A good deal of attention has been given in recent philosophy of religion to the question of whether we can sensibly attribute to God a form of knowledge which the 16th-century Jesuit theologian Luis de Molina termed "middle knowledge". Interest in the doctrine has been spurred by a recognition of its intimate connection to certain conceptions of providence, prophecy, and response to petitionary prayer. According to defenders of the doctrine, which I will call "Molinism", the objects of middle knowledge are (...) all the true counterfactuals of the form, 'If C were to occur, then S would freely do A', where C specifies a particular set of circumstances. (Such propositions are usually referred to as "counterfactuals of freedom" - hereafter "CFs".)' I will try to assess the current status of two distinct questions: (1) Are there true CFs? and (2) Could God know true CFs if there are any? Obviously, Molinism must answer both questions affirmatively. Most of the discussion has focused on the first question. I hope to show that this approach has been a dialectical mistake on the part of critics of Molinism, for the difficulties of giving an affirmative answer to question (2) are, I believe, far easier to state clearly and convincingly. (shrink)