Opponent: Professor Scott Jarvis, The University of Utah, USA
Supervisors: Professor Lázaro Moreno Herrera, professor Camilla Bardel, ass. professor Staffan Nilsson, PhD Ali Reza Majlesi, Stockholm University

Abstract

A long-standing pedagogical implication of the traditional monolingual perspective in research and education has been to discourage any use of non-target languages in the L2 classroom as an attempt to reduce ‘interferences’ (i.e., transfer) between learners’ languages. Within the current multilingual paradigm shift, however, it has become clear that multilingualism can be an asset and the positive effects of transfer are acknowledged. The recent theoretical proposal of ‘pedagogical translanguaging’ suggests that teachers may help learners find opportunities to transfer aspects from their languages intentionally, as a resource or creative strategy, by raising their awareness of similarities between their languages.

Yet, from an empirical perspective very little is known about (un)intentionality in transfer. This thesis investigates how multilingual language learners use transfer unintentionally versus intentionally and how several potentially important variables may be related to (un)inten-tionality in transfer. The focus is on lexical transfer in the written production of multilingual learners of Spanish in Sweden.

Written transfer data were collected from 78 students in heterogeneously multilingual classrooms, including learners with different language backgrounds. Words that were reported (intro- or retrospectively) by a learner to have been transferred from a background language were coded as instances of intentional transfer. Conversely, when non-target-like words that could be traced back to a background language by the researcher were used but not com-mented on by the learner, they were coded as instances of unintentional transfer.

A quasi-Poisson regression analysis was employed to test the significance of transfer factors on learners’ amounts of unintentional and intentional transfer. A qualitative lexical analysis was conducted, moreover, to explore how the different types of lexical transfer were manifested depending on whether they were produced unintentionally or intentionally. As a complement to the data, the learners’ teachers were interviewed on their attitudes towards the use of non-target languages in the Spanish lessons.

The learners produced lexical transfer both unintentionally and intentionally. A significant tendency was found for intentional transfer to come from L2s rather than from L1s, possibly due to a higher level of metalinguistic knowledge in L2s. A significant tendency was also found for unintentional transfer to occur as a result of crosslinguistic morphological similarity. Additionally, a trend was observed for more multilingual learners to exhibit smaller amounts of unintentional transfer, possibly due to an enhanced conscious control of language use. (Un)intentionality was also found to relate to the different types of lexical transfer (e.g., foreignising, borrowing) in varying ways. The learners’ teachers felt positive about a multilingual instructional approach and often tried to make crosslinguistic comparisons.

Apart from shedding light on the dimension of (un)intentionality in lexical transfer, this thesis elaborates on pedagogical translanguaging and the idea of using languages as resources. It suggests that pedagogical linguistic awareness-raising activities should aim not only to help learners find opportunities to use transfer as a strategy, but also to help them avoid certain instances of unintentional transfer, and offers insights into which linguistic aspects such activities could focus on.