the SNOOT fallacy

at issue is this article by David Foster Wallace (who i typically like and agree with). it's quite lengthy but well worth the read; Wallace is a very interesting writer and the topic is the politics of usage and dictionary-making. what could be more at home here on invented usage? there's a lot here i'd like to talk about and argue with, but i'll try to focus.

the article, which appeared in Harper's, i believe, amounts to a glowing review of Bryan Garner's A Dictionary of Modern American Usage. true to form, though, Wallace also gives us all the backstory including the 'seamy underbelly' of lexicography: the descriptivist/precriptivist divide. descriptivists would like a dictionary to reflect language the way people actually use it, insofar as possible, while prescriptivists tend to believe that a dictionary should tell us what's right and how to use words properly. i found it somewhat surprising that Wallace himself comes down firmly on the prescriptive side, proudly calling himself a SNOOT. that's a nice, cute term for something i've called a 'grammar nazi' or a 'big fat jerk' elsewhere on this blog. i'm sure he has his reasons for capitalizing it (and for once mispelling it as 'SNOT,' i'd like to add).

his precious reasons. wallace, garner, and snoots of their ilk are fond of reasoning about language. and the basic premise of 99% of logical arguments about language is... drumroll please... "the purpose of language is communication." in fact, this maxim is taken so keenly for granted that it is embedded only in parentheticals on linguistics departments' webpages. (Brown's for instance: How does the function of language (to communicate) interact with its structure?) so of course wallace feels justified in defending at least the prescriptivist rules that aid 'clarity and precision.'

i'd like to submit (usage liberal that i am) that language has more purposes than just communication. i even believe it goes beyond wallace's observation that the diction/style/accent we use communicates something about us. language is used to confuse, to distract, to entertain, to kill time, to remember, to make art, to perform ceremonies, all of which could be considered communicative under my usually broad definition... but beyond even that, the ways people judge each other based on language use are PART OF LANGUAGE ITSELF. its purpose is also to divide, include, grade and judge. these functions determine who gets listened to, and in extreme cases, who gets listened to is a matter of life and death.

wallace quotes a snoot friend of his: "listening to most people's English feels like watching somebody use a Stradivarius to pound nails." this analogy reveals more than a few things about the prescriptivist (snoot) perspective. first, it reveals that they consider language an object. an object that we all approach the same way from the outside and use for a determined purpose. they believe there is a right function and a wrong function for language, that that function is essentially benign and beautiful. wallace acts as though prejudice, awkwardness, and judgmentality are faults of people, and not built into language in any way.

wallace writes,
These are tense linguistic times. Blame it on Heisenbergian Uncertainty or postmodern relativism or Image Over Substance or the ubiquity, of advertising and P.R. or the rise of Identity Politics or whatever you will — we live in an era of terrible preoccupation with presentation and interpretation. In rhetorical terms, certain long-held distinctions between the Ethical Appeal, Logical Appeal (= an argument's plausibility or soundness), and Pathetic Appeal (= an argument's emotional impact) have now pretty much collapsed — or rather the different sorts of Appeals now affect and are affected by one another in ways that make it almost impossible to advance an argument on "reason" alone.
and i'd like to object. postmodernism is NOT relativism, though this is a common misconception. postmodernism never says 'everything is the same. it doesn't matter how we speak because everything is as good as everything else.' it says 'the distinctions we make in language are arbitrary, but they have to be made. we have to make the choice to have an official academic language, or to read a certain text a certain way. it's just VERY important that we consider why (and from what cultural standpoint) we're making those decisions.' wallace's distinciton between different rhetorical strategies is an old one--one of the oldest, in fact. but language has never been transparent. there has never been a 'logical' argument that was not also rhetorical and emotional. EVERY statement that has content also has form and also has a speaker who speaks from a particular perspective and whose words therefore have a particular emotional impact. as soon as a word enters the language (or is uttered... not the same thing, i guess) it's encoded with information about the social and mental status of those who use it. logic never has been and never will be free of language.

wallace finally praises garner's dictionary of modern usage because garner himself seems not to speak from a normal human position.
It's like he's so bland he's barely there. E.g., as this reviewer was finishing the book's final entry, it struck me that I had no idea whether Bryan Garner was black or white, gay or straight, Democrat or Dittohead. What was even more striking was that I hadn't once wondered about any of this up to now; something about Garner's lexical persona kept me ever from asking where the guy was coming from or what particular agendas or ideologies were informing what he had admitted right up front were "value judgments."
that is, according to wallace, the perfect person to judge language is barely a person at all. garner lays bare his assumptions, but doesn't admit the sociocultural ground he makes them from (although i'll lay 20-to-1 he's white). this is perfection in scientific distance and objectivity. maybe it's a great way to write a text book, but it misses something fundamentally beautiful about what language is: inconsistent and ever-changing, as anything determined by everyone in the world all at once has to be.


Marc André Bélanger said...

I agree with your analysis of snootiness, but must beg to differ about dividing, including, grading and judging being purposes of language. To me, this is like saying that, since I can judge the handiness of someone through their use of a hammer (and by the wear on the hammer) one of the purposes of hammers is to judge handiness. Also, just because I can use a hammer as counterweight, doorstop or smurf pedestal, doesn’t mean these are purposes of hammers.
We have to make the difference between use and purpose.

Cristi said...

thanks for the comment, marc!

i understand your distinction, but i would still argue that those functions (judgment, etc) are a necessary and unavoidable part of language use. the difference between language and a hammer is that language always occurs under observation. it's an essentially social thing. to me, it's hard to distinguish between an essential part of use and a purpose. (although i was being a bit dramatic in my post!)

Seb said...

I'm glad Marc said it first, but I wanted to second his comment. The danger of all "purpose" talk is that it involves value judgments pretending to be facts about nature: it is wrong, it is supposed, to go against the "purpose" of something, but the purpose is something intrinsic to the thing itself.

This sort of teleology is how Aristotle defended slavery ("Some people were just meant to be slaves.")

It's bogus. We need to distinguish between

(a) what something does--what are its causal properties and relationships with other things,


(b) our intentions for something--what we want it to do, or consciously employ it for, and our intentions for something when we create it.

If language is going to be said to have a purpose at all (which I think is a little dubious), it will have to get that purpose through the second kind of stuff (b). And that stuff is going to be contextualized to individual and social cases.

On the other hand, there probably are solid facts about the stuff in (a)--truths about what language does, no matter what we want it to do.

I think it's fair to say that in most cases where language is intentionally employed for something, it is intentionally employed for the purpose of communication. Other things--like dividing, grading, and judging--are, I think, rarely if ever intended. Those seem to often be the results of historical accident. For example, if I try to learn French and have a crappy accent because I can't hack the phonotactics of it, and then somebody in France judges me because I sound like an idiot, that's just something that happened to us. It's like I dropped a hammer and it hit me in the foot. It's not the purpose of a hammer to hit me in the foot--it's just that the world and hammers work in such a way that in some circumstances, I get hit in the foot.

That took longer to say than I meant it to.

Anyway, I wanted to say that I'm not a SNOOT, and wholeheartedly agree that natural language dictionaries ought to be descriptive and not hamper natural language's growth.

HOWEVER, I do think that there are some unnatural languages--technical languages, for example--that really are designed for communication, and for communication alone. While they may have other effects in the world, they really have been constructed for a purpose. People made them to hit nails, and no matter how good they may be at being doorstops, a presciptive dictionary--let's call it a glossary--established for the purpose of maintaining conventions, could be very important to maintaining that technical langugae's usefulness as a tool to achieve its intended purpose.


For the sake of being predictable, I don't know if I agree with your section of logic. For example, when you say:

logic never has been and never will be free of language.

I think this depends a lot on what you mean by language. Certainly there is logic that is independent of natural, oral or written language. If you extend language to "all representations," then the case gets easier, but that doesn't seem to be how "language" is being used by, say, Wallace, and I think there are useful distinctions to be made between the two meanings of the term.

Anonymous said...

you can have your logic in an ivory tower if you wish, but then what's the point?

Marc André Bélanger said...

Cristi: "it's hard to distinguish between an essential part of use and a purpose." Ah, but just because something is hard to do, don’t mean we shouldn’t try…

Seb: "I think it's fair to say that in most cases where language is intentionally employed for something, it is intentionally employed for the purpose of communication." Actually, I’d say for the purpose of expressing meaning. That is what it does first and foremost, everything else is (b). And I think we should be talking more of a prime function than a purpose.

As for prescriptivism, it doesn’t have to be technical or specialized languages; if I want to be understood by people for widely different places and backgrounds, I have to tone down my particular dialect (especially in French) and adopt a more normative “international” (albeit not native to anyone) variety, which would be prescribed by dictionaries and grammar books. That is actually what we learn at school: the normative language.

Seb said...

Marc: I agree with your analysis. Yes, "express meaning" does cover a broader but still appropriate set of functions than mere "communication." I like that a lot.

To the anonymous:

If you were directing that comment at me, when I was thinking about logic as used without natural written or oral language, I was thinking about things like formal mathematics, which I would argue uses logic either in a non-linguistic way (using a narrow definition of language) or in a very specifically purposeful way (using a broad definition of language) such that ethical and pathetical (is that the word?) effects are deliberately minimized.

"but then what's the point?"

Well, math and other applications of formal logic have lots of practical applications that help us do wonderful things. The internet, for example, required the work of thousands of engineers and theorists over time. That work--that logical progress--wouldn't have been possible without rigorous, logical, arguably nonlinguistic thinking.

Now we've gotten to the point where we can have computers do a lot of logical work for us. Sometimes we do that by making machines that can passionlessly perform precisely the mechanical steps that human beings used to do without accidentally dropping the emotional hammer on their feet, so to speak.

Again, this has lots of important consequences. Computational biology, for example. Which contributes now to biomedical research. So "ivory tower" logic, among other things, saves a lot of lives.

That's the point.

Cristi said...

thanks a ton to all commenters. sorry for my absence from the discussion, I just had my wisdom teeth out!

i guess i've kind of worked myself into a corner here: part of the original fallacy i wanted to point out was the fact that i don't think language is actually a 'tool' in any sense. at least not in the hammer or stradivarius sense. that is, we can't pick it up and use it and examine it from the outside. it's not designed for any purpose at all.

but i've left myself defending my 'inclusion/exclusion etc. are the purposes (or inherent uses) of language' position, which contradicts what i originally wanted to say. i guess what i mean is that we build terms as we use them. by even speaking the same language we constantly reinforce a certain way of using terms (even if that doesn't mean agreeing on their definitions or related concepts!). and by speaking the same language we exclude those who don't speak it. and i believe that these things are always occurring whenever we use language... whether they can be called its 'purpose' or 'function' is something i don't think we'll ever know for certain.

i guess a lot of my beliefs about language are somewhat faith-based. we're born into it, we die before we see the endpoints of the changes that happen to it (if there are endpoints!), we're educated in it by certain people (not ourselves!), so that we never have pure unmediated access to the whole thing as an object.

i feel like i'm way off on a tangent here... sorry to change the subject. i'll post again soon. thanks again for the interesting discussion!

Related Posts with Thumbnails