Social Science Statistics Blog

I'm being a bit snarky here, but why should we trust your "instincts?" There's an extensive, if not dispositive, political science literature arguing that factors like precedent are not, in fact, influential except in very specific (and typically uninteresting, from a substantive viewpoint) circumstances. This is a methods blog, not an Empirical Legal Studies-type blog, so we don't need to get too far into this, but it bears mentioning.

It's a bit ironic that this is posted on a day where the Supreme Court essentially reverses a line of precedent that was upheld relatively recently. Some might argue that this is a direct result of a move in the median of the Court; in any case, it is hard to argue that precedent (such as Carhart or Casey) honestly constrained how the justices decided this particular case.

Hi, Anon, personally, I don't find your question snarky at all.

There's no question that courts do on occasion overrule or "overrule" (meaning depart from a prior case without acknowledging that it is doing so) precedent. I wonder what it says about the role of precedent that the court-watching and legal communities seem to notice when a court acts in this manner.

With respect to the political science literature you mention, I'd place against it the nature of law school curricula, particularly in the first year. That's when virtually all law schools in the country beat precedent-based thinking and argument style into their students, and most judges have been to law school. I'd also place against it the appellate practitioners who attempt to persuade judges to rule a certain way. Their briefs and oral arguments suggest that they view precedent as important. (In my view, the odd Brandeis brief is the exception that prove the rule.) Finally, the opinions of the judges themselves suggest a need to show how a present decision is consistent with earlier ones. Experience in these areas (as a law student, a practicing lawyer, and a clerk who drafted appellate opinions) is the source of the "instincts" I referred to above.

I'm not denying that lawyers are taught to argue from precedent, but I wonder if what they actually learn (particularly the clever ones) is how to justify an outcome by appealing to precedent. It seems that, if adhering to precedent mattered, the findings in Segal and Epstein 1996 (for example) would have been decidedly different. As would have yesterday's decision. Anyway, I think we're moving far afield from issues of methods.

I'll need to read the Segal and Epstein reference to comment on it (which will have to wait on my dissertation defense, alas), but I had two quick thoughts. First, it could be that for cases in certain extremely high-profile subject areas such as abortion, precedent matters less, although I would still expect it to matter. But in a dataset consisting of appellate cases, how many are extremely high-profile? And even in an area like abortion: would the Supreme Court justices have decided Griswold, Eisenstadt, and Roe in the same way if these cases had arisen (and been decided) in the reverse order?

Second, I don't think we're afield from methods at all. In my view, the applied researcher contributes at least as much to an understanding of how the world does and should operate by scrutinizing the assumptions underpinning a chosen mathematical algorithm than by applying that algorithm.

There's an argument that precedent is just a beard for preferences, but I think the criticism of the study is the failure to control for what may possibly be an effect of precedent -- you can't just assume it away.

The optimal place to do the research on appellate courts is probably circuit splits. Here you have courts of different ideological makeup resolving the same particular issue, such as the interpretation of a statutory provision. It's possible that this will be tainted by prior circuit precedent but these are often cases of first impression.

Or, you can try to study precedent itself to see its effect. There's good evidence that new Supreme Court precedents have a powerful effect on circuit court outcomes independent of ideology.