Constraints on conventions: Resolving two puzzles of conventionality

Abstract

Conventions play a fundamental, yet contested, role in social reasoning from childhood to adulthood. Conventions about how to eat, dress, speak, or play are often said to be alterable, contingent on authorities or consensus, specific to contexts, and-thereby-distinct from moral concerns. This view of conventional norms has faced two puzzles. Children and adults judge that (a) some conventions should not be adopted and (b) some violations of conventions would be wrong even if the conventions were removed. The puzzles derive, in part, from the notion of "pure" conventions: conventions detached from non-conventional concerns. This paper proposes and examines a novel solution to the two puzzles, termed the constraint view. According to the constraint view, children and adults deem conventions as alterable within constraints imposed by non-conventional concerns. The present research focused on constraints imposed by concerns with agents to whom the norms apply and concerns with others affected by the norms. Findings from four studies with preschoolers and adults supported the constraint view. Participants evaluated actions and norms based on concerns with effects on agents and others, deeming conventions to be alterable insofar as the altered norms did not negatively impact agents or others. The constraint view offers a new framework for research on how children and adults integrate conventional and non-conventional concerns when they evaluate norms and acts.

Keywords: Conventions; Moral cognition; Moral development; Social domains.

Figure 1.

Visualization of the constraint view. Assume that there are five possible norms (e.g., about basketball team jerseys). Norms 3 and 5 have negative effects on the agent (e.g., does not solve the coordination problem efficiently), Norms 4 and 5 have negative effects on others (e.g., causes psychological harm), and Norms 2 and 3 are equivalent solutions to the coordination problem. The arbitrary choice between Norms 1 and 2 constitutes the conventional element of the decision. The dashed line indicates the constraints on this conventional element. For decisions deemed to be under personal jurisdiction (e.g., which ice cream flavor to pick), the introduction of any norm may be perceived as negatively affecting agents (Nucci, Killen, & Smetana, 1996). For other decisions (e.g., whether to use unprovoked violence against persons), there may be only one permissible norm. The latter norms would have no conventional element.

Figure 2.

The figure represents the relation between the constraint view and cross-domain coordination. Theorizing about cross-domain coordination (bottom) aims to explain how an individual balances existing conventional norms against non-conventional considerations, e.g., about agents and others. In contrast, the constraint view (top) offers an explanation of how an individual forms evaluations regarding the (possible or actual) conventions by themselves.

Figure 3.

Alterability of evaluations. The radius of each circle is proportional to the percentage of participants who said they would still negatively evaluate the act even if most people saw nothing wrong with the act. For comparison, the circle in the upper right corner indicates 100%.

Figure 4.

Acceptability of rules (Studies 2–4). The radius of each circle is proportional to the percentage of participants who said the rule was not okay (T = alternative variants, A = affect agents variants, O = affect others variants). The location of the center of the circle represents the proportion of participants who referenced effects on agents (y-axis) and effects on others (x-axis) when explaining why the rule was not okay.