(Below a follow-up on a previous post about Dennett’s ‘Consciousness Explained’. This note draws on discussions with the study group on this text at the TU Delft)
In chapter 12 of Consciousness Explained, Dennett disposes of ‘Qualia’, – they are ‘disqualified’. A possible complaint is that he is not particularly clear as to how the argument works. I don’t plan on defining qualia at this point, but below a second line of reasoning to make Dennett more palatable.
1. Residual Properties
One of Dennett’s main goals in chapter 12 is to show how experiences can be reduced to responses of an organism to its surrounding (that is my reconstruction at least), and thereby ultimately close an apparent gap between on the one hand the world of external phenomena and brain-processes, and on the other the world of subjective experiences.
Qualia are problematic because, according to Dennett, they are almost invented to resist such a re-interpretation (perhaps I’m allowed to say ‘reduction’?). Supposedly knowing whats going on in the brain is something else than knowing qualia, and supposedly knowing the way an organism responds to its surrounding is completely irrelevant to the issue of qualia. Thus qualia would be a field of study forever outside of the scope of ‘regular’ science.
So as Dennett puts this in Intuition Pumps, qualia are the supposed residual properties left after all properties have been handed over to scientific investigation. They are the one thing about subjective experience which science cannot study. The unique private and subjective ‘feel’ of how things appear. Therefore they cannot be something which figures in a materialist ontology.
2. Behaviourism and intuitions
According to Dennett, the intuition that something like a residual property might exist, something like ‘the peculiar way this experience just seems to me’, is pumped by several thought experiments. These thought experiments supposedly proof that qualia are not just part of the responses a subject has. For instance:
“That [combination of behaviour and dispositions] cannot be all there is to it […] for while that complex of mere dispositions might be the basis or somehow, for my particular quale of pink, they could all be changed without changing my intrinsic quale. or my intrinsic quale could change, without changing that manifold of mere dispositions. For instance, my qualia could be inverted without inverting all my dispositions. I could have all the reactivities and associations that I now have for green to the accompaniment of the quale I now have for red, and vice versa.” (389)
Or perhaps more dramatically, all of us might have different qualia connected to the same responses to a colorful object:
“There are the ways things look to and sound to and smell to me, and so forth. That much is obvious. I wonder, if the ways things appear to me are the same as the ways things appear to other people.” (389)
So it seems that the way ‘red’ appears to me, subjectively, privately, might just not be the way red appears to you. Perhaps your red is my green. Even if we respond similarly by calling it ‘red’ and noticing it to be the same color as the firetruck, it might still be the case that ‘red’ appears different, because this experience is independent of these particular responses.
At this point Dennett complains that this intuition is flawed. It presupposes that we actually separate our concept of ‘experience’ from dispositions to behave in certain ways. And in fact, we don’t do this. We are much more behaviorist than we think.
(This is both the Rylean and Wittgensteinean bit of Dennett)
3. Turning knobs
To notice this, Dennett proposes we tweak the experiment a bit. We can reduce behavioral dispositions to brain processes. Qualia are not reducible. So supposedly they have nothing to do with behavior. So behaviour should not matter in any way whatsoever (392). So can we truly image a case of qualia-flipping without flips in the behavior? Would we even ‘intuit’ that a flipping of qualia has occurred in those cases?
For instance, we can imagine that someone’s qualia have been ‘flipped’, for some reason. He now perceives red as blue, and blue as red. In the past, the subject got hit by a blue car, and clearly remembers this being blue. We might expect him to say things like:
‘I have a clear memory of a blue car. Yet if you show me this photograph of the car, the color of the car on the picture looks red to me. Or that is, look at this teapot. Only yesterday it looked red to me, but now it looks blue. It is the same color I remember the car to be, namely blue. And I remembered the teapot to have been red. Rather like what the color of this car looks to me now. It is as if these colors have exchanged the way they look. This is mighty confusing.’
What drives our sense that a flipping has occurred in this case? Well, that we are so mightily surprised by the way the world looks to us now. But this sense of surprise clearly presupposes certain associations and expectations we had of the world are still intact. It is the mismatch between our memory and what our eyes tell us that drives the sense of surprise. And can we not just analyse this mismatch in terms of behavioral dispositions?
Imagine, for instance, that not only our perception, but also all our memories and associations are ‘flipped’ as well. In your perception, the blue car appears to be ‘red’, but all your memories have adapted just as well. The car that hit you is the same color ‘in’ your memory and on the photo. In fact, the ‘redness’ of the car is quite vivid, and completely unlike the ‘blueness’ of a firetruck. You also remember learning in school how the spell the word associated with the color ‘red’: ‘B L U E’, that’s how you spell ‘red’. And everyone of course agrees, since that’s what we call the color of the sky.
Now, being faced with a picture of the car would not prompt a surprise: your memory matches the color on the picture. There would be no way of telling from the outside whether a qualia-flipping has occurred, because the subject behaves as if nothing has changed. But if your memory has ‘flipped’, and every association you could possible have with a color has ‘flipped’ as well, would you ‘from the inside’ even be able to tell that a flipping has occurred?
The question above is hard. And that is exactly Dennett’s point. Thought-experiments concerning qualia-inversion are designed to trigger our intuition that qualia are ‘irreducible’. But they do so in a sneaky way, which evades the question whether modifying the thought-experiment in a way which should be irrelevant, might not yield completely different intuitions. Work is being done by parts of the thought-experiments which are supposed to be irrelevant.
Dennett has a hard time believing that if we truly exclude all behavioral dispositions of an organism from our theory, this will still leave anything for the qualia to be. They can just be mobbed up in these behavioral dispositions to remember, to speak, to act, based on associations, words, cultural practices, etc.
“[Do Qualia Exist? Perhaps ‘no’] Or, yes, because if you really understood everything about the functioning of the nervous system, you’d understand everything about the properties people are actually talking about when they claim to be talking about their qualia.” (46)
– —- HvS
(N.B., it is not evident that the possibility to ‘identify’ everything in consciousness with a brain state makes it impossible to separate consciousness from the brain on a non-ontological level. Dennetts discusses this as well, stating that these sorts of strategies to deal with qualia look for an epi-phenomenal theory. It should be noted that this might not be the only way to agree with materialism and yet exclude consciousness from it)