scientific article; zbMATH DE number 4039890

From MaRDI portal
Publication:3778746





The two principal `unifying' elements meant in the title go back to the sixties. First, the bulk of the kind of mathematics (more precisely, of its set-theoretic transcriptions) considered in hackneyed proof theory holds in theories for admissible sets. In fact, this is implicit in - for example, von Neumann's - work in the twenties, where functions instead of classes are used as primitives (so that definition by recursion corresponds to \(\Sigma_ 1\)-replacement). Secondly, the methods used for hackneyed proof-theoretic results, familiar from predicate logic and extended to formal arithmetic by use of proof trees of ordinal \(<\epsilon_ 0\), extend further by means of suitable ordinal structures; in fancy language, functors in the category of orderings with order- preserving maps; especially, when the systems considered are interpreted in ramified hierarchies. `Hackneyed' means that the results refer only to provability, not to structural properties of proofs, and so are usually marginal refinements of corresponding model-theoretic work. But granted this limitation of the subject, the exposition is most efficient; both professional and not too dense, hence easy to leaf through; in short, perfect for readers with a mature understanding of the subject and wanting a precis of an earlier stage. But for others almost the reverse is true. Without exaggeration the book gives a less adequate idea of the potential either of (i) the notion of admissible set or (ii) of - methods developed under the heading - proof theory than the writings of the pioneers some 20 years ago. Samples: (i) analysis of priority arguments in degree theory in terms of admissible segments of well-founded orderings (instead of less familiar non-standard models); (ii) use of the ordinal machinery for lower bounds on realizations of \(\forall \exists\) theorems (without dragging in formal systems to which all \(\Pi^ 0_ 1\) theorems are added as axioms). For the observer of the broader mathematical scene it may be of interest to trace the author's narrow conception of the subject back to a logical tradition, and not merely to contingencies: a superficial meaning of `unification' in contrast to such discoveries as, say, of the zeta- function long before logical foundations, let alone, to developments in this century, which have contributed to unification in a realistic sense in the theory of prime numbers, resp. in wider branches of mathematics. Glib slogans about the principle of economy, which is called `parsimony' on p. 3(2), obscure such achievements, besides being literally wrong. Simply ignoring the question which axioms of set theory are used in a piece of mathematics is a better bargain than verifying that they hold for admissibles as long as the pay-off for the latter consists in the kind of unification provided here.




Cited in
(59)








This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3778746)