Functionalism and multiple realizability

In philosophy of mind, the multiple realizability argument is one of the most popular arguments against type identity theory and in favour of functionalism. However, functionalism faces a multiple realizability problem very similar to the one raised against type identity theory. In this video I outline the functional multiple realizability problem and then briefly discuss four possible responses.

[@Ansatz66]

There are two parts to the meaning of any word: the meaning in the mind of the person who uses the word, and the meaning in the mind of the person who hears or reads the word. We try to get these two parts to resemble each other as closely as we can, but in the case of things like colors and pains, we have no way to do so. When I use the word pain it refers to my experience of pain, and when you hear the word pain it refers to your experience of pain, and we can only hope that these two experiences are the same and are also shared by most other people. It is fundamentally unknowable whether a masochist feels pain, because we can never truly see the world from the perspective of another person and witness their subjective experience. In the case of a masochist, her behavior gives us more reason than usual to wonder about the reliability of generalizing our experience of pain to other people, but that's the most we can fairly say. Therefore we cannot know whether a functionalist would be wrong to bite the bullet and deny that a masochist feels pain, and in the same way we can't know whether we should deny that an octopus feels pain. It seems fair to suppose that our subjective experiences don't exist in some totally inaccessible magical soul. If we can achieve the technical sophistication required to examine and understand the brain is sufficient detail, we might one day gain the power to fully understand and document the processes that underlie our subjective experiences, and we may even be able to answer whether someone else feels the pain that we do as a purely technical question. Until then, the debate between functionalism and identity theory has no real hope of leading anywhere. We just don't have the facts necessary to reach a conclusion.

[@Stefan-ut8ff]

Because of your instructions, I finished my philosophy notes, which I missed that class. thx!

[@Brian.001]

I think your second response, at 12:12, begins to focus in on the real problem, which is that there is no compelling evidence for MR in brain states / experiential states. Different creatures, with completely different brains, can all exhibit the same functional state that we are associating with 'Pain', but that doesn't show that each of these creatures has experiences of a single type - 'Pain'. As long as the experience is unpleasant enough, it will occasion avoidant behaviour (for example). That doesn't indicate that different creatures experience the same pain type. Think of the experience of colours: John might grow up with inverted spectra - experiencing red as the rest of us experience green, but his functional role when seeing red will still be the same as ours. He might then experience green as we experience red. Functional profile is incapable of pinning down a unique experience-type. Hence, functionalism is just false. If we stick with the original identity theory, and assume that identity implies that a single type of brain state is identical with a single type of experience, the whole problem goes away.

[@doovstoover9703]

I don't really understand the problem with this definition, that pain is domain-specific. What we're talking about is physical sensation, and the point on the scale at which that sensation crosses over into the realm of discomfort. We understand that tolerance for pain varies from person to person, and can be interpreted differently according to context (ie the pain of lifting a heavy weight in the gym; a slap in a fight vs a slap in the bedroom...). What then is the problem with defining pain functionally as that point at which physical sensation becomes unpleasant to the entity experiencing it?

[@Brian.001]

@@doovstoover9703 Well, for me, the problem is that such a functional analysis of pain fails to say anything about the subjective experience of pain.

[@doovstoover9703]

@@Brian.001 isn't it the opposite? The functionalist view of pain, or happiness or jealousy or whatever mental process, is that it is entirely down to the subject's interpretation of their own experience. My experience of a slap or a colour or a loud noise might be totally different to yours, but we interpret our individual experiences to be the same through context, language etc. If a slap to you felt like a tickle to me but my brain still interpreted the tickle as painful, then there is functionally no difference. The part that isn't accounted for is the objective experience, but that's the whole point.

[@Brian.001]

@@doovstoover9703 It depends on what you think functionalism is trying to achieve. The mind-body identity theory was claiming that our subjective mental states (e.g. Pain) are identical with particular brain states (e.g. c-fibre stimulation). Functionalism is motivated by the 'intuition' that this identity statement cannot be true; a pain state can be realised by a variety of brain states (e.g. from one species to another). If this were true, we would then be free to identify pain states with a functional state, rather than a brain state This is token-identity, rather than type identity, with brain states. The problem is that this move to a functional conception of mental states tells us nothing at all about the physical nature of mental states. We are told that a variety of pain states can realise pain, but the question of what pain is, in physical terms, has been abandoned. The move to functional definitions is just a semantic move; it says nothing about the physical nature of mental states. Going back to the first step, multiple instantiation of a pain state, we had no evidence to support that claim in the first place.

[@doovstoover9703]

@Brian well yes, but that's what functionalism *is*. It refutes the importance of specific physical states in the definition of what makes something a mind, and instead places the emphasis on output. Thus octopus pain or alien pain can have entirely different physical mechanisms to human pain but still be defined as pain. Denying multiple realisability only really makes sense if you insist on limiting the definition of a given mental state to the way in which humans specifically experience it - in which case questions like 'does an octopus experience pain', 'can a computer think' cease to make sense.

[@childintime6453]

hey, can you make a video about causal theory of mind? at first I thought it was same as functionalism, but like you’ve said Lewis (who with Armstrong developed so called causal theory) kinda rejects or criticizes functionalism as flawed theory. So what’s the causal theory about? if you can’t make a vid please just respond with a short answer

[@eclecticism1019]

Hello dear Kane, Is there is a possibility for us to have a copy of your PowerPoint slides, please? Kind regards

[@fukpoeslaw3613]

Octopodees? Not octopods? Pronunciation I mean.

[@KaneB]

I've never heard anyone use the term "octopods". "Octopuses" and "octopi" are the most common pluralizations, but "octopodes" is more fun to say. It's actually one of my favourite words. So I say "octopodes".

[@fukpoeslaw3613]

@@KaneB so you stress the last syllable, ok.