Dennett, Consciousness Explained: Three Theses

Curtis Brown

Dennett has various targets in his book; they all seem to get lumped together, but in fact some seem distinct from others. Here are three that it might be useful to distinguish.

1. Special Place in the Brain. 

Many of Dennett's arguments are focussed on "Cartesian materialism" or the "Cartesian theater," the idea that there's a place in the brain where brain states become conscious. Well, in principle there could be. Why does Dennett think there isn't?

A. Partly because of a kind of a priori argument that if consciousness makes a difference to behavior, it would be very inefficient to have to funnel all information through a particular place in the brain before it had an effect on behavior. Given that it takes time for signals to move from one part of the brain to another, we'd react to things more slowly if all signals had to make it to a particular spot before they had their behavioral effects. Since that would be an inefficient way for the brain to work, it's unlikely that this is the kind of brain evolutionary pressures have produced.

Interestingly, as far as I can see, this argument loses its force if consciousness is epiphenomenal (in the good scientific sense, not the bad philosophical one). If consciousness is just a by-product of the brain processes that produce behavior, then the brain could go ahead and produce behavior right away, while consciousness of the state that produced the behavior could wait until the signals got to the right place.

B. In addition to the a priori argument, there are more empirical arguments to the effect that no particular part of the brain looks like a good candidate for the place where consciousness takes place.

2. Time of Occurrence of Conscious States. 

The special-place-in-the-brain view is closely connected with the view that there is a determinate time at which brain events become conscious. Why does Dennett think that?

A. Since it takes time for information to move from one part of the brain to another, we can't give a determinate figure for how long it takes inputs to become conscious unless we can identify a place in the brain at which they become conscious. So arguments against a precise location for conscious experience become also arguments against a precise time for conscious experience.

(There seems to be an assumption here that as far as I can see isn't necessarily true, though it certainly could be, namely that consciousness couldn't occur, so to speak, in different parts of the brain at different times. At one moment I have a conscious state in one place, at the next moment I have a conscious state in a different place. Two problems about this possibility: 1. Then what makes some states conscious and others not? 2. If states in different places can be conscious at different times, why not at the same time? But then we'd lose the serial or "von Neumanesque" aspect of consciousness.)

B. There is also Dennett's verificationist argument that since in certain cases involving very short time intervals, we can't have evidence that will decide whether states at certain times are conscious or not, therefore there can't be a fact of the matter about whether those states at those times are conscious or not. I'm worried about the verificationism here; it seems perfectly coherent to me that I could make judgments about my own (prior) conscious experiences which are determinately true or false even though I can't (now) determine whether they are true or false.

3. No Phenomenology, Only Heterophenomenology.

One of the most puzzing things about the book is that many of the detailed arguments apply only to very small distances and very short time intervals. Suppose we agreed with Dennett that there's no Cartesian Theater inside the brain, and that over very short intervals there's no fact of the matter about precisely when states become conscious. So far that still leaves open the possibility that there is a bigger place where consciousness takes place, namely the brain as a whole, and that there are determinate facts about when conscious experiences occur provided that we don't try to achieve levels of precision down to tens of nanoseconds. But for Dennett the Cartesian theater attacks and the attacks on the Orwellian/Stalinesque distinction are just the thin edge of the wedge; the ultimate goal is to convince us that in a sense there are no facts at all about conscious experience, that there is no phenomenology: there is only heterophenomenology and brain stuff, no genuine phenomenology in between. Why does he think that?

A. Argument from scientific method. All the evidence we have about phenomenology is evidence about our heterophenomenological interpretation of verbal reports. Therefore we can't ever discover the facts about real phenomenology, only the facts about heterophenomenology. Therefore there are no facts about real phenomenology. Both "therefores" look problematic to me. Why can't we learn about real phenomenology by gathering heterophenomenological evidence and subjecting it to critical scrutiny? And even if we can't, the second "therefore" seems to involve verificationism again.

B. Argument from avoidance of skepticism. Rosenthal has formalized the common-sense view of consciousness. But on Rosenthal's view it looks like we can make all sorts of mistakes about our conscious experience. We can't. So Rosenthal's view is mistaken. So common sense is mistaken. So there's no such thing as consciousness. (Or something like that. A million questionable steps here!)



Last update: September 4, 2001. 
Curtis Brown  |  Philosophy of Mind   |  Philosophy Department  |   Trinity University
cbrown@trinity.edu