The zombie idea haunts the literature dedicated to the
philosophy of mind and consciousness.  The philosophical zombie is an exact
physical duplicate of a human being and lives in a world essentially like ours,
but without subjectivity or ‘inner world’.  Among philosophers of mind we find
both supporters and detractors of the zombie idea; Robert Kirk, by his own admission
a former ‘zombie freak’ himself (p.25, fn.1), offers a thorough and cogent
refutation of the concept of zombie as understood above.   In the first four
chapters of the book, the reader is presented with a compelling case for the
inconceivability of zombies.   The remaining seven chapters offer a fresh view
of phenomenal consciousness which is based on Kirk’s own brand of
functionalism.

In chapter 2, ‘Zombies and Minimal
Physicalism’, Kirk argues that ‘if zombies are even possible, physicalism is false’
(p.7).   The Author puts forward his ‘strict implication thesis’, which is
necessary for a commitment to minimal physicalism (i.e., no one who professes
to be a physicalist can consistently withdraw adherence to this thesis and
still remain a physicalist).  Strict implication is defined as follows: ‘A
statement A strictly implies a statement B just in case "not- (if A then
B)" is inconsistent or incoherent for broadly logical or conceptual
reasons’ (p.10).  On the basis of this definition, if P stands for the ‘conjunction
of all actually true statements’ in the ‘austere vocabulary of idealised
contemporary physics (p.9), and Q stands for the ‘conjunction of the totality
of actually true statements in psychological language about the individuals
whose existence physicalists suppose to be provided for by P’ (p.10), then the
strict implication thesis can be stated thus: ‘P strictly implies Q’.  In other
words:

"P and not-Q"
involves inconsistency or other incoherence of a broadly logical or conceptual
kind… (p.10).

That
we cannot easily imagine or conceive of how P entails Q or, alternatively, that
we can easily conceive P and not-Q, is no valid argument against such
entailment (p.11).  Further, Kirk states the ‘redescription thesis’, which runs
as follows:

Any true statements about
the world not expressible in the austere physical vocabulary of P are pure
redescriptions of the world specified by P. (p.9).

The above allows
Kirk to argue that physicalists are not committed to the impossible task of
finding austerely physical equivalents of psychological descriptions.  Finally,
although the strict implication thesis is necessary for minimal physicalism, it
is not sufficient; it is to be supplemented by the following: ‘nothing exists
other than what is strictly implied to exist by P’ (p.17).  Although whether P
implies Q is a question of logical entailment, the strict implication thesis
itself is empirical in character: ‘This is because P includes empirical
statements that just happen to be true in our world.’ (p.11).  More will be
said about this crucial point later.   The consequence of the zombie idea for
the strict implication thesis is that, since P implies Q, then, if zombies were
possible, this would contravene the strict implication thesis, hence (minimal)
physicalism would be false.   

Chapter 3, ‘The Case for Zombies’,
examines some previous arguments for the existence of zombies by such
well-established philosophers in the field as Chalmers, Block, Jackson, Nagel,
as well as by Kirk himself.  The central point Kirk makes is that if something
does seem conceivable it does not follow that it is therefore possible. 
Further, conceivability itself is constrained by possibility in the following
sense:

Conceivability in the
relevant sense needs to be an epistemic matter.  Arguing from conceivability to
possibility makes sense only so long as you don’t already know that the
situation in question is impossible… (pp.27-8).

In
Chapter 4, ‘Zapping the Zombie Idea’, Kirk points out that zombists fall prey
to what he calls the ‘jacket fallacy’:

[Zombists] mistakenly assume that
phenomenal consciousness is a property, which can be stripped off while leaving
the individual’s other main properties intact. (p.39).

Firstly,
Kirk points out that underpinning the zombie idea is an incoherent view of
phenomenally conscious experience, which he calls the ‘e-qualia story’. 
Undermining the e-qualia story by showing that it is inconceivable will
logically undermine the conceivability of zombies.  The e-qualia story’s
central point is that what makes individuals phenomenally conscious is that ‘they
stand in some relation to a special kind of non-physical properties, e-qualia;
e-qualia are caused by physical properties but have no effect in the physical
world (causal closure of the physical): the e-qualia could disappear and
nothing would change in the physical world as a consequence (i.e., jacket
fallacy); human beings are made of functioning bodies and their e-qualia;
finally, human beings do enjoy what Chalmers, a supporter of the conceivability
of the zombie idea, calls ‘epistemic intimacy’ with their e-qualia, that is,
they ‘are able to notice, attend to, think about , and compare their e-qualia.’
(p.40).  The incoherence to which Kirk points is that the last essential requisite,
namely, epistemic intimacy, is logically incompatible with the other components
of the e-qualia story listed above.  To prove the incoherency of the e-qualia
story, Kirk introduces the ‘sole-pictures’ thought experiment.  In short, if
Zob, Kirk’s zombie twin, were suddenly endowed with qualia, by the e-qualia
story he would be conscious.  Kirk enjoins us to imagine the following:  by a
change in the laws of nature, those neural processes which according to the
e-qualia story cause his visual e-qualia, are mirrored in Zob’s brain, only
those visual processes give rise to constantly changing pictures on the soles
of Zob’s feet.  Is it conceivable — asks Kirk — that Zob’s cognitive
processes are at all epistemically relevant to the sole pictures?  Zob does not
even notice the sole pictures and the latter have no effect whatsoever on his
perceptual and cognitive processes, which by hypothesis mirror Kirk’s (not Zob’s). 
Sole pictures have been modelled according to the structure of e-qualia as they
are conceptualised in the e-qualia story (at least in the relevant respects),
hence highlighting that story’s internal  inconsistency: ‘the story allows
individuals no more epistemic access to their e-qualia than Zob has to his sole
pictures …that is, none.’ (p.48).  The inconsistency of the e-qualia story
can be assessed by a priori reflection, which makes it not conceivable.  Kirk
points out that the e-qualia story is presupposed by parallelism and
epiphenomenalism, hence from the incoherence of the e-qualia story the
misconception of the view of consciousness entailed by both these metaphysical
positions follows.  Regarding dualistic interactionism, Kirk admits that it is ‘incompatible
with the scientific evidence’, but that he knows of no ‘a priori refutation of
it’ (p.56). 

Chapter 5, ‘What Has To Be Done’, paves the way to the
second part of Kirk’s project of working out necessary and sufficient
conditions for phenomenal consciousness.  The refutation of the zombie idea
leads to the suggestion that descriptions of the qualitative aspect of
experience, or in technical language ‘qualia’, are ways of talking about
physical processes.   Kirk examines Nagel’s well-known distinction between
first- and third-person perspective, or, in Kirk’s preferred terminology, between
viewpoint-relative and viewpoint-neutral concepts.  Viewpoint-relative concepts
can be fully grasped only by creatures endowed with the appropriate point of
view, which might be said to involve, among other things, a certain sensory
apparatus.  Further, grasping such kind of concepts ‘also requires actual
experiences in the right sensory capacities, perhaps even, in some cases,
experiences of the specific kinds to which the concepts apply.  At least it
requires the ability to create in imagination something akin to the experiences
in question.’ (p.62).  Viewpoint-neutral concepts are those which ‘are
accessible to any sufficiently intelligent creature — human, Martian, robot —
regardless of the specific nature of their perceptual systems’ (p.61), e.g.,
logico-mathematical concepts, concepts of physical theory, shape, distance,
mass, etc.  Nagel’s point is that it is impossible to convey the viewpoint
relative concepts or the ‘what-it-is-like’ perspective in viewpoint-neutral
language and, for this reason, a functional-scientific account of an organism,
e.g., a bat, cannot convey knowledge of what it is like to be that organism;
Nagel sees this as constituting serious problems for physicalism.  It is Kirk’s
view that Nagel has conflated two distinct problems.  Specifically:

One is the problem [Nagel] has emphasized: whether we
can get from a knowledge of relevant viewpoint-neutral facts to a knowledge of
the character of the bat’s conscious experiences: a knowledge of what it is
like (for the bat).  That is the what-is-it-like problem.  The other
problem is whether we can get from a knowledge of those same viewpoint-neutral
facts to a knowledge of whether the bat is phenomenally conscious at all.  That
is the is-it-like-anything problem.  (p.63).

Kirk
aims to tackle the second problem, that is, ‘what does it take — or what is
it — for something to be perceptually conscious?’ (p.63).  Although
solving the what-it-is problem does not warrant any solution to the
what-it-is-like problem, what we would expect to achieve is ‘to explain how,
assuming the universe is a purely physical system, the purely physical facts
about us can necessitate phenomenal consciousness’ (p.72); by doing so, ‘we
should have cracked a major component of the mind-body problem.  We should also
have largely vindicated physicalism’ (p.64).  The direction Kirk takes consists
in working out ‘necessary and sufficient conditions for perceptual-phenomenal
consciousness in terms which we can understand pretty well’ (p.74); namely, ‘in
terms of everyday psychology (p.76).  What Kirk calls ‘moderate  realism’
 is the feature of everyday psychology to which he appeals, which requires ‘no
more than that there be some processes going on which constitute the
system’s working out its own response to the situation as it assesses it.’
(p.75).

In Chapter 6, ‘Deciders’, Kirk begins the work of
solving the ‘what-it-is’ problem.  Firstly, he points out that what matters in
regard of differentiating between phenomenally conscious and non-conscious
creatures is not simply a question of appearance.  Obviously, patterns of behavior
play a central role in the working of our everyday psychological concepts.  In
what is purportedly a neutral, non-question-begging, language, that is, in
language that does not presuppose the object to be defined (p.95),
Kirk presents a core cluster of concepts which figure as necessary conditions
for phenomenal consciousness; these are  ‘Decider’ and ‘basic package’.  A
decider possesses the capacities included in the basic package.  Namely:

(i)                
Initiate and control its own behavior on the basis of incoming and
retained information: information that it can use.

(ii)              
Acquire and retain information about its environment.

(iii)            
Interpret information.

(iv)            
Assess its situation.

(v)              
Choose between alternative courses of action on the basis of
retained and incoming information (equivalently, it can decide on a
particular course of action).

(vi)            
Have goals. (89).

Each
one of the above capacities entails and is entailed by the others, although not
all of them must be present to the same degree in a single entity for it to
qualify as phenomenally conscious.  Further, the capacities included in the
basic package must be integrated, otherwise it would be inappropriate to
qualify the behavior exhibited by the system in question as being its own or to
say that the system is in control of its own behavior.  It is not sufficient
that information gets into the system one way or another, but the information
must be for the system.  The unity of the basic package is of a
functional rather than a natural kind, although the working of the inner
processing is of central importance for the appropriateness of the functional
description, which, Kirk points out, distinguishes his approach from the
markedly behaviorist one proposed by Dennett (pp.91-2).  Kirk’s detailed
arguments in defense of the basic package as a necessary condition for
perceptual-phenomenal consciousness (and as both necessary and sufficient for
non-conscious perception) proceeds by analyzing some apparent counter-examples,
among which are Strawson’s ‘Weather Watchers’, Alzheimer’s sufferers, and paralyzed
people.   

Chapter 7, ‘Decision, Control, and Integration’
develops and refines the discourse on deciders and the basic package.  Kirk
shows how the basic package works in practice as we attempt to apply it to ‘indeterminate
cases’ where our intuitions about the presence of conscious perception in a
system fail us.  Among the systems in question, Kirk mentions simple organisms
such as protozoa, bees, etc., human systems such as embryo, fetus and neonate,
split-brain patients, and artificial systems such as robots.  In addition,
creatures popping up from the philosophical world of thought-experiments are
discussed in relation to the basic package idea, such as the artificial giant,
zombies (again), Block machines and Commander Data.  Kirk notices that when we
apply our ordinary concepts of deciding, interpreting and assessing to a
system, if we do so carefully and on reflection, ‘we will use these notions
only in connection with systems with something like the basic package.’
(p.100).

Chapter 8, ‘De-sophisticating the Framework’, answers
a possible objection to the basic package solution proposed in the previous
chapters.  Namely:

If being a decider requires not only the capacity to
acquire and use information, but to do so in a sense which involves the ability
to represent the world and to have concepts, and in some sense to think about
what it is doing, then a decider is a sophisticated system.  …it is difficult
to understand how [that ability] could be possessed without language.  If that
is right, then…either perceptual consciousness does not require the basic
package, or else only creatures with language can be perceptually conscious. 
(pp.119-20).

The
above objection appears to be seriously damaging to Kirk’s view, but Kirk does
counter it by pointing out that its acceptance is based on unwarranted
assumptions regarding what it is involved in representation, belief, and
rationality more generally.  The prevalent views are strongly rationalistic and
Kantian in their inspiration; in discussing such proposals as Evans’, Bermúdez’
and Davidson’s, among others, Kirk points out that they set unnecessarily
strong conditions on concept possession and believing which cannot be met by
non-linguistic creatures.  

In  Chapter 9, ‘Direct Activity’, Kirk points out that
phenomenal-perceptual consciousness must present some further condition that
distinguishes it from cases such as blindsight, subliminal perception, and
ordinary recollection of our perceptual experiences, all of which exclude
phenomenal consciousness and may be carried out by means of the basic package. 
In evolutionary terms, the ‘point [of perceptual consciousness] is to provide
the organism with instantly utilizable information about events in its current
environment regardless of whether it chooses to summon up this or that
particular item.’ (p.141).  Kirk advances the idea of ‘direct activity’ as
supplement to the basic package, so that direct activity plus the basic package
are to be seen as providing a priori individually necessary and jointly
sufficient conditions for perceptual-phenomenal consciousness, and the entity
in possession of the relevant capacities is called ‘decider-plus’.  Direct
activity consists of ‘instantaneity’ and ‘priority’, which to Kirk are best conceptualized
as ‘the same property viewed in two ways’ (p.151).  By ‘instantaneity’, Kirk
means that ‘it pretty well instantaneously endows the system with certain kinds
of capacities’ such as those of describing the kinds of sensory experiences one
undergoes, recognizing the object of previous perceptions in similar
circumstances, being able ‘spontaneously to produce appropriate non-verbal behavior.’ 
The crucial point is that ‘the sense in which these capacities are acquired
instantaneously is that exercising them doesn’t require any special acts of
recall, any guessing, or any popping up.’ (p.151).  With regard to ‘priority’,
Kirk explains that ‘those events have priority if they act on the organism’s
central processes of interpretation, assessment, and decision-making regardless
of the information’s relevance to whatever goals the organism may currently
have.  … information coming in with priority in this sense "forces
itself" on the system’s central processes, since it affects them in a way
it cannot control … regardless of its relevance to its current goals, and
regardless of whether it actually gets used.’  This ‘enables the system to
alter or modify those goals, and may prompt it to do so.’  (p.153).  In the
same context, Kirk tackles a problem which particularly besets higher-order
theories of perception such as those examined in the previous chapter.  The
problem has its sources in viewing conscious perception in terms of the
dichotomy between concept-processing mechanisms and sensory information, which
gives rise to the dilemma of either having to postulate a ‘Cartesian Theatre’
where the two components of conscious perception come together or having to
explain conscious perception in terms of sub-processes which somehow
ensue in a conscious experience.  Kirk’s brilliant solution is the ‘holistic
approach’.  Namely: 

Instead, we need some way of conceiving of perceptual
consciousness according to which it contributes to the working of the system as
a whole. … Give up thinking of non-conceptual representations as distinct
from but ‘poised’ or ‘available to’, and processed by, an Evans-type
concept-exercising and reasoning system.  Instead, conceive of certain
large-scale complex processes as wholes, whose coordinated activity constitutes
the system’s taking-account-of-directly-active-perceptual-information.’
(pp.154.55).

In Chapter 10, ‘Gap?  What Gap?’, Kirk goes on to
reinforce the point made in the previous chapter by showing that the basic
package-plus is not only nomologically sufficient to perceptual-phenomenal
consciousness, but sufficient ‘in such a way that contradiction or other
incoherence would be involved in a decider-plus not being phenomenally
conscious.’ (p.164).  He proposes to do so by an extension of the sole-pictures
argument to an ‘arbitrary decider-plus’, and concludes that ‘if the basic
package-plus satisfies all the purely functional conditions necessary for
perceptual-phenomenal consciousness, then it satisfies all the conditions
necessary for perceptual-phenomenal consciousness.’ (p.166).  The remainder of
the chapter is dedicated to answering a number of foreseeable objections.

Chapter 11, ‘Survival of the Fittest’, presents a
critical overview of alternative approaches to the problem of
perceptual-phenomenal consciousness.  This includes brief discussions of
neuroscientific accounts, e.g., Edelman and Tononi, Damasio, metaphysical
doctrines, i.e., dualism and physicalism, the philosophies of Wittgenstein and
Sartre, Behaviorism, other kinds of functionalist approaches, Dennett’s ‘multiple
drafts’ and ‘Joycean machine’ model, pure representationalism, higher-order
perception and higher-order thought theories.  The shortcomings of these
alternatives are brought to the fore and the soundness and fruitfulness of Kirk’s
approach are highlighted. 

Kirk’s compelling sole-pictures argument does make its
point rather forcefully, although the book ends with the admission that ‘the
gulf between being the observer of a conscious subject and being that subject,
and the associated gulf between viewpoint-neutral and viewpoint-relative
concepts, are both wide.  … [The zombie idea] will stay shimmering there,
poised to dazzle and confuse.’ (p.218).   

Kirk is to be credited for painstakingly reviewing a
considerable number of objections to which his position might have been open;
however, there is still some scope for a couple of general points. 

First of all, one objection in the book points to the
fact that the language describing the basic package-plus is full of ‘implications
of consciousness’ (p.173).   In his answer Kirk explains that his language is
neutral, and that it is possible to assess whether to apply the concept of
direct activity-plus to a creature without knowing in advance whether the
creature is conscious or not.  Further, he adds that some philosophers, notably
Chalmers, do keep separate the psychological from the phenomenal and claim that
possession of the former does not entail possession of the latter; in fact, zombies
are an exemplification of this very same conceptual distinction.  Perhaps, it
might be worth pursuing the objection a little further.  For one thing, the
neutrality of the language constitutes what is in question in the objection, so
it cannot provide an answer.  Secondly, assessing whether a concept may or may
not be applied to a certain entity is not the same as elaborating the content
of the concept itself: in the present case, one may be in doubt whether an
entity is or is not conscious when assessing whether the basic package-plus
applies to that entity, but this does not mean that in elaborating the concept
of basic package-plus as entailing phenomenal consciousness one can possibly do
without using, implicitly or explicitly, what one already knows about
phenomenally conscious creatures, their behavior, and internal processing
according to the ‘moderate realism of everyday psychology’.   Finally, it might
seem odd that Kirk is here happy to build the very same distinction between
psychological capacities and phenomenal consciousness and the very same zombie
idea which his book is designed to relinquish into the presuppositions of the
methodology he adopts in constructing the central concepts of his position.  In
sum: it seems at least doubtful that one can plausibly conceive of a situation
being for a system as a whole, of a system assessing and acting on the
basis of instantly available incoming information which is for it, or of
a concept like ‘decision/decider’, without necessarily (logically) implying
some idea of consciousness.  Apart from its undeniable merits, I fail to see
how the basic package-plus provides a step forward towards clarifying how a
description of what is purely physical entails a description of the
phenomenal.  Furthermore, if I understand Kirk correctly, he is not satisfied
with a nomological truth, that is, a law of nature regarding the relation
between the physical and the psychological: his conditions specifying the basic
package-plus for phenomenal consciousness are meant to be necessary ones in a
logical sense.   In fact, a purely empirical claim would, for example, weaken
the structure of one of Kirk’s crucial arguments, that is, the extended
sole-pictures argument, which heavily relies on the core assumption that the basic
package-plus satisfies ‘all necessary functional conditions’ for
perceptual-phenomenal consciousness (p.165).  But, exclusively on the basis of
empirical evidence, we could hardly know for sure that all conditions
are satisfied at all times?  Kirk’s view on causal relations is in fact
that ‘[they] are contingent; anything which actually has certain effects could
conceivably have failed to have them even if they and all other processes and
events were held constant.’ (p.49).  It is worth noticing that the basic
package-plus is partly meant to justify the strict implication thesis for
minimal physicalism, hence it cannot itself rely on that thesis for support. 
This brings me to my last point, which is concerned with the strict implication
thesis itself.

It is important to bring to light how Kirk is
justified in making his entailment claim from the totality of physical truths
to the totality of psychological truths.  Kirk puts forward various examples to
illustrate his point, a point he makes extensively also in some of his other
publications.  One of his examples is the digital camera one according to which
if ‘my camera produces something identifiable as an image of a duck’, then ‘a
specification of that image in terms of pixels will strictly imply that description’
(p.20).  Here entailment cannot be established by analyzing the meaning of the
terms involved, since it is clear that ‘we could fully understand the
pixel-language specifications without being able to infer from them that they
were of duck-images’.  Consequently, ‘it will not generally be possible to
establish such cases of strict implication by constructing a deductive
argument.’ (p.20).   Kirk’s crucial move in establishing his entailment claim
is that ‘given [that particular image] is of a duck’, could it ‘have
fitted the specification S and have failed to be of a duck’ ?  ‘The
answer to that question is a firm non-empirically based "No".’
(p.21).  In Mind & Body (2003), Kirk explains that the strict
implication thesis is necessary in the same sense in which the following
sentence is necessary: ‘(A) the number of rabbits in our garden at this moment
is greater than 5’.  Although, as a contingent fact, there could be any number
of rabbits in the garden, still it has to be admitted that the ‘number of
rabbits in the garden at this moment’ corresponds to a definite number.  So, if
we suppose that there are 10 rabbits in the garden at this precise moment, then
it is not contingent but necessary that 10 is greater than 5.  Both
illustrations do highlight the idealistic presuppositions of the strict
implication thesis; an odd predicament if no physicalist, according to Kirk,
can avoid committing to such thesis without self-contradiction or other logical
incoherence.  In particular, the entailment claim in the above examples works
only on the implicit assumption that a conscious entity is there either to make
the decision to build a digital camera in such a way that the duck-picture is
specified by a certain pixel configuration, or to ascertain the number of
rabbits in the garden at any given time; without such a conscious presence the
entailment does not hold.  Because by Kirk’s admission the structure of the
strict implication thesis is the same as the structure of the above examples,
it follows that a conscious (God-like) entity is similarly presupposed for the
entailment of Q by P and for the truth that nothing else exists which is not
implied by P.

Not an easy read, this excellent book is highly
reco

Categories: Philosophical, Psychology