When are Supreme Court justices most willing to stand up for free speech? According to a new study, it depends on whether a liberal or conservative is doing the speaking:
The study, by Professors Lee Epstein, Christopher Parker, and Jeffrey Segal, found that conservative justices were more likely to protect the speech of conservatives, and liberals were more likely to do so for liberal speakers. "Justices are opportunistic free speechers," the authors write. "They are willing to turn back regulation of expression when the expression conforms to their values and uphold it when the expression and their preferences collide."
Epstein and her colleagues pulled data for all First Amendment cases that the Supreme Court resolved after argument between 1953 and 2010. This turned out to be 516 cases, in which 33 different justices cast a total of 4,519 votes. They also used existing data on the ideological orientation of justices.
Traditional political science research has tended to treat decisions supporting free speech as liberal, and decisions opposing free speech as conservative. But Epstein and her co-authors decided to focus on whether the controversial speech in each case was liberal or conservative, and code it as such.
For instance, a student who held up a banner reading "Bong Hits 4 Jesus" at a school event would be coded as a liberal speaker, as would a whistleblower trying to expose police corruption. But pro-life protesters outside an abortion clinic, or the Boy Scouts of America arguing for their rights not to admit gays, would be conservative speakers.
So the authors then checked whether, after controlling for other variables, the ideological orientation of a justice plus the ideological orientation of the speaker would predict the justice's vote in a free speech case. That is, whether conservative justices would defend the speech of conservatives more than liberals.
The study found a stark difference between the justices' votes on free speech cases, depending on whether liberals or conservatives were doing the speaking. "Though the results are consistent with a long line of research in the social sciences, I still find them stunning - shocking, really," Epstein told the New York Times. This chart shows the results of their model:
The chart shows that, in general, conservative justices are less likely to defend free speech than liberal ones — conservatives defend it in one-third of cases, and liberals defend it in about two-thirds. But justices from either group are much more likely to defend the speech of ideologically similar speakers, and less likely to defend the speech of ideological opponents.
Note particularly how far apart the blue dots are. It shows that free speech cases involving liberal speakers divide justices the most. A liberal justice is overwhelmingly likely to support the speaker, while a conservative justice is overwhelmingly likely not to.
Overall, the results suggest that when justices "face a conflict between their standard ideological positions on the First Amendment and their preferences regarding the speaker's ideological grouping, they place significant weight on the latter," the authors write. They add, "this is precisely the result in-group bias accounts anticipate."
The crucial data on whether the speech at issue in a case is liberal or conservative, was hand-coded by study authors Epstein and Parker. "There was almost no disagreement in their codings," they write. But ideally, in a study about bias, the authors wouldn't be using data they themselves assembled.