🧩 Narrative Cluster Processor (Reusable Method)

Notion page 2ef3ba80ce814b94aca29e5d4065d6a8 · https://app.notion.com/p/2ef3ba80ce814b94aca29e5d4065d6a8 Page snapshot 2026-05-12T18:26:28Z · cached 2026-08-13 This is the machinery behind the "Narrative Cluster Processing" step of ../../protocol.md and § III.B of film-media-as-knowledge-practice.md.

🧩 A reusable method for turning raw material (calls, meetings, notes, interviews, research) into narrative clusters, implied meanings, and multiple downstream outputs (briefs, letters, decks, databases).

What this processor does

• Converts content → themes → implications → story arc → actionable artifacts
• Separates facts from meaning from decisions
• Produces consistent outputs you can reuse across projects

Inputs (what you provide)

1. Source material (transcript, notes, doc, article, recording summary)
2. Context (who/what/why now)
3. Constraints (deadline, audience, format requirements)
4. Desired outputs (choose from the library below)

Step-by-step method

1) Extract narrative clusters (theme segmentation)

Create 4–8 clusters that each answer:

• What is this part really about?
• What role does it play (gatekeeper, urgency, ritual, identity, etc.)?

Cluster template

• Label:
• Evidence (verbatim lines / bullets):
• What changes because of this (shift):

2) Derive implied meanings (subtext)

For each cluster, write 1–3 implications:

• Implied meaning: (what the facts suggest)
• Assumption: (what must be true for that implication)
• Risk if wrong: (how it fails)

3) Build a story spine (arc)

Pick the protagonist + stakes + turning point.

Story spine template

• Protagonist:
• Goal:
• Stakes:
• Constraints / antagonists: (systems, time, ambiguity, scarcity)
• Turning point (what begins now):
• First milestone:

4) Choose a direction (output selection)

Select one or more outputs based on audience and intent.

Output library (directions you can go)

A) Theme cards (analysis-ready)

For each cluster: Theme · Evidence · Implication · Decision created · Next action

B) Narrative roadmap (3 phases)

• Phase 1: Orientation
• Phase 2: Initiation / execution
• Phase 3: Coherence / integration

C) Letter-of-intent backbone (portfolio-ready)

• What I'm building
• Why this program / platform
• Why now
• Proof (past work → future work)
• Contribution

D) One-page narrative brief (stakeholder-ready)

• Situation
• What's true
• What's implied
• What matters
• What we're doing next

E) Database-ready structure (for reuse)

Suggested properties:

• Source (relation/files/link)
• Cluster (multi-select)
• Evidence (text)
• Implications (text)
• Stakes (text)
• Decision (text)
• Next action (checkbox + person + date)
• Output type (select)

Feedback protocol (fast iteration)

Reply using this format (≤55 words each section):

1. Clusters: keep/rename/split/merge + what's missing
2. Implications: agree/soften/replace + your sentence
3. Story spine: protagonist / stakes / turning point (1 sentence each)
4. Tone: pick a voice + 3 adjectives
5. Output: pick A–E (or custom)

Notes / operating principles

• Favor human-readable labels over academic taxonomy.
• Keep clusters mutually distinct; overlap is allowed but should be acknowledged.
• Treat implications as hypotheses until validated by the user.
• Outputs should be designed so they can be pasted into Notion pages or mapped into database properties.