On the complexity of discourse complexity: A multidimensional analysis

The present study uses a theory‐based statistical approach to investigate the dimensions of discourse complexity in English, analyzing the distribution of 33 surface linguistic markers of complexity across 23 spoken and written registers. The study shows that discourse complexity is a multidimensional construct, that different types of structural elaboration reflect different discourse functions, and that different kinds of texts are complex in different ways (in addition to being more or less complex). Building on earlier exploratory research, the study identifies a particular five‐dimensional model as the most adequate with respect to the surface linguistic features considered here. These dimensions are labeled to reflect their functional and grammatical underpinnings: Reduced Structure and Specificity, Structural Elaboration of Reference, "Framing" Structural Elaboration, Integrated Structure, and Passive Constructions. Analysis of the relative complexities of spoken and written registers with respect to these dimensions enables an assessment of the relative importance of each dimension, further refinement of their functional interpretations, and a multidimensional assessment of the discourse complexity of particular texts and registers. This analysis identifies a fundamental distinction between the discourse complexities of written and spoken registers: Whereas written registers exhibit many complexity profiles, differing widely in both the extent and the kinds of complexity, spoken registers manifest a single major pattern differing only in extent.