University of Miami Thinking About Consciousness Essay

Like in the first paper, you must identify the argument accurately from the texts I provide and report it clearly. This paper adds one element to the first essay, you must come up with an objection to the new and then evaluate the debate. . This can be your own idea or taken from one of our other readings.

What  paper must contain: Paragraph One: the thesis. Identify the view you are discussing and provide a quote where your chosen author states the core claim of the view. Next two to four paragraphs: Present the author’s argument. Over the next two to four paragraphs, present the author’s argument for the answers you stated in the first paragraph. You must support and explain their argument using quotes from the text I provided. Please do not use any other sources or texts other than those I provide. Do not cite other versions or editions of these texts, either. You may also cite my handouts and lecture, but these cannot replace the text. Objection: Spend one paragraph presenting what you take to be the strongest single objection to the argument you just presented in the previous paragraphs. You can come up with your own objection, or take one from our readings. You do not need to quote in this paragraph, but do cite your sources. Evaluation. Finally, evaluate the argument and the objection. Does the objection defeat the argument? Or does the original view have the resources to resist or respond to the objection? Give reasons why you think the objection succeeds or fails.

The Copyright Law of athe United States (17 U.S.C. 101
et.seq.as amended) limits the use of this material to
Instructional Use.
Author of Article of Chapter: David Papineau
Title of Journal or Book: Thinking About Consciousness
Title of Article of Chapter: The Case for Materialism
Publisher: Oxford
Volume and Issue:
Place of Publication: New York
Date of Publication: 2002
Course: Phil 103
Subject: PHIL103-The Case for Materialism-Papineau
Instructor: Alison McIntyre
Page Numbers: 13-46
Page 1
Chapter
I
THE CASE FOR
MATERIALISM
r. r Introduction
Books on consciousness often begin by distinguishing between
different kinds of consciousness. We are told about self­
co11,sdousness and sentience, creature consciousness and state
consciousness, phenomenal consciousness and access conscious­
ness, perceptual consciousness, higher-order consciousness, and
so on. I’d rather leave all this until later. Some of these
distinctions will become significant in due course, and will be
explained when they are needed. Others will not matter to my
discussion.
For the moment, all I want to say is that I am concerned with that
aspect of consciousness that makes it so philosophically interesting.
Namely, that having a conscious experience is like something, in
Thomas Nagel’s striking phrase (r974). It has become standard
to use ‘phenomenal’ or 1subjective’ to focus on this feature of
consciousness, and I shall adopt these usages in what follows.
The idea is best introduced by examples rather than definitions. (‘If
you gotta ask, you’re never gonna know.’) Compare the difference
between having your eyes shut and having them open, or between
having your teeth drilled with and without an anaesthetic. When
your eyes are open, you have a conscious visual experience, and
13
THE CASE FOR MATERIALISM
THE CASE FOR MATERIALISM
when your teeth are drilled without an anaesthetic, you have a
worth taking some care about this, for there are a number of different
conscious pain. It is like something for you to have these experiences.
It is not like that when you close your eyes, or when the anaesthetic
defences of materialism on offer in the contemporary literature, and
takes effect. What you lose in these latter cases are elements of
phenomenal or subjective consciousness. 1 From now on, when I say
‘conscious’, I shall mean this kind of consciousness.
Much of what follows will be concerned with a particular
philosophical puzzle about consciousness: namely, the puzzle of
not all of them are equaliy compelling. However, I think that there is
one definitive argument for materialism. I shall call this ‘the causal
argument’, and the burden of this first chapter will be to develop this
argument and distinguish· it from some less effective defences of
materialism.
The puzzle can be posed simply. On the one hand, there is a strong
There is a further reason for laying out the argument for
materialism carefully. Many contemporary philosophers harbour
grave suspicions about materialism. Thus some philosophers
contend that the whole idea of materialism is somehow empty, on
the grounds that there is no proper way of characterizing the
‘physical’ realm. (Crane and Mellor 1990, Crane r99r, Segal
2000). And others suggest that contemporary materialism about
argument for adopting a materialist view of conscious states, for
the conscious mind rests on nothing but fashion or prejudice,
supposing that conscious states must be part of the physical world,
unsupported by serious argument (Burge 1993, Clark 1996).
I intend to show that these attitudes are mistaken. The question of
how to define ‘physical’ in the context of the mind-brain debate
how consciousness relates to the physical world. There are other
philosophical puzzles about consciousness, but this seems to me the
most immediate. We will be ill placed to understand anything about
consciousness if we cannot understand its relation to the physical
realm.
that they must be identical to brain states, or something similar. Yet,
on the other hand, there are also strong arguments (and even
stronger intuitions) which suggest that conscious states must be
distinct from any material states.
I believe that in the end the materialist argument wins. Conscious
states are material states. This is not to belittle the anti-materialist
arguments and intuitions. They are deep and important. We will not
does raise a number of interesting points, but there is no great
grasp consciousness properly unless we understand how to answer
difficulty about pinning down a sense precise enough for the
purposes at hand. It will prove easier do this, however, after we
have rehearsed the argument for materialism. Accordingly, I shall
ne>t worry about the meaning of ‘physical’ at this stage, but simply
begin by outlining the case for materialism. Once we have seen what
them. Still, I think that careful analysis will show that they are
is at issue, it will become clearer how materialists can best understand
flawed, and that the right solution is to embrace materialism.
the meaning of ‘physical’, and I shall return to this issue at the end of
the chapter.
There is one terminological point which I do need to address at this
point, however. When I do fix a meaning for ‘physical’ at the end of
the chapter, I shall read this term in a relatively strict sense, as
standing roughly for the kinds of first-order properties studied by the
physical sciences. Under the heading of ‘materialism’, on the other
hand, I shall include not only the doctrine that conscious states are
identical with physical states in this strict sense, but also the doctrine
that they are identical with ‘physically realized functional states’, or
with some other kind of physically realized but not strictly physical
states (these possibilities will be explained further in section r.6
I shall begin by putting the materialist argument on the table. It is
1
Some philosophers assume that ‘phenomenal’ is meant to contrast with
intentional, and on this basis hold that much recent discussion of consciousness,
especially that surrounding David Chalmers’s ‘hard problem’ (1996), is
invalidated by an implicit supposition that subjectivity is independent of
intentionality (cf. Eilan 1998). It is perhaps worth emphasizing that I don’t
intend ‘phenomenal’ to imply ‘non-intentional’. I simply mean it as a non­
committal term for subjective ‘what-it’s-likeness’. Nothing yet rules out the
possibility that all, or only, intentional states involve phenomenal consciousness.
Moreover, since Chalmers also understands ‘phenomenal’ in this way, his ‘hard
problem’ of phenomenal consciousness will still arise even if phenomenality is not
independent of intentionality.
I4
IS
THE CASE FOR MATERIALISM
THE CASE FOR MATERIALISM
below). It is true that the causal argument can be read as
supporting the stricter identification with physical states, and
indeed this is how I shall first present it in the next section. But, as
the history of the causal argument, and in particular the question of
why it has become persuasive only recently.
we shall see, the causal argument can also be construed as
supporting the less strict identification of conscious states with
functional or other physically realized states. Since both the strict
r.2 The Causal Argument
and the less strict identifications tie conscious states constitutively
Let me now outline what I take to be the canonical argument for
materialism. Setting to one side all complications, which can be
to the physical world, few of the arguments in this book will
require me to decide between them. So it will be useful to have a
term which covers both options, and I have adopted ‘materialism’
for this purpose. Correspondingly, a ‘material’ state will mean
either a physical state in the strict sense or some functional ·or
other physically realized state.
In addition to suspicions about the meaning of ‘physical’, there is
the further allegation mentioned above, that contemporary materi­
alism is nothing but a modish fad. I take the causal argument to be
outlined in this chapter to rebut this allegation. The causal argument
may not be conclusive, but it certainly shows that the case for
materialism goes beyond mere fashion or prejudice.
Some may think that the charge of modishness is supported by
historical considerations. Widespread philosophical materialism is a
relatively recent phenomenon, largely a creature of the late
twentieth century. This recent provenance may seem to support
the accusation that contemporary materialism owes its popularity
more to fashion than to any serious argument. ‘If the case is so
substantial’, anti-materialists can ask, how come it took so long for
1
philosophers to appreciate it?’ I take this to be a good historical
question. But I think there is also a good historical answer: namely,
that a key premiss in the argument for materialism rests on empirical
evidence that only became clear-cut during the course of the
twentieth century.
However, I shall not complicate the analysis of this chapter by
overlaying it with historical commentary. The issues are complicated
enough without the added burden of tracing historical strands.
Accordingly, this chapter will focus on the structure of the argument
for materialism, not its history. For those who are interested in the
historical dimension, the Appendix at the end of this book discusses
r6
discussed later, it can be put as follows.
Many effects that we attribute to conscious causes have full physical causes.
But it would be absurd to suppose that these effects are caused twice over.
So the conscious causes must ·be identical to some part of those physical
cau;ses.
To appreciate the force of this argument, consider some bodily
behaviour which we would standardly attribute to conscious causes.
For example, I walk to the fridge to get a beer, because I consciously
feel thirsty. Now combine this example with the thought that,
according to modern physical science, such bodily movements are
fully caused by prior physical processes in brains and nerves. The
obvious conclusion is that the conscious thirst must be identical with
some part of those physical processes.
Let me now lay out the above argument more formally. This will
help us to appreciate both its strengths and its weaknesses.
As a first premiss, take:
(r) Conscious mental occurrences have physical effects.
As I said, the most obvious examples are cases where our conscious
feelings and other mental states cause our behaviour.
Now add in this premiss (‘the completeness of physics’ henceforth):
(2)
All physical effects are fully caused by purely physical prior
histories. 2
2
What about quantum indeterminacy? A stricter version of (2) would say that
the chances of physical effects are always fully fixed by their prior physical histories,
and would reformulate the rest of the argument accordingly (with ( r) then as
‘Conscious mental occurrences affect the chances of physical effects’, and so on). I
shall skip this complication in most of what follows.
17
THE CASE FOR MATERIALISM
In particular, this covers the behavioural effects of conscious causes
to which our attention is drawn by premiss r. The thought behind
premiss 2 is that such physical behaviour will always be fully caused
by physical contractions in your muscles, in tum caused by electrical
messages travelling down your nerves, themselves due to physical
activity in your motor cortex, in tum caused by physical activity in
your sensory cortex, and so on.
At first sight, premisses 1 and 2 seem to suggest that a certain range
of physical effects (physical behaviour) will have two distinct causes:
one involving a conscious state (your thirst, say), and the other
consisting of purely physical states (neuronal firings, say).
Now, some events are indeed overdetermined in this way, like the
death of a man who is simultaneously shot and struck by lightning.
But this seems the wrong model for mental causation. After all,
overdeterminatibn implies that even if one cause had been absent, the
result would still have occurred because of the other cause (the man
would still have died even if he hadn’t been shot, or, alternatively,
even if he hadn’t been struck by lightning). But it seems wrong to say
that I would still have walked to the fridge even if I hadn’t felt
thirsty (because my neurons were firing), or, alternatively, that I
would still have gone to the fridge even if my neurons hadn’t been
firing (because I felt thirsty). So let us add the further premiss:
(3) The physical effects of conscious causes aren’t always overdetermined by distinct causes.
Materialism now follows. Premisses 1 and 2 tell us that certain effects
have a conscious cause and a physical cause. Premiss 3 tells us that
they don’t have two distinct causes. The only possibility left is that
the conscious occurrences mentioned in (1) must be identical with
some part of the physical causes mentioned in (2). This respects both
(1) and (2), yet avoids the implication of overdeterrnination, since
(1) and (2) no longer imply distinct causes.
r. 3 The Ontology ofCauses
The causal argument focuses on the way in .which conscious
occurrences operate as causes. It says that conscious causes must
18
THE CASE FOR MATERIALISM
be identical to physical causes. However, there are different
philosophical theories of causation, and in particular about the
kinds of things that can feature as causes. On one view, causes are
facts, or instantiations of properties. Candidate causes on this view
would be my being in pain, or my having active nociceptive-specific
neurons. On an opposed view, causes are basic particulars, or events,
abstracted from any conscious or physical properties they might
have. The causal argument as stated above will generate different
conclusions, depending on which view of causation you adopt. In
particular, it will generate a stronger conclusion on the former view,
that causes are facts, than on the latter view, which has causes as
basic particulars. Still, this will be of no great moment, since a
rephrasing of the argument will still allow us to generate the stronger
conclusion, even on the assumption that causes are basic particulars.
Let me take this a bit more slowly. I myself favour the view that
causes are facts (cf. Mellor 1995). A restricted variant of this view,
which will perhaps be more familar to some readers, is that causes are
instantiations of properties by particulars, or ‘Kim-events’ (cf. Kim
19 7 3). In what follows, I shall standardly use the term ‘state’ to refer
to this kind of item-that is, to the possession of a property by some
particular. Now, on the view that causes are facts (Kim-events, states)
the causal argument given above implies that conscious properties
(being thirsty, say) must be identical with physical properties (having
a certain brain feature). For, in requiring that conscious causes be
identical with physical causes, the argument will now require that
conscious facts (Kim-events, states)-such as that I am thirsty, say­
are identical to certain physical facts (Kim-events, states)-! have a
certain brain feature, say-and these two facts (Kim-events, states)
cannot be identical unless the properties they involve-being thirsty,
having that brain feature-are themselves identical.
The alternative view of causation is that causes are basic particulars
(cf. Davidson r 980). Then the causal argument, as phrased above,
won’t itself carry you to the identity of conscious and physical
properties, since the identity of conscious Davidson-events with
physical ones requires only the far weaker conclusion that the
relevant conscious and physical properties are instantiated in the
same particular, not that the properties themselves are identical.
19
THE CASE FOR MATERIALISM
THE CASE FOR MATERIALISM
Still, as I said, we can rephrase the argument so as to iegenerate the
stronger conclusion. Let us take premiss r’ to be the claim that all
conscious events cause some. physical events in virtue of their
conscious properties; premiss 2 1 says that all physical events are
caused by prior physical events in virtue of the latter’s physical
properties; and premiss 3′ says that the physical effects of conscious
causes aren’t always caused twice over, in virtue of t’\-yo different
properties of the prior circumstances. In order to make these
consistent, we then need once more to identify the conscious
properties of the causes with their physical properties.
The causal argument as presented in the last section thus argues for
the identification of conscious properties with physical properties.
It is worth noting at this stage, however, that this argument for
property identity proceeds on an abstract, existential level, and is not
concerned with any detailed identifications. It tells us that each
conscious property must be identical with some physical property,
but it doesn’t tell us which specific physical property any given
conscious property may be identical with. 3
To establish any such specific property identity, more detailed
empirical information is needed. It is not enough to know that
conscious causes can always be identified with some part of the
full physical histories behind their effects. To pin down specific
property identities, we need more detailed evidence about correla­
tions between specific conscious properties and the different parts of
those physical histories. We need to know that pain, say, or
thirst, or seeing an elephant, are found when such-and-such
brain areas are active, but not when others are. In Chapter 7 I
shall consider this kind of detailed research, and the kinds of
results it can be expected to bring. But for the moment, I shall
concentrate on the more abstract· existential claim that every
conscious property must be identical with some, as-yet-to be­
identified physical property.
0
3
Similarly, on the alternative construal of the causal argument to be developed
in section I. 7, we will have an abst!act a~gument for the identity of conscious
properties with material properties (even if not with strictly physical properties),
but again this argument on its own win not tell us which material properties any
specific conscious property should be identified with.
20
We can usefully think of this abstract claim and the detailed
correlational research as complementing each other. The abstract
claim doesn’t by itself tell us which physical property a given
conscious property should be paired up with. And the correlational
research, while promising to establish specific pairings, can’t by itself
establish that the paired properties are identical, as opposed to
regularly accompanying each other. The abstract claim is important,
then, since it is needed to license the move from detailed empirical
correlations to property identifications. It tells the empirical
researchers that conscious properties aren’t just correlated with the
physical properties they are regularly found with, but must be
identical with them.
r. 4 Epiphenomena/ism and Pre-established Harmony
All this assumes, however, that the abstract claim does follow from
the causal argument. Let us now examine this argument more
closely.
As laid out above, the causal argument seems valid. 4 So, to deny
the conclusion, we need to deny one of the premisses. All of them
can be denied without contradiction. Indeed, all of them have been
denied by contemporary philosophers, as we shall see. At the same
time, they are all highly plausible, and their denials have various
unattractive consequences.
Let me start with premiss r. This claims that, as a matter of
empirical fact, particular conscious states have particular physical
effects. This certainly seems plausible. Doesn’t my conscious thirst
cause me to walk to the fridge? Or, again, when I have a conscious
headache, doesn’t this cause me to ingest an aspirin?
Still, the possibility of denying this premiss is familiar
enough, under the guise of ‘epiphenomenalism’ or ‘pre-established
harmony’.
The first philosopher to embrace this option was Leibniz. Unlike
4 However Sturgeon (1998) argues that the argument trades on an equivocation
between the everyday sense of ‘physical’ (in premiss I) and a quantum-theoretical
. sense (in premiss 2). I shall comment on Sturgeon’s claim in section I.IO below.
2I
THE CASE FOR MATERIALISM
THE CASE FOR MATERIALISM
most other philosophers prior to the twentieth century, Leibniz was
committed to the causal completeness of physics (see Appendix).
But he was not prepared to accept the identity of mind with brain. So
he opted for a denial of our premiss
I,
and concluded that mind and
matter cannot really influence each other, and that the appearance of
interaction must be due to pre-established hannony. By this Leibniz
meant that God must have arranged things to make sure that mind
and matter always keep in step. In reality, they do not interact, but
are like two trains running on separate tracks. But God fixed their
starting times and speeds so as to ensure they would always run
smoothly alongside each other.
Some contemporary philosophers (for example, Jackson r982)
follow Leibniz in avoiding mind-brain identity by denying
premiss I. But they prefer a rather simpler way of keeping mind
and matter in step. They allow causal influences ‘upwards’ from
brain to mind, while denying any ‘downwards’ causation from mind
to brain. This position is known as epiphenomena/ism. It respects the
causal completeness of physics, in that nothing non-physical
causally influences the physical brain. But it avoids the theological
complications of Leibniz’s pre-established harmony, by allowing the
brain itself to cause conscious effects.
Epiphenomenalism is not a particularly attractive position. For a
start, it would require us to deny many apparently obvious truths,
such as that my conscious thirst caused me to fetch a beer, or that my
conscious headache caused me to swallow an aspirin. According to
epiphenomenalism, my behaviour in both these cases is caused
solely at the physical level. These physical causes may be accompan­
ied by conscious thirst or a conscious headache, but these conscious
states no more cause resulting behaviour than falling barometers
cause rain. 5
5
Chalmers (1996: esp. I 34-6), following Russell (I 92 7) and Lockwood(r989),
argues that there is a way for dualism to avoid this epiphenomenalist ipefficacy
while respecting the completeness of physics. This is to identify phenomenal
properties with the intrinsic properties of the physical realm. Chalmers’s idea is that
physical science picks out properties like mass and charge only extrinsically, via
their relations to observable features of the world. So maybe phenomenal
properties can be identified with the intrinsic nature of such properties, suggests
Chalmers, and thereby have their causal efficacy restored. This seems an entirely
22
That epiphenomenalism has these odd consequences is not in
itself decisive. The theoretical truth can often overturn claims which
were previously regarded as the merest common sense. Moreover,
there is nothlng incoherent about epiphenomenalism. As I shall
have occasion to stress in what follows, there is nothing conceptually
contradictory in the idea of conscious states which exert no causal
powers themselves. Still, epiphenomenalism is surely an empirically
implausible position, by comparison with the materialist view that
conscious states are simply identical to brain states. ·
If epiphenomenalism were true, then the relation between mind
and brain would be like nothing else in nature. After all, science
recognizes no other examples of ‘causal danglers’, ontologically
independent states with causes but no effects. So, given the choice
between epiphenomenalism and materialism, standard principles of
scientific theory choice would seem to favour materialism. If both
views can accommodate the empirical data equally well, then
ordinary scientific methodology will advise us to adopt the simple
view that unifies mind and brain, rather than the ontologically more
profligate story which has the conscious states dangling impotently
from the brain states.
There remains the possibility that the anti-materialist arguments
to be examined later will show that conscious mind and brain cannot
be identical. If this is so, then one of the premisses of the causal
argument must be false. And in that case premiss r seems as likely a
candidate as any. Certainly most contemporary philosophers who
are persuaded by the anti-materialist arguments have opted for
epiphenomenalism and the denial of premiss r, rather than for any
other way out of the causal argument.
sensible view to me. But, pace Chalmers, I would say that it is simply a version of
materialism. My reaction is that the intrinsic features of the physical world with
which Chalmers wants to identify phenomenal properties are themselves simply
basic physical properties. Thus I am happy to agree with Chalmers that scientific
theory picks out these intrinsic physical properties only via descriptions which
refer to observable features of the world. Moreover, I agree that conscious
properties should be identified with arrangements of such intrinsic physical
properties, and thus that it is like something to have these arrangements of
intrinsic properties. Indeed, I find it hard to see what a sensible materialism could
amount to, except this combination of views. So, from my point of view,
Chalmers’s suggested position is simply the optimal formulation of materialism.
23
r
THE CASE FOR MATERIALISM
THE CASE FOR MATERIALISM
But this does not invalidate the criticisms I have levelled against
execution is consciously countermanded. Given this, it seems that the
epiphenomenalism. My concern at the moment is not to prejudge
the anti-materialist case, but merely to assess the causal argument.
· And the point remains that, in the absence of further considerations,
conscious decision is part ofthe cause of the finger movement after all.
The initial cortical activity does not determine the finger movement
on its own, but only puts the motor cortex in a state of ‘readiness’,
which leads to action in just those cases where the conscious decision
is added. This then allows us to reason, as before, via the causal
argument, that .conscious decisions could not play a part in so
influencing physical movements, were they not themselves physical.
In any case, even if conscious decisions did not contribute causally
to the actions normally attributed to them, it would not follow that
they had no physical effects of any kind. For instance, they will still
presumably be causes of the sounds I make, or the marks I put on
paper, when I later report my earlier conscious decisions. So they will
still satisfy premiss r, which requires only that conscious causes have
some physical effects, and not that they have all the physical effects
with which they are normally credited by common sense. So once
more the causal argument will run.
The other worry concerned the possibility of conscious states with
‘broad’ representational contents. The possession of such ‘broad
contents’ hinges on matters outside subjects’ heads. For example,
Hilary Putnam suggests that the representational state thinking about
water hinges on what natural kind is actually water in your
environment, and Tyler Burge argues that thinking about arthritis
hinges on facts about other members of your community (Putnam
197 5, Burge r979, r982).
Now the worry, in the present context, is that-if any conscious
states are representational in this broad way, then this will not sit
happily with premiss r’s claims about causal efficacy. For how can
states which hinge on matters outside your head exert a causal
influence on your bodily movements? Surely your bodily move­
ments are causally influenced solely by matters inside your skin, not
by how matters are outside you.
The possibility of broadly representational conscious states raises
any number of tricky issues, not all of which I can pursue here
(though see section 7. 7 below). However, they seem to me to pose no
real threat to .the causal argument for materialism. Let me content
myself with two comments.
it seems clearly preferable to identify mind with brain than to
condemn conscious states to the status of causal danglers. It may be
that further anti-materialist considerations will yet require us to
reconsider this verdict, but so far we have seen no reason to deny
premiss r, and good reason to uphold it.
Before leaving the issue of epiphenomenalism, it may be worth
addressing some more local worries about premiss r. Even if the
blanket epiphenomenalist refusal to credit any conscious states with
physical effects is methodologically unattractive, there may be some
more specific reasons for doubting whether particular sorts of
conscious states have the physical effects they are normally credited
with. In particular, I am thinking here of conscious decisions, and
doubts about their causal efficacy arising from the experimental
results associated with Benjamin Libet, and of conscious states which
are representational, and doubts about their causal efficacy arising
from the possibility that they may have ‘broad contents’. Let me deal
with these in turn.
In a series of well-known experiments, Libet asked subjects to
decide spontaneously to move their fingers, and simultaneously to
note the precise moment of their decision, as measured by a large
stop-watch on the wall. Libel also used scalp electrodes to detect the
onset of motor cortical activity initiating the finger movement.
Amazingly, he found that this neural activity started a full ¼to ½
second before the subjects were aware of making any conscious
decision (Libel r993).
At first sight, this certainly suggests that such conscious decisions
are epiphenomena! with respect to the actions we normally attribute
to them: since the conscious decisions come later, it looks as if they
must be effects of, rather than identical with, the brain processes that
give rise to the action. But in fact this interpretation is not clear-cut.
Libel himself points out that the conscious decisions still have the
power to ‘endorse’ or ‘cancel’, so to speak, the processes initiated
by the earlier cortical activity: no action will result if the action’s
24
25
THE CASE FOR MATERIALISM
First, I am open to the possibility that some, indeed all, conscious
states may be essentially representational (cf. n. I above); moreover,
it seems plausible that representation in general is a broad matter.
Even so, it would seem odd to allow that conscious properties in
particular, as opposed to representational properties in general, can
depend on broad matters outside the skin. Could two people really be
internally physically identical, yet nevertheless feel different, because
things are different outside them? (Cf. Introduction, n. 2.) Given
this, the natural strategy for those who seek to equate some (or all)
conscious properties with representational properties is to shear off
some species of narrow representation from the general run of broad
representational properties, and to equate representational con­
scious properties with these narrow representational properties. And
then, to return to the matter at hand, there will cease to be any
reason to doubt that these conscious properties have physical effects
such as bodily movements, however it may be with representational
properties in general.
Second, even if you do wish to insist that some conscious
properties are indeed broadly representational (a possibility to
which I shall return in section 7. 7), it will not follow that such
broad conscious properties do not cause any physical effects. For they
may have physical effects outside my body. For example, my
consciously thirsting for water might affect which liquid I put into a
glass, and my consciously worrying about arthritis might affect where
the doctor will poke me when I complain of it. If this is right, then the
causal argument will run as before, and imply that any such broad
conscious properties must also be identical with physical properties,
if their instantiations are to have such physical effects-though these
physical properties will now presumably stretch outside bodies, as
well as inside.
1.
s
Accepting Overdetermination
There remain the two other premisses to the causal argument. It
will be convenient to relegate the discussion of premiss 2, the
completeness of physics, to the last section of this chapter and the
26
THE CASE FOR MATERIALISM
Appendix. So let me now briefly consider premiss 3, the one ruling
out overdetermination.
To reject this premiss is to accept that the physical effects of mental
causes are always overdetermined by distinct causes. This is some­
times called the ‘belt and braces’ view (make doubly sure you get
the effects you want), and is defended by D. H. Mellor (1995:
ro3-5).
At first sight, this position seems to have the odd consequence that
you would still have gone to the fridge for a beer even if you hadn’t
been thirsty (because your cortical neurons would still have been
firing), and that you would still have gone to the fridge even if your
cortex hadn’t been firing (because you would still have been thirsty).
These counterfactual implications seem clearly mistaken.
However, defenders of the belt and braces view maintain that
such implications can be avoided. They argue that the distinct
mental and physical causes may themselves be strongly counter­
factually dependent (that is, they hold that, if you hadn’t been
thirsty, your sensory neurons wouldn’t have fired either, and vice
versa).
Still, this then raises the question of why such causes should always
be so counterfactually dependent, if they are ontologically distinct. 6
Why wouldn’t my neurons have fired, even in the absence of my
conscious thirst? Similarly, why shouldn’t I still have been thirsty,
even if my neurons hadn’t fired? Now, it is not impossible to imagine
mechanisms which would ensure such counterfactual dependence
between distinct causes. Perhaps the conscious thirst occurs first, and
then invariably causes the cortical activity, with both causes thus
available to overdetermine the behaviour. Alternatively, the cortical
activity could invariably cause the thirst. Or, again, the conscious
decision and the cortical activity might be joint effects of some prior
common physical cause. But such mechanisms, though conceptually
coherent, seem highly implausible, especially given that they need to
6
Note that this is only a problem if the causes are genuinely ontologically
distinct, and not if they are merely related as role state and physical realizer. As we
shall see in the next section, the existence of ‘two’ causes in this latter sense does
not threaten overdetermination, precisely because of their ontological inter­
dependence. So I have no objection to versions of the belt and _braces view which
intend only parallel causes in this weak sense. Cf. Segal and Sober 1991.
27
THE CASE FOR MATERIALISM
ensure that the conscious state and the brain state always accompany
each other.
The relevant point is analogous to one made in the last section. We
don’t find any ‘belt and braces’ mechanisms elsewhere in nature­
that is, mechanisms which ensure that certain classes .of effects
invariably have two distinct causes, each of which would suffice by
itself. As with the epiphenomenalist model, a belt and braces model
requiring such peculiar brain mechanisms would seem to be ruled
out by general principles of scientific theory choke. If the simple
picture of mental causation offered by materialism accommodates
the empirical data as well as the complex mechanisms required by
the belt and braces option, then normal methodological principles
would seem to weigh heavily against the belt and braces view.
As with the corresponding argument for epiphenomenalism, this
appeal to principles of scientific theory choice is defeasible. Perhaps
in the_ end the anti-materialist arguments will force us to accept
mind-brain distinctness. In that case, the belt and braces view might
be worth another look. True, it is even more Heath-Robinsonish than
epiphenomenalism. On the other hand, it does at least have the
virtue of retaining the common-sense view that conscious states
characteristically cause behaviour. In any case, my present purpose is
not to decide this issue finally, but only to point out that, as things
stand so far, we have good reason to uphold premiss 3, and none to
deny it.
r.6 Functionalism and Epiphobia
Many contemporary philosophers will feel that the causal argument
as elaborated so far is rather too strong. This argument has claimed
that conscious properties are identical to physical properties. But the
majority of contemporary materialists would probably prefer to
identify mental properties in general, and conscious properties in
particular, with physically realized functional properties, or properties
which supervene on physical properties, or perhaps properties which
are disjunctions of physical properties, rather than with strictly
physical properties themselves.
28
THE CASE FOR MATERIALISM
Let me start with functional properties. I shall come back to the
other possibilities in a moment. A functional property is a higher­
order property-of-having-some-property-which-satisfies-condition­
R, where R specifies some requirement on an instantiation of a
first-order property. In line with this, ‘functionalism’ in the
philosophy of mind is the view that any given mental property
should be identified with some property-of-having-a-first-order­
property-which-bears-certain-causal-relationships-to perceptual
ii1puts, behavioural outputs, and other mental states. For example,
the property of being in pain might be identified, at first pass, with
the property-of-having-some-property-which-arises-from-bodily­
damage-and-gives-rise-to-a-desire-to-avoid-the-source-of-that­
damage.
The advantage of this functionalist account of mental states is that
it allows beings who have quite different instrinsic physical proper­
ties nevertheless to share mental properties. For example, it seems
plausible that octopuses, whose neurology is physically quite
different from human neurology, can nevertheless share the
property of being in pain with humans. But, if this is so, the property
of being in pain cannot be identical with any physical property, for
no suitable physical property will be common to humans and
octopuses. On the other hand, both humans and octopuses will
share the higher-order property-of-having-some-property-which­
arises-from-bodily-damage-and-gives-rise-to-a-desire-to-avoid-the­
source-of-that-damage. The physical properties which play this role
will be different in the two cases, but the higher-order property itself
will be common to octopuses and humans.
Now, how does functionalism stand with respect to the causal
argument? If we take the causal argument at face value, then they
seem inconsistent. For, as we have seen, the causal argument
promises to establish that conscious properties are identical with
strictly physical properties, which is just the claim that functionalism
is designed to avoid.
This tension with the causal argument puts functionalism under
some pressure. If functionalism is inconsistent with the causal
argument, it must deny one of its premisses. And, on reflection, it
could well be held to deny premiss r, the one that says that conscious
29
THE CASE FOR MATERIALISM
causes have physical effects. For, if conscious properties are not
identical with physical properties, but rather with certain higher­
order properties, then conscious causes will not be identical with the
physical causes which premisses 2 and 3 tell us are the only causes of
behavioural effects. So it would seem to follow that conscious states
don’t cause behavioural effects after all. This line of thought is
sometimes said to generate ‘epiphobia’, a condition in which
functionalists are overcome with anxiety about how their view
differs from epiphenomenalism.
For this reason, and perhaps others, some philosophers have
recently become uneasy about ‘higher-order’ properties. They object
that it is profligate to posit substantial new properties for every way
of characterizing objects as possessors of some (first-order) property
which R (cf. Kim r998: ch. 4).
I have some sympathy with this point of view. However, it is
important to realize that, even if we reject higher-order properties on
these grounds, the underlying dilemma highlighted by functional­
ism remains. For we will still need to .decide whether conscious
properties should be identified (a) with those strictly physical
properties whose instantiations are paradigm physical causes, yet
are not shared by humans and octopuses, or (b) with other first-order
properties of a kind which can be shared by humans and octopuses,
but are in danger of being outcompeted as serious causes.
Suppose, to illustrate the point, that we admit no properties except
genuinely first-order properties. But suppose that we also continue to
feel the pull of the thought that both humans and octopuses can be
in pain. Given the physical differences between humans and
octopuses, we might seek to respect this thought by construing
pain as a disjunctive condition, requiring P1 or P2 or … where the
various P; s are the different strictly physical properties which are
causally active when different beings are in pain. But now epiphobia
returns to trouble us once more. For my human arm movement is
presumably caused by my human P1 (my nociceptive-specific
neurons firing, say). But P1 itself isn’t identical with the disjunction
P1 or P2 or . .. -that-is, with pain. So, if P1 causes my movement, the
disjunction presumably doesn’t, and thus it seems to follow once
more that the property of being in pain is inefficacious. The dilemma
30
THE CASE FOR MATERIALISM
remains: if you want to have different creatures sharing pain, then
you seem to end up rendering pains causally inefficacious.
A similar point can be made about views which replace higher­
order functional properties, not by disjunctions of physical proper­
ties, but by properties which ‘supervene’ on physical properties. (For
readers unfamiliar with this notion, it .is explained in section r.8.)
This alternative will again leave us with the choice between identi­
fying conscious properties with (a) physical properties themselves, or
(b) with the properties which supervene on physical properties. And
again we will face the dilemma that only (b) seems to allow
physically different beings to share conscious properties, but only
(a) seems to allow conscious properties to be causally efficacious.
In the next section I shall consider whether this dilemma can be
resolved. However, it would be tiresome to have to address the issue
separately for all the different ways in which conscious properties
can be identified with properties which are not strictly physical­
that is, for functional higher-order properties or disjunctions of
physical properties or supervenient properties. So let me adopt the
general term ‘higher’ property to cover all these alternatives.
Correspondingly, when I speak of a ‘higher’ property being ‘realized’
by a physical property, I shall mean either that a functional higher­
order property is instantiated because some physical property is, or
that a disjunction of physical properties is instantiated because one
of its disjuncts is, or that a supervenient property is instantiated
because some physical property which determines it is. 7
7
It might seem that for completeness I should also consider the possibility that
conscious properties can be identified with macro-properties that are composed of
physical micro-properties. However, the mereological notion of a macro-whole
being composed of micro-parts seems to me orthogonal to the notion of one kind
of property being realized by another kind of property. I would say that
composition can occur within the strictly physical, and accordingly that a
macro-property composed of strictly physical micro-properties is itself a strictly
physical property. So identifying conscious properties with properties composed
of strictly physical properties is itself to identify them with strictly physical
properties, not with some species of ‘higher’ property realized by physical
properties. There are interesting questions about the compositional relation
betw_een micro-parts and macro-wholes-in particular, do the parts have some
kind of causal primacy over wholes?-but they are independent of the arguments
in this book. (For convincing arguments against the causal primacy of micro-parts,
see Hiittemann, forthcoming.)
31
r

THE .CASE FOR MATERIALISM
I. 7
A Possible Cure for Epiphobia
Perhaps there is a cure for epiphobia. We don’t have to agree that the
only respectable kind of causation involves strictly physical causes
having physical effects. For it is arguable that there is a perfectly
normal sense of ’cause’ in which higher states cause the effects that
their realizers cause. On this account, even if pain is a higher
property, differently realized in octopuses and humans, my taking
an aspirin can still be caused by the pain in my head, in virtue of
being caused by whichever strictly physical state realizes that pain
in me.
If we adopt this generous notion of causation, functionalism
becomes consistent with premiss r of the causal argument after all.
The fact that mental states are not identical with strictly physical
states does not mean that they cannot cause the behaviour which is
caused by those strictly physical states. In the generous sense of
’cause’, they will do so as long as they are higher states which are
realized by those strictly physical states.
Indeed, if we look at things in this way, we in effect have another
version of the causal argument, one which reads cause’ generously
throughout, and which ends up with the conclusion that conscious
properties, if not strictly physical properties, must at least be
physically realized higher properties. The argument now runs:
1
(r*) Conscious causes have physical effects, at least in the generous
sense.
(2)
All physical effects are fully caused by purely physical prior
histories.
(3)
The physical effects of conscious causes aren’t overdetermined
by distinct causes.
And the conclusion is now that:
(4*) Conscious causes must at least be higher states which are
realized by the physical causes of their physical effects.
For otherwise (3) would be violated, with the physical effects of
conscious causes being caused twice over, first by their conscious
32
THE CASE FOR MATERIALISM
causes as in ( r’), and second by the distinct physical causes
guaranteed by (2).
Note here how we do in a sense end up with two causes of the
relevant behavioural effects. For we now have both (a) the higher
state with which we are now identifying the conscious state and {b)
the realizing physical state which directly causes the behavioural
result. (Cf. Segal and Sober r99r.)
But the important point is that these two ’causes’ are not now
ontologically distinct, and so do not genuinely overdetermine any
resulting behaviour. The higher cause is present only in virtue of the
physical cause which realizes it. In the circumstances, the one would
be absent if the other were. And because of this, we have no trouble
with the counterfactuals which would be indicative of genuine
overdetermination. It is not true that the behavioural result would
still have been caused even if the physical realizer had been absent,
for the higher state would then have been absent too;• and similarly,
if the higher state had been absent in some particular case, there
would again have been no alternative cause for the behavioural
result, since the physical realizer would have had to be absent too.
Note that it is not essential to this rejigged version of the causal
argument that we start with any assumption that conscious states are
higher states. I shall be considering alternative arguments for
materialism shortly, and in particular a form of argument that
begins with a functionalist assumption of just this sort, taken to be
derivable a priori from the structure of our concepts of conscious
states. If you begin with this kind of a priori functionalism, a variant
.of the causal argument can still serve an important purpose: namely,
that of establishing that higher mental states are physically realized,
as opposed to being realized by some distinctive non-physical
8
Mightn’t this be false?• Since different physical properties can realize a given
higher state, isn’t it possible that a different realizer would have beei:i present if the
actual realizer had been absent? There ~re some delicate issues here (cf. Yablo
1992). But for present purposes the ·significant point is that a given creature
certainly wouldn’t have had a given higher property if it hadn’t had the kind of
physical property that features as a realizer when it itself has that higher property.
For example, I certainly wouldn’t have been in pain if I didn’t have the killd of
physical property that realizes pain in me (since, aftei all, there isn’t any question
of my having the realizer that occurs in octopuses or other creatures).
33
THE CASE FOR MATERIALISM
mind-stuff (cf. Lewis 1966). But this is not how I am thinking of the
rejigged causal argument.
Rather, I intend it to establish both that conscious states must at
least be higher states, if not strictly physical, and that they must be
physically realized. That is, I am taking the identity of conscious
properties with higher (or physical) properties to be the conclusion of
my argument, not a premiss. The premisses are simply (1*), (2), and
(3), which make no claims, a priori or otherwise, about the specific
nature of conscious states, and the conclusion is that, if conscious
properties are not strictly identical with physical properties, then
they must at least be identical with higher-properties-which-are­
physically-realized, otherwise we will be driven to deny that
conscious states cause their effects in any sense or, alternatively, to
accept that those effects are genuinely overdetermined by quite
distinct causes. 9
So far in this section I have shown how functionalism and other
‘higher property’ versions of materialism can respect the premisses of
the causal argument, and indeed can use the rejigged version as an
argument in their favour. However, I have not intended this as a
defence of such views. This is because I am not sure whether they can
really be cleared of the charge of epiphenomenalism.
The issue here hinges on whether we can seriously allow that
higher states cause what their realizers cause. I am not sure what to
say about this. Sometimes I think that this is not a serious notion of
causation, and certainly not one which does justice to the way in
which my thirst causes me to drink a beer. Surely, one feels, my thirst
itself is efficacious in getting me to move, in just the same strict way
as physical causes produce their effects, and not merely in the
second-hand sense that it is realized by some other state which
causes in this strict sense.
When I am in this mood, I am inclined to read the causal argument
as employing a strict notion of causation throughout, and in
particular in premiss I’s assertion that conscious states cause physical
9
I take this construal of the causal argument to rebut Tim Crane’s (1995)
complaint that the causal argument must assume a form of mental causation
which will then be denied by any functionalist-style higher-property version of
materialism.
34
THE CASE FOR MATERIALISM
effects. This then drives me to the conclusion that conscious
properties must be identical to strictly physical properties, and that
any higher properties are merely epiphenomena!. The cost of this
strictly physicalist position, of course, is that I will not share
conscious properties with octopuses or other physically distinct
beings. But perhaps this isn’t as bad as it seems. After all, it doesn’t
mean that octopuses don’t have any conscious properties at all. And I
will still share some properties with them, albeit not the conscious
properties that strictly cause our respective behaviours. (In particular,
I will share some higher properties, which is perhaps why we can
both count as in ‘pain’.)
At other times I feel less fussy about causation. In particular, I
sometimes worry that we will be left with precious few causes, ifwe are
going to hold that higher states are pre-empted as causes whenever
they have realizers in virtue of which they cause. For, if applied
strictly, this principle threatens to block the causal efficacy of even
such eminently respectable causal states as pressures and tem­
peratures. After all, on any particular occasion the effects of
temperatures and pressures will also be caused by specific
molecular movements. These specific movements will realize the
relevant pressures or temperatures, but won’t be identical to them,
since the pressures and temperatures can also be realized
differently. So the pressures and temperatures won’t count as
causes, if they can’t cause what their realizers cause.
This seems odd, and argues against dismissing higher states from
the realm of serious causes_, and in favour of a generous reading of
premiss r. On this reading, my thirst will still be a serious cause of my
going to the fridge, even if it has a realizer in virtue of which it causes.
And then the causal argument will simply yield the conclusion that it
must be a physically realized higher state, not that it must be strictly
physical itself.
As I said, I am not sure what to say about this issue. It is a
complicated matter, and it is not clear how best to resolve it.
Fortunately, nearly all the arguments in the rest of this book will be
insensitive to this issue. We can identify conscious properties either
with strictly physical properties or with physically realized higher
properties. Whichever choice we make, we will still have an identity
35
THE CASE FOR MATERIALISM
between conscious properties and properties which are innocent of
any of the obscurities which surround consciousness. This is the
important point, and beyond that it will not matter too much
whether conscious properties are identified with strictly physical or
with physically realized higher properties. 10
r. 8 Intuition and Supervenience
Let me now distinguish the causal argument we have been
examining from some other ways of defending materialism that
can be found in the recent literature.
To start with, it is sometimes suggested that materialism about
consciousness can be established by a priori intuition alone. This is a
feeble thought, as will become clear shortly, but its deficiencies have
sometimes been obscured by the fashion for thinking of materialism
about the mental in terms of 1supervenience 1 : that is, in terms of the
doctrine that any two beings who share all physical properties must
also share all mental properties.
I myself find the notion of supervenience more trouble than it’s
worth. The notion of supervenience has proved far less straightfor­
ward than it at first seemed, and has generated a huge amount of
technical literature (mostly focusing on the ‘must’ in ‘if … physically
identical … must also be mentally identical’). I would argue that any
benefits offered by the notion of supervenience are more easily
gained simply by identifying mental properties directly with higher­
order properties or disjunctions of physical properties. Accordingly,
the notion of supervenience will not play a prominent part in the rest
of this book.
I mention it here only because supervenience formulations of
10
In Chapter 7 I shall r~tum to the choice between functionalist-style higher­
property views and strict physicali_sm, and consider whether empirical research
into conscious properties can help resolve the issue.-By that stage I shall also have
developed an extensive account of the structure of our concepts of conscious
properties. But I shall argue that none of this delivers a resolution, and my eventual
conclusion will be that our concepts of conscious properties are vague, in that it is
indeterminate whether they refer to higher properties or to the physical properties
which realize them in humans.
THE CASE FOR MATERIALISM
materialism can create the spurious impression that materialism is a
purely intuitive matter. After all, there is a sense in which a priori
intuition does tell us that the conscious realm supervenes on the
physical realm. Everybody has strong intuitions about the correla­
tion between mind and brain. If I made a molecule-for-molecule
physical copy of you using a Star Trek-style teletransporter, for
example, wouldn’t your physical twin automatically have all the
same feelings that you have?
However, this intuition-based supervenience falls far short of
anything worth calling materialism. To see why, note that the
teletransporter thought-experiment is consistent with epipheno­
menalism: perhaps the copy feels like the original simply because its
brain states causally generate extra conscious states, in just the same
way as the original’s brain states do. Here the conscious states would
be distinct from the brain states, but would regularly accompany
them, in virtue of laws by which brain states cause conscious states. A
merely epiphenomenalist mind-brain correlation like this clearly
doesn’t amount to materialism.
In the technical terminology into which we are forced by the
apparatus of supervenience, the point is that the teletransporter
thought-experiment shows only that physical identity guarantees
conscious identity across natural possibilities, possibilities which
share all our natural laws, including any brain-mind epiphenomena!
laws. However, to establish a supervenience amounting to genuine
materialism, we would need to show that physical duplicates
couldn’t possibly be mentally different, whatever the laws of
nature, not just that they aren’t different in worlds which do share
our laws. We need to establish supervenience of the mental across all
metaphysically possible worlds. Only this promises to ensure that
the mental is ontologically inseparable from the physical, and not
just correlated with it.
If you find this obscure, the point can be put more directly in terms
of ontological relations between mental and physical properties.
Mere supervenience across naturally possible worlds doesn’t amount
to materialism, because it doesn’t rule out· the epiphenomenalist
possibility that conscious properties are ontologically quite distinct
from physical properties, albeit constantly correlated with them by
37
THE CASE FOR MATERIALISM
THE CASE FOR MATERIALISM
epiphenomena! laws in this actual world and those nearby worlds
the causal argument, so to speak, to deliver the materialist
that share our natural laws. Supervenience across all possible worlds,
conclusion. The a priori analysis tells us that conscious states have
on the other hand, does arguably suffice for materialism, precisely
because an ontological dependence of mental on physical properties
seems the only thing that will enable physical identity to necessitate
a causal role, and hence have physical effects. The completeness of
physics tells us that these physical effects must have full physical
histories. The denial of overdetennination tells us that these physical
mental identity, whatever laws may obtain.
Now that we see which version of supervenience is required to
ensure genuine materialism, it should be clear that intuition alone
will fail to deliver the materialist goods. It is not at all intuitively
effects aren’t caused twice over. Thus, once more we reach the
obvious that physical duplicates must necessarily be conscious
duplicates, that a physical doppelganger couldn’t possibly have
different experiences. Even a dyed-in-the-wool materialist, like
myself, feels the pull of the intuition that there could be a
‘zombie’, say, who is physically just like me but has no feelings-in
a possible world, so to speak, where any epiphenomena! laws relating
brain states to conscious states have broken down.
It may in fact be true that zombies are impossible, and indeed this
is something for which I shall argue at length in due course. My
present point is only that a priori intuition alone cannot establish
their impossibility. If anything, it suggests just the opposite.
r. 9 An Argument from A Priori Causal Roles
Let me now consider one further form of argument for materialism.
This shares some of the structure of the causal argument. But in place
of premiss I or r*, which simply states that, as a matter of fact,
conscious causes have physical effects, this argument appeals instead
to a putative a priori analysis of our concep”ts of conscious states.
According to this line of thought, our concepts of conscious states,
like pain, or thirst, are each associated a priori with the specification
of some causal role linking that state to physical causes and effects
(cf. Lewis 1966). So, as above, our concept of pain would be linked a
priori with bodily damage as cause and a desire to avoid the source of
the pain as effect. Again, our concept of thirst would be linked to lack
of water as cause and a desire to drink as effect.
This kind of a priori analysis can then be plugged into the rest of
conclusion that conscious states cannot be ontologically distinct
from the physical causes of their physical effects.”
There may seem no great distance between the causal argument
discussed earlier and this argument appealing to a priori analyses of
our concepts of conscious states. However, it is crucially important
that the causal argument discussed earlier rests on no such a priori
assumptions. While that causal argument assumed that conscious
causes have physical effects, it offered this as a straightforward
empirical truth, not as a conceptual matter.
In line with this, note how I have been happy to allow the
conceptual possibility that conscious states may lack effects alto­
gether. This point arose earlier in my discussion of epiphenomen­
alism. My reason for dismissing epiphenomenalism was not that its
denial of mental efficacy violated any conceptual truths, but simply
that it amounted to an empirically far less plausible story than the
simple identities postulated by materialism.
I shall have a lot more to say about our concepts of conscious states
in what follows. Without wanting to pre-empt that analysis, let me
simply say at this stage that it will amply confirm that there are no
11
Interestingly, this form of argument can be used to deliver either a
functionalist version of materialism-conscious states are identical with physically
realized higher-order states-or a more strictly physicalist conclusion-conscious
states are identical with the physical states themselves. We will get the former
conclusion if we take it to be given a priori that conscious properties are higherw
order properties-that is, that they are identical to properties like thewpropertywof­
havingwsomewpropertywwhichwplayswsuch-andwsuchwawcausalwrole. The overall argu­
ment will then lead us to the conclusion that these higher-order properties must be
realized physically, as opposed to being realized by some special mindwstuff. But we
can also take the causal roles associated a priori with concepts of conscious states to
fix reference to whichever firstworder properties actually play those roles, rather
than to the roles themselves. The a priori argument then delivers the conclusion
that conscious properties are identical to strictly physical properties. This is in fact
how the argument is run by David Lewis.
39
THE CASE FOR MATERIALISM
a priori associations between concepts of conscious states and
specifications of causal roles.
On this conceptual issue, I am thus in agreement with a number of
recent writers who have argued that the a priori style of argument for
materialism doesn’t work (cf. Levine 1983, Chalmers 1996). They
object to the initial a priori claim about concepts of conscious states.
Our concepts of conscious states are not a priori related to any
specifications of causal roles, they protest. So there is no conceptual
route, they conclude, from the fact that any causal roles must be
filled by physical states to the conclusion that conscious states are
material.
I accept this criticism of the a priori style of argument for
materialism. To repeat, I agree that our concepts of conscious states
are not associated a priori with causal roles. But this isn’t as bad for
materialism as Levine and Chalmers suggest. If the a priori argument
were the only argument for materialism about consciousness, then
materialism would indeed be in trouble. However, it is not the only
argument. There is also the original causal argument as I have
presented it, which does not depend on any particular assumption
about our concepts of conscious states.
1.10
Whatis ‘Physics’?
Let me now address a terminological issue flagged earlier, an issue
that may have been worrying readers for some time. How exactly is
‘physics’ to be understood in this context of the causal argument? An
awkward dilemma may seem to face anyone trying to defend the
crucial second premiss, the completeness of physics. If we take
‘physics’ to mean the subject-matter currently studied in depart­
ments of physics, discussed in physics journals, and so on, then it
seems pretty obvious that physics is not complete. The track record
of past attempts to list all the fundamental forces and particles
responsible for physical effects is not good, and it seems highly likely
that future physics will identify new categories of physical cause. On
the other hand, if we mean by ‘physics’ the subject-matter of such
40
THE CASE FOR MATERIALISM
future scientific theories, then we seem to be in no position to assess
its completeness, since we don’t yet know what it is.
This difficulty is more apparent than real. If you want to use the
causal argument, it isn’t crucial that you know exa

Calculate your order
Pages (275 words)
Standard price: $0.00
Client Reviews
4.9
Sitejabber
4.6
Trustpilot
4.8
Our Guarantees
100% Confidentiality
Information about customers is confidential and never disclosed to third parties.
Original Writing
We complete all papers from scratch. You can get a plagiarism report.
Timely Delivery
No missed deadlines – 97% of assignments are completed in time.
Money Back
If you're confident that a writer didn't follow your order details, ask for a refund.

Calculate the price of your order

You will get a personal manager and a discount.
We'll send you the first draft for approval by at
Total price:
$0.00
Power up Your Academic Success with the
Team of Professionals. We’ve Got Your Back.
Power up Your Study Success with Experts We’ve Got Your Back.
WeCreativez WhatsApp Support
Our customer support team is here to answer your questions. Ask us anything!
👋 Hi, how can I help?