Tuesday, January 02, 2007

Another Test for Objectivity [Preliminary]

In my last post, I defined subjectivity:
An attribute of a thing is subjective when it cannot be determined that the attribute is a feature of the thing itself rather than a consequence of our mental faculties.
I then proposed a test for objectivity that requires a high-precision external comparator. I still think that test is an excellent one, but there is a simpler one.

How do I know if an attribute of a thing that I detect is really in that thing, or whether it is being subjectively painted on by my faculties? If I smell a flower, how do I know that the scent is an attribute of the flower and not just my subjective impressions of the other attributes of the flower (e.g., it's shape, mass, color, sweetness, etc.)?

[The following is inspired from a gigantic thread over at Thinking Christian.]

Suppose that I have two baskets, one containing pungent cheese, the other containing flowers. Both are covered in cloth.

If our sense of smell were merely a subjective reaction to the other attributes of cheese and flowers, then we ought not be able to sniff out the contents of each basket without knowing the other attributes of the basket contents.

Of course, we can predict which basket is which just from the scent of each basket. I can effectively hide all the other attributes of a thing but scent, and still identify the hidden object.

By this method, we find that color is objective, too. Instead of baskets, let's use boxes with frosted glass on top. Thanks to the glass, you can isolate the color from the other attributes of the object. If you know that we are dealing with a yellow moon and a green clover, you can predict which is in which box by the color seen through the frosted glass. You can test that your sense of color predicts the owner of the attribute.

However, the equivalent procedure for aesthetics is quite different. We don't sense a beautiful painting without seeing the painting first. Rather, we see the painting, and then we know whether it is beautiful to us.

The same is true for morality. Can we "smell a rat" even when an evil act is cloaked from us? Nope. We only feel moral distaste when the other attributes of the moral act are revealed to us. Only when the con man's ploy is exposed in detail do we react with disgust. But, if morality were like scent, shouldn't we sense the disgust in the surrounding "ether," then predict that a morally disgusting act was taking place?

Now, imagine what our sense of smell would be like if it behaved the same way as our morality. We would be unable to predict the contents of the baskets by smell. We would only know the smell of the contents when they were otherwise revealed (visually, by touch, by verbal description, etc).

Indeed, there's a fun parallel with classifications of moral acts and classifications of odors. Suppose again that sense of smell is like sense of morality. If instead of revealing the contents of the basket, our colleague simply tells us that the basket contains a fresh-picked flower of some unspecified species, we would immediately smell a floral aroma. Just by having been provided information about the contents of the basket! Even if our colleague has a dead rat in the basket, we would still smell a floral scent. Would we then insist that smell was objectively in the cheese and the flower (as opposed to our reaction to their other properties)? I don't think we would.

We never get to test our sense of morality by sensing moral disgust, then predicting what the act was that made us feel that way. It's always the other way around. Morality is always a reaction, subsequent to all the other facts of the case. It's not the case that we sense evil, then predict the crime. It only happens in reverse order.

To summarize:
I propose that an attribute be objective when there is symmetry between (i) my detecting a thing and predicting the corresponding attribute, and (ii) my detecting the attribute and predicting the thing that owns that attribute.
This requires that we be able to isolate the test attribute of a thing from the other attributes of that thing.


Franklin Mason said...

I think I have a handle on your distinction between, say, smell and moral disgust. But I'm unsure why you think it maps onto the distinction between subjective and objective. We can predict the object smelled from the smell itself, but this does not imply that the smell is itself objective if we mean by that something like "out there in the world independently of our perception". Smells can be subjective - qualities of our perception and not "out there" - and yet it be the case that we can reliably infer the existence of certain external objects from them if those subjective states are reliably correlated with those external objects; and presumably smells are, for in general they have quite specific external causes.

Moreover, the mere fact that the evil of some act is not something that can be somehow perceptually detected before the act is known does not seem to me to imply that evil is subjective. Why can't one hold that evil is quite literally "out there" in independence of all we think or perceive, and yet it supervenes upon certain other qualities of things, qualities that themselves are not moral? Why can't one hold that knoweldge of evil - something itself perfectly objective - is always an inference from those other qualities? Why oppose "inferred quality" with "objective quality"?

Doctor Logic said...

Hi Franklin,

We can predict the object smelled from the smell itself, but this does not imply that the smell is itself objective if we mean by that something like "out there in the world independently of our perception".

We can always theorize that the things we perceive are "out there" or "not out there," but this seems like a totally useless distinction to me.

I don't see why I ought to care if the world is an illusion if every experience I will ever have confirms the illusion as real. I think that the concept of an illusion is meaningless if there's no experience that will expose the illusion.

For example, is Jupiter "out there"? Or do we all just share the delusion that it's out there? Or maybe other people delusions too? I think the "out there" distinction is useless unless there's a test that will show which is the illusion. If in principle there's no test of which is which, then the question is a meaningless one (because there's no truth value to any answer).

All we care about is that Jupiter acts "as if it is independent of our awareness." This is a practical and meaningful distinction, and one that we can assess using the method that I'm describing.

Why can't one hold that knowledge of evil - something itself perfectly objective - is always an inference from those other qualities? Why oppose "inferred quality" with "objective quality"?

Good question. I'm writing up the answer to this in a separate posting.