The ensemble hypothesis proposes that language is but one of five cognitive capacities that separate human cognition qualitatively from other animal cognition as a result of their interactions. The ensemble consists of an episodic memory capable of mental time travel, mentalizing to augment social cognition, language for overt communication, advanced executive attention for governing working memory, and inner speech for thinking in the form of causal inference. The order in which each of these components arose in hominin evolution is addressed here. It is proposed that the flourishing of symbolic artifacts during the Upper Paleolithic occurred because, for the first time, all five components were in place and interacting in anatomically modern Homo sapiens.