Modern observers tend to simplify the complex process of textual transmission and imagine that in a manuscript culture texts were handed down by scribes copying manuscripts in a long line of succession extending for generations. It is less commonly recognized, however, that manuscript copies were also routinely made from non-handwritten material, such as printed works or stone inscriptions. This paper looks at dated copies of stele inscriptions among the Dunhuang manuscripts, in an attempt to demonstrate the inherent difficulties in dating and establishing provenance for such copies. One of the main questions is whether the date in the colophon refers to the time when the text was carved into stone or the moment of creating the manuscript copy. The analysis reveals that there is no automatic answer to this problem, and the decision has to be made on a case-by-case basis. An additional lesson is that in many cases manuscripts are composite objects the components of which had a history of their own. The panels comprising a typical scroll often came from different locations and were written decades or more apart. It is through analyzing the interrelation of the texts and panels that we begin to uncover the complex process of the manuscript’s creation and the different layers of time and locations.