Abstract

The paper argues that renormalization in quantum field theory was not a radically new - and possibly ad hoc - technique to save a badly flawed theory, but rather the culmination of a methodological strategy that physicists had been applying for a long time. The strategy was to obtain reliable results from unreliable theories by making the derivation of the results independent of possible future modifications of the theory. Examples of this practice include Bohr's use of the Correspondence Principle and Heisenberg's S-matrix theory