This brings me to Schlicht’s second and third claims. Schlicht is absolutely right, of course, to press me on the need for an account of the synchronic and diachronic unities of consciousness and for an account of the individuation of episodes and streams of consciousness (this collection, p. 5). I bracketed such worries because I have not worked out any such accounts to my own satisfaction. I do, however, disagree with Schlicht on the idea that regarding the organism as the subject of consciousness can help with individuation (this collection, p. 6). It is, in my view, not much easier to specify the metaphysical individuation conditions for an organism than it is to do so for an episode of consciousness. And it is problematic to assume that the brain has, so to speak, figured out what these conditions are for us. It is true that the brain must regulate a certain set of functions and processes in order to facilitate the maintenance of homeostasis; and I can even grant that doing so involves “representation” in a teleosemantic or functional-role sense. But this is orthogonal to any issues about the metaphysical individuation of organisms. While it may be easier to say what sort of processes an organism must involve (see above) and to roughly localize those processes than it is to say when one episode of consciousness begins and another ends, it is no easier to provide the ultimate metaphysical individuation conditions that ground the identities of the more basic physical processes that both organisms and consciousness depend upon. At the end of the day we are always left turning our spades with the thought that there just exists a plurality of things in the cosmos—this proton (or bare particular or property or location) is not that one—end of story!
In the case of consciousness there is no special problem of metaphysical individuation, if what we are talking about is the fact that these episodes over here are in “my” head (in this brain), and those over there are in “your” head (in that brain). From a purely epistemic point of view, it does indeed seem to be the case that we individuate episodes of consciousness by reference to individuated organisms. But so what? This is a mere contingency. If, say, we saw conscious processes (or their correlates) first, and only with great effort could we locate the organism to which the processes were attached, we would individuate organisms by referring to the streams of consciousness that “own” those organisms. There is what we might call the “epistemic relativity of individuation”. This does not mean that there are no mind-independent facts about what ultimately individuates things. It just means that, since we have no access to what the ultimate individuators are, no particular way we individuate something should be regarded as privileged. We are guided by practical and interest-relative considerations. We might as well talk about the acorns’ squirrel rather than the squirrel’s acorns, but squirrels are more entertaining to us.
In any case, for any physical process, once you drink the metaphysical individuation Kool-Aid, you won’t come back to normal. At some point you’ll just find yourself saying “this is not that”. And anything you scratch (from universes, to stars and planets, to organisms, to molecules and particles) will fall apart in this connection. I believe there are real unities in nature and that conscious mental states are such real unities (it follows that I think certain brain processes are too), but from the epistemic and conceptual point of view Nagarjuna and the Madhyamikas seem to be right—we cannot pinpoint the basic individuators anywhere, they just seem to dissolve upon analysis one way or another (see e.g., Westerhoff 2009). I believe such individuators exist but that they are inaccessible to us. And, as I argued in the paper, it is a serious confusion to think that in being self-acquainted you ipso facto have access to what it is that ultimately individuates you. You are aware of the unity and individuality of your episodes of consciousness, but those features, in turn, could depend on other unities and individuation conditions that you have no access to (cf. Williford 2011, pp. 202-203).
I am as happy as the next bloke to claim that conscious mental episodes arise and eventually give way to the next conscious mental episodes. Perhaps they overlap somehow to form a stream. Or perhaps they are punctate and we could empirically determine their temporal boundaries; I do not know. It is, however, clear that my conscious episode of reading H.P. Lovecraft for nostalgia’s sake before bed is not the conscious episode of eating yogurt for breakfast that I began the day with. But is it the same organism that eats in the morning and reads in the evening? It is indeed, given our normal (possibly partly pragmatic) individuation criteria for organisms. One would not have to be Heraclitus, however, to notice that the being that started the day is quite different from the one that unwisely decided to read Lovecraft before bed. And at a certain very fine-grained level of analysis, it is a radically different being.
We say things like “I ate yogurt this morning” and “I read Lovecraft at the age of 14” and “I am reading Lovecraft right before bed tonight”. We take this “I” to refer to the same organism and to the same “autobiographical self” to use Damasio’s term or same “ego” in Sartre’s sense of the word. But as Sartre pointed out (2004, pp. 7-9), no such temporally extended and dubitable entity could be entirely present in consciousness (so as to serve as its subjective character). Instead we have only a set of processes that remain more or less constant (and of course a causal-historical chain that is not broken). But the whole causal-historical chain does not exist at the present moment. It cannot be packed into a single conscious episode (though it could, of course, be represented in one). Nor can it, as a real pole of identity, exist throughout all conscious episodes. There is no transcendental ego. Nor is there any transcendental organism. An appeal to the organism does not, just by itself, help us with the diachronic unity problem or the individuation problem. And though the brain may somehow unconsciously always “represent” us as organisms (in perhaps the teleosemantic sense of “represent”), it is evident that we are not always conscious of ourselves as organisms (whatever exactly that is supposed to mean). Yet subjective character is there whenever consciousness is.
Thus, I do not see how the organismal theory helps with either the individuation problem for conscious episodes and streams or the related diachronic unity problem. But it is also not obvious to me how the organismal theory could help us with the synchronic unity issue either—the issue of how different phenomenal and representational contents get integrated into whole, unified conscious mental states or episodes. I fully agree with Schlicht that a brain process (e.g., an “unconscious perception” of something) can be integrated into consciousness, making a new whole, and then possibly slip out again. And I too like Edelman’s and Tononi’s “Dynamic Core” idea (see Edelman & Tononi 2000, Part IV) as a way of conceptualizing this integration and dis-integration. And I would emphatically reject the “micro-consciousness” idea of Zeki (see Zeki 2007). It does not seem to help to imagine many consciousnesses in the brain that somehow meld together to make a bigger one. (At least I hope it is not like that!)
In connection with this, I, like most people, believe that normally there is only one stream of consciousness existing in a given brain (with split-brain cases perhaps being an exception). I prefer the idea that consciousness is a type of process that has certain generic, essential structural features (temporality, subjective character, phenomenal character, representational character) and certain variable features (this comes down to variability of phenomenal and representational characters at different times). Due to some constantly fluctuating integration process, the phenomenal and representational characters of consciousness are always in flux, while temporality and subjective character remain invariant. Moreover, phenomenal and representational characters are such that they can, so to speak, expand and contract.
I can be hearing Bach’s Musikalisches Opfer while staring at a Jackson Pollock painting. I can then close my eyes so that only the beauty remains in my consciousness. If I open them again, then the visual horror will be reintegrated into it. When I closed my eyes there was “contraction”; when I opened them again, “expansion” back to the “size” the experience was before I closed them. There must indeed be something that accounts for this integration process and the resulting synchronic, differentiated unity of consciousness.
I completely agree with Schlicht that we need a theory that allows us to understand how something enters consciousness and how it gets integrated into a whole with other things that have already entered consciousness, but I do not know what that theory. Moreover, I do not know if it is possible to have a conscious experience of but one sensory quality in one modality (say, the auditory consciousness of a pure C tone) without any other thoughts, imaginings, perceptions, or anything else. I would say, though, that if one could, that episode of consciousness would still have subjective character (and temporality).
Though I agree that we all need an integration story, it is quite unclear to me that that story alone will give us a story about subjective character, unless, like Van Gulick (see e.g., 2004; cf. Metzinger 1995), one thinks that subjective character as a kind of reflexivity emerges out of integration. That may be, though I have never understood how, exactly, that is supposed to happen on Van Gulick’s view (though I do have some sense of what he means). It is not clear to me how this is supposed to happen on Schlicht’s view either (see this collection, p. 11). But I would be quite pleased if an account like this could be made to work, since deriving reflexivity from integration would, it seems to me, be a theoretical advance.
In any case, for me, subjective character as reflexivity is a phenomenological given. It is part of the data set from which I begin. It is my “Phenomenological Muse”. I could be wrong about its importance, of course, but until I am shown that, I will explore the model space that is appropriate to that intuition and leave it to empirical testing to determine whether or not the Muse was lying to me. (I will add, in a purely psychologistic vein, that many philosophers are allergic to reflexivity for purely irrational reasons. They just find it odd or too complicated or too puzzling. So they see it as an advantage if they can offer an account that gets around the need for it. For me this is like taking Marlon Brando out of Apocalypse Now or preferring decaf coffee to the real deal.)
As I see it, subjective character is like a universal or form. It is just the “reflexive” structure of consciousness. It is not to be reified into an entity (or refried like a bean, for that matter). If we think of it as an entity, we will find ourselves puzzling over questions about momentary subjects and how all these different subjects relate to each other over time and at a time. This is a confusion in my view. Yes, our normal use of language and the naïve ontology it encodes demand entities and substances to correspond to our nouns! But consciousness is not an entity or substance. It is a process; it is more like a wave than the medium the wave requires. It has a certain structure and a certain dynamical profile. Subjective character is, like temporality, an ever-replicated form that, in my view, is necessary for all consciousness. There is no subject entity strictly speaking. When there is an episode of looking back down the tunnel of previous conscious episodes that are connected in the normal way to that very episode of conscious looking, individuated subjective character is always seen. Just in terms of subjective character, all the episodes are qualitatively identical. This helps reinforce the illusion of a stable, continuous subject entity.[3] Again, there is no such entity. There is just a common form or structure living in the many different tokens. After Parfit (1984) and the Buddhists, we might say that this at once helps dissolve the thing we once thought so substantial and important and draws us closer to other tokens, no matter what stream they happen to be in.
Finally, it seems to me that Schlicht (this collection, p. 7) must take “creature-consciousness” to be more fundamental than “state-consciousness” (or “episode-consciousness”), whereas I would adhere to the usual idea that phenomenally conscious creatures are just those that host episodes (or states) of consciousness. I don't see how reversing this order helps.