SCOTUS cites CGEL (Props to Justice Gorsuch and the Supreme Court library)

« previous post | next post »

(Cross-posted from LAWnLinguistics.)

When grammatical questions come up in legal cases, the lawyers and judges will want to support their arguments and analyses with citations to books about grammar. Most of the time, they cite books intended for a general audience, such as the McGraw-Hill Handbook of English Grammar and Usage, The Elements of Grammar, Strunk & White, and various books by Bryan Garner; and books intended for lawyers, such as The Grammar and Writing Handbook for Lawyers and (again) various books by Bryan Garner.

Unfortunately, none of those books gives an adequate description of English grammar, which is a subject that is much more complex than most people realize. If you're looking for a book that can speak about the subject reliably and with authority, the leading candidates are two books that are both known by the initials CGEL.

One is the Cambridge Grammar of the English Language (2002), whose main co-authors are Rodney Huddleston and Geoffrey Pullum. The other is the Comprehensive Grammar of the English Language (1985), by Randolph Quirk, Sidney Greenbaum, Geoffrey Leech, and Jan Svartvik. If you're dubious about my statement regarding the complexity of English grammar, you should note that both of these books are enormous: the text of Huddleston & Pullum exceeds 1,700 pages, and Quirk et al.'s exceeds 1,600 pages. If you ever have occasion to hold a copy of either book in your hands, be careful not to drop it on your toes.

Up until today, Huddleston & Pullum had been cited by U.S. courts 24 times, and Quirk et al., 12 times. The courts in question included the federal courts of appeals for the Fifth and Sixth Circuits and the supreme courts of Colorado, Minnesota, Oregon, Utah, and Washington. But not the U.S. Supreme Court.

That failure on the part of SCOTUS has now been remedied—as to one of the CGELs, anyway: Huddleston & Pullum. The case in question is Murphy v. Smith, which was decided today, in an opinion by Justice Gorsuch:

As always, we start with the specific statutory language in dispute. That language (again) says “a portion of the judgment (not to exceed 25 percent) shall be applied to satisfy the amount of attorney's fees awarded.” § 1997e(d)(2). And we think this much tells us a few things. First, the word “shall” usually creates a mandate, not a liberty, so the verb phrase “shall be applied” tells us that the district court has some nondiscretionary duty to perform….Second, immediately following the verb we find an infinitival phrase (“to satisfy the amount of attorney's fees awarded”) that specifies the purpose or aim of the verb's non-discretionary duty. Cf. R. Huddleston & G. Pullum, Cambridge Grammar of the English Language, ch. 8, §§ 1, 12.2, pp. 669, 729–730 (2002).

Notwithstanding this citation, the book didn't play any real role in the decision. I seriously doubt that Justice Gorsuch and his clerks felt that research was necessary before they could conclude that the function of the clause to satisfy the amount of attorney's fees awarded was to "specify the purpose or aim" of the duty imposed by the statute. Rather, the citation is there primarily to satisfy the norm of legal analysis that important statements in opinions should be supported by citations to appropriate authority.

Nevertheless, the citation is significant because it can't help but increase awareness of the book among lawyers (and judges, and law schools). Maybe some of them will even think about buying a copy. I can personally attest to the book's usefulness in dealing with language-related issues that arise in legal interpretation. (I'm sure that Quirk et al. contains much of the same information, but I'm less familiar with it and it's not what I reach for first.)

While Huddleston & Pullum isn't cheap, by any means ($250 new on Amazon, used for about $180), it's in the same ballpark as many practice-oriented books marketed to lawyers.

I wonder if the American Bar Association would be interested in negotiating with Cambridge University Press for a discounted bulk rate…



  1. Philip Taylor said,

    February 22, 2018 @ 4:42 pm

    Interesting. You write "citations to", I always write "citations of". Is this an <Am.E>/<Br.E> difference, or am I simply wrong in using "of" ? Incidentally, my copy of Quirk (case-bound in slip case) cost me around £65 new. I never fail to recommend it to those with a genuine interest in the English language.

    [NG: I can't speak to what the usage is in BrE, but I know that in the States, cite to, as in the court cited to Coyote v. Roadrunner, is common enough that Justice Scalia peeved about it—he thought the to didn't belong. That doesn't quite answer your question, since you asked about the noun citation and not the verb cite, but I don't have a story to tell about the verb.]

  2. Rubrick said,

    February 22, 2018 @ 5:43 pm

    Someone needs to do a study on the overrepresentation of the first name Geoffrey among linguists. Give me my p-values!

  3. Y said,

    February 22, 2018 @ 8:37 pm

    Is there an official legal definition or definitions of shall?

  4. Bartleby said,

    February 22, 2018 @ 9:51 pm

    Interesting that Gorsuch referred to an infinitival phrase, whereas I'm pretty sure CGEL would refer to it as an infinitival clause.

  5. Neal Goldfarb said,

    February 22, 2018 @ 11:31 pm

    I think that's right, but I'm prepared to cut them some slack this time.

  6. Thomas Rees said,

    February 23, 2018 @ 3:23 am

    Ah, but we’re textualists, and eager to peeve over the distinction between ‘phrase’ and ‘clause’.

  7. Guan Yang said,

    February 23, 2018 @ 10:16 am

    @Philip Taylor: I have mostly seen “citation to” or “cite to” used by lawyers, and rarely outside the legal profession.

  8. RfP said,

    February 23, 2018 @ 11:29 am

    I would like to put in a good word for A Student’s Introduction to English Grammar, which you might think of as the concise edition of the Cambridge grammar.

    Decades ago, during my short tenure as a linguistics dropout, I studiously avoided syntax. Now that I scriven for a livin’, it has slowly started to dawn on me that it might not hurt to grok the structure of this medium that I massage.

    I may someday graduate to the unabridged version, but in the meantime, this little nugget fits my backpack and my pocketbook. I don’t know how I lived without it.

  9. Andrew Usher said,

    February 23, 2018 @ 11:46 am

    Someone beat me to it – 'citation to' is legal, 'citation of' more standard. In either case the verb should be transitive.

  10. Gene Buckley said,

    February 23, 2018 @ 3:49 pm

    I wonder if Neil Gorsuch has any awareness of the irony here: The only reason he is a member of the Supreme Court is that Mitch McConnell refused to help Barack Obama fulfill his own constitutional mandate under the same interpretation of "shall". The wording of the Constitution does not apply this same mandate to the Senate, though I doubt that would have made any difference.

  11. Guy said,

    February 23, 2018 @ 5:55 pm

    This reminds me of a court case that cited a dictionary to support on argument that “on” must mean physically on top of, I remember thinking that a citation to the section of CGEL discussing “on” would have provided much more insightful analysis on the point.

  12. James Wimberley said,

    February 25, 2018 @ 5:20 am

    In exchange, Cambridge University could seelk a bulk discount from the American Bar Association. I assume they offer bourbon and tequila, as well as overpriced craft beers.

  13. Bartleby said,

    February 26, 2018 @ 12:26 am

    @Thomas Rees: Actually, I wasn't peeving. I just found it interesting that Gorsuch purported to use the CGEL but got the terminology wrong. Did he knowingly switch from clause to phrase, perhaps to be more in line with traditional grammatical terminology, or did he not actually use the CGEL?

  14. Graeme said,

    February 26, 2018 @ 3:18 am

    Y asked about 'official' definitions of 'shall'.

    Modern Act Interpretation Acts, aka Interpretation of Legislation Acts, and drafting manuals, try to avoid 'shall' as ambiguous. So they define 'may' in distinction to 'must'.

    More traditionally "shall' is defined – but with apparent redundant circularity – as being mandatory IF indicating a 'duty'. Which defers any ambiguity to resolving whether the context and purpose of the provision is to impose a duty, or a power more like a conjunction of 'should' but no breach of duty if not done and 'may be essential' to the point of breach of duty, depending on the facts facing that person.

RSS feed for comments on this post