Traditional database query languages are based on set theory and crisp logic. Many applications however, need similarity of retrieval like queries producing reslults with truth values from the interval [0,1]. Such truth value can be regarded as continous membership values of tuples experssing how strongly a query is matched. Formulating queries, we generalize the classical realtional domain calculus by incorporating fuzzy operations and user weights. Besides defining syntax and semantics we show how to map any calculus experssion onto a corresponding similarity algebra experiments. In this way we present a theoretical foundation for a declarative query langauge combining retrieval functionality an traditional relational database