Tuesday, July 12, 2011

The Science of Evil

Psychologist Simon Baron-Cohen’s The Science of Evil: On Empathy and the Origins of Cruelty is a fascinating, clear, short excursion into the current science of empathy, autism and their implications. (And if the name seems vaguely familiar, yes, he is Borat’s cousin.)

The book starts with a discussion of human evil and cruelty, using some salient examples, suggesting that while ‘evil’ as a concept has little explanatory value, and is not amenable to scientific study, empathy has both. That human evil and cruelty are the result of a lack, or erosion of, empathy. Empathy being defined as:
our ability to identify what someone else is thinking or feeling and to respond to their thoughts and feelings with an appropriate emotion (p.16).
Using the word ‘appropriate’ in a definition is troublesome unless you “cash it out” later, but we get the point. If someone lacks empathy entirely, then they are “trapped in their single focus” on themselves (p.18): such people being the subject of the book.

About empathy
Before we get there, Baron-Cohen covers how empathy can be measured, and how people (permanently or temporarily) vary in empathy according to, as is so common in human traits, a Bell-curve distribution (Pp19ff). Even though empathy is a continuous dimension, as a useful heuristic, he divides people into various levels of empathy from level 0 (none) to level 6 (very high levels) according to their EQ (Emotional Quotient), a scale he and his fellow researchers devised based on questionnaires (Pp23ff).

Baron-Cohen notes the problems of self-reporting but argues these “occasional inaccuracies” are cancelled out in sufficiently large samples (p.22). His discussion covers current findings in neuroscience about relevant areas of the brain to empathy (Pp27ff). One of the things that impresses me about the book is that Baron-Cohen is careful not to over-claim, either in his discussion of the neuroscience or his later discussion of genetic connections. He does not fall foul of this critique of connecting neuroscience to psychology: his point is the more basic one of establishing that there is a biological basis for the psychology he is outlining.

When it is all about them
We then move to discussing Zero-Negative empathy; that is, people with zero empathy who behave in ways dangerous to others. Throughout the book, he uses revealing examples: in this section, of people with Borderline, Psychopathic and Narcissistic personality disorders (Type B, Type P and Type N). A weakness in this use of examples is that Baron-Cohen naturally uses particularly intense examples: one needs to pay careful attention to the discussion of general characteristics and not take his examples as templates.

It is nasty to be around a borderline, but it is not good to be one either: they are about 2% of the population but about 15% of those who turn up for counselling or psychiatric help, about 33% of those who commit suicide, and may be as many of half of those seeking clinical help for eating disorders, alcoholism and/or eating disorders; with a high propensity to commit, attempt or threaten suicide (p.55). Borderlines:
rage at those they love. … Despite all this rage, they describe themselves as “empty” inside (p.56).
An emptiness they attempt to fill with impulsive behaviour and which leaves them with a lack of core identity. Marilyn Monroe was a borderline personality (Pp57ff).

So, what causes it? There is a great deal of evidence that childhood deprivation affects brain development and helps generate personality disorders. The causal connection is complicated: 80% of those who suffer childhood sexual abuse do not grow up to be borderline personalities, but 40-70% of borderline personalities report childhood sexual abuse, 60-80% report a history of physical abuse, early separation through divorce, emotional neglect, indifference, deprivation and rejection (p.62). So childhood abuse and neglect creates a strong propensity to being borderline, but is not of itself determinative. Borderline brains show distinctive patterns that include abnormalities in the brain’s “empathy circuit” (Pp62-4).

Next are the psychopaths, or Type P. These are the classic “evil individuals”. They also lack empathy, but are willing to do whatever will satisfy their desires (p.64). About 3 percent of males and 1 percent of females have antisocial personality disorder (which includes psychopaths), but about half of all male prison inmates and a quarter of female prison inmates have antisocial personality disorder (p.67). The classic identified symptoms of psychopaths are:
superficial charm
lack of anxiety or guilt
undependability and dishonesty
inability to form lasting intimate relationships
failure to learn from punishment
poverty of emotions
lack of insight into the impact of their behaviour
failure to plan ahead (p.68).
Economics predicts that criminals will tend to be people with short time horizons, which fits. Baron-Cohen is more specifically interested in the lack of empathy involved in so many of the above symptoms. Part of the danger of the psychopaths is that they may not be physically aggressive, but can be more subtly so. Baron-Cohen accepts that the cliché “snakes in suits” is an apt description of how they can be camouflaged.

Psychopaths respond to parental rejection with rage. Blocked from being expressed towards their parents, the rage builds up, to be vented later in life, unchecked by any empathy (Pp69ff).

Baron-Cohen briefly sets out child psychiatrist John Bowlby’s attachment theory, which generates predictions that are supported by considerable evidence. The theory holds that an infant uses the primary care-giver as a “secure base” from which to explore the world:
By giving praise, reassurance, and a feeling of safety, the caregiver’s affections helps the child manage his or her anxiety, develop self-confidence, and trust in the security of their relationship (p.71).
Baron-Cohen paraphrases this as successful parenting gives a child an “internal pot of gold” that fills the child up with positive emotions. Psychopaths typically have a higher rate of what Bowlby (who developed his theory from work with juvenile delinquents) called “insecure attachment” (Pp 71ff).

Psychologist Jeffrey Gray developed the theory that psychopaths have an under-active “Behavioural Inhibition System” (BIS), the structure that allows learning about the emotional consequences of actions, while anxious people have an over-active BIS. Hence psychopaths do not learn to fear punishment: with clear consequences. Psychopaths also show brain abnormalities in their “empathy circuit” (Pp79ff).

Then it is on to narcissists (Type N), the least studied of the three main types of personality disorders. They lack empathy or humility, think themself much better than other people, typically tending to monologues rather than conversations; though narcissists can vary greatly in their mode of social interaction from outgoing to shy and withdraw. A massive sense of entitlement is a common feature, however. It is speculated that excessive parental admiration, praise or indulgence is a factor in the creation of narcissists. They are about 1 percent of the population but about 16% of those attending clinics for mental health issues while a half to three-quarters are male (Pp88ff).

Baron-Cohen discusses how empathy can be temporarily or permanently depressed (Pp 93-4). This is of particular interest to me, because I have come to the conclusion that bigotry works by exempting people from being objects of empathy. Putting them outside the ambit of empathy.

Clearly, emptiness, rage and entitlement can occur in response to seriously unbalanced parenting and yet not create borderlines, psychopaths or narcissists. The essence of Baron-Cohen’s theory is that it is those things plus zero-empathy which makes Type B, P or N personality disorders.

All about patterns
Then it is on to an even more enlightening discussion of how zero empathy can be positive, in the case of people in the autism spectrum (particularly Aspergers). For empathy is not the sole route to developing a moral code (p.95). These people can be Zero-Positive for two reasons:
First, in their case their empathy difficulties are associated with having a brain that processes information in a way that can lead to talent. Second, their the way their brain processes information paradoxically leads them to be supermoral rather than immoral (p.96).
People in the autism spectrum show underactivity in the empathy circuit (Pp100ff).

What distinguishes these people from the Zero-Negatives is that they systemise to a remarkable degree. The social world lacks rules they can grasp, but the physical world is full of them:
People with Asperger Syndrome have brains that are exquisitely tuned to notice patterns (p.104).
Patterns are basic to acting, to predicting, to making things work, to find new ways to make things work. By identifying patterns we can see what is the case and what can be made to be the case. In a sense, seeing patterns allows us to step outside time because we can identify enduring realities. We can systemise by observing patterns, or by observing and operating on them: we can then create new ways of doing things (Pp105ff).

Calling those parts of the brain that can perceive patterns in changing information the “Systemising Mechanism”, Baron-Cohen points out that these are also distributed in a Bell curve, from people with very low systemising ability to those with very high systemising ability. The problem for the hyper-systemisers, is that all they see are patterns, so any unexpected change is toxic. They seek complete control so everything is completely ordered, with any disruption threatening collapse. The upside is that they see patterns no one else has discovered. Such people live in a black-and-white world of things being true or false, never shades of grey. For them truth is precision. And the emotional world does not “fit”. Their systemising mechanism is set so high, there is no place for empathy. But there can be great intellectual and other creativity, as they spot and use patterns no one has previously noticed (Pp112ff).

Classic autism, in Baron-Cohen’s analysis, is this to the max. Autistics have no sense of empathy at all, no sense of other people except as objects. But everything needs to be systematic, patterned or else it becomes toxic change they cannot deal with, that they have no place to “put” it. But without the obsessive systemisers, Baron-Cohen argues, humans would have achieved far less (Pp118ff).

What makes the hyper-systemisers' zero-empathy profoundly different from the Zero-Negative borderlines, antisocials (including psychopaths) and narcissists is that the hyper-systemisers come to morality, not through empathy, but through a very strong sense that behaviour should be governed by rules. They can, indeed, be super-moral because they are so strongly tuned to fairness as a basic principle. So, they are Zero-Positive both in the sense of being often highly creative and in being moral, even hyper-moral, without empathy (Pp 121ff).

Wrestling with cause
All of which is fascinating and enlightening, but still leads to the question of causes. Yes, parental abuse and neglect or otherwise seriously unbalanced upbringing can lead to destruction of empathy coupled with emptiness, rage or entitlement but most people with abusive or otherwise seriously unbalanced upbringings do not develop into borderlines, psychopaths or narcissists. Conversely, some people develop in to Zero-Negatives without suffering such neglect or abuse. If environment is not sufficient explanation, then interaction with genetics (and Baron-Cohen stresses interaction and that genes are just mechanisms for coding proteins) has to be considered (Pp125ff).

But any genetic tendencies to low empathy that manifests in the Zero-Negatives are going to be different genes from the Zero-Positives, given that the latter go down the empathy distribution as they go up the systemising distribution (p.127).

Which brings us, inevitably, to twin studies and which traits are common in identical twins and different in non-identical twins. Baron-Cohen discusses (in a very accessible way) the evidence on genes for aggression, emotional recognition and genes associated with EQ and autistic traits. Baron-Cohen then reminds us that causes are very mixed: genes are not determinative (Pp128ff).

If empathy is in part genetic, there should be evidence for empathy in animals. Which there is, some of it unexpected (in rats, for example)(Pp143ff). Nevertheless:
whatever glimmerings of empathy we can discern (or imagine we discern) in other species, the level of empathy that humans show is qualitatively different from that seen in any other species (p.145).
As are our systemising abilities.

The final chapter is Reflections on Human Cruelty, where Baron-Cohen returns to making what people call evil amenable to scientific analysis. (He carefully makes it clear he is not buying into any Dawksian anti-religion agenda.) He summarises the ten new ideas he hopes to introduce into the debate:
we all lie somewhere on the empathy spectrum;
at one end is zero empathy;
the empathy circuit in the brain will be abnormal in such cases;
treatment of zero empathy should target the empathy circuit;
John Bowlby’s notion of early secure attachment can be usefully characterised as an internal pot of gold able to be drawn upon;
there are genes for empathy;
while most forms of zero empathy are negative, one is positive;
Zero-Positive comes from a mind striving to “step out of time” to see the repeating patterns in nature;
the Zero-Positive mind finds change toxic;
empathy is the most valuable resource (Pp147ff).
I was particularly struck by the comment that:
Calling the brain types personality disorders leads to debates about whether personality can be changed, especially if personality is defined as a fixed set of traits. Calling them Zero-Negative opens up new avenues for intervention (p.150).
Some of his tentative suggestions for treatment echo what Albert J. Bernstein in Emotional Vampires suggests for narcissists.

Baron-Cohen identifies various outstanding puzzles and suggests psychiatry needs to give much more attention to empathy. He considers the alleged “banality of evil”, what an underactive empathy circuit means and the potential for change. Finally, he considers the possibility of super-empathy and empathy as an under-utilised resource (Pp153ff). The book includes an Appendix incorporating questionnaires for measuring EQ in adults and in children (Pp187ff) and a second Appendix listing characteristics of Type B, P and N Zero-Negatives (Pp197ff), a very useful summary.

I found The Science of Evil a very enlightening book. I would particularly recommend it for anyone who is on the autism spectrum or has to deal with someone on the autism spectrum.


  1. Two disclosures before I start on this: I have Asperger's Syndrome, and am therefore autistic, and I have not read Baron-Cohen's book, and only know what other people have written about it.

    Baron-Cohen is a controversial figure in the Autistic world: for someone who has spent so much time studying us, it is disturbing to see his theories rooted so strongly in what we see to be fundamental misunderstandings of our experience.

    For a start, there is the EQ measure, and the systematising/empathising ratio. I score highly on both these measures, but even then I can see questions featuring false oppositions: Friendships and relationships are just too difficult, so I tend not to bother with them. The question is wrong: what if maintaining friendships is hard, but I see the value in persevering anyway? There are many who see that as one of the characteristics of autism: seeing what was said, sometimes at the expense of what was meant.

    Anyway, one of his theories on autism is that it is a "hypermale" brain. Which implies by extension that empathy and feelings and sociability and fuzziness are the characteristics of a female brain, and also that mechanical interests are unfeminine, but leave that aside. For a start, his theories at best bemuse female autists, and at worst lead to systematic underdiagnosis of female autists, as it does tend to present differently in girls (whether this is a biological effect or sociological is another question).

    For a second point, there are aspects of empathy which I suspect are being glossed over, either by Baron-Cohen himself, or by his readers. Empathy can be broken into three aspects:
    1. Cognitive Empathy, or knowing when empathy is appropriate, being aware of someone else's emotional state. This one is definitely one which autists do badly at. It can be learned, but the process is slow, ad hoc, idiosyncratic, and prone to catastrophic failure. Also associated with this is the condition called Alexithymia, which is where we are not cognitively aware of our own emotional state.
    2. Demonstrative Empathy: being able to make the appropriate response. Smiling when someone's happy, crying when they're sad, being able to give a shoulder to cry on, say the right thing, that sort of thing. There is, of course, a huge variation between folk, but here, again, autists cluster at the non-functional end of ability. We just don't feel the need to shout and cheer at a football game, or crowd around and backslap someone who has won; it's all a bit bewildering and slightly threatening. It doesn't make sense, it doesn't feel natural.


  2. ... cont

    3. Affective Empathy: actually experiencing empathic reactions, feeling someone else's pain, being happy when someone else is happy, sad when they're sad. And here's where Baron-Cohen, I think, gets it 180° wrong: we seem to experience these things more strongly than most. Injustice and unfairness are not just intellectually to be avoided, they cause us somatic discomfort. (How "injustice" and "unfairness" are interpreted is another thing: some may think of the rules as the arbiter of fair, others may see the rules as social constructs, and spend a lot of time thinking about personal codes of ethics.) There are a lot of instances where the NT world displays far less empathy than the autistic world does: the very existence of {nation's} Funniest Home Videos is an example: I find it difficult to watch. The various falls and groinshots cause me the echo of actual somatic pain. It hurts me to watch it. Similarly with any number of sitcoms based on the comedy of cruelty and humiliation.

    The break comes when we are unable to tell that someone else is hurt. If we know that someone is upset, we are upset with them. But if we don't know, then how can we read their mind?

    When Baron-Cohen says "[Classic, nonverbal] Autistics have no sense of empathy at all, no sense of other people except as objects", I submit he has exactly the wrong idea of what's going on. They feel intensely, but they lack the capability of extracting enough meaning from people to have anything to empathise with. They are surrounded by aliens, whom they can't decipher. No wonder then that they can't display empathy as an NT would expect. We on the high-functioning end aren't that bad, but we still lack an instinctive skill for interpreting other people, and have to do it laboriously and consciously.

    As far as Psychopathy goes, I submit that it is the opposite of autism. Where autists have high affective empathy, and low cognitive and demonstrative, psychopaths and their kin are the opposite: they have finely tuned skills in reading other people and reacting appropriately when required -- they are, after all, famously charming when they want to be -- but they simply don't care. They don't have the affective empathy to be able to make an emotional connection.

    There is, as it happens, a blog which started recently about the experience of autistic people with empathy, basically fighting Baron-Cohen's and Autism Now's assertions that we don't have any.
    Autism and Empathy.

    Baron-Cohen is increasingly seen by autists as someone who won't let mere actual autistic experience get in the way of his beautiful theories.

  3. For what it's worth, the men in my family with Asperger's are manipulative, exploitative and have a profound sense of entitlement. My brother is the father of twins with Asperger's, and I suspect these common traits have more to do with the kids being brought up by an utter prick than being the result of their Asperger's, however all three use the diagnosis to justify their appalling behaviour.

    My father was never diagnosed but ticked all the boxes. His sense of entitlement was focused on his utter (and unfounded) faith in gambling, but otherwise he was 'super-moral' as described above.

  4. @p-cat

    Two things I like to say about this sort of situation:

    It's an explanation, not an excuse, and
    Autism and Arseholery are orthogonal.

    Being an Aspie does not make one a saint, nor does it give a get out of jail free card for bad behaviour.

  5. May I suggest a further resources to learn more about empathy and compassion.
    The Center for Building a Culture of Empathy
    The Culture of Empathy website is the largest internet portal for resources and information about the values of empathy and compassion. It contains articles, conferences, definitions, experts, history, interviews,  videos, science and much more about empathy and compassion.

    Also, we invite you to post a link to your article about empathy to our Empathy Center Facebook page.

  6. Catsidhe: your comments are thought-provoking and informative: particularly on the three types of empathy. I am not sure that what you say is all that far from what Baron-Cohen has to say: more a correction than a full rejection.

    The Science of Evil does not broach the "hypermale" issue, so I cannot comment (though I have heard of the theory).