RKD STUDIES

The composition of The rommel-pot player is known in about 30 painted variants, as well as partial painted or drawn copies, and at least four different engravings.¹ Until today, Hofstede de Groot's 1910 assessment still remains valid: ‘A recognized original of this composition cannot now be traced. There are a number of replicas which at best date from the time of Frans Hals, while some perhaps come from his studio’.² Largely, two compositional types can be distinguished: the most frequently repeated version comprises five children [244] [245][246], and the other version features six [247]. In the former, the girl on the left looks towards the viewer, whereas in the latter, she gazes up to the rommel-pot player and is positioned on the right hand side of the boy seen in profile. In the first variant, the boy standing on the right reaches out with his hand, – perhaps trying to pick the musician’s pocket – and in the second variant he appears to have already committed his crime, seemingly hiding something in his hands. Across all known versions the quality of execution varies, with the depiction of the protagonist being most accurately maintained [248] [249] [250] [251] [252][253].

Repeated attempts have been made to declare one or the other version to be the original, or at least single one out as a reliable replica of the lost prototype by Hals himself. Most recently, Slive presented the best preserved picture – the version in the Kimbell Art Museum in Fort Worth – as the original [244].³ However, the painting has noticeable weaknesses in many areas, as becomes evident when studying the detailed photograph of the children's heads. When looking at the boy in the lower left corner, we observe dotted and swept light and dark accents that have been applied with a soft brush [254]. These differ from Hals’s brushwork, typically applied with a flat brush, blending the colors in the flow of a brushstroke. This also applies to the two heads in the door opening on the far left, whose brushwork does not display Hals’s typical soft color-transitions, but rather shows thick dabs and swipes of the brush. This could be the same assistant’s hand as the one we can distinguish in the heads of the protagonists in Merrymakers at Shrovetide (A3.1) and Young man and woman in an inn (A3.3). Nonetheless, a higher level of quality can be observed in the head of the painting’s central figure, the elderly rommel-pot player. But even there the execution does not show the brilliant autograph brushstroke which can be found for example in the facial features of The smoker (A3.2) or the Lute player (A1.15). The still reasonably good painterly quality of the picture in Fort Worth is on a par with that in the similarly sized version kept in a private collection [245].

Remarkably, the production of variants of the Rommel-pot player probably went on for years, or even decades, with later versions also being created outside Hals’s workshop. If we exclude the possibility that this elaborate production was based on a readily available large-scale original held in stock at the workshop, there must have been different methods of creation in use over a long period of time. Accordingly, we can assume the existence of a compositional template made by Hals, either drawn or sketched in color as a form of ricordo, for which he had made individual motif sketches. On the basis of these models, new variants could be steadily produced and assigned to different assistants. Interestingly, the figure of the rommel-pot player is preserved in a small copy that was inserted into a still life by David Bailly (1584-1657) from c. 1625 (B3) [255]. Hals’s preparatory study – which must have certainly existed – can be imagined to have been a colored sketch on paper, similar to the modelli for engravings. It would have remained in the workshop together with the other studies and compositional sketches, while the finished pictures were sold and went out of reach.⁴ In 1792, another drawing by Bailly was offered at auction in Utrecht, dated 1624 and described as ‘[...] Boontje [...] in those days a well-known fool in Haarlem, playing the Rommel-pot’.⁵ If these pieces of the puzzle actually fit together, not only a likely date is set for the creation of these motifs, but also the content of the depiction can be identified more precisely. Boontje is marked as a fool by the fox's tail. The provoking noise of the rommel-pot – created by moving a stick that has been poked through a pig’s bladder which has been stretched over a pot or jug – is reminiscent of the grunting of pigs. The horrible sound is beloved only by children. As the inscription on Jan van de Velde’s (1593-1641) engraving (C5) – which renders Hals’s figure in a slight variation – describes: ‘Many fools run around at Shrovetide / To make a half-penny grunt on a Rommel-pot’.⁶ The subject would then be the foolishness of the world, complemented by the theme of robbing the fool who collects pennies from children. The grinning man that was captured in such a pointed way is unlikely to have been created by Hals on the basis of some random model sketches. An exact likeness of a living troublemaker is more probable, captured by Hals in the same way as Piero (A1.3), Verdonck (A1.34) and Malle Babbe (A1.103).

A comparison of the repetitions and variants permits an insight into the range and duration of the workshop production and the scope of the painterly quality. In addition, the frequent re-use of the expressive motif of the rommel-pot player alone also demonstrates how the Hals workshop adjusted to the buyers’ wishes [256]. A large part of these paintings would have been created in the workshop, where the templates were kept. But many other ones would have been made later, and thus elsewhere, as copies after already existing copies and variants. A definition of what was produced under the oversight of Frans Hals would require technical inspections of as many supports, paint layers, and production techniques as possible. An important summary of such investigations was published by Karin Groen and Ella Hendriks; Slive noted the findings of Claire Barry when she restored the painting in the Kimbell Art Museum in 1988; and in 1994, Ella Hendriks carried out a comparative examination of the painting technique of one of the versions in private ownership [245].⁷

With regard to composition and format, the versions in Fort Worth [244] and private ownership [245] mostly correspond. In the representation of the main figure, they are almost identical; the five children’s heads are different, but executed in a similar sketchy manner. Notably, in the privately owned painting, the left and right hand edges are wider, preserving slightly more of the figures. In the Fort Worth version, the paint layers are better preserved; whereas in the other one, much of the surface is abraded. Photographs taken during cleaning even show the white ground shining through. The technical analysis of this painting conforms to that of the one in Fort Worth.⁸ In her examination conducted in May 1994, Ella Hendriks found an underdrawing in the privately owned picture, which is not typical for Hals’s fleetingly sketched autograph works, but characterizes a copy. This underdrawing fixates the figure of the rommel-pot player in detail, suggesting that a model was precisely transferred onto the canvas. Accordingly, potential forms of underdrawing in the other versions would need to be examined, together with the extent to which they differ from one another. With regard to painterly quality, the two faces of the rommel-pot players are alike [248] [249]. As close as they are to Hals in the observation of the motif, they do not show his autograph brushwork. Therefore, both paintings must be considered to be workshop productions, based on the same templates. In addition, in both versions the faces and hands of the secondary figures are clearly painted in a more fleeting manner than the main protagonist, and there are differences among them in the arrangement and quality of execution. This could mean that there was a binding, very detailed model available for the main figure, while the remainder of the scene was only based on less precise preparatory sketches, leaving more room for an individual manner of execution.

248
Detail of fig. 244
workshop of Frans Hals (I)
Rommel-pot player, c. 1622-1624
Fort Worth, Kimbell Art Museum

249
Detail of fig. 245
workshop of Frans Hals (I)
Rommel-pot player, c. 1622-1624
Munich, private collection

250
Detail of cat.no. A4.2.1j
after Frans Hals (I)
Rommel-pot player, 17th or 18th century
private collection

251
Detail of cat.no. A4.2.1f
workshop of Frans Hals (I)
Rommel-pot player, c. 1622-1624
sale Amsterdam (Christie’s), 7 May 1997, lot 68

252
Detail of fig. 246
workshop of Frans Hals (I), possibly Judith Leyster
Rommel-pot player, c. 1626-1630
Chicago, The Art Institute of Chicago

253
Detail of cat.no. A4.2.1i
after Frans Hals (I)
Rommel-pot player, 17th or 18th century
private collection

254
Detail of fig. 244
workshop of Frans Hals (I)
Rommel-pot player, c. 1622-1624
Fort Worth, Kimbell Art Museum