Why connoisseurship matters (ctd.)
March 14 2013
Pictures: The Bowes Museum/BG/Your Paintings
Thanks for all your emails and comments about The Culture Show programme. It was fun to make, and I'm always glad to have a chance to evangelise about two of my favourite subjects; Van Dyck and connoisseurship. I promised a more detailed note about the picture, so here goes. I’m afraid it’ll be a little rushed, so don’t expect a Burlington type write-up.
Condition
I'll start with condition. At first sight, the picture looked a bit of a mess, and it was easy to see why it had been passed over as a copy for many years. One of the most disfiguring aspects of the portrait was the sitter's left eye, which did not seem to point in the right direction. With a portrait, small damages in a face can make the viewer question the whole image. We tend to look at portraits almost as human faces - and if the eyes are wonky, we assume that the whole portrait must be, in effect, also wonky.
However, as is often the case with condition issues, things looked worse than they in fact were. The wonky eye in question, which at first I thought had been over-painted, was merely missing a dark glaze over the pupil, and a tiny white highlight. Both of these had been cleaned off in a previous campaign of over-zealous restoration. Delicate glazes and pigments like those in an eye on a portrait can be easy to accidentally remove. Possibly, this was done centuries ago, for cleaning pictures used to be the job of the house keeper. Sliced potatoes, stale urine, and worse were used to wipe down paintings, sometimes with disastrous consequences. Wipe too vigorously, off comes a highlight, and suddenly an eye loses its direction.
Elsewhere in the picture, it was the usual story of old layers of dirt and varnish making the paint strokes and colours unreadable. There were also a few holes and some areas of abrasion. Although the picture had been over-cleaned it had not (and this is most unusual) been 'restored'. That is, there were no layers of old over-paint covering the losses and holes. Often, areas of old over-paint can be very hard to remove, especially if applied in oil. Even more fortunately, the picture was unlined, which meant that the original surface of the canvas was actually in excellent condition. I don't recall dealing with an un-lined Van Dyck before. Consequently, the paint layers had not been pressed or flattened in an old lining process (they used to use hot irons to bond the two canvases together, therefore melting and flattening the paint), and all the impasto was just as the artist had intended it. The picture had a fine texture, especially in the drapery. So despite appearances, the painting was in relatively good condition.
There was, however, one area where it had been dramatically altered by a later intervention, and this was in the curious grey, oval additions at the top and bottom. I've not seen these on a Van Dyck before, and again they must have been another reason to doubt the painting in the past. It was fairly easy to see that the edges were additions, especially at the bottom of the picture, as the remains of the sitter's sleeves were visible beneath the later paint. Our paint analysis also confirmed that, at the top, the grey background extended underneath the oval, and so we could safely rule out any question of the oval being original to the picture. In the past, it was not uncommon for owners to add ovals like this if a portrait was intended to be hung as part of a decorative set, perhaps in an architectural feature. When we cleaned the picture, it was decided to leave the oval additions on. It might have been possible to remove them, but they formed part of the picture's history. Below you can see a not very good effort by me at removing them on Photoshop, to give you an idea of how the portrait would originally have looked.
Cleaning the picture was a delicate but enjoyable experience. I was lucky that the Bowes Museum entrusted the picture to our care, and we were able, with Simon Gillespie's help, to use all our experience of conserving Van Dycks (over 20 so far, and many more studio works) to full advantage. Unless you really know what you're dealing with, cleaning Van Dycks can be a fraught business, given the extremely complex and delicate glazes he used. It is very easy to get things wrong, especially in areas with darker pigments like the hair. (If I may say so, the case demonstrates how sometimes the art trade and commercial restorers can have a greater understanding of how to conserve a painting than the museum world. Because we're portrait specialists here at Philip Mould & Company, with a particular expertise in Van Dyck, we have dealt with, researched and restored more Van Dycks in the last few years than a museum conservator might do in a lifetime.) Simon and I decided that the best approach would be to intervene minimally, and so where possible we have left on a layer of the oldest, possibly original, varnish over the whole picture. After the cleaning, there was some re-touching required, for example in areas of abrasion in the drapery, and most notably in the sitter's left eye, where a highlight was replaced. Fortunately, we had a useful guide for any re-touching with a good quality studio copy of the picture at Lamport Hall (below).
Sitter
Although the picture was of an unidentified sitter when John Bowes bought it in 1866, curators at the Bowes museum had more recently suggested Olivia Porter (d.1663) as an identification, by comparing it to other portraits of her by Van Dyck. And they were right. The copy at Lamport Hall had originally been acquired in the late 17th Century as an unknown sitter, but was subsequently identified as Dorothy, Countess of Leicester. However, some further research, including a trip to the Heinz Archive at the National Portrait Gallery and some help from Olivia Porter's descendants, allowed us to prove conclusively that the Leicester identification was wrong, and that the sitter was indeed Olivia (or Olive, as she called herself). Olive was the wife of Van Dyck's closest friend in England, Endymion Porter, one of Charles I's key courtiers. Porter was the only person whom Van Dyck painted himself with [below, Museo Prado]. Olive was a lady-in-waiting to Henrietta Maria, and later, in 1637, converted to Catholicism with such zeal that she was eventually ordered to leave the country by Parliament.
The Bowes picture was of such high quality that I think we can safely say it was done from life. The portrait was not only exceptionally well painted (as Professor Christopher Brown said, ‘this is Van Dyck at his best’), but carried real authority in terms of characterisation and overall human presence. It's dangeours to be subjective about these things, but it feels as if it was someone Van Dyck knew intimately, and liked. The sketchy and unfinished nature of the drapery further suggests that the picture was conceived as a portrait from life, probably done with the intention of being able to use the likeness in the other portraits of Olive that Van Dyck was to paint. The same head, with a slightly different direction of gaze, was used again by Van Dyck in a larger three quarter length portrait now at Syon House (below, Duke of Northumberland collection), a picture which has been in the Northumberland collection since at least 1652.
The late Sir Oliver Millar, author of the section of the 2004 Van Dyck catalogue raisonne devoted to the artist's English works, dated the Syon House picture to c.1637, which I would agree with, and which also seems a most likely date for the Bowes Museum picture. Given Van Dyck's own strong Catholic faith, it is interesting to speculate whether the portraits of Olive done at this time were in any way linked to her conversion. Van Dyck also painted a group portrait of her with her husband and child [Private Collection - a copy (perhaps that recorded as being made by Mary Beale in 1672) is at Dunham Massey], but this is more difficult to date. A less securely identified portrait of Olive by Van Dyck was formerly at Shrubland Park. Two smaller copies of the Bowes picture exist, on panel. One was formerly at Balnagowan Castle, and was later sold at Christie’s as a portrait of Henrietta Maria, and the other remains in the private collection of Olive's descendants.
After filming was over, I was subsequently alerted (again by Olive's descendant) of another important likeness of her at Lacock Abbey. This portrait, above, is an early copy probably by Theodore Roussel (1614-1689) after the head of Olive in Van Dyck's group portrait of her with her family. The Lacock Abbey copy is important because Van Dyck’s original group portrait is in bad condition (even George Vertue in 1751 records this fact), and consequently the likenesses are not reliable as the picture has been substantially over-painted*. So the Lacock Abbey copy, done soon after the original was completed, is another useful guide to what Olive looked like. For more information on Olive's life and the history of some of her portraits, the best source is Gervas Huxley's ‘Endymion Porter: the Life of a Courtier’ (London, 1959).
Provenance
John Bowes bought the portrait of Olive in Paris in 1866, from one of his regular dealers, Madame Lapautre. A receipt records that he bought it with another portrait then attributed to Van Dyck, of Henrietta Maria. The Henrietta Maria picture is also still at the Bowes Museum, but has sadly been very heavily over-cleaned, and badly restored (many years ago). It is hard to tell the quality due to the paint loss, but I would say that it was probably painted in Van Dyck’s studio. The earlier history of Olive’s portrait was unknown, but I found the remains of a wax collector’s seal (below) on the back of the un-lined canvas.
It's hard to make out from the photo, but what you can see is a coronet, the top of a shield with 'mascles' or lozenges, and part of the chain of the order of the Holy Spirit, France's highest order of chivalry (as denoted by the tiny ‘H’ in the chain). All of these combined meant that I was looking for a titled (the Coronet) member of the Rohan family (a coat of arms with nine mascles, since the shield was undivided) who was a member of the order of the St Esprit. With help from Dr. Clive Cheeseman, Richmond Herald at the College of Arms, and Hervé, Baron Pinoteau of the Académie Internationale d'Héraldique, we were able to establish that the arms belonged to either Henri, 2nd Duc de Montbazon (d.1654) (below, with his arms in the engraving), or his son Louis (d.1667).
Hopefully, further research in any Montbazon archives might yield further clues, but it was decided not to do this in the programme. What the wax seal does tell us, however, is that the picture was in France by the middle of the 17th Century. It is likely, therefore, that the Porters took the painting with them when they fled England into exile after Charles I lost the Civil War, and probably sold it soon afterwards. We know that the Porters were in dire financial straits when in exile. The supposition is that the picture remained in France until John Bowes bought it in Paris in 1866.
You can see a larger image of the cleaned painting here on the Your Paintings website. In addition to Professor Christopher Brown, the attribution to Van Dyck is also supported by the Rev. Dr. Susan Barnes, a renowned Van Dyck scholar who was one of the original authors of 'Van Dyck: A Complete Catalogue of the Paintings' (New Haven and London, 2004)
* Incidentally, if you own the group picture, and would like some advice on possibly restoring it...
