Our guest blogger today is Ethan Charles Blocher-Smith , an IUPUI chemistry graduate student studying the effect of gene knockouts on lipid synthesis in yeast. His favorite painting is Botticelli's "The Birth of Venus," with Caravaggio's "Medusa" as a close second.
Part Two in a series of blog posts on the ongoing technical examination of a purported 1874 Alfred Sisley Impressionist painting.
Is the IMA’s Alfred Sisley Impressionist painting a fraud? In an earlier blog post we covered the curatorial concerns regarding this painting and the efforts of students in an innovative IUPUI Chemistry Department seminar to help resolve the mystery. Today’s post reveals the results of a recent imaging campaign using the IMA conservation department’s state-of-the-art cameras for color photography, as well as UV-induced visible fluorescence, infrared, and radiographic imaging.
By way of a cursory physical description of the artwork, House in a Village (Fig. 1, left) is a canvas painting on a wooden stretcher with a central cross member support bar. The reverse of the painting shows that it was relined onto a new support fabric at some point prior to 1952 when the museum acquired it. This is evident from the pristine nature of the canvas backing. Unfortunately, the tacking margins where the painting is nailed to the wooden stretcher have been covered with a thick, aged masking tape, thereby obscuring any relevant details from the canvas border. This is regrettable since it is often the case that the artist’s paints extend over the front of the canvas onto the margins, which are not visible when the painting is in its frame and hence are excellent locations for taking small paint samples for analysis. But we are getting ahead of ourselves.
A raking light image (Fig. 1, right) taken with the painting illuminated obliquely from the left reveals numerous distortions in the canvas, including ripples along the top and bottom edges due to poor stretching and tented cleavage of paint and ground away from the canvas along aging cracks. A large depression in the center of the canvas is also visible. The cause of this surface anomaly is immediately obvious when the painting is photographed under a blacklight. Artists’ materials develop fluorescence as they age, while more recent painting, such as that from a restorer’s retouching, hasn’t yet had the time to develop the same level of fluorescence. As a result, the presumed original oil paint will glow softly while the restoration areas will appear much darker. The central depression is shown in the UV-induced visible fluorescence image (Fig. 2, left) to be the result of improper filling of a large damage. A right angle tear is seen as dark against a fluorescent background, and another smaller tear exists below it. Other areas of the painting show dark patches of missing original paint under UV examination. The reason for the relining of this canvas is now obvious; the work at some point has suffered a serious blow leading to paint loss and tears in the artwork.
Radiography helps to delineate the tear further. These x-rays are akin to the sort of medical imaging done in a hospital, except instead of recording bones and flesh, the radiograph of our painting shows the distribution of radio-opaque pigments, metal tacks, and wooden supports. Many artists’ paints include heavy metal pigments, which block the x-rays and appear in bright contrast on the x-ray. Most
pre-20th century paintings begin with a white or lightly toned priming layer on the canvas called a ground. The principal white pigment at this time was lead white, a basic lead carbonate, and so areas where the lead-containing ground is missing, for instance in the repaired tears, do not show up in the x-ray image. Additionally, we can see numerous heavy strokes of lead white paint in the radiograph that are used as highlights in the village scene. However, some of these bright brushstrokes, for instance those in the upper left-hand corner, do not correspond to features in the surface image. What can account for these rogue bits of paint?
Another imaging technique called near infrared (NIR) transmittography helps to clarify this situation. This type of image is collected by illuminating the back of the canvas with an incandescent lamp and capturing the transmitted radiation with a camera only sensitive to infrared wavelengths. Low energy NIR photons transmit through many materials, but not through paints containing carbonaceous pigments, like charcoal or graphite. This imaging technique is often utilized to look through painted layers to identify the cartoon or sketch done by the artist prior to applying paints. These “underdrawings,” which were used to plan out the composition, were often executed in charcoal or pencil and hence show up in dark contrast on the NIR image. Infrared imaging can also reveal abandoned or overpainted pictures that have used black pigments. In our painting, the infrared transmittogram when viewed on its side reveals a sub-surface painting, perhaps a landscape (Fig 3, left). Traces of bright green paint peeking through the upper layers of blue sky also suggest a composition below the surface painting. Although the features of the lower work are obscured by those of the upper painting, one is tempted to visualize the outstretched branches of a tree on the left, a horizon line, and numerous wispy brush strokes that could indicate sporadic clouds populating the sky (Fig. 3, right). Was this a period canvas reused by a forger in order to provide a convincing reverse for a fake, or is it perhaps a previous work abandoned by Sisley – or another painter – and overpainted with House in a Village? Answers to these questions may be revealed with further analysis.