Introduction III

Problems of authenticity (about 1850–2009) continued

At the opposite extreme stand the drawings. More poorly documented than the paintings or etchings, and rarely signed, they received almost no individual comment at all in Rembrandt’s own lifetime. Before the advent of Benesch’s catalogue, the standard work by Cornelis Hofstede de Groot, which had no illustrations (although many had been reproduced by Lippmann and Kleinmann), contained only cursory discussions of each drawing. Produced in 1906, it listed 1,600 sheets, some of which are doubted in the author’s cryptic notes. Benesch could also refer to several museum catalogues that had brought the discussion of individual drawings to a more sophisticated level. A landmark in this respect was the posthumously published catalogue of 1920 of the drawings by Rembrandt and his school in the Nationalmuseum in Stockholm, compiled by John Kruse. His entries were particularly thorough and included mention of all the opinions held by the chief investigators of the period, with whom Kruse corresponded. As is stated in the introduction by Carl Neumann, who edited the catalogue, Kruse was influenced above all by Fritz Saxl’s methodology, never published in extended form (parts appeared in his dissertation on Rembrandt of 1912), which sought to apply rigid criteria to attributional questions. Benesch also frequently cites Hind’s 1915 catalogue of the British Museum’s drawings, Elfried Bock and Jakob Rosenberg’s catalogue of the Dutch drawings in Berlin of 1930, Frits Lugt’s of the Louvre’s of 1933, W.R. Valentiner’s two Klassiker der Kunst volumes and to a lesser extent M.D. Henkel’s catalogue of the Rijksmuseum’s drawings published in 1942.

One of the most positive aspects of Benesch’s corpus was that it was so richly illustrated, reproducing almost all the drawings that he accepted as well as some that he rejected. But in other respects, as noted above, it was already somewhat outdated when it first appeared in the 1950s. Rejecting less than 10% of the drawings described by Cornelis Hofstede de Groot in 1906, Benesch relied heavily on the previous work he had done on his thesis on Rembrandt’s early drawings, completed in 1921, as well as on his detailed entry on the artist for the Thieme-Becker dictionary of painters and on his 1935 guide to the corpus, Rembrandt. Werk und Forschung. [2] The six-volume catalogue of the drawings was written largely while Benesch was exiled from Vienna at Harvard University during the Second World War, cut off from almost all the original material. It was compiled at a time when perhaps only one other scholar, Frits Lugt, had a comparable – or even superior – knowledge of the corpus of drawings as a whole. [3] Benesch’s methods were also insecure, for he attributed drawings serially; i.e., he would use any drawing that he accepted as a starting point for further attributions, regardless of the stylistic distance he had travelled from the best documented sheets.

The only way forward from Benesch involves reopening the discussion on the authenticity of each drawing. Every sheet’s documented and critical history needs to be reassembled and a fresh eye cast on its technique and style. The core of the present catalogue therefore lies in the lengthy bibliographical and provenance sections of the individual entries, as well as in the close inspection of the materials the artist employed. The individual bibliographies record the opinions on which the status of each work has depended and against which new ideas must stand. Provenance details also include attributions, in catalogues and inventories. Usually, no more reliable documentation exists and there is no other guide to the degree of credence that should be given to the attribution of each drawing. Any consensus that now prevails is usually built on these earlier, published opinions. The bibliographies also provide a ‘historiography’ of each drawing’s critical fortunes, documenting changes in taste and perceptions.

No sheet is so well documented that its attribution is entirely beyond question. For example, a signature could be misleading, as has been discovered with Rembrandt’s paintings; pupils could have produced drawings that influenced Rembrandt, giving rise to the assumption that they were his own preparatory sketches. Yet few have questioned the attribution, for example, of the drawings that have been indented for transfer to Rembrandt’s copper plates ( Cat. nos 5 and 31). Nor have the drawings with reliable-looking signatures often been doubted ( Cat. nos 7a, 11, 31 and 38). [4]  But little value should be attached to other inscriptions by the artist, such as that on the late drawing of a Child being taught to Walk (Cat. no. 53), as in theory he could have used pupils’ sketches to draught the written notes on the verso.

The earlier literature quoted in the entries also serves to expose the flimsiness of the tradition on which attributions have often been based. A typical example is the Museum’s study of a Nude surrounded by Drapery (Johannes Raven, cat. no. 1). Though doubted twice in the 19th century (by Middleton in 1878 and Seidlitz in 1894), it was until recently generally considered to be a drawing of fundamental importance – the preparatory sketch for Rembrandt’s etching of 1661, The Woman with the Arrow (Bartsch 202, Hind 303; for an impression of the first state, see 1848,0911.102). The model is there similarly posed but seen in reverse and therefore in the same direction as she would have appeared on Rembrandt’s etching plate. Securely attributable drawings from Rembrandt’s last decade are particularly scarce and the temptation to believe in the Nude overruled the evidence presented to the eye. Most authorities, on the basis of the comparison with the etching, deemed it to be so certainly authentic as to render further argument superfluous. Differences in quality, in pose, in the angle from which the model is seen and in the governing mentality of the draughtsman were ignored or brushed aside, and several other similar drawings by the same hand were therefore incorporated into the Rembrandt corpus. [5]

 In 1985 a new, if tentative attribution for the drawing was advanced, to Johannes Raven, an obscure follower of Rembrandt. Raven could have made the drawing as Rembrandt worked on the plate for the etching, such sessions of group study from the model being a feature of Rembrandt’s studio practice. [6] The drawing’s style and technique differ substantially from other figure studies of the same period that can be given to Rembrandt with greater certainty (such as the three surviving preparatory sketches for his painting of the Syndics of the Drapers’ Guild of 1662, Benesch, cat. nos 1178–80); therefore, the rejection of the sheet from Rembrandt’s oeuvre seems unlikely to be reversed.

The reattribution of this one drawing cannot be made in isolation. The entire group of nude studies attributed to Rembrandt by virtue of their analogies with the Museum’s sheet have followed suit, two of them in the British Museum’s collection (here as by Johannes Raven, Cat. nos 2 and 3). The reaction among those involved in the study of Rembrandt’s drawings varies from uncertainty in the new attribution to surprise that the nudes could have masqueraded as autograph Rembrandts for so long. The sense of loss is tempered by the resurrection, however tentative, of the artistic personality of Johannes Raven, whose entire corpus of drawings had been reduced to just two items, a signed sheet in the Staatliche Graphische Sammlung in Munich and a Study of a Youth in Basel – which has Rembrandt’s name written on it. [7]

 Johannes Raven is by no means the only Rembrandt follower whose works have dwindled to insignificant numbers. The drawings of Willem Drost, who was Rembrandt’s pupil in around 1650, are now resurfacing in considerable quantities (although the attributions remain somewhat tentative), largely from the mass of works wrongly attributed to his master. Drost’s name still appears in the early 18th-century catalogue of Jan Pietersz. Zomer’s collection in Amsterdam, but disappears almost entirely thereafter. He is now credited with some celebrated canvases that were given to Rembrandt until the 20thcentury. With misconceptions of this order it is hardly surprising that no reasonably dependable catalogue, uncluttered by school works and imitations, has ever been published. The muddle, like an intractable ball of tangled yarn, remains to be fully disentangled and suffers from a further complication caused by the success of some of Rembrandt’s pupils in imitating their master’s style. The drawings here catalogued under the names of Govert Flinck, Samuel van Hoogstraten, Nicolaes Maes, Willem Drost and Johannes Raven bear witness to this fact, as many of them were accepted as Rembrandt’s work by Otto Benesch.


  • ^ [2] - The Rembrandt. Werk und Forschung was in fact worked up as a supplement to the dictionary entry.
  • ^ [3] - See Haverkamp-Begemann, 1989, especially p. 106, for the background to Benesch’s corpus. Lugt did not himself review the catalogue but his own opinions are known through his notes, now kept at the Rijksbureau voor Kunsthistorische Documentatie in The Hague.
  • ^ [4] - The signature on Cat. no. 10 has been dismissed by some scholars, but is in my view only somewhat rubbed.
  • ^ [5] - Until its reattribution (see below) it was doubted in recent years only by Gerson, 1957, and Sumowski, 1961, 21, the former on the unexpected grounds that it is too close to the etching.
  • ^ [6] - See Cat. no. 71 for this practice. The attribution to Raven was first proposed by Schatborn in Amsterdam, 1985, under no. 69.
  • ^ [7] - See Sumowski, IX, 1985, nos 2141 and 2142x.