Scans in Art Work Appraisal
October 25, 2005
Since the late 1970s, CT scanners have probed patients’ bodies to help identify the causes of their illnesses. Just as the technique of computed tomography imaging revolutionized the practice of medical diagnosis in its time, its contemporary use in the art world could ultimately change the way some works are appraised. The quality and reliability of the images produced by a CT scanner – also called a computed axial tomography scanner or a CAT scan – literally “undress” the art work and reveal its internal structure.
The CT scanner, or “CT”, as it is more briefly known, provides a more accurate measurement of the density of the component parts of the object under examination, thereby dissociating parts that are usually merged on a conventional X-ray film. It therefore has its place among the various scientific disciplines used to clarify the history of art works: manufacturing techniques, initial functions, later uses, preservation, etc.
In conventional radiology, the X-ray beam projects onto the film the accumulated shadows of the component parts of the object it goes through. Low-density areas are completely masked by the shadow of denser parts. CT avoids this drawback by enabling each part to be viewed separately. The principle is to record a series of “slices” or sections of the object. The images are recorded in digital format and special image processing software is used to construct sections on any spatial plane desired. The sections can be combined to give a global view of the object and the transparency of the component parts can be modified at will. These operations reveal valuable information about the object’s background:
- manufacturing techniques,
showing for example, whether pottery was modeled from a clay coil, turned
on a wheel or stamped,
The technique of CT scanning, when combined with a pertinent interpretation of the images obtained, is a powerful diagnostic tool which can provide proof of the inner state of an art work. Although it can help establish the history of the piece, when used in conjunction with other techniques of observation and analysis, it is not a dating test.
Like all sophisticated equipment, the CT has its limits. It was designed principally to examine the human body so the three-dimensional objects scanned must not exceed a diameter of 50 cm or the weight of a man. Although a medical CT can handle most materials, an industrial CT should be used for metal items.
CT scanning of antiquities is not recent: in 1979, the year when the Nobel Prize for Medicine was awarded to Allan M. Cormack and Sir Godfrey N. Hounsfield for inventing the CT scanner, Dr Derek Harwood-Nash published the first article on his use of a CT to study an Egyptian mummy. He was already fully aware of the advantages that this essentially non-destructive technique had to offer scientific fields such as Egyptology, Paleontology and Archeology.
Since then, precisely because of its non-destructive nature, CT scanning has been widely used to examine objects from the past, as it leaves them intact for future generations. For example, a CT scan of an Egyptian sarcophagus can pinpoint the ceramic or stone scarab hidden in the mummy’s thorax and give a precise view of the underside. Any hieroglyphs engraved there can then be photographed, enabling an Egyptologist to decipher the history of the deceased without violating the sarcophagus.
Detailed reports of other applications in the cultural heritage field have been published in the press, in specialized magazines or on the Internet: CT scans have assisted in the study of stringed instruments (violins, cellos and guitars), archeological pottery vases, antique glasses, wooden and ivory netsuke, real and fake fossils, and dinosaur eggs. More recently, CT scanning has proved useful in making models of ancient musical instruments, identifying art works for insurance purposes, and making three-dimensional polymer reproductions of art works (stereolithography).
The value of CT for art works lies in the fact that each particle of material making up an object has a measurable density, which differs to some extent from the adjacent particles, because the composition of non-synthetic materials is seldom homogeneous. The density of these particles remains perfectly stable over time because it is mainly dependent on the concentration and average atomic weight of its atoms. When a physical or chemical change occurs in the object it will show up unfailingly as a change in the density of the material. For example, oxidation will cause a lowering in the density of the oxidized material in relation to the non-oxidized particles.
Since it is designed to measure the density of matter, the CT is particularly appropriate for detecting any change in the state of a material. The result of the examination is usually presented as an image in which the densest particles appear dark grey or black, while those of lesser density are of a lighter grey. Unfortunately, these images, displayed in a scale of greys, do not always speak for themselves! They must be interpreted, just as, in the medical field, a radiologist is needed to interpret a scan of the spinal column, for instance, to see whether or not the vertebrae are fractured. The quality of this interpretation obviously depends on several criteria such as the quality of the CT, its initial calibration, the adjustment of the parameters used for the examination, and the type of algorithm used to construct the images.
Apart from its non-destructive character, the CT has another substantial advantage over other scientific tests in that it examines the entire object rather than just a sample.
When used under optimal
conditions, its resolution power, that is the minimal diameter of a particle
of material for which the CT can measure a density value, is about 0.05
mm. This resolution power is therefore largely sufficient to detect, for
instance, a crack in a wooden, ivory, pottery or stone sculpture, even
if the crack is invisible to the naked eye and does not show up on a conventional
Wood is an ideal material for studying with a CT, firstly because it is an organic material that is partly dehydrated and therefore of low density, and secondly because it has growth rings of varying density, arranged in a regular pattern which is disturbed by the least tampering. A CT study of growth rings in a wooden sculpture can be a valuable aid for detecting the assembly of different pieces of wood, whether they are of the same or different species, as well as any gluing or breaks; it can even be used to identify the species. It can be still done if the sculpture has been covered with opaque varnish, or a thick layer of patina or paint, even if the latter contains metallic pigments such as white lead. This is often the case for polychrome wood sculptures, on which paint and patina may be used to hide defects.
For wooden sculptures, CT scanning has the added advantage of showing the extent of damage caused by borer or termites and even detecting the eggs or larvae of living wood-boring insects. Such a diagnosis is valuable in recommending a conservation treatment such as anoxia in a nitrogen environment.
In the tribal art field, ritual stigmata, fetishes or magic charges are sometimes concealed inside the sculpture or under a layer of condensed organic matter, and non-invasive investigation may shed light on their nature and original function.
Objects made of other low density organic matter such as plant fibers, leather, and horn, can also be studied by a CT, despite the lack of visible growth rings. The weave of concealed fabrics is also clearly shown by this technique.
Bone and ivory are the densest organic materials. Yet they can easily be CT scanned because the size of the piece is usually within acceptable bounds.
A CT scan of the Burmese ivory stupa (fig.3, above), illustrated opposite, reveals its mysterious contents: a standing statue of Buddha, finely carved by removing the debris through the openwork screen. The Buddha and the screen were carved from a single piece of ivory, as is proven by the unbroken root canal which runs through the entire sculpture. Indeed, the ivory comes from a hypertrophied incisor, necessarily innervated during the elephant’s lifetime.
A close study of the Ekoi head, illustrated further on, shows a human skull under the layer of antelope skin. This suggests that it dates from a period prior to the British protectorate in Nigeria (late 19th century). Moreover, an anatomical study of the skull shows that teeth from a carnivorous mammal have been inserted in the upper jaw, giving the head a particularly frightening aspect as tradition demands. Lastly, an examination of the frontal sinuses reveals a magical charge, containing some metal, hidden under the bony wall of the sinuses and covered with clay and antelope hide. It is probably because the charge was completely invisible that it was not removed before the head reached the European market.
Of all the materials examined, terracotta is the one for which the CT yields the most information on manufacturing techniques.
Since clay is initially soft, it keeps a trace of everything it has been in contact with before firing, whether it is the stand on which it was modeled, the potter’s fingers or tools. Because of its relatively adhesive nature, fresh clay sometimes incorporates dust or residue of varying density. These marks enable the radiologist to trace the sequence of steps in the creation of the work and to pinpoint any inconsistencies. A plausible explanation must be found for these inconsistencies, in a field where fakes are legion, as was confirmed by many of the experts recently interviewed by Thomas Fuller.
A CT scan of a terracotta object also permits a study of the granulometry of the metal flecks it contains as well as of its overall density. Generally speaking, both the granulometry and global density of the clay are constant in the same sculpture, as the artist theoretically shapes his work from the same stock of clay. Moreover, the CT gives information about the metal or organic supporting structures used by the artist during modeling, whether or not they survived the firing process. Lastly, it reveals the way stamped pieces were assembled, the quality of the glaze, changes made before firing and traces of early paintwork.
On the other hand, when previously fired terracotta material – excavated bricks, for example – is cut, carved, hollowed out, assembled, scraped, sanded, painted, etc., to produce a sculpture which looks like an original piece, a whole range of other signs are revealed by the CT which frequently prove its recent assembly.
According to an investigation carried out by Sheila Farr and published in the Seattle Times early in 2003, the reconstruction of pottery objects is mainly intended to fool unsuspecting connoisseurs who have blind faith in thermoluminescence (TL) testing to authenticate a work.
It should not be forgotten that TL is used only to determine when a piece of pottery was last fired. The test is carried out on tiny samples taken by drilling. Although it will certainly establish the age of the material, it cannot prove that the object had its present shape when the analyzed material was fired.
The question of the influence that CT scanning may have on a later TL test has been raised. To investigate this, we worked in conjunction with Archéolabs to scan ten objects and eleven samples of pottery which had all been age-tested by thermoluminescence. The TL tests were repeated under the same conditions, after the CT scan, although Archéolabs was unaware of the X-ray exposure the specimens had received. The findings showed that the age determined by TL, both for the ten objects and the eleven samples, was the same as before they went through the CT.
This experiment therefore confirms unpublished experiments made by other laboratories indicating that CT scanning antique terracotta, under appropriate testing conditions, in no way modifies the TL results.
No literature has yet described CT scanning of monolithic sculptures. The reason for this is probably to be found in the apparent incompatibility between the relatively low penetrating power of the X-rays used in medical CT (compared with industrial CT) and the high density of the stone.
Yet experience shows that medical CT scanners are powerful enough for stone sculptures under 40 cm in diameter. A CT scan is even a valuable source of information on the inner state of the material, which cannot be established with the other types of scientific analysis, because they mostly study the surface of the sculpture or samples taken from it.
Thus the CT can show whether a sculpture has been constructed from one block or several. If there are several, it can show whether or not they are of the same nature by analyzing their density, the direction of their veins or sedimentary strata, and the quantity of natural metals they contain.
If repairs have been made, separate elements such as metal rods, drill holes, cement joins and injected resin deep within the stone show up clearly.
Fakes can also be detected, such as a head attached to a body made of a different stone.
Lastly, the CT can give information about the outer crust and even find deep-lying causes for surface anomalies. Thus a crack running around the statue does not necessarily mean that the stone has been broken and glued together again, but may be due to the natural erosion of an oxidized sedimentary stratum.
When a collector is interested in an art work, one of his primary concerns is to establish its authenticity. This is based on several subjective factors (experience, pedigree, expert opinion) to which are progressively added, depending on the importance of the work, criteria based on a number of scientific studies: stylistic analysis, thermoluminescence or carbon 14 dating, a dendrochronological study, spectroscopic or microscopic analysis, etc. Alongside these technical tests, which focus mainly on the visible parts of the work or on a few samples, CT scanning is an absolutely non-destructive test that has the advantage of describing the inner state of the object, examined this time as a whole.
CT scanning – or computed tomography – can therefore provide valuable information about an art work’s background by:
- revealing its contents,
CT scanning is thus a valuable aid for people with various interests in art, such as collectors, dealers, art experts and historians, museum curators, anthropologists, ethnologists, paleontologists, stringed instrument makers, and scientific archivists, but it can also be of use in a wider context, in restoration workshops, auction rooms, legal offices, insurance companies, and even forensic science laboratories.
Dr Emmanuel Agneessens, Jean-Luc Berrier, Claire Boullier, Philippe Bourgoin, Robert Courtoy, Bernard de Grunne, Bernard and Catherine Decamp, Prof. Jacques Devière, Georges Dewispelaere, Prof. Robert F. Dondelinger, John Eskenazi, Serge Estiévenart, Marc Léo Félix, the Ghysels family, Marc and Denyse Ginzberg, Philippe Guimiot, Karim and Isabelle Grusenmeyer-Bilquin, Baronne Dora Janssen, Anne-Catherine Kenis, Ralf Kotalla, Olivier Langevin, Michel Leveau, Pierre Loos, Dr Jacques Mathieu, Jean-Pierre Mohen, Antoine Moons, Dr Donat Nicod, Yves and Anne Peemans, Dr Thierry Puttemans, Myriam Serck-Dewaide, Bernd Schnakenberg, René and Anne Vanderstraete, Dr Jean-Hubert Vandresse who were the first to encourage me in this field.
Isabel Ollivier (English translation).
Archéolabs TL, Le Châtelard, Saint Bonnet de Chavagne, France
Siemens AG, Medical Division - Computed Tomography, Forchheim, Germany
Dr Marc Ghysels, has a degree in medicine, with a specialization in radiology. He comes from a family of artists and collectors and has set up a business in Brussels conducting radiological appraisals of art works.
all text & images © Dr Marc Ghysels