We have one more paper session on the final day of the Social Media & Society conference in Glasgow, and it’s a panel coordinated by the great Risto Kunelius that I’m also presenting a paper in. This panel presents outcomes from the CLOUD-C project, which explored discussions around the COP26 (in 2021, which of course took place in Glasgow, too) and COP27 (in 2022, in Sharm El-Sheik) UN climate change summits from a number of perspectives.
Risto kicks us off by introducing the panel itself: data for the project were gathered from Twitter using a set of keywords related to the two summits, and our interest here is in the political debates happening on social media around these events. This explores, then, the accountability function of social media platforms for such events – but it is also a function that no longer exists in this way, since the Twitter that enabled it also does not exist any more. One question emerging from our work on these case studies, then, is how any of this – the political debates themselves, and our analyses of these political debates – might happen now. Results from this work will be published soon in the International Journal of Communication.
Risto’s introduction is followed by Dayei Oh, whose interest is in the triadic polarisation in climate discourse around these events. Often, polarisation is seen as dyadic only: between climate change believers and sceptics. But there are additional divisions within the climate community: between more radical climate activists and more conservative climate bureaucrats, in addition to the climate change deniers.
All of this is overlaid with broader culture war logics, which also introduce other concerns (gender relations, postcolonial struggles, ‘wokeness’, etc.) into climate debates. This paper explored this by applying Louvain modularity algorithms to the Twitter networks in order to identify three broad camps in the two cases, and then conducted topic modelling and sentiment analysis to determine their discussion themes.
For COP26, this identified radical climate activists (24%), radical climate contrarians (22%), and mainstream-moderate actors (40%); for COP27, those distributions changed somewhat (26%, 23%, 37%), and community structures changed as well. The mainstream community shrunk, then, while the radical groups expanded; in COP27, the emphasis shifted from climate activism itself towards broader human rights and related issues.
Topics divided into two common clusters: anti-elite skepticism, and the culture war-ification of climate debates. Anti-elite skepticism is used by both radical groups, but they attack the moderates in different ways. Culture war topics are predominantly used by contrarians, but activists then also push back on these issues.
Negative emotions such as disgust and fear were key emotions for these radical groups, though they are differently distributed across activists and contrarians. Amongst climate activists, negative emotions are substantially greater at COP27. This is also linked to an increase in moral rhetoric amongst radicals on both sides, highlighting the ‘evil’ of the other side. Insulting and toxic language did not increase from one event to the other, but the targets shifted: in COP26, they both attacked the moderates, while in COP27 they attacked each other.
This shows the value of the triadic model of climate debates, and the changes in how the components of that triad interact with each other. One key question here is whether the centre can hold its epistemic authority, given the way it is attacked from both sides but for different reasons; another is whether like so many other topics climate change debates are now overwhelmed by broader culture war battles which are driven by moralising ‘us vs. them’ debates. One response to this might be to cultivate more emotionally intelligent strategies.
