The Evolution of Cooperation (1984) explains how cooperation can emerge and persist among self-interested individuals—even when defection is the dominant short-term strategy. Using the iterated Prisoner’s Dilemma tested via computer tournaments, it shows that cooperation becomes rational when interactions repeat and future consequences matter.

Core Principles of Cooperation

• Cooperation beats defection in long-term exchanges
• The longer the time horizon, the more cooperation outperforms
• Cooperation needs power and ‘retaliatory capacity’ to keep cheaters in check
• One-off encounters & impending relationship dissolution boost the risks of cheating: future value-giving is the best promoter of cooperation (‘shadow of the future’)

Core insight: Cooperation can evolve spontaneously among self-interested actors if interaction is repeated because cooperation satisfies both parties’ self-interest.

Notes

About the Authors:
Robert Axelrod is an American political scientist, he holds a B.A. in mathematics and a Ph.D. in political science (Yale University).
William D. Hamilton is a Darwinian evolutionary biologist famous for inclusive fitness theory (ie.: altruism weighed by genetic relatedness and benefit/cost calculations (Hamilton, 1964 part I and part II).

Foreword (by Richard Dawkins)

Dawkins, author of The Selfish Gene, wrote the foreword.

THIS IS A BOOK OF OPTIMISM. But it is a believable optimism, more satisfying than naive, unrealistic hopes of pie in the sky (or rapture in the revolution).
To be believable, an optimism must first acknowledge fundamental reality, including the reality of human nature, but also the nature of all life. Life as we know it, and probably throughout the universe if there is life elsewhere, means Darwinian life. In a Darwinian world, that which survives survives, and the world becomes full of whatever qualities it takes to survive. As Darwinians, we start pessimistically by assuming deep selfishness at the level of natural selection, pitiless indifference to suffering, ruthless attention to individual success at the expense of others. And yet from such warped beginnings, something can come that is in effect, if not necessarily in intention, close to amicable brotherhood and sisterhood. This is the uplifting message of Robert Axelrod’s remarkable book.

Chapter 1: The Problem of Cooperation

Core problem: In life, selfish individuals find it hard to cooperate even if it is in their collective best interest. 

Question: Under what conditions will cooperation emerge in a world of egoists without central authority?

The book seeks to answer this question with mathematical models derived from the Prisoner’s Dilemma, a model where individual rationality leads to a collectively inferior outcome. In this game, two players face a choice between cooperation and defection:

In order of what’s best for the selfish individual:

• T (Temptation to Defect): The highest payoff, when you defect and the other cooperates.
• R (Reward for Mutual Cooperation): What you get if both cooperate.
• P (Punishment for Mutual Defection): What you get if both defect.
• S (Sucker’s Payoff): The lowest payoff, when you cooperate and the other defects.

For the game to be a true Prisoner’s Dilemma, the payoffs must satisfy two conditions:

• Condition 1: T > R > P > S
• Condition 2: R > (T + S) / 2

Because defection provides a higher individual payoff regardless of the opponent’s choice, “rational” egoists are driven to defect, resulting in mutual punishment (P) rather than the superior mutual reward (R). Cooperation can only emerge when the game is iterated, allowing for the development of stable patterns based on future expectations.

Chapter 2: The Success of TIT FOR TAT (The 4 Properties for Success)

The effectiveness of various strategies was evaluated through a computer tournament involving a round-robin format of the iterated Prisoner’s Dilemma, and the TIT FOR TAT was the winning strategy.

TIT FOR TAT won with a simple policy: cooperate on the first move and subsequently mimic the opponent’s previous action.

The success of this strategy is derived from four robust functional properties:

1. Niceness: never defect first. This prevents the initiation of unnecessary and costly conflict, ensuring that the maximum possible “Reward” for mutual cooperation is maintained as long as the opponent remains cooperative.
2. Provocability to retaliate immediately: This serves as a deterrent, signaling that exploitation will not be tolerated and will be met with a “Punishment” payoff
3. Forgiveness to resume win-win: After retaliating for a defection, the strategy immediately returns to cooperation if the opponent does. This can prevent the relationship from collapsing into a permanent cycle of mutual defection (but it needs the other strategy to make the first step)
4. Clarity to help recognize reciprocity and cooperate: The strategy is transparent and predictable. This allows the opponent to easily recognize the pattern of reciprocity, making it easier for them to adapt their own behavior toward mutual cooperation for their own self-interest

TPM Note: Utilitarian thinking
This approach aligns well with our concepts of ‘power intelligence‘, Machiavellianism, and ‘successful psychopathy‘ (removing the destructive and predatory aspects of psychopathy).

A Nicer Strategy Would Have Won Round 1, But Get Exploited In Round 2

TIT FOR TWO TATS defects only if the other player defected on the previous two moves. It is a more forgiving version of TIT FOR TAT in that it does not punish isolated defections.

Says the author:

The excellent performance of this TIT FOR TWO TATS rule highlights the fact that a common error of the contestants was to expect that gains could be made from being relatively less forgiving than TIT FOR TAT, whereas in fact there were big gains to be made from being even more forgiving. The implication of this finding is striking, since it suggests that even expert strategists do not give sufficient weight to the importance of forgiveness.

Two other rules that would have won were ‘LOOK AHEAD’ and ‘OPTIMISTIC DOWNING’.

However, the first tournament environment was relatively simple. TESTER, a strategy designed to probe and exploit overly forgiving opponents, would have exploited TIT FOR TWO TATS. Because TESTER never defects twice in a row, TIT FOR TWO TATS never triggers retaliation. As a result, it continues cooperating while TESTER periodically defects and gains the temptation payoff.

Round 2 Shows It Pays to Cooperate, But Not to Be Too Nice

What seems to have happened is an interesting interaction between people who drew one lesson and people who drew another from the first round. Lesson One was: “Be nice and forgiving.” Lesson Two was more exploitative: “If others are going to be nice and forgiving, it pays to try to take advantage of them.” The people who drew Lesson One suffered in the second round from those who drew Lesson Two.

This confirms what Machiavelli said: ‘those who are nice are ruined among the many who are not’

Best Quotes:

• Even expert strategists do not give sufficient weight to the importance of forgiveness.
• Even expert strategists (…) made the systematic errors of being too competitive for their own good, not being forgiving enough, and being too pessimistic about the responsiveness of the other side.

Chapter 3: The Chronology of Cooperation

The emergence and maintenance of cooperation are governed by three distinct evolutionary phases:

1. Initial Viability: In a population of individuals who always defect, cooperation cannot emerge through isolated individuals. However, it can take hold if it starts in “clusters” of individuals who use reciprocity (like TIT FOR TAT) and have a significant proportion of their interactions with each other.
2. Robustness: Once a strategy is present, its robustness is measured by how well it performs in a diverse environment containing many different types of strategies. Reciprocity is robust because it can elicit cooperation from many other strategies and protect itself against exploitation.
3. Stability: A strategy is “collectively stable” if it can resist invasion by a rare mutant or a small group of individuals using a different strategy.
   • TIT FOR TAT STABILITY: In an evolutionary sense, TIT FOR TAT is stable because it prevents “mean” mutations from gaining a foothold, as it immediately punishes defection while rewarding the cooperation of its own “kin” or similar strategists.

A critical requirement for this stability is the “Shadow of the Future.” This is mathematically represented by the discount parameter ($w$), which reflects the importance of the next interaction relative to the current one. If the probability of meeting again is sufficiently high, the long-term gains of cooperation outweigh the short-term gains of defection.

Chapter 4: The Live-and-Let-Live System in Trench Warfare

The “Live-and-Let-Live” system during World War I provides a historical demonstration of cooperation emerging in a high-stakes, hostile environment without formal agreements. In the static environment of trench warfare, the months-long time spent in close quartiers with the enemies created an iterated game. The author refers to high “shadow of the future“, meaning that the anticipation of future interactions drives current cooperation.

Cooperation manifested through several mechanisms:

• Mutual Restraint: Units would refrain from firing on specific supply lines or during mealtimes, provided the other side reciprocated.
• Ritualization: Artillery and snipers would demonstrate their accuracy by hitting specific, non-human targets (like a specific brick in a wall) at predictable times. This signaled that they could kill, but chose not to, as long as the peace was maintained.
• The Role of Retaliation: Immediate, proportional response also maintained the truce. If one side broke the informal truce, the other side would respond with an equal or greater volume of fire to restore the “balance of terror.”

🙋🏼‍♂️Lucio’s Note: This is a highly relevant example, and it’s important to contextualize it and remember that cooperation was the exception rather than the norm

Chapter 5: The Evolution of Cooperation in Biological Systems

Chapter 5 is written in cooperation with William D. Hamilton, providing his experitse on biology and evolution.

To account for cooperation, evolutionary theory proposes two approaches that could facilitate cooperation:

• Genetical kinship theory: Close relatedness of players permits true altruism—sacrifice of fitness by one individual for the benefit of another.
• Reciprocity theory by superior payoffs among cooperators

Cooperation does not require intelligence or conscious intent and it can evolve through the mechanical process of natural selection. In biological terms, a strategy’s “payoff” is reflected in the fitness boost it provides: increasing the odds of survival and reproduction.

The authors claim that reciprocity should start with clusters. A cluster provides a nontrivial proportion of the interactions with other cooperators.

Key biological insights include:

• Discrimination is necessary for cooperation: For cooperation to evolve, individuals must be able to recognize each other and remember the outcomes of past interactions. This prevents “cheaters” from exploiting the population indefinitely.
• Repeated interactions increase win-win (Mutualism): Inter-species cooperation (such as cleaner fish and their hosts) relies on the fact that both parties receive a higher fitness benefit from the interaction than they would from defection.
• One-off interactions increase the odds of defection and predation (Impermanence): impermanence of association tends to destabilize symbiosis and promote win-lose and predation
• Cooperation collapses and turn exploitative when future interactions are unlikely: Partners and parasites—including microbes—shift from mutualism to exploitation when the probability of continued interaction drops, such as during illness, injury, or aging.
  Example: pathogens are highly aggressive and exploitative during the “infectious stage” to maximize spreading to new hosts, but are less aggressive during a “chronic phase” once they are settled in, keeping the host alive only as long as a continued association remains profitable.

Also read:

• The Origins of Virtue, a more recent account focused on evolution by Matt Ridley

Chapter 6: How to Choose Effectively (Advice for Players)

Based on the findings of the computer tournaments, four strategic principles are recommended for individuals in reciprocal situations:

• Don’t be envious: Avoid comparing your score to the other player’s score. TIT FOR TAT never actually “wins” a single game (it either ties or loses slightly), yet it wins the tournament overall because it maximizes its total absolute score by encouraging mutual cooperation.
• Don’t be the first to defect: Maintaining a “nice” reputation is functionally superior because it avoids the heavy costs associated with mutual punishment.
  Exception: when the future is not important enough compared to immediate gains, or when the other player is likely to never be seen again
• Reciprocate both cooperation and defection: A strategy must be “provocable” to prevent exploitation. Pure cooperation (being a “Sucker”) is not a stable strategy because it invites aggression.
• Don’t be too clever: Overly complex strategies are often perceived as random or “shifty” by others. Simplicity allows the other player to understand your reciprocity and coordinate with you effectively.

Don’t Be Envious

On the importance of not letting envy stand in the way of win-win:

TIT FOR TAT achieves either the same score as the other player, or a little less. TIT FOR TAT won the tournament, not by beating the other player, but by eliciting behavior from the other player which allowed both to do well. TIT FOR TAT was so consistent at eliciting mutually rewarding outcomes that it attained a higher overall score than any other strategy.
So in a non-zero-sum world you do not have to do better than the other player to do well for yourself. This is especially true when you are interacting with many different players. Letting each of them do the same or a little better than you is fine, as long as you tend to do well yourself. There is no point in being envious of the success of the other player, since in an iterated Prisoner’s Dilemma of long duration the other’s success is virtually a prerequisite of your doing well for yourself.

Note that the behavior still allowed TIT FOR TAT to win overall, even when not always scoring more tha various individual strategies it engaged with.

Reciprocate Both Cooperation and Defection

Best insights and takeaways:

• Reciprocity is a necessity: The extraordinary success of TIT FOR TAT leads to some simple, but powerful advice: practice reciprocity.
• Provocability is necessary for nice approaches: Proposition 4 demonstrates that being provocable is actually required for a nice rule like TIT FOR TAT to resist invasion.
• Retaliation optimum balance: What is clear is that extracting more than one defection for each defection of the other risks escalation. On the other hand, extracting less than one-for-one risks exploitation. (…) While the exact balance will be hard to determine in a given environment, the evidence of the tournament suggests that something approaching a one-for-one response to defection is likely to be quite efFective in a wide range of settings.
• Retaliate calibrating to the environment: the precise level of forgiveness that is optimal depends upon the environment. In particular, if the main danger is unending mutual recriminations, then a generous level of forgiveness is appropriate. But, if the main danger is from strategies that are good at exploiting easygoing rules, then an excess of forgiveness is costly.
  Later on, the author acknowledges that ‘limited provocability’ enhances returns when stable cooperation is the goal

Don’t be too clever

• Make it easy to predict cooperation: In chess, it is useful to keep the other player guessing about your intentions (…) the less efficient will be his strategy (…) to your benefit. But in a non-zero-sum setting (…) you benefit from the other player’s cooperation. The trick is to encourage that cooperation. A good way to do it is to make it clear that you will reciprocate. Words can help (…) but actions speak louder than words.
• Don’t be too calculative: unlike the chess opponent, the other player in a Prisoner’s Dilemma should not be regarded as someone who is out to defeat you. The other player will be watching your behavior for signs of whether you will reciprocate cooperation or not, and therefore your own behavior is likely to be echoed back to you.

This is why some of the popular literature on power and manipulation is often misleading when it states among its laws of ‘keeping others in suspended terror’ as a leadership tactic.

Chapter 7: How to Promote Cooperation (Advice for Reformers)

For those in positions of authority or systemic design, cooperation can be encouraged by altering the environment of the interaction:

• Enlarge the shadow of the future: Make the future more important relative to the present. Ie..: Make interactions more frequent, durable, and predictable. When players know they will interact many times, the incentive to cooperate increases. The two main strategies are:
  • More durable: for example, marriage
  • More frequent: when the next interaction happens sooner, the next move looms larger
• Change the payoffs: Use regulations or incentives to make the Reward (R) for cooperation more attractive or the Punishment (P) for defection more severe.
• Teach people to care about each other: While not strictly necessary for cooperation, social norms that value the other’s well-being can supplement the strategic incentives.
• Teach reciprocity: Explicitly promoting TIT FOR TAT-like behavior helps individuals realize that cooperation is in their long-term self-interest.
• Improve recognition abilities: Systems that increase transparency and allow for the easy identification of defectors (such as credit ratings or reputations) make it harder for exploitation to occur.

Chapter 8: The Social Structure of Cooperation

Social structures provide additional layers of stability or conflict to cooperative systems. These include:

• Labels/Stereotypes: Stereotypes or observable traits (race, gender, status) can be used as proxies for expected behavior. This can lead to self-fulfilling prophecies where certain groups are discriminated against or where hierarchies of “dominance” and “deference” are maintained through the threat of retaliation.
• Reputation for provocability: A player’s history with others affects how a new partner will treat them. This makes it vital to maintain a reputation for being provocable but nice.
• Territoriality: When players interact mostly with their neighbors, cooperative clusters can survive even if the global environment is hostile. This “local” reinforcement allows cooperation to spread outward like a wave.
  • The advantage of great success when diffusion by imitation rules: in a social system based on diffusion by imitation, there is a great advantage to being able to attain outstanding success, even if it means that the average rate of success is not outstanding. This is because the occasions of outstanding success win many converts.
• Regulation is most effective when obedience is convenient: Governments cannot enforce any rule it chooses and must instead elicit compliance from the majority. So they are most effective when they align with the natural incentives for reciprocity already present in the system.

The author here touches upon social power dynamics and status classes/castes (see ‘Social Dominance Theory‘ for more).
He says that marginalized minorities who are short-changed by the majority may be better seeking defensive isolation and sticking to cooperation against isolated defections rather than facing unending punishment.

A Bully’s Reputation to Stave Off Defection & ‘Maximum Return’s Defection Rate’

Having a firm reputation for using TIT FOR TAT is advantageous to a player, but it is not actually the best reputation to have. The best reputation to have is the reputation for being a bully. The best kind of bully to be is one who has a reputation for squeezing the most out of the other player while not tolerating any defections at all from the other.

Note that this may not apply to humans because of social dynamics that create push-backs against bullying (see Lucio’s analysis below).
Even then, a bully reputation is not easy says the author because it requires lots of defection, escalations (including with other wannabe bullies), and costs:

Fortunately, it is not easy to establish a reputation as a bully. To become known as a bully you have to defect a lot, which means that you are likely to provoke the other player into retaliation. Until your reputation is well established, you are likely to have to get into a lot of very unrewarding contests of will. For example, if the other player defects even once, you will be torn between acting as tough as the reputation you want to establish requires and attempting to restore amicable relations in the current interaction.

But this Machiavellian approach makes sense from a perspective of pure maximization of returns:

The way to squeeze the most out of the other is to defect so often that the other player just barely prefers cooperating all the time to defecting all the time. And the best way to encourage cooperation from the other is to be known as someone who will never cooperate again if the other defects even once.

And the author ends again with the advantages of TIT FOR TAT’s ‘assertiveness‘:

Using TIT FOR TAT is an effective way of holding still and letting the other player do the adaptation. It refuses to be bullied, but does not do any bullying of its own.

Chapter 9: The Robustness of Reciprocity

The study of cooperation leads to a cautiously optimistic conclusion. Cooperation based on reciprocity can emerge from small clusters, thrive in diverse environments, and protect itself once established.
This “ratchet effect” suggests that once the conditions for reciprocity are met—primarily a sufficiently long “shadow of the future”—cooperation can become a stable and dominant feature of a system. This logic applies across scales, from the micro-level of biological cells to the macro-level of international relations, proving that mutual gain is possible even among individuals who are purely self-interested.

The Risk of Eliminating or Bankrupting Your Opponent

Axelrod seems to refer to US policy towards the former USSR here:

Occasionally a political leader gets the idea that cooperation with another major power should not be sought because a better plan would be to drive them into bankruptcy. This is an extraordinarily risky enterprise because the target (…) would also have a strong incentive to escalate the conflict before it was irreversibly weakened.
Japan’s desperate gamble at Pearl Harbor, for example, was a response to powerful American economic sanctions aimed at stopping Japanese intervention in China (Ike 1967; Ho-soya 1968). Rather than give up what it regarded as a vital sphere, Japan decided to attack America before becoming even further weakened. Japan (…) decided that the cumulative effects of the sanctions made it better to attack rather than to wait for the situation to get even more desperate.

More Win-Win Games Than Competitions?

We are used to thinking about competitions in which there is only one winner, competitions such as football or chess. But the world is rarely like that. In a vast range of situations mutual cooperation can be better for both sides than mutual defection. The key to doing well lies not in overcoming others, but in eliciting their cooperation.

Propositions

SHARPEST INSIGHTS

These are the insights that struck us the most.
We use the authors’ quotes and research, but the titles and interpretations are our own—and may not reflect the authors’ intent.
Credit for the insight goes to them; blame for the interpretation is on us.

Cooperation Requires Power & Continued Value-Provisioning

From chapter 3, on cooperation requiring future interactions and ‘the ability to provide value’:

if the other player is unlikely to be around much longer because of apparent weakness, then the perceived value of w falls and the reciprocity of TIT FOR TAT is no longer stable.

And the author tightly links it to a real-life scenario from a historical power struggle:

We have Caesar’s explanation of why Pompey’s allies stopped cooperating with him. “They regarded his [Pompey’s] prospects as hopeless and acted according to the common rule by which a man’s friends become his enemies in adversity” [translated by Warner 1960, p. 328].

And to a modern business scenario:

Another example is the case where a business is on the edge of bankruptcy and sells its accounts receivable to an outsider called a factor. This sale is made at a very substantial discount because once a manufacturer begins to go under, even his best customers begin refusing payment for merchandise, claiming defects in quality, failure to meet specifications, tardy delivery, or what-have-you. The great enforcer of morality in commerce is the continuing relationship, the belief that one will have to do business again with this customer, or this supplier, and when a failing company loses this automatic enforcer, not even a strong-arm factor is likely to find a substitute. (Mayer, 1974, p. 280)

And the author, with one final example from politics:

Similarly, any member of Congress who is perceived as likely to be defeated in the next election may have some difficulty doing legislative business with colleagues on the usual basis of trust and good credit

The author ends by adding that cooperation is also easier in smaller towns or ethnic neighborhoods, although I believe ethnicity is a confounding factor.

Cooperation Is Often Good… Unless It’s For a Clique, to Society’s Detriment

Smaller alliances within larger society can promote a small group’s interest at the cost of everyone else:

There are times when one wants to retard, rather than foster, cooperation between players. Collusive business practices are good for the businesses involved but not so good for the rest of society. In fact, most forms of corruption are welcome instances of cooperation for the participants but are unwelcome to everyone else. So, on occasion, the theory will be used in reverse to show how to prevent, rather than to promote, cooperation.

This is also the case with mafias and various ‘secret societies’.

Reciprocity is Honorable & Supports Societal Functioning

Unconditional cooperation can not only hurt you, but it can hurt other innocent bystanders with whom the successful exploiters will interact later. Unconditional cooperation tends to spoil the other player; it leaves a burden on the rest of the community to reform the spoiled player, suggesting that reciprocity is a better foundation for morality than is unconditional cooperation.

And then a crucial point we wholeheartedly agree with here at TPM:

Reciprocity is certainly not a good basis for a morality of aspiration. Yet it is more than just the morality of egoism. It actually helps not only oneself, but others as well. It helps others by making it hard for exploitative strategies to survive.

But there is also a return for the self:

But you also have a private advantage from another person using reciprocity even if you never interact with that person: the other’s reciprocity helps to police the entire community by punishing those who try to be exploitive. And this decreases the number of uncooperative individuals you will have to deal with in the future.

QUOTES

It’s value exchanges above trust:

The foundation of cooperation is not really trust, but the durability of the relationship.

On the importance of continued interaction for cooperation to emerge and yield benefits:

The very possibility of achieving stable mutual cooperation depends upon there being a good chance of a continuing interaction

On the four basic strategies of successful cooperation:

The analysis of the data from these tournaments reveals four properties which tend to make a decision rule successful: avoidance of unnecessary conflict by cooperating as long as the other player does, provocability in the face of an uncalled for defection by the other, forgiveness after responding to a provocation, and clarity of behavior so that the other player can adapt to your pattern of action.

On the ‘power of kindness’:

in fact there were big gains to be made from being even more forgiving. The implication of this finding is striking, since it suggests that even expert strategists do not give sufficient weight to the importance of forgiveness.

On the importance of starting off cooperative:

The lesson is that not being nice may look promising at first, but in the long run it can destroy the very environment it needs for its own success.

🙋🏼‍♂️ Lucio’s Analysis

Sometimes Over-extends Prisoner’s Dilemma Implications Without Considering Alternative Dynamics (Psychology, Biases, etc.)

I’ve noticed this a few times.
As an example, providing real-life applications of how weakness and non-repeated interactions boost the odds of defection, the author writes:

Conversely, a visiting professor is likely to receive poor treatment by other faculty members compared to the way these same people treat their regular colleagues.

This is often not the case because of different psychological and social-psychological dynamics.
For example, many people, and especially men with a strong ingroup-outgroup psychology, want to show the best possible time to high-value outsiders because it reflects well on their ingroup. It’s a status boost if others think well about ‘their’ group/university/country.

Short-term visits also allow people to present their best selves in those brief interactions, and there are no considerations about possible competition for promotion.

Example 2: Shooting Back / Not Shooting is Also Tied to Psychology

For example, the author writes that people in war shoot back because ‘mutual defection is preferred to unilateral restraint’. That is a good point, but there are also basic psychological dynamics at play, like self-defense and, as the author later also notes, revenge-seeking.

Conversely, not shooting is most certainly in good part because mutual restraint is better, but also because many people are uncomfortable with causing harm and killing. It’s a well-known phenomenon that many soldiers don’t shoot in situations where it makes little difference and they can freeride on other shooters.

Conflates Mobility as Cause of Defection Instead of Mobility As Enabler of Defection

Writes the author:

Therefore, if the other player is not likely to be seen again, defecting right away is better than being nice.
This fact has unfortunate implications for groups who are known to move from one place to another. An anthropologist finds that a Gypsy approaches a non-Gypsy expecting trouble, and a non-Gypsy approaches a Gypsy suspiciously, expecting double-dealing.

However, rather than being forced into a defector strategy because of moving, some cheater strategies may move because they adopt a cheater strategy.

For example, psychopathy has been linked to frequent moving, and I doubt that moving is the cause of psychopathy.

Unaccounted Costs for Bully’s Reputation?

The author writes:

The best reputation to have is the reputation for being a bully. The best kind of bully to be is one who has a reputation for squeezing the most out of the other player while not tolerating any defections at all from the other.

While this may apply to digital programs, the considerations are different with humans.
The author is correct in acknowledging later that creating such a reputation is costly, but I believe it goes beyond simple ‘loss of returns from defection’.
The reputation for a bully may also decrease future cooperation once the reputation is established because other cooperators avoid the bully. And it may boost escalations and power showdowns as others approach him ready for an escalation and a fight, rather than cooperation.

It may also encouarage what have been called ‘leveling coalitions’, or coalitions of smaller players who ‘gang up’ against the bully.

We have seen similar dynamics in real world politics, see for example:

• US foreign policy analysis: How US forceful approach strengthens others and decreases US global influence

Chapter 5 Forgets About the Need For Power/Retaliation

This crucially important concept is directionally, but in my opinion, it could have been better fleshed out and misses an important mechanism of power dynamics:

Other mutualisms are also characteristic of situations where continued association is likely, and normally they involve quasi-permanent pairing of individuals, or of inbred or asexual stocks, or of individuals with such stocks (Hamilton 1972 and 1978). Conversely, conditions of free-mixing, and transitory pairing conditions where recognition is impossible, are much more likely to result in exploitation—parasitism, disease, and the like.
(…)
Again, it is seen that impermanence of association tends to destabilize symbiosis.

Axelrod and Hamilton’s argument is generally correct, particularly in cases where interests perfectly converge, such as vertical transmission through the egg and when the host’s health boosts the parasite’s own survival and reproduction.
In these specific scenarios, the guest’s future is biologically tied to the host’s reproductive success, making any defection or “cheating” self-harming and potentially self-destructive for the parasite in case of the host’s death.

However, the logic fails to account for power asymmetry and the importance of retaliation.

If a host lacks the mechanical ability to retaliate or expel a parasite/predator, it becomes a permanent “Sucker,” allowing the parasite to stabilize an equilibrium of optimized exploitation, rather than neutrality or optimized win-win.
Since the payoff for the temptation to defect (T) can be higher in real life than the reward for mutual cooperation (R), an unpunished parasite may evolve to extract the maximum sustainable resources rather than transition to a win-win outcome.

In these power-asymmetric relationships, permanence does not facilitate cooperation but only provide a longer timeline for one-sided extraction.

REVIEW

The Evolution of Cooperation is a seminal work on strategy and a foundational text for The Power Moves. It formalizes how cooperation, retaliation, and long-term incentives shape outcomes in repeated interactions.

While such an important work would have benefited from tighter editing, we are grateful to the author because it remains a core resource for understanding strategic interaction and supports many of this website’s frameworks, strategies, and principles.

Read some in:

• Effective cynicism
• Effective calculativeness, which also applies the ‘law of reciprocity’
• Smart cooperation: the main article and strategy based on cooperation