Errors in inflectional morphemes as an index of linguistic competence of Korean Heritage language learners and American learners of Korean

Abstract

This study examined the linguistic competence in Korean of Korean heritage language learners (HLLs), compared to English-speaking non-heritage language learners (NHLLs) of Korean. It is unclear and controversial as to whether heritage languages learners are exposed to early but are interrupted manifest as L1 competence or share more characteristics with development in L2/FL competence. However, a common misconception is that HLLs outperform NHLLs in overall language skills even though Korean HLLs in Korean as a Foreign Language (KFL) classes do not make better progress than NHLLs despite their comparatively stronger aural interpretive abilities. This study was designed to investigate whether HLLs have an advantage over NHLLs in learning distinctive parametric values in Korean language, through comparing occurrences and sources of grammatical errors exhibited by two groups taking university-level KFL classes. This study addresses Korean inflectional morphemes, with a focus on case and postposition markers and affixal connectives. Data was collected from error analysis (EA) of inflectional morpheme errors and its source on semi-guided and self-generated writing samples, and grammaticality judgment in a word completion (GJWC) test using the same inflectional morphemes used for the EA. Schlyter's Weak language (WL) as L2, Montrul's WL as L1, and the Missing Surface Inflection Hypothesis (MSIH) provided theoretical frameworks. The EA data was coded using the Systematic Analysis of Language Transcript program. The EA and GJWC data were analyzed using a 2-way ANOVA and, when there was a significant interaction effect between heritage status and language proficiency level, a 1-way ANOVA. This study's results confirmed Schlyter's hypothesis, but did not support Montrul's hypothesis from either the EA or GJWC. MSIH failed in explaining underlying linguistic competence of HLLs. Significantly higher error rates caused by omitting necessary subject and object markers among HLLs imply their Korean morphological data stays at the level of Korean child's morphology. Significantly higher error rates in instrument marker in the GJWC test by advanced level of HLLs imply impaired Korean morphology of HLLs. Linguistic variation is more prominent among HLL group. Findings are further discussed in relation to their theoretical, methodological, and pedagogical implications. Differentiated instructional and curricular approaches for HLL and NHLL groups are suggested.