4.22.2006

Son of 'On Like Usage'

Once upon a time, I posted this classic: On Like Usage, which briefly shot Invented Usage to the top of search engine results everywhere and sparked a most interesting debate about the meaning and usage of "like."

As I draw close to the conclusion of my first actual syntax class, I have a few more tools with which to discuss 'like' and why it illustrates a challenge to the contemporary study of linguistics.

Categorial Grammar (CG) operates on the assumption that we can tell what type of word we're dealing with by where it's found in relation to other words. That is to say, we know 'ran' is a certain type of verb because it follows a noun phrase and provides a sentence. Another way to think of these relationships is as input-output functions. 'Gave' requires three noun phrases as inputs, and then provides a sentence (or, more accurately, a truth value) as output.

in the post linked above, I illustrated the myriad ways 'like' can be distributed throughout a sentence. it can modify nouns, adjectives prepositions, verbs... pretty much anything, now that i think of it. most linguists probably wouldn't consider this a problem, since they would claim that 'like' does not contribute to truth conditions. that is, that 'she went to the store' and 'she went, like, to the store' are true in exactly the same instances. if that's the case, 'like' doesn't contribute any meaning, and doesn't have to be accounted for in the grammar. linguistics adheres to a strict separation between language as a formal system (which syntax can describe) and language as it is used by people.

but 'like' DOES change the meaning, and in some cases the truth conditions (i don't believe they're the same thing) of sentences. it's one of many reasons why I think language use cannot be separated from language itself.

while syntax is interesting as an attempt to understand the patterns that arise from language use, the way it is actually implemented seems to forget that purpose. all linguists can do, at this point, is describe (to a limited degree) an idealized language that no one actually speaks.

4 comments:

Marc André Bélanger said...

I like your example of “gave” as requiring three noun phrases as inputs. Because it illustrates that this kind of statement should never be categorical. “Gave” usually needs three inputs, but one could easily say: “I gave and gave, and never got anything in return.” By playing with the grammatical “requirements” of a word, we can gain expressivity.

“linguistics [actually, not all linguistics, but I guess that which you are being taught] adheres to a strict separation between language as a formal system (which syntax can describe) and language as it is used by people” and in that it is wrong. Language is first and foremost something used by people. And if the formal system, i.e. the theoretical model, does not conform to use, than it is wrong. The linguistics school I was taught in always tries to start with observation (“We can explain to the extent that we have understood. We can understand to the extent that we have observed”) Actually, it thrives on those example that undermine current conceptions, the small percentage of cases that don’t fit. I use to say “it’s the 1% that kills you” (referring to a show where one of the characters had to go through a combat simulation, and avoided 99% of the gunshots).

As for like, I don’t think it so much modifies phrases or nouns, but rather the utterance. It does not add grammatical or semantic information to a phrase; it provides back-channelling, a way to connect with the listener , a pause to make sure that they’re following, that they’re not objecting.

This is not the same use as “He was like, ‘What do you mean?’ and all,” where it serves as a narrative way to present the quote and attitude of the person.

Cristi said...

great comment, Marc.

i like your analysis of 'like', and i think we posted something similar on this blog long ago. but the more i think about it, the less i'm convinced that it really is a 'checking' mechanism. it's become too fluid a part of speech to actually serve this purpose anymore (assuming it once did).

and i think in my example it does seem to provide some truth-conditional content. 'She went, like, to the store', seems to protect the speaker from asserting that 'she' went only to the store, or that she went to the store at all. 'she went, like, to the store' might be considered true (or true enough) if 'she' went to the dry cleaners and the vet, but not the store.

Marc André Bélanger said...

"to protect the speaker from asserting" nicely put. Like adds vagueness. But at the same time, the 'checking' mechanism can become a way to (rhetorically) give the listener the opportunity to agree with the 'true enough' factor. Like saying "She went to the store, right?" where "right?" can mean "do you follow me" or "if you agree to call it that".

gwodder said...

HOLY SHIT CRISTI

My heart swells, for serious,

because we're long lost siblings after that little essay,

I never met ANYONE who noticed that the interjection 'like' carried any greater import than a simple pause or "Ummmm" ... . . . .

SIBLINGS.

Now, knowing this, we can watch as the children of this generation struggle with the realization that the language they were given isn't NEARLY expressive enough to communicate the depth of their experience.

Like a message header meant to communicate that "what I'm saying is not what I'm talking about... it's just LIKE what I'm talking about."

Without common ground we cannot employ metaphor and without metaphor we can't deploy language and this is what every young American is unconsciously saying all of the time : STOP trying to be so scientific and precise all the time and LET US HAVE SOME SLACK

Related Posts with Thumbnails