This study of an old bearded man with a sad, forlorn expression was acquired as a Rembrandt by Peter A. B. Widener from the London art market in 1905. When Wilhelm Valentiner cataloged Widener’s paintings in 1913, he dated it about 1645 and emphasized the painting’s “broad, powerful brushwork and deep thoughtful expression which characterize the artist’s later style.” ^[1] Ensuing assessments, however, have been less enthusiastic. In most subsequent catalogs of Rembrandt’s paintings the picture has been doubted, rejected, or omitted entirely. Martin questioned the attribution as early as 1921, and, though Bredius included the picture in his 1935 catalog, he expressed his doubts in a note: “The picture is known to me only from a photograph, and I am not entirely convinced of its authenticity.” ^[2] Bauch subsequently rejected it, as did Gerson. ^[3] Rosenberg is the only modern Rembrandt scholar to accept it as authentic. ^[4] The National Gallery of Art changed its attribution to “Style of Rembrandt” in 1984.

This painting is one of a large number of rapidly executed oil sketches that Valentiner introduced into Rembrandt’s oeuvre in the early years of the twentieth century. Most of these attributions have now been rejected. Indeed, the painting has only the vaguest resemblance to Rembrandt’s work. The figure type is uncharacteristic for Rembrandt: the anatomy of the head is not understood, and the superficial modeling of the skin and hair is foreign to his style.

A date of execution for the painting is difficult to establish. Dendrochronological examination [see dendrochronology ^[5]] has determined that the tree from which the panel was made was felled in 1666 plus or minus five years. ^[6] The head, however, is painted over another rendering of a head of a man. This figure, visible in the upside-down X-radiographs [see X-radiography ^[7]], is seen in profile and wears a hat ^[fig1]. Because the handling of paint in this figure is quite different from that in the surface image ^[fig2], it seems unlikely that both heads were painted by the same artist. Almost certainly, the head we see today was executed on an old panel after Rembrandt’s death, in emulation, or imitation, of the master’s work.

The support is a vertically grained oak panel composed of a single beveled board fitted with its original strips of wood to square off the beveled edges. Dendrochronology gives the panel a felling date of 1666.[1] A small vertical check in the bottom edge has been repaired. A thin off-white ground is visible through thinly painted passages.[2]

The X-radiographs reveal the presence of another portrait, turned 180 degrees, lying below the present portrait (fig. 1). The earlier portrait is partially visible through the sketchy unfinished beard. Raking light reveals the outlines of the image, a head with a cap, smaller in size than in the current portrait. Cross-sections show no intermediate ground layer between the two paintings.

The paint in the present portrait was applied thinly, with impasted highlights, and hair curls incised with the butt end of a brush. The background was painted first with a reserve left for the head. Paint loss is minimal: inpainting is confined to the edges and abrasion is minor. A moderately discolored varnish is present. No conservation work has been carried out since acquisition.

[1] Dendrochronology was performed by Dr. Josef Bauch, Universität Hamburg (see report dated November 29, 1977, in NGA Conservation department files).

[2] The paint and ground layers were analyzed by the NGA Scientific Research department using cross-sections and X-ray fluorescence spectroscopy (see report dated July 1991 in NGA Conservation department files). Pigments identified in the upper painting were available during the seventeenth century.