Interpreting technical and analytical evidence in historical context
Head of the Department of Conservation and Technology, The Courtauld Institute of Art
Making Art in Tudor Britain
Abstract of a paper presented at Tudor and Jacobean Painting: Production, Influences and Patronage
Funded by the British Academy and The Paul Mellon Centre for Studies in British Art
Interpretation of the results of technical analysis of sixteenth-century paintings in an historical context presents particular challenges that are a central theme of the conference. It is important to qualify the application of technology to this kind of study. While the methods are based on science, their application to study works of art does not satisfy the fundamental criteria necessary for interpretation of data from experimental scientific study. The sample size - the number of paintings investigated - is not large, and the extant works may not be representative of sixteenth-century artistic output. Technical study is selective, and often not reproducible. There may be practical and ethical limitations on gathering similar data from all the works in a study, and thus the results are not statistically robust.
The paintings from this period that survive vary significantly in quality and condition and for many of the works there is little or nothing known about the date, sitter, provenance or specific function. The few works for which there is documentary evidence and other historical references - for example, artists’ treatises and documents related to the purchase of materials and artists’ workshop practices - are thus important points of reference for interpretation of new technical evidence.
The Courtauld Institute of Art
© Aviva Burnstock
An example of the interpretation of the evidence from a technical study in an historical context is provided by a painting from the Courtauld Gallery Netherlandish Man with a Skull, which is a typical example of a work made in the middle of the sixteenth century about which little is known. The study provides a vehicle for introducing the methods, their limitations and the results, and how the initial focus on historical questions of identity and authorship evolved through interpretation of the technical evidence, to consider new questions of purpose and function.
The painting was bequeathed by Lord Lee of Fareham to the Courtauld Gallery in 1932. It is dated 1550 - by an inscription - but neither the sitter nor the painter is known. Scholars suggest it is Netherlandish, leading to its current attribution. The support consists of three boards of Baltic oak, joined with the grain vertical, with original wood tool marks on the reverse. The panel compares closely to other sixteenth-century panels from the National Portrait Gallery, made from similar oak imported from the Baltic region, that were used for painting English sitters. This physical evidence supports what is known about the trade and supply of oak that was used for panel painting in both the England and the Netherlands, but we cannot say with confidence where the panel was made. Expert analysis (by Ian Tyers) of the patterns of growth rings present in cross-sections of timbers from which panel paintings are made provides information about the felling date of the tree from which the boards were cut. This technique has provided an earliest possible date for some paintings and has proved decisive in some issues of the dates of copies. Ian Tyers’ knowledge of the wood trade has advanced our ability to identify marks on the back of panels that relate to their origins as shipments of plans that were used for may purposes - such as wainscoting or coffins - and with accumulated knowledge it may be possible to identify planks from the same forest, or even individual trees, in related panel paintings.
© Aviva Burnstock
X-rays can penetrate the entire thickness of a panel (the dark passages) and where the image looks light the x-rays passing through have been absorbed by the painting materials. The contrast is largely determined by the presence of lead white pigment (that is highly absorbing) and the thickness of the panel. The radiograph of the Netherlandish Man provides information about three oak boards that are butt joined, originally using wooden dowels and later additional buttons. The flesh paint is thinly applied with sparing use of lead white pigment in the flesh, which registers just discernable the face and hands of the sitter, superimposed by wide sweeping strokes associated with the priming layer that also contains some lead white. We cannot identify specific pigments used for the work, but we can see small paint losses where the paint is absent, which are not visible through an opaque yellow varnish on the surface of the painting. This is useful as an indication of the condition of the work in relation to possible future conservation treatment.
Detailed examination of the surface
Examination of the paint surface using a light microscope can provide detailed information about the original materials and techniques and the order of painting, and the presence of later additions to the work, including inscriptions and signature. The deterioration of original materials and characterization of campaigns of restoration can be identified using a combination of microscopy and material analysis. Light microscopic examination provided evidence for the originality of part of the inscription on the Netherlandish Man and areas of later strengthening. A microscopic image of a coat of arms on the sitter’s ring shows the quartering of identical shields, which is useful for heraldic identification.
© Aviva Burnstock
© Aviva Burnstock
Characterization of paint layers; identification of pigment and medium
The preparation of minute paint samples provides an opportunity for identifying specific or unusual materials and techniques used for painting, trends in the composition of underlying preparatory layers and characteristic deterioration of pigments. Identification of earlier restorations that are part of the physical history of the painting allows interpretation of trends in interpretation and display of the work at different times in its history. A sample from the dark-coloured background of the Netherlandish Man prepared as a cross-section shows
a chalk ground, particles of a carbon black pigment in a fluid medium that is associated with underdrawing beneath the paint, covered by a thin imprimatura of lead white. The dark paint of the background contains particles of the copper carbonate blue pigment azurite that typically darkens the oil medium that binds the particles. This could suggest that the background that now appears brown was originally blue. Organic analysis requires small samples and separation of media components using instrumental techniques such as chromatography and infrared spectroscopy. Using different techniques it is possible to identify with some precision the organic binding media. The limitations are related to the use of mixtures applied in thin layers. Characterization of different organic binding media, such as the use of oil or egg tempera, resin or protein may explain how the artist achieved a variety of painterly effects, and also some aspects of the condition of the paint, such as the loss of adhesion between layers.
A very useful technique for sixteenth-century paintings, generally thinly painted, where carbon black-containing underdrawing (in dry or fluid media) contrasts with the white - typically chalk - preparation layer. Infrared radiation penetrates paint and is reflected by the white ground, except where carbon black is used for drawing. This might be applied freehand or show more regular characteristic lines that are indicative of the use of tracings or transfer of designs from face patterns. Other examples include visible drawing that deviates or remains close to the painted image, or drawing that is mechanical or freely applied. Drawing in media other than carbon black (applied over a white background), or where the drawing is covered by paint that strongly absorbs in the infrared range of the infrared detector (such as areas of azurite paint), will be obscured or rendered invisible.
Showing another composition beneath the portrait
© Aviva Burnstock
Infrared images of the Netherlandish Man show a complex composition beneath the portrait, freely applied in a fluid medium that contains carbon black. There is no evidence of drawing for the portrait, but the composition visible under the paint in infrared shows figures in an architectural setting that are of sufficient detail to compare with drawings. The stylistic comparison with extant drawings is challenging and imprecise, however, similarity between the underdrawing and the vidimus, in the Bowdoin College Art Museum, USA, and other drawings in the British Museum, by Dirk Jacobsz Vellert (Dirk van Staren c.1480-c.1549? painter, engraver and designer of stained glass windows; registered as a Master of the Antwerp Guild of St Luke in 1511). Vellert possibly worked in England between 1526 and 1538, and certainly designed windows for Kings College Chapel, Cambridge. Though this comparison has clear limitations, it illustrates the possible use of technical evidence and its interpretation.
The evidence from technical study of Netherlandish Man with a Skull gathered so far has shaped new historical questions about workshop practices in the sixteenth century.
For example: Were panels used for different purposes and then re-used, perhaps by other artists working together? How did Netherlandish material practices influence English artists working here and in the Netherlands – and did English and foreign painters/etchers/stained-glass designers and other artists/craftsmen collaborate, share studios and materials?
The original questions of the authorship of the painted portrait and the identity of the sitter have not so far been clearly illuminated by technical study. While the comparison of underdrawing visible in infrared images with Vellert’s drawings might be compelling for me, you might reasonably argue that the comparison is made only because they survive, and are available at reasonably high resolution from the British Museum/Bowdoin website that facilitates the comparison. There may be much closer comparisons to extant drawings by Antwerp artists that are unattributed, or perhaps there is nothing that survives that is comparable.
The limitations of analytical methods and temporal advances in technology applied to study art are another significant variable – it is possible that improved infrared imaging methods would reveal more information that could be critical.
Critical mass of evidence is important: it seems likely that with the accumulation of data from a number of new technical studies of paintings from this era patterns will emerge. Interpretation of this evidence by collaboration between specialists with different expertise has proven successful in informing historical questions and forming new hypotheses.