Madonna and Child with Saint John the Baptist, Saint Peter, and Two Angels, c. 1290

The enthroned Madonna supports her son, seated on her left knee, with both hands. The child, in a frontal position, blesses with his right hand, holding a roll of parchment in his left. The composition is a variant of the type of the Hodegetria Virgin; in the present case, she does not point towards her son, as in the Byzantine prototype, but instead presents him to the spectator.[1]  [1]
Evelyn Sandberg-­Vavalà, L’iconografia della Madonna col bambino nella pittura italiana del Dugento (Siena, 1934), 45–48, classified this subtype as the “seconda modificazione del tipo Odigitria” (second modification of the Hodegetria type). On the iconography of the Hodegetria (a term translatable as “indicator of the way”), cf. the exhaustive treatment by Gregor Martin Lechner, “Maria,” in Reallexikon zur byzantinischen Kunst, ed. Klaus Wessel, 7 vols. (Stuttgart, 2005), 6:58–71. The variants of the protagonists’ poses in this iconographic type are understood as attempts to augment the realistic effect of the image, i.e., expressions of vivacity and of the affection between mother and child. Her affectionate maternal pose is thus given precedence over her more ritual and impersonal role. But Mary’s pose perhaps has another sense: she seems to be rearranging her son’s small legs to conform them to the cross-legged position considered suitable for judges and sovereigns in the Middle Ages.[2]  [2]
That, at any rate, is how I think Mary’s gesture should be interpreted. On the significance of the pose of crossed legs in medieval iconography, cf. Johan Jakob Tikkanen, Die Beinstellungen in der Kunstgeschichte: Ein Beitrag zur Geschichte der künstlerischen Motive (Helsingfors, 1912), 153, while François Garnier, Le langage de l’image au Moyen Age, vol. 2, Grammaire des gestes (Paris, 1989), 158–159, added that “cette attitude est par excellence celle des personagges qui remplit les fonctions du Juge” (this is a pose par excellence of people who perform the duties of a judge).

The panel is painted in a rapid, even cursory manner. The artist omits the form of the throne’s backrest, which remains hidden by the cloth of honor, supported by angels and decorated with bold motifs of popular taste (an interlocking pattern of quatrefoils and octagons). He designs the form of the throne itself in an incongruous manner, apparently semicircular at the seat and rectangular at the base. According to a convention widespread in Tuscany in the second half of the thirteenth century, the painter presents the seat of the Madonna as if it were seen not frontally but from the left.[3]  [3]
In a study on the types of thrones in thirteenth-century paintings, James Stubblebine (1954–1957) distinguished four types, of which the third, exemplified by such works as Guido da Siena’s Maestà in San Domenico in Siena and Duccio’s Madonna Rucellai, commissioned in 1285 and now in the Uffizi, Florence, is a structure conceived as a “three-dimensional block” and represented so that “the front and back are parallel to the picture plane, while the sides run diagonally.” This typology common to paintings dating to the last quarter of the Duecento naturally implies some indication of the chronological position of the panel discussed here. See James H. Stubblebine, “The Development of the Throne in Dugento Tuscan Painting,” Marsyas 7 (1954–1957): 25–39. This is suggested not only by the position of the step but also by the fact that while the figure of the Baptist is seen in its entirety, that of Saint Peter to the right is partially hidden by the throne. That the patron must have been a person of modest means is suggested both by the cursory execution and by the eschewal of gilding on the frame; instead, the artist adopts the unusual expedient of painting it red and decorating it with a frieze of daisies.[4]  [4]
It seems to me that the flowers cursorily painted on the frame can be identified as daisies or marguerites (leucanthemum vulgare), symbols, according to Mirella Levi D’Ancona, of the “blessed souls in heaven” as well as of the “incarnation of Christ.” See Mirella Levi D’Ancona, The Garden of the Renaissance: Botanical Symbolism in Italian Painting (Florence, 1977), 124–126. The decoration of the frame discussed here vaguely recalls that of the panel with Saint Catherine and stories of her legend, no. 1583 in the Museo Nazionale in Pisa; cf. Lorenzo Carletti, in Cimabue a Pisa: La pittura pisana del Duecento da Giunta a Giotto, ed. Mariagiulia Burresi and Antonino Caleca (Pisa, 2005), 193.

Carlo Lasinio’s attribution of the painting to CimabueCimabue
 (Cenni [Benciviene] di Pepo) (c. 1240–before July 14, 1302) Italian painter and mosaicist. His nickname means either “bull-head” or possibly “one who crushes the views of others,”’ (It. cimare: ‘top, shear, blunt’), an interpretation matching the tradition in commentaries on Dante that he was not merely proud of his work but contemptuous of criticism. Filippo Villani and Vasari assigned him the name Giovanni, but this has no historical foundation. He may be considered the most dramatic of those artists influenced by contemporary Byzantine painting through which antique qualities were introduced into Italian work in the late 13th century. His interest in classical Roman drapery techniques and in the spatial and dramatic achievements of such contemporary sculptors as Nicola Pisano, however, distinguishes him from other leading members of this movement. As a result of his influence on such younger artists as Duccio and Giotto, the forceful qualities of his work, and its openness to a wide range of sources, Cimabue appears to have had a direct personal influence on the subsequent course of Florentine, Tuscan, and possibly Roman painting. —Robert Gibbs, Grove Art © Oxford University Press, accepted in the sale catalog of 1835, remained ignored in the art historical literature until the panel reappeared at Patterdale Hall in the Lake District in 1934, and even then it did not meet with approval.[5]  [5]
In the 1934 sale, the painting was classified as “Italian school.” Byam Shaw, in his 1963 letter (see Provenance note 4), noted that the inscription on the label affixed to the back of the panel confirmed Lasinio’s ascription to Cimabue, adding, “not the highest quality, but undoubtedly a genuine Pisan work of the time.” Following the panel’s purchase by the Florentine dealer and collector Alessandro Contini-­Bonacossi, however, several connoisseurs of late medieval Italian art concurred with the Cimabuesque authorship of the work in their correspondence with its then owner.[6]  [6]
Copies of the expertises by Roberto Longhi, Evelyn Sandberg-­Vavalà, Wilhelm Suida, Raimond van Marle, and Adolfo Venturi are preserved in the NGA curatorial files: they were unanimous in confirming the authorship of Cimabue for the painting. Of these experts, only Longhi provided a chronological indication, maintaining that the panel must antedate Cimabue’s Madonna (no. 8343) now in the Uffizi, Florence. Roberto Longhi, on publishing the painting in 1948, confirmed this view, dedicating a brief but penetrating comment to the painting; in his opinion the image was a product of Cimabue’s first period of residence in Pisa.[7]  [7]
“La Madonna come un falchetto inciprignito,” Roberto Longhi wrote in 1948, “ il Bambino con aria di censore prepotente​ . . . ​gli angeli che impugnano, quasi manovelle di timone, le cocche del drappo​ . . . ​i due santi collerici​ . . . ​che sbuffano ai lati del trono. In un tema puramente sacramentale ce n’è abbastanza per alludere ai drammi imminenti del transetto di Assisi” (The Madonna like an irritated bird of prey, the child with the air of an overbearing censor, the angels who grab the corners of the cloth like the handle of a boat’s tiller, the two irascible saints snorting at the sides of the throne. In a purely sacramental theme there is enough to foreshadow the imminent drama of the transept at Assisi). Longhi referred, of course, to the frescoes in the upper church of San Francesco in Assisi, which he dated to the years between the end of the eighth and the first half of the ninth decade of the thirteenth century. See Roberto Longhi, “Giudizio sul Duecento,” Proporzioni 2 (1948): 16. Luigi Coletti, in 1941, took a different view and argued that the panel was better placed in the circle of the Magdalen Master.[8]  [8]
Luigi Coletti, I Primitivi, vol. 1, 120 tavole (Novara, 1941), 33–34.  Subsequently, once the Madonna had entered the Kress Collection in 1948, the attribution to Cimabue would continue to be supported,[9]  [9]
Supporting Cimabue’s authorship were Ugo Galetti and Ettore Camesasca, Enciclopedia della pittura italiana, 3 vols. (Milan, 1951), 1:672; Ferdinando Bologna, “Un capola­voro giovanile di Duccio,” Paragone 11, no. 125 (1960): 9–10, who proposed its execution before 1280; Luiz Marques, La peinture du Duecento en Italie centrale (Paris, 1987), 282, who thought that the panel was painted c. 1270–1275; Adam S. Labuda, in Opus sacrum: Catalogue of the Exhibition from the Collection of Barbara Piasecka Johnson, ed. Józef Grabski (Vienna, 1990), 30; Nicholas H. J. Hall, ed., Colnaghi in America: A Survey to Commemorate the First Decade of Colnaghi, New York (New York, 1992), 134; and Enzo Carli, La pittura a Pisa dalle origini alla Bella Maniera (Pisa, 1994), 23, who placed its execution after the Maestà in the Louvre, Paris, a work that, like the Washington Madonna, also comes from the church of San Francesco in Pisa. but most scholars preferred to leave it in anonymity or ascribed it to Cimabue’s shop.[10]  [10]
The painting was classified as “attributed to Cimabue” in National Gallery of Art, Paintings and Sculpture from the Kress Collection (Washington, DC, 1956), 50–51; National Gallery of Art, Summary Catalogue of European Paintings and Sculpture (Washington, DC, 1965), 28; Fern Rusk Shapley, Paintings from the Samuel H. Kress Collection: Italian Schools, XIII–XV Century (London, 1966), 5–6; Fern Rusk Shapley, Catalogue of the Italian Paintings, 2 vols. (Washington, DC, 1979), 1:133–134; Dorothy Lygon and Russell Francis, “Tuscan Primitives in London Sales: 1801–1837,” The Burlington Magazine 122 (1980): 113, 116; and National Gallery of Art, European Paintings: An Illustrated Catalogue (Washington, DC, 1985), 89. Scholars who doubted the attribution, although without formulating any alternative proposals, include Sergio Samek Ludovici, Cimabue (Milan, 1956), 24, 32; Roberto Salvini, “Cimabue,” in Enciclopedia Universale dell’Arte, ed. Istituto per la collaborazione culturale, 15 vols. (Florence, 1958–1967), 3 (1960): 472; Miklós Boskovits, “Cenni di Pepe (Pepo), detto Cimabue,” in Dizionario biografico degli italiani (Rome, 1979), 23: 542; and Angelo Tartuferi, “Un libro e alcune considerazioni sulla pittura del Duecento in Italia centrale,” Arte cristiana 76 (1988): 434, 440 n. 26. Those who believed that it was produced in the artist’s workshop or by his immediate following include John Pope-­Hennessy, “Review of Proporzioni ii by Roberto Longhi,” The Burlington Magazine 90 (1948): 360; Bernard Berenson, Italian Pictures of the Renaissance: Florentine School, 2 vols. (London, 1963), 1:50; Burton B. Fredericksen and Federico Zeri, Census of Pre-­Nineteenth-­Century Italian Paintings in North American Public Collections (Cambridge, MA, 1972), 54. The painting was defined as the work of an anonymous late thirteenth-century Florentine master by Edward B. Garrison, Italian Romanesque Panel Painting: An Illustrated Index (Florence, 1949), 18, 100; Cesare Brandi, Duccio (Florence, 1951), 135 n. 13; Carlo Volpe, “Preistoria di Duccio,” Paragone 5, no. 49 (1954): 18; Carlo Ludovico Ragghianti, Pittura del Dugento a Firenze (Florence, 1955), 124, 127; Enio Sindona, L’opera completa di Cimabue e il momento figurativo pregiottesco (Milan, 1975), 117, 118; and Angelo Tartuferi, “Pittura fiorentina del Duecento,” in La Pittura in Italia: Le origini, ed. Enrico Castelnuovo (Milan, 1985), 236, 240 n. 43. Roberto Salvini, “Postilla a Cimabue,” Rivista d’arte 26 (1950): 54; and Enzo Carli, Pittura medievale pisana (Milan, 1958), 64, thought it the work of an anonymous Pisan master. Also, Jens T. Wollesen, Hasten to My Aid and Counsel: The Answers of the Pictures; Private Devotional Panel Painting in Italy around 1300 (New York, 2005), 87–88, 182–183, classified is as Cimabuesque, of c. 1250–1300.  The present writer (1976) tentatively associated the panel with the group of paintings that he had gathered around the name of Gaddo Gaddi,[11]  [11]
Miklós Boskovits, Cimabue e i precursori di Giotto: Affreschi, mosaici e tavole (Florence, 1976), n.p. [9], no. 22, pointed out that the style of the Gallery's panel was close to that of the group of works he himself had referred to Gaddo Gaddi. His reconstruction of Gaddo’s oeuvre consisted of the fragmentary Madonna in the church of San Remigio in Florence; the crucifix no. 1345 in the Galleria dell’Accademia in Florence; another painted crucifix in the church of Santo Stefano a Paterno (Bagno a Ripoli, Florence); and a small portable cross in the Fogg Art Museum in Cambridge, Massachusetts (no. 1929.250), in addition to the mosaic of the Coronation of the Virgin on the inner façade of the Duomo in Florence. Vasari cited the latter as a work by Gaddo Gaddi in his vita of that master, together with part of the mosaic on the façade of Santa Maria Maggiore in Rome (in general ascribed by the art historical literature to Filippo Rusuti); the mosaic Assumption in the apse of the transept of Pisa Cathedral, in actual fact executed on the basis of a design by Simone Martini (Sienese, active from 1315; died 1344); and some other works, which today appear stylistically incongruous or are lost. See Giorgio Vasari, Le vite dei più eccellenti pittori, scultori ed architettori, ed. Gaetano Milanesi, 9 vols. (Florence, 1878–1885), 1(1878):346–347. In more recent times, Marco Ciatti and Cecelia Frosinini (1993) and Ada Labriola (1998) cited the panel now in the National Gallery of Art as a work by Gaddi. Cf. Marco Ciatti and Cecilia Frosinini, ed., La Madonna di San Giorgio alla Costa di Giotto: Studi e restauro, Problemi di conservazione e restauro (Florence, 1995), 36; Ada Labriola, “Gaddi, Gaddo,” in Dizionario biografico degli italiani (Rome, 1998), 51:155. and Luciano Bellosi (1998) suggested an attribution to the Florentine Azzo di Mazzetto;[12]  [12]
Accepting Alessandro Bagnoli’s suggestion, Luciano Bellosi, Cimabue, ed. Giovanna Ragionieri (Milan, 1998), 267, thought that the master of the frescoes of the “Sala di Dante” in the Palazzo Pubblico of San Gimignano might be identified with Azzo di Mazzetto, an artist documented in Florence in 1282. He attributed to the same hand a painted crucifix in the Museo Civico in San Gimignano and the present panel of the National Gallery of Art. Silvia Giorgi accepted this proposal in “Azzo di Mazzetto,” in La pittura in Europa: Il dizionario dei pittori, ed. Carlo Pirovano, 3 vols. (Milan, 2002), 1:39, and it was cited without comment by Ada Labriola, “Lo stato degli studi su Cimabue e un libro recente,” Arte cristiana 88 (2000): 344; and Sonia Chiodo, “Attorno a un dipinto inedito di Ambrogio Lorenzetti,” Arte cristiana 91 (2003): 1–6. no certain works by either of these painters have come down to us. It is worth pointing out, in this regard, that documents record a painter by the name of Azzo who worked for the comune of San Gimignano in the years between 1288 and 1293 and who painted “cameram novi palatii comunis” (the room of the new municipal palace) in that town in 1291.[13]  [13]
Cf. Pio Rajna, “Pittura e pittori a San Gimignano intorno all’anno 1300,” Miscellanea storica della Valdelsa 18 (1920): 1–13. The subject of his paintings is not recorded, but it seems plausible to link his name with the frescoes in the Palazzo del Popolo that commemorate the privileges granted to San Gimignano by Charles of AnjouAnjou, House of
 French dynasty of rulers, patrons, and collectors. The first House of Anjou was founded by Charles of Anjou (1266–1285) and was active mainly in Italy, notably as kings of Naples and Jerusalem. Members of the second House of Anjou lost Naples to the House of Aragon, but continued to style themselves as kings of Naples, Sicily, and Jerusalem until the death of Charles, 5th Duke of Anjou, in 1481, when the titular kingdom passed to Louis XI, King of France. —Joan Isobel Friedman, Grove Art © Oxford University Press and bear an inscription with the date 1292. The massive, fleshy figures in these frescoes, often hesitant in movement, only vaguely recall the style of the delicate little panel in the National Gallery of Art.[14]  [14]
Antonino Caleca (1986) placed in doubt Vasari’s claim that Gaddo was the master of the mosaic of the Coronation in the Florentine Duomo. He argues that the execution of the mosaic should be given instead to Francesco da Pisa, documented author of the Christ in Majesty in the apse of Pisa Cathedral. See Antonio Caleca, “Pittura del Duecento e del Trecento a Pisa e a Lucca,” in La Pittura in Italia: Il Duecento e il Trecento, ed. Enrico Castelnuovo, 2 vols. (Milan, 1986), 1:243; and Antonio Caleca, “Francesco da Pisa,” in La Pittura in Italia: Il Duecento e il Trecento, ed. Enrico Castelnuovo, 2 vols. (Milan, 1986), 2:570. For my part, ever since 1984, I have observed that the Coronation of the Virgin reveals features a good deal more archaic than the most advanced artistic tendencies in Florence at the close of thirteenth century and have conjectured that the work originally was intended for a site other than Arnolfo di Cambio’s Duomo. Miklós Boskovits, A Critical and Historical Corpus of Florentine Painting: The Fourteenth Century, sec. 3, vol. 9, The Miniaturist Tendency (Florence, 1984), 18–19 n. 31. Later (1993), I expressed my readiness to accept the proposal of Monica Bietti (1983), who identified with Gaddo Gaddi the artist whose oeuvre had been reconstructed under the conventional name of the Master of Saint Cecilia. See Miklós Boskovits, A Critical and Historical Corpus of Florentine Painting: The Fourteenth Century, sec. 3, vol. 9, The Miniaturist Tendency (Florence, 1984), 18–19 n. 31; Miklós Boskovits, A Critical and Historical Corpus of Florentine Painting, sec. 1, vol. 1, The Origins of Florentine Painting, 1100–1270 (Florence, 1993), 143 n. 314; Monica Bietti, “Gaddo Gaddi: un’ipotesi,” Arte cristiana 71 (1983): 49–52. For their part, Annamaria Giusti (1994) and Luciano Bellosi (1998) accepted, at least as a working hypothesis, Gaddo’s authorship of the Coronation mosaic but denied that the mosaicist had any hand in the panel paintings attributed to him; cf. Annamaria Giusti, “I mosaici dei coretti, dei parapetti del matroneo, e del tamburo della cupola,” in Il Battistero di San Giovanni a Firenze, ed. Antonio Paolucci, 2 vols. (Modena, 1994), 1:325, 327 et passim; Luciano Bellosi, Cimabue, ed. Giovanna Ragionieri (Milan, 1998), 129, 137. For a discussion of the Gaddo = St. Cecilia Master hypothesis, see Miklós Boskovits, “Un nome per il maestro del Trittico Horne,” Saggi e memorie di storia dell’arte 27 (2003): 57–70. The attribution to Francesco Pisano has been reproposed in recent times by Mariagiulia Burresi and Antonino Caleca, “Pittura a Pisa da Giunto a Giotto,” in Cimabue a Pisa: La pittura pisana del Duecento da Giunta a Giotto, ed. Mariagiulia Burresi and Antonino Caleca (Pisa, 2005), 82–84, whereas the present writer gives the Coronation to the author of part of the mosaics of the dome of the baptistery, called the “Penultimate Master,” in Miklós Boskovits, A Critical and Historical Corpus of Florentine Painting, sec. 1, vol. 2, The Mosaics of the Baptistery of Florence (Florence, 2007), 591–594.

The attribution of the work to Gaddo Gaddi was based on a tradition handed down by Vasari, according to whom this master was responsible for the mosaic of the Coronation of the Virgin on the inner façade of the cathedral of Santa Maria del Fiore in Florence. The authenticity of this tradition was accepted down to recent times, until art historians gradually began to perceive that the figurative style of the mosaic appears not to have been that of an artist of the generation of Giotto (Florentine, c. 1265 - 1337), as Gaddo Gaddi would have been, according to the evidence of documents, but that of a somewhat older master. In recent times, in fact, the identification of Gaddo with the Giottesque Master of Saint Cecilia has been proposed, while the Coronation of the Virgin on the inner façade of the Duomo has been either cited with the traditional attribution or attributed to Francesco Pisano, an artist belonging to Cimabue’s generation, or given to the so-called Penultimate Master of the mosaics in the baptistery in Florence.  

Some of the paintings formerly associated with the name of Gaddo still seem to me stylistically homogeneous, even if not easy to refer to a particular artist. It is with this group of works that the panel now in Washington should, I believe, be most profitably compared. The agile, nervous figures, with their jerky movements, flashing eyes, and beaklike noses, which Longhi described so brilliantly, invite comparison both with the figures painted in the portable cross in the Fogg Museum in Cambridge, Massachusetts, and with those in the large painted crucifix in Santo Stefano a Paterno in Florence.[15]  [15]
Roberto Longhi, “Giudizio sul Duecento,” Proporzioni 2 (1948): 16. In this latter painting, in particular, the grieving Madonna [fig. 1]   [fig. 1] Detail of Mourning Madonna, Penultimate Master of the Baptistery, Painted Crucifix, 1280–1290, tempera on panel, Church of Santo Stefano a Paterno, Bagno a Ripoli, Florence. Edizioni Brogi photograph, National Gallery of Art, George Martin Richter Archive, vol. 3, folder 3, sheet 17 recalls the protagonist of the painting in the Gallery [fig. 2]   [fig. 2] Detail of Madonna, Tuscan thirteenth century, Madonna and Child with Saint John the Baptist, Saint Peter, and Two Angels, c. 1290, tempera on panel, National Gallery of Art, Washington, Samuel H. Kress Collection, while the severe and ascetic blessing Christ in the cimasa is comparable to the Baptist in the panel in Washington, bearing in mind, of course, the differences in pictorial technique deriving from their very different dimensions. Drapery forms in these paintings, enlivened by sudden darting highlights and furrowed by sharp, deeply undercut folds so that they seem made of twisted sheet metal instead of fabric, also compare closely. A further observation may be made about the presence of the rare frieze of daisies on the hem of the Madonna’s cloak in a painting now in the Allen Memorial Art Museum in Oberlin, Ohio (inv. no. 1945.9). Generally referred to an anonymous follower of Duccio or of Cimabue, recently (Boskovits 2007) it was attributed to the Master of the Cortona-­Loeser Crucifixes.[16]  [16]
Miklós Boskovits, “Da Duccio a Simone Martini,” in Medioevo: La chiesa e il palazzo; Atti del convegno internazionale di studi, Parma, September 20–24, 2005, ed. Arturo Carlo Quintavalle (Milan, 2007), 566–569. That artist’s use of the same motif decorating the frame of the Washington Madonna suggests that he too belonged to the ambience of the painter of our panel. Unfortunately, the comparisons listed here do not help us to establish the date of the Gallery's panel, but they do, more generally, permit its insertion into the context of Florentine painting at the close of the thirteenth century.[17]  [17]
For a broad discussion of the chronology of this decorative enterprise, cf. Luciano Bellosi, Cimabue, ed. Giovanna Ragionieri (Milan, 1998), 162–182, 278–281.

The Pisan provenance of the panel, small and easily transportable, does not imply that it was executed in that city. The master who painted it must have been trained under the influence of Cimabue, and probably at the time of his fresco decoration of the upper church of San Francesco in Assisi. In particular some figures of angels that populate the right transept of that church, with their tense, frowning faces and ruffled garments, suggest that the execution of the panel in the National Gallery of Art should be placed around 1290.  

Miklós Boskovits (1935–2011)

March 21, 2016