Wilfrid Sellars

London, March 1956. An American professor from Minnesota stands before a University of London audience to deliver three special lectures with a strange title: "The Myth of the Given." He speaks in long, coiling sentences, without notes, the argument circling back on itself like a symphony returning to a theme. By the time he finishes, he has taken the founding assumption of three hundred years of empiricism — that knowledge rests on a foundation of pure experience — and dismantled it from the inside, brick by brick.

Almost nobody outside a lecture hall noticed. Almost nobody outside philosophy ever has.

Ask a stranger on a bus to name a twentieth-century philosopher and you might hear Sartre, Russell, maybe Wittgenstein. You will not hear "Wilfrid Sellars." Yet ask professional philosophers to rank the century's most important minds and Sellars appears again and again near the very top. One eminent commentator called those London lectures, published as Empiricism and the Philosophy of Mind, the most influential paper of its era. Like Bernard Lonergan a few chapters ago, Sellars is a philosopher's philosopher: little known outside the profession, revered within it, and quietly present inside arguments still to come in this book.

Why the gap? Partly the prose — we will be honest about that. Partly the problem: Sellars asked not "what should I do?" or "does God exist?" but something more abstract and, once you see it, more unsettling. When the world presses on your senses, what turns that pressure into knowledge? When does a wired-up creature reacting to light become a person who knows something and can be asked to defend it?

His answer changed epistemology — the theory of knowledge — permanently. And it began at a dinner table in Michigan, where the food came with philosophy whether you wanted it or not.

The Life

Wilfrid Stalker Sellars was born in Ann Arbor, Michigan, on May 20, 1912, into the only major father-and-son pair in the modern stretch of this book. His father, Roy Wood Sellars, was a philosophy professor at the University of Michigan and a leader of the movement called critical realism — the view that we genuinely know a mind-independent world, but always through the mediation of our ideas, never by naked contact. Wilfrid grew up marinated in this. The dinner-table arguments of his childhood became a lifelong dialectic: his mature philosophy is, among other things, a decades-long conversation with his father's realism — accepting the realist conclusion, rebuilding the argument from the ground up.

The schooling that followed was almost absurdly cosmopolitan. The family's stays abroad put the teenage Sellars in Paris lycées, where he did his adolescent reading of Marx in French and absorbed a rigor of argument he never lost. A stretch in Munich around 1930–31 gave him a street-level view of the Nazi movement on its way to power — a lesson in what happens when reasons give way to force. Back home he took his degree at Michigan, added a master's at Buffalo, and then won a Rhodes Scholarship to Oriel College, Oxford, where he read Philosophy, Politics and Economics and took first-class honors. Oxford steeped him in the long shadow of its nineteenth-century idealists — T. H. Green's ghost still haunted the syllabus — and in the realists who fought that ghost; Sellars would spend his career, in a sense, refereeing their quarrel. He began a doctorate at Oxford on Kant, moved to Harvard, passed his exams — and never finished a dissertation anywhere. The man who would reshape American philosophy did it, wryly enough, as a permanent ABD: all but dissertation, all but stoppable.

His teaching career began at the University of Iowa in 1938. The war interrupted it: Sellars served in the Navy from 1943, doing intelligence work on anti-submarine warfare — pattern-hunting of a very practical kind. In 1946 he moved to the University of Minnesota, and there the institution-builder emerged. With Herbert Feigl — the Vienna Circle emigré we met in Moritz Schlick's chapter, now his close ally — he founded Philosophical Studies in 1949, the first journal devoted specifically to analytic philosophy, and helped build the Minnesota Center for Philosophy of Science into a powerhouse. The alliance matters: Sellars worked with the logical empiricists, published in their venues, shared their scientific temper — and spent his whole career transforming their empiricism from within, like a renovator who loves the house too much to let it stand as built.

The 1956 London lectures made him famous among philosophers. After a brief period at Yale, he moved in 1963 to the University of Pittsburgh — and built the department that became, and remains, the capital of Sellarsian philosophy. That same year his collected essays appeared as Science, Perception and Reality; Oxford's Locke Lectures and the Carus Lectures followed. At Pittsburgh his lectures were events: the flowing, blackboard-free delivery, the pipe, the argument unspooling from memory in perfect paragraphs — philosophy happening, not being reported.

His writing was another matter, and honesty requires a line. Sellars' prose is legendarily difficult — jungle-dense, self-referring, thick with invented terminology; "Sellarsian" became professional shorthand for a style. But the difficulty is load-bearing, not decorative: he was saying genuinely new things, and the reader who persists finds a system of rare depth. The profession's verdict has been to keep rereading him, which is its own review.

The later decades had their shadows: Sellars struggled with drinking, and his last years were a diminuendo of the great mid-century output. He kept teaching at Pittsburgh nearly to the end, and died there on July 2, 1989. By then two of his colleagues' students and successors — Robert Brandom and John McDowell — were turning his ideas into the next generation's battleground, and the "Pittsburgh School" had become a name every philosopher knew.

The Big Ideas

The Myth of the Given: the foundation that wasn't there. Start with a dream the empiricists dreamed for three centuries. From Hume through Schlick's Vienna Circle to Ayer's sense-data, empiricism was built on one comforting picture: whatever else may be doubtful, experience itself delivers certainties. Before you interpret anything, before you learn any words, the world simply gives you something — a red patch in your visual field, a stab of pain — and your awareness of it is self-authenticating. It presupposes no other knowledge. It cannot be wrong. On this bedrock of givenness, the empiricists believed, the whole tower of human knowledge is built. Schlick called the bedrock moments Konstatierungen — confirmations; Ayer called them sense-data; the common creed was that something, at the bottom, is just given.

Sellars called this "the Myth of the Given," and his attack on it is the chapter's core, so let's take it slowly.

The given, to do its job, has to satisfy two demands at once. First, it must be independent of concepts — available prior to any learning or language. It has to be, since it's supposed to be the foundation everything else (including concepts) rests on, and since infants and animals sense the world without concepts at all. Second, it must be able to justify beliefs — to serve as evidence. That's what a foundation is for.

Now the dilemma, and it is one of the cleanest in modern philosophy. Justification is a relation between conceptual items — things with content, things that can be true or false, things that can stand in logical relations like supporting and contradicting. A mere sensation, concept-free, is just an event, like a knot in your stomach or a flash on a photodiode. It happens; it doesn't claim anything. So: if the given is truly free of concepts, it can't justify anything — a wordless twinge can't be evidence any more than a rock can be an argument. But if the given is conceptual enough to justify — if sensing a red patch can support "this is red" — then it already presupposes the concept red, which had to be learned, which means it presupposes other knowledge and isn't the presuppositionless foundation it was hired to be. Nothing can play both roles. Sensing a red patch justifies "this is red" only for a creature that already has "red" — and the regress the given was invented to stop reopens behind it.

At the center of the argument sits the distinction Sellars made famous, in the one sentence of his that every philosopher can quote: in characterizing an episode as one of knowing, he wrote, we are not giving an empirical description of it; we are placing it "in the logical space of reasons, of justifying and being able to justify what one says." Unpack that. There are two ways to talk about what happens when light hits your eyes and you say "red." One is the space of causes: photons, retinas, neurons firing — the wiring. The other is the space of reasons: claims, evidence, entitlement — the game of asking "how do you know?" and being answerable. Both descriptions are legitimate. But they are different kinds of description, and the myth of the given is, at bottom, the attempt to smuggle an item from the causal story into the justifying story without paying the toll. Causation isn't justification. The wiring explains why you believe; it cannot, by itself, make you entitled to believe.

The consequences are large. If nothing is simply given, then knowledge has no self-certifying foundation — and yet Sellars was no skeptic. Empirical knowledge is rational, he argued, not because it rests on bedrock but because it is a self-correcting enterprise, one that "can put any claim in jeopardy, though not all at once." Any belief can be revised — but only while standing on the others. You have met this picture twice before, and the convergence is the striking thing: it is Neurath's boat from the Vienna Circle chapter, rebuilt plank by plank at sea, and it is Quine's web of belief. But Sellars arrived by a different road. Quine got there through Duhem and logic; Sellars got there through Kant — through the insight that experience without concepts is blind, now stripped of Kant's fixed categories and made social and revisable. Three routes, one destination: knowledge is holist, communal, and foundationless, all the way down.

Has the myth stayed dead? Honest scoreboard: its death has been announced repeatedly and contested repeatedly. A lively literature on "nonconceptual content" argues that perception does carry fine-grained content beneath our concepts; phenomenal conservatives argue that seemings confer at least some justification just by seeming. The defenders of givenness are fewer, but they are not gone — which is roughly what Sellars' own picture of inquiry would predict.

The manifest and scientific images: how does everything hang together? Sellars gave philosophy its best modern job description in the opening of his great framework essay, "Philosophy and the Scientific Image of Man": the aim of philosophy is to understand how things, "in the broadest possible sense of the term hang together in the broadest possible sense of the term." Things: not just atoms but numbers, duties, deaths, persons. Hang together: one coherent picture, not a filing cabinet of disconnected subjects.

The problem is that we currently have two pictures. The astronomer Arthur Eddington famously described his two tables: the solid, brown, familiar table he wrote on, and the physicist's table — a cloud of particles, almost entirely empty space. Sellars generalized the puzzle. The manifest image is the framework of persons: a world of colored objects, purposes, reasons, promises — the world as encountered by common sense. Don't hear "manifest" as "naive": Sellars insisted this image is itself a sophisticated achievement, refined over generations, with its own internal standards. The scientific image is the framework built by postulation: imperceptible entities — particles, fields, neurons — introduced to explain why the perceptible world behaves as it does. The two images clash, and not politely. The manifest table is solid and brown; the scientific table is mostly void, and nothing in physics is brown.

Which wins? Here Sellars planted his most provocative flag, in the passage philosophers call the scientia mensura: in the dimension of describing and explaining the world, "science is the measure of all things, of what is that it is, and of what is not that it is not." Readers of this book's early chapters will hear the deliberate echo — Protagoras, back in chapter 15, made man the measure of all things; Sellars inverted him. When the question is what exists and why it happens, the final court is not common sense but the sciences' self-correcting enterprise.

But read the fine print: in the dimension of describing and explaining. That clause is the whole game. The manifest image is not, at its core, a rival description to be outvoted; it is the framework of persons — of intentions, standards, oughts — and the space of reasons lives there. So the task, Sellars argued, is not to let one image devour the other but to achieve a stereoscopic view: like your two eyes fusing two flat pictures into one scene with depth, philosophy must join the scientific description of what we are with the manifest framework of what we owe, intend, and mean. Persons and norms survive — not as ghostly extra objects, but as the indispensable dimension of prescription laid over the scientific description.

Does the fusion work? He never finished it, and his heirs split over it. The Churchlands — his radicalized students — argued the scientific image should simply replace talk of beliefs and desires, as chemistry replaced alchemy. John McDowell, from the other side, has argued the manifest world of reasons and values is more real and less optional than Sellars' science-first rhetoric allows. The stereoscope remains philosophy's unfinished homework, and Sellars would have said so.

Psychological nominalism and the Myth of Jones: where "inner life" comes from. The second half of Empiricism and the Philosophy of Mind is under-read and wonderful: having demolished the given, Sellars rebuilds the mind — with a story. Imagine our ancestors as what he playfully called Ryleans (the homage to Gilbert Ryle, chapter 223, is deliberate): they have a rich language, but only for public things — objects, behavior, dispositions. No talk of thoughts, no talk of experiences. Nothing "inner." Now enter a genius, whom Sellars names Jones. Jones notices that his fellows behave intelligently even when silent, and he does what scientists do: he postulates unobservables. Suppose, says Jones, that inside people there are episodes — call them thoughts — that are like sayings, only unsaid: inner speech, modeled on outer speech. Suddenly silent intelligence is explicable. Later Jones postulates impressions — inner states modeled on objects — to explain why people sometimes misperceive: why a red-lit white wall looks red.

Now the payoff, and it is a beautiful reversal. Jones teaches others to apply his theory to themselves — first by observing their own behavior, then, with training, directly: people learn to report "I'm thinking about lunch" without self-observation, the way you learn to report "that's red" without inference. Our famous "privileged access" to our own minds turns out to be real but learned — a social role acquired for theoretical terms — not a private theater we were born watching. Descartes' inner stage, dismantled in chapter 127's aftermath by Ryle's mockery, is here dismantled constructively: the inner is real, but it is constituted through the outer. Notice the convergence with Wittgenstein's private-language argument — the same conclusion, that inner talk depends on public practice, reached by construction rather than therapy.

The frame around the story is what Sellars called psychological nominalism: the thesis that all awareness of kinds and sorts — all classifying, all awareness that something is such-and-such — is a linguistic affair. Even our thoughts are about things because they inherit the semantic properties of the speech they're modeled on. And the legacy of this one thought-experiment is astonishing: it is the charter of functionalism — the now-dominant view that mental states are defined by their roles, not their stuff (worked out formally with the sentence-method of Frank Ramsey, chapter 227); it seeded the "theory-theory" in developmental psychology, on which children learn minds the way scientists learn theories; and it set up Robert Brandom's inferentialism — meaning as inferential role — the Pittsburgh School's flagship.

Norms, rules, and the synthesis: Kant naturalized. Step back and the spine of the whole system shows: Sellars was the twentieth century's great naturalizer of Kant. From Kant he took the lesson that concepts are rules — to have the concept red is to have mastered a norm for using it, not to have collected private samples — and that the "ought" of the space of reasons cannot be reduced to the "is" of the space of causes. From modern science he took the conviction that persons are, without remainder, natural beings. Holding both at once — normativity irreducible, humanity fully natural — is the whole Sellarsian balancing act: scientific realism with a Kantian face. Meaning, on his account, is functional classification: to say what a word means is to say what role it plays, a role that different noises in different languages can share (his notorious dot-quote machinery — •red• — is just notation for "whatever plays the role our word 'red' plays"; the idea matters, the machinery is for connoisseurs). Truth he treated as correct assertability within the rules. Beneath the norms he kept a doctrine of picturing — a causal, map-like correspondence between language and world — his most contested idea, and flagged as such. And in his late ethics he extended the same move to morality: the moral point of view as we-intentions, the shared intentionality of a community of rational beings.

That community is also his legacy. The Pittsburgh School's two great heirs read him in opposite directions — the profession jokes about "right-wing" and "left-wing" Sellarsians. The right wing (Brandom, and the Churchlands beyond him) presses the scientific image and the scientia mensura; the left wing (McDowell) presses the space of reasons and the manifest image. That family quarrel — inferentialism versus conceptual realism, replacement versus re-enchantment — is arguably the most productive debate in contemporary philosophy, and every side of it is playing with Sellars' pieces.

Before & After

Backward, the threads of this whole modern section gather here. From Hume (chapter 151) Sellars inherited empiricism itself — and transformed it from within rather than abandoning it. From Kant (chapter 155) he took the deep structure: concepts as rules, experience as always-already conceptual, now naturalized and socialized — Kant without the fixed categories. He liked to hint at a further step, describing his masterpiece as "incipient Méditations Hégéliennes" — a nod to Hegel (chapter 164) and the social character of spirit, in deliberate contrast to Descartes' solitary meditations (chapter 127), whose inner theater the Myth of Jones rebuilds as a public achievement. From Peirce (chapter 180) comes the self-correcting community of inquiry; from Schlick (chapter 201) and the Vienna Circle, via his ally Feigl, the logical-empiricist home he spent a career renovating; from Carnap (chapter 214), the apprenticeship in syntax and semantics; from Ryle (chapter 223), the behaviorist ancestors his Jones outgrows; from Wittgenstein (chapter 208), the private-language conclusion reached by another road. With Quine (chapter 236) the comparison is one sharp line: both demolished foundations, but Quine dissolved meaning into behavior while Sellars normativized it into rules. Ayer (chapter 239) supplied the sense-data theory that EPM buried deepest. And Austin (chapter 240), whom we just left, is the perfect hinge: two reformers born a year apart, both auditing empiricism — Austin its language, Sellars its epistemology; the ordinary road and the theoretical road out of the given.

Forward: Richard Rorty (chapter 271) will take the myth of the given as the first crack in philosophy's "mirror of nature" and use it to challenge the whole enterprise — watch for it. Brandom and McDowell carry the Pittsburgh inheritance into the present; the Churchlands carry the eliminativist dare; functionalism carries the Myth of Jones into cognitive science; and the nonconceptual-content wars keep the given's ghost in play. Next, though, comes Sellars' exact contemporary Alan Turing (chapter 242) — born the same year, 1912 — who asked the question Sellars' distinction makes urgent: can a machine think?

Why It Still Matters

Start with the argument you are probably having already. When a large language model produces a fluent answer, is it justifying its claims or merely caused to emit them? That is Sellars' distinction, verbatim, applied to 2026: the space of reasons versus pattern-causation. Whether an AI system can genuinely occupy the space of reasons — be answerable for what it says, not just statistically apt — is now a live question in labs and law courts, and no one can discuss it precisely without Sellars' vocabulary. It is the sharpest single application of a 1956 lecture to the technology in your pocket.

The two images clash daily in courtrooms. When a defense lawyer argues "his brain made him do it," she is playing the scientific image against the manifest one — causation against responsibility. Sellars' stereoscopic program says the deep mistake is thinking one description must simply cancel the other: neuroscience describes what we are; responsibility lives in the framework of persons; the hard, unfinished work is holding them together. Judges now do Sellarsian philosophy whether they know it or not.

His picture of knowledge as a self-correcting social enterprise — no claim sacred, but not all challengeable at once — is the best available description of how science actually earns trust, replication crises and all. It also explains what goes wrong in a "do your own research" culture: you can audit any plank of the boat, but only while standing on the others; step off entirely and you're not skeptical, just at sea.

And the Myth of Jones quietly underwrites modern psychology: the idea that self-knowledge is a learned skill, not an infallible inner gaze, is the working assumption of cognitive therapy — you can be wrong about your own feelings, and better trained in reporting them. Introspection as an achievement rather than a birthright: that is Sellars, in the clinic.

Over to You

An infant and a thermostat both respond reliably to temperature. At what point — and by what magic — does a responding system become something that knows it's cold? What has to be added?

If your "privileged access" to your own mind is a learned social skill, as the Myth of Jones suggests, then a community could in principle train it differently. Does that liberate you or unsettle you?

The manifest table is solid; the scientific table is mostly empty space. Which one are you touching right now — and can you actually believe both answers at once, or do you just alternate?

Where to Go Next

Start with the source: Wilfrid Sellars, Empiricism and the Philosophy of Mind — the Harvard edition, with Richard Rorty's introduction and Robert Brandom's study guide, which together act as sherpa gear for the famous difficulty.

A modern way in: Willem A. deVries, Wilfrid Sellars — the clearest full guide to the system, written by the profession's most patient Sellars explainer.

The thread forward: Sellars asked what separates a knower from a mere responder; next comes his exact contemporary Alan Turing, who built the mathematics of the responders — the universal machine — and asked, with a directness philosophy never forgave or forgot, whether a machine might cross that line.