| HOME | ABOUT ME | EXAMPLES OF WORK | FEES AND SERVICES | LINKS | CONTACT ME | FEEDBACK |
The
type identity theory of physicalism is the idea, as outlined by David Lewis
(Journal of Philosophy, 63, pp 17-25), that “every experience is identical
with some physical state.” Experiences and states in this sense are to be
taken as universals, and not as particulars, which is where the type identity
theory differs from the token identity theory.
In
this essay I will look at the type identity theory as outlined by Lewis, and
also the problem of whether this theory can accommodate the idea that mental
states are variably realisable in animals of different physical natures. This
problem is talked about in the article “The Nature of Mental States” by
Hilary Putnam (1979).
The
main argument that Lewis gives is that the definitive property of any experience
is its causal role. Lewis believes that these causal roles, which belong to
experiences, actually belong to physical states. And since these states have the
properties of the experiences within them, they must be the experiences
themselves. Lewis tells us that he is not going so far as to say that physical
states are the objects that the experiences are of, but just the experiences
themselves. It seems slightly odd that he is willing to say that experiences are
in fact physical states, but not to say that the objects of experiences are not
definitely physical. In my opinion, if the type identity theory is true, then
surely the only way the experiences we have could be physical states is if the
objects that are the subject of our experience must also be physical. However,
this does not damage the theory as a whole; because Lewis does not deny the
physicality of the objects, he just does not give a definite answer either way.
Moving
on from this, I will now look at Lewis’ argument in more detail, beginning
with his first premise that “the definitive characteristic of any experience
as such is its causal role” (Lewis, 1966, p19). This causal role is
expressible by a set of conditions that specify its typical causes and effects
under certain circumstances. These conditions are true of the experience, but
are also not part of the experience, as stated by analytic necessity. In other
words, the first premise of Lewis’ argument is that all experiences are
defined by their causal roles.
This
principle, as Lewis says, is not in itself a materialist principle. It is merely
saying that experiences are in some way real and that they are efficacious
outside their own realm. It does not say anything about the actual nature of the
reality of the experience, or the nature of its efficacy.
The
second premise that Lewis talks about is the idea that in today’s world, we
have a set of scientific theories that provide a true account of all physical
phenomena. This seems like a plausible idea, although whether we have this body
of theories at our disposal right now remains to be seen. It seems unlikely that
we can truly explain everything that is describable in physical terms. However,
Lewis also believes, and most other philosophers would agree, that present
physics is not complete. It does seem at least possible though that our future
physics will be able to tell us everything we need to know about the world, and
in particular, causation. The only problem with this is that it seems that it
might always be the case that the present physics of any time is not complete,
and that future physics cannot be used because it has not been discovered yet.
However, the idea does allow for the existence of non-physical phenomena, but
does not allow these non-physical phenomena to be efficacious with respect to
the physical phenomena in any way. This is basically saying that although there
maybe are non-physical things, they are not needed to explain any of the
physical things within the world. This means that the non-physical phenomena
cannot, according to the identity theory, be experiences of any sort. This is
because they are inefficacious with respect to physical phenomena. This is a
problem because, as physicalism says, all experiences have behavioural
manifestations as their effects that involve physical phenomena. From this it
follows that experiences are some sort of physical phenomena.
This
argument does seem appealing in many ways, because it seems hard to imagine any
physical phenomena other than mental states that could occupy the causal roles,
or so Lewis says. This does seem like a reasonable point to me. Furthermore, it
also seems that non-physical phenomena are ruled out as well if we hold such a
high confidence in physical phenomena. Lewis believes the identity theory is
right, not just from the arguments he makes, but from the fact that no one has
seemingly come up anything else to occupy the causal roles.
However,
this has been attempted by Putnam (1979). Putnam proposes an identity theory of
sorts as well, and is trying to show how the identity theory of physicalism
cannot accommodate for the fact that mental states are variably realisable in
creatures with different physical natures. For the rest of this essay I will
look at whether this is the case.
Putnam,
instead of identifying mental states with physical states, identifies mental
states with functional states of each individual system or organism. For this
reason the theory is often known as functionalism. Functionalism is not a fully
type-type identity theory, for it does not make a relation between mental states
and physical states, even if it does reduce the mental states to functional ones
instead. It is a species of type-type identity theory, but one that identifies
the mental with 2nd order functional properties, which are in turn
realised by the 1st order mental properties that Lewis describes.
There can be a different physical state instantiated each time the same
functional state occurs.
So,
why if this is also an idea that the mental states we have are physical in some
way, does it pose a problem for the type identity theory of physicalism? Well,
the reason is that in the type identity theory, we have to say that if you have
a mind, you must have a brain. This discounts the possibility that we can have
brains made of other things, such as silicon, or alien brains. It also leaves us
with a problem of how we describe something like the pain of an animal. If we
say that the mental state of pain is identical with a specific type of neural
state, then we are suggesting that not only does the animal have a brain exactly
like ours, but also they are in exactly the same neurological state as we are in
when we are in pain. This seems a very strange thing to say, as it does not seem
plausible that animals such as insects and birds and molluscs have the same sort
of brain as we do, or that they have anywhere near the same neurological states
as we do. However, Putnam’s functionalism can solve this problem, because it
allows animals to be in different physical states, even if they are in the same
functional state of, for example, pain. The trouble with the identity theory is
that not only for pain, but also for every mental state, the physical state of
that mental state is the same in every animal. This seems highly unlikely, and
if just one mental state can be proven to have different physical states, the
identity theory will collapse. The idea that the pain is a functional state of
the whole system rather than a particular brain state seems far more intuitive,
because we view the mental states of animals by their behaviour, which points
more to a similarity in functional states than a similarity of brain states.
However,
Putnam’s theory has some difficulties too. These difficulties are discussed in
the paper “Mad Pain and Martian Pain” (David Lewis, 1980). Lewis tries to
combine both the identity theory and the functionalist theory to account for all
types of pain, both ‘mad’ and ‘Martian’. Lewis believes that neither of
the two theories can accommodate both of these types of pain. Mad pain, namely
pain whose causes and effects diverge from those of the pain causal role, cannot
be accommodated by functionalism. This is because, for functionalism, pain is
simply to be in a state that occupies the pain role. It also seems that the
identity theory cannot accommodate Martian pain. Martian pain is pain whose
physical realisation is different to ours, for example their pain is foot cavity
inflation rather than c-fibre firing. This is not compatible with the identity
theory because, according to the identity theory, pain is whatever state
occupies the pain role for us. It seems that the identity theory is too
chauvinistic, in that it does not allow animals that clearly have mental states
to in fact truly have mental states. However, Putnam’s theory seems far too
liberal, in that it will allow virtually anything that can be in a state of some
functional kind to have a mental state, even something like a tin can!
Lewis
tries to solve these difficulties by combining a functionalist account of pain
within a given population, with an identity theory account of pain for
individual members of the given population. According to Lewis, a state S is
pain for a population if and only if, with some exceptions, whenever a member of
P is in S, it being in S has the sorts of causes and effects given by the pain
role (Lewis, 1980, p219).
This
would make c-fibre activation pain for the human population and foot cavity
inflation pain for the Martian population. An individual X from a given
population will only have pain if the state they are in is the appropriate pain
state for their population.
This
seems like a very appealing solution to the problems encountered so far because
it seems to have a way to show how each individual of a different physical
nature experiences pain. However, complications in the theory arise when trying
to find out the appropriate population (P) for a given individual (X). Lewis
gives four criteria to try and accomplish this:
(1)P
is the human population.
(2)X
is a member of P.
(3)P
is a population in which X is unexceptional.
(4)P
is a natural kind.
Lewis
gives equal weight to these criteria
and, so, the population that satisfies the greatest number of criteria for a
given individual is the appropriate population for that individual. An example
is as follows:
“If
X is our Martian, we are inclined to say that he is in pain when the cavities in
his feet are inflated; and so says the theory, provided that criterion (1) is
outweighed by the other three, so that the appropriate population is taken to be
the species of Martians to which X belongs.” (Lewis, 1980, p219).
If
Lewis uses this method to characterize the criteria, it seems there are some
problems. Take the example above. The Martian population does satisfy criteria
(2), (3), and (4) for a Martian for whom foot-cavity inflation occupies the
pain-role. After all, it is a population which is a natural kind and of which
our ordinary Martian is an a normal member. But the Human population does not
merely satisfy criterion (1), but criterion (4) as well. And so while Lewis’s
theory does imply that the appropriate population for our ordinary Martian is
the Martian population, it has to be because (2), (3), and (4) outweigh (1) and
(4) together, and not simply (1) by itself as Lewis suggests. But how can (4)
outweigh itself? There seems to be a slight inconsistency in the method used by
Lewis.
However,
the real problems occur when looking at the case of a mad Martian. The example
is as follows:
“If
X is a mad Martian, I would be inclined to say that he is in pain when the
cavities in his feet are inflated; and so says our theory provided that criteria
(2) and (4) together outweigh either (1) or (3) by itself.” (Lewis, 1980,
p220).
It
does seem that the Martian population does satisfy criteria (2) and (4) for a
Martian for whom foot-cavity inflation fails to occupy the pain-role. But it
also seems the human population does not merely satisfy criterion (1) for the
exceptional Martian, but criteria (1) and (4). And the Mad Martian population
does not merely satisfy criterion (3) in this case, but criteria (2) and (3).
This means that Lewis does not really show that the appropriate population for
the mad Martian is the Martian population as opposed to the Human or mad Martian
population. Lewis may respond to this by saying that it does not matter, and
that you can just say you are unsure whether the Mad Martian is in pain or not.
But to me, it just seems obvious that he is, and the ambiguity is not in whether
he is in pain, but in determining what population he belongs in to determine his
pain.
It
seems each of the three populations has an equal weighting to be the appropriate
population in the case of the mad Martian. So it seems Lewis cannot accommodate
the problem of mad Martian pain.
The
root of the problem is that criteria (1) and (3) have been given the same weight
as criteria (2) and (4). It seems less weight should be given to (1) and (3).
This is because, for (1), the human population is a natural kind, and so
automatically satisfies two criteria for any individual. Furthermore, for (3),
any physical realization of a causal role will have a population of individuals
who are similar in terms of the realization of the causal role, but whose
members could seemingly have very different physical natures. But such a mixed
population will again satisfy two criteria for any individual.
There
are two possible ways I can see of dealing with this problem. One is to just get
rid of (1) and (3). This idea, however, has problems as well. For example, if
humans were exceptional in terms of the physical realization of pain within a
larger a larger genus, which is also a natural kind, then getting rid of (1) and
(3) would have the effect of giving the human population and the genus the same
weighting to be the population for any individual human.
Perhaps
another way to sort out the problems would be to give less weighting to (1) and
(3). This would allow for the correct results in terms of mad Martian pain and
other such problematic examples. However, there are other problems with this
solution. It seems there is a contradiction, in that it seems there are reasons
for both (1) to outweigh (3) and (3) to outweigh (1). Unless (1) outweighs (3),
there will be no grounds for taking the appropriate population for Lewis’
madman to be the human population as opposed to a more inclusive genus in which
C fibre activation is the occupant of the same causal role it occupies for the
madman. But second, unless (3) outweighs (1), there will be no grounds for
taking the appropriate population for, for example, an artificially intelligent
robot to be the population of similarly designed robots (no natural kind) as
opposed to the human population.
It
seems to me that all of the theories I have mentioned have serious problems. The
identity theory is flawed because it cannot accommodate the realisability of
mental states within creatures of different physical natures. The attempt to
solve this through Putnam’s functionalist model also fails because it cannot
accommodate mental states such as mad pain, where the pain is not caused by the
normal pain causal role. Finally, Lewis’ attempt to combine both of these
theories also seems doomed to failure. In conclusion, it seems that none of
these theories can accommodate all types of mental state within all types of
creature.
Lewis,
David. 1966. ‘An Argument for the Identity Theory’, Journal of Philosophy
63: 17-25.
Lewis,
David.1980. ‘Mad Pain and Martian Pain’, Readings in the Philosophy of
Psychology, Vol.1, N. Block, ed., Harvard University Press, pp216-222.
Lewis,
David. ‘Review of Putnam’, Readings in the Philosophy of Psychology,
Vol.1, N. Block, ed., Harvard University Press, pp232-233.
Putnam,
H. 1979. ‘The Nature of Mental States’, in Mind Language and Reality.
Philosophical Papers, Vol.2. Cambridge: Cambridge University Press.
©
David Friesen 2004