Another Syntactic-Ambiguity Cautionary Tale (As If We Really Needed One): The Supreme Court’s Opinion in Lockhart

Here we go again.

You might recall that syntactic ambiguity involves uncertainty over what part of a sentence a phrase modifies, or what part of a phrase a word modifies. If you want a whole bunch of examples of syntactic ambiguity, just search for “syntactic” on this blog.

Well, the most recent Supreme Court opinion, Lockhart v. U.S. (opinion PDF here), involves syntactic ambiguity. Here’s the language at issue:

relating to aggravated sexual abuse, sexual abuse, or abusive sexual conduct involving a minor or ward

The question was whether the trailing modifier “involving a minor or a ward” modifies just “abusive sexual conduct” or also modifies “aggravated sexual abuse” and “sexual abuse”. As a citizen, I’m sure I care deeply. But as a contract drafter, I don’t give a flying f*ck, because I’m paranoid about excluding this sort of confusion from my contracts.

In her opinion, Justice Sotomayor offers evidence for the majority’s conclusion that the trailing modifier modifies just the last element. That’s as it should be. But she also trots out that most depressing of subjects, the rule of the last antecedent, which says that a limiting clause or phrase should ordinarily be read as modifying only the noun or phrase that it immediately follows.

The problem with the rule of the last antecedent is the problem with all principles of interpretation: because they rely on generalized notions of now people understand text, they’re arbitrary when applied to any one bit of text. As Justice Kagan demonstrates in her dissent, it’s a simple matter to come up with examples in which any reasonable reader would conclude that the writer intended for the trailing modifier to modify all the preceding elements.

So the rule of the last antecedent is, like other principles of interpretation, a convenient fiction that is put aside whenever it proves inconvenient. Principles of interpretation are a mess. (Those who favor principles of interpretation are prone to calling them “canons,” presumably in the hope that the ecclesiastical-law origins of that word will give principles of interpretation a sanctified glow.)

It’s amusing that in her opinion, Justice Sotomayor describes the rule of the last antecedent as “timeworn.” But I think she meant time-tested, or some such, whereas timeworn means worn out, threadbare. So Justice Sotomayor inadvertently did choose an apt word.

The best discussion of the rule of the last antecedent that I’ve encountered is by Joe Kimble, in his article in the most recent issue of The Scribes Journal of Legal Writing, The Doctrine of the Last Antecedent, the Example in Barnhart, Why Both Are Weak, and How Textualism Postures (PDF here). For a linguistics take on Lockhart, see this post by Neal Goldfarb. And if you want to see where the rule of the last antecedent really goes off the rails, check out my article Bamboozled by a Comma: The Second Circuit’s Misdiagnosis of Ambiguity in American International Group, Inc. v. Bank of America Corp., from the same issue of Scribes as Joe’s article (PDF here).

Again, the lesson for those who draft and review contracts is to become attuned to this sort of confusion and purge it from your contracts. It would be disastrous to rely on the courts and the rule of the last antecedent to bail you out. MSCD contains a chapter on syntactic ambiguity; if anyone know of a better treatment of the subject, I’d like to hear about it.

While I’m at it, I noticed this article on Lockhart by Harvard Law School professor Noah Feldman. Two problems: First, despite its title, the opinion has nothing to do with stuffy language, but with stuffy interpretation of language. And second, the language at issue in Lockhart involves a trailing modifier, not a dangling modifier. Dangling modifiers are something else entirely; see this Grammar Bytes item. Why point this out? Because there’s already enough confusion surrounding the rule of the last antecedent, it’s too bad to have a prominent commentator add to it.