green numbers give full details | back to texts | unexpand these ideas
10775 | The axiom of choice now seems acceptable and obvious (if it is meaningful) |
Full Idea: The main objection to the axiom of choice was that it had to be given by some law or definition, but since sets are arbitrary this seems irrelevant. Formalists consider it meaningless, but set-theorists consider it as true, and practically obvious. | |||
From: Leslie H. Tharp (Which Logic is the Right Logic? [1975], §3) |
10766 | Logic is either for demonstration, or for characterizing structures |
Full Idea: One can distinguish at least two quite different senses of logic: as an instrument of demonstration, and perhaps as an instrument for the characterization of structures. | |||
From: Leslie H. Tharp (Which Logic is the Right Logic? [1975], §2) | |||
A reaction: This is trying to capture the proof-theory and semantic aspects, but merely 'characterizing' something sounds like a rather feeble aspiration for the semantic side of things. Isn't it to do with truth, rather than just rule-following? |
10767 | Elementary logic is complete, but cannot capture mathematics |
Full Idea: Elementary logic cannot characterize the usual mathematical structures, but seems to be distinguished by its completeness. | |||
From: Leslie H. Tharp (Which Logic is the Right Logic? [1975], §2) |
10769 | Second-order logic isn't provable, but will express set-theory and classic problems |
Full Idea: The expressive power of second-order logic is too great to admit a proof procedure, but is adequate to express set-theoretical statements, and open questions such as the continuum hypothesis or the existence of big cardinals are easily stated. | |||
From: Leslie H. Tharp (Which Logic is the Right Logic? [1975], §2) |
10762 | In sentential logic there is a simple proof that all truth functions can be reduced to 'not' and 'and' |
Full Idea: In sentential logic there is a simple proof that all truth functions, of any number of arguments, are definable from (say) 'not' and 'and'. | |||
From: Leslie H. Tharp (Which Logic is the Right Logic? [1975], §0) | |||
A reaction: The point of 'say' is that it can be got down to two connectives, and these are just the usual preferred pair. |
10776 | The main quantifiers extend 'and' and 'or' to infinite domains |
Full Idea: The symbols ∀ and ∃ may, to start with, be regarded as extrapolations of the truth functional connectives ∧ ('and') and ∨ ('or') to infinite domains. | |||
From: Leslie H. Tharp (Which Logic is the Right Logic? [1975], §5) |
10774 | There are at least five unorthodox quantifiers that could be used |
Full Idea: One might add to one's logic an 'uncountable quantifier', or a 'Chang quantifier', or a 'two-argument quantifier', or 'Shelah's quantifier', or 'branching quantifiers'. | |||
From: Leslie H. Tharp (Which Logic is the Right Logic? [1975], §3) | |||
A reaction: [compressed - just listed for reference, if you collect quantifiers, like collecting butterflies] |
10773 | The Löwenheim-Skolem property is a limitation (e.g. can't say there are uncountably many reals) |
Full Idea: The Löwenheim-Skolem property seems to be undesirable, in that it states a limitation concerning the distinctions the logic is capable of making, such as saying there are uncountably many reals ('Skolem's Paradox'). | |||
From: Leslie H. Tharp (Which Logic is the Right Logic? [1975], §2) |
10777 | Skolem mistakenly inferred that Cantor's conceptions were illusory |
Full Idea: Skolem deduced from the Löwenheim-Skolem theorem that 'the absolutist conceptions of Cantor's theory' are 'illusory'. I think it is clear that this conclusion would not follow even if elementary logic were in some sense the true logic, as Skolem assumed. | |||
From: Leslie H. Tharp (Which Logic is the Right Logic? [1975], §7) | |||
A reaction: [Tharp cites Skolem 1962 p.47] Kit Fine refers to accepters of this scepticism about the arithmetic of infinities as 'Skolemites'. |
10765 | Soundness would seem to be an essential requirement of a proof procedure |
Full Idea: Soundness would seem to be an essential requirement of a proof procedure, since there is little point in proving formulas which may turn out to be false under some interpretation. | |||
From: Leslie H. Tharp (Which Logic is the Right Logic? [1975], §2) |
10763 | Completeness and compactness together give axiomatizability |
Full Idea: Putting completeness and compactness together, one has axiomatizability. | |||
From: Leslie H. Tharp (Which Logic is the Right Logic? [1975], §1) |
10770 | If completeness fails there is no algorithm to list the valid formulas |
Full Idea: In general, if completeness fails there is no algorithm to list the valid formulas. | |||
From: Leslie H. Tharp (Which Logic is the Right Logic? [1975], §2) | |||
A reaction: I.e. the theory is not effectively enumerable. |
10772 | Compactness blocks infinite expansion, and admits non-standard models |
Full Idea: The compactness condition seems to state some weakness of the logic (as if it were futile to add infinitely many hypotheses). To look at it another way, formalizations of (say) arithmetic will admit of non-standard models. | |||
From: Leslie H. Tharp (Which Logic is the Right Logic? [1975], §2) |
10771 | Compactness is important for major theories which have infinitely many axioms |
Full Idea: It is strange that compactness is often ignored in discussions of philosophy of logic, since the most important theories have infinitely many axioms. | |||
From: Leslie H. Tharp (Which Logic is the Right Logic? [1975], §2) | |||
A reaction: An example of infinite axioms is the induction schema in first-order Peano Arithmetic. |
10764 | A complete logic has an effective enumeration of the valid formulas |
Full Idea: A complete logic has an effective enumeration of the valid formulas. | |||
From: Leslie H. Tharp (Which Logic is the Right Logic? [1975], §2) |
10768 | Effective enumeration might be proved but not specified, so it won't guarantee knowledge |
Full Idea: Despite completeness, the mere existence of an effective enumeration of the valid formulas will not, by itself, provide knowledge. For example, one might be able to prove that there is an effective enumeration, without being able to specify one. | |||
From: Leslie H. Tharp (Which Logic is the Right Logic? [1975], §2) | |||
A reaction: The point is that completeness is supposed to ensure knowledge (of what is valid but unprovable), and completeness entails effective enumerability, but more than the latter is needed to do the key job. |