Absolute Color
My red might (kind of, sort of) be your red
A classic children’s playground conversation topic, could my red be your blue? Yuxi (https://yuxi.ml/essays/posts/hole-argument-inverted-qualia/) proposes a solution to this that I largely agree with. The core is that what a color “is” is exactly and only defined by the relational structure of qualia space.
The idea then is that there is no extra data to say that a particular point on this relational structure “is” inherently red, so the question is ill-posed. Still, we want to ask the reasonable version of this question. Is there any way we can talk about particular colors as identifiable points on the relational structure of color space, using just the structure of the space? Yes! For instance, at high lightness, the maximally saturated hue tends to be yellowish. The classic inversion question assumes you can separate the identity of a color from its functional role, but this isn’t the case. Yellow is one such example of this, and there are more for the rest of color space.
In general, color space has curvature that is not uniform, so you can use the geometry of the space to “landmark” particular colors. Other properties of the space can be used as well; perceived distances show diminishing returns, so large perceptual differences may be smaller than the sum of many small steps along the same path. This too can be used to talk about “what color is what” using the structure of color space.
Furthermore, if the phenomenology of color is but the metric geometry of color space, then we have not only dispelled the inverted qualia problem, but also have dispelled the problem of other minds. We would be able to scientifically answer Nagel’s question of what it is like to be a bat. The hard problem of being a bat dissolves, leaving the easy problem. It simply requires us to construct the geometry of a bat’s color space from its neurophysiology.
I quite like this! I have for a long time held that a sufficiently good description of what it’s like to be a bat would allow you, with some work, to actually experience that. Similarly, I think that a blind person should be able to create color experience with a good enough description of how to do so, plus a heaping helping of effort. I personally think that sufficiently good meditation instructions are probably enough for a very good approximation, but perhaps technological intervention would be required if the hardware differences are big enough and insurmountable. Remains to be seen.
As for whether our reds are basically the same, yes, given all that. If what a color *is* is exactly its relational role with the rest of experience, we just have to determine what the relational role is and if we have similar ones. The geometry of color space alone works, but in principle we can even use other things. If the human color space was instead spherical, we would have no way *in color space alone* to say that some point on it is red. If color space alone underdetermined color identities, we could still use sensorimotor/semantic couplings to supply the required structure.
It is probably the case that the color red evokes particular tactile anticipations and sensations as a result of evolutionary tuning of our brains for this. The relational roles required to say “this point is red” could be done by the associations between the color space and any other qualia space. We probably have more than enough built-in responses and associations to colors that we can anchor color space in this way. That is to say, we kind of have the same red and can identify it as the same because it results in similar tactile/semantic/etc associations.
Even if color space was symmetric on its own, we could still use the couplings to the rest of experience to identify red. Really, the couplings to the rest of experience given some color *would definitionally be red*.
If you look down at your body and see some color, especially flowing or pooling, especially with pain, this probably warrants a great deal of attention and concern. This pattern of association is almost certainly *in the hardware*, and we call that color red. Red just isn’t a thing-in-itself, it is a pattern of relationship with the rest of qualia space. And importantly, in this setup, we have a relatively inherent red. If standard human hardware allows independent people to derive the same relational role (red), great! That’s fairly inherent, even if many of the associations are learned in lifetime.
We can even get identification of colors from how they interact with attention under various conditions. “Red” is the color that draws your attention in the above situation! Because of this, you can’t even “just attend to pure red”, as the coupling between attention and red is part of what red is. In general, how readily and in what ways a color couples to attention is part of the relational role that constitutes that color. There is no way to have “pure attention” perceive a “pure color”, as it presupposes the possibility of an unworkable separation.