Philosophical Zombies (for IEP)

Philosophical Zombies- IEP Amber Ross

1

Amber Ross

Philosophical Zombies

Internet Encyclopedia of Philosophy,

Under Revision- Please do not quote or circulate.

Introduction

The ‘philosophical zombie’ is a thought experiment in philosophy of mind that

tests our intuitions about the relation between our conscious experience and our

physical nature. Philosophical zombies look and act just like ordinary human beings and

by all outward appearance zombies and humans are indistinguishable. Zombies

resemble us in all our physical respects: they act as we act and speak as we speak. Each

one of us has a hypothetical ‘zombie-twin’, a molecule-for-molecule duplicate of

ourselves. The only difference between human beings and their zombie-twins is that

none of the zombie’s physical activity–inner or outer–is accompanied by any conscious

experience. Our zombie-twins’ senses gather all the same information from the world

that ours gather, but–unlike ourselves–they have no experience of it whatsoever. As our

behavioral twins, zombies will certainly seem to make judgments that are about

conscious experience, judgments indistinguishable from our own, although they have no

conscious experience themselves.

Related thought experiments appear in Gottfried Wilhelm Leibniz (1720) and a

recent forerunner is Keith Campbell’s ‘imitation man’ (1970), whereas the contemporary

notion of the philosophical zombie was introduced by Robert Kirk (1974a, 1974b) and

employed to the greatest extent by David Chalmers (1996), who uses them to argue for a

‘naturalistic’ form of property dualism. The zombie thought experiment also brings to


2

light issues regarding the relationship between conceivability (an epistemological

notion) and both metaphysical possibility and logical possibility.

Table of Contents

1. Zombies and the Metaphysics of Mind

2. Support for the Conceivability of Zombies: The Explanatory Gap

3. The Anti-Physicalist Argument from the Conceivability of Zombies a. The Notion of Conceivability in the Zombie Conceivability Argument

4. Physicalist Responses to the Zombie Conceivability Argument a. Taking the Hard Line: Denying the Conceivability of Zombies

i. Do Philosophical Zombies Employ on a Faulty Notion of Consciousness? b. Taking the Soft Line: Grant Conceivability, Deny Possibility

i. The Phenomenal Concept Strategy

5. Questioning the Foundations: Can Conceivability be a Genuinely Rational Notion? a. Yablo’s Criticism of Ideal Conceivability: Any Conceiver could be Mistaken b. Attempting to Idealize Conceivability

i. Potential Problems with Idealizing Conceivability

6. The Anti-Zombie: Has the Philosophical Zombie Met its Match?

7. References and Further Reading

1. Zombies and the Metaphysics of Mind

The philosophical zombie is a curious, yet philosophically powerful creature. It

does seem as though we can imagine at least to some degree the zombie and its zombie

world,. The mere possibility of this kind of zombie may have the power to overthrow


3

what was–at least for the second half of the 20th century–an almost universally accepted

position in philosophy of mind: physicalism.

Physicalism, generally speaking, is the view that everything that exists, at its most

fundamental level, is composed of (or reduces to or supervenes upon) only one kind of

property. Physical properties are the kind of properties that compose everything we

think of as ordinary physical objects. If physicalism is true, then any creature physically

identical to a human being will be identical to a human being in all its respects,

including its conscious experience. The world must also be similar to ours in relevant

respects, since, for example, a creature physically identical to a human being, but in a

world with the wrong properties, may be inert rather than conscious. It is widely

thought that, regardless of one’s position on the metaphysics of mind, zombies are at

least conceivable. According to physicalists, zombies are merely conceivable; one among

many conceivable impossibilities.

In addition to defending the conceivability of philosophical zombies, mind-body

dualists argue for the metaphysical possibility of zombies. The possibility of zombies is

most often used to argue for a position called ‘property dualism’. Property dualists

believe that our minds consist of two fundamentally different types of mental states,

each with its own distinct kind of metaphysical properties. On the one hand, there are

the functional, information processing, problem solving, cognitive kinds of mental

properties and these reduce to certain physical, functional and structural properties that

are more fundamental than those cognitive mental states themselves. But, in contrast to

the physicalist view, property dualists argue that the fundamental features of our world

are not exhausted by physical properties alone. They hold that there must be at least one

other type of property, a special kind of mental property called ‘qualia’ or the


4

‘phenomenal properties’ of our mental states. These qualia or phenomenal properties

are the properties that constitute the character of our conscious experience and,

according to dualists, they are metaphysically different in kind from our cognitive,

information processing and functional mental states. These special mental properties

neither reduce to nor metaphysically supervene upon any physical, functional, structural

properties or other non-phenomenal properties. Phenomenal properties are

metaphysically fundamental. They are the only kind of property that, by hypothesis, you

do not share with your philosophical-zombie-twin. Having or lacking these phenomenal

properties, the properties that constitute the character of our conscious experience, will

determine whether you are conscious or whether your mind is like that of the zombie–

‘all silent and dark within.’

The most popular contemporary form of mind-body dualism is ‘naturalized

property dualism’. This kind of dualism assumes that there are psychophysical laws in

the actual world, which govern the relationship between our cognitive mental states and

our phenomenal mental states such that, as a matter of empirical fact, conscious

experience will be connected to the physical, functional and structural properties of the

world in a law-like way. But property dualists argue that the phenomenal facts in the

actual world–facts about the character of our conscious experience–are not logically or

metaphysically necessarily connected with physical facts. For phenomenal facts to be

necessarily connected– logically or metaphysically–to physical facts, it would have to be

logically impossible (conceptually incoherent or inconceivable) for the character of one’s

conscious experience to change without some change in the physical properties of the

world. There is a logically necessary connection, for example, between biological facts

and physical facts. We could say, following the dualist David Chalmers, who borrows an


5

image from Kripke (1972), that once God fixed all the physical facts of the world, the

biological facts came along for free. But after setting the physical facts in place, God had

‘more work to do to make sure there is a law relating the [physical] facts and the

[phenomenal] facts’ (1996: 38). According to property dualism, biological facts,

chemical facts and facts about our cognitive mental states are logically necessarily

connected to physical facts; whereas phenomenal facts–facts about the character of

conscious experience–stand apart as metaphysically and conceptually independent of

any other kind of fact, including facts about other kinds of mental states.

Physicalism is not merely a mainstream philosophical view but a fundamental

assumption in the scientific study of the mind. If physicalism is false, then our minds

must be more than brains-in-action. Our conscious experience may be nomologically

connected to our cognitive, functional and structural states as a matter of empirical fact,

but if zombies are possible then our world could have turned out differently. If our

world could have turned out differently–physically just the same but phenomenally

different–that is, different where consciousness is concerned, there must be some

additional feature of our world that could have been absent from it, but is in fact

present. Perhaps this feature emerges from our brain activity, but nevertheless it is a

fundamentally different kind of property than our physical properties and it could only

be this additional, non-physical property that accounts for our conscious experience.

The zombie thought-experiment is at root a philosophical puzzle, but since the

beginning of the 21st century the idea of philosophical zombies has become common

currency in cognitive science and many have taken to heart the notion that, if zombies

are conceivable, there really is some aspect of consciousness that stands outside the

scope of what a thoroughly physicalist science of the mind can explain. These scientific


6

explanations do not have room for a non-functional, non-physical and non-structural

fundamental property yet. A basic assumption of empirical science is that the

phenomena of our world can be explained without appealing to anything beyond the

structure and relations between certain physical properties. If zombies are possible then

our current form of scientific explanation, one that can in principle provide a sufficient

explanation for every other fundamental aspect of our world, could not explain

conscious experience. If property dualism is true, the mind stands outside the scope of

empirical investigation. Zombies are so philosophically powerful that their mere

possibility would illustrate that science-as-we-know-it is unable to explain the conscious

mind.

2. Support for the Conceivability of Zombies: The Explanatory Gap

If we assume that facts about the character of our conscious experience are not

logically determined by facts about our physical nature, then it is conceivable for a

subject’s physical properties to be exactly as they are now, while her conscious

experience differs in some essential way. It would be conceivable for you to feel a slight

ache in your left foot that you are not actually feeling now, even though your physical

body–including your brain activity–is in a state that is molecule-for-molecule identical

to its current state. That ache would be a phenomenal property of your experience. If it

is conceivable for there to be a small change, such as this, in your phenomenal

properties, while your physical properties remain the same, then by the same reasoning

large changes should be conceivable as well.

If physical facts do not logically entail phenomenal facts, then it is conceivable for

phenomenal properties–the properties of conscious experience–to change or differ


7

without a change or difference in physical properties. If it is conceivable that such

properties could change, it is also conceivable that phenomenal properties could be

absent from a world altogether, while the physical properties of that world may be

identical to those of the actual world.

Worlds that are physically identical to our own but lack phenomenal properties

are ‘zombie-worlds’. According to the proponents of the zombie conceivability argument

(see below), as long as zombie-worlds are genuinely conceivable, it follows that they are

metaphysically possible. And if zombies are possible, physicalism is false. If zombies are

possible, then it is possible that the phenomenal consciousness we find present in the

actual world–the feature that separates our world from zombie-worlds–could have been

absent, while the physical facts remain just as they are now. This is precisely what the

physicalist denies.

If it is impossible in principle for our current form of scientific explanation to

account for conscious experience, there will be a persistent and unbridgeable

explanatory gap between truths about the physical world and truths about the conscious

mind. The more compelling the conceivability of philosophical zombies is, the harder it

is for physicalism to close the gap through giving a satisfying explanation of conscious

experience in physical terms or even to dismiss the gap as being insignificant. This

insignificance could arise from some feature of our concept of the mind and conscious

experience, regardless of the metaphysical nature of consciousness itself. The dualist

takes the persistence of the explanatory gap–the failure of physicalism to give a

transparent explanation of the connection between facts about consciousness and

physical facts–as conclusive evidence for the falsity of physicalism.


8

3. The Anti-Physicalist Argument from the Conceivability of Zombies

The anti-physicalist argument from the conceivability of zombies may be the

most compelling argument for mind-body dualism since the beginning of this century.

In its simplest form, it can be stated thus:

1. Zombies are conceivable.

2. Whatever is conceivable is possible.

3. Zombies are possible.

4. Physicalism is false.

Following David Chalmers (2010: 142), the argument can also be laid out more

technically as follows:

1. P&~Q is conceivable.

2. If P&~Q is conceivable, P&~Q is metaphysically possible.

3. If P&~Q is metaphysically possible, physicalism is false.

4. Physicalism is false.

As the conceivability argument is laid out here, P is ‘the conjunction of all

microphysical truths about the universe’ and Q is ‘an arbitrary phenomenal truth’ (ibid:

142). A ‘phenomenal truth’ is a particular fact about conscious experience; it can be a

fact about a particular individual’s conscious experience or about consciousness in

general.

Premise (1) of the conceivability argument states that you can conceive of a

world, in which the phenomenal facts–facts about the character of conscious

experience–are not necessitated by the physical facts. Most physicalists accept premise

(1), agreeing that we can conceive of our world’s physical facts being just as they are


9

now, whereas the phenomenal facts differ in some way. Premise (2) of the conceivability

argument is the conceivability-possibility principle. According to this principle,

whatever is genuinely conceivable is possible. If zombies are genuinely conceivable, then

zombies are possible and if zombies are possible, physicalism must be false.

a. The Notion of Conceivability in the Zombie Conceivability Argument

Premise (2) of the conceivability argument is the most controversial instance of

the conceivability-possibility principle, claiming that any state of affairs that is

conceivable is also possible. Spelled out in these simple terms, it seems clearly false,

since myriad instances exist, in which conceivable states of affairs are metaphysically

impossible. For the most part, conceivability is a good, yet fallible guide to possibility.

But why should conceivability be a perfect guide to possibility in the zombie case?

Chalmers’ defense of the conceivability-possibility principle (and the zombie

conceivability argument) involves dividing the notion of conceivability into several

varieties. Positing careful distinctions within the concept of conceivability itself,

Chalmers reformulates the principle in a way that avoids the most obvious objections.

Rather than the overly simplistic and clearly false claim that could be read as ‘whatever

satisfies any criteria for conceivability is possible,’ the conceivability-possibility

principle is the more refined claim that whatever is ideally conceivable is possible or,

more precisely:

Ideal primary positive conceivability entails primary possibility.

The ideal variety of conceivability above is contrasted with prima facie

conceivability; the positive variety is contrasted with negative conceivability and


10

primary conceivability with secondary. The distinction between prima facie and ideal

conceivability is intended to separate those scenarios that merely seem conceivable at

first from those, which are actually conceivable. The notion of prima facie

conceivability is fairly straightforward: S is prima facie conceivable for a subject, when S

is conceivable for that subject on first appearance (Chalmers, 2010: 143).

The dimension of positive conceivability and negative conceivability is concerned

with creation and elimination. A scenario is positively conceivable, when it can be

‘created in (modal) imagination’, while ‘negative notions of conceivability hold that S is

conceivable when S is not ruled out’ (ibid: 143). A scenario is prima facie negatively

conceivable, when initial consideration of the scenario reveals no obvious conceptual

contraction within that scenario or hypothesis. A scenario is ideally negatively

conceivable only if ‘the hypothesis expressed by S cannot be ruled out a priori even on

ideal rational reflection’ (2010: 143) and will fail to be ideally negatively conceivable,

when the prima facie conceivability of S is ‘undermined by further reflection showing

that the tests that are criterial for conceivability are not in fact passed’ (ibid: 144).

The notion of forming a ‘positive conception of a situation’ is somewhat more

complicated, since it involves imagination, but not ‘imagination’ in an ordinary sense. S

is ‘positively conceivable,’ when a subject can coherently ‘modally imagine’ a situation

that verifies S. Hence, ‘when one can coherently imagine a situation in which S is the

case’ (2010: 144), which is ‘to in some sense imagine a specific configuration of objects

and properties’ (ibid: 145) and ‘fill in arbitrary details in the imagined situation such

that no contradiction reveals itself’ (ibid).

As the word suggests, imagination is often imagistic or in some way sensorial and

genuinely conceivable scenarios exist that cannot be ‘imagined’ in these ways. Descartes’


11

example was ‘imagining’ a chiliagon (Meditation 6) in visual terms: a polygon with 1000

sides indistinguishable from a circle, even though the concepts ‘chiliagon’ and ‘circle’ are

clearly distinct, while both are conceptually coherent. Modal imagination cannot be

perceptual or sensory, as neither can be imagined from the first-personal subjective

perspective. To ‘modally imagine a scenario’ is not to imagine viewing some scene, but

rather to mentally arrange a set of objects and properties in a certain way that would

either verify or fail to verify a proposition that represents a possible world or potential

scenario.

Insofar as we can conceive of a difference between zombies and conscious human

beings, the scenario cannot be positively conceived by comparing perceptions of the two

creatures in imagination, since a zombie world is identical to a human world in every

observable respect. Neither we nor the zombies could observe that zombies lack

phenomenal states. Zombies themselves cannot observe that they are in any way

different from humans. Zombies, by definition, do not observe any difference between

themselves and their human twins, since one cannot observe a subjective state that is

simply absent, hence the zombie is unable to observe a difference.

Zombie-minds will be prima facie positively conceivable, when a subject believes

that all the details of the zombie scenario can be filled in without revealing a

contradiction or when a subject believes she can ‘imagine a situation with certain

important features specified, notes that a situation of this kind appears to verify S, and

judges that the remaining details are not crucial’ (Chalmers, 2002a: 153).

Property dualists and some physicalists claim that zombies are ideally

conceivable, because the zombie-scenario is conceptually coherent and displays no

contradictions concealed within it. The proponents of the zombie conceivability


12

argument believe that, if there is no latent conceptual incoherence in the zombie

scenario, we can infer the possibility of zombies from their conceivability alone. And, if

zombies are possible, physicalism must be false, since, according to physicalism, any

creature, in any possible world that is molecule-for-molecule physically identical to a

human will have human mental properties, both cognitive and phenomenal. This means

that the mere possibility of a zombie-world makes physicalism an untenable position:

physicalism would be false in our world, as well as in all possible worlds. If zombies are

possible, physicalism is necessarily false.

4. Physicalist Responses to the Zombie Conceivability Argument

The common consensus among philosophers is that the zombie-scenario is

conceptually coherent. Property dualists, as well as the majority of physicalists, hold

that they can ‘coherently modally imagine’ zombies and zombie-worlds without

difficulty by ‘fill[ing] in arbitrary details in the imagined [zombie] situation such that no

contradiction reveals itself’ (Chalmers, 2010: 145). Most philosophers would agree that

their positive conception of the zombie-scenario withstands the test for ideal

conceivability, a point crucial for accepting premise (1) of the argument, since no

‘variety’ short of ideal conceivability qualifies as genuine conceivability, as the property

dualist intends it to be understood in the zombie conceivability argument and

interpreted in the conceivability-possibility principle. Very few philosophers argue that

philosophical zombies are not ideally conceivable, but the most compelling arguments

against the conceivability of zombies can be found in Robert Kirk (1974, 2008), Daniel

Dennett (1995, 1999, 2005, 2013), Sydney Shoemaker (1975), Nigel Thomas (1998),


13

David Braddon-Mitchell (2003), Allin Botterell (2001), Keith Frankish (2007), Richard

Brown (2010) and Marcus (2004).

a. Taking the Hard Line: Denying the Conceivability of Zombies

A subject may make the prima facie judgment that she can fill in the details of

scenario S, even though the world S described is in fact conceptually incoherent (for

example, inconceivable objects, such as Roger Penrose’s ‘impossible triangle’ or many

drawings by M.C. Escher). [Image of the “impossible triangle” approximately here.]

Most philosophers rejecting the conceivability of the zombie-scenario argue against its

positive conceivability, through attempting to show that zombies are only prima facie

positively conceivable. According to these philosophers, subjects who believe they are

conceiving of a zombie are actually making a mistake of some sort in their imagination:

possibly by ‘filling in’ the scenario with improper details or by failing to notice ‘holes’ in

the details of their imagined creature. (For compelling arguments of this sort, see Kirk

(2008), Dennett on zombies (1991, 2005) and on the knowledge argument (1991), as

well as Marcus (2004) and Ross Inconceivable Minds (2013) chapter 4.)

Popular attempts to explain away the zombie’s apparent conceivability appeal to

empirical facts about the psychology of imagination, arguing that when we take

ourselves to be imagining a philosophical zombie, we imagine, or refrain from

imagining, different aspects of the zombie-scenario. (See Hill (1997), Hill and

McLaughlin (1999), Block and Stalnaker (1999) and Nagel (1974, fn11), among others.

For an overview of this type of response to the conceivability argument, see Tyler


14

Doggett & Daniel Stoljar (2010).) It is worth quoting Nagel’s account of imagination

here at some length:

We may imagine something by representing it to ourselves either

perceptually [or] sympathetically. … To imagine something perceptually,

we put ourselves in a conscious state resembling the state we would be in if

we perceived it. To imagine something sympathetically, we put ourselves

in a conscious state resembling the thing itself. …When we try to imagine a

mental state occurring without its associated brain state, we first

sympathetically imagine the occurrence of the mental state; that is, we put

ourselves in a state that resembles it mentally. At the same time, we

attempt to perceptually imagine the non-occurrence of the associated

physical state, by putting ourselves into a state unconnected with the first:

one resembling that which we would be in if we perceived the non-

occurrence of the physical state. Where the imagination of mental features

is sympathetic, it appears to us that we can imagine any experience

occurring without its associated brain state, and vice versa. The relation

between them will appear contingent even if it is necessary, because of the

independence of the disparate types of imagination (1979: 175, fn. 11).

It is no small feat to show that the creature we ordinarily imagine, when we

believe we are conceiving of a zombie, does not have all the properties of an actual

philosophical zombie, but is merely a superficially similar creature, the conceivability of

which is irrelevant to the truth or falsity of physicalism. Nor is it a small task to give a

compelling explanation of our apparent ability to imagine philosophical zombies via an


15

account of the psychology of imagination. But the problem with these and similar

attempts to identify the ways, in which we might fail to properly positively conceive of a

zombie, is that they are not actually objections to the zombie conceivability argument.

The proponent of the zombie conceivability argument claims that there is a

genuine sense of ‘modal imagination,’ in which one can ‘imagine an arrangement of

objects and properties’ that form a scenario, in addition to perceptual and sensory or

first-person-subjective imagination (imagining what S is like from the perspective of a

subject in scenario S). If this is correct, then it is irrelevant whether our ordinary

methods of imagination will produce a positive conception of a zombie.

Eric Marcus (2004) argues against the conceivability of zombies by claiming that,

when we take ourselves to conceive of zombies, we are not actually forming a positive

conception of a non-phenomenally-conscious creature physically identical to ourselves.

Instead, we are positively and perceptually imagining something physically identical to a

human being and then refraining from imagining its conscious experience. Since

‘refraining from imagining’ does not suffice for forming a positive conception of a

scenario and the only way to conceive of an absence of subjective experience is to refrain

from imagining subjective experience, he claims we cannot positively conceive of

zombies.

In order to positively conceive of the absence of a subjective experience, we would

need to ‘sympathetically imagine’ the zombie’s experience (to use Nagel’s terminology)

and to ‘put ourselves in a state that resembles it mentally’. And exactly this is

impossible, since there is no state that resembles the zombie’s subjective state mentally,

given that no subjective state for it to mentally resemble exists. As Marcus says:


16

The problem for this thought-experiment is not that there is something

imaginable, only we can’t quite conjure it up. The problem is that there is

nothing to be imagined. To ‘imagine’ creatures that are objectively

identical to us with all subjectivity removed is neither an act of third-

person imagining, nor an act of first-person imagining. No, to ‘imagine’ a

zombie is not really to imagine at all (2004: 483).

The problem with Marcus’ argument against the conceivability of zombies is that

neither ‘sympathetic imagination’ nor ‘first-person imagination’ are modal imagination

and this is the kind of imagination at issue in the zombie conceivability argument.

According to Marcus, ‘it is crucial for the argument in favor of the possibility of zombies

that the conceivability of zombies be in part a matter of first-person imagining’ (ibid:

483), whereas a proponent of the zombie conceivability argument would deny this. In

order to modally imagining a scenario, a subject must be able to imagine particular

properties themselves having certain properties. To modally imagine a zombie world, a

subject must imagine that the properties that constitute conscious experience are or at

least could be such that a creature could share all our physical properties but lack our

conscious experience. And conceiving of mental properies as bearing, or not bearing,

certain relations to other properties (for example physical properties) does not require

any first-person imagination, as it only requires conceptual analysis.

Marcus’ interpretation of Chalmers’ notion of ‘modal imagining’ highlights the

notion’s deceptive simplicity:

[A]ccording to Chalmers, positive imagining need not be perceptual. I can

imagine what is ‘beyond the scale of perception: for example, molecules of


17

H2O or Germany winning the Second World War’ [Chalmers 2002, 151]. In

such cases, there are no mental images, instead ‘we have an intuition of (or

as of) a world in which S, or at least of (or as of) a situation in which S,

where a situation is (roughly) a configuration of objects and properties

within a world’ [Ibid., 151]. In the case of modal imagining, then, the

mediating object is an intuition that represents the possibility in a way

analogous to that of the image in the case of perceptual imagining (Marcus

2004, 479).

The footnote following Marcus’ quote states that ‘I don’t see exactly how an

intuition could be an object analogous to an image; but I won’t digress by disputing that

here’ (ibid: 479). In failing to see how an intuition is analogous to an image, Marcus

reveals something important about modal imagination: images and intuitions may both

be a sort of mental representation, but beyond this there is no illuminating analogy

between the two. The terms intuition, image and imagination all serve to make the

extraordinary task of modally imagining a scenario look like the commonplace act of

imagining a character in a scene, through attempting to precisify the content of certain

concepts (such as consciousness, conscious experience, physicality and so forth) and

judge whether certain properties could bear certain relations to other properties.

Regardless of whether Marcus’ argument addresses the zombie conceivability argument

as Chalmers poses it, Marcus’ main point is still significant: to positively conceive of the

zombie-scenario is in fact harder than it seems.


18

i. Do Philosophical Zombies Employ on a Faulty Notion of Consciousness?

Philosophical zombies are defined as void of conscious experience. But different

interpretations of the meaning of consciousness and conscious experience will generate

different conclusions about the conceivability of zombies. These different

interpretations not only separate physicalists from dualists, but account for the division

between different physicalist positions as well. All dualists and the majority of

physicalists interpret having a conscious experience as something like ‘being in a state

that has a certain phenomenal feel’ or ‘there being something it’s like to be in this

particular state.’

A minority position within physicalism called ‘thin materialism’, according to

Graham and Horgan’s (2000) delineation of physicalist positions, takes the notion of

‘seeming’ to be essential to the concept of conscious experience. These physicalists

believe for a subject to have a conscious experience means to be in a state that seems to

the subject to have a certain character, quality or feel to it. In the descriptions of zombie-

worlds, zombies claim to be conscious, they write about consciousness, complain when

they feel pain in their lower back and so forth. All parties to the zombie-debate agree

that there is a philosophically significant sense, in which it seems to the zombie that she

has conscious experience.

If seeming plays an essential role in the concept of conscious experience itself,

the definition of a philosophical zombie will be incompatible with the concept of

conscious experience, since it will certainly seem to the zombie that she is conscious and

has conscious experiences. If it seems to the zombie that she is having a conscious

experience, then, according to this interpretation of the meaning of consciousness, the

zombie is having a conscious experience. Thus the notion of a philosophical zombie


19

contradicts the thin physicalist’s concept of consciousness, since if all that is required for

consciousness to exist is for it to seem to the subject that she is conscious then zombies

are conceptually incoherent and inconceivable. (It should be noted that such accounts

are typically categorized as ‘anti-realist’ accounts of conscious experience, even though

proponents of thin materialism would disagree with this characterization.)

Thin physicalists deny the primary assumption, on which the conceivability

argument is based, namely the claim that zombies are genuinely conceivable. The

apparent conceivability of zombies may stem from a conceptual confusion about the

nature of consciousness, the latent disagreements over what it means for a scenario to

be conceivable or from a lack of imagination. The thin physicalist holds that in some way

or another the ease, with which one can describe a philosophical zombie is obscuring the

actual complexity of the zombie-scenario and this complexity in turn hides the zombie’s

conceptual incoherence. These physicalists think that zombie-stories induce something

resembling a conceptual hallucination, in which what you seem to be conceiving is not

actually what you conceive. According to thin physicalism, such conceptual

hallucinations might be corrected through the discovery of a yet unknown, but in

principle possible, a priori connection between phenomenal facts about conscious

experience and physical facts. Though prima facie philosophical zombies seem to make

sense, the thin physicalist holds that they are not conceptually coherent. This seems to

be Dennett’s (1995) take on the notion of zombies. In The Unimagined

Preposterousness of Zombies, he compares the concept of consciousness to that of

health:

Supposing that by an act of stipulative imagination you can remove

consciousness while leaving all cognitive systems intact — a quite standard


20

but entirely bogus feat of imagination — is like supposing that by an act of

stipulative imagination, you can remove health while leaving all bodily

functions and powers intact. … Health isn't that sort of thing, and neither is

consciousness.

b. Taking the Soft Line: Grant Conceivability, Deny Possibility

Dualists and the majority of physicalists take the meaning of consciousness to be

something like ‘having phenomenal feel’. This group of physicalists is called ‘thick

materialists’, according to Graham and Horgan (2000) or ‘type-b materialists’,

according to Chalmers (2002b) and endorses premise (1) of the conceivability

argument, granting that philosophical zombies are conceptually coherent. According to

type-b physicalists, there is no conceptual contradiction within the zombie-scenario, in

which it seems to a subject that she is having a conscious experience with a certain

phenomenal content, despite lacking conscious experience whatsoever. If having a

conscious experience means ‘being in a state with a certain phenomenal feel’ and

seeming to be in a state with a phenomenal feel is not conceptually sufficient for having

a conscious experience veridical or otherwise. Then a subject may believe that she is in a

state that has a phenomenal feel, but holding this belief will be conceptually consistent

with the subject not actually being conscious. Despite the subject’s potential to hold a

false belief about her mental state, the scenario itself is still conceptually consistent.

Type-b physicalists insist that zombies are merely conceivable, but that it does

not follow that zombies are metaphysically possible, targeting premise (2) of the

conceivability argument. Even though facts about possibility and conceivability are


21

related, they are separate matters. Type-b physicalists reject the claim that the

conceivability of scenario S straightforwardly entails that S is possible, stipulating

conceivability as a guide to possibility, but nonetheless a defeasible one.

Premise (2) of the zombie conceivability argument, ‘whatever is (ideally

positively) conceivable, is possible,’ is the most widely disputed premise. The plausibility

of premise (2) depends on the inconceivability of metaphysical impossibilities. Most

type-b physicalists hold that premise (2) is undermined by Kripke’s notion of a

posteriori necessity.

According to type-b physicalists, we can conceive of zombies, not because they

are possible, but because the structure of our cognitive economy isolates our

phenomenal concepts (for example, the concept of what it is like to see red) from our

physical, functional, cognitive, and otherwise non-phenomenal concepts (for example,

the physical concept brain event b or the cognitive concepts focusing attention and

identifying a percept). Our ability to conceive of zombies reflects only the nature of our

concepts, just as our ability to conceive of water as distinct from H2O does not entail the

metaphysical possibility that water is not H2O. Therefore our ability to conceive of

zombies says nothing about the metaphysical possibility of zombies. According to type-b

physicalists, the conceivability of the zombie-scenario does not itself show that

physicalism is false in our world.

The most widely accepted view among physicalists is that physicalism is

contingently true of our world. They hold that the mental facts of our world supervene

upon the physical facts, but maintain that physicalism is not the only conceptually

coherent theory of the relationship between mind and world. Most physicalists admit

that it is conceivable that physicalism might have been false of our world and a smaller


22

set of physicalists treat it as a necessary or conceptual truth (for example, Davidson

(1970)).

Whether the psychophysical conditional (‘P →Q’, where P is the set of all physical

facts in a world and Q is the set of all phenomenal facts in that world) is intended to be

taken as a necessary or as a contingent truth of our world, the fact that certain mental

states supervene upon or are identical to certain physical brain states certainly appears

to be contingent. In either case, those who interpret physicalism as either a necessary or

contingent truth widely appeal to Kripke’s (1972) introduction of ‘a posteriori necessity’

to defend physicalism against the zombie conceivability argument.

To borrow Frege’s example: the terms ‘Morning Star’ and ‘Evening Star’ both

refer to the planet Venus, but both pick out their referent through different modes of

presentation or descriptions. Morning Star picks out its referent by means of some

description such as ‘the last heavenly body we see in the morning’, while Evening Star

picks out its referent by the description ‘the first heavenly body we see in the evening’.

The difference between the way these two terms pick out their shared referent indicates

a difference in the meaning of these two terms; that is, a difference between the

concepts’ content.

Since identity is a necessary relation, the Morning Star is necessarily the Evening

Star. Still, the identity of the Morning Star with the Evening Star may appear

contingent, because this identity is not an a priori truth. The fact that the Morning Star

is the Evening Star is not true in virtue of the content of the constituent concepts alone.

Since their identity is an empirical fact rather than a priori truth, it will seem as though

conditions could have turned out differently. It seems as though the Morning Star might

have turned out to be some heavenly body other than Venus. But an identity discovered


23

a posteriori is still an identity and since the relation is necessary it holds in every

possible world.

A posteriori necessary truths seem contingent, because we must first gain

empirical knowledge, in order to see their truth. A person who does not know that the

Morning Star and Evening Star are Venus, who does not know that ‘the last heavenly

body we see in the morning’ is the very same thing as ‘the first heavenly body we see in

the evening’, can conceive of the Morning Star and the Evening Star being distinct

objects. It is consistent with everything this person knows that the Morning Star and

Evening Star may be two separate entities.

Whether or not one is actually aware that these ‘stars’ are the planet Venus, there

is still a sense in which it is conceivable for the Morning Star to be distinct from the

Evening Star. Within the modality of possible worlds, we can express the divergence of

conceivability from possibility by postulating two independent sets of worlds: the first

containing ‘conceptually possible worlds’ (for example, worlds that are merely

conceivable), while the second set contains only ‘metaphysically possible worlds’, the

kind of ‘worlds’ that matter to the truth or falsity of physicalism. Because the identity

relation is metaphysically necessary, the Morning Star will be identical with the

Evening Star in every metaphysically possible world. But because their identity is not a

priori, it is not a conceptually necessary truth and as such there will be conceptually

possible worlds (conceivable scenarios or epistemically possible worlds), in which the

Morning Star and the Evening Star are distinct entities.

Most physicalists are a posteriori physicalists in the sense that they claim we can

conceive of brain events occurring in the absence of phenomenal events, even though

the brain event and phenomenal event are identical. They furthermore claim that we can


24

conceive of zombie-scenarios even though they are impossible, for the same reason that

we can conceive of the Morning Star and Evening Star being distinct entities, even

though their identity is metaphysically necessary. Just as the concepts Morning Star

and Evening Star have different content, but both refer to planet Venus, the concepts

we use to denote our conscious experiences (for example, ‘what it is like to see red’) and

the concepts we use to refer to brain or cognitive events (for example, ‘identifying the

red percept’) have different content, but refer to the same properties. Appealing to two

independent sets of possible worlds allows type-b physicalists to claim that the identity

of the conscious experience ‘seeing-red’ with the brain or cognitive event ‘identifying-

the-red-percept’ may be an empirical fact, only known a posteriori, yet nonetheless

metaphysically necessary.

We might say that the type-b physicalist’s modal universe contains two types of

worlds: the merely conceivable and the genuinely possible. The property dualist cannot

accept this distinction between two sets of worlds, because, in order to defend the

zombie conceivability argument, he must maintain his claim that conceivability entails

possibility by showing how every conceivable world can be a metaphysically possible

world. If the type-b physicalist is correct and some conceivable worlds are not

metaphysically possible, the conceivability of zombies may have no metaphysical

implication.

i. The Phenomenal Concept Strategy

Physicalists hold that the conceivability of zombies and the existence of the

explanatory gap only provide prima facie support for the claim that there is an


25

ontological difference between physical and phenomenal mental properties and that the

evidence thereof is actually defeasible. One might ‘explain away’ this prima facie

evidence against physicalism by showing that the conceivability of zombies and

existence of the explanatory gap are merely benign consequences of the way our mind

organizes and draws connections between our concepts of consciousness and

physicality. Similar to the appeal to ‘a posteriori necessity,’ this sort of alternative

explanation would prohibit the dualist from making a direct inference from the

conceivability of zombies and the existence of the explanatory gap to their metaphysical

possibility and the falsity of physicalism.

For the possibility of a scenario S to be entailed by the conceivability of S, further

criteria must be satisfied beyond the ideal positive conceivability of S. In particular,

there can be no viable alternative explanations of S’s conceivability. There may be

‘blockers’, such as multiple scenarios, which are all ideally positively conceivable

explaining the conceivability of zombies without appealing to an additional class of

ontologically properties, thus ‘blocking’ the inference from zombie-conceivability to

their possibility.

Dualists and type-b physicalists agree that phenomenal and non-phenomenal

concepts are isolated from one another. It follows that facts couched in phenomenal

terms cannot be inferred from facts couched in entirely physical terms and that

statements expressed in phenomenal terms cannot be translated into statements

expressed in entirely physical or functional (non-phenomenal) terms without some loss

of meaning. Some physicalists, applying what has come to be called “The Phenomenal

Concept Strategy” (Loar, 1990), claim that this conceptual isolation explains why Mary

the Color Scientist cannot know what it is like to see red simply in virtue of knowing all


26

the physical facts about her world. Physical facts are facts expressed with physical

concepts and while physical concepts and phenomenal concepts may have the same

referent, they do not share their meaning. Therefore, no knowledge of facts expressed

with physical concepts will close the explanatory gap. Physicalists who support the

phenomenal concept strategy believe that the isolation of phenomenal concepts from

physical concepts can account for the explanatory gap between physical facts and facts

about conscious experience.

There is no single definitive version of the phenomenal concept strategy (see

Stoljar (2005)), but all forms have two basic steps in common. The first is to claim that

the psychophysical conditional (‘P →Q’, where P is the set of all physical and Q is the set

of all phenomenal facts in that world) is a necessary a posteriori truth and grant that it

appears to be contingent. The second step is to acknowledge the conceptual isolation of

phenomenal concepts from physical concepts and argue that conceptual isolation

accounts for both (1) the absence of any a priori entailment of phenomenal facts from

physical facts and (2) the apparent contingency of the psychophysical conditional.

If the phenomenal concept strategy works, that is if the source of the explanatory

gap is simply the isolation of physical concepts from phenomenal concepts, then the

existence of irreducible, non-physical and phenomenal properties is not entailed by the

explanatory gap or by the conceivability of zombies simpliciter. If this is correct, then

the nature of the relationship between our physical and phenomenal concepts will

deflate the metaphysical significance of the gap. This solution to the zombie problem is

neither intended to close the explanatory gap nor should it. If our physical concepts and

phenomenal concepts are isolated from each other, we should expect to see an epistemic

gap, even if the phenomenal states, to which those concepts refer, are identical to, or


27

supervene upon, physical states. If the phenomenal concept strategy succeeds here,

neither the conceivability of zombies nor the explanatory gap more generally would

provide evidence for the claim that property dualism is true of our world.

5. Questioning the Foundations: Can Conceivability be a Genuinely Rational

Notion?

Marcus’ argument against the conceivability of zombies has shown that, at the

very least, positively conceiving of a zombie is a much more demanding task than it first

appears. But, in principle, whether or not the zombie is ideally positively conceivable

does not depend on our fallible powers of imagination. In practice, an ordinary

epistemic agent’s conceivability judgments are ‘tied to [that] subject’s contingent

cognitive limitations’ (Chalmers, 2010: 143), but genuine conceivability is not dictated

by our limited cognitive capacities. ‘Ideal conceivability’, the ‘variety’ of conceivability at

work in the conceivability-possibility principle, ‘abstracts away from those limitations’

(ibid: 143). In ‘modally imagining’ a scenario, we are attempting to determine whether

an ideal conceiver would judge that certain properties could bear certain relations to

other properties.

A fundamental assumption of the zombie conceivability argument is that the

concept of ‘ideal conceivability’ is coherent, without which a rational notion of

conceivability would fail the conceivability-possibility principle and there would be no

link between epistemic and metaphysical facts. Chalmers writes that

if we are looking for a notion of conceivability such that conceivability

tracks possibility perfectly, we must focus on ideal conceivability. In this


28

sense conceivability is not a merely psychological notion; it is a rational

notion, in much the same way that a priority and rational entailment are

rational notions. If there is to be a plausible epistemic/modal bridge, it will

be a bridge between the rational and modal domains (2002a: 160

emphasis original).

Facts about epistemic agents are irrelevant to a priority and entailment. The a

priori truth of a proposition is neither relative to the subject evaluating the proposition

nor is entailment relative to a subject’s beliefs about the relation between propositions.

Though conceivability is (in some cases) relative to a speaker or thinker, for there to be

any instance, in which conceivability entails metaphysical possibility, there must be

some variety of conceivability, according to which the conceivability of S is not relative

to an epistemic agent’s beliefs. For the conceivability-possibility principle to hold, there

must be some ‘variety’ of conceivability that is a rational, rather than a psychological or

subjective notion.

Carving conceivability into several distinct notions gives the proponent of the

zombie conceivability argument the opportunity to explain and explain away the

appearance of subjectivity in conceivability-judgments. If there are in fact several types

of conceivability, it might be plausible that only certain varieties are subject-dependent.

The truth of some conceivability-judgments are clearly relative to a conceiver (for

example, prima and secondi facie conceivability-judgments), but the more finely we

discriminate between types of conceivability, the more plausible it will seems that there

could be a conceiver-independent variety of conceivability, according to which the

conceivability of S can be knowable a priori. This conceiver-independent variety of


29

conceivability would allow us to judge whether scenario S is conceivable simply in virtue

of the concepts’ content that compose the scenario.

a. Yablo’s Criticism of Ideal Conceivability: Any Conceiver could be

Mistaken

Positive conceivability involves imagining a certain configuration of objects and

properties and to then decide whether that modal imagining verifies a particular

scenario, by making a considered judgment whether all the details of that scenario could

be ‘filled in’ without any contradiction revealing itself. But any ordinary epistemic agent

can be ignorant of some fact relevant to the scenario she believes she can conceive.

Yablo (2003) objects to the immediate inference from conceivability to metaphysical

possibility across the board:

If ignorance of an individual's essential properties can generate modal

error, why not ignorance of a property's essential properties? Imagine that

my grasp of a property S fails to reflect the fact that it is essentially

uninstantiable (S might be the property of being sodium-free salt).

Nothing to prevent me, then, from conceiving it as possible that Ss should

exist: a de dicto conceivability error rather than a de re one. Likewise the

de dicto impossibility that some Qs are Rs will be conceivable, if my

understanding of Q omits its essential property of having no Rs in its

extension. Probably there is no proposition for which a worry like this

cannot be raised (Yablo, 2003:17).


30

Nothing is to prevent a subject from believing she can ‘modally imagine’ a

scenario, if she is ignorant of some essential property of a property or concept involved.

No epistemic agent can prevent herself from making such a mistake and no community

of likewise-ignorant epistemic agents could correct that mistake. Chalmers

acknowledges this problem with the rational notion of conceivability in certain cases.

Indeed, it is arguable that one can modally imagine S when S involves an a

priori contradiction. An example may be a case in which one imagines a

geometric object with contradictory properties. In cases like this, one

imagines a situation in something less than full detail (2002a).

The overarching problem is that a subject’s intrinsic cognitive limitations will

always prevent her from knowing with certainty whether she has in fact ideally

positively conceived of a scenario that verifies scenario S or whether she merely believes

that she has positively conceived of S and that her conception of S passes the test for

ideal conceivability. In the same vein, a subject’s intrinsic cognitive limitations will

prevent her from knowing whether she has imagined a situation in full detail or she has

left pertinent details unspecified (allowing a priori contradictions to go unnoticed). As

Yablo (2003) warns, there is probably no scenario for which such worries cannot be

raised.

This is why positive conceivability must be idealized. But the question remains as

to how ideal conceivability ought to be explained.


31

b. Attempting to Idealize Conceivability

So far the notion of ‘ideal conceivability’ has been left under-determined, but

defending its status as a rational notion (like a prioricity and entailment) depends on

there being some method of idealizing conceivability that ‘abstracts away’ from the

subjective judgments of limited and fallible epistemic agents. At the same time, it would

be of no use to idealize conceivability by appealing to an omniscient conceiver, since, by

hypothesis, whatever an omniscient agent believes is true. Introducing an omniscient

conceiver runs the risk of reducing ideal conceivability to a triviality; the claim that an

omniscient conceiver would find scenario S conceivable is tantamount to simply

stipulating the coherence of the concepts that compose the scenario, granting it ideal

conceivability by fiat rather than by argument.

There are several methods, from which one might explicate the notion of ideal

conceivability. Chalmers (2002a) considers two: the first is to appeal to an ideal, though

not omniscient conceiver; and the second to an ideally reasonable epistemic agent.

One could try to define ideal conceivability in terms of the capacities of an

ideal reasoner — a reasoner free of all contingent cognitive limitations.

Using this notion, we could say that S is ideally conceivable if an ideal

reasoner would find it to pass the relevant tests (if an ideal reasoner could

not rule out the hypothesis expressed by S a priori, for example) (2002a:

148, parenthetical remark original).

Chalmers’ explicit worry with this first account is that an ideal reasoner may be

impossible and even incoherent. He reasons that ‘it may be that for every possible

reasoner, there is a more sophisticated possible reasoner’ (ibid: 148), rejecting the


32

notion of an ideal reasoner and pursuing the second elucidation of ideal conceiving. This

is the one Chalmers eventually endorses, which does not appeal to the judgment of an

epistemic agent, but rather to the notions of justification and defeasibility, properties

that appear to be reasoner-independent.

[O]ne can dispense with the notion of an ideal reasoner, and simply invoke

the notion of undefeatability by better reasoning. Given this notion, we can

say that S is ideally conceivable when there is a possible subject for whom

S is prima facie conceivable, with justification that is undefeatable by

better reasoning (2002: 148).

Employing the notion of ‘undefeatability by better reasoning’ may avoid the

regress of ‘more sophisticated reasoners’, since ideal conceivability would rely on a

reasoner only insofar as there must be some epistemic agent who makes the judgment

that S is prima facie conceivable.

The idea is that when prima facie conceivability falls short of ideal

conceivability, then the claim that the relevant tests are passed will either

be unjustified, or the justification will be defeatable by further reasoning.

For ideal conceivability, one needs justification that cannot be rationally

defeated (ibid).

i. Potential Problems with Idealizing Conceivability

There are several potential problems with Chalmers’ rejection of the ideal

reasoner explanation of ideal conceivability. The first is that the account he endorses

instead–invoking the notions of ‘undefeatability’ and ‘justification’–may be open to a


33

regress analogous to the problem, for which the ideal reasoner option was rejected in

the first place. The strength of justification for a particular belief (whether it is

‘defeatable by further reasoning’ or ‘cannot be rationally defeated’) is determined by

several factors, one of which will be the means by which the subject acquires the belief.

This may open the possibility of a regress of ‘better means of belief acquisition’,

analogous to the regress of ‘the more sophisticated reasoner’. For any given means by

which a subject acquires a belief, it is possible (or at least conceivable) for there to be

some superior means by which that belief could be acquired: a means that would create

a stronger rational connection between the believer and the object of her belief,

conferring stronger justification on that belief.

To stop this regress of ‘better means of belief acquisition’, a proponent of ideal

conceivability might appeal to a notion of acquaintance, such as Russell’s account of

acquaintance with sense data. Subjects may be related to the object of their phenomenal

beliefs (for example, the phenomenal properties of their conscious experience) by being

acquainted with them, whereas the dualist might claim that knowledge by acquaintance

is the ultimate justification for a belief. Whether an appeal to ‘acquaintance’ will

successfully stop the regress of ‘better means of belief acquisition’ will depend on one’s

views on acquaintance in general.

Defining ideal conceivability in terms of ‘undefeatable justification’ rather than

what ‘an ideal reasoner could not rule out,’ does not conclusively remove the conceiving

subject from conceivability. This may be unavoidable, since the notion of a reasoner may

be conceptually necessarily connected to the notion of conceivability. If that is the case,

then ideal conceivability stands apart from the exemplary rational notions of ‘a

prioricity’ and ‘entailment’. ‘Being true a priori’ and ‘being entailed by’ are properties of


34

propositions themselves. In the first case, the property can belong to an individual

proposition, whereas in the second case, the property belongs to a set of propositions

(for example, proposition P entails proposition Q). To be a ‘rational notion’, it seems

that ideal conceivability should be a property of propositions themselves, as Chalmers

(2002a) has claimed it to be. But unlike purely rational notions, conceivability–even

ideal conceivability–is not a property of propositions or sets of propositions in

themselves. The property of ‘being justified’ belongs to a belief, claim or judgment.

Beliefs, claims and judgments are not propositions, but rather propositional attitudes

and as such they conceptually require belief-holders, claim-holders and judgment-

makers. These require a subject who forms and holds those beliefs and in whose

cognitive economy that belief can be instantiated.

Insofar as beliefs are the kind of things that can be justified, beliefs are not

merely propositions. ‘Being justified’ is a property of a proposition, only insofar as that

proposition is a belief. And the kind of justification that belongs to a particular belief, or

the extent to which that belief is justified, depends on the kind of epistemic relation that

holds between a believer and the object of her belief. The believer is not something with

which we can dispense without losing precisely that entity to which the property ‘being

justified’ belongs: the belief itself.

Acts of conceiving and reasoning–ideal or otherwise–are epistemic acts. As such,

the notion of an epistemic agent or subject is integrated into the very concepts of

conceiving and reasoning. Conceiving, reasoning, and modally imagining require some

subject who possesses concepts and can make inferences. Thus it is unavoidable that a

coherent notion of ideal conceivability will be one, in which there is a conceiver.

Therefore, the problem Yablo poses for ideal conceivability is still present: nothing is to


35

prevent a subject from believing she can modally imagine a scenario and finding it

ideally conceivable, if she is ignorant of some essential property of a property or concept

involved.

6. The Anti-Zombie: Has the Philosophical Zombie Met its Match?

At root, the zombie conceivability argument relies on the intuition that zombies

and non-physical mental properties are conceivable. This is the starting point for the

argument that zombies are conceivable in either the ideal or rational sense.

Sophisticated philosophical arguments follow this initial intuitive judgment, but without

the intuition that mental properties are non-physically conceivable, the zombie

conceivability argument would have no foundation. As a response, several ‘anti-zombie’

parity arguments have emerged, which use the same resources as the zombie

conceivability argument to construct arguments from the conceivability of anti-zombies

to the truth of physicalism (see Brown (2010), Frankish (2007), Sturgeon (2000) and

Marton (1998)).

Anti-zombies are molecule-for-molecule duplicates of human beings, who are

made conscious by physical facts alone. The anti-zombie is nothing over and above the

picture that results from embracing physicalism about the mind. If physicalism is true,

then we are made conscious by our physical facts alone and anti-zombies are merely

human beings viewed from a physicalist perspective. The conceivability of zombies has

intuitive support, but the intuitive support for anti-zombies may be equally strong. As

long as physicalism is no less intuitively plausible than property dualism (or the

possibility of zombies), the same conceivability argument that shows zombies to be

possible, will equally show anti-zombies to be possible.


36

The property dualist claims that the zombie is ideally positively conceivable; that

is, if any reasonable epistemic agent were to modally imagine the zombie-scenario, she

could fill in all the details of that scenario without encountering a contradiction. If there

is no reason not to extend that same charity to the physicalist, then any reasonable

epistemic agent should be able to modally imagine the anti-zombie scenario without

encountering a contradiction. Assuming that the conceivability-possibility principle

holds, the conceivability of anti-zombies entails their possibility. Therefore, it would be

possible for creatures physically identical to ourselves to be made conscious by physical

facts alone. If physicalism and property dualism are both intuitively plausible, and if the

conceivability-possibility principle holds, then both zombies and anti-zombies are

possible.

But this result is a contradiction, because if physical facts alone make an anti-

zombie conscious, and the same physical facts hold in the zombie world, then the

physical facts of the zombie-world would make the zombie conscious. And the concept

conscious zombie is clearly internally inconsistent.

At this point, there are three options. To avoid contradiction, those who find both

zombies and anti-zombies intuitively conceivable may:

(a) reject the conceivability-possibility thesis

(b) deny the genuine conceivability of physicalism

(c) deny the genuine conceivability of dualism

Most physicalists endorse (a) by denying that conceivability implies possibility,

hence being able to grant that both zombies and anti-zombies are conceivable without


37

contradiction. The only option available to the proponent of the zombie conceivability

argument is to endorse (b) by denying the conceivability of physicalism. In response to

the anti-zombie argument, Chalmers (2010: 180) accepts this and denies the

conceivability of physicalism in any philosophically significant way:

It may be prima facie negatively conceivable that materialism is true about

consciousness, but [the truth of physicalism] is not obviously conceivable

in any stronger sense. Many people have noted that it is very hard to

imagine that consciousness is a physical process. I do not think that this

unimaginability is so obvious that it should be used as a premise in an

argument against materialism, but likewise, the imaginability claim [i.e.

the imaginability of materialism] cannot be used as a premise [in an

argument for materialism], either (2010: 180).

The anti-zombie argument shows that endorsing the genuine conceivability of

zombies leads to a stronger commitment than might have been initially obvious. Rather

than merely denying the truth of physicalism, one must deny that physicalism is a

conceptually coherent position. Proponents of the zombie conceivability argument must

allow that the dualist’s prima facie judgment, that zombies are conceivable, can support

the stronger claim that zombies are conceivable in the rational sense, which is required,

in order for the conceivability-possibility principle to establish the truth of dualism. At

the same time, they must deny that the physicalist’s prima facie judgment that

physicalism is conceivable can support the stronger claim that physicalism is

conceivable in the rational sense, which in turn is required, in order for the

conceivability-possibility principle to establish the truth of physicalism. If physicalism


38

were conceptually coherent, then according to the conceivability-possibility principle,

physicalism would be logically possible and the zombie conceivability argument would

collapse in contradiction.


39

7. References and Further Reading

Block, N. and Stalnaker, R. (1999). ‘Conceptual Analysis, Dualism, and the Explanatory

Gap.’ Philosophical Review 108: 1–46.

Botterell, A. (2001). ‘Conceiving What is Not There.’ Journal of Consciousness Studies

8: 21-42.

Braddon-Mitchell, D. (2003). ‘Qualia and Analytical Conditionals.’ Journal of

Philosophy 100: 111–135.

Brown, R. (2010). ‘Deprioritizing the A Priori Arguments Against Physicalism.’ Journal

of Consciousness Studies 17 (3-4): 47-69.

Campbell, K. (1970). Body and Mind. London: Macmillan.

Chalmers, D. J. (1996). The Conscious Mind: In Search of a Fundamental Theory. New

York and Oxford: Oxford University Press.

Chalmers, D. J. (2002a). ‘Does Conceivability Entail Possibility?’ Conceivability and

Possibility. T. S. Gendler and J. Hawthorne (eds.). Oxford: Clarendon Press.

Chalmers, D. J. (2002b). ‘Consciousness and its Place in Nature.’ Philosophy of Mind:

Classical and Contemporary Readings. D. Chalmers (ed.). Oxford: Oxford

University Press.

Chalmers, D. J. (2010). The Character of Consciousness. Oxford: Oxford University

Press.

Davidson, D. (1970). ‘Mental Events.’,Essays on Actions and Events. D. Davidson.

Oxford: Oxford University Press: 207–223.


40

Dennett, D. C. (1991). Consciousness Explained. Boston: Little, Brown, & Co.

Dennett, D. C. (1995). ‘The Unimagined Preposterousness of Zombies.’ Journal of

Consciousness Studies 2.4: 322-326.

Dennett, D. C. (1999). ‘The Zombic Hunch: Extinction of an Intuition?’ Royal Institute

of Philosophy Millennial Lecture.

Dennett, D. C. (2005). Sweet Dreams: Philosophical Obstacles to a Science of

Consciousness. Cambridge, MA: MIT Press/Bradford Books.

Dennett, D. C. (2013). Intuition Pumps and Other Tools for Thinking. New York:

Norton.

Descartes, R. (1640/1984). ‘Meditations on First Philosophy.’ The Philosophical

Writings of Descartes: Vol. 2. J. Cottingham, R. Stoothoff and D. Murdoch,

Trans. Cambridge: Cambridge University Press: 1-50.

Doggett, T. and Stoljar, D. (2010). ‘Does Nagel’s Footnote Eleven Solve the Mind Body

Problem?’ Philosophical Issues 20 (1): 125-143.

Frankish, K. (2007). ‘The Anti-Zombie Argument.’ Philosophical Quarterly 57(229):

650–666.

Graham, G. and Horgan, T. (2000). ‘Mary Mary Quite Contrary.’ Philosophical Studies

99: 59-87.

Hill, C. S. (1997). ‘Imaginability, Conceivability, Possibility and the Mind-Body

Problem.’ Philosophical Studies 87: 61–85.

Hill, C. and McLaughlin, B. (1999). ‘There are Fewer Things in Reality Than are Dreamt

of in Chalmers’s Philosophy.’ Philosophy and Phenomenological Research 59:

445–454.


41

Kirk, R. (1974a). ‘Sentience and Behaviour.’ Mind 83: 43–60.

Kirk, R. (1974b). ‘Zombies v. Materialists.’ Proceedings of the Aristotelian Society 48

(Supplementary): 135–152.

Kirk, R. (2008). ‘Zombies and Consciousness.’ Philosophical Books. Y. Nagasawa (ed.).

49.2: 170-171.

Kripke, S. (1972). ‘Naming and Necessity.’,Semantics of Natural Language. G. Harman

and D. Davidson (eds.). Dordrecht-Holland/Boston-USA: Reidel: 253-355.

Leibniz, G. W. (1720/1925). The Monadology. D. Garter and R. Aries (trans.).

Indianapolis: Hackett.

Loar, B. (1990). ‘Phenomenal States.’ Philosophical Perspectives 4: Action Theory: 81-

108.

Marcus, E. (2004). ‘Why Zombies are Inconceivable.’ Australasian Journal of

Philosophy 82.3: 477-490.

Marton, P. (1998). ‘Zombies vs. Materialists: The Battle for Conceivability.’ Southwest

Philosophy Review 14: 131–38.

Nagel, T. (1976). ‘What Is It Like To Be A Bat?’ Reprinted in Nagel (1979).

Nagel, T. (1979). Mortal Questions. Cambridge: Cambridge University Press.

Shoemaker, S. (1975). ‘Functionalism and Qualia.’ Philosophical Studies 27: 291-315.

Stoljar, D. (2005). ‘Physicalism and Phenomenal Concepts.’ Mind and Language 20.2:

296-302.

Sturgeon, S. (2000). Matters of Mind: Consciousness, Reason and Nature. London and

New York: Routledge.


42

Thomas, N. J. T. (1998). ‘Zombie Killer.’ Toward a Science of Consciousness II: The

Second Tucson Discussions and Debates. S.R. Hameroff, A.W. Kaszniak, and A.C.

Scott (eds.). Cambridge, MA: MIT Press/Bradford Books.

Yablo, S. (2003). ‘Causal Relevance.’ Noûs-Supplement: Philosophical Issues 13: 316-

328.

Philosophical Zombies (for IEP)

Documents

Transcript of Philosophical Zombies (for IEP)