My Stuff

https://umass-my.sharepoint.com/:f:/g/personal/rwolff_umass_edu/EkxJV79tnlBDol82i7bXs7gBAUHadkylrmLgWbXv2nYq_A?e=UcbbW0

Coming Soon:

The following books by Robert Paul Wolff are available on Amazon.com as e-books: KANT'S THEORY OF MENTAL ACTIVITY, THE AUTONOMY OF REASON, UNDERSTANDING MARX, UNDERSTANDING RAWLS, THE POVERTY OF LIBERALISM, A LIFE IN THE ACADEMY, MONEYBAGS MUST BE SO LUCKY, AN INTRODUCTION TO THE USE OF FORMAL METHODS IN POLITICAL PHILOSOPHY.
Now Available: Volumes I, II, III, and IV of the Collected Published and Unpublished Papers.

NOW AVAILABLE ON YOUTUBE: LECTURES ON KANT'S CRITIQUE OF PURE REASON. To view the lectures, go to YouTube and search for "Robert Paul Wolff Kant." There they will be.

NOW AVAILABLE ON YOUTUBE: LECTURES ON THE THOUGHT OF KARL MARX. To view the lectures, go to YouTube and search for Robert Paul Wolff Marx."





Total Pageviews

Tuesday, March 19, 2013

GUEST POST BY PATRICK WOLFF


What It Is Like to Be a Bat


There is a confusion in the philosophy of mind concerning the possibility of offering a scientific explanation of the  nature of consciousness. As this confusion seems best embodied in Thomas Nagel's famous essay, “What Is It Like to Be a Bat?” these remarks are framed as a direct response to that question.

The confusion resides in the distinction between subjective and objective facts. An objective fact is one that can be completely described by language. A subjective fact is one that has an experiential aspect to it. Of course, it can be described by language (since language can be deployed to describe anything); however, it cannot be completely described by language because there is “something that it is like” within this fact, and therefore a complete representation requires the experience as well. (There are, no doubt, tricky philosophy-of-language issues that are raised by the above, but this simple definition will serve us for this discussion.)

Qualia are subjective facts. (Whether all subjective facts are qualia need not concern us here; the key point is that all qualia are subjective facts.) Hence, qualia cannot be completely described by language. It is easy to grasp this intuitively when we think about our everyday experience. For example, when someone describes some subjective experience, we understand that in order to grasp more fully what is being described, one must imagine oneself in that person’s experience. The description facilitates the effort, but the effort is necessary nonetheless. That we are (sometimes) successful in making this active effort is explained by the essential similarity we all share as human beings.

Now comes the confusion. It has become respectable to argue that the subjective nature of qualia suggests that a purely physicalist account of consciousness may not (or should not, or cannot) be possible. I believe this is clearly wrong. It is true that we do not at present have a physicalist account of consciousness, and may also be true that such an account is far off. (We will almost surely not have one in my lifetime, for example.) It is even possible that there will turn out to be some reason why a physicalist account may not (or should not, or cannot) be possible. But we have no reason at present to believe that a physicalist account would not be possible. Instead, we have every reason to believe that such an account should be possible, at least in principle, because we are all physical beings in the world and consciousness is a physical phenomenon.

To further clear up this confusion, let’s describe how we would determine what it is like to be a bat.

It is surely undeniable that bats have qualia. But bats, unlike people, cannot describe their qualia. And furthermore, we are enough dissimilar to a bat (e.g. we have no sense that is analogous to a bat’s sonar) that even if somehow we were presented with a description of a bat’s qualia, it seems unlikely that we could imagine ourselves into a bat’s experience. A more fundamental approach is necessary.

Before describing this fundamental approach, it may be useful to elaborate upon the normal process of imagining what another person’s qualia are like. Let’s take a simple example, like someone explaining what a meal you did not have tasted like. She might describe the tastes and compare it to other foods that you have recently eaten. You might close your eyes to eliminate competing physical sensations and then direct yourself to imagine the food. Somehow, your conscious effort might allow you to reconstruct and imagine (albeit faintly and inadequately) the relevant gastronomic qualia. At the time of this writing, we do not have anything like a complete physical description for how this process happens. But does anyone seriously doubt that it is at root a physical process? The sounds of the words hit the eardrum and are converted into signals that are processed (somehow) by the brain; the conscious faculty decides (somehow) to imagine the physical sensations conveyed by the signals; the brain (somehow) simulates the physical sensations and (somehow) conjoins those sensations to its understanding of the words, etc. Even though we cannot now provide any sort of adequate physical description of the entire process, do we have any reason to doubt that it is entirely a physical process? And furthermore, do we have any reason to believe that such a description is inherently impossible? I believe the answer to both of these questions is no.

If I am right that the last two questions are correctly answered in the negative, then it follows that we could in principle replicate scientifically what normal, human empathy approximates interpersonally all the time. Furthermore, this replication could be applied far beyond the limits of human empathy, both in terms of scope (applying to bats as well as humans) and representativeness (going far beyond what our normal imagination is capable of). Of course, there might be some scientific reason why such laboratory replication turns out to be inherently impossible. You never know what you will learn until you try: nobody living before the twentieth century would have been likely to guess the inherent physical limitations of the speed of light and the uncertainty principle, for example. Perhaps some physical limitation applies to the replication of consciousness.

But the possibility of a physical limitation of our ability to replicate qualia is completely different from the philosophical claim that “every subjective phenomenon is essentially connected with a single point of view, and it seems inevitable that an objective, physical theory will abandon that point of view.” [Nagel, http://organizations.utep.edu/portals/1475/nagel_bat.pdf, page 2] I contend that this seeming inevitability stems simply from a lack of philosophical imagination.

So now let’s describe the fundamental approach that would result in a physical theory that would meet our needs.

The first step is to develop a sufficient understanding of a bat’s brain (and associated neurological system) to create whatever qualia are desired. Let us suppose that we want to know what it is like for a bat to perceive the wall of a cave using sonar. Then we must understand the bat’s brain in sufficient detail to know precisely what physical processes are associated with “perceiving this wall under these circumstances using sonar”. In doing so, we must learn precisely where the conscious states that are associated with such perception are located, and we must identify every single relevant feature of the brain that is associated with such conscious states.

Of course, we are nowhere close to being able to do anything of the sort. It would require a body of knowledge and an empirically verified theory of bat brain physiology that is completely beyond us today. But I do not know any reason why this would be inherently impossible; it just means that our science is many, many years away from such understanding. And anyway, this is the easy part! For once we have a complete physical description of bat qualia in question, verified (however it would be verified) by experiment and theory, we must move to the second step. The qualia must be replicated in a human brain in such a way that it can be experienced by a person.

The hypothesis here is that bat brain physiology and human brain physiology are sufficiently similar that specific bat qualia can be recreated in a human brain. Of course, this may not be true: but again, we have no reason to assert with any philosophical justification that it cannot be done. The key would be to identify whatever bat stimuli are associated with specific conscious states and then reproduce those stimuli under conditions that are both theoretically justified and empirically verifiable. Under such conditions, we would have at least some reason to believe that whatever qualia would then be perceived by our human subject would represent how the qualia are perceived by the bat. Our human subject would know what it is like to be a bat.

"But wait a minute!" one might imagine our philosopher objecting. "How would we know that the qualia as experienced by the human subject would be the same as that experienced by the bat? Science is wrong all the time (indeed, that is the very point of the scientific method) so no one can know for sure that the recreated qualia in the human subject are the same as in the bat. Only the bat has unmediated access to its own qualia." Well, sure. But the special epistemological status of one's own mental states has been known since Descartes, yet the general response is not to slide into solipsism. Furthermore, we don't have to relitigate whether it is possible to have synthetic a priori knowledge to make a convincing scientific case. If we have sufficient empirical knowledge and a sufficiently justified theory to explain and predict bat qualia and the recreation of such qualia in a human subject, then we would have sufficient evidence that the qualia as experienced by the human subject would be the same as the qualia as experienced by the bat.

The question at hand was whether it is possible to provide a justified physicalist account of conscious mental states. Our answer is as follows. “In practice: not today. In theory: we have every reason to believe so and no obvious reason to think not, although we won’t know for sure until we try. Of course, since subjective facts cannot be fully described objectively, our objective, physical theory will require a subject to experience the qualia, and only that subject will have the experience. But that is just the nature of subjective facts.”

What is it like to be a bat? We don’t know today, but we have every reason to believe that it is in theory knowable. More importantly, until we learn far more than we know today we are wasting our time raising philosophical objections to the possibility of such knowledge.

 

4 comments:

John S. Wilkins said...

What bothers me about qualia is that we assert that they exist but can't describe them or define them except privatively - they are what language can't describe about our experiences. It looks for all the world that we are making inferences from locutions like "what it is like to be" as if that had to have some denotation. Subjectivity is not all that mysterious - it is a perspectival property of observer systems. The added weight that qualia are supposed to bear is where a hard problem becomes Hard, but why think it has any traction at all?

It seems to me that this is what Wittgenstein would have called a case of being bewitched by language. If we do not assume there even is a "what it is like" apart from just being that observer, the issue dissipates.

Eggs Maledict said...

I think this piece has a rather large problem, in terms of formally begging the question.

"But we have no reason at present to believe that a physicalist account would not be possible. Instead, we have every reason to believe that such an account should be possible, at least in principle, because we are all physical beings in the world and consciousness is a physical phenomenon."

The precise question at hand is whether or not consciousness is a physical phenomenon, and so relying on this claim as a premise is question-begging.


"Even though we cannot now provide any sort of adequate physical description of the entire process, do we have any reason to doubt that it is entirely a physical process?"

Certainly - the failure of a physicalist account to adequately deal with qualia problems (e.g. inverted qualia), and the empirical experience of mental states as not being physical both give us at least some reason for such doubts.

On a related note, regarding assumptions/stipulations:

"It is surely undeniable that bats have qualia."

It doesn't appear to be. Certainly we have no way of testing whether bats have qualia. It seems very likely that they do, but a high likelihood is not the same as being undeniable.

Further:

"Let us suppose that we want to know what it is like for a bat to perceive the wall of a cave using sonar. Then we must understand the bat’s brain in sufficient detail to know precisely what physical processes are associated with “perceiving this wall under these circumstances using sonar”. In doing so, we must learn precisely where the conscious states that are associated with such perception are located, and we must identify every single relevant feature of the brain that is associated with such conscious states."

There are more potential problems here, mostly epistemic. Given that we do not have access to the bat's qualia to begin with, how are we supposed to know what conscious states are present, let alone which brain states are associated with them. And even if we did know this, we would still need to show that the brain states fully explained and caused the mental states. I do not say that this is utterly impossible, but it certainly isn't obviously possible, even in principle.

"But the special epistemological status of one’s own mental states has been known since Descartes, yet the general response is not to slide into solipsism. Even if I taste the food that my friend raved about, all I can really know is what it tastes like for me; I can never know for sure what it was like for her. So what?"

So, it presents a problem for physicalist accounts.Currently the physicalist 'gold standard' in terms of philosophy of mind is functionalism, and functionalism has notorious problems dealing with qualia issues. If a better physicalist account were to be developed, we could engage with it - but currently, the lack of a good physicalist account of these issues is a good reason not to believe that physicalism is true.

This is not the stronger claim that physicalism is impossible in principle - just that we do not currently have a convincing argument that it really is the case.

"What is it like to be a bat? We don’t know today, but we have every reason to believe that it is in theory knowable."

This just hasn't been established. We have reason to believe that it may be knowable, i.e. reasons not to dismiss the concept out of hand, but physicalism also has difficult problems which have not been adequately dealt with.

"More importantly, until we learn far more than we know today we are wasting our time raising philosophical objections to the possibility of such knowledge."

Nonsense. If we were to show that physicalism contains a contradiction in itself, that would prove that it is impossible. And even if this is not the case, we might learn something about epistemic limitations, or find a good a priori argument for physicalism (or against the other contenders).

Unknown said...

Dear Wolff,
I've been reading your blog for about a year now and although I've never commented, I come back faithfully almost everyday, and always enjoy what you have to say.
I just wanted to say, in reply to Eggs Maledict, that I don't think the arguement in this post is begging the question. Yes, Patrick Wolff assumes that physicalism is true, presumably because he is convinced by good and ample evidence. But that physicalism is true is not a premise of his argument. All he's saying is that for all we know, physicalism might be able to explain qualia.

I also wanted to ask a question to R. P. Wolff himself: I was reading Rawls and an old post of yours came to my mind, where you said that once his project of providing a theorem in game theory failed, Rawls's attachment to his two principles was only comparable to faith in the Biblical word. My question is: doesn't the Difference Principle have a strong 'intuitive force'? I don't mean it in the sense in which the first principle might have intuitive force. The latter could simply be due to the 'success' of liberalism in the last few hundred years. What I mean is that if you think inequality is bad, it's prima facie a brilliant solution to allow inequality only if it benefits those that are worse off (incidentally, it might turn out that virtually no inequality at all is the only situation that satisfies this requirement; I actually think that the greatest problem with Rawls is that to know which economic system satisfies his requirements, we would need a 'scientific' economics AND we would need perfect knowledge of the present situation, of how different situations affect judgements, etc., which just seems to be impossible).
Anyway, I just assume that inequality is bad (I do not care for Nozickisms, and I don't understand why you call his a 'delightful book'; it's simply evil, because it represents the sort of thought that made it ok once again to blame poverty on the poor. Although it didn't CAUSE it, it is complicit in the destruction of the welfare state all over the Western world), and I just think Rawls's proposal 'sounds good'.
Again, thanks for your blog. It's great to have the oppprtunity to discuss these things with you!

neroden@gmail said...

For a rather excellent example, we *used to* not know how to describe other animals' perception of colors.

But now that we have analyzed their eyes and brains in great detail, we actually can construct a picture of how other animals *see* the world.

We know now.