Clitic-doubled left dislocation in L2 Spanish: The effect of processing load at the syntax-discourse interface

This project examines whether second-language (L2) learners can converge on a native-like pattern at the interface between syntax and discourse under low and high processing pressure, using Spanish clitic-doubled left dislocation (CLLD) as a test case. The original version of the Interface Hypothesi...

Full description

Saved in:
Bibliographic Details
Published in:Language acquisition Vol. 27; no. 3; pp. 306 - 330
Main Authors: Sequeros-Valle, Jose, Hoot, Bradley, Cabrelli, Jennifer
Format: Journal Article
Language:English
Published: Mahwah Routledge 02.07.2020
Psychology Press
Subjects:
ISSN:1048-9223, 1532-7817
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This project examines whether second-language (L2) learners can converge on a native-like pattern at the interface between syntax and discourse under low and high processing pressure, using Spanish clitic-doubled left dislocation (CLLD) as a test case. The original version of the Interface Hypothesis (IH) predicts that L2 competence on syntax-discourse interface structures may diverge from that of L1 speakers , yet there is experimental evidence of native-like offline judgments of clitic-doubling and its correlation to relevant discourse contexts in L2 Spanish. However, the most updated version of the IH argues that problems at the syntax-discourse interface are not due to divergent competence but rather to L2 processing limitations. To isolate the potential source of divergence, in the present study L2 Spanish learners completed an Acceptability Judgment Task (low processing pressure) and a Speeded Production Task (high processing pressure). Group results show convergence on the native-like pattern on the Acceptability Judgment Task and divergence on the Speeded Production Task. The first finding suggests that L2 learners have linguistic knowledge of discourse-related clitic-doubling, challenging the original version of the IH. The second finding suggests that the processing pressure induced by time constraints and production is correlated with L2 divergence, as predicted by the new version of the IH. We conclude by considering whether this divergence is most appropriately attributed to processing at the interfaces or could alternatively be explained as a result of real-time production specifically.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1048-9223
1532-7817
DOI:10.1080/10489223.2020.1769628