ABSTRACT
Conference of the Parties (COP) meetings on climate change are opportunities for social movements and other non-state actors to engage in climate change communication and mobilization. We focus on Instagram as an online arena of climate communication and mobilization during COP26, held in Glasgow. Instagram is a distinct arena for eco-political communication and activism because of its visual focus and generally younger user base. Using a Discourse Network Analysis approach, we analyse 2417 posts to examine relationships and alignments across imagery, discourses, and actors. Instagram serves as a space to articulate critical counter-discourses of climate justice, Indigenous rights, and individual action as a response to perceived failures of COP processes and formal agreements. At the same time, images of celebrities and politicians structure much of the Instagram discourse network. This highlights how Instagram contributes to a celebritization of climate politics, with individual political actors positioned as climate heroes or villains.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1 Future research in this area could take a more collaborative approach to retrieving material from researchers located in multiple regions and using diverse languages to further mitigate the potential for unintended sampling bias based on location or language settings. A collaborative approach would also facilitate data collection across a broader range of time-zones.
2 Social network analysis methods can also consider three or more modes, though this is challenging in practice (Knoke et al., Citation2021).
3 Discourse Network Analysis often examines ties based on agreement or disagreement between actors and discursive themes. This reflects the evolution of DNA as a network-based method for studying policy debate where agreement and disagreement are important bases for congruence networks that identify advocacy coalitions or other subgroups that share policy orientations (Leifeld, Citation2017). Ties between actors and themes can also be analysed through co-references to the same topics/imagery as affiliation networks that are not dependent on concept agreement or disagreement (Leifeld, Citation2020). The latter approach is appropriate for our purposes, as we are more concerned with affiliations among imagery, discursive themes, and organizational actors, rather than with analysing congruence or conflict networks among actors.
4 Centrality was calculated using the percentage degree function in Visone. For the two-mode networks, centrality was calculated separately for each of the modes of data.
5 We did not implement normalization or standardization for calculating tie weights. As coding was done with individual posts as the unit of analysis, we did not have potential issues of overcounting duplicate actors or themes within the same post.
6 Likewise, we did not see evidence of posts taking an explicit climate denial or delay position. In addition, there were no counter-movement actors in the dataset. While climate skeptic or oppositional actors and discourses were likely mobilized around COP26, it appears they were marginal to the #COP26 hashtag to the point they fell below our threshold for sampling inclusion.
7 At the same time, we need to be cautious about the openness and accessibility of social media platforms like Instagram. Social media algorithms privilege certain actors in users’ timelines. Conversely, there remain persistent urban-rural and other digital divides in who has high-quality internet access, and thus access to these social media platforms (e.g. Weeden & Kelly, Citation2021).