That creative history was, in fact, celebrated in a splendid recent exhibition at the Louvre, D'apres l'Antique - which featured not only a dazzling array of contemporary artists' reworkings of classical themes (including modern art photographs of Fuseli's foot, and no fewer than five of Salvador Dali's full size variations on the "Venus de Milo"), but also a well-chosen series of earlier restorations, replicas and reinventions of Greek and Roman sculpture.

The superbly illustrated catalogue makes an excellent introduction to the aims and methods of Renaissance restorers as well as to the often unexpected Nachleben of some of the most famous pieces of ancient statuary (in the last fifty years, the "Venus de Milo", for example, has been used to advertise fast cars, gas cookers, Levi jeans, mineral water and support hosiery!) Restoration, of course, still goes on. We may think we have learned to admire ancient sculpture in its fragmentary state - but only up to a point. The "Venus de Milo", tantalizingly unrestored, minus her arms, is one thing; the vast sculpture groups discovered in the 1950s in a cave near the village of Sperlonga, south of Rome, smashed into several thousand small pieces, are quite another. And inevitably the re-created masterpieces now on show in the Sperlonga museum contain as much plaster and resin as they do original marble; they are no less creative reinventions than any of the ambitious projects by Bernini or Thorvaldsen. The puzzle is not that these restorations took place (who, after all, would bother to visit Sperlonga to see a pile of marble chippings?); but rather that art historians, increasingly interested in the principles and practice of Renaissance intervention in ancient sculpture, have by and large turned a trusting blind eye to the activities of contemporary restorers - as if the "science" of restoration was now above suspicion, unlike the creative fictions of the previous regime.

The classic case of this blindness is the Vatican Museums' restored "Laocoon" group. Laocoon was that doomed priest of Troy, who, in Virgil's Aeneid, failed to persuade his countrymen of the dangers of wooden horses and "Greeks bearing gifts" and ended up throttled to death, along with his two sons, by serpents sent by his divine enemy Athena.

The flamboyant marble version in the Vatican, showing father and boys hopelessly grappling with the menacing snakes, became from the moment of its rediscovery in Rome in 1506 one of the best-known and most influential works of art ever, ancient or modern. It prompted some of the most important debates at the very origin of the modern discipline of art history (notably between Winckelmann and Lessing). It was soon established as an enduring household image across Europe and later America: for Karl Marx, it provided a symbol of the evils of capitalism; for Dickens a picture of Scrooge wrestling with his stockings; for generations of cartoonists, an instantly recognizable schema for all kinds of political trouble (Nixon strangled by his tapes, or plucky "Mac" in 1960 ensnared in "the European problem"). It remains at the top of the academic agenda, and is the subject of two recent books, each by a leading figure in classical art history: My Laocoon: Alternative claims in the interpretation of artworks by Richard Brilliant, Professor of Art and Archaeology at Columbia University, and Laocoonte: Fama e stile by Salvatore Settis, now of the Scuola Normale at Pisa, previously of the Getty Institute.

"Laocoon" appeared, in 1506, with an excellent literary pedigree. As both Brilliant and Settis emphasize, much of the statue's immediate impact came from its obvious links with the Aeneid and with the description of a statue in Pliny the Elder's encyclopaedic Natural History. Pliny had written of a "Laocoon in the palace of the Emperor Titus . . . (showing) Laocoon, his children and the wonderful clasping coils of snakes carved from a single block . . . by Hagesander, Polydorus and Athenodorus, all from Rhodes". The match with the statue that Pliny had seen and described was almost too good to be true. That said, the "Laocoon", as it had been discovered, was far from complete. It was missing not just the usual extremities and the odd bits of snake; but each of the figures had lost their right arms. A huge debate followed (orchestrated by Michelangelo and Raphael among others) as to how those arms were to be restored - and particularly the arm of the central figure of Laocoon himself: bent back or extended straight up? By the 1530s, they had agreed that it should stretch up in the air; and each of the (many) succeeding restorations kept to this model, restoring the sons to match. This became the canonical image of the sculpture.

So far, nothing out of the ordinary. But, in the twentieth century, the story took a surprising twist (carefully told in an appendix by Ludovico Rebaudo in Settis's Laocoonte). For, in 1906, Ludwig Pollak, a German archaeologist-cum dealer, wandering around a mason's yard in Rome, spotted a fragment of a bent marble arm, with bulging muscles similar in style to the "Laocoon". He presented it to the Vatican Museums where it stayed in the stores until the 1950s - when the museum authorities decided that it belonged to the original "Laocoon" after all, dismantled the statue, removed the traditional restorations and inserted the Pollak arm. There are, in fact, very strong arguments against this: the new arm does not directly join with the father's broken shoulder (a wedge of plaster has had to be inserted); it appears to be on a smaller scale and in a slightly differently coloured marble; Pollak himself believed only that it came from a statue like the "Laocoon"; not to mention the fact that the circumstances of its discovery are vague at best, at worst suspicious - no more or less believable than the stories of "serendipitous" finds of missing parts in the Renaissance.

Not surprisingly the new restoration as a whole has not caught the popular imagination. (Cartoonists wanting an instantly recognizable "Laocoon" still thrust his arm straight up.) For some reason, however, professional art historians have almost universally taken it on trust; loath, it seems, to subject the work of modern museum restorers to the same kind of hard-headed analysis as is given to that of leading Renaissance sculptors.