Friday, October 28, 2016

North American Silk Road Collections: From Germany to North America

In my previous IDP blog post, I talked about the inscription on the back of the Penn Museum’s fragmentary mural from Turfan. The inscription clearly indicates that it was once a part of the German collection, or to be exact, the Turfan collection originally housed in the Museum of Ethnology (Museum für Völkerkunde). How then did this and other similar pieces ended up in North American collections?

Many archival records of such pieces bear the name of Albert von Le Coq (1860-1930) as the source of the pieces. Albert von Le Coq participated in the second through the fourth German expeditions. Le Coq started his academic career in his forties as a volunteer researcher at the Museum of Ethnology and served as the director of the museum’s department of Indian art from 1923 to 1925. His presence in the archival records indicates that he was partially, if not entirely, responsible for the transfer of these German pieces to North America.

Albert von Le Coq and Mamasit Mirab in front of caves of the 'Eastern Main Group' including 'Cave 163' in Kizil, MIK B 1070 © Museum für Asiatische Kunst, Kunstsammlung Süd-, Südost- und Zentralasien, Staatliche Museen zu Berlin.

The issue of the dispersion of the German pieces, especially of mural fragments, has been recognized for a long time, and several scholars have already discussed it from the perspectives of both art history and collection history (see works of Lee, Morita, Schlingloff, Ueno, and Zhao in the references). Among them, a recent study by Professor Sonya Lee of the University of Southern California especially details the process through which the pieces in question were brought from Germany to North America.

Around fifty Central Asian mural fragments in the United States are confirmed to have been removed from Germany through their sale in the 1920s. This was a difficult decision for Le Coq, who was planning an exhibition of the Turfan collection in the time of a depreciating German mark and concomitant inflationary pressures. The pieces, selected mainly based on their relative dispensability when compared to those in better conditions remaining in the Museum, were sold through the hands of various dealers. These included Edgar Worch, who was an agent from the firm called Ludwig Glenk in Berlin, and Abel William Bahr, a collector and dealer of Chinese art in North America. (Lee 2015, 11-12)

The same study by Professor Lee also sheds light on the shift of function and meanings of such Central Asian pieces that resulted from these sales and dispersions. Moving into the collections of people with different agendas, some of the German pieces were transformed into art objects serving the goals of their new owners. In the Museum of Ethnology, these Central Asian pieces were exhibited in accord with the museum’s educational agenda for the general public and Le Coq’s intention to tell the narrative of the classical antiquity’s eastern genealogy.

Some of the German pieces ended up in private collections, while many pieces were purchased by North American museums through dealers after the pieces had left Germany. Sixteen pieces of Kizil mural paintings were purchased and later donated to what is now the Smithsonian American Art Museum by John Gellatly (1852–1931). As he was primarily known for his collection of American Art, the Kizil pieces were appreciated aesthetically as art works and exhibited without any indication of their Central Asian contexts. (Lee 2015, 5-11; 12-14)

In other cases, the new owners sometimes appreciated Central Asian pieces as expressing a fundamental unity in artistic or religious works across cultures and times. This was the case for the French author, art historian, and statesman, André Malraux (1901-1976). Malraux saw a striking similarity between Buddhist heads from Afghanistan and the sculpted heads of Notre-Dame de Reims of the French Gothic which led him to promote his Afghan Buddhist heads as the 'Gothic-Buddhist' works, embodying the sentiment found in the French Gothic art works (Levine 2012). Interestingly, according to Ernst Waldschmidt, Malraux's pieces possibly originated from the same source that supplied pieces to the Berlin Museum of Ethnology (Waldschmidt 1932, 3; Levine 2012, 637-638).

Cropped from Buddhist visual narratives of cave shrines and reframed as independent pieces, many buddhas, bodhisattvas, and celestial beings with their serene expressions were appreciated in new ways in new contexts.


Levine, Gregory P. A. 2011. “Malraux’s Buddha Heads.” In A companion to Asian Art and Architecture, edited by Rebecca M. Brown and Deborah S. Hutton, 629-646. Chichester, West Sussex, UK: Wiley-Blackwell.

Lee, Sonya S. 2015. “Central Asia Coming to the Museum: The Display of Kucha Mural Fragments in Interwar Germany and the United States.” Journal of the History of Collections. Advanced Access published October 17, 2015. Accessed February 4, 2016. doi: 10.1093/jhc/fhv031.

Morita, Miki. 2015. “The Kizil Paintings in the Metropolitan Museum.” The Metropolitan Museum Journal 50: 115-135.

Schlingloff, Dieter. 2011. Albert von Le Coq und die Wandmalereien von Kizil (Addendum zu der Denkschrift: T III MQR, Eine ostturkistanische Klosterbibliothek und ihr Schicksal). Leipzig: private print.

Ueno Aki 上野アキ. 1978. “Kijiru nihonjin dō no hekiga: Ru・Kokku shūshū saiiki hekiga chōsa, 1” キジル日本人洞の壁画: ル・コック収集西域壁画調査 1 [Mural paintings from Japaner Höhle in Kizil: Research on the mural paintings from the Western Regions collected by Le Coq, 1]. Bijutsu kenkyū 美術研究 [Journal of Art Studies] 308 (October):113–20.

― 1980a. “Kijiru dai san ku maya dō hekiga seppō zu - jō: Ru・Kokku shūshū saiiki hekiga chōsa, 2” キジル第三区マヤ洞壁画説法図—上: ル・コック収集西域壁画調査 2 [Mural paintings of preaching scenes in Māyāhöhle, 3. Anlage, first part: Research on the mural paintings from the Western Regions collected by Le Coq, 2]. Bijutsu kenkyū 美術研究 [Journal of Art Studies] 312 (February):48–61.

― 1980b. “Kijiru dai san ku maya dō hekiga seppō zu - jō (zoku): Ru・Kokku shūshū saiiki hekiga chōsa, 2” キジル第三区マヤ洞壁画説法図—上 (続): ル・コック収集西域壁画調査 2 [Mural paintings of preaching scenes in Māyāhöhle, 3. Anlage, first part (sequel): Research on the mural paintings from the Western Regions collected by Le Coq, 2]. Bijutsu kenkyū 美術研究 [Journal of Art Studies] 313 (March):91–97.

Waldschmidt, Ernst. 1932. “Die Stuckplastik der Gandhära-Schule (Zu Einigen Neuerwerbungen des Museums Für Völkerkunde).” Berliner Museen 53 (1):1-9.

Zhao Li 赵莉. 2009. “Kezi’er shiku bufen liushi bihua yuanwei kaozheng yu fuyuan” 克孜尔石窟部分流失壁画原位考证与复原 [Historical retrospect on mural outflow in Kizil Grottoes and restoration in its original site]. Zhongguo wenhua yichan 中国文化遗产 [China Cultural Heritage] 2009(3):88–99.

Note: I would like to express my sincere gratitude to Dr. Michelle C. Wang for her insightful suggestion for this blog post.

Friday, September 30, 2016

North American Silk Road Collections: In Search of Provenance

One of my tasks as a research fellow for the Georgetown-IDP project for the North American Silk Road Collections is to locate the pieces’ original locations, including their placement within the archaeological sites. Such information can be derived from various sources, such as stylistic analyses, the materials used, published archaeological reports, and archival records. And sometimes, such provenance information has been inscribed on the pieces by the archaeologist.

Reverse of Object C412. Photographer: Miki Morita

The picture above shows the back of one of the fragmentary murals which I introduced in my last blog post (Penn Museum, C412). The inscription incised directly into the stucco plaster base reads as follows:

III Reise M. Ŏ. M.
Hŏhle I
im Schutt gefunden.

This simple inscription carries much information about this piece’s provenance. First, it is written in German, and we know from colleagues in the German collections that the German expeditions used various abbreviations to denote the provenance. So the first line, 'III Reise' (the third tour), indicates that this fragment was obtained during the third of the four German expeditions held between 1902 and 1914. The third expedition (1905–1907) covered sites around the areas of Kucha and Turfan.

The rest of the inscription gives information on where the fragment was discovered. This part of the inscription is usually straightforward, but sometimes confusing due to abbreviations and errors. In the case of this fragment at the Penn Museum, Professor Adam Smith of the University of Pennsylvania, Stephen Lang of the Penn Museum, and I worked together to reconstruct and translate this inscription, also consulting colleagues at IDP Germany.

The first two letters 'M. Ŏ.' are most probably an abbreviation for 'Ming-Öi', namely 'thousand houses'. This was a general term often used by the locals for Buddhist cave temples. However, it can be confusing. Aurel Stein, for example, used 'Ming-Öi' to refer to the Buddhist cave temples near Shikchin. In the case of some other fragments from the Kizil caves, these characters are followed by 'Q', indicating the transcription used by the Germans for Kizil, namely 'Qyzil'. Accordingly, the third 'M' should indicate a cave temple site visited in the third German expedition. In fact 'M' is most commonly used by the German expeditions to refer to 'Murtuk' (or Murtuq as transcribed by the Germans), and this is also reinforced by the style of the bodhisattva.

Although there are cave temples known as Murtuk, another cave temple site nearby was also included under this designation, namely that of Bezeklik. It was visited during the third German expedition, and 'Höhle I' (Cave 1) of the German numbering of the caves corresponds to Cave 9 in the current numbering. In Altbuddhistische Kultstätten in Chinesisch-Turkistan by Albert Grünwedel, the archaeological report for the third German expedition, we are able to find a record of the rear wall of 'Höhle I', which is filled with rows of praying bodhisattvas (Grünwedel 1912: 231).

Wall painting of adoring bodhisattvas from “Höhle I” (Asian Art Museum, National Museums in Berlin, MIK IB 8492 [lost during the Second World War]; retrieved from the IDP database [Le Coq 1926: 23, pl. 23; Dreyer, Sander, and Weis 2002: 152])

Fortunately, parts of these bodhisattva paintings on the back wall and also on the sides of niches of the corridor remain in-situ in Cave 9 (Höhle I), and they show great similarities to the bodhisattva on the Penn Museum’s piece. Therefore, we concluded that the reconstruction and translation of the inscription should be '3rd expedition, Ming-Öi, Murtuk; Cave 1; Found in the rubble', and this piece most probably originates from Cave 9 (Höhle I) of the Bezeklik caves.

It is confirmed that several museums in North America hold fragments of similar bodhisattva heads. Although I have not seen their inscriptions, the stylistic features and archival information of some of the pieces suggest that they most likely belong to the same Bezeklik cave. While each piece consists of a small bodhisattva’s head, together they would complete the beautiful wall of adorning bodhisattvas.

The question now arises of how pieces from the German state-sponsored expeditions, most of which are now in museum and libraries in Germany (and Russia), found their way to North America. I will explore this in a future blog post.


Dreyer, Caren, Lore Sander, and Friederike Weis. 2002.Staatliche Museen zu Berlin, Dokumentation der Verluste, vol. 5. Berlin: Staatliche Museen, Preußischer Kulturbesitz.

Gabsch, Toralf. (ed.). 2012. Auf Grünwedels Spuren: Restaurierung und Forschung an zentralasiatischen Wandmalereien. Leipzig: Koehler & Amelang.

Grünwedel, Albert. 1912. Altbuddhistische Kultstätten in Chinesisch-Turkistan; Bericht über archäologische Arbeiten von 1906 bis 1907 bei Kuča, Qarašahr und in der oase Turfan. Berlin: G. Reimer.

Härtel, Herbert, and Marianne Yaldiz. 1982. Along the Ancient Silk Routes: Central Asian Art from the West Berlin State Museums: an exhibition lent by the Museum für Indische Kunst, Staatliche Museen Preussischer Kulturbesitz, Berlin, Federal Republic of Germany. New York: Metropolitan Museum of Art.

Le Coq, Albert von. 1926. Die buddhistische Spätantike in Mittelasien, vol. V Neue Bildwerke. Berlin:Reimer u. Vohsen.

The Kucha Academy and the Asian Art Museum, National Museums in Berlin, are currently also engaged in an an international project to locate the mural fragments from the Kucha region. Their aims include the reconstruction of the murals in selected caves, in which these inscriptions found on the back offer important clues. Thanks to colleagues in Berlin for their help with this work.

Wednesday, July 27, 2016

Cataloguing North American Collections

My name is Miki Morita, and I am a new postdoctoral fellow for a joint project between IDP and Georgetown University in Washington D.C.* In this role I will be collecting data and conducting research on Chinese Central Asian manuscripts, art works, and archaeological artefacts in North American collections for inclusion on IDP online.

Very few North American items have been studied extensively, and even fewer have been incorporated into the IDP database. When I was cataloguing some mural fragments from the Kizil Caves (Baicheng County, Xinjiang) in the collection of the Metropolitan Museum of Art, I came to notice the existence of less-known Chinese Central Asian works of art and archaeological artefacts, and felt a strong need to create a universal catalogue for scholarly purposes. IDP was, of course, aware of such Chinese Central Asian materials and had already been working with some of North American institutions. I feel very fortunate that I can take part in this project, which allows me to pursue my research interest in Chinese Central Asian pieces in North America.

The most thrilling part of this project is that we do not know what can be found in the North American collections. Some of the pieces have been recognised and studied in the past, such as a major collection of manuscripts in the Library of Congress and Kizil mural paintings at the Smithsonian Institution. On the other hand, there are pieces that have not yet attracted scholarly attention, such as the following mural fragment (C411) from the collection of the University of Pennsylvania Museum of Archeology and Anthropology.

Object C411. Courtesy of the Penn Museum.

This fragment, containing busts of three Buddhist deities in the Indo-Iranian style II of Kizil mural paintings, was displayed in one of the galleries, yet was not known and studied by art historians until recently. It turns out to be a part of a lunette of Cave 38 in the Kizil Caves in Xinjiang, and was originally collected during the fourth German expedition led by Albert von le Coq in the early twentieth century.

While the Cave 38 fragment of the Penn Museum is a relatively major piece with three figures, many Chinese Central Asian pieces in the North American collections could be small and fragmentary pieces. For example, the same museum also owns two small mural fragments (C412, C413B) that each depict the head of a Buddhist deity.

Object C412. Courtesy of the Penn Museum

Object C413B. Courtesy of the Penn Museum

Despite their size, such small pieces are part of a limited number of remaining Chinese Central Asian pieces and represent very important pieces in the effort to complete a picture of the history and culture of this region. Moreover, each piece comes with unique provenance information that collectively offers a perspective on the formation of the Chinese Central Asian collections in North America.

Fingers crossed that there remain many more pieces residing within the North American collections! I am very excited to see what new scholarly developments can be made based on the outcome of this cataloguing project.

*Thanks to the Henry Luce Foundation for funding this post, and to the Dunhuang Foundation US for funding the training visit of Dr Morita to IDP at the British Library.

Friday, June 17, 2016

A guide to orientating your tomb

Among the manuscripts found in the library cave of the Mogao Grottoes complex, near Dunhuang, there are a number of fascinating divination works. These include the scroll Or.8210/S.3877, recently conserved by colleagues Wong Wing-hui and Vania Assis (see related blog post).

Made of thin yellow paper and written in a rather rough hand, this manuscript was probably intended for personal, rather than more official, use. It includes extracts from different titles, as well as a lay society circular and contracts, some of which are dated from 897, 902 and 909. Its sketches are of particular interest for us as they illustrate a form of divination crucial in ancient China: one that focused on where best to build a tomb.

Geomancy, sometimes referred to as 'siting', dictated the positioning of both domestic and funerary structures, from palaces to graves. It led to practices often better-known nowadays under the term fengshui, literally translated as "wind-water" and thought to go back to the Song Dynasty. Its primary focus, nonetheless, remained on the deceased.

The front of scroll Or.8210/S.3877 is entirely filled by a drawing depicting four different topographical configurations. This appears to be a concrete example of how to determine the appropriate location for a funerary site. First is a group of mounds evoking hills. A patchy inscription, of which only the characters "大吉" (daji) survive, indicates a "very lucky, highly auspicious" spot.

Another caption at the centre of this mountainous formation states that a sepulture positioned there would bring unending riches and honour: "葬得此地,富貴不絕" (zang dei cidi, fugui bujue).

The other three landforms, though not as easily identifiable, are all named as mountain ridges: Baozi Gang (抱子崗), shown in the photograph below, Sangai Shangang (散盖山崗), and Xionglong Shangang (雄龍山崗).

Again, auspicious sites are designated by inscriptions: chu erqian dan 出二千石, lingzhang 令長, chu jiuqing xiang 出九卿相, chu fangbo 出方伯. Unfavourable sites for a sepulture are equally singled out by the character xiong 凶, meaning "ominous, inauspicious".

Two mysterious human figures are also represented. Who are they? Despite looking very similar, it seems that each of them is engaged in a different type of activity. One is bare foot, while the other is wearing boots. It is hard to know what the first one is doing because of the fragmentary nature of the document, but the second one seems to be holding something.

Could they be the geomancer? Or are they the individual who commissioned the document? My guess is as good as yours, so if you have any thoughts please let us know!

On the back of the same manuscript is an incomplete diagram of an auspicious familial gravesite, with several scribbled notes.

Three circles indicate the respective grave mounds of a grand-father, '祖父' (zufu) and of two of his descendants, both buried with their three children. The tombs are arranged across a square plot of land, which is delineated by an open-topped enclosure, probably the entrance, and marked in each corner by what could be watch-towers.

Such a distribution of the sepultures must have been fairly popular during that period, as scroll Or.8210/S.2263 - also in the Stein collection - possesses a very similar representation.

As demonstrated in this manuscript, extra care was thefore paid to the location and orientation of familial graves. Concern with divination as a device to assure a proper burial for one's parents already appeared in the Classic of Filial Pity 孝經 (Xiaojing), during the Western Han dynasty (206BCE-8CE): "[The filial son] determines the burial place [of his parents] by divination and puts them to rest." Over the following centuries, Chinese people increasingly started to believe that an auspicious burial site would also bring good fortune to succeeding generations, and geomancy came to be seen as a way of influencing the future.

Thursday, May 26, 2016

A Gilgit manuscript at the British Library

The Gilgit manuscripts, which were found in the village of Naupur in the 1930s (now in the Gilgit-Baltistan region of Pakistan), are one of the most finds of important Asian manuscripts. The cache was first discovered in 1931 by locals in an ancient ruin, which may have been the residence of a Buddhist monk. They are thought to be the remnants of a Buddhist library, dating from the 5th to 7th centuries AD.

The explorer Aurel Stein, who was passing through the area at the time the manuscripts were first discovered, reported the find in a newspaper article, and several excavations followed. The majority of the Gilgit manuscripts are now held the the National Archives in New Delhi and Shri Pratap Singh Museum in Srinagar (see this essay for more details). The British Library also has a small selection of the manuscripts.

In a letter, Stein wrote:

Meanwhile I have sent some well preserved leaves of two mss. which had been secured from the hands of villagers to Dr. Barnett at the British Museum as a temporary deposit. I have left it to him either to examine them himself or to pass them into competent hands. Kindly put yourself into touch with him, in case you thought it desirable to take up this limited task.

The two manuscripts mentioned by Stein are:

(1) Or.11878A: Eleven folios of a birchbark manuscript containing the major part of the Saṅgharakṣitāvadāna (Divyāvadāna XXIII), and a part of the monastic regulations of the Mulasarvāstivāda school of Buddhism.

(2) Or.11878B: Seven folios of a manuscript containing the Sanskrit text of the Lotus Sutra (Saddharmapuṇḍarīka).

While the vast majority of the Gilgit manuscripts are made from birch-bark, the pages containing the Lotus Sutra (pictured above) are made from paper. The white appearance of the paper is caused by the use of gypsum to 'size' the paper before it was written on. The manuscript had probably travelled west from one of the Buddhist kingdoms of the Silk Road, such as Kucha, where many manuscripts of this type have been found.


Shayne Clarke, Gilgit Manuscripts in the National Archives of India: Facsimile Edition. Volume I. Vinaya Texts. National Archives of India and IRIAB, Soka University, 2014.

Oskar von Hinuber, "The Gilgit Manuscripts: An Ancient Buddhist Library in Modern Research." In Paul Harrison and Jens-Uwe Hartmann (eds.), From Birch Bark to Digital Data: Recent Advances in Buddhist Manuscript Research, Verlag der Österreichischen Akademie der Wissenschaften, Vienna, 2013. 79-135.

Noriyuki KUDO, "Gilgit Saddharmapuṇḍarīkasūtra Manuscript in the British Library, Or.11878B–G." In Annual Report of The International Research Institute for Advanced Buddhology at Soka University 28 (2015), 197-213.

Friday, April 29, 2016

Publication: The Three Hares, A Curiosity Worth Regarding


Hardback, 368 pp., 326 illustrations
ISBN : 9780993103926
England: Skerryvore Productions Ltd, 2016
Price: £30.00
Order online here

From fifteenth-century rural churches in deepest Devon to sixth-century cave temples on the edge of the Gobi desert in China, this book follows its three authors on the tantalising trail of a mysterious medieval motif - three hares running in a circle sharing three ears which form a triangle at the centre of the design.

Along the way, a modern Devon myth is exposed, and the Three Hares in the sacred art of Christianity, Judaism and Buddhism are explored, and tentatively explained, before the trail leads into the Islamic world, and the great Mongol Empire.

Friday, April 22, 2016

Conserving a Chinese scroll

Vania Assis is Conservator of the Dunhuang scrolls at the British Library, and works on various projects supporting IDP's activities. Here is a post about one of her latest conservation jobs.

My colleague Wong Wing-hui and I recently worked on the Chinese scroll Or.8210/S.3877. Like other items in the Stein collection, it had been previously treated during its life as a collection item.

In the past, various materials were used to strengthen and repair manuscripts. In the case of our scroll, silk gauze was pasted on both sides with animal glue. There were, sometimes, several layers on top of each other. Heavy and thick paper was also applied to reinforce weak areas, such as edges, tears and missing areas.

Gauze covering the surface of scroll Or.8210/S.3877

As these materials aged, they became more unstable, causing the item to distort and transferring acidity to the paper. Higher acidity meant that the document became discoloured, which when combined with the texture of the gauze meant that it was difficult to perceive the original aspect of the scroll. In addition, a lower pH also made the item more brittle, making safe handling problematic.

Scroll Or.8210/S.3877 before conservation

Removing these materials proved very challenging: first, because they heavily adhered to the most vulnerable areas; second, because the paper used to make this scroll was particularly thin and transparent.

We worked on a section at a time, using hot water to reactivate the animal glue. We then removed the gauze with tweezers, carefully pulling it away from the paper. One of the most time-consuming processes was to remove the residual animal glue, which had been used in very large quantities. We did so by scraping it with a spatula, while it was damp. During this stage, we also removed old repairs, as they easily peeled away from the original material.

To repair the scroll's countless small tears and lacunae, we used Japanese paper, which is not only more sympathetic to the original paper, but also light weight and acid-free.

Detail of scroll Or.8210/S.3877 before conservation
Detail of scroll Or.8210/S.3877 after conservation

After all treatments, the scroll was lightly pressed for a week, to flatten any distortions. Finally, we rolled it onto an archival quality core support, and it is now ready to be digitised and handled!

Scroll Or.8210/S.3877 after conservation