Do Justices Vote For Those They Can Relate To?

August 7th, 2013

For some time, my intuition has been that ideology alone does not explain how the Court rules on important constitutional law cases. If it did, all the 5-4s would go as expected–this is seldom the case. And I’m not just talking about  Kennedy. Scalia, Breyer, Ginsburg, and others frequently throw us curve balls. Rather, something else informs the decision. The normal human impulse is to place yourself into the case, and assess the situation from that vantage point. Judges are not immune from this. But the ability to transpose oneself into a situation varies, widely. For starters, whose shoes do you walk in?

To use the Second Amendment as an example, I think most people would agree that Justice Scalia’s opinions are written from the point of view of someone who would use a handgun for self-defense. In contrast, Justice Breyer’s opinions are written from the point of view of someone who may be shot by a handgun. I have previously discussed these opinions in terms of social cost, but viewed through what political scientists call “in-group” bias, judges may vote with those they can relate to. (I think you’ll find similar breakdowns when discussing “Stand Your Ground Laws”–are you more likely to be the one standing your ground, or shot).

Another related example is United States v. Jones, the GPS case. During argument, the Justices–Roberts in particular–were very concerned about the prospect of the police attaching a GPS to his own car.

Look how the Chief phrased this question–he used the word “our,” referring to the Justices:

CHIEF JUSTICE ROBERTS: You think there would also not be a search if you put a GPS device on all of our cars, monitored our movements for a month? You think you’re entitled to do that under your theory?

MR. DREEBEN: The Justices of this Court?


Mr. Dreeben: “Under our theory and under this court’s cases, the justices of this court, when driving on public roadways, have no greater expectation…”

Chief Justice Roberts: “So your answer is yes, you could tomorrow decide that you put a GPS device on every one of our cars, follow us for a month, no problem under the Constitution?”

Mr. Dreeben: “Well, equally, Mr. Chief Justice, if the FBI wanted to, it could put its team of surveillance agents around the clock on any individual and follow that individual’s movements as they went around on the public streets and they would thereby gather…”

After the case was argued, I blogged about Linda Greenhouse’s comment that reflects this dynamic:

In any event, it’s not implausible to suppose that the outcome of the GPS case will depend in large part on the justices’ view of reasonable government behavior toward a citizenry that includes themselves. In fact, it’s implausible to suppose otherwise.

In close cases, the ability to relate to the situation–that is, could this happen to me!?–may help decide the issue.

A new article by Lee Epstein, Christopher Parker, and Jeffrey Segal puts this claim to the empirical test in the context of the First Amendment–do justices, apart from ideology, vote to protect speech they view more favorably? Here is the abstract for Do Justices Defend the Speech They Hate? In-Group Bias, Opportunism, and the First Amendment:

While scholars debate whether liberals or conservatives on the Supreme Court are more supportive of First Amendment rights (compare Pritchett [1948] and Epstein, Landes, and Posner [2013] to Volokh [2011] and Sullivan [1992]), we argue that the Court’s First Amendment jurisprudence may fruitfully be viewed through the lens of In-Group Bias, a fundamental norm in social psychology that holds that humans tend to evaluate our own group and its members more favorably than outsiders.

From this perspective we argue that liberal (conservative) justices will evaluate speech claims made by liberal (conservative) parties more favorably than they will speech claims made by conservative (liberal) parties, after controlling for other salient factors. After correcting the coding in Spaeth’s Supreme Court database for ideological direction in cases in which the First Amendment issue is not the main issue, our multilevel hierarchical model finds a substantial in-group bias in the votes of U.S. Supreme Court justices from the 1953-2010 Terms of the Court. The findings hold up through a series of robustness checks.

The authors also address the claim that the Justices support speech they like.

Just as there is some evidence consistent with in-group bias in existing studies, suggestions in the journalistic and legal literature provide some basis for the idea that justices tend to regulate the speech and speakers they hate, to put it succinctly if crassly. Bazelon (2011) writes of Alito’s “empathy” to “people who are …like him”— the father burying his son amid Westboro Baptist Church protestors13 or the law students who want to exclude gay students from their club.14 Winkler too critiques the Roberts Court for“strongly protect[ing] speech that it likes, while allowing regulation of speech it disfavors” (quoted in Stohr, 2011). To our knowledge, though, these ideas—reflective of in-group bias—have never been put to a systematic test.

Epstein, et al, view the in-group bias in terms of “litigant favoritism”–do the Justices like the litigants.

This is where in-group bias enters the picture—as a possible means for resolving tension in the First Amendment literature. Rather than modeling cases along the standard left- right dimension that political scientists continue to use but that legal academics reverse (or outright reject), in-group bias commends adding a second: the ideological nature of the litigants. Relating this dimension to the context of freedom of expression,12 leads to the hypothesis that judges engage in opportunistic behavior following from litigant favoritism: judicial votes are not reflexively pro- or anti-the First Amendment but rather pro- or anti- the nature of the speech (speaker).

And this phenomenon is likely not limited to the First Amendment:

This evidence relates most directly to judging in the free expression context but our study’s implications may be broader. For one thing, it’s hard to imagine that the in-group bias we identify here is cabined to the First Amendment. And, actually, the studies we mentioned earlier examining race, gender, and ethnicity suggest that it is not (e.g. Abrams, Betrand and Mullainathan, 2012; Shayo and Zussman, 2011). We could say the same for research on taxation (Staudt, Epstein and Wiedenbeck, 2006), criminal procedure (Epstein, Landes and Posner, 2013), and federalism (Parker, 2011).

Very interesting research.