Pre-, post-, or amillennialism?
Once you’ve wrestled with the details of Revelation 20 itself, a bigger question is usually waiting right behind it: which overall system does this chapter fit into? Historically, Christians have answered that question in three main ways, named for where they place Christ’s second coming relative to the 1,000 years: premillennialism (before), postmillennialism (after), and amillennialism (the number describes something other than a future golden age).
It is tempting to set these out as three lists of arguments and count which list is longest. That is not how the disagreement actually works. All three views are held by careful readers who know the same texts, and each of them can answer the popular objections to it in its sleep. The disagreement lives in four specific places, and if you understand those four, you understand the debate — including why it has not been settled in eighteen centuries.
The three views, briefly
Premillennialism. Jesus returns first, defeats the beast and the false prophet, and then reigns bodily on earth for 1,000 years before Satan is released for one final rebellion and the new creation arrives. It comes in two main varieties — historic premillennialism, where the church goes through the tribulation, and dispensational premillennialism, which adds a pre-tribulation rapture and a separate future for ethnic Israel.
Postmillennialism. The 1,000 years, taken as a long period rather than a literal count, describes an age in which the gospel progressively succeeds and the nations are, in large part, actually discipled — before Jesus returns to a world his gospel has already substantially won.
Amillennialism. The 1,000 years symbolically describes the whole church age, from Christ’s first coming to his second — now — during which Satan is restrained from deceiving the nations as he once did, and believers already reign with Christ in a real but not yet visible sense. This is the reading the previous article argues for.
None of the three is a modern invention. But the historical picture is not evenly balanced either, and it is worth being honest about that: the earliest identifiable readings of Revelation 20 that we can still name — Papias, Justin Martyr, Irenaeus, Tertullian — are millennial in something like the premillennial sense, and Justin is explicit that he holds it while acknowledging that other orthodox Christians do not (Dialogue with Trypho 80). Several of them wrote in or near Asia Minor, the book’s own home region. The amillennial reading became the majority position later, above all through Augustine, who had held a millennial view himself before rejecting it.
Crux 1: two comings to life, one verb
This is the sharpest point in the whole debate, and it is where premillennialism is strongest.
Revelation 20:4–5 says that the martyrs came to life and reigned with Christ, and then, a few words later, that the rest of the dead did not come to life until the thousand years were ended. The Greek verb is the same in both clauses: ἔζησαν.
The premillennial argument. Everyone agrees that the second one means bodily resurrection from the grave. So the first must too — and if the first is a bodily resurrection, it happens before the thousand years, and there is a thousand-year gap between two bodily resurrections. Henry Alford’s nineteenth-century formulation is still the one people reach for: if the same verb, in the same sentence, can mean spiritual rising in one clause and bodily rising in the next, then language has stopped conveying anything definite at all. Premillennialists then point at the book’s own usage, which is narrower than it looks. This aorist form occurs only seven times in the whole New Testament, and four of them are in Revelation. One of those four is not in dispute at all: he became dead and came to life, of Christ. A second is the beast whose sword-wound healed, and it came to life — a deliberate counterfeit of it, though Beale reads that one figuratively, of a political revival rather than a body. Either way, the book uses this verb for a return from death, and uses it of Christ himself. Aune’s commentary — the standard critical treatment, and one of this page’s own sources — reaches the same verdict and presses it further: the verb here carries the sense “raised from the dead, resurrected,” and however awkward the notion of two resurrections may be, it is difficult to avoid concluding that the author has formulated exactly that.
The amillennial answer comes in four parts, and it is more substantial than it is usually given credit for.
- The subject of the verb in verse 4 is not bodies but souls — John sees “the souls of those who had been beheaded.” Souls coming to life and reigning is a natural description of what happens to a martyr at death, not a description of a body leaving a grave.
- The word “first” is doing structural work, not counting. Revelation uses first and second to mark two orders, not two items in a series: the first heaven and the first earth are the old order that passes away. So the “first resurrection” is the one belonging to the old order — the believer’s passage at death into life with Christ — and the “second death” is the death belonging to the new order. The two pairs are deliberately crossed, which is why the passage can promise that over those in the first resurrection the second death has no power. Beale sets this out as a deliberate crossing, following Meredith Kline: the saints’ first death is physical and their first resurrection spiritual, while the wicked’s second resurrection is physical and their second death spiritual. The pairs are meant to be read across each other, not counted off in sequence.
- A verb can keep one meaning — to come alive — while what comes alive differs. Souls come alive at death; bodies come alive at the end.
- And Revelation does use this verb of spiritual life elsewhere. To Sardis: you have a name that you are alive, and you are dead — the same pairing of living and dead that 20:4–5 makes, and unmistakably not about bodies.
And the New Testament already speaks this way. The sharpest objection to all of the above is that it sounds invented for the occasion — spiritualising a resurrection because a theory needs it. But raising, being made alive and being enthroned are what the apostles ordinarily say about Christians who are still walking around. Paul tells the Ephesians that God, when they were dead in trespasses, made them alive together with Christ, raised them up with him and seated them with him in the heavenly places. He tells the Colossians they have been raised with Christ, and the Romans to reckon themselves dead to sin but alive to God and to present themselves as those alive from the dead. John says the believer has already passed from death into life.
Two details make that more than a loose parallel. The verb behind “seated them with him” in Ephesians 2:6 occurs only twice in the New Testament, and its single theological use is that verse — believers enthroned with Christ already; Revelation 20:4 opens with the same picture, thrones and people sitting on them. And Romans sets alive against dead of living Christians in exactly the way Revelation does at Sardis and again at 20:4–5.
This is also where the position came from. Augustine, who gave amillennialism its currency, argued from those very verses — Colossians 3:1 and Ephesians 2:6 — that the saints’ reign with Christ is the present condition of Christians rather than a future expectation.
A third answer, and it is available to everyone. Both readings above assume that the verb names a mode of existence — you exist either spiritually or bodily — and then argue about which is meant. Alford’s objection depends entirely on that assumption. If the verb names a status instead, one sense covers both clauses without switching: the martyrs entered vindicated life, and the rest did not.
The verse invites this, because it opens by quoting Daniel. Revelation has already used Daniel 7:21 at 13:7 — the beast “makes war with the saints and conquers them,” almost word for word. Revelation 20:4 then supplies Daniel’s next verse: “judgment was given to them” is Daniel 7:22, where the Ancient of Days reverses a verdict the saints had lost and they receive the kingdom. In Daniel the phrase does not mean the saints were handed authority to judge others; it means the court found for them. Beale reads 20:4 against Daniel 7:22 in just this way and summarises the verse as the saints’ vindication.
Revelation uses the noun that way itself at 18:20 — “God has judged your case against her” — and the martyrs’ cry, how long until you judge and avenge our blood?, is answered in those same two words at 19:2.
On this reading 20:5 says that throughout the thousand years the rest of the dead received no such verdict, and the “until” marks the limit of the statement rather than promising them one afterwards. Alford then has nothing to bite on: the sense is constant and only the recipients differ. What this does not do is settle the millennium — a premillennialist can accept every word and hold that the vindication takes the form of a bodily raising. What it removes is the charge that amillennialism must make one verb mean two things.
One more piece of evidence, and both sides claim it. The one place in Revelation where God’s servants plainly come back from death within the narrative is the two witnesses — and John describes it without this verb at all: a breath of life from God entered them, and they stood on their feet. He is quoting Ezekiel 37:10, where the Greek runs “the breath entered them, and they came to life, and they stood on their feet.” John keeps the entering and the standing almost word for word, and drops the verb in the middle. So this is not his fixed idiom for a body raised, and choosing it at 20:4 was not forced — but equally, you cannot argue that he would have written 20:4 the way he wrote 11:11 if he had meant the same thing there.
Underneath all of it sits one question, and the commentators split on it. Who has a share in the first resurrection? Aune reads 20:4 as a single group of actual martyrs, executed for the testimony. Beale reads it as everyone who kept faith to the end, “whether or not by martyrdom,” on the grounds that 20:6’s promise of immunity from the second death is not something the book restricts to a subset of believers. The choice decides who “the rest of the dead” are in 20:5 — the ungodly, on Beale’s reading, or believing and unbelieving dead together, on Aune’s.
Where that leaves it. Premillennialism still has the more immediate reading of the sentence, and that counts for something. But the amillennial answer is not the ad hoc move it is often taken for: it uses the New Testament’s ordinary language for Christian existence, it has a worked account of first and second, and on the vindication reading it does not equivocate at all. Readers who weigh the plain sequence of the sentence most heavily will still land premillennial, and they are not being careless when they do.
What actually carries the theme
Tracking one verb can mislead, because Revelation carries death-and-life mostly through a picture rather than a word: slain, and then standing.
The image is set at the centre of the book — a Lamb standing as though it had been slain — and then handed on to everyone who follows him. The martyrs under the altar get the very participle used of the Lamb: the souls of those who had been slain. The countless multitude out of the great tribulation is standing before the throne. The two witnesses stood on their feet. The 144,000 are standing with the Lamb on Mount Zion, those who conquered the beast are standing on the sea of glass, and at the end the dead are standing before the throne.
Standing is the posture of vindication after being killed, and it belongs first to Jesus and then to his people.
The counterfeit runs alongside it — and notice which of the two uses the plain word. The beast has a head as though slain to death, and its death-wound was healed, and it came to life. The beast’s resurrection is stated; the church’s is shown.
This is worth holding onto through the argument that follows, because it means neither side settles Revelation 20 out of a concordance. Whatever 20:4 turns out to mean, the book’s hope for its readers is not stored in that one verse. It is distributed across every scene in which somebody killed for the testimony ends up on their feet.
Crux 2: what “bound” means
Revelation 20:1–3 piles up four verbs on Satan: an angel seized him, bound him, threw him into the abyss, and shut and sealed it over him.
The premillennial argument. No one seals a door over someone who is still walking around outside. Four verbs of total incapacitation, plus an abyss, plus a seal, do not naturally describe a restriction on one department of a devil’s activity. And the New Testament goes on describing him as active: he prowls like a roaring lion, looking for someone to devour, and he is called the god of this age, who has blinded the minds of unbelievers.
The amillennial answer. The text supplies its own limit, in a purpose clause most summaries leave out: he is bound “so that he should not deceive the nations any longer” — one named effect, not a general disappearance. And that effect is exactly what changed. Before Christ, the knowledge of God was largely confined to one nation; after Pentecost the gospel goes to the nations and they come in. Jesus described his own ministry in precisely these terms: no one can plunder a strong man’s house unless he first binds the strong man, and I saw Satan fall like lightning from heaven. On this reading the world mission is not evidence against the binding; it is the binding, visible.
The rejoinder to Peter’s roaring lion follows from the same distinction: Peter is describing danger to individuals, and 20:3 is about deceiving nations as such. Both can be true at once.
Where that leaves it. This crux runs the other way from the first one. The purpose clause is in the text, and it does limit the scope; premillennialism has to treat a stated limitation as if it were not there. But the sheer weight of the imagery is a real argument, and “bound, but only in one respect” will always sound like less than the picture shows.
Crux 3: does chapter 20 follow chapter 19?
The premillennial argument is usually reported as resting on the little word “and” that opens chapter 20. That is the weakest form of it. There are two stronger ones.
First, 20:10 appears to look back at 19:20. The devil is thrown into the lake of fire “where the beast and the false prophet are also” — a phrase that seems to presuppose that the events of 19:20 have already happened. (The Greek has no verb there; translators supply one. Amillennialists press exactly that point: an elided verb carries little temporal weight.)
Second, and more substantially, Revelation 20 tracks the order of Ezekiel. Ezekiel moves from the resurrection of the dry bones (37), to the assault of Gog and Magog (38–39), to the temple and the restored city (40–48). Revelation 20 moves from coming to life and reigning (20:4), to Gog and Magog (20:8), to the city (21–22) — naming Gog and Magog outright, which is not a common Old Testament phrase. If John is following a source’s sequence, that is evidence of sequence.
The amillennial answer. Recapitulation is not a device invented for this chapter; it is the book’s normal habit, argued at length in the previous article. Chapter 12 has already told the story of Satan’s fall and the church’s protected period, and chapter 20 tells it again from a different angle. The decisive point is that the battle of chapter 19 and the battle of chapter 20 both draw on the single final battle of Ezekiel 38–39: if they are two different battles, John has used one Old Testament battle twice for two separate events. As for Ezekiel’s order, the correspondence is thematic — John draws on Ezekiel throughout the book without following him chapter by chapter. And there is a concrete instance: Ezekiel’s valley of dry bones is the resurrection chapter that opens the sequence, and John has already spent it at 11:11, nine chapters before the millennium, quoting it almost word for word. If he were following Ezekiel’s order, he used its first movement early.
Where that leaves it. The Ezekiel argument cuts both ways and is a good illustration of why this debate persists: the same set of allusions supports “two battles in sequence” and “one battle told twice” about equally well, and which one you find natural depends on how much recapitulation you already think the book does.
Crux 4: how much of the promise lands inside history?
This is where postmillennialism lives, and its case is much wider than the Great Commission.
The postmillennial argument. A long line of Old Testament promises describes an outcome inside history, not after it: the nations given to the Son as his inheritance; all the ends of the earth remembering and turning to the LORD; all kings falling down before him, all nations serving him; the nations streaming to the mountain of the LORD and beating swords into ploughshares; the earth full of the knowledge of the LORD as the waters cover the sea; the stone that becomes a great mountain and fills the whole earth.
The New Testament picks the same thread up, though less cleanly than the case usually admits, and the difference is worth marking. Everything quoted above is a declaration — the earth shall be full, the stone becomes a mountain. Much of the New Testament material is not.
The Great Commission is a command. Make disciples of all nations is an imperative, and the promise attached to it is Christ’s presence, not a rate of success: it commissions a task without forecasting how much of it gets done. The leaven parable does end in completion — the woman hides it until the whole was leavened — but its subject is what the kingdom is like, hidden and pervasive and out of all proportion to its size, and a parable makes one point rather than taking a census.
Romans 11 is the strongest of the three, and it repays quoting precisely. A hardening has come upon Israel in part, until the fullness of the Gentiles comes in — and if their stumbling meant riches for the world, how much more their fullness, when all Israel will be saved. Two different fullnesses are in play, the Gentiles’ in verse 25 and Israel’s in verse 12, and the argument does turn on a future larger than the present.
Revelation gives its own answer to “all nations,” and it is partitive. Twice it describes the redeemed as drawn out of every tribe and tongue and people and nation and out of every nation and tribes and peoples and tongues — out of: some from each, rather than all of each. That is this book’s own description of the outcome. It does not settle the question, but a postmillennial reading has to reckon with it.
The two standard objections, and what postmillennialists say back.
“Revelation shows evil escalating, not receding — the beast makes war on the saints and conquers them.” This objection assumes a particular reading of Revelation, and most contemporary postmillennialists do not hold it. They are partial preterists: they read the beast, the tribulation, and the judgments as describing the first century and the fall of Jerusalem in AD 70, already past. On that reading Revelation is not a forecast of our future at all, and the objection does not reach them. Whether partial preterism is right is a separate argument — see preterist, historicist, futurist or idealist — but it has to be made, not assumed.
“History hasn’t gone that way.” Postmillennialists reply that this measures a claim about the whole age against one or two centuries, and usually against Europe. The same period contains the largest numerical expansion of Christianity in its history, in Africa, Asia and Latin America. They also grant setbacks and long reversals; the claim is about trajectory across the age, not a smooth upward line.
The strongest reply to postmillennialism is not “history disagrees” but Jesus’s own parable of the wheat and the weeds, where both grow together until the harvest and separation is explicitly deferred to the end. Postmillennialists note that the field is the world and that the parable fixes no proportions — it forbids premature separation rather than promising a permanent stalemate. That is a fair reply, though it concedes that the parable is at least not evidence for the view.
And where premillennialism sits on this crux. Dispensational premillennialism answers the same question differently: the land and nation promises to ethnic Israel are concrete and unrepealed, so they need a concrete earthly kingdom to land in, and the millennium is it. Amillennialism’s usual reply is that the promises pass to the church as Israel’s heir. There is a more direct answer, and it is the one this site takes: Israel was never an ethnic category to begin with.
The covenant is chosen against biology in the very passage that establishes it. God blesses Ishmael, multiplies him and makes him a great nation — and in the next breath, but my covenant I will establish with Isaac. Abraham’s own firstborn is blessed and outside the covenant in consecutive verses. Esau and the sons of Keturah go the same way.
The door opened outward just as early. At the Exodus itself a great mixed crowd went up with them, and the law provides that a foreigner who is circumcised may keep the Passover and shall be as the native of the land. Isaiah states the qualification in covenant terms: foreigners who hold fast to my covenant are brought to the holy mountain, and the house is called a house of prayer for all the nations.
So when Paul says not all who are of Israel are Israel, and that the children of the promise rather than the children of the flesh are counted as offspring, he is not transferring anything. He is applying the membership rule the covenant always had — one that admitted outsiders who joined and excluded insiders who did not. Those who belong to Christ are Abraham's offspring, heirs according to promise, and all the promises of God find their Yes in him; they land finally in the new creation, which is where Isaiah’s long life and ordinary work and fruitful ground actually belong.
This is worth stating carefully, because the standard objection to amillennialism is that it “replaces” Israel with the church. On this reading nothing is replaced. The covenant community was always defined by covenant, and it is now what it always was. See Israel and the church.
That last exchange also settles the Isaiah 65 question that premillennialism raises: the passage describes long life and ordinary work but people still dying, which sounds like neither this age nor the final state — except that Isaiah introduces it himself as new heavens and a new earth, the same phrase Revelation later uses for the final state.
And there is a general principle underneath that particular answer. The New Testament characteristically takes an Old Testament promise and raises it rather than simply repeating it — what the literature calls typology with escalation. The clearest single case: God promises Abraham this land, and Paul restates the promise as his being heir of the world. The same move shows up as an omission when Paul quotes the fifth commandment. Exodus promises long life “in the good land which the LORD your God is giving you”; Paul quotes it word for word up to long life on the earth — and stops there. The land becomes the earth by what he leaves out.
Read that way, Isaiah 65 is not sketching an interim stage. It describes the end in the only currency available to its first readers — an idealised version of the life they knew, in which nobody dies young — and the New Testament raises that same promise to the abolition of death itself: death shall be no more. Long life becomes deathless life, exactly as the land became the world.
This is not a device invented to deal with Revelation 20. It is how the New Testament handles the temple, the Sabbath rest, Jerusalem and the throne of David as well, and it is the framework Beale works in throughout.
Where this site lands, and what it costs
The previous article argues for amillennialism, mainly on cruxes 2 and 3: the binding has a stated scope, and the same Ezekiel battle standing behind both chapter 19 and chapter 20 is hard to explain if they are two events. Add to that how closely chapter 20 tracks chapter 12’s account of Satan’s fall, and the recapitulation reading is the one that asks least of the rest of the book.
The honest cost is crux 1. One verb, twice, a few words apart, meaning two different kinds of coming-to-life is the hardest thing amillennialism asks of a reader, and the answer above — souls rather than bodies, first and second marking two orders — is an argument, not a knock-down. It should be said plainly that the two commentaries cited on this page divide over it: Aune decides crux 1 against the reading argued here, and Beale decides it for. A reader who weighs that sentence more heavily than the book’s recapitulation habit will land premillennial, and will be reasoning properly.
Bock, Three Views on the Millennium and Beyond is the place to see each case made at full strength by someone who actually holds it, with the others responding — which is the only format in which this argument is worth having.
A few questions worth sitting with:
- In Revelation 20:4–5, does the same verb have to carry the same sense both times — and if it does, what follows?
- Which of the three pictures of “1,000 years” makes the best sense of Revelation’s other numbers — the 144,000, the 42 months, the 1,260 days? Are those head-counts and calendars, or symbols?
- If Satan is bound, what exactly is the binding stopping — and can you point at the difference it has made?
- How much of what the Old Testament promises to the nations do you expect to see happen inside history, and how much only afterwards? Your answer to that question probably decides your answer to all the others.