Four experiments tested the idea that people distinguish between biconditional, conditional, and enabling intention conditionals by thinking about counterexamples. The experiments examined intention conditionals that contain different types of reasons for actions, such as beliefs, goals, obligations, and social norms, based on a corpus of 48 intention conditionals established through an extensive materials test (n = 136). Experiment 1 (n = 19) showed that retrieved alternative reasons suppress the "affirmation of the consequent" and "denial of the antecedent" inferences from conditional intentions, whereas retrieved disabling reasons suppress the "modus ponens" and "modus tollens" inferences from enabling intentions. Experiment 2 (n = 61) showed that the suppression effects also occur for explicitly provided alternatives and disablers, for a large corpus of 80 intention conditionals. Experiment 3 (n = 60) showed that the suppression effects also occur for unfamiliar content, for which participants cannot rely on prior knowledge or beliefs about probabilities. Experiment 4 (n = 26) showed that participants retrieve alternatives and disablers readily for intentions just as they do for causal conditionals. The implications of the results for alternative accounts based on possibilities and probabilities are discussed.