Dunbar's Number and the Limits of Gossip
The Intuitive Assumption
If ant colonies can coordinate thousands of individuals without leaders, and ants have nearly no intelligence, then surely humans—with our massive brains, language, and memory—should coordinate even better?
We should be able to organize groups of thousands, even millions, through pure social intelligence. No chiefs, no laws, no institutions. Just smart people figuring it out.
This assumption is half right.
Intelligence does improve coordination. But it also introduces entirely new problems that mindless insects never face: deception, coalition politics, reputation management, and the strategic manipulation of social information.
Ants don't lie. Ants don't form conspiracies. Ants don't gossip about who's sleeping with whom.
Humans do all of these things. And that changes everything.
The Coordination Toolkit: What Primates Use Instead of Institutions
Primates coordinate through a set of interlocking mechanisms. None of them require formal structures. All of them rely on direct social relationships, memory, and reputation.
1. Grooming as Social Currency
Grooming isn't just hygiene. It's the fundamental social transaction in primate groups.
What grooming does:
- Establishes alliances
- Signals investment in relationships
- Reduces tension after conflict
- Maintains coalition bonds
The economy:
- Limited time → grooming is scarce → grooming becomes valuable
- Higher-status individuals receive more grooming than they give
- Grooming debts are tracked and reciprocated
- Failure to reciprocate damages reputation
Why this matters: Grooming creates a reputation-based credit system without money, contracts, or enforcement institutions.
3. Coalition Formation: The Game-Changer
Here's where primate politics gets interesting.
In most mammals, the strongest individual wins. In primates, coalitions beat strength.
How coalitions work:
Individual A (strong) threatens Individual B (weak)
Individual B calls for help
Individual C (moderate strength) joins B
A+B > Individual A
Individual A backs down
Why this matters:
- Physical dominance alone is insufficient
- Social relationships become critical
- Alliances must be maintained (through grooming, sharing, reconciliation)
- Betrayal is possible and strategically valuable
- Counter-betrayal must be deterred
The coalition arms race:
- A forms coalition with B to dominate C
- C forms coalition with D to resist A+B
- A tries to split C and D
- B betrays A for better position with C
- All participants track who did what to whom
This is primate politics. And it requires cognitive machinery that ants don't have.
5. Reciprocal Altruism and Reputation
Primates help each other. But not randomly.
The logic:
- I help you today
- You help me tomorrow
- We both benefit over time
The requirement:
- We must interact repeatedly
- We must remember past interactions
- Cheaters must be identifiable and punishable
The problem: What if you take my help and never reciprocate?
The solution: Reputation systems and punishment.
Punishment gradient in primate groups:
1\. Withheld cooperation ("I won't groom you")
2\. Social exclusion ("You're not in our coalition")
3\. Active aggression ("I'll threaten or attack you")
4\. Coalition-based punishment ("We'll all exclude/attack you")
Why this works: The cost of being excluded exceeds the benefit of cheating—IF the group is small enough that exclusion matters.
The Hard Ceiling: Dunbar's Number
All of these mechanisms—grooming, gossip, reputation, reciprocity—have a critical dependency:
You must personally know everyone in the group.
Robin Dunbar identified the constraint: the human brain can maintain stable social relationships with approximately 150 people. While the precise number is debated, the existence of a sharp cognitive limit on intimate networks is the critical insight. Why 150?
| Constraint | Limit |
|---|---|
| Cognitive: Tracking relationships | ~150 individuals |
| Time: Maintaining grooming bonds | ~150 individuals |
| Memory: Remembering reputations | ~150 individuals |
| Gossip bandwidth: Who did what | ~150 individuals |
What happens beyond 150:
Group size: 50 → Everyone knows everyone → Gossip works → Reputation enforces norms Group size: 150 → Barely manageable → Gossip strained → Some anonymity creeps in Group size: 300 → Many strangers → Gossip fails → Reputation unreliable → Free-riding increases Group size: 1,000 → Informal mechanisms fragment locally; system-wide coordination fails
The structural problem:
- At 50 people: You know Bob cheated Alice
- At 500 people: You don't know half the people or what they did
- Reputation-based enforcement collapses
- Gossip can't propagate reliably
- Cheaters can escape to subgroups where they're unknown
What This Explains
This framework clarifies:
Why small communities feel different:
- Everyone knows everyone
- Reputation matters intensely
- Gossip is powerful
- Formal rules feel unnecessary
Why cities feel anonymous:
- Strangers everywhere
- Reputation doesn't follow you
- Gossip is localized and unreliable
- Formal enforcement becomes necessary
Why organizations fragment at certain sizes:
- Startups at 150 people hit coordination crisis
- Military units kept to ~150 (company size)
- Hutterite colonies split at ~150
- Hunter-gatherer bands rarely exceed 150
Why humans evolved such large brains: Not just for tool use. For tracking complex social relationships in groups approaching Dunbar's limit.
The Limits of This Analogy
What primate studies can tell us:
- Basic coordination mechanisms
- Cognitive constraints on group size
- The logic of reputation systems
- Why informal enforcement has limits
What primate studies cannot tell us:
- Why human societies scaled beyond primate limits
- How institutions emerged
- Why some societies stayed small and others didn't
- The role of technology, language, or culture
We're building foundations. Not explaining everything yet.
What's Next
We've now seen two coordination systems: 1. Ants: Mindless, stigmergic, scalable—but brittle and limited 2. Primates: Intelligent, social, flexible—but capped at ~150 individuals
Both work. Both have limits.
But humans bypassed these constraints—not by evolving past them biologically, but by externalizing coordination into culture and institutions. For most of our existence, we didn't. Hunter-gatherer societies stayed small, mobile, and egalitarian. They solved the Dunbar problem by never exceeding Dunbar's number. Bands would split when they got too large. Conflict was resolved by walking away.
This equilibrium lasted for tens of thousands of years.
And then it broke.
Next question: Hunter-gatherer societies solved these limits for 200,000 years. How did they do it—and why did it eventually fail?
Next explainer: "The Original Affluent Society: Why Hierarchy Was Optional"