Corpus-based analysis of genetically modified seed discourse

While ecolinguistics studies interactions between language and the natural world, the theme of genetic modification highlights challenges in developing frameworks for the analysis of ecological discourses. Given the ability of humans to alter life at the molecular level, one could argue that frameworks of anti-essentialism and discursive constructivism have expanded from the cultural and social realms (the traditional focus of discourse analysis) to encompass the natural world. This paper uses multilevel analysis to draw out various dimensions of debates related to genetically modified (GM) seed. Analysis is applied to a web-based corpus using both computational, quantitative methods as well as qualitative critical and interpretive approaches. Keyword analysis, concordance lines, and collocation are used to explore whether sides of the debate are reflected in the semantic structure of the text. Implicature and conceptual blending point to differences at the cognitive level. The analyses give data-based evidence for differing epistemologies, worldviews, and situated contexts that may underlie misunderstandings. Results suggest the multilevel methodology can be an effective way to integrate ecological context into discourse analyses. Findings also lead to a discussion of conceptual frameworks for ecolinguistic discourse analysis.