Linguistische Berichte Heft 285

The use of commas in German is an orthographic area in which deviations from the official standard and variants are frequently observed. They correspond to a pronounced uncertainty among human writers, which seems mostly to be due to the limited applicability of static grammatical knowledge. Since the application of grammatical knowledge is largely required to look up comma rules and put them into practice in an individual text, uncertain writers often lack help with their comma questions. For this problem, Artificial Intelligence (AI) seems to be a solution, as numerous freely available AI-based applications on the internet promise to correct punctuation in free written texts. This article presents an empirical study with 10 open-source AI (correction) tools: On the basis of a text with 386 comma positions it examines the extent to which these AI tools deliver on the promise of standard-compliant comma placement. By systemizing the observable deviations from the official standard, it is argued that the inves-tigated tools adhere – as well as humans do – to lexical text criteria when placing commas. This lexical view on comma placement helps explain why AI outperforms humans in some comma domains and underperforms in others.

This study proposes a novel theoretical interpretation of the so-called Dehnungs- in Ger-man orthography. In contrast to the traditional view that the grapheme merely marks vowel length, the analysis suggests that serves a structural purpose: it signals the non-syllabic realization of the following sonorant. This function accounts for the restriction that only one sonorant may follow and explains why sequences such as are inadmissible within monomorphemic words. The study argues that functions as a grapho-tactic boundary, blocking morphological extensions and preventing homonym formation be-tween monosyllabic and morphologically complex bisyllabic forms. Only when and the sonorant belong to different morphemes is a sequence like permissible. Thus, the study reveals a previously overlooked function of , aligning graphematic, phonologi-cal, and morphological constraints within a unified structural framework.

The present study explores the use of commas by students who did not receive explicit orthog-raphy instructions during their studies, comparing handwriting and keyboard typing as writing mediums. Twenty-two mathematics students participated in the study and were presented with two texts containing similar comma-sensitive passages via a dictation app. They then had to either handwrite or type these texts. The qualitative and descriptive statistical results show minimal differences regarding the writing medium: in the handwritten part, there are only more extreme variations in the correct placement of commas, both upwards and downwards. The evaluation of comma categories and type of comma-sensitive structures provided particu-larly interesting findings: in both writing conditions – handwriting and keyboard typing – prepositional groups were strikingly often separated by commas, seemingly for emphasis or as insertions. Syntactic Vorfeld structures were irregularly comma-sensitive; in some cases, espe-cially infinitive groups were not correctly separated by commas.