For sale - 'the earliest Vermeer'
June 10 2014
It's looking like an interesting week for Old Master attributions; first a Rembrandt self-portrait confirmed in the UK, and now news of Vermeer's 'earliest known painting', a 1655 St Praxedis (above) signed 'Meer', has been announced. The picture has been researched by Christie's and the Rijksmuseum, and will be sold this summer in London with an estimate of £6m-£8m (which seems a little on the low side, but we'll see). More in The Guardian, the NL Times, and USA Today, which reports:
The work was tentatively attributed to Vermeer after it appeared in an exhibition at New York's Metropolitan Museum in 1969, and the authorship was reinforced in 1986, when leading Vermeer scholar Arthur Wheelock argued it was authentic.
But other experts remained skeptical. The painting was not included in a Young Vermeer exhibition in The Hague in 2010, but was displayed in a 2012 show of the artist's work in Rome.
Christie's said Friday it was declaring the work a Vermeer after scientists at Amsterdam's Rijksmuseum and Free University carried out isotope analysis on its lead white — a coveted but toxic type of paint made with lead and vinegar.
"They're able to basically DNA-test lead white," said Henry Pettifer, Christie's head of Old Master paintings.
The tests found that the lead white was a precise match for that used in another early Vermeer, Diana and her Companions — "So precise as to suggest that the same batch of paint could have been used," Pettifer said.
He said the research, along with analysis of the date and signature on the painting, amounted to "a compelling endorsement" of Vermeer's authorship.
Update II - a reader writes:
I’ve known the picture for some time, it was (I believe) first proposed by Stanley Spencer, a dealer of OMP in NYC, and has been the subject of considerable debate. Less well known is the work by the Florentine Ficherelli on which the painting can be said to owe no small debt
This painting it has been argued never left Italy, and it has been proposed that Lead White pigment travels much more easily than either artists or paintings, and that possibly the paint itself was made in batches and sold to various artists. (At best this so-called Vermeer can only be described as juvenilia).
But your last few posts beg an intriguing point. Somewhere on the spectrum is a gifted Scholar with a “golden eye”, who through connoisseurship honed by decades of study, can identify an autograph work by a master others have overlooked. But on one extreme is a museum curator convinced that Western White, Male hegemony has corrupted art history and the art “experience”, and shows work without identification, context, or labels. On another extreme, a technologist is advancing an attribution offering laboratory analysis for evidence, perhaps even in the face of less than convincing visual similarities. I guess I want to ask, “can the center hold”?
All excellent points. As regular readers will know (and as I discussed in my paper at the Mellon Centre conference) I am not entirely convinced that technical analysis is the silver bullet many art historians think it is, when it comes to attribution. We can only confidently assert that the lead white used in this picture is proof of Vermeer's authorship once we have tested hundreds and hundreds of lead whites in pictures by his contemporaries, to build up a valid database. At the moment, paint analysis results have come overwhelmingly through testing the works of great artists, since those are the ones that people want to investigate, both in the art trade and the museum world. So it is often not surprising that the paint analysis database says a certain paint was used in a number of works by artist X - when by and large only works by artist X have been tested, and not those by his inferior followers, associates and students.
Update III - a painter writes:
I am not at all convinced by tests on lead white paint being used to support an attribution to any particular artist.
I am a painter using lead white (there is no modern substitute for lead for impasto work) from a large tube (circa 1927) inherited from my painter grandmother who may well have used the same colour man as her teacher Sickert for buying her paints.
If this was the case, then using the same tests applied to the Vermeers it should be possible to prove that my one Sickert, all my own paintings and my grandmother's paintings are by the same painter on the basis that the lead white used was from the same 'batch' of paint with the same 'DNA'.
One also thinks of the attribution difficulties caused by Rembrandt and his pupils sharing the same models, batches of paper and boards and presumably batches of paint including lead white. Apprentices learned to mix paint as an important and responsible part of their basic training. Consistency between batches, resulting from following exact formulae would have been a basic requirement, especially considering the cost of some of the ingredients. Different studios and individual artists would often follow the same recipes. ( See Daniel V. Thomson 'The materials and techniques of medieval painting'. Dover 1956).
Clever fakers can also easily buy C.19 boxes of watercolours and antique paper which would pass all scientific tests for say a mid nineteenth century water colour by any number of different artists. This wouldn't help the faker to make a convincing Cotman which would pass any connoisseurship test.
Scientific tests on the constituents of any paint can only prove that that that paint is consistent with paints of a particular, quite broad period. They cannot indicate when a painting was made, without other types of additional evidence, or by whom the paint was used. So the correct attribution of the 'earliest Vermeer' will always be a matter for debate and opinion rather than scientific certainty.
The Sickert tube sounds fascinating.