In the previous section I argued that there are good reasons for resisting Schwitzgebel’s case for global pessimism. However, we also saw that there are domains in which our introspective access to phenomenal consciousness is rather less secure than we might have pre-theoretically assumed. In other words, we saw that there is reason to think that certain kinds of phenomenal states are introspectively elusive. In this final section I want to sketch an account of why certain types of phenomenal states are elusive and others are not.
Let me begin by distinguishing the form of phenomenal elusiveness with which I am concerned from another notion of phenomenal elusiveness that I want to set to one side. In a recent paper, Kriegel uses the label “elusive phenomenology” to describe phenomenal states “whose very essence requires the absence of introspective attention” (2013, p. 1171). Among the examples that he gives of elusive phenomenology are the phenomenal states that occur at the fringes or margins of consciousness. As Kriegel notes, such states are elusive in that any attempt to make them the object of attentive introspection would change their nature. Although Kriegel’s notion of elusiveness is closely related to the one that I employ here, the two notions are not identical. (One way of seeing that they are distinct is that Kriegel’s elusiveness is primarily a matter of the phenomenology, whereas my elusiveness is a matter of one’s introspective access to the phenomenology.) Unlike Kriegel, I am interested in a type of elusiveness that is independent of attention. Consider again visual imagery. Although particular instances of visual imagery might be elusive in Kriegel’s sense because they happen to occupy the margins of consciousness, I am interested here in the fact that visual imagery as such appears to be introspectively elusive.[14]
Why might certain types of phenomenal states be elusive in a way that other types of phenomenal states are not? Broadly speaking, there are two places in which we might look for an answer to this question. On the one hand we might appeal to intrinsic features of the phenomenal states themselves. Perhaps there is something inherent in the very nature of certain kinds of phenomenal states that renders them relatively opaque to introspective access. Another possibility is that the elusiveness of certain types of phenomenal states has nothing to do with their intrinsic nature but instead reflects the structure of our introspective capacities. Just as our perceptual system is geared toward the identification of certain kinds of environmental states rather than others, so too it is possible that our introspective system is geared towards the identification of certain kinds of phenomenal states rather than others. On this view, the fact that our introspective access to some types of phenomenology is more secure than it is to others tells us more about introspection than it tells us about phenomenal consciousness (as it were).
It is, I think, premature to speculate which of these two accounts might be the more plausible; indeed, it is possible that a full explanation of elusiveness will have to draw on both ideas. But rather than pursue that thought, I want instead to sketch one way in which the structural features of introspection might go some way towards explaining why certain types of introspective judgments are more secure than others. The account in question appeals to a distinction between two kinds of introspective judgments: scaffolded judgments and freestanding judgments (Bayne & Spener 2010). The distinction is perhaps best grasped by means of examples. Contrast an introspective judgment that is directed towards one’s visual experience of looking at a red tomato with an introspective judgment that is directed towards an experience of visual imagery involving a red tomato in front of one. In the former case, there is a perceptual judgment that one is disposed to make (“There is a red tomato in front of me”) whose content corresponds (broadly speaking) to the content of one’s introspective judgment (“I have an experience as of a red tomato in front of me”). In the latter case, however, there is no such first-order judgment that one is disposed to make whose content might correspond to the content of one’s introspective judgment. In a sense, the former judgment is “scaffolded” by a perceptual disposition in a way that the latter judgment is not.
I suggest that scaffolded judgments are typically more secure than freestanding ones precisely because they are scaffolded. At the very least, it is a striking fact that many of the most epistemically insecure introspective judgments appear to be freestanding. Further, one can tell an attractive story about why introspective scaffolding might contribute to epistemic security. In making scaffolded judgments, the subject is able to both exploit the resources that it has for making freestanding judgments and calibrate those resources by drawing on its dispositions to make first-order perceptual judgments.[15] Just as beliefs that are derived from multiple (independent) sources are typically more secure than beliefs derived from just a single source, so too scaffolded introspective judgments might typically be more secure than their freestanding brethren.