Psyche 7(3): Have We Neglected Phenomenal Consciousness? by William G. Lycan

	Charles Siewert's book, The Significance of Consciousness, can be purchased from Amazon.Com

Have We Neglected Phenomenal Consciousness?

William G. Lycan
Philosophy
University of North Carolina
Chapel Hill, NC 27599-3125
U.S.A.

(919) 962-7291

ujanel@email.unc.edu

PSYCHE, 7(03), March 2001
http://psyche.cs.monash.edu.au/v7/psyche-7-03-lycan.html

KEYWORDS: Consciousness, experience, introspection, phenomenal character, qualia, "what it's like."

COMMENTARY ON: Charles Siewert. (1999). The Significance of Consciousness. Princeton University Press. 392 pp. ISBN: 0691027242. Price: $US 42.50 hbk.

ABSTRACT: Charles Siewert's The Significance of Consciousness contends that most philosophers and psychologists who have written about "consciousness" have neglected a crucial type or aspect that Siewert calls "phenomenal consciousness" and tries carefully to define. The present article argues that some philosophers, at least, have not neglected phenomenal consciousness and have offered tenable theories of it.

1. Introduction

Charles Siewert has produced a work of large scope and immense intellectual energy. At the same time, his approach is careful and painstaking. If you feel a tension between those two characterizations, you should read the book.

It contains much good stuff. My own favorite is Chapter 7, in which Siewert carries out a sort of ontological assay of visual experience and its intentionality, with fruitful attention to nuances of phenomenology. (I very much like Siewert's general approach to intentionality, understanding it in terms of assessability for truth or accuracy.<1>) And Chapter 8 is the best discussion I know of how nonsensory thinking can have a phenomenology without involving imagery.

But the book's main concern is "phenomenal consciousness" and its neglect by some philosophers of mind. (With characteristic irony, Siewert is using the word "neglect" roughly in its neurological sense, as in hemi-neglect and other such clinical deficits.) "Consciousness ... tends to get lost in certain writings one might have presumed were devoted to it" (p. 151). Siewert insists throughout on respect for the first-person perspective; the kind of consciousness he means is, above all, knowable from that perspective.

2. Characterizing Phi-Consciousness

To Siewert's credit, he recognizes that by now the "c"-word is completely up for grabs. Even with the modifier "phenomenal," there are several different things it not only could mean but has actually been used to mean. In the literature of the past fifteen years, finer distinctions have had to be made. And the unreflective use of expressions like "subjectivity," "qualia" and "what it's like," as if they had single meanings and everyone knew those meanings, has concretely harmed the dialectic.

Accordingly, Siewert sets out very painstakingly to stipulate what he is going to mean by "phenomenal consciousness." (Hereafter I shall abbreviate it as "Phi-consciousness," to mark it as his term of art and to keep our readers from reinterpreting it according to their own entrenched linguistic dispositions with respect to the hot-button words "phenomenal" and "consciousness.") Siewert explains the term primarily in two ways.

The First Way is positively: he stipulates that Phi-consciousness is what is shared by silent speech, imagery and sense-experience.

The Second Way is negatively, in terms of the sort of superblindsight cases made familiar by Ned Block (1995),<2> the fictional descendants of Lawrence Weiskrantz' famous patient D.B. Siewert's character Belinda (p. 105) is perhaps the most useful. Belinda sees nothing in the left half of her visual field,<3> but, amazingly, she can make fairly reliable unprompted verbal discriminations of color, shape, position etc. on her left side. (As with D.B., obviously some visual information is getting through, but it does not present itself to her subjectively in the form of sight.) We may suppose, Siewert adds later on, that for that matter Belinda's visual(-type) information is accompanied by a Rosenthal-style higher-order thought; intuitively, that feature can be added without making Belinda's achievement seem any more like normal visual experience. Siewert contrasts Belinda with Connie, whose sight is normal though very poor and who makes exactly the same visual-type discriminations, but who does so in the normal way, by seeing.<4> Phi-consciousness is what Belinda distinctively lacks with respect to vision on her left side. Accordingly, a theory of mind according to which there could not conceivably be such a subject as Belinda is a theory that denies the existence of Phi-consciousness in Siewert's sense.

But at this point there is a crucial shortfall. The trouble is that even working together, those two pincers, the First and Second Ways, do not pin down a unique sense for his special term "phenomenal [Phi-]consciousness." On the positive side, there are several things that are shared by conscious silent speech, imagery and sense-experience: having intentional content; having in some sense qualitative content (which may or may not differ from the intentional content); the subject's awareness of the goings-on; and whatever overlap there is in causal or functional role. (Siewert may say that the last of those is not first-person-knowable, which would disqualify it from helping to characterize Phi-consciousness; but I myself think some of it is first-person-knowable.) Those features are mutually independent, conceptually speaking, and sometimes they come apart in the real world as well.

The blindsight via negativa helps a bit, but not entirely. The Belinda case seems to indicate that awareness is required for Phi-consciousness, since visual awareness and/or awareness of her visual information is what she conspicuously lacks, and that what the awareness is of should be qualitative character or content in some sense. But so far, that is all we've got to go on. (Note that experiences differ along each of the latter dimensions -- i.e., two experiences can share the same style and degree of awareness but differ in qualitative content, and two experiences can have much the same qualitative content but differ in style or degree of awareness.)

The book does offer a third possible handle on "Phi-consciousness." Siewert defines a sense of the term "phenomenal character."

We might identify the sort of phenomenal character we supposed missing by saying that it does not seem to the blindsighter the way it seems to us normally sighted people, for it to look to us as if there is an X on our left. So I might distinguish between my [merely] visually judging that there is an X, and its seeming to me the way it does for it to look to me as if there is an X.... [T]he way it seems to me for it to look as if there is an X on my left is the phenomenal character of the experience I would report by saying, "It seems to me as it does for it to look as if ... etc." (p. 86, italics original).

Now, is it true for Siewert that an experience is Phi-conscious if and only if it has "phenomenal character" in the sense defined? The right-to-left half of this biconditional is affirmed on p. 99, where Siewert also adds, "whatever state differs from another in phenomenal character, is a conscious experience." (And in correspondence, Siewert has endorsed the entire biconditional.) So we can add, helpfully, that an experience is Phi-conscious when and only when it involves: a way of seeming to its subject for it to look/sound/smell/feel/ ... as if [such-and-such a qualitative character is present]." (Some readers will see the phrase "way of seeming" and with a rush of happy recognition translate it directly as "what it's like." I have the impression Siewert would resist this, and I believe he would be right to. The dreaded phrase is at least two ways ambiguous, and in the past ten or fifteen years its use has certainly caused more trouble and confusion than clarification or assistance.<5>) Siewert goes on explicitly to tie the notion of Phi-consciousness to that of differences in ways of seeming:

Finally -- and here is the crucial point -- not only does the way it seems to you typically differ when you consciously see differently colored things, so that one has different kinds of conscious experience, but also any state one is in, or event in one's brain, that differed from others just as these experiences differ among themselves, would be a conscious experience. (p. 87)

So let us count the appeal to such ways of seeming as a Third Way of pinpointing Siewert's notion of "phenomenal consciousness."

Are there experiences that are not Phi-conscious? The foregoing biconditional suggests there may be. It is not obviously contradictory to speak of an "unconscious experience," though I don't talk that way myself. But for brevity, in which follows, I'm going to use "experience" to mean conscious experience and trust that for Siewert this will entail Phi-consciousness. I don't think much hangs on this, so I doubt Siewert will object.

3. Phi-Consciousness and AQ Consciousness

If I were to try to say what "Phi-consciousness" is, in terms more familiar to me, my best guess is that it is: being internally, introspectively or at any rate first-person-ishly, and directly, aware of either a specifically sensory quality or a qualitative character more broadly speaking. (Siewert holds that there is nonsensory Phi-consciousness.) My best guess is not perfect; there is at least a little textual evidence against that interpretation. Also, Siewert deliberately shies away from current philosophy-of-mind jargon such as "introspectively aware," "qualitative character," "qualia," "content," etc., for fear of contamination by tendentious philosophical background assumptions. I very well appreciate that, because in this area there is constant danger of such contamination; but, unfortunately, it leaves the reader without resources for speculating about what Siewert does further intend to require of "Phi-consciousness" and not to require of it. (Also, as I said earlier, in Chapter 8 Siewert makes the very interesting suggestion that thoughts can have phenomenal character, even if they do not involve either imagery or emotion. But -- assuming the Third Way -- that makes it sound as though just being aware of the thoughts suffices for their being Phi-conscious. Is awareness all that's strictly required for Phi-consciousness, then? I don't think so, but I am not sure.)

It is exasperating to have to take so much trouble understanding the fine ontological structure, or ontological fine structure, of experience. It is even more exasperating to try to follow someone else's taking that much trouble in terms other than one's own favorite terms, and as you can tell, I have found this part of Siewert's book fairly tough going. Surprisingly or not, Siewert has assured me in correspondence that my notions of "awareness", "sensory quality", etc. are just as opaque to him as his preferred concepts have been to me. (Georges Rey, Joe Levine, Ned Block and I have mutually exasperated each other in this same way for decades.<6>) The fault, I have become convinced, is in the reality: the ontological fine structure of experience itself. The fault is not Siewert's, or mine.

Unlike most writers in this area, Siewert has been as careful as one needs to be, not to miss subtle but important distinctions; much harm has resulted from the literature's general sloppiness in this regard. Siewert has also been cautious in trying to avoid tendentious or premature metaphysical commitment. Both these aims are praiseworthy; but, given the second, there is small wonder that he leaves doubts about what he means.

So why do I risk misunderstanding, and possibly giving offense, by trying to translate Siewert's jargon into mine? Because I have a personal stake in his arguments against those whom he convicts of consciousness neglect and whose views are thereby shown to be inadequate: for reasons I shall explain below, I believe Siewert takes me to be among those enemies. If I have been thus refuted, I need to find that out. And the only way for me to do that is to attempt the translation. Siewert can blow the whistle if and when I go wrong.

So, onward: what is qualitative character or content?

One thing qualitative content might be, and this at least is clear, is just a quale in roughly the sense of C.I. Lewis. A quale in that sense is the distinctive, introspectible, seemingly monadic qualitative feature of an apparent phenomenal individual, such as the redness of a patch in one's visual field, the perceived pitch of a tone one hears, or the particular stinging character of a pain. (Russell of course believed in actual phenomenal individuals, calling them "sense data," so for him Lewisian qualia were the actual properties of sense data. But one need not thus reify sense data in order to talk of the phenomenal properties that figure in sensory experience; one may find a different ontological location for the redness, the pitch, or the feel of the pain.)

Since it is almost completely uncontroversial both that there are qualia in that sense and that they pose a nasty problem for any philosopher of mind or other metaphysician, one is going to have to address them at some point anyway. But there are also, or so I have argued elsewhere (1998b), components of an experience's overall phenomenal character or "feel" besides the Lewisian quale involved -- e.g., the affective aspect of pain. So it's a complicated business. Siewert's "phenomenal character" certainly outruns Lewisian qualia, because it can be nonsensory.

In commenting on Siewert, I will try not to presuppose details of my own view; but in any case, it seems that "Phi-consciousness" in Siewert's sense is a complex. (He sometimes calls it "a feature," but if it is a feature it's a complex feature.) In his own terms, "phenomenal character" is a way of seeming for it to look [or sound or ...] as if such-and-such a shape [or pitch or ...] is present.

For the sake of discussion I'm going to assume that Phi-consciousness is just something like conscious awareness of a phenomenal or qualitative content -- call that "AQ consciousness" to mark it as stipulatively defined.

I still don't know exactly what "Phi-consciousness" means for Siewert, and it might later turn out to mean something objectionable to me. Not so far, though. If it does mean something close to AQ consciousness, then I militantly believe in it and am on Siewert's side against the neglecters; I am also very sympathetic to his first-person perspective. And in my own works (1987, 1996) I have gradually fashioned an overall theory of AQ consciousness.

4. Targets

Who is the enemy? Siewert's main targets are people who offer theories "of consciousness" that neglect Phi-consciousness. And indeed many people do that, some because they're simply not addressing phenomenal consciousness in any sense in the first place, but others because they are confused and they think they're addressing phenomenal consciousness when they're not. David Chalmers (1996) gives a nice catalogue of psychologists and neuroscientists who do the latter. Siewert's own chief examples are Dennett (1991) and Rosenthal (1986, 1993). On paper at least, Dennett is guilty. (I think one can massage his view in order to make it into a theory of AQ consciousness, anyway, but he doesn't do that.) I believe Rosenthal is innocent, at least as regards his HOT theory per se, but I will let Rosenthal fight his own battle.

More targets are listed generically on p. 338:

Conscious visual experience of a certain kind is not just a certain discriminatory capacity regarding visual stimuli, nor more generally can it rightly be identified with some disposition to have other manifest features -- those of a sort we can with warrant be judged to have without observation beneath our surfaces. This reveals that certain functionalist conceptions of mind embody a neglect of consciousness, to be found also in views (whether professedly 'functionalist' or not) that fail to distinguish consciousness from self-representation, or from the access of information to one's capacities to verbalize it or use it to control or plan behavior.

Now, given this Enemies List and Siewert's occasional unlaudatory references to works of mine, I suspect he thinks that I am a Phi-consciousness neglecter like the others. Another reason for that suspicion is that he classifies Michael Tye (1995) as a neglecter, and my view is similar to Tye's though it is more complex. A third reason is that unless I've missed it, Siewert does not mention any going theories of mind that he thinks do engage Phi-consciousness. (Hasn't anyone engaged it? Never mind whether anyone has offered a theory that adequately explains Phi-consciousness and avoids all reasonable objections, which would be a bit much to ask and is seldom achieved anywhere in philosophy, but you'd think someone would have managed not to just neglect Phi-consciousness.)

So: Let me briefly sketch my theory of AQ consciousness, and I will claim to have explained "Phi-consciousness" to the limited degree that I know what it is. Then I'll suppose that Siewert does think me a neglecter and that he will not agree that my theory explains Phi-consciousness. Then I will pull some arguments from Siewert's book that look as though they might refute my theory, and argue seriatim that they don't.

5. Lycan on AQ Consciousness

My theory of AQ consciousness is, in brief: (1) Phenomenal/qualitative character is a combination of one or more (Lewis-)qualia and some other components. (2) I am a representationalist about the qualia themselves; that is, I hold that qualia are intentional and representational contents of sensory states. (Representationalism is now a familiar view, though it wasn't when I (1987) dragged it over from the works of Anscombe (1965) and Hintikka (1969).) I am a functionalist about the qualia's respective sensory modalities (i.e., about whether a given quale is visual, aural, or whatever). (3) I am a functionalist about all other components of overall qualitative character. But (4) both qualia and functional components can occur without the subject's being aware of them; there are unfelt pains, unnoticed visual sensations, and such.<7> If sensory experience (and Siewert's Phi-consciousness) presuppose awareness, something further has to be added.

(5) I add ... awareness! -- that is, I require the subject's awareness of the relevant mental state's qualitative character. (I'm using the term in what I think is its ordinary sense and in as un-theory-laden a way as I can manage. Without pretending that it is beautifully clear or unproblematic, I think most people would agree that I am aware of some of my own mental states and unaware of others.) (6) Following Armstrong (1981), I offer an "inner sense" or higher-order-perception (HOP) theory of awareness. This theory applies to any mental state (qualitative or not) whose subject can be aware of being in that state. N.B., it is not itself a theory of qualia or of anything else to do with qualitative character; qualitative character was taken care of in points (2) and (3) above. (7) Being representations, higher-order perceptions induce distinctive modes of presentation, special introspective ones. This allows me to address Nagel's (1974) and Jackson's (1982) Knowledge Argument and Levine's (1983) Explanatory Gap. (Here is where the first-person perspective is especially important to me.)

Thus, on my view, AQ is a complex, as I think Phi-consciousness must be for Siewert. It is qualitative character, which is itself tripartite, plus contingent awareness of that, in the form of higher-order perception.

6. Saving Belinda

My theory is a good one, I think, though of course there are tenable objections to it (see especially Neander, 1998); I know any number of people, even intelligent, sophisticated people, who are less than fully persuaded by it. But vis-a-vis Siewert's book, the important thing is that my theory is immune to Belinda. I very firmly agree with Siewert that Belinda is possible, even real-world possible. (So I pass at least that test for being a non-neglecter.) On my view, Belinda's problem is that although she may have incoming qualitative contents of some kind -- maybe even Lewis-qualia, though Siewert gives us little guidance here and in any case the exact mechanism of blindsight is still vigorously disputed<8> -- and she may have some higher-order thoughts about them, she is unaware of any. So she lacks AQ consciousness on my view as she does Phi-consciousness on Siewert's. As per (6), I think that is because she has no higher-order perception directed upon her qualitative contents. (Of course, if she doesn't have any qualitative contents in the first place, that would explain it even more decisively.) And my theory of awareness delivers an account of "ways of seeming", though I have never used that expression: in terms of the introspective mode(s) of presentation under which the qualitative contents are perceived (but it is a long story).

As I said, I suppose that despite all that, Siewert will rank me among the neglecters. So it's time to anticipate objections that he might make, based on objections that he does make to Rosenthal and to Tye, and on other arguments that he gives in various connections.

7. Hyperblindsight?

First and most obvious would be to extend the superblindsighter technique from his Belinda-based attack on Rosenthal's view (pp. 129-30) to mine. Couldn't there be a hyper- or even hyperdyperblindsighter that had all of Belinda's abilities plus higher-order perception directed upon a qualitative content, but still was not conscious or aware of the content in the way required for AQ?

I have a swift and sure answer to that one: No. My theory is that a conscious sensory experience just is higher-order perception of a qualitative content. Though Belinda is possible, her anti-Lycan successor is metaphysically impossible. My theory is a theory of a scientific nature, of a natural (though complex) kind; so if it is true, it is necessarily true.

But also, notice, I could concede the objection, but take it as showing only that the HOP view is a bad theory of awareness. (However, I would have to substitute a theory of awareness that did not rule out Belinda.)

8. Conceivability and Possibility

Second objection: just so; if the theory is true, it is necessarily true. And it is not necessarily true, because we can conceive or imagine otherwise. That is, we can conceive or imagine the hyperblindsighter.

I suspect Siewert would be willing to generalize that, and maintain that no purely informational notion of a "sensory quality" combined with a purely informational higher-order notion of awareness could (by itself) guarantee Phi-consciousness, again because we can always conceive the absence of Phi-consciousness no matter how much representation and information are flying about.

I tend to pooh-pooh this. We all know that conceivability/imaginability does not show metaphysical possibility. Of course, we all know also that that is hardly the end of it. Starting with Max Black (in Jack Smart's famous footnote 13; Smart, 1959) and followed by Saul Kripke (1972) -- and then, years later, by Stephen White (1986), Bill Hart (1988), Frank Jackson (1993), George Bealer (1994), David Chalmers (1996) and others -- many have contended that a posteriori necessity does not come for free. There is a condition that must be met, and although that condition is met by standard scientific examples of a posteriori necessity, the alleged a posteriori necessity of the reduction of conscious experience does not meet it. Siewert extensively anticipates my sort of pooh-poohing (pp. 155-61), and appeals to a mixture of Kripkean considerations with the stubborn view that conceivability creates an enormous presumption in favor of metaphysical possibility, a presumption that can seemingly be defeated only by a theory-independent good reason for thinking that the conceivable hypothesis is not metaphysically possible (p. 163).

Reply: This is a large subject, and for now I can only cite what I've said about it elsewhere. (Against Black, 1987, pp. 134-35; against Kripke, 1987, Chapter 2; against Jackson and Chalmers, 1998a; etc.) I note, though, that I reject the requirement of a theory-independent good reason for doubting the metaphysical possibility.

9. Differences in Ways of Seeming

The third objection is a very compressed one found on pp. 85-89, 92, and 219-220. I don't know that I understand this one fully; Siewert's sentences in these passages are particularly hard to parse. But I think it's something like this: my theory, like Dennett's and Rosenthal's, explicates AQ consciousness as a bipartite affair, as the combination of a qualitative content (a matter of first-order representation and some functional stuff) with an only contingently related higher-order perception. But this seems to imply that two conscious experiences could differ only in what their bottom halves represented. As I said before, two experiences can share the same style and degree of awareness but differ in qualitative content, and two experiences can have much the same qualitative content but differ in style or degree of awareness. But Siewert argues that two Phi-conscious experiences cannot differ from each other only in the ways in which their blindsightish nonconscious counterparts would differ from each other, say in what they represented. Recall his claim that "any state one is in, or event in one's brain, that differed from others just as these experiences differ among themselves, would be a conscious experience" (p. 87). Also:

[I]ts looking to you as if there is something yellow on the left does not differ from its looking to you as if there is something blue on the left, in only the way in which a representation of something yellow differs from a representation of something blue. After all one's visual judgments would differ in this way -- but they do not differ from one another just as the visual experiences do. (p. 88)

Because two Phi-conscious experiences cannot differ from each other only in the ways in which their blindsightish nonconscious counterparts would differ from each other, there is no one ingredient that could be added to each of Belinda's subconscious sensory representations that would make each of them Phi-consciousness; yet my theory implies that there is such an ingredient: awareness.

True, its (consciously) looking to me as if there is something yellow on the left does not differ from its looking to me as if there is something blue on the left, in only the way that representing yellow and representing blue differ from each other. They differ in their distinctive "ways of seeming" as well. But the HOP part of my theory actually predicts that. For the first-order visual representations of yellow and blue are themselves represented by "inner sense" only under distinctive, introspective modes of presentation, which adds a second dimension of difference.

On this view, the difference in "ways of seeming" is still bipartite, a combination of the quality represented and the special mode of presentation under which that first-order representing is itself represented in or by the higher-order perception that makes it into a conscious experience. Siewert may balk at that. He may at least feel like insisting that the difference between its looking to me as if there is something yellow on the left and its looking to me as if there is something blue on the left is a single, unified feature, not a jury rig of representational components that are only contingently related to each other. But to insist on that, without further defense, would beg the question.

10. Higher-Order Perception

Fourth objection: my view is or contains a HOP theory, and hence a higher-order-representation theory. But Siewert has a general objection against what he calls "self-reflection" views of Phi-consciousness (Chapter 6). He argues that Phi-conscious experiences can occur without their subjects' having higher-order thoughts about them, and for that matter without their subjects' having any sort of higher-order representations of them -- as he puts it, without having a mentally self-reflexive intentional feature whose conditions of truth or accuracy require that one have that Phi-conscious experience.

Quite right. But my own HOP theory does not require that AQ-conscious (or otherwise phenomenally conscious) experience be itself represented. The higher-order perception in question is immanent to the experience itself. The experience itself is -- consists of -- an awareness of a qualitative content, and the latter awareness is for me a matter of higher-order perception. But it is only the qualitative content that is represented, not the experience as a whole.

Siewert's Chapter 6 contains an entire section (6.3) on what he calls "the 'conscious-of' trap," a fallacy that "is responsible ... for a lot of false starts in discussions of consciousness" (p. 197). Chiefly, he attributes it to Rosenthal, but I too am listed among the perpetrators/victims.

Some vague and faintly plausible claim to the effect that conscious experience is experience we are conscious of (or 'aware of') is assumed or asserted, and then a dubious construction on this is slipped in unexamined, taking the object of the preposition 'of' to indicate what some mental event is 'directed toward'.... In this way, we are hustled (or hustle ourselves) into supposing there is something obvious or intuitively compelling about the notion that consciousness is some sort of higher-order mentalness, or self-representation, or 'inner scanning.' (p. 197)

Steady on, Siewert: Rosenthal and I maintain that a conscious state (not, experience) is a state one is conscious or aware of being in. (For me that is a terminological stipulation, not a substantive claim.) And we do understand consciousness/awareness of X as there being a mental state "directed toward" X in the sense of having X as its intentional object or representatum. It surprises and interests me that Siewert would find the latter view a "dubious construction"; if the "of" in "conscious of" and "aware of" is not the "of" of intentionality, what is it? (That's a bonafide, not a rhetorical, question.) But neither Rosenthal nor I either is hustled, hustles himself, or in another way arrives at the notion that AQ consciousness or Phi-consciousness is (merely) higher-order mentalness; each of us firmly rejects that notion. Though my own theory of AQ consciousness in part appeals to a HOP theory of awareness, neither that HOP theory itself nor Rosenthal's higher-order thought theory of awareness is (alone) a theory of AQ consciousness or Phi-consciousness at all.

11. HOP vs. HOT

But (fifth) in any case, what is this higher-order perception or "inner sense"?

[W]e would have to know more about what this kind of scanning or self-modeling might be, and just what distinguishes it from higher-order thought, if we are to understand what it might mean to identify it with phenomenal consciousness. (p. 130)

Well, first, neither Armstrong nor I nor any other philosopher I know does identify higher-order perception with Phi-consciousness. But Siewert is right that we still do have to know more about what it is and what distinguishes it from mere higher-order thought.

On my view, inner sense is (indeed) a kind of internal monitoring, a species of proprioception, conducted by a scanning device. It is an internal attention mechanism more perception-like than merely thought-like, particularly in delivering representations that are nonconceptual. It is also sometimes under voluntary control, as you can verify for yourself and as should make it hard for you to doubt that we do have such internal attention mechanisms. It is just introspection, which I take to be psychologically and neurologically real. Siewert himself should agree that he introspects his own qualitative contents in that way; a person who emphasizes the first-person perspective as firmly as he does should not get caught denying that he introspects.

But one need not accept any of these things to hold a higher-order-representation theory of awareness, and for that matter one can appeal to awareness without holding a higher-order-representation theory of it.

12. An Appearance-Reality Gap for Appearances

Sixth: Despite Siewert's lack of sympathy with the views of Dennett, he joins Dennett in objecting to a certain consequence of HOP theories: that there is in principle an appearance/reality gap for appearances themselves. If awareness of one's own mental state is a matter of higher-order perception of that state, then like any other perceptual or quasi-perceptual faculty, the internal attention mechanism is a mechanism and can break down. Thus, an object might seem red to me without seeming to seem red to me, or, worse -- a false positive -- it might seem to seem red to me without (in fact) seeming red to me. But does this even make sense? On my view, it should. But Dennett (1991, pp. 132-33) and Siewert (pp. 212-13) deplore that.

My opinion, not just a biting of the bullet, is that the introspective appearance/reality gap does make sense, strange as it may sound. I think our experiences do sometimes introspectively appear to us differently from the way they actually are. My visual field at some time seems to me much larger than it actually is at that time (Frisby, 1979; what creates the illusion is the constant saccades being made by my eyes). Then there are the usual fraternity-boy examples of mistaking sensations of cold, produced by application of an ice cube after suitable priming about branding irons, etc., for sensations of burning and/or pain. (I don't think the victim merely makes a false judgment.) For that matter, there is the common mistaking of fear for pain and seeming to feel pain when one does not actually feel any. All these cases are disputable, of course.

The problem about false positives is, I admit, worse; see Neander (1998) and Lycan (1998b). But so far as I can see, the idea at least is conceptually coherent.

I trust I have said enough to convince you that Siewert's book is thought-provoking and challenging by any professional standard. You should read it and make sure you are not a neglecter -- especially if you are a materialist who has offered a theory "of phenomenal consciousness."

Notes

<1>. For the details, see pp. 188-92.

<2>. Though it seems Siewert himself published the device first (1993).

<3>. I here omit the qualification that Belinda's left field contains an "amblyopic" portion.

<4>. This is in part to counter Dennett's (1991) suggestion that conscious seeing is just the ability to make unprompted discriminations of color, shape and position so long as these discriminations are very rich in content. "BeLINDa" =, I presume, blindsighter; "CONnie" = conscious seer.

<5>. Though David Chalmers has recently persuaded me in correspondence that one of the three most common usages is conspicuously more natural and less tendentious than the other two, and should be regarded as the standard use.

<6>. In November of 1998, Rey, Levine and I agreed to meet for a day at the National Humanities Center, to converse about our respective theories of mind and our views on conscious experience in particular, with the very limited aim of mere mutual understanding. The three of us have much the same training and philosophical background; we look at the issues in almost exactly the same way; and we even hold very similar views. Nonetheless, with the best will in the world despite a balky coffee machine, it took us some hours just to stipulate a basic vocabulary for identifying and fine-tuning the range of phenomena to which our theories are addressed.

<7>. Siewert expresses distaste for this view on pp. 118-19, but does not argue against it.

<8>. Siewert evinces some skepticism about this on pp. 122-23. It should be noted that psychologists disagree sharply among themselves as to what account should be given of real-world blindsight. Some even believe it is so shallow a thing as a deficit in reporting ability. For a good review of that literature, see Natsoulas (1997).

References

Anscombe, G.E.M. (1965). The intentionality of sensation: A grammatical feature. In R.J. Butler (Ed.), Analytical philosophy: Second series. Oxford: Basil Blackwell.

Armstrong, D.M. (1981). What is consciousness? In The nature of mind. Ithaca, NY: Cornell University Press.

Bealer, G. (1994). Mental properties. Journal of Philosophy, 91, 185-208.

Block, N.J. (1995). On a confusion about a function of consciousness. Behavioral and Brain Sciences, 18, 227-47.

Chalmers, D. (1996). The conscious mind. Oxford: Oxford University Press.

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

Frisby, J.P. (1979). Seeing: Illusion, brain and mind. Oxford: Oxford University Press.

Hart, W.D. (1988). Engines of the soul. Cambridge: Cambridge University Press.

Hintikka, K.J.J. (1969). On the logic of perception. In N.S. Care and R.H. Grimm (Eds.), Perception and personal identity. Cleveland, OH: Case Western Reserve University Press.

Jackson, F. (1982). Epiphenomenal qualia. Philosophical Quarterly, 32, 127-36.

Jackson, F. (1993). Armchair metaphysics. In J. O'Leary-Hawthorne and M. Michael (Eds.), Philosophy in mind. Dordrecht: Kluwer Academic Publishing.

Levine, J. (1983). Materialism and qualia: The explanatory gap. Pacific Philosophical Quarterly, 64, 354-61.

Lycan, W.G. (1987). Consciousness. Cambridge, MA: Bradford Books / MIT Press.

Lycan, W.G. (1996). Consciousness and experience. Cambridge, MA: Bradford Books / MIT Press.

Lycan, W.G. (1998a). Against the new a priorism in metaphysics. Lecture to the Twenty-Fourth Annual Meeting of the Society for Philosophy and Psychology, Minneapolis, MN.

Lycan, W.G. (1998b). In defense of the representational theory of qualia (replies to Neander, Rey and Tye). In Tomberlin (1998).

Nagel, T. (1974). What is it like to be a bat? Philosophical Review, 82, 435-56.

Natsoulas, T. (1997). Blindsight and consciousness 1. American Journal of Psychology, 110, 1-33.

Neander, K. (1998). The division of phenomenal labor: A problem for representational theories of consciousness. In Tomberlin (1998).

Rosenthal, D. (1986). Two concepts of consciousness. Philosophical Studies, 94, 329-59.

Rosenthal, D. (1993). Thinking that one thinks. In M. Davies and G. Humphreys (Eds.), Consciousness. Oxford: Basil Blackwell.

Siewert, C.P. (1993). What Dennett can't imagine and why. Inquiry, 36, 96-112.

Smart, J.J.C. (1959). Sensations and brain processes. Philosophical Review, 68, 141-156.

Tomberlin, J.E. (ed.) (1998). Language, mind, and ontology (Philosophical Perspectives, vol. 12). Atascadero, CA: Ridgeview Publishing.

Tye, M. (1995). Ten problems of consciousness. Cambridge, MA: Bradford Books / MIT Press.

White, S. (1986). Curse of the qualia. Synthese, 68, 333-68.