last month
next month

Centre for the Study of Manuscript Cultures (CSMC)

06/2013 manuscript  of the month

Mantras, Mercury and Manuscripts

Earlier this year, a team of CSMC researchers from the humanities and the natural sciences stayed in Kathmandu for three weeks. The aim of this trip was to examine manuscripts stored there for the first time by means of state-of-the-art technical devices. Among the items examined in this way was, among others, a palm-leaf manuscript containing the esoteric Buddhist text “Commentary on the Tantra of the Deity Trisamayarāja” (Trisamayarājaṭīkā). This work is written in Sanskrit and is only extant in the form of this single manuscript. The material analysis yielded remarkable results: To begin with, it turned out that the manuscript contains two elements – i.e. arsenic (on the leaves) and mercury (in the ink), which point to the use of extraordinarily poisonous substances. Moreover, by means of multispectral imaging textual elements became legible which had been undecipherable to the naked eye. But that’s not all. In one place, we even discovered text which had been completely invisible before. Is it possible that the features mentioned above are in some way related to the fact that the manuscript contains a text which belongs to the esoteric form of Buddhism? In other words, have measures been taken by the producers of the manuscript to protect it from unauthorized use?

In most regions of South and South-East Asia, palm-leaves were the preferred writing-support; only in the second millennium of the common era paper gradually took their place - at least in North India and Nepal. The oblong format of the present manuscript (dimensions: 57 x 6 cm) is quite typical for the older material. The leaves are covered with writing in carbon ink on both sides throughout. Each page contains seven lines which are running parallel to the long edges of the leaves.

illustration no. 1: Back-side of folio 5 of the Trisamayarājaṭīkā manuscript (NAK 5-20). > Enlarge

The manuscript is kept in the National Archives of Nepal in Kathmandu. However, Nepal is not the place where the manuscript was produced. Rather, it originates from the famous Buddhist monastic university Vikramaśīla, which was situated in North-Eastern India near the Ganges. This provenance is not explicitly indicated on the manuscript but can be deduced almost certainly from the fact that a “scholar-monk” (paṇḍitabhikṣu) called Jinaśrīmitra is mentioned as the person who ordered the production of the manuscript. In another manuscript, which is quite similar in many regards, a reference to the same scholar is accompanied by the indication that the place of production was Vikramaśīla.

This monastery was known as a stronghold of Tantric Buddhism. It was destroyed at about 1200 of the common era. Judging from the script, it is very likely that the manuscript was produced in the 12th century. Obviously, it is still extant because it was brought to Nepal before or during the destruction of the monastery.

Certainly, the manuscript originally contained a complete copy of the text mentioned above. However, nowadays some of the leaves are missing, whereas 20 leaves from other manuscripts of which more parts are preserved in different locations were added erroneously. The fourth leaf of the text under consideration here even changed place with the corresponding leaf of another manuscript which was found in Tibet in modern times. How can this disorder be explained?

illustration no. 2: Marginal note on the front-side of folio 9;
top: as it appears on a conventional colour photograph;
bottom: the same part of the leaf as it appears on one of
the multi-spectral photographs. > Enlarge

It was customary to punch one or two holes in the leaves before writing on them (see illustration no. 1) in order that they could be held together by strings. Nowadays, however, the strings are very often missing - which is also true for the present case -, be it because they got lost in the course of time or because they were not attached right from the start. At any rate, the leaves were also piled up, placed between two wooden boards and wrapped in a large piece of cloth; finally, the whole bundle was tied up. In the same way the manuscript under consideration is packaged; only the book covers got lost for unknown reasons. If one works with more than one manuscript at a time, the cloth wrappings must be removed, so that the loose leaves can get mixed up. As a matter of course, this can happen in particular if format, lay-out and the scripts used are comparable or even identical - and this is precisely what can be observed in the present case. Moreover, the subject-matter of the texts preserved on the leaves is similar as well, that is to say, they all deal with Buddhist Tantric topics. Finally, it should be noted that one of the manuscripts inserted had been commissioned by the same learned monk of the monastery Vikramaśīla who had also commissioned our manuscript. It is probably not very far-fetched that at least the place of production of the remaining manuscripts is identical as well. However, in these other manuscripts neither the persons involved nor any places are mentioned; therefore, this conclusion must remain, for the time being, a mere working hypothesis.

illustration no. 3: Marginal note on the front-
side of folio 8; one conventional colour
photograph, followed by two images with
the text which had been invisible before.
> Enlarge

At any rate, all these different manuscripts of esoteric texts also have in common that they contain arsenic and mercury. This might corroborate the hypothesis mentioned above. Moreover, this raises the above question once again: Is there a connection between the use of poisonous substances, invisible text and the esoteric contents of the texts transmitted by these manuscripts?

The material examination revealed that, obviously, the leaves had been intentionally brushed with arsenical substances before the writing process started. Orpiment is responsible for the yellowish colour of the leaves (see illustration no. 1). However, treating the traditional Indian writing-supports or cloth wrappings with arsenical substances was, as can be deduced from many textual sources, a wide-spread phenomenon in the Indian manuscript culture, and this is definitely not confined to manuscripts with magical or esoteric contents. Generally, this treatment of the materials is meant to protect the leaves from ravenous insects. This precaution is easily comprehensible, if one considers the fact that in large parts of the Indian sub-continent the climatic conditions are favourable to the reproduction of insects.

It is true that the admixture of mercury to the ink is mentioned every now and then in the primary and secondary literature as well; however, it was apparently not as common as the use of arsenic - accordingly, it is also missing in the late marginal remarks of our manuscripts which were certainly only added after the manuscripts had been transferred to Nepal. The reasons for the use of mercury in the ink still awaits further investigation. However, for many reasons it is highly unlikely that the mercury (and the arsenic) served the purpose to protect the manuscripts from unauthorized use. To begin with, if that was the case one would have to clarify how it was possible to ensure that only unauthorized users were exposed to the poisonous substances. Furthermore, the dose of poison contained in the leaves and the ink is hardly sufficient to pose a serious threat to the health of human beings - at least if the manuscripts were used in an ordinary way, that is to say, if they were simply read.

illustration no. 3: Multi-spectral imaging camera of
CSMC at the National Archive. > Enlarge

The initial assumption also seems to be rather unlikely with regard to the illegible and invisible text. This phenomenon only occurs in a small amount of text. Predominantly, it only affects additions in the margins of our manuscript, and at present there seems to be no indication that these were notes of a secret nature . This is particularly obvious with regard to the marginal remark in illustration no. 2: The note consists of a mere identification of the text title which, judging from the script used, is rather recent. Moreover, the identification of the text is wrong, and exactly this might be the reason why there had been attempts to remove this note. The text shown in illustration no. 2, which is completely invisible to the naked eye, is very hard to decipher even after the use of our sophisticated technical device. However, it seems to be an ordinary marginal correction of a scribal mistake in line 3. It seems likely that later on this marginal note was erased as well. If this is true, then it might be the result of a scribe’s decision to correct the text in the line rather than in the margin so that the original marginal correction became redundant.

It is true that the initial assumption turned out to be unfounded in the ultimate analysis. Nevertheless, the scientific examinations of manuscripts in Kathmandu have been pathbreaking. The combination of methods from the humanities and the sciences is particularly helpful if one is interested in manuscripts as material products of the Indian culture.

BENDALL, Cecil (1883): Catalogue of the Buddhist Sanskrit Manuscripts in the University Library, Cambridge: With Introductory Notices and Illustrations of the Palaeography and Chronology of Nepal and Bengal. Cambridge: University Press.
LOSTY, Jeremiah P. (1982): The Art of the Book in India. London: The British Library.

Ms. NAK 5-20.
34 palm-leaves altogether (among which 14 contain text from the “Commentary on the Tantra of the Deity Trisamayarāja”), covered throughout with text in seven lines on both sides; dimensions: 57 × 6 cm.
Probable date of production: 12th century.
Provenance: The monastery Vikramaśīla in Bihar, India (this place of origin is, however, only certain in the case of 15 leaves).

Text by Martin Delhey
© for Illustrations: 1 - National Archives, Kathmandu / 2,3 and 4 - CSMC