Gilbert Harman argues that logic as a science of consequence relations (what he calls 'implications') is not the same thing as reasoning in the sense of a procedure for 'reasoned change in view'. Harman's position is an orthodox position among contemporary philosophers and logicians. In fact, Harman takes himself to be explicating what everyone already believes rather than arguing for his position. In this paper, I will demonstrate two things. First, I will show that a paraconsistent logician (someone who rejects ex contradictione quodlibet (ECQ) (A, ~A |= B for any A, B) as invalid) should reject Harman's position. Second, based on my analysis of paraconsistent logic, I will argue against Harman's position. If it is true that Harman's position represents a modern orthodoxy, its rejection has a number of consequences. I will briefly discuss some of them in the end.