Initially developed by scholars familiar with European and American manuscripts, the TEI is now being pressed into service for the encoding of non-western texts. Will the TEI develop as a closed fraternity of western computer programmers and editor/scholars, or will it emerge as an open, egalitarian community of global scholars, who are interested in applying digital technology to the world’s literature? On the one hand, there are practical issues faced by encoders of non-western texts, such as the lack of reliable Unicode for all language scripts and the economic difficulties associated with training non-western scholar/editors in TEI. On the other hand, there are hermeneutic issues raised by TEI going global which are prompted by the general understanding of texts as information rather than as the materially embodied corpus of a culture. Some of the specific issues raised by Tibetan texts in this regard will be used to examine the theoretical tailoring of TEI without presuming that non-western texts will wear TEI comfortably.
Tibetan texts present several challenges to the TEI. Among the practical issues are these: as an endangered language, Tibetan has no government support for its preservation or for its study by the international community of digitally trained scholars. As a minority language, Tibetan waits on a Unicode version guaranteed to represent all of its scripts. Despite these practical problems, because the Tibetan cultural heritage is text-based, it is an obvious candidate for TEI. Even if TEI cannot restore Tibetan texts to their original cultural function, the preservation and encoding of Tibetan manuscripts and transcripts can at least keep one of the most distinctive Asian cultural traditions on 'life support'.
There are, however, other problems raised by the modern encoding of traditional Tibetan texts—problems that are hermeneutic in nature, pertaining to our underlying understanding of what a text is and what needs to be encoded in a text. Informatics, as used by Haraway and Hayles, is a theoretical and operational paradigm for treating a text as abstract information that can be taken out of its material medium and its cultural context. The TEI aims to describe a text so that it can be searched, stored and circulated digitally without a loss of its most meaningful features, and in doing so TEI favors the information in the text that can abandon its material medium and perhaps even its cultural role. But information—i.e., 'informing readers'—is not what many Tibetan texts were meant to provide; the texts were meant to transform readers, helping them become enlightened. It is not just that Tibetan texts are sacred; they are practice texts that demand specific kinds of mental preparation, specific kinds of bodily handling and gestures linked to their materiality, as well as ongoing religious commitment from the readers.
With regard to readers' mental preparation, traditional Tibetan libraries contain many 'restricted' texts, which are not given wide circulation because of the original esoteric meaning of these texts. Certain advanced texts were 'classified', so to speak; they were meant only for meditator/scholars who met certain conditions: they had achieved the right qualifications for understanding the meaning of the texts, had received permission from their teacher to study or use the texts, and had made a commitment to practice the meditation methods described in the texts. These advanced Vajrayana texts are, to this day, not made accessible by traditional Tibetan teachers to any reader for the asking. (In some sense, this problem of Tibetan esotericism is similar to the esotericism of the TEI itself: the TEI tags are a hidden code, developed by, and meant for, a fairly limited group of advanced practitioners of XML.) The TEI, which is a digital publishing tool as much as it is a tool for textual analysis, may make it easier to bring formerly restricted Tibetan texts into public circulation. Although encoding these texts would help Tibetan scholars and advanced practitioners analyze the texts, encoding also makes it more likely that unqualified readers will 'break into' some of the advanced, restricted Tibetan texts. With Tibetan culture in danger, will its textual resources be pillaged by unqualified, spiritual treasure-seekers, and will TEI become a tool for such digital plunder?
A second issue raised by traditional Tibetan texts concerns the ritualistic manner of their reading. Because many of these texts are practice texts, which lay out the particular steps that a meditator would follow in his or her daily meditation practice, the traditional method for reading these practice texts is a training style. It is unlike the reading style used for most western texts, because it requires precision of pronouncing each word (at least in one’s head), accurate counting of the chants requiring repetition, and accompanying visualizations of a detailed nature. Within the Tibetan tradition, many manuscripts are not meant to be read through once from start to finish, but are meant to be scripts for daily meditation practices, involving ritualized bodily and mental gymnastics.
To encode a Tibetan practice text as though it is to be read straight through, in the way that most western texts are read, would neglect its most important function, which is to train the reader in meditation. The solution is not to trivialize Tibetan meditation practices by encoding loops for repeated chants or by encoding inserted graphical images for visualizations, for this would not respect accomplishments expected of the reader/practitioner. It is the reader/practitioners who must contribute the repetition of chants or the visualization of a meditation deity to their reading of the practice text, and a software program that supplied these would sabotage the training that the text instantiates.
Finally, a third issue concerns whether the TEI codes can distinguish between different audiences for the encoded manuscript: can TEI tags be designed to discriminate between a reader who is practicing meditation with the text and a reader who is reading the text for standard western research purposes? A reader who has no interest in meditating or in Tibetan beliefs may read the text as a work of literature, philosophy or history; this reader would benefit from TEI codes that mark the structure and bibliographic details of the text. The TEI tags would allow this first kind of reader to read the text 'western style'. But a second kind of reader, who is actively engaged in using the text in the traditional Tibetan way as a meditation practice text, may be looking for TEI encoding that will map out the most important clues for how to succeed at the meditation practice. Is this a reasonable project for TEI encoding?
The TEI’s application to multilingual, multicultural projects is not a simple, uniform expansion but is a hermeneutic exercise in acknowledging and facing its horizons. Textual practices that are unproblematic within one’s own culture may be problematic in another culture, because the farther one travels from one’s own culture, one finds that texts function in radically different ways. These multicultural horizons need not be limitations on the TEI, but they should be kept within the broad, panoramic view of what the TEI is trying to accomplish.