Recently a number of philosophers have begun to promote what are broadly referred to as attributionist or real self views of moral responsibility. According these views a person is responsible for a thing just in case it is indicative or expressive of her judgments, values, or "world-directed" attitudes. These philosophers have focused a great deal of attention on dissolving the apparent tension between our commonsense intuitions concerning the connection between control and responsibility, on the one hand, and our lack of voluntary control over our values, beliefs and attitudes on the other. In attempting to relieve this tension, many of them have introduced various forms of non-voluntary control or agency we are said to exercise with respect to things such as our values, beliefs and attitudes. I argue that these supposed forms of non-voluntary agency are untenable because they typically rest on a failure to adequately distinguish between two ways in which we make up our minds; in short, they rest on a failure to adequately distinguish theoretical from practical reasoning. Once certain fundamental differences between theoretical and practical reasoning are brought back to the fore of the discussion, it becomes much harder to sustain some sort of unique species of agency that can be said to apply to beliefs and certain other world-directed attitudes. Without such forms of non-voluntary agency, however, proponents of attributionists accounts of moral responsibility seem to face a dilemma; they must either: sneak volition in through the backdoor or commit to holding people responsible for things with respect to which they are passive. The thesis falls into four main sections. In the first section, I introduce the problem by describing an ongoing debate between defenders of attributionist accounts of moral responsibility and defenders of what have been termed volitionist accounts of moral responsibility. In the second section, I explicate Pamela Hieronymi's construal of the form of non-voluntary agency she calls "evaluative control." In section three, I critique Hieronymi's account of evaluative control by pointing to two predominant points of divergence between theoretical and practical reasoning. In the fourth section, I examine the upshots of the absence of non-voluntary for attributionist accounts of moral responsibility; I do so by examining each horn of the dilemma mentioned above.