If a subject is having a phenomenally conscious experience, there is something that it is like to be them. The subjective qualities that characterize phenomenally conscious experiences are what philosophers refer to as phenomenal character. What is the nature of phenomenal character? According to some philosophers, the phenomenal character of an experience can be explained in terms of “qualia”. Qualia, these philosophers claim, are intrinsic, non-intentional features of experience that we can become aware of via introspection. What an experience is like, on this view, can be understood in terms of its qualia.
However, qualia realism (the view that qualia exist), faces a powerful objection on the basis of the “transparent” character of experience. Michael Tye formulates one version of the argument from transparency against qualia realism as follows:
1. If one is aware of one’s visual experience or any of its qualities when one introspects, then in normal cases one can attend to one or more of those items directly on the basis of that awareness.
2. But one cannot so attend
3. So, one is not aware of one’s visual experience and/or its qualities when one introspects in normal cases.
4. So, qualia realism is false. (https://www.academia.edu/26902720/Transparency_Qualia_Realism_and_Representationalism)
When we introspect, the only features that we are able to attend to are external properties of our environment such as size, shape, color, smell, and sound. We cannot shift our attention from these features to intrinsic, qualitative properties. Consequently, qualia are not among the features that we are aware of when we introspect. This entails that qualia realism is false.
If we reject qualia realism, how should we understand the phenomenal character of our experiences? The most straightforward hypothesis is that the phenomenal qualities of an experience are one and the same as the worldly properties that the experience presents to us. The phenomenal character of an experience, on this view, is a complex of worldly properties that is related to us by our conscious state. If two experiences differ in their phenomenal character, this simply amounts to a difference in the worldly properties that those experiences present.
This proposal, if it is endorsed, provides strong support for “strong representationalism” about phenomenal character. According to strong representationalism, the phenomenal character of an experience is identical to the complex of properties that the experience represents. Why should we endorse this view? Consider two characteristic features of experience: First, experiences have accuracy conditions. In other words, an experience can be more or less accurate in its presentation of reality. Second, experiences provide us with cognitively significant information about our environment. When we experience certain environmental features, we are in a position to learn about and respond to those features on the basis of our experience. Representationalism provides a simple and compelling explanation for both of these aspects of experience.
Furthermore, we already know that our brains contain representations of our environment. It is plausible to suppose that, whenever we experience some worldly property, that property is represented by some state in our brain. If the only properties that experiences present to us are properties that are represented by our brain states, the best explanation for this fact is that experiences themselves are representations. Therefore, we should conclude that the phenomenal character of an experience is the same as the complex of properties that the experience represents (for further arguments in favor of representationalism, see the following: https://plato.stanford.edu/entries/consciousness-representational/ )
If strong representationalism is true, what explains why an experience represents some properties rather than others? Historically, the most popular explanation among representationalists has appealed to causal tracking under normal conditions. On this sort of account, an experience represents some property X iff the experience causally tracks the presence of X in normal conditions. Depending on the philosopher in question, what qualify as “normal conditions” is explained in terms of statistical frequency or evolutionary history. However, in either case, what ultimately account for the representational content of an experience are the properties that the experience causally tracks.
Despite its widespread appeal, however, this account of representational content faces serious difficulties. The most powerful objection to causal-tracking versions of strong representationalism is what has been called the “internal-dependence argument”. The internal-dependence argument is an empirical objection to externalist forms of strong representationalism that appeals to two discoveries: First, there is a bad correlation between the structure of our sensory experiences and the structure of the properties that those experiences track. Second, there is a good correlation between the structure of our conscious experiences and the structure of our internal, neural states. According to causal-tracking versions of strong representationalism, if two experiences phenomenally resemble one another in some respect, that similarity must be correlated with some similarity in the properties that those experiences causally track under normal conditions. However, when we investigate the properties tracked by our experiences of sensory properties such as color, sound, and smell, we find no such correlation. On the contrary, sensory experiences that involve similar colors, sounds, and smells are often correlated with vastly different environmental properties. Consequently, it is reasonable to conclude that the phenomenal character of our experiences cannot be explained in terms of the properties that those experiences causally track under normal conditions.
While the bad external correlation between sensory experiences and environmental properties undermines causal-tracking representationalism, it does not undermine representationalism as such. In light of the good correlation between sensory experiences and internal, neural states, we should instead opt for an internalist version of strong representationalism which claims that representational content depends, at least in part, on the internal character of our neural states. On this sort of view, if two experiences have the same representational content, they must also resemble one another in terms of some internal characteristics (for more on internal-dependence and internalist representationalism, see the following: https://www.academia.edu/41692787/How_Can_Brains_in_Vats_Experience_a_Spatial_World_A_Puzzle_for_Internalists )
However, internalist versions of representationalism face an explanatory challenge that is not faced by externalist versions: If representational content is determined by the internal character of our neural states, rather than by causal-tracking relations, why do experiences represent properties that they casually track in so many cases? When someone has an experience of something square, this is normally caused by the presence of something that is, in fact, square. The same goes for many other non-sensory properties that we experience such as size and distance. But how can this be anything other than an unbelievably lucky coincidence if internalist representationalism is true? What is it about the internal character of our neural states that ensures some degree of similarity between many properties that those states track and the properties that they represent?
In order to identify a solution to this puzzle, consider Bertrand Russell’s account of the relationship between spatial experience and the account of space provided by fundamental physics:
“The ‘real’ shape is a shape in physical space, which has no more resemblance to visual shape than light-waves have to colour. Shape as [we perceive it] is in just the same position as colour. I suggest that the apparent shape “corresponds” as a rule to a real shape, due to relations having similar logical properties. But it is a case of correspondence, not identity, just as in the case of colours and their physical correlates” (Russell, 1913)
On Russell’s view, visual experiences of shape are able to help us navigate in the real world because there is a sort of structural correspondence between visual space and the “real” space described by physics. Without endorsing Russell’s views on the natures of visual and “real” space, I would like to suggest that his notion of a structural correspondence between experience and reality provides the key to a viable internalist representationalism.
I believe that the most promising solution to the problem posed above is to hold that the contents of our experiences are determined by our neural states in virtue of a structural matching relation between the elements of those states and the properties that our experiences represent. For example, phenomenal representations of size and shape will depend upon some sort of structural match between the sizes and shapes of worldly entities and some elements of our neural states. Furthermore, representations of sensory properties such as color and sound will depend on elements of neural states that share similar structural relations to those properties as well. (Crucially, this structural matching relation can obtain even if some sensory properties are not actually instantiated in the real world.)
This account provides a straightforward explanation of the tie between, for instance, the spatial properties that our experiences causally track and the spatial properties that they represent: In order for us to remain in touch with our environment, we must have neural states that track and map practically relevant features of our environment such as size and shape. As Russell recognized, this requires those states to stand in structural relations that roughly match the structural relations of size and shape properties. Therefore, on the view under consideration, the conditions that must be met for states to causally track size and shape will guarantee that those states accurately represent size and shape as well. In summary, explaining the representational content of experience in terms of a structural matching relation between neural states and represented properties enables us to explain why experiences accurately represent many properties that they causally track while also allowing for representations of sensory properties in the absence of any causal tracking relation.
In conclusion, considerations such as the transparency of experience along with its accuracy conditions and informativeness provide an abductive case for strong representationalism about phenomenal character. In light of the empirical evidence for internal dependence, we should reject causal-tracking forms of strong representationalism in favor of internalist representationalism. Finally, in order to account for the correlations between represented and causally tracked properties such as size and shape, we should hold that the representational content of experience is determined by a structural matching relation between the elements of internal states and represented properties.
No comments:
Post a Comment