1 portrait on display in Room 2 at the National Portrait Gallery
- Tudor and Jacobean Portraits Database
by Unknown English artist
oil on panel, circa 1595
30 3/8 in. x 24 5/8 in. (771 mm x 625 mm)
Click on the links below to find out more:
New attribution: Unknown English artist
Key findings: Changes to the position of the facial features during the painting process suggest that the portrait was painted from the life.
The portrait was purchased by the Gallery by Private Treaty Sale in 2006. It was bequeathed by John Donne to his friend Robert Ker, 1st Earl of Ancram and the painting remained within the familys possession until its sale. Whilst owned by the Earl of Ancram and his descendants it was hanging in one of the family homes: Monteviot in Roxburghshire and Newbattle Abbey outside Dalkeith; the portrait was previously known as the Newbattle portrait. For some years it was mislabelled as John Duns Scotus, the medieval poet. This misidentification was corrected by the Gallery in 1959.
This portrait is the earliest and most important lifetime portrait in oil of John Donne, and it shows him playing the role of a melancholic lover. Dr Thomas Morton, Bishop of Durham, recalled seeing a portrait of Donne in the chambers of one of Donnes friends at Lincolns Inn, all envelloed with a darkish shadow, his face and feature hardly discernable, which likely relates to this image, or an earlier version of it (Cooper, 2006, p. 175).
The inscription reads: Illumina Tenebr(as)/ Nostra domina (O Lady lighten our darkness) and is a reworking of psalm 17:29 in the Vulgate Bible: Deus meus illumina tenebras meas (O Lord lighten my darkness).
Notes on likely authorship and justification
The painting technique is characteristic of an English painter. However, as the handling of certain elements, including the position of the facial features and the angle of the body, have proved challenging, the artist may not be a practised portrait painter. Given the nature of the pose and format, the portrait must have been carefully orchestrated by Donne; the inscription suggests that it may originally have been painted for a lover or for a friend. It is possible that the painter was a friend or associate perhaps a painter working on theatrical events at the Inns of Court who took instruction directly from the young poet about the nature of the composition.
Commentary on painting style, technique and condition
There is significant damage to the paint surface in several areas, particularly to the background at the right side, where a large area of missing paint has been restored. The paint surface on the face and costume has also been abraded. However, it is evident that the original painting is executed in a simple and straightforward way but appears to lack the confident understanding of the handling of oil paint found in painters trained in traditional workshop methods. In comparison to the face, the flesh paint for the hand appears dark because it was thinly applied over the costume, allowing the dark paint beneath to play a role in the tonal modelling.
Conservation treatment undertaken in 2012 indicated that there had been several previous campaigns of restoration and that the panel had previously split into three parts and been rejoined, perhaps in the eighteenth or nineteenth century, indicating that it was recognised as an important historic object worthy of restoration.
Justification for dating
The technique and materials used are entirely consistent with a work from this period. Analysis of the wooden panel by dendrochronology has indicated that the wood derives from a tree which cannot have been felled before 1581. This is consistent with the interpretation of the historical evidence which suggests a date of c. 1595.
Drawing and transfer technique
No carbon underdrawing is evident. Infrared reflectography has shown that the positions of the mouth, left eye and hat were changed during the painting process. This suggests that this portrait was painted from the life rather than from a pattern. The position of the mouth was raised a little and the left eye was previously painted both above and below the final position.
Relevance to other known versions
There are no other known versions of this portrait.
Baddeley, Richard, Life of Dr Thomas Morton, 1666, pp. 101-2
Colclough, David, Donne, John (1572-1631), ODNB, 2004
Cooper, Tarnya, Searching for Shakespeare, National Portrait Gallery, London, 2006, pp. 39-40, 57, 94,175-6
Cooper, Tarnya, Elizabeth I and Her People, National Portrait Gallery, London, 2013, pp. 212-3, no. 71
Davis, Nick, Melancholic Individuality and the The Lothian Portrait of Donne, American Notes and Queries, vol. 26, issue 1, 2013, pp. 5-12
Edwards, David, L., John Donne: Man of Flesh and Spirit, 2001, pp. 38-40
Frost, Kate Gartner, The Lothian Portrait: A Prolegomenon, John Donne Journal, 15, 1996, pp. 95-125
Keynes, Geoffrey, A Bibliography of Dr John Donne Dean of Saint Pauls, 1973, pp. 373-4
Patterson, Annabel, Donne in Shadows: Pictures and Politics, John Donne Journal, 16, 1997, pp. 1-35
Piper, David, The Image of the Poet: British Poets and their Portraits, 1982, pp. 24-31
Sabine, Maureen, Illumina Tenebras Nostras Domina Donne at Evensong, John Donne Journal, 19, 2000, pp. 19-44
Strong, Roy, Tudor and Jacobean Portraits, 1969, p.65-66
Strong, Roy, The English Icon: Elizabeth and Jacobean Portraiture, 1969, pp. 36-7
Strong, Roy, The Elizabethan Malady. Melancholy in Elizabethan and Jacobean Portraiture, The Tudor and Stuart Monarchy: Pageantry Painting, Iconography, vol.2: Elizabethan, 1995, pp. 295-302
Thornton, Dora, and Jonathan Bate, Shakespeare: Staging the World, British Museum, 2012, no. 26
The panel has a slight warp but is in a stable condition. It is evident that the panel boards have been separated and rejoined. The paint surface has suffered considerable losses of ground and paint layers down the joins, along the top edge, the right part of the lower edge, and in the background to the right of the shoulder on the right. There are other smaller losses, including a small loss just below the eye on the right. Interpreting the paint layers is made difficult by the various restoration campaigns which obscure parts of the original paint. It is difficult to differentiate between original glazing and later intervention. There are thin opaque glazes of restoration on the hat and the cloak and more complex restoration on parts of the reddish brown costume. There are areas where the paint is a little unstable and raised in small blisters, associated with the extensive areas of old paint loss.
Conservation treatment in 2012 at the Hamilton Kerr Institute
The relatively recent varnish and contemporaneous retouching were removed, along with a moderately yellowed natural resin coating and extensive mismatched retouching, which had been applied widely across areas of intact original paint, and areas of older filling. The cleaning revealed the detail of the doublet sleeves more clearly and the original outline of the entire shoulder and arm, which had previously only been approximately positioned in the previous campaigns of restoration and retouching. The original profile of the hat on the left-hand side was recovered from beneath overpaint. The oval background was found to have variations in colour and tone, which were presumably intended to give an illusion of a space behind the sitter. The visible inscription is original and no retouching or reconstruction were carried out.
Number of boards: 3
Panel Orientation: Vertical
Panel condition observations
The panel has been slightly thinned on either side of the panel joins. Uneven areas of wood on the reverse have been filled with brown putty. Some of this putty has been lost and in other areas is raised and loose. Two small areas of damage at the edge of the panel can be seen on the right-hand side (from the back). An area of split wood at this edge is slightly loose. Two old drill holes are visible in the lower right-hand corner and upper half of the left-hand board.
An examination of tree rings, which can help to provide the earliest possible felling dates for the wood used for the panel. The technique can also indicate the geographical origin of the wood.
Number of boards: 3
Last date of tree ring: 1573
For the purposes of analysis, the three board panel was labelled A-C from left to right (from the front). No sapwood is present on any of the boards. Due to the brittle nature of the edges of the central board, it was not analysed. The analysis of the left and right boards found them to derive from the same tree. The last tree rings identified for boards A and B were 1572 and 1573 respectively. Adding the minimum expected number of sapwood rings gives an earliest possible felling date of 1581. Applying an LEHR-usage range to these boards provides a conjectural usage date of 1581-1613.
A technique used to identify changes in composition beneath the surface of the paint layers and to understand the physical structure of a panel or canvas.
The closely aligned vertical wood grain and broad, horizontally applied lead white priming are clearly evident. The rapid method for applying the priming is evident in the brushwork. The brushwork of the mouth appears confused due to the changes in the position of the mouth (see x-ray mosaic 01).
The level of damage to the paint layer is clearly visible; along the panel joins, lower right- hand corner and both the upper and lower edges, and including small isolated paint losses throughout. In addition, a large irregularly shaped loss, approximately 130 mm x 130 mm in size is evident to the right of the shoulder on the right. The x-ray clearly shows how old fill and restoration extends beyond the paint loss in many areas, covering intact original paint, particularly along the upper edge and left-hand join (from the front) which indicates that full restoration could improve the current appearance. Most of the filling material appears very dense in the x-ray, indicating that the material contains a proportion of lead. But the large loss in the background to the right of the shoulder on the right and some of the smaller losses, including the loss below the eye on the right, do not appear dense and evidently do not contain lead. The non-original, diamond-shaped wooden buttons on the reverse, and the old inventory number present in the lower-left corner on the front are also visible in the x-ray.
A technique used to observe the layers beneath the paint surface which can reveal underdrawing and changes to the initial design.
No carbon underdrawing is evident using infrared reflectography.
The change in the position of the mouth can be seen clearly with infrared reflectography. The position of the mouth was raised a little. The first moustache was shorter than the final one. Below the final moustache the line between the first lips runs through the current lower lip. The curl added to the ends of the final moustache appears to be restoration. The position of the eye on the right has been changed. It is evident with infrared reflectography that the eye has been painted both a little lower and a little higher than the final position. The damage in and below the eye intrudes into the lower pentiment. The grey glazes of overpaint on the face and the strengthening of the nose contour can also be seen using infrared reflectography (see infrared reflectogram mosaic 01).
Investigation into microscopic pigment samples or samples of other media in order to help with dating, to reveal the order of the paint layers and help to understand the painting techniques used.
Paint samples were taken to analyse the pigments and layer structure of the painting in October 2009.
Paint sampling showed that the whole panel was prepared as a rectangle at the same time, and that the spandrels beyond the feigned oval have the same ground and priming layers as the rest of the painting.
The flesh in the opening at the top of the shirt seems to contain the same type and character of pigments seen elsewhere, and therefore seems original, although it has a slightly different grey appearance
to the shadow further up the neck.
There is a thick chalk ground and a relatively substantial priming containing lead white, traces of charcoal black and reddish orange particles which appear to be red lead, giving it a warm grey hue.
The flesh is painted with strong red, both opaque and translucent.
Sample 11: One tiny sample contained fine lead white, vermilion, strong red lake, and traces of yellow ochre.
Samples 2 and 7: The black coat contains plant black with quite large angular particles (sample 2). Bone black with rounded particles was used on the sleeve on the right (sample 7), this may be soft shadow.
Sample 6: Dispersion of the red in the sleeves contains red lake, lamp black and some red ochre.
Sample 3: Lamp black and possible yellow lake was found in original paint beneath restoration.
Sample 9: The gold fringe glove is painted with a mixture of vermilion and lead-tin yellow.
Background within the oval
Sample 14: Painted in two layers. The first is a dark layer with lamp black, red ochre, occasional brownish red lake, lead white, occasional very bright yellow which seems likely to be lead-tin yellow. There were traces of translucent yellow earth (possible sienna) in dispersion. The upper layer is more a more translucent brown with a higher proportion of red, some black and apparently no white or yellow. A translucent brownish red may be organic red lake.
Background in spandrels outside oval
Samples 1 and 13: Painted with a mixed dark layer.
Samples 3, 4 and 5: The same layer structure beneath filling material, overpaint and varnish.
An examination of the construction of paint layers, glazes and condition often using a microscope. This method provides important evidence concerning an artist’s technique and paint handling and can reveal a specific artist’s characteristic painting style.
Painting style and method
It was difficult to interpret the details of the original technique due to the condition of the original paint surface and the level of restoration. Forming a clear understanding of the layer structure and level of original paint present in the background and costume was particularly problematic, due to the presence of fine drying craquelure, lead soaps, paint loss, abrasion and multiple campaigns of restoration.
The painting is executed in a simple and straightforward way, and lacks the confident understanding of the handling of oil paint found in painters trained in traditional workshop methods. There is a small inventory number, 180, at the lower-left corner.
There is a chalk ground and a priming layer which contains lead white with occasional particles of red lead and charcoal black.
A large number of lead soaps are visible on the paint surface, which originate from the lead-based priming (see micro 12). There is also a fine network of drying craquelure throughout, which appears to be associated with the ground and priming interface (see micro 03). This could perhaps be due to the presence of oil in the ground layer, drying at a slower rate to the fast drying lead-based priming above.
The position of the eye on the right has been changed a little. This can be seen using infrared reflectography. It is also confirmed with microscopy, where it is evident that the eye was painted both below and above the current position. The pupils and the whites of the eyes have been reinforced with restoration, especially in the eye on the right. There is also a restored loss in the lower part of the eye on the right (see micro 01 and micro 02).
Further examination during conservation treatment in 2012 suggested that the alteration to the position of the eye occurred at an early stage as the pentiment is covered by the pale paint of the face prior to the addition of the grey shadows.
The position of the mouth has been raised a little (see Infrared reflectography). This can be seen in the paint surface where the initial line between the lips runs across the current lower lip, and the moustache was in the position of the line dividing the lips. The lips are painted with a lot of vermilion, red lake, white and black. The definition and parting has been reinforced but this is original. This contains a lot of red lake and the same mixture with additional charcoal black (see micro 06).
Further examination during conservation treatment in 2012 suggested that the change to the position of the mouth occurred at a relatively late stage because the earlier position appears to only have been covered by the uppermost grey modelling layer.
Face and hand
The flesh paint in the face was thinly applied over the light priming layer. The horizontal application of the priming can be seen in the flesh paint where the upper layers were thinly built up over it. The face appears dark because grey shadow was applied over many parts. The initial flesh paint layer contains lead white, vermilion, a little black and red lake and the shadow and contour were applied over this using more black, red lake and some yellow ochre. The charcoal black and other pigments are large in this darker flesh tone. There is restoration in many parts of the face over the thin but abraded original paint. The shadow in the forehead was thinly applied over the flesh colour and seems to have been brought across from the hat colour. The opaque grey shadow on the face is restoration, but the brownish tone is original (see micro 05). The nose is considerably restored. The curl of the moustache has been exaggerated with restoration. The light grey highlight on the upper chest at the base of the neck, between the open edges of the collar, appears to be an original highlight (see micro 16). The flesh paint for the hand appears dark because it was thinly applied over the costume, allowing the dark paint beneath to play a role in the tonal modelling. Minor adjustments were made to the contour. The paint appears thickest in areas of highlight, such as in the nails where there is a high proportion of lead white (see micro 09). The flesh paint is a mixture of lead white, black, vermilion, red lake and traces of yellow ochre. A very fine lead white was found in one paint sample (see Paint sampling). In areas of shadow the paint is more thinly applied and appears to contain a higher proportion of black, with a higher medium content. More red and lake pigments were used in warmer areas. Veins on the back of the hand are described in the paint surface.
There are many losses in the costume with considerable restoration. The costume is difficult to interpret and this area could be significantly improved as a result of full restoration. It appears to be painted with a brownish red underlayer with a red lake glaze (see micro 10). This can be seen most clearly under the lace collar where there is no overpaint. The reddish brown layer has a fine craquelure and is evidently rich in oil medium. The creases in the sleeve fabric are modelled with a lighter red mixture, with a thinly painted pattern of regularly spaced slashes across the creases (see micro 18). The brown braid attached to the upper part of the arms has been considerably strengthened with restorations and overpaint. The original tone of the braid was evidently much less dense than the overpaint. Red lake can be seen beneath small areas of abrasion. There appears to have been a thin brown glaze over this, but this is not clear due to the condition and to the layers of restoration. There are fine tufts round the edges of the braid which appear to be original (see micro 15). Paint sampling showed red lake, lamp black and some red ochre in the sleeves. Plant black and bone black were also found in the costume paint (see Paint sampling).
During conservation treatment in 2012 small areas with an intense red lake tone were exposed that had previously been protected by early overpaint or putty. The underlayer appeared to be a moderately deep grey rather than a brownish red. This would imply that the original appearance was more purplish than is now apparent.
The gold fringe (gold covered thread) on the glove cuff is painted with fine brushstrokes of lead-tin yellow and vermilion (see micro 04).
The lace was thinly applied with pale grey, containing black and white, using a small soft brush. Dark brown/black details were painted with a very fine brush.
There are three collars, each lying over the one beneath (see detail 06 and micro 19). There are small blue areas, which appear to be small splashes of indigo, on the hanging collar tie (on each side, more or less opposite each other). The collars are painted in a similar way to the shirt cuffs, with lead white, black and some ochre (see micro 08).
The hat was painted before the background. The top of the hat was lowered a little during painting, this can be seen in the paint surface. The hat has considerable restoration and the top layer of restoration is thin and opaque.
The inscription seems to be painted mostly with lead white with some lead-tin yellow, the paint mixture also contains a little vermilion and red lake (see micro 03).
The background is very difficult to read due to the discoloured varnish layers, the restoration and the fine craquelure (see micro 12 and micro 13). Despite this, it appears to be painted in one or two layers above the pale priming layer (see micro 13).
Paint sample analysis has shown that the background within the oval seems to be painted with two layers with complex mixtures. The dark lower layer contains lamp black, red ochre, lead white, occasional brownish red lake, translucent yellow (probably sienna), and occasional bright yellow particles (probably lead-tin yellow). The upper layer is darker, reddish and more translucent, and it contains a higher proportion of red, some black and apparently no yellow or white. There seems to be some organic red lake.
Analysis of the background in the spandrels outside the oval showed only the lower dark grey layer.
Order of construction
- Chalk ground
- Costume and hat
- Hands painted over costume
- Highlight on the flesh visible between the open collar edges
- Feigned oval
Lead white, charcoal black, lamp black, bone black, plant black, red lead, vermilion, red lake, earth pigments, lead-tin yellow
Changes in composition/pentimenti
The position of the mouth has been raised. The position of the eye on the right has been changed and during the search for position it was painted both above and below the current position.
There is considerable old paint loss, especially on the panel joins, and there is considerable overpaint and restoration on some parts. This has confused the form in some areas, such the sleeve on the left.
Some parts of the paint surface are a little raised and unstable. The surface is uneven and the varnish is glossy.
A method which helps to reveal past restoration and the extent of varnish layers.
Examination in ultra violet light shows that there is a great deal of old natural resin varnish, which appears opaque in ultra violet light. The tone of fluorescence varies at the edges of the varnish surface where it has been protected by the frame. Considerable restoration is visible in ultra violet light. It is evident that there have been several campaigns of restoration, carried out without removing the varnish. Small areas of the most recent restoration appear darkest and seem to be above the natural resin varnish. However, most of the restoration appears less dark and is evidently covered by natural resin varnish. This can be seen on the hat, on the sleeve on the left where there are old paint losses down the join, on the cloak over the shoulder on the right and several areas in the lower right part of the painting. Other areas with a lighter tone than these restorations are evidently beneath further layers of the natural resin varnish; such as the large loss in the background, losses down the panel joins and near the centre of the right edge. Small spotty restorations in the face are also have a lighter tone than the more recent retouchings. The hand on the right appears to have less varnish over it than the face (see UV 01).