[unfortunately, the yahoo feed simply will not work. Sending manually, but determined not to be secretary.] 1.21 Each can be the case or not be the case and all else stay the same. 6. In his correspondence with Russell, W says that though he may not know WHAT the constituents of thought are or HOW they correspond with objects in the world, he knows THAT they do have constituents that correspond to the objects occurring in facts. Why? Because they MUST. [Ansc. at 28] 7. Since objects form the substance of the world, they can't be complex. If the world had no substance, one prop's making sense would depend on another prop being true. [2.021, 2.0211, Ansc. at 29] 8. If (7) weren't true and there were an infinite regress, it would be impossible to devise a picture of the world -- true or false. I.e., we would not be able to come up with props we could understand. [2.0212, Ansc. at 29, and, of course Leibniz!] 9. If we analyze props, eventually we must end up with atomic props, i.e., names in immediate combination. Thus, even if we can't give a single example of an elementary prop, an atomic fact or a simple object, W insists that each must exist. [NBs, 4.221 & Ansc. at 29] 10. Every statement about complexes can be analyzed into statements about their constituents along with the props that completely describe the complexes [2.0201 & Stern, Wittgenstein on Mind and Language at 56] 11. From the above, we can infer that, if there were no simples we could not picture the world--language would be impossible. [Stern at 57] 12. For any prop p, if it can be analyzed into props q1 & q2, then the sense of p (what p means) is dependent on the truth of q1 and q2. If q1 & q2 are themselves analyzable into props r1 - r4, then the sens of p will depend of the truth of the r props, etc. According to Stern's version of W, if there were no terminus to such analyses , the sense of p would be indeterminate, since the analysis would go on forever and the sense of p would "never stop expanding" i.e., "always be open to revision." Each subsequent analysis would be expected to alter the sense of p ("and possibly even the truth-value" ). [Stern at 57-58] 13. Stern thinks the above argument relies on the suppressed premise that "every significant proposition has a determinate sense" and points out that in the NBs, W wrote "The demand for simple things IS the demand for determinacy of sense." [Stern at 58-59, citing NB 6/18/15] 14. In the Prototractus W wrote "Before a prop can have a sense, it must be completely settled what props follow from it." Stern adds, "For a prop to have determinative sense is just a matter of establishing which props follow from it and that is only possible if analysis comes to an end. [Stern at 58, citing Prot. 3.20102-3.20103] 15. In the NBs, W also wrote, "It seems t hat the idea of the simple is already to be found contained in that of the complex and in the idea of analysis, and in such a way that we come to this aidea quite apart from any examples of simple objects, or of props which mention them, and we realize the existence of the simple abject -- a priori -- as a logical necessity. 16. According to Stern, when W wrote TLP, he assumed that "the true form of all logical relations between props is truth-functional. But truth-functional logic deals with the logical relations between props that are logically independent of one another--in other words, cases where the truth of falsity of any one prop is not affected by the truth or falsity of any other prop." [Stern at 65] 17. We know there must be atomic props because "we can draw conclusions from a false prop." We can, i.e., devise or invent a prop and know what it means without first discovering the facts which hold in regard to its subject matter." [Ansc. at 31] 18. Anscombe says that the following are all true of elementary props: (i) They're mutually independent; (ii) They're positive; (iii) There's only one way for them to be true or false; (iv) There is in them no distinction between an internal and an external negation [e.g., "The present king of France is bald can be false in two ways."]; and (v) They are concatenations of names (simple signs). And, presumably, all of these features must follow from the very fact that we can understand language. [Ansc. at 31] Walto