Plasma Pool

In an early episode of CBS’s The New Adventures of Old Christine, the title character, a generally ignorant, politically liberal single mother played by Julia Louis-Dreyfus, complains to her ex-husband that the “meanie moms” at their son Richie’s exclusive Los Angeles private school are “scary, snobby, and judgmental.”"And you’re not judgmental?” he replies. Christine’s brilliant retort: “It’s not called ‘judgmental’ when you’re right. It’s called ‘right.’”

Judge Sonia Sotomayor and President Barack Obama | *image: CBC

I’m reminded of this exchange in reviewing political scientist Joshua Tucker’s note at The Monkey Cage on a New York Times op-ed by Tom Goldstein in which Goldstein “reports on his analysis of the almost 100 times that [Supreme Court nominee Sonia] Sotomayor weighed in on issues related to race while serving as an appellate judge.” Goldstein:

In addition to Ricci v. DeStefano, Judge Sotomayor has participated in 97 race-related cases. Of these, the court of appeals rejected the claim of discrimination roughly 80 times and agreed with it 10 times. (The remaining cases involved other kinds of claims or dispositions.) In the 10 cases in which the court of appeals favored claims of discrimination, nine resulted in unanimous rulings and seven involved at least one Republican-appointed judge. In the single time a judge dissented from a ruling in which Judge Sotomayor participated, the dissent was over a technical question, not race discrimination.

In total, Judge Sotomayor has disagreed with her colleagues in race-related decisions—a fair measure of whether she is an outlier—only five times in 11 years. In that entire time, Judge Sotomayor has only twice dissented from a ruling on a substantive question of race discrimination.

Goldstein’s “bottom line,” according to Tucker: “after reviewing every single one of these race-related cases, he concluded that “Judge Sotomayor does not allow bias to infect her decision-making.” Tucker, like other Monkey Cage contributors, is concerned with applying the insights of political science research to contemporary public policy, which is today too dominated by specious economic models of human behavior. Although he acknowledges that Goldstein’s piece makes “an important contribution to our debate over Judge Sotomayor’s candidacy,” Tucker says that “as a social scientist,” he “would have liked to have heard a little bit more about [Goldstein's] standards for making this conclusion (e.g., what evidence would have led him to accept the claim that bias did influence her decision making?).” This is an excellent question of the genre that rarely get asked of public arguments purporting to use quantitative data to demonstrate policy points. As economics has slid into a role as the ur-discipline of policy thinking, the supposed integrity of numbers-as-numbers has multiplied to the point that citing numbers can and often does imply stability where nothing but ambiguity exists, for “experts” and generalists alike.

This sort of undue slavishness to the number in public discourse and policy is, indeed, is one of the things The Monkey Cage seems to hope to remedy. But Tucker could have gone a bit farther. The answer to the question he asks of Goldstein—What would constitute evidence that Sotomayor’s rulings on race had been influenced by bias?—is simple: aberration. If, in Goldstein’s schema, Sotomayor’s near constant concurrence with the majority in race-related cases constitutes proof of a lack of bias, then deviation from the majority would necessarily indicates the presence of bias. What precisely constitutes that bias, however—or, if we’d like to see it linearly, at what point along the scale between constant assent and constant dissent Sotomayor ceases to be unbiased and becomes biased—is mysterious.

I plunge the ambiguity of Goldstein’s metric of bias—Sotomayor’s relative concurrence or dissent from various sets of other judges in race-related cases—not to deny its efficacy but to highlight its instability. All Goldstein can tell us is that most of the time, Sotomayor agreed with other people in cases involving race. What he doesn’t, and can’t, tell us is why the rulings of Sotomayor’s fellow judges should be taken, in aggregate, as a reliable, bias-free norm against which to compare her.

We’re not asking here, really, whether or not Sotomayor is biased. We’re asking how often she has agreed with her colleagues, which is a very different question. Goldstein elides these two questions into one. He posits the average of Sotomayor’s colleagues as a fixed point of no-bias. But again: where is this point? What constitutes it? What allows us to say that the contingent result of Sotomayor’s intellectual collaborations with dozens of individuals, absent Sotomayor’s views as such, constitutes a universal norm? Why should we presume that the average of the biases of multiple others represents a default lack of bias? Who made all the judges she’s worked with the collective embodiment of a lack of bias when it comes to race?

What seems to be very solid in Goldstein’s analysis, what seems to be perhaps the sole given—the notion of bias as reliably measurable—is actually clumsy and precarious. There is no good reason to presume that the contingent result of multiple individuals’ intermingling biases is the embodiment of any sort of neutrality, much less any sort of correctness vis-a-vis “bias.” The assertion of this spectral neutral average is no more justified than Christine’s assertion that her own subjective judgment of the meanie moms is not judgmental because it’s “right.” The first is far more complicated, but skeletally they are the same, both relying for cohesion on the assertion of a neutral, bias-free average, each taking for itself the right to demarcate the judgmental from the universal. Christine’s assertion is excusable. It’s on TV and, from the perspective of the show, clearly facetious. Discussion of “bias” in judicial nominations is equally free of foundation, but its discussants—like lovable dolt Christine—are quite sincere.