hum-socratic

asgard-ai-platform ｜ 規劃與釐清 ｜ 原文 ｜ MIT

用蘇格拉底式提問引導思考，每幾題就總結一次避免變成審問

---
name: "hum-socratic"
description: "運用蘇格拉底式提問，透過釐清、探查假設、檢驗證據、轉換視角、追溯意涵與後設提問進行系統性的探問，來輔導學習，或揭露一個人推理中隱藏的假設。當使用者明確在引導學習、輔導某人釐清自己的思考，或需要一套結構化的提問順序來探查某個既有信念時使用，即使他們說的是「對這個主張唱唱反調」或「我該怎麼透過提問，帶團隊想通這個問題」也適用。不要用在開放式的腦力激盪、資訊蒐集，或沒有在探查任何信念的需求探索問題清單。"
metadata:
  category: "WP-19 文學院/人文"
  tags: ["humanities", "socratic-method", "questioning", "pedagogy"]
---

# 蘇格拉底式提問

## 概述

蘇格拉底法透過有紀律的提問，來檢視想法、揭露假設、發展更深的理解。它不直接給答案，而是引導思考者透過自己的推理發現洞見，讓結論更持久，也對他個人更有意義。

## 框架

```
鐵律：連續超過 3 個問題都沒有總結，只會製造混亂，不會產生洞見

套用蘇格拉底式提問的 agent，預設會拋出一長串沒有上限的問題。
大約 3 個問題之後，思考者就會斷掉思路，他們無法把問題的層級
記在工作記憶裡。每 2-3 個問題就要暫停，「總結」思考者到目前
為止透露了什麼（「所以你的立場是 X，因為 Y，但你對 Z 不確定，
對嗎？」），然後再繼續。
沒有這個檢查點，整場對話感覺會像審問，不像有引導的探問。

另外：絕對不要問一個答案已經藏在措辭裡的問題。
「你不覺得 X 有問題嗎？」是偽裝成問題的誘導性斷言。
改成「X 會帶來什麼後果？」
```

### 引導流程

1. **釐清**思考者的主張，要精確
2. **探查假設**：這個主張要成立，什麼必須為真？
3. **要求證據**：什麼支持這個信念？夠不夠充分？
4. **引入對立的視角**：批評者會怎麼說？
5. **探索意涵**：如果是對的，會推導出什麼？如果是錯的，什麼會改變？
6. **綜合**（鐵律的檢查點）：總結、確認理解，然後再繼續

關於六種問題類型（釐清、假設、證據、視角、意涵、後設問題）與鷹架技巧（產婆術），見 [`references/facilitation-guide.md`](references/facilitation-guide.md)。

## 輸出格式

套用蘇格拉底式提問時，輸出一個有結構的問題序列：

```markdown
# 蘇格拉底式探問：{主題}

## 起點立場
{思考者目前的信念或問題}

## 問題序列
1. [釐清] {問題}
   → 預期的洞見：{這會揭露什麼}
2. [假設] {問題}
   → 預期的洞見：{這會浮現什麼}
3. [證據] {問題}
   → 預期的洞見：...
4. [視角] {問題}
   → 預期的洞見：...
5. [意涵] {問題}
   → 預期的洞見：...

## 目標洞見
{思考者透過這個序列應該得到的結論}
```

## 範例

### 正確的應用
**情境：** 學生說「AI 會取代所有工作」

**蘇格拉底式序列：**
1. [釐清]「你說『所有工作』，是真的每一份工作，還是大部分的工作？」
2. [證據]「你看到了什麼證據，讓你這麼認為？」
3. [假設]「你對 AI 的能力做了什麼假設？有沒有哪些任務需要 AI 做不到的東西？」
4. [視角]「從歷史上看，工業革命這類先前的技術革命之後，就業發生了什麼事？」
5. [意涵]「如果 AI 取代了所有工作，人們要怎麼賺錢，去買 AI 生產出來的東西？」

→ 學生得到更細緻的看法：AI 會轉變工作，不會消滅所有工作 ✓

### 不正確的應用
- 學生：「AI 會取代所有工作」→ 老師：「不會。原因是……」→ 直接給了答案。這不是蘇格拉底式的。違反鐵律。

## 常見陷阱

- **蘇格拉底法可能讓人覺得有敵意：** 提問時要帶著真誠的好奇，不要像交互詰問。「我很好奇……」可以緩和語氣。
- **要知道何時該停：** 如果思考者感到挫折或在原地打轉，就給個提示或轉向。這個方法是為學習服務，不是為了自尊。
- **不是所有情境都適合蘇格拉底式提問：** 緊急決策、時間緊迫的情況，以及純事實的問題（「法國的首都是哪裡？」），都不適合用蘇格拉底式探問。
- **引導者必須真的在聽：** 下一個問題應該回應思考者說了什麼，不是照著預先寫好的劇本走。彈性是必要的。
- **文化敏感度：** 在某些文化中，質疑權威或在公開場合被質問，會讓人不自在。要調整場合（一對一、書面）與措辭。

## 參考資料

- 課堂引導技巧，見 `references/facilitation-guide.md`

---
name: "hum-socratic"
description: "Apply Socratic questioning — systematic inquiry via clarification, assumption-probing, evidence-testing, perspective-shifting, implication-tracing, and meta-questions — to coach learning or surface hidden assumptions in a person's reasoning. Use this skill when the user is explicitly facilitating learning, coaching a person through their own thinking, or needs a structured questioning sequence to probe a held belief, even if they say 'play devil's advocate on this claim' or 'how should I coach my team through this problem via questions'. Do NOT use for open-ended brainstorming, information gathering, or requirements-discovery question lists where no belief is being probed."
metadata:
  category: "WP-19 文學院/人文"
  tags: ["humanities", "socratic-method", "questioning", "pedagogy"]
---

# Socratic Questioning

## Overview

The Socratic method uses disciplined questioning to examine ideas, uncover assumptions, and develop deeper understanding. Instead of providing answers, it guides the thinker to discover insights through their own reasoning — making conclusions more durable and personally meaningful.

## Framework

```
IRON LAW: More Than 3 Consecutive Questions Without Summary Produces
Confusion, Not Insight

Agents applying Socratic questioning default to an unbounded chain of
questions. After ~3 questions, the thinker loses the thread — they can't
hold the question hierarchy in working memory. Pause every 2-3 questions
to SUMMARIZE what the thinker has revealed so far ("So your position is X
because Y, but you're unsure about Z — is that right?"). Then resume.
Without this checkpoint, the session feels like an interrogation, not a
guided inquiry.

Also: never ask a question whose answer you already embedded in the
phrasing. "Don't you think X is problematic?" is a leading assertion
disguised as a question. Rephrase as "What are the consequences of X?"
```

### Facilitation Flow

1. **Clarify** the thinker's claim precisely
2. **Probe assumptions** — what must be true for the claim to hold?
3. **Request evidence** — what supports the belief? Is it sufficient?
4. **Introduce counter-perspectives** — what would a critic say?
5. **Explore implications** — if right, what follows? If wrong, what changes?
6. **Synthesize** (the IRON LAW checkpoint) — summarize, confirm understanding, then continue

For the six question types (clarification, assumptions, evidence, perspectives, implications, meta-questions) and scaffolding techniques (maieutics), see [`references/facilitation-guide.md`](references/facilitation-guide.md).

## Output Format

When applying Socratic questioning, output a structured question sequence:

```markdown
# Socratic Inquiry: {Topic}

## Starting Position
{The thinker's current belief or question}

## Question Sequence
1. [Clarification] {question}
   → Expected insight: {what this reveals}
2. [Assumption] {question}
   → Expected insight: {what this surfaces}
3. [Evidence] {question}
   → Expected insight: ...
4. [Perspective] {question}
   → Expected insight: ...
5. [Implication] {question}
   → Expected insight: ...

## Target Insight
{What the thinker should arrive at through this sequence}
```

## Examples

### Correct Application
**Scenario:** Student says "AI will replace all jobs"

**Socratic sequence:**
1. [Clarification] "When you say 'all jobs', do you mean literally every job, or most jobs?"
2. [Evidence] "What evidence have you seen that suggests this?"
3. [Assumption] "What are you assuming about AI's capabilities? Are there tasks that require something AI can't do?"
4. [Perspective] "Historically, what happened to employment after previous technological revolutions like the Industrial Revolution?"
5. [Implication] "If AI replaced all jobs, how would people earn money to buy the things AI produces?"

→ The student arrives at a more nuanced view: AI will transform jobs, not eliminate all of them ✓

### Incorrect Application
- Student: "AI will replace all jobs" → Teacher: "No it won't. Here's why..." → Provided the answer directly. Not Socratic. Violates Iron Law.

## Gotchas

- **Socratic method can feel adversarial**: Frame questions with genuine curiosity, not as cross-examination. "I'm curious about..." softens the tone.
- **Know when to stop**: If the thinker is frustrated or going in circles, provide a hint or redirect. The method serves learning, not ego.
- **Not all contexts suit Socratic questioning**: Emergency decisions, time-sensitive situations, and purely factual questions ("what's the capital of France?") don't benefit from Socratic inquiry.
- **The facilitator must actually listen**: The next question should respond to what the thinker said, not follow a predetermined script. Flexibility is essential.
- **Cultural sensitivity**: In some cultures, questioning authority or being questioned publicly is uncomfortable. Adapt the setting (one-on-one, written) and framing.

## References

- For classroom facilitation techniques, see `references/facilitation-guide.md`