Philosophy Compass 6 (4), 2011, 267-281

Phenomenal concepts
Pär Sundström, Umeå University

Abstract: It's a common idea in philosophy that we possess a peculiar kind of "phenomenal concepts"
by which we can think about our conscious states in "inner" and "direct" ways, as for example, when I
attend to the way a current pain feels and think about this feeling as such. Such phenomenal ways of
thinking figure in a variety of theoretical contexts. The bulk of this article discusses their use in a
certain strategy – the phenomenal concept strategy – for defending the physicalist view that conscious
states are reducible to brain states. It also considers, more briefly, how phenomenal concepts have
been used to defend dualism about consciousness, and how they have been used to explain our special
access to our consciousness. It concludes with a discussion about whether, and in what more precise
sense of the term, we at all possess "phenomenal concepts" of our conscious states.

1 Introduction

'Phenomenal concept' is a technical term that is specified in different ways in the

literature, but the following general idea can serve as our initial focus. We can think
about conscious states – the idea is – under both phenomenal and nonphenomenal
concepts. The former are concepts we employ to think about conscious states in
“inner” and “direct” ways; the latter concepts we employ to think about them in
“outer” and “indirect” ways. For example, I employ a phenomenal concept of pain
when I inwardly attend to the way a current pain feels and think about this feeling as
such. For an example of a nonphenomenal concept, consider a case where I think
“what bothers Sarah must be unpleasant”, not knowing whether Sarah is bothered by a
pain or something else. If she is in fact bothered by a pain, then my thought concerns
that pain, but my concept "gets to" the pain through an outer and indirect route and is
therefore "nonphenomenal".
 2

Phenomenal ways of thinking about consciousness figure in different contexts in

philosophy. At present, they are most strongly associated with a certain strategy – the
phenomenal concept strategy – for defending the physicalist view that conscious
states are reducible to brain states. To think about consciousness in inner and direct
phenomenal ways is very different from thinking about the brain in terms of its
biological properties. According to the phenomenal concept strategy, this difference
in ways of thinking can deceive us into supposing that the thinkings must concern
different and dissociable things. But there is no reason to so suppose, because,
familiarly, different ways of thinking might concern the same thing; as when I think
about myself in an inner way, as I, and also think about (what turns out to be) myself
as the shopper who set off the alarm.

However, phenomenal ways of thinking about consciousness have been invoked for
purposes other than defending physicalism as well. In direct opposition to the
phenomenal concept strategy, they have been invoked to defend dualism about
consciousness. They have also been invoked to explain our special access to our own
consciousness.

The bulk of this article concerns the (physicalist) phenomenal concept strategy.
Section 2 explains what the strategy is, and section 3 discusses some challenges to it.
Section 3 also illustrates, in passing, the use of phenomenal concepts in arguments for
dualism. Section 4 briefly considers how phenomenal concepts have been invoked to
explain our special access to our own consciousness. Section 5 summarises and
extends a bit through a discussion about whether, and in what more precise sense of
the term, we at all possess "phenomenal concepts" of our conscious states.

A terminological note: Phenomenal concepts are variously said to be concepts of

“conscious states”, “phenomenal properties”, or of ways that things are subjectively
“like” for a subject. For present purposes, we may take these to be equivalent: a
phenomenal property may be understood as a type of conscious state, or, equivalently,
as a way that things might be subjectively like for a subject.
 3

2 The phenomenal concept strategy

2.1 Physicalism and its challenges

According to physicalism, conscious states are reducible to brain states, or other

physical states.1 Reduction is sometimes understood in terms of identity between
conscious states and brain states; sometimes in terms of necessary connections
between the two. But on either understanding, the view is that conscious states are as
intimately tied to brain states as water is to H2O, liquidity to loose molecular bonding,
or the global pattern of a pixel grid to the activation of its individual pixels (cf. Lewis
1994). It's not just that consciousness is connected to the brain in some regular or
law-like way; rather, it is in some sense "nothing over and above" what goes on in the
brain.

It might seem counter-intuitive that our rich inner world of consciousness should be
thus reducible to brain states. This (counter-)intuition has also been developed in
various arguments against or challenges for physicalism, prominent among which are
the following three.

The explanatory gap (Levine 1983; 2001). Theoretical reductions seem typically to
be somehow explanatory. For example, the "plasma-like" cohesiveness and
malleability characteristic of liquidity seems to be explained by the loose molecular
bonding that underlies or is identical with it. By contrast, it seems that no account of
what goes on in the brain while I experience red could explain what this is like
subjectively.

The "appearance of contingency" (see Kripke 1972/1980, 144-155; Chalmers 1996;

2010). On the face of it, it seems that the brain state I'm in right now could obtain
without the conscious state I'm in right now. If physicalism is true, this is an illusion:

1
Henceforth, 'brain state' will abbreviate 'brain state or other physical state'. Physical states that are
not confined to the brain but that might be part of what consciousness reduces to include states of the
body (O'Regan and Noë 2001; Noë 2004), and relations to the environment (Dretske 1995, chap. 5; Tye
1995, chap. 5; 2000, chap. 3; Lycan 2001). Physical states should here be understood in a broad sense
that includes for example biological and functional states.
 4

the two are in fact inseparable. But if this is so, how might the illusion of a merely
contingent connection between the two arise? What could explain that reality
presents itself to us in this distorted way?

The challenge for physicalism here is that a standard way of explaining this kind of
illusion seems not to be available in the present case. Consider the fact that it seems
possible for rapid molecular motion to obtain without heat, even though (as it is
commonly held) this is impossible since heat is molecular motion, and something
can't fail to obtain without itself. In this case we can explain the illusory possibility
by appealing to the idea that we are prone to confusing heat itself, the physical
phenomenon, with the sensation of heat, which is the way heat appears to us. It is
really possible that there could be rapid molecular motion without the sensation of
heat. This real possibility together with our tendency to confuse heat itself with the
way it appears to us can explain our illusory impression that there could be rapid
molecular motion without heat.

However, in the case of consciousness, we can't distinguish the way it really is from
the way it subjectively appears; consciousness just is what it's like for a subject. Thus,
when I'm under the impression that my current brain state could obtain without my
current conscious state, there seems to be no other scenario (such as, the brain state
without the subjective appearance of consciousness) that I plausibly confuse with this
one. Therefore, if it is an illusion that consciousness and the brain are merely
contingently connected, we need a different account of how this illusion arises.

The knowledge argument (Jackson 1982; 1986). As illustrated by Frank Jackson's

Mary, it seems in principle possible for someone to (a) know all the physical facts
about colour vision, and be able to derive all that follows from these facts, while (b)
not being able to figure out what it's like to see red. This might seem to show that
what it's like to see red isn't reducible to physical facts – because if it were, Mary
should be able to figure out what it's like to see red in virtue of knowing all the
physical facts and being able to derive all that follows from them.
 5

2.2 The phenomenal concept strategy: The general idea

The phenomenal concept strategy is a strategy for responding to such challenges for
physicalism. It is a strictly defensive strategy: it operates on the assumption that there
are good reasons to accept physicalism and aims to fend off apparent reasons for
rejecting it.2

What is it to apply the phenomenal concept strategy to a given challenge to

physicalism? As I understand the strategy here, it is to make the following
Acknowledgment, and develop the following kind of Account of Acknowledgment:

Acknowledgment: The relevant challenge brings out a way in which the

reduction of conscious states to brain states is exceptional among
theoretical reductions.

Account: However, the exceptional character of the case can be exhaustively

explained in terms of our phenomenal concepts of consciousness. What's
importantly exceptional about the case are the concepts involved; not the
phenomena. (Which is not to say that these concepts are in any way
deficient; they are just special.)

For example, the phenomenal concept strategy vis-à-vis the explanatory gap is to
concede that a peculiar gap pertains to the reduction of consciousness to the brain, but
offer an explanation of this in terms of our phenomenal conceptions. By the
physicalist hypothesis, conscious states are physical states, and as physically
conceived, they can be explained in the way typical of theoretical reductions. It is
only as phenomenally conceived that they can't be so explained, but the reason for this
lies entirely in the phenomenal conceptions.

The approach to the appearance of contingency is similarly to acknowledge that there

is a peculiar, illusory appearance of contingency about the brain-consciousness

2
Phenomenal concept theorists typically accept physicalism on the grounds that (a) some conscious
states have physical effects and (b) all physical effects have complete physical causes. See for example
Papineau (2002, chap. 1 and appendix), and Levine (2001, chap. 1)
 6

connection – an appearance that can't be explained in terms of some real contingent

connection between how the brain or consciousness really are and how they present
themselves to us – but urge that this appearance arises about conscious states only as
phenomenally conceived, and that it can be exhaustively explained by the character of
the phenomenal conceptions.

In the case of the knowledge argument, the phenomenal concept strategy is to accept
that the case of consciousness allows, in a way that's exceptional among theoretical
reductions, the Mary-style possibility of knowing all the physical facts without being
able to figure out that it is, subjectively, the way it is to experience red. The
explanation offered for this is that someone in Mary's situation would lack the
phenomenal concept required to know – or even think – the relevant proposition
concerning what it's like to experience red. And this is compatible with her knowing
other propositions, expressible in physical terms, that concern this same fact. The
resulting suggestion is that Mary's situation is like that of someone who knows that
there are two things in front of her but doesn't know – for lack of the relevant concept
– that the number of things in front of her is the square root of four.

2.3 Other physicalist views

The phenomenal concept strategy should be distinguished from two other types of
physicalist view.

In the first place, Acknowledgment distinguishes the phenomenal concept strategy

from what we may call nonexceptionalist physicalism. According to this view,
consciousness is reducible in a way that is not in any significant way exceptional
among theoretical reductions. One example of this view is a broadly applied
"analytical functionalism" (Lewis 1966; 1972; 1980; Armstrong 1968). This view
says that a theoretical reduction is typically accomplished by a "causal-role analysis"
of some ordinary concept (for example, of liquidity as whatever it is that has such-
and-such typical causes and effects), followed by an empirical discovery of which
physical phenomenon in fact has those causes and effects (in this case: loose
molecular bonding). Broadly applied, the view says that conscious states can be
 7

similarly reduced with the help of causal-role analyses of our ordinary concepts of
them.

Second, the phenomenal concept strategy is distinct from what we may call lack-of-
understanding physicalism. This view agrees that there is something exceptional
about the reduction of consciousness, and that this can be accounted for in terms of
our concepts or theories, but its explanation is that there is something radically
missing in our current concepts or theories. The reason that the physical basis of
consciousness is not perspicuous to us is, according to this view, that our concepts or
theories fail to adequately capture the relevant phenomena. Varieties of this view
have been proposed by Thomas Nagel (1974; 1998; 2000), Colin McGinn (1989;
2001), and Daniel Stoljar (2005; 2006).

It's worth noting that, while the phenomenal concept strategy and lack-of-
understanding physicalism agree on Acknowledgment as formulated above, there is an
important underlying disagreement here. According to lack-of-understanding
physicalism, there is a theory from the point of view of which the connection between
the brain and consciousness is explanatory and transparently necessary in the way that
the connection between, say, liquidity and loose molecular bonding is.3 The reduction
of consciousness is thus in principle "alignable" with the other theoretical reductions,
whether or not we are now or ever in a position to fully grasp this. The phenomenal
concept strategy, by contrast, denies that there is a theory that unifies the brain and
consciousness in the way familiar from other reductions. It promises instead an
explanation of why this is so.

The contrasts with other views bring out some appealing characteristics of the
phenomenal concept strategy. By acknowledging that the case of consciousness is
exceptional among theoretical reductions the strategy pledges to take consciousness
and the frequently felt puzzlement about it as seriously as anyone can ask; it thus
avoids a common complaint about nonexceptionalist physicalism. In suggesting that
the exceptional features of the case can be explained in terms of a difference among
(non-deficient) concepts, it simultaneously promises a theoretically conservative

3
This is emphasised by Nagel (1998, sect. 7; 2000, sect. 6) and McGinn (1989, 353, 361-2; 2001, 299-
301 and passim).
 8

account of the case; it thus avoids a common complaint about lack-of-understanding

physicalism.4

2.4 Phenomenal concepts: Varieties and commonalities

What might phenomenal concepts be, that they can support the explanatory aims of
the phenomenal concept strategy? The literature presents a rich flora of proposals,
among them that phenomenal concepts are:

(i) recognitional concept that one possesses (partly) in virtue of being able to
recognise particular instances as being of the same kind (Loar 1990/1997;
2003; Tye 2000, chap. 2; 2003; Carruthers 2000; 2004; Perry 2001; Levin
2007a; 2007b);
(ii) "quotational" concepts that one employs by using the very conscious state
that one is thinking about (Papineau 1993, chap. 4; 2002, chap. 4; 2007;
Balog 1999; forthcoming a and b; Melnyk 2002; Block 2007 draws on a
similar idea);
(iv) concepts that are distinguished by the special conditions under which one
is justified in applying them (Sturgeon 1994; 2000, chap. 2; Hill 1997;
Hill and McLaughlin 1999);
(iii) indexical concepts, similar to I and now (Tye 1995, chap. 6; Lycan 1996,
sect. 3.3);
(v) concepts that don't just label "we know not what" but that present
conscious states in a peculiarly "substantive and determinate way"
(Levine 2001, 84);
(vi) "conditional" concepts that refer to nonphysical states if we have
appropriate states of this kind, and otherwise refer to physical states
(Hawthorne 2002; Braddon-Mitchell 2003);5

4
Of course, the phenomenal concept strategy is theoretically conservative only to the extent that its
account of phenomenal concepts is. Phenomenal concept theorists typically suppose that phenomenal
concepts can be accounted for in relatively familiar terms. However, Levine (2001) is a salient
exception. As I read him, Levine combines (a) a phenomenal concept strategy to the problem of
consciousness with (b) a lack-of-understanding physicalism about phenomenal concepts, urging that
our current theoretical resources are inadequate to understanding the latter in physical terms. On the
grounds for doubting that phenomenal concepts can be physically explained, see section 3.3 below.
 9

(vii) concepts that utilise cognitive structures that underlie "sensory

concepts", inheriting a kind of representational simplicity of the latter
(Aydede and Güzeldere 2005)7

I can't do justice, here, to the variety of ways in which phenomenal concept theorists
have addressed challenges to physicalism on the basis of these proposals. I shall
instead identify two ideas that are common to most or all varieties of the strategy.

First, phenomenal concept theorists generally agree that phenomenal concepts –

however they should exactly be understood – are inferentially isolated from physical
concepts in the sense that one can't infer a phenomenal characterisation of
consciousness from a purely physical characterisation, however extensive. This is to
reject, among other things, the analytical functionalist view that these concepts can be
analysed in terms of the typical causes and effects of their referents.

This claim is a fundamental component of all phenomenal concept treatments of the

explanatory gap and the appearance of contingency. It is commonly agreed that, if
one could infer phenomenal characterisations of consciousness from physical
characterisations of the brain, then there would be no explanatory gap or appearance
of contingency between the two. To that extent, at least, the inferential isolation
between the two kinds of concept explains the gap and the appearance of
contingency.8

5
Stalnaker (2002) develops a similar suggestion, but unlike Hawthorne and Braddon-Mitchell,
Stalnaker suggests (251-2) that the conditional character is not distinctive of our concepts of
consciousness, but pertains to concepts that we employ in other theoretical reductions as well; for
example water. Since the case of consciousness is thus not claimed to be exceptional, Stalnaker’s
suggestion is not an instance of the phenomenal concept strategy as understood here (cf. section. 2.2-
2.3).
7
These are proposals about what a phenomenal concept might be. But what is a concept to begin with?
It is useful to distinguish at least three different uses of this term. (i) On one use, 'concept' denotes
inner mental symbols that we, according to one view, use to think with (Fodor 1975; 1998). (ii) On
another use, 'concept' stands for certain abstract entities, Fregean Sinne, that compose into propositions
and make up what we think (for this use of the term, see Peacocke 1992; for more on Sinne, see Frege
1918/1997). (iii) On a less committal use of the term, to say that someone "employs a concept" is just
to say that the person thinks something. And to say that someone possesses a concept is to say that she
is able to think something. This use involves no commitment either about what we think when we
think, or about what is involved in thinking (cf. Byrne 2005, sect. 1.1; Sundström forthcoming, sect.
2.4). For present purposes, we can live with this ambiguity, since one can make sense of the relevant
issues on each of the three uses.
8
The phenomenal concept strategy can at this point be developed in two different ways. (1) One might
claim that the inferential isolation of phenomenal concepts by itself suffices to account for the gap and
appearance of contingency. This may be the view of Tye (2000; 2003); he at any rate does not appeal
 10

Second, phenomenal concept theorists typically claim that phenomenal concepts are
experience-dependent in the sense that, in order to possess a phenomenal concept of
some conscious state S one needs to oneself have experienced S.

This claim is a fundamental component of phenomenal concept treatments of the

knowledge argument. If phenomenal concepts are experience-dependent, that can
explain how an experience-deprived subject could know all the physical facts
(physically conceived) while not knowing – for lack of the relevant concept – what it's
like to see red (phenomenally conceived).

3 Worries about the strategy

3.1 Overgeneration

One kind of worry – which one may have either about the strategy in general, or about
some particular version of it – is that it may seem to overgenerate, in the sense of
entailing false predictions about what identity or necessity claims we find
troublesome.

A version of this concern can be extracted from Stoljar (2005; 2006, sect. 9.6.2).
Stoljar remarks (in effect) that when some peculiar feature F pertains to some concept
C, F is often inherited by complex concepts that C partly constitutes. For example, if
C is inferentially isolated from all concepts of a certain kind, it's plausible that the

to any feature of phenomenal concepts other than inferential isolation in accounting for the gap (2000,
32-5; 2003, § 7) or for concerns around the appearance of contingency (2000, 29-32; 2003, § 5). It
may also be the view of Diaz-Leon (forthcoming, sect. 3.2 and 3.4). Alternatively, (2) one might claim
that the inferential isolation partly accounts for the gap and the appearance of contingency, but that
something further must be identified to provide the full account. This is held by Loar (1990/1997),
Levine (2001), and Papineau (2002). The motivation for this view is that there seem to be cases where
(a) an identity or necessary connection is established between phenomena, and (b) our characterisations
of these are inferentially isolated from one another, but where (c) we don't find any explanatory gap or
appearance of contingency; at any rate, no gap or appearance of contingency that we find bothersome.
This seems to be so, for example, in cases involving demonstrative reference, like "this stuff is
CH3CH2OH" (cf. Loar 1990/1997, 608).
 11

concept not-C is inferentially isolated from these concepts as well. Now consider the
following statement:

NumberNoRed: If something is a number then it's not an experience of red.

We can stipulate that (i) experience of red should here be conceived phenomenally.
Moreover, it seems plausible that (ii) whatever "disparity" there is between
phenomenal concepts and physical concepts, there is also between phenomenal
concepts and mathematical concepts. Given (i) and (ii), it would seem that: if, as the
phenomenal concept strategy says, it is some disparity between phenomenal concepts
and physical concepts that makes the brain-consciousness connection appear to us
contingent, then NumberNoRed should also appear to us contingent. However,
NumberNoRed doesn't appear to us contingent. It is, and appears necessary.
Therefore, the phenomenal concept strategy seems to entail an incorrect prediction.9

Sundström (2008) raises the same type of objection against Papineau's "quotational"
phenomenal concept account of the intuition of brain-consciousness distinctness.

3.2 Conceiving consciousness as it is in itself and the threat of "deconstruction"

As noted above, it seems that in the case of consciousness there is no distinction

between how it subjectively appears and how it really is. Phenomenal concept
theorists tend to accept this; it is part of what they typically concede to the challenge
from the appearance of contingency, for example (cf. sections 2.1 and 2.2 above).

Terence Horgan and John Tienson (2001) argue that the phenomenal concept strategy
– or what they call "new wave materialism" – "deconstructs" under this concession.
Their argument runs as follows:

9
See Diaz-Leon (2008, sect. 3) for a response to this worry.
 12

1 When a phenomenal property is conceived under a phenomenal concept,

this property is conceived otherwise than as a physical-functional
property.
2 When a phenomenal property is conceived under a phenomenal concept,
this property is conceived directly, as it is in itself.
3 If (i) a property P is conceived, under a concept C, otherwise than as a
physical-functional property, and (ii) P is conceived, under C, as it is in
itself, then P is not a physical-functional property.
Hence,
4 Phenomenal properties are not physical-functional properties (2001, sect.
3).

The argument is valid, and its conclusion contradicts physicalism. According to

Horgan and Tienson, "new wave materialists are committed to premises 1 and 2".
They can therefore contest only premise 3. Their problem, however, is that 3 "seems
virtually tautologous" (2001, sect. 3).11

I suspect that Horgan and Tienson are only partly right about where new wave
materialists or phenomenal concept theorists may contest the deconstructive
argument.12 I believe there is a crucial disagreement about the interpretation under
which premise 2 is correct, and that this disagreement underlies any disagreement
about 3.

It is true that phenomenal concept theorists tend to accept wordings like that of
premise 2. For example, Brian Loar says that a phenomenal concept picks out its
referent "directly and essentially" (1990/1997, 600). Katalin Balog says that
phenomenal concepts "provide a grasp of the phenomenal properties in a way that
reveals their essence" (forthcoming b, sect. 2). And Brian McLaughlin, in a response

11
Goff (forthcoming) develops a similar objection to the phenomenal concept strategy.
12
There is room for discussion about whether "new wave materialism", as Horgan and Tienson
understand it, is exactly the same as or slightly different from the "phenomenal concept strategy" as
understood here. For present purposes we can disregard most of the (potential) differences between the
views. But one should note that Horgan and Tienson's argument targets strategies for defending
identity physicalism specifically. Due to limitations in space, I can't here discuss how the argument
might be extended to challenge the view that brain states and conscious states are (merely) necessarily
connected.
 13

to Horgan and Tienson's argument, concedes that: "New wave materialists … embrace
premise 2: it is one of the central tenets of their view" (2001, 324).

However, it turns out that Loar, Balog, and McLaughlin all have serious reservations
about the sense in which our phenomenal conceptions "reveal the essences" of
phenomenal properties. McLaughlin urges that phenomenal concepts "do not
conceptually reveal anything about the essential nature of phenomenal properties:
they simply name or demonstrate them" (2001, 324), and that phenomenal concepts
"present phenomenal properties only in the sense that they directly refer to
phenomenal properties (2001, 328). Loar distinguishes two different uses of 'capture
the essence of':

On one use, it expresses a referential notion that comes to no more than

'directly rigidly designate'. On the other, it means something like 'be
conceptually interderivable with some theoretical predicate that reveal the
internal structure of the designated property'" (1990/1997, 603).

Loar only ever acknowledges that phenomenal concepts "capture the essence" of
phenomenal qualities in the former sense (see, e.g., 608-9). And Balog states that
phenomenal concepts "will not afford any clue as to the fundamental nature of the
referent". While they "afford an insight into the essence of the referent", the sense in
which they do so is "by exemplification"; i.e., in the sense that phenomenal concepts
use phenomenal properties to think about phenomenal properties (forthcoming b, sect.
2).

Given these reservations, it is not clear that Loar, McLaughlin and Balog accept that
we conceive phenomenal properties "as they are in themselves" in the sense that
Horgan and Tienson intend in premise 2. Horgan and Tienson have in mind, I
believe, something like an insight into, or understanding of the essential nature of
phenomenal properties. But to say that we refer to phenomenal properties "directly"
or "rigidly" or "by exemplification" is not to accept any such thing. We can plausibly
(a) refer to something directly or rigidly or by exemplification while (b) lacking
insight into the essential nature of what we refer to. Consider for example a set of
differently shaped cardboards attached to sticks. I may lift one of the cardboards by
 14

its stick without looking at it and say, "now look at this (type of) shape". It is
plausible, I think, that I can in this case refer directly, and rigidly, and by
exemplification to a type of shape while having little insight into its essential nature.
It seems natural to interpret Loar's, McLaughlin's and Balog's reservations about the
"essence revealing" character of phenomenal concepts, and their emphases instead on
"directness", "rigidity" and "exemplification", as nods towards some such model for
understanding how we grasp conscious states under phenomenal concepts.13

This dispute concerning premise 2 underlies, I think, any disagreement concerning

premise 3, which says, to repeat:

3. If (i) a property P is conceived, under a concept C, otherwise than as a

physical-functional property, and (ii) P is conceived, under C, as it is in
itself, then P is not a physical-functional property.

If our phenomenal conceptions should display phenomenal properties "as they are in
themselves" in the sense that Horgan and Tienson are after, this might well be
"virtually tautologous". But if phenomenal conceptions display phenomenal
properties "as they are in themselves" only in the sense that Loar, McLaughlin and
Balog grant, there may be little reason to accept 3. At any rate, to the extent that
phenomenal concept reference should be modelled on the kind of "directness",
"rigidity" or "exemplification" illustrated by the cardboard case, there seems to be
little reason to accept it. It seems that I can in this case refer to a certain shape
directly, rigidly and by exemplification, and conceive it "otherwise than as" three-

13
The model may be imperfect. In the cardboard case, the direct, rigid, and exemplificatory reference
is to a high degree "cognitively blind"; I just refer to whatever shape it is that I in fact display.
Meanwhile, Loar, McLaughlin, and Balog emphasise that phenomenal concept reference involves a
"substantive", or "non-blind", or "non-naked" mode of presentation of conscious states (Loar
1990/1997, sect. 4-5; McLaughlin 2001, 326; Balog forthcoming b, sect. 2). But if that's the idea, there
may be no close analogies from other cases to appeal to. It is not easy (for me) to think of a case such
that: (a) I have a "non-blind" grasp of some property, and (b) this property is grasped as it is in itself
(rather than in terms of some contingent, superficial appearance), while (c) I nonetheless lack the kind
of insight into the essential nature of the property that Horgan and Tienson appear to be after in premise
2. Loar, McLaughlin and Balog may well suppose – or be committed to supposing – that phenomenal
concept reference uniquely exemplifies (a)-(c). For an argument that phenomenal concept theorists
should construe phenomenal concept reference as "cognitively blind", see Levin (2008). Hawthorne
(2002, 44-5) may perhaps be read as suggesting this as well.
 15

sided, in the argument's intended sense;14 but for all that, three-sidedness may be part
of the essence of the shape referred to.

We may note in passing that Horgan and Tienson's argument in effect illustrates the
above mentioned idea of invoking phenomenal concepts to defend dualism. While
presented as an ad hominem argument against "new wave materialists", whom Horgan
and Tienson portray as wedded to premises 1 and 2, one might promote the argument
and its anti-physicalist conclusion as plausible independently of new wave materialist
commitments. A dualist argument in this spirit is developed by Nida-Rümelin (2007).
As both Horgan and Tienson, and Nida-Rümelin note, their arguments are similar to
other influential arguments for dualism, such as Kripke's (1972/1980) "modal
argument", Chalmers' (1996; 2010) "two-dimensional argument", and the "property
dualism argument" (White 1986; 2007; see also Smart 1959, 148-50).

3.3 Are phenomenal concepts themselves physically explicable?

The phenomenal concept strategy aims to show that concerns about the physical
nature of consciousness can be explained in terms of phenomenal concepts. But what
about the phenomenal concepts themselves: can we give a satisfactory account of
their nature in physical terms?

Joseph Levine (2001; 2007) argues that phenomenal concepts provide us with a
peculiarly thick and substantive mode of presentation of conscious experiences, and
that we presently lack the resources to understand how this kind of representation
could be physically realised. A physicalist is according to Levine confined to explain
representation in terms of causal and nomological notions. And it seems hopeless to
explain the kind of access that we have to our conscious experiences in these terms.

David Chalmers (2007) provides a related "master argument" against all varieties of
the phenomenal concept strategy. The conclusion aimed for is:

14
Of course, I don't in this case conceive the shape as not three-sided. But I conceive it and I don't
conceive it as three-sided. And that suffices, it seems, for conceiving it "otherwise than as three sided"
 16

Chalmers' Conclusion: Phenomenal concepts are bound to be characterised

either (i) so "thinly" that they fail to account for our "epistemic situation"
with regard to consciousness, or (ii) so "thickly" that they can't be
explained in physical terms.

The argument is developed around the question whether it is coherently conceivable

that there be physical duplicates of us that lack our phenomenal concepts. If this is
coherently conceivable, then, Chalmers argues, phenomenal concepts are not
physically explicable; they themselves generate the kind of explanatory gap they were
supposed to explain (sect. 3.1). Suppose on the other hand that it is not coherently
conceivable that physical duplicates of us lack phenomenal concepts. That would be
to say that zombies (physical duplicates of us without consciousness) are guaranteed
to have whatever phenomenal concepts we have. But, Chalmers argues, zombies are
not in the kind of epistemic situation that we are in with regard to consciousness. For
example, when zombie-Mary leaves her black-and-white room, she does not gain
knowledge that is as "cognitively significant" as the knowledge that conscious Mary
acquires in the corresponding situation. Therefore, on this supposition, phenomenal
concepts don't explain our epistemic situation with regard to consciousness (sect. 3.2).

Responses to Chalmers' argument have typically in one way or other developed the
following idea. By the physicalist hypothesis, conscious states as well as our
phenomenal conceivings of conscious states are physical phenomena. And, just as we
can conceive of our conscious states in both phenomenal and nonphenomenal ways,
so we can conceive of our phenomenal conceivings in these two ways. As an instance
of this, Chalmers' Conclusion can be read in two ways, depending on whether it
employs a phenomenal or a nonphenomenal concept of phenomenal concepts.

One way to develop this idea is to urge that Chalmer's Conclusion is false if
phenomenal concepts are physically conceived. Physically conceived, phenomenal
concepts can be physically explained. But they can also explain our epistemic
situation, according to this suggestion. Although zombie-Mary has no conscious

in Horgan and Tienson's sense. They stress that conceiving of a property "otherwise than as" so-and-so
is "different from, and weaker than" conceiving it "as otherwise than" so-and-so (note 4).
 17

experience, she acquires on release from her confinement concepts that are novel to
her to a degree that parallels Jackson's Mary (see Carruthers and Veillet 2007;
Papineau 2007, sect. 5.4; Levin 2008, sect. 2).

Another response, developed by Balog (forthcoming b), urges that Chalmers'

Conclusion is true but harmless for the phenomenal concept strategy on both the
phenomenal and the nonphenomenal reading. Chalmers' Conclusion is true if
phenomenal concepts are physically conceived because, while phenomenal concepts
thus conceived can be physically explained, they don't explain our epistemic situation.
But this is something that a phenomenal concept theorist "should insist" on (sect. 3).
Chalmers' Conclusion is true if phenomenal concepts are phenomenally conceived
because, while phenomenal concepts thus conceived explain our epistemic situation,
they can't be physically explained. But this new explanatory gap – between physical
facts (physically conceived) and phenomenal concepts (phenomenally conceived) –
"doesn't pose a challenge to the physicalist over and above the original explanatory
gap" (ibid.).

Diaz-Leon (forthcoming) also suggests that Chalmers’ Conclusion is, even if true,
harmless for the phenomenal concept strategy. A phenomenal concept theorist need
not, she argues, account for our “whole epistemic situation” with regard to
consciousness, but only for the “inferential disconnection” between physical and
phenomenal concepts (sect. 3.2).

4 First-person psychology and epistemology

Besides being invoked to defend physicalism or dualism, phenomenal concepts have

been invoked, by both physicalists and dualists, to explain psychological and
epistemic phenomena such as:

(a) our peculiar direct acquaintance with our own consciousness;

(b) the fact that we seem to gain knowledge about our own consciousness in
ways not accessible from a third-person perspective;
 18

(c) the fact that we seem to be infallible or incorrigible about some

judgments about our own consciousness; for example, judgments that one
would naturally express in terms like, "I have this kind of experience
right now".

Balog (forthcoming a, sects. 3-4) offers an explanation of such phenomena based on

the "constitutional account" according to which phenomenal concepts use phenomenal
properties to refer to phenomenal properties. Chalmers (2003, sect. 4.1) draws on a
similar account to justify an incorrigibility thesis according to which a certain class of
phenomenal beliefs can't be false.15

The merits of these accounts can't be discussed here. However, it is worth noting that,
contrary to what is sometimes suggested,16 it is not apparent that phenomena like (a)-
(c) should be explained in terms of phenomenal concepts, or in terms of concepts at
all. Another possibility is that they should be explained by some non-concept
involving access, like an inner experience, on the basis of which we apply concepts to
our conscious states.17

5 So, do we possess phenomenal concepts of our conscious states?

So, do we possess phenomenal concepts of our conscious states? Whether we should

say so depends, of course, on what we mean by 'phenomenal concept'.

On a broad understanding, a concept is phenomenal if and only if it concerns a

conscious state. It is relatively uncontroversial that we possess phenomenal concepts
in this sense.18

15
See also Papineau (2002, sect. 4.12) for a similar suggestion.
16
See, e.g., Balog (2009, 299).
17
This is analogous to the proposal that our access to our outer environment is established by non-
concept involving perceptions, on the basis of which we apply concepts. For general overviews of such
"conceptualist" versus "nonconceptualist" debates, see Toribio (2007) and Bermúdez and Cahen
(2008). Gunther (2003) is an important collection of texts.
18
Exactly how controversial it is depends on what one means by 'concept'; cf. note 7 above.
 19

But as we have seen, philosophers typically mean something more specific than this
by 'phenomenal concept': the term is strongly associated with the idea that we possess
concepts of our conscious states that are in one way or other very special, or even
unique. We have encountered several proposals about what this specialness might
consist in. Three important suggestions are that these concepts:

(i) are inferentially isolated from physical concepts;

(ii) are experience-dependent, in the sense that, to have a concept of a type of
conscious state S, one needs oneself to have experienced S;
(iii) provide us with a grasp of what conscious states are in themselves.

It should be recognised, I believe, both that (i)-(iii) are all controversial, and that it's
controversial whether either one entails that the relevant concepts are unique. I will
conclude by summarising and discussing the suggestions, in a somewhat reshuffled
order.

Regarding the suggestion (i) that we possess concepts of consciousness that are
inferentially isolated from physical concepts, this is disputed by at least some
"analytical functionalists" (cf. section 2.3). But even supposing it is true, it is hard to
see that this would make the relevant concepts unique. As several phenomenal
concept theorists have pointed out, indexical and demonstrative concepts are plausibly
inferentially isolated from physical concepts as well, and the same may be true of,
say, natural kind concepts like water and heat (Loar 1990/1997, 608; Levine 2001, 82-
3, and sect. 2.4-2.5).

Regarding the suggestion (iii) that the relevant concepts provide us with a grasp of
what consciousness is in itself, I have suggested that verbal agreements over such
phrases often hide substantive disagreements. Some take this to mean that
phenomenal concepts provide us with something like an insight into the essence of
consciousness. Others, it seems, take it to mean only that phenomenal concepts refer
to conscious states directly or rigidly, or that their uses exemplify consciousness.

On either of these interpretations of (iii), it is again doubtful that, if true, it would

make phenomenal concepts unique. We plausibly refer directly and rigidly by means
 20

of proper names and demonstratives,19 and by exemplification when we use quoted

words to refer to words. It may also be plausible that we have theoretical concepts –
like an educated person’s concept of electron – that amount to insights into the
essences of what we conceive. The same might be true of many mundane concepts,
like sofa or friend.

The experience-dependence claim, (ii), is also controversial. A long-standing

objection to this kind of claim draws on the possibility of certain "duplication"
scenarios. Take the physical state that I was in last night as I fell asleep in the dark. It
is in principle possible that a molecule-for-molecule copy of me in that state should be
created, by design or accident, out of "nothing". The result would be a creature that
had never experienced any colour. But it seems plausible that the creature would
possess any concepts that I possess of what it's like to experience colours (Unger
1966; Dennett 2005, chap. 5).

This objection targets theses to the effect that it is absolutely impossible to have to
relevant kind of concept without the relevant kind of experience. It doesn't threaten
nomological theses, like the claim that this kind of concept-acquisition is ruled out by
the psychological laws of normal human development.20

But we should recognise this "anthropological" experience-dependence thesis to be

controversial as well. Analytical functionalists can contest it on the grounds that
causal role concepts can plausibly in general be acquired in ways other than by
experience. Another recent challenge comes from (overlapping) arguments by
Michael Tye (2009, sect. 3.6) and Derek Ball (2009).21 Tye and Ball argue that our
concepts of consciousness can be possessed even if one only "partially understands"
them, and that one can therefore acquire them from others even if one has not been in
the states they concern.22

19
Cf. Kripke (1972/1980), and Kaplan (1989).
20
Compare: Duplication possibilities may show that it is in principle possible to become a passable
speaker of Japanese without being exposed to Japanese. But it still seems to be a reliable, law-like,
psychological generalisation that all speakers of Japanese have been exposed to Japanese.
21
Tye is of course a former phenomenal concept theorist, but has turned into a critic in his 2009 book.
 21

A further challenge to the anthropological experience-dependence view comes from

Hume's missing shade of blue. If you experience a wide enough range of colours, that
plausibly suffices to acquire any concept of light blue that I possess, even if you
haven't experienced light blue in particular. Indeed, this may be how we typically, in
fact acquire our concepts of sensory states: given a varied enough sample we "fill in"
the rest. If so, it is not the case, even as an anthropological matter, that you must have
been in conscious state S to have a phenomenal concept of S.23

Finally, even if (ii) should be true – either as an absolute or an anthropological matter

– there is again room for doubt that this would make phenomenal concepts of
consciousness unique. Insofar as it's plausible that we have concepts of conscious
states that require having experienced these states, something analogous may well be
plausible for certain (demonstrative, or "perceptual") concepts of features of our
environment.24

So, as often in philosophy, there is little uncontroversial ground to occupy in the

territory around "phenomenal concepts".25

References

Armstrong, David. 1968. A Materialist Theory of the Mind. London: Routledge.

Aydede, Murat, and Güven Güzeldere. 2005. Cognitive architecture, concepts, and
introspection: An information-theoretic solution to the problem of phenomenal
consciousness. Noûs 39, no. 2: 197-255.
Ball, Derek. 2009. There are no phenomenal concepts. Mind 118, no. 472: 935-962.
Balog, Katalin. Forthcoming a. Acquaintance and the mind-body problem. In Identity

22
The suggestion is inspired by Burge (1979). However, as Tye notes (2009, 71) Burge himself (2003,
413-414) seems to disagree with the suggestion. McDonald (2004, sect. 3) argues against the
experience-dependence of phenomenal concepts in a way that partly overlaps with Tye and Ball.
23
See Hume (1739, book 1, part 1, sect. 1). For responses to this challenge on behalf of experience-
dependence claims, see Hume (ibid.); Cummins (1978), Morreall (1982), Fogelin (1984), Garrett
(1997, chap. 2), and Sundström (in preparation).
24
For such proposals, see for example Peacocke (2001, 242), and Papineau (2002, sect. 4.5 and 6.3;
2007, sect. 2)
25
Work on this article was funded by Riksbankens Jubileumsfond (RJ). Some supplementary support
was provided by the Wenner-Gren Foundations. Thanks to Tim Bayne, Alex Byrne, Frederique de
Vignemont and Daniel Stoljar for helpful comments and discussions.
 22

Theory, ed. Christopher Hill and Simone Gozzano. Cambridge: Cambridge

UP. Online at andromeda.rutgers.edu/~kbalog.
.———. Forthcoming b. In Defense of the Phenomenal Concept Strategy. In
Philosophy and Phenomenological Research. Online at
and rom ed a.rutgers.ed u/ ~kbalog.
———. 1999. Conceivability, possibility, and the mind-body problem. Philosophical
Review 108, no. 4: 497-528.
———. 2009. Phenomenal Concepts. In Oxford Handbook in the Philosophy of Mind,
ed. Brian McLaughlin, Ansgar Beckermann, and Sven Walter. Oxford: Oxford
UP.
Bermúdez, José, and Arnon Cahen. 2008. Nonconceptual mental content. The
Stanford Encyclopedia of Philosophy, ed. Edward Zalta. Online at
http://plato.stanford.edu/entries/content-nonconceptual/.
Block, Ned. 2007. Max Black's objection to mind-body identity. In Phenomenal
Concepts and Phenomenal Knowledge: New Essays on Consciousness and
Physicalism, ed. Torin Alter and Sven Walter, 249-306. Oxford: Oxford UP.
Braddon-Mitchell, David. 2003. Qualia and analytical conditionals. Journal of
Philosophy 100, no. 3: 111-135.
Burge, Tyler. 1979. Individualism and the mental. In Midwest Studies in Philosophy
4, ed. Peter French, Theodore Uehling, and Howard Wettstein, 73-121.
Minneapolis: University of Minnesota Press.
———. 2003. Qualia and intentional content: Reply to Block. In Reflections and
Replies: Essays on the Philosophy of Tyler Burge, ed. Martin Hahn and Bjørn
Ramberg, 405-415. Cambridge, MA: MIT Press.
Byrne, Alex. 2005. Perception and conceptual content. In Contemporary Debates in
Epistemology, ed. Ernest Sosa and Matthias Steup, 231-250. Oxford:
Blackwell.
Carruthers, Peter. 2000. Phenomenal Consciousness: A Naturalistic Theory.
Cambridge: Cambridge UP.
———. 2004. Phenomenal concepts and higher-order experiences. Philosophy and
Phenomenological Research 68, no. 2: 316-336.
Carruthers, Peter, and Benedicte Veillet. 2007. The phenomenal concept strategy.
Journal of Consciousness Studies 14, no. 9: 212-236.
Chalmers, David. 1996. The Conscious Mind: In Search of a Fundamental Theory.
 23

Oxford: Oxford UP.

———. 2003. The content and epistemology of phenomenal belief. In Consciousness:
New Philosophical Perspectives, ed. Quentin Smith and Aleksandar Jokic.
Oxford: Oxford UP.
———. 2007. Phenomenal concepts and the explanatory gap. In Phenomenal
Concepts and Phenomenal Knowledge: New Essays on Consciousness and
Physicalism, ed. Torin Alter and Sven Walter, 167-194. Oxford: Oxford UP.
———. 2010. The two-dimensional argument against physicalism. In The Character
of Consciousness, 141-191. Oxford: Oxford UP.
Cummins, Robert. 1978. The missing shade of blue. Philosophical Review 87, no. 4:
548-565.
Dennett, Daniel. 2005. Sweet Dreams: Philosophical Obstacles to a Science of
Consciousness. Cambridge, MA: MIT Press.
Diaz-Leon, Esa. Forthcoming. Can phenomenal concepts explain the epistemic gap?
Mind. Online at http://umanitoba.academia.edu/EsaDiazLeon.
———. 2008. Defending the phenomenal concept strategy. Australasian Journal of
Philosophy 86, no. 4: 597 – 610.
Dretske, Fred. 1995. Naturalizing the Mind. Cambridge, MA: MIT Press.
Fodor, Jerry. 1975. The Language of Thought. Cambridge, MA: Harvard UP.
———. 1998. Concepts: Where Cognitive Science Went Wrong. Oxford: Oxford UP.
Fogelin, Robert. 1984. Hume and the missing shade of blue. Philosophy and
Phenomenological Research 45, no. 2: 263-272.
Frege, Gottlob. 1918/1997. Thought. In The Frege Reader, ed. Michael Beaney, 325-
345. Oxford: Blackwell.
Garrett, Don. 1997. Cognition and Commitment in Hume's Philosophy. Oxford:
Oxford UP.
Goff, Philip. Forthcoming. Aposteriori physicalists get our phenomenal concepts
wrong. Australasian Journal of Philosophy. Online at
http://herts.academia.edu/PhilipGoff.
Gunther, York. 2003. Essays on Nonconceptual Content. Cambridge, MA: MIT Press.
Hawthorne, John. 2002. Advice for physicalists. Philosophical Studies 109, no. 1: 17-
52.
Hill, Christopher. 1997. Imaginability, conceivability, possibility, and the mind-body
problem. Philosophical Studies 87, no. 1: 61-85.
 24

Hill, Christopher, and Brian McLaughlin. 1999. There are fewer things in reality than
are dreamt of in Chalmers's philosophy. Philosophy and Phenomenological
Research 59, no. 2: 445-454.
Horgan, Terence, and John Tienson. 2001. Deconstructing new wave materialism. In
Physicalism and its Discontents, ed. Carl Gillett and Barry Loewer, 307-318.
Cambridge: Cambridge UP.
Hume, David. 1739/1978. A Treatise of Human Nature. Ed. Peter Nidditch. Oxford:
Oxford UP.
Jackson, Frank. 1982. Epiphenomenal qualia. Philosophical Quarterly 32, no. 127:
127-136.
———. 1986. What Mary didn't know. Journal of Philosophy 83, no. 5: 291-5.
Kaplan, David. 1989. Demonstratives: An essay on the semantics, logic, metaphysics,
and epistemology of demonstratives and other indexicals. In Themes from
Kaplan, ed. Joseph Almog, John Perry, and Howard Wettstein, 481-564.
Oxford: Oxford UP.
Kripke, Saul. 1972/1980. Naming and Necessity. Originally published in Semantics of
Natural Language, ed. Donald Davidson and Gilbert Harman, 253-355, 763-
769. Reprinted as a book with an added preface in 1980, Cambridge, MA:
Harvard UP.
Levin, Janet. 2007a. What is a phenomenal concept? In Phenomenal Concepts and
Phenomenal Knowledge: New Essays on Consciousness and Physicalism, ed.
Torin Alter and Sven Walter, 87-110. Oxford: Oxford UP.
———. 2007b. Nagel vs. Nagel on the nature of phenomenal concepts. Ratio 20, no.
3: 293–307.
———. 2008. Taking type-b materialism seriously. Mind and Language 23, no. 4:
402-425.
Levine, Joseph. 1983. Materialism and qualia: The explanatory gap. Pacific
Philosophical Quarterly 64: 354-361.
———. 2001. Purple Haze: The Puzzle of Consciousness. Oxford: Oxford UP.
———. 2007. Phenomenal concepts and the materialist constraint. In Phenomenal
Concepts and Phenomenal Knowledge: New Essays on Consciousness and
Physicalism, ed. Torin Alter and Sven Walter, 145-166. Oxford: Oxford UP.
Lewis, David. 1966. An argument for the identity theory. Journal of Philosophy 63,
no. 2: 17-25.
 25

———. 1972. Psychophysical and theoretical identifications. Australasian Journal of

Philosophy 50, no. 3: 249-58.
———. 1980. Mad pain and martian pain. In Readings in the Philosophy of
Psychology 1, ed. Ned Block, 216-232. Cambridge, MA: Harvard UP.
———.1994. Reduction of mind. In A Companion to the Philosophy of Mind, ed.
Samuel Guttenplan, 412-431. Oxford: Blackwell.
Loar, Brian. 1990/1997. Phenomenal states. Originally published in Philosophical
Perspectives 4: Action Theory and Philosophy of Mind, ed. James Tomberlin,
81-108. Atascadero, CA: Ridgeview, 1990. Reprinted in substantially revised
form in The Nature of Consciousness: Philosophical Debates, ed. Ned Block,
Owen Flanagan, and Güven Güzeldere, 597-616. Cambridge, MA: MIT Press,
1997. Page references to the 1997 version.
———. 2003. Qualia, properties, modality. Philosophical Issues 1, no. 1: 113-29.
Lycan, William. 1996. Consciousness and Experience. Cambridge, MA: MIT Press.
———. 2001. The case for phenomenal externalism. Philosophical Perspectives 15:
17-35.
McDonald, Cynthia. 2004. Mary meets Molyneux: The explanatory gap and the
individuation of phenomenal concepts. Noûs 39, no. 3: 503-524.
McGinn, Colin. 1989. Can we solve the mind-body problem? Mind 98, no. 391: 349-
66.
———. 2001. How not to solve the mind-body problem. In Physicalism and its
Discontents, ed. Carl Gillett and Barry Loewer. Cambridge: Cambridge UP.
McLaughlin, Brian. 2001. In defence of new wave materialism: A response to Horgan
and Tienson. In Physicalism and its Discontents, ed. Carl Gillett and Barry
Loewer, 319-330. Cambridge: Cambridge UP.
Melnyk, Andrew. 2002. Papineau on the Intuition of Distinctness. SWIF Philosophy
of Mind Review 4. Online at
http://www.swif.uniba.it/lei/mind/forums/melnyk.htm.
Morreall, John. 1982. Hume's missing shade of blue. Philosophy and
Phenomenological Research 42, no. 3: 407-415.
Nagel, Thomas. 1974. What is it like to be a bat? Philosophical Review 83, no. 4: 435-
50.
———. 1998. Conceiving the impossible and the mind-body problem. Philosophy 73,
no. 285: 337-52.
 26

———. 2000. The psychophysical nexus. In New Essays on the A Priori, ed. Paul
Boghossian and Christopher Peacocke. Oxford: Oxford UP.
Nida-Rümelin, Martine. 2007. Grasping phenomenal properties. In Phenomenal
Concepts and Phenomenal Knowledge: New Essays on Consciousness and
Physicalism, ed. Torin Alter and Sven Walter. Oxford: Oxford UP.
Noë, Alva. 2004. Action in Perception. Cambridge, MA: The MIT Press.
O'Regan, Kevin, and Alva Noë. 2001. A sensorimotor account of vision and visual
consciousness. Behavioral and Brain Sciences 24, no. 5: 939-973.
Papineau, David. 1993. Philosophical Naturalism. Oxford: Blackwell.
———. 2002. Thinking About Consciousness. Oxford: Oxford UP.
———. 2007. Phenomenal and perceptual concepts. In Phenomenal Concepts and
Phenomenal Knowledge: New Essays on Consciousness and Physicalism, ed.
Torin Alter and Sven Walter, 111-144. Oxford: Oxford UP.
Peacocke, Christopher. 1992. A Study of Concepts. Cambridge, MA: MIT Press.
———. 2001. Does perception have a nonconceptual content? Journal of Philosophy
98, no. 5: 239-264.
Perry, John. 2001. Knowledge, Possibility, and Consciousness. Cambridge, MA: MIT
Press.
Smart, J. J. C. 1959. Sensations and brain processes. Philosophical Review 68, no. 2:
141-56.
Stalnaker, Robert. 2002. What is it like to be a Zombie? In Conceivability and
Possibility, ed. Tamar Szabó Gendler and John Hawthorne. Oxford: Oxford
UP. Reprinted in Stalnaker,Ways a World Might Be: Metaphysical and Anti-
Metaphysical Essays, Oxford: Oxford UP, 2003, 239-252. Page references to
reprint.
Stoljar, Daniel. 2005. Physicalism and phenomenal concepts. Mind and Language 20,
no. 2: 296-302.
———. 2006. Ignorance and Imagination: The Epistemic Origin of the Problem of
Consciousness. Oxford: Oxford UP.
Sturgeon, Scott. 1994. The epistemic basis of subjectivity. Journal of Philosophy 91,
no. 5: 221-35.
———. 2000. Matters of Mind: Consciousness, Reason and Nature. London:
Routledge.
Sundström, Pär. Forthcoming. On imagism about phenomenal thought. Philosophical
 27

Review 120, no. 1, 2011.

———. In preparation. The missing shade of blue and the prospects of concept
empiricism.
———. 2008. Is the mystery an illusion? Papineau on the problem of consciousness.
Synthese 163, no. 2: 133-143.
Toribio, Josefa. 2007. Nonconceptual content. Philosophy Compass 2, no. 3: 445-460.
Tye, Michael. 1995. Ten Problems of Consciousness: A Representational Theory of
the Phenomenal Mind. Cambridge, MA: MIT Press.
———. 2000. Consciousness, Color, and Content. Cambridge, MA: MIT Press.
———. 2003. A Theory of Phenomenal Concepts. Royal Institute of Philosophy
Supplement 53: 91-105.
———. 2009. Consciousness Revisited: Materialism Without Phenomenal Concepts.
Cambridge, MA: MIT Press.
Unger, Peter. 1966. On experience and the development of the understanding.
American Philosophical Quarterly 3, no. 1: 48-56.
White, Stephen L. 1986. Curse of the qualia. Synthese 68, no. 2: 333-68.
———. 2007. Property dualism, phenomenal concepts, and the semantic premise. In
Phenomenal Concepts and Phenomenal Knowledge: New Essays on
Consciousness and Physicalism, ed. Torin Alter and Sven Walter. Oxford:
Oxford UP.