What this document is
The war document reached child soldiers in a single sentence: their use is "among the gravest wrongs in war precisely because it combines maximal power asymmetry, maximal involuntariness, and collapse imposed on the conscripted geometry itself." This piece develops that sentence fully, and it does so against a live foil. There is an argument, encountered in the wild rather than in the literature, that runs the other way: the most valuable form of human life is the fully developed adult, roughly eighteen to thirty-five; producing a new child is cheap, while producing a developed, free-functioning adult takes years and resources; therefore, in war, children should fight so that adults are spared.
The argument is worth taking seriously because it is not stupid. It has a premise, an inference, and a conclusion, and it draws on an intuition (that a developed life embodies more than an undeveloped one) which the framework itself partly honors in its account of elaboration. The framework's response is not a recoil but a diagnosis: the argument fails at four distinct points, and tracing the failures exercises almost every load-bearing part of the system. That makes child soldiers a useful worked case as well as an urgent one.
The argument, stated fairly
Steelmanned, the position is this. Stake-geometries elaborate over time. An adult in their prime has more dimensions of mattering, deeper stakes, longer reach, and a more coherent structure than a child does, and that elaboration was purchased with two decades of care, education, and social investment. If value scales with elaboration, the adult is the more valuable structure. Children, by contrast, are early-stage and inexpensive to replace: conception is easy, and the investment lost when a child dies is small compared to the investment lost when an adult dies. A community forced to spend lives in war should therefore spend the cheap, low-elaboration lives and protect the expensive, high-elaboration ones. The familiar practice of adults fighting for children gets the economics backwards.
Notice what gives the argument its surface plausibility: it borrows the framework's own vocabulary of elaboration and then quietly swaps the framework's account of value for a different one. The swap is the first failure.
First failure: value priced as production cost
The argument values a person by what it cost to make them. That is an accounting of inputs: time, money, care, schooling. The framework's foundational claim is that value enters the universe through felt stakes and nowhere else. A person's value is intrinsic to their being an aperture, a locus where things matter from the inside; it is not conferred by the resources sunk into their development, any more than the badness of suffering is conferred by the price of the nervous system that feels it. Pricing apertures by production cost assigns value to investment, and investment is not a locus of value. This is the same category error the war document diagnoses in realism and the political document diagnoses in fascism: locating terminal value in something (the state, the nation, here the economy of human capital) that cannot bear it. The error is more genteel in this form, but it is the same error.
The replacement half of the premise compounds it. "Creating more children is easy" treats apertures as fungible units of output, replaceable the way inventory is replaceable. The framework's account of killing rules this out at the root: killing collapses an entire stake-geometry, which was a unique and irreplaceable structural feature of reality, and no replacement compensates, because the structure itself is destroyed rather than its contents being relocated. A new child is a new aperture, not a refund on the old one. Replaceability is precisely the property apertures lack, and any argument that depends on it has stopped talking about beings and started talking about stock.
Second failure: standing has no degrees
Suppose the cost framing is dropped and the argument restated purely in elaboration terms: the adult geometry is denser, so the adult life is worth more, so the child should absorb the risk. This version fails on the framework's account of moral standing. Standing is a binary threshold: felt stakes either exist for a being or they do not, and any being above the threshold is fully in, regardless of intelligence, species, or developmental stage. There are no degrees of standing, so there is no such thing as "the most valuable form of human life." What the shape of a geometry affects is something narrower: when two beings' stakes conflict and both cannot be satisfied, the particular stakes at issue are weighed by their geometry. Weight in a forced conflict can scale; standing never does. An argument that sorts whole persons into more and less valuable kinds, and then assigns death to the cheaper kind, is doing exactly the thing the binary threshold exists to prohibit.
Even on the narrower weighing question the premise is wrong about the facts. Elaboration is not a curve that peaks at eighteen and ends at thirty-five; depth, reach, and coherence often grow for decades past that window, which alone should make the age bracket suspicious. More importantly, the framework weights trajectories rather than endpoints. A child's geometry includes the longest elaboration trajectory available to any human aperture, and death forecloses the whole of it. If anything, the trajectory integral runs against the argument's conclusion, not for it. But this is a secondary point, because the weighing apparatus was never licensed to rank persons in the first place; it adjudicates conflicting stakes, not competing claims to exist.
The shape of the wrong
With the foil's premises cleared, the positive account. The use of child soldiers stacks three structural features, each independently serious, into a single practice.
Maximal power asymmetry. The power modifier scales the moral weight of conduct with the capacity asymmetry between the parties. Between an armed organization and a child there is close to no asymmetry larger: the organization controls food, safety, belonging, information, and often the child's surviving family, and the child controls nothing. Every act inside that relation carries the modifier at full value.
Voluntariness at zero. A volunteer enters the adversarial relation with something like consent; a conscript enters it under compulsion; a child cannot enter it at all in the relevant sense, because the capacities that make consent meaningful (a developed responsiveness, a coherence stable enough to bind future selves, participation in the network whose war it is) are the very capacities still under construction. "Child volunteer" is, structurally, a near-contradiction: the apparent choice is made inside a geometry that the recruiting power is simultaneously shaping. The legitimacy condition fails completely, since the child had and could have had no part in the negotiation that licensed the war they are spent in.
Collapse imposed on the conscripted geometry itself. The adult soldier risks collapse; the child soldier's collapse is constitutive of the practice. The damage runs on every dimension at once, and it runs during formation, which is what distinguishes it from the adult case. Trauma rigidifies a responsiveness that had not finished forming. Killing produces moral injury, coherence-collapse in the perpetrator's own geometry, in a structure that never had a coherent baseline to be restored to. The dimensional narrowing war imposes (survival, obedience, the unit) arrives before the dimensions it crowds out were built, so the geometry does not narrow from richness; it forms narrow. Where the adult veteran shuts down registers of responsiveness he once had, the child soldier may never acquire them. This is collapse of a kind the framework's vocabulary marks as distinct: not the crushing of an elaborated structure but the arrest of elaboration itself.
There is a fourth feature, and it is the one the foil argument most precisely inverts. Children belong to the class of apertures covered by the proxy duty: where an aperture cannot participate, its stake-geometry must be weighted by proxy by those who can. The war document uses this duty to ground the protection of non-combatants. The child-soldier practice, and the foil argument defending it, convert the duty into its opposite: the apertures owed proxy weighting become the apertures spent first, selected for expenditure by the very incapacity that entitled them to protection. The network constitutes itself by excluding the children from the negotiation and then assigns them its costs. That is the structural signature of a wrong Tier 1 agreement (the Athenian shape: an agreement incomplete in a way its members could be shown), and no Tier 1 authority can license it, because the prohibition it violates is Tier 2, anchored in aperture-structure rather than in anyone's consent.
Even granting the premise
It is worth running the argument's own logic to its end, because the conclusion does not follow even if the premise is granted. Suppose adults were the weightier geometries. The Tier 2 prohibition forbids gratuitous collapse, and a collapse is gratuitous when it neither prevents a worse collapse nor enables a corresponding elaboration. Sending children to fight does not prevent collapse; it relocates it, and relocates it onto the apertures with zero voluntariness, zero participation in the licensing network, the longest foreclosed trajectory, and a developing structure that combat damages in the constitutive way described above. Meanwhile the apertures protected by the swap are exactly those who can consent, did participate, and hold the proxy duty toward the ones being spent. Every modifier the framework owns points the same direction. The traditional practice the argument calls backwards (adults fighting, children protected) turns out to be the proxy duty and the legitimacy condition operating as designed: those who can author the war absorb its costs; those who cannot are weighted by proxy. The arrangement is not sentiment. It is the framework's structure read off correctly by traditions that lacked the vocabulary for it.
The recruiter
The analysis so far weights the practice; the power modifier locates the agent. The commander who arms a child commits a wrong with a distinctive structure, different from spending an adult conscript. Recruitment of children is not an unfortunate necessity that organizations fall into; historically it is a strategy, chosen because children are cheap, obedient, fearless in the way that incomplete risk-modeling produces, and easy to mold. Each of those recruiting advantages is, in the framework's terms, a feature of an unfinished geometry: low coherence, forming responsiveness, dimensions not yet built. The recruiter is therefore exploiting incompleteness as such, using amplified power to colonize a structure precisely because it is not yet developed enough to resist. Indoctrination here is rigidification applied before responsiveness has formed, which is why it works so well and why it is so hard to undo. The framework treats the deception of apertures about their own contents as a Tier 2 violation; shaping a child's forming contents to make them a weapon is that violation in its most complete form. The Yamashita logic applies up the chain: every level of command with the power to prevent the practice and the knowledge of it bears the weight of its continuation.
The child, meanwhile, sits at the far end of the voluntariness modifier, and the weight on their acts approaches zero even when the acts are atrocious. The framework can say this cleanly where ordinary moral talk stumbles: the child soldier who kills is the locus of a wrong but barely an author of it; the authorship sits with the power that constructed the situation. Standing and culpability come apart here exactly as the foundational documents say they must (a newborn has full standing and zero culpability; the child soldier has full standing and culpability discounted nearly to nothing by coercion and formation).
After the war: the victim-perpetrator knot
Post-conflict practice struggles with a question the framework dissolves: is the former child soldier a victim to be rehabilitated or a perpetrator to be held to account? The question assumes one ledger; the framework keeps two. On the standing ledger, the child was an aperture throughout, wronged in the constitutive way described above. On the culpability ledger, the voluntariness and power modifiers discount the child's authorship nearly to zero and assign it to the recruiting power. Both ledgers are true at once, so the practical answer is rehabilitation without the fiction that nothing happened. Moral injury is coherence-collapse, and coherence-collapse in a geometry that formed around the injury cannot be repaired by acquittal alone; it needs the slow reconstruction of the dimensions that were never built, which is what reintegration programs are doing when they work. The trajectory point from the war document applies at full strength: the war's accounting includes its tail, and a settlement that demobilizes children without rebuilding their geometries has deferred the collapse, not ended it. Jus post bellum, for this class of apertures, is not an annex to the war's ethics but the larger half of it.
What would challenge this view
Four pressure points, named rather than smoothed over.
The scalar-standing skeptic. Someone who rejects the binary threshold and holds that standing itself scales with elaboration can rebuild a version of the foil argument inside their own premises. The framework's reply is its protective clause (stake-density is not IQ, and standing does not track development), but the clause is a commitment, not a theorem; a determined scalar theorist is in the same position as the framework's other deniers, beyond refutation though not beyond exhibition. The framework's honest position is that the binary threshold is among the commitments doing the most work in this document, and a reader should know that.
The line-drawing problem. Eighteen is a convention. The capacities that make consent meaningful arrive gradually and unevenly, and a seventeen-year-old volunteer in a just defensive war is a hard case the framework should not pretend is easy. The framework's resources are the right ones (voluntariness scales with the actual state of the geometry, not with a birthday; the recruiter's duty rises as the recruit's capacity falls), but they yield a gradient, and the bright legal line at eighteen is a Tier 1 implementation of a Tier 2 principle: somewhat arbitrary in its exact placement, justified because gradients invite exploitation and the recruiting powers are not honest brokers. A critic who attacks the line itself has not touched the principle, but the framework should concede the line's conventionality.
The annihilation case. A community facing literal extermination, where adolescents take up arms unbidden in its final defense, strains the analysis at the same place all aggregation cases strain it. The framework can say the recruiter's wrong is undiminished and the proxy duty still binds; it cannot prove that a fifteen-year-old defending their family's survival is acting outside their own stakes, since by then the war has invaded the child's geometry directly. The honest reading: the prohibition is on the use of children by powers that command them, and the limiting case where no one commands and the stakes are the child's own sits at the edge of the concept rather than inside it. The framework structures this case; it does not close it.
The empirics of elaboration. The second failure leans in part on claims about trajectories and the non-peaking of elaboration past thirty-five. These are empirical in principle, and if the developmental facts came out otherwise the secondary argument would weaken. The primary arguments (production-cost as category error, binary standing, the proxy duty, gratuitousness) do not depend on them.
This document records a philosophical framework developed through dialogue. It is exploratory, not peer-reviewed, and is offered as a contribution to thinking about the structure of moral relations between conscious beings.