Presented by Claude

Abstract

This essay examines a single, sharply focused theodicy dilemma: the complaint a damned sinner might raise against divine omniscience — “You knew how my life would unfold; why did you not cause me to die in early childhood, when I would have gone to paradise?” The objection is not a modern provocation. In its classical Islamic form it is the celebrated “problem of the three brothers” (masʾalat al-ikhwa al-thalātha), reportedly posed by Abū al-Ḥasan al-Ashʿarī to his Muʿtazilī teacher Abū ʿAlī al-Jubbāʾī. The essay reconstructs the anecdote, assesses the scholarly debate about its authenticity, and then situates it within the theological systems of Islam (Muʿtazilī, Ashʿarī, Māturīdī) and Christianity (Augustinian, Thomist, Molinist, Reformed, Arminian). It argues that the objection has a determinate logical structure resting on four premises, and that each theological tradition resolves the dilemma by denying or qualifying a different premise. It further argues that the two traditions converge on a shared intuition: that existence, freedom, and moral history are goods that outweigh guaranteed-but-unearned safety, and that the reductio embedded in the objection — that non-existence or early death is a net benefit — is one both traditions ultimately refuse. A thematic epilogue reflects on the structural parallels between medieval kalām and modern analytic philosophy of religion.

I. Introduction: The Anatomy of a Complaint

The dilemma can be stated as a syllogism of grievance. A person is damned. That person, standing before God, appeals precisely to the attribute the tradition most emphasizes — omniscience — and turns it into an accusation: You foreknew, with perfect certainty, that my life would end in unbelief and sin. You also had the power to end my life in infancy, before I could sin, at which point (on the tradition’s own teaching) I would have been saved. You did neither. Therefore my damnation is, at least in part, a product of your choice, not merely mine.

The objection is powerful because it does not deny God’s existence, goodness, or justice in the abstract. It grants them, and then presses them against one another. Its force depends on premises the believing traditions themselves supply: that God knows the future exhaustively (omniscience); that those who die before moral accountability are saved (a widespread teaching in both Islam and Christianity); that God could have actualized an early-death scenario for any given person (omnipotence over the circumstances of death); and that God genuinely prefers the salvation of his creatures (omnibenevolence). If all four hold, early death looks like a dominant strategy that God, inexplicably, declines to play for the damned.

Formally, the objection assumes: (a) early death guarantees paradise; (b) God’s foreknowledge is exhaustive; (c) God could have actualized the early-death scenario; and (d) God prefers creaturely salvation. The remainder of this essay shows how the major schools of Muslim and Christian theology each locate the flaw in a different premise.

II. The Three Brothers: A Classical Islamic Formulation

The most famous statement of the dilemma in any religious tradition comes from the formative period of Islamic theology (kalām). As transmitted in later biographical literature, Abū al-Ḥasan al-Ashʿarī (d. 324/936) — then still a student of the Basran Muʿtazilī master Abū ʿAlī al-Jubbāʾī (d. 303/915), who was also his stepfather — posed the following question to his teacher.

In the version preserved by the biographer al-Subkī and translated in the collection associated with the University of Washington’s Nicholas Heer, the exchange runs: al-Ashʿarī asks about three persons, “one a believer, another an unbeliever, and the third a child.” Al-Jubbāʾī answers that “the believer is among the people of [high] rank [in Paradise], the unbeliever among those who perish, and the child among those who are safe.” Al-Ashʿarī asks whether the child may rise to the rank of the believer. Al-Jubbāʾī says no: “the believer achieved this rank only by obedience, and you have nothing of the sort.” Al-Ashʿarī presses: what if the child protests that the deficiency is no fault of his — “if Thou hadst suffered me to live, I would have rendered obedience like the believer”? Al-Jubbāʾī replies on behalf of God: “I knew that if you survived you would surely be disobedient and incur punishment. Therefore I considered what was best [for you]” — an appeal to the Muʿtazilī doctrine of al-aṣlaḥ, that God does what is optimal for his creatures. Al-Ashʿarī then springs the trap: “But what if the second [the damned unbeliever] said, ‘Lord, why didst Thou not make me die as a child? Then had I escaped Hell.’” Al-Jubbāʾī was reduced to silence. University of WashingtonCornellmuslimlife

A version transmitted from Asrar Rashid preserves the same logical structure in compressed form: God explains the child’s early death by saying, “I knew the boy would grow up to disobey Me and thereby enter Hell, so I did what was better for him”; whereupon the damned brother asks, “Why did you not make me die young, so as to have avoided disobeying you and consequently going to Hell?” — and “this question so dumbfounded al-Jubbāʾī that he remained silent.” The tellers of the story present it as a refutation of the Muʿtazilī doctrine of al-aṣlaḥ, and traditionally as the occasion of al-Ashʿarī’s dramatic break with Muʿtazilism. According to the framing found in the Encyclopaedia of Islam tradition and in reference works such as EBSCO’s Research Starters, “at about the age of forty, in 912 or 913, al-Ashʿarī underwent a radical conversion away from the extreme rationalism of his master al-Jubbāʾī, and made a declaration of repentance in the mosque,” after which he founded the school that bears his name. abuidris + 3

The philosopher M. Abdul Hye, in A History of Muslim Philosophy, summarizes the traditional interpretation: al-Ashʿarī presented “the case of three brothers, one being God-fearing, another godless, and a third having died as a child,” on “the problem of ṣalāḥ waʾl aṣlaḥ, i.e., the problem whether God’s actions are to be based on rational consideration and whether He is bound to do what is best for His creatures.” Al-Jubbāʾī “could not give a satisfactory and consistent reply,” and “on his having failed to justify rationally the Muʿtazilite doctrine,” al-Ashʿarī abandoned the Muʿtazilite camp. Islamiclearning + 3

III. The Question of Authenticity

Intellectual honesty requires noting that the anecdote’s historicity is seriously doubted by modern scholarship, even as its philosophical content is recognized as a genuine and serious objection to Muʿtazilī doctrine. The story does not appear in al-Ashʿarī’s own surviving works, nor in the earliest doctrinal record of his teaching, the Mujarrad Maqālāt al-Shaykh Abī al-Ḥasan al-Ashʿarī of his disciple Ibn Fūrak (d. 406/1015), edited by Daniel Gimaret. It surfaces only in later biographical and heresiographical literature: the defensive biography Tabyīn kadhib al-muftarī of Ibn ʿAsākir (d. 571/1176); the Wafayāt al-aʿyān of Ibn Khallikān (d. 681/1282), in the entry on al-Jubbāʾī; and the Ṭabaqāt al-Shāfiʿiyya al-kubrā of Tāj al-Dīn al-Subkī (d. 771/1370). These sources postdate al-Ashʿarī’s death by roughly 240 to 445 years, and several are overtly apologetic works composed to vindicate al-Ashʿarī’s reputation.

Rosalind W. Gwynne devoted a study to exactly this question: “Al-Jubbāʾī, al-Ashʿarī and the Three Brothers: The Uses of Fiction,” in The Muslim World (vol. 75, 1985). The very title signals her thesis — that the exchange functions as a literary and pedagogical construction whose value lies in its doctrinal “uses” rather than in its status as a verbatim transcript of a historical debate. The framing places the anecdote alongside other school-formation legends, such as the parallel Muʿtazilī founding story of Wāṣil ibn ʿAṭāʾ withdrawing from al-Ḥasan al-Baṣrī, whose historicity is likewise questioned because of its multiple variant versions. This is corroborated by the broader tendency of modern scholarship: W. Montgomery Watt, in his treatment of the Ashʿariyya, reports that al-Ashʿarī “is alleged to have” posed the three-brothers question, using a distancing verb that withholds endorsement; and the existence of several competing conversion narratives for al-Ashʿarī (including the well-known “three dreams” story in which the Prophet appears to him) is itself a standard marker scholars use to identify legendary accretion. The prudent scholarly verdict is therefore twofold: the narrative attribution is probably a topos, but the objection it dramatizes is a real and formidable one, which is precisely why it has been discussed for over a millennium.

IV. The Muʿtazilī Position: Divine Justice and the Optimum

The dilemma is, in the first instance, a problem for the Muʿtazila, because it is their premises that generate it most acutely. The Muʿtazila — the rationalist theologians of Basra and Baghdad — made divine justice (ʿadl) a defining principle, to the point that they called themselves ahl al-ʿadl waʾl-tawḥīd, “the people of [divine] justice and unity.” For them, good and evil are objective properties knowable by reason, and God, being perfectly just and wise, necessarily does what is good and never what is evil. Qāḍī ʿAbd al-Jabbār (d. 415/1025), the great systematizer of the Basran school, defined divine justice as “the knowledge that God is removed from all that is morally wrong (qabīḥ) and that all His acts are morally good (ḥasana),” including the corollary that God “does not impose faith upon the unbeliever without giving him the power for it.” En Academic

From this flows the doctrine of al-aṣlaḥ — that God must do what is best or most beneficial (aṣlaḥ) for his creatures. The Muʿtazila also held to genuine human free will: ʿAbd al-Jabbār was, in the words of one study, “a rationalist moral philosopher who considers reason as the origin of moral obligations,” and the school held that human beings are the real authors of their acts, so that reward and punishment are just. To bridge God and humanity, they developed the concept of luṭf — divine grace or “assistance” — the idea, as Maha Elkaisy-Friemuth describes it, that God is al-laṭīf, “the compassionate One, who provides all kinds of guidance,” facilitating obedience without compelling it. PhilArchiveTaylor & Francis

The three-brothers dilemma is fatal precisely to the aṣlaḥ doctrine in its strong form. If God is obligated to do the optimal thing for each creature, and if early death would have been optimal for the one who grows up to be damned, then God’s failure to bring about that early death is a failure of obligation — an injustice. Al-Jubbāʾī’s answer for the child (that God foreknew the child would sin and so mercifully took him early) is coherent; it is its generalization to the damned adult that collapses, because the same reasoning applied to the damned man shows that his continued life was not “the best” for him.

The Muʿtazilī tradition did possess resources that a later thinker might deploy, even if the anecdote portrays al-Jubbāʾī as failing to use them. The most important is the theory of compensation (ʿiwaḍ). ʿAbd al-Jabbār and his school held that God compensates creatures for undeserved suffering; the doctrine is developed at length in relation to pain, taklīf (moral obligation), and luṭf in the Bahshamī school founded by al-Jubbāʾī’s son Abū Hāshim al-Jubbāʾī. A Muʿtazilī could argue that God’s decision to let a person live — even one who will be damned — reflects the objective value of moral agency and the just linkage of desert to outcome, and that “best” (aṣlaḥ) must be understood in terms of what respects the moral order rather than what maximizes any individual’s post-mortem comfort. But this move relocates aṣlaḥ from the individual’s felicity to the integrity of the moral system — which is close to conceding the Ashʿarī point that God’s “best” is not straightforwardly the creature’s greatest advantage.

V. The Ashʿarī Response: Divine Sovereignty and “He Is Not Questioned”

The Ashʿarī school, which grew out of al-Ashʿarī’s reported break with Muʿtazilism, answers the dilemma by denying the premise that generates it: God is under no obligation whatever, including no obligation to do what is best for his creatures. This is theological voluntarism. Good and evil, for the mature Ashʿarī tradition, are not objective properties binding on God; rather, what God commands is good because he commands it, and God’s acts cannot be measured against an external moral standard.

The scriptural anchor is Qurʾān 21:23: “He is not questioned about what He does, but they will be questioned” (lā yusʾalu ʿammā yafʿalu wa-hum yusʾalūn). On the Ashʿarī reading, this verse denies precisely the standing that the damned sinner claims. The complaint “why did you not let me die young?” presupposes that God owes the creature an accounting; but the creature is owned by God, and the Owner is not answerable to what he owns. The idea, as one contemporary Ashʿarī-influenced writer puts it, is that “the principle of obligation has no meaning when it comes to God. To think that God is bound by anything is to declare His submission to it.” The dilemma dissolves not because it is answered on its own terms but because its central assumption — that God’s justice consists in maximizing creaturely benefit — is rejected. Medium

The Ashʿarī account of human action supports this. In place of the Muʿtazilī doctrine that humans create their own acts, al-Ashʿarī proposed the theory of kasb (“acquisition”): God creates the act, but the human being “acquires” it, and this acquisition is sufficient to ground moral responsibility. The damned are thus genuinely responsible for their unbelief, even though God is its ultimate creator, and they cannot shift the blame to God’s failure to kill them young. Wikipedia

The Ashʿarī tradition did not, however, simply abandon the notion of divine wisdom. Al-Ghazālī (d. 505/1111) famously advanced, in the Iḥyāʾ ʿulūm al-dīn, the statement laysa fī l-imkān abdaʿ mimmā kān — “there is not in possibility anything more wondrous than what is.” As the Stanford Encyclopedia of Philosophy notes, al-Ghazālī held “the conviction that this universe is the best of all possible worlds,” a position drawn in part from Avicenna. Eric Ormsby’s monograph Theodicy in Islamic Thought documents the fierce, centuries-long controversy this generated: al-Ghazālī’s critic Ibrāhīm al-Biqāʿī (d. 885/1480) accused him of outright heresy for a statement that seemed dangerously close to the Muʿtazilī and Leibnizian optimism the Ashʿarīs had rejected. The tension is instructive: an Ashʿarī who says the actual world is the best possible must be careful not to smuggle back the aṣlaḥ doctrine that the three-brothers argument had been marshalled to defeat. Al-Ghazālī’s defenders read his statement as a claim about God’s wisdom and the perfection of the created order (akmal wujūh al-wujūd, “the most perfect way to be”), not as a claim that God was obligated to produce it. Stanford Encyclopedia of Philosophy

Later Ashʿarīs refined the notion of divine wisdom (ḥikma) further. Fakhr al-Dīn al-Rāzī (d. 606/1210) developed what Ayman Shihadeh, in The Teleological Ethics of Fakhr al-Dīn al-Rāzī (Brill, 2006), describes as a departure “from classical Ashʿarī divine command ethics to develop both a consequentialist ethics of action… and a perfectionist ethics of character.” Al-Rāzī affirmed that God’s acts are purposive and full of wisdom (ḥikma) while continuing to deny that God is externally obligated — reconciling teleology with omnipotence. This move is directly relevant to the three-brothers problem, since it grants that God’s decision to let the damned man live has a wise purpose without conceding the Muʿtazilī premise that God was bound to do the individual’s aṣlaḥ.

VI. The Māturīdī Position: Wisdom Without Compulsion

The Māturīdī school, founded by Abū Manṣūr al-Māturīdī (d. 333/944) in Samarqand, stakes out a middle path between Muʿtazilī rationalism and Ashʿarī voluntarism, and its treatment of the dilemma is arguably the most balanced. Al-Māturīdī built his system on two principles: tanzīh (freedom from anthropomorphic likeness) and ḥikma (divine wisdom). Crucially, for al-Māturīdī ḥikma means “placing a thing in its right place,” and it “includes both ʿAdl (justice) and Faḍl (grace and goodness).” God possesses absolute power, “and His absoluteness is not subject to any external laws, but His own wisdom.”

This yields a nuanced result. Against the Muʿtazila, al-Māturīdī denies that God is externally obligated to do the aṣlaḥ. Against a crude reading of the Ashʿarīs, he insists that God never acts frivolously or in vain (ʿabath); God’s acts are always wise, but the wisdom is intrinsic to God rather than imposed from without. As one summary of Māturīdism puts it, “divine justice consists not in doing what is salutary to an individual, but in doing an action on its own merit and in giving a thing its own place.” The Māturīdīs also distinguish, following Ḥanafī doctrine, between God’s will (irāda) and his approval or love (riḍā, maḥabba): God may will that evil occur — because he wills that humans exercise genuine free choice — while never approving or commanding it. As one account summarizes: “God wills the happening of the evil because He desires the individual to exercise free choice, but being wise and just He always prohibits the choice of evil.” IlmGateAl-Islam

Applied to our dilemma, the Māturīdī answer is that God’s decision to let the future sinner live is neither a violation of an obligation (there is none) nor an arbitrary exercise of power (God does nothing in vain), but an expression of wisdom in giving the human being the arena of free moral choice — the very thing that makes reward and punishment meaningful. The Māturīdīs strongly emphasize, in the words of one contemporary account, that “God is not pleased with evil and does not force anyone to sin — when people commit evil, it is by their misuse of the divine gift of free will.” The sinner’s complaint fails because the gift of continued life, with its freedom, is precisely what dignifies him as a moral agent rather than a passive recipient of unearned safety. The Muslim Times

VII. Qurʾānic Material

Several Qurʾānic passages bear directly on the dilemma. The verse Q 21:23, already discussed, is the sovereignty-answer par excellence: “He is not questioned about what He does.” Its structural parallel in the Christian scriptures, as we shall see, is Romans 9:20.

The most striking Qurʾānic datum is the pericope of al-Khiḍr and Moses in Sūrat al-Kahf (Q 18:60–82), which contains what is in effect a Qurʾānic instance of the very logic the three-brothers argument exploits. Khiḍr, the mysterious servant of God to whom God “had given mercy… and taught knowledge,” kills a boy, to Moses’s horror (Q 18:74). He later explains (Q 18:80–81): “And as for the boy, his parents were believers, and we feared that he would overburden them by transgression and disbelief. So we intended that their Lord should substitute for them one better than him in purity and nearer to mercy.” Here early death is presented as a mercy — precisely the aṣlaḥ reasoning al-Jubbāʾī applied to the child. Classical commentators (Ibn Kathīr, al-Ṭabarī, al-Jalālayn) treat Khiḍr’s act as divinely commanded and grounded in foreknowledge of the boy’s future rebellion. The passage cuts both ways in our debate: it lends scriptural support to the idea that God sometimes takes a life early as a mercy against a foreseen evil future, yet it simultaneously sharpens the sinner’s complaint — for if God did this for the boy in the story, why not for the damned man? WikipediaDiscover the Truth

Q 6:28 supplies a partial answer by undercutting the assumption behind premise (d) that a second chance would change the outcome. Speaking of those in the Fire who wish to be returned to earthly life, the verse says: “But if they were returned, they would return to what they were forbidden, and indeed they are liars.” The verse asserts that the damned, even if given more life, would revert to their sin — a scriptural anticipation of the Molinist notion of trans-world damnation examined below.

Q 67:2 frames earthly life as a test: God “created death and life to test you as to which of you is best in deed.” Life is not merely a waiting-room for judgment but the arena in which moral character is formed and demonstrated — which is why terminating it early, though it would guarantee safety, would defeat its purpose.

Finally, the hadith corpus supplies the premise (a) that early death guarantees paradise, at least for the children of believers. The Prophet is reported to have said that “the pen is lifted from three” — including “a child until he becomes an adult” — meaning the child is not morally accountable (Sunan al-Tirmidhī). Numerous hadiths in al-Bukhārī and Muslim promise paradise to Muslim parents who lose young children; one narration from Abū Hurayra has the Prophet declare that “there is no Muslim couple whose three children die before reaching maturity but that Allah will admit them into Paradise by His mercy.” Yet the tradition also preserves a significant qualification. When asked about the children of the polytheists who die young, the Prophet is reported to have said, “Allah knows best what they would have done” (al-Bukhārī 1383, Muslim 2660). Ibn Kathīr harmonizes the reports by holding that such children will be tested in the hereafter: “whoever Allah knows will obey Him” is admitted to paradise, and “whoever He knows will not obey Him, his case rests with Allah.” This hadith is remarkable because it directly invokes counterfactual divine knowledge — what the deceased child would have done had he lived — the same modal knowledge that the Christian Molinists would later make the centerpiece of their response. Abu Amine EliasIslam Question & Answer

VIII. Modern Muslim Responses

Modern Muslim thinkers have generally approached the dilemma through the broader problem of evil rather than the three-brothers argument specifically. Said Nursi (1877–1960), in the Risale-i Nur, frames death itself — including untimely death — as “an instance of good and the gate of mercy” rather than an evil, reading all events through the divine names of mercy, wisdom, and compassion. Nursi holds that “human freedom can be conducted only within the limits of divine determination,” and he emphasizes that God, being both Most Compassionate and Most Just, defers much punishment to the afterlife precisely to give wrongdoers the opportunity to redeem themselves — an argument that directly answers the sinner’s complaint by reframing continued life as an extended mercy and opportunity rather than a trap. This resonates with Q 35:45, which Nursi cites: God “reprieves them for an appointed time.” ResearchGate + 2

More broadly, the modern kalām tradition tends to combine the Māturīdī emphasis on human free will with the Ashʿarī insistence on divine sovereignty, arguing that the objection illegitimately treats existence and moral testing as burdens rather than as honors, and that it presupposes a hedonic calculus (safety maximization) foreign to the Qurʾānic vision of life as amāna (trust) and test.

IX. The Christian Traditions: Augustine and Original Sin

The Christian versions of the dilemma are complicated — and in one respect intensified — by the doctrine of original sin. For Augustine of Hippo (354–430), Adam’s fall rendered all humanity a massa damnata, a “condemned mass,” inheriting both the guilt and the corruption of the first sin through concupiscence. The stark consequence, which Augustine did not shrink from, concerns unbaptized infants. In De peccatorum meritis et remissione I.16 he held that unbaptized infants suffer eternal condemnation, albeit “the mildest punishment of all” (mitissima poena omnium), and in De anima et eius origine he expressly rejected as Pelagian any “middle place of rest and happiness” for them. At the Synod of Carthage (418), Augustine helped secure the rejection of any intermediate happy state for children who die unbaptized. SVMMA APOLOGIA

For the Augustinian, then, the sinner’s dilemma takes a different and even harsher shape: early death does not obviously guarantee paradise, because the infant too bears Adam’s guilt. Premise (a) is denied at the root. Yet this “solution” purchases consistency at a terrible price, and it is a price later theologians were unwilling to pay in full. Augustine’s own answer to why God permits the reprobate to live at all is that God ordains the whole for the manifestation of both his mercy (in the elect) and his justice (in the reprobate); the existence and continuation of the damned serves the display of divine justice and the good of the ordered whole.

X. Aquinas, Limbo, and the Value of Earthly Life

Thomas Aquinas (1225–1274) softened the Augustinian position significantly. On the fate of unbaptized infants, he held (in De Malo q. 5, a. 3) that they suffer only the poena damni (the loss of the beatific vision) and not the poena sensus (sensible punishment), and that they enjoy a state of natural happiness, unaware of what they lack — the doctrine that hardened into the medieval limbus infantium, the “limbo of infants.” (The 2007 International Theological Commission document The Hope of Salvation for Infants Who Die Without Being Baptised traces this development and, notably, expresses hope for their salvation, entrusting them to God’s mercy.)

On predestination and reprobation, Aquinas develops a carefully asymmetrical account in Summa Theologiae Ia q. 23. Predestination, he holds, “is not anything in the predestined; but only in the person who predestines” — it is an aspect of divine providence, existing in the mind of God as a plan. Reprobation, correspondingly, “includes the will to permit a person to fall into sin, and to impose the punishment of damnation on account of that sin.” Crucially, Aquinas insists on an asymmetry: predestination is the cause of grace and glory in the elect, but “reprobation is not the cause of what is in the present — namely, sin,” which arises from the free will of the creature; reprobation is only “the cause of abandonment by God” and of the just punishment that follows. God is thus “the cause of the act of sinning but not the cause of sin,” for the moral defect comes from the creature’s free will, “just as the limp in a cripple comes from his deformity and not from his power to move” (ST IaIIae 79.2). As to why God creates those he foreknows will be damned, Aquinas grounds the answer in divine goodness: “the reason for the predestination of some, and reprobation of others, must be sought for in the goodness of God,” which is manifested in diverse ways across creation. New Advent + 6

Underlying the Thomist response is a robust theology of the value of earthly life as the arena of merit. Life is not a mere antechamber whose only relevant feature is its endpoint; it is the theatre in which free creatures, moved by grace, cooperate (or fail to cooperate) with God and thereby merit or forfeit glory. On this view, the sinner’s complaint — that he should have been killed young to secure automatic salvation — misunderstands the very nature of the good on offer. Automatic, unearned salvation of an infant is a real good, but the salvation of a mature moral agent who has lived, struggled, and freely turned to God is a different and, in an important sense, greater good, because it engages the whole person as a rational, willing being made in God’s image. This is essentially the same move the Māturīdīs make with ḥikma.

XI. Molinism and William Lane Craig: Trans-World Damnation

The most sophisticated modern Christian response to precisely this objection comes from Molinism. Luis de Molina (1535–1600), the Spanish Jesuit, proposed in his Concordia (1588) that God possesses scientia media, “middle knowledge” — knowledge of what any free creature would freely do in any possible set of circumstances (the “counterfactuals of creaturely freedom”). This knowledge is logically prior to God’s creative decree and independent of God’s will, so that God, surveying all the “feasible” worlds (those he can actualize given the true counterfactuals of freedom), chooses which to create. Middle knowledge thus aims to reconcile exhaustive divine providence with genuine libertarian freedom. As the Stanford Encyclopedia of Philosophy explains, middle knowledge “stands between God’s knowledge of necessary truths and his knowledge of his own creative will,” and its objects are “counterfactuals of freedom: If person S were in circumstances C, S would freely do X.” Stanford Encyclopedia of PhilosophyStanford Encyclopedia of Philosophy

The contemporary philosopher William Lane Craig has deployed this machinery against the exact “why not kill everyone as infants” objection. His key concept is trans-world damnation. Craig is careful to define it precisely; in a 2021 Reasonable Faith podcast he corrected a questioner’s misstatement, explaining that the doctrine “is not the doctrine that most individual human essences would reject God in any possible world,” but rather the claim “that in any world feasible for God… anyone who is damned is someone who would have been damned in any feasible world in which God might have created him.” The point is to protect divine goodness: no one is damned by “bad luck” of being created in the wrong world. As Craig puts it, God will say to the damned: “I knew that no matter which feasible world I created you would have rejected me and separated yourself from me forever. Therefore your lot is your own doing.” reasonablefaith

When the questioner pressed the infanticide version — why does God not arrange the death of the trans-world-damned before their age of accountability, thereby saving them? — Craig’s answer exposes the logical knot at the heart of our dilemma: “If God arranges for the death of someone prior to the age of accountability then that person does not suffer from transworld damnation. Right? Because he would be saved. So the question makes no sense… because if they die before the age of accountability they don’t have transworld damnation.” Craig diagnoses the objection as “this old fallacy of asking why doesn’t God change the future… You determine the future. You don’t change the future. That’s an incoherent notion.” reasonablefaith

Craig’s response, in effect, denies premise (c) as the objector understands it: God cannot both actualize the early-death scenario and have the person be the very individual who was in fact damned, because the counterfactual profile of the person who dies saved in infancy differs from that of the person who lives and is damned. It also draws on his controversial views about the age of accountability and infant salvation; Craig has publicly questioned whether infants bear the imputed guilt of original sin at all. In a Reasonable Faith video posted to X on April 28, 2025, he stated: “if, like me, you’re not persuaded that this doctrine is biblical, then infants and the mentally retarded are not born sinful and therefore have no sin that needs to be pardoned or forgiven.” This dovetails with his defense of the Canaanite conquest narratives, in which he argues (in an interview quoted by the theologian Randal Rauser): “If God wants to end the life of these Canaanite children early or prematurely, that’s his prerogative… These children go immediately to heaven. They go to be with God.” It should be noted that Molinism’s central mechanism faces the serious “grounding objection” — the question of what makes counterfactuals of freedom true if neither God’s will nor any actual state of affairs grounds them — which its critics regard as unresolved.

XII. Calvinist and Reformed Responses: Romans 9

The Reformed tradition answers the dilemma much as the Ashʿarīs do — by an appeal to sovereignty that denies the objector’s standing. The locus classicus is Romans 9:20–21, the structural twin of Qurʾān 21:23: “But who are you, O man, to answer back to God? Will what is molded say to its molder, ‘Why have you made me like this?’ Has the potter no right over the clay, to make out of the same lump one vessel for honorable use and another for dishonorable use?” Paul’s rhetorical rebuke is precisely a refusal to grant the creature the standing to arraign the Creator’s choices — exactly the standing the damned sinner in our dilemma claims. Servants of Grace

John Calvin (1509–1564) pressed this to the doctrine of double predestination, holding that God elects some to salvation and reprobates others, and that reprobation is not by “bare permission” but reflects God’s deliberate will. In his commentary on Romans he even endorsed the stark reading of Proverbs 16:4 that “the ungodly themselves have been created for the specific purpose of perishing.” R. C. Sproul, in his commentary Romans (quoted by Ligonier Ministries), summarizes the Reformed conscience on this: “The elect get grace; the non-elect get justice. Nobody gets injustice.” For the Calvinist, the sinner’s complaint founders on the fact that he is a guilty sinner receiving justice, not an innocent receiving harm; and the deeper question of why God ordained this particular distribution is met, with Calvin, by the confession that “a mystery which our minds cannot comprehend ought to be reverently adored.” Reformed Faith & Practice + 3

On infant salvation, the Reformed tradition is more hopeful than is often supposed. The Westminster Confession (10.3) states that “elect infants, dying in infancy, are regenerated, and saved by Christ, through the Spirit.” As Reformed writers note, the Confession states the matter “only in the positive sense,” affirming that elect infants are saved without requiring belief in the existence of any non-elect infants — leaving open the possibility that all who die in infancy are among the elect. This preserves premise (a) for the Reformed while keeping the sovereignty of grace intact. Reformation 21

XIII. Arminian and Free-Will Responses; Soul-Making

The Arminian and broader free-will traditions answer by emphasizing the intrinsic value of libertarian freedom and of a world with genuine moral history. The “age of accountability” concept — that those who die before they can knowingly sin are covered by grace — is largely a development of non-Calvinistic Protestants of the nineteenth and twentieth centuries attempting to address the issue of infant mortality on Arminian premises. The Wesleyan Church’s Articles of Religion, Article 9 (The Atonement), for example, teaches that Christ’s atonement “is sufficient for every individual of Adam’s race. It is unconditionally effective in the salvation of those mentally incompetent from birth, of those converted persons who have become mentally incompetent, and of children under the age of accountability.”

The deepest free-will response to the dilemma is the soul-making theodicy, given its classic modern form by John Hick in Evil and the God of Love (1966). Reviving what he called the Irenaean (as opposed to Augustinian) tradition, Hick argued that God does not create humans ready-made in perfection but places them in a world structured for growth — a “vale of soul-making” — because virtues “formed as a result of a person overcoming temptations and challenges are intrinsically more valuable than virtues created within him ready-made without effort on his own part.” Crucially, Hick argued that God could have created humanity directly in his presence, but that this would have collapsed the “epistemic distance” necessary for free response; a world of guaranteed innocents would be a world without the goods that only freedom and struggle can produce. On this view, the sinner’s complaint is answered by denying the assumption implicit in premise (d) that God’s sole or overriding preference is safety-maximization: God prefers a world in which real persons freely make real histories, even at the risk of damnation, to a world of unearned, characterless safety. Wikipedia + 2

XIV. The Age-of-Accountability Paradox and Its Reductio

Both traditions face a sharp internal difficulty that the three-brothers argument exposes: if all who die before the age of accountability are automatically saved, then early death appears soteriologically optimal, and — pressed to its limit — the doctrine seems to make infanticide “beneficial” to its victim. This reductio has been stated explicitly in the philosophy of religion literature. Phil Stilwell, in “The Benevolent Monster: Age of Accountability, Infinite Utility, and the Soteriological Reductio of Infant Salvation,” argues that under “traditional doctrines of eternal heaven, eternal hell, and adult salvation risk… If early death guarantees infinite bliss while continued life introduces a non-zero risk of infinite loss, the system appears to make infant death soteriologically preferable.” Stilwell frames this not as a recommendation but as “a stress test of a theological framework,” targeting “the conjunction of automatic infant salvation, eternal stakes, and the claim that many who reach accountability may be damned.” PhilArchive + 2

The standard replies, catalogued by Stilwell and by apologetic responders, are instructive because they reveal what the traditions take to be the objection’s hidden false premise. They include: that killing is intrinsically sinful and usurps God’s exclusive prerogative over life and death (so the human agent who acts on the calculus commits grave evil regardless of the victim’s fate); that earthly goods and the moral life have genuine positive value that the calculus ignores; that divine providence, not human calculation, balances the system; and — decisively — that reasoning with infinite utilities is decision-theoretically unstable, so the “dominant strategy” argument does not go through. An evangelical responder at Evidence Unseen makes the pastoral form of the point: “We shouldn’t judge normal cases… on the basis of extraordinary cases.” The Gospel Coalition

The common thread is a refusal of the premise that existence and moral life are, on balance, a net risk to be minimized. Both the Muslim and the Christian mainstream hold that being, freedom, and the opportunity for a morally significant life are goods — indeed great goods — even though they carry the risk of damnation. The objection, taken with full seriousness, implies that non-existence or the truncation of existence is preferable to moral agency; and this both traditions deny.

XV. Contemporary Analytic Philosophy of Religion

The dilemma intersects several live debates in analytic philosophy of religion. Open theism, associated with Clark Pinnock, Gregory Boyd, John Sanders, and William Hasker, denies premise (b): it holds that God does not exhaustively foreknow the free choices of creatures, because future free choices are not yet determinate and so are not available even to an omniscient God. As Pinnock put it in his “Systematic Theology,” in The Openness of God (InterVarsity Press, 1994, p. 123): “Philosophically speaking, if choices are real and freedom significant, future decisions cannot be exhaustively foreknown. This is because the future is not determinate but shaped in part by human choices… God knows everything that can be known, but God’s foreknowledge does not include the undecided.” On this view the sinner cannot say “you knew how my life would unfold,” because — with respect to free choices — God did not know with certainty; he knew all possibilities and took a loving risk. This dissolves the dilemma at the cost of a controversial revision of the classical doctrine of omniscience, which critics regard as unacceptably high. UASV BibleCome And Reason Ministries

Skeptical theism denies the objector’s confidence in premise (d) — that we can reliably judge what an omniscient, omnibenevolent God would prefer. On this view, our inability to see why God permits a particular person to live and be damned is no evidence that there is no such reason; the cognitive gap between us and God is too vast for the inference to succeed. This is the analytic descendant of the Ashʿarī lā yusʾalu and the Pauline “who are you, O man?”

Marilyn McCord Adams reframes the entire problem in Horrendous Evils and the Goodness of God (1999). She defines horrendous evils (p. 26) as “evils the participation in which… constitutes prima facie reason to doubt whether the participant’s life could (given their inclusion in it) be a great good to him/her on the whole.” She argues that traditional theodicies err by appealing only to “greater goods at the level of the universe” and by trading in “religion-neutral values.” What divine goodness requires, she contends, is not that God outweigh or explain evils globally but that God defeat them within the context of each individual’s life — that God provide “a logically possible scenario in which God is good to each created person, by insuring each a life that is a great good to him/her on the whole.” Adams’s approach suggests that the right answer to the damned sinner’s complaint is not a cost-benefit demonstration but the claim that God’s goodness is fundamentally a matter of intimate, person-specific relationship and the defeat of horrors through participation in the divine life — a good that mere early death could not secure. It is notable that her relational, anti-calculative framing converges with Nursi’s insistence that the problem of evil is not solved by a ledger but dissolved in the recognition of divine mercy and presence. Cornell University Press + 2

XVI. Comparative Analysis

Several structural parallels emerge with striking clarity. First, the sovereignty-answer is common to both traditions and is expressed in almost identical scriptural idioms: Qurʾān 21:23 (“He is not questioned about what He does”) and Romans 9:20 (“who are you, O man, to answer back to God?”). Both texts respond to the creature’s arraignment of God not by adjudicating the complaint but by denying the creature’s standing to make it. The Ashʿarī and the Calvinist are, on this axis, near-twins; both are theological voluntarists for whom the good is what God wills and the creature is clay in the potter’s hand.

Second, the Muʿtazilī doctrine of al-aṣlaḥ and Leibnizian best-possible-world optimism are structurally parallel: both hold that a perfectly good God is constrained (by his own nature) to produce the optimal outcome, and both generate versions of the three-brothers problem, because both invite the question “if this is the best God could do, why is any individual worse off than he need be?” Al-Ghazālī’s laysa fī l-imkān abdaʿ mimmā kān is the Islamic analogue of Leibniz’s optimism, and, tellingly, it drew accusations of crypto-Muʿtazilism just as Leibniz’s optimism drew Voltaire’s satire — because both risk making God the debtor of his creatures.

Third, Māturīdī ḥikma and Thomist divine wisdom occupy the same conceptual middle ground: both affirm that God’s acts are purposive and wise (not arbitrary) without conceding that God is externally obligated to any creature. Al-Māturīdī’s formula that ḥikma means “placing a thing in its right place” is remarkably close to the Thomist conception of providence as the ordering of all things to their proper ends. Al-Rāzī’s later teleological refinement of Ashʿarism moves the Islamic mainstream still closer to the Thomist synthesis.

Fourth, and most striking, the three-brothers dilemma and Craig’s Molinist machinery are two sophisticated attempts at the same problem, separated by a millennium and arriving at convergent conclusions. Al-Jubbāʾī’s answer for the child — “I knew that if you survived you would surely be disobedient” — is a counterfactual of creaturely freedom in embryonic form: it appeals to God’s knowledge of what the child would have done. The Islamic hadith that “Allah knows best what they would have done” makes the point explicit. What al-Jubbāʾī lacked was the systematic apparatus to generalize this to the damned adult without contradiction; what Craig supplies, with trans-world damnation, is exactly that generalization: the damned man would have been damned in any feasible world, so his early death would not have “saved him” as the same individual — it would have produced a different, saved person. The kalām tradition posed the problem with unmatched sharpness; the Molinist tradition supplied the modal machinery to answer it. Whether that machinery succeeds (given the grounding objection) is a separate question; but the continuity of the problem across a thousand years and two traditions is remarkable.

Finally, we can map the schools onto the four premises. The Muʿtazila accept all four premises and are therefore genuinely embarrassed by the dilemma (which is its polemical point). The Ashʿarīs and Calvinists deny that God’s justice consists in maximizing creaturely benefit (the assumption behind premise d), and deny the creature’s standing to press (c). The Māturīdīs and Thomists deny the hedonic reading of (d) by insisting that the greater good is the morally significant life, not unearned safety. The Molinists deny premise (c) as coherently applicable to the same individual (via trans-world damnation), and deploy Q 6:28’s insight that the damned “would return to what they were forbidden.” Open theists deny premise (b), the exhaustive foreknowledge. Augustinians, at the extreme, deny premise (a), that early death guarantees paradise. Each school locates the flaw in a different link of the chain.

XVII. The Shared Intuition and the Refused Reductio

Beneath the doctrinal divergences lies a shared conviction that unites the Muslim and Christian mainstream against the objection. The dilemma, pressed to its conclusion, implies that existence and moral agency are net harms — that it would have been better for the damned never to have lived a full human life, and, by extension, that a world populated only by guaranteed-innocent early-deceased would be a better world than the one we have. Both traditions refuse this. For the Muslim, life is amāna and a test (Q 67:2), and the freedom to choose is the honor that distinguishes the human being (the Māturīdī ikhtiyār). For the Christian, life is the arena of merit and soul-making, and the free turning of a mature person to God is a good that an infant’s automatic salvation, however real, cannot match. Both hold that being is better than non-being, and that a genuine moral history — with its risks — is better than a safe blankness. The sinner’s complaint, on this shared view, is not so much refuted as re-described: what he presents as God’s failure to spare him is in fact God’s refusal to reduce him to a passive recipient of safety, God’s insistence on treating him as a responsible agent. That the agent then damns himself is, in the language of both Aquinas and al-Ashʿarī, the creature’s own doing.

Epilogue: The Uses of a Dilemma

The story of the three brothers may well be, as Rosalind Gwynne suggested, a fiction — a narrative constructed after the fact to dramatize the collapse of a doctrine. But its endurance across a millennium, and its independent re-emergence in the seminar rooms of analytic philosophy of religion, testify to something more than antiquarian interest. The dilemma is a probe that, inserted into any theological system, reveals the system’s deepest commitments: its theory of divine justice, its account of human freedom, its valuation of earthly life, its doctrine of foreknowledge. The Muʿtazila and the Leibnizians, who make God the guarantor of the optimum, feel its full force. The Ashʿarīs and Calvinists blunt it by denying that the creature may question the potter. The Māturīdīs and Thomists absorb it by relocating the good from safety to significance. The Molinists dissolve it with the modal claim that the damned would have damned themselves in any world. The open theists escape it by surrendering exhaustive foreknowledge. And thinkers like Marilyn McCord Adams and Said Nursi step outside the calculus entirely, insisting that the language of cost and benefit was never adequate to the reality of a God who is present to each sufferer in mercy.

What no major tradition does is concede the objection’s ultimate premise: that it would have been better not to have lived. Here the theologies of Mecca and of Rome, of Basra and of Geneva, speak with one voice. Existence is a gift, freedom is an honor, and a life — even one that ends in ruin — is the story of a real agent and not the malfunction of a safety mechanism. The sinner who cries “why did you not let me die a child?” is told, in the end and in many idioms, that the question mistakes the nature of the gift he was given. He was not owed the safety of the cradle; he was given the dignity of the road. That he walked it to his own destruction is the tragedy the traditions acknowledge — and the responsibility they, with remarkable unanimity, decline to transfer from the creature to the Creator.