Can Chunking Reduce Syntactic Complexity of Natural Languages?

Qian Lu,Chunshan Xu,Haitao Liu
DOI: https://doi.org/10.1002/cplx.21779
IF: 2.3
2016-01-01
Complexity
Abstract:Natural language is a complex adaptive system with multiple levels. The hierarchical structure may have much to do with the complexity of language. Dependency Distance has been invoked to explain various linguistic patterns regarding syntactic complexity. However, little attention has been paid to how the structural properties of language to minimize dependency distance. This article computationally simulates several chunked artificial languages, and shows, through comparison with Mandarin Chinese, that chunking may significantly reduce mean dependency distance of linear sequences. These results suggest that language may have evolved the mechanism of dynamic chunking to reduce the complexity for the sake of efficient communication. © 2016 Wiley Periodicals, Inc. Complexity 21: 33–41, 2016
What problem does this paper attempt to address?