This paper deals with the notions and terminology that figure in the syntactic works of Bloomfield, Fries, Hockett, Gleason, and early Chomsky. Notwithstanding Bloomfield’s commitment to constituent structure and his profound influence on syntactic research in the United States, constituency had a surprisingly peripheral role in such works as Fries (1952) “Immediate constituents” (is the last of its syntactic chapters) and notions of dependency structure a much more central role. Many false generalizations by descriptivists (e.g., treatments of Therer-insertion as inversion) result from a failure to consider complex expressions as constituents of the various constructions. Notwithstanding descriptivists’ denunciations and generativists’ endorsements of traditional grammar, it is the descriptivists whose syntactic category notions came closer to those of traditional grammar. The unusual category scheme of Fries did not deviate all that much from traditional schemes, and its innovations were not applied consistently. 1960s generative syntax shared with Fries’s approach a conception of gender features and referential indices in English as borne by Ns rather than by NPs, and a failure to treat inter- and intra-saentential anaphora uniformly. Gleason (1965) is the most honorable exception to the dismal quality of this era’s literature on parts of speech.