Dennett starts off with a discussion of how hallucinations are possible that introduces both his subject and the approach he is going to use. He then describes the metaphysical framework he is working within, basically a common-sense materialist worldview without recourse either to behaviourist blindness or dualist miracle substances. He also states what it is that he is actually trying to explain — what he feels he needs to achieve in order to have provided an explanation of consciousness.
Rather than ignoring people's testimony as to their own emotions, feelings, and "mental states" (the behaviourist approach) or granting them privileged access to these things, Dennett introduces what he calls heterophenomenology. In the heterophenomenological approach, what people say about their internal subjective experiences is allowable as evidence, but as evidence of how things feel to them, not as direct evidence of "things as they actually are". Peoples' feelings about their own consciousness have to be explained, but they don't have any kind of mysterious "special access".
The core of his theory of consciousness is the "multiple drafts" model. In the multiple drafts model consciousness is not a unitary process but rather a distributed one (just as a novel in preparation may exist in multiple drafts at any one time and is only afterwards "finalised"). Sequential timing of events breaks down at small (millisecond) time scales within the brain, and the events that make up consciousness cannot be ordered. In short there is no central place in the brain/mind where everything is presented and decisions are made (the fallacy of the "Cartesian Theatre"). The evidence for this view of consciousness is a whole series of results from experiments in cognitive psychology and neuroscience.
The rest of the book is devoted to discussing the consequences of this view of consciousness for several well known thought experiments and problems in the philosophy of mind (zombies, inverted spectra). It includes a rousing rejection of the whole concept of qualia, a chapter which I found particularly enjoyable. Unlike many philosophers Dennett keeps his feet firmly on the ground at all times, and doesn't get himself stuck in hermeneutic wrangling over obscure details. His thinking is solidly based on experimental results from neuroscience, cognitive psychology and evolutionary biology.
It seems clear to me that Dennett's theory of consciousness must be basically correct. Not only does it appear to explain a large number of otherwise mysterious phenomena, but the alternatives seem to be either completely confused or lacking in explanatory power. Whether the details of his theory prove correct or not, I believe he has made it clear that it is possible to try and explain consciousness without resorting to obfuscation — that there is no metaphysical mystery about consciousness, even if many of its features still remain to be fully elucidated.
Consciousness Explained, as well breaking new ground towards an explanation of consciousness, is also a very good introduction to the philosophy of mind. It is a book that will make you think about your everyday experience of the world and the nature of self and identity. Dennett has clearly gone to a lot of trouble to make his work approachable to a popular audience, and I feel it deserves a wider readership than it is likely to get. Go out and read this book now!
April 1993 [updated January 1995]