| |
Charles Siewert's book, The Significance of Consciousness, can be purchased from Amazon.Com |
|
|---|
Taking the First-Person Approach:
Two Worries for Siewert's Sense of 'Consciousness'
Robert W. Lurz
Humanities Department
Indian River Community College
Ft. Pierce, FL 34981
U.S.A.
rlurz@ircc.cc.fl.us
Copyright (c) Robert W. Lurz 2001
PSYCHE, 7(14), July 2001
http://psyche.cs.monash.edu.au/v7/psyche-7-14-lurz.html
KEYWORDS: consciousness, first-person approach, phenomenal beliefs, sense-datum, terroir.
COMMENTARY ON: C. Siewert (1998) The Significance of Consciousness. Princeton: Princeton University Press. x + 374pp. ISBN: 0-691-02724-2. Price: $42.50 hardcover.
ABSTRACT: There are two things about Siewert's (1999) project that worry me. First, it's not clear to me that by taking Siewert's first-person approach, we can come to grasp what he means by 'consciousness'. And second, even if we are able to come to grasp what he means by this term, it's not clear to me that all the "consciousness-neglectful theoreticians of mind" - for example, Dennett, Rosenthal, and Tye - have failed to give an account of the property which Siewert's term picks out.
1. Introduction
Siewert's project in (1999) is to show that there is an important sense of 'consciousness' which certain prominent theoreticians in the philosophy of mind - for example, Dennett, Rosenthal, and Tye - have failed to grasp and give an account of, and which we can come to grasp by taking a particular first-person approach, which consists of considering (from the first-person perspective) certain uncontroversial examples of the presence of consciousness and contrasting (from the first-person perspective) these examples with hypothetical (but conservative) examples of the absence of consciousness. I am quite sympathetic to Siewert's project, and I agree with much of what he says against certain contemporary theories of consciousness (especially, his criticisms against higher-order theories). But, there are two things about Siewert's project that worry me. First, it's not clear to me that by taking Siewert's first-person approach, we can come to grasp what he means by 'consciousness'. And second, even if we are able to come to grasp what he means by this term, it's not clear to me that all the prominent theoreticians mentioned above have neglected to give an account of the property which Siewert's term picks out. I shall present these two worries in more detail below.
2. First Worry
There is a particular flavor - what the French call 'terroir' - that is the sine qua non for the finest Burgundies and clarets. Many would argue that the most certain way to grasp the meaning of 'terroir' is first-hand: by tasting the terroir in the finest Burgundies and clarets and contrasting it with its absence in lesser Burgundies and clarets. Of course, not everyone who follows this (expensive) first-hand approach will succeed in grasping the meaning of 'terroir': some people will confuse the terroir with other features of fine Burgundies and clarets, such as flavors of fruit or wood. But, this failure, it seems, is more a shortcoming of these people's discriminatory abilities than of the first-hand approach.
In many ways, Siewert's notion of 'consciousness' is to the mind what terroir is to the finest Burgundies and clarets: The most certain way to grasp its meaning is to take up a particular first-person approach, and upon grasping its meaning one grasps something essential about the mind. However, like in the case of terroir, not everyone who takes up Siewert's first-person approach will succeed in grasping his sense of 'consciousness': some people (such as myself) will be unable to distinguish Siewert's sense of 'consciousness' from other possible senses of this term. However, it seems to me that, unlike in the case of terroir, this failure has more to do with the shortcomings of the first-person approach than with these people's semantic discriminatory abilities - or so I intend to show. To see this, let us consider an illustration that Siewert gives of his first-person approach.
Siewert illustrates his first-person approach by asking us to consider, from the first-person point of view, the case of Connie, the legally blind achromatope, and the case of Belinda, the hypothetical reflective blindsighter. In the former case, we are asked to consider that to Connie (who has just had a bright light flashed in the left region of her visual field) there looks to be a spot of light to her left; and, hence, that Connie has a conscious visual experience of a spot of light to her left. In the latter case, we are asked to consider that to Belinda (who has just had an identical light flashed in the left region of her visual field where her scotoma lies) there does not look to be a spot of light to her left; and, hence, that Belinda does not have a conscious visual experience of a spot of light to her left.
Let us call the sense of 'looks' being used in these two cases Siewert's sense of 'looks'. It seems clear to me that we will not be able to grasp Siewert's sense of 'consciousness' through these cases unless we grasp Siewert's sense of 'looks', since the two concepts seem to be intimately connected. So, the question is whether in taking the first-person approach - illustrated by the cases of Connie and Belinda - we will be able to come to grasp what Siewert means by 'looks'? I'm not certain that we will. For to grasp what Siewert means by 'looks', we must be able to distinguish it from other sense of 'looks', and I am not sure we can do this simply through the cases of Connie and Belinda. To demonstrate this, consider the following four senses of 'looks' and 'consciousness':
1. To say that it looks to Connie, but not to Belinda, to be a spot of light on her left (and, hence, that Connie, but not Belinda, possesses a conscious visual experience) is to say that Connie, but not Belinda, is able to visually discriminate the presence and the absence of the light source on her left. (Call this the discriminatory sense of 'looks' and 'consciousness,' or simply 'looks(d)' and 'consciousness(d)'.)<1>
2. To say that it looks to Connie, but not to Belinda, to be a spot of light on her left (and, hence, that Connie, but not Belinda, possesses a conscious visual experience) is to say that Connie, but not Belinda, believes (as the direct result of the light stimulating her visual system) that there is a spot of light on her left. (Call this the visual-belief sense of 'looks' and 'consciousness,' or simply 'looks(vb)' and 'consciousness(vb)'.)<2>
3. To say that it looks to Connie, but not to Belinda, to be a spot of light on her left (and, hence, that Connie, but not Belinda, possesses a conscious visual experience) is to say that Connie, but not Belinda, believes (as the direct result of the light stimulating her visual system) that it looks to be a spot of light on her left. (Call this the phenomenal-belief sense of 'looks' and 'consciousness,' or simply 'looks(pb)' and 'consciousness(pb)'.)<3>
4. To say that it looks to Connie, but not to Belinda, to be a spot of light on her left (and, hence, that Connie, but not Belinda, possesses a conscious visual experience) is to say that Connie, but not Belinda, is directly aware of a sense datum that has the features of being spot-shaped and luminescent. (Call this the sense-datum sense of 'looks' and 'consciousness,' or simply 'looks(sd)' and 'consciousness(sd)'.)<4>
Can we use the cases of Connie and Belinda to discriminate between these senses of 'looks' and Siewert's sense of 'looks'? Certainly, we can use these cases to eliminate the possibility that what Siewert means by 'looks' is simply looks(d) or looks(vb). For, according to the stipulated conditions of the cases, both Connie and Belinda have equal visual discriminatory abilities with respect to items placed within the left region of their visual fields, and they both have (as the direct result of the light stimulating their visual systems) the visual belief that there is a spot of light on their left. Consequently, to both Connie and Belinda there looks(d) and looks(vb) to be a spot of light on their left, but to only Connie does there look (in Siewert's sense) to be a spot of light on her left. It seems, then, that we can use the cases of Connie and Belinda to distinguish Siewert's sense of 'looks' and 'consciousness' from the discriminatory and visual-belief senses of 'looks' and 'consciousness'.
But, the cases of Connie and Belinda appear incapable of helping us to eliminate the possibility that what Siewert means by 'looks' is simply looks(pb) or looks(sd). For, according to the stipulated conditions of these cases, Connie has the phenomenal belief that there looks to be a spot of light on her left, whereas Belinda does not; and there does not appear to be anything about these cases that precludes the possibility that Connie is, whereas Belinda is not, directly aware of a spot-shaped, luminescent sense datum. Consequently, the cases themselves appear to offer us no way to distinguish Siewert's sense of 'looks' from either the phenomenal-belief sense or the sense-datum sense of 'looks'. But, if these cases cannot help us to distinguish between these senses, then they cannot help us to come to know what Siewert has in mind by 'looks' and 'consciousness'. It is as if Siewert has pointed to a number of different things all at once and said, "that's what I mean by 'looks' and 'consciousness'." But, such ambiguous pointing is of no help to those who are not already in the know. It seems, then, that by taking up Siewert's first-person approach, we are left in a state of confusion over what this sense of 'looks' and 'consciousness' is that we are supposed to grasp by conceiving of the cases of Connie and Belinda from the first-person perspective.
Siewert does, of course, attempt to demonstrate that what he means by 'consciousness' is not consciousness(pb) or consciousness(sd). And he does this by presenting arguments which purport to show that these other senses of 'consciousness' have certain problems which his sense of 'consciousness' avoids. Briefly, Siewert offers his "ultra-shoes" argument in Chapter 4 to demonstrate that the phenomenal-belief sense of 'consciousness' entails that there is no genuine distinction between conscious mental states and non-conscious mental states; whereas, his sense of 'consciousness' has no such entailment. And he offers an argument in Chapter 7 to demonstrate that the sense-datum sense of 'consciousness' "commit[s] one to a conception of sensory error incompatible with the facts any would-be account of it needs to acknowledge" (p. 288); whereas, his sense of 'consciousness' has no such commitment. The issue that I have here with these arguments is not whether they are valid or sound, but whether Siewert is entitled to use them to help us home in on his sense of 'consciousness'. I do not see that he is. For in order for Siewert to give any of these arguments, he must already have grasped the distinction between his sense of 'consciousness' and the phenomenal-belief sense and the sense-datum sense of 'consciousness'- otherwise, he would not know whether the arguments he presents are against his own sense of 'consciousness' or against these other senses. However, as we saw, the cases of Connie and Belinda fail to give us a way to distinguish these different senses; and so it is difficult to see how Siewert is able to make such a distinction if we are not.
Siewert could, of course, attempt to distinguish his sense of 'consciousness' from these others by appealing to the cases of Connie and Belinda. But, to do this, he would have to change the original cases so that for both Connie and Belinda, there look(pb) and look(sd) to be a spot of light on their left, but only for Connie does there look (in Siewert's sense) to be a spot of light on her left. The original cases, in other words, would have to be changed so that both Connie and Belinda would have the phenomenal belief that there looks to be a spot of light to their left and be immediately aware of a spot-shaped, luminescent sense datum, but only to Connie would there look (in Siewert's sense) to be a spot of light to her left. But, with these changes, I'm not sure I can conceive of being someone like Belinda: Someone who believes that there looks to be a spot of light to her left, who is aware of a spot-shaped, luminescent sense datum, but who is such that there does not look (in Siewert's sense of 'look') to be a spot of light to her left. Such changes to the original cases would seem to render their conceivability controversial - something that Siewert expressly wants to avoid through his conservative first-person approach. I do not see, then, that Siewert can use the contrastive cases of Connie and Belinda to help us home in on what he has in mind by 'consciousness.'
One might argue that Siewert could rely upon his positive cases of consciousness - cases of silent speech, aural imagery, visual imagery, and other modalities of sense experience - as a way of helping us to focus in on what he means by 'consciousness'. But, I am not sure that these positive cases fare any better than the contrastive cases of Connie and Belinda. For insofar as 'consciousness(pb)' and 'consciousness(sd)' are applicable to Connie's visual experience, these terms are also applicable to the mental states of the subjects in Siewert's positive cases - at least, there is nothing about these positive cases, conceived of from the first-person perspective, that rules out their application. Silently speaking to oneself, for example, may be (in Siewert's sense) "an occurrence or episode of consciousness" (p. 68), but there appears to be nothing to appeal to (from the first-person perspective) to rule out that one's silent speech is an occurrence or episode of consciousness(pb) or consciousness(sd) (e.g., that one has a phenomenal belief that one's silent speech sounds a certain way, or that one is immediately aware of a mental image that has certain acoustical properties). It seems that the problem we faced in the contrastive cases of Connie and Belinda resurfaces in these positive cases: These cases do not adequately differentiate between Siewert's sense of 'consciousness' and the phenomenal-belief sense or the sense-datum sense of 'consciousness'.
3. Second Worry
But let us suppose, for the sake of argument, that we come to grasp what Siewert means by 'consciousness', and that we come to understand that what he means by this term is not consciousness(pb) or consciousness(sd). Now, even if we come to these realizations, it's still open to us to offer an explanation of consciousness (the very property to which Siewert's terms refers) in terms of (say) the property consciousness(pb) by theorizing that these putatively distinct properties are one and the same - much in the same way that one might try to explain what water is by identifying it with H20, admitting that 'water' does not mean the same things as 'H20'. In fact, as noted above (footnote 3), Dennett (1991) seems (on one interpretation) to put forward such a view, which he suggests in an imaginary dialogue with his critic, Otto:
[Otto:] Look. I don't just mean it. I don't just think there seems to be a pinkish glowing ring: there really seems to be a pinkish glowing ring!
[Dennett:] Now you've done it. You've fallen into a trap, along with a lot of others. You seem to think there's a difference between thinking (judging, deciding, being of the heartfelt opinion that) something seems pink to you and something really seeming pink to you. But there is no difference. (pp. 363-364).
So, according to the phenomenal-belief view, consciousness is a type of phenomenal belief: the fact that x looks (or seems) F to S just is the fact that S believes that x looks (or seems) F to him.
There are two points that I wish to make with regard to this suggested explanation of consciousness. First, the phenomenal-belief account is an explanation of consciousness that does not deny the possibility of Belinda, the reflective blindsighter. In fact, the phenomenal-belief account offers an explanation of why there does not look to Belinda to be a spot of light to her left: Belinda (unlike Connie) does not possess the phenomenal belief that there looks to be a spot of light to her left.<5> So, by Siewert's definition of 'conscious neglect,' the phenomenal-belief account is not an account that neglects giving an explanation of consciousness. Furthermore, since Dennett can be interpreted as holding a version of the phenomenal-belief account, Siewert may have incorrectly classified Dennett as belonging to the group of consciousness-neglectful theoreticians.
However, one might argue, as Siewert does (pp. 132-133), that the demerits of the phenomenal-belief account are so great that it should not even be counted as an explanation of consciousness, not even a bad explanation. I'm not convinced, however, that Siewert's argument here succeeds; and this is the second point I wish to make. Very roughly, Siewert argues that the phenomenal-belief account is not an account of consciousness (not even a bad account) since it entails that there is no genuine distinction between conscious states and non-conscious states; and, hence, it denies, in effect, the very reality of the phenomenon - consciousness - that it purports to explain. Siewert appears to rest his argument upon one of two general principles. The first, which he explicitly states, is the principle that
two genuinely distinct features will not be distinct solely in virtue of an alleged difference between the thought that something has the one feature, and the thought that something has the other (p. 132).
The second (and weaker) principle, which Siewert indirectly states, is the principle that
two genuinely distinct features will not be distinct solely in virtue of an alleged difference between the thought that something has one feature, and the thought that something has the other, unless there is a genuine difference between the thoughts.
Siewert suggests this second principle in the following imaginary conversation, which he uses to illustrate his point against the phenomenal-belief account:
Siewert's concludes that "if the distinction between consciously seeing something and blindseeing it amounted to nothing but [what the phenomenal-belief account states - namely, that the difference is simply that between believing and not believing that one consciously sees something], it would be just as fake as the distinction between shoes and ultra-shoes. There would be no genuine distinction of the sort for which [the phenomenal-belief theorist] proposed [his] account" (p. 133).
I am suspicious, however, of whether either of Siewert's two principles can be used to support his conclusion. With regard to his first principle, I am inclined to think that it begs the question against the phenomenal-belief account. Here is why. Consider that if (i) the distinction between having the phenomenal belief that (say) there looks to be a spot of light on the left and not having this phenomenal belief <6> is a genuine distinction, and if (ii) this genuine distinction is all there is to the distinction between consciously seeing a spot of light on the left and blindseeing a spot of light on the left (as the phenomenal-belief account maintains), then the distinction between consciously seeing a spot of light on the left and blindseeing a spot of light on the left must be a genuine distinction as well, which, if true, would falsify Siewert's first principle; for then there would be two genuinely distinct features - consciously seeing a spot of light on the left and blindseeing a spot of light on the left - that are distinct solely in virtue of a difference between the thought that something has the one feature, and the thought that something has the other. So, the truth of Siewert's first principle appears to require either that (i) is false or that (ii) is false. However, since Siewert's first principle does not state that the distinction between the thoughts that figure in it is a fake (i.e., non-genuine) distinction, it is not unreasonable to suppose that the principle does not require that the distinction be a fake distinction - that is, that it does not require that (i) is false. But, if this is so, then Siewert's first principle must require that (ii) is false, which would beg the question against the phenomenal-belief account; for the falsity of (ii) is simply the falsity of the phenomenal-belief account of consciousness.
Siewert's second principle, however, escapes the charge of begging the question by explicitly stating that it requires that the distinction between the thoughts that figure in it is fake (i.e., non-genuine). For recall that the second principle explicitly states that no two genuinely distinct features are distinct solely in virtue of the difference between the thought that something has the one feature, and the thought that something has the other, unless there is a genuine difference between the thoughts (i.e., provided that the difference between the thoughts is not genuine (i.e., fake)). But, the problem that Siewert's second principle seems to have is that it alone does not support his conclusion that "there would be no genuine distinction of the sort for which [the phenomenal-belief theorist] proposed [his] account." At most, Siewert's second principle supports the more modest conclusion that there would be no genuine distinction of the sort for which the phenomenal-belief theorist proposed his account, provided that the distinction between phenomenal and non-phenomenal beliefs contained in the account is a fake distinction. But, it would seem that this modest conclusion could be accepted by any phenomenal-belief theorist who takes the distinction between phenomenal and non-phenomenal beliefs contained in his account to be a genuine distinction, which, I assume, all phenomenal-belief theorist do. So, it seems, in order for Siewert to support his more ambitious conclusion with his second principle, he must show that the distinction between phenomenal and non-phenomenal beliefs contained in the phenomenal-belief account is not genuine.
Siewert, however, does not explicitly offer any reason to show that the distinction between phenomenal and non-phenomenal beliefs contained in the phenomenal-belief account is not genuine, but he does appear to suggest such a reason in his ultra-shoes case (above). In that case, Siewert suggests that the only way to distinguish ultra-shoe beliefs from regular-shoe beliefs is by appealing to the distinction between ultra-shoes and regular-shoes, which, of course, is what is at issue in the case and (hence) cannot be appealed to, making the distinction between the beliefs fake. Perhaps, Siewert wishes to give a similar argument in support of his claim against the phenomenal-belief account - namely, that the only way to distinguish phenomenal from non-phenomenal beliefs is by appealing to the distinction between conscious and non-conscious states, which, of course, is what is at issue and (hence) cannot be appealed to, making the distinction between phenomenal and non-phenomenal beliefs fake.
But, if this is the argument that Siewert wishes to give for his claim that the distinction between phenomenal and non-phenomenal beliefs is fake, its strength is questionable. For why believe that the only way to distinguish phenomenal from non-phenomenal beliefs is by appealing to the distinction between conscious and non-conscious states? To say that the latter distinction is ontologically more fundamental than the former and, therefore, ought to be used to individuate the former is, of course, to beg the question against the phenomenal-belief account, which takes these two distinctions to be ontologically equivalent. So the question remains: Is the only way to distinguish phenomenal from non-phenomenal beliefs by appealing to the distinction between conscious and non-conscious states? I'm not sure that the answer to this question is obviously 'yes'. In fact, I'm inclined to think that a phenomenal-belief theorist could reasonably answer the question in the negative. To see this, consider the following two suggestions on behalf of phenomenal-belief account.
First Suggestion. I see no reason why the phenomenal-belief theorist could not reasonably maintain that the distinction between phenomenal and non-phenomenal beliefs is a brute distinction, one which does not admit of any further explanation and one which can only be grasped from the first-person point of view. Certainly, it is not unreasonable to maintain that there exist fundamental physical distinctions - say, the distinction between fundamental micro-physical properties - which are genuine. So, I see no reason in principle why the phenomenal-belief theorist could not reasonably claim that there exists a fundamental distinction between phenomenal beliefs which is genuine. Of course, such a view may run counter to materialism and functionalism - for it admits of mental facts that are not further accounted for my physical or functional facts - but this alone is no reason for Siewert to reject the view.
Second Suggestion. I see no reason why the phenomenal-belief theorist could not reasonably maintain that the distinction between phenomenal and non-phenomenal beliefs is accounted for in largely functional terms - that is, in terms of the functional role that phenomenal beliefs play within the mind. What would this role be? Well, here is a suggestion. It seems that, in the absence of countervailing considerations, our phenomenal beliefs normally function to ground our perceptual beliefs. The grounds that someone like Connie (but not Belinda) has for believing that there is a spot of light to her left (a perceptual belief), for example, is that she believes that there looks to be a spot of light to her left (a phenomenal belief) and is unaware of any countervailing considerations. It seems plausible, then, to give a functionalist account of Connie's phenomenal belief (and phenomenal beliefs in general) in terms of it being the sort of internal state that, in the absence of countervailing considerations, normally plays the grounds for the sort of perceptual belief that Connie has; and more generally that the distinction between phenomenal beliefs and non-phenomenal beliefs is (very roughly) the distinction between mental states that, in the absence of countervailing considerations, normally ground perceptual beliefs and mental states that do not.
So, there appear to be some plausible reasons to doubt Siewert's suggestion that the only way to distinguish phenomenal from non-phenomenal beliefs is to appeal to the distinction between conscious and non-conscious states. Therefore, it seems that Siewert has not given us any grounds for believing that the distinction between phenomenal and non-phenomenal beliefs contained in the phenomenal-belief account is fake. And without such grounds, Siewert's second principle cannot be used to support the conclusion that there would be no genuine distinction between conscious and non-conscious states for which the phenomenal-belief theorist proposed his account.
In the end, it appears that Siewert's two general principles leave the phenomenal-belief account standing as a challenge to his project in (1999). As someone sympathetic to this project, I find this disappointing. I would have liked to have seen Siewert remove the challenge of the phenomenal-belief account through the use of his first-person approach. But, because of the conservative limits that Siewert places on this approach - namely, that one need not conceive of anything more metaphysically extravagant than the distinction between Connie and Belinda<7> - the approach is unable to eliminate the possibility that the phenomenal-belief account explains the nature of the property to which Siewert's notion of 'consciousness' refers. In effect, Siewert has set limits that appear to prevent his project from succeeding - at least, by way of his first-person approach. Of course, there is always the chance of demonstrating (without taking up Siewert's first-person approach) that the phenomenal-belief account fails to offer a good explanation of consciousness. But, such a demonstration - though valuable - would not show that the phenomenal-belief account is a variety of consciousness neglect, which Siewert's project seems to require.
Notes
<1> Dretske (1997) defends a discriminatory sense of 'looks'.
<2> Armstrong (1968) and Pitcher (1971) defend a visual-belief sense of 'looks'.
<3> Dennett (1991), on one interpretation, defends a phenomenal-belief sense of 'looks' (pp. 126-134; 363-365).
<4> Jackson (1977), among many others, defends a sense-datum sense of 'looks'.
<5> I do not wish to suggest that I endorse the phenomenal-belief account as a good explanation of consciousness. In fact, I am suspicious of whether it is a good explanation; for it seems to me that we simply do not have a phenomenal belief for each and every way an object may look to us. However, this criticism does not show that the phenomenal belief account fails to give an explanation of consciousness; at most, it shows that the account fails to give a good explanation of consciousness.
<6> Or simply having (as Belinda does) the belief that one sees a spot of light on the left, where 'sees' here is taken, according to Siewert's definition, to mean: "caused by internal states that are typically the effect of light stimuli."
<7> Siewert's reason for setting such limits is to avoid the use of controversial cases, such as cases of functionally equivalent, phenomenal zombies, as a way to home in on what he means by 'consciousness.'
References
Armstrong, D. (1968). A Materialist Theory of Mind. New York: Humanities Press.
Dennett, D. (1991). Consciousness Explained. Boston: Little Brown.
Dretske, F. (1997). Naturalizing the Mind. Cambridge, MA.: MIT Press.
Jackson, F. (1977). Perception. Cambridge: Cambridge University Press.
Pitcher, G. (1971). A Theory of Perception. Princeton, NJ: Princeton University Press.
Siewert, C. P. (1999). The Significance of Consciousness. Princeton, NJ: Princeton University Press.
