Crossings as a side effect of dependency lengths

TL;DR

This paper investigates whether the tendency of dependencies not to cross in sentence structures is due to an independent syntactic principle or a side effect of dependency lengths, favoring the latter for its simplicity.

Contribution

It provides evidence rejecting the traditional independent principle hypothesis and supports the dependency length hypothesis as a more parsimonious explanation.

Findings

Crossings are not solely due to an independent syntactic principle.

Shorter dependency lengths correlate with fewer crossings.

The dependency length hypothesis explains crossing patterns across languages.

Abstract

The syntactic structure of sentences exhibits a striking regularity: dependencies tend to not cross when drawn above the sentence. We investigate two competing explanations. The traditional hypothesis is that this trend arises from an independent principle of syntax that reduces crossings practically to zero. An alternative to this view is the hypothesis that crossings are a side effect of dependency lengths, i.e. sentences with shorter dependency lengths should tend to have fewer crossings. We are able to reject the traditional view in the majority of languages considered. The alternative hypothesis can lead to a more parsimonious theory of language.