www.hks.harvard.edu Risky Business: Commissioning Portraits in Renaissance Italy Faculty Research Working Paper Series Jonathan K. Nelson Syracuse University Florence and Harvard Kennedy School Richard Zeckhauser Harvard Kennedy School July 2019 RWP19-024 Visit the HKS Faculty Research Working Paper Series at: https://www.hks.harvard.edu/research-insights/publications?f%5B0%5D=publication_types%3A121 The views expressed in the HKS Faculty Research Working Paper Series are those of the author(s) and do not necessarily reflect those of the John F. Kennedy School of Government or of Harvard University. Faculty Research Working Papers have not undergone formal review and approval. Such papers are included in this series to elicit feedback and to encourage debate on important public policy challenges. Copyright belongs to the author(s). Papers may be downloaded for personal use only. https://www.hks.harvard.edu/research-insights/publications?f%5B0%5D=publication_types%3A121 1 Risky Business: Commissioning Portraits in Renaissance Italy‡ Jonathan K. Nelson∗ and Richard Zeckhauser† Abstract Portraits served as a form of social media in the Renaissance. Prominent individuals commissioned portraits to convey their accomplishments and relationships, not merely their images. Political and church leaders, in particular, used the images to bolster their role, but these commissioned works entailed risks, importantly including risks to reputation. A portrait could be unflattering or unrecognizable. It could also be judged to be indecorous, especially if the portrait was perceived as an attempt to elevate an individual above his or her station. The artist-patron relationship was one between principal and agent. The time gap between commission and delivery brought risks. The work might be delayed, or simply not delivered. Both were significant risks with both Leonardo da Vinci and Michelangelo. Portraits by other artists might turn out to be of low quality or violate decorum. In either case, the reputation of both patron and artist would suffer. A number of salient examples of portraits gone wrong are analyzed in this essay. We mention two here. Elisabetta, mistress of Roberto Malatesta, the last of a long line of rulers of Rimini, became regent when Roberto died unexpectedly. She was guardian of their two minor sons and succeeded to power when Roberto’s brother’s plot to murder her and her elder son failed. Elisabetta then commissioned an altarpiece by Domenico Ghirlandaio for a famed local church. It was intended not only to show her devotion, but also, and strategically important, to publicly affirm her status as ruler, and her source of legitimacy as mother. She had herself depicted in the place of honor, in an outward gaze (unusual for a female patron) and nearly a head taller than her son. The deviations from the norm in the way she was portrayed, if widely accepted, would elevate Elisabetta, but such deviations elevated risks as well. Domenico died and his brother, hardly his equal in skill, took over. Elisabetta took another risk and demanded a price reduction, then accepted binding arbitration. The arbiter granted that quality had suffered and cut the price, but he also noted the departures from decorum in the composition. The reputation of the painting, artist, and presumably the patron, suffered. About twenty years later, when the Malatestas were driven from Rimini, the donors’ portraits were painted out. Francesco del Giocondo commissioned a portrait of his wife, (Mona) Lisa, which has become the most famous portrait of all time. But the story does not end happily for Francesco. Instead of delivering the painting to the original patron, Leonardo kept it and later described it to Cardinal ‡ In Mitteilungen des Kunsthistorischen Institutes in Florenz, thematic issue, “Bad Reception: Negative Reactions to Italian Renaissance Art,” ed. by Diletta Gamberini, Jonathan K. Nelson, and Alessandro Nova, forthcoming. ∗ Syracuse University Florence, and Harvard Kennedy School Research Fellow † Harvard Kennedy School 2 Luigi of Aragon as commissioned by Giuliano de’Medici. The portrait was criticized in its time for many of its highly innovative features that are prized today. Scandal also surrounded the portrait. It served as a prototype for the so-called Mona Vanna, a topless portrait, also due to Leonardo. A final unfortunate tale reemerged a few years ago, in a letter from 1515. Two prominent Florentines, including Duke Giuliano’s nephew Lorenzo, de facto ruler of Florence, made advances to Lisa del Giocondo. She refused. But the suggestion arises that the Mona Vanna was made for the erotic pleasure of Giuliano or Lorenzo de Medici. Commissioning a portrait, as a venture into social media, always brought risks in the Renaissance. Those risks were magnified then, as they are today, when sex and power mixed together with the social medium message. 3 The history of portraiture in the Italian Renaissance is usually presented as a happy sequence of successes. Prominent individuals would commission art that showcased their best qualities. The works would delight the patrons, the artists, and the viewers. The cultural patrimony of the world would be increased and, centuries later, fondly celebrated. But just as accounts of wars are disproportionately written by the victors, the successful fruits of Italian Renaissance patronage tended to be prominently displayed, approvingly admired, discussed by the patrons and artists, and written up favorably by their contemporaries and ours. Works that dissatisfied their patrons, by contrast, tended to get lost to history. Giorgio Vasari, in his Lives of the Artists, discussed far more successful commissions than he did failures. Though individual portraits that caused their subjects displeasure or shame are occasionally mentioned in studies of the patrons or artists, such unsuccessful works have never been studied systematically as a group. Consequently, standard assessments of Italian Renaissance commissions convey an unrealistically favorable picture. That is, the roster of familiar portraits (and other images) vastly exaggerates the ratio of successes to failures, because the failures of the patronage process tended to be hidden and forgotten. In response, the risks inherent in having one’s portrait rendered in Renaissance Italy have been woefully underestimated.1 Risk, which in common parlance refers to the possibility of losing something of value, includes two elements: an event’s probability, or likelihood of happening, and its consequences.2 For this reason, we identify a different conception of the term ‘risk’ than does Suzanne Blier, in her recent book on Art and Risk in Ancient Yoruba. Her central thesis is that, “Risk spurred […] artists (and patrons) into thinking about materials, techniques, and art forms in striking new 1 Informal tallies tend to ignore what decision theorists call selection bias. 2 Economists and decision theorists often employ ‘risk’ in a more technical fashion, referring to situations where the outcome is unknown but the probabilities of possible outcomes are known, such as flipping a coin. ‘Uncertainty’ applies when probabilities are unknown, such as whether a patron will approve a commissioned portrait. ‘Ignorance’ describes situations where even the possible outcomes are unknown, such as the fate of the Mona Lisa, discussed below. See Richard J. Zeckhauser, “Investing in the Unknown and Unknowable”, in: Capitalism and Society I (2006), pp. 1-39. 4 ways.” 3 We are familiar with risks that involve money, such as investments that go bad or losses that are readily compensable with money, such as wrecked cars. However, some losses, such as the loss of one’s good name, or what Pierre Bourdieu famously called ‘symbolic capital,’ are difficult to measure in monetary terms. For both patrons and artists in Renaissance Italy, the loss of reputation was a consequential risk associated with a portrait. A portrait that would be on public view, as in a palazzo, church, or town square, entailed a particularly heavy risk of reputational loss. To facilitate our comparison of highly heterogeneous examples in Italy, from the late fifteenth- through the late sixteenth-century, we establish three broad categories of possible risks to patrons: the quality of the representation, the timeliness of its completion, and the acceptability of its message. First, a prime goal of portraits was to present the individual in a manner both recognizable and favorable. A Renaissance term synonymous with “to portray” was contraffare, which implies an exact reproduction.4 Nevertheless, Renaissance patrons knew what Aristotle had written about good portrait painters in his Poetics (1454b. 9-11): “They, while reproducing the distinctive form of the original, make a likeness which is true to life and yet more beautiful.” The verisimilitude of a portrait was closely wrapped up with the question of quality, as noted in already ancient poems.5 The requirement that the sitter’s features be both accurate and flattering frequently created tension. The second category of risk to the patron involved the timeliness of the portrait’s completion. Many commissioned works of art were delivered late, and some not at all. Most documented criticisms of Italian Renaissance portraits that disappointed their patrons address the quality or the timeliness of the commission. 3 Suzanne Preston Blier, Art and Risk in Ancient Yoruba: Ife History, Power, and Identity, c.1300, Cambridge UK/New York, 2015, p.17. 4 Joost Keizer, “Portrait and Imprint in Fifteenth-Century Italy”, in: Art History XXXVIII (2015), pp. 10-37:31. He cites a 1493 letter of Isabella d’Este Isabella d’Este, who complained about the difficulty of finding portrait painters who “contrafaciano el vulto naturale”. 5 See anonymous for the Greek distich by Leonidas of Alexandria, and the Roman epigram by Lucillius. Also see discussion of an anonymous Greek epigram that criticized a portrait because it accurately depicted the unattractive features of the sitter. 5 The third type of patronal risk involved a somewhat different type of requirement: paintings and sculptures needed to convey an appropriate message. At times, however, portraits were deemed to be confusing, unpopular, or indecorous. Such failings often resulted from the selection, pose, placement, or attributes of the individuals portrayed. Patrons and their contemporaries who perceived failings in any of these perhaps subtle and subjective matters might express disapproval by criticizing, rejecting, damaging, or even destroying a portrait. An unacceptable message could be conveyed by an otherwise excellent portrait that honored a political leader who had fallen out of favor or even been deposed. To cite but two prominent examples, drawings alone survive of Leonardo da Vinci’s statue of Francesco Sforza, Duke of Milan; even less remains of Michelangelo’s bronze statue of Pope Julius II.6 On occasion, political upheavals could lead to a form of damnatio memoriae, the condemnation of memory. In 1330, a marble tomb in the cathedral of Arezzo celebrated events from the life of Guido Tarlati, Lord of Arezzo.7 A decade later, when popular sentiment turned against this ruler, his image was chipped away from the monument. Risks to Italian Renaissance artists were inextricably linked to those of their patrons. That interweaving is captured by the principal-agent model, which plays a prominent role in such fields as economics and political science.8 Modern lawyers and dressmakers, like Renaissance artists, are agents whose responsibility is to act on behalf of the principals. The difficulty in all three cases is that the agent, once engaged, has substantial latitude, and the work in preparation is difficult or impossible to monitor directly. The contracted product does not yet exist and can only be delivered later, often years after the commission. Quality and an agreed approach may suffer in the interim, due to what decision theorists call ‘moral hazard’. Moral hazard arises when the incentive for an agent to skimp on effort when his payoff does not depend on the outcome, such as when an insured individual drives too carelessly, or an artist with an agreed price for a 6 For the former, see Leonardo da Vinci’s Sforza Monument Horse: the Art and the Engineering, ed. by Diane Cole Ahl, Bethlehem 1995; for latter, see anonymous. 7 Georgina Pelham, “Reconstructing the programme of the tomb of Guido Tarlati, Bishop and Lord of Arezzo”, in: Art, Politics and Civic Religion in Central Italy, 1261 – 1352, ed. by Joanna Cannon and Beth Williamson, Aldershot 2000, pp. 71-115. 8 For discussion, in relation to artistic production, see Jonathan K. Nelson/Richard J. Zeckhauser, The Patron’s Payoff: Conspicuous Commissions in Italian Renaissance Art, Princeton 2008, pp. 17-28. 6 commission passes too much work to underlings. The principal-agent relationship of reliance is not part of traditional immediate beneficial exchange. Renaissance portrait artists faced further risks: mainly losses of compensation, reputation, or future work. A painter might have to choose between the Scylla and Charybdis of rendering a homely subject with accuracy or with flattery. Perhaps not surprisingly, many patrons claimed that an unflattering portrait was inaccurate, as we see in the letters of Isabella d’Este, discussed below. Thus, it is often difficult for us to evaluate the nature of an artist’s proclaimed shortcomings. When the quality of a Renaissance portrait, or any other work of art, was deemed unacceptable, the principal was likely to take action. Thus, dissatisfied patrons, at a minimum, might display the work in a less prominent location than originally planned. More significantly, they could withhold or reduce payment, adapt or destroy the oeuvre, besmirch the artist’s reputation, or take some combination of these actions. One the one hand, such actions might also undermine a patron’s reputation, and thus the ability to commission works from artists in great demand. Vasari, for example, recounts that Donatello destroyed a bronze head when his patron, a Genovese merchant, refused to pay the requested sum.9 On the other hand, the threats of such patronal actions could help to ensure quality, and their implementation could enhance the credibility of future threats. Surely, the potential for negative reactions by patrons helped to promote the quality, timeliness and appropriateness of commissions, and thereby dampened the risks of commissioning a portrait. By isolating and identifying those risk elements, we can better understand the strategies that patrons and artists employed to control them. This approach also casts new light on old favorites, such as the Mona Lisa, an example discussed below. To understand how the production of Italian Renaissance portraits imposed risks on both the patron and the artist, we can adapt the etic approach examined in a recent article that presented Pope Sixtus V’s Fontana dell’Acqua Felice as a “failed communication channel.”10 In 9 Giorgio Vasari, Le vite dei più eccellenti pittori, scultori, ed architettori nelle redazioni del 1550 e 1568, ed. by Paola Barocchi and Rosanna Bettarini, Florence 1966–1984, III, p. 212f. Similarly, Vasari’s account of Michelangelo’s Doni tondo illustrates the high cost to a patron for attempting to underpay a celebrated artist. 10 Tamar Cholcman/Dafna Maharshak, “Advertising Gone Wrong: Sixtus V in the Image of Moses: The Fontana dell’Acqua Felice as a Failed Communication Channel”, in: Studies in Visual Arts and Communication I (2014), URL: http://journalonarts.org/wp-content/uploads/2015/05/SVACij-Vol1_No1_2014-CHOLCMAN_T- MAHARSHAK_D-Advertising-gone-wrong.pdf (accessed on 11 July 2018). http://journalonarts.org/wp-content/uploads/2015/05/SVACij-Vol1_No1_2014-CHOLCMAN_T-MAHARSHAK_D-Advertising-gone-wrong.pdf http://journalonarts.org/wp-content/uploads/2015/05/SVACij-Vol1_No1_2014-CHOLCMAN_T-MAHARSHAK_D-Advertising-gone-wrong.pdf 7 Advertising Gone Wrong, the authors tweaked an analytic framework, initially used to evaluate advertisements, in order to explore how public art conveys a message and how such conveyance could go wrong. Those authors define three key advertising concepts, of image, headline, and text, and their three parallels in public art, namely visibility, central message, and elaboration. A prime source of risks for both advertising and public art is what the authors call noise, and which we call interference. Elements that interfere distract from, disrupt, or even counter the main message. Attention to disruptive elements allows us to reevaluate the donor portraits in the Malatesta Altarpiece (Rimini, Museo della Città, fig. 1), painted by Domenico Ghirlandaio and his subordinates in 1493–1496.11 Figure 1. Domenico Ghirlandaio and workshop, Malatesta Altarpiece, tempera on panel (Rimini, Museo della Città) 11 This account of the altarpiece, and the document quoted below, derives Jonathan K. Nelson, “Breaking Conventions: Donor Portraits in Ghirlandaio’s Malatesta Altarpiece”, in: The Art and Language of Power in Renaissance Florence: Essays for Alison Brown, ed. by Amy R. Bloch, Carolyn James, and Camilla Russel, Toronto 2019, pp. 221–249. That analysis, however, does not address the question of risk. 8 We start with the key concepts of visibility, central message, and elaboration. First, to reconstruct the conspicuousness of the Malatesta altarpiece to its intended audience, one must imagine the chapel of Elisabetta Aldobrandini, in the prominent church of San Cataldo in Rimini. Local visitors surely knew about the chapel of the de facto ruler of the city, where the altarpiece was prominently displayed. The only modern Florentine work in Rimini, the altarpiece was commissioned from Domenico Ghirlandaio, one of the most famous painters in central Italy. Second, the central message of the altarpiece is expressed by the three standing saints, Vincent Ferrer, posed between Sebastian and Roche; and by the four kneeling donors, from the left, Elisabetta Aldobrandini, her elder son Pandolfo Malatesta, his wife Violante Bentivoglio, and his younger brother Carlo. Third, the lunette above, the fictive bronze reliefs in the background, and the predella below further elaborate the main message. The message of the altarpiece would have been clear to the intended local audience, in light of recent political developments. When Roberto Malatesta, Lord of Rimini, had died unexpectedly in 1482, his mistress Elisabetta had become the guardian of their two sons, who were then minors and next in line to rule. She became regent after Roberto’s brother, her main rival for power, had sought but failed to murder Elisabetta and her sons. The altarpiece thus served as a public affirmation of her status as both ruler and mother. Unexpectedly, the female donor is depicted in the place of honor, to the right of the central holy figure, and thus on the viewers’ left. She faces outward, an unusual pose for a female patron. Most surprisingly, Elisabetta appears nearly a head taller than her son. A Renaissance artist would hardly have decided unilaterally to deviate from such norms in the position, angle, and height of a patron’s portrayal. Elisabetta surely had conveyed her desire to appear prominently in the painting, and the painter informed her that request did not follow the norm. Obvious deviations like these raised the stakes for all. Though patrons making such a request would surely have expected greater benefits if all went well, they amplified the risks of a dampened or negative reception. In this instance, the norm-breaking contributed to the negative comments about the Malatesta altarpiece portraits. 9 Domenico Ghirlandaio died in 1493. Soon after, the altarpiece was completed by his workshop, led by his brother Davide. Given the lesser artist, Elisabetta refused to pay nearly 20 percent of the stipulated price. In return, Davide refused to send the painting, and arbitration ensued. The surviving documents from 1496 provide one of the very rare legal records about a bad reception of an Italian portrait in the fifteenth century. After the arbiter had the altarpiece evaluated by several painting experts, Ghirlandaio’s heirs were granted less than half of the payment outstanding, for three reasons. First, the quality had been downgraded because of the great difference between Domenico Ghirlandaio and the artists who completed the altarpiece. Second, “certain figures of the illustrious lords of Rimini, who ought to have been drawn from nature, do not at all correspond to their persons or to their appearance.”12 The latter shortcoming presents no surprises; patrons wanted portraits to be recognizable. Third, and more unusual and interesting, the arbiter objected to the persons, that is the roles played by the donors. This is presumably a reference to the excessive elevation of Elisabetta, the former regent, in relation to the ruler Pandolfo, who had come of age in 1495. Three important lessons emerge from the analysis of this work for a more general study of the risks in Italian Renaissance portraiture. First, patrons who were portrayed in the works incurred significant risks, our principal thesis. Second, due to the principal-agent structure of their relationship with the patron, the artists incurred significant risks as well. Thus, Davide Ghirlandaio suffered a pay cut, even though we can assume that the patron herself had requested some of the major elements that led to the arbitrated price reduction. By the judgments of the painting experts, Davide also suffered a loss of reputation for having painted those elements. Third, even though the patron ended up paying a lesser price, this hardly made up for her losses. The inferior quality of the altarpiece, together with the lack of decorum and verisimilitude in the portraits, interfered with viewers’ ability to extract the work’s main message that Elisabetta ruled the city. Elisabetta’s reputation took another blow a few decades after her death in 1497. The donor portraits were all painted out, most probably as part of a damnatio memoriae following the expulsion of the Malatesta family from Rimini in 1528. 12 For translation and discussion of the awkward Latin phrase, see ibidem, p. 236f. 10 The concept of risk, as the potential loss of something of value, was already well-known in Medieval Italy.13 In this study, we thus aim to use what anthropologists would call an emic approach, one which reflects the point of view of the patrons and artists. Formal insurance mechanisms are thought to have originated in Genoa in the fourteenth century; by the fifteenth century, marine insurance was well established for trade in the Mediterranean. Economic historian Giovanni Ceccarelli, describing maritime commerce in Renaissance Florence, explained that insurers evaluated both structural and contingent risks.14 Wealthy merchants, who shipped their goods and worked with insurance brokers, were the same individuals who commissioned many portraits. The risks associated with maritime expeditions and the delivery of goods involved material items, items that if lost could be directly compensated with money. The most severe risks of portraiture, by contrast, involved reputational loss. Money cannot make one whole from such losses of symbolic capital, and not surprisingly, reputational losses were never insured against. Patrons were surely aware of the risks when they commissioned paintings and sculptures. Merchants in Renaissance Italy, like many of their counterparts today, were experts in prices and obsessed with calculations for the future. Especially in Florence, where the ruling class expressed the mercantile culture, an extraordinary number of record books survive which attest to a keen awareness of uncertain future outcomes and the need to take risks. Though some patrons, such as popes and nobles, might have had less helpful commercial experience, they too were surely aware of risks. That hardly implies that either group always made rational or well- informed decisions, but they were surely alert to possible dangers.15 Filippo Strozzi explained, in a letter to his brother, that commissions for family tombs should be done only if they induced 13 For discussions of risk in medieval and early modern Italy, see Giovanni Ceccarelli, Un mercato del rischio: assicurare e farsi assicurare nella Firenze rinascimentale, Venice 2013; and Giovanni Ceccarelli, Il gioco e il peccato: economia e rischio nel tardo Medioevo, Bologna 2003. 14 Giovanni Ceccarelli, “Stime senza probabilità. Assicurazione a rischio nella Firenze rinascimentale”, in: Quaderni storici CXXXV (2010), pp. 651-702. 15 The field of behavioral decision, which has received two Nobel Memorial Prizes in Economics Sciences, has a primary finding that individuals often make clearly non-rational decisions, even when dealing with monetary losses. See, for example, Decision Making: Descriptive, Normative, and Prescriptive Interactions, ed. by David E. Bell, Howard Raiffa, and Amos Tversky, New York 1988. 11 honor, otherwise they should be dropped because they led to shame.16 Certainly, Baccio Valori reaped more shame than honor when he decorated the façade of his home with a series of expensive marble portraits. To his dismay, they were ridiculed for having ugly visages, or visacci, thus giving his home its nickname of the Palazzo dei Visacci; this insult focused on the skill of the sculptor. More importantly, as Robert Williams observed, Valori’s learned contemporaries wrote sonnets indicating that “the very idea of herm portraits in such a setting is tasteless or inappropriate.”17 Period observers also objected to images of citizens publicly ‘impaled’ a violation of decorum. Any work of art ran the risk of showing a lack of skill or decorum, but reputational loss was a particular risk for the primary person depicted. Consider Tommaso Rangone, the wealthy physician who financed, at great cost, the reconstruction of the façade of the Venetian church of S. Giuliano. For this, he commissioned a bronze portrait of himself by a renowned artist, Alessandro Vittoria.18 (fig. 2) The bronze was festooned with depictions of various learned attributes and an inscription, Thomas Philologus Ravennas. This ostentatious presentation provoked an insult by a contemporary, who probably objected to the indecorous display by someone who was neither local nor noble. A portrait of the poet Bartolomeo Aragazzi met a similar fate. Leonardo Bruni, his learned contemporary, put his criticism of this commission in the mouth of a workman, who complained that Aragazzi, “well known to be stupid and puffed up with conceit, ordered a marble tomb to be made for himself.”19 This led Bruni himself to mischievously inquire “what actions, what deed will you have inscribed? That your father drove asses and goods around the fairs?” In these two examples, patronal choices merited 16 For discussion, see Eve Borsook, “Documents for Filippo Strozzi’s Chapel in Santa Maria Novella and Other Related Papers”, in: The Burlington Magazine CXII (1970), pp. 737-745: 738 n. 20. 17 Robert Williams, “The Facade of the Palazzo dei ‘Visacci’”, in: I Tatti Studies in the Italian Renaissance V (1993), pp. 209-244: 225. 18 For discussion, see Martin Gaier, Facciate sacre a scopo profano: Venezia e la politica dei monumenti dal Quattrocento al Settecento, Venice 2002, p. 93. 19 For the Latin text and discussion, see Creighton E. Gilbert, L’arte del Quattrocento nelle testimonianze coeve, Florence 1988, p. 101. 12 disapprobation; they had commissioned portraits too lofty for their status in a highly class- conscious society. Figure 2. Alessandro Vittoria, Tommaso Rangone, bronze (Venice, church of S. Giuliano) Renaissance works were also at risk of censure for a lack of skill apparent in their execution. The poet Andrea Michièli, called Strazzola, ridiculed an unspecified painting on this score. His satirical piece has the painting itself say that since it is so poorly painted, no one can restrain himself from laughing.20 Criticism of a portraitist’s skill also came in much gentler forms, and at times from a patron. King Philip II was pleased to accept a splendid portrait by Titian. (Madrid, Museo Nacional del Prado, fig. 3) In 1551, the king wrote a letter referring to several portraits, including a good likeness of himself in armor, but noted that it was made with excessive haste, and if time had permitted he would have asked the painter to redo the work. This criticism was presumably a reference to the loose brushstrokes, typical of late works by Titian but unusual in Spain at this time.21 20 For discussion, see ibidem, p. 78. 21 For discussion, see Miguel Falomir in: Tiziano, exh. cat., ed. by Miguel Falomir, Madrid 2003, p. 218f: and anonymous. 13 Figure 3. Titian Vecelli, King Philip II in armor, oil on canvas (Madrid, Museo Nacional del Prado) An artist’s dissatisfaction with the commission process could also lead to insult, indeed creative mockery, of the patron. In his book on painters, Karl van Mander recounts a delightful example of an artist skewering his patron.22 A Flemish painter carried out a now-lost portrait of an English captain, who then refused to pay for the work. In response, the painter used watercolors to place jail bars in front of the captain and then threatened to put the portrait on public display! After the patron settled his debt, the artist washed out the bars to free the captain from debtor’s prison. On first reading, this story sounds like an invention, and perhaps it was. But given that the account was published as a truthful account, it must have captured elements of the reality of 22 For discussion and original text (from in Karl Van Mander’s 1604 Schilder-boeck) see Lorne Campbell, Renaissance Portraits: European Portrait-painting in the 14th, 15th, and 16th centuries, New Haven 1990, p. 142. 14 the risks that principals faced should they anger their agents.23 Moreover, patrons who read Vasari’s Life of Michelangelo knew how the artist avenged the insults on the Last Judgment, expressed by the Vatican master of ceremonies, Biagio da Cesena, whom Michelangelo placed in hell as Minòs. Biagio complained in vain to the pope, as both Vasari and Lodovico Domenichi recounted in the sixteenth century.24 Obviously, the offended church official wanted the portrait removed; but for that to happen, part of the fresco would have had to be repainted or otherwise destroyed. It survived, and Biagio remains on view in the Sistine Chapel. Certainly, some Renaissance portraits were destroyed. A letter from Ludovico Gonzaga, the marquis of Mantua, tells us that Galeazzo Maria Sforza, the duke of Milan, was so annoyed with a portrait drawing made by Andrea Mantegna that he had the sheet burned.25 Presumably, this lost work failed to flatter the sitter sufficiently. In the same letter, Ludovico faintly praised his court artist, saying that Mantegna was good in many things, but that his portraits could have more grace. Portraits of particular individuals were sometimes removed at a patron’s request from works that were still in the planning stage. When Giorgio Vasari presented the Duke Cosimo de Medici with a finished drawing for Cosimo Planning the War on Siena, intended to go on the ceiling of the Room of the 500, he included a number of counselors. However, Cosimo told Vasari that these figures were not necessary, because he had done the planning alone.26 Vasari then painted the scene without the unwanted advisors. In commissioning a painting, as in planning a banquet, a patron faced the difficult decision of whom to invite. Then the artist, in the spirit of an event organizer, needed to arrange the guests in 23Another example of ridicule, not related to portraits, appears in Franco Sacchetti’s Trecentonovelle, and is repeated Vasari (note 8), II. p. 120f. Giotto refused to paint a shield with the arms of a nouveau riche patron, and instead had his assistant depict pieces of armour. 24 For discussion and sources see Norman E. Land, “A Concise History of the Tale of Michelangelo and Biagio da Cesena”, in: Source: Notes in the History of Art XXXII (2013), pp. 15-19. For another example of artist painting an unflattering portrait to get revenge see Federico Zuccari (1539/40 - 1609) e le vendette d’artista, exh. cat., ed. by Cristina Acidini Luchinat and Elena Capretti, Florence 2009. 25 Rodolfo Signorini, “Federico III e Cristiano I nella Camera degli Sposi”, in: Mitteilungen des Kunsthistorischen Institutes XVIII (1974), pp. 227-250: 232. 26 Randolph Starn/Loren Partridge, Arts of Power: Three Halls of State in Italy, 1300-1600, Berkeley 1992, p. 183. 15 a manner that reflected their status. This process could result in bruised egos, as they did for some observers of Mantegna’s Camera picta (Mantua, Ducal Palace, fig. 4). A letter from the patron, Ludovico Gonzaga, noted that Galeazzo Maria Sforza had objected to the selection of figures.27 Specifically, the duke expressed disapproval because he had not been included in the Camera picta, although Mantegna had represented both the Holy Roman emperor and the king of Denmark. The marquis justified his decision, explaining that he had felt required to include the emperor, who was his superior, as well as the king, his brother-in-law. Most interestingly, he added that since these portraits had been seen by many people, it would be very awkward to remove them. Ludovico claimed to have entertained the possibility of demolishing the two portraits, in order to resolve a diplomatic incident, which suggests that he thought Sforza wanted such an action. Perhaps the duke of Milan just wanted his portrait to be added, but in the end, Mantegna left the fresco unchanged. Figure 4. Andrea Mantegna, Camera picta (detail), fresco (Mantua, Ducal Palace) 27 For the related letters see Rodolfo Signorini, Opus hoc tenue. La camera dipinta di Andrea Mantegna. Lettura storica iconografica iconologica. Mantua, 1985, p. 306f. nos. 23-24. 16 Almost all of the accounts above relate to portraits requested by their patron-subjects that were accepted and put on display, but which nevertheless brought embarrassment upon them. Surely the portraits that were refused, hidden, or destroyed were, on average, judged to be much less satisfactory. Unfortunately, very few of those images have been passed down for posterity, and even our knowledge of them is very limited. Some have suggested that Ludovico Gonzaga was less than sincere when he criticized Mantegna as a portraitist, positing that the marquis of Mantua only wanted to calm down Sforza, his powerful ally in Milan. But Ludovico appears in the Camera picta with rolls of fat on his thick neck, a large fleshy ear, and an inelegant pose, each of which departed strongly from Renaissance standards of male beauty. Moreover, about twenty years later, Isabella d’Este expressed disapproval of Mantegna, and seemingly for similar reasons. In 1493, the marchesa of Mantua decided to exchange portraits with another noblewoman. However, she was dissatisfied with Mantegna’s portrait of her because, she claimed, the portrait was done poorly and did not resemble her at all.28 She did accept a second painting, by Giovanni Santi, but once again claimed that the resemblance was poor. Of a third portrait, by yet a third artist, she complained that it exaggerated her weight. Surely, the artists could have created an accurate image of Isabella. Indeed, Ludovico il Moro, Duke of Milan, even told Isabella that the third portrait was a good likeness, recalling her appearance when they had last met. He noted that the portrait depicted Isabella as looking heavier than he recalled, but added that perhaps she had gained weight since he had last seen her. Isabella evidently disapproved of the portraits because their flattery fell short. Her goal, we can assume, was a portrait comparable to that of King Antigonus, who was blind in one eye, as described by Pliny (Nat. Hist. 35. 85-90). Apelles, the marchesa could have read, painted Antigonus in profile in order to portray the king without revealing any defects. Isabella wanted a portrait that would capture her appearance unflawed yet unmistakable. 28 For the portraits of Isabella by several artists, discussed below, see the discussions in Alessandro Luzio, La Galleria dei Gonzaga venduta al Inghilterra nel 1627-8, Milan 1913, pp. 183-123; and Sally Hickson, “‘To see ourselves as others see us’: Giovanni Francesco Zaninello of Ferrara and the Portrait of Isabella d’Este by Francesco Francia”, in: Renaissance Studies XXIII (2009), pp. 288-310. 17 One portrait of which Isabella did approve was painted several years later by Francia, who had never set eyes on the marchesa. The artist based this work on yet another portrait, together with a verbal description of the sitter. Isabella wrote to Francia that he had made her appear far more beautiful in his painting than nature ever had. But a still further enhancement was desired; she asked the painter to make her eyes a bit lighter. Twenty-five years later, she gave this improved painting by Francia as a model to Titian. Isabella observed that Titian’s portrait was so pleasing that the sitter herself doubted that she had appeared so beautiful at the age in which he depicted her. Titian enjoyed extraordinary success among European rulers. That success stemmed in part from his ability to create flattering images that nevertheless captured sufficient resemblance to be recognizable, a pleasing combination of expedience and skill. The Holy Roman emperor, Charles V, ordered a portrait from Titian of his late wife, the empress Isabel of Portugal.29 Because the subject was deceased, this work was based on an earlier painting, in this instance by a minor artist. In a fascinating letter, the patron asked Titian to retouch the nose, not because Titian had rendered it poorly, but rather because Titian had painted it accurately. Titian complied, and produced an ‘improvement’ on the earlier portrait, giving the empress a straight, idealized nose. Images of the empress made during her lifetime, by contrast, reveal her nose to be aquiline. Titian similarly idealized his representations of the emperor himself. Charles V suffered from an extremely pronounced version of the so-called Hapsburg jaw. Contemporary accounts relate that his chin jutted so far forward that fully closing his mouth was impossible.30 However, Titian’s many portraits of the ruler reveal not a trace of his royal line’s underbite problem. By contrast, Christoph Amberger created a portrait of Charles V closer to reality, to judge from a surviving drawing of the emperor’s skull. But Charles preferred a heroic image, one that sacrificed anatomical accuracy to patronal pride. Charles gave Titian, not Amberger, the extraordinary honor of being the only artist who was allowed to portray his image 29 Miguel Falomir in: Tiziano (note 21), p. 208f. 30 For this problem, and portraits of the emperor, see Diane H. Bodart, “Il mento ‘posticcio’ dell’imperatore Carlo V”, in: Estremità e escrescenze del corpo, ed. by Clelia Arcelli, Florence 2012, pp. 465-483. 18 On occasion, artists idealized to the extreme. In a letter of 1544 about the Medici portraits in the New Sacristy, Niccolò Martelli wrote that Michelangelo did not depict Duke Lorenzo and Giuliano from nature, but showed them with the proportions, decorum, and grace which he thought would bring them more praise.31 Seen against the background of Medieval writers’ expressed disapproval of Renaissance portraits and their abundant penned attacks on the lack of verisimilitude, Martelli’s praise, known from a volume published in 1546, sounds like an attempt to preempt potential criticism of Michelangelo’s sculptures. If so, his attempt was successful. Martelli’s letter seems to have inspired a similar observation by Vincenzo Borghini about Michelangelo’s statues. In his notes on Aristotle’s Poetics (9.1451b), this sixteenth-century scholar wrote that the historian differs from a poet in the same way that a portraitist who works from nature differs from one who makes stylized works. As an example, he cited the marble statues of Giuliano and Lorenzo in the New Sacristy. Borghini describes them as poetic because of their majesty, grandeur, and heroic beauty. Through such embellishments, Michelangelo created images of ideal rulers. We have no record of the Medici objecting to this lack of verisimilitude. A consideration of the motives for patronal disapproval of portraits enables us to surmise the probable reaction of Francesco del Giocondo to Leonardo’s portrait of his wife, (Mona) Lisa. At least three facts about this mysterious work are reasonably clear. 32 First, Vasari stated that it was begun in Florence. Second, Agostino Vespucci, discussed below, described the portrait of Lisa del Giocondo as unfinished in 1503. Third, it is almost certainly the work described as a completed portrait of a Florentine woman by Antonio de Beatis, secretary to Cardinal Luigi of Aragon, when the two visited Leonardo’s studio in France. De Beatis, presumably passing along information provided by the artist himself, reported that the painting had been made for Giuliano de’ Medici, Duke of Nemours, and brother of Pope Leo X. Most probably, but here scholars disagree, Leonardo put the final touches on the portrait in Rome, when he worked for Giuliano 31 For discussion of the sculptures, and the quotes by Martelli and Borghini discussed below, see Jonathan K. Nelson, “Poetry in Stone: Michelangelo’s Ducal Tombs in the New Sacristy”, in: San Lorenzo: A Florentine Church, ed. by Robert W. Gaston and Louis A. Waldman, Florence 2017, pp. 450-480. 32 For a recent summaries of the vast literature on this painting, see Josephine Rogers Mariotti, Monna Lisa: la Gioconda del magnifico Giuliano, Florence 2009 and Rab Hatfield, The Three Mona Lisas, Milan 2014. 19 from 1513 until the death of the Medici nobleman in 1516. The portrait remained with the artist, and Leonardo evidently described it to the Cardinal, who probably had known Giuliano, as a commission from this distinguished patron, not from the obscure Florentine merchant Francesco del Giocondo. Presumably, this would not have pleased the original patron, who remained very much alive.33 The Mona Lisa was criticized by Vespucci in the earliest known reference to the painting. In a volume of Cicero’s letters, where that ancient author had mentioned that Apelles often left his works incomplete, Vespucci wrote in the margin, dated 1503: “So does Leonardo da Vinci in all his pictures, such as, for instance, the head of Lisa del Giocondo.”34 One of the most cited risk factors for all artistic commissions was a failure to deliver on time or even deliver, as Francesco del Giocondo learned. In about 1525, the humanist Paolo Giovio observed that Leonardo completed very few of the many commissions he received, due both to his unstable character and to his tendency to lose interest in his works. Giovio also tells us that Michelangelo, as well, had a reputation for not completing commissions.35 Francesco could have avoided this risk by commissioning a portrait of his wife from a younger artist active in Florence, Raphael. For his most important patrons, at least, this reliable painter from Urbino presented work that earned great praise. Though Michelangelo, according to Giovio and other contemporaries, was often boorish, Raphael, like Leonardo, received praise for his charm. This quality was highly valued for portraits, given that patrons often complained about the boredom of sittings. Raphael, wrote Giovio, earned the respect of the powerful “through careful observance of civilized behavior, no less than the nobility of his works.”36 With Raphael, 33 For the improbable suggestion that Francesco del Giocondo “gave or sold it [the Mona Lisa] to the Magnificent Giuliano, no doubt for the sake of political advantage” see Hatfield (note 32), p. 123; for a similar view, see Rogers Mariotti (note 32), pp. 11f., 20-24. 34 For facsimile, transcription, translation, and discussion of the Latin text see Hatfield (note 32), pp. 106, 203 fig. 54. 35 Paolo Giovio, Scritti d’arte: Lessico ed ecfrasi, ed. by Sonia Maffei, Pisa 1999, pp. 234, 247. 36 For translation and original see John K. G. Shearman, Raphael in Early Modern Sources (1483-1602), New Haven/London 2003, I, p. 807, doc. 1525/15. 20 patrons knew that decorum would be adeptly displayed. He understood well the principal-agent aspect of the patronage tradition. Raphael deftly removed risk from his patrons; accordingly, he was paid more than his contemporaries for comparable works.37 We close our analysis of the risks of ordering portraits by returning to Francesco del Giocondo. If a commissioned painting became one of the most famous in the world, one might assume that the patron began with a high level of satisfaction. Actually, difficulties with his wife’s portrait continually worsened, and for three reasons. First, Leonardo produced a nude version of the Mona Lisa, known as the Mona Vanna; though the face is somewhat different, the pose and overall appearance remain the same. Recent analyses have demonstrated that the finished drawing in Chantilly, perhaps by the master’s hand, served as the prototype for the numerous painted versions.38 Making matters worse for Francesco del Giocondo and his wife, the clothed and unclothed versions were both known in Rome. Indeed, they inspired Raphael to paint not only La Velata, a female portrait clearly based on the Mona Lisa, but also La Fornarina, a version with her upper body unclothed.39 Second, the Mona Lisa and Mona Vanna entered the French royal collection in the sixteenth century, where both of the works were misidentified as a mistress of King Francis I. Presumably these comments responded to the very same features that are celebrated today: Leonardo’s revolutionary decision to show a woman in full face, smiling, and looking straight out at the viewers. The patron had tried to reduce his risk by hiring a highly respected artist, but the very quality and innovation of the Mona Lisa led to a significant reputational loss, at least in the eyes of the patron and his descendants. Some early modern viewers also found that these same features, combined with the lack of a wedding ring (and of clothes, in the Mona Vanna), showed a lack of decorum. 37 Jonathan K. Nelson/Richard J. Zeckhauser, “Raphael, Superstar, And His Extraordinary Prices”, in: Source: Notes in the History of Art XXXVIII (2018), pp. 15-23. 38 See the website of the current (2019) exhibition, la Joconde nue, at Le musée Condé de Chantilly. URL: http://www.domainedechantilly.com/fr/event/la-joconde-nue/ (accessed 5 June 2019). 39 David Alan Brown/Konrad Oberhuber, “Monna Vanna and Fornarina: Leonardo and Raphael in Rome”, in: Essays presented to Myron P. Gilmore, ed. by Sergio Bertelli and Gloria Ramakus, Florence 1978, II, pp. 25-86. http://www.domainedechantilly.com/fr/event/la-joconde-nue/ 21 Third, a recently published but little-known letter records a prurient scandal about Francesco’s wife.40 In 1515, Filippo di Filippo Strozzi wrote that he informed Lorenzo de’ Medici, Giuliano’s nephew in Florence, that Francesco del Giocondo had heard unsettling rumors that both Filippo and Lorenzo had made advances on the merchant’s wife, which Lisa had rebuffed. Evidently, Leonardo’s model was considered to be a vulnerable paragon of beauty. This leads to a new suggestion, namely that the Mona Vanna was painted for the erotic pleasure of Giuliano or his nephew Lorenzo. Francesco and his wife could not possibly have anticipated that the fame of the portrait he commissioned would inspire a topless version, or lead future scholars to pursue a salacious story about her. Such were the risks in commissioning a portrait in Renaissance Italy. 40 See Hatfield (note 32), pp. 129-131. wp_cover_19_024 30June_Risky_Business_final_draft