Peter Newmark denies the possibility of a science of translation and the existence of a single method of translating. This article argues against this vision. On the basis of a number of examples, the author endeavours to show that the method best applied in translating is - or should be - on and the same, regardless of whether, at the re-expression stage, the translator chooses to follow the semantic or communicative or literal or any other approach. The author also tries to prove that the method itself provides the criteria for giving partial or total preference to any specific approach or combination thereof.