John Perry has written a very detailed but small book, Knowledge,
Possibility and Consciousness (KP&C). It is based
on his Jean Nicod lectures, given in Paris in 1999. These lectures,
as the Forward indicates, are given by a leading philosopher of
mind or philosophically aware cognitive scientist, so you can
know right away what kind of book this is and the kind of audience
at which it is aimed. This is complex, academic philosophy. It
pits physicalism (the idea that every thing can be explained in
physical terms) against dualism (there are two kinds of substance,
mental and physical), and so the volume lies within the purview
of philosophy of mind. The conceptual machinery used to develop
his arguments is borrowed from the philosophy of language, which
itself is used to challenge the epistemological (i.e., to do with
the theory of knowledge) assumptions of a couple of dualist arguments.
Writing a review for such a technical book itself must be technical.
In effect this review will mirror some of the main themes of the
book with a few personal observations at the end. Critical responses
to KP&C will undoubtedly appear in philosophical journals.

Perry’s own position is physicalist and he is an
identity theorist. He quotes Herbert Feigl, also a physicalist,
at the beginning of most of the chapters. Feigl brought the identity
theory into contemporary philosophy, arguing that mental states
are identical to brain states. Perry’s particular stance he calls
antecedent physicalism. Antecedent physicalism isn’t
a kind or species of physicalism, but an attitude towards it.
Think of this attitude as a common sense prima facie acceptance
of physicalism, a default position. If you intuitively lean towards
physicalism then you are an antecedent physicalist. You may have
considered possible arguments against your physicalism but none
has led you to give it up.

Perry wants to defend physicalism against versions of what he
calls neo-dualist arguments. Think of ‘neo-dualist’ as
property dualist – the idea that physical things possess two distinct
kinds of properties, e.g., for minds, mental and physical properties.
The neo-dualists and their arguments are David Chalmers’ “Zombie
argument”, Frank Jackson’s version of the “Knowledge
argument”, and Saul Kripke’s argument against the Identity
theory.

Here are abbreviated versions of these arguments to familiarize
the reader:

The Zombie argument: Physicalism is the view that everything
can be explained in physical terms. So even consciousness can
be explained in physical terms. But there is a logically possible
world (it’s conceivable) where there a beings, zombies, who are
physically indiscernible from us, yet lack consciousness. If physicalism
were correct, these zombies would also possess consciousness since
they are indiscernible from us. Therefore consciousness must be
a further fact about the world over and above the physical facts.
Consequently, physicalism is inadequate to explain consciousness.

The Knowledge argument: Mary is an expert who knows all there
is to know about color but has been raised in a room that is completely
black and white. She has never seen any color until she leaves
the room and sees a ripe tomato. She learns something new, exclaiming,
‘this is what it is like to see red!’ Therefore, there must be
facts about experiencing color that are not physical facts. Mary’s
learning a new fact about the world shows that physicalism is
inadequate.

Kripke’s argument: Physicalists rely on the identity between
sensations and brain states. All identity is necessary. That is,
if one fact is identical to another fact, then there is no possible
situation where they are not the same. But even the physicalist
admits that the relation between sensation and brain state is
contingent. There is, however, no contingent identity. Therefore,
sensations are not brain states. The identity theory is false
and physicalism is false.

Out of the seven chapters that comprise KP&C, there
is a very good introduction to the main themes in chapter 1. Chapter
2 outlines Perry’s stance on physicalism. There are three chapters
devoted to the Knowledge argument, and one each to the Zombie
argument and the Modal argument.

An apt motif for the physicalist viewpoint is taken from a 1960’s
science fiction movie, Fantastic Voyage. Neurosurgeons
were shrunk and injected into a very important person who was
dying of a brain clot, with the aim of operating on him from the
inside. While they were in this person’s brain, a puff of blue
smoke appeared (it was a 1960s movie). One of the neurosurgeons,
played by actress Raquel Welch, exclaimed something like “Look!
We are the first to see a human thought!” This exclamation
expresses the fundamental belief of the physicalist perspective
on consciousness – that we should be able to literally see, with
our own eyes (and with a little help from the right technology),
other people’s thoughts since physicalists believe that
everything, including people’s thoughts, is physical. The dualist
and neo-dualist disagree – there’s more to the picture than meets
the eye. There is a gap between the sensation or experience (the
qualia) of say, tasting chocolate and the corresponding brain
state. The puff of blue smoke is only half the story. Something
else is going on and we just don’t know what it is; at least the
physicalist can’t know by just looking at it. This gap between
brain state and sensation is the basis of the experience gap
argument, which presents prima facie difficulties for
the physicalist who cites an identity between brain state and
sensation. The three arguments of the neo-dualists are all sophisticated
versions of the experience gap argument.

I will outline what Perry has to say about the failings of the
Knowledge argument, since this one takes up a large slab of KP&C.
The Zombie argument and the Modal argument are problematic for
similar reasons, according to Perry.

Now, physicalists can respond to the Knowledge argument in various
ways. The kinds of responses tend to fall within the following
categories. (a). Deny that Mary learnt anything knew, (she did
after all, ‘know all there is to know about color’); (b). Affirm
that Mary did learn something but it was not a new fact but the
acquisition of an ability- a kind of knowing how; (c).
Affirm that Mary learned something new but only by virtue of seeing
something from a new or different perspective.

This last approach has recently become popular among the many
responses to the Knowledge argument within the academic press.
It basically rejects the presumed application of Leibniz’s law
by the dualist in his attempt to defeat the physicalist. Leibniz’s
law in this context is the idea that, given two facts, if the
first fact is the same as the second fact, and you know the first
fact then you should also know the second fact. So, if physicalism
was correct, then every fact about color experience should be
the same as neurological facts. Mary knew all the neurological
facts about color, therefore, she should have known all about
the sensation of experiencing color, which she didn’t. But it
is easy to find a counterexample to Leibniz’s law especially when
constructed within the contexts of beliefs. For example, aubergines
are eggplants, but it doesn’t follow that you know that the vegetable
used in moussaka is an aubergine on the basis that you knew that
it contained eggplant. You have still learned something new about
the same thing. So the application of Leibniz’s law fails in intentional
or epistemic contexts (cases concerning what we believe or know
about). The approach used in KP&C to refute the Knowledge
argument is a species of this last ‘perspectival’ method. And
here Perry shows how the Knowledge argument relies on an epistemological
error. The mistake is to think that when we learn something new
about the world, we learn a new fact that we didn’t know before,
rather than knowing the same thing but in a different way. This
‘oversimple conception of knowledge’ (p16) is put aside for a
better epistemology that can explain Mary’s new knowledge without
threatening physicalism.

Perry’s underlying strategy throughout the book is to explain
the apparent ‘twoness’ inherent in discussions that focus on consciousness.
This duality or twoness consists of both, for example, the brain
state of seeing red and the subjective experience, or qualia –
that what is it like to see red quality. This topic isn’t
for the reader who has a morbid fear of the number two. Though
it’s too easy in philosophical life to bifurcate our thoughts
and problems into this and that, here we are left with no choice
given the way the problem of consciousness has been constructed
by both sides of the debate.

Whereas the dualist sees this twoness as implying that there are
two different things known, the physical and the non-physical
(i.e., consistent with property dualism), Perry and the antecedent
physicalist insist that this twoness only amounts to two ways
of knowing about one (physical) thing (p206). Put another
way, the dualist sees that there are two different ontological
objects or things that exist – the physical and the mental – while
the physicalist sees that there is only one ontological object
but two epistemic objects or ways of knowing. This ‘two ways of
knowing’ is nominated as the two ways strategy. If
the physicalist can demonstrate that, in the Knowledge argument,
what is happening is that there are two ways of thinking of a
single state, ‘he ought to be able to block the inference
that there must be two states… a physical and a non-physical‘
(p19). ‘The twoness does not occur in the subject matter [ie.,
the thing itself] but in the way of knowing’ (p206).
(For the readers familiar with these issues, the author’s
motivation for this strategy arises from the works of Brian Loar
and William Lycan. However, Perry’s innovation to this debate
is the development and use of the idea of reflexive content.
Perry has written extensively on reflexivity and cognate notions
like indexicality elsewhere [like in problems of reference]. Here,
he applies these tools to an epistemic understanding of the second
kind of knowledge that Mary acquires after leaving the black and
white room).

Molyneaux’s problem is used as an analogy in KP&C.
This problem – concerning whether a blind person, after gaining
their sight, could visually recognize one particular shape that
they already knew by touch – parallels the antecedent physicalist
position. In both cases there is a true and informative identity
that could not be determined without experiencing both sensations
– sight and touch. There are two different ways of knowing, but
only one thing known (e.g., recognizing a cube by sight versus
recognizing it by touch). So the physicalist has a more parsimonious
explanation by identifying the brain state with the sensation.

Consider the proposition this brain state is this sensation.

The important thing to note is that ‘the brain state that is known
by inner attention is not known by causing a
sensation, but by being one’, (p206).

This ‘two ways’ of understanding is an undercurrent theme throughout
the book. The simplicity of this message is admitted by Perry
himself. In his conclusion to the book, he writes, ‘As philosophy
books go, this one is not so long. But perhaps it is too long,
given the simplicity of the message’ (p202).

What is not so simple is Perry’s deeper analysis of his two ways
of knowing. This analysis consists of the subject matter assumption
and the subject matter fallacy, which Perry claims the neo-dualist
commits. This is to think that there is a change in subject matter,
or what her belief is about, when Mary acquires new knowledge.
Instead there is a change in reflexive content, as Perry calls
it. My understanding of this is that when you come to learn that
moussaka contains aubergine, you don’t learn two different facts
about two different things, rather, there are two ways of knowing
one thing.

On one hand Perry writes that the message of the book is simple;
but, paradoxically, he writes of the subject matter assumption
that it’s a bit hard to state and difficult to relate to physicalism.
This is an understatement; I found the sections on the subject
matter assumption and reflexive content lacking the usual perspicuity
found in other examples of Perry’s writing. I put this down to
the subject matter of the subject matter assumption (no pun intended),
rather than a lapse in clearness of authorship. I kept thinking
that I was lacking the necessary apparatus to grasp the point.
Either what was written is so easy and obvious that I could not
attain the realization that I had learnt something new, or it
was so difficult and abstruse that no amount of backtracking helped.
This resulted in a paradoxical intuition – that I think he is
right, but I’m not sure why.

His misgivings about the other neo-dualist arguments are much
clearer. His reasons for rejecting the Zombie argument seem sound
though a bit surprising. For instance he writes, ‘What may be
somewhat surprising…is that the possibility of a Chalmers’
zombie world really has nothing to do with the issue of physicalism
versus dualism. It is a test for epiphenomenalism versus the efficacy
of the conscious’ (p28). Yes, I was amazed to read this and wondered
why hasn’t someone thought of this before. My first reaction being
the question of whether he has won the case by redefining epiphenomenalism;
though admittedly it is quite a malleable doctrine.

Turning to a specific criticism about the volume, I found the
Index quite annoying. All the really hard concepts that Perry
declared as his innovation do not get a mention. I’m talking about
recognitional knowledge (not discussed here), reflexive
content, and his views on identity. Nor are other terms
like epiphenomenalism, dualism, or property dualism
mentioned, though they do occur throughout the book. When I first
came across these concepts I went to the index to find out what
else he had written about them, but found no entry. I did however,
find four references to Raquel Welch, one reference to
orgasm – a favorite example of the philosopher of mind
of that ‘what is it like to feel something’- and twelve mentions
of chocolate chip cookies. I don’t want to dismiss the
book on the basis of easy humor, but it would have been extremely
helpful to find a categorization and perhaps a glossary of these
technical terms, which are so fundamental to his argument.

Finally, my impression of the dialectical strategy of the book
is as follows. Perry’s broad approach is like a big ‘either-or’
argument. He doesn’t so much offer direct arguments for
physicalism but instead undermines neo-dualism. What he does for
physicalism is to offer a kind of best explanation argument of
why antecedent physicalism is preferable over neo-dualism. But
I cannot see why Perry’s arguments don’t amount to just a kind
of epistemic possibility: sure, neo-dualism has been undermined,
you might say, there aren’t two things known but two ways
of knowing. For all I know, this is correct given the power
of Perry’s analysis in the context that there is a change in ‘reflexive
content and not a change in subject matter’. But we are still
a long way from direct demonstration that physicalism is correct.
Perry asks us to choose between two competing explanations of
consciousness. Yet in accepting the more parsimonious explanation,
one is already committed to preferring the simplest, (two ways
of knowing), rather than the more difficult (two things known).
It is for this reason that I think the antecedent physicalist
case has only been made consistent or warranted for the person
already leaning towards physicalism. Until we can ride with Rachel
Welch in the miniature submarine and see the evidence for ourselves,
maybe that’s all that is possible for us.





© 2001 Robert Anderson

This book is available from Barnes & Noble.com: Knowledge,Possibility,and Consciousness