Nowadays, it is not uncommon for a discussion, whether in real life or online, to go off the rails when participants strongly disagree. There is no shortage of examples concerning climate change, health, immigration and identity where people end up at loggerheads (or worse). In such disagreements, it seems as though the participants are unable to find their way back to a normal level of conversation.
To investigate what expectations language users have of conflicts, of resolving disagreements and of assistance (in the case of bystanders), we are conducting research into online de-escalation strategies.
How are discussions on X (formerly Twitter) resolved (for example, by webcare staff at large companies who respond to online complainants)? How do Wikipedians resolve disagreements on Wikipedia’s talk pages, and how do people reach agreement on subreddits?
In quasi-experiments, we examine how people react to online verbal abuse, how they themselves would react, and whether they would intervene if it happened to them or if they witnessed it happening.
We have developed an initial lesson plan for secondary schools to be used in Dutch or Social Studies lessons to discuss with the class how to resolve a dispute that has got out of hand. The lesson plan is called ‘De-escalation can be learnt.’ We are continuing to develop follow-up lesson plans: responding to visual memes, recognising fallacious arguments, etc.