Behaviourism

Behaviourism begins from a methodological fact: another person's mind is not observed independently of what that person does. Speech, movement, learning, practical choice, and response to the environment are the public evidence from which mental attributions are made. The philosophical question is whether behaviour is merely evidence for inner states or whether talk of inner states can be analysed entirely in behavioural terms.

The answer separates three doctrines often grouped under one name. Methodological behaviourism is a policy for scientific inquiry, psychological behaviourism is an empirical research programme about learning and behavioural control, and logical behaviourism is a thesis about what mental statements mean. The first can survive the failure of the third. That distinction is behaviourism's lasting lesson: rejecting a reduction of mind to behaviour does not make behaviour optional as evidence.

Three behaviourisms

Methodological behaviourism restricts scientific explanation to publicly observable and operationally defined variables. Early psychology sought to replace introspection with measurements of stimulus and response. The policy does not by itself deny inner states; it denies them a role unless they enter testable relations.

Psychological or radical behaviourism, associated especially with Skinner, explains behaviour through histories of conditioning, reinforcement, stimulus control, and environmental contingencies. Skinner did not simply deny events within the skin. He denied that positing inner representations or autonomous mental causes supplied the right explanatory level when behaviour-environment relations could do the work.

Logical or analytical behaviourism, associated with Ryle and influenced by logical positivism, makes the constitutive claim. To be in a mental state is to behave or be disposed to behave in appropriate ways. Mental vocabulary does not name a private object behind behaviour; it characterises patterns, capacities, tendencies, and their contexts.

Only the last is directly a position on the mind-body problem. It promises dissolution rather than interaction: if mind-talk is a way of talking about bodily dispositions, no extra mental entity or property must be connected to the body.

Ryle and the category mistake

Ryle attacked what he called the dogma of the Ghost in the Machine. The Cartesian picture treats mind as one object and body as another, then asks how the objects interact. Ryle's diagnosis is that this treats dispositional and organisational concepts as if they named episodes in a hidden theatre.

His model example is the visitor shown colleges, libraries, and laboratories who then asks where the university is. The university is not one more building alongside its parts. Likewise, intelligence is not a hidden process added to intelligent activity; it is displayed in how activity is conducted - flexibly, aptly, and with sensitivity to error and context.

This is not the crude claim that thinking consists only in visible muscular movement. Ryle's analyses appeal to multi-track dispositions. To believe that ice is thin is to be disposed, across indefinitely many circumstances, to warn others, avoid stepping on it, answer relevant questions, update plans, and draw appropriate inferences. The open-ended pattern matters, not one stereotyped response.

The category-mistake diagnosis has real force against the image of an inner duplicate performing every intelligent act. Positing an inner rule-reading episode to explain intelligent performance creates a regress: applying the rule intelligently would require another act of interpretation. Some know-how is manifested in competent action rather than derived from prior contemplation of propositions.

The attraction of public criteria

Behaviourism promises three methodological gains.

Public testability. Mental attribution is anchored in evidence available to more than one observer. Psychology need not depend on inaccessible introspective objects.

Parsimony. No non-physical substance or irreducible property is added. The mind-body relation is not a causal bridge between domains; mental descriptions organise patterns in bodily and environmental interaction.

Attention to context. A movement is not behaviour in the relevant sense apart from its circumstances and counterfactual relations. Raising a hand may answer a question, hail a vehicle, vote, stretch, or occur involuntarily. Behavioural meaning lies in a structured pattern, not in motion alone.

These insights persist in functionalism, dispositional theories of belief, embodied cognition, and contemporary operational practice. Behaviourism helped dislodge the assumption that mental explanation must consist of introspected inner objects.

Pretence, suppression, and the actor

The simplest counterexamples separate a state from its ordinary manifestation.

• A trained actor displays pain behaviour without pain.
• A "super-Spartan" experiences intense pain while suppressing every ordinary sign.
• A person believes it will rain but wants to deceive an observer and leaves the umbrella behind.
• Paralysis blocks action while leaving perception, intention, and experience intact.

A sophisticated behaviourist responds with dispositions rather than actual behaviour. The actor is not disposed to respond as a person in pain across the full range of counterfactual circumstances; the super-Spartan would respond under conditions where the motive to suppress was removed.

The response risks becoming unfalsifiable. For any absent behaviour, a background belief, desire, bodily limitation, or context can mask the disposition. If the relevant disposition is "to behave as if in pain unless something prevents it," identifying the preventing conditions may require prior use of pain, belief, desire, and intention. The analysis no longer reduces the mental vocabulary; it redescribes a network of it.

The circularity problem

Mental states do not map one by one onto behaviour. What a belief causes depends on desires and other beliefs:

$$\text{belief}+\text{desire}+\text{context}⟶\text{behaviour}.$$

Believing that the train leaves at noon may produce running, waiting calmly, cancelling the trip, or no movement, depending on one's goals and further information. No finite list of behavioural dispositions defines the belief independently of the rest of the mental economy.

Logical behaviourists can embrace holism: the whole network is analysed at once. But a behavioural description rich enough to distinguish every belief-desire network must include sensitivity to reasons, representations, learning, and counterfactual states. At that point it approaches functionalism. Functionalism retains the causal organisation while allowing internal states to mediate inputs and outputs; it does not require translation into purely behavioural language.

Phenomenal experience

Conscious experience supplies the most intuitive objection. A person can distinguish between feeling pain and merely behaving as if in pain, and the distinction appears to concern what the state is like, not an unmanifested tendency. Even a perfect behavioural duplicate seems conceivably to lack experience.

Rylean analyses are strongest for abilities, intelligence, and dispositional attitudes; they are less natural for occurrent sensation. One can define pain through aversion, attention, report, and withdrawal, but the analysis appears to leave out hurting. The behaviourist may deny that an ineffable residue has coherent identity conditions apart from its role. That reply anticipates illusionism, but it needs an account of why the seeming of a residue occurs and why it is systematically reported.

The problem is not that experience is wholly private and behaviour irrelevant. Reports, discrimination, and action are indispensable evidence. It is that evidence for a state does not ordinarily constitute the state. Smoke is evidence of fire without being what fire means; pain behaviour may stand in an evidential relation to pain of the same form.

Methodological behaviourism after logical behaviourism

Psychology did not return from behaviourism to unrestricted introspection. It retained operational definitions, controlled tasks, learning theory, and suspicion of explanatory homunculi. The cognitive revolution added internal representations and computational processes because they supported prediction, unification, and novel intervention, not because privacy became scientifically respectable by declaration.

Contemporary experiments still define variables through performance and report. The method and evidence distinction is crucial: an operational measure provides a procedure for detecting a target under assumptions; it does not define the target for every theoretical purpose. Reaction time can index processing difficulty without difficulty being identical with reaction time.

Behaviour analysis also remains powerful where environmental contingencies explain action more directly than speculative inner stories. Reinforcement histories, affordances, and social practices shape behaviour whether or not one accepts the behaviourist theory of mind. The error would be to infer from explanatory usefulness to metaphysical completeness.

Behaviour and other minds

Behaviourism's strongest legacy concerns epistemology. Other minds are known through embodied interaction, not through direct inspection of a private medium. Rich, context-sensitive behaviour can justify mental attribution even when anatomy or substrate differs.

This matters for animals and machines. Requiring human speech would build species or architecture into the evidence. Accepting any matching output would be too liberal. The relevant evidence includes learning, transfer, flexible correction, cross-modal integration, sensitivity to reasons, and stable behaviour across interventions.

No behavioural profile is theory-neutral proof of phenomenal consciousness. But the opposite claim - that behaviour supplies no evidence because consciousness is private - would make ordinary knowledge of other human minds equally impossible. Behaviourism fails as a complete metaphysics while identifying the public basis any account of other minds must explain.

From behaviourism to functionalism

Functionalism preserves behaviourism's relational insight and rejects its restriction to the organism's surface. A mental state is characterised through typical causes, relations to other internal states, and effects on behaviour. The intervening states can explain why the same stimulus produces different actions and why one state remains unmanifested.

The shift can be represented schematically:

$$\text{stimulus}\to \boxed{\text{internal causal organisation}}\to \text{behaviour}.$$

Behaviourism analyses the box through the pattern connecting its edges. Functionalism treats places within the box as real organisational states while remaining neutral about their material realisers. This solves circularity by defining mental terms simultaneously through a causal network, but inherits liberalism and phenomenal objections in a new form.

Assessment

Logical behaviourism failed because mentality can be masked, simulated, and mediated by other mental states, and because phenomenal character appears not to reduce to a disposition. Its failure did not restore a Cartesian theatre. It cleared the way for a more adequate relational theory - functionalism - and left a permanent constraint: claims about hidden minds must answer to public patterns of action and interaction.

Selected references

• Hempel, Carl G. "The Logical Analysis of Psychology" (1935).
• Ryle, Gilbert. The Concept of Mind (1949).
• Skinner, B. F. Science and Human Behavior (1953).
• Skinner, B. F. "Behaviorism at Fifty" (1963).
• Watson, John B. "Psychology as the Behaviorist Views It" (1913).