The paper defends the idea that when we evaluate whether agents deserve praise or blame for their actions, we evaluate both whether their action was intentional, and whether it was voluntary. This idea can explain an asymmetry in blameworthiness and praiseworthiness: Agents can be blamed if they have acted either intentionally or voluntarily. However, to merit praise we expect agents to have acted both intentionally and voluntarily. This asymmetry between demands of praise and blame offers an interpretation of the Knobeeffect: in the well-known experiment people blame a company chairman because, although he harmed the environment unintentionally, he did so voluntarily. In turn, praise is withheld, because the chairman did not benefit the en...