Abstract
Usage-based approaches assume that children’s early utterances are item-based. This has been demonstrated in a number of studies using the traceback method. In this approach, a small amount of “target utterances” from a child language corpus is “traced back” to earlier utterances. Drawing on a case study of German, this paper provides a critical evaluation of the method from a usage-based perspective. In particular, we check how factors inherent to corpus data as well as methodological choices influence the results of traceback studies. To this end, we present four case studies in which we change thresholds and the composition of the main corpus, use a cross-corpus approach tracing one child’s utterances back to another child’s corpus, and reverse and randomize the target utterances. Overall, the results show that the method can provide interesting insights—particularly regarding different pathways of language acquisition—but they also show the limitations of the method.
Dokumententyp: | Zeitschriftenartikel |
---|---|
Keywords: | construction grammar, language acquisition, traceback, usage-based model |
Fakultät: | Sprach- und Literaturwissenschaften > Department 1 > Deutsch als Fremdsprache |
Themengebiete: | 400 Sprache > 410 Linguistik
400 Sprache > 430 Deutsch, germanische Sprachen allgemein |
ISSN: | 1613-7027 |
Bemerkung: | Ahead of print, just accepted |
Sprache: | Englisch |
Dokumenten ID: | 84578 |
Datum der Veröffentlichung auf Open Access LMU: | 20. Jan. 2022, 07:27 |
Letzte Änderungen: | 20. Jan. 2022, 07:27 |