Unformed minds: juveniles, neuroscience, and the law

Unformed Minds: Juveniles, Neuroscience, and the Law Oren Harman Evan Miller was 14 years old when he and his friend Colby Smith, 16, discovered that their neighbor in the Country Living Trailer Park between Speake and Morgan County, Alabama, used to own a baseball card shop and still sported a pretty impressive collection. While he was in the Miller family’s trailer, using their telephone, drunk and apparently uninvited, they snuck into his trailer and came out with wads worth of cards. Later that evening, July 15, 2003, they returned to his trailer and beat the hell of out him with a baseball bat, stealing his wallet. They came back a few times after that, each time aggravating the assault and searching out more cash, $350 in all. At one point, late into the night, Evan Miller placed a shirt over his weeping neighbor’s head and said: “Cole, I am God and I come to take your life”. Then the two boys set the trailer on fire. The next day, the country coroner determined that the cause of death of Cole Cannon, 52, was blunt force trauma, multiple rib fractures, and smoke inhalation. In October 2006, tried as an adult by an Alabama jury, Evan Miller was found guilty of capital murder during the course of first-degree arson, and sentenced to life in prison without parole.1 From the eighteenth-century “boy murderer” William York, 10, who mercilessly carved up his 5-year-old bedmate, Susan Matthew, in the Eyke poorhouse of Suffolk, all the way to the Columbine High School shooting and Newtown Connecticut massacre, young murderers have proved as ubiquitous as they are shocking. According to United States Department of Justice statistics for 1996, one of five child murders were committed by children2, and the FBI Uniform Crime Report of 2009 cited 1,494 child homicides in the United States, all committed by kids until the age of 18.3 Adolescent murderers are a tragic scourge, and, unfortunately, nothing new. What is new, however, is a certain kind of defense being marshaled to get them off the hook. Neurobiology has in the last decade learned a lot about the brain, and adolescents, specialists claim, don’t have an adult one. Increasingly, science is being brought into the courtroom to argue that, when it comes to their actions, adolescents cannot be blamed for not being entirely in control. I. It seems to make good sense. Science tells us what the world is like and how things behave, law tells us what the world should be like, and how people ought to behave. Science observes and analyses, law considers and decides. Science is international, law parochial. Scientists use instruments, calculations, experiments and equations; lawyers and judges use precedents and words. In searching out the truth and then acting upon it with moral conviction for the Seth Burkett, “Jury Finds Teen Guilty of Capital Murder”, The Decatur Daily, October 21, 2006. http://www.ojp.usdoj.gov/bjs/pub/ascii/cvvoatv.txt 3 United States Department of Justice, Federal Bureau of Investigation (2009) “Crime in the United States: Uniform Crime Report 2008”. Retrieved from http://www.fbi.gov/ucr/cius2008/offenses/expanded_information/data/shrtable_02.html 1 2 © Elsevier Ltd. 2013 2 benefit of society, a more complementary relationship could hardly be imagined. Take a simple example: milk. After the German physician Robert Koch discovered the bacterium responsible for tuberculosis in 1882, and Pasteur a way to kill it, national statutory schemes could be devised authorizing courts of law to punish milk producers who do not pasteurize. At the flick of a flame and the drop of a gavel, deaths due to tuberculosis plummeted. It was a marriage made in heaven: marching hand in hand, science and the law would make a better world. And yet the use of scientific knowledge by legislators and science-based adjudication are hardly one and the same. In his 2004 book Laws of Men and Laws of Nature: The History of Scientific Expert Testimony in England and America, the historian of science Tal Golan showed convincingly that when scientists began to be called into the courtroom, all hell broke loose.4 In the eighteenth century, the first Earl of Leicester, otherwise known as “Farmer” Coke, reclaimed salt marshes near the harbour at Wells-Next-the-Sea in Norfolk, building embankments which the Harbour Commissioners argued caused the harbour to silt up. The Earl was no pushover, though, and summoned Robert Mylne, surveyor of St. Paul’s, and John Smeaton, renowned civil engineer and designer of the third Eddystone Lighthouse, to testify as experts on his behalf. Both were Fellows of the Royal Society, Britain’s sanctum sanctorum of privileged knowledge. But how was the testimony of these men to be taken? Was their evidence even admissible? On what grounds should their pristine theories trump the experience of local harbour-masters, men of the field? When it came to figuring out what really was happening in the harbour, could Mylne and Smeaton swear an oath to the validity of the laws of motion of tides and silt? Ultimately, the magister, Lord Mansfield, deemed that they could. “In matters of science,” he ordained, “no other witnesses can be called.” The decision blurred the distinction between experts appointed by the courts and partisan witnesses summoned by the adversaries, opening up English courts to endless battles between what Golan calls “proto-scientists who functioned like skilled professionals but cogitated like natural philosophers.”5 Soon, men of science found themselves at each other’s throats in bitter cases involving fire insurance claims, industrial pollution, patents, even murder. Before the early 1900s, for example, scientists disagreed about whether human and animal (mammalian) blood could be distinguished; in one famous case, a man’s life hinged on whether bloodstains at the crime scene were those of his victim or of a goat. And what about X-rays, first introduced in medical negligence claims? Unlike ordinary photographs, X-ray images reproduced more than any witness could ever swear they saw, and yet experts eruditely pontificated to juries, often in elevated tones, on behalf of each side. Soon, informed by developments in the science of psychology, the very admissibility of witness testimony was put on the stand; after all, science was showing, memory is fallible, a dangerous, tricky beast. How then could the truth ever be established? Experts were paid, often handsomely, to disagree. Tal Golan, Laws of Men and Laws of Nature: The History of Scientific Expert Testimony in England and America (Cambridge, Mass: Harvard University Press, 2004) 5 Laws of Men and Laws of Nature, p. 44. 4 © Elsevier Ltd. 2013 3 From the beginning, money was an issue. Speaking in the later nineteenth century, Sir Norman Lockyer, the founder of Nature magazine, lamented the “alien spirit repugnant to students of pure science in this country”, attacking professionals who made their living from expert testimony for thinking only of “self and pelf”, and stooping so low as to become “persona grata to limited companies.” Science, he thought, should be above this. But professionals of another class, mocking gentleman aristocrats and their stuffy learned societies, maintained that only in the courtroom could progress in science be “so clearly epitomized.” They’d be happy to give expert testimony for a buck. That experts charge large sums to appear in court raises serious questions about the validity of their testimony: after all, providing unequivocal opinions can go a long way towards securing one’s next court appearance, and check. But as serious as this problem may be, there looms an even greater one, far more fundamental. Increasingly, legal systems are being asked to determine not just what the scientific evidence shows, but what counts as scientific evidence in the first place After the lie detector or “polygraph” was invented in 1921 — by a medical student who happened to double as a police officer — it didn’t take long before its use at the bench was put before the Court of Appeals of the District of Columbia. In its decision in the 1923 case Frye v. the United States, the court ruled that expert testimony must be based on scientific methods that are sufficiently established and have “general acceptance” in the relevant scientific community, thereby prohibiting the lawyers of James Alphonso Frye from using the controversial lie detector in his defense (Alphonso went to jail for second degree murder). That was how things stood until 1993, when the US Supreme Court considered the case of Daubert v. Merrell Dow Pharmaceuticals (prompting the amusing academic title: “Frye, Frye Again: The Past, Present and Future of the General Acceptance Test”).6 Tragically, Mrs. Daubert had taken the company’s prescription drug Bendectin when she was pregnant, and her son Jason was born with serious birth defects. Experts from both sides argued over whether the defects could be pinned to the drug. Ruling that the evidence in the case did not meet the standard for admitting expert testimony in a federal trial, the court nevertheless refrained from determining a sine qua non definition, offering a “flexible” list of relevancies instead. That a theory be falsifiable and testable seemed to the judges important, as did its reliability and error rate. But, treading the thin line between admitting “irrational pseudoscientific assertions” and succumbing to a “stifling scientific orthodoxy”, the judges unanimously held that prior publication and peer review are not indispensible prerequisites for admission of scientific evidence. “Scientific conclusions are subject to perpetual revision,” Justice Blackmun concluded. “Law, on the other hand, must resolve disputes finally and quickly.”7 And so, bye bye Frye (the wordplay possibilities never end!). Facts, according to the court, don’t just become facts when a scientific community says so. By David E. Bernstein, BNS Expert Evidence Report, 2, no. 3, February 18, 2002. For a discussion of this ruling see George J. Annas, “Scientific evidence in the courtroom – the death of the Frye rule”, Legal Issues in Medicine, 440, no. 14, 1994, pp. 2018-2021. 6 7 © Elsevier Ltd. 2013 4 II. Which brings us back to Evan Miller, the 14-year-old killer. What exactly is being claimed, in his and other minors’ defense, about adolescent brains? A recent article in the journal Issues in Science and Technology by Laurence Steinberg, a distinguished professor of psychology at Temple University, lays out the argument.8 The maturing brain undergoes marked structural changes: first, there is a decrease in gray matter in the prefrontal cortex – a kind of “pruning” of unused connections, or reduction in unnecessary “noise,” associated with improvement in basic cognitive abilities and logical reasoning. Second, substantial changes begin in the density and distribution of dopamine receptors in two places: in the pathways connected to the limbic system, where emotions are processed and rewards and punishments experienced; and in the prefrontal cortex, the brain’s chief executive officer. Since dopamine plays an important role in how pleasure is experienced, these changes influence sensation-seeking. Thirdly, fatty tissue called myelin wraps around neurons, speeding up signal transmission, especially between different brain regions, which is important for the ability to plan ahead, weigh risks and rewards, and the making of complicated decisions. Finally, during adolescence there is a strengthening of the connections between the higher brain (the prefrontal cortex) and the more basic animal brain (the limbic system) – connections that are crucial for developing control over our emotions. All these changes, it was once believed, were more or less complete by childhood. But recent research shows that they continue well into early adulthood. Functional as well as structural changes characterize adolescent brain development. Functional MRI studies have shown that, in the course of tasks requiring self-control, adults’ brains make use of a wide network of brain regions, whereas the brains of adolescents employ fewer regions. A further set of studies shows that when the brains of adolescents are scanned before the playing of a game in which attractive rewards – such as piles of coins, or pictures of happy faces – are shown to them, the reward centers in their brains activate much more strongly than those of either children or adults. At the same time, there is comparatively very little activity in the prefrontal cortex, the region associated with decision-making and impulse control. In sum, then, heightened sensitivity to anticipated rewards, combined with lower levels of self control, may be responsible for motivating adolescents to engage in risky behavior with potentially high pleasure returns: unprotected sex, for example, or fast driving, or experimenting with drugs. A teenager is thus like a car with a fancy accelerator but no brakes. “With powerful impulses under poor control,” Steinberg writes, “the likely result is a crash”. In the late 1980s and early 1990s, following a number of high-profile school shootings and a rise in violent youth crime, the legal system in the United States began cracking down on juvenile criminals, increasingly trying offenders as Laurence Steinberg, “Should the science of adolescent brain development inform public policy?”, Issues in Science and Technology, Spring 2012, pp. 67-78. 8 © Elsevier Ltd. 2013 5 adults and focusing on punishment rather than rehabilitation. Now, however, advocates for adolescent rights have adopted the new neuroscience of the adolescent brain in their struggle to reverse the trend. And it seems to be working. In 2004, 17-year-old Christopher Simmons admitted to the planning and murder of Shirley Cook, whom he tied up with duct tape and electrical wire and threw off a bridge. Writing for the majority, which in a 5 to 4 decision found it unconstitutional to sentence the minor to death, US Supreme Court Justice Kennedy noted that there is a body of sociological and scientific research showing that juveniles don’t share the same degree of a sense of responsibility for their actions as adults. Immature development corroborated by “scientific research” was cited again in Graham v. Florida, 2010, in which the court, in a 6 to 3 decision, deemed life without parole for juveniles for crimes other than homicide as “cruel and unusual punishment” – and so a violation of the Eighth Amendment and hence unconstitutional. “Developments in psychology and brain science,” Justice Breyer wrote, “continue to show fundamental differences between juvenile and adult minds.”9 While not quoted directly, briefs submitted by the American Medical Association and the American Psychological Association on behalf of the defendant, explaining the current state of research on adolescent brain immaturity, had clearly informed the judges’ verdict. Many scientists and advocates for juveniles were thrilled. Here was a clear victory for neuroscience in the courtroom, they thought -- a landmark decision as profoundly important in ending neurological injustice as the 1954 Brown v. Board of Education of Topeka, ending segregation in public schools, had been in ending racial injustice. Though it is not widely remembered now, that case too was a victory for the advanced psychology of the day, as the decision is thought to have been heavily influenced by a study of children of different races playing with dolls of different colors, indicating that segregation had a markedly negative effect on black students’ self-esteem. III. But all is not quite as it seems. To begin with, Christopher Simmons, the 17-yearold murderer of Shirley Cook, admitted that, days before the murder, he had told friends that he was planning on killing someone by tying the victim up and throwing her off a bridge. How could brain-scan evidence showing that adolescents are more likely to act impulsively have anything to do with this obvious premeditation? Consider too that comparison with Brown v. Board of Education. As pointed out in a news feature by Lizzie Buchen in Nature magazine last April,10 the parallel with the Simmons case may extend in a direction that is far from reassuring. The doll study at the time, it transpires, had been more or less ripped to pieces by sociologists even before the ruling – and the judges knew it. In fact, an admission to this effect was tucked away in a small footnote in the official ruling. Clearly, in both cases, something other than science was driving the law. 9 Graham v. Florida, 560 US, 2010. Lizzie Buchen, “Science in court: arrested development”, Nature 484, April 18, 2012. 10 © Elsevier Ltd. 2013 6 Another source of concern is how little the recent neuroscientific consensus adds to what was already known. It will hardly be news to parents and teachers, after all, that adolescents are impulsive, susceptible to peer pressure, thrill-seeking, and altogether rather weak at weighing instant pleasures against slightly longerterm goals. Nor is it news to sociologists and behavioral psychologists, who have been researching the area for decades. What, then, does the neuroscience contribute? Some, commentators, among them Steinberg, argue that present research is a step towards a future neuroscience capable of giving to courts the power of distinguishing between those who couldn’t stop themselves from committing a crime and those who could stop themselves but chose to commit it anyway – a rather important distinction when it comes to culpability before the law. Such power would be amazing, perhaps even fantastic. But the day science solves the problem of free will, alas, is still rather far away. More immediately relevant, it seems, is the claim that correlations found in the recent neuroscience studies are indications of the true causes of behavior. But here too there is room for doubt. Do we really know that mylenated neurons are the cause of our ability to plan ahead, or that axon “pruning” during development is what allows us to think logically? The honest answer has to be that we don’t yet know, in any secure sense, anything of the sort. A correlation, however robust, falls far short of a tested causal mechanism -- especially one sturdy and exclusive enough to build a murder case upon. To think otherwise is, as Stephen J. Morse has said, to succumb to “brain overclaim syndrome.” In this same mocking- scientism-withscientism spirit, the cure he recommends for this seemingly uncontrollable impulse to claim that observed differences in the adolescent brain cause certain adolescent behaviors is a dose of “cognitive jurotherapy.”11 It is too rarely noticed that, in the studies cited in these court cases, scientists do not image the brains of people in the act of committing a crime. Instead, in the safety of their labs, they ask volunteers to refrain from looking at a light shining to their left. Learning that adolescents are not quite as good at controlling their impulse to peek, the researchers conclude that self-control, in general, is compromised, on average, in teens. Dr. Bea Luna, who has been performing these experiments in her lab at the University of Pittsburgh for the past decade, recently shared with me how satisfied she feels knowing that her work might benefit society, even in a small way.12 Proudly, she sent a commissioned report on the state of research to the Supreme Court as it prepared to deliberate over the appeal in Evan Miller’s case. But Luna is also conflicted, for she appreciates that, while we might know quite a bit about the brain circuitry involved in stopping a reflexive response, delinquently glancing at a light is not exactly the same as beating a helpless man with a bat and setting his trailer on fire. Responsibly, Luna makes it clear in her reports that scientific evidence refers to adolescents as a group, never to a particular individual.13 But how then are Stephen J. Morse, “Brain overclaim syndrome and criminal responsibility: a diagnostic note”, Ohio State Journal of Criminal Law, 3, 2006, pp. 397-412. 12 Personal communications with the author, October 28-29, 2012. 13 See Kayla Pope, Beatriz Luna, Christopher Thomas, “Developmental neuroscience and the courts: how science is influencing the disposition of juvenile offenders”, Journal of the American Academy of Child and Adolescent Psychiatry, 51, 4, April 2012, pp. 341-342. 11 © Elsevier Ltd. 2013 7 judges and juries to decide when this group-level data is relevant to a given criminal case. Isn’t that tantamount to using statistics for the prevalence of black crime as evidence against a black man? And how, furthermore, will it be decided who counts as an adult? In distinguishing between people who are ready for the rights and responsibilities of adulthood and those who are not, the bar could be set as low as age 15 and as high as age 22. But setting the lower bar would result in treating many immature individuals as adults, which is dangerous; while setting it at 22 would result in treating many mature individuals as children, which is unfair. What do we do? In their 2010 book Rethinking Juvenile Justice14, Steinberg and Elizabeth Scott argue that instead of randomly picking the mid-point of the range, or deciding the threshold for adulthood on an issue-by-issue basis (driving at 15, say, voting at 18, drinking at 21), the binary classification system child/adult should be abandoned altogether in favor of a third legal category: the adolescent. In effect, this is already happening to some degree in some states in America, where a graduated model allows kids to drive at a certain age, but only with parents; or in connection with child labor laws, where kids are allowed to work in certain jobs but not in others, and only for a limited number of hours. Steinberg and Scott think that this principle could be extended to drinking laws, too (beer and wine at 18, whiskey and vodka at 21), though it is harder to imagine how a three-tier system would work for voting. Nevertheless, in their view, adolescents should not be treated in the same way as adults when it comes to punishment for crime; the mounting neuroscientific evidence, the authors claim, cannot be ignored. But again, we should ask: is that true? Let’s return to the footnote on the doll study in Brown v. Board of Education. The justices buried it there for a reason, knowing full well that the study’s conclusions were less than firm. The point can and should be stated much more generally. In the final analysis, science cannot always come to the rescue to solve our problems. One reason is that scientists often simply don’t have the answer. They don’t know that because the nucleus accumbens – a brain region that seems to play a role in experiencing reward – lights up more in adolescents than in adults in particular lab experiments, that this means that adolescents are biologically more impulsive and therefore more likely to do something stupid under peer pressure. They don’t know this for adolescents as a group, and they definitely don’t know this for any particular teenager. What is known, on the other hand, and from years of solid research not in neuroscience but in social science, is that teen criminals have a better chance of reforming; that life sentences do little to deter adolescents from committing crimes; and that there are serious racial inequalities when it comes to lifewithout-parole sentencing. So long as the width of myelin sheaths covering neurons in the cortex cannot be translated into solid predictions concerning crime – and that will probably never happen – then, when it comes to considering the question of sentencing adolescents for life, a case can be made that neuroscience can be ignored. Elizabeth S. Scott Laurence Steinberg, Rethinking Juvenile Justice (Cambridge Mass.: Harvard University Press, 2010). 14 © Elsevier Ltd. 2013 8 There is a second, more fundamental reason, why science shouldn’t be considered a panacea in such situations. Imagine we knew that a given brain will lead a given male adolescent to commit a given crime with a ninety percent certitude (once again, something never likely to happen). The knowledge would undoubtedly be informative. But whether we decide to incarcerate this potential criminal for precautionary reasons, or invest heavily in his education, or, once a crime has been committed, mitigate his sentence due to diminished responsibility, are determinations to be made based on legal and ethical, not scientific, considerations. For, again, science may overreach, substituting correlations for causations, for example, or assuming that just because a particular system in the brain causes a behavior that it cannot be changed. Daubert v. Merrell Dow Pharmaceuticals is, in this light, and on balance, probably a good thing. Just because a scientific community deems something to be true doesn’t mean it’s ordained in heaven. And in any case, as Wittgenstein taught, “even if all scientific questions be answered, the problems of life have still not been touched at all”.15 Scientists cannot make our moral decisions for us. But, oh, how we’d like them to! Otherwise, why would we turn so often to science to resolve our moral dilemmas? And why would we – Supreme Court justices included – feel compelled to cite ostensibly “hard” neuroscience, which is really quite crude, alongside “soft” social science research, which is actually quite reliable? Surely, this tendency is a mark of the cultural power that science has gained in our times, fulfilling the most optimistic hopes of Magister Lord Mansfield, if not of ‘Farmer’ Coke at the Norfolk harbour docks. We do well to remember, however, that the sword of science is double-edged. The same advocates who claim neuroscientific grounds to mitigate punishment for teen crime can also, drawing on similar sorts of evidence, fight the battle to allow juveniles to make their own decisions about abortion. Which is it then? Are adolescents constitutionally too impulsive to be judged as adults when they murder, or are they sufficiently adult to be considered fit to deal with the consequences of having sex? Science can bring about changes in values. But usually it’s already prevailing values that dictate how science gets used, and to what ends. IV. On June 25, 2012, in a 5 to 4 decision, the US Supreme Court decided that sentencing juveniles to life without parole for homicide is unconstitutional. Delivering the opinion of the court, Justice Kagan quoted from Graham v. Florida that “developments in psychology and brain science continue to show fundamental differences between juvenile and adult minds” - for example, in “parts of the brain involved in behavior control.”16 Since the court reasoned then that such findings both lessened a child’s “moral culpability” and enhanced the prospect that, as the years go by and neurological development occurs, his Ludwig Wittgenstein, Tractatus Logico-Philosophicus, trans. D.F. Pears and B.F. McGuinness (London: Routledge, 1974), 6.52 16 Miller v. Alabama, 567 US, 2012. 15 © Elsevier Ltd. 2013 9 “‘deficiencies will be reformed,” they saw no reason now, in Miller v. Alabama, why this should be true for assault or larceny, and not for murder. Depending on your politics, this development may or may not be a good thing. What is certain is that the tales behind it are far from happy. Evan Miller was an abused and neglected kid, in and out of foster care. He began drinking before he was ten, was a drug addict, and hardly went to school. Cole Cannon, the man he murdered, was his mother’s drug dealer. Perhaps it’s important to consider brain development when sentencing young people to jail. Surely it isn’t entirely irrelevant. But even when done properly, it can never be more than one consideration, among many, in helping us inch closer to justice and, ultimately, a better world. Most adolescents don’t murder, however unformed their minds, after all. On the other hand, the problems of drug addiction, poverty, abuse, poor education, and gun control are real enough, and tough, and they won’t go away over night. As alluring as a fancy MRI picture of a brain may be, we must never give up looking those who struggle with these problems straight in the face. And continue fighting the good fight. Oren Harman is the Chair of the Graduate Program in Science Technology and Society at Bar Ilan University, and the author of The Price of Altruism: George Price and the Search for the Origins of Kindness (W.W. Norton, 2010), winner of the 2010 Los Angeles Times Book Prize. © Elsevier Ltd. 2013