Part XI

The Matter of Language:

There have been many variations on an idea from Modernism, that art is essentially a pastiche of materials, that what matters is the exciting interaction between elements, not whether there are plausible relationships between them.   In painting, an art looking for a mission since the invention of color photography by the Lumiere Brothers (an idea the writer has been pushing since the 1970's), collage, mixed media, jumbled subject matter, whether in Dada or in a dozen related movements, have been standard fare.   And given our human propensity for filling inkblots, clouds or stellar spaces with cartoon heroes and villains, as well as with a panoply of unrelated imaginings, one supposes this a fitting mission for the painter, given the fact that he or she is no longer required to create sustainable images of a given time, place or person.

This urge spilled over into poetry early on.  Why is less certain.  But perhaps it was because writers began to assume the language they used was as "obsolete" as a painted portrait.  We all like to be "modern" after all.  Beginning with the Imagists and with Eliot, who cared less about the coherence and intelligiblity of complex figures, such as metaphor, than about their appeal to the irrational,  by the 1960's this path had developed popularly into the "like, wow!" factor, where conjunctions of impossibilities replaced the plausible fictions of earlier metaphors.  The first generation's restraint kept most of their better work intelligible.

However, by the time of John Ashberry and Robert Bly, there was the additional, political claim that coherent, logical expression was in and of itself a tool of political oppression.  This would have been news to reporters following the Vietnam war (no plan for winning or losing, just a theory about impressing audiences with force that informed men whose cinematic counterpart was Dr. Strangelove, out to "save" a village, as modern counterparts decimate Colombia through surrogates in the "drug war.")  It would have also been news to the beneficiaries of science, where rational coherence had raised undeveloped world crop yields over 1000% while reducing damage to forest and trees, as well as provided women with the means to undo the sociobiological tyranny of unplanned motherhood.  Those two alone far outweighed the outlandish game of nuclear terror or the contributions of techno-hacks to the arms industry.  These seem not to have mattered to an idiot's view of war and empire, which proposed that only a rational, coherent people, using reasoned argument and calm persuasion would conduct empirial wars (and that, therefore, an artist must be irrational, incoherent, rave instead of argue reasonably, and choose methods of persuasion more familiar to fascist pep rallies than to salons, methods still standard fare among "the left" and its heirs).

But one can depend on a certain kind of stupidity sustaining itself.  All nature needs to maintain a species is that it function well enough to have sex and live long enough so that its children will know enough to survive.  This has not been a problem among the theorists and practitioners who suck the tit of political theory in the arts or lately, who campaign for Senate in New York State.

Another stream of thought that's undermined the old authority of language comes from computer science.  In a huge project run by a group of scientists, including Richard Feynman's son, there is an effort to combine in one memory all of one language's words, their likely combinations (this is possible; there are limiting factors), their etymology and meanings, and a functioning grammar.  This, it is claimed, will be able to meet Turing's test and carry on a conversation with a human being without the human being knowing he or she is talking to a computer.  There is a significant problem with the assumptions underpinning this project.   The gravest likely error is the presumption that the abstract symbols of language comprise an independent entity.  Independent from what?

It is more than arguable that words are fundamentally bound to nonverbal associations, themselves comprised of all varieties of memory, whether from the senses, the integration of the senses in images and observation, and the further integration into thought that weighs, judges and sends out messages to react, act or remain standing.  At the base level of existence, all of that nonverbal stuff is required to survive.  But for a hunter on a savannah working in concert with other hunters, all that's needed in language is signal shouting ("hey!" is about the same as "Look here," "I think this is what we're after," "I'm in trouble," "Good shot," "Great rib steak," "Watch out for the pterodactyl," "Great looking," "Nice cave," "Ugly baby!") a shout conveying meaning principally through the context in which it's used.  Once the signs evolve into specific words, there have to be direct associations with what's already, at least in part, in the brains of both speakers in a conversation.  And lacking human sense memory and all the structure that depends on it, what would a computer know about the scent of perfume in a candlelit cafe near the Seine on an evening in early spring?  Or about the moral considerations of armed intervention in Serbia?   It would be passing nothing but an imitation of speech, having no basis to say anything connected to the great unseen of language, the nonverbal data and associations that constitute most of  memory and sensation.

(Now, on a level that a computer can understand, such as tables of trig ratios, functions, and the symbolic logic of algrebra and the calculus, or of game theory, such as chess, they not only can talk but can offer original thought by combining what they know.   But lacking the facilities and memories of five senses, a computer cannot possibly understand most human utterance except by responding with conventionally associated words and meanings.  It wouldn't be much different from the original online psychotherapist Eliza.  Call it Eliza with a PhD.)

Accepting this, and it is impossible not to, requires a poet to give back authority to language that linguist and politicos have taken away.   Why?  Because language is as much of the flesh as the mind is, and as inseparable.  Further, it is arguable that most of the assaults on language have been motivated by poor exercise of its expressive power, whether in propaganda or in bad poetry.  Poets have no choice but invest not only in the power of language but in the energy they apply to making sure that what words they choose to employ alone or in combination actually convey something that someone else can understand.   Language has no other function, although it has vast spaces in which to play.  As the medium of poetry is language, poets cannot use it for other means without producing the kind of gibberish that some scientists and political theorists claim to be the best that language can do.

This is not a new idea.   When Auden's OED was found after his death, it was in tatters, the covers falling off, the binding broken apart, the end papers ripped, the pages dogeared, stained and sometimes mangled.   The editor of Oxford Press, on hearing this, was delighted: "this is exactly what we hope will happen to OED's." It meant to him that the book had been used to its very ruin.  Go through an Auden poem and try to find clumsy usage, or a gross error in meaning.  It's not likely that you'll find much.  Auden was relentless in pursuing definitions, etymology, and history of the usage of words (for which the OED is uniquely suited).  He is himself cited many times for usage in the new OED.  Part of this was great sport for Auden; the greater part, one assumes, was that Auden intended that what he had to say would be what people perceived it to be.   Given lurking, unconscious associations and the imprecision of many words, and the more difficult imprecisions of figurative language, this was a mission bound to partial failure, but it is one that most poets prior to the twentieth century took seriously.   To proceed as if language can have genuine meaning was their guideline.

Do the same.  Computers can do lots of tricks, including writing extravagantly weird, grammatically correct and absolutely meaningless sentences and paragraphs.  They can talk as if they know what they're saying (common to politicians as well, and perhaps more seductive).  But be assured that for the vast proportion of human experience computers and software have yet to take the bat off the rack, nevermind step up to the plate in a game they are aware of in front of people they know and sounds they comprehend.  For poets, such should be second nature, if we take care to use any source available, including that OED.   Language can have genuine meaning.  Poetry can convey an artist's abstraction of a person, an event or a scene to another.   Why else would we write it?