Storm Over the Nile Understanding the Arabic Twitter Discussion on the Civil War in Sudan

This research, funded by KPSRL's Knowledge Management Fund, explores how the Sudanese conflict between the Sudanese Armed Forces (SAF) and the paramilitary Rapid Support Forces (RSF) has been discussed on Arabic-speaking X (previously known as Twitter), focusing on the extent and causes of polarization, the participation of bots, and the involvement of Sudanese and non-Sudanese users. The study analyses 139,487 tweets collected from April 2023 to the end of the year, manually coding a training set of randomly extracted 900 tweets. Machine learning techniques, particularly a neural network with GPT-4 embeddings, were employed to categorize the tweets into several topics and identify user origins and bot activity. Findings reveal that the X discourse on the Sudanese conflict is highly polarized, with users divided into pro-SAF, pro-RSF and pro-peace camps. Support for the RSF decreased over time, while pro-peace sentiments increased, highlighting a growing awareness of the conflict’s impact on civilians. Polarization is primarily driven by Sudanese users, who constitute 77.6 per cent of the identified accounts, while non-Sudanese users view the conflict through the lens of their own geopolitical interests. Bots account for 10.9 per cent of the total tweets, aligning with political agendas of Gulf countries, but their influence does not significantly alter the overall discourse trends. The study underscores the need for an alternative narrative on social media to strengthen the chances of a peaceful resolution of the conflict, focusing on empowering the active role of Sudanese civil society in shaping the narrative. The methodology also provides a replicable framework for analysing polarization in civil conflicts, offering insights into the dynamics of online discourse and the potential for machine learning in such studies.