A Penn-style Treebank of Middle Low German

Proceedings of the 12th Conference on Language Resources and Evaluation (LREC 2020), pages 766–775 Marseille, 11–16 May 2020 c⃝ European Language Resources Association (ELRA), licensed under CC-BY-NC 766 A Penn-style Treebank of Middle Low German Hannah Booth, Anne Breitbarth, Aaron Ecay, Melissa Farasyn Ghent University Blandijnberg 2, B-9000 Ghent, Belgium {hannah.booth, anne.breitbarth, melissa.farasyn}@ugent.be, aaronecay@gmail.com Abstract We outline the issues and decisions involved in creating a Penn-style treebank of Middle Low German (MLG, 1200-1650), which will form part of the Corpus of Historical Low German (CHLG). The attestation for MLG is rich, but the syntax of the language remains relatively understudied. The development of a syntactically annotated corpus for the language will facilitate future studies with a strong empirical basis, building on recent work which indicates that, syntactically, MLG occupies a position in its own right within West Germanic. In this paper, we describe the background for the corpus and the process by which texts were selected to be included. In particular, we focus on the decisions involved in the syntactic annotation of the corpus, specifically, the practical and linguistic reasons for adopting the Penn annotation scheme, the stages of the annotation process itself, and how we have adapted the Penn scheme for syntactic features specific to MLG. We also discuss the issue of data uncertainty, which is a major issue when building a corpus of an under-researched language stage like MLG, and some novel ways in which we capture this uncertainty in the annotation. Keywords: annotation, historical treebank, Low German, parsed corpus, uncertainty 1. Introduction The development of diachronic parsed corpora in recent decades has played a crucial role in facilitating quantita- tive studies of syntactic change and reproducible findings which have a strong empirical basis; see e.g. Pintzuk et al. (2017) for an overview. Historical parsed corpora now ex- ist for a wide range of languages, including English (Taylor et al., 2003; Kroch and Taylor, 2000; Kroch et al., 2004), Icelandic (Wallenberg et al., 2011), French (Martineau et al., 2010) and Portuguese (Galves et al., 2017). One such resource currently under development is the Corpus of His- torical Low German (‘CHLG’), with support from the Her- cules Foundation/FWO.1 The CHLG is planned as a di- achronic parsed corpus spanning Old Low German/Old Saxon (OLG, c.800-1050) and Middle Low German (MLG, c.1250-1600).2 The OLG component is already available as a self-contained corpus, the HeliPaD (Walkden, 2015). The HeliPaD comprises 46,067 words from a single (the largest) OLG text, the Heliand, which is annotated according to the Penn standard for historical English (Santorini, 2010); for more details, see Walkden (2016). In this paper, we report on the second part of the CHLG, the MLG component, which is currently under development, and will be searchable online at www.chlg.ugent.be. The attestation for MLG is rich, but the syntax of the lan- guage remains relatively understudied, in part due to the extant texts not being readily accessible, nor presented in a way to facilitate corpus-based syntactic studies. Contrary to earlier assumptions that MLG does …

Content truncated to 500 words