I have to reply short as I am working on another long text now.
I still think that sum of your claims about the brain is physicalism in nutshell.
The fact that I see the red colour and the fact that I can’t explain in words what is the difference between red feeling and green feeling is the empirical basis of qualia.
Even if we will find that qialia is corresponding to some kind of nuronal state, it will disprove existance of exact zombies and inverted color zombies, but will not help us to answer the reason why 1234 state of neuron is “red”. I call this problem “the problem of table of correspondents”. It correspondents different states of neurons to different subjective feelings. But it is itself mystical thing as we could ask why 1234 is red, and 4321 is green, and we could imagine many different corresponding tables. Sorry to be a little bit sketchy here.
(Please feel free to take your time if you have other things to be doing.)
I am, as I think I have already said, a physicalist. If you take everything I have said and lump it together, you doubtless get physicalism. But that doesn’t mean that everything I say assumes physicalism, and in particular my main criticism of the “zombie” and “inverted spectrum” scenarios does not presuppose physicalism.
The fact that you can’t explain something in words seems like awfully thin evidence on which to base the idea that they might be different in ways that have no physical foundation.
I think the question of just what physical states correspond to what subjective states, and why given physical differences correspond to the subjective-state differences they do, is a very difficult and important problem (conditional on physicalism or something like it, of course; the question doesn’t arise if e.g. qualia are actually magical properties of your immaterial soul). I don’t think “mystical” is the right word for it, though, and in particular I see no reason in principle why it shouldn’t have (what would be for me) a satisfactory answer.
For instance, imagine that after a vast expenditure of effort and ingenuity a team of cognitive scientists, neurologists, etc., comes up with something like this:
A detailed neuron-level map of a typical human brain.
Higher-level explanations of structures within it at different scales, larger scales corresponding broadly to higher levels of abstraction.
Much analysis of how activity in these various structures (generally at the higher levels) corresponds to particular mental phenomena (surprise, sadness, awareness-of-red, multiplying small integers, imagining conversations with a friend, …).
A detailed breakdown of what happens on a typical occasion when the person whose brain it is sees (say) a tomato:
Low-level description, right down to the firing of individual neurons.
Descriptions at higher levels, each one linked to the descriptions at slightly higher and lower levels that overlap with it and to whatever abstractions are appropriate to the level being described.
“So this bit here is something we see whenever he sees something he might eat in a salad. It doesn’t fire for meat, though. We think it represents healthy food somehow.”
“This one fires in that distinctive pattern for anything that makes him think of sex. It’s not firing very strongly here, which is probably a good thing. You can see feeding into it some representations of lips and breasts, which are probably related to the colour and shape of the tomato, respectively. Men are weird.”
“These circuits here, and here, and here all tend to turn on when he sees anything red. The details of their timing and relative strength depends on the brightness of the red and how it’s located in his visual field. It doesn’t seem to be downstream of anything more sophisticated, but if you compare with what happens when he sees someone bleeding badly you’ll notice that that tends to turn this on more strongly and that it feeds into the processes that raise his autonomic arousal levels when he sees red.”
“This is a part of the language-generating subsystem. It does things whenever he says, or thinks about saying, words like ‘red’ and ‘crimson’. It also operates when he hears or reads those words, but in a slightly different way. The activity you see here tends to occur shortly before that happens—it looks like his brain is getting ready to talk about the redness it sees, if it needs to.”
A detailed breakdown of what happens when you ask him about what he’s experiencing, including the links from structures like the ones sketched above into structures that are used when introspecting and when describing things.
Of course the above is an outrageous oversimplification, and the thing these people would have to produce would presumably be a vastly complicated computer model with some currently-inconceivable user interface for looking at different parts of it at different levels of abstraction, tracing what happens in the brain at different speeds, etc.
Anyway, if I were shown something like this and could use it to follow the processes by which seeing a tomato turns into saying “What a beautiful rich red colour” (etc., etc., etc.), and likewise for the processes that happen on seeing a green apple and commenting on its colour, and to observe the parallels and differences between these processes—I would be pretty well satisfied that your “table of correspondents” problem had been adequately solved.
Anyway, for example, the fact is that there are people who are tetrachromatics—they have 4th basic colour. And assuming that i could know everything about their brain, each neuron, each connection, i still don’t know how they feel the 4th colour. It is not “correspondent table”, because on the right self of the table should be experiences.
You say: “The fact that you can’t explain something in words seems like awfully thin evidence on which to base the idea that they might be different in ways that have no physical foundation.”
I think that for our discussion is useful to distinguish two main thesis:
1) Do qualia exist? I think: yes, sure.
2) How they are connected to physical? I think: I don’t know. All attempts to create such connection results in ugly constructions, like zombies, inverted spectrum, epiphenomenalism, corresponding table—or in denial of existence of qualia.
I also want to remind my point about the post: EY tried to prove that zombies are impossible. The way he do it doesn’t work for the thought experiment with inverted spectrum. This approach doesn’t work.
I think it is clear the difference between to two types of theses: a theorem is wrong and the prove of the theorem is wrong.
There may be some things in common between what I describe and the “Mary’s room” experiment, but I’m certainly not recreating it—my position is pretty much the opposite of Jackson’s.
I agree that your and my experience of colour are probably importantly different from those of some tetrachromats. (How different depends on how far their 4th cone’s peak is from the others.) For that matter, different trichromats have cone response functions that aren’t quite the same, so even two people with “normal” colour vision don’t have the exact same colour qualia. In fact, they wouldn’t have even without that difference, because their past experiences and general psychological makeup aren’t identical. I don’t see why any of this tells us anything at all about whether qualia are physical or not, though.
I don’t understand how your “corresponding table” is an “ugly construction”. I mean, if for some reason you actually had to write down such a table then no doubt it would be ugly, but the same is true for all sorts of things we can all agree are real. Is there something about it that you think is a reason not to believe in physicalism?
I don’t think Eliezer was exactly trying to prove that zombies are impossible. He was trying to knock down an argument (based on the idea that obviously you can imagine a world just like this one but where the people are zombies) for zombies being possible, and to offer a better set of intuitions suggesting that they probably aren’t. And it doesn’t seem to me that replacing zombies with inverted qualia in any way refutes Eliezer’s argument because (1) his argument was about zombies, not about inverted qualia and (2) for the reasons I’ve already given above, I think a very similar argument does in fact apply to inverted qualia. It’s not as clear-cut as for zombies, but it seems very convincing to me.
It seems i start to understand where is the difference between our positions.
Qualia are not about cones in the eye. Because I could see colour dreams. So qualia are somewhere in the brains.
You said: “For that matter, different trichromats have cone response functions that aren’t quite the same, so even two people with “normal” colour vision don’t have the exact same colour qualia.”
(I could also see visual images if I press finger on my eye.)
So the brain use qualia to represent colours in outside world, but qualia are not actual colours.
So qualia themselves are like variables in equation. M represent mass, and F presents force in second law of Newton.
F=Ma.
But “M” is not mass, and “F” is not force, they is just variables. And if we say that now “M” is force, and “F” is mass, we will have the same equation. (It is like an experiment with inverted spectrum, btw)
M=Fa
So, qualia are variables which the brain use to denote external experiences. The same way “F” is letter from latin alphabet, which we use to denote force. The latin alphabet is completely different entity than physical forces. And when we discuss “F” we should always remember what we a speaking about—latin alphabet or force.
EDITED:
If we continue this analogy, we could imagine a computer which calculate force. It could tell us everything about results of calculations, but it can’t tell which variables it uses in its internal process.
What we conclude from here:
It could be infinitely many different variables which the computer could use. They could be inverted.
But it has to use some kind of variables, so it can’t be a zombie. Bingo! We just got new argument against p-zombie.
(The longer version of this new anti p-zombie argument is the following: thinking is impossible without variables and variables must be qualia, because the nature of qualia is that they are simple, different and unbreakable in parts, that is why I also called them “atoms of experience”—but it may need longer elaboration as it is very sketchy)
The programer of the computer chose which variables to use in this particular computer. So he created the table of correspondence in which he stated: “F is force, and M is mass”.
But qualia (at least colour qualia) are often caused by what happens to cones in the eye, and the nature of your colour qualia (even ones that occur in dreams) will depend on how your visual system is wired up, which in turn will depend on the cones in your eyes.
qualia are not actual colours
Of course not. That would be a category error. Qualia are what happens in our brains (or our immaterial souls, or wherever we have experiences) in response to external stimulation, or similar things that arise in other ways (e.g., in dreams).
qualia are variables which the brain uses to denote external experiences
This seems like a dangerous metaphor, because the brain presumably uses kinda-variable-like things at different levels, some of which have no direct connection to experience.
They could be inverted
I think how plausible this is depends on what sort of variables you’re thinking of, and is markedly less plausible for the sorts of variables that could actually correspond to qualia.
I mean, you can imagine (lots of oversimplification going on here, but never mind) taking some single neuron and inverting what happens at all its synapses, so that the activation of that neuron has the exact opposite meaning to what it used to be but everything else in the brain carries on just as before. That would be an inversion, of course. But it would be the exact opposite of what’s supposed to happen in the “inverted spectrum” thought experiment. There, you have the same physical substrate somehow producing opposite experiences; but here we have the same experiences with part of the physical substrate inverted.
But a single neuron’s internal state is not in any way a plausible candidate for what a quale could be. Qualia have to be things we are consciously aware of, and we are not consciously aware of the internal states of our neurons any more than a chess program is making plans by predicting the voltages in its DRAM cells.
I have to reply short as I am working on another long text now.
I still think that sum of your claims about the brain is physicalism in nutshell.
The fact that I see the red colour and the fact that I can’t explain in words what is the difference between red feeling and green feeling is the empirical basis of qualia.
Even if we will find that qialia is corresponding to some kind of nuronal state, it will disprove existance of exact zombies and inverted color zombies, but will not help us to answer the reason why 1234 state of neuron is “red”. I call this problem “the problem of table of correspondents”. It correspondents different states of neurons to different subjective feelings. But it is itself mystical thing as we could ask why 1234 is red, and 4321 is green, and we could imagine many different corresponding tables. Sorry to be a little bit sketchy here.
(Please feel free to take your time if you have other things to be doing.)
I am, as I think I have already said, a physicalist. If you take everything I have said and lump it together, you doubtless get physicalism. But that doesn’t mean that everything I say assumes physicalism, and in particular my main criticism of the “zombie” and “inverted spectrum” scenarios does not presuppose physicalism.
The fact that you can’t explain something in words seems like awfully thin evidence on which to base the idea that they might be different in ways that have no physical foundation.
I think the question of just what physical states correspond to what subjective states, and why given physical differences correspond to the subjective-state differences they do, is a very difficult and important problem (conditional on physicalism or something like it, of course; the question doesn’t arise if e.g. qualia are actually magical properties of your immaterial soul). I don’t think “mystical” is the right word for it, though, and in particular I see no reason in principle why it shouldn’t have (what would be for me) a satisfactory answer.
For instance, imagine that after a vast expenditure of effort and ingenuity a team of cognitive scientists, neurologists, etc., comes up with something like this:
A detailed neuron-level map of a typical human brain.
Higher-level explanations of structures within it at different scales, larger scales corresponding broadly to higher levels of abstraction.
Much analysis of how activity in these various structures (generally at the higher levels) corresponds to particular mental phenomena (surprise, sadness, awareness-of-red, multiplying small integers, imagining conversations with a friend, …).
A detailed breakdown of what happens on a typical occasion when the person whose brain it is sees (say) a tomato:
Low-level description, right down to the firing of individual neurons.
Descriptions at higher levels, each one linked to the descriptions at slightly higher and lower levels that overlap with it and to whatever abstractions are appropriate to the level being described.
“So this bit here is something we see whenever he sees something he might eat in a salad. It doesn’t fire for meat, though. We think it represents healthy food somehow.”
“This one fires in that distinctive pattern for anything that makes him think of sex. It’s not firing very strongly here, which is probably a good thing. You can see feeding into it some representations of lips and breasts, which are probably related to the colour and shape of the tomato, respectively. Men are weird.”
“These circuits here, and here, and here all tend to turn on when he sees anything red. The details of their timing and relative strength depends on the brightness of the red and how it’s located in his visual field. It doesn’t seem to be downstream of anything more sophisticated, but if you compare with what happens when he sees someone bleeding badly you’ll notice that that tends to turn this on more strongly and that it feeds into the processes that raise his autonomic arousal levels when he sees red.”
“This is a part of the language-generating subsystem. It does things whenever he says, or thinks about saying, words like ‘red’ and ‘crimson’. It also operates when he hears or reads those words, but in a slightly different way. The activity you see here tends to occur shortly before that happens—it looks like his brain is getting ready to talk about the redness it sees, if it needs to.”
A detailed breakdown of what happens when you ask him about what he’s experiencing, including the links from structures like the ones sketched above into structures that are used when introspecting and when describing things.
Of course the above is an outrageous oversimplification, and the thing these people would have to produce would presumably be a vastly complicated computer model with some currently-inconceivable user interface for looking at different parts of it at different levels of abstraction, tracing what happens in the brain at different speeds, etc.
Anyway, if I were shown something like this and could use it to follow the processes by which seeing a tomato turns into saying “What a beautiful rich red colour” (etc., etc., etc.), and likewise for the processes that happen on seeing a green apple and commenting on its colour, and to observe the parallels and differences between these processes—I would be pretty well satisfied that your “table of correspondents” problem had been adequately solved.
Hi! It seems like you are reinventing here or suggesting a variant of Mary room thought experiment, https://en.wikipedia.org/wiki/Knowledge_argument
Anyway, for example, the fact is that there are people who are tetrachromatics—they have 4th basic colour. And assuming that i could know everything about their brain, each neuron, each connection, i still don’t know how they feel the 4th colour. It is not “correspondent table”, because on the right self of the table should be experiences.
You say: “The fact that you can’t explain something in words seems like awfully thin evidence on which to base the idea that they might be different in ways that have no physical foundation.”
I think that for our discussion is useful to distinguish two main thesis: 1) Do qualia exist? I think: yes, sure. 2) How they are connected to physical? I think: I don’t know. All attempts to create such connection results in ugly constructions, like zombies, inverted spectrum, epiphenomenalism, corresponding table—or in denial of existence of qualia.
I also want to remind my point about the post: EY tried to prove that zombies are impossible. The way he do it doesn’t work for the thought experiment with inverted spectrum. This approach doesn’t work.
I think it is clear the difference between to two types of theses: a theorem is wrong and the prove of the theorem is wrong.
There may be some things in common between what I describe and the “Mary’s room” experiment, but I’m certainly not recreating it—my position is pretty much the opposite of Jackson’s.
I agree that your and my experience of colour are probably importantly different from those of some tetrachromats. (How different depends on how far their 4th cone’s peak is from the others.) For that matter, different trichromats have cone response functions that aren’t quite the same, so even two people with “normal” colour vision don’t have the exact same colour qualia. In fact, they wouldn’t have even without that difference, because their past experiences and general psychological makeup aren’t identical. I don’t see why any of this tells us anything at all about whether qualia are physical or not, though.
I don’t understand how your “corresponding table” is an “ugly construction”. I mean, if for some reason you actually had to write down such a table then no doubt it would be ugly, but the same is true for all sorts of things we can all agree are real. Is there something about it that you think is a reason not to believe in physicalism?
I don’t think Eliezer was exactly trying to prove that zombies are impossible. He was trying to knock down an argument (based on the idea that obviously you can imagine a world just like this one but where the people are zombies) for zombies being possible, and to offer a better set of intuitions suggesting that they probably aren’t. And it doesn’t seem to me that replacing zombies with inverted qualia in any way refutes Eliezer’s argument because (1) his argument was about zombies, not about inverted qualia and (2) for the reasons I’ve already given above, I think a very similar argument does in fact apply to inverted qualia. It’s not as clear-cut as for zombies, but it seems very convincing to me.
It seems i start to understand where is the difference between our positions.
Qualia are not about cones in the eye. Because I could see colour dreams. So qualia are somewhere in the brains.
You said: “For that matter, different trichromats have cone response functions that aren’t quite the same, so even two people with “normal” colour vision don’t have the exact same colour qualia.”
(I could also see visual images if I press finger on my eye.)
So the brain use qualia to represent colours in outside world, but qualia are not actual colours.
So qualia themselves are like variables in equation. M represent mass, and F presents force in second law of Newton.
F=Ma.
But “M” is not mass, and “F” is not force, they is just variables. And if we say that now “M” is force, and “F” is mass, we will have the same equation. (It is like an experiment with inverted spectrum, btw)
M=Fa
So, qualia are variables which the brain use to denote external experiences. The same way “F” is letter from latin alphabet, which we use to denote force. The latin alphabet is completely different entity than physical forces. And when we discuss “F” we should always remember what we a speaking about—latin alphabet or force.
EDITED: If we continue this analogy, we could imagine a computer which calculate force. It could tell us everything about results of calculations, but it can’t tell which variables it uses in its internal process.
What we conclude from here:
It could be infinitely many different variables which the computer could use. They could be inverted.
But it has to use some kind of variables, so it can’t be a zombie. Bingo! We just got new argument against p-zombie.
(The longer version of this new anti p-zombie argument is the following: thinking is impossible without variables and variables must be qualia, because the nature of qualia is that they are simple, different and unbreakable in parts, that is why I also called them “atoms of experience”—but it may need longer elaboration as it is very sketchy)
The programer of the computer chose which variables to use in this particular computer. So he created the table of correspondence in which he stated: “F is force, and M is mass”.
But qualia (at least colour qualia) are often caused by what happens to cones in the eye, and the nature of your colour qualia (even ones that occur in dreams) will depend on how your visual system is wired up, which in turn will depend on the cones in your eyes.
Of course not. That would be a category error. Qualia are what happens in our brains (or our immaterial souls, or wherever we have experiences) in response to external stimulation, or similar things that arise in other ways (e.g., in dreams).
This seems like a dangerous metaphor, because the brain presumably uses kinda-variable-like things at different levels, some of which have no direct connection to experience.
I think how plausible this is depends on what sort of variables you’re thinking of, and is markedly less plausible for the sorts of variables that could actually correspond to qualia.
I mean, you can imagine (lots of oversimplification going on here, but never mind) taking some single neuron and inverting what happens at all its synapses, so that the activation of that neuron has the exact opposite meaning to what it used to be but everything else in the brain carries on just as before. That would be an inversion, of course. But it would be the exact opposite of what’s supposed to happen in the “inverted spectrum” thought experiment. There, you have the same physical substrate somehow producing opposite experiences; but here we have the same experiences with part of the physical substrate inverted.
But a single neuron’s internal state is not in any way a plausible candidate for what a quale could be. Qualia have to be things we are consciously aware of, and we are not consciously aware of the internal states of our neurons any more than a chess program is making plans by predicting the voltages in its DRAM cells.