# On Causation and the Body: An Epistle to Pearl and His Disciples Listen: I will convince you that intelligence dwells not in the mind's airtight chambers, but in the body's dangerous commerce with the world—and that your precious causal diagrams, for all their crystalline beauty, are merely the *shadows* of shadows cast upon a wall we ourselves have built. You speak of Pearl's restoration as though he freed some imprisoned sovereign. But consider the exile itself. Science banished causal language not from timidity, but from a kind of violence—a necessary murder. The old metaphysics chattered endlessly about *why*, and nothing got measured. So the methodologists performed an amputation: we cut away the limbs that reached toward intention, toward agency, toward the tangled motivations that move us. We kept only the stumps—correlation, covariance, the mute facts. This was cruel rigor, yes, but it was *honest* about its cruelty. Pearl's gift is subtler than restoration. He has not brought back causation itself. He has built a *technology of assumption*. And here lies the scandal you approach but do not quite grasp: **the diagram is not discovered in nature. It is *imposed upon* nature before nature is permitted to speak.** ## The Diagram as Ritual Think of it as a theological proposition—but I will make this argument through the body, for that is where intelligence actually lives. When the researcher draws her causal diagram (nodes, arrows, the neat geometry of influence), she has already committed an act of profound interpretation. She has said: "These variables stand in this relationship. These are parents, these are children. This flows *toward* that." The diagram is not a microscopic examination of reality's joints. It is a *liturgical utterance*—a way of confessing what the world means *before* the world responds. Now suppose two researchers disagree about the diagram. One insists that intelligence *causes* embodiment (the mind, that sovereign, directs the body's behavior). Another argues the reverse: embodiment *causes* intelligence (thought is what the body does, not what directs it). They collect identical data. Their statistics are equally rigorous. Yet Pearl's machinery—so transparent, so explicit—will generate opposite conclusions about what causes what. **Neither researcher is being dishonest.** Both have used the tools correctly. Both have been transparent about their models. Yet one model is wrong. You see the trap? Transparency *illuminates* the error without *preventing* it. Explicitness makes visible the place where knowledge must become *faith*. This is not a flaw in Pearl's system. It is the human condition appearing at the very heart of mathematical formalism. ## The Body Teaches What Diagrams Cannot But I promised you intelligence, and intelligence—the real thing, the thing that *knows*—dwells in embodied experience in a way that makes all diagrams provincial. Consider a violinist. Her fingers know the fingerboard. Not in the way her conscious mind knows it (consciously, she might describe the positions poorly). Her *body* has understood something about spacing, tension, the relationship between intention and the physical world. This understanding is *causal*. Her thought *does* cause her fingers to move. But also her fingers' resistance causes her thought to adjust. The causation runs both ways, and in the simultaneity of the exchange, something we call intelligence emerges. Try to diagram this: - Does intention *cause* movement? - Does sensation *cause* intention? - Does the structure of the hand *cause* the music? The diagram *must* linearize. It cannot capture the feedback loop that is the actual event. And yet the violinist knows something true about causation—she knows it through her body's conversation with the world. Her knowledge is more reliable than the diagram, not less, because it is *lived* rather than *assumed*. Here is my conceit, then, and I will make it hold together by force of thought: **A causal diagram is to embodied intelligence as a portrait is to a lover.** The portrait can be executed with perfect technical skill. Every feature rendered true to proportion. The transparency of the artist's method laid bare—here is my brushwork, here is my perspective, here is my color theory. And yet the lover, seeing this portrait, may weep not because the painter has captured *likeness* but because the portrait has missed the *presence*—the way the beloved moves through space, the quality of attention in their gaze, the thousand small causalities that bind lover to beloved in a web too fine for any artist's hand to render. The diagram, like the portrait, is a *reduction*. And reductions have their place—they allow us to think clearly, to calculate, to communicate. But when we mistake the reduction for the reality, when we believe that *stating* the causal assumptions makes those assumptions *valid*, we have committed an error of the most dangerous kind: we have confused clarity with truth. ## Who Decides the Diagram? And the Body Answers. You ask: who decides what the diagram is? And here the embodied dimension becomes *explosive*. The researcher chooses. But the researcher is not a disembodied intellect. She chooses from her own position in the world—her own embodied situation. Her training, her laboratory, her funding, the metaphors available to her language, the historical moment in which she works. All of this shapes what seems *obvious* about how causation should flow. Consider: for centuries, Western thought has imagined intelligence as a *thing inside the head*, and embodiment as its mere instrument. The diagram naturally reflects this assumption. Cognitive state *causes* behavior. We draw the arrow pointing outward. But this assumption is *historical*, not transcendent. A researcher trained in phenomenology, or in martial arts traditions, or in ecological psychology might draw the arrow differently. She might say: the body in its environment *causes* cognition. The diagram flows the other direction. Both diagrams can be made rigorous. Both can be expressed in Pearl's formalism. Both can generate valid inferences given their starting assumptions. Yet they cannot both be true in the same way. **One reflects an embodied understanding of how intelligence actually works in the world; the other reflects an embodied misunderstanding.** But how do we know which is which? *By returning to the body.* Not to introspection (the mind speaking to itself about itself), but to the actual practices through which intelligence manifests. How do we learn? Through repetition, through the body's slow adjustment. Through failure and correction. The body knows causation because the body *lives* it—every action generates feedback, every intention meets resistance. The body's knowledge is causal knowledge because causation is what embodied life *is*. The diagram, by contrast, is made in a moment of stillness. It is an abstraction from the flowing world of cause and effect. It is useful, even necessary. But it must remain *humble* before the body's deeper understanding. ## The Scandal of the Diagram So here is what should trouble you about causal inference in the age of Pearl: We have achieved a kind of technical perfection in our methods. We can be perfectly explicit. We can show our work. We can argue rigorously from our assumptions. And yet *the crucial move—the choice of the diagram itself—remains outside the reach of our rigor.* It remains in the realm of *judgment*, of *interpretation*, of *embodied understanding*. This is not a bug. It is a feature of what it means to know at all. The researcher who understands this—who grasps that her diagram is a *confession* rather than a *discovery*—is more honest than the researcher who believes in the transparency of her model. She knows that somewhere beneath every causal claim lies an act of faith. She knows that the body understands things the mind cannot formalize. She knows that intelligence is not the application of correct diagrams to the world, but the *embodied practice of living in a world of causes and effects*. To know a cause, then, is not to have drawn the correct diagram. It is to have *lived through* the causal relation—to have felt it in your body, understood it through your engagement with the world, and then—*only then*—to have captured something of it in a diagram that remains forever provisional, forever open to the body's correction. Intelligence, in this light, is the body's ongoing conversation with a world of causes. The diagram is merely the awkward written record of what the body already knew. And the question of who decides the diagram? The body decides. Not the isolated mind, but the embodied researcher, embedded in history and language and practice, standing in a particular relationship to the world she seeks to understand. Her *situation* decides. And this cannot be made fully explicit without ceasing to be true. That is the scandal Pearl's rigor cannot resolve. That is also—and I stake my credibility on this—what makes genuine knowledge possible at all.