In the philosophy of mind, logical behaviorism (also known as analytical behaviorism)[1] is the thesis that mental concepts can be explained in terms of behavioral concepts.[2]
Logical behaviorism was first stated by the Vienna Circle, especially Rudolf Carnap.[2] Other philosophers with sympathies for behaviorism included C. G. Hempel, Ludwig Wittgenstein, and W. V. O. Quine (1960).[2] [3] A more moderate form of analytical behaviorism was put forward by the Oxford philosopher Gilbert Ryle in his book The Concept of Mind (1949).[4] [2]
Generally speaking, analytic behaviourism is the view that propositions about the mind, or about mental states more generally, are reducible to propositions about behaviour. For example, a dualist would take 'Finbarr is in pain' to refer to a private, non-physical mental state within Finbarr's mind. But a behaviourist would say that 'Finbarr is in pain' simply refers to Finbarr's behaviour, or his disposition to behave in a certain way. So, the behaviourist might argue that if Finbarr was crying, and this was the reason that the mental state of pain was attributed to Finbarr, then 'Finbarr is in pain' reduces to 'Finbarr is crying'. In other words, 'Finbarr is in pain' means the same thing as 'Finbarr is crying' as, for a behaviourist, statements about mental states merely refer to people's behaviour, or their dispositions to show certain behaviour (pain behaviour, in Finbarr's case).[5]
Following Hempel's behaviourist theory (sometimes called hard behaviourism), which alleged that all propositions about mental states were reducible, without loss of meaning, to propositions about bodily states and behaviour, Gilbert Ryle produced a modified, less extreme form of behaviourism (sometimes called soft behaviourism). Ryle sets out in The Concept of Mind to destroy the illusion of Cartesian Dualism, which he says has produced a widespread acceptance of the 'dogma of the ghost in the machine'- the belief that the mind is an immaterial 'thing' caged within a body. To introduce his behaviourism, Ryle proposes his great criticism of Cartesianism: that it performs a category mistake. Ryle believes that mind-body dualism mistakenly puts the mind in the category of a 'thing', a non-physical entity that exists, driving our actions. But, says Ryle, the mind is not a thing. It is simply a way of talking about behaviour, specifically the dispositions of people to act in certain ways. So, whereas for Hempel 'Finbarr is in pain' reduces to 'Finbarr is crying', Ryle's soft dispositional analysis might say that it means 'Finbarr has a disposition to cry, or shout in pain, or hold onto something for support'. In other words, the mind is not a thing- propositions about mental states are instead a way to express the dispositions of people to act in certain ways.
Hilary Putnam criticises behaviourism by arguing that it confuses the symptoms of mental states (behaviour) with the mental states themselves. Mental states, says Putnam, are distinct from behaviour, and this is something that behaviourism overlooks. Putnam proposes a thought experiment to show the distinctness of mental states from behaviour, and therefore show behaviourism to be false. In Brains and Behaviour, Putnam gives the example of 'X-Worlders', sometimes called 'super-super Spartans'. These are great warriors who have so strongly repressed the urge to display signs of pain that they no longer have any pain behaviour, nor any disposition to display pain behaviour. When an X-Worlder is stabbed, they feel a terrible pain, and yet they show no pain behaviour, nor do they feel any disposition to show their pain. This, says Putnam, shows behaviourism to be false- in this situation, there is no corresponding behaviour with the X-Worlder's pain, showing mental states to be distinct from behaviour.
It has also been argued that behaviourist analysis of mental state terms can never truly be adequately completed. This is due, in large part, to the fact that mental states are multiply realisable in behaviour. In other words, the same mental state can be manifested by an infinite number of behaviours. For example, when Finbarr feels pain, he can scream. Or shout. Or do nothing. Or cry himself to sleep. The number of pain behaviours that Finbarr can display is infinite. But, in order to accurately define a mental state in terms of behaviour, all the possible ways in which a mental state might be manifested need to be taken into account. But the ways mental states can be reflected in behaviour are infinite, showing that mental states cannot be adequately defined in terms of behaviour. This is also an issue for Ryle's soft behaviourism, as someone who is, for example, angry has dispositions to manifest this behaviour in an infinite number of ways.
Another issue, on top of multiple realisability, is the fact that behavioural analysis of mental states becomes circular. According to its critics, behaviourism neglects that whether a mental state is manifested in behaviour depends on its interaction with other mental states. For example, my mental state of wanting to drink my glass of water might not be manifested in behaviour (ie. by me drinking the water) if I believe that my water is poisoned. Ryle says that 'John wants to drink' the water means 'John will drink the water if... or if... or if...', where the 'if...' expresses a condition for John drinking the water- for example, John might drink the water if he's thirsty, or if he's tired, or if he's hallucinating and thinks that it's magical water. But whether John drinks the water is not as straightforward as one disposition to drink water or not- the desire to drink the water also depends on other mental states. If John thinks the water is poisoned, he will not drink. If John does not want to be needing the toilet within the hour, he will not drink the water. Behaviourism needs to recognise all of these other mental states which affect our dispositions, and yet this simply introduces more mental state terminology into the behavioural analysis. This mental state terminology must then be analysed in terms of behaviour to complete the analysis, which will in turn introduce more mental state terminology, as these other mental states that have been introduced also depend on other mental states. Therefore, the analysis from mental state to behaviour is circular and cannot be completed.
Behaviourists hold that propositions about mental states are about behaviour. But this does not seem to account for the asymmetry between how I talk about my own mental states and the mental states of others. For a hard behaviourist, the proposition 'I'm afraid' is apparently reducible to a proposition about behaviour- in other words, it means the same thing as a proposition about behaviour. But when I say that I'm afraid, I am not referring to my behaviour. I don't mean 'I'm shivering and crying and have a frown on my face.' I'm referring to my mental state of fear. Therefore, it would seem that the way I talk about my own mental states and the mental states of others are radically different. I do not need to observe my own behaviour to know I'm in pain. I just know, through introspection, which would show that a purely behavioural analysis of mental states is inadequate. I apply mental states to myself not because of my behaviour but because I am experiencing the mental states themselves.
Logical behaviorism is called "logical", after the idea adopted by Bertrand Russell, that mathematics can be described in terms of formal logic, using Set Theory, and thus make it "scientific", "provable", "specific", consistent and "truthful". In a similar way, it was thought by the Vienna Circle that the phenomena of human mental states such as feelings, perceptions, imaginations etc. can be described in terms of a tendency to behave in a certain way, which could then be tested and explained scientifically through the methods of Behaviorism, whereby everything consists of stimulus-response pairs, with various types of origins and different types of reinforcement.[6]