Tomasello's target article addresses the developmental origins of obligation in children's participation in shared goals and joint plans of action. However, shared intentionality and joint action depend on the ability to construe behaviors as actions, as rational and reasonable. Rationality is also at the heart of the origin story of obligation. As noted in the target article, an obligated individual must be a rational agent with “the capacities to make and regulate oneself by normative judgments about what attitudes are warranted” (Darwall Reference Darwall2013c, p.47). This idea of warranted attitudes provides important conceptual resources for the construction of obligation. Being warranted is already a kind of normative relation.

From a very early age, infants make sense of behaviors using something like a rational frame (Gergley & Csibra Reference Gergley and Csibra2003). Agents have reasons to do things. If the goal is to reach an object, an agent has a reason to move toward that object. To do otherwise is a mistake, a violation of some kind. By the time children can talk, they justify and explain in terms of rational action. Of course the person seeking their dog will look where it is: it would be silly to do otherwise (Bartsch & Wellman Reference Bartsch and Wellman1995). Indeed, children's appreciation of humor is a key marker of this perspective. Nothing is funnier than to see an adult (intentionally) do something wrong. Talking into a shoe like it is a phone is nonsensical, but only in relation to a rational, correct way to use a shoe.

Children's early conceptions of rational action are more objective, depersonalized, non-representational than are adults’. Reasons are not qualities of individuals as much as they are features of situations. Perner and Roessler (Reference Perner, Roessler, Aguilar and Buckareff2010) described this as a “teleological” perspective (see also Gergley & Csibra Reference Gergley and Csibra2003; Kalish Reference Kalish2005 on “natural status”). The dog-seeker has a reason to look under the bed, because that is where the dog is. A more sophisticated conception of action personalizes reasons. Actors have reasons “in their own lights,” as they represent the world to be. The person who does not know the dog is under the bed has no reason to look there. The philosophical literature distinguishes “justifying reasons” used to evaluate actions from “explanatory reasons” used to causally account for actions. Young children seem prone to take justifying reasons as explanatory. This is often characterized as a form of egocentrism. If an act makes sense from the child's point of view, it must make sense for the actor as well.

At first blush, the hypothesis that children's conception of rational action moves from general (“the reason”) to second-party (“their reason”) seems exactly contrary to Tomasello's account of the dyadic then later generalization of obligation. However, the kind of normative commitment involved in rational action is not obligation. Indeed, it may be the personalization of reason that establishes the dyadic context of obligation (and vice versa).

The big problem with children's early egocentric construction of rational action is conflict. Classically, the focus has been on conflict between representation and reality (e.g., false belief). Conflict between perspectives is equally challenging, although only certain kinds of incompatibility actually conflict. People have different goals, reasons to do different things. Even in a clear zero-sum situation – say, two people in a boat desiring different destinations (Rakoczy et al. Reference Rakoczy, Warneken and Tomasello2007) – there is no necessary incompatibility of perspective. One can think, “I have a reason to go to X, and you have a reason to go to Y.” Only one goal will be satisfied, but that is commonplace: goals frequently go unfulfilled, plans are thwarted. That you have a reason to go to Y is accidentally, empirically incompatible with my reason to go to X. A deeper conceptual conflict arises only after we have established a common goal, a joint intention. If we have agreed, “We will go to X together” and you steer the boat to Y, now there is a problem.

The problem of coordination underlies Tomasello's account of the origin of obligation. Obligation is a cognitive mechanism to establish, maintain, and enforce shared goals and action plans. In this role, obligation provides a kind of answer to the question of “What went wrong?” when coordination breaks down. Especially in efforts to restore coordination, obligation focuses attention on the mental processes involved in an other's behavior. Recognizing that someone is doing something wrong or silly entails understanding “warranted attitudes,” but does not necessarily involve understanding what causes wrong or silly behavior. This is just an evaluation. Obligation is part of a causal story. To understand someone as obligated is to understand them as responsive to reasons in a certain way. Reasons have to cause (or at least motivate) their behavior. That is why citing the obligation is an effective way of restoring coordination. I remind you of the reason, and expect you to adjust your behavior in light of it.

It may be in contexts of shared intentionality that children start to think about the causal processes involved in acting on reasons. When coordinating with another, a child needs to track and repair specific individual attributions of goals, and reasons. It is where the observation, “Sometimes people don't do what they should” turns to “Why don't they?” Participating in joint action requires concern with maintaining shared perspective. The depersonalized teleological or justifying understanding of reasoned action (people do what makes sense) becomes enriched with a personal causal/explanatory understanding.

Tomasello's target article has identified a critical nexus in social cognitive development. The construction of a situational and dyadic notion of obligation draws on an earlier general and depersonalized notion of “acting for a reason.” In coming to understand obligations, children are revising their understanding of action and moving to a more fully representational theory of mind. Understanding people as acting for reasons always involves a tension between a general, justifying “what makes sense” form, and an individual, explanatory “what they think” form. Obligation in joint action turns out to be a critical meeting point of these two forms both conceptually and developmentally.