Looted art at Christie's?

At Christie's, they sell on 20 May 2015 a "North Italian School, 17th century" porrtrait of a mother and child, estimated at £3,000 to £5,000, unframed, and measuring 231 by 102 cm. UPDATE: sold for £8,750.

The "North Italian School" seems dubious, the painting has all characteristics of a Dutch portrait instead, but that is not the major story with this painting.

When searching more information on this painting, I stumbled upon a website of the ERR project: "Cultural Plunder by the Einsatzstab Reichsleiter Rosenberg: Database of Art Objects at the Jeu de Paume", an inventory of artworks stolen by the Nazis from mainly French, Belgian and Dutch Jewish art collectors and collected at the Musée du Jeu de Paume in Paris before being resold or distributed to German leaders and museums. Much of it has been reclaimed and returned to their rightful owners after the war, much was put into Western museums instead, and a lot was destroyed during the war. But the fate of a significant number of these works is not known. Selling them at auction is not allowed (until it has been first returned to the heirs of the original owners, that is), and major auction houses are supposed to check that the work they sell isn't Nazi loot.

One page on that site showed a painting owned until the war by Edmond de Rothschild, wealthy member of the French Rothschild banking family. Sized 178 by 118 cm, it sports the exact same inscription as the one for sale: 'NUN EX; SILVA / SILVESTRIS' (upper left) and 'AETA.SU. 18 / Ao 1639' (central right) (transcription of the Christie's site) vs. 'Monex Silva Silvestris' and 'AETA. SU. 18 Ao. 1639' (Rothschild). (Whether it is Monex or Nun ex is hard to read)

The basic painting is obviously exactly the same in composition, execution and size (of the sitter). The quality of painting, while hard to compare between an older black-and-white picture and a recent colour one, seems identical as well.

But there are some important differences. A part on the left side is missing, but it is higher at the top and has an added child portrait (with inscription) on the right. Normally, this would seal it as a different version of the same painting, but I'm not sure. It looks to me, when looking at the Christie's version, that there is a distinct horizontal "seam" or paint rupture or colour difference just above the shield at the top (i.e. just above what would have been the top in the looted version: pictured above). This would be consistent with the painting being resized, with a part added at the top. Resizing would also explain the strange leaves at the left side, which are remnants of the larger Rothschildt version. Similarly, the strange "step" at the bottom in the Christie's version serves no purpose and can easily be an addition.

The paintings are 178 cm vs. 231 cm high, which means that some 53 cm should have been added to the Christie's one for my hypothesis to make any sense. This seems very roughly to correspond with the part above the shield and the step at the bottom combined. I have tried to show what I mean above, by removing these two parts from the Christie's painting.

Which leaves us with the child. Can it be either a new addition, or something that was there in the past, was hidden beneath overpaint and has been re-revealed between 1940 and now? The child certainly has one ugly face, but that's hardly conclusive evidence of post-1940 tampering: it may be that a new version of the work was painted after a child was born, but it seems unlikely (but is it more unlikely than someone altering the picture in this way?). The style of the childs clothing and posture generally corresponds with other portraits of children from that period, so this doesn't rule out that it is a genuine portrait of the time. The inscription, which is curiously omitted from the description on the Christie's website, reads "AETAT SUAE II MENSIUM 1640", or "aged 2 months (sic!), 1640". 

Both the child and the inscription are clearly by a different hand than the original portrait, which would seem to indicate that two identical versions of the painting were made in 1639, but then one was changed to include the child portrait in 1640. This seems weird, but weirder things have happened in the history of art. But to me the child looks more like a Van Meegeren-like addition than a real 1640 painting. That face...

So, why would anyone alter this painting this drastically? Cutting down on the left, adding parts on the top and bottom, and adding a child? No matter if it is the same painting or a copy, the first three questions remain the same (the current dimensions are clearly less harmonious, but perhaps they were needed for a specific location). Only the added child makes sense in a 1640 painting, and perhaps less so in a 1940+ change. Unless someone wanted to alter the appearance of the painting after the war, to reduce the chance of being recognised from the description.

The painting for sale lists no provenance, a dubious description (North Italian?), and no indication of the connection to the looted Rothschildt painting (which, even if it is not the same work, is undeniable). The lack of a frame is also a bit strange for such a large and otherwise well-preserved work. But of course after resizing it wouldn't fit in the original frame any longer... Asking Christie's to provide provenance information and a picture of the reverse may shed some light here. But the contact information provided by Christie's at the auction page sends an out-of-office that she is away until Monday 11th April (sic!). So I guess I'll not get the necessary information before posting this blog.