September 16, 2007

Language Log is strong

A small point, while I think of it, at the risk of seeming a tiny bit pedantic, concerning how to make reference to Language Log. You may have noticed, from other websites or our occasional direct quotations from them, that there are many people who write things like "I really enjoy the Language Log". To take a random example, this page from the website of the radio program Here and Now says The "Language Log" is an online hub where linguists trade thoughts on all aspects of language. And another site said (and we really are flattered and grateful): the website of record for die-hard language buffs is the Language Log, acknowledging in the following sentence: The Language Log, I admit, is not for the faint of heart (see it here). Many thanks for the praise; but for the non-faint of heart, it's "Language Log", not "the Language Log". If I may use the terminological distinction drawn in The Cambridge Grammar of the English Language in pp. 517ff, recently mentioned here), Language Log is a strong proper name, not a weak one.

What I mean by that is simply that syntactically it falls into line with Arizona rather than with (the) Azores; it is like Kansas City rather than (the) Old City (in Jerusalem): it takes no definite article. What you're enjoying is Language Log. Thank you.

Oh, and by the way, if you're wondering whether this is an entirely prescriptive judgment: no, it isn't really. True, I did just recommend using the correct name for our site, so I am making a point about what your future writing behavior should be, if you want to fall into line with the usual practice among those who know the correct name of our site. But the generalization I'm making reference to here is entirely descriptive. It's not like the Language Log has become widely used all over the place and I'm some sort of atavistic reactionary trying to deny that the language is the way it is. 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. That's a descriptive fact. At least at the moment it is. Things could be different in fifty years. But right now, I'm telling you that saying the Language Log is like saying the Iraq: it is just a mistake.

Being anti-prescriptivist doesn't mean refusing to admit that there can ever be such a thing as a linguistic mistake. It means being interested in what's a mistake and what isn't, rather than bull-headedly sticking with ideas of correctness that cannot possibly be correct.

I grant you that this is a rather subtle point. I have seen discussions in the past that have convinced me that many people cannot see any middle ground between two extremes: for them, it's "everything is correct" versus "nothing is relevant". That is, either (they think) there are no rules or standards at all and everything is as grammatical as everything else, or else the rules are the rules no matter what, and it doesn't matter one whit what the educated usage of native speakers and writers tells us about the language. I regard both extremes as utter whacked-out idiocy. Of course some sequences of words are correct Standard English and others are not. But of course that doesn't mean something can be ungrammatical in Standard English despite the fact that all educated speakers take it to be grammatical and normal usage.

In studying English (or any language) one can make the mistake of following usage too closely, and fail to distinguish sporadic errors of speech from systematic patterns of syntax; but one can also make the mistake of not following it closely enough, maintaining a quasi-superstitious belief in rules that don't really state generalizations about the structure of the language at all. Determining whether you are straying toward one of these methodological errors or the other is a matter that calls for deep reflection, close attention to large bodies of data, careful statement of tentatively proposed rules, a constant willingness to reconsider and revise proposals about what an accurate grammar for the language should say. It's not easy at all. It's the business that defines descriptive English grammar, and linguistics more generally. It's what those of us who write for Language Log actually do for a living. I am not in any way trying to suggest to you that's it's straightforward or easy to draw the distinctions in the right places, or to describe languages correctly. But I know I've got this little piece of English right.

Posted by Geoffrey K. Pullum at September 16, 2007 07:34 AM