Strangely enough, after reading a book that quite persuasively advocates mind-body dualism, I'm more of a materialist that I've ever been.
Chalmers's The Conscious Mind is good, very good.
The hard problem of consciousness is easy enough to explain. Here goes: no matter how well we explain the functionining of the brain--even if we could flawlessly predict and map the position and performance of every constituent atom--it will still be an open question why there should be a sense of awareness within it. In other words, the phenomenal (those first person aspects of consciousness) fails to logically supervene on the physical. Things logically supervene when it's inconceivable that one set of facts can exist without entailing another. For instance, imagine a world physically identical to ours, but with different biological facts. We can't--thus the biological supervenes on the physical.
But we can imagine a world physically identical to ours, but with different (or no!) phenomenal facts. We can imagine a person without consciousness who nevertheless functions precisely the same as his twin with consciousness. (This is the zombies argument; Chalmers actually admits that the inverted spectrum argument, i.e. we can imagine a twin who sees different colors in response to the same stimuli, is stronger.)
Hence the conclusion, consciousness is something more than the physical. This is simply property dualism.
Past that, Chalmers advances tentative starts on theories. He suggests consciousness arises from information--this leads him to consider panpsychism, the view that anything that contains information (his example is a thermostat) is conscious, even if the "brilliance" of that consciousness may vary with the complexity of the information.
Most of the arguments here proceed by means of imagining a spectrum between two poles: say, between me and the thermostat. Simplify me piece by piece until I reach the simplicity of a thermostat--isn't believing that consciousness magically "winks out" somewhere along this gradient arbitrary? Is it not more reasonable to think consciousness simply fades but does not vanish? He use the same strategy against Searle's Chinese Room thought experiment; it is the first criticism of the argument I've found persuasive, beating out any of Dennett's attempts at refutation.
The biggest objection is that, if the phenomenal fails to supervene on the physical, it's implied that the physical is causally closed. That's fine, but by relegating consciousness to a non-physical status, consciousness itself must be causally irrelevant. (This is garden variety epiphenomenalism.)
Now Chalmers has a few replies. He points out that we really don't understand causation, so consciousness may well be causally relevant in some way we've yet to figure out. That's possible, but I think if that ever turned out to be the case, it would eat the heart out of his opening supervenience argument. Second, Chalmers says, even if it does not, epiphenomenalism isn't that bad anyway. So our consciousness has no causal role--or, at best, it's redundant--so be it. The flaw is not fatal.
And I agree the flaw isn't fatal. But it's fairly obvious to me that my consciousness is causally relevant--I do something because the mental side of me wants to do that thing. Now that intuition might be wrong, but it takes an extraordinary argument to defeat it. What's more, I hold that intuition nearly as tightly as I hold the fact that I'm conscious. Some deny consciousness exists--Chalmers's only response to this is that it seems obvious consciousness does exist (Hofstadter, on the other hand, thinks consciousness is a mirage). If he'll accept that intuition, it's not clear why he can reject the causal relevance of consciousness, as that intuition runs pretty deep too.
That objection aside, we take dualism as fundamental. There is matter, there is space, there is time, and now there's consciousness (of course, consciousness could be comprised of more fundamental units, but we've established those units will not be matter, space, or time).
But once we take that as fundamental, I don't see the need to take anything else as such. No need for ethical facts, or aesthetic facts, or any non-material fact besides.
I suppose I always thought in for a penny in for a pound, so far as consciousness goes. "We have to accept some kind of mystical 'mental stuff,' and if we have to do that, why not other mystical things? Like 'moral stuff' and the such?" But as Chalmers sketches it, consciousness is really no more mystical than time or space. It's just a brute fact that any information-bearing system will be conscious, just like F = MA. But there's no compelling reason to believe in any further non-physical fact.
Even (previously?) identifying as a moral realist, I don't mind this. I enjoy having my mind changed. (Moreover, I don't think it's particularly important whether or not one's sense of right or wrong is objective or subjective--whatever the source, it still is and one acts accordingly.)
Still, I'm not sure this works. Even if consciousness isn't mysterious, knowledge is. If one has to take knowledge as fundamental--i.e., you either know something or you don't know something, and there is no breaking knowledge down into smaller parts like "falsifiable" or "justifiable" or "true"--then knowledge remains a mystical thing. And you do have to take knowledge to be such. If I have to believe in that spooky thing, then maybe I'm still in for a penny.
It's a great book. Structured, powerful, no rhetorical flourishes, dry, addressing nearly any counterargument you can think up. I offer only my perpetual complaint: why endnotes instead of footnotes?