July 03, 2006

Linguifying

I'm back in Santa Cruz after an exhausting two weeks of packing and moving from Cambridge (with no time to write, and barely enough just to read Language Log once a day as everybody should). I'm pleased to see that the flood of tedious and patronizing messages about my post on Daniel Gilbert has died down a bit. So let me try again to explain what struck me about Gilbert's claim ("Movies, theater, parties, travel — those are just a few of the English nouns that parents of young children quickly forget how to pronounce"). We need a new term for what is going on; although I don't in general think you can only grasp concepts that you have words for, I have learned to my cost that at least some people find it hard to get the hang of a new concept if they have no word for it, and Mark agrees that there is no term already in use. I therefore take the step of coining a new lexeme: linguify. It is a term relating to the writer's art, and in particular to journalism. Definition: To linguify a claim about things in the world is to take that claim and construct from it an entirely different claim that makes reference to the words or other linguistic items used to talk about those things, and then use the latter claim in a context where the former would be appropriate.

I note in passing that linguifying a claim is usually (but not always) done in such a way that the new claim is false instead of true, and it is often (but not necessarily) done with the intention of achieving a humorous effect.

You're going to need an example, of course. I have already given several on Language Log, but those times most of you weren't paying attention. So shut up, switch off your cell phones, put your comic books away, and listen. I'm giving you an example. It's from my earlier post "Bisexual chic". A writer named Alexis Long apparently wanted to say that bisexuality was increasingly being seen by mainstream news media as fashionable. But what he actually wrote (in an Australian newsletter for bisexuals) was:

It's difficult to find a piece of writing in the mainstream press which mentions the word 'bisexual' without finding that it is immediately followed by the word 'chic'.

Instead of talking about mainstream media attitudes, he linguified the claim, constructing a new statement about obligatory word adjacency in running text.

How do I know he didn't mean exactly what he said? Because he couldn't possibly have thought it was true. I searched on Google straight away to check on the linguified claim. And the number of occurrences of the word "bisexual" that Google found in mainstream news sources followed by the word "chic" was zero. And extending the search to all news recorded by Google News at the time, I also got zero, out of 984 news pages containing the word "bisexual".

The original claim might well be true; but the linguified claim was staggeringly, outrageous false. In fact it just couldn't be any more false: what Alexis said was difficult to find in an occurrence of "bisexual" (that it was not immediately followed by "chic") is actually found in 100% of the occurrences. And what Alexis says occurs almost all the time is instead never found at all.

But I am not particularly interested in the literal falsity of the linguified claim. What I'm interested in is why anyone ever linguifies a claim at all.

In the case of Daniel Gilbert's linguification, the linguified claim was obviously intended as humorous. He didn't (surely) think it it was true. Nor did I ever think he did, though apparently many of my readers imagined that I did). The linguified claim is certainly false; and it is certainly not an exaggeration of the underlying real claim in the sense that if it were true the underlying claim would be all the more true. But that's secondary. My interest is in why Gilbert would think linguifying was a good idea — why he thought it funnier, more interesting, or whatever.

Let me give another example. This one was sent by Rob Chametzky, apparently the only Language Log reader who saw clearly what I meant and sent me a new example (thank you, Rob; you truly hear the music; you have restored my faith in humanity). In The New York Times Magazine on Sunday, June 4, 2006, in an article called "Mass Natural", Michael Pollan wrote:

We have already seen what happens when the logic of the factory is applied to organic food production. The industrialization of organic agriculture, which Wal-Mart's involvement will only deepen, has already given us "organic feedlots" -- two words that I never thought would find their way into the same clause.

But he doesn't really mean he thought the adjective "organic" and the noun "feedlots" would never occur in the same clause. He couldn't possibly have thought that there would never be any attested sentences like this one:

Similarly, meat from organic grain-fed beef has the same nutritional profile as meat from the largest Kansas feedlot.
[http://www.eatwild.com/articles/whygrassfed.html]

The two underlined words are in the same clause. But Pollan never meant anything about clauses at all. He clearly meant that he never thought people would start regarding feedlots as being among the practices that characterize organic food production. His thought was about things, specifically farming practices; but he linguified it.

Here is one more. It's from Cynthia Gorney's article "Reversing Roe" in The New Yorker, June 26, 2006, page 50:

He seemed to have come to terms with the fact that a lot of literate people in his state now use his name and "sodomized virgin" in the same sentence.

Gorney is not really intending to tell us that South Dakota state senator Bill Napoli has come to terms with the co-occurrence of certain lexemes in certain uttered sentences. Her underlying claim is the following. A lot of people in South Dakota, when talking about Bill Napoli, make reference to his infamous appearance on PBS, where he was asked for an example of a situation where he would allow that an abortion should be legal even though giving birth would not kill the mother. What he said was that it would have to be a case of a virgin who had been saving her virginity for marriage but got brutalized, raped, and sodomized.

It is clear that Gorney does not think that people only mention these facts in terms that include "Napoli" and "sodomized virgin" within the boundaries of a single sentence. She has linguified her claim. (And by the way, the form of words "use X and Y in the same sentence" is fast becoming a snowclone of linguification.)

So, all you readers who went on and on at me about the Daniel Gilbert case — all you correspondents who told me finger-waggingly that nouns that parents of young children quickly forget how to pronounce "was obviously meant as a mildly humorous trope" (thank you M.S. of Madison, Wisc.), or that it "is just a figure of speech to indicate how rarely they would have cause to use those words" (thank you, Wangden), or that it is used "to indicate — jokingly — discomfort with or lack of knowledge of a concept" (thank you, Michele), or that "the underlying assumption may be that familiarity with a subject leads to (relative) expertise, which in turn implies that one is able to communicate effectively about the subject" (thank you, Matthew), or that it means parents "are not capable of enjoying such things as movies, theater, parties, travel- BECAUSE THEY HAVE NO WORDS FOR THEM" (thank you, Dr Pepper), or that perhaps "it isn't a rhetorical device, but a humor device: silliness" (thank you, Chele), or that "it's metaphor. Not an isolated metaphorical assertion ... but a whole metaphorical paradigm in the Lakoff & Johnson sense" (thank you, Rachael), or any of you others who wrote in: none of you have seen the point. I'm not failing to see the attempted humor, I have not become dyspeptic and humorless, I'm not taking it literally, I'm not incapable of understanding it, I am not unacquainted with the ideas of Benjamin Lee Whorf, I'm not ignorant of the (irrelevant) existence of hyperbole or metaphor or synechdoche.

What I wanted to draw attention to was simply the strange practice of publishing linguified claims: for example, saying the name X is invariably followed by the phrase Y when it isn't, or saying X is always accompanied by the qualifier Y when it isn't, and so on and so on. Why linguify? I have no idea. It just doesn't look like a good writing idea to me.

Posted by Geoffrey K. Pullum at July 3, 2006 06:49 PM