How English Develops From Primary 1 to Secondary 4

What exactly changes in English between Primary 1 and Secondary 4?

The obvious answer is:

the vocabulary becomes harder;

the passages become longer;

the writing becomes more sophisticated;

the questions become more difficult.

All of that is true.

But it may describe only the surface.

Underneath those visible changes, something more interesting appears to be happening.

A Primary 1 child is not simply doing an easier version of Secondary 4 English.

The operation itself changes.

The learner begins by learning to represent something clearly.

Then information has to travel.

Then relationships have to be maintained.

Then unstated meaning has to be inferred.

Then language has to be adapted for another person.

Then several capabilities have to operate together.

Secondary school opens the representational field even further.

Different versions become visible.

Several interpretations become plausible.

Those interpretations have to be selected among.

Then challenged.

Updated.

Evaluated.

Revised.

That gives us the developmental spine behind this entire Tampines and Pasir Ris Living World series:

P1 — REPRESENT

P2 — TRANSMIT

P3 — RELATE

P4 — INFER

P5 — ADAPT

P6 — INTEGRATE

S1 — DIFFERENTIATE

S2 — SELECT

S3 — UPDATE

S4 — EVALUATE AND REVISE

This is the spine.

Everything else in the ten-part Living World series hangs from it.

Why Use the Living World?

Tampines and Pasir Ris give us an unusually useful environment for testing English development.

Pasir Ris naturally gives us:

coast;

mangrove;

water;

habitat;

signal;

response;

variation;

adaptation;

interaction.

Tampines gives us:

transformation;

settlement;

planning;

movement;

connectivity;

changing environments;

human systems.

Together they allow us to use the Living World as an intellectual lens.

But there is an important boundary.

This is not biology tuition.

Primary 1 children are not being asked to study Darwin.

Secondary students are not being taught evolutionary theory inside English lessons.

Instead, ideas such as:

observation;

signal;

connection;

adaptation;

variation;

selection;

interaction;

change

give us different ways to expose the underlying operations of English.

The Living World is the module.

English development remains the curriculum.

The Thermometer Must Stay Fixed

This is the most important rule in the series.

If Primary 1 receives a complicated Darwin lesson simply because Darwin is the intellectual source, the experiment fails.

If Secondary 2 is pushed into Secondary 4 evaluation because the text is interesting, the experiment fails.

The level must stay calibrated.

So:

P1 temperature remains P1.

P2 temperature remains P2.

…

S2 temperature remains S2.

…

S4 temperature remains S4.

Only the Living World fraction changes.

That is what allows us to compare the levels properly.

The Full P1–S4 Spine

Now we can walk through the entire development.

Primary 1 — REPRESENT

OBSERVATION

The first English problem is remarkably small.

What is there?

A young child notices:

a bird;

a leaf;

water;

a person;

an action;

a difference.

Then English has to capture it.

The movement is:

world

↓

attention

↓

distinction

↓

word

↓

sentence

A child who says:

Bird.

has made one representation.

A child who says:

The small bird is on the branch.

has built a higher-resolution representation.

The world has not changed.

The child’s access to it has.

Primary 1 English therefore begins by helping the child make relevant distinctions visible enough to put into language.

The Living World fraction is:

OBSERVATION.

The English operation is:

REPRESENT.

Primary 2 — TRANSMIT

SIGNAL

Once something can be represented, the next problem appears.

Representation has somewhere to go.

Someone speaks.

Someone listens.

Someone writes.

Someone reads.

Someone waves.

Another person understands.

A sign appears.

A person changes direction.

The structure becomes:

sender

↓

signal

↓

receiver

↓

response

Primary 2 English therefore begins moving from isolated meaning towards connected meaning.

Consider:

The bird made a sound.

Then:

The bird called, and another bird turned towards it.

Something has travelled across a connection.

The same thing happens with human language.

A child begins understanding that communication is not complete merely because words were produced.

The receiver has to reconstruct the meaning.

The Living World fraction is:

SIGNAL.

The English operation is:

TRANSMIT.

Primary 3 — RELATE

CONNECTION

By Primary 3, isolated pieces become increasingly insufficient.

The learner has to understand relationships.

Not simply:

It rained.

And:

The path was wet.

But:

The path was wet because it had rained.

Now two representations are connected.

English increasingly has to encode:

cause;

consequence;

sequence;

contrast;

purpose;

reference;

condition.

A passage stops behaving like a pile of sentences and begins behaving like a small network.

Consider:

rain

↓

water level rises

↓

path becomes wet

↓

people change route

The learner is beginning to track a structure.

That matters for both reading and composition.

A story needs connected events.

A comprehension passage needs relationships to remain active across sentences.

The Living World fraction is:

CONNECTION.

The English operation is:

RELATE.

Primary 4 — INFER

CLUE → HIDDEN MEANING

Primary 4 creates an important shift.

The text no longer has to state everything directly.

A child sees:

someone looking at the floor;

a test paper being hidden;

a voice becoming quiet.

The passage never says:

He was embarrassed.

But the learner may infer it.

Now English requires:

visible evidence

↓

connection

↓

background knowledge

↓

possible hidden meaning

↓

check against the rest of the text

This is inference.

And the Living World gives us the same structural operation.

A visible feature may suggest a function.

Tracks may suggest something passed through.

Behaviour may indicate danger.

A condition may imply what happened earlier.

But inference must remain constrained.

It is not:

invent anything.

It is:

What does the evidence allow me to conclude?

The Living World fraction is:

CLUE → HIDDEN STATE OR FUNCTION.

The English operation is:

INFER.

Primary 5 — ADAPT

REPRESENTATION MUST FIT THE RECEIVER

Primary 5 turns the direction around.

The learner is no longer only receiving and reconstructing information.

The learner increasingly has to control what they send.

The same event may be represented differently for:

a friend;

a teacher;

a younger child;

a public notice;

a story;

a formal message.

For example:

To a friend:

Don’t go that way. The path is blocked.

To an authority:

A fallen tree has completely blocked the eastern footpath.

Same underlying event.

Different representation.

Why?

Because communication occurs inside an environment.

That environment includes:

purpose;

audience;

context;

relationship;

format.

Primary 5 therefore introduces a crucial question:

Is this representation suitable here?

Not merely:

Is it grammatically correct?

The Living World fraction is:

ADAPTATION.

The English operation is also:

ADAPT.

Primary 6 — INTEGRATE

SYSTEM

By Primary 6, the child may already possess many individual capabilities.

Vocabulary.

Grammar.

Inference.

Reading.

Writing.

Purpose.

Audience awareness.

Planning.

But the problem becomes:

Can they all remain operational together?

This is different from merely learning another skill.

A child may be able to answer inference questions during isolated practice yet struggle during a full paper.

Why?

Time.

Attention.

Question switching.

Working memory.

Pressure.

Editing.

Several operations now compete for limited resources.

The system must coordinate:

access

↓

connection

↓

inference

↓

selection

↓

representation

↓

language control

↓

monitoring

↓

time

This is why Primary 6 behaves like a closure point.

The earlier Primary capabilities have to become one functioning system.

The Living World fraction is:

SYSTEM.

The English operation is:

INTEGRATE.

The Primary Spine

We can now compress Primary English into six movements:

P1 — REPRESENT

Something becomes available to language.

↓

P2 — TRANSMIT

Meaning crosses between minds.

↓

P3 — RELATE

Pieces become connected.

↓

P4 — INFER

Connections support hidden meaning.

↓

P5 — ADAPT

Representation changes for receiver and purpose.

↓

P6 — INTEGRATE

The capabilities operate together.

That is the first developmental arc.

Then Secondary school changes the structure.

Secondary 1 — DIFFERENTIATE

VARIATION

Secondary 1 introduces more representational freedom.

The same event can now be described differently.

A person can be:

confident;

arrogant;

brave;

reckless;

determined;

stubborn.

The same sentence can change dramatically through one word.

The man walked towards the counter.

versus:

The man marched towards the counter.

The event remains similar.

The representation changes.

Different voices.

Different tones.

Different perspectives.

Different forms.

Different emphases.

The learner begins to discover:

Difference itself carries information.

This means English becomes less compressible.

Words that once looked like synonyms are not always interchangeable.

Two descriptions of the same event may both be valid but produce different effects.

The Living World fraction is:

VARIATION.

The English operation is:

DIFFERENTIATE.

Secondary 2 — SELECT

MULTIPLE CANDIDATES

Secondary 2 may be one of the most important transitions in the whole spine.

Because once variation increases, several candidate representations can remain active simultaneously.

A character appears confident.

But another clue suggests nervousness.

One interpretation fits one paragraph.

Another explains more of the passage.

Several details seem relevant.

Only two matter to the question.

Now the learner cannot simply retrieve the first plausible answer.

They have to:

generate candidates;

compare them;

weigh evidence;

notice contradiction;

suppress weaker options;

commit.

That gives us:

candidate A

candidate B

candidate C

↓

evidence

↓

weighting

↓

selection

This may explain why Secondary 2 can feel surprisingly unstable.

The learner may actually be seeing more, not less.

But more possible representations create a stronger demand for discrimination.

A student says:

Both answers seem right.

That may not be confusion in the simple sense.

It may be a selector problem.

The Living World fraction is:

SELECTION.

The English operation is:

SELECT.

Secondary 3 — UPDATE

INTERACTION

At Secondary 3, the candidates stop sitting still.

They begin to affect one another.

A claim receives a counterclaim.

A character acts.

Another character responds.

That response changes the first character.

New evidence appears.

An interpretation must be modified.

The structure becomes:

state A

↓

interaction

↓

feedback

↓

state A′

This changes argumentative writing dramatically.

A counterargument should not simply be mentioned.

It should pressure the original claim.

Initial claim:

Homework should be abolished.

Counterargument:

Some independent practice may still be necessary.

Updated claim:

Excessive repetitive homework should be reduced while targeted independent practice remains where it clearly supports learning.

The representation has changed because it encountered opposition.

That is interaction.

The Living World fraction is:

INTERACTION.

The English operation is:

UPDATE.

Secondary 4 — EVALUATE AND REVISE

EVOLUTION

By Secondary 4, change itself is no longer enough.

The learner has to ask:

Did the representation become better?

A claim may be revised.

But was the revision:

more accurate?

better supported?

more coherent?

more precise?

better adapted to the task?

The S4 loop becomes:

represent

↓

attack

↓

find weakness

↓

revise

↓

retest

↓

retain

↓

transfer

Consider:

Social media is harmful.

After testing:

Heavy or poorly regulated social media use can create serious risks when it displaces sleep, concentration or healthy offline relationships.

The revised claim is narrower.

But it may be stronger.

It has lost breadth and gained validity.

This is mature representational control.

The Living World fraction is:

EVOLUTION.

The English operation is:

EVALUATE AND REVISE.

The Secondary Spine

We can now compress Secondary English into four movements.

S1 — DIFFERENTIATE

Several meaningful representations become available.

↓

S2 — SELECT

The learner decides which deserves greater weight.

↓

S3 — UPDATE

Representations change after interacting with evidence and opposition.

↓

S4 — EVALUATE / REVISE

The learner decides what should be retained and carried forward.

That is an elegant Secondary arc:

VARIATION

↓

SELECTION

↓

INTERACTION

↓

REVISION

The Living World terminology makes the structure easy to see.

But the English operations underneath it matter more.

Two Major Closure Points

The P1–S4 spine now appears to contain two important closures.

PRIMARY 6 — INTEGRATION

Can the English machine operate as a coordinated system?

Then Secondary school opens that system again.

Multiple representations appear.

They compete.

They interact.

They change.

Then:

SECONDARY 4 — EVALUATION

Can the learner judge the outputs of the system and deliberately improve them?

So P6 and S4 are not doing the same job.

P6 asks:

Can everything work together?

S4 asks:

Can I decide how the whole representation should become better?

That is a major difference.

What Happens at Secondary 2?

The spine also gives us a much more precise way to ask why Secondary 2 may become difficult for some students.

The usual explanation might be:

the work becomes harder.

Perhaps.

But the structure suggests something more specific.

By S1, the learner has begun to see:

multiple possible words;

multiple voices;

multiple interpretations;

multiple perspectives;

multiple ways of framing something.

Then S2 asks:

Choose.

That creates what we might call:

REPRESENTATIONAL ABUNDANCE.

The learner’s problem may shift from:

I cannot think of an answer.

to:

I can think of several answers and cannot determine which deserves to survive.

That is a very different difficulty.

More Capability Can Create Temporary Volatility

This is important.

A learner who sees only one possibility can answer confidently.

A learner who sees three may hesitate.

The second learner may actually have the richer representational system.

But without a developed selector:

more possibilities can produce:

overthinking;

answer changing;

irrelevant detail;

weak prioritisation;

difficulty committing;

unstable comprehension responses.

So development is not always a straight line from:

less capability → more confidence.

Sometimes:

more capability → more internal competition → temporary instability → better selection.

That is a useful way to think about Secondary 2.

The English Machine

The full spine now looks less like a syllabus list and more like a machine gradually acquiring new operations.

STAGE 1 — ACQUIRE A REPRESENTATION

P1.

STAGE 2 — MOVE IT

P2.

STAGE 3 — CONNECT IT

P3.

STAGE 4 — EXTRACT MORE FROM IT

P4.

STAGE 5 — REDESIGN IT FOR ANOTHER MIND

P5.

STAGE 6 — OPERATE THE WHOLE SYSTEM

P6.

Then:

STAGE 7 — GENERATE / RECOGNISE ALTERNATIVES

S1.

STAGE 8 — CHOOSE AMONG THEM

S2.

STAGE 9 — UPDATE THROUGH FEEDBACK

S3.

STAGE 10 — JUDGE AND IMPROVE THE SYSTEM’S OUTPUT

S4.

That is the full P1–S4 developmental machine we are testing.

Why This Matters for Tuition

If this structure is approximately right, then English tuition should not treat every weak answer as the same kind of problem.

A Primary 3 child may have:

a connection problem.

A Primary 4 child may have:

an inference problem.

A Primary 5 child may know the content but fail:

receiver adaptation.

A Primary 6 student may possess the individual skills but lose them:

under coordination load.

A Secondary 1 student may collapse useful distinctions.

A Secondary 2 student may generate candidates but weight them poorly.

A Secondary 3 student may fail to update after challenge.

A Secondary 4 student may revise language without revising thinking.

These require different interventions.

“Do more English” is too coarse.

A Better Diagnostic Question

Instead of asking only:

What did the student get wrong?

Ask:

Which operation failed?

Did they fail to:

represent?

receive?

connect?

infer?

adapt?

coordinate?

differentiate?

select?

update?

evaluate?

Now the error becomes diagnostic.

That creates a much more precise repair system.

The Spine Also Explains Why Earlier Skills Never Disappear

Secondary 4 does not replace Primary 1.

A Secondary 4 student still has to notice accurately.

Still has to decode signals.

Still has to maintain relationships.

Still has to infer.

Still has to adapt to audience.

Still has to coordinate.

The later operations sit on top of the earlier ones.

So the structure is cumulative.

A weakness far down the spine can destabilise much later performance.

A Secondary student with weak vocabulary access may appear to have an evaluation problem when the failure actually began much earlier.

That matters enormously for tuition diagnosis.

The Same Passage Can Be Used at Different Temperatures

This is another useful consequence.

Imagine one photograph of a mangrove.

P1

What can you see?

P2

What signal might an animal detect?

P3

How are the organism and environment connected?

P4

What can you infer from this visible feature?

P5

How would you explain this to a younger child?

P6

Can you understand, select and communicate the relevant information accurately under a complete task?

S1

How could two observers represent this environment differently?

S2

Which interpretation is best supported by the available evidence?

S3

How would one position change after another person challenges it?

S4

After considering all the evidence and alternatives, what is the strongest representation you can defend?

Same world.

Different English temperature.

That is what calibration means.

Tampines and Pasir Ris Become More Than Locations

This is why the geographical pairing matters.

Pasir Ris gives us an environment where:

land meets water;

organisms meet conditions;

signals produce responses;

variation becomes visible.

Tampines gives us another kind of transformation:

human settlement;

planning;

connectivity;

environmental redesign;

changing social space.

One gives us:

organism ↔ environment.

The other gives us:

people ↔ designed environment.

Together they reinforce a central idea running through the series:

English is also an interface.

Learner ↔ world.

Speaker ↔ listener.

Writer ↔ reader.

Text ↔ interpretation.

Claim ↔ counterclaim.

Representation ↔ evidence.

The learner spends ten years becoming increasingly capable of operating across those interfaces.

The Living World Series

The ten individual levels therefore fit inside one developmental spine.

Primary 1 — Observation

SEE.

Primary 2 — Signal

RECEIVE.

Primary 3 — Connection

CONNECT.

Primary 4 — Inference

INFER.

Primary 5 — Adaptation

ADAPT.

Primary 6 — System

COORDINATE.

Secondary 1 — Variation

DIFFERENTIATE.

Secondary 2 — Selection

SELECT.

Secondary 3 — Interaction

UPDATE.

Secondary 4 — Evolution

EVALUATE AND REVISE.

That is the complete Living World developmental sequence.

But This Is Also a Research Instrument

The series has another job.

We can take a completely different intellectual module and pour it through the same P1–S4 column.

Chaucer produces one set of surface fractions.

Shakespeare produces another.

The Living World produces another.

If the underlying English operations keep appearing in the same places despite radically different modules, we may have discovered something more stable than the themes themselves.

That is why the spine matters.

The module can change.

The thermometer stays fixed.

Then we compare the fractions.

The Current Candidate Invariant

After the first three tubes, the hidden sequence looks like this:

REPRESENT

↓

TRANSMIT

↓

RELATE

↓

INFER

↓

ADAPT

↓

INTEGRATE

↓

DIFFERENTIATE

↓

SELECT

↓

UPDATE

↓

EVALUATE / REVISE

We should not treat this as a universal law.

It is a working developmental model.

And its value comes from being testable.

What Happens Next?

The next step is not to protect the model.

It is to attack it.

Run a radically different module.

Sport.

Music.

Mathematics.

Architecture.

Navigation.

Markets.

Machines.

Law.

Something far away from Chaucer, Shakespeare and Darwin.

Then keep the P1–S4 temperatures fixed.

If the same developmental fractions keep returning, the spine becomes more interesting.

If they do not, we change it.

That is how the model improves.

English Tuition Tampines & Pasir Ris

For parents and students, however, the practical result is straightforward.

English development should not be understood only as:

more vocabulary;

harder grammar;

longer comprehension;

better compositions.

Those are visible outputs.

Underneath, the learner may be developing a sequence of increasingly powerful operations:

learning to represent the world;

learning to transfer meaning;

learning to connect information;

learning to infer;

learning to communicate for another mind;

learning to coordinate a complete English system;

learning to preserve meaningful differences;

learning to select among alternatives;

learning to update after feedback;

and finally learning to judge and improve the representations they themselves produce.

That is a much richer picture of what happens between Primary 1 and Secondary 4.

And it gives us a much better question when a child struggles.

Not simply:

Is this child weak in English?

But:

Where in the English system is the signal breaking?

Once we know that,

tuition stops being a pile of extra work.

It becomes targeted repair and deliberate development.

The Full P1–S4 Spine

P1 — REPRESENT
Turn relevant experience into language.

P2 — TRANSMIT
Track meaning between sender and receiver.

P3 — RELATE
Build relationships across information.

P4 — INFER
Use evidence to reconstruct unstated meaning.

P5 — ADAPT
Change representation for purpose, audience and context.

P6 — INTEGRATE
Coordinate multiple English capabilities under constraint.

S1 — DIFFERENTIATE
Recognise meaningful variation among representations.

S2 — SELECT
Discriminate and weight competing possibilities.

S3 — UPDATE
Modify representations through interaction and feedback.

S4 — EVALUATE / REVISE
Judge what works, repair what does not and carry stronger representations forward.

From the first word a Primary 1 child uses to represent something they have noticed,

to the Secondary 4 student who can challenge, revise and defend an increasingly precise representation,

the entire developmental journey can be expressed in ten movements:

REPRESENT → TRANSMIT → RELATE → INFER → ADAPT → INTEGRATE → DIFFERENTIATE → SELECT → UPDATE → EVALUATE.

That is the spine of the Tampines and Pasir Ris Living World series.

And now that the spine exists, every new English module can be tested against it.

EDUKATESG_ENGLISH_FRACTIONATION_PROTOCOL:
version: “2026-08-13.TPPR.LIVINGWORLD.v1”
status: “WORKING_RESEARCH_MODEL”
purpose: “Preserve the exact experimental architecture developed in this branch and prevent future model drift/hallucinated reinterpretation.”

NON_NEGOTIABLE_BOUNDARIES:
– “This is NOT an official MOE P1-S4 developmental taxonomy.”
– “MOE syllabus/curriculum is an external calibration source; the latent P1-S4 spine below is an eduKateSG research hypothesis.”
– “Do NOT claim Darwin, Chaucer, Shakespeare, or any named intellectual figure is taught to students unless explicitly stated.”
– “Modules are lenses/liquids, NOT curriculum replacements.”
– “Locations provide grounded thematic/contextual environments; they do NOT determine the English developmental temperature.”
– “Do NOT rewrite a level because a module produces a more attractive metaphor.”
– “Do NOT infer universality from 3 tubes.”
– “Do NOT call correlations/convergences developmental laws.”
– “S2 volatility/selector hypothesis remains provisional and requires further cross-module testing.”
– “Surface fraction labels and latent English operations are separate variables.”
– “Later levels contain earlier capabilities; stages are cumulative, not replacements.”

EXPERIMENT:
metaphor:
COLUMN: “Fixed P1→S4 English developmental calibration”
TEMPERATURE: “Level-specific English demand”
LIQUID: “Intellectual module”
FRACTION: “Module-specific manifestation at that fixed English level”
INVARIANT: “Candidate latent English operation recurring across structurally distant modules”

method:
  - "CALIBRATE_COLUMN"
  - "POUR_MODULE_BLIND"
  - "EXTRACT_LEVEL_FRACTIONS"
  - "DO_NOT_COMPARE_DURING_EXTRACTION"
  - "COMPLETE_FULL_P1_S4_TUBE"
  - "HORIZONTAL_COMPARE_AFTER_COMPLETION"
  - "SEPARATE_SURFACE_LABEL_FROM_LATENT_OPERATION"
  - "SEARCH_FOR_CONVERGENCE"
  - "ATTACK_CONVERGENCE_WITH_DISTANT_MODULES"
  - "REVISE_SPINE_IF_COUNTEREVIDENCE_APPEARS"

FIXED_COLUMN:
P1:
latent_operation: “REPRESENT”
calibrated_condition: “Access / noticing / elementary representation”
canonical_question: “What is there?”
P2:
latent_operation: “TRANSMIT”
calibrated_condition: “Connected meaning / transmission / reception”
canonical_question: “What is this telling me?”
P3:
latent_operation: “RELATE”
calibrated_condition: “Relationship-building across information”
canonical_question: “How does this connect to that?”
P4:
latent_operation: “INFER”
calibrated_condition: “Evidence-supported reconstruction beyond direct statement”
canonical_question: “What can I work out from the evidence?”
P5:
latent_operation: “ADAPT”
calibrated_condition: “Purpose / audience / context-sensitive representation”
canonical_question: “How should I represent this for this receiver/task?”
P6:
latent_operation: “INTEGRATE”
calibrated_condition: “Integrated Primary control under combined constraint”
canonical_question: “Can the English system keep working together?”
S1:
latent_operation: “DIFFERENTIATE”
calibrated_condition: “Abstraction + meaningful representational variation”
canonical_question: “What differences matter?”
S2:
latent_operation: “SELECT”
calibrated_condition: “Discrimination / weighting among viable representations”
canonical_question: “Which candidate deserves more weight?”
S3:
latent_operation: “UPDATE”
calibrated_condition: “Critical interaction / feedback / state change”
canonical_question: “What changes after these representations interact?”
S4:
latent_operation: “EVALUATE_REVISE”
calibrated_condition: “Integrated judgement / evaluation / refinement”
canonical_question: “What stronger representation should be retained?”

CURRENT_LATENT_SPINE:
ordered:
– “P1:REPRESENT”
– “P2:TRANSMIT”
– “P3:RELATE”
– “P4:INFER”
– “P5:ADAPT”
– “P6:INTEGRATE”
– “S1:DIFFERENTIATE”
– “S2:SELECT”
– “S3:UPDATE”
– “S4:EVALUATE_REVISE”

primary_arc:
  sequence: ["REPRESENT", "TRANSMIT", "RELATE", "INFER", "ADAPT", "INTEGRATE"]
  interpretation: "Build and coordinate the representational machine."
  closure:
    level: "P6"
    type: "SYSTEM_INTEGRATION"
    question: "Can multiple previously acquired English operations function together under constraint?"
secondary_arc:
  sequence: ["DIFFERENTIATE", "SELECT", "UPDATE", "EVALUATE_REVISE"]
  interpretation: "Open, compare, interact with, and improve representations."
  closure:
    level: "S4"
    type: "SYSTEM_EVALUATION"
    question: "Can the learner judge, revise, retain and transfer stronger representations?"

MODULES_COMPLETED:
CHAUCER:
type: “literary/social-perspectival”
exact_surface_fraction_labels: “Use prior Chaucer branch as source of truth; do NOT reconstruct missing labels from memory.”
broad_mechanisms:
– “voice”
– “social position”
– “perspective”
– “story”
– “human plurality”

SHAKESPEARE:
  type: "literary/dramatic/interiority"
  exact_surface_fraction_labels: "Use prior Shakespeare branch as source of truth; do NOT reconstruct missing labels from memory."
  broad_mechanisms:
    - "speech"
    - "performance"
    - "inner/outer representation"
    - "conflict"
    - "ambiguity"
    - "competing interpretation"
LIVING_WORLD_DARWIN:
  type: "non-literary control tube"
  location_context: ["Tampines", "Pasir Ris"]
  explicit_boundary: "Not Darwin biography; not biology tuition."
  surface_fractions:
    P1: "OBSERVATION"
    P2: "SIGNAL"
    P3: "CONNECTION"
    P4: "CLUE_TO_HIDDEN_STATE_OR_FUNCTION"
    P5: "ADAPTATION"
    P6: "SYSTEM"
    S1: "VARIATION"
    S2: "SELECTION"
    S3: "INTERACTION"
    S4: "EVOLUTION"

LIVING_WORLD_MAPPING:
P1:
surface: “OBSERVATION”
latent: “REPRESENT”
rule: “Notice relevant difference and encode it.”
P2:
surface: “SIGNAL”
latent: “TRANSMIT”
rule: “Track source → representation/signal → receiver → possible response.”
P3:
surface: “CONNECTION”
latent: “RELATE”
rule: “Build entity/event/environment/consequence relationships.”
P4:
surface: “CLUE_TO_HIDDEN_STATE_OR_FUNCTION”
latent: “INFER”
rule: “Use visible evidence to support constrained hidden meaning.”
P5:
surface: “ADAPTATION”
latent: “ADAPT”
rule: “Change representation according to audience, purpose, context, form.”
P6:
surface: “SYSTEM”
latent: “INTEGRATE”
rule: “Coordinate access, relation, inference, selection, output, monitoring, time.”
S1:
surface: “VARIATION”
latent: “DIFFERENTIATE”
rule: “Preserve meaningful differences among words, voices, forms, perspectives, representations.”
S2:
surface: “SELECTION”
latent: “SELECT”
rule: “Maintain multiple candidates; weight by task/evidence; suppress weaker candidates; commit.”
S3:
surface: “INTERACTION”
latent: “UPDATE”
rule: “Track representation/state A → challenge/response → feedback → revised state A’.”
S4:
surface: “EVOLUTION”
latent: “EVALUATE_REVISE”
rule: “Represent → attack → identify weakness → revise → retest → retain → transfer.”

S2_SPECIAL_HYPOTHESIS:
status: “PROVISIONAL_HIGH_INTEREST”
do_not_state_as_fact: true
candidate_mechanism:
– “S1 expands representational degrees of freedom.”
– “Multiple viable candidates become simultaneously available.”
– “S2 requires discrimination and weighting.”
– “Weak/immature selection may manifest as hesitation, overthinking, answer-changing, irrelevant inclusion, or unstable commitment.”
concise_form:
“REPRESENTATIONAL_ABUNDANCE -> CANDIDATE_COMPETITION -> WEIGHTING_REQUIREMENT -> SELECTION”
important_distinction:
P4: “CLUE -> INFERENCE”
S2: “MULTIPLE_CLUES + MULTIPLE_INTERPRETATIONS -> WEIGHTING -> SELECTION”
falsification_test:
“Run structurally distant modules through the fixed column. If S2 repeatedly fails to generate discrimination/weighting/selection, weaken or replace the hypothesis.”

S1_S2_S3_S4_BOUNDARIES:
S1:
main_problem: “Several representations become visible.”
operation: “Differentiate.”
forbidden_drift: “Do not make S1 primarily about choosing the best candidate.”
S2:
main_problem: “Several representations remain viable.”
operation: “Select/weight.”
forbidden_drift: “Do not make S2 full integrated evaluative judgement.”
S3:
main_problem: “Representations/actions affect one another across time.”
operation: “Update through interaction/feedback.”
forbidden_drift: “Do not reduce S3 to static comparison.”
S4:
main_problem: “Judge revised representations after pressure/testing.”
operation: “Evaluate, revise, retain, transfer.”
forbidden_drift: “Do not equate revision with merely changing wording.”

CRITICAL_DISTINCTIONS:
– id: “P5_vs_S4”
P5: “Fit representation to current receiver/purpose/context.”
S4: “Evaluate and improve underlying representation after testing across constraints.”
– id: “P6_vs_S4”
P6: “Integrate multiple capabilities.”
S4: “Evaluate and improve outputs produced by the integrated system.”
– id: “S1_vs_S2”
S1: “Candidate-space expansion / meaningful difference.”
S2: “Candidate weighting / selection.”
– id: “S2_vs_S3”
S2: “Static comparative selection can suffice.”
S3: “State changes because of interaction/feedback.”
– id: “P4_vs_S2”
P4: “Infer from constrained evidence.”
S2: “Choose among multiple plausible evidence-supported candidates.”

LOCATION_LOGIC:
PASIR_RIS:
use_as:
– “coast”
– “mangrove”
– “habitat/interface”
– “signal/response”
– “organism-environment relation”
– “living-system interaction”
prohibition:
– “Do not reduce location to ‘nature decoration’.”
TAMPINES:
use_as:
– “transformation”
– “planned environment”
– “connectivity”
– “human-environment reorganisation”
– “changing systems”
prohibition:
– “Do not invent historical/local claims without verification.”
pair_logic:
“PASIR_RIS: organism adapts within environment + TAMPINES: environment is transformed/reorganised -> learner/representation ↔ English environment”

ARTICLE_ARCHITECTURE_COMPLETED:
spine_article:
title: “English Tuition Tampines & Pasir Ris | The Living World P1–Secondary 4 Developmental Spine”
role: “Canonical parent-facing explanation of the full P1-S4 latent spine.”
level_articles:
P1: “Observation”
P2: “Signal”
P3: “Connection”
P4: “Inference”
P5: “Adaptation”
P6: “System”
S1: “Variation”
S2: “Selection”
S3: “Interaction”
S4: “Evolution”
horizontal_comparison_article:
role: “Compare completed Chaucer × Shakespeare × Living World tubes only after each tube is complete.”

THREE_TUBE_RESULT:
status: “EARLY_CONVERGENCE_NOT_PROOF”
strongest_current_signal: “S2”
candidate_cross_module_invariants:
P1: “Something becomes distinct enough to represent.”
P2: “Meaning traverses a sender/representation/receiver relationship.”
P3: “Information becomes relational/networked.”
P4: “Surface evidence supports hidden reconstruction.”
P5: “Representation becomes receiver/task-sensitive.”
P6: “Multiple prior operations require coordination.”
S1: “Meaningful representational variation proliferates.”
S2: “Multiple viable candidates require discrimination/weighting.”
S3: “Representations update through interaction/feedback.”
S4: “Representations are evaluated, revised and selectively retained.”

NEXT_EXPERIMENT_RULE:
objective: “Attack the candidate invariant with maximum structural distance.”
preferred_future_modules:
– “sport”
– “mathematics”
– “music”
– “architecture”
– “navigation”
– “markets”
– “machines”
– “law”
selection_criterion:
“Choose modules because they are structurally distant from Chaucer/Shakespeare/Living World, not merely because they make attractive tuition-page themes.”
blind_requirement:
“Derive each full P1-S4 module without using prior module-specific fractions as templates; compare only after full tube completion.”

ANTI_HALLUCINATION_RUNTIME:
BEFORE_WRITING_ANY_NEW_LEVEL_ARTICLE:
– “Load FIXED_COLUMN[level].”
– “Lock latent_operation.”
– “Load selected module.”
– “Generate only module-specific manifestation compatible with locked level.”
– “Reject any idea that raises/lowers developmental temperature.”
– “Keep module/source behind the English lesson unless explicitly relevant.”
– “State uncertain claims as hypotheses.”
– “Verify current/local/MOE facts externally when used.”

BEFORE_HORIZONTAL_COMPARISON:
  - "Require completed P1-S4 tubes."
  - "Retrieve exact prior labels; never reconstruct missing ones."
  - "Compare latent operations, not surface vocabulary."
  - "Record divergence as evidence, not error."
  - "Do not force convergence."
  - "Search for possible alternative explanations."
  - "Mark confidence per level."
WHEN_MODEL_MEMORY_IS_INCOMPLETE:
  action: "STOP_RECONSTRUCTION"
  instruction: "Retrieve prior branch/context before asserting exact labels or conclusions."
WHEN_NEW_EVIDENCE_CONTRADICTS_SPINE:
  action: "UPDATE_MODEL"
  instruction: "Do not preserve elegance at the expense of evidence."

CANONICAL_ONE_LINE:
“FIXED_COLUMN(P1-S4) + VARIABLE_MODULE -> LEVEL_FRACTIONS -> CROSS_MODULE_COMPARE -> CANDIDATE_INVARIANTS -> ADVERSARIAL_NEW_TUBES -> REVISE”

CANONICAL_SPINE:
“REPRESENT -> TRANSMIT -> RELATE -> INFER -> ADAPT -> INTEGRATE -> DIFFERENTIATE -> SELECT -> UPDATE -> EVALUATE_REVISE”

CONFIDENCE:
overall: “PROVISIONAL”
living_world_internal_coherence: “HIGH”
three_tube_cross_module_convergence: “MODERATE”
S2_selector_hypothesis: “MODERATE_AND_TEST_WORTHY”
universal_developmental_claim: “NOT_ESTABLISHED”

NEVER_DO:
– “Never present the latent spine as an official MOE progression.”
– “Never change the thermometer to make a module fit.”
– “Never backfill missing Chaucer/Shakespeare details from intuition.”
– “Never interpret metaphor as literal cognitive mechanism without evidence.”
– “Never infer causation from thematic convergence.”
– “Never collapse all S2 student difficulties into the selector hypothesis.”
– “Never assume a pretty 10-stage sequence is correct because it is coherent.”
– “Never compare incomplete tubes as if their fractions were final.”
– “Never let SEO/location writing overwrite the experimental architecture.”
– “Never protect the model from falsification.”