In an essay entitled, “Philosophy and the Scientific Image of Man,” the American philosopher Wilfrid Sellars argued that our allegiances are divided between the manifest and the scientific image of the world.1 On the manifest image, the world comprises people, pains, and puzzles; on the scientific image, nothing more than various quantum fields. It is the scientific image that takes ontological pride of place. But if colorless quantum fields are really real, what are we to say about colored roses, which are really red?
There is an obvious gap between the manifest and scientific image of the world. How could roses be red? One venerable philosophical answer is that roses are red in virtue of how they are experienced. But experience is itself no less mysterious than color. We are subjects of experience; quantum fields are not. Questions of this sort have persuaded David Chalmers tentatively to embrace what he calls naturalistic dualism. For all that, dualism is still dualism. Many philosophers have resisted following him into the badlands. The stay-behinds hope to make sense of consciousness in a way that does justice to its richness while somehow embedding it in the scientific side of things.
In his new book, Daniel Dennett argues that all parties to this discussion, while rarely in doubt, are nonetheless in error. There is no point in bridging the gap between the manifest and scientific image of the world because no such gap exists. The manifest image, Dennett argues, is a user illusion.
Consciousness, too? Consciousness, too.
This is a daring move on the great chessboard. Consider an internet favorite: the American flag illusion. Under just the right circumstances of attention, distraction, and relaxation, a red, white, and blue American flag seems to appear suddenly on the computer screen; and within the flag, an alternating series of red stripes. Yet, Dennett argues, there are no red stripes on the retina, or on the screen, or in the brain. It just seems that there are. In many circumstances, that is good enough to warrant, from the way things seem, a judgment about the way things are. Judgments about red stripes are often caused by red stripes, and if not caused by them directly, then justified in virtue of them. Often, but not always. In the case of the American flag illusion, there is nothing there.
Why, then, do observers unfailingly report seeming to see those red stripes?
Why, indeed? One long-standing answer is that there must be something red if something seems to be red. There are no red stripes on the screen—granted. But something is in front of my eyes, and it seems to be red. The door has now swung open to the classical theory of sense-data or qualia. Dennett is having none of it. He is concerned to shut that door. The red stripes that one seems to see do not exist. One can judge that something seems to be a red stripe without ever toppling into the illicit inference that something is both striped and red. It is, of course, tempting to topple. Better not. Bad theories are their own punishment. A special inner theater is required in which to display imaginary red stripes. An inner observer is obligatory, as well.
It would be better by far, Dennett argues, to resist temptation. With temptation resisted, there is no need to hold on to the Cartesian theater, due, in any case, to end its run after more than four hundred years. If all this is to go, what is to remain? In the case of those by-now mythical red stripes, what remains is a representation of red stripes, but one that is inaccurate. The representation exists; the stripes do not. That is why the representation is inaccurate. Although the representation represents the stripes as having a certain shape and color, the representation itself is neither striped nor red. There is nothing in this to provoke alarm. The American eagle represents the United States, but while the United States is large, the American eagle is not.
Dennett is right to emphasize the role that representations play in consciousness; it is a useful idea. He is right to deny that there are subjective stripes involved in the American flag illusion; and he is right to scruple at the Cartesian theater. Still, Dennett’s larger project of demoting the manifest image of the world succeeds only to the extent that consciousness is itself an illusion. It is a striking claim. How could it be true? Illusions occur when things appear other than they really are. The woman on stage appears headless as part of a magic trick; the straight stick in water appears bent. When something appears some way, there is something it is like. But if consciousness is an illusion, then nothing can appear any way at all. To whom, or what, would it be appearing? All the authority of experience is against it. There is, it seems to me, nothing in the world more certain than that I experience pains and itches, that I experience the world as full of colored shapes and noises and smells. Any argument or view to the contrary cannot possibly be convincing.
Dennett dismisses this response by deriding the papal authority of the first person. Up to a point, he has a point. It is certainly true that first-person accounts are revisable. The line between a gentle slap and a caress is not all that clear; the standards under which I might describe myself as zestful are indistinct. But even when revisions are possible, they are conscious revisions. Whatever the papal mistakes, there is no getting around the pope.
Why not stop well before giving up consciousness and say simply that seeming to see a red stripe is itself a perfectly legitimate state of mind? Such states are representational states of the world. They become conscious experiences via the distinctive role they play within a central, global workspace. The manifest experiences need not be denied or demoted—just understood in other terms.
Dennett has written a clever and a provocative book. He has a special knack for offering many scientists what they most want to hear: that there is nothing in philosophy that might cause them to lose sleep. They are right not to lose sleep, but not for the reasons Dennett provides.
