DEEP READING · READ 73
Francisco Varela · Evan Thompson · Eleanor Rosch · 1991 (revised edition 2016)
You assume the world is already out there, fully specified, and that your brain's job is to photograph it accurately. This book says no: there is no ready-made world waiting to be known, and no ready-made "you" inside doing the knowing — world and mind are brought forth together, in the traffic between a body's actions and its perceptions. Follow that thought far enough and you hit something more unsettling: there is no self at the bottom either. The book's real subject is how to stand somewhere with no floor — and not fall apart.
The three authors are an unusual team. Francisco Varela (1946–2001) was a Chilean neurobiologist who, with Humberto Maturana, developed the theory of autopoiesis. Evan Thompson is a philosopher. Eleanor Rosch is the psychologist whose experiments demolished the classical view that categories are defined by necessary and sufficient conditions, replacing it with prototype theory: "bird" is not a checklist in your head but a graded cloud centred on a robin, thinning out toward the edges. Their 1991 book is the founding text of what became embodied and enactive cognitive science — the whole "4E" territory (embodied, embedded, enacted, extended) traces back through it — and it was the first serious scientific work to treat the Buddhist tradition of contemplative inquiry as an epistemological partner rather than exotic decoration. The 2016 revised edition carries new introductions by Thompson and Rosch (Varela died in 2001); both revise parts of the original framing, which is itself a sign that this is a living programme rather than a settled doctrine.
Three claims, and they form a chain:
You have to see what is being denied, or "enaction" is just a pretty word.
The default picture in cognitive science is representationalism: out there is an objective world with determinate properties (length, colour, dangerousness); the senses sample them; the brain builds an internal representation — think of a map standing in for terrain; thinking is computation over that map; action is the output. The picture reigned for decades because it can be written as a program: input → internal symbol manipulation → output. That is cognitivism.
Buried in it, the authors argue, is a premise nobody ever established — it was simply assumed. Call it the pregiven world: the properties of the world were already carved up before any creature arrived, and sit there waiting to be represented correctly or incorrectly. Accept that and only two roads remain, both dead ends. Either mind faithfully mirrors an outside world (objectivism) — in which case explain why different species inhabit visibly different worlds; or the world is a projection of mind (idealism, subjectivism) — which slides into "anything goes." The book calls this the chicken-and-egg problem: does the world specify the mind, or the mind the world? Neither answer works, because the question presupposes that the two arrive separately and only then negotiate.
The authors' third road is blunt: they were never separate. Everything that follows unpacks that.
The book's key term is enaction — from "enact," as a play is enacted: a thing that exists only in the doing of it. The authors' two-part definition, in paraphrase: first, perception consists in perceptually guided action; second, cognitive structures emerge from the recurrent sensorimotor patterns that let action be perceptually guided.
That is a mouthful for a simple loop. You move; the sensory flow changes; the change shapes how you move next; which changes the flow again. Go round that loop long enough and stable patterns settle out — and those patterns are your concepts. Concepts are not rubbings taken from the world; they grow out of what you do with it. The book's epigraph is Antonio Machado: Caminante, no hay camino, se hace camino al andar — "Wanderer, there is no path; the path is made by walking." That is the best one-line gloss on enaction there is.
The sharpest evidence is Held and Hein's 1963 kitten carousel. Two kittens raised in darkness are yoked to opposite ends of a rotating arm. The only difference: one walks under its own power, the other rides in a gondola and is towed by the first. The visual input is nearly identical — same patterns, same motion. Weeks later the active kitten behaves normally: it avoids a visual cliff, it puts out its paws to brace for a surface. The passive one behaves as if blind, and only recovers after some days of walking around itself. The conclusion is brutal: vision does not grow by having light poured in; it has to be woven by your own movement. Seeing is a kind of doing.
Schematic: above, the one-way street being dismantled; below, the loop replacing it — neither end of which comes first.
The biology under that loop is Varela and Maturana's autopoiesis ("self-producing"). What makes a living cell peculiar is that its metabolic network manufactures the very molecules it is made of, including the membrane that encloses it — and that membrane is what makes the metabolism possible. It is a circle producing itself, so its boundary is not drawn by anyone else; it is maintained by the thing itself. From this follows structural coupling: a cell does not "receive information" from its environment. The environment merely perturbs it, and the cell's own structure decides what the perturbation means. A sugar molecule counts as food not because sugar carries "foodness," but because a bacterium's metabolism makes it food. Meaning is not issued by the world; it is conferred by an organism in the act of keeping itself alive. A single cell is already enacting a world of its own.
One distinction is regularly missed: "embodiment" in the title is double. The body is both a biological mechanism open to third-person study, and a lived structure known from the inside — the posture, ache and breath you can feel right now. The book's wager is that both bodies must be held at once, with neither reduced to the other. Neural mechanism alone loses half; phenomenal description alone loses the other half.
Abstractions slide past you, so the authors spend a chapter on a worked example — and it is the book's most persuasive single stretch.
What is red? Objectivism answers: a physical property of the surface (some distribution of reflected wavelengths) that the eye measures. This has hard problems. The same patch under different illuminants reflects wildly different spectra, and you still see one red (colour constancy); conversely, two physically different spectra can look identical. Worse is the comparative biology: most humans are trichromats, with three cone types, while many birds are tetrachromats and see into the ultraviolet, and bees' sensitivity is shifted so far up that a flower which reads pure white to you carries a landing strip pointing at the nectar. If colour is a property the flower owns, whose flower is the real one? There is no answer — not because the measurement is unfinished, but because the question is malformed.
Then say colour is purely subjective, invented by the brain? That fails too. Basic colour vocabularies across the world's languages show strong regularities (Berlin and Kay's 1969 cross-linguistic survey found the order in which basic colour terms appear to be highly consistent, and speakers of very different languages point to nearly the same chip as the best red). Nor can you decide to see red as green. Colour is constrained by biology and yet is not a property of the object.
The authors' answer: colour is enacted — a dimension brought forth jointly by a visual system's structure and an environment's reflectance, coupled over evolutionary time. It is real relative to a kind of body, and that is not the same as arbitrary. The distinction matters and gets abused constantly: relative is not optional. "Relative to your body" is a fiercely binding phrase — your cone absorption curves are not up for negotiation. (An honest footnote: the chapter is not the last word. Colour realists such as Byrne and Hilbert still argue that colours are classes of surface reflectance; the debate is open.)
If the pregiven world has been dislodged on the perception side, it still has a harder redoubt on the evolutionary side: adaptationism. The environment poses well-defined problems — stay warm, catch prey — and selection finds optimal solutions, so organisms are answers to pre-set puzzles and the environment's properties remain pregiven after all.
The authors take up criticisms already circulating inside evolutionary biology and knock this out too, under the heading natural drift: selection eliminates what cannot survive; it does not pick the best. What remains only has to be good enough (satisficing rather than optimising), and within the space of good-enough, which path you take is largely history and accident. Three supports:
Put the three together and the environment stops being a well-posed exam, and the organism stops being the right answer. In staying alive, creatures are also defining what counts as their environment — earthworms remake soil, trees create the shade beneath them, humans build cities and agriculture, and their descendants are then selected in the world their ancestors altered. Species and environment specify each other: the evolutionary version of the conclusion reached on the perception side. Nothing is laid out in advance; there is only mutual shaping all the way down.
Here the book turns, from "the world isn't ready-made" to "neither are you" — which is where it was always heading.
It starts from a fissure inside science: the deeper cognitive science goes, the less it finds a unified self. There is no control room in the brain, only crowds of parallel, semi-autonomous, competing processes; a split-brain patient's two hemispheres give contradictory answers, and the talking hemisphere cheerfully invents reasons for what the other half did; our reports about why we act are routinely after-the-fact narration. Science finds the self to be an emergent, transient pattern — and the scientist goes home and lives as if it were bedrock. That gap, the authors argue, cannot be closed by theory, because it is not a theoretical problem. It is an experiential one. You have to go and look.
Look how? They bring in the Buddhist discipline of mindfulness/awareness (the two wings of the classical pairing: calm abiding, which steadies attention, and insight, which uses that steadiness to examine how experience actually unfolds). What they want is not Buddhist belief or metaphysics but its method — a trainable, transmissible, two-thousand-year-refined technology of first-person examination. Their verdict on phenomenology (Husserl's philosophical study of the structures of experience) turns on the same point: right target, but it stayed at the level of theoretical reflection, with no practice for actually steadying attention enough to see.
And then the finding: sit and look carefully and you will see thoughts, feelings, images and impulses arising and passing, but you will not find the one who has them. The early Buddhist analysis of mind, Abhidharma, breaks experience into the five skandhas — roughly body, the pleasant/unpleasant tone of feeling, recognition, habitual impulse, and awareness — and observes that no one of them is the self, and there is nothing outside them holding them.
One enormous misreading has to be blocked here: the conclusion is not "there is no self," but "there is no independent, unchanging self." The self is real the way a whirlpool is real. A whirlpool is unmistakably there; it can pull a boat under. But you cannot scoop out a whirlpool-thing: it is a shape the water is making, and it ends when the water stills — which does not make it an illusion. "I" is exactly such a self-sustaining pattern: real, but not real the way a stone is. And most of our suffering, the authors argue, comes from gripping a stone-like self and spending our lives defending, comparing and provisioning for it.
The moment you notice there is no floor, panic is the reflex. Borrowing a phrase from the philosopher Richard Bernstein, the authors call it the Cartesian anxiety: either there is an absolutely secure foundation — a fixed world, a fixed self, fixed truths — or there is darkness, chaos and nothing at all. One or the other. A great deal of Western thought is a swing between those poles: no objective ground, so jump to a subjective one; that collapses too, so declare everything constructed and nothing binding.
The either/or is itself the disease. Against it the authors set Madhyamaka, the "middle way" school founded by Nāgārjuna around the second century. Everything arises dependently — in reliance on conditions — and therefore lacks svabhāva, an independent essence of its own. That absence is what emptiness means. The crucial point: "empty" says "lacking independent essence," not "non-existent." Nāgārjuna is at pains to rule out the nihilist pole, and insists that emptiness itself must not be reified into a new foundation. Dependently arisen things are entirely real; they simply have no free-standing way of existing.
An analogy: a football match is completely real — it raises your pulse, empties the streets, costs managers their jobs — yet there is no object on the pitch that is "the match." It is held up by rules, players, ground, crowd and clock, and remove any of them and it is gone. To call the match empty is emphatically not to say nobody is playing. World and self are like that: not fake, just bottomless.
So the book lands here: groundlessness is not a predicament to be overcome, it is how things are; and only by being able to rest in it can a person actually relax. That is not a literary flourish — it follows from every step: perception has no pregiven world under it, evolution has no pregiven problem under it, the self has no unchanging core under it. Three lines meeting at one point.
The last concept gets skipped most often, and it is what separates this book from every merely deconstructive philosophy. If there is no ground, why not "then nothing matters"?
Two answers. First, nihilism is itself a symptom of not having let go of the self: behind "none of it means anything" stands a disappointed self that had demanded an absolute guarantee from the world, and has merely flipped from "there must be a floor" to "there is nothing" while staying inside the same anxiety. Genuinely loosening the grip does not produce coldness; it releases the energy that was going into defending the self, and that energy turns outward.
Second — the theme Varela later gave a whole small book to (Ethical Know-How, 1999) — real ethical action is a product of skill, not of reasoning. Watch an experienced nurse, or someone who has raised several children: in the decisive moment they do not consult a principle and derive a conclusion; they see what is needed and the body moves. It has the structure of any expertise: novices work from rules, experts respond immediately and in context. Ethics can therefore be trained, and the direction of training is the removal of what "how do I come out of this?" puts in the way — with less in the way, what the situation asks for becomes visible.
Strip away everything else and the book is a single chain:
Put differently, what the book establishes is this: the scientific question "what is mind?" and the practical question "how should I live?" are not two questions. Their separation is one of the consequences of the pregiven-world assumption.
1. "Cognition is not the representation of a pregiven world by a pregiven mind but is rather the enactment of a world and a mind on the basis of a history of the actions that a being performs in the world." (paraphrased) — the whole book in a sentence.
2. "Wanderer, there is no path; the path is made by walking." (Machado) You do not follow a path that was already there — walking is what lays it down, and cognition works the same way.
3. To test whether a theory of mind is smuggling in a pregiven world, ask it one question: were these properties already carved up before any creature arrived?
4. Two kittens receive the same light; the one that walks can see, the one that is towed behaves as if blind. Seeing is a kind of doing: perception has to be woven by your own movement.
5. Colour is neither on the flower nor freestanding in your head — the ultraviolet landing strip the bee sees and the white blossom you see are both real. Real relative to a body is not the same as anything goes.
6. Selection removes the unviable; it does not pick the optimal. What survives only has to be good enough, and the rest is history and accident. Concluding that an organ was selected for the job it does well is mistaking the mosaics in St Mark's for the reason the spandrels are there.
7. Meaning is not issued by the world; it is conferred by an organism keeping itself alive — sugar is food because something has to have it to keep its own circle turning.
8. Look inward carefully and you find thoughts, feelings, impulses, but never the one who has them — which is not "I don't exist" but "I am a whirlpool": real, forceful, and with no core you can lift out.
9. "Either an absolute foundation or total nihilism" — that either/or, the Cartesian anxiety, is itself the illness. Dependently arisen, therefore empty; empty, therefore not nothing. No bottom does not mean no world.
10. So groundlessness is not a hole to be patched but the situation we were always in; and the ethics that grows from it is not a set of rules but a skill — with less of yourself in the way, what this moment asks of you becomes visible.