Carnap’s analysis of the language of science had presupposed too close a connection between the semantics and testability. The core problem of the logical empiricist tradition was to show how to provide the interpretation of theoretical terms and hence the explanation of their application to observable entities by means of observation terms. It is argued that the utilization of a much more expressive semantic theory which identifies meanings with hyperintensional entities leads to a clarification of the competencies of semantics and methodology. It is claimed that the observability or unobservability of term’s referents does not determine the meaning of the term. Carnap’s reduction sentences are therefore conceived only as methodological devices for the determination of unobservables via the theory-presupposed observable aspects.