THE RATIONALITY of a belief or action is a matter of its responsiveness to the reasons for and against, and of the process by which those reasons are generated. Why does rationality involve reasons? Here is one answer: beliefs and actions are to have certain properties (such as truth or satisfying desire), and it is more likely that they will if they are responsive to all the reasons for and against. (Might some other process that does not involve considering or weighing reasons at all be even more reliable in achieving that goal?) Whether or not considering reasons is the most effective or the most reliable method for achieving our cognitive goals, why is it effective at all? What connects reasons for and against to these goals? What makes something a reason? Among the welter of information there is, what constitutes something as a reason for or against a belief or an action?
We come to hold beliefs through some general process for arriving at, maintaining, and revising beliefs. For different kinds of beliefs, or even for the same kind on different occasions, we may employ different processes. Following a particular process upon a particular occasion may lead us to a belief that is true; using that process may (invariably or probabilisticly) cause us to believe the truth. The utilization of reasons can play a role in a process's being a probabilistic cause of believing the truth, and differences among reasoning procedures can affect the process's efficiency or effectiveness. It seems plausible to say that believing h for reason r will be conducive to reaching our cognitive goal of true belief only when there is some connection between the truth of r and the truth of h. It is this connection between reasons and the truth of what they are reasons for that explains the connection, such as it is, between believing for reasons and believing the truth. What, then, is the nature of the connection between reasons and what they are reasons for?
Concerning reasons for a belief, the philosophical literature contains two views. One, the a priori view, holds that a reason r for an hypothesis h stands in some relation R to h, such that the faculty of reason can