Abstract: Inequalities in human social groups contribute in important ways to the initiation and perpetuation of violence. Because both social inequalities and interpersonal aggression emerge in early life, it is important to study their developmental origins in the collective and individual behavior of young children. As young as 2 years of age, children assemble stable, linearly transitive dominance hierarchies when brought together in novel social groups. Just as the consequences of socioeconomic status may be due in part to experiences of social ordering per se, subjective childhood experiences of occupying a lesser or greater position on a scale of social influence may similarly affect health, safety, and behavior in early development. Children's experiences with social ordering may represent a first, formative encounter with the hierarchical social relations that affect health and susceptibility to violence over the human lifespan. We have studied the health correlates of group stressors, dominance positions, and biobehavioral reactivity within cohorts of preschool children and semi free-ranging rhesus macaques. Social position is ascertained using naturalistic observations of critical dyadic events, and stress-reactivity is measured using assessments of behavioral and biological responses to standardized challenges. Outcome measures, including indicators of mental and physical health and interpersonal injuries, have been assessed—in the case of children—with parent- and teacher-report questionnaires, child self-report protocols, and frequent, objective physical examinations of the child and—in the case of rhesus monkeys—with daily observations of violence-related injuries. Results to date have suggested the following provisional conclusions: (1) Children and monkeys form stable, linearly transitive social hierarchies in which identifiable subgroups of individuals occupy dominant and subordinate social positions. (2) Individuals occupying subordinate positions within the social group show exaggerated adrenocortical and/or autonomic reactivity to challenge and disproportionate rates of chronic medical conditions or violent injuries. (3) Stress-reactivity, naturally occurring stressors, and social position are interactively predictive of rates of morbidity. Monkeys who displayed high biobehavioral reactivity sustained disproportionate numbers of violent injuries during a confinement stressor. Children who were high in the dominance hierarchy and low in reactivity showed significantly higher rates of externalizing mental health symptoms, while those who were in subordinate social positions and high in reactivity showed higher prevalences of internalizing symptoms. In conclusion, the prevention of youth violence—and other forms of early disorder—will require a deeper, developmental understanding of the “headwaters” of aggression and victimization in early childhood and a stronger accounting of how early social inequalities set trajectories toward healthy or disordered behavior.