Art authentication
And then there were none
Hugh Sebag-Montefiore investigates the case of the disappearing Rembrandts
Last week the National Gallery let the public in on a well-kept secret with the opening of its Rembrandt exhibition called Art in the Making, the first of a series sponsored by Esso. The exhibition, which runs until 17 January, is designed to show how Rembrandt painted and illustrates the tests applied to the paintings by the Gal- lery. One reason for these tests is to establish the paintings' authenticity.
It is a courageous endeavour. Only last year, the National Gallery's curator Christ- opher Brown revealed that an influential group of art historians, the Rembrandt Research Project of Amsterdam, thought that eight paintings in the exhibition might not be by Rembrandt. It is a claim which makes a mockery of the desire to explain Rembrandt's painting 'system' to the lay- man. How can you identify an artist's system from a group of paintings when you are not sure that all the works being examined are by the same man?
In 1982 the RRP published its condem- nation of two National Gallery paintings, `A Scholar in a Lofty Room' and `Tobit and Anna', in Volume 1 of its work, A Corpus of Rembrandt Paintings. But the paintings were still exhibited as Rem- brandts. It was only at the end of last year that Christopher Brown announced that they were not by Rembrandt. He also deattributed another painting, 'An Old Man in an Armchair', to school-of- Rembrandt status. These three paintings will not be on show at the National Gallery exhibition. Brown has given his opinions reluctantly. It is not unusual for curators to hold their counsel until the publication of the updated catalogue. Brown's new cata- logue for Dutch Masters will not be pub-
lished until next year. 'I wanted to wait until then so that I could give my detailed arguments for deattributing the paintings,' he told me. This practice of 'concealing' doubts concerning the authorship of a painting may look like a cover-up to the layman; he has reason to feel a little miffed when he discovers that the Rembrandt which he has been worshipping turns out to be by a relatively unknown pupil. Given the fact that it is quite common for scholars to hold different views on a painting, it is surprising to find that the RRP's views have generated an air of controversy throughout the art world. So what makes them so special? The RRP is a team of art historians funded by the Netherlands Organisation for the Advancement of Pure Research to estab- lish the authenticity of Rembrandt pictures around the world. They came together in Amsterdam in 1968 with a mission. They wanted to examine the 630 Rembrandts identified by Bredius, a previous scholar, and to use the most up-to-date scientific techniques to determine once and for all what was the real thing. So far they have published their views on the 195 pictures painted by Rembrandt before 1634 when Rembrandt was 26; according to them, 82 are not by Rembrandt.
There is every reason to respect their opinions. Unlike most art historians they have devoted their lives to the study of one artist. As a result, they have probably examined more of his paintings than their less specialised colleagues. Also, unlike previous studies of the Dutch master, theirs makes use of an impressive array of scientific tests (in many cases carried out by galleries as part of their own research programmes) in addition to the traditional means of attribution which rely solely on connoisseurship. Their work includes the use of neutron-activation auto-radiography and the examination of infra-red photo- graphs and X-rays of the paintings as well as the analysis of microscopically enlarged samples of paint. They have even counted the threads in the picture's canvases to establish which come from the same bolt.
The extra dimension given to their inves- tigation by the sophisticated scientific aids at their disposal has, however, been dis- appointing. For instance, they have not been able to find any use of anachronistic paint, that is colours which were only invented after Rembrandt's death. If they had, they could have been sure that the painting in question was a modern fake. As far as they can tell, all the works studied were painted in the 17th century, which shows the impotence of scientific analysis. Science cannot differentiate the master painter's work from that of his pupils when all concerned used the same materials. All this means that the RRP has had to fall back on connoisseurship just like everyone else.
`The problem is that the RRP has drawn the parameters of Rembrandt's oeuvre too tightly,' says Brown in his wistful voice. He is a pleasantly rounded epicurean with the languid air of a Roman emperor presiding over his shrinking empire. When he is not feeling persecuted by the RRP, his voice booms out resonantly like that of an off-duty opera singer. For him, as for many of the other experts around the world, the RRP's philosophy is anathema. The RRP are basically saying that if a painting is not brilliant, it cannot be by Rembrandt. Brown counters: 'Many of these paintings are weak, but I see nothing to suggest that they are not by Rembrandt in a less inspired moment. They were all executed at a time when he was running a very busy portrait practice. This explains why less care is exhibited.'
So far, many of the skirmishes concern- ing the National Gallery paintings have dealt with part of the pictures only. Thus, the RRP has doubted the horse in the 'Portrait of Frederick Riehel on Horse- back': in their opinion, Rembrandt would never have painted an animal which looks like a rocking horse. Also the use of light and shadow in this picture is unusually harsh for Rembrandt. Yet Ashok Roy, the Scientific Officer at the National Gallery, has pointed out that the paint used for the background of the portrait has changed colour since it was painted, which makes any judgment based on contrasting colours difficult to believe.
The ruff in the 'Portrait of Philip Lucasz' is demoted because it is not executed as carefully as the ruffs in other Rembrandt paintings. Here a more sketchy, impress- ionistic technique is used rather than the 'A Portrait of Hendrickje Stoffels': does the X-ray help to prove it is by Rembrandt? exact positioning of the white lace lobes over the black coat. The latter technique has been identified by the RRP as being typically Rembrandtesque. Brown explains away the difference in technique by sug- gesting that this painting had to be com- pleted in a hurry; Lucasz had to go abroad in the year when the picture was painted. This necessitated a 'quick' portrait which did not give Rembrandt time to apply his usual system to the collar.
More is at stake in the case of 'A Portrait of Hendrickje Stoffels'. Van der Wetering of the RRP still has not made up his mind whether the painting is by Rembrandt: 'The eye wanders around far less on paintings by Rembrandt than on paintings by his pupils as a result of the use of light; Rembrandt uses a hierarchy of lights and darks so that you can easily concentrate on the focal point of the picture. This is missing in Hendrickje Stoffels. Secondly the fur coat does not hang in an under- standable way; there is no bulge where the arm goes under the coat. I also have a problem with the eye socket. The X-ray shows that a continuous oval was left for the socket. Normally four highlights split it up.'
Brown has an explanation for the abnor- mal features in the painting. 'He was using a model close to home, his wife. So he felt free to try out a portrait composition. He did not feel constrained to finish it in the same way as a commission.' David Born- ford, the restorer at the National Gallery, has identified the typical Rembrandtesque structure in the X-ray of Hendrikje Stof- fels. He applies connoisseurship to the lead white brushstrokes used by the painter to sketch in the painting's basic structure. These brushstrokes are invisible on the surface of the painting but they are visible in the X-ray. 'Each brushstroke is serving a purpose in relation to what is seen on the surface,' he told me. 'Underneath the swirling strokes on the left there are hints of the same coherent structure.' Bomford thinks that the left arm was originally in a different position. The swirling strokes cover up the original design. He uses this piece of evidence to conclude that the picture is not a copy; if it were the artist would have been unlikely to make such drastic changes.
In the course of preparing the National Gallery Exhibition, Brown, Roy and Bom- ford have had doubts about only two of the paintings, the smaller 'Portrait of Mar- garetha de Geer' and the 'Franciscan Friar'. In view of this uncertainty, it is interesting to recall the entry that was put beside the 'Franciscan Friar' in the last catalogue of Dutch pictures: 'Despite the serious damage it has suffered, there is enough of the picture in good or reliable condition to put Rembrandt's authorship beyond doubt.' Which all goes to show that when an art historian says, 'It's definitely a Rembrandt', he really means, 'That's my opinion but I may well be incorrect'.
Previous page
