Conflict and violence are crucial phenomena for understanding human action. Yet, they remain remarkably difficult to define. In a pioneering reader on writings about violence, Nancy Scheper-Hughes and Phillipe Bourgois (2004: 1) summarized that “violence is a slippery concept — nonlinear, productive, destructive, and reproductive” (their emphasis). They admit that despite their extensive efforts to compile a broad and eclectic anthology, “in the end we cannot say we ‘know’ what violence is. ‘It’ cannot be objectified and quantified so that a ‘checklist’ can be drawn up with positive criteria for defining each particular act as violent or not” (p. 2). Scholars in linguistic pragmatics, sociolinguistics and related fields, however, are accustomed to engaging with complex, hard-to-define phenomena. Often, such problems can be analyzed contextually, using tools like participants’ reflexive discourse (Silverstein 1993), contextualization cues (Gumperz 1982), and the mutual construction of meaning and orientation (Duranti 1997: 314–321; Hanks 1996: Chapter 1). For instance, any utterance can in principle be contextually violent (i.e., injurious, aggressive, racist). While it is impossible to create an infallible, universal checklist of what constitutes an offensive utterance, elements such as the target’s uptake, subsequent actions of the participants, and their metapragmatic commentary can help us grasp the possible violence at stake.