Can we evaluate a logic program declaratively? That is, can a logic pr
ogram be evaluated correctly and efficiently, independent of query mod
es and rule/predicate ordering, finding a complete set of answers, and
terminating properly? the answer could be ''yes'', at least for a goo
d subclass of logic programs, based on our investigation and experimen
tation using a deductive database approach. In this paper, an n-queens
problem, a classical logic program, is used as a running example to d
emonstrate the methodology. Our analysis shows that binding analysis a
nd constraint exploration are two essential issues in the realization
of declarative logic programming. The limitations of our methodology a
re also discussed in the paper. (C) 1998 Elsevier Science Inc. All rig
hts reserved.