In today’s blog, Dr Caroline Rae, Lecturer in Technical Art History, University of Glasgow (former Caroline Villers Research Fellow, Courtauld Institute of Art), talks about her exciting discovery of a lost portrait of Mary, Queen of Scots
In 2016-7, I was honoured to be awarded a Caroline Villers Research Fellowship at The Courtauld Institute of Art (CIA) to carry out a research project in collaboration with The National Galleries of Scotland (NGS), and The National Trust (NT). The main aim of this research project was to revisit some of the extant scholarship in the field of Jacobean Scottish portraiture utilising the interdisciplinary methodology of technical art history. As a technical art historian/ painting conservator and former research team member of the Making Art in Tudor Britain project, I knew the potential of the methodology to revise scholarship and add to our collective knowledge in the field. However, I wasn’t expecting to make a significant historical discovery which would soon circulate around the globe!
As a focus for the project, and after discussions with colleagues, I decided to centre on two significant Netherlandish émigré artists who lived and worked in Scotland in the late sixteenth and early seventeenth centuries, Adrian Vanson, one of James VI’s Crown painters, and Adam de Colone. I examined works attributed to these artists from the NGS collections in the painting conservation studio of NGS, who also generously contributed X-radiography, photography and dendrochronology to the project.
A portrait of Sir John Maitland, from the collection of the National Trust at Ham House (left) came in to the Courtauld Institute for examination in mid-December 2016. This work had been recently attributed to Adrian Vanson as a result of research into a series of portraits of this sitter, by David Taylor. I examined the portrait using Infra-red reflectography, paint sampling, microscopy and X-radiography. I processed samples from portraits from both collections at CIA and undertook inorganic analysis of components of paint and ground layers (SEM-EDX) in the Physics Department, Kings College.
At the Courtauld at this time, physical X-ray plates were developed in a dark room, and so I first saw a glimpse of the face of the underlying woman on a X-ray test plate under a dim red light, when expecting to see the face of Maitland. My first thought was that I had developed the wrong plate. However, I soon recognised the queen’s iconic face and cap when I saw the image in full light, which was a very exciting moment! After consulting with my NT and NGS collaborative colleagues, it was agreed that this identification indeed seemed likely. Although there were no inscriptions, the striking similarities in the features and also the pose and costume were agreed to be sufficient to confirm tentatively the sitter’s identity. Although dendrochronology has not yet been carried out on the panel support, which could provide further verification of the date of construction, the materials and techniques used to create the portrait and overlying work, and also many pentimenti present in the overlying work, which inspired several painted and engraved copies, all strongly point to the inscribed date on the overlying work being genuine.
At this time, I also began working with Kate Anderson, Lesley Stevenson and other NGS colleagues on an exhibition of the overall findings from the fellowship into works by Vanson and De Colone (“Art and Analysis”) which was displayed at the Scottish National Portrait Gallery for two and a half years and which was a great opportunity. The portrait of Maitland was kindly loaned by NT for the exhibition. A further notable finding was that I was able to argue in the exhibition that both the underlying portrait of a woman and the overlying portrait of Maitland could be attributed to Vanson, on the grounds of similarities in panel construction and preparation, and characteristic painting techniques to another work from the NGS collection, which can be strongly associated with this artist, thus confirming the earlier attribution by Taylor.[i] Vanson can only be documented to have been present in Scotland from 1581 providing a potential date range for the execution of the Maitland portrait, if the attribution is indeed correct, of 1581 – 1589.
The exhibition at SNPG included a mosaiced transparency of the X-radiograph next to the NT portrait, displayed on a light box. It’s no exaggeration to say that the new discovery caused a media sensation as it was widely reported both nationally and internationally. However, despite several TV and radio interviews on opening day, no-one wanted to question the magic by doubting the veracity of the identification. Could the underlying woman really be a hidden portrait of Mary, Queen of Scots, unseen for 400 years, or were we all, like so many previous enthusiasts, getting carried away by the enduring appeal of this sitter?
Comparing face tracings taken from the underlying portrait to painted and drawn portraits of Mary which can be argued to be “authentic” during her life/ in the early 17th century provided a second “eureka” moment, when the tracing made of the features and outline of the face of the underlying woman’s face matched very closely to a similar tracing taken from one “authentic” portrait in particular, the Blairs Memorial Portrait (Blairs Museum, Aberdeen). As the latter work can be argued to represent an “authentic” likeness of the sitter, having been commissioned by one of the sitter’s ladies in waiting during the last years of her life, Elizabeth Curle, the identification of the female sitter in the underlying portrait as Mary, Queen of Scots, can be strongly argued to be correct. The underlying portrait, produced in Scotland in the 1580s, and the Blairs Memorial Portrait, produced after the death of the sitter (<1618), were clearly both based on a further portrait source. In order to further investigate the potential source for the likeness, tracings of both works were compared to the other painted/ drawn extant authentic portraits, including early portraits of the sitter, in order to ascertain which (if any) were the closest match and therefore could possibly have performed the function of model for the full size portraits.
A finding of this comparison, as has previously been noted by earlier scholars in relation to the Blairs memorial portrait, was that the Rijksmuseum and Blairs miniatures are the closest likenesses. Of the two, the tracing of the Rijksmuseum miniature proved the closest match to the features of both the Blairs Memorial portrait and the underlying portrait. That being said, the tracings of the two full-sized portraits are a much closer match to each other than to the Rijksmuseum miniature. This may suggest that a further (now lost) face pattern or miniature, may have been in circulation during the late sixteenth/ early seventeenth century. It would not be implausible to hypothesise that this might be the miniature famously gifted to Elizabeth Curle by Mary on the day of her execution, or a copy thereof.
Further inferences about the underlying female portrait can be gained by consideration of the costume worn by the sitter in comparison to the miniatures and French court portraiture. The latter aspect, visual comparison of tracings, and possible reasons for the abandonment of the underlying portrait, will be discussed further in the presentation. Using similar methodology, in conjunction with archival research and results, where accessible from technical examination, it has also been possible to add to the extant debate surrounding further “authentic” portraits of this sitter created during her period of captivity (including the “Sheffield” full length and Hilliard miniature portraits), and to begin to probe the international circulation of face patterns of this sitter in the period.
|With thanks to Professor Aviva Burnstock at the Courtauld Institute of Art and the other Trustees of the Caroline Villers Fellowship for awarding me nine months funding to complete this research, and also to colleagues at The National Galleries of Scotland and The National Trust without whose generosity and collaborative spirit this research would also not have been possible.|
If you are interested in learning more, watch out for the forthcoming book chapter on this topic by Caroline Rae and David Taylor
[i] David Taylor, ‘Questions of Attribution’ in National Trust Historic & Collections Annual (Apollo) 2013, pp. 55-9.