Category: Rembrandt and His Pupils

Rembrandt painted more self-portraits than any artist before him, somewhere between forty and eighty depending on where one draws the line between a self-portrait and a tronie using the artist’s own face as a convenient model. Over a forty-year career he cast himself as soldier, burgher, man in an exotic turban, apostle Paul, the ancient painter Zeuxis. Some of these costumed performances continued into the 1660s: the Self-Portrait as the Apostle Paul dates from 1661, the Self-Portrait as Zeuxis from around 1662. But the Self-Portrait with Two Circles c. 1665–69, does something different. The roles have been put down. There is no historical disguise, no borrowed identity. A painter holds his tools and looks at himself in a mirror, and that is all.

He stands in shallow space before a pale wall on which two large arcs are faintly inscribed. He wears a fur-lined tabard over a crimson doublet, a plain linen cap on his head, and holds a palette, brushes, and mahlstick in his left hand. His right hand disappears into an unresolved blur at his hip. At the extreme right edge of the canvas, a narrow vertical strip reveals the edge of a painting he is working on, turned away from the viewer. There is no signature. The face is the most worked area of the surface: built up in dense, layered strokes of ochre, grey, and warm brown, with subtle glazes around the eye sockets and sharper impasto catching the light on the nose and ear. The lower body, by contrast, remains schematic, the tabard blocked in with broad, flat passages that were clearly never brought to completion.

X-ray examination has revealed that Rembrandt substantially altered the composition during execution. In the earlier design, he was turned further to his left with his arm raised in the act of painting, tools in his right hand, faithfully copying his reversed reflection in the mirror. He then corrected for the reversal, moved the palette and brushes to his left hand, realigned his body with the picture plane, and tucked the empty right hand out of sight. In the first version, the subject was a man painting. In the second, the subject is a painter. The distinction is not trivial. Rembrandt revised the composition away from activity and towards identity, from something he was doing to something he was.

By 1665, painting was very nearly the only thing left. He had been declared insolvent in 1656, and his house on the Sint Anthonisbreestraat and his extensive collection of art, antiquities, and curiosities were sold at auction over the following years. To circumvent the legal consequences of the bankruptcy, his partner Hendrickje Stoffels and his son Titus established an art-dealing business in 1660 in which Rembrandt was technically their employee, working under their names. Hendrickje died in 1663. Titus married in February 1668 and died that September, at twenty-seven. Rembrandt survived him by just over a year, dying on 4 October 1669 and buried in an unmarked rented grave in the Westerkerk. He continued to receive significant commissions during these years, including The Conspiracy of Claudius Civilis for the new Amsterdam Town Hall (1661–62) and The Syndics of the Drapers’ Guild (1662), so the story is not simply one of abandonment. But the world that had sustained him, the prosperity, the collection, the household, the people, had fallen away, and what remained was the work.

The two circles on the wall behind him have generated an extraordinary volume of interpretation for marks that may or may not carry any iconographic freight. The most persistent reading connects them to Vasari’s account of Giotto drawing a perfect circle freehand to prove his skill to a papal emissary, a story that itself echoes Pliny the Elder’s account of the contest between the ancient painters Apelles and Protogenes, in which artistic supremacy was demonstrated through the drawing of increasingly fine lines. On this reading, the circles are a late, defiant claim to mastery. Others have proposed that they represent the hemispheres of a double-hemisphere world map, a familiar furnishing in Dutch interiors, implying a claim to universality. More speculative suggestions invoke the rota Aristotelis or even kabbalistic symbolism. The most prosaic possibility is that they are compositional devices placed to balance the off-centre figure, or incidental studio marks left visible in a painting that was never resolved. Digital analysis has confirmed that the arcs are geometrically precise, almost certainly drawn with a compass rather than freehand, which at the very least complicates the Giotto interpretation in its simplest form. The circles remain genuinely opaque, and it is just impossible to know for sure what they mean, if they mean anything at all.

What is less ambiguous is what the painting says about its maker without recourse to symbolism. Rembrandt at sixty stares out, or rather stares at himself in a mirror, with an expression that resists the easy adjectives (defiant, resigned, serene) that have been applied to it. There is an alertness in the face, a professional attentiveness, that belongs to the act of painting rather than to mood. He is studying himself in order to get himself right on the canvas. The painting records that act and, by remaining unfinished, preserves it permanently in progress. Everything theatrical has been discarded. The rich costumes and historical disguises of the earlier self-portraits are gone. The tabard is working dress. The tools are held openly, as attributes. The circles, whatever they are, belong to the studio. And the gaze, which is not directed at us but at a mirror, is the gaze of a man whose relationship with his own face is entirely professional: it is the surface he is painting, nothing more and nothing less.

This is what gives the picture its particular force, and what separates it from the pathos that later centuries have projected onto it. It is tempting to read the painting as a confession of suffering, a record of endurance through loss. But the image itself does not suffer. It works. The face is the most intensely painted passage in the picture because it is the passage Rembrandt cared most about getting right, and the care is visible in every loaded stroke. The rest could wait, and in the end it waited forever. What the Self-Portrait with Two Circles finally shows is a man for whom painting was no longer a profession or a livelihood or a means of display, but the last and most durable form of being he had. Everything else could be taken away, and was. This could not.

References

Chapman, H.P. (1990) Rembrandt’s Self-Portraits: A Study in Seventeenth-Century Identity. Princeton: Princeton University Press

Alpers, S. (1988) Rembrandt’s Enterprise: The Studio and the Market. Chicago: University of Chicago Press

Bryant, J. (2003) Kenwood: Paintings in the Iveagh Bequest. New Haven and London: Yale University Press

Chenault Porter, J. (1988) ‘Rembrandt and His Circles: More about the Late Self-Portrait in Kenwood House’, in Fleischer, R. E. and Munshower, S. S. (eds.) The Age of Rembrandt: Studies in Seventeenth-Century Dutch Painting. University Park, PA: Pennsylvania State University Press, pp. 189–212

De Winkel, M. (2006) Fashion and Fancy: Dress and Meaning in Rembrandt’s Paintings. Amsterdam: Amsterdam University Press

Schama, S. (1999) Rembrandt’s Eyes. New York: Alfred A. Knopf

Van de Wetering, E. (2005) A Corpus of Rembrandt Paintings, vol. IV: Self-Portraits. Dordrecht: Springer

Rembrandt van Rijn (1606–1669), Portrait of Philips Lucasz., 1635. Oil on oak, 79.5 × 58.9 cm. The National Gallery, London

In the autumn of 1629 Batavia was under siege for the second time in two years, its wells fouled and its garrison thinned by cholera, and on 21 September the governor-general, Jan Pieterszoon Coen (1587-1629), died in the thick of it (Ricklefs 1993). Jacques Specx (1588/89–1652) reached the town in time for the funeral and stepped straight into the vacant office, and among those who had made the long passage out with him was his sister-in-law, Petronella Buys (c. 1605–1670). She found her husband in the Indies. Philips Lucasz (c. 1598–1640), already high in the Company’s service, married into the same family connection, and the couple came home to Amsterdam late in 1633 (Yeager-Crasselt 2019). The portrait Rembrandt painted of him a few years later is, before anything else, a picture about the sea and what Company service cost the men who lived by it.

By the time that commission reached him, Rembrandt van Rijn was the busiest portraitist in Amsterdam and, in a sense, not yet quite his own master. Since about 1631 he had lodged with the dealer Hendrick van Uylenburgh (c. 1587–1661) and run his portrait ‘academy’, a well-drilled shop that turned out likenesses for the city’s merchant class and trained younger hands such as Ferdinand Bol and Govert Flinck; he worked there until 1635, the very year of the Lucasz panel, when he set up on his own (van de Wetering 2015). He had married into the same world in 1634, taking Uylenburgh’s cousin Saskia (1612–1642) as his wife, and the preacher who married Lucasz and Petronella in the Nieuwe Kerk, Jan Cornelisz Sylvius, belonged to the Uylenburgh connection too, which gives some sense of how tightly these commissions and these households were laced together. Lucasz and his wife did not linger over the results. On 2 May 1635 they sailed again for the East, leaving the two panels behind with the family. Whatever else the picture is, its haste can be read in two registers at once, as the pressure of a departure that would not wait and as the disputed evidence of how much of it Rembrandt painted himself.

The man who set the commission in motion was no ordinary client. Specx had helped found the Dutch trading post at Hirado in Japan in 1609 before his spell governing the Indies, and by the mid-1630s he was back in Amsterdam, wealthy, and buying pictures with real ambition; among the works plausibly from his collection were Rembrandt histories, including the Abduction of Europa of 1632, now at the Getty, and the Storm on the Sea of Galilee later stolen from the Gardner in Boston (Yeager-Crasselt 2019). Whether he ordered the pendants outright or was given them by the sitters before they sailed was already an open question to I.H. van Eeghen, who leaned toward the second reading (van Eeghen 1956); either way, a patron of that kind, at ease with Rembrandt’s history painting, is the last man one would expect to fob off with a routine studio portrait, which is among the better arguments for taking both panels as autograph. There is a further point about function that tends to get lost in the attribution wrangling. These were pendants made to be left behind. A portrait that hangs in the family house while its subject is at sea for years does a particular kind of work, standing in for an absent man who may not come back, and in Lucasz’s case that is exactly how things fell out.

The likeness gives Lucasz the sober uniform of his rank: black tunic, broad white collar, and a heavy gold chain across the chest. The chain is the one element a contemporary would have read at once, since the Company’s directors, the Heeren XVII, awarded such chains to commanders who had brought the homeward fleet safely in, and Lucasz had steered exactly such a fleet into port in December 1633 (MacLaren and Brown 1991). What has kept cataloguers arguing is the handling. The collar is not built up in the ordinary way, pale lace over a dark ground; instead the painter laid down a solid block of lead white and scored the pattern of the lace into it in black, in places before the white had set, and the chain is struck in with the same economy, quick yellow strokes over the tunic that sit on top of the white where the links cross behind the collar rather than beneath it (Bomford et al. 2006). An X-radiograph shows the composition was once different again: the sitter’s left hand rested on the chain and was afterwards painted out, probably in the seventeenth century (MacLaren and Brown 1991). Someone later papered the back of the panel with fragments of a printed list of French naval appointments running from 1641 to 1673, a coincidence of subject the object seems to have attracted rather than sought.

For most of its recorded life the portrait passed without question as Rembrandt’s, and it took the Rembrandt Research Project to unsettle it. In the third volume of their Corpus, Josua Bruyn and Bob Haak proposed that Rembrandt had painted only the face and left the costume, the collar and the whole of Petronella’s pendant to an assistant, the flat modelling and summary lace being, on their reading, the marks of a lesser hand (Bruyn et al. 1989). Ernst van de Wetering (1938–2021) moved the line back and forth over the following decades and, in the revised Corpus of 2015, returned rather more of the work to Rembrandt (van de Wetering 2015), while the National Gallery’s own team held that both panels were autograph throughout and that the strange collar was ‘a kind of brilliant shorthand’, the sign of a fluent hand working at speed (Bomford et al. 2006). Gary Schwartz, Christian Tümpel and Walter Liedtke were among those never persuaded by the demotion (Yeager-Crasselt 2019). The disagreement has more bite than it first appears, because 1635 is precisely the year Rembrandt was closing out his time in Uylenburgh’s portrait shop and setting up alone, so the question of how much of a 1635 portrait a workshop hand might have done is not an abstraction imposed by later connoisseurs; it is a question about how the studio actually ran in the months this panel left it. A recent micro-XRF study of the closely related Portrait of Maertgen van Bilderbeecq has shown that the white-first, black-openings way of painting lace is fundamentally Rembrandt’s own habit (Schütt and Gerken 2025), which does weaken the case for an assistant in the collar; and yet a shop built to reproduce the master’s manner is exactly the place where such a habit would be taught, so the technical evidence narrows the question without closing it. Speed explains how the picture looks. It cannot, on its own, tell whose speed it was.

It is tempting to make Rembrandt himself a man of the East India trade, and worth resisting the temptation, because his connection to the Company was always at one remove. He never sailed and never held a post; what he had was an Amsterdam saturated by the Company’s cargoes, a clientele that included its officials, and, much later, a collection stocked with the things its ships carried home. The inventory drawn up at his insolvency in 1656 lists Asian curiosities, weapons and an album of Indian miniatures, and in the 1650s he made some two dozen drawings after Mughal portraits on Japanese paper, the one sustained encounter in his work with an art from outside Europe (Schrader 2018). Set the Lucasz portrait of 1635 beside those late sheets and the shape of the relationship comes clear. In the 1630s the Company was a source of sitters and money; by the 1650s it had become a source of images and ideas. The portrait of a Company officer with a gold chain and the drawings after a Mughal emperor sit at the two ends of one long thread, and Specx, who had lived in Japan and governed Batavia, is as good a figure as any to stand at its near end.

What sharpens all of this is that the sitter did not have long. Lucasz went back to the work the chain commemorated, and he died at sea early in 1640, on the passage back to Batavia from the Ceylon campaign that had just taken the Portuguese fort at Negombo ; the year is often given as 1641, though the voyage ties his death to the fighting of 1640. The hurried likeness of 1635 became, without anyone intending it, the last there would ever be. Petronella came home a widow and married again. The pendants were parted before the nineteenth century and have seldom been seen together since. Between them they record a plain and brutal arithmetic, in which a gold chain and a fortune were what the Company paid out for voyages that killed a fair number of the men who made them. As for the hand that once lay on that chain and was quietly taken away, it is hard not to wonder who removed it, and what they supposed they were tidying out of the picture.

References

Bomford, D., Kirby, J., Roy, A., Rüger, A. and White, R. (2006) Art in the Making: Rembrandt. New edn. London: National Gallery Company / New Haven, CT: Yale University Press, pp. 82–88 (cat. no. 5).

Bruyn, J., Haak, B., Levie, S.H., van Thiel, P.J.J. and van de Wetering, E. (1989) A Corpus of Rembrandt Paintings, Volume III: 1635–1642. Dordrecht: Martinus Nijhoff, cat. no. A 115.

Hofstede de Groot, C. (1913) ‘Rembrandts portretten van Philips Lucasse en Petronella Buys’ [Rembrandt’s portraits of Philips Lucasz and Petronella Buys], Oud Holland, 31(5), pp. 236–240.

Leupe, P.A. (1855) ‘Philip Lucasz., raad en directeur-generaal van Nederlandsch-Indië’ [Philip Lucasz, councillor and director-general of the Dutch Indies], Zeeland almanac.

MacLaren, N. (1991) National Gallery Catalogues: The Dutch School, 1600–1900. Revised and expanded by C. Brown. 2nd edn, 2 vols. London: National Gallery, pp. 343–346 (cat. no. 850).

Ricklefs, M.C. (1993) A History of Modern Indonesia since c. 1300. 2nd edn. Stanford, CA: Stanford University Press

Schrader, S. (ed.) (2018) Rembrandt and the Inspiration of India. Los Angeles: J. Paul Getty Museum

Schütt, F. and Gerken, M. (2025) ‘Micro-XRF Meets Maertgen van Bilderbeecq: Investigating an Early Portrait by Rembrandt’, ArtMatters: International Journal for Technical Art History, Special Issue 2, pp. 33–45.

van de Wetering, E. (ed.) (2015) A Corpus of Rembrandt Paintings, Volume VI: Rembrandt’s Paintings Revisited – A Complete Survey. Dordrecht: Springer, cat. no. 132a.

van Eeghen, I.H. (1956) ‘De portretten van Philips Lucas en Petronella Buys’ [The portraits of Philips Lucas and Petronella Buys], Maandblad Amstelodamum, 43, p. 116.

Yeager-Crasselt, L. (2019, updated 2026) ‘Portrait of Petronella Buys (1605–1670)’, in Wheelock, A.K. Jr and Nogrady, E. (eds) The Leiden Collection Catalogue. New York. Available at: https://www.theleidencollection.com/artwork/portrait-of-petronella-buys-1605-1670/ (Accessed: 12 Junw 2023).

van de Wetering, E. ‘A Painter in his Workshop: Rembrandt Investigates the Usefulness of Linear Perspective’. Academia.edu. Available at: https://www.academia.edu/42833749/A_painter_in_his_workshop_Rembrandt_investigates_the_usefulness_of_linear_perspective (Accessed: 11 June 2023).

RKD – Netherlands Institute for Art History (n.d.) Rembrandt van Rijn, Portrait of Philips Lucasz., 1635. RKDimages, image no. 32160. The Hague: RKD – Netherlands Institute for Art History. Available at: https://rkd.nl/en/explore/images/32160 (Accessed: 11June 2023).

RKD – Netherlands Institute for Art History (n.d.) Rembrandt van Rijn, Portrait of Petronella Buys, 1635. RKDimages, image no. 32161. The Hague: RKD – Netherlands Institute for Art History. Available at: https://rkd.nl/en/explore/images/32161 (Accessed: 11 June 2023).