Reflections on the Primary-Secondary Quality Distinction, and Our Ideas of Them

In previous posts, I've offered explanations of the Primary-Secondary Quality Distinction and our Ideas of Primary and Secondary Qualities.

But now it's time to scrutinize these ideas (if you'll permit the pun!) a bit.

In particular, I want to consider:

(1) Why should believe any of this?

(2) Why is it important or significant?

Let's start with (1). To make progress on this question, I think we'd better distinguish a number of more specific questions that are mashed together in (1). Here are some of the more specific questions we might ask:

(I) Why should we believe that there are any Primary and Secondary qualities, as those qualities have been defined?

(II) Granted the definitions of Primary and Secondary qualities, why should we believe that the examples given of each kind really are instances of those kinds? In particular, why believe that color, sound, scent, etc. are Secondary qualities? Why not suppose they are Primary qualities?

(III) Why should we believe that our Ideas of Secondary qualities do not resemble any quality in objects?

There other questions we might list here, but the above is plenty for this post. Having noted question (I), I'm going to set it aside--I think it loses much of its interest once we've answered (II) and (III). I'm going to begin with question (III), since I think question (II) is easier to answer once we've answered question (III).

So, why should we suppose that our ideas of secondary qualities, such as color, sound, taste, smell and so forth do not resemble any quality in the object? I'm going to focus on the case of color for two reasons: first, it will simplify the discussion; second, it is the most controversial example (and so, if we can make the case with respect to color, we can make it for the other qualities as well).

It is important to remember that your idea of color--or, to be more precise, your idea of a specific instance of a specific color, e.g., your idea of the color of the apple (or whatever) you are seeing right now--is the color as you are currently aware of it; the quality apparently "out there" on the apple. That is your "idea of color". It isn't a very natural usage these days.

Indeed, it isn't a very natural think to think, is it? After all, isn't that color "out there" on the apple just the property red? Well, not according to advocates of the view we are discussing. According to them, the apple as it is in itself, apart from any particular way of experiencing it, isn't colored (in the way we usually think of things as being colored) at all.

One reason is simply that appeals to color, as such, do not appear to play any essential role in explaining anything about how objects interact with other objects. Sure, we appeal to color in everyday explanations--"Why did you buy that shirt?" "I liked the color"--but when we want to get technical, when we want a full explanation, appeals to color drop out of explanations in favor of talk of wavelength distributions, behavior of electrons, and neural activity. (I am not claiming that we can explain everything about color perception, by the way!) If we take Ockham's Razor seriously, we shouldn't believe in properties that play no role in our explanations, then we shouldn't believe that color is a surface property of objects (notice that I am not quite claiming that we shouldn't believe in color at all).

A second argument is this. All object-surface we perceive as colored are, ultimately, composed of atoms. But atoms, of course, are not colored. It is, furthermore, difficult to see how a colored surface could be built up out of ultimately colorless parts. So object surfaces aren't really colored. They simply appear colored because of the way they affect us (via the light they reflect--light which is not itself colored either). Strictly speaking, this argument commits the Fallacy of Composition. Still, insofar as it really is difficult to see how to make a colored surface from colorless parts, this argument creates a challenge for anyone who would maintain that color is a real property of surfaces of objects.

I suspect that most people who doubt that color is a property of objects are persuaded by some version of an argument that we can the "Argument from the Contingency of Color Experience". It goes something like this. The way we experience the color of an object is dependent upon three things: the character of the surface of that object, the character of the light reflected off of it, and the character of our perceptual systems. Change any one of these enough, and the object will appear to have a different color. (Of course, there are problem cases even here: objects that appear black; cases where there is no surface at all--the night and daytime skies; films; and colors of light sources, but let's set those aside for now.) However, it seems to be a contingent fact (a fact that might have been otherwise) that we have the perceptual systems we have, and that the light in our environment has the character it has. Moreover, had we lived in an environment with different ambient light, or had we developed different perceptual systems, we would have different beliefs about the colors of objects--and these beliefs would have seemed perfectly natural to us (as natural as our actual beliefs seem). In light of this, it can seem to quite unlikely that we just happen to (a) have the right sort of perceptual systems and (b) live in a world with just the right sort of ambient light so that the true colors of objects are revealed to us. Why should both (a) and (b) be so, given that things (it seems) could have been otherwise? Our inability to explain this can make our ways of experiencing the colors of objects seem quite contingent--a kind of historical/galactic accident--and this, in turn, can undermine our confidence that we perceive the "true colors" of objects. If you combine this line of reasoning with the arguments offered in the previous two paragraphs, skepticism about the view that objects are colored can begin to seem quite plausible.

Other arguments could be offered for the view that objects are not really colored in the way we ordinarily think they are, but I think I have said enough to convey the overall drift of the arguments.

Now let's recall the point of offering these arguments: it was to answer question (III), above. I think we can see that the arguments offered provide an answer to that question. For if we accept that objects are not colored in the ways we perceive them to be colored, then it follows that our Ideas of Secondary Qualities do not resemble any property in those objects.

Now let's turn to question (II), which is relatively easy to answer at this point. If you accept that no property in objects resembles any of our Ideas of Secondary Qualities, then I don't think there's any reason to maintain that color (sound, smell, taste, etc) is a primary quality. The only reason--so it seems to me--to think that color (etc.) is a primary quality (a property the object has "in and of itself") is to suppose that objects possess colors in just the way we experience them. It is hard to see why you would call any other property color!

Of course, you might say that we should call a property of an object a color because it is a property of the object that causes us to experience it as colored--but that is to say that the property is a secondary quality. And this is just what Locke and many others are claiming. Once you give up the view that color is a primary quality, the only way to continue to maintain that color is a property of objects at all is to treat color as a secondary quality, a quality that objects possess because they cause us to perceive them a certain way.

So much for questions (I), (II), and (III), and hence for question (1). Let us, at long last, briefly touch upon question (2): What is significant about all of this?

This post is already far too long, so I'll try to be brief. The most obvious reason this view about secondary qualities and our ideas of them is important is that, if true, it shows that our experience is, in a fairly straightforward sense, systematically misleading. Experience presents objects and events as having properties that they do not really have. (That said, it is interesting to try to think how experience could be otherwise, particularly if you accept the view that objects are not really colored.)

A somewhat less obvious consequence--though it is one that many early modern philosophers inferred--is that we are never immediately aware of objects outside the mind. The immediate objects of awareness, on this view, are something like mental images ("Ideas"). For if the qualities we are aware of--color, sound, scent, etc.--are not qualities of objects outside the mind, then they must be qualities of objects in the mind. At any rate, this is how many have reasoned.

Now, once one has inferred that the immediate objects of awareness are ideas, it is a short step to Veil of Perception Skepticism about our ability to know anything at all about the external world. For if we are never immediately aware of anything but our ideas, it can be hard to see how we could ever justify any claim about the nature--or existence--of anything other than our ideas.

Ok, enough already! Let's hear what you think. Are the arguments offered above any good? Do you have other arguments for the claims I've discussed? Objections? Counterarguments? Let's hear it!