Sweet Dreams: Philosophical Obstacles to a Science of Consciousness

  • Daniel C. Dennett
MIT Press: 2005. 216 pp, $28, £18.95 0262042258 | ISBN: 0-262-04225-8

Has the devil lost his horns? Dan Dennett has been demonized because of his tough materialist stance on consciousness and his claim that we are conscious machines with nobody at home inside. But now he updates his theory and, I think, presents a softer version of it.

Sweet Dreams is a collection of essays and lectures written between 1999 and 2005 in which Dennett tries to freeze time and present a ‘best’ version of his evolving ideas. There are problems with this, especially as some chunks are somewhat disconcertingly repeated from one chapter to the next, but the overall picture gives a good idea of where Dennett's thinking has been going.

He describes his task as being to explain away qualia-based intuitive objections to materialism, and this he does using some of his favourite examples. One is the ever-popular philosopher's zombie, an imaginary creature who looks, acts and speaks like a normal person but has no subjective experience or qualia. It's easy to imagine a zombie — or at least to think you are imagining one — says Dennett, and he calls this “falling for the zombic hunch”, which traps people into believing that consciousness is separate from brain function. Dennett has tried to murder the zombie before, explaining how people fail to follow the rules when they think they are imagining one, but now with the concept of the zombic hunch he explores the damage done by this false intuition. Don't worry, he says, if you are patient and open minded it will pass, or mutate into a less virulent form. Just as we still feel as though the Earth stands still, in the future people may still feel the zombic hunch, but they won't believe it. They will know that mechanistic theories of consciousness do the whole job, so we don't need the concept of qualia.

Demystifying consciousness is Dennett's forte, and is probably the main reason for his status as the devil — after all, people like mysteries. Some even think that if consciousness is explained they will be diminished as people, being turned into mere ‘things’. They don't like the idea that there is no one at home inside their head, no audience watching the magic show of conscious experiences from the safety of the cartesian theatre. Dennett contrasts those theorists to whom it is obvious that a theory that leaves out the subject cannot explain consciousness, with those to whom it is obvious that the subject has to vanish. The first type must be wrong, he says: “A good theory of consciousness should make a conscious mind look like an abandoned factory.”

All these ideas will be familiar to lovers and haters of Dennett, as will his attacks on Mary the colour-scientist (who knows all there is to know about colour perception but lives in a black-and-white world) and the idea of first-person science. But he does add some new twists and, as ever, delights with his quirky turns of phrase. I laughed out loud at his example of a folk theorem as ludicrous as some of the ones that people claim are obvious: “If you burp, sneeze, and fart all at the same time, you die.” But let's have some evidence, he says.

What may be less familiar is Dennett's updating of his theory of consciousness, from the original ‘multiple drafts’ theory, through the delightfully named ‘fantasy echo’, to his latest ideas of ‘fame in the brain’ or ‘cerebral celebrity’.

In a paper originally published as an overview for a special 2001 issue of Cognition, he joins in the “gathering consensus” that favours ‘global workspace’ theory. Originally proposed by Bernard Baars in A Cognitive Theory of Consciousness (Cambridge University Press, 1988), this theory is explicitly based on a theatre metaphor in which the contents of consciousness are illuminated by a spotlight of attention, shining on the stage of working memory. The idea is that many specialist brain systems contribute information to the global workspace; its contents are then broadcast to the rest of the system and this global availability is experienced as a conscious state.

Doesn't this go against everything Dennett has been fighting for? He tries to explain why not. Consciousness, like fame, is not an intrinsic property of brain processes but is more like political influence or clout; conscious events are the ones that have widespread effects in the brain. So we must not think that becoming famous in the brain magically ignites the glow of conscious qualia or lets pictures into the cartesian theatre to be watched by the conscious subject; the effects are enough. Fame in the brain does not lead to consciousness — it is consciousness.

Many global-workspace theorists don't see it this way, and Dennett says they are “surrendering just when victory is at hand”. The critical point hinges on what he calls the ‘hard question’: “And then what happens?”. (This is not to be confused with Chalmers' ‘hard problem’, which Dennett thinks is illusory.) If you think that something has to happen next, such as entering consciousness or becoming conscious, then you are still wallowing in mystery.

But if the devil had stuck with his original multiple-drafts theory, he would surely have gone further. One of its most startling claims, made by Dennett in Consciousness Explained (Little, Brown, 1991), is that “there are no fixed facts about consciousness independent of particular probes”. I took this to mean something like this: at any time there are multiple versions of representations in the brain, none of which is conscious or unconscious. Only when the system is probed — for example by asking a question or requiring an action — does one of the drafts have consequences that cause us to say, after the fact, that it was conscious. In other words, there are no conscious and unconscious streams; nothing ever ‘enters consciousness’; there is no “crucial finish line or boundary somewhere in the brain, marking a place where the order of arrival equals the order of ‘presentation’ in experience”.

Yet the global-workspace theorists believe in conscious and unconscious streams, in contents becoming conscious, and in looking for the neural correlates of the ‘movie in the brain’. I think they still need the old devil.