The Individual Psychology of Evil: What Actually Drives People to Harm Others
As the summer without my Psych of Evil class continues, I wanted to do another post related to the material. So, here is the next major lecture!
In the last post, we established that evil is a label, not a personality type, and that the vast majority of harmful acts are committed by ordinary people operating through ordinary psychological processes. That framing is important, because it opens the door to a question that actually has answers: if evil is something people do rather than something people are, what individual factors make them more likely to do it? That is what this post is about.
I want to be clear upfront that we are not looking for excuses here. Understanding why someone did something harmful is not the same as saying the harm was acceptable. What we are doing is trying to build a real psychological picture of how people arrive at harmful behavior, because that picture is a lot more useful, clinically and socially, than the cartoon version we usually get.
Why Do People Harm Others? Motivation for Evil
The first and most fundamental question is motivation. What moves a person toward harmful action in the first place? If you think about human behavior through the lens of behaviorism, specifically B. F. Skinner's framework of reinforcers and punishments, all behavior is oriented toward obtaining something desirable or avoiding something aversive. Maslow's hierarchy extends this by organizing human needs into levels: biological needs like food, water, and shelter at the base, and ascending through safety, social belonging, esteem, and self-actualization at the top (Maslow, 1943). The key insight here is that the same drives that push people toward prosocial behavior, toward cooperation, toward community, can under certain conditions push people toward harmful behavior as well.
Skinner distinguished between primary reinforcers, things that directly satisfy biological needs like food, warmth, and sex, and secondary reinforcers, things that have acquired value through their association with primary reinforcers. Money is the obvious example of a secondary reinforcer. You cannot eat money or build shelter with it directly, but it functions as a near-universal tool for obtaining things you can. And this is relevant because greed, the pursuit of material wealth and power as ends in themselves, is one of the most historically consistent drivers of what we call evil. Wars of conquest, exploitation, theft, fraud: much of the harm people inflict on one another throughout history can be traced to the pursuit of resources and capital.
Burris describes this through the concept of mind over matter: the calculus that individuals, consciously or not, run when deciding whether to pursue something harmful (Burris, 2022). When the perceived reward for an action outweighs the perceived risk or punishment, the social and psychological brakes that usually inhibit harmful behavior can begin to slip. The Mongol Empire under Genghis Khan is a useful and dramatic example. Their campaign of extreme violence was not, according to Baumeister's analysis, simply wanton cruelty for its own sake. It was instrumental. Cities that surrendered avoided destruction. Cities that resisted were annihilated, and that outcome was deliberately broadcast to the next city down the road. The violence served a strategic purpose, and the cost-benefit calculus of conquest made it, from the perpetrators' perspective, rational (Baumeister, 1997). This is uncomfortable but important: instrumental evil, harm as a means to an end, is far more common than sadistic evil, harm as the end in itself.
Two other motivational structures are worth flagging here. The first is the prisoner's dilemma, a classic model from game theory that illustrates how rational self-interest can produce harmful outcomes even when cooperation would benefit everyone involved. The scenario involves two people who cannot communicate and must each decide independently whether to betray the other or stay silent. If both stay silent, both receive reduced punishment. If one betrays and the other does not, the betrayer goes free and the other faces maximum consequences. If both betray, both are punished. The incentive structure pushes toward betrayal even when mutual loyalty would produce a better collective outcome. This applies to aggression in real life more than you might think: a significant portion of violence is preemptive, people harming others because they fear being harmed first, acting in what they experience as self-defense before any actual attack has occurred.
The second motivational structure is the frustration-aggression relationship. When something blocks a person from obtaining what they want or need, the result is frustration. And frustration, particularly when it is perceived as unjust, is a reliable primer for aggression (Dollard et al., 1939; Harris, 1974). It is worth noting that the relationship is correlational rather than strictly causal: frustration does not automatically produce violence, but it creates a state of emotional readiness in which violence becomes more likely if other conditions are present. Perceived injustice is a particularly potent form of frustration, because it activates not just thwarted goals but also wounded pride and a sense of being under attack, which in turn triggers the felt need to defend oneself.
Empathy: The Behavioral Brake That Can Fail
One of the most important individual-level factors in harmful behavior is empathy, or rather the erosion of it. Empathy is not a simple thing, and researchers distinguish between two broad forms. Cognitive empathy is the intellectual recognition that other people have thoughts and feelings, that they are conscious subjects with their own inner lives. Emotional empathy is the felt experience of other people's emotional states, the thing that makes you wince when someone else is in pain. These are related but separable capacities, and they implicate different neural systems, with the frontal lobe involved in the more deliberate cognitive processes and the limbic system contributing to the more automatic emotional responses (Decety & Jackson, 2004).
Baron-Cohen proposes that empathy exists on a continuum, distributed across the population something like a bell curve (Baron-Cohen, 2011). Most people cluster in the middle ranges, capable of both understanding and feeling the emotional states of others with reasonable accuracy. People at the high end of the scale tend to be altruistic, socially attuned, and highly motivated to avoid causing harm. People at the low end may be indifferent or oblivious to others' suffering without necessarily being aggressive. It is worth being precise here: low empathy does not automatically produce violence. There are plenty of people with limited empathic capacity who do not hurt anyone. The relationship is more conditional than that.
What Baron-Cohen calls empathy erosion is the relevant process for understanding harmful behavior. His argument is that evil, in the sense of deliberately harmful action, requires a reduction in empathy, not necessarily a permanent absence of it (Baron-Cohen, 2011). Something has to reduce the inhibitory effect that empathy usually provides, whether that is situational pressure, ideology, substance use, or repeated exposure to violence. Blum (1994) offers a useful counterpoint here, noting that empathy's relationship to moral behavior is more complicated than a simple more-empathy-means-better-outcomes equation, in part because empathic distress can actually be self-focused and lead people to withdraw from painful situations rather than address them (Blum, 1994). But the core insight remains: whatever empathy is doing, its erosion is consistently present in accounts of harmful behavior, from the first murder a serial killer commits to the organizational dynamics of genocide.
This connects to something worth saying about emotional responses to violence more broadly. Human beings are not, in general, comfortable with harming other people. The first instances of violent action that perpetrators describe, whether we are talking about individual offenders, soldiers, or participants in mass atrocities, are almost always accompanied by significant psychological distress: revulsion, anxiety, guilt, shame. This is not a small thing, and it matters for understanding how harmful behavior escalates over time.
The Self, Esteem, and Why Wounded Pride Is Dangerous
Self-esteem is one of Maslow's esteem needs, and it turns out that the need to feel valued, competent, and respected is a powerful motivator for behavior in both prosocial and antisocial directions. Low self-esteem is associated with increased frustration, and, as we have already established, frustration primes aggression. But the relationship between self-esteem and violence is more nuanced than a simple low-self-esteem-equals-aggression model would suggest.
For one thing, low self-esteem is also associated with depression, social withdrawal, and victimization. It is not a reliable predictor of aggression on its own. What does seem to be more specifically linked to aggression is unstable or fragile self-esteem: a self-concept that is high in surface presentation but brittle underneath, easily threatened, and prone to defensive explosion when challenged. This is the structure Baumeister associates with the most dangerous egotism, where the individual has an inflated and rigid sense of their own importance, and any perceived slight or obstacle reads as an attack that must be repelled (Baumeister, Smart, & Boden, 1996). The concept of narcissistic injury, which we will revisit below, is closely related: when a person whose self-worth depends heavily on external validation encounters humiliation or dismissal, the emotional response can be disproportionately intense and can tip into aggression.
Related to this is the concept of the just-world hypothesis, the widespread belief that the world is fundamentally fair and that people get more or less what they deserve (Lerner, 1980). This worldview has genuine psychological functions; it supports a sense of personal agency and the belief that virtuous behavior will be rewarded. But it has a destructive corollary: if bad things are happening to a person or group, it must be because of something they did. This logic reliably produces victim blaming, enables dehumanization, and has been closely associated with prejudice, discriminatory behavior, and, at its most extreme, ideological justification for genocide. Individualistic cultures, which tend to emphasize personal responsibility and self-reliance above collective accountability, show stronger tendencies toward just-world thinking, and individualism itself is associated with higher rates of interpersonal violence even while it also correlates with resistance to authoritarianism. Culture matters, in other words, and not in a simple way.
Personality Factors: From Orientations to Disorders
Certain stable personality features and orientations are consistently associated with aggressive and harmful behavior. Three ideological clusters deserve mention before we get to clinical personality disorders. Social dominance orientation is the worldview that life is inherently competitive and hierarchical, that the strong dominate the weak as a matter of natural order, and that one must fight constantly to protect and expand one's position. People high in social dominance orientation tend to endorse inequality, show hostility toward outgroups, and support aggressive policies toward those below them in perceived social hierarchies (Pratto et al., 1994).
The authoritarian personality describes individuals who are highly deferential to authority, drawn to rigid hierarchies, and prone to black-and-white thinking about morality and social deviance. Authoritarian personalities tend to be harsh toward perceived rule-breakers and outsiders while remaining compliant with authority figures above them. Right-wing authoritarianism (RWA) extends this into a specific ideology oriented around the belief that the world is dangerous, that traditional hierarchies and values are the primary protection against chaos, and that deviation from those traditions is potentially threatening and deserving of punishment. Both orientations have been consistently linked to prejudice and to support for institutional violence directed at stigmatized groups (Altemeyer, 1996).
At the clinical level, the Cluster B personality disorders include antisocial personality disorder, narcissistic personality disorder, borderline personality disorder, and histrionic personality disorder. Antisocial and narcissistic personality disorders are most directly associated with aggression and predatory behavior, though borderline personality disorder also deserves attention because its hallmark features include highly unstable empathic responses, intense sensitivity to perceived rejection, and the capacity for significant cruelty toward people to whom the individual is closely attached. These are not simple or straightforward relationships, and diagnostic labels carry serious risks of stigmatization and overgeneralization, but the patterns are worth understanding.
The Dark Triad, a framework developed by Paulhus and Williams (2002) that we’ve discussed before, identifies three personality traits that individually predict antisocial behavior and in combination are particularly associated with predatory violence: psychopathy, narcissism, and Machiavellianism (Paulhus & Williams, 2002). Psychopathy involves shallow emotional affect, callousness, impulsivity, and very low empathy. Narcissism, at clinical levels, involves grandiosity, entitlement, and the dangerous vulnerability of the narcissistic injury. Machiavellianism involves calculated manipulation and a cynical exploitation of others as instruments for personal gain. When all three are present together, you get someone who is simultaneously remorseless, entitled, and strategically manipulative, which is a combination that produces significant harm.
Sadism also deserves mention here, both as a personality trait and as a clinical construct. Sadism is the derivation of pleasure from inflicting pain or suffering on others. It was recognized in DSM-III as sadistic personality disorder before being removed in later editions, though it remains acknowledged as both a personality trait and a sexual paraphilia (sexual sadism). What is important to note, and what I will return to below in the context of desensitization, is that sadism is not a fixed and static feature only present in a small number of people from birth. It can develop and intensify through experience, particularly through repeated exposure to violence without adequate psychological processing.
Aggression, Desensitization, and How Violence Becomes Habitual
The frustration-aggression hypothesis, first formally proposed in the 1930s and revised by Harris in the 1970s, proposes that frustration, particularly when experienced as unjust, creates a readiness for aggression (Dollard et al., 1939; Harris, 1974). Displaced aggression occurs when the actual source of frustration cannot be safely targeted, so aggression is redirected toward a more accessible, lower-risk substitute. This is the mechanism behind scapegoating: when people are economically stressed, politically powerless, or otherwise frustrated in ways they cannot directly address, the aggression gets displaced onto a convenient outgroup. History is saturated with examples. Triggered displaced aggression is a related phenomenon in which accumulated frustrations that have not been discharged build up and then produce a disproportionate explosive response to a later, often relatively minor, provocation.
Social learning theory, particularly Bandura's work on modeling and observational learning, establishes that people learn to be aggressive in part by watching others be aggressive and observing whether it produces desired outcomes (Bandura, 1973). When violence is modeled, normalized, and seen to be effective, people are more likely to incorporate it into their own behavioral repertoire. Early exposure to violence in the family or community environment is one of the most reliable predictors of later aggressive behavior, though it is a probabilistic relationship rather than a deterministic one: most people who grow up in violent environments do not become perpetrators, and many do become victims.
Baumeister makes the important observation that violence is habit-forming. Once a person uses violence to obtain something they want and it works, the behavioral reinforcement makes it more likely they will do it again. Each repetition lowers the activation threshold for the next one. This is the micro-level version of what desensitization describes at a broader psychological scale.
Desensitization is a natural response to repeated exposure to any stimulus: over time, the initial emotional reaction diminishes. In therapy, we use this deliberately: exposure therapy for phobias and trauma works precisely by allowing people to confront feared stimuli repeatedly until the anxiety response habituates. But in the context of violence, desensitization means that the guilt, revulsion, fear, and distress that normally function as inhibitors of harmful behavior progressively weaken with each repetition. The members of the Polish police battalions who participated in mass killings during the Second World War reported that the first days were psychologically devastating, vivid in memory and deeply distressing for those who continued. Over weeks and months, the faces blurred together, the distress diminished, and for some the violence shifted into something that was not just tolerable but accompanied by a grotesque form of pleasure, soldiers competing in killing games, finding humor in the destruction. This trajectory from distress to numbness to sadistic enjoyment represents one of the most disturbing psychological progressions in the literature on mass violence.
Alcohol merits a specific mention in this context. It is implicated in roughly half of all violent crime and sexual assault worldwide, a staggering statistic that alone makes it worth understanding. Alcohol suppresses frontal lobe activity, which is precisely the part of the brain most involved in impulse control, empathy processing, and the kind of deliberate moral reasoning that inhibits harmful action. It artificially inflates confidence and threat perception, narrows thinking to crude and immediate categories, and, critically, dulls the guilt and remorse that might otherwise inhibit or curtail violent behavior. The Polish police battalions I mentioned above used alcohol ritually before their killing operations. Ted Bundy was frequently intoxicated during his crimes. These are not coincidences.
Avoiding the Mark of Cain: How Perpetrators Justify Themselves
One of the most psychologically interesting features of harmful behavior is how rarely people who commit it describe themselves as evil. Returning to Burris's concept of the Mark of Cain, the label of evil carries severe social consequences, marking someone as outside the moral community and deserving of exclusion (Burris, 2022). People will go to remarkable lengths to avoid this designation, and the strategies they use are consistent enough to describe as a taxonomy.
Denial is the most basic: “I did not do it, or my role was minimal”. The Nuremberg defendants were extraordinary in their creativity here, finding ways to insist they had been peripheral, following instructions, unaware of the full scope of what was happening. Minimization involves acknowledging that something happened while reducing its moral weight: the victims were not that badly hurt, they suffered less than you think, or they would have died anyway. Justification involves reframing the harmful act as necessary, defensible, or righteous: I was following orders, I was protecting my community, they were the real threat. And victim-blaming locates the source of the harm in the victims themselves, depicting them as dangerous, inferior, deserving, or complicit in their own victimization.
These are not strategies unique to obvious monsters. They are standard features of human moral psychology, used by people at every level of harm from interpersonal cruelty to industrial-scale atrocity. The Nazis at Nuremberg and the abusive partner offering rationalizations after the fact are drawing on the same psychological toolkit. Understanding that is part of what makes this field genuinely unsettling and genuinely important.
Why Is There Evil? Baumeister's Four Roots
To close, it is worth returning to the broad question that opened this series: why is there evil? Baumeister proposes four primary roots, and they map onto much of what we have covered in this post (Baumeister, 1997). The first is material gain: harm committed in pursuit of resources, wealth, power, or status. The second is egotism: harm committed to protect or restore a threatened sense of self, including the violent responses to narcissistic injury and wounded pride we discussed. The third is idealism: harm committed in service of a belief system, whether religious, political, or ethnic, in which the victims are cast as enemies, infidels, or subhuman obstacles to a greater good. Ideological violence is particularly dangerous because it provides moral cover that removes rather than suppresses empathy. And the fourth is sadism: harm committed because the suffering of others has become a source of pleasure, whether through innate tendency or, more commonly, through the desensitization process described above.
It is worth noting that humans have also evolved substantial inhibitions against violence. Empathy, guilt, social norms, legal systems, community bonds: these are all forces that push against harmful behavior, and they work most of the time for most people. The problem, as technology continues to multiply the destructive capacity available to individuals and groups, is that the same forces that can be weakened, eroded, or deliberately dismantled. The study of evil is, in this sense, the study of what removes those brakes, and there is nothing more important to understand.
References
Altemeyer, B. (1996). The authoritarian specter. Harvard University Press.
Bandura, A. (1973). Aggression: A social learning analysis. Prentice-Hall.
Baron-Cohen, S. (2011). The science of evil: On empathy and the origins of cruelty. Basic Books.
Baumeister, R. F. (1997). Evil: Inside human violence and cruelty. W. H. Freeman.
Baumeister, R. F., Smart, L., & Boden, J. M. (1996). Relation of threatened egotism to violence and aggression: The dark side of high self-esteem. Psychological Review, 103(1), 5-33.
Blum, L. (1994). Moral perception and particularity. Cambridge University Press.
Burris, C. T. (2022). Evil in mind: The psychology of harming others. Oxford University Press.
Decety, J., & Jackson, P. L. (2004). The functional architecture of human empathy. Behavioral and Cognitive Neuroscience Reviews, 3(2), 71-100.
Dollard, J., Miller, N. E., Doob, L. W., Mowrer, O. H., & Sears, R. R. (1939). Frustration and aggression. Yale University Press.
Harris, M. B. (1974). Mediators between frustration and aggression in a field experiment. Journal of Experimental Social Psychology, 10(6), 561-571.
Lerner, M. J. (1980). The belief in a just world: A fundamental delusion. Plenum Press.
Maslow, A. H. (1943). A theory of human motivation. Psychological Review, 50(4), 370-396.
Paulhus, D. L., & Williams, K. M. (2002). The dark triad of personality: Narcissism, Machiavellianism, and psychopathy. Journal of Research in Personality, 36(6), 556-563.
Pratto, F., Sidanius, J., Stallworth, L. M., & Malle, B. F. (1994). Social dominance orientation: A personality variable predicting social and political attitudes. Journal of Personality and Social Psychology, 67(4), 741-763.
