JRS, > > The scope of "logic" in TLP may not be clear on this point. If he is > > describing the propositional calculus, then what Godel and Turing have to > > say about second and higher orders of the predicate calculus would be > > irrelevant. > > Sure, but by what principle would one separate out only first-order > statements? But I don't think that's quite the direction that Wittgenstein > would (did) take. Three of the principles used in the Tractatus (all of which would be rejected by the later Wittgenstein) are: 1. the treatment of universal quantification as an abbreviation for a conjunction and existential quantification as an abbreviation for a disjunction; 2. the replacement of identity statements (which standardly involve quantification over classes) with the requirement that in a logically perfect language, each symbol corresponds to one and only one object; and 3. the assumption that a complete analysis of a proposition will yield elementary propositions, each of whose truth-value is independent of the truth-value of every other elementary proposition. All of these are problematic, as every Wittgenstein scholar should know, but if one were to suppose them true, the idea an analysis requiring only first-order predicate logic becomes more plausible. I agree with you that he was wrong on this but I am more interested in why he's wrong and why he might have made such a supposition (aside from the face that insights such as Turing's and Godel's came later). ========================================= Need Something? Check here: http://ludwig.squarespace.com/wittrslinks/