“Children are different” (Miller v. Alabama, 132 S. Ct. 2455, 2470 (2012)). Few beliefs are as strongly held. “Any parent knows” children are less culpable and that they are less well suited to a justice system designed by and for adults (Roper v. Simmons, 125 S. Ct. 1183, 1195 (2004); Miller, 132 S. Ct. at 2464). The claim is certainly as old as the first juvenile court,Footnote 1 but the intuition is far older. Because children cannot consent, for example, contracts with children were unenforceable at common law (Brewer Reference Brewer2012). At the extremes, we are told, there is little scientific or social controversy. While the boundaries between child, adolescent, and adult are difficult to define, there is a consensus that adolescents and adults are different in kind. The juvenile justice system, at least in its ideal form, is designed to treat children as the partial- or semi-citizens that they are (Cohen Reference Cohen2009, 189f).
Extreme acts of violence put pressure on this consensus. Children that kill are exceptional; they, in many people’s minds, create a kind of border problem for juvenile justice. Law and order politicians push for “adult time” for “adult crime.” The proliferation of juvenile transfer laws in the 1990s suggests the salience of this approach, and a late recognition by legislatures of the destructive consequences of these laws has ultimately lead to their retrenchment.
Insofar as the law is concerned with proportional punishment, then, it seems to walk a tightrope. The status of “child” motivates some of the law’s most pressing proscriptions, and hence may justify the use of force.Footnote 2 At the same time, as many reform advocates have insisted, that status firmly limits both the severity and kinds of punishment available to legal institutions. That public opinion tends to align with the claim that children should be given a breakFootnote 3 belies a deeper set of confusions. It matters why. On what grounds should a seventeen-year-old that kills be treated more leniently than his eighteen-year-old counterpart?
The Roberts Court has wrestled with defining the significance of childhood under the Eighth Amendment in a series of important juvenile justice decisions—most notably, Miller v. Alabama.Footnote 4 Part of the answer, for the Court majority, is to ground its doctrine in a “developmental approach.” Briefly, advocates of this approach argue that adolescents form a relatively well-defined group, whose development follows a roughly predictable course to maturity, and whose criminal choices are affected predictably in ways that mitigate culpability. The force of a developmental foundation for juvenile justice jurisprudence, advocates of the approach argue, has been to slowly remove from discussion the question of whether adolescents and adults are, in fact, different (Scott and Steinberg Reference Scott and Steinberg2008).
This consensus, however, papers over slowly accumulating anomalies and contradictions that threaten the general viability of the approach in practice. A number of law and society scholars have noted that this line of reasoning puts pressure on the Court’s jurisprudence in other areas of the law—ranging from contracts to access to abortion and general medical care (Ward Reference Ward2006, 435; Zimring Reference Zimring2005). On the one hand, sympathetic critics affirm the relevance of new insights from developmental psychology, but counsel against an approach to the criminal law that treats children’s capacities at certain ages as ascertainable and fixed (Buss Reference Buss2009a, Reference Buss2009b). On the other, harsher skeptics have gone as far as to suggest that childhood is too varied and too internally complex a category to support substantive practical conclusions around “development” at all.Footnote 5 At best, developmental arguments should be relegated to a buttressing role in future lawmaking (Maroney Reference Maroney2009).
One could simply resign oneself to a hodgepodge of inconsistent doctrine. The burden therefore lies with the juvenile justice reformer to explain how a principled concern with proportional punishment simultaneously justifies and limits the legal response to children that kill. This article does not attempt to meet this burden directly, but it does suggest why the most common approach to the problem, the approach used and extended by the Court, is philosophically confused.
The Court mistakes socially contingent but pervasive features of childhood for the way things must be. In so doing, I suggest, it reifies politically generated dynamics that socio-legal scholars ought to question. To make headway in explaining the peculiar place of adolescents in the criminal law, the empirical and normative claims of the developmental approach must be pried apart. Ultimately, I argue advocacy both in and out of the courts needs to move beyond Miller’s shaky rationale. While developmental arguments have helped secure differential treatment for adolescents in the courts, those same arguments, if extended, risk undermining that accomplishment.Footnote 6
I begin by describing how the court has come to embrace the developmental approach, along with its general appeal. Next, I explain why that view suffers from a series of important internal inconsistencies, omissions, and ambiguities that ought to undermine its allure. More damning, I continue, is how it sets the terms of the discussion. It not only obscures the underlying normative debate, but it directs our attention to a largely unhelpful set of questions. To indicate how one might escape these shortcomings, I then sketch an alternative account that may be able to address the problem without sacrificing the category of “juvenile” itself. To punish children as adults, I conclude, is an attempt to reap the benefits of paternalism without bearing the accompanying political, social, and moral costs.
MILLER v. ALABAMA!
The line of argument I aim to unsettle flows from an assumption that maturity refers to an intrinsic set of abilities that are given independently of the legal regime. I will refer to this assumption as the “natural fact” view of maturity. Before poking and prodding at that assumption, a few words on the Miller decision itself.
In Miller v. Alabama, two fourteen-year-olds convicted of murder challenged the imposition of mandatory life without parole (LWOP) sentences (132 S. Ct. at 2470).Footnote 7 While it did not categorically ban these sentences for juveniles, the Court did hold that mandatory LWOP for those under the age of eighteen at the time of their crimes violates the Eighth Amendment prohibition on cruel and unusual punishment. The Court’s expectation that the sentence should be “uncommon” may create a self-fulfilling prophesy (Miller, 132 S. Ct. at 2469 (Roberts, dissenting)). This decision, extended last year in Montgomery v. Louisiana, relied on a line of reasoning from Roper v. Simmons (in which the Court rejected the imposition of the death penalty for a crime committed by a juvenileFootnote 8) to Graham v. Florida (holding that no minor could be sentenced to LWOP for a non-homicide offense) that juveniles are different from adults and that those differences make a difference for calculations of equity and proportionality in the justice system.
The Court majority in Miller, led by Justice Kagan, relied on two general lines of precedent. The first, the strand that has received the lion’s share of academic commentary, concerns the relationship between LWOP sentences and the death penalty. Graham, from two years prior, lowered the bar for categorical rules against excessive punishment based on the nature of the offense or the kind of offender. While Graham imported the proscription against disproportionate punishment from the Court’s death penalty jurisprudence into its non-capital jurisprudence, Miller reached much farther in applying that proscription. Unlike Thompson, Roper, or, arguably, Graham, the majority in Miller largely sidelined the Court’s traditional confused “evolving standards of decency” Eighth Amendment analysis (Miller, 132 S. Ct. at 2455 (Thomas and Scalia, dissenting)). “Death-is-different” moved to a more general limitation on excessive sentences—at least for juveniles.Footnote 9
The second strand is a categorical ban on sentencing practices based on a mismatch between the culpability of a class of offenders and the seriousness of the penalty (ibid., 2463). For example, children, like the cognitively disabled, are not eligible for the death penalty.Footnote 10 It is this line of analysis that I would like to examine in more detail.
Drawing from Roper, Justice Kagan presented three arguments for treating children as having diminished culpability. The first is that children have a “‘lack of maturity and an underdeveloped sense of responsibility’” that leads to risk-seeking behavior (ibid., 2464). Second, children are more vulnerable to “‘negative influences and outside pressures’” and have “limited ‘contro[l] over their own environment’” (ibid.). And the third, “a child’s character is not as ‘well formed’ as an adult’s; his traits are ‘less fixed’ and his actions are less likely to be ‘evidence of irretrievabl[e] deprav[ity]’” (ibid.).Footnote 11
These arguments ultimately underpin the Court’s proportionality analysis; finding that children are less blameworthy,Footnote 12 children are less able to be deterred,Footnote 13 and that rehabilitation cannot justify a life without parole sentence because “incorrigibility is inconsistent with youth” (ibid., 2465, citing Graham, 130 S. Ct. at 2029). Citing JDB v. North Carolina 131 S. Ct. 2394 (2011), the majority adds that the “the features that distinguish juveniles from adults also put them at a significant disadvantage in criminal proceedings.”
A vision of a “mitigation continuum” informs their analysis—the idea that a “justice regime grounded in mitigation corresponds to the developmental reality of adolescence and thus is compatible with the law’s commitment to allocating punishment fairly on the basis of blameworthiness” (Steinberg and Scott Reference Steinberg and Scott2003). The incapacities of a young child, for example, are functionally equivalent to the incapacity of mental illness. If they are a danger to society, civil commitment is more appropriate than criminal prosecution. While young children are simply different in kind from adults, adolescents are in a period of transition—not quite a child, not yet an adult.Footnote 14
Taken as a whole, the lesson the Court takes from the emerging body of doctrine is that in “imposing a State’s harshest penalties, a sentencer misses too much if he treats every child as an adult” (Miller, 132 S. Ct. at 2468). If “‘death is different,’ children are different too” (ibid., 2470). As a consequence, the Court concludes, the Eighth Amendment forbids a sentencing scheme that mandates life in prison without the possibility of parole for juvenile offenders. Miller, on some accounts, has turned the idea that “children are different” into a wider constitutional principle (Scott Reference Scott2013).
The majority not only grounds evolving juvenile justice doctrine in intuition and legal reasoning, but also science: “Our decisions rested not only on common sense… but on science and social science as well” (Miller, 132 S. Ct. at 2464). The Miller majority contends that the scientific consensus has only gotten stronger since Roper. Whether or not one takes Justice Kagan at her word about following the science (and there are strong reasons to be skepticalFootnote 15), socio-legal scholarship, child advocates, and juvenile justice reformers have largely embraced the “developmental approach” of the Court.Footnote 16
THE DEVELOPMENTAL APPROACH
The larger historical development of juvenile law has left the link between immaturity and culpability largely ignored, leaving a conceptual and doctrinal void (Scott and Steinberg Reference Scott and Steinberg2008, 119). A core complication, Franklin Zimring notes, is that separate treatment of children seems intuitively right in a way that does not invite further scrutiny by advocates (Zimring Reference Zimring2005, 56). The lesser maximum punishments of serious crime “can be seen as testimony to the belief in youthful diminished culpability, but this set of practices is at best mute testimony, lacking any statement of principles that can be analyzed and criticized” (ibid., 57). Development, for many, provides the backbone of a more principled account of penal proportionality for juveniles.
Note that existing doctrine around “excuse” and “justification” is, in theory, capacious enough to include the three factors (decision-making capacity, susceptibility to external circumstances, unfixed character) described by the Court for all defendants, regardless of age (Yaffe Reference Yaffe2018). Advocates, however, resist this straightforward conclusion. For Laurence Steinberg and Elizabeth Scott, systematic differences in the culpability of adolescents and adults are evidence that adolescents should be treated as categorically different. Unlike adults, they argue, mitigating conditions for adolescents are “characteristic of a relatively well-defined group, whose development follows a roughly predictable course to maturity and whose criminal choices are affected predictably in ways that are mitigating of culpability” (Scott and Steinberg Reference Scott and Steinberg2008, 139). An age category is justified by the fact that it applies to most (though not all) members of the group, and that it creates substantial institutional efficiencies. Gideon Yaffe helpfully describes this line of argument as having an empirical dependency. That is, “it provides support for the policy of giving kids a break only if competing, implementable policies do not offer better mixes of false- and true positives and negatives” (Yaffe Reference Yaffe2018, 30). I would add that this argument also has a narrative dependency. Which facts about children are taken to be significant, as well as how those facts relate to each other, are shaped by a wider story about what it means to “grow up” in a given society. These two dependencies, as I detail in the sections that follow, put the argument for leniency on a troubled foundation.
EMPIRICAL LIMITS
Setting aside general concerns about the Court’s use of scientific evidence,Footnote 17 the “developmental approach” articulated in Roper, Graham, and Miller ought to leave us uneasy, even on its own terms. If the approach fails as empirical description, it will be yet another blemish on its viability as legal prescription.
Part of the intuitive appeal of developmental model is that the period of life marked by fifteen to eighteen years-old does not fit cleanly into the categories of “child” or “adult.” The creation of an intermediate category—adolescence—accords with the widely shared intuition that culpability should increase in step with maturity. The next step in the analysis, however, creates controversy; to treat “child,” “adolescent,” and “adult” as different kinds of legal beings. This seems to be the bud nurtured by the Court since Roper and comes into full bloom in Miller and Montgomery. As noted above, advocates justify this approach by describing adolescents as a “relatively well-defined group,” whose development follows a “roughly predictable course.” However, neither claim is obviously correct. Nor is it clear that the Court’s reasoning always counsels leniency. Consider the following issues.
For one, a small but significant point: the data on child development reviewed by the Court actually suggests that the immaturities associated with adolescence extend well beyond eighteen to one’s mid-twenties (Leo and Drizin Reference Leo and Drizin2004; Moffitt Reference Moffitt, Lahey, Moffitt and Caspi2003; Steinberg Reference Steinberg2013; Steinberg and Scott Reference Steinberg and Scott2003). And the argument offered by the Court that individuals at eighteen are able to move from their “situation” (Roper, 125 S. Ct. at 1195), and thus become responsible for their context, would ring hollow to any urban sociologist.Footnote 18 All the reasons articulated by Kennedy that “adult” punishments for fifteen-year-olds are disproportionate apply with equal force to punishments for early twenty-somethings. Likewise, it is not clear that children under fifteen entirely lack the capacity to be found criminally culpable, as Steinberg and Scott claim (Ward Reference Ward2006). Many advocates of the developmental approach recognize these ambiguities, but do not appreciate the general significance of these shortcomings. A “nuanced” view of adolescence comes in stark contrast to a decidedly not nuanced view of both adulthood and childhood. The different treatment appears morally arbitrary. It is unclear that arguments about efficiency or workability provide a strong enough retort to this shortcoming.Footnote 19
That said, it is not clear how nuanced the developmental model is, exactly. If one disaggregates offending rates by gender, for example, one finds substantially different patterns by age group (Hockenberry and Puzzanchera Reference Hockenberry and Puzzanchera2017, 12). This has led some to embrace “girls’ courts” (Heipt Reference Heipt2014). More generally, age is only a rough proxy for a series of as-of-yet unspecified chemical and biological influences (cortisol, hormones, gene expression, neurological change, among others). The Court is treating scientific evidence as if it is complete or fixed in time. If one is concerned with reducing false negatives (those under eighteen years-old that, in fact, meet the criteria of maturity described by the Court) development seems to push toward finer distinctions within the category “adolescent.” We are left without a principled reason to limit developmental reasoning to age, simpliciter. As our biological knowledge improves, age will likely become an increasingly poor proxy for the underlying phenomena it is intended to represent.
Compounding the problem, the claim by Steinberg that the “age-crime curve” supports the use of age as a proxy is also a bit short-sighted (Steinberg Reference Steinberg2013). While trends in juvenile violent crime may roughly map onto the Court’s current doctrine, it is also true that the peak age for homicide has shifted in different time periods within the United States and also varies among industrialized nations (Ulmer and Steffensmeier Reference Ulmer, Steffensmeier, Beaver, Barnes and Boutwell2014). This is not to deny the considerable evidence in favor of the general “dual systems” model of adolescent risk-taking that describes a period in the life-course characterized by “heightened sensation seeking” and “still-maturing self-regulation.” While the general pattern described by this model (an inverted U-shape in risk-seeking, a linear increase in self-regulation) seems to hold across most, but not all countries, adolescent risk-taking manifests itself to different degrees and forms around the globe.Footnote 20 Steinberg and colleagues write, “the broader context in which adolescents develop exerts a powerful impact on the extent to which young people engage in risky and health compromising behavior” (Steinberg et al. Reference Steinberg2017).
That observation is a small example of a larger point: the alignment of empirical description and moral intuition is historically contingent, not essential or inevitable. Most worrisome perhaps, is where the logic of development seems to violate widely shared moral intuitions about children’s desert. Most would bristle at the conclusion that young people who develop faster—e.g., girls, or precocious kids—ought to be treated more harshly by the criminal law than their “slower” counterparts.Footnote 21
In addition to concerns that the Court’s developmental approach is, in various ways, descriptively inaccurate, there are at least two wider issues that stem from a common domain problem. For one, it is not clear that developmental psychology necessitates greater leniency for all youth. While the Court has yet to embrace the distinction, the evidence provided by Moffitt (Reference Moffitt, Lahey, Moffitt and Caspi2003) concerning life-course-persistent and adolescent-transient offending provides an opportunity for criminal courts to ratchet up penalties for a subset of young lawbreakers. The majority in Miller notes,
appropriate occasions for sentencing juveniles to this harshest possible penalty [life without the possibility of parole] will be uncommon. That is especially so because of the great difficulty we noted in Roper and Graham of distinguishing at this early age between the juvenile offender whose crime reflects unfortunate yet transient immaturity, and the rare juvenile offender whose crime reflects irreparable corruption (Miller, 132 S. Ct. at 2469).
If the science improves (we are led to believe), there is nothing stopping the Court from revisiting the point. More generally, there is nothing about the developmental approach that is intrinsically tied to leniency; one can imagine a Roper-like decision that uses similar reasoning to argue for aggravation, instead of mitigation, for a subset of young offenders. (To stop, for instance, what John DiIulio notoriously referred to as “super-predators” in the 1990s.) This is particularly true for the third feature of juveniles mentioned by the Court since Roper: that adolescents’ characters are “not fixed.”
For another, as Justice Scalia’s dissent in Roper underscores, the developmental model has implications for child rights generally—raising troubling questions about children’s rights to make medical decisions or decisions about sexual health, among a host of others. In 1990, the American Psychological Association filed a brief in Hodgson v. Minnesota, 110 S. Ct. 2926 (1990), arguing that a “rich body of research” showed juveniles are mature enough to decide whether to obtain an abortion without parental involvement. While Steinberg and Scott (Reference Steinberg and Scott2003) argue that the science has moved forward since then, it is unclear what the developmental approach means for, among other things, abortion rights (Ward Reference Ward2006). Others have wondered about the significance of the developmental model for the mature minor doctrine which, as one contributor to the pages of this journal notes, is a gateway for transgender youth to access identity-affirming treatment in the face of dissenting, or even abusive parents (Carroll Reference Carroll2009). By treating adolescents as a different kind of Constitutional subject, a variety of domains concerning the choices of young people are immediately called into question. It is unclear why the analysis in Roper, Graham, and Miller should be cabined to the criminal justice context.
None of these critiques are damning, but they are suggestive. What they show, at minimum, is that the “foundation” provided by developmental approach is less sturdy, and less coherent, than advocates often claim. Crucially, the success of developmental arguments depends on how closely the Court’s doctrine tracks current scientific research, a potentially precarious position. Compounding matters, while neuroscientific claims are thought to have particularly strong rhetorical force in policy discussions about children, the evidence about their practical effectiveness in the courtroom is at best mixed (Maroney Reference Maroney2009, 170–71).
Few theories, of course, escape anomalies, contradictions, and inefficiencies in practice. One could argue, as some advocates of the developmental view have done, that a number of the worries described above are overdrawn (Scott & Steinberg Reference Scott and Steinberg2008). The tension between the brute facts of socio-biological development and doctrine, however, is only one source of ambiguity. Even if jurisprudence, scientific evidence, and moral intuitions all happen to align, “childhood” is also a source of narrative conflict. A deeper concern, as I describe below, is that the developmental approach takes for granted a particular story of what it means to come of age (Silva Reference Silva2012).
NARRATIVES OF DEVELOPMENT
Advocates of the developmental approach tend to hold a “natural fact” view of maturity. That is, they view competence as an intrinsic feature of personhood that conditions how we should act with respect to a given individual. For those concerned with development, mistreatment is a kind of mis-assessment. A court system that sentences a fifteen-year-old who kills as if she had the maturity of an adult acts unjustly, as would a court that sentences her as if she had the competence of a child in primary school. And what is being assessed (at least in theory) is not agreement with a particular political settlement, but agreement with a scientific one.
While the Miller Court did apply the developmental approach to utilitarian proportionality arguments (“children are less able to be deterred”), the reasoning is more retributive than utilitarian. The focus shifts away from any actual harms suffered (or inflicted) by those punished and onto questions of whether the legal system has failed to correctly interpret the underlying nature of the defendant.
Maturity, however, is not so straightforward. The brute fact that someone is less intelligent, for example, or that another is more susceptible to the influence of peers is not, relevant to that assessment. What is required is an account of how otherwise neutral features of our world come to have practical significance. And it is the conflict among these accounts, I suggest, that is obscured by the developmental approach. The approach has, as I noted earlier, a “narrative” dependency in addition to an empirical one.Footnote 22
Consider the liberal paternal view, perhaps the most influential narrative about the practical significance of childhood. In this account childhood constitutes a special status, and that status has a dual character. On the one hand, children need protection; they are dependent on others for their basic care, and they need time to experience, learn, and play without facing serious consequences for their choices. On the other, this means children live under a form of guardianship where their voices and actions have only consultative value. As children develop they are understood to have greater say in their government.Footnote 23 Andrew von Hirsch, for instance, emphasizes the semi-autonomy of children. He writes that the expectation that children “try out” making their own decisions counsels leniency for mistakes.Footnote 24
It is this LockeanFootnote 25 move that gives rise to the assumption I wish to examine here. When public institutions recognize competence appropriately, it tells them how they should or should not treat those who possess it. Just action becomes a matter of assessing a person’s competence, interpreting their authority as a speaker (or actor) relative to that assessment, then responding accordingly. A straight line is drawn from maturity to just or unjust political arrangements.
But there are other accounts available. To dramatize the point, consider an extreme alternative narrative of the significance of development: child liberation.Footnote 26 My intention is not to map all variants of the argument, just what I take to be the most significant.Footnote 27 Here, the basic claim is that the line between adult and child is overdrawn, and everyone—regardless of age—is entitled to a full schedule of liberty rights. One version of the argument proceeds as follows. There are many ways in which the capabilities of children are conceived, and those ways are intimately tied to historical circumstance. While biological immaturity is a universal and natural feature of human groups, ways of understanding this period of human life—the institution of childhood—varies through time and across cultures (James and Prout Reference James and Prout1990, 3). In his influential book, Centuries of Childhood, historian Philippe Ariès argues that the concept of childhood as a distinct period of human life did not exist in the fifteenth century. Likewise, “adolescence” and “youth” have a relatively recent history.Footnote 28 These gradations are not necessarily reflections of increasing humanity or understanding. “Child,” “adolescent,” and “youth” are useful categories to particular professions and to particular interests, and they serve as sites for the production of society-specific meanings.Footnote 29 The contemporary disqualification of children from wider civic participation is not a necessary feature of liberalism, only a contingent one.
The most extreme form of the liberationist view is that there is no reason to presume children are incapable, as the liberal paternal view does. While capabilities do develop and vary, the presumption of incompetence does not respect how capabilities work. Taking one strand of this claim, authors argue that the emphasis on competence ignores that incapacity is a social and relational phenomenon. Recalling John Stuart Mill on the subjection of women, for example, one could argue that the features understood to disqualify children—dependence, frivolousness—are a product of a particular form of social prejudice, not intrinsic features of youth or childhood. If we now accept that women are not, to pull an argument from the early twentieth century, “child-like” and best suited to the private sphere, perhaps children themselves are not quite as childlike as commonly held. Furthermore, in various aspects of our adult lives we experience incompetence. We confront financial decisions, medical decisions, and legal decisions, for instance, for which we are ill-equipped. And what do we do? We enlist the aid of accountants, doctors, and lawyers. We seek out fiduciary relationships. Likewise, holders of the liberationist view argue children can “borrow” the relevant capacities needed for participation should they, in fact, lack them (Cohen Reference Cohen1980, 56–61). It is less clear that this view implies mitigation for children as a class.
In this approach, mistreatment is less about a mis-assessment of an independent truth and more about evaluations of accessibility or social autonomy.Footnote 30 Feld’s advocacy for the abolition of the juvenile court, perhaps, could be read in this vein (Feld Reference Feld1997). Instead of drawing a line from maturity to blameworthiness, the relationship is flipped. ‘Maturity’ is the outcome of how a person is treated. It is determined by the ways in which others value and engage with young people.Footnote 31
Like the liberal paternal view, the questions asked by the child liberationist view are not primarily utilitarian. In this case they shift attention to whether a community, or a legal regime, has created the circumstances necessary for a lawbreaker to be treated as mature. While one account draws a straight line from biology to politics, the other offers an (admittedly heroic) account of the ability of collective organization to shape or overcome human differences. Each, importantly, can be built from the same natural facts. The liberationist account casts in relief what the liberal paternal view leaves in shadow. Underneath ascriptions of maturity is not simply a scientific consensus, but also a narrative one.
A neuroscientific foundation for juvenile justice jurisprudence, advocates of the developmental approach argue, has slowly removed from discussion the question of whether adolescents and adults are actually different. If true, the victory might prove pyrrhic. Privileging one narrative of growing up over the other either elides the very real needs of children or obscures the complex social ecology that shapes the capabilities of children.
One response to this apparent dilemma is to take Goldilocks’s counsel—find the pleasant porridge of middle ground. Whereas the developmental approach treats proportionality as a required response to a preexisting natural fact, one might opt for the more moderate stance that maturity is the result, at least in part, of wider institutional and social practices. While we might have a better account of what maturity is after such a diagnosis, a side effect is that the age-category child (or juvenile, or adolescent) is either abandoned or stretched beyond practicable limits.
Worse, we are no closer to a satisfactory account of why children as a general class should be treated more leniently than their eighteen-year-old counterparts. Even if we have a shared narrative about childhood, grounded in uncontroversial facts, the developmental approach is only of limited use. The reason is relatively straightforward. “Child” is not simply a descriptive category, it is also a normative one. Advocates mistake socially contingent, but widespread features of our political order for the way things must be. Public choices concerning the political status of children are understood as natural, inevitable, and necessary. As I detail below, the developmental approach takes for granted a political theory of democracy that socio-legal scholars ought to question.
CIVIC PUNISHMENT
The notion that age marks “special status” (Goodin Reference Goodin1988) emerged in criminal law relatively recently. Before the turn of the eighteenth century, age was not particularly significant in criminal court proceedings (Brewer Reference Brewer2012, 181f). With the rise of republican political thought, however, “questions of accountability for crime and of membership in the political community were asked and answered in the same breath” (ibid., 184). Children’s liability for crime, Brewer writes, was a part of that same transformation. Consent, intent, and appropriate punishment became intertwined with questions of democracy and political obligation. It is this connection to political theory, and democratic theory specifically, that the developmental approach neatly sidesteps.Footnote 32
Specialized procedures for child lawbreakers are a particular instance of a more general historical phenomena. The political rights of children are unbundled and, in various ways, are curtailed or channeled.Footnote 33 This extends to membership in the political community itself. Nationality is conferred at birth, without consent, for instance and children are not granted the right to vote until reaching the age of majority. Under the aegis of parens patriae, the state responds to children’s dependence and reshapes it.Footnote 34 The paradigmatic instance of the exercise of this power is mandatory schooling, but there are a variety of other examples. As Elizabeth Scott herself notes, family law itself can be understood as shaping a set of fiduciary responsibilities toward children (Scott and Scott Reference Scott and Scott1995).
This leads to the crux of the matter. Excluding children from core citizenship rights has significant costs for both administrative rationality and democratic legitimacy. Conflating children with their parents, or only addressing the adults that children will eventually become, leads to administrative difficulty in governing children as independent beings (Cohen Reference Cohen2009, 187). And if children have no voice in the laws, it is unclear how can they be legitimately bound by them.Footnote 35
The first juvenile court, founded in Chicago in 1899, provides one possible answer to these twin deficits. A special, remedial identity—“juvenile”—is given to children in the justice system as a compromise (ibid., 189). In this light, mistreatment involves more than misdescription. Children and the democratic polity are inextricably bound together in a relationship of obligation and dependence. What the state owes to child lawbreakers it owes not simply because they are “immature,” but because of the choice the polity has made to disqualify children from a full schedule of democratic rights. Having made this choice, the state is equally obliged—without exception—to every child, whether Evan Miller, Kuntrell Jackson, or Barron Trump.Footnote 36 Instead of asking, as the developmental approach does, on what grounds ought children be treated as different (e.g. biology, access to institutions), we ought to ask what adolescents are owed in virtue of the political consensus that they have a different and special status.
Any answer to this question, surely, will be controversial. Advocates of the developmental approach, perhaps, will see this as a disadvantage. I disagree. For one, the dangers of “penal populism” have been overstated and oversold (Dzur, Loader, and Sparks Reference Dzur, Loader and Sparks2016). An emerging body of scholarship convincingly calls into the question the presumption that the only way to scale back punitiveness is to insulate criminal justice policy from public discussion and community control (Barker Reference Barker2009; Harcourt Reference Harcourt2014; Miller Reference Miller2016; Pfaff Reference Pfaff2017).
For another, non-response, or displacing the question of political obligation entirely, puts children in a precarious political position. Curtailment of participation rights, and paternalism more generally, is justified by the need to protect children from society and the need to protect society from the actions of children.Footnote 37 The latter relationship, as many have noted, creates a kind of “loophole” in assessments of penal proportionality. Children under the age of eighteen may be innocent, but dangerous, and should—some claim—be punished regardless of their blameworthiness.Footnote 38 The developmental approach does not offer much in response.
As I see it, a focus on democratic choice, age, and disqualification brings an otherwise obscured question of fairness into view. To punish children as adults is to reap the benefits of paternalism without bearing the accompanying political, social, and moral costs.Footnote 39 A democratic deficit remains unpaid. The purchase price of paternalism, minimally, is an affirmative obligation to maintain children’s physical and psychological well-being. And while it is beyond the scope of this article, lengthy sentences, solitary confinement, and lack of access to basic educational programming, on their face, appear irreconcilable with this duty.Footnote 40 Likewise, the terms of this bargain readily translate into specialized procedures for juvenile lawbreakers and a “youth discount” on sentences for children like Evan Miller who are convicted of homicide, although I can only gesture toward that argument here.Footnote 41
While this view of paternalism, this view of what children are owed, is a change in emphasis, it is not a radical departure from the Court’s current doctrine. Nor is it incompatible with developmental arguments. One can remain agnostic about the authorization for paternalistic policies—whether it is neuroscience, common sense, or a majority vote—while insisting that there are burdens triggered by the use of those policies. If age is understood not simply as a “gross proxy,” as critiqued above, but as marking the boundary of a democratic pre-commitment to care for young people, it is easier, for instance, to justify drawing a firm line at eighteen or establish the consistency of Roper with the mature minor doctrine. Clearly, quite a few details need to be filled in. One might worry, for instance, that this view of paternalism pits the extension of rights and the case for leniency against each other. By pushing for legal recognition of children’s choices, advocates undermine the more general case for lesser punishment. While correct, this trade-off appears normatively desirable. Insofar as a polity treats a particular class of individuals differently, the obligation to justify and respond to that different treatment ought to be heightened—the converse should also hold.Footnote 42 Putting democratic obligations into view, I suspect, puts scholars, policymakers, and the courts in a better position to make more fine-grained judgments wisely.
More generally, a collective pre-commitment to treat youth differently is particularly important in the context of violent crime. The facts in Miller are both horrifying and tragic. Consider the crime of fourteen-year-old Evan Miller:
One night in 2003, Miller was at home with a friend, Colby Smith, when a neighbor, Cole Cannon, came to make a drug deal with Miller’s mother. The two boys followed Cannon back to his trailer, where all three smoked marijuana and played drinking games. When Cannon passed out, Miller stole his wallet, splitting about $300 with Smith. Miller then tried to put the wallet back in Cannon’s pocket, but Cannon awoke and grabbed Miller by the throat. Smith hit Cannon with a nearby baseball bat, and once released, Miller grabbed the bat and repeatedly struck Cannon with it. Miller placed a sheet over Cannon’s head, told him “I am God, I’ve come to take your life,” and delivered one more blow. The boys retreated to Miller’s trailer, but soon decided to return to Cannon’s to cover up evidence of their crime. Once there, they lit two fires. Cannon eventually died from his injuries and smoke inhalation (Miller, 132 S. Ct. at 2462).
Justice Kagan offers the following background:
No one can doubt that he and Smith committed a vicious murder. But they did it when high on drugs and alcohol consumed with the adult victim. And if ever a pathological background might have contributed to a fourteen-year-old’s commission of a crime, it is here. Miller’s stepfather physically abused him; his alcoholic and drug-addicted mother neglected him; he had been in and out of foster care as a result; and he had tried to kill himself four times, the first when he should have been in kindergarten (ibid., 2469).
Circumstances like these provoke no small amount of cognitive dissonance. And deciding, ex post, how to take into account an offender’s youth requires a heroic commitment to impartiality. Many might agree with the dissenters’ disgust in Miller, seeing the sentence of death or life without parole as the only fitting response to acts that are, at their core, “adult.” Others side with the Court majority, that a proportionate response to violence, even heinous violence, requires considering the socio-biological context of childhood.
Recasting the problem in terms of a polity’s fiduciary obligation—the outcome of a kind of democratic, paternalist bargain—shores up the strength of this pre-commitment in the face of an ambivalent political and social culture on how tough to get with adolescent lawbreakers.Footnote 43 Most importantly, it allows us to say that among the causes the Eighth Amendment ought to be dedicated is democratic legitimacy.Footnote 44 This is not just a matter of assessing whether children are, in fact, different. The criminal law should be thought of as one of the more important institutionalized powers by which popular democratic mandates are given form. Not only is proportionate punishment about bringing the individual to account for her sins, but it is also about reconciling the democratic polity’s ledger. We shouldn’t demand one debt be discharged without requiring the other.
CONCLUSION
Children experience a profound lack of fit between themselves and the wider world. This lack of fit can be literal; the size of furniture in an office, the height of buttons in an elevator. And this lack of fit can be metaphorical; the stress of living in a neighborhood with pervasive gun violence,Footnote 45 the cognitive demands of navigating a court proceeding. The Court’s modern Eighth Amendment jurisprudence, at least in its ideal form, is precisely designed to address this lack of fit–to adjust and to adapt to the needs of children as children.
That the general public tends to agree with the claim that children should be treated more leniently in the courts papers over a wider controversy about why. Intuition, we have seen, can interfere with analysis.
The answer provided by the developmental approach is clear, simple, and misleading. The claim that the criminal choices of adolescents, as a group, are affected predictably in ways that should mitigate culpability is not a straightforward truth. Various and sundry contradictions, empirical and narrative dependencies, along with the displacement of debates over political obligation all undercut its general viability.
This not to deny the biological reality of immaturity, the social science of risk-taking, or the fact of dependence. Nor is to deny that developmental arguments ought to have a role in reasoned law-making. There are a variety of empirical reasons why children ought to be singled out for special treatment—at least under existing social conditions. However, by conflating political choices with empirical description we are left with narrowed and confused grounds to define and assert the rights of children in the criminal law.
