"Theories without facts may be barren, but facts without theories are meaningless"

We have now indicated the broad field of fact within which economic analysis functions. We have yet to define the specific task of economic analysis itself. The purpose of any analytical treatment of material is to provide a body of principles according to which facts can be selected and interpreted. The complaint is frequently heard that people want "facts", not "theories". The complaint may be justified in protest against theories which have no basis       in fact, but usually it arises from a misunderstanding of the true relationships of facts and theories. Theories without facts may be barren, but facts without theories are meaningless. It is only "theory"- i.e., a body of principles - which enables us to approach the bewildering complexity and chaos of fact, select the facts significant for our purposes, and interpret the significance.

Indeed, it is hardly too much to claim that without a theory to interpret it there is no such thing as a "fact" at all. It is a "fact", for instance, that Oliver Cromwell had a wart on his nose. But what constitutes this supposed "fact"? To the chemist it is a certain conglomeration of atoms and molecules. To the physicist it is a dizzy mass of unpredictably excitable electrons. To the biologist it is a certain impropriety in the behavior of cells. To the psychologist it may be the key to the interpretation of Cromwell's character, and a fact of overwhelming importance. The historian may consider it an insignificant detail or an important causative factor, according to whether he follows economic or psychological interpretations of history. To the economist the wart may be of negligible importance unless Cromwell were prepared to pay a good round sum for its removal. What, then, is the "fact" about the wart? It may be any or all of the above, depending on the particular scheme of interpretation into which it is placed.

Kenneth Boulding, Economic Analysis, 1941 - "The Task of Interpretation"