Investigating connective use in L2 German: A corpus study

Abstract

Writing a cohesive text, that is, the ability to connect sentences, paragraphs, and ideas via the use of a range of grammatical and lexical devices, must be taught explicitly in foreign language (L2) writing classes because the preferred devices used to express cohesion differ between languages (Kunz et al. 2017). L2 writers struggle with cohesion since they tend to rely on native language (L1) strategies to create cohesive texts. This has been shown in studies on L2 English (e.g., Appel & Szeib 2020; Hinkel 2001; Johnson 2017; Stemmer 1991). To date, the most comprehensive monolingual study of cohesion is Halliday and Hasan’s (1976) Cohesion in English, which has served as a point of departure for other languages, including German. The authors list five categories of cohesion: (1) co-reference, (2) substitution, (3) ellipsis, (4) conjunction, and (5) lexical cohesion. While English cohesive devices have garnered considerable attention in second language acquisition (SLA) research (e.g., Das et al. 2017 (connectives); Tanskanen, 2006 (lexical cohesion)), less attention has been paid to the use of cohesive devices in German (e.g., Belz 2005; Strobl 2020; Walter 2007). Conjunction is the cohesive category that has received the most attention from scholars investigating L1 German. For example, Stede (2016) analysed connectives in the Potsdam Commentary Corpus, using a self-developed tool (ConAno) for semi-automated connective extraction and analysis. In addition, Walter (2016) investigated aspects of academic writing in the Korpus Akademisches Deutsch, including the distribution of subordinating conjunctions. A contrastive study, which also includes an intra-language comparison between genres and focuses on connectives, the overarching category of cohesive devices which includes conjunctions, performed by Kunz et al. (2021) revealed that German academic texts contain more temporal (e.g., bevor) and expansion (e.g., anhand) connectives compared with contingency (e.g., aufgrund) and comparison (e.g., dagegen) connectives. In stark contrast to the growing research interest in connectives in L1 German, research into cohesion in L2 German to date has been scarce, with a handful of available studies focusing on texts produced by writers with heterogeneous L1 backgrounds (e.g., Strobl 2020; Walter 2007). Given the impact of the L1 on L2 learners’ cohesion-building patterns, there is a dire need to investigate cohesion in texts produced by L2 German writers with a homogeneous L1 background.

The present study aims to close this research gap, by investigating cohesion in L2 German texts written by learners with L1 Dutch. The first category we will focus on is the conjunction, since it is the category that has received the most attention in research on cohesion in German to date. This will allow us to compare our results with previous studies, shedding light on L1-specific aspects of conjunction in learner writing. This analysis will be based on the Belgisches Deutschkorpus (Beldeko) (Strobl 2020), which has recently been built for this specific purpose. Beldeko consists of 301 summaries (70774 tokens) written by advanced learners of L2 German in an academic writing course. The corpus has been pre-processed and automatically annotated with PoS-tags and lemmata. Furthermore, connectives have been pre-annotated automatically according to guidelines based on PDTB3 (Webber et al. 2019) in combination with DimLex, a database containing German connectives and their corresponding PDTB3 tags (Scheffler & Stede 2016; Stede 2002).

A preliminary descriptive analysis of the automatically pre-annotated data via R shows higher use of temporal and expansion connectives compared with contingency and comparison connectives. This ties in with Kunz et al.’s (2021) results for L1 German. In another study, Konjevod (2012) stated that L2 German learners do not use concessive, conditional, and disjunctive connectives and, furthermore, restrict their use of additive connectives to und. In addition, Walter and Schmidt (2008) concluded that und is mostly used in sentence-initial position by learners. Concerning L2 English research, Martinez (2002) showed that L2 English learners use only a restricted set of connectives, disregarding others. The preliminary result, as well as the hypotheses from earlier research on connectives in learner language, will be further investigated with manual annotation, using the online annotation platform Inception (Klie et al. 2018). In conclusion, we will analyse whether (1) L2 German learners use connectives from all categories, (2) they use a restricted set of connectives per semantic category, (3) they restrict specific connectives to certain positions in sentences, and (4) they use more temporal and expansion connectives than contingency and comparison connectives.

References

Appel, R., & Szeib, A. (2018). Linking adverbials in L2 English academic writing: L1-related differences. System, 78, 115–129. https://doi.org/10.1016/j.system.2018.08.008

Belz, J. A. (2005). Corpus-driven characterizations of pronominal da-compound use by learners and native speakers of German. Die Unterrichtspraxis/Teaching German, 38(1), 44–60. https://doi.org/10.1111/j.1756- 1221.2005.tb00041.x

Das, D., Scheffler, T., Bourgonje, P., & Stede, M. (2018). Constructing a lexicon of English discourse connectives. In Proceedings of the 19th Annual SIGdial Meeting on Discourse and Dialogue (pp. 360–365). Association for Computational Linguistics. https://doi.org/10.18653/v1/W18-5042

Halliday, M. A. K., & Hasan, R. (1976). Cohesion in English. Longman.

Hinkel, E. (2001). Matters of cohesion in L2 academic texts. Applied Language Learning, 12(2), 111–132. Johnson, M. (2017). Improving cohesion in L2 writing: A three-strand approach to building lexical cohesion. English Teaching Forum, 55, 2–13.

Klie, J. C., Bugert, M., Boullosa, B., de Castilho, R. E., & Gurevych, I. (2018). The inception platform: Machine- assisted and knowledge-oriented interactive annotation. In Proceedings of the 27th International Conference on Computational Linguistics: System Demonstrationsm (pp. 5–9). Association for Computational Linguistics.

Konjevod, A. (2012). Connectives in student writing: A learner corpus study. Strani Jezici, 41, 1.

Kunz, K., Degaetano-Ortlieb, S., Lapshinova-Koltunski, E., Menzel, K., & Steiner, E. (2017). English–German contrasts in cohesion and implications for translation. In G. De Sutter, M.-A. Lefer, & I. Delaere (Eds.), Empirical translation studies: New methodological and theoretical traditions (pp. 265–312). De Gruyter Mouton. https://doi.org/10.1515/9783110459586-010

Kunz, K., Lapshinova-Koltunski, E., Martínez Martínez, J., Menzel, K., & Steiner, E. (2021). GECCo - German– English Contrasts in Cohesion: Insights from corpus-based studies of languages, registers and modes. De Gruyter Mouton. https://doi.org/10.1515/9783110711073

Martínez, A. C. L. (2002). The use of discourse markers in EFL learners’ writing. Revista alicantina de estudios ingleses, 15(4), 123–132. https://doi.org/10.14198/raei.2002.15.08

Scheffler, T., & Stede, M. (2016). Adding semantic relations to a large-coverage connective lexicon of German. In Proceedings of LREC (pp. 1008–1013). ELRA.

Stede, M. (2002). DiMLex: A lexical approach to discourse markers. In A. Lenci & V. Di Tomaso (Eds.), Exploring the lexicon: Theory and computation (pp. 1–15). Edizioni dell’Orso.

Stede, M. (2016). Konnektoren und Argumente. In M. Stede (Ed.), Handbuch Textannotation: Postdamer Kommentarkorpus 2.0 (pp. 111–131). Universitätsverlag Potsdam.

Stemmer, B. (1991). Kohäsion im gesprochenen Diskurs deutscher Lerner des Englischen. J. Groos.

Strobl, C. (2020). Darum sind Pronominaladverbien eine Herausforderung für Deutschlerner: Eine korpusbasierte kontrastive Interimssprachenanalyse hierzu. Germanistische Mitteilungen, 45 (1), 89–111.

Tanskanen, S. K. (2006). Collaborating towards coherence: Lexical cohesion in English discourse. John Benjamins. https://doi.org/10.1075/pbns.146

Walter, M. (2007). Hier wird die Wahl schwer, aber entscheidend: Konnektorenkontraste im Deutschen. In H.-J. Krumm (Ed.), Theorie und Praxis - Österreichische Beiträge zu Deutsch als Fremdsprache (pp. 145–161). StudienVerlag.

Walter, M., & Schmidt, K. (2008): “Und das ist auch gut so”: Der Gebrauch des satzinitialen und bei fortgeschrittenen Lernern des Deutschen als Fremdsprache. In B. Ahrenholz, U. Bredel, W. Klein, M. Rost- Roth, & R. Skiba (Eds.), Empirische Forschung und Theoriebildung. Beiträge aus Soziolinguistik, Gesprochene-Sprache- und Zweitspracherwerbsforschung: Festschrift für Norbert Dittmar zum 65. Geburtstag (pp. 331–342). Peter Lang.

Walter, M. (2016). In der Kürze liegt die Würze: Lexikalisch-grammatische Strukturen im akademischen Schreiben. In H. Schweiger, V. Ahamer, C. Tonsern, T. Welke, & N. Zuzok (Eds.), In die Welt hinaus: Festschrift für Renate Faistauer (pp. 201– 217). Praesens.

Webber, B., Prasad, R., Lee, A., & Joshi, A. (2019). The penn discourse treebank 3.0 annotation manual. University of Pennsylvania.