This sketch-like and swiftly executed study, presumably painted from life, was used for the head of an Apostle looking up towards the ascending Virgin in Rubens’ The Assumption of the Virgin in the Liechtenstein Collection, Vienna (fig. 1).¹ This large-scale work on canvas was commissioned in circa 1635 as an altarpiece for the Carthusian Church in Brussels and was probably completed by circa 1637. It was already in the Liechtenstein collection by 1643, probably acquired by Prince Karl Eusebius, and hung as the high altar in the Schlosskirche of Feldsberg in southern Moravia from the late seventeenth century. A sketch-like modello for it, often dated circa 1636–37, is in the Yale University Art Gallery, New Haven.

Rubens’ use of study heads for his larger multi-figural compositions is well documented. Rubens would make the studies from life, often from a variety of angles and, earlier in his career, sometimes with multiple heads on a single panel, providing him with a cast of vivid characters with which he could populate his history paintings.

It is a moot point whether this head sketch was made in preparation for the Liechtenstein Assumption, or used from an existing repertoire, but the latter is more likely, for several reasons. There are small differences between the position and angle of the heads, but a larger one in the hair and beard: the model for the sketch has grey hair with flecks of even paler grey, and a greyish-white beard, whereas the Apostle in the altarpiece has darker hair and beard, albeit with some strands of grey. In the present sketch Rubens is particularly interested in the hair, also the face, but the beard and neck are rapidly added. The brushwork is free and energetic throughout, creating a head of impressive vivacity. The grey bearded man to the left of the corresponding head in the altarpiece also appears to be the same model.

Dendrochronology also points towards a slightly earlier span of dates, not because of the inevitably imprecise dating that the hardwood rings yield, but because panels made from planks sawn from the same Western European oak tree that supplied the plank that comprises the lower three-quarters of the present panel were also used as the supports of seven other oil sketches by Rubens.² Three of these date from circa 1630–32, one from circa 1632–33, two from 1635 and one from circa 1635, which points to a date of execution of this sketch of circa 1630–35, as the dendrochronology report concludes.³ This chronology does not rule out the present sketch being a derivation from the figure in the altarpiece, but viewed in the context of the marked differences between the heads depicted in each, it makes it much less likely. While it might initially seem odd that all these works are sketches and not finished paintings, on the whole Rubens seems to have used finer grained Baltic oak for the panels of finished works, and coarser and irregularly cut planks of Western European oak for sketches, which require a less finished support.

This also makes it implausible that this sketch could be by Van Dyck, to whom it was formerly attributed.⁴ Van Dyck left Antwerp for London in 1632. He was no longer closely associated with Rubens and his activity as a painter of oil sketches in his second Antwerp period (and his preceding Italian sojourn) is, as far as we know, limited to loose compositional oil sketches en brunaille. He painted figural and head sketches when active in Rubens’ atelier in the late 1610s and around 1620, whereas the last hardwood ring of the tree from which the panel was cut is 1611, and the likelihood is that it was felled after 1619, and used for this sketch, as the evidence strongly suggests, circa 1630–35.

We are grateful to Dr Christopher Brown for endorsing the attribution to Rubens.

¹ See J. Kräftner et al., Peter Paul Rubens 1577–1640. The Masterpieces from the Viennese Collections, Vienna 2004, pp. 360–66, no. 90, reproduced.

² Dendrochronological Consultancy Report 1303, Ian Tyers, 21 July 2021. A copy of this report is available upon request. The uppermost quarter of the present panel comprises two irregularly cut planks that are too small to yield dendrochronological information. The outermost hardwood ring of the principal plank dates from 1611, so the earliest conceivable date of use would be after circa 1619.

³ These are three sketches for the Life of Achilles tapestry cycle of circa 1630–32: Achilles instructed by Chiron, Achilles dipped into the River Styx and Thetis receiving armour from Hephaestus, all Museum Boymans-van Beuningen, Rotterdam; Cecrops’ Daughters finding Erichthonius, circa 1632–33, in the National Museum, Stockholm; Hercules and Minerva expelling Mars, in the Museum voor Schone Kunsten, Antwerp; The Voyage of the Prince from Barcelona to Genoa, in The Harvard University Art Museums; and The Meeting of The Two Ferdinands at Nördlingen, in the Getty Museum, Malibu (the last three all circa 1635). A plank sawn from the same tree also served as the support of a landscape by Rubens’ friend Lucas van Uden.

⁴ Max J. Friedländer wrote a certificate for it on 4 November 1954, as Anthonis van Dyck.