Exploring Grammatical Complexity across Proficiency Levels in Argumentative Essays by Iranian EFL Learners

Different measures have been proposed for predicting syntactic complexity in second/foreign language writing. While some have advocated the use of T-unit and clausal subordination measures, others have argued for the use of phrasalcomplexity as an index of syntactic development, especially when analysing academic prose. This study seeks todetermine which of these two measures can best explain syntactic development in the genre of argumentative writing.For this purpose, a corpus of argumentative essays by learners from three different levels of proficiency (i.e., preintermediate,intermediate and upper-intermediate) was compiled. The corpus (including 60,986 words) was then coded and analysed for both phrasal and clausal features. The phrasal features were manually coded based on the schemehypothesized by Biber, Gray and Poonpon (2011) to predict developments in academic writing and the clausal featureswere analysed using the online L2 syntactic complexity analyser developed by Lu (2011). A separate ANOVA test wasused to compare the three groups of essays in terms of each of the phrasal and clausal features. The results showed thattwo features (that relative clauses with animate head nouns and nonfinite relative clauses) from the noun phrasal complexity index and five features (mean length of T-unit, verb phrase per T-unit, mean length of clause, complex nominal per T-unit, and complex nominal per clause) from the T-unit- and clausal-based measures increased with proficiency. The findings indicate that phrasal features cannot be used as an index for tracing syntactic complexity development in the genre of argumentative essays. Our findings question whether the argumentative essay could be taken as a sub-register of academic writing

Keywords:

complex noun phrases , corpus linguistics , argumentative essays , EFL learners , T-unit