Burge on Perceptual Systems and Veridicality (Part 2)
In my previous post, I attributed to Burge the thesis that that the perceptual system, though a product of evolution by natural selection, has a representational function that is apriori connected to normative notion of veridicality. However, this prima facie seems like a problematic claim since natural selection is not itself concerned with normative notions, such as veridicality, and so one may wonder how our perceputal systems come to acquire such a normative concern. In this post, I will attempt to articulate one way this objection may be further unpacked.
According to Darwinian assumption, we have the particular perceptual systems that we do because they promoted the biological fitness of our ancestors. However, this seems to present a problem for Burge’s account. If we accept his insistence that a system which is unreliable may still promote biological fitness, then we seem faced with the following question: what reason do we have to think that our perceptual systems are reliable? In other words, there appears to be a mismatch between the practical mechanism (i.e., natural selection) that produced the perceptual system and the normative role the system is supposed to play. This seems to open up a lacuna between what the perceptual system has been “designed” for—namely, to promote biological fitness—and its representational function—namely, to arrive at true beliefs.
The above objection may be seen as a challenge to Burge’s claim that there is an apriori connection between the representational function of the perceptual system and verdicality. Given that these representational systems were produced by natural selection, it is plausible that they should be apriori connected to biological fitness. However, since natural selection is not apriori concerned with normative concerns, such as veridicality, it remains unclear why we should think that something natural selection has produced would be so concerned. The claim that the perceptual system may be apriori connected to veridicality presupposes that a bodily system may be apriori connected to a goal or function for which it was not “designed”. But why should we accept such an assumption?
Given Burge’s teleological framework, it would seem at least plausible that there is an apriori connection between the cardiac system and pumping blood or the respiratory system and respiration. After all, in both cases, the goals of the systems coincide perfectly with what the systems were designed for by natural selection. But since our perceptual systems were not designed by natural selection to represent veridically (given CIT), then why should we think that there is an apriori connection between the perceptual system and veridicality. Burge’s teleological framework does not seem to offer us a clear answer.
In response to the objection limned above, it may be argued that it is simply part of our concept of what it means for something to be a representational system that it aim at veridicality. But that simply pushes the present line of questioning one step back. The question now becomes, what reason do we have to think that the perceptual system is a representational system in the above (strong) sense? As we noted earlier, natural selection did not design it to be such a system. So what grounds do we have for taking it as such? Whereas before it was suggested that there is a lacuna between a system being a representational system and it aiming to represent veridically, the lacuna is now located between a system being a perceptual system and it being a representational system, in the robust sense just described.
Burge may respond by pointing out that (1) we regularly presuppose that our perceptual system is a representational system and (2) that psychologist often make this very assumption in the course of their theorising. However, it is not immediately clear why (1) or (2) should make any difference. Since we (including psychologists) did not design our own perceptual systems or those of other creatures, we hardly seem entitled to decide what type of system the perceptual system is by definitional fiat. To do so, without supporting argumentation, would be like simply stipulating that in addition to its circulatory function, the cardiovascular system fulfils some other function for which it was not designed. Such a stipulation would hardly seem warranted, even if it would allow us to fulfil particular philosophical desiderata. At the very least, the point would require substantive argumentation. But stipulation is a far cry from argumentation and it is not clear that Burge has offered any cogent argument in defence of the aforementioned proposal.
The above objection may be seen as a challenge to Burge’s claim that there is an apriori connection between the representational function of the perceptual system and verdicality. Given that these representational systems were produced by natural selection, it is plausible that they should be apriori connected to biological fitness. However, since natural selection is not apriori concerned with normative concerns, such as veridicality, it remains unclear why we should think that something natural selection has produced would be so concerned. The claim that the perceptual system may be apriori connected to veridicality presupposes that a bodily system may be apriori connected to a goal or function for which it was not “designed”. But why should we accept such an assumption?
Given Burge’s teleological framework, it would seem at least plausible that there is an apriori connection between the cardiac system and pumping blood or the respiratory system and respiration. After all, in both cases, the goals of the systems coincide perfectly with what the systems were designed for by natural selection. But since our perceptual systems were not designed by natural selection to represent veridically (given CIT), then why should we think that there is an apriori connection between the perceptual system and veridicality. Burge’s teleological framework does not seem to offer us a clear answer.
In response to the objection limned above, it may be argued that it is simply part of our concept of what it means for something to be a representational system that it aim at veridicality. But that simply pushes the present line of questioning one step back. The question now becomes, what reason do we have to think that the perceptual system is a representational system in the above (strong) sense? As we noted earlier, natural selection did not design it to be such a system. So what grounds do we have for taking it as such? Whereas before it was suggested that there is a lacuna between a system being a representational system and it aiming to represent veridically, the lacuna is now located between a system being a perceptual system and it being a representational system, in the robust sense just described.
Burge may respond by pointing out that (1) we regularly presuppose that our perceptual system is a representational system and (2) that psychologist often make this very assumption in the course of their theorising. However, it is not immediately clear why (1) or (2) should make any difference. Since we (including psychologists) did not design our own perceptual systems or those of other creatures, we hardly seem entitled to decide what type of system the perceptual system is by definitional fiat. To do so, without supporting argumentation, would be like simply stipulating that in addition to its circulatory function, the cardiovascular system fulfils some other function for which it was not designed. Such a stipulation would hardly seem warranted, even if it would allow us to fulfil particular philosophical desiderata. At the very least, the point would require substantive argumentation. But stipulation is a far cry from argumentation and it is not clear that Burge has offered any cogent argument in defence of the aforementioned proposal.
Labels: Epistemology, Perception, Tyler Burge
posted by AVERY ARCHER at
11:01 AM
4 Comments:
"This seems to open up a lacuna between what the perceptual system has been “designed” for—namely, to promote biological fitness—and its representational function—namely, to arrive at true beliefs."
Definitions!!! A term like "true beliefs" is impossibly vague. And the word "designed," even in quotes is troublesome. Are "true beliefs" those beliefs which when acted upon successfully procure food and enhance opportunities for procreation? If so, what's the problem? If not, then just what are they?
I realize that Burge is bound to have addressed these questions, but your text suggests that he has done so poorly. When an organism's perceptual apparatus gives it a survival advantage, it has either attained "true beliefs," by defintion, or the term is arguably meaningless. Admittedly, should that organism be a human being of the 21st century the context of existence is so much more sophisticated, and so much more social, that the "value" assigned to the variable "True Beliefs" is utterly changed, but the variable itself remains the same. The fact that the reward for "true beliefs" is delayed and the "value" greatly modified by repression, abstraction and a host of other psychological processes, only reflects the difficulty level of the enormously more sophisticated competition for survival.
Why not just show that in fact, sensory systems are typically extremely accurate indicators of what is going on in the world? E.g., I can measure activity in a neuron, and observe that it tracks stimulus S with such-and-such degree of accuracy. I'm saying this shouldn't necessarily be a philosophical argument: the first empirical question is, How accurate, in fact, are sensory systems in tracking the world?
Indeed, Bill Bialek, a neuroscientist, has gotten a good deal of empirical mileage out of assuming that sensory neurons optimize information transmission, and uses this to predict various features of the neuronal response.
Since perceptual abilities rest on the abilities of our sensory systems, the leap from accurate sensory systems to reliable-enough perception seems less precipitous.
Gilbert: a belief is true if what is says is happening in the world is actually happening in the world. It isn't all that vague.
E.g., 'Mars has no mass' is not true. The problem some people have with evolutionary approaches to reliable belief fixation is that promotion of fitness and truth are not necessarily the same, as the post says. For a silly example: a critter could evolve the belief that lightning is fire. This would probably promote its survival, but would be false.
The posts are about perceptual systems (human, generally speaking) and there are many degrees of separation between the the mass of Mars, as object, and the human perceptual system. I suggest that a better example might be my perception of the light pole at one corner of the property here. I can see it. It is probably some 18' tall, made of wood, brown, etc. Now let's take a closer look at what it has to offer our conversation.
First, being able to see the "truth" of it would seem to have almost no value for the "ability to pass on one’s genes". Or doesn't it? To see brown once provided one's ancesters the ability to make out animals as they passed through underbrush. To be able to make out the brown and texture of wood -- often a difference of the slightest shades of the color -- prevented one from being paralyzed with fear that there was a bear behind every tree. Being able to tell height at a distance allowed one to navigate through the forest, distinguish distant objects from close. It also let one know how far to stand away in order not to be killed by a tree that is being felled.
"But don't we live in cities now?" it might be asked. "Aren't these reasons left behind?" Now the distinctions that were genetically selected serve us in other capacities. We still need to gauge distance from height. We need to be able to make out brown cars from their backgrounds. If there were no longer survival value to these aspects of perception they would be de-selected in favor of the perceptual system of those in the population who have other more useful perceptual abilities.
But let's look once more at this light pole. Hearing is a perceptual sense, as well. But my hearing it seems is an "untruthful" sense. It lies to me. That pole is making bunches of noise. It is evening here and the temperature change is causing the pole to contract. It is making inaudible (to humans), as the rule, crackling sounds. These compliment the crackling and the pity-pat of the tiny legs of the ants that have taken up residence in the pole. One can hear none of these things. Our hearing betray us!
But does it? There is no "reason" to hear these noises. It is only the very rare instance that a human being would have to hear them in order to pass along their genes. Hearing can only “lie” when it fails to assist one to pass along one’s genes.
My eyes don't see the infrared signature of the cooling either. Dear God, how did we ever get this far? But again, we are here because our ancesters needed to see in the visual spectrum, not the infrared, in order to have the highest chance of survival. The visual spectrum is "true".
When all is said and done, the "truth" about this light pole is our truth. Although, in the present context, that truth has no meaningful direct effect on one's ability to pass along one's genes, every quality by which we identify it does. If evolution hadn’t selected for any of those qualities we wouldn’t be able to see the tree from which the pole was made, much less the pole (which, of course, wouldn’t exist). This conversation would be entirely changed but the perception that underlay the new conversation would be equally true.
Furthermore, the brown in a painting is there because brown is one of the colors our evolution has selected for. One can see the brown in a photograph because multi-color emulsions (or color dots) were not considered sufficient unless they could render brown. It's all resoundingly normative. The only other building blocks that might be available are perceived qualities or entities that might persist because they are genetically neutral.
Post a Comment
Subscribe to Post Comments [Atom]
<< Home