Is Philosophical Behaviourism Correct?
Philosophical behaviourism is not correct. The multiple realisability objection defeats hard behaviourism but soft behaviourism can avoid it. The Super-Spartans objection threatens both forms but leaves a residual line of defence. The asymmetry between self-knowledge and knowledge of others is fatal to both forms with no convincing reply — privileged introspective access is irreconcilable with behaviourism.
- Philosophical behaviourism is the view that propositions about mental states can be reduced to propositions about behaviour. There are two forms. Hard behaviourism, associated with Hempel, holds that all folk psychological propositions can be translated without loss of meaning into propositions about publicly observable bodily states and movements. Soft behaviourism, associated with Ryle, holds that mental state propositions are propositions about behavioural dispositions — tendencies to behave in certain ways under certain conditions - Both forms are motivated by the verification principle of the logical positivists: a statement is only meaningful if it is either analytic or empirically verifiable. Since mental states are private and unobservable, the only way to preserve the meaningfulness of folk psychological language is to translate it into language about publicly observable behaviour - A strength of behaviourism is its scientific credentials — it grounds the philosophy of mind in what is publicly observable and empirically accessible, avoiding the metaphysical mysteries of dualism and the interaction problem - I will argue that the multiple realisability objection seriously threatens hard behaviourism but can be avoided by soft behaviourism, that the Super-Spartans objection threatens both forms but has a partial reply, and that the asymmetry between self-knowledge and knowledge of others’ mental states is fatal to both forms with no convincing reply. I will conclude that philosophical behaviourism is not correct
Section 1: Multiple Realisability
- The multiple realisability objection is a strong and convincing objection to hard behaviourism. Hempel’s project requires a complete analytic reduction, and the open-ended diversity of behavioural expression across individuals and cultures makes this impossible in principle. The further reduction to purely physical terms compounds the problem - However the objection does not threaten behaviourism as a whole, because soft behaviourism can avoid it. Ryle’s dispositional account is explicitly designed to accommodate the variability of behavioural expression, and the multiple realisability point is entirely compatible with his view
: The Multiple Realisability Objection to Hard Behaviourism
- Hard behaviourism attempts to give a complete analytic reduction of mental state terms into behavioural terms — a finite, semantically lossless translation. The multiple realisability problem shows this is impossible - A single mental state type can be realised in a vast and potentially infinite range of different behaviours across different people, cultures, and circumstances. Consider anger: some people shout, some go quiet, some cry, some laugh, some seek revenge, some sulk. There is no single finite list of behaviours that captures every instance of anger. This means hard behaviourism cannot provide the complete analytic translation it requires - The problem strikes twice. First, the initial reduction from folk psychological language to behavioural language fails because there is no finite list of behavioural sentences equivalent to a mental state term. Second, the further reduction from behavioural language to purely physical language fails because any given behaviour can itself be realised in an infinite variety of physical movements
Ryle: Soft Behaviourism’s Response — Multi-Tracked Dispositions
- Ryle can avoid this objection. He does not attempt to provide a finite list of specific behaviours equivalent to a mental state. Instead, he identifies mental states with multi-tracked dispositions — complex tendencies to behave across a wide and open-ended range of circumstances. He explicitly accepts that mental states can give rise to an indefinite range of different behaviours, and his account accommodates this precisely because it operates at the level of dispositions rather than specific behavioural translations
Section 2: The Distinctness of Mental States from Behaviour
- The Super-Spartans objection is a strong and significant problem for both forms of behaviourism. It shows that the connection between pain and pain behaviour is at most contingent, and Ryle’s dispositional response, while initially plausible, is defeated by the Super-Super-Spartans case - However the objection does not seem entirely conclusive. A committed behaviourist can insist that one cannot genuinely be in pain without having at least some disposition to exhibit pain behaviour — that the Super-Super-Spartans scenario, on reflection, is incoherent because pain just is a certain kind of dispositional state. The objection is very powerful but leaves open a residual line of defence
Putnam: Putnam’s Super-Spartans
- Both hard and soft behaviourism identify mental states with behaviour or behavioural dispositions. Putnam’s Super-Spartans argue that this identification fails because mental states and behaviour can come apart - Putnam imagines a community of Super-Spartans who have been culturally conditioned to suppress all voluntary pain behaviour. They do not wince, scream, flinch, or take pain-killers. Yet they do genuinely feel pain and dislike it just as we do — it simply takes a great effort of will to suppress the behaviour. If this scenario is conceivable, then the connection between pain and pain behaviour is contingent rather than necessary, and behaviourism is false - The case of perfect actors makes the complementary point: someone can exhibit all the behavioural criteria for pain without being in pain at all. Behaviour is neither necessary nor sufficient for the mental state
Ryle: Ryle’s Response and the Super-Super-Spartans Counter
- Ryle can partially reply by appealing to dispositions. A Super-Spartan who suppresses pain behaviour still has the disposition to exhibit pain behaviour — they are inclined to express it but are overridden by a stronger disposition not to appear weak. The dispositional account accommodates the absence of occurrent behaviour without abandoning the identification of mental states with behavioural dispositions - However Putnam develops the argument further. He imagines Super-Super-Spartans — beings who have been Super-Spartans for so long that they have eliminated even the disposition to exhibit pain behaviour. They do not even have language about pain. Each individual may think privately about their pain, but they have no inclination to report or express it in any way under any conditions whatsoever. If Super-Super-Spartans are conceivable, then a person can be in pain with no behavioural disposition to express it — and Ryle’s dispositional account fails along with Hempel’s
Section 3: The Asymmetry Between Self-Knowledge and Knowledge of Others
- Ryle’s response is not without initial force — it is true that our greater familiarity with our own behaviour accounts for some of the difference. But this is not the core of the asymmetry objection and does not address it - The deeper point stands: self-knowledge via introspection is a distinct method of knowing, not merely a quantitative advantage in behavioural observation. No amount of behavioural data — however extensive — could give another person the direct first-person knowledge I have of my own pain simply by feeling it. Behaviourism cannot account for this and has no convincing reply - This objection applies equally to both hard and soft behaviourism. Hempel has no resources to accommodate privileged introspective access, and Ryle’s dispositional account does not help because the asymmetry concerns the method of self-knowledge, not its content
: The Asymmetry Objection
- Both hard and soft behaviourism imply that all knowledge of mental states is mediated through behaviour. If mental states just are behavioural states or dispositions, then the only way to know about any mind — including one’s own — is by observing behaviour. Since behaviour is publicly observable, one’s access to one’s own mental states should be no different in kind from one’s access to the mental states of others - But this is obviously false. There is a clear asymmetry between self-knowledge and knowledge of others’ mental states. If I imagine a coloured shape, no one but me can tell what colour I am imagining — even if I choose never to tell them and never exhibit any relevant behaviour. I have privileged access to my own mental states that others simply do not have. This privileged access is not explicable in terms of behaviour or behavioural dispositions, because no observable behaviour need be involved at all
Ryle: Ryle’s Response and Why It Fails
- Ryle replies that we simply observe more of our own behaviour than we observe the behaviour of others. We are with ourselves at all times, we attend to our own behaviour more carefully, and this greater familiarity with our own behaviour may explain why our self-knowledge appears richer and more direct - However Ryle’s response misses the point. The asymmetry is not merely a matter of the quantity of knowledge we have about ourselves versus others. It is a matter of the methods available for gaining that knowledge. Even if we were to observe another person’s behaviour the exact same amount as we observe our own, we would still expect them to know less about their own mind than they themselves do. This is because we have introspective access to our own mental states — we can know our pain by feeling it — whereas for others we only ever have their behaviour to go on. Ryle’s point only addresses the asymmetry in the amount of knowledge; it does not address the asymmetry in the methods of knowing. The privileged access we have to our own mental states via introspection is simply not available through behavioural observation of any kind
- The multiple realisability objection seriously damages hard behaviourism — there is no finite translation of mental state terms into behavioural terms — but soft behaviourism avoids it through its dispositional account - The Super-Spartans objection threatens both forms more seriously: the connection between pain and pain behaviour is at most contingent, and the Super-Super-Spartans case defeats even Ryle’s dispositional response. However a determined behaviourist can still maintain that Super-Super-Spartans are not genuinely conceivable, leaving a residual line of defence - The asymmetry between self-knowledge and knowledge of others is fatal to both forms with no convincing reply. Ryle’s response addresses only the quantity of self-knowledge, not the distinct method of introspective access. We can know our own pain by feeling it — a form of knowledge entirely unavailable through behavioural observation of any kind. Behaviourism cannot accommodate this and therefore fails - Philosophical behaviourism is not correct