Minding the Sciences — Death of a Science Academy

The National Academies of Sciences, Engineering, and Medicine forget … Science

In July 2020, just two months after the killing of George Floyd, chairwoman of the U.S. House Committee on Science, Space, and Technology, 84-year-old civil-rights pioneer Eddie Bernice Johnson, asked the National Academies of Sciences, Engineering, and Medicine (NASEM)

[T]o take action on research and policy related to diversity, equity, inclusion, and the racial biases in the nation’s systems that disadvantage people from marginalized backgrounds in pursuit of science, engineering, and medical studies and careers. … Remarks from the National Academies leadership reflected a commitment to removing the barriers to participation in STEMM rooted in deep systemic and structural inequities [emphases added].

In response, under the heading CALLS FOR URGENT ACTION, National Academy of Sciences (NAS) president Marcia McNutt convened a “national summit” in July 2021, which

highlighted how racism operates at different levels in science, technology, engineering, mathematics, and medicine (STEMM) settings; reviewed policies and practices for confronting systemic racism; and explored ways to advance diversity, equity, and inclusion (DEI) in STEMM settings.

This “summit” comprised nine members, including Susan Fiske (Princeton University, psychology) and Gilda Barabino (Olin College of Engineering, biomedical and chemical engineering). The selection of experts was not random or even representative of the range of relevant disciplines. Most members are involved in DEI work; none, as far as I can see, is an expert in the study of individual and group behavioral differences.

The congresswoman’s request should have placed the Academies in a difficult position, because it contains assumptions about racial bias and disadvantage that are not settled science. A proper response would have been to convene a balanced group of researchers who could accurately describe the current state of knowledge on those issues. Or, President McNutt could have declined to comment, on the grounds that the science is too uncertain and the topics too political to be proper subjects for an NAS position paper. Of course, she did neither, because Ms. Johnson’s position is no different from her own, as her earlier comments complaining about too many white males in STEM and vowing “to change the face of the Academy” reveal.¹

The Mission of the National Academies is to “provide independent, trustworthy advice and facilitate solutions to complex challenges by mobilizing expertise, practice, and knowledge in science, engineering, and medicine,” and their Vision is “A nation and a world that rely on scientific evidence to make decisions that benefit humanity.” Not too specific, but, clearly, scientific evidence and knowledge in science are important. So, one might expect, as a first step, that the scientists on this committee would look at some of the assumptions in Congresswoman Johnson’s request rather than simply echoing them, as Dr. McNutt’s remarks suggest.

A committee asked to look at water pollution, say, would not take for granted an initial suggestion that mercury is the cause. Before recommending action, it would analyze water samples and look into the possibly toxic effects of all the chemicals they contain. In the same way, we might expect this committee to examine Johnson’s assumptions about bias and systemic racism. Is there racial discrimination? In what sense is it “systemic,” and so on? Are there other causes for racial disparities in STEMM? It should find the causes before recommending remedies. But, since Johnson’s assumptions are also McNutt’s, it did not.

Of course, social science is less settled than chemistry. It is much harder to decide on the causes of, say, the underrepresentation of certain racial groups in STEMM than on the causes of an ailment traceable to the toxic effects of a known chemical. There are many contributors to academic performance, and their effects are likely to be long delayed. Experiment is usually impossible. No matter: the very difficulty of the task just underlines the importance of advancing understanding before recommending action.

But this is not at all how the National Academies’ workshop reacted. “Racial bias” was illustrated by a single study showing that “bias in hiring is so strong that Black men’s resumes were only viewed as competitive as their White counterparts if the White applicant’s history included a criminal history,” which does indeed look like bias. On the other hand, the response of the employer was just a callback, not an actual hiring decision, and there is no evidence that the employer was wrong in his estimate of potential employee quality. Above all, this is not evidence of discrimination in STEMM, which is the topic of the report. The committee found no evidence of racial bias in STEMM, other than a statistical underrepresentation of some groups that were promptly labeled disadvantaged. The committee wholeheartedly accepted Congresswoman Johnson’s premise that people from “marginalized backgrounds” are “disadvantaged,” rather than different from the majority in some other way. In short, it abdicated its duty to establish the scientific truth of the assumptions presented to it before recommending action.

But don’t despair. The little-known discipline of “diversity science” is available:

Diversity science recognizes both the individual-level psychological tendencies that occur in individuals and the broader sociocultural frameworks that explain the complex implications for behavior.

The language is jarring, but the ideas are basic. Yes, everyone recognizes that individual interests and abilities—“individual-level psychological tendencies”—as well as the societal environment—“broader sociocultural frameworks”—exist and can potentially “explain … behavior.” Unfortunately, this report focuses exclusively on the environment, on discrimination. Individual differences in aptitude, interests, education, etc. get no attention whatever. It is as if a committee assigned to study water pollution obsessed over mercury, ignoring all other possibilities.

[Related: “Psych! You Don’t Have the Job.”]

In the end, the 2021 recommendation was hortatory: “The Summit has been a call to ensure that people from all backgrounds have equitable support and resources to pursue their educational and professional goals.” Who can object to that (although more detail on just what “equitable” means would be helpful)?

The 2023 Report

The 2021 report is preliminary and relatively short, just twelve pages. But now, two years later, NASEM has issued a much longer and even less balanced report, the 450-page Advancing Antiracism, Diversity, Equity, and Inclusion in STEMM Organizations: Beyond Broadening Participation. The four editors of the report were Drs. Fiske and Barabino, mentioned earlier, and two NASEM staff members. Two committees were involved: the Committee on Advancing Antiracism, Diversity, Equity, and Inclusion in STEM Organizations (18 members) and the Board on Behavioral, Cognitive, and Sensory Sciences (12 members). Eleven reviewers were consulted: so, a total of almost fifty people. I’m not sure who the actual writers were.

The weight of credential-prestige (Referenzengewicht?—there must be a German word for it!) behind this second report is considerable; diversity, in the sense of a range of expertise and values, much less so. As with its predecessor, most of the professionals involved seem to have a major interest in racial-equity issues and, correspondingly, an allegiance to what one might call the “standard DEI model.” The committee “elected to focus on challenges facing Black Americans in STEMM.”

Again, there seem to be no IQ experts, no one interested in behavioral differences between groups. Indeed, the only reference I could find to the intelligence issue pertains to slavery-era claims of racial differences in “intelligence, industriousness, ingenuity, sexuality, and criminal behavior,” which were justified by “pseudoscientific intelligence testing, or measuring brain sizes—measurements that were later shown to be fraudulent (e.g., Gould, 1978).” This refers primarily to the work of physician Samuel Morton. The committee—or, at least, one of its eleven reviewers—should have known that Stephen Jay Gould’s claim is the false one. Morton’s brain-size measurements have been fully substantiated (New York Times, 2011). Gould’s attack on IQ measurement in the revised edition of his book The Mismeasure of Man has also been thoroughly debunked. The one factual claim in the report in support of the standard “disparities are the result of systemic racism” model is not true, and alternative accounts are totally omitted.

The NASEM committee doesn’t just promote Gould’s mendacious criticism—it seems not to understand the issue. In what sense are intelligence tests ‘pseudoscientific,’ for example? They do not pretend to measure some kind of Platonic essence of intellectual capacity. They are just tests that make useful predictions: a high IQ, for example, predicts better academic performance than a low one, just as high blood pressure predicts a higher risk of stroke. There’s nothing ‘pseudoscientific’ about predictions like these, which are as true for blacks as they are for whites.

In fact, any possibility that behavioral differences might play a role in below-average STEMM outcomes for “disadvantaged people” is simply ignored in the report. There are hundreds of references (including one to a paper co-authored by notorious fraudster D. A. Stapel), but nary a one on individual and group differences in attitudes, interests, and abilities. (The report does, however, add to the expanding DEI vocabulary. In addition to “minoritized,” there is “latine,” an alternative to the now-unacceptable “latinx.” ‘Controversially,’ the report capitalizes “White,” even though it “risks legitimizing White supremacy.” Linguistic purity seems to be more important to the writers than scientific scrupulousness.)

Neither space nor attention span permits a line-by-line analysis of this weighty tome. A summary will have to suffice.

Let’s start with definitions. Systemic racism is a very popular term in this literature. The report defines it thusly:

Systemic and structural racism are forms of racism that are pervasively and deeply embedded in and throughout systems, laws, written or unwritten policies, entrenched practices, and established beliefs and attitudes that produce, condone, and perpetuate widespread unfair treatment of minoritized people (Bonilla-Silva, 1997).

This is not a usable definition, because the term racism is not itself defined, nor do we know what “pervasively and deeply embedded …” means. (Eduardo Bonilla-Silva is best known for his book Racism without Racists, in which he claims that colorblindness is itself racist, a paradoxical, not to mention nonsensical, claim that has nevertheless caught on.)

The next paragraph claims that “Structural racism describes ‘cultural values in a society that are so ingrained in daily life that they are seen simply as the way things are,’” which is also not very helpful. The systemic racism literature never reveals how this supposedly omnipresent force can be measured. The best I can find is the comment a few sentences later that “Institutional racism denotes policies and practices within and across institutions that, intentionally or not, produce outcomes that chronically favor White people,” which sounds usable but raises a few questions: What if “Black people” are favored? Is institutional racism defined solely by disparities? Is that what is meant by “chronically favor”? Is any racial disparity proof of racism? This is, in fact, assumed throughout the report. Disparities mean racism, and their reduction, by almost any means, therefore counts as “antiracism” even if this requires discrimination against a majority.

The sad history of anti-black racism in the U.S. is treated at length. Its causal relevance to present-day problems is questionable, but its emotional effect is undeniable. Anti-Asian racism is mentioned early on, but the emphasis is on anti-black racism, perhaps because Asian Americans have, in fact, surpassed whites by most socioeconomic metrics in recent decades. The report does not speculate on the relative success of Asians compared to other ‘minoritized’ people. The assumption is that the uniquely tragic history of anti-black racism bears all the blame for any disparities.

[Related: “Science is Rotting from the Top”]

There is a lengthy chapter on race and ethnicity in the report, even though the actual issues do not pertain to race, as such, but to the relative social success of different self-identified groups. The genetics of race play no part in these issues, but the topic serves to distract and alarm.

There is a whole section on “Representation by Race and Ethnicity in U.S. STEMM Higher Education,” which concludes that “Persons identifying as Black or Latine or Hispanic are underrepresented among medical school attendees compared to the percentage of 18–24 years who identify with those racial and ethnic identities.” However, “Persons who identify broadly as Asian are overrepresented numerically in medical schools.” Which leads to a key conclusion of the report:

Although the representation of minoritized persons in STEM higher education is increasing, the collective attainment of science and engineering degrees for Black people, Indigenous people, and Latine people does not reflect their corresponding growth in the U.S. population.

In other words: unless “the representation of minoritized persons in STEM higher education” matches the population proportion, a remedy is needed. This is nonsense that violates common sense as well as basic morality. Unless there is clear evidence of racial discrimination, a discrepancy of any kind is no cause for action. Many professions are favored by different groups—nursing and women, engineering and men, etc. There is absolutely no reason why STEM should be any different. But the NASEM report ignores reason and continues its quest for an unreachable, unfair, and unconstitutional balance.  More, much more, racial data is needed to forward this quest:

data collection organizations … should collect and share with the public information on the demographics of students entering college planning to study STEM and their subsequent educational outcomes, disaggregated by race and ethnicity, gender, and field of study …

What on earth is the point of this obsession with disparities? Disparities by themselves settle nothing. They may raise questions; they never provide answers. They are uninterpretable. Nevertheless, this report, like so many others, always draws the same unwarranted conclusion: It’s racism, stupid!

Instead of scientific analysis, the next section is Chapter 4: Lived Experiences and Other Ways of Knowing in STEMM, a title that most actual scientists will find jaw-dropping. The text explains:

… this chapter provides evidence from the lived experience and other crucial sources of information beyond traditional quantitative methods [emphasis added].

Has NASEM made a breakthrough, finding a new method of inquiry beyond science? Let’s digress for a moment to consider what separates science from other forms of knowledge and belief, an obligation that should be first among the priorities of the National Academies.

Science depends on beliefs that are not themselves scientific. They cannot be proved by the methods of science. Yet, without them science ceases to exist. Unfortunately, these beliefs are rarely made explicit. They are also relatively new; few believed them before the sixteenth and seventeenth centuries in Europe. They have since been taken for granted, but because scientists are rarely made aware of them, they are easily forgotten. They include a belief in open discussion, honesty, curiosity, and single truth, attainable through rational, empirical methods that are accessible to anyone and always subject to revision in light of new knowledge. Standpoint epistemology, the amusing idea that each of us has his own truth, has no place in science, although at times the report seems to verge on accepting it.

Doing science, especially human science in a hot-button area like race, also demands dispassion, the ability to separate the scientific value of a fact from its emotional effects. The only relevant property of a scientific claim is its truth or falsity, as best as these can be assessed by empirical methods open to any rational person. A scientist faced with the claim that, say, blacks are less intelligent than whites may, wearing his non-scientific hat, be revulsed. But as a scientist, he must ask questions: What do you mean by “black”? By what measure are blacks allegedly “less intelligent”? Do you mean all blacks or only some? And so on … This view of science is clearly not shared by the writers of this report.

[Related: “Minding the Sciences — Science Should Leave the University”]

The facts of science must be third-party accessible; personal conviction is not enough. “Lived experience” is not “third-party-accessible” data, so it’s not part of science. In a recent interview, British politician Ann Widdecombe reported hearing a woman MP in the UK Parliament complain, “the men are so rude to me!” She was not comforted when the response was, “Well, they are just as rude to each other.” A black job applicant bemoans that he is being discriminated against; the response is, “You failed your examination!” What these folks say counts as data; their “feelings” do not. So, by third-party criteria, the woman MP is not being victimized for her sex, and the black applicant is not being discriminated against for his race. If they feel bad, the solution lies with them, not the environment about which they complain.

“Belonging” is a major topic in Chapter 4. Someone may feel they don’t belong in an organization because of correctible customs—or because they really don’t belong, because of conflicting interest or inadequate preparation. But somehow, only the first possibility is considered in the report. Apparently, anyone should belong anywhere.

The final section of this long chapter is on “Indigenous approaches, oral and community traditions, and interviews to capture lived experience.” Going beyond written records is important, apparently. In fact, “indigenous knowledge” is either a matter of faith or, potentially, a part of science: practices that are subject to test, and beliefs that are not. For example, quinine, used by the Quechua people as a muscle relaxant, turned out to be useful in curing malaria. That’s now a scientific fact. Indigenous knowledge can be a useful source of hypotheses.

But some beliefs are not susceptible to test. The existence of the Quechua god Apu, like the Christian Trinity, cannot be scientifically proven. Indigenous knowledge that can be tested is potentially part of science; other beliefs are matters of faith. “Indigenous knowledge” is not science.

The report admits the tenuous connection between the material in this chapter and actual science, with the caveat that

the committee does not provide specific conclusions or recommendations associated with this chapter beyond an acknowledgment of the value of soliciting more voices to deepen our understanding of the lived experience of Black people and other systematically marginalized groups in STEMM organizations.

In other words, this out-of-place material is there for its emotional effect. Apparently, this has become common practice at the NASEM (see footnote 1 in Chapter 4). This long chapter doesn’t belong in a scientific report. Nevertheless, the committee recommends: “These methods should be continued and expanded,” and recommends in Chapter 5 a redesign of “STEMM curriculum to incorporate Indigenous ways of knowing, and actively involve Indigenous communities in the development of this process.” This is politics, not science.

Chapter 5 of the report pretends that the previous material has proved that “minoritized individuals have faced numerous systemic barriers … that have negatively impacted their opportunities, representation, and ability to thrive in STEMM careers,” while it has done nothing of the sort. The chapter is a recitation of some of the worst social science available, which comes from a plentiful supply.

Social science fallacies, like stereotype threat and implicit bias, are trotted out once again. (Even the inventors of implicit bias say that it should not be used as a predictor; stereotype threat does seem to impair black test performance, but removing the threat still leaves blacks below the white level.) High academic standards are referred to as White, which apparently is not a racist thing to say.

[Related: “Against Merit”]

Toward the end of the chapter, the authors caution against a “deficit-based framework,” which “may lead to the assumption that the cause of the problem is the character of minoritized individuals rather than the persistence of structural barriers.” Translation: don’t assume that the failure of a black student is due to anything other than racism. Imagine this—if some students fail in STEMM, it may be their fault! This possibility is studiously ignored in the report, but rarely as frankly as it is in this passage.

The report lauds the formation of racially segregated science organizations, such as “the American Indian Science and Engineering Society” and the “National Society for Black Engineers.” Yes, these are indeed “pioneer organizations,” and very bad ones, too, contradicting as they do the foundational idea that science is universal and that scientific organizations should be about … science.

* * *

In summary, the National Academies of Sciences, Engineering, and Medicine’s report Advancing Antiracism, Diversity, Equity, and Inclusion in STEMM Organizations: Beyond Broadening Participation is about as far from actual science as can be imagined. It is a crypto-totalitarian political document promoting an enforced equality of outcome and a collectivist science organization. It is based on the following premises:

• True diversity means the population-proportional representation of “minoritized groups” in STEMM.

• Blacks, whites, Asians, and other groups are all identical in terms of their cognitive skills.

• The underrepresentation of blacks in STEMM is entirely due to systemic and other forms of racism—because there is no proof that “gatekeepers” are not racist.

• Organizations must change so as to eliminate racial disparities. They should “evaluate the institution’s values and norms and identify specific ways to address norms that impede diversity.” In other words, racial balance trumps individual merit. (Regular “climate surveys” and “cultural audits” will check that these steps are being taken. No deviation is to be permitted!)

• “White professional norms in culture, dress and appearance” should be “de-centered.”  (Why? What norms will be substituted?)

• The idea “that science is a meritocracy so it does not matter who is conducting the research …” is false.

Let’s give credit where credit’s due: the report amply shows that its idea of diversity is totally incompatible with science as it has been understood for the past three centuries. The insistence on racial balance also seems to violate the recent Supreme Court ruling against affirmative action. Most importantly, the report reveals the utter corruption of NASEM in its present form and its unfitness for the advisory role it is supposed to perform.

Editor’s Note: This piece is part of a new Minding the Campus article series called Minding the Sciences, wherein we are renewing our focus on the sciences given the many threats it faces in modern academia. Click here to learn more.

Image: Jorge Mendoza, Wikimedia Commons, Creative Commons CC0 1.0 Universal Public Domain Dedication