A common view among ancient historians about Roman attentiveness to children’s psychological development needs reconsideration. The view holds that Romans ignored children’s cognitive ontogeny, perceiving early childhood as a largely undifferentiated life stage. A separate but related issue is the problematic claim that Roman childhood was entirely a matter of social construction. I present evidence from over four hundred years of Roman writing to make three points against Roman neglect of children and radical social construction. First, a consensus (granted our limited number of sources) about “developmental psychology” prevailed from the late republic to late antiquity. Second, this consensus is consistent with modern findings and theories. In particular, the Roman material maps onto a developmental timeline featuring a “2-month revolution,” when the infant engages in dyadic protoconversations with adults, and a “9-month revolution,” when the infant shares attention triadically with adults toward third objects. In the Roman consensus, both dyadic protoconversations and triadic joint attention were central to cultural learning. Third, when the Roman consensus about children’s minds did change, it was because of the Christian doctrine of original sin, according to which even newborns were psychologically corrupt. Nonetheless, I argue, this rupture in the Roman consensus does not entail that we should suppose childhood to have been entirely a social construction or merely a projection of a religious or other ideology. Rather, any construction of childhood–even the late-antique Christian one–will necessarily rest upon a foundation of, and thus be constrained by, certain inescapable biological and psychological “givens.” (PsycINFO Database Record (c) 2019 APA, all rights reserved)