Ancient writers, including philosophers such as Aristotle, often depict friendship as a source of pleasure; by contrast, in his Laelius de amicitia, Cicero describes such relationships as sweet and delightful, but never connects them with uoluptas, which for him is a largely negative term reserved for Epicurean doctrine. This paper argues that there is more to this pointed use of language than Cicero’s well-known dislike of Epicureanism. Considering first the Latin philosophical vocabulary of pleasure and then the vexed question of what exactly qualifies as pleasure according to the Epicurean system, the paper makes the case that Cicero believed (probably correctly) that the pleasures of friendship as conceived of by himself and many ordinary language-users would not in fact qualify as instances of Epicurean uoluptas. If, as Epicurus appears to have held, all pleasures are either bodily or mental, and all mental pleasures are derived from bodily ones, then many activities experienced as pleasurable in and of themselves—including many traditional elements of friendship—are not in fact Epicurean pleasures.