�
behaviourism, functionalism and mental states
Greg Detre
Wednesday, 25 October, 2000
Mr Snowdon, Philosophy of Mind II
Explain and
discuss behaviourism and functionalism and teleo-functionalism. Analyse the
arguments which were used to support them. Are any satisfactory accounts of (a)
experiences (b) propositional attitudes?
Notes � Snowdon, Philosophy of Mind II
Notes �
Churchland, Matter and Consciousness (pg 37)�
Notes �
Putnam, �Brains and behaviour� in Mind, Language and Reality
Notes �
Galen Strawson, Mental Reality, Chapter 9
Notes -
Shoemaker, �Functionalism and qualia�
Dictionary of
Philosophy of Mind � functionalism (1) and (2)
functionalism, like Lewis� Identity theory, characterises the mental in terms of its causal role
NO, in terms of �relational properties�???
it too, then, is a physicalist theory, insofar as most people leave no room in the physical world for non-physical causal effects???
functionalism broadens the ramifications (of physicalism???) though, by arguing that the same arrangement (in processing, i.e. functional and causal, terms) could be instantiated in any (physical, we usually assume) medium
functionalism differs from behaviourism, although:
they both look at behaviour in terms of: output as a function of input
but functionalism allows for there to be a series of hidden, intermediary (mental/functional) states which cannot be eliminated from the causal chain
is the difference between functionalism vs Identity theory that:
Identity theory identifies mental processes with (certain??? type or token???) physical processes, with that identity failing to hold for an alien
whereas functionalism is like a type Identity theory in that it has greater explanatory power for that identity, and moreover, allows that it could hold for a similarly functioning alien/robot
the alien�s functional (cognitive) economy could be physiologically systematically different from ours, but retain a parallel set of mutual relations between mental states, e.g. a pain state that plays an identical functional role (relating e.g. body damage, discomfort and the deisre for the pain to stop, saying ouch and rubbing the wound)
= a functional isomorph (identical structure of internal activities being sustained by the matter)
the problem that this poses for Identity theory is that there is no single physical type for the intertheoretic taxonomy to relate to a given mental type
but can�t we intertheoretically reduce between a given mental state and the functional relations which underlie it???
I spose that�s what functionalism is � ???
I spose when Identity theorists talk about types of physical processes, they�re talking about particular physical laws operating on particular formations of matter, whereas functionalism is instantiation-independent
aahhh, functionists are Identity theorists � token Identity theorists, perhaps???
- they simply reject the usual type Identity theory
thus, psychology = methodologically autonomous from the natural sciences
comparing computational models with human behaviour is an attempt to recreate (and so learn about) the underlying functional organisation implemented in the brain
though functionalism (unlike behaviourism) allows for incorporating mental states� relations with each other in its explanation of behaviour, it still sees mental states like cogs (with teeth of complex, perhaps non-deterministic??? (mathematical???) functions), gearing our senses to our motor signals
- what about qualia?
inverted spectrum � red could play the same functional role, despite being a different sensation to me, as green to another person� and in a functionalist account, we would never know that our two mental states which functionalism would identify as both being �red�, are phenomenologically/qualitatively different, i.e. functionalism would have missed the whole point of the mental (just like all materialist/causal theories)
i.e. functional isomorphs need not be phenomenological isomorphs
e.g. an alien may share the functionally identical state of pain with us, but pain may feel quite different for the alien
if such a spectrum inversion is conceivable to us, but inexplicable in functionalist terms then functionalism can�t be viable/adequate
similarly, if the people of China were to interact like neuronal activity, complex functional states might arise, but the organism cannot be conceived as having qualia
= the �absence of qualia� objection = a stronger form of the objection that functionalism cannot account for (or incorporate???) qualia into its depiction of the mental
is Churchland then saying that we can try to rescue functionalism by stipulating a token identity relation between a particular functional state and the qualia that arise???
so: physical processes can be seen in terms of their functional organisation. these functional states are mental states. qualia arise from some contingent-seeming but actually systematically related way from the physical processes, and will vary from person to person and certainly according to the form of physical instantiation.
but are functional mental states and my qualitative mental states the same???
perhaps qualia arise from some non-functional aspect of physical laws, in which case a functional state will not necessarily correspond to a qualitative mental state
or, they could be more closely related, like in a token identity theory � a functional mental state does feel like something, but you can�t predict what unless you know everything about how the quale arose from the physical instantiation as well as the functional organisation of that instantiation � so you could relate them via the physical, but if all you could see is the qualia and the functional state, it would appear to be a contingent relationship
is this supervenience??? is this what
Davidson is talking about with his funny relationship between mental and
physical where the neural and linguistic descriptions can�t be reconciled
thus, �redness� plays a functional role, but its associated qualia is a contingent epiphenomenon
in fact, these qualia are consistent with each other for any given individual, i.e. though redness need not feel that way (for you or for me), it always will for that person
is this to serve as a means of identification of a mental state � phenomenology as a label (for introspective discrimination of mental states)
but why is it???
thus, functionalists would be admitting the ineliminable reality of qualia. but then qualia need to be fitted in with a materialist/physicalist world view (though the job is easier if qualia are causally inert � can you do that??? can you see qualia as being somehow independent and detachable from the functional state they happen to represent???)
identify the qualia with physical properties of the physiological instantiation, e.g. sensation-of-red corresponds to a particular pattern of distributed firing of APs
thus, functional isomorphs with different physical instantiations may (or indeed may not, depending on exactly what pattern or principle in our physiology is giving rise to the qualia) experience different qualia for the same functional mental state
thus (and this may be taking a step backwards towards carbon chauvinism) it might be a particular property of our brain that gives rise to qualia, which is lacking from our robot being controlled by 109 Chinamen, so a functional isomorph may have different, or indeed, wholly absent, qualia
again, this leaves the hard problem unsolved of where qualia come from??? what, then, is the difference between functionalism (which allows for, but makes no explanation of, qualia � as causally inert token identities with functional states) and logical behaviourism? as far as I'm concerned, such a functionalism is explaining mental states simply as hidden steps in a behavoiurist chain that we speculate to bridge the gap between input and output currently filled by folk psychology (another imposed predictive/explanatory construct of assumptions � like gravity, if we take it as a given that all bodies attract each other, it all makes sense�) � thus functional state are not really mental states at all, and what we really regard as mental states are pushed back to contingent, causally intert epiphenomena.
thus, two physiological isomorphs will obviously be functional isomorphs, and if qualia are related systematically to the physical instantiation, then they�ll have the same qualia, right???
no, any functionally isomorphic system will have qualia, even the 109 Chinamen robot???
�those features at the objective focus of the system�s discriminatory mechanisms, those are its sensory qualia�??? � pg 40/41
so it should be possible to �see�, or deductively imagine what sensory qualia will result, with enough knowledge about a brain (either alien or human)
this last form of functionalism blurs the distinction with reductive materialism
next objection:
temperature = supposedly reduced to mean kinetic energy of constituent molecules�
this may work for a gas, but the temperature of a solid/plasma/vacuum are quite different
i.e. the physical property of temperature has �multiple instantiations�, like psychological properties
but �temperature� is a wholly abstract artificial concept, unlike mental states, which (are really the only things in the world that) we have evidence for
yet thermodynamics is not considered an �autonomous science� separate from the other natural sciences, as psychology is proposed to be. why not/what�s the difference?
because reductions are domain-specific:
temperature-in-a-gas = definition �
temperature-in-a-vacuum = definition �
so, similarly:
joy-in-a-human = hippocampal whatsits etc �
joy-in-a-Martian = �
so there are some type/type reductions of mental � physical states, though narrower than first expected
and so functionalist claims of the autonomy of psychology are false, and functionalism is not so different from Identity theory after all???
arg against Identity theory � neural activity is not �a cause� of a mental state. maybe another reason token Identity theory is preferable is because one can talk of neural activity in the brain as a whole, rather than areas/processes within it, as you would in order to be able to abstract to �types�
super-spartans and X-worlders. pain, and whether it exists when it�s suppressed, or indeed not behaviourally manifested in any way.
�Behaviourism is dead. No one still believes that mental
concepts can be satisfactorily analysed just in terms of behaviour and
dispositions to behaviour.
Neobehaviourism survives � the view that mental life is linked to behaviour in such a way that reference to behaviour enters essentially and centrally into any adequate account of the nature of almost all, if not all, mental states and occurrences.� - Preface
He�s trying to attack neobehaviourism�s approach to the mental by conceiving of beings with rich mental lives but exhibit no behaviour. He is extending the term �behaviour� beyond its usual sense of action (both willed and volitionless) to include mental behaviour, e.g. mental arithmetic. Mental arithmetic, if we never announce or write down our conclusions, is an example of an unobservable activity. Moreover, it�s a sort of mental action, it has intention(???). The only evidence of performing mental arithmetic is in neural activity, which I don�t think he�s classing under �behaviour�. So he�s got these mental agents, and he gives various backgrounds and explanations of how non-active mental beings could be or become so (including the Rooting Story, of how they used to be active agents who slow down in old age, until they lose all capacity for action as well as any memories of their frenetic heyday). He gives these Weather Watchers desires and hopes as well as sensory stimulation, so they aren�t like Buddhists (observing without mental comment???).
He doesn�t agree with the first, and is now trying to attack the second of Armstrong�s two premises:
It becomes crucial then that although the Weather Watchers desire, believe and hope, they don�t intend to act or seek to influence (it may or may not even occur to them that the capacity for action exists), i.e. he believes that it is not contradictory to say that they can desire that something be a certain way, without wishing that they could step in to make it so. He defends this at length, including a set of scenarios from a human who becomes paralysed but doesn�t realise, to a human who really cannot move (through both neurophysiological and psychological reasons) and shows that the antecedent of the action cannot include anything which alters the agent�s capability for action (otherwise wet clay could be said to have a disposition to be fragile since it would be if it was put in a kiln). He argues that the Weather Watchers may or may not possess the concept of action at all � to inactive creatures, the desire for X to happen need not be indisseverable from the desire to do something to make X happen. Indeed, even humans watching the weather, or more convincingly, the World Chess Championships, might wish for an outcome without wanting to intervene to fulfil that desire. Thus, it is not constitutive of desire that it include a disposition to action, even though on Earth the two are almost always conjoined � to induce from that that they must be is to subscribe to the tenuous doctrine of logical naturalism (if non-magical evolution can�t have made it that way, it can�t be that way � this is relevant to the Weather Watchers in general).
He defends the Weather Watchers from the claims that they could not have any sort of linguistic thought without means to vocalise it, that one can desire without desiring to make it so, and that desire cannot be separated from affect. This is intended to complete his refutation of neobehaviourism (started in the previous chapter, regarding sensations) that desires, beliefs and propositional attitudes (i.e. mental states) can be held in the absence of action or behaviour.
He adds a footnote reference to Searle, and his formulation of how desires and action are linked, which seems to conflict with Strawson�s separation of the two, but Strawson seems happy with it.
In the conclusion of the book, he describes himself as a sort of naturalist Cartesian. I take this to mean that though he would place his bets on a materialist explanation (in the broad, laws-of-physics sense of �materialism� which extends beyond just matter), of mind, though he expects our notions of the physical world to change. I think Strawson thinks similarly to how Prof Tas described Nagel�s position, i.e. that our understanding of matter will extend and alter so radically that matter will come to encompass mind in such a way that the mind-body distinction will collapse or be shown/seen to be conceptually defunct.
Strawson�s non-standard use of �behaviour� could be seen in terms of �physical� and �agential� behaviour. Physical behaviour broadly includes any change in an agent�s body while agential behaviour is restricted to behaviour controlled by the agent�s volition. There can also be physical and agential behavioural dispositions. It is necessary to clarify how inclusive these definitions should be (should they include the firing of neurons, or mental arithmetic, for instance).
believes that Block and Fodor have comprehensively refuted FSIT � wants to pick up on their point of the inadequacy of its explanation of qualia, and apply it to functionalism in general
3 types of functionalism:
decompositional functionalism � the whole system is explained in terms of the functions of its parts
computation-representation functionalism � �computer-as-mind� analogy
metaphysical functionalism � mental states simply are functional states
Notes � Block, �Troubles for functionalism�
Notes - Searle, �The Rediscovery of the Mind�, Chapter 3
Notes � Papineau, �Reality and Representation�, Chapters 3 and 4
functionalism(1) - The view that the physical realization of a functional component is not, in some sense, its essence. Rather, what makes a functional component the type it is, is characterized in terms of its role in relating inputs to outputs and its relations to other functional components.
functionalism (2) - An explanatory approach to behavior and the constitution of cognitive states that regards particular behaviors and cognitive structures and capacities as playing functional roles in particular domains or contexts.
functionalism
/"fV<ng>(k)<longs>(<schwa>)n(<schwa>)lIz(<schwa>)m/ n.E20. [f. prec. + -ISM.]
1 Social Sciences. The study or interpretation of phenomena in terms of the functions which they fulfil (esp. within an overall system). E20.
2 Consideration for the function and purpose of a building etc. as regulating its design.
functionalist n. & a.
(a) n. an adherent or student of functionalism;
(b) adj. exhibiting or pertaining to functionalism
Does functionalism exist as a concept outside philosophy of mind?
Do other people talk about 2 types of functionalism? Ooops, who does?
What are teleo-functionalism and causal functionalism?
Why is behaviourism lumped under the same essay?
Because functionalism is a form of behaviourism.
������ or vice versa???
What are the hard/harder problems of consciousness?
Can we talk about the same
�functional organisation� across instantiations with wholly different
representations?
Is neobehaviourism = philosophical behaviourism?
Is there a parallel between what Strawson is trying to do with the behaviourally inert Weather Watchers and Putnam�s super-spartans and X-worlders who are able to suppress all behaviour effected by pain? i.e. attacking behaviourism by chopping off one end of the input-output loop to demonstrate that the mental must exist independently and as more than just an intermediary to cause and effect.
Why is it important for Strawson to include stuff like �mental arithmetic� in his non-standard extended definition of �behaviour�? Is it because mental arithmetic has no outward effect, other than neural activity?
In fact, does neural activity come under his definition of �behaviour�.
Could we not say that the Weather Watchers definitely have some sort of mental life if their processing capacity evidently exceeds their behaviour, i.e. if they have a nervous system at all. No, Strawson rejects that � why???
Do they have qualia?
They certainly seem to have qualia (I'm not sure if he says it explicitly, but it�s very implicit in his idea of mental life). Yes, they definitely have qualia.
He says that their thought is probably akin to our linguistic mental representation, yet language is a social, communicative act � where would it come from?
In fact, I'm not convinced that mentality such as the Weather Watchers� could ever emerge (or exist) at all (at least in biologically plausible terms) (since you need embodiment and action as a kind of training period for the neural network).
I suppose their brains need not be neural networks like ours, or they could be pre-programmed, or indeed the Weather Watchers could be like alien Alzheimer�s patients (which he considers as a possibility in order to satisfy die-hard dispositional thesis holders).
What is the �fink� objection that he dismisses?
What is radical Epictetianism?
What is �functional state identity theory�?
functionalism � it is the information processing properties of our neural mechanism, rather its biological basis, which matters in giving rise to mind. something about mental states?
Behaviourism is the precursor to functionalism. Functionalism is a kind of more adequate formulation of behaviourism, where the function operating on input to produce output can be more than one step. However, they both see the mental simply in terms of functions that we need to posit in order to understand how our complex, adaptable, seemingly free and unpredictable actions can arise � we have to ascribe desires and beliefs to agents, because their behaviour cannot be understand in terms of first-order stimulus response. Thus, functionalism is akin to a higher-order behaviourism, where:
stimulus � ( functional state � (functional state � ( � ) ) ) response
to talk of mental states in the same way that one talks of understanding �meaning� is a behaviourist mistake
in Putnam�s example, how could they know the cultural rule of not showing pain without either being berated once for doing so, or having already evolved this acculturation so that spartanism is innate, in which case how can they have a word for it??? there is no escape from this without giving the evidence that a logical behaviourist needs to ascribe them pain. hmmm, maybe