`Ma il tempo crudele richiude sulla
figura reale della Gioconda le sue
porte infinite che l'arte sola può
sforzare e dischiudere perchè ci si
mostrino i baleni dell'eternità.'
Il Marzocco, 21 December 1913
Frank Zöllner
Leonardo's Portrait of Mona Lisa del Giocondo*
Leonardo da Vinci's portrait of Lisa del Giocondo in the Louvre, the so-called `Mona Lisa' (or `La Gioconda), is arguably one of the best known images in Western culture (Fig.1). Having said that, there are few other incontestable statements to be made about the painting. Not surprisingly, recent publications on portraiture avoid dealing with the `Mona Lisa', because most problems regarding this portrait must still be considered unresolved. The identification of the sitter, traditionally based on Vasari's description of the painting, is still a matter of debate, because Vasari had probably never seen the portrait and his account is contradicted by at least one early source, the so-called Anonimo Gaddiano. Moreover, some scholars consider Vasari's reliability challenged by two additional sources, namely by Antonio de Beatis' account of his visit to Leonardo's studio in October 1517 and by some of Enea Irpino's poems, written c. 1525. De Beatis mentions a woman's portrait Leonardo produced for Giuliano de' Medici and Irpino praises Leonardo's portrait of a woman with a black veil. Both sources have prompted numerous alternative identifications for the Louvre painting. These attempts at alternative identifications and the idealisation of `Mona Lisa' have, in many, if not in most cases, distracted scholarly attention from more profound efforts to understand the portrait itself in its historical context.
The most recent alternative identification has been proposed by Carlo Vecce, who believes that the Louvre painting actually depicts Isabella Gualanda. Vecce's hypothesis is far from unassailable, since the reliability of his major sources - the Anonimo Gaddiano, Antonio de Beatis and Ernea Irpino - can be questioned, and he has no new evidence to corroborate his identification. At the same time, the traditional naming of the portrait as `Mona Lisa' has been reinforced by Grazioso Sironi's and Janice Shell's publication of a document, suggesting that Leonardo's painting of `La Joconda' (or `La Honda') was in Milan as early as 1525. Yet the document, extremely valuable for our knowledge of the early provenance of Leonardo's paintings, does not provide conclusive proof for the traditional naming of the portrait as `Mona Lisa', and it raises a further question about how the painting passed from Milan in 1525 into the collection of Francis I. at some later date. Moreover, some of the suggestions in this important article need reconsideration: neither the year proposed for the portrait's commission, 1500, nor the interpretation of Lisa's predominantly dark garments as indicative of mourning can be accepted. Finally, the authors' assertion that the sitter's identity would `not matter a great deal' is likely to hamper our understanding of the Louvre painting.
In the course of this article, I shall discuss briefly the proposals just mentioned. But chiefly I will present and re-examine the information already known, discuss some new material and concentrate on the following issues: the identity of `Mona Lisa'; reliable information about the sitter and the patron; the precise date, exact circumstances, and most likely motive for the portrait's commission; the domestic environment for which the portrait was created; the pattern of patronage and the tradition of female portraiture in which Lisa's portrait could be placed; and Leonardo's artistic intentions in creating the painting. From what follows, I hope that it will become clear that `Mona Lisa' is the correct identification of the Louvre painting and that our understanding of this portrait can be enhanced considerably with the information discussed.
I. The evidence
Since the beginning of this century, the `Mona Lisa' has acquired the status of an universal icon stripped of its historical context, which seems both to provoke and to defy scholarly attempts to understand it. Yet the painting remains the portrait of an individual person and it is as such that it has been understood in the history of portraiture. Therefore, one would expect that significant consideration had been dedicated to the identity, social status and biography of both patron and sitter. However, as we have seen above, the identity of the sitter is by now regarded a matter of little importance and the available information has neither been presented in the form it deserves, nor has it been discussed thoroughly. It is, therefore, worth recalling the most important early sources: The so-called Anonimo Gaddiano, Giorgio Vasari and three archival documents of somewhat obscure origin.
The Anonimo Gaddiano, writing around 1540, states that Leonardo `portrayed from life Piero Francesco del Giocondo'. This has led some authors to confuse Piero Francesco with Francesco del Giocondo or to assume that there was in fact a portrait of Piero Francesco, which Vasari had misinterpreted as being a portrait of a joyful woman, `La Gioconda'. Consequently, according to a recent article but in earlier contributions as well, the Anonimo Gaddiano is taken to undermine Vasari's reliability. On the other hand, others give the Anonimo Gaddiano's report less credence, probably because nothing else was known of Piero Francesco del Giocondo. This scepticism towards the Anonimo Gaddiano is strengthened by our second source, Giorgio Vasari, who speaks of a portrait not of Piero Francesco del Giocondo but of Mona Lisa del Giocondo and who seems altogether more convincing because of his detailed account of the painting. In fact, he describes the portrait at some length and in great detail, even mentions Lisa's smile, implies that it was done in Leonardo's second Florentine period (i.e. 1500 to 1506) and gives the collection of Francis I. at Fontainebleau as the location of the painting. This information seems to be correct because in c. 1542 Leonardo's `Mona Lisa' and other paintings of Italian artists embellished the `Salle du Bain' at Fontainebleau. The earliest surviving inventory of the French Royal Collections and Cassiano dal Pozzo in 1625 also identify the sitter of the portrait as `La Gioconda'. Vasari's account is again confirmed by at least three of Raphael's early Florentine works of c. 1504 to 1506 (Figs.2-4) because they clearly reflect some of the features of Lisa's portrait, which - as we shall see below - Leonardo executed between 1503 and 1506.
The third piece of information stems from the Le Monnier Vasari edition of 1851, which reports the following biographical data: Francesco di Bartolomeo di Zanobi del Giocondo was born in 1460 and held various Florentine public offices in 1499, 1512 and 1524. Before his marriage to Mona Lisa in 1495 he had already been twice married, in 1491 to Camilla di Mariotto Rucellai and in 1493 to Tommasa di Mariotto Villani. He died in the plague of 1528. The editors of the Le Monnier edition fail to give their source, and the information seems to surface here for the first time; up to now its origin has remained obscure.
The fourth piece of historical evidence about Mona Lisa and her husband is first reported by Eugene Müntz, who writes that a daughter of Francesco del Giocondo died on the first of June 1499 and was buried in St. Maria Novella. Müntz indicates the `Libro dei morti' in the `Archivio di Stato' as his source but admits that he never saw it.
The fifth item of information can be found in some publications by Giovanni Poggi, though the most important of these, `Il Marzocco' of December 1913, has become difficult to find today. Poggi indicates the dates for Mona Lisa`s birth (1479) and for her marriage (1495) and also provides the full name of her father, Antonmaria di Noldo Gherardini, who lived in the quarter of St. Spirito and who in 1480 lists among the members of his family his one-year-old daughter Lisa. Poggi gives the `portata del catasto' of 1480 as source for his information.
Archival research confirms most of the information summarized above and reveals more important data. These data prove the first published source, the Anonimo Gaddiano, who speaks of a portrait of Piero Francesco del Giocondo, to be wrong. In fact, Piero Francesco del Giocondo was Lisa's first son, born on 23 May 1496, namely, fourteen months after her marriage with Francesco del Giocondo. Thus, Piero Francesco was only seven or eight years old when Leonardo's activity for Giocondo began c. 1503 - hardly the right age for a child of a middle-class background to be portrayed. Obviously, the Anonimo Gaddiano was mistaken and we can now reconstruct how he erred: Most likely, Piero was not the sitter of the portrait but the source of information for the Anonimo Gaddiano, who had misunderstood the information that Piero, the first son of Mona Lisa del Giocondo, had given to him. In any case, the documentary evidence discredits the Anonimo Gaddiano, so he can not be taken as a witness for Vasari's unreliability.
Vasari was not only a biographer but also a creative writer and his reliability is therefore more difficult to assess. For example, his story about the musicians who kept Lisa smiling sounds like a appealing literary invention and his description of her facial features is at least partly a mixture of fantasy and literary commonplaces. However, he did know some details of Lisa's portrait, probably from artists returning from the French court and from members of the Giocondo family (see below). Both his general knowledge of artistic activities in the first half of the 16th century in Florence and his standard of information about Leonardo's second Florentine period are quite good. For example, his somewhat anecdotal story about Leonardo's complaints that the `Signoria' had paid him for his work on the `Battle of Anghiari' in small currency (quattrini) finds confirmation in other sources. An equally high degree of accuracy is found in Vasari's account of Raphael's artistic relations with Leonardo and in the fact that he did not repeat the Anonimo Gaddiano's earlier confusion about Leonardo's portrait of Piero Francesco del Giocondo. Indeed, Vasari lived in Florence for long periods between 1524 and 1550, he frequently stayed in the Medici palace, close to Francesco's home in the Via della Stufa (see below), and he was acquainted with two of Francesco del Giocondo's cousins. Considering his attempts to gain first-hand information from Florentine citizens, one can assume that he had known both Lisa and her husband Francesco who - according to unpublished documents - died in spring 1539 when he was almost 80 years old. Lisa herself seems to have lived at least until 1551. Thus Vasari's description of Lisa's portrait, written before the death of Francis I. in 1547, may well have been based on first-hand information obtained from Lisa and Francesco del Giocondo.
The biographical data about Francesco del Giocondo as reported in 19th-century Vasari editions stand in need of some emendation. Francesco held four public offices (not three, as has been previously believed), may be considered a responsible citizen, albeit not a leading figure in Florentine government (as some scholars have stated). He was married three times and had three sons and one daughter. His first wife Camilla di Mariotto Rucellai bore his first son Bartolommeo in February 1491 (new date 1492). His two sons by Lisa were Piero (or Pietro), the one mentioned by the Anonimo Gaddiano, and Andrea, born on 12 December 1502. Francesco did not die in 1528 but in 1539, two years after having made his testament in 1537 - which is of some significance for Vasari's reliability (see above). The fourth piece of information about Mona Lisa's daughter's date of death should be corrected from 1 June to 6 June 1499. Finally, Poggi's reference to the `catasto' of 1480 is accurate, and, following this source, we get some insight into the economic situation and the social status of Lisa's family. Her father, Antonmaria di Noldo Gherardini, who filed his tax declaration in the quarter of St. Maria Novella, drew a modest income from some small farms and farmlands in the vicinity of Florence. He owned a country house (casa signorile) in St. Donato in Poggio, a small village about twenty miles south of Florence, and a house in the city near St. Trinita. However, in 1480 Antonmaria and his family were compelled to reside in rented accommodation in the quarter of St. Spirito, because their own house in the city was damaged (rovinata) and therefore unfit for habitation. We may therefore infer that Antonmaria did not have the necessary means to restore his city property.
The `catasto' of 1498 reveals that, at some point between 1480 and 1498, Antonmaria Gherardini moved his city residence to the Via dei Buon Santi, `popolo' of San Simone, in the quarter of St. Croce. Here again they do not live in their own property, but rent half a house for the discreet sum of 11 florins annually from Lionardo Tomaso di Busini, who lives in the other half of the building. The same tax declaration contains an entry regarding a financial transaction consequent on Lisa's marriage: At the beginning of March 1495 (modern date) Lisa Gherardini marries Francesco del Giocondo and Antonmaria's dowry for his daughter consists in a piece of farmland near his country house and 170 large gold florins. Again, this is rather modest, in particular if one considers the importance of a dowry in those days and compares the amount to the financial agreements for other marriages. For example, in 1504 Maddalena Doni had a dowry of 1400 florins, and Francesco del Giocondo in his testament of 1537 states that the daughter of his son Piero, Cassandra, had a dowry of 1440 florins.
The documents disclose a straightforward Florentine marriage of typical middle-class citizens. Equally typical is the fact that before the marriage Francesco was acquainted with the Gherardini family: Lisa's father, Antonmaria Gherardini, had married as his second wife Caterina di Mariotto Rucellai who was the sister of Francesco del Giocondo's second wife Camilla di Mariotto Rucellai. Despite this connection, the slight economical discrepancy between the Gherardini and the Giocondo families should be noted. Francesco del Giocondo, who is called `civis et mercator florentinus' in the notarial document regarding the marriage, came from a better-off Florentine silk and cloth merchant family whereas his wife Lisa, whose father is only `civis florentinus' and `sanza esercjzzio igniuno' (in the `catasto' of 1480), had a more modest economic background. In Western societies, slight economic and social disparities between groom and bride are as normal as the circumstances that the husband is older and that he has been married before. Thus, contrary to popular belief, there was nothing peculiar about Lisa's status or her marriage. On the other hand, the rather modest circumstances of the matrimony may allow us to speculate about the personal relationship between Francesco and his wife. Given the importance of marriage and dowry in Renaissance Florence, it is worth noting that Francesco does not seem to have made significant political or economical gains from this marriage. We might even conclude that he married Lisa for genuine affection and that this affection also had some bearing on his decision to have Leonardo portray his wife.
II. The date of the commission
As noted above, the period between 1503 and 1506 as approximate dates for Lisa's portrait are drawn from Vasari's account and from Raphael's early works in Florence. Any later date, such as 1513, can be excluded once one agrees with the traditional identification of the portrait's sitter as Lisa del Giocondo. Documents, however, both for the commission of the painting and for subsequent payments, if any, are lacking and it is likely that a contract never existed for a minor commission such as a private portrait; neither could there have been significant payments, because the painting was never delivered. Despite this lack of direct documentation, the particular circumstances and the date of the commission can be reconstructed by piecing together Vasari's account, documents regarding Francesco's family and the records for Leonardo's activities during his second stay in Florence.
Leonardo's second stay in Florence lasted from April 1500 to May 1506, with an interruption from mid-June 1502 until the end of February or the beginning of March 1503, when he travelled as an architect and engineer with Cesare Borgia. Recently it has been suggested that Leonardo began Lisa's portrait immediately on his arrival in Florence, that is, in late April 1500. However, this date can be excluded for various reasons: First, on two occasions Padre Pietro da Novellara, answering Isabella d'Este's inquiries if Leonardo had done any paintings recently, gives a detailed account of the paintings present in the painter's studio. In spring of 1501 he describes the St. Anne cartoon and the `Madonna of the Yarnwinder', but he does not mention another painting which could be identified with Lisa's portrait. On the contrary, he explicitly states that Leonardo had not done anything else. Second, as we have seen above, Vasari had first-hand knowledge of Lisa's portrait and he implies that it was begun in 1503. Third, Raphael's drawings after the `Battle of Anghiari' and the `Leda' reveal an intimate knowledge of Leonardo's works. These drawings as well as three other works from this period, which allude to some features of Lisa's portrait, can be dated between 1504 and 1506. In fact, as can be deduced from a preparatory drawing in the Louvre (Fig.2), in 1504 Raphael experimented with flanking columns for the portrait of Maddalena Doni (Fig.4). Those columns also appear in his portrait of a `Lady with the Unicorn' (Fig.3) from roughly the same period. Such flanking columns (with almost identical column bases) seem to have been part of an earlier idea for Lisa's portrait (probably in the form of a cartoon) because vertical slices of those columns still appear at both sides of the panel, and early copies of the portrait still show them. At some point Leonardo must have decided to paint only those vertical fragments rather than the half columns which are known from Raphael's early Florentine works and from copies of Lisa's portrait. Therefore, one can assume that Raphael was present when Leonardo developed and changed his ideas for the flanking columns between 1504 and 1506. Fourth, in the period from 1500 to 1502, before serving Cesare Borgia, Leonardo would hardly have accepted a portrait commission. He was staying at the SS. Annunziata, accommodated and supported by the Servites while he was at work on the Saint Anne cartoon and on the `Madonna of the Yarnwinder' for Florimon Robertet, treasurer to the king of France. Throughout this time, Isabelle d'Este kept hounding him for her portrait, albeit without success. Therefore it seems reasonable to conclude that Leonardo would not have agreed to do a private portrait at this time. On the basis of this argument we may propose that Leonardo could have begun Lisa's portrait at the end of February or at the beginning of March 1503, after he returned from his travels with Borgia. He must have stopped working on it at the beginning of June 1506, when he left Florence for Milan (though he may have finished the painting later, in particular the background). The resulting period could have been at the most three years and four months and it probably was even shorter (see below). It does not correspond exactly to the four years mentioned by Vasari, although it is a tolerable inaccuracy, assuming that Vasari calculated a period of four years following Francesco and Lisa, who may have indicated to him 1503 and 1506 as the relevant dates (counting 1503 through 1506 yields four years).
This reconstruction so far is based on incomplete documentation (e.g. Isabella d'Este may have written letters now lost). It can, however, be usefully checked against the only complete set of documents for the period in question: Leonardo's bank account in the Ospedale di S. Maria Nuova. This documentation supports Vasari's information regarding Leonardo's stay with the Servites, because until November 1501 Leonardo did not withdraw any money from his bank account. We may thus infer that Leonardo had no need to draw off his savings, because he was getting room and board from the Servites. However, this situation changed noticeably in March 1503, after he had left Cesare Borgia's service. Indeed, he was probably earning nothing at all, and in this period Isabella d'Este - at least for some time - had even stopped the requests for a portrait. At this point, Leonardo started to withdraw 50 gold florins about every three months. From these frequent withdrawals in spring and summer 1503, one might hypothesize that Leonardo was not very busy in spring 1503, and was therefore willing to do a private portrait for a Florentine citizen. In fact, his next and much larger commission to follow, the mural of the Battle of Anghiari, began officially in October 1503 and the only employment Leonardo seems to have had prior to this date was his survey of the Arno near Pisa in July 1503. Thus the commission for Lisa's portrait could have been agreed on between the end of February or the beginning of March and July 1503. Following this line of thought, one can further conjecture about the reasons why the painting was not finished before Leonardo's departure for Milan in June 1506. With the start of his work on the `Battle of Anghiari', Leonardo's withdrawals became less frequent and stopped altogether between May 1504 - when the contract was signed and the first substantial payment of 35 gold florins had arrived - and February 1505. Furthermore, from December 1503 onwards, documents about Leonardo's work on the Battle of Anghiari become more frequent and by then he must have been quite busy with the larger commission, leaving less time for Lisa's portrait. Naturally, from October 1503 onwards, he concentrated on the Battle of Anghiari as the more important commission which - according to the contract - would secure regular monthly payments of 15 florins for at least a year. Obviously, an individual portrait like Lisa's would yield a much smaller profit, and the prestige of a small private commission could not match his involvement in an ambitious project for the Florentine government. Thus Leonardo had good reason to proceed slowly on Lisa's portrait, leaving it in a state he considered unfinished. Finally, in spring 1506, the legal dispute about additional payments for the Madonna of the Rocks and the prospect of better employment at the French court took him to Milan. When he returned to Florence for brief periods in 1507 and 1513, there was not much point in reassuming work on Lisa's portrait: as so often before, Leonardo had other things on his mind and the probable motive for the commission had lost its validity. For the rest of his life he carried Lisa's portrait with him, just as he did with other paintings.
III. Motives for the portrait's commission
In the fifteenth and sixteenth centuries the individual portrait of a middle-class person more often than not had a particular purpose. Unfortunately, in many cases we do not know the circumstances for a portrait's commission because most sitters remain unidentified. But where biographical information is available, one can investigate the circumstances of the portrait's commission, which might explain some of its features. For example, Lisa's daughter's death in June 1499 has been suggested as a possible explanation for her dark veil; according to this suggestion, is based on a mistaken interpretation of one source, Lisa's dark veil and dress indicate the state of mourning for her deceased daughter. However, the portrait was commissioned in 1503, when the normal period of mourning must have been over (see above). But even if Leonardo had begun the painting in April 1500, when Lisa could have been in mourning, one has to ask whether there would have been much sense in portraying a grieving woman only two months before the end of the usual period of mourning? Other considerations speak against this hypothesis as well: Given the high mortality rate for children in the Renaissance, it would not have been customary to go into mourning for a deceased child, nor would one want to commemorate this event with the portrait of a grieving woman. We know of no other Florentine portraits from that period of a mother in mourning for her child, and furthermore, a dark veil does not necessarily indicate mourning. And Lisa's smile and her low-necked dress would seem to deny it!
Still, examining the circumstances of the commission and looking for a motive for the portrait could be particularly fruitful in our case, because some historical information about patron and sitter is available. Francesco del Giocondo does not give the impression of a man who would arbitrarily express wishes for works of art (such as: "there is this famous painter, let's have him paint my wife!"). Indeed, he was almost certainly unlike Isabella d'Este, who at any possible moment wanted a portrait or, at some point, even a completely different work from Leonardo. For all the documents tell us, Francesco was a dedicated citizen, who fulfilled his public responsibility in some official government functions. He married more or less within his class, and in his third marriage, a woman known to his family. In his testament, he emerges as a person who cares for his family and who seeks to arrange his affairs in an orderly way. Therefore, we can assume that he would commission a portrait for some reason or at least under particular circumstances. Two standard motives for a portrait's commission, namely marriage and death, do not fit the chronology of the painting. Yet, in the period when Leonardo started to paint, in spring 1503 - after he had returned again to Florence - there was a suitable moment, because a date in this period happens to coincide with an important event for the Giocondo family: On April 5 1503, Francesco bought a house for his own use in the Via della Stufa, next door to the old family home in the same street. The reason for buying the new asset may have been the opportunity to purchase a house in the neighborhood of the old property (as was common in Renaissance Florence), but the birth of the third son in December 1502 may also have been an additional motive. In any case, this acquisition must have been an important step for the family because before April 1503, Francesco, his wife and his children had lived in the old house next door, probably together with other members of the family. Not until spring 1503, did he have a separate home for his own family and set up a new household. In Renaissance Florence, the establishing of a man's new household quite frequently marked an important occasion to purchase furniture and to commission works of art, since it required a substantial rearrangement of a family's environment. We may, therefore, understand Francesco's acquisition of a new home and his establishing of his own household as relevant circumstances for the commission of Lisa's portrait.
As I have speculated above, Francesco's affection for Lisa may have had something to do with his wish for her portrait. Though this must remain a hypothesis, yet another important event for the Giocondo family could make the commission more understandable: On 12 December 1502, Lisa's second son Andrea was born, thus in spring 1503 she was a young mother who had successfully delivered her second son a few months earlier. Lisa's third pregnancy could have provided an additional stimulus for the commission because as an event it was more significant than one would normally expect: Two and a half years earlier, in June 1499, Lisa had lost a baby daughter, and roughly ten years earlier, at some point in 1492 or 1493, Francesco's first wife, Camilla di Mariotto Rucellai, had died not long after giving birth to their first son Bartolomeo. At some point in 1494 his second wife, Tommasa di Mariotto Villani, had also died at an unknown date within only a year of the wedding. The high mortality of women and children in childbirth in those days and the rather rapid death of his two wives, each within roughly a year after marriage, may justify the assumption that both Camilla and Tommasa had died in the course of delivery or only a few months later. Therefore, the happy outcome of birth was particularly worth commemorating for Francesco del Giocondo, who was painfully aware of the frequently tragic outcome of childbirth.
The establishing of a new household or the birth of a child are worthy motives for a portrait's commission and would fit patterns of patronage in the late 15th and early 16th centuries in Florence. Numerous works of art were commissioned in relation to marriage, pregnancy, birth and fertility. These could have been `cassone'-paintings with secular motives, birth trays, religious paintings with subjects such as the `Holy Family' and also portraits. An allusion to pregnancy or to the importance of fertility and childbirth, for instance, can be found in Botticelli's and Raphael's portraits of pregnant or newly married women. Botticelli's painting shows a pregnant woman, thus alluding to her pregnancy, but also, more generally, to virtue with the column behind the sitter. Raphael's portrait of `La Gravida' of c. 1506 shows the sitter's condition, but also her being married - indicated by two rings on her left hand - and her piety - displayed by a devotional book under her right hand. In Raphael's portrait of Maddalena Strozzi (married to Agnolo Doni in 1504) of c. 1505-1506, the imagery on the back of the panel alludes to the importance of birth for the founding of a family and stresses also the wish for the conception of a first-born son. Thus, by the beginning of the sixteenth century, fertility, pregnancy, childbirth and motherhood could have been significant themes in individual portraits of women. These themes can be placed within the more general framework of increasing private patronage in Renaissance Florence where one of the major issues of society, the family, played an important role. Consequently, quite a few of the prominent commissions had to do with the social status, economic affairs and fortune of the families involved, as, for example, Michelangelo's `Holy Family', which was commissioned for the wedding between Agnolo Doni and Maddalena Strozzi in January 1504. Francesco del Giocondo had Leonardo portray his wife in the same period and in exactly the same cultural context. It therefore is not accidental that Lisa's portrait was the model for Raphael's portrait of Maddalena Doni in some formal aspects. In fact, the formal relationship between the two paintings (Figs.1 and 4) is matched by a link between the Giocondo and Doni families; they are both more or less the same class and Maddalena's brother, Marcello Strozzi, had business connections with Francesco del Giocondo.
After having discussed the precise date and the exact circumstances of the commission for Lisa's portrait, the relationship between the patron and the painter of Lisa's portrait remains to be examined. Some information about Francesco's role as a patron of the visual arts is already given by Vasari who, apart from Lisa's portrait, also mentions a painting of St. Francis by Domenico Puligo and a fresco with a `storia de' Martiri' by Antonio di Donnino (or Domino) Mazzieri, both commissioned for Francesco's family chapel in the choir of the SS. Annunziata. Francesco's testament of January 1537 and an inventory from SS. Annunziata of 1521 provide more information. In his testament, Francesco explicitly mentions the decoration for the `capella Martirum' in SS. Annunziata (though without giving the name of the artist) which suggests that Francesco at some earlier date had transferred the family's burial place from Santa Maria Novella to SS. Annunziata. This transfer had probably taken place in 1526 when Puligo painted a new altarpiece for the same chapel, showing Francesco's patron St. Francis of Assisi receiving the stigmata. Indeed, in the same year, the whole chapel seems to have been refurbished, which makes sense at this point, because in 1526, Francesco was about 65 or 66 years old and thus at about the right age to take care of his burial place. Another reason for the new decoration may have been the death of Francesco's third son Andrea, who is documented only until 1524 and mentioned as deceased already in the testament. Thus again, as in the case of Lisa's portrait, we have commissions related to particular family circumstances.
Francesco's rather thoughtful interest in art is confirmed by an inventory of the SS. Annunziata of 1521. This inventory with its roughly eighty entries records the belongings of `maestro' Valerio who had died in January of the same year. It lists household items such as a bed, a bedstead, blankets, mattresses, carpets, tablecloths etc. and about two dozen small works of art. One carpet, a pitcher, some tablecloths and practically all art works or items of artistic value were given to Francesco. Among these are a cartoon with four figures, a head of St. Sebastian, a Virgin Mary, a sculpture made of clay with the Madonna and Child, a Mary Magdalene, a Madonna with six saints, one St. John the Evangelist and another St. John with a dead Christ and one St. Michael painted onto the bedstead. The estimated values for the art works are low, mostly between 2 and 15 lira each. Generally, art collecting may have had some of its origins in similar arrangements for the distribution of an estate, though Francesco seems not to have had a collector's approach. As can be inferred from a note at the end of the inventory, `maestro' Valerio owed money to Francesco del Giocondo and the items given to him were supposed to compensate Valerio's debts. Some of the works seem to have been unfinished, and there is even a panel prepared to be painted on (`una tauola dapignersi di braccia 3'). Because of this panel and the unfinished works, one is tempted to assume that `maestro' Valerio had been an artist, but apart from these few unfinished items Valerios's list reads like a typical Florentine inventory of that period. In fact, some twenty small works of art including two crucifixes and a painting on the bedstead could be found in many Florentine homes around 1520. Thus Francesco, apart from his wish to collect some outstanding debts, must have desired the artistic items as a common embellishment for his own household or for the homes of his children (and in fact, two items, a pitcher and a Madonna with six saints, were given to Francesco's daughter). This seems to be fairly reasonable, because Francesco was not a poor man and, therefore, must have been more interested in art than in second-hand household goods such as used mattresses and old furniture. Moreover, the relatively small items could be moved easily.
The historical evidence about Francesco makes him appear to be fairly well-off with slightly above average aspirations as a patron, putting him into the class of more important families such as the Doni or the Strozzi. The inventory, again, shows the down-to-earth nature of his interest in works of art. It further indicates his very close links with the Servites of SS. Annunziata, which becomes interesting if one considers the fact that Leonardo had stayed with the Servites prior to his commission for Lisa's portrait. The first contact between Leonardo and Francesco del Giocondo may have taken place in the SS. Annunziata.
IV. The setting and content of Lisa's portrait
At the beginning of the 16th century, no work of art was absolutely independent from tradition and from the requirements of its particular commission. As we shall see, Lisa's portrait is no exception to this rule. Thus, her smile and her sitting above a landscape which extends distantly in the background are references to earlier portraits or to other works of art. Lisa's general position recalls Flemish models and in particular the vertical slices of columns at both sides of the panel had precedents in Flemish portraiture. The allusions to Flemish portraiture are obvious, but by 1503 a reference to Flemish art was no novelty. In fact, Flemish paintings were particularly fashionable in the seventies of the 15th century, and in 1503 in Florence Flemish elements in Lisa's portrait probably seemed rather old-fashioned to a contemporary beholder. However, Flemish portraits combined a high degree of realism with piety, and this reference to piety in particular might have been attractive to Florentine patrons in the difficult years after Savonarola's death.
Lisa's portrait is considerably larger than its Flemish predecessors and in this it can certainly be considered a novelty. The unusually large dimensions put Leonardo's painting into a class of grand Florentine portraits such as Botticelli's `Giuliano de' Medici' of 1476 or 1478, Piero Pollaiuolo's `Galeazzo Maria Sforza', painted in Florence in 1479, Botticelli's so-called `Simonetta Vespucci' with its famous antique cornelian `Apollo and Marsyas' from the Medici collection, and Domenico Ghirlandaio's `Giovanna degli Albizzi', wife of Lorenzo Tornabuoni. The large dimensions of these portraits commissioned for or by the Medici, Sforza and Tornabuoni reflect the importance of these families and of their households. Francesco del Giocondo's family was somewhat less important by comparison and, therefore, the rather grand dimensions of his wife's portrait may have been an attempt to approach the social rank of Renaissance Florence's leading families. The large size of `Mona Lisa' may also suggest that Francesco del Giocondo had commissioned this portrait for a particular room in the newly-acquired house. If so, this would be in keeping with the wide-spread practice of commissioning works of art for a specific domestic setting or even for a particular room where the wall-space for a painting was already prepared. Around 1503 in many cases, a portrait's place seems to have been the chamber (`camera') or the adjacent ante-chamber of a house, where it was surrounded chiefly by religious or devotional imagery. Our knowledge about this kind of environment is still limited, but some preliminary suggestions about the characteristic location for portraits may be made. For example, in the Medici inventory of 1492 men's portraits, sometimes in the company of other men's portraits, could be found in settings together with both religious and secular works of art. Women's portraits, on the other hand, shared the company of portraits (other than their husband's) or other secular paintings to a lesser extent; they are more frequently surrounded by religious works and seem to be hung in more private spaces. This, for example, was the case with the portraits of Alfonsina Orsini (wife of Piero de' Medici) in the Palazzo Medici and of Giovanna degli Albizzi. The location of women's portraits in an environment often dominated by religious imagery needs more investigation, but it may be one reason why `Mona Lisa' or other examples from the same period are similar to paintings of the Virgin Mary. Thus a formal reference to images of the Virgin in a portrait could correspond to a particular domestic setting. More generally, this reference concurs with the moral demands on women, who were expected to make the life of the Virgin Mary normative for their moral conduct.
The representation of morally sound social conduct seems to have been an objective in the portraiture of women. In many portraits of that period inscriptions and attributes symbolize moral demands on women or express individual virtues. Among such symbols were a crystal pitcher (generally for virtue), a rosary (devotion), the prayer book (piety), or more exotic items such as a unicorn (chastity or virginity). A striking characteristic of Lisa's portrait is the lack of such meaning-laden attributes, and only her hands seem to articulate something like virtue. In fact, by 1503 in Florence hands as such - if not holding something or in a gesture of prayer or devotion - were not a common feature of women's portraits. To include them in this painting must have been an intentional allusion to moral demands on women, because exactly this particular way of the right hand resting upon the left was recommended as the appropriate gesture for girls or young women to show their morally sound social conduct.
The position of Lisa's hands is not exactly a symbolic expression (comparable to an attribute such as the unicorn) but rather a gesture taken from the requirements in real life. Its `symbolism', if we want to call it such, is therefore a more `realistic' reference to virtue. The same is true of Lisa's dark veil, which need not have signified mourning. Black was not reserved exclusively for cases of grief, and black silk, for example, was the recommended material for the `first nuptial dress.` A dark veil, moreover, was a standard garment of every-day use, covering the hair and indicating the wedded state of women and the desired virtues associated with it, such as chastity, devotion and obedience to God. For this reason, until relatively recently, married women were obliged to wear black veils in church. Thus, the veil was a garment closely related to women's social conduct and to an ideal of virtue - much in the same way as a modest style of dressing was recommended for any virtuous woman. We could, therefore, regard the presentation of Lisa's hands, of her dress and of her black veil as a subtle reference to female virtue and social conduct.
Still, a black veil in a Florentine portrait is certainly unusual and therefore Lisa's veil demands an explanation which goes beyond the general reference to virtue and social conduct. Such an explanation may be found in contemporary fashion; in fact, right at the beginning of the sixteenth century black or dark clothes were `en vogue' and considered a sign of splendor and dignity. This fashion, which originated in Spain, was inaugurated in Italy and most prominently displayed at the wedding of Lucrezia Borgia and Alfonso d'Este in 1502. We can assume that all persons involved in the commission - Leonardo, Francesco del Giocondo and his wife Lisa - had heard of those fashionable and noble black garments. Indeed, in the years preceeding his commission for Lisa's portrait, Leonardo had made the acquaintance of both Lucrezia Borgia's brother (Cesare) and Alfonso d'Este's sister (Isabella). Francesco del Giocondo, as a merchant of silk and cloth, would have been aware of recent trends in fashion, and we can also assume that his wife Lisa - as most women - had heard of those fashionable garments. Thus Lisa's black veil and the predominantly black or dark colors of her dress may have been inspired by a fashion inaugurated a year prior to the commission of her portrait. One should perhaps also consider whether Lisa's darkish garments were intended to avoid a then-topical conflict between the regulations for women's dress on the one hand and fashion on the other. In the 15th and 16th centuries, a modest style of dress was recommended for every woman and sumptuary laws strongly suggested modest garments in order to guarantee the morally sound appearance of the female citizenry. Considering this situation, Lisa's black veil and darkish outfit may have been a compromise between the current requirements for women's dress, a personal wish for expressing her virtue and her desire to be dressed fashionably. Indeed, in her portrait, Lisa could have looked both fashionable and virtuous. We can accordingly propose that general notions about women's dress and virtue, as well as fashion and possibly her own taste, influenced the appearance of Lisa's portrait.
V. The painter's aspirations
In modern literature, the unusually high artistic achievements of Leonardo's portrait of Lisa del Giocondo are taken for granted and the painting has been taken as an ideal image or an ideal portrait. This may be a valid observation in a formal sense and it certainly is not surprising because many Renaissance portraits are idealized. Lisa's portrait seems, in fact, idealized because her facial features resemble a female type realized by Leonardo in other paintings such as the Louvre Saint Anne. Also the lack of nuptial rings on her left hand and the somewhat un-real landscape background suggest an idealization. Moreover, Leonardo himself recommended painting portraits under the softening effects of twilight and this practice certainly helped to idealize Lisa's features. However, the particular function of idealizations and of idealized beauty in a woman's portrait created for a particular domestic setting and for the expectations of the patron need still to be examined.
In the late fifteenth and early sixteenth centuries, beauty, idealized or not, was closely connected with virtue, and at least in Neo-platonic terms, idealized beauty as a mental concept was even preferable to real beauty. Thus a beautiful appearance could have been considered a morally significant notion which was connected with virtue. In Leonardo's portrait of Ginevra Benci (Fig.5), for example, the motto on the back of the panel reads `VIRTUTEM FORMA DECORAT' (Fig.6), which in the context of the painting can be understood as `beauty embellishes virtue'. Thus the inscription links the sitter's beauty to virtue which is symbolized by laurel and juniper, painted on the back of the panel. A similar case is a portrait medal of Giovanna degli Albizzi (Tornabuoni) where beauty and true love are related to chastity. The same link between beauty and the virtue of chastity was celebrated in the `Festa del Paradiso', organized by Leonardo on the occasion of Gian Galeazzo Sforza's wedding with Isabella d'Aragona.
One important point of Ginevra Benci's portrait is its representation of virtue on two different levels: the visually more appealing level is the expression of virtue by means of the beauty depicted, but to articulate this link between beauty and virtue more explicitly, Leonardo added (or was asked to add) both the inscription and the attributes on the back of the panel. If Leonardo wanted to eliminate the visually less appealing level of inscriptions or attributes, he had to visualize the beauty of virtue by purely pictorial means. We can, indeed, assume that in 1503 Leonardo's ideas about the art of painting - as for example verbalized in the `Trattato della pittura' - were too sophisticated to condone traditional devices such as inscriptions and conventional attributes. Moreover, in his second Florentine period, Leonardo had an incentive to surpass the achievements of earlier portraiture and to prove his creative powers. After the long interval of relatively steady employment at the Milanese court, he tried to re-establish himself in his hometown where he faced competition from some major artists. In order to re-enter business he may have felt the need to show his talent - as he obviously did with the Saint Anne cartoon. Consequently, both Leonardo's style of drawing and his way of drawing figures change dramatically during his second Florentine period. In a period of transition and new challenges it is therefore likely that Leonardo would attempt to express his particular artistic ambitions in a private portrait. He also had a strong personal interest in physiology and facial expression (physiognomy) which is most impressively documented in his so-called `grotesque heads'. We can imagine that he felt a challenge to exceed the expressive qualities of earlier portraits, both his own and those of other painters. This challenge virtually existed in Domenico Ghirlandaios's portrait of Giovanna Tornabuoni (neé degli Albizzi) of 1488, probably the most important example of late 15th-century Florentine portraiture (Fig.7). On Ghirlandaio's panel, which is of almost exactly the same size as `Mona Lisa', an inscription explicitly states that the portrait's only shortcoming was the unresolvable challenge to depict the moral conduct or virtue and the soul of a person (mores animumque):
O art, if thou wert able to depict the conduct and the soul,
No lovelier painting would exist on earth.
Ghirlandaio's portrait of Giovanna Albizzi was probably known to Leonardo who had contacts with the Albizzi family in 1503, and portraits, being located in the chamber of a house, were to some extent accessible. In any case, the general question whether an artist could or could not render visible heavenly things (such as a soul and its virtue) seems to have been an issue in the 15th century. Leonardo must have felt qualified to answer this question, because he had a profound anatomical knowledge and because he believed he had understood how the soul would expresses itself through a person's body and face. With this physiological knowledge, he could respond to the challenge posed by Ghirlandajo's portrait of Giovanna degli Albizzi, trying to demonstrate how painting could imitate both the material appearance of a human person and its immaterial qualities. Thus with Lisa's portrait, Leonardo demonstrated both the highest achievements of painting and the most ambitious aspirations of a painter. Moreover, this demonstration was not a means in itself (as most scholars have it) but conceived to satisfy the main object of woman's portraiture: the presentation of virtue, in this case created for a domestic setting where virtue should ideally be at home.
VI. Conclusions
Considering the available information about the Giocondo family, Lisa's portrait need not be regarded a mysterious image that has irreversibly lost its historical context. The circumstances of the portrait's commission do, in fact, fit neatly into the context of Florentine patronage and portraiture at the end of the 15th and beginning of the 16th centuries. Lisa's portrait continues - though in a somewhat more sophisticated way than preceeding examples - the tradition of women's portraiture celebrating female virtue and beauty. Moreover, also a particular type of patronage, situated within the framework of Florentine family matters and linked with the devotional surroundings of the domestic setting, can be assumed for the commission of Lisa's portrait. Finally, Leonardo's very distinct artistic aspirations, at this crucial point in his career in 1503, and his response to the challenge of earlier portraiture, found their way into Lisa's portrait. All these points constitute a perfectly reasonable cultural context for Lisa's portrait. We could, therefore, read this portrait as a typical painting, which only romanticism has taught us to perceive as being enigmatic and transcending human comprehension. Consequently, I would like to suggest that a further appreciation of Leonardo`s portrait of Lisa del Giocondo should be considered within the historical context which I have tried to outline in this article.
LIST OF ILLUSTRATIONS
1. Leonardo da Vinci, Portrait of Lisa del Giocondo, Paris, Musée du Louvre, Phot. Musées Nationaux
2. Raphael, Drawing for a Portrait of a Young Woman, Paris, Musée du Louvre, Phot. Musées Nationaux
3. Raphael, Lady With a Unicorn, Rome, Galleria Borghese, Phot. Soprintendenza di Beni Artistici e Storici, Rome
4. Raphael, Portrait of Maddalena Doni, Florence, Palazzo Pitti, Phot. Soprintendenza di Beni Artistici e Storici, Florence
5. Leonardo da Vinci, Portrait of Ginevra de' Benci, Washington, National Gallery of Art, Phot. National Gallery
6. Leonardo da Vinci, Portrait of Ginevra de' Benci, verso, Washington, National Gallery of Art, Phot. National Gallery
7. Domenico Ghirlandajo, Portrait of Giovanna degli Albizzi, Lugano, Collection Thyssen-Bornemisza, Phot. Collection Thyssen-Bornemisza.
NOTES
* In writing this article I have enjoyed the advice and received the help of many friends and colleagues. In particular I would like to thank Richard Brilliant, Helga Carl, Anna Colliva, Gino Corti, David Franklin, Amanda Lillie, Michael Lingohr, Alessandra Malquori, Michael Rohlmann, Christof Thoenes, Matthias Winner, W. J. Wegener and the staff of the `Archivio di Stato' in Florence, and here in particular Sandra Marsini. Abbreviations used in the following are ASF for `Archivio di Stato' and BNF for `Biblioteca Nazionale Centrale', both in Florence.