- If physicalism is true, all reality is effable (because it can all be expressed in the language of completed physics).
- Qualia are ineffable.
- So, physicalism is not true.
Tuesday, September 11, 2018
A simple version of the Mary argument
The following cute argument is valid.
Subscribe to:
Post Comments (Atom)
13 comments:
I agree (more or less) that the quale of red is an experience whose intentional object is an instance of redness. But this wouldn't show that qualia are effable unless colors are effable. But I'm inclined to think that colors are ineffable. Argument: if colors aren't ineffable, then it should be possible for someone to communicate to Mary what it's like to see red by just telling her: it's an experience that represents red [insert here, if necessary, an explanation in purely physical terms of what red is]. But Mary can't learn (or even come to entertain the truth concerning) what it's like to see red in that way. And this isn't because she has an inadequate grasp of "experience" or "represents." So, it must be because she has an inadequate grasp of "red."
Brian:
Colors are messy and complicated, as you know best of all. I think it's complicated because of the messy sensitivity profiles of our cones, etc.
But it seems implausible that the Mary argument would work in the case of colors but not in the case of some simpler hypothetical sensory stuff. Imagine, thus, that the Grysk are intelligent animals that simply detect some particular electromagnetic wavelength lambda1 in a non-directional way. (They do so, perhaps, because certain kinds of nebulae emit that wavelength and they have evolved to navigate between starts by means of these nebulae.) And now Merhy is a Grysk who has been deprived of her EM antenna until now. It seems that a lambda1 quale just is an experience whose intentional object is an instance of EM radiation of wavelength lambda1. And so Merhy knows all about the lambda1 quale, as long as she knows that. (So, in fact, on my view, Merhy doesn't actually need to know much physics at all. She does need to know about intentionality and experience, though.)
"But Mary can't learn (or even come to entertain the truth concerning) what it's like to see red in that way." Why not?
Hm.. I'm inclined to think there's still an epistemic gap in the Grysk case. Even if I know that the Grysk detect EM waves of wavelength lambda-11, it seems there's still an open question about what it's like for them when they detect these waves (and an open question as to whether it's like anything at all). It's hard to argue for this intuition, but here's an attempt: imagine another species on another planet, the Xs, that are just like Grysks inside the skull, but their sense receptors are tuned to a different wavelength, lambda-2 (which is the wavelength of radiation emitted by the nebulae near the Xs). So, lambda-2 waves cause the same inner state in the Xs that lambda-1 waves cause in the Grysks. Now, if I'm following you, it sounds like your view is committed to the following claim: the scenario I've described *a priori entails* that the Xs and the Grysks have *different* qualia. I find that claim implausible. (In fact, it seems at least pretty plausible to me that they would in fact have the same qualia, but here I'm only claiming that it's not *a priori* that they would have different qualia.)
""But Mary can't learn (or even come to entertain the truth concerning) what it's like to see red in that way." Why not?"
Well, I was helping myself to the (fairly widely accepted) intuition that Mary has no way of learning what it's like to see red from any black and white textbook. I find this intuition plausible, but I guess you'd deny it. I guess I accept this intuition because, very roughly, I think there is a more general epistemic gap between (A) facts purely about spatiotemporal, causal, and mathematical structure, and (B) facts about color phenomenology (or color, for that matter), and because I think (very roughly) that Mary's black and white textbooks can't convey anything beyond type-(A) facts (well, plus whatever can be gained by adding talk of "experience" and "representation" generally, but I don't think that's enough).
I agree with 2. The meaning of a quale in your counterargument seems to be exactly the same as the meaning of an experience.
If we swap "quale" and "experience" we get:
"The experience of red is a quale whose intentional object is an instance of redness."
This seems just as fine to me as the original sentence.
Brian:
It seems to me to be an intrinsic feature of the Grysk that their state has the telos of representing lambda1 radiation, and an intrinsic feature of the Xs that their state has the telos of representing lambda2 radiation. I don't see a reason to think that the two states could be qualitatively alike.
When the Grysk contemplate lambda1 radiation, they can point to their qualia and correctly say: "That's what lambda1 radiation is like" (just as we point to our qualia and say: "That's what red is like"). When the Xs contemplate lambda2 radiation, they point to their qualia and correctly say: "That's what lambda2 radiation is like." But lambda1 and lambda2 radiation are not relevantly alike. So it can't be that the same quale is what lambda1 and what lambda2 radiation is like.
Maybe I am too much of a realist.
Here's perhaps a different way of thinking about this. I am not very confident of this. But try it on for size. Physics describes various functional properties of EM radiation. But plausibly these functional properties do not actually determine the very nature of EM radiation. Plausibly there is another possible world which instead has some other kind of radiation, EM* radiation, that happens to have the same functional properties as EM radiation. (Perhaps that world has to have everything else be different as well. For instance, if it is an essential feature of fireflies that they emit EM radiation around 550nm. If so, then in that world there won't be fireflies, but there may be critters like fireflies that emit the other kind of radiation, at an analogous wavelength.) The physics journals in that world have the same words as the physics journals in our world, but these words denote other features of the world.
Consider now a Grysk in our world and a Grysk* in the other world. The Grysk is detecting lambda1 EM radiation. The Grysk* is detecting lambda1 EM* radiation. My theory is that the Grysk and Grysk* have different qualia, and by means of these different qualia they respectively see something that the physicist does not publish on: they see something relevant to the intrinsic difference between EM and EM* radiation.
It might even be correct to say that Merhy the Grysk has learned something when she first apprehended lambda1 EM radiation: she has learned something about what lambda1 EM radiation is like.
But this does not refute physicalism. It just shows that we should not identify physicalism with the claim that you can learn all about reality from the textbooks of a completed physics. For the textbooks of a completed physics will not tell you everything about the intrinsic character of EM radiation. They will presuppose your ordinary grasp of the word "light", which ordinary grasp is normally mediated by experience. Those who do not have an experience of light do not know something about light. But what they don't know is intrinsically about light, not intrinsically about themselves.
(Of course, there is probably no room in physicalism for the teleology my view presupposes.)
What about dreams? We can experience almost the same qualia (maybe a bit blurried) while sleeping, and a color experienced while Grysk is dreaming wouldn't be directly caused by lambda1.
One might argue that it would be caused by lambda1 indirectly, because Grysk experienced this color in the past directly and it created a memory from which the dream was formed. But if we had a textbook of completed physics, maybe Grysk would be able to learn about colors caused by lambda1 by studing it, without any actual lambda1 radiation, and then experience the qualia in his sleep.
Interesting. Your proposal (where experience presents the intrinsic nature of the phenomena studied by physics) is actually very similar to the view I defend in my recent "Paradise Regained" paper in AJP.
Brian:
Yeah, it looks like your paper is a better worked out version of the view. Nice work!
Do you agree that this view precludes the standard use of the Mary thought experiment against physicalism? For what Mary has learned can be put together from knowledge of the physical world plus knowledge of experience. Mary (unlike the Vampire Mary of my other post) already has knowledge of experience (say, of black and white objects). And what she learns about qualia when she sees the red tomato could have been learned simply by adding to her knowledge of experience knowledge of *tomatoes*, which I take it are thoroughly physical objects. But as it happens, the only way should gain the relevant knowledge of tomatoes is by experience.
Interestingly, the view also implies that someone in our world with an inverted color spectrum is in the throes of illusion. While there could be someone who has an inverted spectrum, he would not be on par with us, as his visual experiences would be defective. (I suspect as an empirical fact that eventually his brain would correct the visual experiences, much as brains seem to correct upside-down sensations after a few days of wearing prism glasses.)
"Do you agree that this view precludes the standard use of the Mary thought experiment against physicalism?"
I think it's a bit complicated. First, I think we'd want to distinguish between "narrowly physical facts" (roughly, the facts about structure and dynamics that physics tells us about) and "broadly physical facts" (= narrowly physical facts + quiddity facts). When Jackson argues that not all information is "physical information," he seems to have in mind *narrowly* physical information. (I think he introduces "physical information" as information of the sort the physical sciences reveal.) So I don't think the view I defend undermines Mary-style reasoning against narrow physicalism. But I guess I agree that, if the view I defend is right, then Mary doesn't know all there is to know about the extra-mental world, because she doesn't fully know (e.g.) what tomatoes are like. Maybe if she knew everything there is to know about what tomatoes (fire engines, etc.) are like, she could work out what it's like to see a tomato (etc.). But I'm not totally sure. What it's like to see red seems to involve standing in an awareness relation to redness. Even if one knows all about the distribution of redness out in the world, and knows the causal relations that hold between instances of redness and my brain, I don't know if one could deduce that I am phenomenally aware of redness. (But here I think that's because the awareness in question is an intentional relation, so we're getting into the problem of intentionality.)
"Interestingly, the view also implies that someone in our world with an inverted color spectrum is in the throes of illusion."
Yeah, I think there can't be illusion-free spectrum inversion. This is pretty much inevitable if we accept, as I do, (i) objectivism about color, (ii) color exclusion principles like "nothing can be red and green at the same time," and (iii) intentionalism about color experience.
Brian:
It's true that Jackson only works to ensure Mary has narrowly physical information. But if the argument is to rule out physicalism in general, then I think Mary needs to have all the broadly physical facts.
There are, in fact, *two* broad physicalist views that the Mary story as given doesn't rule out.
1. Qualia reduce to narrowly physical features of the brain and broadly physical features of external reality. (This is similar to the view we've been discussing.)
2. Qualia reduce to broadly physical features of the brain. (So, an isomorph of our brain made out of stuff with different quiddities will have different qualia.)
"What it's like to see red seems to involve standing in an awareness relation to redness. Even if one knows all about the distribution of redness out in the world, and knows the causal relations that hold between instances of redness and my brain, I don't know if one could deduce that I am phenomenally aware of redness. (But here I think that's because the awareness in question is an intentional relation, so we're getting into the problem of intentionality.)"
I think that's all correct. But in the original story, Mary also has direct phenomenal access to awareness -- just not awareness of red. Perhaps she can abstract the concept of awareness from her awareness of black, of white, of triangular, of squeaky and of stinky, and then combine the abstracted concept with her concept of redness in its causal relations and quiddity. The only way to rule that out is to move to Vampire Mary.
It might be worth writing some of this stuff up. But I fear that the quiddity approach will be implausible to many.
Post a Comment