The Jewish era, the gap at the Persians, and the account that lies beneath it
21 min read
In-depth:
1 · A date with a package insert
A marriage contract is being drawn up. The ketubah lies on the table, the witnesses bend over it, and at the head of the sheet stands the date: in the year such-and-such since the creation of the world. And then, in every such document, for centuries now, there follows a small and curious little phrase. Lifrat she'anu monim. According to the reckoning that we reckon.
Pause there a moment. It is the most careful legal tradition in the world, a people that guards its calendar with a precision astronomers envy — and above every deed, every marriage letter, every contract of sale, they place a reservation. Not: the year since creation. But: the year since creation, according to our way of counting.
Rabbi Shimon Schwab, who wrestled with this question all his life, points it out himself. "This is perhaps the reason", he writes, "that in our documents we always add: minyan she'anu monim — according to the reckoning that we reckon." It is one of the most moving sentences in rabbinic literature. The tradition knows it. It has never concealed it. It only whispers it, in the margin of every contract.
This chapter is about what stands in that margin — and about the fact that the account which lies beneath it is this book's own account too.
2 · Where it goes wrong: the gap at the Persians
The Jewish era counts the years since creation. In the year 2026 the calendar reads 5786. That number does not fall out of the sky; it has been added up — generation by generation, king after king — and that sum stands, since the second century after Christ, in one book: the Seder Olam Rabbah, ascribed to Rabbi Jose ben Halafta.
Up to the exile that count runs reasonably parallel with what historians find. And then comes the Persian period, and something strange happens.
Persian rule over Israel lasted, by the sober reckoning of cuneiform, king-lists and Greek historians, about two hundred and seven years — from Cyrus' conquest of Babylon to Alexander. The Seder Olam gives it fifty-two. Not two hundred and seven. Fifty-two.
There sits the gap. One hundred and sixty-four years, fallen out between two pages.
Workbenchthe two reckonings side by side
Event
Seder Olam
Historical
Difference
Destruction of the first temple
422/421 BC (AM 3338)
587/586 BC
±164
Dedication of the second temple
352/351 BC
516 BC
±165
Length of the Persian period
52 years
±207 years
±155
Length of the second temple
420 years
±586 years
±166
Destruction of the second temple
AD 70 (AM 3828; the conventional calendar calls this year 3830)
AD 70
0, with a seam of 2 years between Seder Olam and calendar
Note the last line, for there sits the sting. At the end the count closes perfectly. That is because the era from creation has been calculated backwards from the year 70. The error therefore does not sit at the surface, where everyone would stumble over it — it lies buried in the middle, at the Persians. Whoever looks only at the present sees nothing. [DOCUMENTED]
The reckoning of the sages themselves, as Schwab cites it (b. Avodah Zarah 9a, 10b, following the Seder Olam): the 420 years of the second temple = 34 years Persian (of a Persian period that "lasted in total only 52 years") + 180 Greek + 103 Hasmonean + 103 Herodian. [DOCUMENTED — primary]
3 · Four kings do the work of ten
How do you press two hundred years into fifty-two? By merging kings.
The Persian king-list that history knows is long: Cyrus, Cambyses, Darius I, Xerxes, Artaxerxes I, Darius II, and a handful more. The Seder Olam knows four. And in the rabbinic exposition names coincide: Koresh, Daryavesh and Artachshasta are not three men but three names for the same man — the first a personal name, the other two honorific titles.
It is, it must be said, a solution of great elegance. It explains at a single stroke why Ezra and Nehemiah seem to be working under three different royal names. It turns a messy epoch into a tight story. And it is almost entirely wrong.
Archivehow one arrived at four kings
The classic equation — Koresh (Cyrus) = Daryavesh (Darius) = Artachshasta (Artaxerxes) as one monarch with three names — in effect makes the decrees of Ezra 1, Ezra 6 and Ezra 7 the decisions of the same man, and has Ezra arriving in Jerusalem only one year after the building of the temple.
And the tradition knew this grated. Schwab cites Biberfeld, who notes that these consequences "were rejected by Rabbi Zerachiah of Lunel as irreconcilable with the word of the Bible". The tension was therefore not invented by Christians; it had already been noticed within the Jewish tradition itself.
The hard side. Against the four kings of the Seder Olam stand dated Babylonian clay tablets from virtually every regnal year of every Persian king, the astronomically anchored Canon of Ptolemy, the Elephantine papyri, and Greek historiography. This is not one source against one source. It is an archive against a scheme. [DOCUMENTED]
4 · The number that explains everything — and accuses no one
Here comes the find that carries the whole case, and it is of a sobering simplicity. No conspiracy is needed. No deceit is needed. All that is needed is a subtraction, and it stands out in the open in the Seder Olam itself.
Take the year in which, according to the Seder Olam, the first temple falls: 3338 since creation. Take the year in which the second temple falls: 3828. Subtract.
Expert490
3828 − 3338 = 490.
Four hundred and ninety years. Precisely the seventy weeks of Daniel 9:24 — "seventy weeks are determined upon thy people and upon thy holy city".
That is no coincidence and no deceit. It is an exposition. The Seder Olam lets Daniel's seventy weeks run from the destruction of the first temple to the destruction of the second. And in order to make those four hundred and ninety years fit between two fixed end points, room had to go in the middle — and that room was taken away at the Persians, the epoch about which Scripture tells us least.
The calendar is therefore not a chronicle but an exegesis. It was not written in the first place to count the years, but to expound Daniel. And that says at once what the consequence is: on this reckoning Daniel's clock can no longer arrive at a Messiah in the midst of the week. It arrives at the year 70. [DOCUMENTED · INTERPRETED]
Let that sink in, for it is the heart of this whole dossier. The question "when do Daniel's seventy weeks end?" and the question "how old is the world?" have become, in the Jewish tradition, the same question. Whoever fixes the one answer fixes the other with it. The era and the messianic expectation are riveted to each other.
And that determines the tone of this chapter. Whoever cries "forgery" here has not understood. There is no bookkeeper who quietly struck years out. There is an expositor who read Daniel as he believed he had to read him — and that exposition has, centuries later, shaped the calendar of a whole people. That is not a crime. That is a tragedy, and a tragedy calls for compassion, not triumph.
5 · What the tradition itself says about it
The most remarkable thing about this matter is that the sharpest criticism comes not from outside but from within.
In the sixteenth century Azariah dei Rossi in his Me'or Einayim sets the rabbinic chronology beside the Greek and the Roman, and concludes that the traditional reckoning for the time before the second temple is not historically accurate: where the length of a period was unknown, the Seder Olam simply assumed the shortest possible one. His book was banned by several authorities. The question had been put, and it was not answered but buried.
And then, in 1962, a rabbi from Washington Heights takes up the thread again.
The thesis. Schwab dares to consider that the sages concealed the years deliberately: "It seems possible that our Sages, for some unknown reason, 'covered up' a certain historic period and purposely eliminated and suppressed [...]" His ground is Daniel himself — the command to "shut up the words, and seal the book" (Dan. 12:4). That gives, he writes, "a perfect right to assume that certain historical events revealed to Daniel were omitted by him on purpose in faithful obedience" to that heavenly command.
The scale he himself names: "to leave out completely from our annals a period of at least 165 years and to correct all data and historic tables in such a fashion that the subsequent chronological gap could escape notice for countless generations."
And then — and this is why he is cited here with reverence — he turns against his own thesis. Three objections, from himself. First: "The events described in a history book have either happened or they haven't [...] A chronological time table is the backbone of any book on history which expects to be accepted by intellectually honest students." Second, and it is almost audible how it pains him: if our forefathers wanted to hide a period at divine command, "who has the right to uncover the carefully concealed secret? [...] we still were not given the right to remove the veil for any reason." He compares himself with the king of the Khazars, who was told in a dream: "Your intentions are acceptable, but your deeds are not."
The honest outcome. Towards the end of his life Schwab withdrew his position — among other reasons because a wilful concealment would have irreparably damaged the halachic calculations themselves (the molad). Whoever adduces Schwab as a witness for a deliberate falsification uses him against his own last word. This book does not do so. [DOCUMENTED — essay consulted in full; the retraction secondary]
See what happens here. An orthodox rabbi sees the gap, names it, dares to consider the gravest explanation — and then withdraws, not because he is afraid of the facts, but because he holds that he may not lift the veil. That is not cowardice. That is a man standing between two kinds of loyalty and unable to resolve it. No mockery fits there. A silence fits there.
6 · The ledger — where the missing years come from
Now the question for which this chapter was written. How many years are missing? And the answer is not one number but an account with four entries — and that is better than one number, for every entry can be tested separately.
This chronology puts creation at about 3967 BC; the rabbinic reckoning at 3761. The difference is two hundred and six years. This is how it is built up:
Workbenchthe entries of the ledger
Segment
This book
Seder Olam
Difference
Creation → Flood
1656
1656
0
Flood → birth of Abraham
352
292
+60
Abraham → exodus
512
500
+12
Exodus → founding of the temple
480
480
0
Founding of the temple → destruction
380
410
−30
Destruction → present
—
165 short
+165
TOTAL
+207
The ledger measures against the Seder Olam, which calls the year 70 AM 3828; the conventional calendar calls it 3830, a seam of 2 years within the rabbinic count. 207 − 2 = 205: the difference in year-numbers (5794 = 5999). Plus one, because the clock begins at year 0 and the calendar at year 1, gives 206: the distance between the starting dates. The ledger closes, but it follows the anchor (the temple in Ziv 966) and is therefore no independent test. [CALCULATED — reproducible from two tables]
The +60 is the Terah key, and it is the best-anchored entry of the four. Genesis 11:26 reads as if Terah was seventy when he begat Abram. But Terah died at two hundred and five (Gen. 11:32), Abram departed from Haran at seventy-five (Gen. 12:4), and that departure was after his father's death (Acts 7:4). Two hundred and five minus seventy-five is one hundred and thirty. The seventy was the beginning of Terah's begetting of children; Abram was the youngest. Sixty years, and they rest on three texts, among them an explicit New Testament exposition.
The −30 is the period of the kings, and it runs the other way — and precisely for that reason it is worth mentioning. The Seder Olam gives the first-temple period four hundred and ten years; the historical length is 380. Here the rabbinic tradition therefore counts too many, not too few, because a naive addition of the regnal years of Judah and Israel runs too long as long as you do not discount the co-regencies. That is exactly what Thiele resolved.
Say that out loud, for it is the test of the honesty of this whole chapter: whoever cries that "the Jews are missing years" must add that in the period of the kings they have a few too many. A set of books with nothing but windfalls is suspect. This one has a shortfall, and that speaks in its favour.
7 · The seven years — the assumption that has to be carried
There remains the third entry: the twelve years between Abraham's birth and the exodus. Five of them are simply explained (this book reckons two hundred and fifteen years in Egypt where the rabbis reckon two hundred and ten). And the remaining seven call for a decision that may not be concealed.
The problem can be stated sharply. The chain up to Abraham's calling is fixed: 1656 to the Flood, plus 352 to Abraham's birth, plus 75 to his calling — together AM 2083. Add the four-hundred-and-thirty-year sojourn to that, and you land on AM 2513. But the exodus stands at AM 2520.
And here a temptation must first be removed, for it is the first one everyone tries. You cannot find those seven years somewhere within the sojourn — not with Joseph, not with the seven years of famine, not with the slavery. Why not? Because every division of the four hundred and thirty years lands on the same result. Whether you split it as 215 + 215, or 208 + 222, or 400 + 30 — the sum remains four hundred and thirty, and the exodus remains 2513. The famine moreover straddles the descent: Jacob goes down in the second year of famine (Gen. 45:6). It is not a block that you can shove in front of it or behind it.
The gap therefore lies not in the sojourn, but at its starting point. And there Scripture itself shows the way.
For Paul counts the four hundred and thirty years not from Abraham's calling, but from the covenant: "the covenant, that was confirmed before of God, the law, which was four hundred and thirty years after, cannot disannul" (Gal. 3:17). And the covenant is an identifiable event — Genesis 15, the ceremony of the divided animals. It is moreover precisely there that the clock is pronounced: "know of a surety that thy seed shall be a stranger [...] four hundred years" (Gen. 15:13). The counting down begins where it is announced.
If Genesis 15 falls in the year in which Abram was 82 (AM 2090), the chain closes exactly: 2090 plus 430 is 2520. But the text does not name that year. The window runs from his seventy-fifth (the calling) to his eighty-fifth (Genesis 16, where he has dwelt ten years in Canaan). Why then eighty-two of all years?
The surprising answer: because the midrash itself hands it over — from the other side.
ArchiveBereshit Rabbah dates the feast itself
At the casting out of Hagar and Ishmael it stands that Abraham sent them away with the child "on her shoulder". And the midrash balks at that, for Ishmael was no infant at the time. Bereshit Rabbah 53:13, verbatim (Sefaria):
"'On her shoulder and the child' — he was twenty-seven years old — and yet you say: 'On her shoulder'?"
The edition annotates it itself: "The Midrash understands that the feast made in honor of Isaac was for his Bar Mitzva, and Yishmael was fourteen years older than Isaac." And that is right: Ishmael was born when Abram was eighty-six (Gen. 16:16), Isaac when he was a hundred (Gen. 21:5) — fourteen years' difference. Ishmael twenty-seven therefore means Isaac thirteen.
The midrash therefore dates the feast of weaning and the casting out to Isaac's thirteenth year — AM 2120. And that is precisely the point at which the four hundred years of Genesis 15:13 must begin in order to close on the exodus of AM 2520. It does not stand there as chronology; it stands there in passing, to explain a detail about a shoulder. But it stands there. [DOCUMENTED — verified in the primary source]
Now turn the derivation round, and see what happens. Take the midrash's dating (the feast in Isaac's thirteenth year, AM 2120). Take the fact that the four hundred of Genesis 15:13 and the four hundred and thirty of Galatians 3:17 must close on the same exodus — that is a difference of thirty years. Then the covenant follows at 2120 minus 30 = AM 2090, and then Abram is 82. The eighty-two is then no longer a chosen number, but a calculated result.
And note the symmetry with the opposing party, for it is perfect. The Seder Olam builds exactly the same bridge of thirty years — only there it runs from the covenant (when Abram was 70) to Isaac's birth, and here from the covenant (when Abram was 82) to Isaac's persecution. Both systems hang the pair 400/430 on a single thirty-year bridge. The only difference is which event closes that bridge. And that the persecution is the starting point is said by Genesis 15:13 itself — it concerns sojourning with affliction — and Paul reads the weaning feast precisely so: "he that was born after the flesh persecuted him that was born after the Spirit" (Gal. 4:29).
Counterpointwhat is not claimed here
Nobody dates Genesis 15 to Abram's eighty-second. Not the Seder Olam (which chooses seventy), not Jubilees (which arrives at about eighty-five), not Josephus, not Augustine (who expressly leaves the dating open). The eighty-two is a reconstruction. What is documented is the window in which it falls — and that Jubilees reads that window in its later half, after the calling, precisely as the order of the narrative requires.
And no ancient source says that the four hundred years begin at the feast. That remains the interpreted link — now with three threads beneath it instead of none: Genesis 15:13 itself defines the four hundred as the sojourning with affliction of the seed; Galatians 4:29 reads the feast as the first persecution of that seed; and the midrash dates that feast to Isaac's thirteenth year.
There is moreover a second midrashic tradition. Pirkei DeRabbi Eliezer 30 knows, besides the twenty-seven, also a manuscript reading of seventeen, and Ussher chose within that computational tradition (weaning at about five years). This book chooses the explicit midrashic chronology over the computational tradition. That is a choice — with a primary source, and the mirror image of the choice Ussher made. It must remain labelled as a choice.
And an honest ±1. The chain closes exactly on the inclusive count ("in his twenty-seventh year"). On the completed count it lands one year later. That difference falls within the noise that the chronology itself already concedes — but it is named here, not hidden away. [INTERPRETED — doubly labelled]
8 · What the sabbatical cycle cannot say — and what the clock at most mirrors
There lies a tempting thought here: let the sabbatical years decide. If creation falls in 3967 or in 3960, surely one of the two must clash with the shemitah cycle?
No. And it is better that this book says so itself than that a critic should find it.
Expertwhy the shemitah can in principle measure nothing here
The two candidates differ by precisely seven years — one full year-week. And every seven-year cycle is invariant under a shift of seven: 3967 and 3960 give exactly the same shemitah positions. Mathematically undecidable.
And the other anchoring does not help either: if one counts the sabbatical years from the entry into the land (Lev. 25), then the BC dates are identical on both chains — those hang on the temple of 966 plus 480, not on the era from creation. Chain-neutral by construction.
That is a neutral result, not a loss: nothing in the sabbatical evidence contradicts this chronology. It simply cannot distinguish it from its competitor. Whoever hoped for a measurement here must go without. [CALCULATED]
A cycle that does set the two chains apart is the nineteen-year one, the Metonic. Seven is not a multiple of nineteen, so a shift of seven years does indeed move the position within the cycle. What that yields is a mirror within the model, not a test.
Workbenchthree hinges, one chain
Hinge
On this chronology (+205)
On the alternative chain (+198)
AD 31
AM 3996 — intercalary position → a Nisan in April → Wednesday 25 April
AM 3989 — ordinary year → a Nisan in March → the Wednesday loses its mirror
2031
AM 5996 — intercalary position
AM 5989 — ordinary year
2034
AM 5999 — intercalary position → the built-in Adar II → the day-count closes
AM 5992 — ordinary year → the calendar figure breaks
Three hinges, three hits on the one chain and three misses on the other. On the alternative chain not only does the year 6000 lose its landing — the Wednesday of AD 31 also loses its calendar mirror, and that is precisely the point where the standard objection sits.
And now the honest bolt, for without it this is a conjuring trick. This is an interpretation within the model, not a measurement. It presupposes the creation-anchoring of the cycle; it shares the assumptions of the chronology it mirrors; and the base rates belong with it (of the nineteen possible positions, four reproduce the complete pattern — roughly one in five). This is a mirror from within, not external proof, and it holds only at +205, the count from year 0: with the +206 between the starting dates (§6), 3997, 5997 and 6000 become ordinary years. It therefore does not count as a test and cannot make the Wednesday of Casus 31 break, because that book grounds the Adar II of 31 in the observational calendar. [INTERPRETED]
9 · The veil
There remains the question that no sum can answer. How can this be? How can the people of Scripture of all peoples, the people that counted the letters, lose two hundred years?
There is one answer that is dishonest and that is not given here, and that is the answer of gloating: look, they have reckoned themselves blind. Whoever says that has not read Romans 11 — and, worse, he has not seen who is standing there weeping.
For Paul, who saw this most sharply, does not handle it triumphantly. He begins by saying that he has great heaviness and continual sorrow in his heart, and that he could wish himself accursed if it could save his brethren (Rom. 9:1-3). Only then does he speak of the hardening. And he calls it neither total nor permanent: "blindness in part is happened to Israel, until the fulness of the Gentiles be come in" (Rom. 11:25). In part. Until. Two words that hold the door open.
And he adds something that puts in his place every Christian who exalts himself above Israel: "as touching the election, they are beloved for the fathers' sakes — for the gifts and calling of God are without repentance" (Rom. 11:28-29). Beloved. It says beloved.
Expertthe curse on those who calculate the end
One thread comes together here, and it is too striking to leave lying. In the Talmud stands a famous imprecation upon those who try to calculate the time of the Messiah: "tippach atzman shel mechashvei kitzin" — may the bones of those who calculate the ends be blasted (b. Sanhedrin 97b). And Daniel himself is told: "shut up the words, and seal the book, even to the time of the end" (12:4) — precisely the text on which Schwab built his "hidden years".
Honestly weighed, this proves nothing about intent; it shows an attitude. A tradition that forbids the calculating of the end has little reason to keep scrupulously pure the years on which such calculating would rest. That is no accusation — it is an explanation of why an error could remain standing without anyone having to will it. [INTERPRETED — attitude, not proof of intent]
And here the smiling must begin somewhere, for otherwise it becomes too heavy. There is something irresistibly human about all this. A people that does not entirely trust its own calendar and that, instead of repairing it, faithfully and for centuries puts a small phrase beneath it: according to the reckoning that we reckon. No denial. No glossing over. Simply an honest, modest footnote, every single time, in every contract, down through all the centuries.
That is almost moving in its integrity. And it is exactly the attitude this book demands of itself — for this book's two hundred and six years also hang on a choice the text does not name. This book too counts as it itself counts. The only difference a book can make is that it writes that down.
10 · What this chapter carries
The balance, briefly and without adornment.
What is established: the Jewish era is not reliable as a historical quantity; the gap sits at the Persians; it amounts to about 165 years; and it arose because the Seder Olam laid Daniel's four hundred and ninety years between two temple destructions. That is not speculation but a subtraction, and it has been acknowledged within the tradition itself for centuries, from Dei Rossi to Schwab.
What follows from it: the number 5786 cannot be an objection against this chronology, for it is itself a theological construction and not a measurement.
And what does not follow from it: that this book's two hundred and six is thereby settled. That rests on four entries, of which three are firm (the Persian gap, the Terah key, the deduction for the kings) and one contains a carried assumption — the seven years, and with them the dating of the covenant to Abram's eighty-second. That assumption has been made as strong as it can be made: the method is that of the tradition itself, the midrash supplies the missing date, and the Metonic clock points three times in the same direction, though that mirror does not count as a test (§8). But it remains an assumption, and it stands here at the head of the register, not in a footnote.
And with a people that God himself calls beloved, a second rule moreover applies, which weighs heavier than all chronology: no triumph.
HeartHEART
There is a veil over the clock, and the veil was not made by human hands. That is the deepest thing to be said here. Moses had to cover his face, and Paul writes that to this day a covering lies upon the reading — "nevertheless when it shall turn to the Lord, the vail shall be taken away" (2 Cor. 3:15-16). Taken away. Not destroyed, not torn off: taken away, as one lifts a cloth from the face of someone who is finally waking up.
So when this book counts the years and comes upon a gap, that is no victory. It is a wound you see on someone you love. The Messiah who was led outside the gate on the fourteenth of Nisan in the year 31 is a Son of this people, born under this law, counted in these genealogies. And he wept over this city (Luke 19:41).
Whoever reads this chapter with a smile of being right has read it wrongly. Read it with the smile of the little phrase at the foot of the ketubah — according to the reckoning that we reckon — and with the tear of Romans 9. And remember the ending: "and so all Israel shall be saved" (Rom. 11:26). So. At the end of the count stands no reproach. There stands a promise.
Core source [in full]: Rav Shimon Schwab, "Comparative Jewish Chronology", in Selected Speeches (1991), pp. 255-270 — full text consulted; all citations verbatim, including his own three objections and the word about the Khazars. His later retraction [secondary]. Rabbinic basis: Seder Olam Rabbah (2nd century); b. Avodah Zarah 9a, 10b (420 = 34 + 180 + 103 + 103, with a Persian period of 52 years); b. Sanhedrin 97b; Dan. 9:24; 12:4 [DOCUMENTED]. Criticism within the tradition: Azariah dei Rossi, Me'or Einayim (1573) [secondary]. Historical control: dated Babylonian tablets; the Canon of Ptolemy; Elephantine; Greek historiography — Persian period ±207 years (539-332 BC) [DOCUMENTED]. Computational core: AM 3338 → AM 3828 = 490 [CALCULATED]. New Testament: Rom. 9:1-3; 11:25-29; 2 Cor. 3:14-16; Luke 19:41.
The ledger [CALCULATED — reproducible]: +165 (Persian gap) + 60 (Terah key, Gen. 11:26/32 + 12:4 + Acts 7:4) + 12 (sojourn) − 30 (period of the kings; Thiele) = +207 against the Seder Olam (AD 70 = AM 3828); minus the seam of 2 years to the conventional calendar (AD 70 = 3830) = 205, the difference in year-numbers; the distance between the starting dates (autumn 3967 against autumn 3761) is 206. Source: the timeline of God's Algorithm beside the chronological table in Schwab, p. 259.
The seven years [INTERPRETED, triply labelled]: Gal. 3:17 (the 430 from the covenant, not the calling) + Gen. 15:13 (the 400 as sojourning with affliction) + Gal. 4:29 (the feast as the first persecution of the seed) + Bereshit Rabbah 53:13 (verified in the primary source via Sefaria: Ishmael "twenty-seven years old" at the casting out; with Gen. 16:16 and 21:5 → Isaac thirteen; the edition itself annotates the feast as a bar mitzvah). From this the covenant follows at AM 2090, when Abram was 82. Not claimed: no source dates Gen. 15 to 82 (Seder Olam: 70; Jubilees: ±85; Augustine leaves it open); no source says that the 400 begin at the feast; Pirkei DeRabbi Eliezer 30 knows, besides the 27, also a manuscript reading of 17 (within which Ussher chose). Inclusive versus completed counting gives ±1 year.
The test [CALCULATED] and the mirror [INTERPRETED]: the shemitah cycle cannot in principle distinguish the two chains (they differ by exactly one year-week; and the land-anchored count is chain-neutral). The Metonic cycle can do so, and points three times (AD 31, 2031, 2034) in the same direction, but only at +205 (count from year 0), not at +206 (from year 1); it does not count as a test. Base rates obligatory at every use: four of the nineteen positions reproduce the complete pattern. A mirror from within, not external proof.
Tone and limits. This chapter accuses no one of falsification; it shows an exegesis that became a calendar. Schwab is cited with his retraction. The hardening of Israel is read as Paul reads it: partial, temporary, and with tears. Whoever uses this material to set himself above Israel uses it against the intention of this book.
1 · A date with a package insert
A marriage contract is being drawn up. The ketubah lies on the table, the witnesses bend over it, and at the head of the sheet stands the date: in the year such-and-such since the creation of the world. And then, in every such document, for centuries now, there follows a small and curious little phrase. Lifrat she'anu monim. According to the reckoning that we reckon.
Pause there a moment. It is the most careful legal tradition in the world, a people that guards its calendar with a precision astronomers envy — and above every deed, every marriage letter, every contract of sale, they place a reservation. Not: the year since creation. But: the year since creation, according to our way of counting.
Rabbi Shimon Schwab, who wrestled with this question all his life, points it out himself. "This is perhaps the reason", he writes, "that in our documents we always add: minyan she'anu monim — according to the reckoning that we reckon." It is one of the most moving sentences in rabbinic literature. The tradition knows it. It has never concealed it. It only whispers it, in the margin of every contract.
This chapter is about what stands in that margin — and about the fact that the account which lies beneath it is this book's own account too.
2 · Where it goes wrong: the gap at the Persians
The Jewish era counts the years since creation. In the year 2026 the calendar reads 5786. That number does not fall out of the sky; it has been added up — generation by generation, king after king — and that sum stands, since the second century after Christ, in one book: the Seder Olam Rabbah, ascribed to Rabbi Jose ben Halafta.
Up to the exile that count runs reasonably parallel with what historians find. And then comes the Persian period, and something strange happens.
Persian rule over Israel lasted, by the sober reckoning of cuneiform, king-lists and Greek historians, about two hundred and seven years — from Cyrus' conquest of Babylon to Alexander. The Seder Olam gives it fifty-two. Not two hundred and seven. Fifty-two.
There sits the gap. One hundred and sixty-four years, fallen out between two pages.
Workbenchthe two reckonings side by side
Note the last line, for there sits the sting. At the end the count closes perfectly. That is because the era from creation has been calculated backwards from the year 70. The error therefore does not sit at the surface, where everyone would stumble over it — it lies buried in the middle, at the Persians. Whoever looks only at the present sees nothing. [DOCUMENTED]
The reckoning of the sages themselves, as Schwab cites it (b. Avodah Zarah 9a, 10b, following the Seder Olam): the 420 years of the second temple = 34 years Persian (of a Persian period that "lasted in total only 52 years") + 180 Greek + 103 Hasmonean + 103 Herodian. [DOCUMENTED — primary]
3 · Four kings do the work of ten
How do you press two hundred years into fifty-two? By merging kings.
The Persian king-list that history knows is long: Cyrus, Cambyses, Darius I, Xerxes, Artaxerxes I, Darius II, and a handful more. The Seder Olam knows four. And in the rabbinic exposition names coincide: Koresh, Daryavesh and Artachshasta are not three men but three names for the same man — the first a personal name, the other two honorific titles.
It is, it must be said, a solution of great elegance. It explains at a single stroke why Ezra and Nehemiah seem to be working under three different royal names. It turns a messy epoch into a tight story. And it is almost entirely wrong.
Archivehow one arrived at four kings
The classic equation — Koresh (Cyrus) = Daryavesh (Darius) = Artachshasta (Artaxerxes) as one monarch with three names — in effect makes the decrees of Ezra 1, Ezra 6 and Ezra 7 the decisions of the same man, and has Ezra arriving in Jerusalem only one year after the building of the temple.
And the tradition knew this grated. Schwab cites Biberfeld, who notes that these consequences "were rejected by Rabbi Zerachiah of Lunel as irreconcilable with the word of the Bible". The tension was therefore not invented by Christians; it had already been noticed within the Jewish tradition itself.
The hard side. Against the four kings of the Seder Olam stand dated Babylonian clay tablets from virtually every regnal year of every Persian king, the astronomically anchored Canon of Ptolemy, the Elephantine papyri, and Greek historiography. This is not one source against one source. It is an archive against a scheme. [DOCUMENTED]
4 · The number that explains everything — and accuses no one
Here comes the find that carries the whole case, and it is of a sobering simplicity. No conspiracy is needed. No deceit is needed. All that is needed is a subtraction, and it stands out in the open in the Seder Olam itself.
Take the year in which, according to the Seder Olam, the first temple falls: 3338 since creation. Take the year in which the second temple falls: 3828. Subtract.
Expert490
3828 − 3338 = 490.
Four hundred and ninety years. Precisely the seventy weeks of Daniel 9:24 — "seventy weeks are determined upon thy people and upon thy holy city".
That is no coincidence and no deceit. It is an exposition. The Seder Olam lets Daniel's seventy weeks run from the destruction of the first temple to the destruction of the second. And in order to make those four hundred and ninety years fit between two fixed end points, room had to go in the middle — and that room was taken away at the Persians, the epoch about which Scripture tells us least.
The calendar is therefore not a chronicle but an exegesis. It was not written in the first place to count the years, but to expound Daniel. And that says at once what the consequence is: on this reckoning Daniel's clock can no longer arrive at a Messiah in the midst of the week. It arrives at the year 70. [DOCUMENTED · INTERPRETED]
Let that sink in, for it is the heart of this whole dossier. The question "when do Daniel's seventy weeks end?" and the question "how old is the world?" have become, in the Jewish tradition, the same question. Whoever fixes the one answer fixes the other with it. The era and the messianic expectation are riveted to each other.
And that determines the tone of this chapter. Whoever cries "forgery" here has not understood. There is no bookkeeper who quietly struck years out. There is an expositor who read Daniel as he believed he had to read him — and that exposition has, centuries later, shaped the calendar of a whole people. That is not a crime. That is a tragedy, and a tragedy calls for compassion, not triumph.
5 · What the tradition itself says about it
The most remarkable thing about this matter is that the sharpest criticism comes not from outside but from within.
In the sixteenth century Azariah dei Rossi in his Me'or Einayim sets the rabbinic chronology beside the Greek and the Roman, and concludes that the traditional reckoning for the time before the second temple is not historically accurate: where the length of a period was unknown, the Seder Olam simply assumed the shortest possible one. His book was banned by several authorities. The question had been put, and it was not answered but buried.
And then, in 1962, a rabbi from Washington Heights takes up the thread again.
ArchiveRav Shimon Schwab, "Comparative Jewish Chronology" (1962)
The thesis. Schwab dares to consider that the sages concealed the years deliberately: "It seems possible that our Sages, for some unknown reason, 'covered up' a certain historic period and purposely eliminated and suppressed [...]" His ground is Daniel himself — the command to "shut up the words, and seal the book" (Dan. 12:4). That gives, he writes, "a perfect right to assume that certain historical events revealed to Daniel were omitted by him on purpose in faithful obedience" to that heavenly command.
The scale he himself names: "to leave out completely from our annals a period of at least 165 years and to correct all data and historic tables in such a fashion that the subsequent chronological gap could escape notice for countless generations."
And then — and this is why he is cited here with reverence — he turns against his own thesis. Three objections, from himself. First: "The events described in a history book have either happened or they haven't [...] A chronological time table is the backbone of any book on history which expects to be accepted by intellectually honest students." Second, and it is almost audible how it pains him: if our forefathers wanted to hide a period at divine command, "who has the right to uncover the carefully concealed secret? [...] we still were not given the right to remove the veil for any reason." He compares himself with the king of the Khazars, who was told in a dream: "Your intentions are acceptable, but your deeds are not."
The honest outcome. Towards the end of his life Schwab withdrew his position — among other reasons because a wilful concealment would have irreparably damaged the halachic calculations themselves (the molad). Whoever adduces Schwab as a witness for a deliberate falsification uses him against his own last word. This book does not do so. [DOCUMENTED — essay consulted in full; the retraction secondary]
See what happens here. An orthodox rabbi sees the gap, names it, dares to consider the gravest explanation — and then withdraws, not because he is afraid of the facts, but because he holds that he may not lift the veil. That is not cowardice. That is a man standing between two kinds of loyalty and unable to resolve it. No mockery fits there. A silence fits there.
6 · The ledger — where the missing years come from
Now the question for which this chapter was written. How many years are missing? And the answer is not one number but an account with four entries — and that is better than one number, for every entry can be tested separately.
This chronology puts creation at about 3967 BC; the rabbinic reckoning at 3761. The difference is two hundred and six years. This is how it is built up:
Workbenchthe entries of the ledger
The ledger measures against the Seder Olam, which calls the year 70 AM 3828; the conventional calendar calls it 3830, a seam of 2 years within the rabbinic count. 207 − 2 = 205: the difference in year-numbers (5794 = 5999). Plus one, because the clock begins at year 0 and the calendar at year 1, gives 206: the distance between the starting dates. The ledger closes, but it follows the anchor (the temple in Ziv 966) and is therefore no independent test. [CALCULATED — reproducible from two tables]
The +60 is the Terah key, and it is the best-anchored entry of the four. Genesis 11:26 reads as if Terah was seventy when he begat Abram. But Terah died at two hundred and five (Gen. 11:32), Abram departed from Haran at seventy-five (Gen. 12:4), and that departure was after his father's death (Acts 7:4). Two hundred and five minus seventy-five is one hundred and thirty. The seventy was the beginning of Terah's begetting of children; Abram was the youngest. Sixty years, and they rest on three texts, among them an explicit New Testament exposition.
The −30 is the period of the kings, and it runs the other way — and precisely for that reason it is worth mentioning. The Seder Olam gives the first-temple period four hundred and ten years; the historical length is 380. Here the rabbinic tradition therefore counts too many, not too few, because a naive addition of the regnal years of Judah and Israel runs too long as long as you do not discount the co-regencies. That is exactly what Thiele resolved.
Say that out loud, for it is the test of the honesty of this whole chapter: whoever cries that "the Jews are missing years" must add that in the period of the kings they have a few too many. A set of books with nothing but windfalls is suspect. This one has a shortfall, and that speaks in its favour.
7 · The seven years — the assumption that has to be carried
There remains the third entry: the twelve years between Abraham's birth and the exodus. Five of them are simply explained (this book reckons two hundred and fifteen years in Egypt where the rabbis reckon two hundred and ten). And the remaining seven call for a decision that may not be concealed.
The problem can be stated sharply. The chain up to Abraham's calling is fixed: 1656 to the Flood, plus 352 to Abraham's birth, plus 75 to his calling — together AM 2083. Add the four-hundred-and-thirty-year sojourn to that, and you land on AM 2513. But the exodus stands at AM 2520.
And here a temptation must first be removed, for it is the first one everyone tries. You cannot find those seven years somewhere within the sojourn — not with Joseph, not with the seven years of famine, not with the slavery. Why not? Because every division of the four hundred and thirty years lands on the same result. Whether you split it as 215 + 215, or 208 + 222, or 400 + 30 — the sum remains four hundred and thirty, and the exodus remains 2513. The famine moreover straddles the descent: Jacob goes down in the second year of famine (Gen. 45:6). It is not a block that you can shove in front of it or behind it.
The gap therefore lies not in the sojourn, but at its starting point. And there Scripture itself shows the way.
For Paul counts the four hundred and thirty years not from Abraham's calling, but from the covenant: "the covenant, that was confirmed before of God, the law, which was four hundred and thirty years after, cannot disannul" (Gal. 3:17). And the covenant is an identifiable event — Genesis 15, the ceremony of the divided animals. It is moreover precisely there that the clock is pronounced: "know of a surety that thy seed shall be a stranger [...] four hundred years" (Gen. 15:13). The counting down begins where it is announced.
If Genesis 15 falls in the year in which Abram was 82 (AM 2090), the chain closes exactly: 2090 plus 430 is 2520. But the text does not name that year. The window runs from his seventy-fifth (the calling) to his eighty-fifth (Genesis 16, where he has dwelt ten years in Canaan). Why then eighty-two of all years?
The surprising answer: because the midrash itself hands it over — from the other side.
ArchiveBereshit Rabbah dates the feast itself
At the casting out of Hagar and Ishmael it stands that Abraham sent them away with the child "on her shoulder". And the midrash balks at that, for Ishmael was no infant at the time. Bereshit Rabbah 53:13, verbatim (Sefaria):
"'On her shoulder and the child' — he was twenty-seven years old — and yet you say: 'On her shoulder'?"
The edition annotates it itself: "The Midrash understands that the feast made in honor of Isaac was for his Bar Mitzva, and Yishmael was fourteen years older than Isaac." And that is right: Ishmael was born when Abram was eighty-six (Gen. 16:16), Isaac when he was a hundred (Gen. 21:5) — fourteen years' difference. Ishmael twenty-seven therefore means Isaac thirteen.
The midrash therefore dates the feast of weaning and the casting out to Isaac's thirteenth year — AM 2120. And that is precisely the point at which the four hundred years of Genesis 15:13 must begin in order to close on the exodus of AM 2520. It does not stand there as chronology; it stands there in passing, to explain a detail about a shoulder. But it stands there. [DOCUMENTED — verified in the primary source]
Now turn the derivation round, and see what happens. Take the midrash's dating (the feast in Isaac's thirteenth year, AM 2120). Take the fact that the four hundred of Genesis 15:13 and the four hundred and thirty of Galatians 3:17 must close on the same exodus — that is a difference of thirty years. Then the covenant follows at 2120 minus 30 = AM 2090, and then Abram is 82. The eighty-two is then no longer a chosen number, but a calculated result.
And note the symmetry with the opposing party, for it is perfect. The Seder Olam builds exactly the same bridge of thirty years — only there it runs from the covenant (when Abram was 70) to Isaac's birth, and here from the covenant (when Abram was 82) to Isaac's persecution. Both systems hang the pair 400/430 on a single thirty-year bridge. The only difference is which event closes that bridge. And that the persecution is the starting point is said by Genesis 15:13 itself — it concerns sojourning with affliction — and Paul reads the weaning feast precisely so: "he that was born after the flesh persecuted him that was born after the Spirit" (Gal. 4:29).
Counterpointwhat is not claimed here
Nobody dates Genesis 15 to Abram's eighty-second. Not the Seder Olam (which chooses seventy), not Jubilees (which arrives at about eighty-five), not Josephus, not Augustine (who expressly leaves the dating open). The eighty-two is a reconstruction. What is documented is the window in which it falls — and that Jubilees reads that window in its later half, after the calling, precisely as the order of the narrative requires.
And no ancient source says that the four hundred years begin at the feast. That remains the interpreted link — now with three threads beneath it instead of none: Genesis 15:13 itself defines the four hundred as the sojourning with affliction of the seed; Galatians 4:29 reads the feast as the first persecution of that seed; and the midrash dates that feast to Isaac's thirteenth year.
There is moreover a second midrashic tradition. Pirkei DeRabbi Eliezer 30 knows, besides the twenty-seven, also a manuscript reading of seventeen, and Ussher chose within that computational tradition (weaning at about five years). This book chooses the explicit midrashic chronology over the computational tradition. That is a choice — with a primary source, and the mirror image of the choice Ussher made. It must remain labelled as a choice.
And an honest ±1. The chain closes exactly on the inclusive count ("in his twenty-seventh year"). On the completed count it lands one year later. That difference falls within the noise that the chronology itself already concedes — but it is named here, not hidden away. [INTERPRETED — doubly labelled]
8 · What the sabbatical cycle cannot say — and what the clock at most mirrors
There lies a tempting thought here: let the sabbatical years decide. If creation falls in 3967 or in 3960, surely one of the two must clash with the shemitah cycle?
No. And it is better that this book says so itself than that a critic should find it.
Expertwhy the shemitah can in principle measure nothing here
The two candidates differ by precisely seven years — one full year-week. And every seven-year cycle is invariant under a shift of seven: 3967 and 3960 give exactly the same shemitah positions. Mathematically undecidable.
And the other anchoring does not help either: if one counts the sabbatical years from the entry into the land (Lev. 25), then the BC dates are identical on both chains — those hang on the temple of 966 plus 480, not on the era from creation. Chain-neutral by construction.
That is a neutral result, not a loss: nothing in the sabbatical evidence contradicts this chronology. It simply cannot distinguish it from its competitor. Whoever hoped for a measurement here must go without. [CALCULATED]
A cycle that does set the two chains apart is the nineteen-year one, the Metonic. Seven is not a multiple of nineteen, so a shift of seven years does indeed move the position within the cycle. What that yields is a mirror within the model, not a test.
Workbenchthree hinges, one chain
Three hinges, three hits on the one chain and three misses on the other. On the alternative chain not only does the year 6000 lose its landing — the Wednesday of AD 31 also loses its calendar mirror, and that is precisely the point where the standard objection sits.
And now the honest bolt, for without it this is a conjuring trick. This is an interpretation within the model, not a measurement. It presupposes the creation-anchoring of the cycle; it shares the assumptions of the chronology it mirrors; and the base rates belong with it (of the nineteen possible positions, four reproduce the complete pattern — roughly one in five). This is a mirror from within, not external proof, and it holds only at +205, the count from year 0: with the +206 between the starting dates (§6), 3997, 5997 and 6000 become ordinary years. It therefore does not count as a test and cannot make the Wednesday of Casus 31 break, because that book grounds the Adar II of 31 in the observational calendar. [INTERPRETED]
9 · The veil
There remains the question that no sum can answer. How can this be? How can the people of Scripture of all peoples, the people that counted the letters, lose two hundred years?
There is one answer that is dishonest and that is not given here, and that is the answer of gloating: look, they have reckoned themselves blind. Whoever says that has not read Romans 11 — and, worse, he has not seen who is standing there weeping.
For Paul, who saw this most sharply, does not handle it triumphantly. He begins by saying that he has great heaviness and continual sorrow in his heart, and that he could wish himself accursed if it could save his brethren (Rom. 9:1-3). Only then does he speak of the hardening. And he calls it neither total nor permanent: "blindness in part is happened to Israel, until the fulness of the Gentiles be come in" (Rom. 11:25). In part. Until. Two words that hold the door open.
And he adds something that puts in his place every Christian who exalts himself above Israel: "as touching the election, they are beloved for the fathers' sakes — for the gifts and calling of God are without repentance" (Rom. 11:28-29). Beloved. It says beloved.
Expertthe curse on those who calculate the end
One thread comes together here, and it is too striking to leave lying. In the Talmud stands a famous imprecation upon those who try to calculate the time of the Messiah: "tippach atzman shel mechashvei kitzin" — may the bones of those who calculate the ends be blasted (b. Sanhedrin 97b). And Daniel himself is told: "shut up the words, and seal the book, even to the time of the end" (12:4) — precisely the text on which Schwab built his "hidden years".
Honestly weighed, this proves nothing about intent; it shows an attitude. A tradition that forbids the calculating of the end has little reason to keep scrupulously pure the years on which such calculating would rest. That is no accusation — it is an explanation of why an error could remain standing without anyone having to will it. [INTERPRETED — attitude, not proof of intent]
And here the smiling must begin somewhere, for otherwise it becomes too heavy. There is something irresistibly human about all this. A people that does not entirely trust its own calendar and that, instead of repairing it, faithfully and for centuries puts a small phrase beneath it: according to the reckoning that we reckon. No denial. No glossing over. Simply an honest, modest footnote, every single time, in every contract, down through all the centuries.
That is almost moving in its integrity. And it is exactly the attitude this book demands of itself — for this book's two hundred and six years also hang on a choice the text does not name. This book too counts as it itself counts. The only difference a book can make is that it writes that down.
10 · What this chapter carries
The balance, briefly and without adornment.
What is established: the Jewish era is not reliable as a historical quantity; the gap sits at the Persians; it amounts to about 165 years; and it arose because the Seder Olam laid Daniel's four hundred and ninety years between two temple destructions. That is not speculation but a subtraction, and it has been acknowledged within the tradition itself for centuries, from Dei Rossi to Schwab.
What follows from it: the number 5786 cannot be an objection against this chronology, for it is itself a theological construction and not a measurement.
And what does not follow from it: that this book's two hundred and six is thereby settled. That rests on four entries, of which three are firm (the Persian gap, the Terah key, the deduction for the kings) and one contains a carried assumption — the seven years, and with them the dating of the covenant to Abram's eighty-second. That assumption has been made as strong as it can be made: the method is that of the tradition itself, the midrash supplies the missing date, and the Metonic clock points three times in the same direction, though that mirror does not count as a test (§8). But it remains an assumption, and it stands here at the head of the register, not in a footnote.
And with a people that God himself calls beloved, a second rule moreover applies, which weighs heavier than all chronology: no triumph.
HeartHEART
There is a veil over the clock, and the veil was not made by human hands. That is the deepest thing to be said here. Moses had to cover his face, and Paul writes that to this day a covering lies upon the reading — "nevertheless when it shall turn to the Lord, the vail shall be taken away" (2 Cor. 3:15-16). Taken away. Not destroyed, not torn off: taken away, as one lifts a cloth from the face of someone who is finally waking up.
So when this book counts the years and comes upon a gap, that is no victory. It is a wound you see on someone you love. The Messiah who was led outside the gate on the fourteenth of Nisan in the year 31 is a Son of this people, born under this law, counted in these genealogies. And he wept over this city (Luke 19:41).
Whoever reads this chapter with a smile of being right has read it wrongly. Read it with the smile of the little phrase at the foot of the ketubah — according to the reckoning that we reckon — and with the tear of Romans 9. And remember the ending: "and so all Israel shall be saved" (Rom. 11:26). So. At the end of the count stands no reproach. There stands a promise.
Appendix · Sources and verification status
Status: [in full] primary source fully consulted · [secondary] via reliable report · [CALCULATED] our own, reproducible · [INTERPRETED] carried assumption.
Core source [in full]: Rav Shimon Schwab, "Comparative Jewish Chronology", in Selected Speeches (1991), pp. 255-270 — full text consulted; all citations verbatim, including his own three objections and the word about the Khazars. His later retraction [secondary]. Rabbinic basis: Seder Olam Rabbah (2nd century); b. Avodah Zarah 9a, 10b (420 = 34 + 180 + 103 + 103, with a Persian period of 52 years); b. Sanhedrin 97b; Dan. 9:24; 12:4 [DOCUMENTED]. Criticism within the tradition: Azariah dei Rossi, Me'or Einayim (1573) [secondary]. Historical control: dated Babylonian tablets; the Canon of Ptolemy; Elephantine; Greek historiography — Persian period ±207 years (539-332 BC) [DOCUMENTED]. Computational core: AM 3338 → AM 3828 = 490 [CALCULATED]. New Testament: Rom. 9:1-3; 11:25-29; 2 Cor. 3:14-16; Luke 19:41.
The ledger [CALCULATED — reproducible]: +165 (Persian gap) + 60 (Terah key, Gen. 11:26/32 + 12:4 + Acts 7:4) + 12 (sojourn) − 30 (period of the kings; Thiele) = +207 against the Seder Olam (AD 70 = AM 3828); minus the seam of 2 years to the conventional calendar (AD 70 = 3830) = 205, the difference in year-numbers; the distance between the starting dates (autumn 3967 against autumn 3761) is 206. Source: the timeline of God's Algorithm beside the chronological table in Schwab, p. 259.
The seven years [INTERPRETED, triply labelled]: Gal. 3:17 (the 430 from the covenant, not the calling) + Gen. 15:13 (the 400 as sojourning with affliction) + Gal. 4:29 (the feast as the first persecution of the seed) + Bereshit Rabbah 53:13 (verified in the primary source via Sefaria: Ishmael "twenty-seven years old" at the casting out; with Gen. 16:16 and 21:5 → Isaac thirteen; the edition itself annotates the feast as a bar mitzvah). From this the covenant follows at AM 2090, when Abram was 82. Not claimed: no source dates Gen. 15 to 82 (Seder Olam: 70; Jubilees: ±85; Augustine leaves it open); no source says that the 400 begin at the feast; Pirkei DeRabbi Eliezer 30 knows, besides the 27, also a manuscript reading of 17 (within which Ussher chose). Inclusive versus completed counting gives ±1 year.
The test [CALCULATED] and the mirror [INTERPRETED]: the shemitah cycle cannot in principle distinguish the two chains (they differ by exactly one year-week; and the land-anchored count is chain-neutral). The Metonic cycle can do so, and points three times (AD 31, 2031, 2034) in the same direction, but only at +205 (count from year 0), not at +206 (from year 1); it does not count as a test. Base rates obligatory at every use: four of the nineteen positions reproduce the complete pattern. A mirror from within, not external proof.
Tone and limits. This chapter accuses no one of falsification; it shows an exegesis that became a calendar. Schwab is cited with his retraction. The hardening of Israel is read as Paul reads it: partial, temporary, and with tears. Whoever uses this material to set himself above Israel uses it against the intention of this book.