Today, variation is taken seriously in almost all linguistic sub-disciplines and is studied as a phenomenon in its own right; this only applies to a limited extent to graphemics. There is hardly any relevant work available, especially for written Standard German. Graphematic variation is also particularly important for practical reasons, because it should be fundamental to the formulation of the standard. At the end of project, graphematic variation will be systematically analysed using a corpus of A-level papers from the last 90 years. This type of text has three advantages: Firstly, the writers are maximally educated, finished writers who - at least institutionally - are unlikely to receive any more orthographic training. Secondly, the influence of aids can be controlled in this text type. In contrast, texts that are written on a computer are potentially corrected by word processing programmes, sometimes even automatically; texts that are published are potentially edited by final or copy editors or proofreaders. Thirdly, Abitur exams have been written under very similar conditions for decades. A total of 1,397 Abitur papers have already been compiled and are now being digitally recorded, edited and morphologically and syntactically annotated.

The project has three central goals:

• Firstly, the synchronous graphemic variation is to be recorded. The result is an inventory of the areas in which variation can be observed and to what extent.
• Secondly, the diachronic development of this variation is to be described.
• Thirdly, a corpus is to be created with which these goals can be worked on and which can be used in a variety of ways. A secondary aim is to analyse the relationship between variation in usage and the respective valid standard.