Curriculum
We give back the rationale the script always had and modern teaching lost. A native speaker's feel for the language is an implicit grasp of yin-yang and four-image pairings; once that is made explicit, non-native learners have rules to follow too. Axiom
Principle: education as return
Education is not to remake people as tools fitted to society but to help them regain what they originally are and awaken their innate wisdom. For the script, learning characters became a memory burden because modern teaching lost their yin-yang and four-image properties and rationales; education as return makes them explicit again.
Seven levels Planned
Levels follow the generative logic of the New Six Scripts, not stroke count or frequency:
Learning retraces creation
For native and non-native learners
L0 Beginnings
L1 Strokes · two modes
L2 Components · groups
L3 Single-unit · core
L4 Compounds · layout
L5 Words · trigrams
L6 Rationale · both readings
The five thinking primitives Method
The standard for teaching modules is not induced from classroom experience; it comes from the isomorphism of the thought that made the characters (generation track), the thought fixed in them, and the thought of learning them (existence track). Their common core is five thinking primitives, each with an observable measure:
| Primitive | Observable behaviour | Dimension |
|---|---|---|
| Mutual arising | Pairs a target character with its yin-yang partner and says why | Acquisition |
| Rotor composition | Reads the whole meaning from given parts and layout; infers the form from a given meaning | Acquisition, retention |
| Circular recognition | Explains a character family as a whole | Retention |
| Layer traversal | Moves both ways between form, meaning and rationale | Insight |
| Predictive analogy | Predicts the meaning of an unseen character from learned rationales and states the chain of reasoning | Transfer |
Three rules of assessment
- Transfer is the primary measure. If learners' predictions for new characters beat chance clearly and approach the machine-side baseline of 0.70–0.75, that is human-side evidence that rationales are learnable compositions.
- Insight is scored separately, collected before the ablation is revealed; the hit rate before the reveal is the operational definition of insight.
- The framework's axioms are not on the list of things to test.


