Does philosophical behaviourism give the correct account of mental states? (25)
Philosophical behaviourism is a physicalist theory which gives a linguistic, rather than metaphysical, account of mind:
behaviourists argue that the language we use to talk about the mind, i.e. mental concepts, can be reduced to
physical concepts. Behaviourism is the claim that we can explain mental concepts in terms of behaviour, which is
physical. This eliminates the ontological question of whether mind and body are distinct, because ‘the mind’ simply
refers to behaviours – it is not something distinct at all. There are two types of behaviourism: Carl Hempel’s ‘hard’
behaviourism, which claims that all mental states can be reduced without loss of meaning to behaviours, and Gilbert
Ryle’s ‘soft’ behaviourism, which claims that propositions about mental states are propositions about behavioural
dispositions. I will argue that, while Ryle’s form of behaviourism is more convincing, ultimately behaviourism does
not give a correct account of mental states because it is defeated by the objections from a) multiple realisability and
circularity, b) asymmetry, and c) Putnam’s ‘Super-Spartans’ thought experiment.
One objection to behaviourism is that it fails to explain the fact that the same mental state can be realised through
multiple behaviours, even in the same circumstances. For example, both person A and person B experience the
mental state of fear when they see a snake, but while A runs, B freezes. This presents a problem for hard
behaviourism in that a complete explanation of mental states in terms of behaviour is not possible, because there
are potentially infinite conditions of verification (behaviours) to determine any mental state. For soft behaviourism,
if people with the same mental state in the same circumstances can have different dispositions to behaviour, mental
states cannot be reduced to behavioural dispositions.
Moreover, the multiple realisability of behaviours is determined by other mental states. For example, person B’s
disposition to freeze rather than to run depends on their belief that this type of snake responds to movement, on
their recognition of the type of snake etc. This objection can be extended to argue that behaviourism invokes
circularity in its attempt to define mental states. In analysing mental states, the behaviourist will always have to refer
to other mental states – they cannot specify the conditions of verification for fear without referring to other mental
states, such as ‘if person B believes that the snake responds to movement, they will freeze.’ This is problematic
because circular analysis is unsatisfactory and behaviourism fails to give an adequate account of what mental states
are.
This objection particularly targets hard behaviourism because it is reductive and claims mental states can be
translated into behaviours. Hempel can respond that while behaviour varies, the brain states remain the same for all
human beings experiencing that mental state. Hempel’s behaviourism includes neurological processes as behaviours.
This also resolves the issue from circularity, because other mental states such as beliefs do not have to be factored in
– mental states can be precisely defined as brain states. However, by reducing mental states to brain states,
Hempel’s theory collapses into a linguistic version of mind-brain type identity theory; furthermore, this theory still
faces the objection from multiple realisability, because different species may realise the same mental state (e.g.
pain) with different brain states. For example, an octopus can realise pain, but they don’t have c-fibres. The issues
from circularity and multiple realisability have shown that Hempel’s hard behaviourism is incorrect.
Ryle is better able to respond to this objection, because soft behaviourism is not reductive, and claims that mental
concepts can be explained in terms of dispositions to behaviour, rather than actual behaviours. Disposition
statements are ‘open’ and cannot be replaced with a complete set of hypothetical statements (e.g. fear is the
disposition to have an increased heart rate and inability to focus - this does not require you to state all the
conditions under which you would behave this way, because it is only a disposition to behaviour). Ryle is able to get
around the issue of circularity in this way. Moreover, he agrees that mental concepts are ‘indefinitely
heterogeneous’, meaning they may be realised through indefinite behavioural dispositions. Ryle concedes that
mental concepts cannot be replaced with physical concepts and his theory does not aim to offer complete
translations of mental concepts. However, we may object that the second problem from multiple realisability
remains a problem because he fails to explain how we can meaningfully talk of mental states if they can be realised
by indefinite behavioural dispositions. Ryle can respond by claiming that in reality, people with the same mental
state have very similar dispositions and there is sufficient overlap to identify mental states.