mcpField
H
hum-discourse

hum-discourse

github.com/asgard-ai-platformその他24340 回インストール21 週間前に更新
リポジトリ

hum-discourse について

言語が論点の枠組みやアイデンティティ、テキスト内の権力関係をどう形作るかを検討する談話分析スキルです。メディア、企業広報、政策文書、政治演説を具体的な表現に基づいて分析します。

主な機能

  1. 01論点の枠組みと語句の検討
  2. 02主体の位置づけと前提の特定
  3. 03他のテキストへの参照の追跡
  4. 04強調または省略される声の把握
  5. 05文脈と構成の分析

活用シーン

  1. 01企業の人員削減発表の表現分析
  2. 02政策文書に含まれる前提の検討
  3. 03メディアや政治演説での人物描写の分析

インストール

コマンドをコピーしてターミナルで実行します。

$npx skills add https://github.com/asgard-ai-platform/skills --skill hum-discourse

インストール後、Claude Code で/hum-discourseを入力して読み込まれたか確認します。

同梱ファイルexamples/sample_scenario.mdreferences/cda-framework.md

README(原文)English

プロジェクトの README より

この README の日本語訳はまだありません。以下は英語の原文です。日本語の紹介はページ上部をご覧ください。

Discourse Analysis

Overview

Discourse analysis examines how language shapes (not just reflects) reality. It reveals how texts construct identities, power relationships, and social norms through word choice, framing, and what is left unsaid. Especially valuable for analyzing media, corporate communications, policy documents, and political speech.

Framework

IRON LAW: Language Constructs Reality, Not Just Describes It

The choice of words is never neutral. "Restructuring" vs "layoffs",
"enhanced interrogation" vs "torture", "undocumented workers" vs
"illegal aliens" — same events, different realities constructed.

Discourse analysis examines WHAT language does, not just what it says.
Key Analytical Dimensions
  1. Framing: How is the issue defined? What metaphors are used? What's included/excluded?
  2. Subject positions: Who is positioned as agent (active doer) vs patient (passive receiver)?
  3. Presupposition: What is taken for granted without being stated?
  4. Intertextuality: What other texts/discourses does this reference or draw upon?
  5. Power relations: Whose voice is amplified? Whose is silenced?
Analysis Steps
  1. Select the text: What specific communication are you analyzing?
  2. Context: Who produced it? For whom? In what setting? What's the purpose?
  3. Lexical analysis: What word choices are notable? What alternatives were available?
  4. Structural analysis: How is the text organized? What comes first? What's emphasized?
  5. What's absent: What is NOT said? Who is NOT represented?
  6. Power mapping: Who benefits from this particular framing?

Output Format

# Discourse Analysis: {Text/Document}

## Context
- Producer: {who created this text}
- Audience: {intended recipients}
- Purpose: {stated and unstated goals}
- Genre: {press release / policy doc / speech / ad}

## Framing Analysis
| Element | In the Text | Alternative Framing | Effect |
|---------|------------|-------------------|--------|
| {key term} | "{actual language}" | "{what could have been said}" | {how this shapes perception} |

## Subject Positions
- Agent (active): {who does things}
- Patient (passive): {who has things done to them}
- Absent: {who is not mentioned}

## Presuppositions
- {what the text assumes without stating}

## Power Analysis
- This framing benefits: {who}
- This framing disadvantages: {who}

## Key Findings
{What the discourse analysis reveals that a surface reading misses}

Examples

Correct Application

Scenario: Analyzing a tech company's layoff announcement

Text: "We are making the difficult decision to right-size our organization to better position ourselves for long-term growth."

ElementTextAlternativeEffect
"right-size"Implies current size is wrong"lay off 500 employees"Euphemism hides human impact
"our organization"Company as abstract entity"our colleagues"Removes personal connection to affected people
"difficult decision"Positions management as suffering"decision that will displace 500 families"Redirects sympathy toward decision-makers
"long-term growth"Forward-looking justificationNo mention of what caused over-hiringSkips accountability for the situation

Subject positions: Management = agents making "difficult decisions" (sympathetic). Employees = completely absent as subjects. ✓

Incorrect Application
  • "This press release is biased" → Too vague. Which words? What framing? Bias toward what? Discourse analysis requires specific textual evidence. Violates Iron Law: analyze what language DOES.

Gotchas

  • All language is "constructed": Discourse analysis doesn't mean the text is "lying." Even honest communication makes framing choices. The question is what those choices reveal and conceal.
  • Analyst's own discourse: Your analysis is also a discourse with its own framing and assumptions. Be reflexive about your own position.
  • Context is everything: The same words mean different things in different contexts. "We need to move fast" in a startup vs in a hospital has very different implications.
  • Don't over-read: Not every word choice is strategic. Sometimes "right-size" is just corporate jargon the writer learned, not a deliberate framing strategy. Consider intentionality.
  • Discourse analysis is not fact-checking: It examines how meaning is constructed, not whether claims are true. Pair with evidence-based analysis for a complete picture.

References

  • For Fairclough's Critical Discourse Analysis framework, see references/cda-framework.md

コメント