ABSTRACT
Previous studies of dependency distance as a measure of, or a proxy for, syntactic complexity do not consider factors such as sentence length and root distance. In the present study, we propose a new algorithm, i.e. Normalized Dependency Distance (NDD), that takes sentence length and root distance into consideration. Our analysis showed that exponential distribution fit well the distribution model of NDD as it did with Mean Dependency Distance (MDD), the algorithm used in previous studies. Findings indicated that NDD is significantly less dependent on sentence length than MDD is, which suggests that the new algorithm may have, to some extent, addressed the issue of MDD’s dependency on sentence length. It is argued that NDD may serve as a measure of syntactic complexity, which is a kind of universality limited by the capacity of human working memory.
Acknowledgments
This work was supported by National Social Science Fund of China (Grant No. 15BYY179). The authors appreciate the reviewers’ insightful comments and suggestions.
Disclosure statement
No potential conflict of interest was reported by the authors.