The Virgin Crowned by the Infant Christ

…; sale, Jean-Adrien Snyers (1775-1841), Antwerp, sold on the deceased’s premises (auction house not known), scheduled for 4 May 1818, but brought forward to 27 April, no. 116 (‘ANT. VAN DYCK. Deux Anges, dont l’un chante et l’autre joue de la violoncello, amusent de leur musique la Vierge et l’enfant Jésus répresentés dans une gloire. […] Sur toile, haut 3 pieds 3 pouces, large 4 pieds 9 pouces [height and width in inverted order; 137.8 x 93.6 cm]’), bought in; purchased from Jean-Adrien Snyers by Willem I, King of the Netherlands, for the museum, 1822¹¹NHA, RMA, IS, inv. 10, no. 18 (20 June 1821); NHA, RMA, IS, inv. 10, no. 22 (24 July 1821); NHA, RMA, Kop, inv. 36, p. 114 (28 July 1821); NHA, RMA, IS, inv. 10, no. 29 (23 April 1822); NHA, RMA, IS, inv. 10, no. 32 (3 May 1822). E.W. Moes and E. van Biema, De Nationale Konst-Gallery en Het Koninklijk Museum, Amsterdam 1909, p. 134. P. van Vliet, ‘Spaanse schilderijen in het Rijksmuseum, afkomstig van schenkingen van Koning Willem I’, Bulletin van het Rijksmuseum 14 (1966), pp. 131-49, esp. pp. 134, 143, note 9.

ObjectNumber: SK-A-598

Credit line: Gift of Z.K.H. Willem I

The wax-lined support of this painting is made up of three horizontally joined strips of canvas of different weaves. The strip at the top is made up of two pieces vertically joined, measuring 14.5-15.0 cm in height. They are joined together and to the main, middle canvas by lining. The central canvas is a near square, 105.7-107 cm high. The tacking edges at the left, right and bottom have been cut away. It is stitched to the bottom strip, which is 14-15.5 cm high. The central canvas has a double ground of a layer of chalk, and above, one of lead white mixed with black and earth pigments. The upper piece has a single ground of a layer of chalk. There is cusping towards all the edges of the central canvas apart from at the top. The imprint of the original strainer, about 6-8 cm wide, is evident at all its edges except the top where it is only 3-4.5 cm wide, indicating that the canvas has been reduced by 2-3 cm. No underdrawing has been detected. Dilute, brown undermodelling is present beneath the sleeve of the Virgin with sketchy, white brushstrokes to indicate the folds. The composition was built up working from the back using reserves. The paint surface is smooth and opaque, yet applied thinly with passages of wet-in-wet blending.

There are a good many pentiments discernible in infrared reflectography; most likely these were not all made during the same campaign. A) Pentiments made during the initial campaign: the Virgin’s forehead was enlarged; the cloth on which the Christ Child stands was extended at its base; the instrument-playing angel’s profile, proper left arm and right shoulder have been adjusted, and the foreshortening of his proper right arm has been reduced; the fall of drapery on his back was lowered. B) Pentiments made at a subsequent stage: the fall of the Virgin’s veil was reduced; her lowered look to the left was altered to a frontal gaze; more of her chest and bosom was displayed; the fingers of her proper right hand were extended, where more of the white cloth was first depicted; the extent of Christ’s hair and that of the instrument-playing angel was enlarged. The red drapery below and to the left of the viola da gamba was overpainted when the lower strip of canvas was added, at which point, too, the clouds were expanded without reserves.⁵⁵Object file, RMA. Notes made by Gwen Tauber are the basis for the foregoing and give a fuller account.

Three main issues concerning the present painting have been the cause of confusion, and clarification is set out in the following order. The iconography unusually combines at least two separate aspects of Christian belief. The attribution of the work, which has been in the Rijksmuseum collection since the early nineteenth century, has only been settled in recent decades. Conservation has exposed the problematic character of its composition due to old interventions; however, this does not jeopardize the essential integrity of this physically complex work inspired as it most probably was by a lost painting by Anthony van Dyck (1599-1641).

The description in the 1818 sale catalogue confined itself to an account of the image, and in the early years at the museum, the painting was described as the Holy Family. From 1858, the subject was given as the Glorification of Mary, although there is no such subject derived from the Bible or scriptural exegesis.⁶⁶L. Réau, Iconographie de l’art chrétien, 3 vols., Paris 1955-59, II(2), pp. 597-99. The 1976 museum catalogue described it as the Apothesis [sic] of the Virgin. The key to the main intended meaning lies in the rubric to Schelte Adamsz Bolswert’s (1584/88-1659) engraving after a lost painting by Peter Paul Rubens (1577-1640).⁷⁷D. Bodart, Rubens e l’incisione nelle collezione del Gabinetto Nazionale delle Stampa, exh. cat. Rome (Villa della Farnesina alla Lungara) 1977, no. 36. This engraving shows the crowned Virgin with her Child in her arms, trampling on a serpent coiled round the globe of the world. Christ points down to the serpent. The rubric to the print reads ‘Ipsa conteret caput tuum. genesis 3’, this is a reference to the sentence (in the Vulgate) in Genesis 3:15, in which God admonishes the serpent for beguiling Eve to eat the fruit of the tree of knowledge of good and evil and states that ‘she will bruise your head’.

As Mâle relates, the gender of the pronoun was disputed between Catholics and Protestants. The former favoured the feminine reading, which was taken to refer to the ‘new Eve’, the Virgin, whereas Protestants opted for the masculine, which was understood to allude to Christ.⁸⁸E. Mâle, L’art religieux de la fin du XVIe siècle, du XVIIe siècle et du XVIIIe siècle: étude sur l’iconographie après le Concile de Trente, Italie, France, Espagne, Flandres, Paris 1951 (ed. princ. 1932), pp. 37-38. Rubens’s interpretation depended on the feminine reading, while Van Dyck opted for the alternative.⁹⁹See Vey in S.J. Barnes, N. de Poorter, O. Millar and H. Vey, Anthony van Dyck: A Complete Catalogue of the Paintings, New Haven (Conn.)/London 2004, under no. III, 14. In the museum picture the now faint serpent on the globe is not actually suppressed. But it may be understood that the Virgin and Child participate in subduing it, though not, for instance, as deliberately and specifically as is shown in Caravaggio’s Palafrenieri altarpiece (Galleria Borghese), in which the serpent is trampled on.¹⁰¹⁰M. Marini, Io, Michelangelo da Caravaggio, Rome 1974, pp. 409-10, no. 56.

In the Rijksmuseum painting it may seem that as much emphasis, if not more, is placed on the second intended meaning conveyed by the Christ Child crowning the Virgin with a wreath of white and red roses. In Rubens’s paintings of the Assumption of the Virgin,¹¹¹¹R. Oldenbourg, P.P. Rubens: Des Meisters Gemälde, Stuttgart/Berlin 1921, pp. 120, 193, 206, 301, 352. angels offer her wreaths, and in the unused modello at St Petersburg for an Assumption and Coronation of the Virgin,¹²¹²J.S. Held, The Oil Sketches of Peter Paul Rubens: A Critical Catalogue, 2 vols., Princeton 1980, no. 374; N. Gritsay and N. Babina, State Hermitage Museum Catalogue: Seventeenth- and Eighteenth-Century Flemish Painting, St Petersburg 2008, no. 295. the risen Christ crowns his Mother with a chaplet of red and white roses. But no precedent has been found for the Infant Christ’s playing this role. The roses refer to the traditional concept of the Virgin as the mystic rose, the rose without thorns; she is both the rosa candida (white rose) and the rosa rubicunda (red rose). The chaplet alludes to the Crown of Thorns and the Virgin’s grief at Christ’s sacrifice.¹³¹³R. Wittkower, ‘Domenichino’s Madonna Della Rosa’, The Burlington Magazine 90 (1948), pp. 220-23, esp. p. 221. McGrath’s interpretation differs; in her view Christ crowns the Virgin as a bride as she enters heaven.¹⁴¹⁴Private communication. Rubens introduced a choir of music-making angels in the St Petersburg modello; an example such as this and the Netherlandish tradition, referred to by Vey,¹⁵¹⁵Vey in S.J. Barnes, N. de Poorter, O. Millar and H. Vey, Anthony van Dyck: A Complete Catalogue of the Paintings, New Haven (Conn.)/London 2004, p. 399, under no. III.A4. of the Virgin and Child accompanied by music-making angels, would have been a reason for Willeboirts Bosschaert’s introduction of the angel playing the viola da gamba and his singing companion.

Cardinal-Archbishop Federico Borromeo (1564-1631) suggested that the Virgin Immaculate should be depicted ‘seated in some great brilliance’ (in magno aliquo splendore sedente).¹⁶¹⁶J.B. Knipping, Iconography of the Counter Reformation in the Netherlands: Heaven on Earth, 2 vols., Nieuwkoop 1974 (ed. princ. Hilversum 1940), I, p. 249. And in the present work the Virgin and Child are set against the brightness of Christ’s radiance. Healy points out the association of a related composition with the Immaculate Conception,¹⁷¹⁷F. Healy, ‘Images of the “Madonna and Child” and “The Holy Family” in Van Dyck’s Oeuvre’, in H. Vlieghe (ed.), Van Dyck 1599-1999: Conjectures and Refutations, Turnhout 2001, pp. 89-112, esp. p. 106. and no doubt this was in the mind of certain devout viewers, though this is not the subject of the Rijksmuseum picture.¹⁸¹⁸Cited in J. Douglas Stewart, ‘Thomas Willeboirts and Pieter Thijs: A Tale of the Tangled Antwerp Painters; with an Excursus on Van Dyck’s St Felix of Cantalice’, in H. Vlieghe (ed.), Van Dyck 1599-1999: Conjectures and Refutations, Turnhout 2001, pp. 271-88, esp. p. 285, under note 5.

The museum was doubtful about the painting’s early ascription to Van Dyck, with which appellation it had been a gift from King Willem I. The 1853 museum catalogue first associated the painting with the Spanish School; but it was Bredius, some thirty years later, who proposed the Madrid artist José Antolinez (1635-1675) as the painter, which attribution has been retained in all the museum’s printed catalogues. Some art historians concerned with Antolinez have accepted the attribution,¹⁹¹⁹A. de Beruete y Moret, The School of Madrid, London/New York 1909, p. 225 (as less Spanish in style); V. von Loga, Die Malerei in Spanien vom XIV. bis XVIII. Jahrhundert, Berlin 1923, p. 386. although Angulo Iñiguez omitted it from his monograph of 1957,²⁰²⁰D. Angulo Iñiquez, José Antolínez, Madrid 1957. perhaps because he accepted Salazar’s earlier attribution to Miguel Manrique that was followed by Gaya Nuño in 1958.²¹²¹J.A. Gaya Nuño, La pintura español fuera de españa: Historia y catálogo, Madrid 1958, p. 227, no. 1691; Manrique, also known as Miguel de Amberes, active in Malaga in the second half of the seventeenth century, had been supposedly a pupil of Rubens, these few details are given by A. von Wurzbach, Niederländisches Künstler-Lexikon, 3 vols., Leipzig/Vienna 1906-11, II, following Von Boehm in H. Thieme and F. Becker (eds.), Allgemeines Lexikon der bildenden Künstler von der Antike bis zur Gegenwart, 33 vols., Leipzig 1907-50. Ten years later Diáz Padrón recognized it as the work of Willeboirts Bosschaert,²²²²M. Díaz Padrón, ‘Miscelánea de pintura flamenca del siglo XVII fuera de España’, Archivo Español de Arte 41 (1968), pp. 237-44, esp. pp. 237-38. The relevant volume of the Archivo Epsañol de Arte (XLI) is missing from the museum’s run. This might explain the museum’s subsequent failure to acknowledge the rejection of its, by then, traditional attribution to Antolinez. and was to publish a version in the Prague National Gallery as also by him.²³²³M. Díaz Padrón, ‘Un lienzo de Thomas Willeboirts Bosschaert identificado en la Galeria Narodni de Praga’, Goya Revista de Arte 185 (1985), pp. 290-92. Heinrich, in his recent monograph on the artist, accepts the Rijksmuseum picture as the prime original, and describes the Prague picture as a workshop replica.²⁴²⁴A. Heinrich, Thomas Willeboirts Bosschaert (1613/14-1654): Ein Flämische Nachfolger Van Dycks, 2 vols., Turnhout 2003, I, pp. 230-31, no. A52a; J. Douglas Stewart, ‘Thomas Willeboirts and Pieter Thijs: A Tale of the Tangled Antwerp Painters; with an Excursus on Van Dyck’s St Felix of Cantalice’, in H. Vlieghe (ed.), Van Dyck 1599-1999: Conjectures and Refutations, Turnhout 2001, pp. 271-88, esp. p. 285 under note 4, believed the opposite, that the Rijksmuseum painting was a version of that in Prague. In fact there are differences between the two paintings and weaknesses in the Amsterdam picture that point to later interventions. Two other likely workshop replicas are recorded in the RKD.²⁵²⁵Breda, Akademie voor Kunst en Vormgeving St. Joost, set in panelling, shaped, measurements not known; and anonymous sale, Paris (Piasa), 26 March 2010, no. 95, canvas, 130.5 x 119.5 cm. Their status given here is proposed only after judging from electronic images, which in the case of the former is of poor quality.

Conservation undertaken in 2001-03 revealed many pentiments in the Rijksmuseum painting. The support was not assembled as a whole before it was painted. X-radiographs of the two pieces of canvas, which make up the upper strip (showing only ‘sky’), show that they are fragments from another painting. They are fixed to the central canvas by lining, and thus subsequent to the artist’s work. The bottom strip (showing lower part of globe and of the angel’s leg) was stitched to the central canvas; as the paint along the latter’s edge was already dry, it broke up during the process, from which it can be inferred that the stitching took place sometime after the paint was applied.²⁶²⁶Notes by Tauber, RMA object file. These additions would thus have been made at different times.

Wallert and Tauber believe that the outlines of the figures in both the Amsterdam and Prague pictures result from tracing from the same cartoon.²⁷²⁷A. Wallert and G. Tauber, ‘Over herhalingen in de schilderkunst. Het probleem van reproductie’, Bulletin van het Rijksmuseum 52 (2004), pp. 316-27, esp. p. 319. It may be that the Rijksmuseum support bore much of the beginnings of Willeboirts Bosschaert’s painting campaign, which included the major alteration to the white cloth on which Christ stands. Then, it may be surmised, after having turned to a second support, and completing the Prague version (in which no pentiments are visible to the naked eye), he returned to his first version and made small alterations to it (including bestowing a frontal gaze to the Virgin and reducing the fall of her veil).²⁸²⁸Notes by Tauber, RMA object file. The other studio variants might have been made at about the same time. At some later stage, perhaps even after the artist's death, the lower part of the Rijksmuseum painting may well have been damaged and then replaced by stitching on the replacement strip of canvas, on which the design was repeated although allowing for a simplified rendering of the red drapery. The additional strip at the top would have been applied at a later stage.

As Heinrich suggests, the composition was inspired by Van Dyck. Vey believes that his original is lost; he identifies it with the large picture Matthijs Musson offered Princess Amalia van Solms in 1645: ‘A Madonna painting by Anthony van Dyck with two angels; one plays a lute the other a violin, high 5 feet one inch, wide 4 feet and a half [145 x 128 cm] without a frame.’²⁹²⁹Vey in S.J. Barnes, N. de Poorter, O. Millar and H. Vey, Anthony van Dyck: A Complete Catalogue of the Paintings, New Haven (Conn.)/London 2004, under no. III.A4. The original Dutch reads: ‘Een Maerien belt van Antoni van Dyck met twee Enghels; den eenen spelt op de luyt den anderen spelt op de velons, hooch 5 voeten eene duym, de breede 4 voeten ende een half (145 x 128 cm) sonder lyst’. Extant are two versions that accord with this description, neither of which is accepted by Vey; Stewart believes that one of these, with more or less the same dimensions, is by Willeboirts Bosschaert.³⁰³⁰J. Douglas Stewart, ‘Thomas Willeboirts and Pieter Thijs: A Tale of the Tangled Antwerp Painters; with an Excursus on Van Dyck’s St Felix of Cantalice’, in H. Vlieghe (ed.), Van Dyck 1599-1999: Conjectures and Refutations, Turnhout 2001, pp. 271-88, p. 271 and fig. 2. Both versions show prominently the globe of the world and the serpent. The Rijksmuseum composition is far more compact; nevertheless, Van Dyck’s lost composition, or the extant derivations – including the prints after it³¹³¹S. Turner and C. Depauw, The New Hollstein Dutch and Flemish Etchings, Engravings and Woodcuts, 1450-1700: Anthony van Dyck, 8 vols., Rotterdam 2002, VII, no. 552, pp. 232-38. – could well have been influential in its creation.

Heinrich dates the Amsterdam picture between 1646 and 1648 which, in view of what happened to the support, should be taken as referring to the central component. The face of the singing angel is similar to that of the singing nymph in the Aschaffenburg Rinaldo and Armida, as he observed.³²³²A. Heinrich, Thomas Willeboirts Bosschaert (1613/14-1654): Ein Flämische Nachfolger Van Dycks, 2 vols., Turnhout 2003, I, no. A45, where dated c. 1647; he made this observation in A. Heinrich, ‘Thomas Willeboirts Bosschaert, Bergen op Zoom 1613/14-Anvers 1654’, in P. Ramade et al., Meesters van het zuiden: Barokschilders rondom Rubens/Peintres baroques des Pays-Bas du sud, exh. cat. Valenciennes (Musée des Beaux-Arts de Valenciennes)/’s-Hertogenbosch (Noordbrabants Museum) 2000, pp. 101-21, esp. p. 111. Christ’s gesture of crowning is similar, but in reverse, to the putto crowning Amor in Willeboirts Bosschaert’s Love Triumphant, ex-Mauritshuis.³³³³A. Heinrich, Thomas Willeboirts Bosschaert (1613/14-1654): Ein Flämische Nachfolger Van Dycks, 2 vols., Turnhout 2003, I, no. A9, where dated 1643/44.

Gregory Martin, 2022

A. Heinrich, Thomas Willeboirts Bosschaert (1613/14-1654): Ein flämische Nachfolger Van Dycks, 2 vols., Turnhout 2003, I, no. A49

1825, p. 23, no. 87 (as manner of Van Dyck); 1843, p. 18, no. 82, (the Virgin identified as Mary Magdalen, and described as a copy); 1853, p. 35, no. 367 (as an unknown Spanish master, Holy Family, valued at fl. 500); 1858, p. 188, no. 401 (Spanish School, seventeenth century, Glorification of Mary); 1880, p. 398, no. 467 (as attributed to Van Dyck); 1885, p. 69, no. 467 (as Van Dyck, with note that it was probably the work of José Antolinez); 1888, p. 2, no. 11 (as Antolinez); 1903, p. 32, no. 367 (as Antolinez, attribution due to Bredius); 1934, p. 30, no. 367 (as Antolinez, but referring to Allende Salazar’s attribution to Miguel de Amberes); 1960, p. 20, no. 367 (as Antolinez); 1976, p. 85, no. A 598 (as Antolinez, The Apotheosis of the Virgin)

G. Martin, 2022, 'Thomas Willeboirts Bosschaert, The Virgin Crowned by the Infant Christ, in or after c. 1646 - in or after 1646', in Flemish Paintings in the Rijksmuseum, online coll. cat. Amsterdam: hdl.handle.net/10934/RM0001.COLLECT.5797