KeywordsAggressive Behavior Physical Aggression Indirect Aggression Sexual Infidelity Intrasexual Competition
Aggression generally refers to the intended harming or targeted injury of one organism by another.
Among humans and other animals, there exists a variety of behaviors by which individuals try to impose costs on other individuals. Male walruses fight for dominance, lions hunt for prey, a man smashes a mosquito, a woman poisons her abusive husband, etc. Given this variety of aggressive behaviors, aggression is sometimes defined in broad, abstract terms (e.g., the targeted infliction of disorder on one organism by another; Tooby and Cosmides 2010). Sometimes aggression is defined in more specific terms to demarcate a subset of aggressive behaviors (e.g., using the term indirect aggression for inflicting harm via the manipulation of other people or gossip). Researchers in biology, psychology, anthropology, sociology, public health, psychiatry, and political science have worked to document and understand aggressive behavior. Below is an overview of some of this research with a focus on human aggression. The overview follows Tinbergen’s four questions and thus discusses why aggression might have evolved, how aggression might have evolved in humans, how aggression develops over the human life course, and how the psychological mechanisms that regulate aggression work.
Why Has Aggression Evolved?
The short answer to why aggressive behaviors have evolved is that aggression sometimes pays off; sometimes the benefits of violence are greater than the costs. Survival machines that would be more likely to aggress in those circumstances, in which it is likely to yield resources relevant for survival and reproduction, would reproduce more than those that would not aggress in those situations. Over successive generations, such aggressive survival machines would become more numerous because of natural selection (Dawkins 2006). Of course, there are different resources that allow organisms to reproduce, such as food, safe places to sleep, safety of offspring, high social status, or mating opportunities. In general, organisms might use aggression for two distinct goals: organisms can use aggression to remove obstacles to their fitness (e.g., humans might remove snakes from their homes), and they can use aggression as bargaining power (e.g., a stronger man might bargain with a weaker man for a larger than equal share of some resource; Tooby and Cosmides 2010). The costs and benefits of aggressive behavior depend on the resource at stake and the features of the organisms involved. The aggressive hunting behavior of lions (in which a lion and its prey are in a contest over the use of the prey’s meat) has probably evolved for slightly different reasons than the aggressive intrasexual competition among male walruses (in which two male walruses are in a contest over opportunities to mate with certain female walruses).
Some of the adaptive problems relevant to human aggression have been illuminated by theorizing about rational agents. The logic of the costs and benefits of violence is rather complex when aggression occurs between rational agents (Pinker 2002, 2011). In the seventeenth century, Thomas Hobbes described three incentives for violence: competition, diffidence (fear or distrust), and glory (honor or credibility). When there is competition over some resource, the agents will invade (aggress) for gain. When there is distrust among the agents, they will invade for safety. Because of the incentive for credibility, the agents will invade for reputation. Hobbes noted that these three incentives explained why humans might fight over resources like wives and cattle, why they might fight to defend these resources, and why they might fight over insults and other signs of undervalue.
To understand why agents might aggress for reputation, consider how the three incentives are connected: survival machines that obtain more reproductively relevant resources outcompete others. Therefore, survival machines have reason to suspect that others will want to take their resources (e.g., by theft or murder). This sets of a vicious circle. Because each agent has reasons to distrust the other, they both have a reason to attack first, to strike preemptively. This vicious circle is called the Hobbesian trap of the security dilemma. There is a solution to this security dilemma: deterrence or having a credible policy of retaliation. Such a policy can consist of (1) refraining from preemptive attacks, (2) being strong enough to survive an attack by the other, and (3) being strong enough to retaliate in kind. Because this policy involves retaliation, it imposes costs on the attacker and so removes the incentive for the other (the attacker) to invade for gain. Furthermore, because this policy involves being strong enough to survive an attack and retaliate, it removes the incentive for the other to invade because of distrust (because the other no longer has to fear being the target of a preemptive attack). Note that deterrence only works when the policy of retaliation is credible. If one agent doubts that the other is able to retaliate, the agent has reason to suspect a preemptive attack, which sets off the vicious circle. To remove any doubt that one is able to retaliate, the agent could strive to respond to each sign of undervalue – however slight – with retaliation. Hobbes also described another solution to the security dilemma: the Leviathan, a bystander that punishes any use of aggression. Because the agent can expect to be punished when using aggression, there is no longer an incentive to invade for gain. This removes the incentive to invade for safety, which in turn removes the incentive to invade for reputation.
This theorizing about rational agents suggests that humans might have evolved aggressive traits to manage distinct adaptive problems (gaining resources, defense, and reputation). Possible additional adaptive problems for which humans might have evolved aggressive traits are inflicting costs on intrasexual rivals, negotiating dominance hierarchies, deterring mates from sexual infidelity, reducing resource expenditure on unrelated children, and gaining otherwise inaccessible mating opportunities (Buss and Shackelford 1997). Furthermore, recent research has argued that certain aggressive traits may have evolved by sexual selection, meaning that certain aggressive behaviors might have evolved specifically because individuals of the opposite sex perceive these behaviors as attractive (Archer 2009a).
How Has Human Aggression Evolved?
Aggression is ubiquitous among animals, but it also comes in many varieties. This precludes a simple account of the phylogeny of aggression in humans or how aggression has evolved in humans. Many aggressive behaviors (predation, intrasexual competition) are present in many mammal species and might be traced to aggressive behaviors in common ancestors of these species. Other aggressive behaviors that are prevalent among humans, such as war (aggression between two or more multi-individual coalitions), are not common among mammals (Tooby and Cosmides 1988).
A key question regarding the phylogeny of aggression in humans is whether a particular aggressive trait is a homologue of a similar trait in another species (the traits in the two species evolved from a single trait in a common ancestor) or is an analogue of a similar trait in another species (the traits in the two species evolved independently). There have been multiple studies with animals showing that aggressive behaviors (e.g., hunting, defense of territory) are based on some cost-benefit analysis. In species that are phylogenetically distant, such resemblances in the aggressive traits are likely analogous structures (Archer 2009b).
Some violent traits in humans might be homologues of traits observed in other primate species. For example, a particular aggressive trait observed in both humans and chimpanzees might have evolved from a trait present in the common ancestor of both species. However, there is considerable debate about what kind of aggressive behaviors were the characteristic for the common ancestor of humans, chimps, and bonobos (e.g., Boehm 2012).
Although it might appear that humans have evolved few specifically aggressive features – in contrast to deer with antlers and lions with sharp canines, the human body has few features that appear to be exclusively for harming others – this does not mean that ancestral humans were peaceful. A gruesome form of aggression – cannibalism – seems to have been practiced by ancestral humans (Fernández-Jalvo et al. 1999). Substantial research in archeology and anthropology has explored to what extent warfare has occurred over human evolution. Given that uncertainties are greater when interpreting older archeological findings and when extrapolating from current hunter-gatherer tribes to increasingly ancient times, exactly how much warfare there was among ancestral humans remains a topic of debate.
How Does Human Aggression Develop Over the Life Course?
As humans appear to have evolved few or no conspicuous aggressive organs such as antlers, the development of aggression in humans is not linked to the growth of particular organs. However, humans do seem to reliably develop motivations for aggression.
Recent research suggests the following developmental trajectory of physical aggression in humans: infants develop physically aggressive behaviors such as hitting and biting before the age of two. The frequency of such aggression increases till about the age of three. After that, the frequency of physical aggression decreases (Alink et al. 2006). In addition, it seems that men and women develop to be violent in different ways. Various forms of physical aggression (e.g., fist fights) are more common among men than among women, whereas indirect forms of aggression (gossip, social exclusion) seem more common among women than among men (Archer 2004). Notwithstanding the general pattern that humans of both sexes seem to become less violent as they get older, there is evidence that certain forms of high-risk violence such as homicide are most prevalent among young adult men (e.g., Wilson and Daly 1985).
As alluded to in the discussion of why aggression evolved, it is unlikely that natural selection has evolved indiscriminate aggressive behaviors. That is, over the course of human evolution, there has likely been a selection for being aggressive in those situations where it yields benefits. This implies that humans likely have evolved to develop contingent aggressive behaviors and to regulate their aggression based on some cost-benefit analysis. (Even very aggressive individuals are not aggressive all the time toward all others.) Hence, the development of aggressive behaviors might be influenced by factors that affect the development of psychological mechanisms involved in the cost-benefit analysis or in behavioral inhibition. This reasoning might help explain why certain individual difference variables (e.g., psychopathy, self-control) or environmental variables (e.g., having a mother who smoked during pregnancy) are associated with increased aggression.
Clearly, aggression varies across human societies. So it seems that in some cultures, people grow up to be more aggressive (or differently aggressive) than in other cultures. Evolutionary-minded researchers have studied cross-cultural variation in aggression. Some research suggest that variables related to mating strategies influence the development of aggression. For example, data from foraging societies in the standard cross-cultural sample shows that warfare within societies correlates positively with the proportion of polygyny (Marlowe 2003). Another finding is that homicide rates across the states of the USA correlate positively with teenage birth rates, a variable indicative of a fast reproductive strategy (Hackman and Hruschka 2013).
Of course, various forms of learning are involved in the development of aggressive behaviors. But it makes little sense to talk about human aggression in general being either innate or learned (Pinker 2002). Humans evolved capacities to learn certain things more quickly than others, and some forms of aggression (e.g., smashing a mosquito biting your arm) require less or different learning than other forms of aggression (e.g., bullying, armed combat).
How Do the Psychological Mechanisms That Regulate Aggression Work?
Several commonsense explanations or folk-psychological theories of “how aggression works” have intuitive appeal. It might feel as if a series of insults creates some kind of aggressive pressure that will reduce when you retaliate. This implies a hydraulic or pneumatic model of aggression (i.e., describing aggression regulation mechanisms as resembling pumps, communicating vessels, or other devices involving a liquid or gas-like substance under pressure). A prediction of such a hydraulic model is that expressing aggression (catharsis, “letting the anger out”) should reduce aggression. However, there is not much evidence that catharsis reduces aggression (Bushman 2002). Another commonsense notion is that perceiving violence makes people violent. Although learning certainly influences aggression, researchers have argued that it is time to discard general learning models that assume that aggression is an automatic and maladaptive response to perceiving violence (Ferguson and Dyck 2012).
There are different ways to describe how aggression regulation mechanisms work. An information-processing perspective suggests that aggression regulation mechanisms work like other cognitive systems, using fast and frugal “if-then” heuristics (e.g., Todd and Gigerenzer 2007). As humans can engage in aggression for various reasons (e.g., predation, dominance, revenge, jealousy, sadism), there are probably multiple information-processing mechanisms involved in regulating aggressive behavior, each involving specific heuristics and each implemented in particular neurological pathways. Thus, in terms of information processing, each of the aggression regulation mechanisms consists of if-then rules that convert specific inputs into outputs. As alluded to above, the inputs might consist of specific cues about the costs and benefits of aggression.
Other ways to describe aggression regulation mechanisms involve identifying the common physiological or psychological antecedents, for example, the hormones, neural activity, emotions, or beliefs that are associated with aggression. For example, aggression has been associated with low serotonin (Duke et al. 2013), high testosterone (Dabbs and Morris 1990), and consumption of certain psychoactive substances (e.g., alcohol, cocaine; Boles and Miotto 2003). However, aggression is not produced by a single hormone or neurotransmitter. Research in neuroscience has yielded diverse descriptions of the brain systems underlying aggressive behaviors. Such research typically distinguishes two or more complex neural systems, e.g., systems for affective aggression, predatory aggression, and defensive aggression (Adams 2006; Blair 2004).
Several emotions have been related to aggressive behaviors: anger (Sell et al. 2009), hate (Sell 2013), revenge (McCullough et al. 2013), and jealousy (Harris 2003). Although research of nonhuman animals suggests that fear (or fear-like affective states) is involved in defensive aggression, there has been little research of fear-motivated aggression in humans. The recalibrational theory of anger proposed by Sell et al. illustrates that current research aims to explain aggressive behaviors as being the product of regulatory mechanisms. In short, this theory proposes that anger is a negotiation tactic to bargain for better treatment. For anger to be successful as a negotiation tactic, it should correlate with having a better bargaining position. The research by Sell et al. suggests that some of the mechanisms that regulate aggression are influenced by cues of having a good bargaining position: their data showed that among men, physical strength was a predictor of anger and aggression, whereas among women physical attractiveness was a predictor of anger and aggression.
Aggression has also been related to the phenomena of dehumanization and anthropomorphizing. Humans seem to be able to decrease and increase their attributions of humanlike traits such as intentions and suffering to animals (including humans) and nonanimals (Brandt and Reyna 2011). In general, humans might be less prone to aggress toward entities that are perceived as relatively humanlike (e.g., your pet cat or dog) and more prone to aggress toward entities that are perceived as less humanlike (e.g., a fish, a human who committed a heinous crime; Bastian et al. 2013).
Aggression is part of human life: conflict, rape, violence, and weapons are human universals. As all behaviors, aggressive behaviors are amenable to the four kinds of explanations outlined by Tinbergen (ultimate function, phylogeny, development, proximate mechanism). Human aggression has been studied in various disciplines, with each discipline focusing on particular kinds of aggression. There is no unified or general psychological model of human aggression. One way to organize various forms of human aggression is to distinguish aggressive behaviors based on the identities of the victim and the perpetrator (e.g., parent–child, male–female; Campbell 2005). The challenge for evolutionary psychologists is to continue to use insights of related disciplines to build better explanations of the various forms of human aggression.
- Alink, L. R., Mesman, J., Van Zeijl, J., Stolk, M. N., Juffer, F., Koot, H. M., … & Van IJzendoorn, M. H. (2006). The early childhood aggression curve: Development of physical aggression in 10-to 50-month-old children. Child Development, 77, 954–966.Google Scholar
- Campbell, A. (2005). Aggression. In D. Buss (Ed.), Handbook of evolutionary psychology (pp. 628–652). New York: Wiley.Google Scholar
- Dawkins, R. (2006). The selfish gene, 30th anniversary edition. Oxford, UK: Oxford University Press.Google Scholar
- Pinker, S. (2002). The blank slate: The modern denial of human nature. New York: Viking.Google Scholar
- Pinker, S. (2011). The better angels of our nature: The decline of violence in history and its causes. London, England:Penguin.Google Scholar
- Sell, A. N. (2013). Revenge can be more fully understood by making distinctions between anger and hatred. Behavioral and Brain Sciences, 36, London, England. 36–37.Google Scholar
- Todd, P. M., & Gigerenzer, G. (2007). Mechanisms of ecological rationality: Heuristics and environments that make us smart. In R. I. M. Dunbar & L. Barrett (Eds.), The Oxford handbook of evolutionary psychology (pp. 197–210). Oxford/New York: Oxford University Press.Google Scholar
- Tooby, J., & Cosmides, L. (1988). The evolution of war and its cognitive foundations. Institute for Evolutionary Studies Technical Report, 88(1), 1–15.Google Scholar
- Tooby, J., & Cosmides, L. (2010). Groups in mind: The coalitional roots of war and morality. In Human morality and sociality: Evolutionary and comparative perspectives (pp. 191–234). New York: Palgrave MacMillan.Google Scholar