Phenomenal character

State that occupies the characteristic causal role of pain for the appropriate population. This account leads to problems of a very weird technical sort — for example, what to say about someone who is Mad, Martian, and different from others in the population. But there is no need to go into these issues here. Lewis says that maybe the Madman is in pain in one sense of the term pain, whereas the Martian is in pain in another sense of the term, but he also states unequivocally that the theory is meant to be a theory of the phenomenal character of experience. It is unclear whether what it is like to be the Madman is the same as what it is like to be the Martian. Perhaps Lewis would have rejected interpersonal comparisons of this sort (Stalnaker, 1999). 

Another way to help explain the notion of phenomenal character is to reflect on the famous inverted spectrum hypothesis — that is, what it is like for you when you see red things is the same as what it is like for me when I... 

A further way to fix the referent of the term phenomenal character is to say that it is what gives rise to the explanatory gap (Levine, 1983). Tell me everything you like about what goes on physically and functionally in someone who is experiencing red, and, it seems, you still won t have told me what it is like to experience red. For even after I have all the relevant physical and functional information, I can still intelligibly ask, Why do those physical and functional goings-on generate that phenomenal character (the phenomenal character of the experience of red) Why couldn t another phenomenal character be present ... 

I shall call those philosophers who hold that phenomenal character supervenes on internal constitution (where the term phenomenal character is understood as explained earlier) phenomenal internalists. On this view, it is metaphysically impossible for intrinsic duplicates to differ with respect to the phenomenal character of their internal states. Until recently, phenomenal internalism was almost universally accepted in the philosophy of mind. With the rise of representationalism, however, the view has begun to feel some pressure (Byrne, 2001 Dretske, 1995 Jackson, 2002b Lycan, 1996 Tye, 1995, 2000). For if the phenomenal character of a state is (or supervenes on) its representational content (or a representational content the state has that meets certain further conditions, e.g., with respect to the functional role it plays) and content brings in external factors, then prima facie it is metaphysically possible for intrinsic duplicates to differ phenomenally. 

I say prima facie here since there is a weaker form of phenomenal externalism that has it that even though phenomenal character is partly constituted by causal relations to external entities, other downstream factors enter into it that preclude internal duplicates from differing phenomenally when the normal causes of, or normal environments for, their internal states differ. This form of externalism is compatible with phenomenal... 

One reaction the phenomenal internalist might have to my argument is to say so much the worse for physicalism. As will become clear later, this reaction requires the phenomenal internalist to be a dualist not only about phenomenal character but also about the representational properties of mental states. Dualism of any sort is problematic dualism this broad, doubly so. In the present context, however, my aim is not to refute the dualist. I am content to show here that a choice needs to be made either give up phenomenal internalism or give up physicalism. 

Given physicalism, the thesis of phenomenal internalism reduces to the thesis that it is metaphysically impossible for microphysical duplicates to differ with respect to the phenomenal character of their internal states. To refute this thesis, it suffices to produce an example of two entities that are microphysical duplicates in some metaphysically possible world W without the two entities being phenomenal duplicates in W. It is to the task of constructing such an example that I turn in Sections I and II. 

This might be the reaction of Jaegwon Kim. Although Kim does not take up the issue of internalism versus externalism directly in his 2005, it is clear from his remarks (pp. 172-5) that he remains a phenomenal internalist. It is also clear that he rejects physicalism for individual qualia or phenomenal characters. However, my argument creates trouble for Kim, since he is not a dualist about the representational properties of mental states. 

Here is another way to make the point. Suppose I feel a pain in a finger, and I move my finger to a different location relative to my torso. Then my pain feels to be in a different location, and this entails that there is a difference in phenomenal character before and after the movement. By contraposition, then, sameness in phenomenal character entails sameness in felt torso-relative location. Since sameness in felt location necessitates... 

In making these remarks, 1 am not assuming the truth of representation-alism with respect to phenomenal character either generally or more narrowly with respect to the phenomenal character that attaches to the experience of bodily location. According to representationalism in its weakest form, necessarily experiences that have the same representational content have the same phenomenal character. This is not assumed previously even for the special case of bodily location phenomenal character, nor is it a consequence of what I say." As just noted, what my comments entail is only that bodily sensations that feel alike with respect to bodily location (and thus have the same locational phenomenal character) must represent the same torso-relative bodily location. ... 

I shall not press the point here since it is not needed for present purposes, but, in my view, our experiences generally have what might be called a presentational phenomenology. For the appropriate external aspects, experiences with the same phenomenal character present the same aspects of the world to us or the same aspects of our bodies (or sometimes both). I focus on the case of phenomenal location, since it seems especially clear-cut and compelling. 

A second reason is that empirical work on color vision traces the phenomenology of color experiences to opponent processing channels in the brain. Differences in color experience phenomenology are explained by reference to different activation levels in these neuronal channels (Pautz, 2006). So, again, if you fix the brain events, you fix the phenomenal character or at least the phenomenal character of color experience. 

A fourth reason (suggested to me by Cory Juhl) appeals to causal considerations. Consider a microphysical duplicate of the present time slice of our world (call it MD) that (a) is governed by the same physical laws as our world and (b) is the initial slice of a world, W, with no history prior to the present time. If the physical world is causally closed, then W will unfold physically just as the actual world will. So, future behavior in W will be the same as in our world. Now, given that phenomenal externalism imposes some sort of backward-looking tracking requirement on phenomenal character, since MD is the first time slice of I, there is no phenomenal character tokened in Wat the present time. MD, then, is a zombie replica of the current time slice of the actual world. But if this is the case, then... 

A final reason that might be offered for being an internalist about phenomenal character is that the phenomenal character of an experience is an intrinsic property of it. Given that this is so, of course intrinsic duplicates... 

Too fast, 1 respond. How should we understand the term intrinsic in the premise here The term intrinsic sometimes means essential. Take the visual experience I am undergoing now, as 1 view the page before me. It is not implausible to hold that this experience could not have had a different phenomenal character. If I had been having a visual experience with a different phenomenal character, then it would not have been this very experience. If the phenomenal character of my experience is essential to it, then its phenomenal character is intrinsic to it in the preceding sense. If this is how we understand what it is for a property to be intrinsic, then the argument of the qualia internalist is straightforwardly invalid. 

Furthermore, and more importantly, why should we now accept the assumption One reply is that the truth of the assumption is revealed by introspection. But this is not so. Although it is true that introspection does not reveal phenomenal character to be an extrinsic property, this is perfectly compatible with supposing that it is. Of course, those who accept the thesis of revelation with respect to phenomenal character — that the nature of phenomenal character is fully exposed in introspection - will balk at this reply. But the thesis of revelation is a philosophical thesis, not... 

A microphysical duplicate of in a petri dish has no phenomenal character. 

The reasoning behind (4) is simply this. Suppose that there is a microphysical duplicate of in a petri dish. This duplicate will be a certain connected structure of firing patterns in an appropriate group of neurons in the dish. However, there won t be any token experience in the petri dish. For, patently, there is no subject in the dish to have an experience and, as noted in Section 1, experiences cannot exist unowned. But if there is no experience in the dish, then there is no entity in the dish having phenomenal character. And if this is true, then (4) follows. 

If the phenomenal character of f is an intrinsic property of v, then it is irreducibly nonphysical. 

But if the phenomenal character of v is irreducibly nonphysical, then it is hard to see how it can be causally efficacious with respect to behavior. Since patently it is, we have... 

The phenomenon of transparency also counts against the view that phenomenal character is intrinsic. For more on transparency see Moore ([1903] 1922), Harman (1990), and Tye (2002). 

With phenomenal characters, we seem finally to have come face to face with paradigmatic instances of intrinsic properties. The hurtfulness of pain, the acrid smell of sulphur, the taste and flavor of pineapple — these things are intrinsic qualities if anything is. 

What, then, is radical externalism It is the view that experiences are to be sorted into phenomenal types depending on what objects or properties in the environment they are experiences of. It may be boiled down to the following principle of type-individuation ifei and ex are shape experiences, then ei is of the same type or phenomenal character as if and only if the object of ei (i.e., the shape apprehended by ei) has the same geometrical properties as the object of ex. ... 

I think we are to take it that the object of a shape-experience is the apprehended shape rather than the object possessing the shape. Otherwise, CampbelFs principle of type-individuation would imply that seeing a spear point first has the same phenomenal character as seeing a spear presented sideways. 

For the radical externalist... there is no difference in the phenomenal character of shape experience in sight and in touch. The sameness of property perceived in sight and touch is transparent to the subject, and cross-modal transfer is a rational phenomenon.. 

I. If radical externalism is correct, seeing a cube and touching a cube are the same in phenomenal character. 

If seeing a square and touching a square are the same in phenomenal character, then (a) the sameness of the properties perceived will be transparent (obvious) to the subject, (b) cross-modal transfer will occur,... 

---