CAP (Context – Analysis – Practice) is a lesson-planning model for teacher education developed by Jason Anderson, with an optional fourth stage, Evaluation, giving the variant CAPE. Anderson chose plain words deliberately, so that their everyday meanings carry the pedagogic sense without the jargon of "presentation" or "engage."
Context comes first. Meaning and use are established through a text, a presented situation, or the learners' own experience, before any explicit focus on form. Analysis then draws language features out of that context, where they are, in Anderson's words, "noticed and analysed explicitly for meaning, form, pronunciation and usage/use as appropriate." Because the example arrives before the rule, Analysis tends to work more inductively than the teacher-led presentation that opens a classic PPP lesson. Practice has learners use the language along a continuum from controlled to freer activities. The optional Evaluation stage adds self-, peer-, or teacher assessment of what learners produced.
Anderson presents CAP less as an invention than as a description, "the hidden paradigm in contemporary ELT": the pattern that global coursebooks have quietly followed for decades when they introduce new language through a whole text or situation before isolating and practising it. Naming that pattern gives trainee teachers a clearer, context-first alternative to PPP. The move is not a rejection of PPP, which Anderson defends in its own right, but a reframing that puts meaning and learner noticing ahead of teacher presentation.
CAP operates at the level of the single lesson. Anderson's later TATE (Text–Analysis–Task–Exploration) lifts the same instincts to the level of curriculum design, adding an exploration stage and aligning with task-supported teaching. Both frameworks attract the same objection, pressed by Geoff Jordan among others: relabelling the stages does not free them from the synthetic syllabus. Context–Analysis–Practice still rests on pre-selecting language items and presenting them one after another, and so inherits the charge that learners do not acquire language in the order it is served to them.