The Analysis of Mind — Chapter-by-Chapter Outline

Author: Bertrand Russell First published: 1921 Edition covered: The 1921 first edition (London: George Allen & Unwin), as reproduced in the public-domain Project Gutenberg etext #2529 — the standard text used by the Routledge Classics edition (2022, with a new introduction by Thomas Baldwin, ISBN 978-1-003-30893-5) and the Dover edition. The book consists of 15 lectures, based on the William James Lectures Russell gave at Harvard in 1920–21. All lecture titles, order, and quotations below follow the table of contents of the 1921 text as digitized by Project Gutenberg.

---

Central thesis

Mind and matter are not two different kinds of stuff. Russell argues that both the "mental" world studied by psychology and the "physical" world studied by physics are constructed out of the same neutral stuff — a world of transient particulars (sensations, images, and events like them) which are in themselves neither mental nor material. He is trying to harmonize two tendencies of his day: behaviorist psychology, which makes psychology increasingly dependent on physiology and external observation, and the new physics of Einstein and Eddington, which had made "matter" less and less material by deriving it from events. Russell's route is the view of William James and the American new realists: the "stuff" of the world is a "neutral stuff," out of which both mind and matter are built by logical construction.

The book pursues this thesis as a psychological program, not an abstract metaphysics: Russell analyzes believing, desiring, perceiving, remembering, thinking, and willing to show that each reduces to sensations and images in various causal relations. His most radical claims are that consciousness is not the essence of mind, that sensations are not themselves knowledge, that desire is a causal law of behavior rather than an inner state, that introspection reveals nothing intrinsically different from sensation, and that images differ from sensations only in their causation. The book's famous dismissal of "consciousness" is borrowed from William James; its positive doctrine — that the world is composed of particulars that are the intersection of mind and matter — is developed through Russell's own theory of perspectives, mnemic causation, and belief.

The stuff of which the world of our experience is composed is, in my belief, neither mind nor matter, but something more primitive than either. Both mind and matter seem to be composite, and the stuff of which they are compounded lies in a sense between the two, in a sense above them both, like a common ancestor.

---

Lecture 1 — Recent Criticisms of "Consciousness"

Central question

Is "consciousness" — a peculiar relation to objects, or a pervading quality of psychical phenomena — the essence of everything mental?

Main argument

Russell opens by declaring what he wants to analyze: "There are certain occurrences which we are in the habit of calling 'mental.' Among these we may take as typical believing and desiring." He then sets out to refute a theory he himself formerly held: that the essence of everything mental is a peculiar something called "consciousness," conceived either as a relation to objects or as a pervading quality of psychical phenomena.

The act, the content, and the object

The traditional analysis of cognition, given its classic form by Brentano, holds that every psychical phenomenon is characterized by "intentional inexistence of an object" — a direction towards an object. Meinong developed this into a threefold analysis: every thought of an object contains an act (the same for all thoughts), a content (the present mental event that distinguishes this thought), and the object (which need not exist at the time of thinking). Russell rejects all three elements as he works toward his own view. The act, he argues, is unnecessary and fictitious: "I think so-and-so" suggests that thinking is the act of a person, but empirically no act can be discovered — it is "the ghost of the subject, or what once was the full-blooded soul." He suggests we should say "there is a thought in me" rather than "I think." The reference of thoughts to objects is not the simple essential thing Brentano supposed: it is derivative, and "consists largely in beliefs" — beliefs connecting the thought with the elements that make up its object.

James's critique of consciousness

Russell takes up William James's essay "Does 'Consciousness' Exist?", in which James argues that "consciousness," refined down from the soul to a "bare Bewusstheit," has "evaporated to this estate of pure diaphaneity" and is "on the point of disappearing altogether. It is the name of a nonentity." James's positive thesis is that the raw material of the world is one primal stuff — "pure experience" — arranged in different patterns by its inter-relations: "EXPERIENCE, I BELIEVE, HAS NO SUCH INNER DUPLICITY." He illustrates with paint: the same paint in a paint-shop is just saleable matter, while in a picture it performs a spiritual function. One undivided portion of experience plays the part of a knower in one context and of a thing known in another.

The American new realists and neutral stuff

Russell reports that the American realists (Perry, Holt) dropped the term "experience" and speak of "neutral entities" as the stuff out of which both mind and matter are constructed. Holt's definition: the terms of logic, substantialized, are "all strictly of one substance, for which perhaps the least dangerous name is neutral-stuff." Russell's own provisional position: James is right that consciousness is not an entity; sensations are neutral — what is heard or seen "belongs equally to psychology and to physics" — while images belong only to the mental world, and occurrences that form no part of any experience belong only to the physical world. Sensations are subject to both physical and psychological causal laws and are therefore truly neutral.

The indirect arguments: behaviorism and psychoanalysis

The second half of the lecture marshals indirect evidence. The behaviorists (Watson) deny that there is a separate source of knowledge called introspection and treat psychology as the science of behavior. Russell gives the example of the schoolboy asked "What is six times nine?" — all that is observed is a language-habit. He concedes "it is humiliating to find how terribly adequate this hypothesis turns out to be," though he thinks the behaviorists overstate their case. Psychoanalysis (Freud, Jung) shows that actions and beliefs can be wholly dominated by desires of which the subject is unconscious. Russell proposes his own deflationary reading: an "unconscious" desire is merely a causal law of our behavior — "we remain restlessly active until a certain state of affairs is realized, when we achieve temporary equilibrium." The unconscious desire is not a mysterious underground prisoner; it has "exactly the same status as a force in dynamics." Consciousness is not needed to explain anything: "Man has developed out of the animals, and there is no serious gap between him and the amoeba."

Neither materialism nor idealism

Russell warns that his view cannot be classified as either: "matter is not so material and mind not so mental as is generally supposed." The world is to be constructed out of neutral entities which "have neither the hardness and indestructibility of matter, nor the reference to objects which is supposed to characterize mind." This dissolves the Cartesian problem of interaction and the doctrine of psycho-physical parallelism.

Key ideas

• Believing and desiring are taken as the typical mental occurrences that the lectures will analyze.
• Brentano's doctrine of intentionality and Meinong's act/content/object schema are rejected: the act is a fiction, and reference to objects is derivative, consisting largely in beliefs.
• James's "pure experience" and the American realists' "neutral-stuff" supply the framework: one primal stuff, differently arranged.
• Sensations are neutral (subject to both physical and psychological laws); images are mental; occurrences outside all experience are purely physical.
• Behaviorism shows that "knowing" as observable is a set of language-habits; introspection as a separate source of knowledge is denied, with qualifications.
• Psychoanalysis demonstrates unconscious desires, which Russell reinterprets as causal laws of behavior — "the same status as a force in dynamics" — not as a hidden second mind.
• "Consciousness" will disappear from the argument until the lectures on words, where it "will re-emerge as mainly a trivial and unimportant outcome of linguistic habits."
• The old dispute between materialism and idealism is obsolete: mind and matter are different convenient groupings of an underlying material.

Key takeaway

Consciousness is not the essence of mind; mind and matter are both constructions out of a neutral stuff that is in itself neither mental nor material.

---

Lecture 2 — Instinct and Habit

Central question

What distinguishes instinctive from learned behavior, and what does animal behavior reveal about the elements of mind?

Main argument

Russell begins from the hypothesis of mental continuity: "from the protozoa to man there is nowhere a very wide gap either in structure or in behaviour," and therefore probably no wide mental gap. He then considers which way the continuity should be used — whether we should level up animal intelligence to our own or level down our own to what is observable in animals.

External observation as the method

Only bodily phenomena can be directly observed in animals; desires and beliefs are not visible to external observation. Russell argues that although self-knowledge has an advantage for detached facts (we know our own toothache directly), it is extraordinarily fallible — as psychoanalysis has shown — and must be checked by external observation. "On the whole, therefore, there is probably more to be learnt about human psychology from animals than about animal psychology from human beings."

Mechanical and vital movements

Following William James, Russell classifies movements: reflex, voluntary, and mechanical. A movement is "mechanical" when the body proceeds as if only dead matter were involved (a man falling over a cliff describes the same parabola as a corpse). Reflex and voluntary movements are indistinguishable to an outside observer and are grouped together as "vital" movements — those in whose causation the special properties of the nervous system are involved. The distinction between vital and mechanical movements parallels the physical distinction between macroscopic and microscopic equations: a human body, like a mass of dynamite, contains stored energy in unstable equilibrium, so that a physically small disturbance (a spoken word) can release vast effects.

Instinct defined

Russell adopts and modifies Lloyd Morgan's definition: an instinctive movement is one that is, on its first occurrence, independent of prior experience. It requires no prevision of its biological end — the first time a bird builds a nest it cannot know eggs will be laid in it. Instinct is not the infallible, preternaturally wise mechanism of popular belief: chicks follow any moving object remotely resembling their mother; Fabre's and Bergson's claims about the unerring sting of the Ammophila wasp are refuted by the Peckhams' observations; the Lomechusa beetle larva is tended by ants as if it were their own young. Russell states five points about instinct: (1) it requires no prevision of its end; (2) it is only adapted to achieve its end in the usual circumstances; (3) instinctive processes often improve with experience; (4) instinct supplies the impulses to the random experimental movements required for learning; (5) nascent instincts are easily modifiable.

Habit and learning

Learning consists in the acquisition of habits, studied in animals: the cat in the puzzle-cage lifting the latch, the rat in the Hampton Court maze. Watson's theory — the successful movement recurs more often than any other, and so comes to dominate — is partly right but cannot explain improvement at the second trial. Thorndike's "Law of Effect" (responses followed by satisfaction are strengthened) and "Law of Exercise" (responses are strengthened by repetition) are accepted with Russell's own explanation of "satisfaction" and "discomfort" to come in Lecture III. The essential point: "Learning is only possible when instinct supplies the driving-force" — the middle-aged man goes to the City every day to get fish into his larder just as the cat lifts the latch.

Key ideas

• The hypothesis of mental continuity throughout organic evolution is the test of any suggested theory of mind.
• Introspection is fallible and must be checked by external observation; animal behavior is the simpler, more analyzable datum.
• "Vital" movements (reflex and voluntary together) differ from mechanical movements in involving the nervous system's stored energy — like a spark applied to dynamite.
• An instinctive movement is one independent of prior experience; it acts without foresight and is rough and ready, not infallible.
• Instinct is easily misled (chicks, the Ammophila wasp, the Lomechusa beetle) and matures gradually.
• Learning (habit) arises from random instinctive movements and follows Thorndike's laws of effect and exercise.
• Instinct supplies the driving force of all learning; language does not alter the essential character of learning.

Key takeaway

Instinct is a mechanism for acting without foresight in ways usually advantageous biologically, and it supplies the raw driving force out of which all habit and learning are formed.

---

Lecture 3 — Desire and Feeling

Central question

What is desire — an attitude towards an imagined end, as common sense supposes, or something quite different?

Main argument

Russell announces that true views of desire require "an almost complete reversal of the ordinary unreflecting opinion." Common sense treats desire as an attitude towards something imagined — the end or object of the desire, which then causes the actions that aim at satisfying it. Russell argues the reverse: the discomfort comes first, the actions are pushed from behind, and the "end" is a retrospective description.

The argument from psychoanalysis

Psychoanalysis shows a conflict between the outside observer and the patient's consciousness: a person avows end A while his actions realize end B. The whole tendency of psychoanalysis, which Russell says is "entirely right," is to trust the outside observer rather than introspection — which "demands a re-statement of what constitutes desire, exhibiting it as a causal law of our actions, not as something actually existing in our minds." His example: we have an impulse to inflict pain on those we hate; we therefore believe they are wicked and that punishment will reform them, enabling us to act on the impulse while believing we act on the desire for reform.

Desire as a behavior-cycle

From the study of animals, Russell defines desire observationally. A "behavior-cycle" is a series of voluntary or reflex movements tending to cause a certain result and continuing until that result is caused. The "purpose" of a cycle is the result that brings it to an end, normally by quiescence; an animal "desires" the purpose of its behavior-cycle while the cycle is in progress. The marks of a cycle — appropriateness of actions and continuance until the result is achieved — are matters of degree: rivers "desire" the sea in a loose sense, and a falling animal shows the same marks of restless movement followed by quiescence without our attributing desire.

Discomfort and pleasure as causal properties

The primitive non-cognitive element in desire is "a push, not a pull, an impulsion away from the actual, rather than an attraction towards the ideal." Discomfort is a property of a sensation or other mental occurrence consisting in the fact that it stimulates movements tending to produce a change involving its cessation; pleasure is the property of stimulating no such movements, or only such as tend to prolong the occurrence. Russell follows Dr. Henry Head in distinguishing discomfort from the sensation of pain ("pain is a distinct sensory quality equivalent to heat and cold"). This makes pleasure and discomfort causal characteristics, not separate mental contents — a point argued against Berkeley, who used the ambiguity of "pain" for subjective idealism.

Conscious and unconscious desire

"Conscious" desire consists of desire proper together with a true belief as to its purpose — the state of affairs that will bring quiescence. Such beliefs are often mistaken, since only experience shows what will relieve a discomfort; the mistakes that arise are a large part of what Freud attributes to the "censor." Beliefs that we desire something tend to generate real secondary desires — the jilted lover who, unable to acknowledge revenge, develops a genuine desire to change his profession or go round the world. "Desire for beliefs" (vanity, optimism, religion) is a potent special case: we satisfy, by believing, desires too large to achieve.

Key ideas

• Desire is a causal law of behavior — a characteristic of a series of actions — not an existent state in the mind; like force in mechanics, it is a convenient fiction.
• A behavior-cycle is the observable unit: movements tending to a result, continuing until the result is achieved, then quiescence.
• Discomfort is the prime mover of action — an impulsion from behind, not an attraction from the future; the bird does each act because it has an impulse to that act, not because it foresees the nest and the eggs.
• Discomfort and pleasure are causal properties of occurrences (defined by the movements they stimulate), not separate mental contents; pain-the-sensation must be distinguished from discomfort.
• Conscious desire = desire plus a true belief about its purpose; all primitive desire is unconscious.
• False beliefs about our desires generate secondary desires that are real but do not bring complete satisfaction — the source of "the vanity of human wishes."
• Desires for beliefs are satisfied by believing rather than by facts, which explains much self-deception, vanity, and religion.

Key takeaway

Desire is not an inner attitude towards an imagined end but a causal law of our behavior — a discomfort that pushes us through behavior-cycles until quiescence, with consciousness adding only beliefs (often false) about what will bring that quiescence.

---

Lecture 4 — Influence of Past History on Present Occurrences in Living Organisms

Central question

How can past occurrences enter into the causation of present behavior in living organisms?

Main argument

Russell states the characteristic that broadly distinguishes living organisms from dead matter: the response of an organism to a given stimulus is very often dependent upon the past history of the organism. The saying "a burnt child fears the fire" embodies it: the burn may leave no visible traces, yet it modifies the child's reaction. It is customary to assume the past operates by modifying brain structure — but Russell insists this is a hypothesis, not an observed fact.

Mnemic phenomena

Following a suggestion from Richard Semon's Die Mneme, Russell calls "mnemic phenomena" those responses which can only be brought under causal laws by including past occurrences as part of the cause of the present response. His example: you smell peat-smoke and recall a former occasion; the cause of your recollection is both the present peat-smoke and the past occasion — the same stimulus will not produce the same recollection in another man who did not share your experience.

Russell surveys the classes of mnemic phenomena: (a) acquired habits — the cat in the cage, the capital of France; (b) images — copies of past sensations, as when "New York" calls up an image; (c) association — a present experience calls up the context of the former experience; (d) non-sensational elements in perception — when we see a penny we seem aware of its "real" circular shape, and we fill out sensation with the usual correlates, as the artist must unlearn in learning to draw; (e) memory as knowledge — remembering is caused by both the present stimulus and the past occurrence; (f) experience itself — "the essence of 'experience' is the modification of behaviour produced by what is experienced," and a biography is a series of occurrences linked by mnemic causation.

Semon's theory of the engram

Russell summarizes Semon: a stimulus leaves an "engram" — a difference between the primary and secondary indifference-states of the organism; "mnemic phenomena" are those due to engrams; the recurrence of part of an engram-complex "ekphores" the whole. Semon's Law of Engraphy (simultaneous excitements form a connected complex which works engraphically) and Law of Ekphory (partial return of the energetic situation works ekphorically) are presented. But Russell declines to adopt the engram as a working tool: since nothing is known of its nature, he prefers to formulate "mnemic causation" directly in observable terms, with one law: if a complex stimulus A has caused a complex reaction B in an organism, the occurrence of a part of A on a future occasion tends to cause the whole reaction B.

Mnemic causation and the metaphysics of cause

In mnemic causation, the cause is not "X now causes Y now" but "A, B, C… in the past, together with X now, cause Y now." The past items are the "mnemic cause," X the "occasion" or "stimulus," Y the "reaction." A "disposition" (like understanding a word) is not something actual but "merely the mnemic portion of a mnemic causal law." This restores the maxim "same cause, same effect" against Bergson, who denied psychological causation because the same stimulus repeated does not have the same consequences. The old metaphysical notion of cause — which wants to "feel a connection" — is what makes mnemic causation look paradoxical; science knows only "observed uniformities of sequence."

The case for the engram

Russell nevertheless finds it probable, though not certain, that mnemic causation is not ultimate: mnemic phenomena are found in physiology and even plants (Sir Francis Darwin); the brain-lesion evidence proves the brain is necessary for memory but not sufficient; and we cannot yet see differences between the brain of a man who speaks English and one who speaks French. His conclusion: "I am myself inclined, as a working hypothesis, to adopt the belief in question, and to hold that past experience only affects present behaviour through modifications of physiological structure. But the evidence seems not quite conclusive."

Key ideas

• Mnemic phenomena are responses whose proximate cause includes a past event; "a burnt child fears the fire" is the type-case.
• Habit, association, images, perception's non-sensational elements, memory, and experience itself all fall under mnemic phenomena.
• Semon's engram and laws of engraphy and ekphory provide the classic theory, but the engram is a hypothesis, not an observed fact.
• Mnemic causation states the observable law: part of a past stimulus-complex tends to arouse the whole reaction.
• A disposition (e.g., knowing French) is not an actual mental content but the mnemic portion of a causal law.
• Bergson's denial of psychological causation is answered by including the past in the cause.
• Whether mnemic causation is ultimate or reducible to physiological structure is left undecided; the materialistic answer is treated as the more probable working hypothesis.

Key takeaway

Past occurrences enter into the proximate causation of present behavior — mnemic causation — and this single characteristic underlies habit, association, images, perception, memory, and experience itself.

---

Lecture 5 — Psychological and Physical Causal Laws

Central question

What is causation, and what really distinguishes the laws of psychology from those of physics?

Main argument

Russell begins with a destruction of the traditional conception of cause and effect, replacing it with "laws of change." The traditional notion — a particular event A necessarily followed by a particular event B — has never been found in nature. Everything in nature is in continuous change, so cause and effect must be shortened indefinitely until we are left with "a certain direction of change at each moment": differential equations. Empirically we only know approximate generalizations — "A is usually followed by B" — which are not universal, not necessary, and not unique. Any event has many nearly invariable antecedents (all the hooters in London sound at twelve o'clock; any may be called the cause of men leaving the factory for dinner).

The table as a system of particulars

Russell then applies this deflationary view of causation to the physical object. "The" table which several people simultaneously see is not a single existent behind its appearances: "instead of looking for an impartial source, we can secure neutrality by the equal representation of all parties." The table is the whole set of particulars which would naturally be called its "aspects" from different points of view, collected by similarity and the laws of perspective and reflection — a first approximation that is later refined.

Perspectives and the two classifications

The stellar photography example shows the two ways of collecting particulars: (1) all the happenings in one place — the "perspective," the view of the world from that place, as in a photograph; (2) all the appearances of one object in different places — the physical object, e.g., Sirius as the system of its appearances. Physics treats as a unit the whole system of appearances of a piece of matter; psychology is interested in certain appearances themselves — perceptions are "the appearances of objects at places from which sense-organs and the suitable parts of the nervous system form part of the intervening medium." A "piece of matter" is defined, more precisely, as the limiting set of appearances obtained by moving the defining appearance nearer and nearer to the object, so that the distorting effect of the medium tends to zero.

The difference between physics and psychology

Physics groups particulars by their "active" place (the object from which they emanate); psychology groups them by their "passive" place (where they occur — a brain). This difference of grouping, not a difference of substance, is the fundamental distinction between the sciences: "The laws which psychology seeks cannot be so stated, since the particulars themselves are what interests the psychologist."

Key ideas

• The traditional notion of cause — invariable, necessary, unique — is replaced by laws of change, ideally differential equations.
• Empirical causal laws are approximate "usually follows" generalizations, and no event has a single "the cause."
• A physical object is a logical construction: the set of its correlated appearances or aspects, not an inferred hidden existent.
• Particulars are classified in two ways: by object (the way of physics) and by place or perspective (the way of psychology).
• Perceptions are appearances of objects at places where sense-organs and nervous system form part of the intervening medium — like a photographic plate with a telescope as part of the medium.
• A piece of matter is defined as the limiting set of appearances free from the distorting influence of the medium.
• Physics and psychology differ in their causal laws and groupings, not in their subject matter.

Key takeaway

Causation is law-governed change among particulars, matter is a construction from appearances, and physics and psychology are distinguished by how they group those particulars — by object versus by perspective.

---

Lecture 6 — Introspection

Central question

Is introspection a separate and authoritative source of knowledge about mental processes?

Main argument

Corresponding to the supposed duality of matter and mind, orthodox psychology posits two ways of knowing: sensation and external perception for matter, "introspection" (Kant's "inner sense") for mental processes. Russell aims to show that the stuff of mental life consists wholly of sensations and images, and that no special way of knowing is required for the latter.

Three questions, three answers

Russell distinguishes three senses in which introspection might be a source of knowledge: (1) Can we observe anything about ourselves which others cannot observe (privacy)? (2) Does everything observable obey the laws of physics? (3) Can we observe anything differing intrinsically from the constituents of the physical world? His answers: (1) Yes — some data are private, but privacy is a matter of degree (a toothache is private, a clap of thunder public, and even the most private sensation has correlations others can observe); privacy does not make a datum unamenable to scientific treatment. (2) Tentatively yes — images, as introspective data, cannot at present be brought under the laws of physics, e.g., a visual image of a friend in a chair that is physically empty. (3) No — "I propose to argue that thoughts, beliefs, desires, pleasures, pains and emotions are all built up out of sensations and images alone."

The critics of introspection

Knight Dunlap's "The Case against Introspection" is examined: Stout's view requires an illicit single observer, and Dunlap holds that "I am never aware of an awareness." Russell replies that the statement "I am aware of a colour" is itself known, and that beings aware of objects but not of their own awareness (children, animals) cannot make such statements — "To my mind the fact (which he admits) that we know there is awareness, is ALL BUT decisive against his theory." Dunlap is right, though, that what is called introspection really consists of awareness of images and bodily sensations — "instances of knowing" built from such constituents suitably related.

The fallibility of introspection

Introspection is not more trustworthy than external perception: psychoanalysis shows its fallibility as to what we desire; an autobiography confronted with documentary evidence is full of inadvertent errors; a forgotten letter shows our opinions were more foolish than we remembered. "It is necessary to construct hypotheses and test them by their consequences, just as we do in physical science."

Key ideas

• The stuff of mental life, as opposed to its relations and structure, consists wholly of sensations and images.
• Introspection is valid in the sense of privacy (some observations are private to one observer) but not in the sense of revealing intrinsically different stuff.
• Images are private in a way sensations are not, and presently cannot be brought under the causal laws of physics; the objection of localization (an image of a friend in an empty chair) supports this.
• The subject or "act" is a logical fiction; "consciousness of" is an external and rather remote relation, like uncle and nephew.
• Awareness of awareness is admitted against Dunlap: "we can be aware of an awareness."
• Introspection is exceedingly fallible and must be checked by external observation; it gives no privileged certainty.
• Knowing is not an unanalysable relation but various complex relations, all involving sensations and images.

Key takeaway

Introspection reveals no stuff intrinsically different from sensation — only images, which differ from sensations causally — and it enjoys no special trustworthiness.

---

Lecture 7 — The Definition of Perception

Central question

What is perception, once mind and matter are both constructions out of particulars?

Main argument

Russell develops the twofold classification of particulars into "things" (sets of aspects collected by the laws of perspective) and "perspectives" or "biographies" (sets of particulars collected by time-relations), preparing definitions of sensation and image.

Perspectives, biographies, and places

A "momentary thing" is a set of particulars collected by the generalized laws of perspective; a thing is a series of such sets collected by the laws of dynamics. Alternatively we can first collect the successive aspects presented to one spectator (a biography) and then the set of such series — the actor on the stage seen from many seats. Two aspects of different things belong to the same perspective when they are simultaneous; because there is no universal time except by construction, simultaneity can be defined within one biography. Every particular relevant to physics is associated with two places: the "active" place (where its object is) and the "passive" place (where it occurs). When I look at a star, my sensation is a member of the group that is the star and of the group that is my biography — "subjectivity in the point of view is not a distinctive peculiarity of mind: it is present just as much in the photographic plate."

Perception and sensation defined

Adopting the physical standpoint, a "perception" of an object is the appearance of the object from a place where there is a brain (with sense-organs and nerves in the intervening medium), distinguished from other appearances by two characteristics: they give rise to mnemic phenomena, and they are themselves affected by mnemic phenomena (a spoken sentence has different effects on those who know the language). "Sensation" is defined as that part of a perception which proceeds without mnemic influences out of the character of the object — "the sensation is a theoretical core in the actual experience; the actual experience is the perception."

Regular and irregular appearances

The star example: in vacuo a star's appearances are "regular," derived from the star by the laws of perspective, and are members of the system that is the star. Irregular appearances (through mist, atmosphere, an eye) are caused by the object together with the intervening medium; when a particular cannot be traced back to one object but blends several, it is a "confused appearance" — and "all actual perception is confused to a greater or less extent." Perception, so defined, is not a precise notion.

The prejudice of the external world

Russell notes that everything outside one's own biography is hypothesis — the belief in things outside my experience "exists antecedently to evidence" and is justified practically by the simplification it introduces into the laws of physics. "With this proviso, I propose to continue yielding to the prejudice."

Key ideas

• Particulars are collected into momentary things and things by perspective and dynamical laws, or into perspectives and biographies by time-relations.
• Every particular has an active place (where its object is) and a passive place (where it occurs); subjectivity belongs to photographic plates as much as to minds.
• A perception is an appearance of an object at a brain-involving place, distinguished by giving rise to and being affected by mnemic phenomena.
• Sensation is the theoretical non-mnemic core of perception; perception is the actual experience including mnemic filling-out.
• Regular appearances are members of the object (the star); irregular appearances are caused by the object plus the medium.
• All actual perception is vague and confused to a greater or less extent; perception is not a precise notion.
• Belief in the world outside one's biography is a practically justified prejudice, not a demonstrated truth.

Key takeaway

Perception is the appearance of an object at a place where a brain is part of the medium, and sensation is its non-mnemic core — so perception already contains a mnemic element supplied by past experience.

---

Lecture 8 — Sensations and Images

Central question

How are sensations and images to be distinguished, and what does the distinction show about mind and matter?

Main argument

Russell quotes James's example of the imagined fire: "mental fire is what won't burn real sticks" — but he corrects James by noting that images do have effects, only according to different laws. If physical and psychological causal laws could be sharply distinguished, we could define sensations as events with physical causes and mental effects, and voluntary movements as events with mental causes and physical effects. Since the distinction of causal laws is not sharp, Russell falls back on the definition prepared in Lecture VII: a sensation is the non-mnemic element in a perception.

Sensation as the intersection of mind and matter

Pure sensation is hard to isolate — in the theater at home we seem to hear equally well anywhere because we unconsciously fill out what we hear with inferences; a friend mistook a burst tire for a pistol and insisted he had seen the flash. But there is always a sensational core. The crucial step: sensation is not itself a cognition. Russell previously regarded it as such, distinguishing the seeing from what is seen — the sensation as "awareness of the colour" and the colour as the "sense-datum." But the subject appears to be a logical fiction, "like mathematical points and instants"; dispensing with the subject, "the possibility of distinguishing the sensation from the sense-datum vanishes." The sensation we have when we see a patch of colour simply is that patch of colour — "an actual constituent of the physical world." If we admit that the patch of colour may be both physical and psychical, the reason for distinguishing sense-datum from sensation disappears. "Sensations are what is common to the mental and physical worlds; they may be defined as the intersection of mind and matter." This is Mach's view (1886) and James's and Dewey's ("perceptions... as pure natural events, not cases of awareness").

Three attempted criteria for images

Russell considers how to distinguish images from sensations: (1) By less vividness — Hume's "impressions" and "faint ideas," Stout's modified "aggressiveness": both fail, since a just visible star is a faint sensation, while Macbeth's "horrid image" can unfix his hair. (2) By absence of belief in physical reality — circular: we only feel images unreal when we already know they are images; in dreams images feel just as real. (3) By their causes and effects — valid: images have mnemic causes (they are occasioned, through association, by a sensation or another image), while sensations are caused by stimuli external to the brain. "Images have been defined as 'centrally excited sensations,'" a phrase Russell accepts with caution since it assumes a proximate physiological cause.

The defense of images against behaviorism

Watson denies images outright, replacing them with faint sensations, especially subvocal speech. Russell replies that inner speech might indeed be experimentally decided (a delicate instrument in the mouth), but visual and auditory images cannot be explained away: the image of a friend sitting in an empty chair "contradicts all the usual physical laws" and occupies the same area of the visual field as a sensation. Galton's investigations showed education tends to kill visual imagery — the Fellows of the Royal Society had less of it than their wives. "If you try to persuade any uneducated person that she cannot call up a visual picture of a friend sitting in a chair... she will conclude that you are mad."

Images as copies

Finally, Russell affirms Hume's principle: simple images are copies of simple sensations, which is what makes an image an image "of" something. The recognizable resemblance of images and sensations is of fundamental importance for memory and knowledge generally — the subject of the next lecture.

Key ideas

• The mental/physical distinction is not primarily about stuff but about causal laws — and the causal-law distinction is not sharp.
• A sensation is the non-mnemic core of a perception; it is not itself a cognition but supplies the data for knowledge.
• With the subject discarded as a logical fiction, the sensation and the sense-datum collapse into one: the patch of colour is both physical and psychical.
• "Sensations are what is common to the mental and physical worlds; they may be defined as the intersection of mind and matter."
• Images cannot be defined by faintness (Hume, Stout) or by felt unreality (circular), but only by their causes: mnemic causes, as opposed to external stimuli.
• Behaviorism's denial of images fails against visual and auditory images, which contradict physical laws and cannot be reduced to small sensations.
• Images are, in their simple parts, copies of antecedent sensations (Hume's principle), and this resemblance underwrites meaning, memory, and thought.

Key takeaway

Sensations are the neutral intersection of mind and matter — not cognitive in themselves — and images are distinguished from them only by their causal laws, not by any intrinsic difference.

---

Lecture 9 — Memory

Central question

What is memory as a form of knowledge, and what actually occurs when we remember?

Main argument

Memory introduces the analysis of knowledge, which will occupy the book to Lecture XIII. Two preliminary points: everything constituting a memory-belief is happening now — it is not logically necessary that the event remembered occurred, or that the past existed at all (the hypothesis that the world sprang into being five minutes ago is logically tenable); and images without beliefs are insufficient for memory, as are habits.

The features of memory-images

Memory-images must have two characteristics: feelings of familiarity (which lead us to trust our memories — and familiarity is a definite feeling, capable of existing without an object, as in Turgenev's Smoke with the smell of heliotrope) and feelings of pastness (which assign places in the time-order). Russell distinguishes familiarity, recognition in the sense of knowing what a thing is (merely habit), and recognition as "this has existed before" (a belief, hence memory in a minimal sense). Recognition shows the peculiarity of psychological causal laws: its cause is the two occasions of the stimulus, either alone insufficient.

Habit-memory and true memory

Following Bergson's Matter and Memory, Russell distinguishes habit-memory (learning a lesson by heart — a gramophone could relate its past with suitable records) from "true memory," the recollection of a unique past event. True memory is knowledge of the past obtained "immediately, not inferred, not abstract." Between sensation and memory-image there is "immediate memory" — the retention of the immediate past in a condition intermediate between sensation and image, included in the "specious present"; fading sensations are "akoluthic" sensations. True memory applies only to events beyond the period of fading.

The analysis of true memory

True memory consists of an image accompanied by a belief-feeling expressed by "this happened." Memory-images and imagination-images do not differ intrinsically; the belief-feeling is the distinctive thing. There are at least three belief-feelings: memory, expectation, and bare assent — "the pastness lies, not in the content of what is believed, but in the nature of the belief-feeling." In primitive memory we judge "this occurred": the word "this" is a vague word covering both the image and its prototype, so the judgment is "vague, but not false." Vagueness and accuracy are defined instrumentally — an instrument is accurate in proportion as it reacts differently to slightly different stimuli; a vague belief has a better chance of truth but less value. The analysis is summarized: memory demands (a) an image, (b) a belief in past existence expressible as "this existed"; the believing is a specific feeling or complex of sensations; the content includes the image, a feeling akin to respect translated as "real," and a relation between them; the time-determination lies in the belief-feeling. "This analysis of memory is probably extremely faulty, but I do not know how to improve it."

Key ideas

• Memory-beliefs are present occurrences; their reference to the past is not logically guaranteed (the world might have begun five minutes ago).
• Feelings of familiarity and pastness, not intrinsic vividness, attach memory-images to the past and to their prototypes.
• Bergson's distinction between habit-memory and true memory is vital but hard to apply in practice; words can easily counterfeit true memory.
• Immediate memory and the specious present bridge sensation and image, giving us experience of succession.
• Recognition ("this has existed before") is a belief whose cause is two occasions of the stimulus — a genuinely psychological causal law.
• The essence of memory is a special belief-feeling; memory, expectation, and bare assent are three belief-feelings with different time-references.
• Primitive memory's "this occurred" is vague but not false; vague identity (close similarity) has been a source of philosophical confusion.
• Accuracy is a matter of degree, defined by the ratio of difference of response to difference of stimulus.

Key takeaway

True memory is a present image bound to the past by a distinctive belief-feeling ("this existed") — the pastness lies in the feeling, not in the content believed.

---

Lecture 10 — Words and Meaning

Central question

What is the relation of "meaning" that connects a word (or image) to what it designates?

Main argument

Russell investigates the relation we express when we say the word "Napoleon" means a certain person. A word, considered physically, is not a simple thing: it is a set of similar series of movements (spoken) or sounds (heard). "Meaning" is what distinguishes a word from other sets of similar movements, and it remains to be defined.

What words can mean

Language is not conventional at base — no parliament of speechless elders agreed to call a cow a cow; the association of words with their meanings grew up naturally. The essence of language is the employment of fixed associations so that something now sensible calls up the "idea" of something else: the present sensible thing is a "sign" or "symbol," and what it calls up is its "meaning." What words mean: proper names mean series of occurrences bound by causal laws (Napoleon was a series of gradually changing appearances — "first a squalling baby, then a boy, then a slim and beautiful youth, then a fat and slothful person"); general names (man, cat) mean classes of such collections; verbs and words like "rain" or "lightning" denote sets of similar occurrences; quality words (white, round) denote static features; relation words (in, above, before) are the most abstract and logically simplest.

Understanding words

Understanding is a habit, "more like understanding cricket than knowing dictionary definitions": a person understands a word when (a) suitable circumstances make him use it (active understanding) and (b) hearing it causes suitable behavior (passive understanding). Dogs have passive understanding only. The famous example: "Look out, there's a motor coming" — the Englishman jumps aside without any "mental" intermediary; the Frenchman with slight English goes through inner speech; the child needs the object pointed out. Russell lists six ways words can mean: (1) used properly on suitable occasions; (2) hearing them produces appropriate action; (3) association with another word (translation); (4) association with an object when first learnt; (5) description or recall of a memory-image; (6) description or creation of an imagination-image — the narrative and imaginative uses. The last two, the use of words in "thinking," depend at origin on images and cannot be fully dealt with on behaviorist lines: "the problem of the meaning of words is brought into connection with the problem of the meaning of images."

The meaning of images and the causal theory

Images have meaning, and their meaning is more primitive than that of words. An image "of" St. Paul's has some of the effects the object would have — especially associative and emotional effects; a vague image of a nondescript dog may mean "dog" in general if it has the associations all dogs share. Berkeley and Hume, in attacking general ideas, wrongly assumed every image has the definiteness of a physical object. The general theory: "IN MNEMIC CAUSATION, an image or word, as stimulus, has the same effect (or very nearly the same effect) as would belong to some object... In that case we say that the image or word means that object." A word's meaning is wholly constituted by mnemic causal laws, not likeness. Words acquire through association "some of the same causal efficacy as the object" — the word "motor" can make you leap aside though it cannot break your bones. Telescoped processes (A causes B, B causes C, so A comes to cause C directly) explain why familiar words act without images. Words excel images for thought — easy to produce, concrete though abstract in meaning, instances highly similar — but "the tyranny of traditional words is dangerous, and we have to be on our guard against assuming that grammar is the key to metaphysics."

Key ideas

• A word is a set of similar occurrences; meaning is what distinguishes words from other movement-sets.
• Language's basis is natural association, not convention; the essence of language is the use of signs to call up ideas.
• Proper names denote causal series of particulars; general names denote classes; verbs and "rain"-words denote sets of similar occurrences.
• Understanding a word is a habit of active use and passive response — like understanding cricket, not like knowing definitions.
• Words have demonstrative, narrative, and imaginative uses; the narrative and imaginative uses (thinking) depend at origin on images.
• Meaning is constituted by mnemic causal laws: a word or image means what shares its mnemic effects.
• Telescoped processes let words produce directly the effects once produced through images.
• Grammar is not a key to metaphysics; words should be checked, occasionally, by direct contemplation of facts through images.

Key takeaway

Meaning is a causal relation — a word or image means an object when it shares that object's mnemic effects — and thinking is largely the use of words and images as signs for what is remote in time or space.

---

Lecture 11 — General Ideas and Thought

Central question

Must we postulate abstract ideas to explain the intelligent use of abstract words, or is thought built from images and words alone?

Main argument

Russell poses the question: seeing that there are words with abstract meanings which we can use intelligently, what mental content must be assumed? The logical answer: "absolutely nothing in the way of abstract mental content is inferable from the mere fact that we can use intelligently words of which the meaning is abstract" — a machine could be built that says "There is a dog" when a dog appears.

Berkeley on abstract ideas

Russell rehearses Berkeley's onslaught on "abstract ideas" (against Locke's general triangle): every idea is particular — "I cannot by any effort of thought conceive the abstract idea above described." Berkeley's solution: an idea which in itself is particular becomes general "by being made to represent or stand for all other particular ideas of the same sort" — the drawn inch-line used to demonstrate the bisection of any line. Russell agrees in substance but finds the view inadequate because images are "as a rule not of one definite prototype, but of a number of related similar prototypes."

The generic image

Semon's account is quoted at length: the ekphored picture of a familiar face is "something so to speak generalized," like "those American photographs which seek to display what is general about a type by combining a great number of photographs of different heads over each other on one plate" — a "physiological abstraction" produced by "mnemic homophony." Russell distinguishes the vague from the general: a composite image alone yields only vagueness; it functions as a general idea when, in addition, we have particular images of the instances and knowingly contrast them with the generalized image. "A general idea differs from a vague one in a way analogous to that in which a memory-image differs from an imagination-image" — the difference consists in the presence of a certain accompanying belief or judgment.

Imageless thinking

Russell examines the experimental psychology of thought — Titchener's summary of Watt, Messer, and Bühler — and dismisses the supposed discovery of "imageless thinking" (Bühler's "Bewusstheit dass"). The method relies on fallible introspection; a trained psychologist "is at least equally likely to have acquired a habit of seeing what his theories require." What looks like imageless thinking is probably "telescoped processes, in which habit has caused a great many intermediate terms to be elided." "I remain, therefore, entirely unconvinced that there is any such phenomenon as thinking which consists neither of images nor of words."

Consciousness of universals

On Brentano's view there must be a special mental content whose object is a universal — the universal cat is "conceived" while a particular cat is "perceived." Once the relation of mental occurrence to object is seen as indirect and causal, this disappears: a horse reacts to the smell of a bear derived from a bearskin without being "conscious of" the universal, and a man who says "I smell a bear" reacts no differently. "A universal never appears before the mind as a single object in the sort of way in which something perceived appears" — universals are inferred parts of the structure of the world, not data. Behaviorist psychology can therefore account for all the observable behavior regarded as evidence that people think; introspective data must be accepted "simply on account of the fact that we seem to perceive them." But the behaviorist philosophy of physics fails: since the physical world itself is "infected through and through with subjectivity" (the theory of relativity), we are brought back to the necessity of trusting private observations.

Key ideas

• Logic shows no abstract mental content is inferable from correct speech; a machine could manage it.
• Berkeley and Hume were right against abstract ideas: every image is particular; generality is a matter of use.
• Semon's generic image — superposition of similar perceptions — yields vagueness; adding contrasted particular images makes it function as a general idea.
• A general idea differs from a vague one by an accompanying belief or judgment, as memory differs from imagination.
• The Würzburg experimental work on "imageless thinking" rests on unreliable introspection; such thinking is likely telescoped habit.
• Thinking consists of images and words, never of a third kind of content; "ideas" need not be added to sensations and images.
• Consciousness of universals is a reflective judgment of similarities and differences; universals are inferred, not data.
• Behaviorism fails as a philosophy of psychology because it rests on an inadequate philosophy of physics.

Key takeaway

General thought needs no abstract ideas: a generic image plus a judgment contrasts particulars with the general, and all thinking is built from images and words.

---

Lecture 12 — Belief

Central question

What is belief — the central problem in the analysis of mind — in its intrinsic nature, apart from truth and falsehood?

Main argument

"Believing seems the most 'mental' thing we do, the thing most remote from what is done by mere matter." Russell lays down requisites any theory of belief must satisfy: (1) beliefs are characterized by truth or falsehood, which consist in a relation to something outside the belief — the "fact," its "objective," with the relation called "reference"; (2) we must distinguish believing from what is believed — the believing is "just the same" across contents; (3) both what is believed and the believing are present occurrences — the content of "Caesar crossed the Rubicon" is present even though its objective is past; (4) the gulf between content and objective does not defeat knowledge — "knowing" is "a very external and complicated relation... involving no more unity than there is between a signpost and the town to which it points"; (5) the objective reference of a belief derives from the meanings of the words or images in its content.

The content of belief

The content of a belief is always complex — "Plato preceded Aristotle" and "Aristotle preceded Plato" contain the same constituents but are different contents. Contents may consist of words only, images only, or mixtures, plus sensations as constituents (the tram-noise example: hearing a noise and saying "tram"). Word-contents are what logic calls "propositions"; Russell extends "proposition" to image-contents, distinguishing "image-propositions" from "word-propositions." Image-propositions are more primitive and may ante-date language; many of our beliefs are "unconscious" in the sense that we have never told ourselves we have them — the cock pigeon expecting the obnoxious bird in the looking-glass experiences a belief deserving the name.

Three theories of believing

The same content can be entertained in different attitudes — expecting, remembering, doubting, desiring an "egg for breakfast." Russell tests two rival theories: (1) The view that belief's differentia is causal efficacy in producing voluntary movements — inadequate, since beliefs often occur without issuing in movement, and believing differs from merely considering the same content; (2) The Spinozistic view (endorsed by James: "ANY OBJECT WHICH REMAINS UNCONTRADICTED IS IPSO FACTO BELIEVED") that belief is the mere uncontradicted image — attractive for dreams and hallucinations, but it cannot explain memory or mathematical belief. Russell's own theory: there are at least three kinds of belief — memory, expectation, and bare assent — "each of these I regard as constituted by a certain feeling or complex of sensations, attached to the content believed." The analysis in the case of bare assent: (a) a proposition of interrelated images (possibly with sensations), (b) the feeling of assent, (c) a relation between them making the proposition what is assented to. Words mislead here: "I ate my breakfast" and "Caesar conquered Gaul" have the same verbal form but are psychologically different — memory versus assent. The belief-feeling can even occur unattached and heightened (James's nitrous-oxide intoxication, "revelation or mystic insight"): "the belief-feeling, in abnormal strength, attaches itself, more or less accidentally, to some content."

Key ideas

• Belief is the central problem: psychology, theory of knowledge, and metaphysics revolve about it.
• Truth and falsehood consist in a belief's relation to its "objective" (the fact that makes it true or false); the intrinsic nature of belief is treated independently.
• The content of a belief is always complex and consists of words and/or images (with possible sensations as signs); "believing in God" is really believing that God exists.
• Word-contents and image-contents are propositions; image-propositions are more primitive and may ante-date language.
• Belief cannot be defined by causal efficacy alone, nor as an uncontradicted image; both views capture part of the truth.
• There are at least three belief-feelings — memory, expectation, bare assent — each a feeling or complex of sensations attached to a content by a specific relation.
• Language conceals the difference: memory and assent to a past proposition have the same verbal form but different psychology.
• The belief-feeling may occur detached from any content, as in James's nitrous-oxide cases and alleged revelation.

Key takeaway

Belief is a specific belief-feeling (memory, expectation, or assent) attached by a relation to a complex content or proposition — an analysis that makes believing an experienced occurrence, not an act of a subject.

---

Lecture 13 — Truth and Falsehood

Central question

What makes a belief true or false, and can truth be defined without recourse to anything outside the belief?

Main argument

Truth and falsehood lie strictly outside the analysis of mind — a belief is rendered true or false by a relation to a fact that may lie outside the believer's experience — yet the question whether minds yield knowledge is so vital that any analysis of mind must be examined in relation to it. Russell divides the discussion into four parts.

I. Knowledge as accuracy and appropriateness of response

Viewed from outside, a man is an instrument making responses to stimuli. A response displays knowledge when it has two characteristics: accuracy and appropriateness — distinct and sometimes incompatible (when pursued by a tiger, accuracy is furthered by turning round to look, appropriateness by running away). The two boys and the weather-cocks: one boy always answers "four" to every multiplication question (like a weather-cock stuck fast — no variation of response with variation of stimulus); the other answers at random (like a weather-cock that changes without change of wind). Accuracy requires: different responses to relevantly different stimuli, and the same response to stimuli not relevantly different. But accuracy alone is not knowledge: a calculating machine gives perfect answers; a person who always believes falsely is as sensitive an instrument as one who always believes truly — "The observable and practical difference between them would be that the one who always believed falsely would quickly come to a bad end." The missing element is appropriateness, suitability for realizing one's purpose.

II. The failure of intrinsic criteria

Philosophers have sought an intrinsic mark of truth: (1) Self-evidence (Meinong) fails — what seems self-evident to one man (men at the Antipodes falling off) is dubitable for another; "self-evidence is useless as a practical criterion"; mathematical self-evidence is explained by the fact that symbols mean what we choose. (2) Coherence (Joachim, the idealists) fails — it requires that from one truth the whole of truth be inferable, and "a consistent fairy-tale is a different thing from truth, however elaborate it may be."

III. Verifiability

There is no infallible criterion; but there are relative criteria. "One of the plainest cases of verification, perhaps ultimately the only case, consists in the happening of something expected" — first images with belief, then sensations: expectation is verified when the sensation which is the "meaning" of the image succeeds it. All verification is ultimately of this sort, even for the past (manuscripts, inscriptions, causal laws verified by future experience). "It affords no theoretical refutation of the sceptic, whose position must remain logically unassailable."

IV. The formal definition

The objective reference of a proposition is a function of the meanings of its component words. Since there are true and false propositions but no true and false facts, two contradictory beliefs have the same objective; the difference lies in direction — "when to-day is Tuesday, your belief that it is Tuesday points TOWARDS the fact, whereas when to-day is not Tuesday your belief points AWAY FROM the fact" (a way of looking at the matter owed to Ludwig Wittgenstein). In the simplest cases a true proposition corresponds formally to its objective: the same relation relates the images as relates their prototypes (the window to the left of the door). Complications: word-propositions have more terms than their facts ("precedes" is a solid word, not a relation); negative propositions and negative facts (there are negative facts, Russell asserts); atomic propositions are verified by replacing each word by its meaning. The formal theory is true but inadequate: it does not explain our preference for true beliefs, which requires the causal efficacy of beliefs and their appropriateness to purposes.

Key ideas

• Truth and falsehood consist in a belief's relation to an outside fact (its objective); no purely psychological means can distinguish them.
• Knowledge as behavior = accuracy of response plus appropriateness to purpose; both are needed and they are distinct.
• Accuracy is defined instrumentally: different responses to relevantly different stimuli, same response to irrelevant differences.
• Self-evidence and coherence fail as intrinsic criteria of truth.
• Verification is fundamentally the happening of something expected; there is no absolute criterion, and scepticism remains logically unassailable.
• A true proposition "points towards" its objective, a false one away from it — the direction of pointing, not the fact alone, determines truth.
• Word-propositions have more terms than the facts they assert; there are negative facts; the simplest correspondence is found in image-propositions.
• The formal definition of truth is inadequate without the causal efficacy of beliefs and appropriateness.

Key takeaway

Truth is a formal correspondence between belief and fact — a true belief points towards its objective — while knowledge additionally requires the accuracy and appropriateness that only purpose and causal efficacy supply.

---

Lecture 14 — Emotions and Will

Central question

Can emotions and will be analysed into sensations and images, as the book's main thesis requires?

Main argument

Russell has nothing original to say here; he treats the two topics only to complete the thesis that "all psychic phenomena are built up out of sensations and images alone."

Emotions and the James–Lange theory

An emotion is essentially complex; the question is whether it contains anything not reducible to sensations and images and their relations. The more important topic is the physiological causation of emotions. James's statement of the James–Lange theory: "THE BODILY CHANGES FOLLOW DIRECTLY THE PERCEPTION OF THE EXCITING FACT, AND THAT OUR FEELING OF THE SAME CHANGES AS THEY OCCUR IS THE EMOTION" — "we feel sorry because we cry, angry because we strike, afraid because we tremble." Sherrington's experiments (severed spinal cords in dogs, with emotional expression preserved) and Cannon's work on adrenin (pain, fear, rage affect the supply of adrenin; artificial injection can produce the symptoms of fear) are weighed; following Angell's defence, Russell concludes James's essential contentions are not materially affected, though he does not need to decide the doubtful physiological details.

Emotions as confused perceptions

Whatever the physiology, Russell's thesis holds: if James is right, an emotion involves a confused perception of the viscera concerned in its causation; if Cannon and Sherrington are right, a confused perception of its external stimulus. Either way it contains elements that are sensations by the definitions of Lectures VII and VIII. An emotion in its entirety is a process with dynamic elements — motor impulses, desires, pleasures and pains — and these, per Lecture III, are characteristics of processes, not separate ingredients. "The ingredients of an emotion are only sensations and images and bodily movements succeeding each other according to a certain pattern."

Will and voluntary movement

A voluntary movement is defined via James: it is one accompanied by kinaesthetic sensations that tends to be caused by the images of those sensations — "no movement can be made voluntarily unless it has previously occurred involuntarily." Volition in the emphatic sense — decision after deliberation — adds a judgment ("this is what I shall do") and a sensation of tension during doubt followed by a different sensation at the moment of deciding. Conflict of desires supplies incompatible kinaesthetic images until one becomes exclusive. "Thus will seems to add no new irreducible ingredient to the analysis of the mind."

Key ideas

• Emotions are complex processes; the analysis turns on their physiological causation.
• The James–Lange theory reverses common sense: bodily changes follow the perception, and the feeling of those changes is the emotion.
• Sherrington's and Cannon's experiments do not refute James's essential contentions, on Angell's defence.
• On either physiological theory, an emotion contains elements that are sensations by definition — a confused perception of viscera or of external stimulus.
• Desires, pleasures, and pains in emotion are properties of processes, not separate ingredients.
• A voluntary movement is one caused by the image of its own kinaesthetic sensations; nothing voluntary is possible without prior involuntary performance.
• Volition adds a judgment and the sensations of tension and decision, but no new stuff.
• Emotions and will add no irreducible ingredients to the analysis of mind.

Key takeaway

Emotions are processes whose cross-sections are confused perceptions — of the viscera or of the stimulus — and will is kinaesthetic imagery guiding movement, so neither introduces any stuff beyond sensations and images.

---

Lecture 15 — Characteristics of Mental Phenomena

Central question

What characterizes mind as opposed to matter, and how is psychology to be distinguished from physics?

Main argument

Russell reviews the suggested marks of mind. The provisional answer given at the outset was that psychology and physics are distinguished by their causal laws, not their subject matter — with images as the one subject matter subject only to psychological laws.

Consciousness is too narrow

Consciousness cannot be the essence of mind. Russell now defines it: consciousness is of something, and consists in the relation of an image of a word to an object which was defined as "meaning," together with belief. "A sensation which is remembered becomes an object of consciousness as soon as it begins to be remembered" — but sensation itself is not an instance of consciousness. Sensation "is the sort of thing of which we MAY be conscious, but not a thing of which we MUST be conscious"; there is no class of occurrences of which we are always conscious. The practical function of consciousness and thought is "that they enable us to act with reference to what is distant in time or space" — reference to absent objects through association and habit. Consciousness is "far too complex and accidental to be taken as the fundamental characteristic of mind."

Mnemic causation is too wide

The power of acquiring experience does not define mind, for mnemic phenomena are found in physiology and plants — "perhaps it is the nerves that acquire experience rather than the mind." The same applies to memory: its causal peculiarities may have a physiological explanation in the brain's modification by past events.

Subjectivity

Subjectivity comes much nearer to what is required. A perspective (all the appearances of different things at a given place and time) and a biography (their history through time) give "the view of the world from a certain place." Subjectivity is exhibited by a photographic plate — it involves none of the other marks of mind. But it is an essential element in the definition of mind, presupposed in mnemic causation (the causal unit is a whole perspective in a place where there is nervous tissue) and in perception.

Data and causal laws

The data of the physical sciences are also psychological data: outside psychology, data ultimately consist of sensations (to say "we see the nerves" is like saying "we hear the nightingale"). Beliefs, desires, and volitions are complex phenomena of sensations and images interrelated — constructed or inferred, not data. The ultimate data of psychology are only sensations and images and their relations. Physics groups particulars by their "active" place; psychology by their "passive" place; images have no active place and belong exclusively to psychology. Whether mental events are causally dependent on physical events (materialism in one sense) "I do not profess to know" — but "I think the bulk of the evidence points to the materialistic answer as the more probable." Psychological laws such as association are rough generalizations (Wohlgemuth's experiments on the direction of associations), and precision seems to require the nervous system.

The fundamental science

Russell's closing vision: an ultimate scientific account of the world, if it were ascertainable, "would resemble psychology rather than physics" — it would state causal laws in terms of the particulars themselves, not in terms of matter, which is "a logical fiction." "Causal laws so stated would, I believe, be applicable to psychology and physics equally; the science in which they were stated would succeed in achieving what metaphysics has vainly attempted, namely a unified account of what really happens." The final summary runs: I. Physics and psychology are not distinguished by their material; mind and matter alike are logical constructions, physics grouping particulars by active places, psychology by passive places. II. The essential characteristics of psychological causal laws are subjectivity and mnemic causation, which are connected. III. Habit, memory, and thought are developments of mnemic causation, probably derivative from physical causation in nervous tissue. IV. Consciousness is a complex and far from universal characteristic. V. "Mind is a matter of degree, chiefly exemplified in number and complexity of habits." VI. All our data, in both physics and psychology, are subject to psychological causal laws; but physical causal laws can only be stated in terms of matter, which is inferred and constructed — "In this respect psychology is nearer to what actually exists."

Key ideas

• Consciousness is defined as meaning-plus-belief — a sensation becomes an object of consciousness only when remembered or used as a sign; it is complex and accidental, not fundamental.
• Mnemic causation is too wide to define mind: nerves may acquire experience, and memory may be physiological.
• Subjectivity — the perspective or biography giving the view of the world from a place — is essential to mind but not sufficient (photographic plates have it).
• Data ultimately consist of sensations; the data of physical science are also psychological data, while beliefs and desires are constructions, not data.
• Physics and psychology differ by grouping particulars: active place versus passive place; images belong exclusively to psychology.
• Whether mental events depend causally on physical events is undecided; the bulk of evidence points to the materialistic answer as more probable.
• Psychological laws like association are rough generalizations; precision seems to require the nervous system.
• The ideal science states causal laws of particulars, applicable to physics and psychology alike, with matter a logical fiction — "psychology is nearer to what actually exists."

Key takeaway

Mind and matter are both logical constructions from neutral particulars, and psychology differs from physics only in its causal laws and groupings — mind being a matter of degree, chiefly in the number and complexity of habits.

---

The book's overall argument

1. Lecture 1 (Recent Criticisms of "Consciousness") — establishes that consciousness is not the essence of mind, that mind and matter are both built from a neutral stuff, and that believing and desiring are the typical mental occurrences to be analyzed.
2. Lecture 2 (Instinct and Habit) — establishes the continuity of mind with animal life and that instinct and habit, the raw materials of behavior, need no consciousness to explain them.
3. Lecture 3 (Desire and Feeling) — establishes that desire is a causal law of behavior (a behavior-cycle driven by discomfort) rather than an inner attitude toward an imagined end, and that pleasure and discomfort are causal properties.
4. Lecture 4 (Influence of Past History on Present Occurrences in Living Organisms) — establishes mnemic causation: past occurrences enter into the proximate causes of present responses, unifying habit, association, images, perception, memory, and experience.
5. Lecture 5 (Psychological and Physical Causal Laws) — establishes that causation is law-governed change, that matter is a construction from appearances, and that physics groups particulars by object while psychology groups them by perspective.
6. Lecture 6 (Introspection) — establishes that introspection reveals no stuff intrinsically different from sensation, only private data (chiefly images), and enjoys no privileged certainty.
7. Lecture 7 (The Definition of Perception) — establishes that perception is the appearance of an object at a brain-involving place and that sensation is its theoretical non-mnemic core.
8. Lecture 8 (Sensations and Images) — establishes that sensations are the intersection of mind and matter and that images differ from sensations only in their causal laws, being defined by mnemic causes.
9. Lecture 9 (Memory) — establishes that true memory is a present image plus a distinctive belief-feeling of pastness, with the time-reference in the feeling rather than the content.
10. Lecture 10 (Words and Meaning) — establishes that meaning is a causal relation — a word or image means what shares its mnemic effects — and that thinking is the use of signs for what is remote.
11. Lecture 11 (General Ideas and Thought) — establishes that no abstract ideas are needed: a generic image plus a judgment of similarity and difference constitutes generality, and all thinking consists of images and words.
12. Lecture 12 (Belief) — establishes that belief consists of a belief-feeling (memory, expectation, or bare assent) attached to a complex content or proposition.
13. Lecture 13 (Truth and Falsehood) — establishes that truth is a formal correspondence or "pointing towards" an objective fact, that no intrinsic criterion works, and that knowledge additionally requires accuracy and appropriateness.
14. Lecture 14 (Emotions and Will) — establishes that emotions are processes containing confused perceptions and that will adds no new ingredient beyond kinaesthetic imagery and judgment.
15. Lecture 15 (Characteristics of Mental Phenomena) — establishes the final answer: mind and matter are logical constructions from neutral particulars, distinguished by causal laws (subjectivity and mnemic causation), with mind a matter of degree.

---

Common misunderstandings

Misunderstanding: Russell denies the existence of the mind. He analyzes the mind, he does not eliminate it. Believing, desiring, remembering, and willing are all real occurrences; his claim is only that they are built out of sensations and images, not that they are illusions.

Misunderstanding: "Consciousness does not exist" means we are never conscious. James's and Russell's point is narrower: "consciousness" as an entity — a stuff or a peculiar relation — does not exist. We do have experiences and thoughts; but being "conscious of" something turns out to be a complex relation involving images and beliefs, not a fundamental quality. Sensation itself is not consciousness.

Misunderstanding: Neutral monism is a disguised materialism. Russell insists the reverse: "matter is not so material and mind not so mental as is generally supposed." Matter is a logical fiction constructed from appearances; physics describes particulars grouped by active place. If anything, the final lecture says psychology is nearer to what actually exists, because physical laws can only be stated in terms of the constructed entity "matter."

Misunderstanding: Russell is a behaviorist. He sympathizes with behaviorist method and borrows its critique of introspection, but he rejects its central denials — that there are no images, that introspection is no source of knowledge — and he concludes that behaviorism fails because it rests on an inadequate philosophy of physics.

Misunderstanding: Sensation is a kind of knowledge. Russell explicitly denies this: the pure sensation is not cognitive. It becomes knowledge only through memory, belief, and the correlations that turn it into perception.

Misunderstanding: Desire is an inner state that causes action. The book reverses this: desire is a causal law of behavior. A behavior-cycle is pushed from behind by discomfort; the "end" is a retrospective description of what brings quiescence. Conscious desire is just desire plus a (possibly false) belief about its purpose.

Misunderstanding: Images are faint sensations. Faintness (Hume) and lack of "aggressiveness" (Stout) fail as criteria — a just-visible star is a faint sensation, and Macbeth's image can unfix his hair. Images differ from sensations in their causes and effects (mnemic causation), not intrinsically.

Misunderstanding: The "unconscious" is a mysterious Freudian realm. Russell demythologizes it: an unconscious desire is merely a desire not accompanied by a true belief about its purpose. It has the same status as a force in dynamics — a convenient way of describing behavior, not an underground prisoner.

Misunderstanding: Truth is subjective or psychological. Truth and falsehood depend on a belief's relation to a fact outside the believer. No intrinsic mark — self-evidence, coherence, or certainty — distinguishes true from false beliefs; only verification by the happening of what is expected gives a practical, fallible guide.

---

Central paradox / key insight

The book's most counterintuitive idea is that mind and matter are not opposed substances at all. The same particulars — a patch of colour, a noise — are members of the physical world when collected into the systems we call objects and members of the mental world when collected into the perspective of a perceiver. There is no impassable gulf between the two because there was never a gulf in the first place: both are different groupings of one neutral stuff. Russell states the paradox as early as the first lecture and never abandons it.

The stuff of which the world of our experience is composed is, in my belief, neither mind nor matter, but something more primitive than either. Both mind and matter seem to be composite, and the stuff of which they are compounded lies in a sense between the two, in a sense above them both, like a common ancestor.

The deepest consequence: "consciousness," which philosophy had made the essence of mind, is a complex and accidental outcome of linguistic habits; and matter, which physics had made the essence of the physical, is a logical fiction constructed from appearances. The two most certain things in the traditional picture — that I am conscious and that matter is real — are both demoted in favor of a world of transient particulars whose causal laws admit of a single, unified science.

---

Important concepts

Neutral stuff / neutral monism — The doctrine, taken from James and the American new realists, that the "stuff" of the world is neither mental nor material: mind and matter are both constructed from it. Sensations are subject to both physical and psychological causal laws and are therefore truly neutral; entities subject only to physical laws are purely material, those subject only to psychological laws purely mental.

Particulars — The transient, brief existents that make up the world — such as occur in seeing and hearing, together with images more or less resembling them. They are "relative particulars" in that they may themselves be complex; ultimate things like points and instants are logical fictions constructed from them.

Sense-datum — On the old view, the object of sensation distinguished from the act of sensing it. Russell argues that once the subject (the "act") is discarded as a logical fiction, the sense-datum and the sensation collapse into one: the patch of colour simply is our sensation of it.

Sensation — The non-mnemic core of a perception; the intersection of mind and matter. It is not itself knowledge, but it supplies the data for knowledge of the physical world, including our own bodies.

Image — An occurrence distinguished from a sensation only by its causation: images have mnemic causes (they are occasioned by association, by a sensation or another image), while sensations are caused by stimuli external to the brain. Images are private, are copies of antecedent sensations in their simple parts, and are essential to thinking, memory, and meaning.

Perception — The appearance of an object from a place where there is a brain (with sense-organs and nerves in the intervening medium), distinguished by giving rise to and being affected by mnemic phenomena. All actual perception is vague and confused to some degree.

Perspective / biography — A perspective is the set of particulars that are appearances of different things at a given place and time — "the view of the world from that place"; a biography is the series of perspectives through time. Subjectivity is the characteristic of perspectives and biographies.

Active / passive place — Every particular relevant to physics has two places: the active place of the object whose appearance it is, and the passive place where it occurs. Physics groups particulars by active place; psychology by passive place.

Mnemic causation — Causation in which the proximate cause includes a past event: "A, B, C,... in the past, together with X now, cause Y now." The past items are the mnemic cause, X the occasion, Y the reaction. A disposition is the mnemic portion of a mnemic causal law.

Engram — Semon's hypothetical material alteration in the organism left by a stimulus, in virtue of which the recurrence of part of a stimulus-complex "ekphores" the whole reaction. Russell treats it as a useful but unverified hypothesis, preferring the observable formulation of mnemic causation.

Behavior-cycle — A series of voluntary or reflex movements tending to cause a certain result and continuing until that result is caused; the purpose of a cycle is the result that brings it to an end, normally by quiescence; an animal desires the purpose of its behavior-cycle while it is in progress.

Discomfort / pleasure — Causal properties of mental occurrences: discomfort stimulates movements tending to produce its cessation; pleasure stimulates no such movements, or only such as prolong the occurrence. Pain, the sensation, must be distinguished from discomfort.

Instinct / habit — An instinctive movement is a vital movement independent of prior experience; a learnt movement or habit is due to previous experience of similar situations. Instinct supplies the driving force of learning.

Meaning — The relation of a word or image to what it designates, constituted by mnemic causal laws: a word or image means an object when it has the same (or nearly the same) mnemic effects as the object.

Proposition — The content of a belief expressed in words (a word-proposition) or consisting of images (an image-proposition); propositions are what is true or false.

Belief-feeling — The feeling or complex of sensations which constitutes believing: at least three kinds — memory, expectation, and bare assent — differing in their time-reference, attached to a content by the relation expressed by saying the content is what is believed.

Content / objective — The content of a belief is the present complex occurrence believed (words and/or images); the objective is the fact, which may lie outside the believer's experience, that makes the belief true or false; the relation between them is the belief's reference.

Consciousness — Defined as the relation of an image of a word to an object (meaning) together with belief; a sensation becomes an object of consciousness when remembered or used as a sign of something else. It is complex and far from universal, "mainly a trivial and unimportant outcome of linguistic habits."

Immediate memory / specious present — The retention of the immediate past in a condition intermediate between sensation and image; fading sensations are "akoluthic." This gives us experience of succession and bridges sensation and image.

Familiarity / pastness — The two feelings by which memory-images are trusted and assigned places in the time-order: familiarity leads us to trust memories; pastness gives the sense of temporal remoteness.

Accuracy / appropriateness — The two characteristics of a knowledge-response: accuracy is different responses to relevantly different stimuli and the same response to irrelevant differences; appropriateness is suitability for realizing one's purpose. Both are required for knowledge.

Data — The propositions (and objects) whose truth is known without demonstration; ultimately, outside psychology, data consist of sensations, and the data of the physical sciences are also psychological data.

---

References and Web Links

Primary text (full text, public domain)

• Project Gutenberg: The Analysis of Mind by Bertrand Russell, eBook #2529 — https://www.gutenberg.org/ebooks/2529
• Plain-text version (UTF-8) of the same etext — https://www.gutenberg.org/ebooks/2529.txt.utf-8
• Project Gutenberg author page for Bertrand Russell (25 ebooks) — https://www.gutenberg.org/ebooks/author/355

Editions and publisher pages

• Routledge Classics edition (2022), with an introduction by Thomas Baldwin, ISBN 978-1-003-30893-5 — https://www.taylorfrancis.com/books/9781003308935
• Digital object identifier for the Routledge edition — https://doi.org/10.4324/9781003308935
• The 1921 first edition was published by George Allen & Unwin, London; the Dover edition (2005) reprints the same 15-lecture text.

Scholarly background (Stanford Encyclopedia of Philosophy)

• "Neutral Monism" — the entry covering Mach, James, and Russell, with extensive quotation from The Analysis of Mind — https://plato.stanford.edu/entries/neutral-monism/
• "Bertrand Russell" — the entry on Russell's life and philosophy, including a section on his theory of neutral monism — https://plato.stanford.edu/entries/russell/

Overview and context

• Wikipedia: "The Analysis of Mind" — article on the book's context, contents, and reception, with the full reference list — https://en.wikipedia.org/wiki/The_Analysis_of_Mind

Supplementary commentary (secondary source)

• Roy Lotz, "Review: The Analysis of Mind" (2017) — a general reader's critical review, useful as a supplement but not a substitute for the primary text — https://lotzintranslation.com/2017/10/08/review-the-analysis-of-mind/

Works Russell draws on directly (mentioned in the book)

• William James, "Does 'Consciousness' Exist?" (1904), reprinted in Essays in Radical Empiricism — the source of the critique of consciousness that Russell adopts.
• Ernst Mach, The Analysis of Sensations (1886; English translation, Open Court, 1914) — "a book of fundamental importance" for the neutrality of sensations.
• Richard Semon, Die Mneme (1904) and Die mnemischen Empfindungen (1909) — the sources of the engram theory and mnemic phenomena.