1. There is an important argument which can be traced back to Kant's second and third Critiques, and which has been defended by a number of distinguished modern philosophers.1 It goes as follows. Moral judgments are universalizable; that is, I am logically committed to making the same moral judgment about all relevantly similar cases. If I refuse to make the same moral judgment about two relevantly similar cases, then either I believe that they are relevantly different, or I have changed my moral views between examining the first case and examining the second, or I am simply irrational and not a proper subject for discussion. In contrast, however, aesthetic judgments are not universalizable; works of art are necessarily unique. If I say that a painting is aesthetically pleasing or successful or important or striking or whatever, I am not committed to making the same judgment about any relevantly similar work. Occasionally no doubt I might make the same judgment about a relevantly similar work, but I am in no way logically committed to doing so. Indeed in certain cases—the cases which are the topic of this paper—I am logically committed to making an entirely different aesthetic judgment of a relevantly similar work. Since works of art are necessarily unique, copies, fakes, forgeries, pastiches and ‘works in the style of …’, however plausible, however skilful, however close to the original, can never have the same aesthetic merit. Even if my Athena print of a Canaletto were qualitatively identical with the original, it could not have the same aesthetic merit. Even if a modern symphony reproduced perfectly the style of Mozart and even if it were in general of comparable musical quality, it could not have the same aesthetic merit as (most of) the forty-one originals.