Skip to main content

Verified by Psychology Today


Beyond Revenge: The Evolution of the Forgiveness Instinct

Psychology, science and cultural evolution

For those familiar with Aristotle’s Golden Mean, the Buddhist Middle Way, or Bertalanffy’s principle of dynamic equilibrium, the emergence of a new synthesis of positive and negative psychology will be no surprise. A more balanced approach to scientific analysis of the ‘good’ and the ‘bad’ in human nature is now emerging, and few researchers approach this balanced approach to analysis better than does Michael McCullough in his book, Beyond Revenge: The Evolution of the Forgiveness Instinct.

McCullough begins by placing side-by-side two classics texts on ape society: Frans de Waal’s book, Good Natured, and Wrangham and Peterson’s book, Demonic Males. An obvious and yet striking contrast emerges. Ape society includes highly moral behaviors: setting and enforcing rules that benefit the group, sharing with the needy, sympathizing with those that suffer, offering consolation to the vanquished, and returning favors to the generous. Nevertheless, coalitional bonds between males within the ape community often serve to fuel an attack-on-sight ethic toward individuals from other communities, and McCullough reminds us that this tendency to maintain one set of rules for kith and kin, and a very different set of rules for outsiders is similarly prominent in human societies.

More generally, McCullough argues that revenge is not a disease and forgiveness the cure: both revenge and forgiveness are aspects of human nature, instinctual patterns of behavior that are context sensitive, and if we want to make the world a less vengeful, more forgiving place we need to make social environments less abundant in the factors that evoke the desire for revenge and more abundant in the factors that evoke the forgiveness instinct.

McCullough highlights a number of important factors:

First, people are more likely to forgive those whom they feel close, care about; or those who are suffering and who cannot help themselves.

Second, people forgive to the extent that they perceive their relationship with the transgressor to be a valuable one.

Third, people forgive to the extent that they perceive the transgressor as unwilling or unable to harm them in the future. McCullough believes that these three variables – careworthiness, expected value, and perceived safety – are open to manipulation in projects of social redesign.

The key challenge in promoting forgiveness involves creating the social conditions that signal and activate careworthiness, value, and safety amongst ‘strangers’ -- because ‘friends’ generally signal and activate successfully in this respect -- and one of the key influences involved in the process of transforming ‘strangers’ into ‘friends’ is the extent to which society creates opportunities for people to work together toward common goals. Working toward common goals facilitates cooperation, and cooperation is a foundation stone for friendship, forgiveness, and the kinds of group dynamics that activate and maintain forgivness behaviours over time.

From an evolutionary perspective, McCullough sees forgiveness and revenge as two sides of the same coin, a coupled pair of instincts that have evolved as an evolutionarily stable strategy (ESS), a natural byproduct of an evolutionary winnowing process. He draws upon game theory simulations to argue that tit-for-tat strategies of cooperation and retaliation are critical for ensuring inclusive fitness in a group -- cooperate if your partner cooperates, retaliate if your partner defects, and forgive your partner and return to cooperation if they repent. There are different versions of same strategy (e.g., generous tit-for-tat, which unconditionally forgives one third of the time after a defection; and contrite tit-for-tat, which forgives righteous anger), but all game theory simulations point to the same conclusion: only organisms that can forgive those they work with will survive the evolutionary winnowing process.

However, the results of any game theory simulation depend on the assumptions used in the mathematical specification of game contingencies. For example, if we add the assumption that we are ‘stuck with’ those around us (family, neighbors, work colleagues) and thus suffer more as a result of retaliation (i.e., if we model functional proximity contingencies), game theory simulations find evidence for success of the strategy known as ‘very generous tit-for-tat’. Very generous tit-for-tat will forgive ‘buddies’ unconditionally about two-thirds of the time. Furthermore, if we add assumptions about the spread of gossip and the consequences of having a ‘good’ reputation (he cooperates with group members) or a ‘bad’ reputation (he takes advantage of the group), another set of optimal play strategies emerge. Gossip often helps to inhibit the bad behavior of individuals in a group, as individuals seek to avoid 'revenge' from the friends of the person they have hurt.

Although game theory simulations assume that behavior is under the control of rational contingencies, researchers have found that information that is independent of the formal structure of an experimental game can impact on players’ behavior. For example, one study reported that oblique eye contact or light taps on a shoulder or arm significantly increasing contributions by males in an economic game (Kurzban, 2001). Likewise, game play strategies are a function not only of game play contingencies, but also of how games are cognitively framed. For example, people are more generous when a game is framed as a community social event than when it is framed as an economic investment (Pillutla and Chen, 1999). Thus, unlike mathematicians, most people do not see these games as abstract structures that can be logically analyzed; some cognitive framing is required to determine the context that agents find themselves in.

Finally, in the context of rational manipulation of game play contingencies and subsequent levels of cooperation/forgiveness and conflict in a group, both McCullough and game play theorists need to consider two related issues: group size and problem complexity. McCullough touches upon the problem of expanding small group game solutions (e.g., tit-for-tat) to large scale issues of mass behavior and international affairs, but one problem for game theory currently is that it is not applicable to the analysis of human action in a mass behavior context, particularly in situations where there are a continuum of agents, each of whom operate in different sub-groups, with different sub-cultures, different roles, and different rules governing their behavior. Consistent with McCullough’s view, some of the same core principles of behavior change may be critical in shaping cooperation and forgiveness in this context – signal and activate careworthiness, expected value, and perceived safety throughout the whole social network -- but other variables also come into play, and this takes us to the second, related issue: problem complexity.

Many of the problems that drive conflict are complex problems and when working with a group it is important to help a group understand the nature of the complex problems they face. To my mind, the future of effective conflict resolution – a fundamental theme of McCullough’s book – rests in the hands of those with knowledge of applied systems science. Such systems science methods are available for dealing with complex problems (Warfield, 1974, Warfield, 2006, Warfield and Cárdenas, 1994), and have been successfully applied to problems of conflict resolution (Broome, 2006). However, too few psychologists and sociologists are aware of these methods, or have experience using them (see also my earlier blog post on Designing our Children’s Future).

But none of this is designed to detract in any way from Michael McCullough’s book. This is a truly outstanding book: clear, concise, balanced, well-argued, insightful, enlightening. A must read for anyone interested in the current state of the world.

Follow Michael Hogan on Twitter


BROOME, B. J. 2006. Applications of Interactive Design Methodologies in Protracted Conflict Situations. Facilitating group communication in context: Innovations and applications with natural groups., Hampton Press.

KURZBAN, R. 2001. The social psychophysics of cooperation: Nonverbal communication in a public goods game. Journal of Nonverbal Behavior, 25, 241-259.

PILLUTLA, M. M. & CHEN, X. P. 1999. Social norms and cooperation in social dilemmas: The effects of context and feedback. Organizational Behavior and Human Decision Processes, 78, 81-103.

WARFIELD, J. N. 1974. Structuring complex systems, [Columbus, Ohio,, Battelle Memorial Institute.

WARFIELD, J. N. 2006. An introduction to systems science, Singapore, World Scientific.

WARFIELD, J. N. & CÁRDENAS, A. R. 1994. A handbook of interactive management, Ames, Iowa, Iowa State University Press.

More from Michael Hogan Ph.D.
More from Psychology Today