Syntactic complexity in professional architectural discourse: Insights from native and non-native English speakers’ texts

This study investigates the syntactic differences between texts authored by native (NES) and non-native English-speaking (NNES) professional architects on ArchDaily, the world's most visited architecture website. We focused on established indices of syntactic complexity hypothesized to differentiate between first language (L1) and second language (L2) writing. The corpus consisted of randomly selected residential architecture texts by Iranian and British architects. Data analysis was conducted using Coh-Metrix Core Desktop Beta (2023), with results processed in SPSS. Preliminary MANOVA analysis revealed significant differences in the mean scores of the two groups, and follow-up analyses indicated that syntactic complexity indices, including left-embeddedness, minimal edit distance for words, and the number of modifiers per noun phrase, exhibited medium to large effect sizes in distinguishing the texts. Additionally, among the syntactic density indices, agentless passive voice, negation, gerund, and noun phrase density were most effective in differentiating native from non-native compositions. The findings revealed that while both NES and NNES architects demonstrate sophisticated writing skills, they exhibit distinct patterns of syntactic complexity in their professional discourse. NES architects tend towards richer modification and cohesive flow, while NNES architects lean towards longer and denser sentences. These differences, influenced by both linguistic proficiency and discourse-specific conventions, highlight the need for continued attention to clarity and readability in architectural communication and underscore the importance of considering both linguistic proficiency and discourse-semantic motivations when analyzing syntactic complexity in professional writing.