Using computational text analysis applied to approximately 7 billion token words contained in over 38 million articles from the Web of Science (1900 – 2016), the authors explore the correlates of the diffusion of new scientific ideas.