Age effects in L1 attrition: The role of input

Silvina Montrul, University of Illinois at Urbana-Champaign

Studies of maturational effects in language development have mostly been concerned with L1 and adult L2 acquisition. In this talk, I argue that age effects are also relevant in normal L1 attrition, or the loss of linguistic skill in a bilingual environment. In essence, the younger the individual when reduction to language input and lack of use of the L1 takes place, the more severe the extent of language loss at the grammatical level. In the first part of the talk I will review studies showing how the effects of attrition in childhood are more dramatic than in adulthood. And within childhood, I will show that simultaneous bilingual children are more vulnerable to attrition and incomplete acquisition under reduced input conditions than sequential bilingual children. In addition to stressing the role of age, I will also emphasize the role of input, since the long-term effects of reduced input in childhood appear to be less dramatic for the developing L1 grammar than the effects of interrupted input. To illustrate cases of interrupted input, the second part of the talk focuses on the unique language learning experiences of internationally adopted children. I will present behavioral data from an ongoing case study of Alicia, a Guatemalan adult adopted by an American family at age 9, that challenges a recent proposal by Pallier, Dehaene, Poline, LeBiham, Argenti, Depoux & Mehler (2003) and Ventureyra, Pallier and Yoo (2004). Pallier and collaborators found that Korean adults adopted as children by French families between the ages of 3 and 8 years of age exhibited complete loss of Korean, while their French was native-like. They suggested that total L1 loss allows a resetting of the neural networks that normally subserve L1 retention and hence permits native-like acquisition of an L2. Unlike Pallier's findings, the behavioral data from my research project is consistent with the idea that there are age effects for L1 attrition, and that a 8-9 year old is unlikely to undergo complete L1 loss. Nonetheless, interrupted L1 input after age 9 has long-lasting and significant effects in the adult linguistic competence, as revealed from an analysis of Alicia's Spanish nominal and verbal morphology elicited with a variety of oral and written measures.