Online conflict tends to advance social movements rather than undermine them, according to new research from McGill's Desautels Faculty of Management.

“Anyone who has watched an online debate turn heated assumes arguing kills collective action,” said Emmanuelle Vaast, Professor of Information Systems and the study's senior author. “Our study shows the opposite: the same social media dynamics that produce conflict also enable millions of strangers to discover they are telling the same story and act together. It explains how a teenager’s solo school strike could become a worldwide mobilization in a matter of months.” 

The study was carried out in collaboration with colleagues at Northeastern University and the University of Massachusetts.  

The researchers examined social media activity related to the 2019 global climate strikes sparked by Swedish activist Greta Thunberg – the largest climate mobilization in history – to understand how distinct causes, from pipeline opposition to the protection of monarch butterflies and wolves, fused into a single global “movement of movements.” They found that this process happens in three phases: exposure (movements discover each other’s issues), entanglement (conversations become tangled and contentious) and coalescence (participants converge on shared goals and joint action).  

“Conflict did not weaken the movement,” Vaast said. “Co-operation and disagreement evolved together, like the two strands of a double helix. Arguing was part of how a shared agenda took shape.” 

Previous research has mostly examined single movements and treated online conflict as a threat to collective action, the researchers said. 

The team used computational text-analysis methods to analyze about 1.9 million Twitter messages posted between April and September 2019. Vaast noted that by combining semantic network analysis and natural language processing at this scale, the team could observe movement dynamics and identify patterns that may have been obscured by other methodologies, such as interviews or case studies. 

Although social media platforms have shifted since 2019 from largely chronological feeds to recommendation-driven ones optimized to hold users' attention, the researchers said this does not necessarily impede movement building. 

“Recommendation algorithms can actually accelerate discovery. For instance, TikTok surfaces [social] causes to people who never sought them out. So, coming together on social media is still possible today. It just runs through more opaque mechanisms,” Vaast said. 

“That's exactly why the 2019 moment is worth studying: it's a benchmark for what today's algorithms are enabling.”  

“Entangling and Coalescing on Social Media: A Study of Issue Boundary Spanning in Movement of Movements,” by Carol Lee (Northeastern University), Pratyush Bharati (University of Massachusetts) and Emmanuelle Vaast (McGill), was published in Information Systems Journal.