Thanks for the comments.

Mohan: I think that the distinction you point to is relevant, and in fact may provide some support for the position that we can individuate experiences more finely than Tye claims. Given that these visual records will have different individuation conditions from the states of consciousness, and that they are also at least in principle separable from one another (though I think this depends on whether or not the spatial relations between the objects are represented, i.e. is the computer represented as being to the right of the cup, and if so, then it may not be separable when we apply the counterfactual test), then this perhaps puts pressure on Tye's claims that experiences are not divisible into smaller parts.

Also, we might use this distinction to say that regardless of which of these (visual record or state of consciousness) we call the 'experience' (Tye will claim that it is the state of consciousness), we still need some explanation of how it is that these visual records are unified, and so the question of phenomenal unity might re-appear as there are still parts that need to be unified to form a whole.


Also, for what it's worth, I think that of the possible criteria for experience individuation I mentioned above, the counterfactual test is the most promising (and more than one of the other criteria collapse into this one). The results of this test when we apply it within a sensory modality will depende on how we think the relations between the objects of our experience are represented though: for example, if the computer is represented as being to the right of the cup, and this 'to the right' is part of the representation, then if we ask 'could I have had the very same experience of the computer, in the absence of the cup, then we might think the answer is 'no'. This may mean that within a sensory modality, the counterfactual test will still always yield the result that we have only one experience at a time. Across modalities however, it seems far more likely that the counterfactual test will tell us that we have six experiences (or an experience per sensory modality). If we can individuate experiences this finely, then the question to Tye or a representationalist of a similar stripe would be 'don't we still need to account for the unity of these six experiences?'