This dissertation presents an analysis of the diachronic development and the licensing conditions of finite auxiliary ellipsis in Early Modern German, in particular the so-called afinite construction.

The historical development of the phenomenon has rather peculiar properties as compared to other known processes of language change. Usually, such processes go to completion after their initiation and a period of transition. EMG auxiliary ellipsis, however, disappears from language use before it could entirely replace constructions with auxiliaries in the relevant contexts. This curious behaviour is explained by a combination of formal and functional factors. Certain parametric changes in the formal marking of embedded clauses are argued to be crucial input conditions for the rise of the afinite construction. These changes concern the C-system as well as the placement of verbs in embedded clauses. It is argued that the afinite construction develops as a means of marking clausal dependency. The spread and demise of the construction on the other hand are attributed to stylistic factors, having to do with shifts in text complexity.

Also the discussion of the licensing conditions combines formal and functional considerations. As the ellipsis of the finite auxiliary never fully replaces constructions with overt auxiliaries, it must have been optional even at the time when it was most 'established'. Therefore, any account of its licensing conditions also has to take this optionality into consideration. It is argued that the ellipsis develops as a marker of subordination, expressing the lack of assertion.