The acquisition of English as an L3 by Catalan/Spanish bilinguals. A study of crosslinguistic influence in third language acquisition

Abstract

The present study focuses on the acquisition of English as a third language by Catalan/Spanish bilinguals. The aim is to find evidence of crosslinguistic influence from Catalan and/or Spanish in their L3 English by testing the properties of the Definiteness Effect (DE), according to which the copula cannot be followed by a definite DP in existential constructions in English and Spanish; VOS word order, which is possible in Spanish and Catalan but not in English; and VSO word order, which is only possible in Spanish. Our research suggests that the participants transfer from Catalan in the way they rate the DE condition; however, there is not enough evidence to argue in favor of crosslinguistic influence from either of the languages in the word order condition. The results obtained have been contrasted against a series of predictions formulated according to four models of L3 acquisition. Our data suggest non-facilitative influence from Catalan regardless of whether this is the L1 or L2 of the participants. Therefore, we conclude that the results do not corroborate the predictions made for two of the models: The Cumulative-Enhancement (Flynn, Foley, & Vinnitskaya, 2004); according to which crosslinguistic influence should be facilitative or remain neutral and can come from either the L1 or L2; and the L2 Status Factor (Bardel & Falk, 2007, 2012), according to which crosslinguistic influence should come from the last acquired language. The results do not allow us to argue in favor of the Typological Primacy Model (Rothman, 2010, 2013, 2015) either. This model predicts full transfer from the most typologically similar language to the L3; however, there is lack of evidence of full transfer from Catalan, the language which is considered to be more typologically similar to English in this case. The same conclusion is reached regarding the Linguistic Proximity Model (Westergaard, Mitrofanova, Mykhayly & Rodina, 2016), which predicts crosslinguistic influence can come from either the L1 or L2, takes place property by property by property and can be facilitative or non-facilitative, as our results do not provide evidence of crosslinguistic influence from Spanish. However, neither of these two last models is fully ruled out. The results also suggest a possible effect of Catalan in the Spanish of some of the participants.