This week we investigated the evidence of crosslinguistic influences in language learning. For Tuesday, we read chapter 3 in Ortega’s Understanding SLA. Some of the things we talked about in class are: “error” vs. “transfer solution”, our own experiences with crosslinguistic influences, and L1 transfer vs. UG. L1 can speed up or slow L2 learning down, so there can be both positive and negative effects of transfer. Positive transfer is typically less-studied because positive transfer is difficult to pinpoint.
Early researchers thought that most mistakes in L2 were directly related to the learner ‘leaning on’ the grammar of the L1. We now know that this is sometimes the case but it is not a given. Therefore, in researching this phenomenon we focus on the nature of the cognitive processes in the mind and the systematic grammar rules that language learners tend to make, the learner’s perceptions of the relationship between the L1 and L2, and the learner’s proficiency level.
Learners tend to acquire language in a pattern. For example, they tend to acquire question forms in a certain order. The L1 can put learners on the fast track or slow track to learning the L2, depending on how marked the L1’s form is versus the markedness of the L2 form. Learners also tend to learn in a U-shaped pattern, performing well when unaware of grammar structure, making more errors when becoming aware, and improving after learning the form. When learners learn three or more languages, they will tend to transfer patterns from the language that is more typologically similar.
On Thursday, we discussed the Spada and Lightbown study of francophone children in an intensive English program in Canada. We began by discussing the stages of question acquisition. Each stage builds upon the other, adding to the repertoire of question forms that the learner can use. It seems that only when learners use a certain stage (around stage 5) a domino effect can occur, so if the learner acquires that stage first, the other stages will naturally be acquired.
In this study, the researchers used the “input flood” treatment in the classroom. Most of the treatment was input related, while only some was in eliciting the target form. In the testing, two of the tasks were input related, and two were production related.
They found that the learners improved more on written production tasks than oral production tasks. They were also surprised to find that there seemed to be a sub-stage in the acquisition pattern related to the inversion of the noun phrase. They infer that this is due to L1 transfer because French has this rule and English does not. However, as they note at the end, this conclusion cannot be reached unless there is more proof that learners from different L1 backgrounds do not make the same transfer error.
Subscribe to:
Post Comments (Atom)
1 comment:
HI Ann,
Thanks for your commentary of week 2. Just one clarification: The domino effect happens with relative clauses and other areas of language that are in a markedness relationship, but it doesn't happen with questions. For questions, the development occurs from one stage upwards to the next, and learners do not skip stages. Teaching a stage that is too high in development to students who are still at a lower developmental stage is not advisable, because most students would miss the instruction as "too much over their heads." The findings in Spada and Lightbown (1999) show this, for the most part (that is, at least for the large majority of learners the development is constrained upwards, no domino effect).
When we read Chapter 6, we will revisit the topic. For now, I hope our discussion made it clear that the influences of the L1 on the L2 are many, but mostly subtle, non-deterministic, and quite difficult to interpret... Nothing like transferring wholesale word-by-word and structure-by-structure preferences :-)
Post a Comment