With Scalia Gone, Who’s Wearing the Thick Grammarian’s Spectacles Now?

In 1991, the Supreme Court decided West Virginia University Hospitals v. Casey (opinions here).  The case determined that the right to recover “a reasonable attorney’s fee” if successful in a civil rights case against a state did not include the right to recover the cost of expert witness services.  Writing for a majority of six, Justice Scalia focused on the importance of interpreting the United States Code as a coherent body of law, listing many examples of fee shifting statutes that do mention expert fees, and inferring that their absence in the civil rights law implies that Congress did not intend for the winning party to recover them. 

Justice Stevens dissented.  The fee-shifting statute was actually a congressional reaction to an earlier stingy ruling by the Supreme Court.  Offering once again the least generous interpretation of the civil rights law, Stevens argued, would undermine the congressional effort to override the Court.  Stevens then noted that congressional overrides of the Supreme Court come primarily when “the Court has put on its thick grammarian's spectacles and ignored the available evidence of congressional purpose and the teaching of prior cases construing a statute.”   And that is exactly what happened.  Congress quickly overrode this decision too, making it clear that expert fees were to be recoverable by a winning plaintiff in a civil rights case.

Now that Justice Scalia is gone, who has the spectacles?  We got a surprising hint last week when Justices Sotomayor and Kagan fought over the application of grammatical principles in Lockhart v. United States (here).  It appears that all of the justices will continue in Scalia’s tradition of close textual analysis, although in this case neither side did it very well. 

Lockhart was convicted of possession of child pornography.  The statute carries a mandatory minimum sentence for those who have already been convicted of certain federal statutes involving sexual abuse, or have been convicted “under the laws of any State relating to aggravated sexual abuse, sexual abuse, or abusive sexual conduct involving a minor or ward.”  Lockhart had earlier been convicted in New York for having sexually abused his then 53-year old girlfriend.

The case revolved around the words “involving a minor or ward.”  If that phrase modifies only “abusive sexual conduct,” then Lockhart’s conviction for abusing his adult girlfriend puts him within the sentence enhancement.  But if “involving a minor or ward” modifies all three crimes, then Lockhart would not be subject to the longer prison term since his crime did not involve a  minor.

Sotomayor’s majority opinion relied on 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 rule is subject to exception when context demands it, and Kagan’s dissent argued that this was such a case.  Rather, the dissent argued, the final phrase modifies the entire list of three crimes.

My own analysis is that the rule itself applies ambiguously, and thus makes no contribution to resolving the case.  Consider the following sentence:

              Bill saw a man and a woman with a young child

There are three readings:

 1.      The young child was with the woman;

              2.      The young child was with the man and the woman;
              3.      The young child was with Bill.

The last antecedent rule tells us to disregard interpretation 3.  Bill is clearly not the last antecedent.  But Lockhart was, in essence, about choosing between 1 and 2.  “A woman” is the last noun before “with a young child,” and “a man and a woman” is the last phrase before “with a young child.”  Indeed, as linguists recognize, we can understand “with a young child” either way.  Both sides present examples from ordinary speech to support their position:  the modifier refers to the last member of the adjacent phrase (Find "a defensive catcher, a quick-footed shortstop, or a pitcher from last year’s World Champion Kansas City Royals”); or the modifier refers to the entire adjacent phrase (I hope to meet “an actor, director, or producer involved with the new Star Wars movie.”).  Why this disarray?  A combination of context and the lengths of the list and the modifier seem to contribute to our preferred understanding.

The non-linguistic arguments on both sides, in contrast, were persuasive.  The majority noted that the three types of state crimes mimicked virtually verbatim the list of federal crimes that would also result in an enhanced sentence, and that only the third of the federal crimes involves minors.  Why would Congress create an asymmetry between state and federal triggers for sentence enhancement when a perfectly natural reading of the law would increase the penalty for prior convictions regardless of which court heard the case?  The dissent, in contrast, cited legislative history that strongly suggested congressional concern with individuals who have repeatedly preyed on children.  After all, Lockhart was convicted of violating the child pornography statute, but had never been convicted of a crime involving the abuse of children.

In some ways, nothing much has changed since 1991.  The substantive disagreement was over whether a general preference for coherence should trump detailed inquiry into the history of the particular statute in dispute.  Once again, coherence won – this time, by a vote of 6 – 2.  How would Scalia have voted?  Perhaps he would have sided with the dissenters, relying on the rule of lenity in cases of linguistic ambiguity in criminal statutes.  He frequently dissented on those grounds.  Or perhaps he would have stuck with the majority’s rather persuasive coherence argument, the same kind of argument he had used a quarter century earlier in Casey.  Only the remaining justices and their clerks know the answer to that one.