The cultural intersection where Henry Walters, Bernard Berenson, and Marcello Massarenti met at the beginning of the 1900s was paved with endless copies of Italian old-master paintings. From one generation of artists to another extending from the fifteenth to the nineteenth centuries, the imitation of old-master paintings became an inseparable part of the fabric of Italy’s cultural life. Painters at all levels participated in the practice. Italian masters often made signature copies of their own paintings. Devoted apprentices employed by a master in his workshop and other followers dutifully copied his paintings. Other painters unassociated with a master made additional copies both to pay homage to him and to identify themselves with established standards of greatness. Students, as a means of learning their trade, endlessly made copy after copy. And to all of the copies made for legitimate reasons were added the work of the forgers whose motivation was purely to deceive a never-ending flow of eager purchasers. Over centuries the copies, like a house of cards, were layered one upon the other to the point of their inevitable collapse. It was this phenomenon that led inexorably to the hundreds of copies attributed to Italian old masters that were scattered throughout the Massarenti collection, and it was the challenge presented by these copies that brought Walters and Berenson together.
In the tradition of the ancient Roman practice of repetitively copying the statuary of earlier Greek masters such as Praxiteles and Polyclitus, the idea of copying venerated works of art had become both widespread and ethically acceptable during the Renaissance and continued for hundreds of years thereafter. A faithful copy of a masterpiece was nothing to frown upon. To the contrary, sometimes a skillfully made copy not only was considered to be as valuable as the original but also, paradoxically, was viewed as having attained a higher level of perfection than the original itself. This point is illustrated by a legendary story recorded in 1568 in the second edition of Vasari’s The Lives of Artists. According to Vasari, the Duke of Mantua, while visiting Florence early in the sixteenth century, saw a painting of Pope Leo X by Raphael in the Medici palace. He asked Pope Clement to obtain the painting for him, a request that was conveyed to Ottaviano de’ Medici, a member of the family who served as the painting’s custodian. Unwilling to part with the Raphael, Ottaviano summoned Andrea del Sarto, a painter whose prodigious skills almost equaled Raphael’s, and commissioned him to copy Raphael’s painting. This copy was sent to the Duke of Mantua, who, unaware of the deception, was fully satisfied with his “Raphael.” After learning about this, Vasari confidentially informed the painter Giulio Romano, one of Raphael’s most gifted disciples, that the painting in Mantua was by Andrea del Sarto, not Raphael. Although surprised, Romano replied that it made no difference to him because “I value it [the Andrea del Sarto painting] even more than if it was by Raphael, for it is extraordinary that one great master should so exactly imitate the style of another.”1
Another instructive story about the Renaissance art of imitation involves Baccio Bandinelli, a sixteenth-century sculptor whose prominence rested on his skill at replicating classical sculpture. As retold by Leonard Barkan, in his Unearthing the Past, certain ambassadors of the French King Francis I, while visiting two cardinals at the Vatican, saw the Laocoön, probably the most venerated ancient sculpture in Rome, and suggested that it would make a wonderful gift for their king. One of the cardinals, Giulio de’ Medici, replied, “There shall be sent to his majesty either this one [the Roman original] or one so like it that there shall be no difference.” Bandinelli was then summoned and asked whether he could carve a statue that was equal to the original. He replied memorably that “he could make one not merely equal to it, but even surpassing it in perfection.”2 After Bandinelli completed his replica, Guilio adored it so much that he sent it to his home in Florence instead of delivering it to France. While the ancient Roman version of the Laocoön has remained firmly ensconced in the Vatican, Bandinelli’s marvelous copy has for generations been on display at the Uffizi, where its adoration has not been diminished by the absence of originality.
The idea that a copy could be as desirable as the original work of art was at the heart of negotiations in 1664 between a member of the Medici royal family and a notable dealer, Annibale Ranuzzi, for a painting purportedly by Michelangelo. As negotiations between Leopoldo de’ Medici and Ranuzzi progressed, Ranuzzi admitted that Michelangelo was not the author of the painting, but this did not end the prospect of making the sale. Ranuzzi argued that the painting he sought to sell was so exquisite that it not only could pass for a Michelangelo but was “more beautiful than if it were by Michelangelo, in a style that will not be difficult to baptize as by the hand of Michelangelo, especially if it were in your Highness’s room.”3
These stories demonstrate that, as works of art by Renaissance and Baroque masters became increasingly rare, members of Italy’s aristocracy permitted copies of their best paintings to be made and given to friends or allies as special gifts or, conversely, requested that copies be made for their collection if the originals were unavailable. Often, the master himself would participate in creating a second or duplicate signature copy of a significant painting at the request of an important donor or donors. As a result, the question of what constituted an “original” or “signature” painting by a master and what constituted a “copy” became increasingly difficult to define. The problem is illustrated by Guido Reni’s practice of repeatedly copying his most illustrious paintings. To a casual viewer, there was often no perceptible difference between Reni’s initial composition, which served as a prototype, and his subsequent copies. Reni’s goal, however, was not simply to “stamp out” pictures but instead to achieve a more perfect image of an earlier composition by making subtle changes in color, texture, and brushwork.4 While Reni became known for the quantity of replicas that came from his studio, practically every notable Renaissance and Baroque artist engaged to some extent in the practice of making more than one “signature” painting of the same picture. Perhaps the three most illustrious examples of this practice are Leonardo’s Virgin of the Rocks, copies of which are at the Louvre and the National Gallery in London;5 Giorgione’s Adoration of the Shepherds, versions of which are at the National Gallery of Art in Washington, D.C. (the Allendale Nativity) and the Gemäldegalerie in Vienna; and Titian’s Portrait of Filippo Archinto, copies of which are in the Johnson Collection at the Philadelphia Museum of Art and the Altman Collection at the Metropolitan Museum of Art.
Many Renaissance and Baroque masters, seeking to capitalize on the market for their work, employed dozens of assistants to copy an earlier painting or complete a painting based on only a concept or sketch by the master, further complicating the identity of the actual artist and leading ultimately to the common “school of” attributions used in museums today.6 Vasari reported that Raphael “kept a great number of artisans at work . . . and was never seen leaving home . . . without fifty painters, all worthy and good men, accompanying him.”7 Titian’s large workshop of over thirty artists included his son, his nephews, and several other artists who resided with Titian in his home for many years and who became consequently so intimately familiar with Titian’s manner of painting that their versions and Titian’s own paintings became hardly distinguishable. In the later years of Titian’s long life, members of his workshop began to sell copies painted solely by them while pretending that Titian was also involved.8 The production of workshops became so widespread and prodigious that John Walker, former director of the National Gallery of Art, has suggested that “virtually all Italian Renaissance paintings were the product of a shop with assistants almost invariably at work on some part of the picture.”9
By reason of the express terms of most commissions, those who contracted with a master to paint a picture were well aware that that the artists employed in the master’s studio would have a significant role in its production. Most commissions called for the painting to be done “by his [the master’s] hand” (di sua mano). While this phrase might connote to modern readers a contractual obligation imposed on the master to complete the painting solely with his own hand, it meant something significantly different to the contracting parties in the fifteenth and sixteenth centuries. The phrase di sua mano meant only that the master was obligated to design the work, retouch the work of his subordinates where necessary, paint the most important or difficult aspects of the painting, and approve of the final product. The master was free to delegate most of the painting to other artists employed within his workshop.10 Some master painters permitted a painting to be considered an “original” by them if they simply retouched the painting with a few brush strokes. Titian, for example, was legendary for completing the paintings of his workshop with only a few strokes of his brush.11 Guido Reni, to cite another example, was reported to have responded to a request from a cardinal for a copy of one of his paintings by stating, “I not only entreat Your Eminence to have a copy made of it, but I promise you, that, without any profit, I will retouch and finish it all in such a way that it will not have to envy the original.”12 Some contracts even allowed the master to subcontract a painting or part of a painting to an independent artist so long as he approved and signed the final product.
Paintings authored by Italian masters were endlessly copied not only by their students and followers but also by other distinguished masters. These copies were not intended to deceive a purchaser or to devalue the original painting. Instead, the painters making the copies sought to demonstrate their own skill in being able to replicate the work of another great artist while at the same time paying homage to him. As reflected in Vasari’s tribute to Raphael, the practice of imitation was highly encouraged throughout the Renaissance. Referring to Raphael, Vasari wrote that his successors should imitate him, “for anyone who imitated him [Raphael] discovered that he had taken refuge in a secure port, and likewise, those painters, who in the future will imitate his efforts in the art of painting will be honored in the world.”13 In keeping with Vasari’s advice, copies of Raphael’s famous portrait of Pope Julius II were made not only by artists in his studio but also by Titian, another great master. Today, Raphael’s portrait of Julius II is in London’s National Gallery; the copy by his studio, in the Uffizi Gallery; and the copy by Titian, in the Pitti Palace in Florence.14 In the same spirit that prompted Titian to copy Raphael, Rubens made numerous copies of paintings by Titian, and the cycle of one outstanding painter copying another continued for generations.15 Because countless copies of Titian paintings were made over the centuries, it is not surprising that seven paintings erroneously attributed to Titian found a home in Massarenti’s collection in the late nineteenth century and were later purchased by Walters.
In the sixteenth and seventeenth centuries, copies of old-master paintings also proliferated in teaching academies throughout Italy. Milan’s Ambrosiana Art Academy, founded around 1620 by Archbishop Federico Borromeo, commissioned artists to paint copies of exemplary paintings having sacred themes, such as Leonardo’s Last Supper, Correggio’s Coronation of the Virgin, and Raphael’s Adoration of the Magi. The copies were used both as models for students, who were admitted and graded on the basis of their ability to replicate them, and for devotional purposes, as called for by the Council of Trent. According to Borromeo, “It was a praiseworthy thing to procure copies provided that they are worked with extreme diligence and taken from the most excellent models.”16 By integrating his copies into his splendid collection, Borromeo created one of the earliest and greatest full-scale teaching museums in Europe.17 It was a model that Henry Walters might have had in mind when acquiring many copies of Italian masterpieces in the early 1900s to augment the paintings he purchased from Massarenti and to satisfy the educational goals of his own collection.
As the making and acquisition of copies became an increasing phenomenon in the burgeoning market for Italian paintings between the fifteenth and eighteenth centuries, many talented painters achieved their fame not through creativity or originality but as artist-entrepreneurs who satisfied the taste for Renaissance and Baroque art by beautifully replicating and imitating the art of others. Vincenzo Camuccini was among the most gifted of these Italian copyists. Beginning at the age of thirteen, he supplemented his income by painting replicas of Italian old masters and selling these copies as originals to primarily unsuspecting English aristocrats who had traveled to Rome as part of their grand tour. He focused his talents on copying paintings by famous Baroque artists, such as Annibale Carracci, Caravaggio, Guido Reni, and Guercino.18
By the middle of the nineteenth century, after centuries of copying, it had become exceedingly difficult to distinguish between an Italian painting authored by a master, a painting made by subordinate artists in the master’s workshop, a close version of a masterful painting made by another notable artist as a tribute to the master, an imitation made by a very good student or follower of the master, and a forgery or fake made by a copyist for the purpose of deceiving a purchaser into believing that it was painted by the master.19 The passage of time had simply obliterated, for the most part, both the identities and the motivations of the artists who made the copies. Churches, museums, and private collections throughout Italy were full of paintings that lacked any provenance and had uncertain authorship. Although it is impossible to quantify the number of such anonymous paintings, one art historian has estimated that half of the notable paintings in Italy were without a reliable attribution.20 As a result, Italy’s rich cultural heritage epitomized by its Renaissance and Baroque paintings was in danger of becoming as fragmented and anonymous as the pieces of classical Roman sculpture that had been recovered from its soil.
Following Italy’s unification in 1861, Giovanni Morelli, a member of the Italian Parliament and Italy’s foremost connoisseur, undertook the monumental task of changing this. His mission was to establish a national inventory of Italy’s art treasures by identifying the authors of the massive number of misattributed Italian Renaissance and Baroque paintings, to promote their cultural and economic value, to place these paintings in newly established museums, and to prevent the most precious of these works of art from being gobbled up by hungry collectors and newly established museums in Germany and England. In pursuing these goals, he became a giant in his field.
Morelli’s fame is derived primarily from the scientific methodology he devised for identifying the authors of Renaissance paintings. Trained in medicine and the natural sciences, Morelli’s understanding of human anatomy led him to examine works of art for unnoticed, anatomical details of the painter’s subject, such as his or her ears and hands. These details, Morelli insisted, were spontaneously repeated by artists in their paintings and served like fingerprints to link painters to their works of art. Morelli contended that the evidence imbedded in the painting itself, which could be detected by close visual analysis, provided a more confident basis for determining the painter’s author than did collateral documentary evidence, which was often misleading.21 Morelli, however, did not believe that accurate attributions were essential for their own sake. Instead, he thought they were a means for understanding that the art of Italy had evolved “organically” based on the distinct natural and cultural qualities of the different regions of Italy and the different schools of art they produced.22 To Morelli, connoisseurship was an inseparable component of art history.
Morelli’s approach to connoisseurship and his willingness to challenge the conventional attributions given to Italian paintings by earlier art historians and museum curators had a dramatic effect on Berenson.23 After studying Morelli’s explanation of his scientific approach in 1889 and briefly meeting with him in 1890, Berenson became his most dedicated disciple. Berenson referred to Morelli as his “revered master.” He expressed the belief that the brilliance of Morelli’s scientific approach to art was comparable to if not greater “than Winckelmann’s to antique sculpture or Darwin’s to biology.”24 In two books written around the turn of the century, Berenson explained to a broader audience the importance of the Morellian method of connoisseurship and his plan to develop a similar approach to the study of art history.25
Morelli, however, had one failing which, like a virus, seemed difficult to resist. It was caused by allowing his independent judgment to be compromised by his pursuit of financial gain. Despite Morelli’s public opposition to the export of Italian Renaissance paintings, from the 1850s to the time of his death in 1891, he engaged actively as a dealer in the marketing of Italian paintings to wealthy patrons abroad, especially in England. In return for commissions, he authenticated, estimated the value, and sold paintings not only from outside sources but also from his own collection. Sometimes he exaggerated the importance of the paintings in his collection, leading scholars to question in hindsight the objectivity of many of his attributions.26 His willingness to compromise his intellectual judgment for economic gain soiled his otherwise splendid reputation.27 Part brilliant connoisseur, part self-serving commercial entrepreneur, Morelli became a twisted model of the conflicts and contradictions that subsequently confronted many of his followers, especially Bernard Berenson.
Beginning in the sixteenth century, successive generations of wealthy English art collectors became the principal victims of the Italian copyists.28 The copies that they unwittingly purchased were proudly displayed in their country homes as if they were original Renaissance and Baroque masterpieces. In November 1894, an extensive exhibition of Venetian old masters owned by members of the English aristocracy opened at the New Gallery in London. The paintings purportedly included thirty-three paintings by Titian, seventeen by Giorgione, fourteen by Giovanni Bellini, and thirteen by Paolo Veronese. Berenson, who was relatively unknown at the time, attended the London exhibition. Applying the scientific theory of connoisseurship that he had learned from Morelli, Berenson wrote, in March 1895, a devastating critique challenging most of the attributions in the exhibition. With regard to the thirty-three paintings attributed to Titian, Berenson wrote that “no other name, it is true, is so recklessly abused as Titian’s,” and he concluded that only one of the thirty-three was painted by Titian. Turning to the seventeen paintings allegedly by Giorgione, Berenson focused on one, Portrait of a Lady Professor of Bologna, and caustically stated that it was “neither of a Lady, nor of a Professor, nor of Bologna.” He claimed that neither this painting nor any others in the exhibition were by Giorgione. With regard to the paintings by Giovanni Bellini, Berenson expressed the opinion that throughout Europe only one in six paintings attributed to Giovanni was actually by him and, consistent with this ratio, only three of the eighteen paintings in the exhibition were painted by him. And with regard to the paintings by Veronese, none, according to Berenson, were by him.29 Much to the chagrin of the owners of the paintings on display, Berenson’s analysis of the exhibition at London’s New Gallery helped launch his career and quickly made him the most well-known, albeit controversial, connoisseur of Italian art.30
Berenson’s critique of the exhibition of Venetian artists in London was republished in 1901 in volume 1 of The Study and Criticism of Italian Art. In this volume, Berenson observed that, due to the haphazard practices of what he characterized as the “old connoisseurship,” attributions to famous Italian masters were often abused and not worthy of trust.31 This warning was followed in 1902 by Berenson’s republication of his “Rudiments of Connoisseurship,” in which he cautioned that many Italian masters delegated the execution of their conceptions to members of their workshops, that the names of the masters placed on these paintings should be treated with suspicion, and that the attribution given to any picture purportedly by an Italian master should be subjected to “severe criticism before it is accepted.” Berenson then proceeded to describe in detail the morphological standards he used in assigning a name to a painting, suggesting that he was among the very few who were able not only to reliably identify the artist but also to rate the quality of the painting. Blending his extraordinary knowledge about art with his interest in self-promotion, Berenson concluded that connoisseurship to him was not just a science but an art as well.32
Berenson was not the first or only connoisseur to warn prospective buyers against the peril of copies and false attributions in the Italian art market. In 1860, James Jackson Jarves, in an article in the Atlantic Monthly, “Italian Experience in Collecting Old Masters,” observed that, due to the pervasive practice among Florentine artists and dealers of forging copies of old-master paintings and documents designed to deceive unwary buyers about the paintings’ provenance, there was in Florence an unending quarry of copied paintings that had no value. Jarves warned that “no purchaser . . . should give heed to any statement about the history or authenticity of the works offered to him through such channels [Florentine art dealers], but rely both for value and facts upon his own resources.”33 Ironically, although Jarves recognized that Florence had turned into a “vast picture shop,” Jarves himself fell prey to purchasing many works of art by purported masters whose attributions proved to be untrustworthy, including a painting allegedly by Raphael that turned out to be a forgery.34
As a result of the Italian culture of reproduction that continued for four hundred years, American millionaires who ventured into the market for Italian master paintings at the turn of the last century had to tiptoe through a veritable minefield of copies and fake paintings as well as contrived documents seeking to prove the contrary. Risk was an inherent component of collecting Italian old masters. Benjamin Ives Gilman, the influential director of the Boston Museum of Fine Arts, warned that “the flood of reproductions threatens to drown out the seeds of artistic culture” and that it was hard to distinguish great Italian masterpieces from “the chaos of reproductions.”35 An article in the New York Times Magazine similarly cautioned that a virtual flood of fictitious “old master” paintings had been arriving in the United States, causing America’s art lovers to “meet with copies everywhere.”36
At the beginning of the twentieth century, wealth provided American millionaires with no shield from the perils of collecting Italian paintings. While Isabella Stewart Gardner, with the assistance of Berenson, was able to assemble a significant collection of Italian old masters before the turn of the century, other titans struggled. Andrew Mellon, for example, allegedly spent over $400,000 (the equivalent of $10 million today) between 1899 and 1905 without purchasing a single painting that he later found worthy of hanging in the National Gallery of Art.37 In 1903, John G. Johnson, a wealthy and renowned Philadelphia attorney and one of America’s earliest collectors of Italian Renaissance paintings, was fooled by an Italian dealer who purported to sell him twenty-three paintings by Renaissance masters.38 After reviewing Johnson’s collection, Berenson sent him a letter that brutally challenged the attributions of some of Johnson’s most precious paintings. With regard to Johnson’s painting attributed to Botticelli, Berenson sarcastically noted, “Never in the world”; with regard to his Carpaccio, Berenson wrote, “most certainly not”; with regard to his Perugino, Berenson noted, “Not Perugino”; and with regard to his Lorenzo di Credi, Berenson caustically observed, “No and not even Florentine.”39 When, in 1904, Berenson visited the collection of Joseph Widener, he found “mostly horrors masquerading under great names,” and when he returned in 1908, he similarly found “nothing of importance among the Italians.”40 A Titian purchased by Widener for $40,000 was discovered to be a copy, and a painting attributed to Botticelli was downgraded to a workshop piece.41 In 1909, J. P Morgan paid $200,000 for a Madonna and Child, purportedly by Raphael, only to have its authenticity questioned by Berenson and others. This resulted in the painting being reattributed to “School of Raphael” and subsequently sold by Morgan’s heirs for only $2,500, a small fraction of its original price.42 In light of the publicity about the perils of collecting Italian Renaissance paintings and the misadventures of some of his fellow millionaires, Walters should have been keenly aware of the risk he was taking when venturing into this field.