The behavioral processes through which people react appropriately to verbal descriptions remain poorly understood. I argue here that these processes are Pavlovian. Common objections to a Pavlovian account of symbolic behavior evidence a lack of familiarity with the relevant data or misunderstandings of operant theory. Although much remains to be done to explore the relation between verbal understanding and simpler forms of Pavlovian conditioning, the similarity of these two sets of phenomena has received increasing support in recent years. The difficulties of operant principles in accounting for symbolic behavior have a more general implication: Operant models, which focus on the maintenance of responding, must incorporate principles of behavioral induction that explain the origins of novel environment-behavior relations. Pavlovian processes are likely to contribute.