National Gallery of Art

The three panels in the National Gallery of Art collection (this work, Joachim and Anna Giving Food to the Poor and Offerings to the Temple, and The Presentation of the Virgin in the Temple) form part of a larger series of scenes from the childhood of Mary, of which a fourth component is also known: Joachim Leaving Jerusalem now in the Keresztény Múzeum at Esztergom in Hungary [fig. 1]   [fig. 1] Andrea di Bartolo, Joachim Leaving Jerusalem, c. 1400/1405, tempera on panel, Keresztény Múzeum, Esztergom, Hungary.[1]  [1]
No. 55.148; 46 × 34 cm. The painting, its provenance unknown, was purchased by Cardinal János Simor, perhaps in Rome, for the museum in Esztergom between 1867 and 1878; see Miklós Boskovits et al., Christian Art in Hungary: Collections from the Esztergom Christian Museum (Budapest, 1965), 52. Its original gold ground was evidently removed at an early date, and it, like the Washington panels, was regilded during a nineteenth-century restoration. Since two of the episodes, Joachim and Anna Giving Food to the Poor and Offerings to the Temple and Joachim Leaving Jerusalem, are seldom found represented in art, there are good reasons to assume that the sequence would have originally comprised at least four other, more commonly illustrated scenes, namely, the Expulsion of Joachim from the Temple, the Annunciation to Joachim (and / or to Anna), the Meeting at the Golden Gate, and the Betrothal of the Virgin.[2]  [2]
On the role of Joachim in Italian cycles of the childhood of Mary, cf. Jacqueline Lafontaine-­Dosogne, Iconographie de l’enfance de la Vierge dans l’Empire byzantin et en Occident, 2 vols. (Bruxelles, 1964–1965), 2:154–159; Gertrud Schiller, Ikonographie der christlichen Kunst, 6 vols. (Gütersloh, 1966–1990), 4, pt. 2: 38–54. These (and possibly other) scenes would have accompanied a central image of the Madonna and Child,[3]  [3]
As an example we may cite the vita-icon, name piece of the Master of San Martino in the Museo Nazionale in Pisa, in which the main image at the center is flanked on either side by six superimposed stories from the life of the Virgin. See Enzo Carli, Il Museo di Pisa (Pisa, 1974), 41–43 and fig. 48. or of Saint Anne with the Madonna and Child,[4]  [4]
Cf. the retable of Bernat de Puig in the Museu Nacional d’Art de Cata­lunya, Barcelona, repro. in Josep Gudiol and Santiago Alcolea i Blanch, Pintura gótica catalana (Barcelona, 1986), fig. 57. a Coronation of the Virgin,[5]  [5]
As in the polyptych commissioned from Bartolo di Fredi for the church of San Francesco at Montalcino; cf. Gaudenz Freuler, Bartolo di Fredi Cini: Ein Beitrag zur sienesischen Malerei des 14. Jahrhunderts (Disentis, 1994), 188–221. or a theme such as the Annunciation or Pentecost.[6]  [6]
The Annunciation is placed at the center of an illuminated page with scenes from the childhood of Mary, executed by the Bedford Master, in a Book of Hours in the Oesterreichisches Nationalbibliothek in Vienna, cod. 1855; Gertrud Schiller, Ikonographie der christlichen Kunst, 6 vols. (Gütersloh, 1966 – ​1990), 4, pt. 2: pl. 467. The scene of Pentecost occupies a central position in the cycle of frescoes by the young Bartolo di Fredi in the church of Sant’Agostino at San Gimignano. See Gaudenz Freuler, Bartolo di Fredi Cini: Ein Beitrag zur sienesischen Malerei des 14. Jahrhunderts (Disentis, 1994), 32 – ​45. It is difficult, therefore, to reconstruct the dismembered and dispersed altarpieceAltarpiece
 An image-bearing structure set on the rear part of the altar, abutting the back of the altarblock, or set behind the altar in such a way as to be visually joined with the altar when viewed from a distance. It is also sometimes called a retable, following the medieval term retrotabulum. The altarpiece was never officially prescribed by the Church, but it did perform a prescribed function alternatively carried out by a simple inscription on the altarblock: to declare to which saint or mystery the altar was dedicated. In fact, the altarpiece did more than merely identify the altar; its form and content evoked the mystery or personage whose cult was celebrated at the altar. This original and lasting function influenced the many forms taken by the altarpiece throughout its history. —Alexander Nagel, Grove Art © Oxford University Press of which our three panels would have formed part, also because we do not know exactly how the surviving scenes from the life of the Virgin were related to the main image of the altarpiece. The fact that the grain of the wooden support is vertical would seem to exclude the proposition that they are fragments of a predellaPredella
 A horizontal band, cut from a single plank, below the main panels of an altarpiece. The appearance of the predella can be seen as part of the development of the altarpiece from a single panel to a large, multilevel polyptych. The small figures or scenes painted on the predella formed part of the integrated program of the altarpiece, providing a visual commentary on the major images above and at the same time physically raising the main panels, thus improving their visibility. —Ronald Baxter, Grove Art © Oxford University Press,[7]  [7]
On the structural characteristics of the predella in late medieval altarpieces in Tuscany, see Monika Cämmerer-­George, Die Rahmung der toskanischen Altarbilder im Trecento (Strasbourg, 1966), 93−94; Christoph Merzenich, Vom Schreinerwerk zum Gemälde: Florentiner Altarwerke der ersten Hälfte des Quattrocento (Berlin, 2001), 55. and the hypothesis advanced in the past, that our panels could be fragments of a reliquary cupboard, seems to have no foundation.[8]  [8]
Cesare Brandi formulated this hypothesis, and various scholars accepted it, in Cesare Brandi, Quattrocentisti senesi (Milan, 1949), 243. They could have been fragments of a vita-icon type panel, with a whole-length figure of the Madonna and Child flanked by a vertically arranged series of scenes of her life.[9]  [9]
Joanna Dunn of the National Gallery of Art conservation department tells me that, judging from the cracks lining up, it seems a “strong possibility” that The Nativity of the Virgin was placed above The Presentation of the Virgin in the Temple. However, the x-radiographs do not show the wood grain clearly enough to prove this. Such an image, however, would appear decidedly old-fashioned in Siena after the mid-fourteenth century. A more likely alternative format is suggested by the cases in which Sienese painters of the late fourteenth century, such as Taddeo di Bartolo or Andrea di Bartolo himself,[10]  [10]
On the typology of vita-icon in Tuscany, see Victor M. Schmidt, “Tipologie e funzioni della pittura senese su tavola,” in Duccio: Siena fra tradizione bizantina e mondo gotico, ed. Alessandro Bagnoli et al. (Cinisello Balsamo, Milan, 2003), 541 – ​544; Rita Sauer, “ ‘. . . so be ye holy in all manner of conversation’: The Vita Retable,” in Kult Bild: Das Altar- und Andachtsbild von Duccio bis Perugino (Petersberg, 2006), 131 – ​176. For vita-icons with horizontally arranged scenes, see Taddeo di Bartolo’s panel San Geminianus Enthroned and Eight Stories of His Legend, or Andrea di Bartolo’s panel with the Crucifixion at the center and eight scenes from the life of Christ and of saints to the sides, formerly in the Stoc­let collection in Brussels. Cf. Raimond van Marle, The Development of the Italian Schools of Painting, vol. 2, The Sienese School of the 14th Century (The Hague, 1924), fig. 354; Pierre Bautier, “I primitivi italiani della collezione Stoclet a Bruxelles,” Cronache d’arte 4 (1927): 315 and fig. 6.
produced paintings in a form similar to thirteenth-century dossals, with a large-scale representation at the center, flanked by narrative scenes in two superimposed orders [fig. 2]   [fig. 2] Reconstruction of a dispersed altarpiece by Andrea di Bartolo (color images are NGA objects): a. Joachim and Anna Giving Food to the Poor and Offerings to the Temple; b. Expulsion of Joachim (?), lost; c. Joachim Leaving Jerusalem (fig. 1); d. Annunciation of the Birth of Mary (?), lost; e. Lost; f. Meeting at Porta Aurea (?), lost; g. The Nativity of the Virgin; h. The Presentation of the Virgin in the Temple; i. Marriage of the Virgin (?), lost (see also Reconstruction). A round-arched termination, enriched on the inside with cusped moldings, would be very appropriate for this kind of altarpiece. In Siena in this period, and in Tuscany in general, the wood grain of the support in a vertical panel is invariably aligned vertically, and in a horizontal panel horizontally. It cannot be excluded, of course, that the painting was realized during the artist’s stay in Venice or in the Marche and not in Siena, as Laurence Kanter, in correspondence, suggests. He points out that carpentry practice in the Venetian territories frequently aligned panels parallel to the shorter axis, so in the case of a horizontal altarpiece the wood grain would run vertically.[11]  [11]
Taddeo di Bartolo’s San Geminianus dossal (see note 10 above) has been transferred to a modern panel support with a vertical wood grain and so cannot be adduced for comparison. He further notes that the incised profiles of the original frame moldings on our panels argue for a Venetian provenance. In Tuscany, engaged frames were applied before the panels were gessoed or gilt. In Venice, they were added afterward, and their profiles are often found inscribed on the picture surface as a guide to the painter.

The scenes from the life of the Virgin painted by Andrea are based on an apocryphal text called De Ortu Beatae Mariae et Infantia Salvatoris, attributed to the evangelist Matthew. Later sources enriched this narrative with additional episodes. According to the legend, the marriage of Joachim (father of the Virgin Mary) and Anna remained childless for many years, a state that was interpreted by the high priest of the temple in Jerusalem as punishment for grave sins. Therefore, Joachim’s offering of a sacrificial lamb was rejected, and he was expelled from the temple. The scene represented in this work is usually identified as Joachim and the Beggars but refers instead to a previous episode in the life of Mary’s parents. A version of the legend, evidently familiar in Tuscany, recounts that Joachim and Anna lived in a particularly charitable way, dividing all their worldly goods into three parts: a third was allocated to the poor, another third to the temple, and only a third was kept for their own needs.[12]  [12]
The narrative of the Pseudo-­Matthew was supplemented with other episodes by a German priest named Wernher in the mid-twelfth century. His Driv liet von der maget (Three Books on the Virgin) is the first source to mention the story of the division of Joachim’s worldly goods for charitable purposes; see Gertrud Schiller, Ikonographie der christlichen Kunst, 6 vols. (Gütersloh, 1966 – ​1990), 4, pt. 2: 54 – ​55; W. J. Hoffman, “Wernher,” in Marienlexikon, ed. Remigius Bäumer and Leo Scheffczyk, 6 vols. (St. Ottilien, 1994), 6:716 – ​717.
In the panel at the Gallery, we see, to the left, Joachim distributing loaves of bread to the poor, while his wife is presiding over the delivery of sacks of grain to the temple, where a priest receives them. This episode would have been followed by the lost scene of the priest’s rejection of the offering of a sacrificial animal and the Expulsion from the Temple, the premise for Joachim’s Abandonment of the City, which is described in the painting now in Esztergom.

At this point in the sequence, other episodes usually illustrated in cycles of the childhood of Mary are likely to have followed: namely, the Angel’s Annunciation of the Birth of Mary both to Joachim and to Anna, and the Return of Joachim to the City, linked with the Meeting of Husband and Wife at the Golden Gate. In the following scene of the Nativity of the Virgin, Andrea faithfully followed the model proposed by his father, Bartolo di Fredi, in the cycle of frescoes in the church of Sant’Agostino at San Gimignano and elsewhere:[13]  [13]
The scene frescoed by Bartolo in Sant’Agostino at San Gimignano was repeated with small variations by the same painter in the predella fragment, part of the altarpiece commissioned from Bartolo for the church of San Francesco at Montalcino and now in the local Museo Civico. See Gaudenz Freuler, Bartolo di Fredi Cini: Ein Beitrag zur sienesischen Malerei des 14. Jahrhunderts (Disentis, 1994), figs. 34, 37. The young Andrea probably collaborated in the execution of this part of the Montalcino altarpiece. in the foreground at the center we see a young woman seated on the ground, supporting with one arm the newborn child who stands on her lap, back turned to the viewer, while another woman, also crouched on the ground, is gesturing with both hands towards the child, as if inviting the baby girl to come to her arms. Further in the background we see two standing women: one is just entering the room through a door in the rear wall, bearing a bowl of food in her hands; the other is pouring water into a basin for the child’s mother to wash her hands. Anna is shown reclining on the skillfully foreshortened bed to the right, its curtain drawn back. On the other side of the scene, Joachim and another elderly man are seated in a barrel-vaulted loggia adjacent to the room of the childbirth, awaiting news of the event. The following scene, The Presentation of the Virgin in the Temple, also resembles the corresponding fresco by Bartolo in San Gimignano,[14]  [14]
Reproduced in Gaudenz Freuler, Bartolo di Fredi Cini: Ein Beitrag zur sienesischen Malerei des 14. Jahrhunderts (Disentis, 1994), fig. 27. but in this case both paintings reveal the influence of a celebrated prototype frescoed by Pietro and Ambrogio Lorenzetti on the facade of the Ospedale della Scala in Siena.[15]  [15]
This lost cycle on the façade of the Ospedale della Scala is now known only from descriptions in the sources. Various scholars have proposed the involvement in it not only of the Lorenzetti brothers but also of Simone Martini (Sienese, active from 1315; died 1344); cf. Daniela Gallavotti Cavallero, “Pietro, Ambrogio e Simone, 1335, e una questione di affreschi perduti,” Prospettiva 48 (1987): 69 – ​74, and for a recent summary of the status questionis, see Wolfgang Loseries, in Maestri senesi e toscani nel Lindenau – ​Museum di Altenburg, ed. Miklós Boskovits and Johannes Tripps (Siena, 2008), 130 n. 17. The scene represents the episode of the three-year-old Mary being taken by her parents to the temple; the child spontaneously ascends the flight of steps to the temple, where she would reside until the age of fourteen. By painting the temple at an angle to the picture plane, displaced to the right side of the composition, Andrea seems, however, more faithful to his father’s more dynamic and “modern” composition than to the Lorenzettian model.

After initial attempts to attribute the three panels to Bartolo di Fredi,[16]  [16]
In manuscript expertises (some of them dated 1934), Giuseppe Fiocco, Raimond van Marle, Osvald Sirén, Wilhelm Suida, and Adolfo Venturi all proposed an attribution of the three panels to Bartolo di Fredi. According to the manuscript opinion of F. Mason Perkins, however, the panels were attributable not to Bartolo himself but to an “exceptionally close and as yet unidentified pupil.” See Fern Rusk Shapley, Catalogue of the Italian Paintings, 2 vols. (Washington, DC, 1979), 1:4, and copies of the expertises in NGA curatorial files.

art historians in general accepted them (and also the fourth now in Esztergom) as the work of Andrea.[17]  [17]
Roberto Longhi made the attribution to Andrea di Bartolo in an undated expertise most likely written in 1934, like the others cited in the previous note; see Fern Rusk Shapley, Catalogue of the Italian Paintings, 2 vols. (Washington, DC, 1979), 1:4. Bernard Berenson endorsed this proposal (in an autograph annotation written on the back of a photograph of The Nativity), though with the specification “in great part.” Copies in NGA curatorial files. The generally accepted date for them is c. 1400 or shortly thereafter.[18]  [18]
The dating “c. 1400” already appeared in the entry relating to the three panels in the National Gallery of Art, Preliminary Catalogue of Paintings and Sculpture (Washington, DC, 1941), 3, and was confirmed in the various successive catalogs of the Gallery, as well as by Fern Rusk Shapley, Paintings from the Samuel H. Kress Collection: Italian Schools, xiii – ​xv Century (London, 1966), 65 – ​66; Enzo Carli, I pittori senesi (Siena, 1971), 138; Enzo Carli, La pittura senese del Trecento (Milan, 1981), 238; Giulietta Chelazzi Dini, in Il gotico a Siena: Miniature, pitture, oreficerie, oggetti d’arte (Florence, 1982), 317; Valentina Maderna, ed., Il polittico di Andrea di Bartolo a Brera restaurato (Florence, 1986), 17; and Philippe Lorentz, “De Sienne a Strasbourg: Posterité d’une composition d’Ambrogio Lorenzetti, la Nativité de la Vierge de l’Hôpital Santa Maria della Scala à Sienne,” in Hommage à Michel Laclotte: Etudes sur la peinture du Moyen Age et de la Renaissance (Paris, 1994), 125 repro., 126, 127, 130 – ​131 n. 45. Miklós Boskovits, Early Italian Panel Paintings (Budapest, 1966), 40; András Mucsi, Katalog der Alten Gemälde­galerie des Christlichen Museums zu Esztergom (Budapest, 1975), 42; Eberhard Kasten, “Andrea di Bartolo,” in Allgemeines Künstlerlexikon: Die bildenden Künstler aller Zeiten und Völker, ed. Günter Meißner (Munich, 1992), 3:512, 514, proposed, instead, a less specific date for the three panels in the first decade of the fifteenth century. A proposal to insert them into the catalog of Giorgio di Andrea[19]  [19]
Laurence B. Kanter, “Giorgio di Andrea di Bartolo,” Arte cristiana 74 (1986): 22. Corrado Fratini rejected this rather audacious attribution in Pinacoteca comunale di Città di Castello, vol. 1, Dipinti, Catalogo regionale dei beni culturali dell’Umbria, ed. Francesco Federico Mancini (Perugia, 1987), 143 – ​144; see also Elisabetta Avanzati, in La Sede storica del Monte dei Paschi di Siena: Vicende costruttive e opere d’arte, ed. Francesco Gurrieri and Luciano Bellosi (Florence, 1988), 276 – ​280; Michel Laclotte and Esther Moench, Peinture italienne: Musée du Petit Palais, Avignon (Paris, 2005), 56; and Andrea De Marchi, in Pinacoteca Nazionale di Bologna, catalogo generale, vol. 1, Dal Duecento a Francesco Francia, ed. Jadranka Bentini, Gian Piero Cammarota, and Daniela Scaglietti Kelescian (Venice, 2004), 183 – ​184. has found no acceptance in the literature. G. Fattorini described the three Washington panels as akin to the Adoration of the Magi in the Salini collection (Castello di Gallico, near Asciano, Siena), which he dated to the first decade of the fifteenth century.[20]  [20]
According to G. Fattorini, in La collezione Salini: Dipinti, sculture e oreficerie dei secoli XII, XIII, XIV e XV, ed. Luciano Bellosi, 2 vols. (Florence, 2009), 1:238, the painting was possibly decorated with the very same punches as the Washington stories of the Virgin.

From a stylistic point of view, the scenes from the childhood of Mary can be compared with such paintings as the six stories of Saint Galgano now divided between the Museo Nazionale in Pisa and the National Gallery in Dublin (these, too, most likely originated as parts of a panel in the form of a dossal);[21]  [21]
Enzo Carli, Il Museo di Pisa (Pisa, 1974), 61–62. Suggested dating: probably before 1400. various portable triptychs in the museums of Altenburg,[22]  [22]
Francesca Pasut, in Maestri senesi e toscani nel Lindenau–Museum di Altenburg, ed. Miklós Boskovits and Johannes Tripps (Siena, 2008), 104–106, with dating c. 1395–1400. Philadelphia,[23]  [23]
No. JC Cat. 99; see Carl Brandon Strehlke, Italian Paintings, 1250 – ​1450, in the John G. Johnson Collection and the Philadelphia Museum of Art (Philadelphia, 2004), 37 – ​41. Suggested dating: c. 1394. Prague,[24]  [24]
Nn. O11.919 – ​O11.921. See Olga Pujmanová, Italienische Tafelbilder des Trecento in der Nationalgalerie Prag (Berlin, 1984), no. 20. Suggested dating: c. 1400. and Siena;[25]  [25]
No. 133; see Piero Torriti, La Pinacoteca Nazionale di Siena, vol. 1, I dipinti dal xii al xv secolo (Genoa, 1977), 203, with dating to the early years of the fifteenth century. or the paintings on a casket in the Palazzo Pubblico in Siena.[26]  [26]
Alessandro Cecchi, in Il gotico a Siena: Miniature, pitture, oreficerie, oggetti d’arte (Florence, 1982), 328–331, with dating to c. 1400–1410. Unfortunately, none of these paintings is securely dated. Since the only documented works of the painter have been lost, the one secure point of reference for the chronology of his career remains the fragmentary polyptych in the Church of the Osservanza at Siena, dated 1413.[27]  [27]
See Raimond van Marle, The Development of the Italian Schools of Painting, vol. 2, The Sienese School of the 14th Century (The Hague, 1924), 576, and Antonio Vannini, L’Osservanza di Siena: Guida artistica della chiesa e del convento di San Bernardino all’Osservanza (Siena, 2004), 40–41. The lack of other secure points of reference explains why the chronological reconstruction of Andrea’s works remains so beset by uncertainty. For example, his signed Assumption of the Virgin (Virginia Museum of Fine Arts, Richmond) sometimes is considered to belong to his early period, and sometimes to his full maturity.[28]  [28]
Hendrik W. van Os, “Andrea di Bartolo’s Assumption of the Virgin,” Arts in Virginia 2 (1971): 5, dated the painting now in Richmond (no. 54.11.3; 230.2 × 85 cm) “in the seventies” of the fourteenth century. Gaudenz Freuler, “Andrea di Bartolo, Fra Tommaso d’Antonio Cafarini, and Sienese Dominicans in Venice,” The Art Bulletin 69 (1987): 584, pushed its date forward to the latter years of the century, given that Ser Palamedes, in whose memory the painting was commissioned, was still alive in 1394. Art historians in general, however, have continued to regard the panel as a youthful work of the artist. See Elisabetta Avanzati, in La Sede storica del Monte dei Paschi di Siena: Vicende costruttive e opere d’arte, ed. Francesco Gurrieri and Luciano Bellosi (Florence, 1988), 282; Carl Brandon Strehlke, Italian Paintings, 1250 – ​1450, in the John G. Johnson Collection and the Philadelphia Museum of Art (Philadelphia, 2004), 27. The present writer has long maintained (and still believes) that Andrea di Bartolo’s stories of Christ, divided between the museums of Toledo in Ohio, Bologna, and private collections, probably belong to the Richmond Assumption. It is generally recognized as a work of the artist’s full maturity; cf. Andrea De Marchi, in Pinacoteca Nazionale di Bologna, catalogo generale, vol. 1, Dal Duecento a Francesco Francia, ed. Jadranka Bentini, Gian Piero Cammarota, and Daniela Scaglietti Kelescian (Venice, 2004), 183 – ​184, with a suggested dating of c. 1420. The common origin of the dispersed predella and the Richmond panel is suggested both by their stylistic character and their size. The width of the panel with the Assumption (measuring 230.2 × 85 cm) matches that of the Toledo Crucifixion (50 × 84.3 cm) that would in origin have been placed below it, at the center of the predella. See Miklós Boskovits and Serena Padovani, The Thyssen-­Bornemisza Collection: Early Italian Painting 1290 – ​1470 (London, 1990), 21. Some clues for a reconstruction of the artist’s career can, I believe, be deduced from the miniatures painted by Andrea for the choir-books of the Eremo di Lecceto near Siena, probably during the 1390s.[29]  [29]
The miniatures of antiphonary H.1.7 now in the Biblioteca Comunale of Siena are dated c. 1400; cf. Giulietta Chelazzi Dini, in Il gotico a Siena: Miniature, pitture, oreficerie, oggetti d’arte (Florence, 1952), 320 – ​321; Gaudenz Freuler, “La miniatura senese degli anni 1370 – ​1420,” in La miniatura senese 1270 – ​1420, ed. Cristina De Benedictis (Milan, 2002), 182 – ​186. The illuminations of a missal, G.III.7 in the same library, are at least in part probably attributable to Andrea in a slightly earlier phase, as Chelazzi Dini proposed in Il gotico a Siena: Miniature, pitture, oreficerie, oggetti d’arte (Florence, 1982), 317 – ​318. The strong compositional simplification and charged color of these miniatures reveal significant affinities with the scenes from the life of the Virgin being discussed here, and thus seem to confirm that they belong to a relatively precocious phase in Andrea’s career.

Comparisons of the Gallery’s panels with the figures of saints in the Church of the Osservanza in Siena (1413), on the other hand, show that the latter belong to a more advanced phase in the artist’s career. Some lateral panels of polyptychs, such as that in Tuscania Cathedral, of which the predella has also survived, are easier to compare with the Osservanza saints. In contrast to the tall and slender saints of the Osservanza, who wear draperies furrowed by long, close-set, sharply undercut folds, those of Tuscania are more robust in physique and more placid in expression; their statuesque figures seem to indicate an earlier date of execution, somewhat closer in style to the group of miniatures Andrea probably realized in the last decade of the fourteenth century.[30]  [30]
On the Tuscania polyptych, cf. Luisa Mortari, in La pittura viterbese dal xiv al xvi secolo: Catalogo delle opere, ed. Italo Faldi and Luisa Mortari (Viterbo, 1954), 29 – ​30, and Laurence B. Kanter, “Giorgio di Andrea di Bartolo,” Arte cristiana 74 (1986): 17 – ​24. Kanter’s proposed reconstruction of the altarpiece, and his addition to it of components now situated elsewhere, seems correct, but it is difficult to share his attribution of it to Giorgio d’Andrea or his dating to the 1420s. This date, formerly accepted also by the present writer, now seems to me too late and should, I believe, be modified to 1405 – ​1410; cf. Miklós Boskovits and Serena Padovani, The Thyssen-­Bornemisza Collection: Early Italian Painting 1290 – ​1470 (London, 1990), 21. Gabriele Fattorini came to similar conclusions concerning the date of the Tuscania polyptych in La collezione Salini: Dipinti, sculture e oreficerie dei secoli XII, XIII, XIV e XV, ed. Luciano Bellosi, 2 vols. (Florence, 2009), 1:242 – ​249. If this conclusion is correct, and if therefore the crowded scenes thronged with corpulent and largely immobile figures in the predella in Tuscania testify to Andrea’s art around 1405 – ​1410, it seems reasonable to propose a dating to the very first years of the Quattrocento for the Gallery’s scenes from the life of the Virgin. The compositions in these panels are reduced to essentials, and no signs are yet visible either of the more spacious layout of the scenes or of the greater liveliness of the figures that can be seen in the stories of Christ in the now dispersed predella that should probably be connected with the Assumption in Richmond and in other altarpieces reasonably considered later than the Osservanza saints.[31]  [31]
Apart from polyptych no. 220 of the Pinacoteca in Siena, generally considered a late work of Andrea, I refer to the polyptych in Sant’Angelo in Vado, now divided between the Pinacoteca di Brera in Milan and the Galleria Nazionale in Urbino, and the dispersed predella reconstructed around the Crucifixion no. 12.6 in the Metropolitan Museum of Art, New York. For these paintings, see Piero Torriti, La Pinacoteca Nazionale di Siena, vol. 1, I dipinti dal XII al XV secolo (Genoa, 1977), 207; Valentina Maderna, “Il polittico di Andrea di Bartolo a Brera,” in Il polittico di Andrea di Bartolo a Brera restaurato, ed. Valentina Maderna (Florence, 1986), 9 – ​15; Federico Zeri, in Federico Zeri and Elizabeth E. Gardner, Italian Paintings: Sienese and Central Italian Schools; A Catalogue of the Collection of the Metropolitan Museum of Art (New York, 1980), 1 – ​​2; and Miklós Boskovits and Serena Padovani, The Thyssen-­Bornemisza Collection: Early Italian Painting 1290 – ​​1470 (London, 1990), 16 – ​​21. In my discussion of this predella in 1990, I wrongly connected it with two lateral panels of paired saints from a dispersed triptych, now in a private collection. These panels, Saints Louis of Toulouse and John the Baptist and John the Evangelist and Francis, now seem to me relatively early works of the artist, probably datable to the first decade of the fifteenth century, whereas the predella I had placed in the period c. 1415 – ​​1420 ought to be closer in date to the latter end of this spectrum.

The closest stylistic affinities of the Gallery’s panels therefore are with works whose figures are more robust and more sedate in character. Paintings that fall into this category — ​​apart from the polyptych in Buonconvento and the altarpiece now in the museum in Murano, both datable to the last decade of the fourteenth century — ​​include the fragment with the Virgin Annunciate formerly in the Minneapolis Institute of Arts and the fragmentary Saint Michael Archangel in the Pinacoteca Nazionale in Siena (no. 63), for both of which Laurence Kanter (1986) proposed a provenance from the same altarpiece of which the Gallery’s panels originally formed a part.[32]  [32]
Laurence B. Kanter, “Giorgio di Andrea di Bartolo,” Arte cristiana 74 (1986): 22 – ​24. The Minneapolis fragment was deaccessioned, and its present whereabouts are unknown to me. For the fragment no. 63 in the Pinacoteca Nazionale in Siena, cf. Piero Torriti, La Pinacoteca Nazionale di Siena, vol. 1, I dipinti dal XII al XV secolo (Genoa, 1977), 205. Apparently, during these years — ​​that is, the first fifteen years of the fifteenth century — ​​Andrea especially painted small-scale works for private devotion, such as the abovementioned portable triptych no. 133 in the Pinacoteca of Siena; this resembles our scenes from the life of Mary not only in the proportions and physiognomic types of the figures but also in its peculiar compositional devices.[33]  [33]
The somewhat naive compositional device proposed by the painter in the Nativity of the central panel of the triptych in Siena, namely that of displacing sharply to the left the little tree in the background to avoid its branches being concealed by the cusped border of the frame, recalls the improbable displacement — ​​​for the same reason — ​​​of the dome to the left margin of the temple in Presentation of the Virgin in the Temple and Joachim and Anna Giving Food to the Poor and Offerings to the Temple in the National Gallery of Art. In another triptych datable to this period, that of the Lindenau-­Museum in Altenburg, the cloak of the young female saint of the left leaf is closely comparable with that of the majestic Saint Anne of The Presentation of the Virgin to the Temple. Another comparable work [34]  [34]
We may further recall as works exemplifying Andrea’s phase at the turn of the century a portable triptych in the Brooklyn Museum in New York, no. 34. 839; see Carl Brandon Strehlke, Italian Paintings, 1250 – ​1450, in the John G. Johnson Collection and the Philadelphia Museum of Art (Philadelphia, 2004), 41, and the little panels with saints, formerly belonging to the museums in Altenburg, Oslo, Oxford, and in private collections; see Francesca Pasut, in Maestri senesi e toscani nel Lindenau – ​Museum di Altenburg, ed. Miklós Boskovits and Johannes Tripps (Siena, 2008), 107 – ​111. is the Adoration of the Magi now in the Salini collection, which recalls the Washington panels both in the statuesque pose of its figures and in the characteristics of its architectural backdrop.[35]  [35]
Gold Backs: 1250 – ​1480 (Turin, 1996), 122, 124; Gabriele Fattorini, in La collezione Salini: Dipinti, sculture e oreficerie dei secoli XII, XIII, XIV e XV, ed. Luciano Bellosi, 2 vols. (Florence, 2009), 1:236 – ​249. The painting, cut on all sides, measures 25.5 × 21 cm. The fragment at first sight might also seem the companion panel of these in the Gallery, but its original size must have been slightly smaller than that of the Washington panels.

Miklós Boskovits (1935–2011)

March 21, 2016