Psyche 7(11): First Person Warrant: Comments On Siewert's <i>The Significance Of Consciousness</i> by Fred Dretske

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

First Person Warrant
Comments On Siewert's The Significance Of Consciousness

Fred Dretske
Philosophy Department
Duke University
Durham, NC 27708
U.S.A.

fred.dretske@duke.edu

PSYCHE, 7(11), April 2001
http://psyche.cs.monash.edu.au/v7/psyche-7-11-dretske.html

KEYWORDS: functionalism, representationalism, experience, perception, first-person warrant, knowledge.

COMMENTARY ON: C. Siewert (1998) The Significance of Consciousness. Princeton: Princeton University Press. x + 374pp. ISBN: 0-691-02724-2. Price: $42.50 hardcover.

ABSTRACT: I agree with Siewert's claims about the special character and importance of phenomenal consciousness and the impossibility of providing a satisfactory functionalist reduction of it. I question, however, his dismissal of a representational (non-functionalist) theory of conscious (e.g., perceptual) experience. I also question his account of how conscious agents are supposed to know, or enjoy first person warrant, for their belief that they are conscious.

There is so much I agree with in Siewert's book that I hesitated to contribute to this symposium. Like other philosophers I get nervous when I can't find things to criticize. I take it as a sign that I'm not paying attention. Not this time. I paid attention. I found a few things to pick at, but, on the whole, I agree with David Chalmers (agreeing with Chalmers also makes me nervous): this is a marvelous book. Some philosophers -- most conspicuously Dan Dennett -- won't be thrilled by it, but almost everyone else will love it.

Siewert and I agree about a lot. So I'll start here -- with a quick recitation of things we have in common. I will then register puzzlement over what the objection is supposed to be to my own (non-functionalist) theory of consciousness. I'll conclude with a question about the epistemology of consciousness: how, on Siewert's view of things, do we know we are conscious?

Siewert argues that functionalism about phenomenal consciousness is false (pp. 141-2). Having a conscious experience is not simply a matter of possessing discriminatory talents or abilities, manifest (p. 140) or hidden (p. 147), directed outward or inward (p. 133); it is not a matter of making higher order judgments (pp. 82-84); it is not a matter of using, or having a capacity to use, nor a special way of coding (p. 135), information (however rich -- p. 107) about the world or about one's own discriminatory powers (p. 129). Neither is it a matter of having self-directed thoughts of a special kind (p. 197, p. 214). Creatures who have conscious experiences will differ in a variety of ways from those who are not conscious (p. 141) -- the ones who are conscious will, for example, think and say they are conscious -- but neither the behavioral difference nor the tendency to manifest it is plausibly identified with consciousness. Thinking and saying one is conscious is not the same as being conscious; like most other discriminatory talents we exhibit, these cognitive and behavioral differences are not to be identified with, but are rather to be understood as the result of, and explained by, consciousness (p. 137). I agree with all this.

To Siewert's credit he makes this argument with extraordinary sophistication and subtlety without putting excessive demands on the reader's powers of imagination. Though he thinks zombies (self-moving humanoids who lack consciousness) are possible (p. 309), he thinks they are difficult to imagine (footnote 6, p. 357; footnote 1, p. 363) and he never bases his argument on their possibility. The most one is asked to imagine (from the first person point of view) is low-grade blindsight, someone who doesn't actually see anything on the left but who nonetheless manifests, for objects on the left, the unprompted powers of discrimination of a person with bad eyesight (to whom things in that part of the visual field look blurry and indistinct). Siewert isn't asking us to imagine "premium quality" (Ned Block's super-duper) blindsighters -- watch-makers and needlepointers who can't see anything. Just someone who manifests a low-grade ability to identify things he has no conscious experience of -- the kind of ability I exhibit at recognizing the words I see on a page without my reading glasses. That is something I can certainly imagine. I bet most readers who don't have a functionalist theory to defend will be able to imagine it too.

It is important to understand that although Siewert rejects functionalism, he does not deny the causal efficacy or explanatory relevance of consciousness. Seeing a ripe apple (i.e., having a conscious experience of a ripe apple) causes me to reach out and grasp it. Had I not seen it, I would not have reached in just the way I did. Seeing it explains why I reached. There may be other, non-conscious, internal states that bring about the same behavior. In the case of Siewert's imagined amblyopic blindsighters, for instance, their (unprompted) responses may be caused by non-conscious encodings of information or non-conscious visual representations (p. 83). The fact that their behavior is produced by non-conscious internal events doesn't mean our behavior isn't caused -- and therefore that it isn't explained -- by the fact that we actually see things, by the fact that we are conscious. The fact that consciousness is not the only thing that can cause X doesn't mean that, in us, it doesn't cause X (p. 309).

An apparent tension develops, but Siewert neatly disposes of it. If non-conscious events can produce the same effects as conscious events (in Siewert's imagined low-grade blindsighters, for example), maybe they can always, even in premium-quality blindsight, produce the same effects. If so, that not only creates a problem about other minds, but a problem about one's own, a problem about first-person warrant. Why can't a non-conscious event -- one of Siewert's non-conscious visual representations -- produce in us a belief that we are conscious? If it can, how can we be sure it doesn't? How can we be sure we are actually conscious and not just being caused to believe we are by non-conscious events? If we can't be sure, first-person warrant vanishes. If, on the other hand, non-conscious events cannot always produce the same effects as conscious ones, why can't a functionalist exploit this fact in specifying a functional role for consciousness: consciousness is whatever produces, or is apt to produce, that where that (e.g., a belief that one is conscious) is what only conscious events can produce.

Siewert rejects the second horn of this dilemma: non-conscious events cannot always produce the same effects as conscious ones, but this cannot be exploited by the functionalist. Non-conscious events cannot, for instance, produce the belief that one is conscious, but it doesn't make sense to identify consciousness with whatever produces, or is apt to produce, a belief that one is conscious (p. 142, p. 145). That would be like saying (p. 132) that ultra shoes are special shoes in only one respect -- they make their owner believe they are ultra shoes. That, Siewert tells us, isn't a way that shoes can differ. For the same reason, the difference between a conscious and a non-conscious event cannot consist in believing and not believing it (the event) is conscious. That is not a way things can differ.

So much for functionalism.

Though I was aggressive in my rejection of functionalism in (Dretske, 1995), Siewert groups me (Section 4.8) with functionalists in neglecting phenomenal consciousness. He disposes of my representational theory of conscious experience in a few lines (pp. 144-145). I find these criticisms pretty anemic, but Siewert probably thinks his criticisms of Michael Tye, another representationalist, apply as well to me: he is, as it were, killing, two representational birds with one stone. If this was his intent, I don't think a single stone will do. At least I didn't feel any impact. So just for the record, and in order to elicit a clarifying response, let's be sure we have the objection straight. I argued in (Dretske, 1995) that conscious experiences are internal representations that derive their powers of representation from a process of natural selection. The systems that give rise to perceptual experience, for instance, were selected to do a certain job, to provide information (= indicate) how things stand with respect to external objects (in the case of exterocepton) and (in proprioception) parts of the body. Qualia, how things seem (look, smell, feel) to a person, is how those representations represent (possibly misrepresent) things to be. As I understand it, Siewert's criticism is that one can imagine a blindsight subject having such representations of objects without being conscious of those objects. So conscious experience of objects cannot be simply a matter of harboring such representations of them. End of story.

Siewert's powers of imagination must be greater than mine because I don't find it that easy to imagine this. To my ear this is like imagining a person without a heart who nonetheless has an organ in his chest that not only pumps blood (this, of course, needn't be a heart) but has pumping blood as its biological function, as the thing it was evolved to do. Can we imagine this?

Think about an animal -- a honey bee, for instance -- equipped with anatomical structures whose biological function it is to pick up and process information about the color of nearby objects. This information is not only received and processed, but, when the need arises, used to control and guide the bee's foraging behavior. Do these bees have color vision? Biologists tell us they do. If they are right about this (it certainly sounds like vision to me), then since color vision is seeing colors, and seeing colors is (visual) awareness of color, the bees must be conscious. Of colors. Asking one to imagine a bee with a properly functioning receptor system of this sort that lacks consciousness is asking one to imagine them as blind. I can't do that.

I can imagine zombie bees (bees that, though lacking color vision, behave as though they had it) as well as the next guy. I find this as easy to do as imagining a device -- perhaps even another bodily organ -- in the chest of a human being pumping blood that is not really a heart. But I imagine this by imagining something there that doesn't belong there, that doesn't have the biological function of doing what hearts have the function of doing (either because it doesn't have the function of doing it or because the function isn't biological -- e.g., an artificial pump). But how am I supposed to imagine my bees as blind? Unlike fake or artificial hearts, or devices that, for whatever reason, merely function as hearts, everything in the bee that tells it the color of things is working as it is supposed to, exactly the way it was "designed" to work. What, therefore, is one being asked to imagine when asked to imagine the bees as blind? Are we supposed to imagine these bees as lacking conscious experience, as buzzing blindsighters? But exactly what properties do conscious experiences of color have that the bees' internal biological representations of color lack? Both conscious experiences as ordinarily understood and the representations I describe are of colors. They are both accurate (veridical) or inaccurate (illusory or misleading). The representations possess the subjective features usually associated with perceptual experiences -- they, for example, generate a point of view for the animal in which they occur: the objects whose color is represented systematically change as the bee flies around. The representations, just like experiences, are distinguishable from conceptual representations (e.g., beliefs or judgments) about colors. The representations are, like experiences, distinguishable (as state types) from the physical states that realize them. It is possible (at least logically) to have organisms who are functionally (and perhaps even physically) identical who nonetheless differ in the kinds of representations occurring in them. So we can, with respect to these representations, conceive of inverted spectra phenomena (even, perhaps, zombies) -- thus capturing the sort of intuitions that sustain Siewert's "first person" point of view. I could extend this list. The point of compiling such a list is that if Siewert's case against representationalism rests on the fact that he can imagine bees (or people) having biological representations of the sort I describe without having conscious experiences, he is, I think, obliged to tell us what property he is imagining these experiences to have that these biological representations lack. What does the first-person point of view provide that is not provided for the creatures in whom these representations occur? Until we know this, we won't know whether we can imagine there to be representations of the sort I describe without conscious experience.

Let me turn, finally, to an epistemological issue. I have a question about exactly what Siewert takes to be the relation between phenomenal experience and the knowledge (or first-person warranted belief) that one has it, the knowledge that (unlike blindsight) one actually sees something off on the left. Since he rejects a perceptual model of introspection, he doesn't think we are aware of our own conscious experiences (of objects on the left) in the way these experiences make us aware of external objects on the left (p. 214). So, when I see an apple, how do I find out that I see it, that I'm conscious of it? What warrants me in believing that, unlike one of Siewert's blindsighters I see the apple in front of me? Nothing I am perceptually aware of (in this case, only the apple) tells me I'm aware of it. The apple, being a physical (mind independent) object, would presumably be the same if I weren't aware of it. But if nothing I am perceptually aware of tells me that I am aware of something, and I am not aware of the internal machinery that makes me aware of the apple, where do I get my warrant for believing that I am aware of something, that I actually see an apple? What is the source of this warrant?

At times (e.g., p. 39, p. 172) Siewert suggests that it is not what one is conscious of but, simply, the fact that one is conscious that provides first-person warrant (or is essential to the warrant -- p. 20) for the belief that one is conscious. This would mean that in the special case of consciousness, P warrants you in believing P. This doesn't mean that if you are conscious you necessarily think you are since Siewert believes -- quite correctly it seems to me -- that animals and small children are conscious without believing that they are -- without even (in the case of animals) the capacity to believe they are. One can have a mind without the capacity to think one has a mind (p. 203). The thought seems to be, rather, that if one thinks one is conscious (or not conscious, for that matter -- p. 172) the fact that one is conscious (or not conscious) provides first-person warrant for the belief. These beliefs enjoy a kind of first-person warrant that other beliefs lack.

I find this a little mysterious -- mysterious enough to ask for additional clarification. As I understand it, the fact that S is conscious provides S, but not me, with a warrant (first-person warrant) for thinking S is conscious. What I need to understand is why my belief that S is conscious isn't warranted by the same fact -- that S is conscious -- that warrants his belief. The answer we typically get to questions like this in epistemology -- questions like: Why is S's belief that P warranted but my belief that P is not? -- is that S is aware of some fact, F, that I am not and this fact justifies his belief. S knows F and I don't. This answer is not available here since the only fact that we are given to warrant S's (but not my) belief is the fact that S is conscious -- exactly the fact that we cannot, without begging the question, assume that S knows but I don't.

So I don't understand the source of first-person warrant. The reason I press this issue is because although I, like Siewert, think we enjoy a special kind of first-person authority with respect to (large chunks) of our mental life, I do not think we enjoy the distinctive first-person warrant described in this book. What we enjoy special authority about is not the fact that we occupy mental states having content, the fact that we have beliefs and experiences, but, rather, facts relating to the content of these mental states -- what it is these states make us consciously aware of. We have to learn from others -- usually parents -- that we are conscious, and most of us don't learn it until we are three or four years old. Conscious animals, I suspect, never learn it. What we know, and what we have special warrant for is not that we have beliefs and experiences, conscious states with a content that can be either true or false, accurate or inaccurate. What we know and what we have privileged access to is what it is we believe and what it is we experience -- that, for example, Daddy is home, the dog is loose, and that there are cookies in the jar. A child knows what it thinks long before it knows that it thinks. We are authorities about what is, so to speak, in our mind, not about the fact that we have a mind.

This, though, isn't the place to drag in my own dog and pony show (those who want to view the dog and pony can look in (Dretske, 1993, 1995, 1999). I don't want to drag Siewert to places he doesn't want to go, or doesn't think he has to go (he says that some of these epistemological issues would take another several books -- p. 172). I do think, though, that he needs to clarify the special connection between phenomenal consciousness and first-person warrant by telling us what the special relation is that he stands to the fact that he is conscious that gives him, but not others, first person warrant for the belief that he is conscious. After all, if he has freckles, he enjoys no first-person warrant for thinking he has them. He either found out he had them by looking in a mirror or by someone telling him. What makes the fact that he is conscious so different? If I knew this I would better understand Siewert's views on consciousness and first-person warrant.

References

Dretske, F. (1993). Conscious experience. Mind, 102, 1-21.

Dretske, F. (1995). Naturalizing the Mind. Cambridge, MA; MIT Press.

Dretske, F. (1999). The mind's awareness of itself. Philosophical Studies, 95, 103-124.