In early June, Merriam-Webster announced its intention to update its entry on “racism.” The change was prompted by Kennedy Mitchum, a young woman frustrated that dictionaries usually don’t include a definition representative of the way “racism” is now used in certain circles, mostly academic and left-leaning, to name systems which produce disparities along racial lines – so-called “systemic racism.”
Merriam-Webster’s current entry on “racism” (as of August 7, 2020) gives three, related definitions:
D1. a belief that race is the primary determinant of human traits and capacities and that racial differences produce an inherent superiority of a particular race
D2. (a) a doctrine or political program based on the assumption of racism and designed to execute its principles, (b) a political or social system founded on racism
D3. racial prejudice or discrimination
As Mitchum explains it, she suggested the change because, in online debates over systemic racism, “[p]eople were copy-and-pasting the definition to her in an attempt to prove racism could only exist if you believe your race to be superior to another,” as per D1. What’s needed is a definition showing that racism is “prejudice combined with social and institutional power… a system of advantage based on skin colour.” But wait: isn’t this basically what D2 already says? Indeed, Merriam-Webster’s editor-at-large, Peter Sokolowski, has stated that D2 “covers the sense that Ms. Mitchum was seeking,” and that the update will consist only in “mak[ing] its wording even more clear.” So, why couldn’t Mitchum simply lob D2 back at her interlocutors and claim vindication?
I haven’t seen a clear explanation from Mitchum, but another interview provides the following hint:
Mitchum said she takes issue with Merriam-Webster’s current definition only focusing on prejudice due to the color of a person’s skin as well as racial hierarchy. “There’s so much more systems at play,” she noted. “We can see it in health care … mass incarceration.”
Anyone familiar with the academic and professional literature on “systemic racism” in these domains can guess at what’s behind this statement. When critical social justice theorists talk about “racism,” they describe it as a matter of a social system’s being organized in such a way that it creates and perpetuates racial inequalities regardless of the conscious beliefs, attitudes, or intentions of those who inhabit the system. Although they also make much of purported unconscious biases in the propagation of racism, even in systems, their criterion for diagnosing systemic racism is entirely consequentialist: “disparate impact” along racial lines is its sole necessary and sufficient condition. For example, in White Fragility, Robin DiAngelo asserts that “[b]y definition, racism is a deeply embedded historical system of institutional power (24), “a system of unequal institutional power,” (125) “a network of norms and actions that consistently create advantage for whites and disadvantage for people of color,” (27–28), “a far-reaching system that no longer depends [as per D2] on the good [or bad] intentions of individual actors; it becomes the default of the society and is reproduced automatically,” (21) i.e., without conscious intent. Journalist Radley Balko’s gloss on “systemic racism” captures the idea perfectly:
Of particular concern to some on the right is the term “systemic racism,” often wrongly interpreted as an accusation that everyone in the system is racist. In fact, systemic racism means almost the opposite. It means that we have systems and institutions that produce racially disparate outcomes, regardless of the intentions of the people who work within them.
In light of such statements, D2 would seem to fall short by failing to make a clean separation between human psychology (beliefs, intentions, etc.) and the quasi-mechanistic, or “automatic,” operations of social systems. D2 speaks of systems “based on” or “founded on” racism in the psychological sense (D1), and “designed to execute its principles.” “Design” implies intent. But critical social justice ideology requires a clean break with human psychology – beliefs, motives and the like, are at most incidental to racism, and are therefore excluded from its definition. No one has made this clearer than Ibram X. Kendi, who claims that “’Institutional racism’ and ‘structural racism’ and ‘systemic racism’ are redundant. Racism itself is institutional, structural, and systemic,” and its defining mark is the production of “racial inequity.” (How to Be an Antiracist, 18) If Kendi is correct, then the problem with dictionary definitions of “racism” isn’t merely that D2 isn’t clearly separated from D1; it’s that D1 exists in the first place. If racism simpliciter is inherently systemic or structural, and if its defining feature is the production of inequalities along racial lines, then D1 itself is wrong – the relic of a benighted age, which ought to be pulled from the dictionary like a Confederate statue from its pedestal.
But whether Mitchum’s objective was to remove D1 or merely to revise D2 in a way that detaches it from D1, her mission hits a snag. As Merriam-Webster explains in a note appended to its current entry for “racism,” “[t]he lexicographer’s role is to explain how words are (or have been) actually used, not how some may feel that they should be used.” As it turns out, D1 and D2 in their current forms capture the actual use of “racism” by ordinary English speakers very well. While terms like “systemic racism” and “structural racism” have made inroads into the working vocabularies of many English speakers, they aren’t used with consistent meanings, and they are usually assimilated to D1 or D2 in ways that are inconsistent with critical social justice theory.
In recent weeks, several acquaintances – one a criminologist – pointed me to this harrowing report on the once overtly-racist culture of the Detroit Police force as an example of “systemic racism.” But it describes a system inhabited by overt racists (as per D1) and neither a system founded on racist views or designed to perpetuate them (as per D2), nor a system that perpetuates racial inequalities regardless of people’s intentions (as per DiAngelo and Kendi). Replace the bad actors with good ones, and the system would be transformed.
The same is true of the Walter MacMillian case, which forms the central narrative of Bryan Stevenson’s book-turned-movie, Just Mercy. MacMillian was wrongly convicted of murder and sentenced to death, but Stevenson eventually succeeded in having him exonerated. In reading (or watching) the story, one is struck by just how badly the deck is stacked against MacMillian and Stevenson – it’s as if “the whole system” is against them. Presumably this is why Warner Bros. allowed free streaming of the movie for the entire month of June “to educate viewers on systemic racism.” In a statement released on Twitter, they said:
Our film Just Mercy, based on the life work of civil rights attorney Bryan Stevenson, is one resource we can humbly offer to those who are interested in learning more about the systemic racism that plagues our society.
But here again we have a case in which the intentions of specific people made all the difference to the way the system malfunctioned: the intentions of the witnesses who gave false testimony against MacMillian, of the police who coerced them to do so, and of the district attorney who withheld exculpatory evidence. It’s unclear just how far these acts were motivated by overt (D1) racism, and just how far they were motivated by other factors, like the prosecutor’s desire to win – an unfortunate byproduct of our adversarial approach to law. Regardless, the injustice exemplified in the MacMillian case is not “systemic racism” in the sense explained by DiAngelo or Kendi. Replace the bad actors with good ones, and McMillian would never have been convicted. The problem was with individual people, not “the system.”
In fact, confusion about the meaning of “systemic racism” is so widespread that Kendi seems to have given up on it, preferring to speak of “racist policies” instead of “racist systems.” For Kendi, a “racist policy” is “any measure,” including “written and unwritten laws, rules, procedures, processes, regulations, and guidelines that govern people,” “that produces or sustains racial inequity between racial groups.” (How to Be an Antiracist, 18). And why does he prefer this term? Here’s his explanation:
Racist policies have been described by other terms: “institutional racism,” “structural racism,” and “systemic racism,” for instance. But those are vaguer terms than “racist policy.” When I use them I find myself having to immediately explain what they mean. (How to Be an Antiracist, 18)
But this is wrong. The term “systemic racism” is not vague at all. Its stipulated meaning is perfectly clear. The problem is that it’s such a departure from the ordinary meaning of “racism” that people who haven’t gone through a process of re-education as part of their academic training spontaneously misunderstand the term in a way that assimilates it to D2, and by extension to D1.
In a recent piece in The Atlantic, linguist John McWhorter diagnoses the difficulty in getting ordinary English speakers to embrace the meaning that critical social-justice theorists attach to “racism.” Amusingly, he calls this “racism 3.0” to mark its advance beyond definitions like D1 and D2, which he calls “racism 1.0” and “racism 2.0,” respectively. As McWhorter explains, the historical progression from racism 1.0/D1 to racism 2.0/D2 was uncontroversial because the latter is readily perceived as a natural extension of the former. But not so with the move from racism 2.0 to 3.0. “The 3.0 usage implies that calling racial disparities ‘racism’ is natural because it is indisputable that racial disparities stem from bias-infused barriers.” But the role of racial bias (racism 1.0/D1) in the etiology of these disparities is rarely indisputable. McWhorter himself has disputed the role of racial bias in generating a disparity central to the case for systemic racism in policing: the fact “that black people are killed [by police] at a rate disproportionate to their percentage of the population.” Like Adolph Reed, McWhorter argues that an objective look at the data reveals that poverty, not race, is the key to understanding patterns of police violence. Others, like Heather MacDonald have argued that the key is not poverty but the prevalence of violent crime in the black community.
As McWhorter notes, this type of disagreement obstructs the ‘”pathway toward society-wide consensus” around racism 3.0. And it does so precisely because most English speakers are unwilling to extend the concept of “racism” to phenomena in which racism 1.0 plays no active role, particularly in the present. “It is especially challenging to that consensus,” he observes, “that some of the barriers in question, even if founded in racism 1.0 and 2.0, were in the past rather than the present,” because people often “resist a definition of racism that encompasses the actions and attitudes of people now long gone.” But this is precisely why critical social-justice theorists want to sever the connection to the attitudes and actions of people: it would be much easier to make the charge of “racism” stick if the sole criterion for making it was disparate impact in the present. It is essentially the same strategy at work in the choice to define “racial microaggressions” as including both intentional and unintentional slights, as captured in Robin DiAngelo’s mantra “think impact, not intention.” Intention is hard to prove, and genuine racists (1.0 style) can hide their malign intent behind ambiguities of language and behavior. Making intent irrelevant allows one to sidestep that problem, and to level the damning charge of “racism” on supposedly clearer and more objective behaviorist and consequentialist grounds.
The frustration behind this strategy is understandable. It would be a great boon to people of goodwill to be able to prevent people of ill-will from concealing their villainous motives. But declaring intention irrelevant to racism is not the right move. Racism 3.0 involves a huge departure from common usage, and one that results in consuming the innocent along with the guilty. No wonder it gets pushback from many speakers of ordinary English – the very fact that prompted Mitchum to write Merriam-Webster in the first place.
Given these facts, it is astonishing that McWhorter’s comments appear in a piece arguing in favor of bringing the dictionary definition closer to racism 3.0. “The Dictionary Definition of Racism Has to Change,” he says, “not because of ideological pressures emanating from “the great Awokening,” but because “sociopolitics drew the usage of the word racism beyond the dictionary definition long ago.” As a result, he claims, “legions of people, especially educated ones,” now use the term in a new way, i.e., racism 3.0. But this overestimates the extent to which racism 3.0 has infiltrated ordinary discourse. As McWhorter notes, racism 3.0 “is a tenet of much social science “ and “[i]t is a usage of racism that one often acquires in college classes in the social sciences.” The fact is that, until the recent proliferation of anti-racist reading lists and the related buying (and, I presume, reading) frenzy, this was just about the only way one could acquire this use of “racism.” But as this bit of specialized nomenclature has migrated beyond its native habitat in left-leaning academic circles in the humanities and social sciences, it has entered the vocabulary of the average English speaker without a single, clear meaning. As the above examples show, it tends to be used as a catch-all for just about anything that anyone finds objectionable pertaining to the manner in which anything that can reasonably be construed as a “system” interacts with race. In addition, it is usually assimilated to racism 1.0 and 2.0. This is true even of ostensibly “educated” people. As we saw above, in the very act of advocating for a dictionary update, Mitchum herself failed to clearly articulate the sense of “racism” omitted from the current entry, with the result that Merriam-Webster’s Sokolowski understood D2 to cover “the sense that Ms. Mitchum was seeking,” and to respond with a promise of something that sounds more like racism 2.1 than 3.0. Racism 3.0 is not about “prejudice combined with social and institutional power,” as Mitchum put it. It is about the effects of social and institutional power minus the prejudice.
Moreover, whatever meaning they attach to “systemic racism,” most English speakers still make a distinction between it and racism simpliciter, where the latter retains its traditional person- and belief-focused meaning (racism 1.0/D1). As evidence of this, consider Robin DiAngelo’s lament that white people regularly come away from her workshops still insisting that they aren’t racists, that they continue to respond with outrage and indignation at the suggestion that they are. (See chs. 1 and 7–9 of White Fragility) She chalks this up to a supposed tendency on the part of whites to be very uncomfortable talking about race issues generally, and she invented the term “white fragility” to name this tendency. I find it much more plausible to suppose that whites object not to talk about race-issues generally, but to her kind of talk about race, which – in accordance with the behaviorist and consequentialist leanings of racism 3.0 – requires them to bear the label “racist” and “white supremacist” simply in virtue of being white in a social context where, due in part to the genuine racism of the past, their status as white benefits them in unearned ways. One can acknowledge some truth in these claims but still be unwilling to bear the label “racist,” because as a matter of current socio-linguistic fact, “racism” is still understood mainly along the lines of racism 1.0/D1, and hence as a damning label.
Tellingly, DiAngelo acknowledges that “the dominant conceptualization of racism” is focused on “terrible people who consciously don’t like people of color,” and she says:
If your definition of a racist is someone who holds conscious dislike of people because of race [racism 1.0/D1], then I agree that it is offensive for me to suggest that you are racist when I don’t know you. I also agree that if this is your definition of racism, and you are against racism, then you are not racist. (13)
Good news! If you’re a competent English-speaker who uses the words “racism,” “racist,” etc. with their ordinary meanings, then, unless you consciously dislike people of color, you’re not a racist! But here’s the bad news: DiAngelo insists that “the dominant conceptualization of racism” (racism 1.0/D1) is “misinformed,” (123) and a quick glance at the section headings in chapter 1 of her book reveal that she sees those of us who accept it as possessed of “uniformed” and “simplistic” views about race and racism.
But the average English speaker isn’t having it. Again, most seem to be using “systemic racism” in a way that assimilates it to racism 1.0/D1 or 2.0/D2, and most are still distinguishing between it and racism simpliciter, where the latter is understood as racism 1.0. Both of these linguistic practices are inconsistent with racism 3.0. This undermines McWhorter’s contention that a dictionary update along the lines of racism 3.0 would “reflect how language is actually used” by ordinary speakers. While it may be true, as McWhorter claims, that an understanding of “racism” along the lines of racism 3.0 is “shared by legions of people,” it is equally true that legions upon legions do not share that understanding.
Presumably this is why McWhorter felt the need to soft-pedal the radical nature of racism 3.0 in his Atlantic piece, describing it as focusing “less on attitudes than results,” and claiming that it refers to “societal disparities between white people and others … as racism, as a kind of shorthand for the attitudinal racism [racism 1.0] creating the disparities.” Either this is disingenuous, or McWhorter himself misunderstands racism 3.0; for, as the above passages from DiAngelo and Kendi show, racism 3.0 focuses solely on results while excluding attitudes. It is not a matter of focusing more on one and less on the other, of foregrounding one and backgrounding the other. Likewise, it is not a matter of shorthand meant to capture the relationship between racism 1.0 and disparity-producing systems. To the contrary, it is meant to sever that relationship. Racism 3.0 is a radical redefinition of “racism” made for political purposes. This is the standard M.O. of those coming out of the “school of resentment” and “grievance studies” traditions. As the philosopher Richard Rorty observed two decades ago, practitioners in these fields are “resentful specialists in subversion” who treat literature and philosophy, and indeed language itself, as tools to be used for political purposes. Hurling the damning label “racist” at people and systems that don’t deserve it in order to incite revolutionary outrage is exactly the kind of subversive linguistic manipulation prescribed in their playbook.
Those of us who care about such matters need to keep a watchful eye on Merriam-Webster’s update, which is anticipated as early as August 2020. For, despite the reassuring statements from Merriam –Webster and from Sokolowski about remaining true to the actual usage of words, and limiting the update to a clarification of D2, Sokolowski has also said that “[a]ctivism doesn’t change the dictionary, … [a]ctivism changes the language,” and that “the people working on the new definition will be consulting the work of experts in black studies.” These statements may suggest that, like McWhorter, Sokolowski overestimates the extent to which full-blown “racism 3.0” has entered ordinary discourse. Likewise, his response to the internet rumor that Merriam-Webster plans to update its definition of “racism” to say or imply that only whites can be racist suggests he does not understand that exactly this view is entailed by racism 3.0, as used by “experts in black studies” like DiAngelo and Kendi. Should the update turn out to be more like racism 3.0 than 2.1, it will count as a case of activist lexicography that will only strengthen the hold of critical social justice theory on our culture.