This paper presents a technique for interpreting output demands by a natural language sentence generator in a formally transparent and efficient way. These demands are stated in a logical language. A network knowledge base organizes the concepts of the application domain into categories known to the generator. The logical expressions are interpreted by the generator using the knowledge base and a restricted, but efficient, hybrid knowledge representation system. This design has been used to allow the NIGEL generator to interpret statements in a first-order predicate calculus using the NIKL and KL-TWO knowledge representation systems. The success of this experiment has led to plans for the inclusion of this design in both the evolving Penman natural generator and the Janus natural language interface.