Climate change and Twitter in Belgium

Pizarro Pedraza, Andrea [USL-B] Cougnon, Louise-Amélie [UCL] Catellani, Andrea [UCL] Samofalova, Yuliya [UCL] De Schutter, Olivier [UCL]

This report is part of the project 2o2cm Overcoming Obstacles and Disincentives to Climate Change Mitigation: A cross-cutting approach by human and social sciences, funded by Solstice (JPI Climate). The project “has the theoretical and transformative aim to improve the scientific understanding of obstacles and disincentives to climate change mitigation and to understand how multimodal devices can operate on them, through the interdisciplinary collaboration between linguistics and social psychological approaches”. This report belongs specifically to the WP2, which focuses particularly on the analysis of text and image from social media, such as Twitter, Reddit and Instagram. Here, we present the results of the linguistic analysis of Twitter data in French-speaking Belgium related to climate change. The goals of this analysis are to understand “ how climate change is depicted and communicated in Twitter ; what are the obstacles to the adoption of attitudes and behaviours of climate change mitigation: forms of resistance, deformation, and denial; and what textual sources are mostly read, cited and commented on by laypeople and authorities online.” We will tackle these questions from a linguistic perspective, and particularly with a combination of methods from corpus linguistics and discourse analysis (which will be explained in detail in section 2.). This will translate very specifically in the following objectives: Objective 1: Collect a corpus of Belgian French, Climate change-related tweets Objective 2: Analyze the distribution of the data across segments, time and particular accounts (most active, most mentioned) Objective 3: Analyze the presence and distribution of several important axes related to Climate change that have a crucial impact on individual life (food, transport, energetic efficiency) Objective 4: Extract and analyze the expressions of difficulty in the corpus at different linguistic levels.