I can't find where McCawley says the universal quantifier has existential import. Is it distinct from his advocacy of restricted quantification? --And.