14. Defining a Romano-Egyptian Painting Workshop at Tebtunis

The collections of the Phoebe A. Hearst Museum of Anthropology, University of California, Berkeley, include eleven mummy portraits excavated between 1899 and 1900 from Tebtunis, Egypt.¹ This group constitutes one of the largest assemblages of Roman-period mummy portraits to remain both together and unrestored since excavation; as such, it presents a rare opportunity to explore the local practices of an ancient painting workshop. The group also contains, on the back of an effaced portrait, direct evidence of painting practice: a sketch with annotations detailing how the image should be completed (fig. 14.1).² The APPEAR project inspired the collaborative technical study of the eleven Tebtunis portraits as well as a single, additional portrait purchased from Theodor Graf for the University of California (PAHMA 5-2327; fig. 14.2), which provides a comparative example of a stylistically different portrait attributed to the site of Kerke.³

Sketch, shown in NIR illumination, with instructions in Greek for completing the portrait, found on the reverse of a nearly effaced portrait, Tebtunis, Egypt, second century AD. Encaustic paint and ink on sycomore fig panel, 34.8 x 22.5 cm (13 11/16 x 8 7/8 in.). Berkeley, University of California, Phoebe A. Hearst Museum of Anthropology, 6-21378a.

fullscreen Figure 14.1

Mummy Portrait, Romano-Egyptian, second century AD. Attributed to Kerke, Egypt. Glue tempera paint on linden panel, 35 x 19 cm (13 ¾ x 7 ½ in.). Berkeley, University of California, Phoebe A. Hearst Museum of Anthropology, 5-2327.

fullscreen Figure 14.2

A multimodal approach based on noninvasive analytical techniques—including in situ X-ray fluorescence spectroscopy, Fourier transform infrared spectroscopy, hyperspectral imaging, and photometric stereo imaging—was used to compare the portraits and determine whether the use of materials and techniques across the group defines a local workshop.⁴ Noninvasive analyses then targeted the selection of a limited number of paint microsamples to confirm material identifications and provide information about pigment distribution within the layers.⁵ Wood substrates were identified via scanning electron microscopy of tiny samples from the panels.⁶

Distinctive features of nine of the Tebtunis portraits (PAHMA 6-21374, 6-21376, 6-21377, 6-21378a [see fig.14.1], 6-21378b [see fig. 14.3], 6-21379, 6-21381, 6-21382, and 6-21383) set them apart from the rest of the paintings currently in the APPEAR corpus. As the corpus grows, these attributes may identify related paintings in other collections. These portraits are unique in the APPEAR corpus, as they have gypsum chalk white preliminary sketches on the obverse and/or Greek script on the reverse. Additionally, these nine Tebtunis portraits are on thick (1.2–1.5 cm) sycomore fig panels; thick panels and the use of sycomore fig are less common within the APPEAR corpus. Sycomore fig was used for fewer than 20 percent of the panels identified for the APPEAR project, including this group of nine.⁷ Finally, two pairs of portraits within this group of nine share features that are identical in both appearance and execution and are unlike any portraits outside the group. Two male subjects (PAHMA 6-21377 and 6-21378b [see fig. 14.3]) wear gilded wreaths that were painted first with an indigo-based paint, and two female subjects (PAHMA 6-21381 and 6-21383) wear on their third fingers gold bands rendered in yellow ochre and gold leaf.

Mummy Portrait, Romano-Egyptian, second century AD. Tebtunis, Egypt. Encaustic paint on sycomore fig panel, 34.8 x 22.5 cm (13 11/16 x 8 7/8 in.). Berkeley, University of California, Phoebe A. Hearst Museum of Anthropology, 6-21378b.

fullscreen Figure 14.3

Although additional features unite these nine paintings, none are unique to the Tebtunis portraits. The paint on all eleven Tebtunis portraits has a beeswax-based binder that was applied with brushes and metal tools. Lead white is a component of most colors, and nearly all tones are based on a mixture of iron earth pigments, including hematite, goethite, jarosite, and a manganese-rich “umber.” Egyptian blue has been identified within the APPEAR corpus in the shading and outlines of faces, the backgrounds, and in blue and blue-based colors. Among the tighter group of nine Tebtunis portraits, Egyptian blue appears only in the neutral background color.⁸ Blue, purple, and pink shades are all based on the organic pigments indigo and madder. Certain pigments detected on other mummy portraits—namely, orpiment, realgar, and cinnabar—were not detected on these nine images.

Marked differences between the Tebtunis portraits and the portrait purchased from Graf (PAHMA 5-2327; see fig. 14.2) show the range possible in materials among known mummy portraits.⁹ The Graf example is painted on a thin (0.3 mm) sycomore fig panel with an animal glue–based paint¹⁰ over a white, calcium sulfate foundation layer; it has no discernible underdrawing. Its palette relies even more heavily than the Tebtunis portraits on iron earth pigments. No lead white, madder, indigo, Egyptian blue, cinnabar, orpiment, or gold leaf was detected.

APPEAR project participants are finding that the portrait painters, wherever they worked, had a similar range of materials available to them. Workshop practice within the Tebtunis portraits is defined and differentiated from the rest of the APPEAR corpus not by unique materials, but by specific choices the painters made from readily accessible tools and the subtle differences in how these artists employed the materials. With the possibility not only to identify materials but also to map how and where they are used in a painting, we can begin to distinguish the practice of a workshop.