Everyday life is full of situations in which we act not only as individuals, but also with others, as parts of informal groups or structured collectives. For example, a group of friends spend a day hiking a trail together, a guardian offers a hand to stabilize a young child’s tentative first steps, or an academic department draws up policies that facilitate student instruction. Philosophers call this category of action “joint action,” and agree that it is an important and distinctive way in which we exercise our agency. However, there has been a tendency in the philosophical literature to explain the phenomenon by analyzing highly structured and cooperative cases of joint action like two or more agents painting a house together or moving a heavy piece of furniture together. On these analyses, two or more agents are engaged in joint, rather than individual action when they jointly and intentionally pursue goals that they have in common, all while abiding by principles of rationality which demand that they act in ways that are both conducive to achieving the shared goal and responsive to the actions of their co-participants. In this dissertation, I argue that these standard analyses of joint action do not provide us with a comprehensive understanding of the phenomenon, because they ignore important cases of joint action that do not fit into the highly structured and cooperative paradigm. To address this gap in the literature and bring us closer to a comprehensive understanding of joint action, my dissertation uses case studies of the speech act of telling and early childhood development to show that joint action also occurs in situations which involve little to no cooperation or strongly shared goals. Instead, as we see when we investigate these atypical cases, joint action is consistent with competition and antagonism, and does not require participants to have the sophisticated theory of mind needed to track and adjust to the goals and actions of their co-participants in a way that coheres with principles of rationality.