I needed to reread an essay DFW published in Harper’s called “Tense Present.” It’s an essay about authority, grammar, rhetoric, democracy, and other stuff. It is alternately hugely amazing and hugely upsetting. The reason I bring it up is that Wallace grounds the central thrust of his argument on Wittgenstein’s Philosophical Investigations. Wallace basically summarizes the PI and then pivots out to talk about the argument’s potential relevance for grammar and politics.
The formatting here is weird; if you click the links for the footnotes, they will take you to another site; or I’ve copied the text with footnotes below so you can just scroll up and down if you like. The man loved footnotes, what can I say. Footnote 23 is where the summary of PI resides. Excerpt:
Even if, as a thought experiment, we assume a kind of nineteenth-century scientific realism-in which, even though some scientists’ interpretations of natural phenomena might be biased the natural phenomena themselves can be supposed to exist wholly independent of either observation or interpretation — no such realist supposition can be made about “language behavior,” because this behavior is both human and fundamentally normative. To understand this, you have only to accept the proposition that language is by its very nature public — i.e., that there can be no such thing as a Private Language and then to observe the way Methodological Descriptivists seem either ignorant of this fact or oblivious to its consequences, as in for example one Charles Fries’s introduction to an epigone of Webster’s Third called The American College Dictionary:A dictionary can be an "authority" only in the sense in which a book of chemistry or of physics or of botany can be an "authority": by the accuracy and the completeness of its record of the observed facts of the field examined, in accord with the latest principles and techniques of the particular science.
This is so stupid it practically drools. An “authoritative” physics text presents the results of physicists’ observations and physicists’ theories about those observations. If a physics textbook operated on Descriptivist principles, the fact that some Americans believe that electricity flows better downhill (based on the observed fact that power lines tend to run high above the homes they serve) would require the Electricity Flows Better Downhill Theory to be included as a “valid” theory in the textbook — just as, for Dr. Fries, if some Americans use infer for imply, the use becomes an ipso facto “valid” part of the language. Structural linguists like Gove and Fries are not, finally, scientists but census-takers who happen to misconstrue the importance of “observed facts.” It isn’t scientific phenomena they’re tabulating but rather a set of human behaviors, and a lot of human behaviors are — to be blunt — moronic. Try, for instance, to imagine an “authoritative” ethics textbook whose principles were based on what most people actually do.
Norm-wise, let’s keep in mind that language didn’t come into being because our hairy ancestors were sitting around the veldt with nothing better to do. Language was invented to serve certain specific purposes: “That mushroom is poisonous”; “Knock these two rocks together and you can start a fire”; “This shelter is mine!” And so on. Clearly, as linguistic communities evolve over time, they discover that some ways of using language are “better” than others — meaning better with respect to the community’s purposes. If we assume that one such purpose might be communicating which kinds of food are safe to eat, then you can see how, for example, a misplaced modifier might violate an important norm:
“People who eat that kind of mushroom often get sick” confuses the recipient about whether he’ll get sick only if he eats the mushroom frequently or whether he stands a good chance of getting sick the very first time he eats it. In other words, the community has a vested practical interest in excluding this kind of misplaced modifier from acceptable usage; and even if a certain percentage of tribesmen screw up and use them, this still doesn’t make m.m.’s a good idea.
Maybe now the analogy between usage and ethics is clearer. Just because people sometimes lie, cheat on their taxes, or scream at their kids, this doesn’t mean that they think those things are “good.” The whole point of norms is to help us evaluate our actions (including utterances) according to what we as a community have decided our real interests and purposes are. Granted, this analysis is oversimplified; in practice it’s incredibly hard to arrive at norms and to keep them at least minimally fair or sometimes even to agree on what they are (q.v. today’s Culture Wars). But the Descriptivists’ assumption that all usage norms are arbitrary and dispensable leads to — well, have a mushroom.
The connotations of arbitrary here are tricky, though, and this sort of segues into the second argument Descriptivists make. There is a sense in which specific linguistic conventions are arbitrary. For instance, there’s no particular metaphysical reason why our word for a four-legged mammal that gives milk and goes Moo is cow and not, say, prtlmpf. The uptown phrase for this is “the arbitrariness of the linguistic sign,” and it’s used, along with certain principles of cognitive science and generative grammar, in a more philosophically sophisticated version of Descriptivism that holds the conventions of SWE to be more like the niceties of fashion than like actual norms. This “Philosophical Descriptivism” doesn’t care much about dictionaries or method; its target is the standard SNOOT claim supra — that prescriptive rules have their ultimate justification in the community’s need to make its language meaningful.
22. (“EVIDENCE OF CANCER LINK REFUTED BY TOBACCO INSTITUTE RESEARCHERS”)
23. This proposition is in fact true, as is interpolatively demonstrated below, and although the demonstration is extremely persuasive it is also, as you can see from the size of this FN, lengthy and involved and rather, umm, dense, so that again you’d probably be better off simply granting the truth of the proposition and forging on with the main text.
INTERPOLATIVE DEMONSTRATION OF THE FACT THAT THERE IS NO SUCH THING AS A PRIVATE LANGUAGE
It’s sometimes tempting to imagine that there can be such a things as Private Language. Many of us are prone to lay-philosophising about the weird privacy of our own mental states, for example, and from the fact that when my knee hurts only I can feel it, it’s tempting to conclude that for me the word pain has a very subjective internal meaning that only I can truly understand. This line of thinking is sort of like the adolescent pot-smoker’s terror that his own inner experience is both private and unverifiable, a syndrome that is techinically known as Cannabalic Solipsism. Eating ChipsAhoy! and staring very intently at the television’s network PGA event, for instance, the adolescent potsmoker is struck by ghastly possibility that, e.g., what he sees as the color green and what other people call “the color green” may in fact not be the same color experiences at all The fact that both he and someone else call Pebble Beach’s fairways green and a stoplight’s GO signal green appears to guarantee only that there is a similar consistency in their color experience of fairways and GO lights, not that the actual subjective quality of those color experiences is the same; it could be that what the ad. pot-smoker experiences as green everyone else actually experiences as blue, and what we “mean” by the Word blue is what he “means” by green, etc., etc., until the Whole line of thinking gets so vexed and exhausting that the a.p.-s, ends up slumped crumb-strewn and paralyzed in his chair.
The point here is that the idea of a Private Language, like Private Colors and most of the other solipsistic conceits with which this particular reviewer has at various times been afflicted, is both deluded and demonstrably false.In the case of Private Language, the delusion is usually based on the belief that a word such as pain has the meaning it does because it is somehow “connected” to a feeling in my knee. But as Mr. L. Wittgenstein’s Philosophical Investigations proved in the 1950s, words actually have the meanings they do because of certain rules and verification tests that are imposed on us from outside our own subjectivities, viz., by the community in which we have to get along and communicate with other people. Wittgenstein’s argument, which is admittedly very complex and gnomic and opaque, basically centers on the fact that a word like “pain” means what it does for me because of the way the community I’m part of has tacitly agreed to use “pain”.
If you’re thinking that all this fuss is not only abstract but also pretty irrelevant to the Usage Wars or to anything you have any real interest in at all, you are very much mistaken. If words’ meanings depend on transpersonal rules and these rules on community consensus, language is not only conceptually non-Private but also irreducibly public, political, and ideological. This means that questions about our national consensus on grammar and usage arc actually bound up with every last social issue that millennial America’s about — class, race, gender, morality, tolerance, pluralism, cohesion, equality, fairness, money: You name it.
24. Norms, after all, are just practices people have agreed on as optimal ways of doing things for certain purposes. They’re not laws, but they’re not laissez-faire, either.