Search

Frame shift is a cross-linguistic phenomenon in translation which results in corresponding pairs of linguistic material evoking different frames. The ability to predict frame shifts enables automatic creation of multilingual FrameNets through annotation projection. Here, we propose the Frame Shift Prediction task and demonstrate that graph attention networks, combined with auxiliary training, can learn cross-linguistic frame-to-frame correspondence and predict frame shifts.

In this paper, Watson, Meehan, Lynch, Nave, and Dennis discuss their investigations of how video evidence is used in criminal trials for police officers charged in on-duty shootings. Although still ongoing, their study has already yielded several lessons worth sharing. The authors consider how making sense of the violence seen on video is no simple task. They conclude that to understand videos of police violence they needed to turn to courts of law rather than directly analysing videos of police violence. In this sense, their research led to an investigation of investigations.

In this article we present an ethnomethodological study of a controversial case of ‘friendly fire’ from the Iraq War in which leaked video footage, war on video, acquired particular significance. We examine testimony given during a United States Air Force (USAF) investigation of the incident alongside transcribed excerpts from the video to make visible the methods employed by the investigators to assess the propriety of the actions of the pilots involved. With a focus on the way in which the USAF investigators pursued their own analysis of language-in-use in their discussions with the pilots about what had been captured on the video, we turn attention to the background expectancies that analytical work was grounded in. These ‘vernacular’ forms of video analysis and the expectancies which inform them constitute, we suggest, an inquiry into military culture from within that culture. As such, attending to them provides insights into that culture., + Sprache: eng