Nothing Said of Anyone

A discursive theological article on hell: what has actually been defined and by which acts, the pain of loss and the pain of sense, the eternity of punishment and the texts it rests on, the patristic range, the Thomistic treatment read at its own loci, the question of who is in fact damned, and the universal salvific will. The Four Last Things, Part III

A Short File

The Catechism of the Catholic Church states the doctrine of hell in five paragraphs, and hangs the load-bearing sentence of all five on a single footnote. The sentence is at paragraph 1035: “The teaching of the Church affirms the existence of hell and its eternity.” The footnote is a string of eight references and one abbreviation: Cf. DS 76; 409; 411; 801; 858; 1002; 1351; 1575; Paul VI, CPG #12.¹

That footnote is the file. It is not a summary of the file; it is not a selection from a larger file the Catechism could have cited; it is, so far as the Church’s own most recent comprehensive statement is concerned, the documentary basis on which a Catholic is asked to hold that hell exists and does not end. Eight numbers. It takes about two hours to read all of them at their own loci, in the languages in which they were promulgated, and the experience of doing so is not what a reader who has absorbed the popular presentation of this doctrine expects. It is not that the file is weaker than advertised. On the central assertions it is considerably firmer than a modern Catholic is usually led to believe — these are conciliar and papal acts of the highest solemnity, repeated across eleven centuries, and there is no honest way to read them as metaphor or as a pastoral exaggeration that the Church has quietly retired. It is that the file is so much shorter than the doctrine as it is ordinarily preached, painted, and imagined. Almost everything a reader thinks he knows about hell is not in it.

There is no fire in it whose nature is specified. There is no geography, no location, no duration of anything other than the state itself, no description of the sufferings, no account of what the damned experience or think or do. There is no number: not a proportion, not an estimate, not a hint. And — this is the point at which the shortness of the file stops being a curiosity and becomes the governing fact of the subject — there is no name. In two thousand years the Church has canonized many thousands of individuals, declaring of each that he is in fact in heaven. She has never once, of any human being who has ever lived, made the corresponding declaration in the other direction. Not of Judas. Not of Nero, or Arius, or Simon Magus, or any of the men whose names the tradition uses as bywords for damnation. The doctrine is defined; the census is empty.

This essay is about that shape — maximal on the fact, minimal on the mode, and silent on every person — and about why the shape is not an evasion but the doctrine itself, correctly stated. The third part of a four-part series has an awkward inheritance: the essay on death establishes that the choice of a life is closed by dying, and the essay on judgment establishes that the closed choice is met by a verdict. This essay takes up what the verdict can be when it goes the other way, and it is the hardest of the four to write honestly, because the temptation runs in both directions at once. On one side there is the lurid tradition: the tortures itemized, the population estimated, the whole apparatus of a doctrine that has been made to say a great deal more than it ever said, until a reader who discovers how little of it is defined feels he has been sold something. On the other side there is the quieter and now more common failure, in which the doctrine is retained in form, evacuated in substance, and mentioned only in a tone that makes clear the speaker does not think it applies to anyone. Both are ways of not reading the file.

So the method here is simple to the point of pedantry: go to each source, read it whole, in its own language, at its own locus, and report what it actually says — including where it says less than the tradition later assumed and where it says more than a modern reader is prepared for. Claims are marked by weight throughout. Definitive means an act the Church proposes as revealed and to be held by faith; authoritative but not definitive means genuine magisterial teaching that binds religious assent without being defined; common theological opinion means a position widely held in the schools, often by saints, which the Church has never made her own; disputed means what it says; and project synthesis means a judgment of this essay, argued from the checked sources and attributable to no one else. Where those categories cut across each other — and on this subject they constantly do — the essay says so rather than smoothing them into a single voice.

One preliminary distinction has to be made now, because English hides it and the confusion it causes runs through the whole popular literature. The word “hell” in older English, and infernus in Latin, and Hades and Sheol behind them, name in the first instance the abode of the dead as such, without any judgment about their state. That is the sense in which the Creed says that Christ descended into hell; nobody supposes he descended into the place of the damned to be punished. Scripture uses a second word, Gehenna — from the valley of Hinnom outside Jerusalem — for the place of final loss, and it is that second word, not the first, which carries the doctrine.² An enormous amount of confident argument on both sides of this subject — that hell is not really in the Old Testament, that the New Testament contradicts itself, that the Church invented the doctrine later — turns out on inspection to be an equivocation between those two senses. This essay uses “hell” throughout in the second sense, the state of final and definitive loss, and says so wherever a source uses it in the first.

The essay does not argue that hell exists. That is not something an essay can argue; it is a revealed claim, received on the authority of the one who made it, and a reader who does not grant that authority will not be moved by a better arrangement of the same sentences. What the essay can do, and what it undertakes, is to establish exactly what has been claimed, on exactly whose authority, with exactly what force — and to be as scrupulous about the wide margin of undefined territory as about the narrow core. That margin turns out to be where nearly all the real theological work of the tradition has happened, and where nearly all the damage of the popular presentation has been done.

What the Acts Say

Read in order of date, the eight items in the Catechism’s footnote fall into three unequal groups: one creed of uncertain authorship, one set of canons whose conciliar standing is a genuine historical question, and six acts of councils and popes whose standing is not in question at all. Taking them in that order is not a way of grading them from weak to strong. It is a way of showing that the doctrine did not arrive by a single decree and does not stand on one.

The creed nobody wrote

The oldest item is the Quicumque, the creed that begins “Whosoever wishes to be saved,” known for a thousand years as the Athanasian Creed and not written by Athanasius. Denzinger’s number 76 covers its second half, and its penultimate clause is the sentence the Catechism wants: et qui bona egerunt, ibunt in vitam aeternam; qui vero mala, in ignem aeternum — “and they who have done good shall go into life everlasting; but they who have done evil, into everlasting fire.”³

An anonymous creed is a strange thing to put at the head of a dossier, and its force is easy to state exactly. It is not an act of a council or a pope. What it is, is a text the Latin Church took into her worship and her catechesis and recited as her own for more than a millennium — which is precisely the mechanism by which the ordinary and universal magisterium does most of its work. A doctrine held everywhere, taught everywhere, and prayed everywhere does not become less binding for lacking a promulgation date. It is also worth noticing what the sentence does and does not do. It puts the two destinies in exactly parallel grammar, with the same adjective on both sides. It does not describe the fire.

The canons of 543, and a widespread misattribution

Denzinger’s numbers 403 to 411 are nine canons anathematizing propositions drawn from Origen. The ninth is the one that matters here: Si quis dicit aut sentit, ad tempus esse daemonum et impiorum hominum supplicium, eiusque finem aliquando futurum, sive restitutionem et redintegrationem esse daemonum aut impiorum hominum, anathema sit — if anyone says or holds that the punishment of demons and of impious men is temporary and will one day have an end, or that there will be a restoration and reintegration of demons or of impious men, let him be anathema.⁴

Two things must be said about this text, and they pull in opposite directions.

The first is that it is the sharpest and most explicit statement in the whole file. Nothing else in the dossier addresses the eternity of punishment head-on as its own question. Everything else states the doctrine while doing something else; this canon states it as the thing being decided, names the contrary opinion in both of its forms — a punishment that ends, and a general restoration — and anathematizes both. It also does something the later Latin acts never do: it extends the sentence explicitly to the demons.

The second is that the canons of 543 are very frequently attributed to the Second Council of Constantinople of 553, the fifth ecumenical council, and Denzinger does not print them there. He prints them under 543, as canons of a local synod at Constantinople convened at the initiative of the emperor Justinian, who had himself composed the Book against Origen from which they are drawn. The acts of the ecumenical council of 553, which Denzinger prints immediately afterwards, are the fourteen anathemas concerning the Three Chapters. Origen appears in them exactly once, in the eleventh, in a list: “If anyone does not anathematize Arius, Eunomius, Macedonius, Apollinaris, Nestorius, Eutyches, and Origen, together with their impious writings, and all other heretics…”⁵ The council names Origen among the heretics. It does not say which of his propositions it means, and it certainly does not repeat canon 9.

The distinction is not a technicality that lets anyone off. The 543 canons were confirmed by the pope, entered the Church’s doctrinal record, were cited as authoritative for fourteen centuries, and are cited by the Catechism in 1992 as evidence for the existence and eternity of hell. That is real teaching authority. But it is authority of a different kind from an ecumenical definition, and a reader told flatly that “an ecumenical council defined the eternity of hell against Origen in 553” has been told something the sources do not support in that form. The correction cuts one way only: it does not weaken the doctrine, which rests on much more than these canons. It weakens a particular argument that is often made for it, and the difference matters because arguments that will not survive inspection are a liability to the thing they defend.

Lateran IV, and the councils of the East and West reunited

With the Fourth Lateran Council of 1215 the file moves onto ground where nothing is disputed. Firmiter — the opening constitution, “Firmly we believe and simply confess” — is a definition of an ecumenical council, and its closing clause runs: all men shall rise with their own bodies, which they now bear, to receive according to their works, whether these have been good or evil, illi cum diabolo poenam perpetuam, et isti cum Christo gloriam sempiternam — “the former perpetual punishment with the devil, the latter everlasting glory with Christ.”⁶

Three features of this text repay attention. It states the doctrine inside a creedal chapter, as part of the shape of Catholic faith and not as a separate topic. It states it symmetrically, again, with the destinies of the wicked and the elect in the same sentence and the same construction — a pattern that will turn out to matter enormously when the essay reaches the exegesis. And it says cum diabolo, with the devil: the same association the canon of 543 made, and the same one the sentence of Matthew 25:41 makes.

Fifty-nine years later, the Second Council of Lyons received the profession of faith of the emperor Michael Palaeologus, and it contains the most precise sentence in the entire dossier: Illorum autem animas, qui in mortali peccato vel cum solo originali decedunt, mox in infernum descendere, poenis tamen disparibus puniendas — “but the souls of those who depart in mortal sin, or with original sin only, descend immediately into hell, to be punished, however, with unequal punishments.”⁷

Every word of that is load-bearing. Mox: immediately, not at the general judgment — the state begins at death, which is why the essay on judgment in this series has to distinguish the particular judgment from the general one. In mortali peccato vel cum solo originali: two classes, not one, and the second class is those who die with original sin only, which is to say without any personal sin whatever. Poenis tamen disparibus: unequal punishments, with tamen — “however” — doing the work of a concession, marking that the two classes share the word infernum but not the same suffering. And what is not there is as striking as what is: this definition, the most exact in the file about who goes and when, contains no word for eternity at all.

Benedict XII, and a definition that does not say what it is thought to say

The same is true, and more conspicuously so, of the constitution Benedictus Deus of 1336, which is the single most solemn act in the dossier — a papal definition, in the strict sense, made after a controversy in which Benedict XII’s own predecessor had preached a contrary opinion. Its clause on hell reads: Diffinimus insuper, quod secundum Dei ordinationem communem animae decedentium in actuali peccato mortali mox post mortem suam ad inferna descendunt, ubi poenis infernalibus cruciantur — “We define moreover that, according to God’s general ordinance, the souls of those who depart in actual mortal sin descend immediately after their death into hell, where they are tormented with infernal punishments.”⁸

Now compare the clause four lines above it, on the blessed, which the same act defines with the same verb: their vision and enjoyment of the divine essence, once begun, has continued and will continue without interruption usque ad finale iudicium et ex tunc usque in sempiternum — until the final judgment and from then on for ever. The pope who was willing to write in sempiternum about heaven, in the same document, four lines apart, did not write it about hell. He defined that the descent is immediate, that the state is real, that it involves punishment. He did not there define its duration.

This is not a discovery that undermines anything, and it must not be reported as one. The eternity of hell is stated by the Quicumque, by canon 9 of 543, by Firmiter, by Florence’s decree for the Jacobites, by Trent, and by Paul VI; it is taught by the ordinary and universal magisterium; it is asserted flatly by the Catechism in the sentence this essay began with. Nobody needs Benedictus Deus for it. But the observation does establish something about the shape of the file, which is the subject of this essay: the two acts in it that define most exactly what happens to a soul at death — Lyons and Benedictus Deus — are concerned with immediacy and with disparity, and leave duration to the acts that had already settled it. A reader who has been told that the tradition is monotonously repetitive on this subject has not looked at it. The tradition is economical. Each act adds what its own controversy required and no more.

Florence on the Jacobites, at full strength

The hardest text in the dossier is the bull Cantate Domino of 1442, the Council of Florence’s decree for the Jacobites, and it should be quoted without softening: the Roman Church firmiter credit, profitetur et praedicat, nullos extra catholicam Ecclesiam exsistentes, non solum paganos, sed nec Iudaeos aut haereticos atque schismaticos, aeternae vitae fieri posse participes; sed in ignem aeternum ituros, “qui paratus est diabolo et angelis eius” (Mt 25,41), nisi ante finem vitae eidem fuerint aggregati — firmly believes, professes, and preaches that none of those existing outside the Catholic Church, not only pagans but also Jews or heretics and schismatics, can become partakers of eternal life, but will go into the eternal fire prepared for the devil and his angels, unless before the end of life they are joined to it.⁹

That is the text. It is an act of an ecumenical council, and it is not a text a Catholic may set aside. It is also a text on which the Church’s own authoritative interpretation has done a great deal of work, and the interpretation is itself magisterial and cannot be set aside either. The Second Vatican Council states the same doctrine with the condition made explicit: whoever, “knowing that the Catholic Church was made necessary by Christ, would refuse to enter or to remain in it, could not be saved”; and, two paragraphs later, “those also can attain to salvation who through no fault of their own do not know the Gospel of Christ or His Church, yet sincerely seek God and moved by grace strive by their deeds to do His will as it is known to them through the dictates of conscience.”¹⁰ The condition on which Florence’s sentence turns — nisi ante finem vitae eidem fuerint aggregati, unless they are joined to the Church — is read by the later magisterium as reached in more than one way, and the culpability of the separation is what carries the weight.

The honest description of the relationship is neither “Florence was wrong” nor “nothing has changed.” It is that a fifteenth-century council stated a doctrine in a formula shaped by the controversy in front of it, in which the persons contemplated were people who knew the Church and had left or refused her; and that the twentieth-century magisterium, facing a different question, stated the same doctrine with the operative condition spelled out. Development of that kind is ordinary in the history of dogma and is not concealment. What would be concealment is quoting only one of the two.

Trent, and Paul VI

The seventh item, DS 1575, is the weakest citation in the footnote, and saying so is not a criticism of the Catechism. It is canon 25 of the sixth session of the Council of Trent (1547), and its subject is not hell but the good works of the justified: anathema on anyone who says that the just man sins at least venially in every good work, or, more intolerably, mortally, and therefore merits eternal punishments (poenas aeternas mereri), and is not damned only because God does not impute those works to damnation.¹¹ Eternal punishment appears in the canon as a shared premise between Trent and its opponents, not as the thing being defined. It is real evidence — a doctrine assumed as common ground in a definitive act is genuinely attested by that act — but it is evidence of a different order from Firmiter, and a reader deserves to be told which is which.

The last item is the most recent: Paul VI’s Credo of the People of God, pronounced on 30 June 1968 at the close of the Year of Faith, whose twelfth paragraph professes that Christ will come again in glory to judge the living and the dead, “each according to his merits — those who have responded to the love and piety of God going to eternal life, those who have refused them to the end going to the fire that is not extinguished.”¹² Two clauses are worth noting. “To the end” — usque in finem in the tradition’s phrase — makes explicit what the older texts leave implicit: what damns is not the refusal but the final refusal. And “the fire that is not extinguished” is scriptural language, taken from Mark, rather than a description; the most recent solemn profession of the doctrine describes hell in exactly the words Christ used and adds nothing.

The aggregate

Set beside each other, the eight acts assert five things and no more. That hell exists. That it is entered immediately at death and not at the general judgment. That its punishments are unequal, and that at least two classes are distinguished within them. That the state does not end. And — by a related act the Catechism cites in the next paragraph rather than this one — that God predestines nobody to it: the Second Council of Orange in 529 anathematized the contrary with unusual heat, saying that those who wish to believe so great an evil are anathematized cum omni detestatione, and Trent repeated the anathema a thousand years later.¹³

The Catechism adds one further statement of its own, in the same paragraph and in the summary that follows: “The chief punishment of hell is eternal separation from God, in whom alone man can possess the life and happiness for which he was created and for which he longs.”¹⁴ That sentence is not in any of the eight acts. It is the Church’s own authoritative gloss, and it identifies which of the two classical punishments is the principal one — which is where this essay has to go next, because the distinction it presupposes is both older than the Catechism and much less well known than the fire.

Two Punishments, and Which One Is the Doctrine

Around the year 180, in the last book of a long refutation of Gnostic teaching, Irenaeus of Lyons wrote a paragraph about the judgment which is, so far as this essay could establish, the earliest sustained Christian statement of what hell consists in. It is worth quoting at length, because everything the later tradition does with this question is already in it:

And to as many as continue in their love towards God, does He grant communion with Him. But communion with God is life and light, and the enjoyment of all the benefits which He has in store. But on as many as, according to their own choice, depart from God, He inflicts that separation from Himself which they have chosen of their own accord. But separation from God is death, and separation from light is darkness; and separation from God consists in the loss of all the benefits which He has in store. Those, therefore, who cast away by apostasy these forementioned things, being in fact destitute of all good, do experience every kind of punishment. God, however, does not punish them immediately of Himself, but that punishment falls upon them because they are destitute of all that is good. Now, good things are eternal and without end with God, and therefore the loss of these is also eternal and never-ending. It is in this matter just as occurs in the case of a flood of light: those who have blinded themselves, or have been blinded by others, are for ever deprived of the enjoyment of light. It is not, however, that the light has inflicted upon them the penalty of blindness, but it is that the blindness itself has brought calamity upon them.¹⁵

Three claims are packed into that, and they are not the same claim. First, that what hell is is separation, and that the separation is chosen by the one separated rather than imposed on him from outside. Second, that God does not punish the lost “immediately of Himself,” the suffering falling upon them from the privation itself — Irenaeus’s blindness image is doing exactly the work that image is designed to do: the sun does not blind, and the sun does not withdraw, and the blind man is nonetheless in the dark. Third, that the loss is eternal because what is lost is eternal. Not one word about fire.

That is not because Irenaeus did not believe in the fire. Two paragraphs earlier in the same chapter he had cited the everlasting fire prepared for the devil and his angels, and elsewhere he speaks of it without embarrassment. It is because, asked what the punishment consists in, he answered with the loss.

The two names

The later scholastic tradition gave the two things names, and the names are worth having because they keep the analysis honest: poena damni, the pain of loss — the deprivation of God — and poena sensus, the pain of sense, the positive affliction inflicted on the damned by something other than themselves. The distinction is not a scholastic invention imposed on the Fathers; it is a label for a distinction the Fathers were already making. John Chrysostom, preaching at Antioch in the 390s on the narrow gate and the false prophets, makes it explicitly, first as a piece of exegesis and then as a cry:

And though it appear indeed to be some single judgment, the being burnt up, yet if one examine carefully, these are two punishments. For he that is burnt is also cast of course out of God’s kingdom…

No doubt hell, and that punishment, is a thing not to be borne. Yet though one suppose ten thousand hells, he will utter nothing like what it will be to fail of that blessed glory, to be hated of Christ, to hear “I know you not”… Yea, better surely to endure a thousand thunderbolts, than to see that face of mildness turning away from us, and that eye of peace not enduring to look upon us.¹⁶

“Ten thousand hells” is a rhetorical figure, and Chrysostom is a preacher rather than a systematician; but the figure is doing a precise piece of theological work, and it is the same work Irenaeus’s blindness image did. It ranks the two punishments. Whatever the fire is, it is not the worst thing about hell.

Thomas Aquinas states the same ranking in the language of quantity, and the result is one of the most counterintuitive sentences in the Summa. Asking whether sin incurs a debt of punishment infinite in quantity, he distinguishes the two components of any sin — the turning away from the immutable good, and the disordered turning towards a mutable one — and assigns a punishment to each:

Accordingly, in so far as sin consists in turning away from something, its corresponding punishment is the “pain of loss,” which also is infinite, because it is the loss of the infinite good, i.e. God. But in so far as sin turns inordinately to something, its corresponding punishment is the “pain of sense,” which is also finite.¹⁷

Read that against the popular presentation of this doctrine and the reversal is total. In the imagination formed by the sermon and the fresco, the fire is hell and the loss of God is a refinement for people who have gone soft on the subject. In the sources the ordering is the other way round, and it is not close: the loss is the infinite punishment, and the sense-pain is the finite one; the loss is what Irenaeus names when asked what hell is; the loss is what Chrysostom sets ten thousand hells against and finds them light; the loss is what the Catechism calls the chief punishment. The “spiritual” account of hell is not a modern softening of a harder original. It is the older account, and by the tradition’s own measure it is the more terrible of the two.

How firmly each is held

The two are not held with the same firmness, and the difference is exactly traceable.

The pain of loss is doctrine in the strict sense, though it is stated by implication more often than by name. Every one of the defining acts places the damned somewhere the blessed are not; Lateran IV puts them cum diabolo rather than cum Christo; the sentence of Matthew 25 that all of them quote begins Depart from me. The Catechism states it directly and twice — “the chief punishment of hell is eternal separation from God” — and the underlying claim, that God is the good in which alone the rational creature can rest, is the whole substance of the fourth part of this series. There is no theological position within Catholic orthodoxy that denies the pain of loss, and it is hard to see what one would look like.

There is also a striking indirect witness for it in the definitions themselves, which nobody reading them for the fire ever notices. Lyons and Florence define that souls departing “in mortal sin, or with original sin only” descend immediately into hell, “to be punished, however, with unequal punishments.” The second class is people who have committed no personal sin at all. Medieval theology, reading that clause, concluded that their state consists in the loss of the vision of God without any pain of sense — which is to say that the definitions themselves contemplate a condition in which the poena damni is present and the poena sensus is not, and give it the name infernus. The two punishments are therefore separable, and the Church’s own solemn language separates them.¹⁸

The pain of sense is held less firmly, and the boundary is precise. That there is one is taught: Christ speaks of fire, of the worm, of weeping and gnashing of teeth; the Quicumque says ignem aeternum; Florence says ignem aeternum; Paul VI in 1968 says “the fire that is not extinguished.” A Catholic is not free to hold that the language of fire is a rhetorical device with no referent, because the word stands inside solemn professions of faith and in the mouth of the Lord in three Gospels. What it is has never been defined at all.

The witness for that last claim is, of all people, Augustine — who is the most severe voice in the entire patristic tradition on the doctrine of hell, who wrote a whole book of the City of God defending it against objectors, and who is therefore the last author one would expect to leave a door open. Having laid out the two rival readings of the worm and the fire of Isaias 66:24 and Mark — one referring both to the spirit, as the anguish of a repentance that comes too late, the other referring the fire to the body and the worm to the soul — and having stated his own preference for referring both to the body, he stops and says this:

Let each man, then, choose which he will, whether he thinks that the worm is real and pertains to the body, or that spiritual things are meant by bodily representations, and that it belongs to the soul. But which of these is true will be more readily discovered by the facts themselves, when there shall be in the saints such knowledge as shall not require that their own experience teach them the nature of these punishments.¹⁹

That sentence should be quoted more often than it is. The Father who insisted hardest that the punishment is real, bodily, and unending is the same Father who tells his reader that the mode of it is a question on which he may make up his own mind and that the answer will arrive with the facts. Thomas Aquinas, writing eight centuries later, holds for a corporeal fire and argues for it at length — but he does so in a text that surveys four earlier opinions, attributes one of them to Augustine, and concedes that “there have been many opinions about the fire of hell.”²⁰ A doctrine that has carried four competing accounts of its central image for sixteen centuries, in authors nobody accuses of laxity, is not a doctrine whose mode has been defined.

The most recent papal statement puts the boundary in modern terms. In a general audience of 28 July 1999, John Paul II said that “to describe this reality Sacred Scripture uses a symbolical language which will gradually be explained,” that “the images of hell that Sacred Scripture presents to us must be correctly interpreted,” that “they show the complete frustration and emptiness of life without God,” and that “rather than a place, hell indicates the state of those who freely and definitively separate themselves from God, the source of all life and joy.”²¹

That address was widely reported at the time as a papal announcement that hell is empty, or that it is only a metaphor, and it says neither. Read whole, it is a compressed restatement of exactly the structure this section has been tracing: the doctrine is affirmed without qualification (“damnation remains a real possibility”); the chief punishment is identified as the definitive separation; the imagery is identified as imagery, in a way Augustine had already licensed sixteen centuries earlier; and no claim whatever is made about who is there. The pope who was accused of dissolving the doctrine was in fact stating it with unusual care about which parts of it are defined.

What follows for the way it is spoken

Two practical consequences follow from the boundary, and they cut against opposite habits.

Against the lurid habit: nobody preaching this doctrine has the Church’s authority for a description. Not for the temperature, not for the instruments, not for the sequence of torments, not for the anatomical detail that a certain kind of devotional literature has supplied in enormous quantity. Those descriptions are not doctrine; most of them are not even old; and their effect, where they are believed and then disbelieved, has been to take the doctrine down with them. A reader who discovers that the itemized tortures were nobody’s teaching quite reasonably wonders what else in the package was decoration.

Against the evacuating habit: the boundary does not run where the modern instinct wants it to. It is not the case that the fire is metaphor and the separation is the reality, in the sense that would let a reader keep the second while quietly disbelieving in any positive suffering at all. Aquinas’s article on whether the damned are tormented by fire alone answers no — not because fire is too much but because it is too little. And the pain of loss, which is the part everyone concedes as the sophisticated reading, is the part the tradition calls infinite. Anyone who reaches for the distinction as a way of making the doctrine bearable has misread which side of it is the heavy one.

The Longest Word

The eternity of hell rests, in the end, on one sentence. Everything else in the New Testament is compatible with a long punishment or a terminal one; this sentence is the place where the Church’s reading and the alternative readings actually meet. It closes the last discourse in Matthew, after the sheep and the goats have been separated and the sentence pronounced:

And these shall go into everlasting punishment: but the just, into life everlasting.²²

The two clauses carry the same adjective. In Greek it is aiōnios on both sides; in the Vulgate tradition behind the Douay–Rheims it is rendered by two English forms of one Latin word. Whatever the adjective means in the second clause it means in the first, and the sentence gives no purchase for splitting it.

That is the argument. It is not a subtle argument and it is not a modern one. Augustine had it in the early fifth century, deployed it against the identical objection, and stated it in terms nobody has improved on:

Then what a fond fancy is it to suppose that eternal punishment means long-continued punishment, while eternal life means life without end, since Christ in the very same passage spoke of both in similar terms in one and the same sentence, “These shall go away into eternal punishment, but the righteous into life eternal!” If both destinies are “eternal,” then we must either understand both as long-continued but at last terminating, or both as endless. For they are correlative,—on the one hand, punishment eternal, on the other hand, life eternal. And to say in one and the same sense, life eternal shall be endless, punishment eternal shall come to an end, is the height of absurdity.²³

It is worth pausing on how the argument works, because its structure determines what could defeat it. It is not an argument from the meaning of a word. It is an argument from a symmetry, and it is indifferent to the lexical question: it says that whatever the word means, it cannot mean two different things inside one clause of one sentence, and that a reader who terminates the punishment has undertaken to terminate the life. That is why it survives the lexical objections that follow — and it is also why those objections deserve to be stated at their full strength, since a reader who has met them only as caricature will not understand why serious people have found them serious.

The case against, at full strength

First: the adjective does not mean “endless,” and the New Testament itself proves it. The Greek aiōnios is formed from aiōn, an age or epoch, and its native force is “belonging to the age” or “age-long.” Greek possessed a stronger word for strict endlessness, aidios. The Epistle of Jude uses both, one verse apart. Of the fallen angels: God “hath reserved under darkness in everlasting chains” — desmois aidiois. Of the cities of the plain, in the next sentence: they “were made an example, suffering the punishment of eternal fire” — pyros aiōniou.²⁴ And the fire that fell on Sodom is not burning. It burned once, consumed the cities, and went out; the Dead Sea is not on fire. If aiōnios meant “never-ending” as a description of the burning, Jude’s sentence would be false as a matter of geography. The author who had aidios available and did not use it here evidently did not mean by aiōnios what the later Latin tradition made it mean.

Second: the noun in Matthew 25:46 means corrective chastisement, not retribution. The word is kolasis, from kolazō, whose root sense is to prune or check; it is distinguished in the classical vocabulary from timōria, vengeance. Aristotle states the distinction in the Rhetoric in one line: diapherei de timōria kai kolasis: hē men gar kolasis tou paschontos heneka estin, hē de timōria tou poiountos, hina plērōthēi — “there is a difference between revenge and punishment; the latter is inflicted in the interest of the sufferer, the former in the interest of him who inflicts it, that he may obtain satisfaction.”²⁵ If Matthew’s phrase means “age-long corrective chastisement” rather than “endless retributive torment,” the sentence is not merely compatible with a purgative reading; it invites one. And Aristotle is not a fringe witness: he is the standard authority on Greek usage for the period in which the Gospels were written.

Third: the strongest Pauline text says destruction, not torment. Paul’s only extended statement on the fate of the wicked runs: they “shall suffer eternal punishment in destruction, from the face of the Lord and from the glory of his power.” The Greek is olethron aiōnion — eternal ruin or destruction.²⁶ Matthew supplies a second: fear him who can “destroy both soul and body in hell” — apolesai, to destroy or lose utterly.²⁷ On the annihilationist reading these texts mean what they say: the wicked are destroyed, the destruction is permanent (and so “eternal” in the only sense that matters), and the doctrine of unending conscious torment is a later importation.

Fourth: the Old Testament image behind the fire is a rubbish-heap, not a torture chamber. The single verse that supplies the New Testament’s most vivid phrase reads: “And they shall go out, and see the carcasses of the men that have transgressed against me: their worm shall not die, and their fire shall not be quenched: and they shall be a loathsome sight to all flesh.”²⁸ What is unquenched there is the fire, and what does not die is the worm; the objects on which they work are corpses. The undying worm is the permanence of the disgrace, not the permanence of a sufferer.

Fifth: the Apocalypse’s eternity attaches to the smoke. “And the smoke of their torments shall ascend up for ever and ever” — kai ho kapnos tou basanismou autōn eis aiōnas aiōnōn anabainei.²⁹ A rising column of smoke is what a burnt city leaves behind; the image is of a monument, not of a duration of pain.

Sixth: the received text of the strongest Marcan passage is doubtful. Augustine’s argument in City of God XXI.9, quoted in the previous section, leans on the fact that the Lord “did not shrink from using the same words three times over in one passage”: the refrain “where their worm dieth not, and the fire is not extinguished” appears three times in the Douay–Rheims text of Mark 9. In the modern critical text it appears once. The first two occurrences — verses 44 and 46 in the Vulgate numbering — are marked as omissions in the Nestle–Aland text and are generally regarded as assimilation to the surviving third.³⁰ The rhetorical force Augustine appealed to belongs to a text he inherited rather than to the words of the Lord as the best witnesses transmit them.

Seventh: another whole body of texts points the other way. “And when all things shall be subdued unto him, then the Son also himself shall be subject unto him that put all things under him, that God may be all in all.” “For God hath concluded all in unbelief, that he may have mercy on all.” “As by the offence of one, unto all men to condemnation: so also by the justice of one, unto all men to justification of life.” “And through him to reconcile all things unto himself, making peace through the blood of his cross.” “The times of the restitution of all things.” “Who will have all men to be saved and to come to the knowledge of the truth.”³¹ Every one of those is Scripture, and the reading that makes them all express nothing but an unrealized wish is not obviously the natural one.

The case for, given the case against

None of the seven is frivolous. Two of them are, in this essay’s judgment, straightforwardly correct as far as they go, and saying so costs the doctrine nothing.

The fourth is correct: the Isaian image is of corpses, and the eternity in Isaias is the eternity of the worm and the fire rather than of a conscious sufferer. What follows is not that the New Testament means the same thing but that the New Testament is doing something with the image, which is what quotation of the prophets normally does. The sixth is correct: Augustine’s appeal to the threefold repetition is weakened by the text-critical evidence, and a reader who wants to argue from Mark should argue from 9:47 alone, which stands in all witnesses and is a direct quotation of Isaias.

The others do not hold, and it is worth being exact about why, because the reasons are not the ones usually given.

The Jude objection is the sharpest, and its answer is in the sentence itself. Jude does not say that Sodom’s fire burns for ever. He says that the cities prokeintai deigma — are set forth as an exhibit, a specimen — “suffering the punishment of eternal fire.” The historical burning is the exhibit; the eternal fire is what it is an exhibit of. Read that way the sentence is about the fire of the judgment and not about the geology of the Dead Sea, and the choice of aiōnios over aidios carries no weight, because the two words are used of different things: strict endlessness of the angels’ chains, and the fire of the age to come for the cities’ example. This is the ordinary reading and it is not special pleading; but it is a reading, and a reader should know that it is one.

The kolasis objection proves less than it is asked to prove. Aristotle’s distinction is real and is stated where the objection says it is stated. But it is Aristotle’s technical usage in a handbook of persuasion, offered as a clarification precisely because ordinary speakers ran the two words together; and by the first century the corrective force of kolasis had worn thin enough that the New Testament can use the noun for the pain that accompanies fear, where nothing corrective is in view.³² More decisively, the objection is aimed at the wrong target. Grant kolasis its corrective sense in full: the adjective still governs both members of Matthew 25:46, and a corrective chastisement that is aiōnios in the same sense as the life is a chastisement that never finishes correcting — which is not an improvement.

The olethros objection reads three words and stops before the fourth. Paul does not write “eternal destruction” and leave it there; he writes eternal destruction apo prosōpou tou kyriou, “from the face of the Lord and from the glory of his power.” A thing that has been annihilated is not from anything; “from the face of the Lord” names a relation, and a relation requires two terms. The phrase describes exclusion, which is exactly the poena damni of the previous section, and it is difficult to construe as extinction without deleting the prepositional phrase that gives it its content.

The Apocalypse objection stops one clause early in the same way. The verse continues: “neither have they rest day nor night, who have adored the beast and his image.” It is the smoke that ascends and the worshippers who have no rest, and the second clause has a personal subject. Six chapters later the same book says of the devil, the beast, and the false prophet that “they shall be tormented day and night for ever and ever” — basanisthēsontai hēmeras kai nyktos eis tous aiōnas tōn aiōnōn — with a plural verb and personal subjects and no smoke anywhere in the sentence.³³ This is the passage Augustine cites in XXI.23 beside Matthew 25:41, and it is why he says that in Scripture “everlasting” and “for ever” “are wont to mean nothing else than endless duration.”

The seventh objection — the universal texts — is the one that cannot be disposed of by closer reading, because the closer reading confirms that they say what they appear to say. The reply is of a different kind, and it should be given as what it is: an account of how a canon containing both sets is to be held together, not a demonstration that one set is illusory. Three points carry it. The context of the greatest of them is subjection rather than salvation: 1 Corinthians 15 is about enemies being put under Christ’s feet, and being subjected to a king is not the same as being reconciled to him. The scope of the offer is not the same as the tally of its acceptance: “who will have all men to be saved” states what God wills for all, and the whole question of the last section of this essay is what that willing amounts to when it meets a refusal. And the texts of universal reconciliation stand in the same books as the texts of final loss, often within a few pages: the Paul of Romans 11:32 is the Paul of 2 Thessalonians 1:9, and a reading that keeps one by discarding the other has not solved the problem but chosen a side of it.

What the definitions do to the argument

For a Catholic reader the exegetical dispute is not open in the way the foregoing might suggest, and pretending otherwise would be its own kind of dishonesty. The Church has determined the sense: the Quicumque, the canon of 543, Firmiter, Cantate Domino, and Paul VI all read the texts one way, and the last of them does so in 1968, after every one of the seven objections had been in print for a century or more. The doctrine of the eternity of hell does not rest on the lexicography of aiōnios and could not, because no doctrine rests on lexicography. It rests on the Church’s reading of the whole, exercised as an act of authority, and the exegetical arguments are the reasons that reading gives for itself rather than the ground on which it stands.

But that is a claim about authority, not a claim that the objections are stupid, and the distinction matters practically. A reader who has been told that the alternative readings are the inventions of people who wanted an easier religion will one day encounter Jude 7 and discover that it is not so, and the discovery will cost him more than the objection was worth. It is better to say plainly: the case against is real, parts of it are correct, the whole of it has been available to the Church for sixteen centuries, and she has read the texts otherwise.

Project synthesis — an observation about the form of the definitions

Something in the acts themselves suggests that their drafters had noticed exactly where the argument lies. Matthew 25:46 states the two destinies in one sentence under one adjective. The Quicumque states them in one sentence under one adjective: in vitam aeternam… in ignem aeternum. Firmiter states them in one clause with the two datives balanced: illi cum diabolo poenam perpetuam, et isti cum Christo gloriam sempiternam. Paul VI in 1968 states them in one sentence with the participles balanced. Four acts across fifteen centuries reproduce the grammatical symmetry of the verse rather than paraphrasing it, and each of them could have said simply that the punishment is everlasting without saying it in a construction that couples it to the life. This is an observation about form and it is offered as nothing more; drafters imitate their sources for many reasons, and the pattern is consistent with mere quotation. But it is at least suggestive that the Church’s own formulations have preserved the one feature of the text on which the whole weight of the argument falls.

The Fathers, Named Accurately

Two claims are made about the patristic evidence on hell, and both are false. The first is that the Fathers taught the eternity of punishment with unanimity, so that the doctrine can be read off the second century like a barcode. The second, which has become more fashionable, is that the early Church was substantially universalist and that the doctrine of endless punishment is a later Latin hardening. Each claim is made by people who have read a selection. The actual range is wider than either, and the honest way to report it is to name the positions and the men who held them, without suppressing the ones that embarrass a case and without inflating the ones that support it.

The second century: the doctrine, and the objection to it, arrive together

Justin Martyr, writing to the Roman senate around 155–160, states the doctrine flatly: those who have committed sins “will justly suffer in eternal fire the punishment of whatever sins they have committed,” and the demons, “having been shut up in eternal fire, shall suffer their just punishment and penalty.”³⁴ What is more interesting is the chapter that follows, and the fact that it was needed. Its heading in the received text is “Eternal punishment not a mere threat,” and it opens:

And that no one may say what is said by those who are deemed philosophers, that our assertions that the wicked are punished in eternal fire are big words and bugbears, and that we wish men to live virtuously through fear, and not because such a life is good and pleasant; I will briefly reply to this…³⁵

Within a century of the crucifixion, then, the doctrine is being taught as Christian teaching, and it is already being accused of being a scare story invented to make people behave. The objection is not modern. Neither is the answer, which is that a doctrine’s usefulness for frightening people is not evidence about whether it is true.

Irenaeus, a generation later, holds both the eternity and the account of hell as self-chosen separation that the previous section quoted in full. He is worth naming twice in this connection, because he is the counter-example to the assumption that the “severe” and the “spiritual” readings of hell belong to different parties. The earliest Father to describe hell as the loss of God rather than as an inflicted torment is also one of the earliest to say that the loss is without end, and he gives the reason: the goods lost are eternal, and therefore so is the losing of them.

Origen, read as he asked to be read

Origen of Alexandria wrote De principiis in the 220s, and its sixth chapter, “On the End or Consummation,” is the fountainhead of every Christian universalism since. Before quoting what it says, it is necessary to quote how it says it, because on this point Origen has been more consistently misrepresented than any other Father:

These subjects, indeed, are treated by us with great solicitude and caution, in the manner rather of an investigation and discussion, than in that of fixed and certain decision. For we have pointed out in the preceding pages those questions which must be set forth in clear dogmatic propositions, as I think has been done to the best of my ability when speaking of the Trinity. But on the present occasion our exercise is to be conducted, as we best may, in the style of a disputation rather than of strict definition.³⁶

That is the author’s own statement, at the head of the chapter and before its content, that what follows is exploratory and not doctrinal proposition. He has just distinguished it from what he did with the Trinity. A man who explicitly marks the difference between what he holds and what he is trying out is entitled to have the marking respected, and the tradition’s habit of citing him as though he had promulgated a system is a failure of reading, whether the citer is condemning him or recruiting him.

With that established, here is the content. Origen argues from 1 Corinthians 15 that the end must resemble the beginning; that Christ’s enemies being “put under his feet” names the same subjection by which the apostles and saints are subject to him, which is to say a saving subjection; and that “the goodness of God, through His Christ, may recall all His creatures to one end, even His enemies being conquered and subdued.”³⁷ In the tenth chapter of the second book he offers an account of the fire that is the direct ancestor of every psychological reading of hell since:

By these words it seems to be indicated that every sinner kindles for himself the flame of his own fire, and is not plunged into some fire which has been already kindled by another, or was in existence before himself… when the mind itself, or conscience, receiving by divine power into the memory all those things of which it had stamped on itself certain signs and forms at the moment of sinning, will see a kind of history, as it were, of all the foul, and shameful, and unholy deeds which it has done, exposed before its eyes: then is the conscience itself harassed, and, pierced by its own goads, becomes an accuser and a witness against itself.³⁸

Now the textual problem, which is not a footnote to the Origen question but most of it. De principiis does not survive in Greek. It survives complete only in the Latin translation Rufinus of Aquileia made around 397, and Rufinus states his editorial method in a prologue that is worth reading before any sentence is attributed to Origen on its strength. He praises Jerome for having, in translating Origen’s homilies, “so smoothed and corrected them in his translation, that a Latin reader would meet with nothing which could appear discordant with our belief,” announces that he follows the same rule, and then:

If, therefore, we have found anywhere in his writings, any statement opposed to that view, which elsewhere in his works he had himself piously laid down regarding the Trinity, we have either omitted it, as being corrupt, and not the composition of Origen, or we have brought it forward agreeably to the rule which we frequently find affirmed by himself… we have, in order that the passage might be clearer, added what we had read more fully stated on the same subject in his other works.³⁹

The translator, by his own account, omitted what he judged corrupt and supplied from elsewhere what he judged obscure. He did this specifically on doctrinal grounds. Meanwhile the Greek of De principiis survives only in fragments, and a substantial part of those fragments comes down through authors quoting Origen in order to condemn him — Jerome after his change of mind, and the emperor Justinian, whose Book against Origen of 543 supplied the very canons that anathematized the doctrine. The result is a genuine circle. Our two principal witnesses to what Origen wrote are a translator who removed the objectionable parts and a prosecutor who collected them.

That is not a reason to acquit Origen; the canons of 543 anathematize a proposition, not a man’s biography, and the proposition is condemned whoever held it. It is a reason for exactness. What can be said with confidence is that the proposition “the punishment of demons and impious men is temporary and will one day end” is condemned, and that a version of it is in the Latin De principiis. What cannot be said with the same confidence is exactly how Origen held it, in what Greek words, or with how much of the tentativeness his own preface claims.

Gregory of Nyssa, whom the Church has never censured

The Origen case is complicated by transmission. The next case is not complicated at all, and it is the one that a reader is most likely never to have been told.

Gregory of Nyssa, brother of Basil the Great, bishop, one of the three Cappadocians, whose orthodoxy on the Trinity is one of the foundations of the Church’s own language about God, wrote a catechetical handbook in the 380s. Its twenty-sixth chapter argues that the Incarnation defeated the devil by a stratagem that was just, because it returned to the deceiver his own deception, and salutary, because its aim was healing. Then it says this:

In like manner, when, after long periods of time, the evil of our nature, which now is mixed up with it and has grown with its growth, has been expelled, and when there has been a restoration of those who are now lying in Sin to their primal state, a harmony of thanksgiving will arise from all creation, as well from those who in the process of the purgation have suffered chastisement, as from those who needed not any purgation at all… He accomplished all the results before mentioned, freeing both man from evil, and healing even the introducer of evil himself. For the chastisement, however painful, of moral disease is a healing of its weakness.⁴⁰

“Healing even the introducer of evil himself” is the proposition that canon 9 anathematizes, stated by a canonized Father and doctor of the Church, whose feast is in the calendar, who was never censured for it in his lifetime or afterwards, and whom the Second Council of Nicaea honoured. The editors of the nineteenth-century translation record, at exactly this sentence, a marginal note found in one of the manuscripts collated by Krabinger: “This must be read with caution: it seems to savour of Origen’s opinion.”⁴¹ A scribe, somewhere in the transmission, felt the difficulty and wrote it in the margin instead of deleting the text.

The right way to say what this shows is narrow and should not be widened. It does not show that the doctrine of eternal punishment is doubtful; the Church has defined it, and a Father’s opinion does not compete with a definition. It does not show that Gregory was a heretic; a position held before the Church has ruled on it is not the same act as a position held after. What it shows is how doctrine actually comes to be settled — that the question was genuinely open enough, in the fourth century, for a bishop of unimpeachable standing to take the losing side of it in a work of catechesis, and that the Church, when she closed the question, closed it against the proposition without turning on the man. Both halves of that are facts. Reporting only the first produces a history in which the doctrine was never seriously contested; reporting only the second produces a history in which the Church’s own saints are on the other side. Neither is the history.

Augustine, and the other extreme

The most severe treatment of hell in the patristic corpus is Book XXI of the City of God, and it contains, incidentally, the best evidence available for how widespread the merciful reading was among ordinary Christians in the fifth century. Augustine opens his refutation like this:

I must now, I see, enter the lists of amicable controversy with those tender-hearted Christians who decline to believe that any, or that all of those whom the infallibly just Judge may pronounce worthy of the punishment of hell, shall suffer eternally, and who suppose that they shall be delivered after a fixed term of punishment, longer or shorter according to the amount of each man’s sin.⁴²

“Amicable controversy” with “tender-hearted Christians” is not the language of heresy-hunting, and the chapters that follow argue with these people rather than about them. In the chapters immediately after, Augustine catalogues further mitigating opinions he has encountered — that the intercession of the saints will empty hell at the judgment; that all the baptized will be spared; that all Catholics who persevere in the Church will be spared whatever their lives — and answers each in turn. The picture that emerges from Book XXI is not of a Church united in severity with a few Origenist deviants outside it. It is of a Church in which the merciful reading was common enough among the faithful that the greatest theologian of the age spent a book of his greatest work answering four versions of it.

And Augustine’s own answer overshoots in the other direction, into a position the Church has likewise never made her own:

But many more are left under punishment than are delivered from it, in order that it may thus be shown what was due to all.⁴³

That sentence is not doctrine. It is not in any of the eight acts; it is not in the Catechism; it is not in Trent; it appears nowhere in the magisterium in any form. It is Augustine’s own inference from his own account of original sin, and its influence on Latin preaching for a millennium is one of the principal reasons the popular presentation of this doctrine outran its sources. A Catholic is not obliged to hold it, and the Church, having had fifteen hundred years and dozens of opportunities to endorse it, never has.

What the consent of the Fathers amounts to here

The rule that a doctrine attested by the unanimous consent of the Fathers is thereby certain is a rule about agreement, not an author count, and it is worth stating precisely where the agreement lies on this subject and where it does not.

The pattern is the one this essay has met at every stage. Where the tradition is firm it is very firm, and modern readers underestimate it. Where it is open it is genuinely open, and modern readers, having been told it was closed, tend to discover the openness in the worst possible way — by finding a saint on the wrong side of it and concluding that the whole thing was improvised.

Reading Thomas Directly

Anyone who wants to know what Thomas Aquinas taught about hell is sent to the same place: questions 97 to 99 of the Supplement to the Third Part of the Summa theologiae, on the punishment of the damned, the will and intellect of the damned, and God’s mercy and justice toward them. Those questions are the source of most of what circulates as “Aquinas on hell,” including the material that most embarrasses his modern admirers. Before reading them it is worth knowing what they are, because they are not what the label says.

The Supplement is not the Summa

Aquinas died in 1274 with the Summa unfinished; he had stopped in the middle of the Third Part, at the sacrament of penance. What is printed as the Supplement was assembled after his death by compiling material from his much earlier commentary on the Sentences of Peter Lombard, written at Paris in the early 1250s when he was in his late twenties. The compilation is competent and faithful; it is also, in the strict sense, not a text Aquinas wrote for the Summa, and it represents his mind of twenty years earlier.

The correspondence can be checked rather than taken on report, and this essay checked it. Supplement question 99, article 1, asks whether by divine justice an eternal punishment is inflicted on sinners; it carries six objections, three arguments in the sed contra, a corpus giving one principal reason and four further reasons “assigned by the saints,” and six replies. Book IV, distinction 46, question 1, article 3 of the Scriptum super Sententiis is titled Utrum ex divina justitia inferatur peccatoribus poena aeterna — whether an eternal punishment is inflicted on sinners by divine justice — and carries six objections, three arguments in the sed contra, a corpus giving one principal reason and four further reasons a sanctis assignatae, and six replies, in the same order, with the same content.⁴⁴ The Supplement’s article is that article.

Nothing scandalous follows. Aquinas did not repudiate the position, and the mature Summa treats the same question in the First Part of the Second Part, where he had reached it before he died. But two things do follow, and they matter for how the material should be used. First, when the Supplement gives a school opinion of the 1250s — on the location of hell, on the fire’s physics, on what the blessed feel — it is giving the settled furniture of a thirteenth-century Paris classroom, and it should be read as that rather than as the considered judgment of the author of the Summa. Second, where the mature Summa covers the same ground, the mature text is the one to read. On the eternity of punishment it does.

The mature argument: irreparability

Summa theologiae I-II, question 87, article 3, asks whether any sin incurs a debt of eternal punishment, and its corpus is one paragraph. It does not argue from the infinity of the offended majesty. It does not argue from the sinner’s hypothetical wish to sin for ever. It argues from a structural feature of the disorder itself:

Sin incurs a debt of punishment through disturbing an order. But the effect remains so long as the cause remains. Wherefore so long as the disturbance of the order remains the debt of punishment must needs remain also. Now disturbance of an order is sometimes reparable, sometimes irreparable: because a defect which destroys the principle is irreparable, whereas if the principle be saved, defects can be repaired by virtue of that principle. For instance, if the principle of sight be destroyed, sight cannot be restored except by Divine power; whereas, if the principle of sight be preserved, while there arise certain impediments to the use of sight, these can be remedied by nature or by art… Consequently if a sin destroys the principle of the order whereby man’s will is subject to God, the disorder will be such as to be considered in itself, irreparable, although it is possible to repair it by the power of God. Now the principle of this order is the last end, to which man adheres by charity. Therefore whatever sins turn man away from God, so as to destroy charity, considered in themselves, incur a debt of eternal punishment.⁴⁵

The clause that carries the whole weight is the concessive one: irreparable considered in itself, “although it is possible to repair it by the power of God.” Aquinas is not saying that mortal sin cannot be forgiven. He is saying that it is not the sort of damage that heals from within, because what it destroys is the principle by which healing would proceed — as a blinded eye does not recover by exercise. Everything that is true in the modern formulation of hell as “definitive self-exclusion” is already in that analysis, and so is the reason the formulation is not a softening: an exclusion that could reverse itself would not need excluding.

Two of the replies are worth having. To the objection that punishment should be proportioned to fault and fault is temporal, he answers with Augustine’s argument from the laws — adultery committed in a moment is not punished with a momentary penalty — and then adds Gregory the Great’s formula, that “it is just that he who has sinned against God in his own eternity should be punished in God’s eternity,” glossing it: a man sins in his own eternity not only by sinning throughout his life but because “from the very fact that he fixes his end in sin, he has the will to sin, everlastingly.”⁴⁶ To the objection that punishment is medicinal and eternal punishment can cure nobody, he answers that punishments are not always medicinal for the one punished, and that “the eternal punishments inflicted by God on the reprobate are medicinal punishments for those who refrain from sin through the thought of those punishments.” The damned are not being corrected; the living are.

That is a hard doctrine, and it should be stated as a hard doctrine rather than smoothed. It is also the point at which the Supplement’s parallel article says something that almost nobody who quotes Aquinas on hell has ever quoted. Asked whether by God’s mercy all punishment of men and demons comes to an end, and answering no, he adds:

God, for His own part, has mercy on all. Since, however, His mercy is ruled by the order of His wisdom, the result is that it does not reach to certain people who render themselves unworthy of that mercy, as do the demons and the damned who are obstinate in wickedness. And yet we may say that even in them His mercy finds a place, in so far as they are punished less than they deserve condignly, but not that they are entirely delivered from punishment.⁴⁷

Minus puniuntur quam condigne merentur: they are punished less than they strictly deserve. On the same page, the same author answers the objection from Anselm — that it is unjust for God to permit the utter loss of a creature made for happiness — by conceding a version of it: it is unbecoming to the divine goodness “that a whole genus of creature fail of the end for which it was made: wherefore it is unbecoming for all men or all angels to be damned. But there is no reason why some men or some angels should perish for ever, because the intention of the Divine will is fulfilled in the others who are saved.”⁴⁸ A reader who has been told that scholastic theology took a grim satisfaction in the arithmetic of damnation should sit with those two replies. They are in the same question as the severest statements, by the same author, and they say that mercy operates inside hell and that the damnation of the whole species would be unfitting.

Why the will does not turn back

The account of hell as self-exclusion needs an answer to an obvious question: if the loss is so terrible and the deprivation is felt, why does the damned will not simply change its mind? Aquinas answers in two places, and the answers are of quite different weight.

The angelic case he treats in the First Part, and he treats it as certain, though for a reason worth noticing — he is arguing not from a definition but from what beatitude would have to be: “everlasting stability is of the very nature of true beatitude; hence it is termed ‘life everlasting.’” Origen’s opinion that every created will remains indefinitely reversible “deprives angels and saints of true beatitude,” and is also contrary to Scripture; so “according to Catholic Faith, it must be held firmly both that the will of the good angels is confirmed in good, and that the will of the demons is obstinate in evil.”⁴⁹

The account of what the damned will actually want, by contrast, comes from the Supplement, is school theology, and should be labelled as such. It is nonetheless the most psychologically acute thing in the dossier. Do the damned repent? Yes and no, and the distinction is exact: “he repents of a sin directly who hates sin as such: and he repents indirectly who hates it on account of something connected with it, for instance punishment. Accordingly the wicked will not repent of their sins directly, because consent in the malice of sin will remain in them; but they will repent indirectly, inasmuch as they will suffer from the punishment.”⁵⁰ And do they hate God? Not God as he is in himself, “since He is goodness by His essence,” and “whoever sees Him in His essence cannot hate Him”; but God as met in the effect that afflicts them — “the damned, perceiving God in His punishment, which is the effect of His justice, hate Him, even as they hate the punishment inflicted on them.”⁵¹

Put those two together and the picture is not of a will held down by force. It is of a will that wants the thing it chose, resents the consequence, and has never seen the good it refused — which is why it can go on refusing. Whether that picture is true is not something an essay can settle; it is theological reconstruction, offered as such by its author, and no act of the magisterium has adopted it. But it is the reconstruction that stands behind the Catechism’s phrase “definitive self-exclusion,” and a reader who wants to know where that phrase comes from should know that it comes from here.

Where Thomas is opinion, and should be labelled

Three positions in these questions are commonly cited as “the Catholic teaching” and are nothing of the kind. Reporting them is not an attack on Aquinas; it is the ordinary work of distinguishing a doctor from a definition.

The fire is corporeal, of a species different from ours, and located beneath the earth. The Supplement holds all three. The article on the fire opens by surveying the contrary opinions at length — Avicenna’s, and a reading of Augustine on punishment by images of bodies — before concluding that at least the fire tormenting risen bodies must be corporeal; the article on its location is a piece of ancient cosmology and should be read as one.⁵²

The blessed rejoice in the punishment of the damned. This is the article most often quoted against Aquinas, usually without its distinction, and the distinction is the whole content of the article. He denies the direct form outright: “a thing may be a matter of rejoicing in two ways. First directly, when one rejoices in a thing as such: and thus the saints will not rejoice in the punishment of the wicked.” What he affirms is indirect: they rejoice “by considering therein the order of Divine justice and their own deliverance,” so that “the Divine justice and their own deliverance will be the direct cause of the joy of the blessed: while the punishment of the damned will cause it indirectly.”⁵³ Even so stated, this is common medieval opinion and no more. It appears in no conciliar act, no papal definition, and no paragraph of the Catechism, and a Catholic is entirely free to think that the preceding article — whether the blessed pity the damned — was answered too quickly.

Many are damned. Aquinas inherits Augustine’s proportion in places and does not make it a conclusion of the Summa; and in any case the point stands where the last section left it. No magisterial act has ever taught a proportion, and the next section takes up why.

The general rule for using this material is simple and can be stated in one sentence. Where Aquinas is expounding what the Church has defined, he is a witness of great weight to a doctrine that does not depend on him; where he is reconstructing what the definitions leave open — the physics of the fire, the topography, the interior life of the damned, the emotions of the blessed — he is a theologian of the first rank offering an account, and the reader owes him the compliment of treating it as an account rather than as a decree.

The Census Nobody Has

The Church has a formal procedure for declaring that a particular person is in heaven. It is called canonization, it is used several times a year, it has been used many thousands of times, and its effect is precisely a statement about an individual’s actual condition after death. There is no corresponding procedure in the other direction. There never has been. No pope, no council, and no rite has ever declared of any human being who has ever lived that he is in hell.

That is a negative claim about a very large corpus, and it should be stated with its bounds. This essay did not read every magisterial act; what it did was read the eight acts of the Catechism’s own dossier, the paragraphs of the Catechism on hell and on mortal sin, the relevant canons of Trent, and the most recent papal statement on the subject. None of them names anyone. And the claim happens to be made in that most recent statement, by the pope, as part of the doctrine: “Damnation remains a real possibility, but it is not granted to us, without special divine revelation, to know which human beings are effectively involved in it.”⁵⁴

The rest of this section is about why the census is empty — which turns out not to be a policy of discretion but a consequence of the doctrine itself.

The one place the question is asked

Somewhere on the road to Jerusalem a man asked Jesus the question that everyone since has wanted answered, and Luke records the exchange in two verses:

And a certain man said to him: Lord, are they few that are saved? But he said to them: Strive to enter by the narrow gate: for many, I say to you, shall seek to enter and shall not be able.⁵⁵

Read the grammar. One man, singular, asks a question in the third person about other people: are the saved few? The answer is addressed to a plural audience and is an imperative in the second person: you, strive. Agōnizesthe is the verb of the wrestling floor. The question asked for a number and got a command; the question was about them and the answer was about you.

That is the only place in the Gospels where the population of the saved is put to Christ as a direct question, and the shape of the reply is the shape of the whole doctrine. It is not that the question was rebuked — there is no rebuke in the text. It is that it was answered on a different axis, and that the axis it was answered on is the only one anybody can act on.

The other numerical texts are of the same kind, and it is worth reading them for their grammatical mood rather than their arithmetic. “Enter ye in at the narrow gate: for wide is the gate, and broad is the way that leadeth to destruction, and many there are who go in thereat. How narrow is the gate, and strait is the way that leadeth to life: and few there are that find it!”⁵⁶ The sentence does contain “many” and “few,” and an honest reading does not pretend otherwise; the way to life is described as the one fewer people find. But the sentence is not a report, it is a command with a reason attached, and it is addressed to the hearer about the hearer. Neither text has ever been read by the magisterium as a demographic statement, and the Catechism, quoting the second of them in the paragraph immediately after its statement of the doctrine, glosses it as “an urgent call to conversion.”

Judas, and the limit of what can be said

If the Church were ever going to name anyone, the case is made for her. Of one man the Lord said: “But woe to that man by whom the Son of man shall be betrayed. It were better for him, if that man had not been born.”⁵⁷ Of one man he said, in the priestly prayer: “none of them is lost, but the son of perdition.”⁵⁸ Of one man the apostles said, choosing his successor, that he “hath by transgression fallen, that he might go to his own place.”⁵⁹ Three texts, of unusual severity, about a named individual.

The tradition has overwhelmingly read them as they appear. Augustine, Aquinas, and the general run of Latin theology treat Judas as lost, and a reader who finds that reading natural is in very large company. What has never happened is the next step. No council has defined it, no pope has declared it, and the Catechism, which quotes both Matthew 26 and the Gospel warnings about hell, makes no statement about Judas at all. On the one human being about whom Scripture speaks in this way, the Church’s formal teaching is silent.

Two conclusions are available from that silence and only one of them is warranted. The unwarranted one is that the Church privately doubts the doctrine and is hedging. The warranted one is that the act of declaring an individual damned is not an act she has ever understood herself to be able to perform — which is a claim not about Judas but about the kind of knowledge that would be required.

Why the knowledge is not available

The Church’s own analysis of mortal sin makes the private judgment structurally impossible, and this is the load-bearing point of the whole section. Three conditions must be met together: “Mortal sin is sin whose object is grave matter and which is also committed with full knowledge and deliberate consent.” The first condition is public — grave matter is specified by the commandments, and whether an act falls under one of them is a question anybody can examine. The second and third are not:

Unintentional ignorance can diminish or even remove the imputability of a grave offense… the promptings of feelings and passions can also diminish the voluntary and free character of the offense, as can external pressures or pathological disorders.⁶⁰

Knowledge and consent are interior. Whether a particular man knew what he was doing, and whether the assent he gave was free rather than driven by compulsion, terror, formation, or illness, is not visible from outside and is frequently not visible from inside. And even if it were, there remains the last instant, which is invisible to everyone: the doctrine is about dying unrepentant, and the Church’s phrase is always to the end. John Paul II puts the two together: hell is the state of those who “definitively reject the Father’s mercy, even at the last moment of their life.”⁶¹

The Catechism draws the practical conclusion in the same paragraph in which it states that mortal sin causes “the eternal death of hell”: “However, although we can judge that an act is in itself a grave offense, we must entrust judgment of persons to the justice and mercy of God.”⁶² That is not a counsel of tact. It is a statement about the location of a boundary: acts can be judged, persons cannot, and the doctrine of hell concerns persons.

The same limit is imposed even on self-knowledge, which is the point at which the Church’s discipline becomes genuinely surprising. Trent, having taught that the justified should place their firmest hope in God’s help, adds that concerning the gift of perseverance “let no one promise himself anything as certain with absolute certainty” — nemo sibi certi aliquid absoluta certitudine polliceatur — and directs the reader instead to Paul: let him that thinketh himself to stand take heed lest he fall, and work out his salvation with fear and trembling.⁶³ If a man cannot have absolute certainty about the one case he can see from the inside, the case for which he has continuous first-person evidence and unlimited access, then the epistemic situation with respect to anyone else is not merely difficult. It is closed.

One class about which the Church does speak

Precision requires an exception, and it is instructive. Everything above concerns human beings. About the fallen angels the Church’s language is not conditional at all.

Canon 9 of 543 anathematizes the proposition that the punishment “of demons and of impious men” will end — naming the demons first. Firmiter says the wicked receive perpetual punishment cum diabolo, with the devil. Matthew 25:41 has the fire “prepared for the devil and his angels,” in the perfect participle: it exists, and it exists for them. The Apocalypse says of the devil, the beast, and the false prophet that “they shall be tormented day and night for ever and ever.” John Paul II states it as doctrine in the same address that refuses the human census: “Christian faith teaches that in taking the risk of saying ‘yes’ or ‘no’, which marks the human creature’s freedom, some have already said no. They are the spiritual creatures that rebelled against God’s love and are called demons.”⁶⁴

So the Church does hold that hell is occupied. What she has never said is that any human being is in it. The distinction is not arbitrary: the angelic choice, on the traditional analysis, was made once, with full knowledge, by an intellect that does not deliberate as we do — which is exactly the case in which the three conditions of a mortal sin are transparently met, and exactly the case that is furthest from every human case anyone could examine. The one class about which the doctrine speaks in the indicative is the one class whose interior act is not opaque.

What follows for speech

The consequences for how the doctrine may be spoken are strict and cut in a direction that both parties in the modern quarrel dislike.

Nobody may say of any person, living or dead, that he is damned. Not of an enemy, not of a public criminal, not of a suicide, not of an apostate, not of anyone. This is not sentiment; it is the direct consequence of a rule the Catechism states and Trent enforces even against the speaker’s knowledge of himself. Preaching that consigns named individuals, or identifiable classes of individuals, to hell is not rigorous orthodoxy. It is a claim to information the Church says nobody has.

And nobody may say that the doctrine has no application. The absence of a census is not evidence of an empty hell; it is the absence of a census. The definitions are in the indicative — souls departing in mortal sin descend immediately — and the conditional under which they operate is one that the Church treats as really satisfiable, or the warning would be a pretence and the Roman Canon’s petition to be saved from final damnation would be a formality.

There is a third consequence, which cuts against a modern reflex and is worth stating because it is definitively taught. The fear of hell is not a disreputable motive. Trent anathematized the proposition that the sorrow “which is prepared by the examination, recollection, and detestation of sins,” by which a man weighs “the gravity, multitude, and foulness of his sins, the loss of eternal beatitude, and the incurring of eternal damnation,” together with a purpose of amendment, “is not a true and useful sorrow, nor prepares for grace, but makes a man a hypocrite and a greater sinner.”⁶⁵ The charge that the doctrine produces frightened hypocrites rather than converts is exactly the charge Justin Martyr was answering in the second century, and it is one of the very few objections in this whole subject that the Church has met with a formal anathema.

The Will That Wants All Saved

Everything in this essay so far has been description: what the acts say, how firmly, on what evidence. The last substantive question is not descriptive, and it is the one that makes the doctrine difficult rather than merely severe. Scripture says that God wills the salvation of all men. Scripture also says that the will of God is always accomplished. And the Church teaches that some rational creatures are, in fact, finally lost. Any two of those three are comfortable together. All three are not.

The relevant texts are not obscure or few, and they should be laid out before anything is done with them.

This is good and acceptable in the sight of God our Saviour, who will have all men to be saved and to come to the knowledge of the truth.

Is it my will that a sinner should die, saith the Lord God, and not that he should be converted from his ways, and live?… As I live, saith the Lord God, I desire not the death of the wicked, but that the wicked turn from his way, and live. Turn ye, turn ye from your evil ways: and why will you die, O house of Israel?

The Lord delayeth not his promise, as some imagine, but dealeth patiently for your sake, not willing that any should perish, but that all should return to penance.

But thou hast mercy upon all, because thou canst do all things, and overlookest the sins of men for the sake of repentance. For thou lovest all things that are, and hatest none of the things which thou hast made… But thou sparest all: because they are thine, O Lord, who lovest souls.⁶⁶

Against them stands the psalm verse Aquinas takes as his sed contra when he raises the difficulty formally: “God hath done all things, whatsoever He would.”

Three readings, offered as three

The classical treatment is Summa theologiae I, question 19, article 6, and its manner is as instructive as its content. Aquinas does not give the answer. He gives three, and attributes each:

The words of the Apostle, “God will have all men to be saved,” etc. can be understood in three ways. First, by a restricted application, in which case they would mean, as Augustine says, “God wills all men to be saved that are saved, not because there is no man whom He does not wish saved, but because there is no man saved whose salvation He does not will.” Secondly, they can be understood as applying to every class of individuals, not to every individual of each class; in which case they mean that God wills some men of every class and condition to be saved, males and females, Jews and Gentiles, great and small, but not all of every condition. Thirdly, according to Damascene, they are understood of the antecedent will of God; not of the consequent will.⁶⁷

The first two readings are restrictive: they make the verse say less than it appears to say. The first is Augustine’s, and it is uncomfortable to modern ears — it converts a universal statement into a tautology about the saved. The second is scarcely better; it makes “all men” mean “some of every kind of man.” Aquinas does not endorse either, and he does not reject them; he sets them out as available and moves to the third, which is the one the later tradition took up and which the Catechism presupposes without naming.

The antecedent and consequent will is not a division inside God, and Aquinas says so immediately: “this distinction must not be taken as applying to the divine will itself, in which there is nothing antecedent nor consequent, but to the things willed.” The point is about descriptions. A thing considered absolutely may be good, and the same thing considered with all its circumstances may not be:

Thus that a man should live is good; and that a man should be killed is evil, absolutely considered. But if in a particular case we add that a man is a murderer or dangerous to society, to kill him is a good; that he live is an evil. Hence it may be said of a just judge, that antecedently he wills all men to live; but consequently wills the murderer to be hanged. In the same way God antecedently wills all men to be saved, but consequently wills some to be damned, as His justice exacts.⁶⁸

The objection to that, at full strength

The distinction is standard, it is old, and it does resolve the formal contradiction. It also has a cost, and pretending otherwise would be the kind of smoothing this essay has been trying to avoid.

The cost is in the example. A judge who “antecedently wills all men to live” and hangs the murderer is describing a preference that has been overridden by a consideration external to it — the good of the city, the demands of the law. That is a perfectly intelligible structure for a judge. Applied to God it produces a picture in which the salvific will is a first-order preference that loses to a second-order one, and in which what is described as God’s will for a particular man ends up meaning something considerably weaker than the words of 1 Timothy suggest. Aquinas himself concedes as much when he says that the antecedent will “may be called a willingness rather than an absolute will.” The objection, put bluntly: if “God wills all men to be saved” means only that God has a defeasible inclination in favour of everyone’s salvation which is regularly defeated, then the verse has been paraphrased into something no one would have written it to say.

There is a second and sharper form of the objection, which the tradition also has to face. If God’s consequent will damns, and God’s will is the universal cause, in what sense is the damnation not simply God’s doing? Aquinas has an answer, and it is the most carefully constructed piece of the whole system:

Reprobation differs in its causality from predestination. This latter is the cause both of what is expected in the future life by the predestined—namely, glory—and of what is received in this life—namely, grace. Reprobation, however, is not the cause of what is in the present—namely, sin; but it is the cause of abandonment by God. It is the cause, however, of what is assigned in the future—namely, eternal punishment. But guilt proceeds from the free-will of the person who is reprobated and deserted by grace. In this way, the word of the prophet is true—namely, “Destruction is thy own, O Israel.”⁶⁹

The asymmetry is the point. Predestination causes both the grace and the glory; reprobation causes only the punishment, and the guilt on which the punishment falls is the creature’s. Whether that asymmetry survives pressure — whether “permitting the fall” and “causing the fall” can really be held apart when the permitter sustains the faller in being at every instant — is the oldest question in the treatise on grace, and this essay does not undertake to settle it.

What the Church has and has not decided here

She has decided one thing, twice, with unusual force. The Second Council of Orange in 529 did not merely deny that God predestines anyone to evil; it anathematized the contrary belief with a phrase of open contempt: “that some have been predestined to evil by divine power we not only do not believe, but if there are any who wish to believe so great an evil, we say to them anathema cum omni detestatione, with all detestation.”⁷⁰ The Catechism states the same doctrine flatly and adds the condition: “God predestines no one to go to hell; for this, a wilful turning away from God (a mortal sin) is necessary, and persistence in it until the end.”⁷¹

What she has not decided is the mechanism. This is not a lacuna; it is a decision not to decide, and it was made explicitly. After a decade of formal disputations between Dominican and Jesuit theologians on how efficacious grace and free will are related — the congregatio de auxiliis — Paul V closed the matter in 1607 by sending the participants home, promising a determination that never came, and forbidding either party to censure the other: it was “very seriously forbidden by the same Most Holy Lord that in the treatment of this question anyone should qualify the side opposed to his own or brand it with any censure.”⁷² Four centuries later the determination has still not come. On the precise question of how a God who wills the salvation of all is related to the loss of any, the Church’s formal position is that two incompatible schools may both be held and neither may be censured.

That is a datum of the first importance for the tone in which this doctrine may be discussed, and it is almost never mentioned. Anyone who states confidently how divine sovereignty and human refusal are reconciled has gone beyond what the Church has settled, whichever direction he is going.

The residue

Something has to be said about what remains, and it should be said without the customary flourish that makes an unresolved difficulty sound like a resolved one.

The three propositions can be held together without formal contradiction. The distinction of antecedent and consequent will, or Augustine’s restrictive reading, or the class reading, each removes the inconsistency. None of them removes the difficulty, which is not logical but — for want of a better word — personal: the doctrine asks a believer to hold at once that God desires the salvation of a particular man with the desire attested in Ezechiel, and that the man may nevertheless be lost, and that this is not a defeat. The tradition’s resources for that are real but they are not a solution; they are a set of things that can truthfully be said around a place where an answer is missing.

Benedict XVI’s handling in Spe salvi is, in this essay’s judgment, the most useful modern statement precisely because it does not claim more: “The judgement of God is hope, both because it is justice and because it is grace. If it were merely grace, making all earthly things cease to matter, God would still owe us an answer to the question about justice — the crucial question that we ask of history and of God. If it were merely justice, in the end it could bring only fear to us all.”⁷³ The two things that cannot be given up are that the way a man lives matters, and that mercy is real; a doctrine that dropped either would answer the difficulty by abolishing one of its terms.

One further clarification is needed, because the universal salvific will is regularly made to carry an inference it will not bear. That God wills all men to be saved does not entail that all men will be saved, for a reason internal to the willing itself. What is willed is the salvation of men — of creatures who are free, and whose acceptance is theirs. A salvation that arrived regardless of the creature’s assent would not be the salvation of that creature but its replacement by something else, and a freedom incapable of refusal is not the freedom the doctrine of creation says was given. This is not a proof that anyone refuses. It is the reason the possibility cannot be closed off by appeal to God’s desire, and it is why the last section of this essay is about hope rather than about certainty.

What May Be Hoped

Twentieth-century Catholic theology produced a question that will not go away and has never been settled: whether a Christian may hope that all men are in fact saved. The literature on it is large, modern, and mostly in copyright, and this essay has not read it; nothing said here is a contribution to that debate or a judgment about the positions taken in it.⁷⁴ What can be done, and what is done here, is to ask what the Church’s own acts and her own prayer actually establish about the question — because they establish considerably more than either party to the modern argument usually notices, and the thing they establish is not on either side of it.

What the Church prays

The article of the Catechism on hell ends in a summary of five short paragraphs, and the fourth of them is this:

The Church prays that no one should be lost: “Lord, let me never be parted from you.” If it is true that no one can save himself, it is also true that God “desires all men to be saved” (1 Tim 2:4), and that for him “all things are possible” (Mt 19:26).⁷⁵

Two paragraphs earlier the same summary states that the Church “warns the faithful of the sad and lamentable reality of eternal death.” The Catechism therefore does what a good deal of modern discussion assumes cannot be done: it states the warning and the prayer as two aspects of one teaching, in adjacent sentences, without any sense of strain.

The same pairing is in the Church’s central prayer. Immediately after stating that God predestines no one to hell, the Catechism quotes the Roman Canon: “Father, accept this offering from your whole family. Grant us your peace in this life, save us from final damnation, and count us among those you have chosen.”⁷⁶ And the Church’s oldest instruction about for whom she prays stands two verses before the verse everyone quotes about the universal salvific will: “I desire therefore, first of all, that supplications, prayers, intercessions and thanksgivings be made for all men.”⁷⁷

Project synthesis — what praying for an outcome presupposes

A prayer that no one be lost is intelligible only under two conditions, and it entails both of them.

It entails that the loss is possible. Nobody prays to be spared what cannot happen. If the doctrine of hell described a merely notional danger — a warning attached to nothing — then the petition in the Roman Canon would be a piece of liturgical theatre, and the Church would be teaching her people to ask for something they were never at risk of failing to receive. The prayer is evidence that she does not think so.

And it entails that the loss is not settled. Nobody prays for what is already determined either way. A Church that held the number of the lost to be fixed and known would not ask that no one be lost, any more than she asks that the Incarnation should have taken place. The petition is in the optative because its object is open.

The conjunction of those two is exactly the shape this essay has been tracing from the first page: the doctrine held at full strength, the census left empty. It is not a compromise between severity and mercy. It is the only position from which the Church’s own liturgical act makes sense. This inference is the present essay’s, not a source’s, and it establishes nothing about how many are in fact lost — only that the Church’s practice presupposes a real possibility and an unclosed question.

The two ways of getting it wrong

What that forecloses can be stated exactly, and it forecloses something on each side.

It forecloses the certainty that all are saved. Hope is not knowledge, and a hope that has hardened into a conclusion has stopped being hope and started being a prediction. The definitions are in the indicative and their condition is one the Church treats as really satisfiable; the fallen angels are not a hypothesis; the warnings are addressed to people who are told, in the imperative, to do something about their situation. A reader who leaves this essay believing that the doctrine of hell is a formality that no longer applies to anybody has read it against its evidence.

And it forecloses the certainty that any particular person is lost. That was the subject of the section on the census, and it does not need repeating except to note that the two prohibitions have the same root. Both rest on the fact that the interior act which alone would settle a case is not available for inspection — not another’s, and, Trent insists, not one’s own.

Between them lies the position the Church actually occupies, which has a name in her own vocabulary: the theological virtue of hope, which is neither the presumption that the end is secured nor the despair that it is forfeited, and which is a virtue precisely because its object is uncertain to the one who holds it and certain only in the one it leans on.

How the doctrine is addressed

There is one formal feature of the New Testament’s hell texts that this essay noticed late and that seems to it decisive for how the doctrine may be used. Almost without exception, they are in the second person and the imperative mood.

“Enter ye in at the narrow gate.” “Strive to enter by the narrow gate.” “If thy right eye scandalize thee, pluck it out.” “Fear ye not them that kill the body… but rather fear him that can destroy both soul and body in hell.” The parable of the sheep and the goats is third-person narrative, but it is told to disciples about what will be asked of them, and its content is a list of things to do. The one time a bystander converts the subject to the third person and asks for a number, the answer converts it back.⁷⁸

That is not a device for softening the doctrine. Second-person address is more alarming than third-person description, not less; a warning about you is worse than a statistic about them. But it does explain both the shape of what the Church has defined and the shape of what she has refused to define. Everything the texts do is aimed at a hearer’s decision. Nothing they do requires him to know the mode of the fire, the location of hell, or the number of its inhabitants; and everything the Church has defined is what is needed to make the address intelligible — that the state is real, that it begins at death, that it does not end, and that God does not put anyone there who has not gone.

Project synthesis — the judgment this article reaches

Judgment. The doctrine of hell as the Church has actually defined it is narrow, severe, and not revisable at its core; and the wide territory around that core — the mode of the punishment, the topography, the psychology, the number, and above all the identity of any individual — is not defined at all, has never been defined, and in the case of individuals is undefinable in principle from the Church’s own analysis of what a mortal sin is. The emptiness of the human census is not diplomatic silence and not a modern softening. It is the direct consequence of a doctrine that makes the damning act consist in something interior, free, and final, none of which is available for inspection from outside a soul.

Reasons. (1) The Catechism’s own dossier for the existence and eternity of hell is eight items, and when they are read at their own loci in their own languages they assert five things — that the state exists, that it is entered immediately at death, that its punishments are unequal, that it does not end, and that God predestines no one to it — and nothing else. (2) The tradition’s own ranking of the two punishments is the reverse of the popular one: the pain of loss is the infinite one in Aquinas’s analysis, the earliest sustained patristic account of hell describes only the loss, and the Catechism calls it the chief punishment; the mode of the pain of sense has never been determined, and Augustine, the severest witness, expressly leaves the reader free about it. (3) The argument for eternity that has actually carried the tradition is not lexical but grammatical — the correlativity of Matthew 25:46 — and it survives the lexical objections, some of which are nevertheless correct as far as they go. (4) The patristic range genuinely includes a canonized Father teaching the healing of the devil and another teaching that most of the human race is lost, and the Church defined a point that excludes the first and has never adopted the second. (5) Mortal sin requires grave matter, full knowledge, and deliberate consent; only the first is public, and the Catechism accordingly directs that judgment of persons be entrusted to God, while Trent forbids absolute certainty even about oneself. (6) The Church nevertheless does speak in the indicative about the fallen angels, which shows that the silence about human beings is principled rather than squeamish: it falls exactly where the interior act is opaque. (7) On the mechanism relating the universal salvific will to the loss of any, Paul V in 1607 formally forbade either school to censure the other, and no determination has followed.

Strongest counterargument. That an audit of this kind, however scrupulous, functions in practice as an evacuation. A doctrine about which nothing may be said concerning the mode, nothing concerning the number, and nothing concerning any person is a doctrine that cannot be preached; and a doctrine that cannot be preached is dead within two generations whatever its formal standing. The tradition’s concreteness was not carelessness. It was a recognition that a warning has to be vivid to function as a warning, and the vividness is not an accretion — the fire, the worm, the outer darkness and the gnashing of teeth are in the mouth of Christ, not in a baroque fresco. An essay that translates all of that into the boundary between the defined and the undefined has said nothing false and produced something no one will ever be afraid of, which is precisely what those words were given for. There is a second edge to the same objection: the argument from the empty census proves less than it appears to, because the Church has also never declared of most particular people that they are in heaven, and nobody infers from that silence that heaven’s population is unknowable in principle.

Reply, and what it concedes. The reply to the second edge is that the cases are not parallel: the Church possesses a procedure for the positive declaration and uses it constantly, so the absence of the negative is a fact about what she takes herself to be able to do, not about her interest or her opportunity. The reply to the first edge concedes most of it. This essay is an audit and not a sermon, and an audit is not a substitute for preaching. What the audit removes is narrower than the objection supposes: not the images, which are Christ’s and remain available to anyone who preaches them as Christ used them, but the claim to teach as the Church what the Church has not taught. A preacher who says “the Lord said, where their worm dieth not” has the whole tradition behind him. A preacher who describes the temperature, or estimates the proportion, or names a man, is not being more faithful than the audit; he is saying something the Church has never said, and the discovery of that, when it comes, has historically cost more than the vividness was worth.

What would change this judgment. Four findings, each specific and each checkable. (1) A magisterial act, of any weight, declaring of a named human being that he is damned. That would refute the central negative claim of this essay directly, and the essay’s reading of why the census is empty would have to be abandoned. (2) Documentary evidence that the fifteen Origenist anathemas, or the nine canons of 543, were promulgated in the official sessions of the Second Council of Constantinople. That would not change the doctrine but would materially strengthen the dossier and require the correction of this essay’s account of it. (3) A definition or authoritative determination of the mode of the pain of sense — that the fire is corporeal, or that it is not. That would move a large block of what this essay classes as open into the closed column and would falsify its claim that Augustine’s permission still stands. (4) Evidence that the Catechism’s “the chief punishment of hell is eternal separation from God” was intended and received as more than catechetical synthesis — that it restates a determination made elsewhere. This essay found no such determination and treats the sentence as authoritative teaching rather than as a definition; a source showing otherwise would strengthen the doctrine and weaken the essay’s classification.

The end of this part

The essay on death in this series ends with a closed door: the choice of a life is finished when the life is. The essay on judgment ends with a verdict on what was behind it. This one has been about what the verdict can be when it goes the other way, and it has found, in the end, a doctrine much smaller and much harder than the one that is usually described — smaller, because almost none of the furniture belongs to it; harder, because what does belong to it is not negotiable and has never been withdrawn.

There is no consoling last sentence available here, and manufacturing one would be a lie about the sources. What is available is the thing the Church actually does with the doctrine, which is neither to describe it nor to explain it away but to pray about it, in the plural, at the altar, in the middle of the offering: save us from final damnation, and count us among those you have chosen. That petition presupposes everything this essay has argued — that the danger is real, that no one’s case is closed, that the outcome is not in the petitioner’s gift. It is also, as it happens, the only form in which anybody has ever been given the subject to handle. The fourth part of this series is about what is being asked for.

Scope, Method, and Qualifications

Reader, question, and method

This is a discursive theological article for a serious general reader, able to follow careful distinctions but not presumed to know scholastic terminology, Greek, or Latin. It is the third of four independent essays on the Four Last Things — death, judgment, hell, heaven — each complete in itself. It refers to the others by subject and inherits from them only two claims, both of which they establish on their own evidence: that the moral choice of a life is closed by death, and that the closed choice is met by a judgment. No text, section, or apparatus is shared between the four; each has its own sources, records, and generation metadata.

The governing question is what the Church has actually defined about hell, by which acts, with what force, as against what is common theological opinion or devotional imagery. The claim reached is stated in full in the terminal “Project synthesis” block of the last section and is summarized here: the core is narrow, severe, and unrevisable; the surrounding territory is undefined; and the absence of any declaration that a particular human being is damned follows from the Church’s own analysis of what a mortal sin requires, rather than from discretion.

The argument proceeds by audit. It takes the eight references in the Catechism’s own footnote to its statement that “the teaching of the Church affirms the existence of hell and its eternity” (CCC 1035, note 615), reads each at its own locus in the language in which it was promulgated so far as that language was reachable, and reports the aggregate. It then treats in order: the distinction of the pain of loss and the pain of sense and how firmly each is held; the eternity of punishment, with the principal contrary readings stated at full strength before any reply; the patristic range, including positions later excluded, named with the men who held them; the Thomistic treatment read at its own loci with the status of the Supplement established by direct comparison with the Scriptum super Sententiis; the question of who is in fact damned; the universal salvific will; and what the Church’s own prayer presupposes.

Claims are classified by weight throughout, in the terms set out in the opening section: definitive teaching; authoritative but non-definitive teaching; common theological opinion; disputed opinion; and project synthesis. Where a source is used at a weight lower than the one usually attributed to it — as with the canons of 543, or Trent’s canon 25 — the essay says so at that point rather than in this appendix.

Included and excluded scope

Included are: the documentary basis of the doctrine of hell’s existence and eternity as the Catechism itself identifies it; the distinction of poena damni and poena sensus and the differing firmness with which each is held; the eternity of punishment, its principal scriptural loci, and seven contrary readings stated at full strength with replies; the patristic range from Justin to Augustine, including positions later excluded; the mature Thomistic account of the reason for eternity and the status of the Supplement; the question whether the Church has ever declared an individual damned, and why not; the universal salvific will, the antecedent and consequent will, reprobation, and the Church’s refusal to settle the mechanism; and what the Church’s liturgical prayer presupposes.

Excluded are: any account of death, of the particular or general judgment, or of heaven beyond what a sentence of transition requires, those being the subjects of the other three essays; purgatory, except where a conciliar text distinguishes it from hell in the same clause; the state of those dying with original sin only, beyond noting that the conciliar clause distinguishes classes of punishment and that CCC 1261 leaves the question to God’s mercy; the theology of the resurrection of the body; the doctrine of grace, predestination, and perseverance beyond the loci needed for the salvific-will question; the entire modern controversy over the hope for universal salvation, whose literature was not read; private revelations and visionary descriptions of hell, of any period, none of which is used or assessed here; the exegesis of the descent of Christ to the dead; the history of the doctrine in the Eastern churches and in the Reformation confessions; annihilationism and conditional immortality as systematic positions, as against the scriptural arguments for them, which are stated; and any judgment about the state, culpability, or destiny of any person.

Material qualifications

1. “Hell” is used throughout in the sense of the state of final and definitive loss, not in the older and wider sense of the abode of the dead in which the Creed says Christ descended. Where a source uses it in the wider sense the essay says so.

2. The claim that no magisterial act has ever declared a named human being to be damned is a bounded negative claim. What was searched is stated in the section that makes it: the eight acts of the Catechism’s dossier, the Catechism’s articles on the last things and on sin, the relevant canons of Trent, Lumen gentium 14 and 16, and the most recent papal statement. No exhaustive search of the magisterium was performed, and the claim is offered as correctable. It is independently asserted by John Paul II in the 1999 audience, which is the essay’s principal warrant for it.

3. The account of the canons of 543 reports Denzinger’s own arrangement and the text of canon 11 of Constantinople II. It does not adjudicate the long scholarly question whether the fifteen Origenist anathemas circulating under the council’s name belong to its sessions. Mansi and the critical editions of the conciliar acts were not reached, and that is the ceiling of the claim.

4. The observation that Benedictus Deus uses no word for eternity in its hell clause while using one of heaven four lines earlier is a textual observation about that act, made from the Latin at its locus. It is not an argument that the eternity of hell is doubtful; the essay states in the same paragraph that the doctrine does not depend on that act.

5. The observation that Firmiter uses poenam perpetuam and gloriam sempiternam is likewise a textual fact. The essay judges the variation stylistic and rests nothing on it.

6. Greek is transliterated throughout because the typeface used carries no Greek. Every transliteration is the author’s; the registered edition and the exact locus are given so a reader may check them. The Greek witness used is the Robinson–Pierpont Byzantine Textform in the byztxt Unicode CSV form, whose inline apparatus records divergences from Nestle–Aland and is a divergence note rather than a full critical apparatus. Where the essay reports that Nestle–Aland omits Mark 9:44 and 9:46, it reports that apparatus and does not adjudicate the text.

7. Scriptural chapter and verse are given in the Vulgate numbering the Douay–Rheims follows. Where that differs materially from the numbering of Greek editions and modern Bibles — notably at Mark 9 and Apocalypse 20:9–10 — the essay says so at the point of use.

8. Origen’s positions are reported from the Latin translation of Rufinus, who states in his own prologue that he omitted what he judged corrupt and supplied from elsewhere what he judged obscure. Nothing is attributed to Origen’s Greek. The essay reports what the received Latin text says and marks the transmission problem where it matters.

9. Gregory of Nyssa is quoted from a nineteenth-century English translation; the Greek was not consulted. The marginal note in a Krabinger manuscript is reported from that edition’s footnote at second hand and was not inspected.

10. Augustine’s City of God is quoted from the Dods 1871 translation, checked in this library’s tracked transcription of the 1871 printing at registered passages with exact line ranges. That is a nineteenth-century translation, not a critical edition, and the Latin was not collated.

11. The Supplement to the Summa is identified as a posthumous compilation from Aquinas’s Scriptum super Sententiis. The identification was checked by comparing the structure and content of Supplementum q. 99 a. 1 with Scriptum IV d. 46 q. 1 a. 3 at their own loci; the Latin of the Supplement was not consulted, so the comparison is of structure and content and not a collation.

12. The distinction of antecedent and consequent will is presented with the objection to it stated at strength. The essay does not adopt a position in the controversy de auxiliis, which Paul V forbade either school to censure in 1607 and which has never been determined.

13. “Definitive” as used here means an act the Church proposes as revealed and to be held by faith. The essay does not undertake to grade every act on the technical scale of theological notes, and a reader who needs that precision for a particular text should consult a competent manual.

14. The pastoral consequences drawn in the section on the census are drawn from the Catechism’s own directions and from Trent; they are not pastoral advice, and nothing here diagnoses any person’s state, guides any confession, or addresses scrupulosity, despair, or bereavement, all of which require a competent pastor and not an essay.

15. Where this essay says a popular presentation outruns its sources, the claim is always about a specific attribution and is documented at that point. It is never a claim that the doctrine is weaker than the Church says.

Notes, translations, rights, currentness, and review

The numbered Notes carry exact citations, edition and translation identifications, verification dates, transliterations, and the qualifications that would interrupt the argument; no indispensable premise lives only in a note. Scripture is quoted from the public-domain Douay–Rheims (Challoner revision) in the source library’s tracked verse-text derivative of the Project Gutenberg text, and in transliterated Greek from the tracked Robinson–Pierpont Byzantine Textform CSV files, three of which were registered for this essay. Denzinger’s Latin and its older English are read in a dated web transcription that does not identify its printed exemplar and is used as a finding aid and convenient witness for the underlying acts; printed Denzinger editions retain their own editorial rights and none was collated. Vatican web texts are quoted briefly with attribution and remain the property of the Libreria Editrice Vaticana. Patristic texts are quoted from identified public-domain nineteenth-century translations, delivered by copyrighted hosts whose markup and editorial revisions are not claimed here. The Aristotle passage is quoted in Ross’s Greek and Freese’s 1926 English as delivered by the Perseus Digital Library, whose markup is stated to be licensed CC BY-SA 3.0 United States. Official texts and third-party translations remain outside the project’s CC BY 4.0 grant; project-created prose, tables, images, and organization are project content.

All online witnesses were checked on 2026-07-25. No mutable civil or canonical law is relied on and no discipline is described as current beyond the acts cited at their own dates. This revision received internal argumentative, source-consistency, quotation, rights, and production review by the authoring agent. Independent scriptural, patristic, Thomistic, dogmatic, theological, and ecclesiastical review is outstanding; no imprimatur, nihil obstat, or ecclesiastical approval is claimed, and internal checking is not independent review.

References

Scripture

• English: Matthew 5:22, 29–30; 7:13–14; 10:28; 13:41–42, 49–50; 18:8–9; 25:31–46; 26:24. Mark 9:42–47 (Vulgate numbering). Luke 13:23–28; 16:19–31. John 17:12. Acts 1:25; 3:19–21. Romans 5:18–19; 11:32. 1 Corinthians 15:22–28. Colossians 1:19–20. 2 Thessalonians 1:6–10. 1 Timothy 2:1–6. Jude 6–7. Apocalypse 14:9–11; 20:9–15; 21:8. Isaias 66:24. Ezechiel 18:23, 32; 33:11. Wisdom 11:24–27. 2 Peter 3:9. 1 John 4:18. All quoted from the public-domain Douay–Rheims translation (Challoner revision) in Vulgate chapter and verse numbering, read at their own loci on 2026-07-25 in the source library’s tracked verse-text derivative of Project Gutenberg eBook 1581 (registered edition challoner-gutenberg-1581). That edition’s wording differs from the 1899/1914 American printing in a registered list of verses; none of the verses used here was checked against that list, and the translation is an identified working witness, not a critical edition.

• Greek: Matthew 7:13–14; 10:28; 25:41, 46. Mark 9:43–49 (Greek numbering). Luke 13:23–24; 16:23–26. 2 Thessalonians 1:8–9. Jude 6–7. Apocalypse 14:10–11; 20:10, 14–15. Read on 2026-07-25 in the source library’s tracked CSV form of the Robinson–Pierpont Byzantine Textform (edition rp2018-byztxt-unicode-csv); the artifacts for Mark, 2 Thessalonians, and the Apocalypse were registered for this essay from the same pinned repository state as the existing ones. Quoted only in transliteration. The inline {NA} and {Byz} braces record divergences and are a divergence note, not a critical apparatus.

Conciliar and papal acts

• Quicumque (Athanasian Creed), v. 41, DS 76; canons against Origen, Constantinople 543, DS 403–411, esp. 409 and 411; Second Council of Orange (529), conclusion, DS 397; Second Council of Constantinople (553), canon 11, DS 433; Fourth Lateran Council (1215), constitution 1 Firmiter, DS 801; Second Council of Lyons (1274), profession of Michael Palaeologus, DS 858; John XXII to the Armenians (1321), DS 926; Benedict XII, Benedictus Deus (1336), DS 1000–1002; Council of Florence, Laetentur caeli (1439), DS 1304–1306, and Cantate Domino (1442), DS 1351; Council of Trent, session 6 (1547), ch. 13 and canons 17 and 25, DS 1541, 1567, 1575, and session 14 (1551), canon 5, DS 1705; Paul V, formula ending the disputations de auxiliis (5 September 1607), DS 1997. Latin read 2026-07-25 in the patristica.net web transcription of the Latin Enchiridion symbolorum, which prints the revised and older numbering in parallel and does not identify its printed exemplar; the same site’s Deferrari English of the older numbering was used as a cross-check for DS 403–411 and DS 801. The compilation is a finding aid to its underlying documents and a convenient witness, not a substitute for them; no printed Denzinger and no conciliar critical edition was collated, and that is the stated ceiling of every claim resting on it.

• Second Vatican Council, dogmatic constitution Lumen gentium (21 November 1964), 14 and 16; official Vatican English web text, vatican.va, checked 2026-07-25.

• Paul VI, Solemni hac liturgia (Credo of the People of God), 30 June 1968, n. 12; official Vatican English web text, vatican.va, checked 2026-07-25. The Latin of Acta Apostolicae Sedis was not consulted.

• John Paul II, general audience of 28 July 1999, nn. 1–4; official Vatican English web text, vatican.va, checked 2026-07-25. Delivered in Italian; the Italian was not collated. A general audience is an act of the ordinary papal magisterium in its least solemn form.

• Benedict XVI, encyclical Spe salvi (30 November 2007), nn. 45–48; official Vatican English web text, vatican.va, checked 2026-07-25.

• Catechism of the Catholic Church 1030–1037 (purgatory and hell, with note 615 to 1035), 1051–1060 (In Brief), 1261 (children who die without baptism), 1854–1864 (mortal and venial sin); official Vatican English web text at vatican.va (hell) and the adjacent archive pages, checked 2026-07-25. Quoted briefly with attribution; the Catechism remains the property of the Libreria Editrice Vaticana.

Patristic witnesses

• Justin Martyr, Second Apology 7–9 (eternal fire; eternal punishment not a mere threat); Ante-Nicene Fathers, vol. 1 (Buffalo, 1885/1887); public domain; read 2026-07-25 at New Advent and cross-read in this library’s tracked OCR of the Buffalo printing. Chapter numbering varies between editions.

• Irenaeus of Lyons, Adversus haereses V.27.2 (separation from God is death; the loss of eternal goods is eternal; the blindness image); trans. Alexander Roberts and William H. Rambaut, Ante-Nicene Fathers, vol. 1 (Buffalo, 1887); public domain; read 2026-07-25 in this library’s tracked OCR of that printing at artifact lines 72024–72062, registered as a passage, and cross-checked at New Advent. The OCR is a locator aid, not textual control.

• Origen, De principiis I.6.1–2 (the end like the beginning; all recalled to one end; the author’s own disclaimer of genre) and II.10.4–6 (the sinner kindles his own fire; the medicinal reading); trans. Frederick Crombie, Ante-Nicene Fathers, vol. 4 (Buffalo, 1885); public domain; read 2026-07-25 at New Advent (book I) and New Advent (book II). Rufinus of Aquileia, Prologue to the same translation (his editorial rule), printed page 237, read the same day in the Christian Classics Ethereal Library delivery of the same volume, ccel.org, which reproduces front matter the New Advent delivery omits. The treatise survives complete only in Rufinus’s Latin.

• Gregory of Nyssa, Oratio catechetica magna (The Great Catechism) XXVI (the restoration of those now lying in sin; the healing of the introducer of evil); trans. William Moore and Henry Austin Wilson, Nicene and Post-Nicene Fathers, 2nd series, vol. 5 (New York, 1893); public domain; read 2026-07-25 at ccel.org. The Greek was not consulted; the manuscript marginal note is reported from that edition’s own footnote.

• John Chrysostom, Homilies on the Gospel of Saint Matthew 23, section 9 (two punishments in one judgment; ten thousand hells); trans. George Prevost, revised M. B. Riddle, Nicene and Post-Nicene Fathers, 1st series, vol. 10 (New York, 1888); public domain; read 2026-07-25 at New Advent. Section divisions are editorial; the Greek was not consulted.

• Augustine of Hippo, De civitate Dei XXI.9 (the worm and the fire; “let each man choose which he will”), XXI.11 (duration of penalty not proportioned to duration of offence), XXI.12 (the greater part left under punishment), XXI.17 (the tender-hearted Christians; Origen more indulgent), XXI.23 (the correlativity of Matthew 25:46); trans. Marcus Dods, The City of God, vol. II (Edinburgh: T. & T. Clark, 1871); public domain; each chapter read whole on 2026-07-25 in this library’s tracked transcription of the 1871 printing, and each registered with this revision as a passage with exact artifact line ranges.

Thomistic and philosophical sources

• Thomas Aquinas, Summa theologiae I, q. 19, a. 6 with ad 1 (the will of God always fulfilled; three readings of 1 Tim 2:4; antecedent and consequent will); q. 23, a. 3 with ad 2–3 (reprobation and its causality); q. 64, a. 2 (the obstinacy of the demons); I-II, q. 87, a. 3 with ad 1–2 (irreparability; the medicinal reply) and a. 4 (pain of loss infinite, pain of sense finite); Supplementum, q. 94, aa. 1–3 (the saints and the damned); q. 97, aa. 1, 5–7 (not fire alone; the fire’s nature and place); q. 98, aa. 1–5 (the will of the damned); q. 99, aa. 1–2 with replies (eternity of punishment; mercy in hell; Anselm’s principle). English from the public-domain translation of the Fathers of the English Dominican Province at New Advent; every question listed was read whole on 2026-07-25. The Supplement is a posthumous compilation and is labelled as such wherever used.

• Thomas Aquinas, Scriptum super libros Sententiarum IV, d. 46, q. 1, a. 3 (Utrum ex divina justitia inferatur peccatoribus poena aeterna) with q. 2 on divine mercy; Latin read 2026-07-25 at Corpus Thomisticum, paragraphs 21988–22024, and compared article by article with Supplementum q. 99.

• Aristotle, Rhetoric I.10.17 (Bekker 1369b12–14), the distinction of kolasis and timōria; Greek from the text ed. W. D. Ross (Oxford, 1959) and English from the translation of J. H. Freese (Loeb, 1926, public domain in the United States by date), both read 2026-07-25 in the Perseus Digital Library, whose markup is stated to be licensed CC BY-SA 3.0 United States.

Last revised (UTC): 2026-07-26T03:17:43Z

Reuse and rights. To the extent Triptych holds the rights, project-created content and design are licensed under CC BY 4.0. Scripture, liturgical or official texts, received prayers or hymns, quotations, fonts, and other third-party material retain their own status; public-domain material remains public domain. Identify changes. Attribution implies neither Triptych nor ecclesiastical approval. See LICENSE and THIRD_PARTY.md in the source.