Descartes on colors and shapes

As is well known, Descartes argues that the sensation of white in our minds when we look at snow does not resemble whatever it is in the snow that produces this sensation in us. (He puts this point in different ways in different places; e.g., sometimes he says that our sensory awareness of whiteness leaves us "wholly ignorant" of what the snow is like (Principles of Philosophy, I.68).) The same holds for many other sensory qualities: the pain we feel when we put our finger in the fire does not resemble anything in the fire, the sweet scent we have of honey does not resemble anything in honey, and so on.

But what about my sensory awareness of the shape of a snowball, a fireplace, or a honey jar? In these cases, Descartes takes a different line: "We know size, shape, and so forth in quite a different way from the way in which we know colors, pains and the like" (PP, I.69). What is this difference? Descartes writes: "there are many features, such as size, shape, and number which we clearly perceive to be actually or at least possibly present in the in objects in a way exactly corresponding to our sensory perception or understanding" (PP, I.70).

So the obvious question here is: what makes our sensory perception of shape different from our sensory perception of color, so that the former but not the latter can 'correspond to' or resemble the thing represented? Descartes' argument in the final quotation above strikes me as weak. Descartes says that we clearly perceive that our sensory perceptions of shapes either (i) actually resemble or (ii) possibly resemble something in the objects themselves. Regarding (i), I strongly doubt that we can clearly and distinctly perceive anything about the relationship between the ideas in our minds and the objects outside of us -- we would need to be able to 'step outside of our minds,' as it were, to survey and compare both the contents of our minds and objects as they really are. And if we take (ii), then it at least seems possible to me that my sensory awareness of white resembles some property in the object itself. Of course, that would be a fortunate coincidence, but coincidences are not impossible. (Perhaps Descartes' notion of possibility rules out more than our modern one(s)?)

So, is there a way to save Descartes' position that our sensory perceptions of shapes can/do resemble something in the objects themselves, whereas our sensor perceptions of colors can/ do not? Perhaps the piece of wax section in Meditation 2 could be of some help here?

Update: I had forgotten that this very problem also arises, perhaps more expiciltly, in Locke's Essay: Locke says that our ideas of primary qualities (shape, mobility, solidity, extension, and number) really do "resemble" their causes in the objects that we perceive (II.viii.15). And perhaps because this claim is more front-and-center in Locke than Descartes, commentators on the Essay from Berkeley through today have had difficulty making good sense of this claim. Berkeley brings out the problem clearly: is the idea square in my mind actually square-shaped? Is my idea of motion itself moving?


Ian said...

I don't know the reasons Descartes would give for this position, but I can attempt a justification of my own:

Objects in my visual field have geometrical properties of size, shape, and (phenomenal) color. It is reasonable to suppose, as Descartes must have, that the size and shape of a distal object cause (or cause in part) it to appear of a size and shape "corresponding to" or "resembling" them. But the properties of distal objects that cause them to look, say, red and green, are not (phenomenal) redness or greenness. The property of reflecting certain frequencies of light, or whatever Descartes believed to cause the appearance of color, is simply not like the property of seeming red.

Kenny said...

I don't know how relevant this is to the argument, but one motivation at least for our belief of this sort is that size, shape, and number are available to both our visual and tactile senses, while most other features are limited to a single sense. (Or two closely related senses, like smell and taste.)

Greg Frost-Arnold said...

Ian: I'd like to know explicitly why (as you say) "it is reasonable to suppose" that mental/ internal shapes somehow resemble something in the distal objects, if we assume (with Descartes) that it is UNreasonable to suppose that mental/ internal colors resemble something in external objects. My gut feeling/ intuition is that there IS something very reasonable aobut Descartes' claim -- I just want to get explicit about what that is.

Kenny: This suggestion (which sounds like Moritz Schlick's epistemological "point-coincidence" argument) is appealling -- I can't find anything like it in Descartes, but I don't have the 10 volumes of Adam and Tannery committed to memory, so it could be there. I'm really out of my depth here, but here's my knee-jerk reaction.
(1) It seems like some sensory qualities "are available to" multiple sensory modalities: I both hear and feel loud music (tones and vibrations).
(2) Is it really the case that the 'shape' of a baseball made available through our visual system is one and the same thing as the 'shape' made available via our tactile system?

Finally, I just wanted to mention that Kant pretty clearly does not follow Descartes (and the other mechanical philosophers) on this point -- I wonder whether something like the line of thought in the original post had something to do with his position.

Ian said...

Obviously, the silhouettes of distal objects from my perspective play a role in the shapes those objects take on in my visual field. Most of the time, the shape of the silhouette is more or less identical to the shape of the object in my visual field. This is all simply true. "Reasonable to suppose" was probably not the right phrase.

The distal object and the phenomenal "representation" have at least one clear, unequivocal property in common: their shape in outline. If one thing has the same shape as another thing, it seems clear enough to me that they "resemble" one another. Regarding color, I just can't think of anything analogous. What property could distal color and phenomenal color have in common, such that the one resembles the other? (Is there even such a thing as distal color? We would have to invent it.) The properties of distal objects that make them seem, say, red don't make sense when applied to phenomenal colors. How could phenomenal redness resemble the property of absorbing light-waves of frequencies X, Y, and Z? Properties of phenomenal color and "color-producing" properties of distal objects seem to be of radically different kinds. And if the relevant properties are of radically different kinds, no one of them can hold both for the distal object and the phenomenal representation. And if no one of them holds for both, then neither one can resemble the other. That's how I would justify Descartes' claim, anyway.

Kenny said...

Ian - it seems to me that the distinction you'd like to draw between distal and proximal color is available here for shape as well. At least, this is the possibility that Greg seems to be trying to draw our attention to. Your claim that "the shape of the silhouette is more or less identical to the shape of the object in my visual field" is exactly what's in question here.

There is a way in which your point is (seems?) entirely correct - the shape of the colored patch in my field of vision is more or less identical to the shape of the silhouette. However, my perception of the shape isn't obviously the same thing as the shape of the colored patch in my visual field. (Is there further experience when I see a black box, beyond the blackness in a certain shape, which is my perception of its boxness? I don't know how to introspect on that, but admittedly it seems redundant and possibly counterproductive.) And regarding my hedge above, we also can't get an independent means of comparing the shape of a colored patch in a visual field with the shape of an object, any more than we can get an independent means of comparing the perception of the color with the actual color.

But of course, there is still some level on which Descartes seems to be right. The only explanation I can think of is that this property seems to be available to multiple sensory modalities, though Greg is right to point out that it's not obvious that it's really the same representational property in both cases.

me said...

Hi Greg, this is Frank N P. (remember me?) Just a little question: Where can i find the document RCC (the one that you've written in that extract of you dissertation about Tarski)? Thank you in advance. and All the Best

Greg Frost-Arnold said...

Hi Frank--

Unfortunately, the documents (RCC 090-16 and 102-63) are not currently publicly available. If you can read German, you can request them from the Archives of Scientific Philosophy in the University of Pittsburgh library system. Also (here comes a shameless self-promotional plug), my translation of and commentary on those notes are being published as a book from Open Court Press, hopefully appearing in early 2007. Also, highlights from the notes will appear in an article by Paolo Mancosu that should appear in History and Philosophy of Logic pretty soon.

Greg Frost-Arnold said...

Another update: we don't need to go all the way to Kant to find the worry that I raised in the original post about Descartes' view. I think Leibniz has it already in section 12 of his Discourse on Metaphysics.

"It is even possible to demonstrate that the notions of size, shape, and motion are not as distinct as is imagined, and that they contain something imaginary and relative to our perception, as do (though to a greater extent) color, heat, and other similar qualities, qualities about which one can doubt whether they are truly found in the nature of things outside ourselves."

Leibniz does draw a distinction between shape and size on the one hand, and other sensory qualities on the other -- but for Leibniz, unlike Descartes, this is a difference of degree, not of kind.

Frank said...