September 30, 2007

Language Log only pretty strong

A little while ago, Geoff Pullum firmly laid down the law about the name of this blog: it is "strong", that is, anarthrous, lacking the definite articke the:

One hundred percent of the references to Language Log by the people who actually write for Language Log say Language Log. None of us call the site the Language Log. And what made us arbiters of good taste? Well, we created Language Log, and coined its name. We coined it as a strong proper name. The sporadic use of the Language Log by others is a sign of imperfect learning.

But alert reader Tim Leonard has observed that one contributor to Language Log sometimes uses the arthrous version of the name; from my own website:

I am an occasional contributor to the Language Log ... ; to the American Dialect Society mailing list; and to Chris Waigl's eggcorn database.

What's going on here?  Something subtler than my having learned the name imperfectly.

Although I believe that my writing on Language Log itself uniformly treats the blog's name as anarthrous, occasionally in writing for other audiences (as above) I go arthrous.  Here are three more instances:

Several contributors to the Language Log do not share Garner's animus towards however, regarding it as an acceptable alternative to but.  (link)

... in postings to the American Dialect Society mailing list and the Language Log (a group blog on linguistic issues) ...  (link)

For fun, I did occasional postings to the American Dialect Society mailing list and to the Language Log.  (link)

In three of these, the pairing with the arthrous the American Dialect Society mailing list might have promoted my use of the, but I don't think I would necessarily have omitted the article otherwise.  In any case, my practice is variable, though with a very strong preference for anarthrousness.

I believe that my inclination to occasionally use the article stems from the generalization that proper names with singular count common nouns as their heads are mostly arthrous -- arthrousness is the default, though there are a number of islands of anarthrousness  -- and that this generalization is very strong for certain types of proper nouns, in particular names of organizations (the Federal Bureau of Investigation, the American Academy of Arts and Sciences), institutions (the Massachusetts Institute of Technology, the Frick Museum), and publications (the San Francisco Chronicle, the Daily Telegraph, the New York Observer).  The generalization makes sense, because singular count nouns require a determiner to be usable as argument NPs, and for a proper name this determiner would naturally be definite -- a possessive (Craig's List) or the definite article the.

A blog has something of the character of an organization, an institution, AND a publication, and the word log is a count noun, so we'd expect the proper name Language Log to be arthrous, like the Destin Log (a newspaper in Destin, Florida), the HobbySpace Log (a website for space enthusiasts), and the Cruise Log (on USA Today, with information on cruise travel).  There's then a good reason why outsiders (people not directly connected to the blog) often treat the blog name as arthrous, and why I sometimes go arthrous when talking to outsiders.

Further complexity.  Sometimes a normally arthrous name occurs in a truncated form, for quick reference or in a heading.  So there's a webpage with the heading Science Log (truncated), on which we find an arthrous full version:

The Science Log was compiled by Dee Davis, Science Officer for the USS Texas, International Federation of Trekkers.

An outsider might easily take the heading Language Log to be a truncated version of an arthrous name.

Meanwhile, there's genuine variation in article use, as on the Famosa Slough website, where this San Diego wetland is referred to both with the article and without:

The Famosa Slough is a 37-acre wetland between Ocean Beach and the San Diego Sports Arena area.

FRIENDS OF FAMOSA SLOUGH is a group of concerned citizens whose goal is to restore Famosa Slough as a natural wetland preserve.

An outsider might guess that Language Log was like Famosa Slough, with both variants acceptable.  The fact that references to Language Log ON Language Log are consistently anarthrous could easily escape a reader's notice.  Why should anyone be keeping track of such things?

Still further complexity.  Some proper names are anarthrous by local custom: though we'd expect the names to have an article, locals conventionally use the shorter version.  As a slogan: familiarity (sometimes) breeds anarthrousness.  Since I last wrote about (an)arthrousness, in acronyms and initialisms, people have been writing me about the initialism CIA, which would be expected to be arthrous on general principles, and is indeed so used by outsiders, but which has been widely reported to be anarthrous for those associated with the agency.  Nathan Austin notes that Harry Matthews's recent book (part memoir, part fiction) about the agency is entitled My Life in CIA and has the narrator saying (p. 66):

I asked Patrick if there was anything particularly useful he could pass on to me "about the CIA." "The first thing to remember is that nobody connected with the agency calls it the CIA.  It's plain CIA."

So even if you noticed the consistent use of Language Log on Language Log, you might think that that was just an insider thing, and that the other variant was allowable.

In fact, when Mark and Geoff started the blog, they chose an unexpectedly anarthrous name for it, and now Geoff would like to stipulate that only this usage is acceptable.  But I can't see how people should have been expected to LEARN this.

One further level of complexity.  Perhaps Mark and Geoff were thinking of Language Log not as an ordinary proper name but as a title -- like the title of a book, movie, newspaper column, musical group, etc.  If so, then the form of the proper NP is entirely up for grabs; as Geoff Pullum noted in The Great Eskimo Vocabulary Hoax, titles don't even have to be constituents (Geoff lists book titles like If on a Winter's Night a Traveler and Dancer from the Dance).  So you could name a rock group Statue of Liberty or Federal Bureau of Investigation even when these expressions would require an article when used as ordinary proper names.  So maybe Language Log is a title like these, in which case the person who chooses the title gets to stipulate its form.  The Stanford linguistics department's rock band is Dead Tongues; the Dead Tongues is just wrong -- not far off, but wrong.  Much like Geoff said about the Language Log above.  But then the (an)arthrousness of proper names would not be relevant to the question.

(There's a whole lot to be said about titles, and about articles in them, but I'll save that for another day.)

Posted by Arnold Zwicky at September 30, 2007 02:59 PM