On Tuesday, I went to hear Ray Tallis and Robin Dunbar speak at the British Academy, under the title "Neuroscience and Neuromania". The point of departure was Ray's new book, Aping Mankind: Neuromania, Darwinitis and the Misrepresentation of Humanity. Ray's thesis was that while neuroscience was great, we should not expect it to explain everything about us. In particular, the key to understanding what is special about us, our self-consciousness and our capacity for creative thought and action, is not to be found there. Furthermore, this is not just a matter of complexity, of the fact that a neuron-by-neuron description of a human being would be hopelessly unwieldy. I agree with his conclusion, to the extent that I would certainly place my bet for it, rather than against it. But I do not think that his reasons are sufficient to make the conclusion even close to certain.
The humanities are definitely not going to become redundant as keys to our collective self-understanding any time soon. But I would not go so far as to say "never". There was some discussion of whether Ray underestimated the potential of neuroscience. It has, after all, made great strides in recent decades, and no-one knows how it might progress in the next fifty years. As Ray pointed out, we cannot place much weight on such promises, any more than we can on an assurance that "your cheque is in the post". And if we look at current neuroscience and expect more of the same - greater refinement in distinguishing the physical correlates of mental states, for example - then we can easily make a case that the contribution to our self-understanding of future neuroscience will be pretty well as limited as the contribution of present neuroscience.
But that would be to neglect the possibility of fundamental reconceptions, within neuroscience. We cannot yet see what such a reconception might be like, but we can draw a parallel. It would be natural to think that the unity of consciousness depended on the activity of some locatable part of the brain. But it now seems that it is more likely to be a consequence of some overall co-ordination that is located nowhere in particular within the brain. (See for example chapters 44 and 47 of the Blackwell Companion to Consciousness, ed. Velmans and Schneider, 2007.) If we can make other leaps of this kind, from looking at concrete neurons to looking at abstract structures, without losing touch with the biochemical underpinnings that make the subject neuroscience rather than anything else, the explanatory reach of the subject might turn out to be far greater than current work would lead us to expect.
I have another worry about Ray's case against the pretensions of neuroscience in the hands of the neuromaniacs. To set it out, I must outline the two central themes in his case.
The first theme was that we can only understand what we do if we do not have too narrow a focus. A little action is part of a whole life. He gave the example of the Libet experiments. An experimental subject moves a finger, as the experimenter directs. But the subject also got himself to the laboratory. Before that, he decided to participate. He looked in his diary, and saw that an hour at the laboratory would fit nicely in between the other things he had planned for the day, perhaps some shopping and a drink with friends. And so on, perhaps right back to an education that made him feel that it was worth contributing to the advancement of knowledge, and perhaps right forward to an idea that one day, he might be able to tell his grandchildren that he participated in experiments that proved to be of great significance.
The second theme was that certain things, absolutely crucial to our condition and our experience, are missing from what the natural sciences provide. These are points of view, intentionality and secondary qualities. The natural sciences describe the world without describing it as it would appear from any time or place, they find no place for the outward-directed gaze that underlies intentionality ("I am thinking about, looking at, acting upon this glass of water"), and they do not recognize secondary qualities such as colours. We, on the other hand, are keenly aware of our locations in space-time, locations that allow us certain experiences and actions, but make others impossible. We also think about, pay attention to and act upon particular things, and our experience is replete with secondary qualities.
Now consider the matter-maniac, a replacement for the neuromaniac. This matter-maniac takes the point about wider context, and acknowledges that we need to look way beyond the contents of someone's head. We need to look at the whole planet, over a long stretch of time, in order to understand what people do. But, the matter-maniac maintains, we can do all we need by talking in terms of particles and their trajectories.
We can identify thoughts with certain dispositions of particles, not just at the moment of thought but before and after (in order to provide the context that makes sense of thoughts). We can do the same for actions. We can identify points of view by working out what information can reach a given person, and what lies within her influence - at the most extensive, her past and future light-cones. We can accommodate intentionality by identifying relations between pre-existing brain states, their origins in prior events, current dispositions that would mean that any changes in objects that were the focus of attention would produce certain responses in the subject, and the practical effects of changes of brain states that would normally be identified as decisions to act in certain ways. Finally, we can correlate secondary qualities with the influences of some particles (those that reflect green light, for example) with others (in brains when people experience green).
The problems with the matter-maniac's approach are legion. I do not think it is a sensible approach at all. But anyone who wants to see off the neuromaniac decisively without falling into the clutches of the matter-maniac, and to do so on the basis of the need for broad focus and the need to accommodate points of view, intentionality and secondary qualities, must show that it is a bad approach. Furthermore, it would be dangerous to do so on the basis of the many specific difficulties in a particle-by-particle approach. The matter-maniac's approach is easily generalized, to something that says "We only need to find one description, of any sort, that is couched in the terms of physics, chemistry and biology, that has a broad focus and that somehow accommodates points of view, intentionality and secondary qualities. Then we can conclude that the neuromaniac has only been shown to be mistaken in detail, not in principle". Such a general threat must be seen off with a general response.
One general response that was offered after the formal proceedings is interesting but it does not, in my view, do the trick. This is the response that the matter-maniac has helped himself to the notion of a point of view, in describing the world in material terms. That is, the matter-maniac must conceive himself as standing somewhere and looking at the world. He relies on the notion of a point of view, and does so in a way that he cannot support merely by identifying the consequences of the people within the world that he conceives having the spatio-temporal locations that they have.
This response fails if, as I think is the case, we can conceive and discuss a world without, even in imagination, perceiving it. If, for example, we want to set out a structure of space-time, we give the mathematical formulation. Then we can give data on how the space-time is populated (beyond any population that is necessary to give the structure itself). Just to see off a challenge to my counter to the response, I take it that we can set out some mathematical data without worrying about where they are written down or the point of view from which we contemplate them. Mathematical structures (including sets of data) are not spatio-temporally located, even though representations of them are. This may look like dangerous Platonism about mathematics. I regard it as harmless Platonism. Mathematical structures just are, without being here or there, although physical instantiations of them are here or there. If one thought of the structures as here or there, the flavour of serious mathematical thought would be vastly different from what it is.
Suppose that this counter to the response failed. Then there would be another counter. Even if the matter-maniac must perceive the world he conceives, whether in imagination or in reality, he could perceive without bringing the fact of his perception into consciousness and analyzing it. He would not need to recognize the existence of his point of view. He would therefore have no need of the notion of a point of view, applied to himself, logically prior to the points of view that he analyzed out of his data on the conceived world.
If neither of these counters worked, then the independence of the natural sciences from more humanistic disciplines would be under threat. Reductionist ambitions that would create dependence of the humanities on the natural sciences may very well be excessive. But to create the reverse dependence would be implausible in the extreme.