It is a truism that semantic concepts (concepts of meaning, denotation, reference and even truth, etc.) are relative to language. I distinguish two kinds of them in accordance to their relativity to language L; the relativity is either explicit (“true in L”), or implicit (“true L”). If language is explicated, the concepts of the former kind can be easily explicated in a plausible way and we resist semantic paradoxes. In the case of the latter concepts, the explication is also accessible and paradox-free. One can find then new interesting facts concerning famous Tarski’s theorem.