In this post, I want to explore Alonzo Fyfe's analysis of value. (The appropriate section of that lengthy page can be identified by the phrase "A Robust Theory". I recommend using your browser's 'find' feature!) For the sake of simplicity, I'll concentrate on non-moral value here, leaving the extension to morality for a future post.
From my overview of Desire Utilitarianism (with some minor editing):
There is no value without a valuer; no intrinsic value in the universe - no 'goodons' and 'badons' to complement protons and electrons. Fortunately, however, there are valuers (us!), so it is possible for objects or states of affairs to be assigned subjective value.
Value is a relationship between an object (or action or state of affairs) and a specific set of desires.
There are 4 dimensions to all value statements:
(1) A class of objects (or whatever) to be evaluated.
(2) A set of desires to evaluate them against.
(3) Whether the relationship between them is direct ("pleasing") or indirect ("useful").
(4) Whether the object to be evaluated thwarts ("bad") or fulfills ("good") the desires.
As an example, the value tasty evaluates objects put in one's mouth against the desires of the individual doing the tasting. The relationship is a direct one (cf. the value nutritious would be 'indirect' here instead) and the desires in question are fulfilled - i.e. it tastes 'good'.
Beautiful is analysed similarly, except that we instead evaluate 'things seen and heard'. This seems a bit incomplete to me - I would have thought that we were not only evaluating different objects, but evaluating them in a different way. So perhaps the key difference here is really along dimension (2), rather than (1) as Alonzo originally suggested.
Alonzo's example of illness is also illustrative:
We only call something an 'illness' if there is a reason to recommend avoiding it. If we discover that there is no reason to avoid it, we cease to classify it as an illness. If you have a bunch of microbes running around inside of you that give you added strength, you're not 'ill'. Only if they make you weaker are you called 'ill.' This shows that 'illness' is a value-laden concept. Now, let's break 'illness' down into its four components.
The class of objects to be evaluated is mental or physical functioning of the person to whom illness is being ascribed. This, of course, distinguishes mental and physical illness (though the line between these two is becoming increasingly fuzzy).
The set of desires that functioning is evaluated against are those of the being to whom the illness is ascribed. I am not sick in virtue of suffering symptoms that you don't like; I am sick in virtue of feeling symptoms that I do not like. Those who might want to call homosexual desires an illness might want to take note of this.
Which relationships are relevant? Again, just like with harm, both direct and indirect relationships are relevant here. If it causes one great pain, or if it merely makes one weak or tire easily, it can qualify as an illness.
And, of course, illnesses thwart desires. That is to say, they are bad things.
That should be enough to give you the basic idea of how this 4-dimensional analysis is supposed to work. I particularly like how it sheds light on the relational aspects of value. We all know that a single object can be simultaneously 'good' for some purposes, and 'bad' for others. This is explained because various different desires can be plugged in for the second parameter. It's also interesting to see the similarities and differences between various forms of value that this formal treatment highlights.
So, I think this analysis is both powerful and flexible. Though perhaps a little vague in some respects. I would be very interested to hear what others think of it, particularly with regard to the following:
1) Are there any counter-examples? That is, are there any (real) values that cannot be analysed within this framework, such as desire-independent values?
2) Could it be improved upon (e.g. by adding another dimension, thus allowing another level of analysis)?
Also, it's not immediately clear how to reconcile this with my thesis that value-judgements are made relative to some framework of standards. I guess we would need to re-describe those standards in terms of desire fulfillment. A topic for a future post, perhaps...?