DISCOURSE COMPLEXOLOGY: PARADIGM SHIFT OR TERMINOLOGY TRANSFORMATION?
Marina Ivanovna Solnyshkina, Gulnoza N. Shoeva · Vestnik Tomskogo gosudarstvennogo universiteta · 2024
The research is aimed at providing readers with a concise summary of a century-long history of views on “text complexity” and “comprehension difficulty” as the main research objects in discourse complexology. Both concepts are of high relevance for automated assessment of complexity predictors and subsequent adaptation of a text for the target audience. The article is structured into four sections, each addressing different aspects of discourse complexology. In the Introduction, we highlight the similarity of two research paradigms, i.e. text complexity assessment and text adaptation. We also indicate the importance of text-reader match in numerous spheres indicating that avoidance of communication failures is especially significant in institutional discourse. We offer a historical review of three main approaches in the paradigm of discursive complexology: parametric, neural and criterial, which have been developing and reshaping the intensions of the concept of complexity for over a century. We register three paradigm shifts in integrated studies of text complexity, i.e. in the 1920s after the introduction of statistical methods, in the 1970s after the acceptance of Walter Kintsch's comprehension model and in the late 2010s after the implementation of machine learning and neural nets. We also justify the introduction of the term “discourse complexology” by broadening the scope of studies in the area due to representative datasets and effective tools at the disposal of modern researchers. The part “Complexity as a Determinant of Text Adaptation/Simplification” discusses the foundation theories of text complexity, evolution of complexity assessment methods, and practical applications of these theories in text adaptation and simplification. We examine various methods of text adaptation including semantic, structural and compositional modifications, and highlight the importance of readers' linguistic and cognitive abilities in effective communication. The article further discusses modern approaches to the assessment of text complexity and comprehension difficulty, which incorporate interdisciplinary tools and perspectives. We emphasize the emergent nature of text as a construct where the whole is more than the sum of its parts thus challenging the traditional additive theory on text complexity. The discussion extends to the implications of the existing theories on text comprehension and the role of individual reader's abilities in text perception, comprehension and recall. The article concludes with a call for continued interdisciplinary research to refine models of text complexity and adaptation, emphasizing the need for an integrated approach. Future research should focus on developing more sophisticated methods for predicting text complexity and adapting texts to meet different readers' needs, thereby enhancing communication across domains. The research findings may be employed to predict specifics of different categories of readers' perception and in natural language processing. The authors declare no conflicts of interests.