My third and final point concerns the classification of cognitive imaginings. Cognitive imaginings are usually considered to be non-sensory in the sense of not having a sensory phenomenal character or indeed any phenomenal character at all. An example of cognitive imagination is to imagine that there is a largest prime number. Dokic and Arcangeli suggest that this orthodox classification may be misguided, since one can plausibly argue that cognitive imaginings have a certain phenomenology, namely a cognitive one (this collection, pp. 10–11). Therefore, the authors claim, we could classify them as experiential imaginings as well.
I think the idea of ascribing a certain cognitive phenomenology to cognitive imaginings is very attractive, since it acknowledges the idea of a cognitive phenomenology in general and allows us to classify all kinds of imaginings according to one single feature, which is their phenomenal character (see also section 5). However, I am unsure about the classification of cognitive imaginings as experiential imaginings. Here is why: in the beginning of the paper, the authors define one important feature of the kinds of imaginings that they consider experiential: they involve an “experiential perspective” and are (in this sense) “from the inside” (see Dokic & Arcangeli this collection, p. 3). It is not spelled out in detail how we should understand the notion of an experiential perspective but, as I interpret it, this involves at least that things are oriented “within egocentric space” (Martin 2002, p. 408), to use Martin’s expression. Martin only speaks about visual perceptual experiences, but it seems to me that one can plausibly expand this notion to all kinds of experiences: they involve an egocentric perspective. As I understand Dokic and Arcangeli, they consider this egocentric perspective to be a defining feature of the phenomenology of experiential imaginings that re-create experiences.
If cognitive imaginings are considered to be experiential imaginings, and if experiential imaginings are considered to involve an egocentric perspective, one would expect cognitive imaginings to also have this egocentric perspective. However, it seems to me that the phenomenal character of cognitive imaginings does not involve the perspective of an experience. If I imagine that the earth is flat (and according to the authors thereby re-create the belief that the earth is flat) it seems that imagining this does not involve any egocentric perspective in the sense given above. If at all, cognitive imaginings incorporate a very specific kind of perspective that is distinct from any experiential perspective. Consequently, even if cognitive imaginings have a phenomenal character, this seems quite different from the phenomenal character of experiences (given that the latter is considered to involve an experiential perspective). If the authors endorse a different notion of experiential phenomenal character and having an experiential phenomenal character is, for example, just a synonym for having a phenomenal character, then my point is not valid. However, if Dokic and Arcangeli indeed think that having an experiential phenomenal character means that an egocentric perspective is involved (as in the case of experiences), I suggest that we need to reconsider the classification of cognitive imaginings as provided here. While I find the idea that cognitive imaginings may have some kind of phenomenal character convincing, it seems less convincing to me that they have an experiential phenomenal character in the sense discussed here. Therefore, I propose that we instead classify cognitive imaginings as a different kind of imagination with a specific cognitive phenomenal character.