James Cahill dies at 87; scholar of Chinese art

74. Rokushô-yake: Green Pigment Burns Silk

74. Rokushô-yake: Green Pigment Burns Silk

For many years I carried on a running argument with a Chinese friend who is a collector-scholar, and whom I won’t name because I will argue here that he was wrong. It was about a chemical phenomenon we could observe in old paintings, especially those done on silk: a deterioration or eating-away of the silk brought about chemically by the action of some pigment used by the artist on the silk support. He maintained that it was a yellow pigment that caused this, and insisted that it was a well-known phenomenon, familiar to all Chinese collector-connoisseurs; he was much given to lecturing me with bits of wisdom that “we Chinese” all knew, and this was another. But as readers of my writings know, I have long been a skeptic of those old familiar Chinese “truths,” believing that they often mislead us more than they inform us, or are only partial truths that block acceptance of other truths outside them. This was such another such case. I argued, on the strength of what I had learned from Japanese collectors and mounters, that the deterioration was caused instead by the mineral green pigment made from malachite and called in Japanese rokushô, so that the phenomenon had come to be known to the Japanese as rokushô-yake, “malachite-pigment burning.” We continued this argument in the presence of students and others in October, 1981, when we were together in a viewing at the Palace Museum, Beijing, seeing a handscroll landscape by Li T’ang, his “River Temple in the Long Summer” (see Richard Barnhart’s section of our 3,000 Yrs of Ch Ptg book fig. 120, p. 130, the latter half of the scroll), a work that has notably suffered from this affliction. Neither of us would give way, and the dispute remains unresolved. But I mean here to offer, as a poor substitute for a scholarly article I meant to write and never did, the evidence I have gathered for my belief. I do this without having followed up, for more than twenty years, the leads and clues I had gathered then, so that what follows is surely outdated. Still, it should have some interest and value.

I had learned about rokushô-yake in the course of going around among Japanese collectors, dealers, and mounters, over many years, looking at paintings with them. (I did the same with Chinese collectors and dealers, and learned a great deal from them as well.) Notable paintings on which it can be seen, besides the Li T’ang handscroll, include the Liaoning Museum version of the “Nymph of the Lo River” handscroll composition attributed to Ku K’ai-chih (3,000 Yrs fig. 43A, p. 14), and the “Summer Rainstorm” painting, one of three surviving landscapes-with-figures of the four seasons, in the Kuonji, Yamashiro (Siren Ch Ptg, III, Pl. 243.) All have suffered serious damage and disfigurement from this chemical deterioration. I could add a number of others, including Japanese Nanga-school works of the 18th century.

In May, 1982 I wrote a long letter about this problem to Dr. Thomas Chase, who had succeeded John Gettens as the head of the Freer Gallery of Art’s Technical Laboratory. I described the viewing in Beijing, and quoted the argument of my Chinese friend that it was a yellow pigment, t’eng-huang or gamboge, that caused the deterioration. I had talked earlier with Tom Chase about this question, and been told that it was definitely the malachite pigment that was at fault. I quoted from this earlier conversation in which he had told me—I was asking for written confirmation—that while the yellow pigment was poisonous (“one reads admonitions against painters using it and licking their brushes”), it doesn’t damage the painting support in this way.

Tom Chase passed on my letter to John Winter, his associate at the Freer, who had done research on this and related problems, and who in July 1982 wrote a long response to my letter. (I have preserved all this correspondence; the whole thick folder, holding also copies of the technical articles listed below, will go to the Archive in my name at the Freer Gallery, and can eventually be consulted there by interested people.) About gamboge, he wrote: “This is a yellow-colored vegetable resin, and as with many organic resins, it appears possible for it to migrate slowly, at least through paper. Occasional examples of yellow areas appearing faintly on the back of a scroll may be encountered. This must, however, be distinguished from a chemical attack on the painting support, and I have no evidence that gamboge is responsible for such attacks.”

He continues: “There is plenty of evidence that green copper-based pigments will sometimes eat into either silk or paper. This effect is by no means confined to East Asian paintings, and is notorious, for example, in the case of Persian and Indian miniatures. It is also known in some western graphics and watercolors, including, I think, works on parchment.” He goes on to note two papers presented at recent scientific gatherings that attempted to analyze examples, but remarks that no one seems to understand yet quite how it happens. He then attempts a tentative explanation based on his own observation and a paper he recently heard:

“There are a few clues. Cupric ions in solution are known to catalyse the degradation of cellulose, as are various other transition-metal ions, such as ferric compounds. Since there are plenty of examples of paintings having malachite (a basic copper carbonate) as the green pigment that do not show support degradation in the green areas, we assume that the insolubility of malachite usually prevents such an effect occurring. Malachite can, however, be decomposed in acidic conditions, and this may make the copper more catalytically active. Alum, which is sometimes used in support sizing and so on, can give acids in the presence of water, and the paper (enclosed) by Emoto . . . seems to suggest that malachite in the presence of alum may give rise to problems.”

So, there is the hypothesis, and it makes sense: alum is indeed used as a sizing in East Asian paintings, and its interaction with the malachite may dissolve it and bring about the chemical decomposition of the silk support. If further research during the twenty-plus years since then has more completely clarified the problem, I am unaware of it—but I haven’t tried to follow up this correspondence.

Winter’s letter, after putting forth this hypothetical explanation, went on to ask a question: “If this should be so, it would be surprising if similar effects did not also sometimes occur with azurite (also a basic copper carbonate) as the blue pigment. Have you ever noticed examples of support degradation in blue areas?” I could not think of cases where something like rokushô-yake had occurred in areas of mineral blue, but on my next visit to Tokyo I put this question to the great Tokyo mounter Meguro Sanji, studio name Kôkakudô (=Huang-hao Lou, taken with pride over having remounted a Wang Meng painting for Chang Ta-ch’ien). Meguro was not only the most accomplished mounter I knew (although less famous than two others in Kyoto, and scarcely known to most specialists and collectors), but also a highly knowledgeable and articulate authority on the physical properties of Chinese and Japanese painting—he had been trained by Chang Ta-ch’ien in Chinese techniques, and was employed for their most important jobs by Chang, by the collector Cheng Chi, and by Laurence Sickman (for his “Playing Music in the Garden” scroll ascribed to Chou Fang, for one). I spent a lot of time talking with him, and learned a lot. About the problem of whether azurite also caused degradation, he gave me a detailed answer (quoting from a letter I wrote John Winter in September, 1984):

“He [Meguro] said that the problem occurred with iwa e-no-gu, or mineral pigments, in particular gunjô and rokushô, which are azurite and malachite. As for why the effect is produced in some cases and not in others, he first suggested that it might be some effect of the glue, which is particularly heavy in the mineral pigments, and later said that the alum that is mixed with the glue and put on the paper or silk as a sizing can also be a contributing factor. This is what was suggested in one of the papers you sent me. Finally, he said that the deterioration can be stopped by putting gôsei jushi, which I think must mean something like synthetic resin (?), thinly on the paper from the back. This is also done, he said, to  keep the ink from running or spreading when a painting is remounted.”

Winter responded (letter of April, 1986) expressing interest at hearing Meguro’s report on cases of gunshô-yake or azurite burning, and adding “I am still hoping to find an example myself.”

In my talk with Meguro, as reported to Winter,  I had raised another pigment problem: “the tendency of lead white, used chiefly in later  paintings (I think), to turn to a silvery-grey. I don’t know what causes this, but have seen it fairly often.” I described an album by Hua Yen I had bought that had this problem, and my attempt to get it fixed by having it remounted and telling the mounter (not Meguro—a lesser person) what I had heard: that “the pigment could be restored to its original whiteness by the use of hydrogen peroxide.” The mounter tried this, but too cautiously—there was little change.

Winter responded: “It is fairly well-known for this [lead white] to blacken to a greater or lesser extent. The conventional wisdom is that atmospheric pollutants give rise to the formation of lead sulphide (PbS), which is black, and which can be oxidized to the white lead sulphate (PbSo4) with a suitable oxidizing agent. Hydrogen peroxide has indeed been used to effect this oxidation, the the process has been described in the literature, for example in H. J. Plenderleith & A.E. Wenner, The Conservation of Antiquities and Works of Art, 2nd ed., 1971, p. 87.” Winter adds that he is cautious about this explanation, since he knows of no actual evidence for lead sulphide formation, but that it seems to work.

Finally, another observation about pigments in later Chinese painting. In looking closely at paintings, chiefly flower paintings, by a few artists of the 17th-18th century, notably Yûn Shou-p’ing and Li Shan, I had noticed that some areas of heavy mineral pigment, especially in lighter colors, seen close-up exhibited a curious reticulation, or puddling, making tiny, very subtle natural patterns within the heavy-pigment areas. It was to be seen also in paintings by more recent masters such as Jen Po-nien and Wu Ch’ang-shih, who used it consistently enough to indicate that it was a deliberate effect. I asked various people about this, over the years, without getting a useful answer. One Japanese friend who worked on late Chinese painting expressed great interest, and when his next book appeared it contained a description of the phenomenon. But when I turned excitedly to the endnote to learn his source, it was to see that it was myself, “conversation with James Cahill.” Dead end.

At last, some time in the 1980s, I put this problem to my painter-friend Cheng Shifa. He wrote out an explanation from his own understanding as artist and collector—his Chinese-language note is in the folder, along with an English summary reading: “Painters often added glue in the pigment. They used water to dissolve the glue before blending in the pigment. When the glue used is enough, it does not create sediments. Later painters used less glue, that’s why there are traces of the sediments. Some painters found the effect useful and adopted it in painting. Wu Ch’ang-shih and Jen Po-nien used this technique quite often.”

So, there is the best explanation I was able to get, over quite a few years, for this deliberate puddling of heavy mineral pigments in some later Chinese paintings. I will conclude by listing below the learned papers of which John Winter sent me copies, and that are in the folder. Those seriously interested can look them up; the foregoing amateur’s account is intended to satisfy the more casual interest of non-specialist readers, and is, surely and admittedly, badly out of date.

John Winter, “Deterioration Mechanisms in East Asian Paintings: Some Considerations of Microscopic Structure and Mechanical Failure Modes.” Reprinted from: Proceedings of the Third ISCRCP, Conservation of Far Eastern Art Objects (1980).

Gerhard Banik and Johann Ponahlo, “Some Aspects Concerning Degradation Phenomena of Paper Caused by Green Copper Containing Pigments.” Also, four authors (these same two and others): “The Destruction of Paper by Green Copper Pigments Demonstrated by a Sample of Chinese Wallpaper.”

John Winter, “’Lead White’ in Japanese Paintings” (in: Studies in Conservation 26, 1981.)

Yoshimichi Emoto, “Deterioration of a Japanese Wooden Panel Painting—An Approach to Study of Malachite Staining, ‘Rokushô-yake.’” International Symposium on the Conservation and Restoration of Cultural Property—Conservation of Far Eastern Art Objects. Nov. 26-29, 1979, Tokyo.

• < Prev
• Next >