By “introspection” I mean an unmediated judgment that has as its intentional object a current psychological state of one’s own. Introspection can take as its object a wide variety of psychological states, but here I am concerned only with the introspection of phenomenal states—states that there is “something it is like” for the subject in question to be in. In principle one could have any number of reasons for self-ascribing a phenomenal state—for example, it is possible to self-ascribe pain on the basis of neural or behavioural evidence—but introspection involves the self-ascription of phenomenal states on the basis of seemingly “direct” contact with them.[3]
There are many aspects of consciousness with respect to which we clearly have little to no introspective access. For example, introspection is clearly not a source of information about the neural basis of consciousness or its functional role. But surely, one might think, introspection can provide trustworthy answers to such questions as, “Am I now in a conscious state with such-and-such a phenomenal character?” Roughly speaking, to regard introspection as able to reveal the phenomenal character of one’s conscious states is to have an optimistic attitude towards it. But there is more than one sense in which introspection might be said to reveal the character of consciousness, and thus more than one way to be an introspective optimist.
One way in which introspection can reveal a phenomenal state is by allowing one to discriminate it from its phenomenal neighbours. I take discrimination to be bound up with the capacity to single the state out from amongst the other experiences—e.g., thoughts, perceptual experiences, and bodily sensations—that happen to populate one’s field of consciousness. Discriminative access to an experience allows one to direct one’s attention towards it and to thus make it the potential target of demonstrative thought—“I wish that this experience would stop”. A second mode of introspective access to consciousness involves the deployment of categories. To categorize a phenomenal state is to locate it within a taxonomy of some kind. Categorical access to the experience of an itch, for example, involves recognizing it as a phenomenal state of a certain type—a state, perhaps, that has a certain intensity, bodily location, and relations to other experiences. Categorical access is a more sophisticated form of access than discriminative access. Just as it is possible to discriminate a bird from its surroundings without being able to recognize it as a bird—perhaps all one can do is bring it under the demonstrative, “that thing there in the sky”—so too it may be possible to discriminate a phenomenal state without being able to recognize it as the kind of phenomenal state it is. Mature human beings enjoy some degree of categorical and discriminative access to their phenomenal states, but many conscious creatures—non-linguistic animals and young children, for example—may enjoy only discriminative access to consciousness.[4]
With this in mind, we can distinguish two forms of introspective optimism. Moderate introspective optimism holds that being in a phenomenal state typically brings with it the capacity to discriminate that state from its phenomenal neighbours, while a more radical form of introspective optimism holds that being in a phenomenal state typically brings with it the capacity to both discriminate and accurately categorize it. By the same token, introspective pessimism can be more or less radical depending on whether its scope is restricted to categorical access (moderate) or includes both categorical and discriminative access (radical). In what follows, I use the terms “introspective optimism” and “introspective pessimism” to refer to the moderate versions of these views unless noted otherwise.
Although introspective optimism is often assumed rather than explicitly argued for, I think it is possible to discern two lines of argument for it in the literature. Neither argument is conclusive, but taken together they go some way towards justifying the widespread endorsement of introspective optimism.
The first argument is phenomenological: introspection seems to reveal itself as providing a trustworthy source of information about consciousness. In other words, the epistemic security of introspection seems to be something that is manifest in its very phenomenology. Consider Brie Gertler’s description of what it is like to attend to the experience that is generated by pinching oneself:
When I try this, I find it nearly impossible to doubt that my experience has a certain phenomenal quality—the phenomenal quality it epistemically seems to me to have, when I focus my attention on the experience. Since this is so difficult to doubt, my grasp of the phenomenal property seems not to derive from background assumptions that I could suspend: e.g., that the experience is caused by an act of pinching. It seems to derive entirely from the experience itself. If that is correct, my judgment registering the relevant aspect of how things epistemically seem to me (this phenomenal property is instantiated) is directly tied to the phenomenal reality that is its truthmaker. (2012, p. 111)
I suspect that Gertler’s comments will strike a chord with many readers—they certainly resonate with me. Introspection seems not merely to provide one with information about one’s experiences, it seems also to “say” something about the quality of that information. This point can be illuminated by contrasting introspection with other forms of access to consciousness. Suppose that you believe that you have the phenomenology associated with anger because a friend has pointed out that you are behaving angrily. In cases like this, testimony provides one with a form of access to one’s phenomenal states, but this access surely lacks the epistemic security that introspective access typically possesses—or at least seems to possess. It would be very odd to put more faith in “third-person” evidence concerning one’s own conscious states than “first-person” evidence.
Now, one might think that even if the phenomenological consideration just surveyed can explain why optimism seems so compelling, it surely can’t provide any justification for it. Appealing to introspection itself in order to establish its epistemic credentials would be as futile as attempting to pull oneself up by one’s own shoelaces. If it’s introspection itself that is in the dock, how could its own testimony exonerate it?
In considering this objection we need to distinguish two questions. One question is whether introspection makes claims about its own veracity. A second question is what to make of such claims should they exist—that is, whether to regard them as providing additional reasons for thinking that introspection is trustworthy. Beginning with the first question, it seems to me not implausible to suppose that introspection could bear witness to its own epistemic credentials. After all, perceptual experience often contains clues about its epistemic status. Vision doesn’t just provide information about the objects and properties present in our immediate environment, it also contains information about the robustness of that information. Sometimes vision presents its take on the world as having only low-grade quality, as when objects are seen as blurry and indistinct or as surrounded by haze and fog. At other times visual experience represents itself as a highly trustworthy source of information about the world, such as when one takes oneself to have a clear and unobstructed view of the objects before one. In short, it seems not implausible to suppose that vision—and perceptual experience more generally—often contains clues about its own evidential value. As far as I can see there is no reason to dismiss the possibility that what holds of visual experience might also hold true of introspection: acts of introspection might contain within themselves information about the degree to which their content ought to be trusted.
The foregoing addresses the first of the two questions identified above but not the second, for nothing in what I have said provides any reason to think that introspection is a reliable witness to its own veracity. It is one thing for introspection to represent its deliverances as trustworthy but it is another for those deliverances to be trustworthy. But this being noted, it seems to me not unreasonable to think that the claims introspection makes on its own behalf should be afforded some degree of warrant. In general, we regard perceptual testimony as innocent unless proven guilty, and even if introspection is not itself a form of perception it seems reasonable to apply that same rule here. (After all, it is not clear why we would have acquired a cognitive capacity if its deployment routinely led us astray.) The phenomenological argument certainly doesn’t provide any kind of proof for introspective optimism, but it seems to me to do more than merely explain why optimism is so attractive: it also provides it with some degree of justification.
A rather different argument for optimism takes as its point of origin the very notion of a phenomenal state. By definition, a phenomenal state is a state that there is “something that it’s like” for the subject in question to be in. Conscious creatures enjoy mental states of many kinds, but it is only phenomenal states that bring with them a subjective perspective. But—so the argument runs—if a phenomenal state is a state that there is something it is like to be in, then the subject of that state must have epistemic access to its phenomenal character. A state to which the subject had no epistemic access could not make a constitutive contribution to what it was like for that subject to be the subject that it was, and thus it could not qualify as a phenomenal state. Call this the conceptual argument.[5]
How compelling is this argument? It seems to me that a lot depends on what is implied by the notion of “epistemic access”. There is little to recommend the conceptual argument if “epistemic access” is understood in terms of categorization, for it seems fairly clear that a subject need not possess the capacity to accurately categorize its phenomenal states in order for them to contribute to its phenomenal perspective. Of necessity any phenomenal state will fall under categories of various kinds, but the nature of these categories need not be transparent to the creature experiencing it.
But suppose that we construe epistemic access in terms of categorization, rather than identification. Might the conceptual argument justify a moderate form of optimism, according to which subjects must have discriminative access to their phenomenal states? To make this clearer, suppose that it is possible for phenomenal states to occur within the modules of early vision of the kind that are concerned with determining (say) texture or colour constancy. Such phenomenal states—assuming that they are possible—would be completely inaccessible to the subject in question. The creature in question would be unable to contrast the phenomenal character of these states with the phenomenal character of any of its other experiences; it would be unable to single such states out for attention, and it would be unable to make them the objects of demonstrative thought. As such, it seems to me that it is very plausible to hold that they couldn’t be genuinely ascribed to the subject in question, but could at best be ascribed only to one of the subject’s perceptual modules. The root of this intuition, I suspect, lies with the thought that a phenomenal state to which the subject has no discriminative access couldn’t be anything “to” the subject—that in the relevant sense of the phrase there couldn’t be anything “that it’s like” for the subject to have the relevant experiences.
Although attractive, this argument is not without its problems. One challenge comes in the form of creatures that lack introspective capacities. A creature without introspective capacities might be able to use its conscious states to discriminate some features of the world from others, but it would not be able to make its conscious states themselves objects of its own discriminative activities. And yet—the objection runs—it would be implausible to hold that creatures that lack the capacity for introspective discrimination cannot have phenomenal states. Intuitively, having phenomenal states is one thing and being able to discriminate one’s phenomenal states for each other is another—and more sophisticated—thing. Thus—the argument runs—discriminative access to a phenomenal state cannot be a necessary condition for being in that state.
I certainly agree that it would be implausible to restrict phenomenal states to creatures that possess introspective capacities, but perhaps the objection can be met without making such a restriction. What we can say is that when a creature does acquire introspective capacities those capacities bring with them the ability to discriminate its phenomenal states from one another (at least under epistemically benign conditions). So, we can grant that being in a phenomenal state doesn’t require discriminative access to that state, but also hold that creatures with introspective capacities will be able to discriminate their phenomenal states from one another (again, at least when conditions are epistemically benign).
A second objection to the conceptual argument concerns states that occupy the “margins” of consciousness—such as the unnoticed hum of the refrigerator or the background phenomenology of mood experiences. It is arguable that in some cases experiences like this not only fail to fall within the scope of introspection but in fact cannot be brought within its scope, for to attend to them would be to bring them into the “centre” of consciousness and thus change their phenomenal character. Such states serve as potential counter-examples to the claim that creatures with introspective capacities must be able to discriminate their phenomenal states from one another.
In response, one might grant that even if the phenomenal states that occur in the margins of consciousness cannot be singled out for introspective attention, there is still a sense in which they can be the objects of discrimination. Not only can they be discriminated from one another, they can also be discriminated from those phenomenal states that do fall within the scope of attention. Indeed, if such states cannot be discriminated from their phenomenal neighbours in any way then it is unclear what reason we could have for thinking of them as falling within the margins of consciousness at all, rather than being completely unconscious.
Where do these considerations leave us? I have suggested that the phenomenological argument provides some reason to take at least a moderate form of optimism seriously. It doesn’t, of course, establish that our access to all kinds of phenomenal states is robust—indeed, one might even appeal to phenomenological considerations to motivate the idea that our epistemic access to significant regions of phenomenal space is very poor. (I return to this topic shortly.) The conceptual argument provides little reason to think that we will always be able to categorize our phenomenal states, but it does provide some motivation for the idea that being in a phenomenal state brings with it the ability to discriminate that phenomenal state, at least when it comes to creatures with introspective capacities. In short, optimism of at least a moderate form is not merely a holdover from Cartesianism but can be provided with some degree of support. With these considerations in mind let us turn now to the case for pessimism.