Mandoku – An Incubator for Premodern Chinese Texts – or How to Get the Text We Want: An Inquiry into the Ideal Workflow

Premodern Chinese texts pose problems that are difficult to accommodate with the current TEI text model, which bases the main hierarchy of a text on its structural content, rather than on a hierarchy that models the pages, lines and character positions. For the TEI, this is a sensible decision and has led to the abolishment of elements like and in the latest release of the Guidelines. For premodern Chinese texts however, especially texts that are transmitted as manuscripts or woodblock printings and have not yet seen a modern edition printed with movable type (let alone as, more recently, computerized typesetting), establishing the structural hierarchy of the text content is, together with the even more daunting question of establishing the proper characters of the text (on which see below), an important part of the research question that motivates the digitization of the text. Requiring an answer to this question before a proper electronic text can be created makes this intractable in the digital medium and glosses over an important leap of faith in the creation of a TEI encoded text. In this paper, I will try to trace some of the implications and propose an approach that allows different models of the text for different stages in the encoding process, thus closer modeling the process of the creation of an electronic text.