The success of a paradigm may not be for the reasons it claims (i.e. accurate description of the world). This is good.