These highlights are from the Kindle version of Facing Reality: Two Truths About Race in America: by Charles Murray.
You have to be quite old to remember how uncomplicated it seemed to many of us, White and Black alike, in 1963. African Americans had been wronged for centuries, during slavery and after. It was time to set things right. Ten months later, the Civil Rights Act of 1964 was signed by Lyndon Johnson.
Less than a year later, President Lyndon Johnson announced “the next and profound stage of the battle for civil rights” – namely, “not just equality as a right and a theory but equality as a fact and equality as a result.” It marked the beginning of a process whereby the founding ideals of the American creed were recast as the struggle for social justice.
The twenty-first century saw the growth of a new ideology that repudiated the American creed altogether. It began in academia as intersectionality and critical race theory conjoined with a bastardized vision of socialism. By 2016, it exerted significant influence within the left wing of the Democratic Party. As I write, the new ideology still goes by several names. “Woke” originated within the African American community. “Critical race theory” and “anti-racism” are the most widely used terms. But there’s one label that covers it all: identity politics.
The core premise of identity politics is that individuals are inescapably defined by the groups into which they were born – principally (but not exclusively) by race and sex – and that this understanding must shape our politics. Identity politics turns the American creed on its head. Treating people as individuals is considered immoral because it ignores our history of racism and sexism. Remedying America’s systemic racism and omnipresent White privilege requires that people of color be treated preferentially. The power of the state not only may legitimately be used to this end, it must be so used, and sweepingly.
The new administration came to office in January 2021 with the support of American elites who had largely accepted that the ideals of colorblindness and America as a melting pot were not just outdated, they were evidence of the racism still embedded in the White consciousness. Within a week of his inauguration, President Biden signed four executive orders intended to promote “racial equity,” promising that “we’re going to make strides to end systemic racism, and every branch of the White House and the federal government will be part of that.”
In some ways, there’s nothing new here. The Biden administration is acting on an assumption that has been incorporated into law for more than fifty years: It is appropriate for the government to play racial favorites, to dispense favors and penalties according to the group to which individuals belong.
In the 2019 ACS, 66 percent of all people who self-identified as Latino also self-identified as White, while 29 percent self-identified either as “Other Single Race” or as a combination of two or more races not including White, Black, Asian, Filipino, or Polynesian.
The answers to the Census Bureau’s questions are consistent with the 23andMe genetic findings about Latinos. In that sample, the self-identified Latinos showed 65.1 percent European ancestry, 6.2 percent African ancestry, and 18.0 percent Native American ancestry, leaving 10.7 percent for the rest. But these numbers are far from evenly distributed across the self-identified Latinos.
All this means that it is problematic to lump Latinos into a single group when analyzing either cognitive ability or crime. People who self-identify as Latino span the range from those who have been living in what is now the United States for centuries to first-generation immigrants; from Latinos who are genetically 100 percent European to Latinos who are 100 percent descendants of a specific indigenous pre-Columbian population.
Whites pushed for the label Native American, but most of the people it was intended to please still call themselves Indians.
I substitute European for White, African for Black, Latin for Latino, and Amerindian for Native American. Asian remains Asian.
From the first census in 1790 through the 1850 census, the population within America’s settled regions was 82–84 percent European and the rest was African. Subsequent tides of immigration increased the proportion of Europeans. As of the 1960 census, America was about 87 percent European, 11 percent African, something more than 1 percent Latin, and something less than 1 percent Asian.3 America was effectively a biracial country with one race overwhelmingly dominant in numbers as well as dominant politically, economically, and culturally.
America’s big cities have been transformed by immigration over the past several decades. In 1960, New York was the most cosmopolitan city in America but its population was still more than three-quarters European. New York City went from 77 percent European in 1960 to 32 percent in 2019. That’s a transformation by any definition.
Other major cities changed even more than New York did. From 1960 to 2019, Los Angeles went from 85 percent European to 29 percent; Chicago went from 82 to 34 percent; Houston went from 77 to 23 percent. All the rhetoric about the racial diversity of America is true – for big cities.
Three obvious geographic groupings of American zip codes: Africans in the states that once formed the Confederacy; Latins in the southern half of Florida, the Southwest, much of California, and a few other western concentrations; and Europeans everywhere else.
Big-city America is authentically multiracial, far more so than the major cities of Europe or Asia. The Deep South is biracial, with a European majority and a large African minority, but hardly any Asians and relatively few Latins outside the big cities. The Southwest and California up through the Central Valley are also biracial, but with a different pair of races and an even larger minority, Latins being more than a third of the total population.
IQ denotes the mental agility that lets some people collate disparate bits of information and then infer and deduce from that information better than other people, whether the task is to decipher a corporate balance sheet, analyze Middle-march, or determine why the car won’t start.
Four years later, Section 402 of the Civil Rights Act of 1964 required the commissioner of the Office of Education to conduct a survey “concerning the lack of availability of equal educational opportunities for individuals by reason of race, color, religion, or national origin in public educational institutions.” The sample was massive: 645,000 students in 4,000 schools. The report of the results, known as “the Coleman Report” after its principal investigator, the sociologist James S. Coleman, was submitted in July 1966.
The Coleman Report published the racial means on the study’s cognitive test battery. Subsequent analyses refined the results, finding that the European–African difference was about 15 points for ninth-graders and 18 points for twelfth-graders, but the size of the difference in test scores was ignored in the midst of the consternation created by the Coleman Report’s central conclusion that the quality of a school played almost no role in explaining the performance of African students. This was heresy – the point of the survey had been to prove that poor education was to blame for the problems of African students and better education would be the solution. But the Coleman Report found that the far more important factors were aspects of a student’s family background.
For the rest of the 1960s and throughout the 1970s, race differences in cognitive ability were seldom part of the policy conversation. A notable exception was Arthur Jensen’s 1969 article in the Harvard Educational Review arguing that educational programs were unlikely to close the gap because it was substantially genetic. But that was followed in 1972 by Christopher Jencks’s Inequality: A Reassessment of the Effect of Family and Schooling in America, which made the case for environmental explanations of the gap. Policy analysts embraced Jencks’s analysis and dismissed Jensen’s.
A score that puts someone at the 99th percentile is obviously high, and it is natural to think it doesn’t make a lot of difference to get more precise. But that’s illusory. Every starter on every men’s basketball team for every major university in the country is probably in the 99th percentile on a scale of male basketball ability – but that’s the same percentile that LeBron James is in. Percentiles are grossly inadequate for conveying differences in performance at the high end.
The top figure should also serve as an object lesson in the necessity of judging people as individuals, not members of groups. If you rely on the difference in means you are going to make a huge number of mistakes about individuals.
Among people of the four races with IQs of 100, 70 percent are European or Asian. For IQs of 115, 85 percent. For IQs of 125, 90 percent. For IQs of 140, 96 percent.
The mean differences separating European teenagers from African teenagers in math and reading haven’t diminished since the last half of the 1980s. That’s more than three decades during which hundreds of billions of dollars have been poured into attempts to improve the education of disadvantaged children, including the intense effort to reduce test-score differences through No Child Left Behind.
The best way to avoid becoming a victim of violent crime is to stay away from the parts of town where the violent crime is concentrated at the times of day when the most crime is committed – if you’re affluent enough not to live in those parts of town.
The FBI reports national totals of arrests by race and it reports arrests by city, but the FBI does not report arrests by race for cities.
There is a technical literature about this question, summarized in the note, that supports the validity of arrest rates.5 For example, here’s the conclusion of the largest and most rigorous study, which examined 335,619 incidents of violent crime in which the victim saw the offender: “Multivariate logistic regression results show the odds of arrest for white offenders is approximately 22% higher for robbery, 13% higher for aggravated assault, and 9% higher for simple assault than they are for black offenders.”
The African/European ratios for murder arrests are larger than the ratios for overall violent crimes in all the cities with the single exception of Fort Lauderdale. So too with the Latin/European ratios with the exceptions of Fort Lauderdale and Tucson. In other words, the crime that gets the most careful police attention shows larger racial disproportions in 18 of the 21 comparisons available. This is contrary to expectations if it is thought that the police are getting away with wrongly arresting Africans and Latins for less-scrutinized crimes. In most cases, the ratios for murder arrests were not just somewhat larger than the ones for overall violent crime but substantially larger.
Eight of the thirteen cities that have released arrest data have also released their datasets for reports of crime. Of these, only the New York dataset includes the race of the reported perpetrator. The African/European ratio of reported perpetrators in New York was 14.8, higher than the arrest-based ratio of 11.6. The Latin/European ratio was 3.9, fractionally lower than the arrest-based ratio of 4.1. Neither result is consistent with the hypothesis that arrest data exaggerate minority crime.
Even in zip codes where Africans constitute less than a quarter of the population, African victims identified 79 percent of the suspects as African and 17 percent as Latin – a total of 96 percent of the suspects. In zip codes where Latins account for less than a quarter of the population, Latin victims identified 62 percent of the suspects as African and 31 percent as Latin – a total of 93 percent of suspects.
Many African lives have been taken by violence, but of the 1,906 African deaths in the New York shootings database for which the race of the perpetrator is known, 89 percent were killed by Africans. Ten percent were killed by Latins. Just 0.6 percent were killed by Europeans. Of the 7,858 Africans who were wounded in shootings, 90 percent were shot by Africans, 9 percent by Latins, and 0.4 percent by Europeans.
Across thirteen American cities, including four of the nation’s most important ones, the African arrest rate for violent crime was usually around 9 to 11 times the European rate and the Latin arrest rate for violent crime was usually around 2 to 3 times the European rate.
IQ is not everything. Cognitive ability is nonetheless important in the workplace, and not just for a few intellectually demanding jobs. You have probably observed this for yourself.
You will find brilliant performers of every race in any occupation. That doesn’t negate the relevance of these considerations to group means.
If you’re an employer and want to know whether it’s worth the trouble to give cognitive tests to job applicants, information about the correlation of IQ and productivity for a specific occupation makes it a straightforward matter to calculate the dollar value of hiring someone with an IQ of 100 versus someone with an IQ of 115.
Universities do their best to hide what’s going on. They refuse to reveal mean SAT scores by race, proclaiming that “everyone we admit can do the work.” What they don’t acknowledge is that the admitted African and Latin students, as groups, will be concentrated in the bottom of their classes – and that the people making the admissions decisions know it in advance.
Testimony in the recent case charging Harvard with discrimination against Asian applicants included evidence that the same profile of test scores, GPA, and extracurricular activities that gave an Asian applicant a 25 percent chance of admission gave an African applicant a 95 percent chance and a Latin a 77 percent chance.
The numbers of test takers with a combined verbal and math score of 1500+ were around 900 for Africans and around 3,300 for Latins. Meanwhile, the numbers for Europeans and Asians with scores in that range were about 27,500 and 20,000 respectively.
The granting of certification is typically based on pass/fail with no gradations. You either get your certification or you don’t. This is appropriate for establishing a minimum level of competence, but it means that the people who pass the test represent a wide range of performance, from the barely qualified to the superbly qualified. If they could, most hospitals and most school systems would presumably want to choose the most qualified. But to do so would leave them vulnerable to lawsuits and investigation by the Equal Employment Opportunity Commission if the people they hired turned out to be disproportionately European.
Why are there so few minorities in these high-prestige jobs? It’s a numbers game in which the odds against a Latin achieving one of those positions are high and the odds against an African are prohibitive, even if we assume that there is no racism whatsoever among the employers for high-prestige jobs.
To illustrate, I’ll use the cohort of young Americans ages 25–29, the age at which the potential candidates for such jobs are coming out of law schools, medical schools, business schools, and graduate STEM departments. In 2019, there were 23.2 million Americans in that age group. About 228,000 people in that age group can be expected to have IQs of 135 or higher.
Why aren’t there more Africans and Latins in senior positions in the corporations and institutions that did hire such talented people? Again, it’s a numbers game. Let’s say that an elite IT company in Silicon Valley snags 100 new hires from the 135+ pool in the racial proportions of the pool as a whole. That means 70 are European, 22 are Asian, 4 are Latin, 1 is African, and 3 are a mixture of races or “other.” What percent of new hires of any race in any company rise to senior positions? It depends on the organization, and the definition of senior, but in any case the four Latins are competing against 96 others and the one African is competing against 99 others to become one of the chosen few. Those are daunting odds.
Unless you are familiar with race differences in cognitive ability, you might reasonably be convinced that the absence of African and Latin faces in the highest ranks of the American private sector means that you live in a systemically racist country. If instead you are familiar with those differences and you still want to expose racism wherever it does exist, you will start your inquiries aware that it is inevitable that a large majority of employers of people with extremely high cognitive ability will have no Africans or Latins in those coveted jobs. Among those that do have some Africans or Latins, only a small minority will have even one in the upper echelons of the organization. These results will occur in the absence of any racism whatsoever.
Now, the question is whether differences in cognitive ability translate into significant differences in job performance and productivity. What does it mean for the quality of customer service when one group of retail salespersons has a mean IQ 13 points lower than another group? What does it mean for children’s education if one group of K–12 teachers has a mean IQ that is 15 points lower than another group?
The two most comprehensive meta-analyses are by Philip L. Roth, Allen Huffcutt, and Philip Bobko, “Ethnic Group Differences in Measures of Job Performance: A New Meta-Analysis,” published in the Journal of Applied Psychology (2003); and by Patrick F. McKay and Michael A. McDaniel, “A Reexamination of Black-White Mean Differences in Work Performance: More Data, More Moderators,” published in the Journal of Applied Psychology (2006).
Among accountants, race differences in the pass rate for the Certified Public Accountant exam are commensurate with the race differences in cognitive ability. In the legal profession, the race differences in pass rates for the bar exam are commensurate with race differences in cognitive ability. So are differences in the percentage of attorneys who have been the subject of repeated complaints in California. In the medical profession, race differences in board certification for a medical specialty are commensurate with race differences in cognitive ability.
In big-city America, disproportionate minority crime rates raise the costs of doing business for retailers of all kinds. It is often alleged that large commercial chains avoid putting stores in minority neighborhoods. The empirical part of the allegation is sometimes true, but the inference that racism is to blame does not follow.
Shoplifting is far more common in many big-city minority neighborhoods than elsewhere. It often doesn’t make economic sense for big chain stores, which have business models based on low profit margins, to locate in such neighborhoods. Either they won’t make a profit or they will have to charge higher prices, leaving themselves open to accusations of racist price gouging.
If they take measures to apprehend shoplifters, they risk charges of racism and financial shakedowns through the threat of lawsuits. Actions taken to prevent shoplifting can also put employees at risk of violent confrontations. It’s a no-win situation. Opening a store in a big-city minority neighborhood is often not economically rational. Racism need not have anything to do with the decision.
Attempts to stimulate economic growth in places with high crime rates work only in places that are gentrifying or can be gentrified.
Being a logger or a firefighter carries intrinsic physical risks, but only two ways of making a lawful living are intrinsically dangerous because other human beings may deliberately assault or kill you – being a member of a combat unit in the military or being a police officer.
The differences in the group rates of violence are real and large, and it is human nature, not racism, to take precautions accordingly.
The new ideologues of the far left are akin to the Red Guards of Mao’s Great Proletarian Cultural Revolution of the 1960s, and they are coming for all of us. The comparison is not overblown – not when students demand that an art professor at Skidmore be fired for briefly watching a “Back the Blue” demonstration and successfully intimidate other students into dropping his classes; not when the University of Southern California places a professor on leave after student protests because he used a common Chinese term that sounds something like the n-word; not when a Yale lecturer is subjected to ugly demonstrations over an email suggesting that Yale students should be allowed to make their own decisions about Halloween costumes.
Recent examples of victims as I write include James Bennet, formerly opinion editor at the New York Times, who was forced out of his position because he authorized an op-ed column by Senator Tom Cotton that offended the Times’s “woke” staff. Other examples are the decisions of Andrew Sullivan (New York magazine), Matthew Yglesias (Vox), Bari Weiss (New York Times), and Glenn Greenwald (The Intercept) to leave their positions because of the ideological conformity that was demanded of them.
Jonah Goldberg has described the fragility of the American system by comparing it to a garden hacked out of a tropical jungle. A garden surrounded by jungle is unnatural. The gardeners must tend it with unremitting care lest the jungle return.
The truly grave danger of refusing to confront race differences in means is that it leads in a straight line to thinking that the only legitimate evidence of a nonracist society is equal outcomes. It appears that the Biden administration already accepts that logic. If that’s what the people in power truly believe, and if those equal outcomes continue to elude them, the logical conclusion is that the state must force equal outcomes by whatever means necessary. Once the state is granted the power to engineer equal outcomes by dispensing opportunities preferentially and freedoms selectively, it will be one group versus another, “us” against “them.”
Working-class and middle-class Whites who now see themselves as second-class citizens in the eyes of the government aren’t making it up. Of course they are not enduring anything remotely comparable to the legally sanctioned inequalities that Blacks faced until 1964. But they are now told – by government officials, college administrators, and corporate human resources managers – to get in line behind minority applicants for admission to elite colleges and for employment and promotion in attractive white-collar jobs. Well-to-do Whites can find ways to circumvent this problem, but working-class and middle-class Whites cannot. It has long been my view, first expressed in these words long ago, that aggressive affirmative action is a poison leaking into the American experiment.
Aggressive affirmative action is practiced most sweepingly for government jobs at all levels. At the city level, it affects the selection and promotion of police, prosecutors, public defenders, correctional officers, personnel in the social welfare bureaucracies, healthcare workers on the public payroll, and K–12 teachers in the public schools.
The solution: eliminate all forms of government-sponsored preferential treatment by race. It is not within any government’s power to force racial harmony on its citizens, but it is within the government’s power to strip away the legal and administrative incentives and requirements for preferential treatment according to race.
In 1782, the founders of the newly independent United States chose the motto novus ordo seclorum – “a new order of the ages” – to inscribe on the Great Seal. They were right to do so. The creation of a nation dedicated to the proposition that every individual has the same rights to liberty and the same innate human dignity as everyone else was an unprecedented world-historical event. But I fear that we are nearing a point of no return. We must reaffirm the American creed explicitly and quickly, or this country will become just another big power like other big powers, governed with all the historic oppressions that America tried to cast off. In that event, we will not only have sacrificed our heritage. We will lose foundational freedoms and jeopardize even the rule of law.