Hidden: Unseen Paintings Beneath Tudor Portraits
Research undertaken as part of the Making Art in Tudor Britain project has revealed unexpected information about some of the Gallery’s sixteenth-century portraits. Analytical techniques such as infrared reflectography and x-radiography make it possible to examine the layers beneath the surface and have shown that some of the wooden panel supports on which the portraits were painted had been reused
The Protestant spymaster and the Virgin Mary
The Flagellation of Christ and the English statesman
Back to front: The wing of an altarpiece
The queen’s two faces: A hidden portrait
Renaissance recycling: Wall panelling to portrait?
Sir Francis Walsingham
by an unknown artist oil on panel, late sixteenth century. NPG 1704
Infrared photograph of NPG 1704.
X-ray of NPG 1704.
Sir Francis Walsingham established and ran the Elizabethan secret service and his spies operated mainly against Roman Catholic conspirators. This portrait is a reversed version of the only known portrait type of Walsingham, which is associated with the artist John de Critz the Elder, who received extensive patronage from Walsingham in the 1580s. Although the sitter and/or the individual who commissioned this portrait would probably have been unaware of it, technical analysis has shown that this panel has been reused. This has revealed the fascinating juxtaposition of a portrait of the Protestant spymaster superimposed over a small devotional image that appears to show the Virgin and Child.
Detail from the infrared photograph showing the preparatory drawing for the portrait and the outline of two figures beneath the sitter’s face.
This technique of analysis is undertaken with a specially modified camera and a bright light source. Infrared radiation is part of the spectrum of normal light. It is absorbed by carbon black, and as a result the images can reveal evidence of preparatory drawings made by the artist using a carbon medium, such as charcoal. In the portrait of Sir Francis Walsingham the preparatory drawing that marks out the features of the face is clearly evident in the infrared photograph, and the confidence and placement of the marks indicates that a pattern has been used. The infrared photograph also revealed the partial outlines of two figures painted beneath the portrait. The loose brushstrokes that mark out the profile of the figure on the left suggest that this underlying composition may be unfinished.
The underlying composition
Detail from the x-ray showing more detail in the underlying figures.
X-radiography revealed more information about the composition beneath the portrait. The pose of the figure on the right and the loose long hair suggests that it is likely to be a depiction of the Virgin Mary. The figure on the left could be St. Joseph or an angel. Stylistically, the composition appears to be Northern European. Similar images were widely circulated in the form of prints by artists such as Albrecht Dürer and Martin Schongauer. There is no evidence of any damage to the underlying painting; the damage that is evident in the x-ray occurred when the painting was a portrait and not before. Dating the panel The panel is constructed from a single board of Eastern Baltic oak. Dendrochronological analysis has been used to date the wood, and suggests that the support was first used between 1547 and 1579, whilst the portrait of Walsingham is dated to the mid-1580s
1st Earl of Dorset
by an unknown artist,
oil on panel, 1601. NPG 4024
Digital infrared reflectogram of NPG 4024.
X-ray mosaic of NPG 4024.
Thomas Sackville was a poet, playwright and statesman, and the owner of Knole house in Kent. He was made Lord Treasurer in 1599 and Lord High Steward in 1601; the rod of office that he holds in this portrait probably refers to his appointment as Steward. Sackville was created Earl of Dorset in 1604.
This analytical technique is capable of recording a wider range of wavelengths than infrared photography. It clearly shows the underdrawing that marks out the details of the sitter’s face in a carbon medium. The drawing follows a pattern for the features, but there are also freehand marks that model the face in a more three-dimensional manner. There are also suggestions of an underlying composition evident in the face. This is much clearer elsewhere in the portrait; for example, two figures are revealed beneath the fur of the sitter’s gown.
Detail from the digital infrared reflectogram showing the underdrawing in the face, and also the suggestion of an underlying composition.
Detail from the digital infrared reflectogram showing the figures in the underlying composition.
Digitally balanced x-ray mosaic detail showing the figure of Christ at the column.
X-rays are obstructed by heavy metals such as lead, which is present in certain paint mixtures such as lead-tin yellow and lead white. As a result, x-radiography can be used to examine changes in composition beneath the surface of the paint layers. However, the final image can be difficult to interpret because all of the paint layers are superimposed over each other. In this case the x-ray revealed enough detail for it to be possible to identify the source for the composition as the depiction of the Flagellation of Christ by the Italian artist, Sebastiano del Piombo, which is in the Borgherini Chapel, San Pietro in Montorio, Rome.
This composition became more widely known in Europe through various copies and engravings. The figure of Christ at the column stands out particularly clearly in the x-ray because the paint mixture in these areas contains a higher proportion of lead white. Some parallel marks across Christ’s torso are evident in the x-ray, and these could be deliberate score marks that were intended to damage the paint surface. However, although religious imagery was restricted in post-Reformation England, certain religious images could be quite legitimate if they were kept away from places of worship and were not used in ways that were considered to be idolatrous. The overpainting of the Flagellation is, therefore, not necessarily an example of the iconoclastic destruction of an image; it could have been produced within a studio and then failed to sell.
Dating the panel
Dendrochronological analysis suggests that the boards used to make this panel derive from trees felled after 1590. The portrait of Lord Sackville celebrates the culmination of his political career, his appointment as Lord Treasurer in 1599, and it is inscribed in the top left ‘1601 æta: 63’. This is inaccurate because he was said to be 72 years old at his death in 1608, but it is possible that the artist incorrectly noted the detail of the inscription and put the sitter’s age at 63 instead of 65. The characteristics of the paint handling suggest that the portrait was produced in the early seventeenth century and there is therefore a very narrow gap, of no more than eleven years, between the first use of the panel for The Flagellation and the production of the portrait. This might suggest that the same artist’s studio could have been responsible for the production of both works.
Detail photograph of the inscription.
Decorative scheme, front of the panel, NPG 844
Unknown man, reverse of panel
by an unknown artist, oil on panel, mid sixteenth century NPG 844
This panel was given to the Gallery in 1890 when it was thought to be a portrait of the Protestant theologian Richard Hooker (1554-1600). The identity of the sitter is unknown but it is probably a Dutch or Flemish patron. The composition on the other side of the portrait relates to the resurrection, with the trumpeting angel raising the dead man from the tomb.
Detail photograph of the upper-right corner, showing the unpainted strip at the edge and one of the notches
It was thought that this portrait had been painted on a reused panel; however, close examination suggested that both sides of the panel were painted at the same time. A thin strip of unpainted wood is evident on the right-hand side of the decorative scheme and also on the corresponding side of the portrait. This has a series of regular notches cut into it, which may record where pins were placed to attach the panel into a supporting structure or frame. The other edges of the panel are much rougher, which suggests that the panel has been trimmed at some point, possibly in order to remove it from a framework. As a result some of the details, such as the top of the architrave, have been lost.
A triptych altarpiece?
The proportions of the monochrome composition, and the incomplete inscriptions, suggest that the panel is a fragment of something that was originally at least twice as large. However, adding a second panel to complete the composition would mean that the portrait, which from the other side would form the right-hand side of the two panels, would have been awkwardly posed facing outwards. As a result, it seems more likely that the panel was part of a door, or wing. This is a structure common to small altarpieces and the pose of the sitter is very similar to that of a donor.
Detail image of the background of the portrait.
Dating the panel
The painted background of the portrait was previously thought to date from the eighteenth or nineteenth century. However, sampling showed that the paint mixtures contained the pigments azurite, lead-tin yellow and smalt, which suggests that the paint is likely to date from the sixteenth or seventeenth century rather than later. This dating was further refined by dendrochronological analysis, which showed that the tree used to make the oak panel support was felled after 1517, and was likely to have been used at some point between 1517 and 1549. The costume of the sitter suggests that the later end of this time bracket is most likely.
Detail image of the inscription on the cartouche.
The Dutch inscription on the cartouche shows that the monochrome composition is clearly incomplete.
What remains reads: O HERE – (O Lord -)
ŇS INI – (our -)
MET VWĒ – (with thy -)
Detail image of the inscription on the architrave.
In full, the Latin inscription in the architrave may have read ‘NON MORIAR SED VIVAM’, which is from the seventeenth verse of the 118th Psalm: ‘I shall not die, but live, and declare the works of the Lord’.
Other portraits painted on reused panels
Queen Elizabeth I
by an unknown artist,
oil on panel, 1580-90.
Infrared reflectogram mosaic of NPG 200.
X-ray of NPG 200.
This portrait of Queen Elizabeth I is not painted from the life but based on a pattern and numerous versions of this portrait type would have been painted, although very few survive today. Elizabeth is shown wearing the Lesser George, the medal signifying membership of the Order of the Garter, on a ribbon around her neck.
As has been seen, infrared reflectography not only reveals carbon underdrawing. The long wavelength of the infrared radiation penetrates pigmented paint layers to varying degrees, which can provide information about what lies beneath the surface of the paint layers. In this portrait, the infrared reflectogram mosaic clearly shows a snake that was originally painted in Elizabeth’s hand. This was an emblem of wisdom, prudence and reasoned judgment. However, in the Christian tradition serpents also signify original sin and temptation by the Devil. The ambiguity of the emblem may have resulted in it being painted out at a late stage of the painting process, and replaced with a bunch of flowers. The snake is now partly evident on the surface of the painting as a result of the degradation of the paint layer. The infrared reflectogram mosaic also shows that the panel has been reused; the features of another woman are clearly evident beneath the portrait of the queen.
Infrared reflectogram mosaic detail of the overpainted snake.
Detail from the infrared reflectogram mosaic, showing the features of the underlying portrait.
Detail from the x-ray mosaic, showing the overlaid portraits.
The x-ray provided more information about the painting beneath the portrait of Elizabeth. It shows a female head in a higher position, facing in the opposite direction. The identity of the original sitter is unknown, but the unfinished portrait appears to have been very competently painted, possibly by a different artist. The woman appears to have been depicted wearing a French hood of a type that was fashionable in the 1570s and 1580s, and dendrochronological analysis provides a conjectural usage-date of 1572-82 for the panel. This suggests that there may have been a period of a few years before the panel was re-used.
This painting does not feature in the display ‘Hidden’, and is currently on display at Kenilworth Castle, Warwickshire.
An unknown woman, possibly Margaret Douglas, Countess of Lennox, by an unknown artist,
oil on panel, c. 1560-1565. NPG 401
X-ray mosaic of NPG 401
Reverse of NPG 401.
The identification of the sitter in this portrait is uncertain. However, comparison with other known portraits suggests that it is possibly Margaret Douglas, Countess of Lennox. Her son, Lord Darnley, married Mary, Queen of Scots and she was the grandmother of King James VI of Scotland and I of England.
X-ray detail showing the rose.
X-ray revealed the outlines of roses that resemble the heraldic ‘Tudor’ rose. These had been painted on the reverse of the panel and could be seen because x-rays penetrate both the paint layers and the support. The resulting x-radiograph contains all of the information superimposed onto one image. The roses were more evident in the x-ray than on the surface of the reverse because the pigments contain heavy metal; most likely lead in lead white and red lead.
Dating the panel
Dendrochronological analysis showed that the wood used to make the boards came from the same slow growing tree. The boards are also comparatively narrow, which suggests that they have been trimmed. As the tree was slow growing, even a small amount of trimming could have removed evidence for a significant number of rings and it is, therefore, difficult to date the panel. The last heartwood ring was dated to 1424, which suggests that the tree was felled at some point after 1432. The conjoined red and white rose was a heraldic emblem that signified the union of the houses of York and Lancaster through the marriage of Henry VII and Elizabeth of York in 1486, and if the painting on the reverse does depict ‘Tudor’ roses, it must be later than this. The costume of the sitter in the portrait dates to the 1560s.
A reused panel?
The x-ray also shows the unusual horizontal alignment of the two boards used to make the panel. This suggests that the panel may have been reused in order to make the portrait, particularly when coupled with the discovery of the heraldic roses on the reverse. The panel may initially have formed part of a larger decorative scheme, perhaps as a section of wall panelling. However, it is possible that that the two sides of the panel were painted at the same time as it is relatively unusual to find a panel that has been prepared with a smooth painting surface on both sides and there are surviving examples of portraits with heraldic emblems relating to the sitter on the reverse. For example, the 1430s Portrait of a Man holding a Book, possibly Guillaume Fillastre by the Workshop of Rogier van der Weyden, in the Courtauld Gallery, which has the sitter’s device of a holly branch painted on the reverse.
This painting does not feature in the display ‘Hidden’.
Case study 1 - Portraits of two medieval kings
Case study 2 - 16th Century double portrait
Case study 3 - Four portraits of Edward VI
Case study 4 - Henry VII
Case study 5 - Portrait of Bishop Foxe
Case study 6 - Portrait of John Astley
Case study 7 - Portraits based on designs by Hans Holbein the Younger
Case study 8 - Portrait of Sir Thomas Gresham
Case study 9 - Portrait of Sir Henry Lee
Case study 10 - The Phoenix and the Pelican
Case study 11 - A portrait set of early English Kings and Queens
Case study 12 - Two Portraits of Robert Dudley, 1st Earl of Leicester
Case study 13 - Portraits of Henry VIII