[Aiwendil]

Looks like it's going to be "The Nature of Art: Aiwendil vs. Kalessin, part III".

Anyway:

I am not so sure that a sophisticated model that predicts what will be pleasing or not is a good thing.

Surely the question is not whether it is a good thing; it is whether it can exist.

It sounds to me as though the record company releasing music by Britney Spears is obviously in possession of such a thing.

This is completely different. The model of "art" that they possess does not predict what will be good art; it predicts what will be popular. There is no reason to assume that what is popular is good, and I have never done so.

And I still haven't quite reconciled the semantics behind "I like something therefore it is good".

I've never said "I like something therefore it is good"; in fact, I've said rather the opposite. To make an analogy: suppose I measure the mass of an object on a balance. If I said "The balance reads 5 grams, therefore this thing has a rest mass of 5 grams," I would be reversing the causality. Properly, I should say "The object has a rest mass of 5 grams, therefore the balance reads 5 grams." In this simple case, there is really no distinction between "reading, therefore mass" and "mass, therefore reading". But suppose the balance were incorrectly calibrated, or suppose it was made in an environment with a different gravitational acceleration. Then the five gram object might produce a reading of, say ten grams. The object has not changed; whether we think its mass is 5 or 10 grams does not affect its actual nature. Suppose two people perform the measurement, one with a correctly calibrated balance and one with an incorrectly calibrated one. They get results of 5 grams and 10 grams, respectively. It is to be expected that a contention arises between them, and each argues his or her position. They might invoke other, cruder methods of measurement, real or fictitious: "Look, it sinks in water, so it must be heavy"; "It looks just like another 5 gram object"; "It's square, and square things are always 10 grams." They might eventually agree that they simply have different subjective opinions about mass, or that a real determination is impossible. However, nothing changes the fact that the object is 5 grams; certainly the fact that one person (honestly) believes it to be 10 grams does not change its mass.

Now (obviously), the object = art; the mass = the quality of the art; the observers = us; the scales = our innate evaluation of what makes good art; the appeal to cruder measurements = hypotheses about popularity, comparison with other works of art, etc.; the poor calibration of the scale = factors such as reputation, accessibility, and familiarity that interfere with our ability to assess the aesthetic value of art.

That's a lengthy analogy, but I think it demonstrates pretty well my views on art. So, just because a person thinks a work of art is good (or weighs ten grams), that doesn't mean that it must be good. That person's assessment may be inaccurate for any of various reasons (i.e., the person's scale is poorly calibrated).

Sorry for the length of that analogy. And so far, I've only tackled your first three sentences!

Your example of perfect cadence illustrates my point. Traditional music theory perhaps, but only in relation to the diatonic tradition. If you factor in musical theory appropriate to classical Indian raga, or African pentatonic composition, or Balinese gamelan, and so on ... which is more valid?

This is what I was getting at when I said that, for example, an imperfect cadence is justified in certain circumstances. A complete theory of art would be able to explain under what circumstances an imperfect cadence will sound good. Similarly, a complete theory of art would take into account Indian ragas, African pentatonic scales, and so forth. Diatonic theory is certainly not a complete theory of music. But it is a succesful theory, within limits. To make another (briefer) analogy, Newtonian mechanics is not a complete theory; in very high-velocity cases it is wrong. A more generalized theory, though (like general relativity) encompasses both the low-velocity and high-velocity scenarios. Thus, a complete theory of music would encompass all music that is aesthetically pleasing. Note that I do not think a complete theory could ever be formulated; music (or any art) is far too complex and the theory would be impossibly complicated.

The terms in which we can describe and understand are indeed universal ... but our experience is not.

This actually makes some sense to me. However: don't the terms in which we describe art arise from our experience? If our diverse experiences are not at least very similar, how could a common understanding of form, and specifically, of good form arise?

This may sound like a reduction to subjectivity, but that is not my intention.

Yet it certainly does in some sense lead to subjectivity. This only works if you accept (which I think you do) the disconnection of quality from experience. But this inevitably leads to the conclusion that the quality of art has nothing to do with how deeply it moves a person, or what that person's subjective reaction to it is; it depends only on some objective standard that cannot relate to experience (because experience is subjective).

The line of absolute external objectivity in art is one where you end up painting yourself into a corner, and saying "well, because this (let's say Britney) meets the precise set of laws required to be good, I am compelled to like it.

No. If you don't think something is good, but a certain formulation of the "laws of art" say that it is good, then either: 1. the formulation of the laws is incorrect or incomplete, or 2. there are interfering factors that prevent you from liking the work. Just to show how 2 might work: suppose I'm miserably ill the first time I hear the album Kind of Blue. My subsequent dislike for the album may have little or nothing to do with my liking/appreciation of its aesthetic qualities; I may dislike it for non-artistic reasons. (Note: I don't really dislike that album, in fact it's among my favorites).

If you prefer something else (say, Wagner) you are wrong, because only I truly understand the laws and the proof of that is that I do not prefer Wagner to Britney".

Your example seems to be aimed less at the idea of "laws of music" than at the irrational behavior of a hypothetical person. First of all, the laws of music would never say that Britney is superior to Wagner. Second, your hypothetical person is using circular logic: A is true because A predicts B and B is true because B predicts A. That person's formulation of the laws seems to be arbitrary. Third, the person assumes that the laws are fully known, which they can never be.