Connecting input to comprehension: First language acquisition of active transitives and suffixal passives by Korean-speaking preschool children

Shin, Gyu-Ho
O’Grady, William
Journal Title
Journal ISSN
Volume Title
Starting Page
Ending Page
Alternative Title
A central issue in language acquisition is the contribution of input to the development of linguistic knowledge. In this dissertation project, I investigate the developmental trajectories of two constructions that express transitive events in Korean (active transitives and suffixal passives) for Korean-speaking preschool children. Three major grammatical factors affect interpretation of Korean sentences: word order through the relative position of arguments in a sentence, case marking via designated markers attached to arguments, and voice by way of verbal morphology. Each factor induces particular comprehension heuristics (i.e., a strategic way, acquired probabilistically through exposure, that a comprehender employs in the course of comprehension). Literature on the interaction of the three factors in the acquisition of the two construction types remains relatively thin, particularly for distributional properties of the relevant input and the (asymmetric) contributions of these factors to children’s comprehension. Throughout the dissertation, I made use of corpus analysis through Natural Language Processing techniques and picture selection experiments in order to investigate this issue. I first conducted a semi-automatic analysis of caregiver input using the entire Korean child-directed speech data in the CHILDES database. Four major findings were reported as follows: (1) Of the core constructional patterns with no omission of arguments and case marking, the canonical active transitive occurred far more frequently than its scrambled counterpart, and passives in general were extremely rare, regardless of canonicity. (2) Of the three passive types found in Korean—suffixal (p. 3, (5)), lexical (p. 3, (6)), and paraphrastic (p. 4, (7)), the suffixal passive was the most frequent of all instances of the passive (with or without argument / case marking omission). (3) The degree of association between individual markers and thematic roles in constructional patterns expressing a transitive event was asymmetric: the nominative case marker was a very strong cue for agenthood (and vice versa), the accusative case marker was a moderately good cue for themehood (and vice versa), and the dative marker was not likely to occur with the agent (and vice versa). (4) When two overt arguments are attested in active transitives, the NOM-marked argument tends to occur before its ACC-marked counterpart. I also carried out a series of picture selection experiments, by devising a novel methodology in which parts of test sentences were obscured by way of acoustic masking with child-friendly contexts. Given the experimental setting (i.e., reversible stimuli with two animate arguments), it was found that three factorsword order, case marking, and voiceinteracted with one another in children’s comprehension of the two constructions in the following ways: (1) The word-order-related heuristic (Agent-First) operates reliably only in conjunction with other grammatical cues such as the presence of a second argument and case marking. (2) The case-marking-related heuristics (NOM-as-Agent; ACC-as-Theme), which apply locally to a single noun, work more reliably for comprehension than the word-order-related heuristic (Agent-First). (3) The voice-related heuristics (Theme-First; DAT-as-Agent) are less influential in comprehension than the word order and case marking heuristics, which frequently override them. Children’s performance in this experiment was interpreted in combination with input properties and postulated features of a child processor. By and large, characteristics of each comprehension heuristic mirrored properties of caregiver input, which suggests a close connection between what children are exposed to and how knowledge related to these factors emerges and grows. Despite the scope of investigation (i.e., patterns expressing transitive events with animate agents and themes), the nature of input provided a reasonably clear indication that children develop particular heuristics in relation to each factor and apply them to comprehension. This finding aligns well with usage-based and emergentist approaches to language development, pointing towards a substantial contribution of input to child language development.
Linguistics, Caregiver input, Case marking, Comprehension, Voice, Word order
178 pages
Geographic Location
Time Period
Related To
Table of Contents
All UHM dissertations and theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission from the copyright owner.
Rights Holder
Local Contexts
Email if you need this content in ADA-compliant format.