Suppose a copy is made of your favourite painting, say Vermeer’s Girl with a Pearl Earring, that is so good experts cannot, with the naked eye, tell it apart from the original. Would looking at the copy give you the same satisfaction as looking at the original? (If you say yes, I believe—forgive the chutzpah—that you are wrong about yourself). Now the reason we would still prefer to look at the original is that our appreciation of a painting is affected not just by its causally efficacious properties—how it reflects light to our retina in particular—but also by our knowledge of some of its causally inert properties, in particular its history (knowledge of causally inert properties is itself causally potent, of course). We find ourselves located in different causal chains when we look at the original and at the copy, and not only do we care, we care in a way that is part of our aesthetic appreciation. The same is true, if anything to a greater degree, when we compare looking at a genuine Vermeer or at a forgery such as Van Megeeren's false Vermeer Lady and Gentleman at the Spinet, even if taken by the greatest experts at the time as genuine and praised.

Now to music. There is here a greater variety of possible causal chains: compare listening to a Beatles' melody sung live by them, listening to a record of their performance, listening to just the melody of their song performed by others, listening to the performance of a melody that, in a sense, they could have written (and it would have been a good one) but they did not, a computer did. In the last case, we are not provided with any satisfactory historical relationship to the Beatles, however pleasant the listening otherwise.

Conclusion pro tem – If a human being can produce it, so can, in principle, another kind of machine. But the output is historically different and it matters to our aesthetic appreciation. Is this objectionable, a kind of snobbery we should overcome? I don’t think so. But cases as that evoked by Noga do certainly challenge most common views of what is a work of art, be it a painting or a song. And this reminds me of a web conference on art and cognition at interdisciplines.org organised in 2002-2003 by Noga and by Gloria Origgi where similar issues came up (and archived on line here)

Before Jason Brown, there has been many attempts at mimicking the "Beatles sound". The irony is that the worse ersatz ever produced came from the Beatles themselves. Free as a bird (1995) resurrects an old demo by Lennon, with lyrics completed by McCartney and arrangements by all 3 remaining beatles. The sound was specially reengineered to produce some kind of genuine "Abbey Road" feeling, and the videoclip is full of historical allusions to the band. The song is not unpleasant, but it walks a very thin line between the authentic and the tacky, and falls on the wrong side.

http://www.youtube.com/watch?v=0D196-oXw2k

Is "Free as a bird" a genuine Beatles' song? The historical relation between the Beatles and the song is close to satisfying. After all, we accept as genuine many paintings that got the finishing touch from assistants, as well as paintings from groups with a variable composition, like Monsu Desideriu. The problem with "Free as a bird", I think, is that it is so self-consciously a Beatles song. The heavily marketed historical relation between the Beatles and the song ruined it: it is too genuine to be true.