Cluster Convergence Theorem

Abstract

A power-counting theorem is presented, that is designed to play an analogous role, in the proof of a BPHZ convergence theorem, in Euclidean position space, to the role played by Weinberg's power-counting theorem, in Zimmermann's proof of the BPHZ convergence theorem, in momentum space. If $x$ denotes a position space configuration, of the vertices, of a Feynman diagram, and $\sigma$ is a real number, such that $0 < \sigma < 1$, a $\sigma$-cluster, of $x$, is a nonempty subset, $J$, of the vertices of the diagram, such that the maximum distance, between any two vertices, in $J$, is less than $\sigma$, times the minimum distance, from any vertex, in $J$, to any vertex, not in $J$. The set of all the $\sigma$-clusters, of $x$, has similar combinatoric properties to a forest, and the configuration space, of the vertices, is cut up into a finite number of sectors, classified by the set of all their $\sigma$-clusters. It is proved that if, for each such sector, the integrand can be bounded by an expression, that satisfies a certain power-counting requirement, for each $\sigma$-cluster, then the integral, over the position, of any one vertex, is absolutely convergent, and the result can be bounded by the sum of a finite number of expressions, of the same type, each of which satisfies the corresponding power-counting requirements.