Prominent scholar and activist Dr. Ibram X. Kendi has claimed, “To oppose reparations is to be racist. To support reparations is to be anti-racist.” But is Kendi right? One way to tell is to apply signal detection theory (SDT). SDT is a decision-making framework that attempts to distinguish between truthful signals and noise.
Let’s imagine two distributions: racists and anti-racists. In each, any randomly chosen individual generates, with some probability, unique responses to questions about whether she is racist. Each question is designed to elicit a signal about the person’s racist or anti-racist status. The answer is then compared to a “true” criterion that distinguishes between racist and anti-racist. One could imagine a spectrum of responses to such questions, responses that provide information allowing us to designate the person racist or anti-racist. We could then compare these with Kendi’s binary criterion of whether the person supports reparations.
Presumably, whether a randomly selected person supports or opposes reparations can depend on the individual circumstances of the case presented, such as whether the reparations policy is based on corrective or distributive justice. Anti-racists might all support reparations to some degree. They might become pro-reparations activists, or merely vote for a candidate who supports reparations. Racists might all oppose reparations, but to extents ranging from arguing against them on the news to simply voting for a candidate who opposes them.
The existence of this spectrum of responses suggests that support for reparations is not a foolproof criterion demarcating a clear division between racists and anti-racists. What if our racist and anti-racist groups overlap? For example, is a person who votes for a pro-reparation candidate, but declines to engage in direct activism racist—or insufficiently anti-racist? What about a person who chooses not to vote for a pro-reparation candidate for reasons unrelated to the politician’s stance on reparations, but joins a political organization that advocates for reparations?
If the distributions of racists and anti-racists overlap, Kendi’s binary might become a quadrant comprising of hits, misses, false alarms and correct rejects in the following way.
(1) A person both joins a pro-reparations organization and votes for a pro-reparations candidate. This is a hit. The person is correctly identified as anti-racist.
(2) A person does not join a pro-reparations organization but votes for a pro-reparations candidate for reasons unrelated to the politician’s position on reparations. This is a miss. A racist is falsely designated as an anti-racist.
(3) A person joins a pro-reparations organization but does not vote for a pro-reparations candidate for reasons unrelated to the politician’s position on reparations. This is a false alarm. An anti-racist is falsely designated as a racist.
(4) A person does not join a a pro-reparations organization or vote for a pro-reparations candidate. This is a correct reject. A racist is correctly designated as a racist, i.e. correctly rejected from the anti-racist category.
|Supports politician who advocates reparations||Joins organization that advocates for reparations||Designation||Truth|
So what do we mean when we say we support or oppose reparations and is supporting or opposing reparations a function of racism or anti-racism? This partly depends on how we define racism, a term whose meaning has expanded a great deal. But it also probably depends on a full assessment of the reasons why people support or oppose reparations. Kendi’s binary framework may obscure a more complex interplay between facts, principles and motives.
A number of theoretical and practical questions can influence public support or opposition to reparations. Should reparations be based on corrective or distributive justice? Does support stem from a sense of victimisation, disingenuous victimhood signaling, self-interest or impartial moral conviction? Does opposition reflect principled doubts about the validity of the claims of the descendants of the victims, especially since a lot of time has elapsed? Does it reflect system justifications or endowment effects of white privilege? Does it depend on the race, class or gender of the victims or perpetrators?
Are any of these postures rooted in political or ethical beliefs about some of the trade-offs involved, such as higher taxes? Are the supporters or opponents unaware of or reticent about their real motives?
Opinions on reparations may also be influenced by beliefs about the extent to which US society is or should be a meritocracy that provides equality of opportunity. Opinions may also vary over the course of time. They may be affected by the race or gender of the victims. There may also be more (or less) support for a reparations policy based on a tort model derived from the principle of corrective justice, as opposed to one based on an atonement model derived from the principle of distributive justice.
Support for any given reparations program, then, may depend on many factors that Kendi’s binary framework obscures.