The book deals with the concept of fragmentation as applied to languages and their documentation. It focuses in particular on the theoretical and methodological consequences of such a fragmentation for the linguistic analysis and interpretation of texts and, hence, for the reconstruction of languages. Furthermore, by adopting an innovative perspective, the book aims to test the application of the concept of fragmentation to languages which are not commonly included in the categories of 'Corpussprache', 'Truemmersprache', and 'Restsprache'. This is the case with diachronic or diatopic varieties - of even well-known languages - which are only attested through a limited corpus of texts as well as with endangered languages. In this latter case, not only is the documentation fragmented, but the very linguistic competence of the speakers, due to the reduction of contexts of language use, interference phenomena with majority languages, and consequent presence of semi-speakers.