The Argument from Reason is a name given to what is actually a family of arguments. The common thread is arguing that some philosophical position (usually Naturalism or Physicalism) is incompatible with the (supposed) reality of rational inference. In this thesis I analyze versions of the argument from reason presented by William Hasker and Victor Reppert, ultimately concluding that while their arguments are at best incomplete there is a plausibly sound version of the argument which I go on to defend. After laying some groundwork, I define physicalism and microphysicalism and present the latter as the most plausible target of my version of the argument. I then distil two distinct versions of the argument from reason as found in the works of Hasker and Reppert. The first I call the ontological argument from reason which argues that various features or necessary conditions of an event's being a case of rational inference (intentionality, mental causation, truth, unity of consciousness, or the reliability of our rational faculties) are incompatible with microphysicalism. I conclude that their arguments in each case either add nothing to the ongoing debate on each issue, are wholly misguided (particularly in their understanding of the supervenience relation), or at best incomplete with no clear road forward to resolution. I then consider what I call the epistemological argument from reason by Hasker and argue with him that in order for rational inference to justify our beliefs it must be the case that logical relationships between propositions are able to play a causal role in the forming of the inference. I argue that microphysicalism's ontological commitments preclude logical relationships from playing such a role and as such they preclude justification via rational inference. I then consider possible objections and replies, chief of which is the intelligibility of logical relationships playing any causal role. In response I argue that the causation involved is not mechanistic/efficient and then offer as an alternative a brief sketch of a non-reducible conscious linguistically capable agent. I contend that such an agent would be capable of grasping logical relationships in order to make them suitably causally efficacious.