the dungeon has...

Why might one think that Functionalism is better than Behaviourism?

To assess whether functionalism is any significant improvement on behaviourism we need to first consider both the theories in regard to their strengths and weaknesses. Once that has been done it will be possible to consider whether functionalism has a stronger position, or provides a better understanding of the mind, than behaviourism.

It is easiest, at least for me, to define Behaviourism in two forms, psychological behaviourism and philosophical behaviourism. Psychological behaviourism springs from the problem Wittgenstein raised with his ‘Beetle in the box’ analogy in his Philosophical Investigations. In this everyone had a box containing something, but they could only look inside their own individual box. All of them referred to the thing inside the box as a ‘beetle’, and understood each other when they referred to their own. However, the contents of the box may be different for each person, in which case they would not be seen to be referring to the same thing. This example suggests that due to the subjective nature of experience, although our language would seem to be referring to the same thing, it might not necessarily be doing so. This was applied by behaviourists to talk about the mind:

And if anyone insists that there are inner subjective criteria for psychological expressions as well, we should reply, argues the behaviourist, that even if such existed, they (like Wittgenstein’s beetles) could not be part of the meanings that can be understood and shared by different persons.[i]

This meant, for the psychological behaviourists, that it is not correct to refer to the inner workings of the mind at all, and that such things should be referred to instead entirely through observation of behaviour. A given situation would produce typical behaviour, reducing human actions to input/output behaviour. Since mental states were considered identical to our behaviour, descriptions of inner states could be deemed irrelevant:

Behaviourism held that mental states were identical with sets of actual and counterfactual overt behaviours and that inner states of the subject, though no doubt causally implicated in such behaviours, were not theoretically important to understanding what it is to be in certain mental states.[ii]

Philosophical behaviourism conversely was less inclined to dismiss the mind altogether. It talks instead about dispositions to behave. For example a brittle vase may be deemed disposed to smashing, but we do not call it the cause of the vase breaking. Examples like this can be applied to people in order to suggest how they might behave given certain dispositions. They concede that these dispositions may never be manifested at all, for example a brittle vase that never breaks. For example, Sam believes that it is going to rain. From this we would anticipate that he would look for his raincoat before leaving the house. Or, conversely, his looking for his raincoat signifies that he thinks it is going to rain. This version of behaviourism is not so quick to discard all talk of mental events, but it does still suffer from some of the same criticism.

The first and obvious argument is that talk of behaviour in such a way appears to make any talk of inner mental processes irrelevant. This would seem strange, as I would consider there to be far more to me than can be described purely in terms of my behaviour. John Heil summarises this point quite nicely:

If your having a headache is solely a matter of your behaving, or being disposed to behave, in a particular way, then the intrinsic qualitative nature of whatever is responsible for your so behaving, or being disposed to behave, is irrelevant.[iii]

This is exactly the point that seems to be being made by the psychological behaviourists. The example of the ‘beetle in the box’ means that ‘the intrinsic qualitative nature’ of our mental states is so indescribable in public terms that we should discard it in terms of something more accessible to everyday language.

However, it is not so easy to define behaviour in this way, as behaviour itself is subject to very many variations. For example if I take behaviour to mean, purely, bodily motions, without reference to intentions, then I would find myself very hard put to describe any action or type of action. Every time I go to drink tea, the specific motions of my arms etc. vary. I would also run into difficulty trying to describe actions such as looking in a mirror, as to describe someone in this way would be to presume that there is something mental occurring. As Donald Davidson said: ‘No matter how we patch and fit the non-mental conditions, we always find the need for an additional condition (provided he notices, understands, etc.)’[iv] This would appear to rule out behaviourism as a good theory of mind, purely due to the fact that it is difficult to even describe some forms of behaviour without recourse to mental states.

However dispositions, in this respect, would seem to be causing a different type of problem for the philosophical behaviourist. It would be good to say that given certain mental states certain behaviour would occur, according to the person’s dispositions. But, the type of reduction, which they would intend, does not seem at all feasible given the sheer number of possible dispositions to behave. Predictions of behaviour would become impossible, unless an extra clause were added. It would be necessary to add a ceteris paribus clause or something similar, in order to counteract the number of conflicting possible beliefs and desires that could have an effect on a person's behaviour. This would make such prediction however, merely trivial. For example, if I thought I was hungry and I believed that there was food in front of me, you would expect, given the extra clause, that I would eat it. Without that there may be the possibility that I thought it was poisoned, or believed myself allergic, or any of a large number of beliefs that could cause me to act in a different way.

In behaviourism beliefs and desires are not seen to be the cause of behaviour, since as the two are not distinguished separately, one cannot be seen to be the cause of the other. This causes problems as it indicates that when I am in pain I am disposed to shout in pain and hop on one leg, rather than the pain being the cause of such behaviour. This can be seen by what Hilary Putnam said:

…Both the dualist and the materialist would want to argue that, although the meaning of ‘pain’ may be explained by reference to overt behaviour, what we mean by ‘pain’ is not the presence of a cluster of responses, but rather the presence of an event or condition that normally causes those responses.[v]

This was a big problem for behaviourism, and to all intents and purposes it is the problem that functionalism can most clearly be seen to resolve. Instead of being a purely input/output theory like psychological behaviourism, functionalism tries ‘To illustrate the causal relationships between brain processes, mental processes, and externally observable behaviour.’[vi] It does not deny the existence of mental states, or even to equate such mental states with specific brain processes, as the identity theory did. Functionalism has ‘multiple realisability’, meaning that it is not restricted to a certain type of physical thing having beliefs or desires or the like. This means that it is not dependant on our biology, making it compatible with both dualist and physicalist models. Because of this it also appears to bypass the problem with mind-body interaction, as it doesn’t specify the types of thing that are interacting. It considers mental states in terms of their function rather than their make-up.

Functionalism would appear to benefit from being able to refer to behaviour, and making behavioural propositions, yet still allowing mental states to be the cause of behaviour. It also allows for at least some predictions of behaviour based on this causal role:

Our view is that the concept of pain, or indeed of any other experience or mental state, is the concept of a state that occupies a certain causal role, a state with certain typical causes and effects.[vii]

Mental states would seem to receive inputs for the environment, creating new mental states, or causing an interaction of mental states, which would then lead to a behavioural output. The theory itself developed with the advent of computers, which showed that manipulation was needed as an intermediary step between inputs and outputs.

John Searle raised concern with such claims that computers model cognition ‘…Specifically the claim that the appropriately programmed computer literally has cognitive states and that the programs thereby explain human cognition.’[viii] He felt that such a strong comparison would cause difficulty in accepting the existence of mental states. A similar argument was given by Ned Block, who conceived of the whole of China acting in the same way as a human as described by the functionalist view. This would ‘function’ in the same way, if only for a short period of time, and would have inputs and outputs. Functionalism could be seen to grant this system mental states, as it has not limited itself to the substance making up the thought process:

What makes the Homunculi-headed system just described a prima facie counterexample to (machine) functionalism is that there is prima facie doubt as to whether it has any mental states at all – especially whether it has what philosophers have variously called ‘qualitative states’, ‘raw feels’, or ‘immediate phenomenological qualities’.[ix]

This can be seen as one of the main problems with functionalism, in that it is so ‘wildly liberal,’[x] that it will allow for almost anything to be considered a mind, if it functions in a similar way, however such a thing may lack qualitative states. This was called by Putnam ‘Absent Qualia’.

One argument was made by David Lewis. He considered two cases, the pain of a madman and the pain of a Martian. The madman has different causes of pain, and different behaviour, which results from it. The Martian has similar causes of pain, and similar behaviour resulting from it, but he does not have a physical brain in the same sense as we do due to having a completely alien biology. David Lewis argued that a theory of mind is necessary that would allow for both the madman and the Martian to be described as being in pain. ‘A simple identity theory straightforwardly solves the problem of mad pain. It goes just as straightforwardly wrong about Martian pain. A simple behaviourism or functionalism goes the other way: right about the Martian, wrong about the madman.’[xi] He suggested that there needed to be a slight alteration to the functionalist conception in order for it to allow for both to be considered to be in pain. He said that we should consider the state to be ‘apt for having certain causes and effects.’[xii] This version therefore would allow for both, considering the madman to be an exception to the general rule of pain behaviour, but still in pain

The problem with ‘absent Qualia’ still hangs over the functionalist theory, but David Lewis again has a reply. He says that ‘a theory of what it is for a state to be pain is inescapably a theory of what it is like to be in that state, of how that state feels, of the phenomenal character of that state,’[xiii] and he considers this to be also true of the Functionalist position. Although this is not a definitive argument against ‘Absent Qualia’, it suggests that qualitative states, at least in his version of functionalism, should be taken into account. It is also possible to think that although functionalism cannot account for qualia that may not be important, as it could be considered that qualia are so inconsequential in psychological explanation that this is no great problem. Even if functionalism does leave qualia out, it does not leave out sensations like pain, pleasure, and so on, since these are defined not in terms of their qualitative character but in terms of their functional role.

There are some similarities between Functionalism and Behaviourism. The difference is that Functionalism takes mental states more seriously than Behaviourism. Functionalism believes that looking at inputs and outputs is not enough; in fact it is essential to also include the relations between different states. Functionalism's main difference from (philosophical) behaviourism is that it thinks the same thing is true of mental states: they cannot be fully defined in terms of behaviour, the definition of a mental state must also make reference to other mental states. To illustrate this it appears easiest to use the example of pain.

Hilary Putnam wrote: ‘Imagine a community of ‘super-Spartans’ or ‘super-stoics’ – a community in which the adults have the ability to successfully suppress all involuntary pain behaviour.’[xiv] It would be a community in which there is no behavioural reaction to pain – in some sense similar to the ‘mad pain’ example above – and therefore no external way of telling that they are in pain.

The behaviourist theory would state that they are not in pain, as they are not exhibiting any pain behaviour. This seems to go against our conception of pain; we would want to say that they are in pain, even if they don’t show it. Functionalism in this instance would allow that they are still experiencing pain. What is common to all pains in the Behaviourist model is that people in pain tend to act alike. Functionalism, however, says that the common factor which all pains share is their functional role, more specifically their relation to inputs, outputs, and other functionally defined mental states.

Functionalism then, would appear to solve many of the problems with behaviourism. It allows for mental states to be the cause of behaviour, whatever the biological or non-biological make-up. We do not have a problem in attributing states, such as pain, to a person whose behaviour differs, or is even non-apparent. It must be held in mind however, that the problem with functionalism being a bit too vague is still true, leaving it slightly shaky. In addition, the problems that Functionalism encounters with qualia must also be considered. However, it can still be considered to be a superior theory than behaviourism, as mental states cannot be fully defined in terms of behaviour; the definition of a mental state must also make reference to other mental states.

- - - - - - - - - - - - -

[i] Jaegwon Kim, “Philosophy of Mind”, Westview Press 1998, p30-1
[ii] Andy Clark, “Microcognition”, MIT Press 1993, p22
[iii] John Heil, “Philosophy of Mind”, Routledge Press 1998, p65
[iv] Donald Davidson, “Essays on Actions and Events”, Oxford University Press 1980, p217
[v] Hilary Putnam, Brains and Behaviour, “Readings in Philosophy of Psychology Volume One”, Ed. N Block, Harvard University Press 1980, p27
[vi] John Searle, Breaking the Hold, “The Nature of Consciousness”, Ed N. Block, MIT Press 1998, p494
[vii] David Lewis, Mad Pain and Martian Pain, “Readings in Philosophy of Psychology Volume One”, Ed. N Block, Harvard University Press 1980, p218
[viii] John Searle, Minds, Brains, and Programs, “The Mind’s I”, Eds D Hofstadter and D Dennett, Penguin Books 1982, p353-354
[ix] Ned Bock, Troubles with Functionalism, “Readings in Philosophy of Psychology Volume One”, Ed. N Block, Harvard University Press 1980, p278
[x] Ned Bock, Troubles with Functionalism, “Readings in Philosophy of Psychology Volume One”, Ed. N Block, Harvard University Press 1980, p294
[xi] David Lewis, Mad Pain and Martian Pain, “Readings in Philosophy of Psychology Volume One”, Ed. N Block, Harvard University Press 1980, p217
[xii] David Lewis, Mad Pain and Martian Pain, “Readings in Philosophy of Psychology Volume One”, Ed. N Block, Harvard University Press 1980, p219
[xiii] David Lewis, Mad Pain and Martian Pain, “Readings in Philosophy of Psychology Volume One”, Ed. N Block, Harvard University Press 1980, p222
[xiv] Hilary Putnam, Brains and Behaviour, “Readings in Philosophy of Psychology Volume One”, Ed. N Block, Harvard University Press 1980, p29

TrackBack URL for this entry: http://bluejoh.com/cgi-bin/mt/mt-tb.cgi/407