Materialism and the subjectivity of experience

AUTHOR'S PROOF JrnlID 11406_ArtID 9276_Proof# 1 - 02/09/2010 Philosophia DOI 10.1007/s11406-010-9276-3 1 3 2 4 Reinaldo J. Bernal Velásquez 5 F Materialism and the Subjectivity of Experience O O Received: 10 August 2010 / Accepted: 15 August 2010 # Springer Science+Business Media B.V. 2010 C O R R EC TE D PR Abstract The phenomenal properties of conscious mental states happen to be exclusively accessible from the first-person perspective. Consequently, some philosophers consider their existence to be incompatible with materialist metaphysics. In this paper I criticise one particular argument that is based on the idea that for something to be real it must (at least in principle) be accessible from an intersubjective perspective. I argue that the exclusively subjective access to phenomenal contents can be explained by the very particular nature of the epistemological relation holding between a subject and his own mental states. Accordingly, this subjectivity does not compel us to deny the possibility that phenomenal contents are ontologically objective properties. First, I present the general form of the argument that I will discuss. Second, I show that this argument makes use of a criterion of reality that is not applicable to the case of subjective experience. Third, I discuss a plausible objection and give an argument for rejecting observation models of self-knowledge of phenomenal contents. These models fall prey to the homunculus illusion. U N Keywords Consciousness . Explanatory gap . Materialism . Subjectivity . Metaphysics of mind 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 Introduction 27 Kripke (1972) famously claimed that every identity statement between terms that are rigid designators is, if true, a necessary truth. Therefore, if psycho-physical 28 29 R. J. Bernal Velásquez (*) Institut Jean-Nicod, Pavillon Jardin, École Normale Supérieure, 29 rue d’Ulm, 75005 Paris, France e-mail: reynaldobernalv@yahoo.com R. J. Bernal Velásquez Département de Philosophie, UFR de Sciences humaines, Bât. ARSH, UPMF (Grenoble II), 1281 avenue Centrale, domaine universitaire, 38400 Saint-Martin d’Hères, France Q1 AUTHOR'S PROOF JrnlID 11406_ArtID 9276_Proof# 1 - 02/09/2010 Philosophia F statements (PPSs) of the sort “pain is the firing of C-fibers” are true at all, they are necessarily true.1 However, PPSs seem to be contingent: e.g., we (may) have the intuition that there is a possible world where there is pain and yet no C-fibers firing. Levine (1983) argued that the appearance of contingency of PPSs is due to an ‘explanatory gap’: they don’t explain why, e.g., pain feels the way it does, or even why there should be any feeling associated with C-fibers firings at all. My thesis is that the explanatory gap seems unbridgeable, at least in part, because we may, explicitly or implicitly, take the right-hand term of a PPS as referring to an ontologically objective property and the left-hand one as referring to an ontologically subjective one. That this is often so can be illustrated with the following cases. 30 31 32 33 34 35 36 37 38 39 40 Searle (2002) explicitly states that consciousness is ontologically subjective. In terms of a PPS, C-fiber firing would be an ontologically objective physical phenomenon, while pain is an ontologically subjective one. This raises the question, how something can be real and caused by biological systems—as Searle claims conscious experience is—and yet be ontologically subjective?2 Chalmers (1996) claims that there is a metaphysically possible world where every objective fact is the same as in the actual world and yet there is no (subjective) experience. In terms of a PPS, a world where there is (say) C-fiber firing but no pain. Humphrey (2006) says that PPSs only establish correlations, which could not turn out to be identities. The reason is that the terms in both sides of the PPS do not share the same ‘dimension’. In terms of PPSs, I take this to mean that while C-fiber firings are objective physical phenomena, occurrences of pain are not. 41 42 43 44 45 46 47 48 49 50 51 52 53 54 In this paper I present an argument against the claim that PPSs cannot express metaphysical identities. I will try to show that phenomenal properties could be (or supervene on) physical properties despite their subjective character. The main point is that the subjectivity of conscious experience is due to an epistemological condition, which does not entail ontological subjectivity (whatever this could mean). If ‘pain’ is a rigid designator, and pain can be (or supervene on) physical properties despite their subjective character, the explanatory gap in PPSs will seem less insurmountable. First, I will present one intuitive argument against the physical reality of phenomenal properties of conscious mental states. The argument is based on the idea that for something to be real it must (at least in principle) be accessible from an intersubjective perspective, which is not the case for phenomenal properties. Second, I will show that this argument makes use of a criterion of reality that is not applicable in the case of subjective experience. Finally, I will develop my position by addressing a plausible objection. Thereby, I will reject observation models of selfknowledge of phenomenal properties. 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 D U N C O R R EC – TE – PR O O – 1 Kripke (1972) argues that both “pain” and “firing of C-fibers” are rigid designators. Indeed, it is not easy to give a positive account of ‘ontological subjectivity’. Defining it as the condition of being real, but only from a subjective point of view, falls short. 2 AUTHOR'S PROOF JrnlID 11406_ArtID 9276_Proof# 1 - 02/09/2010 Philosophia 71 According to Kim (2005, p.30): “It may well be that the [mind-body] problem is an inexorable consequence of the tension between the objective world of physical existence and the subjective world of experience”. In fact, a tension arises with the attempt to reconcile the following statements. 72 73 74 75 (a) Every real entity and property has a physical nature (materialist monism3): Every property is or supervenes on a physical property, i.e., if every physical property4 is fixed, all the remaining properties—psychological, social, etc.—are fixed as well. (b) Phenomenal properties of conscious mental states are exclusively accessible from a subjective perspective.5 76 77 78 79 80 81 O F Subjectivity and Objectivity 82 83 (1) Phenomenal properties are real properties [supposition]. (2) Given (a), (1) implies that phenomenal properties have a physical nature, i.e., they are physical properties or supervene on physical properties [from (a) and (1)] (3) Physical entities and properties have—by definition—an objective reality [premise]. (4) Every objective entity or property is—at least in principle—accessible from an intersubjective perspective [supposition]. (5) If phenomenal properties are real properties, they must be accessible from an intersubjective perspective [conclusion from (1) to (4)]. (6) However, given (b), phenomenal properties are exclusively accessible from a subjective perspective. 84 85 86 87 88 89 90 91 92 93 94 95 R R EC TE D PR O The tension between (a) and (b) can be illustrated by the following line of reasoning. O How to resolve the tension illustrated by statements (5) and (6)? One may: Deny (a): propose some kind of dualism, idealism or panpsychism.6 Deny (b): claim that phenomenal properties are not exclusively accessible from the subjective perspective. iii) Deny (1): claim that phenomenal properties are not real properties (eliminativism). iv) Deny (4): claim that some entities or properties may be (objectively) real and yet not be accessible from an intersubjective perspective. U N C i) ii) 3 I use ‘materialism’ instead of ‘physicalism’, for the latter is often taken to be the stronger thesis that every real phenomenon can be captured in terms of the language of the natural sciences. 4 ‘Every physical property’ includes not only microphysical properties, but properties at every level. It is an open question whether macroscopic properties logically supervene on microscopic ones, which I would answer on the negative. Cf. Kistler (2010). 5 I will only be concerned with conscious phenomenal properties. Rosenthal (2005) claims there are unconscious ones. Note that statement (b) plays a key role in the famous “knowledge argument” (Jackson 1986): Mary, before leaving the black-and-white room, is said not to know the phenomenal property ‘red’ (what it is like to see red), since she has never had the experience of seeing red. To have had the experience of seeing red is tantamount to having had subjective access to the phenomenal property ‘red’. 6 Note that G. Strawson (2006) develops a version of panpsychism that he claims is compatible with physicalism and hence with (a). 96 97 98 99 100 101 102 AUTHOR'S PROOF JrnlID 11406_ArtID 9276_Proof# 1 - 02/09/2010 Philosophia I shall accept both (a) and (b) without argument. I take every property to be (or supervene on) a physical property of our (material) world, and I take it that phenomenal properties are only accessible from the subjective perspective. In fact, even if mental contents can be reduced in behaviourist, functional, or physical terms, phenomenal properties of conscious mental states are exclusively accessible from the subjective point of view (Nagel 1974).7 Accordingly, with respect to (5), we are left with the following disjunction: 103 104 105 106 107 108 109 (iii) Phenomenal properties are not real properties. 110 or 111 112 113 114 Prima facie, (iv) seems implausible. Accepting the first disjunct may appear to be the better option. However, in an attempt to reconcile materialism with realism about phenomenal properties, I will explore the possibility of accepting the second disjunct. My thesis is that, concerning phenomenal contents, the entailment from the epistemological condition of being exclusively accessible from the subjective perspective to the metaphysical one of not being physical (or physically supervenient) is invalid. I will argue that phenomenal contents are only accessible from the subjective perspective in virtue of the exclusivity of the relation that each subject has to his own mental states. 115 116 117 118 119 120 121 122 123 124 The Intersubjective-availability Criterion of Reality 125 R R EC TE D PR O O F (iv) Phenomenal properties are exclusively accessible from the subjective perspective and, nonetheless, they are physical properties or supervene on physical properties. U N C O Whenever different subjects are involved in an epistemological relation with the same entity X, they have good reason to believe in the objective reality of this entity. It is unreasonable to deny the existence of, say, the moon, since every subject (under suitable conditions) can see it in the sky. But do we only have epistemic reason to believe in the objective reality of some entity or property X when it is possible (in principle) to have intersubjective epistemic access to it?8 7 Even though some structural characteristics of ‘phenomenal spaces’ can be inferred from behavioural analyses, this does not contradict the statement that phenomenal contents are only accessible from the subjective perspective. Certainly, if phenomenal properties are physical properties, they are accessible in the same sense that the property ‘mass’ or ‘electric charge’ is accessible. But besides this third person point of view type of access, I will claim that there is another type of epistemological relation a subject can establish with—and only with—his own mental states, which is the one that presents phenomenal properties as phenomenal properties. 8 Scientific realists and antirealists disagree about the existence of unobservables. However, the problem posed by unobservables in the realism/anti-realism debate is orthogonal to the present one. There, unobservability is not related to the exclusivity of a subjective access. Were our senses powerful enough, we would have intersubjective access to, e.g., a microscopic particle. Nevertheless, it can be claimed that, from the scientific point of view, phenomenal properties are to be considered sui generis unobservable properties. 126 127 128 129 130 131 132 Q2 AUTHOR'S PROOF JrnlID 11406_ArtID 9276_Proof# 1 - 02/09/2010 Philosophia U N C O R R EC TE D PR O O F Certainly, if there is an entity or instantiated property X in a given place at a given time, it is possible (in principle) for different subjects to establish an epistemological relation to it. Accordingly, if a subject S claims he perceives an X, but no one else can perceive this putative something, there is good reason to suppose that the X does not exist (i.e., has no objective reality) and believe instead that S is hallucinating an X or misperceiving something (taking a Y for an X). Therefore, it seems right to state that we only have epistemic reason to believe in the objective reality of some entity or property X when it is possible (in principle) to have intersubjective epistemic access to it. Let us then adopt the following ‘intersubjective accessibility’ criterion of reality: (IACR) If a putative entity X is not intersubjectively accessible, there is no epistemic reason to believe in its objective reality. Now, are there any restrictions on the applicability of this criterion? For the IACR to be applied to a putative X, we need the different subjects to be in equivalent epistemic conditions to establish an epistemic relation with X. This seems to be an uncontroversial requirement, a restriction, for the rational applicability of the criterion. In fact, it makes no sense to demand intersubjective access to an X if it is not possible for every subject to be properly placed or equipped to access X.9 Now, if the putative X happens to be a property of a mental state of a subject S, there is clearly an asymmetry between the kinds of epistemic relations that S, on the one hand, and other subjects, on the other hand, may establish to X. Think about self-knowledge of the position of my limbs. I know it in a way that is exclusive to me. I do not have ‘to look’ at my hand to know where it is, in contrast to the mechanism others subjects may use to find this out.10 Mental states being embodied, they hold an exclusive kind of relation with the embodying system.11 Because of this epistemological asymmetry between the kinds of epistemic relations that a given subject S, on the one hand, and other subjects, on the other hand, may establish to S’s mental states, the IACR should not be applied in the case of phenomenal properties of conscious mental states. It is not just the case that phenomenal properties happen to fail the IACR; phenomenal properties are out of the scope of the IACR. Accordingly, the IACR entailment from the epistemological condition of being exclusively accessible from the subjective perspective to the metaphysical one of not being objectively real is invalid when applied to phenomenal contents. To argue that phenomenal properties are not real, but fictions or illusions, a different criterion of reality is required. I will continue by discussing a possible reply by the eliminativist: Self-knowledge of phenomenal properties is not (they claim) epistemically well grounded. This is best explained by the claim that there are no such properties. 9 Note that even if a subject, in order to be able to establish an epistemological relation with an X, may require special equipment and skills, it is (in principle) possible for him to meet these requirements. 10 I am indebted to Max Kistler for the analogy between self-knowledge of mental states and selfknowledge of body positions. 11 For mental states to be ‘embodied’ does not mean ‘within’ or ‘contained in’ a body. It means that they are (for materialists) physical states belonging to the body they are part of. 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 AUTHOR'S PROOF JrnlID 11406_ArtID 9276_Proof# 1 - 02/09/2010 Philosophia 172 Consider the following objection: “Certainly, if conscious mental states had phenomenal properties, a subject S would occupy a particular and exclusive epistemic perspective towards these properties of his mental states. However, the exclusive subjective character of the epistemic access to these putative properties should not preclude the possibility for each subject of observing, analysing and describing them in a satisfactory manner. In fact, our observation-based knowledge is always supported by subjective epistemic access. Each of us has a particular and exclusive perspective on the world, from which he relates to objective reality. Despite this, we can agree about the description we give of, say, a table, the moon or a brain. Now, psychology teaches us that no satisfactory descriptions of phenomenal contents can be given. Several well-known experiments show that we are usually wrong in our judgements about the phenomenal contents of our own conscious mental states.”12 Eliminativism takes the alleged impossibility of describing phenomenal contents in a consistent manner to be due to the non-existence of such properties13 and not to very exceptional epistemological circumstances. My reply to this objection includes three points. First, psychological studies certainly show that the descriptions that one may give of the phenomenal contents of one’s conscious experiences are not to be (uncritically) trusted. But this is no argument against the metaphysical claim that phenomenal contents may be objectively real. The ontological objectivity of some entity does not depend on our ability to achieve agreement on a description for it. Second, the conceptual and methodological difficulties (perhaps insurmountable) one faces when trying to describe (putatively real) phenomenal contents can be accounted for precisely in virtue of the very particular epistemological relation that subjects have to their own conscious mental states. Moreover, if phenomenal contents happen to be non-conceptual contents14 there are reasons to expect phenomenal knowledge not to be fully translatable in conceptual terms. Finally, and more importantly, the objection rests on the supposition that the relation between subjects and (the putative) phenomenal contents is analogous, or similar enough, to the relation between subjects and objects like chairs, the moon or brains. I believe this supposition is mistaken because—among other problems—it seems motivated by the homunculus illusion: the idea that there is the self on one side, and his mental states on the other, and that the former somehow ‘observes’ the latter (with the ‘mind’s eye’?). I will conclude by developing this last point through a brief critique of observation models of self-knowledge of conscious mental states. 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 U N C O R R EC TE D PR O O F The Ontological Deflationism Alternative 12 The interpretation of the experiments used by eliminativists to support the claim that no satisfactory descriptions of phenomenal contents can be given is controversial. For the sake of the argument I will accept that in fact we are usually misled or laden by folk psychology concerning phenomenological descriptions. 13 For instance, Dennett (1991, chap. 11) seems to argue in this direction. 14 This is indeed the position I will defend in forthcoming work. AUTHOR'S PROOF JrnlID 11406_ArtID 9276_Proof# 1 - 02/09/2010 Philosophia 211 It is quite common to find in the literature expressions of the form “I have access to mental state x”—therefore x is conscious—and “I don’t have access to mental state y”—therefore y is unconscious. This way of talking can mislead us into thinking that there is something ‘there’ with this strange property of being only accessible to me. Moreover, one could reach the conclusion that phenomenal properties (if there are such things) are ontologically subjective. This is the view I am criticizing. Think about the expression “I have access to some of my mental states”. What does this ‘I’ amount to? If we put to one side all the mental states (conscious and unconscious) a person is in, what is supposed to remain in the other side as the ‘I’ which does or does not have ‘access’ to the mental states?15 I consider any account of self-knowledge that presents subjects as having access to some mental states—in any ordinary sense of the term ‘access’—to be inappropriate. I will call these accounts ‘observation models’ of self-knowledge of conscious mental states. Certainly, in a sense we can say that there is ‘access’ to the phenomenal contents of our conscious mental states. By means of self-knowledge abilities, somehow, we can reflect on and describe our experiences. However, this should not mislead us to consider that there is a subject-object relation between the subject and his experiences. To be sure, there is no experience without an experiencer, and there is no experiencer without experiences; experiencer and experiences are not detachable.16 By contrast, in an ordinary subject-object epistemological relation the relata are detachable.17 This is a substantial difference. 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 EC TE D PR O O F Psychological Subjects and the Unity of Consciousness Psychological Subjects 15 U N C O R R It is characteristic of persons that they have minds. Each mind, at a given moment, can be identified with the totality of the person’s mental states (or mental process stages) occurring at that moment.18 I will call this bunch of mental states ‘the psychological subject’. This concept is not supposed to capture what a person is, or what ‘the self’ amounts to. It is only supposed to be useful for the purposes at hand: Daniel Dennett (2005) would say that, precisely, there is no ‘I’. There is no homunculus enjoying the “Cartesian Theatre” show. I agree with this claim, but for different reasons. First, I take conscious mental states to have the non-illusory property that there is something it is like to be in those (somehow bound) states. Second, Dennett also talks as if there were a subject who would or not have access to some mental states or properties of mental states: “ […] you do not “have access to” the intrinsic qualities of your own experiences in any interesting sense, any more than outside observers do. You have access only to the relations between them that you can detect” (2005, p. 81, italics in the original). 16 In G. Strawson (2009, p.59) words: “There can’t be experience without a subject of experience simply because experience is necessarily experience for—for someone-or-something. Experience necessarily involves experiential ‘what it-is-likeness,’ and experiential what-it-is-likeness is necessarily what-it-islikeness for someone-or-something. Whatever the correct account of the nature of this experiencing something, its existence cannot be denied” (italics in the original). 17 Even when you are looking at yourself in a mirror, subject and object are detachable; you may not know that it is you that you are looking at. 18 A mind can be considered as the complete history of someone’s mental states, or as a (wide enough) time-slice of this history. In discussing conscious experience, I am using the second sense. 233 234 235 236 237 238 AUTHOR'S PROOF JrnlID 11406_ArtID 9276_Proof# 1 - 02/09/2010 Philosophia 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 The Unity of Consciousness 270 U N C O R R EC TE D PR O O F clarifying the nature of the relation between the mind and the experiences of a person.19 Now, we can classify mental states into conscious and unconscious ones. Certainly, this classification can be made using different criteria. We may use functional criteria and talk of ‘access consciousness’ or experience-based criteria and talk of ‘phenomenal consciousness’. And we may distinguish between a creature being conscious and a mental state being conscious. Here, I am using the term ‘conscious’ in the phenomenal sense and to characterise some mental states. Following Nagel (1974), I take it that a mental state is conscious if and only if there is something it is like to be in that state for the relevant creature. Since ex hypothesi the psychological subject is constituted, at any given time, by the totality of the person’s mental states, whenever a person is experiencing she is in some conscious mental states (as well as the unconscious ones) which are a constitutive part of the corresponding psychological subject. Likewise, whenever a person engages in self-knowledge activities (however they might proceed), since these are psychological activities, they involve mental states that are also constitutive parts of the psychological subject. Therefore, both self-knowledge activities and the phenomenal properties of conscious mental states belong to the same psychological subject. In this sense, the relation between the experiencing subject and the experiences he might know about is an internal one. This is certainly a trivial point, but it has epistemic implications that may be overlooked. To say that the experiencer-experience relation is an internal one does not just mean that each person has a privileged first person perspective towards her own conscious mind, like the one you would have towards an external object that you and only you (for whatever reason) could perceive. This is a metaphysical claim: the mind and the totality of mental states (some of which can be self-known) are one and the same psychological entity. Therefore, the subjective character of phenomenal properties is an epistemological condition that can be explained by the fact that the experiences are constitutive parts of the experiencer.20 Someone’s experiences cannot be intersubjectively accessible in the same way that they are subjectively accessible. It can be replied that the ‘psychological subject’ is a composed system that can be divided in such a way that some part of it (some mental states) would have epistemological access to another part of it (other mental states). When I say “I have access to some of my mental states”, these mental states do not correspond to the whole psychological subject. Some mental states could have access to other mental states, such that the ‘I’ would be constituted by the former bunch of states. There are two options when making the proposed division. First, suppose we put to one side all the conscious mental states. In that case, there remains no conscious 19 If materialism is true—and I have supposed that it is—the psychological subject supervenes on some physical system undergoing some neural processes. 20 Note that consequently if the experiencer is real, so are the phenomenal properties of his conscious mental states. 271 272 273 274 275 276 277 278 AUTHOR'S PROOF JrnlID 11406_ArtID 9276_Proof# 1 - 02/09/2010 Philosophia 21 U N C O R R EC TE D PR O O F ‘I’ that would have ‘access’ to some of his mental states. We would have an epistemological subject lacking any conscious mental states. This is quite problematic; for one thing, it is clear that epistemological activities—including self-knowledge—involve paradigmatically conscious mental states on the side of the subject.21 Second, suppose we separate only some conscious mental states, in order for the ‘I’ to remain a conscious psychological entity. In this case, there would be conscious mental states towards which we have no access (since they belong to the ‘I’). This option also seems unacceptable. Even worse: we would have two conscious minds (two bunches of mental states) belonging to a single psychological subject.22 In fact, one of the most salient characteristics of consciousness is its unity, as Kant claimed.23 We experience our conscious mind as a single mind, even though we can be in several conscious mental states at the same time (i.e., when conscious, we enjoy different experiences simultaneously). Every attempt to model self-knowledge of conscious mental states under a subject-object schema, where a conscious subject accesses some (conscious) mental states, seems committed to violate this unity. It can be replied that the former argument supposes the conscious character of a mental state to be an intrinsic feature of the state,24 while it can result from a relational property holding between different states. In fact, Higher Order Thought (HOT) theories of consciousness, like Rosenthal’s (2005) and Carruthers’ (2005), claim that the ‘what is it likeness’, i.e., the phenomenal experience, results from the actual formation (Rosenthal) or the potential possibility of forming (Carruthers) a HOT which represents a lower level thought. An [n]-level thought becomes conscious (and reveals its phenomenal qualities) when it is (or can be) the object of an [n+1]-level thought. Introspection abilities would result from our capacity to be (or produce) HOTs. This view could be used to claim that, after all, we have access to some of our mental states under a subject-object model: the HOTs (the ‘subject’) represent the lower order mental states (the ‘object’) they access. But notice, firstly, that the relation between higher and lower order mental states remains an internal one, where the higher order thought logically requires the existence of the lower order one. They are not detachable, by contrast with what happens with an ordinary subject-object knowledge relation. The signalled dissimilarity between self-knowledge of conscious mental states and observational It could be argued that conscious mental states are not necessary for the epistemological activity of a (human) subject obtaining perceptual knowledge. For instance, subjects who suffer blindsight claim to be completely blind regarding some part of their visual field. However, experiments show that there is some perception of the items belonging to the blindsight field: when asked to ‘guess’ about some features of these items, subjects are well above chance for the correctness of their answers. But does this kind of unconscious perception gives rise to knowledge? I think it does not: these perceptual contents do not produce (or amount to) beliefs, since the subjects think they are randomly guessing; neither are these contents used in appropriate ways (for instance, the subjects do not spontaneously grasp a blindsighted object). 22 Note that even if the subset of conscious mental states corresponding to the ‘I’ was not permanent, for the mental states constituting the ‘I’ could change, each mental state while belonging to the ‘I’ would still be conscious and nevertheless inaccessible. 23 I am indebted to Conor McHugh for the suggestion of this point. 24 Indeed, this is the thesis I will defend in forthcoming work: the property for a mental state being conscious, i.e., that there is something it is like to be in that state, is an intrinsic (and emergent) property of a complex physical state. 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 AUTHOR'S PROOF JrnlID 11406_ArtID 9276_Proof# 1 - 02/09/2010 Philosophia 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 R R EC TE D PR O O F knowledge remains. Secondly, HOT theories may account for access consciousness, but they face serious difficulties in accounting for phenomenal consciousness. In fact, it is mysterious why phenomenology would appear in virtue of the (actual or potential) formation of HOTs. Why would it not be enough to be in first order mental states with phenomenal properties? Why would an intentional relation between first order and second order mental states render the first order one phenomenally conscious?25 Furthermore, it remains unclear how the different conscious mental states are bounded to result in the unity of consciousness. Now, since HOT theories propose an explanation for phenomenal consciousness, when we reject this explanation, the subject-object model they provide loses its grounds. We plainly can reflect upon and describe our experiences, and the question of how we acquire self-knowledge of them is certainly difficult. Moreover, it is controversial whether self-knowledge, and in particular self-knowledge concerning phenomenal qualities, is a genuine kind of knowledge and thus deserves this label (Cf. Alter & Walter 2009). It is beyond my purposes here to present or defend some positive account of self-knowledge. I just want to say that, however self-knowledge of phenomenal contents may work, an observational model—with the ‘observation’ turned ‘inward’—is implausible. Self-knowledge is reflexive in a strong sense: the object is part of the subject. In short, talk of one having ‘access’ to some of one’s mental states is misguided. Whatever the nature of the epistemological relation between a subject and his conscious mental states is, it is an internal one in the sense of obtaining within a single psychological subject. The psychological subject does not ‘access’ in any ordinary sense of the term some of his mental states, for they are part of what he is. To think otherwise is to fall prey to the homunculus illusion. O Conclusion U N C I have argued that because the knowledge one has of the (putative) phenomenal contents of one’s conscious mental states is of a very particular kind, the inferences from epistemological considerations to metaphysical conclusions we usually make when analysing ordinary subject-object epistemological relations are no longer sound. In particular, the application of the IACR to the question of the objective reality of phenomenal properties is invalid: the exclusive subjectivity of the access one has to the phenomenal properties of one’s conscious mental states is no conclusive reason to deny the possibility of these properties being (or supervening on) physical properties. I did not give an argument for the claim that phenomenal properties of conscious mental states are ontologically objective. I only tried to reconcile this possibility with their subjective character. By criticising the observation model of self-knowledge, I tried to give support to the idea that the experiencer and his experiences are not detachable. This, in turn, could explain why experiences have a subjective character. 25 Moreover, it seems plausible to attribute conscious experience to some non-human animals, despite their lack of metarepresentational capacities, against what HOTs entail. 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 AUTHOR'S PROOF JrnlID 11406_ArtID 9276_Proof# 1 - 02/09/2010 Philosophia O O F Note that I did not simply rely on the distinction between two different epistemic perspectives, the so-called first-person and third-person points of view (McGinn 1989), to account for the subjective character of phenomenal properties. The first person and the third person points of view are not independent epistemological perspectives (cf. Davidson 2001). But the metaphysical distinction between being a constitutive part of a psychological subject and being another psychological subject is between exclusive conditions. I believe that the latter distinction—the metaphysical one—has deeper epistemological consequences than the former, as my argument was intended to show. Given materialist monism, if phenomenal contents are real objective properties, they are (or supervene on) physical properties of (in the case of earth life) some neurological processes. Now comes the hardest question: how is it possible for a material being to be such that there is something it is like to be that being? D PR Acknowledgments I am indebted to Max Kistler, Joëlle Proust, Conor McHugh, Kirk Michaelian, Santiago Echeverry, the participants of the “Carnap lectures 2010” workshop (Bochum-Germany), the participants of the “3rd CEU Philosophy Graduate Conference” (Budapest-Hungary), and an anonymous reviewer of Philosophia for very useful and enlightening discussions and comments. TE References U N C O R R EC Alter, T., & Walter, S. (Eds.). (2009). Phenomenal concepts and phenomenal knowledge. Oxford: Oxford University Press. Block, N. (2007). Consciousness, function, and representation. Cambridge: MIT Press. Carruthers, P. (2005). Consciousness. Oxford: Oxford University Press. Chalmers, D. (1996). The conscious mind. Oxford: Oxford University Press. Davidson, D. (2001). Subjective, intersubjective, objective. Oxford: Oxford University Press. Dennett, D. (1991). Consciousness explained. Boston: Little, Brown and Co. Dennett, D. (2005). Sweet dreams. Cambridge: MIT Press. Humphrey, N. (2006). Seeing red. Cambridge: The Belknap Press of Harvard University Press. Jackson, F. (1986). What Mary didn’t know. Journal of Philosophy, 83, 291–295. Kim, J. (2005). Physicalism, or something near enough. Princeton: Princeton University Press. Kistler, M. (2010). Réduction et émergence. Forthcoming. Kripke, S. (1972). Naming and necessity. Cambridge: Harvard University Press. Levine, J. (1983). Materialism and qualia: the explanatory gap. Pacific Philosophical Quarterly, 64, 354–361. McGinn, C. (1989). Can we solve the mind-body problem? Mind, 98, 349–366. Rosenthal, D. (2005). Consciousness and mind. Oxford: Oxford University Press. Searle, J. (2002). Consciousness and language. Cambridge: Cambridge University Press. Strawson, G. (2006). Consciousness and its place in nature. Exeter: Imprint Academic. Strawson, G. (2009). On the sesmet theory of subjectivity. In D. Skrbina (Ed.), Mind that abides (pp. 57–64). Amsterdam: John Benjamins Publishing Company. 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 Q3