---
id: wiki-2026-0508-role-of-conflict-in-narrative
title: Role of Conflict in Narrative
category: 10_Wiki/Topics
status: verified
canonical_id: self
aliases: [Narrative Conflict, Story Conflict, Dramatic Conflict]
duplicate_of: none
source_trust_level: A
confidence_score: 0.9
verification_status: applied
tags: [storytelling, writing, narrative, drama]
raw_sources: []
last_reinforced: 2026-05-10
github_commit: pending
tech_stack:
  language: english-korean
  framework: narrative-craft
---

# Role of Conflict in Narrative

## 매 한 줄
> **"매 story 의 engine 의 conflict — desire vs obstacle"**. Aristotle 의 *Poetics* (335 BCE) 의 *agon* 의 origin, 매 modern beat-sheet (Save the Cat, Hero's Journey) 의 same axis 의 reuse. 매 conflict 의 absence 의 narrative 의 dead — character change 의 vehicle 의 obstacle 의 friction.

## 매 핵심

### 매 Conflict Types (Classical)
1. **Person vs Person** — 매 antagonist clash (Iliad, Breaking Bad).
2. **Person vs Self** — 매 internal moral struggle (Hamlet, Crime and Punishment).
3. **Person vs Society** — 매 institutional pressure (1984, Handmaid's Tale).
4. **Person vs Nature** — 매 survival (The Old Man and the Sea, The Martian).
5. **Person vs Technology** — 매 modern AI/system conflict (Black Mirror, Ex Machina).
6. **Person vs Fate/Supernatural** — 매 destiny challenge (Oedipus, Final Destination).

### 매 Functional Roles
- **Reveals character** — 매 pressure 의 true self.
- **Drives plot** — 매 cause-effect chain.
- **Creates stakes** — 매 reader 의 emotional investment.
- **Forces choice** — 매 character agency 의 demonstrate.

### 매 Structure (3-act + conflict escalation)
1. **Act 1 setup** — 매 inciting incident 의 conflict 의 introduce.
2. **Act 2 confrontation** — 매 rising stakes, midpoint reversal.
3. **Act 3 resolution** — 매 climax (peak conflict) → denouement.

### 매 응용
1. Novel/screenplay drafting — 매 each scene 의 micro-conflict 의 require.
2. Game narrative — 매 quest design 의 obstacle 의 core.
3. Marketing storytelling — 매 customer (hero) vs problem (villain) framing.
4. Therapy / personal narrative — 매 reframing internal conflict.

## 💻 패턴

### Beat sheet template (Markdown)
```markdown
# Story: <Title>

## Protagonist
- Name, want (external goal), need (internal lack).

## Antagonist / Obstacle
- Force opposing the want.

## Beats
1. Opening Image — status quo.
2. Inciting Incident — conflict introduced (page 10).
3. Plot Point 1 — commit to journey (page 25).
4. Midpoint — false victory or false defeat.
5. Plot Point 2 — all-is-lost moment.
6. Climax — peak conflict resolution.
7. Closing Image — mirror of opening, transformed.
```

### Conflict density check (Python)
```python
import re
def conflict_score(scene_text: str) -> float:
    # crude heuristic: conflict verbs per 100 words
    verbs = re.findall(r"\b(argue|fight|refuse|attack|defy|resist|escape|flee|kill|hate)\b",
                       scene_text, re.I)
    words = len(scene_text.split())
    return len(verbs) / max(words, 1) * 100

print(conflict_score(open("scene_1.txt").read()))
```

### Scene goal-conflict-disaster (template)
```markdown
**Scene 12 — The Confrontation**
- Goal: Hero wants to retrieve the key.
- Conflict: Antagonist has set a trap.
- Disaster: Hero gets the key but loses ally.
```

### Dialogue subtext (Luau-flavored example for game dev)
```lua
-- NPC reaction system
local function reactToPlayer(npc, action)
    if npc.relationship < 30 and action == "ask_favor" then
        npc:say("After what you did? Get out.")  -- conflict surfacing
    elseif npc.relationship > 70 and action == "ask_favor" then
        npc:say("Anything for you.")            -- conflict resolved
    end
end
```

### LLM-assisted conflict generator (Python, Anthropic SDK)
```python
import anthropic
client = anthropic.Anthropic()

resp = client.messages.create(
    model="claude-opus-4-7",
    max_tokens=400,
    messages=[{
        "role": "user",
        "content": "Given a protagonist who fears commitment and an antagonist who is their estranged sibling, generate 3 escalating conflict scenes."
    }]
)
print(resp.content[0].text)
```

## 매 결정 기준
| 상황 | Conflict type |
|---|---|
| Character study | Person vs Self |
| Thriller | Person vs Person |
| Dystopia | Person vs Society |
| Survival | Person vs Nature |
| Tech ethics | Person vs Technology |
| Tragedy | Person vs Fate |

**기본값**: 매 layered — 매 external (P vs P) + internal (P vs Self) 의 same character 의 simultaneous.

## 🔗 Graph

## 🤖 LLM 활용
**언제**: scene 의 brainstorm, conflict beat 의 escalate, antagonist motivation 의 deepen.
**언제 X**: real human conflict 의 mediation — 매 fictional craft 의 tool, not therapy.

## ❌ 안티패턴
- **Manufactured conflict**: 매 character 의 act stupidly 의 plot 의 force — reader 의 feel manipulated.
- **No internal conflict**: 매 external action 의 only — 매 character 의 flat.
- **Resolved 의 too early**: 매 act 2 의 conflict 의 deflate — sustain 의 climax 의 reach.
- **Antagonist 의 motiveless**: 매 generic villain — 매 antagonist 의 own internal logic 의 require.

## 🧪 검증 / 중복
- Verified (Aristotle *Poetics*, McKee *Story* 1997, Snyder *Save the Cat* 2005).
- 신뢰도 A.

## 🕓 Changelog
| 날짜 | 변경 |
|---|---|
| 2026-05-08 | Phase 1 |
| 2026-05-10 | Manual cleanup — 6 conflict types + beat sheet + scene template |