Chapter 2

Method: How to Read Christians Across Two Thousand Years

On the discipline required to hear what the dead are actually saying.

The Bishop's Catalogue

In the year 421, in the African coastal city of Hippo Regius, an aging bishop sat down to write a short book for his friend Laurentius. The bishop was Augustine. He was almost seventy. He had been a Christian for thirty-five years and a bishop for twenty-five. The book was the Enchiridion ad Laurentium — “the handbook for Laurentius” — a short summary of Christian faith, hope, and love.¹ Augustine intended it to be brief. He did not quite manage. The handbook ran to thirty-some chapters before he reached the doctrine of the last things. And when he reached the last things, something curious happened.

He stopped to argue.

In chapters 110 through 113, Augustine paused his summary and turned to refute, at unusual length, a series of softer eschatological positions held by his fellow Christians. He called these fellow Christians the misericordes — pronounced “mih-zair-ee-CORE-dayz,” a Latin word meaning “the merciful ones.” The word is gentle. Augustine’s tone is firm but not unkind. He treats the misericordes as sincere Christians whose error is rooted in genuine human compassion. He never calls them heretics. He never calls them novel. He never says their views are unknown in the Church. He simply argues with them.

He argues, in fact, with at least five distinct softer positions held by various groups of his Christian contemporaries. Some say all baptized Catholics will eventually be saved, no matter how they have lived. Others say all Christians who have received the Eucharist will be saved. Others say all who have ever confessed Christ — including heretics — will eventually be saved. Others say all human beings of every religion will eventually be saved through divine mercy. The catalogue is long. The argument runs for pages. The very length is the witness: these positions were widespread enough, in the Christianity of Augustine’s own day, to require this much careful refutation.²

Now here is the strange part. For most of the next sixteen hundred years, Augustine’s readers in the Latin West did not read this catalogue carefully. They read Augustine’s conclusion — eternal conscious torment for the unredeemed, no postmortem opportunity, no exceptions — and quietly forgot that he was arguing against fellow Christians. The catalogue itself dropped out of view. Within a generation or two of Augustine’s death, the Western theological imagination had settled into a strange double conviction: that Augustine’s position was the position of the Church, and that no other position had ever been seriously held by serious Christians.

Both halves of that conviction are false. Augustine’s position was a position; it became dominant for understandable reasons; but it was not the universal Christian position of the late fourth century, and Augustine himself is the witness against the claim that it was. We have, in the Enchiridion, a roll-call of dissent — written by the most committed defender of the strict position — that names the very diversity later writers would deny existed.

This is what I mean when I say that doing historical theology well is harder than it looks.

Most of the time, our memory of the Christian tradition is shorter than the tradition itself. We inherit not the tradition but a selection from the tradition — usually the selection that won the argument in our part of the Church. We then call that selection “what Christians have always believed.” The trouble is that the selection is not the tradition. The losers in old arguments often left more behind than we have noticed. Sometimes the winners themselves left careful records of who they were arguing against — records we have not bothered to read.

This chapter is about how to read the tradition without that selective forgetting. It is about the discipline required to hear what dead Christians are actually saying, rather than the echo of what someone in my own century said they said. It is, in a sense, the foundation on which the rest of this book stands. The chapters that follow will trace what Christians have actually believed about the fate of the dead, century by century. But none of that tracing matters if we do not first agree on how to read.

Let me put my methodological cards on the table. I am writing as a conservative evangelical scholar with a theological position of my own. I hold conditional immortality with postmortem opportunity. I am genuinely open to a conservative biblical universalism. I reject eternal conscious torment as incompatible with the character of God revealed in Christ. Those are my convictions. They will inevitably shape what I notice and what I emphasize. I am not pretending otherwise.

But in this book I am bound by an iron discipline: the figures speak for themselves. When Origen taught the eventual restoration of the devil, that is what Origen taught — not a softer position I might prefer to find. When Augustine taught eternal conscious torment with no postmortem opportunity, that is what Augustine taught — not a more hopeful position later readers have wished for. My job is to describe the historical record, not to fix it. The reader is welcome to disagree with my theology. The historical map, I hope, stands on its own.

What follows are seven commitments I have made for reading carefully. Each is simple enough to state. Each is harder to keep than it looks.

Seven Commitments for Reading the Tradition

1. Read Primary Sources Before Secondary Sources

This first commitment seems obvious. It is not.

A great deal of contemporary writing about historical figures depends on a chain of secondary references that has lost touch with what the figure actually wrote. Someone reads a textbook that quotes another textbook that quotes a third textbook that quoted, several decades ago, a translation of the original. By the time the claim reaches us, it has been filtered through five sets of interpretive hands. Each set of hands had its own concerns. Each compression dropped something. What we receive may bear only a passing resemblance to what the original figure said.

I have seen this happen with Origen. I have seen it happen with Augustine. I have seen it happen with Calvin, Wesley, Barth. The reader who wants to know what these figures actually taught has only one reliable option: read them. Not the secondary literature first, then the figure. The figure first, then the secondary literature.

The trouble is that this is hard work. Origen wrote in Greek (and some of his Greek survives only in a Latin translation by a man, Rufinus, who admits he edited the text he was translating).³ Augustine wrote in Latin. The Cappadocians wrote in Greek. Isaac of Nineveh wrote in Syriac. The medieval mystics wrote in Latin, Middle High German, and Middle English. Calvin wrote in Latin and French. To go to the primary sources is to commit to translation work, to working with the best modern English versions where the original is beyond us, to checking the translations against each other and against the underlying text whenever possible.

I do not pretend to read every primary source in its original language. Few scholars can. But where the original-language scholarship matters — where a Greek or Latin or Syriac word carries weight that an English translation might miss — I will say so. And throughout, I will quote primary sources at length, in reliable translation, so that the reader can hear the figures’ own voices and not just my paraphrase of them.

This commitment cuts both ways. It cuts against the apologists for my own position. The contemporary universalist or PO advocate is often tempted to read selected sentences from Origen or Gregory of Nyssa as if those sentences settled the modern question. They do not. The careful reader must read the whole text, in context, with the figure’s own argument in view. The same goes for the traditionalist. The Augustinian who quotes City of God 21 without reading Enchiridion 110–113 is missing something that Augustine himself documented.

Let me give one concrete illustration of how the chain of secondary references can go wrong. A common popular claim about Origen is that he taught the eventual salvation of the devil and that this was the specific doctrine condemned by the Council of 553. Each link of that claim deserves examination. Did Origen teach the salvation of the devil? The honest answer, as we have already seen, is that the textual evidence is mixed and the scholarship is divided. Was the alleged doctrine the specific doctrine condemned at 553? The anathemas of 553 actually targeted a particular Evagrian construction tied to the pre-existence of souls and the eventual identity of all rational creatures with Christ — a construction Origen himself did not hold in that specific form. Was the council itself an ecumenical council that bindingly condemned the doctrine? The conciliar status of the anti-Origenist anathemas is itself contested, as we will see. By the time the popular claim reaches us, each link of the chain has lost detail that the original sources still contain. The reader who returns to the primary sources finds something messier and more interesting than the popular shorthand. That is what reading primary sources first is for.

Primary sources first. Then secondary sources. Then judgment.

2. Resist Anachronism

The modern reader brings modern categories to the ancient text. The categories I bring to Origen — “universalism,” “conditional immortality,” “postmortem opportunity,” “inclusivism,” “exclusivism,” “annihilationism” — were not the categories Origen used. He had his own categories: apokatastasis pantōn (the restoration of all things), the aiōnes (ages) of God’s economy, the medicinal nature of postmortem fire, the freedom of rational creatures. To force Origen into the boxes of a twenty-first-century evangelical debate is to make him answer questions he was not asking.

The temptation to anachronism is most subtle when the modern category is partly right. Origen was a universalist in a meaningful sense — that is, he taught the eventual restoration of all rational creatures. So “Origen was a universalist” is not exactly false. But it is misleading. Modern universalism (Hart, Talbott, Parry, Hronich) is not Origen’s universalism. The modern position usually rests on a few specific exegetical and philosophical moves — the lexical case on aiōnios, the universal-scope readings of the Pauline texts, the argument from God’s nature as love — that are not exactly Origen’s moves.⁴ Origen’s case is woven into a larger speculative system about the pre-existence of souls, the freedom of all rational creatures, the medicinal economy of the aiōnes, and the eventual subjection of all things to Christ. To call Origen a universalist and stop is to flatten a rich, strange, three-dimensional theology into a two-dimensional modern label.

I will use the modern categories. I have to. The reader needs a vocabulary to handle the historical material, and the modern terms (universal restoration, postmortem opportunity, conditional immortality) are clearer than any historical alternative. But I will flag when the modern category is being applied to a premodern figure, and I will name the slippage when slippage occurs. “Origen was a universalist in the patristic sense, which is not identical to what modern universalists mean by the term,” is a sentence I will write more than once.

The same caution applies to the term “conditional immortality.” The phrase as a fixed theological label is no older than the nineteenth century — emerging in the 1840s and 1850s in the work of Edward White, Henry Constable, and others.⁵ But the substantive position — that immortality is a gift of God rather than an inherent property of the soul, and that the unsaved finally cease to exist — has antecedents in Arnobius of Sicca in the late third century, and arguably in some Apologists and in the Socinians of the sixteenth and seventeenth centuries. To call Arnobius a “conditionalist” is descriptively useful. To imply that Arnobius held the position in the same form Edward White did is anachronistic. Both can be true at once if we are careful.

Quentin Skinner, in a classic essay on historical interpretation, called this the danger of “the mythology of doctrines” — the assumption that every historical figure must have held a position on questions our era happens to find pressing.⁶ Skinner’s warning applies with full force here. Some premodern figures gave no thought to the questions a modern reader most wants to ask them. We must respect that silence. Not every silence is a vote.

A specific example will help. One of the most contested words in the whole tradition is the Greek aiōnios — pronounced “eye-OH-nee-oss,” an adjective formed from aiōn (“age” or “eon”). The word appears in Matthew 25:46, where Jesus speaks of those going “to aiōnios punishment” and the righteous “to aiōnios life.” The traditional English translation — “eternal,” “everlasting” — assumes that aiōnios denotes endless duration. The lexical case against that assumption has been made vigorously by John Wesley Hanson in the nineteenth century, by Ilaria Ramelli and David Konstan in their 2007 monograph Terms for Eternity, and by a number of contemporary evangelical writers, who argue that aiōnios describes a thing as pertaining to an age, or as having the character of the age, rather than as enduring endlessly through time.²⁸

The lexical question itself is patristic Greek territory and gets full treatment in Chapter 8, where Origen’s reading of aiōnios is the locus classicus. Here I want only to note the methodological point. Augustine read Matthew 25:46 in Latin, where aeternus carries a stronger connotation of endless duration than the underlying Greek aiōnios necessarily does. Augustine acknowledged that he did not read Greek well. His Latin reading became the dominant Western reading for over a thousand years. But the question of what aiōnios meant to Origen, or to Gregory of Nyssa, or to the Greek translators of the Septuagint, is a question that has to be settled in Greek, not retrojected from Augustine’s Latin.²⁹

The danger of anachronism here is not just lexical. It is theological. To assume that aiōnios in third-century Alexandrian Greek must mean what aeternus in fifth-century African Latin meant — or what “eternal” means in twenty-first-century English — is to project later semantic categories back onto earlier vocabulary. The honest historian listens for the meanings the figures themselves were using. Sometimes those meanings cluster around what modern English calls “eternal.” Sometimes they do not. The careful reader holds the question open.

3. Distinguish What a Figure Held from What Later Interpreters Claim They Held

This third commitment is closely related to the second, but it goes further.

The reception history of any major figure is itself a subject of historical inquiry. There is what Augustine taught. There is what the Augustinian tradition taught after him. The two are related, but they are not identical, and a great deal can be lost in the gap between them.

Take the case of Calvin. There is a substantial body of scholarship — beginning with the so-called “Calvin against the Calvinists” thesis associated with R. T. Kendall, Brian Armstrong, and others — that argues that the Reformed scholasticism of the seventeenth century departed in significant ways from Calvin’s own positions, especially on the extent of the atonement.⁷ One does not have to accept every claim of that scholarship to see the basic point: Calvin and Calvinism are not the same thing. The careful reader who wants to know what Calvin taught has to read Calvin, not Beza or Turretin or Hodge.

The same goes for Augustine. The Augustinian tradition that grew up around his work — through Caesarius of Arles, Gregory the Great, the Carolingian theologians, the medieval doctors, and on into the Reformation — was not always faithful to Augustine’s own nuances. Augustine himself wavered, developed, and qualified. The mature Augustine is not identical to the young Augustine. The polemical Augustine is not identical to the pastoral Augustine. To read “Augustinianism” back into Augustine’s mind is a basic interpretive mistake.

The same goes for Barth. The phrase “Barth taught universalism” is a common one in popular discussions of twentieth-century theology. It is also wrong, or at least misleading. Barth refused to teach universalism. He also refused to teach its denial. What he taught was something more complicated, rooted in his doctrine of election and his refusal to limit grace in advance. The Barthian tradition — Jüngel, Torrance, Bromiley — has read Barth in various ways. But the careful reader has to read Barth, and the careful reader will find that “Barth taught universalism” flattens a position more subtle than the slogan suggests.⁸

The Barth case is worth a sentence more, because it illustrates how easily even sophisticated readers can let the slogan stand in for the position. In Church Dogmatics IV/3, §70, Barth explicitly takes up the question of universal salvation and explicitly refuses it — not because he is committed to the doctrine of eternal conscious torment, but because he refuses to make any prediction at all about the final scope of grace. He insists that the Church has no warrant either to teach that all will be saved or to teach that any particular person will be lost. The position is genuinely irreducible to either side of the modern debate. To make Barth a universalist is to flatten the strange shape of what he actually said. To make him a strict traditionalist is to miss the radical universal scope of his doctrine of election in Church Dogmatics II/2. Either move flattens. The careful reader holds the position open in the shape Barth himself gave it.

4. Distinguish the Figure's Primary Commitments from Their Speculative Reflections

Origen knew the difference. In the preface to De Principiis (On First Principles), Origen sets out a careful distinction between what the apostles handed down as the “rule of faith” — the non-negotiable doctrinal core — and what they left open for the inquiry and reflection of later thinkers.⁹ The first category is fixed: the existence of one God, the divinity of Christ, the inspiration of Scripture, the resurrection of the body, the freedom of the will, the existence of angels and demons. The second category is open: the origin of the soul, the nature of the world’s eternity, the precise mechanics of the eschaton.

Origen is explicit that his speculative reflections — about pre-existence, about the eventual restoration, about the angelic hierarchy — fall in the second category. He is offering possibilities, not pronouncing dogma. He says so in his own words. The reader who treats every Origenist speculation as Origen’s firm doctrine has missed the framing Origen himself supplied.

This distinction matters across the tradition. Augustine’s pastoral letters, written to grieving relatives or anxious correspondents, are not at the same level of formal commitment as his polemical works against the Pelagians. His De Cura Pro Mortuis Gerenda (“On the Care to Be Taken for the Dead”), addressing whether prayers for the deceased could help them, is in a softer register than City of God 21. Calvin’s sermons are not the Institutes. Aquinas’s commentaries are not the Summa. Edwards’s late notebooks, where he wrestled privately with hard questions, are not his polished sermons.

The careful reader makes these distinctions. When I quote Origen’s speculative reflections on apokatastasis, I will note that Origen himself flagged them as speculative. When I quote Augustine’s polemical works, I will distinguish polemical context from pastoral context. The figure had a register. We owe them attention to which register they were speaking in.

Augustine in particular rewards this kind of careful register-tracking. The Augustine of the late polemical works against Pelagius and Julian of Eclanum is a different rhetorical voice from the Augustine who wrote letters of comfort to recently bereaved widows. The Augustine of the 411 anti-Donatist sermons is not exactly the Augustine of the early De Vera Religione. And the Augustine of the recently rediscovered Divjak letters (1981) and Dolbeau sermons (1990) has shown himself, on still further reading, to be a more nuanced and more pastoral figure than the polemical-Augustine caricature once allowed.²⁶ All of this matters for eschatology specifically. When Augustine writes City of God 21 against the misericordes, he is in his polemical register, marshalling his argument against fellow Christians whose softer positions he regarded as dangerous. When he writes De Cura Pro Mortuis Gerenda to the bishop Paulinus of Nola, asking whether prayers and burial near the martyrs can help the dead, he is in his pastoral register, more cautious, more conscious of the limits of what we can know. To collapse these voices into a single “Augustinian doctrine” is to lose what Augustine himself was doing.

5. Acknowledge Contested Scholarship

Some historical questions have settled answers. Gregory of Nyssa taught universal restoration. This is not in dispute among serious scholars. Brian Daley, no friend of contemporary universalism, treats it as established.¹⁰ Ilaria Ramelli, an outspoken contemporary universalist, treats it as established. Michael McClymond, the major recent critic of the universalist tradition, treats it as established. They argue about how to evaluate Gregory’s teaching, but not about what Gregory taught.

Other questions are genuinely contested. Did Origen teach the eventual salvation of the devil? Origen says contradictory things, and the surviving textual record is complicated by Rufinus’s editing of De Principiis. Some scholars (Ramelli) read Origen as affirming the devil’s eventual restoration. Others (Mark Edwards, Henri Crouzel in his more cautious moments) hold that Origen denied this explicitly. The textual evidence cuts both ways.¹¹ The honest historian does not pretend the question is settled.

Did Augustine, in his earliest writings, hold a softer position on the fate of the lost? Robert O’Connell argued in a series of books that the early Augustine showed traces of Neoplatonic and Origenist influence that the mature Augustine rejected.¹² Others have resisted O’Connell’s readings. The young Augustine of De Libero Arbitrio (“On Free Will”) is not exactly the old Augustine of the Enchiridion, but how to characterize the difference is itself a scholarly puzzle.

What does Maximus the Confessor’s relationship to apokatastasis amount to? He uses the term. He uses it differently from Origen. Polycarp Sherwood, in a 1955 study, argued that Maximus preserved a “mystical” universalism behind his anti-Origenist rhetoric. Hans Urs von Balthasar developed this reading. Andrew Louth is more cautious. Brian Daley has written specifically on what he calls Maximus’s “honorable silence” on the question.¹³ The honest answer is: scholars are still working it out.

Where the historical record is contested, I will say so. I will name the scholars on each side. I will explain why the disagreement exists. And I will let the reader follow the argument rather than pretending the case is closed when it is not.

This is the most important commitment to keep when one is writing from a particular theological position. The temptation, when an unsettled scholarly question favors my position, is to treat the favorable side as settled. The temptation, when an unsettled question goes against my position, is to insist on its contested status. Both temptations distort. The discipline is to describe the actual state of scholarly disagreement, regardless of whose side it favors.

6. Distinguish Formal Teaching from Devotional Speech

Christian communities have done many kinds of speaking and writing. The Summa Theologiae is not the same kind of text as a Eucharistic prayer. The Council of Trent is not the same kind of speech as the Dies Irae. A papal encyclical is not the same kind of utterance as a saint’s spontaneous prayer in extremis. To read all of these as if they were equally formal pronouncements of doctrine is a category error.

This matters for the long hope tradition. Patristic prayers for the dead, medieval liturgies for the departed, the Eastern Orthodox memorial services, the prayers of countless ordinary Christians at the gravesides of unconverted loved ones — these are devotional speech. They reach for hope. They sometimes name a hope the formal dogmatic statements of the same community did not authorize. The Anglican Book of Common Prayer’s burial office, the various Roman Catholic prayers for those who have died “in our day,” the Orthodox prayers for “those whose names the Lord knows” — these are not dogmatic statements that the Church teaches universalism or postmortem opportunity. But they are also not nothing. They are a witness, in the register of devotional speech, that the Church’s hope has reached further than its formal theology has often dared to articulate.

The point cuts both ways. Reading devotional speech as if it were dogma overstates the long hope tradition. Reading dogma as if it were devotional speech understates the strictness of figures like Augustine and Aquinas. Both are mistakes. The careful reader keeps the genres separate.

This is especially important when we turn to the Eastern Orthodox tradition. The liturgical life of the Orthodox East has prayed for the dead, including for the dead of other faiths and for those whose state at death is uncertain, in ways that the formal dogmatic statements of the Eastern Church have not exactly authorized. Andrew Louth and Kallistos Ware have both written carefully about how the Orthodox liturgical imagination has held open a hope that the dogmatic texts have not always articulated.²⁷ The same pattern shows up in the Latin West — the medieval and post-medieval Catholic prayers for the dead, the Anglican burial office, the Lutheran funeral hymns, the Methodist prayers for those who have not heard the gospel. None of this is dogma in the strict sense. None of it amounts to a formal Church teaching of postmortem opportunity or of universalism. But the cumulative liturgical witness is striking. Christians at prayer have, century after century, asked God to do more than the dogmatic theologians have allowed themselves to teach. That is a fact worth registering carefully — not as a knock-down argument for one position, but as evidence of where the living devotional imagination of Christian communities has reached when the formal restraints of polemical theology were not in view.

A related caution: poetic and homiletic speech is not systematic theology. Dante’s Inferno is a great work of theological imagination, but it is not the doctrine of the Roman Catholic Church. Jonathan Edwards’s “Sinners in the Hands of an Angry God” is a great revival sermon, but it is not Edwards’s systematic theology of hell (and Edwards’s actual systematic theology of hell is more interesting and more nuanced than the sermon suggests). The careful reader takes account of the rhetorical situation. Sermons are not summae. Poems are not catechisms.

7. Avoid Both Apologetic and Polemical Distortion

The seventh and last commitment is the hardest. When we read premodern figures, the temptation is always to read them in service of whatever modern argument we are making. If I am a contemporary universalist, the temptation is to make Origen and Gregory of Nyssa say more than they said in favor of my position. If I am a contemporary traditionalist, the temptation is to dismiss them when they do not say what I want — or to read them through the lens of their later opponents, accepting Augustine’s polemical caricatures rather than reading the figures themselves.

Both temptations distort. Both are common. Both reduce the figures to ammunition in a modern debate they did not enter.

I have committed myself to resisting both. When Origen wrote things that do not fit easily with modern universalism — and there are several — I will quote them. When Augustine wrote things that do not fit easily with the strict modern Augustinianism that claims him — and there are several — I will quote them. The figures are not my allies. They are not my opponents. They are voices in a long Christian conversation about what becomes of the dead, and my job is to hear them.

This commitment means something else, too. It means generosity toward figures whose positions I do not share. Augustine, Aquinas, Calvin, Edwards — these are great thinkers who deserve serious engagement. I disagree with central aspects of their eschatology. I also believe they have things to teach me. Augustine’s worry about cheap grace is a real worry. Aquinas’s metaphysical seriousness is a real virtue. Calvin’s sense of God’s sovereignty is something the contemporary universalist tradition has sometimes attenuated and ought to recover. The strict tradition has its strengths. I will name them.

And generosity in the other direction, too. Origen, Gregory of Nyssa, Isaac of Nineveh, Julian of Norwich, the German mystics, Wesley, Barth, Balthasar, Moltmann — these are also great thinkers who deserve serious engagement, and they have been treated cheaply by much modern evangelical writing.³⁰ I will take them seriously as Christian theologians, name their genuine contributions, and resist the temptation to read them merely as champions of my own position. They were doing their own work, on their own terms. The least I can do is meet them where they were.

An ancient example illuminates how this kind of irenic balance can look in practice. In the early fifth century, John Cassian — the Latin-speaking monk who had spent years in Egypt absorbing the desert tradition — found himself trying to do justice to two extremes at once. The Egyptian Origenist monks were on one side, with their strongly spiritualizing exegesis. The Augustinian Latins were on the other, increasingly committed to a strict doctrine of grace that left little room for human cooperation. Cassian’s Conferences — written in the 420s, addressed to Latin-speaking monks — attempted to mediate the question of grace and human will in ways that satisfied neither party fully. His Conference 13 became controversial almost immediately. Augustine’s heirs would later label Cassian a “semi-Pelagian,” a label he would have rejected. Cassian did not see himself as compromising between extremes. He saw himself as articulating a position deeper than either extreme — a position rooted in the Eastern monastic tradition’s long experience of grace at work in real human lives.²⁴ Whatever we make of Cassian’s specific theological positions, his methodological instinct is the right one for the historical theologian today. The two competing camps in a given controversy may both have something to say. The truth is sometimes wider than either side allows. Hearing them both — refusing the convenience of an easy alliance — is part of the discipline.

Test Cases: Where the Method Matters

Methodological commitments stated in the abstract sound fine. The question is whether they cash out in actual interpretive practice. Let me sketch four test cases — places where the method I have just described changes what we conclude about the historical record. Each will be developed at full length in later chapters; here I want to show, briefly, how the method works.

Test Case 1: 1 Peter 3:18–4:6 and the Descent of Christ

One of the most contested passages in the New Testament is 1 Peter 3:18–4:6, where Peter speaks of Christ “preaching to the spirits in prison” (3:19) and of the gospel being “preached even to those who are dead” (4:6). The patristic Church read this passage extensively. The doctrine of Christ’s descent to the dead (the “harrowing of hell”) became one of the central loci of patristic eschatological reflection, and is enshrined in the Apostles’ Creed. Most Greek Fathers read the descent as a real ministry of Christ to the souls of the departed, including in some form an offer of salvation to those who had not heard the gospel in life.¹⁴

The modern interpretive landscape is contested. Some contemporary evangelical readers (following Augustine’s influential reading in Letter 164 to Evodius) hold that Christ in His pre-incarnate state preached through Noah to Noah’s contemporaries, who were “in prison” of disobedience during their earthly lives — no postmortem opportunity in view. Other readers (following the broader patristic consensus) hold that 1 Peter does in fact describe a postmortem ministry of Christ to the departed.¹⁵

The temptation, on either side, is to retroject the modern debate into the patristic text. The modern PO advocate is tempted to find more explicit endorsement of postmortem opportunity in the Fathers than the Fathers actually offer. The modern denial of PO is tempted to read Augustine’s minority Latin reading as the patristic consensus, when in fact it was not. The careful reader, applying the method, reads the actual patristic texts — Clement of Alexandria, Origen, Cyril of Alexandria, Gregory the Great, John of Damascus — and lets them say what they say. The patristic descent tradition is rich, varied, and on the whole more friendly to some form of postmortem ministry than Augustine’s Latin reading would suggest. That is a fact worth noting without overclaiming. I develop the patristic descent tradition at length in Chapter 11, and the Augustinian Latin reading in Chapter 12.

What does the method actually surface here? Several things the popular reading tends to miss. Clement of Alexandria, writing around the turn of the third century, treats the descent as a genuine postmortem proclamation of the gospel to the dead — and reads it as having a saving effect, not merely transferring the souls of the patriarchs from one waiting-place to another. Origen takes up the same reading and develops it further. Cyril of Alexandria, in the fifth century, ties the descent explicitly to 1 Peter 3:19. The pattern across the Greek Fathers is that the descent is preaching, the preaching has the character of gospel, and the response of the souls who hear it bears on their salvation in some way the Fathers do not fully specify but clearly intend. Whether that adds up to a doctrine of “postmortem opportunity” in the modern sense is a separate question, and the answer depends on what one is willing to read in. But it certainly does not add up to Augustine’s Latin reading, on which the descent is essentially a metaphor for Christ’s pre-incarnate preaching through Noah, and the souls in 1 Peter 3:19 are the souls of Noah’s contemporaries during their earthly lives.

The method does not require us to conclude that the Greek Fathers taught modern PO. It does require us to notice that the Augustinian Latin reading is not the patristic reading, that it is a minority Western position whose later dominance was a Western achievement, and that anyone who claims “the Fathers” for the strict reading is overstating what the historical record will bear. This is the method working — surfacing what the popular reading suppresses, without overclaiming what the historical record actually says.

Test Case 2: The Augustinian Misericordes Catalogue

I have already used this case as the opening vignette of this chapter. Let me say a bit more about it as a methodological test.

The misericordes catalogue is in the public domain. The text is available in any standard edition of the Enchiridion and in the NPNF first-series English translation. The basic facts of the case — that Augustine catalogues five distinct softer positions, that he treats them as held by sincere fellow Christians, that he never calls them heretical or unprecedented — are not in scholarly dispute.

And yet, in popular evangelical writing on the doctrine of hell, the catalogue is almost never mentioned. The standard popular treatments cite Augustine on the eternity of punishment without noting that Augustine is arguing against fellow Christians who held softer views. The result is a reader who has been told, by people citing Augustine, that “the Church has always taught” eternal conscious torment — when the very Augustinian text being cited demonstrates the opposite, that the diversity of Christian eschatology was alive and well in the late fourth century.

Notice what Augustine concedes. The opponents of eternal punishment are “some — indeed very many.” They are not heretics. They do not reject Scripture; they are reading the same Scripture Augustine is reading and reaching different conclusions. They are characterized as yielding to “human feelings” — that is, to compassion, to moral seriousness about the character of God, to a refusal to soften their hearts against the suffering of the lost. Augustine is firm in opposition. He is not dismissive of who they are or how many they are. The text is a witness against the modern claim that this kind of position only emerged in modern times.

Reading the catalogue, applying the method, what do we conclude? Several things at once. First, Augustine taught eternal conscious torment — that is not in doubt. Second, Augustine was opposing fellow Christians, in substantial numbers, who taught softer positions — that is not in doubt either. Third, the popular evangelical claim “the Church has always taught eternal conscious torment with no exceptions” is not what the historical record actually says. Fourth, the dominant Western position became Augustine’s for understandable historical reasons, but the dominance was an outcome, not a recovery of original Christian teaching. Fifth, the modern recovery of softer eschatological positions is not, properly speaking, an innovation. It is the return of a tradition that was alive, in substantial numbers, in the late fourth century — and that Augustine himself documents.

The full development of the catalogue belongs to Chapter 12, where I treat Augustine’s eschatology at length. The point here is methodological: a reader who has been told one thing by popular writers, and who actually opens the primary source, will find a different picture. The method works.

Test Case 3: The Reception of Origen and the Council of 553

The standard popular objection to patristic universalism is “an ecumenical council condemned it.” The council in question is the Second Council of Constantinople in 553. The condemnations are the fifteen anti-Origenist anathemas. The objection sounds decisive. It is much less decisive than it sounds.

Applying the method, several things come into view at once. First, the historical record of how the fifteen anathemas were produced is genuinely murky. They may have been issued by an extra-conciliar gathering of bishops in Constantinople before the Council formally convened, rather than enacted by the Council itself. Norman Tanner, in the standard scholarly edition Decrees of the Ecumenical Councils, has noted the textual and procedural difficulties.¹⁷ Modern Catholic theologians (Balthasar, Rahner, John Sachs) have distinguished between “Origenism” — the specifically Evagrian construction tied to the pre-existence of souls and the eventual identity of all rational creatures with Christ — and apokatastasis as a broader doctrine. The condemnation, on this reading, targets the Evagrian construction, not apokatastasis simpliciter.

Second, the specific anti-apokatastasis anathema reads, roughly: “If anyone shall say that the punishment of the demons and of impious people is temporary and shall have an end, and that apokatastasis will take place of the demons and the impious, let him be anathema.” The phrase is specific. It condemns a particular construction tying the restoration of the demons to the temporary nature of punishment. It does not condemn every form of universalism — for instance, it does not condemn Gregory of Nyssa, who taught a universal restoration of human beings without committing himself to the same Evagrian framework. Gregory was never declared a heretic. He remained a doctor of the Church.

Third, several major figures we will encounter — Maximus the Confessor in the seventh century, John Scotus Eriugena in the ninth — continued to engage apokatastasis in nuanced ways after 553. The notion that the council closed the question is not borne out by the post-553 history of Greek and Carolingian theological reflection.

The careful reader applying the method does not conclude that 553 was a non-event. It was a real event. But the council’s condemnation is more specific, more procedurally complicated, and more limited in its target than the popular evangelical citation suggests. I develop the controversy in detail in Chapter 11. The point here, again, is methodological: the popular shorthand will not survive a careful reading of the actual record.

Test Case 4: Calvin on the Descent of Christ

The fourth test case is more compressed but no less instructive. Calvin, in the Institutes 2.16.8–11, treats the descent of Christ to the dead.¹⁸ Calvin’s reading is famously non-literal. He treats the descent not as Christ’s spatial journey to a place called Hades but as the depth of His suffering on the cross — the soul-anguish of bearing the wrath of God. This is a striking move, and Reformed commentators have read it in two very different ways ever since.

One reading — call it the strict reading — treats Calvin as effectively dissolving the doctrine of the descent into the cross. There is no real descent. The clause in the Apostles’ Creed becomes a metaphor for the Crucifixion. On this reading, Calvin has departed substantially from the patristic and medieval understanding of the descent.

The other reading — call it the moderate reading — treats Calvin as preserving the descent but reframing what it means. The descent is real; it is part of the work of Christ; but its significance is the substitutionary depth of the suffering, not a literal harrowing-of-hell narrative.

Which reading is correct? Both have substantial textual support. The Reformed tradition itself has divided. The Heidelberg Catechism (Question 44) reads Calvin moderately. Some later Reformed theologians have read Calvin more strictly. The question, applying the method, is not which modern theological position we want Calvin to have held; it is what Calvin actually argued in the Institutes, in his sermons, and in his commentary on 1 Peter. The honest answer is that Calvin’s position is genuinely complicated and that the disagreement among his readers reflects a real tension in his own work.¹⁹ I treat Calvin at length in Chapter 15.

These four test cases — the patristic descent tradition, the Augustinian misericordes, the Council of 553, and Calvin on the descent — could be multiplied. The 1845 dispute over John Henry Newman’s essay on doctrinal development, the modern recovery of the Anabaptist soul-sleep tradition, Wesley’s sermons on universal salvific will, Barth’s refusal of universalism in Church Dogmatics IV/3 — each is a place where the casual reading of a popular textbook will diverge sharply from what the figures themselves wrote. The same pattern shows up over and over. The popular consensus is thinner than the historical record. The historical record is messier than the popular consensus. Reading the primary sources, attending to the registers, acknowledging contested scholarship — these are not optional refinements for an obsessive historian. They are the only way to read the tradition at all without distorting it.

What the Method Does Not Do

Before I close, I want to be clear about what the method I have just described is not designed to do.

It is not designed to settle the contemporary debate over the fate of the dead. The historical record is not the same thing as Scripture, and even the historical record cannot decide the modern theological question by itself. Christians have held diverse positions on these matters across two thousand years. The diversity is itself a fact. The fact that a position has historical pedigree does not make it correct. The fact that a position is held by a powerful minority — or even a powerful majority — does not make it correct. The question of what is true belongs to a different kind of argument than the question of what Christians have believed.

This is a documentary book. Its argument is about the shape of the tradition. It is not, in this volume, an exegetical defense of any particular eschatological position. I have written elsewhere about the exegetical case for postmortem opportunity, and others have written far more extensively about the exegetical case for universalism and for conditional immortality.²⁰ Those arguments stand on their own. Whether postmortem opportunity, conditional immortality, or universal restoration is theologically true is a question Scripture must answer, not a question history can settle.

What the method can do is dismantle a specific rhetorical move: the claim that the modern recovery of these positions is a modern innovation, foreign to the historic Christian tradition. That claim is false. It is the claim most often used to dismiss postmortem opportunity, conditional immortality, and hopeful universalism in contemporary evangelical settings. Demolishing that claim is the modest but important work of this book.

The methodological tradition I am drawing on, in attempting this work, is not new. Jaroslav Pelikan, in the opening volume of his five-volume The Christian Tradition, articulated the distinction between “the history of dogma” — the official pronouncements of councils and creeds — and “the history of Christian doctrine,” which includes the broader development of Christian thinking. The history of doctrine, on Pelikan’s account, includes positions that were never officially endorsed but remained part of the Christian conversation. The history of dogma is narrower. Confusing the two — treating “what the councils said” as “what Christians have believed” — is one of the standard pitfalls of popular church history.²² Pelikan has shaped this book’s approach more than any single methodological source.

So too has the work of Carlos Eire, the Yale historian whose A Very Brief History of Eternity and Reformations have demonstrated, again and again, how easily contemporary readers project our own categories onto premodern texts. Eire is especially good at showing how the meaning of basic Christian vocabulary — “eternity,” “heaven,” “hell,” “soul” — has shifted across centuries in ways that surprise the modern reader. We do not all mean the same things by these words. Premodern Christians did not all mean the same things by them either. The historian who fails to track these shifts misses much of what the figures themselves were saying.²³

And John Henry Newman, in his 1845 Essay on the Development of Christian Doctrine, supplied the vocabulary in which the question I am pursuing in this book can be addressed. Newman’s seven “notes” for distinguishing genuine doctrinal development from corruption are, in their way, methodological criteria for evaluating which strands of the tradition can claim continuity with the apostolic faith and which represent departures.²⁵ I do not share Newman’s Roman Catholic ecclesiology, and my own use of his criteria is necessarily Protestant in shape. But the question Newman set himself — what is genuine development, what is corruption, and how do we tell the difference? — is precisely the question this book is pursuing, applied specifically to the doctrine of the fate of the dead.

There is also something else the method does. It restores the figures themselves to us. When we read Origen carefully, not as a champion of modern universalism but as a third-century Alexandrian theologian wrestling with the deep questions of his own day, we meet someone richer and stranger than the slogan. When we read Augustine carefully, not as a stand-in for modern strict traditionalism but as a fourth-century North African bishop arguing with his neighbors, we meet someone more complicated and more honest than the caricature on either side. When we read Gregory of Nyssa carefully, we meet a philosopher-theologian of remarkable depth whose universalism is bound up with his account of human nature in ways that should give every reader — universalist or not — something to think about.

The figures of the tradition are not partisans in a twenty-first-century evangelical debate. They were doing their own work. The least we owe them, after we have inherited so much from them, is the courtesy of reading them as they wrote.

A Personal Note in Closing

I want to close by saying something personal. I came to this project after years of pastoral and academic work on the questions of the afterlife. I have buried friends. I have sat with grieving families. I have wrestled with the question of what to say to a Christian whose unconverted parent has just died. I have studied near-death experiences and the question of what the soul is and what happens at death.²¹ I have wrestled, for many years, with how to reconcile the love of God for all human beings with what the New Testament says about the seriousness of final judgment.

In all of that wrestling, the single most useful thing I have done for myself, theologically, has been to read more widely in the Christian tradition than my evangelical training had prepared me to read. The American evangelical tradition I grew up in handed me a thin selection of the Christian past: Augustine for the doctrine of sin and grace, Calvin and Edwards for the doctrine of God’s sovereignty, Spurgeon and Lewis as the great pastoral voices, and a vague sense that everyone else in church history had either gotten the basics wrong or were too Catholic to bother with.

What I found, when I started reading more widely, was that the Christian conversation about the fate of the dead is far richer than my training had suggested. Origen was not a Gnostic heretic; he was the most serious early Christian theologian of the Hellenistic age. Gregory of Nyssa was not a footnote; he was one of the four great Eastern Fathers, and his universalism is bound up with his Trinitarian theology in ways the West has rarely understood. Isaac of Nineveh, whom I had never heard of before seminary, turns out to have written some of the most beautiful and theologically careful pages on the mercy of God in the whole Christian tradition. Wesley’s sermons on grace are not what American evangelicalism inherited from him. Barth on election cracks open the Reformed tradition in ways most evangelical Reformed readers have not registered.

The shift was not from one position to another. I did not move from strict traditionalism to confident universalism. I moved from a thin reading of the tradition to a thicker one. The thicker reading did not give me confidence in a single eschatological position; it gave me a sense of how complicated the Christian conversation about these questions has always been, and how much honest work the question still requires. When I encountered Augustine after reading Origen, Augustine was richer than I had thought. When I encountered Calvin after reading the medieval mystics, Calvin was sharper and more interesting than the textbook version. The figures of the tradition, taken on their own terms, are larger than the labels we attach to them.

That experience is part of what shaped this book. I have tried to write the chapters that follow in a way that gives the reader something like the experience I have had — the experience of meeting the figures as figures, not as proxies in a contemporary debate. If the reader who finishes this book comes away with a more vivid sense of Origen, of Gregory of Nyssa, of Augustine, of Calvin, of Wesley, of Barth, of Balthasar, of Moltmann — and of all the smaller voices clustered around them — then I will have done my job. The contemporary theological argument will still need to be made on its own terms. But the historical map will be more accurate, and the conversation in the present will have more partners than American evangelical writing has usually allowed for.

None of this is heresy hunting in the other direction. None of it requires me to throw out Augustine or Calvin or Edwards. The discipline I have tried to model in this book is the discipline of reading them all — the strict tradition and the long hope tradition alike — as serious Christian theologians doing serious Christian work. They are not all correct. They cannot all be correct, because they disagree. But they are all worth reading. They all repay careful attention. And they all, in their different ways, were trying to be faithful to the same Scripture, the same Lord, the same gospel.

That is the spirit in which I have tried to write the chapters that follow. I will fail in places. Every historian fails in places; the material is too vast and too contested for any single writer to do justice to all of it. But the discipline I have just sketched is the discipline I have tried to maintain. Primary sources first. No anachronism. Distinguish the figure from the reception. Distinguish the primary from the speculative. Acknowledge contested scholarship. Distinguish formal teaching from devotional speech. Refuse both apologetic and polemical distortion.

If I have done my job, the reader will finish this book with a more accurate picture of the Christian tradition than the one most of us were handed in seminary or in the popular evangelical press. The reader will know that the long hope of postmortem opportunity — in some form — has been alive in the Christian conversation from the beginning. The reader will know that it was held by figures of remarkable seriousness, by men and women of unimpeachable orthodoxy on other questions, by theologians who lived close to the apostles and others who lived close to the great Christian centuries afterward. The reader will know that the dominant Augustinian Western position was achieved, not given; that it was won in argument against fellow Christians, not handed down from an undisputed early consensus; and that the modern recovery of softer positions is not the introduction of something foreign but the resumption of a long-standing Christian conversation.³¹

That is the historical map this book aims to draw. Whether the long hope is true — whether God will in the end save more than the strict tradition has allowed for, by means richer than death has been allowed to close — is a question for Scripture and theology to answer. The historical map is the modest gift this book intends to offer.

So now we begin. The next chapter will distinguish the major positions we will encounter — universal restoration, postmortem opportunity, purgatorial purification, the descent of Christ, conditional immortality, inclusivism, soul-sleep — so that as we trace the historical record, we will know which kind of position each figure is holding. After that, we turn to the Apostolic Fathers — the generation closest to the apostles — and listen carefully to what they actually said about the dead.

The dead are still speaking. Our job is to learn to listen.