Ryle's The Concept of Mind: The Ghost in the Machine

Michael BrenndoerferUpdated June 24, 202513 min read

Gilbert Ryle uses category mistakes and the distinction between knowing how and knowing that to challenge the Cartesian picture of mind.

PhilosophyAll chapters
Reading tools Notes & comments
Notes & comments
Part 7 of 12Mind, Self, and Experience
Chapter 81Mind and category mistakes › What Did the Visitor Miss?
Reading the full chapter · 8 sections · 13 min8 sections · 13 min

All 8 sections · 13 min read

Mind and category mistakes › What Did the Visitor Miss?Full chapter · Sections in this chapter1 / 8 ↓

What Did the Visitor Miss?

A visitor to Oxford or Cambridge has seen the colleges, libraries, playing fields, laboratories, and offices. After the tour he asks where the university is. He has understood the guide to be listing institutions of different kinds and expects one more stop, reserved for the university itself. Pointing him toward another building would leave his misunderstanding intact. What he needs to understand is how the places he has seen belong to and are organized as a university.

Gilbert Ryle uses this example in the opening chapter of The Concept of Mind (1949). His target is a picture that makes even familiar descriptions of people sound puzzling. We say that a colleague understands a question, that a student hesitated before answering, or that a friend acted tactfully. A philosophical theory can take such descriptions as indirect reports of a second, private career running alongside the person's public career. It then has to explain how we know what is happening in that second career and how its events can affect the first. Ryle asks whether those puzzles arise from the way the theory initially sorts the descriptions. The university example is an invitation to examine that possibility in a case where the mistake is easier to see. Chapter I, §2, “The Absurdity of the Official Doctrine”

The Two Careers of the Official Doctrine

Ryle calls his target the official doctrine and associates it chiefly with Descartes. The distinction matters. The opening chapter sets out a philosophical picture and tests its consequences. You can follow the test without assuming that every detail of Ryle's reconstruction is a fair account of Descartes.

On this picture a person has a public bodily history. Bodies occupy space, move, and are available to observation. The person also has a mental history, the episodes of which are available directly to that person alone. Other people can at best infer a mental history from the movements and utterances they observe. Describing a neighbour as anxious then seems to require a conjecture about a hidden episode. The neighbour's access to that episode is supposed to be of a different order from ours. Ryle calls the picture the ghost in the machine. Chapter I, §§1–2

A question about other minds follows. Suppose I infer that my neighbour is concerned because she keeps returning to a problem and asks questions about its consequences. Under the doctrine Ryle describes, her actions are evidence for an episode that I can never consult directly. Analogies with my own experience may make my conjecture plausible, but I cannot inspect her inner stream to see whether the conjecture succeeded. The philosophical picture makes that restriction part of the meaning of psychological description, even for the teachers, friends, and critics who already use those descriptions and sometimes correct them. Ryle takes the gap between the picture and these ordinary practices seriously. It is a reason to investigate how the descriptions actually work. Chapter I, §1, “The Official Doctrine”

The doctrine faces a different problem when an intention appears to issue in a bodily action. The two careers have been described in terms of different kinds of processes. An account of how one process brings about another must explain why the connection is intelligible. In Ryle's reconstruction, theorists have treated the mental and the bodily as competing kinds of cause from the beginning. The later interaction puzzle thus also provides a reason to ask what licensed the initial classification. Chapter I, §3, “The Origin of the Category-Mistake”

What a Category Mistake Changes

The visitor to the university could have made a mundane error. He might have asked for the library and been pointed toward the registrar's office. A map could have corrected that mistake because the library is the kind of place for which that question makes sense. But his request to see the university as one more institution alongside all the others has a different structure. He has placed the university in the same list as its constituent establishments. Giving him more addresses would not explain how those establishments are related.

Ryle brings out another version of the confusion with team spirit. Someone watching a cricket match might understand what the bowlers, batters, fielders, and officials do, yet ask whose separate job it is to supply the team's spirit. An explanation might draw the spectator's attention to how people carry out their own tasks, support one another, and respond when a teammate makes an error. Those observations can matter even when they fail to identify a further player. In the example, the trouble lies in the spectator's expectation about what counts as an answer. Chapter I, §2

Category Mistake

A category mistake occurs when a question or description presupposes the wrong logical type. Searching for team spirit as a separate position in the lineup is one kind of mistake. Misidentifying a player who actually occupies a position is another. Learning a player's name could settle the second case while leaving the first confusion untouched.

Applied to the mind, the proposal needs care. The examples show that words can describe real features of a situation without naming a separate item of the kind a puzzled spectator is looking for. They also show how treating a description as the name of a separate item can generate an apparently deep mystery about its whereabouts. By themselves they cannot establish that every mental description behaves like university or team spirit. Ryle has to make the case for his specific target by studying psychological descriptions and the circumstances in which people use them. This is where his treatment of intelligent action earns its place in the book.

Knowing How and Knowing That

Imagine a chess learner who can accurately repeat the rules and has memorized some useful maxims. Even so, she blunders a piece when the position takes a form she has not seen before. Her opponent notices a threat and adapts to it. Both may be able to state the rule that governs a move, but the same ability to reproduce it tells us little about the difference in their play. The learner still needs to recognize when a maxim applies, which considerations to weigh, and when an exception matters.

Knowing that describes grasp of a fact or proposition. Knowing how describes a competence. The difference is plain when a teacher asks a pupil to recite a rule and then asks the pupil to use it to solve an unfamiliar problem. It would be equally misleading to assume that the competent pupil never consults a rule or thinks through a problem in words. A lesson can improve a student's practice by giving them a rule to use. The question is whether the intelligent application of a rule can always be explained by a prior intelligent operation of consulting one. Chapter II, §§2–3, “Intelligence and Intellect” and “Knowing How and Knowing That”

Ryle's regress tests precisely that proposal. Suppose a chess move is intelligent only because the player first intelligently considers what a rule says to do. The first consideration is itself something the player must do intelligently. On the proposed account she therefore first has to consider a rule for considering the rule. That prior consideration is subject to the same demand. The explanation postpones what makes any application intelligent at every step. Ryle's conclusion concerns the requirement that an intelligent operation must always inherit its intelligence from a separate prior theoretical operation. He can grant that deliberation sometimes contributes to a good move and that deliberation itself can be clever or clumsy. Chapter II, §3

There is a simple way to see why merely inserting a thought does so little work. Picture a student who correctly thinks, “Look for a threat,” then checks only the moves that suit the attack she already wants to make. Her maxim is sensible. The way she used it was poor. Something similar can happen in the other direction: a player who has practiced a pattern spots the overlooked defender immediately and explains the move only after being asked. We can distinguish the quality of the play from the timing of an explicit explanation. The distinction leaves room for the thought itself to be assessed, both as planning and as a subsequent explanation. An explanation after the move can also show the teacher whether the student noticed the defender or arrived at the best move by chance.

Recognizing Skill Without Watching a Second Performance

Consider Ryle's clown. An ordinary stumble and a practiced fall can resemble one another closely. The practiced fall may hinge on where the spectators are looking, when they expect the performer to regain balance, and whether the performer can fall again safely on cue. We learn to describe what the clown did by attending to those circumstances and to what the performer is capable of doing. A camera can record the stumble. The competence is not a separate event of the sort that another camera angle could reveal. The recorded fall can be an exercise of it. Chapter II, §4, “The Motives of the Intellectualist Legend”

This also explains why a single observation can mislead. A novice may accidentally do the right thing. A practiced performer may make a mistake. What happened matters, and so do the reasons for it, the conditions under which it was done, and how the person handles a change in those conditions. Suppose a learner tests a proof on two simple examples and finds that it seems to work. A teacher can ask the learner to produce a case in which one premise fails, or to explain why the next step still follows when a diagram is redrawn. Those questions help distinguish remembering a route through a familiar example from being able to work with the argument.

A competence can survive interruptions and change its mode of exercise. A chess player knows how to play while asleep. When she wakes, she may demonstrate that competence by making a move, identifying a blunder in someone else's play, or explaining a line she would choose under a different rule. These possibilities clarify why a description of a skill cannot be read simply as a prediction that a given movement will follow a fixed cue. The player has to be responsive to circumstances. One choice can display that ability even if it would have been a mistake in a slightly different position. Chapter II, §6, “The Positive Account of Knowing How”

How Far Does the Diagnosis Reach?

There is a tempting jump from Ryle's examples to a theory on which people have no experiences that are difficult for others to observe. Think of two patients in a waiting room who both sit still. One could be in pain and keeping quiet; the other could be comfortable. The visible posture does not tell us which is which. We might ask what else we can learn about their situations and why a person's own report can matter. The opening diagnosis supplies no reason to discount such a report. The mistake Ryle is trying to expose is a philosophical inference from our accounts of thinking, feeling, and acting to a parallel life of events that the accounts are supposed always to describe. Chapter I, §2

A related objection comes from the inside. If I concentrate in order to remember an address, it seems possible that I do some mental work before I say anything. A Rylean can grant the recollection. The question is what follows about an account of memory. The fact that I silently go over part of the address and then say the rest makes for a different episode than immediately blurting it out. But establishing how the episode unfolded does not yet tell us that competence in remembering addresses must always consist in going through that episode. Separating those points leaves space for both psychological inquiry and the conceptual distinction about skill.

An objector might also reject the strong premise of the regress while defending a role for internal representations. An explanation could say that unconscious states help produce an action without requiring that each internal state be preceded by a further act of intelligent deliberation. Ryle's regress would not by itself refute that proposal. Its premise targets an account on which an earlier intelligent act of rule consideration is always needed. A defender of internal mechanisms has a different task: specify what the mechanisms contribute and why their operation helps explain the pattern of competent conduct.

That leaves a fruitful distinction between asking how to recognize an intelligent action and asking what makes it physically possible. Try describing a chess player's move in terms of a threatened piece and then in terms of the movement of a hand. These descriptions answer different questions. If we want to know why the move avoided a loss, the first is relevant. If we want to know why the player had difficulty lifting the piece, the second might help. The mere fact that a description omits the information supplied by another description does not show that the two descriptions concern two parallel actions. Chapter I, §§2–3

A Test for Reading or Teaching

Take the statement “She understands the rule.” It is easy to feel that we must either locate an inner act of understanding or settle for recording what she did. Ryle's framework gives us more to look at. Ask the student to explain where the rule comes from, apply it to a new case, or recognize when it is irrelevant. Each request samples a different way a competence can show itself. One successful response can be rehearsed, and one failed response can reflect an ambiguous question. Together, responses in suitably different circumstances give a teacher more to go on. We can make this test without pretending that everything important about a learner's experience is visible to us. Chapter II, §§2–3, 6, and 8–9

The same exercise helps with philosophical theories. When a theory proposes an additional event to explain what a person did, ask what independent work the event is supposed to do. If the theory treats an intelligent move as the product of a prior intelligent consultation, try applying its criterion of intelligence to the consultation itself. If a concept is instead being used to describe an ability, ask how one could tell what counts as success, accident, or error. These questions return us to the visitor at the university. There was something real to understand about the institution, and the last stop on the tour was never going to supply it.

Further Reading

Start with “Descartes' Myth,” chapter I of The Concept of Mind. Read Ryle's statement of the official doctrine before his university example, then follow the example through the beginning of §3. This makes it easier to see how he wants the example to bear on the account of mind. Next read chapter II on knowing how and knowing that, moving from the regress to the clown and then to the account of skill. Compare the different jobs done by the example, the argument, and the positive proposal.

Join the discussionAsk a question or share what you learned.

Comments

Add to the conversation

A question, a correction, or an example from your own work.

No comments yet. Be the first to share your thoughts!

A different angle
All writing
Philosophy

Aristotle's De Anima: Soul, Perception, and Intellect

Aristotle treats soul as the form of a living body and asks how living things perceive, move, and think.

14 min read
Philosophy

Husserl's Cartesian Meditations: Transcendental Phenomenology

Edmund Husserl's five meditations on transcendental phenomenology, from the epoché to the experience of other minds and a shared world.

14 min read
Philosophy

Husserl's Ideas I: The Epoché and Transcendental Phenomenology

Edmund Husserl's 1913 systematic presentation of transcendental phenomenology, introducing the epoché (bracketing) and eidetic reduction.

13 min read

About the author

Reference

Citation details

Cite or share this article.

BIBTEXAcademic
@misc{brenndoerfer2025ryles, author = {Michael Brenndoerfer}, title = {Ryle's The Concept of Mind: The Ghost in the Machine}, year = {2025}, url = {https://mbrenndoerfer.com/writing/ryle-concept-mind-ghost-machine-behaviorism-guide}, organization = {mbrenndoerfer.com}, note = {Accessed: 2026-11-19} }
APAAcademic
Michael Brenndoerfer (2025). Ryle's The Concept of Mind: The Ghost in the Machine. Retrieved from https://mbrenndoerfer.com/writing/ryle-concept-mind-ghost-machine-behaviorism-guide
MLAAcademic
Michael Brenndoerfer. "Ryle's The Concept of Mind: The Ghost in the Machine." 2026. Web. November 19, 2026. <https://mbrenndoerfer.com/writing/ryle-concept-mind-ghost-machine-behaviorism-guide>.
CHICAGOAcademic
Michael Brenndoerfer. "Ryle's The Concept of Mind: The Ghost in the Machine." Accessed November 19, 2026. https://mbrenndoerfer.com/writing/ryle-concept-mind-ghost-machine-behaviorism-guide.
HARVARDAcademic
Michael Brenndoerfer (2025) 'Ryle's The Concept of Mind: The Ghost in the Machine'. Available at: https://mbrenndoerfer.com/writing/ryle-concept-mind-ghost-machine-behaviorism-guide (Accessed: November 19, 2026).
SimpleBasic
Michael Brenndoerfer (2025). Ryle's The Concept of Mind: The Ghost in the Machine. https://mbrenndoerfer.com/writing/ryle-concept-mind-ghost-machine-behaviorism-guide
Newsletter

Stay up to date

Get new articles, book updates, and notes on AI and technology.

Unsubscribe at any time. Privacy policy