Cross-linguistic Syntactic, Lexical and Phonetic Influence in the Acquisition of L3 Spanish

Abstract

The main ongoing debate within the recently-established field of Third Language (L3) Acquisition revolves around which of the previously-acquired languages prevails as a source of cross-linguistic influence (CLI) during production in the target language. Whereas several factors have been found to promote CLI, the existing body of L3 research points to a potential stronger rivalry between two of them: typology, or the relative distance among the languages involved, and second language (L2) status, also known as foreign language effect. In fact, they are at the core of two models of multilingual transfer that stem from the area of L3 morphosyntax, namely the Typology Primacy Model (TPM; Rothman, 2010, 2011, 2013, 2015), and the L2 Status Factor Model (L2 SFM; Bardel & Falk, 2007, 2012). For the most part, claims backing a more determinant role for typology come from studies investigating lexis. Whether or not typology overrides the effect of the L2 with regards to syntax and phonology as well is less clear. On the one hand, studies focusing on syntactic CLI can be divided into those whose findings suggest a crucial role either for the L1 (Na Raong & Leung, 2009) or for the L2 (Bardel & Falk, 2007), those in which findings point to typology as the deciding factor in determining a source of CLI (Rothman, 2011), and those that fail to show any influence from the L2 on the L3 (Håkansson, Pienemann, & Sayehli, 2002) even when the L2 and the L3 are typologically close (Martínez Adrián, 2005). On the other hand, results from previous research investigating phonological CLI also yield three main distinct findings: i) a more marked L2 effect (Llama, Cardoso & Collins, 2010), ii) a predominant influence from the first language (L1; Llisterri & Poch, 1987), and iii) combined CLI, that is, the L1 and the L2 concur in influencing the L3 (Wrembel, 2014). The primary goal of this dissertation is to add to the debate by looking into all three sub-areas among trilinguals at a high-intermediate to advanced level of proficiency, which has been the least targeted proficiency level until now. The selected topics, per area, are: lexical inventions, relative clause attachment preferences (RCA), and voice onset time (VOT). The secondary goal was to compare the relative influences of typology and L2 status across sub-areas, and to examine all results in light of three of the current L3 multilingual transfer models, the TPM, the L2 SFM, and the Cumulative Enhancement Model (CEM; Flynn, Foley, & Vinnitskaya, 2004). The results obtained are in clear agreement with previous reports in the area of lexis. However, the interplay of our trilinguals’ linguistic systems with regard to RCA and VOT is more complex. Findings in these areas seem to suggest that other factors (language exposure in the case of RCA, and the L1, in the case of VOT) may trump the two under investigation.