’s precisely such loaded, negative language that makes it impossible for people calmly to consider the facts. If you had a brother who was less intelligent than yourself, would you say he was “intellectually inferior”?]

Good point about “mainstream conservatives” by which I assume you mean Sen. Lindsay Graham who (in spite of his misgivings about what she’s written and her decisions) is going to stab other conservatives in the back by confirming her. I call them “faux conservatives.”

You know, the often under fire Prof. William Shockley (who was made into a pariah by the left because he found through his studies that blacks were inferior to whites in terms of IQ) lived five houses from our house at Stanford. My liberal father loathed him as he did all Republicans and conservatives.

If blacks ARE inferior to whites IQ-wise, then we need more columns like yours and the discussion needs to get out there. Thanks for the other links. I’ll check them out tomorrow.

Rose writes:

Excellent post, but I believe that our inability to admit the truth about black intelligence must be located in a larger cultural context. Intelligence in general is so overvalued by contemporary society that any implication that blacks are inferior to whites in this regard to many people would imply that they are inferior human beings. When liberals claim that the acknowledgement of racial differences will lead to mistreatment of blacks, I have to wonder if this is because in a way they feel that such ill-use would be justified if the black race’s lower intelligence were indisputably proven. It should be emphasized instead that blacks do not have to earn their right to be treated with respect by passing tests and earning degrees, for it is theirs by value of their humanity.

You write:

“Mac Donald then slips further from away from the truth, by speaking of ‘black underachievement.’ As though it were blacks’ fault. As though the problem were that black haven’t tried hard enough. That’s not the case, and it’s not fair to blacks to say that it is.”

This is a very important point. It is also cruel to black children to expect that they all attend college and to consider them failures if they do not become credits to their race. I’m sure that many black students realize the impossibility of their achieving these lofty goals and that this leads to hopelessness. I have long been in favor of allowing all children the option of choosing an apprenticeship over high school. “No race can prosper till it learns that there is as much dignity in tilling a field as writing a poem,” said Booker T. Washington, and black children (especially black boys) should be taught that there is no less dignity in aspiring to simple blue-collar work than to be a professional.

Andrew E. writes:

Your entry today on Heather MacDonald’s piece about the Ricci decision reminded me of an article written last week by Andrew McCarthy over at NRO responding to Ponnuru’s New York Times op-ed concerning the same. Recall that Ponnuru was arguing that constitutional originalists could not argue, with respect to the 14th Amendment, against that kind of discrimination at issue in the Ricci case (together but unequal, as you say) just as originalists do not argue that the 14th Amendment prohibits separate but equal types of discrimination. You have argued that the proper and classic originalist understanding of the 14th Amendment would prohibit together but unequal, but not separate but equal arrangements. Well, relying on a 1995 academic article written by then University of Chicago law professor now federal appeals judge, Michael McConnell, McCarthy argues that the 14th Amendment does indeed have an equality mandate, thus rendering unconstitutional both separate but equal and together but unequal arrangements. I’m not well read enough in American jurisprudence to speak authoritatively but the classic understanding that you’ve put forth always made the most sense to me. Rather than try to summarize McCarthy’s (and McConnell’s) reasoning, I’ll simply link to the article below. Nonetheless, I find it somewhat humorous and not a little ironic that originalists have found ways to cover all the bases here.

LA replies:

I’ll have to read it. On the face of it, McConnell’s position is absurd, since from the writing of the 14th amendment in 1868 until the Brown decision in 1954 it was universally understood that the amendment did not impinge upon education in particular and that it allowed for separate but equal arrangements in general. The case law was so well settled that the Court in the Brown decision did not challenge it directly, but did an end run around it, by declaring a new principle that anything that disadvantages black people (in this case, anything that makes them feel inferior by keeping them in separate schools from whites, a fact “proved” by the finding that black girls preferred white dolls to black dolls showing that blacks felt put down), by that fact violeates the 14th amendment.

Diana Hull writes:

Thanks for always speaking the truth. But this truth is particularly distasteful and I think it could be modified some while still being consistent with the evidence. It is also a true statement that the within group difference are greater than the between group differences.

LA replies:

The statement, “The within group difference are greater than the between group differences,” is a frequently used statement which seems to mean something but is meaningless. It’s meaningless because the between-group differences are differences between the mean of one group and the mean of another group. But the mean of a single group cannot be different from itself. Any differences within a group are differences between individuals belonging to that group. So it’s an apples and oranges situation.

Further, the differences within a single group do not lessen or change in the slightest the significance of the changes between group. 16 percent of whites have IQ above 115. 2.3 percent of blacks have IQ above 115. The reason there is a vastly lower proportion of blacks working in intellectually demanding fields is that the proportion of blacks capable of working in those fields is vastly lower. Absorb that monumental fact, and you will see that the statement, “The within group difference are greater than the between group differences,” does not alter it in the slightest.

Posted by Lawrence Auster at July 15, 2009 01:51 AM | Send