As has already been noted here, Dan Kahan dropped by the law school earlier this week and gave three fascinating presentations to the law school community. One, which Michael commented on earlier, was on his paper (co-authored with David Hoffman and Donald Braman) criticizing the Supreme Court’s decision in Scott v. Harris, entitled “Whose Eyes are You Going to Believe? Scott v. Harris and the Perils of Cognitive Illiberalism.”

In brief, Kahan and his co-authors argue that the Supreme Court went awry in Scott by refusing to credit the views of “an identifiable subcommunity” as being within the realm of those held by “reasonable jurors.” This refusal to credit such beliefs with reasonableness, they argue, is potentially destructive of the legitimacy of the justice system.

It’s a fascinating argument, backed by a novel empirical approach to assessing the views of “reasonable jurors” in a use of force case like Scott. But I’m left with a question about the theory, and a question about Scott: Today, I want to focus on the theory: How are judges to tell when the views of “an identifiable subcommunity” are at issue, making summary judgement less appropriate? Monday, I’ll focus on Scott: I’m not certain that the Scott holding is as Kahan et al. describe it, which way may mute their concern.

Scott v. Harris was the case from last term that involved the reasonableness of a police officer’s decision to run a fleeing speeder off the road. Harris, the speeding driver, sued Scott, the police officer who rammed him, for violating the Fourth Amendment. (What really should have been at issue was the police department’s policies and procedures allowing an untrained officer to attempt such a dangerous maneuver on a non-violent offender, but the Supreme Court eliminated the possibility of such suits long ago.) Ordinarily, in deciding whether to grant summary judgement, the court must interpret the disputed facts in the light most favorable to the non-moving party–here, Harris, the fleeing motorist. The wrinkle in the case was that there were two videos of the chase, captured by the police’s dashboard cameras, adding a twist to the usual summary judgement issue of what the “genuine issues of material fact” were. As Justice Scalia, for the majority, described the problem, “The videotape quite clearly contradicts the version of the story told by [Harris] and adopted by the Court of Appeals.” Scalia characterizes the video as resembling “a Hollywood-style car chase of the most frightening sort, placing police officers and innocent bystanders alike at great risk of serious injury.” Think Ronin. (Although frankly, the facts seemed a bit more Raising Arizona to me.) “[Harris’s] version of events is so utterly discredited by the record that no reasonable jury could have believed him. The Court of Appeals should not have relied on such visible fiction; it should have viewed the facts in the light depicted by the videotape.”

Kahan et al. argue that in reversing the denial of summary judgement for the police officer, the Supreme Court ran roughshod over the views of a substantial subcommunity that viewed the video far differently than did the majority–one that they have identified and measured through public opinion polling. It’s not that a jury verdict would likely have come out the other way–indeed, given the data, it looks unlikely that Harris would have prevailed before a jury. But the court, by awarding summary judgement, didn’t give the jury that chance. Kahan et al. argue that summary judgement for Scott “was wrong precisely because doing so denied a dissenting group of citizens the respect they are owed, and hence denied the law the legitimacy it needs, when the law adopts a view of the facts that divides citizens on social, cultural, and political lines.”

Kahan and his co-authors are careful to note, however, that they are not arguing that summary judgement is inappropriate whenever some reasonable juror exists who might disagree about the facts. Courts can still grant summary judgement even in the knowledge that “outliers” exist who would disagree with the court’s conclusions as to the “genuine issues of material fact.” Rather, they argue that it is only when such outliers cluster around certain identity-defining characteristics–“demographic, cultural, political, or otherwise”–that the court should pause to consider whether summary judgement is appropriate.

How is a judge to determine whether the outliers share such “identity-defining characteristics,” in the absence of polling data? Kahan et al. suggest that the judge should “attend to cues that a cultural subcommunity will react with outrage” to a given statement of undisputable fact. (Which raises the question: Has Scott v. Harris produced any outrage?) For example, if the issue raised is “a potentially divisive matter in our society,” or if “numerous public interest groups [have] filed briefs” in the matter, or if “coverage of the case in the media . . . suggest[s] the decision would be controversial,” then the judge should second-guess his or her intuitive appraisal of the facts. The judge should also try to imagine who might object to a description of the facts and see if they share certain identifying characteristics–that is, if they form a “subcommunity”.

I don’t see this being a workable test, however. The three rough measures Kahan et al. propose strike me as likely to be present in many, many cases, and certainly most Supreme Court cases. That leaves the issue of whether the outliers cluster in a certain “subcommunity.” I have two difficulties there. First, I am skeptical that a judge who believes that the interpretation of the facts is beyond the zone of jury discretion would be able to identify particular “subcommunities” that would disagree with him or her. Second, I’m confused about what qualifies as a “subcommunity” that would trigger heightened review. What “identifying characteristics” count? For example, let’s say that race car drivers are particularly likely to view the facts contrary to the way the majority does. Or less ridiculously, males in their late teens. Let’s say 80% of the outliers fall into one of those two groups. Do those identifying characteristics count? Should the question now go to the jury, so that race car drivers and teenage males can have their say (and likely be voted down)? Kahan et al. don’t really answer this, but I suspect the answer has to be no. It’s not just any old characteristics that matter. It’s characteristics that define groups that we are particularly sensitive to being excluded from juries (or judgeships)–racial minorities, women, the poor. Age and profession are just not subcommunities that generate the same sort of intuitive weight as the communities that Kahan et al. allude to when they say that “coercive, near-deadly encounter[s] between police and a citizen” are “always a potentially divisive matter in our society.” (Emphasis added.)

A related question is, how many characteristics should count for purposes of a single inquiry? Above, I grouped race car drivers and teenage males together. Let’s say you only get to a substantial number of outliers by clumping various groups together–should that clump generate concern, or not? That is, do the outliers have to share a single characteristic to generate heightened review, or will multiple characteristics do the trick? Again, Kahan et al. are a bit fuzzy on this, but their use of their data suggests the latter. Although they use artificial “types” to represent the different views of Scott, they are very careful to muddy the demographic waters as to who disagrees with the majority’s view of the facts. Although blacks and women are both associated with some disagreement with the Scott majority (a fact they allude to by using “Linda,” a black female Eastern liberal Democrat social worker, to represent a Scott dissenter), they also make clear that Scott dissenters cut across various demographic lines. Indeed, while the pro-Scott forces are represented by a single icon, “Ron,” a rich white Western Republican man, the dissenters have two: in addition to “Linda,” there’s also “Bernie,” a white male Eastern liberal Democrat university professor. Are Linda and Bernie part of an “identifiable subcommunity”? Identifiable how? They’re both liberal Democrats, but it’s not clear that they represent a significant portion of liberal Democrats. As far as I can tell, the essential characteristic that identifies them is that they are more likely to take a dim view of the police running speeders off the road. In other words, I’m not sure what identifies them other than the fact that they may be outliers on the Scott issue.

Kahan et al., drawing on the work of Mary Douglas and Aaron Wildavsky, classify the Scott outliers as “bets,” or what Douglas & Wildavsky would describe as “egalitarian communitarians.” (For more, see Kahan’s Boden Lecture.) But this seems like an extremely abstract subcommunity to me. There are no meetings of the Egalitarian Communitarian Society. I don’t see how a judge would be able to tell whether their description of the facts would be more likely to offend a higher proportion (but perhaps not a majority) of “egalitarian communitarians”. That’s not likely to show up in the news, or amicus briefs, or protests, or even public opinion polls. Furthermore, the category is so broad that I’m skeptical it retains much significance if all of the outliers happen to be “egalitarian communitarians.”

Rather, it does seem to me to be significant if a judicial statement of the “facts not reasonably subject to dispute” was one that, say, blacks or women as a group would be highly likely to disagree with, while non-blacks or men would be highly likely to agree with it, and the judge him- or herself was in the latter categories. Given my concerns above, I don’t know that I have a good way to address that issue, however, except for two suggestions: first, make sure the courts stick to a principled application of the summary judgement standard, and second, make sure the judiciary represents a variety of demographic backgrounds. And as to the first suggestion, at least, I believe Scott is vulnerable, as I’ll discuss in my next post.