# On Causation and the Body: A Meditation in the Manner of Donne ## I. The Conceit Established Intelligence is a body that thinks it is a diagram, and the diagram is a body that refuses to think itself at all. Consider what we have done: we expelled *cause* from the kingdom of knowledge as though it were a courtier caught in sedition, and for one hundred years we called this banishment *rigor*—as though the absence of a thing were the same as the presence of discipline. We measured correlations like misers counting coins that were never ours to begin with, and we called this *science*. The body moved; the mind recorded; the diagram remained forever innocent, forever sterile, forever *safe*. Then Pearl arrived—that technician of the real—and said: No. The diagram must *speak* first. You must assume the causal skeleton before the data can even dress itself. You must commit to the architecture of cause before you measure a single thing. And we celebrated this as liberation. But Pearl's restoration was not an escape from assumption; it was a *formalization* of it. It was turning the hidden scaffolding of all thinking into an *explicit* lattice—a diagram you must lay down like a prayer before you can read what God has written in the world. The transparency is perfect. The validity remains uncertain. You can see exactly how you are wrong. This is not comfort. This is clarity without redemption. ## II. The Body as the Banished Thing Here lies the scandal that no one speaks of plainly: **the body is the first thing expelled from causal reasoning, and the body is where causation actually lives.** When we ask "what causes intelligence?" we have already committed a series of murders: We have murdered the hand that writes the diagram. We have murdered the eye that reads it. We have murdered the viscera that feel whether the diagram is *true*, not whether it is *formally valid*. The body does not think in DAGs—directed acyclic graphs, those bloodless flowcharts of influence. The body thinks in *entanglement*. My hand does not wait to receive a causal signal from my brain before it knows the coffee is too hot; the hand and the heat and the nerve and the impulse are *one act*, not a chain of causes executing in sequence. The body is not a diagram. The body is what refuses to become a diagram. When we speak of *embodied* cognition, we are speaking of something prior to causation itself: we are speaking of the **simultaneity** of knowing and being. The body does not *receive* information about the world. The body is the world's way of *touching itself*. Yet every causal diagram separates what the body has joined. It draws a line between the knower and the known. It insists that one must *cause* the other, as though they were ever separate things. ## III. The Assumption You Must Make Pearl is right to demand transparency about assumptions. But he has not solved the deepest problem; he has merely *named* it. Here is what he requires: **You must draw your diagram before you know what caused what.** This is not an empirical procedure. This is an *act of faith*. You look at intelligence and you decide: Is it caused by genes? Environment? The interaction between them? The timing of neural development? The structure of language available to the child? The particular way that body meets world in the first six months of life? You must choose your skeleton before you can measure anything. You must assume the **direction** of influence—that A causes B and not B causes A—in the very moment when the causal structure is most obscure. And here is the terrible beauty of it: **You are always right to make this assumption, and you can never know if you are right.** The diagram is not a discovery. It is an *interpretation*. It is a way of carving reality at its joints—but reality has infinitely many joints, and you have chosen which ones to cut. Who decides? The researcher, naturally. The researcher with her prior knowledge, her disciplinary prejudices, her funding, her hope, her fear that the diagram will not work. The researcher brings her body to the work—her embodied understanding of what intelligence *feels* like from the inside—and then she abstracts it away and calls it rigor. This is not dishonesty. It is the unavoidable human condition of knowledge-making. ## IV. The Body Decides, But Speaks in Whispers Here is what embodied cognition teaches us if we listen carefully: **The body knows causation before the mind can diagram it.** When you learn to play the violin, you do not first internalize a causal model of how finger position produces pitch, how pressure produces tone. Your body *becomes* the diagram. Your hand and the string and the wood and the air and your ear are all causing each other simultaneously in ways no DAG could ever capture. The causation is real. It is also *pre-diagrammatic*. This is where Pearl's framework becomes provincial. It is magnificent for questions like "Does smoking cause cancer?" (where the causal structure is relatively stable, where bodies are sufficiently similar, where the outcome is unambiguous). But for intelligence? For the question of how a body becomes a mind? The body *already knows* whether your causal diagram is true. A child develops. A mind emerges. A person becomes intelligent. And at every moment, the body *feels* what is working and what is not. The embodied researcher—the one who has *lived* intelligence, who has struggled with it, who has felt its absence and its presence—brings that knowledge to the work. It lives in her hands, in her intuitions, in her refusals and her insistences. But then she must *translate* this knowing into a diagram. And in that translation, something dies. The question is not whether to include the body in our causal models. The question is whether a causal model can ever be true to the body's way of knowing. ## V. Who Decides? The Honest Answer When you ask "who decides what diagram is attached to the research?" you are asking the question that science has always refused to answer directly. The answer is: *the embodied researcher, making an assumption that cannot be empirically justified, betting her reputation and her time on a picture of causation that might be entirely wrong.* She decides by: - Her training (which is social, not empirical) - Her intuition (which is embodied, not abstract) - Her hopes (which are personal, not universal) - Her judgment (which is particular, not generalizable) And then—and this is the crucial move—she *makes her assumption transparent*. She draws the diagram plainly. She says: "Here is what I am assuming about how intelligence is caused." She does not hide her skeleton in footnotes. She does not pretend the diagram emerged from pure reason. This does not make the diagram *true*. It makes the diagram *honest*. And honesty is not the same as validity. A researcher can be completely transparent about a model that is completely wrong. But the transparency means that another researcher—another body, another mind, another set of embodied intuitions—can see exactly where to push back. Can see exactly which assumption to question. Can propose a different diagram, a different carving of reality at different joints. The diagram is not a fact. It is an *invitation to disagreement*. ## VI. The Embodied Dimension: Where Causation Begins Intelligence, understood from the embodied perspective, is not a thing that is *caused*. It is a *way of being in the world* that emerges from the body's constant conversation with its environment. Your neurons do not cause your intelligence; they *embody* it. Your environment does not cause your intelligence; it *invites* it forth. Your genes do not cause your intelligence; they *constrain and enable* it. The causal diagram separates these things into vectors and nodes. But embodied intelligence knows them as *aspects of a single unfolding*. A child reaches for a toy. The reaching is not caused by prior mental events; it *is* the beginning of mind. The hand knows before the brain knows. The body understands causation by *living* it, not by diagramming it. When we ask "what is the cause of intelligence?" we have already made an embodied choice—the choice to abstract away from the body, to treat it as a mere vessel for the real business of causation. We have expelled the body from the diagram and then wondered why our diagrams feel so bloodless, so inhuman, so far from the way intelligence actually *lives* in the world. Pearl's tools are provably correct *given a diagram you must assume*. But the diagram itself—the skeleton before the data—emerges from embodied understanding. It comes from bodies thinking, bodies feeling, bodies knowing the world. To be rigorous about causation, we must be rigorous about the body. We must ask: **What embodied understanding am I encoding in this diagram? What am I assuming about how a body can know and cause and be known?** The answer will be contested. It should be. Because intelligence is not a problem with a diagram. It is a problem that only a body can solve, and every body solves it differently. --- *In the end, Pearl gave us transparency without salvation. He gave us the skeleton without the flesh. The diagram is honest, but the honesty reveals only how much we do not—and cannot—know about what we are assuming when we assume that intelligence can be diagrammed at all.*