Rapid Mobilisation and Organisational Fragmentation: The Telegram Ecosystem during the 2026 Italian Student Protests
Lorenzo Alvisi, Maurizio Tesconi
Abstract
In October 2026, social media pages named Italia in Piazza used several platforms to call for demonstrations in Italian cities. The mobilisation brought participants into a rapidly assembled network of Telegram chats, but disagreements over its direction and protest tactics became visible after the first demonstrations. We study how this online environment formed and subsequently fragmented. We first constructed an ecosystem of chats connected to the protests, then compared it with established conspiracy, radical-left and radical-right ecosystems. Few links or observed participants connected the protest ecosystem to these communities. Its messages nevertheless expressed a broader mix of political positions, brought together by a shared focus on the protests. Within this ecosystem, geographically distributed discussions were linked through a small administrative core and connections to TikTok and Instagram. The mobilisation subsequently developed into two distinct spaces: Movimento 17 Ottobre and a re-formed Italia in Piazza chat. The former contained more content classified as promoting violence or hate, while the latter retained part of the original staff and advocated organised, non-violent protest. Their observed populations showed little overlap. After the original chats became inaccessible, discussion was rebuilt on Telegram, while Movimento 17 Ottobre encouraged migration to Signal. The case connects the fragmentation of a shared mobilisation with questions of deplatforming and community governance: the loss of discussion spaces need not end collective activity, while the environments that emerge can develop markedly different norms around protest and violence.
Create a lesson
Related papers
Constructing an Artworld Influencers Network by Mining Social Media
Amalia Foka
Theoretical relations between basic centrality measures in real-world networks
Mikhail Tuzhilin
No Note Is an Island: Probing LLMs' Understanding of Crowdsourced Fact-Checking Environments
Weijian Fan, Chichen Lin, Xiao Han et al.
Agent-Based Disruption Simulation on an Edge-Level Product-Annotated Supply Network of the Automotive Industry
Benjamin Rolf, Hiroyasu Inoue, Sebastian Lang
Similarity-Weighted Leiden-CPM Clustering of a 7.4-Million-Node Biomedical Citation Graph: Weight Alignment, Scale, and Sparsity
Hossein Mohasel Arjomandi
Building Weighted PubMed Citation Graphs at Scale: A Pipeline and an Audit of Edge-Similarity Features
Hossein Mohasel Arjomandi