This paper focuses on Lithuanian L2 verb forms that do not exist in the target language. The data is drawn from the Lithuanian L2 corpus and includes spontaneous speech data by 17 speakers of different L1s and all three acquisition varieties (pre-basic, basic and post-basic). The non-existent verb forms make only a small part of the overall verb forms used, however, when distinguished from occasional mistakes, the forms that indicate grammatical development that are analysed show notable systematicity among different speakers. Five distinct and most frequent formants are analyzed in detail, namely -u, -a, -o, -(i)au, ė, that are used by several speakers with seemingly similar function. The analysis reveals specific functions and discusses different factors that have an effect on the acquisition of these formants. In many cases frequency is an important factor, however, with formants -u and -ė transparency and regularity seem to have a stronger effect and thus facilitate the acquisition. Formant -a, being nearly the most common and extremely multifunctional (i.e. rather unmarked), is incorporated into the system firstly as part of the base form and holds this position till a very advanced stage of Lithuanian L2 acquisition. Verb stem type also has an effect with the regularly suffixed verbs being acquired the easiest, while the mixed type and the primary ones cause many difficulties for the speakers. Overall, the systematicity discovered shows the creative structuring of the interlanguage that only occurs more actively in the post-basic variety of acquisition.
L2 Chinese Learners' Lexical and Grammatical Development of Result-state Resultative Verb Compounds: A Usage-based Corpus Study
