Scholarly life, at its height, is about dialog and debate. When an author publishes a work, they are inviting others to join a wider conversation to find faulty premises, resolve gaps in reasoning, push for conceptual clarity, and extend general understanding. I’m grateful to the editors of Law & Social Inquiry for not only publishing my essay, but providing a space to have an exchange around its core claims. I also want to thank Elizabeth Scott, Laurence Steinberg, David Tanenhaus, and James Backstrom for their thoughtful replies to my essay, and for their prior work and advocacy from which my essay drew.
At this point, there are three core issues that appear to be in contention. The first concerns how damning, exactly, the critique I’ve offered of the developmental approach is. The second asks about the relationship between the developmental approach and the vision of paternalism I offer in the final pages of my essay. The third is about the history, contemporary viability, and future consequences of that view. While full answers to these questions will surely require a broader analysis than I can provide here, I’ll offer a brief sketch, devoting most of my attention to the first issue.
To start, and to underscore a comment made by David Tanenhaus, it’s hard to overstate the foresight and the accomplishments of activists and scholars ahead of Roper v. Simmons. My worry, as I note in my article, is that the arguments used in Roper, if extended, risk undermining those achievements. That worry is not simply, as Scott and Steinberg suggest in their response, that age is a poor proxy for development. Nor is it that we’re “in the grip” of a liberal paternalist narrative of childhood. I’m well aware that legal boundaries are often vague and imprecise, but that’s not the thrust of my criticism. I’m questioning the basis on which the boundary should be drawn, and whether that line drawing should be left as an exercise for science. Their response misses this mark, in significant part, because they don’t acknowledge the implicit normative assumptions that scaffold their own view.
In my article, I offer criticism in two related parts. The first forwards the claim that the internal coherence of the view of childhood presented by advocates is overstated, not simply that an age rule (e.g., sixteen years old to drive) might be under- or overinclusive. And the second part, roughly, highlights how even if the developmental approach passes the squint test, it doesn’t uniformly suggest leniency for children who kill. Multiple accounts of what it means to be a child can be anchored to the same set of empirical facts.
Whether we’re talking about population-level variation in law-breaking, patterns of youthful offending by crime type (and even within types, such as “violent crime”), demographic features (gender, intelligence), or the influence of welfare state institutions (schools and childcare, among others) on the expression of traits that are understood to mark maturity, to name just a handful of wrinkles, the developmental approach offers crude answers, unhelpful ambiguities, or normatively suspect counsel. Perhaps one can add epicycles to account for these anomalies, as Scott and Steinberg attempt, but the issues I (and others) raise ought to decrease our confidence in its general reliability—even on its own terms.
Another, softer criterion for coherence is whether a given account is consistent with how the same phenomenon—childhood—is treated in other domains of law. Steinberg and Scott believe the developmental approach can meet that standard. I do not. They argue that children’s decision making in other areas of life is not imperiled by a focus on development because, among other factors, one can empirically distinguish between neutral and non-neutral decision-making settings. We can, therefore, rely on the developmental approach to respect children’s choices in other domains—access to contraception, medical care, and so on.
There are a number of issues with this distinction, but I’ll stress two. To start, large amounts of law-breaking can be persuasively described as being conceived, planned, or even executed in so-called neutral cognitive settings. For another, outside the laboratory, it’s not clear where one might draw the line around neutrality in real-world decision-making contexts. Choices about sex, consent, and reproduction, in particular, seem like poor candidates for this kind of crude distinction. As I see it, the result is a potential double bind for juvenile justice and for children’s rights more generally. On the one hand, if we can distinguish between neutral and non-neutral contexts, and only the latter counsels mitigation (as with duress, involuntary intoxication, and so on), the case for leniency erodes for a potentially large group of child law-breakers. On the other, if we can’t distinguish between those contexts for children in practice, just as we can’t (as Scott and Steinberg as well as the Supreme Court claim) distinguish between adolescent-transient and adolescent-persistent offending, children’s choices in other domains are left without substantive protection.
Perhaps this dilemma can be sidestepped; maybe there is yet another, more granular distinction that can provide a more consistent account of childhood and responsibility. As I point out in my article, however, this still doesn’t offer much solace. Even if one can make the account of childhood in the developmental approach internally coherent, or—with a lower bar—coherent enough, the data is consistent with multiple kinds of stories of what it means to come of age, each of which has different consequences for the case for leniency. This observation is what I refer to in my article as the “narrative dependency” of the developmental approach.
To be clear, I don’t have an answer about which story we should ultimately pick. And the developmental approach, I’ll admit, certainly does better than a number of its predecessors. However, my aim is not to offer a first-order alternative to the question of how, precisely, we should go about drawing lines. This, I suspect, is the heart of the confusion. On my account, a legal regime might tell a story about childhood that is rooted in hard facts or in useful fictions. Perhaps a blend of both. Whatever the case, if a polity accepts a particular account that curtails the rights of children, that choice comes with a debt that society has a moral obligation to pay. Or so I argue.
It is not an entailment of my view, as Scott and Steinberg imply, that extending a particular right to young children (say, to drive) necessarily entails imposing an obligation of full citizenship (registering for military service). I don’t have something so mechanical in mind. 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. Of course, some restrictions might still meet that higher standard. As I suggest in the final pages of my article, putting democratic obligations into view puts scholars, policy makers, and the courts in a better position to make more fine-grained judgments wisely.
The central omission in my argument, Scott and Steinberg contend, is a substantive account of why societies decide to give children a break in the first place. Children have a special status because they are developmentally immature—the justification for disparate treatment, in the final analysis, is derived from the biological fact of immaturity. The recent past, however, seems to counsel otherwise. David Tanenhaus rightly notes that the idea that “children are different” was not the prevailing view as recently as the 1990’s. Views of childhood change from one historical period to the next sometimes because of, sometimes in spite of, and most of the time unrelated to, new scientific knowledge.
Even when all parties accept the same facts, as I show through the toy example of “child liberationism” (which, to be clear, is just one extreme in a spectrum of possibilities), there are multiple ways of giving those otherwise neutral observations about our natural world practical significance. To justify any one particular narrative moves us outside the realm of more straightforward scientific hypothesis testing into more complex normative questions. A technocratic discussion of human biology won’t suffice. By sussing out the assumptions in the stories about childhood that are accepted in our moment, we can call them out for what they are: implicit political choices, not natural imperatives. And as choices, as human decisions, they invite a discussion of collective obligation, paternalism, and democracy—a conversation, I’ve argued, confounded by the developmental approach.
Each response essay, in a different voice, asks about the origins and consequences of the view of paternalism I forward in my essay. I agree that my view of the normative demands of paternalism does not, in principle, conflict with the developmental approach. One might gloss over the criticisms of development that I offer in the body of the essay, but accept the bargaining argument presented in the final pages as yet another justification for leniency for children who kill. While possible, this position is a bit awkward. It begs the question of why we would seek to extend an increasingly shaky approach.
Tanenhaus’s larger point in his response is well-taken; the view of paternalism I offer has its roots in a tradition at least as old as the 1960s, perhaps far older,Footnote 1 I’d add, and wrestling with that intellectual heritage would likely provide for a more convincing alternative. Here, I’ll offer a truthful, but surely unsatisfying reply: I agree, and I intend to take up this task in future work. Similarly, he points our attention to the social and legal history that foregrounded Miller, and notes that I gesture toward this history, but do not give a sufficient accounting of the rise of the developmental approach. Fair enough. There is a lot to learn from even the thumbnail sketch Tanenhaus offers in his response, and his earlier work is also illuminating for understanding the origins of modern juvenile courts. In offering a sympathetic account of the appeal of the developmental approach, my eyes were trained more on its conceptual and jurisprudential appeal and less on the historical conditions that facilitated its rise, and the tools I brought to that discussion were largely drawn from political theory. To my ears, Tanenhaus’s reply is more of a call for methodological pluralism and interdisciplinary inquiry, and less of a direct criticism. This symposium, which includes a legal historian, a theorist, a neuroscientist, and a practitioner, in effect, begins to answer that call.
Interdisciplinarity, odd alliances, and deep disagreements are all on display in James Backstrom’s response. The ability to distinguish right from wrong, he notes, is different from weighing the risks of a given action. When the former is present, as it is for most adolescents, children should be eligible for punishment. The rub, of course, is what constitutes “appropriate accountability,” to use Backstrom’s phrase, and who decides what is appropriate. The case for leniency, we agree, does not simply rest on the capacities of children. We disagree, however, about the place of prosecutorial discretion in making that assessment. Perhaps if the institutional incentives prosecutors face were properly aligned with consensus community goals, and if publicity in charging, bargaining, and trial decisions were ensured, a stronger case could be made for entrusting decisions about the status of children to prosecutors’ offices (Pfaff Reference Pfaff2017). Decisions about the appropriate treatment of children in the criminal law should not remain on such precarious footing. On this point, Scott, Steinberg, and Tanenhaus, I suspect, would agree.
I’ll leave it to the audience of this symposium to assess the strength of the reasoning in my article, the responses of my critics, and my short reply here. In deciding whether to accept a given idea, we are always looking to what, independent of us, might make our supporting reasons good ones. That, too, is typically in contention, and the conversation then goes another round. I, for one, look forward to continuing the debate.
