Though the logic of Bloomfield's statement ['The existence of even a single over-differentiated paradigm implies homonymy in the regular paradigms'] cannot be denied, I have not followed it because it seems to violate the basic functional operation of the language. Productive patterns, analogical constructions, and the structurally principal forms and combinations of forms do not appear to be based on the most diverse paradigmatic set or sets.
- Nida (1948), Pag. 425