1 Introduction

Here is the opening of the summary of an amicus brief in Roper v. Simmons, the 2005 United States Supreme Court decision holding the death penalty unconstitutional for all juvenile capital offenders. It was filed by, inter alia, the American Medical Association, the American Psychiatric Association, the American Academy of Child and Adolescent Psychiatry, and the American Academy of Psychiatry and the Law:

The adolescent’s mind works differently from ours. Parents know it. This Court [the United States Supreme Court] has said it. Legislatures have presumed it for decades or more.Footnote 1

This blinding insight was confirmed in Miller v. Alabama, which held that mandatory life imprisonment for juvenile offenders who had committed homicide crimes is unconstitutional. In dissent, Chief Justice Robert said, “[…] teenagers are less mature, less responsible, and less fixed in their ways than adults—not that a Supreme Court case was needed to establish that.”Footnote 2

Precisely. The common law of immaturity in a wide variety of criminal and civil contexts and the establishment of a juvenile court system in the 19th C. confirm that the law has long responded to the psychological differences between juveniles and adults. Indeed, Roman law instantiated the difference and the English common law of infancy has its roots in the 14th C.Footnote 3 The age of majority has varied from time to time. In the mature common law of criminal responsibility, kidsFootnote 4 younger than 7 were conclusively presumed to be not responsible, kids 7–13 were rebuttably presumed not responsible, and kids 14 and older were presumed responsible. Today, the dividing line in the United States for most purposes is 18 and varies for criminal responsibility; in English law, the age of criminal responsibility is 12. But the important culpability and competence differences between those older and younger than the dividing line were always assumed.

It was equally and always assumed, implicitly or explicitly, that those differences and the differential legal treatment that followed were based on differences in the capacities of the two groups, such as the alleged greater average impulsiveness of kids, which could vary according to the context. Committing a serious crime and deciding whether to have an abortion are different behaviors that seem to require different capacities for responsibility. Differences may be more important for some legal categories than for others. Age is allegedly a good but imperfect proxy for these differences. Virtually all important juvenile law scholars, such as David Brink, Barry Feld, Elizabeth Scott, Laurence Steinberg, Franklin Zimring, and a host of others agree. This was, and still is, the received wisdom.Footnote 5 Let us refer to this as The Received Wisdom [TRW], recognizing that it may have different forms. TRW has been substantially bolstered in recent years by increasingly sophisticated behavioral science that confirms substantial behavioral differences on average between adults and adolescents, many of which are relevant to classic responsibility criteria such as rationality broadly conceived, and by neuroimaging studies demonstrating that the brain’s anatomical maturation does not end at any of the usual legal dividing lines but continues into the mid-20s.

Arguing against TRW, Gideon Yaffe’s The Age of Culpability [AC], an immensely rigorous account of why kids should be given a break for criminal responsibility, suggests that everyone else has provided the wrong rationale for differential criminal justice treatment. TRW is so entrenched that, to the best of my knowledge, AC is the only serious counterargument ever offered for giving kids a break. According to AC, the reason kids deserve a break does not depend on different capacities, but instead is better justified on the basis that kids generally have much less say about the legal criteria for criminal responsibility, primarily because they do not have the right to vote. AC argues that kids should be given a break because Yaffe’s reason-responsive view of criminal responsibility entails that they have fewer legal reasons to obey than adults and are therefore less culpable. AC argues that age has political meaning rather than significance because it betokens behavioral capacity differences.

This creative and intricately argued volume has many interesting chapters that deploy original and nuanced arguments that develop AC’s theory of responsibility, especially the chapters on culpability and desert. These are challenging and interesting in themselves and less (but still quite) controversial than the argument about giving kids a break. They should be read by all those interested in criminal culpability generally. I suspect that this wider theory of responsibility is AC’s major quarry and deserves its own review. Nonetheless, the general theory is nominally in aid of the volume’s basic goal of re-orienting how we should think about giving kids a break. Because the primary goal is re-thinking why the criminal justice system should treat kids differently, this essay will therefore focus on that question.

The question for criminal law theorists is whether TRW or AC’s rationale is a better positive explanation and more normatively desirable. AC’s argument is labyrinthinely complicated. Ockham’s razor is dulled with every stroke. It is so densely argued that a review that does it justice would be as long as the book, and perhaps longer. This is unsurprising when one is seeking to upend the legal, scholarly and commonsense wisdom (or lack of it) of millennia by philosophical argument rather than by empirical disproof. A great deal of heavy lifting will be necessary. By itself, the intricacy of the necessary argumentation and the hoary countervailing history do not mean that AC is wrong. But AC clearly has the burden of persuasion on historical and methodological grounds. In this case, I am not persuaded: TRW is a better fit to the data and furnishes a more appealing ground for giving kids a break than lack of say over the criminal law. TRW holds that the law gives kids less say because on average they are less rational and experienced than adults. Indeed, that is the response that virtually all people have when hearing AC’s argument.

AC rightly and usefully notes that there are three independent positive arguments for treating kids differently: age is a proxy for diminished culpability, kid will be kids, and kids will grow out of it. It also correctly observes that some advocates for TRW conflate the three. Nevertheless, the proxy argument is the most common and important position. After addressing a few preliminary issues, I therefore discuss AC’s negative argument against the validity of TRW, focusing on the diminished culpability claim, especially because the structure of the negative argument against the other two is structurally similar and because it is unclear that kids will be kids is a genuinely independent argument. I conclude that the negative argument does not succeed.

If the negative case against TRW fails, as I claim, then the only issue is whether AC’s alternative is desirable. Again, the essay argues that it is not, and concludes by offering three positive arguments for the traditional rationale: (1) coherence and simplicity; (2) a benignly definitional argument that survives the negative argument and supports giving all juveniles a break in the exceedingly unlikely event that the empirical assumptions of the traditional rationale are proven incorrect; and (3) a proposal for individualization of the culpability assessments of juveniles who commit serious crimes so that the criminal justice system blames and punishes them proportionately to their culpability. Throughout, my primary goal is not to take issue with AC, although it will seem that way, but to offer what I hope are helpful thoughts about how kids should be treated in the criminal justice system.

2 Preliminaries

AC’s argument is ahistorical. As the Introduction noted, kids have been treated differently from adults in civil and criminal law going back to Roman law. The difference in contractual obligations is a primary example. The explanation is virtually always that kids have lesser capacities than adults. To continue the example of contracts, kids are more likely to have bad judgment and to be more easily taken advantage of because they are less rational and experienced. It is highly unlikely that kids may avoid their contracts in most instances because they have little say over contract law. It is of course possible that criminal liability is distinguishable from civil duties and protections, but where the law seems uniform in differential treatment and for uniform reasons—to wit, some version of TRW—drawing that distinction would have strengthened AC’s case.

It is important to clarify a point that advocates for kids who base their arguments on TRW often confuse. Finding an average psychological or biological difference between kids and adults, even a big one, does not entail different legal treatment. First, the difference must be relevant to the issue at hand, criminal responsibility in this case. Second, even if the difference is relevant, it must be sufficiently large normatively to justify differential legal treatment. The most egregious example of the first problem is the belief that neuroanatomical maturation differences between adolescents and adults entail lesser responsibility for the former. Neuroanatomical maturation is not a relevant responsibility characteristic, however. The criteria for responsibility are behavioral, namely acts and mental states. The claim based on neuroanatomy is a category error. AC scrupulously and correctly does not make this mistake. It says that neural facts, if they matter at all, only do so if they underlie a psychological state or capacity that does matter.Footnote 6 An example of the second is impulsivity. The Supreme Court and many others seem to think that impulsivity diminishes responsibility, but it is not a defense to criminal conduct and it would not be a standard mitigating factor in adult sentencing. Consequently, there seems to be a relevance problem. But let’s assume that the Supreme Court is right. In that case, is the average impulsivity difference between adolescents and adults large enough to justify lesser criminal responsibility for adolescents as a class? Maybe or maybe not, depending on what the legislature or the courts believe is the minimum for full responsibility. Even if TRW is the best justification for treating kids differently, these two issues must be faced forthrightly without begging the question.

AC addresses the proxy argument with a clarifying analysis of the costs and benefits of any legal scheme that is perhaps the most penetrating that I have read. It correctly recognizes, as all sensible TRW advocates concede, that the behavioral characteristics that are relevant to responsibility will be both over-inclusive and under-inclusive no matter where we draw the dividing line, at least among adolescents. Some kids below the line will have the relevant responsibility characteristics and some adults over it will not, raising the possibility of true and false positives and negatives in decision-making. AC cautions that any adequate policy analysis must therefore include a full assessment of all the true and false positives and negatives and cannot simply focus on the virtues of the true positives, as AC accuses the Supreme Court of doing in Miller.

Where, given our values, the age line should be drawn in any legal context, such as criminal responsibility, will depend on the full costs and benefits. Nevertheless, because no proxy is perfect, there will be undoubted costs wherever the line is drawn. We should recognize that no legislature will be perfectly able to do that full accounting. After holding hearings and presumably listening to experts as well as pressure groups, the legislature will fly by the empirical seat of its pants. Among “equally implementable” alternatives, to use AC’s phrase—and in this instance the age lines are all equally implementable—we can never be absolutely sure which is the best, nor should we expect such certainty. As Bismarck observed, watching legislation and sausage being made are very similar endeavors. Neither is much like developing armchair theory.

Before turning to AC’s substantive criticism of the proxy argument, it is important to recognize what a false positive means in juvenile cases and its systemic effects. A psychologically mature minor who does not really deserve a break according to TRW will get one, but that does not mean the juvenile goes entirely free. Instead the juvenile will receive diminished blame and punishment, but he or she will still be punished and often substantially for serious crimes. Thus, retribution, deterrence and incapacitation in appropriate cases will be served, albeit not as fully as if the kid did not get a break. But how much marginal retribution, deterrence and incapacitation will be lost with such false positives? I do not know the answer and neither does anyone else, but would it be so objectionable if some young people were not subjected to the full deserved afflictive imposition of state blame and punishment?

3 The Negative Argument Against TRW

AC’s negative argument, which it terms the “empirical dependence” claim, starts with an intuition Professor Yaffe believes is widely shared: Even if the best behavioral science were to indicate unquestionably that there were no substantial responsibility-relevant differences between kids and adults or even if there were an equally implementable policy with a better mix of gains and losses that would not give all kids a break, we would still want to give all kids below the age cut-off line a break. AC concedes that a “no differences” finding is exceedingly unlikely scientifically; indeed the increasing accumulation of scientific information continues to confirm the differences. But if it turned out to be true, it would make no difference. We would still want to give all kids a break. AC confidently asserts that the proxy for culpability argument therefore must be false and that the true ground for categorically giving kids a break is to be found in some other rationale.

With respect, I do not share that intuition, nor did all but one of the students in my “Freedom and Responsibility” seminar with upper division Penn Law students. What AC terms “our implicit and unarticulated rational for giving kids a break”Footnote 7 is perhaps not nearly as widely-shared as AC believes and empirical dependence is not nearly as problematic as it avers. Average capacity differences are not a metaphysical or conceptual truth about kids of any age. They are empirically grounded and could turn out to vanish upon further investigation. (If they did, would you believe the studies or your lying eyes?) As a practical matter, however, differential capacity is not contingent although in principle, as AC rightly argues, it is. It is a safe prediction that these differences will endure empirically, so perhaps it’s too easy to claim that I would stop giving all kids a break because I know I will never have to face the possibility. But I do not think that is true for a reason prompted by the second pillar of AC’s intuition: that we would not give up on giving kids a categorical break even if there were an equally implementable policy that had a better mix of losses and gains. I think such a policy already exists, however, individuation, and I would adopt it. I will discuss it in the last section below, but under this policy some but not all kids would get a break.

But for those who accept TRW and want to give all kids a break, will AC force them to concede that the proxy for culpability ground fails? I think not. AC begins by hypothesizing a not completely hypothetical scheme by which all boys under the age of 18 get a break, but girls only get a break if they are under 16 on the ground that girls mature faster and therefore do not differ from adults at an earlier age than boys. AC admits that no one knows if, all things considered, this scheme would be preferable in gains and losses to a gender-neutral scheme. It also allows that it might well be better and says that if it is, TRW adherents should accept it. AC assumes that most people would nevertheless recoil from this scheme, but if they do, it claims, they should reject the proxy argument or at least demand that it be supplemented. After all, the state should not be inflicting blame and punishment according to some proxy if there is a superior proxy. Gender is of course a controversial variable to use to improve the proxy argument, but it may be the most useful if one is aiming for accuracy and efficiency in identifying those kids who deserve a break. AC gives other similar examples,Footnote 8 but let us use the gender example because the argument structure is similar in each case. That is, it seeks to identify every possible argument that might support the proxy for diminished culpability and then demonstrate that it cannot be right.

AC notes that attachment to giving all kids a break might be irrational if a superior proxy were found, but tries to determine if qualms about the gender-based classification might be both rational and consistent with the proxy argument. In another words, like the excellent philosopher that he is, Professor Yaffe puts maximum pressure on his own argument. For example, appealing to more general bad consequences of a policy, such as discriminating against women leading to the perpetuation of negative stereotypes, misses the point. Consequences are not all that matters to criminal liability: desert is crucial. The aversion to a gender-sensitive classification for criminal liability cannot be properly based on this ground if gender is superior for ascribing blame and punishment. Another possibility is that gender-based classification would unequally distribute benefits and burdens. But, as AC points out, even giving all kids a break does that because those defendants who are adults are apportioned heavier burdens than kids, and our society believes that is just because there is a genuine moral difference between adults and kids. It follows that if gender classification is superior, then there is a moral justification for the gender inequality that is as sound as the adult/youth inequality. Just blame and punishment are essential government functions, AC rightly notes, and doing it better should lead us to adopt the superior policy of gender-sensitivity. AC completes its gender discussion by asking whether there is an exclusionary reason not to adopt the gender-sensitive classification if it is more accurate, and suggests that if there is, that is why we must recoil from this type of scheme. It concludes that there is not such a reason because there are permissible contexts for gender-specific policies, such as TSA employees who must pat down passengers and because oppression based on gender is not a feature of criminal punishment.

After showing why the objections to the gender proxy can be defeated, AC says this:

[…] my point is not that all this shows the we ought not to give breaks to girls between the ages of 16 and 18. Far from it. I think the gender sensitive policy should be laughed out of contention. My point is that the advocate of the Proxy for Culpability argument cannot laugh at it, but must instead entertain the serious possibility that it is more justified by the standards informing the Proxy for Culpability argument in the first place. This is an objection to the Proxy for Culpability argument because the gender-sensitive policy is unjustified even it if offers superior classification than the policy of giving all kids a break regardless of gender.Footnote 9

With respect, I do not think that gender classifications for purposes of ascribing blame and punishment are necessarily laughable, but instead are sometimes plausible. Consider Section 54 of the English Coroners and Justice Act that Parliament passed in 2009 and went into effect in 2010,Footnote 10 which replaced the traditional provocation/passion doctrine for reducing an intentional homicide from murder to manslaughter with a new “loss of control” mitigation. If there is a “qualifying trigger” for the loss of control, a broader doctrine than traditional provocation, the question is what degree of control can be expected of the defendant. Subsection1(c) answers that question as follows: “(c) a person of D’s sex and age, with a normal degree of tolerance and self-restraint and in the circumstances of D, might have reacted in the same or in a similar way to D. [emphasis supplied]”.Footnote 11 In other words, English law assumes, perhaps incorrectly but perhaps correctly, that sex is morally relevant to the degree of self-control to expect of ordinary people because it affects morally relevant categories. Further, with respect, the final conclusion that the gender-based policy is unjustified even if superior, is simply an ipse dixit based on the potentially incorrect assumption that we think all kids should be given an equal break.

AC’s assumption that we all want to give all kids a break is simply an assertion not backed by evidence and there is counter-evidence. I do not think the proxy for culpability argument has been decisively refuted.

4 The AC Rationale for Giving All Kids a Break

AC says that the thesis of the book can be summarized by a slogan: “Kids ought to be given a break because they are disenfranchised.”Footnote 12 The positive argument can be stated as follows. Kids have little say over the content of criminal law because they cannot vote, although they can influence the law by speech (if their parents or other controlling institutions allow them to express themselves). Therefore, when deciding whether to violate a criminal law, they have less reason than adults to recognize, weigh and respond to the legal reasons not to offend, even if the offense is a clear moral violation.Footnote 13 As people who can influence the law, adults are complicit in giving themselves legal reasons to obey and in the state’s treatment of them if they offend. Because kids are not so complicit, they have fewer supporting reasons to obey. As a result of the absence of legal reasons, kids are less culpable and deserve a corresponding break.

On average kids have much less say or potential say about the law than adults. Of course there are many adults who have little say, and some kids who may have substantial say because they are the children of legislators or because they become involved in a public controversy, such as the issues of gun control in the wake of school shootings or of climate change. But kids cannot vote, full stop, and most adult citizens can. Moreover, parents have significant authority over how much their children can speak. If culpability differences depend in large measure on how much say kids have compared to adults, the culpability differences and the breaks owed must be very large. The assessment of culpability differences and the breaks given in our system are not so large, however. For example, after Miller, kids convicted of homicide crimes can be sentenced to life without the possibility of parole although such a sentence may not be mandatory. AC does argue that, all things being equal, mitigation is always appropriate,Footnote 14 but perhaps the implication of AC’s argument is that the breaks should be much larger than they are now. After all, as every criminal lawyer knows, the more important question for most defendants is not whether they will be convicted, but if they will then go to jail or prison, and if so, for how long. The size of breaks matters a lot and deserved more attention in AC.

The last paragraph began with the empirical assertion AC adopts that kids have less say about the law. This is not a conceptual truth and thus it itself raises AC’s empirical dependence concern. Suppose we had some measure that operationalized a consensual meaning of “having a say.” In that case, we would have a continuum comparison between kids and adults that could range from kids having lots more say by the influence of their speech to kids having essentially no say. As AC discusses, voting is not the only way to have say, although it is very important. I recognize that the two extreme positions are about as empirically likely as there being no differences in responsibility characteristics between kids and adults, but AC would have to concede it is a possibility and should consider what follows, but does not. Consider the possibilities. If kids have no say at all, then are they culpable at all? If the answer is affirmative, then the disenfranchisement account cannot possibly be doing all the work for why kids should be given a break. It at least must be supplementing TRW or something else. The reverse is also true. If kids have a lot more say, they are more culpable on that ground than adults. If AC still wants to give kids a break, the enhanced culpability resulting from say must be swamped by some other variable that reduces’ kids’ culpability. Once again, the variable must be TRW or something like it. At the least, even if AC’s disenfranchisement account is plausible, it needs supplementation.

The question is why disenfranchisement reduces culpability? What further reason does the law provide beyond the moral reasons not to do wrong? Juveniles are not prosecuted for regulatory, malum prohibitum crimes. They are prosecuted for acts that are at the core of the criminal law: force, fraud and theft in the absence of a compelling justification, which are clearly moral wrong as any ordinary kid knows. Of course, there cannot be criminal blame and punishment if there is no violation of a legal prohibition, and even at the core, law and morality can come apart, although seldom. The state can only impose a legal punishment for a criminal law violation, and kids do not have a say in what those punishments are. Nonetheless, starkly immoral behavior like homicide, aggravated assault, sexual assault, and robbery, and even less serious moral wrongs such as vandalism, would be prohibited no matter who did or did not have a say in the content of the core of the criminal law. Moreover, the threat of a punishment is alone a very weak normative reason not to offend among those who manifest proper “modes of transaction with reasons” (AC’s somewhat ungainly terminology) compared to those underlying the claim that an action is morally wrong.

Even a kid understands what a punishment is, but because they have less maturity, less rationality and less life experience, they may not fully and properly weigh the moral and practical reasons not to offend. If this is correct, then it lends support to the proxy for culpability argument rather than the disenfranchisement position. Kids are less culpable because they, or at least some of them, have diminished responsibility characteristics.

A crucial step in AC’s argument is to explain why kids are not entitled to vote, why they don’t have a say. Suppose for example that the voting age were lowered to 14. Would we now think 14-year olds do not deserve a break? AC’s basic answer is that we do need an age of majority for voting to permit parents to influence the future course of the law by doing their best to pass on their values that kids will promote when they are adults and can vote. The age of majority is when we assume that kids can now be trusted to exercise judgment independent of their parents despite being influenced by their parents.

As always, AC has careful, detailed arguments about this position, but it hardly seems parsimonious or even plausible as an account of why kids don’t have a say. There is merit in the point about people needing to pass on their values to influence the future course of the law and to bind themselves to the continuous polity in that way. Nevertheless, the more fundamental, common sense reason for parental control is that kids need guidance as they develop because they do not possess the rationality and experience to be sufficiently self-directive. Parents have always had control over their kids and tried to inculcate their values, even in polities that did not grant the vote to the parents either. Parents in such polities had no motivation to influence the course of the law, but inculcated values nonetheless.

The less developed capacities of most kids and the lack of experience of all kids are also more convincing reasons why they cannot vote until they have gained sufficient rationality and experience.Footnote 15 Roman and common law and general parental practice grant kids increasing responsibility and holds them increasingly responsible as they grow older not because they are closer to having a full say the closer they get to the age of majority, but because their capacities for rational self-direction increase. Once more, a diminished rationality or responsibility argument seems a simpler and more plausible explanation for the differential treatment of kids than competing theories.

5 The Virtues of TRW

This section first provides a brief general defense of TRW on the grounds of coherence and simplicity. Then it suggests a version of TRW that would guarantee a break to all kids that would not be subject to the empirical dependence argument. Because I am untroubled by that argument and by not giving all kids a break, I offer this version of TRW as a friendly amendment and not as a desperate effort to save TRW. Finally, I offer another version of TRW that is also not subject to empirical dependence but that also does not give all kids a break.

Perhaps the greatest virtue of TRW is that it provides a satisfying and simple, coherent account that ties together sensibly all the different legal contexts in which kids are treated differently by the law. It fits the data not only of the criminal law of juveniles, but of virtually any doctrine that treats kids differently. It is possible, of course, that criminal culpability is distinguishable from all other doctrines, but doing so would require yet another intricate argument. This does not mean that the age of responsibility must be the same across all contexts. For example, I consider the imposition of state blame and punishment such a grave imposition that I would happily raise the age of full responsibility even if kids could vote or make other independent decisions at an earlier age.

The clear intuition is that kids are treated differently because they have different behavioral capacities. AC is correct, however, that the proper task of the theoretician is to furnish a good argument to justify the intuition by showing that the intuition is a rational conclusion to the premises.Footnote 16 Any set of arguments will be open to counterarguments, but as long as the argument itself seems a good one and leads to a sensible result, it will stand until decisively refuted. In the spirit of AC, I wish to offer a TRW argument that conforms to AC’s intuition that all kids should get a break when charged and convicted of crime.

The major premise in the argument is that legally different treatment of classes of people concerning responsibility and competence in both criminal law and civil law depend upon relevant differences in behavioral capacities. Here are some classic examples. Some people with major mental disorder are excused from criminal responsibility, are legally insane, because their cognitive or self-regulation capacities are severely impaired at the time of the crime (this includes minors who face criminal charges). A defendant is incompetent to stand criminal trial if the defendant cannot understand the nature of the charges and proceedings or cannot rationally assist counsel. A person may be involuntarily civilly committed as a mentally abnormal sexually violent predator if the person has serious difficulty controlling his sexual offending. A person is incompetent to contract if the person lacks the capacity to understand the meaning and the effects of the words making up the transaction or contract, in other words, he lacks the capacity to know what he is doing. To make a valid will, the person must know the nature/extent of their property, the natural objects of the property, the disposition that the will is making, and must have the ability to connect these elements to form a coherent plan. It is important to recognize, however, that in virtually all cases, the relevant decision will be individualized and not based on membership in a class. Minors are an exception in competence to contract. Except in exceptional circumstances, they can always avoid a contract because they lack understanding.Footnote 17

The question then is what behavioral characteristic relevant to responsibility do minors have that would justify a categorical approach and give all kids a break in the criminal justice system without regard to their individual behavioral capacities?Footnote 18 The answer is that those who have not reached majority do not have the capacity to fully “own” and to fully take responsibility for their individual mode of recognizing, weighing and responding to reasons.Footnote 19 The capacity to “own” takes time. Anyone with fewer than 18 years of experience (or however many that a polity adopts) hasn’t had enough time to test his reason-responsiveness against experience. Even minors with seemingly fully developed rational capacities lack the experience of self-direction necessary for full responsibility. Youth is a time when we are trying ourselves on for size and deciding what fits and what does not. Sometimes kids are reflective about this, but most probably are not. Nonetheless, it is a process that always occurs implicitly or explicitly. In a profound sense, this process continues throughout life, but youth is when it is most important. This is not an argument for a character theory of responsibility and excuse. Like AC, I reject such theories. It is a claim that a certain amount of life experience is necessary to have the capacity to take responsibility for oneself, especially time in mid to late adolescence when full formal rational capacities are achieved. In a similar vein, AC says, “Parents enjoy legal entitlements the function of which is to entitle them to exert influence over who their children are and will become….”Footnote 20 In short, minors are still becoming.

In a certain sense, this is a definitional argument. Full moral agency requires that people have 18 years of experience because 18 years of experience produces full moral agency in virtue of permitting people fully to take responsibility for themselves. In this case, however, I think the circularity is benign rather than vicious. It is a normative and not empirical claim, although empirical facts about developmental psychology might well influence the age at which majority is reached. Of course, at age 18, some people have more “ownership” of themselves than others, but our polity has simply decided that by age 18, all else being equal, everyone has enough to be treated as a fully moral agent.

Patricia (Patty) Hearst furnishes a good analogous, intuitive example of the importance of time in becoming responsible for one’s self. Many will recall that the 19-year old heiress was kidnapped by a radical group, The Symbionese Liberation Army (SLA), and “coercively persuaded” (i.e., “brainwashed”) while isolated in captivity to adopt their values, attitudes and political agenda. Hearst became a member of the SLA and assumed the revolutionary name, “Tania.” Not long after becoming Tania, she eagerly and actively participated in serious criminal behavior, including armed bank robbery.Footnote 21 Assume that she was genuinely coercively persuaded, that it was in no way up to her to become Tania, and that the time between that process ending and the criminal conduct was very short.Footnote 22 Tania clearly intentionally acted, intended the criminal conduct, was not threatened with dire consequences if she did not participate, and suffered from no mental infirmity. In short, the prima facie case was airtight and Tania had no affirmative defense of duress or legal insanity. Her rationality was fully intact. Nonetheless, did she deserve some break for her crimes?

If you think Hearst deserves a break, I suggest that the explanation just given for why we might give all kids a break—the need for sufficient time to pass to test one’s reasons against experience in order to be fully responsible—does seem plausible for Hearst. She needed to be in the world as Tania for some amount of time in order to own being Tania and to become fully responsible. We can argue as a normative matter how much time is necessary, much as a polity might put the age of majority at a somewhat lower or higher age than 18. In a real sense, kids are “coercively indoctrinated” by their parents, but they have a great deal of time as they mature in the world to test themselves, including their modes of reason.

Suppose, like me, you are unpersuaded by the “ownership” argument and do not want to give either Hearst or Christopher SimmonsFootnote 23 a break. Is there still a good TRW argument for giving many kids a break and perhaps almost all of them? I think individualization based on rational capacity with a break for underdeveloped rational capacity or diminished rationality would serve well. This was the approach recommended by the dissent in Roper and the majority in Miller. This underdeveloped or diminished capacity doctrine would be accompanied by a strong presumption that it should apply that the prosecution would have to overcome in presenting its case. This presumption should do much to alleviate the thorny problem of not giving a break to those kids who really deserve one. Presumably, the characteristics kids lack should be the same as those we think adults possess. For example, if impulsivity is a reason why kids should be given a break, then similar impulsivity in adults should be mitigating also. But I do not insist on this. Some adherents of TRW believe that there is not a continuum of rationality between youth and adulthood and that adolescence is a unique developmental phase. Nonetheless, that phase does not magically always end at the age of majority and individualization could still be accomplished.

Difficult questions for proponents of individualization are how it is to be accomplished as a practical matter and whether the costs of individuation are worth it if most kids do deserve a break. Testing the relevant variables would be much more difficult than, say, testing IQ to determine if a defendant is intellectually disabled. Commonsense, intuitive judgments are also likely to be unreliable. For reasons like these, when I proposed a generic “diminished responsibility” mitigation based on diminished rationality,Footnote 24 I suggested that the diminishment had to be quite substantial. Also, these variables are likely to be arrayed along a continuum at most ages, but I believe we lack the tools to individuate precisely. Therefore, I also suggested that there be a single class of those deserving mitigation. In the case of kids, that would presumably be most kids. I am open to various schemes concerning how much of a break should be given. It could be uniform or there could be a sliding scale depending on age because older kids usually have more of the “right stuff.” Because the criminal justice system is always balancing individual culpability and public protection, another possibility is an inverse sliding scale, with the size of the break dependent on the seriousness of the crime.

Individualization is time-consuming and expensive. Is it worth it for kids, most of whom will probably deserve a break? I think it is because I believe some kids are fully responsible. Christopher Simmons was presumably in this category. Nevertheless, I would limit individualization to only the most serious crimes that involve danger to persons, such as homicide, aggravated assault, serious sex crimes, robbery, and perhaps burglary of homes. These are such morally wrongful offenses that even kids have the strongest possible reasons, including legal reasons given the penalties for these crimes, not to offend. If such a kid is fully responsible, the kid should get the full condign sentence. Judges and juries may be hesitant to conclude that a particular juvenile defendant accused of serious crime deserves a break, but the strong presumption of giving one should guide the inquiry and limit the number of cases in which kids who deserve break are wrongly denied one. This would be especially true for younger kids. All kids who commit less serious crimes should be given a break. I include theft in this category because it is clearly wrong to steal, but the threat of personal injury makes crimes that do so far more heinous and the need for public safety protection is less powerful.

6 Conclusion

AC is a tour de force that must be read by everyone concerned with criminal culpability in general and with juvenile responsibility in particular. Such a formidable challenge to TRW deserves the widest possible readership. Even if AC fails to persuade, its arguments must be questioned and answered by those who want to offer a principled defense of giving kids a break.