Syntactic ambiguity (Part 3 of Scalia and Garner on Statutory Interpretation)

[Updated. Part 1; Part 2]

One of the recurring problems in interpreting statutes and other texts is that of syntactic (i.e., grammatical) ambiguity. Reading Law sets out three canons of interpretation that are intended to deal with this problem: the Last-Antecedent Canon, the Series-Qualifier Canon, and the Nearest-Reasonable Referent Canon. I’m going to take a look at these canons, but I think that it would be helpful if I first say a few words about syntactic ambiguity and how to analyze it.

An expression is syntactically ambiguous when the string of words can plausibly be assigned more than one syntactic structure. Consider the sentence I once shot an elephant in my pajamas.  The sentence can be read in two ways: (1) ‘I once shot an elephant, who was wearing my pajamas’ and (2) ‘Wearing my pajamas, I once shot an elephant.’ Each of these readings is associated with a different phrase structure, as shown in the following diagrams (S = sentence, NP = noun phrase, VP = verb phrase, Adv = adverb, V = verb, PP = prepositional phrase):

Reading 1:

Reading 2:

The classic legal case involving syntactic ambiguity is Liparota v. United States, a Supreme Court decision from 1985. The statute there imposed criminal penalties on anyone who “knowingly uses, transfers, acquired, alters, or possesses [food stamps] in any manner not authorized by [law].”The question was whether someone who used, transferred, acquired, altered, or possessed food stamps illegally could be convicted without proof that they knew that their use, transfer, etc. was illegal.

The statute could be read in two ways, each of them associated with a different phrase structure. [Update: See the note at the end of the post.] The government argued for an interpretation in which knowingly modified only the phrase uses…food stamps:

The defendant, on the other hand, argued that knowingly modified the larger phrase uses…food stamps in any manner not authorized by law:

These alternative phrase structures illustrate one of the insights reflected in the use of these kind of diagrams (known as “tree diagrams”). The overall expression that I’ve been focusing on consists of phrases that are themselves composed of phrases—sometimes phrases of the same kind. Thus, uses…food stamps is a verb phrase (VP), and it is embedded in the larger VP uses food stamps in a manner not authorized by law.

Linguists refer to the kind of ambiguity displayed here as “attachment ambiguity”; the ambiguity arises because there are two different ways in which the modifier (knowingly) is integrated into the overall phrase. In the government’s interpretation, the adverb is attached to the lower-level VP, while in the defendant’s interpretation it is attached to VP one level up (referred to as the bottom-level VP’s “parent”). These alternatives are referred to, logically enough, as “low attachment” and “high attachment.”

The syntactic relationships shown by the tree diagrams correspond to the semantic relationships that go into how the overall phrase is understood. So with the low-attachment phrase structure, the adverb’s scope is limited to the bottom-level VP that it is attached to, meaning that it modifies only that VP. With the high-attachment structure, on the other hand, it has scope over (and therefore modifies) the bottom-level VP’s parent.

Update: My statement that each of the two interpretations of the statute in Liparota is “associated with a different phrase structure” is a bit of an oversimplification. Even with the high-attachment structure, the statute is ambiguous between (1) a reading on which the government must prove that the defendant knew that the manner in which he used/transferred/etc. food stamps was illegal and (2) a reading in which the government must prove that the defendant was aware of the manner in which he used/transferred/etc. food stamps, and the use/transfer/etc. was illegal, but not that the defendant was aware of the illegality.

Justice White argued for the latter interpretation in a dissenting opinion:

Even accepting that “knowingly” does extend through the sentence, or at least that we should read § 2024(b)(1) as if it does, the statute does not mean what the Court says it does. Rather, it requires only that the defendant be aware of the relevant aspects of his conduct. A requirement that the defendant know that he is acting in a particular manner, coupled with the fact that that manner is forbidden, does not establish a defense of ignorance of the law. It creates only a defense of ignorance or mistake of fact. Knowingly to do something that is unauthorized by law is not the same as doing something knowing that it is unauthorized by law.

This point is demonstrated by the hypothetical statute referred to by the majority, which punishes one who “knowingly sells a security without a permit.” . . . Even if “knowingly” does reach “without a permit,” I would think that a defendant who knew that he did not have a permit, though not that a permit was required, could be convicted.

Section 2024(b)(1) is an identical statute, except that, instead of detailing the various legal requirements, it incorporates them by proscribing use of coupons “in any manner not authorized” by law. This shorthand approach to drafting does not transform knowledge of illegality into an element of the crime. As written, § 2024(b)(1) is substantively no different than if it had been broken down into a collection of specific provisions making crimes of particular improper uses. For example, food stamps cannot be used to purchase tobacco.  The statute might have said, inter alia, that anyone “who knowingly uses coupons to purchase cigarettes” commits a crime. Under no plausible reading could a defendant then be acquitted because he did not know cigarettes are not “eligible food.” But in fact, that is exactly what § 2024(b)(1) does say; it just does not write it out longhand.

The Court’s opinion provides another illustration of the general point: someone who used food stamps to purchase groceries at inflated prices without realizing he was overcharged. I agree that such a person may not be convicted, but not for the reason given by the majority. The purchaser did not “knowingly” use the stamps in the proscribed manner, for he was unaware of the circumstances of the transaction that made it illegal.

[citations and footnote omitted.]

This ambiguity results from the distinction between “de dicto” and “de re” interpretations. That’s a topic that deserves a post of its own, but if you want a general introduction to the topic, you can go here or here.

This distinction is highly relevant to several areas of the law, but it’s one that judges, lawyers, and legal academics are generally unaware of. The few discussions of the distinction in a legal context include the following:

Jill C. Anderson, Just Semantics, The Lost Readings of the Americans with Disabilities Act, 117 Yale L.J. 992 (2008).

Robert E. Rodes, Jr., De Re and De Dicto, 73 Notre Dame L. Rev. 627 (1998) (SSRN).

Gideon Yaffe, “Trying to Kill the Dead: De Dicto and De Re Intention in Attempted Crimes” in Philosophical Foundations of Language in the Law (Andrei Marmor & Scott Soames, eds.2011) (ms.).

3 responses to “Syntactic ambiguity (Part 3 of Scalia and Garner on Statutory Interpretation)

  1. Hi, Neal! This was a great piece and I really enjoyed reading it. I was a bit confused by the update, though; is it possible you meant to indicate that even given the *high-attachment* structure, the statute is ambiguous? I suspect this might be the case only because Justice White begins with ” . . . accepting that ‘knowingly’ does extend through the sentence”, and such an interpretation is consistent with the adverb “knowingly” taking scope over the rest of the tree. I could be wrong, since I’ve always had trouble with the de dicto/re distinction.


  2. You’re right. I’ve corrected the update. Thanks for the catch.

  3. Pingback: RELINE – Annual seminar 2012 « From Words to Deeds: translation & the law

Leave a Reply