Generative syntax embodies three complementary goals, two of which are adopted by all practitioners. The first two goals characterize what a 'possible human language' might be and provide formal grammars of individual languages. Generative syntacticians have not been very concerned with methodology. Chomsky set the tone for this lack of interest in Syntactic Structures. The generative methodology section focuses on the relative merits of introspective versus conversational data. The methodology section evaluates the recent trend to admit more and more types of semantic data as evidence in syntactic theorizing. All formal generative approaches to syntax outside of P-and-P have their roots in the lexicalist hypothesis, first proposed in Chomsky. The typological goal has in general played a much more important role in Cognitive-Functional Linguistics than in generative grammar. Cognitive-functional linguists tend to prioritize conversational and experimental data over introspective, though their day-to-day practice generally relies on the latter.