Electronic Thesis and Dissertation Repository

Thesis Format



Doctor of Philosophy


Hispanic Studies


Bruhn de Garavito, Joyce


Until fairly recently, most researchers assumed that the acquisition of a second language (L2) and of a third (or subsequent) (L3/Ln) language were indistinguishable. This is not the case, as knowledge of two or more previous languages adds complexity to non-native acquisition. This study addresses the issue of crosslinguistic influence between three languages in view of two theories: (a) the L2 will always be the cause of crosslinguistic influence in an L3 (Bardel & Falk, 2012); (b) the determining factor for transfer is typological similarity between the languages in question (Rothman, 2010).

This study focuses on the L3 acquisition of Spanish by speakers whose L1 is Brazilian Portuguese (BP) and L2 is English. A comparison group consisted of learners of L2 Spanish whose L1 is BP. In particular, I examined the acquisition of the choice of word order in Spanish, subject-verb or verb-subject, which depends on two factors: the type of verb (morphosyntax), and the information structure of the sentence (pragmatics), whether focused or unfocused. The problem for learners consists of processing the interface of two linguistic modules: morphosyntax and pragmatics. Interfaces have been found to be difficult in acquisition (Sorace, 2011), adding an additional wrinkle to the problem faced by learners.

The three languages chosen differ in relation to word order: Portuguese distinguishes between verb types, but pragmatics is not a factor; English exhibits fixed subject-verb order; and Spanish takes both verb class and pragmatics into consideration. Because of this, we are able to distinguish between possible influence from English, which predicts rejection of verb-subject order in all circumstances, and typology, which predicts learners using verb class to distinguish word order but not pragmatics.

Participants completed a preference task (Lozano, 2006), a production task, a vocabulary quiz, and a linguistic profile questionnaire. Findings support the hypothesis that typology plays a role in L3 transfer, as learners are able to adjust their interlanguage to accept the inverted order in instances where the L2 is inflexible. However, both L2 and L3 learners also seem to integrate pragmatic constraints in their interlanguage showing that they are on their way to acquiring a Spanish native-like grammar.

Summary for Lay Audience

This study examines the acquisition of Spanish as a third language and whether the second language of a speaker is the main source of influence when learning a third, or whether crosslinguistic influence comes from the language that is typologically similar to the language being acquired. The main participants were native speakers of Brazilian Portuguese (BP) who had learned English as a second language and were learning Spanish as a third. A comparison group consisted of speakers of BP who were learning Spanish as a second language.

The objective was to analyze how learners deal with word order alternations in Spanish. While English is a strictly subject-verb word order language, Spanish and BP allow subject-verb inversion. In BP, the factor that determines word order is the type of verb used, while in Spanish both the type of verb and the focus (pragmatics) of the sentence are important. Therefore, if the source of influence in Spanish is BP, the typologically similar language, we expect learners to accept verb-subject order relying mainly on verb class. However, if it is English, we predict that L3 learners will only accept English word order.

Structures in which learners have to process two complex elements, verb class and pragmatics, are particularly difficult because, even though syntactic constraints (verb class) seem to be in place since early stages of the learning process, pragmatics (focus) seems to cause difficulties even at advanced levels.

Participants completed a written task, a production task, a vocabulary quiz, and a linguistic profile questionnaire. Few instances of outright rejection of the verb-subject order were found, which lends support to the prediction that the source of transfer is the typologically similar language, and not necessarily the second or more recently acquired one. However, learners were also able to acquire pragmatic constraints in L3 Spanish, which shows the importance of input. The level of proficiency also seems to play an important role in the way learners accept and produce sentences with inverted order. This study therefore contributes to the issues related to L3 acquisition and also to our understanding of the acquisition of complex interface phenomena.