Adrian Vermeule and Eric Posner have a pair of blog posts sketching out a new paper idea. Their idea is to explore the question: “Under what conditions should judges take into account the information contained in the votes of other judges?” For instance, if judges disagree over how to interpret a statute, does that mean that either interpretation is reasonable?

I’m very interested in reading what Posner and Vermeule work out about this. My own instinct is that it only makes sense for judges to take into account the votes of other judges when those other judges share relevant elements of an interpretative framework. A strong textualist’s view of the statute’s meaning is largely beside-the-point for a strong purposivist, and vice versa. Similarly, lower-court judges who approach the Anti-Terrorism and Effective Death Penalty Act in very different ways will not learn very much from one another’s votes.

Towards the end of their first post, Posner and Vermeule discuss this issue:

So far we have assumed that all Justices are using a common interpretive theory (in the examples, we have assumed that the Justices are all trying to determine the ordinary meaning of the text). But puzzles also arise at the meta-level of competing approaches to interpretation. In case (1), suppose that the five Justices are purposivists who think that purposive sources clearly suggest X, and the four are textualist who think that the ordinary meaning clearly indicates Y (or vice-versa). Does that undermine the argument for ambiguity? Or should all nine Justices recognize that reasonable minds can disagree about the proper approach to interpretation, and then say that the agency has second-order discretion to choose among reasonable interpretive approaches? On that logic, the agency wins as well, not because the statute is ambiguous within any particular interpretive approach, but because there is second-order ambiguity in the choice of interpretive approaches.

I think the answer to these questions are probably “yes” and “no,” respectively. Disagreement at the meta-level about the proper method of interpretation generally should not produce deference to the agency or other actor. Many textualists and purposivists believe that their method of interpretation is correct, and in any event they get no new information from the fact that the textualist judges keep being textualists and the purposivist judges keep being purposivist.

The importance of methodological disagreement can be drawn from some of the legal materials — for instance, the much-debated ninth footnote of Chevron says that “If a court, employing traditional tools of statutory construction, ascertains that Congress had an intention on the precise question at issue, that intention is the law and must be given effect.” I read that footnote to say that agency deference comes only after interpretive choice.

The point can instead be framed in terms of conditional probabilities — whether Judge A should update his or her vote and confidence level after learning about the vote of Judge B depends on Judge B’s confidence level, but also P(Judge B correct|Judge B’s confidence level). For judges who are not prone to certain cognitive biases and are pursuing the same interpretive project, updating might make sense, but I’m not so sure how often those circumstances describe real-world debates among real-world judges.

One final thought. The second post closes this way:

There is much more work to be done, on all these questions. But our tentative judgment is that the potential costs of such a system, while real, do not necessarily and invariably justify throwing away relevant information — the information contained in the votes of other judges.

This might well turn out to be right in the final analysis. But it’s worth remembering that sometimes “throwing away relevant information” is indeed a very good idea, because the information is noisy enough that it’s not worth the decision costs. The point that extremely simple interpretive theories are sometimes better was made in a profound book by … Adrian Vermeule.