Daniel Dennett, disqualifying qualia, softening up the hard problem, fullness of vacuity, dysfunctional functionalism
Around track 2 of disc 9 of Intuition Pumps, Dennett offers what I would call an argument from vacuity. He argues that David Chalmers unwittingly plays a magic trick on himself and others by placing a set of issues under the one umbrella called the "hard problem of consciousness." None of these issues is really , in Dennett's opinion, a hard problem. But in naming them thus, Chalmers (says Dennett) is like a magician who seems to be playing the same card trick over and over again, but is really playing several different ones. In this analogy, expert magicians watch what they think is the same trick played over and over again. They find it unusually difficult to determine which trick he is playing because they take these performances as iterations of the same trick when each is in fact different from the one that came before. Furthermore, each of the tricks that he plays is actually an easy one, so it is precisely because they are looking for "the" one hard trick that they fail to detect what is really going on.
Philsophers have been thrown off track in a like manner by David Chalmers's characterized many easy problems as examples of "the" hard problem of consciousness. The deception begins with the word "the" title of his famous article: it focuses our attention on a commonality that isn't really there. And it can be very hard to solve a problem when one has failed to recognize that it is in fact a pseudo-problem.
Having prepared us to look for easy solutions, Dennett then turns our attention to one of the most famous instances of the hard problem: qualia. Interestingly, Dennett never gives the reader any details of what Chalmers has to say about this term (to do so would undermine his attempt to make it look easy to solve). He instead lists what he thinks are their properties: ineffable, subjective, intrinsic and private (could this be a bit of a straw-man?)
Let me add, however, that by "qualia" Chalmers means having an experience. It is something that can't be measured. It escapes description in terms of function. That is, my seeing red is not something that you can isolate in terms of what it does for you. Once you think about what qualia is supposed to be, you can see how Dennett's argument moves in a circle. So, armed with that clue, let's review Dennett's argument that there might as well not be any qualia.
In any case, after listing the aforementioned four properties, Dennett has us imagine a subject whose ability to discern colors and whose emotional responses have been extensively measured. Then, after all of these measurements have already been taken, he he wakes up one day to find that the colors that used to give negative affective responses now seem pleasant, and the colors that used to seem pleasant now affect him negatively. He then asks what he believes is a damning question: how do qualia figure in the explanation of this change? DD argues that they have zero relevance, and this irrelevance in turn indicates that they are not real.
He argues toward this conclusion by asking us whether we can tell whether the explanation for the observed change in the subject's attitude is the change in qualia or whether it is the change in quality. The subject himself thinks the qualia are the same (and for my money, this follows quite surely from the fact that qualia are intrinsic). But Dennett insists that it is reasonable to ask how we might be able to determine whether or not the subject is correct regarding the persistence of qualia.
At this point, it might help to mention a term he introduces earlier: heterophenomenology. That is his name for the testing the veracity of first-person reports by means of a third-person perspective. Dennett's questioning of whether the change in qualia caused the change in affect or vice versa is an example of heterophenomenologcial inquiry.
Back to the subject with new affect. As I said, this subject reports that things have the same colors that they had before, but he now has a different affective response to those colors (in the thought experiment conjured by Dennett, the subject's reports of emotional responses to color were carefully registered both before and after the change). Dennett proposes, however, that the reverse may have happened: the new condition may perhaps be the result of a change in qualia combined with the affective response remaining constant. He then points out that there is no way to tell which of these two hypotheses is the case: same qualia with new affect or new qualia with same affect. That is because there is no functional difference between the two. Since, qualia's role in this change makes no functinal difference, it might as well not be there. After all, it is not pulling any functional weight: affective response (which is something we can observe) is doing all the work. So we might as well get rid of qualia in our account of how the subject responds to his environment and explain the change in terms of change of affect.
Using language he introduced earlier, we might say that the third-person perspective of heterophenomenology cannot find a way of answering that question. From that perspective, then, you might say that the concept of qualia is vacuous. It is a pseudophenomenon.
But isn't this argument circular? That is, DD's task is to argue against the claim, made by proponents of qualia, that there is something more to consciousness than the functional features that be registered using the third person perspective. He sets about this task by arguing that since qualia cannot be distinguished by means of the functional third person perspective, it must not be real. He assumes that the third person perspective is the only one can use to judge the reality of things. But in adopting his "heterophenomenological" method, he assumes precisely that which he is trying to establish in this argument. His argument moves in a circle.
Case dismissed!
I suggest that he read the part about circular argumentation in chapter one of G K Chesterton's Orthodoxy, appropriately titled "The Maniac."
The funny thing is that somewhere earlier in this blog I have already constructed an argument from vacuity against the very concept of functionalism (or at least against DD's version thereof). So I am wondering if I'm guilty of similar faux pas. Gotta check that out...
My argument goes something like this: a tool can perform the same way in many different contexts while being indifferent to those contexts. A hammer being used to push down a nail doesn't "notice" whether it is the nail on a coffin or a coffee table. Similarly, the very same computer program can be used to direct traffic or play chess (okay, I admit it's far fetched, but not logically impossible). So it seems rather plausible to say that the computer doesn't know which is being done because it doesn't know anything at all. Knowing doesn't make a difference to function. Therefore, one might as well dispense with it. You know, Ockham's razor and all that...
Methinks I don't commit the same mistake as DD. But I am inclined to thank him for providing a kind of legitimation of my own modus operandi.
Philsophers have been thrown off track in a like manner by David Chalmers's characterized many easy problems as examples of "the" hard problem of consciousness. The deception begins with the word "the" title of his famous article: it focuses our attention on a commonality that isn't really there. And it can be very hard to solve a problem when one has failed to recognize that it is in fact a pseudo-problem.
Having prepared us to look for easy solutions, Dennett then turns our attention to one of the most famous instances of the hard problem: qualia. Interestingly, Dennett never gives the reader any details of what Chalmers has to say about this term (to do so would undermine his attempt to make it look easy to solve). He instead lists what he thinks are their properties: ineffable, subjective, intrinsic and private (could this be a bit of a straw-man?)
Let me add, however, that by "qualia" Chalmers means having an experience. It is something that can't be measured. It escapes description in terms of function. That is, my seeing red is not something that you can isolate in terms of what it does for you. Once you think about what qualia is supposed to be, you can see how Dennett's argument moves in a circle. So, armed with that clue, let's review Dennett's argument that there might as well not be any qualia.
In any case, after listing the aforementioned four properties, Dennett has us imagine a subject whose ability to discern colors and whose emotional responses have been extensively measured. Then, after all of these measurements have already been taken, he he wakes up one day to find that the colors that used to give negative affective responses now seem pleasant, and the colors that used to seem pleasant now affect him negatively. He then asks what he believes is a damning question: how do qualia figure in the explanation of this change? DD argues that they have zero relevance, and this irrelevance in turn indicates that they are not real.
He argues toward this conclusion by asking us whether we can tell whether the explanation for the observed change in the subject's attitude is the change in qualia or whether it is the change in quality. The subject himself thinks the qualia are the same (and for my money, this follows quite surely from the fact that qualia are intrinsic). But Dennett insists that it is reasonable to ask how we might be able to determine whether or not the subject is correct regarding the persistence of qualia.
At this point, it might help to mention a term he introduces earlier: heterophenomenology. That is his name for the testing the veracity of first-person reports by means of a third-person perspective. Dennett's questioning of whether the change in qualia caused the change in affect or vice versa is an example of heterophenomenologcial inquiry.
Back to the subject with new affect. As I said, this subject reports that things have the same colors that they had before, but he now has a different affective response to those colors (in the thought experiment conjured by Dennett, the subject's reports of emotional responses to color were carefully registered both before and after the change). Dennett proposes, however, that the reverse may have happened: the new condition may perhaps be the result of a change in qualia combined with the affective response remaining constant. He then points out that there is no way to tell which of these two hypotheses is the case: same qualia with new affect or new qualia with same affect. That is because there is no functional difference between the two. Since, qualia's role in this change makes no functinal difference, it might as well not be there. After all, it is not pulling any functional weight: affective response (which is something we can observe) is doing all the work. So we might as well get rid of qualia in our account of how the subject responds to his environment and explain the change in terms of change of affect.
Using language he introduced earlier, we might say that the third-person perspective of heterophenomenology cannot find a way of answering that question. From that perspective, then, you might say that the concept of qualia is vacuous. It is a pseudophenomenon.
But isn't this argument circular? That is, DD's task is to argue against the claim, made by proponents of qualia, that there is something more to consciousness than the functional features that be registered using the third person perspective. He sets about this task by arguing that since qualia cannot be distinguished by means of the functional third person perspective, it must not be real. He assumes that the third person perspective is the only one can use to judge the reality of things. But in adopting his "heterophenomenological" method, he assumes precisely that which he is trying to establish in this argument. His argument moves in a circle.
Case dismissed!
I suggest that he read the part about circular argumentation in chapter one of G K Chesterton's Orthodoxy, appropriately titled "The Maniac."
The funny thing is that somewhere earlier in this blog I have already constructed an argument from vacuity against the very concept of functionalism (or at least against DD's version thereof). So I am wondering if I'm guilty of similar faux pas. Gotta check that out...
My argument goes something like this: a tool can perform the same way in many different contexts while being indifferent to those contexts. A hammer being used to push down a nail doesn't "notice" whether it is the nail on a coffin or a coffee table. Similarly, the very same computer program can be used to direct traffic or play chess (okay, I admit it's far fetched, but not logically impossible). So it seems rather plausible to say that the computer doesn't know which is being done because it doesn't know anything at all. Knowing doesn't make a difference to function. Therefore, one might as well dispense with it. You know, Ockham's razor and all that...
Methinks I don't commit the same mistake as DD. But I am inclined to thank him for providing a kind of legitimation of my own modus operandi.
Comments