The interplay between genre variation and syntax in a historical Low German corpus

Abstract

In this chapter, we focus on the choice of different genres in the Middle Low German part of the tagged and parsed Corpus of Historical Low German and its implications for syntax. We discuss how the inclusion or exclusion of genres has an impact on the study and the discovery of syntactic phenomena in Middle Low German, such as null referential subjects, resumptive pronouns, relative particles and gaps in coordinations. This interplay between genre and syntax also influences parsing decisions. Furthermore, we look at the influence of (sparse) punctuation on (automatically) tagging the corpus itself, and how a closer study of genrespecific syntactic elements contributes to the improvement of the accuracy of automatic classifiers.

Crossing the Borders between Meter, Syntax and Information Structure. Some Methodological Notes

“…Latest investigations have moreover shown that the genre of a text plays an important role in defining language change (Whitt 2018). For instance, the investigation by Farasyn et al (2018) shows that the inclusion or exclusion of genre as a variable yields different results for the study of different language change phenomena in Low German. Fuß and Hinterhölzl (2023) moreover show that the rise of German expletive es is first witnessed in legal texts.…”

Towards a Historical Treebank of Middle and Early Modern Welsh, Part I: Workflow and POS Tagging

“…A series of corpora based around the various Penn parsed corpora of historical English, such as the Penn-Helsinki Parsed Corpus of Middle English (Kroch and Taylor 2000), have adopted a broadly common framework, providing a parallel format for part-of-speech information, and adopting a constituency-based system for representing syntactic structure. Examples include the French MCVF (Modéliser le changement: Les voies du français) corpus (Martineau 2010), the Portuguese Tycho Brahe corpus (Galves, de Andrade and Faria 2017), the Icelandic IcePaHC corpus (Wallenberg et al 2011), the Irish POMIC corpus (Lash 2014), and the ongoing Corpus of Historical Low German (Koleva et al 2017, Farasyn et al 2018.…”