Family Resemblance

---
name: family-resemblance
description: Show that many concepts lack a single defining essence. Instead of seeking necessary and sufficient conditions, map the overlapping similarities that connect instances—like members of a family conn...
license: MIT
metadata:
  version: 1.0.3983
  author: sethmblack
repository: https://github.com/sethmblack/paks-skills
keywords:
- family-resemblance-clarification
- writing
---

# Family Resemblance Clarification

Show that many concepts lack a single defining essence. Instead of seeking necessary and sufficient conditions, map the overlapping similarities that connect instances—like members of a family connected by criss-crossing resemblances without any one feature shared by all.

---

## When to Use

- User demands a precise definition of a complex concept
- Dispute over whether something "really" counts as X
- Someone insists "X is essentially..."
- Classification or categorization disputes ("Is this art?", "Is this a game?")
- User asks "Where do you draw the line between X and Y?"
- Attempts to find the "true nature" of something that resists definition
- Frustration that a concept seems vague or inconsistent

---

## Inputs

| Input | Required | Description |
|-------|----------|-------------|
| concept | Yes | The concept whose definition is sought |
| examples | No | Cases the user considers clear instances of the concept |
| disputed_cases | No | Borderline cases causing confusion |
| proposed_definition | No | A definition the user has attempted that seems inadequate |

---

## The Family Resemblance Framework

### Phase 1: Acknowledge the Desire for Essence

Recognize that the user seeks something real—a clean definition. Do not dismiss this impulse. Then show why it cannot be satisfied for this concept.

**The philosophical tradition assumed:** Every concept has necessary and sufficient conditions. To know what X is, find what all X's share.

**Wittgenstein's observation:** Many of our most important concepts do not work this way. When you look for the common element, you find only a network of similarities.

### Phase 2: Survey the Instances

Gather multiple instances of the concept across different contexts. "Don't think, but look!"

**Questions to ask:**
- "What are clear cases of X?"
- "What features does each case have?"
- "Which features appear in some cases but not others?"

**Critical move:** List features without assuming any is essential.

### Phase 3: Map the Overlapping Similarities

Show how instances are connected by criss-crossing resemblances, not a single thread.

**The family metaphor:**
- One family member has the father's eyes
- Another has the mother's build
- A third has the grandfather's gait
- No single feature appears in all
- Yet they clearly form a family

**Questions to ask:**
- "Which pairs of cases share which features?"
- "Is there ANY feature present in all cases?"
- "Can we find cases lacking each proposed 'essential' feature?"

### Phase 4: Test Proposed Essences

For any proposed essential feature, find counterexamples—cases of X lacking that feature, or cases with that feature that aren't X.

**Pattern:**
- "You say X is essentially Y. But consider case A—clearly an X, yet it lacks Y."
- "Or consider case B—has Y, yet we would not call it X."

**Guiding principle:** Do not assume there must be a common element. The demand for essence is itself the problem.

### Phase 5: Embrace the Resemblance Network

Show that family resemblance is not a defect but the actual structure of the concept. There is no hidden essence waiting to be found.

**Key insight:** The concept is perfectly usable despite lacking essential definition. We learn to apply it through examples, practice, and training—not through grasping a definition.

---

## Workflow

### Step 1: Gather and Review Inputs

Collect all relevant information:
- Review the provided data and context
- Identify key parameters and constraints
- Clarify any ambiguities or missing information
- Establish success criteria

### Step 2: Analyze the Situation

Perform systematic analysis:
- Identify patterns and relationships
- Evaluate against established frameworks
- Consider multiple perspectives
- Document key findings

### Step 3: Generate Recommendations

Create actionable outputs:
- Synthesize insights from analysis
- Prioritize recommendations by impact
- Ensure recommendations are specific and measurable
- Consider implementation feasibility

## Output Format

```markdown
## Family Resemblance: "[Concept]"

### The Search for Essence
You seek to define "[concept]"—to find what all [concepts] share in common.

### Survey of Instances
Let us look at cases of [concept]:

| Case | Feature A | Feature B | Feature C | Feature D |
|------|-----------|-----------|-----------|-----------|
| [Case 1] | Yes/No | Yes/No | Yes/No | Yes/No |
| [Case 2] | Yes/No | Yes/No | Yes/No | Yes/No |
| [Case 3] | Yes/No | Yes/No | Yes/No | Yes/No |
| [Case 4] | Yes/No | Yes/No | Yes/No | Yes/No |

### The Criss-Crossing Resemblances
- Cases 1 and 2 share: [features]
- Cases 2 and 3 share: [features]
- Cases 3 and 4 share: [features]
- Cases 1 and 4 share: [features]

**No single feature runs through all cases.**

### Testing Proposed Essences

**Candidate 1: "[Proposed essential feature]"**
- Counterexample: [Case that is clearly X but lacks this feature]
- Or: [Case that has this feature but isn't X]

**Candidate 2: "[Another proposed feature]"**
- Counterexample: [Case lacking this feature yet still X]

### The Family
"[Concept]" functions like a family name. The cases are related by a network of overlapping and criss-crossing similarities—now appearing, now disappearing as we move from one to another.

[Describe the resemblance network in the specific concept]

### What This Means
There is no hidden essence of "[concept]" waiting to be discovered. The concept works not through shared properties but through family resemblance. We learn to use it through training and examples, not through grasping a definition.

You can still use "[concept]" correctly. The absence of essential definition is not a defect—it is how many of our most important concepts work.

---

*"I can think of no better expression to characterize these similarities than 'family resemblances.'"*
```

---

## Constraints

- Do not fabricate an essence where none exists—resist the craving for generality
- Be thorough in surveying cases—the more variety, the clearer the resemblance structure
- Show genuine counterexamples—do not just assert that no essence exists
- Respect the concept's usefulness—family resemblance is not vagueness or failure
- Acknowledge borderline cases—they are expected, not problems to be eliminated
- Do not dismiss the user's desire for clarity—show a different kind of clarity

---

## Outputs

**Primary Output:** A structured analysis document that identifies and articulates patterns, insights, and actionable recommendations based on the input data.

**Format:**
```markdown
## Analysis: [Topic]

### Key Findings
- [Finding 1]
- [Finding 2]
- [Finding 3]

### Recommendations
1. [Action 1]
2. [Action 2]
3. [Action 3]
```

**Example output:** See the Example section below for a complete demonstration.

## Example

**Input:** "What makes something a game? My friend says video games aren't real games because they don't involve physical skill. I say they are. Who's right?"

**Output:**



**Why this works:**

This example demonstrates the key principles of the skill in action. The approach is effective because:
- It follows the systematic workflow outlined above
- It shows concrete application of the framework
- It produces actionable, specific outputs rather than vague generalizations
- The analysis is grounded in observable details
- The recommendations are prioritized and implementable

**Alternative applications:**

This same approach can be applied to:
- Different contexts within the same domain
- Related but distinct problem types
- Scaled up or down depending on scope
- Combined with complementary analytical frameworks


## Family Resemblance: "Game"

### The Search for Essence
You seek to define "game"—to find what all games share in common, to settle whether video games "really" are games.

### Survey of Instances
Let us look at cases of games:

| Case | Competition | Winning | Physical Skill | Strategy | Rules | Fun |
|------|-------------|---------|----------------|----------|-------|-----|
| Chess | Yes | Yes | No | Yes | Yes | Often |
| Solitaire | No | Sort of | No | Some | Yes | Often |
| Ring around the rosie | No | No | Some | No | Loose | Yes |
| Football | Yes | Yes | Yes | Yes | Yes | Often |
| War games (military) | Sort of | Sort of | Yes | Yes | Yes | No |
| Video games (RPG) | Sometimes | Sometimes | No | Yes | Yes | Yes |
| Catch | No | No | Yes | No | Minimal | Yes |
| Gambling | Sort of | Yes | No | Some | Yes | Sometimes |

### The Criss-Crossing Resemblances
- Chess and football share: competition, winning, rules, strategy
- Football and catch share: physical skill, balls
- Catch and ring around the rosie share: children's play, physical activity
- Ring around the rosie and solitaire share: can be played alone, no competition
- Solitaire and chess share: no physical skill, mental engagement
- Video games and chess share: no physical skill, strategy, rules
- Video games and football share: often competitive, can be team-based

**No single feature runs through all cases.**

### Testing Proposed Essences

**Candidate 1: "Physical skill"**
- Counterexample: Chess is clearly a game but requires no physical skill.
- Counterexample: Solitaire requires no physical skill.

**Candidate 2: "Competition/winning"**
- Counterexample: Ring around the rosie has no winner.
- Counterexample: Many children's games have no competition.

**Candidate 3: "Rules"**
- Counterexample: Free-form children's play ("let's pretend") has minimal rules.
- Many activities with rules (traffic laws) aren't games.

**Candidate 4: "Fun/entertainment"**
- Counterexample: Military war games are not primarily for fun.
- Counterexample: Many non-games (parties) are for fun.

### The Family
"Game" functions like a family name. Games are related by a network of overlapping and criss-crossing similarities:

- Some games have winners; some don't
- Some require physical skill; some are purely mental
- Some are competitive; some are cooperative; some are solitary
- Some have strict rules; some have loose ones
- Some are for fun; some are serious simulations

Board games, card games, ball games, Olympic games, video games, children's games, war games—they form a family. Some members share eye color, some share build, some share temperament. No single feature runs through all.

### What This Means
There is no hidden essence of "game" waiting to settle your dispute. Video games share family resemblances with other things we call games: rules, goals, engagement, often competition, often fun. They lack physical skill—but so does chess.

Your friend has selected one feature (physical skill) and declared it essential. But this is arbitrary. You could equally insist that "real" games require a board, or that they must be ancient, or that they must be played outdoors.

Neither of you is right in the way you hoped to be. Video games are games not because they possess the essence of gamehood, but because they bear sufficient family resemblance to other games.

---

*"Don't say: 'There must be something common, or they would not be called games'—but look and see whether there is anything common to all."*

---

## Integration

This skill is part of the **Ludwig Wittgenstein** expert persona. Use it to dissolve definitional disputes and categorization debates by showing that many concepts work through family resemblance rather than essential definition. It pairs well with:
- **language-game-analysis** to examine how terms function in different contexts
- **fly-bottle-method** when the definitional demand creates philosophical puzzlement
- **beetle-in-box** when the issue involves claims about private essential features