Reconsidering Vermeer’s Perfectionism
What can we learn by examining Vermeer's paintings?
oil on panel
painted surface: 20 x 17.8 cm (7 7/8 x 7 in.)
framed: 39.7 x 37.5 x 5.1 cm (15 5/8 x 14 3/4 x 2 in.)
1942.9.98
This image is in the public domain.
Read our full Open Access policy for images
.
Gazing at the viewer from beneath a striking hat that casts a strong shadow over much of her face, the subject of Girl with a Flute occupies a shallow space defined by a patterned tapestry and a chair with lion-head finials. She rests her left arm on a ledge in the foreground and holds a recorder loosely in that hand. She wears a fur-trimmed jacket known as a jak or manteltje, a garment commonly found in domestic genre paintings by
When Girl with a Flute came to the National Gallery of Art in 1942 as part of a massive donation from
Yet despite its enthusiastic initial reception, the painting’s attribution was soon called into question by scholars, who noted the painting’s many compositional weaknesses.
Similarities in scale, subject matter, and wood panel support have caused some scholars to consider Girl with the Red Hat
Despite its modest appearance, Girl with a Flute has a more complex relationship to Vermeer than Wheelock has proposed. A combination of traditional connoisseurship, understanding of historical painting practice, and the latest results of an array of technical analyses shows that Girl with a Flute was made not by Vermeer, but by someone who, despite understanding that artist’s distinctive working methods, was unable to achieve the same level of finesse: conceivably, an unknown studio associate.
Several factors relate Girl with a Flute to paintings by Vermeer, and specifically to Girl with the Red Hat. The overall conception of the two paintings is undeniably close. Both are tronies: informal, anonymous head studies that offered artists a vehicle for exploring physiognomy, facial expressions, unusual costumes, and striking effects of light and shadow.
Girl with a Flute is indisputably a 17th-century painting. Dendrochronology has determined a felling date for the wood in the early 1650s,
However, at every stage of the painting process, the execution of Girl with a Flute is less assured than that of Girl with the Red Hat or any other painting by Vermeer.
Poor technique in the painting’s underlayers, presumably the mark of an inexperienced artist, caused paint defects that impacted the surface appearance. These include wide drying cracks in dark areas and wrinkles in white areas that had to be scraped down to level the surface before the final paint was applied
The final paint of Girl with a Flute not only reveals an intimate knowledge of Vermeer’s idiosyncratic painting practice, but also betrays the artist’s struggle in mastering his process and technique. In both Girl with a Flute and Girl with the Red Hat, the dull, greenish shadow cast by the hat over most of the face includes green earth, an unusual pigment choice for flesh tones in Dutch paintings. Although Vermeer used green earth for flesh tone shadows at times throughout his career, the practice is virtually unknown among his contemporaries. In both paintings, this greenish shadow is interrupted on one cheek by broad swaths of bright pink highlights. But while Vermeer modulated flesh tones with delicate strokes of his brush, the painter of Girl with a Flute formed the cheek highlights with hard-edged planes of thick, unvaried pink or gray paint and applied the greenish shadows with a similarly heavy hand
Such qualitative disparities are also evident in the painting of the chair’s lion-head finial, where the artist dutifully followed Vermeer’s process for crafting diffuse highlights to suggest an abstracted play of light. Yet the curves and dashes with which Vermeer magically conjured up the carved head of the chair’s finial in Girl with the Red Hat
The many similarities with Vermeer’s Girl with the Red Hat and the evident differences in material knowledge and skill of execution suggest that Girl with a Flute represents a student’s or workshop associate’s earnest response to that vibrant and engaging painting, or another similar, now lost painting of its type. There is, however, no record of a studio in which Vermeer trained or was assisted by another artist or artists, and most scholars have deemed such an arrangement unlikely and unnecessary for an artist who probably executed just one or two paintings per year. Vermeer registered no pupils with the Delft painter’s guild, and the brief notes made by visitors to his studio make no mention of assistants.
In sum, while the theory that Girl with a Flute was begun by Vermeer and completed by another artist cannot be sustained, it simply is not possible to know who did paint the work or under what circumstances it was painted. It is clearly dependent upon, and probably closely postdates, Girl with the Red Hat. With present knowledge, we cannot be sure whether it was created in honest emulation or with the deliberate intent to fool a discerning 17th-century Dutch art market, as it fooled connoisseurs in the early 20th century.
Marjorie E. Wieseman
October 7, 2022
Possibly Pieter Claesz van Ruijven [1624-1674], Delft; possibly by inheritance to his wife, Maria de Knuijt [d. 1681]; possibly by inheritance to her daughter, Magdalena van Ruijven [1655-1682], Delft; possibly by inheritance to her husband, Jacobus Abrahamsz. Dissius [1653-1695], Delft;[1] (sale, Amsterdam, 16 May 1696, probably no. 39 or 40).[2] possibly the van Son family; Jan Mahie van Boxtel en Liempde and his wife, Geertruida van Boxtel en Liempde [née van Son, d. 1876], 's-Hertogenbosch; purchased from the estate by their daughter, Jaqueline Gertrude Marie de Grez [Dowager de Grez, née Mahie van Boxtel en Liempde, d. 1917], Brussels, wife of Jonkheer Jean de Grez [1837-1910]; sold by 1911 to (Antiquar E. Jonas, Paris);[3] sold 16 June 1911 to (Thos. Agnew & Sons, Ltd., London); half-share sold June 1911 to (M. Knoedler & Co., New York, and P. & D. Colnaghi, London); Agnew's half-share sold 21 November 1913 to (M. Knoedler & Co., New York);[4] Knoedler's three-quarter share sold July 1915 to (P. & D. Colnaghi, London). August Janssen [1863-1918], Amsterdam, after August 1916;[5] his estate; sold 1919 with the entire Janssen collection to (Jacques Goudstikker, Amsterdam);[6] purchased jointly April 1921 by (M. Knoedler & Co., New York, and Frederick Muller & Co., Amsterdam);[7] sold February 1923 to Joseph E. Widener; inheritance from Estate of Peter A. B. Widener by gift through power of appointment of Joseph E. Widener, Elkins Park, Pennsylvania, after purchase by funds of the Estate; gift 1942 to NGA.
The support is a single, vertically grained oak panel with beveled edges on the back. Dendrochronology gives a tree felling date in the early 1650s.
Full-bodied paint is applied thinly, forming a rough surface texture in lighter passages. In many areas, particularly in the proper left collar and cuff, a distinctive wrinkling, which disturbs the surface, seems to have been scraped down before the final paint layers were applied. Still-wet paint in the proper right cheek and chin was textured with a fingertip, then glazed translucently. The X-radiograph shows extensive design modifications: the proper left shoulder was lowered and the neck opening moved to the viewer’s left; the collar on this side may have been damaged or scraped down before being reworked in a richer, creamy white. The earring was painted over the second collar. These adjustments preceded the completion of the background tapestry. The proper left sleeve was longer, making the cuff closer to the wrist. Probably at the same time, the fur trim was added to the front of the jacket, covering the lower part of the neck opening. Infrared reflectography at 1.1 to 1.8 microns shows that changes also were made to the shape of the hat and contour of the arm on the figure’s proper right side.
The panel has a slight convex warp, a small check in the top edge at the right, and small gouges, rubs, and splinters on the back from nails and handling. The paint is rather abraded in several areas including the decoration of the hat, the sitter’s proper left arm, and the girl’s necklace. There are a number of small losses and areas of abrasion in the background and there is a large loss in the upper portion of the sitter’s proper right collar. The painting was treated in 1995 to remove discolored varnish and inpainting. It had last been treated in 1933 by Louis de Wild.
Arthur K. Wheelock Jr. based on conservation reports by David Bull, Catherine Metzger, and Melissa Katz
April 24, 2014