Support (one object preventing another from falling) is linguistically encoded by adults and children in a highly structured and differentiated way, with basic locative expressions or Light verbs (e.g., in English, the block is on/put on the box) encoding Support-from-Below, and lexical verbs (e.g., she stuck the block on the box) encoding Mechanical Support. We ask whether parents use such highly differentiated language when describing support configurations to their young children and whether the age of the child moderates this effect. We found that when parents of children 6-months to 3.5 years of age described events of Support-from-Below and Mechanical Support to their children, they did use distinct linguistic expressions, encoding Support-from-Below with Light verbs (e.g., put on) and Mechanical Support with other lexical verbs (e.g., stick on/to). This differentiation was especially prominent for parents of the oldest children, who used more lexical verbs (stick), and fewer Light verbs (put), to encode Mechanical Support. These findings are the first to demonstrate that parents’ spatial language to their young children distinguishes between two distinct types of support and suggest that parent input may play an important role in helping children learn to linguistically carve up the semantic domain of support. This sets the stage for future research to test the causal connection between the two.