Dissertation

Relexification and Lexical Relativity

in the Second Language Acquisition of the Dative Alternation

Defended 17 June, 2022

Abstract:

English ditransitive verbs show a complex alternation between the double object construction (DOC, (1)) and prepositional object datives (POD, (2)). This dissertation examines the acquisition, representation, and learnability of the dative alternation among L2 English learners, presenting experimental data from participants with L1 backgrounds of Mandarin Chinese (n = 14), Korean (n = 19), and Spanish (n = 21), as well as L1 English (n = 41) speakers. While argument structure alternations have played an important role in L2 acquisition, this study differs from previous work in that it focuses on the abstract syntax of datives.

The DOC exhibits an asymmetric structural pattern that has been a key point of discussion in syntactic research (e.g., Citko, Emonds, and Whitney, 2017, henceforth CEW). A-bar movement (wh-questions and relative clauses) of the goal argument is ungrammatical (3), while A-bar movement of the theme is acceptable (4). Conversely, A-movement of the theme argument is ungrammatical (5) with non-pronominal arguments, while A-movement of the goal (6) is grammatical. CEW have argued that these structural contrasts are best explained through a syntactic derivation of the DOC from the POD that promotes the indirect object to a direct object position. This proposal, based on Emonds (1993), links English DOCs to Marantz’s Generalization, a cross-linguistic pattern identified by Baker (1988) in which indirect objects surface productively as applied direct objects.

This experiment tests whether L2 learners’ grammatical intuitions show the syntactic hallmarks of Marantz’s Generalization. Participants judged the acceptability of sentences formed with twelve verbs (give, send, tell, text, buy, make, return, push, say, yell, purchase, create) appearing in each of seven syntactic contexts, which included passives, questions, and relative clauses targeting both direct objects and indirect objects. The study also includes a detailed contrastive analysis of ditransitive constructions in Mandarin Chinese, Korean, and Spanish, showing that the ditransitive structures found in these languages differ considerably from English in their distribution and structure.

A Generalized Linear Mixed Model analysis was developed to test whether the CEW analysis makes accurate predictions for the acceptance of sentences among L1 and L2 English speakers. Notably, both L1 and L2 English speakers showed sensitivity to the ungrammaticality of A-movement of themes and A-bar movement of goals. These results point to a strikingly asymmetric pattern, in which verbs that form double objects productively surface with distinct syntax. Another interaction was identified between L2 English proficiency and rejection of sentences with non-DOC verbs, modeled at the individual participant level, incorporating Stringer’s (2010) notion of Lexical Relativity. The syntactic regularity observed across L1 groups thus coexisted with another pattern of lexical variation.

The results lend empirical support to syntactically derived accounts of the dative alternation. From an acquisitional perspective, the data are best situated within a theory of L2 acquisition that assumes both pervasive L1 influence at the level of individual lexical items, and adult access to universal grammatical constraints, as proposed in Sprouse’s (2006) model of the initial state of L2 acquisition in terms of Relexification.

Examples

(1) Mary gave the student a book.

(2) Mary gave a book to the student.

(3) *Which student did Mary give a book?

(4) What book did Mary give the student?

(5) *The book was given a student.

(6) The student was given a book.