Win,
Are you assuming that Mary's full set of physical facts includes the neural connections established when a normal person sees color?
--------------------------------------------------------------------------------
No. Look, I think there's been a serious misunderstanding at work here. When Mary sees red for the first time, if reductive materialism is true, she'll say, yeah, yeah, nothing surprising here; I already knew exactly what this would be like.
No, she wouldn't. Because as I already pointed out, she will not have the memory of seeing red.
Let's look at this another way. Let's imagine this exact same thought experiment under the framework of property dualism.
Mary has never seen red. She is presented with all the physical facts about seeing red. We will assume that she is actually capable of understanding and processing all of this information.
Now may sees red for the first time. Does she gain anything new?
The dualist would clearly say "yes, she gains the knowledge of what it is like to see red". But is that all she gains?
No. Even under property dualism, she will gain the memory of having seen red. And even under property dualism, that memory is a physical structure in her brain. She is still physically different after seeing red than before. Knowing all the physical facts about red doesn't give her the memory of seeing red.
If this is a contradiction under physicalism, then why is it not a contradiction under property dualism?
Let's go a step further. Let's put a p-zombie in the room instead.
You have claimed that a p-zombie is physically identical to a person, except that it does not have phenomenal consciousness. You have also claimed that such critters are logically possible.
So, we do the same experiment, when the p-zombie sees red for the first time, will it gain anything?
Of course it will. It will gain the memory of seeing red.
The apparent dualism here is the fact that there are two types of knowledge: empirical and abstract. Even in property dualism, a person (or zombie) can only get abstract knowledge from a book. They have to actually see red to get the empirical knowledge. Even the p-zombie has empirical knowledge, so the empirical knowledge is not phenomenal.
There are only two conclusions that can be drawn. Either the existence of both empirical and abstract knowledge do not invalidate physicalism, or they invalidate property dualism too.
As I said before, none of this has anything to do with philosophy. It has to do with how the brain works. The thought experiment never even gets to the point of dealing with the phenomenal. It falls apart as soon as it assumes that Mary could possibly gain the memory of seeing red by reading about red in a book.
This is the reason I gave the alternative thought experiment, where a robot alters Mary's brain surgically. After all, the real question this thought experiment is trying to ask is: If Mary was physically identical to somebody who has had the experience of seeing red, would she gain anything new by actually seeing red.
Naturally, I claim she would not. Do you think she would? If so, what? Would her brain be aware of it? Could her brain be aware of it?
If nothing else, I think this highlights the central problem I have with epiphenomenalism. If I altered Mary's brain is such a way as to be identical to the way it would be if she had experienced seeing red, her brain could not possibly tell the difference. But you must assert that she would be able to tell the difference. Could she vocalize that fact? If so, then you must assert that the p-zombie would also claim that it could tell the difference. But that is a contradiction, since the p-zombie is physically identical to the way it would have been if it had actually seen red, in which case it would not claim it could tell the difference.
Dr. Stupid
