Annunciation Triptych (Merode Altarpiece), ca. 1427–1432
Workshop of Robert Campin (South Netherlandish, active by 1406, died 1444)
South Netherlands (modern Belgium), Tournai
Oil paint on oak; Central panel 25 1/4 x 24 7/8 in. (64.1 x 63.2 cm); each wing 25 3/8 x 10 3/4 in. (64.5 x 27.3 cm)
The Cloisters Collection, 1956 (56.70)
All too often, museum visitors make the assumption that the paintings in our galleries are unchanged from the day that they were created hundreds of years ago. This is not the case, and there are distinctions to be made between the various types of changes that have occurred over timebetween those that have happened naturally and those that have resulted from human intervention. The assessment of a given alterationbe it an addition or a subtractionoften necessitates a reconsideration of the painting's history, which in turn, can reveal helpful information about the function and importance of the work in its own time. This essay will discuss alterations found in Netherlandish paintings dating from the fifteenth and sixteenth centuries in The Metropolitan Museum of Art.
The presence of halos as identifying appurtenances of holy figures in early Netherlandish paintings is as rare an occurrence as it is commonplace in Italian paintings of the same period. To assess whether halos are original to a picture or added later, close visual scrutiny of the painting's surface is essential. Netherlandish panels in which the authenticity of halos has been brought into question include the Portrait of a Carthusian (49.7.19) and A Goldsmith in His Shop (Saint Eligius?) (1975.1.110), both by Petrus Christus (active by 1444, died 1474/76), and the Portrait of a Young Man by Hans Memling (active by 1465, died 1494) (1975.1.112). The halos in the works by Christus were questioned by Max J. Friedländer as early as 1916 and by William Martin Conway in 1921. In preparation for the 1994 exhibition Petrus Christus: Renaissance Master of Bruges, the issue of the halos was again raised. They were determined through technical examination to be later additions, and they were removed. With the removal of the halos, the aesthetic intentions of the artist were restored. Christus was among the first early Netherlandish painters to break through the barrier of the plain, dark background that was conventionally employed in portraiture by providing an illusionistic space to surround the figures. The addition of the halos, by contrast, introduced an element that forced the viewer to focus on the foreground, discouraging further investigation of the space beyond the picture plane. The restoration thus allowed for a renewed discussion of the function and meaning of the paintings. Neither panel was originally conceived as a religious image; the false halos had altered their intended function as secular portraits.
The identification of the Carthusian in Christus' portrait as a saint had long been doubted, as there was no known Carthusian lay brother who rose to sainthood in the south Netherlands around 1446, when the painting was made. (The sitter is not a monk, as he has no tonsure, and is not, by the rule of the order, clean-shaven.) Perhaps it was Bruno (ca. 10301101), the founder of the Carthusian order, formally canonized in 1623, whom a later owner of the painting wished to identify by the addition of the halo. The alteration may have been made in Spain; the panel was in the collection of Don Ramon de Oms, viceroy of Majorca by 1911, and two nineteenth-century copies of the painting were known in Valencia collections.
In the Portrait of a Goldsmith (1975.1.110), the removal of the halo similarly prompted discussion about the identity of the central figure. The sitter had been thought to represent Saint Eligius, the patron saint of goldsmiths, and until 1998 the picture was known by that eponymous title. That year, Martha Wolff and Hugo van der Velden each noted, independently, that the painting did not conform to any of the standard representations of the saint, which show him performing miracles, and they argued convincingly that the identification of the sitter with Saint Eligius occurred in modern times. Early mentions of the painting refer to it only as a representation of a goldsmith in his shop. It was not until 1817 that the subject of the picture was described as "a goldsmith or rather the patron of goldsmithsSaint Eligius." It may have been at this time, when the painting also received a new gold frame, that the halo was added. Van der Velden even proposed an alternative identification for the sitter, namely, Willem van Vlueten, who is documented as having paid for his citizenship in Bruges in 1433 and later rose to prominence as a celebrated goldsmith at the court of Duke Philip the Good. If Van der Velden is correct, the portrait would be among the earliest and largest professional portraits known in early Netherlandish painting. This would, in turn, certainly call for a reassessment of the development of early portraiture.
Memling's Portrait of a Young Man (1975.1.112) was considered, when first mentioned in 1857 by Gustav Waagen, as a Saint Sebastian, since the figure then appeared with two added features-a halo and an arrow held between the thumb and forefinger of his left hand. The landscape view through the window remarkably appears in two Italian paintings of about 1480, a Self-Portrait by Pietro Perugino (active by 1469, died 1523; Galleria degli Uffizi, Florence) and a Virgin and Child from the circle of Andrea del Verrocchio (possibly Domenico Ghirlandaio [1449-1494]); Musée du Louvre, Paris). The quotation indicates that the panel must have traveled to Italy soon after it was painted. This would explain the Italianate foreshortened oval form of the halo, traces of which may still be seen when viewing the picture in raking light. By the time Philip Lehman acquired the painting in 1915, both the halo and the arrow had been removed in a cleaning and restoration of 1912 in London. Evidence of this is visible in a stripped-state photo.
Other examples of alterations appear to be roughly contemporary with the date of the painting, perhaps requested by the first or a subsequent owner to bring the painting up-to-date or to render it more meaningful. How else may we explain the addition of the arched latticework and grape arbor to Hans Memling's Virgin and Child with Saints Catherine of Alexandria and Barbara (14.40.634)? There is no underdrawing for this feature, as there is for the rest of the work. Close examination of the surface reveals that the greens used for the arbor and employed for the trees in the background are of different values and intensities. Furthermore, the brushwork used to describe the leaves in the arbor is broader and more descriptive of nature than are the schematic strokes that characterize the stylized trees and bushes of the background landscape. Because the paint film for the arbor appears, under high magnification, to be very old and integral with the original paint layers, indicating that it was added very early on to the painting, another explanation must be sought.
Although the identity of the hand responsible for the arbor is difficult to determine, the probable impetus for the alteration may perhaps be more readily explained. Grapes are a well-known symbol for the Eucharist and for the divine salvation that was bestowed through Christ's sacrifice on the Cross. The iconography of the grapes is especially significant in its connection to Memling's hometown of Bruges, where, still today, a relic of Christ's blood is housed in the Basilica of the Holy Blood. The Holy Blood had been granted papal justification by Pope Sixtus IV (r. 147184) in 1467 and 1472, which gave rise to a popular devotion in Bruges and elsewhere. The addition of the arbor to Memling's Virgin and Child with Saints Catherine of Alexandria and Barbara is evidence of this devotion. Whether the male donor, shown kneeling at the left of the assembly of holy figures, requested the alteration himself or whether it was made by a later owner is not known. But the fact that it appears in an early (probably sixteenth-century) copy of the painting (Galleria dell'Accademia, Venice) suggests that it was added not long after the completion of the painting.
Several early Netherlandish paintings record through their alterations the most significant events of lifemarriage, birth, and death. The Annunciation Triptych (56.70), dating to ca. 1425, is one case in point. Sometime after the completion of the left wing, two figures were added, presumably when the donor, Peter Engebrechts (shown kneeling), was married. The two figures are the town messenger at the back gate and Engebrecht's new wife. It was common practice for noble families thus to document weddings and births, adding portraits of new family members. The wings of an altarpiece by the Master of the Saint Barbara Legend (32.100.56a) acknowledge the deaths of four members of the guild, lay confraternity, or civic group that commissioned the triptych. Small crosses were later painted in above the praying hands of these kneeling donors, who may have died in an outbreak of the plague of 148990.
Examples such as these are readily explained; others are more problematic, and a word of caution is advised. It had been maintained that on the wings of Gerard David's Nativity triptych (49.7.20a-c), the attributes of the pig and of the wheel and sword were later additions to the original work, made in order to transform the kneeling donor figures into saints. This could have occurred when the triptych was acquired by a collector for whom the identities of the donors were unknown and for whom their conversion into saintsspecifically, Saints Anthony Abbot and Catherinehad particular personal relevance. The explanation is logical, but the conclusion was contradicted by the results of technical analysis. Indeed, these findings call for a reconsideration of the period and context in which the triptych was made. Under the high magnification of a microscope, it is very clear that the image of the pig, while painted over the black coat of the donor, is integral with the original paint layers and was thus part of the initial composition. The instance of the female donor even more clearly demonstrates the evolution of the picture. X-radiography shows that an area was left in reserve for both the wheel and the sword, indicating that these traditional attributes of Saint Catherine were from the outset intended to be associated with the figure. Catherine's crown was also part of the original plan, although it has been restored. Unfortunately we do not know the identities of the donor figures, but it is likely that they are an Anthony and a Catherine who wished to model their daily lives after those of their name saints.
The representation of mortals as saints, reflecting a desire to emulate the lives of the saints and to follow their sacrifices and devotion to Christ, becomes increasingly apparent in the early sixteenth century. The practice reflects an extension of the beliefs of and widespread adherence to the Modern Devotion, a popular reform movement begun by Geert Grote in the late fourteenth century and initially practiced by the Brothers of the Common Life. This religious movement encouraged an individual to imitate Christ by identifying with his life and sufferings, renouncing the world, and embracing virtue. The more readily identified examples of this phenomenon in painting are those depicting royalty or nobility. Margaret of Austria (14801530) and Prince Juan (14781497) were portrayed as Saints Margaret and John in one panel of a lost diptych documented in Margaret's inventory of 1516. Juan de Flandes (active by 1496, died 1519) most likely intended the same royal pair to be identified as the couple celebrating their marriage in his intimate Marriage Feast at Cana (1982.60.20). Margaret of Austria also had her portrait painted as Mary Magdalene (e.g., Mary Magdalene, ca. 151020, by the Master of the Magdalene Legend, Staatliche Galerie, Schleissheim, on loan from the Alte Pinakothek, Munich). The Virgin in Michel Sittow's (14691525) Virgin and Child, ca. 151518, in Berlin (Gemäldegalerie), and the Magdalene in his Mary Magdalene, ca. 1518, in the Detroit Institute of Arts, may be portraits of Catherine of Aragon. Jan Gossart (active by 1503, died 1532) also used disguised portraiture in his paintings. For Anna van Bergen (14921541), wife of Adolf of Burgundy (ca. 14891540), marquis of Veere, identification with the Virgin Mary as the exemplar of ideal motherhood had a very personal significance. According to Karel van Mander, in his Schilderboeck of 1604, Anna and her son were the models for Gossart and his workshop in their production of the Virgin and Child, ca. 1525, of which multiple versions were made (17.190.17).
Occasionally, the close scrutiny of early Netherlandish paintings yields quite another findingthat portraits of figures, and sometimes figures in their entirety, were reworked or initially painted by a different hand. Here the physical evidence hints at developments in workshop structure and procedure, and perhaps also at contractual stipulations. The observation that the Virgin's head is clearly the work of a different or superior hand in several panel paintings and manuscript illuminations of the same period (that is, the late fifteenth and early sixteenth centuries) has often led to the conclusion that they were overpainted at a later date. However, technical examination has in some cases proven otherwise. With regard to the production of panel paintings, we know that they were either part of series of like images offered on the open market or the collaborative effort of a team of artists on a large-scale work commissioned for export. In each case, the wishes of the client and the specific requirements of the contract undoubtedly came into play.
Unfortunately, relatively few contracts survive. However, of those that do, occasionally certain ones stipulate that the heads and other parts of a painting must be painted by a hand superior to the rest of the work. On August 31, 1487, the church masters of Saint Bavo in Haarlem commissioned Mourijn and Claas van Waterlant to paint additional portions of an altarpiece on which they had already worked, and to be assisted in their work by "as good a master as can be found in Holland" for some faces and other parts. An additional document concerning a lawsuit against Albert (or Aelbrecht) Cornelis (active in Bruges before 1513, died 1531) may be helpful to consider. It pertains to Cornelis' only documented painting, the Coronation of the Virgin (Groeningemuseum, Bruges), which was commissioned by the Guild of Saint Francis in 1517 for their chapel in the Church of Saint James in Bruges. The contract stipulates that Cornelis himself must paint the main parts of the image, including the faces and hands of the figures. In the suit brought against him, Cornelis was accused of subcontracting portions of the painting. In his defense, he argued that his sole obligation was to paint only the faces within a designated period of time. The suit was resolved, but not without Cornelis having to rework some sections of the painting that had been originally subcontracted.
From this brief inquiry into the practice of intentional alterations in early Netherlandish painting, it is clear that identifying such changes is only the first step toward understanding the meaning and function of a work of art in its own time. Technical examination, comparative looking, and a close reading of period contracts, guild regulations, and workshop practices are also necessary to place the work in its proper context. Toward this end, it becomes increasingly apparent that, ultimately, it is only through an interdisciplinary approach that new discoveries can be made.
Ainsworth, Maryan W. "Intentional Alterations of Early Netherlandish Painting". In Heilbrunn Timeline of Art History. New York: The Metropolitan Museum of Art, 2000–. http://www.metmuseum.org/toah/hd/neth/hd_neth.htm (January 2008)
Related exhibitions and online features
These related Museum Bulletin or Journal articles may or may not represent the most current scholarship.