Having noted the initial plausibility of the general outline of Schooler’s account, I pointed out some problems and expressed some general reservations about its scope. First, I argued that the postulation of a third kind of conscious but not accessed or reflected state is not justified. As a result, the account is too narrow, because one of the underlying general assumptions is not justified. This assumption causes a number of problems and a few misunderstandings. However, the assumption seems conceptually independent of the main project, which is to allow us to bridge the gap between first- and third-person criteria for consciousness. I suggested that the main distinction is underspecified and needs further clarifications of the elements involved: access, reportability, and levels of awareness.
Second, although it is tempting to attribute a first-order account to Schooler, a more convincing alliance would actually be certain functional accounts, especially higher-order accounts and global workspace accounts. And I argued that we should replace the introduced dichotomy by a finer-grained distinction of different kinds of meta-cognitive processes and meta-reflections in several dimensions.
In discussing support for the underlying conceptual framework, I then argued that the evidence offered is actually about complex cases. As exciting as the empirical results are, they seem not to be about individual states, but rather about the connection between many states or even the stream of consciousness. The project is about creature consciousness, not state consciousness—though the initial distinction suggests otherwise. This is the first result of my commentary.
I would suggest giving up the idea that the account offers a new meta-perspective, which for Schooler is a preferable alternative to reductive physicalist accounts. I do not think there is a need for this ontological element in his account, and it does not seem to fit with the rest of the methodological project. In addition, the claim seems independent of the rest of the project and there are reductive accounts available that fit very nicely with his project. This is the second result.
In essence I suggested a few conclusions and recommendations, mostly based on conceptual considerations, which clarify and strengthen the main project, with which I sympathize.
We keep many main insights of the paper:
The account is still be a cognitive account, and we allow that cognitive factors help to get a grasp on consciousness; the project is still to bridge the gap between the first- and third-person perspective.
We also keep the insight that further processing and certain kinds of further processes might either change the state itself and/or the state’s content. But we acknowledge that we need to consider the embeddedness of the state to determine the experience. In other words, we focus on processes and phenomena, instead of individual states. This allows Schooler to associate his project with either a hybrid account or a version of a functional account, more specifically a workspace account or higher-order account. Which in turn helps to specify the dimensions of meta-processing in more detail and get a better grasp of the necessary conceptual clarifications. Nonetheless, we still see meta-awareness and consciousness as distinct phenomena. I take this to be the driving idea in his initial distinction.
The proposed list of potential criteria is still extremely useful, since it helps to determine these very reflective dimensions and factors, which determine both experience and the activities of the mind. For example, the behavioral criteria[24] will be caused by these very meta-processes, which we try to identify in more detail.
Finally, we keep the insight that factors accessible through the third-person perspective can give us insight into what is going on in the mind, as well as in conscious processes.
The remaining task, then, is to specify the aspects and dimensions that are relevant, and the kinds of meta-processes, access, or reflection in question. I suggested building blocks for an improved taxonomy of different kinds of reflections and “taking stock”. I suggested that awareness itself might come in degrees and at different levels of representation. By distinguishing dissociable levels of access differentiated at hierarchical representational levels, we allow for partial awareness. In effect, this allows for a fine-grained perspective on conscious experience. Instead of just unconscious, conscious, and a meta-reflective level of awareness, we have different dimensions of experience. And this is done at the sub-personal level by a specification of the term “access”. But we should restrain from simply postulating a third category, namely a state that is unaccessed (or un-accessible) but conscious, thereby avoiding the problems associated with the postulation of this third category. The resulting finer-grained taxonomy allows an improved understanding of how exactly meta-awareness and conscious experience differ. Of course there is a price to pay if we accept this change of focus. While we can still claim that the criteria give an insight into what is going on in the mind, “the mind” includes unconscious states, conscious states, and several levels of re-representational processes.
There are a number of advantages of a view like this. First, it is not in conflict with some of the most promising candidates for philosophical theories of consciousness. Moreover, one can still account for the similarity of an unconscious state and its conscious counterpart. And third, one can keep the initial idea behind Schooler’s distinction between the experienced state and a meta-reflective level of awareness of “knowing that one is in this state”, but would substitute it with a finer-grained conceptual framework of multiple differences among several dimensions.