Is racism an individual or systemic problem? Traditionally, racism was broadly recognized as an interpersonal phenomenon: reflexive antipathy towards an identifiable “other,” supported by the negative cultural tropes and stereotypes used to inflame resentment and justify discrimination. This was the definition used by history’s most prominent anti-racist figures, from Frederick Douglass through W.E.B. Du Bois to James Baldwin, in their scathing critiques of slavery and Jim Crow. In this telling, racism is a disposition informing the beliefs and behavior of the people who operate society’s structures and institutions. It is a bottom-up process and ultimately exacts untold harms on both oppressor and oppressed.

Yet this definition has undergone a phase transition in modern progressive discourse. Rather than an emergent psycho-social phenomenon, racism today is usually understood by theorists, analysts, and activists in structural and institutional terms that don’t require the prevalence of individual racist attitudes to explain recurrent disparities between racial groups. Bestselling author Ibram X Kendi is one of a number of contemporary writers who exemplifies this outlook: “‘Racist policy’ says exactly what the problem is and where the problem is. ‘Institutional racism’ and ‘structural racism’ and ‘systemic racism’ are redundant. Racism itself is institutional, structural, and systemic.” This account is characteristically imprecise, but it suggests that racism is a top-down structural force transmitted from social and political institutions to people, who are somehow beyond the scope of individual agency or intent.

Although individual racist attitudes have been in precipitous decline for half a century, the “racist” epithet is as common as it ever was. Indeed, white liberals now report seeing more racism against blacks than blacks themselves do. This is partly because the definition of “racism” has expanded to include banalities like asking where someone’s family is from, while dubious schools of thought such as Robin DiAngelo’s “white fragility” theory have been popularized in media and academic circles. These developments have seen the concomitant ascendance of implicit association testing and anti-bias training. This discrepancy between the reality of racial progress and the ever-expanding definition of racism in modern life is responsible for much of the confusion in a mainstream debate that routinely conflates racism with racial inequality.

For example, when the term “white supremacy” is employed in public discourse, it implicates persons, institutions, and policies in a vast web of oppression held to be fundamental to the life of the nation itself. Nikole Hannah-Jones, head of the New York Times’s 1619 Project tells us that “anti-black racism runs in the very DNA of this country,” a metaphor that commentators have noted implies permanence. In his 2017 book We Were Eight Years In Power, Ta-Nehisi Coates argued that white supremacy is “a force so fundamental to America that it is difficult to imagine the country without it.” In this rendering, “white supremacy” describes (and purports to explain) the mass privileging of whites at the collective expense of nonwhites in perpetuity.

If we are to make further meaningful progress against racism, there needs to be broad agreement about what it is. There is nothing approaching that at present, and it is difficult to achieve because challenging the prevailing view tends to elicit a startling degree of moralizing invective. But if racism can’t be clearly identified in particular policies or individuals, there isn’t an obvious third place for it to be found.

Crisis in Levittown

The term “institutional racism” was first invoked by Black Panther leader Stokely Carmichael (aka Kwame Ture) in his 1967 book Black Power: The Politics Of Liberation:

Racism is both overt and covert. It takes two, closely related forms: individual whites acting against individual blacks, and acts by the total white community against the black community. We call these individual racism and institutional racism… The second type is less overt, far more subtle, less identifiable in terms of specific individuals committing the acts. But it is no less destructive of human life. The second type originates in the operation of established and respected forces in the society.

Carmichael provides an example of each kind: “When a black family moves into a home in a white neighborhood and is stoned, burned, or routed out, they are victims of an overt act of individual racism which many people will condemn—at least in words. But it is institutional racism that keeps black people locked in dilapidated slum tenements, subject to the daily prey of exploitative slumlords, merchants, loan sharks, and discriminatory real estate agents.” Over time, “institutional racism” came to be understood using a “systems theory” framework, in which a system is a conglomeration of interrelated and inter-dependent factors more powerful than the sum of its parts. Here, individual racism isn’t necessary for the recreation of racial inequality. Instead, a systemically racist structure churns out such pervasive inequality that it cannot be explained by the sum total of racist individuals and actions within it.

Carmichael’s example of housing segregation is worth considering, particularly the historic practice of “redlining” whereby black neighborhoods were systematically denied good credit. When black Americans migrated north after the failure of southern Reconstruction and the rise of Jim Crow, whites feared the property values of their neighborhoods would depreciate upon their arrival. So, they took a number of steps to prevent it—from propagandizing to firebombing to literally paying black people to walk up and down their street to stoke fears of an impending influx. The upshot was that a majority of blacks were made to live in economically deprived and socially blighted communities.

In his groundbreaking essay for the Atlantic entitled “The Case For Reparations,” Ta-Nehisi Coates unpacks the systemic element of this practice. Along with individual acts of racism such as “blockbusting” and selective price raising, the Federal Housing Administration (FHA) adopted a system of maps by which neighborhoods were rated according to their potential value. This allowed them to determine whether mortgages would be insured and loans distributed. This is how the word redlining was coined:

On the maps, green areas, rated “A,” indicated “in demand” neighborhoods that, as one appraiser put it, “lacked a single foreigner or negro.” These neighborhoods were considered excellent prospects for insurance. Neighborhoods where black people lived were rated “D” and were usually considered ineligible for FHA backing. They were colored in red. Neither the percentage of black people living there nor their social class mattered. Black people were viewed as a contagion.

Levittown, Pennsylvania, was a planned community in the metropolitan area built during the 1950s. FHA financing was only made available to those of “the Caucasian race,” as stipulated in sales agreements and deed covenants. William J. Levitt, the real estate developer behind the various Levittown projects, didn’t consider himself racist despite these appalling policies, believing that housing and race relations were entirely separate matters. This seems to provide a clear illustration of an autonomous racist system, able to generate racially disparate results without necessitating individual racist attitudes to reproduce itself.

But this analysis misplaces responsibility for the more direct forms of racism that allowed such historical injuries to occur in the first place. It was true that when blacks moved into a neighborhood, property values often declined, but this had everything to do with northern whites’ racist perceptions of blacks, which, at the time, were so commonplace as to be banal. Without these perceptions, it is unlikely that the laws and institutional practices Coates describes would have existed at all. Conversely, it is likely that racial segregation would result without the role of federal government if predominant anti-black attitudes and norms remained.

There was certainly an institutional incentive to keep neighborhoods white, but redlining was an outgrowth of interpersonal racist dynamics, emanating from culturally enshrined perceptions, and executed by individual agents—loan officers, realtors, sellers, federal housing administrators, and citizens—without whom, housing segregation couldn’t have occurred. Rare footage from a short documentary on Levittown made in the 1950s shows the degree to which racist beliefs motivated white homeowners in the northern states. It was filmed in the wake of the infamous integration experience of Bill and Daisy Myers which culminated in violent riots. Although not every resident interviewed was virulently anti-black, many were, and conspiratorial rumors about the Myers couple rapidly proliferated through the community:

What unfolded in Levittown and across the country was a product of racist attitudes held by individuals, not a subtle consequence of an implicit bias or structural imbalance. The racial divisions engendered by redlining practices were not more than the sum of their parts, but what we would expect from people with the views they held.

Systemic racism vs. racial inequality

Distinguishing the structural from the interpersonal roots of racial inequality is crucial to understanding the obstacles to self-realization experienced by historically injured groups. And that understanding can help mitigate the polarization that makes it difficult to address those inequalities that persist. Racism and racial inequality are separate problems, and the assumption that the former always explains the latter disrupts our ability to develop solutions to the complex social problems that disproportionately impact minority communities.

The problem is that many of the examples of systemic racism commonly offered by progressive analysts are poorly conceived and defined, and possess little predictive power. For instance, an infamous study in which black “looking” names received fewer call-backs from job recruiters provides prima facie evidence of discrimination on the part of the recruiters but no evidence that this is systemic. Furthermore, replications of the study have found similar results for Asian names. Given the astounding success of East Asian and Indian Americans, the original study’s findings can explain very little about the causes of the nationwide achievement gaps between blacks and whites.

“Systemic racism” is a term often used to explain the social and economic legacy of historical injustice and white supremacy, its cultural heritage, and the ongoing consequences of each. Indeed, Coates’s entire argument for reparations is based on the compounding impacts of redlining practices, even though the 1968 Fair Housing Act made them illegal. This concern is no doubt sincere and substantiated with reference to black/white differentials in intergenerational transfers of wealth, disproportionate rates of single parent homes in the black community, or demographic differences in sentencing irrespective of past criminality (which likely has to do with the kinds of neighborhoods many black boys are brought up in through no fault of their own). There is an undeniable weight to our history, and although its impact may sometimes be overstated, it is a moral and an analytical mistake to dismiss its lingering effects altogether.

But does the term “racism” usefully illuminate these phenomena? I don’t think so. When most people hear that word, they aren’t thinking about the complex system of interactions between groups in society that advantages some and disadvantages others, nor the present inequalities produced by historical racism, but rather interpersonal and individual forms of racial resentment and hatred, traditionally understood. Expanding the concept to incorporate any inequality in outcome or undesirable policy is a category error with a potentially heavy cost. Labeling policy preferences we dislike racist can increase support for those policies among certain demographics and make skeptics more receptive to populist arguments. And by waging a cultural war against an intangible and all-pervasive structural force, we fail to identify and shame specific acts of racism, and shroud legitimate claims in uncertainty, confusion, and doubt.

“Systemic racism” is a profoundly unhelpful idea that refuses to acknowledge the progress made in stigmatizing prejudice and promoting norms of mutual respect and tolerance. Contra Ibram X Kendi, racism in America today is not systemic, beyond the fact that some people are racist and people exist in social systems. The term “racial inequality” more accurately describes ongoing disparities in outcome, but it does not ascribe blame a priori. We are therefore more likely to be encouraged to investigate the causes of these disparities rather than to be deterred by accusatory terms like “systemic racism” and “white supremacy” that purport to explain everything but in fact explain nothing. Conflating the universal human propensity for racial bias with the bloody history of discarded systems and structures such as slavery and Jim Crow is a way of taking neither seriously. We can address continuing racial inequality in our society while acknowledging that racism no longer defines our cultural and political landscape.