SS officers and staff relax at Solahütte, a resort for camp personnel 19 miles from Auschwitz-Birkenau. The photo is from an album taken by Karl-Friedrich Höcker, adjutant to the Auschwitz Kommandant, Richard Baer. Photo courtesy the USHMM


How evil happens

Why some people choose to do evil remains a puzzle, but are we starting to understand how this behaviour is triggered?

by Noga Arikha + BIO

SS officers and staff relax at Solahütte, a resort for camp personnel 19 miles from Auschwitz-Birkenau. The photo is from an album taken by Karl-Friedrich Höcker, adjutant to the Auschwitz Kommandant, Richard Baer. Photo courtesy the USHMM

In 1941, en route from a ghetto to a concentration camp in Ukraine, a Nazi soldier beat my grandfather to death. My father witnessed this murder. His is just one of millions of similar stories, of course, and I grew up aware of how death hovered on the other side of life, and brutality on the underside of humanity. The ‘sapiens’ in Homo sapiens does not fully describe our species: we are as violent as we are smart. This might be why we are the only Homo genus left over in the first place, and why we have been so destructively successful at dominating our planet. But still the question nags away: how are ordinary people capable of such obscene acts of violence?

This duality is also a puzzle to ourselves, at the heart of cosmologies, theologies and tragedies, the motor of moral codes and the tension at the heart of socio-political systems. We know light and we know dark. We are capable of doing terrible things, but also of asking ourselves contemplatively and creatively how that is. The self-consciousness that characterises the human mind is nowhere more baffling than in this problem of evil, which philosophers have been discussing since Plato. An obvious place to look for explanations of evil is in the patterns of behaviour that those who commit atrocities display.

This is what the neurosurgeon Itzhak Fried at the University of California, Los Angeles did with his article ‘Syndrome E’ (1997) in The Lancet. A syndrome is a group of biological symptoms that together constitute a clinical picture. And E stands for evil. With Syndrome E, Fried identified a cluster of 10 neuropsychological symptoms that are often present when evil acts are committed – when, as he puts it, ‘groups of previously nonviolent individuals’ turn ‘into repetitive killers of defenceless members of society’. The 10 neuropsychological symptoms are:

1. Repetition: the aggression is repeated compulsively.
2. Obsessive ideation: the perpetrators are obsessed with ideas that justify their aggression and underlie missions of ethnic cleansing, for instance that all Westerners, or all Muslims, or all Jews, or all Tutsis are evil.
3. Perseveration: circumstances have no impact on the perpetrator’s behaviour, who perseveres even if the action is self-destructive.
4. Diminished affective reactivity: the perpetrator has no emotional affect.
5. Hyperarousal: the elation experienced by the perpetrator is a high induced by repetition, and a function of the number of victims.
6. Intact language, memory and problem-solving skills: the syndrome has no impact on higher cognitive abilities.
7. Rapid habituation: the perpetrator becomes desensitised to the violence.
8. Compartmentalisation: the violence can take place in parallel to an ordinary, affectionate family life.
9. Environmental dependency: the context, especially identification with a group and obedience to an authority, determines what actions are possible.
10. Group contagion: belonging to the group enables the action, each member mapping his behaviour on the other. Fried’s assumption was that all these ways of behaving had underlying neurophysiological causes that were worth investigating.

Note that the syndrome applies to those previously normal individuals who become able to kill. It excludes the wartime, sanctioned killing by and of military recruits that leads many soldiers to return home (if they ever do) with post-traumatic stress disorder (PTSD); recognised psychopathologies such as sociopathic personality disorder that can lead someone to shoot schoolchildren; and crimes of passion or the sadistic pleasure in inflicting pain. When Hannah Arendt coined her expression ‘the banality of evil’ in Eichmann in Jerusalem (1963), she meant that the people responsible for actions that led to mass murder can be ordinary, obeying orders for banal reasons, such as not losing their jobs. The very notion of ordinariness was tested by social psychologists. In 1971, the prison experiment by the psychologist Philip Zimbardo at Stanford University played with this notion that ‘ordinary students’ could turn into abusive mock ‘prison guards’ – though it was largely unfounded, given evidence of flaws in the never-replicated experiment. Still, those afflicted with Syndrome E are indeed ordinary insofar as that they are not affected by any evident psychopathology. The historian Christopher Browning wrote of equally ‘ordinary men’ in the 1992 book of that name (referenced by Fried) who became Nazi soldiers. The soldier who killed my grandfather was very probably an ordinary man too.

Today, biology is a powerful explanatory force for much human behaviour, though it alone cannot account for horror. Much as the neurosciences are an exciting new tool for human self-understanding, they will not explain away our brutishness. Causal accounts of the destruction that humans inflict on each other are best provided by political history – not science, nor metaphysics. The past century alone is heavy with atrocities of unfathomable scale, albeit fathomable political genesis. But it was the advent of ISIS and the surge in youthful, enthusiastic recruits to it that gave Fried’s hypothesis a new urgency, and prompted him to organise, with the neurophysiologist Alain Berthoz at the Collège de France in Paris, three conferences around Syndrome E that between 2015 and 2017 gathered cognitive neuroscientists, social psychologists, neurophysiologists, psychiatrists, terrorism specialists and jurists, some of whose theories and insights I share here. Syndrome E is a useful provocation to an innovative, interdisciplinary discussion of this old problem – and a powerful example of how to frame neuroscientific output in human terms. Already this approach is giving rise to interesting hypotheses and explanations.

As the brain’s functional anatomy reveals itself in increasingly precise ways, neuroscience is growing in its ability to address the complexities underlying our behaviour, violence included. But since we are evolved animals, to investigate the biological bases for behaviour is to look both at the embodied results of evolutionary time and at historical time – at how the evolved circuits of the brain are recruited by cultures, as well as producing cultures. Given that we evolved as inherently social, interactive creatures, neuroscience requires dialogue with other disciplines – the brain has not evolved in isolation, and action always takes place at a moment in time in a particular place with particular meaning. The psychological and cultural environment is central in determining whether and how given biological processes will play out. The traits enumerated by Fried thus encompass a combination of neurological and environmental conditions.

Central to Syndrome E is the symptom of ‘diminished affect’. Most people – except, precisely, psychopaths – shy away from or are extremely reluctant to inflict pain, let alone kill. As the psychiatrist Robert Jay Lifton has shown, it takes brainwashing and coercion to dull our emotional response and to overcome our reticence to cross the line beyond which ‘habituation’ sets in – the Syndrome E symptom whereby the repetition of the act makes it easier to perform. Perpetrators of mass murder and torturers can also love and want the best for their children, while feeling nothing for their victims – an instance of the ‘compartmentalisation’ symptom of Syndrome E. This was probably the case for the anonymous Nazi soldier who killed my grandfather. Family belonging and social belonging are separate. When they meet, as happened in Bosnia and Rwanda when families turned on each other, the group identity prevails. Empathy is rarely universal.

The social neuroscientist Tania Singer at the Max Planck Institute in Leipzig in Germany defines empathy as the ability to ‘resonate’ with the feelings of the other. It develops from babyhood on – as imitation at first, then joint attention – into the ability to adopt the point of view of another, along with a shift in spatial perception from self to other, as if one were literally stepping into another’s shoes. This requires an ability to distinguish between self and other in the first place, an aspect of the so-called ‘theory of mind’ that one acquires over the first five years of life. The developmental psychologist Philippe Rochat at Emory University in Atlanta has shown how children develop an ethical stance by that time as well, and become aware of how their actions can be perceived by others.

But while empathy ensures the cohesion of a group or a society, it is also biased and parochial. Revenge thrives on it. The social psychologist Emile Bruneau at the University of Pennsylvania has demonstrated how it is easily directed at an ‘in-group’ at the expense of an ‘out-group’ that can then be targeted as an enemy, and dehumanised. Its selectivity also explains how we can walk by a homeless person without feeling the need to offer help, or rejoice in nasty gossip about a disliked absentee. Inevitably, we all practise selective empathy, its absence manifest in everyday, non-lethal instances of violence that occur in social and family life, in business and politics. What the psychologist Simon Baron-Cohen at the University of Cambridge calls ‘empathy erosion’ in The Science of Evil: On Empathy and the Origins of Cruelty (2011) is therefore not a sufficient ingredient in the outbreak of extreme violence. But it is a necessary one, opening the way to discrimination and ultimately genocide. As the social neuroscientist Jean Decety at the University of Chicago put it, ‘our hypersociality has a dark side’.

This developmental account can dispel, in part, the mystery of our two faces – of our ability at once to help each other and to kill each other, or to argue ourselves into ‘just wars’. In common with other hominins such as chimpanzees, we have evolved the capacity to cement relationships, communicate and cooperate with those in our immediate environment – and also to attack outsiders and members of other tribes. But our evolved self-consciousness is what defines our humanity even apart from other hominins. What remains puzzling is our continued ability to destroy even as we are able to understand ourselves and to create sophisticated scientific models of our own minds.

Under given circumstances, 70 per cent of a population can take part in crimes as part of a group

Neuroscience gives an interesting physiological model of the emotion of empathy as a complex, dynamic process that unites executive, premotor and sensorimotor functions. It recruits, in particular, the ventromedial prefrontal cortex (vmPFC) and the orbitofrontal context (OFC), with which the vmPFC overlaps in part, and which is crucial for the processing of emotions generated in the amygdala – an evolutionary ancient structure within the limbic system. Lesion to the OFC impairs emotional feeling – and with it, decision-making. With his ‘somatic marker hypothesis’, the neuroscientist Antonio Damasio at the University of Southern California in Los Angeles has shown how bodily feelings that participate in signalling emotions, processed in the OFC and vmPFC, enable appropriate, socially situated decision-making, thereby informing our evaluations of the world, including our moral sense.

In the phenomenon of diminished affect, hyperactivity in these same areas of the frontal lobe inhibits activation of the amygdala. Studies have shown dysfunctional activity of the OFC in people with obsessive-compulsive disorder. It thus might also be involved in the obsessive nature of ideas about one group that justify murderous intent against its members. And the sense of elated hyperarousal – such as that induced by cocaine – that entrains action upon these ideas involves processing in the medial prefrontal cortex (mPFC). In short, in cases of Syndrome E emotional pathways in the brain no longer regulate judgment and action. A breakdown occurs in the feedback between the amygdala and higher, cognitive cortical structures. The acting self splits away from the feeling self, a phenomenon that Fried calls ‘cognitive fracture’. He believes that, under given circumstances, about 70 per cent of the population can be subject to it and be able to take part in crimes as part of a group – as might have happened in the Stanford prison experiment, despite caveats regarding its results.

The acting self of the individual with cognitive fracture feels no empathy. But empathy is not always a reliable guide to appropriate behaviour – we don’t feel empathy for the insects dying because of climate change, for instance, but we can decide rationally to act against the disaster. It can even lead to bad decisions with regard to those at whom it is directed – a surgeon who feels empathy for the patient under drapes should really not operate. There is such a thing as a surfeit of feeling. The psychologist Paul Bloom at Yale University has argued ‘against empathy’, in a 2016 book of that title and elsewhere, suggesting that ‘rational compassion’ is a better barometer with which to evaluate our environment and how we should act upon it. That is to say, members of a group whose mission is to kill its perceived enemies might have the ability for emotional empathy for their group, and no rational compassion for their perceived enemy.

An account of the inability to feel any emotion for such perceived enemies can take us closer to understanding what it is like to have crossed the line beyond which one can maim and kill in cold blood. Observers at the International Criminal Court (ICC) at the Hague note frequently the absence of remorse displayed by perpetrators. The clinical psychologist Françoise Sironi, who assesses perpetrators for the ICC and treats them and their victims, has directly seen what Lifton called the ‘murder of the self’ at work – notably with Kang Kek Iew, the man known as ‘Duch’, who proudly created and directed the Khmer Rouge S-21 centre for torture and extermination in Cambodia. Duch was one of those who felt absolutely no remorse. His sole identity was his role, dutifully kept up for fear of losing himself and falling into impotence. He did not comprehend what Sironi meant when she asked him: ‘What happened to your conscience?’ The very question was gibberish to him.

Along with what Fried calls this ‘catastrophic’ desensitisation to emotional cues, cognitive functions remain intact – another Syndrome E symptom. A torturer knows exactly how to hurt, in full recognition of the victim’s pain. He – usually he – has the cognitive capacity, necessary but not sufficient for empathy, to understand the victim’s experience. He just does not care about the other’s pain except instrumentally. Further, he does not care that he does not care. Finally, he does not care that caring does, in fact, matter. The emotionally inflected judgment that underlies the moral sense is gone.

Such a state involves the fusion of identity with a larger system within which occurs the splitting of the feeling self and the cognitive self, and the concomitant replacement of individual moral values with that system’s norms and rules. Chemistry is operative throughout, as it is in all cerebral and somatic functions – and tweakable by pharmaceuticals. The neuroscientist Trevor Robbins at the University of Cambridge has studied ‘pharmacoterrorism’, and how, for instance, the amphetamine Captagon – used, inter alia, by ISIS members – affects dopamine function, depletes serotonin in the OFC, and leads to rigid, psychopathic-like behaviour, increasing aggression and leading to the perseverance that Fried lists among the Syndrome E symptoms. It shuts off social attachment, and disables all emotional feeling (empathy included), a condition called alexithymia.

This is one simplified neurological account of how murderous action becomes possible. The neuroscience of value and action can help to further explain what might be going on. The OFC is exceptionally developed in humans and primates. As Edmund Rolls at the Oxford Centre for Computational Neuroscience has shown, it plays a crucial role in representing reward value in response to a stimulus: we make choices based on the assignation of value – to an object, an idea, an action, a norm, a person. Our emotions are value-rich, and our actions vary and can be updated according to how they are met in the world, in turn motivating us to seek or avoid a stimulus. Our behaviour can continue in the search of an absent reward – this would be one account of compulsive action, a Syndrome E symptom. The neuroscientist Mathias Pessiglione and his team in Paris have also shown a central role for vmPFC in value-attribution to a stimulus or an idea, whereby we choose to undertake an action based on its attractive reward or its aversive outcome. But when this function is overstimulated, new inputs – such as cries for mercy – have no impact on the attribution of value to the idea, for instance that ‘all you people deserve to die’, and action cannot change. It becomes automatic, controllable by an external agent or leader, independently of any sense of value.

Coercion switches off the sense of responsibility – a chilling finding

But these neurological events signify criminal action only under particular environmental circumstances. The psychiatrist David Cohen and his team at the Pitié-Salpêtrière hospital in Paris evaluated teenage candidates for radicalisation. They found that certain socio-psychological conditions in childhood – such as an absent father or an unstable mother, and a history of foster care – affected the development of identity, in some cases eventually leading to the need to subsume it into a wider group with a transcendental message. Again, group trumps family. As the anthropologist Scott Atran has shown, conflicts are often intractable and non-negotiable because they are conducted in the name of absolute, spiritual values – secular or religious – and not for any utilitarian outcome. These values can seem highly attractive – stronger than family ties.

In her novel Home Fire (2017), the British Pakistani writer Kamila Shamsie showed how a loving, innocent but maladjusted and lost young man of Pakistani origin could fall prey to an ISIS recruiter’s siren call to rejoin a lost father and find fulfilment and belonging in a community depicted as devoted to a greater good. Our narratives, inner and outer, inform and justify the choices we make, conferring on them a coherence that is reassuring and can seem good and right. Coherence rides on the moral sense and masquerades as it, bringing on a cognitive dissonance ‘between what we think and what we do’, as Zimbardo once put it – between what we convince ourselves was an appropriate action, and our deeply held, prior beliefs. Shamsie’s character soon regrets his choice and tries to get away from a violence he cannot stomach, unable to withstand the cognitive dissonance. Not so Nazi doctors, say, who convinced themselves that they were acting for a greater good – in a perverse twist to the equivalence of morality with a concern for the good of others. Heinrich Himmler’s speech in Poznan in 1943 is a chilling instance of this high-minded justification of criminal behaviour: ‘We have the moral right, we had the duty to our people to do it, to kill this people who wanted to kill us.’ Once moral justification is divorced from an emotionally calibrated response to the other, violence can be deployed on rationalised grounds. This has happened time and again throughout history.

But ‘ordinary men’ must cross a line into that zone where the Syndrome E symptoms operate – pushed by circumstance. A noteworthy insight into what happens during the crossing is provided by the neuroscientist Patrick Haggard at University College London. He has shown how powerful is the initial coercion that allows us to step beyond the line. In the wake of the 1961 trial in Jerusalem of Adolf Eichmann, who invoked the ‘Nuremberg Defence’ that he was ‘just obeying orders’ – so-called because it was first used by the Nazi defendants in the Nuremberg Trials of 1945-46 – the psychologist Stanley Milgram at Yale University showed, or rather exaggeratedly claimed, that most people will obey orders from an authority even if the order is to harm another person. Milgram was interested in obedience. Haggard, who has been studying the sense of agency – the sense that we initiate and own our actions, which is central to our lives, and also to legal arguments about criminal accountability – asked instead what it feels like to be coerced and have one’s autonomy removed to some degree. Through an experiment that partly takes its cue from Milgram’s (but addresses some of its ethical and methodological issues) and uses the intentional binding effect, Haggard found that people do feel a notable reduction in their sense of agency when they are coerced into an action. Coercion switches off the sense of responsibility – a chilling finding.

The neurological correlates of what can lead to our worst actions do not indicate a clinical condition. Syndrome E is not a disease, nor quite a disorder eligible for integration into the Diagnostic and Statistical Manual of Mental Disorders or the International Statistical Classification of Diseases and Related Problems. If it were officialised as such, it would have intricate juridical ramifications: the use of neurological evidence in court is problematic, as the jurist Jean-Paul Costa, a former president of the European Court of Human Rights, has pointed out, because it requires the expert reading of imprecise and opaque data. It is hard to establish exactly which brain events – including those underlying the sense of agency – could or should constitute legally mitigating factors.

But introducing, as Fried has done, a set of features that characterise our most beastly nature, and kickstarting a wide-ranging discussion across the fields relevant to their study, particularly in the area of neuroscience, can only help to enrich programmes of prevention and remediation at a time when these are sorely needed. The devil might be dead, but evil actions will always exist. The ‘Why?’ remains a metaphysical puzzle, and I am one of the millions whose life is lived under this question mark, passed on by my survivor father. But at least some answers to the ‘How?’ are now within our reach.