Disparities do not Equal Discrimination
The idea underlying it all.
“We have a hard time recognizing that racial discrimination is the sole cause of racial disparities in this country and in the world at large.” - Ibram X. Kendi1
“When I see racial disparities, I see racism.” - Ibram X. Kendi2
Disparities do not equal discrimination.
To some, this might sound like anathema, to others, like a totally innocuous and incontrovertible statement, one barely worth even paying attention to.
Sadly, however, this statement is of vital importance today. It lies at the heart of virtually every political discussion and disagreement when it comes to analyzing diverse demographic groups. Either knowingly or unknowingly, it is implicit in nearly every political discussion in a diverse nation.
Why are certain protected demographics3 underrepresented or overrepresented in different walks of life?
From prisons, to education, to median income, disparities are ubiquitous, indeed you would be hard-pressed to find anything that does not have a disparity (as Thomas Sowell4 and Gail Heriot5 have noted).
What’s the reason for this, the cause of this pernicious, and persistent, ailment of ours?
If you consult the intelligentsia, those consisting of academics, journalists, and other prominent intellectuals, they've already arrived at a consensus. Seemingly, the answer is quite clear: discrimination.
Discrimination in all its ugly shapes and forms. It is the univariate.
The existence of a disparity is taken as ipso facto evidence of discrimination.
The sentiment that the mere existence of a disparity sufficiently demonstrates discrimination is invoked as an article of faith. An oath of loyalty. A statement so axiomatic, so self-assured, that dissent, more than just being beyond belief, in the minds of its adherents, is but a figment.
Let’s provide some clarifications on terminology. As every student of statistics knows, correlation does not imply causation. Sometimes this is written as ‘correlation does not equal causation’.6 What does that mean exactly?
In precise terms, this means that observing a correlation is not a sufficient condition for establishing causation. ‘Correlation does not equal causation' rejects the notion of ‘if p is true, then q follows’ (if p then q) which underlies beliefs in systemic racism and the patriarchy (See Kendi’s quotes above). Alternative explanations are suppressed or deemed inappropriate in polite society, creating a tentative taboo, a fatal faux pas.7
This mantra that disparities equal discrimination is repeated breathlessly. When people of color have an unfavorable disparity (relative to whites) it is self-evidently attributed to racism8.
When women have an unfavorable disparity (relative to men) it is self-evidently attributed to sexism9.
And this way of thinking marches onward relentlessly, ad infinitum. We see new oppressed classes burst onto the surface of American social life by the decade, including discrimination against transgender individuals (transphobia) being the sole acceptable reason of suicide disparities and so on.
Whenever it is deemed that there is an oppressed group, intersectionality, with its power-based dichotomy, necessitates an oppressor.
Considering the ostensible academic consensus, is the case closed? Has the matter been settled? Not quite. Far from it, the classical rules of logical inference, the understanding of multivariate analysis in complex social science, and yes, simple common sense, still apply.
One way to expose the illogical nature of this is to invert the typical dichotomy of intersectionality. Are there disparities that favor ‘oppressed’ groups? Such as disparities in favor of women, or certain POCs?
The answer is a resounding yes.
For instance, men are only 50% of the population, but 93% of the prison population10. Is this disparity due to sexism against men?
Senior citizens are 17% of the US population11, but only 2.8% of the federal prison population12. Is this disparity due to discrimination in favor of senior citizens? No, it’s probably just because grandpa can’t rob a bank like you or I can. The implied inverse of course, is also apparent. young adults between the ages of 18-35 are 20% of the population13 but are 34% of the population in prisons14. Is this unfavorable disparity evidence of rampant ageism against young adults?
Asian-Americans are 6% of the US population, but nearly 20% of doctors15 and 25% of students at elite universities16. Yet, we are constantly berated with the message that we live in a society stricken with white privilege and white supremacy. This is hard to reconcile with the persistent finding that Asian-Americans outperform whites. On income17, life expectancy18, SAT scores19, MCAT scores20, infant mortality rates21, teenage pregnancy rates22, homicide rates23 and credit scores24, Asians have a favorable disparity relative to whites, time and time again. Indeed, many Asian nationalities in the United States, such as Indian and Korean-Americans, have a median income nearly twice that of whites25. Are these favorable disparities due to some mysterious ‘Asian privilege’ or 'Asian supremacist’ society? Doubtful, to say the least…
These indisputable findings contradict and challenge intersectional concepts of ‘privilege’ and ‘oppression’. They call for multivariate explanations that take into account a complex array of factors, not simplistic monocausal Manichaean reasoning26.
These findings upend the conventional understanding of ‘privileged’ and ‘oppressed’ identities in intersectionality27. It casts tremendous doubt on the axiomatic belief that discrimination is solely responsible for certain demographics' unfavorable disparities. And perhaps most crucially, it paves a way forward for us.
Disparities exist. They are, by all available and plausible evidence, incredibly resilient and persistent. They can be reduced, that is certain. They can never be erased, this is inescapable28. Eventually, we must get to a point in which we can fashion a typology of disparities:
Legitimate disparities: a legitimate disparity is a disparity that does not arise out of discrimination
Illegitimate disparities: an illegitimate disparity is a disparity that arises out of discrimination
I want to depart with a pithy rebuttal when one is confronted with the archetypal leftist take on matters of diversity and representation in American society. When you hear one of the fashionable mantras, the questionable boilerplate bromides, try this: disparities do not equal discrimination.29
Kendi, I. X. (2017). Stamped from the Beginning: The Definitive History of Racist Ideas in America. (p. 19). Nation Books.
Modern understandings of protected classes come from Title VII of the 1964 Civil Rights Act and federal case law, they include: race, religion, sex, color and nation of origin. Originally designed for employment discrimination, these protections now span the scope of everything the federal government touches, including education, housing, and health care.
Sowell, T. (2018). Discrimination and Disparities. Basic Books.
Heriot, G. L. (2019). TITLE VII DISPARATE IMPACT LIABILITY MAKES ALMOST EVERYTHING PRESUMPTIVELY ILLEGAL. NYUJLL. “everything or nearly everything has a disparate impact on some race, color, religion, sex, or national origin group.”
Hirsi Ali, A. (2021). Prey: Immigration, Islam, and the Erosion of Women's Rights. HarperCollins. This is downstream of the ‘iron rule of politics’, “…if there are real problems in society and responsible parties don’t deal with it, the irresponsible parties will jump on them.”
https://www.usatoday.com/story/news/politics/2021/04/29/sen-tim-scott-says-america-isnt-racist-data-suggests-otherwise/4887356001/ This article is very telling, it attempts to rebut the notion “America isn’t racist” just by showing disparities! It is literally disparities = proof of discrimination. See Kendi quotes above as well.
https://www.bop.gov/about/statistics/statistics_inmate_gender.jsp Similar disparities are found in state prisons, men are overrepresented.
https://www.census.gov/quickfacts/US senior citizens are adults aged 65 and above.
https://www.bop.gov/about/statistics/statistics_inmate_age.jsp Similar disparities are found in state prisons, senior citizens (65+) are underrepresented, young adults are overrepresented
https://bjs.ojp.gov/content/pub/pdf/revcoa18.pdf Asians are 6% of the population but only 1% of those who commit homicide :).
Virgil’s Georgics. "Happy the man, who, studying nature's laws, Thro' known effects can trace the secret cause"
I will sometimes use ‘applied intersectionality’. Intersectionality, as illustrated above, denotes that certain identities are privileged or oppressed, and the intersection of these identities can create unique additional oppression (like how the intersection of Asian + female get the stereotype of being bad drivers), applying this theory tells you any identity that has a favorable disparity got it through ‘privilege’, and the necessary inverse that any identity that has an unfavorable/underperforming disparity got it through ‘oppression’.
Clark, G. (2007). A Farewell to Alms: A Brief Economic History of the World. Princeton University Press.
Clark, G. (2014). The Son Also Rises: Surnames and the History of Social Mobility. Princeton University Press.
https://www.nber.org/digest/aug20/riches-rags-and-back-again-impact-chinas-revolutions “One of the most extreme attempts in history to eliminate advantages of the elite and eradicate economic and educational inequality succeeded only in the short term.”
Some might prefer ‘disparities do not imply discrimination’ to keep it more in line with ‘correlation does not imply causation’ but as I show in footnote 6, ‘does not equal’ is often used in this aphorism as well. Regardless, both just mean “A isn’t sufficient evidence of B, there are other potential variables and causes”, and so it comes down to a matter of preference. I also use DED as an acronym for this!
Draft: *Disparities exist. They are, by all available and plausible evidence, remarkably resilient and painfully persistent.*
Good article on South Africa and how widespread corruption/nepotism was used to eliminate discrimination with incredible results:
https://thepsmiths.substack.com/p/review-south-africas-brave-new-world
I use exactly the same examples. If racism is the only explanation for differences in group outcomes by race, then how do you explain the differences between Asians and whites?
I'm sure racism accounts for some differences in outcomes between racial groups, but the assumption that racism must account for100% of the difference is rather strange.