This paper discusses theoretical and methodological problems associated with defining the linguistic norms which emerge from the prescriptive tradition of linguistic research. One acknowledged deficiency of this tradition is the difficulty it has in dealing with linguistic variation and change (section I.). The article will focus on the fundamental distinction between the static concept of explicit standard norms which ignore linguistic variation and change and the dynamic concept of implicit non-standard norms such as those of bilingual varieties which involve such variation and change (section II.). Section III. argues for a specific term of 'code-switching norms' which is able to simultaneously consider the principles of linguistic change and variability as well as that of inherent variability. Finally, it discusses methodological implications for a sampling model.