Online research in philosophy

We give a self-contained and streamlined version of the classification of the provably computable functions of PA. The emphasis is put on illuminating as well as seems possible the intrinsic computational character of the standard cut elimination process. The article is intended to be suitable for teaching purposes and just requires basic familiarity with PA and the ordinals below ε0. (Familiarity with a cut elimination theorem for a Gentzen or Tait calculus is helpful but not presupposed).

For α less than ε0 let $N\alpha$ be the number of occurrences of ω in the Cantor normal form of α. Further let $\mid n \mid$ denote the binary length of a natural number n, let $\mid n\mid_h$ denote the h-times iterated binary length of n and let inv(n) be the least h such that $\mid n\mid_h \leq 2$ . We show that for any natural number h first order Peano arithmetic, PA, does not prove the following sentence: For all (...) K there exists an M which bounds the lengths n of all strictly descending sequences $\langle \alpha_0, ..., \alpha_n\rangle$ of ordinals less than ε0 which satisfy the condition that the Norm $N\alpha_i$ of the i-th term αi is bounded by $K + \mid i \mid \cdot \mid i\mid_h$ . As a supplement to this (refined Friedman style) independence result we further show that e.g., primitive recursive arithmetic, PRA, proves that for all K there is an M which bounds the length n of any strictly descending sequence $\langle \alpha_0,..., \alpha_n\rangle$ of ordinals less than ε0 which satisfies the condition that the Norm $N\alpha_i$ of the i-th term αi is bounded by $K + \mid i \mid\cdot inv(i)$ . The proofs are based on results from proof theory and techniques from asymptotic analysis of Polya-style enumerations. Using results from Otter and from Matou $\breve$ ek and Loebl we obtain similar characterizations for finite bad sequences of finite trees in terms of Otter's tree constant 2.9557652856... (shrink)

We survey a selection of results about majorization hierarchies. The main focus is on classical and recent results about the comparison between the slow and fast growing hierarchies.

It is shown that the so called slow growing hierarchy depends non trivially on the choice of its underlying structure of ordinals. To this end we investigate the growth rate behaviour of the slow growing hierarchy along natural subsets of notations for Γ 0 . Let T be the set-theoretic ordinal notation system for Γ 0 and T tree the tree ordinal representation for Γ. It is shown in this paper that (G α ) α ∈ T matches up with (...) the class of functions which are elementary recursive in the Ackermann function as does (G α ) α ∈ T tree (by folklore). By thinning out terms in which the addition function symbol occurs we single out subsystems $T* \subseteq T$ and $T^{tree*} \subseteq T^{tree}$ (both of order type not exceeding ε 0 ) and prove that (G α ) α ∈ T tree* still matches up with (G α ) α ∈ T tree but (G α ) α ∈ T* now consists of elementary recursive functions only. We discuss the relationship between these results and the Γ 0 -based termination proof for the standard rewrite system for the Ackermann function. (shrink)

In this article we show how to extract with the use of the Buchholz-Cichon-Weiermann approach to subrecursive hierarchies from Rathjen's 1991 ordinal analysis of KPM a characterization of the provably total number-theoretic functions of KPM and some of its (most prominent) subsystems in a uniform and direct way.

Inspired by Pohlers' local predicativity approach to Pure Proof Theory and Howard's ordinal analysis of bar recursion of type zero we present a short, technically smooth and constructive strong normalization proof for Gödel's system T of primitive recursive functionals of finite types by constructing an ε 0 -recursive function [] 0 : T → ω so that a reduces to b implies [a] $_0 > [b]_0$ . The construction of [] 0 is based on a careful analysis of the Howard-Schütte (...) treatment of Gödel's T and utilizes the collapsing function ψ: ε 0 → ω which has been developed by the author for a local predicativity style proof-theoretic analysis of PA. The construction of [] 0 is also crucially based on ideas developed in the 1995 paper "A proof of strongly uniform termination for Gödel's T by the method of local predicativity" by the author. The results on complexity bounds for the fragments of T which are obtained in this paper strengthen considerably the results of the 1995 paper. Indeed, for given n let T n be the subsystem of T in which the recursors have type level less than or equal to n+2. (By definition, case distinction functionals for every type are also contained in T n .) As a corollary of the main theorem of this paper we obtain (reobtain?) optimal bounds for the T n -derivation lengths in terms of ω n+2 -descent recursive functions. The derivation lengths of type one functionals from T n (hence also their computational complexities) are classified optimally in terms of $ -descent recursive functions. In particular we obtain (reobtain?) that the derivation lengths function of a type one functional a ∈ T 0 is primitive recursive, thus any type one functional a in T 0 defines a primitive recursive function. Similarly we also obtain (reobtain?) a full classification of T 1 in terms of multiple recursion. As proof-theoretic corollaries we reobtain the classification of the IΣ n+1 -provably recursive functions. Taking advantage from our finitistic and constructive treatment of the terms of Gödel's T we reobtain additionally (without employing continuous cut elimination techniques) that PRA + PRWO ( $\varepsilon_0) \vdash \Pi^0_2$ - Refl(PA) and PRA + PRWO ( $\omega_{n+2}) \vdash \Pi^0_2$ - Refl(I Σ n+1 ), hence PRA + PRWO( $\epsilon_0) \vdash$ Con(PA) and PRA + PRWO( $\omega_{n+2}) \vdash$ Con(IΣ n+1 ). For programmatic reasons we outline in the introduction a vision of how to apply a certain type of infinitary methods to questions of finitary mathematics and recursion theory. We also indicate some connections between ordinals, term rewriting, recursion theory and computational complexity. (shrink)

Inspired by Pohlers' proof-theoretic analysis of KPω we give a straightforward non-metamathematical proof of the (well-known) classification of the provably total functions of $PA, PA + TI(\prec\lceil)$ (where it is assumed that the well-ordering $\prec$ has some reasonable closure properties) and KPω. Our method relies on a new approach to subrecursion due to Buchholz, Cichon and the author.

Let Ω be the least uncountable ordinal. Let K(Ω) be the category where the objects are the countable ordinals and where the morphisms are the strictly monotonic increasing functions. A dilator is a functor on K(Ω) which preserves direct limits and pullbacks. Let $\tau \Omega: \xi = \omega^\xi\}$ . Then τ has a unique "term"-representation in Ω. λξη.ω ξ + η and countable ordinals called the constituents of τ. Let $\delta and K(τ) be the set of the constituents of τ. (...) Let β = max K(τ). Let [ β ] be an occurrence of β in τ such that τ [ β] = τ. Let $\bar \theta$ be the fixed point-free version of the binary Aczel-Buchholz-Feferman-function (which is defined explicitly in the text below) which generates the Bachman-hierarchy of ordinals. It is shown by elementary calculations that $\xi \mapsto \bar \theta(\tau \lbrack \gamma + \xi \rbrack)\delta$ is a dilator for every $\gamma > \max\{\beta. \delta.\omega\}$. (shrink)

Let $\Omega:= \aleph_1$ . For any $\alpha \Omega:\xi = \omega^\xi\}$ let EΩ (α) be the finite set of ε-numbers below Ω which are needed for the unique representation of α in Cantor-normal form using 0, Ω, +, and ω. Let $\alpha^\ast:= \max (E_\Omega(\alpha) \cup \{0\})$ . A function f: εΩ + 1 → Ω is called essentially increasing, if for any $\alpha < \varepsilon_{\Omega + 1}; f(\alpha) \geq \alpha^\ast: f$ is called essentially monotonic, if for any $\alpha,\beta < \varepsilon_{\Omega + (...) 1}$; $\alpha \leq \beta \wedge \alpha^\ast \leq \beta^\ast \Rightarrow f(\alpha) \leq f(\beta).$ Let Clf(0) be the least set of ordinals which contains 0 as an element and which satisfies the following two conditions: (a) $\alpha,\beta \epsilon \mathrm{Cl}_f(0) \Rightarrow \omega^\alpha + \beta \epsilon \mathrm{Cl}_f(0)$ , (b) $E_\Omega\alpha \subseteq \mathrm{Cl}_f(0) \Rightarrow f(\alpha) \epsilon \mathrm{Cl}_f(0)$ . Let ϑεΩ + 1 be the Howard-Bachmann ordinal, which is, for example, defined in [3]. The following theorem is shown: If f:εΩ + 1 → Ω is essentially monotonic and essentially increasing, then the order type of Clf(0) is less than or equal to ϑεΩ + 1. (shrink)