The History of Human Cooperation and Competition
This page gives the historical background for the behaviorally anchored model introduced on the front page. The aim is not to reconstruct a single linear story of human origins. The evidence does not support that level of certainty. The aim is more modest: identify the recurring features of human social life that a realistic model of cooperation and competition should be able to reproduce.
The central claim is that human cooperation did not evolve as pure altruism. It evolved as conditional, plastic, reputation-sensitive, norm-governed, group-bounded cooperation under ecological pressure. Competition evolved alongside it, not outside it. Humans cooperate, but often within groups, coalitions, families, alliances, institutions, and moral communities that also compete with other actors.
From Primate Sociality to Human Social Life
Human group behavior did not begin from nothing. It developed from a primate baseline that already included group living, kin bonds, dominance relations, alliances, grooming, reconciliation, aggression, and coalition behavior. Chimpanzees and bonobos are useful comparisons, although neither should be treated as a direct model of early humans.
The important point is that several ingredients of human social behavior are older than humans themselves:
- kin-biased support
- alliance formation
- status competition
- conflict and reconciliation
- social learning
- repeated interaction
- sensitivity to rank and relationship history
Human cooperation therefore built on an inherited social substrate. What changed was not that cooperation appeared for the first time, but that it became more flexible, more symbolic, more norm-governed, and more capable of scaling across groups.
Cooperative Child-Rearing and Dependence
One major evolutionary pressure was the cost of raising human children. Humans have long childhoods, large brains, extended dependency, and unusually demanding development. That makes child-rearing difficult to explain as a mother-only system.
Human infants and children are commonly supported by wider networks: fathers, grandparents, siblings, kin, and other group members. This pattern is often discussed under cooperative breeding or alloparenting. In this view, cooperation is not merely a tool for occasional collective tasks. It is part of the developmental environment in which humans become human.
For modeling, this matters because it suggests that human cooperation is not only adult exchange. It includes dependency, care, teaching, tolerance, and long-term investment in immature group members.
Foraging Bands and Flexible Group Life
For most of Homo sapiens history, human social life was probably organized around small-scale foraging societies rather than states, markets, or formal institutions. These groups should not be imagined as fixed, isolated "tribes." Ethnographic and archaeological work instead points toward flexible networks: small residential groups embedded in wider social landscapes, with people moving, visiting, marrying, trading, splitting, and rejoining.
Several features are especially important:
- food sharing
- cooperative childcare
- repeated interaction
- reputation tracking
- coalition formation
- norm enforcement
- mobility between groups
- fission-fusion grouping, where groups split and reassemble
- cooperation among both kin and non-kin
This is likely the most important historical reference point for the behaviorally anchored model. Human cooperation did not evolve only in anonymous one-shot encounters. It evolved in social worlds where people remembered one another, depended on one another, sanctioned one another, and could sometimes leave bad relationships.
Language, Norms, and Shared Reality
Language changed the scale and structure of cooperation. It allowed humans to coordinate around absent objects, future plans, past obligations, imagined outcomes, promises, accusations, reputations, and shared rules.
With language, cooperation becomes more than direct reciprocity. People can say:
- that person shared fairly
- that person cannot be trusted
- people like us do not behave that way
- this resource belongs to that family
- we owe them help
- they violated the rule
This makes human cooperation symbolic and normative. It is not only a matter of what two individuals do in the moment. It becomes a matter of what the group remembers, expects, permits, punishes, and teaches.
Cooperation and Competition Scale Together
Cooperation and competition should not be treated as separate historical stages. They scale together. Humans cooperate inside families, bands, coalitions, villages, firms, armies, parties, and states. Those same cooperative units may compete with other individuals or groups.
This creates a recurring pattern:
- internal cooperation can support external competition
- punishment can protect cooperation by opposing free riders
- group identity can expand cooperation within a boundary
- the same boundary can sharpen competition with outsiders
The behavioral implication is that human cooperation is often bounded. It depends on who is treated as kin, partner, neighbor, guest, ally, stranger, rival, or enemy. Any model that makes cooperation universal and unconditional will miss this structure.
Agriculture, Settlement, and Institutions
Agriculture and sedentary settlement changed the social problem. Stored resources, land, inheritance, population growth, and permanent settlements made cooperation larger and more institutionally organized. They also intensified inequality, hierarchy, coercion, and organized conflict.
This period did not simply increase cooperation. It changed the mechanisms that stabilized cooperation:
- property rules
- marriage and inheritance systems
- leadership and ranking
- religious or ritual authority
- third-party punishment
- taxation and redistribution
- organized defense and warfare
Compared with small-scale foraging societies, agricultural and state societies rely more heavily on institutions. Cooperation becomes less dependent on direct personal knowledge and more dependent on rules, offices, records, enforcement, and shared symbolic systems.
States, Markets, and Cooperation Among Strangers
Modern large-scale societies depend on cooperation among strangers. People routinely coordinate with others they will never meet again: buying food, using roads, paying taxes, obeying laws, trusting money, following contracts, and relying on institutions.
This is not simply a larger version of small-group cooperation. It depends on cultural and institutional machinery:
- law
- money
- writing
- bureaucracy
- contracts
- markets
- formal education
- shared national, religious, or civic identities
These mechanisms extend cooperation beyond face-to-face relationships. They also introduce new forms of competition: class conflict, market rivalry, political competition, bureaucratic power struggles, and large-scale war.
What the Behaviorally Anchored Model Should Capture
The behaviorally anchored model should therefore begin from a historically informed picture of human social life. It should not assume that humans are simply cooperative, selfish, peaceful, violent, egalitarian, hierarchical, altruistic, or competitive. Human social behavior contains all of these tendencies, expressed conditionally.
A useful model should include at least some of the following capacities:
| Capacity | Why it matters historically | Modeling implication |
|---|---|---|
| Repeated interaction | Small-scale social life often involved known partners and remembered histories. | Agents should update expectations from past encounters. |
| Reputation | Language and memory allow information about behavior to travel beyond direct interaction. | Agents should respond to observed and reported reliability. |
| Norm enforcement | Groups stabilize cooperation by sanctioning cheats, free riders, and rule-breakers. | Punishment and exclusion should be possible, even when costly. |
| Group boundaries | Cooperation often expands within social boundaries and weakens across them. | Agents should distinguish partners, allies, strangers, and rivals. |
| Social learning | Human behavior is transmitted culturally as well as learned individually. | Agents should learn from others, not only from private reward. |
| Ecological pressure | Scarcity, risk, mobility, and resource distribution change the payoff of cooperation. | Cooperation should vary with environmental conditions. |
Python Implementation
The corresponding Python implementation lives in the EvolvedCooperation
repository:
- package:
behaviorally_anchored_model/ - runtime:
behaviorally_anchored_model/behaviorally_anchored_model.py - config and proof scenarios:
behaviorally_anchored_model/config/behaviorally_anchored_config.py
That model implements the small-scale family and foraging-band layer of this page: repeated interaction, reputation, norm enforcement, group boundaries, kin/household child care, ecological pressure, and social learning. It does not try to model the later institutional layers discussed above, such as agriculture, states, law, money, writing, bureaucracy, markets, or formal education.
Methodological Role
This page supports the richer side of the dual modeling strategy. The minimal generative model asks how far cooperation can emerge from sparse assumptions. The behaviorally anchored model asks what happens when agents are given capacities that human social history suggests are important.
The role of the behaviorally anchored model is not to prove how cooperation originally evolved. Its role is to generate realistic cooperative and competitive behavior, then test which assumptions are actually necessary. If reputation, norm enforcement, group boundaries, or social learning can be removed without changing the results, they are not load-bearing in that model. If removing them destroys cooperation, they become candidates for deeper investigation in the minimal model.
In that sense, the two approaches should discipline each other. The behaviorally anchored model keeps the project close to recognizable human behavior. The minimal generative model keeps the project honest about what has actually been explained.
References
- Boehm, C. (1999). Hierarchy in the Forest: The Evolution of Egalitarian Behavior. Harvard University Press.
- Boyd, R., & Richerson, P. J. (1985). Culture and Evolutionary Process. University of Chicago Press. https://press.uchicago.edu/ucp/books/book/chicago/C/bo5970597.html
- Gurven, M., & Hill, K. (2009). Why Do Men Hunt? Current Anthropology, 50(1), 51-74. https://doi.org/10.1086/595620
- Henrich, J. (2015). The Secret of Our Success: How Culture Is Driving Human Evolution, Domesticating Our Species, and Making Us Smarter. Princeton University Press.
- Hrdy, S. B. (2009). Mothers and Others: The Evolutionary Origins of Mutual Understanding. Harvard University Press.
- Kelly, R. L. (2013). The Lifeways of Hunter-Gatherers: The Foraging Spectrum (2nd ed.). Cambridge University Press.
- Tomasello, M. (2009). Why We Cooperate. MIT Press. https://mitpress.mit.edu/9780262013598/why-we-cooperate/