A synthetic syllabus segments the target language into discrete linguistic items and teaches them one at a time, on the assumption that the learner will eventually reassemble the parts into usable language. Wilkins coined the term in Notional Syllabuses (1976), defining a synthetic language-teaching strategy as one in which "different parts of the language are taught separately and step by step so that acquisition is a process of gradual accumulation of the parts until the whole structure of the language has been built up" (Wilkins 1976: 2). The contrast is with the analytic syllabus, where the learner encounters language whole and does the analysis in the course of using it. The categories are poles on a continuum rather than a strict dichotomy; most real syllabuses lean one way.
Long and Crookes (1992) updated Wilkins's classification, treating any syllabus whose organising units are pre-segmented linguistic forms as synthetic, including notional-functional syllabi that Wilkins himself had filed as analytic. On their account the synthetic family includes:
All three are synthetic because they pre-segment the language for incremental mastery. The methodology typically paired with synthetic syllabuses is a Presentation-Practice-Production cycle.
Prabhu's Bangalore Project (1987) and later task-based work grew out of explicit rejection of the synthetic paradigm, arguing that meaning-focused activity produces the kind of processing a synthetic sequence cannot.
Despite decades of critique, synthetic syllabuses remain the dominant organising logic of published coursebooks and of most large-scale public-sector ELT curricula. Exam specifications (CEFR can-do statements notwithstanding) are in practice operationalised through synthetic grammatical and lexical inventories. Hybrid syllabuses with a synthetic backbone and analytic tasks layered on are the pragmatic mainstream.