U.S. weapons transfers make up
the lion’s share of the global arms trade.
Increasing political pushback against arms deals with countries like Saudi Arabia suggest an interplay between public discourse, national politics and consequently arms trade. Nonetheless, little is known about the kind of discourse that accompanies such deals.
This project looks at Reddit discourse from 2008-2021 alongside public data on U.S. arms transfers during the same period to fill this gap. We analyze themes of discourse using a topic model and the framing of them using a novel neural network classifier. In particular, we look at the relative moralization of discourse about weapons transfers, as moral frames have been
shown to affect perceptions of bias and hence empathy in contentious domains.