Russell put forward his 1905 theory of denoting on the basis of evidence which he said is "derived from the difficulties which seem unavoidable if we regard denoting phrases as standing for the genuine constituents of the propositions in whose verbal expressions they occur."¹ What are these difficulties? In developing his theory of sense and reference, Frege had discussed the failure of the substitutivity of identity (SI) in intensional context and the related problem of why identity-statements can be informative. Russell discussed both these problems, but at the same time he broadened the range of paradigmatic semantical problems to also include the problem of empty terms, which can, among other things, manifest itself in the form of a failure of existential generalization (EG) in intensional contexts.