Summer is here and I have finally started on my summer reading list. First up was Michael Gazzaniga’s new book The Consciousness Instinct. Gazzaniga is of course we’ll known for his work on split brain patients and for helping to found the the discipline of cognitive neuroscience. I was very excited to read the book but after having done so I am very disappointed. There are some interesting ideas in the book but overall it does not strike me as a serious contribution to the study of consciousness.
The book begins with the standard potted history of the mind body problem with Descartes invoked as the primary villain. It was Descartes who initiated the-brain-is-a-machine ethos and Gazzaniga thinks that is a mistake. This part of the book was well written but could be found almost anywhere. He then goes completely off the rails and invokes quantum mechanics as a non-mechanical foundation for solving the mind-body problem. In particular he invokes the notion of complementarity as his solution. According to him Quantuum mechanics tells us that a physical system can be in two different states at once (p. 175). So the brain can be a mechanical system and also a mind at the same time. No problem.
I am of course no expert on quantuum mechanics (though I have put it in a fair amount of time trying to figure it out). But as far as I understand it this is a gross misuse of the idea of complimentary. Quantum mechanics does not say that a physical system can be in two contradictory states at the same time! Rather what it says is that the state of the system *before measurement* cannot be described by classical concepts like ‘wave’ or ‘particle’ yet once a measurement is made (and depending on the type of measurement we make) we will find that it does have one of these properties (and had we done the measurement different we would have found that it had the other property). How, then, should we think of poor Schrodinger’s cat? Isn’t the poor cat both dead and alive (as Gazzaniga says on p. 181)? Not as I understand it! When we have a vector, represented by |A> and we add it to another vector |B> then, yes, we do get a new vector that represents the state the system has entered but saying that 1/2|Alive> + 1/2|Dead> represents the cat’s state before measurement doesn’t mean it is both dead AND alive; it means that when we measure it it will EITHER be dead OR alive (with probabilities given by the 1/2).
But what about before we measure it? What state is the cat in then? As far as I understand it quantum mechanics (i.e. the mathematical formalism) is silent on that question but the reply I prefer is that the cat has no determinate properties before the measurement.
But all of this is highly controversial and does not help us at all with the mind-body problem! Suppose, as Gazzaniga assumes, that the mind and brain are two irreducible complimentary descriptions of the single system, then we would only be able to know (i.e. measure) one of the at a time, at the expense of the other. But that is manifestly not the situation. We can measure our own brain activity even as we are having conscious experiences produced by/identical with that neural activity. No complementarity required.
I am leaving out a lot of the details, and as I said some of his views are interesting, but what is it about consciousness that drives people to these kinds of extreme intellectual gyrations? Why do people trust their intuitions so much that they are ready to jettison all of the progress made by psychology and neuroscience as wasted time?
Richard Brown 
Shame to hear that about Gazzaniga’s book—it seems he’s confused superposition with complementarity. Now, the two aren’t completely separate, either, but simply pointing to a superposed state (such as that of the cat) doesn’t do complementarity justice: the real idea is that there are observables A, B such that if a system is in a determinate state regarding A, it can’t be in a determinate state regarding B. This can also be expressed in terms of complete sets of commuting observables—such a set gives you all the information you can simultaneously hold about a system. In classical physics, there is only one such set, and it tells you all there is to know about a system; in quantum physics, many such sets exist, and having complete information about one entails incomplete information about others. (The connection to wave / particle complementarity comes because waves are eigenstates of the momentum observable, while particles—point-like localizations—are eigenstates of the position observable.)
I do think that one could at least take some inspiration for the philosophy of mind from this relationship, because it yields a way for two phenomena to be related that doesn’t reduce to those you could draw up a priori, from simply considering combinatorial possibilities. That is, if you have two sets of phenomena X and Y, you could:
1. consider both separate (dualism)
2. consider both to overlap (interactionism)
3. consider either a subset of the other (materialism / idealism)
4. consider only one of them ‘real’ (eliminativism)
5. consider them identical (panpsychism, dual aspect theory)
6. consider both to stem from some third set of phenomena (neutral monism)
And that seems to be pretty much it. Yet, the discovery of complementarity has taught us that there are more than just those ways for two domains to be related: you could draw up an analogue of the above list for the relationship between waves and particles (say), and neither of these possibilities would actually capture what we observe. So one could make either of two arguments for the relationship of mind and brain: they could be related in the same way as complementary properties in quantum mechanics, or, there could be yet another way of being related we haven’t thought of so far. But I’ve never really seen this idea worked out in a way that I found convincing.