National Gallery of Canada / Musée des beaux-arts du Canada

Bulletin 22, 1973

Annual Index
Author & Subject

Bartolommeo Veneto and His Portrait of a Lady

by Creighton Gilbert

Pages  1  |  2  |  3  |  4  |  5  |  6  |  7  |  8  |  9  


13 Giovanni Morelli, loc. cit.; E. Michalski, Zeitschrift für bildende Kunst, vol. LXI (1927-1928), p. 280.

14 One of these, the Fitzwilliam portrait of "The Labyrinth Man," was mentioned above; its stylistic date must be about 1506-1508 in any case, and the strange costume may be explained as a device of the Gonzaga, the family ruling Mantua, whose Marchioness was by birth a member of the ruling Este family of Ferrara. The second is the fine drawing in the Galleria Estense, Modena, where the bulk of the collection comes, as its name implies, from the same Ferrarese ruling family. The third is an unpublished portrait of a cardinal, identifiable as Cardinal Ippolito d'Este by the coat of arms on the back. A copy of this, without the coat of arms, has been published as a work of Lotto by Bernard Berenson, Lorenzo Lotto (London: Phaidon Press, 1956), pl. 63.

15 One portrait of this group, the Houston Young Man, has a now indecipherable date that has been read both as 1512 and 1520. Though 1520 has been chosen more often, 1512 was the earlier reading, and also seems preferable because in 1520 Bartolommeo was painting works in an entirely different style, notably the dated Girl with the Lute. If that gives a date for the group, a location in Padua can be deduced from the portrait now in Budapest inscribed with the name, city, and age of the sitter, Girolamo Dondi dell' Orologio of Padua, aged twenty-four. It was wrongly attributed to Bartolommeo by Hevesy (loc. cit.) and should instead be recognized as a poor copy of the Rome portrait of this group. The face and even the costume folds are tracings, while the hands and landscape have been redesigned. It is curious that Hevesy linked it, rather shakily, to other works of Bartolommeo as of the same sitter at a later date, and not to its direct prototype in Rome, Bartolommeo's most famous work, which thus acquires a name for its sitter and a probable place of execution.


After this article had been completed, it was possible for me through the courteous help of Dr Lucio Grossato to study manuscript histories of the Dondi dell'Orologio family. The most significant of these, Notizie storiche riguardanti la famiglia Orologio, Padua, Biblioteca Civica, Ms. 283, which summarizes public documents from 1374 to 1633 and was evidently composed just after the latter date, includes just two references to the name Girolamo. These are in 1503, when a family member so named is mentioned as the future heir of an uncle and this might have been a child, and 1509, when he appears on a civic list. The ordinary likelihood that these are about one and the same man is enhanced by the fact that the name was not used by the family; it would be remarkable if it produced two Girolamos in one decade and otherwise none in centuries. This person, as the subject of Bartolommeo's portrait when twenty-four, thus is happily consistent with the stylistic proposal above that he was painted about 1512. A later date (possible in a different view of Bartolommeo's style) is excluded by the documentary dates; an earlier date (possible on a documentary basis) is excluded by the stylistic place of the work in Venetian painting.

16 Usually the last record of Bartolommeo's life is thought to be of 1530 (apart from the erroneous reference to 1555 already discussed). This 1530 terminus is based on two paintings, the Mentmore Lady and the Portrait of Lua vico Martinengo in the National Gallery, London, both with inscriptions that have regularly been read as containing that date. The latter picture shows the date in two lines, in the form "MDXXX/XVI ZUN," which since Cavalcaselle first discussed it has been read without argument as "June 16, 1530" until recently, in the National Gallery's catalogue of sixteenth-century Venetian pictures, Cecil Gould proposed to read it "June, 1546" on the ground that the costume is in a fashion worn at the latter date, of which he gives an excellent comparative example. That would extend Bartolommeo's working life sixteen more years. Yet the reading 1530 is a much better fit for a number of other kinds of evidence; indeed it is remarkable how various elements, independent of each other, all seem to lead us to 1530, and can only be re-read as 1546 by accepting the occurrence of distinctly less usual events. A very simple element is that in this period, just as today, it is much less common to give dates with the month and year only (as "June, 1546") than with the day, month, and year (as in "June 16,1530"); thus the other eight dates written by Bartolommeo on panels include none of the former style but two of the latter. More weightily, there survive six other inscribed works by Bartolommeo in the ten years 1520-1530, four with dates; it is obviously easy to accept a seventh of 1530, but a strain on probability to suppose that he then worked fifteen more years from which the surviving similarly inscribed works are zero, followed by just one in the sixteenth year 1546. In addition, Bartolommeo constantly changed his style under eclectic influences at intervals of five years or so, about five times up to 1530, when indeed he was using the style in which he painted this portrait; it is very hard to suppose, then, that when he was about fifty, he shifted his habits to maintain one style for sixteen years, this picture being the sole evidence. It is the odder that when we accept 1546, we have to accept these latter alternatives cumulatively - he must have written the less common date form and by accident all the late works are lost and he shifted his work patterns in middle age, the less likely option coming up right every time - and it is stranger still that the series of more likely options we must consistently reject do not merely diverge from 1546 but fit the same wrong answer, of 1530. Historians do not like to exclude solutions that several independent types of evidence support, and this is probably why Gould's point has been greeted with doubt. (See the reaction of Bassi, in the entry in Dizionario biografico degli Italiani, loc. cit.)

One certainly wants more data. Thus it was most generous of Dr Camillo Boselli, at my request, to survey archival and early printed sources in Brescia - the town fifty miles east of Milan where the Martinengo were an active aristocratic family - to search for the birth date of Ludovico, the sitter of this portrait.  It is usual to accept age of sitters as good evidence for dating their portraits, and this sitter appears to modern observers about twenty or even less (an inscription once added to the surface and now cleaned off said he was 26; it might have been based on family records). Dr Boselli found four members of the family named Ludovico in the sixteenth century, all in the early part, the two youngest being born in 1506 and 1509. The latter was assumed to be the sitter by two recent contributors to the Storia di Brescia (Brescia: 1963-1964 [vol. II, pp. 312, 981]) with whom Boselli concurs. This finding again gives the most natural kind of support to the date 1530, when the sitter would be close to twenty just as we would have wished to find him, and this time it is even difficult to find a strained explanation to fit 1546. It is not plausible to propose a fifth Ludovico of whom no records have survived; this was an important family, and for one of its male adult members to appear in no records is as little possible as it would be for a baronial family in Tudor England. One could suggest a posthumous portrait; that is, if the sitter died at twenty, he could be painted at that age later, in 1546. (It seems to be accepted that a man who died at a more advanced age would not be commemorated with a portrait showing him as he was at twenty.) But the posthumous portrait idea can be excluded in the case of Ludovico born in 1509, who was still living in 1546. It is barely possible for Ludovico born in 1506, who is last recorded alive in 1534, and who died sometime between then and a record of his widow in 1568. Our portrait could then be a posthumous portrait in 1546 of this Ludovico, if we suggest that he died in the earliest of the thirty-four possible years, and also take it that he is meant to look twenty-eight, contrary to our visual impression, and if we add this to our other series of less likely but adopted options. This chain of less likely or almost unbelievable circumstances which we must accept will recall the type of "unfair" detective story, in which the author's explanation, through improbable coincidences etc., serves to prevent readers from solving the puzzle. On the other hand, if any part of the less likely series is disallowed, the painting is of 1530.

To all this of course there is the proper answer that improbabilities must be accepted because the wearing of a 1546 fashion in 1530 is an impossibility. And yet it seems to me that we cannot be quite so confident about the close dating of costumes, even though this is a helpful technical aid in art history which Mrs Newton's own work has much advanced by making a great collection of dated paintings. The inexactitude comes in as the costumes compared are not identical but simply similar, to an extent which is as debatable as the other kinds of evidence, so that the close dating also must be left a little flexible. Mrs Newton, thus, in the Ottawa portrait, is rightly careful to offer a five-year span for the costume element she considers most helpful (i.e., most subject to change) and notes that five years earlier still an "early form of the type" is found. We need only hypothesize a matching "late form" five years after it, or the appearance of the form prior to dated examples of it, or delays in fashions reaching other regions, or a costume element less helpful because less rapidly changed - all unstrained possibilities - or a combination of some of these factors each accounting for a part of the time spread, to arrive at a sixteen year wearing period for Ludovico Martinengo's type of suit, and it would then not require us to vote for the other improbable circumstances. (I am much indebted to Mr Gould for discussing these factors with me.)

Next PageNote 17

  |  2  |  3  |  4  |  5  |  6  |  7  |  8  |  9

Top of this page

Home | Français | Introduction | History
Annual Index | Author & Subject | Credits | Contact

This digital collection was produced under contract to Canada's Digital Collections program, Industry Canada.

"Digital Collections Program, Copyright © National Gallery of Canada 2001"