Chapter 17

Can the Devil Be Redeemed?

The Title Argument, the Hardest Question, and the Scope of the Hope

A. The Argument on the Table

McClymond named his two-volume work The Devil’s Redemption. That title is not decoration. It is the argument.

Pick up the book. Read the spine. The message is plain before you open the cover: if universalism is true, then the devil himself must be saved. And if that sounds absurd, then universalism is absurd. I have spent the last sixteen chapters answering McClymond’s best arguments, one by one. But this is the one that gives the book its name. This is the argument he wants you to remember when you set his two volumes down.

It is the hardest chapter in this book. I will not dodge it. I will not hide behind technicalities or wave the question away. The title argument deserves a straight answer, and I mean to give one.

The logic runs like this. If grace really does win every will in the end, the principle cannot stop at the human race. Satan and the fallen angels are also rational beings with wills. If God’s love is truly final and truly irresistible-in-the-end, then it must reach the demons too. But the redemption of the devil is so clearly contrary to Scripture and to the deep instinct of nearly every Christian who has ever lived that it serves as a reductio ad absurdum (a proof by absurd consequence). The entailment, McClymond argues, refutes the whole project.¹

McClymond frames his proposition with care. He notes that universalism amounts to the claim that there are no creatures who, given a choice, will fail to believe and turn to God.² And creatures, on this framing, includes demons. The title is the reductio. The work is the case behind it.

I want to honor this argument before I answer it. It is not a throwaway line. It is the punch McClymond has been building toward across more than 1,200 pages of careful scholarship. And it lands on something real: some universalists really have taught the salvation of the devil. The entailment is not invented.

B. Why It Has Force

The reductio has force because it is historically grounded. Origen, the father of Christian universalism, did seem to extend the hope to the fallen angels.⁴ In On First Principles, Origen wrote that the destruction of the last enemy (death) does not mean the enemy ceases to exist. Rather, “the hostile purpose and will which proceeded not from God but from itself will come to an end.” He added: “For to the Almighty nothing is impossible, nor is anything beyond the reach of cure by its Maker.”⁵ The process, he said, would extend “even to the last enemy.”

Origen was not alone. McClymond documents that Radical Reformers like Clement Ziegler appealed to Origen for the idea that pagans and demons would be saved.⁶ The Russian theologian Sergius Bulgakov went further. He wrote that “Satan’s very being, his createdness by the omniscient God, is, so to speak, an ontological proof of the inevitability of his future salvation.”⁷ Bulgakov even described a coming “duel” between Satan and God after the Second Coming, one that Satan would finally lose.

So the charge is not a straw man. Real universalists, serious ones, have said exactly what McClymond accuses them of saying. McClymond did not invent the entailment. He found it in the record and held it up for all to see. And the instinct against it runs deep. The council that condemned Origen’s system in 553 did so partly because it linked the restoration of all things with the restoration of the demons (see Chapter 16). Even the medieval Cathars, themselves universalist about human souls, drew the line at Satan. If even other universalists have balked at this step, the instinct must be taken seriously.

De Young presses a further angle. He argues that the basis for human salvation does not even exist for the angelic. The author of Hebrews says that Christ took on human nature, not angelic nature, to save (Hebrews 2:16). De Young puts it bluntly: Christ “died to save people by becoming the God-man. He did not die to save angels by becoming the God-angel.”⁸ If there is no atonement for the angelic, there is no mechanism for their redemption. The instinct that says “the devil cannot be saved” is backed by a real theological argument about the nature of the Incarnation.

James Beilby adds a related point. The Christian tradition has been nearly universal in affirming that the demons are not candidates for salvation, and Beilby identifies a plausible reason. Angels, unlike human beings, walk by sight rather than by faith. Where humans grope forward through ambiguity and incomplete knowledge (Hebrews 11:1), the angels possessed a direct experience of God before their rebellion. Their choice, on this view, was made with full information. What new realization could reach them that they did not already have?⁹ That is a question I cannot easily answer, and its difficulty is part of why this chapter is the hardest in the book.

I take all of these points seriously. The historical record is real. The Hebrews argument is real. The weight of Christian tradition is almost entirely on McClymond’s side here. The question is whether his reductio actually reaches the position this book defends.

C. The Response

1. The Hope This Book Defends Is for Human Persons

Here is the line I want to draw, and I want to draw it clearly. The hope defended in this book is for human persons. Every person God has made will, in the end, freely come to saving faith in Jesus Christ. That is the claim. The scope is the human race redeemed in Christ.¹⁰

Look at the biblical texts that carry the universal hope. God “desires all people to be saved and to come to the knowledge of the truth,” Paul writes, and the “all people” is immediately linked to the “one mediator between God and men, the man Christ Jesus, who gave himself as a ransom for all” (1 Timothy 2:4–6).¹¹ The word is “people” (Greek anthrōpous, “human beings”). Paul names the scope: the human family. And the ransom he points to is the death of a man for humans.

The great Adam-Christ parallel in Romans 5 and 1 Corinthians 15 drives the same point. “As in Adam all die, so also in Christ shall all be made alive” (1 Corinthians 15:22). The “all” who die in Adam are human beings. The “all” made alive in Christ are the same set.¹² The demons never fell in Adam. They are not part of the Adam-Christ parallel. The entire structure of the universal-hope texts is built on human-directed promises.

De Young is right on a point that helps my case: the Incarnation is for humans. Christ took on human flesh to save human beings (Hebrews 2:14–17). The cross is the sacrifice of the God-man for human sin.¹³ This is actually a reason the human hope is well-grounded. It does not follow that the same certainty extends to creatures who fell by a different kind of rebellion, before Adam, outside the scope of the Incarnation and the cross.

2. The Powers Are a Different Question

I am not saying the powers cannot be saved. I am saying their fate is a different question from the human hope, and the human hope does not depend on the answer.

Scripture treats the demonic differently from the human, consistently. Satan is “the ancient serpent” (Revelation 12:9), the accuser, the adversary. He is named as an enemy throughout the biblical story, from Genesis 3 to Revelation 20. The demons “believe—and shudder” (James 2:19), but Scripture never calls that belief “saving faith.”¹⁴ Jesus says the eternal fire was “prepared for the devil and his angels” (Matthew 25:41). Revelation speaks of the devil being “thrown into the lake of fire” and “tormented day and night forever and ever” (Revelation 20:10).¹⁵

These are hard texts. I do not dismiss them. The age-language question applies here as it does everywhere (see Chapter 22), and the lake of fire receives its full treatment in Chapter 24. But the point for this chapter is simpler: the biblical picture of the demonic is not the same as the biblical picture of fallen humanity. Humans are made in the image of God. They were given the promise of a Redeemer in the very hour of the Fall (Genesis 3:15) and pursued by a Savior who became one of them. The powers have no such promise, no such Incarnation, and no such Redeemer who shares their nature.

Consider how differently the New Testament speaks of human sin and angelic rebellion. Paul tells the Romans that “while we were still sinners, Christ died for us” (Romans 5:8). Nowhere does Paul say Christ died for the sins of the angels. The entire atonement vocabulary of the New Testament is shaped around human sin, human death, and human rescue. When Peter speaks of those “angels who sinned,” he says God cast them into chains of darkness, “to be kept until the judgment” (2 Peter 2:4). The language is confinement and waiting, not redemption and hope.

That does not settle their fate. It means their fate is genuinely open in a way that the human hope, built on specific promises, is not.

3. Three Honest Options, Held Loosely

When I look at what Scripture says about the fallen powers and what honest Christians have concluded, I see three live possibilities.

First, the powers may be finally destroyed. If the lake of fire is the consuming presence of God, the demonic forces may be consumed entirely. The evil is burned up. The beings themselves cease to exist. This view has roots in conditional immortality and in the way some early fathers read “the last enemy to be destroyed is death” (1 Corinthians 15:26).¹⁶

Second, the powers may be a permanent exception to the restoration. God’s saving purpose is for human persons. The demonic, having fallen from a different kind of knowledge and by a different kind of rebellion, remain the permanent enemies of God, kept under judgment, while the human family is brought home. Their defeat is Christ’s victory. Their conversion is not required for God to be “all in all.”¹⁷ Paul says Christ must reign “until he has put all his enemies under his feet” (1 Corinthians 15:25). “Under his feet” is the language of conquest, not of persuasion. The enemies are subdued. God’s order is restored. That may be the final word on the powers.

Third, following Origen and some moderns, even the powers may be finally restored. The love of God has no limit. If even the most deeply fallen will could, given enough time and enough grace, freely turn, then perhaps even the angels who fell could come home. This is the boldest option, and the one McClymond takes as the reductio.¹⁸

Robin Parry, writing as Gregory MacDonald in The Evangelical Universalist, tentatively offers two creative ways to hold the third option. One is to read the devil not as a personal being but as a personification of evil. If so, the “torment” in Revelation 20:10 is the permanent defeat of a system, not the suffering of an individual. Death and Hades, after all, are also thrown into the lake of fire, and they are not conscious beings (Revelation 20:14).¹⁹ The other proposal is bolder: the devil could be punished forever while Lucifer, the fallen angel behind the persona, is finally restored. Just as the “old self” must be put to death so the new self can live (2 Corinthians 5:17), the “devil” as an identity of evil could be destroyed while the angel is reborn. Parry offers an analogy: the redemption of Anakin Skywalker required the permanent death of Darth Vader.²⁰ He does not claim certainty. But these proposals show that the options are richer than the either/or McClymond assumes.

I will not dogmatize on any of these. I hold them loosely. I lean toward the first or second, not because the third is impossible, but because Scripture gives us so little to build on. The human hope has thick biblical grounding. The angelic question does not. I can hold my confident hope for humanity without needing to resolve the question of the powers. And that is the key: the human hope stands on its own, regardless of which option is correct.²¹

4. Why the Entailment Fails

McClymond’s reductio depends on a hidden premise. It assumes that universalism is a single abstract principle: “all rational wills are saved.” If that is the principle, then yes, it must include angelic wills. The reductio follows.

But that is not the principle this book defends. The hope defended here is not an abstract axiom about rational wills. It is a set of human-directed biblical promises. It grows from specific texts, a specific atonement, and a specific Incarnation. The cross of Christ is the ground of human salvation. The Adam-Christ parallel defines the scope. The Bible’s saving promises name their recipients: “all people,” “all who fell in Adam,” “the world.” The demons are not in Adam, were not ransomed by the God-man, and are not named in those promises.²²

This is also where the book’s non-necessitarianism pays off (see Chapter 5). If I held, as some do, that universal salvation is metaphysically necessary because no rational will can finally resist God, then the reductio would bite. An angelic will is still a rational will, and a metaphysical necessity makes no exceptions. But I do not hold that. I hold that God’s love and wisdom are sufficient to win every human heart, through the specific means Scripture names, without coercion. The human hope is built on promises, not on metaphysics.²³

McClymond himself frames the issue in a way that exposes the gap. He writes that universalism amounts to claiming “there are no intelligent creatures who, given a choice, will ultimately fail to believe, repent, and turn to God.”²⁴ But that is the necessitarian version. The biblical version is narrower: there are no human persons whom God will fail to bring to faith in Christ, because God has promised it and the cross has accomplished it. The jump from “all people” to “all intelligent creatures” is a jump the texts do not make.

Think of it this way. A father promises to bring all his children home for a reunion. That promise does not commit him to bringing home every stranger in the country. The scope is defined by the family. In the same way, God’s saving promises are defined by their recipients: the human family, fallen in Adam, redeemed in Christ.

5. The Free-Will Residue Applies Most Here

I have been honest throughout this book that the bare logical possibility of final free refusal cannot be ruled out by force (see the author’s position statement in the introduction, and the fuller treatment in Chapter 26). A genuine “yes” requires a possible “no.” For human beings, I believe the love of God is sufficient to win every heart, because God knows what each person would freely choose in every situation He could arrange, and He arranges the best possible appeal for each soul.²⁵

But if any creatures could finally, freely refuse God forever, the fallen powers are the likeliest candidates. They sinned with greater knowledge than any human being. They saw God’s glory and chose rebellion anyway. They have persisted in that rebellion across the whole span of human history. If the free-will residue is real anywhere, it is real here.²⁶

Talbott captures the strangeness of this kind of rebellion through Milton’s Paradise Lost. Milton’s Satan is both defiant and tormented. In Book IV, an awakened conscience leads him to consider repentance: “Is there no place / Left for repentance, none for Pardon left?” But the fear of shame and the illusion that he can still rule in hell trap him. He plunges into despair: “Evil be thou my Good.” Talbott observes that this is deeply irrational, no less absurd than choosing torment as pleasure.²⁷ Whether such a grip of self-deception could last forever, or whether even it must yield, is the question no one on earth can answer with certainty.

This does not trouble the human hope. It actually strengthens the boundary. The author’s libertarian honesty (see Chapter 5) means that the non-restoration of the fallen powers is entirely coherent with the confident hope for humanity. If human universalism were a metaphysical necessity, it could not survive the possible non-restoration of the angelic. But because it rests on promises, atonement, and the God who became human, the human hope can stand while the angelic question remains open.

Let me put it another way. I believe God knows how to reach every human heart (see Chapter 26 on middle knowledge). He has eternity to work. He has the cross to offer. He has His own unveiled presence to show. For human beings, these resources are sufficient. For fallen angels, who sinned from a higher vantage point and with a different kind of will, I do not know if the same tools apply. And I do not need to know, because the human hope does not depend on the answer.

6. Pastoral Candor

I want to name the discomfort, because I think honesty is more helpful than a neat answer in either direction.

Some readers will want me to say, “Yes, even the devil will be saved.” I will not say that. Scripture does not give me enough to build on. The texts that address the demonic point more toward judgment than toward restoration. I do not know the devil’s final fate, and I will not pretend to.

Other readers will want me to say, “The devil is lost forever, case closed.” I am not sure I can say that either, because I worship a God for whom nothing is impossible (Luke 1:37), and because Origen’s instinct that God’s creative power cannot be finally defeated by any creature is not easily dismissed.²⁸

What I can say is this: the question of the devil’s fate is not the question of this book. The question of this book is whether every human being will be brought to saving faith in Christ. That question has thick biblical answers. The demonic question does not. And the two questions can be separated. They should be separated. A good surgeon does not refuse to operate on the patient he can save because there is another patient he cannot.

I am content to rest where Scripture rests: confident of Christ’s victory, confident of the human family’s salvation, and honest about what lies beyond the horizon of what God has told us.²⁹

D. Engagement with Specific Scholars

McClymond’s reductio gains its power from the historical record, and the record is real. Origen did extend the hope to the powers. He wrote that “the process will go on through the innumerable ranks of those who are making progress and becoming reconciled to God from their state of enmity, until it reaches even to the last enemy, who is death, in order that he, too, may be destroyed and remain an enemy no longer.”³⁰ Bulgakov went further, calling the devil’s createdness “an ontological proof of the inevitability of his future salvation.”³¹ McClymond is right to name these voices. They exist. They said what he says they said.

But the universalist family is not uniform on this point, and McClymond knows it. Thomas Talbott, one of the most careful philosophical defenders of universalism, frames his case around three propositions: that God wills the reconciliation of all sinners, that God has the power to accomplish this, and that some sinners will never be reconciled. McClymond himself explains that Talbott’s strategy is to deny the third proposition while holding fast to the first two.³² But Talbott’s denial is aimed at human persons. His argument, as McClymond reports it, is a variation on the problem of evil applied to the specific evil of sinners who remain unreconciled, not to the fate of the angelic. Talbott asks why any person who truly came to know God would choose eternal misery, and he builds his case on human psychology, human freedom, and human encounter with divine love. The demonic is not where his evidence concentrates.

Talbott also presses the Colossians 1:20 question. He insists that Paul’s reconciliation language is explicitly redemptive. Against scholars like Peter O’Brien, who read the reconciliation of the powers as mere subjugation, Talbott argues that forced subjugation is not what Paul means by peace-making through the blood of the cross.³³ This is a serious point. But it does not undermine my central claim: the human hope rests on human-directed promises, whatever one concludes about the cosmic scope of Colossians 1:20.

Andrew Hronich, working in the same philosophical tradition, represents a similar range: the human hope is where the biblical evidence points most clearly, and the demonic question is an open one that the human case does not require. Hronich draws attention to a detail that strengthens the distinction: the subjection language in 1 Corinthians 15 and Ephesians 1 does not necessarily exclude salvation. The church is also “subjected” to Christ, yet that subjection is entirely compatible with redemption. As Christ is subjected to the Father in 1 Corinthians 15:28, so the cosmos is subjected to Christ; the focus is on the re-establishment of God’s rule and harmony, not on the annihilation of the subjected.³⁴

Robin Parry, in The Evangelical Universalist, presses the Colossians text further. The poem in Colossians 1:15–20 is built on a structural parallel between creation and reconciliation. The “all things” created “in him” in verse 16 are the same “all things” reconciled “through him” in verse 20. The powers are named in both halves. Parry argues that reading their reconciliation as punishment strains the word, since Paul connects reconciliation with “making peace” through the cross, and in Romans 5:1 and 5:10, reconciliation and peace with God clearly express salvation.³⁵ He also notes that the powers’ defeat at the cross (Colossians 2:15) opens the way for their reconciliation; they had to be overcome before they could be restored.³⁶ I find this compelling, though I stop short of claiming it settles the question. Colossians 1:20 may carry a broader scope than I can prove. What I can prove is that the human hope does not depend on resolving that scope.

David Bentley Hart, whose That All Shall Be Saved makes the strongest recent case for the restoration of all things, would extend the scope more broadly. Hart argues from the doctrine of creation out of nothing (creatio ex nihilo) that if God freely creates, knowing the outcome, then no creature can be finally lost without that loss becoming part of God’s positive intention. For Hart, this means God must save all rational creatures, angelic and human alike, or the Kingdom is incomplete.³⁷ Here I part company with Hart in the way I have parted with him before (see Chapter 5). His case rests on a metaphysical necessity I do not share. My case rests on promises and atonement. Hart’s exegetical and historical work remains among the most valuable in print, and I use it gratefully. But his scope exceeds what Scripture explicitly warrants for the demonic.

De Young’s final move, though, is the one I dispute. He concludes that if hell is permanent for the devil, it must be permanent for humans too. But that does not follow. The Incarnation is the very thing that makes the human situation different. Christ became human. He died for human sin. He ransomed human beings. The permanence of angelic judgment does not require the permanence of human judgment, because the basis of salvation is different for the two kinds of creatures. De Young’s own argument from the Incarnation actually supports the distinction.³⁹

De Young also presses the question: if fallen angels cannot be restored, does universalism not hold God hostage to His enemies? I understand the worry. But the biblical picture cuts the other way. God’s victory over the powers is described as their defeat, not their conversion (Colossians 2:15). Their subjugation is God’s triumph. What the human hope adds is that among the defeated, the human rebels are also healed. Their defeat by fire is also their rescue through fire. The powers may receive only the first. Humans receive both.

E. Counter-Objections

“You cannot believe love wins every human will and then exempt the angelic will.”

This objection assumes that the ground of the hope is a universal principle about love and wills. But the ground is the cross, the promises, and the Incarnation in human flesh. The principle “love wins every will” is an inference some universalists draw, but it is not the foundation. No principle forces the human hope to scale to the angelic.⁴⁰ A doctor may have the skill to heal every patient in a hospital, but his promise to heal applies only to those under his care. God’s love is unlimited. His promises have a named scope.

“Colossians 1:20 says ‘all things,’ so you must include the powers or abandon the verse.”

Colossians 1:20 says that God was pleased “through him to reconcile to himself all things, whether on earth or in heaven, making peace by the blood of his cross.” This is a sweeping text, and I take it seriously. Parry has shown that the structural parallel with verse 16 makes it hard to narrow the scope.⁴¹ But “reconcile all things” need not carry the same form for every creature. The reconciliation of the cosmos can include the defeat of the hostile powers, the restoration of order, and the healing of every human rebel. Even Parry acknowledges that the powers’ reconciliation follows their defeat; they had to be overcome before they could be restored.⁴²

A full treatment belongs in Chapter 25 and Chapter 29. For this chapter: even if Colossians 1:20 includes the powers, the form of their reconciliation may differ from the human. Pacification is a form of reconciliation too. I believe the human rebels will be converted. For the angelic, the text leaves more than one option open.

“If God’s love has no limit, it cannot have a limit at the angelic.”

God’s love has no limit. But the promise has a named scope. I believe God’s love extends to every corner of creation. I do not know what that love will finally do with the fallen powers. I know what it will do with human beings, because Scripture tells me. The limit is not in God’s love. The limit is in what God has revealed.⁴³

F. Conclusion and the Boundary

This is the hardest chapter in the book, and I have tried to be honest. McClymond is right that some universalists have extended the hope to the devil. They have. Origen did. Bulgakov did. The entailment is not invented. But the entailment follows only from the abstract principle, not from the biblical promises. The universalism this book defends is not a metaphysical system that must save every rational creature by logical force. It is a hope grounded in the cross, the Incarnation, and the specific saving promises of God for human beings.

The devil’s fate is an open question. I lean toward the first or second of the three options I laid out: destruction or permanent judgment. I do not claim to know. What I do claim, with confidence, is that the question of the devil’s fate does not have the power to refute the human hope. The promises stand. The cross holds. The Savior who became human to save humans will finish what He started.

Think about what McClymond is asking the reader to do. He is asking you to give up the hope that your neighbor, your friend, your lost child will be brought to Christ, because you cannot be certain about the fate of fallen angels. That is a strange trade. The human hope is thick with biblical support. The angelic question is thin and unresolved. I will not surrender the thick for the thin.

Grace for sinners is not refuted by mystery about angels. The title of McClymond’s book names a hard question. It does not name a refutation.⁴⁴

The cumulative case continues in Chapter 40. The answer to McClymond’s title will be simple: the universalism he named his book after is not the one this book holds. The devil’s redemption belongs to Origen and Bulgakov. The human family’s redemption belongs to the gospel. That is the strand that survives.

There are things given to us to know, and things that are not. The salvation of the human family in Christ is given. The rest, we leave in the hands of a God whose judgments are true and whose mercy we trust even where His purposes pass beyond our sight.