Optimizing scientific communication: the role of relative clauses as markers of complexity in English and German scientific writing between 1650 and 1900
Saarland University, Saarbruecken, Germany, 2023.
The aim of this thesis is to show that both scientific English and German have become increasingly optimized for scientific communication from 1650 to 1900 by adapting the usage of relative clauses as markers of grammatical complexity. While the lexico-grammatical changes in terms of features and their frequency distribution in scientific writing during this period are well documented, in the present work we are interested in the underlying factors driving these changes and how they affect efficient scientific communication. As the scientific register emerges and evolves, it continuously adapts to the changing communicative needs posed by extra-linguistic pressures arising from the scientific community and its achievements. We assume that, over time, scientific language maintains communicative efficiency by balancing lexico-semantic expansion with a reduction in (lexico-)grammatical complexity on different linguistic levels. This is based on the idea that linguistic complexity affects processing difficulty and, in turn, communicative efficiency. To achieve optimization, complexity is adjusted on the level of lexico-grammar, which is related to expectation-based processing cost, and syntax, which is linked to working memory-based processing cost. We conduct five corpus-based studies comparing English and German scientific writing to general language. The first two investigate the development of relative clauses in terms of lexico-grammar, measuring the paradigmatic richness and syntagmatic predictability of relativizers as indicators of expectation-based processing cost. The results confirm that both levels undergo a reduction in complexity over time. The other three studies focus on the syntactic complexity of relative clauses, investigating syntactic intricacy, locality, and accessibility. Results show that intricacy and locality decrease, leading to lower grammatical complexity and thus mitigating memory-based processing cost. However, accessibility is not a factor of complexity reduction over time. Our studies reveal a register-specific diachronic complexity reduction in scientific language both in lexico-grammar and syntax. The cross-linguistic comparison shows that English is more advanced in its register-specific development while German lags behind due to a later establishment of the vernacular as a language of scientific communication.