Abstract

In the first part of this paper the author restates arguments made earlier against well-known criticisms of a logical nature leveled (by C. Hempel and others) against the so-called verifiability principle, which purport to show that it is at once both too restrictive and too permissive: including as cognitively meaningful, statements intuitively lacking this property, and excluding others that are generally admitted to possess it. The author claims to show that the charge that the verifiability principle is unduly permissive will not stand, because of certain logico-semantical blunders made by the critics; and that the charge of being too restrictive can be removed by introducing the notion of guasi-truth-conditions. He then moves on to an examination and elaboration of the concept of cognitive use, in terms of which he attempts to turn the flank of the argument purporting to show that the verifiability principle is too restrictive, explaining that even if one does not invoke the concept of quasi-truth-conditions, that of cognitive use enables one to save for science and cognition in general sentences that lack truth-values (cognitive meaning). A version of the verifiability principle that denies cognitive meaning to non-finitistic, non-analytic mixed-quantifications does not therefore do any injury to science, frequent claims to the contrary notwithstanding.