Subjective character is often described as a certain “for-me-ness,” “mineness,” or even “me-ishness” that is phenomenologically manifest and, presumably, always accompanying, even if in a muted or background form, any consciousness whatsoever (see e.g., Zahavi 2005; Levine 2001; Kriegel 2009 and Block 1995). F and related theorists point out that it also seems that one can become so absorbed in one’s actions, at one extreme, and perhaps so dulled at the other that one loses all sense of oneself (see e.g., Tononi & Koch 2008, pp. 240–241). Moreover, they might argue that it does not seem reasonable to suppose that worms and bees have a sense of self at all, and yet they may be conscious. A common reply from the defenders of subjective character to the first claim is that we are not talking about focusing on oneself or one’s current mental state as an object of attention or concern, and that, if they tried harder, F theorists would realize that even in the most dulled or, at the other pole, absorbed state, they are still aware at some level of themselves (or the very experiential state they are in). To the second objection, the typical reply is that the sort of subjective character we are envisaging does not require the sort of conceptual sophistication or reflective capabilities that would make it impossible for dogs (or even bees and worms) to count as conscious beings (see e.g., Gennaro 2012, chs. 7 & 8). Of course, the replies can be replied to, and so on. And we won’t enter into these debates here. Suffice it to say that, unsurprisingly, those who think that subjective character is essential to consciousness have ways of answering objections, just as do those who deny its essentiality. As commonly happens, the answers drive us back to questions that are themselves at least as hard to settle as the ones we began with. Moreover, appeals to the neuroscientific and psychological literature in the attempt to decide these issues sometimes get what plausibility they have from interpretations of the experiments and results that are as questionable as the claims they are supposed to support.
My view here is that one should follow the modeling path inspired by one’s “phenomenological muse” and give up fighting phenomenological intuition wars. If you find subjective character to be essential, develop models of consciousness that encode that, and see where they lead. If you don’t find it essential but find other things to be more important (multimodal information integration or availability in the global workspace or whatever), model those. And let’s not forget that we might all be working on different parts of the same elephant, so perhaps we will be able to combine models fruitfully one day. Eventually we may have ways of more or less decisively testing the different models.[22]
Different intuitions about what is essential to a phenomenon drive different models of the phenomenon. As long as enough people (and don't ask for a number) share one’s phenomenological intuitions, one’s project won’t be, we hope, insane or unmotivated. In regard to the present bifurcation point, many otherwise sane, rigorous, and careful thinkers in many widely distributed traditions and disciplines have had some version of the intuition that consciousness, somehow, involves a sense of self or sense of itself.[23]
Now, how should we characterize subjective character at the phenomenological level? It does not add much to say that it is a “sense of self.” What sort of a sense of self are we talking about? To say that it is “mineness” or “for-me-ness” makes it seem as though we are talking about the ownership of experiences. But this is probably just a certain analogy based on the ownership of property. Yes, for all that matters here, it may well be the case that, always, if I am in a position to know, without having to observe any behavior, that there is a pain in the room, then I am in a position to know that it is my own pain in the room. But it does not do much good to say that “me-ishness” or “mineness” adheres to my experiences like a property or haecceity. It is not as if I just see that my experiences have Willifordhood instead of Zahavihood or Gallagherhood, and thereby know whose are whose—like distinguishing two otherwise qualitatively identical coats by different name tags on the inner pockets.
Note that looking for a special property of the experience is not that different from seeking out its relation to a special object (its owner or The Self) that one may be directly acquainted with. In both cases we are looking for a special something that individuates the experience. There is no interesting difference here between a special unique property that only my experiences have and a special unique self-object to which they all relate.
Subjective character should probably not be thought of as a matter of a constant relation to a self-object or as a special property of mineness or me-ishness that all experiences come with, all the more is this so if it is possible to misattribute ownership to certain sensations.[24] The first-personal dimension (Zahavi), the sense of self in the act of knowing (Damasio), for-me-ness, me-ishness (Block), ipseity, être-pour-soi (Sartre), Selbstvertrautheit, and so on—these are all suggestive names for the phenomenon in question. But we’d like to know if there is not an at least somewhat less ambiguous way of characterizing it.
One name for it that I do rather like depends on a grammatical analogy that can be fleshed out a bit more. Every experience, we may say, involves the appearance of something to something (or someone). The former can be called the genitive of manifestation (appearance-of), the latter the dative of manifestation (appearance-to).[25] The genitive of manifestation corresponds to the intentionality of consciousness—its directedness at objects; the dative of manifestation corresponds to subjective character. The identification of subjective character and the dative of manifestation may not at first be so obvious.
The primary intuition here is that there is no such thing as the mere non-relational phenomenal appearance of an object or quality. Objects and qualities don’t just phenomenally manifest—full stop. Rather, anything that phenomenally appears, appears to someone or something (cf. Strawson 2011, pp. 41–46). If this were false, phenomenal consciousness would be more like a monadic property of its objects than like a relation between a subject and an object of some sort (see Butchvarov 1979, p. 250). The idea that consciousness could be phenomenally manifest but manifest to no one is either incoherent or, at best, strains credulity. Yet this seems to be exactly what F and related theorists are committed to—aches and pains that can appear (be phenomenally conscious) but appear to no one.
If we accept that there is a dative of manifestation, that objects and qualities appear to someone or something, we are closer to but not quite up to subjective character just yet. Subjective character, recall, is supposed to be something phenomenologically detectable. And one might raise the following sort of worry. Suppose phenomenally manifest objects and properties are manifest to something or someone. It does not follow from this alone that that to which they are manifest is itself manifest or even manifestable. Nor does it follow that the fact that they are manifest to something is manifest or even manifestable. In other words, there could indeed be a dative of manifestation and yet no direct phenomenological evidence of this at all. In fact, Hume’s famous failure to find his own self and Moore’s similar but more tentative musings on this issue can be taken as expressions of the intuition that we do not find a distinct subject relatum in experience.[26] And surely it is true that we do not find a little ubiquitous homunculus—the constant and ever-present thing Hume might have been seeking, like the little face at the bottom of old first-person video games like Quake—to which all our experiences relate—nor do we find a self-haecceity forever re-instantiated by our conscious episodes.
There is, however, this strong intuition that phenomenal consciousness is relational, that it involves a subject-object polarity. And the strong intuition that we do not find any entity or special criterial property that could be a self-entity, me-haecceity, me-ish quale, or subject-relatum is in some apparent tension with this intuition of relationality. Moreover, a hidden subject-relatum would not account for the phenomenology of subjective character, evidently. There is a real question here. How is it that consciousness seems to have a subject-object relational structure, and yet we do not seem to be able to find the subject-relatum, one of the relata of the relation? Isn’t it the case that if something non-inferentially seems relational, then we are non-inferentially aware of its (at least) apparent relata? Speaking naïvely and barring certain irrelevant counterexamples, if I see that the cup is on the table, don’t I see the cup and see the table too? In the case of the subject-object polarity, do we imagine or project this relation? Is it a product of reflection and memory?
It seems to me that the F theorist should say that it is somehow a product of higher cognition that is projected onto normal adult human conscious experience. But if one is really committed to the intuition that subjective character is an essential and hence ubiquitous feature of conscious experience, then one will simply have to abandon self-relatum and self-haecceity accounts as characterizations of the phenomenology (and as explanatory models, for that matter). What we need is an account of how it is that consciousness manifestly and non-inferentially appears to have a relational structure even though one of the relata is, in a certain sense, invisible.
Here the view that consciousness is self-manifesting can save the day. An episode or perhaps stream of consciousness, on this view, appears to itself at every moment while other things appear to it as well. This will require more unpacking, but at present we just want to clarify the putative phenomenological content of the claim as best we can. We leave the notion of appearing or of phenomenally manifesting undefined. Or, if you prefer, we define it ostensively by inner ostension and hope that our interlocutors know what we are talking about and have similar conscious minds (cf. Fales 1996, pp. 147–148).
Let’s say that phenomenal manifestation is just the appearance to/in consciousness of something. Let’s leave it open what that something is (qualities, facts, objects). We all can know what phenomenal manifestation is, in this purely phenomenological sense, if we are conscious and capable of normal reflection, attention, memory, and conceptual cognition. If we have tasted coffee, then the taste of coffee has been phenomenally manifest to us. If we haven’t, then it has not. And think of this generically—it’s what experiencing the taste of coffee has in common with seeing the blue sky and with feeling one’s own existence.[27] Now, the claim is that an episode of consciousness is phenomenally manifest to itself whenever anything else is phenomenally manifest to/in that episode. Whenever anything else appears to consciousness, that act or episode or stream of consciousness appears to itself as well. And it is important to remember that this does not mean that one is reflecting on one’s experience or that one has any propositional attitude towards that experience or that one is paying any attention to that experience as such.
Now, let us suppose that this is the case. Can we recover a notion of subjective character from this in a way that accounts for both the Humean intuition that the subject-relatum is, in some sense, invisible and that, nevertheless, consciousness has a subject-object relational structure that is phenomenally manifest and non-inferentially knowable? Yes, we can, and at a relatively low price.
The subject-relatum, on the current proposal, is just the episode of consciousness itself. The episode appears to/in the episode. Other things (qualities, objects, etc.) appear in/to the episode as well. The episode is a unified whole, the differentiated qualities and objects appearing in/to it are like its parts (stressing “like”—it’s an analogy).[28] We do not find episodes that do not have parts (except perhaps in some very special circumstances), but it is foolhardy to look for some special entity or haecceity that is separable from all the other parts or like a part among the parts. There is no such thing. And that, arguably, is the sort of thing Hume was failing to find. No such subject is given, hence we don’t find it. Nonetheless, the true subject-relatum, the episode of consciousness itself, is not invisible. It is manifest.
The main price to pay here is that we must try to wrap our heads around the idea that an episode of consciousness could be the phenomenological subject of consciousness. I say, and say truly, that such and such appears to me or that I see, feel, hear, or am conscious of such and such. If I am a subject of consciousness and all subjects of consciousness are just so many episodes, then am I just an episode of consciousness?! I’ve seen the incredulous stares with my own eyes and have been told that the sentence expressing the view that the subject of consciousness is the episode of consciousness has the same status as sentences like, “Pink dreams sleep furiously.”
Indeed, this claim seems wildly counterintuitive at first. But once we realize that there is a certain temporal element connoted in our usage of “I,” then this can be ameliorated. “I” normally refers not just to the present experience but to a whole history of connected experiences and much else besides. So it would be a mistake to infer from “I’ve seen the incredulous stares” the claim that “Incredulous stares were seen in/by this current episode of consciousness.” Instead, in the spirit of Four Dimensionalism, one should translate thus: There was a past series of conscious episodes suitably connected to each other and to the present one; incredulous stares were seen by/in them for some time; and the episodes are being recalled in/by the present conscious episode, which bears the same relation (transitively conceived), or some suitable analogue thereof, in the case of broken streams, to that sequence of earlier episodes.
Note, however, that fundamentally the use of “I” is anchored in moment-by-moment, self-manifesting conscious experience. Imagine a person with severe anterograde amnesia and retrograde amnesia as well. Such a person might think, from moment to moment, “I am seeing this,” “I am feeling that,” but beyond a certain perhaps necessary amount of working memory, they may not carry any of that information into their future. We can imagine truly minimal subjects that have only the minimal amount of working memory required for consciousness, supposing that some amount is required. On the view proposed here, such a conscious being’s consciousness would still have subjective character. It would simply fail to be more or less automatically enriched by memory, projection, familiarity with one’s body and dispositions, autobiographical idealizations and distortions, etc., that is, by the autobiographical representational grid through which our experience is normally spontaneously filtered. Perhaps such a person could not think “I” in the sense in which we normally think it. They may lack an “autobiographical self” and even “extended consciousness”, as Damasio would put it (see Damasio 1999 and 2010). But their experience would be self-manifest and other things (“parts”) would be manifest in/to that experience as well.
Still, isn’t it a bit too odd to hold that the whole episode is conscious of its “parts”—however we end up construing these? Or that the “parts” are phenomenally manifest to/in the whole they belong to? Doesn’t this still seem like a totally bizarre thing to say? We have to remind ourselves that there is no thing in consciousness, no ego entity, no homunculus that these qualities could be manifest to. We don’t find any such thing; and no hidden thing could allow us to account for the phenomenology. However, we agreed (I hope) that consciousness has a relational, subject-object structure and that this structure is itself phenomenally manifest and not inferred.
Another way to put it is to say that there is a kind of contrast present in our experience all the time. Something is before me, and it is not me. Something is present to consciousness, but it is not that consciousness. Given our mereological analogy, this contrast is a bit like that between a whole and its proper parts. The whole is not a proper part. Yet, at a suitably generic level, it bears the same relation to itself that it bears to its constituents (everything is a part of itself too, though an improper part).
Assuming that this relational structure is not projected onto the experience in reflection, assuming that is, that this is a genuine “prepredicative” structure of experience, the contrast between the subject-pole and the object-pole is manifest, even if it normally remains unthematized or attended to as such. On the hypothesis that consciousness is always self-manifesting, there is no problem here. The relevant contrast is like the contrast between the parts and their unified whole. The parts are manifest. The whole is manifest (self-manifest). So all the needed elements are present for their relations (of differentiation, unification, and inclusion) to be manifest.
Moreover, the idea that the difference between the parts and the whole is prepredicatively manifest is no more implausible than the idea that the difference between parts and other parts is prepredicatively manifest, something almost no one would deny. If I see a red patch on a black background, I have a differentiated, contrastive visual experience. The same goes for differences between the sensory modalities: we see and hear simultaneously, etc. If those sorts of contrasts can figure into the ground level of experience, why not the contrast between the unified self-manifesting whole and all its manifest “parts”—the totality of simultaneously manifesting qualities (however we understand them exactly) in all modalities (sensory and possibly cognitive, conative, and affective)?
Subjective character then, on this view, is just the self-manifesting character of an episode of consciousness. This view has the nice feature that it allows us to simultaneously account for the Humean-Buddhist “no-substantial-self” intuition and the intuition of relationality, with its attendant minimalist “sense of self”—as subject-pole.[29] It does this with less metaphysical cost than self-entity and self-haecceity theories, even supposing that those theories are not entirely phenomenologically implausible and explanatorily bankrupt. Let’s remember, however, that this is meant as a phenomenological claim fundamentally: consciousness is self-manifest just as the unified totality of sensory qualities (etc.) is manifest; and their contrast is manifest too, just as the contrast between such qualities (etc.) is manifest. This phenomenological claim has an ontological significance only if we accept that consciousness is indeed how it seems to be upon reflection. A claim that I accept in this case, but one need not accept it to appreciate the phenomenological point and the virtues of this way of articulating it.