In the past I've read many summaries of and glanced over Nagel's article "What's it like to be a Bat?" but never read it all of the way through. Yesterday, I got my first chance to do so (or rather, "Yesterday, curiosity finally overcame laziness and procrastination..."). Here's my attempt to synthesize the points that I found interesting--many of which surprised me, even though I had read and/or listened to multiple summaries.
The article argues that the kind of reduction of consciousness to physical processes that has been offered up to that time; instead, an "objective phenomenology" (to be defined later) is needed, although what that would be like is left open.
First, let's get to the failure of reductionism. By "reduction" he means the explanation of a process that we observe by redescribing it in terms of a process that we do not observe. Take the way water freezes, for example. We can reduce this process by talking of the molecules interlocking with each other.
The initial description includes sensible qualities, such as the way liquid, then frozen water feels. The reductive description leaves those sensible qualities behind, and offers measurable features instead.
The characteristics found in the reduced descriptions are more generic as well. Think of the explanation of how flavor changes during cooking. That could redescribed at the molecular level without any mention of flavor. Instead, the reduced description would talk of attraction, repulsion, movement, energy, size, number, mass, etc. These same terms could be used in the reduced description of other processes, for example, how TNT explodes. The reductive terms, therefore, are more generic in the sense that these same terms account for a vast array of processes in nature that seem, when we look at them, to have no qualities in common.
Nagel aims to show us that it is a mistake to try to understand consciousness by reducing it in the manner just described. To use that method for this subject matter is to take the easy way out, to fail to give due recognition to the subjective nature of experience.
What is wrong with reductive analyses of consciousness? Well, for one, they fail to "capture" the subjective character of experience because "all of them are logically compatible with the absence of the mental." One can perhaps analyze robots that behave like people in terms of their functional states (what they are good for doing) or in terms of their intentional states (what they are "trying," as it were, to do [methinks that at this point Nagel has in mind Daniel Dennett's understanding of "intentional"]). But robots or the like, when thus analyzed, can do what they do without having experience. If the same reductive analysis would be given of human consciousness and of the operation of a non-conscious thing, then that analysis has failed to capture the difference between the two.
Reductive analyses are attempted because they work so well for things other than experience. The latter sort of reduction starts with an understanding of nature that is very much located in the phenomenal. Think, for example, of water: how it tastes, feels, and looks. Chemical reduction gives us a description of water in which the phenomenal is totally removed, leaving us only characteristics that are objective--characteristics that could be recognized even by a rational creature relying on different sensory channels than we are familiar with.
Even though they work so well in describing things other than experience, they can't describe experience, says Nagel because it is "essentially connected with a single point of view and it seems inevitable that an objective, physical theory will abandon that point."
Nagel illustrates this point with the example of a bat using echolocation to perceive its surroundings. We have "no reason to suppose that it is subjectively like anything we can experience or imagine." We access what is distinctive about a bat's sonar experience by extrapolating from our own sensations. We might try to do this by imagining ourselves flying around with wings, etc. To do so might tell us what it would be like for you or me to act like a bat, but not what it would be like for the bat while it is acting that way. So the extrapolation from our own experience to sonar echolocation is going to be incomplete, generic, analogous: it will fail to capture bat experience's "specific, subjective character." Consequently, we are left believing in something mysterious, "in facts beyond the reach of human concepts"; in the truths of propositions not "expressible in a human language."
Nagel caveats that his argument is not focusing on the private nature of experience. That is, the problem he is focusing on is not the fact that individual experiences are private to their possessors (i.e., the fact that I can't access your act of experiencing in the same way that you can): rather, it is because the type of experience had by bats is inaccessible to us humans. He is making a point about types, not tokens.
Contrast our ignorance of bat echolocation with what we know about each other's experience. We can talk in objective terms about what its like for other humans to have experiences because we are similar enough to be able to adopt their (specific) point of view. We are able to know what sort of experience others have by moving, as it were, from the "first person" to the "third person." When it comes to our knowledge of bats, however, we have no comparable first person perspective to rely upon.
Nagel then cashes in the points he has made about bat experience to addresses the mind-body problem:
"IF the facts of experience... are accessible ONLY from the one [type of] point of view; THEN it is a mystery how the true character of experience could be revealed in the physical operation of that organism. The latter is a domain of objective facts par excellence--the kind that can be observed and understood from any point of view and by individuals with different perceptual systems." [my caps]
He cautions us that the sketch of this problem that he has given so far "is not by itself an argument against reduction." I think his point is that he is not arguing for dualism (it is clear elsewhere that he thinks that, while the sort of reductions tried so far have failed, there may be some other sort of reduction that may work).
Then he supports his point by talking about Martians (I'm not sure how his example helps but here goes). Suppose both a Martian (or any rational being with a sufficiently different visual apparatus) and a human understand the nature of lightning: they both understand the same objective nature even though they grasp it from different points of view (that is, with different "visual phenomenolog[ies]"). There are not two different things being understood but two different ways of experiencing the same thing. His point: the more objective our cognition, the more we are focused on that which is external to our point of view. The more subjective our cognition, on the other hand, the more we are focused on the point of view from which that which is internal to our point of view. In order to understand the objective character of lightening, one must get away from the a strictly human viewpoint. But it is difficult to see what would be left to understand regarding human experience if one who sought to understand that experience went about it by removing the strictly human viewpoint (that's a polite way of saying nothing would be left). Hence the rhetorical question:"[W]hat would be left of what it was like [for a bat] to be a bat if one removed the viewpoint of the bat?"
Nagel then goes on discuss how reductive accounts of things differ from prescientific descriptions thereof (Actually, I went over this comparison earlier because I thought it would help clear things up, so this will sound redundant, but here goes). Normally, reduction moves from from dependence upon our species-specific point of view to descriptions of the same reality that are given "in terms of more general effects and of properties detectable by means other than the human senses." Think here of how sound, which is a physical event as it is understood by those who possess hearing, comes to be understood, thanks to reduction, in terms of waves of compression and decompression in a medium. It is precisely by leaving behind the consideration of the distinctively human experience of sound that we attain a "fuller understanding of the external world." Note also, as Nagel does, that the human viewpoint, inasmuch as it has been left behind, remains unreduced.
And that unreduced subjectivity is worth knowing. As Nagel says, "Since it is the essence of the internal world and not merely a point of view of it." In other words, there is something objectively true about subjectivity. There is a need for an objective phenomenology to account for this. Such would enable us to ask better questions and offer better answers about the relation of the physical to the mental. It could show how the physicalist hypothesis might be true. At present, however, we have inadequate hypotheses. We have, for example, Donald Davidson's reasonable assertion that since mental events have physical causes and effects, "certain physical events have irreducible mental properties" or conversely, mental events must themselves have physical descriptions (he has intentions in mind rather than experience, but his comment seems applicable to experience as well) . He says this even though we are unable to produce a "general psychophysical theory" (i.e., offer an explanation of how the two are related to each other). So Nagel finds important reasons for saying that mental states are states of the body, while confessing that it is mysterious how the two terms, being so disparate, could be identified with each other. We talk about this identity like a pre-Socratic philosopher might talk about the hypothesis that matter is energy. Sure, but how? When we try to use the usual model of explanation, we treat subjective events as "effects through which mental reference to physical is secured, or else we get a false account of how mental terms refer (for example, a causal behaviorist one)." These ways don't work [because...?] In simply asserting this identity, we are like someone who left a caterpillar in an enclosed space and returned to find a butterfly. Such a person knows that the new butterfly "was once a caterpillar... without having any idea in what sense this might be so." Well, it just happened. Somehow. Don't know how, but it did, and that's all there is to it. That's not all there is to it! To think otherwise is to settle for less than a complete explanation.
Nagel closes by proposing that we develop the aforementioned objective phenomenology. Instead of trying something exotic, like conceiving the sonar experiences of bats, let's try to figure out how we might describe to blind people what it's like to see. We might come up with structural features of perceptions; or we try to develop some other features. In any case, once we came up with these objective descriptions of subjectivity, we'd be properly equipped to take on the question of how the mental and physical are to be identified: "Aspects of subjective experience that admitted this kind of objective description might be better candidates for objective explanations of a more familiar sort." But meanwhile, in the absence of good grasp of the problem of the relation of the subjective and the objective, "we cannot even pose the mind-body problem without sidestepping it."
The article argues that the kind of reduction of consciousness to physical processes that has been offered up to that time; instead, an "objective phenomenology" (to be defined later) is needed, although what that would be like is left open.
First, let's get to the failure of reductionism. By "reduction" he means the explanation of a process that we observe by redescribing it in terms of a process that we do not observe. Take the way water freezes, for example. We can reduce this process by talking of the molecules interlocking with each other.
The initial description includes sensible qualities, such as the way liquid, then frozen water feels. The reductive description leaves those sensible qualities behind, and offers measurable features instead.
The characteristics found in the reduced descriptions are more generic as well. Think of the explanation of how flavor changes during cooking. That could redescribed at the molecular level without any mention of flavor. Instead, the reduced description would talk of attraction, repulsion, movement, energy, size, number, mass, etc. These same terms could be used in the reduced description of other processes, for example, how TNT explodes. The reductive terms, therefore, are more generic in the sense that these same terms account for a vast array of processes in nature that seem, when we look at them, to have no qualities in common.
Nagel aims to show us that it is a mistake to try to understand consciousness by reducing it in the manner just described. To use that method for this subject matter is to take the easy way out, to fail to give due recognition to the subjective nature of experience.
What is wrong with reductive analyses of consciousness? Well, for one, they fail to "capture" the subjective character of experience because "all of them are logically compatible with the absence of the mental." One can perhaps analyze robots that behave like people in terms of their functional states (what they are good for doing) or in terms of their intentional states (what they are "trying," as it were, to do [methinks that at this point Nagel has in mind Daniel Dennett's understanding of "intentional"]). But robots or the like, when thus analyzed, can do what they do without having experience. If the same reductive analysis would be given of human consciousness and of the operation of a non-conscious thing, then that analysis has failed to capture the difference between the two.
Reductive analyses are attempted because they work so well for things other than experience. The latter sort of reduction starts with an understanding of nature that is very much located in the phenomenal. Think, for example, of water: how it tastes, feels, and looks. Chemical reduction gives us a description of water in which the phenomenal is totally removed, leaving us only characteristics that are objective--characteristics that could be recognized even by a rational creature relying on different sensory channels than we are familiar with.
Even though they work so well in describing things other than experience, they can't describe experience, says Nagel because it is "essentially connected with a single point of view and it seems inevitable that an objective, physical theory will abandon that point."
Nagel illustrates this point with the example of a bat using echolocation to perceive its surroundings. We have "no reason to suppose that it is subjectively like anything we can experience or imagine." We access what is distinctive about a bat's sonar experience by extrapolating from our own sensations. We might try to do this by imagining ourselves flying around with wings, etc. To do so might tell us what it would be like for you or me to act like a bat, but not what it would be like for the bat while it is acting that way. So the extrapolation from our own experience to sonar echolocation is going to be incomplete, generic, analogous: it will fail to capture bat experience's "specific, subjective character." Consequently, we are left believing in something mysterious, "in facts beyond the reach of human concepts"; in the truths of propositions not "expressible in a human language."
Nagel caveats that his argument is not focusing on the private nature of experience. That is, the problem he is focusing on is not the fact that individual experiences are private to their possessors (i.e., the fact that I can't access your act of experiencing in the same way that you can): rather, it is because the type of experience had by bats is inaccessible to us humans. He is making a point about types, not tokens.
Contrast our ignorance of bat echolocation with what we know about each other's experience. We can talk in objective terms about what its like for other humans to have experiences because we are similar enough to be able to adopt their (specific) point of view. We are able to know what sort of experience others have by moving, as it were, from the "first person" to the "third person." When it comes to our knowledge of bats, however, we have no comparable first person perspective to rely upon.
Nagel then cashes in the points he has made about bat experience to addresses the mind-body problem:
"IF the facts of experience... are accessible ONLY from the one [type of] point of view; THEN it is a mystery how the true character of experience could be revealed in the physical operation of that organism. The latter is a domain of objective facts par excellence--the kind that can be observed and understood from any point of view and by individuals with different perceptual systems." [my caps]
He cautions us that the sketch of this problem that he has given so far "is not by itself an argument against reduction." I think his point is that he is not arguing for dualism (it is clear elsewhere that he thinks that, while the sort of reductions tried so far have failed, there may be some other sort of reduction that may work).
Then he supports his point by talking about Martians (I'm not sure how his example helps but here goes). Suppose both a Martian (or any rational being with a sufficiently different visual apparatus) and a human understand the nature of lightning: they both understand the same objective nature even though they grasp it from different points of view (that is, with different "visual phenomenolog[ies]"). There are not two different things being understood but two different ways of experiencing the same thing. His point: the more objective our cognition, the more we are focused on that which is external to our point of view. The more subjective our cognition, on the other hand, the more we are focused on the point of view from which that which is internal to our point of view. In order to understand the objective character of lightening, one must get away from the a strictly human viewpoint. But it is difficult to see what would be left to understand regarding human experience if one who sought to understand that experience went about it by removing the strictly human viewpoint (that's a polite way of saying nothing would be left). Hence the rhetorical question:"[W]hat would be left of what it was like [for a bat] to be a bat if one removed the viewpoint of the bat?"
Nagel then goes on discuss how reductive accounts of things differ from prescientific descriptions thereof (Actually, I went over this comparison earlier because I thought it would help clear things up, so this will sound redundant, but here goes). Normally, reduction moves from from dependence upon our species-specific point of view to descriptions of the same reality that are given "in terms of more general effects and of properties detectable by means other than the human senses." Think here of how sound, which is a physical event as it is understood by those who possess hearing, comes to be understood, thanks to reduction, in terms of waves of compression and decompression in a medium. It is precisely by leaving behind the consideration of the distinctively human experience of sound that we attain a "fuller understanding of the external world." Note also, as Nagel does, that the human viewpoint, inasmuch as it has been left behind, remains unreduced.
And that unreduced subjectivity is worth knowing. As Nagel says, "Since it is the essence of the internal world and not merely a point of view of it." In other words, there is something objectively true about subjectivity. There is a need for an objective phenomenology to account for this. Such would enable us to ask better questions and offer better answers about the relation of the physical to the mental. It could show how the physicalist hypothesis might be true. At present, however, we have inadequate hypotheses. We have, for example, Donald Davidson's reasonable assertion that since mental events have physical causes and effects, "certain physical events have irreducible mental properties" or conversely, mental events must themselves have physical descriptions (he has intentions in mind rather than experience, but his comment seems applicable to experience as well) . He says this even though we are unable to produce a "general psychophysical theory" (i.e., offer an explanation of how the two are related to each other). So Nagel finds important reasons for saying that mental states are states of the body, while confessing that it is mysterious how the two terms, being so disparate, could be identified with each other. We talk about this identity like a pre-Socratic philosopher might talk about the hypothesis that matter is energy. Sure, but how? When we try to use the usual model of explanation, we treat subjective events as "effects through which mental reference to physical is secured, or else we get a false account of how mental terms refer (for example, a causal behaviorist one)." These ways don't work [because...?] In simply asserting this identity, we are like someone who left a caterpillar in an enclosed space and returned to find a butterfly. Such a person knows that the new butterfly "was once a caterpillar... without having any idea in what sense this might be so." Well, it just happened. Somehow. Don't know how, but it did, and that's all there is to it. That's not all there is to it! To think otherwise is to settle for less than a complete explanation.
Nagel closes by proposing that we develop the aforementioned objective phenomenology. Instead of trying something exotic, like conceiving the sonar experiences of bats, let's try to figure out how we might describe to blind people what it's like to see. We might come up with structural features of perceptions; or we try to develop some other features. In any case, once we came up with these objective descriptions of subjectivity, we'd be properly equipped to take on the question of how the mental and physical are to be identified: "Aspects of subjective experience that admitted this kind of objective description might be better candidates for objective explanations of a more familiar sort." But meanwhile, in the absence of good grasp of the problem of the relation of the subjective and the objective, "we cannot even pose the mind-body problem without sidestepping it."
Comments