Learning a generative syntax from transparent syntactic atoms in the linguistic input

Area of Science:

• Psycholinguistics
• Developmental Linguistics
• Computational Linguistics

Background:

• Understanding how children acquire complex syntactic structures is a fundamental question in linguistics and psychology.
• Previous theories have proposed various innate mechanisms or complex input requirements for language acquisition.

Purpose of the Study:

• To investigate the hypothesis that simple, two-word parental utterances containing core syntactic relations enable children to derive the basic components of hierarchical syntax.
• To identify variables within these utterances that may serve as generative syntactic rules.

Main Methods:

• Analysis of a large English-language parental corpus to identify two-word utterances with verbs and their core arguments (subject, object, indirect object).
• Examination of these utterances for transparent information regarding binary dependency/merge relations.
• A second study analyzed early child sentences to compare verb usage and grammatical relations with parental input.

Main Results:

• Parental two-word sentences demonstrably contain clear information about the binary dependency/merge relation crucial for syntactic connectivity.
• The identified syntactic atoms in parental input naturally incorporate variables, facilitating generalization to other contexts.
• Children in the study predominantly used the same verbs in their early sentences, expressing similar core grammatical relations as observed in parental input.

Conclusions:

• The findings support a learning model where children acquire foundational syntax through exposure to parental two-word sentences.
• Early linguistic input, even in simplified forms, plays a critical role in bootstrapping complex syntactic development.