One of the most extraordinary statistics to emerge from the Harvard University admissions scandal cannot be properly understood without some background on the scandal itself.

The Harvard Admissions Case

The Harvard Admissions Case

In 2014, Harvard University became embroiled in a major admissions controversy when Students for Fair Admissions (SFFA) sued the institution, alleging that its admissions policies discriminated against Asian-American applicants while favouring other racial groups. The plaintiffs’ allegation was that Harvard systematically raises the bar for Asian-Americans and systematically lowers it for whites, African-Americans and Hispanics, and that it operates a quota limiting the number of Asians.

The ALDC Finding

The case is relevant here because of the race realist Nathan Cofnas. As part of the case, in 2019, three economists published a paper titled Legacy and Athlete Preferences at Harvard. The researchers made an astonishing discovery. More than 43% of white students admitted to Harvard belong to a category known as ALDC. Among African-American, Asian-American and Hispanic-American admits, the proportion was less than 16%.

That was not even the most extraordinary finding. According to the researchers’ admissions model, approximately three quarters of white ALDC admits would have been rejected had they been treated like ordinary applicants without ALDC status. Three quarters is 75% of the 43% admitted under these special categories. Basic arithmetic shows that only 25% of the white ALDC population would survive, reducing the total white population by a third, while the share of every other racial group increased or remained unchanged.

This finding matters because Cofnas has made a great deal of noise about the supposed destruction of academic meritocracy by wokeness and by diversity, equity and inclusion (DEI), the dreaded DEI monster accused of ruining the very fabric of society. He claims that racial preferences allow black students to occupy places they would never obtain in a genuinely meritocratic system:

“Without any affirmative action, and I point out, you know, just the data show that representation of certain groups, particularly African-Americans, would just fall to a degree that very few people understand.”

Yet here is published research, not speculation, based on Harvard’s own admissions data, demonstrating that a remarkably large proportion of its white population belongs to categories enjoying preferential treatment. Where is Cofnas’s passionate denunciation of these preferences? Where is his outrage about the academically stronger applicants potentially displaced by legacy admissions? Where are the sweeping conclusions about the intellectual capabilities of white Americans, drawn from the discovery that so many white admits benefited from special consideration? If the intellectual merit of entire populations is to be measured through Harvard’s admissions decisions, this information is surely important.

Nepotism and Selective Outrage

ALDC stands for Athletes, Legacies, Deans’ interests and Children of staff. Athletes require little explanation. Legacies are applicants whose parents attended Harvard. Deans’ or directors’ interests are largely special interests and people connected to donors or influential individuals. Children of Harvard faculty and staff are likewise self-explanatory.

Of this list, athletic achievement can certainly represent genuine merit, and nobody suggests otherwise. Being born to someone who attended Harvard, however, is not an academic accomplishment. That is nepotism. Having wealthy or influential connections is not evidence of intellectual superiority. That is also nepotism. Being the child of a Harvard employee does not make a student more deserving. It is nepotism again. ALDC is, in effect, a fancy aristocaucasian way of saying nepotism.

These institutional preferences gave substantial advantages to particular applicants, which may be why Cofnas and the anti-woke brigade have never protested the existence of this anti-meritocratic system. The researchers found that a non-ALDC applicant with a 10% probability of admission could see that probability move to near certainty if they qualified for this preferential category. If preferential treatment in the form of affirmative action is evidence of the corruption of meritocracy, the same principle should apply when it benefits white applicants. Cofnas’s argument also takes for granted that Harvard has shown preferential treatment for black students on the basis of race, which has never actually been proven, only speculated.

Why, then, does Cofnas’s account of Harvard’s admission system focus purely on denigrating black meritocracy while apparently overlooking one of the most substantial examples of unmerited favour, enjoyed by his own particular caste? Not every white ALDC student was academically unqualified, a leap in logic best left to the statistically challenged race realism gang. The finding does, however, irrefutably demonstrate that many owe their admission to advantages that would not have been available to otherwise comparable applicants. That is an enormous problem for anyone presenting Harvard’s existing racial admission statistics as a straightforward reflection of intellectual merit. 30% of whites, gone. Just like that.

None of this is surprising in itself, but for a race realist such as Cofnas it should be devastating. If academic ranking is the gold standard, it should be applied consistently. If unfair advantage, such as inherited privilege and DEI, is incompatible with meritocracy, then both should be condemned wholly. One set of rules cannot be used to question the legitimacy of black academic achievement while another leaves white representation conveniently unexamined, even when it is published in black and white. That is what Cofnas does, consistently. It is disingenuous and dishonest and, frankly, his calling card: bad faith and bad mathematical research. Poor application of data is not unusual for race realists and eugenicists. It is somewhat par for the course, and separate work on IQ testing addresses the point.

Merit Beyond Academic Scores

According to Cofnas, the ideal university admission system is simply a giant spreadsheet. Sort by academic score, take the top 2,000, and a perfect meritocracy results. This is a remarkably narrow conception of human achievement, and it is not how any institution, educational or vocational, has ever or will ever select candidates. Two students might have identical grades, but one may have founded a successful business, another may be an accomplished musician, another may have demonstrated extraordinary leadership, and another may have achieved exceptional academic results despite substantial educational disadvantage.

Harvard’s admissions process assessed academic performance alongside extracurricular achievements, athletic accomplishments, recommendations, personal characteristics and other contextual factors. There is nothing inherently anti-meritocratic about that. If merit is defined exclusively in terms of grades, then of course the applicants with the highest grades will appear the most meritorious. But that is a chosen definition, a single metric, one portion of a CV, and it seldom suffices to secure success in any institution. It can never be an objective discovery about the totality of an individual’s capability until their full résumé is taken into account.

This leads to one of the most controversial elements of the Harvard litigation, the personal rating. Harvard assessed applicants on qualities including leadership, integrity, confidence and other personal characteristics, graded from one to six, with one being the very highest. SFFA argued that Asian-American applicants were disadvantaged by this system because they received lower average personal ratings from Harvard admissions officers than whites and other applicants. Among applicants in the highest academic index decile, the very best academic performers, 47% of African-American applicants received a personal rating of one or two. For white applicants the figure was 29.6%, and for Asian-American applicants it was 22.2%.

This demonstrates how rankings can change when a university evaluates something other than academic results, and it ultimately offers a meritocratic explanation for the higher admission rates of black students. Cofnas attempts to reduce these accomplishments, and it is the height of bad faith to suggest that an African-American applicant who has performed comparatively well in the personal rating assessment is proof of DEI, affirmative action or some other favourability scheme.

Harvard argued that an applicant’s experiences, including experiences related to race, could legitimately inform an individual assessment. That is meritocratic, because a poor demographic cannot simultaneously be told to pull itself up by the bootstraps and then be ignored, or even penalised, when it does. It seems that black Americans perhaps disproportionately show attractive values beyond the academic that make them desirable candidates. And if only academic scores are to be treated as legitimate, it must be explained why leadership, creativity, resilience and exceptional accomplishments outside the classroom suddenly cease to count as merit in Cofnas’s analysis.

From Applicants to Professors

The most extraordinary extrapolation in Cofnas’s wider argument comes next. In an aneurysm-inducing leap of logic, Cofnas argues that in a meritocracy Harvard faculty would be recruited from the best of the best students, which means the number of black professors would approach 0%. The question is how he arrived there. Apparently, the academic achievements of a university professor can be inferred from the examination results of teenagers applying to Harvard.

The leap involved deserves attention. On one side are school students applying for undergraduate admission. On the other are established academics with years of education, research, teaching, publication and professional experience. University professors are not appointed on the basis of their SAT scores, as Cofnas must know. That stops being important after the master’s degree, mostly. Professors are assessed according to the requirements of the academic position: research, scholarship, teaching, academic leadership, contribution to their discipline, and sometimes substantial professional expertise acquired outside conventional academic pathways. These are all matters with which Cofnas is presumably unfamiliar, being a poor employee and a terrible professor.

To determine whether a particular professor was qualified, that professor’s actual qualifications and achievements would need to be examined. What cannot be done is to take the projected racial composition of an undergraduate admissions class and pretend it provides the answer, or any answer. There is no automatic mathematical relationship between the racial composition of Harvard freshmen and the racial composition of the qualified academic labour market, and there is certainly no scientific justification for applying such a relationship to universities across the world. The obvious question is what Cofnas was thinking, but since this ridiculous extrapolation has met with virtually zero pushback, the only available response is to congratulate him on pulling the wool over so many people’s eyes.

A Failure of Inference

If this quasi-interpretation and extrapolation of data is accepted, it need not stop there. The findings bear repeating:

“Asian American applicants are significantly stronger than white applicants on the observable characteristics outside of race that Harvard values when making admissions decisions.“| Asian American discrimination in Harvard admission

“Among those who are in the top 10% of applicants based on grades and test scores, Asian American applicants outnumbered white applicants by more than 45%.“| Asian American discrimination in Harvard admission

“Among white admits over 43% are ALDC. Among admits who are African-American, Asian-American, and Hispanic, the share is less than 16%. Our model of admissions shows that roughly 3/4 of white ALDC admits would have been rejected if they had been treated as non-white.“| Asian American discrimination in Harvard admission

Applying the same logic, the deleterious effect of a colour-blind meritocracy would be catastrophic for the white student population of Harvard. By Cofnas’s logic, this would prove that only Asians should be teaching as professors in universities. Asians would rise to around 60% of the Harvard population, and whites would fall to around 26%. None of this is credible, because it is not science. It is nothing more than a fundamental failure of inference.

The Harvard argument does not exist in isolation. Cofnas has written extensively, almost exclusively, in defence of hereditarian explanations for differences in cognitive ability between racial groups. He has also argued that a genuinely colour-blind meritocracy would produce levels of racial representation that many people would find unacceptable. The problem lies in the selectivity and pseudo-scientific analysis behind the meaning he attempts to extract from the data. A university’s applicant pool is not a representative sample of an entire population, and never has been. Academic scores are not a direct measurement of genetic intelligence, and admissions preferences are not limited to affirmative action. Undergraduate admissions outcomes do not determine the qualifications of future university professors, even in a meritocracy. An academic who wishes to make scientifically defensible claims should know this.

The research that Cofnas’s argument brings to attention, but hopes will go unexamined, shows a rather different picture. 43% of white admits belong to a special preferred category. The majority of those white ALDC admits would not have been admitted without their special status, according to the researchers’ model. Asian-American applicants outperform white applicants on many of the observable criteria Harvard valued, and they faced a modelled admission disadvantage relative to comparable white applicants. None of these findings established a racial hierarchy of innate intelligence.

The lesson is statistics 101. A statistic can be accurate and still be used to construct a completely unrelated and misleading argument. Anyone claiming to have discovered a universal truth about racial intelligence in a Harvard admissions spreadsheet has almost certainly left something much bigger out.

---