en français


Portrait of Maximilian, Archduke of Austria

Workshop of Peter Paul Rubens (Flemish 1577-1640)

Maximillian, Archduke of Austria, 1621-1640

oil on canvas

36.6 x 30 cm

Gift of Herman Levy, Esq., O.B.E., 1984


To view an interactive image comparison of this artwork:

Click here


Harry Axelsohn Johnson, Stockholm

Sven Boström, Stockholm

London, England

Christie’s London, “Pictures by Old Masters,” 15 April 1966. Catalogued as by Del Mazo, “Head of a Man with Grey Beard and White Ruff,” no. 125.

Purchased from Roland, Browse & Delbanco, London, England, 1966

Herman Levy, Esq., O.B.E.


Exhibition History

The Herman H. Levy Collection, McMaster Museum of Art, Togo Salmon Gallery, 11 June 1994 – 9 November 2006

Levy Series, Northern Art in the Age of Cock, Durer, Rubens & Rembrandt, McMaster Museum of Art, Levy Gallery, 10 May – 25 August 2007

The Last Things Before the Last, McMaster Museum of Art, Levy Gallery, 24 May – 18 August 2012

Worldly Possessions: Visualizing Ownership in the Age of the Baroque, McMaster Museum of Art, Levy Gallery, 27 August 2013 – 25 January 2014


Ihor Holubizky

Peter Paul Rubens was born in the German city of Siegen. The family moved to Cologne, where Rubens’s father, Jan, died in 1587.[1] Two years later, Rubens’s mother and family moved to Antwerp, where Rubens received a humanist education in Latin and classical literature before commencing his painting apprenticeship at the age of fourteen, first with the Flemish artist Tobias Verhaecht, followed by others.[2] To further his education, Rubens lived and worked in Italy between 1600 and 1608, where he studied the work of Renaissance artists and received his first commissions. He returned to Antwerp in 1609, establishing a studio and workshop (the most prominent of his apprentices was Anthony van Dyck). Working in the Baroque style, Rubens produced allegorical and mythological subjects, portraits, history paintings, commissioned altarpieces, and landscapes.[3] As Hans Vlieghe noted, “After his return to Antwerp…Rubens not only created a new, Baroque style of history painting but also liberated the art of portraiture from a certain schematicism which had beset it since the middle of the sixteenth century.”[4] His mature period work and “animation by the flesh”[5] has generated the common usage term “Rubenesque.” The Rubens workshop also produced copies of paintings by other artists (a popular and accepted practice in the sixteenth and early seventeenth centuries),[6] as well as prints, engravings, and woodcuts from Rubens’s paintings, for which the artist established copyright in the Low Countries, England, France, and Spain.[7]

Rubens was an astute manager of his prolific workshop production, and was a collector of art and antiquities, which enabled him to amass considerable wealth[8] and achieve the status of diplomat, a consequence of his connections to nobility and his own personal ambitions beyond art alone.[9] Rubens lived and worked in Spain and England after 1621, before returning to Antwerp in 1630, where he died 10 years later.

The two outstanding questions regarding the McMaster painting are, is it attributable to Peter Paul Rubens and who is the subject?

The sitter’s identity was determined through a comparison with a three-quarter portrait of Maximilian illustrated in the Corpus Rubenianum Ludwig Burchard Part XIX (2) (catalogue raisonné no. 122[10]): the subject is Maximilian III, Archduke of Austria (1558–1618). It is described as a copy of Rubens’s original (no. 120[11]), which was possibly painted from life in 1615. A subsequent commissioned copy from Rubens (no. 121[12]) was most likely executed in 1619 for Maximilian’s successor, Archduke Leopold-Wilhelm.[13] The current whereabouts for both are unknown, and they are presumably lost.

A further comparison between the published and the McMaster painting was inconclusive on several counts. The comparison was based solely on a black and white image, and as the Rubens cataloguer Hans Vlieghe noted for the catalogued copy, “the portrait certainly displays the severity of design characteristic of Rubens’s style from c. 1615 to c. 1630 [but] I do not believe this to be an autograph [by Rubens’s hand] example [and] I believe the painting to be a studio replica untouched by Rubens.”[14] Vlieghe also noted that he had never examined the work, and his assessment was based on a photograph.[15]

The technical examination of the McMaster painting could not establish that it was done prior to Rubens’s death in 1640, only that the materials are consistent with the period as well as with paintings done after the artist’s time. The “burden” of the attribution, therefore, returns to connoisseurship, and it is complicated, as noted: working from photographic evidence and the recovery of useful documents (the paper trail chase).

We know that Rubens kept the originals of his “political” portraits (as well as other subjects) to be executed by apprentices and, depending on the merit or value of the commission, completed or finished by Rubens himself. They could also be done in variations.[16] The continuing investigation, therefore, turned to the ten catalogued Maximilian “copies”; six of these are paintings (adding to the confusion, no. 122 was included in this list). In due course, black and white photographs were uncovered for copy 3 and copy 5. The former, a same-composition three-quarter portrait, was formerly in the Bavarian State collections, and presumably destroyed in 1943 during the Second World War[17]. Copy 5 was sold from the collection of the 8th Duke of Buccleuch, Scotland, at Christie’s London in 1959.[18]

A comparison of the two copies of the three-quarter copy portraits of Maximilian show strong facial similarities (the McMaster painting is the face only), but there are differences in the background details and modelling of the face. Nonetheless, it is possible to suggest that the painters of both three-quarter portraits, if by workshop apprentices, had access to the Rubens original,[19] and the different compositions can be attributed to variations for clients. The unanswerable question is, in which order were they painted?

Other comparisons can be extrapolated from engravings after the Rubens portrait of Maximilian by Lucas Vorsterman (1595–1675), Pieter de Jode II (1606–74), and Jonas Suyderhoef (1613–86).[20] The features of the Vorsterman engraving are the closest to the paintings cited (although it is a reversed image, as a consequence of the engraving print process), and to that of the McMaster work. It offers a tempting connection as Vorsterman worked in Rubens’s studio from 1618–21, and therefore he could have had access to the original painting.[21] But the Vorsterman-Rubens working relationship ended with a dispute over copyright, and may offer an explanation as to why the Vorsterman engraving was not published until 1650, ten years after Rubens’s death.[22] However, it must be emphasized that the Vorsterman engraving only confirmed the identity of the sitter as well as evidence of facial similarities to the McMaster painting.

The link between the McMaster painting and the Corpus Rubenianum research came unexpectedly. Correspondence with Rubens House in Antwerp (which holds the Rubens catalogue raisonné files) that sought to locate a colour image of no. 122 uncovered an image of the McMaster painting. It is copy 6, whose provenance is recorded as Sven Boström, Stockholm, and a near-conclusive connection to the workshop of Peter Paul Rubens. The dimensions, very close to the McMaster painting, had been noted early on, but this “accidental” discovery gives evidence of the often protracted and circuitous route in research rather than a flawed process or methodology.

Questions still remain unanswered. Why did the Rubens catalogue raisonné research, published in 1987, “end” with the Boström provenance? How did it get to a London gallery, and why was it catalogued as a work by the Spanish painter Juan Bautista Martínez del Mazo in the Christie’s sale of 1966, yet suspected to be an early Rubens study by Henry Roland of the London dealers Roland, Browse and Delbanco, who purchased it at auction. Roland presented the work to the trustees of the National Gallery, London, for purchase acquisition shortly after. The offer was declined, and the minutes of the trustees’ meeting record the following: “It had turned out not to be a sketch of a head, but apparently had been cut from a larger portrait.”[23] This cropping, from a larger work, was also proposed in the technical examination of the McMaster painting prior to the recovery of the National Gallery records. Could the McMaster fragment have been cut from one of the other three recorded copies (whereabouts unknown), or perhaps cut from a yet undiscovered copy?

The research into the McMaster painting, while still unable to establish a definitive answer, is not a failure of the process. It highlights the ever-present challenges in art history: incomplete and fractional records over a long period of time, factors of connoisseurship and the “authority of the eye,” and how much data is retrievable from scientific methods. The case is not “closed” on the McMaster painting, but for now it appears to be the only version for which the whereabouts are known of Rubens’s 1615 original.

[1] Christopher White, Peter Paul Rubens, Man and Artist (New Haven, CT: Yale University Press, 1987), 1–51; Pierre Cabanne, Rubens (London: Thames and Hudson, 1967), 18–20.

[2] Other artists cited in Rubens’s early apprenticeship are Adam van Noort and Octave van Veen. Cabanne, Rubens, 29. Also see White, Peter Paul Rubens, 4–8.

[3] In essence, the majority of Rubens’s paintings are commissions, rather than what we think of as “exhibition” works. See Francis Haskell, “Feast Days and Salerooms,” in The Ephemeral Museum, Old Master Paintings and the Rise of the Art Exhibition (New Haven, CT: Yale University Press, 2000), 8–9.

[4] Vlieghe, Rubens Portraits, 21.

[5] Cabanne, Rubens, 188.

[6] Jeffry M. Muller, Rubens: The Artist as Collector (Princeton, NJ: Princeton University Press, 1989), 59.

[7] Julius S. Held, Rubens and His Circle (Princeton, NJ: Princeton University Press, 1982), 115.

[8] The 1645 accounting of Rubens’s estate amounted to 283,987 florins. Muller, Rubens, 59. A current exchange equivalent is difficult to determine, but it could be in excess of USD15 million, keeping in mind that there were fewer millionaires in seventeenth century Europe in comparison to today’s global industrial economy. See “Rubens’ Remuneration,” last modified March 2012, http://ringlingdocents.org/rubens/dutchmoney.htm. As another comparison, Rubens’s inheritance from his mother was 200 florins. Muller, Rubens, 49.

[9] See C. V. Wedgwood, The Political Career of Peter Paul Rubens (London: Thames and Hudson, 1975), 13, where he contextualizes the period: “The span of Rubens’s life…covers the latter end of the Counter-Reformation, an epoch in which religious wars, which were always half political and often wholly so, repeatedly involved most of western Europe. [Furthermore] the prolonged power struggle [of] the Thirty Years War covered the latter part of his life and figures prominently in his correspondence.” It should also be noted that the status of diplomats and nature of diplomacy had a very different contour in the volatile political scene of Europe in the first half of the seventeenth century. Rubens was knighted by Phillip IV, King of Spain, and Charles I, King of England. In essence, he was a trusted free-agent confidant to many sides. Muller also notes that Rubens’s sale of antiquities from his collection to the 1st Duke of Buckingham allowed him the financial independence to pursue diplomatic work. Muller, Rubens, 62. See also “Rubens’ First Brush with Britain,” Phaidon News, accessed March 22, 2015, http://ca.phaidon.com/agenda/art/events/2011/november/22/rubens-first-brush-with-britain/.

[10] Vlieghe, Rubens Portraits, 135–36. This version is recorded as the collection of Heinz Kisters (1912–1977), a German art dealer and collector who specialized in old masters. The connection between the McMaster painting and the Rubens portrait of Maximilian was first identified by Dr. Jeremy Wood, University of Nottingham, and Dr. Gerlinde Gruber, Kunsthistorisches Museum, Vienna, in correspondence with Nenagh Hathaway in early 2013. Why this seemingly obvious evidence was previously overlooked is addressed in the publication essay “Limits of the Eye and the Engine of Curiosity.”

[11] Ibid., 134.

[12] Ibid., 135–36.

[13] In a circa 1641 David Teniers painting of the “gallery” of Archduke Leopold-Wilhelm in Brussels (collection of Staatsgalerie Schleissheim) Rubens’s portrait of Maximilian appears in the lower-right corner. Vlieghe, 134, cited as copy (1); illustrated as fig. [14]5.

[14] Vlieghe., 135.

[15] Ibid.

[16] Muller, Rubens, 60; Vlieghe, Rubens Portraits, 31.

[17] Dr. Andrea Christine Bambi, senior curator and head of art provenance research for the Bayerische Staatsgemäldesammlungen, Munich, e-mail message to author, September 9, 2014. Vlieghe records the Munich Maximilian painting as having been “lost since 1945” (Vlieghe, Rubens Portraits, 134), whereas Bayerische Staatsgemäldesammlungen records indicate that it was presumably destroyed in 1943; there was an RAF bombing of Munich in October 1943. The discrepancy in dates can be explained as a “margin of error” on Vlieghe’s part.

[18] Christie’s London, e-mail message to author, August 1, 2014; Buccleuch Collection archivist, e-mail message to author, September 3, 2014.

[19] The Kisters and Munich paintings are also very close in size 115 x 93 cm and 117.5 x 95.5 cm, respectively. It should be noted, however, that both can only be regarded as “sight measurements” and would have to be remeasured to confirm accuracy, which is, of course, impossible for the lost Munich painting.

[20] The Jode II and Suyderhoef are “identical,” although oriented differently. The facial handling differs from that of Vorsterman. Jode II did engraving work for Anthony van Dyck, but there is no other Jode connection to the Rubens workshop. Suyderhoef’s first dated engraving is 1641, after Rubens’s death. Examples of the Jode II, Suyderhoef, and Vorsterman engravings are in the collection of the British Museum, last modified February 14, 2015, http://www.britishmuseum.org/research/collection_online/search.aspx.

[21] Held, Rubens and His Circle, 114.

[22] Copyright for prints was extended beyond Rubens’s death in Spain. See Held, 115.

[23] National Gallery archives: ref. NG1/14 NG, Board Minutes, 5 May 1966. Alan Crookham, research centre manager, the National Gallery, London, e-mail message to author, July 23, 2014. Letters from Henry M. Roland of Roland, Browse and Delbanco Gallery, London, to Herman Levy on April 21, 1966, and on May 10 and 12, 1966 (invoice sent May 6), confirms the purchase arrangement. McMaster Museum of Art files.


Nenagh Hathaway

A gentleman with a white-grey beard and large moustache, depicted against a dark background, stares directly at the viewer (Fig. 1). He wears a fine ruff, the only indication of costume in this portrait. The relatively small work is linked to several other portraits of this sitter produced by the Rubens workshop, including a larger-format painting and an engraving by Lucas Vorsterman, a printmaker who worked for Rubens.

Rubens was a prolific artist celebrated for his monumental history paintings, altarpieces, landscapes, and portraits. While the latter was not a subject he preferred, as is clear from his correspondence, Rubens and his large workshop created many portraits over the course of his illustrious career.[3] The participation of workshop assistants was common, and in surviving documents Rubens himself acknowledges their important role in reference to the cost of his paintings. By employing assistants, Rubens could sell works at a range of costs depending on the degree of his personal involvement in their creation. Assistants and pupils would imitate Rubens’s technique quite closely, sometimes to the point that their participation is difficult to discern. The McMaster portrait does not display the characteristic appearance of a mature Rubens portrait, which suggests that if it was produced by an assistant in the master’s studio it was probably not intended for one of his wealthier clients.

The portrait depicts Maximilian III (1558–1618), Archduke of Austria and Count of Tyrol. This identification was accomplished through comparisons between the subject of the McMaster painting and other surviving portraits.[4] One such example is a highly similar knee-length picture of Maximilian (Fig. 2). Hans Vlieghe, a noted Rubens expert, considers this portrait, whose current location is unknown, to be a studio replica of a lost original by Rubens.[5] Although the knee-length version is considerably larger than the McMaster portrait, the dimensions of the heads in the two works compare well, and share extremely similar facial features. The remarkably close resemblance between the sitters establishes a connection between the McMaster painting and Rubens’s studio.[6] The robust similarity suggests that the painter of the McMaster portrait was well aware of the larger version, or of the lost original by Rubens, which could have served as the model for both.

An engraved likeness of Maximilian III by Vorsterman is also very close to the McMaster portrait, especially when we are reminded that an image engraved in the copper plate becomes mirrored when it is printed (Fig. 3).[7] Vorsterman (1595–1675) was a member of Rubens’s studio from about 1617–18 until the early 1620s. Rubens was able to generate additional income for himself from Vorsterman’s reproductions. The engraving includes the shoulders and chest of the sitter, which links the work to the larger-format painting. However, it is possible that the McMaster painting originally included these details found in the engraving and the knee-length portrait. The X-radiograph (Fig. 4) reveals significant damages around the edges of the McMaster portrait, which indicates that it was indeed cropped on all sides.[8] However, at the moment it remains impossible to determine the precise relationship between the engraving, the lost Rubens original, the McMaster portrait and Fig. 2.

On close inspection, the dark background of the McMaster painting is actually green-blue in colour. Coincidentally, this colour is similar to Vlieghe’s description of the stiffly hanging curtain behind Maximilian in the knee-length portrait.[9] Rubens used this visual strategy for emphasizing the sitters in other portraits.[10] Whether or not the McMaster painting ever featured a curtain behind Maximilian’s head cannot be confirmed. A cross section from the area demonstrates that the blue-green paint in the McMaster portrait is original, although it may have darkened slightly over time (Fig. 5).

A version of the knee-length portrait is depicted in the “gallery painting” of Archduke Leopold Wilhelm’s collection by David Teniers the Younger (Fig. 6), but it is again unclear whether this represents Fig. 2, the lost original by Rubens, or yet another version. The existence of what appear to be serial reproductions of varied quality attests to the fact that portraits of Maximilian were in high demand. It seems quite likely that an autograph Rubens portrait of Maximilian existed and that this likeness was popularized and circulated by copies—and copies based on copies—in different media.[11] These versions could be distributed to various locations for personal, political, or other motivations. A commemorative function is also possible, which would imply that the portrait was painted after November 2, 1618, the day Maximilian III died.[12]

[1] Hans Vlieghe, Rubens Portraits of Identified Sitters Painted in Antwerp (London: Harvey Miller Publishers, 1987), 134.

[2] The vendor to the Christie’s sale in April 1966 was a London gallery, but there are no further details at the time of writing because of a Christie’s 50-year confidentiality policy. Christie’s London, e-mail message to author, May 18, 2015.

[3] In a letter to Annibale Chieppio, Rubens writes, “Then I should not have to waste more time, travel, expenses, salaries (even the munificence of His Highness will not repay all this) upon works unworthy of me [portraits], and which anyone can do to the Duke’s taste.” See Ruth Saunders Magurn, trans. and ed., The Letters of Peter Paul Rubens (Cambridge, MA: Harvard University Press, 1955), 38.

[4] This identification was made by Dr. Jeremy Wood, University of Nottingham, and confirmed by Dr. Gerlinde Gruber, Kunsthistorisches Museum, Vienna.

[5] Vlieghe acknowledges that his analysis is based on a photograph and, presumably, the descriptions of others who had seen the work—or a colour image of the portrait—in person. Hans Vlieghe, “Rubens. Portraits of Identified Sitters Painted in Antwerp,” Corpus Rubenianum Ludwig Burchard, vol. 19, 2 (London and New York: Oxford University Press, 1987), 135–136, no. 122, fig 146.

[6] Vlieghe also notes a resemblance between the hairstyle and facial expression of the Corpus portrait and a portrait of a (much younger) Maximilian III by Martino Rota (c. 1580) at the Kunsthistorisches Museum, Vienna. Attempts by Ihor Holubiszky to secure a colour photograph of the larger-format portrait resulted in the identification of the McMaster painting as copy number 6 in Vlieghe’s Corpus list for CR 121.

[7] Two other prints of Maximilian are known, one produced by Pieter de Jode and the other engraved by Jonas Suyderhoef. These do not resemble either the McMaster or Corpus portraits as closely as the Vorsterman image.

[8] This had been noted around the time of purchase by the Director of the National Gallery, London, Sir Philip Hendy, who also thought the work was created in the early part of Rubens’s career. See the archives at the Research Centre, National Gallery London, ref. NG1/14 NG, Board Minutes 5 May 1966.

[9] Vlieghe, “Rubens,” 135.

[10] The Infanta Isabel Clara Eugenia at the Prado is a good example.

[11] The Corpus Rubenianum (vol. 19, 134–135) includes three entries for portraits of Maximilian, Archduke of Austria. The portrait of Maximilian catalogued as CR 120 is listed as a work from the Rubens estate of 1640. Vlieghe considers this to be the original Rubens painting from which the other non-autograph works are derived. These copies are found under the entries CR 121 and 122. The provenance description for CR 121 (also Corpus vol. 19, 134) states that the work may have been painted for Archduke Leopold-Wilhelm of Austria in 1619. Vlieghe also inserts a list of copies in the CR 121 entry, one of which (number 6) is the McMaster painting. This list includes paintings, engravings, and drawings.

[12] This also coincides with the presence of Vorsterman in Rubens’s studio from about 1617–18.