The Anatomy of Violence

by Adrian Raine

  In 1994 I published our findings on 4,269 live male births occurring at the Rigshospitalet in 1959.11 Birth complications were assessed by obstetricians assisted by midwives. Examples of delivery complications included things like forceps extraction, breech delivery, umbilical-cord prolapse, preeclampsia,12 and long birth duration. One year later, social workers went around to all the homes of the mothers and conducted interviews. Had she wanted the pregnancy? Did she ever make an attempt to abort the fetus during pregnancy? Was her child placed in a public institution for any reason for at least four months in the first year of life? These three indicators of maternal rejection of the child were duly noted. When these babies were eighteen years old, we conducted a national search of all court records in Denmark to find out which of the baby boys had been arrested for a violent crime.13 We then classified them into four groups. Those with neither birth complications nor maternal rejection of the child in the first year of life were the normal controls. Some had birth complications, but had not been rejected by their mothers. Some were rejected, but had a normal birth. And the fourth group had the double whammy—birth complications and rejection by their mothers in the first year of life.

  The results were striking. As you can see in the top half of Figure 6.1, the first three groups did not differ significantly from each other, with rates of violence at about 3 percent. It was the fourth biosocial group—the one with both the biological and the social hits—that had the highest rates of violence. This group had three times the average of the other three groups—9 percent of them became violent offenders. Furthermore, although only 4.5 percent of the population had both birth complications and early child rejection, this small group accounted for 18 percent of all violent crimes perpetrated by the entire sample of 4,269—four times higher.14 It’s a classic case of early biological factors interacting with social factors very early in life to shape adult violence.

  A lot of violence is committed after the age of eighteen. Would this biosocial interaction also explain this later violence, or is it especially important in explaining early violence? We reassessed the entire birth cohort at age thirty-four for arrests for violent crimes. This resulted in a tripling of the sample size of violent offenders, allowing us to conduct more detailed analyses.15 The results indicated that the biosocial interaction was specific to violent crime with an early onset. It did not explain violence that started later in life. In addition, we found the interaction to be specific to violent offending—it did not explain nonviolent criminal offending. It seems that a violent birth makes for violent behavior in particular.

  Looking back at the three components of “maternal rejection,” were there any that were particularly important? Two of them were. First, rearing in a public-care institution in the first year of life was critical. Second, an attempt to abort the fetus also came up trumps. These were the two elements of maternal rejection that interacted with birth complications in producing later violence. In contrast, if the mother simply did not want the pregnancy but took no action, it did not seem to affect long-term outcome.16 Furthermore, the interaction was found to be specific to more serious forms of violence like robbery, rape, and murder—but not for less serious forms like threats of violence. It seems, then, that birth complications conspire with more severe forms of maternal rejection to launch particularly violent criminal careers.

  The problem with our Copenhagen study is that the sample is made up almost exclusively of white babies. More than that, they are white babies from a European country with relatively low levels of homicide. Are these findings some peculiarity of an idiosyncratic Danish culture? What about black babies and other nationalities? These questions were first addressed by two American criminologists who tested the “bad birth and bad mother” hypothesis using a cohort of 867 male and female African-American babies who made up the Philadelphia Collaborative Perinatal Project. Birth complications had been collected on this sample at the times of the babies’ births.17 Criminologists Alex Piquero and Steven Tibbetts followed our original design and broke the larger sample down into the same four groups.18 Their results are shown in the lower half of Figure 6.1. They’re visually striking, showing almost identical findings. Yet again, it was found that those with both birth complications and a disadvantaged family environment were much more likely to become adult violent offenders. The first findings, from Denmark, were not a fluke.

  The birth-biosocial interaction is found in Denmark and the United States. What about other countries? So far the interaction appears to be holding up. Pregnancy complications interacted with poor parenting in predicting adult violence in a very large Swedish sample of 7,101 men.19 In a Canadian sample of 849 boys, an interaction was found between increased serious obstetric complications and family adversity in raising the likelihood of violent offending at age seventeen.20 In a Finnish sample, perinatal risk interacted with being an only child21 in raising the odds of adult violent offending by a factor of 4.4 in a sample of 5,587 males.22 Furthermore, interactions between birth complications and negative home environments in predisposing children to antisocial behavior have been found in Hawaii23 and Pittsburgh.24 Almost wherever you go in the world, you find the same effect.25 The combination of birth complications and adverse home environments appears to be a useful biosocial key that can help open the lock on the causes of violence.

  Figure 6.1 Birth complications interacting with negative early home environments predispose to adult violence

  You may nevertheless be asking how exactly birth complications and negative home environments like maternal rejection combine to shape adult violence. If we look at birth complications first, the likely pathway is that they have a negative impact on the brain. Take my birth as an example. I was born at home as a “blue baby” without intensive-care treatment. As an adult I have always been hopeless at finding my way around new places—I have a very poor spatial sense. Some babies are blue at birth because they suffer from a birth complication termed “hypoxia”—a partial lack of oxygen. Our brains need oxygen to metabolize glucose—a fuel that provides energy for brain cells. Without oxygen, brain cells will start to die in a few minutes. Particularly sensitive to this destructive process is the hippocampus, a part of the brain centrally involved in spatial ability as well as short-term memory, capacities that have been found to be impaired in those who are persistent offenders throughout their lives.26 Hypoxia at birth was also found in one study to be the best predictor of a lack of self-control,27 a key behavioral risk factor for crime and especially for explosive, impulsive aggression. As we saw in a previous chapter, the hippocampus is structurally and functionally impaired in violent offenders.28 Other birth complications such as preeclampsia, maternal bleeding, and maternal infection cause a reduction in blood supply to the placenta, resulting in cell loss not just to the hippocampus but also to other brain areas including the frontal cortex. Consequently, birth complications have multiple neural pathways to a violent outcome.

  A more specific pathway by which birth complications can result in behavior problems in children was shown by Jianghong Liu in her analysis of data from the large birth cohort in Mauritius. She demonstrated interconnections between three key processes—birth complications, low IQ, and antisocial behavior. We had assessed prenatal, perinatal, and postnatal birth complications and also measured at age eleven both IQ and externalizing behavior problems—aggression, delinquency, and hyperactivity. Jianghong showed that birth complications were significantly related to increased externalizing behavior problems.29 She also found that birth complications were associated with lower IQ at age eleven.30 Low IQ in turn was associated with externalizing behavior problems. The triangulation of relationships was complete. Low IQ mediated the relationship between birth complications and later behavior problems—birth complications result in lower IQ, and this in turn results in problem behavior in later childhood—more specifically, aggression, antisocial behavior, and hyperactivity. IQ is predicated on a well-functioning brain, and like othe
r neurocognitive measures it acts as a proxy for brain functioning.

  At least five other studies besides Jianghong’s have observed direct links between birth complications and behavior problems, delinquency, and adult violence.31 For example, in Holland two separate studies showed direct relationships between birth complications and externalizing behavior problems in boys and girls.32 These and other studies, however, did not test the biosocial hypothesis, and some studies have not found direct links between birth complications and violence or have obtained only partial support.33 At the same time, even more studies such as the one we conducted in Copenhagen do not find a direct path between birth complications and problem behavior. Instead, social processes are critical, seemingly acting as a trigger for the dormant birth-complications risk factor for violence.

  In our Copenhagen study we found that a critical component of “maternal rejection of the child” was being institutionalized for at least four months in the first year of life. Why was this component of the social risk factor so important in our study? The life of a young English boy born during the Edwardian era, in 1907, offers a poignant insight. John Bowlby was a Londoner who saw his mother for just one hour a day. She thought that a child could be spoiled by too much attention and affection. When he was seven years old Bowlby was packed off to a boarding school, and by his own account had a terrible time there. In his words, “I wouldn’t send a dog away to boarding school at age seven.”34

  This early experience and poor bonding with his mother proved to be pivotal for John Bowlby and was to shape his future career. After graduating in psychology from Cambridge University he worked with delinquent children before training as a psychoanalyst and psychiatrist, going on to pioneer a new approach to attachment theory. His classic book, written at the end of World War II, brings together his own early experiences and his knowledge of delinquent boys and helps explain why maternal rejection was so important in our Copenhagen study.

  Entitled Forty-four Juvenile Thieves, Bowlby’s book was an in-depth analysis of the early home backgrounds of forty-four juveniles who turned out to be offenders.35 In those early days of delinquency research he made the innovative argument that the lack of a continuous and loving relationship between mother and infant resulted in the inability of the infant to develop a normal personality, and the inability to form normal interpersonal relationships. His case studies highlighted in the lives of these forty-four thieves the prolonged separation from their mothers early in life. This resulted in the absence of a warm, continuous, and intimate relationship between the mother and her infant. The result? What he termed “affectionless psychopathy.” Some of his illustrations were graphic and dramatic. In two of the affectionless psychopaths, they had each spent nine months in a hospital without any visits from either parent.

  This social perspective on crime and delinquency was to be fine-tuned a little by other scholars in later years. What transpires is that there is a critical period early in life when being connected with the mother really counts. In humans this starts at about six months and ends after about two years. For this reason breakage of the mother-infant bonding process for at least four months in the first year of life—as experienced by some of our Copenhagen babies—freezes the social-interpersonal development of the infant. That freezing results in the glacial, emotionless psychopath that we see in adulthood. Recall that “Jolly” Jane Toppan was orphaned and institutionalized until the age of five and went on to become a killer nurse—she had exactly this risk factor for psychopathic violence.

  As for Bowlby himself, what became of this little boy with his early maternal deprivation and ruthless parental separation? He was likely spared an outcome of “affectionless psychopathy” because right at the get-go, despite the absence of his mother, he had all the attentions of a caring nanny. As others went on to argue, the decisive issue is whether you have someone to bond with—anyone at all.36 It could be a stand-in mother figure not genetically related to you, like a nanny, or your father, or even an elder sibling who takes on the caregiving role. As long as you have the opportunity to consistently bond with any human early in life, you derive the basis for appropriate social relationships.

  We’ve seen here that, as with Peter Sutcliffe, indicators of later violence can emerge by the time we take our first breath of air. For Peter it was not just birth complications, but also schizophrenia, a genetically based mental illness that we’ll return to later. But in our developmental search for the origins of violence, is the nine months spent in the womb already too late? Scientists are beginning to trace back the origins of potential evil to points not too long after the moment of conception. The anatomy of violence moves on to events that occur before birth, and our genetic-like perspective here is fittingly found in Genesis and a fable about the origins of humankind.

  THE MARKS OF CAIN

  Cain, a son of Adam and Eve, has the dubious distinction of being the world’s first murderer—and the killer of his own brother. Cain’s story is a fitting beginning to the history of homicide. After all, about 20 percent of all homicides take place within the family, and of these about two-thirds can be viewed as reactive aggression37—responding aggressively to an upsetting or provoking external stimulus.

  Cain was one of these cases. He was absolutely furious with God. God had accepted his brother Abel’s sacrificial offering of a sheep—but had rejected Cain’s offering of crops. In a fit of rage, Cain displaced his aggression onto Abel and slew him. As punishment, the story goes, God placed a mark on Cain as a curse, and Cain was destined to become a restless wanderer who would walk the earth, never again able to cultivate crops.38

  The search for a real-life mark of Cain in criminals was a goal of early criminologists, and as we saw earlier, Lombroso, the father of the discipline, was adamant that it could be found. In his thousands of painstaking physical observations of criminals in Italy, Lombroso believed he saw physiological signs of the “born criminal,” and witnessed multiple hallmarks of Cain that he called “atavistic stigmata” and that he fervently believed set criminal offenders apart biologically from the rest of us.

  Do you have the mark of Cain? Take your right hand, lay it palm up, and relax it. Fold the fingers of this hand a little toward you. Can you see one continuous crease that goes all the way across the top of your palm? Or do you see two main creases that do not join together? If you have a single palmar crease, bad luck. According to Lombroso, you have the atavistic stigmata that makes you an evolutionary throwback to lower species.

  Now take off your shoes and socks, stand up, and look down at your feet. Do you see a big gap between the first and second toe? If you do things are not looking good—another strike against you. There are others. If you want to see if you have one that I have, stick your tongue out and look at it in the mirror. Do you see a fissure—a line running down the middle of it? Another mark of Cain.

  It sounds completely ridiculous—yet there is some veracity to Lombroso’s claims. The “stigmata” outlined above are just three of a number of what are now called “minor physical anomalies.” These anomalies have been associated with disorders of pregnancy and are thought to be a marker for fetal neural maldevelopment at about the third or fourth months of pregnancy. For example, during fetal development your ears sit relatively low on your head, but they begin to drift up to their normal positions at about four months of development. If disruption occurs to fetal brain development at this time, there is incomplete embryonic migration of the ear anlage—essentially, the ears will not migrate to their normal position, resulting in low-seated ears.39 These anomalies are viewed as indirect markers of abnormal brain development. If you want to have a quick look in the mirror at yourself, is the point where your ear connects to your head below your eyes? If so take another gulp.

  In case you are getting worried, other minor physical anomalies include adherent (attached) earlobes, electrostatic hair, and curved little fingers. It is believed that they may be caused by environmental teratogeni
c influences acting on the fetus, factors such as anoxia, bleeding, infection—or fetal exposure to alcohol.40 Don’t worry if like me you have only one or two minor physical anomalies—it’s having a handful that really counts.

  Minor physical anomalies—like many other markers for violence—have not been systematically assessed in serial killers. But they have been systematically assessed in research studies on a wide variety of antisocial populations of different ages, ranging from troublesome toddlers to violent adults. Beginning with a breakthrough paper in Science, minor physical anomalies have even been linked to peer aggression as early as age three.41 Again, in another study at the preschool level, more minor physical anomalies have been found in aggressive and impulsive boys.42 Moving on a little into elementary school, boys with problem behaviors have more anomalies.43 Transferring to secondary school and to the troublesome teens, minor physical anomalies in boys assessed at age fourteen predicted violent delinquency at age seventeen. Interestingly, at this age, the relationship was specific to violent offending—it was not observed for nonviolent forms of delinquency.44 In this study the effects could not be attributed to potential confounds such as family adversity. At about the same age—but from a biosocial perspective—minor physical anomalies in seven-year-olds combined with environmental risk factors in predisposing the children to conduct disorder at age seventeen.45 This highlights again the biosocial key that we saw when we looked into birth complications—the interaction between a biological and a social factor in predisposing someone to antisocial behavior.