This article revisits earlier proposals that language learning is, in essence, the learning of formulaic sequences and their interpretations; that this occurs at all levels of granularity from large to small; and that the language system emerges from the statistical abstraction of patterns latent within and across form and function in language usage. It considers recent research in individual differences, the psycholinguistics of language processing, and longitudinal studies of first (L1) and second (L2) language acquisition. The first section reviews studies of individual differences in phonological short-term memory (PSTM) and working memory (WM) and their correlations with vocabulary and grammar acquisition in L2. The second section summarizes evidence that language processing is sensitive to the statistical properties of formulaic language in terms of frequency and transitional probability. The third section examines the definition of formulas and formulaicity using different statistical metrics. The fourth section evaluates longitudinal research in L1 and L2 into the putative developmental sequence commonly proposed in usage-based approaches, from formula to low-scope pattern to creative construction. The final section weighs the implications of the statistical distributions of formulaicity in usage for developmental sequences of language acquisition. Zipf's law and the “phrasal teddy bear” explain the paradox whereby formulas seed language acquisition and yet learner language is formula-light in comparison to native norms.