Most philosophers have typically used the
term ‘experience’ to refer to the phenomenon of phenomenal consciousness;
‘phenomenal consciousness’ and ‘experience’ are used interchangeably. The
mainstream view (Chalmers 1996, Block 2002, Kim 2005) has it that our sensory
qualities, such as a pain or an itch in one’s finger are intrinsically
conscious. These qualities are typical examples of phenomenal properties, which
are taken to be distinct from any cognitive, intentional or functional
properties. Jaegwon Kim (2005) for instance, argues that these properties
cannot be functionalized even in principle.

David Rosenthal
argues in this book that this view is mistaken. The main reason that
consciousness appears unamenable to a physicalist reduction is that it is
mistakenly construed as an intrinsic and therefore unanalysable property. In
the case of phenomenal consciousness, this construal rests heavily on the
assumption that although many types of mental states such as thoughts, desires
and beliefs can occur unconsciously, our sensory states/qualities are
intrinsically conscious. Rosenthal notices quite rightly that that phenomenal
consciousness appears to be inexplicable is due to its apparent self-intimating
nature, which makes one think that an intrinsic theory or that a necessary
connection between consciousness and sensory qualities appears more plausible.
He argues throughout his book that all types of mental states including sensory
qualities are not invariably conscious. Regarding the latter, one of the
examples he offers to this effect is that we normally do speak of having the
same headache all afternoon, even though the awareness of our pain is
intermittent. Whereas the ‘headaching’ quality endures all afternoon says
Rosenthal, sometimes we have a higher-order thought towards it and sometimes
not. Rosenthal concludes that during these intervals, that is, when we do not
have a higher-order thought towards the headache, there is nothing it is like
to have it.

The book is a
collection of essays some of which appear here for the first time and it is
divided into four parts. In part I, Rosenthal proposes his higher-order thought
view. He suggests that we must draw a three-fold distinction among unconscious
states, nonintrospectively and introspectively conscious states. He argues that
not only our desires, thoughts and beliefs can occur unconsciously but also our
sensory states. One is in a conscious state such that there’s something it’s
like for one to be in it only if one is aware of being in that state. This
awareness is in turn explained in terms of one’s having a higher-order thought
(HOT) to the effect that one is in that state. In order that is, to have a pain
sensation, I need to have a HOT about this sensation. But HOT needn’t be itself
conscious. That would require a third-order thought (introspection).
Rosenthal’s idea is that consciousness is a relational property (as opposed to
an intrinsic property). Since both the first-order state (sensory state) and
the HOT are not conscious, consciousness is analysable and amenable to a
physicalist reductive explanation.    

The view owes its main attraction to the
fact that these higher-order thoughts by virtue of which one becomes conscious
are occurrent episodes in immediate contrast with being dispositional states.
Peter Carruthers (2000) for instance, proposes that a state becomes conscious
by virtue of being disposed to give rise to a higher-order thought and not by
virtue of being the actual target of such a thought. Thus nonintrospectively
conscious states are dispositional states not occurrent. While it may be true
that as Rosenthal puts the point, a dispositional thought is just a disposition
for a thought to occur and cannot make a first-order state conscious, occurrent
episodes come with a price; because at any given moment, we would be receiving
an enormous amount of perceptual inputs, and thus we would be representing all
of them on a higher-order cognitive level, it is hard to see how all these
occurrent thoughts or beliefs wouldn’t result in a ‘cognitive overload’.
Moreover Rosenthal’s three-fold distinction seems oversimplified. It would
sound more plausible to suggest that there are degrees of awareness rather than
states that suddenly become conscious or as Rosenthal says states that suddenly
‘light up’. Take the often-cited case of the absent-minded driver (Armstrong,
1968). According to Armstrong, when one is driving very long distances in
monotonous conditions one can come to at some point and realize that one has
driven many miles without being conscious of the driving. Why not say that in
this case the driver is minimally conscious of the road instead of saying that
he suddenly becomes aware of his surroundings and where he currently is? 

In part II,
Rosenthal proposes his homomorphism theory of mental qualities. Although his
HOT theory is representationalist, Rosenthal takes the view the sensory
qualities (i.e. properties of our first-order sensory states) are non-representational.
Rosenthal argues that our distinct sensory qualities resemble and differ from
one another in ways that parallel the similarities and differences among
perceptible properties of the physical objects of the environment.  He suggests
that we should fix the family of mental qualities characteristic of each
sensory modality by reference to the family of perceptual properties to which
the modality in question enables us to have perceptual access. So the idea is
that our sensory states must themselves resemble and differ from one another in
ways that are homomorphic to the similarities and differences among the
physical colors that the organism can discriminate. On this view, physical red
for instance, is determined by the characteristic way it resembles and differs
from other discriminable properties in the quality space of perceptible colors.
So if homomorphism is correct it enables us to fix each mental property partly
by reference to the relevant family of perceptible properties and in comparative
terms. Recall that according to Rosenthal sensory qualities are not invariably
conscious. It appears then that this view enables us to individuate
conceptually the mental qualities that appear both in perceptual discrimination
and in conscious qualitative experience thereby refuting the view that sensory
qualities are ineffable. Further, granted that sensory qualities can be defined
in terms of their perceptual roles cases where molecular/functional duplicates
might nonetheless have inverted color sensory qualities — spectrum inversion
scenarios- lose their intuitive force. If sensory qualities can occur
unconsciously then they can be defined in objective terms. And if sensory
qualities are to be identified with functional roles then spectrum inversion scenarios
don’t seem to be possible since functional roles cannot be inverted.   

In part III,
Rosenthal observes that verbally expressed thoughts are conscious and explores
the relation between consciousness and language. He draws an important
distinction between reporting and expressing and he argues that the content of
a conscious state is not determined by the nonassertoric state but by the
assertoric state. A thought according to him, is an occurrent belief; any
episodic intentional state with an assertoric mental attitude. As in the case
where one doubts that an object is blue one needs to be conscious of the
object, when one doubts or wonders whether a mental state has a particular
property one must similarly be conscious of that mental state. If I doubt that
this object is blue I must thing that the object is there. Likewise,
non-assertoric attitudes like desiring, wondering and doubting don’t seem to
make one conscious of being in that state. Assertions have truth conditions.
According to Rosenthal, the truth conditions for p and ‘I think that p‘
differ. The truth conditions of the statement ‘it’s raining’ for instance, are
not the same as the truth conditions of the statement ‘I think that it’s
raining’. This is because the latter doesn’t express my first-order thought
that ‘it’s raining’ but my higher-order thought that ‘I think it’s raining’.
Therefore it has different propositional content. Rosenthal claims plausibly
that propositions of the form ‘not-p, but I think that p‘ or ‘p,
but I think that not-p‘ although cannot rationally be asserted we can
consider what it would be for them to be true. In an anti-Wittgensteinian
spirit, it is not incoherent for one to suppose for the sake of the argument
that it’s not raining but I think that it is. Rosenthal says that
‘[Wittgenstein’s] exclusive reliance on use and the consequent assimilation of
reporting to expressing led to his rejection of mental states that aren’t
conscious’ (p. 293). He concludes this Part by arguing quite convincingly that
we should not appeal to any monitoring or quasi-perceptual mechanism to explain
the relative accuracy of the way we are conscious or our mental states. He
argues that when the individual has the thought that it’s raining, ‘the way it
behaves and its knowledge of the ambient situation tend to lead to its also
having the thought, ‘I have the thought that it’s raining’ (p.16). We initially
says Rosenthal come to think of ourselves as being in perceptual states by
noting occasional errors about the things we take to be in our immediate
environment. Taking note of our perceptual errors leads to our thinking of
ourselves as being in those states. 

In the final
part, Rosenthal attempts to explain the most important characteristic of a
higher-order thought. The thought must be occurrent, assertoric, noninferential
(in the sense that it is not the result of conscious inference from a
first-order thought — since HOT itself needn’t be conscious) and most
importantly must be ‘of oneself’. That is a HOT is a de se belief. The
content of such belief is roughly ‘I’m in F’ (where F indicates the first-order
state). Rosenthal appeals to John Perry’s (1979) well-known example to argue
that self-consciousness is not required for such a thought, ‘the reason has to
do with the special way in which we sometimes refer to ourselves when we speak
using the first-person pronoun and frame thoughts using the mental analogue of
that pronoun (p.348). Roughly speaking, the ‘of oneself’ requirement is
explained in terms of the occurrence of essentially indexical reference with
each HOT.

Overall, I think, David Rosenthal’s
Consciousness and Mind presents us with the most plausible higher-order
representationalist account in the market. The book is engaging and very
clearly written. Rosenthal is one of the very few philosophers who hold the
view that our sensory qualities are not invariably conscious and he does a very
good job in bringing up the advantages of such a view. What’s more, he gives us
good reason to think that consciousness is explicable and analysable and I
think that if there is a possibility to this effect we should better explore
it.

© 2006 Dimitris
Platchias

Dimitris
Platchias is studying for a Ph.D. in philosophy in the University of Glasgow.
 His main interests lie in the Philosophy of Mind and especially in the
Philosophy of Perception and Consciousness.

Categories: Philosophical, Psychology