If practical reasoning deserves its name, its form must be different from that of ordinary (theoretical) reasoning. A few have thought that the conclusion of practical reasoning is an action, rather than a mental state. I argue here that if the conclusion is an action, then so too is one of the premises. You might reason your way from doing one thing to doing another: from browsing journal abstracts to reading a particular journal article. I motivate this by sympathetically re-examining Hume's claim that a conclusion about what ought to be done follows only from an argument one of whose premises is likewise about what ought to be done.