Consider our old friend Mary, who grew up in a black and white room, learned all of physics, and then saw a red tomato, allegedly learning a new fact about the world, what red looks like. Suppose Mary now went back to her black and white room and returned to contemplating the foundations of quantum mechanics, just as she did before. At that point, clearly Mary knows what red looks like. But unless she is visualizing red stuff, she is not having any red qualia at that point. So:
- One can know what red is like without having any red qualia.
Moreover, presumably whatever state her mind has—regardless of whether the mind is physical or not—in her black and white room after seeing the red tomato is a state that could have been induced in her (by a neurosurgeon or a demon) without her having had any red qualia. One might worry whether that induced state would count as knowledge, but if one adds that she gets testimonial evidence that her mental representations of qualia are correct despite based on false memories, it could be knowledge. Thus:
- One can know what red is like without having or having had any red qualia.
It is possible to agree to (2) while holding that the knowledge of qualia argument against physicalism is a good argument. For one might hold that the state of mind that allows for (2) is not a state that can simply come from learning all the physical facts. It is a state that might require some kind of neurosurgical or supernatural intervention. But it seems to me that when one accepts (2), it becomes significantly less plausible that one cannot learn what red is like just by learning all the physical facts.
There is another move the defender of the knowledge argument can make. They can deny (1) and (2), holding that when Mary is back to thinking about quantum mechanics, she doesn’t know what red is like, but that we are inclined to incorrectly say that she knows it because she has the skill of coming to know it at a moment’s notice by visualizing something red. This is a good move, but it has a pitfall: it makes knowledge of what red is like significantly disanalogous to ordinary knowledge, such as of multiplication tables, which one has even when it is merely dispositional, when one is not thinking about it. But if knowledge of what is red is like has this significant disanalogy to ordinary knowledge, that makes it less likely that it is factual knowledge—which the argument requires it to be.
9 comments:
I have a very impoverished olfactory imagination - try as I might, I can’t imagine what apples smell like. Of course, when I do actually smell apples, they are instantly recognizable.
Do I know what apples smell like? It’s hard for me to give a straight answer. On the one hand, I can identify apples by smell, and when I do, I seem to experience associated qualia. On the other hand, my access to what apples smell like feels quite different from my access to other knowledge. I can only “retrieve it” in certain situations. If I think to myself “What is the capital of the United States?” the answer pops out - not so for “What do apples smell like?”. Perhaps it is better to say I forget what apples smell like when I’m not smelling them. Other people have aphantasia - they can’t visualize things. Perhaps they don’t know what red looks like when they’re not experiencing it.
Suppose a neurologist implanted in me the ability to recognize by scent some kind of exotic fruit I have never smelled before. It seems strange to say that I have learned a new fact, or that my epistemic situation has improved somehow. I wouldn’t be able to imagine what it smelled like it even if I tried. I have gained a new ability though.
Very interesting.
Normally, I can't visualize specific faces, even of my closest family members. (On rare occasions, I get a brief mental flash of a distorted version of one of their faces. And if I make a mental effort, I can visualize parts of generic faces: a fat nose, or a pair of eyes above a nose bridge, though I don't think I can visualize the eyes together with the whole nose.) I have no trouble recognizing them, though.
The inability to imagine a face seems to me to be just that: an inability, akin to the inability to draw a face. (Interestingly, I can draw a face that is recognizably in the genus homo, maybe even in the species homo sapiens, though not recognizably any particular human.)
Do I know what my family members look like? One test of whether someone knows what someone looks like is whether they can pick them out of a lineup. I wouldn't have any trouble with that. But another test is whether they can describe them. And I'd do VERY poorly there.
Your argument concerning point 2 is very interesting, as I see it as a very plausible argument against accepting the knowledge argument's conclusion.
Firstly, it is It seems entirely plausible that a neurosurgeon could bring about the relevant mental state, for example by stimulating the optic nerve to fire in the same way it would on Mary seeing a red tomato, or by stimulating the nerves it connects to in the way they are stimulated by the optic nerve. It is also plausible that if the neurosurgeon brought about all, and only, the persistent physical changes that would result from Mary seeing a red tomato, the result would be the same.
Conversely, it is implausible that Mary would be able to do this from any amount of studying books. We do not know how to address specific neurons in our brain to bring about specific changes through mental effort, and there is no reason to believe that the brain is wired up in a way to enable this.
So, Mary could know at least as much as the neurosurgeon about what changes to make in her brain to bring about this change, yet be unable to bring them about. She could know all the relevant physics, yet still learn something from seeing colors for the first time, without this being a challenge to physicalism of the mind.
Maybe I'm missing something, but why would this present any challenge to the knowledge argument?
As I understand it, this is what you're saying: it is possible to "know what red is like without having or having had any red qualia", by inducing a state that is identical to one where one had previously experienced red qualia.
But I think most people, including proponents of the knowledge argument, would already accept this as self-evidently true, so I'm not sure how this would affect the knowledge argument.
(If I induce in person P a mental state S that includes 'knowledge that x', with P being justified in trusting the authenticity of her own mental states, then of course P will know x. )
More explicitly, none of the premises of the argument get challenged by these considerations:
let's say Mary studies a lot and learns all the physical facts about vision, but she does this in her black and white room. After she finishes all the studying, she will be in a state where she knows every relevant physical fact, and yet there is something she doesn't know. Let's also say that after this, a demon or a neurosurgeon get's her mind to be in a state where she has knowledge of what she didn't know before. But this later course of action that the demon/surgeon took doesn't change the earlier state of affairs, where she both knew every (relevant) physical fact yet didn't know everything- and therefore, doesn't affect the conclusion of the argument either: that there is a non-physical fact.
That's a nice response.
I guess I was thinking that the reason why we thought that Mary didn't know what red is like was some principle like "You can't know what red is like if you haven't experienced red." But if you just *directly* intuit that in the story as given Mary didn't know what red is like, then you can run the argument while rejecting the principle. And that does actually seem a better way to think about this stuff. Good.
That said, I think this state of "knowing what red is like" is a puzzling one, especially in light of aphantasiacs who say that they know what red is like but can't visualize it.
I don't have aphantasia, but I'm honestly not certain whether I can visualise what red is like with my eyes closed. I'm pretty sure I do _know_ what it's like, tho. Things get weird when one tries to mentally imagine or visualise stuff, and not just about colors. I agree that this is puzzling.
It's very puzzling indeed.
Suppose you and I are neither able to visualize colors, but we both know what red and blue are like. However, while you know that (ripe) tomatoes are red, I think they are blue. I take it then you not only know what red and blue are like, but you know what tomatoes look like. And I know what red and blue are like, but I don't know what tomatoes look like. Furthermore, let's suppose neither of us has the *words* "red" and "blue"--our thought of what tomatoes look like isn't a verbal thought like "Tomatoes are red/blue", but at least partly a non-verbal thought like "Tomatoes are *like this*."
Now suppose you make your knowledge occurrent and I make my mistake occurrent. Is there any difference in what it is like for you to think the true thought "Tomatoes are *like this*" and what it is like for me to think the false thought "Tomatoes are *like this*"? If we were visualizing colors, there would be a difference in what it is like for us to think the two thoughts. But we're not, because I assume we can't visualize colors. So if there is a difference, and it's not in words nor in visualized colors, what is it a difference in?
But if there is no difference, how can it be that (a) we're both thinking occurrently (and hence consciously) while (b) you're right and (c) I'm wrong?
This is very puzzling indeed!
Under the assumption that we can't visualise colors, I think we can still say that by thinking "tomatoes are *like this*", we are referencing what is, or what we perceive to be (in the induced experience-scenario), an earlier color experience of ours. I might make a reference to a tomato I've actually seen, and you might make the reference "tomatoes look like this blue tennis ball I have seen last month". If I never saw any color, I couldn't make such a reference.
Very cool solution, if I understand it rightly. So, on this hypothesis, our thoughts might be internally exactly alike, but have different "like this" content, because the "like this" content comes from a _de re_ reference in the thought. If I am reading you right, this would be just like someone on Earth thinking that water is wet and someone on Twin Earth thinking that XYZ is wet--the two thoughts are internally the same but distinguished externally semantically. However, in the "like this" case, the de re reference is to an earlier mental state, while in the water/XYZ case, the de re reference is to a physical substance.
If that's right--and I kind of like it--then that may undercut the suggestion that I was exploring in my post, namely that you can give knowledge of what red is like to someone who hasn't experienced it by inducing memories. For now it seems that the memory itself doesn't internally encode what the red experience was like, but simply carries a _de re_ reference to the experience. This would be similar to this point: If Twin Earth has no water but only XYZ, you can't induce in someone from Twin Earth a fake memory of swimming in water. For if you induce in them a copy of your memory of swimming in water, in them the de re reference shifts, and their intrinsic duplicate of your mental state represents swimming in XYZ. And if they aren't from Twin Earth but from a planet that has nothing at all like water/XYZ, then if you give them an intrinsic duplicate of your mental state, that duplicate will simply fail in de re representation of what one is swimming in.
Post a Comment