With piercing, deep-set eyes, this bearded man leans forward and stares off to his right. He wears a wide, floppy beret and a red-and-yellow patterned robe draped over his shoulders. While this mysterious and intense figure has traditionally been identified as “The Jewish Philosopher,” this designation is undoubtedly fanciful. Nevertheless, the image clearly depicts a concerned individual who seems to be actively brooding over his thoughts.

From 1639 until 1656 Rembrandt lived in a large house on the Jodenbreestraat on the edge of the Jewish quarter in Amsterdam. During those years, and particularly from the late 1640s, he frequently depicted Jewish models in his paintings. As Rosenberg has suggested, Rembrandt probably found in the picturesque faces of the Ashkenazi Jews an intense spirituality that suggested to him the spirit of the people who populated the ancient world. ^[1] At a time when he was searching for a deeper emotional understanding of biblical and historical figures, he found in these care-worn faces an underlying philosophical awareness of human existence. Although a painting such as this was undoubtedly executed from life, it was not considered a portrait in the conventional sense, but rather a tronie, a bust-length figure study that was an imaginative evocation of the model.

This man, with his sad eyes and sharply chiseled features, is seen again in one of Rembrandt’s most memorable figure studies, A Bearded Man in a Cap from 165[7] (National Gallery, London) ^[fig1]. He was also the model Rembrandt used for his 1653 masterpiece, Aristotle Contemplating a Bust of Homer ^[fig2], as well as for The Apostle Paul from about 1657.

Thus, although this work is neither signed nor dated, it must have been created in the mid-1650s, and, perhaps somewhat earlier, because the figure looks slightly younger. In all three of these other paintings, moreover, Rembrandt has given the figure a fuller beard than is apparent in The Philosopher. The differences between The Philosopher and these other works, however, are more profound than those of age and beard size. In the latter paintings Rembrandt has suggested a more thoughtful individual both by emphasizing the wrinkles in his forehead and by throwing the upper portion of his face into shadow. In contrast to Aristotle, in particular, the expression of The Philosopher lacks subtlety and psychological understanding. The differences are in part ones of intent, but they also suggest that the works were created by different artistic personalities.

A close examination of the painting techniques in The Philosopher indicates that this work, though executed with great sensitivity, cannot be by the master. The primary difference between it and comparable works by Rembrandt from the early 1650s is that here the features are more sharply defined and articulated. The eyes are particularly distinctive because of the pink accents along the lower portion of the lid. The nose, likewise, is forcefully modeled, with thick impastos along the bridge and thin translucent paints that reveal the ocher ground beneath in the shadow. Other areas, particularly the beard, are painted with feathery strokes that are unlike Rembrandt’s brushwork. Also unusual in the beard is the way that the edges have been softened with strokes of white from the white shirt beneath it.

A more marked difference in handling from that seen in Rembrandt’s own works is the rather superficial indication of the colored pattern of the man’s robe. The contour of the robe, moreover, is not sensitively conceived. Not only does it not define a logical form, but the nuances of shading that one finds along such a contour in Rembrandt’s paintings are absent. Finally, the hands lack structure.

The awkward appearance of the hands must have bothered an early collector or restorer. From the time that the painting first entered the Rembrandt literature in 1905 until it was treated in 1983–1984, the hands were covered by two layers of overpaint, a gray layer with a dark brown resinous one over it ^[fig3]. Just when the hands had been overpainted could not be determined by technical examination, but it was clearly done at a relatively late date because the overpaint covered old-age crackle and paint losses. Quite possibly the overpaint was applied during the eighteenth century, for in 1772 a larger version of the composition without the hands was auctioned in Paris. ^[2]

Although no trace of the painting from this sale has ever been found, another version of The Philosopher, also without hands but on canvas, appeared on the art market in London at about the same time that the Washington painting appeared in Paris. In 1911 this version passed through the collection of Maurice Kann in Paris, the same collector who had owned The Philosopher in 1905, the year before P. A. B. Widener bought it. In 1914 Kann sold the recently discovered version to the Berlin collector Marcus Kappel, whose collection was cataloged by Wilhelm von Bode. Bode, who had published The Philosopher in his corpus on Rembrandt paintings in 1906, reversed himself in his catalog of the Kappel Collection and argued that the Kappel painting was the original. ^[3] Bode’s assessment of the Kappel version has found little support in the literature. ^[4] The Kappel painting changed hands three more times and is now in the collection of the Haggerty Museum of Art as “attributed to Rembrandt van Rijn” ^[fig4]. The National Gallery of Art painting was accepted as a Rembrandt by all Rembrandt scholars until it was rejected in 1969 by Gerson. ^[5]

The questions concerning the hands are of interest because The Philosopher is painted on two different panels. While most of the image is painted on a walnut panel, joined to it along the bottom edge is an oak strip measuring approximately five centimeters in width on which the hands are painted. Although no difference in execution or in pigments is evident in the treatment of the hands on the main panel and on the strip, the grounds are not identical. In the ground ^[6] of the main panel appear the elements mercury and tin, indicating the presence of vermilion and possibly lead-tin yellow, elements that are missing on the smaller strip. The differences in the grounds suggest that the bottom strip was added after the composition had been planned on a smaller scale, and thus, presumably, without hands. ^[7] Along with this alteration are a number of other design changes. The artist raised the model’s right shoulder slightly and changed the shape of the hat at least three times. Initially he painted it substantially larger, then reduced it to the size of a skullcap, before painting it in its present size.

A number of stylistic similarities exist between this work and paintings by Drost, Willem, who, according to Houbraken, was a pupil of Rembrandt’s. ^[8] Although the dates of his apprenticeship to Rembrandt are not known, a number of signed and dated works from the early 1650s indicate that in these years he was strongly influenced by the master. ^[9] This period corresponds to the time this work was probably executed. One characteristic of Drost’s paintings of male sitters that parallels the pose of the man in The Philosopher is that his figures often stare very intently out of the picture plane. ^[10] Facial features tend to be firmly modeled, although he frequently had problems depicting hands. Not only do many of them lack structure, but the wrists join awkwardly with the foreshortened arms. ^[11] Finally, he favored red and orange colors and patterned robes such as that worn by the subject in The Philosopher. A comparable example is his painting A Young Woman in the Wallace Collection, from about 1654 ^[fig5]. The stylistic connection between his works and The Philosopher seems sufficiently strong to suggest that he may have depicted this striking image. ^[12]

The cradled panel support is composed of two vertically grained boards of wood joined horizontally through the figure’s hands. The join is 5.5 cm from the bottom edge. The main board is walnut, and the lower extension is oak.[1] Edging strips have been added to the top and sides.[2]

A thin white or beige ground layer is present on both upper and lower panel boards, with variations in composition. Density in the X-radiographs indicates the presence of a small amount of white lead in the ground in the main panel but not in the extension. A thin, dark, translucent red layer was laid directly on both panel boards, but there are variations in the pigment composition between the two areas.[3]

The paint was applied thickly in the light passages, with low impasto and loose brushmarking, and more thinly in the dark passages and the background. The imprimatura color was incorporated into the radiating lines on the hat, and into the flesh tones, where mid-tones were created by thinly glazing the red underlayer.

Several changes are visible as pentimenti, in infrared reflectography,[4] and in the X-radiographs. The contour of the proper right shoulder was raised slightly, and the hat was initially larger. The gray background was then drawn over the hat to decrease its size, followed by a repainting of the hat in its present size. The painting was treated in 1983-1984 to remove discolored varnish and overpaint that covered the lower part of the sitter’s arms and his hands.

[1] The characterization of the wood was accomplished by Dr. Peter Klein. (See note dated March 1986 in Conservation department files.)

[2] The exact method used to encase this painting is difficult to determine because of the presence of the cradle. Dating the panel was not possible because of the complex construction. (See report from Dr. Joseph Bauch, University of Hamburg, dated November 29, 1977 in NGA Conservation department files).

[3] The ground and paint in the main panel and the bottom extension were analyzed by the NGA Scientific Research department using cross-sections and X-ray fluorescence spectroscopy (see resports dated August 22, 1984; August 24, 1984; April 24, 1986, in NGA Conservation department files). The analysis indicated the presence of vermilion in the ground on the main panel but not on the extension.

[4] Infrared reflectography was performed using a Hammamatsu c/1000-03 vidicon camera fitted with a lead sulphide tube.