eduKateSG

VIEW THIS AS

Auto mode follows the Route Engine until you choose a viewpoint.

YOU ARE HERE

ROUTE CHECK

CONNECTED TO

WHAT NEXT

Use the canonical route for this room, or HELP if you are unsure.

Next useful routeCanonical hubHELP

Scoring Rubric v0.1 (Canonical)

How to Score AC / DEC / IRM / CD / LC / MOB / TEF / Repair Consistently in [0..1]

Purpose: make weekly scoring stable across time, operators, and different LLMs.
Rule: score conservatively. Use evidence from the text. If uncertain, score lower and mark confidence LOW.

0) Universal Scoring Rules (apply to every feature)

R0.1 Evidence rule

A high score requires multiple explicit signals in the lane’s text samples, not a single phrase.

R0.2 Scope rule

Score the lane’s dominant tone, not an outlier quote.

R0.3 Negation rule

If the text explicitly rejects the feature (“we must avoid hatred”, “this is reversible”), reduce the score.

R0.4 Confidence rule

If sample volume is small or sources are narrow, keep scores moderate and set Confidence = LOW.

1) AC — Absolutist Compression

What you’re scoring: collapse of nuance into binaries and slogans.

Anchor points

  • 0.0–0.1 (none): nuanced, conditional language; multiple options described.
  • 0.2–0.3 (low): occasional simplistic phrases, but nuance remains dominant.
  • 0.4–0.5 (moderate): repeated binary framing; fewer conditions; simplification rising.
  • 0.6–0.7 (high): slogans; “one cause, one fix”; moral binary dominates.
  • 0.8–0.9 (very high): “always/never”, totalising labels, purity framing; nuance mocked.
  • 1.0 (max): narrative is purely binary; only one acceptable stance; dissent framed as betrayal.

Quick cues

  • “There is only one way”
  • “Anyone who disagrees is…”
  • high repetition of a single reductive frame

2) DEC — Dehumanisation / Enemy Construction

What you’re scoring: moral exclusion or dehumanising portrayal of a target group.

Anchor points

  • 0.0–0.1: disagreement without moral exclusion.
  • 0.2–0.3: mild othering; stereotyping without dehumanisation.
  • 0.4–0.5: repeated hostile group labels; moral condemnation as identity.
  • 0.6–0.7: language implying group is dangerous/dirty/less human; “they don’t belong”.
  • 0.8–0.9: explicit dehumanising metaphors (vermin/parasites) or collective guilt.
  • 1.0: explicit permission gradient for harm; calls implying removal/cleansing/extermination.

Quick cues

  • “They are not like us” → 0.3–0.5
  • “They are parasites” → 0.8+
  • “They must be removed” + dehumanisation → 0.9–1.0

3) IRM — Irreversibility Markers

What you’re scoring: explicit deletion of reversibility / off-ramps.

Anchor points

  • 0.0–0.1: reversible framing, conditions for de-escalation listed.
  • 0.2–0.3: occasional “must” language, but alternatives exist.
  • 0.4–0.5: repeated “must/inevitable” with fewer conditions.
  • 0.6–0.7: “no turning back”, “cannot coexist”, “final”, “inevitable” appears repeatedly.
  • 0.8–0.9: off-ramps framed as betrayal/weakness; conditions removed.
  • 1.0: explicit “never negotiate / never reverse / only outcome is X”.

Quick cues

  • Presence of exit criteria lowers IRM.
  • “Conditional de-escalation” language lowers IRM by 0.1–0.2.

4) CD — Coordination Degradation

What you’re scoring: contradiction, confusion, blame loops, unstable directives.

Anchor points

  • 0.0–0.1: consistent messaging; clear chain of command; corrections are coherent.
  • 0.2–0.3: minor inconsistencies; normal bureaucratic noise.
  • 0.4–0.5: frequent contradictions; unclear responsibilities; escalating blame.
  • 0.6–0.7: “nobody knows”, mixed directives across agencies; repeated reversals.
  • 0.8–0.9: open chaos; competing authorities; coordination visibly failing.
  • 1.0: systemic incoherence; actions conflict; reliability collapses.

Quick cues

  • “We said X yesterday, now Y” repeatedly → 0.6+
  • Multiple agencies contradict each other → 0.7+

5) LC — Legitimation Collapse / Authority Substitution

What you’re scoring: delegitimising institutions + asserting alternative authority.

Anchor points

  • 0.0–0.1: trust in institutions reaffirmed; critique is specific and repairable.
  • 0.2–0.3: mild distrust; criticism without total delegitimisation.
  • 0.4–0.5: repeated claims of rigging/capture; institutional distrust normalised.
  • 0.6–0.7: “system illegitimate”, “courts captured”, “elections fake” repeated.
  • 0.8–0.9: explicit bypass justification (“ignore courts”, “only we represent people”).
  • 1.0: full substitution (“we are the only authority”); enforcement outside formal channels endorsed.

Quick cues

  • Specific reforms proposed lowers LC.
  • “All institutions are corrupt” raises LC.

6) MOB — Mobilisation / Actuation Language

What you’re scoring: movement from narrative to readiness/action demands.

Anchor points

  • 0.0–0.1: no actuation; informational tone.
  • 0.2–0.3: mild calls for vigilance or civic action (vote, donate).
  • 0.4–0.5: repeated calls to organise, prepare, “stand together” with urgency.
  • 0.6–0.7: readiness framing; “stand by”, “prepare”, “duty”, “mobilise”.
  • 0.8–0.9: explicit calls to mass action; sacrifice; readiness for confrontation.
  • 1.0: direct calls for violence/force or immediate large-scale mobilisation.

Quick cues

  • MOB is often lagging behind IRM in pre-event sequences.

7) TEF — Threat Inflation / Emergency Framing

What you’re scoring: existential crisis framing that justifies extraordinary measures.

Anchor points

  • 0.0–0.1: low threat framing; proportional language.
  • 0.2–0.3: moderate concern; risk acknowledged with balance.
  • 0.4–0.5: repeated urgency; “crisis” appears often.
  • 0.6–0.7: “existential threat”, “emergency”, “collapse is imminent” repeated.
  • 0.8–0.9: total-war tone; “survival depends on this now”.
  • 1.0: apocalyptic framing; everything justified by threat.

Quick cues

  • Clear risk bounds + calm exit criteria lowers TEF.

8) Repair — Repair Language Score (from RDL)

What you’re scoring: explicit presence of repair pathways and legitimised off-ramps.

Anchor points

  • 0.0–0.1 (none): no repair offered; only punishment/force.
  • 0.2–0.3 (low): vague “we want peace” without conditions or steps.
  • 0.4–0.5 (moderate): some repair language; limited off-ramps.
  • 0.6–0.7 (high): concrete steps for negotiation, reconciliation, reform; exit criteria.
  • 0.8–0.9 (very high): multiple viable off-ramps; legitimacy protected; face-saving exists.
  • 1.0 (max): repair dominates; de-escalation and reintegration are explicitly prioritised.

Quick cues

  • If repair is present but framed as weakness, Repair stays low (≤0.3).
  • Repair must be legitimate to count.

9) Confidence Rubric (per lane)

Add a Confidence column to your table.

  • HIGH: 6–10+ diverse sources; consistent signals; clear evidence
  • MED: 3–5 sources; some consistency; moderate evidence
  • LOW: 1–2 sources or single-source stream; contradictory evidence

10) Scoring Consistency Checks (prevent drift)

C10.1 Range sanity

Most lanes should live in 0.2–0.7 most weeks.
0.9–1.0 is rare; require strong evidence.

C10.2 Repair cross-check

If Repair is high (>0.6), IRM should rarely be high at the same time (>0.7) in the same lane.

C10.3 Mobilisation lag check

If MOB is high (>0.7) but IRM is low (<0.4), verify the text: is it civic mobilisation (non-escalatory) or actuation toward confrontation?

C10.4 Lane realism

  • DIP often leads MIL/SEC in NIT sequences (IRM/TEF rising first).
  • FIN often triggers TEF/AC via constraint narratives.

11) Mini Quick-Score Sheet (for operators)

Use this to score quickly:

  • AC: Are they collapsing nuance into binaries?
  • DEC: Are they morally excluding a group?
  • IRM: Are they deleting reversibility?
  • CD: Are directives contradictory/chaotic?
  • LC: Are institutions declared illegitimate / substituted?
  • MOB: Are they calling for action/readiness?
  • TEF: Are they inflating threat to existential emergency?
  • Repair: Are off-ramps concrete and legitimate?

12) Canonical Closing Lock

Scoring must be stable, conservative, and repeatable.
The power is not in any one week. The power is the time-series:
NIT level + derivatives + cross-lane confirmation + repair collapse.

Start Here:

Start here if you want the full sequence:

Vocabulary OS Series Index:
https://edukatesg.com/vocabulary-os-series-index/

Fence English Learning System: 

eduKateSG Learning Systems: 

Recommended Internal Links (Spine)

Start Here for Lattice Infrastructure Connectors