Climate change vs global warming: A corpus-driven approach to climate talk of the decade 2010-2019

Abstract

Ecolinguistics, a strand of research pioneered by Halliday (1990) focusing on the impact of language on the environment, has traditionally taken an approach similar to critical discourse analysis (CDA) (Baker & Ellece, 2011; Fill & Mühlhäusler, 2001). In the last decade, the application of corpus-assisted methods in ecolinguistics studies began to draw attention. This presentation extends the talk by Law and Matthiessen (2019) with an aim to look deeper into the lexicographical patterns in climate talk between 2010 and 2019. A 227,499-word eco-corpus was created from a collection of 275 randomly selected texts on the topic of environment published on the internet within this period. These texts consist of media reports, online magazines, transcripts from TV talk shows, and public speeches. A corpus linguistics analysis emphasizing on 1-gram, 2-word concgrams, and 3-word concgrams and their respective concordances was performed using ConcGram 1.0 (Greaves, 2009). Findings revealed the top 40 unique words, unique nouns, meaningful 2-word and 3-word concgrams, as well as distinctive trends in word choices. This includes the preference for climate change over global warming, the low frequencies of occurrence for words related to wildlife (e.g., whale, bee, frog, fish, bird, reef), and the more frequent use of neutral and negative words (e.g. risk, cause, issues) than positive ones (e.g. reafforestation, healthy, promote, ratify). The concgrams were also grouped by cause, effect, time and places/people. These results provide a snapshot of the language use by the media outlets in construing climate talk in the 2010s. Alternative words to approach climate talk in the 2020s will be discussed. Keywords: ecolinguistics, climate talk, climate change, global warming, concgram