from Amazon.Com
|
|
William Robinson's book Computers, Minds, and
Robots may be purchased from Amazon.Com |
|
|---|
A Theory of Phenomenal Consciousness?
William S. Robinson
Department of Philosophy
Iowa State University
USA
wsrob@iastate.edu
Copyright (c) William S. Robinson 1999
PSYCHE, 5(4), April, 1999
http://psyche.cs.monash.edu.au/v5/psyche-5-04-robinson.html
KEYWORDS: Carruthers, hard problem, higher order thoughts, phenomenal consciousness, recognition, seeming.
COMMENTARY ON: Carruthers, P. (1998) "Natural theories of consciousness," European Journal of Philosophy, 6(2), 203-222.
ABSTRACT: Peter Carruthers offers a model that embraces first order representations (FORs) and higher order representations (HORs) or higher order thoughts (HOTs). His model stipulates certain features of FORs and HOTs. Carruthers agrees with qualia realists that the FORs of his model are not adequate for phenomenal consciousness, and invokes HOTs to supply the required addition. It is argued that Carruthers' HOTs fail to provide anything that will enable him to account for phenomenal consciousness, i.e., that his model fails to include phenomenal consciousness and thus provides no help in understanding human consciousness, animal consciousness, or the "hard problem".
In "Natural Theories of Consciousness" Peter Carruthers describes a model that he claims will show the "hard problem" to be not so hard after all. To evaluate this claim, we should begin with the model, which contains the following elements and stipulations.
(1) First order representings (FORs) -- brain events that track properties of physical objects.
a. Neither these events nor anything they cause has
nonrelational properties corresponding to phenomenal properties
of experience. (See Carruthers' note 12.)
b. For bodily sensations, the tracked properties are
properties of our own body parts.
c. FORs are analogue.
d FORs causally contribute to behavior appropriate to the
tracked properties.
(2) Higher order representations (HORs; in particular, higher order thoughts, or HOTs) -- brain events that track first order representations.
a. HOTs are not analogue.
b. HOTs contribute to making the appearance/reality
distinction, e.g., to judgments about a thing's seeming red
while known not to really be red.
Some comments on this model are in order. First, "tracking" is Tye's (1995) useful metaphor, but I am using it to cover a wider class of similar explications of the representing relation. Neither Carruthers' article nor this commentary commits to which member of this class will prove to be most defensible. Second, Tye also provides an example that is convenient for understanding tracked properties, namely, (certain) triples of surface reflectance percentages. I will use this example, but only as an example of the kind of account we should have in mind. Some other similar account may prove to be better in the end, but such a development will not affect either Carruthers' argument or this commentary. Third, analogousness is not much elaborated in Carruthers' article; but I shall take it as satisfied wherever the physical properties of representing events are ordered isomorphically to the properties that are represented. For example, more intense stimulations might be tracked by events involving more neurons, or larger patches of a color might be tracked by events that involve a larger area of visual cortex.<1>
Carruthers and qualia realists agree that what is provided under (1) is not sufficient for phenomenal consciousness. Qualia realists will want to add particular occurrences of phenomenal consciousness that literally resemble and differ in non-relational ways. (This addition is, of course, a rejection of (1a).) They will hold, e.g., that triples of percentages of light of certain key wavelengths reflected from ripe mangoes, oranges, and summer leaves, respectively, have a way that they look, and that these ways of looking have nothing "triple-ish" about them. They will hold that the way ripe mangoes look intrinsically resembles the way oranges look more than it does the way summer leaves look. It is the qualities of these ways of looking, in virtue of which they intrinsically resemble and differ, that will be the what-it-is-like of consciousness.<2>
Such qualities are rejected by Carruthers. In his model, the addition to (1) that we must make to obtain phenomenal consciousness is higher order thoughts (HOTs). The question we must ask is how this addition might be supposed to provide phenomenal consciousness. I consider a series of answers. (A1) Perhaps the key lies in the fact that what is tracked is internal. But this answer cannot be right. Many untoward bodily conditions are wholly internal. If internality of object tracked were the source of phenomenal consciousness, then first order representations of certain wholly internal bodily irregularities (e.g., broken ribs) should themselves produce phenomenal consciousness. But dogs can, presumably, have first order representations of broken ribs while, on Carruthers' account, lacking phenomenal consciousness.
(A2) Perhaps the key to phenomenal consciousness in Carruthers' model lies in the fact that the relevant HOTs are ones that apply recognitional concepts. (A2a) Perhaps what this means is that the application of concepts in HOTs is non-inferential. That would be to say that HOTs are directly conditioned to occurrences of first order analogue tracking events; and that they persist despite the presence of conflicting thoughts (e.g., to the effect that the object before one isn't really red).
But first level trackings are likewise not based on inference, and they persist in the face of top-down influence. (If they did not, optical illusions ought to disappear when we know about them.) This holds as well for first level trackings of internal events. So, it does not seem that non-inferentiality and persistence of representation can be the properties that are responsible for phenomenal consciousness, if, as Carruthers holds, first order representations are not adequate for phenomenal consciousness.
(A2b) It seems likely that (A2a) offers too reductive a view of what is involved in recognitional application of a concept. Carruthers seems to give us a richer account of recognition: "When I recognise in myself an experience as of red, what I recognise is, in fact, a perceptual state which represents worldly redness, and which underpins, in turn, my capacity to recognise, and to act differentially upon, red objects" (8). But this apparently richer description actually adds nothing. The words that follow "in fact" give a theoretical (i.e., not experiential) description that merely repeats what is contained in (1) and (1d), which Carruthers agrees to be insufficient for phenomenal consciousness.
(A3) Perhaps the key to phenomenal consciousness lies in the referents of the judgments for which HOTs are critical -- that is, in the seeming of something to be, e.g., red, or in the character of the way red is represented (at the first level). "What gets added by the presence of a HOR-system is a dimension of seeming or appearance of that very same first-order content.... [W]hen I focus on my experience of a colour I can, in a sense, do other than focus on the colour itself -- I can focus on the way that colour seems to me, or on the way it appears; and this is to focus on the subjectivity of my experiential state" (6).
Here, indeed, we have come upon something that answers to the provision of phenomenal consciousness. Unfortunately, Carruthers has no place in his model for the appeal he makes in the quoted passage. A qualia realist can offer to identify the way a colour seems or appears with a non-relational quality of an experience. Carruthers explicitly denies such qualities. What is left? The appearance of first-order content (e.g, red) cannot be the tracked physical property (a triple of reflectance percentages, as it may be); neither can it be the neural properties of the FOR, since red things do not appear as a multitude of action potentials. We have considered various aspects of HOTs, and concluded similar inadequacy. It may seem that we can locate the key fact about HOTs simply in the fact that they lead to "seems" judgments. But this will not do, because many "seems" judgments are unrelated to phenomenal consciousness -- witness the first occurrence of "seem" in the preceding sentence. In light of this fact, we need seemings that can ground the difference between "seems" judgments that involve phenomenal consciousness and "seems" judgments that do not. The judgments themselves evidently cannot provide such a ground.
(A4) Perhaps Carruthers can supply what is needed through his distinction between what the world is like for S and what an experience is like for S. Referring to the latter, he says that "this would seem to require subjects to possess information about, and to make discriminations amongst, their own states of experience" (5). But information is preserved by tracking, and discrimination by differential tracking; and these, on Carruthers' own view, do not imply phenomenal consciousness. (That is why, above, we had to look for more in "recognition" than noninferential response to first order events.) Thus, Carruthers is right to highlight the need for a distinction between what aspects of the world are tracked by a subject (what the world is like to S), and what experience is like; but despite his intentions, his model fails to include the latter.
I believe that we have now examined all the plausible answers to our question, "How can what Carruthers adds in (2) provide for phenomenal consciousness?" None have proven adequate. I conclude that Carruthers has presented a model which may permit construction of a naturalistic theory of many aspects of mental function, but which has no place for phenomenal consciousness. Consequently, the naturalistic theory that is built on the model cannot enlighten us in any way about phenomenal consciousness, whether in ourselves or in nonhuman animals. Nor, a fortiori, can it be regarded as an advance on the `hard problem'.
<1> For elaboration of analogue representation, see Robinson (1995).
<2> For explanation and defense of qualia realism see Robinson (1996, 1997).
Robinson, W. S. (1995). Direct representation, Philosophical Studies, 80, 305-322.
Robinson, W. S. (1996). The hardness of the hard problem, Journal of Consciousness Studies, 3, 14-25. [Reprinted in J. Shear (Ed.) Explaining consciousness -- the "hard problem". Cambridge, MA: MIT Press, 1995-97.]
Robinson, W. S. (1997). Intrinsic qualities of experience: Surviving Harman's critique, Erkenntnis, 47, 285-309.
Tye, Michael (1995) Ten problems of consciousness: A representational theory of the phenomenal mind. Cambridge, MA: MIT Press.
