Are you a Bayesian Racist? Uhlman et. al. (2010) coin the term “Bayesian Racism” as “the belief that it is rational to discriminate against individuals based on stereotypes about their racial group” and they offer the following test to identify them:
[Expected benefit of an additional unit of information is less than the expected
The authors entertain the idea that stereotyping may be “epistemically rational” but then conduct a poll that purports to show that many “Bayesian Racists” are not motivated by accuracy and so are indeed irrational. The description of the poll methodology is remarkably thin. For example, no non-response rate is presented and there appears to be no validation of the Bayesian racist scale. Still, I am relieved to report that not all apparatchiks are as innumerate as Newsweek’s Shankar Vedantam.
A good Bayseian, like Walter Williams, simply recognizes statistical averages (e.g inner city blacks commit a high rate of crime) and then calculates that, under certain circumstances (e.g night), the expected benefit of an additional unit of information is less than the expected cost of acquiring that unit (e.g being the victim of a crime).
As for rationality and the Uhlman et. al. (2010) article, Phillip Tetlock and others have conducted a few studies on this matter (Tribe, 1971; Koehler, 1996; Tetlock et al, 2000). Tetlock (2003) concluded that Liberal Egalitarians are the logically incoherent ones:
In one study of forbidden base rates (i.e Bayesian estimates concerning protected groups), Tetlock et al. manipulated observers’ beliefs about the correlation between the distribution of ﬁres across neighborhoods and racial composition of neighborhoods. Liberal-egalitarian observers responded with outrage at executives who used the race-correlatedbase rates in setting premiums but not at executives who used race-neutral base rates. In a follow-up experiment, participants role-played executives who had been lured into inadvertently using a forbidden base rate. Thrust into this predicament, egalitarians were more likely to cleanse themselves morally by volunteering for good causes, especially anti-racist ones.
Specifically, with regards to hypothesis that Uhlman et. al. (2010) tested, tetlock et al. found that:
the opposite effect, using base rates to justify harsh reactions to Blacks, did not materialize at all in Experiment 3, even among the most conservative, and materialized only among a small minority of conservatives in Experiment 4. This “dog-that-did-not-bark” is contrary to the prediction of theories of racial policy reasoning that depict many, even most, Americans as covert or symbolic racists who are quick to seize on pretexts for denying opportunities to Blacks (cf. Sniderman & Piazza, 1993). Indeed, the pattern is more consistent with a view of liberals as “symbolic antiracists” (who change their views about the acceptability of inequality as soon as it implicates historically oppressed groups) than it is of conservatives as symbolic racists (who are always looking for justifications for thwarting the aspirations of oppressed groups).
I would have three comments on this: First, the liberal-left bias in the Uhlman et. al. (2010) study is palpable; unsurprisingly, the findings above, which directly contradict those of the poorly done Uhlman et. al. (2010) study, went unmentioned.
Second, the liberal-left interpretation of non-rational (i.e value or emotion driven) discrimination is even more slanted. Often the pejorative term “irrational,” which suggests illogical or nonsensical, is flung around. The concept of “irrationality” is based on an enlightenment epistemology, which classifies the non-rational as irrational, without considering the possible meta-rationality that situation. For example, in a secular universe, an appetite for life is non-rational– there’s no objective telos and the value of living can not be derived from any prior principle. The non rational valuing of life (or race or anti-discrimination), however, is evolutionarily, societally, and individually meta-rational; that is, it has a functional purpose, given the dynamics of the local system. Much of our thinking, particularly when it comes to values, is, in this sense, “irrational.” With regards to Liberal-egalitarians, I grant that they are only irrational relative to their own epistemology. From my perspective they are non-rational high priests of Multiculturaldom. Given their prior valuations, it’s meta rational for them dragoon the plebs into compliance and enforce the “irrational, unjust, and tyrannical” sacred value of anti-discrimination. In short, it’s meta-rational for them to be irrational. It just so happens that their sacred values and vision clash with mine and that I consider their values to be decadent.
Third, the moral-theological and fundamentally non-rational nature of liberal-egalitarianism accounts for why the unprincipled exception of making Bayesian estimates with regards to conservative Europeans is deemed rational.