Conscious experience, reduction and identity: many explanatory gaps, one solution

Abstract

This paper considers the so‐called explanatory gap between brain activity and conscious experience. A number of different, though closely related, explanatory gaps are distinguished and a monistic account of conscious experience, a version of Herbert Feigl's “dual‐access theory,” is advocated as a solution to the problems they are taken to pose for physicalist accounts of mind. Although dual‐access theory is a version of the mind‐body identity thesis, it in no way “eliminates” conscious experience; rather, it provides a parsimonious and explanatorily fruitful theory of the consciousness‐body relation which faithfully preserves the nature of conscious experience while going quite far in “bridging” the various explanatory gaps distinguished below.

Notes

Liam P. Dempsey, Dalhousie University, 5252 Tobin Street, Apt. 704, Halifax, Nova Scotia, Canada B3H 4K2, email: [email protected]

It is not only proponents of the explanatory gap that resist the identification of qualia with neurophysiological properties of the human central nervous system; any version of “non‐reductive physicalism” denies the reducibility of mental properties, although it holds that all events or particulars are physical. Some non‐reductive physicalists construe the relationship between mental and physical properties as one of supervenience. For instance, Marras (1993, 2000) holds that qualia, like other mental properties, supervene on, but are not identical to, neurophysiological properties. The reasons for the widespread rejection of mind‐body reduction are varied; this paper seeks to deal with those who object specifically to the reduction of qualia to their neurophysiological co‐variants on the basis of the supposed explanatory gap taken to exist between the two sets of properties. Non‐reductive physicalists are motivated more by the supposed multiple realizability of the mental: if a mental property is realizable in a wide variety of physically disparate organisms, then it cannot be identical with any one physical property (see also Hill, 1991).

Other proponents of what I am calling a reductive explanatory gap include Kim (1998), Chalmers (1996) and Jackson (1993, 1997). For a critique of the sort of reduction scheme used by the advocates of the explanatory gap, see Block & Stalnaker (1999), Hill & McLaughlin (1999) and Marras (2002, 2000). The model of reduction advocated by the above writers has many similarities with the “analytic functionalism” originally proposed by Lewis (1966) and Armstrong (1968). On this view, mental properties are given a functional definition—are defined in terms of a causal role—and are then identified with whatever (physical) property is found to occupy that role.

This is Hillary Putnam's (1975) example of a hypothetical molecular composition of “twin water”: waterish stuff that is identical to real water except for its molecular composition.

In Levine's words: The story goes something like this. Molecules of H₂O move about at various speeds. Some fast‐moving molecules that happen to be near the surface of the liquid have sufficient kinetic energy to escape the intermolecular attractive forces that keep the liquid intact. These molecules enter the atmosphere. That's evaporation. The precise value of the intermolecular attractive forces of H₂O molecules determines the vapor pressure of liquid masses of H₂O, the pressure exerted by molecules attempting to escape into saturated air. As the average kinetic energy of the molecules increases, so does the vapor pressure. When the vapor pressure reaches the point where it is equal to atmospheric pressure, large bubbles form within the liquid and burst forth at the liquid's surface. The water boils. (1997, p. 549)

There is a problem, however: folk theory and chemical theory have disparate vocabularies. “Boil” is not part of chemical theory. How then could the fact that water boils at 100(C be deduced from chemical theory? The language of the folk theory and of the chemical theory must be bridged. Hence, “bridge principles” (Levine, 1993, p. 550) will be required. These bridge principles take the form of definitions of the terms “into the proprietary vocabularies of the theories appealed to in the explanation” (Levine, 1993, p. 551). Thus, for Levine, bridge principles allow the causal roles definitive of the pretheoretic macro features of the explanadum to be identified with the underlying chemical mechanisms of the explanans. So, for example, colorlessness is a superficial property of water, but not a chemical property. To explain why water is colorless in terms of its molecular structure, then, “we need to reduce colorlessness to a property like having a particular spectral reflectance function” (Levine, 1993, p. 551). This particular spectral reflectance function plays the causal role definitive of colorlessness, and so, since H₂O has this spectral reflectance function, we have an explanation of why water is colorless: colorlessness is a particular spectral reflectance function; water is H₂O and H₂O has this spectral reflectance function.

Kim, Chalmers and Jackson, on the other hand, believe that metaphysical consequences follow from this line of reasoning. Kim (1998), for example, gives essentially the same account of reductive explanations, and like Levine, he agrees that such a strategy will not work in the case of qualia. From these sorts of considerations, Kim concludes that qualia are quite likely irreducible epiphenomena. But for Kim, unlike Chalmers (1996), such epiphenomenalism amounts to eliminativism. According to what Kim calls Alexander's Dictum, genuine (non‐Cambridge) properties must have causal powers and the irreducibility of qualia precludes their causal efficacy. As Kim puts the point, if you stay with physicalism, you come to [a] choice point: either you retain supervenient and yet irreducible (that is, nonfunctionalizable) mental properties, say qualia, but accept their causal impotence, or you embrace mental eliminativism and deny the reality of these irreducible properties. … [But] on Alexander's criterion of what is real, eliminativism and epiphenomenalism both come to pretty much the same thing: mental irrealism. (1998, p. 119) So not only might this version of the explanatory gap deny the physical nature of qualia, it may even challenge their reality; the qualities of conscious experience have failed to find a place in the causal economy of the natural world.

Both quotes are taken from Güzeldere's (1997, pp. 47–8) very informative survey of the historical and contemporary consciousness literature; Tyndall is there quoted at greater length.

Güzeldere (1997, p. 4) suggests that the puzzlement over the nature of conscious experience is taken—by those who feel such puzzlement—to be the result of either (1) the limitations of the materialist paradigm or (2) our own cognitive limitations. As Güzeldere (1997, p. 48) points out, Nagel acknowledges the sort of position propounded by McGinn and Tyndall (see Nagel, 1986, pp. 48–9).

In a similar vein, I think, Leibniz (1714) points out that if a brain were made as large as a mill such that one could walk around inside it, one would still not see any thoughts or feelings—thoughts and feelings are not the things we see when we observe a brain.

For contemporary accounts of the mind‐body relation that are similar to Feigl's dual‐access theory, see Perry's (2001) “two‐way” reflexive theory of sentience; see also Loar (1997, fn. 14) who advocates a mode of presentation account of first‐person experience. Although I may demur with some of the details, I agree with the general tack of these approaches. My aim here is to demonstrate how this elegant and parsimonious account of the qualia‐body relation dissolves the various explanatory gaps distinguished above.

For HOP theory see Armstrong (1968) and Lycan (1997); for HOT theory see Carruthers (2000).

What's more, I am sympathetic with Perry's skepticism concerning the theoretical benefits a HOT or HOP theory brings to the “mystery of sentience” and the resolution of the explanatory gaps. The amazement that a brain state should feel like something for its owner is not, I believe, dissipated by positing a HOT or HOP that represents it. As Perry puts the point: One may say that it is somewhat amazing and mysterious that it can be like something to be in a state. That is correct, but however amazing it may be, it is true. We gain nothing by pushing the mystery somewhere else in the mind. The states of our body, often carrying information about the external world, put our brains in states it is like something to be in. Amazing, but true. The mystery of sentience does not come when we perceive those states, or think about them, or know them; it comes when we are in them. (2001, p. 46) If it were the case that a first‐order state is only conscious when there is a HOT or HOP representing it, then I would be inclined to identify qualia—the raw feels of conscious experiences—not with the first‐order state, but with the HOT or HOP that makes it conscious. But again, what have we gained by pushing the locus of experience back to the second‐order thought or percept? Is it not just as mysterious and amazing that the neural activity underpinning a HOT or HOP should feel like something for its owner?

For a review and critique of Tye's PANIC model of consciousness, see Levine (1997).

According to Loar, what Mary gains is a new concept, specifically a phenomenal concept. Phenomenal concepts are recognitional concepts and they may very well pick out or classify the same things as certain neuro‐functional concepts.

In defense of his qualia‐externalism, Tye points to the apparent transparency of conscious experience. Consider the perception of a red sunset. When we turn our focus to the color experience, Tye contends, we inevitably end up attending to the external object of the experience, not something internal or intrinsic to the percipient. “This is why,” Tye writes, you cannot find any technicolor qualia, any raw feels, by peering around inside the brain (with or without a flashlight). They simply are not in there. To discover what it's like, you need to look outside the head to what the brain states represent.… So systems that are internally physically identical do not have to be phenomenally identical. (1995, p. 151) In the final section of this paper, I give the dual‐access account of why we apparently do not see any technicolor qualia while peering around the head. However, space does not permit a detailed critique of Tye's argument from transparency. Nonetheless, two points are worthy noting here. First, as Levine (1997) points out, one can recognize the representational nature of sensory qualia without externalizing them. Second, there are many cases in which conscious experiences are had in the absence of perceiving any external objects (e.g. day and night dreams), and indeed, even perceptual experiences can be evoked without any external objects, as when one's brain is stimulated directly by a neurosurgeon. In these cases, it is far from clear that there is any transparency in Tye's sense.

For example, Levine (1997, pp. 109–12) considers a race of aliens that are neurophysiologically identical to humans except for the fact that they have color‐inverting lenses—lenses that invert the color information reaching the retina. Because, for Tye, qualitative character is a function of the distal properties it represents, he is committed to the view that these creatures have qualitatively identical color experiences as humans when presented with the same distal stimuli. And, one might query, what of creatures that represent these same distal stimuli through the use of radically distinct sense organs? Arguments from the inverted and absent qualia thought experiments are much more manageable on a view, like dual‐access theory, that construes qualia as intrinsic to the organism. Levine (1997, p. 105) makes this point when he writes that, “if you opt for a view of qualia as intrinsic physical properties, as the identity theorist does, the inverted and absent qualia aren't so much of a problem. … It is quite plausible to maintain that physical duplicates must share all of their (intrinsic) properties.” By hypothesis, my physical duplicate has C‐fibers if I do, and if pain = excited C‐fibers, then there is little reason to doubt my physical duplicate's capacity for the experience of pain.

Hardin's remarks are based on experiments first conducted by Hurvich et al. (1968).

Interestingly, Levine (1997, pp. 111–12) argues that Tye's qualia externalism leads to some very counter‐intuitive consequences concerning one's introspective access to one's own qualia. However, a discussion of these issues goes beyond the scope of this paper.

Nor should we confuse dual‐access with Block's (1997) notion of “access consciousness” which he distinguishes from phenomenal consciousness; it is phenomenal consciousness that we are concerned with here.

Marras (2000, 2002) has noted a number of problems with the reductive explanation model, especially Kim's (1998) version, and has suggested that, once properly understood, a functional model of reduction is not substantially different from the Nagelian model (Nagel, 1961).

Of course, beliefs about our own conscious experiences represent what I have called cognitive access to those experiences. And since referential opacity is only relevant for doxastic contexts, these points about the referential opacity of phenomenal descriptions applies only to cognitive access—to our beliefs about, and descriptions of, our conscious experiences.

Pyramidal cell activity is Block & Stalnaker's (1999) hypothetical reduction base for conscious experience.

To paraphrase the insightful definition of consciousness in Drever's (1964) A Dictionary of Psychology. Of course, the visual metaphor is misleading, except, perhaps, in the case of cognitive access. What is important here is the recognition that some brain states can be accessed from more than one perspective.

Block & Stalnaker make this point by analogy when they write that “[i]f we believe that heat is correlated with but not identical to molecular kinetic energy, we should regard as legitimate the question of why the correlation exists and what its mechanism is. But once we realize that heat is molecular energy, questions like this will be seen as wrongheaded” (Block & Stalnaker, 1999, p. 24).

Additional information

Notes on contributors

Liam P. Dempsey

Liam P. Dempsey, Dalhousie University, 5252 Tobin Street, Apt. 704, Halifax, Nova Scotia, Canada B3H 4K2, email: [email protected]