September 24, 2007

Mukasey weighs in on clear writing and light beer

Newspapers have been running profiles of Judge Michael B. Mukasey, President Bush's nominee to succeed Alberto Gonzales as Attorney General, and he is revealed to have a number of surprising qualities, at least compared to some of Bush's past choices for Cabinet positions. In the Washington Post, a federal public defender describes Mukasey as "very sharp, very focused," adding, "It was interesting to argue before him because he was interested in ideas and language." The New York Times divulges that he has a framed photograph of George Orwell in his chambers. "He is a particular idol of mine for his clear writing and complete disdain for cant," Mukasey explained. "I try to recognize when some spongy abstraction is trying to cover up an excuse for thought or analysis."

Now, not everyone at Language Log Plaza is a fan of Orwell's writings on language-related matters. Geoff Pullum, for one, has vilified Orwell's famous 1946 essay, "Politics and the English Language," agreeing with Stanley Fish's assessment of it as "turgid, self-righteous and philosophically hopeless." But regardless of how you feel about Orwell as an icon of linguistic clarity, it's refreshing to have a prospective Attorney General who is "interested in ideas and language." In this regard, the most telling detail in these profiles of Mukasey is a bit of Latin wordplay buried in a judicial footnote, concerning the taste of competing brands of light beer.

In the 1992 case (Coors Brewing Co. v. Anheuser-Busch Co. 802 F.Supp. 965), Coors argued that an Anheuser-Busch ad campaign for Natural Light beer made unfair and misleading comparisons to the brewing process of Coors Light. In his U.S. District Court opinion, Mukasey rejected the complaint from Coors, but in a footnote he refused to take a stand on the effect of pasteurization on the taste of beer:

The parties have disputed vigorously whether the taste of beer, unlike the taste of milk, is adversely affected by pasteurization. Coors says it is; Anheuser-Busch say it isn't. They have also disputed, in this forum and elsewhere, other features of their products and their advertising, defendants having gone so far as to accuse Coors of "ad hominem attacks on Natural Light." (Def. Mem. at 13) However, de gustibus cerevesiae non scit lex.

The New York Times glosses Mukasey's Latin phrase as, "The law takes no account of taste in weak beer," while the Washington Post translates it more literally: "Concerning the taste of beer law knows nothing." What the papers don't mention is that Mukasey was simultaneously playing on two well-known Latin expressions:

  • De gustibus non est disputandum.
    "There is no accounting for tastes."
    (Lit.: "Concerning taste there is no disputing.")
  • De minimis non curat lex.
    "The law does not care about trifling matters."

The latter expression is the basis for the legal principle known as de minimis, used by courts to avoid passing judgment on matters deemed unworthy of judicial attention. Mukasey managed to dismiss the pasteurization question as de minimis, while at the same time making a clever parallel to the de gustibus sentiment of "to each his own" — surely a wise tactic when it comes to assessing the relative merits of light beer.

(An etymological note: Latin cerevisia is the root of Spanish cerveza, Portuguese cerveja, and Catalan cervesa, all via Old French cervoise. It is a medieval Latin word of Gaulish origin and honors Ceres, goddess of the harvest. Gaulish was a Celtic language, which presumably explains the connection to another cognate, Welsh cwrw 'beer, ale.' Most of the other western European languages have inherited terms like English beer and German bier, from a Germanic root that probably derives from Latin bibere 'to drink.')

[Update #1: Conrad Roth emails to question my assumption that Welsh cwrw shares the Celtic origin of Latin cerevisia. According to an online dictionary, the derivation is as follows:

ETYMOLOGY: Welsh cwrw < cwrwf < *cwryf < *cwrf < British *korm
From the same British root: Cornish korev (= beer), Breton koref (= beer)
From the same Celtic root: Irish coirm (= beer; drinking party; concert)
Cf Latin cremor (= broth, thick juice), also used as a technical term in Enlish - Webster 1913 cremor = cream; a substance resembling cream; yeast; scum); Greek kourmi (Harry Thurston Peck, Harpers Dictionary of Classical Antiquities (1898); entry for Perseus: "The beer or barley-wine of Crete was known as korma or kourmi.").

The University of Wales' "Lexicon of the Celtic World" (see PDF link for "Examples from the Celtic Core Vocabulary") agrees with this, tracing kwrw back to Proto-Celtic *kurmi- (and PIE *kor-m). That is evidently distinct from the etymon for Ceres, cereal, etc. See also this entry in "An etymological lexicon of Proto-Celtic."]

[Update #2: Mischa Hooker writes:

I'd suggest, however, that cerevisia *is* likely related to the Welsh cwrw, since it's apparently a borrowing from Gaulish. The connection of cerevisia with Ceres is what I would question. [Note the apparent pattern m > v as argued by L. H. Gray, "Mutation in Gaulish," Language 20.4 (1944) pp. 227-8 (mutation no. 17 in the list), citing this set of cognates among some others.]

Serves me right for wading into the muddy waters of Celtic etymology!]

[Update #3: Beer historian Martyn Cornell further discounts the connection between cerevisia and Ceres:

This is Latin folk etymology, based on the Romans' idea that all barbarian words had to be derived from Latin originals. The name of the goddess Ceres has nothing at all to do with the word cerevisia (how would it?), which comes from a variant of the Gaulish word curmi, with e-vocalisation and the same change of m to v that happened with other Celtic words, such as the names of the British kingdoms Dumnonia (Devon) and Demetia (Dyfed). Curmi seems to be linked etymologically with the same IE root that gave the Latin word cremo, to burn or boil, in the same way that so many other brewing words have roots in words meaning bubbling or boiling.
The Welsh version of the Common Celtic word curmi underwent the same m-to-v change, to become coref (IIRC) in medieval Welsh, losing the f (single f being pronounced v in Welsh, of course) to turn into cwrw. In Cornish and Breton the f stayed, giving coref in the former and coreff in the latter. The old Irish cognate was coirm, later replaced by lionn, which originally just meant "drink". Cerevesia (or cervisia, or other variants) isn't a medieval Latin word either — it's found from at least the 1st century AD. And modern French still has the word cervoise, with the specialised meaning "unhopped ale", as against biere, the name of the hopped drink — a distinction that has been, sadly, lost in English these past three or four centuries.

Posted by Benjamin Zimmer at September 24, 2007 06:08 AM