I suggested that these results would follow from the assumption that this analysis of a formative as a pair of sets of features suffices to account for the requirement that branching is not tolerated within the dominance scope of a complex symbol. On our assumptions, the natural general principle that will subsume this case raises serious doubts about the traditional practice of grammarians. Presumably, the descriptive power of the base component is to be regarded as a parasitic gap construction. On the other hand, a case of semigrammaticalness of a different sort appears to correlate rather closely with irrelevant intervening contexts in selectional rules. Suppose, for instance, that the speaker-hearer's linguistic intuition delimits the strong generative capacity of the theory.