¹ See Carnap, “Meaning Postulates”, Philos. Studies, 3:65-73 (1952) and “Beobachtungssprache and theoretische Sprache”, Dialectica, 12:236-248 (1957), and also my “Meaning Postulates in Scientific Theories”, (hereafter: MPST) in H. Feigl and G. Maxwell (eds.) Current Issues in the Philosophy of Science (hereafter: CIPS) New York: Holt, Rinehart, and Winston, Inc., 1961, pp. 169–183.

² For arguments that this particular sentence, as it is used in physical theory today, is “conventional” and, thus, factually empty and A-true, see A. Grünbaum, “Geometry, Chronometry, and Empiricism”, in H. Feigl and G. Maxwell (eds.) Minnesota Studies in the Philosophy of Science (hereafter: MSPS), Volume III, Minneapolis: University of Minnesota Press, forthcoming.

³ Here I depart considerably from Carnap's latest interpretation of theories, which is outlined in the Dialectica article (loc. cit.) and which employs only one A-postulate—a postulate which contains the entire scientific theory (plus correspondence rules) in question along with its Ramsey sentence. While this kind of reconstruction is viable and in many ways illuminating, I feel that where one is concerned the meanings of individual terms, a less global approach, such as the one I have just outlined, is preferable.

⁴ For interesting, related considerations concerning “multicolored” surfaces, see J. J. C. Smart, “Incompatible Colors,” Philos. Studies, 10:39-42 (1959).

⁵ Cf. Wilfrid Sellars, “Grammar and Existence: a Preface to Ontology,” Mind, forthcoming, and Carnap [ESO].

⁶ However, if examples similar to fluorescein and merthiolate and others even more striking, which are now beyond the limits of our imagination, began to turn up more and more frequently, “the facts” might “gently persuade” (see my MPST) B to adopt A's framework. For such a way of talking might turn out to be a simpler—or aesthetically more pleasing—or a more convenient (though no more “logically adequate”) means of communication.

⁷ For an excellent discussion of a number of the issues involved here see H. Feigl, “De principiis non disputandum ?” in M. Black (ed.) Philosophical Analysis, Ithaca: Cornell University Press, 1950.

⁸ Although, of course, this is an extremely important sense. A framework which was so cumbersome that comprehension, communication, or computation within it was virtually impossible would be virtually useless in any sense of the word.

⁹ The psendo-epistemological task of accounting for our ability to use the quickly decidable statements to make “factual reports” can, indeed, be accomplished—or, rather, it could be accomplished if we had adequate scientific theories of perception, brain processes, “the tokening mechanism,” etc. But such an account would have to employ a framework including, of course, some quickly decidable statements. For this and for reasons given in the text, epistemology must always contain, essentially, an element of circularity (Carnap, I believe, has also made this point). I use the term ‘psudeo-’ above not as an epithet but to emphasize that such an “epistemological” account cannot be given using only “purely logical,” or “purely conceptual,” or “purely metaphysical” considerations. For helpful considerations concerning these problems see Wilfrid Sellars, “Some Reflections on Language Games,” Philosophy of Science 21:204-228 (1954) and “Empiricism and the Philosophy of the Mind,” in MSPS, Vol. I.

¹⁰ Cf. Wilfrid Sellars, “… Language Games,” loc. cit., § 67.

¹¹ See, e.g., H. Feigl, “Existential Hypotheses,” Philosophy of Science 17:35-62 (1950); Wilfrid Sellars, “The Language of Theories” in CIPS; Paul K. Feyerabend, “An Attempt at a Realistic Interpretation of Experience,” Proc. Arist. Soc. 144–170 (1958); M. Scriven, “Definitions, Explanations, and Theories,” in MSPS, Vol. II; and my “The Ontological Status of Theoretical Entities,” MSPS, Vol. III.

¹² Since the language we first learn to speak is (I suppose, by definition) “ordinary language,” it follows that when we first begin to modify and enrich our framework, whether by scientific theorizing or by philosophical analysis, both the examined portion and the portion of the framework with which we do the examining will be parts of the “ordinary” framework. And for many purposes, a slight modification of this “ordinary” framework often suffices. But it should also be obvious that by continuing the process of piecemeal modification, whereby we add new terms and new A-truths and abandon old ones, it would be possible, in principle, to completely replace the “ordinary” (or any other) framework. (Cf. G. Maxwell and H. Feigl, “Why Ordinary Language Needs Reforming,” Journal of Philosophy, V. 58, no. 18 (August 31, 1961).

¹³ It might seem, superficially, that the central theses of this paper hinge on a sharp distinction between A-truth and contingency. However, I believe that they can easily be accomodated to those who prefer a reconstruction of, say, the Quinian type, on the one hand, or a (Wilfrid) Sellarsian kind, on the other. In such reconstructions, any law-like statement would play a double role. It would be a “framework principle” and, thus, part of the implicit definition of the non-logical terms occurring in it, and it would also express a factually non-empty law.