Language transfer in L2 academic writings: a dependency grammar approach

Area of Science:

• Linguistics
• Computational Linguistics
• Second Language Acquisition

Background:

• Dependency distance (DD) is crucial for language processing and comprehension.
• Prior research explored DD in L2 writing, but the influence of native language remains understudied.
• Understanding native language influence on DD in L2 academic writing is essential for pedagogical development.

Purpose of the Study:

• To investigate the influence of L1 on dependency distance (DD) in L2 academic writing.
• To analyze DD patterns in a large corpus of L2 academic abstracts across different language backgrounds.
• To determine if native language transfer affects DD in English L2 academic writing.

Main Methods:

• Analysis of over 2.2 million academic abstracts from Scopus (Arts & Humanities, Social Sciences).
• Utilized Stanford CoreNLP 4.5.5 syntactic parser to analyze dependency distance patterns.
• Compared mean dependency distance (MDD) across L2 writers from various language backgrounds and native English writers.

Main Results:

• Native language significantly influences the mean dependency distance (MDD) in L2 English academic writing.
• L2 writers whose native languages have a longer MDD exhibit longer MDDs in their English academic writing compared to native English writers.
• A clear correlation exists between L1 MDD and L2 English MDD, indicating language transfer.

Conclusions:

• Native language transfer demonstrably affects dependency distance in L2 academic writing.
• Findings provide insights into syntactic patterns in L2 academic discourse.
• Results can inform pedagogical strategies for L2 academic writing instruction, focusing on syntactic complexity and native language interference.