Knowledge Argument redux

[Paul C. Anagnostopoulos]

Folks,

Since we had that grueling discussion about the Knowledge Argument, I've been chatting with a fellow named Martine Nida-Ruemeline at Universite De Fribourg. We've gone back and forth a few times and I've tried to make it clear to him that there is subjective nonconscious information gained when Mary finally sees color, so that she did not have all the information before she left the room. (Note how I'm avoiding the use of the word fact). He finally said:
It is a premise of the argument that there are no physical facts that you
can only learn by actually seeing red. It is not denied that you can learn
physical facts by seeing red, but it is assumed that for every physical fact
that you can learn by seeing red there also is another way to konw the same
fact in another way without seeing it . ... You obviously deny this premise. So: What
are the facts you have in mind that Mary can only learn by actually seing
red? A simple concrete example would be helpful.

Well, I'm not a neurphysiologist, so I can only imagine some of the neural training that goes on when a baby learns to see color. (I find it difficult to believe that others can't imagine it.) A little searching found Karen Dobkins at UCSD. She is going to provide me with a paper on how infants use color for motion processing, but how this ability fades as they grow older.

She also mentioned something else interesting. Someone (sorry, can't remember who) suggested that a person deprived of color when he was a baby would never be able to see color correctly. Experiments suggest this is true of tree shrews. In these experiments, the shrews were raised in red light.

~~ Paul

[Paul C. Anagnostopoulos]

Here is the Dobkins paper:

http://www.windfall.com/Dobkins-paper.pdf

~~ Paul

[AmateurScientist]

Originally posted by Paul C. Anagnostopoulos
(I find it difficult to believe that others can't imagine it.)
~~ Paul

For what it's worth, Paul, I find it difficult to believe others can't imagine it too.

Again, in order to make the argument of those denying your premise, one has to assume the conclusion. That is, there is no physical fact about red that cannot be gained without actually seeing red. It's question begging again for the 6,583rd time.

I've lost patience with the dualists and their specious reasoning. I will no longer indulge their condescension or foolish arrogance. Fools, all of them.

AS

[Argo Nimbus]

Originally posted by Paul C. Anagnostopoulos
Since we had that grueling discussion about the Knowledge Argument, I've been chatting with a fellow named Martine Nida-Ruemeline at Universite De Fribourg. We've gone back and forth a few times and I've tried to make it clear to him that there is subjective nonconscious information gained when Mary finally sees color, so that she did not have all the information before she left the room.

Paul:

Suppose we try this with twin sisters, Mary and Beth, who are identical except that Mary is colorblind and Beth has normal vision. Both have been raised in the real world and each knows that objects come in different colors. Now place both in the same room with two of the usual charts used to check for color blindness. That is, people with normal vision will see the word "Bingo" on the first chart and "Hotdog" on the second, but colorblind people will not be able to distinguish the words. Beth is shown the first chart and passes the test by shouting, "Bingo". Mary is shown the second chart and says (truthfully) that she doesn't see a word. Doesn't Beth know something that Mary doesn't? Curing Mary of colorblindness (thus giving her the same ability as Beth to distinguish colors) wouldn't change the argument. Mary, like Beth, would then know something that colorblind Mary didn't.

--- Argo

[Argo Nimbus]

Originally posted by Argo Nimbus
Curing Mary of colorblindness (thus giving her the same ability as Beth to distinguish colors) wouldn't change the argument.

On second thought, curing our hypothetical Mary of color blindness would have some advantages. First we show colorblind Mary the second chart and she fails to see the word "Hotdog". Next we cure Mary's colorblindness by repairing whatever physical damage there may be. Then we again show her the second chart. This time she sees not only colors, but the word "Hotdog". She hasn't yet learned to associate the word "red" with the color red, so she couldn't point to a red object if we requested it. However, she remembers that before her cure the chart looked one way and after the cure it looks different and she knows what the qualitative difference is whether she could find words to describe it or not. She was raised in the real world with her sister Beth, so she knows the word "hotdog" when she sees it.


Martine Nida-Ruemeline was quoted as saying:
What are the facts you have in mind that Mary can only learn by actually seing red? A simple concrete example would be helpful.


In the example I've just given, Mary can only learn that the second chart contains the word "Hotdog" by actually seeing colors.

--- Argo

[Paul C. Anagnostopoulos]

Argo, I'm not sure what point you're trying to make. Beth "knows" things that Mary doesn't in the sense of (a) having brain machinery that Mary doesn't; (b) therefore having neural training that Mary doesn't. I'm not sure where to go from here.

~~ Paul

[Argo Nimbus]

Originally posted by Paul C. Anagnostopoulos
I'm not sure what point you're trying to make. Beth "knows" things that Mary doesn't in the sense of (a) having brain machinery that Mary doesn't; (b) therefore having neural training that Mary doesn't. I'm not sure where to go from here.


Paul:

I was responding to this: "I've tried to make it clear to him that there is subjective nonconscious information gained [by which I assume you mean stored memories, i.e. knowledge] when Mary finally sees color, so that she did not have all the information before she left the room." I thought that the task of making this clear would be easier if we used an example that is closer to real life than p-zombies.


Martine Nida-Ruemeline was quoted as saying:
It is a premise of the argument that there are no physical facts that you can only learn by actually seeing red.


Is this Nida-Ruemeline's premise or yours? Colorblind Mary provides a counter-example, so I would say that this premise is false. The test for colorblindness provides a concrete example of information that can only be learned by seeing color: the memory of seeing the word "Hotdog". Of course, someone could explain the test to Mary and tell her that people with normal vision can see the word "Hotdog". What would that prove? Colorblind Mary still can't see the word, so there is still information that she can't learn except by seeing colors. To claim otherwise, one would have to say that the memory of being told the word is indistinguishable from the memory of seeing the word (which I don't find plausible).


Martine Nida-Ruemeline was quoted as saying:
... it is assumed that for every physical fact that you can learn by seeing red there also is another way to konw the same fact in another way without seeing it. ... You obviously deny this premise


Okay, you deny this premise. So would I. What does Nida-Ruemeline say in defense of it?

Of course, maybe I'm completely missing the point of the argument. If memory of an experience doesn't count as information, but memory of the word "hotdog" does, then Mary would know the same information no matter what experience was necessary for her to learn it, i.e no matter whether she learned it from you or saw it for herself. Is this the real point? If it is, then the real premise of the argument is that memories of first person experiences do not count as information, even if I can distinguish between such memories.

--- Argo

[Paul C. Anagnostopoulos]

Argo said:I was responding to this: "I've tried to make it clear to him that there is subjective nonconscious information gained [by which I assume you mean stored memories, i.e. knowledge] when Mary finally sees color,
I'm thinking more of training in her visual pathways: neural connections and weightings that are formed while experiencing color. These are memories in the sense that everything in your brain is a memory, but not memories in the usual sense of the word.

Is this Nida-Ruemeline's premise or yours? Colorblind Mary provides a counter-example, so I would say that this premise is false. The test for colorblindness provides a concrete example of information that can only be learned by seeing color: the memory of seeing the word "Hotdog". Of course, someone could explain the test to Mary and tell her that people with normal vision can see the word "Hotdog". What would that prove? Colorblind Mary still can't see the word, so there is still information that she can't learn except by seeing colors. To claim otherwise, one would have to say that the memory of being told the word is indistinguishable from the memory of seeing the word (which I don't find plausible).
Aha! It is that distinction that many people don't seem to make. (It is N-R's statement of the premise of the Knowledge Argument.)

Okay, you deny this premise. So would I. What does Nida-Ruemeline say in defense of it?
He asks for a specific example of neural training that can't be gained by reading a book. Seems obvious to me that there are many such examples, but I found that paper by Dobkins as an example.

Of course, maybe I'm completely missing the point of the argument. If memory of an experience doesn't count as information, but memory of the word "hotdog" does, ...
In our long discussion of the KA a couple of weeks ago, we spent a lot of time trying to define fact, information, knowledge. Most people disagreed with me that neural training should be called subjective physical facts. Even Stimpy disagreed, though we agreed that the KA is flawed. Max Deutsch agrees with me: http://neologic.net/rd/chalmers/mdeutsch.html. Well, in fairness, I should say that I agree with him. If we can't call these things facts, I'm not sure what word to use.

~~ Paul

[Paul C. Anagnostopoulos]

Continuing discussion with Nida-Ruemelin has brought a couple of things to light. First of all, he said "Of course Mary doesn't have all the neural training ...". That made me wonder why he asked for a concrete example.

However, it now becomes clear that he does not consider neural training to be physical facts. He asks "... I would have to know which physical fact you identify with the content of what Mary learns after release." He thinks that when I say Mary doesn't have all the physical facts, I'm talking about book-learned facts that correspond to neural training. Of course I am talking about the neural training itself. The discussion continues.

So here is another example where excluding subjective physical information from the term facts causes confusion and perpetuates a muddied conversation. But I guess I've lost this argument.

So now I ask: Do philosophers realize that neural training is missing from the premise to the KA, in which case the KA addresses a straw man definition of physicalism? Or does the official definition of physicalism really exclude such things as neural training, in which case it is absurd?

~~ Paul

[ChuckieR]

Originally posted by Paul C. Anagnostopoulos
She also mentioned something else interesting. Someone (sorry, can't remember who) suggested that a person deprived of color when he was a baby would never be able to see color correctly. Experiments suggest this is true of tree shrews. In these experiments, the shrews were raised in red light.
~~ Paul Paul,

I believe it was me who brought up the developmental issue in the Materialism thread. I find it a little annoying that, for the purposes of this abstract thought experiment, the KA asks us to "suspend our disbelief" that an adult raised in a color-deprived environment will be able to perceive colors normally when she is finally exposed to them. But it seems to me that this is the heart of the problem. The shrew study you found seems to back this up.

I believe the point that the KA is trying to make is that there is something about "redness" that goes beyond the physical, book-learned "facts" (not a well defined term, as we found out). I think everyone agrees on this, even though the thought experiment itself is somewhat flawed. Materialists and non-materialists simply draw a different conclusion from this. For non-materialists, since the "redness of red" cannot be adequately described in words, materialism is flawed. Materialists would likely counter that the "redness of red" is simply a result of the activity of the brain, and that differentiating between colors is all that matters. "Redness" itself is irrelevant and meaningless by itself. "Can you tell red from green from blue" is what matters. It is the "differential", the "relative", the "comparison", the "stimulated" vs. "unstimulated" that matters.

N-R's quote seems to show how this quickly degrades into a "word" argument, where the term "physical facts" has a specific (though, to me, poorly defined) meaning, and when you start talking about training neurons, he will counter something like, "well, that's not physical facts". Like AmateurScientist, I too have a difficult time understanding how these folks don't seem to "get" that book-learned knowledge and neural-training "experience" are two very different things, and that reading a book will not train your visual processing system to process colors. They want to have this philosophical discussion that seems to have absolutely no connection to how the brain actually works. I say, let them have the discussion. The rest of us see very clearly that it has little or nothing to do with reality.

[ChuckieR]

Originally posted by Paul C. Anagnostopoulos
Continuing discussion with Nida-Ruemelin has brought a couple of things to light. First of all, he said "Of course Mary doesn't have all the neural training ...". That made me wonder why he asked for a concrete example.

However, it now becomes clear that he does not consider neural training to be physical facts.:) I should have read ahead! This is exactly what I thought would happen. It becomes word-play. These people love to argue, but they don't seem to like to "figure things out". I'm the opposite. I love to figure things out, but I can't stand it when people argue just for the sake of argument.

He asks "... I would have to know which physical fact you identify with the content of what Mary learns after release." He thinks that when I say Mary doesn't have all the physical facts, I'm talking about book-learned facts that correspond to neural training. Of course I am talking about the neural training itself. The discussion continues.

So here is another example where excluding subjective physical information from the term facts causes confusion and perpetuates a muddied conversation. But I guess I've lost this argument.I don't think you've lost the argument, you've just shown that the argument is disconnected from reality. But that's just what the KA is all about! i.e., there is something "unreal" (non-physical) about "the redness of red". Of course, since the argument itself does not seem to adequately connect itself to physicalism, it of course is in no position to refute it.

So now I ask: Do philosophers realize that neural training is missing from the premise to the KA, in which case the KA addresses a straw man definition of physicalism? Or does the official definition of physicalism really exclude such things as neural training, in which case it is absurd?

~~ Paul Obviously physicalism doesn't exclude neural training (I know you said this just to show its absurdity), so I suppose that leaves the first assertion.

In fainess, I don't want to pretend that I don't appreciate what the KA is trying to demonstrate - namely, that the "redness of red" cannot be put into words. I must admit to sometimes pondering the redness of red and wondering how red "looks" (feels?) to others. But I don't feel any need to jump to non-physical explanations of color processing in order to ponder the question.

[Paul C. Anagnostopoulos]

Chuck, I wasn't lamenting that I'd lost the overall argument, just that I'd lost the argument that physical facts should include the state of Mary's brain after training. Many people, including physicalists, want to restrict fact to mean conceptual fact.

So the story continues. N-R has made the clear distinction between knowing about her physical state and having the physical state. So now we must decide whether the premise "Mary knows all the physical facts" includes having the physical state. Stay tuned . . .

~~ Paul

[ChuckieR]

Originally posted by Paul C. Anagnostopoulos
Chuck, I wasn't lamenting that I'd lost the overall argument, just that I'd lost the argument that physical facts should include the state of Mary's brain after training. Many people, including physicalists, want to restrict fact to mean conceptual fact.Ah, sorry I misunderstood.

So the story continues. N-R has made the clear distinction between knowing about her physical state and having the physical state. So now we must decide whether the premise "Mary knows all the physical facts" includes having the physical state. Stay tuned . . .

~~ Paul I think it might be useful in your dialog with him to ask directly whether he believes that Mary will be able to distinguish between, say, red, green, and blue objects when she leaves the room (or equivalently, whether she would be able to pass a color-blindness test). This will avoid some of the word games, and get to the heart of what N-R thinks is going on.

I can imagine several answers to this question:

Obviously, she will be able to distinguish between different colored objects .
I assume she would be able to. Of course, if she couldn't, then the KA wouldn't make much sense [admitting the weakness of the KA].
That's not the point of the argument [oh, but it is].
I don't know whether she would be able to (or No, she won't be able to), but anyway it doesn't matter for this argument [again, this [i]is the argument].

[Paul C. Anagnostopoulos]

Not a word from the fellow in three days. Stay tuned . . .

~~ Paul