Eschatology and Israel

"Not one stone will be left on another. All will be torn down!" — Mark 13:2, NET¹

"Many of the orthodox criticise it as an attempt to seize the prerogatives of their God." — William Blackstone, Christian Zionist, on Zionism, 1908²

Summary

This article asks three questions.  Did Jesus predict the fall of Jerusalem before it happened?  Did he say the end of the world would come within his own generation, and was he wrong?  Does the founding of modern Israel in 1948 fulfil prophecy?  Each claim is scored by the eight tests of How to Test a Fulfilled Prophecy, cited here by number.

1.  The fall of Jerusalem:  a real hit, worth less than it is usually claimed.  Mark 13, Matthew 24 and Luke 21 have Jesus say the temple's buildings will be thrown down, and Luke adds a siege with an embankment, encirclement, slaughter and captives.  All of that happened in AD 70, as Josephus, who was there, and Tacitus report.³  The prediction passes Test 4, because Jesus's followers did not destroy the temple;  the Romans did.  It passes Test 5 for the event.  But it passes Test 1 only if Mark was written before the war, and the site's Gospel dating article finds that undecided:  Mark falls somewhere between the 40s and AD 70.  Two facts favour a genuine prediction.  The discourse leaves out what every eyewitness account of the siege dwells on:  the fire in the sanctuary, the famine, and the faction fighting.  And a saying of Jesus against the temple was remembered at his trial as false testimony, which no one writing after the event would invent, so some such saying goes back before the Gospels.  Two facts limit its weight.  Every detail of Luke's siege is stock language from the Greek Old Testament's accounts of 586 BC, as C. H. Dodd showed in 1947.⁴  And another Jesus, the son of Ananias, cried "Woe, woe to Jerusalem!" through the city's streets from about AD 62 until a Roman stone killed him during the siege.³  A man with no claim to foreknowledge made the same forecast and was right.  Priced from the record of the two centuries before, a well-informed forecaster in AD 30 might have given the razing about one chance in seven.  So even with an early Mark, the prediction is evidence of the strength of a good forecast, a ratio of about seven to one, not of a miracle.

2.  "This generation will not pass away":  either a miss or a hit on the temple alone, and the best rescue costs the hit.  Mark 13:30 says "this generation will not pass away until all these things take place," a few verses after the Son of Man arrives "in the clouds with great power and glory."¹  The phrase "this generation" occurs nineteen times in the first three Gospels, and outside the three disputed verses it always means the people living at that time, and the standard lexicon files these verses there, noting only that "the meaning nation is advocated by some."⁵  So the reading that rescues the saying by making "generation" mean "the Jewish race" fails.  Albert Schweitzer, Bertrand Russell, Bart Ehrman and Dale Allison read the saying as a prediction of the end within a lifetime, and conclude Jesus was wrong.  C. S. Lewis called it "the most embarrassing verse in the Bible."  N. T. Wright and R. T. France answer that "all these things" are the temple's fall and its signs, which did come within forty years, and that the cosmic language is the prophets' way of describing the fall of a city, as Isaiah 13:10 uses it of Babylon.  That reading has real support in Mark, where 13:32 turns to "that day or hour" which no one knows, and it is weakest in Matthew, where the cosmic signs come "immediately after" the siege.  The method's symmetry rule settles what each reading is worth.  If the coming of 13:26 was meant to happen within the generation, the saying failed.  If it was not, the saying is a hit only on the temple's fall, worth the ratio in point 1 and no more.  It cannot be both.  The first Christians felt the problem:  2 Peter 3:4 answers scoffers asking "Where is his promised return?"

3.  Every dated end-time prediction counted here failed.  Preterism holds that most of the discourse was fulfilled in AD 70.  Historicism reads Daniel and Revelation as a map of church history, counting a "day" as a year.  Futurism places most of it in a future that began, for many twentieth-century writers, with Israel in 1948.  Historicists and futurists have produced dated predictions, and the article tabulates every one it could check in the interpreter's own words, from Hippolytus's AD 500 through William Miller's 1843 and 1844, the Watch Tower's 1914, 1925 and 1975, Hal Lindsey's forty years from 1948, Edgar Whisenant's 88 Reasons and Harold Camping's 2011.  None of the fifteen came true.  That count is a verdict on the interpreters and their day-for-a-year and forty-year arithmetic, not on the biblical texts, which set no such dates.  It is also the plainest evidence for Test 8:  Christians have been the most frequent victims of loose prophetic reasoning.

4.  Israel in 1948:  a gathering happened, but it fails Test 4, and most of what the texts promise did not happen with it.  Isaiah 11, Jeremiah 16, Amos 9 and Ezekiel 36–37 promise a return of the scattered people.  In their own setting each ties the return to more:  a Davidic king, the reunion of Israel and Judah, a new heart, and in Deuteronomy 30 a return that follows repentance.  The first return their books point to is the return from Babylon.  Christians did expect a modern return long before Zionism, so the expectation passes Test 3.  But the people who expected it worked to make it happen.  In 1840 Lord Shaftesbury lobbied the Foreign Secretary and called the result "a prelude to the Antitype of the decree of Cyrus."⁶  In 1891 William Blackstone petitioned the President of the United States to "give Palestine back" to the Jews.⁷  Rabbinic tradition read the Song of Songs as an oath forbidding a mass return by human initiative, and orthodox critics in 1908 called Zionism "an attempt to seize the prerogatives of their God."  A return sought deliberately by readers of the prophecy is the textbook case of a fulfilment within human reach.  Calculations that make Ezekiel 4 point to 1948 fail Test 2, because they need several free choices of starting point, multiplier and length of year.

Verdict.  The temple prediction passes the tests that do not depend on Mark's date and is a genuine, modest hit if Mark is early.  The generation saying is either a miss or a hit on the temple alone, and the text does not settle which.  The dated predictions all failed, and they are the interpreters' failures.  Israel's founding is a remarkable event, but as a fulfilled prophecy it fails Test 4.

Part 1: The Olivet Discourse and AD 70

What the texts predict

Jesus and his disciples leave the temple.  One of them points to the stones and buildings.  Jesus answers, "Do you see these great buildings?  Not one stone will be left on another.  All will be torn down!"¹  On the Mount of Olives the disciples ask, "Tell us, when will these things happen?  And what will be the sign that all these things are about to take place?"¹  The rest of Mark 13 is the answer, and Matthew 24 and Luke 21 repeat it with changes.

The parts of the answer that can be checked against history are few.  Mark warns that "when you see the abomination of desolation standing where it should not be" (let the reader understand), "then those in Judea must flee to the mountains."¹  Luke puts a plainer phrase in the same frame:  "when you see Jerusalem surrounded by armies, then know that its desolation has come near."¹  Luke adds that "they will fall by the edge of the sword and be led away as captives among all nations.  Jerusalem will be trampled down by the Gentiles until the times of the Gentiles are fulfilled."¹  An earlier lament in Luke is the most detailed:  "your enemies will build an embankment against you and surround you and close in on you from every side.  They will demolish you, you and your children within your walls, and they will not leave within you one stone on top of another."¹

So the discourse commits to five checkable things:  the temple's buildings thrown down;  a siege with an embankment and a ring around the city;  slaughter, including children;  captives taken to many nations;  and a warning to flee Judea.  Mark and Matthew add the timing:  "this generation will not pass away until all these things take place."¹  That verse is the subject of Part 2.

What happened

Rome's war with Judea began in AD 66.  Titus besieged Jerusalem in the spring of 70.  When his earthworks were burned, he ringed the city with a wall.  Josephus, who watched from the Roman side, gives its length as "forty furlongs, one only abated," about eight kilometres, with thirteen forts, and says "the whole was completed in three days."³  The sanctuary burned in August.  Josephus counts ninety-seven thousand captives taken in the war and gives a figure of eleven hundred thousand dead in the siege.³  Those over seventeen were sent in chains to the mines of Egypt or to die in provincial arenas, and those under seventeen were sold.³  Tacitus, a Roman senator with no sympathy for either Jews or Christians, puts the besieged at six hundred thousand.⁸  Both figures are thought to be inflated, as the editor of Eusebius already noted in 1890.⁹  Then Titus "gave orders that they should now demolish the entire city and temple," sparing three towers and part of the western city wall as a camp for the garrison, and the rest "was so thoroughly laid even with the ground by those that dug it up to the foundation, that there was left nothing to make those that came thither believe it had ever been inhabited."³

The match is good.  Every checkable element of Luke 19:43–44 and 21:20–24 happened.  "Not one stone will be left on another" is hyperbole, as Josephus's own "laid even with the ground" is, with stated exceptions.  The Western Wall of today's photographs is part of the Temple Mount's retaining wall, not one of the "great buildings" Jesus pointed to, and Josephus does not mention it here.  The site's Gospel dating article covers that point in detail.

Test 1: was it written before?

Everything turns on when the Gospels were written.  The site's Gospel dating article argues the question at length.  Its finding, in brief:  outside evidence sets latest possible dates (Luke by about 144, Matthew by 155, Mark by 170) but no earliest ones.  Mark falls somewhere between the 40s and AD 70, Luke–Acts in the early 60s or the 80s, Matthew in the 70s or 80s.  The return-of-Christ half of the discourse was already in circulation by about AD 50, because Paul's first letter to Thessalonica uses its words for the "coming," the trumpet, the thief in the night and the birth pangs.  But Paul has the return, not the ruin.  Nothing in his letters predicts the temple's fall.

The critical case, in its own words.  The strongest argument for writing after the event concerns Luke.  J. M. Creed's 1930 commentary on Luke, still the classic statement, granted that "In itself c. xxi. might be compatible with this early date."  But when Luke is set beside Mark, "it becomes impossible to give a convincing interpretation of the Lucan text except on the assumption that the siege was past history, and that the evangelist has modified his source in the light of an event, well known to all his readers."  He concluded:  "The Gospel, then, was certainly written after A.D. 70, and probably not immediately after."¹⁰  In his notes on chapter 21 he gave the main reason:  "The 'abomination of desolation' standing where it ought not is replaced by 'Jerusalem encircled with armies.'"¹⁰  Creed did not apply the same argument to Mark.  He judged it "probable that the discourse as a whole is prior to the fall of Jerusalem," because Mark's distresses of "those in Judaea" run straight on into the return of the Son of Man.¹⁰

The answer, in its own words.  C. H. Dodd replied in 1947.  He did not argue for an early Luke;  his article opens by saying that "the Third Gospel was in fact produced after the Fall of Jerusalem is on other grounds fairly certain."⁴  He argued that Luke's siege language is no evidence either way.  "If you want to say in Greek 'Jerusalem will be besieged', the choice of available expressions is strictly limited," and the operations Luke describes "are no more than the regular commonplaces of ancient warfare."⁴  Worse for the critical case, the things that "caught the imagination of Josephus, and, we may suppose, of any other witness" are missing:  "the fantastic faction-fighting which continued all through the siege, the horrors of pestilence and famine (including cannibalism), and finally the conflagration in which the Temple and a large part of the city perished."  Dodd:  "Nothing is said of them here."⁴  Every word Luke uses comes from the Greek Old Testament's prophecies and histories of the Babylonian siege.  "So far as any historical event has coloured the picture, it is not Titus's capture of Jerusalem in A.D. 70, but Nebuchadrezzar's capture in 586 B.C.  There is no single trait of the forecast which cannot be documented directly out of the Old Testament."⁴  The Gospel dating article tested this word by word against the whole Septuagint and confirmed it.  Isaiah 29:3 has Luke's two signature words, "encircle" and "rampart," in one verse addressed to Jerusalem.

What the exchange settles.  Dodd removes Luke 21:20 as evidence for a late date.  He does not supply an early one, because a writer in 85 glossing Mark for Gentile readers would reach for the same Septuagint words.  The silence about fire, famine and faction is the strongest point for a prediction written before the event, in all three Gospels.  It is an argument from silence, and a writer in 85 had no duty to itemise.  So Test 1 is open for Luke and for Matthew.  For Mark it is open too, but critics as early as Creed placed the bulk of Mark 13 before 70.

A saying remembered at the trial.  One more line of evidence bears on whether Jesus spoke against the temple at all, apart from any Gospel's date.  Mark reports that at his trial "some stood up and gave this false testimony against him:  'We heard him say, "I will destroy this temple made with hands and in three days build another not made with hands."'"¹  Passers-by mock him with it at the cross.¹  Acts has false witnesses accuse Stephen of saying "that Jesus the Nazarene will destroy this place."¹  John gives a different form of the saying and applies it to Jesus's body.¹  A writer composing after 70 to show that Jesus foresaw the fall would not put the saying in the mouths of false witnesses.  So a saying against the temple goes back before the Gospels.  What it does not show is that the saying predicted a Roman destruction within a generation, which is the form the discourse gives it.

Test 2: what did it commit to?

Scored element by element, the discourse commits to the temple's fall (high specificity:  one building, one city), within a generation (medium:  a window of about forty years), with the siege, slaughter and captivity of any ancient sack (low:  these happen in every sack).  It names no enemy, no year and no commander.  The flight command fits the one flight that is recorded badly.  Eusebius, writing about 260 years later, says the Jerusalem church left for Pella "by a revelation, vouchsafed to approved men there before the war,"⁹ not by obeying the Gospel, and Pella lies across the Jordan, not in "the mountains" of Judea.  Only Epiphanius, about AD 375, says "Christ had told them" to leave.¹¹  The tradition grows toward the Gospel over time, which makes Eusebius's version the older one and leaves the flight unconfirmed as a fulfilment.

The "abomination of desolation" has several candidates and fits none cleanly.  Josephus records an old oracle that the city would fall "when a sedition should invade the Jews, and their own hand should pollute the temple of God," and says the Zealots who occupied the temple in 67–68 "made themselves the instruments of their accomplishment."³  The Romans set up their standards in the temple court and sacrificed to them, but only once the sanctuary was already burning, too late to be a signal to flee.³  A sign that cannot be fixed lowers the Test 2 score.

Test 8: the son of Ananias

Josephus tells of another Jesus.  "There was one Jesus, the son of Ananus, a plebeian and a husbandman, who, four years before the war began, and at a time when the city was in very great peace and prosperity," came to the Feast of Tabernacles and "began on a sudden to cry aloud, 'A voice from the east, a voice from the west, a voice from the four winds, a voice against Jerusalem and the holy house, a voice against the bridegrooms and the brides, and a voice against this whole people!'"  The Roman governor Albinus had him "whipped till his bones were laid bare", and "at every stroke of the whip his answer was, 'Woe, woe to Jerusalem!'"  He "continued this ditty for seven years and five months," until a stone from a Roman engine killed him on the wall during the siege.³

This man is the control Test 8 asks for.  He predicted the city's and the temple's doom, his prediction is reported by a historian who was no follower of his, it was made years before the event, and it came true.  No one treats it as evidence of foreknowledge.  Josephus himself records it among portents, alongside a heifer that gave birth to a lamb in the temple.³  His cry is built from Jeremiah's stock phrase for a doomed city, "the voice of the bridegroom and the voice of the bride."  It is as generic as Mark 13, and from the same prophetic source.

Josephus also shows what failed predictions looked like in the same war.  On the day the temple burned, about six thousand people died in a portico because "a false prophet" had proclaimed "that God commanded them to get upon the temple, and that there they should receive miraculous signs of their deliverance."³  Josephus adds that "there was then a great number of false prophets" urging the people "that they should wait for deliverance from God."  A fair count of first-century Jerusalem prophecy includes those misses alongside the hits.

Price

How unlikely was the temple's fall to a well-informed observer in AD 30?  In the two centuries before, the temple had been profaned by Antiochus IV, entered by Pompey, stripped of its gold by Crassus, stormed by Herod and the Roman general Sosius, and partly burned by the troops of Sabinus.  Five violent episodes in about 196 years, and none razed the sanctuary.  On that record, with the standard rule for an event never yet seen, a forecaster would put the chance of a razing within forty years at about one in seven.¹²  A true foreknower would be right every time, so a correct prediction made in AD 30 is evidence for foreknowledge by a factor of about seven.  That is real evidence, but modest.  It is also the most it can be worth.  If Mark wrote during the war, when revolt was already under way, a forecaster's chance would be one in two or better and the factor two or less.  The son of Ananias, forecasting in AD 62, shows how far that kind of foresight could reach.

Scored:  Test 1 open (rests on Mark's date);  Test 2 medium for the temple and its timing, low for the siege details;  Test 3 does not apply, since the saying is a direct prediction;  Test 4 pass;  Test 5 pass for the event;  Test 6 must count the generation saying (Part 2);  Test 7 about seven to one at most;  Test 8 passed by a non-Christian control, which caps the weight.

Part 2: "This generation will not pass away"

The sayings

Five sayings give a time limit.

• Mark 13:30, after the cosmic signs and the Son of Man "arriving in the clouds with great power and glory": "this generation will not pass away until all these things take place."¹ Matthew 24:34 and Luke 21:32 repeat it.
• Mark 9:1: "there are some standing here who will not experience death before they see the kingdom of God come with power."¹ Matthew 16:28 has "before they see the Son of Man coming in his kingdom."¹
• Matthew 10:23, to the Twelve sent out on a mission: "you will not finish going through all the towns of Israel before the Son of Man comes."¹

Paul, writing about AD 50, counted himself among those who might be alive at the end:  "we who are alive, who are left until the coming of the Lord, will surely not go ahead of those who have fallen asleep."¹  And a later New Testament letter answers people who had noticed the delay.  2 Peter 3 warns of scoffers "saying, 'Where is his promised return?  For ever since our ancestors died, all things have continued as they were from the beginning of creation,'" and replies that "a single day is like a thousand years with the Lord" and that "the Lord is not slow concerning his promise, as some regard slowness, but is being patient toward you."¹  So the problem is not a modern invention.  It was felt within the New Testament itself.

The case that Jesus was wrong, in its own words

Albert Schweitzer made the argument famous in 1906.  He read Matthew 10:23 at face value:  Jesus "tells them in plain words (Matt. x. 23), that He does not expect to see them back in the present age."  And "It is equally clear … that this prediction was not fulfilled.  The disciples returned to Him; and the appearing of the Son of Man had not taken place."¹³  For Schweitzer this was the turning point of Jesus's career:  "the non-fulfilment of Matt. x. 23 is the first postponement of the Parousia."¹³

Bertrand Russell, in his 1927 lecture "Why I Am Not a Christian," made it a charge against Jesus's wisdom.  "He certainly thought that His second coming would occur in clouds of glory before the death of all the people who were living at that time."  He quoted Matthew 10:23 and 16:28, and concluded:  "In that respect, clearly He was not so wise as some other people have been, and He was certainly not superlatively wise."¹⁴

Bart Ehrman, in Jesus:  Apocalyptic Prophet of the New Millennium (1999), built a historical case.  "Jesus appears to have thought that this coming judgment of God through the cosmic Son of Man is imminent.  It is right around the corner.  In fact, it is to happen within his own generation."  He calls this theme "independently attested in all of our earliest sources."¹⁵  Summing up Schweitzer's reading, which he adopts, he writes:  "It turns out that Jesus was wrong."¹⁵

Dale Allison, the most thorough scholar on this side, wrote the bluntest sentence.  "Jesus' generation, however, passed away.  They all tasted death.  And it is not the kingdom of God that has come but the scoffers who ask, Where is the promise of his coming? … Jesus the millenarian prophet, like all millenarian prophets, was wrong:  reality has taken no notice of his imagination."¹⁶  His reason is the Bible's own test.  "If Deut 18:21-22 is anything to go by, biblical prophecies were, as a rule, sufficiently short-term that their truth or falsity could be judged by a prophet's contemporaries."¹⁶

C. S. Lewis, a believer, put the critic's charge in its sharpest form in 1960:  "He said in so many words, 'this generation shall not pass till all these things be done.'  And he was wrong."  Lewis's own comment:  "It is certainly the most embarrassing verse in the Bible."  But he turned it into evidence of the Gospels' honesty, since the same passage has Jesus confess that the Son does not know the day, and no one inventing sayings would have kept either:  "This passage (Mark 13:30-32) and the cry 'Why hast thou forsaken me?' (Mark 15:34) together make up the strongest proof that the New Testament is historically reliable."¹⁷

The case that he was not, in its own words

N. T. Wright argues that the critics misread the genre.  In Daniel 7:13 "one like a son of man" comes "with the clouds" and goes up "to the Ancient of Days."¹  The movement is toward God, not toward earth.  So "the coming of the Son of Man on the clouds of heaven was never conceived as a primitive form of space travel, but as a symbol for a mighty reversal of fortunes within history and at the national level."¹⁸  On Schweitzer:  "This, as we have seen, is a straightforward if devastating misreading of apocalyptic language."¹⁸  Wright's own reading keeps the time limit and changes the event:  "the divine kingdom will be manifest within a generation, when Jesus and his followers are vindicated in and through the destruction of Jerusalem.  The generation that rejects Jesus must be the last before the great cataclysm."¹⁸

R. T. France makes the same argument from the Greek commentary.  The "coming of the Son of Man" in Mark 13:24–27 "is language not about an eschatological descent of Jesus to the earth but, as in the vision of Daniel from which it derives, about the vindication and enthronement of the Son of Man at the right hand of God."  What is described "is a change of government:  the temple and all that it stood for is out, and the Son of Man is in."  He is frank about why this matters.  It removes "the single most embarrassing feature of chapter 13 for traditional Christian interpretation, the quite unequivocal and very emphatic statement in v. 30 that the events just described will take place before this generation has passed.  They did!"¹⁹  France divides the chapter at verse 32:  his outline heads 13:24–31 "The coming of the Son of Man within 'this generation'" and 13:32–37 "An unknown day and hour:  be ready."¹⁹

The prophets' cosmic language supports part of this.  Isaiah's oracle "about Babylon" says "the stars in the sky and their constellations no longer give out their light; the sun is darkened as soon as it rises, and the moon does not shine."¹  Isaiah 34 has the sky "roll up like a scroll" over the judgment of Edom,¹ and Ezekiel 32:7 darkens the sun over Egypt's Pharaoh.  Babylon, Edom and Egypt fell without the stars going out.  So the darkened sun and falling stars of Mark 13:24–25 are, in the prophets' own usage, a way to describe the fall of a power.

What the Greek and the structure support

"Generation" means the people alive at one time.  The phrase "this generation" occurs nineteen times in the first three Gospels.²⁰  In every use outside the disputed verses it means Jesus's contemporaries:  the generation that asks for a sign, that will be judged with Nineveh, on which the blood of the prophets will come.  Matthew 23:36 says "this generation will be held responsible for all these things", a dozen verses before 24:34 and with the same two Greek phrases.¹  The standard lexicon files 24:34, 13:30 and 21:32 under "contemporaries" and notes only that "the meaning nation is advocated by some."⁵  The reading that rescues the saying by making genea mean "the Jewish race," which the Scofield Reference Bible made popular,²¹ fails on usage.  It also fails on sense:  a promise that the Jewish people would still exist when "all these things" happen would be true however long the delay, and so commits to nothing.

"All these things" starts as the temple question.  In Mark the disciples ask only, "when will these things happen?  And what will be the sign that all these things are about to take place?", about the buildings Jesus has just said will be thrown down.¹  The answer's last word on timing echoes the question:  "until all these things take place."  Mark 13:29 says "when you see these things happening, know that he is near", and what one sees before he is near cannot include his arrival.  Then 13:32 turns:  "But as for that day or hour no one knows it."¹  In Mark, France's division is a fair reading of the text.

Matthew is harder.  Matthew's disciples ask a joined question:  "when will these things happen?  And what will be the sign of your coming and of the end of the age?"¹  And Matthew puts the cosmic signs "Immediately after the suffering of those days", with the Son of Man appearing so that "all the tribes of the earth will mourn", and angels gathering the elect "from one end of heaven to the other."¹  France accepts that in Matthew too the time limit is the living generation.  Wright and France must read Matthew's gathering of the elect as the spread of the church after AD 70, and the tribes' mourning as the same vindication.  That is possible, but it is a reading the text has to be argued into.

The other sayings.  Matthew 10:23 is the hardest.  France reads "the towns of Israel" as the Twelve's mission in Galilee and the coming as Daniel's enthronement,²² and he names the alternative he rejects:  "that Jesus mistakenly expected an immediate parousia."²²  Mark 9:1 is followed at once, "six days later" in Mark and Matthew, by the Transfiguration, where three of those "standing here" see Jesus in glory.  Whether the evangelists meant that scene as the saying's fulfilment, the text does not say.  Paul's "we who are alive" shows that the first generation expected to see the end, not that Jesus promised it.

The price of each reading

The method's symmetry rule decides what the saying is worth.  Two readings are live.

• The cosmic reading. Mark 13:30 promised that the visible coming of the Son of Man would happen within Jesus's generation. It did not. On this reading the saying is a clear miss, and it outweighs the temple prediction, because it is the claim the discourse makes with the most solemnity: "Heaven and earth will pass away, but my words will never pass away."
• The historical reading. "All these things" are the temple's fall and the events leading to it, and the Son of Man's "coming" is his vindication. The time limit was kept: the temple fell about forty years later. But on this reading the discourse made no prediction about the end of the world at all, and its "hit" is the temple's fall, which Part 1 priced at about seven to one at most.

A defender cannot take the second reading to clear the saying of failure and then count the first reading's cosmic coming as a further fulfilment still to come.  If the probability that Jesus meant the cosmic coming within the generation is h, the saying as a whole is worth about (1 − h) × 7.4 to the believer's side.  At h = 0.5 it is worth about 3.7;  at h = 0.86 it is worth nothing;  above that it counts against.²³

Where does h lie?  The Greek of genea settles one question, against the "race" reading.  The structure of Mark gives the historical reading a real foothold, and the prophets' cosmic language shows it is not special pleading.  Matthew's "immediately after," the joined question about "your coming," Matthew 10:23, Paul's "we who are alive," and the scoffers of 2 Peter all push the other way, and show that the first readers took the coming to be near.  An honest estimate puts h somewhere near even.  On that estimate the generation saying adds little to the temple prediction and cannot be presented as a fulfilled prophecy of anything beyond it.  It is not a refutation either.  Lewis's point stands:  a church inventing sayings after the fact would not have kept this one.

Part 3: Three systems, and every dated prediction they produced

The three systems

Christian readings of the discourse, Daniel and Revelation fall into three families.  Each starts from a different answer to the question Part 2 asks.

Preterism ("past-ism") holds that most or all of the discourse was fulfilled in the first century.  Partial preterists, such as Kenneth Gentry and Gary DeMar, hold that the Olivet Discourse was fulfilled in AD 70 and that Christ's final, visible return is still to come.  Gentry:  the events of Matthew 24 "are said to be coming upon 'this generation' (v. 34), i.e., the very generation which rejected and 'pierced' Him."²⁴  DeMar argues from usage:  "'this generation' always means the generation to whom Jesus was speaking."²⁵  Full preterists, following Max King, go further and hold that the second coming itself took place in AD 70:  "The coming of Christ is the fall of Jerusalem, or the destruction of the temple."²⁶  Full preterism is rejected by most Christian traditions, because it leaves no future bodily return or resurrection.

Historicism reads Daniel and Revelation as a map of church history from the apostles to the end, counting each prophetic "day" as a year.  Protestant historicists identified Revelation's beast with the papacy and counted the 1,260 days of Revelation 12:6 as 1,260 years.  E. B. Elliott's standard commentary of 1844 fixed on AD 533 "as at least the primary epoch of the commencement of the 1260 predicted years of Papal supremacy."²⁷  William Miller stated the rule plainly in 1838:  "I then proved that days were to be counted years, by the command of God, by the example of Jacob, and by the fulfilment of a part of the vision."²⁸  Applied to the "2,300 evenings and mornings" of Daniel 8:14, from a decree he dated to 457 BC, it gave him his year:  "take 457 from 2300, and it will leave A. D. 1843."²⁸

Futurism, and its best-known form, dispensationalism, places most of the discourse and of Revelation in a future seven-year tribulation before Christ's visible return.  The 1917 Scofield Reference Bible split the discourse into two sieges.  Luke 21:20–24 "refers to the siege by Titus, A.D. 70, when the city was taken, and verse 24 literally fulfilled", while Matthew's and Mark's "abomination" refers "to the final tribulation siege."²¹  Scofield also read the generation saying as the "race" reading Part 2 rejects:  genea's "primary definition" is "race, kind, family, stock, breed," so "the generation — nation, or family of Israel — will be preserved unto 'these things'."²¹  Later futurists kept the future tribulation but dropped Scofield's reading of "generation," and tied it instead to the founding of Israel.  Hal Lindsey's The Late Great Planet Earth (1970) read Israel's rebirth as the fig tree of Matthew 24:32 putting out leaves, and then asked:  "What generation?  Obviously, in context, the generation that would see the signs — chief among them the rebirth of Israel.  A generation in the Bible is something like forty years.  If this is a correct deduction, then within forty years or so of 1948, all these things could take place."²⁹  In 1980 he wrote, "The decade of the 1980's could very well be the last decade of history as we know it."³⁰

The count

Test 6 asks for every prediction from a source, chosen by a rule fixed before scoring.  The rule here:  every end-time prediction with a stated date or a closed window, by a Christian interpreter or body, that this article could read in the interpreter's own words, and whose date has passed.  Newton's private note that the end would come "not before 2060" is a lower bound whose date has not arrived, so it is left out.

Fifteen dated predictions over eighteen centuries, from ten named writers, the anonymous Joachite school and the Watch Tower Society.  None came true.

What the count shows, and what it does not

It does not show that the Bible failed.  None of these dates is in the Bible.  Each comes from an interpreter's arithmetic:  a day for a year, a generation of forty years, 6,000 years of world history, seventy jubilees, 2,520 years from a chosen start.  The texts themselves warn against the exercise.  Mark 13:32 says "as for that day or hour no one knows it."¹  An honest scorer charges these misses to the method, not to the texts.

It does show three things.  First, the methods have a record of zero for fifteen, and any new calculation using them starts with that record.  Grant Jeffrey's 1948 date, examined in Part 4, uses the same 2,520-year device as Russell's 1914.  Second, the record answers the claim that the generation saying must be about 1948:  the writers who held that view set dates, and the dates passed.  Third, it is the plainest evidence for Test 8.  The tests were not invented to embarrass the Bible.  Christians themselves have been the most frequent victims of loose prophetic reasoning, and when a date failed, the movement often reinterpreted it rather than give it up.  The Watch Tower kept 1914 by making the return invisible, and in 1995 it set aside the "calculations of what is the lifetime of a generation since 1914," while saying two years later that this "offered no change at all in our fundamental teaching about 1914."⁴⁴⁴⁵

Preterism escapes the count, at a price.  A preterist sets no future date and so cannot miss one.  But full preterism's claim that the second coming already happened, unseen, in AD 70 cannot fail either, because nothing visible was predicted to happen.  Under Test 2 a claim that no observation could have contradicted is worth nothing as evidence, in either direction.

Part 4: Israel in 1948

What the texts say in their own setting

On 14 May 1948 the State of Israel declared its independence.  Christian writers have called it the fulfilment of several prophecies.  Each should first be read where it stands.

• Isaiah 11:11–12: "At that time the Lord will again lift his hand to reclaim the remnant of his people from Assyria, Egypt, Pathros, Cush, Elam, Shinar, Hamath, and the seacoasts. He will lift a signal flag for the nations; he will gather Israel's dispersed people and assemble Judah's scattered people from the four corners of the earth."¹ "At that time" points back to 11:1, "A shoot will grow out of Jesse's root stock," the coming Davidic king. The chapter ends with the Lord drying up the Egyptian Sea and the Euphrates and making "a highway leading out of Assyria for the remnant of his people, just as there was for Israel, when they went up from the land of Egypt."¹ "Again" counts the first exodus as the first gathering.
• Isaiah 66:7–8: "Before she goes into labor, she gives birth! Before her contractions begin, she delivers a boy! Who has ever heard of such a thing? Who has ever seen this? Can a country be brought forth in one day? Can a nation be born in a single moment?"¹ The image is a birth without labour pains.
• Ezekiel 36:24–28: "I will take you from the nations and gather you from all the countries; then I will bring you to your land. I will sprinkle you with pure water, and you will be clean from all your impurities. … I will give you a new heart, and I will put a new spirit within you."¹ The gathering comes first and the cleansing after it, which is why dispensational writers expect a return in unbelief.
• Ezekiel 37, the valley of dry bones, which "are all the house of Israel". The chapter explains the vision: "I will make them one nation in the land, on the mountains of Israel, and one king will rule over them all. They will never again be two nations," and "My servant David will be king over them."¹
• Amos 9:14–15: "I will bring back my people, Israel; they will rebuild the cities lying in rubble and settle down. … I will plant them on their land, and they will never again be uprooted from the land I have given them."¹
• Jeremiah 16:14–15: people will swear by "the Lord lives who delivered the people of Israel from the land of the north and from all the other lands where he had banished them. At that time I will bring them back to the land I gave their ancestors."¹ Jeremiah's own book names the first return it means: "Only when the seventy years of Babylonian rule are over will I again take up consideration for you."¹
• Deuteronomy 30:1–5 sets the condition the others assume: "if you and your descendants turn to the Lord your God and obey him … the Lord your God will reverse your captivity and have pity on you. He will turn and gather you from all the peoples among whom he has scattered you."¹

Read in their own setting, these texts promise a return of the scattered people joined to other things:  a Davidic king, the reunion of the northern and southern kingdoms, a new heart and obedience, and in Deuteronomy a return that follows repentance.  The first return their own books point to is the return from Babylon under Cyrus.  The founding of Israel in 1948 matches the gathering.  It does not match the king, the reunion of two kingdoms that no longer exist as separate peoples, or a national turning to God before the return.  A modern reader can assign those to a later stage, and dispensational writers do.  But then 1948 is a partial fit, and the parts that fit are the parts any return would have.

Two of the texts also conflict on the order of events.  Deuteronomy 30 puts repentance before the gathering;  Ezekiel 36 puts the gathering before the cleansing.  A corpus that allows either order can be fitted to either kind of return, which lowers its Test 2 score.  Amos 9:15 adds a further problem.  If it described the return from Babylon, it failed, because the people were uprooted again in AD 70 and 135.  Assigning it to 1948 because the first return was followed by another exile is a reading made after the event, which is what Test 3 rules out as evidence.

Isaiah 66:8 is the verse most often quoted, because the state was declared in a single day.  Scored on its own words, it fits poorly.  The image is a birth that comes before labour, and 1948 came after fifty years of organised settlement, and the declaration was followed the next day by war.  Nor is "in one day" specific:  almost every modern state came into being by a declaration or treaty on a stated date.

Was a modern return expected before it happened?

Yes, and that is the strongest point for the claim.  Christian readers expected a return of the Jews to their land long before Zionism, which passes Test 3 for the general expectation.

In 1840 Lord Shaftesbury, the evangelical reformer, "had long cherished the belief, founded upon an earnest and diligent study of the prophecies contained in Holy Scripture, that the Jews were to return to their inheritance in the Holy Land, and it appeared to him that the time was ripe for the accomplishment of those prophecies," as his biographer put it.⁶  His diary for July of that year reads:  "Everything seems ripe for their return to Palestine."⁶  Earlier still, the Boston minister Increase Mather argued in 1669 that the true meaning of Daniel 12:2 might be "That after the Jews are brought into their own Land again, and there disturbed with Gog and Magog" at Armageddon.⁴⁶  The 1917 Scofield Reference Bible, the most widely read dispensational study Bible, set out in its note on Deuteronomy 30 a "Restoration to the land."²¹  Rabbinic tradition also expected a final ingathering, but in God's time, as the next paragraph shows.

But the expectation was not one thing.  The Babylonian Talmud records a ruling that "Anyone who ascends from Babylonia to Eretz Yisrael transgresses a positive mitzva," and reads the three oaths of the Song of Songs as binding Israel "that the Jews should not ascend to Eretz Yisrael as a wall," "that they should not rebel against the rule of the nations of the world," and the nations "that they should not subjugate the Jews excessively."⁴⁷  On that reading the prophecies promised a return in God's time and forbade a mass return by human initiative.  In the twentieth century the Satmar rebbe, Joel Teitelbaum, made the oaths absolute:  "These oaths may not be violated even on pain of death."⁴⁸  Orthodox opponents of Zionism said so in its first decade.  William Blackstone, a supporter, reported in 1908:  "Many of the orthodox criticise it as an attempt to seize the prerogatives of their God.  While others say that God will not work miracles to accomplish that which they can do themselves."²

Even the Christian Zionists did not read Zionism as the promised restoration.  Blackstone wrote that the movement "marked a wonderful innovation in the attitude of the Jews and a closer gathering of the dry bones of Ezekiel," and in the next breath:  "this godless national gathering of Israel is not the fulfilment of the glorious divine restoration, so glowingly described by the prophets."²  He also offered a date:  "It is significant that this first Zionist congress assembled just 1,260 years after the capture of Jerusalem by the Mohammedans in A. D. 637."²  That is one more day-for-a-year reckoning, landing on 1897, for the count in Part 3.

Test 4: a return within human reach

The founding of Israel fails Test 4 more plainly than any case in the method article.  The return was sought, planned, financed and fought for by people who had read the prophecies, and Christian readers of the same prophecies lobbied governments to bring it about.

Shaftesbury dined with the Foreign Secretary, Lord Palmerston, on 1 August 1840 and "Propounded my scheme, which seemed to strike his fancy."  He noted that he had to argue on Palmerston's terms:  "I am forced to argue politically, financially, commercially."  On 24 August Palmerston told him that he had instructed the ambassador at Constantinople to seek protection for Jewish settlers.  Shaftesbury wrote:  "This is a prelude to the Antitype of the decree of Cyrus."⁶  He saw himself, in other words, as arranging the modern counterpart of the decree that ended the Babylonian exile.

In March 1891 Blackstone presented a petition to President Benjamin Harrison.  It asked, "Why not give Palestine back to them again?  According to God's distribution of nations it is their home, an inalienable possession from which they were expelled by force,"⁷ and it asked the President to secure "an international conference to consider the condition of the Israelites and their claims to Palestine as their ancient home."⁷  Its signatories included the Chief Justice, J. Pierpont Morgan and John D. Rockefeller.

The British government that issued the Balfour Declaration in November 1917 acted on mixed motives, and its Prime Minister named both.  David Lloyd George wrote that Chaim Weizmann "appealed to my deep reverence for the great men of his race who were the authors of the sublime literature upon which I was brought up."  He also wrote that "the actual time of the declaration was determined by considerations of war policy.  It was part of our propagandist strategy for mobilizing every opinion and force throughout the world which would weaken the enemy and improve the Allied chances."⁴⁹  A biblical upbringing was one motive among strategic ones.  Either way, the Declaration was a decision taken by men who knew the texts.

Religious Jews argued the same way before Zionism existed.  In 1862 Rabbi Zvi Hirsch Kalischer wrote that "THE REDEMPTION OF ISRAEL, for which we long, is not to be imagined as a sudden miracle," and:  "On the contrary, the Redemption will begin by awakening support among the philanthropists and by gaining the consent of the nations to the gathering of some of the scattered of Israel into the Holy Land."⁵⁰  Kalischer still expected the full redemption to be a miracle.  But the beginning, he said, was for people to make.

The method article draws its line at exactly this point.  Theudas and the Egyptian prophet did what was in their power, gathering crowds at the right place;  what was not in their power, the parting of the Jordan and the fall of Jerusalem's walls, did not happen.  A national return, by contrast, is the kind of event that determined people with money, votes and arms can bring about.  It was brought about by people who believed it was predicted, among others, and opposed by others who read the same texts as forbidding it.  That is the textbook case of a fulfilment within reach of those who knew the prophecy.  It does not show the return was not providential.  It shows it cannot be used as evidence of foreknowledge to someone who does not already believe it.

What a skeptic should grant is narrower and real.  A people dispersed for eighteen centuries kept its identity, its language of prayer and its attachment to one land, and returned to it.  That is historically unusual.  Whether it is improbable enough to need explaining is a separate question about Jewish history, owned by Are Today's Jews Descended from Ancient Israel? and Judaism, not a test of a prediction.

The calculation from Ezekiel 4

Grant Jeffrey's Armageddon:  Appointment with Destiny (1988) argued that Ezekiel predicted the year.  He wrote that Israel was "reborn as a nation — precisely to the year as foretold by the prophet Ezekiel."⁴²  His steps, in his own words:

1. Ezekiel lies on his side 390 days for Israel and 40 for Judah, "one day for each year."¹ Jeffrey: "Out of a total decreed punishment of 430 years for Israel's and Judah's sin (390 years plus 40 years = 430 years); once you deduct the 70 years of the Babylonian captivity which ended in 536 B.C.; there is a total of 360 years of further punishment beyond 536 B.C."⁴²
2. Leviticus 26:18 warns, "I will discipline you seven times more on account of your sins."¹ Jeffrey reads this as a multiplier of time: "the end of the punishment and restoration to the land would be accomplished in 2,520 biblical years of 360 days each."⁴²
3. Converting: "2,520 biblical years x 360 = 907,000 days. Converting this figure into our calendar year of 365.25 days and dividing 365.25 into 907,200 days we reach a total of 2,483.8 calendar years." Counted from "the Spring of 536 B.C.," this is supposed to land on 14 May 1948.⁴²

The calculation fails Test 2 because nearly every step is a free choice.  The subtraction of seventy years is not in Ezekiel.  "Seven times more" in Leviticus 26 reads naturally as sevenfold severity, and the chapter applies it four times, so the multiplier could as well be seven to the fourth power.  The 360-day year is imported from a separate reckoning of Daniel.  The start in 536 BC is one of several dates for the end of the exile.  And the arithmetic does not reach its target.  907,200 days from 21 March 536 BC ends on 18 January 1949, about eight months after 14 May 1948.⁵¹  The same 2,520-year device served Charles Taze Russell in 1889, counted from a different start:  "the 'Times of the Gentiles' is a period of 2520 years, from the year B. C. 606 to and including A. D. 1914."³⁶  A method that delivers 1914 from one starting point and 1948 from another has not predicted either.

Luke 21:24's "Jerusalem will be trampled down by the Gentiles until the times of the Gentiles are fulfilled" is also claimed for 1948.¹  It does not fit that year, since the Old City of Jerusalem remained under Jordanian rule until 1967.  Anyone who applies the verse to modern Israel must move its fulfilment to the Six-Day War, and a verse that can be moved from one date to another when the first does not fit has not committed to either.

Scored:  Test 1 pass;  Test 2 low (the gathering fits;  the king, the reunion, the new heart, the painless birth and the date do not, and the calculations need free choices);  Test 3 pass for a general Christian expectation of a return, mixed for Jewish readers;  Test 4 fail;  Test 5 pass for the event;  Test 6 must count Blackstone's 1897 and the moved dates for Luke 21:24;  Test 7 not priceable while Test 4 fails;  Test 8:  a return of any exiled people, planned by its own leaders, would not be counted as fulfilled prophecy, and this one should not be either.

Objections and responses

"Mark 13 was written after AD 70, so the temple prediction was written after the event."  That may be true, and this article does not claim otherwise.  Test 1 is open, and the site's Gospel dating article places Mark anywhere from the 40s to 70.  Three points keep the question open rather than settled against the prediction.  All three Gospels omit the fire, famine and faction that dominate every eyewitness account.  J. M. Creed, the classic critical voice for a late Luke, judged Mark's discourse "probable" to be "prior to the fall of Jerusalem."¹⁰  And a saying against the temple was remembered at Jesus's trial as false testimony, which no writer after 70 would invent.

"Anyone could have predicted it.  Jesus son of Ananias did."  Agreed, and the article says so.  The prediction is priced as a good forecast, worth at most about seven to one if Mark is early and two to one or less if Mark wrote during the war.  It is not presented as a miracle.

"Not one stone was left on another?  The Western Wall is still standing."  Jesus pointed at "these great buildings," the temple's structures.  The Western Wall is part of the platform's retaining wall.  Josephus uses the same kind of hyperbole, saying the city was laid "even with the ground," while naming the towers and the stretch of western city wall that were spared.³

"Genea means race.  Jesus promised the Jewish people would survive until his return."  The usage is against it:  every other use of "this generation" in the first three Gospels means the people then alive.  The standard lexicon lists the three disputed verses under "contemporaries."⁵  And a promise that a people will still exist at some unstated future date commits to nothing that could fail, so it would not count as a fulfilled prophecy even if it were the meaning.

"The cosmic signs are literal.  By Deuteronomy 18:22 Jesus is a false prophet."  If the cosmic reading is right, the saying failed.  This article does not deny that.  It argues that the reading is not certain:  the prophets used the same darkened sun and falling stars for the fall of Babylon, Edom and Egypt, and in Mark the time limit can attach to the temple and its signs.  It also notes Lewis's point that a church inventing sayings would not have kept a verse that looked like a failure, next to a confession that the Son did not know the day.

"Preterism is a rescue invented after the delay, so it fails Test 3."  Partly true, and the article prices it that way.  The preterist reading is allowed to clear the saying of failure only by giving up any claim that the discourse predicted the end of the world.  That is the symmetry rule, and it leaves the saying worth what the temple's fall is worth, no more.

"The failed date-setters prove nothing about the Bible."  Agreed.  The article charges the misses to the interpreters' arithmetic.  The point of the count is that the methods have failed every time they were tried, so any new calculation from them, including the ones aimed at 1948, deserves no confidence.

"Isaiah 66:8 fits exactly:  Israel was born in one day, 14 May 1948."  The verse's image is a birth before labour, and the state came after fifty years of settlement and was followed at once by war.  "In one day" fits any state founded by a dated declaration, which is almost every modern state.  That is a low Test 2 score.

"God can use human means.  That Christians and Zionists worked for the return does not stop it from being a fulfilment."  True, and the article does not deny providence.  Test 4 asks a narrower question:  whether the event is evidence of foreknowledge to someone who does not already believe.  If any outcome that believers worked to bring about can count, then Theudas's crowd at the Jordan would count too.  The test has to exclude what readers of the prophecy could do themselves, or it excludes nothing.

"Jeffrey's Ezekiel calculation lands precisely on 14 May 1948."  It does not.  His own figures, 907,200 days from the spring of 536 BC, end in January 1949, about eight months later.  The calculation also needs a subtraction Ezekiel does not make, a multiplier read from a verse about severity, and a 360-day year taken from another book.

"The article is too generous.  An after-the-event text deserves no credit at all."  Every number in Part 1 is conditional on Mark being early, and the scorecard says so.  If Mark wrote after 70, the temple prediction is worth nothing, and the generation saying becomes the evangelist's own expectation, which the cosmic reading still has to answer.

Issues and unanswered questions

1. Mark's date decides Part 1. This article takes the range from the site's Gospel dating article and does not re-argue it. The masculine participle in Mark 13:14 and the Caligula crisis of AD 40, which some read as the "abomination," are not covered there or here.
2. The value of h is a judgement. No datum fixes how likely it is that Jesus meant a visible coming within his generation. The article puts it near even and shows what each value implies. A reader who sets it higher should conclude the saying counts against.
3. The base rate for the temple's fall is crude. Five episodes in two centuries is a small sample, and Laplace's rule is a convention. A forecaster who weighed Rome's destruction of Carthage and Corinth, or the unrest of the 30s, could reasonably set a higher prior and a lower ratio.
4. The Pella tradition was not weighed in full. Gerd Lüdemann's 1980 case against it and Craig Koester's 1989 defence were not read. The article uses only what Eusebius and Epiphanius say.
5. Gaps in the count. Chuck Smith's 1981 date (Future Survival, 1978) could not be obtained. The Watch Tower's own admissions of the 1925 and 1975 failures were not located. Dale Allison's later statements (The Historical Christ and the Theological Jesus, 2009) were not obtained. Adding any of these would not change the result, since the count has no hits.
6. Balfour's own account of his motives and the Satmar rebbe's Vayoel Moshe (1961) in its own Hebrew are not quoted here; Teitelbaum is quoted through Aviezer Ravitzky's translation. Lloyd George's memoir gives both a religious and a strategic motive; how much each weighed is a historians' question this article does not settle.
7. Isaiah 53 and Daniel 9 are tested in their own articles. This article does not use Daniel 9's endpoint, which some read as the temple's destruction in AD 70.

Verdict

The fall of Jerusalem:  passes Tests 4 and 5;  Test 1 is open and rests on Mark's date;  low specificity outside the temple and the time limit.  If Mark is early, it is a genuine hit worth about seven to one, the strength of a very good forecast.  The son of Ananias shows that forecast was within human reach.

"This generation":  undecidable between two readings, and the symmetry rule fixes what each is worth.  On the cosmic reading the saying failed.  On the historical reading it was fulfilled, but only in the temple's fall.  The "race" reading of genea fails.  The saying cannot be counted as a fulfilled prophecy of the end of the world on any reading.

The dated predictions:  fifteen dates over eighteen centuries, none fulfilled.  These are the interpreters' failures, and they are the clearest Test 8 lesson in the Prophecy section.

Israel in 1948:  passes Tests 1, 3 and 5;  fails Test 4, because the return was sought and lobbied for by readers of the prophecies;  low on Test 2, because it fits the gathering and little else the texts promise.  It is a remarkable chapter of Jewish history.  It is not evidence of foreknowledge.

Sources