Jakob Hohwy
Department of Philosophy
Monash University

Consciousness. We have come to expect science to be able to explain all sorts of phenomena in the world (global warming, hereditary diseases, life – you name it). Consciousness is an anomaly in the success story of science for there is a real question whether science, in particular neuroscience, can explain much about what consciousness is. A good question to ask is how and to what extent consciousness resists scientific explanation. That might tell us something about what is special about consciousness.

It is easy enough to indicate what consciousness is. Consciousness is experience. It is the kind of thing that happens to you when you wake up in the morning after a dreamless sleep, and sensations and perceptions begin to flood in. It is what it’s like to have thoughts about world peace, to see a red apple, to have a pain, or a pang of jealousy.

These are all examples of what consciousness is. To appreciate this indication of what consciousness is you need to imagine yourself having the exemplified experiences. It is not possible to give a concise definition of consciousness (somewhat as one could give a concise definition of what it takes for something to be a smoke alarm). Try as you might, a precise definition always seems to leave out the really important part about consciousness, that it feels like something to be conscious.

This is a first indication that consciousness is set apart from most other fields of scientific study. Normally, scientists will have at least a rudimentary, working definition of the phenomenon they wish to study (even the vitalists would have operated with a vague definition of life; they just couldn’t imagine anything in the world that could explain life).

Neuroscience is an interdisciplinary effort comprising biomedical sciences, psychology and computational theory, with the understanding that they all converge on cognitive functions. How can neuroscience then study consciousness if it doesn’t have a firm functional definition? How does it pin down consciousness? In practice, the neuroscience of consciousness simply assumes that the subjects being studied have the right kinds of conscious experiences at the right times (namely, when in the scanner). There is therefore an element of trusting the subject. Scientists trust the subjects’ reports that certain cognitive functions (such as discriminating faces from houses) are accompanied by consciousness.

Some view this as a serious problem for viewing consciousness science as an objective endeavour. Experience is inherently subjective, so we have no way of performing an independent check on whether subjects do indeed have the experiences they claim to have. I don’t think the subjectivity of experience is serious enough to be what prevents scientific explanation of consciousness. Science is used to operate with unobservables (e.g., events beyond the light cone) and we can treat conscious states as unobservables. We can come to rationally believe in certain unobservables if postulating them forms part of the best explanation of the evidence that we do have (similar to how one might believe a burglar has been around because that is the best explanation of the broken window). So, we can come to rationally believe in a certain interpretation of a subject’s introspective reports if that interpretation is part of the best explanation of those reports and the subject’s other behaviour (the circularity here is benign and commonplace, as the burglar example shows). Subjectivity and privacy is a salient aspect of consciousness but I believe it is not what should make us think consciousness is forever beyond the ken of science.

There is a deeper problem here. It may be that the subjectivity of consciousness and the lack of a formal definition do not prevent neuroscience from finding the neural correlates of consciousness. But, having found the correlates, what happens next? Ideally, neuroscience should explain facts about consciousness in terms of facts about the neural activity. If not then we are only rationally justified in believing that brain and mind are correlated, not that consciousness is in fact a wholly natural phenomenon. Neuroscience can of course be used to explain many things. For instance, memory in sea slugs is explainable in terms of synaptic strength and gene expression. But it seems only things defined in functional terms can be explained (e.g., one finds the neural mechanism that explains memory function). The problem is then clear: consciousness cannot be given a functional definition so it cannot be neuroscientifically explained.

I’ll motivate this a little more and then indicate what I think it shows about consciousness. Assume that we have discovered some neural mechanism. In fact, assume we have discovered all the neural mechanisms that there are. These mechanisms will explain a whole host of psychological functions (memory, representation, etc). Among these there will be functions we associate with consciousness. For example, the function of bodily damage to bring us in an internal state that causes us to withdraw our hand form the fire, where we associate the internal state with being in the conscious state of feeling pain. The problem is that the neural mechanisms only concern the causal transactions among states not the nature of the states themselves. That is, the neural story fails to capture what is distinctive about pain, namely that it hurts. We can put this differently. The neural explanation of the supposed pain would work just as well for a creature that is like us physically but that doesn’t feel any hurt when it “suffers” bodily damage, internal states, and subsequent withdrawal behaviour. This shows that the neural explanation is blind to consciousness: its best explanations can ever only capture input-out causal profiles.

To many, something like this is the problem of consciousness – the mind-body problem (though it is not always expounded in terms of explanation). There are no easy solutions to it. But I don’t think it captures our notion of explanation entirely. We rarely attempt to explain phenomena in their entirety and all the way down. Instead, we assume various things about them, and then we use a particular contrast (“why e rather than e*?”) to pinpoint a particular aspect of the phenomenon in question. For a conscious state we assume the creature is conscious as such, and we assume it is in, say, pain, and then we may search for the neural mechanism that explains why it is in a sharp pain rather than a dull pain. Sometimes there will be an isomorphism between the neural mechanism and the particular aspect of the conscious state. Such isomorphisms many people find genuinely explanatory. This contrasts sharply with how many people agree to the case for the unexplanatoriness of neuroscience that I outlined above. This suggests that neuroscience may explain aspects of consciousness, even if it cannot ultimately place consciousness in the material world.

The considerations so far add up to a somewhat messy picture. In order to be within the purview of scientific explanation, consciousness must have some functional aspect and yet people argue that it hasn’t. Instead, they point out, conscious states are intrinsic since we can conceive of someone having such states without any functional scaffolding. Whereas we have good reason to believe consciousness is a material phenomenon since it correlates so well with neural activity, we also have reason to believe the material and the conscious belong to different levels.

These considerations pull in different directions and seem to preclude a clear conception of consciousness, quite apart from worries about the metaphysical question of its nature. I think a more clear conception can be had, if we think of consciousness as a dispositional property. Dispositions are captured in functional terms (being fragile is being disposed to break when struck), they have intrinsic bases (a certain molecular structure of the fragile cup), mechanisms involving the base can explain the manifestation of the disposition (changes in the molecular bonding under impact). Finally, the manifestation of the disposition (the breaking) and its base are in some sense at different levels. A dispositional account of consciousness could therefore unify most of the problematic elements we have discussed for consciousness. Being conscious would stand to consciousness as breaking stands to fragility, and having a neural correlate of consciousness stands to consciousness as having a certain molecular basis stands to fragility.

On this dispositional conception, we can account for why it seems as if science can explain consciousness even if it cannot place it in the natural world. For science is well-placed to investigate relations between dispositional bases and their manifestations, just as science is well-placed to investigate relations between causes and effects. This is so even if science cannot explain the manifestation relation itself, or, indeed, explain causality itself. The ultimate question for consciousness science, on this conception, is therefore “what is it like to be a manifesting disposition?”

Recommended literature:

Examples of neuroscience:
Bartels, A. and S. Zeki 2004. The neural correlates of maternal and romantic love. NeuroImage 21(3), 1155-1166.
Blakemore, S.-J., Oakley, D.A. and Frith, C.D. 2003: Delusions of alien control in the normal brain. Neuropsychologia, 41, 1058–1067.
Frith, C., S.-J. Blakemore, et al. (2000). “Explaining the symptoms of schizophrenia: Abnormalities in the awareness of action.” Brain Research Reviews 31:
357-363.
Leopold, D. A. and N. K. Logothetis (1999). “Multistable phenomena: changing views in perception.” Trends Cogn. Sci. 3: 254-264.
Tong, F., K. Nakayama, et al. (1998). “Binocular rivalry and visual awareness in human extrastriate cortex.” Neuron 21: 753-759.
Rees, G. 2007. Neural correlates of the contents of visual awareness in humans Phil.Trans R Soc 362: 877-886.

Methodology in the science of consciousness:
Frith, C., R. Perry, et al. (1999). “The neural correlates of conscious experience: an experimental framework.” Trends in Cognitive Sciences 3(3): 105.
Driver, J., Haggard, P., Shallice, T. 2007. Mental processes in the human brain. Phil.Trans R Soc 362: 756-9.
Hohwy, J. and C. D. Frith (2004). “Can neuroscience explain consciousness?” Journal of Consciousness Studies 11(7-8): 180-198.
Hohwy, J. The Search for Neural Correlates of Consciousness. Philosophy Compass 2/3: 461-474, 2007.
Jack, A. I. and A. Roepstorff, Eds. (2003). Trusting the Subject Vol I. Exeter, Imprint Academic.
Jack, A. I. and A. Roepstorff, Eds. (2004). Trusting the Subject Vol II. Exeter, Imprint Academic.
Lipton, P. (2004). Inference to the Best Explanation. London, Routledge.

The philosophical problems:
Chalmers, D. (1996). The Conscious Mind. Harvard, Oxford University Press.
Gold, I. & Stoljar, D. (1999), ‘A neuron doctrine in the philosophy of neuroscience’, Behavioral and Brain Sciences 22, pp. 809–69.
Levine, J. (2001). Purple Haze: The Puzzle of Consciousness. Oxford, Oxford University Press.
Nagel, T. (1974). “What is it like to be a bat?” Philosophical Review 83: 435-450.