The postulation of diachronic universals derives from certain conclusions reached in the investigation of synchronic universals, i.e., that there are natural (universal) phonological subsystems in languages or dialects, and that there are natural (universal) structural relationships between the elements in these subsystems. In essence, a hypothesis about a particular diachronic universal is a claim that a shared natural state in various languages or dialects is the product of the same diachronic process (a sound change, series of linked sound changes, etc.). A counterproposal to this hypothesis would specify that there are multiple diachronic routes which can link two parallel states in the history of two or more innovating languages or dialects. If the latter view is correct, and I believe that it is, then the theory of diachronic universals will require refinement if it is to retain theoretical value.