German Impaired Grammar: The Clause Structure Revisited

Abstract

Based on spontaneous data from 50 German children with specific language impairment (SLI), we explore several aspects of impaired clause structure. Our findings are that children with SLI use more finite than nonfinite verb forms (57% vs. 36%). In declarative main clauses they prefer the verb in clause-final position (44%) over genuine (3%) or subject-initial (27%) verb-second patterns; the vast majority (80%) of wh-questions and subordinate clauses are not target consistent. Several current models of clause structure deficits are tested against these findings, especially the Missing Agreement Hypothesis, the Optional Tense Hypothesis, the Truncation Hypothesis, and the Minimal Default Grammar Hypothesis. It is shown that only the latter can account for the entirety of error patterns found in German SLIs. The model is further explored with respect to tense marking in the speech of German SLIs, which is shown to be correct in the presence of overt temporal adverbs.