On the Definition of Racism

Bobvis (here and here) and his co-bloggers are trying to hammer out a definition of racism.

Bobvis:

"I believe that the term racist should apply only to beliefs that work to create a society wherein the members of certain races are disadvantaged relative to the members of other races.

"My definition applied to comments that "work to create a society wherein the members of certain races are disadvantaged". Comments that approve of such a society work to create such a society.

"Like it or not though, the race of a person can be useful for creating prior expectations about certain things, and I occasionally use them. Race and sex affect my expectations, and therefore my behavior towards people. I don't think this makes me racist."

Trumwill:

"If one uses factual and accurate statistics that suggest, for instance, that blacks in general have lower IQs, are more prone to crime, or whatever... the veracity of the data doesn't absolve the racist intent and, if carried out into action, racist actions.

"I would say that professing (approvingly) that the world would be a better place if minorities were disadvantaged is a racist comment."

Unfortunately, Bobvis and Trumwill find themselves encumbered in this project by their own intelligence and fair-mindedness. Encumbered, because they are smart enough to know that the races of man differ phenotypically in operationally significant ways, and they also want to avoid calling behavior "racist" unless they think it deserves the moral opprobrium normally attached to that word. In contrast, most other writers find it convenient to dispense with these concerns when worrying themselves about racism.

Φ's Definition

For my part, I will avoid tripping over my residual intelligence and fair-mindedness by making a rhetorically sweeping generalization:

"Racism" is morally neutral: it neither asperses a lawful act nor justifies an unlawful one.

People should be free to go about their associations with their fellow citizens with whatever motivations they happen to possess. We choose to buy or not buy, sell or not sell, employ and discharge or refrain from doing either, based on whatever preferences. The fact that our different races affect how we interact with each other may be wise or foolish, efficient or inefficient, but it is not, in and of itself, either good or bad morally speaking.

So Jesse Jackson, for instance, is free to be relieved that the footsteps behind him on a dark night in the city belong to a white person instead of a black person. The home-buyer is free to prefer a neighborhood of his own race. The McDonald's manager is free to staff his registers with people of the race preferred by his customers. Our immigration laws may (and should) take race and nation-of-origin into account to make sure that such immigrants as we choose to admit will benefit the existing citizenry as a whole.

In contrast, no one may use race to justify depriving someone else of his rights under the law. If I murder someone, I am not justified in my murder by this or that claim about race. I am not excused from boorish social behavior on the grounds that I am exercised about race.

In another example, the police may be (and, in fact, are) perfectly justified in suspecting a white person driving slowly through a black neighborhood of seeking to buy drugs, for instance, or a group of blacks driving through a white neighborhood of looking to rob someone. The police should be free to subject such people to a higher level of scrutiny; they may not, however, pre-emptively arrest them without a warrant or a showing of probable cause.

Of course, the reaction to this can be anticipated: Φ doesn't care that the burden of abolishing race-based Civil Rights laws will fall primarily on blacks.

Φ's Justification

It all comes down to prior probabilities. As Brandon Burg points out in the comments, our background knowledge of the prevalence of a certain behavior in the aggregate mathematically factors in determining the proper threshold for determinig its occurence in an individual case. Brandon Burg applies it to assessing certain traits in individuals of a particular race; however, the concept also applies in making determinations of racism.

Our Civil Rights laws were written for the world of 1958, when invidious racial discrimination was widespread, and more often than not enforced by legal or extra-legal violence. And in the minds of many supporters of the status-quo, it is always 1958, with evil racists around every corner.

But that is not the world of 2008. Such racial discrimination as remains is largely prudential, and based on reasonable calculations of the probability of dysfunctional behavior on the part of particular racial minorities. Indeed, if we have a widespread social problem with regard to race, it is our bending over backwards to ignore evidence of this dysfunction in the creation of our social policies, with disastrous results.

So while my sweeping claim may not adequately address every hard case imaginable, it is operationally justified under the conditions in which we actually find ourselves, and our pressing need to expand the domain of freedom and prudence.