Who Wants War? The Science of Evil

These cute little animals are prairie voles. Don’t they look happy, cuddled up to each other, sharing a warm nest and some yummy food?

Now here’s a montane vole. This is a separate species, but closely related to the prairie vole. Cousins, you might say!

Sort of hard to tell them apart, but they have very different behaviours.

Whereas prairie voles live in colonies, and stick with one mating partner for life, montane voles live alone and will mate with any receptive female. They’re promiscuous, in contrast to monogamous prairie voles.

Parenting behaviour differs also, with both prairie vole parents involved in raising their pups, while only the mother is involved with montane vole offspring. Now, clearly, a prairie vole father cannot suckle his infants, but he guards the nest and huddles over newborn pups to keep them warm and safe.

When a prairie vole pair is separated, each individual mopes, and their stress hormone levels increase. In contrast, montane voles prefer solitude. Once a male prairie vole is bonded to a female, the male will attack intruders into the nest, including other female prairie voles.

From a biological point of view, what is behind these huge differences in behaviour between these closely related vole species? Two hormones: oxytocin and vasopressin.

This is a model of the oxytocin molecule. It’s a nonapeptide, which means it’s a string consisting of only nine amino acids. Amino acids, you may recall, are the building blocks of proteins. So oxytocin is like a very small protein.

You’ve probably heard of oxytocin being used to stimulate contractions in women giving birth. But it is important in many other aspects of reproduction, including lactation and sexual activity. Some people call it the love hormone, others the “tend-and-befriend” hormone.

It turns out that those clingy prairie voles have a high density of oxytocin receptors in a part of their brains called the nucleus accumbens, compared to a low density in brains of those solitary montane voles.

Vasopressin is also a nonapeptide. Almost the same as oxytocin, but two of its nine amino acids are different. From an evolutionary point of view, the family of nonapeptides including vasopressin and oxytocin (called the vasotocin family) is extraordinarily ancient, found in organisms living at least 500 million years ago.

Another name for vasopressin is ADH, which stands for “AntiDiuretic Hormone”, so-called because it stimulates the kidneys to reabsorb water which would otherwise be excreted as urine. It has medical uses, including treatment of bedwetting in children. A synthetic analogue, desmopressin, is taken as a nasal spray.

Getting back to our voles, it turns out that male prairie voles have a very high density of vasopressin receptors in a part of the brain called the ventral pallidum. In contrast, montane voles have higher receptor densities in the lateral septum part of the brain.

So, how do we know that these differences in receptor densities are responsible for the behaviour differences between these two vole species? By giving extra oxytocin or vasopressin to animals and monitoring behaviour; by giving substances which block the receptors; or by modifying the genes to produce animals lacking in receptors or having more receptors. All four techniques have been applied to study vole behaviour. So, blocking oxytocin receptors suppresses pair-bonding, while giving oxytocin increases pair-bonding. In male prairie voles, increasing vasopressin receptor density increases attachment behaviours like grooming and cuddling. Male montane voles with increased vasopressin receptor density behave more like prairie voles.

But it’s not just voles! It turns out that the vasotocin family plays a role in multiple social species: mammals including primates and dogs; gregarious birds, lizards, and fishes, as well as some social insects such as ants and wasps. Interestingly, though, not honey bees!

As Patricia Churchland, the Canadian-American neurophilosopher, concludes, “In highly social mammals, oxytocin is released in the brain in positive social situations, such as grooming, cuddling, sex, and food sharing. At least in highly social animals, this release, supported by the cannabinoids, tends to intensify social attachment. It results in reduced vigilance and anxiety, and an increased sense of trust and well-being.”¹

No wonder, then, that the vasotocin family of hormones is associated with empathy. However, empathy is such a vague concept. So hard to even pin down the meaning of the word. However, here is a useful working definition, quoted from Wikipedia: “Empathy is the capacity to understand or feel what another person is experiencing from within their frame of reference, that is, the capacity to place oneself in another’s position”.

I described above how desmopressin can be administered by the intranasal route. And most of the experiments in which oxytocin is given to human subjects also used intranasal administration. Another short peptide, insulin, can also be given nasally. It appears that these relatively small molecules are able to bypass the blood-brain barrier by migrating through the skull into the brain, courtesy of the olfactory system.

No one questions the importance of what’s called chemosensory communication in nonhuman species. All of the important functions that organisms carry out, whether it’s plants, animals, bacteria, or others, are based on being able to recognize that what you’re dealing with is something to eat, or something that may eat you, or something to reproduce with. And this is done by recognizing chemicals, usually proteins, by their shapes, either through touch, taste, or smell. In animals, mostly smell. And none of this requires any conscious awareness on the part of the animal doing what its nose tells it to.

In humans, however, scientists have taken the unjustifiable attitude that smell is not very important. They base this on the finding that when you ask a person if they can smell something, they seem to have a much lower sensitivity than, say, a dog. But the scientists don’t ask the dog; they just monitor the dog’s physiologic or behavioural responses to an odour. So, apples and oranges.

There is every reason to believe that olfactory communication works perfectly well in humans, and it does so without requiring conscious awareness. So, there was no reason for evolution to connect olfaction to the centres of conscious awareness.

And the olfactory system can be amazingly sensitive²! Bottom line: it may be that empathy is to some extent an olfactory phenomenon. In other words, having an empathic response depends on being close enough to someone so that we can smell them, even if we are not consciously aware of their smell. It’s high time that this was studied by scientists!

Here is a hypothetical example: we know that the fetus secretes oxytocin into its urine. This oxytocin will thus be in high concentrations in the amniotic fluid. When the fetus is delivered, a whopping dose of oxytocin is thus available to the mother (and to the father, if present!) which may help pair-bonding between the mother and child (or father and child).

The oxytocin/vasopressin system is not just about bonding and caring; it’s also about aggressive and violent behaviour. Male prairie voles defend their nests; fish and birds engage in group conflict and aggression. This is thought to be because oxytocin and its analogues amplify concern for the interests of genetically related or culturally similar ‘in-group’ others and increase willingness to defend against outside intruders and enemy conspecifics ³. In human males, the combined administration of testosterone and vasopressin affects aggressive behaviour⁴.

So, what does all this have to do with wars and with evil! I thought you’d never ask!

So we have empathy and aggression, both based on the same, evolutionarily ancient hormone system. And we’ve seen with the example of the two vole species, how genetically determined differences in vasopressin and oxytocin signalling results in huge behavioural differences.

Could this apply in humans also? What if some humans were genetically different from other humans in terms of oxytocin and vasopressin signalling, different enough that they would behave differently? Not a disorder, like a genetic disease; just a genetic difference, like a subspecies within the human species.

My hypothesis is that such a genetic variant exists, and people in this group have lower levels of oxytocin signalling. Therefore, lower levels of empathy. Or, more specifically, lower levels of affective empathy.

But what does that mean? Empathy is frequently subdivided into cognitive empathy and affective (or emotional) empathy.

Affective empathy (again, from Wikipedia) is the ability to respond with an appropriate emotion to another’s mental states. For example, experiencing sympathy and compassion in response to their suffering, or even personal distress. Such empathic feelings can give rise to empathic anger in a situation where someone else is being hurt. That anger may stimulate our desire to help, or to punish.

Cognitive empathy is the ability to understand another’s perspective or mental state. It is generally independent from affective empathy.

It appears that women are more likely to manifest affective empathy than men, consistent with the role that women have in nurturing offspring.

In what follows, I will be referring to individuals with high oxytocin signalling as affective empathy types, and the people with the genetic variant resulting in lower levels of oxytocin signalling, as cognitive empathy types.

Earlier I said that cognitive empathy types have lower levels of affective empathy. However, they may have a greater capacity for cognitive empathy, that is, they are better able to understand, intellectually, how someone else may be feeling, without sharing those feelings.

These differences, I believe, are important in helping us understand conscience. What is conscience? From a neurobiological point of view, Patricia Churchland has devoted a whole book to this topic. Acknowledging that there is no precise definition for the word conscience, she provides a practical, working formulation: conscience is an individual’s judgment about what is morally right or wrong. Unfortunately, that introduces another ambiguous word: morals.

Morals and ethics are two ideas often used interchangeably. But I believe they’re quite distinct. Morality is subjective, unique to the individual, often changes little over a lifetime, and forms a strong basis for how a person behaves. It may also form the core for an individual’s sense of belonging to a community that shares the same moral standards.

Ethics, in contrast, are externally imposed, for example by one’s professional affiliations. Physicians, lawyers, engineers, and so on, are expected to adhere to a code of ethics which may explicitly state what behaviours are permitted or not permitted in particular situations. Ethical standards change over time; for example, attitudes to slavery and whether it is ethically acceptable to own slaves.

Whereas infractions of one’s moral standards may lead to feelings of guilt and shame, infractions of a code of ethics may not, even if the infraction is exposed and punished. A physician who has sex with a patient may feel neither guilt nor shame, even if the consequence is loss of the license to practise medicine. However, it is certainly possible that one’s moral code requires adherence to the code of ethics to which one has promised to adhere, in which case an ethical violation will also be a moral one, with guilt and shame resulting from the breach of one’s moral standards.

Thus one can consider ethics as a societal construct aiming to harness the conscience of individuals in a community to encourage behaviours helpful for the community or society and discourage harmful behaviours.

So, back to conscience. Judgment about what is right or wrong. When we contemplate an action or behaviour to decide whether to do it or not, we have an emotional reaction: either pleasure if it’s something that we would feel good about doing, or guilt or shame if it’s a behaviour that we consider morally wrong. Our emotions guide our actions.

Where do those emotions come from? From our interactions with our earliest caregivers. For example, if we do something that displeases mom, we may sense her disappointment and hurt. And assuming that our affective empathy is intact, we will experience personal distress at mom’s hurt. Making the connection between our having caused that hurt, turns our personal distress into feelings of guilt and shame.

Just to throw a bit of psychiatry into this: we don’t actually respond to the real mom’s hurt, we respond to the feelings of our internalized representation of mom. The internal object is the term used by psychoanalysts.

We carry that internalized representation of mom in our brains all our lives, usually without any conscious awareness since the process began well before we develop the capacity for conscious recall of stored memories, typically around two to four years of age. Nevertheless, even without that conscious awareness, our emotional reaction continues to induce feelings of pleasure, or of guilt and shame. Thus, in a very real sense, our internalized mother is our conscience.

Of course, it is unlikely to be only our mothers. We also internalize other caregivers and authority figures, including ones that never had any reality for us, such as God, or the monster under the bed or in the closet. I was born in the Netherlands, where Saint Nicholas’s sidekick, Zwarte Piet, kept track of who was naughty or nice. And we are often socialized or acculturated to fear other people, such as the police.

The important issue, though, is having intact affective empathy, which depends on high levels of oxytocin signalling. So, what about people with the genetic variant resulting in decreased oxytocin signalling and therefore decreased affective empathy? As you might guess, their consciences may not work as well. Because they experience less, or maybe no, emotional distress themselves when causing someone else hurt or pain.

Do we know people with little or no conscience? Almost certainly. Extreme examples we call psychopaths, or similar terms like sociopath, antisocial personality disorder, conduct disorder.

Here’s the problem. Because the research on people with empathy deficits has focused pretty well exclusively on criminal populations, it really underestimates the number of individuals in this group. When applying the criteria to label someone as a psychopath, using diagnostic criteria developed from studies of people in prison, research suggests that about 1% of the general population, but 20–30% of the prison population are found to have a psychopathic personality⁵. This way of tackling the issue is based on considering low levels of affective empathy and of conscience as pathological, as a disorder, an illness, a disease, something to treat or attempt to fix. Adopting this point of view is easy if you apply a definition that applies mostly to criminals.

But as the example of the montane vole and the prairie vole demonstrate, it is possible to have genetic variants that affect levels of empathy, genetic variants that do not indicate an abnormality or disorder, simply a different species or subspecies. So, what would happen if we applied the same approach to humans? What would be the result if we stopped looking at low levels of affective empathy and of conscience as pathological?

One result would be that the proportion of the general population making up the cognitive empathy types subspecies could be much higher than 1%. 20%? 30%? No one knows, because no one, it seems, has looked at the issue in this way before.

There is also something else to keep in mind. Montane voles and prairie voles are two separate species out of the approximately 155 vole species, and therefore do not ordinarily interbreed, and the high vs low oxytocin signalling dichotomy remains. Humans are different. So there are likely to be quite a few people who have gradations of oxytocin signalling, somewhere between the cognitive empathy type and the affective empathy type extremes.

Now, neither cognitive empathy types or affective empathy types, or the people in between, tend to believe that there is a large group of individuals who think and behave differently from them. That’s a consequence of scientists, including sociologists and anthropologists, focusing on the criminal population and labelling people with low empathy as abnormal, diseased, disordered, or pathological. That’s why we use terms like sociopathy, psychopathy, or antisocial personality disorder.

If you’re an affective empathy type, your tendency is to “default to truth” as described so well by Malcolm Gladwell in his book “Talking to Strangers”. I highly recommend the audiobook version, which includes actual interviews. Here is a quote from the Wikipedia article about the book: “‘Default to truth’ is used throughout the book to observe how human beings are by nature trusting, not only of people or technology, but of everything. Sometimes this kind of behavior, the lack of understanding each other, leads to disastrous and tragic outcomes, as elaborated by Gladwell in the stories he brings. Gladwell notes how there are evolutionary social reasons why we trust more than suspect – the need for cooperation being one.”

On the other hand, if you’re a cognitive empathy type, your mantra might be “Everyone’s in it for themselves.” This mindset would lead you to consistently try to get ahead of the pack, with little concern about hurting others. Decisions to follow norms of social behaviour, such as codes of ethics or laws, may be predicated on the calculated risk of getting caught and/or punished for not doing so.

Have a look at this video clip from my dashcam. Notice the grey Porsche that passes me on the right. Its driver goes right through the red light. And then the driver again goes through the next red light! My daughter always says, “Oh, it’s a pregnant woman in labour and they’re going to the hospital”. But then, he stops at the next red light and waits for the green. So, it’s not an emergency. Just someone who feels entitled. Now, I wonder, driving an expensive, flashy car like a Porsche; is that part of that cognitive empathy type mindset where status and influence are important drivers of behaviour?

Many of you will have seen this video, taken on board a Sunwing flight from Montreal to Mexico on the day before New Year’s Eve, 2021. Passengers are ignoring public health measures, not wearing masks, jumping and dancing in the aisle, vaping and openly passing around a bottle of hard liquor. The flight had been chartered by party organizer for a Quebec group of social media influencers and reality show stars. Entitled much?

I invite you to consider the people you hear about in the news. Heads of big business, politicians, social media influencers, religious leaders. Do they often behave in ways which suggest that hurting others is not high on their list of concerns? cognitive empathy types?

Consider also rich and powerful people of centuries gone by: slave owners, colonizers, kings, popes, crusaders, explorers. We may be ready to tear down their works, monuments and statues now, but in their days these individuals were lauded and even revered. Do they fit the cognitive empathy type mold?

Now, many of us would like to have power, wealth, and fame. But many of us, the affective empathy types, are unwilling to hurt others to get that power, wealth, and fame. Affective empathy and conscience get in the way.

For cognitive empathy types with less affective empathy and conscience, the path to riches and influence is less fraught. And the easiest way to get rich is to have other people generate wealth for you. Now, how do cognitive empathy types convince affective empathy types to do that, to work hard, to pay taxes or tithes, to give cognitive empathy types a portion of what they grow or produce? If as a cognitive empathy type you can charm affective empathy types into doing so, great! If not, violence or the threat of violence is extremely effective. Seduction, or intimidation. Or both.

Seduction or intimidation are also effective to get power over others, whether as a politician in a democracy, a dictator in an authoritarian regime, or an empire-builder waging war.

I said earlier that people with lower levels of oxytocin signalling, the cognitive empathy types, may have a greater capacity for cognitive empathy, that is, they are better able to understand, intellectually, how someone else may be feeling, without sharing those feelings. It is possible to develop this capacity, to master what is called “emotional contagion”, to become skilful at “infecting” others with one’s enthusiasm, energy, and creative ideas, but also with one’s paranoia, hatred, and lusts.

Adolf Hitler was very skilled at creating emotional contagion on a large scale, with mass rallies where people yelled “Heil Hitler!” in unison. Donald Trump also holds large rallies where his supporters chant “Make America great again”, or “Lock her up!”, or “Build that wall!”. These are seductions, writ large. Huge!, as Trump would say.

At the same time, low levels of affective empathy make these individuals resistant to being infected by emotional contagion. It’s like they’ve been vaccinated! But others with higher levels of affective empathy are particularly susceptible to this kind of seduction.

And here’s where the evil creeps in. Prairie voles also manifest high levels of vasopressin signalling in certain parts of the brain, particularly in males. This makes them behave aggressively towards intruders. They protect the nest, just like social wasps and other social species do. Human affective empathy types similarly want to protect their mates, their families, their loved ones, their communities.

Some cognitive empathy types learn just how to exploit this. When building an empire, they convince the affective empathy types to go to war against other nations, or to enslave others, to commit genocide with smallpox-infested blankets, deliberate starvation, or bombing civilians; they destroy the livelihoods of others by putting fences around the village commons, flooding hunting grounds to build hydroelectric dams, or burning jungles to raise cattle; and they seek to destroy cultures with residential schools or repressive language laws. And don’t forget eugenics and concentration camps!

Evil cognitive empathy types convince affective empathy types to participate in these activities by labeling the targets as subhuman and as threats. The affective empathy types’ natural inclination to not want to hurt others is easily overridden by convincing them that those others are not human, not kin, and are dangerous to their loved ones. Whether the others be Jews, Muslims, blacks, LGBTQ, indigenous, immigrants, Romas, the intellectually challenged: it doesn’t matter. The process is much the same.

Here is Putin attacking Russians who fled the country after his invasion of Ukraine: “Any people, and particularly the Russian people, are able to distinguish true patriots from scum and traitors, and simply spit them out like a fly that flew into their mouths.”⁶

So, Who Wants War? Not the affective empathy types, unless they can be convinced that there is an enemy, that the enemy is different from them (as in subhuman), and that the enemy poses a threat. And I can’t imagine that just being a cognitive empathy type and possessing the skills to seduce and intimidate affective empathy types, will lead you to want to wage wars! After all, there have to be easier and less dangerous ways to get power and riches!

I suspect that there is something else. Perhaps the evil cognitive empathy types are also responding to an internalized object, but an internalized object that they fear and hate, and possibly want to destroy. What if some cognitive empathy types, the ones who truly behave in evil ways, were themselves bullied, abused, or tortured as children? It’s certainly possible and perhaps even likely; one or both parents would also have to be cognitive empathy types in order to transmit that genetic variant to their offspring.

What do we know of the early childhoods of people who later in life would wage war in their attempt to gain power and riches and fame? People love to compile lists, and here is a list of the ten most evil people in the history of the world. I will leave it as an exercise for you to look up their childhood histories, perhaps on Wikipedia. Hint: both Hitler and Stalin were beaten by their fathers.

I don’t want to give the impression that evil cognitive empathy types are all rich and infamous. What about those people who are dishonest with their spouses, who threaten their tenants with renovictions, who harass their employees, who sell shoddy goods at high prices, who construct high-rise apartments using dangerously flammable materials? There are doctors who accept gifts from pharmaceutical companies, tech giants using old-school union-busting tactics, politicians doing their utmost to prevent certain groups from being able to vote. Most of these behaviours are supposedly being regulated by government agencies, but those agencies seem unwilling to interfere, a phenomenon called “regulatory capture”.

There are good reasons to believe that the proportion of the population who are cognitive empathy types, including evil cognitive empathy types, is increasing. One reason is that with increases in productivity, those people who actually generate wealth through their labours are able to support more “rentiers”. Another reason is that the capacity to wreak violence has changed over the centuries due to technological advances. This is well described by James Dale Davidson and Lord William Rees-Mogg in their books⁷; the first of these technological advances was the development of metal weapons, which only the wealthy could afford, and allowed them to enslave people without means. Then, around 500 AD, the invention of the stirrup. “Stirrups enabled heavily armed knights to keep their balance while fighting at full gallop. The result was a tremendous increase in the effectiveness of armored cavalry—whose charges could no longer be resisted by infantry”⁸. However, heavily armed knights and their horses were expensive, and the authors believe that this led to a devolution of central authority and a rise of local centres of power—ie, feudalism.

Around a thousand years later, the use of gunpowder became widespread. In 1494 siege cannon were able to reduce the walls of a fortress to rubble in only 8 hours; a few years earlier, that same fortress had been able to withstand a siege of 7 years! The gunpowder revolution gave powerful advantages to the offense while making defenses more vulnerable. This raised the scale of warfare and led to larger political entities, thus introducing the modern era, and eventually the Industrial Revolution.

The industrial revolution made possible the large-scale production of inexpensive, reliable handguns. This again changed the balance of power, possibly contributing to the end of slavery, and even to the enfranchisement of women!

In the last century, the development of nuclear arms again swung the balance of power towards megapolitical entities. And more recently we have witnessed the development of intelligent weapons such as drones which can accurately drop tank-busting munitions or remote-guided attacks on heads of state.

Whatever the technology, the issue remains the same: whoever controls violence gets to be in charge. While evil cognitive empathy types have always known this, ordinarily peace-loving affective empathy types mistakenly believe that the rule of law will protect them, and are caught by surprise when it does not.

Of course, wars are only the most extreme example of cognitive empathy types taking advantage of affective empathy types. Many governments, including democratically elected ones, have figured out how to transfer wealth from those that create it to those that covet it. Some of these ways are highly creative. Welfare systems, for example, give taxpayer money to the indigent, who flow through all of these funds to merchants and landlords. The COVID-19 pandemic increased the welfare scale enormously, what with vaccines, tests, expensive treatments, bail-out funding for employers and income supports for employees, all paid for by taxpayers, and all flowing through to large corporations, enriching their investors and executives. During the pandemic, the wealth of the world’s 10 richest billionaires doubled. At the same time, superwealthy Americans pay almost no income tax! I’ll bet that these people are doing better financially by being law-abiding citizens (following laws which they helped design) than the world’s worst dictators who seized power by force!

The second issue is: how can we identify cognitive empathy types, so as to avoid voting for them, or marrying them, or fighting wars for them, or further enriching them by buying their products or services? Science has learned to distinguish montane voles from prairie voles using a variety of indicators, and we can learn to do the same with humans. I’m working on this, using Artificial Intelligence (AI). So stay tuned!

1. Churchland P. Conscience: The Origins of Moral Intuition. W.W. Norton & Company; 2019:272. ↩
2. “The moth with the most developed sense of smell, however, is that of…the Indian luna moth (Actias selene). The male of this species is so sensitive to the female’s sex pheromone that he can trace a female via her scent from as far away as 6 1/2 miles (11 km). In experiments in which male specimens were released this distance away from caged females, 26 percent successfully located the females, while 46 percent of the males located the females if released 2 1/2 miles (4.1 km) from them.” Shuker KPN. The Hidden Powers of Animals: Uncovering the Secrets of Nature. Marshall Editions; 2001:240. ↩
3. Triki Z, Daughters K, De Dreu CKW. Oxytocin has ‘tend-and-defend’ functionality in group conflict across social vertebrates. Philos Trans R Soc Lond B Biol Sci. 2022;377:20210137. PMID 35369742 ↩
4. Akkoc Altinok DC, Votinov M, Henzelmann F et al. A Combined Administration of Testosterone and Arginine Vasopressin Affects Aggressive Behavior in Males. Brain Sci. 2021;11:1623. PMID 34942928 ↩
5. van Dongen JDM. The Empathic Brain of Psychopaths: From Social Science to Neuroscience in Empathy. Front Psychol. 2020;11:695. PMID 32477201 ↩
6. Shaun Walker, “‘I feel so much freer’ The gifted young Russians making a new life in Armenia”, The Guardian, 2022-5-13, p19 ↩
7. Davidson JD, Rees-Mogg W. The Sovereign Individual: How to Survive and Thrive During the Collapse of the Welfare State. Simon and Schuster; 1997:416.
   Davidson JD, Rees-Mogg W. The Great Reckoning: Protecting Yourself in the Coming Depression. Simon and Schuster; 1994:608. ↩
8. p69, Davidson JD, Rees-Mogg W. The Great Reckoning: Protecting Yourself in the Coming Depression. Simon and Schuster; 1994:608. ↩

• ← Why do people attempt suicide?
• Our Tandem Bicycle →