Art by Jin Suk
The Dream of the Ring
The ancient pulse of germ and birth
Was shrunken hard and dry,
And every spirit upon earth
Seemed fervourless as I.
At once a voice arose among
The bleak twigs overhead
In a full-hearted evensong
Of joy illimited;
An aged thrush, frail, gaunt, and small,
In blast-beruffled plume,
Had chosen thus to fling his soul
Upon the growing gloom.
—Thomas Hardy, from “The Darkling Thrush”
I’m writing this at the end of the first week of January. It’s an anxious time. The news is full of the incomprehensible devastation of the L.A. fires and the upcoming inauguration of a new president who has promised to turn all government institutions and precedents on their heads. By the time this sees print, we will have gotten a taste of what’s in store.
These immediate events naturally affect my tone and approach here, but I don’t know if my take would be any different without them. There are so many other things a worrier can worry about. A short list might include climate change, the Ukraine war, the Gaza genocide, immigration and deportation crises, corporate grift and the instability of our justice system, terrorism, homelessness, inflation, official and unofficial surveillance, threats of new diseases . . .
This list is partial, and I’m struck as I make it that twenty-five years ago, at the turn of the millennium, before the World Trade Center attacks, things really did look different. Wars, yes—always—and there were certainly climate concerns, but they were not yet everywhere in the news. Our faith in justice systems was fairly intact, the specter of global disease was still mostly hypothetical. The world as we knew it was just not the same as now. I was there; the satchel of human cares seemed lighter, and the future open. We had two young children, and we were not girding them to face all the things that might befall them.
How do we measure these collective states, how do we filter out nostalgia and rosy memories, so we can speak honestly about the various changes and their effects on us?
I saved out one item from my list. For me it’s every bit as consequential as the others. I mean the first emergence of digital technology and, after just a few years, the great leap toward artificial intelligence. Worry attends here, too.
These advances have significantly changed how we do things—how we live, and even how we conceive of ourselves. On an immediate, practical level, AI innovations have infiltrated the job market and economy, eliminating many middle-level positions while creating a spectrum of new job needs, altering how corporations do business and how people interact. A substantial part of the human interface has been eliminated: banking is done at machines or at a keyboard, communications between people and organizations and businesses of every kind are brokered by automated systems, and so on. We live with a growing apprehension about simulation: Who is on the other end, am I dealing with a person here? And we are so obviously being tracked: Hey, I just mentioned kitty litter a minute ago and look at this—I have five ads on my screen for kitty litter.
I think of this as alienation. The imposition of the public interface is one thing, but large parts of our social lives are lived behind a scrim. Zooms, chats, text messages intercede, in many instances supplanting face-to-face interchange. We’ve grown so accustomed to these modes that we may not recognize the personal shortfall from earlier times. Most of us just embrace it. My children tell me that neither they nor their peers like dealing voice-to-voice. Texting and DM-ing are enough. Is what I call alienation even experienced as alienation, and if it isn’t, then what do we call it?
The digital encroachment feels coextensive with all aspects of our living. Boundaries between former ways and the new are smudged. What has changed? How can we tell the difference? Simple: turn out the lights, put the iPhone aside, step far into the night, and look up: how strange it feels, the ancient Other!
~
I always ask: how close can AI come to simulating humanness? Programs can already convincingly impersonate us on many fronts, do the things we do, and, for instance in the field of medicine, do things we can’t. Writing-bots, ever more refined, can get top grades in university classes, where it is not unheard of that their work is being evaluated by grading programs.
What AI can do is one question, what it can’t is another.
I recently listened to a podcast with writer Lawrence Weschler in which he developed a revealing analogy. He asked the listener to imagine a triangle drawn inside a circle, points touching the perimeter, then to replace the triangle with a square, then a hexagon, and onward, expanding the number of segments as far as the mind can imagine, so far that to all appearances the inside figure is identical to its enclosure. Except that it’s not. His point is that even at the most extreme imaginable iteration absolute identity cannot be achieved.
It’s Zeno’s paradox updated. The lines, digital products of 1s and 0s, are rational, while the shape they aspire to is organic, irrational. No matter how minute the subdivisions, they will never make the full transformational leap. The line can never achieve curvature. And in the same way, the program based on 1s and 0s can never achieve the ultimately nonlinear essence of the human.
~
The diminishing margin between segments and circle may appear to be nearly invisible in the visual graphic I’ve sketched here, but what lies within that margin matters greatly. It is nothing less than what’s commonly called the unconscious.
As the body is said to be about sixty percent water, so the psyche is a sizable percent unconscious. Dreams and day dreams; untended meanderings of the mind. The philosopher Henri Bergson calls this the “deep self”—the core of a person’s being, accessed through intuition, which is to the mind what instinct is to the animal self. He characterizes the deep self as a continuous action of self-creation, distinct from what we might call the outer “daily self.” We can point to this intangible core entity, but we also need to consider how much its unruly energies impinge on our daily actions—our ways of speaking, of writing.
AI does not possess an unconscious, plain and simple. While there are programs that can very nearly replicate the neural pathways of the conscious mind, they have no access to the mind behind the mind, which of course influences the conscious at every turn. This is why words generated by predictive AI are a kind of a Potemkin village of expression. Genuine mental activity is at no point predictable. Even the most sophisticated large language models will never produce artistic writing, at least as I imagine it: language hemmed around by irrational colorations, open to leaps and divagations in its progress.
~
Style is a very simple matter; it is all rhythm. Once you get that you can’t use the wrong words. But on the other hand, here I am sitting after half the morning, crammed with ideas, with visions, and so on, and can’t dislodge them, for lack of the right rhythm. Now this is very profound, what rhythm is, and goes far deeper than words. A sight, an emotion, creates this wave in the mind, long before it makes words to fit it.
—Virginia Woolf, letter to Vita Sackville-West
Woolf ’s words come closer than other assertions I’ve read to capturing how writing works. She identifies the primacy of rhythm, rhythm preceding language—body before mind—proposing in effect that our bodily instinct is our original way of being in the world, of knowing. The rhythm finds the words. It doesn’t happen to me with any regularity, but it does happen, and I know that many writers have the experience—called “inspiration” or “being in the zone”—when the words seem to arrive before the thought, when they make the thought, when the keys are clicking to keep up, when the writer knows that these words are the right words, their way of falling into place being the assurance.
AI, even as it draws on a mountainous database of works by writers (unacknowledged and unpaid) and is programmed for the most supple responsiveness, cannot find the physical rhythms that tap the unconscious, that discover associations. . . Programs are not bodies. Programs don’t grow strong on mother’s milk. This is insultingly obvious, but it needs to be asserted every now and then.
~
With digital expression, there is of course the issue of “close enough” or “good enough.” What AI produces, at the tap of a finger, is prose that passes muster. And that achievement, significant, has collateral consequences. As more and more of our texts—in business, media, advertising, and academia—are generated thus, we are slowly coached to acceptance, and a new ethos begins to infiltrate our culture.
In some ways this can be seen as a betterment, bringing about a massive gain in efficiency. Prose is stripped for parts, parts turned into units that can be manipulated according to need. But on another level—the level where my worries live—this new ease is a slippery slope, what feels like the beginning of a culture-wide regression. To call it a dumbing down may be too strong. Rather, it is a re-prioritizing, an acceptance of mere adequacy, a scrapping of nuance.
This is not the end of the world, I understand. But “the end of the world” is not my measuring stick. I want to think in terms of what is best in us. I value our aspirational selves, which have nothing to do with convenience or the so-called bottom line. We are better than that, and we know we are. My hope is that as our lives are increasingly taken over by high-functioning programs and mediated contacts, the aspirational part of our natures will not atrophy, but will, among other recourses, turn to the expressive arts—look for some verification that there is a larger and deeper view of things to be had. It would seem a natural tropism. Do we believe our lives are pointed toward meaning, some personal consummation? If we don’t, then what are we to do with ourselves? Do we just scroll away our days until the reel runs out?
~
One of the delights of the essay form is that it allows us writers certain liberties. The root of the word, as we are forever being reminded by essayists, is “to try, to attempt.” In that spirit, I would like to change things up, to pause my horses midstream and, taking a deep breath, go meta. I’m inspired to bring the process of formulating this essay to bear on the subject itself.
There is the well-known story of August Kekulé, the German chemist who in the mid-nineteenth century was intent on discovering the structure of benzene. He worked for a very long time without success, but one night, asleep, he had a vivid dream of a snake eating its own tail—the archetypical ouroboros—and he woke up with his answer: the benzene ring.
As I was developing the main thrust of what I’m writing here, I suddenly remembered hearing that story. It arrived via the familiar writerly route, by unpremeditated association. And I thought, Of course! Hadn’t I just been describing Weschler’s analogy based on the image of a circle?
How does the associative psyche work? The Kekulé rings, I’m convinced, prepared for or even invited a follow-up image, which, when it came to me soon after, let me see much more clearly what I’d been pushing toward—as if the deeper mind knew. It appeared unannounced in the mind’s eye like a gift.
I’ve been trying to make the case that the digital in its full-scale incursion steadily diminishes what I think of as the human portion of our lives.
Think of all the ways digital systems intervene and mediate, how they change and simplify so many of our tasks but in doing so distance us—brokering our sense of the weather, our spatial orientation, and our financial grasp on things as 0s and 1s dissolve the cash nexus. Further, each mediation alters our sense of personal gravity. We feel less tethered to the slow material world. A new ease and efficiency lead to willing subjugations, a significant giving over of agency.
~
The image, as I said, came in a flash—a daytime flash, not a dream. As Kekulé saw his snake, that circular inspiration, so I had, from out of seeming nowhere, a distinct mental image of da Vinci’s famous Vitruvian Man: the human figure drawn inside a circle, his four limbs extended, each extended digit touching points on the circumference, like Weschler’s segments.
Since the time of the Renaissance, that figure has become an archetype, exalting the human and influencing our collective self-image. It articulates the age-old belief that the human is the right measure of things.
The flash was an instigation, and in its unexpected arrival a double confirmation. It made—enacted—the case for the role of the unconscious in the writing process. And at the same time it showed the very thing that will never be within digital reach.
In it, I identified my core concern, which goes deeper than virtual usurpation. The more fundamental worry became immediately obvious when I put Weschler’s visual analogy side by side with da Vinci’s figure. The two images communicate the human portion, then and now. Vitruvian Man is the time-honored paradigm: the free-standing individual, confident, and, as the fact of the circle suggests, complete. Beside it: the segmented iteration of a subjugated counterpart.
This shift needs to be contemplated—the decline from grandeur to capitulation, a gradual sacrifice of individual initiative.
~
This is a pessimistic assessment, I admit, but it’s not a hopeless one. I’ve always been drawn to the idea of fraught survival—the tiniest indestructible essence, the light bleeding through the crack, as a countering energy and a stay against the forces that would destroy.
There’s the song of Hardy’s thrush, yes, but nowhere is it more beautifully expressed than in Eugenio Montale’s poem “The Eel,” which embodies the power of compression and announces the lasting resilience of spirit.
Montale describes the eel, that slight slithering creature, leaving its cold Baltic waters for the warmer southern seas, and from there making its way up through estuaries and rivers, higher and higher, “deep under opposing tides / of branch narrowing into more slender / branch,” pushing its way right into the dry wastes:
filtering like bright water through
rivulets of mere mud until one day
light flashed from chestnut leaves
lights up the quiver in a dead pool. . . .
“Love’s arrow on earth,” Montale calls it,
the green spirit who seeks
life there only
where drought and desolation gnaw,
the spark that says
everything begins where everything seems
charcoal, a burnt-down stump;
brief rainbow, iris, twin
to the glance mounted within your lashes
which you keep sparkling and untouched
in the midst of the sons
of Man, all sunk in your mire—Can you
not see she’s your sister?
(trans. Millicent Bell, AGNI 51)
I console myself with these words in our parched-out days. The conviction that there is a surviving generative spark—an essence apart from all the systems we have wrapped around ourselves. This essence, where does it live? Where can we find it? People will search in their own places, but for me it’s the reason for art. Nietzsche put it succinctly: “We have art so that we shall not die of reality.” I find that essence in the written work that taps the deepest part of the core, that mines our daydreams and longings, our private passions and emotional formations—in the process revealing all that no code ever will.
—S. B.

Sven Birkerts is coeditor of AGNI. He is the author of eleven books: An Artificial Wilderness: Essays on 20th Century Literature (William Morrow), The Electric Life: Essays on Modern Poetry (William Morrow), American Energies: Essays on Fiction (William Morrow), The Gutenberg Elegies: The Fate of Reading in an Electronic Age (Faber & Faber), Readings (Graywolf), My Sky Blue Trades: Growing Up Counter in a Contrary Time (Viking, 2002), Reading Life (Graywolf, 2007), Then, Again: The Art of Time in the Memoir (Graywolf, 2008), The Other Walk (Graywolf, 2011), Changing the Subject: Art and Attention in the Internet Age (Graywolf, 2015), and The Miro Worm and the Mysteries of Writing (Arrowsmith Press, 2024). He has edited Tolstoy’s Dictaphone: Writers and the Muse (Graywolf) as well as Writing Well (with Donald Hall) and The Evolving Canon (Allyn & Bacon).
He has received grants from the Lila Wallace-Reader’s Digest Foundation and the Guggenheim Foundation. He was winner of the Citation for Excellence in Reviewing from the National Book Critics Circle in 1985 and the Spielvogel-Diamonstein Award from PEN for the best book of essays in 1990. Birkerts has reviewed regularly for The New York Times Book Review, The New Republic, Esquire, The Washington Post, The Atlantic, Mirabella, Parnassus, The Yale Review, and other publications. He has taught writing at Harvard University, Emerson College, Amherst College, Mt. Holyoke College, and the graduate Bennington Writing Seminars, which he directed for ten years. He lives in Amherst, Massachusetts. (updated 4/2025)