On Monday I attended a discussion of James Stazicker‘s paper Attention, Visual Knowledge, and Psychophysics. I have talked about Block’s argument before that recent experimental work on attention suggests that there is mental paint (i.e. that there is more to phenomenology than what’s in the world or in our representations of it). In this paper Stazicker wanted to offer an account of the representational contents of vision that denied any kind of illusion (like Block) but at the same time rejected Block’s argument for mental paint (Stazicker says that his view is compatible with mental paint but not with the argument that Block gives for it).

The basic idea that Stazicker wants to develop is that vision represents determinable properties rather than determinate properties. That is to say that our visual representations while we are looking at a line of (say) 5 centimeters will be something to the effect of a line that is, say, 4.5-5.25 centimeters long. If this is right then there are many different ways of veridically representing the line of 5 cm length. We could represent it, as above, or as 4.75-5.5 or many others. Stazicker wants to maintain that these two different representations will produce different phenomenologies but that each is perfectly veridical. If this is what happens in Carrassco type cases then we can still say that our belief are veridical even though we agree that attention changes our phenomenology. He also argues that we have independent reason to think that vision does deal in determinable representations stemming from considerations about the limited spatial resolution of our representations.

Block is aware of this kind of objection and responds to it in Attention and Mental Paint. As he says,

The problem with this proposal is that it if the phenomenology of perception flows from representational content, then indeterminacy in content would have to be reflected in an indeterminacy of look. But there need be no such indeterminacy.

If our experience represents something indeterminately as, say, 4.5-5.25 cm in length then we should expect the phenomenology to be indeterminate as well, but since our phenomenology isn’t this way we have some evidence that there is more to it than the indeterminate representation; there is also the phenomenological mode of presentation, that is what it is like for the subject to have that conscious experience.

Stazicker responds to this line of argument in the paper. He argues that there is no problem with saying that our phenomenology is indeterminate. He denies that saying that conscious experience is indeterminate is the same as saying that it is blurry (though blurriness does involve indeterminacy, it also involves something else, something like the phenomenology of blurriness), nor is it the same as representing a disjunct of possibilities. He rather appeals to notions of seeing things in the distance. When one sees something that is far away one’s representations are indeterminate but without being burry or fuzzy or disjointed.

During the discussion Dan Shargel brought up the issue of how we can tell if our normal conscious vision is blurry or not. He reported an experience of having his prescription updated on his glasses. Suddenly he realized that his vision had been blurry before but had not realized that it was that way before the update. Perhaps that is what conscious vision is like for us. Block argued that there is a phenomenological difference between seeing a clear image blurrily and seeing a blurry image clearly. In each case one would be tempted to say that a subject would draw the same ‘pixel array’ even though there is a distinct phenomenological difference (the difference between feeling like you see it clearly or blurrily). Block also argued that one could not cash this out merely in terms of determinable versus determinate contents. Chalmers suggested that you might be able to capture that difference representationally in the following way. In the case of seeing the blurry image clearly one has a visual experience which represents the various smudges in a very determinate way (so one has a determinate representation of the indeterminate thing itself), whereas when one see a clear image blurrily one has an indeterminate visual experience in that one represents the determinate thing in a smudged way. Block insisted that this did not meet his objection since one would end up drawing the same thing in both cases.

I have to admit that I lean towards Block view here. It does seem to me that conscious visual experience presents things as being some determinate way. So, when I look at the frame of a painting it seems to me that the frame has some determinate length even though I am unsure what that length is. One interpretation I have long been attracted to is to see higher-order thought-like states as (phenomenal) modes of presentations for first-order sensory qualities. The higher-order states may very well represent the indeterminate first-order states as being determinate. This would allow one to endorse Block’s view that experience doesn’t seem indeterminate while also taking the empirical evidence to suggest that first-order states are indeterminate.