A recent study conducted by Maya Sen, a political scientist at the University of Rochester, and Adam Glynn, a government professor at Harvard, shows that judges who have at least one female child may be more likely to rule in favor of women in certain types of cases. The report “Identifying Judicial Empathy: Does Having Daughters Cause Judges to Rule for Women’s Issues?” finds that having at least one daughter corresponds to a 7 percent increase in the proportion of cases in which a judge will vote in a feminist direction. The study further finds that having one daughter as opposed to one son is linked to a 16 percent increase in the proportion of “gender-related cases decided in a feminist direction.” The study found the “daughter effect” was more dramatic in judges appointed by Republican presidents than in those appointed by Democrats.
Sen told the New York Times in a recent interview, “By having at least one daughter, judges learn what it’s like to be a woman, perhaps a young woman, who might have to deal with issues like equity in terms of pay, university admissions or taking care of children.” Sen and Glynn consider other causal explanations for their findings, but conclude that learning is the mechanism at play. For example, they rule out the possibility that parents of daughters feel compelled to rule in ways that would protect their female children Sen and Glynn saw an effect only in gender-related civil cases, not a conservative shift among gender-related criminal cases like sexual assault.
The problem with the study is not that the data are wrong. The problem is that too often those who use data like these mean to either exempt the judgments from moral consequence altogether or to praise particular judicial motivations that they happen to like. In the first instance, they justify legal realism with data, omitting any reflection on whether the observed effects can or should be minimized. They gloss over too the overwhelming number of cases that are decided by mundane, less-subjective methods. In the second variation, they celebrate the phenomenon as “empathy” with some results, while condemning it as “bias” in others . . . .
Sen and Glynn describe their use of case-level data on pages 7 and 8 of their report. They collected all Courts of Appeals cases, published and unpublished, decided from 1996 to 2002 that had “gender,” “pregnancy,” or “sex” in the LexisNexis case classification headings. They then pared the set down to cases that explicitly involved employment discrimination on the basis of gender by private actors, employment discrimination on the basis of pregnancy by private actors, reproductive rights or abortion, and Title IX claims. They eliminated cases where plaintiffs were male or the issue turned on LGBT interests. Nine-hundred ninety cases remained, most of them employment discrimination cases.
The researchers coded the judges’ votes as being “(1) antifeminist or (2) partially or entirely feminist.” What did they count as “feminist”? They write: “The feminist decision would be one in which the judge voted in favor of the female plaintiff or in favor of the plaintiff representing women’s interest (e.g., if the Equal Employment Opportunity Commission brought a suit on behalf of female employees). The antifeminist decision would be one in which the judge voted in favor of defendant employers or in favor of sustaining a restrictive abortion statute. The partially feminist decision would be one in which the judge’s vote was mixed, for example, an instance in which the plaintiff won some claims but lost others.” Note that the directionality of the decisions was coded in a binary way — mixed decisions counted as “feminist.”
The terms “feminist” and “anti-feminist” as applied here are galling for anyone whose brand of feminism does not include abortion on demand or genuflection to every employment discrimination claim made by other women. Unfortunately, conservative feminists are pretty accustomed to the “no true Scotsman” response to our arguments that these sorts of decisions should not be blithely labeled “feminist.” So, let’s momentarily overlook the rhetorical implication of the labels.
More substantively, consider this: If judges with daughters rule in favor of female plaintiffs in employment discrimination cases more frequently than judges with sons, that does not make those judges more “feminist.” It makes them biased toward women plaintiffs. This may be true. Its truth, however, is no more worth celebrating as enlightenment than judges whose bias runs slightly in favor of other classes of plaintiffs.
Professors Sen and Glynn appear too responsible to make grandiose or overtly partisan claims about their findings, though they understandably wish to draw media attention to their work. Nevertheless, some media outlets are perfectly happy to do that for them. A headline asks, “How Do We Get More Female-Friendly Courts?” Readers then get snapshots like this: “If we want there to be more feminist courtroom rulings, we just need to appoint more judges with daughters. Apparently, it really is that simple [ . . . ] So, here’s to hoping we see more judges with daughters in the court room….or with adopted homosexual daughters.” Yay! Three cheers for bias when it favors me or people I like!
Even more circumspect reports seem to interpret these findings favorably, as evidence that this phenomenon simply reflects a benevolent appreciation for the plight of others. For example, Sharon McCloskey writes:
“It’s been widely assumed though not necessarily documented that life experiences shape a judge’s world views.
And indeed that notion makes perfect sense. If you’ve walked the walk of a person on unemployment, or a person who’s endured on the job discrimination, you’ve learned a little about that experience.
U.S. Supreme Court Justice Ruth Bader Ginsburg thought as much when she attributed former Chief Justice William Rehnquist’s shift on women’s issues to his dealings with a daughter struggling as a newly-divorced mother with a demanding job.
But now political scientists have proof.”
Many of the reports feature this quip from Justice Ruth Bader Ginsburg in which she attributes the vote of Chief Justice Rehnquist in Nevada v. Hibbs to his experiences with his daughter. Rehnquist authored an opinion in that case, joined by the liberal members of the Court, that seemed to betray his usual states’-rights philosophy. If, indeed, the Chief decided that case based on compassion for his daughter, rather than law, why is that laudable? There’s an implication here that it was a good thing, if shocking one, that Rehnquist leaned liberal in Hibbs. I guess Justice Kennedy, who authored the dissent, just had not learned enough from his daughter Kristin. Neither apparently did Justices Thomas or Scalia (another father of daughters!) who joined him.
Would commentators respond with smug satisfaction if we learned that liberal judges deviate from the party line when they have children or loved ones who are members of the military or law enforcement? Would they consider it “empathy” if judges whose children are business owners frequently side with the Chamber of Commerce in cases where it has a stake? Or would that count as impermissible bias? Before glomming onto a cheerful interpretation of this study’s findings, readers should examine their own biases . . . I mean “empathy.”
Tamara Tabo is a summa cum laude graduate of the Thurgood Marshall School of Law at Texas Southern University, where she served as Editor-in-Chief of the school’s law review. After graduation, she clerked on the U.S. Court of Appeals for the Fifth Circuit. She will be working at the Center for Legal Pedagogy at Texas Southern University during the 2013-2014 academic year. She looks forward to a career of teaching and writing about, but never practicing, law. You can reach her at firstname.lastname@example.org