We can say, I think, that writing is sometimes as difficult as it is because it strains our vision, as if with our inner eye—our insight—we are trying to see the meaning, the significance of something we come up against, an object, an event, an imaginative world of our own making. But the just-right word or phrase or sentence we may eventually find depends just upon that vision-power, and good writers who command it may show us what it can produce.

The Nobel Laureate Odysseus Elytis is considered one of the three major modern Greek poets of the last century. He said of himself that he was a poet only, not a writer of prose, but the collection of what essays he did write, entitled Open Papers, shows what visionary powers he possessed to lay bare the significance of small fragments of life around him. In an essay there entitled “First Things First,” he speaks about his poetic work, and says ironically and outright that “my powers of observation are largely absent, and every attempt at description bores me to death.” If we can’t call it observation, imagine, then, how lucid a perception this passage of his a few pages later illustrates:

The way a bird leans to one side, or the yogurt vendor calls a little louder on the downhill at dusk, or the way an odor of burnt grass billows through the open window (from where?), the subtlest, most invisible marks assume their entire meaning, as though their only mission was to convince me that at any moment the beloved arrives.

This is a poet speaking for a time in prose. Prose wants to describe the husk, but poetry wants the seed. We could spend a good while trying to understand the sophisticated grammar and composition of that one sentence, but too much of that would deny Elytis what he wants to give us, his vision:

Here is the smallest canvas where my life’s ideogram can be embroidered; if you think it worth examining, it would be enough to yield a space whose meaning lies not in the natural elements that compose it but in their extensions and correlations inside us to our farthest limits, so that, in order to become easily read and understood, the entire significance of the vision is finally concentrated in the psychic clarity it presupposes and needs.

Vision in a psychic clarity. English has two difficult adjectives which together sum up, I believe, what Elytis is saying here. Something is perspicuous which is clear and lucid because of its precise constrction; and someone or something (like a poem) which is so discerning and keen that it sees through such a perspicuous structure is perspicacious. Poet and reader, artist and spectator must be these qualities to each other, in order to, as Elytis says again, “recompose the world, literally and metaphorically, so that the more its desires are actualized the more they contribute to the materialization of a Good acceptable to all humans.” Can we think of anything to be prized more at the moment than a philological way to materialize the Good?

Elytis won the Nobel Prize in Literature in 1979. His acceptance speech, as well as the Nobel Foundation’s award speech, are very much worth the while to read. Both are available online.

***

One try, one draft—no, that never works in writing. It is one of the great mysteries of our already mysterious enough human mind why putting words to the ideas we have is so often so difficult. We try to get past that obstacle with the get-it-done-with-one method, but the first and final draft that method produces never seems to be worth the effort.

One explanation for all that unproductive work is that we’re trying to do too much at once. There seem to be vast tracks of land behind our eyes as well as in front of them. We’re watching, waiting for an animal thought to cross our path, but most of the time thoughts travel in packs. And how do we catch a klatch of roaming thoughts? We usually think a compound sentence will trip them up, or even better, a complex sentence will dig a ditch deep enough for all of them to fall at once. But even if we snare them, we’ve got another problem on our hands: a throng of unruly ideas, one nipping at the next.

What to do? Thin out the herd before setting the trap. Here’s what that looks like. You’re on vacation, you’ve rented a small cottage on an island off the coast, and one lazy afternoon you decide to write a paragraph describing the landscape around you. You throw the net and happen to catch this:

I would like to describe the charming view off my porch here at my cabin in the woods where I am for a short vacation this summer, after I did not take a vacation in a long time. There is a very tall pine tree to the east of my cabin and it is so close to the cabin that its branches touch the roof. Some birds are in the tree every morning and sometimes when I am watching them I can see the ocean through the branches when there is no fog, although there is fog almost every morning here.

And so you go. That net is three sentences wide by eleven clauses high, ample enough to tow in more thoughts than a reader can digest at one sitting. To cull that haul, let’s begin by counting the clauses of the first sentence, this on the principle that each clause (a group of words with a subject and a predicate) speaks a thought we wish to communicate. That’s its logical purpose, but this same technique will also give us a chance to decide whether a particular clause-thought is worth stating at all. And by any good measure, the first thought of the first sentence here is not worth the saying: I would like to describe amounts to saying you are going to say something; it’s warming up, not playing the game, and that’s reason enough to delete the clause. But the view was charming, you might object. Well, then, begin with that: The view off my porch here at my cabin in the woods was charming.

But that same first sentence has two more clauses whose ideas are not relevant to the purpose of the piece—that you are on a short vacation and that you haven’t taken a vacation in a long time—neither of which ideas will affect the view you wish to describe (unless by some chance you are intending to write a psychological study on the relation between fatigue and the perception of beauty). So let’s reduce this first complex sentence to a simple one (The view off my porch here at my cabin in the woods was charming) and then go on to count the clauses of the second sentence. There we find another complex sentence with yet another three clauses. This technique of isolating the clauses within a sentence also gives us the chance to look closely at the grammatical construction we chose in our first drafting, and here we see that two of the three clauses employ the verb is—a weak choice when we’re describing a landscape, because the verb to be depicts stability, not movement. But for all that it appears otherwise, the natural world is never standing still, and strong sentences face what’s happening: so not, there is a very tall pine tree to the east of my cabin, but a very tall pine stands on the east side of my cabin; or better, a pine tree towers over the east side of my cabin. And note in that last version, the adverb very has no longer become necessary, because anything that is not very tall would not tower. It’s a good rule of thumb to guard against overusing such adverbs of degree. They make exaggeration easy, and amplification to the point of overstatement always reads as fishily uncritical. If we then abridge the second clause, we can trim this second sentence into a sharper picture: A pine tree towers over the east side of my cabin, so close that its branches touch the roof.

The third sentence is yet again complex, this time with no thinner a herd than five clauses. This gives us an opportunity to see how combining them can restrict the scope and tighten the grip of a long sentence. The first clause speaks of birds and the second of watching them, so why not find a stronger verb for the first verb and then combine the two: I watch the birds perch in the tree every morning. Then add the fifth clause to establish the physical condition: and if the fog has lifted; and then what happens as a result: I can see the ocean through the branches. This effectively deletes the need for the fifth clause, but we could attach the closing idea it contains into our new third clause, producing this final version: I watch the birds perch in the tree every morning, and if the fog, which is there every morning, has finally lifted, I can see the ocean through the green branches. We add the adverb finally to extend the clause by three syllables and avoid the abruptly clipped has lifted.

With all this, the observation to be made is that one draft is never the final draft. Human perception is too fine and human language too rough to ever get the two to match without a close inspection of our first try. And that is all to the good, because it compels us to look more closely at what we think we see, outside at the world and inside our mind. The craft of language will help us adjust our vision, and as we do, we might just find all manner of hidden subtleties that escaped our notice at first.

***

There is a meaningful distinction to be made between publicity and advertising. To publicize is to make known what in fact exists. To advertise is to make known what may or may not in fact exist. I might publicize a meeting by stating its subject, time, and place, but to advertise a product I wish to sell, I must do more than state the relevant facts: I must suggest an association, a likeness, between my product and other objects—a relationship whose power will depend upon your imaginative belief, whether veridical or not.

And therein begins the trouble. When an actor reading advertising copy tells me she can’t believe how much time she has saved using a new digital calendar, or when another actor, capably manufactured this time for a phone call, tells me her department is in the final stages of approving my unrequested loan request, what is the reality I am being induced to believe? Who, in other words, is speaking to me? And from what world? Is a human presence who has configured language to communicate one’s thoughts and reflections addressing me, another human subjectivity? And if not (of course not), am I confusing myself in engaging with the simulacrum? I know that a serious actor on stage is portraying a fiction; I have agreed to suspend my disbelief in order to enter another intelligible world. Am I expected to assume the same off stage? And to whose benefit am I involuntarily seized there?

Such thoughts come to mind in thinking more about artificial intelligence, specifically about large language model technology, because there is an analogy to be made between publicity and advertising on the one hand, and true human presence and mechanical verisimilitude on the other. For there is no logical way (though imaginatively there are many) to maintain that machine-assembled language is human language. Speech, the deliberate use of language written or oral to convey one’s living thoughts and emotions, is human for the very reason that it arises from human awareness. Its recognized elements and patterns take shape from the kaleidoscopic reflections which ideas and events make directly on the human mind. One witnesses all that and then speaks or writes. And that is where the difference between the dignity of human speech and the indignities of its deliberate impersonation is to be found.

And that difference is poignant. For am I not the less in letting something else speak for me, something bereft of that interior human perception and wonder? If I cannot find the words to say what I want to say—about what has passed before my mind, what has made its way through my thoughts and perhaps even etched at my heart—well, then I need another human being to help me learn the way words work in my language and another to tell me more about the manner in which language, like music, can so mysteriously convey meaning. In that way I may become more than I presently am. But what will not help my character enlarge is turning to an artificial someone to tell me what to say or do or think—unless, I suppose, I deem that humanoid good enough for me. But such a price to be paid. The language of human speech is not a fabrication, inconceivably sophisticated though it be. It is a making known of what directly presents itself to one’s mind, one’s own precious share of awareness. To short that connection is to accept the imitation for the genuine, and to accept gladly the parodic for the real.

There, I think, is where the conversation about this kind of AI needs to be had. It’s the same place as any serious conversation about education, about politics, about the economy needs to be had: what is a human being and what is the purpose of being human? If the answer is nothing more important, really, than anything else and so, to be honest, no distinguishable purpose at all, then indeed we should accept our wretched lot and get on as best we can. But if we answer in any way which points beyond our self-asserting selves and beyond the barren, violent world those selves can alone only make, then being told what to think and to be given the very sentences to say what we as priceless persons mean and feel in our own being—well, that will not be good enough for us any longer. Instead, each of us will find our dignity in awaking to our own bright mind and the language that brings that uncontrived presence among other minds in a human—not artificial—community worth gathering and preserving in its preeminence.

***

We forget sometimes that we have to know what we want before we can know what to do, that we have to see the big picture before we can make a considered judgment about how to proceed. In matters of education this means taking the time to understand the theory of a subject, and in the study of language and literature, it means learning the ways by which we can configure our sentences and paragraphs. What is called a formal education is one that is concerned to understand just these theoretical configurations—the forms, or patterns, which shape the specific elements of a subject into the meaningful whole we see.

The patterns of words into which we can shape sentences are traditionally called figures of speech. These figures, or configurations, can be thought of as linguistic designs by which we direct our readers’ attention into and then through the ideas we are presenting for their consideration. Arthur Quinn, author of a both learned and practical manual for writers called Figures of Speech: Sixty Ways to Turn a Phrase, says in his preface that the phrase figures of speech “is misleading in its static, passive form. It should be the ‘figurings of speech’—or, better yet, simply ‘figuring speech.’” “We are confronted, inescapably,” he concludes, with the intoxicating possibility that we can make language do for us almost anything we want.” Whereupon there follows definitions and illustrations of figures with such lusty names as anastrophe, anadiplosis, and polyptoton.

As it happens, the English novelist Elizabeth Taylor (not the actress) wrote a short story entitled “A Red-Letter Day,” where in a paragraph about halfway through there appears this one sentence happily illustrating, in the brief scope of twenty words, these same three figures of speech. Mrs. Lancaster, the headmaster’s wife, is greeting parents on Visting Day at an English public (that is, private) school:

As smooth as minnows were Mrs. Lancaster’s phrases of welcome; she had soothed so many mothers, mothered so many boys.

When a writer chooses to invert the standard English word order of subject then predicate, as Taylor does here in the first clause of this sentence, the unexpectedly reversed word order is a figure called anastrophe, the purpose of which is to awaken the reader’s attention (which was no doubt expecting the regularly ordered Mrs. Lancaster’s phrases of welcome were as smooth as minnows) with a sudden little verbal pinch. And had the author reordered the elements to the point of disorder, for example, Mrs. Lancaster’s phrases of welcome as smooth as minnows were, she would have been employing (unsuccessfully) another figure of speech called hyperbaton, sometimes defined as “the violent displacement of words across a line. In citing a successful hyperbaton, Quinn quotes the poet W. H. Auden: About suffering they were never wrong, / The old masters. Such configurations require, as Taylor astutely displays in her sentence, a subtle sense of when enough is enough.

The remainder of this one sentence, from the semicolon to the period, combines both anadiplosis, and polyptoton. When the last word of one clause (mothers) is used again in some form to begin the next clause (mothered), the figure is called anadiplosis, which in the original Greek of the term means nothing more mysterious than to double up again. And when that same word makes an appearance in different grammatical forms as it does here (mothers as noun versus mothered as verb), the configuration is called polyptoton, Greek, again, for many falls, meaning many occurrences, or grammatical inflections. (Students of Latin or Greek or German will recognize the term case, which in those and other highly inflected languages means the change a word undergoes in spelling to indicate a change in grammatical function. Case derives from the Latin casus, a fall, which is what is pictured to happen when a word falls, or changes itself, through its many grammatical forms.)

Every subject has its own set of specialized terminology like this (called its argot), the purpose of which is to direct our attention when we are learning to a uniform general notion, so that we can recognize the diverse particular instances of that notion when we come upon them. What we might call a vertical arrangement to thinking typifies formal learning, and when we think that way, we can move our mind in both directions, from first learning the notion and then seeing it exemplified in the particular, or from observing a particular instance and then tracing it back up to its general source. What we should not do is move only horizontally, from particular to particular, never trying to find the larger, explanatory idea under which each instance has its existence. To think scientifically about something first means to think about generalities, all-embracing ideas which account for the specifics flowing from them. Formal thought, and formal education, follows the traditional axiom that “there is no science of singulars.” Science, then, means more than a knowledge of discrete facts.

And the worth of this way of learning might be of good consequence to us at the moment, because merely collecting facts in an ever-speeding world of events does not automatically produce a system, which is an organized body of facts. To organize facts, we need more than practicality: we need reason, purpose, intelligible end. Such things satisfyingly answer the human need for meaning, which usefulness alone, no matter its flamboyance, cannot.

***

Perhaps along with other American Transcendentalists, Ralph Waldo Emerson is not a writer whose paragraphs are considered models of structured composition. His sentences, though, are another matter. Robert Richardson, in his First We Read, Then We Write: Emerson on the Creative Process, declares that “for Emerson the sentence—not the paragraph and not the essay—is the main structural and formal unit,” and that this preoccupation “steered him away from narrative, from logic, from continuity, from formal arrangement and effect.” One can understand, then, the reluctance of writing teachers to hold up Emerson’s extended prose as an exemplar of formal paragraph construction.

But all the thought that was in those sentences! Richardson continues:

When his sentences work, which is often enough, his success can be traced to his taking endless pains with sentence mechanics. He liked sentences that had a little bite or pop, a flash-point, and he had several different ways of achieving this effect, which we may distinguish as the whip-crack, the back-flip, the brass ring (hole in one), and the mousetrap.

Richardson goes on to give an example of each type from Emerson’s writings: “Every man is wanted, but no man is wanted much” (the whip-crack, where the final position snaps the idea); “Nothing astonishes men so much as common sense and plain dealing” (the back-flip, where an unexpected logic is displayed); “Hitch your wagon to a star” (the brass ring, where an idea sounds its profundity); and “Our chief want in life is somebody who shall make us do what we can” (the mousetrap, where the sentence catches an idea whole).

Yet for all that, it was the poet, not the prose writer, whom Emerson thought truly representative of the nature of human nature, because our prose sentences, even his, are still after the fact of the poet’s vision. In his essay “The Poet,” Emerson says:

For poetry was all written before time was, and whenever we are so finely organized that we can penetrate into that region where the air is music, we hear those primal warblings and attempt to write them down, but we lose ever and anon a word or a verse and substitute something of our own, and thus miswrite the poem.

So finely organized. Hours could be spent on the implications of that phrase alone. Organized administratively, as in well planned? Surely not. Organized as integrated, whole, organically cognate with? Perhaps we’re closer to his meaning there. Rich human depths to reflect upon. Robert Richardson’s short study of Emerson, still available, has much to say about reading and writing and language and poetry and art. He has written, as well, a biography of Emerson (Emerson: The Mind on Fire), and a biography too of Emerson’s worthy contemporary Thoreau (Henry Thoreau: A Life of the Mind). Perhaps the one unanticipated lesson Emerson will teach us is how to savor the long, slow time it takes to read and ponder him. Much like a long, slow, savory meal with a friend whose insight we’ve come to rely upon.

***