If you are a Canadian faculty member, there is a reasonable chance that you recently received an email or letter from Statistics Canada. The Survey of Postsecondary Faculty and Researchers was designed to assess what has come to be known as “diversity” among the groups targeted, in consequence of a commitment made by the three Canadian research granting councils, under guidance from the federal Liberal government, to increase “diversity” among those receiving funding. It has long been the case that research funding was dependent, as much as possible, on two factors, both intensely meritocratic: the research record of the applicant and the quality of the proposed research. That appears about to change.

The fact of this occurrence motivated me to try my hand at writing a critique of the concept of diversity, which is a very slippery term. What it truly means is “let’s aim for fewer white men in positions of authority,” which would be a fine idea if race and sex were reasonable criteria by which to judge applicants, and if it wasn’t motivated by a broad set of “progressive” beliefs, which include the idea that we live in an oppressive patriarchy and that men who work now should be required to step back so that a litany of hypothetical, definable and prejudicial historical wrongs might be righted (this even though those who do the righting weren’t those who committed the prejudicial crimes, so to speak, and those who benefit not those who were the victims). There was even a recent article in Nature, a magazine that was once, with Science, one of the two unquestionably most influential scientific journals, suggesting male scientist should voluntarily delay their career advancement so that their underprivileged colleagues (underprivileged despite their status as university professors) could catch up and justice properly served.

“Diversity” is a word that, on the face of it, masquerades as something positive—because it is positive, in some of its manifestations. It’s obviously not helpful to set up an organization where everyone thinks alike, or solely in the approved manner. It is necessary, for example, for healthy organizations to ally the conservative tendency to preserve with the more liberal tendency to transform. But that begs the question: where is diversity to be found? Among the ideologues who were pushing the “progressive” doctrine that it’s part of, most frequently including “inclusivity, equity and intersectionality,” it is to be found in a set of immutable characteristics that typify different groups, including race, sex, gender (because that is distinguished by those same ideologues from sex) and sexual proclivity, above all.

There are real problems with this agenda, however. The first is that it’s dangerous, in exactly the manner it is hypothetically designed to fight. The argument made by those who are truly prejudiced has always been that the differences between groups are so large that discrimination, isolation, segregation and even open conflict, including war and genocide are necessary, for the safety of whatever group they are part of and are hypothetically protecting. Why is it any less risky for the argument to be made in the reverse manner? The claim that group-based differences are so important that they must take substantive priority during hiring and promotion merely risks validating the opposite claim.

There’s a second problem, too—and it’s particularly interesting, because it has been made by the same ideologically-oriented groups on the left that are pushing the diversity agenda: considering race, say and gender when making diversity decisions is not sufficient. Diversity that focuses on females is insufficient, because black, Asian or Hispanic women, for example, face more egregious prejudice that white women. This brings us to the last word of the progressive set—“intersectionality.” For the ideologues of intersectionality, true diversity cannot be limited to the features we have already considered—race and the like—because many people are alienated or, in the jargon, “marginalized,” from the broader culture by more than one oppressed minority feature. In consequence, the “intersection” between the groups must be considered for any real justice to make its appearance as a consequence of policy.

This is an extremely problematic theory, practically speaking (and this is the second problem, in addition to the danger just outlined), in that there appear to be no limits, practically or philosophically, to the number of group memberships that have to be taken into account for true diversity to establish itself (and I mean this non-ironically). It doesn’t take much thought—just a little arithmetic—to determine the nature of the problem: There are just too many potential intersectional categories. Let’s break it down. I’m going to use American statistics, which are much easier to come by than their Canadian equivalents, but it doesn’t matter, because it’s the principle that is in question here, and not the precise statistics themselves.

There’s race and sex, for starters—and plenty of attention is paid to both—and, following that, gender, which seems to come in something approximating second place in terms of import. But how many races, sexes and genders is it required to consider? Assume, for the sake of argument (and this is what the modern science suggests) that there are five major human subpopulations: African, European/Middle Eastern, East Asian, inhabitants of Oceania, and denizens of the New World. Let’s assume two sexes and three genders—although many of those concerned with diversity, etc., would insist that there are a much larger number of the latter.  So that’s 5X3X2 =30. Then we might as well add to that disabilities, which are extraordinarily common (particularly when you consider that many people who are not actively ill in some major manner, physical or mental, are faced with the exceptional stress that comes with caring for a family member who is). I don’t know how to calculate the appropriate number here, although but according to the National Center for Education Statistics, 20% of undergraduates[1] reported a disability in 2015-2016. These include “those who reported that they had one or more of the following conditions: blindness or visual impairment that cannot be corrected by wearing glasses; hearing impairment (e.g., deaf or hard of hearing); orthopedic or mobility impairment; speech or language impairment; learning, mental, emotional, or psychiatric condition (e.g., serious learning disability, depression, ADD, or ADHD); or other health impairment or problem.”[2] So that’s an additional eight categories, which brings us to 240. I can’t see why class/economic origin shouldn’t be taken into account as well. According to the U.S. Census Bureau estimates, 12% of Americans live below the poverty line. So, we need at least an additional two categories to account for economic disparity, and that’s a very coarse grained measure. And that brings us to almost 500 (2X240). If we are serious about diversity, and are willing to attribute it to group identity, and are going to apply its dictates to hiring, placement and promotion for every position, then we have a minimum of five hundred different categories to consider—and there are many other categories of exclusion that are arguably of equal import (such that it is difficult to determine why it was that race, sex and gender occupy so much attention). There’s height, strength and attractiveness, which all arguably provide an unequal starting place in the race for success. There’s intelligence, native language and education. There’s age, marital status and—of critical importance—presence or absence of dependent children. I can’t for the life of me see why these are less important if diversity as it is currently defined is going to be seriously pursued. So that brings the number of relevant categories up to far more than 500. If we assume just two categories for each of these additional group differences—and there’s no reason to limit the differentiation to two—then we are faced with 500 time 2 nine times or 256000 categories. And why not? Who’s to say, given that elimination of discrimination is hypothetically the goal, that one more important than another? I say this in all seriousness: Isn’t that just another form of discrimination?

And there are great problems even with the categories that seem, on the surface, relatively simple—let’s say, race. The geneticists, as I mention, consider five (and even those don’t overlap perfectly with the categories that are used politically: Hispanic, for example—a favorite unique to the US—is  more of a linguistic category, although those described that are often of mixed European and New World origin). But why is it reasonable to stop at five? Let’s take the case of “black,” for example. First, we could note that blacks who immigrate to the US do better, in general, than blacks who are born American. This is true for education[3] (black immigrants are more likely than Americans, in general, to have a college degree) as well as income ($43,800, somewhat lower than immigrants to the US in general ($48,000, a figure inflated by the outsized economic attainments of Asians: $70,600) but substantially higher than U.S. born blacks ($33,500). (Americans overall average $52,000). Do we therefore differentiate blacks on the basis of their place of birth, and add another category to the diversity pool? And we could also make a very strong scientific case for even further differentiation. Are we truly to be satisfied with the claim that all blacks are the same, even independently of birthplace?  For example, there is much more genetic diversity among Africans—the putative home of humanity—than among all other non-African populations.[4] [5] This is in large part because the apparently small number of migrations out of Africa to the rest of the world produced a variability bottleneck: a relatively small number of people moved, and so a relatively small amount of genetic diversity existed. By what logic, therefore, is it reasonably to cluster all these people together, call them “black,” and assume that organizational diversity justice has been served, say, by their increased rates of hiring, promotion and placement?

And let’s return to the beginning. As far as I am concerned, unless you accept it as a dogmatic given (and this would be if you were an advocate of the “equity” doctrine, which means that all outcomes for all groups in all professions must be identical, and which therefor runs into the same arithmetical problem that diversity encounters, as described previously) university hiring and granting practices are remarkably meritocratic. In the university departments I have worked within (McGill, Harvard and the University of Toronto) it was obvious to everyone that within the limits of human error, which are of course manifold, people were promoted when they deserved it and obtained research grant money for the same reasons. In both cases, the more productive people had a pronounced edge, which is exactly how it should be if scientific research is important enough to garner investment, be it from private or public funding sources. The three granting agencies are as meritocratic as our somewhat (and inevitably) flawed measures of research productivity can make them, and the universities themselves bend over backwards and tie themselves in knots (both clichés are necessary) to right past wrongs—to the point where, according to the well-respected social scientists Wendy Williams and Stephen Ceci in their demonstrated a 2:1 hiring advantage for female candidates for open science, technology, engineering and mathematical positions.

In consequence, I would like to suggest that the proper way to determine who gets what slice of which pie in a given organization is the manner in which employers are legally bound to hire: first, they must conduct an analysis of the job to determine and list its requirements; then, with certain exceptions they are required to hire, place or promote the person who is most qualified to undertake that job, regardless of attributes that are not relevant to the task. These include, in my opinion, the differences in race, sex, gender, and their combinations (as well as the other intrinsic differences we discussed) that are pushed so assiduously, self-righteously and thoughtlessly by the progressives who think they can replace comparatively well-functioning meritocracies, aimed at the solution of serious problems by the most qualified people with candidates chosen on the basis of attributes that would clearly be viewed as prejudicial if they were used as grounds for rejection, failure to promote, and firing.

A final observation. The fact of the endless multiplication of categories of victimization, let’s say—or at least difference—was actually solved long ago by the Western emphasis on the individual. We essentially assumed that each person was characterized by so many differences than every other person (the ultimate in “intersectionality”) that it was better to concentrate solely on meritocratic selection, where the only difference that was to be considered was the suitability of the person for the specific and well-designed tasks that constituted a given job. That works—not perfectly, but less imperfectly than anything else that has been contemplated or worse, implemented. We toy with it at our peril.

via Jordan B. Peterson


[1] U.S. Department of Education, National Center for Education Statistics. (2019). Digest of Education Statistics, 2017 (2018-070), Chapter 3 (table available online at https://nces.ed.gov/fastfacts/display.asp?id=60)
[2] Ibid.
[3]Tauriac, J.J & Liem, J.H. (2012). Exploring the divergent academic outcomes of U.S.-origin and immigrant-origin Black undergraduates. Journal of Diversity in Higher Education, 5, 244-258.
[4] Michael C. Campbell, M.C. & Tishkoff, S.A. (2008).  African genetic diversity: implications for human demographic history, modern human origins and complex disease mapping.  Annual Review of Genomics and Human Genetics, 9 403-433.
[5] Tishkoff, S.A., Reed, F.A. Frieldaender, F.R. et al. (2009). Science, 324, 1035–1044.

Leave a Reply

  • (not be published)