I think part of the disagreement over this is that the usual, simple model of qualia is too limited. That’s an easy mistake to make, given that the normal examples of qualia are things that are directly reported by sensory organs, but in practice qualia have much more to do with our brains than our eyes and ears and so on, and don’t have a 1:1 correspondence with sensory information in most cases.
Take lukeprog’s coin ellipse/circle example. It’s pretty trivial to show that the outline of a coin from most perspectives is an ellipse, but that doesn’t mean that a person’s brain is going to generate only that quale for the shape of the coin: It knows, in some sense, that the coin is circular, and reports that, either along with the ellipse quale (perhaps on a different ‘channel’), or, more likely, instead of it.
This is more obvious when considering sensory information that’s ‘read’. I find it to be very obvious in speech, for example—I actually perceive three separate ‘channels’ of information when I’m listening to speech; one for the actual sound, one for the emotional tone-of-voice content, and one for the parsed words. Importantly, the latter two ‘channels’ are also communicated as streams of qualia; they’re just not communicated as streams of sound-qualia, but as visual and conceptual qualia respectively. (Have I mentioned that I’m synesthetic? That may be relevant, especially to the emotional/visual channel.) Reading text and reading body language also seem to involve the same effect; I rarely-to-never consciously notice the shapes of words that I’m reading, but mostly just perceive a stream of concept-qualia, and fluently reading body language seems to involve a similar stream of emotion-qualia without, or with only secondary, awareness of the relevant details of how a person looks.
These derived qualia seem likely to vary considerably from one person to another, for obvious reasons, but it seems likely that most people use them more than they use first-order qualia, and often instead of them. Thus, it does make sense and would not necessarily be incorrect for someone to say that they perceive a coin as a circle rather than an ellipse, since it’s nontrivially likely that they are in fact perceiving a ‘circular’ quale, generated by their brain to report the coin’s shape in 3d space.
I think part of the disagreement over this is that the usual, simple model of qualia is too limited. That’s an easy mistake to make, given that the normal examples of qualia are things that are directly reported by sensory organs, but in practice qualia have much more to do with our brains than our eyes and ears and so on, and don’t have a 1:1 correspondence with sensory information in most cases.
Take lukeprog’s coin ellipse/circle example. It’s pretty trivial to show that the outline of a coin from most perspectives is an ellipse, but that doesn’t mean that a person’s brain is going to generate only that quale for the shape of the coin: It knows, in some sense, that the coin is circular, and reports that, either along with the ellipse quale (perhaps on a different ‘channel’), or, more likely, instead of it.
This is more obvious when considering sensory information that’s ‘read’. I find it to be very obvious in speech, for example—I actually perceive three separate ‘channels’ of information when I’m listening to speech; one for the actual sound, one for the emotional tone-of-voice content, and one for the parsed words. Importantly, the latter two ‘channels’ are also communicated as streams of qualia; they’re just not communicated as streams of sound-qualia, but as visual and conceptual qualia respectively. (Have I mentioned that I’m synesthetic? That may be relevant, especially to the emotional/visual channel.) Reading text and reading body language also seem to involve the same effect; I rarely-to-never consciously notice the shapes of words that I’m reading, but mostly just perceive a stream of concept-qualia, and fluently reading body language seems to involve a similar stream of emotion-qualia without, or with only secondary, awareness of the relevant details of how a person looks.
These derived qualia seem likely to vary considerably from one person to another, for obvious reasons, but it seems likely that most people use them more than they use first-order qualia, and often instead of them. Thus, it does make sense and would not necessarily be incorrect for someone to say that they perceive a coin as a circle rather than an ellipse, since it’s nontrivially likely that they are in fact perceiving a ‘circular’ quale, generated by their brain to report the coin’s shape in 3d space.