The Nature of Phenomenal Consciousness
Well it has been a month since I gave my talk at the Graduate Center. I have been meaning to write something on this, but have been swamped with the beginning of the semester. I will try to reconstruct some of the discussion, as I am finalizing my slides for my upcoming talk at the Metro-Area Research Group on Awareness and Meditation, which will include some of the stuff from this one (hopefully improved by the discussion of course! )
I finally edited and uploaded a rehearsal version of the talk, which you can view here.
During the discussion there were several very interesting themes, but I will focus on the stuff relating to HOROR theory. One theme, brought up by David Chalmers, was that on my view first-order states are never phenomenally conscious. Phenomenal consciousness, on the view I was defending, just is a higher-order representation. But this seems very odd! How could a first-order pain, say, never be phenomenally conscious?!?! I agree that this is counter-intuitive. But if that is where the path of inquiry takes us, then so be it. It is also, by the way, counter-intuitive that I am currently in motion as i sit in my chair and type this, but I am. It is also counter-intuitive that there is no absolute simultaneity, but there are good reasons to think that this is the case none the less. So, I agree that if there were no evidence at all for this view then the counter-intuitiveness of it would count against it. But there is good evidence for it, at least enough to see that it is a legitimate possibility. We have philosophical evidence from thinking about overflow and misrepresentation and we have empirical evidence from Lau’s results.
Another theme, brought up by David Rosenthal and is in some ways the flip side of Dave’s worry, was that the term ‘phenomenal consciousness’ brings with it the assumption that we are talking about a first-order property. If that is the case, that is, if it is the case that ‘phenomenal consciousness’ is defined in such a way so as to guarantee that it is a first-order property, it is contradictory, or nonsensical, to argue that it is really a higher-order property. In response I think it is important that we start with a conception of the data that is neutral about these kind of metaphysical assumptions. I think it is common for people to think of phenomenal consciousness as the property of there being something that it is like for one to be in certain states. It may be true that people then go on to make the metaphysical assumption that this is a property of first-order states but that is something additional. To say that a state is phenomenally conscious is just to say that there is something that it is like for me to be in that state. It is then an open question whether this property is a property of first-order states or a property of higher-order states. We can quibble about which words to use, and for which reasons, but there is no doubt that there is something that it is like for me to have a conscious pain and we want to know the nature of that property.
This brings up another very interesting theme of the discussion, which was what reason we have for thinking that the higher-order state is the phenomenally conscious state. Why not say, as David Rosenthal does, that the phenomenally conscious state is the state that you are conscious of yourself as being in (i.e. the first-order state)? This avoids the two previous problems. I find this move hard to accept for the following reasons. First, in the case of empty higher-order states one would then have to say that the phenomenally conscious state is not the first-order state, but rather the notional state. That is pretty weird, and I have talked about the weirdness before
. To say that there are phenomenally conscious states that have no neural correlates is very, very unsettling! Of course, it may still be true (counter-intuitiveness not all by itself a strike, etc). But, just to be clear it seems to me that if one says this in the empty case, then one must also say it in the normal case. Isn’t it extremely ad hoc to say that in the good case the first-order state has the property of being phenomenally conscious but in the bad case it is the notional state. So, I think it has to be the notional state in all cases, but then no first-order state is every phenomenally conscious! Only notional states are! In addition to this I think there is a good reason to think that it is the higher-order state that is phenomenally conscious (I mean, according to the higher-orde view). When we ask which state is phenomenally conscious we want to know ‘which state is it that there is something that it is like for the creature to be in?’ That is, we are looking for the state in virtue of which there is something that it is like for the subject. According to the higher-order view this is just the higher-order representation.
Related to this, Dan Shargel brought up the following worry. I identify the higher-order conception of phenomenal consciousness with metal appearances. Phenomenal consciousness just is a how one’s mental life appears to one. Dan suggested that this in itself pushed towards phenomenal conscious being a property of first-order states. If phenomenal consciousness is a matter of mental appearances, then it should be a matter of what appears to me, and what appears to me is my first-order states. To an extent this is right. When I have an appropriate higher-order representation I am conscious of myself as being in some first-order state. So, the way my mental life appears to me is as though I am in the first-order state. This is in fact why it is that it seems to us common sensically that phenomenal consciousness goes with the first-order states. And this is exactly what the higher-order representation is supposed to do! And we know (or at least suspect) that it can do this in the absence of the first-order state. This is why Rosenthal has said that it is a mistake to think of the higher-order state as conferring some new property onto the first-order state. The empty higher-order representation argument shows us this. So I agree that the higher-order representation makes it the case that it appears to me as though I am in some first-order state, and which state I appear to be in is just the content of the higher-order representation but I deny that this means that the first-order state comes to have some new property that it did not have before. if anything, the person has the new property, as Jake likes to point out, but of course the person has that property in virtue of being in the higher-order state, which is all that matters to me!
In many ways I see this debate as analogous to the debate between the representationist and the naive realist and there is a lot more to say about this and the other interesting questions (e.g. Cressida asked an interesting question about the ‘argument from concept acquisition’ (I think she asked how one picked out the sensory quality if one didn’t know what to look, or whether acquiring the concept required having a phenomenally conscious experience in the first place) and Rosie asked about mental appearances (basically she pointed out that I phrased my argument as ‘all phenomenal consciousness is mental appearance’ but what I needed was ‘all mental appearances is phenomenal consciousness’ since without that one could hold, as Rosenthal does, that there are mental appearances that are not involved in phenomenal consciousness (e.g. HOTs about cognitive states on his view), and Peter Godfrey-Smith asked about my notion of ‘what it is likeness’ and what I would say about fish) but I have to get to work! Hopefully I can come back to those other issues at a later date…at some point I am going to write this up as a paper but that will have to wait a bit…