L3 acquisition of Portuguese clefts by L1-Mandarin L2-English speakers

This study explored the trajectory of L3 acquisition (L3A) of three structurally different clefts in European Portuguese (EP) by L1-Mandarin Chinese (MC) L2-English learners, within the framework of the Feature Reassembly Hypothesis (Lardiere, 2008, 2009). This study also considers the predictions of L3A models that differ with respect to the role of the previously acquired languages on the acquisition of a target L3. Regarding the analyzed structures, L1 factor models (e.g., Na Ranong & Leung, 2009) would predict an early acquisition of pseudoclefts, and a delay in acquiring é-que clefts and standard clefts. Models such as the Cumulative-Enhancement Model (Flynn et al., 2004), the Typological Primacy Model (Rothman, 2011), the Scalpel Model (Slabakova, 2017) and the Linguistic Proximity Model (Westergaard et al., 2017) would predict a facilitating L2 effect on the acquisition of standard clefts and pseudoclefts. 60 MC speakers across three proficiency levels and 21 EP speakers completed an acceptability judgment task (AJT) centered on EP clefts. The learners also performed an AJT of English clefts at a later time. The results suggested a scale of difficulty (standard cleft > pseudocleft > é-que cleft) in development up to advanced stages. The syntactic structure of standard clefts, essentially the left peripheral focus, could be mapped from the L2 onto the L3 lexical items before L3ers reach low-intermediate (B1),¹ while the association of a [+ focus] feature on specific L3 items appears to cause difficulty in acquiring é-que clefts until the learners reach a more advanced (C1)² level. The results supported a facilitating L2 effect on the acquisition of standard clefts, which contradicts predictions that would be made by L1 factor models.