In The Color of Ivory: Polychromy on Byzantine Ivories, Carolyn L. Connor argues that modern scholarship devoted to Late Antique and Byzantine ivory carvings perpetuates a fundamental misunderstanding. The impression the objects create today, she avers, results from systematic cleaning in the modern era. With handling and seasonal changes in temperature and humidity, the painted surfaces of carved ivories became rubbed or flaked; rather than restore the losses, owners found it easier to simply remove the paint, although many pieces remain visibly stained where dyes penetrated the surface. The wholesale cleaning of Late Antique and Byzantine ivories may also reflect the imposition of an early modern aesthetic based on conceptions of what is appropriate to the artistic media, for sculpture an unpainted surface. In the past scholars have recognized that at least some gilding and painting were the final stages in preparing an ivory for use. Richard Delbrueck, in his study of consular diptychs (Die Consulardiptychen und verwandte Denkmaeler [Berlin, 1929]), discussed the issue generally and specifically reported where traces of paint or gold were located on the surfaces of those works that comprised the catalogue of diptychs. More recently, Anthony Cutler ( The Hand of the Master [Princeton, 1994]) has noted that some Byzantine ivories have been cleaned to remove what may have been original paint; he discusses the use of gold and polychrome but is careful to leave open the question of how extensive the painting of figures might have been. Connor pushes aside issues of dating, provenance, iconography and workshop practices in order to focus the reader's attention on the question of painted surfaces. She opens on a rhetorical note, claiming that we resist envisioning the ivories as painted despite the evidence for it.
Connor sees her first task as presenting undeniable proof of the painting of ivory. She has examined, under microscope, one hundred ivories in the United States, Europe and Russia (listed in an appendix, pp. 84-87). The specimens range in date from the early fifth to the eleventh century and comprise consular diptychs, icons, caskets, pyxides and some miscellaneous pieces. She reports that of the one hundred ivories examined, ninety-five had some remaining color or gilding. From the sample, Connor isolates about a dozen works that she presents as "case studies." These she describes in order to give the reader some sense of where the traces of paint, dye or gold were observed (garments, imperial regalia, inscriptions, background detail and so on). For the vast majority, Connor only reports the different colors she saw through the microscope, not where on the surface she observed them or how extensive the remains were. The findings serve mainly to undergird a general thesis, rather than to offer information that others might use in discussing individual works.
In the second chapter, Connor turns to the difficult problem of how to verify the contemporaneity of carving and painting. Her first approach is based on the relative uniformity of the colors red, pink, orange, green, blue-green and blue; Connor establishes this uniformity by comparing the colors seen under a microscope with those of Munsell charts. The ranges are published in a simple list (p. 24), but individual values are not reported in the appendix, where they might have been valuable. Did lighting conditions affect the observations? Were the observations from Late Antique ivories in any way different from those made from Byzantine ivories? The argument can be summarized as follows. If medieval ivories were not painted until modern times, after they had entered collections on three continents, then we might expect observable variations in the colors used. The uniformity she reports is reassuring but does not constitute an ironclad argument for the contemporaneity of carving and painting. We should bear in mind the use of the word "color" in two different contexts. What we think of as the color of a garment or background element, a uniform blue or green, is a perception formed by observation at a distance of a foot or more. A painted area seen through a microscope can appear to consist of a number of materials of quite different color values. A cluster of crystals or clump of pigment may not be sufficient grounds from which to infer the shade of a tunic or color of a sandal. Many of the Renaissance artist's pigments were the same as those available to the Byzantine painter, yet each used the materials to create colors that appear to be different when seen at a normal viewing distance. The speck of pigment adhering to the surface of a cleaned ivory may not always be subject to date using a microscope; nor may it invariably indicate the color the artist intended us to see.
The second argument is more difficult to assess. At Connor's request, members of the Department of Objects Conservation of the Metropolitan Museum analyzed color samples from two Byzantine ivories in the collection: the Joshua casket fragments (acc. no. 17.190.137a-c) and a complete casket (acc. no. 17.190.238). Connor was also able to take one or more samples from the blue background of the Forty Martyrs triptych in the Hermitage Museum, St. Petersburg; these were analyzed in New York and the results are compared in a chart (p. 27). The analytical method employed was energy dispersive X-ray spectrometry, and excerpts of the reports written by those who performed the tests are published in the text. When interpreted, the test data reveal percentage amounts of elements. The percentages are then compared with those derived from samples of materials known to be in use in the Middle Ages. The knowledge of materials derives from two sources. One is the recipe books surviving from Antiquity and the Middle Ages (though not from medieval Byzantium). The second source is the data collected from the analytical study of medieval works of art. Interpreting the data is hardly straightforward, and, like most of the readers of this book, I am not competent to criticize the analytical findings. I would, however, like to offer a brief critique of the design of the experiment.
If we accept Connor's contention that there is widespread reluctance to acknowledge the extent to which Late Antique and Byzantine ivories were painted--rather than, say, scholarly reticence in the face of matters for which there is little unambiguous evidence--then we must ask if she has succeeded in breaking down resistance by means of a compelling case. The results from tests performed on two ivories and samples of a single color taken from a third present a narrow base for argument. Two of the three works subjected to analysis, the Hermitage Triptych and Metropolitan casket, have a marked difference between the amount of polychrome surviving on the background and that remaining on the figures. After the samples were taken from the Metropolitan casket, it was stripped of much of its paint (prior to exhibition in the The Glory of Byzantium. Art and Culture of the Middle Byzantine Era: A.D. 843-1261, catalogue eds. H. Evans and W. Wixom [New York, 1997], cat. no. 78); the decision to remove the paint represents a verdict on the date when it was applied. It is reasonable to ask if the two ivories were painted or repainted after they left the Byzantine world. Anthony Cutler (1994) has suggested that the backgrounds of both works were applied in the eighteenth or nineteenth century, in response to what he calls "`medievalizing' impulses." If repainting is even a remote possibility, then neither work can provide the conclusive evidence that Connor seeks to establish; using such results may even damage an otherwise promising case. It would, I think, have been useful to try to correlate the results of the tests with those that have been performed on Byzantine and Armenian manuscripts, which are undeniably medieval. What begins with a reasonable sample of one hundred works in the end comes down to analytical data presented on one trustworthy piece, the Joshua Casket fragments.
At the conclusion of chapter 2 the medieval painting of ivory is taken as proved. The remainder of Connor's study is a survey of sources. Middle and New Kingdom Egypt, ancient Crete and Mycenae, Syria and Phoenicia, Greece and Rome all offer evidence in the form of either objects or, more often, prose writings and poetry. Some of the sources are as much as five millennia before the time of the ivories under discussion. Connor is careful not to offer this material as evidence for Byzantine practice, but it can serve no other purpose. Chapter 4, "The Testimony of Ancient and Medieval Texts," promises information more relevant to Late Antique and Byzantine ivories, but the authors cited are mainly ancient ones: Homer, Pliny, Pausanias, Cicero, Claudian and others. The Late Antique source noted, the Augustan History, does not mention the use of color on ivories. The Byzantine testimony is a late ninth-century reference, in Philotheos's Kletorologion, to imperial diplomas held in ivory diptychs, like drawings in a portfolio. Philotheos's mention of color pertains to the diplomas--perhaps their background color, but more likely the imperial signature--not to the ivories. Nevertheless, Connor relates this passage to the Dumbarton Oaks diptych leaf (acc. no. 37.18), which contains traces of color that K. Weitzmann (Catalogue of the Byzantine and Early Mediaeval Antiquities in the Dumbarton Oaks Collection, III, Ivories and Steatites [Washington, 1972], 56) thought were a Western addition. No evidence cited in this chapter supports Connor's contention that Byzantine ivories were brightly painted objects. Silence does not vitiate the argument. Anyone familiar with Byzantine literature will know that it is barren ground for those seeking references of the kind needed to confirm the painting of ivory. Ancient sources are incomparably richer, but what Homer and Pliny may have said or what was found at Nimrud and in King Tutankhamen's tomb may not be directly relevant to Late Antiquity and Byzantium. The argument for the value of ancient sources is made on the grounds of Byzantine admiration of Antiquity, but this is by no means a straightforward matter, particularly as it pertains to sculpture.
In the last substantive chapter, 5, Connor turns to color as an essential element of the Byzantine world, in clothing, on the walls of churches and devotional objects. This is necessarily a superficial survey, yet several specific parallels are presented. A metalwork object from the San Marco Treasury is cited as evidence of the effect that the maker of the Harbaville Triptych might have sought. The blue background of the Joshua casket fragments is compared with that of a manuscript (Cod. Sinai. gr. 512). More of these kinds of comparisons, pursued in detail and with a concern for date, might have been helpful if used in conjunction with the findings made through the microscope. Comparing an ivory with a manuscript has merit, though the example cited may not be the first comparison that comes to mind. The Joshua Casket is the one Byzantine ivory that can be related with confidence to a nearly contemporary work of art, the Joshua Roll (Cod. Vat. Pal. gr. 431). Connor would have performed an immense service, and perhaps advanced her argument, if she had compared the distribution of the color on the ivory panels with that of the roll. The lurid color reconstruction (pl. IV) does not resemble the treatment of the Joshua Roll (perhaps any more than the polychromed sculpture on the exterior of the Philadelphia Museum of Art reflects that of ancient pediments). One wonders if the limited use of color by the painter of the Joshua Roll might not have been a factor in its having been chosen as a source for the ivory. In the conclusion, Connor insists that brilliant and heavy coloration was a regular feature of Byzantine art, including ivory sculpture. The art of book illumination shows us that this is not invariably true. A number of manuscripts have miniatures executed with thinned pigments and a restricted palette.
Connor's book raises an issue that deserves to be settled. Doing so will require more investigation, and even then it remains to be seen if the analytical methods currently available can offer evidence that will allow us to compare ivories to better preserved works in other media. Much of this slim volume has been devoted to irrelevant testimony. More diagrams like those on pp. 90-92 (showing where polychrome was found on the rosette borders of the Joshua Casket) are required in order to make available to a wide audience the evidence needed to debate the issue. Even then, if the paint has been irretrievably lost from many ivories, no amount of imaginative reconstruction will get it back.