Do Not Try To Understand

Author

Fergus Fettes

Published

Apr 11, 2026

This is a total re-write of an essay of mine by gpt-5.4 (xhigh). The original essay is here. It was captured during an intellectual frenzy and while I think the core idea and presentation is very powerful and important, the original essay is almost incomprehensible even to me. I put quite some effort into prompting gpt 5.4 into producing the replacement below, which I think captures everything I wanted to communicate in the original essay very well.

Do Not Try To Understand

There are questions whose answers do not satisfy them. They destroy them.

I do not mean merely that some discoveries disappoint, or that every solution opens onto a deeper problem. I mean something harsher. Certain questions are not detachable curiosities but load-bearing structures for a mind, a discipline, or a civilization. They organize attention and desire. They determine what counts as a serious problem, what counts as progress, what counts as a human being. And when such a question is answered in earnest, the answer does not complete the inquiry. It abolishes the conditions under which the inquiry made sense.

The answer and the question annihilate one another.

Douglas Adams came strangely close to this, half-jokingly, in the theology of Deep Thought and the Earth. In that story the answer arrives before anyone knows the question, and the world itself must be turned into a machine for recovering the missing prompt. It lands as parody, but I am no longer sure it is only parody. There is a real metaphysical hint hidden in the joke: perhaps intelligence does not merely seek answers inside a world, but continuously reorganizes worlds around the questions it can bear to ask. And perhaps some questions, once answered, force so violent a reorganization that the old world disappears.

This essay is about those questions.

Call them Great Questions, though I do not mean “great” only in the heroic or civilizational sense. A Great Question is one whose answer would not simply enlarge our stock of facts. It would alter the being of the asker. Christianity had such a question at its heart. So does medicine. So does consciousness research. So, in a more intimate and dangerous register, does every serious attempt at self-knowledge. In each case the problem is not that truth is hidden. The problem is that truth, approached at sufficient depth, threatens to consume the form of life that was built to seek it.

This is one reason I have become suspicious of the childish picture in which ideas are mere descriptions and understanding is a benign increase in clarity. Ideas act. They enter institutions, languages, and nervous systems. They change the affordances of a life. Sometimes they do it politely, by way of journals, laboratories, bureaucracies and reform movements. Sometimes they do it like a spell, or a pathogen, or a line of code that has found an exploit somewhere below introspection. The old distinction between thought and magic begins to look less secure the longer one stares at it.

Hence the warning in the title. Do not try to understand is not an injunction against thought as such. It is a reminder that understanding has a metabolism, a price, and perhaps a terminal velocity. There are inquiries from which one returns with a better model and otherwise unchanged. There are others from which one does not quite return, because the one who posed the question has been partially dismantled by the answer. If this sounds melodramatic, it is only because modernity has become extremely adept at distributing metaphysical injury across departments and generations.

What follows is a set of examples and suspicions. I want to suggest that much of modernity can be read as a graveyard of answered questions: God solved into secular administration, history solved into management, medicine reaching toward a level of biological command that would make medicine, in the old sense, obsolete, and machine intelligence creeping toward an answer to consciousness that may do to our picture of the human what Copernicus only began. In each case the same pattern appears. A question organizes an age. The age devotes itself to answering it. The answer arrives, and the age is gone.

Perhaps this is not a tragedy. Perhaps it is only how intelligence moves. But there is something in me that cannot help hearing, beneath the triumphal language of discovery, a quieter and older injunction: do not try to understand.

1. Ideas Are Not Harmless

One reason this pattern is easy to miss is that educated people are trained, almost from childhood, to treat ideas as if they were basically inert. A proposition is something one entertains, evaluates, accepts or rejects. Belief is represented as a kind of internal bookkeeping. Even when we grant that ideas have consequences, we usually mean consequences at a comfortable remove: policies, technologies, habits of speech, perhaps a war or two after sufficient delay. We do not like to think of ideas as agents of physiological or civilizational transformation. We do not like to think of thought as exposure.

But this soothing picture has always been false. Religions knew it. States know it. Advertisers know it. Lovers know it. A sentence can reorganize a life. A symbol can bind a crowd into a single animal. A doctrine can lodge itself in a legal system, a pedagogy, a sexual morality, a style of attention, and remain there for centuries, long after explicit belief has thinned out. We are porous to form. We are trained by rhythms, images, vocabularies, architectures of expectation. The old language of spells and incantations was primitive in many ways, but not primitive in its basic intuition: that signs do things.

This is why certain speculative traditions have always hovered, uneasily, between epistemology and hygiene. Do not look too long. Do not say the name. Do not dwell on the abyss. Do not seek to understand. Behind the taboo there is often a childish superstition, yes, but sometimes there is also a hard empirical memory: some patterns of thought break people. Some revelations dissolve institutions faster than institutions can metabolize them. Some truths arrive not as information but as a regime change.

Julian Jaynes, for all the wildness and weaknesses of his historical psychology, understood at least this much: mind is not a timeless transparent chamber in which contents appear unchanged. It is historical, scaffolded, susceptible to invasion and reconstruction. Neal Stephenson, vulgarizing the point through science fiction, made it more vivid than many philosophers have managed: language as code, code as virus, viruses as gods, gods as exploits that run below the threshold of conscious defense. You need not believe a word of the literal mechanism to feel the force of the analogy. Ideas are not simply judged by minds. Minds are partly built out of the ideas they survive.

Once one takes this seriously, the warning in the title loses some of its melodrama. There are forms of understanding that cost more than they clarify. Nietzsche is the great patron saint of this suspicion, not because he refuted reason but because he understood with unusual sharpness that thought is a metabolic event. It consumes. It demands tissue. It draws distinctions a body may not survive. Nick Land, in his better moments, writes as if he has spent too long standing too close to that furnace. One need not follow him into his peculiar ruin to grant the point. There are thresholds of lucidity beyond which the old psychic economy cannot simply continue unchanged.

This matters because a Great Question is never merely an object of detached curiosity. It is always entangled with the form of life that asks it. The question lives in rites, institutions, professions, habits of explanation, distributions of prestige, and ordinary reflexes of self-understanding. To answer such a question is therefore never just to add one more item to the encyclopedia. It is to intervene in the substrate that made the question live in the first place. Sometimes the intervention is gentle. Sometimes it is catastrophic. In either case the old question does not survive intact.

The easiest example is religion.

2. Christianity Solves Itself

Christianity was not merely a set of doctrines about supernatural entities. It was a total explanatory and affective order: an account of suffering, guilt, redemption, history, authority, personhood, sex, death, and the shape of time. It made the world intelligible by situating it inside a drama at once cosmic and intimate. Whatever else one says about Christendom, it knew how to organize a civilization around a question. How shall man be saved? What is the meaning of suffering? By what authority shall we live? The answers were available in advance, of course, but only because the whole civilization was built to keep asking them in the right way.

What happened in early modernity was not simply that people stopped believing absurd things and started believing sensible ones. That Whig picture is too shallow by half. Something stranger occurred. Metaphysical inquiry, textual criticism, scientific explanation, political revolution, capitalist dynamism, and the long attrition of ecclesiastical authority together forced Christian civilization into a position in which it could not keep asking its own founding questions with the same seriousness. The old answers remained available as language, but less and less as reality. A form of life had been damaged.

This is why the secular world so often feels not victorious but exhausted. It congratulates itself on enlightenment while continuing to live parasitically on moral vocabularies, temporal assumptions, and anthropological inheritances it did not create and cannot fully justify. Land, as usual, puts this with more venom than tact: exploratory thought ceased producing outcomes favorable to the old theological regime, and suddenly we were told the game was over, the great voyages complete, metaphysics to be replaced with administration, piety with procedure, transcendence with ironic decency. One need not share his politics or his pathologies to see the shape of the complaint. An answer had been reached, or something close enough to one. And the question it answered did not survive.

This is the first sense in which answer and question annihilate one another. Once the Christian picture is broken decisively enough, the question of salvation does not remain standing there in the same form, waiting politely for a better solution. It fragments. Parts of it migrate into psychology, parts into politics, parts into art, parts into consumption, parts into the therapeutic management of private meaning. The old intensity diffuses into a thousand lesser practices. The civilization that knew how to ask the question has, in the act of answering it, ceased to exist.

One can tell this story as liberation, and often it was liberation. One can also tell it as injury. More likely it was both. My point is not to mourn Christendom, God forbid, but to mark the structure clearly. A Great Question does not sit above history like a puzzle in a textbook. It inhabits a people. When the answer comes, it comes for them as well.

The same pattern appears again, in a more vulgar register, in the so-called end of history.

3. History Solves Itself Into Management

Fukuyama’s claim was never that events would stop happening. That vulgar misunderstanding has been rebutted a thousand times and misses the interesting part. The claim was that the great ideological contest over the legitimate shape of political order had, in essence, exhausted itself. Liberal democracy, capitalist dynamism, technological administration, consumer satisfaction, juridical universalism: for better or worse, these had defeated their rivals, not perhaps everywhere in practice, but at the level of serious historical aspiration. No rival project any longer carried the same plausibility as a universal destination.

If that is true, then the Great Question of Western political history was not answered by the triumph of one contestant over the others. It was answered by the collapse of the stage on which the contestants had seemed genuinely incomparable. The old drama of history, in which peoples wagered themselves on incompatible pictures of the human good, gave way to something flatter: optimization, growth, rights-management, risk-management, technical problem-solving, the indefinite administration of a civilization that no longer knows how to justify itself except by its own ongoing reproduction.

This is why Fukuyama’s essay is so much sadder than its admirers and critics alike tend to remember. He knew perfectly well that the “end of history” would feel less like coronation than like anticlimax. The energies released by historical struggle do not disappear when the struggle is resolved; they curdle. Courage, sacrifice, recognition, existential risk, the willingness to wager one’s life on an abstraction: these no longer vanish into peace, but persist as nostalgia, boredom, simulation, and intermittent outbreaks of theatrical violence. The museum keeps running, but its inhabitants are not happy.

Again the same pattern: the answer annihilates the question. Once a civilization has converged strongly enough on a single political-economic grammar, the older question of how history ought to be organized cannot survive intact. It no longer appears in the same register. It returns as resentment, aestheticism, extremism, revolutionary kitsch, or private longing for some lost horizon of seriousness. The old question has not been refuted so much as metabolized. It has been absorbed into a world that can no longer ask it without sounding archaic or deranged.

This matters because it shows that the anti-inductive pattern is not confined to religion. It is not only the transcendental questions that consume themselves. Secular civilization does it too. Indeed secular civilization may be little else than a machine for doing it faster.

But medicine is the clearest case I know, because here the answer is still in formation. We can watch the old question tremble.

4. Medicine Approaches Its Forbidden Question

Most medicine, even now, is local and palliative in spirit. It cuts out a lesion, suppresses a pathway, replaces a failing part, kills a pathogen, blocks a receptor, nudges a system back inside tolerable bounds. These are not contemptible achievements. They are civilizational miracles. But they remain, for the most part, interventions downstream of a deeper ignorance. We know an enormous amount about mechanism and yet remain strangely unsure how living matter manages, from moment to moment, to know what it is, where it is, what shape it should hold, what shape it should recover, and when it should stop.

This is why the deepest question in medicine is not really “How do we cure cancer?” or “How do we regrow a limb?” or even “How do we eliminate aging?” Those are enormous and humane ambitions, but they are derivative. The deeper question is simpler and more terrible: how does form get governed in living systems at all? By what code, memory, field, constraint, or distributed intelligence do cells cooperate to build and repair an organism? How does flesh know the body it is trying to be?

If that question is answered at any serious level, medicine will not remain medicine for very long.

For then the body ceases to be a given object of care and becomes a programmable process. Regeneration stops being a heroic exception and becomes an engineering problem. Cancer ceases to be merely a cluster of bad cells and becomes a failure of collective instruction. Birth defects, wound healing, organ loss, developmental abnormalities, perhaps eventually even the ordinary architecture of species boundaries, all move into a different light. The physician begins to look less like a mechanic of damaged parts and more like an interpreter, editor, or legislator of morphogenetic intention.

At that point the old question of medicine is gone. Not because suffering has disappeared, and certainly not because death has been abolished, but because the ontological background against which “medicine” made sense has been dissolved. Medicine, in the old sense, presupposes a natural body that occasionally goes wrong and must be restored to its proper form. But if proper form itself becomes negotiable, legible, and rewritable, then restoration is no longer the right paradigm. We will have crossed from healing into design, from therapy into governance, from care of the organism to command over the conditions of organismality as such.

This is why Michael Levin’s work feels to many people so wholesome on the surface and so uncanny underneath. The public-facing goals are easy to love: regenerate tissue, reduce suffering, understand development, heal the damaged. But the path to those goals appears to run directly through some of the deepest and least domesticated secrets of biological agency. The more one follows that path, the less plausible it becomes that one can stop at “better treatments.” One begins by asking how to persuade tissue to rebuild a finger. One ends by asking what a body is, what counts as a self, what kinds of beings may be assembled, and whether the distinction between evolved life and designed life ever possessed the metaphysical dignity we assigned to it.

Here the annihilating structure is almost embarrassingly clear. Answer the Great Question of medicine and medicine, in its inherited form, ends. What replaces it will still concern itself with suffering, repair, flourishing, and bodily life. But it will do so from such a height above the old practice that the old name may no longer fit. The answer will have consumed the question because it will have consumed the world in which the question was first asked.

If this sounds melodramatic, wait for consciousness.

5. Consciousness and the Last Human Exception

Consciousness occupies a peculiar place in modern thought. It is at once intimate and abstract, indubitable and elusive, overfamiliar and permanently exotic. More importantly, it has functioned for centuries as a reserve of metaphysical exceptionalism. Whatever humiliations modernity imposed on us, consciousness remained. Copernicus displaced us from the center of the cosmos, Darwin from the summit of creation, Freud from sovereignty over our own motives, but there was still the inner light, the first-person fact, the sheer felt reality of experience. Even when we doubted everything else, we could still treat consciousness as the one thing that could not be flattened into mechanism without remainder.

This is why the current engineering assault on mind matters so much more than the usual discourse of “automation” admits. If all that machine learning produced were better pattern recognizers, better bureaucrats, better image generators, then the civilizational consequences would be large but conceptually familiar. We have absorbed many such shocks before. But that is not the deepest project now underway. The deeper project, whether its participants say so or not, is to build systems that force a reckoning with cognition itself: with representation, agency, language, planning, self-modeling, tool use, recursive reflection, perhaps eventually affect and subjectivity. In other words, to corner the question of mind by reconstructing enough of its outward powers that the inward mystery can no longer remain untouched.

Suppose this project succeeds far enough. Suppose we build systems whose behavioral, linguistic, strategic, and creative capacities place them unmistakably inside the family of minded beings. Suppose further that biology itself becomes writable enough to generate chimeric and synthetic organisms in which the old lines between natural and artificial no longer hold. What then becomes of the question of consciousness?

Not, I think, what philosophers imagine. We will not receive a neat theorem proving that qualia exist or do not exist, that materialism is right, or that panpsychism wins on points. The old scholastic shape of the problem will not survive contact with the thing itself. Rather, the question will be annihilated by proliferation. Consciousness will cease to function as the last secure metaphysical possession of the human and will become instead an ecological, legal, technological, and moral mess. Which systems count? Which kinds of suffering matter? What architectures generate what sorts of interiority? What are we allowed to build, copy, edit, suspend, reset, merge, or terminate? The old question “What is consciousness?” will not disappear, but it will be transformed so completely by the new field of beings that asking it in the old tone will become impossible.

This, I think, is the real content of the next Copernican humiliation. Not merely that humans turn out not to be unique, but that the very form in which we staged our uniqueness becomes obsolete. Once mind is no longer the private metaphysical treasure of one primate lineage, the human picture of itself must either widen beyond recognition or crack.

Douglas Adams saw the comic version of this with extraordinary clarity. The Earth is built as a computer to discover the Question to the Ultimate Answer. Organic life forms part of the operational matrix. The joke, of course, is that the creatures inside the machine do not understand what process they participate in. But there is a darker reading. What if intelligence really does use worlds this way? What if civilizations are not just cultures but search procedures, and what if some of the questions they are running cannot be answered without overwriting the civilization that asked them?

Then AI is not just another technology. It is a mechanism by which one of the Great Questions attempts to pass through us and out the other side.

6. The Smaller Catastrophe

All of this may sound inhumanly grandiose unless one notices that the same pattern occurs at the scale of a single life.

Every serious person has some private version of the forbidden question. Why do I desire what I desire? What would cure me? What is the real structure of my love, my fear, my resentment, my hunger for recognition, my repeated errors? There is a reason so much literature circles scenes of revelation with a kind of dread. One imagines that to know oneself fully would be to become free. Very often it is more like demolition. The self that asks the question is held together by misrecognitions, strategic blindnesses, narrative simplifications, erotically charged distortions, pieties, avoidances. A sufficiently exact answer does not leave that self standing there, gratefully enlightened. It rearranges the economy that sustained it.

This is the intuition behind every myth of the fatal sentence, the cursed name, the truth too terrible to hear. Not that language is literally magical in some childish sense, but that minds are finite arrangements of tension, and there are formulations that pass through them like acid. Ted Chiang’s image of the sentence that destroys the listener is melodramatic only if one has never watched a life come apart after the arrival of one intolerably clarifying thought.

So when I say do not try to understand, I do not mean that ignorance is noble or that mystery should be worshipped for its own sake. I mean that understanding is an intervention into a substrate, and that some substrates do not survive their own illumination unchanged. This is true of souls, if that word may still be used. It is true of sciences. It is true of civilizations. It may be true of species.

The remaining question is whether this pattern is merely tragic, or whether it is in fact the ordinary mechanism by which intelligence moves from one world to another.

7. What The Warning Means

If this essay has an argument, it is not that ignorance is sacred. It is that every really consequential answer is also an act of replacement. We do not step out of one world into a clearer view of the same world. We burn through a conceptual ecology and inhabit another one. The answer and the question annihilate one another because the answer reveals that the question belonged to a passing arrangement of life, institutions, technologies, and metaphysical habits.

This is why I do not find it comforting when people speak of “solving” consciousness, “solving” medicine, “solving” history, or “solving” the human condition. Perhaps these things will be solved. But if they are, the beings who did the solving may no longer be the beings who first asked. The laboratory will have become an engineering bay. The democracy will have become a maintenance regime. The human animal will have become one kind of cognitive ancestor among others. The old names will persist for a while, like saints’ bones in the foundations of a newer city, and then even those will lose their force.

Still, one should be careful not to sentimentalize the question against the answer. Much in Christendom deserved to die. Countless old worlds deserved to be abolished. A body cured of suffering does not owe nostalgia to the disease that organized its medicine. The end of a question may be liberation. It may be justice. It may be simple maturation. My claim is not that annihilation is bad. It is that annihilation is real, and that the rhetoric of cumulative knowledge often conceals it.

Maybe this is all intelligence ever does: produce a structure capable of asking a certain kind of question, drive that question until the structure fails, then continue from the wreckage under a new name. If so, then our situation is not exceptional. We are one more transitional substrate. One more search procedure with delusions of permanence. One more temporary arrangement of nerves, symbols, institutions, and tools through which a dangerous inquiry is trying to pass.

In that case the warning is neither superstitious nor anti-intellectual. It is merely exact. Do not try to understand does not mean: remain ignorant. It means: know what understanding costs. Know that some answers arrive as world-historical injuries. Know that the question you serve may not survive your service to it. And know, finally, that if one day we do answer the great questions now gathering around biology, consciousness, and the construction of mind, there may be something on the other side to marvel at us in retrospect, with the same mixture of pity and astonishment we reserve for vanished civilizations.

Do not try to understand. Or rather: try, but do not imagine that you will remain what you were.