Summer 2004

The Copy and Its Evil Twin

Thirteen notes on forgery

Alexander Nagel

Icon of the Virgin Eleousa, Venice, mid-fourteenth century.

You would expect that the crime of art forgery has been around as long as people have been making artifacts. But in fact it is very difficult to find an art forgery in the West before the fifteenth century—in contrast to China, where, as with paper and spaghetti, they were doing it already a millennium earlier. There have been cases of textual forgery ever since documents carried legal authority, and forged coins appear pretty much as soon as coinage was invented, in the early seventh century BCE. But art forgeries appeared only with the advent of the idea that works of art are to be appreciated primarily as singularities, as unrepeatable performances by an author.

Works of art were not always authored performances. The work of art as we see it enshrined in modern museums is the historical aberration. Some of the more interesting experiments in twentieth-century art have made it easier to see this. Media like photography or film that involve mechanical replication suspend the notion of the authentic and unique work of art, as Walter Benjamin saw: it makes no sense to ask for the authentic print from a photographic negative. And then there is the factory-produced readymade, the media-saturated images of Pop Art, the serial applications of Minimalism and Conceptual Art. The digital universe has, finally, introduced the idea of images with no physical existence at all, where each virtual instantiation presents exactly the same information and there is no degradation from one reproduction to the next. 

Before the advent of the cult of art—before the museum, the picture gallery, the connoisseur, the art dealer, and the art forgery—images naturally took the form of copies. The sixteenth-century Neoplatonist, alchemist, and mnemnotechnician Giulio Camillo once described images as just one of the “phases” that a body can take: there is the form of the physical body, which could then be translated into subtler form in a painting (Camillo mentions the portrait made of him by Titian), and to still subtler form in the reflection of a mirror. The idea was an old one, and it fundamentally affected the way images were understood. Images were not merely representations but were understood as translations of the physical form of a person into another medium. By transferring the lineaments of form they actually transmitted something of the essence of the person they depicted. Further translations of forms from image to image were simply an extension of the process.

The Cambrai Madonna (Notre-Dame du Grâce), Italo-Byzantine, ca. 1340. Believed at the time to be painted by St. Luke himself.

The entire tradition of the Byzantine icon was based on the idea that a copy of an image, if done accurately, was another translation of the prototype-form. Portrait icons were thought to have their basis in a miraculous moment of truth when, say, St. Luke painted the portrait of the Virgin Mary from life. Later images were presented as copies of that original portrait, and so were translations of the essential form of the Virgin. This meant that images led a strange and unstable chronological life. An icon made yesterday, if it legitimately transmitted the original image, was treated as an antiquity. The Byzantine prelate Nikephorus, in the ninth century, said this about an icon produced in his own day: “This image of Christ is not a new invention. The picture has the authority of time: it is coeval with the proclamation of the Gospels.” 

The philosopher Nelson Goodman once made a distinction between autographic works, like paintings, that exist as unique products, and “allographic” works, like musical compositions, that exist in multiple applications. Two different performances of Beethoven’s Fifth Symphony are both equally instances of Beethoven’s Fifth Symphony. Texts are also allographic in a basic sense: a Penguin edition of Jane Austen’s Sense and Sensibility offers the same novel as a luxury edition, as long as it follows the same philologically established text. We point at the paperback edition and say, “This novel was written around 1810.” If we put Goodman’s distinction through some historical adaptations, we find that in premodern times images functioned quasi-allographically. They were iterations of a prototype image, and so belonged as much to the time of the content they depicted as to the time of their making. 

In Roman antiquity, famous Greek statues existed in innumerable replicas, which were regarded as legitimate substitutes for the original. The second-century Roman writer Lucian regularly described copies as “Myron’s work” or “Polycleitus’s work.” The work was understood to reside in the basic form and conception, and these could be copied by an able craftsman. Sometimes the copyist would sign his own name on the sculpture, but there is no need to call this fraud; he was simply claiming credit for the technical execution of this work. In his treatise On the Sublime, the third-century writer Longinus was careful to distinguish between literary imitation and plagiarism, and to characterize “good” imitation he compared it to taking casts from beautiful statues, “as it is acceptable to do.” Whereas exact copying of texts was stealing, making an exact replica of a statue was good cultural transmission. Literary authorship and artistic authorship were not equivalent cultural values.

Virgin and Child by Hayne of Brussels, ca. 1455. Flemish interpretation/copy of Cambrai Madonna.

When images inhabit a copy culture, there is no room for forgery. Without a cult of the originally produced work, appreciated as a singular and unrepeatable performance—without a conception of the work as an event—forgery has no function. This is why it is so difficult to find documented cases of art forgery in antiquity, and why the few cases that have been brought forward are inconclusive: in each case what is alleged a forgery can reasonably be explained as a copy. But ultimately we are not talking about this or that case; the question is a systemic one. Either forgery is widespread or it isn’t, either it is a structural feature of a world of art or it isn’t, and in antiquity and the Middle Ages it was not. The Romans were the great lawmakers: they had laws about the forgery of documents and about the forgery of currency, but they had no laws about the forgery of works of art. 

In 1440, a Flemish prelate returned from Rome to Cambrai with a small painting of the Virgin and Child on gold ground. The exotic panel was installed in Cambrai cathedral and soon became famous in its new setting: touted as a portrait of the Virgin painted by St. Luke, it quickly began attracting thousands of pilgrims. In fact, it was painted only about one hundred years earlier in a Byzantine manner by an Italian, probably Sienese, painter. It is itself a copy, an adaptation of a type long held to be an invention of St. Luke, the Virgin of Tenderness, or Virgin Eleousa. And it in turn generated copies: fifteen copies by Petrus Christus and Hayne of Brussels were ordered in the 1450s alone. Like the Cambrai Madonna itself, these copies are not exact: there is no effort to avoid a normal fifteenth-century painting style in the bodies and facial features, and there are even significant changes, such as in the gaze of the Virgin. Other paintings, such as one by Rogier van der Weyden, show even freer adaptations of the model.

In the copying of a text, a misspelled word, or a skipped word, is an alarm signal that the scribe-copyist has made a mistake. In “continuous” media like painting or sculpture there is no way to tell that the handmade copy has veered away from the model. For this reason, the protocols that governed image-substitution were generally tolerant of a certain amount of drift between model and copy. What mattered above all was the transmission of programmatic, essential content. That is why premodern copies are so often far from exact. Only a general typological resemblance links the various copies of Praxiteles’ Aphrodite of Cnidos. It is one of the most famous statues of antiquity, and yet the copies vary so widely that we can only guess at what the lost original looked like. 

Around 1500, the art world begins to bristle with stories of frauds, scandals, and reprisals involving fakes. But this is not to say that forgeries displaced traditional modes of copying overnight. Forgeries came into being fitfully, in certain places and at certain times. A work made as a copy might be passed off as a forgery in another context. The conspiring conditions included newly assertive artist-authors, a new breed of collector and connoisseur, an emerging art market, and a new class of intermediaries later known as dealers. In 1496, Michelangelo makes a Sleeping Cupid in the antique manner. He is told by a proto-dealer to “distress” it so as to sell it as an antique at a higher price. The rich Cardinal who buys it—for about the price of a house!—is incensed when the work is exposed and gets rid of it. It is almost immediately snapped up by one of the great art collectors of the day, Isabella d’Este, and now displayed as a famous Michelangelo forgery. In fact, she sets it alongside a (copy of a) famous Sleeping Cupid by the ancient Greek sculptor Praxiteles, and so stages a mini-exhibition on the subject of copies and forgeries, modernity, and antiquity.

Virgin and Child by Rogier van der Weyden, ca. 1455–1460. Thought to be another copy of the Cambrai Madonna.

Traditional conceptions of the copy don’t just go away. When Reginald Pole, a friend of Michelangelo’s, is asked for his copy of a drawing of a Pietà by the master, he replies that he would be happy to give it away, as he can get another copy from a friend. To this day, scholars debate the attribution of several highly finished Michelangelo drawings that exist in copies. When Isabella d’Este is asked by an aristocratic friend for her painting of Mary Magdalene, she replies that she would be happy to send it, but only asks for time to have a good copy made. Even today, companies dealing in “genuine fakes”—expertly painted copies of famous paintings, sold as copies—do an enormous business. A well-executed copy evidently still carries something of the magic and force of the original. Recently a copy of a Van Gogh was sold as a copy for $200,000. If forgeries are the form copies take in the era of art, this is the form copies take in the era of forgery.

The emergence of art forgery presupposes a culture in which what matters above all is not the content a work of art transmits but the irreducible qualities that make this work an unrepeatable event. Eventually this conception of art would form the basis of a discipline called the History of Art, which devoted its energies to putting each artistic performance on a timeline, and to studying it as the product of an author and a historical moment. When the nineteenth-century priest and art collector Alexander Schnütgen was on his deathbed, one of his crucifixes was brought to him to kiss. He opened his eyes, looked up at it for a moment, and said: “Thirteenth century.” When it came to art, the connoisseur had eclipsed the priest. Always the mimic, forgery conforms to the new criteria of value. No longer content to reproduce only schematic features, it now aims to render the kinds of details that would satisfy a connoisseur. A seventeenth-century copy of the Cambrai Madonna shows the new criteria at work. It is a copy in the era of forgery: exact enough to confuse the critics as to its date until scientific tests were done on it in recent times.

The pairing forgery/copy may be more interesting than the pairing copy/original. Forgery is not merely the criminalized version of what had been in earlier times a legitimate replica. A collateral effect of a system in which performativity is all, the forgery is the copy in metastasized form. It crawls over the surface of art, imitating with obsessive care the appearance of the original. Ultimately, of course, in serving the cult of the authored artifact the forgery aims to subvert it: it is out to prove that an artifact can escape its historical moment, and its author. It claims that the singular can be repeated. The threat of forgery intensifies the pleasurable rituals of art—close looking, the making of fine distinctions—to the point where they turn into paranoid surveillance. Forgery is the harassing bad conscience of the cult of art, shadowing our obsession with originals and mocking our fetishism of the art object. 

Alexander Nagel teaches and writes about Renaissance and contemporary art and divides his time between Toronto, New York, and Rome. From 2004 to 2006 he will be Mellon Professor at the National Gallery in Washington, where his aim is to become a student again.

If you’ve enjoyed the free articles that we offer on our site, please consider subscribing to our nonprofit magazine. You get twelve online issues and unlimited access to all our archives.