On Hilary Putnam’s “The Nature of Mental States”

by Vered Arnon

Part I. Analysis.

Hilary Putnam’s article “The Nature of Mental States” directly addresses the question of whether or not pains are brain states. Putnam argues that this question is not an identity claim arising from empirical reduction. He goes on to propose a functional state theory of pain as a superior hypothesis compared to the brain state hypothesis and the behavioural disposition hypothesis.

Putnam’s article can be divided into five different arguments.

I. Identity Claim Question.

Putnam explains the position behind the assertion that to say “pain states are brain states” is meaningful, and he also explains the position behind the assertion that it is not meaningful to say “pain states are brain states”. One problem with identity claims is the combining of the notions ‘property’ and ‘concept’ into one notion, but obviously one could not reasonably assert that the concept of one thing is identical to the concept of another thing, regardless of whether or not both concepts pertain to the same individual. Another problem with identity claims is empirical reduction. But to discard identity claims in favour of claims of correlation is problematic too, because it opens and excludes what Putnam considers “the wrong questions”. He rejects both positions, and proposes instead the assertion that “pain is A” where ‘pain’ and ‘A’ are distinct, not synonymous, and “pain is A” is not an identity claim.

II. Proposed hypothesis that pain is a functional state of a whole organism.

Putnam introduces the notion of a Turing machine. This machine operates as a probabilistic automaton, and he claims all organisms capable of feeling pain are probabilistic automata. Pain he describes as a subset of the sets of sensory inputs and motor outputs. These inputs and outputs create a kind of functional organisation. Putnam admits that this hypothesis is vague, and adds that one must also be able to distinguish which set of inputs referred to pain, as well as sensory organs that do this distinguishing and a preference function. Putnam claims that this hypothesis is less vague then the physical-chemical state hypothesis.

III. Argument against the brain state theory.

Putnam says that brain state theorists assume only physical properties and consider non-physical properties unintelligible. Brain state theory is also said to be incompatible with dualism and mentalism. While brain state theorists are proud of this, Putnam considers it a weakness in the theory, because functional state theory, while it doesn’t assert dualism or mentalism, is not incompatible with them at all.

Putnam also claims that brain state theory has to specify a universal species-independant physical-chemical state that leads to the same physical correlate of pain. This is impossible to do. Since the brain state theorist maintains that every psychological state is a brain state, one should not be able to find one state which can be clearly applied to two different species when their physical-chemical state is not the same. But Putnam gives an example of applying ‘hungry’ to both a mammal and an octopus. Their physical-chemical states are entirely different, but they can have the exact same psychological state. Putnam concludes that brain state theory fails because neurophysiological laws haven’t been developed to the point of being species-independent.

IV. Argument against the behavioural disposition theory.

Putnam claims that behavioural disposition theory defines the state of being in pain using the notion of pain. He seems to be accusing the theory of being circular. The theory does not adequately specify the behaviour. It is very vague. Animals can exhibit the same behaviour when in pain and when not in pain, the correlation between pain and behaviour is not invariable. He also claims that behavioural disposition theory would require a species independent invariant correlation. The theory equates pain with a behaviour disposition without explaining that behaviour, Putnam charges. He says that functional state theory is more plausible, and while that plausibility is subjective, he seems to imply that it is no more subjective than behavioural disposition theory, which is implausible and vague on top of being subjective. The only valid part of behavioural-disposition theory, Putnam maintains, is also part of functional state theory, but functional state theory doesn’t share the problems and difficulties of behavioural disposition theory.

V. Methodological justification of functional state hypothesis.

Putnam gives three methodological points in favour of functional state theory. Identifying psychological states, including pain states, with functional states allows for clear easy deriving of psychological laws. It also explains pain rather than merely correlating it with something else. Finally, functional state theory prevents inquiry from being distracted towards what Putnam considers the “wrong” sorts of questions.

Part II. Criticism.

I. Putnam’s article is outdated. It was written in 1967. Much research has happened since then in the area of neurophysiology.

II. If pain is a functional state, that does not rule out it being a brain state. Putnam writes that one needn’t know the internal process of the machine to analyse its response to pain stimuli. However, those unseen processes could very well be physical-chemical processes and brain states. The functional state hypothesis is not incompatible with the brain state hypothesis.

III. The demand to find species-independent laws for pain is inappropriate. Pain might be species-dependant. Putnam seems to think that brain state theorists would necessarily have to maintain that pain is exactly the same physical-chemical state across the board in every single individual and species, and occurs in exactly the same configuration in the same type of brain. But since different species have different brain structures and physical-chemical composition, it seems logical to me to allow for different configurations from species to species of the phenomena we call pain.

IV. In advancing his argument against the brain state theory, Putnam seems to be arguing in favour of the behavioural-disposition theory. He writes that “we identify organisms as in pain… on the basis of their behaviour.” But then when arguing against behavioural disposition theory he claims that no behaviour is invariably correlated with pain.

Basically the whole argument in favour of functional state theory over brain state theory in the last paragraph of part III works very well as an argument in favour of behavioural disposition theory.

V. Putnam admits that his hypothesis is vague, and lacks a) specification of the kind of functional organisation that an organism must have to be capable of feeling pain, and b) a way to distinguish which sensory inputs meet the criteria for causing pain opposed to other sensory inputs which don’t, and c) a “preference function” – all of which are not vague at all in most recent neurophysiological research.

At best, Putnam’s machine could provide a model that helps to study and organism’s functional response to pain, but it doesn’t actually answer any questions about the nature of pain itself, and it does not rule out either of the two hypotheses he argues against.

Pain could be a functional state, a brain state, and a behavioural disposition. The three hypotheses are not exclusive, and nothing in Putnam’s article shows that functional state theory is not compatible with, or just another way of examining, the other two.

Leave a Reply

Your email address will not be published. Required fields are marked *