From Towards a Science of Consciousness 3
Section 6: Evolution and
the Function of Consciousness CogNet
Proceedings
If qualia evolved they must belong to the Physical World. Yet current physics and chemistry, which are supposed to describe this world, has no place for them. On the other hand, if qualia evolved we should expect there to be brain machinery devoted to their production-"qualagens"-and a phylogeny of qualia and qualagens to be discovered.
There is a duplicity in terms such as perception, awareness, thought, intelligence, behavior. We should say that these all refer to attributes which are indeed often and importantly to do with consciousness. But they are not necessarily to do with consciousness. They need the adjective conscious or unconscious to make their meanings clear, since there are unconscious forms of each of them. If we concentrate our attention exclusively on the unconscious forms of perception, awareness, etc., we will indeed be able to keep to the language of science-of molecular biology, computing, or whatever. No one will accuse us of being unscientific. But then we will have avoided the question of what consciousness actually is.
For example we may be tempted to say that "consciousness" just means "awareness." That sounds sensible and pragmatic. Indeed it is one of the meanings to be found in the O.E.D. But then we might go on to say that awareness (of a sort) is exhibited by thermostats and burglar alarms, as well as now highly sophisticated computer based recognition systems; adding perhaps that such forms of machine awareness are getting better and better and sooner or later will come to approach human consciousness. Such a line of argument goes wrong at the first step in failing to distinguish between conscious and unconscious awareness.
Or we may be tempted to equate consciousness with thinking or even with intelligence and use the duplicity of these words to slide away from the real problem. It is reasonable to hope that since thinking, intelligence and consciousness are connected then advances in Artificial Intelligence will help in our understanding of consciousness. And it is part of the traditional strategy of science to go first for those parts of a problem that we have the readiest means to tackle-and seemingly daunting problems sometimes dissolve away if we proceed like this. In support of taking such an attitude to the problem of consciousness one might point, as Dennett does, to the way in which advances in molecular biology have made redundant such earlier ideas as the vital force (Dennett 1991). Indeed vital force was an unreal idea. It was a misunderstanding. But qualia are not like that. Agonizing pain is not a misunderstanding. In discussions of consciousness, feelings and sensations should be on center stage. (Chalmers 1995, Humphrey 1992, Cairns-Smith 1996).
I take the following to be examples of essentially conscious attributes, because they are attributes for which, it seems to me, there are no unconscious equivalent forms:
a feeling
an emotion
a sensation
a mood
These are all qualia, and I define consciousness informally, and for present purposes, in these terms: a state of consciousness is some kind of arrangement or organization of qualia.
Perhaps the most significant thing which qualia have in common is precisely that our current hard sciences of physics, chemistry etc., have no place for them. The purpose of this chapter is to show that qualia should have such a place-and that there is both bad news and good news here. The bad news is that there is something deeply deficient about current physics and chemistry. The good news is that qualia should be accessible to science.
There is a natural dualism in the distinction between the known and the unknown. For example there is a distinction between what as scientists we think we understand in principle about the mind, and what we do not understand at all. Such frontiers are forever changing, the dualism I speak of is strictly provisional, but at the moment it lies neatly between the unconscious and conscious parts of our mind.
We now realize that most of what we do is unconscious (Baars 1988). It is this part of our mind, what we might call our Greater Self, that we can in principle understand in terms of nerve cells and circuitry. To say that our Greater Self is a big unconscious control computer is not so far off. The other part of our mind comes and goes: the Evanescent Self, as we might call it, the conscious mind, the bit that is made of qualia, the bit that switches on every morning.
It often looks as if the brain makes qualia as some kind of add-on activity-because it is conscious only sometimes and to different extents: and because at different moments it may be more or less conscious, and in different ways. An example of this kind of thing is in the acquisition of a skill, where the unconscious systems increasingly predominate. Ask yourself how one rides a bicycle: how for example one can ride a bicycle slowly without falling sideways. There is evidently much manipulation of the handle bars needed, but if you ask most people who can ride bicycles which way they should turn them to stop falling, say, to the left they will probably suggest that you should turn the handle bars to the right. Consciously they seem not to know that this will quickly tip them over. Fortunately their unconscious Greater Self knows the correct answer to the question ( or anyway it knows what to do.
Functional brain imaging has revealed distinctive alterations in locations and patterns of brain activity during learning tasks (Posner and Raichle 1994), and both Milner (1998) and Goodale (1998) have given us examples of cases where the brain seems to have distinct pathways for conscious and unconscious forms of the same sort of activity. It seems clear now that when a mental activity is qualia laden (highly conscious) or qualia free (unconscious) the brain is operating somewhat differently.
Yet this now-you-have-it-now-you-don't aspect of our consciousness, together with the now well known delays in conscious perceptions and conscious actions, has helped to give an impression that our consciousness does not actually do anything. No doubt consciousness is less intricately involved in our actions that we might like to imagine, but an argument to the contrary-that "feelings are causes"-was put more that a hundred years ago. It is an argument with devastating implications. Not nearly enough attention has been paid to it (Glynn 1993).
The O.E.D. gives, as one of its examples of uses of the term consciousness, a paragraph written by Thomas Huxley in 1866:
We class sensations along with emotions, and volitions, and thoughts, under the common head of states of consciousness. But what consciousness is, we know not; and how it is that anything so remarkable as a state of consciousness comes about as a result of irritating nervous tissue, is just as unaccountable as the appearance of the Djin when Aladdin rubbed his lamp, or any other ultimate fact of nature.
It seems a reasonable facing of facts, but then in 1874 he expressed a more positive but, to my mind, more dubious view:
. . . it follows that our mental conditions are simply the symbols in consciousness of the changes which take place automatically in the organism; and that to take an extreme illustration, the feeling we call volition is not the cause of a voluntary act, but the symbol of that state of the brain which is the immediate cause of that act. We are conscious automata. (Huxley 1874)
William James would have none of it. In addressing the "automaton" issue he gives a robust defence of the efficacy of feelings:
. . . common-sense, though the intimate nature of causality and of the connection of things in the universe lies beyond her pitifully bounded horizon, has the root and gist of the truth in her hands when she obstinately holds to it that feelings and ideas are causes. (James 1890)
And this was not just oratorical eloquence. By 1878 James had found a wonderful argument for the efficacy of feelings based, paradoxically, on the very idea that Huxley had done so much to promote: Darwin's idea of evolution though natural selection (Glynn 1993, Richards 1987).
In terms of more current language the argument can be put like this. Feelings such as hunger and lust seem so obviously adaptive, "designed" for survival and reproduction, that we can suppose that they evolved. In that case they are effects of genes, effects of DNA molecules, that is, physico-chemical effects. On the other hand feelings such as these have physico-chemical consequences too, in the adaptive activities which they encourage. This cannot be make-believe. In the warnings and encouragements of pain and pleasure it is the feeling that is effective, the feeling itself. If it were only the "neural correlates" of feelings that influenced behavior, not the feelings themselves, then there would be no selective advantage in actually feeling anything. And even if we did have feelings as some kind of accidental side effect there would be no reason to expect them to be appropriate. What would it matter what feelings we had if they could have no effects? The appropriateness of qualia, their whole evolution, would be a mystery.
There seems to be no way round it: if they evolved, qualia must have both physico-chemical causes and physico-chemical effects-and thus good enough entrance qualifications to be admitted to The Physical World.
It is not that the hard sciences cannot deal with qualia: that might be of no consequence. Science has no pretensions to explain everything. What is so devastating about the qualia Bomb is that science ought to be able to deal with qualia. They are part of the physical world, because they evolved; and they are part of the machinery of our behavior, along with ion pumps, action potentials, reflexes and so on. A hundred years ago there was a similar situation in physics. The way in which the colors of glowing coals changed with temperature clearly should have been understandable in physical terms and yet it was to need a revolution in physics for such an understanding to come about. Now it is such things as the sensations of color that are demanding to be let into the Society of Material Things-alongside heredity, and so much else in biology that can now be explained in terms of physics and chemistry and the theory of evolution.
So it is our matter theory that is set to change. Nothing new here, we might say. Think how our ideas of matter have shifted since Greek times. About 420 BC Democritus had a Theory of Everything. There were only Atoms, Motion and Void. The persistence of the Universe depended on the durability of the atoms while the variability of the Universe was to be explained because the atoms could be arranged in different patterns. It was a wonderful idea due to reach a mathematical perfection in the nineteenth-century kinetic theory of gases. But it was not rich enough to explain everything about the material world. Newton was to add forces operating across space and suggested that such would be important in holding the smallest parts of materials together: "Have not the small particles of bodies certain powers, virtues, or forces by which they act at a distance . . ." (1730/1979); Faraday and Maxwell were to see fields to be as real as atoms (1844); Planck and then Einstein would upset the whole idea of material substance: quantum energy became the ultimate stuff whose ultimate grain was action. And then, with Heisenberg, atoms had ceased to be things at all (1959).
But there are other fabrics woven differently from the quantum energy, light for example. A laser beam is produced by and interacts with molecular matter but a laser beam is not made of atoms and molecules. Light is a different way of arranging the quantum energy.
So here is a tentative minimal conjecture: feelings and sensations are yet another way of arranging the quantum energy. Like light they are produced by and interact with molecular matter-although so far only in brains.
To the extent that a state of consciousness is "an organization of qualia," we can claim to be some way along the road to understanding in physical terms what a state of consciousness is: we know quite a lot about the crucial "organization" part of the prescription, about brain pathways and switch-gear for qualia production and control. For example we know quite a lot about what has to happen for pain to be felt, how and why it is wired up the way it is. But how such sensations themselves are made, still eludes us. This may be partly because the mechanisms have not been consistently looked for, in a belief that qualia are irrelevant and/or hopelessly "non physical" concomitants of normal brain activity. But the evolutionary argument shows that qualia are indeed physico-chemical-at least in the sense of having a place in some future physics and chemistry-and if indeed qualia evolved, then there must be machinery that produce these effects in the brain. Let us call such things "qualagens." The odds are that, like most biological machinery, the qualagens are at root molecular, with protein molecules as key components. I am inclined to agree with those who think that feelings and sensations are large scale effects arising from vast numbers of microscopic processes, and that these are macroquantum effects of some sort (Cairns-Smith 1996, Marshall 1989, Lockwood 1989, Penrose 1994, Jibu and Yasue 1995, Yasue 1998).
Maybe there are special proteins and hence genes out of which the qualagens are made. Alternatively qualagenesis might be a combined effect of activities of not only many, but of many kinds of protein molecules that also have other functions; so that one might only be able to say of some brain protein that "when this bit of this molecule wobbles it contributes to making such and such a quale." So perhaps the qualagens are difficult to find because the key proteins are not only widely spread, but have other more obvious functions, as in the Hameroff and Penrose microtubule conjecture (Penrose 1994).
raw perceptual sensations color sensations, smells . . .
interpretative feelings of space, motion, recognition . . .
intellectual feelings feelings of doubt, of certainty . . .
coercive feelings and sensations hunger, fear, pleasures, pains, itches . . .
volitional feelings and sensations curiosity, urges, desires . . .
And then, outside this scheme there are what we might lump together as background qualia: moods, attitudes . . .
Which kinds of feelings and sensations came first? Presumably they would have evolved originally on the back of wholly unconscious nervous systems. To judge from the way new functions usually catch on in evolution, the first step would have been an accidental side effect, a preadaptation with some marginal usefulness. But on the face of it sending neuro-electrochemical signals and making qualia are distinctly different activities, and so it is unlikely that exactly the same equipment would be ideal for each. In that case the first quale having caught on, we would then expect natural selection to have honed increasingly sophisticated devices that became increasingly dedicated to the production of this new class of effect.
We might guess that, to have caught on, the first kinds of qualia must have been both "raw" and "coercive" at the same time, making a direct connection between a simple perception and a simple action. Physical pleasures and pains can be like this. They are nice or nasty in themselves. Smells and tastes are very often like this too, and like pleasure and pain, associated particularly with the brain's more ancient limbic regions.
The in-between qualia of complex perception are presumably more recent-and more neocortical than limbic. Most of our perceptual sensations are not particularly nice or nasty in themselves: patches of color or texture, forms, motions, etc., need interpretation and only begin to push us into action at a high level of recognition-of an apple or a tiger. We often look before we leap.
More remote still would be where we have to think too, perhaps for days, in preparation for action. Then my scheme begins to look a bit too simple, because perhaps part of what keeps us thinking consciously is a coercive emotion, a desire to think it out: "a passion for reason" Demasio calls it (1994). Then we would have to say that coercive qualia are part of conscious thought too-conscious drives, not to immediate action in this case, but to more thinking.
Presumably the qualia of conscious thought are among the most recent. But perhaps we can still see connections in the words we use. Our language often makes connections between abstract thought and bodily sensations. We feel that an idea is wrong; we feel uneasy about it: we weigh up alternatives as if judging them by muscle feel: we like the sound of an idea, or find it distasteful. . . . Again and again we use qualia-words. So perhaps the qualia of contemplation did indeed evolve from more immediate and primitive forms, through successive modifications of qualagenic genes and proteins in Nature's usual branching style. If that is the case there is a tree to be found: a phylogeny of qualia and qualagens.
Baars, B. J. 1988. A Cognitive Theory of Consciousness. Cambridge: Cambridge University Press.
Cairns-Smith, A. G. 1996. Evolving the Mind: on the nature of matter and the origin of consciousness. Cambridge: Cambridge University Press.
Chalmers, D. J. 1995. The puzzles of conscious experience.In Scientific American 273:62-68; 1996.
Chalmers, D. J. 1996. The Conscious Mind: in search of a fundamental theory New York: Oxford University Press.
Damasio, A. R. 1994. Descartes' Error: emotion, reason and the human brain. New York: Avon.
Dennett, D. C. 1991. Consciousness Explained. Harmondsworth: Penguin, p. 25.
Faraday, M. 1844. Matter. London: Library of the Institution of Electrical Engineers.
Glynn, I. M. 1993. The evolution of consciousness: William James's unresolved problem. In Biological Reviews of the Cambridge Philosophical Society 68:599-616.
Goodale, M. 1998. Unconscious visual processing for action: evidence from normal observers. Abstract no. 140, Consciousness Research Abstracts: Toward a science of consciousness: "Tucson III," p. 81.
Heisenberg, W. 1959. Physics and Philosophy: the revolution in modern science. London: Allen & Unwin, chapter 10.
Humphrey, N. 1992. A History of the Mind. New York: Simon and Schuster.
Huxley, T. 1874. On the hypothesis that animals are automata and its history. In Fortnightly Review 22:555-589.
James, W. 1890/1983. The Principles of Psychology. New York: Holt 1890. Reprinted Cambridge Massachusetts: Harvard University Press 1983., p. 140.
Jibu, M., and K. Yasue. 1995. Quantum Brain Dynamics and Consciousness: an introduction.
Lockwood, M. 1989. Mind, Brain and the Quantum: the compound "I." Oxford: Blackwell, chapter 14.
Marshall, I. N. 1989. Consciousness and Bose-Einstein condensates. In New Ideas in Psychology 7:73-83.
Milner, D. 1998. Unconscious visual processing for action: neurophysiological evidence. Abstract no. 135, Consciousness Research Abstracts: Toward a science of consciousness: "Tucson III," p. 80.
Newton, I. 1730/1979. Opticks. (4th edition) Quest. 31. London 1730. Reprinted New York 1979: Dover, pp. 375-406.
Penrose, R. 1994. Shadows of the Mind: a search for the missing science of consciousness. Oxford: Oxford University Press, ch 7.
Posner, M. I., and M. E. Raichle. 1994. Images of Mind. New York: W. H. Freeman, pp. 125-129.
Richards, R. J. 1987. Darwin and the emergence of evolutionary theories of mind and behavior. Chicago: University of Chicago Press.
Yasue, K. 1998. Physics approaches to consciousness. Abstract no. 289, Consciousness Research Abstracts: Towards as science of consciousness: "Tucson III," p. 131.