t is not enough that one mental event cause another mental event in virtue of its propositional content. Someone who engages in rational inference must recognize the correctness of the principle of sound reasoning, which one applies to one's inference. Modus Ponens works, affirming the consequent does not. Our inferences are supposed to be governed by the rules of reasoning we recognize to be correct. However, can these rules of inference ever really govern our reasoning process? According to physicalism, all of our reasoning processes are the inevitable result of a physical substrate that is not governed by reasons. ¶ So we might ask this question: "Which laws govern the activity we call rational inference?" We might stipulate, for the purposes of this discussion, the idea that laws of physics are accounts of the powers and liabilities of the objects in question. If the materialist claims that laws other than the laws of physics apply to the assemblage of particles we call human beings, then those particles are not what (mechanistic) physics says they are, and we have admitted a fundamental explanatory dualism. If however, the laws are the laws of physics, then there are no powers and liabilities that cannot be predicted from the physical level. If this is so there can be a sort of emergence, in that the basic laws governing a sleeping pill will not mention that the pills tend to put you to sleep. Nevertheless, the pill's soporific effectiveness can be fully and completely analyzed in terms of its physical powers and liabilities. If this is so, then we will be rational if and only if the physical configurations of matter guarantee that we are physical, and in the last analysis, the laws of logic do not govern our intellectual conduct.
"THE ARGUMENT FROM REASON" IN THE BLACKWELL COMPANION TO NATURAL THEOLOGY, WILLIAM
LANE CRAIG AND J.P. MORELAND, EDS. (WILEY-BLACKWELL: 2009), PP. 379-80.