Abstract (EN)

To exhibit the differences between different versions of the same text is quite an easy operation for a computer program if it concerns texts written in any modern European language such as English or French, where words are well separated by spaces. On the contrary, if it concerns a language such as Sanskrit, where traditionally no spaces occur between words, and some very long compound words exist, it becomes a difficult problem, especially if there is no valid and reliable computer oriented lexicon available. In this article we intend to show how we proceed to determine, in terms of words, the differences that can exist between the different versions of the same Sanskrit text, with the aim to elaborate a critical edition. Exhibiting these differences, it becomes possible to elaborate a distance between the different versions of the text. Once this step is achieved, the study and determination of relations, which can exist between the versions, becomes possible through the use of phylogenic trees.