The value-set is the set of things that "feel good" to the
entity. To be more precise: imagine a list pairing off
various experiences with their qualitative feel: painful,
pleasurable, etc. Just which qualitative feels should be
regarded as assignments of value is an interesting question.
Maybe "pleasure" and "pain" between them cover everything,
maybe not. I do think there are some "qualitative feels" which
aren't assignments of values - for example, perceptions of color.
Even if you fear blue, perceiving something as blue is not
an assignment of value - fearing the thing because it's blue is.
The value-system is the set of all those principles for *acting*
(or all those dispositions to act, in the case of an unconscious
value-system), which will help determine whether the entity
has "good" or "bad" experiences. For example, "Act so as to feel
good". Another interesting question: does any entity ever make
a choice that's not a product of its value-system? Or is there
a logical identity between "choice" and "act determined by a
value-system"?
What does it mean for something to "have value"? The only
answers I can think of are (i) it appears in a value-set
(as defined above), or (ii) it plays some role in a value-system
(for example, helps bring about situations in the value-set).
So if "being good" is a form of "having value", I don't see
any way to give it meaning except by referring to *some*
entity's experiences or preferences. In particular, I
don't see how one can judge one value-system to be better
than another, unless one has a value-system to judge by.
As far as I can see, this is completely non-paradoxical,
but it does imply that there are no facts of the form
"Morality A is better than morality B", only facts of the
form "Morality A is better than morality B, according to
morality C".
-mitch
http://www.thehub.com.au/~mitch