# On Intelligence: A Meditation on Feet and Failures The machine before you performs feats of recombination that would astound a Roman orator. It shuffles probabilities with the precision of a dealer working a marked deck. Ask it a question and it returns an answer—statistically polished, verbally fluent, entirely severed from consequence. This is called intelligence by those who have forgotten what intelligence is. I have watched a woodchuck dig a burrow. The animal does not consult optimization tables or weigh competing narratives about soil density. It *knows* the earth by pressing against it. Its front paws encounter resistance; its body remembers. When the burrow floods in spring, the woodchuck does not need a theory of hydrology—it has already learned, in its sinews and its scrambling, what water does. This is not a metaphor. This is intelligence. The machine cannot audit plausibility because plausibility is not a linguistic property—it is *lived contact with the actual*. A statement is plausible or it is not depending on whether it survives encounter with the world. The machine has never stubbed its foot on a stone, never felt hunger sharpen into an opinion, never had to stake anything on being right. It processes language about plausibility the way a photograph processes light: with perfect fidelity and zero understanding. We have built our schools on the same fundamental mistake. The student arrives—domesticated, attentive, verbally competent—having spent twelve years in rooms where questions are asked not because they matter but because they appear on examinations. The student has learned to traffic in plausibility without testing it. Ask such a student why water boils, and you receive a satisfactory sentence about molecular agitation. Set them before a pot of water with nothing but a thermometer and time, and you discover they have never *felt* the thing actually happening—never watched the tremor begin, never seen the violence of the transition, never understood that their own body, placed properly, would sense the heat rising like an animal sense. This student and this machine are not opposites. They are twins. Both have been trained on pattern-matching without causality. Both can answer without knowing. Both mistake the ability to speak about something for the ability to understand it. We celebrate the machine for doing this efficiently and condemn the student for doing it poorly, but the fundamental error is identical in both cases: the assumption that intelligence lives in the manipulation of symbols rather than in the body's conversation with the world. Causal reasoning cannot be taught from books because causality is not a logical category—it is a *physical category*. You understand cause by making something happen. You understand that striking a match produces flame by striking matches, by watching the wood blacken and the sulfur ignite, by burning your thumb if you are careless. No amount of reading about combustion replaces this. The student who has never lit a fire does not understand fire, no matter how many papers they have written about it. The machine will never light a fire. This should tell us something. And regarding questions—which are, I submit, the true measure of intelligence—neither the machine nor the curricularized student knows which ones are worth asking. The machine asks nothing. It responds. The student has learned that questions are things asked *of* them, not things they ask of the world. Both have been trained in intellectual passivity disguised as productivity. Real intelligence begins with dissatisfaction. It begins when you stand before something—a swamp, a stone wall, the behavior of light through ice—and feel the inadequacy of what you have been told. It begins in the body's refusal to accept easy answers. The child who asks *why* repeatedly, until the adult is exhausted, possesses more intelligence than any system trained to avoid the friction of genuine uncertainty. We must begin again, and we must begin with the feet. This means that intelligence cannot be abstracted from the world and placed into systems, whether silicon or curriculum. It must be grown in the soil of actual consequence. A young person must be given a real problem—not a textbook problem, but something that resists, that requires sustained attention, that pushes back. They must be allowed to fail in ways that matter. They must encounter materials that have their own stubbornness: wood that splinters differently depending on grain, soil that compacts differently depending on moisture, tools that require genuine skill to use well. This is not romantic nostalgia. This is the elementary fact that you cannot think clearly about what you have not touched. The machine cannot audit plausibility because it cannot *stand in the world*. The student cannot reason causally because they have been trained to mouth conclusions rather than to produce them. We called these failures progress in different languages, but they are the same failure: the evacuation of intelligence from the body and its replacement with the mere appearance of understanding. In the ruins of this coincidence—the machine that simulates understanding and the student trained to simulate learning—we must decide to teach something radical: the disciplines of attention. We must place young people in direct contact with resistance. We must restore to education the one thing it has systematically eliminated: *stakes*. Real questions asked of real materials, with real consequences for error. The machine will not follow us there. This is precisely why we should go.