A corpus-based dependency study of the correlation between nominal group complexity and clause complexity in English academic writing

Linguistic complexity can be analyzed from two perspectives: group complexity and clause complexity. Complex nominal groups (NGs) are widely recognized as a defining characteristic of English academic writing. This study investigates the relationship between NG complexity and clause complexity across three disciplinary groups— Social Sciences (SS), Humanities and Natural Sciences (NS)—through the lens of dependency distance (DD). The results show that participant NGs are most complex in NS texts, whereas clauses exhibit the greatest complexity in Humanities texts. In general, there is a tendency to minimize the insertion of other clausal constituents between the verbal group and its object NG. Among the few inserted constituents, those with shorter DDs from the head verbs (HVs) to the object NGs are more frequent in Humanities texts, while those with longer DDs are more common in NS texts. Shorter in-between constituents typically form phrasal verbs with the HVs, while longer ones are mostly prepositional phrases functioning as circumstantial adjuncts. These findings suggest that NG complexity and clause complexity are not necessarily negatively correlated, both contributing to the overall linguistic complexity of English academic writing, and that adverbial groups functioning as comment adjuncts or intensifiers are not encouraged in English academic writing.