The acquisition of nominal structures (NS) in early Chinese SLA has been shown to be patterned, reflecting the structural and hierarchical complexity of NS (Charters 2013). This study compares the usage patterns of Chinese nominal structures in the form of [X 的 NP] in advanced-level L1 American English (AE) and L1 German (GER) learners of Chinese. The two groups of learners were given the same visual prompt for an independent writing task whereby identical instructions were given that directed them to write an essay of 500 characters to describe and comment on the visual input using the formal register of Chinese. 12 essays in each language were included in the analysis. The results showed that L1 GER and L1 AE NS productions were not significantly different in mean length. However, L1 GER and L1 AE NS productions were significantly different in structural patterns: L1 GER produced more Relative Clause (RC) NS tokens and fewer Attributive/Possessive (A/P) NS tokens than L1 AE. L1 GER produced more OBJ RC tokens and fewer SUB RC tokens than L1 AE. Furthermore, the German sample showed a significantly greater type frequency of NS with RC than the American English sample, suggesting higher productivity of the structure in advanced L1 German learners. Cross-linguistic grammatical influence (Odlin 1989; Wu 2014) was interpreted in terms of a knowledge-based experiential model of learning to account for the cross-L1 differences in the degree of NS complexity and productivity. We suggest that the form-function paired similarity between German Extended Adjective Construction (EAC) and the Chinese RC-like [X_的_N] construction allows L1 GER learners to draw on knowledge of the former in learning the latter and thereby facilitates its acquisition. This argument is consistent with knowledge-based (aka information-theoretical) models of L2/L3 phonological development, e.g. tone/pitch learning (Wu, Tu, and Wang 2012, 2017; Wiener & Goss, 2019). Bohn & Best, 2012; Chang, 2018; Chang & Mishler, 2012), contra the Processability Theory (Pienemann 2005).
This study offers a first corpus-based comparative analysis of cross-linguistic influence in Chinese SLA. Because our samples are small, the argument remain inclusive and awaits further confirmation based on more cross-linguistic data.